diff options
| author | Maxime Ripard <maxime@cerno.tech> | 2020-04-04 11:02:15 +0200 |
|---|---|---|
| committer | Maxime Ripard <maxime@cerno.tech> | 2020-04-04 11:02:15 +0200 |
| commit | cc46c03397c1865a181f1a4f66d4645806e5a943 (patch) | |
| tree | e3a762d8961219a61b96ae6e5ec875ba9fbe2314 /drivers | |
| parent | drm/mm: revert "Break long searches in fragmented address spaces" (diff) | |
| parent | Merge branch 'ttm-transhuge' of git://people.freedesktop.org/~thomash/linux i... (diff) | |
| download | linux-cc46c03397c1865a181f1a4f66d4645806e5a943.tar.xz linux-cc46c03397c1865a181f1a4f66d4645806e5a943.zip | |
Merge drm/drm-next into drm-misc-next-fixes
Alex needs v5.6 into drm-misc-next-fixes to merge a fix for a regression in
the scatterlist processing in PRIME.
Signed-off-by: Maxime Ripard <maxime@cerno.tech>
Diffstat (limited to 'drivers')
1002 files changed, 22338 insertions, 12639 deletions
diff --git a/drivers/acpi/acpi_watchdog.c b/drivers/acpi/acpi_watchdog.c index b5516b04ffc0..6e9ec6e3fe47 100644 --- a/drivers/acpi/acpi_watchdog.c +++ b/drivers/acpi/acpi_watchdog.c @@ -55,12 +55,14 @@ static bool acpi_watchdog_uses_rtc(const struct acpi_table_wdat *wdat) } #endif +static bool acpi_no_watchdog; + static const struct acpi_table_wdat *acpi_watchdog_get_wdat(void) { const struct acpi_table_wdat *wdat = NULL; acpi_status status; - if (acpi_disabled) + if (acpi_disabled || acpi_no_watchdog) return NULL; status = acpi_get_table(ACPI_SIG_WDAT, 0, @@ -88,6 +90,14 @@ bool acpi_has_watchdog(void) } EXPORT_SYMBOL_GPL(acpi_has_watchdog); +/* ACPI watchdog can be disabled on boot command line */ +static int __init disable_acpi_watchdog(char *str) +{ + acpi_no_watchdog = true; + return 1; +} +__setup("acpi_no_watchdog", disable_acpi_watchdog); + void __init acpi_watchdog_init(void) { const struct acpi_wdat_entry *entries; @@ -126,12 +136,11 @@ void __init acpi_watchdog_init(void) gas = &entries[i].register_region; res.start = gas->address; + res.end = res.start + ACPI_ACCESS_BYTE_WIDTH(gas->access_width) - 1; if (gas->space_id == ACPI_ADR_SPACE_SYSTEM_MEMORY) { res.flags = IORESOURCE_MEM; - res.end = res.start + ALIGN(gas->access_width, 4) - 1; } else if (gas->space_id == ACPI_ADR_SPACE_SYSTEM_IO) { res.flags = IORESOURCE_IO; - res.end = res.start + gas->access_width - 1; } else { pr_warn("Unsupported address space: %u\n", gas->space_id); diff --git a/drivers/acpi/acpica/evevent.c b/drivers/acpi/acpica/evevent.c index 8c83d8c620dc..789d5e920aaf 100644 --- a/drivers/acpi/acpica/evevent.c +++ b/drivers/acpi/acpica/evevent.c @@ -265,4 +265,49 @@ static u32 acpi_ev_fixed_event_dispatch(u32 event) handler) (acpi_gbl_fixed_event_handlers[event].context)); } +/******************************************************************************* + * + * FUNCTION: acpi_any_fixed_event_status_set + * + * PARAMETERS: None + * + * RETURN: TRUE or FALSE + * + * DESCRIPTION: Checks the PM status register for active fixed events + * + ******************************************************************************/ + +u32 acpi_any_fixed_event_status_set(void) +{ + acpi_status status; + u32 in_status; + u32 in_enable; + u32 i; + + status = acpi_hw_register_read(ACPI_REGISTER_PM1_ENABLE, &in_enable); + if (ACPI_FAILURE(status)) { + return (FALSE); + } + + status = acpi_hw_register_read(ACPI_REGISTER_PM1_STATUS, &in_status); + if (ACPI_FAILURE(status)) { + return (FALSE); + } + + /* + * Check for all possible Fixed Events and dispatch those that are active + */ + for (i = 0; i < ACPI_NUM_FIXED_EVENTS; i++) { + + /* Both the status and enable bits must be on for this event */ + + if ((in_status & acpi_gbl_fixed_event_info[i].status_bit_mask) && + (in_enable & acpi_gbl_fixed_event_info[i].enable_bit_mask)) { + return (TRUE); + } + } + + return (FALSE); +} + #endif /* !ACPI_REDUCED_HARDWARE */ diff --git a/drivers/acpi/apei/ghes.c b/drivers/acpi/apei/ghes.c index 103acbbfcf9a..24c9642e8fc7 100644 --- a/drivers/acpi/apei/ghes.c +++ b/drivers/acpi/apei/ghes.c @@ -171,7 +171,7 @@ int ghes_estatus_pool_init(int num_ghes) * New allocation must be visible in all pgd before it can be found by * an NMI allocating from the pool. */ - vmalloc_sync_all(); + vmalloc_sync_mappings(); rc = gen_pool_add(ghes_estatus_pool, addr, PAGE_ALIGN(len), -1); if (rc) diff --git a/drivers/acpi/sleep.c b/drivers/acpi/sleep.c index 152f7fc0b200..e5f95922bc21 100644 --- a/drivers/acpi/sleep.c +++ b/drivers/acpi/sleep.c @@ -1006,6 +1006,13 @@ static bool acpi_s2idle_wake(void) return true; /* + * If the status bit of any enabled fixed event is set, the + * wakeup is regarded as valid. + */ + if (acpi_any_fixed_event_status_set()) + return true; + + /* * If there are no EC events to process and at least one of the * other enabled GPEs is active, the wakeup is regarded as a * genuine one. diff --git a/drivers/android/binder.c b/drivers/android/binder.c index a6b2082c24f8..e47c8a4c83db 100644 --- a/drivers/android/binder.c +++ b/drivers/android/binder.c @@ -5228,6 +5228,7 @@ static int binder_open(struct inode *nodp, struct file *filp) binder_dev = container_of(filp->private_data, struct binder_device, miscdev); } + refcount_inc(&binder_dev->ref); proc->context = &binder_dev->context; binder_alloc_init(&proc->alloc); @@ -5405,6 +5406,7 @@ static int binder_node_release(struct binder_node *node, int refs) static void binder_deferred_release(struct binder_proc *proc) { struct binder_context *context = proc->context; + struct binder_device *device; struct rb_node *n; int threads, nodes, incoming_refs, outgoing_refs, active_transactions; @@ -5421,6 +5423,12 @@ static void binder_deferred_release(struct binder_proc *proc) context->binder_context_mgr_node = NULL; } mutex_unlock(&context->context_mgr_node_lock); + device = container_of(proc->context, struct binder_device, context); + if (refcount_dec_and_test(&device->ref)) { + kfree(context->name); + kfree(device); + } + proc->context = NULL; binder_inner_proc_lock(proc); /* * Make sure proc stays alive after we @@ -6077,6 +6085,7 @@ static int __init init_binder_device(const char *name) binder_device->miscdev.minor = MISC_DYNAMIC_MINOR; binder_device->miscdev.name = name; + refcount_set(&binder_device->ref, 1); binder_device->context.binder_context_mgr_uid = INVALID_UID; binder_device->context.name = name; mutex_init(&binder_device->context.context_mgr_node_lock); diff --git a/drivers/android/binder_internal.h b/drivers/android/binder_internal.h index ae991097d14d..283d3cb9c16e 100644 --- a/drivers/android/binder_internal.h +++ b/drivers/android/binder_internal.h @@ -8,6 +8,7 @@ #include <linux/list.h> #include <linux/miscdevice.h> #include <linux/mutex.h> +#include <linux/refcount.h> #include <linux/stddef.h> #include <linux/types.h> #include <linux/uidgid.h> @@ -33,6 +34,7 @@ struct binder_device { struct miscdevice miscdev; struct binder_context context; struct inode *binderfs_inode; + refcount_t ref; }; /** diff --git a/drivers/android/binderfs.c b/drivers/android/binderfs.c index e2580e5316a2..f303106b3362 100644 --- a/drivers/android/binderfs.c +++ b/drivers/android/binderfs.c @@ -154,6 +154,7 @@ static int binderfs_binder_device_create(struct inode *ref_inode, if (!name) goto err; + refcount_set(&device->ref, 1); device->binderfs_inode = inode; device->context.binder_context_mgr_uid = INVALID_UID; device->context.name = name; @@ -257,8 +258,10 @@ static void binderfs_evict_inode(struct inode *inode) ida_free(&binderfs_minors, device->miscdev.minor); mutex_unlock(&binderfs_minors_mutex); - kfree(device->context.name); - kfree(device); + if (refcount_dec_and_test(&device->ref)) { + kfree(device->context.name); + kfree(device); + } } /** @@ -445,6 +448,7 @@ static int binderfs_binder_ctl_create(struct super_block *sb) inode->i_uid = info->root_uid; inode->i_gid = info->root_gid; + refcount_set(&device->ref, 1); device->binderfs_inode = inode; device->miscdev.minor = minor; diff --git a/drivers/atm/nicstar.c b/drivers/atm/nicstar.c index 8db8c0fb5e2d..7af74fb450a0 100644 --- a/drivers/atm/nicstar.c +++ b/drivers/atm/nicstar.c @@ -91,7 +91,7 @@ #ifdef GENERAL_DEBUG #define PRINTK(args...) printk(args) #else -#define PRINTK(args...) +#define PRINTK(args...) do {} while (0) #endif /* GENERAL_DEBUG */ #ifdef EXTRA_DEBUG diff --git a/drivers/auxdisplay/Kconfig b/drivers/auxdisplay/Kconfig index b8313a04422d..48efa7a047f3 100644 --- a/drivers/auxdisplay/Kconfig +++ b/drivers/auxdisplay/Kconfig @@ -111,7 +111,7 @@ config CFAG12864B If unsure, say N. config CFAG12864B_RATE - int "Refresh rate (hertz)" + int "Refresh rate (hertz)" depends on CFAG12864B default "20" ---help--- @@ -329,7 +329,7 @@ config PANEL_LCD_PROTO config PANEL_LCD_PIN_E depends on PANEL_PROFILE="0" && PANEL_LCD="1" && PANEL_LCD_PROTO="0" - int "Parallel port pin number & polarity connected to the LCD E signal (-17...17) " + int "Parallel port pin number & polarity connected to the LCD E signal (-17...17) " range -17 17 default 14 ---help--- @@ -344,7 +344,7 @@ config PANEL_LCD_PIN_E config PANEL_LCD_PIN_RS depends on PANEL_PROFILE="0" && PANEL_LCD="1" && PANEL_LCD_PROTO="0" - int "Parallel port pin number & polarity connected to the LCD RS signal (-17...17) " + int "Parallel port pin number & polarity connected to the LCD RS signal (-17...17) " range -17 17 default 17 ---help--- @@ -359,7 +359,7 @@ config PANEL_LCD_PIN_RS config PANEL_LCD_PIN_RW depends on PANEL_PROFILE="0" && PANEL_LCD="1" && PANEL_LCD_PROTO="0" - int "Parallel port pin number & polarity connected to the LCD RW signal (-17...17) " + int "Parallel port pin number & polarity connected to the LCD RW signal (-17...17) " range -17 17 default 16 ---help--- @@ -374,7 +374,7 @@ config PANEL_LCD_PIN_RW config PANEL_LCD_PIN_SCL depends on PANEL_PROFILE="0" && PANEL_LCD="1" && PANEL_LCD_PROTO!="0" - int "Parallel port pin number & polarity connected to the LCD SCL signal (-17...17) " + int "Parallel port pin number & polarity connected to the LCD SCL signal (-17...17) " range -17 17 default 1 ---help--- @@ -389,7 +389,7 @@ config PANEL_LCD_PIN_SCL config PANEL_LCD_PIN_SDA depends on PANEL_PROFILE="0" && PANEL_LCD="1" && PANEL_LCD_PROTO!="0" - int "Parallel port pin number & polarity connected to the LCD SDA signal (-17...17) " + int "Parallel port pin number & polarity connected to the LCD SDA signal (-17...17) " range -17 17 default 2 ---help--- @@ -404,12 +404,12 @@ config PANEL_LCD_PIN_SDA config PANEL_LCD_PIN_BL depends on PANEL_PROFILE="0" && PANEL_LCD="1" - int "Parallel port pin number & polarity connected to the LCD backlight signal (-17...17) " + int "Parallel port pin number & polarity connected to the LCD backlight signal (-17...17) " range -17 17 default 0 ---help--- This describes the number of the parallel port pin to which the LCD 'BL' signal - has been connected. It can be : + has been connected. It can be : 0 : no connection (eg: connected to ground) 1..17 : directly connected to any of these pins on the DB25 plug diff --git a/drivers/auxdisplay/charlcd.c b/drivers/auxdisplay/charlcd.c index 874c259a8829..c0da3820454b 100644 --- a/drivers/auxdisplay/charlcd.c +++ b/drivers/auxdisplay/charlcd.c @@ -88,7 +88,7 @@ struct charlcd_priv { int len; } esc_seq; - unsigned long long drvdata[0]; + unsigned long long drvdata[]; }; #define charlcd_to_priv(p) container_of(p, struct charlcd_priv, lcd) diff --git a/drivers/auxdisplay/img-ascii-lcd.c b/drivers/auxdisplay/img-ascii-lcd.c index efb928e25aef..1cce409ce5ca 100644 --- a/drivers/auxdisplay/img-ascii-lcd.c +++ b/drivers/auxdisplay/img-ascii-lcd.c @@ -356,7 +356,6 @@ static int img_ascii_lcd_probe(struct platform_device *pdev) const struct of_device_id *match; const struct img_ascii_lcd_config *cfg; struct img_ascii_lcd_ctx *ctx; - struct resource *res; int err; match = of_match_device(img_ascii_lcd_matches, &pdev->dev); @@ -378,8 +377,7 @@ static int img_ascii_lcd_probe(struct platform_device *pdev) &ctx->offset)) return -EINVAL; } else { - res = platform_get_resource(pdev, IORESOURCE_MEM, 0); - ctx->base = devm_ioremap_resource(&pdev->dev, res); + ctx->base = devm_platform_ioremap_resource(pdev, 0); if (IS_ERR(ctx->base)) return PTR_ERR(ctx->base); } diff --git a/drivers/base/core.c b/drivers/base/core.c index 42a672456432..dbb0f9130f42 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -718,6 +718,8 @@ static void __device_links_queue_sync_state(struct device *dev, { struct device_link *link; + if (!dev_has_sync_state(dev)) + return; if (dev->state_synced) return; @@ -745,25 +747,31 @@ static void __device_links_queue_sync_state(struct device *dev, /** * device_links_flush_sync_list - Call sync_state() on a list of devices * @list: List of devices to call sync_state() on + * @dont_lock_dev: Device for which lock is already held by the caller * * Calls sync_state() on all the devices that have been queued for it. This - * function is used in conjunction with __device_links_queue_sync_state(). + * function is used in conjunction with __device_links_queue_sync_state(). The + * @dont_lock_dev parameter is useful when this function is called from a + * context where a device lock is already held. */ -static void device_links_flush_sync_list(struct list_head *list) +static void device_links_flush_sync_list(struct list_head *list, + struct device *dont_lock_dev) { struct device *dev, *tmp; list_for_each_entry_safe(dev, tmp, list, links.defer_sync) { list_del_init(&dev->links.defer_sync); - device_lock(dev); + if (dev != dont_lock_dev) + device_lock(dev); if (dev->bus->sync_state) dev->bus->sync_state(dev); else if (dev->driver && dev->driver->sync_state) dev->driver->sync_state(dev); - device_unlock(dev); + if (dev != dont_lock_dev) + device_unlock(dev); put_device(dev); } @@ -801,7 +809,7 @@ void device_links_supplier_sync_state_resume(void) out: device_links_write_unlock(); - device_links_flush_sync_list(&sync_list); + device_links_flush_sync_list(&sync_list, NULL); } static int sync_state_resume_initcall(void) @@ -813,7 +821,7 @@ late_initcall(sync_state_resume_initcall); static void __device_links_supplier_defer_sync(struct device *sup) { - if (list_empty(&sup->links.defer_sync)) + if (list_empty(&sup->links.defer_sync) && dev_has_sync_state(sup)) list_add_tail(&sup->links.defer_sync, &deferred_sync); } @@ -865,6 +873,11 @@ void device_links_driver_bound(struct device *dev) driver_deferred_probe_add(link->consumer); } + if (defer_sync_state_count) + __device_links_supplier_defer_sync(dev); + else + __device_links_queue_sync_state(dev, &sync_list); + list_for_each_entry(link, &dev->links.suppliers, c_node) { if (!(link->flags & DL_FLAG_MANAGED)) continue; @@ -883,7 +896,7 @@ void device_links_driver_bound(struct device *dev) device_links_write_unlock(); - device_links_flush_sync_list(&sync_list); + device_links_flush_sync_list(&sync_list, dev); } static void device_link_drop_managed(struct device_link *link) diff --git a/drivers/base/memory.c b/drivers/base/memory.c index b9f474c11393..4086718f6876 100644 --- a/drivers/base/memory.c +++ b/drivers/base/memory.c @@ -97,30 +97,13 @@ static ssize_t phys_index_show(struct device *dev, } /* - * Show whether the memory block is likely to be offlineable (or is already - * offline). Once offline, the memory block could be removed. The return - * value does, however, not indicate that there is a way to remove the - * memory block. + * Legacy interface that we cannot remove. Always indicate "removable" + * with CONFIG_MEMORY_HOTREMOVE - bad heuristic. */ static ssize_t removable_show(struct device *dev, struct device_attribute *attr, char *buf) { - struct memory_block *mem = to_memory_block(dev); - unsigned long pfn; - int ret = 1, i; - - if (mem->state != MEM_ONLINE) - goto out; - - for (i = 0; i < sections_per_block; i++) { - if (!present_section_nr(mem->start_section_nr + i)) - continue; - pfn = section_nr_to_pfn(mem->start_section_nr + i); - ret &= is_mem_section_removable(pfn, PAGES_PER_SECTION); - } - -out: - return sprintf(buf, "%d\n", ret); + return sprintf(buf, "%d\n", (int)IS_ENABLED(CONFIG_MEMORY_HOTREMOVE)); } /* diff --git a/drivers/base/platform.c b/drivers/base/platform.c index 7fa654f1288b..b5ce7b085795 100644 --- a/drivers/base/platform.c +++ b/drivers/base/platform.c @@ -363,10 +363,10 @@ static void setup_pdev_dma_masks(struct platform_device *pdev) { if (!pdev->dev.coherent_dma_mask) pdev->dev.coherent_dma_mask = DMA_BIT_MASK(32); - if (!pdev->dma_mask) - pdev->dma_mask = DMA_BIT_MASK(32); - if (!pdev->dev.dma_mask) - pdev->dev.dma_mask = &pdev->dma_mask; + if (!pdev->dev.dma_mask) { + pdev->platform_dma_mask = DMA_BIT_MASK(32); + pdev->dev.dma_mask = &pdev->platform_dma_mask; + } }; /** @@ -662,20 +662,8 @@ struct platform_device *platform_device_register_full( pdev->dev.of_node_reused = pdevinfo->of_node_reused; if (pdevinfo->dma_mask) { - /* - * This memory isn't freed when the device is put, - * I don't have a nice idea for that though. Conceptually - * dma_mask in struct device should not be a pointer. - * See http://thread.gmane.org/gmane.linux.kernel.pci/9081 - */ - pdev->dev.dma_mask = - kmalloc(sizeof(*pdev->dev.dma_mask), GFP_KERNEL); - if (!pdev->dev.dma_mask) - goto err; - - kmemleak_ignore(pdev->dev.dma_mask); - - *pdev->dev.dma_mask = pdevinfo->dma_mask; + pdev->platform_dma_mask = pdevinfo->dma_mask; + pdev->dev.dma_mask = &pdev->platform_dma_mask; pdev->dev.coherent_dma_mask = pdevinfo->dma_mask; } @@ -700,7 +688,6 @@ struct platform_device *platform_device_register_full( if (ret) { err: ACPI_COMPANION_SET(&pdev->dev, NULL); - kfree(pdev->dev.dma_mask); platform_device_put(pdev); return ERR_PTR(ret); } diff --git a/drivers/base/swnode.c b/drivers/base/swnode.c index 0b081dee1e95..de8d3543e8fe 100644 --- a/drivers/base/swnode.c +++ b/drivers/base/swnode.c @@ -608,6 +608,13 @@ static void software_node_release(struct kobject *kobj) { struct swnode *swnode = kobj_to_swnode(kobj); + if (swnode->parent) { + ida_simple_remove(&swnode->parent->child_ids, swnode->id); + list_del(&swnode->entry); + } else { + ida_simple_remove(&swnode_root_ids, swnode->id); + } + if (swnode->allocated) { property_entries_free(swnode->node->properties); kfree(swnode->node); @@ -773,13 +780,6 @@ void fwnode_remove_software_node(struct fwnode_handle *fwnode) if (!swnode) return; - if (swnode->parent) { - ida_simple_remove(&swnode->parent->child_ids, swnode->id); - list_del(&swnode->entry); - } else { - ida_simple_remove(&swnode_root_ids, swnode->id); - } - kobject_put(&swnode->kobj); } EXPORT_SYMBOL_GPL(fwnode_remove_software_node); diff --git a/drivers/block/floppy.c b/drivers/block/floppy.c index cd3612e4e2e1..8ef65c085640 100644 --- a/drivers/block/floppy.c +++ b/drivers/block/floppy.c @@ -853,14 +853,17 @@ static void reset_fdc_info(int mode) /* selects the fdc and drive, and enables the fdc's input/dma. */ static void set_fdc(int drive) { + unsigned int new_fdc = fdc; + if (drive >= 0 && drive < N_DRIVE) { - fdc = FDC(drive); + new_fdc = FDC(drive); current_drive = drive; } - if (fdc != 1 && fdc != 0) { + if (new_fdc >= N_FDC) { pr_info("bad fdc value\n"); return; } + fdc = new_fdc; set_dor(fdc, ~0, 8); #if N_FDC > 1 set_dor(1 - fdc, ~8, 0); diff --git a/drivers/block/null_blk.h b/drivers/block/null_blk.h index bc837862b767..62b660821dbc 100644 --- a/drivers/block/null_blk.h +++ b/drivers/block/null_blk.h @@ -14,9 +14,6 @@ #include <linux/fault-inject.h> struct nullb_cmd { - struct list_head list; - struct llist_node ll_list; - struct __call_single_data csd; struct request *rq; struct bio *bio; unsigned int tag; diff --git a/drivers/block/null_blk_main.c b/drivers/block/null_blk_main.c index 16510795e377..133060431dbd 100644 --- a/drivers/block/null_blk_main.c +++ b/drivers/block/null_blk_main.c @@ -1518,8 +1518,6 @@ static int setup_commands(struct nullb_queue *nq) for (i = 0; i < nq->queue_depth; i++) { cmd = &nq->cmds[i]; - INIT_LIST_HEAD(&cmd->list); - cmd->ll_list.next = NULL; cmd->tag = -1U; } diff --git a/drivers/block/paride/pcd.c b/drivers/block/paride/pcd.c index 117cfc8cd05a..cda5cf917e9a 100644 --- a/drivers/block/paride/pcd.c +++ b/drivers/block/paride/pcd.c @@ -276,7 +276,7 @@ static const struct block_device_operations pcd_bdops = { .release = pcd_block_release, .ioctl = pcd_block_ioctl, #ifdef CONFIG_COMPAT - .ioctl = blkdev_compat_ptr_ioctl, + .compat_ioctl = blkdev_compat_ptr_ioctl, #endif .check_events = pcd_block_check_events, }; diff --git a/drivers/block/virtio_blk.c b/drivers/block/virtio_blk.c index 54158766334b..0736248999b0 100644 --- a/drivers/block/virtio_blk.c +++ b/drivers/block/virtio_blk.c @@ -245,13 +245,20 @@ static blk_status_t virtio_queue_rq(struct blk_mq_hw_ctx *hctx, err = virtblk_add_req(vblk->vqs[qid].vq, vbr, vbr->sg, num); if (err) { virtqueue_kick(vblk->vqs[qid].vq); - blk_mq_stop_hw_queue(hctx); + /* Don't stop the queue if -ENOMEM: we may have failed to + * bounce the buffer due to global resource outage. + */ + if (err == -ENOSPC) + blk_mq_stop_hw_queue(hctx); spin_unlock_irqrestore(&vblk->vqs[qid].lock, flags); - /* Out of mem doesn't actually happen, since we fall back - * to direct descriptors */ - if (err == -ENOMEM || err == -ENOSPC) + switch (err) { + case -ENOSPC: return BLK_STS_DEV_RESOURCE; - return BLK_STS_IOERR; + case -ENOMEM: + return BLK_STS_RESOURCE; + default: + return BLK_STS_IOERR; + } } if (bd->last && virtqueue_kick_prepare(vblk->vqs[qid].vq)) diff --git a/drivers/block/xen-blkfront.c b/drivers/block/xen-blkfront.c index e2ad6bba2281..9df516a56bb2 100644 --- a/drivers/block/xen-blkfront.c +++ b/drivers/block/xen-blkfront.c @@ -213,6 +213,7 @@ struct blkfront_info struct blk_mq_tag_set tag_set; struct blkfront_ring_info *rinfo; unsigned int nr_rings; + unsigned int rinfo_size; /* Save uncomplete reqs and bios for migration. */ struct list_head requests; struct bio_list bio_list; @@ -259,6 +260,18 @@ static int blkfront_setup_indirect(struct blkfront_ring_info *rinfo); static void blkfront_gather_backend_features(struct blkfront_info *info); static int negotiate_mq(struct blkfront_info *info); +#define for_each_rinfo(info, ptr, idx) \ + for ((ptr) = (info)->rinfo, (idx) = 0; \ + (idx) < (info)->nr_rings; \ + (idx)++, (ptr) = (void *)(ptr) + (info)->rinfo_size) + +static inline struct blkfront_ring_info * +get_rinfo(const struct blkfront_info *info, unsigned int i) +{ + BUG_ON(i >= info->nr_rings); + return (void *)info->rinfo + i * info->rinfo_size; +} + static int get_id_from_freelist(struct blkfront_ring_info *rinfo) { unsigned long free = rinfo->shadow_free; @@ -883,8 +896,7 @@ static blk_status_t blkif_queue_rq(struct blk_mq_hw_ctx *hctx, struct blkfront_info *info = hctx->queue->queuedata; struct blkfront_ring_info *rinfo = NULL; - BUG_ON(info->nr_rings <= qid); - rinfo = &info->rinfo[qid]; + rinfo = get_rinfo(info, qid); blk_mq_start_request(qd->rq); spin_lock_irqsave(&rinfo->ring_lock, flags); if (RING_FULL(&rinfo->ring)) @@ -1181,6 +1193,7 @@ static int xlvbd_alloc_gendisk(blkif_sector_t capacity, static void xlvbd_release_gendisk(struct blkfront_info *info) { unsigned int minor, nr_minors, i; + struct blkfront_ring_info *rinfo; if (info->rq == NULL) return; @@ -1188,9 +1201,7 @@ static void xlvbd_release_gendisk(struct blkfront_info *info) /* No more blkif_request(). */ blk_mq_stop_hw_queues(info->rq); - for (i = 0; i < info->nr_rings; i++) { - struct blkfront_ring_info *rinfo = &info->rinfo[i]; - + for_each_rinfo(info, rinfo, i) { /* No more gnttab callback work. */ gnttab_cancel_free_callback(&rinfo->callback); @@ -1339,6 +1350,7 @@ free_shadow: static void blkif_free(struct blkfront_info *info, int suspend) { unsigned int i; + struct blkfront_ring_info *rinfo; /* Prevent new requests being issued until we fix things up. */ info->connected = suspend ? @@ -1347,8 +1359,8 @@ static void blkif_free(struct blkfront_info *info, int suspend) if (info->rq) blk_mq_stop_hw_queues(info->rq); - for (i = 0; i < info->nr_rings; i++) - blkif_free_ring(&info->rinfo[i]); + for_each_rinfo(info, rinfo, i) + blkif_free_ring(rinfo); kvfree(info->rinfo); info->rinfo = NULL; @@ -1775,6 +1787,7 @@ static int talk_to_blkback(struct xenbus_device *dev, int err; unsigned int i, max_page_order; unsigned int ring_page_order; + struct blkfront_ring_info *rinfo; if (!info) return -ENODEV; @@ -1788,9 +1801,7 @@ static int talk_to_blkback(struct xenbus_device *dev, if (err) goto destroy_blkring; - for (i = 0; i < info->nr_rings; i++) { - struct blkfront_ring_info *rinfo = &info->rinfo[i]; - + for_each_rinfo(info, rinfo, i) { /* Create shared ring, alloc event channel. */ err = setup_blkring(dev, rinfo); if (err) @@ -1815,7 +1826,7 @@ again: /* We already got the number of queues/rings in _probe */ if (info->nr_rings == 1) { - err = write_per_ring_nodes(xbt, &info->rinfo[0], dev->nodename); + err = write_per_ring_nodes(xbt, info->rinfo, dev->nodename); if (err) goto destroy_blkring; } else { @@ -1837,10 +1848,10 @@ again: goto abort_transaction; } - for (i = 0; i < info->nr_rings; i++) { + for_each_rinfo(info, rinfo, i) { memset(path, 0, pathsize); snprintf(path, pathsize, "%s/queue-%u", dev->nodename, i); - err = write_per_ring_nodes(xbt, &info->rinfo[i], path); + err = write_per_ring_nodes(xbt, rinfo, path); if (err) { kfree(path); goto destroy_blkring; @@ -1868,9 +1879,8 @@ again: goto destroy_blkring; } - for (i = 0; i < info->nr_rings; i++) { + for_each_rinfo(info, rinfo, i) { unsigned int j; - struct blkfront_ring_info *rinfo = &info->rinfo[i]; for (j = 0; j < BLK_RING_SIZE(info); j++) rinfo->shadow[j].req.u.rw.id = j + 1; @@ -1900,6 +1910,7 @@ static int negotiate_mq(struct blkfront_info *info) { unsigned int backend_max_queues; unsigned int i; + struct blkfront_ring_info *rinfo; BUG_ON(info->nr_rings); @@ -1911,20 +1922,16 @@ static int negotiate_mq(struct blkfront_info *info) if (!info->nr_rings) info->nr_rings = 1; - info->rinfo = kvcalloc(info->nr_rings, - struct_size(info->rinfo, shadow, - BLK_RING_SIZE(info)), - GFP_KERNEL); + info->rinfo_size = struct_size(info->rinfo, shadow, + BLK_RING_SIZE(info)); + info->rinfo = kvcalloc(info->nr_rings, info->rinfo_size, GFP_KERNEL); if (!info->rinfo) { xenbus_dev_fatal(info->xbdev, -ENOMEM, "allocating ring_info structure"); info->nr_rings = 0; return -ENOMEM; } - for (i = 0; i < info->nr_rings; i++) { - struct blkfront_ring_info *rinfo; - - rinfo = &info->rinfo[i]; + for_each_rinfo(info, rinfo, i) { INIT_LIST_HEAD(&rinfo->indirect_pages); INIT_LIST_HEAD(&rinfo->grants); rinfo->dev_info = info; @@ -2017,6 +2024,7 @@ static int blkif_recover(struct blkfront_info *info) int rc; struct bio *bio; unsigned int segs; + struct blkfront_ring_info *rinfo; blkfront_gather_backend_features(info); /* Reset limits changed by blk_mq_update_nr_hw_queues(). */ @@ -2024,9 +2032,7 @@ static int blkif_recover(struct blkfront_info *info) segs = info->max_indirect_segments ? : BLKIF_MAX_SEGMENTS_PER_REQUEST; blk_queue_max_segments(info->rq, segs / GRANTS_PER_PSEG); - for (r_index = 0; r_index < info->nr_rings; r_index++) { - struct blkfront_ring_info *rinfo = &info->rinfo[r_index]; - + for_each_rinfo(info, rinfo, r_index) { rc = blkfront_setup_indirect(rinfo); if (rc) return rc; @@ -2036,10 +2042,7 @@ static int blkif_recover(struct blkfront_info *info) /* Now safe for us to use the shared ring */ info->connected = BLKIF_STATE_CONNECTED; - for (r_index = 0; r_index < info->nr_rings; r_index++) { - struct blkfront_ring_info *rinfo; - - rinfo = &info->rinfo[r_index]; + for_each_rinfo(info, rinfo, r_index) { /* Kick any other new requests queued since we resumed */ kick_pending_request_queues(rinfo); } @@ -2072,13 +2075,13 @@ static int blkfront_resume(struct xenbus_device *dev) struct blkfront_info *info = dev_get_drvdata(&dev->dev); int err = 0; unsigned int i, j; + struct blkfront_ring_info *rinfo; dev_dbg(&dev->dev, "blkfront_resume: %s\n", dev->nodename); bio_list_init(&info->bio_list); INIT_LIST_HEAD(&info->requests); - for (i = 0; i < info->nr_rings; i++) { - struct blkfront_ring_info *rinfo = &info->rinfo[i]; + for_each_rinfo(info, rinfo, i) { struct bio_list merge_bio; struct blk_shadow *shadow = rinfo->shadow; @@ -2337,6 +2340,7 @@ static void blkfront_connect(struct blkfront_info *info) unsigned int binfo; char *envp[] = { "RESIZE=1", NULL }; int err, i; + struct blkfront_ring_info *rinfo; switch (info->connected) { case BLKIF_STATE_CONNECTED: @@ -2394,8 +2398,8 @@ static void blkfront_connect(struct blkfront_info *info) "physical-sector-size", sector_size); blkfront_gather_backend_features(info); - for (i = 0; i < info->nr_rings; i++) { - err = blkfront_setup_indirect(&info->rinfo[i]); + for_each_rinfo(info, rinfo, i) { + err = blkfront_setup_indirect(rinfo); if (err) { xenbus_dev_fatal(info->xbdev, err, "setup_indirect at %s", info->xbdev->otherend); @@ -2416,8 +2420,8 @@ static void blkfront_connect(struct blkfront_info *info) /* Kick pending requests. */ info->connected = BLKIF_STATE_CONNECTED; - for (i = 0; i < info->nr_rings; i++) - kick_pending_request_queues(&info->rinfo[i]); + for_each_rinfo(info, rinfo, i) + kick_pending_request_queues(rinfo); device_add_disk(&info->xbdev->dev, info->gd, NULL); @@ -2652,9 +2656,9 @@ static void purge_persistent_grants(struct blkfront_info *info) { unsigned int i; unsigned long flags; + struct blkfront_ring_info *rinfo; - for (i = 0; i < info->nr_rings; i++) { - struct blkfront_ring_info *rinfo = &info->rinfo[i]; + for_each_rinfo(info, rinfo, i) { struct grant *gnt_list_entry, *tmp; spin_lock_irqsave(&rinfo->ring_lock, flags); diff --git a/drivers/bus/sunxi-rsb.c b/drivers/bus/sunxi-rsb.c index be79d6c6a4e4..1bb00a959c67 100644 --- a/drivers/bus/sunxi-rsb.c +++ b/drivers/bus/sunxi-rsb.c @@ -345,7 +345,7 @@ static int sunxi_rsb_read(struct sunxi_rsb *rsb, u8 rtaddr, u8 addr, if (ret) goto unlock; - *buf = readl(rsb->regs + RSB_DATA); + *buf = readl(rsb->regs + RSB_DATA) & GENMASK(len * 8 - 1, 0); unlock: mutex_unlock(&rsb->lock); diff --git a/drivers/bus/ti-sysc.c b/drivers/bus/ti-sysc.c index f702c85c81b6..440019655fbb 100644 --- a/drivers/bus/ti-sysc.c +++ b/drivers/bus/ti-sysc.c @@ -1266,6 +1266,8 @@ static const struct sysc_revision_quirk sysc_revision_quirks[] = { SYSC_QUIRK("gpu", 0x50000000, 0x14, -1, -1, 0x00010201, 0xffffffff, 0), SYSC_QUIRK("gpu", 0x50000000, 0xfe00, 0xfe10, -1, 0x40000000 , 0xffffffff, SYSC_MODULE_QUIRK_SGX), + SYSC_QUIRK("lcdc", 0, 0, 0x54, -1, 0x4f201000, 0xffffffff, + SYSC_QUIRK_SWSUP_SIDLE | SYSC_QUIRK_SWSUP_MSTANDBY), SYSC_QUIRK("usb_otg_hs", 0, 0x400, 0x404, 0x408, 0x00000050, 0xffffffff, SYSC_QUIRK_SWSUP_SIDLE | SYSC_QUIRK_SWSUP_MSTANDBY), SYSC_QUIRK("usb_otg_hs", 0, 0, 0x10, -1, 0x4ea2080d, 0xffffffff, @@ -1294,7 +1296,6 @@ static const struct sysc_revision_quirk sysc_revision_quirks[] = { SYSC_QUIRK("gpu", 0, 0xfe00, 0xfe10, -1, 0x40000000 , 0xffffffff, 0), SYSC_QUIRK("hsi", 0, 0, 0x10, 0x14, 0x50043101, 0xffffffff, 0), SYSC_QUIRK("iss", 0, 0, 0x10, -1, 0x40000101, 0xffffffff, 0), - SYSC_QUIRK("lcdc", 0, 0, 0x54, -1, 0x4f201000, 0xffffffff, 0), SYSC_QUIRK("mcasp", 0, 0, 0x4, -1, 0x44306302, 0xffffffff, 0), SYSC_QUIRK("mcasp", 0, 0, 0x4, -1, 0x44307b02, 0xffffffff, 0), SYSC_QUIRK("mcbsp", 0, -1, 0x8c, -1, 0, 0, 0), @@ -1400,7 +1401,7 @@ static void sysc_init_revision_quirks(struct sysc *ddata) } /* 1-wire needs module's internal clocks enabled for reset */ -static void sysc_clk_enable_quirk_hdq1w(struct sysc *ddata) +static void sysc_pre_reset_quirk_hdq1w(struct sysc *ddata) { int offset = 0x0c; /* HDQ_CTRL_STATUS */ u16 val; @@ -1488,7 +1489,7 @@ static void sysc_init_module_quirks(struct sysc *ddata) return; if (ddata->cfg.quirks & SYSC_MODULE_QUIRK_HDQ1W) { - ddata->clk_enable_quirk = sysc_clk_enable_quirk_hdq1w; + ddata->clk_disable_quirk = sysc_pre_reset_quirk_hdq1w; return; } diff --git a/drivers/cdrom/gdrom.c b/drivers/cdrom/gdrom.c index 886b2638c730..c51292c2a131 100644 --- a/drivers/cdrom/gdrom.c +++ b/drivers/cdrom/gdrom.c @@ -519,7 +519,7 @@ static const struct block_device_operations gdrom_bdops = { .check_events = gdrom_bdops_check_events, .ioctl = gdrom_bdops_ioctl, #ifdef CONFIG_COMPAT - .ioctl = blkdev_compat_ptr_ioctl, + .compat_ioctl = blkdev_compat_ptr_ioctl, #endif }; diff --git a/drivers/char/ipmi/ipmi_si_platform.c b/drivers/char/ipmi/ipmi_si_platform.c index c78127ccbc0d..638c693e17ad 100644 --- a/drivers/char/ipmi/ipmi_si_platform.c +++ b/drivers/char/ipmi/ipmi_si_platform.c @@ -194,7 +194,7 @@ static int platform_ipmi_probe(struct platform_device *pdev) else io.slave_addr = slave_addr; - io.irq = platform_get_irq(pdev, 0); + io.irq = platform_get_irq_optional(pdev, 0); if (io.irq > 0) io.irq_setup = ipmi_std_irq_setup; else @@ -378,7 +378,7 @@ static int acpi_ipmi_probe(struct platform_device *pdev) io.irq = tmp; io.irq_setup = acpi_gpe_irq_setup; } else { - int irq = platform_get_irq(pdev, 0); + int irq = platform_get_irq_optional(pdev, 0); if (irq > 0) { io.irq = irq; diff --git a/drivers/char/tpm/Makefile b/drivers/char/tpm/Makefile index 5a0d99d4fec0..9567e5197f74 100644 --- a/drivers/char/tpm/Makefile +++ b/drivers/char/tpm/Makefile @@ -21,9 +21,11 @@ tpm-$(CONFIG_EFI) += eventlog/efi.o tpm-$(CONFIG_OF) += eventlog/of.o obj-$(CONFIG_TCG_TIS_CORE) += tpm_tis_core.o obj-$(CONFIG_TCG_TIS) += tpm_tis.o -obj-$(CONFIG_TCG_TIS_SPI) += tpm_tis_spi_mod.o -tpm_tis_spi_mod-y := tpm_tis_spi.o -tpm_tis_spi_mod-$(CONFIG_TCG_TIS_SPI_CR50) += tpm_tis_spi_cr50.o + +obj-$(CONFIG_TCG_TIS_SPI) += tpm_tis_spi.o +tpm_tis_spi-y := tpm_tis_spi_main.o +tpm_tis_spi-$(CONFIG_TCG_TIS_SPI_CR50) += tpm_tis_spi_cr50.o + obj-$(CONFIG_TCG_TIS_I2C_ATMEL) += tpm_i2c_atmel.o obj-$(CONFIG_TCG_TIS_I2C_INFINEON) += tpm_i2c_infineon.o obj-$(CONFIG_TCG_TIS_I2C_NUVOTON) += tpm_i2c_nuvoton.o diff --git a/drivers/char/tpm/tpm2-cmd.c b/drivers/char/tpm/tpm2-cmd.c index 13696deceae8..760329598b99 100644 --- a/drivers/char/tpm/tpm2-cmd.c +++ b/drivers/char/tpm/tpm2-cmd.c @@ -525,6 +525,8 @@ static int tpm2_init_bank_info(struct tpm_chip *chip, u32 bank_index) return 0; } + bank->crypto_id = HASH_ALGO__LAST; + return tpm2_pcr_read(chip, 0, &digest, &bank->digest_size); } diff --git a/drivers/char/tpm/tpm_tis_spi.c b/drivers/char/tpm/tpm_tis_spi_main.c index d1754fd6c573..d1754fd6c573 100644 --- a/drivers/char/tpm/tpm_tis_spi.c +++ b/drivers/char/tpm/tpm_tis_spi_main.c diff --git a/drivers/clk/clk.c b/drivers/clk/clk.c index f0f2b599fd7e..95adf6c6db3d 100644 --- a/drivers/clk/clk.c +++ b/drivers/clk/clk.c @@ -4713,7 +4713,7 @@ EXPORT_SYMBOL(of_clk_get_by_name); * * Returns: The number of clocks that are possible parents of this node */ -unsigned int of_clk_get_parent_count(struct device_node *np) +unsigned int of_clk_get_parent_count(const struct device_node *np) { int count; @@ -4725,7 +4725,7 @@ unsigned int of_clk_get_parent_count(struct device_node *np) } EXPORT_SYMBOL_GPL(of_clk_get_parent_count); -const char *of_clk_get_parent_name(struct device_node *np, int index) +const char *of_clk_get_parent_name(const struct device_node *np, int index) { struct of_phandle_args clkspec; struct property *prop; diff --git a/drivers/clk/imx/clk-imx8mp.c b/drivers/clk/imx/clk-imx8mp.c index f6c120cca0d4..cf192907b7dc 100644 --- a/drivers/clk/imx/clk-imx8mp.c +++ b/drivers/clk/imx/clk-imx8mp.c @@ -560,7 +560,7 @@ static int imx8mp_clocks_probe(struct platform_device *pdev) hws[IMX8MP_CLK_MEDIA_AXI] = imx8m_clk_hw_composite("media_axi", imx8mp_media_axi_sels, ccm_base + 0x8a00); hws[IMX8MP_CLK_MEDIA_APB] = imx8m_clk_hw_composite("media_apb", imx8mp_media_apb_sels, ccm_base + 0x8a80); hws[IMX8MP_CLK_HDMI_APB] = imx8m_clk_hw_composite("hdmi_apb", imx8mp_media_apb_sels, ccm_base + 0x8b00); - hws[IMX8MP_CLK_HDMI_AXI] = imx8m_clk_hw_composite("hdmi_axi", imx8mp_media_apb_sels, ccm_base + 0x8b80); + hws[IMX8MP_CLK_HDMI_AXI] = imx8m_clk_hw_composite("hdmi_axi", imx8mp_media_axi_sels, ccm_base + 0x8b80); hws[IMX8MP_CLK_GPU_AXI] = imx8m_clk_hw_composite("gpu_axi", imx8mp_gpu_axi_sels, ccm_base + 0x8c00); hws[IMX8MP_CLK_GPU_AHB] = imx8m_clk_hw_composite("gpu_ahb", imx8mp_gpu_ahb_sels, ccm_base + 0x8c80); hws[IMX8MP_CLK_NOC] = imx8m_clk_hw_composite_critical("noc", imx8mp_noc_sels, ccm_base + 0x8d00); @@ -686,7 +686,7 @@ static int imx8mp_clocks_probe(struct platform_device *pdev) hws[IMX8MP_CLK_CAN1_ROOT] = imx_clk_hw_gate2("can1_root_clk", "can1", ccm_base + 0x4350, 0); hws[IMX8MP_CLK_CAN2_ROOT] = imx_clk_hw_gate2("can2_root_clk", "can2", ccm_base + 0x4360, 0); hws[IMX8MP_CLK_SDMA1_ROOT] = imx_clk_hw_gate4("sdma1_root_clk", "ipg_root", ccm_base + 0x43a0, 0); - hws[IMX8MP_CLK_ENET_QOS_ROOT] = imx_clk_hw_gate4("enet_qos_root_clk", "enet_axi", ccm_base + 0x43b0, 0); + hws[IMX8MP_CLK_ENET_QOS_ROOT] = imx_clk_hw_gate4("enet_qos_root_clk", "sim_enet_root_clk", ccm_base + 0x43b0, 0); hws[IMX8MP_CLK_SIM_ENET_ROOT] = imx_clk_hw_gate4("sim_enet_root_clk", "enet_axi", ccm_base + 0x4400, 0); hws[IMX8MP_CLK_GPU2D_ROOT] = imx_clk_hw_gate4("gpu2d_root_clk", "gpu2d_div", ccm_base + 0x4450, 0); hws[IMX8MP_CLK_GPU3D_ROOT] = imx_clk_hw_gate4("gpu3d_root_clk", "gpu3d_core_div", ccm_base + 0x4460, 0); diff --git a/drivers/clk/imx/clk-scu.c b/drivers/clk/imx/clk-scu.c index fbef740704d0..b8b2072742a5 100644 --- a/drivers/clk/imx/clk-scu.c +++ b/drivers/clk/imx/clk-scu.c @@ -43,12 +43,12 @@ struct imx_sc_msg_req_set_clock_rate { __le32 rate; __le16 resource; u8 clk; -} __packed; +} __packed __aligned(4); struct req_get_clock_rate { __le16 resource; u8 clk; -} __packed; +} __packed __aligned(4); struct resp_get_clock_rate { __le32 rate; @@ -84,7 +84,7 @@ struct imx_sc_msg_get_clock_parent { struct req_get_clock_parent { __le16 resource; u8 clk; - } __packed req; + } __packed __aligned(4) req; struct resp_get_clock_parent { u8 parent; } resp; @@ -121,7 +121,7 @@ struct imx_sc_msg_req_clock_enable { u8 clk; u8 enable; u8 autog; -} __packed; +} __packed __aligned(4); static inline struct clk_scu *to_clk_scu(struct clk_hw *hw) { diff --git a/drivers/clk/qcom/dispcc-sc7180.c b/drivers/clk/qcom/dispcc-sc7180.c index dd7af41e47eb..0a5d395bce93 100644 --- a/drivers/clk/qcom/dispcc-sc7180.c +++ b/drivers/clk/qcom/dispcc-sc7180.c @@ -592,24 +592,6 @@ static struct clk_branch disp_cc_mdss_rot_clk = { }, }; -static struct clk_branch disp_cc_mdss_rscc_ahb_clk = { - .halt_reg = 0x400c, - .halt_check = BRANCH_HALT, - .clkr = { - .enable_reg = 0x400c, - .enable_mask = BIT(0), - .hw.init = &(struct clk_init_data){ - .name = "disp_cc_mdss_rscc_ahb_clk", - .parent_data = &(const struct clk_parent_data){ - .hw = &disp_cc_mdss_ahb_clk_src.clkr.hw, - }, - .num_parents = 1, - .flags = CLK_IS_CRITICAL | CLK_SET_RATE_PARENT, - .ops = &clk_branch2_ops, - }, - }, -}; - static struct clk_branch disp_cc_mdss_rscc_vsync_clk = { .halt_reg = 0x4008, .halt_check = BRANCH_HALT, @@ -687,7 +669,6 @@ static struct clk_regmap *disp_cc_sc7180_clocks[] = { [DISP_CC_MDSS_PCLK0_CLK_SRC] = &disp_cc_mdss_pclk0_clk_src.clkr, [DISP_CC_MDSS_ROT_CLK] = &disp_cc_mdss_rot_clk.clkr, [DISP_CC_MDSS_ROT_CLK_SRC] = &disp_cc_mdss_rot_clk_src.clkr, - [DISP_CC_MDSS_RSCC_AHB_CLK] = &disp_cc_mdss_rscc_ahb_clk.clkr, [DISP_CC_MDSS_RSCC_VSYNC_CLK] = &disp_cc_mdss_rscc_vsync_clk.clkr, [DISP_CC_MDSS_VSYNC_CLK] = &disp_cc_mdss_vsync_clk.clkr, [DISP_CC_MDSS_VSYNC_CLK_SRC] = &disp_cc_mdss_vsync_clk_src.clkr, diff --git a/drivers/clk/qcom/videocc-sc7180.c b/drivers/clk/qcom/videocc-sc7180.c index c363c3cc544e..276e5ecd4840 100644 --- a/drivers/clk/qcom/videocc-sc7180.c +++ b/drivers/clk/qcom/videocc-sc7180.c @@ -97,7 +97,7 @@ static struct clk_branch video_cc_vcodec0_axi_clk = { static struct clk_branch video_cc_vcodec0_core_clk = { .halt_reg = 0x890, - .halt_check = BRANCH_HALT, + .halt_check = BRANCH_HALT_VOTED, .clkr = { .enable_reg = 0x890, .enable_mask = BIT(0), diff --git a/drivers/clk/ti/clk-43xx.c b/drivers/clk/ti/clk-43xx.c index af3e7805769e..e5538d577ce5 100644 --- a/drivers/clk/ti/clk-43xx.c +++ b/drivers/clk/ti/clk-43xx.c @@ -78,7 +78,7 @@ static const struct omap_clkctrl_reg_data am4_gfx_l3_clkctrl_regs[] __initconst }; static const struct omap_clkctrl_reg_data am4_l4_rtc_clkctrl_regs[] __initconst = { - { AM4_L4_RTC_RTC_CLKCTRL, NULL, CLKF_SW_SUP, "clk_32768_ck" }, + { AM4_L4_RTC_RTC_CLKCTRL, NULL, CLKF_SW_SUP, "clkdiv32k_ick" }, { 0 }, }; diff --git a/drivers/clocksource/hyperv_timer.c b/drivers/clocksource/hyperv_timer.c index 9d808d595ca8..eb0ba7818eb0 100644 --- a/drivers/clocksource/hyperv_timer.c +++ b/drivers/clocksource/hyperv_timer.c @@ -343,7 +343,8 @@ static u64 notrace read_hv_clock_tsc_cs(struct clocksource *arg) static u64 read_hv_sched_clock_tsc(void) { - return read_hv_clock_tsc() - hv_sched_clock_offset; + return (read_hv_clock_tsc() - hv_sched_clock_offset) * + (NSEC_PER_SEC / HV_CLOCK_HZ); } static void suspend_hv_clock_tsc(struct clocksource *arg) @@ -398,7 +399,8 @@ static u64 notrace read_hv_clock_msr_cs(struct clocksource *arg) static u64 read_hv_sched_clock_msr(void) { - return read_hv_clock_msr() - hv_sched_clock_offset; + return (read_hv_clock_msr() - hv_sched_clock_offset) * + (NSEC_PER_SEC / HV_CLOCK_HZ); } static struct clocksource hyperv_cs_msr = { diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c index cbe6c94bf158..808874bccf4a 100644 --- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -1076,9 +1076,17 @@ static int cpufreq_init_policy(struct cpufreq_policy *policy) pol = policy->last_policy; } else if (def_gov) { pol = cpufreq_parse_policy(def_gov->name); - } else { - return -ENODATA; + /* + * In case the default governor is neiter "performance" + * nor "powersave", fall back to the initial policy + * value set by the driver. + */ + if (pol == CPUFREQ_POLICY_UNKNOWN) + pol = policy->policy; } + if (pol != CPUFREQ_POLICY_PERFORMANCE && + pol != CPUFREQ_POLICY_POWERSAVE) + return -ENODATA; } return cpufreq_set_policy(policy, gov, pol); diff --git a/drivers/devfreq/devfreq.c b/drivers/devfreq/devfreq.c index cceee8bc3c2f..7dcf2093e531 100644 --- a/drivers/devfreq/devfreq.c +++ b/drivers/devfreq/devfreq.c @@ -738,7 +738,6 @@ struct devfreq *devfreq_add_device(struct device *dev, { struct devfreq *devfreq; struct devfreq_governor *governor; - static atomic_t devfreq_no = ATOMIC_INIT(-1); int err = 0; if (!dev || !profile || !governor_name) { @@ -800,8 +799,7 @@ struct devfreq *devfreq_add_device(struct device *dev, devfreq->suspend_freq = dev_pm_opp_get_suspend_opp_freq(dev); atomic_set(&devfreq->suspend_count, 0); - dev_set_name(&devfreq->dev, "devfreq%d", - atomic_inc_return(&devfreq_no)); + dev_set_name(&devfreq->dev, "%s", dev_name(dev)); err = device_register(&devfreq->dev); if (err) { mutex_unlock(&devfreq->lock); diff --git a/drivers/dma-buf/dma-buf.c b/drivers/dma-buf/dma-buf.c index f4ace9af2191..ccc9eda1bc28 100644 --- a/drivers/dma-buf/dma-buf.c +++ b/drivers/dma-buf/dma-buf.c @@ -108,6 +108,7 @@ static int dma_buf_release(struct inode *inode, struct file *file) dma_resv_fini(dmabuf->resv); module_put(dmabuf->owner); + kfree(dmabuf->name); kfree(dmabuf); return 0; } diff --git a/drivers/dma/coh901318.c b/drivers/dma/coh901318.c index e51d836afcc7..1092d4ce723e 100644 --- a/drivers/dma/coh901318.c +++ b/drivers/dma/coh901318.c @@ -1947,8 +1947,6 @@ static void dma_tc_handle(struct coh901318_chan *cohc) return; } - spin_lock(&cohc->lock); - /* * When we reach this point, at least one queue item * should have been moved over from cohc->queue to @@ -1969,8 +1967,6 @@ static void dma_tc_handle(struct coh901318_chan *cohc) if (coh901318_queue_start(cohc) == NULL) cohc->busy = 0; - spin_unlock(&cohc->lock); - /* * This tasklet will remove items from cohc->active * and thus terminates them. diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c index c3b1283b6d31..17909fd1820f 100644 --- a/drivers/dma/dmaengine.c +++ b/drivers/dma/dmaengine.c @@ -1151,7 +1151,7 @@ int dma_async_device_register(struct dma_device *device) } if (!device->device_release) - dev_warn(device->dev, + dev_dbg(device->dev, "WARN: Device release is not defined so it is not safe to unbind this driver while in use\n"); kref_init(&device->ref); diff --git a/drivers/dma/idxd/cdev.c b/drivers/dma/idxd/cdev.c index 1d7347825b95..989b7a25ca61 100644 --- a/drivers/dma/idxd/cdev.c +++ b/drivers/dma/idxd/cdev.c @@ -81,9 +81,9 @@ static int idxd_cdev_open(struct inode *inode, struct file *filp) dev = &idxd->pdev->dev; idxd_cdev = &wq->idxd_cdev; - dev_dbg(dev, "%s called\n", __func__); + dev_dbg(dev, "%s called: %d\n", __func__, idxd_wq_refcount(wq)); - if (idxd_wq_refcount(wq) > 1 && wq_dedicated(wq)) + if (idxd_wq_refcount(wq) > 0 && wq_dedicated(wq)) return -EBUSY; ctx = kzalloc(sizeof(*ctx), GFP_KERNEL); @@ -204,6 +204,7 @@ static int idxd_wq_cdev_dev_setup(struct idxd_wq *wq) minor = ida_simple_get(&cdev_ctx->minor_ida, 0, MINORMASK, GFP_KERNEL); if (minor < 0) { rc = minor; + kfree(dev); goto ida_err; } @@ -212,7 +213,6 @@ static int idxd_wq_cdev_dev_setup(struct idxd_wq *wq) rc = device_register(dev); if (rc < 0) { dev_err(&idxd->pdev->dev, "device register failed\n"); - put_device(dev); goto dev_reg_err; } idxd_cdev->minor = minor; @@ -221,8 +221,8 @@ static int idxd_wq_cdev_dev_setup(struct idxd_wq *wq) dev_reg_err: ida_simple_remove(&cdev_ctx->minor_ida, MINOR(dev->devt)); + put_device(dev); ida_err: - kfree(dev); idxd_cdev->dev = NULL; return rc; } diff --git a/drivers/dma/idxd/sysfs.c b/drivers/dma/idxd/sysfs.c index 6d907fe150aa..6ca6e520a2fa 100644 --- a/drivers/dma/idxd/sysfs.c +++ b/drivers/dma/idxd/sysfs.c @@ -124,6 +124,7 @@ static int idxd_config_bus_probe(struct device *dev) rc = idxd_device_config(idxd); if (rc < 0) { spin_unlock_irqrestore(&idxd->dev_lock, flags); + module_put(THIS_MODULE); dev_warn(dev, "Device config failed: %d\n", rc); return rc; } @@ -132,6 +133,7 @@ static int idxd_config_bus_probe(struct device *dev) rc = idxd_device_enable(idxd); if (rc < 0) { spin_unlock_irqrestore(&idxd->dev_lock, flags); + module_put(THIS_MODULE); dev_warn(dev, "Device enable failed: %d\n", rc); return rc; } @@ -142,6 +144,7 @@ static int idxd_config_bus_probe(struct device *dev) rc = idxd_register_dma_device(idxd); if (rc < 0) { spin_unlock_irqrestore(&idxd->dev_lock, flags); + module_put(THIS_MODULE); dev_dbg(dev, "Failed to register dmaengine device\n"); return rc; } @@ -516,7 +519,7 @@ static ssize_t group_tokens_reserved_store(struct device *dev, if (val > idxd->max_tokens) return -EINVAL; - if (val > idxd->nr_tokens) + if (val > idxd->nr_tokens + group->tokens_reserved) return -EINVAL; group->tokens_reserved = val; @@ -901,6 +904,20 @@ static ssize_t wq_size_show(struct device *dev, struct device_attribute *attr, return sprintf(buf, "%u\n", wq->size); } +static int total_claimed_wq_size(struct idxd_device *idxd) +{ + int i; + int wq_size = 0; + + for (i = 0; i < idxd->max_wqs; i++) { + struct idxd_wq *wq = &idxd->wqs[i]; + + wq_size += wq->size; + } + + return wq_size; +} + static ssize_t wq_size_store(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) @@ -920,7 +937,7 @@ static ssize_t wq_size_store(struct device *dev, if (wq->state != IDXD_WQ_DISABLED) return -EPERM; - if (size > idxd->max_wq_size) + if (size + total_claimed_wq_size(idxd) - wq->size > idxd->max_wq_size) return -EINVAL; wq->size = size; @@ -999,12 +1016,14 @@ static ssize_t wq_type_store(struct device *dev, return -EPERM; old_type = wq->type; - if (sysfs_streq(buf, idxd_wq_type_names[IDXD_WQT_KERNEL])) + if (sysfs_streq(buf, idxd_wq_type_names[IDXD_WQT_NONE])) + wq->type = IDXD_WQT_NONE; + else if (sysfs_streq(buf, idxd_wq_type_names[IDXD_WQT_KERNEL])) wq->type = IDXD_WQT_KERNEL; else if (sysfs_streq(buf, idxd_wq_type_names[IDXD_WQT_USER])) wq->type = IDXD_WQT_USER; else - wq->type = IDXD_WQT_NONE; + return -EINVAL; /* If we are changing queue type, clear the name */ if (wq->type != old_type) diff --git a/drivers/dma/imx-sdma.c b/drivers/dma/imx-sdma.c index 066b21a32232..4d4477df4ede 100644 --- a/drivers/dma/imx-sdma.c +++ b/drivers/dma/imx-sdma.c @@ -1331,13 +1331,14 @@ static void sdma_free_chan_resources(struct dma_chan *chan) sdma_channel_synchronize(chan); - if (sdmac->event_id0) + if (sdmac->event_id0 >= 0) sdma_event_disable(sdmac, sdmac->event_id0); if (sdmac->event_id1) sdma_event_disable(sdmac, sdmac->event_id1); sdmac->event_id0 = 0; sdmac->event_id1 = 0; + sdmac->context_loaded = false; sdma_set_channel_priority(sdmac, 0); @@ -1631,7 +1632,7 @@ static int sdma_config(struct dma_chan *chan, memcpy(&sdmac->slave_config, dmaengine_cfg, sizeof(*dmaengine_cfg)); /* Set ENBLn earlier to make sure dma request triggered after that */ - if (sdmac->event_id0) { + if (sdmac->event_id0 >= 0) { if (sdmac->event_id0 >= sdmac->sdma->drvdata->num_events) return -EINVAL; sdma_event_enable(sdmac, sdmac->event_id0); diff --git a/drivers/dma/tegra20-apb-dma.c b/drivers/dma/tegra20-apb-dma.c index 3a45079d11ec..4a750e29bfb5 100644 --- a/drivers/dma/tegra20-apb-dma.c +++ b/drivers/dma/tegra20-apb-dma.c @@ -281,7 +281,7 @@ static struct tegra_dma_desc *tegra_dma_desc_get( /* Do not allocate if desc are waiting for ack */ list_for_each_entry(dma_desc, &tdc->free_dma_desc, node) { - if (async_tx_test_ack(&dma_desc->txd)) { + if (async_tx_test_ack(&dma_desc->txd) && !dma_desc->cb_count) { list_del(&dma_desc->node); spin_unlock_irqrestore(&tdc->lock, flags); dma_desc->txd.flags = 0; @@ -756,10 +756,6 @@ static int tegra_dma_terminate_all(struct dma_chan *dc) bool was_busy; spin_lock_irqsave(&tdc->lock, flags); - if (list_empty(&tdc->pending_sg_req)) { - spin_unlock_irqrestore(&tdc->lock, flags); - return 0; - } if (!tdc->busy) goto skip_dma_stop; diff --git a/drivers/dma/ti/k3-udma-glue.c b/drivers/dma/ti/k3-udma-glue.c index c1511298ece2..4d7561a1b3e3 100644 --- a/drivers/dma/ti/k3-udma-glue.c +++ b/drivers/dma/ti/k3-udma-glue.c @@ -564,12 +564,12 @@ static int k3_udma_glue_cfg_rx_flow(struct k3_udma_glue_rx_channel *rx_chn, if (IS_ERR(flow->udma_rflow)) { ret = PTR_ERR(flow->udma_rflow); dev_err(dev, "UDMAX rflow get err %d\n", ret); - goto err; + return ret; } if (flow->udma_rflow_id != xudma_rflow_get_id(flow->udma_rflow)) { - xudma_rflow_put(rx_chn->common.udmax, flow->udma_rflow); - return -ENODEV; + ret = -ENODEV; + goto err_rflow_put; } /* request and cfg rings */ @@ -578,7 +578,7 @@ static int k3_udma_glue_cfg_rx_flow(struct k3_udma_glue_rx_channel *rx_chn, if (!flow->ringrx) { ret = -ENODEV; dev_err(dev, "Failed to get RX ring\n"); - goto err; + goto err_rflow_put; } flow->ringrxfdq = k3_ringacc_request_ring(rx_chn->common.ringacc, @@ -586,19 +586,19 @@ static int k3_udma_glue_cfg_rx_flow(struct k3_udma_glue_rx_channel *rx_chn, if (!flow->ringrxfdq) { ret = -ENODEV; dev_err(dev, "Failed to get RXFDQ ring\n"); - goto err; + goto err_ringrx_free; } ret = k3_ringacc_ring_cfg(flow->ringrx, &flow_cfg->rx_cfg); if (ret) { dev_err(dev, "Failed to cfg ringrx %d\n", ret); - goto err; + goto err_ringrxfdq_free; } ret = k3_ringacc_ring_cfg(flow->ringrxfdq, &flow_cfg->rxfdq_cfg); if (ret) { dev_err(dev, "Failed to cfg ringrxfdq %d\n", ret); - goto err; + goto err_ringrxfdq_free; } if (rx_chn->remote) { @@ -648,7 +648,7 @@ static int k3_udma_glue_cfg_rx_flow(struct k3_udma_glue_rx_channel *rx_chn, if (ret) { dev_err(dev, "flow%d config failed: %d\n", flow->udma_rflow_id, ret); - goto err; + goto err_ringrxfdq_free; } rx_chn->flows_ready++; @@ -656,8 +656,17 @@ static int k3_udma_glue_cfg_rx_flow(struct k3_udma_glue_rx_channel *rx_chn, flow->udma_rflow_id, rx_chn->flows_ready); return 0; -err: - k3_udma_glue_release_rx_flow(rx_chn, flow_idx); + +err_ringrxfdq_free: + k3_ringacc_ring_free(flow->ringrxfdq); + +err_ringrx_free: + k3_ringacc_ring_free(flow->ringrx); + +err_rflow_put: + xudma_rflow_put(rx_chn->common.udmax, flow->udma_rflow); + flow->udma_rflow = NULL; + return ret; } diff --git a/drivers/dma/ti/k3-udma.c b/drivers/dma/ti/k3-udma.c index ea79c2df28e0..0536866a58ce 100644 --- a/drivers/dma/ti/k3-udma.c +++ b/drivers/dma/ti/k3-udma.c @@ -5,6 +5,7 @@ */ #include <linux/kernel.h> +#include <linux/delay.h> #include <linux/dmaengine.h> #include <linux/dma-mapping.h> #include <linux/dmapool.h> @@ -96,6 +97,24 @@ struct udma_match_data { u32 level_start_idx[]; }; +struct udma_hwdesc { + size_t cppi5_desc_size; + void *cppi5_desc_vaddr; + dma_addr_t cppi5_desc_paddr; + + /* TR descriptor internal pointers */ + void *tr_req_base; + struct cppi5_tr_resp_t *tr_resp_base; +}; + +struct udma_rx_flush { + struct udma_hwdesc hwdescs[2]; + + size_t buffer_size; + void *buffer_vaddr; + dma_addr_t buffer_paddr; +}; + struct udma_dev { struct dma_device ddev; struct device *dev; @@ -112,6 +131,8 @@ struct udma_dev { struct list_head desc_to_purge; spinlock_t lock; + struct udma_rx_flush rx_flush; + int tchan_cnt; int echan_cnt; int rchan_cnt; @@ -130,16 +151,6 @@ struct udma_dev { u32 psil_base; }; -struct udma_hwdesc { - size_t cppi5_desc_size; - void *cppi5_desc_vaddr; - dma_addr_t cppi5_desc_paddr; - - /* TR descriptor internal pointers */ - void *tr_req_base; - struct cppi5_tr_resp_t *tr_resp_base; -}; - struct udma_desc { struct virt_dma_desc vd; @@ -169,7 +180,7 @@ enum udma_chan_state { struct udma_tx_drain { struct delayed_work work; - unsigned long jiffie; + ktime_t tstamp; u32 residue; }; @@ -502,7 +513,7 @@ static bool udma_is_chan_paused(struct udma_chan *uc) { u32 val, pause_mask; - switch (uc->desc->dir) { + switch (uc->config.dir) { case DMA_DEV_TO_MEM: val = udma_rchanrt_read(uc->rchan, UDMA_RCHAN_RT_PEER_RT_EN_REG); @@ -551,12 +562,17 @@ static void udma_sync_for_device(struct udma_chan *uc, int idx) } } +static inline dma_addr_t udma_get_rx_flush_hwdesc_paddr(struct udma_chan *uc) +{ + return uc->ud->rx_flush.hwdescs[uc->config.pkt_mode].cppi5_desc_paddr; +} + static int udma_push_to_ring(struct udma_chan *uc, int idx) { struct udma_desc *d = uc->desc; - struct k3_ring *ring = NULL; - int ret = -EINVAL; + dma_addr_t paddr; + int ret; switch (uc->config.dir) { case DMA_DEV_TO_MEM: @@ -567,21 +583,37 @@ static int udma_push_to_ring(struct udma_chan *uc, int idx) ring = uc->tchan->t_ring; break; default: - break; + return -EINVAL; } - if (ring) { - dma_addr_t desc_addr = udma_curr_cppi5_desc_paddr(d, idx); + /* RX flush packet: idx == -1 is only passed in case of DEV_TO_MEM */ + if (idx == -1) { + paddr = udma_get_rx_flush_hwdesc_paddr(uc); + } else { + paddr = udma_curr_cppi5_desc_paddr(d, idx); wmb(); /* Ensure that writes are not moved over this point */ udma_sync_for_device(uc, idx); - ret = k3_ringacc_ring_push(ring, &desc_addr); - uc->in_ring_cnt++; } + ret = k3_ringacc_ring_push(ring, &paddr); + if (!ret) + uc->in_ring_cnt++; + return ret; } +static bool udma_desc_is_rx_flush(struct udma_chan *uc, dma_addr_t addr) +{ + if (uc->config.dir != DMA_DEV_TO_MEM) + return false; + + if (addr == udma_get_rx_flush_hwdesc_paddr(uc)) + return true; + + return false; +} + static int udma_pop_from_ring(struct udma_chan *uc, dma_addr_t *addr) { struct k3_ring *ring = NULL; @@ -610,6 +642,10 @@ static int udma_pop_from_ring(struct udma_chan *uc, dma_addr_t *addr) if (cppi5_desc_is_tdcm(*addr)) return ret; + /* Check for flush descriptor */ + if (udma_desc_is_rx_flush(uc, *addr)) + return -ENOENT; + d = udma_udma_desc_from_paddr(uc, *addr); if (d) @@ -890,6 +926,9 @@ static int udma_stop(struct udma_chan *uc) switch (uc->config.dir) { case DMA_DEV_TO_MEM: + if (!uc->cyclic && !uc->desc) + udma_push_to_ring(uc, -1); + udma_rchanrt_write(uc->rchan, UDMA_RCHAN_RT_PEER_RT_EN_REG, UDMA_PEER_RT_EN_ENABLE | UDMA_PEER_RT_EN_TEARDOWN); @@ -946,9 +985,10 @@ static bool udma_is_desc_really_done(struct udma_chan *uc, struct udma_desc *d) peer_bcnt = udma_tchanrt_read(uc->tchan, UDMA_TCHAN_RT_PEER_BCNT_REG); bcnt = udma_tchanrt_read(uc->tchan, UDMA_TCHAN_RT_BCNT_REG); + /* Transfer is incomplete, store current residue and time stamp */ if (peer_bcnt < bcnt) { uc->tx_drain.residue = bcnt - peer_bcnt; - uc->tx_drain.jiffie = jiffies; + uc->tx_drain.tstamp = ktime_get(); return false; } @@ -961,35 +1001,59 @@ static void udma_check_tx_completion(struct work_struct *work) tx_drain.work.work); bool desc_done = true; u32 residue_diff; - unsigned long jiffie_diff, delay; + ktime_t time_diff; + unsigned long delay; + + while (1) { + if (uc->desc) { + /* Get previous residue and time stamp */ + residue_diff = uc->tx_drain.residue; + time_diff = uc->tx_drain.tstamp; + /* + * Get current residue and time stamp or see if + * transfer is complete + */ + desc_done = udma_is_desc_really_done(uc, uc->desc); + } - if (uc->desc) { - residue_diff = uc->tx_drain.residue; - jiffie_diff = uc->tx_drain.jiffie; - desc_done = udma_is_desc_really_done(uc, uc->desc); - } - - if (!desc_done) { - jiffie_diff = uc->tx_drain.jiffie - jiffie_diff; - residue_diff -= uc->tx_drain.residue; - if (residue_diff) { - /* Try to guess when we should check next time */ - residue_diff /= jiffie_diff; - delay = uc->tx_drain.residue / residue_diff / 3; - if (jiffies_to_msecs(delay) < 5) - delay = 0; - } else { - /* No progress, check again in 1 second */ - delay = HZ; + if (!desc_done) { + /* + * Find the time delta and residue delta w.r.t + * previous poll + */ + time_diff = ktime_sub(uc->tx_drain.tstamp, + time_diff) + 1; + residue_diff -= uc->tx_drain.residue; + if (residue_diff) { + /* + * Try to guess when we should check + * next time by calculating rate at + * which data is being drained at the + * peer device + */ + delay = (time_diff / residue_diff) * + uc->tx_drain.residue; + } else { + /* No progress, check again in 1 second */ + schedule_delayed_work(&uc->tx_drain.work, HZ); + break; + } + + usleep_range(ktime_to_us(delay), + ktime_to_us(delay) + 10); + continue; } - schedule_delayed_work(&uc->tx_drain.work, delay); - } else if (uc->desc) { - struct udma_desc *d = uc->desc; + if (uc->desc) { + struct udma_desc *d = uc->desc; - uc->bcnt += d->residue; - udma_start(uc); - vchan_cookie_complete(&d->vd); + uc->bcnt += d->residue; + udma_start(uc); + vchan_cookie_complete(&d->vd); + break; + } + + break; } } @@ -1033,29 +1097,27 @@ static irqreturn_t udma_ring_irq_handler(int irq, void *data) goto out; } - if (uc->cyclic) { - /* push the descriptor back to the ring */ - if (d == uc->desc) { + if (d == uc->desc) { + /* active descriptor */ + if (uc->cyclic) { udma_cyclic_packet_elapsed(uc); vchan_cyclic_callback(&d->vd); - } - } else { - bool desc_done = false; - - if (d == uc->desc) { - desc_done = udma_is_desc_really_done(uc, d); - - if (desc_done) { + } else { + if (udma_is_desc_really_done(uc, d)) { uc->bcnt += d->residue; udma_start(uc); + vchan_cookie_complete(&d->vd); } else { schedule_delayed_work(&uc->tx_drain.work, 0); } } - - if (desc_done) - vchan_cookie_complete(&d->vd); + } else { + /* + * terminated descriptor, mark the descriptor as + * completed to update the channel's cookie marker + */ + dma_cookie_complete(&d->vd.tx); } } out: @@ -1965,36 +2027,81 @@ static struct udma_desc *udma_alloc_tr_desc(struct udma_chan *uc, return d; } +/** + * udma_get_tr_counters - calculate TR counters for a given length + * @len: Length of the trasnfer + * @align_to: Preferred alignment + * @tr0_cnt0: First TR icnt0 + * @tr0_cnt1: First TR icnt1 + * @tr1_cnt0: Second (if used) TR icnt0 + * + * For len < SZ_64K only one TR is enough, tr1_cnt0 is not updated + * For len >= SZ_64K two TRs are used in a simple way: + * First TR: SZ_64K-alignment blocks (tr0_cnt0, tr0_cnt1) + * Second TR: the remaining length (tr1_cnt0) + * + * Returns the number of TRs the length needs (1 or 2) + * -EINVAL if the length can not be supported + */ +static int udma_get_tr_counters(size_t len, unsigned long align_to, + u16 *tr0_cnt0, u16 *tr0_cnt1, u16 *tr1_cnt0) +{ + if (len < SZ_64K) { + *tr0_cnt0 = len; + *tr0_cnt1 = 1; + + return 1; + } + + if (align_to > 3) + align_to = 3; + +realign: + *tr0_cnt0 = SZ_64K - BIT(align_to); + if (len / *tr0_cnt0 >= SZ_64K) { + if (align_to) { + align_to--; + goto realign; + } + return -EINVAL; + } + + *tr0_cnt1 = len / *tr0_cnt0; + *tr1_cnt0 = len % *tr0_cnt0; + + return 2; +} + static struct udma_desc * udma_prep_slave_sg_tr(struct udma_chan *uc, struct scatterlist *sgl, unsigned int sglen, enum dma_transfer_direction dir, unsigned long tx_flags, void *context) { - enum dma_slave_buswidth dev_width; struct scatterlist *sgent; struct udma_desc *d; - size_t tr_size; struct cppi5_tr_type1_t *tr_req = NULL; + u16 tr0_cnt0, tr0_cnt1, tr1_cnt0; unsigned int i; - u32 burst; + size_t tr_size; + int num_tr = 0; + int tr_idx = 0; - if (dir == DMA_DEV_TO_MEM) { - dev_width = uc->cfg.src_addr_width; - burst = uc->cfg.src_maxburst; - } else if (dir == DMA_MEM_TO_DEV) { - dev_width = uc->cfg.dst_addr_width; - burst = uc->cfg.dst_maxburst; - } else { - dev_err(uc->ud->dev, "%s: bad direction?\n", __func__); + if (!is_slave_direction(dir)) { + dev_err(uc->ud->dev, "Only slave cyclic is supported\n"); return NULL; } - if (!burst) - burst = 1; + /* estimate the number of TRs we will need */ + for_each_sg(sgl, sgent, sglen, i) { + if (sg_dma_len(sgent) < SZ_64K) + num_tr++; + else + num_tr += 2; + } /* Now allocate and setup the descriptor. */ tr_size = sizeof(struct cppi5_tr_type1_t); - d = udma_alloc_tr_desc(uc, tr_size, sglen, dir); + d = udma_alloc_tr_desc(uc, tr_size, num_tr, dir); if (!d) return NULL; @@ -2002,19 +2109,46 @@ udma_prep_slave_sg_tr(struct udma_chan *uc, struct scatterlist *sgl, tr_req = d->hwdesc[0].tr_req_base; for_each_sg(sgl, sgent, sglen, i) { - d->residue += sg_dma_len(sgent); + dma_addr_t sg_addr = sg_dma_address(sgent); + + num_tr = udma_get_tr_counters(sg_dma_len(sgent), __ffs(sg_addr), + &tr0_cnt0, &tr0_cnt1, &tr1_cnt0); + if (num_tr < 0) { + dev_err(uc->ud->dev, "size %u is not supported\n", + sg_dma_len(sgent)); + udma_free_hwdesc(uc, d); + kfree(d); + return NULL; + } cppi5_tr_init(&tr_req[i].flags, CPPI5_TR_TYPE1, false, false, CPPI5_TR_EVENT_SIZE_COMPLETION, 0); cppi5_tr_csf_set(&tr_req[i].flags, CPPI5_TR_CSF_SUPR_EVT); - tr_req[i].addr = sg_dma_address(sgent); - tr_req[i].icnt0 = burst * dev_width; - tr_req[i].dim1 = burst * dev_width; - tr_req[i].icnt1 = sg_dma_len(sgent) / tr_req[i].icnt0; + tr_req[tr_idx].addr = sg_addr; + tr_req[tr_idx].icnt0 = tr0_cnt0; + tr_req[tr_idx].icnt1 = tr0_cnt1; + tr_req[tr_idx].dim1 = tr0_cnt0; + tr_idx++; + + if (num_tr == 2) { + cppi5_tr_init(&tr_req[tr_idx].flags, CPPI5_TR_TYPE1, + false, false, + CPPI5_TR_EVENT_SIZE_COMPLETION, 0); + cppi5_tr_csf_set(&tr_req[tr_idx].flags, + CPPI5_TR_CSF_SUPR_EVT); + + tr_req[tr_idx].addr = sg_addr + tr0_cnt1 * tr0_cnt0; + tr_req[tr_idx].icnt0 = tr1_cnt0; + tr_req[tr_idx].icnt1 = 1; + tr_req[tr_idx].dim1 = tr1_cnt0; + tr_idx++; + } + + d->residue += sg_dma_len(sgent); } - cppi5_tr_csf_set(&tr_req[i - 1].flags, CPPI5_TR_CSF_EOP); + cppi5_tr_csf_set(&tr_req[tr_idx - 1].flags, CPPI5_TR_CSF_EOP); return d; } @@ -2319,47 +2453,66 @@ udma_prep_dma_cyclic_tr(struct udma_chan *uc, dma_addr_t buf_addr, size_t buf_len, size_t period_len, enum dma_transfer_direction dir, unsigned long flags) { - enum dma_slave_buswidth dev_width; struct udma_desc *d; - size_t tr_size; + size_t tr_size, period_addr; struct cppi5_tr_type1_t *tr_req; - unsigned int i; unsigned int periods = buf_len / period_len; - u32 burst; + u16 tr0_cnt0, tr0_cnt1, tr1_cnt0; + unsigned int i; + int num_tr; - if (dir == DMA_DEV_TO_MEM) { - dev_width = uc->cfg.src_addr_width; - burst = uc->cfg.src_maxburst; - } else if (dir == DMA_MEM_TO_DEV) { - dev_width = uc->cfg.dst_addr_width; - burst = uc->cfg.dst_maxburst; - } else { - dev_err(uc->ud->dev, "%s: bad direction?\n", __func__); + if (!is_slave_direction(dir)) { + dev_err(uc->ud->dev, "Only slave cyclic is supported\n"); return NULL; } - if (!burst) - burst = 1; + num_tr = udma_get_tr_counters(period_len, __ffs(buf_addr), &tr0_cnt0, + &tr0_cnt1, &tr1_cnt0); + if (num_tr < 0) { + dev_err(uc->ud->dev, "size %zu is not supported\n", + period_len); + return NULL; + } /* Now allocate and setup the descriptor. */ tr_size = sizeof(struct cppi5_tr_type1_t); - d = udma_alloc_tr_desc(uc, tr_size, periods, dir); + d = udma_alloc_tr_desc(uc, tr_size, periods * num_tr, dir); if (!d) return NULL; tr_req = d->hwdesc[0].tr_req_base; + period_addr = buf_addr; for (i = 0; i < periods; i++) { - cppi5_tr_init(&tr_req[i].flags, CPPI5_TR_TYPE1, false, false, - CPPI5_TR_EVENT_SIZE_COMPLETION, 0); + int tr_idx = i * num_tr; - tr_req[i].addr = buf_addr + period_len * i; - tr_req[i].icnt0 = dev_width; - tr_req[i].icnt1 = period_len / dev_width; - tr_req[i].dim1 = dev_width; + cppi5_tr_init(&tr_req[tr_idx].flags, CPPI5_TR_TYPE1, false, + false, CPPI5_TR_EVENT_SIZE_COMPLETION, 0); + + tr_req[tr_idx].addr = period_addr; + tr_req[tr_idx].icnt0 = tr0_cnt0; + tr_req[tr_idx].icnt1 = tr0_cnt1; + tr_req[tr_idx].dim1 = tr0_cnt0; + + if (num_tr == 2) { + cppi5_tr_csf_set(&tr_req[tr_idx].flags, + CPPI5_TR_CSF_SUPR_EVT); + tr_idx++; + + cppi5_tr_init(&tr_req[tr_idx].flags, CPPI5_TR_TYPE1, + false, false, + CPPI5_TR_EVENT_SIZE_COMPLETION, 0); + + tr_req[tr_idx].addr = period_addr + tr0_cnt1 * tr0_cnt0; + tr_req[tr_idx].icnt0 = tr1_cnt0; + tr_req[tr_idx].icnt1 = 1; + tr_req[tr_idx].dim1 = tr1_cnt0; + } if (!(flags & DMA_PREP_INTERRUPT)) - cppi5_tr_csf_set(&tr_req[i].flags, + cppi5_tr_csf_set(&tr_req[tr_idx].flags, CPPI5_TR_CSF_SUPR_EVT); + + period_addr += period_len; } return d; @@ -2517,29 +2670,12 @@ udma_prep_dma_memcpy(struct dma_chan *chan, dma_addr_t dest, dma_addr_t src, return NULL; } - if (len < SZ_64K) { - num_tr = 1; - tr0_cnt0 = len; - tr0_cnt1 = 1; - } else { - unsigned long align_to = __ffs(src | dest); - - if (align_to > 3) - align_to = 3; - /* - * Keep simple: tr0: SZ_64K-alignment blocks, - * tr1: the remaining - */ - num_tr = 2; - tr0_cnt0 = (SZ_64K - BIT(align_to)); - if (len / tr0_cnt0 >= SZ_64K) { - dev_err(uc->ud->dev, "size %zu is not supported\n", - len); - return NULL; - } - - tr0_cnt1 = len / tr0_cnt0; - tr1_cnt0 = len % tr0_cnt0; + num_tr = udma_get_tr_counters(len, __ffs(src | dest), &tr0_cnt0, + &tr0_cnt1, &tr1_cnt0); + if (num_tr < 0) { + dev_err(uc->ud->dev, "size %zu is not supported\n", + len); + return NULL; } d = udma_alloc_tr_desc(uc, tr_size, num_tr, DMA_MEM_TO_MEM); @@ -2631,6 +2767,9 @@ static enum dma_status udma_tx_status(struct dma_chan *chan, ret = dma_cookie_status(chan, cookie, txstate); + if (!udma_is_chan_running(uc)) + ret = DMA_COMPLETE; + if (ret == DMA_IN_PROGRESS && udma_is_chan_paused(uc)) ret = DMA_PAUSED; @@ -2697,11 +2836,8 @@ static int udma_pause(struct dma_chan *chan) { struct udma_chan *uc = to_udma_chan(chan); - if (!uc->desc) - return -EINVAL; - /* pause the channel */ - switch (uc->desc->dir) { + switch (uc->config.dir) { case DMA_DEV_TO_MEM: udma_rchanrt_update_bits(uc->rchan, UDMA_RCHAN_RT_PEER_RT_EN_REG, @@ -2730,11 +2866,8 @@ static int udma_resume(struct dma_chan *chan) { struct udma_chan *uc = to_udma_chan(chan); - if (!uc->desc) - return -EINVAL; - /* resume the channel */ - switch (uc->desc->dir) { + switch (uc->config.dir) { case DMA_DEV_TO_MEM: udma_rchanrt_update_bits(uc->rchan, UDMA_RCHAN_RT_PEER_RT_EN_REG, @@ -3248,6 +3381,98 @@ static int udma_setup_resources(struct udma_dev *ud) return ch_count; } +static int udma_setup_rx_flush(struct udma_dev *ud) +{ + struct udma_rx_flush *rx_flush = &ud->rx_flush; + struct cppi5_desc_hdr_t *tr_desc; + struct cppi5_tr_type1_t *tr_req; + struct cppi5_host_desc_t *desc; + struct device *dev = ud->dev; + struct udma_hwdesc *hwdesc; + size_t tr_size; + + /* Allocate 1K buffer for discarded data on RX channel teardown */ + rx_flush->buffer_size = SZ_1K; + rx_flush->buffer_vaddr = devm_kzalloc(dev, rx_flush->buffer_size, + GFP_KERNEL); + if (!rx_flush->buffer_vaddr) + return -ENOMEM; + + rx_flush->buffer_paddr = dma_map_single(dev, rx_flush->buffer_vaddr, + rx_flush->buffer_size, + DMA_TO_DEVICE); + if (dma_mapping_error(dev, rx_flush->buffer_paddr)) + return -ENOMEM; + + /* Set up descriptor to be used for TR mode */ + hwdesc = &rx_flush->hwdescs[0]; + tr_size = sizeof(struct cppi5_tr_type1_t); + hwdesc->cppi5_desc_size = cppi5_trdesc_calc_size(tr_size, 1); + hwdesc->cppi5_desc_size = ALIGN(hwdesc->cppi5_desc_size, + ud->desc_align); + + hwdesc->cppi5_desc_vaddr = devm_kzalloc(dev, hwdesc->cppi5_desc_size, + GFP_KERNEL); + if (!hwdesc->cppi5_desc_vaddr) + return -ENOMEM; + + hwdesc->cppi5_desc_paddr = dma_map_single(dev, hwdesc->cppi5_desc_vaddr, + hwdesc->cppi5_desc_size, + DMA_TO_DEVICE); + if (dma_mapping_error(dev, hwdesc->cppi5_desc_paddr)) + return -ENOMEM; + + /* Start of the TR req records */ + hwdesc->tr_req_base = hwdesc->cppi5_desc_vaddr + tr_size; + /* Start address of the TR response array */ + hwdesc->tr_resp_base = hwdesc->tr_req_base + tr_size; + + tr_desc = hwdesc->cppi5_desc_vaddr; + cppi5_trdesc_init(tr_desc, 1, tr_size, 0, 0); + cppi5_desc_set_pktids(tr_desc, 0, CPPI5_INFO1_DESC_FLOWID_DEFAULT); + cppi5_desc_set_retpolicy(tr_desc, 0, 0); + + tr_req = hwdesc->tr_req_base; + cppi5_tr_init(&tr_req->flags, CPPI5_TR_TYPE1, false, false, + CPPI5_TR_EVENT_SIZE_COMPLETION, 0); + cppi5_tr_csf_set(&tr_req->flags, CPPI5_TR_CSF_SUPR_EVT); + + tr_req->addr = rx_flush->buffer_paddr; + tr_req->icnt0 = rx_flush->buffer_size; + tr_req->icnt1 = 1; + + /* Set up descriptor to be used for packet mode */ + hwdesc = &rx_flush->hwdescs[1]; + hwdesc->cppi5_desc_size = ALIGN(sizeof(struct cppi5_host_desc_t) + + CPPI5_INFO0_HDESC_EPIB_SIZE + + CPPI5_INFO0_HDESC_PSDATA_MAX_SIZE, + ud->desc_align); + + hwdesc->cppi5_desc_vaddr = devm_kzalloc(dev, hwdesc->cppi5_desc_size, + GFP_KERNEL); + if (!hwdesc->cppi5_desc_vaddr) + return -ENOMEM; + + hwdesc->cppi5_desc_paddr = dma_map_single(dev, hwdesc->cppi5_desc_vaddr, + hwdesc->cppi5_desc_size, + DMA_TO_DEVICE); + if (dma_mapping_error(dev, hwdesc->cppi5_desc_paddr)) + return -ENOMEM; + + desc = hwdesc->cppi5_desc_vaddr; + cppi5_hdesc_init(desc, 0, 0); + cppi5_desc_set_pktids(&desc->hdr, 0, CPPI5_INFO1_DESC_FLOWID_DEFAULT); + cppi5_desc_set_retpolicy(&desc->hdr, 0, 0); + + cppi5_hdesc_attach_buf(desc, + rx_flush->buffer_paddr, rx_flush->buffer_size, + rx_flush->buffer_paddr, rx_flush->buffer_size); + + dma_sync_single_for_device(dev, hwdesc->cppi5_desc_paddr, + hwdesc->cppi5_desc_size, DMA_TO_DEVICE); + return 0; +} + #define TI_UDMAC_BUSWIDTHS (BIT(DMA_SLAVE_BUSWIDTH_1_BYTE) | \ BIT(DMA_SLAVE_BUSWIDTH_2_BYTES) | \ BIT(DMA_SLAVE_BUSWIDTH_3_BYTES) | \ @@ -3361,6 +3586,10 @@ static int udma_probe(struct platform_device *pdev) if (ud->desc_align < dma_get_cache_alignment()) ud->desc_align = dma_get_cache_alignment(); + ret = udma_setup_rx_flush(ud); + if (ret) + return ret; + for (i = 0; i < ud->tchan_cnt; i++) { struct udma_tchan *tchan = &ud->tchans[i]; diff --git a/drivers/edac/synopsys_edac.c b/drivers/edac/synopsys_edac.c index 2d263382d797..880ffd833718 100644 --- a/drivers/edac/synopsys_edac.c +++ b/drivers/edac/synopsys_edac.c @@ -479,20 +479,14 @@ static void handle_error(struct mem_ctl_info *mci, struct synps_ecc_status *p) pinf = &p->ceinfo; if (!priv->p_data->quirks) { snprintf(priv->message, SYNPS_EDAC_MSG_SIZE, - "DDR ECC error type:%s Row %d Bank %d Col %d ", - "CE", pinf->row, pinf->bank, pinf->col); - snprintf(priv->message, SYNPS_EDAC_MSG_SIZE, - "Bit Position: %d Data: 0x%08x\n", + "DDR ECC error type:%s Row %d Bank %d Col %d Bit Position: %d Data: 0x%08x", + "CE", pinf->row, pinf->bank, pinf->col, pinf->bitpos, pinf->data); } else { snprintf(priv->message, SYNPS_EDAC_MSG_SIZE, - "DDR ECC error type:%s Row %d Bank %d Col %d ", - "CE", pinf->row, pinf->bank, pinf->col); - snprintf(priv->message, SYNPS_EDAC_MSG_SIZE, - "BankGroup Number %d Block Number %d ", - pinf->bankgrpnr, pinf->blknr); - snprintf(priv->message, SYNPS_EDAC_MSG_SIZE, - "Bit Position: %d Data: 0x%08x\n", + "DDR ECC error type:%s Row %d Bank %d Col %d BankGroup Number %d Block Number %d Bit Position: %d Data: 0x%08x", + "CE", pinf->row, pinf->bank, pinf->col, + pinf->bankgrpnr, pinf->blknr, pinf->bitpos, pinf->data); } @@ -509,10 +503,8 @@ static void handle_error(struct mem_ctl_info *mci, struct synps_ecc_status *p) "UE", pinf->row, pinf->bank, pinf->col); } else { snprintf(priv->message, SYNPS_EDAC_MSG_SIZE, - "DDR ECC error type :%s Row %d Bank %d Col %d ", - "UE", pinf->row, pinf->bank, pinf->col); - snprintf(priv->message, SYNPS_EDAC_MSG_SIZE, - "BankGroup Number %d Block Number %d", + "DDR ECC error type :%s Row %d Bank %d Col %d BankGroup Number %d Block Number %d", + "UE", pinf->row, pinf->bank, pinf->col, pinf->bankgrpnr, pinf->blknr); } diff --git a/drivers/firmware/efi/efi.c b/drivers/firmware/efi/efi.c index 621220ab3d0e..21ea99f65113 100644 --- a/drivers/firmware/efi/efi.c +++ b/drivers/firmware/efi/efi.c @@ -552,7 +552,7 @@ int __init efi_config_parse_tables(void *config_tables, int count, int sz, seed = early_memremap(efi.rng_seed, sizeof(*seed)); if (seed != NULL) { - size = seed->size; + size = READ_ONCE(seed->size); early_memunmap(seed, sizeof(*seed)); } else { pr_err("Could not map UEFI random seed!\n"); @@ -562,7 +562,7 @@ int __init efi_config_parse_tables(void *config_tables, int count, int sz, sizeof(*seed) + size); if (seed != NULL) { pr_notice("seeding entropy pool\n"); - add_bootloader_randomness(seed->bits, seed->size); + add_bootloader_randomness(seed->bits, size); early_memunmap(seed, sizeof(*seed) + size); } else { pr_err("Could not map UEFI random seed!\n"); diff --git a/drivers/firmware/efi/efivars.c b/drivers/firmware/efi/efivars.c index 7576450c8254..aff3dfb4d7ba 100644 --- a/drivers/firmware/efi/efivars.c +++ b/drivers/firmware/efi/efivars.c @@ -83,13 +83,16 @@ static ssize_t efivar_attr_read(struct efivar_entry *entry, char *buf) { struct efi_variable *var = &entry->var; + unsigned long size = sizeof(var->Data); char *str = buf; + int ret; if (!entry || !buf) return -EINVAL; - var->DataSize = 1024; - if (efivar_entry_get(entry, &var->Attributes, &var->DataSize, var->Data)) + ret = efivar_entry_get(entry, &var->Attributes, &size, var->Data); + var->DataSize = size; + if (ret) return -EIO; if (var->Attributes & EFI_VARIABLE_NON_VOLATILE) @@ -116,13 +119,16 @@ static ssize_t efivar_size_read(struct efivar_entry *entry, char *buf) { struct efi_variable *var = &entry->var; + unsigned long size = sizeof(var->Data); char *str = buf; + int ret; if (!entry || !buf) return -EINVAL; - var->DataSize = 1024; - if (efivar_entry_get(entry, &var->Attributes, &var->DataSize, var->Data)) + ret = efivar_entry_get(entry, &var->Attributes, &size, var->Data); + var->DataSize = size; + if (ret) return -EIO; str += sprintf(str, "0x%lx\n", var->DataSize); @@ -133,12 +139,15 @@ static ssize_t efivar_data_read(struct efivar_entry *entry, char *buf) { struct efi_variable *var = &entry->var; + unsigned long size = sizeof(var->Data); + int ret; if (!entry || !buf) return -EINVAL; - var->DataSize = 1024; - if (efivar_entry_get(entry, &var->Attributes, &var->DataSize, var->Data)) + ret = efivar_entry_get(entry, &var->Attributes, &size, var->Data); + var->DataSize = size; + if (ret) return -EIO; memcpy(buf, var->Data, var->DataSize); @@ -199,6 +208,9 @@ efivar_store_raw(struct efivar_entry *entry, const char *buf, size_t count) u8 *data; int err; + if (!entry || !buf) + return -EINVAL; + if (in_compat_syscall()) { struct compat_efi_variable *compat; @@ -250,14 +262,16 @@ efivar_show_raw(struct efivar_entry *entry, char *buf) { struct efi_variable *var = &entry->var; struct compat_efi_variable *compat; + unsigned long datasize = sizeof(var->Data); size_t size; + int ret; if (!entry || !buf) return 0; - var->DataSize = 1024; - if (efivar_entry_get(entry, &entry->var.Attributes, - &entry->var.DataSize, entry->var.Data)) + ret = efivar_entry_get(entry, &var->Attributes, &datasize, var->Data); + var->DataSize = datasize; + if (ret) return -EIO; if (in_compat_syscall()) { diff --git a/drivers/firmware/imx/imx-scu.c b/drivers/firmware/imx/imx-scu.c index 03b43b7a6d1d..f71eaa5bf52d 100644 --- a/drivers/firmware/imx/imx-scu.c +++ b/drivers/firmware/imx/imx-scu.c @@ -29,6 +29,7 @@ struct imx_sc_chan { struct mbox_client cl; struct mbox_chan *ch; int idx; + struct completion tx_done; }; struct imx_sc_ipc { @@ -100,6 +101,14 @@ int imx_scu_get_handle(struct imx_sc_ipc **ipc) } EXPORT_SYMBOL(imx_scu_get_handle); +/* Callback called when the word of a message is ack-ed, eg read by SCU */ +static void imx_scu_tx_done(struct mbox_client *cl, void *mssg, int r) +{ + struct imx_sc_chan *sc_chan = container_of(cl, struct imx_sc_chan, cl); + + complete(&sc_chan->tx_done); +} + static void imx_scu_rx_callback(struct mbox_client *c, void *msg) { struct imx_sc_chan *sc_chan = container_of(c, struct imx_sc_chan, cl); @@ -149,6 +158,19 @@ static int imx_scu_ipc_write(struct imx_sc_ipc *sc_ipc, void *msg) for (i = 0; i < hdr->size; i++) { sc_chan = &sc_ipc->chans[i % 4]; + + /* + * SCU requires that all messages words are written + * sequentially but linux MU driver implements multiple + * independent channels for each register so ordering between + * different channels must be ensured by SCU API interface. + * + * Wait for tx_done before every send to ensure that no + * queueing happens at the mailbox channel level. + */ + wait_for_completion(&sc_chan->tx_done); + reinit_completion(&sc_chan->tx_done); + ret = mbox_send_message(sc_chan->ch, &data[i]); if (ret < 0) return ret; @@ -247,6 +269,11 @@ static int imx_scu_probe(struct platform_device *pdev) cl->knows_txdone = true; cl->rx_callback = imx_scu_rx_callback; + /* Initial tx_done completion as "done" */ + cl->tx_done = imx_scu_tx_done; + init_completion(&sc_chan->tx_done); + complete(&sc_chan->tx_done); + sc_chan->sc_ipc = sc_ipc; sc_chan->idx = i % 4; sc_chan->ch = mbox_request_channel_byname(cl, chan_name); diff --git a/drivers/firmware/imx/misc.c b/drivers/firmware/imx/misc.c index 4b56a587dacd..d073cb3ce699 100644 --- a/drivers/firmware/imx/misc.c +++ b/drivers/firmware/imx/misc.c @@ -16,7 +16,7 @@ struct imx_sc_msg_req_misc_set_ctrl { u32 ctrl; u32 val; u16 resource; -} __packed; +} __packed __aligned(4); struct imx_sc_msg_req_cpu_start { struct imx_sc_rpc_msg hdr; @@ -24,18 +24,18 @@ struct imx_sc_msg_req_cpu_start { u32 address_lo; u16 resource; u8 enable; -} __packed; +} __packed __aligned(4); struct imx_sc_msg_req_misc_get_ctrl { struct imx_sc_rpc_msg hdr; u32 ctrl; u16 resource; -} __packed; +} __packed __aligned(4); struct imx_sc_msg_resp_misc_get_ctrl { struct imx_sc_rpc_msg hdr; u32 val; -} __packed; +} __packed __aligned(4); /* * This function sets a miscellaneous control value. diff --git a/drivers/firmware/imx/scu-pd.c b/drivers/firmware/imx/scu-pd.c index b556612207e5..af3ae0087de4 100644 --- a/drivers/firmware/imx/scu-pd.c +++ b/drivers/firmware/imx/scu-pd.c @@ -61,7 +61,7 @@ struct imx_sc_msg_req_set_resource_power_mode { struct imx_sc_rpc_msg hdr; u16 resource; u8 mode; -} __packed; +} __packed __aligned(4); #define IMX_SCU_PD_NAME_SIZE 20 struct imx_sc_pm_domain { diff --git a/drivers/fsi/Kconfig b/drivers/fsi/Kconfig index 92ce6d85802c..4cc0e630ab79 100644 --- a/drivers/fsi/Kconfig +++ b/drivers/fsi/Kconfig @@ -55,6 +55,7 @@ config FSI_MASTER_AST_CF config FSI_MASTER_ASPEED tristate "FSI ASPEED master" + depends on HAS_IOMEM help This option enables a FSI master that is present behind an OPB bridge in the AST2600. diff --git a/drivers/gpio/gpiolib-acpi.c b/drivers/gpio/gpiolib-acpi.c index 31fee5e918b7..0017367e94ee 100644 --- a/drivers/gpio/gpiolib-acpi.c +++ b/drivers/gpio/gpiolib-acpi.c @@ -21,18 +21,21 @@ #include "gpiolib.h" #include "gpiolib-acpi.h" -#define QUIRK_NO_EDGE_EVENTS_ON_BOOT 0x01l -#define QUIRK_NO_WAKEUP 0x02l - static int run_edge_events_on_boot = -1; module_param(run_edge_events_on_boot, int, 0444); MODULE_PARM_DESC(run_edge_events_on_boot, "Run edge _AEI event-handlers at boot: 0=no, 1=yes, -1=auto"); -static int honor_wakeup = -1; -module_param(honor_wakeup, int, 0444); -MODULE_PARM_DESC(honor_wakeup, - "Honor the ACPI wake-capable flag: 0=no, 1=yes, -1=auto"); +static char *ignore_wake; +module_param(ignore_wake, charp, 0444); +MODULE_PARM_DESC(ignore_wake, + "controller@pin combos on which to ignore the ACPI wake flag " + "ignore_wake=controller@pin[,controller@pin[,...]]"); + +struct acpi_gpiolib_dmi_quirk { + bool no_edge_events_on_boot; + char *ignore_wake; +}; /** * struct acpi_gpio_event - ACPI GPIO event handler data @@ -202,6 +205,57 @@ static void acpi_gpiochip_request_irqs(struct acpi_gpio_chip *acpi_gpio) acpi_gpiochip_request_irq(acpi_gpio, event); } +static bool acpi_gpio_in_ignore_list(const char *controller_in, int pin_in) +{ + const char *controller, *pin_str; + int len, pin; + char *endp; + + controller = ignore_wake; + while (controller) { + pin_str = strchr(controller, '@'); + if (!pin_str) + goto err; + + len = pin_str - controller; + if (len == strlen(controller_in) && + strncmp(controller, controller_in, len) == 0) { + pin = simple_strtoul(pin_str + 1, &endp, 10); + if (*endp != 0 && *endp != ',') + goto err; + + if (pin == pin_in) + return true; + } + + controller = strchr(controller, ','); + if (controller) + controller++; + } + + return false; +err: + pr_err_once("Error invalid value for gpiolib_acpi.ignore_wake: %s\n", + ignore_wake); + return false; +} + +static bool acpi_gpio_irq_is_wake(struct device *parent, + struct acpi_resource_gpio *agpio) +{ + int pin = agpio->pin_table[0]; + + if (agpio->wake_capable != ACPI_WAKE_CAPABLE) + return false; + + if (acpi_gpio_in_ignore_list(dev_name(parent), pin)) { + dev_info(parent, "Ignoring wakeup on pin %d\n", pin); + return false; + } + + return true; +} + /* Always returns AE_OK so that we keep looping over the resources */ static acpi_status acpi_gpiochip_alloc_event(struct acpi_resource *ares, void *context) @@ -289,7 +343,7 @@ static acpi_status acpi_gpiochip_alloc_event(struct acpi_resource *ares, event->handle = evt_handle; event->handler = handler; event->irq = irq; - event->irq_is_wake = honor_wakeup && agpio->wake_capable == ACPI_WAKE_CAPABLE; + event->irq_is_wake = acpi_gpio_irq_is_wake(chip->parent, agpio); event->pin = pin; event->desc = desc; @@ -1328,7 +1382,9 @@ static const struct dmi_system_id gpiolib_acpi_quirks[] = { DMI_MATCH(DMI_SYS_VENDOR, "MINIX"), DMI_MATCH(DMI_PRODUCT_NAME, "Z83-4"), }, - .driver_data = (void *)QUIRK_NO_EDGE_EVENTS_ON_BOOT, + .driver_data = &(struct acpi_gpiolib_dmi_quirk) { + .no_edge_events_on_boot = true, + }, }, { /* @@ -1341,16 +1397,20 @@ static const struct dmi_system_id gpiolib_acpi_quirks[] = { DMI_MATCH(DMI_SYS_VENDOR, "Wortmann_AG"), DMI_MATCH(DMI_PRODUCT_NAME, "TERRA_PAD_1061"), }, - .driver_data = (void *)QUIRK_NO_EDGE_EVENTS_ON_BOOT, + .driver_data = &(struct acpi_gpiolib_dmi_quirk) { + .no_edge_events_on_boot = true, + }, }, { /* - * Various HP X2 10 Cherry Trail models use an external - * embedded-controller connected via I2C + an ACPI GPIO - * event handler. The embedded controller generates various - * spurious wakeup events when suspended. So disable wakeup - * for its handler (it uses the only ACPI GPIO event handler). - * This breaks wakeup when opening the lid, the user needs + * HP X2 10 models with Cherry Trail SoC + TI PMIC use an + * external embedded-controller connected via I2C + an ACPI GPIO + * event handler on INT33FF:01 pin 0, causing spurious wakeups. + * When suspending by closing the LID, the power to the USB + * keyboard is turned off, causing INT0002 ACPI events to + * trigger once the XHCI controller notices the keyboard is + * gone. So INT0002 events cause spurious wakeups too. Ignoring + * EC wakes breaks wakeup when opening the lid, the user needs * to press the power-button to wakeup the system. The * alternative is suspend simply not working, which is worse. */ @@ -1358,33 +1418,61 @@ static const struct dmi_system_id gpiolib_acpi_quirks[] = { DMI_MATCH(DMI_SYS_VENDOR, "HP"), DMI_MATCH(DMI_PRODUCT_NAME, "HP x2 Detachable 10-p0XX"), }, - .driver_data = (void *)QUIRK_NO_WAKEUP, + .driver_data = &(struct acpi_gpiolib_dmi_quirk) { + .ignore_wake = "INT33FF:01@0,INT0002:00@2", + }, + }, + { + /* + * HP X2 10 models with Bay Trail SoC + AXP288 PMIC use an + * external embedded-controller connected via I2C + an ACPI GPIO + * event handler on INT33FC:02 pin 28, causing spurious wakeups. + */ + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Hewlett-Packard"), + DMI_MATCH(DMI_PRODUCT_NAME, "HP Pavilion x2 Detachable"), + DMI_MATCH(DMI_BOARD_NAME, "815D"), + }, + .driver_data = &(struct acpi_gpiolib_dmi_quirk) { + .ignore_wake = "INT33FC:02@28", + }, + }, + { + /* + * HP X2 10 models with Cherry Trail SoC + AXP288 PMIC use an + * external embedded-controller connected via I2C + an ACPI GPIO + * event handler on INT33FF:01 pin 0, causing spurious wakeups. + */ + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "HP"), + DMI_MATCH(DMI_PRODUCT_NAME, "HP Pavilion x2 Detachable"), + DMI_MATCH(DMI_BOARD_NAME, "813E"), + }, + .driver_data = &(struct acpi_gpiolib_dmi_quirk) { + .ignore_wake = "INT33FF:01@0", + }, }, {} /* Terminating entry */ }; static int acpi_gpio_setup_params(void) { + const struct acpi_gpiolib_dmi_quirk *quirk = NULL; const struct dmi_system_id *id; - long quirks = 0; id = dmi_first_match(gpiolib_acpi_quirks); if (id) - quirks = (long)id->driver_data; + quirk = id->driver_data; if (run_edge_events_on_boot < 0) { - if (quirks & QUIRK_NO_EDGE_EVENTS_ON_BOOT) + if (quirk && quirk->no_edge_events_on_boot) run_edge_events_on_boot = 0; else run_edge_events_on_boot = 1; } - if (honor_wakeup < 0) { - if (quirks & QUIRK_NO_WAKEUP) - honor_wakeup = 0; - else - honor_wakeup = 1; - } + if (ignore_wake == NULL && quirk && quirk->ignore_wake) + ignore_wake = quirk->ignore_wake; return 0; } diff --git a/drivers/gpio/gpiolib.c b/drivers/gpio/gpiolib.c index 4d0106ceeba7..00fb91feba70 100644 --- a/drivers/gpio/gpiolib.c +++ b/drivers/gpio/gpiolib.c @@ -2306,9 +2306,16 @@ static void gpiochip_irq_disable(struct irq_data *d) { struct gpio_chip *chip = irq_data_get_irq_chip_data(d); + /* + * Since we override .irq_disable() we need to mimic the + * behaviour of __irq_disable() in irq/chip.c. + * First call .irq_disable() if it exists, else mimic the + * behaviour of mask_irq() which calls .irq_mask() if + * it exists. + */ if (chip->irq.irq_disable) chip->irq.irq_disable(d); - else + else if (chip->irq.chip->irq_mask) chip->irq.chip->irq_mask(d); gpiochip_disable_irq(chip, d->hwirq); } diff --git a/drivers/gpu/drm/amd/acp/Kconfig b/drivers/gpu/drm/amd/acp/Kconfig index 13340f353ea8..216d932a7831 100644 --- a/drivers/gpu/drm/amd/acp/Kconfig +++ b/drivers/gpu/drm/amd/acp/Kconfig @@ -1,5 +1,6 @@ # SPDX-License-Identifier: MIT menu "ACP (Audio CoProcessor) Configuration" + depends on DRM_AMDGPU config DRM_AMD_ACP bool "Enable AMD Audio CoProcessor IP support" diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index 97dd9596d17a..2992a49ad4a5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -579,6 +579,7 @@ struct amdgpu_asic_funcs { /* invalidate hdp read cache */ void (*invalidate_hdp)(struct amdgpu_device *adev, struct amdgpu_ring *ring); + void (*reset_hdp_ras_error_count)(struct amdgpu_device *adev); /* check if the asic needs a full reset of if soft reset will work */ bool (*need_full_reset)(struct amdgpu_device *adev); /* initialize doorbell layout for specific asic*/ @@ -993,6 +994,8 @@ uint32_t amdgpu_mm_rreg(struct amdgpu_device *adev, uint32_t reg, uint32_t acc_flags); void amdgpu_mm_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v, uint32_t acc_flags); +void amdgpu_mm_wreg_mmio_rlc(struct amdgpu_device *adev, uint32_t reg, uint32_t v, + uint32_t acc_flags); void amdgpu_mm_wreg8(struct amdgpu_device *adev, uint32_t offset, uint8_t value); uint8_t amdgpu_mm_rreg8(struct amdgpu_device *adev, uint32_t offset); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c index bc2e72a66db9..abfbe89e805e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c @@ -29,6 +29,7 @@ #include <linux/module.h> #include <linux/dma-buf.h> #include "amdgpu_xgmi.h" +#include <uapi/linux/kfd_ioctl.h> static const unsigned int compute_vmid_bitmap = 0xFF00; @@ -224,7 +225,7 @@ void amdgpu_amdkfd_gpu_reset(struct kgd_dev *kgd) int amdgpu_amdkfd_alloc_gtt_mem(struct kgd_dev *kgd, size_t size, void **mem_obj, uint64_t *gpu_addr, - void **cpu_ptr, bool mqd_gfx9) + void **cpu_ptr, bool cp_mqd_gfx9) { struct amdgpu_device *adev = (struct amdgpu_device *)kgd; struct amdgpu_bo *bo = NULL; @@ -240,8 +241,8 @@ int amdgpu_amdkfd_alloc_gtt_mem(struct kgd_dev *kgd, size_t size, bp.type = ttm_bo_type_kernel; bp.resv = NULL; - if (mqd_gfx9) - bp.flags |= AMDGPU_GEM_CREATE_MQD_GFX9; + if (cp_mqd_gfx9) + bp.flags |= AMDGPU_GEM_CREATE_CP_MQD_GFX9; r = amdgpu_bo_create(adev, &bp, &bo); if (r) { @@ -501,10 +502,11 @@ int amdgpu_amdkfd_get_dmabuf_info(struct kgd_dev *kgd, int dma_buf_fd, metadata_size, &metadata_flags); if (flags) { *flags = (bo->preferred_domains & AMDGPU_GEM_DOMAIN_VRAM) ? - ALLOC_MEM_FLAGS_VRAM : ALLOC_MEM_FLAGS_GTT; + KFD_IOC_ALLOC_MEM_FLAGS_VRAM + : KFD_IOC_ALLOC_MEM_FLAGS_GTT; if (bo->flags & AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED) - *flags |= ALLOC_MEM_FLAGS_PUBLIC; + *flags |= KFD_IOC_ALLOC_MEM_FLAGS_PUBLIC; } out_put: diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h index b0ad3be0b03f..13feb313e9b3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h @@ -242,6 +242,9 @@ int amdgpu_amdkfd_gpuvm_import_dmabuf(struct kgd_dev *kgd, void amdgpu_amdkfd_gpuvm_init_mem_limits(void); void amdgpu_amdkfd_unreserve_memory_limit(struct amdgpu_bo *bo); +int amdgpu_amdkfd_get_tile_config(struct kgd_dev *kgd, + struct tile_config *config); + /* KGD2KFD callbacks */ int kgd2kfd_init(void); void kgd2kfd_exit(void); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c index 4bcc175a149d..6529caca88fe 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c @@ -79,7 +79,7 @@ static uint32_t get_sdma_rlc_reg_offset(struct amdgpu_device *adev, dev_warn(adev->dev, "Invalid sdma engine id (%d), using engine id 0\n", engine_id); - /* fall through */ + fallthrough; case 0: sdma_engine_reg_base = SOC15_REG_OFFSET(SDMA0, 0, mmSDMA0_RLC0_RB_CNTL) - mmSDMA0_RLC0_RB_CNTL; @@ -319,7 +319,6 @@ const struct kfd2kgd_calls arcturus_kfd2kgd = { .address_watch_get_offset = kgd_gfx_v9_address_watch_get_offset, .get_atc_vmid_pasid_mapping_info = kgd_gfx_v9_get_atc_vmid_pasid_mapping_info, - .get_tile_config = kgd_gfx_v9_get_tile_config, .set_vm_context_page_table_base = kgd_set_vm_context_page_table_base, .get_hive_id = amdgpu_amdkfd_get_hive_id, }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c index ca91fffb8a36..4ec6d0c03201 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c @@ -42,38 +42,6 @@ enum hqd_dequeue_request_type { SAVE_WAVES }; -/* Because of REG_GET_FIELD() being used, we put this function in the - * asic specific file. - */ -static int amdgpu_amdkfd_get_tile_config(struct kgd_dev *kgd, - struct tile_config *config) -{ - struct amdgpu_device *adev = (struct amdgpu_device *)kgd; - - config->gb_addr_config = adev->gfx.config.gb_addr_config; -#if 0 -/* TODO - confirm REG_GET_FIELD x2, should be OK as is... but - * MC_ARB_RAMCFG register doesn't exist on Vega10 - initial amdgpu - * changes commented out related code, doing the same here for now but - * need to sync with Ken et al - */ - config->num_banks = REG_GET_FIELD(adev->gfx.config.mc_arb_ramcfg, - MC_ARB_RAMCFG, NOOFBANK); - config->num_ranks = REG_GET_FIELD(adev->gfx.config.mc_arb_ramcfg, - MC_ARB_RAMCFG, NOOFRANKS); -#endif - - config->tile_config_ptr = adev->gfx.config.tile_mode_array; - config->num_tile_configs = - ARRAY_SIZE(adev->gfx.config.tile_mode_array); - config->macro_tile_config_ptr = - adev->gfx.config.macrotile_mode_array; - config->num_macro_tile_configs = - ARRAY_SIZE(adev->gfx.config.macrotile_mode_array); - - return 0; -} - static inline struct amdgpu_device *get_amdgpu_device(struct kgd_dev *kgd) { return (struct amdgpu_device *)kgd; @@ -805,7 +773,6 @@ const struct kfd2kgd_calls gfx_v10_kfd2kgd = { .address_watch_get_offset = kgd_address_watch_get_offset, .get_atc_vmid_pasid_mapping_info = get_atc_vmid_pasid_mapping_info, - .get_tile_config = amdgpu_amdkfd_get_tile_config, .set_vm_context_page_table_base = set_vm_context_page_table_base, .get_hive_id = amdgpu_amdkfd_get_hive_id, .get_unique_id = amdgpu_amdkfd_get_unique_id, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c index 8f052e98a3c6..0b7e78748540 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c @@ -84,31 +84,6 @@ union TCP_WATCH_CNTL_BITS { float f32All; }; -/* Because of REG_GET_FIELD() being used, we put this function in the - * asic specific file. - */ -static int get_tile_config(struct kgd_dev *kgd, - struct tile_config *config) -{ - struct amdgpu_device *adev = (struct amdgpu_device *)kgd; - - config->gb_addr_config = adev->gfx.config.gb_addr_config; - config->num_banks = REG_GET_FIELD(adev->gfx.config.mc_arb_ramcfg, - MC_ARB_RAMCFG, NOOFBANK); - config->num_ranks = REG_GET_FIELD(adev->gfx.config.mc_arb_ramcfg, - MC_ARB_RAMCFG, NOOFRANKS); - - config->tile_config_ptr = adev->gfx.config.tile_mode_array; - config->num_tile_configs = - ARRAY_SIZE(adev->gfx.config.tile_mode_array); - config->macro_tile_config_ptr = - adev->gfx.config.macrotile_mode_array; - config->num_macro_tile_configs = - ARRAY_SIZE(adev->gfx.config.macrotile_mode_array); - - return 0; -} - static inline struct amdgpu_device *get_amdgpu_device(struct kgd_dev *kgd) { return (struct amdgpu_device *)kgd; @@ -730,7 +705,6 @@ const struct kfd2kgd_calls gfx_v7_kfd2kgd = { .address_watch_get_offset = kgd_address_watch_get_offset, .get_atc_vmid_pasid_mapping_info = get_atc_vmid_pasid_mapping_info, .set_scratch_backing_va = set_scratch_backing_va, - .get_tile_config = get_tile_config, .set_vm_context_page_table_base = set_vm_context_page_table_base, .read_vmid_from_vmfault_reg = read_vmid_from_vmfault_reg, }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c index 19a10db93d68..ccd635b812b5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c @@ -41,31 +41,6 @@ enum hqd_dequeue_request_type { RESET_WAVES }; -/* Because of REG_GET_FIELD() being used, we put this function in the - * asic specific file. - */ -static int get_tile_config(struct kgd_dev *kgd, - struct tile_config *config) -{ - struct amdgpu_device *adev = (struct amdgpu_device *)kgd; - - config->gb_addr_config = adev->gfx.config.gb_addr_config; - config->num_banks = REG_GET_FIELD(adev->gfx.config.mc_arb_ramcfg, - MC_ARB_RAMCFG, NOOFBANK); - config->num_ranks = REG_GET_FIELD(adev->gfx.config.mc_arb_ramcfg, - MC_ARB_RAMCFG, NOOFRANKS); - - config->tile_config_ptr = adev->gfx.config.tile_mode_array; - config->num_tile_configs = - ARRAY_SIZE(adev->gfx.config.tile_mode_array); - config->macro_tile_config_ptr = - adev->gfx.config.macrotile_mode_array; - config->num_macro_tile_configs = - ARRAY_SIZE(adev->gfx.config.macrotile_mode_array); - - return 0; -} - static inline struct amdgpu_device *get_amdgpu_device(struct kgd_dev *kgd) { return (struct amdgpu_device *)kgd; @@ -676,6 +651,5 @@ const struct kfd2kgd_calls gfx_v8_kfd2kgd = { .get_atc_vmid_pasid_mapping_info = get_atc_vmid_pasid_mapping_info, .set_scratch_backing_va = set_scratch_backing_va, - .get_tile_config = get_tile_config, .set_vm_context_page_table_base = set_vm_context_page_table_base, }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c index 7f91feff7c4f..df841c2ac5e7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c @@ -48,28 +48,6 @@ enum hqd_dequeue_request_type { RESET_WAVES }; - -/* Because of REG_GET_FIELD() being used, we put this function in the - * asic specific file. - */ -int kgd_gfx_v9_get_tile_config(struct kgd_dev *kgd, - struct tile_config *config) -{ - struct amdgpu_device *adev = (struct amdgpu_device *)kgd; - - config->gb_addr_config = adev->gfx.config.gb_addr_config; - - config->tile_config_ptr = adev->gfx.config.tile_mode_array; - config->num_tile_configs = - ARRAY_SIZE(adev->gfx.config.tile_mode_array); - config->macro_tile_config_ptr = - adev->gfx.config.macrotile_mode_array; - config->num_macro_tile_configs = - ARRAY_SIZE(adev->gfx.config.macrotile_mode_array); - - return 0; -} - static inline struct amdgpu_device *get_amdgpu_device(struct kgd_dev *kgd) { return (struct amdgpu_device *)kgd; @@ -736,7 +714,6 @@ const struct kfd2kgd_calls gfx_v9_kfd2kgd = { .address_watch_get_offset = kgd_gfx_v9_address_watch_get_offset, .get_atc_vmid_pasid_mapping_info = kgd_gfx_v9_get_atc_vmid_pasid_mapping_info, - .get_tile_config = kgd_gfx_v9_get_tile_config, .set_vm_context_page_table_base = kgd_gfx_v9_set_vm_context_page_table_base, .get_hive_id = amdgpu_amdkfd_get_hive_id, .get_unique_id = amdgpu_amdkfd_get_unique_id, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h index 63d3e6683dfe..aedf67d57449 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h @@ -60,5 +60,3 @@ uint32_t kgd_gfx_v9_address_watch_get_offset(struct kgd_dev *kgd, bool kgd_gfx_v9_get_atc_vmid_pasid_mapping_info(struct kgd_dev *kgd, uint8_t vmid, uint16_t *p_pasid); -int kgd_gfx_v9_get_tile_config(struct kgd_dev *kgd, - struct tile_config *config); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c index e1d1eed7a25f..9dff792c9290 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c @@ -29,6 +29,7 @@ #include "amdgpu_vm.h" #include "amdgpu_amdkfd.h" #include "amdgpu_dma_buf.h" +#include <uapi/linux/kfd_ioctl.h> /* BO flag to indicate a KFD userptr BO */ #define AMDGPU_AMDKFD_USERPTR_BO (1ULL << 63) @@ -400,18 +401,18 @@ static int vm_update_pds(struct amdgpu_vm *vm, struct amdgpu_sync *sync) static uint64_t get_pte_flags(struct amdgpu_device *adev, struct kgd_mem *mem) { struct amdgpu_device *bo_adev = amdgpu_ttm_adev(mem->bo->tbo.bdev); - bool coherent = mem->alloc_flags & ALLOC_MEM_FLAGS_COHERENT; + bool coherent = mem->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_COHERENT; uint32_t mapping_flags; mapping_flags = AMDGPU_VM_PAGE_READABLE; - if (mem->alloc_flags & ALLOC_MEM_FLAGS_WRITABLE) + if (mem->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_WRITABLE) mapping_flags |= AMDGPU_VM_PAGE_WRITEABLE; - if (mem->alloc_flags & ALLOC_MEM_FLAGS_EXECUTABLE) + if (mem->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_EXECUTABLE) mapping_flags |= AMDGPU_VM_PAGE_EXECUTABLE; switch (adev->asic_type) { case CHIP_ARCTURUS: - if (mem->alloc_flags & ALLOC_MEM_FLAGS_VRAM) { + if (mem->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_VRAM) { if (bo_adev == adev) mapping_flags |= coherent ? AMDGPU_VM_MTYPE_CC : AMDGPU_VM_MTYPE_RW; @@ -1160,24 +1161,24 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu( /* * Check on which domain to allocate BO */ - if (flags & ALLOC_MEM_FLAGS_VRAM) { + if (flags & KFD_IOC_ALLOC_MEM_FLAGS_VRAM) { domain = alloc_domain = AMDGPU_GEM_DOMAIN_VRAM; alloc_flags = AMDGPU_GEM_CREATE_VRAM_WIPE_ON_RELEASE; - alloc_flags |= (flags & ALLOC_MEM_FLAGS_PUBLIC) ? + alloc_flags |= (flags & KFD_IOC_ALLOC_MEM_FLAGS_PUBLIC) ? AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED : AMDGPU_GEM_CREATE_NO_CPU_ACCESS; - } else if (flags & ALLOC_MEM_FLAGS_GTT) { + } else if (flags & KFD_IOC_ALLOC_MEM_FLAGS_GTT) { domain = alloc_domain = AMDGPU_GEM_DOMAIN_GTT; alloc_flags = 0; - } else if (flags & ALLOC_MEM_FLAGS_USERPTR) { + } else if (flags & KFD_IOC_ALLOC_MEM_FLAGS_USERPTR) { domain = AMDGPU_GEM_DOMAIN_GTT; alloc_domain = AMDGPU_GEM_DOMAIN_CPU; alloc_flags = 0; if (!offset || !*offset) return -EINVAL; user_addr = untagged_addr(*offset); - } else if (flags & (ALLOC_MEM_FLAGS_DOORBELL | - ALLOC_MEM_FLAGS_MMIO_REMAP)) { + } else if (flags & (KFD_IOC_ALLOC_MEM_FLAGS_DOORBELL | + KFD_IOC_ALLOC_MEM_FLAGS_MMIO_REMAP)) { domain = AMDGPU_GEM_DOMAIN_GTT; alloc_domain = AMDGPU_GEM_DOMAIN_CPU; bo_type = ttm_bo_type_sg; @@ -1198,7 +1199,7 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu( } INIT_LIST_HEAD(&(*mem)->bo_va_list); mutex_init(&(*mem)->lock); - (*mem)->aql_queue = !!(flags & ALLOC_MEM_FLAGS_AQL_QUEUE_MEM); + (*mem)->aql_queue = !!(flags & KFD_IOC_ALLOC_MEM_FLAGS_AQL_QUEUE_MEM); /* Workaround for AQL queue wraparound bug. Map the same * memory twice. That means we only actually allocate half @@ -1680,10 +1681,12 @@ int amdgpu_amdkfd_gpuvm_import_dmabuf(struct kgd_dev *kgd, INIT_LIST_HEAD(&(*mem)->bo_va_list); mutex_init(&(*mem)->lock); + (*mem)->alloc_flags = ((bo->preferred_domains & AMDGPU_GEM_DOMAIN_VRAM) ? - ALLOC_MEM_FLAGS_VRAM : ALLOC_MEM_FLAGS_GTT) | - ALLOC_MEM_FLAGS_WRITABLE | ALLOC_MEM_FLAGS_EXECUTABLE; + KFD_IOC_ALLOC_MEM_FLAGS_VRAM : KFD_IOC_ALLOC_MEM_FLAGS_GTT) + | KFD_IOC_ALLOC_MEM_FLAGS_WRITABLE + | KFD_IOC_ALLOC_MEM_FLAGS_EXECUTABLE; (*mem)->bo = amdgpu_bo_ref(bo); (*mem)->va = va; @@ -2242,3 +2245,25 @@ int amdgpu_amdkfd_remove_gws_from_process(void *info, void *mem) kfree(mem); return 0; } + +/* Returns GPU-specific tiling mode information */ +int amdgpu_amdkfd_get_tile_config(struct kgd_dev *kgd, + struct tile_config *config) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)kgd; + + config->gb_addr_config = adev->gfx.config.gb_addr_config; + config->tile_config_ptr = adev->gfx.config.tile_mode_array; + config->num_tile_configs = + ARRAY_SIZE(adev->gfx.config.tile_mode_array); + config->macro_tile_config_ptr = + adev->gfx.config.macrotile_mode_array; + config->num_macro_tile_configs = + ARRAY_SIZE(adev->gfx.config.macrotile_mode_array); + + /* Those values are not set from GFX9 onwards */ + config->num_banks = adev->gfx.config.num_banks; + config->num_ranks = adev->gfx.config.num_ranks; + + return 0; +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c index 438d10ae343b..af91627b19b0 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c @@ -1208,7 +1208,6 @@ static int amdgpu_cs_submit(struct amdgpu_cs_parser *p, struct amdgpu_fpriv *fpriv = p->filp->driver_priv; struct drm_sched_entity *entity = p->entity; enum drm_sched_priority priority; - struct amdgpu_ring *ring; struct amdgpu_bo_list_entry *e; struct amdgpu_job *job; uint64_t seq; @@ -1261,9 +1260,6 @@ static int amdgpu_cs_submit(struct amdgpu_cs_parser *p, priority = job->base.s_priority; drm_sched_entity_push_job(&job->base, entity); - ring = to_amdgpu_ring(entity->rq->sched); - amdgpu_ring_priority_get(ring, priority); - amdgpu_vm_move_to_lru_tail(p->adev, &fpriv->vm); ttm_eu_fence_buffer_objects(&p->ticket, &p->validated, p->fence); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c index 94a6c42f29ea..6ed36a2c5f73 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c @@ -61,12 +61,24 @@ static int amdgpu_ctx_priority_permit(struct drm_file *filp, return -EACCES; } +static enum gfx_pipe_priority amdgpu_ctx_sched_prio_to_compute_prio(enum drm_sched_priority prio) +{ + switch (prio) { + case DRM_SCHED_PRIORITY_HIGH_HW: + case DRM_SCHED_PRIORITY_KERNEL: + return AMDGPU_GFX_PIPE_PRIO_HIGH; + default: + return AMDGPU_GFX_PIPE_PRIO_NORMAL; + } +} + static int amdgpu_ctx_init_entity(struct amdgpu_ctx *ctx, const u32 hw_ip, const u32 ring) { struct amdgpu_device *adev = ctx->adev; struct amdgpu_ctx_entity *entity; struct drm_gpu_scheduler **scheds = NULL, *sched = NULL; unsigned num_scheds = 0; + enum gfx_pipe_priority hw_prio; enum drm_sched_priority priority; int r; @@ -79,46 +91,51 @@ static int amdgpu_ctx_init_entity(struct amdgpu_ctx *ctx, const u32 hw_ip, const priority = (ctx->override_priority == DRM_SCHED_PRIORITY_UNSET) ? ctx->init_priority : ctx->override_priority; switch (hw_ip) { - case AMDGPU_HW_IP_GFX: - sched = &adev->gfx.gfx_ring[0].sched; - scheds = &sched; - num_scheds = 1; - break; - case AMDGPU_HW_IP_COMPUTE: - scheds = adev->gfx.compute_sched; - num_scheds = adev->gfx.num_compute_sched; - break; - case AMDGPU_HW_IP_DMA: - scheds = adev->sdma.sdma_sched; - num_scheds = adev->sdma.num_sdma_sched; - break; - case AMDGPU_HW_IP_UVD: - sched = &adev->uvd.inst[0].ring.sched; - scheds = &sched; - num_scheds = 1; - break; - case AMDGPU_HW_IP_VCE: - sched = &adev->vce.ring[0].sched; - scheds = &sched; - num_scheds = 1; - break; - case AMDGPU_HW_IP_UVD_ENC: - sched = &adev->uvd.inst[0].ring_enc[0].sched; - scheds = &sched; - num_scheds = 1; - break; - case AMDGPU_HW_IP_VCN_DEC: - scheds = adev->vcn.vcn_dec_sched; - num_scheds = adev->vcn.num_vcn_dec_sched; - break; - case AMDGPU_HW_IP_VCN_ENC: - scheds = adev->vcn.vcn_enc_sched; - num_scheds = adev->vcn.num_vcn_enc_sched; - break; - case AMDGPU_HW_IP_VCN_JPEG: - scheds = adev->jpeg.jpeg_sched; - num_scheds = adev->jpeg.num_jpeg_sched; - break; + case AMDGPU_HW_IP_GFX: + sched = &adev->gfx.gfx_ring[0].sched; + scheds = &sched; + num_scheds = 1; + break; + case AMDGPU_HW_IP_COMPUTE: + hw_prio = amdgpu_ctx_sched_prio_to_compute_prio(priority); + scheds = adev->gfx.compute_prio_sched[hw_prio]; + num_scheds = adev->gfx.num_compute_sched[hw_prio]; + break; + case AMDGPU_HW_IP_DMA: + scheds = adev->sdma.sdma_sched; + num_scheds = adev->sdma.num_sdma_sched; + break; + case AMDGPU_HW_IP_UVD: + sched = &adev->uvd.inst[0].ring.sched; + scheds = &sched; + num_scheds = 1; + break; + case AMDGPU_HW_IP_VCE: + sched = &adev->vce.ring[0].sched; + scheds = &sched; + num_scheds = 1; + break; + case AMDGPU_HW_IP_UVD_ENC: + sched = &adev->uvd.inst[0].ring_enc[0].sched; + scheds = &sched; + num_scheds = 1; + break; + case AMDGPU_HW_IP_VCN_DEC: + sched = drm_sched_pick_best(adev->vcn.vcn_dec_sched, + adev->vcn.num_vcn_dec_sched); + scheds = &sched; + num_scheds = 1; + break; + case AMDGPU_HW_IP_VCN_ENC: + sched = drm_sched_pick_best(adev->vcn.vcn_enc_sched, + adev->vcn.num_vcn_enc_sched); + scheds = &sched; + num_scheds = 1; + break; + case AMDGPU_HW_IP_VCN_JPEG: + scheds = adev->jpeg.jpeg_sched; + num_scheds = adev->jpeg.num_jpeg_sched; + break; } r = drm_sched_entity_init(&entity->entity, priority, scheds, num_scheds, @@ -502,6 +519,29 @@ struct dma_fence *amdgpu_ctx_get_fence(struct amdgpu_ctx *ctx, return fence; } +static void amdgpu_ctx_set_entity_priority(struct amdgpu_ctx *ctx, + struct amdgpu_ctx_entity *aentity, + int hw_ip, + enum drm_sched_priority priority) +{ + struct amdgpu_device *adev = ctx->adev; + enum gfx_pipe_priority hw_prio; + struct drm_gpu_scheduler **scheds = NULL; + unsigned num_scheds; + + /* set sw priority */ + drm_sched_entity_set_priority(&aentity->entity, priority); + + /* set hw priority */ + if (hw_ip == AMDGPU_HW_IP_COMPUTE) { + hw_prio = amdgpu_ctx_sched_prio_to_compute_prio(priority); + scheds = adev->gfx.compute_prio_sched[hw_prio]; + num_scheds = adev->gfx.num_compute_sched[hw_prio]; + drm_sched_entity_modify_sched(&aentity->entity, scheds, + num_scheds); + } +} + void amdgpu_ctx_priority_override(struct amdgpu_ctx *ctx, enum drm_sched_priority priority) { @@ -514,13 +554,11 @@ void amdgpu_ctx_priority_override(struct amdgpu_ctx *ctx, ctx->init_priority : ctx->override_priority; for (i = 0; i < AMDGPU_HW_IP_NUM; ++i) { for (j = 0; j < amdgpu_ctx_num_entities[i]; ++j) { - struct drm_sched_entity *entity; - if (!ctx->entities[i][j]) continue; - entity = &ctx->entities[i][j]->entity; - drm_sched_entity_set_priority(entity, ctx_prio); + amdgpu_ctx_set_entity_priority(ctx, ctx->entities[i][j], + i, ctx_prio); } } } @@ -628,20 +666,53 @@ void amdgpu_ctx_mgr_fini(struct amdgpu_ctx_mgr *mgr) mutex_destroy(&mgr->lock); } + +static void amdgpu_ctx_init_compute_sched(struct amdgpu_device *adev) +{ + int num_compute_sched_normal = 0; + int num_compute_sched_high = AMDGPU_MAX_COMPUTE_RINGS - 1; + int i; + + /* use one drm sched array, gfx.compute_sched to store both high and + * normal priority drm compute schedulers */ + for (i = 0; i < adev->gfx.num_compute_rings; i++) { + if (!adev->gfx.compute_ring[i].has_high_prio) + adev->gfx.compute_sched[num_compute_sched_normal++] = + &adev->gfx.compute_ring[i].sched; + else + adev->gfx.compute_sched[num_compute_sched_high--] = + &adev->gfx.compute_ring[i].sched; + } + + /* compute ring only has two priority for now */ + i = AMDGPU_GFX_PIPE_PRIO_NORMAL; + adev->gfx.compute_prio_sched[i] = &adev->gfx.compute_sched[0]; + adev->gfx.num_compute_sched[i] = num_compute_sched_normal; + + i = AMDGPU_GFX_PIPE_PRIO_HIGH; + if (num_compute_sched_high == (AMDGPU_MAX_COMPUTE_RINGS - 1)) { + /* When compute has no high priority rings then use */ + /* normal priority sched array */ + adev->gfx.compute_prio_sched[i] = &adev->gfx.compute_sched[0]; + adev->gfx.num_compute_sched[i] = num_compute_sched_normal; + } else { + adev->gfx.compute_prio_sched[i] = + &adev->gfx.compute_sched[num_compute_sched_high - 1]; + adev->gfx.num_compute_sched[i] = + adev->gfx.num_compute_rings - num_compute_sched_normal; + } +} + void amdgpu_ctx_init_sched(struct amdgpu_device *adev) { int i, j; + amdgpu_ctx_init_compute_sched(adev); for (i = 0; i < adev->gfx.num_gfx_rings; i++) { adev->gfx.gfx_sched[i] = &adev->gfx.gfx_ring[i].sched; adev->gfx.num_gfx_sched++; } - for (i = 0; i < adev->gfx.num_compute_rings; i++) { - adev->gfx.compute_sched[i] = &adev->gfx.compute_ring[i].sched; - adev->gfx.num_compute_sched++; - } - for (i = 0; i < adev->sdma.num_instances; i++) { adev->sdma.sdma_sched[i] = &adev->sdma.instance[i].ring.sched; adev->sdma.num_sdma_sched++; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c index 3bb74056b9d2..c0f9a651dc06 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c @@ -33,6 +33,7 @@ #include "amdgpu.h" #include "amdgpu_pm.h" #include "amdgpu_dm_debugfs.h" +#include "amdgpu_ras.h" /** * amdgpu_debugfs_add_files - Add simple debugfs entries @@ -178,7 +179,7 @@ static int amdgpu_debugfs_process_reg_op(bool read, struct file *f, } else { r = get_user(value, (uint32_t *)buf); if (!r) - WREG32(*pos >> 2, value); + amdgpu_mm_wreg_mmio_rlc(adev, *pos >> 2, value, 0); } if (r) { result = r; @@ -783,11 +784,11 @@ static ssize_t amdgpu_debugfs_gpr_read(struct file *f, char __user *buf, ssize_t result = 0; uint32_t offset, se, sh, cu, wave, simd, thread, bank, *data; - if (size & 3 || *pos & 3) + if (size > 4096 || size & 3 || *pos & 3) return -EINVAL; /* decode offset */ - offset = *pos & GENMASK_ULL(11, 0); + offset = (*pos & GENMASK_ULL(11, 0)) >> 2; se = (*pos & GENMASK_ULL(19, 12)) >> 12; sh = (*pos & GENMASK_ULL(27, 20)) >> 20; cu = (*pos & GENMASK_ULL(35, 28)) >> 28; @@ -825,7 +826,7 @@ static ssize_t amdgpu_debugfs_gpr_read(struct file *f, char __user *buf, while (size) { uint32_t value; - value = data[offset++]; + value = data[result >> 2]; r = put_user(value, (uint32_t *)buf); if (r) { result = r; @@ -992,18 +993,6 @@ int amdgpu_debugfs_regs_init(struct amdgpu_device *adev) return 0; } -void amdgpu_debugfs_regs_cleanup(struct amdgpu_device *adev) -{ - unsigned i; - - for (i = 0; i < ARRAY_SIZE(debugfs_regs); i++) { - if (adev->debugfs_regs[i]) { - debugfs_remove(adev->debugfs_regs[i]); - adev->debugfs_regs[i] = NULL; - } - } -} - static int amdgpu_debugfs_test_ib(struct seq_file *m, void *data) { struct drm_info_node *node = (struct drm_info_node *) m->private; @@ -1269,9 +1258,43 @@ failure: return 0; } +static int amdgpu_debugfs_sclk_set(void *data, u64 val) +{ + int ret = 0; + uint32_t max_freq, min_freq; + struct amdgpu_device *adev = (struct amdgpu_device *)data; + + if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev)) + return -EINVAL; + + ret = pm_runtime_get_sync(adev->ddev->dev); + if (ret < 0) + return ret; + + if (is_support_sw_smu(adev)) { + ret = smu_get_dpm_freq_range(&adev->smu, SMU_SCLK, &min_freq, &max_freq, true); + if (ret || val > max_freq || val < min_freq) + return -EINVAL; + ret = smu_set_soft_freq_range(&adev->smu, SMU_SCLK, (uint32_t)val, (uint32_t)val, true); + } else { + return 0; + } + + pm_runtime_mark_last_busy(adev->ddev->dev); + pm_runtime_put_autosuspend(adev->ddev->dev); + + if (ret) + return -EINVAL; + + return 0; +} + DEFINE_SIMPLE_ATTRIBUTE(fops_ib_preempt, NULL, amdgpu_debugfs_ib_preempt, "%llu\n"); +DEFINE_SIMPLE_ATTRIBUTE(fops_sclk_set, NULL, + amdgpu_debugfs_sclk_set, "%llu\n"); + int amdgpu_debugfs_init(struct amdgpu_device *adev) { int r, i; @@ -1285,6 +1308,15 @@ int amdgpu_debugfs_init(struct amdgpu_device *adev) return -EIO; } + adev->smu.debugfs_sclk = + debugfs_create_file("amdgpu_force_sclk", 0200, + adev->ddev->primary->debugfs_root, adev, + &fops_sclk_set); + if (!(adev->smu.debugfs_sclk)) { + DRM_ERROR("unable to create amdgpu_set_sclk debugsfs file\n"); + return -EIO; + } + /* Register debugfs entries for amdgpu_ttm */ r = amdgpu_ttm_debugfs_init(adev); if (r) { @@ -1335,35 +1367,19 @@ int amdgpu_debugfs_init(struct amdgpu_device *adev) } } + amdgpu_ras_debugfs_create_all(adev); + return amdgpu_debugfs_add_files(adev, amdgpu_debugfs_list, ARRAY_SIZE(amdgpu_debugfs_list)); } -void amdgpu_debugfs_fini(struct amdgpu_device *adev) -{ - int i; - - for (i = 0; i < AMDGPU_MAX_RINGS; ++i) { - struct amdgpu_ring *ring = adev->rings[i]; - - if (!ring) - continue; - - amdgpu_debugfs_ring_fini(ring); - } - amdgpu_ttm_debugfs_fini(adev); - debugfs_remove(adev->debugfs_preempt); -} - #else int amdgpu_debugfs_init(struct amdgpu_device *adev) { return 0; } -void amdgpu_debugfs_fini(struct amdgpu_device *adev) { } int amdgpu_debugfs_regs_init(struct amdgpu_device *adev) { return 0; } -void amdgpu_debugfs_regs_cleanup(struct amdgpu_device *adev) { } #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.h index b382527e359a..de12d1101526 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.h @@ -32,7 +32,6 @@ struct amdgpu_debugfs { }; int amdgpu_debugfs_regs_init(struct amdgpu_device *adev); -void amdgpu_debugfs_regs_cleanup(struct amdgpu_device *adev); int amdgpu_debugfs_init(struct amdgpu_device *adev); void amdgpu_debugfs_fini(struct amdgpu_device *adev); int amdgpu_debugfs_add_files(struct amdgpu_device *adev, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index a3aaf2e3794c..faa3e7102156 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -306,6 +306,26 @@ void amdgpu_mm_wreg8(struct amdgpu_device *adev, uint32_t offset, uint8_t value) BUG(); } +void static inline amdgpu_mm_wreg_mmio(struct amdgpu_device *adev, uint32_t reg, uint32_t v, uint32_t acc_flags) +{ + trace_amdgpu_mm_wreg(adev->pdev->device, reg, v); + + if ((reg * 4) < adev->rmmio_size && !(acc_flags & AMDGPU_REGS_IDX)) + writel(v, ((void __iomem *)adev->rmmio) + (reg * 4)); + else { + unsigned long flags; + + spin_lock_irqsave(&adev->mmio_idx_lock, flags); + writel((reg * 4), ((void __iomem *)adev->rmmio) + (mmMM_INDEX * 4)); + writel(v, ((void __iomem *)adev->rmmio) + (mmMM_DATA * 4)); + spin_unlock_irqrestore(&adev->mmio_idx_lock, flags); + } + + if (adev->asic_type >= CHIP_VEGA10 && reg == 1 && adev->last_mm_index == 0x5702C) { + udelay(500); + } +} + /** * amdgpu_mm_wreg - write to a memory mapped IO register * @@ -319,8 +339,6 @@ void amdgpu_mm_wreg8(struct amdgpu_device *adev, uint32_t offset, uint8_t value) void amdgpu_mm_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v, uint32_t acc_flags) { - trace_amdgpu_mm_wreg(adev->pdev->device, reg, v); - if (adev->asic_type >= CHIP_VEGA10 && reg == 0) { adev->last_mm_index = v; } @@ -328,20 +346,26 @@ void amdgpu_mm_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v, if ((acc_flags & AMDGPU_REGS_KIQ) || (!(acc_flags & AMDGPU_REGS_NO_KIQ) && amdgpu_sriov_runtime(adev))) return amdgpu_kiq_wreg(adev, reg, v); - if ((reg * 4) < adev->rmmio_size && !(acc_flags & AMDGPU_REGS_IDX)) - writel(v, ((void __iomem *)adev->rmmio) + (reg * 4)); - else { - unsigned long flags; + amdgpu_mm_wreg_mmio(adev, reg, v, acc_flags); +} - spin_lock_irqsave(&adev->mmio_idx_lock, flags); - writel((reg * 4), ((void __iomem *)adev->rmmio) + (mmMM_INDEX * 4)); - writel(v, ((void __iomem *)adev->rmmio) + (mmMM_DATA * 4)); - spin_unlock_irqrestore(&adev->mmio_idx_lock, flags); - } +/* + * amdgpu_mm_wreg_mmio_rlc - write register either with mmio or with RLC path if in range + * + * this function is invoked only the debugfs register access + * */ +void amdgpu_mm_wreg_mmio_rlc(struct amdgpu_device *adev, uint32_t reg, uint32_t v, + uint32_t acc_flags) +{ + if (amdgpu_sriov_fullaccess(adev) && + adev->gfx.rlc.funcs && + adev->gfx.rlc.funcs->is_rlcg_access_range) { - if (adev->asic_type >= CHIP_VEGA10 && reg == 1 && adev->last_mm_index == 0x5702C) { - udelay(500); + if (adev->gfx.rlc.funcs->is_rlcg_access_range(adev, reg)) + return adev->gfx.rlc.funcs->rlcg_wreg(adev, reg, v); } + + amdgpu_mm_wreg_mmio(adev, reg, v, acc_flags); } /** @@ -2718,6 +2742,9 @@ static void amdgpu_device_xgmi_reset_func(struct work_struct *__work) if (adev->asic_reset_res) goto fail; + + if (adev->mmhub.funcs && adev->mmhub.funcs->reset_ras_error_count) + adev->mmhub.funcs->reset_ras_error_count(adev); } else { task_barrier_full(&hive->tb); @@ -3193,6 +3220,12 @@ void amdgpu_device_fini(struct amdgpu_device *adev) flush_delayed_work(&adev->delayed_init_work); adev->shutdown = true; + /* make sure IB test finished before entering exclusive mode + * to avoid preemption on IB test + * */ + if (amdgpu_sriov_vf(adev)) + amdgpu_virt_request_full_gpu(adev, false); + /* disable all interrupts */ amdgpu_irq_disable_all(adev); if (adev->mode_info.mode_config_initialized){ @@ -3235,7 +3268,6 @@ void amdgpu_device_fini(struct amdgpu_device *adev) adev->rmmio = NULL; amdgpu_device_doorbell_fini(adev); - amdgpu_debugfs_regs_cleanup(adev); device_remove_file(adev->dev, &dev_attr_pcie_replay_count); if (adev->ucode_sysfs_en) amdgpu_ucode_sysfs_fini(adev); @@ -3881,8 +3913,15 @@ static int amdgpu_do_asic_reset(struct amdgpu_hive_info *hive, } } - if (!r && amdgpu_ras_intr_triggered()) + if (!r && amdgpu_ras_intr_triggered()) { + list_for_each_entry(tmp_adev, device_list_handle, gmc.xgmi.head) { + if (tmp_adev->mmhub.funcs && + tmp_adev->mmhub.funcs->reset_ras_error_count) + tmp_adev->mmhub.funcs->reset_ras_error_count(tmp_adev); + } + amdgpu_ras_intr_cleared(); + } list_for_each_entry(tmp_adev, device_list_handle, gmc.xgmi.head) { if (need_full_reset) { @@ -3928,6 +3967,8 @@ static int amdgpu_do_asic_reset(struct amdgpu_hive_info *hive, if (r) goto out; + amdgpu_fbdev_set_suspend(tmp_adev, 0); + /* must succeed. */ amdgpu_ras_resume(tmp_adev); @@ -4101,6 +4142,8 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev, */ amdgpu_unregister_gpu_instance(tmp_adev); + amdgpu_fbdev_set_suspend(adev, 1); + /* disable ras on ALL IPs */ if (!(in_ras_intr && !use_baco) && amdgpu_device_ip_need_full_reset(tmp_adev)) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.h index 168579492a55..936d85aa0fbc 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.h @@ -448,6 +448,8 @@ struct amdgpu_pm { /* powerplay feature */ uint32_t pp_feature; + /* Used for I2C access to various EEPROMs on relevant ASICs */ + struct i2c_adapter smu_i2c; }; #define R600_SSTU_DFLT 0 diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index e1c4c2df716b..8ea86ffdea0d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -1121,18 +1121,16 @@ static void amdgpu_pci_remove(struct pci_dev *pdev) { struct drm_device *dev = pci_get_drvdata(pdev); - struct amdgpu_device *adev = dev->dev_private; #ifdef MODULE if (THIS_MODULE->state != MODULE_STATE_GOING) #endif DRM_ERROR("Hotplug removal is not supported\n"); drm_dev_unplug(dev); - drm_dev_put(dev); - amdgpu_debugfs_fini(adev); amdgpu_driver_unload_kms(dev); pci_disable_device(pdev); pci_set_drvdata(pdev, NULL); + drm_dev_put(dev); } static void @@ -1301,24 +1299,55 @@ static int amdgpu_pmops_runtime_idle(struct device *dev) { struct drm_device *drm_dev = dev_get_drvdata(dev); struct amdgpu_device *adev = drm_dev->dev_private; - struct drm_crtc *crtc; + /* we don't want the main rpm_idle to call suspend - we want to autosuspend */ + int ret = 1; if (!adev->runpm) { pm_runtime_forbid(dev); return -EBUSY; } - list_for_each_entry(crtc, &drm_dev->mode_config.crtc_list, head) { - if (crtc->enabled) { - DRM_DEBUG_DRIVER("failing to power off - crtc active\n"); - return -EBUSY; + if (amdgpu_device_has_dc_support(adev)) { + struct drm_crtc *crtc; + + drm_modeset_lock_all(drm_dev); + + drm_for_each_crtc(crtc, drm_dev) { + if (crtc->state->active) { + ret = -EBUSY; + break; + } + } + + drm_modeset_unlock_all(drm_dev); + + } else { + struct drm_connector *list_connector; + struct drm_connector_list_iter iter; + + mutex_lock(&drm_dev->mode_config.mutex); + drm_modeset_lock(&drm_dev->mode_config.connection_mutex, NULL); + + drm_connector_list_iter_begin(drm_dev, &iter); + drm_for_each_connector_iter(list_connector, &iter) { + if (list_connector->dpms == DRM_MODE_DPMS_ON) { + ret = -EBUSY; + break; + } } + + drm_connector_list_iter_end(&iter); + + drm_modeset_unlock(&drm_dev->mode_config.connection_mutex); + mutex_unlock(&drm_dev->mode_config.mutex); } + if (ret == -EBUSY) + DRM_DEBUG_DRIVER("failing to power off - crtc active\n"); + pm_runtime_mark_last_busy(dev); pm_runtime_autosuspend(dev); - /* we don't want the main rpm_idle to call suspend - we want to autosuspend */ - return 1; + return ret; } long amdgpu_drm_ioctl(struct file *filp, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c index 7403588684b3..6b9c9193cdfa 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c @@ -192,6 +192,14 @@ static bool amdgpu_gfx_is_multipipe_capable(struct amdgpu_device *adev) return adev->gfx.mec.num_mec > 1; } +bool amdgpu_gfx_is_high_priority_compute_queue(struct amdgpu_device *adev, + int queue) +{ + /* Policy: make queue 0 of each pipe as high priority compute queue */ + return (queue == 0); + +} + void amdgpu_gfx_compute_queue_acquire(struct amdgpu_device *adev) { int i, queue, pipe, mec; @@ -565,7 +573,6 @@ int amdgpu_gfx_ras_late_init(struct amdgpu_device *adev) int r; struct ras_fs_if fs_info = { .sysfs_name = "gfx_err_count", - .debugfs_name = "gfx_err_inject", }; struct ras_ih_if ih_info = { .cb = amdgpu_gfx_process_ras_data_cb, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h index ca17ffb01301..5825692d07e4 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h @@ -41,6 +41,15 @@ #define AMDGPU_MAX_GFX_QUEUES KGD_MAX_QUEUES #define AMDGPU_MAX_COMPUTE_QUEUES KGD_MAX_QUEUES +enum gfx_pipe_priority { + AMDGPU_GFX_PIPE_PRIO_NORMAL = 1, + AMDGPU_GFX_PIPE_PRIO_HIGH, + AMDGPU_GFX_PIPE_PRIO_MAX +}; + +#define AMDGPU_GFX_QUEUE_PRIORITY_MINIMUM 0 +#define AMDGPU_GFX_QUEUE_PRIORITY_MAXIMUM 15 + struct amdgpu_mec { struct amdgpu_bo *hpd_eop_obj; u64 hpd_eop_gpu_addr; @@ -151,6 +160,8 @@ struct amdgpu_gfx_config { unsigned num_gpus; unsigned multi_gpu_tile_size; unsigned mc_arb_ramcfg; + unsigned num_banks; + unsigned num_ranks; unsigned gb_addr_config; unsigned num_rbs; unsigned gs_vgt_table_depth; @@ -204,6 +215,7 @@ struct amdgpu_gfx_funcs { u32 queue, u32 vmid); int (*ras_error_inject)(struct amdgpu_device *adev, void *inject_if); int (*query_ras_error_count) (struct amdgpu_device *adev, void *ras_error_status); + void (*reset_ras_error_count) (struct amdgpu_device *adev); }; struct sq_work { @@ -278,8 +290,9 @@ struct amdgpu_gfx { uint32_t num_gfx_sched; unsigned num_gfx_rings; struct amdgpu_ring compute_ring[AMDGPU_MAX_COMPUTE_RINGS]; + struct drm_gpu_scheduler **compute_prio_sched[AMDGPU_GFX_PIPE_PRIO_MAX]; struct drm_gpu_scheduler *compute_sched[AMDGPU_MAX_COMPUTE_RINGS]; - uint32_t num_compute_sched; + uint32_t num_compute_sched[AMDGPU_GFX_PIPE_PRIO_MAX]; unsigned num_compute_rings; struct amdgpu_irq_src eop_irq; struct amdgpu_irq_src priv_reg_irq; @@ -361,6 +374,8 @@ void amdgpu_gfx_bit_to_mec_queue(struct amdgpu_device *adev, int bit, int *mec, int *pipe, int *queue); bool amdgpu_gfx_is_mec_queue_enabled(struct amdgpu_device *adev, int mec, int pipe, int queue); +bool amdgpu_gfx_is_high_priority_compute_queue(struct amdgpu_device *adev, + int queue); int amdgpu_gfx_me_queue_to_bit(struct amdgpu_device *adev, int me, int pipe, int queue); void amdgpu_gfx_bit_to_me_queue(struct amdgpu_device *adev, int bit, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c index d42be880a236..4981e443a884 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c @@ -117,12 +117,10 @@ void amdgpu_job_free_resources(struct amdgpu_job *job) static void amdgpu_job_free_cb(struct drm_sched_job *s_job) { - struct amdgpu_ring *ring = to_amdgpu_ring(s_job->sched); struct amdgpu_job *job = to_amdgpu_job(s_job); drm_sched_job_cleanup(s_job); - amdgpu_ring_priority_put(ring, s_job->s_priority); dma_fence_put(job->fence); amdgpu_sync_free(&job->sync); amdgpu_sync_free(&job->sched_sync); @@ -143,7 +141,6 @@ int amdgpu_job_submit(struct amdgpu_job *job, struct drm_sched_entity *entity, void *owner, struct dma_fence **f) { enum drm_sched_priority priority; - struct amdgpu_ring *ring; int r; if (!f) @@ -158,9 +155,6 @@ int amdgpu_job_submit(struct amdgpu_job *job, struct drm_sched_entity *entity, priority = job->base.s_priority; drm_sched_entity_push_job(&job->base, entity); - ring = to_amdgpu_ring(entity->rq->sched); - amdgpu_ring_priority_get(ring, priority); - return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c index 6a1b62bc3dcf..fd1dc3236eca 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c @@ -88,9 +88,6 @@ void amdgpu_driver_unload_kms(struct drm_device *dev) if (adev->rmmio == NULL) goto done_free; - if (amdgpu_sriov_vf(adev)) - amdgpu_virt_request_full_gpu(adev, false); - if (adev->runpm) { pm_runtime_get_sync(dev->dev); pm_runtime_forbid(dev->dev); @@ -175,6 +172,7 @@ int amdgpu_driver_load_kms(struct drm_device *dev, unsigned long flags) else if (amdgpu_device_supports_baco(dev) && (amdgpu_runtime_pm != 0) && (adev->asic_type >= CHIP_TOPAZ) && + (adev->asic_type != CHIP_VEGA10) && (adev->asic_type != CHIP_VEGA20) && (adev->asic_type != CHIP_ARCTURUS)) /* enable runpm on VI+ */ adev->runpm = true; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.c index 676c48c02d77..ead3dc572ec5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.c @@ -32,7 +32,6 @@ int amdgpu_mmhub_ras_late_init(struct amdgpu_device *adev) }; struct ras_fs_if fs_info = { .sysfs_name = "mmhub_err_count", - .debugfs_name = "mmhub_err_inject", }; if (!adev->mmhub.ras_if) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.h index 1cd78940cf82..e89fb35fec71 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.h @@ -26,6 +26,7 @@ struct amdgpu_mmhub_funcs { int (*ras_late_init)(struct amdgpu_device *adev); void (*query_ras_error_count)(struct amdgpu_device *adev, void *ras_error_status); + void (*reset_ras_error_count)(struct amdgpu_device *adev); }; struct amdgpu_mmhub { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.c index 7d5c3a9de9ea..6201a5f4b4fa 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.c @@ -30,7 +30,6 @@ int amdgpu_nbio_ras_late_init(struct amdgpu_device *adev) }; struct ras_fs_if fs_info = { .sysfs_name = "pcie_bif_err_count", - .debugfs_name = "pcie_bif_err_inject", }; if (!adev->nbio.ras_if) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c index 1791c084787d..c687f5415b3f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c @@ -1319,7 +1319,8 @@ void amdgpu_bo_release_notify(struct ttm_buffer_object *bo) amdgpu_amdkfd_unreserve_memory_limit(abo); /* We only remove the fence if the resv has individualized. */ - WARN_ON_ONCE(bo->base.resv != &bo->base._resv); + WARN_ON_ONCE(bo->type == ttm_bo_type_kernel + && bo->base.resv != &bo->base._resv); if (bo->base.resv == &bo->base._resv) amdgpu_amdkfd_remove_fence_on_pt_pd_bos(abo); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c index bc3cf04a1a94..f197f1be0969 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c @@ -92,6 +92,9 @@ void amdgpu_pm_acpi_event_handler(struct amdgpu_device *adev) if (adev->powerplay.pp_funcs->enable_bapm) amdgpu_dpm_enable_bapm(adev, adev->pm.ac_power); mutex_unlock(&adev->pm.mutex); + + if (is_support_sw_smu(adev)) + smu_set_ac_dc(&adev->smu); } } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c index d33f74100094..be50867ea644 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c @@ -24,6 +24,7 @@ */ #include <linux/firmware.h> +#include <linux/dma-mapping.h> #include "amdgpu.h" #include "amdgpu_psp.h" @@ -38,6 +39,9 @@ static void psp_set_funcs(struct amdgpu_device *adev); +static int psp_sysfs_init(struct amdgpu_device *adev); +static void psp_sysfs_fini(struct amdgpu_device *adev); + /* * Due to DF Cstate management centralized to PMFW, the firmware * loading sequence will be updated as below: @@ -136,6 +140,13 @@ static int psp_sw_init(void *handle) return ret; } + if (adev->asic_type == CHIP_NAVI10) { + ret= psp_sysfs_init(adev); + if (ret) { + return ret; + } + } + return 0; } @@ -152,6 +163,10 @@ static int psp_sw_fini(void *handle) release_firmware(adev->psp.ta_fw); adev->psp.ta_fw = NULL; } + + if (adev->asic_type == CHIP_NAVI10) + psp_sysfs_fini(adev); + return 0; } @@ -185,6 +200,7 @@ psp_cmd_submit_buf(struct psp_context *psp, int ret; int index; int timeout = 2000; + bool ras_intr = false; mutex_lock(&psp->mutex); @@ -209,7 +225,8 @@ psp_cmd_submit_buf(struct psp_context *psp, * because gpu reset thread triggered and lock resource should * be released for psp resume sequence. */ - if (amdgpu_ras_intr_triggered()) + ras_intr = amdgpu_ras_intr_triggered(); + if (ras_intr) break; msleep(1); amdgpu_asic_invalidate_hdp(psp->adev, NULL); @@ -222,7 +239,7 @@ psp_cmd_submit_buf(struct psp_context *psp, * during psp initialization to avoid breaking hw_init and it doesn't * return -EINVAL. */ - if (psp->cmd_buf_mem->resp.status || !timeout) { + if ((psp->cmd_buf_mem->resp.status || !timeout) && !ras_intr) { if (ucode) DRM_WARN("failed to load ucode id (%d) ", ucode->ucode_id); @@ -1816,6 +1833,97 @@ static int psp_set_powergating_state(void *handle, return 0; } +static ssize_t psp_usbc_pd_fw_sysfs_read(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + struct drm_device *ddev = dev_get_drvdata(dev); + struct amdgpu_device *adev = ddev->dev_private; + uint32_t fw_ver; + int ret; + + if (!adev->ip_blocks[AMD_IP_BLOCK_TYPE_PSP].status.late_initialized) { + DRM_INFO("PSP block is not ready yet."); + return -EBUSY; + } + + mutex_lock(&adev->psp.mutex); + ret = psp_read_usbc_pd_fw(&adev->psp, &fw_ver); + mutex_unlock(&adev->psp.mutex); + + if (ret) { + DRM_ERROR("Failed to read USBC PD FW, err = %d", ret); + return ret; + } + + return snprintf(buf, PAGE_SIZE, "%x\n", fw_ver); +} + +static ssize_t psp_usbc_pd_fw_sysfs_write(struct device *dev, + struct device_attribute *attr, + const char *buf, + size_t count) +{ + struct drm_device *ddev = dev_get_drvdata(dev); + struct amdgpu_device *adev = ddev->dev_private; + void *cpu_addr; + dma_addr_t dma_addr; + int ret; + char fw_name[100]; + const struct firmware *usbc_pd_fw; + + if (!adev->ip_blocks[AMD_IP_BLOCK_TYPE_PSP].status.late_initialized) { + DRM_INFO("PSP block is not ready yet."); + return -EBUSY; + } + + snprintf(fw_name, sizeof(fw_name), "amdgpu/%s", buf); + ret = request_firmware(&usbc_pd_fw, fw_name, adev->dev); + if (ret) + goto fail; + + /* We need contiguous physical mem to place the FW for psp to access */ + cpu_addr = dma_alloc_coherent(adev->dev, usbc_pd_fw->size, &dma_addr, GFP_KERNEL); + + ret = dma_mapping_error(adev->dev, dma_addr); + if (ret) + goto rel_buf; + + memcpy_toio(cpu_addr, usbc_pd_fw->data, usbc_pd_fw->size); + + /* + * x86 specific workaround. + * Without it the buffer is invisible in PSP. + * + * TODO Remove once PSP starts snooping CPU cache + */ +#ifdef CONFIG_X86 + clflush_cache_range(cpu_addr, (usbc_pd_fw->size & ~(L1_CACHE_BYTES - 1))); +#endif + + mutex_lock(&adev->psp.mutex); + ret = psp_load_usbc_pd_fw(&adev->psp, dma_addr); + mutex_unlock(&adev->psp.mutex); + +rel_buf: + dma_free_coherent(adev->dev, usbc_pd_fw->size, cpu_addr, dma_addr); + release_firmware(usbc_pd_fw); + +fail: + if (ret) { + DRM_ERROR("Failed to load USBC PD FW, err = %d", ret); + return ret; + } + + return count; +} + +static DEVICE_ATTR(usbc_pd_fw, S_IRUGO | S_IWUSR, + psp_usbc_pd_fw_sysfs_read, + psp_usbc_pd_fw_sysfs_write); + + + const struct amd_ip_funcs psp_ip_funcs = { .name = "psp", .early_init = psp_early_init, @@ -1834,6 +1942,21 @@ const struct amd_ip_funcs psp_ip_funcs = { .set_powergating_state = psp_set_powergating_state, }; +static int psp_sysfs_init(struct amdgpu_device *adev) +{ + int ret = device_create_file(adev->dev, &dev_attr_usbc_pd_fw); + + if (ret) + DRM_ERROR("Failed to create USBC PD FW control file!"); + + return ret; +} + +static void psp_sysfs_fini(struct amdgpu_device *adev) +{ + device_remove_file(adev->dev, &dev_attr_usbc_pd_fw); +} + static const struct amdgpu_psp_funcs psp_funcs = { .check_fw_loading_status = psp_check_fw_loading_status, }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h index 37fa184f27f6..297435c0c7c1 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h @@ -114,6 +114,8 @@ struct psp_funcs int (*mem_training)(struct psp_context *psp, uint32_t ops); uint32_t (*ring_get_wptr)(struct psp_context *psp); void (*ring_set_wptr)(struct psp_context *psp, uint32_t value); + int (*load_usbc_pd_fw)(struct psp_context *psp, dma_addr_t dma_addr); + int (*read_usbc_pd_fw)(struct psp_context *psp, uint32_t *fw_ver); }; #define AMDGPU_XGMI_MAX_CONNECTED_NODES 64 @@ -351,6 +353,14 @@ struct amdgpu_psp_funcs { #define psp_ring_get_wptr(psp) (psp)->funcs->ring_get_wptr((psp)) #define psp_ring_set_wptr(psp, value) (psp)->funcs->ring_set_wptr((psp), (value)) +#define psp_load_usbc_pd_fw(psp, dma_addr) \ + ((psp)->funcs->load_usbc_pd_fw ? \ + (psp)->funcs->load_usbc_pd_fw((psp), (dma_addr)) : -EINVAL) + +#define psp_read_usbc_pd_fw(psp, fw_ver) \ + ((psp)->funcs->read_usbc_pd_fw ? \ + (psp)->funcs->read_usbc_pd_fw((psp), fw_ver) : -EINVAL) + extern const struct amd_ip_funcs psp_ip_funcs; extern const struct amdgpu_ip_block_version psp_v3_1_ip_block; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c index 930633a0ed64..3c32a94d2424 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c @@ -281,6 +281,11 @@ static ssize_t amdgpu_ras_debugfs_ctrl_write(struct file *f, const char __user * struct ras_debug_if data; int ret = 0; + if (amdgpu_ras_intr_triggered()) { + DRM_WARN("RAS WARN: error injection currently inaccessible\n"); + return size; + } + ret = amdgpu_ras_debugfs_ctrl_parse_data(f, buf, size, pos, &data); if (ret) return -EINVAL; @@ -394,6 +399,10 @@ static ssize_t amdgpu_ras_sysfs_read(struct device *dev, .head = obj->head, }; + if (amdgpu_ras_intr_triggered()) + return snprintf(buf, PAGE_SIZE, + "Query currently inaccessible\n"); + if (amdgpu_ras_error_query(obj->adev, &info)) return -EINVAL; @@ -721,6 +730,9 @@ int amdgpu_ras_error_query(struct amdgpu_device *adev, if (adev->nbio.funcs->query_ras_error_count) adev->nbio.funcs->query_ras_error_count(adev, &err_data); break; + case AMDGPU_RAS_BLOCK__XGMI_WAFL: + amdgpu_xgmi_query_ras_error_count(adev, &err_data); + break; default: break; } @@ -1110,6 +1122,32 @@ void amdgpu_ras_debugfs_create(struct amdgpu_device *adev, &amdgpu_ras_debugfs_ops); } +void amdgpu_ras_debugfs_create_all(struct amdgpu_device *adev) +{ + struct amdgpu_ras *con = amdgpu_ras_get_context(adev); + struct ras_manager *obj; + struct ras_fs_if fs_info; + + /* + * it won't be called in resume path, no need to check + * suspend and gpu reset status + */ + if (!con) + return; + + amdgpu_ras_debugfs_create_ctrl_node(adev); + + list_for_each_entry(obj, &con->head, node) { + if (amdgpu_ras_is_supported(adev, obj->head.block) && + (obj->attr_inuse == 1)) { + sprintf(fs_info.debugfs_name, "%s_err_inject", + ras_block_str(obj->head.block)); + fs_info.head = obj->head; + amdgpu_ras_debugfs_create(adev, &fs_info); + } + } +} + void amdgpu_ras_debugfs_remove(struct amdgpu_device *adev, struct ras_common_if *head) { @@ -1142,7 +1180,6 @@ static void amdgpu_ras_debugfs_remove_all(struct amdgpu_device *adev) static int amdgpu_ras_fs_init(struct amdgpu_device *adev) { amdgpu_ras_sysfs_create_feature_node(adev); - amdgpu_ras_debugfs_create_ctrl_node(adev); return 0; } @@ -1734,18 +1771,30 @@ static void amdgpu_ras_check_supported(struct amdgpu_device *adev, *hw_supported = 0; *supported = 0; - if (amdgpu_sriov_vf(adev) || + if (amdgpu_sriov_vf(adev) || !adev->is_atom_fw || (adev->asic_type != CHIP_VEGA20 && adev->asic_type != CHIP_ARCTURUS)) return; - if (adev->is_atom_fw && - (amdgpu_atomfirmware_mem_ecc_supported(adev) || - amdgpu_atomfirmware_sram_ecc_supported(adev))) - *hw_supported = AMDGPU_RAS_BLOCK_MASK; + if (amdgpu_atomfirmware_mem_ecc_supported(adev)) { + DRM_INFO("HBM ECC is active.\n"); + *hw_supported |= (1 << AMDGPU_RAS_BLOCK__UMC | + 1 << AMDGPU_RAS_BLOCK__DF); + } else + DRM_INFO("HBM ECC is not presented.\n"); + + if (amdgpu_atomfirmware_sram_ecc_supported(adev)) { + DRM_INFO("SRAM ECC is active.\n"); + *hw_supported |= ~(1 << AMDGPU_RAS_BLOCK__UMC | + 1 << AMDGPU_RAS_BLOCK__DF); + } else + DRM_INFO("SRAM ECC is not presented.\n"); + + /* hw_supported needs to be aligned with RAS block mask. */ + *hw_supported &= AMDGPU_RAS_BLOCK_MASK; *supported = amdgpu_ras_enable == 0 ? - 0 : *hw_supported & amdgpu_ras_mask; + 0 : *hw_supported & amdgpu_ras_mask; } int amdgpu_ras_init(struct amdgpu_device *adev) @@ -1846,8 +1895,6 @@ int amdgpu_ras_late_init(struct amdgpu_device *adev, goto interrupt; } - amdgpu_ras_debugfs_create(adev, fs_info); - r = amdgpu_ras_sysfs_create(adev, fs_info); if (r) goto sysfs; @@ -1856,7 +1903,6 @@ int amdgpu_ras_late_init(struct amdgpu_device *adev, cleanup: amdgpu_ras_sysfs_remove(adev, ras_block); sysfs: - amdgpu_ras_debugfs_remove(adev, ras_block); if (ih_info->cb) amdgpu_ras_interrupt_remove_handler(adev, ih_info); interrupt: @@ -1873,7 +1919,6 @@ void amdgpu_ras_late_fini(struct amdgpu_device *adev, return; amdgpu_ras_sysfs_remove(adev, ras_block); - amdgpu_ras_debugfs_remove(adev, ras_block); if (ih_info->cb) amdgpu_ras_interrupt_remove_handler(adev, ih_info); amdgpu_ras_feature_enable(adev, ras_block, 0); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h index a5fe29a9373e..55c3eceb390d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h @@ -592,6 +592,8 @@ int amdgpu_ras_sysfs_remove(struct amdgpu_device *adev, void amdgpu_ras_debugfs_create(struct amdgpu_device *adev, struct ras_fs_if *head); +void amdgpu_ras_debugfs_create_all(struct amdgpu_device *adev); + void amdgpu_ras_debugfs_remove(struct amdgpu_device *adev, struct ras_common_if *head); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c index ed15b1fa5e98..c0096097bbcf 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c @@ -25,7 +25,6 @@ #include "amdgpu.h" #include "amdgpu_ras.h" #include <linux/bits.h> -#include "smu_v11_0_i2c.h" #include "atom.h" #define EEPROM_I2C_TARGET_ADDR_VEGA20 0xA0 @@ -124,6 +123,7 @@ static int __update_table_header(struct amdgpu_ras_eeprom_control *control, unsigned char *buff) { int ret = 0; + struct amdgpu_device *adev = to_amdgpu_device(control); struct i2c_msg msg = { .addr = 0, .flags = 0, @@ -137,7 +137,7 @@ static int __update_table_header(struct amdgpu_ras_eeprom_control *control, msg.addr = control->i2c_address; - ret = i2c_transfer(&control->eeprom_accessor, &msg, 1); + ret = i2c_transfer(&adev->pm.smu_i2c, &msg, 1); if (ret < 1) DRM_ERROR("Failed to write EEPROM table header, ret:%d", ret); @@ -251,33 +251,18 @@ int amdgpu_ras_eeprom_init(struct amdgpu_ras_eeprom_control *control) .buf = buff, }; + /* Verify i2c adapter is initialized */ + if (!adev->pm.smu_i2c.algo) + return -ENOENT; + if (!__get_eeprom_i2c_addr(adev, &control->i2c_address)) return -EINVAL; mutex_init(&control->tbl_mutex); - switch (adev->asic_type) { - case CHIP_VEGA20: - ret = smu_v11_0_i2c_eeprom_control_init(&control->eeprom_accessor); - break; - - case CHIP_ARCTURUS: - ret = smu_i2c_eeprom_init(&adev->smu, &control->eeprom_accessor); - break; - - default: - return 0; - } - - if (ret) { - DRM_ERROR("Failed to init I2C controller, ret:%d", ret); - return ret; - } - msg.addr = control->i2c_address; - /* Read/Create table header from EEPROM address 0 */ - ret = i2c_transfer(&control->eeprom_accessor, &msg, 1); + ret = i2c_transfer(&adev->pm.smu_i2c, &msg, 1); if (ret < 1) { DRM_ERROR("Failed to read EEPROM table header, ret:%d", ret); return ret; @@ -303,23 +288,6 @@ int amdgpu_ras_eeprom_init(struct amdgpu_ras_eeprom_control *control) return ret == 1 ? 0 : -EIO; } -void amdgpu_ras_eeprom_fini(struct amdgpu_ras_eeprom_control *control) -{ - struct amdgpu_device *adev = to_amdgpu_device(control); - - switch (adev->asic_type) { - case CHIP_VEGA20: - smu_v11_0_i2c_eeprom_control_fini(&control->eeprom_accessor); - break; - case CHIP_ARCTURUS: - smu_i2c_eeprom_fini(&adev->smu, &control->eeprom_accessor); - break; - - default: - return; - } -} - static void __encode_table_record_to_buff(struct amdgpu_ras_eeprom_control *control, struct eeprom_table_record *record, unsigned char *buff) @@ -476,7 +444,7 @@ int amdgpu_ras_eeprom_process_recods(struct amdgpu_ras_eeprom_control *control, control->next_addr += EEPROM_TABLE_RECORD_SIZE; } - ret = i2c_transfer(&control->eeprom_accessor, msgs, num); + ret = i2c_transfer(&adev->pm.smu_i2c, msgs, num); if (ret < 1) { DRM_ERROR("Failed to process EEPROM table records, ret:%d", ret); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h index ca78f812d436..7e8647a05df7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h @@ -44,7 +44,6 @@ struct amdgpu_ras_eeprom_table_header { struct amdgpu_ras_eeprom_control { struct amdgpu_ras_eeprom_table_header tbl_hdr; - struct i2c_adapter eeprom_accessor; uint32_t next_addr; unsigned int num_recs; struct mutex tbl_mutex; @@ -79,7 +78,6 @@ struct eeprom_table_record { }__attribute__((__packed__)); int amdgpu_ras_eeprom_init(struct amdgpu_ras_eeprom_control *control); -void amdgpu_ras_eeprom_fini(struct amdgpu_ras_eeprom_control *control); int amdgpu_ras_eeprom_reset_table(struct amdgpu_ras_eeprom_control *control); int amdgpu_ras_eeprom_process_recods(struct amdgpu_ras_eeprom_control *control, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c index 18e11b0fdc3e..a7e1d0425ed0 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c @@ -151,76 +151,6 @@ void amdgpu_ring_undo(struct amdgpu_ring *ring) } /** - * amdgpu_ring_priority_put - restore a ring's priority - * - * @ring: amdgpu_ring structure holding the information - * @priority: target priority - * - * Release a request for executing at @priority - */ -void amdgpu_ring_priority_put(struct amdgpu_ring *ring, - enum drm_sched_priority priority) -{ - int i; - - if (!ring->funcs->set_priority) - return; - - if (atomic_dec_return(&ring->num_jobs[priority]) > 0) - return; - - /* no need to restore if the job is already at the lowest priority */ - if (priority == DRM_SCHED_PRIORITY_NORMAL) - return; - - mutex_lock(&ring->priority_mutex); - /* something higher prio is executing, no need to decay */ - if (ring->priority > priority) - goto out_unlock; - - /* decay priority to the next level with a job available */ - for (i = priority; i >= DRM_SCHED_PRIORITY_MIN; i--) { - if (i == DRM_SCHED_PRIORITY_NORMAL - || atomic_read(&ring->num_jobs[i])) { - ring->priority = i; - ring->funcs->set_priority(ring, i); - break; - } - } - -out_unlock: - mutex_unlock(&ring->priority_mutex); -} - -/** - * amdgpu_ring_priority_get - change the ring's priority - * - * @ring: amdgpu_ring structure holding the information - * @priority: target priority - * - * Request a ring's priority to be raised to @priority (refcounted). - */ -void amdgpu_ring_priority_get(struct amdgpu_ring *ring, - enum drm_sched_priority priority) -{ - if (!ring->funcs->set_priority) - return; - - if (atomic_inc_return(&ring->num_jobs[priority]) <= 0) - return; - - mutex_lock(&ring->priority_mutex); - if (priority <= ring->priority) - goto out_unlock; - - ring->priority = priority; - ring->funcs->set_priority(ring, priority); - -out_unlock: - mutex_unlock(&ring->priority_mutex); -} - -/** * amdgpu_ring_init - init driver ring struct. * * @adev: amdgpu_device pointer @@ -499,13 +429,6 @@ int amdgpu_debugfs_ring_init(struct amdgpu_device *adev, return 0; } -void amdgpu_debugfs_ring_fini(struct amdgpu_ring *ring) -{ -#if defined(CONFIG_DEBUG_FS) - debugfs_remove(ring->ent); -#endif -} - /** * amdgpu_ring_test_helper - tests ring and set sched readiness status * diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h index 24caff085d00..9a443013d70d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h @@ -167,9 +167,6 @@ struct amdgpu_ring_funcs { uint32_t reg0, uint32_t reg1, uint32_t ref, uint32_t mask); void (*emit_tmz)(struct amdgpu_ring *ring, bool start); - /* priority functions */ - void (*set_priority) (struct amdgpu_ring *ring, - enum drm_sched_priority priority); /* Try to soft recover the ring to make the fence signal */ void (*soft_recovery)(struct amdgpu_ring *ring, unsigned vmid); int (*preempt_ib)(struct amdgpu_ring *ring); @@ -222,6 +219,7 @@ struct amdgpu_ring { struct mutex priority_mutex; /* protected by priority_mutex */ int priority; + bool has_high_prio; #if defined(CONFIG_DEBUG_FS) struct dentry *ent; @@ -258,10 +256,6 @@ void amdgpu_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count); void amdgpu_ring_generic_pad_ib(struct amdgpu_ring *ring, struct amdgpu_ib *ib); void amdgpu_ring_commit(struct amdgpu_ring *ring); void amdgpu_ring_undo(struct amdgpu_ring *ring); -void amdgpu_ring_priority_get(struct amdgpu_ring *ring, - enum drm_sched_priority priority); -void amdgpu_ring_priority_put(struct amdgpu_ring *ring, - enum drm_sched_priority priority); int amdgpu_ring_init(struct amdgpu_device *adev, struct amdgpu_ring *ring, unsigned ring_size, struct amdgpu_irq_src *irq_src, unsigned irq_type); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h index d3d4707f2168..60bb3e8b3118 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h @@ -126,6 +126,9 @@ struct amdgpu_rlc_funcs { void (*stop)(struct amdgpu_device *adev); void (*reset)(struct amdgpu_device *adev); void (*start)(struct amdgpu_device *adev); + void (*update_spm_vmid)(struct amdgpu_device *adev, unsigned vmid); + void (*rlcg_wreg)(struct amdgpu_device *adev, u32 offset, u32 v); + bool (*is_rlcg_access_range)(struct amdgpu_device *adev, uint32_t reg); }; struct amdgpu_rlc { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c index 7854c053e85d..250a309e4dee 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c @@ -93,7 +93,6 @@ int amdgpu_sdma_ras_late_init(struct amdgpu_device *adev, struct ras_ih_if *ih_info = (struct ras_ih_if *)ras_ih_info; struct ras_fs_if fs_info = { .sysfs_name = "sdma_err_count", - .debugfs_name = "sdma_err_inject", }; if (!ih_info) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h index 485335267d78..4b352206354b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h @@ -56,6 +56,7 @@ struct amdgpu_sdma_ras_funcs { void (*ras_fini)(struct amdgpu_device *adev); int (*query_ras_error_count)(struct amdgpu_device *adev, uint32_t instance, void *ras_error_status); + void (*reset_ras_error_count)(struct amdgpu_device *adev); }; struct amdgpu_sdma { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c index fe131c21e8a3..9f44ba7d9d97 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c @@ -968,7 +968,7 @@ static int amdgpu_ttm_tt_pin_userptr(struct ttm_tt *ttm) /* Map SG to device */ r = -ENOMEM; nents = dma_map_sg(adev->dev, ttm->sg->sgl, ttm->sg->nents, direction); - if (nents != ttm->sg->nents) + if (nents == 0) goto release_sg; /* convert SG to linear array of pages and dma addresses */ @@ -1028,7 +1028,7 @@ int amdgpu_ttm_gart_bind(struct amdgpu_device *adev, struct amdgpu_ttm_tt *gtt = (void *)ttm; int r; - if (abo->flags & AMDGPU_GEM_CREATE_MQD_GFX9) { + if (abo->flags & AMDGPU_GEM_CREATE_CP_MQD_GFX9) { uint64_t page_idx = 1; r = amdgpu_gart_bind(adev, gtt->offset, page_idx, @@ -1036,7 +1036,10 @@ int amdgpu_ttm_gart_bind(struct amdgpu_device *adev, if (r) goto gart_bind_fail; - /* Patch mtype of the second part BO */ + /* The memory type of the first page defaults to UC. Now + * modify the memory type to NC from the second page of + * the BO onward. + */ flags &= ~AMDGPU_PTE_MTYPE_VG10_MASK; flags |= AMDGPU_PTE_MTYPE_VG10(AMDGPU_MTYPE_NC); @@ -1837,9 +1840,11 @@ int amdgpu_ttm_init(struct amdgpu_device *adev) *The reserved vram for memory training must be pinned to the specified *place on the VRAM, so reserve it early. */ - r = amdgpu_ttm_training_reserve_vram_init(adev); - if (r) - return r; + if (!amdgpu_sriov_vf(adev)) { + r = amdgpu_ttm_training_reserve_vram_init(adev); + if (r) + return r; + } /* allocate memory as required for VGA * This is used for VGA emulation and pre-OS scanout buffers to @@ -2565,13 +2570,3 @@ int amdgpu_ttm_debugfs_init(struct amdgpu_device *adev) return 0; #endif } - -void amdgpu_ttm_debugfs_fini(struct amdgpu_device *adev) -{ -#if defined(CONFIG_DEBUG_FS) - unsigned i; - - for (i = 0; i < ARRAY_SIZE(ttm_debugfs_entries); i++) - debugfs_remove(adev->mman.debugfs_entries[i]); -#endif -} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h index 7551f3729445..bd05bbb4878d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h @@ -134,6 +134,5 @@ uint64_t amdgpu_ttm_tt_pte_flags(struct amdgpu_device *adev, struct ttm_tt *ttm, struct ttm_mem_reg *mem); int amdgpu_ttm_debugfs_init(struct amdgpu_device *adev); -void amdgpu_ttm_debugfs_fini(struct amdgpu_device *adev); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c index f4d40855147b..9dd51f0d2c11 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c @@ -28,7 +28,6 @@ int amdgpu_umc_ras_late_init(struct amdgpu_device *adev) int r; struct ras_fs_if fs_info = { .sysfs_name = "umc_err_count", - .debugfs_name = "umc_err_inject", }; struct ras_ih_if ih_info = { .cb = amdgpu_umc_process_ras_data_cb, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c index f96464e2c157..a41272fbcba2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c @@ -493,14 +493,9 @@ static int amdgpu_vcn_dec_get_destroy_msg(struct amdgpu_ring *ring, uint32_t han int amdgpu_vcn_dec_ring_test_ib(struct amdgpu_ring *ring, long timeout) { - struct amdgpu_device *adev = ring->adev; struct dma_fence *fence; long r; - /* temporarily disable ib test for sriov */ - if (amdgpu_sriov_vf(adev)) - return 0; - r = amdgpu_vcn_dec_get_create_msg(ring, 1, NULL); if (r) goto error; @@ -527,6 +522,9 @@ int amdgpu_vcn_enc_ring_test_ring(struct amdgpu_ring *ring) unsigned i; int r; + if (amdgpu_sriov_vf(adev)) + return 0; + r = amdgpu_ring_alloc(ring, 16); if (r) return r; @@ -656,15 +654,10 @@ err: int amdgpu_vcn_enc_ring_test_ib(struct amdgpu_ring *ring, long timeout) { - struct amdgpu_device *adev = ring->adev; struct dma_fence *fence = NULL; struct amdgpu_bo *bo = NULL; long r; - /* temporarily disable ib test for sriov */ - if (amdgpu_sriov_vf(adev)) - return 0; - r = amdgpu_bo_create_reserved(ring->adev, 128 * 1024, PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM, &bo, NULL, NULL); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h index daaf909d009a..f0128f745bd2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h @@ -270,6 +270,9 @@ struct amdgpu_virt { #define amdgpu_sriov_runtime(adev) \ ((adev)->virt.caps & AMDGPU_SRIOV_CAPS_RUNTIME) +#define amdgpu_sriov_fullaccess(adev) \ +(amdgpu_sriov_vf((adev)) && !amdgpu_sriov_runtime((adev))) + #define amdgpu_passthrough(adev) \ ((adev)->virt.caps & AMDGPU_PASSTHROUGH_MODE) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c index f44c26c373a1..6d9252a27916 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c @@ -1080,8 +1080,12 @@ int amdgpu_vm_flush(struct amdgpu_ring *ring, struct amdgpu_job *job, struct dma_fence *fence = NULL; bool pasid_mapping_needed = false; unsigned patch_offset = 0; + bool update_spm_vmid_needed = (job->vm && (job->vm->reserved_vmid[vmhub] != NULL)); int r; + if (update_spm_vmid_needed && adev->gfx.rlc.funcs->update_spm_vmid) + adev->gfx.rlc.funcs->update_spm_vmid(adev, job->vmid); + if (amdgpu_vmid_had_gpu_reset(adev, id)) { gds_switch_needed = true; vm_flush_needed = true; @@ -1442,7 +1446,7 @@ static int amdgpu_vm_update_ptes(struct amdgpu_vm_update_params *params, uint64_t incr, entry_end, pe_start; struct amdgpu_bo *pt; - if (flags & AMDGPU_PTE_VALID) { + if (flags & (AMDGPU_PTE_VALID | AMDGPU_PTE_PRT)) { /* make sure that the page tables covering the * address range are actually allocated */ @@ -1599,14 +1603,11 @@ static int amdgpu_vm_bo_update_mapping(struct amdgpu_device *adev, goto error_unlock; } - if (flags & AMDGPU_PTE_VALID) { + if (flags & (AMDGPU_PTE_VALID | AMDGPU_PTE_PRT)) { struct amdgpu_bo *root = vm->root.base.bo; if (!dma_fence_is_signaled(vm->last_direct)) amdgpu_bo_fence(root, vm->last_direct, true); - - if (!dma_fence_is_signaled(vm->last_delayed)) - amdgpu_bo_fence(root, vm->last_delayed, true); } r = vm->update_funcs->prepare(¶ms, resv, sync_mode); @@ -1714,7 +1715,7 @@ static int amdgpu_vm_bo_split_mapping(struct amdgpu_device *adev, AMDGPU_GPU_PAGES_IN_CPU_PAGE; } - } else if (flags & AMDGPU_PTE_VALID) { + } else if (flags & (AMDGPU_PTE_VALID | AMDGPU_PTE_PRT)) { addr += bo_adev->vm_manager.vram_base_offset; addr += pfn << PAGE_SHIFT; } @@ -2584,8 +2585,7 @@ bool amdgpu_vm_evictable(struct amdgpu_bo *bo) return false; /* Don't evict VM page tables while they are updated */ - if (!dma_fence_is_signaled(bo_base->vm->last_direct) || - !dma_fence_is_signaled(bo_base->vm->last_delayed)) { + if (!dma_fence_is_signaled(bo_base->vm->last_direct)) { amdgpu_vm_eviction_unlock(bo_base->vm); return false; } @@ -2762,11 +2762,7 @@ long amdgpu_vm_wait_idle(struct amdgpu_vm *vm, long timeout) if (timeout <= 0) return timeout; - timeout = dma_fence_wait_timeout(vm->last_direct, true, timeout); - if (timeout <= 0) - return timeout; - - return dma_fence_wait_timeout(vm->last_delayed, true, timeout); + return dma_fence_wait_timeout(vm->last_direct, true, timeout); } /** @@ -2839,7 +2835,6 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm, vm->update_funcs = &amdgpu_vm_sdma_funcs; vm->last_update = NULL; vm->last_direct = dma_fence_get_stub(); - vm->last_delayed = dma_fence_get_stub(); mutex_init(&vm->eviction_lock); vm->evicting = false; @@ -2894,7 +2889,6 @@ error_free_root: error_free_delayed: dma_fence_put(vm->last_direct); - dma_fence_put(vm->last_delayed); drm_sched_entity_destroy(&vm->delayed); error_free_direct: @@ -3097,8 +3091,6 @@ void amdgpu_vm_fini(struct amdgpu_device *adev, struct amdgpu_vm *vm) dma_fence_wait(vm->last_direct, false); dma_fence_put(vm->last_direct); - dma_fence_wait(vm->last_delayed, false); - dma_fence_put(vm->last_delayed); list_for_each_entry_safe(mapping, tmp, &vm->freed, list) { if (mapping->flags & AMDGPU_PTE_PRT && prt_fini_needed) { @@ -3209,6 +3201,7 @@ int amdgpu_vm_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) union drm_amdgpu_vm *args = data; struct amdgpu_device *adev = dev->dev_private; struct amdgpu_fpriv *fpriv = filp->driver_priv; + long timeout = msecs_to_jiffies(2000); int r; switch (args->in.op) { @@ -3220,6 +3213,21 @@ int amdgpu_vm_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) return r; break; case AMDGPU_VM_OP_UNRESERVE_VMID: + if (amdgpu_sriov_runtime(adev)) + timeout = 8 * timeout; + + /* Wait vm idle to make sure the vmid set in SPM_VMID is + * not referenced anymore. + */ + r = amdgpu_bo_reserve(fpriv->vm.root.base.bo, true); + if (r) + return r; + + r = amdgpu_vm_wait_idle(&fpriv->vm, timeout); + if (r < 0) + return r; + + amdgpu_bo_unreserve(fpriv->vm.root.base.bo); amdgpu_vmid_free_reserved(adev, &fpriv->vm, AMDGPU_GFXHUB_0); break; default: diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h index d00648ee8d54..06fe30e1492d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h @@ -276,7 +276,6 @@ struct amdgpu_vm { /* Last submission to the scheduler entities */ struct dma_fence *last_direct; - struct dma_fence *last_delayed; unsigned int pasid; /* dedicated to vm */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c index 4cc7881f438c..cf96c335b258 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c @@ -104,12 +104,13 @@ static int amdgpu_vm_sdma_commit(struct amdgpu_vm_update_params *p, if (r) goto error; - tmp = dma_fence_get(f); - if (p->direct) + if (p->direct) { + tmp = dma_fence_get(f); swap(p->vm->last_direct, tmp); - else - swap(p->vm->last_delayed, tmp); - dma_fence_put(tmp); + dma_fence_put(tmp); + } else { + dma_resv_add_shared_fence(p->vm->root.base.bo->tbo.base.resv, f); + } if (fence && !p->direct) swap(*fence, f); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c index 7a89c91f7b80..95b3327168ac 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c @@ -26,7 +26,12 @@ #include "amdgpu_xgmi.h" #include "amdgpu_smu.h" #include "amdgpu_ras.h" +#include "soc15.h" #include "df/df_3_6_offset.h" +#include "xgmi/xgmi_4_0_0_smn.h" +#include "xgmi/xgmi_4_0_0_sh_mask.h" +#include "wafl/wafl2_4_0_0_smn.h" +#include "wafl/wafl2_4_0_0_sh_mask.h" static DEFINE_MUTEX(xgmi_mutex); @@ -36,6 +41,109 @@ static DEFINE_MUTEX(xgmi_mutex); static struct amdgpu_hive_info xgmi_hives[AMDGPU_MAX_XGMI_HIVE]; static unsigned hive_count = 0; +static const int xgmi_pcs_err_status_reg_vg20[] = { + smnXGMI0_PCS_GOPX16_PCS_ERROR_STATUS, + smnXGMI0_PCS_GOPX16_PCS_ERROR_STATUS + 0x100000, +}; + +static const int wafl_pcs_err_status_reg_vg20[] = { + smnPCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, + smnPCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS + 0x100000, +}; + +static const int xgmi_pcs_err_status_reg_arct[] = { + smnXGMI0_PCS_GOPX16_PCS_ERROR_STATUS, + smnXGMI0_PCS_GOPX16_PCS_ERROR_STATUS + 0x100000, + smnXGMI0_PCS_GOPX16_PCS_ERROR_STATUS + 0x500000, + smnXGMI0_PCS_GOPX16_PCS_ERROR_STATUS + 0x600000, + smnXGMI0_PCS_GOPX16_PCS_ERROR_STATUS + 0x700000, + smnXGMI0_PCS_GOPX16_PCS_ERROR_STATUS + 0x800000, +}; + +/* same as vg20*/ +static const int wafl_pcs_err_status_reg_arct[] = { + smnPCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, + smnPCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS + 0x100000, +}; + +static const struct amdgpu_pcs_ras_field xgmi_pcs_ras_fields[] = { + {"XGMI PCS DataLossErr", + SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, DataLossErr)}, + {"XGMI PCS TrainingErr", + SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, TrainingErr)}, + {"XGMI PCS CRCErr", + SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, CRCErr)}, + {"XGMI PCS BERExceededErr", + SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, BERExceededErr)}, + {"XGMI PCS TxMetaDataErr", + SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, TxMetaDataErr)}, + {"XGMI PCS ReplayBufParityErr", + SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, ReplayBufParityErr)}, + {"XGMI PCS DataParityErr", + SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, DataParityErr)}, + {"XGMI PCS ReplayFifoOverflowErr", + SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, ReplayFifoOverflowErr)}, + {"XGMI PCS ReplayFifoUnderflowErr", + SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, ReplayFifoUnderflowErr)}, + {"XGMI PCS ElasticFifoOverflowErr", + SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, ElasticFifoOverflowErr)}, + {"XGMI PCS DeskewErr", + SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, DeskewErr)}, + {"XGMI PCS DataStartupLimitErr", + SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, DataStartupLimitErr)}, + {"XGMI PCS FCInitTimeoutErr", + SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, FCInitTimeoutErr)}, + {"XGMI PCS RecoveryTimeoutErr", + SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, RecoveryTimeoutErr)}, + {"XGMI PCS ReadySerialTimeoutErr", + SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, ReadySerialTimeoutErr)}, + {"XGMI PCS ReadySerialAttemptErr", + SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, ReadySerialAttemptErr)}, + {"XGMI PCS RecoveryAttemptErr", + SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, RecoveryAttemptErr)}, + {"XGMI PCS RecoveryRelockAttemptErr", + SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, RecoveryRelockAttemptErr)}, +}; + +static const struct amdgpu_pcs_ras_field wafl_pcs_ras_fields[] = { + {"WAFL PCS DataLossErr", + SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, DataLossErr)}, + {"WAFL PCS TrainingErr", + SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, TrainingErr)}, + {"WAFL PCS CRCErr", + SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, CRCErr)}, + {"WAFL PCS BERExceededErr", + SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, BERExceededErr)}, + {"WAFL PCS TxMetaDataErr", + SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, TxMetaDataErr)}, + {"WAFL PCS ReplayBufParityErr", + SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, ReplayBufParityErr)}, + {"WAFL PCS DataParityErr", + SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, DataParityErr)}, + {"WAFL PCS ReplayFifoOverflowErr", + SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, ReplayFifoOverflowErr)}, + {"WAFL PCS ReplayFifoUnderflowErr", + SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, ReplayFifoUnderflowErr)}, + {"WAFL PCS ElasticFifoOverflowErr", + SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, ElasticFifoOverflowErr)}, + {"WAFL PCS DeskewErr", + SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, DeskewErr)}, + {"WAFL PCS DataStartupLimitErr", + SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, DataStartupLimitErr)}, + {"WAFL PCS FCInitTimeoutErr", + SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, FCInitTimeoutErr)}, + {"WAFL PCS RecoveryTimeoutErr", + SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, RecoveryTimeoutErr)}, + {"WAFL PCS ReadySerialTimeoutErr", + SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, ReadySerialTimeoutErr)}, + {"WAFL PCS ReadySerialAttemptErr", + SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, ReadySerialAttemptErr)}, + {"WAFL PCS RecoveryAttemptErr", + SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, RecoveryAttemptErr)}, + {"WAFL PCS RecoveryRelockAttemptErr", + SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, RecoveryRelockAttemptErr)}, +}; + void *amdgpu_xgmi_hive_try_lock(struct amdgpu_hive_info *hive) { return &hive->device_list; @@ -490,7 +598,6 @@ int amdgpu_xgmi_ras_late_init(struct amdgpu_device *adev) }; struct ras_fs_if fs_info = { .sysfs_name = "xgmi_wafl_err_count", - .debugfs_name = "xgmi_wafl_err_inject", }; if (!adev->gmc.xgmi.supported || @@ -560,3 +667,99 @@ uint64_t amdgpu_xgmi_get_relative_phy_addr(struct amdgpu_device *adev, return addr + dram_base_addr; } + +static int amdgpu_xgmi_query_pcs_error_status(struct amdgpu_device *adev, + uint32_t value, + uint32_t *ue_count, + uint32_t *ce_count, + bool is_xgmi_pcs) +{ + int i; + int ue_cnt; + + if (is_xgmi_pcs) { + /* query xgmi pcs error status, + * only ue is supported */ + for (i = 0; i < ARRAY_SIZE(xgmi_pcs_ras_fields); i ++) { + ue_cnt = (value & + xgmi_pcs_ras_fields[i].pcs_err_mask) >> + xgmi_pcs_ras_fields[i].pcs_err_shift; + if (ue_cnt) { + dev_info(adev->dev, "%s detected\n", + xgmi_pcs_ras_fields[i].err_name); + *ue_count += ue_cnt; + } + } + } else { + /* query wafl pcs error status, + * only ue is supported */ + for (i = 0; i < ARRAY_SIZE(wafl_pcs_ras_fields); i++) { + ue_cnt = (value & + wafl_pcs_ras_fields[i].pcs_err_mask) >> + wafl_pcs_ras_fields[i].pcs_err_shift; + if (ue_cnt) { + dev_info(adev->dev, "%s detected\n", + wafl_pcs_ras_fields[i].err_name); + *ue_count += ue_cnt; + } + } + } + + return 0; +} + +int amdgpu_xgmi_query_ras_error_count(struct amdgpu_device *adev, + void *ras_error_status) +{ + struct ras_err_data *err_data = (struct ras_err_data *)ras_error_status; + int i; + uint32_t data; + uint32_t ue_cnt = 0, ce_cnt = 0; + + if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__XGMI_WAFL)) + return -EINVAL; + + err_data->ue_count = 0; + err_data->ce_count = 0; + + switch (adev->asic_type) { + case CHIP_ARCTURUS: + /* check xgmi pcs error */ + for (i = 0; i < ARRAY_SIZE(xgmi_pcs_err_status_reg_arct); i++) { + data = RREG32_PCIE(xgmi_pcs_err_status_reg_arct[i]); + if (data) + amdgpu_xgmi_query_pcs_error_status(adev, + data, &ue_cnt, &ce_cnt, true); + } + /* check wafl pcs error */ + for (i = 0; i < ARRAY_SIZE(wafl_pcs_err_status_reg_arct); i++) { + data = RREG32_PCIE(wafl_pcs_err_status_reg_arct[i]); + if (data) + amdgpu_xgmi_query_pcs_error_status(adev, + data, &ue_cnt, &ce_cnt, false); + } + break; + case CHIP_VEGA20: + default: + /* check xgmi pcs error */ + for (i = 0; i < ARRAY_SIZE(xgmi_pcs_err_status_reg_vg20); i++) { + data = RREG32_PCIE(xgmi_pcs_err_status_reg_vg20[i]); + if (data) + amdgpu_xgmi_query_pcs_error_status(adev, + data, &ue_cnt, &ce_cnt, true); + } + /* check wafl pcs error */ + for (i = 0; i < ARRAY_SIZE(wafl_pcs_err_status_reg_vg20); i++) { + data = RREG32_PCIE(wafl_pcs_err_status_reg_vg20[i]); + if (data) + amdgpu_xgmi_query_pcs_error_status(adev, + data, &ue_cnt, &ce_cnt, false); + } + break; + } + + err_data->ue_count += ue_cnt; + err_data->ce_count += ce_cnt; + + return 0; +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h index 2aa61adee459..4a92067fe595 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h @@ -37,6 +37,12 @@ struct amdgpu_hive_info { struct task_barrier tb; }; +struct amdgpu_pcs_ras_field { + const char *err_name; + uint32_t pcs_err_mask; + uint32_t pcs_err_shift; +}; + struct amdgpu_hive_info *amdgpu_get_xgmi_hive(struct amdgpu_device *adev, int lock); int amdgpu_xgmi_update_topology(struct amdgpu_hive_info *hive, struct amdgpu_device *adev); int amdgpu_xgmi_add_device(struct amdgpu_device *adev); @@ -48,6 +54,8 @@ int amdgpu_xgmi_ras_late_init(struct amdgpu_device *adev); void amdgpu_xgmi_ras_fini(struct amdgpu_device *adev); uint64_t amdgpu_xgmi_get_relative_phy_addr(struct amdgpu_device *adev, uint64_t addr); +int amdgpu_xgmi_query_ras_error_count(struct amdgpu_device *adev, + void *ras_error_status); static inline bool amdgpu_xgmi_same_hive(struct amdgpu_device *adev, struct amdgpu_device *bo_adev) diff --git a/drivers/gpu/drm/amd/amdgpu/atom.c b/drivers/gpu/drm/amd/amdgpu/atom.c index dd30f4e61a8c..cae426c7c086 100644 --- a/drivers/gpu/drm/amd/amdgpu/atom.c +++ b/drivers/gpu/drm/amd/amdgpu/atom.c @@ -744,8 +744,8 @@ static void atom_op_jump(atom_exec_context *ctx, int *ptr, int arg) cjiffies = jiffies; if (time_after(cjiffies, ctx->last_jump_jiffies)) { cjiffies -= ctx->last_jump_jiffies; - if ((jiffies_to_msecs(cjiffies) > 5000)) { - DRM_ERROR("atombios stuck in loop for more than 5secs aborting\n"); + if ((jiffies_to_msecs(cjiffies) > 10000)) { + DRM_ERROR("atombios stuck in loop for more than 10secs aborting\n"); ctx->abort = true; } } else { diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c index 8b17f182d532..c8f2aa1db13b 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c @@ -35,6 +35,8 @@ #include "gc/gc_10_1_0_offset.h" #include "gc/gc_10_1_0_sh_mask.h" +#include "smuio/smuio_11_0_0_offset.h" +#include "smuio/smuio_11_0_0_sh_mask.h" #include "navi10_enum.h" #include "hdp/hdp_5_0_0_offset.h" #include "ivsrcid/gfx/irqsrcs_gfx_10_1.h" @@ -52,7 +54,7 @@ * 1. Primary ring * 2. Async ring */ -#define GFX10_NUM_GFX_RINGS 2 +#define GFX10_NUM_GFX_RINGS_NV1X 1 #define GFX10_MEC_HPD_SIZE 2048 #define F32_CE_PROGRAM_RAM_SIZE 65536 @@ -222,6 +224,49 @@ static const struct soc15_reg_golden golden_settings_gc_10_1_2[] = SOC15_REG_GOLDEN_VALUE(GC, 0, mmUTCL1_CTRL, 0xffffffff, 0x00800000) }; +static void gfx_v10_rlcg_wreg(struct amdgpu_device *adev, u32 offset, u32 v) +{ + static void *scratch_reg0; + static void *scratch_reg1; + static void *scratch_reg2; + static void *scratch_reg3; + static void *spare_int; + static uint32_t grbm_cntl; + static uint32_t grbm_idx; + uint32_t i = 0; + uint32_t retries = 50000; + + scratch_reg0 = adev->rmmio + (adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG0_BASE_IDX] + mmSCRATCH_REG0)*4; + scratch_reg1 = adev->rmmio + (adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG1_BASE_IDX] + mmSCRATCH_REG1)*4; + scratch_reg2 = adev->rmmio + (adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG1_BASE_IDX] + mmSCRATCH_REG2)*4; + scratch_reg3 = adev->rmmio + (adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG1_BASE_IDX] + mmSCRATCH_REG3)*4; + spare_int = adev->rmmio + (adev->reg_offset[GC_HWIP][0][mmRLC_SPARE_INT_BASE_IDX] + mmRLC_SPARE_INT)*4; + + grbm_cntl = adev->reg_offset[GC_HWIP][0][mmGRBM_GFX_CNTL_BASE_IDX] + mmGRBM_GFX_CNTL; + grbm_idx = adev->reg_offset[GC_HWIP][0][mmGRBM_GFX_INDEX_BASE_IDX] + mmGRBM_GFX_INDEX; + + if (amdgpu_sriov_runtime(adev)) { + pr_err("shouldn't call rlcg write register during runtime\n"); + return; + } + + writel(v, scratch_reg0); + writel(offset | 0x80000000, scratch_reg1); + writel(1, spare_int); + for (i = 0; i < retries; i++) { + u32 tmp; + + tmp = readl(scratch_reg1); + if (!(tmp & 0x80000000)) + break; + + udelay(10); + } + + if (i >= retries) + pr_err("timeout: rlcg program reg:0x%05x failed !\n", offset); +} + static const struct soc15_reg_golden golden_settings_gc_10_1_nv14[] = { /* Pending on emulation bring up */ @@ -500,29 +545,28 @@ static int gfx_v10_0_ring_test_ib(struct amdgpu_ring *ring, long timeout) struct amdgpu_device *adev = ring->adev; struct amdgpu_ib ib; struct dma_fence *f = NULL; - uint32_t scratch; - uint32_t tmp = 0; + unsigned index; + uint64_t gpu_addr; + uint32_t tmp; long r; - r = amdgpu_gfx_scratch_get(adev, &scratch); - if (r) { - DRM_ERROR("amdgpu: failed to get scratch reg (%ld).\n", r); + r = amdgpu_device_wb_get(adev, &index); + if (r) return r; - } - - WREG32(scratch, 0xCAFEDEAD); + gpu_addr = adev->wb.gpu_addr + (index * 4); + adev->wb.wb[index] = cpu_to_le32(0xCAFEDEAD); memset(&ib, 0, sizeof(ib)); - r = amdgpu_ib_get(adev, NULL, 256, &ib); - if (r) { - DRM_ERROR("amdgpu: failed to get ib (%ld).\n", r); + r = amdgpu_ib_get(adev, NULL, 16, &ib); + if (r) goto err1; - } - ib.ptr[0] = PACKET3(PACKET3_SET_UCONFIG_REG, 1); - ib.ptr[1] = ((scratch - PACKET3_SET_UCONFIG_REG_START)); - ib.ptr[2] = 0xDEADBEEF; - ib.length_dw = 3; + ib.ptr[0] = PACKET3(PACKET3_WRITE_DATA, 3); + ib.ptr[1] = WRITE_DATA_DST_SEL(5) | WR_CONFIRM; + ib.ptr[2] = lower_32_bits(gpu_addr); + ib.ptr[3] = upper_32_bits(gpu_addr); + ib.ptr[4] = 0xDEADBEEF; + ib.length_dw = 5; r = amdgpu_ib_schedule(ring, 1, &ib, NULL, &f); if (r) @@ -530,15 +574,13 @@ static int gfx_v10_0_ring_test_ib(struct amdgpu_ring *ring, long timeout) r = dma_fence_wait_timeout(f, false, timeout); if (r == 0) { - DRM_ERROR("amdgpu: IB test timed out.\n"); r = -ETIMEDOUT; goto err2; } else if (r < 0) { - DRM_ERROR("amdgpu: fence wait failed (%ld).\n", r); goto err2; } - tmp = RREG32(scratch); + tmp = adev->wb.wb[index]; if (tmp == 0xDEADBEEF) r = 0; else @@ -547,8 +589,7 @@ err2: amdgpu_ib_free(adev, &ib, NULL); dma_fence_put(f); err1: - amdgpu_gfx_scratch_free(adev, scratch); - + amdgpu_device_wb_free(adev, index); return r; } @@ -1016,6 +1057,10 @@ static int gfx_v10_0_rlc_init(struct amdgpu_device *adev) return r; } + /* init spm vmid with 0xf */ + if (adev->gfx.rlc.funcs->update_spm_vmid) + adev->gfx.rlc.funcs->update_spm_vmid(adev, 0xf); + return 0; } @@ -1304,7 +1349,7 @@ static int gfx_v10_0_sw_init(void *handle) case CHIP_NAVI14: case CHIP_NAVI12: adev->gfx.me.num_me = 1; - adev->gfx.me.num_pipe_per_me = 2; + adev->gfx.me.num_pipe_per_me = 1; adev->gfx.me.num_queue_per_pipe = 1; adev->gfx.mec.num_mec = 2; adev->gfx.mec.num_pipe_per_mec = 4; @@ -1783,11 +1828,11 @@ static int gfx_v10_0_init_csb(struct amdgpu_device *adev) adev->gfx.rlc.funcs->get_csb_buffer(adev, adev->gfx.rlc.cs_ptr); /* csib */ - WREG32_SOC15(GC, 0, mmRLC_CSIB_ADDR_HI, + WREG32_SOC15_RLC(GC, 0, mmRLC_CSIB_ADDR_HI, adev->gfx.rlc.clear_state_gpu_addr >> 32); - WREG32_SOC15(GC, 0, mmRLC_CSIB_ADDR_LO, + WREG32_SOC15_RLC(GC, 0, mmRLC_CSIB_ADDR_LO, adev->gfx.rlc.clear_state_gpu_addr & 0xfffffffc); - WREG32_SOC15(GC, 0, mmRLC_CSIB_LENGTH, adev->gfx.rlc.clear_state_size); + WREG32_SOC15_RLC(GC, 0, mmRLC_CSIB_LENGTH, adev->gfx.rlc.clear_state_size); return 0; } @@ -1895,6 +1940,11 @@ static int gfx_v10_0_rlc_resume(struct amdgpu_device *adev) if (!amdgpu_sriov_vf(adev)) /* enable RLC SRM */ gfx_v10_0_rlc_enable_srm(adev); } else { + if (amdgpu_sriov_vf(adev)) { + gfx_v10_0_init_csb(adev); + return 0; + } + adev->gfx.rlc.funcs->stop(adev); /* disable CG */ @@ -2395,7 +2445,7 @@ static int gfx_v10_0_cp_gfx_enable(struct amdgpu_device *adev, bool enable) for (i = 0; i < adev->gfx.num_gfx_rings; i++) adev->gfx.gfx_ring[i].sched.ready = false; } - WREG32_SOC15(GC, 0, mmCP_ME_CNTL, tmp); + WREG32_SOC15_RLC(GC, 0, mmCP_ME_CNTL, tmp); for (i = 0; i < adev->usec_timeout; i++) { if (RREG32_SOC15(GC, 0, mmCP_STAT) == 0) @@ -2710,18 +2760,20 @@ static int gfx_v10_0_cp_gfx_start(struct amdgpu_device *adev) amdgpu_ring_commit(ring); /* submit cs packet to copy state 0 to next available state */ - ring = &adev->gfx.gfx_ring[1]; - r = amdgpu_ring_alloc(ring, 2); - if (r) { - DRM_ERROR("amdgpu: cp failed to lock ring (%d).\n", r); - return r; - } - - amdgpu_ring_write(ring, PACKET3(PACKET3_CLEAR_STATE, 0)); - amdgpu_ring_write(ring, 0); + if (adev->gfx.num_gfx_rings > 1) { + /* maximum supported gfx ring is 2 */ + ring = &adev->gfx.gfx_ring[1]; + r = amdgpu_ring_alloc(ring, 2); + if (r) { + DRM_ERROR("amdgpu: cp failed to lock ring (%d).\n", r); + return r; + } - amdgpu_ring_commit(ring); + amdgpu_ring_write(ring, PACKET3(PACKET3_CLEAR_STATE, 0)); + amdgpu_ring_write(ring, 0); + amdgpu_ring_commit(ring); + } return 0; } @@ -2818,39 +2870,41 @@ static int gfx_v10_0_cp_gfx_resume(struct amdgpu_device *adev) mutex_unlock(&adev->srbm_mutex); /* Init gfx ring 1 for pipe 1 */ - mutex_lock(&adev->srbm_mutex); - gfx_v10_0_cp_gfx_switch_pipe(adev, PIPE_ID1); - ring = &adev->gfx.gfx_ring[1]; - rb_bufsz = order_base_2(ring->ring_size / 8); - tmp = REG_SET_FIELD(0, CP_RB1_CNTL, RB_BUFSZ, rb_bufsz); - tmp = REG_SET_FIELD(tmp, CP_RB1_CNTL, RB_BLKSZ, rb_bufsz - 2); - WREG32_SOC15(GC, 0, mmCP_RB1_CNTL, tmp); - /* Initialize the ring buffer's write pointers */ - ring->wptr = 0; - WREG32_SOC15(GC, 0, mmCP_RB1_WPTR, lower_32_bits(ring->wptr)); - WREG32_SOC15(GC, 0, mmCP_RB1_WPTR_HI, upper_32_bits(ring->wptr)); - /* Set the wb address wether it's enabled or not */ - rptr_addr = adev->wb.gpu_addr + (ring->rptr_offs * 4); - WREG32_SOC15(GC, 0, mmCP_RB1_RPTR_ADDR, lower_32_bits(rptr_addr)); - WREG32_SOC15(GC, 0, mmCP_RB1_RPTR_ADDR_HI, upper_32_bits(rptr_addr) & - CP_RB1_RPTR_ADDR_HI__RB_RPTR_ADDR_HI_MASK); - wptr_gpu_addr = adev->wb.gpu_addr + (ring->wptr_offs * 4); - WREG32_SOC15(GC, 0, mmCP_RB_WPTR_POLL_ADDR_LO, - lower_32_bits(wptr_gpu_addr)); - WREG32_SOC15(GC, 0, mmCP_RB_WPTR_POLL_ADDR_HI, - upper_32_bits(wptr_gpu_addr)); - - mdelay(1); - WREG32_SOC15(GC, 0, mmCP_RB1_CNTL, tmp); - - rb_addr = ring->gpu_addr >> 8; - WREG32_SOC15(GC, 0, mmCP_RB1_BASE, rb_addr); - WREG32_SOC15(GC, 0, mmCP_RB1_BASE_HI, upper_32_bits(rb_addr)); - WREG32_SOC15(GC, 0, mmCP_RB1_ACTIVE, 1); - - gfx_v10_0_cp_gfx_set_doorbell(adev, ring); - mutex_unlock(&adev->srbm_mutex); - + if (adev->gfx.num_gfx_rings > 1) { + mutex_lock(&adev->srbm_mutex); + gfx_v10_0_cp_gfx_switch_pipe(adev, PIPE_ID1); + /* maximum supported gfx ring is 2 */ + ring = &adev->gfx.gfx_ring[1]; + rb_bufsz = order_base_2(ring->ring_size / 8); + tmp = REG_SET_FIELD(0, CP_RB1_CNTL, RB_BUFSZ, rb_bufsz); + tmp = REG_SET_FIELD(tmp, CP_RB1_CNTL, RB_BLKSZ, rb_bufsz - 2); + WREG32_SOC15(GC, 0, mmCP_RB1_CNTL, tmp); + /* Initialize the ring buffer's write pointers */ + ring->wptr = 0; + WREG32_SOC15(GC, 0, mmCP_RB1_WPTR, lower_32_bits(ring->wptr)); + WREG32_SOC15(GC, 0, mmCP_RB1_WPTR_HI, upper_32_bits(ring->wptr)); + /* Set the wb address wether it's enabled or not */ + rptr_addr = adev->wb.gpu_addr + (ring->rptr_offs * 4); + WREG32_SOC15(GC, 0, mmCP_RB1_RPTR_ADDR, lower_32_bits(rptr_addr)); + WREG32_SOC15(GC, 0, mmCP_RB1_RPTR_ADDR_HI, upper_32_bits(rptr_addr) & + CP_RB1_RPTR_ADDR_HI__RB_RPTR_ADDR_HI_MASK); + wptr_gpu_addr = adev->wb.gpu_addr + (ring->wptr_offs * 4); + WREG32_SOC15(GC, 0, mmCP_RB_WPTR_POLL_ADDR_LO, + lower_32_bits(wptr_gpu_addr)); + WREG32_SOC15(GC, 0, mmCP_RB_WPTR_POLL_ADDR_HI, + upper_32_bits(wptr_gpu_addr)); + + mdelay(1); + WREG32_SOC15(GC, 0, mmCP_RB1_CNTL, tmp); + + rb_addr = ring->gpu_addr >> 8; + WREG32_SOC15(GC, 0, mmCP_RB1_BASE, rb_addr); + WREG32_SOC15(GC, 0, mmCP_RB1_BASE_HI, upper_32_bits(rb_addr)); + WREG32_SOC15(GC, 0, mmCP_RB1_ACTIVE, 1); + + gfx_v10_0_cp_gfx_set_doorbell(adev, ring); + mutex_unlock(&adev->srbm_mutex); + } /* Switch to pipe 0 */ mutex_lock(&adev->srbm_mutex); gfx_v10_0_cp_gfx_switch_pipe(adev, PIPE_ID0); @@ -3207,6 +3261,22 @@ done: return r; } +static void gfx_v10_0_compute_mqd_set_priority(struct amdgpu_ring *ring, struct v10_compute_mqd *mqd) +{ + struct amdgpu_device *adev = ring->adev; + + if (ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE) { + if (amdgpu_gfx_is_high_priority_compute_queue(adev, ring->queue)) { + mqd->cp_hqd_pipe_priority = AMDGPU_GFX_PIPE_PRIO_HIGH; + ring->has_high_prio = true; + mqd->cp_hqd_queue_priority = + AMDGPU_GFX_QUEUE_PRIORITY_MAXIMUM; + } else { + ring->has_high_prio = false; + } + } +} + static int gfx_v10_0_compute_mqd_init(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; @@ -3332,6 +3402,9 @@ static int gfx_v10_0_compute_mqd_init(struct amdgpu_ring *ring) tmp = REG_SET_FIELD(tmp, CP_HQD_IB_CONTROL, MIN_IB_AVAIL_SIZE, 3); mqd->cp_hqd_ib_control = tmp; + /* set static priority for a compute queue/ring */ + gfx_v10_0_compute_mqd_set_priority(ring, mqd); + /* map_queues packet doesn't need activate the queue, * so only kiq need set this field. */ @@ -3508,6 +3581,7 @@ static int gfx_v10_0_kcq_init_queue(struct amdgpu_ring *ring) /* reset ring buffer */ ring->wptr = 0; + atomic64_set((atomic64_t *)&adev->wb.wb[ring->wptr_offs], 0); amdgpu_ring_clear_ring(ring); } else { amdgpu_ring_clear_ring(ring); @@ -3920,9 +3994,8 @@ static uint64_t gfx_v10_0_get_gpu_clock_counter(struct amdgpu_device *adev) amdgpu_gfx_off_ctrl(adev, false); mutex_lock(&adev->gfx.gpu_clock_mutex); - WREG32_SOC15(GC, 0, mmRLC_CAPTURE_GPU_CLOCK_COUNT, 1); - clock = (uint64_t)RREG32_SOC15(GC, 0, mmRLC_GPU_CLOCK_COUNT_LSB) | - ((uint64_t)RREG32_SOC15(GC, 0, mmRLC_GPU_CLOCK_COUNT_MSB) << 32ULL); + clock = (uint64_t)RREG32_SOC15(SMUIO, 0, mmGOLDEN_TSC_COUNT_LOWER) | + ((uint64_t)RREG32_SOC15(SMUIO, 0, mmGOLDEN_TSC_COUNT_UPPER) << 32ULL); mutex_unlock(&adev->gfx.gpu_clock_mutex); amdgpu_gfx_off_ctrl(adev, true); return clock; @@ -3961,7 +4034,8 @@ static int gfx_v10_0_early_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - adev->gfx.num_gfx_rings = GFX10_NUM_GFX_RINGS; + adev->gfx.num_gfx_rings = GFX10_NUM_GFX_RINGS_NV1X; + adev->gfx.num_compute_rings = AMDGPU_MAX_COMPUTE_RINGS; gfx_v10_0_set_kiq_pm4_funcs(adev); @@ -4209,6 +4283,45 @@ static int gfx_v10_0_update_gfx_clock_gating(struct amdgpu_device *adev, return 0; } +static void gfx_v10_0_update_spm_vmid(struct amdgpu_device *adev, unsigned vmid) +{ + u32 data; + + data = RREG32_SOC15(GC, 0, mmRLC_SPM_MC_CNTL); + + data &= ~RLC_SPM_MC_CNTL__RLC_SPM_VMID_MASK; + data |= (vmid & RLC_SPM_MC_CNTL__RLC_SPM_VMID_MASK) << RLC_SPM_MC_CNTL__RLC_SPM_VMID__SHIFT; + + WREG32_SOC15(GC, 0, mmRLC_SPM_MC_CNTL, data); +} + +static bool gfx_v10_0_check_rlcg_range(struct amdgpu_device *adev, + uint32_t offset, + struct soc15_reg_rlcg *entries, int arr_size) +{ + int i; + uint32_t reg; + + if (!entries) + return false; + + for (i = 0; i < arr_size; i++) { + const struct soc15_reg_rlcg *entry; + + entry = &entries[i]; + reg = adev->reg_offset[entry->hwip][entry->instance][entry->segment] + entry->reg; + if (offset == reg) + return true; + } + + return false; +} + +static bool gfx_v10_0_is_rlcg_access_range(struct amdgpu_device *adev, u32 offset) +{ + return gfx_v10_0_check_rlcg_range(adev, offset, NULL, 0); +} + static const struct amdgpu_rlc_funcs gfx_v10_0_rlc_funcs = { .is_rlc_enabled = gfx_v10_0_is_rlc_enabled, .set_safe_mode = gfx_v10_0_set_safe_mode, @@ -4219,7 +4332,10 @@ static const struct amdgpu_rlc_funcs gfx_v10_0_rlc_funcs = { .resume = gfx_v10_0_rlc_resume, .stop = gfx_v10_0_rlc_stop, .reset = gfx_v10_0_rlc_reset, - .start = gfx_v10_0_rlc_start + .start = gfx_v10_0_rlc_start, + .update_spm_vmid = gfx_v10_0_update_spm_vmid, + .rlcg_wreg = gfx_v10_rlcg_wreg, + .is_rlcg_access_range = gfx_v10_0_is_rlcg_access_range, }; static int gfx_v10_0_set_powergating_state(void *handle, @@ -4414,7 +4530,7 @@ static void gfx_v10_0_ring_emit_ib_gfx(struct amdgpu_ring *ring, if (flags & AMDGPU_IB_PREEMPTED) control |= INDIRECT_BUFFER_PRE_RESUME(1); - if (!(ib->flags & AMDGPU_IB_FLAG_CE)) + if (!(ib->flags & AMDGPU_IB_FLAG_CE) && vmid) gfx_v10_0_ring_emit_de_meta(ring, (!amdgpu_sriov_vf(ring->adev) && flags & AMDGPU_IB_PREEMPTED) ? true : false); } diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c index 8f20a5dd44fe..733d398c61cc 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c @@ -3346,6 +3346,10 @@ static int gfx_v7_0_rlc_init(struct amdgpu_device *adev) return r; } + /* init spm vmid with 0xf */ + if (adev->gfx.rlc.funcs->update_spm_vmid) + adev->gfx.rlc.funcs->update_spm_vmid(adev, 0xf); + return 0; } @@ -3570,6 +3574,18 @@ static int gfx_v7_0_rlc_resume(struct amdgpu_device *adev) return 0; } +static void gfx_v7_0_update_spm_vmid(struct amdgpu_device *adev, unsigned vmid) +{ + u32 data; + + data = RREG32(mmRLC_SPM_VMID); + + data &= ~RLC_SPM_VMID__RLC_SPM_VMID_MASK; + data |= (vmid & RLC_SPM_VMID__RLC_SPM_VMID_MASK) << RLC_SPM_VMID__RLC_SPM_VMID__SHIFT; + + WREG32(mmRLC_SPM_VMID, data); +} + static void gfx_v7_0_enable_cgcg(struct amdgpu_device *adev, bool enable) { u32 data, orig, tmp, tmp2; @@ -4221,7 +4237,8 @@ static const struct amdgpu_rlc_funcs gfx_v7_0_rlc_funcs = { .resume = gfx_v7_0_rlc_resume, .stop = gfx_v7_0_rlc_stop, .reset = gfx_v7_0_rlc_reset, - .start = gfx_v7_0_rlc_start + .start = gfx_v7_0_rlc_start, + .update_spm_vmid = gfx_v7_0_update_spm_vmid }; static int gfx_v7_0_early_init(void *handle) @@ -4338,6 +4355,11 @@ static void gfx_v7_0_gpu_early_init(struct amdgpu_device *adev) adev->gfx.config.mc_arb_ramcfg = RREG32(mmMC_ARB_RAMCFG); mc_arb_ramcfg = adev->gfx.config.mc_arb_ramcfg; + adev->gfx.config.num_banks = REG_GET_FIELD(mc_arb_ramcfg, + MC_ARB_RAMCFG, NOOFBANK); + adev->gfx.config.num_ranks = REG_GET_FIELD(mc_arb_ramcfg, + MC_ARB_RAMCFG, NOOFRANKS); + adev->gfx.config.num_tile_pipes = adev->gfx.config.max_tile_pipes; adev->gfx.config.mem_max_burst_length_bytes = 256; if (adev->flags & AMD_IS_APU) { diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c index fa245973de12..fc32586ef80b 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c @@ -1318,6 +1318,10 @@ static int gfx_v8_0_rlc_init(struct amdgpu_device *adev) return r; } + /* init spm vmid with 0xf */ + if (adev->gfx.rlc.funcs->update_spm_vmid) + adev->gfx.rlc.funcs->update_spm_vmid(adev, 0xf); + return 0; } @@ -1820,6 +1824,11 @@ static int gfx_v8_0_gpu_early_init(struct amdgpu_device *adev) adev->gfx.config.mc_arb_ramcfg = RREG32(mmMC_ARB_RAMCFG); mc_arb_ramcfg = adev->gfx.config.mc_arb_ramcfg; + adev->gfx.config.num_banks = REG_GET_FIELD(mc_arb_ramcfg, + MC_ARB_RAMCFG, NOOFBANK); + adev->gfx.config.num_ranks = REG_GET_FIELD(mc_arb_ramcfg, + MC_ARB_RAMCFG, NOOFRANKS); + adev->gfx.config.num_tile_pipes = adev->gfx.config.max_tile_pipes; adev->gfx.config.mem_max_burst_length_bytes = 256; if (adev->flags & AMD_IS_APU) { @@ -4421,6 +4430,22 @@ static int gfx_v8_0_deactivate_hqd(struct amdgpu_device *adev, u32 req) return r; } +static void gfx_v8_0_mqd_set_priority(struct amdgpu_ring *ring, struct vi_mqd *mqd) +{ + struct amdgpu_device *adev = ring->adev; + + if (ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE) { + if (amdgpu_gfx_is_high_priority_compute_queue(adev, ring->queue)) { + mqd->cp_hqd_pipe_priority = AMDGPU_GFX_PIPE_PRIO_HIGH; + ring->has_high_prio = true; + mqd->cp_hqd_queue_priority = + AMDGPU_GFX_QUEUE_PRIORITY_MAXIMUM; + } else { + ring->has_high_prio = false; + } + } +} + static int gfx_v8_0_mqd_init(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; @@ -4544,9 +4569,6 @@ static int gfx_v8_0_mqd_init(struct amdgpu_ring *ring) /* defaults */ mqd->cp_hqd_eop_rptr = RREG32(mmCP_HQD_EOP_RPTR); mqd->cp_hqd_eop_wptr = RREG32(mmCP_HQD_EOP_WPTR); - mqd->cp_hqd_pipe_priority = RREG32(mmCP_HQD_PIPE_PRIORITY); - mqd->cp_hqd_queue_priority = RREG32(mmCP_HQD_QUEUE_PRIORITY); - mqd->cp_hqd_quantum = RREG32(mmCP_HQD_QUANTUM); mqd->cp_hqd_ctx_save_base_addr_lo = RREG32(mmCP_HQD_CTX_SAVE_BASE_ADDR_LO); mqd->cp_hqd_ctx_save_base_addr_hi = RREG32(mmCP_HQD_CTX_SAVE_BASE_ADDR_HI); mqd->cp_hqd_cntl_stack_offset = RREG32(mmCP_HQD_CNTL_STACK_OFFSET); @@ -4558,6 +4580,10 @@ static int gfx_v8_0_mqd_init(struct amdgpu_ring *ring) mqd->cp_hqd_eop_wptr_mem = RREG32(mmCP_HQD_EOP_WPTR_MEM); mqd->cp_hqd_eop_dones = RREG32(mmCP_HQD_EOP_DONES); + /* set static priority for a queue/ring */ + gfx_v8_0_mqd_set_priority(ring, mqd); + mqd->cp_hqd_quantum = RREG32(mmCP_HQD_QUANTUM); + /* map_queues packet doesn't need activate the queue, * so only kiq need set this field. */ @@ -5589,6 +5615,18 @@ static void gfx_v8_0_unset_safe_mode(struct amdgpu_device *adev) } } +static void gfx_v8_0_update_spm_vmid(struct amdgpu_device *adev, unsigned vmid) +{ + u32 data; + + data = RREG32(mmRLC_SPM_VMID); + + data &= ~RLC_SPM_VMID__RLC_SPM_VMID_MASK; + data |= (vmid & RLC_SPM_VMID__RLC_SPM_VMID_MASK) << RLC_SPM_VMID__RLC_SPM_VMID__SHIFT; + + WREG32(mmRLC_SPM_VMID, data); +} + static const struct amdgpu_rlc_funcs iceland_rlc_funcs = { .is_rlc_enabled = gfx_v8_0_is_rlc_enabled, .set_safe_mode = gfx_v8_0_set_safe_mode, @@ -5600,7 +5638,8 @@ static const struct amdgpu_rlc_funcs iceland_rlc_funcs = { .resume = gfx_v8_0_rlc_resume, .stop = gfx_v8_0_rlc_stop, .reset = gfx_v8_0_rlc_reset, - .start = gfx_v8_0_rlc_start + .start = gfx_v8_0_rlc_start, + .update_spm_vmid = gfx_v8_0_update_spm_vmid }; static void gfx_v8_0_update_medium_grain_clock_gating(struct amdgpu_device *adev, @@ -6094,7 +6133,7 @@ static void gfx_v8_0_ring_emit_ib_gfx(struct amdgpu_ring *ring, if (amdgpu_sriov_vf(ring->adev) && (ib->flags & AMDGPU_IB_FLAG_PREEMPT)) { control |= INDIRECT_BUFFER_PRE_ENB(1); - if (!(ib->flags & AMDGPU_IB_FLAG_CE)) + if (!(ib->flags & AMDGPU_IB_FLAG_CE) && vmid) gfx_v8_0_ring_emit_de_meta(ring); } @@ -6236,104 +6275,6 @@ static void gfx_v8_0_ring_set_wptr_compute(struct amdgpu_ring *ring) WDOORBELL32(ring->doorbell_index, lower_32_bits(ring->wptr)); } -static void gfx_v8_0_ring_set_pipe_percent(struct amdgpu_ring *ring, - bool acquire) -{ - struct amdgpu_device *adev = ring->adev; - int pipe_num, tmp, reg; - int pipe_percent = acquire ? SPI_WCL_PIPE_PERCENT_GFX__VALUE_MASK : 0x1; - - pipe_num = ring->me * adev->gfx.mec.num_pipe_per_mec + ring->pipe; - - /* first me only has 2 entries, GFX and HP3D */ - if (ring->me > 0) - pipe_num -= 2; - - reg = mmSPI_WCL_PIPE_PERCENT_GFX + pipe_num; - tmp = RREG32(reg); - tmp = REG_SET_FIELD(tmp, SPI_WCL_PIPE_PERCENT_GFX, VALUE, pipe_percent); - WREG32(reg, tmp); -} - -static void gfx_v8_0_pipe_reserve_resources(struct amdgpu_device *adev, - struct amdgpu_ring *ring, - bool acquire) -{ - int i, pipe; - bool reserve; - struct amdgpu_ring *iring; - - mutex_lock(&adev->gfx.pipe_reserve_mutex); - pipe = amdgpu_gfx_mec_queue_to_bit(adev, ring->me, ring->pipe, 0); - if (acquire) - set_bit(pipe, adev->gfx.pipe_reserve_bitmap); - else - clear_bit(pipe, adev->gfx.pipe_reserve_bitmap); - - if (!bitmap_weight(adev->gfx.pipe_reserve_bitmap, AMDGPU_MAX_COMPUTE_QUEUES)) { - /* Clear all reservations - everyone reacquires all resources */ - for (i = 0; i < adev->gfx.num_gfx_rings; ++i) - gfx_v8_0_ring_set_pipe_percent(&adev->gfx.gfx_ring[i], - true); - - for (i = 0; i < adev->gfx.num_compute_rings; ++i) - gfx_v8_0_ring_set_pipe_percent(&adev->gfx.compute_ring[i], - true); - } else { - /* Lower all pipes without a current reservation */ - for (i = 0; i < adev->gfx.num_gfx_rings; ++i) { - iring = &adev->gfx.gfx_ring[i]; - pipe = amdgpu_gfx_mec_queue_to_bit(adev, - iring->me, - iring->pipe, - 0); - reserve = test_bit(pipe, adev->gfx.pipe_reserve_bitmap); - gfx_v8_0_ring_set_pipe_percent(iring, reserve); - } - - for (i = 0; i < adev->gfx.num_compute_rings; ++i) { - iring = &adev->gfx.compute_ring[i]; - pipe = amdgpu_gfx_mec_queue_to_bit(adev, - iring->me, - iring->pipe, - 0); - reserve = test_bit(pipe, adev->gfx.pipe_reserve_bitmap); - gfx_v8_0_ring_set_pipe_percent(iring, reserve); - } - } - - mutex_unlock(&adev->gfx.pipe_reserve_mutex); -} - -static void gfx_v8_0_hqd_set_priority(struct amdgpu_device *adev, - struct amdgpu_ring *ring, - bool acquire) -{ - uint32_t pipe_priority = acquire ? 0x2 : 0x0; - uint32_t queue_priority = acquire ? 0xf : 0x0; - - mutex_lock(&adev->srbm_mutex); - vi_srbm_select(adev, ring->me, ring->pipe, ring->queue, 0); - - WREG32(mmCP_HQD_PIPE_PRIORITY, pipe_priority); - WREG32(mmCP_HQD_QUEUE_PRIORITY, queue_priority); - - vi_srbm_select(adev, 0, 0, 0, 0); - mutex_unlock(&adev->srbm_mutex); -} -static void gfx_v8_0_ring_set_priority_compute(struct amdgpu_ring *ring, - enum drm_sched_priority priority) -{ - struct amdgpu_device *adev = ring->adev; - bool acquire = priority == DRM_SCHED_PRIORITY_HIGH_HW; - - if (ring->funcs->type != AMDGPU_RING_TYPE_COMPUTE) - return; - - gfx_v8_0_hqd_set_priority(adev, ring, acquire); - gfx_v8_0_pipe_reserve_resources(adev, ring, acquire); -} - static void gfx_v8_0_ring_emit_fence_compute(struct amdgpu_ring *ring, u64 addr, u64 seq, unsigned flags) @@ -6966,7 +6907,6 @@ static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_compute = { .test_ib = gfx_v8_0_ring_test_ib, .insert_nop = amdgpu_ring_insert_nop, .pad_ib = amdgpu_ring_generic_pad_ib, - .set_priority = gfx_v8_0_ring_set_priority_compute, .emit_wreg = gfx_v8_0_ring_emit_wreg, }; diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c index 6a49fe452f29..37c8231f1407 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c @@ -697,6 +697,11 @@ static const struct soc15_reg_golden golden_settings_gc_9_4_1_arct[] = SOC15_REG_GOLDEN_VALUE(GC, 0, mmSQ_FIFO_SIZES, 0xffffffff, 0x00000f00), }; +static const struct soc15_reg_rlcg rlcg_access_gc_9_0[] = { + {SOC15_REG_ENTRY(GC, 0, mmGRBM_GFX_INDEX)}, + {SOC15_REG_ENTRY(GC, 0, mmSQ_IND_INDEX)}, +}; + static const u32 GFX_RLC_SRM_INDEX_CNTL_ADDR_OFFSETS[] = { mmRLC_SRM_INDEX_CNTL_ADDR_0 - mmRLC_SRM_INDEX_CNTL_ADDR_0, @@ -721,6 +726,59 @@ static const u32 GFX_RLC_SRM_INDEX_CNTL_DATA_OFFSETS[] = mmRLC_SRM_INDEX_CNTL_DATA_7 - mmRLC_SRM_INDEX_CNTL_DATA_0, }; +void gfx_v9_0_rlcg_wreg(struct amdgpu_device *adev, u32 offset, u32 v) +{ + static void *scratch_reg0; + static void *scratch_reg1; + static void *scratch_reg2; + static void *scratch_reg3; + static void *spare_int; + static uint32_t grbm_cntl; + static uint32_t grbm_idx; + + scratch_reg0 = adev->rmmio + (adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG0_BASE_IDX] + mmSCRATCH_REG0)*4; + scratch_reg1 = adev->rmmio + (adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG1_BASE_IDX] + mmSCRATCH_REG1)*4; + scratch_reg2 = adev->rmmio + (adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG1_BASE_IDX] + mmSCRATCH_REG2)*4; + scratch_reg3 = adev->rmmio + (adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG1_BASE_IDX] + mmSCRATCH_REG3)*4; + spare_int = adev->rmmio + (adev->reg_offset[GC_HWIP][0][mmRLC_SPARE_INT_BASE_IDX] + mmRLC_SPARE_INT)*4; + + grbm_cntl = adev->reg_offset[GC_HWIP][0][mmGRBM_GFX_CNTL_BASE_IDX] + mmGRBM_GFX_CNTL; + grbm_idx = adev->reg_offset[GC_HWIP][0][mmGRBM_GFX_INDEX_BASE_IDX] + mmGRBM_GFX_INDEX; + + if (amdgpu_sriov_runtime(adev)) { + pr_err("shouldn't call rlcg write register during runtime\n"); + return; + } + + if (offset == grbm_cntl || offset == grbm_idx) { + if (offset == grbm_cntl) + writel(v, scratch_reg2); + else if (offset == grbm_idx) + writel(v, scratch_reg3); + + writel(v, ((void __iomem *)adev->rmmio) + (offset * 4)); + } else { + uint32_t i = 0; + uint32_t retries = 50000; + + writel(v, scratch_reg0); + writel(offset | 0x80000000, scratch_reg1); + writel(1, spare_int); + for (i = 0; i < retries; i++) { + u32 tmp; + + tmp = readl(scratch_reg1); + if (!(tmp & 0x80000000)) + break; + + udelay(10); + } + if (i >= retries) + pr_err("timeout: rlcg program reg:0x%05x failed !\n", offset); + } + +} + #define VEGA10_GB_ADDR_CONFIG_GOLDEN 0x2a114042 #define VEGA12_GB_ADDR_CONFIG_GOLDEN 0x24104041 #define RAVEN_GB_ADDR_CONFIG_GOLDEN 0x24000042 @@ -738,9 +796,9 @@ static void gfx_v9_0_ring_emit_de_meta(struct amdgpu_ring *ring); static u64 gfx_v9_0_ring_get_rptr_compute(struct amdgpu_ring *ring); static int gfx_v9_0_query_ras_error_count(struct amdgpu_device *adev, void *ras_error_status); -static void gfx_v9_0_clear_ras_edc_counter(struct amdgpu_device *adev); static int gfx_v9_0_ras_error_inject(struct amdgpu_device *adev, void *inject_if); +static void gfx_v9_0_reset_ras_error_count(struct amdgpu_device *adev); static void gfx_v9_0_kiq_set_resources(struct amdgpu_ring *kiq_ring, uint64_t queue_mask) @@ -1847,6 +1905,10 @@ static int gfx_v9_0_rlc_init(struct amdgpu_device *adev) break; } + /* init spm vmid with 0xf */ + if (adev->gfx.rlc.funcs->update_spm_vmid) + adev->gfx.rlc.funcs->update_spm_vmid(adev, 0xf); + return 0; } @@ -1917,7 +1979,7 @@ static int gfx_v9_0_mec_init(struct amdgpu_device *adev) static uint32_t wave_read_ind(struct amdgpu_device *adev, uint32_t simd, uint32_t wave, uint32_t address) { - WREG32_SOC15(GC, 0, mmSQ_IND_INDEX, + WREG32_SOC15_RLC(GC, 0, mmSQ_IND_INDEX, (wave << SQ_IND_INDEX__WAVE_ID__SHIFT) | (simd << SQ_IND_INDEX__SIMD_ID__SHIFT) | (address << SQ_IND_INDEX__INDEX__SHIFT) | @@ -1929,7 +1991,7 @@ static void wave_read_regs(struct amdgpu_device *adev, uint32_t simd, uint32_t wave, uint32_t thread, uint32_t regno, uint32_t num, uint32_t *out) { - WREG32_SOC15(GC, 0, mmSQ_IND_INDEX, + WREG32_SOC15_RLC(GC, 0, mmSQ_IND_INDEX, (wave << SQ_IND_INDEX__WAVE_ID__SHIFT) | (simd << SQ_IND_INDEX__SIMD_ID__SHIFT) | (regno << SQ_IND_INDEX__INDEX__SHIFT) | @@ -1993,7 +2055,8 @@ static const struct amdgpu_gfx_funcs gfx_v9_0_gfx_funcs = { .read_wave_vgprs = &gfx_v9_0_read_wave_vgprs, .select_me_pipe_q = &gfx_v9_0_select_me_pipe_q, .ras_error_inject = &gfx_v9_0_ras_error_inject, - .query_ras_error_count = &gfx_v9_0_query_ras_error_count + .query_ras_error_count = &gfx_v9_0_query_ras_error_count, + .reset_ras_error_count = &gfx_v9_0_reset_ras_error_count, }; static const struct amdgpu_gfx_funcs gfx_v9_4_gfx_funcs = { @@ -2004,7 +2067,8 @@ static const struct amdgpu_gfx_funcs gfx_v9_4_gfx_funcs = { .read_wave_vgprs = &gfx_v9_0_read_wave_vgprs, .select_me_pipe_q = &gfx_v9_0_select_me_pipe_q, .ras_error_inject = &gfx_v9_4_ras_error_inject, - .query_ras_error_count = &gfx_v9_4_query_ras_error_count + .query_ras_error_count = &gfx_v9_4_query_ras_error_count, + .reset_ras_error_count = &gfx_v9_4_reset_ras_error_count, }; static int gfx_v9_0_gpu_early_init(struct amdgpu_device *adev) @@ -3310,6 +3374,22 @@ static void gfx_v9_0_kiq_setting(struct amdgpu_ring *ring) WREG32_SOC15_RLC(GC, 0, mmRLC_CP_SCHEDULERS, tmp); } +static void gfx_v9_0_mqd_set_priority(struct amdgpu_ring *ring, struct v9_mqd *mqd) +{ + struct amdgpu_device *adev = ring->adev; + + if (ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE) { + if (amdgpu_gfx_is_high_priority_compute_queue(adev, ring->queue)) { + mqd->cp_hqd_pipe_priority = AMDGPU_GFX_PIPE_PRIO_HIGH; + ring->has_high_prio = true; + mqd->cp_hqd_queue_priority = + AMDGPU_GFX_QUEUE_PRIORITY_MAXIMUM; + } else { + ring->has_high_prio = false; + } + } +} + static int gfx_v9_0_mqd_init(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; @@ -3446,6 +3526,10 @@ static int gfx_v9_0_mqd_init(struct amdgpu_ring *ring) tmp = REG_SET_FIELD(tmp, CP_HQD_IB_CONTROL, MIN_IB_AVAIL_SIZE, 3); mqd->cp_hqd_ib_control = tmp; + /* set static priority for a queue/ring */ + gfx_v9_0_mqd_set_priority(ring, mqd); + mqd->cp_hqd_quantum = RREG32(mmCP_HQD_QUANTUM); + /* map_queues packet doesn't need activate the queue, * so only kiq need set this field. */ @@ -3664,6 +3748,7 @@ static int gfx_v9_0_kcq_init_queue(struct amdgpu_ring *ring) /* reset ring buffer */ ring->wptr = 0; + atomic64_set((atomic64_t *)&adev->wb.wb[ring->wptr_offs], 0); amdgpu_ring_clear_ring(ring); } else { amdgpu_ring_clear_ring(ring); @@ -3963,6 +4048,63 @@ static int gfx_v9_0_soft_reset(void *handle) return 0; } +static uint64_t gfx_v9_0_kiq_read_clock(struct amdgpu_device *adev) +{ + signed long r, cnt = 0; + unsigned long flags; + uint32_t seq; + struct amdgpu_kiq *kiq = &adev->gfx.kiq; + struct amdgpu_ring *ring = &kiq->ring; + + BUG_ON(!ring->funcs->emit_rreg); + + spin_lock_irqsave(&kiq->ring_lock, flags); + amdgpu_ring_alloc(ring, 32); + amdgpu_ring_write(ring, PACKET3(PACKET3_COPY_DATA, 4)); + amdgpu_ring_write(ring, 9 | /* src: register*/ + (5 << 8) | /* dst: memory */ + (1 << 16) | /* count sel */ + (1 << 20)); /* write confirm */ + amdgpu_ring_write(ring, 0); + amdgpu_ring_write(ring, 0); + amdgpu_ring_write(ring, lower_32_bits(adev->wb.gpu_addr + + kiq->reg_val_offs * 4)); + amdgpu_ring_write(ring, upper_32_bits(adev->wb.gpu_addr + + kiq->reg_val_offs * 4)); + amdgpu_fence_emit_polling(ring, &seq); + amdgpu_ring_commit(ring); + spin_unlock_irqrestore(&kiq->ring_lock, flags); + + r = amdgpu_fence_wait_polling(ring, seq, MAX_KIQ_REG_WAIT); + + /* don't wait anymore for gpu reset case because this way may + * block gpu_recover() routine forever, e.g. this virt_kiq_rreg + * is triggered in TTM and ttm_bo_lock_delayed_workqueue() will + * never return if we keep waiting in virt_kiq_rreg, which cause + * gpu_recover() hang there. + * + * also don't wait anymore for IRQ context + * */ + if (r < 1 && (adev->in_gpu_reset || in_interrupt())) + goto failed_kiq_read; + + might_sleep(); + while (r < 1 && cnt++ < MAX_KIQ_REG_TRY) { + msleep(MAX_KIQ_REG_BAILOUT_INTERVAL); + r = amdgpu_fence_wait_polling(ring, seq, MAX_KIQ_REG_WAIT); + } + + if (cnt > MAX_KIQ_REG_TRY) + goto failed_kiq_read; + + return (uint64_t)adev->wb.wb[kiq->reg_val_offs] | + (uint64_t)adev->wb.wb[kiq->reg_val_offs + 1 ] << 32ULL; + +failed_kiq_read: + pr_err("failed to read gpu clock\n"); + return ~0; +} + static uint64_t gfx_v9_0_get_gpu_clock_counter(struct amdgpu_device *adev) { uint64_t clock; @@ -3970,16 +4112,7 @@ static uint64_t gfx_v9_0_get_gpu_clock_counter(struct amdgpu_device *adev) amdgpu_gfx_off_ctrl(adev, false); mutex_lock(&adev->gfx.gpu_clock_mutex); if (adev->asic_type == CHIP_VEGA10 && amdgpu_sriov_runtime(adev)) { - uint32_t tmp, lsb, msb, i = 0; - do { - if (i != 0) - udelay(1); - tmp = RREG32_SOC15(GC, 0, mmRLC_REFCLOCK_TIMESTAMP_MSB); - lsb = RREG32_SOC15(GC, 0, mmRLC_REFCLOCK_TIMESTAMP_LSB); - msb = RREG32_SOC15(GC, 0, mmRLC_REFCLOCK_TIMESTAMP_MSB); - i++; - } while (unlikely(tmp != msb) && (i < adev->usec_timeout)); - clock = (uint64_t)lsb | ((uint64_t)msb << 32ULL); + clock = gfx_v9_0_kiq_read_clock(adev); } else { WREG32_SOC15(GC, 0, mmRLC_CAPTURE_GPU_CLOCK_COUNT, 1); clock = (uint64_t)RREG32_SOC15(GC, 0, mmRLC_GPU_CLOCK_COUNT_LSB) | @@ -4053,6 +4186,101 @@ static const u32 sgpr_init_compute_shader[] = 0xbe800080, 0xbf810000, }; +static const u32 vgpr_init_compute_shader_arcturus[] = { + 0xd3d94000, 0x18000080, 0xd3d94001, 0x18000080, 0xd3d94002, 0x18000080, + 0xd3d94003, 0x18000080, 0xd3d94004, 0x18000080, 0xd3d94005, 0x18000080, + 0xd3d94006, 0x18000080, 0xd3d94007, 0x18000080, 0xd3d94008, 0x18000080, + 0xd3d94009, 0x18000080, 0xd3d9400a, 0x18000080, 0xd3d9400b, 0x18000080, + 0xd3d9400c, 0x18000080, 0xd3d9400d, 0x18000080, 0xd3d9400e, 0x18000080, + 0xd3d9400f, 0x18000080, 0xd3d94010, 0x18000080, 0xd3d94011, 0x18000080, + 0xd3d94012, 0x18000080, 0xd3d94013, 0x18000080, 0xd3d94014, 0x18000080, + 0xd3d94015, 0x18000080, 0xd3d94016, 0x18000080, 0xd3d94017, 0x18000080, + 0xd3d94018, 0x18000080, 0xd3d94019, 0x18000080, 0xd3d9401a, 0x18000080, + 0xd3d9401b, 0x18000080, 0xd3d9401c, 0x18000080, 0xd3d9401d, 0x18000080, + 0xd3d9401e, 0x18000080, 0xd3d9401f, 0x18000080, 0xd3d94020, 0x18000080, + 0xd3d94021, 0x18000080, 0xd3d94022, 0x18000080, 0xd3d94023, 0x18000080, + 0xd3d94024, 0x18000080, 0xd3d94025, 0x18000080, 0xd3d94026, 0x18000080, + 0xd3d94027, 0x18000080, 0xd3d94028, 0x18000080, 0xd3d94029, 0x18000080, + 0xd3d9402a, 0x18000080, 0xd3d9402b, 0x18000080, 0xd3d9402c, 0x18000080, + 0xd3d9402d, 0x18000080, 0xd3d9402e, 0x18000080, 0xd3d9402f, 0x18000080, + 0xd3d94030, 0x18000080, 0xd3d94031, 0x18000080, 0xd3d94032, 0x18000080, + 0xd3d94033, 0x18000080, 0xd3d94034, 0x18000080, 0xd3d94035, 0x18000080, + 0xd3d94036, 0x18000080, 0xd3d94037, 0x18000080, 0xd3d94038, 0x18000080, + 0xd3d94039, 0x18000080, 0xd3d9403a, 0x18000080, 0xd3d9403b, 0x18000080, + 0xd3d9403c, 0x18000080, 0xd3d9403d, 0x18000080, 0xd3d9403e, 0x18000080, + 0xd3d9403f, 0x18000080, 0xd3d94040, 0x18000080, 0xd3d94041, 0x18000080, + 0xd3d94042, 0x18000080, 0xd3d94043, 0x18000080, 0xd3d94044, 0x18000080, + 0xd3d94045, 0x18000080, 0xd3d94046, 0x18000080, 0xd3d94047, 0x18000080, + 0xd3d94048, 0x18000080, 0xd3d94049, 0x18000080, 0xd3d9404a, 0x18000080, + 0xd3d9404b, 0x18000080, 0xd3d9404c, 0x18000080, 0xd3d9404d, 0x18000080, + 0xd3d9404e, 0x18000080, 0xd3d9404f, 0x18000080, 0xd3d94050, 0x18000080, + 0xd3d94051, 0x18000080, 0xd3d94052, 0x18000080, 0xd3d94053, 0x18000080, + 0xd3d94054, 0x18000080, 0xd3d94055, 0x18000080, 0xd3d94056, 0x18000080, + 0xd3d94057, 0x18000080, 0xd3d94058, 0x18000080, 0xd3d94059, 0x18000080, + 0xd3d9405a, 0x18000080, 0xd3d9405b, 0x18000080, 0xd3d9405c, 0x18000080, + 0xd3d9405d, 0x18000080, 0xd3d9405e, 0x18000080, 0xd3d9405f, 0x18000080, + 0xd3d94060, 0x18000080, 0xd3d94061, 0x18000080, 0xd3d94062, 0x18000080, + 0xd3d94063, 0x18000080, 0xd3d94064, 0x18000080, 0xd3d94065, 0x18000080, + 0xd3d94066, 0x18000080, 0xd3d94067, 0x18000080, 0xd3d94068, 0x18000080, + 0xd3d94069, 0x18000080, 0xd3d9406a, 0x18000080, 0xd3d9406b, 0x18000080, + 0xd3d9406c, 0x18000080, 0xd3d9406d, 0x18000080, 0xd3d9406e, 0x18000080, + 0xd3d9406f, 0x18000080, 0xd3d94070, 0x18000080, 0xd3d94071, 0x18000080, + 0xd3d94072, 0x18000080, 0xd3d94073, 0x18000080, 0xd3d94074, 0x18000080, + 0xd3d94075, 0x18000080, 0xd3d94076, 0x18000080, 0xd3d94077, 0x18000080, + 0xd3d94078, 0x18000080, 0xd3d94079, 0x18000080, 0xd3d9407a, 0x18000080, + 0xd3d9407b, 0x18000080, 0xd3d9407c, 0x18000080, 0xd3d9407d, 0x18000080, + 0xd3d9407e, 0x18000080, 0xd3d9407f, 0x18000080, 0xd3d94080, 0x18000080, + 0xd3d94081, 0x18000080, 0xd3d94082, 0x18000080, 0xd3d94083, 0x18000080, + 0xd3d94084, 0x18000080, 0xd3d94085, 0x18000080, 0xd3d94086, 0x18000080, + 0xd3d94087, 0x18000080, 0xd3d94088, 0x18000080, 0xd3d94089, 0x18000080, + 0xd3d9408a, 0x18000080, 0xd3d9408b, 0x18000080, 0xd3d9408c, 0x18000080, + 0xd3d9408d, 0x18000080, 0xd3d9408e, 0x18000080, 0xd3d9408f, 0x18000080, + 0xd3d94090, 0x18000080, 0xd3d94091, 0x18000080, 0xd3d94092, 0x18000080, + 0xd3d94093, 0x18000080, 0xd3d94094, 0x18000080, 0xd3d94095, 0x18000080, + 0xd3d94096, 0x18000080, 0xd3d94097, 0x18000080, 0xd3d94098, 0x18000080, + 0xd3d94099, 0x18000080, 0xd3d9409a, 0x18000080, 0xd3d9409b, 0x18000080, + 0xd3d9409c, 0x18000080, 0xd3d9409d, 0x18000080, 0xd3d9409e, 0x18000080, + 0xd3d9409f, 0x18000080, 0xd3d940a0, 0x18000080, 0xd3d940a1, 0x18000080, + 0xd3d940a2, 0x18000080, 0xd3d940a3, 0x18000080, 0xd3d940a4, 0x18000080, + 0xd3d940a5, 0x18000080, 0xd3d940a6, 0x18000080, 0xd3d940a7, 0x18000080, + 0xd3d940a8, 0x18000080, 0xd3d940a9, 0x18000080, 0xd3d940aa, 0x18000080, + 0xd3d940ab, 0x18000080, 0xd3d940ac, 0x18000080, 0xd3d940ad, 0x18000080, + 0xd3d940ae, 0x18000080, 0xd3d940af, 0x18000080, 0xd3d940b0, 0x18000080, + 0xd3d940b1, 0x18000080, 0xd3d940b2, 0x18000080, 0xd3d940b3, 0x18000080, + 0xd3d940b4, 0x18000080, 0xd3d940b5, 0x18000080, 0xd3d940b6, 0x18000080, + 0xd3d940b7, 0x18000080, 0xd3d940b8, 0x18000080, 0xd3d940b9, 0x18000080, + 0xd3d940ba, 0x18000080, 0xd3d940bb, 0x18000080, 0xd3d940bc, 0x18000080, + 0xd3d940bd, 0x18000080, 0xd3d940be, 0x18000080, 0xd3d940bf, 0x18000080, + 0xd3d940c0, 0x18000080, 0xd3d940c1, 0x18000080, 0xd3d940c2, 0x18000080, + 0xd3d940c3, 0x18000080, 0xd3d940c4, 0x18000080, 0xd3d940c5, 0x18000080, + 0xd3d940c6, 0x18000080, 0xd3d940c7, 0x18000080, 0xd3d940c8, 0x18000080, + 0xd3d940c9, 0x18000080, 0xd3d940ca, 0x18000080, 0xd3d940cb, 0x18000080, + 0xd3d940cc, 0x18000080, 0xd3d940cd, 0x18000080, 0xd3d940ce, 0x18000080, + 0xd3d940cf, 0x18000080, 0xd3d940d0, 0x18000080, 0xd3d940d1, 0x18000080, + 0xd3d940d2, 0x18000080, 0xd3d940d3, 0x18000080, 0xd3d940d4, 0x18000080, + 0xd3d940d5, 0x18000080, 0xd3d940d6, 0x18000080, 0xd3d940d7, 0x18000080, + 0xd3d940d8, 0x18000080, 0xd3d940d9, 0x18000080, 0xd3d940da, 0x18000080, + 0xd3d940db, 0x18000080, 0xd3d940dc, 0x18000080, 0xd3d940dd, 0x18000080, + 0xd3d940de, 0x18000080, 0xd3d940df, 0x18000080, 0xd3d940e0, 0x18000080, + 0xd3d940e1, 0x18000080, 0xd3d940e2, 0x18000080, 0xd3d940e3, 0x18000080, + 0xd3d940e4, 0x18000080, 0xd3d940e5, 0x18000080, 0xd3d940e6, 0x18000080, + 0xd3d940e7, 0x18000080, 0xd3d940e8, 0x18000080, 0xd3d940e9, 0x18000080, + 0xd3d940ea, 0x18000080, 0xd3d940eb, 0x18000080, 0xd3d940ec, 0x18000080, + 0xd3d940ed, 0x18000080, 0xd3d940ee, 0x18000080, 0xd3d940ef, 0x18000080, + 0xd3d940f0, 0x18000080, 0xd3d940f1, 0x18000080, 0xd3d940f2, 0x18000080, + 0xd3d940f3, 0x18000080, 0xd3d940f4, 0x18000080, 0xd3d940f5, 0x18000080, + 0xd3d940f6, 0x18000080, 0xd3d940f7, 0x18000080, 0xd3d940f8, 0x18000080, + 0xd3d940f9, 0x18000080, 0xd3d940fa, 0x18000080, 0xd3d940fb, 0x18000080, + 0xd3d940fc, 0x18000080, 0xd3d940fd, 0x18000080, 0xd3d940fe, 0x18000080, + 0xd3d940ff, 0x18000080, 0xb07c0000, 0xbe8a00ff, 0x000000f8, 0xbf11080a, + 0x7e000280, 0x7e020280, 0x7e040280, 0x7e060280, 0x7e080280, 0x7e0a0280, + 0x7e0c0280, 0x7e0e0280, 0x808a880a, 0xbe80320a, 0xbf84fff5, 0xbf9c0000, + 0xd28c0001, 0x0001007f, 0xd28d0001, 0x0002027e, 0x10020288, 0xb88b0904, + 0xb78b4000, 0xd1196a01, 0x00001701, 0xbe8a0087, 0xbefc00c1, 0xd89c4000, + 0x00020201, 0xd89cc080, 0x00040401, 0x320202ff, 0x00000800, 0x808a810a, + 0xbf84fff8, 0xbf810000, +}; + /* When below register arrays changed, please update gpr_reg_size, and sec_ded_counter_reg_size in function gfx_v9_0_do_edc_gpr_workarounds, to cover all gfx9 ASICs */ @@ -4073,6 +4301,23 @@ static const struct soc15_reg_entry vgpr_init_regs[] = { { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_STATIC_THREAD_MGMT_SE7), 0xffffffff }, }; +static const struct soc15_reg_entry vgpr_init_regs_arcturus[] = { + { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_RESOURCE_LIMITS), 0x0000000 }, + { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_NUM_THREAD_X), 0x40 }, + { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_NUM_THREAD_Y), 4 }, + { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_NUM_THREAD_Z), 1 }, + { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_PGM_RSRC1), 0xbf }, + { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_PGM_RSRC2), 0x400000 }, /* 64KB LDS */ + { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_STATIC_THREAD_MGMT_SE0), 0xffffffff }, + { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_STATIC_THREAD_MGMT_SE1), 0xffffffff }, + { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_STATIC_THREAD_MGMT_SE2), 0xffffffff }, + { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_STATIC_THREAD_MGMT_SE3), 0xffffffff }, + { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_STATIC_THREAD_MGMT_SE4), 0xffffffff }, + { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_STATIC_THREAD_MGMT_SE5), 0xffffffff }, + { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_STATIC_THREAD_MGMT_SE6), 0xffffffff }, + { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_STATIC_THREAD_MGMT_SE7), 0xffffffff }, +}; + static const struct soc15_reg_entry sgpr1_init_regs[] = { { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_RESOURCE_LIMITS), 0x0000000 }, { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_NUM_THREAD_X), 0x40 }, @@ -4141,7 +4386,6 @@ static const struct soc15_reg_entry gfx_v9_0_edc_counter_regs[] = { { SOC15_REG_ENTRY(GC, 0, mmTCC_EDC_CNT2), 0, 1, 16}, { SOC15_REG_ENTRY(GC, 0, mmTCA_EDC_CNT), 0, 1, 2}, { SOC15_REG_ENTRY(GC, 0, mmSQC_EDC_CNT3), 0, 4, 6}, - { SOC15_REG_ENTRY(HDP, 0, mmHDP_EDC_CNT), 0, 1, 1}, }; static int gfx_v9_0_do_edc_gds_workarounds(struct amdgpu_device *adev) @@ -4204,7 +4448,10 @@ static int gfx_v9_0_do_edc_gpr_workarounds(struct amdgpu_device *adev) adev->gfx.config.max_cu_per_sh * adev->gfx.config.max_sh_per_se; int sgpr_work_group_size = 5; - int gpr_reg_size = compute_dim_x / 16 + 6; + int gpr_reg_size = adev->gfx.config.max_shader_engines + 6; + int vgpr_init_shader_size; + const u32 *vgpr_init_shader_ptr; + const struct soc15_reg_entry *vgpr_init_regs_ptr; /* only support when RAS is enabled */ if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__GFX)) @@ -4214,6 +4461,16 @@ static int gfx_v9_0_do_edc_gpr_workarounds(struct amdgpu_device *adev) if (!ring->sched.ready) return 0; + if (adev->asic_type == CHIP_ARCTURUS) { + vgpr_init_shader_ptr = vgpr_init_compute_shader_arcturus; + vgpr_init_shader_size = sizeof(vgpr_init_compute_shader_arcturus); + vgpr_init_regs_ptr = vgpr_init_regs_arcturus; + } else { + vgpr_init_shader_ptr = vgpr_init_compute_shader; + vgpr_init_shader_size = sizeof(vgpr_init_compute_shader); + vgpr_init_regs_ptr = vgpr_init_regs; + } + total_size = (gpr_reg_size * 3 + 4 + 5 + 2) * 4; /* VGPRS */ total_size += @@ -4222,7 +4479,7 @@ static int gfx_v9_0_do_edc_gpr_workarounds(struct amdgpu_device *adev) (gpr_reg_size * 3 + 4 + 5 + 2) * 4; /* SGPRS2 */ total_size = ALIGN(total_size, 256); vgpr_offset = total_size; - total_size += ALIGN(sizeof(vgpr_init_compute_shader), 256); + total_size += ALIGN(vgpr_init_shader_size, 256); sgpr_offset = total_size; total_size += sizeof(sgpr_init_compute_shader); @@ -4235,8 +4492,8 @@ static int gfx_v9_0_do_edc_gpr_workarounds(struct amdgpu_device *adev) } /* load the compute shaders */ - for (i = 0; i < ARRAY_SIZE(vgpr_init_compute_shader); i++) - ib.ptr[i + (vgpr_offset / 4)] = vgpr_init_compute_shader[i]; + for (i = 0; i < vgpr_init_shader_size/sizeof(u32); i++) + ib.ptr[i + (vgpr_offset / 4)] = vgpr_init_shader_ptr[i]; for (i = 0; i < ARRAY_SIZE(sgpr_init_compute_shader); i++) ib.ptr[i + (sgpr_offset / 4)] = sgpr_init_compute_shader[i]; @@ -4248,9 +4505,9 @@ static int gfx_v9_0_do_edc_gpr_workarounds(struct amdgpu_device *adev) /* write the register state for the compute dispatch */ for (i = 0; i < gpr_reg_size; i++) { ib.ptr[ib.length_dw++] = PACKET3(PACKET3_SET_SH_REG, 1); - ib.ptr[ib.length_dw++] = SOC15_REG_ENTRY_OFFSET(vgpr_init_regs[i]) + ib.ptr[ib.length_dw++] = SOC15_REG_ENTRY_OFFSET(vgpr_init_regs_ptr[i]) - PACKET3_SET_SH_REG_START; - ib.ptr[ib.length_dw++] = vgpr_init_regs[i].reg_value; + ib.ptr[ib.length_dw++] = vgpr_init_regs_ptr[i].reg_value; } /* write the shader start address: mmCOMPUTE_PGM_LO, mmCOMPUTE_PGM_HI */ gpu_addr = (ib.gpu_addr + (u64)vgpr_offset) >> 8; @@ -4262,7 +4519,7 @@ static int gfx_v9_0_do_edc_gpr_workarounds(struct amdgpu_device *adev) /* write dispatch packet */ ib.ptr[ib.length_dw++] = PACKET3(PACKET3_DISPATCH_DIRECT, 3); - ib.ptr[ib.length_dw++] = compute_dim_x; /* x */ + ib.ptr[ib.length_dw++] = compute_dim_x * 2; /* x */ ib.ptr[ib.length_dw++] = 1; /* y */ ib.ptr[ib.length_dw++] = 1; /* z */ ib.ptr[ib.length_dw++] = @@ -4342,18 +4599,6 @@ static int gfx_v9_0_do_edc_gpr_workarounds(struct amdgpu_device *adev) goto fail; } - switch (adev->asic_type) - { - case CHIP_VEGA20: - gfx_v9_0_clear_ras_edc_counter(adev); - break; - case CHIP_ARCTURUS: - gfx_v9_4_clear_ras_edc_counter(adev); - break; - default: - break; - } - fail: amdgpu_ib_free(adev, &ib, NULL); dma_fence_put(f); @@ -4401,6 +4646,10 @@ static int gfx_v9_0_ecc_late_init(void *handle) if (r) return r; + if (adev->gfx.funcs && + adev->gfx.funcs->reset_ras_error_count) + adev->gfx.funcs->reset_ras_error_count(adev); + r = amdgpu_gfx_ras_late_init(adev); if (r) return r; @@ -4705,6 +4954,47 @@ static int gfx_v9_0_update_gfx_clock_gating(struct amdgpu_device *adev, return 0; } +static void gfx_v9_0_update_spm_vmid(struct amdgpu_device *adev, unsigned vmid) +{ + u32 data; + + data = RREG32_SOC15(GC, 0, mmRLC_SPM_MC_CNTL); + + data &= ~RLC_SPM_MC_CNTL__RLC_SPM_VMID_MASK; + data |= (vmid & RLC_SPM_MC_CNTL__RLC_SPM_VMID_MASK) << RLC_SPM_MC_CNTL__RLC_SPM_VMID__SHIFT; + + WREG32_SOC15(GC, 0, mmRLC_SPM_MC_CNTL, data); +} + +static bool gfx_v9_0_check_rlcg_range(struct amdgpu_device *adev, + uint32_t offset, + struct soc15_reg_rlcg *entries, int arr_size) +{ + int i; + uint32_t reg; + + if (!entries) + return false; + + for (i = 0; i < arr_size; i++) { + const struct soc15_reg_rlcg *entry; + + entry = &entries[i]; + reg = adev->reg_offset[entry->hwip][entry->instance][entry->segment] + entry->reg; + if (offset == reg) + return true; + } + + return false; +} + +static bool gfx_v9_0_is_rlcg_access_range(struct amdgpu_device *adev, u32 offset) +{ + return gfx_v9_0_check_rlcg_range(adev, offset, + (void *)rlcg_access_gc_9_0, + ARRAY_SIZE(rlcg_access_gc_9_0)); +} + static const struct amdgpu_rlc_funcs gfx_v9_0_rlc_funcs = { .is_rlc_enabled = gfx_v9_0_is_rlc_enabled, .set_safe_mode = gfx_v9_0_set_safe_mode, @@ -4716,7 +5006,10 @@ static const struct amdgpu_rlc_funcs gfx_v9_0_rlc_funcs = { .resume = gfx_v9_0_rlc_resume, .stop = gfx_v9_0_rlc_stop, .reset = gfx_v9_0_rlc_reset, - .start = gfx_v9_0_rlc_start + .start = gfx_v9_0_rlc_start, + .update_spm_vmid = gfx_v9_0_update_spm_vmid, + .rlcg_wreg = gfx_v9_0_rlcg_wreg, + .is_rlcg_access_range = gfx_v9_0_is_rlcg_access_range, }; static int gfx_v9_0_set_powergating_state(void *handle, @@ -4919,7 +5212,7 @@ static void gfx_v9_0_ring_emit_ib_gfx(struct amdgpu_ring *ring, if (amdgpu_sriov_vf(ring->adev) && (ib->flags & AMDGPU_IB_FLAG_PREEMPT)) { control |= INDIRECT_BUFFER_PRE_ENB(1); - if (!(ib->flags & AMDGPU_IB_FLAG_CE)) + if (!(ib->flags & AMDGPU_IB_FLAG_CE) && vmid) gfx_v9_0_ring_emit_de_meta(ring); } @@ -5044,105 +5337,6 @@ static u64 gfx_v9_0_ring_get_wptr_compute(struct amdgpu_ring *ring) return wptr; } -static void gfx_v9_0_ring_set_pipe_percent(struct amdgpu_ring *ring, - bool acquire) -{ - struct amdgpu_device *adev = ring->adev; - int pipe_num, tmp, reg; - int pipe_percent = acquire ? SPI_WCL_PIPE_PERCENT_GFX__VALUE_MASK : 0x1; - - pipe_num = ring->me * adev->gfx.mec.num_pipe_per_mec + ring->pipe; - - /* first me only has 2 entries, GFX and HP3D */ - if (ring->me > 0) - pipe_num -= 2; - - reg = SOC15_REG_OFFSET(GC, 0, mmSPI_WCL_PIPE_PERCENT_GFX) + pipe_num; - tmp = RREG32(reg); - tmp = REG_SET_FIELD(tmp, SPI_WCL_PIPE_PERCENT_GFX, VALUE, pipe_percent); - WREG32(reg, tmp); -} - -static void gfx_v9_0_pipe_reserve_resources(struct amdgpu_device *adev, - struct amdgpu_ring *ring, - bool acquire) -{ - int i, pipe; - bool reserve; - struct amdgpu_ring *iring; - - mutex_lock(&adev->gfx.pipe_reserve_mutex); - pipe = amdgpu_gfx_mec_queue_to_bit(adev, ring->me, ring->pipe, 0); - if (acquire) - set_bit(pipe, adev->gfx.pipe_reserve_bitmap); - else - clear_bit(pipe, adev->gfx.pipe_reserve_bitmap); - - if (!bitmap_weight(adev->gfx.pipe_reserve_bitmap, AMDGPU_MAX_COMPUTE_QUEUES)) { - /* Clear all reservations - everyone reacquires all resources */ - for (i = 0; i < adev->gfx.num_gfx_rings; ++i) - gfx_v9_0_ring_set_pipe_percent(&adev->gfx.gfx_ring[i], - true); - - for (i = 0; i < adev->gfx.num_compute_rings; ++i) - gfx_v9_0_ring_set_pipe_percent(&adev->gfx.compute_ring[i], - true); - } else { - /* Lower all pipes without a current reservation */ - for (i = 0; i < adev->gfx.num_gfx_rings; ++i) { - iring = &adev->gfx.gfx_ring[i]; - pipe = amdgpu_gfx_mec_queue_to_bit(adev, - iring->me, - iring->pipe, - 0); - reserve = test_bit(pipe, adev->gfx.pipe_reserve_bitmap); - gfx_v9_0_ring_set_pipe_percent(iring, reserve); - } - - for (i = 0; i < adev->gfx.num_compute_rings; ++i) { - iring = &adev->gfx.compute_ring[i]; - pipe = amdgpu_gfx_mec_queue_to_bit(adev, - iring->me, - iring->pipe, - 0); - reserve = test_bit(pipe, adev->gfx.pipe_reserve_bitmap); - gfx_v9_0_ring_set_pipe_percent(iring, reserve); - } - } - - mutex_unlock(&adev->gfx.pipe_reserve_mutex); -} - -static void gfx_v9_0_hqd_set_priority(struct amdgpu_device *adev, - struct amdgpu_ring *ring, - bool acquire) -{ - uint32_t pipe_priority = acquire ? 0x2 : 0x0; - uint32_t queue_priority = acquire ? 0xf : 0x0; - - mutex_lock(&adev->srbm_mutex); - soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0); - - WREG32_SOC15_RLC(GC, 0, mmCP_HQD_PIPE_PRIORITY, pipe_priority); - WREG32_SOC15_RLC(GC, 0, mmCP_HQD_QUEUE_PRIORITY, queue_priority); - - soc15_grbm_select(adev, 0, 0, 0, 0); - mutex_unlock(&adev->srbm_mutex); -} - -static void gfx_v9_0_ring_set_priority_compute(struct amdgpu_ring *ring, - enum drm_sched_priority priority) -{ - struct amdgpu_device *adev = ring->adev; - bool acquire = priority == DRM_SCHED_PRIORITY_HIGH_HW; - - if (ring->funcs->type != AMDGPU_RING_TYPE_COMPUTE) - return; - - gfx_v9_0_hqd_set_priority(adev, ring, acquire); - gfx_v9_0_pipe_reserve_resources(adev, ring, acquire); -} - static void gfx_v9_0_ring_set_wptr_compute(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; @@ -6322,10 +6516,13 @@ static int gfx_v9_0_ras_error_count(const struct soc15_reg_entry *reg, return 0; } -static void gfx_v9_0_clear_ras_edc_counter(struct amdgpu_device *adev) +static void gfx_v9_0_reset_ras_error_count(struct amdgpu_device *adev) { int i, j, k; + if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__GFX)) + return; + /* read back registers to clear the counters */ mutex_lock(&adev->grbm_idx_mutex); for (i = 0; i < ARRAY_SIZE(gfx_v9_0_edc_counter_regs); i++) { @@ -6513,7 +6710,6 @@ static const struct amdgpu_ring_funcs gfx_v9_0_ring_funcs_compute = { .test_ib = gfx_v9_0_ring_test_ib, .insert_nop = amdgpu_ring_insert_nop, .pad_ib = amdgpu_ring_generic_pad_ib, - .set_priority = gfx_v9_0_ring_set_priority_compute, .emit_wreg = gfx_v9_0_ring_emit_wreg, .emit_reg_wait = gfx_v9_0_ring_emit_reg_wait, .emit_reg_write_reg_wait = gfx_v9_0_ring_emit_reg_write_reg_wait, diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.c index f099f13d7f1e..cceb46faf212 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.c @@ -893,10 +893,13 @@ int gfx_v9_4_query_ras_error_count(struct amdgpu_device *adev, return 0; } -void gfx_v9_4_clear_ras_edc_counter(struct amdgpu_device *adev) +void gfx_v9_4_reset_ras_error_count(struct amdgpu_device *adev) { int i, j, k; + if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__GFX)) + return; + mutex_lock(&adev->grbm_idx_mutex); for (i = 0; i < ARRAY_SIZE(gfx_v9_4_edc_counter_regs); i++) { for (j = 0; j < gfx_v9_4_edc_counter_regs[i].se_num; j++) { diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.h b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.h index 2e3f6f755ad4..1ffecc5c0f0a 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.h +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.h @@ -32,4 +32,6 @@ int gfx_v9_4_query_ras_error_count(struct amdgpu_device *adev, int gfx_v9_4_ras_error_inject(struct amdgpu_device *adev, void *inject_if); +void gfx_v9_4_reset_ras_error_count(struct amdgpu_device *adev); + #endif /* __GFX_V9_4_H__ */ diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c index e0654a216ab5..cc866c367939 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c @@ -81,24 +81,31 @@ static void gfxhub_v2_0_init_system_aperture_regs(struct amdgpu_device *adev) { uint64_t value; - /* Disable AGP. */ - WREG32_SOC15(GC, 0, mmGCMC_VM_AGP_BASE, 0); - WREG32_SOC15(GC, 0, mmGCMC_VM_AGP_TOP, 0); - WREG32_SOC15(GC, 0, mmGCMC_VM_AGP_BOT, 0x00FFFFFF); - - /* Program the system aperture low logical page number. */ - WREG32_SOC15(GC, 0, mmGCMC_VM_SYSTEM_APERTURE_LOW_ADDR, - adev->gmc.vram_start >> 18); - WREG32_SOC15(GC, 0, mmGCMC_VM_SYSTEM_APERTURE_HIGH_ADDR, - adev->gmc.vram_end >> 18); - - /* Set default page address. */ - value = adev->vram_scratch.gpu_addr - adev->gmc.vram_start - + adev->vm_manager.vram_base_offset; - WREG32_SOC15(GC, 0, mmGCMC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_LSB, - (u32)(value >> 12)); - WREG32_SOC15(GC, 0, mmGCMC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_MSB, - (u32)(value >> 44)); + if (!amdgpu_sriov_vf(adev)) { + /* + * the new L1 policy will block SRIOV guest from writing + * these regs, and they will be programed at host. + * so skip programing these regs. + */ + /* Disable AGP. */ + WREG32_SOC15(GC, 0, mmGCMC_VM_AGP_BASE, 0); + WREG32_SOC15(GC, 0, mmGCMC_VM_AGP_TOP, 0); + WREG32_SOC15(GC, 0, mmGCMC_VM_AGP_BOT, 0x00FFFFFF); + + /* Program the system aperture low logical page number. */ + WREG32_SOC15(GC, 0, mmGCMC_VM_SYSTEM_APERTURE_LOW_ADDR, + adev->gmc.vram_start >> 18); + WREG32_SOC15(GC, 0, mmGCMC_VM_SYSTEM_APERTURE_HIGH_ADDR, + adev->gmc.vram_end >> 18); + + /* Set default page address. */ + value = adev->vram_scratch.gpu_addr - adev->gmc.vram_start + + adev->vm_manager.vram_base_offset; + WREG32_SOC15(GC, 0, mmGCMC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_LSB, + (u32)(value >> 12)); + WREG32_SOC15(GC, 0, mmGCMC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_MSB, + (u32)(value >> 44)); + } /* Program "protection fault". */ WREG32_SOC15(GC, 0, mmGCVM_L2_PROTECTION_FAULT_DEFAULT_ADDR_LO32, @@ -260,18 +267,6 @@ static void gfxhub_v2_0_program_invalidation(struct amdgpu_device *adev) int gfxhub_v2_0_gart_enable(struct amdgpu_device *adev) { - if (amdgpu_sriov_vf(adev)) { - /* - * GCMC_VM_FB_LOCATION_BASE/TOP is NULL for VF, becuase they are - * VF copy registers so vbios post doesn't program them, for - * SRIOV driver need to program them - */ - WREG32_SOC15(GC, 0, mmGCMC_VM_FB_LOCATION_BASE, - adev->gmc.vram_start >> 24); - WREG32_SOC15(GC, 0, mmGCMC_VM_FB_LOCATION_TOP, - adev->gmc.vram_end >> 24); - } - /* GART Enable. */ gfxhub_v2_0_init_gart_aperture_regs(adev); gfxhub_v2_0_init_system_aperture_regs(adev); diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c index 10171acbf3e1..8606f877478f 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c @@ -922,32 +922,25 @@ static int gmc_v9_0_late_init(void *handle) if (r) return r; /* Check if ecc is available */ - if (!amdgpu_sriov_vf(adev)) { - switch (adev->asic_type) { - case CHIP_VEGA10: - case CHIP_VEGA20: - case CHIP_ARCTURUS: - r = amdgpu_atomfirmware_mem_ecc_supported(adev); - if (!r) { - DRM_INFO("ECC is not present.\n"); - if (adev->df.funcs->enable_ecc_force_par_wr_rmw) - adev->df.funcs->enable_ecc_force_par_wr_rmw(adev, false); - } else { - DRM_INFO("ECC is active.\n"); - } - - r = amdgpu_atomfirmware_sram_ecc_supported(adev); - if (!r) { - DRM_INFO("SRAM ECC is not present.\n"); - } else { - DRM_INFO("SRAM ECC is active.\n"); - } - break; - default: - break; - } + if (!amdgpu_sriov_vf(adev) && (adev->asic_type == CHIP_VEGA10)) { + r = amdgpu_atomfirmware_mem_ecc_supported(adev); + if (!r) { + DRM_INFO("ECC is not present.\n"); + if (adev->df.funcs->enable_ecc_force_par_wr_rmw) + adev->df.funcs->enable_ecc_force_par_wr_rmw(adev, false); + } else + DRM_INFO("ECC is active.\n"); + + r = amdgpu_atomfirmware_sram_ecc_supported(adev); + if (!r) + DRM_INFO("SRAM ECC is not present.\n"); + else + DRM_INFO("SRAM ECC is active.\n"); } + if (adev->mmhub.funcs && adev->mmhub.funcs->reset_ras_error_count) + adev->mmhub.funcs->reset_ras_error_count(adev); + r = amdgpu_gmc_ras_late_init(adev); if (r) return r; diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c index ff2e6e1ccde7..6173951db7b4 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c @@ -693,7 +693,7 @@ static int jpeg_v2_0_set_clockgating_state(void *handle, bool enable = (state == AMD_CG_STATE_GATE); if (enable) { - if (jpeg_v2_0_is_idle(handle)) + if (!jpeg_v2_0_is_idle(handle)) return -EBUSY; jpeg_v2_0_enable_clock_gating(adev); } else { diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c index c6d046df4b70..c04c2078a7c1 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c @@ -477,7 +477,7 @@ static int jpeg_v2_5_set_clockgating_state(void *handle, continue; if (enable) { - if (jpeg_v2_5_is_idle(handle)) + if (!jpeg_v2_5_is_idle(handle)) return -EBUSY; jpeg_v2_5_enable_clock_gating(adev, i); } else { diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c index 49a3a56ec017..396c2a624de0 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c @@ -747,7 +747,19 @@ static void mmhub_v1_0_query_ras_error_count(struct amdgpu_device *adev, err_data->ue_count += ded_count; } +static void mmhub_v1_0_reset_ras_error_count(struct amdgpu_device *adev) +{ + uint32_t i; + + /* read back edc counter registers to reset the counters to 0 */ + if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__MMHUB)) { + for (i = 0; i < ARRAY_SIZE(mmhub_v1_0_edc_cnt_regs); i++) + RREG32(SOC15_REG_ENTRY_OFFSET(mmhub_v1_0_edc_cnt_regs[i])); + } +} + const struct amdgpu_mmhub_funcs mmhub_v1_0_funcs = { .ras_late_init = amdgpu_mmhub_ras_late_init, .query_ras_error_count = mmhub_v1_0_query_ras_error_count, + .reset_ras_error_count = mmhub_v1_0_reset_ras_error_count, }; diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c index bde189680521..fb3f228458e5 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c @@ -72,11 +72,18 @@ static void mmhub_v2_0_init_system_aperture_regs(struct amdgpu_device *adev) WREG32_SOC15(MMHUB, 0, mmMMMC_VM_AGP_TOP, 0); WREG32_SOC15(MMHUB, 0, mmMMMC_VM_AGP_BOT, 0x00FFFFFF); - /* Program the system aperture low logical page number. */ - WREG32_SOC15(MMHUB, 0, mmMMMC_VM_SYSTEM_APERTURE_LOW_ADDR, - adev->gmc.vram_start >> 18); - WREG32_SOC15(MMHUB, 0, mmMMMC_VM_SYSTEM_APERTURE_HIGH_ADDR, - adev->gmc.vram_end >> 18); + if (!amdgpu_sriov_vf(adev)) { + /* + * the new L1 policy will block SRIOV guest from writing + * these regs, and they will be programed at host. + * so skip programing these regs. + */ + /* Program the system aperture low logical page number. */ + WREG32_SOC15(MMHUB, 0, mmMMMC_VM_SYSTEM_APERTURE_LOW_ADDR, + adev->gmc.vram_start >> 18); + WREG32_SOC15(MMHUB, 0, mmMMMC_VM_SYSTEM_APERTURE_HIGH_ADDR, + adev->gmc.vram_end >> 18); + } /* Set default page address. */ value = adev->vram_scratch.gpu_addr - adev->gmc.vram_start + @@ -247,18 +254,6 @@ static void mmhub_v2_0_program_invalidation(struct amdgpu_device *adev) int mmhub_v2_0_gart_enable(struct amdgpu_device *adev) { - if (amdgpu_sriov_vf(adev)) { - /* - * MMMC_VM_FB_LOCATION_BASE/TOP is NULL for VF, becuase they are - * VF copy registers so vbios post doesn't program them, for - * SRIOV driver need to program them - */ - WREG32_SOC15(MMHUB, 0, mmMMMC_VM_FB_LOCATION_BASE, - adev->gmc.vram_start >> 24); - WREG32_SOC15(MMHUB, 0, mmMMMC_VM_FB_LOCATION_TOP, - adev->gmc.vram_end >> 24); - } - /* GART Enable. */ mmhub_v2_0_init_gart_aperture_regs(adev); mmhub_v2_0_init_system_aperture_regs(adev); diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c index a5281df8d84f..0d413fabd015 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c @@ -1596,7 +1596,19 @@ static void mmhub_v9_4_query_ras_error_count(struct amdgpu_device *adev, err_data->ue_count += ded_count; } +static void mmhub_v9_4_reset_ras_error_count(struct amdgpu_device *adev) +{ + uint32_t i; + + /* read back edc counter registers to reset the counters to 0 */ + if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__MMHUB)) { + for (i = 0; i < ARRAY_SIZE(mmhub_v9_4_edc_cnt_regs); i++) + RREG32(SOC15_REG_ENTRY_OFFSET(mmhub_v9_4_edc_cnt_regs[i])); + } +} + const struct amdgpu_mmhub_funcs mmhub_v9_4_funcs = { .ras_late_init = amdgpu_mmhub_ras_late_init, .query_ras_error_count = mmhub_v9_4_query_ras_error_count, + .reset_ras_error_count = mmhub_v9_4_reset_ras_error_count, }; diff --git a/drivers/gpu/drm/amd/amdgpu/mmsch_v2_0.h b/drivers/gpu/drm/amd/amdgpu/mmsch_v2_0.h new file mode 100644 index 000000000000..1b5086c7d4e6 --- /dev/null +++ b/drivers/gpu/drm/amd/amdgpu/mmsch_v2_0.h @@ -0,0 +1,338 @@ +/* + * Copyright 2019 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ + +#ifndef __MMSCH_V2_0_H__ +#define __MMSCH_V2_0_H__ + +// addressBlock: uvd0_mmsch_dec +// base address: 0x1e000 +#define mmMMSCH_UCODE_ADDR 0x0000 +#define mmMMSCH_UCODE_ADDR_BASE_IDX 0 +#define mmMMSCH_UCODE_DATA 0x0001 +#define mmMMSCH_UCODE_DATA_BASE_IDX 0 +#define mmMMSCH_SRAM_ADDR 0x0002 +#define mmMMSCH_SRAM_ADDR_BASE_IDX 0 +#define mmMMSCH_SRAM_DATA 0x0003 +#define mmMMSCH_SRAM_DATA_BASE_IDX 0 +#define mmMMSCH_VF_SRAM_OFFSET 0x0004 +#define mmMMSCH_VF_SRAM_OFFSET_BASE_IDX 0 +#define mmMMSCH_DB_SRAM_OFFSET 0x0005 +#define mmMMSCH_DB_SRAM_OFFSET_BASE_IDX 0 +#define mmMMSCH_CTX_SRAM_OFFSET 0x0006 +#define mmMMSCH_CTX_SRAM_OFFSET_BASE_IDX 0 +#define mmMMSCH_CTL 0x0007 +#define mmMMSCH_CTL_BASE_IDX 0 +#define mmMMSCH_INTR 0x0008 +#define mmMMSCH_INTR_BASE_IDX 0 +#define mmMMSCH_INTR_ACK 0x0009 +#define mmMMSCH_INTR_ACK_BASE_IDX 0 +#define mmMMSCH_INTR_STATUS 0x000a +#define mmMMSCH_INTR_STATUS_BASE_IDX 0 +#define mmMMSCH_VF_VMID 0x000b +#define mmMMSCH_VF_VMID_BASE_IDX 0 +#define mmMMSCH_VF_CTX_ADDR_LO 0x000c +#define mmMMSCH_VF_CTX_ADDR_LO_BASE_IDX 0 +#define mmMMSCH_VF_CTX_ADDR_HI 0x000d +#define mmMMSCH_VF_CTX_ADDR_HI_BASE_IDX 0 +#define mmMMSCH_VF_CTX_SIZE 0x000e +#define mmMMSCH_VF_CTX_SIZE_BASE_IDX 0 +#define mmMMSCH_VF_GPCOM_ADDR_LO 0x000f +#define mmMMSCH_VF_GPCOM_ADDR_LO_BASE_IDX 0 +#define mmMMSCH_VF_GPCOM_ADDR_HI 0x0010 +#define mmMMSCH_VF_GPCOM_ADDR_HI_BASE_IDX 0 +#define mmMMSCH_VF_GPCOM_SIZE 0x0011 +#define mmMMSCH_VF_GPCOM_SIZE_BASE_IDX 0 +#define mmMMSCH_VF_MAILBOX_HOST 0x0012 +#define mmMMSCH_VF_MAILBOX_HOST_BASE_IDX 0 +#define mmMMSCH_VF_MAILBOX_RESP 0x0013 +#define mmMMSCH_VF_MAILBOX_RESP_BASE_IDX 0 +#define mmMMSCH_VF_MAILBOX_0 0x0014 +#define mmMMSCH_VF_MAILBOX_0_BASE_IDX 0 +#define mmMMSCH_VF_MAILBOX_0_RESP 0x0015 +#define mmMMSCH_VF_MAILBOX_0_RESP_BASE_IDX 0 +#define mmMMSCH_VF_MAILBOX_1 0x0016 +#define mmMMSCH_VF_MAILBOX_1_BASE_IDX 0 +#define mmMMSCH_VF_MAILBOX_1_RESP 0x0017 +#define mmMMSCH_VF_MAILBOX_1_RESP_BASE_IDX 0 +#define mmMMSCH_CNTL 0x001c +#define mmMMSCH_CNTL_BASE_IDX 0 +#define mmMMSCH_NONCACHE_OFFSET0 0x001d +#define mmMMSCH_NONCACHE_OFFSET0_BASE_IDX 0 +#define mmMMSCH_NONCACHE_SIZE0 0x001e +#define mmMMSCH_NONCACHE_SIZE0_BASE_IDX 0 +#define mmMMSCH_NONCACHE_OFFSET1 0x001f +#define mmMMSCH_NONCACHE_OFFSET1_BASE_IDX 0 +#define mmMMSCH_NONCACHE_SIZE1 0x0020 +#define mmMMSCH_NONCACHE_SIZE1_BASE_IDX 0 +#define mmMMSCH_PDEBUG_STATUS 0x0021 +#define mmMMSCH_PDEBUG_STATUS_BASE_IDX 0 +#define mmMMSCH_PDEBUG_DATA_32UPPERBITS 0x0022 +#define mmMMSCH_PDEBUG_DATA_32UPPERBITS_BASE_IDX 0 +#define mmMMSCH_PDEBUG_DATA_32LOWERBITS 0x0023 +#define mmMMSCH_PDEBUG_DATA_32LOWERBITS_BASE_IDX 0 +#define mmMMSCH_PDEBUG_EPC 0x0024 +#define mmMMSCH_PDEBUG_EPC_BASE_IDX 0 +#define mmMMSCH_PDEBUG_EXCCAUSE 0x0025 +#define mmMMSCH_PDEBUG_EXCCAUSE_BASE_IDX 0 +#define mmMMSCH_PROC_STATE1 0x0026 +#define mmMMSCH_PROC_STATE1_BASE_IDX 0 +#define mmMMSCH_LAST_MC_ADDR 0x0027 +#define mmMMSCH_LAST_MC_ADDR_BASE_IDX 0 +#define mmMMSCH_LAST_MEM_ACCESS_HI 0x0028 +#define mmMMSCH_LAST_MEM_ACCESS_HI_BASE_IDX 0 +#define mmMMSCH_LAST_MEM_ACCESS_LO 0x0029 +#define mmMMSCH_LAST_MEM_ACCESS_LO_BASE_IDX 0 +#define mmMMSCH_IOV_ACTIVE_FCN_ID 0x002a +#define mmMMSCH_IOV_ACTIVE_FCN_ID_BASE_IDX 0 +#define mmMMSCH_SCRATCH_0 0x002b +#define mmMMSCH_SCRATCH_0_BASE_IDX 0 +#define mmMMSCH_SCRATCH_1 0x002c +#define mmMMSCH_SCRATCH_1_BASE_IDX 0 +#define mmMMSCH_GPUIOV_SCH_BLOCK_0 0x002d +#define mmMMSCH_GPUIOV_SCH_BLOCK_0_BASE_IDX 0 +#define mmMMSCH_GPUIOV_CMD_CONTROL_0 0x002e +#define mmMMSCH_GPUIOV_CMD_CONTROL_0_BASE_IDX 0 +#define mmMMSCH_GPUIOV_CMD_STATUS_0 0x002f +#define mmMMSCH_GPUIOV_CMD_STATUS_0_BASE_IDX 0 +#define mmMMSCH_GPUIOV_VM_BUSY_STATUS_0 0x0030 +#define mmMMSCH_GPUIOV_VM_BUSY_STATUS_0_BASE_IDX 0 +#define mmMMSCH_GPUIOV_ACTIVE_FCNS_0 0x0031 +#define mmMMSCH_GPUIOV_ACTIVE_FCNS_0_BASE_IDX 0 +#define mmMMSCH_GPUIOV_ACTIVE_FCN_ID_0 0x0032 +#define mmMMSCH_GPUIOV_ACTIVE_FCN_ID_0_BASE_IDX 0 +#define mmMMSCH_GPUIOV_DW6_0 0x0033 +#define mmMMSCH_GPUIOV_DW6_0_BASE_IDX 0 +#define mmMMSCH_GPUIOV_DW7_0 0x0034 +#define mmMMSCH_GPUIOV_DW7_0_BASE_IDX 0 +#define mmMMSCH_GPUIOV_DW8_0 0x0035 +#define mmMMSCH_GPUIOV_DW8_0_BASE_IDX 0 +#define mmMMSCH_GPUIOV_SCH_BLOCK_1 0x0036 +#define mmMMSCH_GPUIOV_SCH_BLOCK_1_BASE_IDX 0 +#define mmMMSCH_GPUIOV_CMD_CONTROL_1 0x0037 +#define mmMMSCH_GPUIOV_CMD_CONTROL_1_BASE_IDX 0 +#define mmMMSCH_GPUIOV_CMD_STATUS_1 0x0038 +#define mmMMSCH_GPUIOV_CMD_STATUS_1_BASE_IDX 0 +#define mmMMSCH_GPUIOV_VM_BUSY_STATUS_1 0x0039 +#define mmMMSCH_GPUIOV_VM_BUSY_STATUS_1_BASE_IDX 0 +#define mmMMSCH_GPUIOV_ACTIVE_FCNS_1 0x003a +#define mmMMSCH_GPUIOV_ACTIVE_FCNS_1_BASE_IDX 0 +#define mmMMSCH_GPUIOV_ACTIVE_FCN_ID_1 0x003b +#define mmMMSCH_GPUIOV_ACTIVE_FCN_ID_1_BASE_IDX 0 +#define mmMMSCH_GPUIOV_DW6_1 0x003c +#define mmMMSCH_GPUIOV_DW6_1_BASE_IDX 0 +#define mmMMSCH_GPUIOV_DW7_1 0x003d +#define mmMMSCH_GPUIOV_DW7_1_BASE_IDX 0 +#define mmMMSCH_GPUIOV_DW8_1 0x003e +#define mmMMSCH_GPUIOV_DW8_1_BASE_IDX 0 +#define mmMMSCH_GPUIOV_CNTXT 0x003f +#define mmMMSCH_GPUIOV_CNTXT_BASE_IDX 0 +#define mmMMSCH_SCRATCH_2 0x0040 +#define mmMMSCH_SCRATCH_2_BASE_IDX 0 +#define mmMMSCH_SCRATCH_3 0x0041 +#define mmMMSCH_SCRATCH_3_BASE_IDX 0 +#define mmMMSCH_SCRATCH_4 0x0042 +#define mmMMSCH_SCRATCH_4_BASE_IDX 0 +#define mmMMSCH_SCRATCH_5 0x0043 +#define mmMMSCH_SCRATCH_5_BASE_IDX 0 +#define mmMMSCH_SCRATCH_6 0x0044 +#define mmMMSCH_SCRATCH_6_BASE_IDX 0 +#define mmMMSCH_SCRATCH_7 0x0045 +#define mmMMSCH_SCRATCH_7_BASE_IDX 0 +#define mmMMSCH_VFID_FIFO_HEAD_0 0x0046 +#define mmMMSCH_VFID_FIFO_HEAD_0_BASE_IDX 0 +#define mmMMSCH_VFID_FIFO_TAIL_0 0x0047 +#define mmMMSCH_VFID_FIFO_TAIL_0_BASE_IDX 0 +#define mmMMSCH_VFID_FIFO_HEAD_1 0x0048 +#define mmMMSCH_VFID_FIFO_HEAD_1_BASE_IDX 0 +#define mmMMSCH_VFID_FIFO_TAIL_1 0x0049 +#define mmMMSCH_VFID_FIFO_TAIL_1_BASE_IDX 0 +#define mmMMSCH_NACK_STATUS 0x004a +#define mmMMSCH_NACK_STATUS_BASE_IDX 0 +#define mmMMSCH_VF_MAILBOX0_DATA 0x004b +#define mmMMSCH_VF_MAILBOX0_DATA_BASE_IDX 0 +#define mmMMSCH_VF_MAILBOX1_DATA 0x004c +#define mmMMSCH_VF_MAILBOX1_DATA_BASE_IDX 0 +#define mmMMSCH_GPUIOV_SCH_BLOCK_IP_0 0x004d +#define mmMMSCH_GPUIOV_SCH_BLOCK_IP_0_BASE_IDX 0 +#define mmMMSCH_GPUIOV_CMD_STATUS_IP_0 0x004e +#define mmMMSCH_GPUIOV_CMD_STATUS_IP_0_BASE_IDX 0 +#define mmMMSCH_GPUIOV_ACTIVE_FCN_ID_IP_0 0x004f +#define mmMMSCH_GPUIOV_ACTIVE_FCN_ID_IP_0_BASE_IDX 0 +#define mmMMSCH_GPUIOV_SCH_BLOCK_IP_1 0x0050 +#define mmMMSCH_GPUIOV_SCH_BLOCK_IP_1_BASE_IDX 0 +#define mmMMSCH_GPUIOV_CMD_STATUS_IP_1 0x0051 +#define mmMMSCH_GPUIOV_CMD_STATUS_IP_1_BASE_IDX 0 +#define mmMMSCH_GPUIOV_ACTIVE_FCN_ID_IP_1 0x0052 +#define mmMMSCH_GPUIOV_ACTIVE_FCN_ID_IP_1_BASE_IDX 0 +#define mmMMSCH_GPUIOV_CNTXT_IP 0x0053 +#define mmMMSCH_GPUIOV_CNTXT_IP_BASE_IDX 0 +#define mmMMSCH_GPUIOV_SCH_BLOCK_2 0x0054 +#define mmMMSCH_GPUIOV_SCH_BLOCK_2_BASE_IDX 0 +#define mmMMSCH_GPUIOV_CMD_CONTROL_2 0x0055 +#define mmMMSCH_GPUIOV_CMD_CONTROL_2_BASE_IDX 0 +#define mmMMSCH_GPUIOV_CMD_STATUS_2 0x0056 +#define mmMMSCH_GPUIOV_CMD_STATUS_2_BASE_IDX 0 +#define mmMMSCH_GPUIOV_VM_BUSY_STATUS_2 0x0057 +#define mmMMSCH_GPUIOV_VM_BUSY_STATUS_2_BASE_IDX 0 +#define mmMMSCH_GPUIOV_ACTIVE_FCNS_2 0x0058 +#define mmMMSCH_GPUIOV_ACTIVE_FCNS_2_BASE_IDX 0 +#define mmMMSCH_GPUIOV_ACTIVE_FCN_ID_2 0x0059 +#define mmMMSCH_GPUIOV_ACTIVE_FCN_ID_2_BASE_IDX 0 +#define mmMMSCH_GPUIOV_DW6_2 0x005a +#define mmMMSCH_GPUIOV_DW6_2_BASE_IDX 0 +#define mmMMSCH_GPUIOV_DW7_2 0x005b +#define mmMMSCH_GPUIOV_DW7_2_BASE_IDX 0 +#define mmMMSCH_GPUIOV_DW8_2 0x005c +#define mmMMSCH_GPUIOV_DW8_2_BASE_IDX 0 +#define mmMMSCH_GPUIOV_SCH_BLOCK_IP_2 0x005d +#define mmMMSCH_GPUIOV_SCH_BLOCK_IP_2_BASE_IDX 0 +#define mmMMSCH_GPUIOV_CMD_STATUS_IP_2 0x005e +#define mmMMSCH_GPUIOV_CMD_STATUS_IP_2_BASE_IDX 0 +#define mmMMSCH_GPUIOV_ACTIVE_FCN_ID_IP_2 0x005f +#define mmMMSCH_GPUIOV_ACTIVE_FCN_ID_IP_2_BASE_IDX 0 +#define mmMMSCH_VFID_FIFO_HEAD_2 0x0060 +#define mmMMSCH_VFID_FIFO_HEAD_2_BASE_IDX 0 +#define mmMMSCH_VFID_FIFO_TAIL_2 0x0061 +#define mmMMSCH_VFID_FIFO_TAIL_2_BASE_IDX 0 +#define mmMMSCH_VM_BUSY_STATUS_0 0x0062 +#define mmMMSCH_VM_BUSY_STATUS_0_BASE_IDX 0 +#define mmMMSCH_VM_BUSY_STATUS_1 0x0063 +#define mmMMSCH_VM_BUSY_STATUS_1_BASE_IDX 0 +#define mmMMSCH_VM_BUSY_STATUS_2 0x0064 +#define mmMMSCH_VM_BUSY_STATUS_2_BASE_IDX 0 + +#define MMSCH_VERSION_MAJOR 2 +#define MMSCH_VERSION_MINOR 0 +#define MMSCH_VERSION (MMSCH_VERSION_MAJOR << 16 | MMSCH_VERSION_MINOR) + +enum mmsch_v2_0_command_type { + MMSCH_COMMAND__DIRECT_REG_WRITE = 0, + MMSCH_COMMAND__DIRECT_REG_POLLING = 2, + MMSCH_COMMAND__DIRECT_REG_READ_MODIFY_WRITE = 3, + MMSCH_COMMAND__INDIRECT_REG_WRITE = 8, + MMSCH_COMMAND__END = 0xf +}; + +struct mmsch_v2_0_init_header { + uint32_t version; + uint32_t header_size; + uint32_t vcn_init_status; + uint32_t vcn_table_offset; + uint32_t vcn_table_size; +}; + +struct mmsch_v2_0_cmd_direct_reg_header { + uint32_t reg_offset : 28; + uint32_t command_type : 4; +}; + +struct mmsch_v2_0_cmd_indirect_reg_header { + uint32_t reg_offset : 20; + uint32_t reg_idx_space : 8; + uint32_t command_type : 4; +}; + +struct mmsch_v2_0_cmd_direct_write { + struct mmsch_v2_0_cmd_direct_reg_header cmd_header; + uint32_t reg_value; +}; + +struct mmsch_v2_0_cmd_direct_read_modify_write { + struct mmsch_v2_0_cmd_direct_reg_header cmd_header; + uint32_t write_data; + uint32_t mask_value; +}; + +struct mmsch_v2_0_cmd_direct_polling { + struct mmsch_v2_0_cmd_direct_reg_header cmd_header; + uint32_t mask_value; + uint32_t wait_value; +}; + +struct mmsch_v2_0_cmd_end { + struct mmsch_v2_0_cmd_direct_reg_header cmd_header; +}; + +struct mmsch_v2_0_cmd_indirect_write { + struct mmsch_v2_0_cmd_indirect_reg_header cmd_header; + uint32_t reg_value; +}; + +static inline void mmsch_v2_0_insert_direct_wt(struct mmsch_v2_0_cmd_direct_write *direct_wt, + uint32_t *init_table, + uint32_t reg_offset, + uint32_t value) +{ + direct_wt->cmd_header.reg_offset = reg_offset; + direct_wt->reg_value = value; + memcpy((void *)init_table, direct_wt, sizeof(struct mmsch_v2_0_cmd_direct_write)); +} + +static inline void mmsch_v2_0_insert_direct_rd_mod_wt(struct mmsch_v2_0_cmd_direct_read_modify_write *direct_rd_mod_wt, + uint32_t *init_table, + uint32_t reg_offset, + uint32_t mask, uint32_t data) +{ + direct_rd_mod_wt->cmd_header.reg_offset = reg_offset; + direct_rd_mod_wt->mask_value = mask; + direct_rd_mod_wt->write_data = data; + memcpy((void *)init_table, direct_rd_mod_wt, + sizeof(struct mmsch_v2_0_cmd_direct_read_modify_write)); +} + +static inline void mmsch_v2_0_insert_direct_poll(struct mmsch_v2_0_cmd_direct_polling *direct_poll, + uint32_t *init_table, + uint32_t reg_offset, + uint32_t mask, uint32_t wait) +{ + direct_poll->cmd_header.reg_offset = reg_offset; + direct_poll->mask_value = mask; + direct_poll->wait_value = wait; + memcpy((void *)init_table, direct_poll, sizeof(struct mmsch_v2_0_cmd_direct_polling)); +} + +#define MMSCH_V2_0_INSERT_DIRECT_RD_MOD_WT(reg, mask, data) { \ + mmsch_v2_0_insert_direct_rd_mod_wt(&direct_rd_mod_wt, \ + init_table, (reg), \ + (mask), (data)); \ + init_table += sizeof(struct mmsch_v2_0_cmd_direct_read_modify_write)/4; \ + table_size += sizeof(struct mmsch_v2_0_cmd_direct_read_modify_write)/4; \ +} + +#define MMSCH_V2_0_INSERT_DIRECT_WT(reg, value) { \ + mmsch_v2_0_insert_direct_wt(&direct_wt, \ + init_table, (reg), \ + (value)); \ + init_table += sizeof(struct mmsch_v2_0_cmd_direct_write)/4; \ + table_size += sizeof(struct mmsch_v2_0_cmd_direct_write)/4; \ +} + +#define MMSCH_V2_0_INSERT_DIRECT_POLL(reg, mask, wait) { \ + mmsch_v2_0_insert_direct_poll(&direct_poll, \ + init_table, (reg), \ + (mask), (wait)); \ + init_table += sizeof(struct mmsch_v2_0_cmd_direct_polling)/4; \ + table_size += sizeof(struct mmsch_v2_0_cmd_direct_polling)/4; \ +} + +#endif diff --git a/drivers/gpu/drm/amd/amdgpu/navi10_ih.c b/drivers/gpu/drm/amd/amdgpu/navi10_ih.c index cf557a428298..e08245a446fc 100644 --- a/drivers/gpu/drm/amd/amdgpu/navi10_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/navi10_ih.c @@ -32,6 +32,7 @@ #include "soc15_common.h" #include "navi10_ih.h" +#define MAX_REARM_RETRY 10 static void navi10_ih_set_interrupt_funcs(struct amdgpu_device *adev); @@ -284,6 +285,38 @@ static void navi10_ih_decode_iv(struct amdgpu_device *adev, } /** + * navi10_ih_irq_rearm - rearm IRQ if lost + * + * @adev: amdgpu_device pointer + * + */ +static void navi10_ih_irq_rearm(struct amdgpu_device *adev, + struct amdgpu_ih_ring *ih) +{ + uint32_t reg_rptr = 0; + uint32_t v = 0; + uint32_t i = 0; + + if (ih == &adev->irq.ih) + reg_rptr = SOC15_REG_OFFSET(OSSSYS, 0, mmIH_RB_RPTR); + else if (ih == &adev->irq.ih1) + reg_rptr = SOC15_REG_OFFSET(OSSSYS, 0, mmIH_RB_RPTR_RING1); + else if (ih == &adev->irq.ih2) + reg_rptr = SOC15_REG_OFFSET(OSSSYS, 0, mmIH_RB_RPTR_RING2); + else + return; + + /* Rearm IRQ / re-write doorbell if doorbell write is lost */ + for (i = 0; i < MAX_REARM_RETRY; i++) { + v = RREG32_NO_KIQ(reg_rptr); + if ((v < ih->ring_size) && (v != ih->rptr)) + WDOORBELL32(ih->doorbell_index, ih->rptr); + else + break; + } +} + +/** * navi10_ih_set_rptr - set the IH ring buffer rptr * * @adev: amdgpu_device pointer @@ -297,6 +330,9 @@ static void navi10_ih_set_rptr(struct amdgpu_device *adev, /* XXX check if swapping is necessary on BE */ *ih->rptr_cpu = ih->rptr; WDOORBELL32(ih->doorbell_index, ih->rptr); + + if (amdgpu_sriov_vf(adev)) + navi10_ih_irq_rearm(adev, ih); } else WREG32_SOC15(OSSSYS, 0, mmIH_RB_RPTR, ih->rptr); } diff --git a/drivers/gpu/drm/amd/amdgpu/nv.c b/drivers/gpu/drm/amd/amdgpu/nv.c index 2d1bebdf1603..033cbbca2072 100644 --- a/drivers/gpu/drm/amd/amdgpu/nv.c +++ b/drivers/gpu/drm/amd/amdgpu/nv.c @@ -516,7 +516,8 @@ int nv_set_ip_blocks(struct amdgpu_device *adev) !amdgpu_sriov_vf(adev)) amdgpu_device_ip_block_add(adev, &smu_v11_0_ip_block); amdgpu_device_ip_block_add(adev, &vcn_v2_0_ip_block); - amdgpu_device_ip_block_add(adev, &jpeg_v2_0_ip_block); + if (!amdgpu_sriov_vf(adev)) + amdgpu_device_ip_block_add(adev, &jpeg_v2_0_ip_block); break; default: return -EINVAL; diff --git a/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h b/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h index 36b65797434e..a44fd6060d5b 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h +++ b/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h @@ -31,6 +31,9 @@ #define GFX_CMD_RESERVED_MASK 0x7FF00000 #define GFX_CMD_RESPONSE_MASK 0x80000000 +/* USBC PD FW version retrieval command */ +#define C2PMSG_CMD_GFX_USB_PD_FW_VER 0x2000000 + /* TEE Gfx Command IDs for the register interface. * Command ID must be between 0x00010000 and 0x000F0000. */ diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c index 8ab3bf3158a9..0afd610a1263 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c @@ -26,6 +26,7 @@ #include "amdgpu.h" #include "amdgpu_psp.h" +#include "amdgpu_ras.h" #include "amdgpu_ucode.h" #include "soc15_common.h" #include "psp_v11_0.h" @@ -65,6 +66,9 @@ MODULE_FIRMWARE("amdgpu/arcturus_ta.bin"); /* memory training timeout define */ #define MEM_TRAIN_SEND_MSG_TIMEOUT_US 3000000 +/* For large FW files the time to complete can be very long */ +#define USBC_PD_POLLING_LIMIT_S 240 + static int psp_v11_0_init_microcode(struct psp_context *psp) { struct amdgpu_device *adev = psp->adev; @@ -865,6 +869,11 @@ static int psp_v11_0_ras_trigger_error(struct psp_context *psp, if (ret) return -EINVAL; + /* If err_event_athub occurs error inject was successful, however + return status from TA is no long reliable */ + if (amdgpu_ras_intr_triggered()) + return 0; + return ras_cmd->ras_status; } @@ -1109,6 +1118,82 @@ static void psp_v11_0_ring_set_wptr(struct psp_context *psp, uint32_t value) WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_67, value); } +static int psp_v11_0_load_usbc_pd_fw(struct psp_context *psp, dma_addr_t dma_addr) +{ + struct amdgpu_device *adev = psp->adev; + uint32_t reg_status; + int ret, i = 0; + + /* Write lower 32-bit address of the PD Controller FW */ + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_36, lower_32_bits(dma_addr)); + ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_35), + 0x80000000, 0x80000000, false); + if (ret) + return ret; + + /* Fireup interrupt so PSP can pick up the lower address */ + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_35, 0x800000); + ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_35), + 0x80000000, 0x80000000, false); + if (ret) + return ret; + + reg_status = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_35); + + if ((reg_status & 0xFFFF) != 0) { + DRM_ERROR("Lower address load failed - MP0_SMN_C2PMSG_35.Bits [15:0] = %02x...\n", + reg_status & 0xFFFF); + return -EIO; + } + + /* Write upper 32-bit address of the PD Controller FW */ + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_36, upper_32_bits(dma_addr)); + + ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_35), + 0x80000000, 0x80000000, false); + if (ret) + return ret; + + /* Fireup interrupt so PSP can pick up the upper address */ + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_35, 0x4000000); + + /* FW load takes very long time */ + do { + msleep(1000); + reg_status = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_35); + + if (reg_status & 0x80000000) + goto done; + + } while (++i < USBC_PD_POLLING_LIMIT_S); + + return -ETIME; +done: + + if ((reg_status & 0xFFFF) != 0) { + DRM_ERROR("Upper address load failed - MP0_SMN_C2PMSG_35.Bits [15:0] = x%04x\n", + reg_status & 0xFFFF); + return -EIO; + } + + return 0; +} + +static int psp_v11_0_read_usbc_pd_fw(struct psp_context *psp, uint32_t *fw_ver) +{ + struct amdgpu_device *adev = psp->adev; + int ret; + + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_35, C2PMSG_CMD_GFX_USB_PD_FW_VER); + + ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_35), + 0x80000000, 0x80000000, false); + if (!ret) + *fw_ver = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_36); + + return ret; +} + static const struct psp_funcs psp_v11_0_funcs = { .init_microcode = psp_v11_0_init_microcode, .bootloader_load_kdb = psp_v11_0_bootloader_load_kdb, @@ -1133,6 +1218,8 @@ static const struct psp_funcs psp_v11_0_funcs = { .mem_training = psp_v11_0_memory_training, .ring_get_wptr = psp_v11_0_ring_get_wptr, .ring_set_wptr = psp_v11_0_ring_set_wptr, + .load_usbc_pd_fw = psp_v11_0_load_usbc_pd_fw, + .read_usbc_pd_fw = psp_v11_0_read_usbc_pd_fw }; void psp_v11_0_set_psp_funcs(struct psp_context *psp) diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c index e55884d204bd..9159bd46482b 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c @@ -1801,13 +1801,9 @@ static int sdma_v4_0_late_init(void *handle) struct ras_ih_if ih_info = { .cb = sdma_v4_0_process_ras_data_cb, }; - int i; - /* read back edc counter registers to clear the counters */ - if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__SDMA)) { - for (i = 0; i < adev->sdma.num_instances; i++) - RREG32_SDMA(i, mmSDMA0_EDC_COUNTER); - } + if (adev->sdma.funcs && adev->sdma.funcs->reset_ras_error_count) + adev->sdma.funcs->reset_ras_error_count(adev); if (adev->sdma.funcs && adev->sdma.funcs->ras_late_init) return adev->sdma.funcs->ras_late_init(adev, &ih_info); @@ -2572,10 +2568,22 @@ static int sdma_v4_0_query_ras_error_count(struct amdgpu_device *adev, return 0; }; +static void sdma_v4_0_reset_ras_error_count(struct amdgpu_device *adev) +{ + int i; + + /* read back edc counter registers to clear the counters */ + if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__SDMA)) { + for (i = 0; i < adev->sdma.num_instances; i++) + RREG32_SDMA(i, mmSDMA0_EDC_COUNTER); + } +} + static const struct amdgpu_sdma_ras_funcs sdma_v4_0_ras_funcs = { .ras_late_init = amdgpu_sdma_ras_late_init, .ras_fini = amdgpu_sdma_ras_fini, .query_ras_error_count = sdma_v4_0_query_ras_error_count, + .reset_ras_error_count = sdma_v4_0_reset_ras_error_count, }; static void sdma_v4_0_set_ras_funcs(struct amdgpu_device *adev) diff --git a/drivers/gpu/drm/amd/amdgpu/si_dpm.c b/drivers/gpu/drm/amd/amdgpu/si_dpm.c index 4cb4c891120b..0860e85a2d35 100644 --- a/drivers/gpu/drm/amd/amdgpu/si_dpm.c +++ b/drivers/gpu/drm/amd/amdgpu/si_dpm.c @@ -3439,7 +3439,6 @@ static void si_apply_state_adjust_rules(struct amdgpu_device *adev, if (adev->asic_type == CHIP_HAINAN) { if ((adev->pdev->revision == 0x81) || - (adev->pdev->revision == 0x83) || (adev->pdev->revision == 0xC3) || (adev->pdev->device == 0x6664) || (adev->pdev->device == 0x6665) || diff --git a/drivers/gpu/drm/amd/amdgpu/smu_v11_0_i2c.c b/drivers/gpu/drm/amd/amdgpu/smu_v11_0_i2c.c index c902f26cf50d..9bffbab35041 100644 --- a/drivers/gpu/drm/amd/amdgpu/smu_v11_0_i2c.c +++ b/drivers/gpu/drm/amd/amdgpu/smu_v11_0_i2c.c @@ -46,8 +46,7 @@ #define I2C_NO_STOP 1 #define I2C_RESTART 2 -#define to_amdgpu_device(x) (container_of(x, struct amdgpu_ras, eeprom_control.eeprom_accessor))->adev -#define to_eeprom_control(x) container_of(x, struct amdgpu_ras_eeprom_control, eeprom_accessor) +#define to_amdgpu_device(x) (container_of(x, struct amdgpu_device, pm.smu_i2c)) static void smu_v11_0_i2c_set_clock_gating(struct i2c_adapter *control, bool en) { @@ -592,7 +591,8 @@ static uint32_t smu_v11_0_i2c_eeprom_write_data(struct i2c_adapter *control, static void lock_bus(struct i2c_adapter *i2c, unsigned int flags) { - struct amdgpu_ras_eeprom_control *control = to_eeprom_control(i2c); + struct amdgpu_device *adev = to_amdgpu_device(i2c); + struct amdgpu_ras_eeprom_control *control = &adev->psp.ras.ras->eeprom_control; if (!smu_v11_0_i2c_bus_lock(i2c)) { DRM_ERROR("Failed to lock the bus from SMU"); @@ -610,7 +610,8 @@ static int trylock_bus(struct i2c_adapter *i2c, unsigned int flags) static void unlock_bus(struct i2c_adapter *i2c, unsigned int flags) { - struct amdgpu_ras_eeprom_control *control = to_eeprom_control(i2c); + struct amdgpu_device *adev = to_amdgpu_device(i2c); + struct amdgpu_ras_eeprom_control *control = &adev->psp.ras.ras->eeprom_control; if (!smu_v11_0_i2c_bus_unlock(i2c)) { DRM_ERROR("Failed to unlock the bus from SMU"); @@ -630,7 +631,8 @@ static int smu_v11_0_i2c_eeprom_i2c_xfer(struct i2c_adapter *i2c_adap, struct i2c_msg *msgs, int num) { int i, ret; - struct amdgpu_ras_eeprom_control *control = to_eeprom_control(i2c_adap); + struct amdgpu_device *adev = to_amdgpu_device(i2c_adap); + struct amdgpu_ras_eeprom_control *control = &adev->psp.ras.ras->eeprom_control; if (!control->bus_locked) { DRM_ERROR("I2C bus unlocked, stopping transaction!"); @@ -679,7 +681,7 @@ int smu_v11_0_i2c_eeprom_control_init(struct i2c_adapter *control) control->class = I2C_CLASS_SPD; control->dev.parent = &adev->pdev->dev; control->algo = &smu_v11_0_i2c_eeprom_i2c_algo; - snprintf(control->name, sizeof(control->name), "RAS EEPROM"); + snprintf(control->name, sizeof(control->name), "AMDGPU EEPROM"); control->lock_ops = &smu_v11_0_i2c_i2c_lock_ops; res = i2c_add_adapter(control); diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c index 2b488dfb2f21..a40499d51c93 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc15.c +++ b/drivers/gpu/drm/amd/amdgpu/soc15.c @@ -89,6 +89,13 @@ #define HDP_MEM_POWER_CTRL__RC_MEM_POWER_CTRL_EN_MASK 0x00010000L #define HDP_MEM_POWER_CTRL__RC_MEM_POWER_LS_EN_MASK 0x00020000L #define mmHDP_MEM_POWER_CTRL_BASE_IDX 0 + +/* for Vega20/arcturus regiter offset change */ +#define mmROM_INDEX_VG20 0x00e4 +#define mmROM_INDEX_VG20_BASE_IDX 0 +#define mmROM_DATA_VG20 0x00e5 +#define mmROM_DATA_VG20_BASE_IDX 0 + /* * Indirect registers accessor */ @@ -309,6 +316,8 @@ static bool soc15_read_bios_from_rom(struct amdgpu_device *adev, { u32 *dw_ptr; u32 i, length_dw; + uint32_t rom_index_offset; + uint32_t rom_data_offset; if (bios == NULL) return false; @@ -321,11 +330,23 @@ static bool soc15_read_bios_from_rom(struct amdgpu_device *adev, dw_ptr = (u32 *)bios; length_dw = ALIGN(length_bytes, 4) / 4; + switch (adev->asic_type) { + case CHIP_VEGA20: + case CHIP_ARCTURUS: + rom_index_offset = SOC15_REG_OFFSET(SMUIO, 0, mmROM_INDEX_VG20); + rom_data_offset = SOC15_REG_OFFSET(SMUIO, 0, mmROM_DATA_VG20); + break; + default: + rom_index_offset = SOC15_REG_OFFSET(SMUIO, 0, mmROM_INDEX); + rom_data_offset = SOC15_REG_OFFSET(SMUIO, 0, mmROM_DATA); + break; + } + /* set rom index to 0 */ - WREG32(SOC15_REG_OFFSET(SMUIO, 0, mmROM_INDEX), 0); + WREG32(rom_index_offset, 0); /* read out the rom data */ for (i = 0; i < length_dw; i++) - dw_ptr[i] = RREG32(SOC15_REG_OFFSET(SMUIO, 0, mmROM_DATA)); + dw_ptr[i] = RREG32(rom_data_offset); return true; } @@ -831,6 +852,15 @@ static bool soc15_need_full_reset(struct amdgpu_device *adev) /* change this when we implement soft reset */ return true; } + +static void vega20_reset_hdp_ras_error_count(struct amdgpu_device *adev) +{ + if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__HDP)) + return; + /*read back hdp ras counter to reset it to 0 */ + RREG32_SOC15(HDP, 0, mmHDP_EDC_CNT); +} + static void soc15_get_pcie_usage(struct amdgpu_device *adev, uint64_t *count0, uint64_t *count1) { @@ -998,6 +1028,7 @@ static const struct amdgpu_asic_funcs vega20_asic_funcs = .get_config_memsize = &soc15_get_config_memsize, .flush_hdp = &soc15_flush_hdp, .invalidate_hdp = &soc15_invalidate_hdp, + .reset_hdp_ras_error_count = &vega20_reset_hdp_ras_error_count, .need_full_reset = &soc15_need_full_reset, .init_doorbell_index = &vega20_doorbell_index_init, .get_pcie_usage = &vega20_get_pcie_usage, @@ -1243,6 +1274,10 @@ static int soc15_common_late_init(void *handle) if (amdgpu_sriov_vf(adev)) xgpu_ai_mailbox_get_irq(adev); + if (adev->asic_funcs && + adev->asic_funcs->reset_hdp_ras_error_count) + adev->asic_funcs->reset_hdp_ras_error_count(adev); + if (adev->nbio.funcs->ras_late_init) r = adev->nbio.funcs->ras_late_init(adev); diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.h b/drivers/gpu/drm/amd/amdgpu/soc15.h index d0fb7a67c1a3..b03f950c486c 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc15.h +++ b/drivers/gpu/drm/amd/amdgpu/soc15.h @@ -42,6 +42,13 @@ struct soc15_reg_golden { u32 or_mask; }; +struct soc15_reg_rlcg { + u32 hwip; + u32 instance; + u32 segment; + u32 reg; +}; + struct soc15_reg_entry { uint32_t hwip; uint32_t inst; diff --git a/drivers/gpu/drm/amd/amdgpu/soc15_common.h b/drivers/gpu/drm/amd/amdgpu/soc15_common.h index 19e870c79896..c893c645a4b2 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc15_common.h +++ b/drivers/gpu/drm/amd/amdgpu/soc15_common.h @@ -70,10 +70,9 @@ } \ } while (0) -#define AMDGPU_VIRT_SUPPORT_RLC_PRG_REG(a) (amdgpu_sriov_vf((a)) && !amdgpu_sriov_runtime((a))) #define WREG32_RLC(reg, value) \ do { \ - if (AMDGPU_VIRT_SUPPORT_RLC_PRG_REG(adev)) { \ + if (amdgpu_sriov_fullaccess(adev)) { \ uint32_t i = 0; \ uint32_t retries = 50000; \ uint32_t r0 = adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG0_BASE_IDX] + mmSCRATCH_REG0; \ @@ -98,7 +97,7 @@ #define WREG32_SOC15_RLC_SHADOW(ip, inst, reg, value) \ do { \ uint32_t target_reg = adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg;\ - if (AMDGPU_VIRT_SUPPORT_RLC_PRG_REG(adev)) { \ + if (amdgpu_sriov_fullaccess(adev)) { \ uint32_t r2 = adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG1_BASE_IDX] + mmSCRATCH_REG2; \ uint32_t r3 = adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG1_BASE_IDX] + mmSCRATCH_REG3; \ uint32_t grbm_cntl = adev->reg_offset[GC_HWIP][0][mmGRBM_GFX_CNTL_BASE_IDX] + mmGRBM_GFX_CNTL; \ diff --git a/drivers/gpu/drm/amd/amdgpu/umc_v6_1.c b/drivers/gpu/drm/amd/amdgpu/umc_v6_1.c index 793bf70e64b1..14d346321a5f 100644 --- a/drivers/gpu/drm/amd/amdgpu/umc_v6_1.c +++ b/drivers/gpu/drm/amd/amdgpu/umc_v6_1.c @@ -186,6 +186,10 @@ static void umc_v6_1_query_ras_error_count(struct amdgpu_device *adev, if (rsmu_umc_index_state) umc_v6_1_disable_umc_index_mode(adev); + if ((adev->asic_type == CHIP_ARCTURUS) && + amdgpu_dpm_set_df_cstate(adev, DF_CSTATE_DISALLOW)) + DRM_WARN("Fail to disable DF-Cstate.\n"); + LOOP_UMC_INST_AND_CH(umc_inst, ch_inst) { umc_reg_offset = get_umc_6_reg_offset(adev, umc_inst, @@ -199,6 +203,10 @@ static void umc_v6_1_query_ras_error_count(struct amdgpu_device *adev, &(err_data->ue_count)); } + if ((adev->asic_type == CHIP_ARCTURUS) && + amdgpu_dpm_set_df_cstate(adev, DF_CSTATE_ALLOW)) + DRM_WARN("Fail to enable DF-Cstate\n"); + if (rsmu_umc_index_state) umc_v6_1_enable_umc_index_mode(adev); } @@ -228,7 +236,11 @@ static void umc_v6_1_query_error_address(struct amdgpu_device *adev, SOC15_REG_OFFSET(UMC, 0, mmMCA_UMC_UMC0_MCUMC_ADDRT0); } - /* skip error address process if -ENOMEM */ + mc_umc_status = RREG64_PCIE((mc_umc_status_addr + umc_reg_offset) * 4); + + if (mc_umc_status == 0) + return; + if (!err_data->err_addr) { /* clear umc status */ WREG64_PCIE((mc_umc_status_addr + umc_reg_offset) * 4, 0x0ULL); @@ -236,7 +248,6 @@ static void umc_v6_1_query_error_address(struct amdgpu_device *adev, } err_rec = &err_data->err_addr[err_data->err_addr_cnt]; - mc_umc_status = RREG64_PCIE((mc_umc_status_addr + umc_reg_offset) * 4); /* calculate error address if ue/ce error is detected */ if (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Val) == 1 && @@ -288,6 +299,10 @@ static void umc_v6_1_query_ras_error_address(struct amdgpu_device *adev, if (rsmu_umc_index_state) umc_v6_1_disable_umc_index_mode(adev); + if ((adev->asic_type == CHIP_ARCTURUS) && + amdgpu_dpm_set_df_cstate(adev, DF_CSTATE_DISALLOW)) + DRM_WARN("Fail to disable DF-Cstate.\n"); + LOOP_UMC_INST_AND_CH(umc_inst, ch_inst) { umc_reg_offset = get_umc_6_reg_offset(adev, umc_inst, @@ -300,6 +315,10 @@ static void umc_v6_1_query_ras_error_address(struct amdgpu_device *adev, umc_inst); } + if ((adev->asic_type == CHIP_ARCTURUS) && + amdgpu_dpm_set_df_cstate(adev, DF_CSTATE_ALLOW)) + DRM_WARN("Fail to enable DF-Cstate\n"); + if (rsmu_umc_index_state) umc_v6_1_enable_umc_index_mode(adev); } diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c index 71f61afdc655..09b0572b838d 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c @@ -1352,7 +1352,7 @@ static int vcn_v1_0_set_clockgating_state(void *handle, if (enable) { /* wait for STATUS to clear */ - if (vcn_v1_0_is_idle(handle)) + if (!vcn_v1_0_is_idle(handle)) return -EBUSY; vcn_v1_0_enable_clock_gating(adev); } else { diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c index c387c81f8695..ec8091a661df 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c @@ -29,6 +29,7 @@ #include "soc15d.h" #include "amdgpu_pm.h" #include "amdgpu_psp.h" +#include "mmsch_v2_0.h" #include "vcn/vcn_2_0_0_offset.h" #include "vcn/vcn_2_0_0_sh_mask.h" @@ -54,7 +55,7 @@ static int vcn_v2_0_set_powergating_state(void *handle, enum amd_powergating_state state); static int vcn_v2_0_pause_dpg_mode(struct amdgpu_device *adev, int inst_idx, struct dpg_pause_state *new_state); - +static int vcn_v2_0_start_sriov(struct amdgpu_device *adev); /** * vcn_v2_0_early_init - set function pointers * @@ -67,7 +68,10 @@ static int vcn_v2_0_early_init(void *handle) struct amdgpu_device *adev = (struct amdgpu_device *)handle; adev->vcn.num_vcn_inst = 1; - adev->vcn.num_enc_rings = 2; + if (amdgpu_sriov_vf(adev)) + adev->vcn.num_enc_rings = 1; + else + adev->vcn.num_enc_rings = 2; vcn_v2_0_set_dec_ring_funcs(adev); vcn_v2_0_set_enc_ring_funcs(adev); @@ -154,7 +158,10 @@ static int vcn_v2_0_sw_init(void *handle) for (i = 0; i < adev->vcn.num_enc_rings; ++i) { ring = &adev->vcn.inst->ring_enc[i]; ring->use_doorbell = true; - ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 2 + i; + if (!amdgpu_sriov_vf(adev)) + ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 2 + i; + else + ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 1 + i; sprintf(ring->name, "vcn_enc%d", i); r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst->irq, 0); if (r) @@ -163,6 +170,10 @@ static int vcn_v2_0_sw_init(void *handle) adev->vcn.pause_dpg_mode = vcn_v2_0_pause_dpg_mode; + r = amdgpu_virt_alloc_mm_table(adev); + if (r) + return r; + return 0; } @@ -178,6 +189,8 @@ static int vcn_v2_0_sw_fini(void *handle) int r; struct amdgpu_device *adev = (struct amdgpu_device *)handle; + amdgpu_virt_free_mm_table(adev); + r = amdgpu_vcn_suspend(adev); if (r) return r; @@ -203,6 +216,9 @@ static int vcn_v2_0_hw_init(void *handle) adev->nbio.funcs->vcn_doorbell_range(adev, ring->use_doorbell, ring->doorbell_index, 0); + if (amdgpu_sriov_vf(adev)) + vcn_v2_0_start_sriov(adev); + r = amdgpu_ring_test_helper(ring); if (r) goto done; @@ -304,6 +320,9 @@ static void vcn_v2_0_mc_resume(struct amdgpu_device *adev) uint32_t size = AMDGPU_GPU_PAGE_ALIGN(adev->vcn.fw->size + 4); uint32_t offset; + if (amdgpu_sriov_vf(adev)) + return; + /* cache window 0: fw */ if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) { WREG32_SOC15(UVD, 0, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW, @@ -448,6 +467,9 @@ static void vcn_v2_0_disable_clock_gating(struct amdgpu_device *adev) { uint32_t data; + if (amdgpu_sriov_vf(adev)) + return; + /* UVD disable CGC */ data = RREG32_SOC15(VCN, 0, mmUVD_CGC_CTRL); if (adev->cg_flags & AMD_CG_SUPPORT_VCN_MGCG) @@ -606,6 +628,9 @@ static void vcn_v2_0_enable_clock_gating(struct amdgpu_device *adev) { uint32_t data = 0; + if (amdgpu_sriov_vf(adev)) + return; + /* enable UVD CGC */ data = RREG32_SOC15(VCN, 0, mmUVD_CGC_CTRL); if (adev->cg_flags & AMD_CG_SUPPORT_VCN_MGCG) @@ -658,6 +683,9 @@ static void vcn_v2_0_disable_static_power_gating(struct amdgpu_device *adev) uint32_t data = 0; int ret; + if (amdgpu_sriov_vf(adev)) + return; + if (adev->pg_flags & AMD_PG_SUPPORT_VCN) { data = (1 << UVD_PGFSM_CONFIG__UVDM_PWR_CONFIG__SHIFT | 1 << UVD_PGFSM_CONFIG__UVDU_PWR_CONFIG__SHIFT @@ -705,6 +733,9 @@ static void vcn_v2_0_enable_static_power_gating(struct amdgpu_device *adev) uint32_t data = 0; int ret; + if (amdgpu_sriov_vf(adev)) + return; + if (adev->pg_flags & AMD_PG_SUPPORT_VCN) { /* Before power off, this indicator has to be turned on */ data = RREG32_SOC15(VCN, 0, mmUVD_POWER_STATUS); @@ -1215,9 +1246,12 @@ static int vcn_v2_0_set_clockgating_state(void *handle, struct amdgpu_device *adev = (struct amdgpu_device *)handle; bool enable = (state == AMD_CG_STATE_GATE); + if (amdgpu_sriov_vf(adev)) + return 0; + if (enable) { /* wait for STATUS to clear */ - if (vcn_v2_0_is_idle(handle)) + if (!vcn_v2_0_is_idle(handle)) return -EBUSY; vcn_v2_0_enable_clock_gating(adev); } else { @@ -1631,6 +1665,9 @@ int vcn_v2_0_dec_ring_test_ring(struct amdgpu_ring *ring) unsigned i; int r; + if (amdgpu_sriov_vf(adev)) + return 0; + WREG32(adev->vcn.inst[ring->me].external.scratch9, 0xCAFEDEAD); r = amdgpu_ring_alloc(ring, 4); if (r) @@ -1667,6 +1704,11 @@ static int vcn_v2_0_set_powergating_state(void *handle, int ret; struct amdgpu_device *adev = (struct amdgpu_device *)handle; + if (amdgpu_sriov_vf(adev)) { + adev->vcn.cur_state = AMD_PG_STATE_UNGATE; + return 0; + } + if (state == adev->vcn.cur_state) return 0; @@ -1680,6 +1722,215 @@ static int vcn_v2_0_set_powergating_state(void *handle, return ret; } +static int vcn_v2_0_start_mmsch(struct amdgpu_device *adev, + struct amdgpu_mm_table *table) +{ + uint32_t data = 0, loop; + uint64_t addr = table->gpu_addr; + struct mmsch_v2_0_init_header *header; + uint32_t size; + int i; + + header = (struct mmsch_v2_0_init_header *)table->cpu_addr; + size = header->header_size + header->vcn_table_size; + + /* 1, write to vce_mmsch_vf_ctx_addr_lo/hi register with GPU mc addr + * of memory descriptor location + */ + WREG32_SOC15(UVD, 0, mmMMSCH_VF_CTX_ADDR_LO, lower_32_bits(addr)); + WREG32_SOC15(UVD, 0, mmMMSCH_VF_CTX_ADDR_HI, upper_32_bits(addr)); + + /* 2, update vmid of descriptor */ + data = RREG32_SOC15(UVD, 0, mmMMSCH_VF_VMID); + data &= ~MMSCH_VF_VMID__VF_CTX_VMID_MASK; + /* use domain0 for MM scheduler */ + data |= (0 << MMSCH_VF_VMID__VF_CTX_VMID__SHIFT); + WREG32_SOC15(UVD, 0, mmMMSCH_VF_VMID, data); + + /* 3, notify mmsch about the size of this descriptor */ + WREG32_SOC15(UVD, 0, mmMMSCH_VF_CTX_SIZE, size); + + /* 4, set resp to zero */ + WREG32_SOC15(UVD, 0, mmMMSCH_VF_MAILBOX_RESP, 0); + + adev->vcn.inst->ring_dec.wptr = 0; + adev->vcn.inst->ring_dec.wptr_old = 0; + vcn_v2_0_dec_ring_set_wptr(&adev->vcn.inst->ring_dec); + + for (i = 0; i < adev->vcn.num_enc_rings; ++i) { + adev->vcn.inst->ring_enc[i].wptr = 0; + adev->vcn.inst->ring_enc[i].wptr_old = 0; + vcn_v2_0_enc_ring_set_wptr(&adev->vcn.inst->ring_enc[i]); + } + + /* 5, kick off the initialization and wait until + * VCE_MMSCH_VF_MAILBOX_RESP becomes non-zero + */ + WREG32_SOC15(UVD, 0, mmMMSCH_VF_MAILBOX_HOST, 0x10000001); + + data = RREG32_SOC15(UVD, 0, mmMMSCH_VF_MAILBOX_RESP); + loop = 1000; + while ((data & 0x10000002) != 0x10000002) { + udelay(10); + data = RREG32_SOC15(UVD, 0, mmMMSCH_VF_MAILBOX_RESP); + loop--; + if (!loop) + break; + } + + if (!loop) { + DRM_ERROR("failed to init MMSCH, " \ + "mmMMSCH_VF_MAILBOX_RESP = 0x%08x\n", data); + return -EBUSY; + } + + return 0; +} + +static int vcn_v2_0_start_sriov(struct amdgpu_device *adev) +{ + int r; + uint32_t tmp; + struct amdgpu_ring *ring; + uint32_t offset, size; + uint32_t table_size = 0; + struct mmsch_v2_0_cmd_direct_write direct_wt = { {0} }; + struct mmsch_v2_0_cmd_direct_read_modify_write direct_rd_mod_wt = { {0} }; + struct mmsch_v2_0_cmd_direct_polling direct_poll = { {0} }; + struct mmsch_v2_0_cmd_end end = { {0} }; + struct mmsch_v2_0_init_header *header; + uint32_t *init_table = adev->virt.mm_table.cpu_addr; + uint8_t i = 0; + + header = (struct mmsch_v2_0_init_header *)init_table; + direct_wt.cmd_header.command_type = MMSCH_COMMAND__DIRECT_REG_WRITE; + direct_rd_mod_wt.cmd_header.command_type = + MMSCH_COMMAND__DIRECT_REG_READ_MODIFY_WRITE; + direct_poll.cmd_header.command_type = + MMSCH_COMMAND__DIRECT_REG_POLLING; + end.cmd_header.command_type = MMSCH_COMMAND__END; + + if (header->vcn_table_offset == 0 && header->vcn_table_size == 0) { + header->version = MMSCH_VERSION; + header->header_size = sizeof(struct mmsch_v2_0_init_header) >> 2; + + header->vcn_table_offset = header->header_size; + + init_table += header->vcn_table_offset; + + size = AMDGPU_GPU_PAGE_ALIGN(adev->vcn.fw->size + 4); + + MMSCH_V2_0_INSERT_DIRECT_RD_MOD_WT( + SOC15_REG_OFFSET(UVD, i, mmUVD_STATUS), + 0xFFFFFFFF, 0x00000004); + + /* mc resume*/ + if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) { + tmp = AMDGPU_UCODE_ID_VCN; + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, + mmUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW), + adev->firmware.ucode[tmp].tmr_mc_addr_lo); + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, + mmUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH), + adev->firmware.ucode[tmp].tmr_mc_addr_hi); + offset = 0; + } else { + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, + mmUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW), + lower_32_bits(adev->vcn.inst->gpu_addr)); + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, + mmUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH), + upper_32_bits(adev->vcn.inst->gpu_addr)); + offset = size; + } + + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, mmUVD_VCPU_CACHE_OFFSET0), + 0); + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, mmUVD_VCPU_CACHE_SIZE0), + size); + + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, + mmUVD_LMI_VCPU_CACHE1_64BIT_BAR_LOW), + lower_32_bits(adev->vcn.inst->gpu_addr + offset)); + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, + mmUVD_LMI_VCPU_CACHE1_64BIT_BAR_HIGH), + upper_32_bits(adev->vcn.inst->gpu_addr + offset)); + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, mmUVD_VCPU_CACHE_OFFSET1), + 0); + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, mmUVD_VCPU_CACHE_SIZE1), + AMDGPU_VCN_STACK_SIZE); + + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, + mmUVD_LMI_VCPU_CACHE2_64BIT_BAR_LOW), + lower_32_bits(adev->vcn.inst->gpu_addr + offset + + AMDGPU_VCN_STACK_SIZE)); + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, + mmUVD_LMI_VCPU_CACHE2_64BIT_BAR_HIGH), + upper_32_bits(adev->vcn.inst->gpu_addr + offset + + AMDGPU_VCN_STACK_SIZE)); + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, mmUVD_VCPU_CACHE_OFFSET2), + 0); + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, mmUVD_VCPU_CACHE_SIZE2), + AMDGPU_VCN_CONTEXT_SIZE); + + for (r = 0; r < adev->vcn.num_enc_rings; ++r) { + ring = &adev->vcn.inst->ring_enc[r]; + ring->wptr = 0; + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, mmUVD_RB_BASE_LO), + lower_32_bits(ring->gpu_addr)); + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, mmUVD_RB_BASE_HI), + upper_32_bits(ring->gpu_addr)); + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, mmUVD_RB_SIZE), + ring->ring_size / 4); + } + + ring = &adev->vcn.inst->ring_dec; + ring->wptr = 0; + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, + mmUVD_LMI_RBC_RB_64BIT_BAR_LOW), + lower_32_bits(ring->gpu_addr)); + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, + mmUVD_LMI_RBC_RB_64BIT_BAR_HIGH), + upper_32_bits(ring->gpu_addr)); + /* force RBC into idle state */ + tmp = order_base_2(ring->ring_size); + tmp = REG_SET_FIELD(0, UVD_RBC_RB_CNTL, RB_BUFSZ, tmp); + tmp = REG_SET_FIELD(tmp, UVD_RBC_RB_CNTL, RB_BLKSZ, 1); + tmp = REG_SET_FIELD(tmp, UVD_RBC_RB_CNTL, RB_NO_FETCH, 1); + tmp = REG_SET_FIELD(tmp, UVD_RBC_RB_CNTL, RB_NO_UPDATE, 1); + tmp = REG_SET_FIELD(tmp, UVD_RBC_RB_CNTL, RB_RPTR_WR_EN, 1); + MMSCH_V2_0_INSERT_DIRECT_WT( + SOC15_REG_OFFSET(UVD, i, mmUVD_RBC_RB_CNTL), tmp); + + /* add end packet */ + tmp = sizeof(struct mmsch_v2_0_cmd_end); + memcpy((void *)init_table, &end, tmp); + table_size += (tmp / 4); + header->vcn_table_size = table_size; + + } + return vcn_v2_0_start_mmsch(adev, &adev->virt.mm_table); +} + static const struct amd_ip_funcs vcn_v2_0_ip_funcs = { .name = "vcn_v2_0", .early_init = vcn_v2_0_early_init, diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c index 2d64ba1adf99..c6363f5ad564 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c @@ -74,29 +74,30 @@ static int amdgpu_ih_clientid_vcns[] = { static int vcn_v2_5_early_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - if (adev->asic_type == CHIP_ARCTURUS) { - u32 harvest; - int i; - - adev->vcn.num_vcn_inst = VCN25_MAX_HW_INSTANCES_ARCTURUS; - for (i = 0; i < adev->vcn.num_vcn_inst; i++) { - harvest = RREG32_SOC15(UVD, i, mmCC_UVD_HARVESTING); - if (harvest & CC_UVD_HARVESTING__UVD_DISABLE_MASK) - adev->vcn.harvest_config |= 1 << i; - } - - if (adev->vcn.harvest_config == (AMDGPU_VCN_HARVEST_VCN0 | - AMDGPU_VCN_HARVEST_VCN1)) - /* both instances are harvested, disable the block */ - return -ENOENT; - } else - adev->vcn.num_vcn_inst = 1; if (amdgpu_sriov_vf(adev)) { adev->vcn.num_vcn_inst = 2; adev->vcn.harvest_config = 0; adev->vcn.num_enc_rings = 1; } else { + if (adev->asic_type == CHIP_ARCTURUS) { + u32 harvest; + int i; + + adev->vcn.num_vcn_inst = VCN25_MAX_HW_INSTANCES_ARCTURUS; + for (i = 0; i < adev->vcn.num_vcn_inst; i++) { + harvest = RREG32_SOC15(UVD, i, mmCC_UVD_HARVESTING); + if (harvest & CC_UVD_HARVESTING__UVD_DISABLE_MASK) + adev->vcn.harvest_config |= 1 << i; + } + + if (adev->vcn.harvest_config == (AMDGPU_VCN_HARVEST_VCN0 | + AMDGPU_VCN_HARVEST_VCN1)) + /* both instances are harvested, disable the block */ + return -ENOENT; + } else + adev->vcn.num_vcn_inst = 1; + adev->vcn.num_enc_rings = 2; } @@ -1672,7 +1673,7 @@ static int vcn_v2_5_set_clockgating_state(void *handle, return 0; if (enable) { - if (vcn_v2_5_is_idle(handle)) + if (!vcn_v2_5_is_idle(handle)) return -EBUSY; vcn_v2_5_enable_clock_gating(adev); } else { diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c index 8d56afd76eb3..0ec5f25adf56 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c @@ -1169,7 +1169,7 @@ static int kfd_ioctl_get_tile_config(struct file *filep, if (!dev) return -EINVAL; - dev->kfd2kgd->get_tile_config(dev->kgd, &config); + amdgpu_amdkfd_get_tile_config(dev->kgd, &config); args->gb_addr_config = config.gb_addr_config; args->num_banks = config.num_banks; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c index 692abfd2088a..77ea0f0cb163 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c @@ -1734,7 +1734,7 @@ static int allocate_hiq_sdma_mqd(struct device_queue_manager *dqm) retval = amdgpu_amdkfd_alloc_gtt_mem(dev->kgd, size, &(mem_obj->gtt_mem), &(mem_obj->gpu_addr), - (void *)&(mem_obj->cpu_ptr), true); + (void *)&(mem_obj->cpu_ptr), false); return retval; } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_events.c b/drivers/gpu/drm/amd/amdkfd/kfd_events.c index 1f8365575b12..15476fca8fa6 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_events.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_events.c @@ -187,7 +187,7 @@ static int create_signal_event(struct file *devkfd, if (p->signal_mapped_size && p->signal_event_count == p->signal_mapped_size / 8) { if (!p->signal_event_limit_reached) { - pr_warn("Signal event wasn't created because limit was reached\n"); + pr_debug("Signal event wasn't created because limit was reached\n"); p->signal_event_limit_reached = true; } return -ENOSPC; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c b/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c index bb77b8890e77..78714f9a8b11 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c @@ -316,7 +316,7 @@ static void kfd_init_apertures_vi(struct kfd_process_device *pdd, uint8_t id) { /* * node id couldn't be 0 - the three MSB bits of - * aperture shoudn't be 0 + * aperture shouldn't be 0 */ pdd->lds_base = MAKE_LDS_APP_BASE_VI(); pdd->lds_limit = MAKE_LDS_APP_LIMIT(pdd->lds_base); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c index 436b7f518979..48cda3073b70 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c @@ -87,9 +87,21 @@ static struct kfd_mem_obj *allocate_mqd(struct kfd_dev *kfd, int retval; struct kfd_mem_obj *mqd_mem_obj = NULL; - /* From V9, for CWSR, the control stack is located on the next page - * boundary after the mqd, we will use the gtt allocation function - * instead of sub-allocation function. + /* For V9 only, due to a HW bug, the control stack of a user mode + * compute queue needs to be allocated just behind the page boundary + * of its regular MQD buffer. So we allocate an enlarged MQD buffer: + * the first page of the buffer serves as the regular MQD buffer + * purpose and the remaining is for control stack. Although the two + * parts are in the same buffer object, they need different memory + * types: MQD part needs UC (uncached) as usual, while control stack + * needs NC (non coherent), which is different from the UC type which + * is used when control stack is allocated in user space. + * + * Because of all those, we use the gtt allocation function instead + * of sub-allocation function for this enlarged MQD buffer. Moreover, + * in order to achieve two memory types in a single buffer object, we + * pass a special bo flag AMDGPU_GEM_CREATE_CP_MQD_GFX9 to instruct + * amdgpu memory functions to do so. */ if (kfd->cwsr_enabled && (q->type == KFD_QUEUE_TYPE_COMPUTE)) { mqd_mem_obj = kzalloc(sizeof(struct kfd_mem_obj), GFP_KERNEL); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c index 22abdbc6dfd7..fe0cd49d4ea7 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c @@ -327,10 +327,10 @@ err_alloc_mem: static int kfd_process_device_reserve_ib_mem(struct kfd_process_device *pdd) { struct qcm_process_device *qpd = &pdd->qpd; - uint32_t flags = ALLOC_MEM_FLAGS_GTT | - ALLOC_MEM_FLAGS_NO_SUBSTITUTE | - ALLOC_MEM_FLAGS_WRITABLE | - ALLOC_MEM_FLAGS_EXECUTABLE; + uint32_t flags = KFD_IOC_ALLOC_MEM_FLAGS_GTT | + KFD_IOC_ALLOC_MEM_FLAGS_NO_SUBSTITUTE | + KFD_IOC_ALLOC_MEM_FLAGS_WRITABLE | + KFD_IOC_ALLOC_MEM_FLAGS_EXECUTABLE; void *kaddr; int ret; @@ -641,6 +641,11 @@ static void kfd_process_notifier_release(struct mmu_notifier *mn, /* Indicate to other users that MM is no longer valid */ p->mm = NULL; + /* Signal the eviction fence after user mode queues are + * destroyed. This allows any BOs to be freed without + * triggering pointless evictions or waiting for fences. + */ + dma_fence_signal(p->ef); mutex_unlock(&p->mutex); @@ -692,8 +697,9 @@ static int kfd_process_device_init_cwsr_dgpu(struct kfd_process_device *pdd) { struct kfd_dev *dev = pdd->dev; struct qcm_process_device *qpd = &pdd->qpd; - uint32_t flags = ALLOC_MEM_FLAGS_GTT | - ALLOC_MEM_FLAGS_NO_SUBSTITUTE | ALLOC_MEM_FLAGS_EXECUTABLE; + uint32_t flags = KFD_IOC_ALLOC_MEM_FLAGS_GTT + | KFD_IOC_ALLOC_MEM_FLAGS_NO_SUBSTITUTE + | KFD_IOC_ALLOC_MEM_FLAGS_EXECUTABLE; void *kaddr; int ret; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c index 5303877c081a..aa0bfa78a667 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c @@ -490,7 +490,7 @@ static ssize_t node_show(struct kobject *kobj, struct attribute *attr, dev->node_props.num_sdma_queues_per_engine); sysfs_show_32bit_prop(buffer, "num_cp_queues", dev->node_props.num_cp_queues); - sysfs_show_64bit_prop(buffer, "unique_id", + sysfs_show_64bit_prop(buffer, "unique_id", dev->node_props.unique_id); if (dev->gpu) { diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c index 1c0449adcef6..d3674d805a0a 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -132,9 +132,6 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev); /* removes and deallocates the drm structures, created by the above function */ static void amdgpu_dm_destroy_drm_device(struct amdgpu_display_manager *dm); -static void -amdgpu_dm_update_connector_after_detect(struct amdgpu_dm_connector *aconnector); - static int amdgpu_dm_plane_init(struct amdgpu_display_manager *dm, struct drm_plane *plane, unsigned long possible_crtcs, @@ -410,8 +407,9 @@ static void dm_vupdate_high_irq(void *interrupt_params) if (acrtc) { acrtc_state = to_dm_crtc_state(acrtc->base.state); - DRM_DEBUG_DRIVER("crtc:%d, vupdate-vrr:%d\n", acrtc->crtc_id, - amdgpu_dm_vrr_active(acrtc_state)); + DRM_DEBUG_VBL("crtc:%d, vupdate-vrr:%d\n", + acrtc->crtc_id, + amdgpu_dm_vrr_active(acrtc_state)); /* Core vblank handling is done here after end of front-porch in * vrr mode, as vblank timestamping will give valid results @@ -461,8 +459,9 @@ static void dm_crtc_high_irq(void *interrupt_params) if (acrtc) { acrtc_state = to_dm_crtc_state(acrtc->base.state); - DRM_DEBUG_DRIVER("crtc:%d, vupdate-vrr:%d\n", acrtc->crtc_id, - amdgpu_dm_vrr_active(acrtc_state)); + DRM_DEBUG_VBL("crtc:%d, vupdate-vrr:%d\n", + acrtc->crtc_id, + amdgpu_dm_vrr_active(acrtc_state)); /* Core vblank handling at start of front-porch is only possible * in non-vrr mode, as only there vblank timestamping will give @@ -525,8 +524,9 @@ static void dm_dcn_crtc_high_irq(void *interrupt_params) acrtc_state = to_dm_crtc_state(acrtc->base.state); - DRM_DEBUG_DRIVER("crtc:%d, vupdate-vrr:%d\n", acrtc->crtc_id, - amdgpu_dm_vrr_active(acrtc_state)); + DRM_DEBUG_VBL("crtc:%d, vupdate-vrr:%d, planes:%d\n", acrtc->crtc_id, + amdgpu_dm_vrr_active(acrtc_state), + acrtc_state->active_planes); amdgpu_dm_crtc_handle_crc_irq(&acrtc->base); drm_crtc_handle_vblank(&acrtc->base); @@ -546,7 +546,18 @@ static void dm_dcn_crtc_high_irq(void *interrupt_params) &acrtc_state->vrr_params.adjust); } - if (acrtc->pflip_status == AMDGPU_FLIP_SUBMITTED) { + /* + * If there aren't any active_planes then DCH HUBP may be clock-gated. + * In that case, pageflip completion interrupts won't fire and pageflip + * completion events won't get delivered. Prevent this by sending + * pending pageflip events from here if a flip is still pending. + * + * If any planes are enabled, use dm_pflip_high_irq() instead, to + * avoid race conditions between flip programming and completion, + * which could cause too early flip completion events. + */ + if (acrtc->pflip_status == AMDGPU_FLIP_SUBMITTED && + acrtc_state->active_planes == 0) { if (acrtc->event) { drm_crtc_send_vblank_event(&acrtc->base, acrtc->event); acrtc->event = NULL; @@ -902,7 +913,7 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) init_data.asic_id.chip_family = adev->family; - init_data.asic_id.pci_revision_id = adev->rev_id; + init_data.asic_id.pci_revision_id = adev->pdev->revision; init_data.asic_id.hw_internal_rev = adev->external_rev_id; init_data.asic_id.vram_width = adev->gmc.vram_width; @@ -1435,6 +1446,73 @@ static void s3_handle_mst(struct drm_device *dev, bool suspend) drm_kms_helper_hotplug_event(dev); } +static int amdgpu_dm_smu_write_watermarks_table(struct amdgpu_device *adev) +{ + struct smu_context *smu = &adev->smu; + int ret = 0; + + if (!is_support_sw_smu(adev)) + return 0; + + /* This interface is for dGPU Navi1x.Linux dc-pplib interface depends + * on window driver dc implementation. + * For Navi1x, clock settings of dcn watermarks are fixed. the settings + * should be passed to smu during boot up and resume from s3. + * boot up: dc calculate dcn watermark clock settings within dc_create, + * dcn20_resource_construct + * then call pplib functions below to pass the settings to smu: + * smu_set_watermarks_for_clock_ranges + * smu_set_watermarks_table + * navi10_set_watermarks_table + * smu_write_watermarks_table + * + * For Renoir, clock settings of dcn watermark are also fixed values. + * dc has implemented different flow for window driver: + * dc_hardware_init / dc_set_power_state + * dcn10_init_hw + * notify_wm_ranges + * set_wm_ranges + * -- Linux + * smu_set_watermarks_for_clock_ranges + * renoir_set_watermarks_table + * smu_write_watermarks_table + * + * For Linux, + * dc_hardware_init -> amdgpu_dm_init + * dc_set_power_state --> dm_resume + * + * therefore, this function apply to navi10/12/14 but not Renoir + * * + */ + switch(adev->asic_type) { + case CHIP_NAVI10: + case CHIP_NAVI14: + case CHIP_NAVI12: + break; + default: + return 0; + } + + mutex_lock(&smu->mutex); + + /* pass data to smu controller */ + if ((smu->watermarks_bitmap & WATERMARKS_EXIST) && + !(smu->watermarks_bitmap & WATERMARKS_LOADED)) { + ret = smu_write_watermarks_table(smu); + + if (ret) { + mutex_unlock(&smu->mutex); + DRM_ERROR("Failed to update WMTABLE!\n"); + return ret; + } + smu->watermarks_bitmap |= WATERMARKS_LOADED; + } + + mutex_unlock(&smu->mutex); + + return 0; +} + /** * dm_hw_init() - Initialize DC device * @handle: The base driver device containing the amdgpu_dm device. @@ -1713,6 +1791,8 @@ static int dm_resume(void *handle) amdgpu_dm_irq_resume_late(adev); + amdgpu_dm_smu_write_watermarks_table(adev); + return 0; } @@ -1826,8 +1906,8 @@ static void update_connector_ext_caps(struct amdgpu_dm_connector *aconnector) caps->aux_min_input_signal = min; } -static void -amdgpu_dm_update_connector_after_detect(struct amdgpu_dm_connector *aconnector) +void amdgpu_dm_update_connector_after_detect( + struct amdgpu_dm_connector *aconnector) { struct drm_connector *connector = &aconnector->base; struct drm_device *dev = connector->dev; @@ -2156,10 +2236,10 @@ static void handle_hpd_rx_irq(void *param) } } #ifdef CONFIG_DRM_AMD_DC_HDCP - if (hpd_irq_data.bytes.device_service_irq.bits.CP_IRQ) { - if (adev->dm.hdcp_workqueue) - hdcp_handle_cpirq(adev->dm.hdcp_workqueue, aconnector->base.index); - } + if (hpd_irq_data.bytes.device_service_irq.bits.CP_IRQ) { + if (adev->dm.hdcp_workqueue) + hdcp_handle_cpirq(adev->dm.hdcp_workqueue, aconnector->base.index); + } #endif if ((dc_link->cur_link_settings.lane_count != LANE_COUNT_UNKNOWN) || (dc_link->type == dc_connection_mst_branch)) @@ -2954,6 +3034,9 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev) if (adev->asic_type != CHIP_CARRIZO && adev->asic_type != CHIP_STONEY) dm->dc->debug.disable_stutter = amdgpu_pp_feature_mask & PP_STUTTER_MODE ? false : true; + /* No userspace support. */ + dm->dc->debug.disable_tri_buf = true; + return 0; fail: kfree(aencoder); @@ -4245,9 +4328,22 @@ create_stream_for_sink(struct amdgpu_dm_connector *aconnector, struct dmcu *dmcu = core_dc->res_pool->dmcu; stream->psr_version = dmcu->dmcu_version.psr_version; - mod_build_vsc_infopacket(stream, - &stream->vsc_infopacket, - &stream->use_vsc_sdp_for_colorimetry); + + // + // should decide stream support vsc sdp colorimetry capability + // before building vsc info packet + // + stream->use_vsc_sdp_for_colorimetry = false; + if (aconnector->dc_sink->sink_signal == SIGNAL_TYPE_DISPLAY_PORT_MST) { + stream->use_vsc_sdp_for_colorimetry = + aconnector->dc_sink->is_vsc_sdp_colorimetry_supported; + } else { + if (stream->link->dpcd_caps.dpcd_rev.raw >= 0x14 && + stream->link->dpcd_caps.dprx_feature.bits.VSC_SDP_COLORIMETRY_SUPPORTED) { + stream->use_vsc_sdp_for_colorimetry = true; + } + } + mod_build_vsc_infopacket(stream, &stream->vsc_infopacket); } } finish: @@ -6469,7 +6565,6 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, uint32_t target_vblank, last_flip_vblank; bool vrr_active = amdgpu_dm_vrr_active(acrtc_state); bool pflip_present = false; - bool swizzle = true; struct { struct dc_surface_update surface_updates[MAX_SURFACES]; struct dc_plane_info plane_infos[MAX_SURFACES]; @@ -6515,9 +6610,6 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, dc_plane = dm_new_plane_state->dc_state; - if (dc_plane && !dc_plane->tiling_info.gfx9.swizzle) - swizzle = false; - bundle->surface_updates[planes_count].surface = dc_plane; if (new_pcrtc_state->color_mgmt_changed) { bundle->surface_updates[planes_count].gamma = dc_plane->gamma_correction; @@ -6726,8 +6818,7 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, amdgpu_dm_link_setup_psr(acrtc_state->stream); else if ((acrtc_state->update_type == UPDATE_TYPE_FAST) && acrtc_state->stream->link->psr_feature_enabled && - !acrtc_state->stream->link->psr_allow_active && - swizzle) { + !acrtc_state->stream->link->psr_allow_active) { amdgpu_dm_psr_enable(acrtc_state->stream); } diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h index 2bec5e6a3054..5cab3e65d992 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h @@ -483,6 +483,9 @@ int amdgpu_dm_update_crtc_color_mgmt(struct dm_crtc_state *crtc); int amdgpu_dm_update_plane_color_mgmt(struct dm_crtc_state *crtc, struct dc_plane_state *dc_plane_state); +void amdgpu_dm_update_connector_after_detect( + struct amdgpu_dm_connector *aconnector); + extern const struct drm_encoder_helper_funcs amdgpu_dm_encoder_helper_funcs; #endif /* __AMDGPU_DM_H__ */ diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c index c4fd148bf6e0..5b70ed3cdb88 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c @@ -412,6 +412,7 @@ static void update_config(void *handle, struct cp_psp_stream_config *config) link->dig_be = config->link_enc_inst; link->ddc_line = aconnector->dc_link->ddc_hw_inst + 1; link->dp.rev = aconnector->dc_link->dpcd_caps.dpcd_rev.raw; + link->dp.mst_supported = config->mst_supported; display->adjust.disable = 1; link->adjust.auth_delay = 2; diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c index 318b474ff20e..c20fb08c450b 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c @@ -400,8 +400,8 @@ bool dm_helpers_dp_mst_start_top_mgr( struct amdgpu_dm_connector *aconnector = link->priv; if (!aconnector) { - DRM_ERROR("Failed to found connector for link!"); - return false; + DRM_ERROR("Failed to find connector for link!"); + return false; } if (boot) { @@ -423,8 +423,8 @@ void dm_helpers_dp_mst_stop_top_mgr( struct amdgpu_dm_connector *aconnector = link->priv; if (!aconnector) { - DRM_ERROR("Failed to found connector for link!"); - return; + DRM_ERROR("Failed to find connector for link!"); + return; } DRM_INFO("DM_MST: stopping TM on aconnector: %p [id: %d]\n", @@ -445,7 +445,7 @@ bool dm_helpers_dp_read_dpcd( struct amdgpu_dm_connector *aconnector = link->priv; if (!aconnector) { - DRM_ERROR("Failed to found connector for link!"); + DRM_ERROR("Failed to find connector for link!"); return false; } @@ -463,7 +463,7 @@ bool dm_helpers_dp_write_dpcd( struct amdgpu_dm_connector *aconnector = link->priv; if (!aconnector) { - DRM_ERROR("Failed to found connector for link!"); + DRM_ERROR("Failed to find connector for link!"); return false; } @@ -483,7 +483,7 @@ bool dm_helpers_submit_i2c( bool result; if (!aconnector) { - DRM_ERROR("Failed to found connector for link!"); + DRM_ERROR("Failed to find connector for link!"); return false; } @@ -538,7 +538,7 @@ bool dm_helpers_is_dp_sink_present(struct dc_link *link) struct amdgpu_dm_connector *aconnector = link->priv; if (!aconnector) { - BUG_ON("Failed to found connector for link!"); + BUG_ON("Failed to find connector for link!"); return true; } @@ -580,6 +580,20 @@ enum dc_edid_status dm_helpers_read_local_edid( /* We don't need the original edid anymore */ kfree(edid); + /* connector->display_info will be parsed from EDID and saved + * into drm_connector->display_info from edid by call stack + * below: + * drm_parse_ycbcr420_deep_color_info + * drm_parse_hdmi_forum_vsdb + * drm_parse_cea_ext + * drm_add_display_info + * drm_connector_update_edid_property + * + * drm_connector->display_info will be used by amdgpu_dm funcs, + * like fill_stream_properties_from_drm_display_mode + */ + amdgpu_dm_update_connector_after_detect(aconnector); + edid_status = dm_helpers_parse_edid_caps( ctx, &sink->dc_edid, diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c index 4c5755633bfd..e8208df420d9 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c @@ -207,7 +207,7 @@ static bool validate_dsc_caps_on_connector(struct amdgpu_dm_connector *aconnecto if (!dc_dsc_parse_dsc_dpcd(aconnector->dc_link->ctx->dc, dsc_caps, NULL, - &dc_sink->sink_dsc_caps.dsc_dec_caps)) + &dc_sink->dsc_caps.dsc_dec_caps)) return false; return true; @@ -262,8 +262,8 @@ static int dm_dp_mst_get_modes(struct drm_connector *connector) #if defined(CONFIG_DRM_AMD_DC_DCN) if (!validate_dsc_caps_on_connector(aconnector)) - memset(&aconnector->dc_sink->sink_dsc_caps, - 0, sizeof(aconnector->dc_sink->sink_dsc_caps)); + memset(&aconnector->dc_sink->dsc_caps, + 0, sizeof(aconnector->dc_sink->dsc_caps)); #endif } } @@ -451,6 +451,7 @@ static void dm_dp_destroy_mst_connector(struct drm_dp_mst_topology_mgr *mgr, aconnector->dc_sink); dc_sink_release(aconnector->dc_sink); aconnector->dc_sink = NULL; + aconnector->dc_link->cur_link_settings.lane_count = 0; } drm_connector_unregister(connector); @@ -530,7 +531,7 @@ static void set_dsc_configs_from_fairness_vars(struct dsc_mst_fairness_params *p memset(¶ms[i].timing->dsc_cfg, 0, sizeof(params[i].timing->dsc_cfg)); if (vars[i].dsc_enabled && dc_dsc_compute_config( params[i].sink->ctx->dc->res_pool->dscs[0], - ¶ms[i].sink->sink_dsc_caps.dsc_dec_caps, + ¶ms[i].sink->dsc_caps.dsc_dec_caps, params[i].sink->ctx->dc->debug.dsc_min_slice_height_override, 0, params[i].timing, @@ -551,7 +552,7 @@ static int bpp_x16_from_pbn(struct dsc_mst_fairness_params param, int pbn) kbps = div_u64((u64)pbn * 994 * 8 * 54, 64); dc_dsc_compute_config( param.sink->ctx->dc->res_pool->dscs[0], - ¶m.sink->sink_dsc_caps.dsc_dec_caps, + ¶m.sink->dsc_caps.dsc_dec_caps, param.sink->ctx->dc->debug.dsc_min_slice_height_override, (int) kbps, param.timing, &dsc_config); @@ -748,14 +749,14 @@ static bool compute_mst_dsc_configs_for_link(struct drm_atomic_state *state, params[count].sink = stream->sink; aconnector = (struct amdgpu_dm_connector *)stream->dm_stream_context; params[count].port = aconnector->port; - params[count].compression_possible = stream->sink->sink_dsc_caps.dsc_dec_caps.is_dsc_supported; + params[count].compression_possible = stream->sink->dsc_caps.dsc_dec_caps.is_dsc_supported; dc_dsc_get_policy_for_timing(params[count].timing, &dsc_policy); if (!dc_dsc_compute_bandwidth_range( stream->sink->ctx->dc->res_pool->dscs[0], stream->sink->ctx->dc->debug.dsc_min_slice_height_override, dsc_policy.min_target_bpp, dsc_policy.max_target_bpp, - &stream->sink->sink_dsc_caps.dsc_dec_caps, + &stream->sink->dsc_caps.dsc_dec_caps, &stream->timing, ¶ms[count].bw_range)) params[count].bw_range.stream_kbps = dc_bandwidth_in_kbps_from_timing(&stream->timing); @@ -837,7 +838,7 @@ bool compute_mst_dsc_configs_for_state(struct drm_atomic_state *state, if (!aconnector || !aconnector->dc_sink) continue; - if (!aconnector->dc_sink->sink_dsc_caps.dsc_dec_caps.is_dsc_supported) + if (!aconnector->dc_sink->dsc_caps.dsc_dec_caps.is_dsc_supported) continue; if (computed_streams[i]) diff --git a/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c b/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c index 2f1c9584ac32..37fa7b48250e 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c +++ b/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c @@ -267,7 +267,7 @@ static struct atom_display_object_path_v2 *get_bios_object( && id.enum_id == obj_id.enum_id) return &bp->object_info_tbl.v1_4->display_path[i]; } - /* fall through */ + fallthrough; case OBJECT_TYPE_CONNECTOR: case OBJECT_TYPE_GENERIC: /* Both Generic and Connector Object ID @@ -280,7 +280,7 @@ static struct atom_display_object_path_v2 *get_bios_object( && id.enum_id == obj_id.enum_id) return &bp->object_info_tbl.v1_4->display_path[i]; } - /* fall through */ + fallthrough; default: return NULL; } diff --git a/drivers/gpu/drm/amd/display/dc/calcs/dce_calcs.c b/drivers/gpu/drm/amd/display/dc/calcs/dce_calcs.c index 5d081c42e81b..2c6db379afae 100644 --- a/drivers/gpu/drm/amd/display/dc/calcs/dce_calcs.c +++ b/drivers/gpu/drm/amd/display/dc/calcs/dce_calcs.c @@ -3265,33 +3265,33 @@ bool bw_calcs(struct dc_context *ctx, bw_fixed_to_int(bw_mul(data-> stutter_exit_watermark[9], bw_int_to_fixed(1000))); - calcs_output->stutter_entry_wm_ns[0].b_mark = - bw_fixed_to_int(bw_mul(data-> - stutter_entry_watermark[4], bw_int_to_fixed(1000))); - calcs_output->stutter_entry_wm_ns[1].b_mark = - bw_fixed_to_int(bw_mul(data-> - stutter_entry_watermark[5], bw_int_to_fixed(1000))); - calcs_output->stutter_entry_wm_ns[2].b_mark = - bw_fixed_to_int(bw_mul(data-> - stutter_entry_watermark[6], bw_int_to_fixed(1000))); - if (ctx->dc->caps.max_slave_planes) { - calcs_output->stutter_entry_wm_ns[3].b_mark = + calcs_output->stutter_entry_wm_ns[0].b_mark = bw_fixed_to_int(bw_mul(data-> - stutter_entry_watermark[0], bw_int_to_fixed(1000))); - calcs_output->stutter_entry_wm_ns[4].b_mark = + stutter_entry_watermark[4], bw_int_to_fixed(1000))); + calcs_output->stutter_entry_wm_ns[1].b_mark = bw_fixed_to_int(bw_mul(data-> - stutter_entry_watermark[1], bw_int_to_fixed(1000))); - } else { - calcs_output->stutter_entry_wm_ns[3].b_mark = + stutter_entry_watermark[5], bw_int_to_fixed(1000))); + calcs_output->stutter_entry_wm_ns[2].b_mark = bw_fixed_to_int(bw_mul(data-> - stutter_entry_watermark[7], bw_int_to_fixed(1000))); - calcs_output->stutter_entry_wm_ns[4].b_mark = + stutter_entry_watermark[6], bw_int_to_fixed(1000))); + if (ctx->dc->caps.max_slave_planes) { + calcs_output->stutter_entry_wm_ns[3].b_mark = + bw_fixed_to_int(bw_mul(data-> + stutter_entry_watermark[0], bw_int_to_fixed(1000))); + calcs_output->stutter_entry_wm_ns[4].b_mark = + bw_fixed_to_int(bw_mul(data-> + stutter_entry_watermark[1], bw_int_to_fixed(1000))); + } else { + calcs_output->stutter_entry_wm_ns[3].b_mark = + bw_fixed_to_int(bw_mul(data-> + stutter_entry_watermark[7], bw_int_to_fixed(1000))); + calcs_output->stutter_entry_wm_ns[4].b_mark = + bw_fixed_to_int(bw_mul(data-> + stutter_entry_watermark[8], bw_int_to_fixed(1000))); + } + calcs_output->stutter_entry_wm_ns[5].b_mark = bw_fixed_to_int(bw_mul(data-> - stutter_entry_watermark[8], bw_int_to_fixed(1000))); - } - calcs_output->stutter_entry_wm_ns[5].b_mark = - bw_fixed_to_int(bw_mul(data-> - stutter_entry_watermark[9], bw_int_to_fixed(1000))); + stutter_entry_watermark[9], bw_int_to_fixed(1000))); calcs_output->urgent_wm_ns[0].b_mark = bw_fixed_to_int(bw_mul(data-> diff --git a/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c b/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c index f0f07b160152..3960a8db94cb 100644 --- a/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c +++ b/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c @@ -712,6 +712,11 @@ unsigned int get_highest_allowed_voltage_level(uint32_t hw_internal_rev, uint32_ case PRID_DALI_DF: case PRID_DALI_E3: case PRID_DALI_E4: + case PRID_POLLOCK_94: + case PRID_POLLOCK_95: + case PRID_POLLOCK_E9: + case PRID_POLLOCK_EA: + case PRID_POLLOCK_EB: return 0; default: break; diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.c index 68a1120ff674..55d09adbf0d9 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.c @@ -115,12 +115,11 @@ void dcn20_update_clocks_update_dpp_dto(struct clk_mgr_internal *clk_mgr, dpp_inst = i; dppclk_khz = context->res_ctx.pipe_ctx[i].plane_res.bw.dppclk_khz; - prev_dppclk_khz = clk_mgr->base.ctx->dc->current_state->res_ctx.pipe_ctx[i].plane_res.bw.dppclk_khz; + prev_dppclk_khz = clk_mgr->dccg->pipe_dppclk_khz[i]; - if ((prev_dppclk_khz > dppclk_khz && safe_to_lower) || prev_dppclk_khz < dppclk_khz) { + if (safe_to_lower || prev_dppclk_khz < dppclk_khz) clk_mgr->dccg->funcs->update_dpp_dto( clk_mgr->dccg, dpp_inst, dppclk_khz); - } } } diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c index 64cbd5462c79..ab267ddd4abe 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c @@ -46,6 +46,7 @@ /* Constants */ #define LPDDR_MEM_RETRAIN_LATENCY 4.977 /* Number obtained from LPDDR4 Training Counter Requirement doc */ +#define SMU_VER_55_51_0 0x373300 /* SMU Version that is able to set DISPCLK below 100MHz */ /* Macros */ @@ -720,6 +721,13 @@ void rn_clk_mgr_construct( } else { struct clk_log_info log_info = {0}; + clk_mgr->smu_ver = rn_vbios_smu_get_smu_version(clk_mgr); + + /* SMU Version 55.51.0 and up no longer have an issue + * that needs to limit minimum dispclk */ + if (clk_mgr->smu_ver >= SMU_VER_55_51_0) + debug->min_disp_clk_khz = 0; + /* TODO: Check we get what we expect during bringup */ clk_mgr->base.dentist_vco_freq_khz = get_vco_frequency_from_reg(clk_mgr); diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c index 6dece1ee30bf..2ffb22177df9 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc.c @@ -1365,7 +1365,7 @@ bool dc_post_update_surfaces_to_stream(struct dc *dc) int i; struct dc_state *context = dc->current_state; - if ((!dc->clk_optimized_required && !dc->wm_optimized_required) || dc->optimize_seamless_boot_streams > 0) + if ((!dc->optimized_required) || dc->optimize_seamless_boot_streams > 0) return true; post_surface_trace(dc); @@ -1378,6 +1378,10 @@ bool dc_post_update_surfaces_to_stream(struct dc *dc) } dc->hwss.optimize_bandwidth(dc, context); + + dc->optimized_required = false; + dc->wm_optimized_required = false; + return true; } @@ -1824,11 +1828,12 @@ enum surface_update_type dc_check_update_surfaces_for_stream( // If there's an available clock comparator, we use that. if (dc->clk_mgr->funcs->are_clock_states_equal) { if (!dc->clk_mgr->funcs->are_clock_states_equal(&dc->clk_mgr->clks, &dc->current_state->bw_ctx.bw.dcn.clk)) - dc->clk_optimized_required = true; + dc->optimized_required = true; // Else we fallback to mem compare. } else if (memcmp(&dc->current_state->bw_ctx.bw.dcn.clk, &dc->clk_mgr->clks, offsetof(struct dc_clocks, prev_p_state_change_support)) != 0) { - dc->clk_optimized_required = true; - } + dc->optimized_required = true; + } else if (dc->wm_optimized_required) + dc->optimized_required = true; } return type; @@ -1867,6 +1872,8 @@ static void copy_surface_update_to_plane( surface->time.index++; if (surface->time.index >= DC_PLANE_UPDATE_TIMES_MAX) surface->time.index = 0; + + surface->triplebuffer_flips = srf_update->flip_addr->triplebuffer_flips; } if (srf_update->scaling_info) { @@ -2198,7 +2205,7 @@ static void commit_planes_for_stream(struct dc *dc, dc->optimize_seamless_boot_streams--; if (dc->optimize_seamless_boot_streams == 0) - dc->clk_optimized_required = true; + dc->optimized_required = true; } } diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link.c b/drivers/gpu/drm/amd/display/dc/core/dc_link.c index 02e1ad318203..67cfff1586e9 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link.c @@ -585,14 +585,14 @@ static void read_current_link_settings_on_detect(struct dc_link *link) LINK_SPREAD_05_DOWNSPREAD_30KHZ : LINK_SPREAD_DISABLED; } -static bool detect_dp( - struct dc_link *link, - struct display_sink_capability *sink_caps, - bool *converter_disable_audio, - struct audio_support *audio_support, - enum dc_detect_reason reason) +static bool detect_dp(struct dc_link *link, + struct display_sink_capability *sink_caps, + bool *converter_disable_audio, + struct audio_support *audio_support, + enum dc_detect_reason reason) { bool boot = false; + sink_caps->signal = link_detect_sink(link, reason); sink_caps->transaction_type = get_ddc_transaction_type(sink_caps->signal); @@ -609,9 +609,8 @@ static bool detect_dp( sink_caps->signal = SIGNAL_TYPE_DISPLAY_PORT_MST; link->type = dc_connection_mst_branch; - dal_ddc_service_set_transaction_type( - link->ddc, - sink_caps->transaction_type); + dal_ddc_service_set_transaction_type(link->ddc, + sink_caps->transaction_type); /* * This call will initiate MST topology discovery. Which @@ -640,13 +639,10 @@ static bool detect_dp( if (reason == DETECT_REASON_BOOT) boot = true; - dm_helpers_dp_update_branch_info( - link->ctx, - link); + dm_helpers_dp_update_branch_info(link->ctx, link); - if (!dm_helpers_dp_mst_start_top_mgr( - link->ctx, - link, boot)) { + if (!dm_helpers_dp_mst_start_top_mgr(link->ctx, + link, boot)) { /* MST not supported */ link->type = dc_connection_single; sink_caps->signal = SIGNAL_TYPE_DISPLAY_PORT; @@ -654,7 +650,7 @@ static bool detect_dp( } if (link->type != dc_connection_mst_branch && - is_dp_active_dongle(link)) { + is_dp_active_dongle(link)) { /* DP active dongles */ link->type = dc_connection_active_dongle; if (!link->dpcd_caps.sink_count.bits.SINK_COUNT) { @@ -665,14 +661,15 @@ static bool detect_dp( return true; } - if (link->dpcd_caps.dongle_type != DISPLAY_DONGLE_DP_HDMI_CONVERTER) + if (link->dpcd_caps.dongle_type != + DISPLAY_DONGLE_DP_HDMI_CONVERTER) *converter_disable_audio = true; } } else { /* DP passive dongles */ sink_caps->signal = dp_passive_dongle_detection(link->ddc, - sink_caps, - audio_support); + sink_caps, + audio_support); } return true; @@ -973,6 +970,9 @@ static bool dc_link_detect_helper(struct dc_link *link, break; } + if (link->local_sink->edid_caps.panel_patch.disable_fec) + link->ctx->dc->debug.disable_fec = true; + // Check if edid is the same if ((prev_sink != NULL) && ((edid_status == EDID_THE_SAME) || (edid_status == EDID_OK))) same_edid = is_same_edid(&prev_sink->dc_edid, &sink->dc_edid); @@ -1498,9 +1498,8 @@ static void enable_stream_features(struct pipe_ctx *pipe_ctx) } } -static enum dc_status enable_link_dp( - struct dc_state *state, - struct pipe_ctx *pipe_ctx) +static enum dc_status enable_link_dp(struct dc_state *state, + struct pipe_ctx *pipe_ctx) { struct dc_stream_state *stream = pipe_ctx->stream; enum dc_status status; @@ -1532,7 +1531,8 @@ static enum dc_status enable_link_dp( pipe_ctx->stream_res.pix_clk_params.requested_sym_clk = link_settings.link_rate * LINK_RATE_REF_FREQ_IN_KHZ; if (state->clk_mgr && !apply_seamless_boot_optimization) - state->clk_mgr->funcs->update_clocks(state->clk_mgr, state, false); + state->clk_mgr->funcs->update_clocks(state->clk_mgr, + state, false); // during mode switch we do DP_SET_POWER off then on, and OUI is lost dpcd_set_source_specific_data(link); @@ -1540,21 +1540,20 @@ static enum dc_status enable_link_dp( skip_video_pattern = true; if (link_settings.link_rate == LINK_RATE_LOW) - skip_video_pattern = false; - - if (perform_link_training_with_retries( - &link_settings, - skip_video_pattern, - LINK_TRAINING_ATTEMPTS, - pipe_ctx, - pipe_ctx->stream->signal)) { + skip_video_pattern = false; + + if (perform_link_training_with_retries(&link_settings, + skip_video_pattern, + LINK_TRAINING_ATTEMPTS, + pipe_ctx, + pipe_ctx->stream->signal)) { link->cur_link_settings = link_settings; status = DC_OK; - } - else + } else { status = DC_FAIL_DP_LINK_TRAINING; + } - if (link->preferred_training_settings.fec_enable != NULL) + if (link->preferred_training_settings.fec_enable) fec_enable = *link->preferred_training_settings.fec_enable; else fec_enable = true; @@ -1766,8 +1765,7 @@ static void write_i2c_retimer_setting( slave_address, buffer[0], buffer[1], i2c_success?1:0); if (!i2c_success) - /* Write failure */ - ASSERT(i2c_success); + goto i2c_write_fail; /* Based on DP159 specs, APPLY_RX_TX_CHANGE bit in 0x0A * needs to be set to 1 on every 0xA-0xC write. @@ -1785,8 +1783,7 @@ static void write_i2c_retimer_setting( pipe_ctx->stream->link->ddc, slave_address, &offset, 1, &value, 1); if (!i2c_success) - /* Write failure */ - ASSERT(i2c_success); + goto i2c_write_fail; } buffer[0] = offset; @@ -1798,8 +1795,7 @@ static void write_i2c_retimer_setting( offset = 0x%x, reg_val = 0x%x, i2c_success = %d\n", slave_address, buffer[0], buffer[1], i2c_success?1:0); if (!i2c_success) - /* Write failure */ - ASSERT(i2c_success); + goto i2c_write_fail; } } } @@ -1819,8 +1815,7 @@ static void write_i2c_retimer_setting( slave_address, buffer[0], buffer[1], i2c_success?1:0); if (!i2c_success) - /* Write failure */ - ASSERT(i2c_success); + goto i2c_write_fail; /* Based on DP159 specs, APPLY_RX_TX_CHANGE bit in 0x0A * needs to be set to 1 on every 0xA-0xC write. @@ -1838,8 +1833,7 @@ static void write_i2c_retimer_setting( pipe_ctx->stream->link->ddc, slave_address, &offset, 1, &value, 1); if (!i2c_success) - /* Write failure */ - ASSERT(i2c_success); + goto i2c_write_fail; } buffer[0] = offset; @@ -1851,8 +1845,7 @@ static void write_i2c_retimer_setting( offset = 0x%x, reg_val = 0x%x, i2c_success = %d\n", slave_address, buffer[0], buffer[1], i2c_success?1:0); if (!i2c_success) - /* Write failure */ - ASSERT(i2c_success); + goto i2c_write_fail; } } } @@ -1870,8 +1863,7 @@ static void write_i2c_retimer_setting( offset = 0x%x, reg_val = 0x%x, i2c_success = %d\n", slave_address, buffer[0], buffer[1], i2c_success?1:0); if (!i2c_success) - /* Write failure */ - ASSERT(i2c_success); + goto i2c_write_fail; /* Write offset 0x00 to 0x23 */ buffer[0] = 0x00; @@ -1882,8 +1874,7 @@ static void write_i2c_retimer_setting( offset = 0x%x, reg_val = 0x%x, i2c_success = %d\n", slave_address, buffer[0], buffer[1], i2c_success?1:0); if (!i2c_success) - /* Write failure */ - ASSERT(i2c_success); + goto i2c_write_fail; /* Write offset 0xff to 0x00 */ buffer[0] = 0xff; @@ -1894,10 +1885,14 @@ static void write_i2c_retimer_setting( offset = 0x%x, reg_val = 0x%x, i2c_success = %d\n", slave_address, buffer[0], buffer[1], i2c_success?1:0); if (!i2c_success) - /* Write failure */ - ASSERT(i2c_success); + goto i2c_write_fail; } + + return; + +i2c_write_fail: + DC_LOG_DEBUG("Set retimer failed"); } static void write_i2c_default_retimer_setting( @@ -1922,8 +1917,7 @@ static void write_i2c_default_retimer_setting( offset = 0x%x, reg_val = 0x%x, i2c_success = %d\n", slave_address, buffer[0], buffer[1], i2c_success?1:0); if (!i2c_success) - /* Write failure */ - ASSERT(i2c_success); + goto i2c_write_fail; /* Write offset 0x0A to 0x17 */ buffer[0] = 0x0A; @@ -1934,8 +1928,7 @@ static void write_i2c_default_retimer_setting( offset = 0x%x, reg_val = 0x%x, i2c_success = %d\n", slave_address, buffer[0], buffer[1], i2c_success?1:0); if (!i2c_success) - /* Write failure */ - ASSERT(i2c_success); + goto i2c_write_fail; /* Write offset 0x0B to 0xDA or 0xD8 */ buffer[0] = 0x0B; @@ -1946,8 +1939,7 @@ static void write_i2c_default_retimer_setting( offset = 0x%x, reg_val = 0x%x, i2c_success = %d\n", slave_address, buffer[0], buffer[1], i2c_success?1:0); if (!i2c_success) - /* Write failure */ - ASSERT(i2c_success); + goto i2c_write_fail; /* Write offset 0x0A to 0x17 */ buffer[0] = 0x0A; @@ -1958,8 +1950,7 @@ static void write_i2c_default_retimer_setting( offset = 0x%x, reg_val= 0x%x, i2c_success = %d\n", slave_address, buffer[0], buffer[1], i2c_success?1:0); if (!i2c_success) - /* Write failure */ - ASSERT(i2c_success); + goto i2c_write_fail; /* Write offset 0x0C to 0x1D or 0x91 */ buffer[0] = 0x0C; @@ -1970,8 +1961,7 @@ static void write_i2c_default_retimer_setting( offset = 0x%x, reg_val = 0x%x, i2c_success = %d\n", slave_address, buffer[0], buffer[1], i2c_success?1:0); if (!i2c_success) - /* Write failure */ - ASSERT(i2c_success); + goto i2c_write_fail; /* Write offset 0x0A to 0x17 */ buffer[0] = 0x0A; @@ -1982,8 +1972,7 @@ static void write_i2c_default_retimer_setting( offset = 0x%x, reg_val = 0x%x, i2c_success = %d\n", slave_address, buffer[0], buffer[1], i2c_success?1:0); if (!i2c_success) - /* Write failure */ - ASSERT(i2c_success); + goto i2c_write_fail; if (is_vga_mode) { @@ -1998,8 +1987,7 @@ static void write_i2c_default_retimer_setting( offset = 0x%x, reg_val = 0x%x, i2c_success = %d\n", slave_address, buffer[0], buffer[1], i2c_success?1:0); if (!i2c_success) - /* Write failure */ - ASSERT(i2c_success); + goto i2c_write_fail; /* Write offset 0x00 to 0x23 */ buffer[0] = 0x00; @@ -2010,8 +1998,7 @@ static void write_i2c_default_retimer_setting( offset = 0x%x, reg_val= 0x%x, i2c_success = %d\n", slave_address, buffer[0], buffer[1], i2c_success?1:0); if (!i2c_success) - /* Write failure */ - ASSERT(i2c_success); + goto i2c_write_fail; /* Write offset 0xff to 0x00 */ buffer[0] = 0xff; @@ -2022,9 +2009,13 @@ static void write_i2c_default_retimer_setting( offset = 0x%x, reg_val= 0x%x, i2c_success = %d end here\n", slave_address, buffer[0], buffer[1], i2c_success?1:0); if (!i2c_success) - /* Write failure */ - ASSERT(i2c_success); + goto i2c_write_fail; } + + return; + +i2c_write_fail: + DC_LOG_DEBUG("Set default retimer failed"); } static void write_i2c_redriver_setting( @@ -2053,8 +2044,7 @@ static void write_i2c_redriver_setting( slave_address, buffer[3], buffer[4], buffer[5], buffer[6], i2c_success?1:0); if (!i2c_success) - /* Write failure */ - ASSERT(i2c_success); + DC_LOG_DEBUG("Set redriver failed"); } static void disable_link(struct dc_link *link, enum signal_type signal) @@ -2960,6 +2950,8 @@ static void update_psp_stream_config(struct pipe_ctx *pipe_ctx, bool dpms_off) config.link_enc_inst = pipe_ctx->stream->link->link_enc_hw_inst; config.dpms_off = dpms_off; config.dm_stream_ctx = pipe_ctx->stream->dm_stream_context; + config.mst_supported = (pipe_ctx->stream->signal == + SIGNAL_TYPE_DISPLAY_PORT_MST); cp_psp->funcs.update_stream_config(cp_psp->handle, &config); } } @@ -3410,7 +3402,7 @@ uint32_t dc_link_bandwidth_kbps( link_bw_kbps *= 8; /* 8 bits per byte*/ link_bw_kbps *= link_setting->lane_count; - if (link->dpcd_caps.fec_cap.bits.FEC_CAPABLE) { + if (dc_link_is_fec_supported(link) && !link->dc->debug.disable_fec) { /* Account for FEC overhead. * We have to do it based on caps, * and not based on FEC being set ready, @@ -3454,3 +3446,11 @@ void dc_link_overwrite_extended_receiver_cap( dp_overwrite_extended_receiver_cap(link); } +bool dc_link_is_fec_supported(const struct dc_link *link) +{ + return (dc_is_dp_signal(link->connector_signal) && + link->link_enc->features.fec_supported && + link->dpcd_caps.fec_cap.bits.FEC_CAPABLE && + !IS_FPGA_MAXIMUS_DC(link->ctx->dce_environment)); +} + diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c index 93127bc90f3c..7cbb1efb4f68 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c @@ -1446,11 +1446,15 @@ enum link_training_result dc_link_dp_perform_link_training( &link->preferred_training_settings, <_settings); - /* 1. set link rate, lane count and spread. */ + /* Configure lttpr mode */ + if (!link->is_lttpr_mode_transparent) + configure_lttpr_mode(link); + if (link->ctx->dc->work_arounds.lt_early_cr_pattern) start_clock_recovery_pattern_early(link, <_settings, DPRX); - else - dpcd_set_link_settings(link, <_settings); + + /* 1. set link rate, lane count and spread. */ + dpcd_set_link_settings(link, <_settings); if (link->preferred_training_settings.fec_enable != NULL) fec_enable = *link->preferred_training_settings.fec_enable; @@ -1460,8 +1464,6 @@ enum link_training_result dc_link_dp_perform_link_training( dp_set_fec_ready(link, fec_enable); if (!link->is_lttpr_mode_transparent) { - /* Configure lttpr mode */ - configure_lttpr_mode(link); /* 2. perform link training (set link training done * to false is done as well) @@ -1669,11 +1671,11 @@ enum link_training_result dc_link_dp_sync_lt_attempt( dp_set_panel_mode(link, panel_mode); /* Attempt to train with given link training settings */ - /* Set link rate, lane count and spread. */ if (link->ctx->dc->work_arounds.lt_early_cr_pattern) start_clock_recovery_pattern_early(link, <_settings, DPRX); - else - dpcd_set_link_settings(link, <_settings); + + /* Set link rate, lane count and spread. */ + dpcd_set_link_settings(link, <_settings); /* 2. perform link training (set link training done * to false is done as well) @@ -2672,9 +2674,12 @@ static void dp_test_send_link_test_pattern(struct dc_link *link) break; } - test_pattern_color_space = dpcd_test_params.bits.YCBCR_COEFS ? - DP_TEST_PATTERN_COLOR_SPACE_YCBCR709 : - DP_TEST_PATTERN_COLOR_SPACE_YCBCR601; + if (dpcd_test_params.bits.CLR_FORMAT == 0) + test_pattern_color_space = DP_TEST_PATTERN_COLOR_SPACE_RGB; + else + test_pattern_color_space = dpcd_test_params.bits.YCBCR_COEFS ? + DP_TEST_PATTERN_COLOR_SPACE_YCBCR709 : + DP_TEST_PATTERN_COLOR_SPACE_YCBCR601; dc_link_dp_set_test_pattern( link, @@ -3436,6 +3441,17 @@ static bool retrieve_link_cap(struct dc_link *link) sink_id.ieee_device_id, sizeof(sink_id.ieee_device_id)); + /* Quirk Apple MBP 2017 15" Retina panel: Wrong DP_MAX_LINK_RATE */ + { + uint8_t str_mbp_2017[] = { 101, 68, 21, 101, 98, 97 }; + + if ((link->dpcd_caps.sink_dev_id == 0x0010fa) && + !memcmp(link->dpcd_caps.sink_dev_id_str, str_mbp_2017, + sizeof(str_mbp_2017))) { + link->reported_link_cap.link_rate = 0x0c; + } + } + core_link_read_dpcd( link, DP_SINK_HW_REVISION_START, @@ -3720,7 +3736,8 @@ static void set_crtc_test_pattern(struct dc_link *link, struct pipe_ctx *odm_pipe; enum controller_dp_color_space controller_color_space; int opp_cnt = 1; - uint16_t count = 0; + int offset = 0; + int dpg_width = width; switch (test_pattern_color_space) { case DP_TEST_PATTERN_COLOR_SPACE_RGB: @@ -3742,33 +3759,30 @@ static void set_crtc_test_pattern(struct dc_link *link, for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) opp_cnt++; + dpg_width = width / opp_cnt; + offset = dpg_width; - width /= opp_cnt; + opp->funcs->opp_set_disp_pattern_generator(opp, + controller_test_pattern, + controller_color_space, + color_depth, + NULL, + dpg_width, + height, + 0); for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) { struct output_pixel_processor *odm_opp = odm_pipe->stream_res.opp; - odm_opp->funcs->opp_program_bit_depth_reduction(odm_opp, ¶ms); odm_opp->funcs->opp_set_disp_pattern_generator(odm_opp, controller_test_pattern, controller_color_space, color_depth, NULL, - width, - height); - } - opp->funcs->opp_set_disp_pattern_generator(opp, - controller_test_pattern, - controller_color_space, - color_depth, - NULL, - width, - height); - /* wait for dpg to blank pixel data with test pattern */ - for (count = 0; count < 1000; count++) { - if (opp->funcs->dpg_is_blanked(opp)) - break; - udelay(100); + dpg_width, + height, + offset); + offset += offset; } } } @@ -3786,11 +3800,12 @@ static void set_crtc_test_pattern(struct dc_link *link, else if (opp->funcs->opp_set_disp_pattern_generator) { struct pipe_ctx *odm_pipe; int opp_cnt = 1; + int dpg_width = width; for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) opp_cnt++; - width /= opp_cnt; + dpg_width = width / opp_cnt; for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) { struct output_pixel_processor *odm_opp = odm_pipe->stream_res.opp; @@ -3800,16 +3815,18 @@ static void set_crtc_test_pattern(struct dc_link *link, CONTROLLER_DP_COLOR_SPACE_UDEFINED, color_depth, NULL, - width, - height); + dpg_width, + height, + 0); } opp->funcs->opp_set_disp_pattern_generator(opp, CONTROLLER_DP_TEST_PATTERN_VIDEOMODE, CONTROLLER_DP_COLOR_SPACE_UDEFINED, color_depth, NULL, - width, - height); + dpg_width, + height, + 0); } } break; @@ -3987,6 +4004,11 @@ bool dc_link_dp_set_test_pattern( default: break; } + + if (pipe_ctx->stream_res.tg->funcs->lock_doublebuffer_enable) + pipe_ctx->stream_res.tg->funcs->lock_doublebuffer_enable( + pipe_ctx->stream_res.tg); + pipe_ctx->stream_res.tg->funcs->lock(pipe_ctx->stream_res.tg); /* update MSA to requested color space */ pipe_ctx->stream_res.stream_enc->funcs->dp_set_stream_attribute(pipe_ctx->stream_res.stream_enc, &pipe_ctx->stream->timing, @@ -3994,9 +4016,27 @@ bool dc_link_dp_set_test_pattern( pipe_ctx->stream->use_vsc_sdp_for_colorimetry, link->dpcd_caps.dprx_feature.bits.SST_SPLIT_SDP_CAP); + if (pipe_ctx->stream->use_vsc_sdp_for_colorimetry) { + if (test_pattern == DP_TEST_PATTERN_COLOR_SQUARES_CEA) + pipe_ctx->stream->vsc_infopacket.sb[17] |= (1 << 7); // sb17 bit 7 Dynamic Range: 0 = VESA range, 1 = CTA range + else + pipe_ctx->stream->vsc_infopacket.sb[17] &= ~(1 << 7); + resource_build_info_frame(pipe_ctx); + link->dc->hwss.update_info_frame(pipe_ctx); + } + /* CRTC Patterns */ set_crtc_test_pattern(link, pipe_ctx, test_pattern, test_pattern_color_space); - + pipe_ctx->stream_res.tg->funcs->unlock(pipe_ctx->stream_res.tg); + pipe_ctx->stream_res.tg->funcs->wait_for_state(pipe_ctx->stream_res.tg, + CRTC_STATE_VACTIVE); + pipe_ctx->stream_res.tg->funcs->wait_for_state(pipe_ctx->stream_res.tg, + CRTC_STATE_VBLANK); + pipe_ctx->stream_res.tg->funcs->wait_for_state(pipe_ctx->stream_res.tg, + CRTC_STATE_VACTIVE); + if (pipe_ctx->stream_res.tg->funcs->lock_doublebuffer_disable) + pipe_ctx->stream_res.tg->funcs->lock_doublebuffer_disable( + pipe_ctx->stream_res.tg); /* Set Test Pattern state */ link->test_pattern_enabled = true; } @@ -4126,8 +4166,7 @@ void dp_set_fec_ready(struct dc_link *link, bool ready) struct link_encoder *link_enc = link->link_enc; uint8_t fec_config = 0; - if (link->dc->debug.disable_fec || - IS_FPGA_MAXIMUS_DC(link->ctx->dce_environment)) + if (!dc_link_is_fec_supported(link) || link->dc->debug.disable_fec) return; if (link_enc->funcs->fec_set_ready && @@ -4162,8 +4201,7 @@ void dp_set_fec_enable(struct dc_link *link, bool enable) { struct link_encoder *link_enc = link->link_enc; - if (link->dc->debug.disable_fec || - IS_FPGA_MAXIMUS_DC(link->ctx->dce_environment)) + if (!dc_link_is_fec_supported(link) || link->dc->debug.disable_fec) return; if (link_enc->funcs->fec_set_enable && diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c index 58634f191a55..51e0ee6e7695 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c @@ -431,6 +431,7 @@ void dp_set_dsc_on_stream(struct pipe_ctx *pipe_ctx, bool enable) dsc_cfg.pic_height = stream->timing.v_addressable + stream->timing.v_border_top + stream->timing.v_border_bottom; dsc_cfg.pixel_encoding = stream->timing.pixel_encoding; dsc_cfg.color_depth = stream->timing.display_color_depth; + dsc_cfg.is_odm = pipe_ctx->next_odm_pipe ? true : false; dsc_cfg.dc_dsc_cfg = stream->timing.dsc_cfg; ASSERT(dsc_cfg.dc_dsc_cfg.num_slices_h % opp_cnt == 0); dsc_cfg.dc_dsc_cfg.num_slices_h /= opp_cnt; @@ -535,6 +536,7 @@ bool dp_set_dsc_pps_sdp(struct pipe_ctx *pipe_ctx, bool enable) dsc_cfg.pic_height = stream->timing.v_addressable + stream->timing.v_border_top + stream->timing.v_border_bottom; dsc_cfg.pixel_encoding = stream->timing.pixel_encoding; dsc_cfg.color_depth = stream->timing.display_color_depth; + dsc_cfg.is_odm = pipe_ctx->next_odm_pipe ? true : false; dsc_cfg.dc_dsc_cfg = stream->timing.dsc_cfg; DC_LOG_DSC(" "); diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c index 572ce3842535..75c7ce4c7581 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c @@ -893,6 +893,7 @@ static void calculate_inits_and_adj_vp(struct pipe_ctx *pipe_ctx) int vpc_div = (data->format == PIXEL_FORMAT_420BPP8 || data->format == PIXEL_FORMAT_420BPP10) ? 2 : 1; bool orthogonal_rotation, flip_vert_scan_dir, flip_horz_scan_dir; + int odm_idx = 0; /* * Need to calculate the scan direction for viewport to make adjustments @@ -924,11 +925,13 @@ static void calculate_inits_and_adj_vp(struct pipe_ctx *pipe_ctx) * stream->dst.width / stream->src.width - src.x * plane_state->dst_rect.width / src.width * stream->dst.width / stream->src.width); - /*modified recout_skip_h calculation due to odm having no recout offset caused by split*/ + /*modified recout_skip_h calculation due to odm having no recout offset*/ while (odm_pipe) { - recout_skip_h += odm_pipe->plane_res.scl_data.recout.width + odm_pipe->plane_res.scl_data.recout.x; + odm_idx++; odm_pipe = odm_pipe->prev_odm_pipe; } + if (odm_idx) + recout_skip_h += odm_idx * data->recout.width; recout_skip_v = data->recout.y - (stream->dst.y + (plane_state->dst_rect.y - stream->src.y) * stream->dst.height / stream->src.height - @@ -2171,10 +2174,10 @@ enum dc_status dc_validate_global_state( if (pipe_ctx->stream != stream) continue; - if (dc->res_pool->funcs->get_default_swizzle_mode && + if (dc->res_pool->funcs->patch_unknown_plane_state && pipe_ctx->plane_state && pipe_ctx->plane_state->tiling_info.gfx9.swizzle == DC_SW_UNKNOWN) { - result = dc->res_pool->funcs->get_default_swizzle_mode(pipe_ctx->plane_state); + result = dc->res_pool->funcs->patch_unknown_plane_state(pipe_ctx->plane_state); if (result != DC_OK) return result; } diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h index bc1220dce3b1..d3ceb39e428e 100644 --- a/drivers/gpu/drm/amd/display/dc/dc.h +++ b/drivers/gpu/drm/amd/display/dc/dc.h @@ -39,7 +39,7 @@ #include "inc/hw/dmcu.h" #include "dml/display_mode_lib.h" -#define DC_VER "3.2.74" +#define DC_VER "3.2.76" #define MAX_SURFACES 3 #define MAX_PLANES 6 @@ -230,6 +230,7 @@ struct dc_config { bool forced_clocks; bool disable_extended_timeout_support; // Used to disable extended timeout and lttpr feature as well bool multi_mon_pp_mclk_switch; + bool psr_on_dmub; }; enum visual_confirm { @@ -389,6 +390,7 @@ struct dc_debug_options { int always_scale; bool disable_pplib_clock_request; bool disable_clock_gate; + bool disable_mem_low_power; bool disable_dmcu; bool disable_psr; bool force_abm_enable; @@ -410,7 +412,6 @@ struct dc_debug_options { bool dmub_offload_enabled; bool dmcub_emulation; bool dmub_command_table; /* for testing only */ - bool psr_on_dmub; struct dc_bw_validation_profile bw_val_profile; bool disable_fec; bool disable_48mhz_pwrdwn; @@ -520,7 +521,7 @@ struct dc { struct dce_hwseq *hwseq; /* Require to optimize clocks and bandwidth for added/removed planes */ - bool clk_optimized_required; + bool optimized_required; bool wm_optimized_required; /* Require to maintain clocks and bandwidth for UEFI enabled HW */ @@ -871,6 +872,7 @@ struct dc_flip_addrs { unsigned int flip_timestamp_in_us; bool flip_immediate; /* TODO: add flip duration for FreeSync */ + bool triplebuffer_flips; }; bool dc_post_update_surfaces_to_stream( @@ -1024,6 +1026,11 @@ struct dc_sink_dsc_caps { struct dsc_dec_dpcd_caps dsc_dec_caps; }; +struct dc_sink_fec_caps { + bool is_rx_fec_supported; + bool is_topology_fec_supported; +}; + /* * The sink structure contains EDID and other display device properties */ @@ -1037,7 +1044,10 @@ struct dc_sink { struct stereo_3d_features features_3d[TIMING_3D_FORMAT_MAX]; bool converter_disable_audio; - struct dc_sink_dsc_caps sink_dsc_caps; + struct dc_sink_dsc_caps dsc_caps; + struct dc_sink_fec_caps fec_caps; + + bool is_vsc_sdp_colorimetry_supported; /* private to DC core */ struct dc_link *link; diff --git a/drivers/gpu/drm/amd/display/dc/dc_link.h b/drivers/gpu/drm/amd/display/dc/dc_link.h index c45c7680fa58..00ff5e98278c 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_link.h +++ b/drivers/gpu/drm/amd/display/dc/dc_link.h @@ -333,4 +333,7 @@ bool dc_submit_i2c_oem( uint32_t dc_bandwidth_in_kbps_from_timing( const struct dc_crtc_timing *timing); + +bool dc_link_is_fec_supported(const struct dc_link *link); + #endif /* DC_LINK_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dc_types.h b/drivers/gpu/drm/amd/display/dc/dc_types.h index 299f6e00f576..0d210104ba0a 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_types.h @@ -231,6 +231,7 @@ struct dc_panel_patch { unsigned int extra_t7_ms; unsigned int skip_scdc_overwrite; unsigned int delay_ignore_msa; + unsigned int disable_fec; }; struct dc_edid_caps { diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c b/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c index 68c4049cbc2a..743042d5905a 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c @@ -645,7 +645,7 @@ bool dce_aux_transfer_with_retries(struct ddc_service *ddc, case AUX_TRANSACTION_REPLY_AUX_DEFER: case AUX_TRANSACTION_REPLY_I2C_OVER_AUX_DEFER: retry_on_defer = true; - /* fall through */ + fallthrough; case AUX_TRANSACTION_REPLY_I2C_OVER_AUX_NACK: if (++aux_defer_retries >= AUX_MAX_DEFER_RETRIES) { goto fail; diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.c b/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.c index 066188ba7949..24adec407972 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.c @@ -267,6 +267,9 @@ static void set_speed( uint32_t xtal_ref_div = 0; uint32_t prescale = 0; + if (speed == 0) + return; + REG_GET(MICROSECOND_TIME_BASE_DIV, XTAL_REF_DIV, &xtal_ref_div); if (xtal_ref_div == 0) @@ -274,17 +277,15 @@ static void set_speed( prescale = ((dce_i2c_hw->reference_frequency * 2) / xtal_ref_div) / speed; - if (speed) { - if (dce_i2c_hw->masks->DC_I2C_DDC1_START_STOP_TIMING_CNTL) - REG_UPDATE_N(SPEED, 3, - FN(DC_I2C_DDC1_SPEED, DC_I2C_DDC1_PRESCALE), prescale, - FN(DC_I2C_DDC1_SPEED, DC_I2C_DDC1_THRESHOLD), 2, - FN(DC_I2C_DDC1_SPEED, DC_I2C_DDC1_START_STOP_TIMING_CNTL), speed > 50 ? 2:1); - else - REG_UPDATE_N(SPEED, 2, - FN(DC_I2C_DDC1_SPEED, DC_I2C_DDC1_PRESCALE), prescale, - FN(DC_I2C_DDC1_SPEED, DC_I2C_DDC1_THRESHOLD), 2); - } + if (dce_i2c_hw->masks->DC_I2C_DDC1_START_STOP_TIMING_CNTL) + REG_UPDATE_N(SPEED, 3, + FN(DC_I2C_DDC1_SPEED, DC_I2C_DDC1_PRESCALE), prescale, + FN(DC_I2C_DDC1_SPEED, DC_I2C_DDC1_THRESHOLD), 2, + FN(DC_I2C_DDC1_SPEED, DC_I2C_DDC1_START_STOP_TIMING_CNTL), speed > 50 ? 2:1); + else + REG_UPDATE_N(SPEED, 2, + FN(DC_I2C_DDC1_SPEED, DC_I2C_DDC1_PRESCALE), prescale, + FN(DC_I2C_DDC1_SPEED, DC_I2C_DDC1_THRESHOLD), 2); } static bool setup_engine( diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_mem_input.c b/drivers/gpu/drm/amd/display/dc/dce/dce_mem_input.c index 8aa937f496c4..51481e922eb9 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_mem_input.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_mem_input.c @@ -479,7 +479,7 @@ static void program_grph_pixel_format( case SURFACE_PIXEL_FORMAT_GRPH_ABGR16161616F: sign = 1; floating = 1; - /* fall through */ + fallthrough; case SURFACE_PIXEL_FORMAT_GRPH_ARGB16161616F: /* shouldn't this get float too? */ case SURFACE_PIXEL_FORMAT_GRPH_ARGB16161616: grph_depth = 3; diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c index 2c932c29f1f9..bc109d4fc6e6 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c @@ -134,17 +134,15 @@ static bool dmub_psr_copy_settings(struct dmub_psr *dmub, int i = 0; for (i = 0; i < MAX_PIPES; i++) { - if (res_ctx && - res_ctx->pipe_ctx[i].stream && - res_ctx->pipe_ctx[i].stream->link && - res_ctx->pipe_ctx[i].stream->link == link && - res_ctx->pipe_ctx[i].stream->link->connector_signal == SIGNAL_TYPE_EDP) { + if (res_ctx->pipe_ctx[i].stream && + res_ctx->pipe_ctx[i].stream->link == link && + res_ctx->pipe_ctx[i].stream->link->connector_signal == SIGNAL_TYPE_EDP) { pipe_ctx = &res_ctx->pipe_ctx[i]; break; } } - if (!pipe_ctx || !&pipe_ctx->plane_res || !&pipe_ctx->stream_res) + if (!pipe_ctx) return false; // First, set the psr version @@ -235,6 +233,6 @@ struct dmub_psr *dmub_psr_create(struct dc_context *ctx) */ void dmub_psr_destroy(struct dmub_psr **dmub) { - kfree(dmub); + kfree(*dmub); *dmub = NULL; } diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c index e441c149ff40..deccab0228d2 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c @@ -830,8 +830,8 @@ static void hubbub1_det_request_size( hubbub1_get_blk256_size(&blk256_width, &blk256_height, bpe); - swath_bytes_horz_wc = height * blk256_height * bpe; - swath_bytes_vert_wc = width * blk256_width * bpe; + swath_bytes_horz_wc = width * blk256_height * bpe; + swath_bytes_vert_wc = height * blk256_width * bpe; *req128_horz_wc = (2 * swath_bytes_horz_wc <= detile_buf_size) ? false : /* full 256B request */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c index 385250e1e3fd..9cc3314966bd 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c @@ -1048,7 +1048,7 @@ void dcn10_plane_atomic_disconnect(struct dc *dc, struct pipe_ctx *pipe_ctx) if (opp != NULL) opp->mpcc_disconnect_pending[pipe_ctx->plane_res.mpcc_inst] = true; - dc->clk_optimized_required = true; + dc->optimized_required = true; if (hubp->funcs->hubp_disconnect) hubp->funcs->hubp_disconnect(hubp); @@ -1099,7 +1099,7 @@ void dcn10_plane_atomic_disable(struct dc *dc, struct pipe_ctx *pipe_ctx) false); hubp->power_gated = true; - dc->clk_optimized_required = false; /* We're powering off, no need to optimize */ + dc->optimized_required = false; /* We're powering off, no need to optimize */ hws->funcs.plane_atomic_power_down(dc, pipe_ctx->plane_res.dpp, @@ -1356,6 +1356,9 @@ void dcn10_init_hw(struct dc *dc) */ if (dcb->funcs->is_accelerated_mode(dcb) || dc->config.power_down_display_on_boot) { hws->funcs.init_pipes(dc, dc->current_state); + if (dc->res_pool->hubbub->funcs->allow_self_refresh_control) + dc->res_pool->hubbub->funcs->allow_self_refresh_control(dc->res_pool->hubbub, + !dc->res_pool->hubbub->ctx->dc->debug.disable_stutter); } for (i = 0; i < res_pool->audio_count; i++) { @@ -2717,30 +2720,20 @@ void dcn10_optimize_bandwidth( hws->funcs.verify_allow_pstate_change_high(dc); if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { - if (context->stream_count == 0) { + if (context->stream_count == 0) context->bw_ctx.bw.dcn.clk.phyclk_khz = 0; - dc->clk_mgr->funcs->update_clocks( - dc->clk_mgr, - context, - true); - } else if (dc->clk_optimized_required || IS_DIAG_DC(dc->ctx->dce_environment)) { - dc->clk_mgr->funcs->update_clocks( - dc->clk_mgr, - context, - true); - } - } - - if (dc->wm_optimized_required || IS_DIAG_DC(dc->ctx->dce_environment)) { - hubbub->funcs->program_watermarks(hubbub, - &context->bw_ctx.bw.dcn.watermarks, - dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000, + dc->clk_mgr->funcs->update_clocks( + dc->clk_mgr, + context, true); } - dc->clk_optimized_required = false; - dc->wm_optimized_required = false; + hubbub->funcs->program_watermarks(hubbub, + &context->bw_ctx.bw.dcn.watermarks, + dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000, + true); + dcn10_stereo_hw_frame_pack_wa(dc, context); if (dc->debug.pplib_wm_report_mode == WM_REPORT_OVERRIDE) diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.h index eb13589b9a81..762109174fb8 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.h @@ -62,11 +62,11 @@ SRI(DP_DPHY_FAST_TRAINING, DP, id), \ SRI(DP_SEC_CNTL1, DP, id), \ SRI(DP_DPHY_BS_SR_SWAP_CNTL, DP, id), \ - SRI(DP_DPHY_INTERNAL_CTRL, DP, id), \ SRI(DP_DPHY_HBR2_PATTERN_CONTROL, DP, id) #define LE_DCN10_REG_LIST(id)\ + SRI(DP_DPHY_INTERNAL_CTRL, DP, id), \ LE_DCN_COMMON_REG_LIST(id) struct dcn10_link_enc_aux_registers { diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c index a9a43b397db9..63acb8ff7462 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c @@ -299,7 +299,6 @@ void optc1_set_vtg_params(struct timing_generator *optc, uint32_t asic_blank_end; uint32_t v_init; uint32_t v_fp2 = 0; - int32_t vertical_line_start; struct optc *optc1 = DCN10TG_FROM_TG(optc); @@ -316,9 +315,8 @@ void optc1_set_vtg_params(struct timing_generator *optc, patched_crtc_timing.v_border_top; /* if VSTARTUP is before VSYNC, FP2 is the offset, otherwise 0 */ - vertical_line_start = asic_blank_end - optc1->vstartup_start + 1; - if (vertical_line_start < 0) - v_fp2 = -vertical_line_start; + if (optc1->vstartup_start > asic_blank_end) + v_fp2 = optc1->vstartup_start - asic_blank_end; /* Interlace */ if (REG(OTG_INTERLACE_CONTROL)) { @@ -1195,7 +1193,7 @@ static void optc1_enable_stereo(struct timing_generator *optc, REG_UPDATE_3(OTG_STEREO_CONTROL, OTG_STEREO_EN, stereo_en, OTG_STEREO_SYNC_OUTPUT_LINE_NUM, 0, - OTG_STEREO_SYNC_OUTPUT_POLARITY, 0); + OTG_STEREO_SYNC_OUTPUT_POLARITY, flags->RIGHT_EYE_POLARITY == 0 ? 0 : 1); if (flags->PROGRAM_POLARITY) REG_UPDATE(OTG_STEREO_CONTROL, diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c index 3b71898e859e..261bdc3a8218 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c @@ -570,7 +570,7 @@ static const struct dc_plane_cap plane_cap = { static const struct dc_debug_options debug_defaults_drv = { .sanity_checks = true, - .disable_dmcu = true, + .disable_dmcu = false, .force_abm_enable = false, .timing_trace = false, .clock_trace = true, @@ -598,7 +598,7 @@ static const struct dc_debug_options debug_defaults_drv = { }; static const struct dc_debug_options debug_defaults_diags = { - .disable_dmcu = true, + .disable_dmcu = false, .force_abm_enable = false, .timing_trace = true, .clock_trace = true, @@ -1233,7 +1233,7 @@ static enum dc_status dcn10_validate_global(struct dc *dc, struct dc_state *cont return DC_OK; } -static enum dc_status dcn10_get_default_swizzle_mode(struct dc_plane_state *plane_state) +static enum dc_status dcn10_patch_unknown_plane_state(struct dc_plane_state *plane_state) { enum dc_status result = DC_OK; @@ -1295,7 +1295,7 @@ static const struct resource_funcs dcn10_res_pool_funcs = { .validate_plane = dcn10_validate_plane, .validate_global = dcn10_validate_global, .add_stream_to_ctx = dcn10_add_stream_to_ctx, - .get_default_swizzle_mode = dcn10_get_default_swizzle_mode, + .patch_unknown_plane_state = dcn10_patch_unknown_plane_state, .find_first_free_match_stream_enc_for_link = dcn10_find_first_free_match_stream_enc_for_link }; diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dccg.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dccg.c index 50bffbfdd394..62cc2651e00c 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dccg.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dccg.c @@ -70,6 +70,8 @@ void dccg2_update_dpp_dto(struct dccg *dccg, int dpp_inst, int req_dppclk) REG_UPDATE(DPPCLK_DTO_CTRL, DPPCLK_DTO_ENABLE[dpp_inst], 0); } + + dccg->pipe_dppclk_khz[dpp_inst] = req_dppclk; } void dccg2_get_dccg_ref_freq(struct dccg *dccg, diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.c index 13e057d7ee93..42bba7c9548b 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.c @@ -369,84 +369,6 @@ void dpp2_set_cursor_attributes( } } -#define IDENTITY_RATIO(ratio) (dc_fixpt_u3d19(ratio) == (1 << 19)) - -bool dpp2_get_optimal_number_of_taps( - struct dpp *dpp, - struct scaler_data *scl_data, - const struct scaling_taps *in_taps) -{ - /* Some ASICs does not support FP16 scaling, so we reject modes require this*/ - if (scl_data->viewport.width != scl_data->h_active && - scl_data->viewport.height != scl_data->v_active && - dpp->caps->dscl_data_proc_format == DSCL_DATA_PRCESSING_FIXED_FORMAT && - scl_data->format == PIXEL_FORMAT_FP16) - return false; - - if (scl_data->viewport.width > scl_data->h_active && - dpp->ctx->dc->debug.max_downscale_src_width != 0 && - scl_data->viewport.width > dpp->ctx->dc->debug.max_downscale_src_width) - return false; - - /* TODO: add lb check */ - - /* No support for programming ratio of 8, drop to 7.99999.. */ - if (scl_data->ratios.horz.value == (8ll << 32)) - scl_data->ratios.horz.value--; - if (scl_data->ratios.vert.value == (8ll << 32)) - scl_data->ratios.vert.value--; - if (scl_data->ratios.horz_c.value == (8ll << 32)) - scl_data->ratios.horz_c.value--; - if (scl_data->ratios.vert_c.value == (8ll << 32)) - scl_data->ratios.vert_c.value--; - - /* Set default taps if none are provided */ - if (in_taps->h_taps == 0) { - if (dc_fixpt_ceil(scl_data->ratios.horz) > 4) - scl_data->taps.h_taps = 8; - else - scl_data->taps.h_taps = 4; - } else - scl_data->taps.h_taps = in_taps->h_taps; - if (in_taps->v_taps == 0) { - if (dc_fixpt_ceil(scl_data->ratios.vert) > 4) - scl_data->taps.v_taps = 8; - else - scl_data->taps.v_taps = 4; - } else - scl_data->taps.v_taps = in_taps->v_taps; - if (in_taps->v_taps_c == 0) { - if (dc_fixpt_ceil(scl_data->ratios.vert_c) > 4) - scl_data->taps.v_taps_c = 4; - else - scl_data->taps.v_taps_c = 2; - } else - scl_data->taps.v_taps_c = in_taps->v_taps_c; - if (in_taps->h_taps_c == 0) { - if (dc_fixpt_ceil(scl_data->ratios.horz_c) > 4) - scl_data->taps.h_taps_c = 4; - else - scl_data->taps.h_taps_c = 2; - } else if ((in_taps->h_taps_c % 2) != 0 && in_taps->h_taps_c != 1) - /* Only 1 and even h_taps_c are supported by hw */ - scl_data->taps.h_taps_c = in_taps->h_taps_c - 1; - else - scl_data->taps.h_taps_c = in_taps->h_taps_c; - - if (!dpp->ctx->dc->debug.always_scale) { - if (IDENTITY_RATIO(scl_data->ratios.horz)) - scl_data->taps.h_taps = 1; - if (IDENTITY_RATIO(scl_data->ratios.vert)) - scl_data->taps.v_taps = 1; - if (IDENTITY_RATIO(scl_data->ratios.horz_c)) - scl_data->taps.h_taps_c = 1; - if (IDENTITY_RATIO(scl_data->ratios.vert_c)) - scl_data->taps.v_taps_c = 1; - } - - return true; -} - void oppn20_dummy_program_regamma_pwl( struct dpp *dpp, const struct pwl_params *params, diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.c index 6bdfee20b6a7..1b1ae9ce2799 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.c @@ -369,6 +369,7 @@ static bool dsc_prepare_config(const struct dsc_config *dsc_cfg, struct dsc_reg_ dsc_reg_vals->pps.block_pred_enable = dsc_cfg->dc_dsc_cfg.block_pred_enable; dsc_reg_vals->pps.line_buf_depth = dsc_cfg->dc_dsc_cfg.linebuf_depth; dsc_reg_vals->alternate_ich_encoding_en = dsc_reg_vals->pps.dsc_version_minor == 1 ? 0 : 1; + dsc_reg_vals->ich_reset_at_eol = (dsc_cfg->is_odm || dsc_reg_vals->num_slices_h > 1) ? 0xF : 0; // TODO: in addition to validating slice height (pic height must be divisible by slice height), // see what happens when the same condition doesn't apply for slice_width/pic_width. @@ -531,7 +532,6 @@ static void dsc_update_from_dsc_parameters(struct dsc_reg_values *reg_vals, cons reg_vals->pps.rc_buf_thresh[i] = reg_vals->pps.rc_buf_thresh[i] >> 6; reg_vals->rc_buffer_model_size = dsc_params->rc_buffer_model_size; - reg_vals->ich_reset_at_eol = reg_vals->num_slices_h == 1 ? 0 : 0xf; } static void dsc_write_to_registers(struct display_stream_compressor *dsc, const struct dsc_reg_values *reg_vals) diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c index 97c0c8ced8e5..233318260da4 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c @@ -307,7 +307,8 @@ void dcn20_init_blank( COLOR_DEPTH_UNDEFINED, &black_color, otg_active_width, - otg_active_height); + otg_active_height, + 0); if (num_opps == 2) { bottom_opp->funcs->opp_set_disp_pattern_generator( @@ -317,7 +318,8 @@ void dcn20_init_blank( COLOR_DEPTH_UNDEFINED, &black_color, otg_active_width, - otg_active_height); + otg_active_height, + 0); } hws->funcs.wait_for_blank_complete(opp); @@ -645,6 +647,9 @@ enum dc_status dcn20_enable_stream_timing( return DC_ERROR_UNEXPECTED; } + if (dc->hwseq->funcs.PLAT_58856_wa && (!dc_is_dp_signal(stream->signal))) + dc->hwseq->funcs.PLAT_58856_wa(context, pipe_ctx); + pipe_ctx->stream_res.tg->funcs->program_timing( pipe_ctx->stream_res.tg, &stream->timing, @@ -974,7 +979,8 @@ void dcn20_blank_pixel_data( stream->timing.display_color_depth, &black_color, width, - height); + height, + 0); for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) { odm_pipe->stream_res.opp->funcs->opp_set_disp_pattern_generator( @@ -985,7 +991,8 @@ void dcn20_blank_pixel_data( stream->timing.display_color_depth, &black_color, width, - height); + height, + 0); } if (!blank) @@ -1656,22 +1663,16 @@ void dcn20_optimize_bandwidth( { struct hubbub *hubbub = dc->res_pool->hubbub; - if (dc->wm_optimized_required || IS_DIAG_DC(dc->ctx->dce_environment)) { - /* program dchubbub watermarks */ - hubbub->funcs->program_watermarks(hubbub, - &context->bw_ctx.bw.dcn.watermarks, - dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000, - true); - dc->wm_optimized_required = false; - } + /* program dchubbub watermarks */ + hubbub->funcs->program_watermarks(hubbub, + &context->bw_ctx.bw.dcn.watermarks, + dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000, + true); - if (dc->clk_optimized_required || IS_DIAG_DC(dc->ctx->dce_environment)) { - dc->clk_mgr->funcs->update_clocks( - dc->clk_mgr, - context, - true); - dc->wm_optimized_required = false; - } + dc->clk_mgr->funcs->update_clocks( + dc->clk_mgr, + context, + true); } bool dcn20_update_bandwidth( diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.c index 6c4f90f58656..1e73357eda34 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.c @@ -111,7 +111,6 @@ static const struct hwseq_private_funcs dcn20_private_funcs = { .enable_power_gating_plane = dcn20_enable_power_gating_plane, .dpp_pg_control = dcn20_dpp_pg_control, .hubp_pg_control = dcn20_hubp_pg_control, - .dsc_pg_control = NULL, .update_odm = dcn20_update_odm, .dsc_pg_control = dcn20_dsc_pg_control, .get_surface_visual_confirm_color = dcn10_get_surface_visual_confirm_color, diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_opp.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_opp.c index 023cc71fad0f..138321e151eb 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_opp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_opp.c @@ -45,7 +45,8 @@ void opp2_set_disp_pattern_generator( enum dc_color_depth color_depth, const struct tg_color *solid_color, int width, - int height) + int height, + int offset) { struct dcn20_opp *oppn20 = TO_DCN20_OPP(opp); enum test_pattern_color_format bit_depth; @@ -92,6 +93,11 @@ void opp2_set_disp_pattern_generator( DPG_ACTIVE_WIDTH, width, DPG_ACTIVE_HEIGHT, height); + /* set DPG offset */ + REG_SET_2(DPG_OFFSET_SEGMENT, 0, + DPG_X_OFFSET, offset, + DPG_SEGMENT_WIDTH, 0); + switch (test_pattern) { case CONTROLLER_DP_TEST_PATTERN_COLORSQUARES: case CONTROLLER_DP_TEST_PATTERN_COLORSQUARES_CEA: diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_opp.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_opp.h index 4093bec172c1..64c5b429c79a 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_opp.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_opp.h @@ -36,6 +36,7 @@ #define OPP_DPG_REG_LIST(id) \ SRI(DPG_CONTROL, DPG, id), \ SRI(DPG_DIMENSIONS, DPG, id), \ + SRI(DPG_OFFSET_SEGMENT, DPG, id), \ SRI(DPG_COLOUR_B_CB, DPG, id), \ SRI(DPG_COLOUR_G_Y, DPG, id), \ SRI(DPG_COLOUR_R_CR, DPG, id), \ @@ -53,6 +54,7 @@ uint32_t FMT_422_CONTROL; \ uint32_t DPG_CONTROL; \ uint32_t DPG_DIMENSIONS; \ + uint32_t DPG_OFFSET_SEGMENT; \ uint32_t DPG_COLOUR_B_CB; \ uint32_t DPG_COLOUR_G_Y; \ uint32_t DPG_COLOUR_R_CR; \ @@ -68,6 +70,8 @@ OPP_SF(DPG0_DPG_CONTROL, DPG_HRES, mask_sh), \ OPP_SF(DPG0_DPG_DIMENSIONS, DPG_ACTIVE_WIDTH, mask_sh), \ OPP_SF(DPG0_DPG_DIMENSIONS, DPG_ACTIVE_HEIGHT, mask_sh), \ + OPP_SF(DPG0_DPG_OFFSET_SEGMENT, DPG_X_OFFSET, mask_sh), \ + OPP_SF(DPG0_DPG_OFFSET_SEGMENT, DPG_SEGMENT_WIDTH, mask_sh), \ OPP_SF(DPG0_DPG_COLOUR_R_CR, DPG_COLOUR0_R_CR, mask_sh), \ OPP_SF(DPG0_DPG_COLOUR_R_CR, DPG_COLOUR1_R_CR, mask_sh), \ OPP_SF(DPG0_DPG_COLOUR_B_CB, DPG_COLOUR0_B_CB, mask_sh), \ @@ -97,6 +101,8 @@ type DPG_HRES; \ type DPG_ACTIVE_WIDTH; \ type DPG_ACTIVE_HEIGHT; \ + type DPG_X_OFFSET; \ + type DPG_SEGMENT_WIDTH; \ type DPG_COLOUR0_R_CR; \ type DPG_COLOUR1_R_CR; \ type DPG_COLOUR0_B_CB; \ @@ -144,7 +150,8 @@ void opp2_set_disp_pattern_generator( enum dc_color_depth color_depth, const struct tg_color *solid_color, int width, - int height); + int height, + int offset); bool opp2_dpg_is_blanked(struct output_pixel_processor *opp); diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c index 78971b6b195c..a67395208991 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c @@ -153,6 +153,7 @@ struct _vcs_dpi_ip_params_st dcn2_0_ip = { .xfc_supported = true, .xfc_fill_bw_overhead_percent = 10.0, .xfc_fill_constant_bytes = 0, + .number_of_cursors = 1, }; struct _vcs_dpi_ip_params_st dcn2_0_nv14_ip = { @@ -220,7 +221,8 @@ struct _vcs_dpi_ip_params_st dcn2_0_nv14_ip = { .xfc_supported = true, .xfc_fill_bw_overhead_percent = 10.0, .xfc_fill_constant_bytes = 0, - .ptoi_supported = 0 + .ptoi_supported = 0, + .number_of_cursors = 1, }; struct _vcs_dpi_soc_bounding_box_st dcn2_0_soc = { @@ -335,6 +337,117 @@ struct _vcs_dpi_soc_bounding_box_st dcn2_0_soc = { .use_urgent_burst_bw = 0 }; +struct _vcs_dpi_soc_bounding_box_st dcn2_0_nv14_soc = { + .clock_limits = { + { + .state = 0, + .dcfclk_mhz = 560.0, + .fabricclk_mhz = 560.0, + .dispclk_mhz = 513.0, + .dppclk_mhz = 513.0, + .phyclk_mhz = 540.0, + .socclk_mhz = 560.0, + .dscclk_mhz = 171.0, + .dram_speed_mts = 8960.0, + }, + { + .state = 1, + .dcfclk_mhz = 694.0, + .fabricclk_mhz = 694.0, + .dispclk_mhz = 642.0, + .dppclk_mhz = 642.0, + .phyclk_mhz = 600.0, + .socclk_mhz = 694.0, + .dscclk_mhz = 214.0, + .dram_speed_mts = 11104.0, + }, + { + .state = 2, + .dcfclk_mhz = 875.0, + .fabricclk_mhz = 875.0, + .dispclk_mhz = 734.0, + .dppclk_mhz = 734.0, + .phyclk_mhz = 810.0, + .socclk_mhz = 875.0, + .dscclk_mhz = 245.0, + .dram_speed_mts = 14000.0, + }, + { + .state = 3, + .dcfclk_mhz = 1000.0, + .fabricclk_mhz = 1000.0, + .dispclk_mhz = 1100.0, + .dppclk_mhz = 1100.0, + .phyclk_mhz = 810.0, + .socclk_mhz = 1000.0, + .dscclk_mhz = 367.0, + .dram_speed_mts = 16000.0, + }, + { + .state = 4, + .dcfclk_mhz = 1200.0, + .fabricclk_mhz = 1200.0, + .dispclk_mhz = 1284.0, + .dppclk_mhz = 1284.0, + .phyclk_mhz = 810.0, + .socclk_mhz = 1200.0, + .dscclk_mhz = 428.0, + .dram_speed_mts = 16000.0, + }, + /*Extra state, no dispclk ramping*/ + { + .state = 5, + .dcfclk_mhz = 1200.0, + .fabricclk_mhz = 1200.0, + .dispclk_mhz = 1284.0, + .dppclk_mhz = 1284.0, + .phyclk_mhz = 810.0, + .socclk_mhz = 1200.0, + .dscclk_mhz = 428.0, + .dram_speed_mts = 16000.0, + }, + }, + .num_states = 5, + .sr_exit_time_us = 8.6, + .sr_enter_plus_exit_time_us = 10.9, + .urgent_latency_us = 4.0, + .urgent_latency_pixel_data_only_us = 4.0, + .urgent_latency_pixel_mixed_with_vm_data_us = 4.0, + .urgent_latency_vm_data_only_us = 4.0, + .urgent_out_of_order_return_per_channel_pixel_only_bytes = 4096, + .urgent_out_of_order_return_per_channel_pixel_and_vm_bytes = 4096, + .urgent_out_of_order_return_per_channel_vm_only_bytes = 4096, + .pct_ideal_dram_sdp_bw_after_urgent_pixel_only = 40.0, + .pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm = 40.0, + .pct_ideal_dram_sdp_bw_after_urgent_vm_only = 40.0, + .max_avg_sdp_bw_use_normal_percent = 40.0, + .max_avg_dram_bw_use_normal_percent = 40.0, + .writeback_latency_us = 12.0, + .ideal_dram_bw_after_urgent_percent = 40.0, + .max_request_size_bytes = 256, + .dram_channel_width_bytes = 2, + .fabric_datapath_to_dcn_data_return_bytes = 64, + .dcn_downspread_percent = 0.5, + .downspread_percent = 0.38, + .dram_page_open_time_ns = 50.0, + .dram_rw_turnaround_time_ns = 17.5, + .dram_return_buffer_per_channel_bytes = 8192, + .round_trip_ping_latency_dcfclk_cycles = 131, + .urgent_out_of_order_return_per_channel_bytes = 256, + .channel_interleave_bytes = 256, + .num_banks = 8, + .num_chans = 8, + .vmm_page_size_bytes = 4096, + .dram_clock_change_latency_us = 404.0, + .dummy_pstate_latency_us = 5.0, + .writeback_dram_clock_change_latency_us = 23.0, + .return_bus_width_bytes = 64, + .dispclk_dppclk_vco_speed_mhz = 3850, + .xfc_bus_transport_time_us = 20, + .xfc_xbuf_latency_tolerance_us = 4, + .use_urgent_burst_bw = 0 +}; + struct _vcs_dpi_soc_bounding_box_st dcn2_0_nv12_soc = { 0 }; #ifndef mmDP0_DP_DPHY_INTERNAL_CTRL @@ -928,7 +1041,7 @@ static const struct resource_caps res_cap_nv14 = { }; static const struct dc_debug_options debug_defaults_drv = { - .disable_dmcu = true, + .disable_dmcu = false, .force_abm_enable = false, .timing_trace = false, .clock_trace = true, @@ -947,7 +1060,7 @@ static const struct dc_debug_options debug_defaults_drv = { }; static const struct dc_debug_options debug_defaults_diags = { - .disable_dmcu = true, + .disable_dmcu = false, .force_abm_enable = false, .timing_trace = true, .clock_trace = true, @@ -1143,6 +1256,7 @@ static const struct encoder_feature_support link_enc_feature = { .max_hdmi_pixel_clock = 600000, .hdmi_ycbcr420_supported = true, .dp_ycbcr420_supported = true, + .fec_supported = true, .flags.bits.IS_HBR2_CAPABLE = true, .flags.bits.IS_HBR3_CAPABLE = true, .flags.bits.IS_TPS3_CAPABLE = true, @@ -1557,7 +1671,7 @@ static void acquire_dsc(struct resource_context *res_ctx, } } -static void release_dsc(struct resource_context *res_ctx, +void dcn20_release_dsc(struct resource_context *res_ctx, const struct resource_pool *pool, struct display_stream_compressor **dsc) { @@ -1617,7 +1731,7 @@ static enum dc_status remove_dsc_from_stream_resource(struct dc *dc, pipe_ctx = &new_ctx->res_ctx.pipe_ctx[i]; if (pipe_ctx->stream_res.dsc) - release_dsc(&new_ctx->res_ctx, dc->res_pool, &pipe_ctx->stream_res.dsc); + dcn20_release_dsc(&new_ctx->res_ctx, dc->res_pool, &pipe_ctx->stream_res.dsc); } } @@ -2041,14 +2155,17 @@ int dcn20_populate_dml_pipes_from_context( /* todo: default max for now, until there is logic reflecting this in dc*/ pipes[pipe_cnt].dout.output_bpc = 12; /* - * Use max cursor settings for calculations to minimize + * For graphic plane, cursor number is 1, nv12 is 0 * bw calculations due to cursor on/off */ - pipes[pipe_cnt].pipe.src.num_cursors = 2; + if (res_ctx->pipe_ctx[i].plane_state && + res_ctx->pipe_ctx[i].plane_state->address.type == PLN_ADDR_TYPE_VIDEO_PROGRESSIVE) + pipes[pipe_cnt].pipe.src.num_cursors = 0; + else + pipes[pipe_cnt].pipe.src.num_cursors = dc->dml.ip.number_of_cursors; + pipes[pipe_cnt].pipe.src.cur0_src_width = 256; pipes[pipe_cnt].pipe.src.cur0_bpp = dm_cur_32bit; - pipes[pipe_cnt].pipe.src.cur1_src_width = 256; - pipes[pipe_cnt].pipe.src.cur1_bpp = dm_cur_32bit; if (!res_ctx->pipe_ctx[i].plane_state) { pipes[pipe_cnt].pipe.src.is_hsplit = pipes[pipe_cnt].pipe.dest.odm_combine != dm_odm_combine_mode_disabled; @@ -2298,6 +2415,7 @@ bool dcn20_validate_dsc(struct dc *dc, struct dc_state *new_ctx) + stream->timing.v_border_bottom; dsc_cfg.pixel_encoding = stream->timing.pixel_encoding; dsc_cfg.color_depth = stream->timing.display_color_depth; + dsc_cfg.is_odm = pipe_ctx->next_odm_pipe ? true : false; dsc_cfg.dc_dsc_cfg = stream->timing.dsc_cfg; dsc_cfg.dc_dsc_cfg.num_slices_h /= opp_cnt; @@ -2384,7 +2502,7 @@ struct pipe_ctx *dcn20_find_secondary_pipe(struct dc *dc, return secondary_pipe; } -void dcn20_merge_pipes_for_validate( +static void dcn20_merge_pipes_for_validate( struct dc *dc, struct dc_state *context) { @@ -2409,7 +2527,7 @@ void dcn20_merge_pipes_for_validate( odm_pipe->prev_odm_pipe = NULL; odm_pipe->next_odm_pipe = NULL; if (odm_pipe->stream_res.dsc) - release_dsc(&context->res_ctx, dc->res_pool, &odm_pipe->stream_res.dsc); + dcn20_release_dsc(&context->res_ctx, dc->res_pool, &odm_pipe->stream_res.dsc); /* Clear plane_res and stream_res */ memset(&odm_pipe->plane_res, 0, sizeof(odm_pipe->plane_res)); memset(&odm_pipe->stream_res, 0, sizeof(odm_pipe->stream_res)); @@ -2447,41 +2565,29 @@ int dcn20_validate_apply_pipe_split_flags( struct dc *dc, struct dc_state *context, int vlevel, - bool *split) + bool *split, + bool *merge) { int i, pipe_idx, vlevel_split; + int plane_count = 0; bool force_split = false; - bool avoid_split = dc->debug.pipe_split_policy != MPC_SPLIT_DYNAMIC; + bool avoid_split = dc->debug.pipe_split_policy == MPC_SPLIT_AVOID; - /* Single display loop, exits if there is more than one display */ + if (context->stream_count > 1) { + if (dc->debug.pipe_split_policy == MPC_SPLIT_AVOID_MULT_DISP) + avoid_split = true; + } else if (dc->debug.force_single_disp_pipe_split) + force_split = true; + + /* TODO: fix dc bugs and remove this split threshold thing */ for (i = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; - bool exit_loop = false; - - if (!pipe->stream || pipe->top_pipe) - continue; - if (dc->debug.force_single_disp_pipe_split) { - if (!force_split) - force_split = true; - else { - force_split = false; - exit_loop = true; - } - } - if (dc->debug.pipe_split_policy == MPC_SPLIT_AVOID_MULT_DISP) { - if (avoid_split) - avoid_split = false; - else { - avoid_split = true; - exit_loop = true; - } - } - if (exit_loop) - break; + if (pipe->stream && !pipe->prev_odm_pipe && + (!pipe->top_pipe || pipe->top_pipe->plane_state != pipe->plane_state)) + ++plane_count; } - /* TODO: fix dc bugs and remove this split threshold thing */ - if (context->stream_count > dc->res_pool->pipe_count / 2) + if (plane_count > dc->res_pool->pipe_count / 2) avoid_split = true; /* Avoid split loop looks for lowest voltage level that allows most unsplit pipes possible */ @@ -2504,11 +2610,12 @@ int dcn20_validate_apply_pipe_split_flags( /* Split loop sets which pipe should be split based on dml outputs and dc flags */ for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; + int pipe_plane = context->bw_ctx.dml.vba.pipe_plane[pipe_idx]; if (!context->res_ctx.pipe_ctx[i].stream) continue; - if (force_split || context->bw_ctx.dml.vba.NoOfDPP[vlevel][context->bw_ctx.dml.vba.maxMpcComb][pipe_idx] > 1) + if (force_split || context->bw_ctx.dml.vba.NoOfDPP[vlevel][context->bw_ctx.dml.vba.maxMpcComb][pipe_plane] > 1) split[i] = true; if ((pipe->stream->view_format == VIEW_3D_FORMAT_SIDE_BY_SIDE || @@ -2521,10 +2628,44 @@ int dcn20_validate_apply_pipe_split_flags( split[i] = true; if (dc->debug.force_odm_combine & (1 << pipe->stream_res.tg->inst)) { split[i] = true; - context->bw_ctx.dml.vba.ODMCombineEnablePerState[vlevel][pipe_idx] = dm_odm_combine_mode_2to1; + context->bw_ctx.dml.vba.ODMCombineEnablePerState[vlevel][pipe_plane] = dm_odm_combine_mode_2to1; } - context->bw_ctx.dml.vba.ODMCombineEnabled[pipe_idx] = - context->bw_ctx.dml.vba.ODMCombineEnablePerState[vlevel][pipe_idx]; + context->bw_ctx.dml.vba.ODMCombineEnabled[pipe_plane] = + context->bw_ctx.dml.vba.ODMCombineEnablePerState[vlevel][pipe_plane]; + + if (pipe->prev_odm_pipe && context->bw_ctx.dml.vba.ODMCombineEnabled[pipe_plane] != dm_odm_combine_mode_disabled) { + /*Already split odm pipe tree, don't try to split again*/ + split[i] = false; + split[pipe->prev_odm_pipe->pipe_idx] = false; + } else if (pipe->top_pipe && pipe->plane_state == pipe->top_pipe->plane_state + && context->bw_ctx.dml.vba.ODMCombineEnabled[pipe_plane] == dm_odm_combine_mode_disabled) { + /*Already split mpc tree, don't try to split again, assumes only 2x mpc combine*/ + split[i] = false; + split[pipe->top_pipe->pipe_idx] = false; + } else if (pipe->prev_odm_pipe || (pipe->top_pipe && pipe->plane_state == pipe->top_pipe->plane_state)) { + if (split[i] == false) { + /*Exiting mpc/odm combine*/ + merge[i] = true; + if (pipe->prev_odm_pipe) { + ASSERT(0); /*should not actually happen yet*/ + merge[pipe->prev_odm_pipe->pipe_idx] = true; + } else + merge[pipe->top_pipe->pipe_idx] = true; + } else { + /*Transition from mpc combine to odm combine or vice versa*/ + ASSERT(0); /*should not actually happen yet*/ + split[i] = true; + merge[i] = true; + if (pipe->prev_odm_pipe) { + split[pipe->prev_odm_pipe->pipe_idx] = true; + merge[pipe->prev_odm_pipe->pipe_idx] = true; + } else { + split[pipe->top_pipe->pipe_idx] = true; + merge[pipe->top_pipe->pipe_idx] = true; + } + } + } + /* Adjust dppclk when split is forced, do not bother with dispclk */ if (split[i] && context->bw_ctx.dml.vba.NoOfDPP[vlevel][context->bw_ctx.dml.vba.maxMpcComb][pipe_idx] == 1) context->bw_ctx.dml.vba.RequiredDPPCLK[vlevel][context->bw_ctx.dml.vba.maxMpcComb][pipe_idx] /= 2; @@ -2566,7 +2707,7 @@ bool dcn20_fast_validate_bw( if (vlevel > context->bw_ctx.dml.soc.num_states) goto validate_fail; - vlevel = dcn20_validate_apply_pipe_split_flags(dc, context, vlevel, split); + vlevel = dcn20_validate_apply_pipe_split_flags(dc, context, vlevel, split, NULL); /*initialize pipe_just_split_from to invalid idx*/ for (i = 0; i < MAX_PIPES; i++) @@ -3026,7 +3167,7 @@ static struct dc_cap_funcs cap_funcs = { }; -enum dc_status dcn20_get_default_swizzle_mode(struct dc_plane_state *plane_state) +enum dc_status dcn20_patch_unknown_plane_state(struct dc_plane_state *plane_state) { enum dc_status result = DC_OK; @@ -3052,7 +3193,7 @@ static struct resource_funcs dcn20_res_pool_funcs = { .add_stream_to_ctx = dcn20_add_stream_to_ctx, .remove_stream_from_ctx = dcn20_remove_stream_from_ctx, .populate_dml_writeback_from_context = dcn20_populate_dml_writeback_from_context, - .get_default_swizzle_mode = dcn20_get_default_swizzle_mode, + .patch_unknown_plane_state = dcn20_patch_unknown_plane_state, .set_mcif_arb_params = dcn20_set_mcif_arb_params, .populate_dml_pipes = dcn20_populate_dml_pipes_from_context, .find_first_free_match_stream_enc_for_link = dcn10_find_first_free_match_stream_enc_for_link @@ -3290,6 +3431,9 @@ void dcn20_patch_bounding_box(struct dc *dc, struct _vcs_dpi_soc_bounding_box_st static struct _vcs_dpi_soc_bounding_box_st *get_asic_rev_soc_bb( uint32_t hw_internal_rev) { + if (ASICREV_IS_NAVI14_M(hw_internal_rev)) + return &dcn2_0_nv14_soc; + if (ASICREV_IS_NAVI12_P(hw_internal_rev)) return &dcn2_0_nv12_soc; diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.h index f5893840b79b..9d5bff9455fd 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.h @@ -119,14 +119,15 @@ void dcn20_set_mcif_arb_params( display_e2e_pipe_params_st *pipes, int pipe_cnt); bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, bool fast_validate); -void dcn20_merge_pipes_for_validate( - struct dc *dc, - struct dc_state *context); int dcn20_validate_apply_pipe_split_flags( struct dc *dc, struct dc_state *context, int vlevel, - bool *split); + bool *split, + bool *merge); +void dcn20_release_dsc(struct resource_context *res_ctx, + const struct resource_pool *pool, + struct display_stream_compressor **dsc); bool dcn20_validate_dsc(struct dc *dc, struct dc_state *new_ctx); void dcn20_split_stream_for_mpc( struct resource_context *res_ctx, @@ -159,7 +160,7 @@ enum dc_status dcn20_build_mapped_resource(const struct dc *dc, struct dc_state enum dc_status dcn20_add_stream_to_ctx(struct dc *dc, struct dc_state *new_ctx, struct dc_stream_state *dc_stream); enum dc_status dcn20_add_dsc_to_stream_resource(struct dc *dc, struct dc_state *dc_ctx, struct dc_stream_state *dc_stream); enum dc_status dcn20_remove_stream_from_ctx(struct dc *dc, struct dc_state *new_ctx, struct dc_stream_state *dc_stream); -enum dc_status dcn20_get_default_swizzle_mode(struct dc_plane_state *plane_state); +enum dc_status dcn20_patch_unknown_plane_state(struct dc_plane_state *plane_state); void dcn20_patch_bounding_box( struct dc *dc, diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hwseq.c index 081ad8e43d58..ada65b1a7eb1 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hwseq.c @@ -112,3 +112,25 @@ void dcn21_optimize_pwr_state( true); } +/* If user hotplug a HDMI monitor while in monitor off, + * OS will do a mode set (with output timing) but keep output off. + * In this case DAL will ask vbios to power up the pll in the PHY. + * If user unplug the monitor (while we are on monitor off) or + * system attempt to enter modern standby (which we will disable PLL), + * PHY will hang on the next mode set attempt. + * if enable PLL follow by disable PLL (without executing lane enable/disable), + * RDPCS_PHY_DP_MPLLB_STATE remains 1, + * which indicate that PLL disable attempt actually didn’t go through. + * As a workaround, insert PHY lane enable/disable before PLL disable. + */ +void dcn21_PLAT_58856_wa(struct dc_state *context, struct pipe_ctx *pipe_ctx) +{ + if (!pipe_ctx->stream->dpms_off) + return; + + pipe_ctx->stream->dpms_off = false; + core_link_enable_stream(context, pipe_ctx); + core_link_disable_stream(pipe_ctx); + pipe_ctx->stream->dpms_off = true; +} + diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hwseq.h b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hwseq.h index 182736096123..26bf24d3b59f 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hwseq.h @@ -44,4 +44,7 @@ void dcn21_optimize_pwr_state( const struct dc *dc, struct dc_state *context); +void dcn21_PLAT_58856_wa(struct dc_state *context, + struct pipe_ctx *pipe_ctx); + #endif /* __DC_HWSS_DCN21_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.c index f9a7e43d66b9..b9ff9767e08f 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.c +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.c @@ -119,7 +119,6 @@ static const struct hwseq_private_funcs dcn21_private_funcs = { .enable_power_gating_plane = dcn20_enable_power_gating_plane, .dpp_pg_control = dcn20_dpp_pg_control, .hubp_pg_control = dcn20_hubp_pg_control, - .dsc_pg_control = NULL, .update_odm = dcn20_update_odm, .dsc_pg_control = dcn20_dsc_pg_control, .get_surface_visual_confirm_color = dcn10_get_surface_visual_confirm_color, @@ -131,6 +130,7 @@ static const struct hwseq_private_funcs dcn21_private_funcs = { .dccg_init = dcn20_dccg_init, .set_blend_lut = dcn20_set_blend_lut, .set_shaper_3dlut = dcn20_set_shaper_3dlut, + .PLAT_58856_wa = dcn21_PLAT_58856_wa, }; void dcn21_hw_sequencer_construct(struct dc *dc) diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c index dce4966eca20..51b5910cd05f 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c @@ -156,7 +156,8 @@ struct _vcs_dpi_ip_params_st dcn2_1_ip = { .xfc_supported = false, .xfc_fill_bw_overhead_percent = 10.0, .xfc_fill_constant_bytes = 0, - .ptoi_supported = 0 + .ptoi_supported = 0, + .number_of_cursors = 1, }; struct _vcs_dpi_soc_bounding_box_st dcn2_1_soc = { @@ -854,7 +855,7 @@ static const struct dc_plane_cap plane_cap = { }; static const struct dc_debug_options debug_defaults_drv = { - .disable_dmcu = true, + .disable_dmcu = false, .force_abm_enable = false, .timing_trace = false, .clock_trace = true, @@ -875,7 +876,7 @@ static const struct dc_debug_options debug_defaults_drv = { }; static const struct dc_debug_options debug_defaults_diags = { - .disable_dmcu = true, + .disable_dmcu = false, .force_abm_enable = false, .timing_trace = true, .clock_trace = true, @@ -1589,6 +1590,7 @@ static const struct encoder_feature_support link_enc_feature = { .max_hdmi_pixel_clock = 600000, .hdmi_ycbcr420_supported = true, .dp_ycbcr420_supported = true, + .fec_supported = true, .flags.bits.IS_HBR2_CAPABLE = true, .flags.bits.IS_HBR3_CAPABLE = true, .flags.bits.IS_TPS3_CAPABLE = true, @@ -1729,6 +1731,19 @@ static int dcn21_populate_dml_pipes_from_context( return pipe_cnt; } +enum dc_status dcn21_patch_unknown_plane_state(struct dc_plane_state *plane_state) +{ + enum dc_status result = DC_OK; + + if (plane_state->ctx->dc->debug.disable_dcc == DCC_ENABLE) { + plane_state->dcc.enable = 1; + /* align to our worst case block width */ + plane_state->dcc.meta_pitch = ((plane_state->src_rect.width + 1023) / 1024) * 1024; + } + result = dcn20_patch_unknown_plane_state(plane_state); + return result; +} + static struct resource_funcs dcn21_res_pool_funcs = { .destroy = dcn21_destroy_resource_pool, .link_enc_create = dcn21_link_encoder_create, @@ -1738,7 +1753,7 @@ static struct resource_funcs dcn21_res_pool_funcs = { .remove_stream_from_ctx = dcn20_remove_stream_from_ctx, .acquire_idle_pipe_for_layer = dcn20_acquire_idle_pipe_for_layer, .populate_dml_writeback_from_context = dcn20_populate_dml_writeback_from_context, - .get_default_swizzle_mode = dcn20_get_default_swizzle_mode, + .patch_unknown_plane_state = dcn21_patch_unknown_plane_state, .set_mcif_arb_params = dcn20_set_mcif_arb_params, .find_first_free_match_stream_enc_for_link = dcn10_find_first_free_match_stream_enc_for_link, .update_bw_bounding_box = update_bw_bounding_box @@ -1785,6 +1800,7 @@ static bool dcn21_resource_construct( dc->caps.force_dp_tps4_for_cp2520 = true; dc->caps.extended_aux_timeout_support = true; dc->caps.dmcub_support = true; + dc->caps.is_apu = true; if (dc->ctx->dce_environment == DCE_ENV_PRODUCTION_DRV) dc->debug = debug_defaults_drv; @@ -1848,7 +1864,7 @@ static bool dcn21_resource_construct( goto create_fail; } - if (dc->debug.psr_on_dmub) { + if (dc->debug.disable_dmcu) { pool->base.psr = dmub_psr_create(ctx); if (pool->base.psr == NULL) { diff --git a/drivers/gpu/drm/amd/display/dc/dm_cp_psp.h b/drivers/gpu/drm/amd/display/dc/dm_cp_psp.h index 626d22d437f4..968c46dfb506 100644 --- a/drivers/gpu/drm/amd/display/dc/dm_cp_psp.h +++ b/drivers/gpu/drm/amd/display/dc/dm_cp_psp.h @@ -32,6 +32,7 @@ struct cp_psp_stream_config { uint8_t otg_inst; uint8_t link_enc_inst; uint8_t stream_enc_inst; + uint8_t mst_supported; void *dm_stream_ctx; bool dpms_off; }; diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h b/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h index a56b611db15e..dfd3be452766 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h +++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h @@ -69,6 +69,7 @@ struct _vcs_dpi_voltage_scaling_st { struct _vcs_dpi_soc_bounding_box_st { struct _vcs_dpi_voltage_scaling_st clock_limits[MAX_CLOCK_LIMIT_STATES]; + unsigned int num_states; double sr_exit_time_us; double sr_enter_plus_exit_time_us; double urgent_latency_us; @@ -111,7 +112,6 @@ struct _vcs_dpi_soc_bounding_box_st { double xfc_bus_transport_time_us; double xfc_xbuf_latency_tolerance_us; int use_urgent_burst_bw; - unsigned int num_states; double min_dcfclk; bool do_urgent_latency_adjustment; double urgent_latency_adjustment_fabric_clock_component_us; @@ -204,6 +204,7 @@ struct _vcs_dpi_ip_params_st { unsigned int LineBufferFixedBpp; unsigned int can_vstartup_lines_exceed_vsync_plus_back_porch_lines_minus_one; unsigned int bug_forcing_LC_req_same_size_fixed; + unsigned int number_of_cursors; }; struct _vcs_dpi_display_xfc_params_st { diff --git a/drivers/gpu/drm/amd/display/dc/inc/core_types.h b/drivers/gpu/drm/amd/display/dc/inc/core_types.h index f285b76888fb..d523fc9547e7 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/core_types.h +++ b/drivers/gpu/drm/amd/display/dc/inc/core_types.h @@ -124,7 +124,7 @@ struct resource_funcs { struct dc *dc, struct dc_state *new_ctx, struct dc_stream_state *stream); - enum dc_status (*get_default_swizzle_mode)( + enum dc_status (*patch_unknown_plane_state)( struct dc_plane_state *plane_state); struct stream_encoder *(*find_first_free_match_stream_enc_for_link)( diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h index 05ee5295d2c1..336c80a18175 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h @@ -27,11 +27,12 @@ #define __DAL_DCCG_H__ #include "dc_types.h" +#include "hw_shared.h" struct dccg { struct dc_context *ctx; const struct dccg_funcs *funcs; - + int pipe_dppclk_khz[MAX_PIPES]; int ref_dppclk; }; diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dsc.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dsc.h index c59740084ebc..7c2a3328b208 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/dsc.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dsc.h @@ -39,6 +39,7 @@ struct dsc_config { uint32_t pic_height; enum dc_pixel_encoding pixel_encoding; enum dc_color_depth color_depth; /* Bits per component */ + bool is_odm; struct dc_dsc_config dc_dsc_cfg; }; diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/link_encoder.h b/drivers/gpu/drm/amd/display/dc/inc/hw/link_encoder.h index fb748f082c56..c2b392a533b1 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/link_encoder.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/link_encoder.h @@ -68,6 +68,7 @@ struct encoder_feature_support { unsigned int max_hdmi_pixel_clock; bool hdmi_ycbcr420_supported; bool dp_ycbcr420_supported; + bool fec_supported; }; union dpcd_psr_configuration { diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/opp.h b/drivers/gpu/drm/amd/display/dc/inc/hw/opp.h index 7575564b2265..2717352eb697 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/opp.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/opp.h @@ -310,7 +310,8 @@ struct opp_funcs { enum dc_color_depth color_depth, const struct tg_color *solid_color, int width, - int height); + int height, + int offset); bool (*dpg_is_blanked)( struct output_pixel_processor *opp); diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer_private.h b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer_private.h index b1d736cbcd5a..52a26e6be066 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer_private.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer_private.h @@ -145,6 +145,8 @@ struct hwseq_private_funcs { const struct dc_plane_state *plane_state); bool (*set_shaper_3dlut)(struct pipe_ctx *pipe_ctx, const struct dc_plane_state *plane_state); + void (*PLAT_58856_wa)(struct dc_state *context, + struct pipe_ctx *pipe_ctx); }; struct dce_hwseq { diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h index 0cb8967f0c45..10b5fa9d2588 100644 --- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h +++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h @@ -50,6 +50,7 @@ enum dmub_cmd_type { DMUB_CMD__REG_REG_WAIT = 4, DMUB_CMD__PLAT_54186_WA = 5, DMUB_CMD__PSR = 64, + DMUB_CMD__ABM = 66, DMUB_CMD__VBIOS = 128, }; @@ -256,6 +257,52 @@ struct dmub_rb_cmd_psr_set_version { struct dmub_cmd_psr_set_version_data psr_set_version_data; }; +struct dmub_cmd_abm_set_pipe_data { + uint32_t ramping_boundary; + uint32_t otg_inst; +}; + +struct dmub_rb_cmd_abm_set_pipe { + struct dmub_cmd_header header; + struct dmub_cmd_abm_set_pipe_data abm_set_pipe_data; +}; + +struct dmub_cmd_abm_set_backlight_data { + uint32_t frame_ramp; +}; + +struct dmub_rb_cmd_abm_set_backlight { + struct dmub_cmd_header header; + struct dmub_cmd_abm_set_backlight_data abm_set_backlight_data; +}; + +struct dmub_cmd_abm_set_level_data { + uint32_t level; +}; + +struct dmub_rb_cmd_abm_set_level { + struct dmub_cmd_header header; + struct dmub_cmd_abm_set_level_data abm_set_level_data; +}; + +struct dmub_cmd_abm_set_ambient_level_data { + uint32_t ambient_lux; +}; + +struct dmub_rb_cmd_abm_set_ambient_level { + struct dmub_cmd_header header; + struct dmub_cmd_abm_set_ambient_level_data abm_set_ambient_level_data; +}; + +struct dmub_cmd_abm_set_pwm_frac_data { + uint32_t fractional_pwm; +}; + +struct dmub_rb_cmd_abm_set_pwm_frac { + struct dmub_cmd_header header; + struct dmub_cmd_abm_set_pwm_frac_data abm_set_pwm_frac_data; +}; + union dmub_rb_cmd { struct dmub_rb_cmd_read_modify_write read_modify_write; struct dmub_rb_cmd_reg_field_update_sequence reg_field_update_seq; @@ -272,6 +319,11 @@ union dmub_rb_cmd { struct dmub_rb_cmd_psr_enable psr_enable; struct dmub_rb_cmd_psr_set_level psr_set_level; struct dmub_rb_cmd_PLAT_54186_wa PLAT_54186_wa; + struct dmub_rb_cmd_abm_set_pipe abm_set_pipe; + struct dmub_rb_cmd_abm_set_backlight abm_set_backlight; + struct dmub_rb_cmd_abm_set_level abm_set_level; + struct dmub_rb_cmd_abm_set_ambient_level abm_set_ambient_level; + struct dmub_rb_cmd_abm_set_pwm_frac abm_set_pwm_frac; }; #pragma pack(pop) diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd_dal.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd_dal.h index ce793f47f234..d37535d21928 100644 --- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd_dal.h +++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd_dal.h @@ -45,4 +45,13 @@ enum psr_version { PSR_VERSION_2_1 = 0x21, // PSR Version 2, includes Y-coordinate support for SU }; +enum dmub_cmd_abm_type { + DMUB_CMD__ABM_INIT_CONFIG = 0, + DMUB_CMD__ABM_SET_PIPE = 1, + DMUB_CMD__ABM_SET_BACKLIGHT = 2, + DMUB_CMD__ABM_SET_LEVEL = 3, + DMUB_CMD__ABM_SET_AMBIENT_LEVEL = 4, + DMUB_CMD__ABM_SET_PWM_FRAC = 5, +}; + #endif /* _DMUB_CMD_DAL_H_ */ diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_srv.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_srv.h index e619fa9cf53a..c2671f2616c8 100644 --- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_srv.h +++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_srv.h @@ -104,7 +104,7 @@ enum dmub_window_id { DMUB_WINDOW_4_MAILBOX, DMUB_WINDOW_5_TRACEBUFF, DMUB_WINDOW_6_FW_STATE, - DMUB_WINDOW_7_RESERVED, + DMUB_WINDOW_7_SCRATCH_MEM, DMUB_WINDOW_TOTAL, }; @@ -316,6 +316,7 @@ struct dmub_srv { enum dmub_asic asic; void *user_ctx; bool is_virtual; + struct dmub_fb scratch_mem_fb; volatile const struct dmub_fw_state *fw_state; /* private: internal use only */ diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c index 45be185ef312..ce32cc7933c4 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c @@ -52,8 +52,11 @@ /* Default tracebuffer size if meta is absent. */ #define DMUB_TRACE_BUFFER_SIZE (1024) +/* Default scratch mem size. */ +#define DMUB_SCRATCH_MEM_SIZE (256) + /* Number of windows in use. */ -#define DMUB_NUM_WINDOWS (DMUB_WINDOW_6_FW_STATE + 1) +#define DMUB_NUM_WINDOWS (DMUB_WINDOW_TOTAL) /* Base addresses. */ #define DMUB_CW0_BASE (0x60000000) @@ -211,9 +214,11 @@ dmub_srv_calc_region_info(struct dmub_srv *dmub, struct dmub_region *mail = &out->regions[DMUB_WINDOW_4_MAILBOX]; struct dmub_region *trace_buff = &out->regions[DMUB_WINDOW_5_TRACEBUFF]; struct dmub_region *fw_state = &out->regions[DMUB_WINDOW_6_FW_STATE]; + struct dmub_region *scratch_mem = &out->regions[DMUB_WINDOW_7_SCRATCH_MEM]; const struct dmub_fw_meta_info *fw_info; uint32_t fw_state_size = DMUB_FW_STATE_SIZE; uint32_t trace_buffer_size = DMUB_TRACE_BUFFER_SIZE; + uint32_t scratch_mem_size = DMUB_SCRATCH_MEM_SIZE; if (!dmub->sw_init) return DMUB_STATUS_INVALID; @@ -256,7 +261,10 @@ dmub_srv_calc_region_info(struct dmub_srv *dmub, fw_state->base = dmub_align(trace_buff->top, 256); fw_state->top = fw_state->base + dmub_align(fw_state_size, 64); - out->fb_size = dmub_align(fw_state->top, 4096); + scratch_mem->base = dmub_align(fw_state->top, 256); + scratch_mem->top = scratch_mem->base + dmub_align(scratch_mem_size, 64); + + out->fb_size = dmub_align(scratch_mem->top, 4096); return DMUB_STATUS_OK; } @@ -334,6 +342,7 @@ enum dmub_status dmub_srv_hw_init(struct dmub_srv *dmub, struct dmub_fb *mail_fb = params->fb[DMUB_WINDOW_4_MAILBOX]; struct dmub_fb *tracebuff_fb = params->fb[DMUB_WINDOW_5_TRACEBUFF]; struct dmub_fb *fw_state_fb = params->fb[DMUB_WINDOW_6_FW_STATE]; + struct dmub_fb *scratch_mem_fb = params->fb[DMUB_WINDOW_7_SCRATCH_MEM]; struct dmub_rb_init_params rb_params; struct dmub_window cw0, cw1, cw2, cw3, cw4, cw5, cw6; @@ -370,7 +379,7 @@ enum dmub_status dmub_srv_hw_init(struct dmub_srv *dmub, dmub->hw_funcs.reset(dmub); if (inst_fb && data_fb && bios_fb && mail_fb && tracebuff_fb && - fw_state_fb) { + fw_state_fb && scratch_mem_fb) { cw2.offset.quad_part = data_fb->gpu_addr; cw2.region.base = DMUB_CW0_BASE + inst_fb->size; cw2.region.top = cw2.region.base + data_fb->size; @@ -396,6 +405,8 @@ enum dmub_status dmub_srv_hw_init(struct dmub_srv *dmub, dmub->fw_state = fw_state_fb->cpu_addr; + dmub->scratch_mem_fb = *scratch_mem_fb; + if (dmub->hw_funcs.setup_windows) dmub->hw_funcs.setup_windows(dmub, &cw2, &cw3, &cw4, &cw5, &cw6); diff --git a/drivers/gpu/drm/amd/display/include/dal_asic_id.h b/drivers/gpu/drm/amd/display/include/dal_asic_id.h index ea7015f869c9..8a87d0ed90ae 100644 --- a/drivers/gpu/drm/amd/display/include/dal_asic_id.h +++ b/drivers/gpu/drm/amd/display/include/dal_asic_id.h @@ -134,11 +134,6 @@ #define PICASSO_A0 0x41 /* DCN1_01 */ #define RAVEN2_A0 0x81 -#define RAVEN2_15D8_REV_94 0x94 -#define RAVEN2_15D8_REV_95 0x95 -#define RAVEN2_15D8_REV_E9 0xE9 -#define RAVEN2_15D8_REV_EA 0xEA -#define RAVEN2_15D8_REV_EB 0xEB #define RAVEN1_F0 0xF0 #define RAVEN_UNKNOWN 0xFF #ifndef ASICREV_IS_RAVEN @@ -149,16 +144,17 @@ #define PRID_DALI_E3 0xE3 #define PRID_DALI_E4 0xE4 +#define PRID_POLLOCK_94 0x94 +#define PRID_POLLOCK_95 0x95 +#define PRID_POLLOCK_E9 0xE9 +#define PRID_POLLOCK_EA 0xEA +#define PRID_POLLOCK_EB 0xEB + #define ASICREV_IS_PICASSO(eChipRev) ((eChipRev >= PICASSO_A0) && (eChipRev < RAVEN2_A0)) #ifndef ASICREV_IS_RAVEN2 #define ASICREV_IS_RAVEN2(eChipRev) ((eChipRev >= RAVEN2_A0) && (eChipRev < RENOIR_A0)) #endif #define ASICREV_IS_RV1_F0(eChipRev) ((eChipRev >= RAVEN1_F0) && (eChipRev < RAVEN_UNKNOWN)) -#define ASICREV_IS_POLLOCK(eChipRev) (eChipRev == RAVEN2_15D8_REV_94 \ - || eChipRev == RAVEN2_15D8_REV_95 \ - || eChipRev == RAVEN2_15D8_REV_E9 \ - || eChipRev == RAVEN2_15D8_REV_EA \ - || eChipRev == RAVEN2_15D8_REV_EB) #define FAMILY_RV 142 /* DCN 1*/ diff --git a/drivers/gpu/drm/amd/display/include/logger_types.h b/drivers/gpu/drm/amd/display/include/logger_types.h index 89a709267019..d66f9d8eefb4 100644 --- a/drivers/gpu/drm/amd/display/include/logger_types.h +++ b/drivers/gpu/drm/amd/display/include/logger_types.h @@ -124,36 +124,37 @@ enum dc_log_type { #define DC_MIN_LOG_MASK ((1 << LOG_ERROR) | \ (1 << LOG_DETECTION_EDID_PARSER)) -#define DC_DEFAULT_LOG_MASK ((1 << LOG_ERROR) | \ - (1 << LOG_WARNING) | \ - (1 << LOG_EVENT_MODE_SET) | \ - (1 << LOG_EVENT_DETECTION) | \ - (1 << LOG_EVENT_LINK_TRAINING) | \ - (1 << LOG_EVENT_LINK_LOSS) | \ - (1 << LOG_EVENT_UNDERFLOW) | \ - (1 << LOG_RESOURCE) | \ - (1 << LOG_FEATURE_OVERRIDE) | \ - (1 << LOG_DETECTION_EDID_PARSER) | \ - (1 << LOG_DC) | \ - (1 << LOG_HW_HOTPLUG) | \ - (1 << LOG_HW_SET_MODE) | \ - (1 << LOG_HW_RESUME_S3) | \ - (1 << LOG_HW_HPD_IRQ) | \ - (1 << LOG_SYNC) | \ - (1 << LOG_BANDWIDTH_VALIDATION) | \ - (1 << LOG_MST) | \ - (1 << LOG_DETECTION_DP_CAPS) | \ - (1 << LOG_BACKLIGHT)) | \ - (1 << LOG_I2C_AUX) | \ - (1 << LOG_IF_TRACE) | \ - (1 << LOG_DTN) /* | \ - (1 << LOG_DEBUG) | \ - (1 << LOG_BIOS) | \ - (1 << LOG_SURFACE) | \ - (1 << LOG_SCALER) | \ - (1 << LOG_DML) | \ - (1 << LOG_HW_LINK_TRAINING) | \ - (1 << LOG_HW_AUDIO)| \ - (1 << LOG_BANDWIDTH_CALCS)*/ +#define DC_DEFAULT_LOG_MASK ((1ULL << LOG_ERROR) | \ + (1ULL << LOG_WARNING) | \ + (1ULL << LOG_EVENT_MODE_SET) | \ + (1ULL << LOG_EVENT_DETECTION) | \ + (1ULL << LOG_EVENT_LINK_TRAINING) | \ + (1ULL << LOG_EVENT_LINK_LOSS) | \ + (1ULL << LOG_EVENT_UNDERFLOW) | \ + (1ULL << LOG_RESOURCE) | \ + (1ULL << LOG_FEATURE_OVERRIDE) | \ + (1ULL << LOG_DETECTION_EDID_PARSER) | \ + (1ULL << LOG_DC) | \ + (1ULL << LOG_HW_HOTPLUG) | \ + (1ULL << LOG_HW_SET_MODE) | \ + (1ULL << LOG_HW_RESUME_S3) | \ + (1ULL << LOG_HW_HPD_IRQ) | \ + (1ULL << LOG_SYNC) | \ + (1ULL << LOG_BANDWIDTH_VALIDATION) | \ + (1ULL << LOG_MST) | \ + (1ULL << LOG_DETECTION_DP_CAPS) | \ + (1ULL << LOG_BACKLIGHT)) | \ + (1ULL << LOG_I2C_AUX) | \ + (1ULL << LOG_IF_TRACE) | \ + (1ULL << LOG_HDMI_FRL) | \ + (1ULL << LOG_DTN) /* | \ + (1ULL << LOG_DEBUG) | \ + (1ULL << LOG_BIOS) | \ + (1ULL << LOG_SURFACE) | \ + (1ULL << LOG_SCALER) | \ + (1ULL << LOG_DML) | \ + (1ULL << LOG_HW_LINK_TRAINING) | \ + (1ULL << LOG_HW_AUDIO)| \ + (1ULL << LOG_BANDWIDTH_CALCS)*/ #endif /* __DAL_LOGGER_TYPES_H__ */ diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.c index 83eaec4c6ad7..e9fbd94f8635 100644 --- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.c +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.c @@ -52,8 +52,8 @@ static uint8_t is_cp_desired_hdcp1(struct mod_hdcp *hdcp) * hdcp is not desired */ for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) { - if (hdcp->connection.displays[i].state != MOD_HDCP_DISPLAY_INACTIVE && - !hdcp->connection.displays[i].adjust.disable) { + if (hdcp->displays[i].state != MOD_HDCP_DISPLAY_INACTIVE && + !hdcp->displays[i].adjust.disable) { is_auth_needed = 1; break; } @@ -73,8 +73,8 @@ static uint8_t is_cp_desired_hdcp2(struct mod_hdcp *hdcp) * hdcp is not desired */ for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) { - if (hdcp->connection.displays[i].state != MOD_HDCP_DISPLAY_INACTIVE && - !hdcp->connection.displays[i].adjust.disable) { + if (hdcp->displays[i].state != MOD_HDCP_DISPLAY_INACTIVE && + !hdcp->displays[i].adjust.disable) { is_auth_needed = 1; break; } @@ -114,6 +114,9 @@ static enum mod_hdcp_status execution(struct mod_hdcp *hdcp, } else if (is_in_hdcp2_dp_states(hdcp)) { status = mod_hdcp_hdcp2_dp_execution(hdcp, event_ctx, &input->hdcp2); + } else { + event_ctx->unexpected_event = 1; + goto out; } out: return status; @@ -325,7 +328,8 @@ enum mod_hdcp_status mod_hdcp_add_display(struct mod_hdcp *hdcp, /* add display to connection */ hdcp->connection.link = *link; *display_container = *display; - status = mod_hdcp_add_display_to_topology(hdcp, display->index); + status = mod_hdcp_add_display_to_topology(hdcp, display_container); + if (status != MOD_HDCP_STATUS_SUCCESS) goto out; @@ -371,10 +375,10 @@ enum mod_hdcp_status mod_hdcp_remove_display(struct mod_hdcp *hdcp, status = mod_hdcp_remove_display_from_topology(hdcp, index); if (status != MOD_HDCP_STATUS_SUCCESS) goto out; - display->state = MOD_HDCP_DISPLAY_INACTIVE; + memset(display, 0, sizeof(struct mod_hdcp_display)); - /* request authentication for remaining displays*/ - if (get_active_display_count(hdcp) > 0) + /* request authentication when connection is not reset */ + if (current_state(hdcp) != HDCP_UNINITIALIZED) callback_in_ms(hdcp->connection.link.adjust.auth_delay * 1000, output); out: @@ -481,10 +485,8 @@ enum mod_hdcp_operation_mode mod_hdcp_signal_type_to_operation_mode( break; case SIGNAL_TYPE_EDP: case SIGNAL_TYPE_DISPLAY_PORT: - mode = MOD_HDCP_MODE_DP; - break; case SIGNAL_TYPE_DISPLAY_PORT_MST: - mode = MOD_HDCP_MODE_DP_MST; + mode = MOD_HDCP_MODE_DP; break; default: break; diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.h b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.h index b09d2f5502b3..60ff1a0028ac 100644 --- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.h +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.h @@ -165,7 +165,6 @@ struct mod_hdcp_auth_counters { /* contains values per connection */ struct mod_hdcp_connection { struct mod_hdcp_link link; - struct mod_hdcp_display displays[MAX_NUM_OF_DISPLAYS]; uint8_t is_repeater; uint8_t is_km_stored; uint8_t is_hdcp1_revoked; @@ -201,6 +200,8 @@ struct mod_hdcp { struct mod_hdcp_config config; /* per connection */ struct mod_hdcp_connection connection; + /* per displays */ + struct mod_hdcp_display displays[MAX_NUM_OF_DISPLAYS]; /* per authentication attempt */ struct mod_hdcp_authentication auth; /* per state in an authentication */ @@ -327,7 +328,7 @@ void mod_hdcp_dump_binary_message(uint8_t *msg, uint32_t msg_size, /* psp functions */ enum mod_hdcp_status mod_hdcp_add_display_to_topology( - struct mod_hdcp *hdcp, uint8_t index); + struct mod_hdcp *hdcp, struct mod_hdcp_display *display); enum mod_hdcp_status mod_hdcp_remove_display_from_topology( struct mod_hdcp *hdcp, uint8_t index); enum mod_hdcp_status mod_hdcp_hdcp1_create_session(struct mod_hdcp *hdcp); @@ -391,13 +392,13 @@ enum mod_hdcp_status mod_hdcp_write_content_type(struct mod_hdcp *hdcp); /* hdcp version helpers */ static inline uint8_t is_dp_hdcp(struct mod_hdcp *hdcp) { - return (hdcp->connection.link.mode == MOD_HDCP_MODE_DP || - hdcp->connection.link.mode == MOD_HDCP_MODE_DP_MST); + return (hdcp->connection.link.mode == MOD_HDCP_MODE_DP); } static inline uint8_t is_dp_mst_hdcp(struct mod_hdcp *hdcp) { - return (hdcp->connection.link.mode == MOD_HDCP_MODE_DP_MST); + return (hdcp->connection.link.mode == MOD_HDCP_MODE_DP && + hdcp->connection.link.dp.mst_supported); } static inline uint8_t is_hdmi_dvi_sl_hdcp(struct mod_hdcp *hdcp) @@ -502,11 +503,6 @@ static inline uint8_t is_display_active(struct mod_hdcp_display *display) return display->state >= MOD_HDCP_DISPLAY_ACTIVE; } -static inline uint8_t is_display_added(struct mod_hdcp_display *display) -{ - return display->state >= MOD_HDCP_DISPLAY_ACTIVE_AND_ADDED; -} - static inline uint8_t is_display_encryption_enabled(struct mod_hdcp_display *display) { return display->state >= MOD_HDCP_DISPLAY_ENCRYPTION_ENABLED; @@ -514,35 +510,24 @@ static inline uint8_t is_display_encryption_enabled(struct mod_hdcp_display *dis static inline uint8_t get_active_display_count(struct mod_hdcp *hdcp) { - uint8_t added_count = 0; - uint8_t i; - - for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) - if (is_display_active(&hdcp->connection.displays[i])) - added_count++; - return added_count; -} - -static inline uint8_t get_added_display_count(struct mod_hdcp *hdcp) -{ - uint8_t added_count = 0; + uint8_t active_count = 0; uint8_t i; for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) - if (is_display_added(&hdcp->connection.displays[i])) - added_count++; - return added_count; + if (is_display_active(&hdcp->displays[i])) + active_count++; + return active_count; } -static inline struct mod_hdcp_display *get_first_added_display( +static inline struct mod_hdcp_display *get_first_active_display( struct mod_hdcp *hdcp) { uint8_t i; struct mod_hdcp_display *display = NULL; for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) - if (is_display_added(&hdcp->connection.displays[i])) { - display = &hdcp->connection.displays[i]; + if (is_display_active(&hdcp->displays[i])) { + display = &hdcp->displays[i]; break; } return display; @@ -555,9 +540,9 @@ static inline struct mod_hdcp_display *get_active_display_at_index( struct mod_hdcp_display *display = NULL; for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) - if (hdcp->connection.displays[i].index == index && - is_display_active(&hdcp->connection.displays[i])) { - display = &hdcp->connection.displays[i]; + if (hdcp->displays[i].index == index && + is_display_active(&hdcp->displays[i])) { + display = &hdcp->displays[i]; break; } return display; @@ -570,8 +555,8 @@ static inline struct mod_hdcp_display *get_empty_display_container( struct mod_hdcp_display *display = NULL; for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) - if (!is_display_active(&hdcp->connection.displays[i])) { - display = &hdcp->connection.displays[i]; + if (!is_display_active(&hdcp->displays[i])) { + display = &hdcp->displays[i]; break; } return display; diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_execution.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_execution.c index 37c8c05497d6..f244b72e74e0 100644 --- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_execution.c +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_execution.c @@ -129,7 +129,7 @@ static inline uint8_t get_device_count(struct mod_hdcp *hdcp) static inline enum mod_hdcp_status check_device_count(struct mod_hdcp *hdcp) { /* device count must be greater than or equal to tracked hdcp displays */ - return (get_device_count(hdcp) < get_added_display_count(hdcp)) ? + return (get_device_count(hdcp) < get_active_display_count(hdcp)) ? MOD_HDCP_STATUS_HDCP1_DEVICE_COUNT_MISMATCH_FAILURE : MOD_HDCP_STATUS_SUCCESS; } diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_transition.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_transition.c index 5bc6706d2af7..f3711914364e 100644 --- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_transition.c +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_transition.c @@ -230,6 +230,9 @@ enum mod_hdcp_status mod_hdcp_hdcp1_dp_transition(struct mod_hdcp *hdcp, (!conn->is_repeater && is_dp_mst_hdcp(hdcp) && input->stream_encryption_dp != PASS)) { fail_and_restart_in_ms(0, &status, output); break; + } else if (conn->hdcp1_retry_count < conn->link.adjust.hdcp1.min_auth_retries_wa) { + fail_and_restart_in_ms(0, &status, output); + break; } if (conn->is_repeater) { set_watchdog_in_ms(hdcp, 5000, output); diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp2_execution.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp2_execution.c index 491c00f48026..549c113abcf7 100644 --- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp2_execution.c +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp2_execution.c @@ -208,7 +208,7 @@ static inline uint8_t get_device_count(struct mod_hdcp *hdcp) static enum mod_hdcp_status check_device_count(struct mod_hdcp *hdcp) { /* device count must be greater than or equal to tracked hdcp displays */ - return (get_device_count(hdcp) < get_added_display_count(hdcp)) ? + return (get_device_count(hdcp) < get_active_display_count(hdcp)) ? MOD_HDCP_STATUS_HDCP2_DEVICE_COUNT_MISMATCH_FAILURE : MOD_HDCP_STATUS_SUCCESS; } diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_ddc.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_ddc.c index ff9d54812e62..bb5130f4228d 100644 --- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_ddc.c +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_ddc.c @@ -65,6 +65,7 @@ enum mod_hdcp_ddc_message_id { MOD_HDCP_MESSAGE_ID_READ_LC_SEND_L_PRIME, MOD_HDCP_MESSAGE_ID_WRITE_SKE_SEND_EKS, MOD_HDCP_MESSAGE_ID_READ_REPEATER_AUTH_SEND_RECEIVERID_LIST, + MOD_HDCP_MESSAGE_ID_READ_REPEATER_AUTH_SEND_RECEIVERID_LIST_PART2, MOD_HDCP_MESSAGE_ID_WRITE_REPEATER_AUTH_SEND_ACK, MOD_HDCP_MESSAGE_ID_WRITE_REPEATER_AUTH_STREAM_MANAGE, MOD_HDCP_MESSAGE_ID_READ_REPEATER_AUTH_STREAM_READY, @@ -101,6 +102,7 @@ static const uint8_t hdcp_i2c_offsets[] = { [MOD_HDCP_MESSAGE_ID_READ_LC_SEND_L_PRIME] = 0x80, [MOD_HDCP_MESSAGE_ID_WRITE_SKE_SEND_EKS] = 0x60, [MOD_HDCP_MESSAGE_ID_READ_REPEATER_AUTH_SEND_RECEIVERID_LIST] = 0x80, + [MOD_HDCP_MESSAGE_ID_READ_REPEATER_AUTH_SEND_RECEIVERID_LIST_PART2] = 0x80, [MOD_HDCP_MESSAGE_ID_WRITE_REPEATER_AUTH_SEND_ACK] = 0x60, [MOD_HDCP_MESSAGE_ID_WRITE_REPEATER_AUTH_STREAM_MANAGE] = 0x60, [MOD_HDCP_MESSAGE_ID_READ_REPEATER_AUTH_STREAM_READY] = 0x80, @@ -135,6 +137,7 @@ static const uint32_t hdcp_dpcd_addrs[] = { [MOD_HDCP_MESSAGE_ID_READ_LC_SEND_L_PRIME] = 0x692f8, [MOD_HDCP_MESSAGE_ID_WRITE_SKE_SEND_EKS] = 0x69318, [MOD_HDCP_MESSAGE_ID_READ_REPEATER_AUTH_SEND_RECEIVERID_LIST] = 0x69330, + [MOD_HDCP_MESSAGE_ID_READ_REPEATER_AUTH_SEND_RECEIVERID_LIST_PART2] = 0x69340, [MOD_HDCP_MESSAGE_ID_WRITE_REPEATER_AUTH_SEND_ACK] = 0x693e0, [MOD_HDCP_MESSAGE_ID_WRITE_REPEATER_AUTH_STREAM_MANAGE] = 0x693f0, [MOD_HDCP_MESSAGE_ID_READ_REPEATER_AUTH_STREAM_READY] = 0x69473, @@ -405,7 +408,7 @@ enum mod_hdcp_status mod_hdcp_read_ake_cert(struct mod_hdcp *hdcp) enum mod_hdcp_status status; if (is_dp_hdcp(hdcp)) { - hdcp->auth.msg.hdcp2.ake_cert[0] = 3; + hdcp->auth.msg.hdcp2.ake_cert[0] = HDCP_2_2_AKE_SEND_CERT; status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_AKE_SEND_CERT, hdcp->auth.msg.hdcp2.ake_cert+1, sizeof(hdcp->auth.msg.hdcp2.ake_cert)-1); @@ -423,7 +426,7 @@ enum mod_hdcp_status mod_hdcp_read_h_prime(struct mod_hdcp *hdcp) enum mod_hdcp_status status; if (is_dp_hdcp(hdcp)) { - hdcp->auth.msg.hdcp2.ake_h_prime[0] = 7; + hdcp->auth.msg.hdcp2.ake_h_prime[0] = HDCP_2_2_AKE_SEND_HPRIME; status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_AKE_SEND_H_PRIME, hdcp->auth.msg.hdcp2.ake_h_prime+1, sizeof(hdcp->auth.msg.hdcp2.ake_h_prime)-1); @@ -441,7 +444,7 @@ enum mod_hdcp_status mod_hdcp_read_pairing_info(struct mod_hdcp *hdcp) enum mod_hdcp_status status; if (is_dp_hdcp(hdcp)) { - hdcp->auth.msg.hdcp2.ake_pairing_info[0] = 8; + hdcp->auth.msg.hdcp2.ake_pairing_info[0] = HDCP_2_2_AKE_SEND_PAIRING_INFO; status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_AKE_SEND_PAIRING_INFO, hdcp->auth.msg.hdcp2.ake_pairing_info+1, sizeof(hdcp->auth.msg.hdcp2.ake_pairing_info)-1); @@ -459,7 +462,7 @@ enum mod_hdcp_status mod_hdcp_read_l_prime(struct mod_hdcp *hdcp) enum mod_hdcp_status status; if (is_dp_hdcp(hdcp)) { - hdcp->auth.msg.hdcp2.lc_l_prime[0] = 10; + hdcp->auth.msg.hdcp2.lc_l_prime[0] = HDCP_2_2_LC_SEND_LPRIME; status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_LC_SEND_L_PRIME, hdcp->auth.msg.hdcp2.lc_l_prime+1, sizeof(hdcp->auth.msg.hdcp2.lc_l_prime)-1); @@ -474,14 +477,27 @@ enum mod_hdcp_status mod_hdcp_read_l_prime(struct mod_hdcp *hdcp) enum mod_hdcp_status mod_hdcp_read_rx_id_list(struct mod_hdcp *hdcp) { - enum mod_hdcp_status status; + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; if (is_dp_hdcp(hdcp)) { - hdcp->auth.msg.hdcp2.rx_id_list[0] = 12; - status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_REPEATER_AUTH_SEND_RECEIVERID_LIST, - hdcp->auth.msg.hdcp2.rx_id_list+1, - sizeof(hdcp->auth.msg.hdcp2.rx_id_list)-1); + uint32_t device_count = 0; + uint32_t rx_id_list_size = 0; + uint32_t bytes_read = 0; + hdcp->auth.msg.hdcp2.rx_id_list[0] = HDCP_2_2_REP_SEND_RECVID_LIST; + status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_REPEATER_AUTH_SEND_RECEIVERID_LIST, + hdcp->auth.msg.hdcp2.rx_id_list+1, + HDCP_MAX_AUX_TRANSACTION_SIZE); + if (status == MOD_HDCP_STATUS_SUCCESS) { + bytes_read = HDCP_MAX_AUX_TRANSACTION_SIZE; + device_count = HDCP_2_2_DEV_COUNT_LO(hdcp->auth.msg.hdcp2.rx_id_list[2]) + + (HDCP_2_2_DEV_COUNT_HI(hdcp->auth.msg.hdcp2.rx_id_list[1]) << 4); + rx_id_list_size = MIN((21 + 5 * device_count), + (sizeof(hdcp->auth.msg.hdcp2.rx_id_list) - 1)); + status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_REPEATER_AUTH_SEND_RECEIVERID_LIST_PART2, + hdcp->auth.msg.hdcp2.rx_id_list + 1 + bytes_read, + (rx_id_list_size - 1) / HDCP_MAX_AUX_TRANSACTION_SIZE * HDCP_MAX_AUX_TRANSACTION_SIZE); + } } else { status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_REPEATER_AUTH_SEND_RECEIVERID_LIST, hdcp->auth.msg.hdcp2.rx_id_list, @@ -495,7 +511,7 @@ enum mod_hdcp_status mod_hdcp_read_stream_ready(struct mod_hdcp *hdcp) enum mod_hdcp_status status; if (is_dp_hdcp(hdcp)) { - hdcp->auth.msg.hdcp2.repeater_auth_stream_ready[0] = 17; + hdcp->auth.msg.hdcp2.repeater_auth_stream_ready[0] = HDCP_2_2_REP_STREAM_READY; status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_REPEATER_AUTH_STREAM_READY, hdcp->auth.msg.hdcp2.repeater_auth_stream_ready+1, sizeof(hdcp->auth.msg.hdcp2.repeater_auth_stream_ready)-1); diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_log.h b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_log.h index 6e844825ad23..d3192b9d0c3d 100644 --- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_log.h +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_log.h @@ -37,10 +37,11 @@ /* default logs */ #define HDCP_ERROR_TRACE(hdcp, status) \ HDCP_LOG_ERR(hdcp, \ - "[Link %d] WARNING %s IN STATE %s", \ + "[Link %d] WARNING %s IN STATE %s STAY COUNT %d", \ hdcp->config.index, \ mod_hdcp_status_to_str(status), \ - mod_hdcp_state_id_to_str(hdcp->state.id)) + mod_hdcp_state_id_to_str(hdcp->state.id), \ + hdcp->state.stay_count) #define HDCP_HDCP1_ENABLED_TRACE(hdcp, displayIndex) \ HDCP_LOG_VER(hdcp, \ "[Link %d] HDCP 1.4 enabled on display %d", \ @@ -111,6 +112,9 @@ sizeof(hdcp->auth.msg.hdcp1.bksv)); \ HDCP_DDC_READ_TRACE(hdcp, "BCAPS", &hdcp->auth.msg.hdcp1.bcaps, \ sizeof(hdcp->auth.msg.hdcp1.bcaps)); \ + HDCP_DDC_READ_TRACE(hdcp, "BSTATUS", \ + (uint8_t *)&hdcp->auth.msg.hdcp1.bstatus, \ + sizeof(hdcp->auth.msg.hdcp1.bstatus)); \ HDCP_DDC_WRITE_TRACE(hdcp, "AN", hdcp->auth.msg.hdcp1.an, \ sizeof(hdcp->auth.msg.hdcp1.an)); \ HDCP_DDC_WRITE_TRACE(hdcp, "AKSV", hdcp->auth.msg.hdcp1.aksv, \ diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_psp.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_psp.c index d9cb2383d6de..836e47954938 100644 --- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_psp.c +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_psp.c @@ -54,7 +54,7 @@ enum mod_hdcp_status mod_hdcp_remove_display_from_topology( dtm_cmd = (struct ta_dtm_shared_memory *)psp->dtm_context.dtm_shared_buf; - if (!display || !is_display_added(display)) + if (!display || !is_display_active(display)) return MOD_HDCP_STATUS_DISPLAY_NOT_FOUND; memset(dtm_cmd, 0, sizeof(struct ta_dtm_shared_memory)); @@ -76,22 +76,18 @@ enum mod_hdcp_status mod_hdcp_remove_display_from_topology( } enum mod_hdcp_status mod_hdcp_add_display_to_topology( - struct mod_hdcp *hdcp, uint8_t index) + struct mod_hdcp *hdcp, struct mod_hdcp_display *display) { struct psp_context *psp = hdcp->config.psp.handle; struct ta_dtm_shared_memory *dtm_cmd; - struct mod_hdcp_display *display = - get_active_display_at_index(hdcp, index); struct mod_hdcp_link *link = &hdcp->connection.link; if (!psp->dtm_context.dtm_initialized) { DRM_ERROR("Failed to add display topology, DTM TA is not initialized."); + display->state = MOD_HDCP_DISPLAY_INACTIVE; return MOD_HDCP_STATUS_FAILURE; } - if (!display || is_display_added(display)) - return MOD_HDCP_STATUS_UPDATE_TOPOLOGY_FAILURE; - dtm_cmd = (struct ta_dtm_shared_memory *)psp->dtm_context.dtm_shared_buf; memset(dtm_cmd, 0, sizeof(struct ta_dtm_shared_memory)); @@ -113,20 +109,21 @@ enum mod_hdcp_status mod_hdcp_add_display_to_topology( psp_dtm_invoke(psp, dtm_cmd->cmd_id); - if (dtm_cmd->dtm_status != TA_DTM_STATUS__SUCCESS) + if (dtm_cmd->dtm_status != TA_DTM_STATUS__SUCCESS) { + display->state = MOD_HDCP_DISPLAY_INACTIVE; return MOD_HDCP_STATUS_UPDATE_TOPOLOGY_FAILURE; + } - display->state = MOD_HDCP_DISPLAY_ACTIVE_AND_ADDED; HDCP_TOP_ADD_DISPLAY_TRACE(hdcp, display->index); - - return MOD_HDCP_STATUS_SUCCESS; + + return MOD_HDCP_STATUS_SUCCESS; } enum mod_hdcp_status mod_hdcp_hdcp1_create_session(struct mod_hdcp *hdcp) { struct psp_context *psp = hdcp->config.psp.handle; - struct mod_hdcp_display *display = get_first_added_display(hdcp); + struct mod_hdcp_display *display = get_first_active_display(hdcp); struct ta_hdcp_shared_memory *hdcp_cmd; if (!psp->hdcp_context.hdcp_initialized) { @@ -177,11 +174,11 @@ enum mod_hdcp_status mod_hdcp_hdcp1_destroy_session(struct mod_hdcp *hdcp) HDCP_TOP_HDCP1_DESTROY_SESSION_TRACE(hdcp); for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) if (is_display_encryption_enabled( - &hdcp->connection.displays[i])) { - hdcp->connection.displays[i].state = - MOD_HDCP_DISPLAY_ACTIVE_AND_ADDED; + &hdcp->displays[i])) { + hdcp->displays[i].state = + MOD_HDCP_DISPLAY_ACTIVE; HDCP_HDCP1_DISABLED_TRACE(hdcp, - hdcp->connection.displays[i].index); + hdcp->displays[i].index); } return MOD_HDCP_STATUS_SUCCESS; @@ -231,7 +228,7 @@ enum mod_hdcp_status mod_hdcp_hdcp1_enable_encryption(struct mod_hdcp *hdcp) { struct psp_context *psp = hdcp->config.psp.handle; struct ta_hdcp_shared_memory *hdcp_cmd; - struct mod_hdcp_display *display = get_first_added_display(hdcp); + struct mod_hdcp_display *display = get_first_active_display(hdcp); hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); @@ -301,14 +298,13 @@ enum mod_hdcp_status mod_hdcp_hdcp1_enable_dp_stream_encryption(struct mod_hdcp for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) { - if (hdcp->connection.displays[i].state != MOD_HDCP_DISPLAY_ACTIVE_AND_ADDED || - hdcp->connection.displays[i].adjust.disable) + if (hdcp->displays[i].adjust.disable) continue; memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); hdcp_cmd->in_msg.hdcp1_enable_dp_stream_encryption.session_handle = hdcp->auth.id; - hdcp_cmd->in_msg.hdcp1_enable_dp_stream_encryption.display_handle = hdcp->connection.displays[i].index; + hdcp_cmd->in_msg.hdcp1_enable_dp_stream_encryption.display_handle = hdcp->displays[i].index; hdcp_cmd->cmd_id = TA_HDCP_COMMAND__HDCP1_ENABLE_DP_STREAM_ENCRYPTION; psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); @@ -316,8 +312,8 @@ enum mod_hdcp_status mod_hdcp_hdcp1_enable_dp_stream_encryption(struct mod_hdcp if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) return MOD_HDCP_STATUS_HDCP1_ENABLE_STREAM_ENCRYPTION_FAILURE; - hdcp->connection.displays[i].state = MOD_HDCP_DISPLAY_ENCRYPTION_ENABLED; - HDCP_HDCP1_ENABLED_TRACE(hdcp, hdcp->connection.displays[i].index); + hdcp->displays[i].state = MOD_HDCP_DISPLAY_ENCRYPTION_ENABLED; + HDCP_HDCP1_ENABLED_TRACE(hdcp, hdcp->displays[i].index); } return MOD_HDCP_STATUS_SUCCESS; @@ -364,7 +360,7 @@ enum mod_hdcp_status mod_hdcp_hdcp2_create_session(struct mod_hdcp *hdcp) { struct psp_context *psp = hdcp->config.psp.handle; struct ta_hdcp_shared_memory *hdcp_cmd; - struct mod_hdcp_display *display = get_first_added_display(hdcp); + struct mod_hdcp_display *display = get_first_active_display(hdcp); if (!psp->hdcp_context.hdcp_initialized) { DRM_ERROR("Failed to create hdcp session, HDCP TA is not initialized"); @@ -421,11 +417,11 @@ enum mod_hdcp_status mod_hdcp_hdcp2_destroy_session(struct mod_hdcp *hdcp) HDCP_TOP_HDCP2_DESTROY_SESSION_TRACE(hdcp); for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) if (is_display_encryption_enabled( - &hdcp->connection.displays[i])) { - hdcp->connection.displays[i].state = - MOD_HDCP_DISPLAY_ACTIVE_AND_ADDED; + &hdcp->displays[i])) { + hdcp->displays[i].state = + MOD_HDCP_DISPLAY_ACTIVE; HDCP_HDCP2_DISABLED_TRACE(hdcp, - hdcp->connection.displays[i].index); + hdcp->displays[i].index); } return MOD_HDCP_STATUS_SUCCESS; @@ -662,7 +658,7 @@ enum mod_hdcp_status mod_hdcp_hdcp2_enable_encryption(struct mod_hdcp *hdcp) { struct psp_context *psp = hdcp->config.psp.handle; struct ta_hdcp_shared_memory *hdcp_cmd; - struct mod_hdcp_display *display = get_first_added_display(hdcp); + struct mod_hdcp_display *display = get_first_active_display(hdcp); hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); @@ -747,10 +743,9 @@ enum mod_hdcp_status mod_hdcp_hdcp2_enable_dp_stream_encryption(struct mod_hdcp for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) { - if (hdcp->connection.displays[i].state != MOD_HDCP_DISPLAY_ACTIVE_AND_ADDED || - hdcp->connection.displays[i].adjust.disable) + if (hdcp->displays[i].adjust.disable) continue; - hdcp_cmd->in_msg.hdcp2_enable_dp_stream_encryption.display_handle = hdcp->connection.displays[i].index; + hdcp_cmd->in_msg.hdcp2_enable_dp_stream_encryption.display_handle = hdcp->displays[i].index; hdcp_cmd->in_msg.hdcp2_enable_dp_stream_encryption.session_handle = hdcp->auth.id; hdcp_cmd->cmd_id = TA_HDCP_COMMAND__HDCP2_ENABLE_DP_STREAM_ENCRYPTION; @@ -759,8 +754,8 @@ enum mod_hdcp_status mod_hdcp_hdcp2_enable_dp_stream_encryption(struct mod_hdcp if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) break; - hdcp->connection.displays[i].state = MOD_HDCP_DISPLAY_ENCRYPTION_ENABLED; - HDCP_HDCP2_ENABLED_TRACE(hdcp, hdcp->connection.displays[i].index); + hdcp->displays[i].state = MOD_HDCP_DISPLAY_ENCRYPTION_ENABLED; + HDCP_HDCP2_ENABLED_TRACE(hdcp, hdcp->displays[i].index); } return (hdcp_cmd->hdcp_status == TA_HDCP_STATUS__SUCCESS) ? MOD_HDCP_STATUS_SUCCESS diff --git a/drivers/gpu/drm/amd/display/modules/inc/mod_hdcp.h b/drivers/gpu/drm/amd/display/modules/inc/mod_hdcp.h index 891bca555e17..eae9309cfb24 100644 --- a/drivers/gpu/drm/amd/display/modules/inc/mod_hdcp.h +++ b/drivers/gpu/drm/amd/display/modules/inc/mod_hdcp.h @@ -102,6 +102,7 @@ enum mod_hdcp_status { struct mod_hdcp_displayport { uint8_t rev; uint8_t assr_supported; + uint8_t mst_supported; }; struct mod_hdcp_hdmi { @@ -110,14 +111,12 @@ struct mod_hdcp_hdmi { enum mod_hdcp_operation_mode { MOD_HDCP_MODE_OFF, MOD_HDCP_MODE_DEFAULT, - MOD_HDCP_MODE_DP, - MOD_HDCP_MODE_DP_MST + MOD_HDCP_MODE_DP }; enum mod_hdcp_display_state { MOD_HDCP_DISPLAY_INACTIVE = 0, MOD_HDCP_DISPLAY_ACTIVE, - MOD_HDCP_DISPLAY_ACTIVE_AND_ADDED, MOD_HDCP_DISPLAY_ENCRYPTION_ENABLED }; @@ -157,7 +156,8 @@ struct mod_hdcp_display_adjustment { struct mod_hdcp_link_adjustment_hdcp1 { uint8_t disable : 1; uint8_t postpone_encryption : 1; - uint8_t reserved : 6; + uint8_t min_auth_retries_wa : 1; + uint8_t reserved : 5; }; enum mod_hdcp_force_hdcp_type { diff --git a/drivers/gpu/drm/amd/display/modules/inc/mod_info_packet.h b/drivers/gpu/drm/amd/display/modules/inc/mod_info_packet.h index 42cbeffac640..13c57ff2abdc 100644 --- a/drivers/gpu/drm/amd/display/modules/inc/mod_info_packet.h +++ b/drivers/gpu/drm/amd/display/modules/inc/mod_info_packet.h @@ -34,8 +34,7 @@ struct dc_info_packet; struct mod_vrr_params; void mod_build_vsc_infopacket(const struct dc_stream_state *stream, - struct dc_info_packet *info_packet, - bool *use_vsc_sdp_for_colorimetry); + struct dc_info_packet *info_packet); void mod_build_hf_vsif_infopacket(const struct dc_stream_state *stream, struct dc_info_packet *info_packet, int ALLMEnabled, int ALLMValue); diff --git a/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c b/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c index 6a8a056424b8..cff3ab15fc0c 100644 --- a/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c +++ b/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c @@ -130,8 +130,7 @@ enum ColorimetryYCCDP { }; void mod_build_vsc_infopacket(const struct dc_stream_state *stream, - struct dc_info_packet *info_packet, - bool *use_vsc_sdp_for_colorimetry) + struct dc_info_packet *info_packet) { unsigned int vsc_packet_revision = vsc_packet_undefined; unsigned int i; @@ -139,11 +138,6 @@ void mod_build_vsc_infopacket(const struct dc_stream_state *stream, unsigned int colorimetryFormat = 0; bool stereo3dSupport = false; - /* Initialize first, later if infopacket is valid determine if VSC SDP - * should be used to signal colorimetry format and pixel encoding. - */ - *use_vsc_sdp_for_colorimetry = false; - if (stream->timing.timing_3d_format != TIMING_3D_FORMAT_NONE && stream->view_format != VIEW_3D_FORMAT_NONE) { vsc_packet_revision = vsc_packet_rev1; stereo3dSupport = true; @@ -153,9 +147,8 @@ void mod_build_vsc_infopacket(const struct dc_stream_state *stream, if (stream->psr_version != 0) vsc_packet_revision = vsc_packet_rev2; - /* Update to revision 5 for extended colorimetry support for DPCD 1.4+ */ - if (stream->link->dpcd_caps.dpcd_rev.raw >= 0x14 && - stream->link->dpcd_caps.dprx_feature.bits.VSC_SDP_COLORIMETRY_SUPPORTED) + /* Update to revision 5 for extended colorimetry support */ + if (stream->use_vsc_sdp_for_colorimetry) vsc_packet_revision = vsc_packet_rev5; /* VSC packet not needed based on the features @@ -269,13 +262,6 @@ void mod_build_vsc_infopacket(const struct dc_stream_state *stream, info_packet->valid = true; - /* If we are using VSC SDP revision 05h, use this to signal for - * colorimetry format and pixel encoding. HW should later be - * programmed to set MSA MISC1 bit 6 to indicate ignore - * colorimetry format and pixel encoding in the MSA. - */ - *use_vsc_sdp_for_colorimetry = true; - /* Set VSC SDP fields for pixel encoding and colorimetry format from DP 1.3 specs * Data Bytes DB 18~16 * Bits 3:0 (Colorimetry Format) | Bits 7:4 (Pixel Encoding) diff --git a/drivers/gpu/drm/amd/include/asic_reg/wafl/wafl2_4_0_0_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/wafl/wafl2_4_0_0_sh_mask.h new file mode 100644 index 000000000000..82b6cc25205e --- /dev/null +++ b/drivers/gpu/drm/amd/include/asic_reg/wafl/wafl2_4_0_0_sh_mask.h @@ -0,0 +1,69 @@ +/* + * Copyright 2020 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ + +#ifndef _wafl2_4_0_0_SH_MASK_HEADER +#define _wafl2_4_0_0_SH_MASK_HEADER + +//PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__DataLossErr__SHIFT 0x0 +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__TrainingErr__SHIFT 0x1 +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__CRCErr__SHIFT 0x5 +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__BERExceededErr__SHIFT 0x6 +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__TxMetaDataErr__SHIFT 0x7 +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__ReplayBufParityErr__SHIFT 0x8 +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__DataParityErr__SHIFT 0x9 +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__ReplayFifoOverflowErr__SHIFT 0xa +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__ReplayFifoUnderflowErr__SHIFT 0xb +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__ElasticFifoOverflowErr__SHIFT 0xc +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__DeskewErr__SHIFT 0xd +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__DataStartupLimitErr__SHIFT 0xf +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__FCInitTimeoutErr__SHIFT 0x10 +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__RecoveryTimeoutErr__SHIFT 0x11 +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__ReadySerialTimeoutErr__SHIFT 0x12 +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__ReadySerialAttemptErr__SHIFT 0x13 +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__RecoveryAttemptErr__SHIFT 0x14 +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__RecoveryRelockAttemptErr__SHIFT 0x15 +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__ClearBERAccum__SHIFT 0x17 +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__BERAccumulator__SHIFT 0x18 +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__DataLossErr_MASK 0x00000001L +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__TrainingErr_MASK 0x00000002L +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__CRCErr_MASK 0x00000020L +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__BERExceededErr_MASK 0x00000040L +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__TxMetaDataErr_MASK 0x00000080L +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__ReplayBufParityErr_MASK 0x00000100L +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__DataParityErr_MASK 0x00000200L +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__ReplayFifoOverflowErr_MASK 0x00000400L +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__ReplayFifoUnderflowErr_MASK 0x00000800L +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__ElasticFifoOverflowErr_MASK 0x00001000L +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__DeskewErr_MASK 0x00002000L +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__DataStartupLimitErr_MASK 0x00008000L +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__FCInitTimeoutErr_MASK 0x00010000L +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__RecoveryTimeoutErr_MASK 0x00020000L +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__ReadySerialTimeoutErr_MASK 0x00040000L +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__ReadySerialAttemptErr_MASK 0x00080000L +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__RecoveryAttemptErr_MASK 0x00100000L +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__RecoveryRelockAttemptErr_MASK 0x00200000L +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__ClearBERAccum_MASK 0x00800000L +#define PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS__BERAccumulator_MASK 0xFF000000L + +#endif diff --git a/drivers/gpu/drm/amd/include/asic_reg/wafl/wafl2_4_0_0_smn.h b/drivers/gpu/drm/amd/include/asic_reg/wafl/wafl2_4_0_0_smn.h new file mode 100644 index 000000000000..4a51a90c611a --- /dev/null +++ b/drivers/gpu/drm/amd/include/asic_reg/wafl/wafl2_4_0_0_smn.h @@ -0,0 +1,29 @@ +/* + * Copyright 2020 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ + +#ifndef _wafl2_4_0_0_SMN_HEADER +#define _wafl2_4_0_0_SMN_HEADER + +#define smnPCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS 0x11cf0210 + +#endif diff --git a/drivers/gpu/drm/amd/include/asic_reg/xgmi/xgmi_4_0_0_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/xgmi/xgmi_4_0_0_sh_mask.h new file mode 100644 index 000000000000..f37712f05b03 --- /dev/null +++ b/drivers/gpu/drm/amd/include/asic_reg/xgmi/xgmi_4_0_0_sh_mask.h @@ -0,0 +1,69 @@ +/* + * Copyright 2020 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ + +#ifndef _xgmi_4_0_0_SH_MASK_HEADER +#define _xgmi_4_0_0_SH_MASK_HEADER + +//PCS_GOPX16_PCS_ERROR_STATUS +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__DataLossErr__SHIFT 0x0 +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__TrainingErr__SHIFT 0x1 +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__CRCErr__SHIFT 0x5 +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__BERExceededErr__SHIFT 0x6 +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__TxMetaDataErr__SHIFT 0x7 +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__ReplayBufParityErr__SHIFT 0x8 +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__DataParityErr__SHIFT 0x9 +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__ReplayFifoOverflowErr__SHIFT 0xa +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__ReplayFifoUnderflowErr__SHIFT 0xb +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__ElasticFifoOverflowErr__SHIFT 0xc +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__DeskewErr__SHIFT 0xd +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__DataStartupLimitErr__SHIFT 0xf +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__FCInitTimeoutErr__SHIFT 0x10 +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__RecoveryTimeoutErr__SHIFT 0x11 +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__ReadySerialTimeoutErr__SHIFT 0x12 +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__ReadySerialAttemptErr__SHIFT 0x13 +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__RecoveryAttemptErr__SHIFT 0x14 +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__RecoveryRelockAttemptErr__SHIFT 0x15 +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__ClearBERAccum__SHIFT 0x17 +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__BERAccumulator__SHIFT 0x18 +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__DataLossErr_MASK 0x00000001L +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__TrainingErr_MASK 0x00000002L +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__CRCErr_MASK 0x00000020L +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__BERExceededErr_MASK 0x00000040L +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__TxMetaDataErr_MASK 0x00000080L +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__ReplayBufParityErr_MASK 0x00000100L +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__DataParityErr_MASK 0x00000200L +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__ReplayFifoOverflowErr_MASK 0x00000400L +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__ReplayFifoUnderflowErr_MASK 0x00000800L +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__ElasticFifoOverflowErr_MASK 0x00001000L +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__DeskewErr_MASK 0x00002000L +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__DataStartupLimitErr_MASK 0x00008000L +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__FCInitTimeoutErr_MASK 0x00010000L +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__RecoveryTimeoutErr_MASK 0x00020000L +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__ReadySerialTimeoutErr_MASK 0x00040000L +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__ReadySerialAttemptErr_MASK 0x00080000L +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__RecoveryAttemptErr_MASK 0x00100000L +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__RecoveryRelockAttemptErr_MASK 0x00200000L +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__ClearBERAccum_MASK 0x00800000L +#define XGMI0_PCS_GOPX16_PCS_ERROR_STATUS__BERAccumulator_MASK 0xFF000000L + +#endif diff --git a/drivers/gpu/drm/amd/include/asic_reg/xgmi/xgmi_4_0_0_smn.h b/drivers/gpu/drm/amd/include/asic_reg/xgmi/xgmi_4_0_0_smn.h new file mode 100644 index 000000000000..6ccbac4ce87e --- /dev/null +++ b/drivers/gpu/drm/amd/include/asic_reg/xgmi/xgmi_4_0_0_smn.h @@ -0,0 +1,29 @@ +/* + * Copyright 2020 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ + +#ifndef _xgmi_4_0_0_SMN_HEADER +#define _xgmi_4_0_0_SMN_HEADER + +#define smnXGMI0_PCS_GOPX16_PCS_ERROR_STATUS 0x11af0210 + +#endif diff --git a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h index abc0eb4ac493..a3c238c39ef5 100644 --- a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h +++ b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h @@ -167,27 +167,6 @@ struct tile_config { #define KFD_MAX_NUM_OF_QUEUES_PER_DEVICE_DEFAULT 4096 -/* - * Allocation flag domains - * NOTE: This must match the corresponding definitions in kfd_ioctl.h. - */ -#define ALLOC_MEM_FLAGS_VRAM (1 << 0) -#define ALLOC_MEM_FLAGS_GTT (1 << 1) -#define ALLOC_MEM_FLAGS_USERPTR (1 << 2) -#define ALLOC_MEM_FLAGS_DOORBELL (1 << 3) -#define ALLOC_MEM_FLAGS_MMIO_REMAP (1 << 4) - -/* - * Allocation flags attributes/access options. - * NOTE: This must match the corresponding definitions in kfd_ioctl.h. - */ -#define ALLOC_MEM_FLAGS_WRITABLE (1 << 31) -#define ALLOC_MEM_FLAGS_EXECUTABLE (1 << 30) -#define ALLOC_MEM_FLAGS_PUBLIC (1 << 29) -#define ALLOC_MEM_FLAGS_NO_SUBSTITUTE (1 << 28) /* TODO */ -#define ALLOC_MEM_FLAGS_AQL_QUEUE_MEM (1 << 27) -#define ALLOC_MEM_FLAGS_COHERENT (1 << 26) /* For GFXv9 or later */ - /** * struct kfd2kgd_calls * @@ -223,8 +202,6 @@ struct tile_config { * @set_scratch_backing_va: Sets VA for scratch backing memory of a VMID. * Only used for no cp scheduling mode * - * @get_tile_config: Returns GPU-specific tiling mode information - * * @set_vm_context_page_table_base: Program page table base for a VMID * * @invalidate_tlbs: Invalidate TLBs for a specific PASID @@ -310,8 +287,6 @@ struct kfd2kgd_calls { void (*set_scratch_backing_va)(struct kgd_dev *kgd, uint64_t va, uint32_t vmid); - int (*get_tile_config)(struct kgd_dev *kgd, struct tile_config *config); - void (*set_vm_context_page_table_base)(struct kgd_dev *kgd, uint32_t vmid, uint64_t page_table_base); uint32_t (*read_vmid_from_vmfault_reg)(struct kgd_dev *kgd); diff --git a/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c b/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c index 9f2428fd98f6..e8b27fab6aa1 100644 --- a/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c +++ b/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c @@ -23,15 +23,12 @@ #include <linux/firmware.h> #include <linux/pci.h> -#include "pp_debug.h" #include "amdgpu.h" #include "amdgpu_smu.h" #include "smu_internal.h" -#include "soc15_common.h" #include "smu_v11_0.h" #include "smu_v12_0.h" #include "atom.h" -#include "amd_pcie.h" #include "vega20_ppt.h" #include "arcturus_ppt.h" #include "navi10_ppt.h" @@ -121,20 +118,20 @@ static int smu_feature_update_enable_state(struct smu_context *smu, if (enabled) { ret = smu_send_smc_msg_with_param(smu, SMU_MSG_EnableSmuFeaturesLow, - feature_low); + feature_low, NULL); if (ret) return ret; ret = smu_send_smc_msg_with_param(smu, SMU_MSG_EnableSmuFeaturesHigh, - feature_high); + feature_high, NULL); if (ret) return ret; } else { ret = smu_send_smc_msg_with_param(smu, SMU_MSG_DisableSmuFeaturesLow, - feature_low); + feature_low, NULL); if (ret) return ret; ret = smu_send_smc_msg_with_param(smu, SMU_MSG_DisableSmuFeaturesHigh, - feature_high); + feature_high, NULL); if (ret) return ret; } @@ -195,21 +192,13 @@ int smu_get_smc_version(struct smu_context *smu, uint32_t *if_version, uint32_t return -EINVAL; if (if_version) { - ret = smu_send_smc_msg(smu, SMU_MSG_GetDriverIfVersion); - if (ret) - return ret; - - ret = smu_read_smc_arg(smu, if_version); + ret = smu_send_smc_msg(smu, SMU_MSG_GetDriverIfVersion, if_version); if (ret) return ret; } if (smu_version) { - ret = smu_send_smc_msg(smu, SMU_MSG_GetSmuVersion); - if (ret) - return ret; - - ret = smu_read_smc_arg(smu, smu_version); + ret = smu_send_smc_msg(smu, SMU_MSG_GetSmuVersion, smu_version); if (ret) return ret; } @@ -218,17 +207,19 @@ int smu_get_smc_version(struct smu_context *smu, uint32_t *if_version, uint32_t } int smu_set_soft_freq_range(struct smu_context *smu, enum smu_clk_type clk_type, - uint32_t min, uint32_t max) + uint32_t min, uint32_t max, bool lock_needed) { int ret = 0; - if (min <= 0 && max <= 0) - return -EINVAL; - if (!smu_clk_dpm_is_enabled(smu, clk_type)) return 0; + if (lock_needed) + mutex_lock(&smu->mutex); ret = smu_set_soft_freq_limited_range(smu, clk_type, min, max); + if (lock_needed) + mutex_unlock(&smu->mutex); + return ret; } @@ -251,7 +242,7 @@ int smu_set_hard_freq_range(struct smu_context *smu, enum smu_clk_type clk_type, if (max > 0) { param = (uint32_t)((clk_id << 16) | (max & 0xffff)); ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMaxByFreq, - param); + param, NULL); if (ret) return ret; } @@ -259,7 +250,7 @@ int smu_set_hard_freq_range(struct smu_context *smu, enum smu_clk_type clk_type, if (min > 0) { param = (uint32_t)((clk_id << 16) | (min & 0xffff)); ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinByFreq, - param); + param, NULL); if (ret) return ret; } @@ -335,12 +326,8 @@ int smu_get_dpm_freq_by_index(struct smu_context *smu, enum smu_clk_type clk_typ param = (uint32_t)(((clk_id & 0xffff) << 16) | (level & 0xffff)); - ret = smu_send_smc_msg_with_param(smu,SMU_MSG_GetDpmFreqByIndex, - param); - if (ret) - return ret; - - ret = smu_read_smc_arg(smu, ¶m); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_GetDpmFreqByIndex, + param, ¶m); if (ret) return ret; @@ -542,7 +529,8 @@ int smu_update_table(struct smu_context *smu, enum smu_table_id table_index, int ret = smu_send_smc_msg_with_param(smu, drv2smu ? SMU_MSG_TransferTableDram2Smu : SMU_MSG_TransferTableSmu2Dram, - table_id | ((argument & 0xFFFF) << 16)); + table_id | ((argument & 0xFFFF) << 16), + NULL); if (ret) return ret; @@ -900,6 +888,7 @@ static int smu_sw_init(void *handle) mutex_init(&smu->sensor_lock); mutex_init(&smu->metrics_lock); + mutex_init(&smu->message_lock); smu->watermarks_bitmap = 0; smu->power_profile_mode = PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT; @@ -943,6 +932,13 @@ static int smu_sw_init(void *handle) return ret; } + if (adev->smu.ppt_funcs->i2c_eeprom_init) { + ret = smu_i2c_eeprom_init(smu, &adev->pm.smu_i2c); + + if (ret) + return ret; + } + return 0; } @@ -952,6 +948,9 @@ static int smu_sw_fini(void *handle) struct smu_context *smu = &adev->smu; int ret; + if (adev->smu.ppt_funcs->i2c_eeprom_fini) + smu_i2c_eeprom_fini(smu, &adev->pm.smu_i2c); + kfree(smu->irq_source); smu->irq_source = NULL; @@ -1155,6 +1154,21 @@ static int smu_smc_table_hw_init(struct smu_context *smu, } } } + + if (smu->ppt_funcs->set_power_source) { + /* + * For Navi1X, manually switch it to AC mode as PMFW + * may boot it with DC mode. + */ + if (adev->pm.ac_power) + ret = smu_set_power_source(smu, SMU_POWER_SOURCE_AC); + else + ret = smu_set_power_source(smu, SMU_POWER_SOURCE_DC); + if (ret) { + pr_err("Failed to switch to %s mode!\n", adev->pm.ac_power ? "AC" : "DC"); + return ret; + } + } } if (adev->asic_type != CHIP_ARCTURUS) { ret = smu_notify_display_change(smu); @@ -1471,21 +1485,26 @@ static int smu_disable_dpm(struct smu_context *smu) } /* - * For baco on Arcturus, this operation - * (disable all smu feature) will be handled by SMU FW. + * Disable all enabled SMU features. + * This should be handled in SMU FW, as a backup + * driver can issue call to SMU FW until sequence + * in SMU FW is operational. */ - if (adev->asic_type == CHIP_ARCTURUS) { - if (use_baco && (smu_version > 0x360e00)) - return 0; - } - - /* Disable all enabled SMU features */ ret = smu_system_features_control(smu, false); if (ret) { pr_err("Failed to disable smu features.\n"); return ret; } + /* + * Arcturus does not have BACO bit in disable feature mask. + * Enablement of BACO bit on Arcturus should be skipped. + */ + if (adev->asic_type == CHIP_ARCTURUS) { + if (use_baco && (smu_version > 0x360e00)) + return 0; + } + /* For baco, need to leave BACO feature enabled */ if (use_baco) { /* @@ -1992,7 +2011,7 @@ int smu_set_mp1_state(struct smu_context *smu, return 0; } - ret = smu_send_smc_msg(smu, msg); + ret = smu_send_smc_msg(smu, msg, NULL); if (ret) pr_err("[PrepareMp1] Failed!\n"); @@ -2056,8 +2075,11 @@ int smu_set_watermarks_for_clock_ranges(struct smu_context *smu, smu_feature_is_enabled(smu, SMU_FEATURE_DPM_DCEFCLK_BIT) && smu_feature_is_enabled(smu, SMU_FEATURE_DPM_SOCCLK_BIT)) { smu_set_watermarks_table(smu, table, clock_ranges); - smu->watermarks_bitmap |= WATERMARKS_EXIST; - smu->watermarks_bitmap &= ~WATERMARKS_LOADED; + + if (!(smu->watermarks_bitmap & WATERMARKS_EXIST)) { + smu->watermarks_bitmap |= WATERMARKS_EXIST; + smu->watermarks_bitmap &= ~WATERMARKS_LOADED; + } } mutex_unlock(&smu->mutex); @@ -2065,6 +2087,29 @@ int smu_set_watermarks_for_clock_ranges(struct smu_context *smu, return 0; } +int smu_set_ac_dc(struct smu_context *smu) +{ + int ret = 0; + + /* controlled by firmware */ + if (smu->dc_controlled_by_gpio) + return 0; + + mutex_lock(&smu->mutex); + if (smu->ppt_funcs->set_power_source) { + if (smu->adev->pm.ac_power) + ret = smu_set_power_source(smu, SMU_POWER_SOURCE_AC); + else + ret = smu_set_power_source(smu, SMU_POWER_SOURCE_DC); + if (ret) + pr_err("Failed to switch to %s mode!\n", + smu->adev->pm.ac_power ? "AC" : "DC"); + } + mutex_unlock(&smu->mutex); + + return ret; +} + const struct amd_ip_funcs smu_ip_funcs = { .name = "smu", .early_init = smu_early_init, @@ -2667,12 +2712,3 @@ uint32_t smu_get_pptable_power_limit(struct smu_context *smu) return ret; } - -int smu_send_smc_msg(struct smu_context *smu, - enum smu_message_type msg) -{ - int ret; - - ret = smu_send_smc_msg_with_param(smu, msg, 0); - return ret; -} diff --git a/drivers/gpu/drm/amd/powerplay/arcturus_ppt.c b/drivers/gpu/drm/amd/powerplay/arcturus_ppt.c index d3c4e7a8c1b1..c6d3bef15320 100644 --- a/drivers/gpu/drm/amd/powerplay/arcturus_ppt.c +++ b/drivers/gpu/drm/amd/powerplay/arcturus_ppt.c @@ -21,7 +21,6 @@ * */ -#include "pp_debug.h" #include <linux/firmware.h> #include "amdgpu.h" #include "amdgpu_smu.h" @@ -42,7 +41,7 @@ #include <linux/pci.h> #include "amdgpu_ras.h" -#define to_amdgpu_device(x) (container_of(x, struct amdgpu_ras, eeprom_control.eeprom_accessor))->adev +#define to_amdgpu_device(x) (container_of(x, struct amdgpu_device, pm.smu_i2c)) #define CTF_OFFSET_EDGE 5 #define CTF_OFFSET_HOTSPOT 5 @@ -374,13 +373,13 @@ arcturus_set_single_dpm_table(struct smu_context *smu, ret = smu_send_smc_msg_with_param(smu, SMU_MSG_GetDpmFreqByIndex, - (clk_id << 16 | 0xFF)); + (clk_id << 16 | 0xFF), + &num_of_levels); if (ret) { pr_err("[%s] failed to get dpm levels!\n", __func__); return ret; } - smu_read_smc_arg(smu, &num_of_levels); if (!num_of_levels) { pr_err("[%s] number of clk levels is invalid!\n", __func__); return -EINVAL; @@ -390,12 +389,12 @@ arcturus_set_single_dpm_table(struct smu_context *smu, for (i = 0; i < num_of_levels; i++) { ret = smu_send_smc_msg_with_param(smu, SMU_MSG_GetDpmFreqByIndex, - (clk_id << 16 | i)); + (clk_id << 16 | i), + &clk); if (ret) { pr_err("[%s] failed to get dpm freq by index!\n", __func__); return ret; } - smu_read_smc_arg(smu, &clk); if (!clk) { pr_err("[%s] clk value is invalid!\n", __func__); return -EINVAL; @@ -553,13 +552,13 @@ static int arcturus_run_btc(struct smu_context *smu) { int ret = 0; - ret = smu_send_smc_msg(smu, SMU_MSG_RunAfllBtc); + ret = smu_send_smc_msg(smu, SMU_MSG_RunAfllBtc, NULL); if (ret) { pr_err("RunAfllBtc failed!\n"); return ret; } - return smu_send_smc_msg(smu, SMU_MSG_RunDcBtc); + return smu_send_smc_msg(smu, SMU_MSG_RunDcBtc, NULL); } static int arcturus_populate_umd_state_clk(struct smu_context *smu) @@ -744,7 +743,8 @@ static int arcturus_upload_dpm_level(struct smu_context *smu, bool max, single_dpm_table->dpm_state.soft_min_level; ret = smu_send_smc_msg_with_param(smu, (max ? SMU_MSG_SetSoftMaxByFreq : SMU_MSG_SetSoftMinByFreq), - (PPCLK_GFXCLK << 16) | (freq & 0xffff)); + (PPCLK_GFXCLK << 16) | (freq & 0xffff), + NULL); if (ret) { pr_err("Failed to set soft %s gfxclk !\n", max ? "max" : "min"); @@ -759,7 +759,8 @@ static int arcturus_upload_dpm_level(struct smu_context *smu, bool max, single_dpm_table->dpm_state.soft_min_level; ret = smu_send_smc_msg_with_param(smu, (max ? SMU_MSG_SetSoftMaxByFreq : SMU_MSG_SetSoftMinByFreq), - (PPCLK_UCLK << 16) | (freq & 0xffff)); + (PPCLK_UCLK << 16) | (freq & 0xffff), + NULL); if (ret) { pr_err("Failed to set soft %s memclk !\n", max ? "max" : "min"); @@ -774,7 +775,8 @@ static int arcturus_upload_dpm_level(struct smu_context *smu, bool max, single_dpm_table->dpm_state.soft_min_level; ret = smu_send_smc_msg_with_param(smu, (max ? SMU_MSG_SetSoftMaxByFreq : SMU_MSG_SetSoftMinByFreq), - (PPCLK_SOCCLK << 16) | (freq & 0xffff)); + (PPCLK_SOCCLK << 16) | (freq & 0xffff), + NULL); if (ret) { pr_err("Failed to set soft %s socclk !\n", max ? "max" : "min"); @@ -1289,12 +1291,11 @@ static int arcturus_get_power_limit(struct smu_context *smu, return -EINVAL; ret = smu_send_smc_msg_with_param(smu, SMU_MSG_GetPptLimit, - power_src << 16); + power_src << 16, &asic_default_power_limit); if (ret) { pr_err("[%s] get PPT limit failed!", __func__); return ret; } - smu_read_smc_arg(smu, &asic_default_power_limit); } else { /* the last hope to figure out the ppt limit */ if (!pptable) { @@ -1498,7 +1499,8 @@ static int arcturus_set_power_profile_mode(struct smu_context *smu, ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetWorkloadMask, - 1 << workload_type); + 1 << workload_type, + NULL); if (ret) { pr_err("Fail to set workload type %d\n", workload_type); return ret; @@ -2188,7 +2190,7 @@ static int arcturus_i2c_eeprom_control_init(struct i2c_adapter *control) control->class = I2C_CLASS_SPD; control->dev.parent = &adev->pdev->dev; control->algo = &arcturus_i2c_eeprom_i2c_algo; - snprintf(control->name, sizeof(control->name), "RAS EEPROM"); + snprintf(control->name, sizeof(control->name), "AMDGPU EEPROM"); res = i2c_add_adapter(control); if (res) @@ -2233,7 +2235,7 @@ static int arcturus_set_df_cstate(struct smu_context *smu, return -EINVAL; } - return smu_send_smc_msg_with_param(smu, SMU_MSG_DFCstateControl, state); + return smu_send_smc_msg_with_param(smu, SMU_MSG_DFCstateControl, state, NULL); } static const struct pptable_funcs arcturus_ppt_funcs = { @@ -2299,7 +2301,6 @@ static const struct pptable_funcs arcturus_ppt_funcs = { .notify_memory_pool_location = smu_v11_0_notify_memory_pool_location, .system_features_control = smu_v11_0_system_features_control, .send_smc_msg_with_param = smu_v11_0_send_msg_with_param, - .read_smc_arg = smu_v11_0_read_arg, .init_display_count = smu_v11_0_init_display_count, .set_allowed_mask = smu_v11_0_set_allowed_mask, .get_enabled_mask = smu_v11_0_get_enabled_mask, diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c index bf04cfefb283..7740488999df 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c @@ -1250,7 +1250,7 @@ static void smu7_set_dpm_event_sources(struct pp_hwmgr *hwmgr, uint32_t sources) switch (sources) { default: pr_err("Unknown throttling event sources."); - /* fall through */ + fallthrough; case 0: protection = false; /* src is unused */ @@ -3698,12 +3698,12 @@ static int smu7_request_link_speed_change_before_state_change( data->force_pcie_gen = PP_PCIEGen2; if (current_link_speed == PP_PCIEGen2) break; - /* fall through */ + fallthrough; case PP_PCIEGen2: if (0 == amdgpu_acpi_pcie_performance_request(hwmgr->adev, PCIE_PERF_REQ_GEN2, false)) break; + fallthrough; #endif - /* fall through */ default: data->force_pcie_gen = smu7_get_current_pcie_speed(hwmgr); break; diff --git a/drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h b/drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h index 97b6714e83e6..ae2c318dd6fa 100644 --- a/drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h +++ b/drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h @@ -362,6 +362,7 @@ struct smu_context struct mutex mutex; struct mutex sensor_lock; struct mutex metrics_lock; + struct mutex message_lock; uint64_t pool_size; struct smu_table_context smu_table; @@ -371,6 +372,9 @@ struct smu_context struct amd_pp_display_configuration *display_config; struct smu_baco_context smu_baco; void *od_settings; +#if defined(CONFIG_DEBUG_FS) + struct dentry *debugfs_sclk; +#endif uint32_t pstate_sclk; uint32_t pstate_mclk; @@ -404,6 +408,7 @@ struct smu_context uint32_t smc_if_version; bool uploading_custom_pp_table; + bool dc_controlled_by_gpio; }; struct i2c_adapter; @@ -514,8 +519,7 @@ struct pptable_funcs { int (*set_last_dcef_min_deep_sleep_clk)(struct smu_context *smu); int (*system_features_control)(struct smu_context *smu, bool en); int (*send_smc_msg_with_param)(struct smu_context *smu, - enum smu_message_type msg, uint32_t param); - int (*read_smc_arg)(struct smu_context *smu, uint32_t *arg); + enum smu_message_type msg, uint32_t param, uint32_t *read_arg); int (*init_display_count)(struct smu_context *smu, uint32_t count); int (*set_allowed_mask)(struct smu_context *smu); int (*get_enabled_mask)(struct smu_context *smu, uint32_t *feature_mask, uint32_t num); @@ -567,6 +571,7 @@ struct pptable_funcs { int (*override_pcie_parameters)(struct smu_context *smu); uint32_t (*get_pptable_power_limit)(struct smu_context *smu); int (*disable_umc_cdr_12gbps_workaround)(struct smu_context *smu); + int (*set_power_source)(struct smu_context *smu, enum smu_power_src_type power_src); }; int smu_load_microcode(struct smu_context *smu); @@ -707,7 +712,7 @@ int smu_get_dpm_level_count(struct smu_context *smu, enum smu_clk_type clk_type, int smu_get_dpm_freq_range(struct smu_context *smu, enum smu_clk_type clk_type, uint32_t *min, uint32_t *max, bool lock_needed); int smu_set_soft_freq_range(struct smu_context *smu, enum smu_clk_type clk_type, - uint32_t min, uint32_t max); + uint32_t min, uint32_t max, bool lock_needed); int smu_set_hard_freq_range(struct smu_context *smu, enum smu_clk_type clk_type, uint32_t min, uint32_t max); int smu_get_dpm_level_range(struct smu_context *smu, enum smu_clk_type clk_type, @@ -715,6 +720,7 @@ int smu_get_dpm_level_range(struct smu_context *smu, enum smu_clk_type clk_type, enum amd_dpm_forced_level smu_get_performance_level(struct smu_context *smu); int smu_force_performance_level(struct smu_context *smu, enum amd_dpm_forced_level level); int smu_set_display_count(struct smu_context *smu, uint32_t count); +int smu_set_ac_dc(struct smu_context *smu); bool smu_clk_dpm_is_enabled(struct smu_context *smu, enum smu_clk_type clk_type); const char *smu_get_message_name(struct smu_context *smu, enum smu_message_type type); const char *smu_get_feature_name(struct smu_context *smu, enum smu_feature_mask feature); diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu_v11_0.h b/drivers/gpu/drm/amd/powerplay/inc/smu_v11_0.h index acccdf621b4e..674e426ed59b 100644 --- a/drivers/gpu/drm/amd/powerplay/inc/smu_v11_0.h +++ b/drivers/gpu/drm/amd/powerplay/inc/smu_v11_0.h @@ -29,6 +29,7 @@ #define SMU11_DRIVER_IF_VERSION_VG20 0x13 #define SMU11_DRIVER_IF_VERSION_ARCT 0x12 #define SMU11_DRIVER_IF_VERSION_NV10 0x35 +#define SMU11_DRIVER_IF_VERSION_NV12 0x33 #define SMU11_DRIVER_IF_VERSION_NV14 0x36 /* MP Apertures */ @@ -182,9 +183,8 @@ int smu_v11_0_system_features_control(struct smu_context *smu, int smu_v11_0_send_msg_with_param(struct smu_context *smu, enum smu_message_type msg, - uint32_t param); - -int smu_v11_0_read_arg(struct smu_context *smu, uint32_t *arg); + uint32_t param, + uint32_t *read_arg); int smu_v11_0_init_display_count(struct smu_context *smu, uint32_t count); @@ -267,4 +267,7 @@ uint32_t smu_v11_0_get_max_power_limit(struct smu_context *smu); int smu_v11_0_set_performance_level(struct smu_context *smu, enum amd_dpm_forced_level level); +int smu_v11_0_set_power_source(struct smu_context *smu, + enum smu_power_src_type power_src); + #endif diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu_v12_0.h b/drivers/gpu/drm/amd/powerplay/inc/smu_v12_0.h index d79e54b5ebf6..7fbebc1979cf 100644 --- a/drivers/gpu/drm/amd/powerplay/inc/smu_v12_0.h +++ b/drivers/gpu/drm/amd/powerplay/inc/smu_v12_0.h @@ -40,14 +40,13 @@ struct smu_12_0_cmn2aisc_mapping { int smu_v12_0_send_msg_without_waiting(struct smu_context *smu, uint16_t msg); -int smu_v12_0_read_arg(struct smu_context *smu, uint32_t *arg); - int smu_v12_0_wait_for_response(struct smu_context *smu); int smu_v12_0_send_msg_with_param(struct smu_context *smu, enum smu_message_type msg, - uint32_t param); + uint32_t param, + uint32_t *read_arg); int smu_v12_0_check_fw_status(struct smu_context *smu); diff --git a/drivers/gpu/drm/amd/powerplay/navi10_ppt.c b/drivers/gpu/drm/amd/powerplay/navi10_ppt.c index 0d73a49166af..9c60b38ab53a 100644 --- a/drivers/gpu/drm/amd/powerplay/navi10_ppt.c +++ b/drivers/gpu/drm/amd/powerplay/navi10_ppt.c @@ -21,7 +21,6 @@ * */ -#include "pp_debug.h" #include <linux/firmware.h> #include <linux/pci.h> #include "amdgpu.h" @@ -31,7 +30,6 @@ #include "amdgpu_atomfirmware.h" #include "smu_v11_0.h" #include "smu11_driver_if_navi10.h" -#include "soc15_common.h" #include "atom.h" #include "navi10_ppt.h" #include "smu_v11_0_pptable.h" @@ -349,7 +347,6 @@ navi10_get_allowed_feature_mask(struct smu_context *smu, | FEATURE_MASK(FEATURE_DS_DCEFCLK_BIT) | FEATURE_MASK(FEATURE_FW_DSTATE_BIT) | FEATURE_MASK(FEATURE_BACO_BIT) - | FEATURE_MASK(FEATURE_ACDC_BIT) | FEATURE_MASK(FEATURE_GFX_SS_BIT) | FEATURE_MASK(FEATURE_APCC_DFLL_BIT) | FEATURE_MASK(FEATURE_FW_CTF_BIT) @@ -393,6 +390,9 @@ navi10_get_allowed_feature_mask(struct smu_context *smu, if (smu->adev->pg_flags & AMD_PG_SUPPORT_JPEG) *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_JPEG_PG_BIT); + if (smu->dc_controlled_by_gpio) + *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_ACDC_BIT); + /* disable DPM UCLK and DS SOCCLK on navi10 A0 secure board */ if (is_asic_secure(smu)) { /* only for navi10 A0 */ @@ -527,6 +527,9 @@ static int navi10_store_powerplay_table(struct smu_context *smu) table_context->thermal_controller_type = powerplay_table->thermal_controller_type; + if (powerplay_table->platform_caps & SMU_11_0_PP_PLATFORM_CAP_HARDWAREDC) + smu->dc_controlled_by_gpio = true; + mutex_lock(&smu_baco->mutex); if (powerplay_table->platform_caps & SMU_11_0_PP_PLATFORM_CAP_BACO || powerplay_table->platform_caps & SMU_11_0_PP_PLATFORM_CAP_MACO) @@ -661,14 +664,14 @@ static int navi10_dpm_set_uvd_enable(struct smu_context *smu, bool enable) if (enable) { /* vcn dpm on is a prerequisite for vcn power gate messages */ if (smu_feature_is_enabled(smu, SMU_FEATURE_VCN_PG_BIT)) { - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_PowerUpVcn, 1); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_PowerUpVcn, 1, NULL); if (ret) return ret; } power_gate->vcn_gated = false; } else { if (smu_feature_is_enabled(smu, SMU_FEATURE_VCN_PG_BIT)) { - ret = smu_send_smc_msg(smu, SMU_MSG_PowerDownVcn); + ret = smu_send_smc_msg(smu, SMU_MSG_PowerDownVcn, NULL); if (ret) return ret; } @@ -686,14 +689,14 @@ static int navi10_dpm_set_jpeg_enable(struct smu_context *smu, bool enable) if (enable) { if (smu_feature_is_enabled(smu, SMU_FEATURE_JPEG_PG_BIT)) { - ret = smu_send_smc_msg(smu, SMU_MSG_PowerUpJpeg); + ret = smu_send_smc_msg(smu, SMU_MSG_PowerUpJpeg, NULL); if (ret) return ret; } power_gate->jpeg_gated = false; } else { if (smu_feature_is_enabled(smu, SMU_FEATURE_JPEG_PG_BIT)) { - ret = smu_send_smc_msg(smu, SMU_MSG_PowerDownJpeg); + ret = smu_send_smc_msg(smu, SMU_MSG_PowerDownJpeg, NULL); if (ret) return ret; } @@ -970,7 +973,7 @@ static int navi10_force_clk_levels(struct smu_context *smu, if (ret) return size; - ret = smu_set_soft_freq_range(smu, clk_type, min_freq, max_freq); + ret = smu_set_soft_freq_range(smu, clk_type, min_freq, max_freq, false); if (ret) return size; break; @@ -1042,7 +1045,7 @@ static int navi10_pre_display_config_changed(struct smu_context *smu) int ret = 0; uint32_t max_freq = 0; - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_NumOfDisplays, 0); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_NumOfDisplays, 0, NULL); if (ret) return ret; @@ -1063,19 +1066,11 @@ static int navi10_display_config_changed(struct smu_context *smu) int ret = 0; if ((smu->watermarks_bitmap & WATERMARKS_EXIST) && - !(smu->watermarks_bitmap & WATERMARKS_LOADED)) { - ret = smu_write_watermarks_table(smu); - if (ret) - return ret; - - smu->watermarks_bitmap |= WATERMARKS_LOADED; - } - - if ((smu->watermarks_bitmap & WATERMARKS_EXIST) && smu_feature_is_supported(smu, SMU_FEATURE_DPM_DCEFCLK_BIT) && smu_feature_is_supported(smu, SMU_FEATURE_DPM_SOCCLK_BIT)) { ret = smu_send_smc_msg_with_param(smu, SMU_MSG_NumOfDisplays, - smu->display_config->num_display); + smu->display_config->num_display, + NULL); if (ret) return ret; } @@ -1102,7 +1097,7 @@ static int navi10_force_dpm_limit_value(struct smu_context *smu, bool highest) return ret; force_freq = highest ? max_freq : min_freq; - ret = smu_set_soft_freq_range(smu, clk_type, force_freq, force_freq); + ret = smu_set_soft_freq_range(smu, clk_type, force_freq, force_freq, false); if (ret) return ret; } @@ -1128,7 +1123,7 @@ static int navi10_unforce_dpm_levels(struct smu_context *smu) if (ret) return ret; - ret = smu_set_soft_freq_range(smu, clk_type, min_freq, max_freq); + ret = smu_set_soft_freq_range(smu, clk_type, min_freq, max_freq, false); if (ret) return ret; } @@ -1400,7 +1395,7 @@ static int navi10_set_power_profile_mode(struct smu_context *smu, long *input, u if (workload_type < 0) return -EINVAL; smu_send_smc_msg_with_param(smu, SMU_MSG_SetWorkloadMask, - 1 << workload_type); + 1 << workload_type, NULL); return ret; } @@ -1465,7 +1460,8 @@ static int navi10_notify_smc_display_config(struct smu_context *smu) if (smu_feature_is_supported(smu, SMU_FEATURE_DS_DCEFCLK_BIT)) { ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetMinDeepSleepDcefclk, - min_clocks.dcef_clock_in_sr/100); + min_clocks.dcef_clock_in_sr/100, + NULL); if (ret) { pr_err("Attempt to set divider for DCEFCLK Failed!"); return ret; @@ -1493,6 +1489,7 @@ static int navi10_set_watermarks_table(struct smu_context *smu, *clock_ranges) { int i; + int ret = 0; Watermarks_t *table = watermarks; if (!table || !clock_ranges) @@ -1544,6 +1541,18 @@ static int navi10_set_watermarks_table(struct smu_context *smu, clock_ranges->wm_mcif_clocks_ranges[i].wm_set_id; } + smu->watermarks_bitmap |= WATERMARKS_EXIST; + + /* pass data to smu controller */ + if (!(smu->watermarks_bitmap & WATERMARKS_LOADED)) { + ret = smu_write_watermarks_table(smu); + if (ret) { + pr_err("Failed to update WMTABLE!"); + return ret; + } + smu->watermarks_bitmap |= WATERMARKS_LOADED; + } + return 0; } @@ -1674,10 +1683,10 @@ static int navi10_set_standard_performance_level(struct smu_context *smu) return navi10_set_performance_level(smu, AMD_DPM_FORCED_LEVEL_AUTO); } - ret = smu_set_soft_freq_range(smu, SMU_SCLK, sclk_freq, sclk_freq); + ret = smu_set_soft_freq_range(smu, SMU_SCLK, sclk_freq, sclk_freq, false); if (ret) return ret; - ret = smu_set_soft_freq_range(smu, SMU_UCLK, uclk_freq, uclk_freq); + ret = smu_set_soft_freq_range(smu, SMU_UCLK, uclk_freq, uclk_freq, false); if (ret) return ret; @@ -1742,10 +1751,10 @@ static int navi10_set_peak_performance_level(struct smu_context *smu) if (ret) return ret; - ret = smu_set_soft_freq_range(smu, SMU_SCLK, sclk_freq, sclk_freq); + ret = smu_set_soft_freq_range(smu, SMU_SCLK, sclk_freq, sclk_freq, false); if (ret) return ret; - ret = smu_set_soft_freq_range(smu, SMU_UCLK, uclk_freq, uclk_freq); + ret = smu_set_soft_freq_range(smu, SMU_UCLK, uclk_freq, uclk_freq, false); if (ret) return ret; @@ -1855,12 +1864,11 @@ static int navi10_get_power_limit(struct smu_context *smu, return -EINVAL; ret = smu_send_smc_msg_with_param(smu, SMU_MSG_GetPptLimit, - power_src << 16); + power_src << 16, &asic_default_power_limit); if (ret) { pr_err("[%s] get PPT limit failed!", __func__); return ret; } - smu_read_smc_arg(smu, &asic_default_power_limit); } else { /* the last hope to figure out the ppt limit */ if (!pptable) { @@ -1900,7 +1908,8 @@ static int navi10_update_pcie_parameters(struct smu_context *smu, pptable->PcieLaneCount[i] : pcie_width_cap); ret = smu_send_smc_msg_with_param(smu, SMU_MSG_OverridePcieParameters, - smu_pcie_arg); + smu_pcie_arg, + NULL); if (ret) return ret; @@ -1946,13 +1955,13 @@ static int navi10_overdrive_get_gfx_clk_base_voltage(struct smu_context *smu, ret = smu_send_smc_msg_with_param(smu, SMU_MSG_GetVoltageByDpm, - param); + param, + &value); if (ret) { pr_err("[GetBaseVoltage] failed to get GFXCLK AVFS voltage from SMU!"); return ret; } - smu_read_smc_arg(smu, &value); *voltage = (uint16_t)value; return 0; @@ -2209,7 +2218,7 @@ static int navi10_run_btc(struct smu_context *smu) { int ret = 0; - ret = smu_send_smc_msg(smu, SMU_MSG_RunBtc); + ret = smu_send_smc_msg(smu, SMU_MSG_RunBtc, NULL); if (ret) pr_err("RunBtc failed!\n"); @@ -2221,9 +2230,9 @@ static int navi10_dummy_pstate_control(struct smu_context *smu, bool enable) int result = 0; if (!enable) - result = smu_send_smc_msg(smu, SMU_MSG_DAL_DISABLE_DUMMY_PSTATE_CHANGE); + result = smu_send_smc_msg(smu, SMU_MSG_DAL_DISABLE_DUMMY_PSTATE_CHANGE, NULL); else - result = smu_send_smc_msg(smu, SMU_MSG_DAL_ENABLE_DUMMY_PSTATE_CHANGE); + result = smu_send_smc_msg(smu, SMU_MSG_DAL_ENABLE_DUMMY_PSTATE_CHANGE, NULL); return result; } @@ -2332,7 +2341,6 @@ static const struct pptable_funcs navi10_ppt_funcs = { .notify_memory_pool_location = smu_v11_0_notify_memory_pool_location, .system_features_control = smu_v11_0_system_features_control, .send_smc_msg_with_param = smu_v11_0_send_msg_with_param, - .read_smc_arg = smu_v11_0_read_arg, .init_display_count = smu_v11_0_init_display_count, .set_allowed_mask = smu_v11_0_set_allowed_mask, .get_enabled_mask = smu_v11_0_get_enabled_mask, @@ -2366,6 +2374,7 @@ static const struct pptable_funcs navi10_ppt_funcs = { .get_pptable_power_limit = navi10_get_pptable_power_limit, .run_btc = navi10_run_btc, .disable_umc_cdr_12gbps_workaround = navi10_disable_umc_cdr_12gbps_workaround, + .set_power_source = smu_v11_0_set_power_source, }; void navi10_set_ppt_funcs(struct smu_context *smu) diff --git a/drivers/gpu/drm/amd/powerplay/renoir_ppt.c b/drivers/gpu/drm/amd/powerplay/renoir_ppt.c index 810994017f49..7bf52ecba01d 100644 --- a/drivers/gpu/drm/amd/powerplay/renoir_ppt.c +++ b/drivers/gpu/drm/amd/powerplay/renoir_ppt.c @@ -24,7 +24,6 @@ #include "amdgpu.h" #include "amdgpu_smu.h" #include "smu_internal.h" -#include "soc15_common.h" #include "smu_v12_0_ppsmc.h" #include "smu12_driver_if.h" #include "smu_v12_0.h" @@ -111,8 +110,8 @@ static struct smu_12_0_cmn2aisc_mapping renoir_clk_map[SMU_CLK_COUNT] = { CLK_MAP(GFXCLK, CLOCK_GFXCLK), CLK_MAP(SCLK, CLOCK_GFXCLK), CLK_MAP(SOCCLK, CLOCK_SOCCLK), - CLK_MAP(UCLK, CLOCK_UMCCLK), - CLK_MAP(MCLK, CLOCK_UMCCLK), + CLK_MAP(UCLK, CLOCK_FCLK), + CLK_MAP(MCLK, CLOCK_FCLK), }; static struct smu_12_0_cmn2aisc_mapping renoir_table_map[SMU_TABLE_COUNT] = { @@ -280,7 +279,7 @@ static int renoir_print_clk_levels(struct smu_context *smu, break; case SMU_MCLK: count = NUM_MEMCLK_DPM_LEVELS; - cur_value = metrics.ClockFrequency[CLOCK_UMCCLK]; + cur_value = metrics.ClockFrequency[CLOCK_FCLK]; break; case SMU_DCEFCLK: count = NUM_DCFCLK_DPM_LEVELS; @@ -342,14 +341,14 @@ static int renoir_dpm_set_uvd_enable(struct smu_context *smu, bool enable) if (enable) { /* vcn dpm on is a prerequisite for vcn power gate messages */ if (smu_feature_is_enabled(smu, SMU_FEATURE_VCN_PG_BIT)) { - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_PowerUpVcn, 0); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_PowerUpVcn, 0, NULL); if (ret) return ret; } power_gate->vcn_gated = false; } else { if (smu_feature_is_enabled(smu, SMU_FEATURE_VCN_PG_BIT)) { - ret = smu_send_smc_msg(smu, SMU_MSG_PowerDownVcn); + ret = smu_send_smc_msg(smu, SMU_MSG_PowerDownVcn, NULL); if (ret) return ret; } @@ -367,14 +366,14 @@ static int renoir_dpm_set_jpeg_enable(struct smu_context *smu, bool enable) if (enable) { if (smu_feature_is_enabled(smu, SMU_FEATURE_JPEG_PG_BIT)) { - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_PowerUpJpeg, 0); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_PowerUpJpeg, 0, NULL); if (ret) return ret; } power_gate->jpeg_gated = false; } else { if (smu_feature_is_enabled(smu, SMU_FEATURE_JPEG_PG_BIT)) { - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_PowerDownJpeg, 0); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_PowerDownJpeg, 0, NULL); if (ret) return ret; } @@ -423,7 +422,7 @@ static int renoir_force_dpm_limit_value(struct smu_context *smu, bool highest) return ret; force_freq = highest ? max_freq : min_freq; - ret = smu_set_soft_freq_range(smu, clk_type, force_freq, force_freq); + ret = smu_set_soft_freq_range(smu, clk_type, force_freq, force_freq, false); if (ret) return ret; } @@ -456,7 +455,7 @@ static int renoir_unforce_dpm_levels(struct smu_context *smu) { if (ret) return ret; - ret = smu_set_soft_freq_range(smu, clk_type, min_freq, max_freq); + ret = smu_set_soft_freq_range(smu, clk_type, min_freq, max_freq, false); if (ret) return ret; } @@ -622,22 +621,24 @@ static int renoir_force_clk_levels(struct smu_context *smu, return ret; ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxGfxClk, soft_max_level == 0 ? min_freq : - soft_max_level == 1 ? RENOIR_UMD_PSTATE_GFXCLK : max_freq); + soft_max_level == 1 ? RENOIR_UMD_PSTATE_GFXCLK : max_freq, + NULL); if (ret) return ret; ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinGfxClk, soft_min_level == 2 ? max_freq : - soft_min_level == 1 ? RENOIR_UMD_PSTATE_GFXCLK : min_freq); + soft_min_level == 1 ? RENOIR_UMD_PSTATE_GFXCLK : min_freq, + NULL); if (ret) return ret; break; case SMU_SOCCLK: GET_DPM_CUR_FREQ(clk_table, clk_type, soft_min_level, min_freq); GET_DPM_CUR_FREQ(clk_table, clk_type, soft_max_level, max_freq); - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxSocclkByFreq, max_freq); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxSocclkByFreq, max_freq, NULL); if (ret) return ret; - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinSocclkByFreq, min_freq); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinSocclkByFreq, min_freq, NULL); if (ret) return ret; break; @@ -645,10 +646,10 @@ static int renoir_force_clk_levels(struct smu_context *smu, case SMU_FCLK: GET_DPM_CUR_FREQ(clk_table, clk_type, soft_min_level, min_freq); GET_DPM_CUR_FREQ(clk_table, clk_type, soft_max_level, max_freq); - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxFclkByFreq, max_freq); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxFclkByFreq, max_freq, NULL); if (ret) return ret; - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinFclkByFreq, min_freq); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinFclkByFreq, min_freq, NULL); if (ret) return ret; break; @@ -681,7 +682,8 @@ static int renoir_set_power_profile_mode(struct smu_context *smu, long *input, u } ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetWorkloadMask, - 1 << workload_type); + 1 << workload_type, + NULL); if (ret) { pr_err_once("Fail to set workload type %d\n", workload_type); return ret; @@ -701,7 +703,7 @@ static int renoir_set_peak_clock_by_device(struct smu_context *smu) if (ret) return ret; - ret = smu_set_soft_freq_range(smu, SMU_SCLK, sclk_freq, sclk_freq); + ret = smu_set_soft_freq_range(smu, SMU_SCLK, sclk_freq, sclk_freq, false); if (ret) return ret; @@ -709,7 +711,7 @@ static int renoir_set_peak_clock_by_device(struct smu_context *smu) if (ret) return ret; - ret = smu_set_soft_freq_range(smu, SMU_UCLK, uclk_freq, uclk_freq); + ret = smu_set_soft_freq_range(smu, SMU_UCLK, uclk_freq, uclk_freq, false); if (ret) return ret; @@ -810,9 +812,10 @@ static int renoir_set_watermarks_table( clock_ranges->wm_mcif_clocks_ranges[i].wm_set_id; } + smu->watermarks_bitmap |= WATERMARKS_EXIST; + /* pass data to smu controller */ - if ((smu->watermarks_bitmap & WATERMARKS_EXIST) && - !(smu->watermarks_bitmap & WATERMARKS_LOADED)) { + if (!(smu->watermarks_bitmap & WATERMARKS_LOADED)) { ret = smu_write_watermarks_table(smu); if (ret) { pr_err("Failed to update WMTABLE!"); @@ -913,7 +916,6 @@ static const struct pptable_funcs renoir_ppt_funcs = { .powergate_vcn = smu_v12_0_powergate_vcn, .powergate_jpeg = smu_v12_0_powergate_jpeg, .send_smc_msg_with_param = smu_v12_0_send_msg_with_param, - .read_smc_arg = smu_v12_0_read_arg, .set_gfx_cgpg = smu_v12_0_set_gfx_cgpg, .gfx_off_control = smu_v12_0_gfx_off_control, .init_smc_tables = smu_v12_0_init_smc_tables, diff --git a/drivers/gpu/drm/amd/powerplay/smu_internal.h b/drivers/gpu/drm/amd/powerplay/smu_internal.h index 7bd200ffcda8..40c35bcc5a0a 100644 --- a/drivers/gpu/drm/amd/powerplay/smu_internal.h +++ b/drivers/gpu/drm/amd/powerplay/smu_internal.h @@ -79,12 +79,13 @@ #define smu_set_default_od_settings(smu, initialize) \ ((smu)->ppt_funcs->set_default_od_settings ? (smu)->ppt_funcs->set_default_od_settings((smu), (initialize)) : 0) -int smu_send_smc_msg(struct smu_context *smu, enum smu_message_type msg); +#define smu_send_smc_msg_with_param(smu, msg, param, read_arg) \ + ((smu)->ppt_funcs->send_smc_msg_with_param? (smu)->ppt_funcs->send_smc_msg_with_param((smu), (msg), (param), (read_arg)) : 0) + +static inline int smu_send_smc_msg(struct smu_context *smu, enum smu_message_type msg, uint32_t *read_arg) { + return smu_send_smc_msg_with_param(smu, msg, 0, read_arg); +} -#define smu_send_smc_msg_with_param(smu, msg, param) \ - ((smu)->ppt_funcs->send_smc_msg_with_param? (smu)->ppt_funcs->send_smc_msg_with_param((smu), (msg), (param)) : 0) -#define smu_read_smc_arg(smu, arg) \ - ((smu)->ppt_funcs->read_smc_arg? (smu)->ppt_funcs->read_smc_arg((smu), (arg)) : 0) #define smu_alloc_dpm_context(smu) \ ((smu)->ppt_funcs->alloc_dpm_context ? (smu)->ppt_funcs->alloc_dpm_context((smu)) : 0) #define smu_init_display_count(smu, count) \ @@ -210,4 +211,7 @@ int smu_send_smc_msg(struct smu_context *smu, enum smu_message_type msg); #define smu_disable_umc_cdr_12gbps_workaround(smu) \ ((smu)->ppt_funcs->disable_umc_cdr_12gbps_workaround ? (smu)->ppt_funcs->disable_umc_cdr_12gbps_workaround((smu)) : 0) +#define smu_set_power_source(smu, power_src) \ + ((smu)->ppt_funcs->set_power_source ? (smu)->ppt_funcs->set_power_source((smu), (power_src)) : 0) + #endif diff --git a/drivers/gpu/drm/amd/powerplay/smu_v11_0.c b/drivers/gpu/drm/amd/powerplay/smu_v11_0.c index c9e5ce135fd4..d19e1d0d56c0 100644 --- a/drivers/gpu/drm/amd/powerplay/smu_v11_0.c +++ b/drivers/gpu/drm/amd/powerplay/smu_v11_0.c @@ -26,7 +26,6 @@ #define SMU_11_0_PARTIAL_PPTABLE -#include "pp_debug.h" #include "amdgpu.h" #include "amdgpu_smu.h" #include "smu_internal.h" @@ -64,7 +63,7 @@ static int smu_v11_0_send_msg_without_waiting(struct smu_context *smu, return 0; } -int smu_v11_0_read_arg(struct smu_context *smu, uint32_t *arg) +static int smu_v11_0_read_arg(struct smu_context *smu, uint32_t *arg) { struct amdgpu_device *adev = smu->adev; @@ -92,7 +91,8 @@ static int smu_v11_0_wait_for_response(struct smu_context *smu) int smu_v11_0_send_msg_with_param(struct smu_context *smu, enum smu_message_type msg, - uint32_t param) + uint32_t param, + uint32_t *read_arg) { struct amdgpu_device *adev = smu->adev; int ret = 0, index = 0; @@ -101,11 +101,12 @@ smu_v11_0_send_msg_with_param(struct smu_context *smu, if (index < 0) return index; + mutex_lock(&smu->message_lock); ret = smu_v11_0_wait_for_response(smu); if (ret) { pr_err("Msg issuing pre-check failed and " "SMU may be not in the right state!\n"); - return ret; + goto out; } WREG32_SOC15(MP1, 0, mmMP1_SMN_C2PMSG_90, 0); @@ -115,10 +116,21 @@ smu_v11_0_send_msg_with_param(struct smu_context *smu, smu_v11_0_send_msg_without_waiting(smu, (uint16_t)index); ret = smu_v11_0_wait_for_response(smu); - if (ret) + if (ret) { pr_err("failed send message: %10s (%d) \tparam: 0x%08x response %#x\n", smu_get_message_name(smu, msg), index, param, ret); - + goto out; + } + if (read_arg) { + ret = smu_v11_0_read_arg(smu, read_arg); + if (ret) { + pr_err("failed to read message arg: %10s (%d) \tparam: 0x%08x response %#x\n", + smu_get_message_name(smu, msg), index, param, ret); + goto out; + } + } +out: + mutex_unlock(&smu->message_lock); return ret; } @@ -262,6 +274,9 @@ int smu_v11_0_check_fw_version(struct smu_context *smu) case CHIP_NAVI10: smu->smc_if_version = SMU11_DRIVER_IF_VERSION_NV10; break; + case CHIP_NAVI12: + smu->smc_if_version = SMU11_DRIVER_IF_VERSION_NV12; + break; case CHIP_NAVI14: smu->smc_if_version = SMU11_DRIVER_IF_VERSION_NV14; break; @@ -671,12 +686,14 @@ int smu_v11_0_notify_memory_pool_location(struct smu_context *smu) ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSystemVirtualDramAddrHigh, - address_high); + address_high, + NULL); if (ret) return ret; ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSystemVirtualDramAddrLow, - address_low); + address_low, + NULL); if (ret) return ret; @@ -685,15 +702,15 @@ int smu_v11_0_notify_memory_pool_location(struct smu_context *smu) address_low = (uint32_t)lower_32_bits(address); ret = smu_send_smc_msg_with_param(smu, SMU_MSG_DramLogSetDramAddrHigh, - address_high); + address_high, NULL); if (ret) return ret; ret = smu_send_smc_msg_with_param(smu, SMU_MSG_DramLogSetDramAddrLow, - address_low); + address_low, NULL); if (ret) return ret; ret = smu_send_smc_msg_with_param(smu, SMU_MSG_DramLogSetDramSize, - (uint32_t)memory_pool->size); + (uint32_t)memory_pool->size, NULL); if (ret) return ret; @@ -757,7 +774,7 @@ int smu_v11_0_set_deep_sleep_dcefclk(struct smu_context *smu, uint32_t clk) int ret; ret = smu_send_smc_msg_with_param(smu, - SMU_MSG_SetMinDeepSleepDcefclk, clk); + SMU_MSG_SetMinDeepSleepDcefclk, clk, NULL); if (ret) pr_err("SMU11 attempt to set divider for DCEFCLK Failed!"); @@ -784,11 +801,13 @@ int smu_v11_0_set_driver_table_location(struct smu_context *smu) if (driver_table->mc_address) { ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetDriverDramAddrHigh, - upper_32_bits(driver_table->mc_address)); + upper_32_bits(driver_table->mc_address), + NULL); if (!ret) ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetDriverDramAddrLow, - lower_32_bits(driver_table->mc_address)); + lower_32_bits(driver_table->mc_address), + NULL); } return ret; @@ -802,11 +821,13 @@ int smu_v11_0_set_tool_table_location(struct smu_context *smu) if (tool_table->mc_address) { ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetToolsDramAddrHigh, - upper_32_bits(tool_table->mc_address)); + upper_32_bits(tool_table->mc_address), + NULL); if (!ret) ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetToolsDramAddrLow, - lower_32_bits(tool_table->mc_address)); + lower_32_bits(tool_table->mc_address), + NULL); } return ret; @@ -819,7 +840,7 @@ int smu_v11_0_init_display_count(struct smu_context *smu, uint32_t count) if (!smu->pm_enabled) return ret; - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_NumOfDisplays, count); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_NumOfDisplays, count, NULL); return ret; } @@ -837,12 +858,12 @@ int smu_v11_0_set_allowed_mask(struct smu_context *smu) bitmap_copy((unsigned long *)feature_mask, feature->allowed, 64); ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetAllowedFeaturesMaskHigh, - feature_mask[1]); + feature_mask[1], NULL); if (ret) goto failed; ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetAllowedFeaturesMaskLow, - feature_mask[0]); + feature_mask[0], NULL); if (ret) goto failed; @@ -862,17 +883,11 @@ int smu_v11_0_get_enabled_mask(struct smu_context *smu, return -EINVAL; if (bitmap_empty(feature->enabled, feature->feature_num)) { - ret = smu_send_smc_msg(smu, SMU_MSG_GetEnabledSmuFeaturesHigh); - if (ret) - return ret; - ret = smu_read_smc_arg(smu, &feature_mask_high); + ret = smu_send_smc_msg(smu, SMU_MSG_GetEnabledSmuFeaturesHigh, &feature_mask_high); if (ret) return ret; - ret = smu_send_smc_msg(smu, SMU_MSG_GetEnabledSmuFeaturesLow); - if (ret) - return ret; - ret = smu_read_smc_arg(smu, &feature_mask_low); + ret = smu_send_smc_msg(smu, SMU_MSG_GetEnabledSmuFeaturesLow, &feature_mask_low); if (ret) return ret; @@ -894,7 +909,7 @@ int smu_v11_0_system_features_control(struct smu_context *smu, int ret = 0; ret = smu_send_smc_msg(smu, (en ? SMU_MSG_EnableAllSmuFeatures : - SMU_MSG_DisableAllSmuFeatures)); + SMU_MSG_DisableAllSmuFeatures), NULL); if (ret) return ret; @@ -923,7 +938,7 @@ int smu_v11_0_notify_display_change(struct smu_context *smu) return ret; if (smu_feature_is_enabled(smu, SMU_FEATURE_DPM_UCLK_BIT) && smu->adev->gmc.vram_type == AMDGPU_VRAM_TYPE_HBM) - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetUclkFastSwitch, 1); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetUclkFastSwitch, 1, NULL); return ret; } @@ -947,30 +962,24 @@ smu_v11_0_get_max_sustainable_clock(struct smu_context *smu, uint32_t *clock, return -EINVAL; ret = smu_send_smc_msg_with_param(smu, SMU_MSG_GetDcModeMaxDpmFreq, - clk_id << 16); + clk_id << 16, clock); if (ret) { pr_err("[GetMaxSustainableClock] Failed to get max DC clock from SMC!"); return ret; } - ret = smu_read_smc_arg(smu, clock); - if (ret) - return ret; - if (*clock != 0) return 0; /* if DC limit is zero, return AC limit */ ret = smu_send_smc_msg_with_param(smu, SMU_MSG_GetMaxDpmFreq, - clk_id << 16); + clk_id << 16, clock); if (ret) { pr_err("[GetMaxSustainableClock] failed to get max AC clock from SMC!"); return ret; } - ret = smu_read_smc_arg(smu, clock); - - return ret; + return 0; } int smu_v11_0_init_max_sustainable_clocks(struct smu_context *smu) @@ -1106,7 +1115,7 @@ int smu_v11_0_set_power_limit(struct smu_context *smu, uint32_t n) return -EOPNOTSUPP; } - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetPptLimit, n); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetPptLimit, n, NULL); if (ret) { pr_err("[%s] Set power limit Failed!\n", __func__); return ret; @@ -1136,11 +1145,7 @@ int smu_v11_0_get_current_clk_freq(struct smu_context *smu, ret = smu_get_current_clk_freq_by_table(smu, clk_id, &freq); else { ret = smu_send_smc_msg_with_param(smu, SMU_MSG_GetDpmClockFreq, - (asic_clk_id << 16)); - if (ret) - return ret; - - ret = smu_read_smc_arg(smu, &freq); + (asic_clk_id << 16), &freq); if (ret) return ret; } @@ -1375,9 +1380,9 @@ int smu_v11_0_gfx_off_control(struct smu_context *smu, bool enable) if (!(adev->pm.pp_feature & PP_GFXOFF_MASK)) return 0; if (enable) - ret = smu_send_smc_msg(smu, SMU_MSG_AllowGfxOff); + ret = smu_send_smc_msg(smu, SMU_MSG_AllowGfxOff, NULL); else - ret = smu_send_smc_msg(smu, SMU_MSG_DisallowGfxOff); + ret = smu_send_smc_msg(smu, SMU_MSG_DisallowGfxOff, NULL); break; default: break; @@ -1515,10 +1520,18 @@ int smu_v11_0_set_xgmi_pstate(struct smu_context *smu, int ret = 0; ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetXgmiMode, - pstate ? XGMI_MODE_PSTATE_D0 : XGMI_MODE_PSTATE_D3); + pstate ? XGMI_MODE_PSTATE_D0 : XGMI_MODE_PSTATE_D3, + NULL); return ret; } +static int smu_v11_0_ack_ac_dc_interrupt(struct smu_context *smu) +{ + return smu_send_smc_msg(smu, + SMU_MSG_ReenableAcDcInterrupt, + NULL); +} + #define THM_11_0__SRCID__THM_DIG_THERM_L2H 0 /* ASIC_TEMP > CG_THERMAL_INT.DIG_THERM_INTH */ #define THM_11_0__SRCID__THM_DIG_THERM_H2L 1 /* ASIC_TEMP < CG_THERMAL_INT.DIG_THERM_INTL */ @@ -1552,6 +1565,9 @@ static int smu_v11_0_irq_process(struct amdgpu_device *adev, break; } + } else if (client_id == SOC15_IH_CLIENTID_MP1) { + if (src_id == 0xfe) + smu_v11_0_ack_ac_dc_interrupt(&adev->smu); } return 0; @@ -1591,6 +1607,12 @@ int smu_v11_0_register_irq_handler(struct smu_context *smu) if (ret) return ret; + ret = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_MP1, + 0xfe, + irq_src); + if (ret) + return ret; + return ret; } @@ -1628,14 +1650,14 @@ int smu_v11_0_set_azalia_d3_pme(struct smu_context *smu) { int ret = 0; - ret = smu_send_smc_msg(smu, SMU_MSG_BacoAudioD3PME); + ret = smu_send_smc_msg(smu, SMU_MSG_BacoAudioD3PME, NULL); return ret; } static int smu_v11_0_baco_set_armd3_sequence(struct smu_context *smu, enum smu_v11_0_baco_seq baco_seq) { - return smu_send_smc_msg_with_param(smu, SMU_MSG_ArmD3, baco_seq); + return smu_send_smc_msg_with_param(smu, SMU_MSG_ArmD3, baco_seq, NULL); } bool smu_v11_0_baco_is_support(struct smu_context *smu) @@ -1704,12 +1726,12 @@ int smu_v11_0_baco_set_state(struct smu_context *smu, enum smu_baco_state state) data |= 0x80000000; WREG32_SOC15(THM, 0, mmTHM_BACO_CNTL, data); - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_EnterBaco, 0); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_EnterBaco, 0, NULL); } else { - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_EnterBaco, 1); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_EnterBaco, 1, NULL); } } else { - ret = smu_send_smc_msg(smu, SMU_MSG_ExitBaco); + ret = smu_send_smc_msg(smu, SMU_MSG_ExitBaco, NULL); if (ret) goto out; @@ -1777,19 +1799,13 @@ int smu_v11_0_get_dpm_ultimate_freq(struct smu_context *smu, enum smu_clk_type c param = (clk_id & 0xffff) << 16; if (max) { - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_GetMaxDpmFreq, param); - if (ret) - goto failed; - ret = smu_read_smc_arg(smu, max); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_GetMaxDpmFreq, param, max); if (ret) goto failed; } if (min) { - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_GetMinDpmFreq, param); - if (ret) - goto failed; - ret = smu_read_smc_arg(smu, min); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_GetMinDpmFreq, param, min); if (ret) goto failed; } @@ -1811,7 +1827,7 @@ int smu_v11_0_set_soft_freq_limited_range(struct smu_context *smu, enum smu_clk_ if (max > 0) { param = (uint32_t)((clk_id << 16) | (max & 0xffff)); ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxByFreq, - param); + param, NULL); if (ret) return ret; } @@ -1819,7 +1835,7 @@ int smu_v11_0_set_soft_freq_limited_range(struct smu_context *smu, enum smu_clk_ if (min > 0) { param = (uint32_t)((clk_id << 16) | (min & 0xffff)); ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMinByFreq, - param); + param, NULL); if (ret) return ret; } @@ -1939,3 +1955,18 @@ int smu_v11_0_set_performance_level(struct smu_context *smu, return ret; } +int smu_v11_0_set_power_source(struct smu_context *smu, + enum smu_power_src_type power_src) +{ + int pwr_source; + + pwr_source = smu_power_get_index(smu, (uint32_t)power_src); + if (pwr_source < 0) + return -EINVAL; + + return smu_send_smc_msg_with_param(smu, + SMU_MSG_NotifyPowerSource, + pwr_source, + NULL); +} + diff --git a/drivers/gpu/drm/amd/powerplay/smu_v12_0.c b/drivers/gpu/drm/amd/powerplay/smu_v12_0.c index 870e6db2907e..169ebdad87b8 100644 --- a/drivers/gpu/drm/amd/powerplay/smu_v12_0.c +++ b/drivers/gpu/drm/amd/powerplay/smu_v12_0.c @@ -20,7 +20,6 @@ * OTHER DEALINGS IN THE SOFTWARE. */ -#include "pp_debug.h" #include <linux/firmware.h> #include "amdgpu.h" #include "amdgpu_smu.h" @@ -50,7 +49,7 @@ int smu_v12_0_send_msg_without_waiting(struct smu_context *smu, return 0; } -int smu_v12_0_read_arg(struct smu_context *smu, uint32_t *arg) +static int smu_v12_0_read_arg(struct smu_context *smu, uint32_t *arg) { struct amdgpu_device *adev = smu->adev; @@ -78,7 +77,8 @@ int smu_v12_0_wait_for_response(struct smu_context *smu) int smu_v12_0_send_msg_with_param(struct smu_context *smu, enum smu_message_type msg, - uint32_t param) + uint32_t param, + uint32_t *read_arg) { struct amdgpu_device *adev = smu->adev; int ret = 0, index = 0; @@ -87,11 +87,12 @@ smu_v12_0_send_msg_with_param(struct smu_context *smu, if (index < 0) return index; + mutex_lock(&smu->message_lock); ret = smu_v12_0_wait_for_response(smu); if (ret) { pr_err("Msg issuing pre-check failed and " "SMU may be not in the right state!\n"); - return ret; + goto out; } WREG32_SOC15(MP1, 0, mmMP1_SMN_C2PMSG_90, 0); @@ -101,10 +102,21 @@ smu_v12_0_send_msg_with_param(struct smu_context *smu, smu_v12_0_send_msg_without_waiting(smu, (uint16_t)index); ret = smu_v12_0_wait_for_response(smu); - if (ret) + if (ret) { pr_err("Failed to send message 0x%x, response 0x%x param 0x%x\n", index, ret, param); - + goto out; + } + if (read_arg) { + ret = smu_v12_0_read_arg(smu, read_arg); + if (ret) { + pr_err("Failed to read message arg 0x%x, response 0x%x param 0x%x\n", + index, ret, param); + goto out; + } + } +out: + mutex_unlock(&smu->message_lock); return ret; } @@ -163,9 +175,9 @@ int smu_v12_0_powergate_sdma(struct smu_context *smu, bool gate) return 0; if (gate) - return smu_send_smc_msg(smu, SMU_MSG_PowerDownSdma); + return smu_send_smc_msg(smu, SMU_MSG_PowerDownSdma, NULL); else - return smu_send_smc_msg(smu, SMU_MSG_PowerUpSdma); + return smu_send_smc_msg(smu, SMU_MSG_PowerUpSdma, NULL); } int smu_v12_0_powergate_vcn(struct smu_context *smu, bool gate) @@ -174,9 +186,9 @@ int smu_v12_0_powergate_vcn(struct smu_context *smu, bool gate) return 0; if (gate) - return smu_send_smc_msg(smu, SMU_MSG_PowerDownVcn); + return smu_send_smc_msg(smu, SMU_MSG_PowerDownVcn, NULL); else - return smu_send_smc_msg(smu, SMU_MSG_PowerUpVcn); + return smu_send_smc_msg(smu, SMU_MSG_PowerUpVcn, NULL); } int smu_v12_0_powergate_jpeg(struct smu_context *smu, bool gate) @@ -185,9 +197,9 @@ int smu_v12_0_powergate_jpeg(struct smu_context *smu, bool gate) return 0; if (gate) - return smu_send_smc_msg_with_param(smu, SMU_MSG_PowerDownJpeg, 0); + return smu_send_smc_msg_with_param(smu, SMU_MSG_PowerDownJpeg, 0, NULL); else - return smu_send_smc_msg_with_param(smu, SMU_MSG_PowerUpJpeg, 0); + return smu_send_smc_msg_with_param(smu, SMU_MSG_PowerUpJpeg, 0, NULL); } int smu_v12_0_set_gfx_cgpg(struct smu_context *smu, bool enable) @@ -196,7 +208,9 @@ int smu_v12_0_set_gfx_cgpg(struct smu_context *smu, bool enable) return 0; return smu_v12_0_send_msg_with_param(smu, - SMU_MSG_SetGfxCGPG, enable ? 1 : 0); + SMU_MSG_SetGfxCGPG, + enable ? 1 : 0, + NULL); } int smu_v12_0_read_sensor(struct smu_context *smu, @@ -262,10 +276,10 @@ int smu_v12_0_gfx_off_control(struct smu_context *smu, bool enable) int ret = 0, timeout = 500; if (enable) { - ret = smu_send_smc_msg(smu, SMU_MSG_AllowGfxOff); + ret = smu_send_smc_msg(smu, SMU_MSG_AllowGfxOff, NULL); } else { - ret = smu_send_smc_msg(smu, SMU_MSG_DisallowGfxOff); + ret = smu_send_smc_msg(smu, SMU_MSG_DisallowGfxOff, NULL); /* confirm gfx is back to "on" state, timeout is 0.5 second */ while (!(smu_v12_0_get_gfxoff_status(smu) == 2)) { @@ -331,17 +345,11 @@ int smu_v12_0_get_enabled_mask(struct smu_context *smu, if (!feature_mask || num < 2) return -EINVAL; - ret = smu_send_smc_msg(smu, SMU_MSG_GetEnabledSmuFeaturesHigh); - if (ret) - return ret; - ret = smu_read_smc_arg(smu, &feature_mask_high); + ret = smu_send_smc_msg(smu, SMU_MSG_GetEnabledSmuFeaturesHigh, &feature_mask_high); if (ret) return ret; - ret = smu_send_smc_msg(smu, SMU_MSG_GetEnabledSmuFeaturesLow); - if (ret) - return ret; - ret = smu_read_smc_arg(smu, &feature_mask_low); + ret = smu_send_smc_msg(smu, SMU_MSG_GetEnabledSmuFeaturesLow, &feature_mask_low); if (ret) return ret; @@ -388,14 +396,11 @@ int smu_v12_0_get_dpm_ultimate_freq(struct smu_context *smu, enum smu_clk_type c switch (clk_type) { case SMU_GFXCLK: case SMU_SCLK: - ret = smu_send_smc_msg(smu, SMU_MSG_GetMaxGfxclkFrequency); + ret = smu_send_smc_msg(smu, SMU_MSG_GetMaxGfxclkFrequency, max); if (ret) { pr_err("Attempt to get max GX frequency from SMC Failed !\n"); goto failed; } - ret = smu_read_smc_arg(smu, max); - if (ret) - goto failed; break; case SMU_UCLK: case SMU_FCLK: @@ -419,14 +424,11 @@ int smu_v12_0_get_dpm_ultimate_freq(struct smu_context *smu, enum smu_clk_type c switch (clk_type) { case SMU_GFXCLK: case SMU_SCLK: - ret = smu_send_smc_msg(smu, SMU_MSG_GetMinGfxclkFrequency); + ret = smu_send_smc_msg(smu, SMU_MSG_GetMinGfxclkFrequency, min); if (ret) { pr_err("Attempt to get min GX frequency from SMC Failed !\n"); goto failed; } - ret = smu_read_smc_arg(smu, min); - if (ret) - goto failed; break; case SMU_UCLK: case SMU_FCLK: @@ -450,7 +452,7 @@ failed: } int smu_v12_0_mode2_reset(struct smu_context *smu){ - return smu_v12_0_send_msg_with_param(smu, SMU_MSG_GfxDeviceDriverReset, SMU_RESET_MODE_2); + return smu_v12_0_send_msg_with_param(smu, SMU_MSG_GfxDeviceDriverReset, SMU_RESET_MODE_2, NULL); } int smu_v12_0_set_soft_freq_limited_range(struct smu_context *smu, enum smu_clk_type clk_type, @@ -458,45 +460,42 @@ int smu_v12_0_set_soft_freq_limited_range(struct smu_context *smu, enum smu_clk_ { int ret = 0; - if (max < min) - return -EINVAL; - switch (clk_type) { case SMU_GFXCLK: case SMU_SCLK: - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinGfxClk, min); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinGfxClk, min, NULL); if (ret) return ret; - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxGfxClk, max); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxGfxClk, max, NULL); if (ret) return ret; break; case SMU_FCLK: case SMU_MCLK: - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinFclkByFreq, min); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinFclkByFreq, min, NULL); if (ret) return ret; - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxFclkByFreq, max); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxFclkByFreq, max, NULL); if (ret) return ret; break; case SMU_SOCCLK: - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinSocclkByFreq, min); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinSocclkByFreq, min, NULL); if (ret) return ret; - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxSocclkByFreq, max); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxSocclkByFreq, max, NULL); if (ret) return ret; break; case SMU_VCLK: - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinVcn, min); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinVcn, min, NULL); if (ret) return ret; - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxVcn, max); + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxVcn, max, NULL); if (ret) return ret; break; @@ -515,11 +514,13 @@ int smu_v12_0_set_driver_table_location(struct smu_context *smu) if (driver_table->mc_address) { ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetDriverDramAddrHigh, - upper_32_bits(driver_table->mc_address)); + upper_32_bits(driver_table->mc_address), + NULL); if (!ret) ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetDriverDramAddrLow, - lower_32_bits(driver_table->mc_address)); + lower_32_bits(driver_table->mc_address), + NULL); } return ret; diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/vega20_smumgr.c b/drivers/gpu/drm/amd/powerplay/smumgr/vega20_smumgr.c index 49e5ef3e3876..16aa171971d3 100644 --- a/drivers/gpu/drm/amd/powerplay/smumgr/vega20_smumgr.c +++ b/drivers/gpu/drm/amd/powerplay/smumgr/vega20_smumgr.c @@ -33,6 +33,8 @@ #include "smu7_smumgr.h" #include "vega20_hwmgr.h" +#include "smu_v11_0_i2c.h" + /* MP Apertures */ #define MP0_Public 0x03800000 #define MP0_SRAM 0x03900000 @@ -406,6 +408,7 @@ static int vega20_smu_init(struct pp_hwmgr *hwmgr) struct vega20_smumgr *priv; unsigned long tools_size = 0x19000; int ret = 0; + struct amdgpu_device *adev = hwmgr->adev; struct cgs_firmware_info info = {0}; @@ -505,6 +508,10 @@ static int vega20_smu_init(struct pp_hwmgr *hwmgr) priv->smu_tables.entry[TABLE_ACTIVITY_MONITOR_COEFF].version = 0x01; priv->smu_tables.entry[TABLE_ACTIVITY_MONITOR_COEFF].size = sizeof(DpmActivityMonitorCoeffInt_t); + ret = smu_v11_0_i2c_eeprom_control_init(&adev->pm.smu_i2c); + if (ret) + goto err4; + return 0; err4: @@ -537,6 +544,9 @@ static int vega20_smu_fini(struct pp_hwmgr *hwmgr) { struct vega20_smumgr *priv = (struct vega20_smumgr *)(hwmgr->smu_backend); + struct amdgpu_device *adev = hwmgr->adev; + + smu_v11_0_i2c_eeprom_control_fini(&adev->pm.smu_i2c); if (priv) { amdgpu_bo_free_kernel(&priv->smu_tables.entry[TABLE_PPTABLE].handle, @@ -560,6 +570,7 @@ static int vega20_smu_fini(struct pp_hwmgr *hwmgr) kfree(hwmgr->smu_backend); hwmgr->smu_backend = NULL; } + return 0; } diff --git a/drivers/gpu/drm/amd/powerplay/vega20_ppt.c b/drivers/gpu/drm/amd/powerplay/vega20_ppt.c index 4ad8d6c14ee5..49ff3756bd9f 100644 --- a/drivers/gpu/drm/amd/powerplay/vega20_ppt.c +++ b/drivers/gpu/drm/amd/powerplay/vega20_ppt.c @@ -21,7 +21,6 @@ * */ -#include "pp_debug.h" #include <linux/firmware.h> #include "amdgpu.h" #include "amdgpu_smu.h" @@ -587,7 +586,7 @@ static int vega20_check_powerplay_table(struct smu_context *smu) static int vega20_run_btc_afll(struct smu_context *smu) { - return smu_send_smc_msg(smu, SMU_MSG_RunAfllBtc); + return smu_send_smc_msg(smu, SMU_MSG_RunAfllBtc, NULL); } #define FEATURE_MASK(feature) (1ULL << feature) @@ -670,13 +669,13 @@ vega20_set_single_dpm_table(struct smu_context *smu, ret = smu_send_smc_msg_with_param(smu, SMU_MSG_GetDpmFreqByIndex, - (clk_id << 16 | 0xFF)); + (clk_id << 16 | 0xFF), + &num_of_levels); if (ret) { pr_err("[GetNumOfDpmLevel] failed to get dpm levels!"); return ret; } - smu_read_smc_arg(smu, &num_of_levels); if (!num_of_levels) { pr_err("[GetNumOfDpmLevel] number of clk levels is invalid!"); return -EINVAL; @@ -687,12 +686,12 @@ vega20_set_single_dpm_table(struct smu_context *smu, for (i = 0; i < num_of_levels; i++) { ret = smu_send_smc_msg_with_param(smu, SMU_MSG_GetDpmFreqByIndex, - (clk_id << 16 | i)); + (clk_id << 16 | i), + &clk); if (ret) { pr_err("[GetDpmFreqByIndex] failed to get dpm freq by index!"); return ret; } - smu_read_smc_arg(smu, &clk); if (!clk) { pr_err("[GetDpmFreqByIndex] clk value is invalid!"); return -EINVAL; @@ -1200,7 +1199,8 @@ static int vega20_upload_dpm_level(struct smu_context *smu, bool max, single_dpm_table->dpm_state.soft_min_level; ret = smu_send_smc_msg_with_param(smu, (max ? SMU_MSG_SetSoftMaxByFreq : SMU_MSG_SetSoftMinByFreq), - (PPCLK_GFXCLK << 16) | (freq & 0xffff)); + (PPCLK_GFXCLK << 16) | (freq & 0xffff), + NULL); if (ret) { pr_err("Failed to set soft %s gfxclk !\n", max ? "max" : "min"); @@ -1215,7 +1215,8 @@ static int vega20_upload_dpm_level(struct smu_context *smu, bool max, single_dpm_table->dpm_state.soft_min_level; ret = smu_send_smc_msg_with_param(smu, (max ? SMU_MSG_SetSoftMaxByFreq : SMU_MSG_SetSoftMinByFreq), - (PPCLK_UCLK << 16) | (freq & 0xffff)); + (PPCLK_UCLK << 16) | (freq & 0xffff), + NULL); if (ret) { pr_err("Failed to set soft %s memclk !\n", max ? "max" : "min"); @@ -1230,7 +1231,8 @@ static int vega20_upload_dpm_level(struct smu_context *smu, bool max, single_dpm_table->dpm_state.soft_min_level; ret = smu_send_smc_msg_with_param(smu, (max ? SMU_MSG_SetSoftMaxByFreq : SMU_MSG_SetSoftMinByFreq), - (PPCLK_SOCCLK << 16) | (freq & 0xffff)); + (PPCLK_SOCCLK << 16) | (freq & 0xffff), + NULL); if (ret) { pr_err("Failed to set soft %s socclk !\n", max ? "max" : "min"); @@ -1245,7 +1247,8 @@ static int vega20_upload_dpm_level(struct smu_context *smu, bool max, single_dpm_table->dpm_state.soft_min_level; ret = smu_send_smc_msg_with_param(smu, (max ? SMU_MSG_SetSoftMaxByFreq : SMU_MSG_SetSoftMinByFreq), - (PPCLK_FCLK << 16) | (freq & 0xffff)); + (PPCLK_FCLK << 16) | (freq & 0xffff), + NULL); if (ret) { pr_err("Failed to set soft %s fclk !\n", max ? "max" : "min"); @@ -1260,7 +1263,8 @@ static int vega20_upload_dpm_level(struct smu_context *smu, bool max, if (!max) { ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinByFreq, - (PPCLK_DCEFCLK << 16) | (freq & 0xffff)); + (PPCLK_DCEFCLK << 16) | (freq & 0xffff), + NULL); if (ret) { pr_err("Failed to set hard min dcefclk !\n"); return ret; @@ -1421,7 +1425,9 @@ static int vega20_force_clk_levels(struct smu_context *smu, } ret = smu_send_smc_msg_with_param(smu, - SMU_MSG_SetMinLinkDpmByIndex, soft_min_level); + SMU_MSG_SetMinLinkDpmByIndex, + soft_min_level, + NULL); if (ret) pr_err("Failed to set min link dpm level!\n"); @@ -1477,13 +1483,13 @@ static int vega20_overdrive_get_gfx_clk_base_voltage(struct smu_context *smu, ret = smu_send_smc_msg_with_param(smu, SMU_MSG_GetAVFSVoltageByDpm, - ((AVFS_CURVE << 24) | (OD8_HOTCURVE_TEMPERATURE << 16) | freq)); + ((AVFS_CURVE << 24) | (OD8_HOTCURVE_TEMPERATURE << 16) | freq), + voltage); if (ret) { pr_err("[GetBaseVoltage] failed to get GFXCLK AVFS voltage from SMU!"); return ret; } - smu_read_smc_arg(smu, voltage); *voltage = *voltage / VOLTAGE_SCALE; return 0; @@ -1956,8 +1962,10 @@ static int vega20_set_power_profile_mode(struct smu_context *smu, long *input, u workload_type = smu_workload_get_type(smu, smu->power_profile_mode); if (workload_type < 0) return -EINVAL; - smu_send_smc_msg_with_param(smu, SMU_MSG_SetWorkloadMask, - 1 << workload_type); + smu_send_smc_msg_with_param(smu, + SMU_MSG_SetWorkloadMask, + 1 << workload_type, + NULL); return ret; } @@ -2029,7 +2037,8 @@ vega20_set_uclk_to_highest_dpm_level(struct smu_context *smu, dpm_table->dpm_state.hard_min_level = dpm_table->dpm_levels[dpm_table->count - 1].value; ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinByFreq, - (PPCLK_UCLK << 16) | dpm_table->dpm_state.hard_min_level); + (PPCLK_UCLK << 16) | dpm_table->dpm_state.hard_min_level, + NULL); if (ret) { pr_err("[%s] Set hard min uclk failed!", __func__); return ret; @@ -2047,7 +2056,7 @@ static int vega20_pre_display_config_changed(struct smu_context *smu) if (!smu->smu_dpm.dpm_context) return -EINVAL; - smu_send_smc_msg_with_param(smu, SMU_MSG_NumOfDisplays, 0); + smu_send_smc_msg_with_param(smu, SMU_MSG_NumOfDisplays, 0, NULL); ret = vega20_set_uclk_to_highest_dpm_level(smu, &dpm_table->mem_table); if (ret) @@ -2074,7 +2083,8 @@ static int vega20_display_config_changed(struct smu_context *smu) smu_feature_is_supported(smu, SMU_FEATURE_DPM_SOCCLK_BIT)) { smu_send_smc_msg_with_param(smu, SMU_MSG_NumOfDisplays, - smu->display_config->num_display); + smu->display_config->num_display, + NULL); } return ret; @@ -2247,7 +2257,8 @@ vega20_notify_smc_display_config(struct smu_context *smu) if (smu_feature_is_supported(smu, SMU_FEATURE_DS_DCEFCLK_BIT)) { ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetMinDeepSleepDcefclk, - min_clocks.dcef_clock_in_sr/100); + min_clocks.dcef_clock_in_sr/100, + NULL); if (ret) { pr_err("Attempt to set divider for DCEFCLK Failed!"); return ret; @@ -2262,7 +2273,8 @@ vega20_notify_smc_display_config(struct smu_context *smu) memtable->dpm_state.hard_min_level = min_clocks.memory_clock/100; ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinByFreq, - (PPCLK_UCLK << 16) | memtable->dpm_state.hard_min_level); + (PPCLK_UCLK << 16) | memtable->dpm_state.hard_min_level, + NULL); if (ret) { pr_err("[%s] Set hard min uclk failed!", __func__); return ret; @@ -2853,8 +2865,10 @@ static int vega20_set_thermal_fan_table(struct smu_context *smu) struct smu_table_context *table_context = &smu->smu_table; PPTable_t *pptable = table_context->driver_pptable; - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetFanTemperatureTarget, - (uint32_t)pptable->FanTargetTemperature); + ret = smu_send_smc_msg_with_param(smu, + SMU_MSG_SetFanTemperatureTarget, + (uint32_t)pptable->FanTargetTemperature, + NULL); return ret; } @@ -2864,15 +2878,13 @@ static int vega20_get_fan_speed_rpm(struct smu_context *smu, { int ret; - ret = smu_send_smc_msg(smu, SMU_MSG_GetCurrentRpm); + ret = smu_send_smc_msg(smu, SMU_MSG_GetCurrentRpm, speed); if (ret) { pr_err("Attempt to get current RPM from SMC Failed!\n"); return ret; } - smu_read_smc_arg(smu, speed); - return 0; } @@ -3137,7 +3149,7 @@ static int vega20_set_df_cstate(struct smu_context *smu, return -EINVAL; } - return smu_send_smc_msg_with_param(smu, SMU_MSG_DFCstateControl, state); + return smu_send_smc_msg_with_param(smu, SMU_MSG_DFCstateControl, state, NULL); } static int vega20_update_pcie_parameters(struct smu_context *smu, @@ -3155,7 +3167,8 @@ static int vega20_update_pcie_parameters(struct smu_context *smu, pptable->PcieLaneCount[i] : pcie_width_cap); ret = smu_send_smc_msg_with_param(smu, SMU_MSG_OverridePcieParameters, - smu_pcie_arg); + smu_pcie_arg, + NULL); } return ret; @@ -3229,7 +3242,6 @@ static const struct pptable_funcs vega20_ppt_funcs = { .notify_memory_pool_location = smu_v11_0_notify_memory_pool_location, .system_features_control = smu_v11_0_system_features_control, .send_smc_msg_with_param = smu_v11_0_send_msg_with_param, - .read_smc_arg = smu_v11_0_read_arg, .init_display_count = smu_v11_0_init_display_count, .set_allowed_mask = smu_v11_0_set_allowed_mask, .get_enabled_mask = smu_v11_0_get_enabled_mask, diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_drv.c b/drivers/gpu/drm/arm/display/komeda/komeda_drv.c index ea5cd1e17304..e7933930a657 100644 --- a/drivers/gpu/drm/arm/display/komeda/komeda_drv.c +++ b/drivers/gpu/drm/arm/display/komeda/komeda_drv.c @@ -146,14 +146,14 @@ static const struct of_device_id komeda_of_match[] = { MODULE_DEVICE_TABLE(of, komeda_of_match); -static int komeda_rt_pm_suspend(struct device *dev) +static int __maybe_unused komeda_rt_pm_suspend(struct device *dev) { struct komeda_drv *mdrv = dev_get_drvdata(dev); return komeda_dev_suspend(mdrv->mdev); } -static int komeda_rt_pm_resume(struct device *dev) +static int __maybe_unused komeda_rt_pm_resume(struct device *dev) { struct komeda_drv *mdrv = dev_get_drvdata(dev); diff --git a/drivers/gpu/drm/bochs/bochs_hw.c b/drivers/gpu/drm/bochs/bochs_hw.c index 952199cc0462..dce4672e3fc8 100644 --- a/drivers/gpu/drm/bochs/bochs_hw.c +++ b/drivers/gpu/drm/bochs/bochs_hw.c @@ -157,10 +157,8 @@ int bochs_hw_init(struct drm_device *dev) size = min(size, mem); } - if (pci_request_region(pdev, 0, "bochs-drm") != 0) { - DRM_ERROR("Cannot request framebuffer\n"); - return -EBUSY; - } + if (pci_request_region(pdev, 0, "bochs-drm") != 0) + DRM_WARN("Cannot request framebuffer, boot fb still active?\n"); bochs->fb_map = ioremap(addr, size); if (bochs->fb_map == NULL) { diff --git a/drivers/gpu/drm/bridge/analogix/analogix-anx6345.c b/drivers/gpu/drm/bridge/analogix/analogix-anx6345.c index 3ec34cf6fdad..2bc6e4f85171 100644 --- a/drivers/gpu/drm/bridge/analogix/analogix-anx6345.c +++ b/drivers/gpu/drm/bridge/analogix/analogix-anx6345.c @@ -210,8 +210,7 @@ static int anx6345_dp_link_training(struct anx6345 *anx6345) if (err) return err; - dpcd[0] = drm_dp_max_link_rate(anx6345->dpcd); - dpcd[0] = drm_dp_link_rate_to_bw_code(dpcd[0]); + dpcd[0] = dp_bw; err = regmap_write(anx6345->map[I2C_IDX_DPTX], SP_DP_MAIN_LINK_BW_SET_REG, dpcd[0]); if (err) diff --git a/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c b/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c index f85c15ad8486..383b1073d7de 100644 --- a/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c +++ b/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c @@ -1624,28 +1624,34 @@ static void hdmi_config_AVI(struct dw_hdmi *hdmi, struct drm_display_mode *mode) frame.colorspace = HDMI_COLORSPACE_RGB; /* Set up colorimetry */ - switch (hdmi->hdmi_data.enc_out_encoding) { - case V4L2_YCBCR_ENC_601: - if (hdmi->hdmi_data.enc_in_encoding == V4L2_YCBCR_ENC_XV601) - frame.colorimetry = HDMI_COLORIMETRY_EXTENDED; - else + if (!hdmi_bus_fmt_is_rgb(hdmi->hdmi_data.enc_out_bus_format)) { + switch (hdmi->hdmi_data.enc_out_encoding) { + case V4L2_YCBCR_ENC_601: + if (hdmi->hdmi_data.enc_in_encoding == V4L2_YCBCR_ENC_XV601) + frame.colorimetry = HDMI_COLORIMETRY_EXTENDED; + else + frame.colorimetry = HDMI_COLORIMETRY_ITU_601; + frame.extended_colorimetry = + HDMI_EXTENDED_COLORIMETRY_XV_YCC_601; + break; + case V4L2_YCBCR_ENC_709: + if (hdmi->hdmi_data.enc_in_encoding == V4L2_YCBCR_ENC_XV709) + frame.colorimetry = HDMI_COLORIMETRY_EXTENDED; + else + frame.colorimetry = HDMI_COLORIMETRY_ITU_709; + frame.extended_colorimetry = + HDMI_EXTENDED_COLORIMETRY_XV_YCC_709; + break; + default: /* Carries no data */ frame.colorimetry = HDMI_COLORIMETRY_ITU_601; + frame.extended_colorimetry = + HDMI_EXTENDED_COLORIMETRY_XV_YCC_601; + break; + } + } else { + frame.colorimetry = HDMI_COLORIMETRY_NONE; frame.extended_colorimetry = - HDMI_EXTENDED_COLORIMETRY_XV_YCC_601; - break; - case V4L2_YCBCR_ENC_709: - if (hdmi->hdmi_data.enc_in_encoding == V4L2_YCBCR_ENC_XV709) - frame.colorimetry = HDMI_COLORIMETRY_EXTENDED; - else - frame.colorimetry = HDMI_COLORIMETRY_ITU_709; - frame.extended_colorimetry = - HDMI_EXTENDED_COLORIMETRY_XV_YCC_709; - break; - default: /* Carries no data */ - frame.colorimetry = HDMI_COLORIMETRY_ITU_601; - frame.extended_colorimetry = - HDMI_EXTENDED_COLORIMETRY_XV_YCC_601; - break; + HDMI_EXTENDED_COLORIMETRY_XV_YCC_601; } frame.scan_mode = HDMI_SCAN_MODE_NONE; diff --git a/drivers/gpu/drm/bridge/tc358767.c b/drivers/gpu/drm/bridge/tc358767.c index 402a690d1fe5..e4c0ea03ae3a 100644 --- a/drivers/gpu/drm/bridge/tc358767.c +++ b/drivers/gpu/drm/bridge/tc358767.c @@ -298,7 +298,7 @@ static inline int tc_poll_timeout(struct tc_data *tc, unsigned int addr, static int tc_aux_wait_busy(struct tc_data *tc) { - return tc_poll_timeout(tc, DP0_AUXSTATUS, AUX_BUSY, 0, 1000, 100000); + return tc_poll_timeout(tc, DP0_AUXSTATUS, AUX_BUSY, 0, 100, 100000); } static int tc_aux_write_data(struct tc_data *tc, const void *data, @@ -641,7 +641,7 @@ static int tc_aux_link_setup(struct tc_data *tc) if (ret) goto err; - ret = tc_poll_timeout(tc, DP_PHY_CTRL, PHY_RDY, PHY_RDY, 1, 1000); + ret = tc_poll_timeout(tc, DP_PHY_CTRL, PHY_RDY, PHY_RDY, 100, 100000); if (ret == -ETIMEDOUT) { dev_err(tc->dev, "Timeout waiting for PHY to become ready"); return ret; @@ -877,7 +877,7 @@ static int tc_wait_link_training(struct tc_data *tc) int ret; ret = tc_poll_timeout(tc, DP0_LTSTAT, LT_LOOPDONE, - LT_LOOPDONE, 1, 1000); + LT_LOOPDONE, 500, 100000); if (ret) { dev_err(tc->dev, "Link training timeout waiting for LT_LOOPDONE!\n"); return ret; @@ -950,7 +950,7 @@ static int tc_main_link_enable(struct tc_data *tc) dp_phy_ctrl &= ~(DP_PHY_RST | PHY_M1_RST | PHY_M0_RST); ret = regmap_write(tc->regmap, DP_PHY_CTRL, dp_phy_ctrl); - ret = tc_poll_timeout(tc, DP_PHY_CTRL, PHY_RDY, PHY_RDY, 1, 1000); + ret = tc_poll_timeout(tc, DP_PHY_CTRL, PHY_RDY, PHY_RDY, 500, 100000); if (ret) { dev_err(dev, "timeout waiting for phy become ready"); return ret; diff --git a/drivers/gpu/drm/bridge/ti-tfp410.c b/drivers/gpu/drm/bridge/ti-tfp410.c index 40c4d4a5517b..e3eb6364c0f7 100644 --- a/drivers/gpu/drm/bridge/ti-tfp410.c +++ b/drivers/gpu/drm/bridge/ti-tfp410.c @@ -150,7 +150,8 @@ static int tfp410_attach(struct drm_bridge *bridge, dvi->next_bridge->type, dvi->next_bridge->ddc); if (ret) { - dev_err(dvi->dev, "drm_connector_init() failed: %d\n", ret); + dev_err(dvi->dev, "drm_connector_init_with_ddc() failed: %d\n", + ret); return ret; } diff --git a/drivers/gpu/drm/drm_client_modeset.c b/drivers/gpu/drm/drm_client_modeset.c index 841794e19eb6..7443114bd713 100644 --- a/drivers/gpu/drm/drm_client_modeset.c +++ b/drivers/gpu/drm/drm_client_modeset.c @@ -951,7 +951,8 @@ bool drm_client_rotation(struct drm_mode_set *modeset, unsigned int *rotation) * depending on the hardware this may require the framebuffer * to be in a specific tiling format. */ - if ((*rotation & DRM_MODE_ROTATE_MASK) != DRM_MODE_ROTATE_180 || + if (((*rotation & DRM_MODE_ROTATE_MASK) != DRM_MODE_ROTATE_0 && + (*rotation & DRM_MODE_ROTATE_MASK) != DRM_MODE_ROTATE_180) || !plane->rotation_property) return false; diff --git a/drivers/gpu/drm/drm_dp_helper.c b/drivers/gpu/drm/drm_dp_helper.c index a4b98f8055f4..c6fbe6e6bc9d 100644 --- a/drivers/gpu/drm/drm_dp_helper.c +++ b/drivers/gpu/drm/drm_dp_helper.c @@ -1280,6 +1280,85 @@ drm_dp_get_quirks(const struct drm_dp_dpcd_ident *ident, bool is_branch) #undef DEVICE_ID_ANY #undef DEVICE_ID +struct edid_quirk { + u8 mfg_id[2]; + u8 prod_id[2]; + u32 quirks; +}; + +#define MFG(first, second) { (first), (second) } +#define PROD_ID(first, second) { (first), (second) } + +/* + * Some devices have unreliable OUIDs where they don't set the device ID + * correctly, and as a result we need to use the EDID for finding additional + * DP quirks in such cases. + */ +static const struct edid_quirk edid_quirk_list[] = { + /* Optional 4K AMOLED panel in the ThinkPad X1 Extreme 2nd Generation + * only supports DPCD backlight controls + */ + { MFG(0x4c, 0x83), PROD_ID(0x41, 0x41), BIT(DP_QUIRK_FORCE_DPCD_BACKLIGHT) }, + /* + * Some Dell CML 2020 systems have panels support both AUX and PWM + * backlight control, and some only support AUX backlight control. All + * said panels start up in AUX mode by default, and we don't have any + * support for disabling HDR mode on these panels which would be + * required to switch to PWM backlight control mode (plus, I'm not + * even sure we want PWM backlight controls over DPCD backlight + * controls anyway...). Until we have a better way of detecting these, + * force DPCD backlight mode on all of them. + */ + { MFG(0x06, 0xaf), PROD_ID(0x9b, 0x32), BIT(DP_QUIRK_FORCE_DPCD_BACKLIGHT) }, + { MFG(0x06, 0xaf), PROD_ID(0xeb, 0x41), BIT(DP_QUIRK_FORCE_DPCD_BACKLIGHT) }, + { MFG(0x4d, 0x10), PROD_ID(0xc7, 0x14), BIT(DP_QUIRK_FORCE_DPCD_BACKLIGHT) }, + { MFG(0x4d, 0x10), PROD_ID(0xe6, 0x14), BIT(DP_QUIRK_FORCE_DPCD_BACKLIGHT) }, +}; + +#undef MFG +#undef PROD_ID + +/** + * drm_dp_get_edid_quirks() - Check the EDID of a DP device to find additional + * DP-specific quirks + * @edid: The EDID to check + * + * While OUIDs are meant to be used to recognize a DisplayPort device, a lot + * of manufacturers don't seem to like following standards and neglect to fill + * the dev-ID in, making it impossible to only use OUIDs for determining + * quirks in some cases. This function can be used to check the EDID and look + * up any additional DP quirks. The bits returned by this function correspond + * to the quirk bits in &drm_dp_quirk. + * + * Returns: a bitmask of quirks, if any. The driver can check this using + * drm_dp_has_quirk(). + */ +u32 drm_dp_get_edid_quirks(const struct edid *edid) +{ + const struct edid_quirk *quirk; + u32 quirks = 0; + int i; + + if (!edid) + return 0; + + for (i = 0; i < ARRAY_SIZE(edid_quirk_list); i++) { + quirk = &edid_quirk_list[i]; + if (memcmp(quirk->mfg_id, edid->mfg_id, + sizeof(edid->mfg_id)) == 0 && + memcmp(quirk->prod_id, edid->prod_code, + sizeof(edid->prod_code)) == 0) + quirks |= quirk->quirks; + } + + DRM_DEBUG_KMS("DP sink: EDID mfg %*phD prod-ID %*phD quirks: 0x%04x\n", + (int)sizeof(edid->mfg_id), edid->mfg_id, + (int)sizeof(edid->prod_code), edid->prod_code, quirks); + + return quirks; +} +EXPORT_SYMBOL(drm_dp_get_edid_quirks); + /** * drm_dp_read_desc - read sink/branch descriptor from DPCD * @aux: DisplayPort AUX channel diff --git a/drivers/gpu/drm/drm_dp_mst_topology.c b/drivers/gpu/drm/drm_dp_mst_topology.c index 4fbec48075d3..70c4b7afed12 100644 --- a/drivers/gpu/drm/drm_dp_mst_topology.c +++ b/drivers/gpu/drm/drm_dp_mst_topology.c @@ -1937,7 +1937,7 @@ static u8 drm_dp_calculate_rad(struct drm_dp_mst_port *port, return parent_lct + 1; } -static bool drm_dp_mst_is_dp_mst_end_device(u8 pdt, bool mcs) +static bool drm_dp_mst_is_end_device(u8 pdt, bool mcs) { switch (pdt) { case DP_PEER_DEVICE_DP_LEGACY_CONV: @@ -1967,13 +1967,13 @@ drm_dp_port_set_pdt(struct drm_dp_mst_port *port, u8 new_pdt, /* Teardown the old pdt, if there is one */ if (port->pdt != DP_PEER_DEVICE_NONE) { - if (drm_dp_mst_is_dp_mst_end_device(port->pdt, port->mcs)) { + if (drm_dp_mst_is_end_device(port->pdt, port->mcs)) { /* * If the new PDT would also have an i2c bus, * don't bother with reregistering it */ if (new_pdt != DP_PEER_DEVICE_NONE && - drm_dp_mst_is_dp_mst_end_device(new_pdt, new_mcs)) { + drm_dp_mst_is_end_device(new_pdt, new_mcs)) { port->pdt = new_pdt; port->mcs = new_mcs; return 0; @@ -1993,7 +1993,7 @@ drm_dp_port_set_pdt(struct drm_dp_mst_port *port, u8 new_pdt, port->mcs = new_mcs; if (port->pdt != DP_PEER_DEVICE_NONE) { - if (drm_dp_mst_is_dp_mst_end_device(port->pdt, port->mcs)) { + if (drm_dp_mst_is_end_device(port->pdt, port->mcs)) { /* add i2c over sideband */ ret = drm_dp_mst_register_i2c_bus(&port->aux); } else { @@ -2172,7 +2172,7 @@ drm_dp_mst_port_add_connector(struct drm_dp_mst_branch *mstb, } if (port->pdt != DP_PEER_DEVICE_NONE && - drm_dp_mst_is_dp_mst_end_device(port->pdt, port->mcs)) { + drm_dp_mst_is_end_device(port->pdt, port->mcs)) { port->cached_edid = drm_get_edid(port->connector, &port->aux.ddc); drm_connector_set_tile_property(port->connector); @@ -2302,14 +2302,18 @@ drm_dp_mst_handle_link_address_port(struct drm_dp_mst_branch *mstb, mutex_unlock(&mgr->lock); } - if (old_ddps != port->ddps) { - if (port->ddps) { - if (!port->input) { - drm_dp_send_enum_path_resources(mgr, mstb, - port); - } + /* + * Reprobe PBN caps on both hotplug, and when re-probing the link + * for our parent mstb + */ + if (old_ddps != port->ddps || !created) { + if (port->ddps && !port->input) { + ret = drm_dp_send_enum_path_resources(mgr, mstb, + port); + if (ret == 1) + changed = true; } else { - port->available_pbn = 0; + port->full_pbn = 0; } } @@ -2401,11 +2405,10 @@ drm_dp_mst_handle_conn_stat(struct drm_dp_mst_branch *mstb, port->ddps = conn_stat->displayport_device_plug_status; if (old_ddps != port->ddps) { - if (port->ddps) { - dowork = true; - } else { - port->available_pbn = 0; - } + if (port->ddps && !port->input) + drm_dp_send_enum_path_resources(mgr, mstb, port); + else + port->full_pbn = 0; } new_pdt = port->input ? DP_PEER_DEVICE_NONE : conn_stat->peer_device_type; @@ -2556,13 +2559,6 @@ static int drm_dp_check_and_send_link_address(struct drm_dp_mst_topology_mgr *mg if (port->input || !port->ddps) continue; - if (!port->available_pbn) { - drm_modeset_lock(&mgr->base.lock, NULL); - drm_dp_send_enum_path_resources(mgr, mstb, port); - drm_modeset_unlock(&mgr->base.lock); - changed = true; - } - if (port->mstb) mstb_child = drm_dp_mst_topology_get_mstb_validated( mgr, port->mstb); @@ -2996,6 +2992,7 @@ drm_dp_send_enum_path_resources(struct drm_dp_mst_topology_mgr *mgr, ret = drm_dp_mst_wait_tx_reply(mstb, txmsg); if (ret > 0) { + ret = 0; path_res = &txmsg->reply.u.path_resources; if (txmsg->reply.reply_type == DP_SIDEBAND_REPLY_NAK) { @@ -3008,14 +3005,22 @@ drm_dp_send_enum_path_resources(struct drm_dp_mst_topology_mgr *mgr, path_res->port_number, path_res->full_payload_bw_number, path_res->avail_payload_bw_number); - port->available_pbn = - path_res->avail_payload_bw_number; + + /* + * If something changed, make sure we send a + * hotplug + */ + if (port->full_pbn != path_res->full_payload_bw_number || + port->fec_capable != path_res->fec_capable) + ret = 1; + + port->full_pbn = path_res->full_payload_bw_number; port->fec_capable = path_res->fec_capable; } } kfree(txmsg); - return 0; + return ret; } static struct drm_dp_mst_port *drm_dp_get_last_connected_port_to_mstb(struct drm_dp_mst_branch *mstb) @@ -3589,13 +3594,9 @@ drm_dp_mst_topology_mgr_invalidate_mstb(struct drm_dp_mst_branch *mstb) /* The link address will need to be re-sent on resume */ mstb->link_address_sent = false; - list_for_each_entry(port, &mstb->ports, next) { - /* The PBN for each port will also need to be re-probed */ - port->available_pbn = 0; - + list_for_each_entry(port, &mstb->ports, next) if (port->mstb) drm_dp_mst_topology_mgr_invalidate_mstb(port->mstb); - } } /** @@ -4858,41 +4859,102 @@ static bool drm_dp_mst_port_downstream_of_branch(struct drm_dp_mst_port *port, return false; } -static inline -int drm_dp_mst_atomic_check_bw_limit(struct drm_dp_mst_branch *branch, - struct drm_dp_mst_topology_state *mst_state) +static int +drm_dp_mst_atomic_check_port_bw_limit(struct drm_dp_mst_port *port, + struct drm_dp_mst_topology_state *state); + +static int +drm_dp_mst_atomic_check_mstb_bw_limit(struct drm_dp_mst_branch *mstb, + struct drm_dp_mst_topology_state *state) { - struct drm_dp_mst_port *port; struct drm_dp_vcpi_allocation *vcpi; - int pbn_limit = 0, pbn_used = 0; + struct drm_dp_mst_port *port; + int pbn_used = 0, ret; + bool found = false; - list_for_each_entry(port, &branch->ports, next) { - if (port->mstb) - if (drm_dp_mst_atomic_check_bw_limit(port->mstb, mst_state)) - return -ENOSPC; + /* Check that we have at least one port in our state that's downstream + * of this branch, otherwise we can skip this branch + */ + list_for_each_entry(vcpi, &state->vcpis, next) { + if (!vcpi->pbn || + !drm_dp_mst_port_downstream_of_branch(vcpi->port, mstb)) + continue; - if (port->available_pbn > 0) - pbn_limit = port->available_pbn; + found = true; + break; } - DRM_DEBUG_ATOMIC("[MST BRANCH:%p] branch has %d PBN available\n", - branch, pbn_limit); + if (!found) + return 0; - list_for_each_entry(vcpi, &mst_state->vcpis, next) { - if (!vcpi->pbn) - continue; + if (mstb->port_parent) + DRM_DEBUG_ATOMIC("[MSTB:%p] [MST PORT:%p] Checking bandwidth limits on [MSTB:%p]\n", + mstb->port_parent->parent, mstb->port_parent, + mstb); + else + DRM_DEBUG_ATOMIC("[MSTB:%p] Checking bandwidth limits\n", + mstb); + + list_for_each_entry(port, &mstb->ports, next) { + ret = drm_dp_mst_atomic_check_port_bw_limit(port, state); + if (ret < 0) + return ret; + + pbn_used += ret; + } + + return pbn_used; +} + +static int +drm_dp_mst_atomic_check_port_bw_limit(struct drm_dp_mst_port *port, + struct drm_dp_mst_topology_state *state) +{ + struct drm_dp_vcpi_allocation *vcpi; + int pbn_used = 0; + + if (port->pdt == DP_PEER_DEVICE_NONE) + return 0; + + if (drm_dp_mst_is_end_device(port->pdt, port->mcs)) { + bool found = false; + + list_for_each_entry(vcpi, &state->vcpis, next) { + if (vcpi->port != port) + continue; + if (!vcpi->pbn) + return 0; + + found = true; + break; + } + if (!found) + return 0; - if (drm_dp_mst_port_downstream_of_branch(vcpi->port, branch)) - pbn_used += vcpi->pbn; + /* This should never happen, as it means we tried to + * set a mode before querying the full_pbn + */ + if (WARN_ON(!port->full_pbn)) + return -EINVAL; + + pbn_used = vcpi->pbn; + } else { + pbn_used = drm_dp_mst_atomic_check_mstb_bw_limit(port->mstb, + state); + if (pbn_used <= 0) + return pbn_used; } - DRM_DEBUG_ATOMIC("[MST BRANCH:%p] branch used %d PBN\n", - branch, pbn_used); - if (pbn_used > pbn_limit) { - DRM_DEBUG_ATOMIC("[MST BRANCH:%p] No available bandwidth\n", - branch); + if (pbn_used > port->full_pbn) { + DRM_DEBUG_ATOMIC("[MSTB:%p] [MST PORT:%p] required PBN of %d exceeds port limit of %d\n", + port->parent, port, pbn_used, + port->full_pbn); return -ENOSPC; } - return 0; + + DRM_DEBUG_ATOMIC("[MSTB:%p] [MST PORT:%p] uses %d out of %d PBN\n", + port->parent, port, pbn_used, port->full_pbn); + + return pbn_used; } static inline int @@ -5090,9 +5152,15 @@ int drm_dp_mst_atomic_check(struct drm_atomic_state *state) ret = drm_dp_mst_atomic_check_vcpi_alloc_limit(mgr, mst_state); if (ret) break; - ret = drm_dp_mst_atomic_check_bw_limit(mgr->mst_primary, mst_state); - if (ret) + + mutex_lock(&mgr->lock); + ret = drm_dp_mst_atomic_check_mstb_bw_limit(mgr->mst_primary, + mst_state); + mutex_unlock(&mgr->lock); + if (ret < 0) break; + else + ret = 0; } return ret; @@ -5490,7 +5558,8 @@ struct drm_dp_aux *drm_dp_mst_dsc_aux_for_port(struct drm_dp_mst_port *port) if (drm_dp_read_desc(port->mgr->aux, &desc, true)) return NULL; - if (drm_dp_has_quirk(&desc, DP_DPCD_QUIRK_DSC_WITHOUT_VIRTUAL_DPCD) && + if (drm_dp_has_quirk(&desc, 0, + DP_DPCD_QUIRK_DSC_WITHOUT_VIRTUAL_DPCD) && port->mgr->dpcd[DP_DPCD_REV] >= DP_DPCD_REV_14 && port->parent == port->mgr->mst_primary) { u8 downstreamport; diff --git a/drivers/gpu/drm/drm_file.c b/drivers/gpu/drm/drm_file.c index c4c704e01961..eb009d3ab48f 100644 --- a/drivers/gpu/drm/drm_file.c +++ b/drivers/gpu/drm/drm_file.c @@ -48,6 +48,11 @@ #include "drm_internal.h" #include "drm_legacy.h" +#if defined(CONFIG_MMU) && defined(CONFIG_TRANSPARENT_HUGEPAGE) +#include <uapi/asm/mman.h> +#include <drm/drm_vma_manager.h> +#endif + /* from BKL pushdown */ DEFINE_MUTEX(drm_global_mutex); @@ -872,3 +877,139 @@ struct file *mock_drm_getfile(struct drm_minor *minor, unsigned int flags) return file; } EXPORT_SYMBOL_FOR_TESTS_ONLY(mock_drm_getfile); + +#ifdef CONFIG_MMU +#ifdef CONFIG_TRANSPARENT_HUGEPAGE +/* + * drm_addr_inflate() attempts to construct an aligned area by inflating + * the area size and skipping the unaligned start of the area. + * adapted from shmem_get_unmapped_area() + */ +static unsigned long drm_addr_inflate(unsigned long addr, + unsigned long len, + unsigned long pgoff, + unsigned long flags, + unsigned long huge_size) +{ + unsigned long offset, inflated_len; + unsigned long inflated_addr; + unsigned long inflated_offset; + + offset = (pgoff << PAGE_SHIFT) & (huge_size - 1); + if (offset && offset + len < 2 * huge_size) + return addr; + if ((addr & (huge_size - 1)) == offset) + return addr; + + inflated_len = len + huge_size - PAGE_SIZE; + if (inflated_len > TASK_SIZE) + return addr; + if (inflated_len < len) + return addr; + + inflated_addr = current->mm->get_unmapped_area(NULL, 0, inflated_len, + 0, flags); + if (IS_ERR_VALUE(inflated_addr)) + return addr; + if (inflated_addr & ~PAGE_MASK) + return addr; + + inflated_offset = inflated_addr & (huge_size - 1); + inflated_addr += offset - inflated_offset; + if (inflated_offset > offset) + inflated_addr += huge_size; + + if (inflated_addr > TASK_SIZE - len) + return addr; + + return inflated_addr; +} + +/** + * drm_get_unmapped_area() - Get an unused user-space virtual memory area + * suitable for huge page table entries. + * @file: The struct file representing the address space being mmap()'d. + * @uaddr: Start address suggested by user-space. + * @len: Length of the area. + * @pgoff: The page offset into the address space. + * @flags: mmap flags + * @mgr: The address space manager used by the drm driver. This argument can + * probably be removed at some point when all drivers use the same + * address space manager. + * + * This function attempts to find an unused user-space virtual memory area + * that can accommodate the size we want to map, and that is properly + * aligned to facilitate huge page table entries matching actual + * huge pages or huge page aligned memory in buffer objects. Buffer objects + * are assumed to start at huge page boundary pfns (io memory) or be + * populated by huge pages aligned to the start of the buffer object + * (system- or coherent memory). Adapted from shmem_get_unmapped_area. + * + * Return: aligned user-space address. + */ +unsigned long drm_get_unmapped_area(struct file *file, + unsigned long uaddr, unsigned long len, + unsigned long pgoff, unsigned long flags, + struct drm_vma_offset_manager *mgr) +{ + unsigned long addr; + unsigned long inflated_addr; + struct drm_vma_offset_node *node; + + if (len > TASK_SIZE) + return -ENOMEM; + + /* + * @pgoff is the file page-offset the huge page boundaries of + * which typically aligns to physical address huge page boundaries. + * That's not true for DRM, however, where physical address huge + * page boundaries instead are aligned with the offset from + * buffer object start. So adjust @pgoff to be the offset from + * buffer object start. + */ + drm_vma_offset_lock_lookup(mgr); + node = drm_vma_offset_lookup_locked(mgr, pgoff, 1); + if (node) + pgoff -= node->vm_node.start; + drm_vma_offset_unlock_lookup(mgr); + + addr = current->mm->get_unmapped_area(file, uaddr, len, pgoff, flags); + if (IS_ERR_VALUE(addr)) + return addr; + if (addr & ~PAGE_MASK) + return addr; + if (addr > TASK_SIZE - len) + return addr; + + if (len < HPAGE_PMD_SIZE) + return addr; + if (flags & MAP_FIXED) + return addr; + /* + * Our priority is to support MAP_SHARED mapped hugely; + * and support MAP_PRIVATE mapped hugely too, until it is COWed. + * But if caller specified an address hint, respect that as before. + */ + if (uaddr) + return addr; + + inflated_addr = drm_addr_inflate(addr, len, pgoff, flags, + HPAGE_PMD_SIZE); + + if (IS_ENABLED(CONFIG_HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD) && + len >= HPAGE_PUD_SIZE) + inflated_addr = drm_addr_inflate(inflated_addr, len, pgoff, + flags, HPAGE_PUD_SIZE); + return inflated_addr; +} +#else /* CONFIG_TRANSPARENT_HUGEPAGE */ +unsigned long drm_get_unmapped_area(struct file *file, + unsigned long uaddr, unsigned long len, + unsigned long pgoff, unsigned long flags, + struct drm_vma_offset_manager *mgr) +{ + return current->mm->get_unmapped_area(file, uaddr, len, pgoff, flags); +} +#endif /* CONFIG_TRANSPARENT_HUGEPAGE */ +EXPORT_SYMBOL_GPL(drm_get_unmapped_area); +#endif /* CONFIG_MMU */ diff --git a/drivers/gpu/drm/drm_gem_shmem_helper.c b/drivers/gpu/drm/drm_gem_shmem_helper.c index a421a2eed48a..df31e5782eed 100644 --- a/drivers/gpu/drm/drm_gem_shmem_helper.c +++ b/drivers/gpu/drm/drm_gem_shmem_helper.c @@ -254,11 +254,16 @@ static void *drm_gem_shmem_vmap_locked(struct drm_gem_shmem_object *shmem) if (ret) goto err_zero_use; - if (obj->import_attach) + if (obj->import_attach) { shmem->vaddr = dma_buf_vmap(obj->import_attach->dmabuf); - else + } else { + pgprot_t prot = PAGE_KERNEL; + + if (!shmem->map_cached) + prot = pgprot_writecombine(prot); shmem->vaddr = vmap(shmem->pages, obj->size >> PAGE_SHIFT, - VM_MAP, pgprot_writecombine(PAGE_KERNEL)); + VM_MAP, prot); + } if (!shmem->vaddr) { DRM_DEBUG_KMS("Failed to vmap pages\n"); @@ -540,8 +545,9 @@ int drm_gem_shmem_mmap(struct drm_gem_object *obj, struct vm_area_struct *vma) } vma->vm_flags |= VM_MIXEDMAP | VM_DONTEXPAND; - vma->vm_page_prot = pgprot_writecombine(vm_get_page_prot(vma->vm_flags)); - vma->vm_page_prot = pgprot_decrypted(vma->vm_page_prot); + vma->vm_page_prot = vm_get_page_prot(vma->vm_flags); + if (!shmem->map_cached) + vma->vm_page_prot = pgprot_writecombine(vma->vm_page_prot); vma->vm_ops = &drm_gem_shmem_vm_ops; return 0; diff --git a/drivers/gpu/drm/drm_lease.c b/drivers/gpu/drm/drm_lease.c index b481cafdde28..825abe38201a 100644 --- a/drivers/gpu/drm/drm_lease.c +++ b/drivers/gpu/drm/drm_lease.c @@ -542,10 +542,12 @@ int drm_mode_create_lease_ioctl(struct drm_device *dev, } DRM_DEBUG_LEASE("Creating lease\n"); + /* lessee will take the ownership of leases */ lessee = drm_lease_create(lessor, &leases); if (IS_ERR(lessee)) { ret = PTR_ERR(lessee); + idr_destroy(&leases); goto out_leases; } @@ -580,7 +582,6 @@ out_lessee: out_leases: put_unused_fd(fd); - idr_destroy(&leases); DRM_DEBUG_LEASE("drm_mode_create_lease_ioctl failed: %d\n", ret); return ret; diff --git a/drivers/gpu/drm/drm_modes.c b/drivers/gpu/drm/drm_modes.c index 10336b144c72..d4d64518e11b 100644 --- a/drivers/gpu/drm/drm_modes.c +++ b/drivers/gpu/drm/drm_modes.c @@ -1698,6 +1698,13 @@ static int drm_mode_parse_cmdline_options(const char *str, if (rotation && freestanding) return -EINVAL; + if (!(rotation & DRM_MODE_ROTATE_MASK)) + rotation |= DRM_MODE_ROTATE_0; + + /* Make sure there is exactly one rotation defined */ + if (!is_power_of_2(rotation & DRM_MODE_ROTATE_MASK)) + return -EINVAL; + mode->rotation_reflection = rotation; return 0; diff --git a/drivers/gpu/drm/drm_prime.c b/drivers/gpu/drm/drm_prime.c index 86d9b0e45c8c..1de2cde2277c 100644 --- a/drivers/gpu/drm/drm_prime.c +++ b/drivers/gpu/drm/drm_prime.c @@ -967,7 +967,7 @@ int drm_prime_sg_to_page_addr_arrays(struct sg_table *sgt, struct page **pages, index = 0; for_each_sg(sgt->sgl, sg, sgt->nents, count) { - len = sg->length; + len = sg_dma_len(sg); page = sg_page(sg); addr = sg_dma_address(sg); diff --git a/drivers/gpu/drm/etnaviv/etnaviv_buffer.c b/drivers/gpu/drm/etnaviv/etnaviv_buffer.c index 32d9fac587f9..76d38561c910 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_buffer.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_buffer.c @@ -12,6 +12,7 @@ #include "common.xml.h" #include "state.xml.h" +#include "state_blt.xml.h" #include "state_hi.xml.h" #include "state_3d.xml.h" #include "cmdstream.xml.h" @@ -233,6 +234,8 @@ void etnaviv_buffer_end(struct etnaviv_gpu *gpu) struct etnaviv_cmdbuf *buffer = &gpu->buffer; unsigned int waitlink_offset = buffer->user_size - 16; u32 link_target, flush = 0; + bool has_blt = !!(gpu->identity.minor_features5 & + chipMinorFeatures5_BLT_ENGINE); lockdep_assert_held(&gpu->lock); @@ -248,16 +251,38 @@ void etnaviv_buffer_end(struct etnaviv_gpu *gpu) if (flush) { unsigned int dwords = 7; + if (has_blt) + dwords += 10; + link_target = etnaviv_buffer_reserve(gpu, buffer, dwords); CMD_SEM(buffer, SYNC_RECIPIENT_FE, SYNC_RECIPIENT_PE); CMD_STALL(buffer, SYNC_RECIPIENT_FE, SYNC_RECIPIENT_PE); + if (has_blt) { + CMD_LOAD_STATE(buffer, VIVS_BLT_ENABLE, 0x1); + CMD_SEM(buffer, SYNC_RECIPIENT_FE, SYNC_RECIPIENT_BLT); + CMD_STALL(buffer, SYNC_RECIPIENT_FE, SYNC_RECIPIENT_BLT); + CMD_LOAD_STATE(buffer, VIVS_BLT_ENABLE, 0x0); + } CMD_LOAD_STATE(buffer, VIVS_GL_FLUSH_CACHE, flush); - if (gpu->exec_state == ETNA_PIPE_3D) - CMD_LOAD_STATE(buffer, VIVS_TS_FLUSH_CACHE, - VIVS_TS_FLUSH_CACHE_FLUSH); + if (gpu->exec_state == ETNA_PIPE_3D) { + if (has_blt) { + CMD_LOAD_STATE(buffer, VIVS_BLT_ENABLE, 0x1); + CMD_LOAD_STATE(buffer, VIVS_BLT_SET_COMMAND, 0x1); + CMD_LOAD_STATE(buffer, VIVS_BLT_ENABLE, 0x0); + } else { + CMD_LOAD_STATE(buffer, VIVS_TS_FLUSH_CACHE, + VIVS_TS_FLUSH_CACHE_FLUSH); + } + } CMD_SEM(buffer, SYNC_RECIPIENT_FE, SYNC_RECIPIENT_PE); CMD_STALL(buffer, SYNC_RECIPIENT_FE, SYNC_RECIPIENT_PE); + if (has_blt) { + CMD_LOAD_STATE(buffer, VIVS_BLT_ENABLE, 0x1); + CMD_SEM(buffer, SYNC_RECIPIENT_FE, SYNC_RECIPIENT_BLT); + CMD_STALL(buffer, SYNC_RECIPIENT_FE, SYNC_RECIPIENT_BLT); + CMD_LOAD_STATE(buffer, VIVS_BLT_ENABLE, 0x0); + } CMD_END(buffer); etnaviv_buffer_replace_wait(buffer, waitlink_offset, @@ -323,6 +348,8 @@ void etnaviv_buffer_queue(struct etnaviv_gpu *gpu, u32 exec_state, bool switch_mmu_context = gpu->mmu_context != mmu_context; unsigned int new_flush_seq = READ_ONCE(gpu->mmu_context->flush_seq); bool need_flush = switch_mmu_context || gpu->flush_seq != new_flush_seq; + bool has_blt = !!(gpu->identity.minor_features5 & + chipMinorFeatures5_BLT_ENGINE); lockdep_assert_held(&gpu->lock); @@ -433,6 +460,15 @@ void etnaviv_buffer_queue(struct etnaviv_gpu *gpu, u32 exec_state, * 2 semaphore stall + 1 event + 1 wait + 1 link. */ return_dwords = 7; + + /* + * When the BLT engine is present we need 6 more dwords in the return + * target: 3 enable/flush/disable + 4 enable/semaphore stall/disable, + * but we don't need the normal TS flush state. + */ + if (has_blt) + return_dwords += 6; + return_target = etnaviv_buffer_reserve(gpu, buffer, return_dwords); CMD_LINK(cmdbuf, return_dwords, return_target); @@ -447,11 +483,25 @@ void etnaviv_buffer_queue(struct etnaviv_gpu *gpu, u32 exec_state, CMD_LOAD_STATE(buffer, VIVS_GL_FLUSH_CACHE, VIVS_GL_FLUSH_CACHE_DEPTH | VIVS_GL_FLUSH_CACHE_COLOR); - CMD_LOAD_STATE(buffer, VIVS_TS_FLUSH_CACHE, - VIVS_TS_FLUSH_CACHE_FLUSH); + if (has_blt) { + CMD_LOAD_STATE(buffer, VIVS_BLT_ENABLE, 0x1); + CMD_LOAD_STATE(buffer, VIVS_BLT_SET_COMMAND, 0x1); + CMD_LOAD_STATE(buffer, VIVS_BLT_ENABLE, 0x0); + } else { + CMD_LOAD_STATE(buffer, VIVS_TS_FLUSH_CACHE, + VIVS_TS_FLUSH_CACHE_FLUSH); + } } CMD_SEM(buffer, SYNC_RECIPIENT_FE, SYNC_RECIPIENT_PE); CMD_STALL(buffer, SYNC_RECIPIENT_FE, SYNC_RECIPIENT_PE); + + if (has_blt) { + CMD_LOAD_STATE(buffer, VIVS_BLT_ENABLE, 0x1); + CMD_SEM(buffer, SYNC_RECIPIENT_FE, SYNC_RECIPIENT_BLT); + CMD_STALL(buffer, SYNC_RECIPIENT_FE, SYNC_RECIPIENT_BLT); + CMD_LOAD_STATE(buffer, VIVS_BLT_ENABLE, 0x0); + } + CMD_LOAD_STATE(buffer, VIVS_GL_EVENT, VIVS_GL_EVENT_EVENT_ID(event) | VIVS_GL_EVENT_FROM_PE); CMD_WAIT(buffer); diff --git a/drivers/gpu/drm/etnaviv/etnaviv_drv.c b/drivers/gpu/drm/etnaviv/etnaviv_drv.c index 6b43c1c94e8f..a8685b2e1803 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_drv.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_drv.c @@ -551,6 +551,7 @@ static int etnaviv_bind(struct device *dev) mutex_init(&priv->gem_lock); INIT_LIST_HEAD(&priv->gem_list); priv->num_gpus = 0; + priv->shm_gfp_mask = GFP_HIGHUSER | __GFP_RETRY_MAYFAIL | __GFP_NOWARN; priv->cmdbuf_suballoc = etnaviv_cmdbuf_suballoc_new(drm->dev); if (IS_ERR(priv->cmdbuf_suballoc)) { diff --git a/drivers/gpu/drm/etnaviv/etnaviv_drv.h b/drivers/gpu/drm/etnaviv/etnaviv_drv.h index efc656efeb0f..4d8dc9236e5f 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_drv.h +++ b/drivers/gpu/drm/etnaviv/etnaviv_drv.h @@ -35,6 +35,7 @@ struct etnaviv_drm_private { int num_gpus; struct device_dma_parameters dma_parms; struct etnaviv_gpu *gpu[ETNA_MAX_PIPES]; + gfp_t shm_gfp_mask; struct etnaviv_cmdbuf_suballoc *cmdbuf_suballoc; struct etnaviv_iommu_global *mmu_global; diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gem.c b/drivers/gpu/drm/etnaviv/etnaviv_gem.c index 6adea180d629..dc9ef302f517 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_gem.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_gem.c @@ -602,6 +602,7 @@ static int etnaviv_gem_new_impl(struct drm_device *dev, u32 size, u32 flags, int etnaviv_gem_new_handle(struct drm_device *dev, struct drm_file *file, u32 size, u32 flags, u32 *handle) { + struct etnaviv_drm_private *priv = dev->dev_private; struct drm_gem_object *obj = NULL; int ret; @@ -624,8 +625,7 @@ int etnaviv_gem_new_handle(struct drm_device *dev, struct drm_file *file, * above new_inode() why this is required _and_ expected if you're * going to pin these pages. */ - mapping_set_gfp_mask(obj->filp->f_mapping, GFP_HIGHUSER | - __GFP_RETRY_MAYFAIL | __GFP_NOWARN); + mapping_set_gfp_mask(obj->filp->f_mapping, priv->shm_gfp_mask); etnaviv_gem_obj_add(dev, obj); diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gem.h b/drivers/gpu/drm/etnaviv/etnaviv_gem.h index 6b68fe16041b..98e60df882b6 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_gem.h +++ b/drivers/gpu/drm/etnaviv/etnaviv_gem.h @@ -105,7 +105,7 @@ struct etnaviv_gem_submit { unsigned int nr_pmrs; struct etnaviv_perfmon_request *pmrs; unsigned int nr_bos; - struct etnaviv_gem_submit_bo bos[0]; + struct etnaviv_gem_submit_bo bos[]; /* No new members here, the previous one is variable-length! */ }; diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gpu.c b/drivers/gpu/drm/etnaviv/etnaviv_gpu.c index 799ec20b267d..a31eeff2b297 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_gpu.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_gpu.c @@ -333,9 +333,13 @@ static void etnaviv_hw_identify(struct etnaviv_gpu *gpu) gpu->identity.revision = etnaviv_field(chipIdentity, VIVS_HI_CHIP_IDENTITY_REVISION); } else { + u32 chipDate = gpu_read(gpu, VIVS_HI_CHIP_DATE); gpu->identity.model = gpu_read(gpu, VIVS_HI_CHIP_MODEL); gpu->identity.revision = gpu_read(gpu, VIVS_HI_CHIP_REV); + gpu->identity.product_id = gpu_read(gpu, VIVS_HI_CHIP_PRODUCT_ID); + gpu->identity.customer_id = gpu_read(gpu, VIVS_HI_CHIP_CUSTOMER_ID); + gpu->identity.eco_id = gpu_read(gpu, VIVS_HI_CHIP_ECO_ID); /* * !!!! HACK ALERT !!!! @@ -350,7 +354,6 @@ static void etnaviv_hw_identify(struct etnaviv_gpu *gpu) /* Another special case */ if (etnaviv_is_model_rev(gpu, GC300, 0x2201)) { - u32 chipDate = gpu_read(gpu, VIVS_HI_CHIP_DATE); u32 chipTime = gpu_read(gpu, VIVS_HI_CHIP_TIME); if (chipDate == 0x20080814 && chipTime == 0x12051100) { @@ -373,6 +376,12 @@ static void etnaviv_hw_identify(struct etnaviv_gpu *gpu) gpu->identity.model = chipModel_GC3000; gpu->identity.revision &= 0xffff; } + + if (etnaviv_is_model_rev(gpu, GC1000, 0x5037) && (chipDate == 0x20120617)) + gpu->identity.eco_id = 1; + + if (etnaviv_is_model_rev(gpu, GC320, 0x5303) && (chipDate == 0x20140511)) + gpu->identity.eco_id = 1; } dev_info(gpu->dev, "model: GC%x, revision: %x\n", @@ -506,7 +515,7 @@ static int etnaviv_hw_reset(struct etnaviv_gpu *gpu) /* read idle register. */ idle = gpu_read(gpu, VIVS_HI_IDLE_STATE); - /* try reseting again if FE it not idle */ + /* try resetting again if FE is not idle */ if ((idle & VIVS_HI_IDLE_STATE_FE) == 0) { dev_dbg(gpu->dev, "FE is not idle\n"); continue; @@ -772,6 +781,14 @@ int etnaviv_gpu_init(struct etnaviv_gpu *gpu) gpu->identity.features &= ~chipFeatures_FAST_CLEAR; } + /* + * If the GPU is part of a system with DMA addressing limitations, + * request pages for our SHM backend buffers from the DMA32 zone to + * hopefully avoid performance killing SWIOTLB bounce buffering. + */ + if (dma_addressing_limited(gpu->dev)) + priv->shm_gfp_mask |= GFP_DMA32; + /* Create buffer: */ ret = etnaviv_cmdbuf_init(priv->cmdbuf_suballoc, &gpu->buffer, PAGE_SIZE); @@ -851,6 +868,13 @@ int etnaviv_gpu_debugfs(struct etnaviv_gpu *gpu, struct seq_file *m) verify_dma(gpu, &debug); + seq_puts(m, "\tidentity\n"); + seq_printf(m, "\t model: 0x%x\n", gpu->identity.model); + seq_printf(m, "\t revision: 0x%x\n", gpu->identity.revision); + seq_printf(m, "\t product_id: 0x%x\n", gpu->identity.product_id); + seq_printf(m, "\t customer_id: 0x%x\n", gpu->identity.customer_id); + seq_printf(m, "\t eco_id: 0x%x\n", gpu->identity.eco_id); + seq_puts(m, "\tfeatures\n"); seq_printf(m, "\t major_features: 0x%08x\n", gpu->identity.features); @@ -930,6 +954,20 @@ int etnaviv_gpu_debugfs(struct etnaviv_gpu *gpu, struct seq_file *m) seq_puts(m, "\t FP is not idle\n"); if ((idle & VIVS_HI_IDLE_STATE_TS) == 0) seq_puts(m, "\t TS is not idle\n"); + if ((idle & VIVS_HI_IDLE_STATE_BL) == 0) + seq_puts(m, "\t BL is not idle\n"); + if ((idle & VIVS_HI_IDLE_STATE_ASYNCFE) == 0) + seq_puts(m, "\t ASYNCFE is not idle\n"); + if ((idle & VIVS_HI_IDLE_STATE_MC) == 0) + seq_puts(m, "\t MC is not idle\n"); + if ((idle & VIVS_HI_IDLE_STATE_PPA) == 0) + seq_puts(m, "\t PPA is not idle\n"); + if ((idle & VIVS_HI_IDLE_STATE_WD) == 0) + seq_puts(m, "\t WD is not idle\n"); + if ((idle & VIVS_HI_IDLE_STATE_NN) == 0) + seq_puts(m, "\t NN is not idle\n"); + if ((idle & VIVS_HI_IDLE_STATE_TP) == 0) + seq_puts(m, "\t TP is not idle\n"); if (idle & VIVS_HI_IDLE_STATE_AXI_LP) seq_puts(m, "\t AXI low power mode\n"); @@ -1805,11 +1843,15 @@ static int etnaviv_gpu_rpm_suspend(struct device *dev) if (atomic_read(&gpu->sched.hw_rq_count)) return -EBUSY; - /* Check whether the hardware (except FE) is idle */ - mask = gpu->idle_mask & ~VIVS_HI_IDLE_STATE_FE; + /* Check whether the hardware (except FE and MC) is idle */ + mask = gpu->idle_mask & ~(VIVS_HI_IDLE_STATE_FE | + VIVS_HI_IDLE_STATE_MC); idle = gpu_read(gpu, VIVS_HI_IDLE_STATE) & mask; - if (idle != mask) + if (idle != mask) { + dev_warn_ratelimited(dev, "GPU not yet idle, mask: 0x%08x\n", + idle); return -EBUSY; + } return etnaviv_gpu_hw_suspend(gpu); } diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gpu.h b/drivers/gpu/drm/etnaviv/etnaviv_gpu.h index 97bb48042b4d..8ea48697d132 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_gpu.h +++ b/drivers/gpu/drm/etnaviv/etnaviv_gpu.h @@ -15,11 +15,11 @@ struct etnaviv_gem_submit; struct etnaviv_vram_mapping; struct etnaviv_chip_identity { - /* Chip model. */ u32 model; - - /* Revision value.*/ u32 revision; + u32 product_id; + u32 customer_id; + u32 eco_id; /* Supported feature fields. */ u32 features; diff --git a/drivers/gpu/drm/etnaviv/etnaviv_hwdb.c b/drivers/gpu/drm/etnaviv/etnaviv_hwdb.c index 39b463db76c9..167971a09be7 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_hwdb.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_hwdb.c @@ -7,8 +7,42 @@ static const struct etnaviv_chip_identity etnaviv_chip_identities[] = { { + .model = 0x400, + .revision = 0x4652, + .product_id = 0x70001, + .customer_id = 0x100, + .eco_id = 0, + .stream_count = 4, + .register_max = 64, + .thread_count = 128, + .shader_core_count = 1, + .vertex_cache_size = 8, + .vertex_output_buffer_size = 1024, + .pixel_pipes = 1, + .instruction_count = 256, + .num_constants = 320, + .buffer_size = 0, + .varyings_count = 8, + .features = 0xa0e9e004, + .minor_features0 = 0xe1299fff, + .minor_features1 = 0xbe13b219, + .minor_features2 = 0xce110010, + .minor_features3 = 0x8000001, + .minor_features4 = 0x20102, + .minor_features5 = 0x120000, + .minor_features6 = 0x0, + .minor_features7 = 0x0, + .minor_features8 = 0x0, + .minor_features9 = 0x0, + .minor_features10 = 0x0, + .minor_features11 = 0x0, + }, + { .model = 0x7000, .revision = 0x6214, + .product_id = ~0U, + .customer_id = ~0U, + .eco_id = ~0U, .stream_count = 16, .register_max = 64, .thread_count = 1024, @@ -43,7 +77,13 @@ bool etnaviv_fill_identity_from_hwdb(struct etnaviv_gpu *gpu) for (i = 0; i < ARRAY_SIZE(etnaviv_chip_identities); i++) { if (etnaviv_chip_identities[i].model == ident->model && - etnaviv_chip_identities[i].revision == ident->revision) { + etnaviv_chip_identities[i].revision == ident->revision && + (etnaviv_chip_identities[i].product_id == ident->product_id || + etnaviv_chip_identities[i].product_id == ~0U) && + (etnaviv_chip_identities[i].customer_id == ident->customer_id || + etnaviv_chip_identities[i].customer_id == ~0U) && + (etnaviv_chip_identities[i].eco_id == ident->eco_id || + etnaviv_chip_identities[i].eco_id == ~0U)) { memcpy(ident, &etnaviv_chip_identities[i], sizeof(*ident)); return true; diff --git a/drivers/gpu/drm/etnaviv/etnaviv_perfmon.c b/drivers/gpu/drm/etnaviv/etnaviv_perfmon.c index 8adbf2861bff..e6795bafcbb9 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_perfmon.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_perfmon.c @@ -32,6 +32,7 @@ struct etnaviv_pm_domain { }; struct etnaviv_pm_domain_meta { + unsigned int feature; const struct etnaviv_pm_domain *domains; u32 nr_domains; }; @@ -410,36 +411,78 @@ static const struct etnaviv_pm_domain doms_vg[] = { static const struct etnaviv_pm_domain_meta doms_meta[] = { { + .feature = chipFeatures_PIPE_3D, .nr_domains = ARRAY_SIZE(doms_3d), .domains = &doms_3d[0] }, { + .feature = chipFeatures_PIPE_2D, .nr_domains = ARRAY_SIZE(doms_2d), .domains = &doms_2d[0] }, { + .feature = chipFeatures_PIPE_VG, .nr_domains = ARRAY_SIZE(doms_vg), .domains = &doms_vg[0] } }; +static unsigned int num_pm_domains(const struct etnaviv_gpu *gpu) +{ + unsigned int num = 0, i; + + for (i = 0; i < ARRAY_SIZE(doms_meta); i++) { + const struct etnaviv_pm_domain_meta *meta = &doms_meta[i]; + + if (gpu->identity.features & meta->feature) + num += meta->nr_domains; + } + + return num; +} + +static const struct etnaviv_pm_domain *pm_domain(const struct etnaviv_gpu *gpu, + unsigned int index) +{ + const struct etnaviv_pm_domain *domain = NULL; + unsigned int offset = 0, i; + + for (i = 0; i < ARRAY_SIZE(doms_meta); i++) { + const struct etnaviv_pm_domain_meta *meta = &doms_meta[i]; + + if (!(gpu->identity.features & meta->feature)) + continue; + + if (meta->nr_domains < (index - offset)) { + offset += meta->nr_domains; + continue; + } + + domain = meta->domains + (index - offset); + } + + return domain; +} + int etnaviv_pm_query_dom(struct etnaviv_gpu *gpu, struct drm_etnaviv_pm_domain *domain) { - const struct etnaviv_pm_domain_meta *meta = &doms_meta[domain->pipe]; + const unsigned int nr_domains = num_pm_domains(gpu); const struct etnaviv_pm_domain *dom; - if (domain->iter >= meta->nr_domains) + if (domain->iter >= nr_domains) return -EINVAL; - dom = meta->domains + domain->iter; + dom = pm_domain(gpu, domain->iter); + if (!dom) + return -EINVAL; domain->id = domain->iter; domain->nr_signals = dom->nr_signals; strncpy(domain->name, dom->name, sizeof(domain->name)); domain->iter++; - if (domain->iter == meta->nr_domains) + if (domain->iter == nr_domains) domain->iter = 0xff; return 0; @@ -448,14 +491,16 @@ int etnaviv_pm_query_dom(struct etnaviv_gpu *gpu, int etnaviv_pm_query_sig(struct etnaviv_gpu *gpu, struct drm_etnaviv_pm_signal *signal) { - const struct etnaviv_pm_domain_meta *meta = &doms_meta[signal->pipe]; + const unsigned int nr_domains = num_pm_domains(gpu); const struct etnaviv_pm_domain *dom; const struct etnaviv_pm_signal *sig; - if (signal->domain >= meta->nr_domains) + if (signal->domain >= nr_domains) return -EINVAL; - dom = meta->domains + signal->domain; + dom = pm_domain(gpu, signal->domain); + if (!dom) + return -EINVAL; if (signal->iter >= dom->nr_signals) return -EINVAL; diff --git a/drivers/gpu/drm/etnaviv/state_blt.xml.h b/drivers/gpu/drm/etnaviv/state_blt.xml.h index daae55995def..0e8bcf9dcc93 100644 --- a/drivers/gpu/drm/etnaviv/state_blt.xml.h +++ b/drivers/gpu/drm/etnaviv/state_blt.xml.h @@ -46,6 +46,8 @@ DEALINGS IN THE SOFTWARE. /* This is a cut-down version of the state_blt.xml.h file */ +#define VIVS_BLT_SET_COMMAND 0x000140ac + #define VIVS_BLT_ENABLE 0x000140b8 #define VIVS_BLT_ENABLE_ENABLE 0x00000001 diff --git a/drivers/gpu/drm/etnaviv/state_hi.xml.h b/drivers/gpu/drm/etnaviv/state_hi.xml.h index 41d8da2b6f4f..deaaa99fa654 100644 --- a/drivers/gpu/drm/etnaviv/state_hi.xml.h +++ b/drivers/gpu/drm/etnaviv/state_hi.xml.h @@ -8,17 +8,17 @@ http://0x04.net/cgit/index.cgi/rules-ng-ng git clone git://0x04.net/rules-ng-ng The rules-ng-ng source files this header was generated from are: -- state.xml ( 26087 bytes, from 2017-12-18 16:51:59) -- common.xml ( 35468 bytes, from 2018-01-22 13:48:54) -- common_3d.xml ( 14615 bytes, from 2017-12-18 16:51:59) -- state_hi.xml ( 30232 bytes, from 2018-02-15 15:48:01) -- copyright.xml ( 1597 bytes, from 2016-12-08 16:37:56) -- state_2d.xml ( 51552 bytes, from 2016-12-08 16:37:56) -- state_3d.xml ( 79992 bytes, from 2017-12-18 16:51:59) -- state_blt.xml ( 13405 bytes, from 2017-12-18 16:51:59) -- state_vg.xml ( 5975 bytes, from 2016-12-08 16:37:56) - -Copyright (C) 2012-2018 by the following authors: +- state.xml ( 26666 bytes, from 2019-12-20 21:20:35) +- common.xml ( 35468 bytes, from 2018-02-10 13:09:26) +- common_3d.xml ( 15058 bytes, from 2019-12-28 20:02:03) +- state_hi.xml ( 30552 bytes, from 2019-12-28 20:02:48) +- copyright.xml ( 1597 bytes, from 2018-02-10 13:09:26) +- state_2d.xml ( 51552 bytes, from 2018-02-10 13:09:26) +- state_3d.xml ( 83098 bytes, from 2019-12-28 20:02:03) +- state_blt.xml ( 14252 bytes, from 2019-10-20 19:59:15) +- state_vg.xml ( 5975 bytes, from 2018-02-10 13:09:26) + +Copyright (C) 2012-2019 by the following authors: - Wladimir J. van der Laan <laanwj@gmail.com> - Christian Gmeiner <christian.gmeiner@gmail.com> - Lucas Stach <l.stach@pengutronix.de> @@ -48,6 +48,9 @@ DEALINGS IN THE SOFTWARE. #define MMU_EXCEPTION_SLAVE_NOT_PRESENT 0x00000001 #define MMU_EXCEPTION_PAGE_NOT_PRESENT 0x00000002 #define MMU_EXCEPTION_WRITE_VIOLATION 0x00000003 +#define MMU_EXCEPTION_OUT_OF_BOUND 0x00000004 +#define MMU_EXCEPTION_READ_SECURITY_VIOLATION 0x00000005 +#define MMU_EXCEPTION_WRITE_SECURITY_VIOLATION 0x00000006 #define VIVS_HI 0x00000000 #define VIVS_HI_CLOCK_CONTROL 0x00000000 @@ -81,6 +84,13 @@ DEALINGS IN THE SOFTWARE. #define VIVS_HI_IDLE_STATE_IM 0x00000200 #define VIVS_HI_IDLE_STATE_FP 0x00000400 #define VIVS_HI_IDLE_STATE_TS 0x00000800 +#define VIVS_HI_IDLE_STATE_BL 0x00001000 +#define VIVS_HI_IDLE_STATE_ASYNCFE 0x00002000 +#define VIVS_HI_IDLE_STATE_MC 0x00004000 +#define VIVS_HI_IDLE_STATE_PPA 0x00008000 +#define VIVS_HI_IDLE_STATE_WD 0x00010000 +#define VIVS_HI_IDLE_STATE_NN 0x00020000 +#define VIVS_HI_IDLE_STATE_TP 0x00040000 #define VIVS_HI_IDLE_STATE_AXI_LP 0x80000000 #define VIVS_HI_AXI_CONFIG 0x00000008 @@ -140,6 +150,8 @@ DEALINGS IN THE SOFTWARE. #define VIVS_HI_CHIP_TIME 0x0000002c +#define VIVS_HI_CHIP_CUSTOMER_ID 0x00000030 + #define VIVS_HI_CHIP_MINOR_FEATURE_0 0x00000034 #define VIVS_HI_CACHE_CONTROL 0x00000038 @@ -237,6 +249,8 @@ DEALINGS IN THE SOFTWARE. #define VIVS_HI_BLT_INTR 0x000000d4 +#define VIVS_HI_CHIP_ECO_ID 0x000000e8 + #define VIVS_HI_AUXBIT 0x000000ec #define VIVS_PM 0x00000000 diff --git a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c index 8428ae12dfa5..1f79bc2a881e 100644 --- a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c +++ b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c @@ -55,6 +55,7 @@ static const char * const decon_clks_name[] = { struct decon_context { struct device *dev; struct drm_device *drm_dev; + void *dma_priv; struct exynos_drm_crtc *crtc; struct exynos_drm_plane planes[WINDOWS_NR]; struct exynos_drm_plane_config configs[WINDOWS_NR]; @@ -644,7 +645,7 @@ static int decon_bind(struct device *dev, struct device *master, void *data) decon_clear_channels(ctx->crtc); - return exynos_drm_register_dma(drm_dev, dev); + return exynos_drm_register_dma(drm_dev, dev, &ctx->dma_priv); } static void decon_unbind(struct device *dev, struct device *master, void *data) @@ -654,7 +655,7 @@ static void decon_unbind(struct device *dev, struct device *master, void *data) decon_atomic_disable(ctx->crtc); /* detach this sub driver from iommu mapping if supported. */ - exynos_drm_unregister_dma(ctx->drm_dev, ctx->dev); + exynos_drm_unregister_dma(ctx->drm_dev, ctx->dev, &ctx->dma_priv); } static const struct component_ops decon_component_ops = { diff --git a/drivers/gpu/drm/exynos/exynos7_drm_decon.c b/drivers/gpu/drm/exynos/exynos7_drm_decon.c index ff59c641fa80..f2d87a7445c7 100644 --- a/drivers/gpu/drm/exynos/exynos7_drm_decon.c +++ b/drivers/gpu/drm/exynos/exynos7_drm_decon.c @@ -40,6 +40,7 @@ struct decon_context { struct device *dev; struct drm_device *drm_dev; + void *dma_priv; struct exynos_drm_crtc *crtc; struct exynos_drm_plane planes[WINDOWS_NR]; struct exynos_drm_plane_config configs[WINDOWS_NR]; @@ -127,19 +128,19 @@ static int decon_ctx_initialize(struct decon_context *ctx, decon_clear_channels(ctx->crtc); - return exynos_drm_register_dma(drm_dev, ctx->dev); + return exynos_drm_register_dma(drm_dev, ctx->dev, &ctx->dma_priv); } static void decon_ctx_remove(struct decon_context *ctx) { /* detach this sub driver from iommu mapping if supported. */ - exynos_drm_unregister_dma(ctx->drm_dev, ctx->dev); + exynos_drm_unregister_dma(ctx->drm_dev, ctx->dev, &ctx->dma_priv); } static u32 decon_calc_clkdiv(struct decon_context *ctx, const struct drm_display_mode *mode) { - unsigned long ideal_clk = mode->htotal * mode->vtotal * mode->vrefresh; + unsigned long ideal_clk = mode->clock; u32 clkdiv; /* Find the clock divider value that gets us closest to ideal_clk */ diff --git a/drivers/gpu/drm/exynos/exynos_drm_dma.c b/drivers/gpu/drm/exynos/exynos_drm_dma.c index 9ebc02768847..619f81435c1b 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_dma.c +++ b/drivers/gpu/drm/exynos/exynos_drm_dma.c @@ -58,7 +58,7 @@ static inline void clear_dma_max_seg_size(struct device *dev) * mapping. */ static int drm_iommu_attach_device(struct drm_device *drm_dev, - struct device *subdrv_dev) + struct device *subdrv_dev, void **dma_priv) { struct exynos_drm_private *priv = drm_dev->dev_private; int ret; @@ -74,7 +74,14 @@ static int drm_iommu_attach_device(struct drm_device *drm_dev, return ret; if (IS_ENABLED(CONFIG_ARM_DMA_USE_IOMMU)) { - if (to_dma_iommu_mapping(subdrv_dev)) + /* + * Keep the original DMA mapping of the sub-device and + * restore it on Exynos DRM detach, otherwise the DMA + * framework considers it as IOMMU-less during the next + * probe (in case of deferred probe or modular build) + */ + *dma_priv = to_dma_iommu_mapping(subdrv_dev); + if (*dma_priv) arm_iommu_detach_device(subdrv_dev); ret = arm_iommu_attach_device(subdrv_dev, priv->mapping); @@ -98,19 +105,21 @@ static int drm_iommu_attach_device(struct drm_device *drm_dev, * mapping */ static void drm_iommu_detach_device(struct drm_device *drm_dev, - struct device *subdrv_dev) + struct device *subdrv_dev, void **dma_priv) { struct exynos_drm_private *priv = drm_dev->dev_private; - if (IS_ENABLED(CONFIG_ARM_DMA_USE_IOMMU)) + if (IS_ENABLED(CONFIG_ARM_DMA_USE_IOMMU)) { arm_iommu_detach_device(subdrv_dev); - else if (IS_ENABLED(CONFIG_IOMMU_DMA)) + arm_iommu_attach_device(subdrv_dev, *dma_priv); + } else if (IS_ENABLED(CONFIG_IOMMU_DMA)) iommu_detach_device(priv->mapping, subdrv_dev); clear_dma_max_seg_size(subdrv_dev); } -int exynos_drm_register_dma(struct drm_device *drm, struct device *dev) +int exynos_drm_register_dma(struct drm_device *drm, struct device *dev, + void **dma_priv) { struct exynos_drm_private *priv = drm->dev_private; @@ -137,13 +146,14 @@ int exynos_drm_register_dma(struct drm_device *drm, struct device *dev) priv->mapping = mapping; } - return drm_iommu_attach_device(drm, dev); + return drm_iommu_attach_device(drm, dev, dma_priv); } -void exynos_drm_unregister_dma(struct drm_device *drm, struct device *dev) +void exynos_drm_unregister_dma(struct drm_device *drm, struct device *dev, + void **dma_priv) { if (IS_ENABLED(CONFIG_EXYNOS_IOMMU)) - drm_iommu_detach_device(drm, dev); + drm_iommu_detach_device(drm, dev, dma_priv); } void exynos_drm_cleanup_dma(struct drm_device *drm) diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.c b/drivers/gpu/drm/exynos/exynos_drm_drv.c index ba0f868b2477..57defeb44522 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_drv.c +++ b/drivers/gpu/drm/exynos/exynos_drm_drv.c @@ -270,7 +270,7 @@ static int exynos_drm_bind(struct device *dev) struct drm_encoder *encoder; struct drm_device *drm; unsigned int clone_mask; - int cnt, ret; + int ret; drm = drm_dev_alloc(&exynos_drm_driver, dev); if (IS_ERR(drm)) @@ -293,10 +293,9 @@ static int exynos_drm_bind(struct device *dev) exynos_drm_mode_config_init(drm); /* setup possible_clones. */ - cnt = 0; clone_mask = 0; list_for_each_entry(encoder, &drm->mode_config.encoder_list, head) - clone_mask |= (1 << (cnt++)); + clone_mask |= drm_encoder_mask(encoder); list_for_each_entry(encoder, &drm->mode_config.encoder_list, head) encoder->possible_clones = clone_mask; diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.h b/drivers/gpu/drm/exynos/exynos_drm_drv.h index d4d21d8cfb90..6ae9056e7a18 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_drv.h +++ b/drivers/gpu/drm/exynos/exynos_drm_drv.h @@ -223,8 +223,10 @@ static inline bool is_drm_iommu_supported(struct drm_device *drm_dev) return priv->mapping ? true : false; } -int exynos_drm_register_dma(struct drm_device *drm, struct device *dev); -void exynos_drm_unregister_dma(struct drm_device *drm, struct device *dev); +int exynos_drm_register_dma(struct drm_device *drm, struct device *dev, + void **dma_priv); +void exynos_drm_unregister_dma(struct drm_device *drm, struct device *dev, + void **dma_priv); void exynos_drm_cleanup_dma(struct drm_device *drm); #ifdef CONFIG_DRM_EXYNOS_DPI diff --git a/drivers/gpu/drm/exynos/exynos_drm_dsi.c b/drivers/gpu/drm/exynos/exynos_drm_dsi.c index 38e43d957ae0..e080aa92338c 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_dsi.c +++ b/drivers/gpu/drm/exynos/exynos_drm_dsi.c @@ -1772,8 +1772,9 @@ static int exynos_dsi_probe(struct platform_device *pdev) ret = devm_regulator_bulk_get(dev, ARRAY_SIZE(dsi->supplies), dsi->supplies); if (ret) { - dev_info(dev, "failed to get regulators: %d\n", ret); - return -EPROBE_DEFER; + if (ret != -EPROBE_DEFER) + dev_info(dev, "failed to get regulators: %d\n", ret); + return ret; } dsi->clks = devm_kcalloc(dev, @@ -1786,9 +1787,10 @@ static int exynos_dsi_probe(struct platform_device *pdev) dsi->clks[i] = devm_clk_get(dev, clk_names[i]); if (IS_ERR(dsi->clks[i])) { if (strcmp(clk_names[i], "sclk_mipi") == 0) { - strcpy(clk_names[i], OLD_SCLK_MIPI_CLK_NAME); - i--; - continue; + dsi->clks[i] = devm_clk_get(dev, + OLD_SCLK_MIPI_CLK_NAME); + if (!IS_ERR(dsi->clks[i])) + continue; } dev_info(dev, "failed to get the clock: %s\n", diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimc.c b/drivers/gpu/drm/exynos/exynos_drm_fimc.c index 8ea2e1d77802..29ab8be8604c 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fimc.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fimc.c @@ -97,6 +97,7 @@ struct fimc_scaler { struct fimc_context { struct exynos_drm_ipp ipp; struct drm_device *drm_dev; + void *dma_priv; struct device *dev; struct exynos_drm_ipp_task *task; struct exynos_drm_ipp_formats *formats; @@ -1133,7 +1134,7 @@ static int fimc_bind(struct device *dev, struct device *master, void *data) ctx->drm_dev = drm_dev; ipp->drm_dev = drm_dev; - exynos_drm_register_dma(drm_dev, dev); + exynos_drm_register_dma(drm_dev, dev, &ctx->dma_priv); exynos_drm_ipp_register(dev, ipp, &ipp_funcs, DRM_EXYNOS_IPP_CAP_CROP | DRM_EXYNOS_IPP_CAP_ROTATE | @@ -1153,7 +1154,7 @@ static void fimc_unbind(struct device *dev, struct device *master, struct exynos_drm_ipp *ipp = &ctx->ipp; exynos_drm_ipp_unregister(dev, ipp); - exynos_drm_unregister_dma(drm_dev, dev); + exynos_drm_unregister_dma(drm_dev, dev, &ctx->dma_priv); } static const struct component_ops fimc_component_ops = { diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimd.c b/drivers/gpu/drm/exynos/exynos_drm_fimd.c index 21aec38702fc..bb67cad8371f 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fimd.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fimd.c @@ -167,6 +167,7 @@ static struct fimd_driver_data exynos5420_fimd_driver_data = { struct fimd_context { struct device *dev; struct drm_device *drm_dev; + void *dma_priv; struct exynos_drm_crtc *crtc; struct exynos_drm_plane planes[WINDOWS_NR]; struct exynos_drm_plane_config configs[WINDOWS_NR]; @@ -1090,7 +1091,7 @@ static int fimd_bind(struct device *dev, struct device *master, void *data) if (is_drm_iommu_supported(drm_dev)) fimd_clear_channels(ctx->crtc); - return exynos_drm_register_dma(drm_dev, dev); + return exynos_drm_register_dma(drm_dev, dev, &ctx->dma_priv); } static void fimd_unbind(struct device *dev, struct device *master, @@ -1100,7 +1101,7 @@ static void fimd_unbind(struct device *dev, struct device *master, fimd_atomic_disable(ctx->crtc); - exynos_drm_unregister_dma(ctx->drm_dev, ctx->dev); + exynos_drm_unregister_dma(ctx->drm_dev, ctx->dev, &ctx->dma_priv); if (ctx->encoder) exynos_dpi_remove(ctx->encoder); diff --git a/drivers/gpu/drm/exynos/exynos_drm_g2d.c b/drivers/gpu/drm/exynos/exynos_drm_g2d.c index 2a3382d43bc9..fcee33a43aca 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_g2d.c +++ b/drivers/gpu/drm/exynos/exynos_drm_g2d.c @@ -232,6 +232,7 @@ struct g2d_runqueue_node { struct g2d_data { struct device *dev; + void *dma_priv; struct clk *gate_clk; void __iomem *regs; int irq; @@ -1409,7 +1410,7 @@ static int g2d_bind(struct device *dev, struct device *master, void *data) return ret; } - ret = exynos_drm_register_dma(drm_dev, dev); + ret = exynos_drm_register_dma(drm_dev, dev, &g2d->dma_priv); if (ret < 0) { dev_err(dev, "failed to enable iommu.\n"); g2d_fini_cmdlist(g2d); @@ -1434,7 +1435,7 @@ static void g2d_unbind(struct device *dev, struct device *master, void *data) priv->g2d_dev = NULL; cancel_work_sync(&g2d->runqueue_work); - exynos_drm_unregister_dma(g2d->drm_dev, dev); + exynos_drm_unregister_dma(g2d->drm_dev, dev, &g2d->dma_priv); } static const struct component_ops g2d_component_ops = { diff --git a/drivers/gpu/drm/exynos/exynos_drm_gsc.c b/drivers/gpu/drm/exynos/exynos_drm_gsc.c index 88b6fcaa20be..45e9aee8366a 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_gsc.c +++ b/drivers/gpu/drm/exynos/exynos_drm_gsc.c @@ -97,6 +97,7 @@ struct gsc_scaler { struct gsc_context { struct exynos_drm_ipp ipp; struct drm_device *drm_dev; + void *dma_priv; struct device *dev; struct exynos_drm_ipp_task *task; struct exynos_drm_ipp_formats *formats; @@ -1169,7 +1170,7 @@ static int gsc_bind(struct device *dev, struct device *master, void *data) ctx->drm_dev = drm_dev; ctx->drm_dev = drm_dev; - exynos_drm_register_dma(drm_dev, dev); + exynos_drm_register_dma(drm_dev, dev, &ctx->dma_priv); exynos_drm_ipp_register(dev, ipp, &ipp_funcs, DRM_EXYNOS_IPP_CAP_CROP | DRM_EXYNOS_IPP_CAP_ROTATE | @@ -1189,7 +1190,7 @@ static void gsc_unbind(struct device *dev, struct device *master, struct exynos_drm_ipp *ipp = &ctx->ipp; exynos_drm_ipp_unregister(dev, ipp); - exynos_drm_unregister_dma(drm_dev, dev); + exynos_drm_unregister_dma(drm_dev, dev, &ctx->dma_priv); } static const struct component_ops gsc_component_ops = { diff --git a/drivers/gpu/drm/exynos/exynos_drm_rotator.c b/drivers/gpu/drm/exynos/exynos_drm_rotator.c index b98482990d1a..dafa87b82052 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_rotator.c +++ b/drivers/gpu/drm/exynos/exynos_drm_rotator.c @@ -56,6 +56,7 @@ struct rot_variant { struct rot_context { struct exynos_drm_ipp ipp; struct drm_device *drm_dev; + void *dma_priv; struct device *dev; void __iomem *regs; struct clk *clock; @@ -243,7 +244,7 @@ static int rotator_bind(struct device *dev, struct device *master, void *data) rot->drm_dev = drm_dev; ipp->drm_dev = drm_dev; - exynos_drm_register_dma(drm_dev, dev); + exynos_drm_register_dma(drm_dev, dev, &rot->dma_priv); exynos_drm_ipp_register(dev, ipp, &ipp_funcs, DRM_EXYNOS_IPP_CAP_CROP | DRM_EXYNOS_IPP_CAP_ROTATE, @@ -261,7 +262,7 @@ static void rotator_unbind(struct device *dev, struct device *master, struct exynos_drm_ipp *ipp = &rot->ipp; exynos_drm_ipp_unregister(dev, ipp); - exynos_drm_unregister_dma(rot->drm_dev, rot->dev); + exynos_drm_unregister_dma(rot->drm_dev, rot->dev, &rot->dma_priv); } static const struct component_ops rotator_component_ops = { diff --git a/drivers/gpu/drm/exynos/exynos_drm_scaler.c b/drivers/gpu/drm/exynos/exynos_drm_scaler.c index 497973e9b2c5..93c43c8d914e 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_scaler.c +++ b/drivers/gpu/drm/exynos/exynos_drm_scaler.c @@ -39,6 +39,7 @@ struct scaler_data { struct scaler_context { struct exynos_drm_ipp ipp; struct drm_device *drm_dev; + void *dma_priv; struct device *dev; void __iomem *regs; struct clk *clock[SCALER_MAX_CLK]; @@ -450,7 +451,7 @@ static int scaler_bind(struct device *dev, struct device *master, void *data) scaler->drm_dev = drm_dev; ipp->drm_dev = drm_dev; - exynos_drm_register_dma(drm_dev, dev); + exynos_drm_register_dma(drm_dev, dev, &scaler->dma_priv); exynos_drm_ipp_register(dev, ipp, &ipp_funcs, DRM_EXYNOS_IPP_CAP_CROP | DRM_EXYNOS_IPP_CAP_ROTATE | @@ -470,7 +471,8 @@ static void scaler_unbind(struct device *dev, struct device *master, struct exynos_drm_ipp *ipp = &scaler->ipp; exynos_drm_ipp_unregister(dev, ipp); - exynos_drm_unregister_dma(scaler->drm_dev, scaler->dev); + exynos_drm_unregister_dma(scaler->drm_dev, scaler->dev, + &scaler->dma_priv); } static const struct component_ops scaler_component_ops = { diff --git a/drivers/gpu/drm/exynos/exynos_hdmi.c b/drivers/gpu/drm/exynos/exynos_hdmi.c index 3e5f1a77286d..1a7c828fc41d 100644 --- a/drivers/gpu/drm/exynos/exynos_hdmi.c +++ b/drivers/gpu/drm/exynos/exynos_hdmi.c @@ -1805,18 +1805,10 @@ static int hdmi_resources_init(struct hdmi_context *hdata) hdata->reg_hdmi_en = devm_regulator_get_optional(dev, "hdmi-en"); - if (PTR_ERR(hdata->reg_hdmi_en) != -ENODEV) { + if (PTR_ERR(hdata->reg_hdmi_en) != -ENODEV) if (IS_ERR(hdata->reg_hdmi_en)) return PTR_ERR(hdata->reg_hdmi_en); - ret = regulator_enable(hdata->reg_hdmi_en); - if (ret) { - DRM_DEV_ERROR(dev, - "failed to enable hdmi-en regulator\n"); - return ret; - } - } - return hdmi_bridge_init(hdata); } @@ -2023,6 +2015,15 @@ static int hdmi_probe(struct platform_device *pdev) } } + if (!IS_ERR(hdata->reg_hdmi_en)) { + ret = regulator_enable(hdata->reg_hdmi_en); + if (ret) { + DRM_DEV_ERROR(dev, + "failed to enable hdmi-en regulator\n"); + goto err_hdmiphy; + } + } + pm_runtime_enable(dev); audio_infoframe = &hdata->audio.infoframe; @@ -2047,7 +2048,8 @@ err_unregister_audio: err_rpm_disable: pm_runtime_disable(dev); - + if (!IS_ERR(hdata->reg_hdmi_en)) + regulator_disable(hdata->reg_hdmi_en); err_hdmiphy: if (hdata->hdmiphy_port) put_device(&hdata->hdmiphy_port->dev); diff --git a/drivers/gpu/drm/exynos/exynos_mixer.c b/drivers/gpu/drm/exynos/exynos_mixer.c index 38ae9c32feef..21b726baedea 100644 --- a/drivers/gpu/drm/exynos/exynos_mixer.c +++ b/drivers/gpu/drm/exynos/exynos_mixer.c @@ -94,6 +94,7 @@ struct mixer_context { struct platform_device *pdev; struct device *dev; struct drm_device *drm_dev; + void *dma_priv; struct exynos_drm_crtc *crtc; struct exynos_drm_plane planes[MIXER_WIN_NR]; unsigned long flags; @@ -894,12 +895,14 @@ static int mixer_initialize(struct mixer_context *mixer_ctx, } } - return exynos_drm_register_dma(drm_dev, mixer_ctx->dev); + return exynos_drm_register_dma(drm_dev, mixer_ctx->dev, + &mixer_ctx->dma_priv); } static void mixer_ctx_remove(struct mixer_context *mixer_ctx) { - exynos_drm_unregister_dma(mixer_ctx->drm_dev, mixer_ctx->dev); + exynos_drm_unregister_dma(mixer_ctx->drm_dev, mixer_ctx->dev, + &mixer_ctx->dma_priv); } static int mixer_enable_vblank(struct exynos_drm_crtc *crtc) diff --git a/drivers/gpu/drm/hisilicon/kirin/kirin_ade_reg.h b/drivers/gpu/drm/hisilicon/kirin/kirin_ade_reg.h index 0da860200410..e2ac09894a6d 100644 --- a/drivers/gpu/drm/hisilicon/kirin/kirin_ade_reg.h +++ b/drivers/gpu/drm/hisilicon/kirin/kirin_ade_reg.h @@ -83,7 +83,6 @@ #define VSIZE_OFST 20 #define LDI_INT_EN 0x741C #define FRAME_END_INT_EN_OFST 1 -#define UNDERFLOW_INT_EN_OFST 2 #define LDI_CTRL 0x7420 #define BPP_OFST 3 #define DATA_GATE_EN BIT(2) diff --git a/drivers/gpu/drm/hisilicon/kirin/kirin_drm_ade.c b/drivers/gpu/drm/hisilicon/kirin/kirin_drm_ade.c index 73cd28a6ea07..86000127d4ee 100644 --- a/drivers/gpu/drm/hisilicon/kirin/kirin_drm_ade.c +++ b/drivers/gpu/drm/hisilicon/kirin/kirin_drm_ade.c @@ -46,7 +46,6 @@ struct ade_hw_ctx { struct clk *media_noc_clk; struct clk *ade_pix_clk; struct reset_control *reset; - struct work_struct display_reset_wq; bool power_on; int irq; @@ -136,7 +135,6 @@ static void ade_init(struct ade_hw_ctx *ctx) */ ade_update_bits(base + ADE_CTRL, FRM_END_START_OFST, FRM_END_START_MASK, REG_EFFECTIVE_IN_ADEEN_FRMEND); - ade_update_bits(base + LDI_INT_EN, UNDERFLOW_INT_EN_OFST, MASK(1), 1); } static bool ade_crtc_mode_fixup(struct drm_crtc *crtc, @@ -304,17 +302,6 @@ static void ade_crtc_disable_vblank(struct drm_crtc *crtc) MASK(1), 0); } -static void drm_underflow_wq(struct work_struct *work) -{ - struct ade_hw_ctx *ctx = container_of(work, struct ade_hw_ctx, - display_reset_wq); - struct drm_device *drm_dev = ctx->crtc->dev; - struct drm_atomic_state *state; - - state = drm_atomic_helper_suspend(drm_dev); - drm_atomic_helper_resume(drm_dev, state); -} - static irqreturn_t ade_irq_handler(int irq, void *data) { struct ade_hw_ctx *ctx = data; @@ -331,12 +318,6 @@ static irqreturn_t ade_irq_handler(int irq, void *data) MASK(1), 1); drm_crtc_handle_vblank(crtc); } - if (status & BIT(UNDERFLOW_INT_EN_OFST)) { - ade_update_bits(base + LDI_INT_CLR, UNDERFLOW_INT_EN_OFST, - MASK(1), 1); - DRM_ERROR("LDI underflow!"); - schedule_work(&ctx->display_reset_wq); - } return IRQ_HANDLED; } @@ -919,7 +900,6 @@ static void *ade_hw_ctx_alloc(struct platform_device *pdev, if (ret) return ERR_PTR(-EIO); - INIT_WORK(&ctx->display_reset_wq, drm_underflow_wq); ctx->crtc = crtc; return ctx; diff --git a/drivers/gpu/drm/i915/Kconfig.profile b/drivers/gpu/drm/i915/Kconfig.profile index c280b6ae38eb..0bfd276c19fe 100644 --- a/drivers/gpu/drm/i915/Kconfig.profile +++ b/drivers/gpu/drm/i915/Kconfig.profile @@ -20,6 +20,9 @@ config DRM_I915_HEARTBEAT_INTERVAL check the health of the GPU and undertake regular house-keeping of internal driver state. + This is adjustable via + /sys/class/drm/card?/engine/*/heartbeat_interval_ms + May be 0 to disable heartbeats and therefore disable automatic GPU hang detection. @@ -33,11 +36,18 @@ config DRM_I915_PREEMPT_TIMEOUT expires, the HW will be reset to allow the more important context to execute. + This is adjustable via + /sys/class/drm/card?/engine/*/preempt_timeout_ms + May be 0 to disable the timeout. -config DRM_I915_SPIN_REQUEST - int "Busywait for request completion (us)" - default 5 # microseconds + The compiled in default may get overridden at driver probe time on + certain platforms and certain engines which will be reflected in the + sysfs control. + +config DRM_I915_MAX_REQUEST_BUSYWAIT + int "Busywait for request completion limit (ns)" + default 8000 # nanoseconds help Before sleeping waiting for a request (GPU operation) to complete, we may spend some time polling for its completion. As the IRQ may @@ -45,6 +55,9 @@ config DRM_I915_SPIN_REQUEST check if the request will complete in the time it would have taken us to enable the interrupt. + This is adjustable via + /sys/class/drm/card?/engine/*/max_busywait_duration_ns + May be 0 to disable the initial spin. In practice, we estimate the cost of enabling the interrupt (if currently disabled) to be a few microseconds. @@ -60,6 +73,9 @@ config DRM_I915_STOP_TIMEOUT that the reset itself may take longer and so be more disruptive to interactive or low latency workloads. + This is adjustable via + /sys/class/drm/card?/engine/*/stop_timeout_ms + config DRM_I915_TIMESLICE_DURATION int "Scheduling quantum for userspace batches (ms, jiffy granularity)" default 1 # milliseconds @@ -73,4 +89,7 @@ config DRM_I915_TIMESLICE_DURATION is scheduled for execution for the timeslice duration, before switching to the next context. + This is adjustable via + /sys/class/drm/card?/engine/*/timeslice_duration_ms + May be 0 to disable timeslicing. diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile index bc28c31c4f78..9f887a86e555 100644 --- a/drivers/gpu/drm/i915/Makefile +++ b/drivers/gpu/drm/i915/Makefile @@ -47,6 +47,7 @@ i915-y += i915_drv.o \ i915_sysfs.o \ i915_utils.o \ intel_device_info.o \ + intel_dram.o \ intel_memory_region.o \ intel_pch.o \ intel_pm.o \ @@ -79,9 +80,11 @@ gt-y += \ gt/debugfs_gt.o \ gt/debugfs_gt_pm.o \ gt/gen6_ppgtt.o \ + gt/gen7_renderclear.o \ gt/gen8_ppgtt.o \ gt/intel_breadcrumbs.o \ gt/intel_context.o \ + gt/intel_context_param.o \ gt/intel_context_sseu.o \ gt/intel_engine_cs.o \ gt/intel_engine_heartbeat.o \ @@ -107,7 +110,8 @@ gt-y += \ gt/intel_rps.o \ gt/intel_sseu.o \ gt/intel_timeline.o \ - gt/intel_workarounds.o + gt/intel_workarounds.o \ + gt/sysfs_engines.o # autogenerated null render state gt-y += \ gt/gen6_renderstate.o \ diff --git a/drivers/gpu/drm/i915/display/icl_dsi.c b/drivers/gpu/drm/i915/display/icl_dsi.c index d842e280699d..17cee6f80d8b 100644 --- a/drivers/gpu/drm/i915/display/icl_dsi.c +++ b/drivers/gpu/drm/i915/display/icl_dsi.c @@ -599,13 +599,13 @@ static void gen11_dsi_gate_clocks(struct intel_encoder *encoder) u32 tmp; enum phy phy; - mutex_lock(&dev_priv->dpll_lock); + mutex_lock(&dev_priv->dpll.lock); tmp = intel_de_read(dev_priv, ICL_DPCLKA_CFGCR0); for_each_dsi_phy(phy, intel_dsi->phys) tmp |= ICL_DPCLKA_CFGCR0_DDI_CLK_OFF(phy); intel_de_write(dev_priv, ICL_DPCLKA_CFGCR0, tmp); - mutex_unlock(&dev_priv->dpll_lock); + mutex_unlock(&dev_priv->dpll.lock); } static void gen11_dsi_ungate_clocks(struct intel_encoder *encoder) @@ -615,13 +615,13 @@ static void gen11_dsi_ungate_clocks(struct intel_encoder *encoder) u32 tmp; enum phy phy; - mutex_lock(&dev_priv->dpll_lock); + mutex_lock(&dev_priv->dpll.lock); tmp = intel_de_read(dev_priv, ICL_DPCLKA_CFGCR0); for_each_dsi_phy(phy, intel_dsi->phys) tmp &= ~ICL_DPCLKA_CFGCR0_DDI_CLK_OFF(phy); intel_de_write(dev_priv, ICL_DPCLKA_CFGCR0, tmp); - mutex_unlock(&dev_priv->dpll_lock); + mutex_unlock(&dev_priv->dpll.lock); } static void gen11_dsi_map_pll(struct intel_encoder *encoder, @@ -633,7 +633,7 @@ static void gen11_dsi_map_pll(struct intel_encoder *encoder, enum phy phy; u32 val; - mutex_lock(&dev_priv->dpll_lock); + mutex_lock(&dev_priv->dpll.lock); val = intel_de_read(dev_priv, ICL_DPCLKA_CFGCR0); for_each_dsi_phy(phy, intel_dsi->phys) { @@ -652,7 +652,7 @@ static void gen11_dsi_map_pll(struct intel_encoder *encoder, intel_de_posting_read(dev_priv, ICL_DPCLKA_CFGCR0); - mutex_unlock(&dev_priv->dpll_lock); + mutex_unlock(&dev_priv->dpll.lock); } static void @@ -1350,15 +1350,15 @@ static void gen11_dsi_get_timings(struct intel_encoder *encoder, static void gen11_dsi_get_config(struct intel_encoder *encoder, struct intel_crtc_state *pipe_config) { - struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); + struct drm_i915_private *i915 = to_i915(encoder->base.dev); struct intel_crtc *crtc = to_intel_crtc(pipe_config->uapi.crtc); struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder); intel_dsc_get_config(encoder, pipe_config); /* FIXME: adapt icl_ddi_clock_get() for DSI and use that? */ - pipe_config->port_clock = - cnl_calc_wrpll_link(dev_priv, &pipe_config->dpll_hw_state); + pipe_config->port_clock = intel_dpll_get_freq(i915, + pipe_config->shared_dpll); pipe_config->hw.adjusted_mode.crtc_clock = intel_dsi->pclk; if (intel_dsi->dual_link) diff --git a/drivers/gpu/drm/i915/display/intel_atomic_plane.c b/drivers/gpu/drm/i915/display/intel_atomic_plane.c index c86d7a35c816..457b258683d3 100644 --- a/drivers/gpu/drm/i915/display/intel_atomic_plane.c +++ b/drivers/gpu/drm/i915/display/intel_atomic_plane.c @@ -133,15 +133,37 @@ intel_plane_destroy_state(struct drm_plane *plane, kfree(plane_state); } +unsigned int intel_plane_pixel_rate(const struct intel_crtc_state *crtc_state, + const struct intel_plane_state *plane_state) +{ + unsigned int src_w, src_h, dst_w, dst_h; + unsigned int pixel_rate = crtc_state->pixel_rate; + + src_w = drm_rect_width(&plane_state->uapi.src) >> 16; + src_h = drm_rect_height(&plane_state->uapi.src) >> 16; + dst_w = drm_rect_width(&plane_state->uapi.dst); + dst_h = drm_rect_height(&plane_state->uapi.dst); + + /* Downscaling limits the maximum pixel rate */ + dst_w = min(src_w, dst_w); + dst_h = min(src_h, dst_h); + + return DIV_ROUND_UP_ULL(mul_u32_u32(pixel_rate, src_w * src_h), + dst_w * dst_h); +} + unsigned int intel_plane_data_rate(const struct intel_crtc_state *crtc_state, const struct intel_plane_state *plane_state) { const struct drm_framebuffer *fb = plane_state->hw.fb; unsigned int cpp; + unsigned int pixel_rate; if (!plane_state->uapi.visible) return 0; + pixel_rate = intel_plane_pixel_rate(crtc_state, plane_state); + cpp = fb->format->cpp[0]; /* @@ -153,7 +175,7 @@ unsigned int intel_plane_data_rate(const struct intel_crtc_state *crtc_state, if (fb->format->is_yuv && fb->format->num_planes > 1) cpp *= 4; - return cpp * crtc_state->pixel_rate; + return pixel_rate * cpp; } int intel_plane_calc_min_cdclk(struct intel_atomic_state *state, diff --git a/drivers/gpu/drm/i915/display/intel_atomic_plane.h b/drivers/gpu/drm/i915/display/intel_atomic_plane.h index 2bcf15e34728..a6bbf42bae1f 100644 --- a/drivers/gpu/drm/i915/display/intel_atomic_plane.h +++ b/drivers/gpu/drm/i915/display/intel_atomic_plane.h @@ -18,6 +18,9 @@ struct intel_plane_state; extern const struct drm_plane_helper_funcs intel_plane_helper_funcs; +unsigned int intel_plane_pixel_rate(const struct intel_crtc_state *crtc_state, + const struct intel_plane_state *plane_state); + unsigned int intel_plane_data_rate(const struct intel_crtc_state *crtc_state, const struct intel_plane_state *plane_state); void intel_plane_copy_uapi_to_hw_state(struct intel_plane_state *plane_state, diff --git a/drivers/gpu/drm/i915/display/intel_audio.c b/drivers/gpu/drm/i915/display/intel_audio.c index 30fb7c887ff0..62f234f641de 100644 --- a/drivers/gpu/drm/i915/display/intel_audio.c +++ b/drivers/gpu/drm/i915/display/intel_audio.c @@ -149,6 +149,10 @@ static const struct { { 74250, AUD_CONFIG_PIXEL_CLOCK_HDMI_74250 }, { 148352, AUD_CONFIG_PIXEL_CLOCK_HDMI_148352 }, { 148500, AUD_CONFIG_PIXEL_CLOCK_HDMI_148500 }, + { 296703, AUD_CONFIG_PIXEL_CLOCK_HDMI_296703 }, + { 297000, AUD_CONFIG_PIXEL_CLOCK_HDMI_297000 }, + { 593407, AUD_CONFIG_PIXEL_CLOCK_HDMI_593407 }, + { 594000, AUD_CONFIG_PIXEL_CLOCK_HDMI_594000 }, }; /* HDMI N/CTS table */ @@ -234,6 +238,7 @@ static const struct hdmi_aud_ncts hdmi_aud_ncts_36bpp[] = { /* get AUD_CONFIG_PIXEL_CLOCK_HDMI_* value for mode */ static u32 audio_config_hdmi_pixel_clock(const struct intel_crtc_state *crtc_state) { + struct drm_i915_private *dev_priv = to_i915(crtc_state->uapi.crtc->dev); const struct drm_display_mode *adjusted_mode = &crtc_state->hw.adjusted_mode; int i; @@ -243,6 +248,9 @@ static u32 audio_config_hdmi_pixel_clock(const struct intel_crtc_state *crtc_sta break; } + if (INTEL_GEN(dev_priv) < 12 && adjusted_mode->crtc_clock > 148500) + i = ARRAY_SIZE(hdmi_audio_clock); + if (i == ARRAY_SIZE(hdmi_audio_clock)) { DRM_DEBUG_KMS("HDMI audio pixel clock setting for %d not found, falling back to defaults\n", adjusted_mode->crtc_clock); @@ -844,7 +852,7 @@ static void glk_force_audio_cdclk(struct drm_i915_private *dev_priv, struct intel_crtc *crtc; int ret; - crtc = intel_get_crtc_for_pipe(dev_priv, PIPE_A); + crtc = intel_get_first_crtc(dev_priv); if (!crtc) return; diff --git a/drivers/gpu/drm/i915/display/intel_bios.c b/drivers/gpu/drm/i915/display/intel_bios.c index 2049cf5b54f3..839124647202 100644 --- a/drivers/gpu/drm/i915/display/intel_bios.c +++ b/drivers/gpu/drm/i915/display/intel_bios.c @@ -26,7 +26,6 @@ */ #include <drm/drm_dp_helper.h> -#include <drm/i915_drm.h> #include "display/intel_display.h" #include "display/intel_display_types.h" diff --git a/drivers/gpu/drm/i915/display/intel_bios.h b/drivers/gpu/drm/i915/display/intel_bios.h index c17199caeff8..e29e79faa01b 100644 --- a/drivers/gpu/drm/i915/display/intel_bios.h +++ b/drivers/gpu/drm/i915/display/intel_bios.h @@ -32,8 +32,6 @@ #include <linux/types.h> -#include <drm/i915_drm.h> - struct drm_i915_private; struct intel_crtc_state; struct intel_encoder; diff --git a/drivers/gpu/drm/i915/display/intel_cdclk.c b/drivers/gpu/drm/i915/display/intel_cdclk.c index 0741d643455b..979a0241fdcb 100644 --- a/drivers/gpu/drm/i915/display/intel_cdclk.c +++ b/drivers/gpu/drm/i915/display/intel_cdclk.c @@ -1868,6 +1868,8 @@ static void intel_set_cdclk(struct drm_i915_private *dev_priv, const struct intel_cdclk_config *cdclk_config, enum pipe pipe) { + struct intel_encoder *encoder; + if (!intel_cdclk_changed(&dev_priv->cdclk.hw, cdclk_config)) return; @@ -1876,8 +1878,28 @@ static void intel_set_cdclk(struct drm_i915_private *dev_priv, intel_dump_cdclk_config(cdclk_config, "Changing CDCLK to"); + /* + * Lock aux/gmbus while we change cdclk in case those + * functions use cdclk. Not all platforms/ports do, + * but we'll lock them all for simplicity. + */ + mutex_lock(&dev_priv->gmbus_mutex); + for_each_intel_dp(&dev_priv->drm, encoder) { + struct intel_dp *intel_dp = enc_to_intel_dp(encoder); + + mutex_lock_nest_lock(&intel_dp->aux.hw_mutex, + &dev_priv->gmbus_mutex); + } + dev_priv->display.set_cdclk(dev_priv, cdclk_config, pipe); + for_each_intel_dp(&dev_priv->drm, encoder) { + struct intel_dp *intel_dp = enc_to_intel_dp(encoder); + + mutex_unlock(&intel_dp->aux.hw_mutex); + } + mutex_unlock(&dev_priv->gmbus_mutex); + if (drm_WARN(&dev_priv->drm, intel_cdclk_changed(&dev_priv->cdclk.hw, cdclk_config), "cdclk state doesn't match!\n")) { diff --git a/drivers/gpu/drm/i915/display/intel_color.c b/drivers/gpu/drm/i915/display/intel_color.c index 36dd52d2a9ee..c1cce93a1c25 100644 --- a/drivers/gpu/drm/i915/display/intel_color.c +++ b/drivers/gpu/drm/i915/display/intel_color.c @@ -348,48 +348,56 @@ static void icl_load_csc_matrix(const struct intel_crtc_state *crtc_state) crtc_state->csc_mode); } -/* - * Set up the pipe CSC unit on CherryView. - */ -static void cherryview_load_csc_matrix(const struct intel_crtc_state *crtc_state) +static void chv_load_cgm_csc(struct intel_crtc *crtc, + const struct drm_property_blob *blob) { - struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); + const struct drm_color_ctm *ctm = blob->data; enum pipe pipe = crtc->pipe; + u16 coeffs[9]; + int i; - if (crtc_state->hw.ctm) { - const struct drm_color_ctm *ctm = crtc_state->hw.ctm->data; - u16 coeffs[9] = {}; - int i; - - for (i = 0; i < ARRAY_SIZE(coeffs); i++) { - u64 abs_coeff = - ((1ULL << 63) - 1) & ctm->matrix[i]; - - /* Round coefficient. */ - abs_coeff += 1 << (32 - 13); - /* Clamp to hardware limits. */ - abs_coeff = clamp_val(abs_coeff, 0, CTM_COEFF_8_0 - 1); - - /* Write coefficients in S3.12 format. */ - if (ctm->matrix[i] & (1ULL << 63)) - coeffs[i] = 1 << 15; - coeffs[i] |= ((abs_coeff >> 32) & 7) << 12; - coeffs[i] |= (abs_coeff >> 20) & 0xfff; - } + for (i = 0; i < ARRAY_SIZE(coeffs); i++) { + u64 abs_coeff = ((1ULL << 63) - 1) & ctm->matrix[i]; + + /* Round coefficient. */ + abs_coeff += 1 << (32 - 13); + /* Clamp to hardware limits. */ + abs_coeff = clamp_val(abs_coeff, 0, CTM_COEFF_8_0 - 1); - intel_de_write(dev_priv, CGM_PIPE_CSC_COEFF01(pipe), - coeffs[1] << 16 | coeffs[0]); - intel_de_write(dev_priv, CGM_PIPE_CSC_COEFF23(pipe), - coeffs[3] << 16 | coeffs[2]); - intel_de_write(dev_priv, CGM_PIPE_CSC_COEFF45(pipe), - coeffs[5] << 16 | coeffs[4]); - intel_de_write(dev_priv, CGM_PIPE_CSC_COEFF67(pipe), - coeffs[7] << 16 | coeffs[6]); - intel_de_write(dev_priv, CGM_PIPE_CSC_COEFF8(pipe), coeffs[8]); + coeffs[i] = 0; + + /* Write coefficients in S3.12 format. */ + if (ctm->matrix[i] & (1ULL << 63)) + coeffs[i] |= 1 << 15; + + coeffs[i] |= ((abs_coeff >> 32) & 7) << 12; + coeffs[i] |= (abs_coeff >> 20) & 0xfff; } - intel_de_write(dev_priv, CGM_PIPE_MODE(pipe), crtc_state->cgm_mode); + intel_de_write(dev_priv, CGM_PIPE_CSC_COEFF01(pipe), + coeffs[1] << 16 | coeffs[0]); + intel_de_write(dev_priv, CGM_PIPE_CSC_COEFF23(pipe), + coeffs[3] << 16 | coeffs[2]); + intel_de_write(dev_priv, CGM_PIPE_CSC_COEFF45(pipe), + coeffs[5] << 16 | coeffs[4]); + intel_de_write(dev_priv, CGM_PIPE_CSC_COEFF67(pipe), + coeffs[7] << 16 | coeffs[6]); + intel_de_write(dev_priv, CGM_PIPE_CSC_COEFF8(pipe), + coeffs[8]); +} + +/* convert hw value with given bit_precision to lut property val */ +static u32 intel_color_lut_pack(u32 val, int bit_precision) +{ + u32 max = 0xffff >> (16 - bit_precision); + + val = clamp_val(val, 0, max); + + if (bit_precision < 16) + val <<= 16 - bit_precision; + + return val; } static u32 i9xx_lut_8(const struct drm_color_lut *color) @@ -399,6 +407,13 @@ static u32 i9xx_lut_8(const struct drm_color_lut *color) drm_color_lut_extract(color->blue, 8); } +static void i9xx_lut_8_pack(struct drm_color_lut *entry, u32 val) +{ + entry->red = intel_color_lut_pack(REG_FIELD_GET(LGC_PALETTE_RED_MASK, val), 8); + entry->green = intel_color_lut_pack(REG_FIELD_GET(LGC_PALETTE_GREEN_MASK, val), 8); + entry->blue = intel_color_lut_pack(REG_FIELD_GET(LGC_PALETTE_BLUE_MASK, val), 8); +} + /* i965+ "10.6" bit interpolated format "even DW" (low 8 bits) */ static u32 i965_lut_10p6_ldw(const struct drm_color_lut *color) { @@ -415,48 +430,34 @@ static u32 i965_lut_10p6_udw(const struct drm_color_lut *color) (color->blue >> 8); } -static u32 ilk_lut_10(const struct drm_color_lut *color) +static void i965_lut_10p6_pack(struct drm_color_lut *entry, u32 ldw, u32 udw) { - return drm_color_lut_extract(color->red, 10) << 20 | - drm_color_lut_extract(color->green, 10) << 10 | - drm_color_lut_extract(color->blue, 10); + entry->red = REG_FIELD_GET(PALETTE_RED_MASK, udw) << 8 | + REG_FIELD_GET(PALETTE_RED_MASK, ldw); + entry->green = REG_FIELD_GET(PALETTE_GREEN_MASK, udw) << 8 | + REG_FIELD_GET(PALETTE_GREEN_MASK, ldw); + entry->blue = REG_FIELD_GET(PALETTE_BLUE_MASK, udw) << 8 | + REG_FIELD_GET(PALETTE_BLUE_MASK, ldw); } -/* Loads the legacy palette/gamma unit for the CRTC. */ -static void i9xx_load_luts_internal(const struct intel_crtc_state *crtc_state, - const struct drm_property_blob *blob) +static u16 i965_lut_11p6_max_pack(u32 val) { - struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); - struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); - enum pipe pipe = crtc->pipe; - int i; - - if (HAS_GMCH(dev_priv)) { - if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_DSI)) - assert_dsi_pll_enabled(dev_priv); - else - assert_pll_enabled(dev_priv, pipe); - } - - if (blob) { - const struct drm_color_lut *lut = blob->data; - - for (i = 0; i < 256; i++) { - u32 word = i9xx_lut_8(&lut[i]); + /* PIPEGCMAX is 11.6, clamp to 10.6 */ + return clamp_val(val, 0, 0xffff); +} - if (HAS_GMCH(dev_priv)) - intel_de_write(dev_priv, PALETTE(pipe, i), - word); - else - intel_de_write(dev_priv, LGC_PALETTE(pipe, i), - word); - } - } +static u32 ilk_lut_10(const struct drm_color_lut *color) +{ + return drm_color_lut_extract(color->red, 10) << 20 | + drm_color_lut_extract(color->green, 10) << 10 | + drm_color_lut_extract(color->blue, 10); } -static void i9xx_load_luts(const struct intel_crtc_state *crtc_state) +static void ilk_lut_10_pack(struct drm_color_lut *entry, u32 val) { - i9xx_load_luts_internal(crtc_state, crtc_state->hw.gamma_lut); + entry->red = intel_color_lut_pack(REG_FIELD_GET(PREC_PALETTE_RED_MASK, val), 10); + entry->green = intel_color_lut_pack(REG_FIELD_GET(PREC_PALETTE_GREEN_MASK, val), 10); + entry->blue = intel_color_lut_pack(REG_FIELD_GET(PREC_PALETTE_BLUE_MASK, val), 10); } static void i9xx_color_commit(const struct intel_crtc_state *crtc_state) @@ -525,6 +526,35 @@ static void skl_color_commit(const struct intel_crtc_state *crtc_state) ilk_load_csc_matrix(crtc_state); } +static void i9xx_load_lut_8(struct intel_crtc *crtc, + const struct drm_property_blob *blob) +{ + struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); + const struct drm_color_lut *lut; + enum pipe pipe = crtc->pipe; + int i; + + if (!blob) + return; + + lut = blob->data; + + for (i = 0; i < 256; i++) + intel_de_write(dev_priv, PALETTE(pipe, i), + i9xx_lut_8(&lut[i])); +} + +static void i9xx_load_luts(const struct intel_crtc_state *crtc_state) +{ + struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); + struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); + const struct drm_property_blob *gamma_lut = crtc_state->hw.gamma_lut; + + assert_pll_enabled(dev_priv, crtc->pipe); + + i9xx_load_lut_8(crtc, gamma_lut); +} + static void i965_load_lut_10p6(struct intel_crtc *crtc, const struct drm_property_blob *blob) { @@ -548,14 +578,38 @@ static void i965_load_lut_10p6(struct intel_crtc *crtc, static void i965_load_luts(const struct intel_crtc_state *crtc_state) { struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); + struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); const struct drm_property_blob *gamma_lut = crtc_state->hw.gamma_lut; + if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_DSI)) + assert_dsi_pll_enabled(dev_priv); + else + assert_pll_enabled(dev_priv, crtc->pipe); + if (crtc_state->gamma_mode == GAMMA_MODE_MODE_8BIT) - i9xx_load_luts(crtc_state); + i9xx_load_lut_8(crtc, gamma_lut); else i965_load_lut_10p6(crtc, gamma_lut); } +static void ilk_load_lut_8(struct intel_crtc *crtc, + const struct drm_property_blob *blob) +{ + struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); + const struct drm_color_lut *lut; + enum pipe pipe = crtc->pipe; + int i; + + if (!blob) + return; + + lut = blob->data; + + for (i = 0; i < 256; i++) + intel_de_write(dev_priv, LGC_PALETTE(pipe, i), + i9xx_lut_8(&lut[i])); +} + static void ilk_load_lut_10(struct intel_crtc *crtc, const struct drm_property_blob *blob) { @@ -566,7 +620,7 @@ static void ilk_load_lut_10(struct intel_crtc *crtc, for (i = 0; i < lut_size; i++) intel_de_write(dev_priv, PREC_PALETTE(pipe, i), - ilk_lut_10(&lut[i])); + ilk_lut_10(&lut[i])); } static void ilk_load_luts(const struct intel_crtc_state *crtc_state) @@ -575,7 +629,7 @@ static void ilk_load_luts(const struct intel_crtc_state *crtc_state) const struct drm_property_blob *gamma_lut = crtc_state->hw.gamma_lut; if (crtc_state->gamma_mode == GAMMA_MODE_MODE_8BIT) - i9xx_load_luts(crtc_state); + ilk_load_lut_8(crtc, gamma_lut); else ilk_load_lut_10(crtc, gamma_lut); } @@ -685,7 +739,7 @@ static void ivb_load_luts(const struct intel_crtc_state *crtc_state) const struct drm_property_blob *degamma_lut = crtc_state->hw.degamma_lut; if (crtc_state->gamma_mode == GAMMA_MODE_MODE_8BIT) { - i9xx_load_luts(crtc_state); + ilk_load_lut_8(crtc, gamma_lut); } else if (crtc_state->gamma_mode == GAMMA_MODE_MODE_SPLIT) { ivb_load_lut_10(crtc, degamma_lut, PAL_PREC_SPLIT_MODE | PAL_PREC_INDEX_VALUE(0)); @@ -708,7 +762,7 @@ static void bdw_load_luts(const struct intel_crtc_state *crtc_state) const struct drm_property_blob *degamma_lut = crtc_state->hw.degamma_lut; if (crtc_state->gamma_mode == GAMMA_MODE_MODE_8BIT) { - i9xx_load_luts(crtc_state); + ilk_load_lut_8(crtc, gamma_lut); } else if (crtc_state->gamma_mode == GAMMA_MODE_MODE_SPLIT) { bdw_load_lut_10(crtc, degamma_lut, PAL_PREC_SPLIT_MODE | PAL_PREC_INDEX_VALUE(0)); @@ -729,9 +783,8 @@ static void glk_load_degamma_lut(const struct intel_crtc_state *crtc_state) struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); enum pipe pipe = crtc->pipe; - const u32 lut_size = INTEL_INFO(dev_priv)->color.degamma_lut_size; + int i, lut_size = INTEL_INFO(dev_priv)->color.degamma_lut_size; const struct drm_color_lut *lut = crtc_state->hw.degamma_lut->data; - u32 i; /* * When setting the auto-increment bit, the hardware seems to @@ -770,8 +823,7 @@ static void glk_load_degamma_lut_linear(const struct intel_crtc_state *crtc_stat struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); enum pipe pipe = crtc->pipe; - const u32 lut_size = INTEL_INFO(dev_priv)->color.degamma_lut_size; - u32 i; + int i, lut_size = INTEL_INFO(dev_priv)->color.degamma_lut_size; /* * When setting the auto-increment bit, the hardware seems to @@ -812,7 +864,7 @@ static void glk_load_luts(const struct intel_crtc_state *crtc_state) glk_load_degamma_lut_linear(crtc_state); if (crtc_state->gamma_mode == GAMMA_MODE_MODE_8BIT) { - i9xx_load_luts(crtc_state); + ilk_load_lut_8(crtc, gamma_lut); } else { bdw_load_lut_10(crtc, gamma_lut, PAL_PREC_INDEX_VALUE(0)); ivb_load_lut_ext_max(crtc); @@ -856,7 +908,7 @@ icl_program_gamma_superfine_segment(const struct intel_crtc_state *crtc_state) const struct drm_color_lut *lut = blob->data; struct intel_dsb *dsb = intel_dsb_get(crtc); enum pipe pipe = crtc->pipe; - u32 i; + int i; /* * Program Super Fine segment (let's call it seg1)... @@ -889,7 +941,7 @@ icl_program_gamma_multi_segment(const struct intel_crtc_state *crtc_state) const struct drm_color_lut *entry; struct intel_dsb *dsb = intel_dsb_get(crtc); enum pipe pipe = crtc->pipe; - u32 i; + int i; /* * Program Fine segment (let's call it seg2)... @@ -948,7 +1000,7 @@ static void icl_load_luts(const struct intel_crtc_state *crtc_state) switch (crtc_state->gamma_mode & GAMMA_MODE_MODE_MASK) { case GAMMA_MODE_MODE_8BIT: - i9xx_load_luts(crtc_state); + ilk_load_lut_8(crtc, gamma_lut); break; case GAMMA_MODE_MODE_12BIT_MULTI_SEGMENTED: icl_program_gamma_superfine_segment(crtc_state); @@ -974,6 +1026,13 @@ static u32 chv_cgm_degamma_udw(const struct drm_color_lut *color) return drm_color_lut_extract(color->red, 14); } +static void chv_cgm_gamma_pack(struct drm_color_lut *entry, u32 ldw, u32 udw) +{ + entry->green = intel_color_lut_pack(REG_FIELD_GET(CGM_PIPE_GAMMA_GREEN_MASK, ldw), 10); + entry->blue = intel_color_lut_pack(REG_FIELD_GET(CGM_PIPE_GAMMA_BLUE_MASK, ldw), 10); + entry->red = intel_color_lut_pack(REG_FIELD_GET(CGM_PIPE_GAMMA_RED_MASK, udw), 10); +} + static void chv_load_cgm_degamma(struct intel_crtc *crtc, const struct drm_property_blob *blob) { @@ -1020,21 +1079,24 @@ static void chv_load_cgm_gamma(struct intel_crtc *crtc, static void chv_load_luts(const struct intel_crtc_state *crtc_state) { struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); - const struct drm_property_blob *gamma_lut = crtc_state->hw.gamma_lut; + struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); const struct drm_property_blob *degamma_lut = crtc_state->hw.degamma_lut; + const struct drm_property_blob *gamma_lut = crtc_state->hw.gamma_lut; + const struct drm_property_blob *ctm = crtc_state->hw.ctm; - cherryview_load_csc_matrix(crtc_state); + if (crtc_state->cgm_mode & CGM_PIPE_MODE_CSC) + chv_load_cgm_csc(crtc, ctm); - if (crtc_state_is_legacy_gamma(crtc_state)) { - i9xx_load_luts(crtc_state); - return; - } - - if (degamma_lut) + if (crtc_state->cgm_mode & CGM_PIPE_MODE_DEGAMMA) chv_load_cgm_degamma(crtc, degamma_lut); - if (gamma_lut) + if (crtc_state->cgm_mode & CGM_PIPE_MODE_GAMMA) chv_load_cgm_gamma(crtc, gamma_lut); + else + i965_load_luts(crtc_state); + + intel_de_write(dev_priv, CGM_PIPE_MODE(crtc->pipe), + crtc_state->cgm_mode); } void intel_color_load_luts(const struct intel_crtc_state *crtc_state) @@ -1660,28 +1722,13 @@ bool intel_color_lut_equal(struct drm_property_blob *blob1, return true; } -/* convert hw value with given bit_precision to lut property val */ -static u32 intel_color_lut_pack(u32 val, u32 bit_precision) +static struct drm_property_blob *i9xx_read_lut_8(struct intel_crtc *crtc) { - u32 max = 0xffff >> (16 - bit_precision); - - val = clamp_val(val, 0, max); - - if (bit_precision < 16) - val <<= 16 - bit_precision; - - return val; -} - -static struct drm_property_blob * -i9xx_read_lut_8(const struct intel_crtc_state *crtc_state) -{ - struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); enum pipe pipe = crtc->pipe; struct drm_property_blob *blob; - struct drm_color_lut *blob_data; - u32 i, val; + struct drm_color_lut *lut; + int i; blob = drm_property_create_blob(&dev_priv->drm, sizeof(struct drm_color_lut) * LEGACY_LUT_LENGTH, @@ -1689,20 +1736,12 @@ i9xx_read_lut_8(const struct intel_crtc_state *crtc_state) if (IS_ERR(blob)) return NULL; - blob_data = blob->data; + lut = blob->data; for (i = 0; i < LEGACY_LUT_LENGTH; i++) { - if (HAS_GMCH(dev_priv)) - val = intel_de_read(dev_priv, PALETTE(pipe, i)); - else - val = intel_de_read(dev_priv, LGC_PALETTE(pipe, i)); - - blob_data[i].red = intel_color_lut_pack(REG_FIELD_GET( - LGC_PALETTE_RED_MASK, val), 8); - blob_data[i].green = intel_color_lut_pack(REG_FIELD_GET( - LGC_PALETTE_GREEN_MASK, val), 8); - blob_data[i].blue = intel_color_lut_pack(REG_FIELD_GET( - LGC_PALETTE_BLUE_MASK, val), 8); + u32 val = intel_de_read(dev_priv, PALETTE(pipe, i)); + + i9xx_lut_8_pack(&lut[i], val); } return blob; @@ -1710,22 +1749,21 @@ i9xx_read_lut_8(const struct intel_crtc_state *crtc_state) static void i9xx_read_luts(struct intel_crtc_state *crtc_state) { + struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); + if (!crtc_state->gamma_enable) return; - crtc_state->hw.gamma_lut = i9xx_read_lut_8(crtc_state); + crtc_state->hw.gamma_lut = i9xx_read_lut_8(crtc); } -static struct drm_property_blob * -i965_read_lut_10p6(const struct intel_crtc_state *crtc_state) +static struct drm_property_blob *i965_read_lut_10p6(struct intel_crtc *crtc) { - struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); - u32 lut_size = INTEL_INFO(dev_priv)->color.gamma_lut_size; + int i, lut_size = INTEL_INFO(dev_priv)->color.gamma_lut_size; enum pipe pipe = crtc->pipe; struct drm_property_blob *blob; - struct drm_color_lut *blob_data; - u32 i, val1, val2; + struct drm_color_lut *lut; blob = drm_property_create_blob(&dev_priv->drm, sizeof(struct drm_color_lut) * lut_size, @@ -1733,51 +1771,42 @@ i965_read_lut_10p6(const struct intel_crtc_state *crtc_state) if (IS_ERR(blob)) return NULL; - blob_data = blob->data; + lut = blob->data; for (i = 0; i < lut_size - 1; i++) { - val1 = intel_de_read(dev_priv, PALETTE(pipe, 2 * i + 0)); - val2 = intel_de_read(dev_priv, PALETTE(pipe, 2 * i + 1)); - - blob_data[i].red = REG_FIELD_GET(PALETTE_RED_MASK, val2) << 8 | - REG_FIELD_GET(PALETTE_RED_MASK, val1); - blob_data[i].green = REG_FIELD_GET(PALETTE_GREEN_MASK, val2) << 8 | - REG_FIELD_GET(PALETTE_GREEN_MASK, val1); - blob_data[i].blue = REG_FIELD_GET(PALETTE_BLUE_MASK, val2) << 8 | - REG_FIELD_GET(PALETTE_BLUE_MASK, val1); + u32 ldw = intel_de_read(dev_priv, PALETTE(pipe, 2 * i + 0)); + u32 udw = intel_de_read(dev_priv, PALETTE(pipe, 2 * i + 1)); + + i965_lut_10p6_pack(&lut[i], ldw, udw); } - blob_data[i].red = REG_FIELD_GET(PIPEGCMAX_RGB_MASK, - intel_de_read(dev_priv, PIPEGCMAX(pipe, 0))); - blob_data[i].green = REG_FIELD_GET(PIPEGCMAX_RGB_MASK, - intel_de_read(dev_priv, PIPEGCMAX(pipe, 1))); - blob_data[i].blue = REG_FIELD_GET(PIPEGCMAX_RGB_MASK, - intel_de_read(dev_priv, PIPEGCMAX(pipe, 2))); + lut[i].red = i965_lut_11p6_max_pack(intel_de_read(dev_priv, PIPEGCMAX(pipe, 0))); + lut[i].green = i965_lut_11p6_max_pack(intel_de_read(dev_priv, PIPEGCMAX(pipe, 1))); + lut[i].blue = i965_lut_11p6_max_pack(intel_de_read(dev_priv, PIPEGCMAX(pipe, 2))); return blob; } static void i965_read_luts(struct intel_crtc_state *crtc_state) { + struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); + if (!crtc_state->gamma_enable) return; if (crtc_state->gamma_mode == GAMMA_MODE_MODE_8BIT) - crtc_state->hw.gamma_lut = i9xx_read_lut_8(crtc_state); + crtc_state->hw.gamma_lut = i9xx_read_lut_8(crtc); else - crtc_state->hw.gamma_lut = i965_read_lut_10p6(crtc_state); + crtc_state->hw.gamma_lut = i965_read_lut_10p6(crtc); } -static struct drm_property_blob * -chv_read_cgm_lut(const struct intel_crtc_state *crtc_state) +static struct drm_property_blob *chv_read_cgm_gamma(struct intel_crtc *crtc) { - struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); - u32 lut_size = INTEL_INFO(dev_priv)->color.gamma_lut_size; + int i, lut_size = INTEL_INFO(dev_priv)->color.gamma_lut_size; enum pipe pipe = crtc->pipe; struct drm_property_blob *blob; - struct drm_color_lut *blob_data; - u32 i, val; + struct drm_color_lut *lut; blob = drm_property_create_blob(&dev_priv->drm, sizeof(struct drm_color_lut) * lut_size, @@ -1785,18 +1814,13 @@ chv_read_cgm_lut(const struct intel_crtc_state *crtc_state) if (IS_ERR(blob)) return NULL; - blob_data = blob->data; + lut = blob->data; for (i = 0; i < lut_size; i++) { - val = intel_de_read(dev_priv, CGM_PIPE_GAMMA(pipe, i, 0)); - blob_data[i].green = intel_color_lut_pack(REG_FIELD_GET( - CGM_PIPE_GAMMA_GREEN_MASK, val), 10); - blob_data[i].blue = intel_color_lut_pack(REG_FIELD_GET( - CGM_PIPE_GAMMA_BLUE_MASK, val), 10); - - val = intel_de_read(dev_priv, CGM_PIPE_GAMMA(pipe, i, 1)); - blob_data[i].red = intel_color_lut_pack(REG_FIELD_GET( - CGM_PIPE_GAMMA_RED_MASK, val), 10); + u32 ldw = intel_de_read(dev_priv, CGM_PIPE_GAMMA(pipe, i, 0)); + u32 udw = intel_de_read(dev_priv, CGM_PIPE_GAMMA(pipe, i, 1)); + + chv_cgm_gamma_pack(&lut[i], ldw, udw); } return blob; @@ -1804,22 +1828,46 @@ chv_read_cgm_lut(const struct intel_crtc_state *crtc_state) static void chv_read_luts(struct intel_crtc_state *crtc_state) { + struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); + if (crtc_state->cgm_mode & CGM_PIPE_MODE_GAMMA) - crtc_state->hw.gamma_lut = chv_read_cgm_lut(crtc_state); + crtc_state->hw.gamma_lut = chv_read_cgm_gamma(crtc); else i965_read_luts(crtc_state); } -static struct drm_property_blob * -ilk_read_lut_10(const struct intel_crtc_state *crtc_state) +static struct drm_property_blob *ilk_read_lut_8(struct intel_crtc *crtc) { - struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); - u32 lut_size = INTEL_INFO(dev_priv)->color.gamma_lut_size; enum pipe pipe = crtc->pipe; struct drm_property_blob *blob; - struct drm_color_lut *blob_data; - u32 i, val; + struct drm_color_lut *lut; + int i; + + blob = drm_property_create_blob(&dev_priv->drm, + sizeof(struct drm_color_lut) * LEGACY_LUT_LENGTH, + NULL); + if (IS_ERR(blob)) + return NULL; + + lut = blob->data; + + for (i = 0; i < LEGACY_LUT_LENGTH; i++) { + u32 val = intel_de_read(dev_priv, LGC_PALETTE(pipe, i)); + + i9xx_lut_8_pack(&lut[i], val); + } + + return blob; +} + +static struct drm_property_blob *ilk_read_lut_10(struct intel_crtc *crtc) +{ + struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); + int i, lut_size = INTEL_INFO(dev_priv)->color.gamma_lut_size; + enum pipe pipe = crtc->pipe; + struct drm_property_blob *blob; + struct drm_color_lut *lut; blob = drm_property_create_blob(&dev_priv->drm, sizeof(struct drm_color_lut) * lut_size, @@ -1827,17 +1875,12 @@ ilk_read_lut_10(const struct intel_crtc_state *crtc_state) if (IS_ERR(blob)) return NULL; - blob_data = blob->data; + lut = blob->data; for (i = 0; i < lut_size; i++) { - val = intel_de_read(dev_priv, PREC_PALETTE(pipe, i)); - - blob_data[i].red = intel_color_lut_pack(REG_FIELD_GET( - PREC_PALETTE_RED_MASK, val), 10); - blob_data[i].green = intel_color_lut_pack(REG_FIELD_GET( - PREC_PALETTE_GREEN_MASK, val), 10); - blob_data[i].blue = intel_color_lut_pack(REG_FIELD_GET( - PREC_PALETTE_BLUE_MASK, val), 10); + u32 val = intel_de_read(dev_priv, PREC_PALETTE(pipe, i)); + + ilk_lut_10_pack(&lut[i], val); } return blob; @@ -1845,6 +1888,8 @@ ilk_read_lut_10(const struct intel_crtc_state *crtc_state) static void ilk_read_luts(struct intel_crtc_state *crtc_state) { + struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); + if (!crtc_state->gamma_enable) return; @@ -1852,21 +1897,19 @@ static void ilk_read_luts(struct intel_crtc_state *crtc_state) return; if (crtc_state->gamma_mode == GAMMA_MODE_MODE_8BIT) - crtc_state->hw.gamma_lut = i9xx_read_lut_8(crtc_state); + crtc_state->hw.gamma_lut = ilk_read_lut_8(crtc); else - crtc_state->hw.gamma_lut = ilk_read_lut_10(crtc_state); + crtc_state->hw.gamma_lut = ilk_read_lut_10(crtc); } -static struct drm_property_blob * -glk_read_lut_10(const struct intel_crtc_state *crtc_state, u32 prec_index) +static struct drm_property_blob *glk_read_lut_10(struct intel_crtc *crtc, + u32 prec_index) { - struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); - int hw_lut_size = ivb_lut_10_size(prec_index); + int i, hw_lut_size = ivb_lut_10_size(prec_index); enum pipe pipe = crtc->pipe; struct drm_property_blob *blob; - struct drm_color_lut *blob_data; - u32 i, val; + struct drm_color_lut *lut; blob = drm_property_create_blob(&dev_priv->drm, sizeof(struct drm_color_lut) * hw_lut_size, @@ -1874,20 +1917,15 @@ glk_read_lut_10(const struct intel_crtc_state *crtc_state, u32 prec_index) if (IS_ERR(blob)) return NULL; - blob_data = blob->data; + lut = blob->data; intel_de_write(dev_priv, PREC_PAL_INDEX(pipe), prec_index | PAL_PREC_AUTO_INCREMENT); for (i = 0; i < hw_lut_size; i++) { - val = intel_de_read(dev_priv, PREC_PAL_DATA(pipe)); - - blob_data[i].red = intel_color_lut_pack(REG_FIELD_GET( - PREC_PAL_DATA_RED_MASK, val), 10); - blob_data[i].green = intel_color_lut_pack(REG_FIELD_GET( - PREC_PAL_DATA_GREEN_MASK, val), 10); - blob_data[i].blue = intel_color_lut_pack(REG_FIELD_GET( - PREC_PAL_DATA_BLUE_MASK, val), 10); + u32 val = intel_de_read(dev_priv, PREC_PAL_DATA(pipe)); + + ilk_lut_10_pack(&lut[i], val); } intel_de_write(dev_priv, PREC_PAL_INDEX(pipe), 0); @@ -1897,13 +1935,15 @@ glk_read_lut_10(const struct intel_crtc_state *crtc_state, u32 prec_index) static void glk_read_luts(struct intel_crtc_state *crtc_state) { + struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); + if (!crtc_state->gamma_enable) return; if (crtc_state->gamma_mode == GAMMA_MODE_MODE_8BIT) - crtc_state->hw.gamma_lut = i9xx_read_lut_8(crtc_state); + crtc_state->hw.gamma_lut = ilk_read_lut_8(crtc); else - crtc_state->hw.gamma_lut = glk_read_lut_10(crtc_state, PAL_PREC_INDEX_VALUE(0)); + crtc_state->hw.gamma_lut = glk_read_lut_10(crtc, PAL_PREC_INDEX_VALUE(0)); } void intel_color_init(struct intel_crtc *crtc) diff --git a/drivers/gpu/drm/i915/display/intel_crt.c b/drivers/gpu/drm/i915/display/intel_crt.c index f49c98f6cb7e..78f9b6cde810 100644 --- a/drivers/gpu/drm/i915/display/intel_crt.c +++ b/drivers/gpu/drm/i915/display/intel_crt.c @@ -32,7 +32,6 @@ #include <drm/drm_crtc.h> #include <drm/drm_edid.h> #include <drm/drm_probe_helper.h> -#include <drm/i915_drm.h> #include "i915_drv.h" #include "intel_connector.h" diff --git a/drivers/gpu/drm/i915/display/intel_csr.c b/drivers/gpu/drm/i915/display/intel_csr.c index 57320c12839f..3112572cfb7d 100644 --- a/drivers/gpu/drm/i915/display/intel_csr.c +++ b/drivers/gpu/drm/i915/display/intel_csr.c @@ -40,8 +40,8 @@ #define GEN12_CSR_MAX_FW_SIZE ICL_CSR_MAX_FW_SIZE -#define TGL_CSR_PATH "i915/tgl_dmc_ver2_04.bin" -#define TGL_CSR_VERSION_REQUIRED CSR_VERSION(2, 4) +#define TGL_CSR_PATH "i915/tgl_dmc_ver2_06.bin" +#define TGL_CSR_VERSION_REQUIRED CSR_VERSION(2, 6) #define TGL_CSR_MAX_FW_SIZE 0x6000 MODULE_FIRMWARE(TGL_CSR_PATH); diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c index 9f7d1d7189ae..73d0f4648c06 100644 --- a/drivers/gpu/drm/i915/display/intel_ddi.c +++ b/drivers/gpu/drm/i915/display/intel_ddi.c @@ -1325,164 +1325,6 @@ intel_ddi_get_crtc_encoder(struct intel_crtc *crtc) return ret; } -static int hsw_ddi_calc_wrpll_link(struct drm_i915_private *dev_priv, - i915_reg_t reg) -{ - int refclk; - int n, p, r; - u32 wrpll; - - wrpll = intel_de_read(dev_priv, reg); - switch (wrpll & WRPLL_REF_MASK) { - case WRPLL_REF_SPECIAL_HSW: - /* - * muxed-SSC for BDW. - * non-SSC for non-ULT HSW. Check FUSE_STRAP3 - * for the non-SSC reference frequency. - */ - if (IS_HASWELL(dev_priv) && !IS_HSW_ULT(dev_priv)) { - if (intel_de_read(dev_priv, FUSE_STRAP3) & HSW_REF_CLK_SELECT) - refclk = 24; - else - refclk = 135; - break; - } - /* fall through */ - case WRPLL_REF_PCH_SSC: - /* - * We could calculate spread here, but our checking - * code only cares about 5% accuracy, and spread is a max of - * 0.5% downspread. - */ - refclk = 135; - break; - case WRPLL_REF_LCPLL: - refclk = 2700; - break; - default: - MISSING_CASE(wrpll); - return 0; - } - - r = wrpll & WRPLL_DIVIDER_REF_MASK; - p = (wrpll & WRPLL_DIVIDER_POST_MASK) >> WRPLL_DIVIDER_POST_SHIFT; - n = (wrpll & WRPLL_DIVIDER_FB_MASK) >> WRPLL_DIVIDER_FB_SHIFT; - - /* Convert to KHz, p & r have a fixed point portion */ - return (refclk * n * 100) / (p * r); -} - -static int skl_calc_wrpll_link(const struct intel_dpll_hw_state *pll_state) -{ - u32 p0, p1, p2, dco_freq; - - p0 = pll_state->cfgcr2 & DPLL_CFGCR2_PDIV_MASK; - p2 = pll_state->cfgcr2 & DPLL_CFGCR2_KDIV_MASK; - - if (pll_state->cfgcr2 & DPLL_CFGCR2_QDIV_MODE(1)) - p1 = (pll_state->cfgcr2 & DPLL_CFGCR2_QDIV_RATIO_MASK) >> 8; - else - p1 = 1; - - - switch (p0) { - case DPLL_CFGCR2_PDIV_1: - p0 = 1; - break; - case DPLL_CFGCR2_PDIV_2: - p0 = 2; - break; - case DPLL_CFGCR2_PDIV_3: - p0 = 3; - break; - case DPLL_CFGCR2_PDIV_7: - p0 = 7; - break; - } - - switch (p2) { - case DPLL_CFGCR2_KDIV_5: - p2 = 5; - break; - case DPLL_CFGCR2_KDIV_2: - p2 = 2; - break; - case DPLL_CFGCR2_KDIV_3: - p2 = 3; - break; - case DPLL_CFGCR2_KDIV_1: - p2 = 1; - break; - } - - dco_freq = (pll_state->cfgcr1 & DPLL_CFGCR1_DCO_INTEGER_MASK) - * 24 * 1000; - - dco_freq += (((pll_state->cfgcr1 & DPLL_CFGCR1_DCO_FRACTION_MASK) >> 9) - * 24 * 1000) / 0x8000; - - if (WARN_ON(p0 == 0 || p1 == 0 || p2 == 0)) - return 0; - - return dco_freq / (p0 * p1 * p2 * 5); -} - -int cnl_calc_wrpll_link(struct drm_i915_private *dev_priv, - struct intel_dpll_hw_state *pll_state) -{ - u32 p0, p1, p2, dco_freq, ref_clock; - - p0 = pll_state->cfgcr1 & DPLL_CFGCR1_PDIV_MASK; - p2 = pll_state->cfgcr1 & DPLL_CFGCR1_KDIV_MASK; - - if (pll_state->cfgcr1 & DPLL_CFGCR1_QDIV_MODE(1)) - p1 = (pll_state->cfgcr1 & DPLL_CFGCR1_QDIV_RATIO_MASK) >> - DPLL_CFGCR1_QDIV_RATIO_SHIFT; - else - p1 = 1; - - - switch (p0) { - case DPLL_CFGCR1_PDIV_2: - p0 = 2; - break; - case DPLL_CFGCR1_PDIV_3: - p0 = 3; - break; - case DPLL_CFGCR1_PDIV_5: - p0 = 5; - break; - case DPLL_CFGCR1_PDIV_7: - p0 = 7; - break; - } - - switch (p2) { - case DPLL_CFGCR1_KDIV_1: - p2 = 1; - break; - case DPLL_CFGCR1_KDIV_2: - p2 = 2; - break; - case DPLL_CFGCR1_KDIV_3: - p2 = 3; - break; - } - - ref_clock = cnl_hdmi_pll_ref_clock(dev_priv); - - dco_freq = (pll_state->cfgcr0 & DPLL_CFGCR0_DCO_INTEGER_MASK) - * ref_clock; - - dco_freq += (((pll_state->cfgcr0 & DPLL_CFGCR0_DCO_FRACTION_MASK) >> - DPLL_CFGCR0_DCO_FRACTION_SHIFT) * ref_clock) / 0x8000; - - if (drm_WARN_ON(&dev_priv->drm, p0 == 0 || p1 == 0 || p2 == 0)) - return 0; - - return dco_freq / (p0 * p1 * p2 * 5); -} - static int icl_calc_tbt_pll_link(struct drm_i915_private *dev_priv, enum port port) { @@ -1505,77 +1347,6 @@ static int icl_calc_tbt_pll_link(struct drm_i915_private *dev_priv, } } -static int icl_calc_mg_pll_link(struct drm_i915_private *dev_priv, - const struct intel_dpll_hw_state *pll_state) -{ - u32 m1, m2_int, m2_frac, div1, div2, ref_clock; - u64 tmp; - - ref_clock = dev_priv->cdclk.hw.ref; - - if (INTEL_GEN(dev_priv) >= 12) { - m1 = pll_state->mg_pll_div0 & DKL_PLL_DIV0_FBPREDIV_MASK; - m1 = m1 >> DKL_PLL_DIV0_FBPREDIV_SHIFT; - m2_int = pll_state->mg_pll_div0 & DKL_PLL_DIV0_FBDIV_INT_MASK; - - if (pll_state->mg_pll_bias & DKL_PLL_BIAS_FRAC_EN_H) { - m2_frac = pll_state->mg_pll_bias & - DKL_PLL_BIAS_FBDIV_FRAC_MASK; - m2_frac = m2_frac >> DKL_PLL_BIAS_FBDIV_SHIFT; - } else { - m2_frac = 0; - } - } else { - m1 = pll_state->mg_pll_div1 & MG_PLL_DIV1_FBPREDIV_MASK; - m2_int = pll_state->mg_pll_div0 & MG_PLL_DIV0_FBDIV_INT_MASK; - - if (pll_state->mg_pll_div0 & MG_PLL_DIV0_FRACNEN_H) { - m2_frac = pll_state->mg_pll_div0 & - MG_PLL_DIV0_FBDIV_FRAC_MASK; - m2_frac = m2_frac >> MG_PLL_DIV0_FBDIV_FRAC_SHIFT; - } else { - m2_frac = 0; - } - } - - switch (pll_state->mg_clktop2_hsclkctl & - MG_CLKTOP2_HSCLKCTL_HSDIV_RATIO_MASK) { - case MG_CLKTOP2_HSCLKCTL_HSDIV_RATIO_2: - div1 = 2; - break; - case MG_CLKTOP2_HSCLKCTL_HSDIV_RATIO_3: - div1 = 3; - break; - case MG_CLKTOP2_HSCLKCTL_HSDIV_RATIO_5: - div1 = 5; - break; - case MG_CLKTOP2_HSCLKCTL_HSDIV_RATIO_7: - div1 = 7; - break; - default: - MISSING_CASE(pll_state->mg_clktop2_hsclkctl); - return 0; - } - - div2 = (pll_state->mg_clktop2_hsclkctl & - MG_CLKTOP2_HSCLKCTL_DSDIV_RATIO_MASK) >> - MG_CLKTOP2_HSCLKCTL_DSDIV_RATIO_SHIFT; - - /* div2 value of 0 is same as 1 means no div */ - if (div2 == 0) - div2 = 1; - - /* - * Adjust the original formula to delay the division by 2^22 in order to - * minimize possible rounding errors. - */ - tmp = (u64)m1 * m2_int * ref_clock + - (((u64)m1 * m2_frac * ref_clock) >> 22); - tmp = div_u64(tmp, 5 * div1 * div2); - - return tmp; -} - static void ddi_dotclock_get(struct intel_crtc_state *pipe_config) { int dotclock; @@ -1601,215 +1372,22 @@ static void ddi_dotclock_get(struct intel_crtc_state *pipe_config) pipe_config->hw.adjusted_mode.crtc_clock = dotclock; } -static void icl_ddi_clock_get(struct intel_encoder *encoder, - struct intel_crtc_state *pipe_config) -{ - struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); - struct intel_dpll_hw_state *pll_state = &pipe_config->dpll_hw_state; - enum port port = encoder->port; - enum phy phy = intel_port_to_phy(dev_priv, port); - int link_clock; - - if (intel_phy_is_combo(dev_priv, phy)) { - link_clock = cnl_calc_wrpll_link(dev_priv, pll_state); - } else { - enum intel_dpll_id pll_id = intel_get_shared_dpll_id(dev_priv, - pipe_config->shared_dpll); - - if (pll_id == DPLL_ID_ICL_TBTPLL) - link_clock = icl_calc_tbt_pll_link(dev_priv, port); - else - link_clock = icl_calc_mg_pll_link(dev_priv, pll_state); - } - - pipe_config->port_clock = link_clock; - - ddi_dotclock_get(pipe_config); -} - -static void cnl_ddi_clock_get(struct intel_encoder *encoder, - struct intel_crtc_state *pipe_config) -{ - struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); - struct intel_dpll_hw_state *pll_state = &pipe_config->dpll_hw_state; - int link_clock; - - if (pll_state->cfgcr0 & DPLL_CFGCR0_HDMI_MODE) { - link_clock = cnl_calc_wrpll_link(dev_priv, pll_state); - } else { - link_clock = pll_state->cfgcr0 & DPLL_CFGCR0_LINK_RATE_MASK; - - switch (link_clock) { - case DPLL_CFGCR0_LINK_RATE_810: - link_clock = 81000; - break; - case DPLL_CFGCR0_LINK_RATE_1080: - link_clock = 108000; - break; - case DPLL_CFGCR0_LINK_RATE_1350: - link_clock = 135000; - break; - case DPLL_CFGCR0_LINK_RATE_1620: - link_clock = 162000; - break; - case DPLL_CFGCR0_LINK_RATE_2160: - link_clock = 216000; - break; - case DPLL_CFGCR0_LINK_RATE_2700: - link_clock = 270000; - break; - case DPLL_CFGCR0_LINK_RATE_3240: - link_clock = 324000; - break; - case DPLL_CFGCR0_LINK_RATE_4050: - link_clock = 405000; - break; - default: - drm_WARN(&dev_priv->drm, 1, "Unsupported link rate\n"); - break; - } - link_clock *= 2; - } - - pipe_config->port_clock = link_clock; - - ddi_dotclock_get(pipe_config); -} - -static void skl_ddi_clock_get(struct intel_encoder *encoder, - struct intel_crtc_state *pipe_config) -{ - struct intel_dpll_hw_state *pll_state = &pipe_config->dpll_hw_state; - int link_clock; - - /* - * ctrl1 register is already shifted for each pll, just use 0 to get - * the internal shift for each field - */ - if (pll_state->ctrl1 & DPLL_CTRL1_HDMI_MODE(0)) { - link_clock = skl_calc_wrpll_link(pll_state); - } else { - link_clock = pll_state->ctrl1 & DPLL_CTRL1_LINK_RATE_MASK(0); - link_clock >>= DPLL_CTRL1_LINK_RATE_SHIFT(0); - - switch (link_clock) { - case DPLL_CTRL1_LINK_RATE_810: - link_clock = 81000; - break; - case DPLL_CTRL1_LINK_RATE_1080: - link_clock = 108000; - break; - case DPLL_CTRL1_LINK_RATE_1350: - link_clock = 135000; - break; - case DPLL_CTRL1_LINK_RATE_1620: - link_clock = 162000; - break; - case DPLL_CTRL1_LINK_RATE_2160: - link_clock = 216000; - break; - case DPLL_CTRL1_LINK_RATE_2700: - link_clock = 270000; - break; - default: - drm_WARN(encoder->base.dev, 1, - "Unsupported link rate\n"); - break; - } - link_clock *= 2; - } - - pipe_config->port_clock = link_clock; - - ddi_dotclock_get(pipe_config); -} - -static void hsw_ddi_clock_get(struct intel_encoder *encoder, - struct intel_crtc_state *pipe_config) -{ - struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); - int link_clock = 0; - u32 val, pll; - - val = hsw_pll_to_ddi_pll_sel(pipe_config->shared_dpll); - switch (val & PORT_CLK_SEL_MASK) { - case PORT_CLK_SEL_LCPLL_810: - link_clock = 81000; - break; - case PORT_CLK_SEL_LCPLL_1350: - link_clock = 135000; - break; - case PORT_CLK_SEL_LCPLL_2700: - link_clock = 270000; - break; - case PORT_CLK_SEL_WRPLL1: - link_clock = hsw_ddi_calc_wrpll_link(dev_priv, WRPLL_CTL(0)); - break; - case PORT_CLK_SEL_WRPLL2: - link_clock = hsw_ddi_calc_wrpll_link(dev_priv, WRPLL_CTL(1)); - break; - case PORT_CLK_SEL_SPLL: - pll = intel_de_read(dev_priv, SPLL_CTL) & SPLL_FREQ_MASK; - if (pll == SPLL_FREQ_810MHz) - link_clock = 81000; - else if (pll == SPLL_FREQ_1350MHz) - link_clock = 135000; - else if (pll == SPLL_FREQ_2700MHz) - link_clock = 270000; - else { - drm_WARN(&dev_priv->drm, 1, "bad spll freq\n"); - return; - } - break; - default: - drm_WARN(&dev_priv->drm, 1, "bad port clock sel\n"); - return; - } - - pipe_config->port_clock = link_clock * 2; - - ddi_dotclock_get(pipe_config); -} - -static int bxt_calc_pll_link(const struct intel_dpll_hw_state *pll_state) -{ - struct dpll clock; - - clock.m1 = 2; - clock.m2 = (pll_state->pll0 & PORT_PLL_M2_MASK) << 22; - if (pll_state->pll3 & PORT_PLL_M2_FRAC_ENABLE) - clock.m2 |= pll_state->pll2 & PORT_PLL_M2_FRAC_MASK; - clock.n = (pll_state->pll1 & PORT_PLL_N_MASK) >> PORT_PLL_N_SHIFT; - clock.p1 = (pll_state->ebb0 & PORT_PLL_P1_MASK) >> PORT_PLL_P1_SHIFT; - clock.p2 = (pll_state->ebb0 & PORT_PLL_P2_MASK) >> PORT_PLL_P2_SHIFT; - - return chv_calc_dpll_params(100000, &clock); -} - -static void bxt_ddi_clock_get(struct intel_encoder *encoder, - struct intel_crtc_state *pipe_config) -{ - pipe_config->port_clock = - bxt_calc_pll_link(&pipe_config->dpll_hw_state); - - ddi_dotclock_get(pipe_config); -} - static void intel_ddi_clock_get(struct intel_encoder *encoder, struct intel_crtc_state *pipe_config) { struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); + enum phy phy = intel_port_to_phy(dev_priv, encoder->port); - if (INTEL_GEN(dev_priv) >= 11) - icl_ddi_clock_get(encoder, pipe_config); - else if (IS_CANNONLAKE(dev_priv)) - cnl_ddi_clock_get(encoder, pipe_config); - else if (IS_GEN9_LP(dev_priv)) - bxt_ddi_clock_get(encoder, pipe_config); - else if (IS_GEN9_BC(dev_priv)) - skl_ddi_clock_get(encoder, pipe_config); - else if (INTEL_GEN(dev_priv) <= 8) - hsw_ddi_clock_get(encoder, pipe_config); + if (intel_phy_is_tc(dev_priv, phy) && + intel_get_shared_dpll_id(dev_priv, pipe_config->shared_dpll) == + DPLL_ID_ICL_TBTPLL) + pipe_config->port_clock = icl_calc_tbt_pll_link(dev_priv, + encoder->port); + else + pipe_config->port_clock = + intel_dpll_get_freq(dev_priv, pipe_config->shared_dpll); + + ddi_dotclock_get(pipe_config); } void intel_ddi_set_dp_msa(const struct intel_crtc_state *crtc_state, @@ -3049,7 +2627,7 @@ static void icl_map_plls_to_ports(struct intel_encoder *encoder, enum phy phy = intel_port_to_phy(dev_priv, encoder->port); u32 val; - mutex_lock(&dev_priv->dpll_lock); + mutex_lock(&dev_priv->dpll.lock); val = intel_de_read(dev_priv, ICL_DPCLKA_CFGCR0); drm_WARN_ON(&dev_priv->drm, @@ -3075,7 +2653,7 @@ static void icl_map_plls_to_ports(struct intel_encoder *encoder, val &= ~icl_dpclka_cfgcr0_clk_off(dev_priv, phy); intel_de_write(dev_priv, ICL_DPCLKA_CFGCR0, val); - mutex_unlock(&dev_priv->dpll_lock); + mutex_unlock(&dev_priv->dpll.lock); } static void icl_unmap_plls_to_ports(struct intel_encoder *encoder) @@ -3084,13 +2662,13 @@ static void icl_unmap_plls_to_ports(struct intel_encoder *encoder) enum phy phy = intel_port_to_phy(dev_priv, encoder->port); u32 val; - mutex_lock(&dev_priv->dpll_lock); + mutex_lock(&dev_priv->dpll.lock); val = intel_de_read(dev_priv, ICL_DPCLKA_CFGCR0); val |= icl_dpclka_cfgcr0_clk_off(dev_priv, phy); intel_de_write(dev_priv, ICL_DPCLKA_CFGCR0, val); - mutex_unlock(&dev_priv->dpll_lock); + mutex_unlock(&dev_priv->dpll.lock); } static void icl_sanitize_port_clk_off(struct drm_i915_private *dev_priv, @@ -3189,7 +2767,7 @@ static void intel_ddi_clk_select(struct intel_encoder *encoder, if (drm_WARN_ON(&dev_priv->drm, !pll)) return; - mutex_lock(&dev_priv->dpll_lock); + mutex_lock(&dev_priv->dpll.lock); if (INTEL_GEN(dev_priv) >= 11) { if (!intel_phy_is_combo(dev_priv, phy)) @@ -3233,7 +2811,7 @@ static void intel_ddi_clk_select(struct intel_encoder *encoder, hsw_pll_to_ddi_pll_sel(pll)); } - mutex_unlock(&dev_priv->dpll_lock); + mutex_unlock(&dev_priv->dpll.lock); } static void intel_ddi_clk_disable(struct intel_encoder *encoder) @@ -3987,8 +3565,9 @@ static void intel_enable_ddi_hdmi(struct intel_encoder *encoder, if (!intel_hdmi_handle_sink_scrambling(encoder, connector, crtc_state->hdmi_high_tmds_clock_ratio, crtc_state->hdmi_scrambling)) - DRM_ERROR("[CONNECTOR:%d:%s] Failed to configure sink scrambling/TMDS bit clock ratio\n", - connector->base.id, connector->name); + DRM_DEBUG_KMS("[CONNECTOR:%d:%s] Failed to configure sink " + "scrambling/TMDS bit clock ratio\n", + connector->base.id, connector->name); /* Display WA #1143: skl,kbl,cfl */ if (IS_GEN9_BC(dev_priv)) { diff --git a/drivers/gpu/drm/i915/display/intel_ddi.h b/drivers/gpu/drm/i915/display/intel_ddi.h index 167c6579d972..55fd72b901fe 100644 --- a/drivers/gpu/drm/i915/display/intel_ddi.h +++ b/drivers/gpu/drm/i915/display/intel_ddi.h @@ -6,8 +6,6 @@ #ifndef __INTEL_DDI_H__ #define __INTEL_DDI_H__ -#include <drm/i915_drm.h> - #include "intel_display.h" struct drm_connector_state; @@ -47,7 +45,5 @@ u8 intel_ddi_dp_pre_emphasis_max(struct intel_encoder *encoder, int intel_ddi_toggle_hdcp_signalling(struct intel_encoder *intel_encoder, bool enable); void icl_sanitize_encoder_pll_mapping(struct intel_encoder *encoder); -int cnl_calc_wrpll_link(struct drm_i915_private *dev_priv, - struct intel_dpll_hw_state *state); #endif /* __INTEL_DDI_H__ */ diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c index 3031e64ee518..346846609f45 100644 --- a/drivers/gpu/drm/i915/display/intel_display.c +++ b/drivers/gpu/drm/i915/display/intel_display.c @@ -41,7 +41,6 @@ #include <drm/drm_plane_helper.h> #include <drm/drm_probe_helper.h> #include <drm/drm_rect.h> -#include <drm/i915_drm.h> #include "display/intel_crt.h" #include "display/intel_ddi.h" @@ -2720,9 +2719,10 @@ u32 intel_plane_fb_max_stride(struct drm_i915_private *dev_priv, /* * We assume the primary plane for pipe A has - * the highest stride limits of them all. + * the highest stride limits of them all, + * if in case pipe A is disabled, use the first pipe from pipe_mask. */ - crtc = intel_get_crtc_for_pipe(dev_priv, PIPE_A); + crtc = intel_get_first_crtc(dev_priv); if (!crtc) return 0; @@ -9542,7 +9542,7 @@ static void ilk_init_pch_refclk(struct drm_i915_private *dev_priv) } /* Check if any DPLLs are using the SSC source */ - for (i = 0; i < dev_priv->num_shared_dpll; i++) { + for (i = 0; i < dev_priv->dpll.num_shared_dpll; i++) { u32 temp = intel_de_read(dev_priv, PCH_DPLL(i)); if (!(temp & DPLL_VCO_ENABLE)) @@ -10129,6 +10129,9 @@ static void bdw_set_pipemisc(const struct intel_crtc_state *crtc_state) BIT(PLANE_CURSOR))) == 0) val |= PIPEMISC_HDR_MODE_PRECISION; + if (INTEL_GEN(dev_priv) >= 12) + val |= PIPEMISC_PIXEL_ROUNDING_TRUNC; + intel_de_write(dev_priv, PIPEMISC(crtc->pipe), val); } @@ -14299,11 +14302,11 @@ verify_single_dpll_state(struct drm_i915_private *dev_priv, if (new_crtc_state->hw.active) I915_STATE_WARN(!(pll->active_mask & crtc_mask), "pll active mismatch (expected pipe %c in active mask 0x%02x)\n", - pipe_name(drm_crtc_index(&crtc->base)), pll->active_mask); + pipe_name(crtc->pipe), pll->active_mask); else I915_STATE_WARN(pll->active_mask & crtc_mask, "pll active mismatch (didn't expect pipe %c in active mask 0x%02x)\n", - pipe_name(drm_crtc_index(&crtc->base)), pll->active_mask); + pipe_name(crtc->pipe), pll->active_mask); I915_STATE_WARN(!(pll->state.crtc_mask & crtc_mask), "pll enabled crtcs mismatch (expected 0x%x in 0x%02x)\n", @@ -14332,10 +14335,10 @@ verify_shared_dpll_state(struct intel_crtc *crtc, I915_STATE_WARN(pll->active_mask & crtc_mask, "pll active mismatch (didn't expect pipe %c in active mask)\n", - pipe_name(drm_crtc_index(&crtc->base))); + pipe_name(crtc->pipe)); I915_STATE_WARN(pll->state.crtc_mask & crtc_mask, "pll enabled crtcs mismatch (found %x in enabled mask)\n", - pipe_name(drm_crtc_index(&crtc->base))); + pipe_name(crtc->pipe)); } } @@ -14359,8 +14362,10 @@ verify_disabled_dpll_state(struct drm_i915_private *dev_priv) { int i; - for (i = 0; i < dev_priv->num_shared_dpll; i++) - verify_single_dpll_state(dev_priv, &dev_priv->shared_dplls[i], NULL, NULL); + for (i = 0; i < dev_priv->dpll.num_shared_dpll; i++) + verify_single_dpll_state(dev_priv, + &dev_priv->dpll.shared_dplls[i], + NULL, NULL); } static void @@ -14743,8 +14748,8 @@ static int intel_atomic_check(struct drm_device *dev, /* Catch I915_MODE_FLAG_INHERITED */ for_each_oldnew_intel_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) { - if (new_crtc_state->hw.mode.private_flags != - old_crtc_state->hw.mode.private_flags) + if (new_crtc_state->uapi.mode.private_flags != + old_crtc_state->uapi.mode.private_flags) new_crtc_state->uapi.mode_changed = true; } @@ -15318,7 +15323,6 @@ static void skl_commit_modeset_enables(struct intel_atomic_state *state) struct intel_crtc *crtc; struct intel_crtc_state *old_crtc_state, *new_crtc_state; struct skl_ddb_entry entries[I915_MAX_PIPES] = {}; - const u8 num_pipes = INTEL_NUM_PIPES(dev_priv); u8 update_pipes = 0, modeset_pipes = 0; int i; @@ -15355,7 +15359,7 @@ static void skl_commit_modeset_enables(struct intel_atomic_state *state) continue; if (skl_ddb_allocation_overlaps(&new_crtc_state->wm.skl.ddb, - entries, num_pipes, pipe)) + entries, I915_MAX_PIPES, pipe)) continue; entries[pipe] = new_crtc_state->wm.skl.ddb; @@ -15393,7 +15397,7 @@ static void skl_commit_modeset_enables(struct intel_atomic_state *state) continue; drm_WARN_ON(&dev_priv->drm, skl_ddb_allocation_overlaps(&new_crtc_state->wm.skl.ddb, - entries, num_pipes, pipe)); + entries, I915_MAX_PIPES, pipe)); entries[pipe] = new_crtc_state->wm.skl.ddb; modeset_pipes &= ~BIT(pipe); @@ -15428,7 +15432,7 @@ static void skl_commit_modeset_enables(struct intel_atomic_state *state) continue; drm_WARN_ON(&dev_priv->drm, skl_ddb_allocation_overlaps(&new_crtc_state->wm.skl.ddb, - entries, num_pipes, pipe)); + entries, I915_MAX_PIPES, pipe)); entries[pipe] = new_crtc_state->wm.skl.ddb; modeset_pipes &= ~BIT(pipe); @@ -16320,7 +16324,6 @@ intel_primary_plane_create(struct drm_i915_private *dev_priv, enum pipe pipe) struct intel_plane *plane; const struct drm_plane_funcs *plane_funcs; unsigned int supported_rotations; - unsigned int possible_crtcs; const u32 *formats; int num_formats; int ret, zpos; @@ -16401,18 +16404,16 @@ intel_primary_plane_create(struct drm_i915_private *dev_priv, enum pipe pipe) plane->get_hw_state = i9xx_plane_get_hw_state; plane->check_plane = i9xx_plane_check; - possible_crtcs = BIT(pipe); - if (INTEL_GEN(dev_priv) >= 5 || IS_G4X(dev_priv)) ret = drm_universal_plane_init(&dev_priv->drm, &plane->base, - possible_crtcs, plane_funcs, + 0, plane_funcs, formats, num_formats, i9xx_format_modifiers, DRM_PLANE_TYPE_PRIMARY, "primary %c", pipe_name(pipe)); else ret = drm_universal_plane_init(&dev_priv->drm, &plane->base, - possible_crtcs, plane_funcs, + 0, plane_funcs, formats, num_formats, i9xx_format_modifiers, DRM_PLANE_TYPE_PRIMARY, @@ -16454,7 +16455,6 @@ static struct intel_plane * intel_cursor_plane_create(struct drm_i915_private *dev_priv, enum pipe pipe) { - unsigned int possible_crtcs; struct intel_plane *cursor; int ret, zpos; @@ -16487,10 +16487,8 @@ intel_cursor_plane_create(struct drm_i915_private *dev_priv, if (IS_I845G(dev_priv) || IS_I865G(dev_priv) || HAS_CUR_FBC(dev_priv)) cursor->cursor.size = ~0; - possible_crtcs = BIT(pipe); - ret = drm_universal_plane_init(&dev_priv->drm, &cursor->base, - possible_crtcs, &intel_cursor_plane_funcs, + 0, &intel_cursor_plane_funcs, intel_cursor_formats, ARRAY_SIZE(intel_cursor_formats), cursor_format_modifiers, @@ -16619,6 +16617,18 @@ static void intel_crtc_free(struct intel_crtc *crtc) kfree(crtc); } +static void intel_plane_possible_crtcs_init(struct drm_i915_private *dev_priv) +{ + struct intel_plane *plane; + + for_each_intel_plane(&dev_priv->drm, plane) { + struct intel_crtc *crtc = intel_get_crtc_for_pipe(dev_priv, + plane->pipe); + + plane->base.possible_crtcs = drm_crtc_mask(&crtc->base); + } +} + static int intel_crtc_init(struct drm_i915_private *dev_priv, enum pipe pipe) { struct intel_plane *primary, *cursor; @@ -16697,6 +16707,8 @@ static int intel_crtc_init(struct drm_i915_private *dev_priv, enum pipe pipe) intel_color_init(crtc); + intel_crtc_crc_init(crtc); + drm_WARN_ON(&dev_priv->drm, drm_crtc_index(&crtc->base) != crtc->pipe); return 0; @@ -17785,11 +17797,9 @@ static void plane_config_fini(struct intel_initial_plane_config *plane_config) i915_vma_put(plane_config->vma); } -int intel_modeset_init(struct drm_i915_private *i915) +/* part #1: call before irq install */ +int intel_modeset_init_noirq(struct drm_i915_private *i915) { - struct drm_device *dev = &i915->drm; - enum pipe pipe; - struct intel_crtc *crtc; int ret; i915->modeset_wq = alloc_ordered_workqueue("i915_modeset", 0); @@ -17814,6 +17824,17 @@ int intel_modeset_init(struct drm_i915_private *i915) intel_fbc_init(i915); + return 0; +} + +/* part #2: call after irq install */ +int intel_modeset_init(struct drm_i915_private *i915) +{ + struct drm_device *dev = &i915->drm; + enum pipe pipe; + struct intel_crtc *crtc; + int ret; + intel_init_pm(i915); intel_panel_sanitize_ssc(i915); @@ -17834,6 +17855,7 @@ int intel_modeset_init(struct drm_i915_private *i915) } } + intel_plane_possible_crtcs_init(i915); intel_shared_dpll_init(dev); intel_update_fdi_pll_freq(i915); @@ -18311,7 +18333,6 @@ static void intel_modeset_readout_hw_state(struct drm_device *dev) struct intel_connector *connector; struct drm_connector_list_iter conn_iter; u8 active_pipes = 0; - int i; for_each_intel_crtc(dev, crtc) { struct intel_crtc_state *crtc_state = @@ -18340,33 +18361,7 @@ static void intel_modeset_readout_hw_state(struct drm_device *dev) readout_plane_state(dev_priv); - for (i = 0; i < dev_priv->num_shared_dpll; i++) { - struct intel_shared_dpll *pll = &dev_priv->shared_dplls[i]; - - pll->on = pll->info->funcs->get_hw_state(dev_priv, pll, - &pll->state.hw_state); - - if (IS_ELKHARTLAKE(dev_priv) && pll->on && - pll->info->id == DPLL_ID_EHL_DPLL4) { - pll->wakeref = intel_display_power_get(dev_priv, - POWER_DOMAIN_DPLL_DC_OFF); - } - - pll->state.crtc_mask = 0; - for_each_intel_crtc(dev, crtc) { - struct intel_crtc_state *crtc_state = - to_intel_crtc_state(crtc->base.state); - - if (crtc_state->hw.active && - crtc_state->shared_dpll == pll) - pll->state.crtc_mask |= 1 << crtc->pipe; - } - pll->active_mask = pll->state.crtc_mask; - - drm_dbg_kms(&dev_priv->drm, - "%s hw state readout: crtc_mask 0x%08x, on %i\n", - pll->info->name, pll->state.crtc_mask, pll->on); - } + intel_dpll_readout_hw_state(dev_priv); for_each_intel_encoder(dev, encoder) { pipe = 0; @@ -18623,7 +18618,6 @@ intel_modeset_setup_hw_state(struct drm_device *dev, struct intel_encoder *encoder; struct intel_crtc *crtc; intel_wakeref_t wakeref; - int i; wakeref = intel_display_power_get(dev_priv, POWER_DOMAIN_INIT); @@ -18676,19 +18670,7 @@ intel_modeset_setup_hw_state(struct drm_device *dev, intel_modeset_update_connector_atomic_state(dev); - for (i = 0; i < dev_priv->num_shared_dpll; i++) { - struct intel_shared_dpll *pll = &dev_priv->shared_dplls[i]; - - if (!pll->on || pll->active_mask) - continue; - - drm_dbg_kms(&dev_priv->drm, - "%s enabled but not in use, disabling\n", - pll->info->name); - - pll->info->funcs->disable(dev_priv, pll); - pll->on = false; - } + intel_dpll_sanitize_state(dev_priv); if (IS_G4X(dev_priv)) { g4x_wm_get_hw_state(dev_priv); @@ -18820,6 +18802,15 @@ void intel_modeset_driver_remove_noirq(struct drm_i915_private *i915) #if IS_ENABLED(CONFIG_DRM_I915_CAPTURE_ERROR) +static bool +has_transcoder(struct drm_i915_private *dev_priv, enum transcoder cpu_transcoder) +{ + if (cpu_transcoder == TRANSCODER_EDP) + return HAS_TRANSCODER_EDP(dev_priv); + else + return INTEL_INFO(dev_priv)->pipe_mask & BIT(cpu_transcoder); +} + struct intel_display_error_state { u32 power_well_driver; @@ -18928,7 +18919,7 @@ intel_display_capture_error_state(struct drm_i915_private *dev_priv) for (i = 0; i < ARRAY_SIZE(error->transcoder); i++) { enum transcoder cpu_transcoder = transcoders[i]; - if (!INTEL_INFO(dev_priv)->trans_offsets[cpu_transcoder]) + if (!has_transcoder(dev_priv, cpu_transcoder)) continue; error->transcoder[i].available = true; diff --git a/drivers/gpu/drm/i915/display/intel_display.h b/drivers/gpu/drm/i915/display/intel_display.h index f92efbbec838..adb1225a3480 100644 --- a/drivers/gpu/drm/i915/display/intel_display.h +++ b/drivers/gpu/drm/i915/display/intel_display.h @@ -26,7 +26,6 @@ #define _INTEL_DISPLAY_H_ #include <drm/drm_util.h> -#include <drm/i915_drm.h> enum link_m_n_set; struct dpll; @@ -40,6 +39,7 @@ struct drm_framebuffer; struct drm_i915_error_state_buf; struct drm_i915_gem_object; struct drm_i915_private; +struct drm_mode_fb_cmd2; struct drm_modeset_acquire_ctx; struct drm_plane; struct drm_plane_state; @@ -47,6 +47,7 @@ struct i915_ggtt_view; struct intel_atomic_state; struct intel_crtc; struct intel_crtc_state; +struct intel_crtc_state; struct intel_digital_port; struct intel_dp; struct intel_encoder; @@ -55,7 +56,6 @@ struct intel_plane; struct intel_plane_state; struct intel_remapped_info; struct intel_rotation_info; -struct intel_crtc_state; enum i915_gpio { GPIOA, @@ -313,10 +313,11 @@ enum phy_fia { }; #define for_each_pipe(__dev_priv, __p) \ - for ((__p) = 0; (__p) < INTEL_NUM_PIPES(__dev_priv); (__p)++) + for ((__p) = 0; (__p) < I915_MAX_PIPES; (__p)++) \ + for_each_if(INTEL_INFO(__dev_priv)->pipe_mask & BIT(__p)) #define for_each_pipe_masked(__dev_priv, __p, __mask) \ - for ((__p) = 0; (__p) < INTEL_NUM_PIPES(__dev_priv); (__p)++) \ + for_each_pipe(__dev_priv, __p) \ for_each_if((__mask) & BIT(__p)) #define for_each_cpu_transcoder_masked(__dev_priv, __t, __mask) \ @@ -614,6 +615,7 @@ intel_format_info_is_yuv_semiplanar(const struct drm_format_info *info, /* modesetting */ void intel_modeset_init_hw(struct drm_i915_private *i915); +int intel_modeset_init_noirq(struct drm_i915_private *i915); int intel_modeset_init(struct drm_i915_private *i915); void intel_modeset_driver_remove(struct drm_i915_private *i915); void intel_modeset_driver_remove_noirq(struct drm_i915_private *i915); diff --git a/drivers/gpu/drm/i915/display/intel_display_debugfs.c b/drivers/gpu/drm/i915/display/intel_display_debugfs.c index 46954cc7b6c0..1e6eb7f2f72d 100644 --- a/drivers/gpu/drm/i915/display/intel_display_debugfs.c +++ b/drivers/gpu/drm/i915/display/intel_display_debugfs.c @@ -920,8 +920,13 @@ static int i915_shared_dplls_info(struct seq_file *m, void *unused) int i; drm_modeset_lock_all(dev); - for (i = 0; i < dev_priv->num_shared_dpll; i++) { - struct intel_shared_dpll *pll = &dev_priv->shared_dplls[i]; + + seq_printf(m, "PLL refclks: non-SSC: %d kHz, SSC: %d kHz\n", + dev_priv->dpll.ref_clks.nssc, + dev_priv->dpll.ref_clks.ssc); + + for (i = 0; i < dev_priv->dpll.num_shared_dpll; i++) { + struct intel_shared_dpll *pll = &dev_priv->dpll.shared_dplls[i]; seq_printf(m, "DPLL%i: %s, id: %i\n", i, pll->info->name, pll->info->id); diff --git a/drivers/gpu/drm/i915/display/intel_display_power.c b/drivers/gpu/drm/i915/display/intel_display_power.c index 6e25a1317161..246e406bb385 100644 --- a/drivers/gpu/drm/i915/display/intel_display_power.c +++ b/drivers/gpu/drm/i915/display/intel_display_power.c @@ -939,11 +939,17 @@ unlock: static void assert_can_enable_dc5(struct drm_i915_private *dev_priv) { - bool pg2_enabled = intel_display_power_well_is_enabled(dev_priv, - SKL_DISP_PW_2); + enum i915_power_well_id high_pg; - drm_WARN_ONCE(&dev_priv->drm, pg2_enabled, - "PG2 not disabled to enable DC5.\n"); + /* Power wells at this level and above must be disabled for DC5 entry */ + if (INTEL_GEN(dev_priv) >= 12) + high_pg = TGL_DISP_PW_3; + else + high_pg = SKL_DISP_PW_2; + + drm_WARN_ONCE(&dev_priv->drm, + intel_display_power_well_is_enabled(dev_priv, high_pg), + "Power wells above platform's DC5 limit still enabled.\n"); drm_WARN_ONCE(&dev_priv->drm, (intel_de_read(dev_priv, DC_STATE_EN) & @@ -2740,7 +2746,7 @@ void intel_display_power_put(struct drm_i915_private *dev_priv, BIT_ULL(POWER_DOMAIN_INIT)) #define TGL_DISPLAY_DC_OFF_POWER_DOMAINS ( \ - TGL_PW_2_POWER_DOMAINS | \ + TGL_PW_3_POWER_DOMAINS | \ BIT_ULL(POWER_DOMAIN_MODESET) | \ BIT_ULL(POWER_DOMAIN_AUX_A) | \ BIT_ULL(POWER_DOMAIN_AUX_B) | \ @@ -3936,7 +3942,7 @@ static const struct i915_power_well_desc tgl_power_wells[] = { .name = "power well 3", .domains = TGL_PW_3_POWER_DOMAINS, .ops = &hsw_power_well_ops, - .id = DISP_PW_ID_NONE, + .id = TGL_DISP_PW_3, { .hsw.regs = &hsw_power_well_regs, .hsw.idx = ICL_PW_CTL_IDX_PW_3, diff --git a/drivers/gpu/drm/i915/display/intel_display_power.h b/drivers/gpu/drm/i915/display/intel_display_power.h index 601e000ffd0d..da64a5edae7a 100644 --- a/drivers/gpu/drm/i915/display/intel_display_power.h +++ b/drivers/gpu/drm/i915/display/intel_display_power.h @@ -100,6 +100,7 @@ enum i915_power_well_id { SKL_DISP_PW_MISC_IO, SKL_DISP_PW_1, SKL_DISP_PW_2, + TGL_DISP_PW_3, SKL_DISP_DC_OFF, }; diff --git a/drivers/gpu/drm/i915/display/intel_display_types.h b/drivers/gpu/drm/i915/display/intel_display_types.h index 0d8a64305464..5e00e611f077 100644 --- a/drivers/gpu/drm/i915/display/intel_display_types.h +++ b/drivers/gpu/drm/i915/display/intel_display_types.h @@ -39,7 +39,6 @@ #include <drm/drm_probe_helper.h> #include <drm/drm_rect.h> #include <drm/drm_vblank.h> -#include <drm/i915_drm.h> #include <drm/i915_mei_hdcp_interface.h> #include <media/cec-notifier.h> @@ -642,6 +641,14 @@ struct intel_crtc_scaler_state { /* Flag to use the scanline counter instead of the pixel counter */ #define I915_MODE_FLAG_USE_SCANLINE_COUNTER (1<<2) +struct intel_wm_level { + bool enable; + u32 pri_val; + u32 spr_val; + u32 cur_val; + u32 fbc_val; +}; + struct intel_pipe_wm { struct intel_wm_level wm[5]; bool fbc_wm_enabled; @@ -650,6 +657,14 @@ struct intel_pipe_wm { bool sprites_scaled; }; +struct skl_wm_level { + u16 min_ddb_alloc; + u16 plane_res_b; + u8 plane_res_l; + bool plane_en; + bool ignore_lines; +}; + struct skl_plane_wm { struct skl_wm_level wm[8]; struct skl_wm_level uv_wm[8]; @@ -1046,6 +1061,32 @@ struct intel_crtc_state { enum transcoder mst_master_transcoder; }; +enum intel_pipe_crc_source { + INTEL_PIPE_CRC_SOURCE_NONE, + INTEL_PIPE_CRC_SOURCE_PLANE1, + INTEL_PIPE_CRC_SOURCE_PLANE2, + INTEL_PIPE_CRC_SOURCE_PLANE3, + INTEL_PIPE_CRC_SOURCE_PLANE4, + INTEL_PIPE_CRC_SOURCE_PLANE5, + INTEL_PIPE_CRC_SOURCE_PLANE6, + INTEL_PIPE_CRC_SOURCE_PLANE7, + INTEL_PIPE_CRC_SOURCE_PIPE, + /* TV/DP on pre-gen5/vlv can't use the pipe source. */ + INTEL_PIPE_CRC_SOURCE_TV, + INTEL_PIPE_CRC_SOURCE_DP_B, + INTEL_PIPE_CRC_SOURCE_DP_C, + INTEL_PIPE_CRC_SOURCE_DP_D, + INTEL_PIPE_CRC_SOURCE_AUTO, + INTEL_PIPE_CRC_SOURCE_MAX, +}; + +#define INTEL_PIPE_CRC_ENTRIES_NR 128 +struct intel_pipe_crc { + spinlock_t lock; + int skipped; + enum intel_pipe_crc_source source; +}; + struct intel_crtc { struct drm_crtc base; enum pipe pipe; @@ -1089,6 +1130,10 @@ struct intel_crtc { /* per pipe DSB related info */ struct intel_dsb dsb; + +#ifdef CONFIG_DEBUG_FS + struct intel_pipe_crc pipe_crc; +#endif }; struct intel_plane { @@ -1235,6 +1280,7 @@ struct intel_dp { int max_link_rate; /* sink or branch descriptor */ struct drm_dp_desc desc; + u32 edid_quirks; struct drm_dp_aux aux; u32 aux_busy_last_status; u8 train_set[4]; @@ -1407,8 +1453,17 @@ vlv_pipe_to_channel(enum pipe pipe) } static inline struct intel_crtc * +intel_get_first_crtc(struct drm_i915_private *dev_priv) +{ + return to_intel_crtc(drm_crtc_from_index(&dev_priv->drm, 0)); +} + +static inline struct intel_crtc * intel_get_crtc_for_pipe(struct drm_i915_private *dev_priv, enum pipe pipe) { + /* pipe_to_crtc_mapping may have hole on any of 3 display pipe system */ + drm_WARN_ON(&dev_priv->drm, + !(INTEL_INFO(dev_priv)->pipe_mask & BIT(pipe))); return dev_priv->pipe_to_crtc_mapping[pipe]; } @@ -1598,11 +1653,15 @@ intel_crtc_has_dp_encoder(const struct intel_crtc_state *crtc_state) (1 << INTEL_OUTPUT_DP_MST) | (1 << INTEL_OUTPUT_EDP)); } + static inline void intel_wait_for_vblank(struct drm_i915_private *dev_priv, enum pipe pipe) { - drm_wait_one_vblank(&dev_priv->drm, pipe); + struct intel_crtc *crtc = intel_get_crtc_for_pipe(dev_priv, pipe); + + drm_crtc_wait_one_vblank(&crtc->base); } + static inline void intel_wait_for_vblank_if_active(struct drm_i915_private *dev_priv, enum pipe pipe) { diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c index 0a5a9197f8f5..0a417cd2af2b 100644 --- a/drivers/gpu/drm/i915/display/intel_dp.c +++ b/drivers/gpu/drm/i915/display/intel_dp.c @@ -40,7 +40,6 @@ #include <drm/drm_edid.h> #include <drm/drm_hdcp.h> #include <drm/drm_probe_helper.h> -#include <drm/i915_drm.h> #include "i915_debugfs.h" #include "i915_drv.h" @@ -2399,7 +2398,7 @@ intel_dp_compute_config(struct intel_encoder *encoder, struct intel_connector *intel_connector = intel_dp->attached_connector; struct intel_digital_connector_state *intel_conn_state = to_intel_digital_connector_state(conn_state); - bool constant_n = drm_dp_has_quirk(&intel_dp->desc, + bool constant_n = drm_dp_has_quirk(&intel_dp->desc, 0, DP_DPCD_QUIRK_CONSTANT_N); int ret = 0, output_bpp; @@ -4515,7 +4514,8 @@ intel_dp_get_dpcd(struct intel_dp *intel_dp) * it don't care about read it here and in intel_edp_init_dpcd(). */ if (!intel_dp_is_edp(intel_dp) && - !drm_dp_has_quirk(&intel_dp->desc, DP_DPCD_QUIRK_NO_SINK_COUNT)) { + !drm_dp_has_quirk(&intel_dp->desc, 0, + DP_DPCD_QUIRK_NO_SINK_COUNT)) { u8 count; ssize_t r; @@ -5682,6 +5682,7 @@ intel_dp_set_edid(struct intel_dp *intel_dp) intel_dp->has_audio = drm_detect_monitor_audio(edid); drm_dp_cec_set_edid(&intel_dp->aux, edid); + intel_dp->edid_quirks = drm_dp_get_edid_quirks(edid); } static void @@ -5694,6 +5695,7 @@ intel_dp_unset_edid(struct intel_dp *intel_dp) intel_connector->detect_edid = NULL; intel_dp->has_audio = false; + intel_dp->edid_quirks = 0; } static int @@ -6449,6 +6451,7 @@ static int intel_dp_hdcp2_config_stream_type(struct intel_digital_port *intel_dig_port, bool is_repeater, u8 content_type) { + int ret; struct hdcp2_dp_errata_stream_type stream_type_msg; if (is_repeater) @@ -6464,8 +6467,11 @@ int intel_dp_hdcp2_config_stream_type(struct intel_digital_port *intel_dig_port, stream_type_msg.msg_id = HDCP_2_2_ERRATA_DP_STREAM_TYPE; stream_type_msg.stream_type = content_type; - return intel_dp_hdcp2_write_msg(intel_dig_port, &stream_type_msg, + ret = intel_dp_hdcp2_write_msg(intel_dig_port, &stream_type_msg, sizeof(stream_type_msg)); + + return ret < 0 ? ret : 0; + } static @@ -7562,8 +7568,8 @@ static bool intel_edp_init_connector(struct intel_dp *intel_dp, edid = drm_get_edid(connector, &intel_dp->aux.ddc); if (edid) { if (drm_add_edid_modes(connector, edid)) { - drm_connector_update_edid_property(connector, - edid); + drm_connector_update_edid_property(connector, edid); + intel_dp->edid_quirks = drm_dp_get_edid_quirks(edid); } else { kfree(edid); edid = ERR_PTR(-EINVAL); @@ -7609,9 +7615,8 @@ static bool intel_edp_init_connector(struct intel_dp *intel_dp, intel_panel_setup_backlight(connector, pipe); if (fixed_mode) { - /* We do not know the orientation, but their might be a quirk */ drm_connector_set_panel_orientation_with_quirk(connector, - DRM_MODE_PANEL_ORIENTATION_UNKNOWN, + dev_priv->vbt.orientation, fixed_mode->hdisplay, fixed_mode->vdisplay); } diff --git a/drivers/gpu/drm/i915/display/intel_dp.h b/drivers/gpu/drm/i915/display/intel_dp.h index 3da166054788..0c7be8ed1423 100644 --- a/drivers/gpu/drm/i915/display/intel_dp.h +++ b/drivers/gpu/drm/i915/display/intel_dp.h @@ -8,8 +8,6 @@ #include <linux/types.h> -#include <drm/i915_drm.h> - #include "i915_reg.h" enum pipe; diff --git a/drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c b/drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c index 48276237b362..3e706bb850a8 100644 --- a/drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c +++ b/drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c @@ -328,15 +328,31 @@ intel_dp_aux_display_control_capable(struct intel_connector *connector) int intel_dp_aux_init_backlight_funcs(struct intel_connector *intel_connector) { struct intel_panel *panel = &intel_connector->panel; - struct drm_i915_private *dev_priv = to_i915(intel_connector->base.dev); + struct intel_dp *intel_dp = enc_to_intel_dp(intel_connector->encoder); + struct drm_device *dev = intel_connector->base.dev; + struct drm_i915_private *dev_priv = to_i915(dev); if (i915_modparams.enable_dpcd_backlight == 0 || - (i915_modparams.enable_dpcd_backlight == -1 && - dev_priv->vbt.backlight.type != INTEL_BACKLIGHT_VESA_EDP_AUX_INTERFACE)) + !intel_dp_aux_display_control_capable(intel_connector)) return -ENODEV; - if (!intel_dp_aux_display_control_capable(intel_connector)) + /* + * There are a lot of machines that don't advertise the backlight + * control interface to use properly in their VBIOS, :\ + */ + if (dev_priv->vbt.backlight.type != + INTEL_BACKLIGHT_VESA_EDP_AUX_INTERFACE && + !drm_dp_has_quirk(&intel_dp->desc, intel_dp->edid_quirks, + DP_QUIRK_FORCE_DPCD_BACKLIGHT)) { + DRM_DEV_INFO(dev->dev, + "Panel advertises DPCD backlight support, but " + "VBT disagrees. If your backlight controls " + "don't work try booting with " + "i915.enable_dpcd_backlight=1. If your machine " + "needs this, please file a _new_ bug report on " + "drm/i915, see " FDO_BUG_URL " for details.\n"); return -ENODEV; + } panel->backlight.setup = intel_dp_aux_setup_backlight; panel->backlight.enable = intel_dp_aux_enable_backlight; diff --git a/drivers/gpu/drm/i915/display/intel_dp_mst.c b/drivers/gpu/drm/i915/display/intel_dp_mst.c index 306c1549a680..44f3fd251ca1 100644 --- a/drivers/gpu/drm/i915/display/intel_dp_mst.c +++ b/drivers/gpu/drm/i915/display/intel_dp_mst.c @@ -50,7 +50,7 @@ static int intel_dp_mst_compute_link_config(struct intel_encoder *encoder, const struct drm_display_mode *adjusted_mode = &crtc_state->hw.adjusted_mode; void *port = connector->port; - bool constant_n = drm_dp_has_quirk(&intel_dp->desc, + bool constant_n = drm_dp_has_quirk(&intel_dp->desc, 0, DP_DPCD_QUIRK_CONSTANT_N); int bpp, slots = -EINVAL; @@ -548,12 +548,41 @@ static int intel_dp_mst_get_ddc_modes(struct drm_connector *connector) return ret; } +static int +intel_dp_mst_connector_late_register(struct drm_connector *connector) +{ + struct intel_connector *intel_connector = to_intel_connector(connector); + int ret; + + ret = drm_dp_mst_connector_late_register(connector, + intel_connector->port); + if (ret < 0) + return ret; + + ret = intel_connector_register(connector); + if (ret < 0) + drm_dp_mst_connector_early_unregister(connector, + intel_connector->port); + + return ret; +} + +static void +intel_dp_mst_connector_early_unregister(struct drm_connector *connector) +{ + struct intel_connector *intel_connector = to_intel_connector(connector); + + intel_connector_unregister(connector); + drm_dp_mst_connector_early_unregister(connector, + intel_connector->port); +} + static const struct drm_connector_funcs intel_dp_mst_connector_funcs = { .fill_modes = drm_helper_probe_single_connector_modes, .atomic_get_property = intel_digital_connector_atomic_get_property, .atomic_set_property = intel_digital_connector_atomic_set_property, - .late_register = intel_connector_register, - .early_unregister = intel_connector_unregister, + .late_register = intel_dp_mst_connector_late_register, + .early_unregister = intel_dp_mst_connector_early_unregister, .destroy = intel_connector_destroy, .atomic_destroy_state = drm_atomic_helper_connector_destroy_state, .atomic_duplicate_state = intel_digital_connector_duplicate_state, diff --git a/drivers/gpu/drm/i915/display/intel_dpll_mgr.c b/drivers/gpu/drm/i915/display/intel_dpll_mgr.c index e5bfe5245276..2d47f1f756a2 100644 --- a/drivers/gpu/drm/i915/display/intel_dpll_mgr.c +++ b/drivers/gpu/drm/i915/display/intel_dpll_mgr.c @@ -45,6 +45,22 @@ * commit phase. */ +struct intel_dpll_mgr { + const struct dpll_info *dpll_info; + + bool (*get_dplls)(struct intel_atomic_state *state, + struct intel_crtc *crtc, + struct intel_encoder *encoder); + void (*put_dplls)(struct intel_atomic_state *state, + struct intel_crtc *crtc); + void (*update_active_dpll)(struct intel_atomic_state *state, + struct intel_crtc *crtc, + struct intel_encoder *encoder); + void (*update_ref_clks)(struct drm_i915_private *i915); + void (*dump_hw_state)(struct drm_i915_private *dev_priv, + const struct intel_dpll_hw_state *hw_state); +}; + static void intel_atomic_duplicate_dpll_state(struct drm_i915_private *dev_priv, struct intel_shared_dpll_state *shared_dpll) @@ -52,8 +68,8 @@ intel_atomic_duplicate_dpll_state(struct drm_i915_private *dev_priv, enum intel_dpll_id i; /* Copy shared dpll state */ - for (i = 0; i < dev_priv->num_shared_dpll; i++) { - struct intel_shared_dpll *pll = &dev_priv->shared_dplls[i]; + for (i = 0; i < dev_priv->dpll.num_shared_dpll; i++) { + struct intel_shared_dpll *pll = &dev_priv->dpll.shared_dplls[i]; shared_dpll[i] = pll->state; } @@ -88,7 +104,7 @@ struct intel_shared_dpll * intel_get_shared_dpll_by_id(struct drm_i915_private *dev_priv, enum intel_dpll_id id) { - return &dev_priv->shared_dplls[id]; + return &dev_priv->dpll.shared_dplls[id]; } /** @@ -103,11 +119,14 @@ enum intel_dpll_id intel_get_shared_dpll_id(struct drm_i915_private *dev_priv, struct intel_shared_dpll *pll) { - if (drm_WARN_ON(&dev_priv->drm, pll < dev_priv->shared_dplls || - pll > &dev_priv->shared_dplls[dev_priv->num_shared_dpll])) + long pll_idx = pll - dev_priv->dpll.shared_dplls; + + if (drm_WARN_ON(&dev_priv->drm, + pll_idx < 0 || + pll_idx >= dev_priv->dpll.num_shared_dpll)) return -1; - return (enum intel_dpll_id) (pll - dev_priv->shared_dplls); + return pll_idx; } /* For ILK+ */ @@ -144,7 +163,7 @@ void intel_prepare_shared_dpll(const struct intel_crtc_state *crtc_state) if (drm_WARN_ON(&dev_priv->drm, pll == NULL)) return; - mutex_lock(&dev_priv->dpll_lock); + mutex_lock(&dev_priv->dpll.lock); drm_WARN_ON(&dev_priv->drm, !pll->state.crtc_mask); if (!pll->active_mask) { drm_dbg(&dev_priv->drm, "setting up %s\n", pll->info->name); @@ -153,7 +172,7 @@ void intel_prepare_shared_dpll(const struct intel_crtc_state *crtc_state) pll->info->funcs->prepare(dev_priv, pll); } - mutex_unlock(&dev_priv->dpll_lock); + mutex_unlock(&dev_priv->dpll.lock); } /** @@ -173,7 +192,7 @@ void intel_enable_shared_dpll(const struct intel_crtc_state *crtc_state) if (drm_WARN_ON(&dev_priv->drm, pll == NULL)) return; - mutex_lock(&dev_priv->dpll_lock); + mutex_lock(&dev_priv->dpll.lock); old_mask = pll->active_mask; if (drm_WARN_ON(&dev_priv->drm, !(pll->state.crtc_mask & crtc_mask)) || @@ -199,7 +218,7 @@ void intel_enable_shared_dpll(const struct intel_crtc_state *crtc_state) pll->on = true; out: - mutex_unlock(&dev_priv->dpll_lock); + mutex_unlock(&dev_priv->dpll.lock); } /** @@ -222,7 +241,7 @@ void intel_disable_shared_dpll(const struct intel_crtc_state *crtc_state) if (pll == NULL) return; - mutex_lock(&dev_priv->dpll_lock); + mutex_lock(&dev_priv->dpll.lock); if (drm_WARN_ON(&dev_priv->drm, !(pll->active_mask & crtc_mask))) goto out; @@ -243,7 +262,7 @@ void intel_disable_shared_dpll(const struct intel_crtc_state *crtc_state) pll->on = false; out: - mutex_unlock(&dev_priv->dpll_lock); + mutex_unlock(&dev_priv->dpll.lock); } static struct intel_shared_dpll * @@ -262,7 +281,7 @@ intel_find_shared_dpll(struct intel_atomic_state *state, drm_WARN_ON(&dev_priv->drm, dpll_mask & ~(BIT(I915_NUM_PLLS) - 1)); for_each_set_bit(i, &dpll_mask, I915_NUM_PLLS) { - pll = &dev_priv->shared_dplls[i]; + pll = &dev_priv->dpll.shared_dplls[i]; /* Only want to check enabled timings first */ if (shared_dpll[i].crtc_mask == 0) { @@ -362,9 +381,9 @@ void intel_shared_dpll_swap_state(struct intel_atomic_state *state) if (!state->dpll_set) return; - for (i = 0; i < dev_priv->num_shared_dpll; i++) { + for (i = 0; i < dev_priv->dpll.num_shared_dpll; i++) { struct intel_shared_dpll *pll = - &dev_priv->shared_dplls[i]; + &dev_priv->dpll.shared_dplls[i]; swap(pll->state, shared_dpll[i]); } @@ -462,7 +481,7 @@ static bool ibx_get_dpll(struct intel_atomic_state *state, if (HAS_PCH_IBX(dev_priv)) { /* Ironlake PCH has a fixed PLL->PCH pipe mapping. */ i = (enum intel_dpll_id) crtc->pipe; - pll = &dev_priv->shared_dplls[i]; + pll = &dev_priv->dpll.shared_dplls[i]; drm_dbg_kms(&dev_priv->drm, "[CRTC:%d:%s] using pre-allocated %s\n", @@ -506,6 +525,19 @@ static const struct intel_shared_dpll_funcs ibx_pch_dpll_funcs = { .get_hw_state = ibx_pch_dpll_get_hw_state, }; +static const struct dpll_info pch_plls[] = { + { "PCH DPLL A", &ibx_pch_dpll_funcs, DPLL_ID_PCH_PLL_A, 0 }, + { "PCH DPLL B", &ibx_pch_dpll_funcs, DPLL_ID_PCH_PLL_B, 0 }, + { }, +}; + +static const struct intel_dpll_mgr pch_pll_mgr = { + .dpll_info = pch_plls, + .get_dplls = ibx_get_dpll, + .put_dplls = intel_put_dpll, + .dump_hw_state = ibx_dump_hw_state, +}; + static void hsw_ddi_wrpll_enable(struct drm_i915_private *dev_priv, struct intel_shared_dpll *pll) { @@ -818,8 +850,8 @@ hsw_ddi_calculate_wrpll(int clock /* in Hz */, } static struct intel_shared_dpll * -hsw_ddi_hdmi_get_dpll(struct intel_atomic_state *state, - struct intel_crtc *crtc) +hsw_ddi_wrpll_get_dpll(struct intel_atomic_state *state, + struct intel_crtc *crtc) { struct intel_crtc_state *crtc_state = intel_atomic_get_new_crtc_state(state, crtc); @@ -846,8 +878,47 @@ hsw_ddi_hdmi_get_dpll(struct intel_atomic_state *state, return pll; } +static int hsw_ddi_wrpll_get_freq(struct drm_i915_private *dev_priv, + const struct intel_shared_dpll *pll) +{ + int refclk; + int n, p, r; + u32 wrpll = pll->state.hw_state.wrpll; + + switch (wrpll & WRPLL_REF_MASK) { + case WRPLL_REF_SPECIAL_HSW: + /* Muxed-SSC for BDW, non-SSC for non-ULT HSW. */ + if (IS_HASWELL(dev_priv) && !IS_HSW_ULT(dev_priv)) { + refclk = dev_priv->dpll.ref_clks.nssc; + break; + } + /* fall through */ + case WRPLL_REF_PCH_SSC: + /* + * We could calculate spread here, but our checking + * code only cares about 5% accuracy, and spread is a max of + * 0.5% downspread. + */ + refclk = dev_priv->dpll.ref_clks.ssc; + break; + case WRPLL_REF_LCPLL: + refclk = 2700000; + break; + default: + MISSING_CASE(wrpll); + return 0; + } + + r = wrpll & WRPLL_DIVIDER_REF_MASK; + p = (wrpll & WRPLL_DIVIDER_POST_MASK) >> WRPLL_DIVIDER_POST_SHIFT; + n = (wrpll & WRPLL_DIVIDER_FB_MASK) >> WRPLL_DIVIDER_FB_SHIFT; + + /* Convert to KHz, p & r have a fixed point portion */ + return (refclk * n / 10) / (p * r) * 2; +} + static struct intel_shared_dpll * -hsw_ddi_dp_get_dpll(struct intel_crtc_state *crtc_state) +hsw_ddi_lcpll_get_dpll(struct intel_crtc_state *crtc_state) { struct drm_i915_private *dev_priv = to_i915(crtc_state->uapi.crtc->dev); struct intel_shared_dpll *pll; @@ -878,6 +949,69 @@ hsw_ddi_dp_get_dpll(struct intel_crtc_state *crtc_state) return pll; } +static int hsw_ddi_lcpll_get_freq(struct drm_i915_private *i915, + const struct intel_shared_dpll *pll) +{ + int link_clock = 0; + + switch (pll->info->id) { + case DPLL_ID_LCPLL_810: + link_clock = 81000; + break; + case DPLL_ID_LCPLL_1350: + link_clock = 135000; + break; + case DPLL_ID_LCPLL_2700: + link_clock = 270000; + break; + default: + drm_WARN(&i915->drm, 1, "bad port clock sel\n"); + break; + } + + return link_clock * 2; +} + +static struct intel_shared_dpll * +hsw_ddi_spll_get_dpll(struct intel_atomic_state *state, + struct intel_crtc *crtc) +{ + struct intel_crtc_state *crtc_state = + intel_atomic_get_new_crtc_state(state, crtc); + + if (WARN_ON(crtc_state->port_clock / 2 != 135000)) + return NULL; + + crtc_state->dpll_hw_state.spll = SPLL_PLL_ENABLE | SPLL_FREQ_1350MHz | + SPLL_REF_MUXED_SSC; + + return intel_find_shared_dpll(state, crtc, &crtc_state->dpll_hw_state, + BIT(DPLL_ID_SPLL)); +} + +static int hsw_ddi_spll_get_freq(struct drm_i915_private *i915, + const struct intel_shared_dpll *pll) +{ + int link_clock = 0; + + switch (pll->state.hw_state.spll & SPLL_FREQ_MASK) { + case SPLL_FREQ_810MHz: + link_clock = 81000; + break; + case SPLL_FREQ_1350MHz: + link_clock = 135000; + break; + case SPLL_FREQ_2700MHz: + link_clock = 270000; + break; + default: + drm_WARN(&i915->drm, 1, "bad spll freq\n"); + break; + } + + return link_clock * 2; +} + static bool hsw_get_dpll(struct intel_atomic_state *state, struct intel_crtc *crtc, struct intel_encoder *encoder) @@ -889,23 +1023,14 @@ static bool hsw_get_dpll(struct intel_atomic_state *state, memset(&crtc_state->dpll_hw_state, 0, sizeof(crtc_state->dpll_hw_state)); - if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_HDMI)) { - pll = hsw_ddi_hdmi_get_dpll(state, crtc); - } else if (intel_crtc_has_dp_encoder(crtc_state)) { - pll = hsw_ddi_dp_get_dpll(crtc_state); - } else if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_ANALOG)) { - if (WARN_ON(crtc_state->port_clock / 2 != 135000)) - return false; - - crtc_state->dpll_hw_state.spll = - SPLL_PLL_ENABLE | SPLL_FREQ_1350MHz | SPLL_REF_MUXED_SSC; - - pll = intel_find_shared_dpll(state, crtc, - &crtc_state->dpll_hw_state, - BIT(DPLL_ID_SPLL)); - } else { + if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_HDMI)) + pll = hsw_ddi_wrpll_get_dpll(state, crtc); + else if (intel_crtc_has_dp_encoder(crtc_state)) + pll = hsw_ddi_lcpll_get_dpll(crtc_state); + else if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_ANALOG)) + pll = hsw_ddi_spll_get_dpll(state, crtc); + else return false; - } if (!pll) return false; @@ -918,6 +1043,16 @@ static bool hsw_get_dpll(struct intel_atomic_state *state, return true; } +static void hsw_update_dpll_ref_clks(struct drm_i915_private *i915) +{ + i915->dpll.ref_clks.ssc = 135000; + /* Non-SSC is only used on non-ULT HSW. */ + if (intel_de_read(i915, FUSE_STRAP3) & HSW_REF_CLK_SELECT) + i915->dpll.ref_clks.nssc = 24000; + else + i915->dpll.ref_clks.nssc = 135000; +} + static void hsw_dump_hw_state(struct drm_i915_private *dev_priv, const struct intel_dpll_hw_state *hw_state) { @@ -929,12 +1064,14 @@ static const struct intel_shared_dpll_funcs hsw_ddi_wrpll_funcs = { .enable = hsw_ddi_wrpll_enable, .disable = hsw_ddi_wrpll_disable, .get_hw_state = hsw_ddi_wrpll_get_hw_state, + .get_freq = hsw_ddi_wrpll_get_freq, }; static const struct intel_shared_dpll_funcs hsw_ddi_spll_funcs = { .enable = hsw_ddi_spll_enable, .disable = hsw_ddi_spll_disable, .get_hw_state = hsw_ddi_spll_get_hw_state, + .get_freq = hsw_ddi_spll_get_freq, }; static void hsw_ddi_lcpll_enable(struct drm_i915_private *dev_priv, @@ -958,6 +1095,25 @@ static const struct intel_shared_dpll_funcs hsw_ddi_lcpll_funcs = { .enable = hsw_ddi_lcpll_enable, .disable = hsw_ddi_lcpll_disable, .get_hw_state = hsw_ddi_lcpll_get_hw_state, + .get_freq = hsw_ddi_lcpll_get_freq, +}; + +static const struct dpll_info hsw_plls[] = { + { "WRPLL 1", &hsw_ddi_wrpll_funcs, DPLL_ID_WRPLL1, 0 }, + { "WRPLL 2", &hsw_ddi_wrpll_funcs, DPLL_ID_WRPLL2, 0 }, + { "SPLL", &hsw_ddi_spll_funcs, DPLL_ID_SPLL, 0 }, + { "LCPLL 810", &hsw_ddi_lcpll_funcs, DPLL_ID_LCPLL_810, INTEL_DPLL_ALWAYS_ON }, + { "LCPLL 1350", &hsw_ddi_lcpll_funcs, DPLL_ID_LCPLL_1350, INTEL_DPLL_ALWAYS_ON }, + { "LCPLL 2700", &hsw_ddi_lcpll_funcs, DPLL_ID_LCPLL_2700, INTEL_DPLL_ALWAYS_ON }, + { }, +}; + +static const struct intel_dpll_mgr hsw_pll_mgr = { + .dpll_info = hsw_plls, + .get_dplls = hsw_get_dpll, + .put_dplls = intel_put_dpll, + .update_ref_clks = hsw_update_dpll_ref_clks, + .dump_hw_state = hsw_dump_hw_state, }; struct skl_dpll_regs { @@ -1230,6 +1386,7 @@ struct skl_wrpll_params { static void skl_wrpll_params_populate(struct skl_wrpll_params *params, u64 afe_clock, + int ref_clock, u64 central_freq, u32 p0, u32 p1, u32 p2) { @@ -1289,14 +1446,15 @@ static void skl_wrpll_params_populate(struct skl_wrpll_params *params, * Intermediate values are in Hz. * Divide by MHz to match bsepc */ - params->dco_integer = div_u64(dco_freq, 24 * MHz(1)); + params->dco_integer = div_u64(dco_freq, ref_clock * KHz(1)); params->dco_fraction = - div_u64((div_u64(dco_freq, 24) - + div_u64((div_u64(dco_freq, ref_clock / KHz(1)) - params->dco_integer * MHz(1)) * 0x8000, MHz(1)); } static bool skl_ddi_calculate_wrpll(int clock /* in Hz */, + int ref_clock, struct skl_wrpll_params *wrpll_params) { u64 afe_clock = clock * 5; /* AFE Clock is 5x Pixel clock */ @@ -1362,14 +1520,15 @@ skip_remaining_dividers: */ p0 = p1 = p2 = 0; skl_wrpll_get_multipliers(ctx.p, &p0, &p1, &p2); - skl_wrpll_params_populate(wrpll_params, afe_clock, ctx.central_freq, - p0, p1, p2); + skl_wrpll_params_populate(wrpll_params, afe_clock, ref_clock, + ctx.central_freq, p0, p1, p2); return true; } static bool skl_ddi_hdmi_pll_dividers(struct intel_crtc_state *crtc_state) { + struct drm_i915_private *i915 = to_i915(crtc_state->uapi.crtc->dev); u32 ctrl1, cfgcr1, cfgcr2; struct skl_wrpll_params wrpll_params = { 0, }; @@ -1382,6 +1541,7 @@ static bool skl_ddi_hdmi_pll_dividers(struct intel_crtc_state *crtc_state) ctrl1 |= DPLL_CTRL1_HDMI_MODE(0); if (!skl_ddi_calculate_wrpll(crtc_state->port_clock * 1000, + i915->dpll.ref_clks.nssc, &wrpll_params)) return false; @@ -1404,6 +1564,64 @@ static bool skl_ddi_hdmi_pll_dividers(struct intel_crtc_state *crtc_state) return true; } +static int skl_ddi_wrpll_get_freq(struct drm_i915_private *i915, + const struct intel_shared_dpll *pll) +{ + const struct intel_dpll_hw_state *pll_state = &pll->state.hw_state; + int ref_clock = i915->dpll.ref_clks.nssc; + u32 p0, p1, p2, dco_freq; + + p0 = pll_state->cfgcr2 & DPLL_CFGCR2_PDIV_MASK; + p2 = pll_state->cfgcr2 & DPLL_CFGCR2_KDIV_MASK; + + if (pll_state->cfgcr2 & DPLL_CFGCR2_QDIV_MODE(1)) + p1 = (pll_state->cfgcr2 & DPLL_CFGCR2_QDIV_RATIO_MASK) >> 8; + else + p1 = 1; + + + switch (p0) { + case DPLL_CFGCR2_PDIV_1: + p0 = 1; + break; + case DPLL_CFGCR2_PDIV_2: + p0 = 2; + break; + case DPLL_CFGCR2_PDIV_3: + p0 = 3; + break; + case DPLL_CFGCR2_PDIV_7: + p0 = 7; + break; + } + + switch (p2) { + case DPLL_CFGCR2_KDIV_5: + p2 = 5; + break; + case DPLL_CFGCR2_KDIV_2: + p2 = 2; + break; + case DPLL_CFGCR2_KDIV_3: + p2 = 3; + break; + case DPLL_CFGCR2_KDIV_1: + p2 = 1; + break; + } + + dco_freq = (pll_state->cfgcr1 & DPLL_CFGCR1_DCO_INTEGER_MASK) * + ref_clock; + + dco_freq += ((pll_state->cfgcr1 & DPLL_CFGCR1_DCO_FRACTION_MASK) >> 9) * + ref_clock / 0x8000; + + if (WARN_ON(p0 == 0 || p1 == 0 || p2 == 0)) + return 0; + + return dco_freq / (p0 * p1 * p2 * 5); +} + static bool skl_ddi_dp_set_dpll_hw_state(struct intel_crtc_state *crtc_state) { @@ -1444,6 +1662,40 @@ skl_ddi_dp_set_dpll_hw_state(struct intel_crtc_state *crtc_state) return true; } +static int skl_ddi_lcpll_get_freq(struct drm_i915_private *i915, + const struct intel_shared_dpll *pll) +{ + int link_clock = 0; + + switch ((pll->state.hw_state.ctrl1 & + DPLL_CTRL1_LINK_RATE_MASK(0)) >> + DPLL_CTRL1_LINK_RATE_SHIFT(0)) { + case DPLL_CTRL1_LINK_RATE_810: + link_clock = 81000; + break; + case DPLL_CTRL1_LINK_RATE_1080: + link_clock = 108000; + break; + case DPLL_CTRL1_LINK_RATE_1350: + link_clock = 135000; + break; + case DPLL_CTRL1_LINK_RATE_1620: + link_clock = 162000; + break; + case DPLL_CTRL1_LINK_RATE_2160: + link_clock = 216000; + break; + case DPLL_CTRL1_LINK_RATE_2700: + link_clock = 270000; + break; + default: + drm_WARN(&i915->drm, 1, "Unsupported link rate\n"); + break; + } + + return link_clock * 2; +} + static bool skl_get_dpll(struct intel_atomic_state *state, struct intel_crtc *crtc, struct intel_encoder *encoder) @@ -1493,6 +1745,25 @@ static bool skl_get_dpll(struct intel_atomic_state *state, return true; } +static int skl_ddi_pll_get_freq(struct drm_i915_private *i915, + const struct intel_shared_dpll *pll) +{ + /* + * ctrl1 register is already shifted for each pll, just use 0 to get + * the internal shift for each field + */ + if (pll->state.hw_state.ctrl1 & DPLL_CTRL1_HDMI_MODE(0)) + return skl_ddi_wrpll_get_freq(i915, pll); + else + return skl_ddi_lcpll_get_freq(i915, pll); +} + +static void skl_update_dpll_ref_clks(struct drm_i915_private *i915) +{ + /* No SSC ref */ + i915->dpll.ref_clks.nssc = i915->cdclk.hw.ref; +} + static void skl_dump_hw_state(struct drm_i915_private *dev_priv, const struct intel_dpll_hw_state *hw_state) { @@ -1507,12 +1778,30 @@ static const struct intel_shared_dpll_funcs skl_ddi_pll_funcs = { .enable = skl_ddi_pll_enable, .disable = skl_ddi_pll_disable, .get_hw_state = skl_ddi_pll_get_hw_state, + .get_freq = skl_ddi_pll_get_freq, }; static const struct intel_shared_dpll_funcs skl_ddi_dpll0_funcs = { .enable = skl_ddi_dpll0_enable, .disable = skl_ddi_dpll0_disable, .get_hw_state = skl_ddi_dpll0_get_hw_state, + .get_freq = skl_ddi_pll_get_freq, +}; + +static const struct dpll_info skl_plls[] = { + { "DPLL 0", &skl_ddi_dpll0_funcs, DPLL_ID_SKL_DPLL0, INTEL_DPLL_ALWAYS_ON }, + { "DPLL 1", &skl_ddi_pll_funcs, DPLL_ID_SKL_DPLL1, 0 }, + { "DPLL 2", &skl_ddi_pll_funcs, DPLL_ID_SKL_DPLL2, 0 }, + { "DPLL 3", &skl_ddi_pll_funcs, DPLL_ID_SKL_DPLL3, 0 }, + { }, +}; + +static const struct intel_dpll_mgr skl_pll_mgr = { + .dpll_info = skl_plls, + .get_dplls = skl_get_dpll, + .put_dplls = intel_put_dpll, + .update_ref_clks = skl_update_dpll_ref_clks, + .dump_hw_state = skl_dump_hw_state, }; static void bxt_ddi_pll_enable(struct drm_i915_private *dev_priv, @@ -1903,6 +2192,23 @@ bxt_ddi_hdmi_set_dpll_hw_state(struct intel_crtc_state *crtc_state) return bxt_ddi_set_dpll_hw_state(crtc_state, &clk_div); } +static int bxt_ddi_pll_get_freq(struct drm_i915_private *i915, + const struct intel_shared_dpll *pll) +{ + const struct intel_dpll_hw_state *pll_state = &pll->state.hw_state; + struct dpll clock; + + clock.m1 = 2; + clock.m2 = (pll_state->pll0 & PORT_PLL_M2_MASK) << 22; + if (pll_state->pll3 & PORT_PLL_M2_FRAC_ENABLE) + clock.m2 |= pll_state->pll2 & PORT_PLL_M2_FRAC_MASK; + clock.n = (pll_state->pll1 & PORT_PLL_N_MASK) >> PORT_PLL_N_SHIFT; + clock.p1 = (pll_state->ebb0 & PORT_PLL_P1_MASK) >> PORT_PLL_P1_SHIFT; + clock.p2 = (pll_state->ebb0 & PORT_PLL_P2_MASK) >> PORT_PLL_P2_SHIFT; + + return chv_calc_dpll_params(i915->dpll.ref_clks.nssc, &clock); +} + static bool bxt_get_dpll(struct intel_atomic_state *state, struct intel_crtc *crtc, struct intel_encoder *encoder) @@ -1936,6 +2242,13 @@ static bool bxt_get_dpll(struct intel_atomic_state *state, return true; } +static void bxt_update_dpll_ref_clks(struct drm_i915_private *i915) +{ + i915->dpll.ref_clks.ssc = 100000; + i915->dpll.ref_clks.nssc = 100000; + /* DSI non-SSC ref 19.2MHz */ +} + static void bxt_dump_hw_state(struct drm_i915_private *dev_priv, const struct intel_dpll_hw_state *hw_state) { @@ -1959,66 +2272,7 @@ static const struct intel_shared_dpll_funcs bxt_ddi_pll_funcs = { .enable = bxt_ddi_pll_enable, .disable = bxt_ddi_pll_disable, .get_hw_state = bxt_ddi_pll_get_hw_state, -}; - -struct intel_dpll_mgr { - const struct dpll_info *dpll_info; - - bool (*get_dplls)(struct intel_atomic_state *state, - struct intel_crtc *crtc, - struct intel_encoder *encoder); - void (*put_dplls)(struct intel_atomic_state *state, - struct intel_crtc *crtc); - void (*update_active_dpll)(struct intel_atomic_state *state, - struct intel_crtc *crtc, - struct intel_encoder *encoder); - void (*dump_hw_state)(struct drm_i915_private *dev_priv, - const struct intel_dpll_hw_state *hw_state); -}; - -static const struct dpll_info pch_plls[] = { - { "PCH DPLL A", &ibx_pch_dpll_funcs, DPLL_ID_PCH_PLL_A, 0 }, - { "PCH DPLL B", &ibx_pch_dpll_funcs, DPLL_ID_PCH_PLL_B, 0 }, - { }, -}; - -static const struct intel_dpll_mgr pch_pll_mgr = { - .dpll_info = pch_plls, - .get_dplls = ibx_get_dpll, - .put_dplls = intel_put_dpll, - .dump_hw_state = ibx_dump_hw_state, -}; - -static const struct dpll_info hsw_plls[] = { - { "WRPLL 1", &hsw_ddi_wrpll_funcs, DPLL_ID_WRPLL1, 0 }, - { "WRPLL 2", &hsw_ddi_wrpll_funcs, DPLL_ID_WRPLL2, 0 }, - { "SPLL", &hsw_ddi_spll_funcs, DPLL_ID_SPLL, 0 }, - { "LCPLL 810", &hsw_ddi_lcpll_funcs, DPLL_ID_LCPLL_810, INTEL_DPLL_ALWAYS_ON }, - { "LCPLL 1350", &hsw_ddi_lcpll_funcs, DPLL_ID_LCPLL_1350, INTEL_DPLL_ALWAYS_ON }, - { "LCPLL 2700", &hsw_ddi_lcpll_funcs, DPLL_ID_LCPLL_2700, INTEL_DPLL_ALWAYS_ON }, - { }, -}; - -static const struct intel_dpll_mgr hsw_pll_mgr = { - .dpll_info = hsw_plls, - .get_dplls = hsw_get_dpll, - .put_dplls = intel_put_dpll, - .dump_hw_state = hsw_dump_hw_state, -}; - -static const struct dpll_info skl_plls[] = { - { "DPLL 0", &skl_ddi_dpll0_funcs, DPLL_ID_SKL_DPLL0, INTEL_DPLL_ALWAYS_ON }, - { "DPLL 1", &skl_ddi_pll_funcs, DPLL_ID_SKL_DPLL1, 0 }, - { "DPLL 2", &skl_ddi_pll_funcs, DPLL_ID_SKL_DPLL2, 0 }, - { "DPLL 3", &skl_ddi_pll_funcs, DPLL_ID_SKL_DPLL3, 0 }, - { }, -}; - -static const struct intel_dpll_mgr skl_pll_mgr = { - .dpll_info = skl_plls, - .get_dplls = skl_get_dpll, - .put_dplls = intel_put_dpll, - .dump_hw_state = skl_dump_hw_state, + .get_freq = bxt_ddi_pll_get_freq, }; static const struct dpll_info bxt_plls[] = { @@ -2032,6 +2286,7 @@ static const struct intel_dpll_mgr bxt_pll_mgr = { .dpll_info = bxt_plls, .get_dplls = bxt_get_dpll, .put_dplls = intel_put_dpll, + .update_ref_clks = bxt_update_dpll_ref_clks, .dump_hw_state = bxt_dump_hw_state, }; @@ -2275,27 +2530,12 @@ static void cnl_wrpll_params_populate(struct skl_wrpll_params *params, params->dco_fraction = dco & 0x7fff; } -int cnl_hdmi_pll_ref_clock(struct drm_i915_private *dev_priv) -{ - int ref_clock = dev_priv->cdclk.hw.ref; - - /* - * For ICL+, the spec states: if reference frequency is 38.4, - * use 19.2 because the DPLL automatically divides that by 2. - */ - if (INTEL_GEN(dev_priv) >= 11 && ref_clock == 38400) - ref_clock = 19200; - - return ref_clock; -} - static bool -cnl_ddi_calculate_wrpll(struct intel_crtc_state *crtc_state, - struct skl_wrpll_params *wrpll_params) +__cnl_ddi_calculate_wrpll(struct intel_crtc_state *crtc_state, + struct skl_wrpll_params *wrpll_params, + int ref_clock) { - struct drm_i915_private *dev_priv = to_i915(crtc_state->uapi.crtc->dev); u32 afe_clock = crtc_state->port_clock * 5; - u32 ref_clock; u32 dco_min = 7998000; u32 dco_max = 10000000; u32 dco_mid = (dco_min + dco_max) / 2; @@ -2327,15 +2567,22 @@ cnl_ddi_calculate_wrpll(struct intel_crtc_state *crtc_state, return false; cnl_wrpll_get_multipliers(best_div, &pdiv, &qdiv, &kdiv); - - ref_clock = cnl_hdmi_pll_ref_clock(dev_priv); - cnl_wrpll_params_populate(wrpll_params, best_dco, ref_clock, pdiv, qdiv, kdiv); return true; } +static bool +cnl_ddi_calculate_wrpll(struct intel_crtc_state *crtc_state, + struct skl_wrpll_params *wrpll_params) +{ + struct drm_i915_private *i915 = to_i915(crtc_state->uapi.crtc->dev); + + return __cnl_ddi_calculate_wrpll(crtc_state, wrpll_params, + i915->dpll.ref_clks.nssc); +} + static bool cnl_ddi_hdmi_pll_dividers(struct intel_crtc_state *crtc_state) { u32 cfgcr0, cfgcr1; @@ -2363,6 +2610,68 @@ static bool cnl_ddi_hdmi_pll_dividers(struct intel_crtc_state *crtc_state) return true; } +static int __cnl_ddi_wrpll_get_freq(struct drm_i915_private *dev_priv, + const struct intel_shared_dpll *pll, + int ref_clock) +{ + const struct intel_dpll_hw_state *pll_state = &pll->state.hw_state; + u32 p0, p1, p2, dco_freq; + + p0 = pll_state->cfgcr1 & DPLL_CFGCR1_PDIV_MASK; + p2 = pll_state->cfgcr1 & DPLL_CFGCR1_KDIV_MASK; + + if (pll_state->cfgcr1 & DPLL_CFGCR1_QDIV_MODE(1)) + p1 = (pll_state->cfgcr1 & DPLL_CFGCR1_QDIV_RATIO_MASK) >> + DPLL_CFGCR1_QDIV_RATIO_SHIFT; + else + p1 = 1; + + + switch (p0) { + case DPLL_CFGCR1_PDIV_2: + p0 = 2; + break; + case DPLL_CFGCR1_PDIV_3: + p0 = 3; + break; + case DPLL_CFGCR1_PDIV_5: + p0 = 5; + break; + case DPLL_CFGCR1_PDIV_7: + p0 = 7; + break; + } + + switch (p2) { + case DPLL_CFGCR1_KDIV_1: + p2 = 1; + break; + case DPLL_CFGCR1_KDIV_2: + p2 = 2; + break; + case DPLL_CFGCR1_KDIV_3: + p2 = 3; + break; + } + + dco_freq = (pll_state->cfgcr0 & DPLL_CFGCR0_DCO_INTEGER_MASK) * + ref_clock; + + dco_freq += (((pll_state->cfgcr0 & DPLL_CFGCR0_DCO_FRACTION_MASK) >> + DPLL_CFGCR0_DCO_FRACTION_SHIFT) * ref_clock) / 0x8000; + + if (drm_WARN_ON(&dev_priv->drm, p0 == 0 || p1 == 0 || p2 == 0)) + return 0; + + return dco_freq / (p0 * p1 * p2 * 5); +} + +static int cnl_ddi_wrpll_get_freq(struct drm_i915_private *i915, + const struct intel_shared_dpll *pll) +{ + return __cnl_ddi_wrpll_get_freq(i915, pll, i915->dpll.ref_clks.nssc); +} + static bool cnl_ddi_dp_set_dpll_hw_state(struct intel_crtc_state *crtc_state) { @@ -2408,6 +2717,44 @@ cnl_ddi_dp_set_dpll_hw_state(struct intel_crtc_state *crtc_state) return true; } +static int cnl_ddi_lcpll_get_freq(struct drm_i915_private *i915, + const struct intel_shared_dpll *pll) +{ + int link_clock = 0; + + switch (pll->state.hw_state.cfgcr0 & DPLL_CFGCR0_LINK_RATE_MASK) { + case DPLL_CFGCR0_LINK_RATE_810: + link_clock = 81000; + break; + case DPLL_CFGCR0_LINK_RATE_1080: + link_clock = 108000; + break; + case DPLL_CFGCR0_LINK_RATE_1350: + link_clock = 135000; + break; + case DPLL_CFGCR0_LINK_RATE_1620: + link_clock = 162000; + break; + case DPLL_CFGCR0_LINK_RATE_2160: + link_clock = 216000; + break; + case DPLL_CFGCR0_LINK_RATE_2700: + link_clock = 270000; + break; + case DPLL_CFGCR0_LINK_RATE_3240: + link_clock = 324000; + break; + case DPLL_CFGCR0_LINK_RATE_4050: + link_clock = 405000; + break; + default: + drm_WARN(&i915->drm, 1, "Unsupported link rate\n"); + break; + } + + return link_clock * 2; +} + static bool cnl_get_dpll(struct intel_atomic_state *state, struct intel_crtc *crtc, struct intel_encoder *encoder) @@ -2457,6 +2804,21 @@ static bool cnl_get_dpll(struct intel_atomic_state *state, return true; } +static int cnl_ddi_pll_get_freq(struct drm_i915_private *i915, + const struct intel_shared_dpll *pll) +{ + if (pll->state.hw_state.cfgcr0 & DPLL_CFGCR0_HDMI_MODE) + return cnl_ddi_wrpll_get_freq(i915, pll); + else + return cnl_ddi_lcpll_get_freq(i915, pll); +} + +static void cnl_update_dpll_ref_clks(struct drm_i915_private *i915) +{ + /* No SSC reference */ + i915->dpll.ref_clks.nssc = i915->cdclk.hw.ref; +} + static void cnl_dump_hw_state(struct drm_i915_private *dev_priv, const struct intel_dpll_hw_state *hw_state) { @@ -2470,6 +2832,7 @@ static const struct intel_shared_dpll_funcs cnl_ddi_pll_funcs = { .enable = cnl_ddi_pll_enable, .disable = cnl_ddi_pll_disable, .get_hw_state = cnl_ddi_pll_get_hw_state, + .get_freq = cnl_ddi_pll_get_freq, }; static const struct dpll_info cnl_plls[] = { @@ -2483,6 +2846,7 @@ static const struct intel_dpll_mgr cnl_pll_mgr = { .dpll_info = cnl_plls, .get_dplls = cnl_get_dpll, .put_dplls = intel_put_dpll, + .update_ref_clks = cnl_update_dpll_ref_clks, .dump_hw_state = cnl_dump_hw_state, }; @@ -2578,7 +2942,7 @@ static bool icl_calc_dp_combo_pll(struct intel_crtc_state *crtc_state, { struct drm_i915_private *dev_priv = to_i915(crtc_state->uapi.crtc->dev); const struct icl_combo_pll_params *params = - dev_priv->cdclk.hw.ref == 24000 ? + dev_priv->dpll.ref_clks.nssc == 24000 ? icl_dp_combo_pll_24MHz_values : icl_dp_combo_pll_19_2MHz_values; int clock = crtc_state->port_clock; @@ -2601,9 +2965,9 @@ static bool icl_calc_tbt_pll(struct intel_crtc_state *crtc_state, struct drm_i915_private *dev_priv = to_i915(crtc_state->uapi.crtc->dev); if (INTEL_GEN(dev_priv) >= 12) { - switch (dev_priv->cdclk.hw.ref) { + switch (dev_priv->dpll.ref_clks.nssc) { default: - MISSING_CASE(dev_priv->cdclk.hw.ref); + MISSING_CASE(dev_priv->dpll.ref_clks.nssc); /* fall-through */ case 19200: case 38400: @@ -2614,9 +2978,9 @@ static bool icl_calc_tbt_pll(struct intel_crtc_state *crtc_state, break; } } else { - switch (dev_priv->cdclk.hw.ref) { + switch (dev_priv->dpll.ref_clks.nssc) { default: - MISSING_CASE(dev_priv->cdclk.hw.ref); + MISSING_CASE(dev_priv->dpll.ref_clks.nssc); /* fall-through */ case 19200: case 38400: @@ -2631,6 +2995,49 @@ static bool icl_calc_tbt_pll(struct intel_crtc_state *crtc_state, return true; } +static int icl_ddi_tbt_pll_get_freq(struct drm_i915_private *i915, + const struct intel_shared_dpll *pll) +{ + /* + * The PLL outputs multiple frequencies at the same time, selection is + * made at DDI clock mux level. + */ + drm_WARN_ON(&i915->drm, 1); + + return 0; +} + +static int icl_wrpll_ref_clock(struct drm_i915_private *i915) +{ + int ref_clock = i915->dpll.ref_clks.nssc; + + /* + * For ICL+, the spec states: if reference frequency is 38.4, + * use 19.2 because the DPLL automatically divides that by 2. + */ + if (ref_clock == 38400) + ref_clock = 19200; + + return ref_clock; +} + +static bool +icl_calc_wrpll(struct intel_crtc_state *crtc_state, + struct skl_wrpll_params *wrpll_params) +{ + struct drm_i915_private *i915 = to_i915(crtc_state->uapi.crtc->dev); + + return __cnl_ddi_calculate_wrpll(crtc_state, wrpll_params, + icl_wrpll_ref_clock(i915)); +} + +static int icl_ddi_combo_pll_get_freq(struct drm_i915_private *i915, + const struct intel_shared_dpll *pll) +{ + return __cnl_ddi_wrpll_get_freq(i915, pll, + icl_wrpll_ref_clock(i915)); +} + static bool icl_calc_dpll_state(struct intel_crtc_state *crtc_state, struct intel_encoder *encoder, struct intel_dpll_hw_state *pll_state) @@ -2645,7 +3052,7 @@ static bool icl_calc_dpll_state(struct intel_crtc_state *crtc_state, ret = icl_calc_tbt_pll(crtc_state, &pll_params); else if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_HDMI) || intel_crtc_has_type(crtc_state, INTEL_OUTPUT_DSI)) - ret = cnl_ddi_calculate_wrpll(crtc_state, &pll_params); + ret = icl_calc_wrpll(crtc_state, &pll_params); else ret = icl_calc_dp_combo_pll(crtc_state, &pll_params); @@ -2768,7 +3175,7 @@ static bool icl_calc_mg_pll_state(struct intel_crtc_state *crtc_state, struct intel_dpll_hw_state *pll_state) { struct drm_i915_private *dev_priv = to_i915(crtc_state->uapi.crtc->dev); - int refclk_khz = dev_priv->cdclk.hw.ref; + int refclk_khz = dev_priv->dpll.ref_clks.nssc; int clock = crtc_state->port_clock; u32 dco_khz, m1div, m2div_int, m2div_rem, m2div_frac; u32 iref_ndiv, iref_trim, iref_pulse_w; @@ -2969,6 +3376,78 @@ static bool icl_calc_mg_pll_state(struct intel_crtc_state *crtc_state, return true; } +static int icl_ddi_mg_pll_get_freq(struct drm_i915_private *dev_priv, + const struct intel_shared_dpll *pll) +{ + const struct intel_dpll_hw_state *pll_state = &pll->state.hw_state; + u32 m1, m2_int, m2_frac, div1, div2, ref_clock; + u64 tmp; + + ref_clock = dev_priv->dpll.ref_clks.nssc; + + if (INTEL_GEN(dev_priv) >= 12) { + m1 = pll_state->mg_pll_div0 & DKL_PLL_DIV0_FBPREDIV_MASK; + m1 = m1 >> DKL_PLL_DIV0_FBPREDIV_SHIFT; + m2_int = pll_state->mg_pll_div0 & DKL_PLL_DIV0_FBDIV_INT_MASK; + + if (pll_state->mg_pll_bias & DKL_PLL_BIAS_FRAC_EN_H) { + m2_frac = pll_state->mg_pll_bias & + DKL_PLL_BIAS_FBDIV_FRAC_MASK; + m2_frac = m2_frac >> DKL_PLL_BIAS_FBDIV_SHIFT; + } else { + m2_frac = 0; + } + } else { + m1 = pll_state->mg_pll_div1 & MG_PLL_DIV1_FBPREDIV_MASK; + m2_int = pll_state->mg_pll_div0 & MG_PLL_DIV0_FBDIV_INT_MASK; + + if (pll_state->mg_pll_div0 & MG_PLL_DIV0_FRACNEN_H) { + m2_frac = pll_state->mg_pll_div0 & + MG_PLL_DIV0_FBDIV_FRAC_MASK; + m2_frac = m2_frac >> MG_PLL_DIV0_FBDIV_FRAC_SHIFT; + } else { + m2_frac = 0; + } + } + + switch (pll_state->mg_clktop2_hsclkctl & + MG_CLKTOP2_HSCLKCTL_HSDIV_RATIO_MASK) { + case MG_CLKTOP2_HSCLKCTL_HSDIV_RATIO_2: + div1 = 2; + break; + case MG_CLKTOP2_HSCLKCTL_HSDIV_RATIO_3: + div1 = 3; + break; + case MG_CLKTOP2_HSCLKCTL_HSDIV_RATIO_5: + div1 = 5; + break; + case MG_CLKTOP2_HSCLKCTL_HSDIV_RATIO_7: + div1 = 7; + break; + default: + MISSING_CASE(pll_state->mg_clktop2_hsclkctl); + return 0; + } + + div2 = (pll_state->mg_clktop2_hsclkctl & + MG_CLKTOP2_HSCLKCTL_DSDIV_RATIO_MASK) >> + MG_CLKTOP2_HSCLKCTL_DSDIV_RATIO_SHIFT; + + /* div2 value of 0 is same as 1 means no div */ + if (div2 == 0) + div2 = 1; + + /* + * Adjust the original formula to delay the division by 2^22 in order to + * minimize possible rounding errors. + */ + tmp = (u64)m1 * m2_int * ref_clock + + (((u64)m1 * m2_frac * ref_clock) >> 22); + tmp = div_u64(tmp, 5 * div1 * div2); + + return tmp; +} + /** * icl_set_active_port_dpll - select the active port DPLL for a given CRTC * @crtc_state: state for the CRTC to select the DPLL for @@ -3201,7 +3680,7 @@ static bool mg_pll_get_hw_state(struct drm_i915_private *dev_priv, hw_state->mg_pll_tdc_coldst_bias = intel_de_read(dev_priv, MG_PLL_TDC_COLDST_BIAS(tc_port)); - if (dev_priv->cdclk.hw.ref == 38400) { + if (dev_priv->dpll.ref_clks.nssc == 38400) { hw_state->mg_pll_tdc_coldst_bias_mask = MG_PLL_TDC_COLDST_COLDSTART; hw_state->mg_pll_bias_mask = 0; } else { @@ -3682,6 +4161,12 @@ static void mg_pll_disable(struct drm_i915_private *dev_priv, icl_pll_disable(dev_priv, pll, enable_reg); } +static void icl_update_dpll_ref_clks(struct drm_i915_private *i915) +{ + /* No SSC ref */ + i915->dpll.ref_clks.nssc = i915->cdclk.hw.ref; +} + static void icl_dump_hw_state(struct drm_i915_private *dev_priv, const struct intel_dpll_hw_state *hw_state) { @@ -3709,18 +4194,21 @@ static const struct intel_shared_dpll_funcs combo_pll_funcs = { .enable = combo_pll_enable, .disable = combo_pll_disable, .get_hw_state = combo_pll_get_hw_state, + .get_freq = icl_ddi_combo_pll_get_freq, }; static const struct intel_shared_dpll_funcs tbt_pll_funcs = { .enable = tbt_pll_enable, .disable = tbt_pll_disable, .get_hw_state = tbt_pll_get_hw_state, + .get_freq = icl_ddi_tbt_pll_get_freq, }; static const struct intel_shared_dpll_funcs mg_pll_funcs = { .enable = mg_pll_enable, .disable = mg_pll_disable, .get_hw_state = mg_pll_get_hw_state, + .get_freq = icl_ddi_mg_pll_get_freq, }; static const struct dpll_info icl_plls[] = { @@ -3739,6 +4227,7 @@ static const struct intel_dpll_mgr icl_pll_mgr = { .get_dplls = icl_get_dplls, .put_dplls = icl_put_dplls, .update_active_dpll = icl_update_active_dpll, + .update_ref_clks = icl_update_dpll_ref_clks, .dump_hw_state = icl_dump_hw_state, }; @@ -3753,6 +4242,7 @@ static const struct intel_dpll_mgr ehl_pll_mgr = { .dpll_info = ehl_plls, .get_dplls = icl_get_dplls, .put_dplls = icl_put_dplls, + .update_ref_clks = icl_update_dpll_ref_clks, .dump_hw_state = icl_dump_hw_state, }; @@ -3760,6 +4250,7 @@ static const struct intel_shared_dpll_funcs dkl_pll_funcs = { .enable = mg_pll_enable, .disable = mg_pll_disable, .get_hw_state = dkl_pll_get_hw_state, + .get_freq = icl_ddi_mg_pll_get_freq, }; static const struct dpll_info tgl_plls[] = { @@ -3780,6 +4271,7 @@ static const struct intel_dpll_mgr tgl_pll_mgr = { .get_dplls = icl_get_dplls, .put_dplls = icl_put_dplls, .update_active_dpll = icl_update_active_dpll, + .update_ref_clks = icl_update_dpll_ref_clks, .dump_hw_state = icl_dump_hw_state, }; @@ -3814,7 +4306,7 @@ void intel_shared_dpll_init(struct drm_device *dev) dpll_mgr = &pch_pll_mgr; if (!dpll_mgr) { - dev_priv->num_shared_dpll = 0; + dev_priv->dpll.num_shared_dpll = 0; return; } @@ -3822,14 +4314,14 @@ void intel_shared_dpll_init(struct drm_device *dev) for (i = 0; dpll_info[i].name; i++) { drm_WARN_ON(dev, i != dpll_info[i].id); - dev_priv->shared_dplls[i].info = &dpll_info[i]; + dev_priv->dpll.shared_dplls[i].info = &dpll_info[i]; } - dev_priv->dpll_mgr = dpll_mgr; - dev_priv->num_shared_dpll = i; - mutex_init(&dev_priv->dpll_lock); + dev_priv->dpll.mgr = dpll_mgr; + dev_priv->dpll.num_shared_dpll = i; + mutex_init(&dev_priv->dpll.lock); - BUG_ON(dev_priv->num_shared_dpll > I915_NUM_PLLS); + BUG_ON(dev_priv->dpll.num_shared_dpll > I915_NUM_PLLS); } /** @@ -3856,7 +4348,7 @@ bool intel_reserve_shared_dplls(struct intel_atomic_state *state, struct intel_encoder *encoder) { struct drm_i915_private *dev_priv = to_i915(state->base.dev); - const struct intel_dpll_mgr *dpll_mgr = dev_priv->dpll_mgr; + const struct intel_dpll_mgr *dpll_mgr = dev_priv->dpll.mgr; if (drm_WARN_ON(&dev_priv->drm, !dpll_mgr)) return false; @@ -3879,7 +4371,7 @@ void intel_release_shared_dplls(struct intel_atomic_state *state, struct intel_crtc *crtc) { struct drm_i915_private *dev_priv = to_i915(state->base.dev); - const struct intel_dpll_mgr *dpll_mgr = dev_priv->dpll_mgr; + const struct intel_dpll_mgr *dpll_mgr = dev_priv->dpll.mgr; /* * FIXME: this function is called for every platform having a @@ -3908,7 +4400,7 @@ void intel_update_active_dpll(struct intel_atomic_state *state, struct intel_encoder *encoder) { struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); - const struct intel_dpll_mgr *dpll_mgr = dev_priv->dpll_mgr; + const struct intel_dpll_mgr *dpll_mgr = dev_priv->dpll.mgr; if (drm_WARN_ON(&dev_priv->drm, !dpll_mgr)) return; @@ -3917,6 +4409,84 @@ void intel_update_active_dpll(struct intel_atomic_state *state, } /** + * intel_dpll_get_freq - calculate the DPLL's output frequency + * @i915: i915 device + * @pll: DPLL for which to calculate the output frequency + * + * Return the output frequency corresponding to @pll's current state. + */ +int intel_dpll_get_freq(struct drm_i915_private *i915, + const struct intel_shared_dpll *pll) +{ + if (drm_WARN_ON(&i915->drm, !pll->info->funcs->get_freq)) + return 0; + + return pll->info->funcs->get_freq(i915, pll); +} + +static void readout_dpll_hw_state(struct drm_i915_private *i915, + struct intel_shared_dpll *pll) +{ + struct intel_crtc *crtc; + + pll->on = pll->info->funcs->get_hw_state(i915, pll, + &pll->state.hw_state); + + if (IS_ELKHARTLAKE(i915) && pll->on && + pll->info->id == DPLL_ID_EHL_DPLL4) { + pll->wakeref = intel_display_power_get(i915, + POWER_DOMAIN_DPLL_DC_OFF); + } + + pll->state.crtc_mask = 0; + for_each_intel_crtc(&i915->drm, crtc) { + struct intel_crtc_state *crtc_state = + to_intel_crtc_state(crtc->base.state); + + if (crtc_state->hw.active && crtc_state->shared_dpll == pll) + pll->state.crtc_mask |= 1 << crtc->pipe; + } + pll->active_mask = pll->state.crtc_mask; + + drm_dbg_kms(&i915->drm, + "%s hw state readout: crtc_mask 0x%08x, on %i\n", + pll->info->name, pll->state.crtc_mask, pll->on); +} + +void intel_dpll_readout_hw_state(struct drm_i915_private *i915) +{ + int i; + + if (i915->dpll.mgr && i915->dpll.mgr->update_ref_clks) + i915->dpll.mgr->update_ref_clks(i915); + + for (i = 0; i < i915->dpll.num_shared_dpll; i++) + readout_dpll_hw_state(i915, &i915->dpll.shared_dplls[i]); +} + +static void sanitize_dpll_state(struct drm_i915_private *i915, + struct intel_shared_dpll *pll) +{ + if (!pll->on || pll->active_mask) + return; + + drm_dbg_kms(&i915->drm, + "%s enabled but not in use, disabling\n", + pll->info->name); + + pll->info->funcs->disable(i915, pll); + pll->on = false; +} + +void intel_dpll_sanitize_state(struct drm_i915_private *i915) +{ + int i; + + for (i = 0; i < i915->dpll.num_shared_dpll; i++) + sanitize_dpll_state(i915, &i915->dpll.shared_dplls[i]); +} + +/** * intel_shared_dpll_dump_hw_state - write hw_state to dmesg * @dev_priv: i915 drm device * @hw_state: hw state to be written to the log @@ -3926,8 +4496,8 @@ void intel_update_active_dpll(struct intel_atomic_state *state, void intel_dpll_dump_hw_state(struct drm_i915_private *dev_priv, const struct intel_dpll_hw_state *hw_state) { - if (dev_priv->dpll_mgr) { - dev_priv->dpll_mgr->dump_hw_state(dev_priv, hw_state); + if (dev_priv->dpll.mgr) { + dev_priv->dpll.mgr->dump_hw_state(dev_priv, hw_state); } else { /* fallback for platforms that don't use the shared dpll * infrastructure diff --git a/drivers/gpu/drm/i915/display/intel_dpll_mgr.h b/drivers/gpu/drm/i915/display/intel_dpll_mgr.h index 2a104c64291d..5d9a2bc371e7 100644 --- a/drivers/gpu/drm/i915/display/intel_dpll_mgr.h +++ b/drivers/gpu/drm/i915/display/intel_dpll_mgr.h @@ -278,6 +278,15 @@ struct intel_shared_dpll_funcs { bool (*get_hw_state)(struct drm_i915_private *dev_priv, struct intel_shared_dpll *pll, struct intel_dpll_hw_state *hw_state); + + /** + * @get_freq: + * + * Hook for calculating the pll's output frequency based on its + * current state. + */ + int (*get_freq)(struct drm_i915_private *i915, + const struct intel_shared_dpll *pll); }; /** @@ -372,15 +381,18 @@ void icl_set_active_port_dpll(struct intel_crtc_state *crtc_state, void intel_update_active_dpll(struct intel_atomic_state *state, struct intel_crtc *crtc, struct intel_encoder *encoder); +int intel_dpll_get_freq(struct drm_i915_private *i915, + const struct intel_shared_dpll *pll); void intel_prepare_shared_dpll(const struct intel_crtc_state *crtc_state); void intel_enable_shared_dpll(const struct intel_crtc_state *crtc_state); void intel_disable_shared_dpll(const struct intel_crtc_state *crtc_state); void intel_shared_dpll_swap_state(struct intel_atomic_state *state); void intel_shared_dpll_init(struct drm_device *dev); +void intel_dpll_readout_hw_state(struct drm_i915_private *dev_priv); +void intel_dpll_sanitize_state(struct drm_i915_private *dev_priv); void intel_dpll_dump_hw_state(struct drm_i915_private *dev_priv, const struct intel_dpll_hw_state *hw_state); -int cnl_hdmi_pll_ref_clock(struct drm_i915_private *dev_priv); enum intel_dpll_id icl_tc_port_to_pll_id(enum tc_port tc_port); bool intel_dpll_is_combophy(enum intel_dpll_id id); diff --git a/drivers/gpu/drm/i915/display/intel_dsb.c b/drivers/gpu/drm/i915/display/intel_dsb.c index 76ae01277fd6..d7a6bf2277df 100644 --- a/drivers/gpu/drm/i915/display/intel_dsb.c +++ b/drivers/gpu/drm/i915/display/intel_dsb.c @@ -52,7 +52,7 @@ static inline bool intel_dsb_enable_engine(struct intel_dsb *dsb) dsb_ctrl = intel_de_read(dev_priv, DSB_CTRL(pipe, dsb->id)); if (DSB_STATUS & dsb_ctrl) { - DRM_DEBUG_KMS("DSB engine is busy.\n"); + drm_dbg_kms(&dev_priv->drm, "DSB engine is busy.\n"); return false; } @@ -72,7 +72,7 @@ static inline bool intel_dsb_disable_engine(struct intel_dsb *dsb) dsb_ctrl = intel_de_read(dev_priv, DSB_CTRL(pipe, dsb->id)); if (DSB_STATUS & dsb_ctrl) { - DRM_DEBUG_KMS("DSB engine is busy.\n"); + drm_dbg_kms(&dev_priv->drm, "DSB engine is busy.\n"); return false; } @@ -115,20 +115,20 @@ intel_dsb_get(struct intel_crtc *crtc) obj = i915_gem_object_create_internal(i915, DSB_BUF_SIZE); if (IS_ERR(obj)) { - DRM_ERROR("Gem object creation failed\n"); + drm_err(&i915->drm, "Gem object creation failed\n"); goto out; } vma = i915_gem_object_ggtt_pin(obj, NULL, 0, 0, 0); if (IS_ERR(vma)) { - DRM_ERROR("Vma creation failed\n"); + drm_err(&i915->drm, "Vma creation failed\n"); i915_gem_object_put(obj); goto out; } buf = i915_gem_object_pin_map(vma->obj, I915_MAP_WC); if (IS_ERR(buf)) { - DRM_ERROR("Command buffer creation failed\n"); + drm_err(&i915->drm, "Command buffer creation failed\n"); goto out; } @@ -203,7 +203,7 @@ void intel_dsb_indexed_reg_write(struct intel_dsb *dsb, i915_reg_t reg, } if (drm_WARN_ON(&dev_priv->drm, dsb->free_pos >= DSB_BUF_SIZE)) { - DRM_DEBUG_KMS("DSB buffer overflow\n"); + drm_dbg_kms(&dev_priv->drm, "DSB buffer overflow\n"); return; } @@ -277,7 +277,7 @@ void intel_dsb_reg_write(struct intel_dsb *dsb, i915_reg_t reg, u32 val) } if (drm_WARN_ON(&dev_priv->drm, dsb->free_pos >= DSB_BUF_SIZE)) { - DRM_DEBUG_KMS("DSB buffer overflow\n"); + drm_dbg_kms(&dev_priv->drm, "DSB buffer overflow\n"); return; } @@ -310,7 +310,8 @@ void intel_dsb_commit(struct intel_dsb *dsb) goto reset; if (is_dsb_busy(dsb)) { - DRM_ERROR("HEAD_PTR write failed - dsb engine is busy.\n"); + drm_err(&dev_priv->drm, + "HEAD_PTR write failed - dsb engine is busy.\n"); goto reset; } intel_de_write(dev_priv, DSB_HEAD(pipe, dsb->id), @@ -322,15 +323,18 @@ void intel_dsb_commit(struct intel_dsb *dsb) (tail - dsb->free_pos * 4)); if (is_dsb_busy(dsb)) { - DRM_ERROR("TAIL_PTR write failed - dsb engine is busy.\n"); + drm_err(&dev_priv->drm, + "TAIL_PTR write failed - dsb engine is busy.\n"); goto reset; } - DRM_DEBUG_KMS("DSB execution started - head 0x%x, tail 0x%x\n", - i915_ggtt_offset(dsb->vma), tail); + drm_dbg_kms(&dev_priv->drm, + "DSB execution started - head 0x%x, tail 0x%x\n", + i915_ggtt_offset(dsb->vma), tail); intel_de_write(dev_priv, DSB_TAIL(pipe, dsb->id), i915_ggtt_offset(dsb->vma) + tail); if (wait_for(!is_dsb_busy(dsb), 1)) { - DRM_ERROR("Timed out waiting for DSB workload completion.\n"); + drm_err(&dev_priv->drm, + "Timed out waiting for DSB workload completion.\n"); goto reset; } diff --git a/drivers/gpu/drm/i915/display/intel_dsi_vbt.c b/drivers/gpu/drm/i915/display/intel_dsi_vbt.c index 694498f4b719..574dcfec9577 100644 --- a/drivers/gpu/drm/i915/display/intel_dsi_vbt.c +++ b/drivers/gpu/drm/i915/display/intel_dsi_vbt.c @@ -36,7 +36,6 @@ #include <drm/drm_crtc.h> #include <drm/drm_edid.h> -#include <drm/i915_drm.h> #include <video/mipi_display.h> diff --git a/drivers/gpu/drm/i915/display/intel_dvo.c b/drivers/gpu/drm/i915/display/intel_dvo.c index 77f3d083b7a1..341d5ce8b062 100644 --- a/drivers/gpu/drm/i915/display/intel_dvo.c +++ b/drivers/gpu/drm/i915/display/intel_dvo.c @@ -30,7 +30,6 @@ #include <drm/drm_atomic_helper.h> #include <drm/drm_crtc.h> -#include <drm/i915_drm.h> #include "i915_drv.h" #include "intel_connector.h" diff --git a/drivers/gpu/drm/i915/display/intel_fbc.c b/drivers/gpu/drm/i915/display/intel_fbc.c index ddf8d3bb7a7d..2e5d835a9eaa 100644 --- a/drivers/gpu/drm/i915/display/intel_fbc.c +++ b/drivers/gpu/drm/i915/display/intel_fbc.c @@ -42,6 +42,7 @@ #include "i915_drv.h" #include "i915_trace.h" +#include "i915_vgpu.h" #include "intel_display_types.h" #include "intel_fbc.h" #include "intel_frontbuffer.h" @@ -320,7 +321,7 @@ static void gen7_fbc_activate(struct drm_i915_private *dev_priv) SNB_CPU_FENCE_ENABLE | params->fence_id); intel_de_write(dev_priv, DPFC_CPU_FENCE_OFFSET, params->crtc.fence_y_offset); - } else { + } else if (dev_priv->ggtt.num_fences) { intel_de_write(dev_priv, SNB_DPFC_CTL_SA, 0); intel_de_write(dev_priv, DPFC_CPU_FENCE_OFFSET, 0); } @@ -508,12 +509,12 @@ static int intel_fbc_alloc_cfb(struct drm_i915_private *dev_priv, fbc->compressed_llb = compressed_llb; - GEM_BUG_ON(range_overflows_t(u64, dev_priv->dsm.start, - fbc->compressed_fb.start, - U32_MAX)); - GEM_BUG_ON(range_overflows_t(u64, dev_priv->dsm.start, - fbc->compressed_llb->start, - U32_MAX)); + GEM_BUG_ON(range_overflows_end_t(u64, dev_priv->dsm.start, + fbc->compressed_fb.start, + U32_MAX)); + GEM_BUG_ON(range_overflows_end_t(u64, dev_priv->dsm.start, + fbc->compressed_llb->start, + U32_MAX)); intel_de_write(dev_priv, FBC_CFB_BASE, dev_priv->dsm.start + fbc->compressed_fb.start); intel_de_write(dev_priv, FBC_LL_BASE, @@ -691,12 +692,37 @@ static bool intel_fbc_cfb_size_changed(struct drm_i915_private *dev_priv) fbc->compressed_fb.size * fbc->threshold; } +static bool intel_fbc_can_enable(struct drm_i915_private *dev_priv) +{ + struct intel_fbc *fbc = &dev_priv->fbc; + + if (intel_vgpu_active(dev_priv)) { + fbc->no_fbc_reason = "VGPU is active"; + return false; + } + + if (!i915_modparams.enable_fbc) { + fbc->no_fbc_reason = "disabled per module param or by default"; + return false; + } + + if (fbc->underrun_detected) { + fbc->no_fbc_reason = "underrun detected"; + return false; + } + + return true; +} + static bool intel_fbc_can_activate(struct intel_crtc *crtc) { struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); struct intel_fbc *fbc = &dev_priv->fbc; struct intel_fbc_state_cache *cache = &fbc->state_cache; + if (!intel_fbc_can_enable(dev_priv)) + return false; + if (!cache->plane.visible) { fbc->no_fbc_reason = "primary plane not visible"; return false; @@ -795,28 +821,6 @@ static bool intel_fbc_can_activate(struct intel_crtc *crtc) return true; } -static bool intel_fbc_can_enable(struct drm_i915_private *dev_priv) -{ - struct intel_fbc *fbc = &dev_priv->fbc; - - if (intel_vgpu_active(dev_priv)) { - fbc->no_fbc_reason = "VGPU is active"; - return false; - } - - if (!i915_modparams.enable_fbc) { - fbc->no_fbc_reason = "disabled per module param or by default"; - return false; - } - - if (fbc->underrun_detected) { - fbc->no_fbc_reason = "underrun detected"; - return false; - } - - return true; -} - static void intel_fbc_get_reg_params(struct intel_crtc *crtc, struct intel_fbc_reg_params *params) { diff --git a/drivers/gpu/drm/i915/display/intel_fbdev.c b/drivers/gpu/drm/i915/display/intel_fbdev.c index 876264fc6560..3bc804212a99 100644 --- a/drivers/gpu/drm/i915/display/intel_fbdev.c +++ b/drivers/gpu/drm/i915/display/intel_fbdev.c @@ -40,7 +40,6 @@ #include <drm/drm_crtc.h> #include <drm/drm_fb_helper.h> #include <drm/drm_fourcc.h> -#include <drm/i915_drm.h> #include "i915_drv.h" #include "intel_display_types.h" diff --git a/drivers/gpu/drm/i915/display/intel_fifo_underrun.c b/drivers/gpu/drm/i915/display/intel_fifo_underrun.c index 470b3b0b9bdb..813a4f7033e1 100644 --- a/drivers/gpu/drm/i915/display/intel_fifo_underrun.c +++ b/drivers/gpu/drm/i915/display/intel_fifo_underrun.c @@ -103,7 +103,7 @@ static void i9xx_check_fifo_underruns(struct intel_crtc *crtc) intel_de_posting_read(dev_priv, reg); trace_intel_cpu_fifo_underrun(dev_priv, crtc->pipe); - DRM_ERROR("pipe %c underrun\n", pipe_name(crtc->pipe)); + drm_err(&dev_priv->drm, "pipe %c underrun\n", pipe_name(crtc->pipe)); } static void i9xx_set_fifo_underrun_reporting(struct drm_device *dev, @@ -123,7 +123,8 @@ static void i9xx_set_fifo_underrun_reporting(struct drm_device *dev, intel_de_posting_read(dev_priv, reg); } else { if (old && intel_de_read(dev_priv, reg) & PIPE_FIFO_UNDERRUN_STATUS) - DRM_ERROR("pipe %c underrun\n", pipe_name(pipe)); + drm_err(&dev_priv->drm, "pipe %c underrun\n", + pipe_name(pipe)); } } @@ -155,7 +156,7 @@ static void ivb_check_fifo_underruns(struct intel_crtc *crtc) intel_de_posting_read(dev_priv, GEN7_ERR_INT); trace_intel_cpu_fifo_underrun(dev_priv, pipe); - DRM_ERROR("fifo underrun on pipe %c\n", pipe_name(pipe)); + drm_err(&dev_priv->drm, "fifo underrun on pipe %c\n", pipe_name(pipe)); } static void ivb_set_fifo_underrun_reporting(struct drm_device *dev, @@ -176,8 +177,9 @@ static void ivb_set_fifo_underrun_reporting(struct drm_device *dev, if (old && intel_de_read(dev_priv, GEN7_ERR_INT) & ERR_INT_FIFO_UNDERRUN(pipe)) { - DRM_ERROR("uncleared fifo underrun on pipe %c\n", - pipe_name(pipe)); + drm_err(&dev_priv->drm, + "uncleared fifo underrun on pipe %c\n", + pipe_name(pipe)); } } } @@ -223,8 +225,8 @@ static void cpt_check_pch_fifo_underruns(struct intel_crtc *crtc) intel_de_posting_read(dev_priv, SERR_INT); trace_intel_pch_fifo_underrun(dev_priv, pch_transcoder); - DRM_ERROR("pch fifo underrun on pch transcoder %c\n", - pipe_name(pch_transcoder)); + drm_err(&dev_priv->drm, "pch fifo underrun on pch transcoder %c\n", + pipe_name(pch_transcoder)); } static void cpt_set_fifo_underrun_reporting(struct drm_device *dev, @@ -246,8 +248,9 @@ static void cpt_set_fifo_underrun_reporting(struct drm_device *dev, if (old && intel_de_read(dev_priv, SERR_INT) & SERR_INT_TRANS_FIFO_UNDERRUN(pch_transcoder)) { - DRM_ERROR("uncleared pch fifo underrun on pch transcoder %c\n", - pipe_name(pch_transcoder)); + drm_err(&dev_priv->drm, + "uncleared pch fifo underrun on pch transcoder %c\n", + pipe_name(pch_transcoder)); } } } @@ -381,8 +384,8 @@ void intel_cpu_fifo_underrun_irq_handler(struct drm_i915_private *dev_priv, if (intel_set_cpu_fifo_underrun_reporting(dev_priv, pipe, false)) { trace_intel_cpu_fifo_underrun(dev_priv, pipe); - DRM_ERROR("CPU pipe %c FIFO underrun\n", - pipe_name(pipe)); + drm_err(&dev_priv->drm, "CPU pipe %c FIFO underrun\n", + pipe_name(pipe)); } intel_fbc_handle_fifo_underrun_irq(dev_priv); @@ -403,8 +406,8 @@ void intel_pch_fifo_underrun_irq_handler(struct drm_i915_private *dev_priv, if (intel_set_pch_fifo_underrun_reporting(dev_priv, pch_transcoder, false)) { trace_intel_pch_fifo_underrun(dev_priv, pch_transcoder); - DRM_ERROR("PCH transcoder %c FIFO underrun\n", - pipe_name(pch_transcoder)); + drm_err(&dev_priv->drm, "PCH transcoder %c FIFO underrun\n", + pipe_name(pch_transcoder)); } } diff --git a/drivers/gpu/drm/i915/display/intel_gmbus.c b/drivers/gpu/drm/i915/display/intel_gmbus.c index 4ef8a81ae0ad..1fd3a5a6296b 100644 --- a/drivers/gpu/drm/i915/display/intel_gmbus.c +++ b/drivers/gpu/drm/i915/display/intel_gmbus.c @@ -32,7 +32,6 @@ #include <linux/i2c.h> #include <drm/drm_hdcp.h> -#include <drm/i915_drm.h> #include "i915_drv.h" #include "intel_display_types.h" @@ -632,8 +631,9 @@ retry: * till then let it sleep. */ if (gmbus_wait_idle(dev_priv)) { - DRM_DEBUG_KMS("GMBUS [%s] timed out waiting for idle\n", - adapter->name); + drm_dbg_kms(&dev_priv->drm, + "GMBUS [%s] timed out waiting for idle\n", + adapter->name); ret = -ETIMEDOUT; } intel_de_write_fw(dev_priv, GMBUS0, 0); @@ -656,8 +656,9 @@ clear_err: */ ret = -ENXIO; if (gmbus_wait_idle(dev_priv)) { - DRM_DEBUG_KMS("GMBUS [%s] timed out after NAK\n", - adapter->name); + drm_dbg_kms(&dev_priv->drm, + "GMBUS [%s] timed out after NAK\n", + adapter->name); ret = -ETIMEDOUT; } @@ -669,9 +670,9 @@ clear_err: intel_de_write_fw(dev_priv, GMBUS1, 0); intel_de_write_fw(dev_priv, GMBUS0, 0); - DRM_DEBUG_KMS("GMBUS [%s] NAK for addr: %04x %c(%d)\n", - adapter->name, msgs[i].addr, - (msgs[i].flags & I2C_M_RD) ? 'r' : 'w', msgs[i].len); + drm_dbg_kms(&dev_priv->drm, "GMBUS [%s] NAK for addr: %04x %c(%d)\n", + adapter->name, msgs[i].addr, + (msgs[i].flags & I2C_M_RD) ? 'r' : 'w', msgs[i].len); /* * Passive adapters sometimes NAK the first probe. Retry the first @@ -680,16 +681,18 @@ clear_err: * drm_do_probe_ddc_edid, which bails out on the first -ENXIO. */ if (ret == -ENXIO && i == 0 && try++ == 0) { - DRM_DEBUG_KMS("GMBUS [%s] NAK on first message, retry\n", - adapter->name); + drm_dbg_kms(&dev_priv->drm, + "GMBUS [%s] NAK on first message, retry\n", + adapter->name); goto retry; } goto out; timeout: - DRM_DEBUG_KMS("GMBUS [%s] timed out, falling back to bit banging on pin %d\n", - bus->adapter.name, bus->reg0 & 0xff); + drm_dbg_kms(&dev_priv->drm, + "GMBUS [%s] timed out, falling back to bit banging on pin %d\n", + bus->adapter.name, bus->reg0 & 0xff); intel_de_write_fw(dev_priv, GMBUS0, 0); /* @@ -926,9 +929,10 @@ void intel_gmbus_force_bit(struct i2c_adapter *adapter, bool force_bit) mutex_lock(&dev_priv->gmbus_mutex); bus->force_bit += force_bit ? 1 : -1; - DRM_DEBUG_KMS("%sabling bit-banging on %s. force bit now %d\n", - force_bit ? "en" : "dis", adapter->name, - bus->force_bit); + drm_dbg_kms(&dev_priv->drm, + "%sabling bit-banging on %s. force bit now %d\n", + force_bit ? "en" : "dis", adapter->name, + bus->force_bit); mutex_unlock(&dev_priv->gmbus_mutex); } diff --git a/drivers/gpu/drm/i915/display/intel_hdcp.c b/drivers/gpu/drm/i915/display/intel_hdcp.c index 229b4e329864..ee0f27ea2810 100644 --- a/drivers/gpu/drm/i915/display/intel_hdcp.c +++ b/drivers/gpu/drm/i915/display/intel_hdcp.c @@ -43,6 +43,7 @@ static int intel_hdcp_read_valid_bksv(struct intel_digital_port *intel_dig_port, const struct intel_hdcp_shim *shim, u8 *bksv) { + struct drm_i915_private *i915 = to_i915(intel_dig_port->base.base.dev); int ret, i, tries = 2; /* HDCP spec states that we must retry the bksv if it is invalid */ @@ -54,7 +55,7 @@ int intel_hdcp_read_valid_bksv(struct intel_digital_port *intel_dig_port, break; } if (i == tries) { - DRM_DEBUG_KMS("Bksv is invalid\n"); + drm_dbg_kms(&i915->drm, "Bksv is invalid\n"); return -ENODEV; } @@ -485,8 +486,8 @@ int intel_hdcp_validate_v_prime(struct intel_connector *connector, return ret; sha_idx += sizeof(sha_text); } else { - DRM_DEBUG_KMS("Invalid number of leftovers %d\n", - sha_leftovers); + drm_dbg_kms(&dev_priv->drm, "Invalid number of leftovers %d\n", + sha_leftovers); return -EINVAL; } @@ -514,11 +515,11 @@ int intel_hdcp_validate_v_prime(struct intel_connector *connector, rep_ctl | HDCP_SHA1_COMPLETE_HASH); if (intel_de_wait_for_set(dev_priv, HDCP_REP_CTL, HDCP_SHA1_COMPLETE, 1)) { - DRM_ERROR("Timed out waiting for SHA1 complete\n"); + drm_err(&dev_priv->drm, "Timed out waiting for SHA1 complete\n"); return -ETIMEDOUT; } if (!(intel_de_read(dev_priv, HDCP_REP_CTL) & HDCP_SHA1_V_MATCH)) { - DRM_DEBUG_KMS("SHA-1 mismatch, HDCP failed\n"); + drm_dbg_kms(&dev_priv->drm, "SHA-1 mismatch, HDCP failed\n"); return -ENXIO; } @@ -537,7 +538,8 @@ int intel_hdcp_auth_downstream(struct intel_connector *connector) ret = intel_hdcp_poll_ksv_fifo(intel_dig_port, shim); if (ret) { - DRM_DEBUG_KMS("KSV list failed to become ready (%d)\n", ret); + drm_dbg_kms(&dev_priv->drm, + "KSV list failed to become ready (%d)\n", ret); return ret; } @@ -547,7 +549,7 @@ int intel_hdcp_auth_downstream(struct intel_connector *connector) if (DRM_HDCP_MAX_DEVICE_EXCEEDED(bstatus[0]) || DRM_HDCP_MAX_CASCADE_EXCEEDED(bstatus[1])) { - DRM_DEBUG_KMS("Max Topology Limit Exceeded\n"); + drm_dbg_kms(&dev_priv->drm, "Max Topology Limit Exceeded\n"); return -EPERM; } @@ -560,13 +562,14 @@ int intel_hdcp_auth_downstream(struct intel_connector *connector) */ num_downstream = DRM_HDCP_NUM_DOWNSTREAM(bstatus[0]); if (num_downstream == 0) { - DRM_DEBUG_KMS("Repeater with zero downstream devices\n"); + drm_dbg_kms(&dev_priv->drm, + "Repeater with zero downstream devices\n"); return -EINVAL; } ksv_fifo = kcalloc(DRM_HDCP_KSV_LEN, num_downstream, GFP_KERNEL); if (!ksv_fifo) { - DRM_DEBUG_KMS("Out of mem: ksv_fifo\n"); + drm_dbg_kms(&dev_priv->drm, "Out of mem: ksv_fifo\n"); return -ENOMEM; } @@ -576,7 +579,7 @@ int intel_hdcp_auth_downstream(struct intel_connector *connector) if (drm_hdcp_check_ksvs_revoked(&dev_priv->drm, ksv_fifo, num_downstream)) { - DRM_ERROR("Revoked Ksv(s) in ksv_fifo\n"); + drm_err(&dev_priv->drm, "Revoked Ksv(s) in ksv_fifo\n"); ret = -EPERM; goto err; } @@ -594,12 +597,13 @@ int intel_hdcp_auth_downstream(struct intel_connector *connector) } if (i == tries) { - DRM_DEBUG_KMS("V Prime validation failed.(%d)\n", ret); + drm_dbg_kms(&dev_priv->drm, + "V Prime validation failed.(%d)\n", ret); goto err; } - DRM_DEBUG_KMS("HDCP is enabled (%d downstream devices)\n", - num_downstream); + drm_dbg_kms(&dev_priv->drm, "HDCP is enabled (%d downstream devices)\n", + num_downstream); ret = 0; err: kfree(ksv_fifo); @@ -642,7 +646,8 @@ static int intel_hdcp_auth(struct intel_connector *connector) if (ret) return ret; if (!hdcp_capable) { - DRM_DEBUG_KMS("Panel is not HDCP capable\n"); + drm_dbg_kms(&dev_priv->drm, + "Panel is not HDCP capable\n"); return -EINVAL; } } @@ -659,7 +664,7 @@ static int intel_hdcp_auth(struct intel_connector *connector) if (intel_de_wait_for_set(dev_priv, HDCP_STATUS(dev_priv, cpu_transcoder, port), HDCP_STATUS_AN_READY, 1)) { - DRM_ERROR("Timed out waiting for An\n"); + drm_err(&dev_priv->drm, "Timed out waiting for An\n"); return -ETIMEDOUT; } @@ -680,7 +685,7 @@ static int intel_hdcp_auth(struct intel_connector *connector) return ret; if (drm_hdcp_check_ksvs_revoked(&dev_priv->drm, bksv.shim, 1)) { - DRM_ERROR("BKSV is revoked\n"); + drm_err(&dev_priv->drm, "BKSV is revoked\n"); return -EPERM; } @@ -706,7 +711,7 @@ static int intel_hdcp_auth(struct intel_connector *connector) /* Wait for R0 ready */ if (wait_for(intel_de_read(dev_priv, HDCP_STATUS(dev_priv, cpu_transcoder, port)) & (HDCP_STATUS_R0_READY | HDCP_STATUS_ENC), 1)) { - DRM_ERROR("Timed out waiting for R0 ready\n"); + drm_err(&dev_priv->drm, "Timed out waiting for R0 ready\n"); return -ETIMEDOUT; } @@ -743,8 +748,10 @@ static int intel_hdcp_auth(struct intel_connector *connector) } if (i == tries) { - DRM_DEBUG_KMS("Timed out waiting for Ri prime match (%x)\n", - intel_de_read(dev_priv, HDCP_STATUS(dev_priv, cpu_transcoder, port))); + drm_dbg_kms(&dev_priv->drm, + "Timed out waiting for Ri prime match (%x)\n", + intel_de_read(dev_priv, HDCP_STATUS(dev_priv, + cpu_transcoder, port))); return -ETIMEDOUT; } @@ -753,7 +760,7 @@ static int intel_hdcp_auth(struct intel_connector *connector) HDCP_STATUS(dev_priv, cpu_transcoder, port), HDCP_STATUS_ENC, ENCRYPT_STATUS_CHANGE_TIMEOUT_MS)) { - DRM_ERROR("Timed out waiting for encryption\n"); + drm_err(&dev_priv->drm, "Timed out waiting for encryption\n"); return -ETIMEDOUT; } @@ -765,7 +772,7 @@ static int intel_hdcp_auth(struct intel_connector *connector) if (repeater_present) return intel_hdcp_auth_downstream(connector); - DRM_DEBUG_KMS("HDCP is enabled (no repeater present)\n"); + drm_dbg_kms(&dev_priv->drm, "HDCP is enabled (no repeater present)\n"); return 0; } @@ -1271,7 +1278,7 @@ static int hdcp2_authentication_key_exchange(struct intel_connector *connector) return ret; if (msgs.send_cert.rx_caps[0] != HDCP_2_2_RX_CAPS_VERSION_VAL) { - DRM_DEBUG_KMS("cert.rx_caps dont claim HDCP2.2\n"); + drm_dbg_kms(&dev_priv->drm, "cert.rx_caps dont claim HDCP2.2\n"); return -EINVAL; } @@ -1280,7 +1287,7 @@ static int hdcp2_authentication_key_exchange(struct intel_connector *connector) if (drm_hdcp_check_ksvs_revoked(&dev_priv->drm, msgs.send_cert.cert_rx.receiver_id, 1)) { - DRM_ERROR("Receiver ID is revoked\n"); + drm_err(&dev_priv->drm, "Receiver ID is revoked\n"); return -EPERM; } @@ -1455,7 +1462,7 @@ int hdcp2_authenticate_repeater_topology(struct intel_connector *connector) if (HDCP_2_2_MAX_CASCADE_EXCEEDED(rx_info[1]) || HDCP_2_2_MAX_DEVS_EXCEEDED(rx_info[1])) { - DRM_DEBUG_KMS("Topology Max Size Exceeded\n"); + drm_dbg_kms(&dev_priv->drm, "Topology Max Size Exceeded\n"); return -EINVAL; } @@ -1463,9 +1470,15 @@ int hdcp2_authenticate_repeater_topology(struct intel_connector *connector) seq_num_v = drm_hdcp_be24_to_cpu((const u8 *)msgs.recvid_list.seq_num_v); + if (!hdcp->hdcp2_encrypted && seq_num_v) { + drm_dbg_kms(&dev_priv->drm, + "Non zero Seq_num_v at first RecvId_List msg\n"); + return -EINVAL; + } + if (seq_num_v < hdcp->seq_num_v) { /* Roll over of the seq_num_v from repeater. Reauthenticate. */ - DRM_DEBUG_KMS("Seq_num_v roll over.\n"); + drm_dbg_kms(&dev_priv->drm, "Seq_num_v roll over.\n"); return -EINVAL; } @@ -1474,7 +1487,7 @@ int hdcp2_authenticate_repeater_topology(struct intel_connector *connector) if (drm_hdcp_check_ksvs_revoked(&dev_priv->drm, msgs.recvid_list.receiver_ids, device_cnt)) { - DRM_ERROR("Revoked receiver ID(s) is in list\n"); + drm_err(&dev_priv->drm, "Revoked receiver ID(s) is in list\n"); return -EPERM; } @@ -1507,25 +1520,27 @@ static int hdcp2_authenticate_repeater(struct intel_connector *connector) static int hdcp2_authenticate_sink(struct intel_connector *connector) { struct intel_digital_port *intel_dig_port = intel_attached_dig_port(connector); + struct drm_i915_private *i915 = to_i915(connector->base.dev); struct intel_hdcp *hdcp = &connector->hdcp; const struct intel_hdcp_shim *shim = hdcp->shim; int ret; ret = hdcp2_authentication_key_exchange(connector); if (ret < 0) { - DRM_DEBUG_KMS("AKE Failed. Err : %d\n", ret); + drm_dbg_kms(&i915->drm, "AKE Failed. Err : %d\n", ret); return ret; } ret = hdcp2_locality_check(connector); if (ret < 0) { - DRM_DEBUG_KMS("Locality Check failed. Err : %d\n", ret); + drm_dbg_kms(&i915->drm, + "Locality Check failed. Err : %d\n", ret); return ret; } ret = hdcp2_session_key_exchange(connector); if (ret < 0) { - DRM_DEBUG_KMS("SKE Failed. Err : %d\n", ret); + drm_dbg_kms(&i915->drm, "SKE Failed. Err : %d\n", ret); return ret; } @@ -1540,7 +1555,8 @@ static int hdcp2_authenticate_sink(struct intel_connector *connector) if (hdcp->is_repeater) { ret = hdcp2_authenticate_repeater(connector); if (ret < 0) { - DRM_DEBUG_KMS("Repeater Auth Failed. Err: %d\n", ret); + drm_dbg_kms(&i915->drm, + "Repeater Auth Failed. Err: %d\n", ret); return ret; } } @@ -1630,6 +1646,7 @@ static int hdcp2_disable_encryption(struct intel_connector *connector) static int hdcp2_authenticate_and_encrypt(struct intel_connector *connector) { + struct drm_i915_private *i915 = to_i915(connector->base.dev); int ret, i, tries = 3; for (i = 0; i < tries; i++) { @@ -1638,10 +1655,10 @@ static int hdcp2_authenticate_and_encrypt(struct intel_connector *connector) break; /* Clearing the mei hdcp session */ - DRM_DEBUG_KMS("HDCP2.2 Auth %d of %d Failed.(%d)\n", - i + 1, tries, ret); + drm_dbg_kms(&i915->drm, "HDCP2.2 Auth %d of %d Failed.(%d)\n", + i + 1, tries, ret); if (hdcp2_deauthenticate_port(connector) < 0) - DRM_DEBUG_KMS("Port deauth failed.\n"); + drm_dbg_kms(&i915->drm, "Port deauth failed.\n"); } if (i != tries) { @@ -1652,9 +1669,10 @@ static int hdcp2_authenticate_and_encrypt(struct intel_connector *connector) msleep(HDCP_2_2_DELAY_BEFORE_ENCRYPTION_EN); ret = hdcp2_enable_encryption(connector); if (ret < 0) { - DRM_DEBUG_KMS("Encryption Enable Failed.(%d)\n", ret); + drm_dbg_kms(&i915->drm, + "Encryption Enable Failed.(%d)\n", ret); if (hdcp2_deauthenticate_port(connector) < 0) - DRM_DEBUG_KMS("Port deauth failed.\n"); + drm_dbg_kms(&i915->drm, "Port deauth failed.\n"); } } @@ -1663,23 +1681,24 @@ static int hdcp2_authenticate_and_encrypt(struct intel_connector *connector) static int _intel_hdcp2_enable(struct intel_connector *connector) { + struct drm_i915_private *i915 = to_i915(connector->base.dev); struct intel_hdcp *hdcp = &connector->hdcp; int ret; - DRM_DEBUG_KMS("[%s:%d] HDCP2.2 is being enabled. Type: %d\n", - connector->base.name, connector->base.base.id, - hdcp->content_type); + drm_dbg_kms(&i915->drm, "[%s:%d] HDCP2.2 is being enabled. Type: %d\n", + connector->base.name, connector->base.base.id, + hdcp->content_type); ret = hdcp2_authenticate_and_encrypt(connector); if (ret) { - DRM_DEBUG_KMS("HDCP2 Type%d Enabling Failed. (%d)\n", - hdcp->content_type, ret); + drm_dbg_kms(&i915->drm, "HDCP2 Type%d Enabling Failed. (%d)\n", + hdcp->content_type, ret); return ret; } - DRM_DEBUG_KMS("[%s:%d] HDCP2.2 is enabled. Type %d\n", - connector->base.name, connector->base.base.id, - hdcp->content_type); + drm_dbg_kms(&i915->drm, "[%s:%d] HDCP2.2 is enabled. Type %d\n", + connector->base.name, connector->base.base.id, + hdcp->content_type); hdcp->hdcp2_encrypted = true; return 0; @@ -1687,15 +1706,16 @@ static int _intel_hdcp2_enable(struct intel_connector *connector) static int _intel_hdcp2_disable(struct intel_connector *connector) { + struct drm_i915_private *i915 = to_i915(connector->base.dev); int ret; - DRM_DEBUG_KMS("[%s:%d] HDCP2.2 is being Disabled\n", - connector->base.name, connector->base.base.id); + drm_dbg_kms(&i915->drm, "[%s:%d] HDCP2.2 is being Disabled\n", + connector->base.name, connector->base.base.id); ret = hdcp2_disable_encryption(connector); if (hdcp2_deauthenticate_port(connector) < 0) - DRM_DEBUG_KMS("Port deauth failed.\n"); + drm_dbg_kms(&i915->drm, "Port deauth failed.\n"); connector->hdcp.hdcp2_encrypted = false; @@ -1938,12 +1958,13 @@ void intel_hdcp_component_init(struct drm_i915_private *dev_priv) static void intel_hdcp2_init(struct intel_connector *connector, const struct intel_hdcp_shim *shim) { + struct drm_i915_private *i915 = to_i915(connector->base.dev); struct intel_hdcp *hdcp = &connector->hdcp; int ret; ret = initialize_hdcp_port_data(connector, shim); if (ret) { - DRM_DEBUG_KMS("Mei hdcp data init failed\n"); + drm_dbg_kms(&i915->drm, "Mei hdcp data init failed\n"); return; } diff --git a/drivers/gpu/drm/i915/display/intel_hdcp.h b/drivers/gpu/drm/i915/display/intel_hdcp.h index 1b2eacaf8949..7c12ad609b1f 100644 --- a/drivers/gpu/drm/i915/display/intel_hdcp.h +++ b/drivers/gpu/drm/i915/display/intel_hdcp.h @@ -8,8 +8,6 @@ #include <linux/types.h> -#include <drm/i915_drm.h> - struct drm_connector; struct drm_connector_state; struct drm_i915_private; diff --git a/drivers/gpu/drm/i915/display/intel_hdmi.c b/drivers/gpu/drm/i915/display/intel_hdmi.c index bdbb5ce3fa81..39930232b253 100644 --- a/drivers/gpu/drm/i915/display/intel_hdmi.c +++ b/drivers/gpu/drm/i915/display/intel_hdmi.c @@ -36,7 +36,6 @@ #include <drm/drm_edid.h> #include <drm/drm_hdcp.h> #include <drm/drm_scdc_helper.h> -#include <drm/i915_drm.h> #include <drm/intel_lpe_audio.h> #include "i915_debugfs.h" @@ -2276,14 +2275,9 @@ static bool hdmi_deep_color_possible(const struct intel_crtc_state *crtc_state, } } - /* Display WA #1139: glk */ - if (bpc == 12 && IS_GLK_REVID(dev_priv, 0, GLK_REVID_A1) && - adjusted_mode->htotal > 5460) - return false; - - /* Display Wa_1405510057:icl */ + /* Display Wa_1405510057:icl,ehl */ if (crtc_state->output_format == INTEL_OUTPUT_FORMAT_YCBCR420 && - bpc == 10 && INTEL_GEN(dev_priv) >= 11 && + bpc == 10 && IS_GEN(dev_priv, 11) && (adjusted_mode->crtc_hblank_end - adjusted_mode->crtc_hblank_start) % 8 == 2) return false; diff --git a/drivers/gpu/drm/i915/display/intel_hdmi.h b/drivers/gpu/drm/i915/display/intel_hdmi.h index d3659d0b408b..8ff1f76a63df 100644 --- a/drivers/gpu/drm/i915/display/intel_hdmi.h +++ b/drivers/gpu/drm/i915/display/intel_hdmi.h @@ -9,8 +9,6 @@ #include <linux/hdmi.h> #include <linux/types.h> -#include <drm/i915_drm.h> - #include "i915_reg.h" struct drm_connector; diff --git a/drivers/gpu/drm/i915/display/intel_hotplug.c b/drivers/gpu/drm/i915/display/intel_hotplug.c index 8af0ae61e1bb..a091442efba4 100644 --- a/drivers/gpu/drm/i915/display/intel_hotplug.c +++ b/drivers/gpu/drm/i915/display/intel_hotplug.c @@ -23,8 +23,6 @@ #include <linux/kernel.h> -#include <drm/i915_drm.h> - #include "i915_drv.h" #include "intel_display_types.h" #include "intel_hotplug.h" @@ -89,29 +87,16 @@ enum hpd_pin intel_hpd_pin_default(struct drm_i915_private *dev_priv, enum port port) { - switch (port) { - case PORT_A: - return HPD_PORT_A; - case PORT_B: - return HPD_PORT_B; - case PORT_C: - return HPD_PORT_C; - case PORT_D: - return HPD_PORT_D; - case PORT_E: - return HPD_PORT_E; - case PORT_F: - if (IS_CNL_WITH_PORT_F(dev_priv)) - return HPD_PORT_E; - return HPD_PORT_F; - case PORT_G: - return HPD_PORT_G; - case PORT_H: - return HPD_PORT_H; - case PORT_I: - return HPD_PORT_I; + enum phy phy = intel_port_to_phy(dev_priv, port); + + switch (phy) { + case PHY_F: + return IS_CNL_WITH_PORT_F(dev_priv) ? HPD_PORT_E : HPD_PORT_F; + case PHY_A ... PHY_E: + case PHY_G ... PHY_I: + return HPD_PORT_A + phy - PHY_A; default: - MISSING_CASE(port); + MISSING_CASE(phy); return HPD_NONE; } } @@ -185,10 +170,13 @@ static bool intel_hpd_irq_storm_detect(struct drm_i915_private *dev_priv, hpd->stats[pin].count += increment; if (hpd->stats[pin].count > threshold) { hpd->stats[pin].state = HPD_MARK_DISABLED; - DRM_DEBUG_KMS("HPD interrupt storm detected on PIN %d\n", pin); + drm_dbg_kms(&dev_priv->drm, + "HPD interrupt storm detected on PIN %d\n", pin); storm = true; } else { - DRM_DEBUG_KMS("Received HPD interrupt on PIN %d - cnt: %d\n", pin, + drm_dbg_kms(&dev_priv->drm, + "Received HPD interrupt on PIN %d - cnt: %d\n", + pin, hpd->stats[pin].count); } @@ -217,7 +205,8 @@ intel_hpd_irq_storm_switch_to_polling(struct drm_i915_private *dev_priv) dev_priv->hotplug.stats[pin].state != HPD_MARK_DISABLED) continue; - DRM_INFO("HPD interrupt storm detected on connector %s: " + drm_info(&dev_priv->drm, + "HPD interrupt storm detected on connector %s: " "switching from hotplug detection to polling\n", connector->base.name); @@ -259,8 +248,9 @@ static void intel_hpd_irq_storm_reenable_work(struct work_struct *work) continue; if (connector->base.polled != connector->polled) - DRM_DEBUG_DRIVER("Reenabling HPD on connector %s\n", - connector->base.name); + drm_dbg(&dev_priv->drm, + "Reenabling HPD on connector %s\n", + connector->base.name); connector->base.polled = connector->polled; } drm_connector_list_iter_end(&conn_iter); @@ -295,11 +285,12 @@ intel_encoder_hotplug(struct intel_encoder *encoder, if (old_status == connector->base.status) return INTEL_HOTPLUG_UNCHANGED; - DRM_DEBUG_KMS("[CONNECTOR:%d:%s] status updated from %s to %s\n", - connector->base.base.id, - connector->base.name, - drm_get_connector_status_name(old_status), - drm_get_connector_status_name(connector->base.status)); + drm_dbg_kms(&to_i915(dev)->drm, + "[CONNECTOR:%d:%s] status updated from %s to %s\n", + connector->base.base.id, + connector->base.name, + drm_get_connector_status_name(old_status), + drm_get_connector_status_name(connector->base.status)); return INTEL_HOTPLUG_CHANGED; } @@ -373,7 +364,7 @@ static void i915_hotplug_work_func(struct work_struct *work) u32 hpd_retry_bits; mutex_lock(&dev->mode_config.mutex); - DRM_DEBUG_KMS("running encoder hotplug functions\n"); + drm_dbg_kms(&dev_priv->drm, "running encoder hotplug functions\n"); spin_lock_irq(&dev_priv->irq_lock); @@ -401,8 +392,9 @@ static void i915_hotplug_work_func(struct work_struct *work) struct intel_encoder *encoder = intel_attached_encoder(connector); - DRM_DEBUG_KMS("Connector %s (pin %i) received hotplug event.\n", - connector->base.name, pin); + drm_dbg_kms(&dev_priv->drm, + "Connector %s (pin %i) received hotplug event.\n", + connector->base.name, pin); switch (encoder->hotplug(encoder, connector, hpd_event_bits & hpd_bit)) { @@ -487,9 +479,10 @@ void intel_hpd_irq_handler(struct drm_i915_private *dev_priv, long_hpd = long_mask & BIT(pin); - DRM_DEBUG_DRIVER("digital hpd on [ENCODER:%d:%s] - %s\n", - encoder->base.base.id, encoder->base.name, - long_hpd ? "long" : "short"); + drm_dbg(&dev_priv->drm, + "digital hpd on [ENCODER:%d:%s] - %s\n", + encoder->base.base.id, encoder->base.name, + long_hpd ? "long" : "short"); queue_dig = true; if (long_hpd) { diff --git a/drivers/gpu/drm/i915/display/intel_hotplug.h b/drivers/gpu/drm/i915/display/intel_hotplug.h index 087b5f57b321..1e6b4fda2900 100644 --- a/drivers/gpu/drm/i915/display/intel_hotplug.h +++ b/drivers/gpu/drm/i915/display/intel_hotplug.h @@ -8,8 +8,6 @@ #include <linux/types.h> -#include <drm/i915_drm.h> - struct drm_i915_private; struct intel_connector; struct intel_encoder; diff --git a/drivers/gpu/drm/i915/display/intel_lpe_audio.c b/drivers/gpu/drm/i915/display/intel_lpe_audio.c index 516e7179a5a4..ad5cc13037ae 100644 --- a/drivers/gpu/drm/i915/display/intel_lpe_audio.c +++ b/drivers/gpu/drm/i915/display/intel_lpe_audio.c @@ -127,7 +127,8 @@ lpe_audio_platdev_create(struct drm_i915_private *dev_priv) kfree(pdata); if (IS_ERR(platdev)) { - DRM_ERROR("Failed to allocate LPE audio platform device\n"); + drm_err(&dev_priv->drm, + "Failed to allocate LPE audio platform device\n"); return platdev; } @@ -190,7 +191,8 @@ static bool lpe_audio_detect(struct drm_i915_private *dev_priv) }; if (!pci_dev_present(atom_hdaudio_ids)) { - DRM_INFO("HDaudio controller not detected, using LPE audio instead\n"); + drm_info(&dev_priv->drm, + "HDaudio controller not detected, using LPE audio instead\n"); lpe_present = true; } } @@ -203,18 +205,19 @@ static int lpe_audio_setup(struct drm_i915_private *dev_priv) dev_priv->lpe_audio.irq = irq_alloc_desc(0); if (dev_priv->lpe_audio.irq < 0) { - DRM_ERROR("Failed to allocate IRQ desc: %d\n", + drm_err(&dev_priv->drm, "Failed to allocate IRQ desc: %d\n", dev_priv->lpe_audio.irq); ret = dev_priv->lpe_audio.irq; goto err; } - DRM_DEBUG("irq = %d\n", dev_priv->lpe_audio.irq); + drm_dbg(&dev_priv->drm, "irq = %d\n", dev_priv->lpe_audio.irq); ret = lpe_audio_irq_init(dev_priv); if (ret) { - DRM_ERROR("Failed to initialize irqchip for lpe audio: %d\n", + drm_err(&dev_priv->drm, + "Failed to initialize irqchip for lpe audio: %d\n", ret); goto err_free_irq; } @@ -223,7 +226,8 @@ static int lpe_audio_setup(struct drm_i915_private *dev_priv) if (IS_ERR(dev_priv->lpe_audio.platdev)) { ret = PTR_ERR(dev_priv->lpe_audio.platdev); - DRM_ERROR("Failed to create lpe audio platform device: %d\n", + drm_err(&dev_priv->drm, + "Failed to create lpe audio platform device: %d\n", ret); goto err_free_irq; } @@ -259,8 +263,8 @@ void intel_lpe_audio_irq_handler(struct drm_i915_private *dev_priv) ret = generic_handle_irq(dev_priv->lpe_audio.irq); if (ret) - DRM_ERROR_RATELIMITED("error handling LPE audio irq: %d\n", - ret); + drm_err_ratelimited(&dev_priv->drm, + "error handling LPE audio irq: %d\n", ret); } /** @@ -278,7 +282,8 @@ int intel_lpe_audio_init(struct drm_i915_private *dev_priv) if (lpe_audio_detect(dev_priv)) { ret = lpe_audio_setup(dev_priv); if (ret < 0) - DRM_ERROR("failed to setup LPE Audio bridge\n"); + drm_err(&dev_priv->drm, + "failed to setup LPE Audio bridge\n"); } return ret; } diff --git a/drivers/gpu/drm/i915/display/intel_lvds.c b/drivers/gpu/drm/i915/display/intel_lvds.c index b7ad0b534790..9a067effcfa0 100644 --- a/drivers/gpu/drm/i915/display/intel_lvds.c +++ b/drivers/gpu/drm/i915/display/intel_lvds.c @@ -37,7 +37,6 @@ #include <drm/drm_atomic_helper.h> #include <drm/drm_crtc.h> #include <drm/drm_edid.h> -#include <drm/i915_drm.h> #include "i915_drv.h" #include "intel_atomic.h" @@ -182,8 +181,9 @@ static void intel_lvds_pps_get_hw_state(struct drm_i915_private *dev_priv, if (INTEL_GEN(dev_priv) <= 4 && pps->t1_t2 == 0 && pps->t5 == 0 && pps->t3 == 0 && pps->tx == 0) { - DRM_DEBUG_KMS("Panel power timings uninitialized, " - "setting defaults\n"); + drm_dbg_kms(&dev_priv->drm, + "Panel power timings uninitialized, " + "setting defaults\n"); /* Set T2 to 40ms and T5 to 200ms in 100 usec units */ pps->t1_t2 = 40 * 10; pps->t5 = 200 * 10; @@ -192,10 +192,10 @@ static void intel_lvds_pps_get_hw_state(struct drm_i915_private *dev_priv, pps->tx = 200 * 10; } - DRM_DEBUG_DRIVER("LVDS PPS:t1+t2 %d t3 %d t4 %d t5 %d tx %d " - "divider %d port %d powerdown_on_reset %d\n", - pps->t1_t2, pps->t3, pps->t4, pps->t5, pps->tx, - pps->divider, pps->port, pps->powerdown_on_reset); + drm_dbg(&dev_priv->drm, "LVDS PPS:t1+t2 %d t3 %d t4 %d t5 %d tx %d " + "divider %d port %d powerdown_on_reset %d\n", + pps->t1_t2, pps->t3, pps->t4, pps->t5, pps->tx, + pps->divider, pps->port, pps->powerdown_on_reset); } static void intel_lvds_pps_init_hw(struct drm_i915_private *dev_priv, @@ -317,7 +317,8 @@ static void intel_enable_lvds(struct intel_encoder *encoder, intel_de_posting_read(dev_priv, lvds_encoder->reg); if (intel_de_wait_for_set(dev_priv, PP_STATUS(0), PP_ON, 5000)) - DRM_ERROR("timed out waiting for panel to power on\n"); + drm_err(&dev_priv->drm, + "timed out waiting for panel to power on\n"); intel_panel_enable_backlight(pipe_config, conn_state); } @@ -332,7 +333,8 @@ static void intel_disable_lvds(struct intel_encoder *encoder, intel_de_write(dev_priv, PP_CONTROL(0), intel_de_read(dev_priv, PP_CONTROL(0)) & ~PANEL_POWER_ON); if (intel_de_wait_for_clear(dev_priv, PP_STATUS(0), PP_ON, 1000)) - DRM_ERROR("timed out waiting for panel to power off\n"); + drm_err(&dev_priv->drm, + "timed out waiting for panel to power off\n"); intel_de_write(dev_priv, lvds_encoder->reg, intel_de_read(dev_priv, lvds_encoder->reg) & ~LVDS_PORT_EN); @@ -398,7 +400,7 @@ static int intel_lvds_compute_config(struct intel_encoder *intel_encoder, /* Should never happen!! */ if (INTEL_GEN(dev_priv) < 4 && intel_crtc->pipe == 0) { - DRM_ERROR("Can't support LVDS on pipe A\n"); + drm_err(&dev_priv->drm, "Can't support LVDS on pipe A\n"); return -EINVAL; } @@ -408,8 +410,9 @@ static int intel_lvds_compute_config(struct intel_encoder *intel_encoder, lvds_bpp = 6*3; if (lvds_bpp != pipe_config->pipe_bpp && !pipe_config->bw_constrained) { - DRM_DEBUG_KMS("forcing display bpp (was %d) to LVDS (%d)\n", - pipe_config->pipe_bpp, lvds_bpp); + drm_dbg_kms(&dev_priv->drm, + "forcing display bpp (was %d) to LVDS (%d)\n", + pipe_config->pipe_bpp, lvds_bpp); pipe_config->pipe_bpp = lvds_bpp; } @@ -833,7 +836,8 @@ void intel_lvds_init(struct drm_i915_private *dev_priv) } if (!dev_priv->vbt.int_lvds_support) { - DRM_DEBUG_KMS("Internal LVDS support disabled by VBT\n"); + drm_dbg_kms(&dev_priv->drm, + "Internal LVDS support disabled by VBT\n"); return; } @@ -852,10 +856,12 @@ void intel_lvds_init(struct drm_i915_private *dev_priv) pin = GMBUS_PIN_PANEL; if (!intel_bios_is_lvds_present(dev_priv, &pin)) { if ((lvds & LVDS_PORT_EN) == 0) { - DRM_DEBUG_KMS("LVDS is not present in VBT\n"); + drm_dbg_kms(&dev_priv->drm, + "LVDS is not present in VBT\n"); return; } - DRM_DEBUG_KMS("LVDS is not present in VBT, but enabled anyway\n"); + drm_dbg_kms(&dev_priv->drm, + "LVDS is not present in VBT, but enabled anyway\n"); } lvds_encoder = kzalloc(sizeof(*lvds_encoder), GFP_KERNEL); @@ -969,7 +975,7 @@ void intel_lvds_init(struct drm_i915_private *dev_priv) */ fixed_mode = intel_encoder_current_mode(intel_encoder); if (fixed_mode) { - DRM_DEBUG_KMS("using current (BIOS) mode: "); + drm_dbg_kms(&dev_priv->drm, "using current (BIOS) mode: "); drm_mode_debug_printmodeline(fixed_mode); fixed_mode->type |= DRM_MODE_TYPE_PREFERRED; } @@ -985,8 +991,8 @@ out: intel_panel_setup_backlight(connector, INVALID_PIPE); lvds_encoder->is_dual_link = compute_is_dual_link_lvds(lvds_encoder); - DRM_DEBUG_KMS("detected %s-link lvds configuration\n", - lvds_encoder->is_dual_link ? "dual" : "single"); + drm_dbg_kms(&dev_priv->drm, "detected %s-link lvds configuration\n", + lvds_encoder->is_dual_link ? "dual" : "single"); lvds_encoder->a3_power = lvds & LVDS_A3_POWER_MASK; @@ -995,7 +1001,7 @@ out: failed: mutex_unlock(&dev->mode_config.mutex); - DRM_DEBUG_KMS("No LVDS modes found, disabling.\n"); + drm_dbg_kms(&dev_priv->drm, "No LVDS modes found, disabling.\n"); drm_connector_cleanup(connector); drm_encoder_cleanup(encoder); kfree(lvds_encoder); diff --git a/drivers/gpu/drm/i915/display/intel_opregion.c b/drivers/gpu/drm/i915/display/intel_opregion.c index dfd78fccd456..cc6b00959586 100644 --- a/drivers/gpu/drm/i915/display/intel_opregion.c +++ b/drivers/gpu/drm/i915/display/intel_opregion.c @@ -30,8 +30,6 @@ #include <linux/firmware.h> #include <acpi/video.h> -#include <drm/i915_drm.h> - #include "display/intel_panel.h" #include "i915_drv.h" diff --git a/drivers/gpu/drm/i915/display/intel_overlay.c b/drivers/gpu/drm/i915/display/intel_overlay.c index 5f1207dec10e..481187223101 100644 --- a/drivers/gpu/drm/i915/display/intel_overlay.c +++ b/drivers/gpu/drm/i915/display/intel_overlay.c @@ -27,7 +27,6 @@ */ #include <drm/drm_fourcc.h> -#include <drm/i915_drm.h> #include "gem/i915_gem_pm.h" #include "gt/intel_ring.h" @@ -324,7 +323,7 @@ static int intel_overlay_continue(struct intel_overlay *overlay, /* check for underruns */ tmp = intel_de_read(dev_priv, DOVSTA); if (tmp & (1 << 17)) - DRM_DEBUG("overlay underrun, DOVSTA: %x\n", tmp); + drm_dbg(&dev_priv->drm, "overlay underrun, DOVSTA: %x\n", tmp); rq = alloc_request(overlay, NULL); if (IS_ERR(rq)) @@ -1069,7 +1068,7 @@ int intel_overlay_put_image_ioctl(struct drm_device *dev, void *data, overlay = dev_priv->overlay; if (!overlay) { - DRM_DEBUG("userspace bug: no overlay\n"); + drm_dbg(&dev_priv->drm, "userspace bug: no overlay\n"); return -ENODEV; } @@ -1093,7 +1092,8 @@ int intel_overlay_put_image_ioctl(struct drm_device *dev, void *data, drm_modeset_lock_all(dev); if (i915_gem_object_is_tiled(new_bo)) { - DRM_DEBUG_KMS("buffer used for overlay image can not be tiled\n"); + drm_dbg_kms(&dev_priv->drm, + "buffer used for overlay image can not be tiled\n"); ret = -EINVAL; goto out_unlock; } @@ -1228,7 +1228,7 @@ int intel_overlay_attrs_ioctl(struct drm_device *dev, void *data, overlay = dev_priv->overlay; if (!overlay) { - DRM_DEBUG("userspace bug: no overlay\n"); + drm_dbg(&dev_priv->drm, "userspace bug: no overlay\n"); return -ENODEV; } @@ -1372,7 +1372,7 @@ void intel_overlay_setup(struct drm_i915_private *dev_priv) update_reg_attrs(overlay, overlay->regs); dev_priv->overlay = overlay; - DRM_INFO("Initialized overlay support.\n"); + drm_info(&dev_priv->drm, "Initialized overlay support.\n"); return; out_free: diff --git a/drivers/gpu/drm/i915/display/intel_panel.c b/drivers/gpu/drm/i915/display/intel_panel.c index 5aead622019c..276f43870802 100644 --- a/drivers/gpu/drm/i915/display/intel_panel.c +++ b/drivers/gpu/drm/i915/display/intel_panel.c @@ -1882,6 +1882,7 @@ static int pwm_setup_backlight(struct intel_connector *connector, struct drm_i915_private *dev_priv = to_i915(dev); struct intel_panel *panel = &connector->panel; const char *desc; + u32 level, ns; int retval; /* Get the right PWM chip for DSI backlight according to VBT */ @@ -1906,8 +1907,12 @@ static int pwm_setup_backlight(struct intel_connector *connector, */ pwm_apply_args(panel->backlight.pwm); - retval = pwm_config(panel->backlight.pwm, CRC_PMIC_PWM_PERIOD_NS, - CRC_PMIC_PWM_PERIOD_NS); + panel->backlight.min = 0; /* 0% */ + panel->backlight.max = 100; /* 100% */ + level = intel_panel_compute_brightness(connector, 100); + ns = DIV_ROUND_UP(level * CRC_PMIC_PWM_PERIOD_NS, 100); + + retval = pwm_config(panel->backlight.pwm, ns, CRC_PMIC_PWM_PERIOD_NS); if (retval < 0) { drm_err(&dev_priv->drm, "Failed to configure the pwm chip\n"); pwm_put(panel->backlight.pwm); @@ -1915,11 +1920,10 @@ static int pwm_setup_backlight(struct intel_connector *connector, return retval; } - panel->backlight.min = 0; /* 0% */ - panel->backlight.max = 100; /* 100% */ - panel->backlight.level = DIV_ROUND_UP( - pwm_get_duty_cycle(panel->backlight.pwm) * 100, - CRC_PMIC_PWM_PERIOD_NS); + level = DIV_ROUND_UP(pwm_get_duty_cycle(panel->backlight.pwm) * 100, + CRC_PMIC_PWM_PERIOD_NS); + panel->backlight.level = + intel_panel_compute_brightness(connector, level); panel->backlight.enabled = panel->backlight.level != 0; drm_info(&dev_priv->drm, "Using %s PWM for LCD backlight control\n", diff --git a/drivers/gpu/drm/i915/display/intel_pipe_crc.c b/drivers/gpu/drm/i915/display/intel_pipe_crc.c index 59d7e3cb3445..a9a5df2fee4d 100644 --- a/drivers/gpu/drm/i915/display/intel_pipe_crc.c +++ b/drivers/gpu/drm/i915/display/intel_pipe_crc.c @@ -441,15 +441,11 @@ display_crc_ctl_parse_source(const char *buf, enum intel_pipe_crc_source *s) return 0; } -void intel_display_crc_init(struct drm_i915_private *dev_priv) +void intel_crtc_crc_init(struct intel_crtc *crtc) { - enum pipe pipe; + struct intel_pipe_crc *pipe_crc = &crtc->pipe_crc; - for_each_pipe(dev_priv, pipe) { - struct intel_pipe_crc *pipe_crc = &dev_priv->pipe_crc[pipe]; - - spin_lock_init(&pipe_crc->lock); - } + spin_lock_init(&pipe_crc->lock); } static int i8xx_crc_source_valid(struct drm_i915_private *dev_priv, @@ -587,7 +583,8 @@ int intel_crtc_verify_crc_source(struct drm_crtc *crtc, const char *source_name, int intel_crtc_set_crc_source(struct drm_crtc *crtc, const char *source_name) { struct drm_i915_private *dev_priv = to_i915(crtc->dev); - struct intel_pipe_crc *pipe_crc = &dev_priv->pipe_crc[crtc->index]; + struct intel_crtc *intel_crtc = to_intel_crtc(crtc); + struct intel_pipe_crc *pipe_crc = &intel_crtc->pipe_crc; enum intel_display_power_domain power_domain; enum intel_pipe_crc_source source; intel_wakeref_t wakeref; @@ -640,7 +637,7 @@ void intel_crtc_enable_pipe_crc(struct intel_crtc *intel_crtc) { struct drm_crtc *crtc = &intel_crtc->base; struct drm_i915_private *dev_priv = to_i915(crtc->dev); - struct intel_pipe_crc *pipe_crc = &dev_priv->pipe_crc[crtc->index]; + struct intel_pipe_crc *pipe_crc = &intel_crtc->pipe_crc; u32 val = 0; if (!crtc->crc.opened) @@ -660,7 +657,7 @@ void intel_crtc_disable_pipe_crc(struct intel_crtc *intel_crtc) { struct drm_crtc *crtc = &intel_crtc->base; struct drm_i915_private *dev_priv = to_i915(crtc->dev); - struct intel_pipe_crc *pipe_crc = &dev_priv->pipe_crc[crtc->index]; + struct intel_pipe_crc *pipe_crc = &intel_crtc->pipe_crc; /* Swallow crc's until we stop generating them. */ spin_lock_irq(&pipe_crc->lock); diff --git a/drivers/gpu/drm/i915/display/intel_pipe_crc.h b/drivers/gpu/drm/i915/display/intel_pipe_crc.h index db258a756fc6..43012b189415 100644 --- a/drivers/gpu/drm/i915/display/intel_pipe_crc.h +++ b/drivers/gpu/drm/i915/display/intel_pipe_crc.h @@ -13,7 +13,7 @@ struct drm_i915_private; struct intel_crtc; #ifdef CONFIG_DEBUG_FS -void intel_display_crc_init(struct drm_i915_private *dev_priv); +void intel_crtc_crc_init(struct intel_crtc *crtc); int intel_crtc_set_crc_source(struct drm_crtc *crtc, const char *source_name); int intel_crtc_verify_crc_source(struct drm_crtc *crtc, const char *source_name, size_t *values_cnt); @@ -22,7 +22,7 @@ const char *const *intel_crtc_get_crc_sources(struct drm_crtc *crtc, void intel_crtc_disable_pipe_crc(struct intel_crtc *crtc); void intel_crtc_enable_pipe_crc(struct intel_crtc *crtc); #else -static inline void intel_display_crc_init(struct drm_i915_private *dev_priv) {} +static inline void intel_crtc_crc_init(struct intel_crtc *crtc) {} #define intel_crtc_set_crc_source NULL #define intel_crtc_verify_crc_source NULL #define intel_crtc_get_crc_sources NULL diff --git a/drivers/gpu/drm/i915/display/intel_psr.c b/drivers/gpu/drm/i915/display/intel_psr.c index 7e754201f54d..fd9b146e3aba 100644 --- a/drivers/gpu/drm/i915/display/intel_psr.c +++ b/drivers/gpu/drm/i915/display/intel_psr.c @@ -304,7 +304,7 @@ void intel_psr_init_dpcd(struct intel_dp *intel_dp) drm_dbg_kms(&dev_priv->drm, "eDP panel supports PSR version %x\n", intel_dp->psr_dpcd[0]); - if (drm_dp_has_quirk(&intel_dp->desc, DP_DPCD_QUIRK_NO_PSR)) { + if (drm_dp_has_quirk(&intel_dp->desc, 0, DP_DPCD_QUIRK_NO_PSR)) { drm_dbg_kms(&dev_priv->drm, "PSR support not currently available for this panel\n"); return; diff --git a/drivers/gpu/drm/i915/display/intel_quirks.c b/drivers/gpu/drm/i915/display/intel_quirks.c index 9d235d270dac..46beb155d835 100644 --- a/drivers/gpu/drm/i915/display/intel_quirks.c +++ b/drivers/gpu/drm/i915/display/intel_quirks.c @@ -82,6 +82,16 @@ static const struct intel_dmi_quirk intel_dmi_quirks[] = { DMI_MATCH(DMI_PRODUCT_NAME, ""), }, }, + { + .callback = intel_dmi_reverse_brightness, + .ident = "Thundersoft TST178 tablet", + /* DMI strings are too generic, also match on BIOS date */ + .matches = {DMI_EXACT_MATCH(DMI_BOARD_VENDOR, "AMI Corporation"), + DMI_EXACT_MATCH(DMI_BOARD_NAME, "Aptio CRB"), + DMI_EXACT_MATCH(DMI_PRODUCT_NAME, "To be filled by O.E.M."), + DMI_EXACT_MATCH(DMI_BIOS_DATE, "04/15/2014"), + }, + }, { } /* terminating entry */ }, .hook = quirk_invert_brightness, diff --git a/drivers/gpu/drm/i915/display/intel_sdvo.c b/drivers/gpu/drm/i915/display/intel_sdvo.c index b0588150752c..637d8fe2f8c2 100644 --- a/drivers/gpu/drm/i915/display/intel_sdvo.c +++ b/drivers/gpu/drm/i915/display/intel_sdvo.c @@ -34,7 +34,6 @@ #include <drm/drm_atomic_helper.h> #include <drm/drm_crtc.h> #include <drm/drm_edid.h> -#include <drm/i915_drm.h> #include "i915_drv.h" #include "intel_atomic.h" diff --git a/drivers/gpu/drm/i915/display/intel_sdvo.h b/drivers/gpu/drm/i915/display/intel_sdvo.h index a66f224aa17d..72065e4360d5 100644 --- a/drivers/gpu/drm/i915/display/intel_sdvo.h +++ b/drivers/gpu/drm/i915/display/intel_sdvo.h @@ -8,8 +8,6 @@ #include <linux/types.h> -#include <drm/i915_drm.h> - #include "i915_reg.h" struct drm_i915_private; diff --git a/drivers/gpu/drm/i915/display/intel_sprite.c b/drivers/gpu/drm/i915/display/intel_sprite.c index 7abeefe8dce5..deda351719db 100644 --- a/drivers/gpu/drm/i915/display/intel_sprite.c +++ b/drivers/gpu/drm/i915/display/intel_sprite.c @@ -37,10 +37,10 @@ #include <drm/drm_fourcc.h> #include <drm/drm_plane_helper.h> #include <drm/drm_rect.h> -#include <drm/i915_drm.h> #include "i915_drv.h" #include "i915_trace.h" +#include "i915_vgpu.h" #include "intel_atomic_plane.h" #include "intel_display_types.h" #include "intel_frontbuffer.h" @@ -284,6 +284,16 @@ int intel_plane_check_src_coordinates(struct intel_plane_state *plane_state) bool rotated = drm_rotation_90_or_270(plane_state->hw.rotation); /* + * FIXME hsub/vsub vs. block size is a mess. Pre-tgl CCS + * abuses hsub/vsub so we can't use them here. But as they + * are limited to 32bpp RGB formats we don't actually need + * to check anything. + */ + if (fb->modifier == I915_FORMAT_MOD_Y_TILED_CCS || + fb->modifier == I915_FORMAT_MOD_Yf_TILED_CCS) + return 0; + + /* * Hardware doesn't handle subpixel coordinates. * Adjust to (macro)pixel boundary, but be careful not to * increase the source viewport size, because that could @@ -297,26 +307,26 @@ int intel_plane_check_src_coordinates(struct intel_plane_state *plane_state) drm_rect_init(src, src_x << 16, src_y << 16, src_w << 16, src_h << 16); - if (!fb->format->is_yuv) - return 0; - - /* YUV specific checks */ - if (!rotated) { + if (fb->format->format == DRM_FORMAT_RGB565 && rotated) { + hsub = 2; + vsub = 2; + } else { hsub = fb->format->hsub; vsub = fb->format->vsub; - } else { - hsub = vsub = max(fb->format->hsub, fb->format->vsub); } + if (rotated) + hsub = vsub = max(hsub, vsub); + if (src_x % hsub || src_w % hsub) { - DRM_DEBUG_KMS("src x/w (%u, %u) must be a multiple of %u for %sYUV planes\n", - src_x, src_w, hsub, rotated ? "rotated " : ""); + DRM_DEBUG_KMS("src x/w (%u, %u) must be a multiple of %u (rotated: %s)\n", + src_x, src_w, hsub, yesno(rotated)); return -EINVAL; } if (src_y % vsub || src_h % vsub) { - DRM_DEBUG_KMS("src y/h (%u, %u) must be a multiple of %u for %sYUV planes\n", - src_y, src_h, vsub, rotated ? "rotated " : ""); + DRM_DEBUG_KMS("src y/h (%u, %u) must be a multiple of %u (rotated: %s)\n", + src_y, src_h, vsub, yesno(rotated)); return -EINVAL; } @@ -355,9 +365,8 @@ static int skl_plane_min_cdclk(const struct intel_crtc_state *crtc_state, const struct intel_plane_state *plane_state) { struct drm_i915_private *dev_priv = to_i915(plane_state->uapi.plane->dev); - unsigned int pixel_rate = crtc_state->pixel_rate; - unsigned int src_w, src_h, dst_w, dst_h; unsigned int num, den; + unsigned int pixel_rate = intel_plane_pixel_rate(crtc_state, plane_state); skl_plane_ratio(crtc_state, plane_state, &num, &den); @@ -365,17 +374,7 @@ static int skl_plane_min_cdclk(const struct intel_crtc_state *crtc_state, if (INTEL_GEN(dev_priv) >= 10 || IS_GEMINILAKE(dev_priv)) den *= 2; - src_w = drm_rect_width(&plane_state->uapi.src) >> 16; - src_h = drm_rect_height(&plane_state->uapi.src) >> 16; - dst_w = drm_rect_width(&plane_state->uapi.dst); - dst_h = drm_rect_height(&plane_state->uapi.dst); - - /* Downscaling limits the maximum pixel rate */ - dst_w = min(src_w, dst_w); - dst_h = min(src_h, dst_h); - - return DIV64_U64_ROUND_UP(mul_u32_u32(pixel_rate * num, src_w * src_h), - mul_u32_u32(den, dst_w * dst_h)); + return DIV_ROUND_UP(pixel_rate * num, den); } static unsigned int @@ -2077,6 +2076,18 @@ vlv_sprite_check(struct intel_crtc_state *crtc_state, return 0; } +static bool intel_format_is_p01x(u32 format) +{ + switch (format) { + case DRM_FORMAT_P010: + case DRM_FORMAT_P012: + case DRM_FORMAT_P016: + return true; + default: + return false; + } +} + static int skl_plane_check_fb(const struct intel_crtc_state *crtc_state, const struct intel_plane_state *plane_state) { @@ -2155,6 +2166,15 @@ static int skl_plane_check_fb(const struct intel_crtc_state *crtc_state, return -EINVAL; } + /* Wa_1606054188:tgl */ + if (IS_TIGERLAKE(dev_priv) && + plane_state->ckey.flags & I915_SET_COLORKEY_SOURCE && + intel_format_is_p01x(fb->format->format)) { + drm_dbg_kms(&dev_priv->drm, + "Source color keying not supported with P01x formats\n"); + return -EINVAL; + } + return 0; } @@ -3011,7 +3031,6 @@ skl_universal_plane_create(struct drm_i915_private *dev_priv, struct intel_plane *plane; enum drm_plane_type plane_type; unsigned int supported_rotations; - unsigned int possible_crtcs; const u64 *modifiers; const u32 *formats; int num_formats; @@ -3066,10 +3085,8 @@ skl_universal_plane_create(struct drm_i915_private *dev_priv, else plane_type = DRM_PLANE_TYPE_OVERLAY; - possible_crtcs = BIT(pipe); - ret = drm_universal_plane_init(&dev_priv->drm, &plane->base, - possible_crtcs, plane_funcs, + 0, plane_funcs, formats, num_formats, modifiers, plane_type, "plane %d%c", plane_id + 1, @@ -3120,7 +3137,6 @@ intel_sprite_plane_create(struct drm_i915_private *dev_priv, { struct intel_plane *plane; const struct drm_plane_funcs *plane_funcs; - unsigned long possible_crtcs; unsigned int supported_rotations; const u64 *modifiers; const u32 *formats; @@ -3205,10 +3221,8 @@ intel_sprite_plane_create(struct drm_i915_private *dev_priv, plane->id = PLANE_SPRITE0 + sprite; plane->frontbuffer_bit = INTEL_FRONTBUFFER(pipe, plane->id); - possible_crtcs = BIT(pipe); - ret = drm_universal_plane_init(&dev_priv->drm, &plane->base, - possible_crtcs, plane_funcs, + 0, plane_funcs, formats, num_formats, modifiers, DRM_PLANE_TYPE_OVERLAY, "sprite %c", sprite_name(pipe, sprite)); diff --git a/drivers/gpu/drm/i915/display/intel_tv.c b/drivers/gpu/drm/i915/display/intel_tv.c index 4f81ee26b7ab..d2e3a3a323e9 100644 --- a/drivers/gpu/drm/i915/display/intel_tv.c +++ b/drivers/gpu/drm/i915/display/intel_tv.c @@ -33,7 +33,6 @@ #include <drm/drm_atomic_helper.h> #include <drm/drm_crtc.h> #include <drm/drm_edid.h> -#include <drm/i915_drm.h> #include "i915_drv.h" #include "intel_connector.h" diff --git a/drivers/gpu/drm/i915/display/intel_vdsc.c b/drivers/gpu/drm/i915/display/intel_vdsc.c index 7cba57ae72fe..95ad87d4ccb3 100644 --- a/drivers/gpu/drm/i915/display/intel_vdsc.c +++ b/drivers/gpu/drm/i915/display/intel_vdsc.c @@ -6,8 +6,6 @@ * Manasi Navare <manasi.d.navare@intel.com> */ -#include <drm/i915_drm.h> - #include "i915_drv.h" #include "intel_display_types.h" #include "intel_dsi.h" diff --git a/drivers/gpu/drm/i915/display/vlv_dsi.c b/drivers/gpu/drm/i915/display/vlv_dsi.c index d07cfad8ce6f..f4c362dc6e15 100644 --- a/drivers/gpu/drm/i915/display/vlv_dsi.c +++ b/drivers/gpu/drm/i915/display/vlv_dsi.c @@ -1591,59 +1591,6 @@ static const struct drm_connector_funcs intel_dsi_connector_funcs = { .atomic_duplicate_state = intel_digital_connector_duplicate_state, }; -static enum drm_panel_orientation -vlv_dsi_get_hw_panel_orientation(struct intel_connector *connector) -{ - struct drm_i915_private *dev_priv = to_i915(connector->base.dev); - struct intel_encoder *encoder = intel_attached_encoder(connector); - enum intel_display_power_domain power_domain; - enum drm_panel_orientation orientation; - struct intel_plane *plane; - struct intel_crtc *crtc; - intel_wakeref_t wakeref; - enum pipe pipe; - u32 val; - - if (!encoder->get_hw_state(encoder, &pipe)) - return DRM_MODE_PANEL_ORIENTATION_UNKNOWN; - - crtc = intel_get_crtc_for_pipe(dev_priv, pipe); - plane = to_intel_plane(crtc->base.primary); - - power_domain = POWER_DOMAIN_PIPE(pipe); - wakeref = intel_display_power_get_if_enabled(dev_priv, power_domain); - if (!wakeref) - return DRM_MODE_PANEL_ORIENTATION_UNKNOWN; - - val = intel_de_read(dev_priv, DSPCNTR(plane->i9xx_plane)); - - if (!(val & DISPLAY_PLANE_ENABLE)) - orientation = DRM_MODE_PANEL_ORIENTATION_UNKNOWN; - else if (val & DISPPLANE_ROTATE_180) - orientation = DRM_MODE_PANEL_ORIENTATION_BOTTOM_UP; - else - orientation = DRM_MODE_PANEL_ORIENTATION_NORMAL; - - intel_display_power_put(dev_priv, power_domain, wakeref); - - return orientation; -} - -static enum drm_panel_orientation -vlv_dsi_get_panel_orientation(struct intel_connector *connector) -{ - struct drm_i915_private *dev_priv = to_i915(connector->base.dev); - enum drm_panel_orientation orientation; - - if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) { - orientation = vlv_dsi_get_hw_panel_orientation(connector); - if (orientation != DRM_MODE_PANEL_ORIENTATION_UNKNOWN) - return orientation; - } - - return intel_dsi_get_panel_orientation(connector); -} - static void vlv_dsi_add_properties(struct intel_connector *connector) { struct drm_i915_private *dev_priv = to_i915(connector->base.dev); @@ -1662,7 +1609,7 @@ static void vlv_dsi_add_properties(struct intel_connector *connector) drm_connector_set_panel_orientation_with_quirk( &connector->base, - vlv_dsi_get_panel_orientation(connector), + intel_dsi_get_panel_orientation(connector), connector->panel.fixed_mode->hdisplay, connector->panel.fixed_mode->vdisplay); } diff --git a/drivers/gpu/drm/i915/gem/i915_gem_client_blt.c b/drivers/gpu/drm/i915/gem/i915_gem_client_blt.c index 81366aa4812b..0598e5382a1d 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_client_blt.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_client_blt.c @@ -217,7 +217,7 @@ static void clear_pages_worker(struct work_struct *work) 0); out_request: if (unlikely(err)) { - i915_request_skip(rq, err); + i915_request_set_error_once(rq, err); err = 0; } diff --git a/drivers/gpu/drm/i915/gem/i915_gem_context.c b/drivers/gpu/drm/i915/gem/i915_gem_context.c index adcebf22a3d3..68326ad3b2e0 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_context.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_context.c @@ -67,10 +67,9 @@ #include <linux/log2.h> #include <linux/nospec.h> -#include <drm/i915_drm.h> - #include "gt/gen6_ppgtt.h" #include "gt/intel_context.h" +#include "gt/intel_context_param.h" #include "gt/intel_engine_heartbeat.h" #include "gt/intel_engine_user.h" #include "gt/intel_ring.h" @@ -243,7 +242,6 @@ static void __free_engines(struct i915_gem_engines *e, unsigned int count) if (!e->engines[count]) continue; - RCU_INIT_POINTER(e->engines[count]->gem_context, NULL); intel_context_put(e->engines[count]); } kfree(e); @@ -256,7 +254,51 @@ static void free_engines(struct i915_gem_engines *e) static void free_engines_rcu(struct rcu_head *rcu) { - free_engines(container_of(rcu, struct i915_gem_engines, rcu)); + struct i915_gem_engines *engines = + container_of(rcu, struct i915_gem_engines, rcu); + + i915_sw_fence_fini(&engines->fence); + free_engines(engines); +} + +static int __i915_sw_fence_call +engines_notify(struct i915_sw_fence *fence, enum i915_sw_fence_notify state) +{ + struct i915_gem_engines *engines = + container_of(fence, typeof(*engines), fence); + + switch (state) { + case FENCE_COMPLETE: + if (!list_empty(&engines->link)) { + struct i915_gem_context *ctx = engines->ctx; + unsigned long flags; + + spin_lock_irqsave(&ctx->stale.lock, flags); + list_del(&engines->link); + spin_unlock_irqrestore(&ctx->stale.lock, flags); + } + i915_gem_context_put(engines->ctx); + break; + + case FENCE_FREE: + init_rcu_head(&engines->rcu); + call_rcu(&engines->rcu, free_engines_rcu); + break; + } + + return NOTIFY_DONE; +} + +static struct i915_gem_engines *alloc_engines(unsigned int count) +{ + struct i915_gem_engines *e; + + e = kzalloc(struct_size(e, engines, count), GFP_KERNEL); + if (!e) + return NULL; + + i915_sw_fence_init(&e->fence, engines_notify); + return e; } static struct i915_gem_engines *default_engines(struct i915_gem_context *ctx) @@ -266,12 +308,10 @@ static struct i915_gem_engines *default_engines(struct i915_gem_context *ctx) struct i915_gem_engines *e; enum intel_engine_id id; - e = kzalloc(struct_size(e, engines, I915_NUM_ENGINES), GFP_KERNEL); + e = alloc_engines(I915_NUM_ENGINES); if (!e) return ERR_PTR(-ENOMEM); - e->ctx = ctx; - for_each_engine(engine, gt, id) { struct intel_context *ce; @@ -305,7 +345,6 @@ static void i915_gem_context_free(struct i915_gem_context *ctx) list_del(&ctx->link); spin_unlock(&ctx->i915->gem.contexts.lock); - free_engines(rcu_access_pointer(ctx->engines)); mutex_destroy(&ctx->engines_mutex); if (ctx->timeline) @@ -492,30 +531,75 @@ static void kill_engines(struct i915_gem_engines *engines) static void kill_stale_engines(struct i915_gem_context *ctx) { struct i915_gem_engines *pos, *next; - unsigned long flags; - spin_lock_irqsave(&ctx->stale.lock, flags); + spin_lock_irq(&ctx->stale.lock); + GEM_BUG_ON(!i915_gem_context_is_closed(ctx)); list_for_each_entry_safe(pos, next, &ctx->stale.engines, link) { - if (!i915_sw_fence_await(&pos->fence)) + if (!i915_sw_fence_await(&pos->fence)) { + list_del_init(&pos->link); continue; + } - spin_unlock_irqrestore(&ctx->stale.lock, flags); + spin_unlock_irq(&ctx->stale.lock); kill_engines(pos); - spin_lock_irqsave(&ctx->stale.lock, flags); + spin_lock_irq(&ctx->stale.lock); + GEM_BUG_ON(i915_sw_fence_signaled(&pos->fence)); list_safe_reset_next(pos, next, link); list_del_init(&pos->link); /* decouple from FENCE_COMPLETE */ i915_sw_fence_complete(&pos->fence); } - spin_unlock_irqrestore(&ctx->stale.lock, flags); + spin_unlock_irq(&ctx->stale.lock); } static void kill_context(struct i915_gem_context *ctx) { kill_stale_engines(ctx); - kill_engines(__context_engines_static(ctx)); +} + +static void engines_idle_release(struct i915_gem_context *ctx, + struct i915_gem_engines *engines) +{ + struct i915_gem_engines_iter it; + struct intel_context *ce; + + INIT_LIST_HEAD(&engines->link); + + engines->ctx = i915_gem_context_get(ctx); + + for_each_gem_engine(ce, engines, it) { + struct dma_fence *fence; + int err = 0; + + /* serialises with execbuf */ + set_bit(CONTEXT_CLOSED_BIT, &ce->flags); + if (!intel_context_pin_if_active(ce)) + continue; + + fence = i915_active_fence_get(&ce->timeline->last_request); + if (fence) { + err = i915_sw_fence_await_dma_fence(&engines->fence, + fence, 0, + GFP_KERNEL); + dma_fence_put(fence); + } + intel_context_unpin(ce); + if (err < 0) + goto kill; + } + + spin_lock_irq(&ctx->stale.lock); + if (!i915_gem_context_is_closed(ctx)) + list_add_tail(&engines->link, &ctx->stale.engines); + spin_unlock_irq(&ctx->stale.lock); + +kill: + if (list_empty(&engines->link)) /* raced, already closed */ + kill_engines(engines); + + i915_sw_fence_commit(&engines->fence); } static void set_closed_name(struct i915_gem_context *ctx) @@ -539,11 +623,16 @@ static void context_close(struct i915_gem_context *ctx) { struct i915_address_space *vm; + /* Flush any concurrent set_engines() */ + mutex_lock(&ctx->engines_mutex); + engines_idle_release(ctx, rcu_replace_pointer(ctx->engines, NULL, 1)); i915_gem_context_set_closed(ctx); - set_closed_name(ctx); + mutex_unlock(&ctx->engines_mutex); mutex_lock(&ctx->mutex); + set_closed_name(ctx); + vm = i915_gem_context_vm(ctx); if (vm) i915_vm_close(vm); @@ -668,23 +757,30 @@ err_free: return ERR_PTR(err); } -static void +static int context_apply_all(struct i915_gem_context *ctx, - void (*fn)(struct intel_context *ce, void *data), + int (*fn)(struct intel_context *ce, void *data), void *data) { struct i915_gem_engines_iter it; struct intel_context *ce; + int err = 0; - for_each_gem_engine(ce, i915_gem_context_lock_engines(ctx), it) - fn(ce, data); + for_each_gem_engine(ce, i915_gem_context_lock_engines(ctx), it) { + err = fn(ce, data); + if (err) + break; + } i915_gem_context_unlock_engines(ctx); + + return err; } -static void __apply_ppgtt(struct intel_context *ce, void *vm) +static int __apply_ppgtt(struct intel_context *ce, void *vm) { i915_vm_put(ce->vm); ce->vm = i915_vm_get(vm); + return 0; } static struct i915_address_space * @@ -722,9 +818,10 @@ static void __set_timeline(struct intel_timeline **dst, intel_timeline_put(old); } -static void __apply_timeline(struct intel_context *ce, void *timeline) +static int __apply_timeline(struct intel_context *ce, void *timeline) { __set_timeline(&ce->timeline, timeline); + return 0; } static void __assign_timeline(struct i915_gem_context *ctx, @@ -806,6 +903,7 @@ void i915_gem_init__contexts(struct drm_i915_private *i915) void i915_gem_driver_release__contexts(struct drm_i915_private *i915) { flush_work(&i915->gem.contexts.free_work); + rcu_barrier(); /* and flush the left over RCU frees */ } static int gem_context_register(struct i915_gem_context *ctx, @@ -971,6 +1069,30 @@ static void cb_retire(struct i915_active *base) kfree(cb); } +static inline struct i915_gem_engines * +__context_engines_await(const struct i915_gem_context *ctx) +{ + struct i915_gem_engines *engines; + + rcu_read_lock(); + do { + engines = rcu_dereference(ctx->engines); + if (unlikely(!engines)) + break; + + if (unlikely(!i915_sw_fence_await(&engines->fence))) + continue; + + if (likely(engines == rcu_access_pointer(ctx->engines))) + break; + + i915_sw_fence_complete(&engines->fence); + } while (1); + rcu_read_unlock(); + + return engines; +} + I915_SELFTEST_DECLARE(static intel_engine_mask_t context_barrier_inject_fault); static int context_barrier_task(struct i915_gem_context *ctx, intel_engine_mask_t engines, @@ -981,6 +1103,7 @@ static int context_barrier_task(struct i915_gem_context *ctx, { struct context_barrier_task *cb; struct i915_gem_engines_iter it; + struct i915_gem_engines *e; struct intel_context *ce; int err = 0; @@ -997,7 +1120,13 @@ static int context_barrier_task(struct i915_gem_context *ctx, return err; } - for_each_gem_engine(ce, i915_gem_context_lock_engines(ctx), it) { + e = __context_engines_await(ctx); + if (!e) { + i915_active_release(&cb->base); + return -ENOENT; + } + + for_each_gem_engine(ce, e, it) { struct i915_request *rq; if (I915_SELFTEST_ONLY(context_barrier_inject_fault & @@ -1028,7 +1157,7 @@ static int context_barrier_task(struct i915_gem_context *ctx, if (err) break; } - i915_gem_context_unlock_engines(ctx); + i915_sw_fence_complete(&e->fence); cb->task = err ? NULL : task; /* caller needs to unwind instead */ cb->data = data; @@ -1215,6 +1344,63 @@ out: return err; } +static int __apply_ringsize(struct intel_context *ce, void *sz) +{ + return intel_context_set_ring_size(ce, (unsigned long)sz); +} + +static int set_ringsize(struct i915_gem_context *ctx, + struct drm_i915_gem_context_param *args) +{ + if (!HAS_LOGICAL_RING_CONTEXTS(ctx->i915)) + return -ENODEV; + + if (args->size) + return -EINVAL; + + if (!IS_ALIGNED(args->value, I915_GTT_PAGE_SIZE)) + return -EINVAL; + + if (args->value < I915_GTT_PAGE_SIZE) + return -EINVAL; + + if (args->value > 128 * I915_GTT_PAGE_SIZE) + return -EINVAL; + + return context_apply_all(ctx, + __apply_ringsize, + __intel_context_ring_size(args->value)); +} + +static int __get_ringsize(struct intel_context *ce, void *arg) +{ + long sz; + + sz = intel_context_get_ring_size(ce); + GEM_BUG_ON(sz > INT_MAX); + + return sz; /* stop on first engine */ +} + +static int get_ringsize(struct i915_gem_context *ctx, + struct drm_i915_gem_context_param *args) +{ + int sz; + + if (!HAS_LOGICAL_RING_CONTEXTS(ctx->i915)) + return -ENODEV; + + if (args->size) + return -EINVAL; + + sz = context_apply_all(ctx, __get_ringsize, NULL); + if (sz < 0) + return sz; + + args->value = sz; + return 0; +} + static int user_to_context_sseu(struct drm_i915_private *i915, const struct drm_i915_gem_context_param_sseu *user, @@ -1562,77 +1748,6 @@ static const i915_user_extension_fn set_engines__extensions[] = { [I915_CONTEXT_ENGINES_EXT_BOND] = set_engines__bond, }; -static int engines_notify(struct i915_sw_fence *fence, - enum i915_sw_fence_notify state) -{ - struct i915_gem_engines *engines = - container_of(fence, typeof(*engines), fence); - - switch (state) { - case FENCE_COMPLETE: - if (!list_empty(&engines->link)) { - struct i915_gem_context *ctx = engines->ctx; - unsigned long flags; - - spin_lock_irqsave(&ctx->stale.lock, flags); - list_del(&engines->link); - spin_unlock_irqrestore(&ctx->stale.lock, flags); - } - break; - - case FENCE_FREE: - init_rcu_head(&engines->rcu); - call_rcu(&engines->rcu, free_engines_rcu); - break; - } - - return NOTIFY_DONE; -} - -static void engines_idle_release(struct i915_gem_engines *engines) -{ - struct i915_gem_engines_iter it; - struct intel_context *ce; - unsigned long flags; - - GEM_BUG_ON(!engines); - i915_sw_fence_init(&engines->fence, engines_notify); - - INIT_LIST_HEAD(&engines->link); - spin_lock_irqsave(&engines->ctx->stale.lock, flags); - if (!i915_gem_context_is_closed(engines->ctx)) - list_add(&engines->link, &engines->ctx->stale.engines); - spin_unlock_irqrestore(&engines->ctx->stale.lock, flags); - if (list_empty(&engines->link)) /* raced, already closed */ - goto kill; - - for_each_gem_engine(ce, engines, it) { - struct dma_fence *fence; - int err; - - if (!ce->timeline) - continue; - - fence = i915_active_fence_get(&ce->timeline->last_request); - if (!fence) - continue; - - err = i915_sw_fence_await_dma_fence(&engines->fence, - fence, 0, - GFP_KERNEL); - - dma_fence_put(fence); - if (err < 0) - goto kill; - } - goto out; - -kill: - kill_engines(engines); -out: - i915_sw_fence_commit(&engines->fence); -} - static int set_engines(struct i915_gem_context *ctx, const struct drm_i915_gem_context_param *args) @@ -1669,14 +1784,10 @@ set_engines(struct i915_gem_context *ctx, * first 64 engines defined here. */ num_engines = (args->size - sizeof(*user)) / sizeof(*user->engines); - - set.engines = kmalloc(struct_size(set.engines, engines, num_engines), - GFP_KERNEL); + set.engines = alloc_engines(num_engines); if (!set.engines) return -ENOMEM; - set.engines->ctx = ctx; - for (n = 0; n < num_engines; n++) { struct i915_engine_class_instance ci; struct intel_engine_cs *engine; @@ -1729,6 +1840,11 @@ set_engines(struct i915_gem_context *ctx, replace: mutex_lock(&ctx->engines_mutex); + if (i915_gem_context_is_closed(ctx)) { + mutex_unlock(&ctx->engines_mutex); + free_engines(set.engines); + return -ENOENT; + } if (args->size) i915_gem_context_set_user_engines(ctx); else @@ -1737,7 +1853,7 @@ replace: mutex_unlock(&ctx->engines_mutex); /* Keep track of old engine sets for kill_context() */ - engines_idle_release(set.engines); + engines_idle_release(ctx, set.engines); return 0; } @@ -1748,7 +1864,7 @@ __copy_engines(struct i915_gem_engines *e) struct i915_gem_engines *copy; unsigned int n; - copy = kmalloc(struct_size(e, engines, e->num_engines), GFP_KERNEL); + copy = alloc_engines(e->num_engines); if (!copy) return ERR_PTR(-ENOMEM); @@ -1852,17 +1968,19 @@ set_persistence(struct i915_gem_context *ctx, return __context_set_persistence(ctx, args->value); } -static void __apply_priority(struct intel_context *ce, void *arg) +static int __apply_priority(struct intel_context *ce, void *arg) { struct i915_gem_context *ctx = arg; if (!intel_engine_has_semaphores(ce->engine)) - return; + return 0; if (ctx->sched.priority >= I915_PRIORITY_NORMAL) intel_context_set_use_semaphores(ce); else intel_context_clear_use_semaphores(ce); + + return 0; } static int set_priority(struct i915_gem_context *ctx, @@ -1955,6 +2073,10 @@ static int ctx_setparam(struct drm_i915_file_private *fpriv, ret = set_persistence(ctx, args); break; + case I915_CONTEXT_PARAM_RINGSIZE: + ret = set_ringsize(ctx, args); + break; + case I915_CONTEXT_PARAM_BAN_PERIOD: default: ret = -EINVAL; @@ -1983,6 +2105,18 @@ static int create_setparam(struct i915_user_extension __user *ext, void *data) return ctx_setparam(arg->fpriv, arg->ctx, &local.param); } +static int copy_ring_size(struct intel_context *dst, + struct intel_context *src) +{ + long sz; + + sz = intel_context_get_ring_size(src); + if (sz < 0) + return sz; + + return intel_context_set_ring_size(dst, sz); +} + static int clone_engines(struct i915_gem_context *dst, struct i915_gem_context *src) { @@ -1991,12 +2125,10 @@ static int clone_engines(struct i915_gem_context *dst, bool user_engines; unsigned long n; - clone = kmalloc(struct_size(e, engines, e->num_engines), GFP_KERNEL); + clone = alloc_engines(e->num_engines); if (!clone) goto err_unlock; - clone->ctx = dst; - for (n = 0; n < e->num_engines; n++) { struct intel_engine_cs *engine; @@ -2026,6 +2158,12 @@ static int clone_engines(struct i915_gem_context *dst, } intel_context_set_gem(clone->engines[n], dst); + + /* Copy across the preferred ringsize */ + if (copy_ring_size(clone->engines[n], e->engines[n])) { + __free_engines(clone, n + 1); + goto err_unlock; + } } clone->num_engines = n; @@ -2033,8 +2171,7 @@ static int clone_engines(struct i915_gem_context *dst, i915_gem_context_unlock_engines(src); /* Serialised by constructor */ - free_engines(__context_engines_static(dst)); - RCU_INIT_POINTER(dst->engines, clone); + engines_idle_release(dst, rcu_replace_pointer(dst->engines, clone, 1)); if (user_engines) i915_gem_context_set_user_engines(dst); else @@ -2388,6 +2525,10 @@ int i915_gem_context_getparam_ioctl(struct drm_device *dev, void *data, args->value = i915_gem_context_is_persistent(ctx); break; + case I915_CONTEXT_PARAM_RINGSIZE: + ret = get_ringsize(ctx, args); + break; + case I915_CONTEXT_PARAM_BAN_PERIOD: default: ret = -EINVAL; @@ -2461,6 +2602,9 @@ i915_gem_engines_iter_next(struct i915_gem_engines_iter *it) const struct i915_gem_engines *e = it->engines; struct intel_context *ctx; + if (unlikely(!e)) + return NULL; + do { if (it->idx >= e->num_engines) return NULL; diff --git a/drivers/gpu/drm/i915/gem/i915_gem_context.h b/drivers/gpu/drm/i915/gem/i915_gem_context.h index 3ae61a355d87..f1d884d304bd 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_context.h +++ b/drivers/gpu/drm/i915/gem/i915_gem_context.h @@ -192,12 +192,16 @@ i915_gem_context_unlock_engines(struct i915_gem_context *ctx) static inline struct intel_context * i915_gem_context_get_engine(struct i915_gem_context *ctx, unsigned int idx) { - struct intel_context *ce = ERR_PTR(-EINVAL); + struct intel_context *ce; rcu_read_lock(); { struct i915_gem_engines *e = rcu_dereference(ctx->engines); - if (likely(idx < e->num_engines && e->engines[idx])) + if (unlikely(!e)) /* context was closed! */ + ce = ERR_PTR(-ENOENT); + else if (likely(idx < e->num_engines && e->engines[idx])) ce = intel_context_get(e->engines[idx]); + else + ce = ERR_PTR(-EINVAL); } rcu_read_unlock(); return ce; @@ -207,7 +211,6 @@ static inline void i915_gem_engines_iter_init(struct i915_gem_engines_iter *it, struct i915_gem_engines *engines) { - GEM_BUG_ON(!engines); it->engines = engines; it->idx = 0; } diff --git a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c index 4f9c1f5a4ded..36d069504836 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c @@ -10,7 +10,6 @@ #include <linux/uaccess.h> #include <drm/drm_syncobj.h> -#include <drm/i915_drm.h> #include "display/intel_frontbuffer.h" @@ -28,6 +27,19 @@ #include "i915_sw_fence_work.h" #include "i915_trace.h" +struct eb_vma { + struct i915_vma *vma; + unsigned int flags; + + /** This vma's place in the execbuf reservation list */ + struct drm_i915_gem_exec_object2 *exec; + struct list_head bind_link; + struct list_head reloc_link; + + struct hlist_node node; + u32 handle; +}; + enum { FORCE_CPU_RELOC = 1, FORCE_GTT_RELOC, @@ -35,17 +47,15 @@ enum { #define DBG_FORCE_RELOC 0 /* choose one of the above! */ }; -#define __EXEC_OBJECT_HAS_REF BIT(31) -#define __EXEC_OBJECT_HAS_PIN BIT(30) -#define __EXEC_OBJECT_HAS_FENCE BIT(29) -#define __EXEC_OBJECT_NEEDS_MAP BIT(28) -#define __EXEC_OBJECT_NEEDS_BIAS BIT(27) -#define __EXEC_OBJECT_INTERNAL_FLAGS (~0u << 27) /* all of the above */ +#define __EXEC_OBJECT_HAS_PIN BIT(31) +#define __EXEC_OBJECT_HAS_FENCE BIT(30) +#define __EXEC_OBJECT_NEEDS_MAP BIT(29) +#define __EXEC_OBJECT_NEEDS_BIAS BIT(28) +#define __EXEC_OBJECT_INTERNAL_FLAGS (~0u << 28) /* all of the above */ #define __EXEC_OBJECT_RESERVED (__EXEC_OBJECT_HAS_PIN | __EXEC_OBJECT_HAS_FENCE) #define __EXEC_HAS_RELOC BIT(31) -#define __EXEC_VALIDATED BIT(30) -#define __EXEC_INTERNAL_FLAGS (~0u << 30) +#define __EXEC_INTERNAL_FLAGS (~0u << 31) #define UPDATE PIN_OFFSET_FIXED #define BATCH_OFFSET_BIAS (256*1024) @@ -220,15 +230,14 @@ struct i915_execbuffer { struct drm_file *file; /** per-file lookup tables and limits */ struct drm_i915_gem_execbuffer2 *args; /** ioctl parameters */ struct drm_i915_gem_exec_object2 *exec; /** ioctl execobj[] */ - struct i915_vma **vma; - unsigned int *flags; + struct eb_vma *vma; struct intel_engine_cs *engine; /** engine to queue the request to */ struct intel_context *context; /* logical state for the request */ struct i915_gem_context *gem_context; /** caller's context */ struct i915_request *request; /** our request to build */ - struct i915_vma *batch; /** identity of the batch obj/vma */ + struct eb_vma *batch; /** identity of the batch obj/vma */ struct i915_vma *trampoline; /** trampoline used for chaining */ /** actual size of execobj[] as we may extend it for the cmdparser */ @@ -276,8 +285,6 @@ struct i915_execbuffer { struct hlist_head *buckets; /** ht for relocation handles */ }; -#define exec_entry(EB, VMA) (&(EB)->exec[(VMA)->exec_flags - (EB)->flags]) - static inline bool eb_use_cmdparser(const struct i915_execbuffer *eb) { return intel_engine_requires_cmd_parser(eb->engine) || @@ -364,9 +371,9 @@ eb_vma_misplaced(const struct drm_i915_gem_exec_object2 *entry, static inline bool eb_pin_vma(struct i915_execbuffer *eb, const struct drm_i915_gem_exec_object2 *entry, - struct i915_vma *vma) + struct eb_vma *ev) { - unsigned int exec_flags = *vma->exec_flags; + struct i915_vma *vma = ev->vma; u64 pin_flags; if (vma->node.size) @@ -375,24 +382,24 @@ eb_pin_vma(struct i915_execbuffer *eb, pin_flags = entry->offset & PIN_OFFSET_MASK; pin_flags |= PIN_USER | PIN_NOEVICT | PIN_OFFSET_FIXED; - if (unlikely(exec_flags & EXEC_OBJECT_NEEDS_GTT)) + if (unlikely(ev->flags & EXEC_OBJECT_NEEDS_GTT)) pin_flags |= PIN_GLOBAL; if (unlikely(i915_vma_pin(vma, 0, 0, pin_flags))) return false; - if (unlikely(exec_flags & EXEC_OBJECT_NEEDS_FENCE)) { + if (unlikely(ev->flags & EXEC_OBJECT_NEEDS_FENCE)) { if (unlikely(i915_vma_pin_fence(vma))) { i915_vma_unpin(vma); return false; } if (vma->fence) - exec_flags |= __EXEC_OBJECT_HAS_FENCE; + ev->flags |= __EXEC_OBJECT_HAS_FENCE; } - *vma->exec_flags = exec_flags | __EXEC_OBJECT_HAS_PIN; - return !eb_vma_misplaced(entry, vma, exec_flags); + ev->flags |= __EXEC_OBJECT_HAS_PIN; + return !eb_vma_misplaced(entry, vma, ev->flags); } static inline void __eb_unreserve_vma(struct i915_vma *vma, unsigned int flags) @@ -406,13 +413,13 @@ static inline void __eb_unreserve_vma(struct i915_vma *vma, unsigned int flags) } static inline void -eb_unreserve_vma(struct i915_vma *vma, unsigned int *flags) +eb_unreserve_vma(struct eb_vma *ev) { - if (!(*flags & __EXEC_OBJECT_HAS_PIN)) + if (!(ev->flags & __EXEC_OBJECT_HAS_PIN)) return; - __eb_unreserve_vma(vma, *flags); - *flags &= ~__EXEC_OBJECT_RESERVED; + __eb_unreserve_vma(ev->vma, ev->flags); + ev->flags &= ~__EXEC_OBJECT_RESERVED; } static int @@ -420,11 +427,11 @@ eb_validate_vma(struct i915_execbuffer *eb, struct drm_i915_gem_exec_object2 *entry, struct i915_vma *vma) { - struct drm_i915_private *i915 = eb->i915; if (unlikely(entry->flags & eb->invalid_flags)) return -EINVAL; - if (unlikely(entry->alignment && !is_power_of_2(entry->alignment))) + if (unlikely(entry->alignment && + !is_power_of_2_u64(entry->alignment))) return -EINVAL; /* @@ -442,14 +449,6 @@ eb_validate_vma(struct i915_execbuffer *eb, } else { entry->pad_to_size = 0; } - - if (unlikely(vma->exec_flags)) { - drm_dbg(&i915->drm, - "Object [handle %d, index %d] appears more than once in object list\n", - entry->handle, (int)(entry - eb->exec)); - return -EINVAL; - } - /* * From drm_mm perspective address space is continuous, * so from this point we're always using non-canonical @@ -472,41 +471,29 @@ eb_validate_vma(struct i915_execbuffer *eb, return 0; } -static int +static void eb_add_vma(struct i915_execbuffer *eb, unsigned int i, unsigned batch_idx, struct i915_vma *vma) { struct drm_i915_gem_exec_object2 *entry = &eb->exec[i]; - int err; + struct eb_vma *ev = &eb->vma[i]; GEM_BUG_ON(i915_vma_is_closed(vma)); - if (!(eb->args->flags & __EXEC_VALIDATED)) { - err = eb_validate_vma(eb, entry, vma); - if (unlikely(err)) - return err; - } + ev->vma = i915_vma_get(vma); + ev->exec = entry; + ev->flags = entry->flags; if (eb->lut_size > 0) { - vma->exec_handle = entry->handle; - hlist_add_head(&vma->exec_node, + ev->handle = entry->handle; + hlist_add_head(&ev->node, &eb->buckets[hash_32(entry->handle, eb->lut_size)]); } if (entry->relocation_count) - list_add_tail(&vma->reloc_link, &eb->relocs); - - /* - * Stash a pointer from the vma to execobj, so we can query its flags, - * size, alignment etc as provided by the user. Also we stash a pointer - * to the vma inside the execobj so that we can use a direct lookup - * to find the right target VMA when doing relocations. - */ - eb->vma[i] = vma; - eb->flags[i] = entry->flags; - vma->exec_flags = &eb->flags[i]; + list_add_tail(&ev->reloc_link, &eb->relocs); /* * SNA is doing fancy tricks with compressing batch buffers, which leads @@ -519,30 +506,23 @@ eb_add_vma(struct i915_execbuffer *eb, */ if (i == batch_idx) { if (entry->relocation_count && - !(eb->flags[i] & EXEC_OBJECT_PINNED)) - eb->flags[i] |= __EXEC_OBJECT_NEEDS_BIAS; + !(ev->flags & EXEC_OBJECT_PINNED)) + ev->flags |= __EXEC_OBJECT_NEEDS_BIAS; if (eb->reloc_cache.has_fence) - eb->flags[i] |= EXEC_OBJECT_NEEDS_FENCE; + ev->flags |= EXEC_OBJECT_NEEDS_FENCE; - eb->batch = vma; + eb->batch = ev; } - err = 0; - if (eb_pin_vma(eb, entry, vma)) { + if (eb_pin_vma(eb, entry, ev)) { if (entry->offset != vma->node.start) { entry->offset = vma->node.start | UPDATE; eb->args->flags |= __EXEC_HAS_RELOC; } } else { - eb_unreserve_vma(vma, vma->exec_flags); - - list_add_tail(&vma->exec_link, &eb->unbound); - if (drm_mm_node_allocated(&vma->node)) - err = i915_vma_unbind(vma); - if (unlikely(err)) - vma->exec_flags = NULL; + eb_unreserve_vma(ev); + list_add_tail(&ev->bind_link, &eb->unbound); } - return err; } static inline int use_cpu_reloc(const struct reloc_cache *cache, @@ -563,14 +543,14 @@ static inline int use_cpu_reloc(const struct reloc_cache *cache, } static int eb_reserve_vma(const struct i915_execbuffer *eb, - struct i915_vma *vma) + struct eb_vma *ev, + u64 pin_flags) { - struct drm_i915_gem_exec_object2 *entry = exec_entry(eb, vma); - unsigned int exec_flags = *vma->exec_flags; - u64 pin_flags; + struct drm_i915_gem_exec_object2 *entry = ev->exec; + unsigned int exec_flags = ev->flags; + struct i915_vma *vma = ev->vma; int err; - pin_flags = PIN_USER | PIN_NONBLOCK; if (exec_flags & EXEC_OBJECT_NEEDS_GTT) pin_flags |= PIN_GLOBAL; @@ -584,11 +564,16 @@ static int eb_reserve_vma(const struct i915_execbuffer *eb, if (exec_flags & __EXEC_OBJECT_NEEDS_MAP) pin_flags |= PIN_MAPPABLE; - if (exec_flags & EXEC_OBJECT_PINNED) { + if (exec_flags & EXEC_OBJECT_PINNED) pin_flags |= entry->offset | PIN_OFFSET_FIXED; - pin_flags &= ~PIN_NONBLOCK; /* force overlapping checks */ - } else if (exec_flags & __EXEC_OBJECT_NEEDS_BIAS) { + else if (exec_flags & __EXEC_OBJECT_NEEDS_BIAS) pin_flags |= BATCH_OFFSET_BIAS | PIN_OFFSET_BIAS; + + if (drm_mm_node_allocated(&vma->node) && + eb_vma_misplaced(entry, vma, ev->flags)) { + err = i915_vma_unbind(vma); + if (err) + return err; } err = i915_vma_pin(vma, @@ -613,8 +598,8 @@ static int eb_reserve_vma(const struct i915_execbuffer *eb, exec_flags |= __EXEC_OBJECT_HAS_FENCE; } - *vma->exec_flags = exec_flags | __EXEC_OBJECT_HAS_PIN; - GEM_BUG_ON(eb_vma_misplaced(entry, vma, exec_flags)); + ev->flags = exec_flags | __EXEC_OBJECT_HAS_PIN; + GEM_BUG_ON(eb_vma_misplaced(entry, vma, ev->flags)); return 0; } @@ -622,10 +607,11 @@ static int eb_reserve_vma(const struct i915_execbuffer *eb, static int eb_reserve(struct i915_execbuffer *eb) { const unsigned int count = eb->buffer_count; + unsigned int pin_flags = PIN_USER | PIN_NONBLOCK; struct list_head last; - struct i915_vma *vma; + struct eb_vma *ev; unsigned int i, pass; - int err; + int err = 0; /* * Attempt to pin all of the buffers into the GTT. @@ -641,44 +627,54 @@ static int eb_reserve(struct i915_execbuffer *eb) * room for the earlier objects *unless* we need to defragment. */ + if (mutex_lock_interruptible(&eb->i915->drm.struct_mutex)) + return -EINTR; + pass = 0; - err = 0; do { - list_for_each_entry(vma, &eb->unbound, exec_link) { - err = eb_reserve_vma(eb, vma); + list_for_each_entry(ev, &eb->unbound, bind_link) { + err = eb_reserve_vma(eb, ev, pin_flags); if (err) break; } - if (err != -ENOSPC) - return err; + if (!(err == -ENOSPC || err == -EAGAIN)) + break; /* Resort *all* the objects into priority order */ INIT_LIST_HEAD(&eb->unbound); INIT_LIST_HEAD(&last); for (i = 0; i < count; i++) { - unsigned int flags = eb->flags[i]; - struct i915_vma *vma = eb->vma[i]; + unsigned int flags; + ev = &eb->vma[i]; + flags = ev->flags; if (flags & EXEC_OBJECT_PINNED && flags & __EXEC_OBJECT_HAS_PIN) continue; - eb_unreserve_vma(vma, &eb->flags[i]); + eb_unreserve_vma(ev); if (flags & EXEC_OBJECT_PINNED) /* Pinned must have their slot */ - list_add(&vma->exec_link, &eb->unbound); + list_add(&ev->bind_link, &eb->unbound); else if (flags & __EXEC_OBJECT_NEEDS_MAP) /* Map require the lowest 256MiB (aperture) */ - list_add_tail(&vma->exec_link, &eb->unbound); + list_add_tail(&ev->bind_link, &eb->unbound); else if (!(flags & EXEC_OBJECT_SUPPORTS_48B_ADDRESS)) /* Prioritise 4GiB region for restricted bo */ - list_add(&vma->exec_link, &last); + list_add(&ev->bind_link, &last); else - list_add_tail(&vma->exec_link, &last); + list_add_tail(&ev->bind_link, &last); } list_splice_tail(&last, &eb->unbound); + if (err == -EAGAIN) { + mutex_unlock(&eb->i915->drm.struct_mutex); + flush_workqueue(eb->i915->mm.userptr_wq); + mutex_lock(&eb->i915->drm.struct_mutex); + continue; + } + switch (pass++) { case 0: break; @@ -689,13 +685,20 @@ static int eb_reserve(struct i915_execbuffer *eb) err = i915_gem_evict_vm(eb->context->vm); mutex_unlock(&eb->context->vm->mutex); if (err) - return err; + goto unlock; break; default: - return -ENOSPC; + err = -ENOSPC; + goto unlock; } + + pin_flags = PIN_USER; } while (1); + +unlock: + mutex_unlock(&eb->i915->drm.struct_mutex); + return err; } static unsigned int eb_batch_index(const struct i915_execbuffer *eb) @@ -732,17 +735,14 @@ static int eb_lookup_vmas(struct i915_execbuffer *eb) unsigned int i, batch; int err; + if (unlikely(i915_gem_context_is_closed(eb->gem_context))) + return -ENOENT; + INIT_LIST_HEAD(&eb->relocs); INIT_LIST_HEAD(&eb->unbound); batch = eb_batch_index(eb); - mutex_lock(&eb->gem_context->mutex); - if (unlikely(i915_gem_context_is_closed(eb->gem_context))) { - err = -ENOENT; - goto err_ctx; - } - for (i = 0; i < eb->buffer_count; i++) { u32 handle = eb->exec[i].handle; struct i915_lut_handle *lut; @@ -787,45 +787,37 @@ static int eb_lookup_vmas(struct i915_execbuffer *eb) i915_gem_object_unlock(obj); add_vma: - err = eb_add_vma(eb, i, batch, vma); + err = eb_validate_vma(eb, &eb->exec[i], vma); if (unlikely(err)) goto err_vma; - GEM_BUG_ON(vma != eb->vma[i]); - GEM_BUG_ON(vma->exec_flags != &eb->flags[i]); - GEM_BUG_ON(drm_mm_node_allocated(&vma->node) && - eb_vma_misplaced(&eb->exec[i], vma, eb->flags[i])); + eb_add_vma(eb, i, batch, vma); } - mutex_unlock(&eb->gem_context->mutex); - - eb->args->flags |= __EXEC_VALIDATED; - return eb_reserve(eb); + return 0; err_obj: i915_gem_object_put(obj); err_vma: - eb->vma[i] = NULL; -err_ctx: - mutex_unlock(&eb->gem_context->mutex); + eb->vma[i].vma = NULL; return err; } -static struct i915_vma * +static struct eb_vma * eb_get_vma(const struct i915_execbuffer *eb, unsigned long handle) { if (eb->lut_size < 0) { if (handle >= -eb->lut_size) return NULL; - return eb->vma[handle]; + return &eb->vma[handle]; } else { struct hlist_head *head; - struct i915_vma *vma; + struct eb_vma *ev; head = &eb->buckets[hash_32(handle, eb->lut_size)]; - hlist_for_each_entry(vma, head, exec_node) { - if (vma->exec_handle == handle) - return vma; + hlist_for_each_entry(ev, head, node) { + if (ev->handle == handle) + return ev; } return NULL; } @@ -837,32 +829,21 @@ static void eb_release_vmas(const struct i915_execbuffer *eb) unsigned int i; for (i = 0; i < count; i++) { - struct i915_vma *vma = eb->vma[i]; - unsigned int flags = eb->flags[i]; + struct eb_vma *ev = &eb->vma[i]; + struct i915_vma *vma = ev->vma; if (!vma) break; - GEM_BUG_ON(vma->exec_flags != &eb->flags[i]); - vma->exec_flags = NULL; - eb->vma[i] = NULL; + eb->vma[i].vma = NULL; - if (flags & __EXEC_OBJECT_HAS_PIN) - __eb_unreserve_vma(vma, flags); + if (ev->flags & __EXEC_OBJECT_HAS_PIN) + __eb_unreserve_vma(vma, ev->flags); - if (flags & __EXEC_OBJECT_HAS_REF) - i915_vma_put(vma); + i915_vma_put(vma); } } -static void eb_reset_vmas(const struct i915_execbuffer *eb) -{ - eb_release_vmas(eb); - if (eb->lut_size > 0) - memset(eb->buckets, 0, - sizeof(struct hlist_head) << eb->lut_size); -} - static void eb_destroy(const struct i915_execbuffer *eb) { GEM_BUG_ON(eb->reloc_cache.rq); @@ -1198,7 +1179,7 @@ static int __reloc_gpu_alloc(struct i915_execbuffer *eb, goto out_pool; skip_request: - i915_request_skip(rq, err); + i915_request_set_error_once(rq, err); err_request: i915_request_add(rq); err_unpin: @@ -1329,11 +1310,11 @@ out: static u64 eb_relocate_entry(struct i915_execbuffer *eb, - struct i915_vma *vma, + struct eb_vma *ev, const struct drm_i915_gem_relocation_entry *reloc) { struct drm_i915_private *i915 = eb->i915; - struct i915_vma *target; + struct eb_vma *target; int err; /* we've already hold a reference to all valid objects */ @@ -1365,7 +1346,7 @@ eb_relocate_entry(struct i915_execbuffer *eb, } if (reloc->write_domain) { - *target->exec_flags |= EXEC_OBJECT_WRITE; + target->flags |= EXEC_OBJECT_WRITE; /* * Sandybridge PPGTT errata: We need a global gtt mapping @@ -1375,7 +1356,8 @@ eb_relocate_entry(struct i915_execbuffer *eb, */ if (reloc->write_domain == I915_GEM_DOMAIN_INSTRUCTION && IS_GEN(eb->i915, 6)) { - err = i915_vma_bind(target, target->obj->cache_level, + err = i915_vma_bind(target->vma, + target->vma->obj->cache_level, PIN_GLOBAL, NULL); if (WARN_ONCE(err, "Unexpected failure to bind target VMA!")) @@ -1388,17 +1370,17 @@ eb_relocate_entry(struct i915_execbuffer *eb, * more work needs to be done. */ if (!DBG_FORCE_RELOC && - gen8_canonical_addr(target->node.start) == reloc->presumed_offset) + gen8_canonical_addr(target->vma->node.start) == reloc->presumed_offset) return 0; /* Check that the relocation address is valid... */ if (unlikely(reloc->offset > - vma->size - (eb->reloc_cache.use_64bit_reloc ? 8 : 4))) { + ev->vma->size - (eb->reloc_cache.use_64bit_reloc ? 8 : 4))) { drm_dbg(&i915->drm, "Relocation beyond object bounds: " "target %d offset %d size %d.\n", reloc->target_handle, (int)reloc->offset, - (int)vma->size); + (int)ev->vma->size); return -EINVAL; } if (unlikely(reloc->offset & 3)) { @@ -1417,18 +1399,18 @@ eb_relocate_entry(struct i915_execbuffer *eb, * do relocations we are already stalling, disable the user's opt * out of our synchronisation. */ - *vma->exec_flags &= ~EXEC_OBJECT_ASYNC; + ev->flags &= ~EXEC_OBJECT_ASYNC; /* and update the user's relocation entry */ - return relocate_entry(vma, reloc, eb, target); + return relocate_entry(ev->vma, reloc, eb, target->vma); } -static int eb_relocate_vma(struct i915_execbuffer *eb, struct i915_vma *vma) +static int eb_relocate_vma(struct i915_execbuffer *eb, struct eb_vma *ev) { #define N_RELOC(x) ((x) / sizeof(struct drm_i915_gem_relocation_entry)) struct drm_i915_gem_relocation_entry stack[N_RELOC(512)]; struct drm_i915_gem_relocation_entry __user *urelocs; - const struct drm_i915_gem_exec_object2 *entry = exec_entry(eb, vma); + const struct drm_i915_gem_exec_object2 *entry = ev->exec; unsigned int remain; urelocs = u64_to_user_ptr(entry->relocs_ptr); @@ -1458,9 +1440,7 @@ static int eb_relocate_vma(struct i915_execbuffer *eb, struct i915_vma *vma) * we would try to acquire the struct mutex again. Obviously * this is bad and so lockdep complains vehemently. */ - pagefault_disable(); - copied = __copy_from_user_inatomic(r, urelocs, count * sizeof(r[0])); - pagefault_enable(); + copied = __copy_from_user(r, urelocs, count * sizeof(r[0])); if (unlikely(copied)) { remain = -EFAULT; goto out; @@ -1468,7 +1448,7 @@ static int eb_relocate_vma(struct i915_execbuffer *eb, struct i915_vma *vma) remain -= count; do { - u64 offset = eb_relocate_entry(eb, vma, r); + u64 offset = eb_relocate_entry(eb, ev, r); if (likely(offset == 0)) { } else if ((s64)offset < 0) { @@ -1510,278 +1490,34 @@ out: return remain; } -static int -eb_relocate_vma_slow(struct i915_execbuffer *eb, struct i915_vma *vma) -{ - const struct drm_i915_gem_exec_object2 *entry = exec_entry(eb, vma); - struct drm_i915_gem_relocation_entry *relocs = - u64_to_ptr(typeof(*relocs), entry->relocs_ptr); - unsigned int i; - int err; - - for (i = 0; i < entry->relocation_count; i++) { - u64 offset = eb_relocate_entry(eb, vma, &relocs[i]); - - if ((s64)offset < 0) { - err = (int)offset; - goto err; - } - } - err = 0; -err: - reloc_cache_reset(&eb->reloc_cache); - return err; -} - -static int check_relocations(const struct drm_i915_gem_exec_object2 *entry) -{ - const char __user *addr, *end; - unsigned long size; - char __maybe_unused c; - - size = entry->relocation_count; - if (size == 0) - return 0; - - if (size > N_RELOC(ULONG_MAX)) - return -EINVAL; - - addr = u64_to_user_ptr(entry->relocs_ptr); - size *= sizeof(struct drm_i915_gem_relocation_entry); - if (!access_ok(addr, size)) - return -EFAULT; - - end = addr + size; - for (; addr < end; addr += PAGE_SIZE) { - int err = __get_user(c, addr); - if (err) - return err; - } - return __get_user(c, end - 1); -} - -static int eb_copy_relocations(const struct i915_execbuffer *eb) +static int eb_relocate(struct i915_execbuffer *eb) { - struct drm_i915_gem_relocation_entry *relocs; - const unsigned int count = eb->buffer_count; - unsigned int i; int err; - for (i = 0; i < count; i++) { - const unsigned int nreloc = eb->exec[i].relocation_count; - struct drm_i915_gem_relocation_entry __user *urelocs; - unsigned long size; - unsigned long copied; - - if (nreloc == 0) - continue; - - err = check_relocations(&eb->exec[i]); - if (err) - goto err; - - urelocs = u64_to_user_ptr(eb->exec[i].relocs_ptr); - size = nreloc * sizeof(*relocs); - - relocs = kvmalloc_array(size, 1, GFP_KERNEL); - if (!relocs) { - err = -ENOMEM; - goto err; - } - - /* copy_from_user is limited to < 4GiB */ - copied = 0; - do { - unsigned int len = - min_t(u64, BIT_ULL(31), size - copied); - - if (__copy_from_user((char *)relocs + copied, - (char __user *)urelocs + copied, - len)) - goto end; - - copied += len; - } while (copied < size); - - /* - * As we do not update the known relocation offsets after - * relocating (due to the complexities in lock handling), - * we need to mark them as invalid now so that we force the - * relocation processing next time. Just in case the target - * object is evicted and then rebound into its old - * presumed_offset before the next execbuffer - if that - * happened we would make the mistake of assuming that the - * relocations were valid. - */ - if (!user_access_begin(urelocs, size)) - goto end; - - for (copied = 0; copied < nreloc; copied++) - unsafe_put_user(-1, - &urelocs[copied].presumed_offset, - end_user); - user_access_end(); - - eb->exec[i].relocs_ptr = (uintptr_t)relocs; - } - - return 0; - -end_user: - user_access_end(); -end: - kvfree(relocs); - err = -EFAULT; -err: - while (i--) { - relocs = u64_to_ptr(typeof(*relocs), eb->exec[i].relocs_ptr); - if (eb->exec[i].relocation_count) - kvfree(relocs); - } - return err; -} - -static int eb_prefault_relocations(const struct i915_execbuffer *eb) -{ - const unsigned int count = eb->buffer_count; - unsigned int i; - - for (i = 0; i < count; i++) { - int err; - - err = check_relocations(&eb->exec[i]); - if (err) - return err; - } - - return 0; -} - -static noinline int eb_relocate_slow(struct i915_execbuffer *eb) -{ - struct drm_device *dev = &eb->i915->drm; - bool have_copy = false; - struct i915_vma *vma; - int err = 0; - -repeat: - if (signal_pending(current)) { - err = -ERESTARTSYS; - goto out; - } - - /* We may process another execbuffer during the unlock... */ - eb_reset_vmas(eb); - mutex_unlock(&dev->struct_mutex); - - /* - * We take 3 passes through the slowpatch. - * - * 1 - we try to just prefault all the user relocation entries and - * then attempt to reuse the atomic pagefault disabled fast path again. - * - * 2 - we copy the user entries to a local buffer here outside of the - * local and allow ourselves to wait upon any rendering before - * relocations - * - * 3 - we already have a local copy of the relocation entries, but - * were interrupted (EAGAIN) whilst waiting for the objects, try again. - */ - if (!err) { - err = eb_prefault_relocations(eb); - } else if (!have_copy) { - err = eb_copy_relocations(eb); - have_copy = err == 0; - } else { - cond_resched(); - err = 0; - } - if (err) { - mutex_lock(&dev->struct_mutex); - goto out; - } - - /* A frequent cause for EAGAIN are currently unavailable client pages */ - flush_workqueue(eb->i915->mm.userptr_wq); - - err = i915_mutex_lock_interruptible(dev); - if (err) { - mutex_lock(&dev->struct_mutex); - goto out; - } - - /* reacquire the objects */ + mutex_lock(&eb->gem_context->mutex); err = eb_lookup_vmas(eb); + mutex_unlock(&eb->gem_context->mutex); if (err) - goto err; - - GEM_BUG_ON(!eb->batch); - - list_for_each_entry(vma, &eb->relocs, reloc_link) { - if (!have_copy) { - pagefault_disable(); - err = eb_relocate_vma(eb, vma); - pagefault_enable(); - if (err) - goto repeat; - } else { - err = eb_relocate_vma_slow(eb, vma); - if (err) - goto err; - } - } - - /* - * Leave the user relocations as are, this is the painfully slow path, - * and we want to avoid the complication of dropping the lock whilst - * having buffers reserved in the aperture and so causing spurious - * ENOSPC for random operations. - */ - -err: - if (err == -EAGAIN) - goto repeat; - -out: - if (have_copy) { - const unsigned int count = eb->buffer_count; - unsigned int i; - - for (i = 0; i < count; i++) { - const struct drm_i915_gem_exec_object2 *entry = - &eb->exec[i]; - struct drm_i915_gem_relocation_entry *relocs; - - if (!entry->relocation_count) - continue; + return err; - relocs = u64_to_ptr(typeof(*relocs), entry->relocs_ptr); - kvfree(relocs); - } + if (!list_empty(&eb->unbound)) { + err = eb_reserve(eb); + if (err) + return err; } - return err; -} - -static int eb_relocate(struct i915_execbuffer *eb) -{ - if (eb_lookup_vmas(eb)) - goto slow; - /* The objects are in their final locations, apply the relocations. */ if (eb->args->flags & __EXEC_HAS_RELOC) { - struct i915_vma *vma; + struct eb_vma *ev; - list_for_each_entry(vma, &eb->relocs, reloc_link) { - if (eb_relocate_vma(eb, vma)) - goto slow; + list_for_each_entry(ev, &eb->relocs, reloc_link) { + err = eb_relocate_vma(eb, ev); + if (err) + return err; } } return 0; - -slow: - return eb_relocate_slow(eb); } static int eb_move_to_gpu(struct i915_execbuffer *eb) @@ -1794,27 +1530,19 @@ static int eb_move_to_gpu(struct i915_execbuffer *eb) ww_acquire_init(&acquire, &reservation_ww_class); for (i = 0; i < count; i++) { - struct i915_vma *vma = eb->vma[i]; + struct eb_vma *ev = &eb->vma[i]; + struct i915_vma *vma = ev->vma; err = ww_mutex_lock_interruptible(&vma->resv->lock, &acquire); - if (!err) - continue; - - GEM_BUG_ON(err == -EALREADY); /* No duplicate vma */ - if (err == -EDEADLK) { GEM_BUG_ON(i == 0); do { int j = i - 1; - ww_mutex_unlock(&eb->vma[j]->resv->lock); + ww_mutex_unlock(&eb->vma[j].vma->resv->lock); - swap(eb->flags[i], eb->flags[j]); swap(eb->vma[i], eb->vma[j]); - eb->vma[i]->exec_flags = &eb->flags[i]; } while (--i); - GEM_BUG_ON(vma != eb->vma[0]); - vma->exec_flags = &eb->flags[0]; err = ww_mutex_lock_slow_interruptible(&vma->resv->lock, &acquire); @@ -1825,8 +1553,9 @@ static int eb_move_to_gpu(struct i915_execbuffer *eb) ww_acquire_done(&acquire); while (i--) { - unsigned int flags = eb->flags[i]; - struct i915_vma *vma = eb->vma[i]; + struct eb_vma *ev = &eb->vma[i]; + struct i915_vma *vma = ev->vma; + unsigned int flags = ev->flags; struct drm_i915_gem_object *obj = vma->obj; assert_vma_held(vma); @@ -1870,10 +1599,9 @@ static int eb_move_to_gpu(struct i915_execbuffer *eb) i915_vma_unlock(vma); __eb_unreserve_vma(vma, flags); - vma->exec_flags = NULL; + i915_vma_put(vma); - if (unlikely(flags & __EXEC_OBJECT_HAS_REF)) - i915_vma_put(vma); + ev->vma = NULL; } ww_acquire_fini(&acquire); @@ -1887,7 +1615,7 @@ static int eb_move_to_gpu(struct i915_execbuffer *eb) return 0; err_skip: - i915_request_skip(eb->request, err); + i915_request_set_error_once(eb->request, err); return err; } @@ -2008,7 +1736,7 @@ static int eb_parse_pipeline(struct i915_execbuffer *eb, if (!pw) return -ENOMEM; - err = i915_active_acquire(&eb->batch->active); + err = i915_active_acquire(&eb->batch->vma->active); if (err) goto err_free; @@ -2025,7 +1753,7 @@ static int eb_parse_pipeline(struct i915_execbuffer *eb, dma_fence_work_init(&pw->base, &eb_parse_ops); pw->engine = eb->engine; - pw->batch = eb->batch; + pw->batch = eb->batch->vma; pw->batch_offset = eb->batch_start_offset; pw->batch_length = eb->batch_len; pw->shadow = shadow; @@ -2067,7 +1795,7 @@ err_trampoline: err_shadow: i915_active_release(&shadow->active); err_batch: - i915_active_release(&eb->batch->active); + i915_active_release(&eb->batch->vma->active); err_free: kfree(pw); return err; @@ -2130,15 +1858,12 @@ static int eb_parse(struct i915_execbuffer *eb) if (err) goto err_trampoline; - eb->vma[eb->buffer_count] = i915_vma_get(shadow); - eb->flags[eb->buffer_count] = - __EXEC_OBJECT_HAS_PIN | __EXEC_OBJECT_HAS_REF; - shadow->exec_flags = &eb->flags[eb->buffer_count]; - eb->buffer_count++; + eb->vma[eb->buffer_count].vma = i915_vma_get(shadow); + eb->vma[eb->buffer_count].flags = __EXEC_OBJECT_HAS_PIN; + eb->batch = &eb->vma[eb->buffer_count++]; eb->trampoline = trampoline; eb->batch_start_offset = 0; - eb->batch = shadow; shadow->private = pool; return 0; @@ -2165,7 +1890,7 @@ add_to_client(struct i915_request *rq, struct drm_file *file) spin_unlock(&file_priv->mm.lock); } -static int eb_submit(struct i915_execbuffer *eb) +static int eb_submit(struct i915_execbuffer *eb, struct i915_vma *batch) { int err; @@ -2192,7 +1917,7 @@ static int eb_submit(struct i915_execbuffer *eb) } err = eb->engine->emit_bb_start(eb->request, - eb->batch->node.start + + batch->node.start + eb->batch_start_offset, eb->batch_len, eb->batch_flags); @@ -2327,15 +2052,22 @@ static int __eb_pin_engine(struct i915_execbuffer *eb, struct intel_context *ce) intel_context_timeline_unlock(tl); if (rq) { - if (i915_request_wait(rq, - I915_WAIT_INTERRUPTIBLE, - MAX_SCHEDULE_TIMEOUT) < 0) { - i915_request_put(rq); - err = -EINTR; - goto err_exit; - } + bool nonblock = eb->file->filp->f_flags & O_NONBLOCK; + long timeout; + timeout = MAX_SCHEDULE_TIMEOUT; + if (nonblock) + timeout = 0; + + timeout = i915_request_wait(rq, + I915_WAIT_INTERRUPTIBLE, + timeout); i915_request_put(rq); + + if (timeout < 0) { + err = nonblock ? -EWOULDBLOCK : timeout; + goto err_exit; + } } eb->engine = ce->engine; @@ -2560,6 +2292,73 @@ signal_fence_array(struct i915_execbuffer *eb, } } +static void retire_requests(struct intel_timeline *tl, struct i915_request *end) +{ + struct i915_request *rq, *rn; + + list_for_each_entry_safe(rq, rn, &tl->requests, link) + if (rq == end || !i915_request_retire(rq)) + break; +} + +static void eb_request_add(struct i915_execbuffer *eb) +{ + struct i915_request *rq = eb->request; + struct intel_timeline * const tl = i915_request_timeline(rq); + struct i915_sched_attr attr = {}; + struct i915_request *prev; + + lockdep_assert_held(&tl->mutex); + lockdep_unpin_lock(&tl->mutex, rq->cookie); + + trace_i915_request_add(rq); + + prev = __i915_request_commit(rq); + + /* Check that the context wasn't destroyed before submission */ + if (likely(!intel_context_is_closed(eb->context))) { + attr = eb->gem_context->sched; + + /* + * Boost actual workloads past semaphores! + * + * With semaphores we spin on one engine waiting for another, + * simply to reduce the latency of starting our work when + * the signaler completes. However, if there is any other + * work that we could be doing on this engine instead, that + * is better utilisation and will reduce the overall duration + * of the current work. To avoid PI boosting a semaphore + * far in the distance past over useful work, we keep a history + * of any semaphore use along our dependency chain. + */ + if (!(rq->sched.flags & I915_SCHED_HAS_SEMAPHORE_CHAIN)) + attr.priority |= I915_PRIORITY_NOSEMAPHORE; + + /* + * Boost priorities to new clients (new request flows). + * + * Allow interactive/synchronous clients to jump ahead of + * the bulk clients. (FQ_CODEL) + */ + if (list_empty(&rq->sched.signalers_list)) + attr.priority |= I915_PRIORITY_WAIT; + } else { + /* Serialise with context_close via the add_to_timeline */ + i915_request_set_error_once(rq, -ENOENT); + __i915_request_skip(rq); + } + + local_bh_disable(); + __i915_request_queue(rq, &attr); + local_bh_enable(); /* Kick the execlists tasklet if just scheduled */ + + /* Try to clean up the client's timeline after submitting the request */ + if (prev) + retire_requests(tl, prev); + + mutex_unlock(&tl->mutex); +} + static int i915_gem_do_execbuffer(struct drm_device *dev, struct drm_file *file, @@ -2572,6 +2371,7 @@ i915_gem_do_execbuffer(struct drm_device *dev, struct dma_fence *in_fence = NULL; struct dma_fence *exec_fence = NULL; struct sync_file *out_fence = NULL; + struct i915_vma *batch; int out_fence_fd = -1; int err; @@ -2586,9 +2386,8 @@ i915_gem_do_execbuffer(struct drm_device *dev, args->flags |= __EXEC_HAS_RELOC; eb.exec = exec; - eb.vma = (struct i915_vma **)(exec + args->buffer_count + 1); - eb.vma[0] = NULL; - eb.flags = (unsigned int *)(eb.vma + args->buffer_count + 1); + eb.vma = (struct eb_vma *)(exec + args->buffer_count + 1); + eb.vma[0].vma = NULL; eb.invalid_flags = __EXEC_OBJECT_UNKNOWN_FLAGS; reloc_cache_init(&eb.reloc_cache, eb.i915); @@ -2656,10 +2455,6 @@ i915_gem_do_execbuffer(struct drm_device *dev, if (unlikely(err)) goto err_context; - err = i915_mutex_lock_interruptible(dev); - if (err) - goto err_engine; - err = eb_relocate(&eb); if (err) { /* @@ -2673,21 +2468,23 @@ i915_gem_do_execbuffer(struct drm_device *dev, goto err_vma; } - if (unlikely(*eb.batch->exec_flags & EXEC_OBJECT_WRITE)) { + if (unlikely(eb.batch->flags & EXEC_OBJECT_WRITE)) { drm_dbg(&i915->drm, "Attempting to use self-modifying batch buffer\n"); err = -EINVAL; goto err_vma; } - if (eb.batch_start_offset > eb.batch->size || - eb.batch_len > eb.batch->size - eb.batch_start_offset) { + + if (range_overflows_t(u64, + eb.batch_start_offset, eb.batch_len, + eb.batch->vma->size)) { drm_dbg(&i915->drm, "Attempting to use out-of-bounds batch\n"); err = -EINVAL; goto err_vma; } if (eb.batch_len == 0) - eb.batch_len = eb.batch->size - eb.batch_start_offset; + eb.batch_len = eb.batch->vma->size - eb.batch_start_offset; err = eb_parse(&eb); if (err) @@ -2697,6 +2494,7 @@ i915_gem_do_execbuffer(struct drm_device *dev, * snb/ivb/vlv conflate the "batch in ppgtt" bit with the "non-secure * batch" bit. Hence we need to pin secure batches into the global gtt. * hsw should have this fixed, but bdw mucks it up again. */ + batch = eb.batch->vma; if (eb.batch_flags & I915_DISPATCH_SECURE) { struct i915_vma *vma; @@ -2710,13 +2508,13 @@ i915_gem_do_execbuffer(struct drm_device *dev, * fitting due to fragmentation. * So this is actually safe. */ - vma = i915_gem_object_ggtt_pin(eb.batch->obj, NULL, 0, 0, 0); + vma = i915_gem_object_ggtt_pin(batch->obj, NULL, 0, 0, 0); if (IS_ERR(vma)) { err = PTR_ERR(vma); goto err_parse; } - eb.batch = vma; + batch = vma; } /* All GPU relocation batches must be submitted prior to the user rq */ @@ -2763,16 +2561,16 @@ i915_gem_do_execbuffer(struct drm_device *dev, * inactive_list and lose its active reference. Hence we do not need * to explicitly hold another reference here. */ - eb.request->batch = eb.batch; - if (eb.batch->private) - intel_engine_pool_mark_active(eb.batch->private, eb.request); + eb.request->batch = batch; + if (batch->private) + intel_engine_pool_mark_active(batch->private, eb.request); trace_i915_request_queue(eb.request, eb.batch_flags); - err = eb_submit(&eb); + err = eb_submit(&eb, batch); err_request: add_to_client(eb.request, file); i915_request_get(eb.request); - i915_request_add(eb.request); + eb_request_add(&eb); if (fences) signal_fence_array(&eb, fences); @@ -2791,17 +2589,15 @@ err_request: err_batch_unpin: if (eb.batch_flags & I915_DISPATCH_SECURE) - i915_vma_unpin(eb.batch); + i915_vma_unpin(batch); err_parse: - if (eb.batch->private) - intel_engine_pool_put(eb.batch->private); + if (batch->private) + intel_engine_pool_put(batch->private); err_vma: if (eb.exec) eb_release_vmas(&eb); if (eb.trampoline) i915_vma_unpin(eb.trampoline); - mutex_unlock(&dev->struct_mutex); -err_engine: eb_unpin_engine(&eb); err_context: i915_gem_context_put(eb.gem_context); @@ -2819,9 +2615,7 @@ err_in_fence: static size_t eb_element_size(void) { - return (sizeof(struct drm_i915_gem_exec_object2) + - sizeof(struct i915_vma *) + - sizeof(unsigned int)); + return sizeof(struct drm_i915_gem_exec_object2) + sizeof(struct eb_vma); } static bool check_buffer_count(size_t count) diff --git a/drivers/gpu/drm/i915/gem/i915_gem_internal.c b/drivers/gpu/drm/i915/gem/i915_gem_internal.c index 9cfb0e41ff06..cbbff81aa0af 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_internal.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_internal.c @@ -8,8 +8,6 @@ #include <linux/slab.h> #include <linux/swiotlb.h> -#include <drm/i915_drm.h> - #include "i915_drv.h" #include "i915_gem.h" #include "i915_gem_object.h" diff --git a/drivers/gpu/drm/i915/gem/i915_gem_mman.c b/drivers/gpu/drm/i915/gem/i915_gem_mman.c index e8cccc131c40..b39c24dae64e 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_mman.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_mman.c @@ -775,7 +775,7 @@ static struct file *mmap_singleton(struct drm_i915_private *i915) struct file *file; rcu_read_lock(); - file = i915->gem.mmap_singleton; + file = READ_ONCE(i915->gem.mmap_singleton); if (file && !get_file_rcu(file)) file = NULL; rcu_read_unlock(); diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.h b/drivers/gpu/drm/i915/gem/i915_gem_object.h index e44a2f40b520..2faa481cc18f 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_object.h +++ b/drivers/gpu/drm/i915/gem/i915_gem_object.h @@ -11,8 +11,6 @@ #include <drm/drm_file.h> #include <drm/drm_device.h> -#include <drm/i915_drm.h> - #include "display/intel_frontbuffer.h" #include "i915_gem_object_types.h" #include "i915_gem_gtt.h" diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object_blt.c b/drivers/gpu/drm/i915/gem/i915_gem_object_blt.c index 70809d8897cd..e00792158f13 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_object_blt.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_object_blt.c @@ -186,7 +186,7 @@ int i915_gem_object_fill_blt(struct drm_i915_gem_object *obj, 0); out_request: if (unlikely(err)) - i915_request_skip(rq, err); + i915_request_set_error_once(rq, err); i915_request_add(rq); out_batch: @@ -196,6 +196,17 @@ out_unpin: return err; } +/* Wa_1209644611:icl,ehl */ +static bool wa_1209644611_applies(struct drm_i915_private *i915, u32 size) +{ + u32 height = size >> PAGE_SHIFT; + + if (!IS_GEN(i915, 11)) + return false; + + return height % 4 == 3 && height <= 8; +} + struct i915_vma *intel_emit_vma_copy_blt(struct intel_context *ce, struct i915_vma *src, struct i915_vma *dst) @@ -237,7 +248,8 @@ struct i915_vma *intel_emit_vma_copy_blt(struct intel_context *ce, size = min_t(u64, rem, block_size); GEM_BUG_ON(size >> PAGE_SHIFT > S16_MAX); - if (INTEL_GEN(i915) >= 9) { + if (INTEL_GEN(i915) >= 9 && + !wa_1209644611_applies(i915, size)) { *cmd++ = GEN9_XY_FAST_COPY_BLT_CMD | (10 - 2); *cmd++ = BLT_DEPTH_32 | PAGE_SIZE; *cmd++ = 0; @@ -385,7 +397,7 @@ out_unlock: drm_gem_unlock_reservations(objs, ARRAY_SIZE(objs), &acquire); out_request: if (unlikely(err)) - i915_request_skip(rq, err); + i915_request_set_error_once(rq, err); i915_request_add(rq); out_batch: diff --git a/drivers/gpu/drm/i915/gem/i915_gem_phys.c b/drivers/gpu/drm/i915/gem/i915_gem_phys.c index b07bb40edd5a..698e22420dc5 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_phys.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_phys.c @@ -194,10 +194,11 @@ int i915_gem_object_attach_phys(struct drm_i915_gem_object *obj, int align) /* Perma-pin (until release) the physical set of pages */ __i915_gem_object_pin_pages(obj); - if (!IS_ERR_OR_NULL(pages)) { + if (!IS_ERR_OR_NULL(pages)) i915_gem_shmem_ops.put_pages(obj, pages); - i915_gem_object_release_memory_region(obj); - } + + i915_gem_object_release_memory_region(obj); + mutex_unlock(&obj->mm.lock); return 0; diff --git a/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c b/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c index 830d3f96e1f6..03e5eb4c99d1 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c @@ -12,7 +12,6 @@ #include <linux/pci.h> #include <linux/dma-buf.h> #include <linux/vmalloc.h> -#include <drm/i915_drm.h> #include "i915_trace.h" diff --git a/drivers/gpu/drm/i915/gem/i915_gem_stolen.c b/drivers/gpu/drm/i915/gem/i915_gem_stolen.c index 491cfbaaa330..5557dfa83a7b 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_stolen.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_stolen.c @@ -13,6 +13,7 @@ #include "gem/i915_gem_region.h" #include "i915_drv.h" #include "i915_gem_stolen.h" +#include "i915_vgpu.h" /* * The BIOS typically reserves some of the system's memory for the exclusive diff --git a/drivers/gpu/drm/i915/gem/i915_gem_tiling.c b/drivers/gpu/drm/i915/gem/i915_gem_tiling.c index 6c7825a2dc2a..37f77aee1212 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_tiling.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_tiling.c @@ -6,7 +6,6 @@ #include <linux/string.h> #include <linux/bitops.h> -#include <drm/i915_drm.h> #include "i915_drv.h" #include "i915_gem.h" diff --git a/drivers/gpu/drm/i915/gem/i915_gem_userptr.c b/drivers/gpu/drm/i915/gem/i915_gem_userptr.c index 63ead7a2b64a..7ffd7afeb7a5 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_userptr.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_userptr.c @@ -10,8 +10,6 @@ #include <linux/swap.h> #include <linux/sched/mm.h> -#include <drm/i915_drm.h> - #include "i915_drv.h" #include "i915_gem_ioctls.h" #include "i915_gem_object.h" diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c index 375d864736f3..54b86cf7f5d2 100644 --- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c +++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c @@ -1004,7 +1004,7 @@ emit_rpcs_query(struct drm_i915_gem_object *obj, return 0; skip_request: - i915_request_skip(rq, err); + i915_request_set_error_once(rq, err); err_request: i915_request_add(rq); err_batch: @@ -1559,7 +1559,7 @@ static int write_to_scratch(struct i915_gem_context *ctx, goto out_vm; skip_request: - i915_request_skip(rq, err); + i915_request_set_error_once(rq, err); err_request: i915_request_add(rq); err_unpin: @@ -1708,7 +1708,7 @@ static int read_from_scratch(struct i915_gem_context *ctx, goto out_vm; skip_request: - i915_request_skip(rq, err); + i915_request_set_error_once(rq, err); err_request: i915_request_add(rq); err_unpin: @@ -1809,7 +1809,6 @@ static int igt_vm_isolation(void *arg) vm_total = ctx_vm(ctx_a)->total; GEM_BUG_ON(ctx_vm(ctx_b)->total != vm_total); - vm_total -= I915_GTT_PAGE_SIZE; count = 0; num_engines = 0; @@ -1828,10 +1827,10 @@ static int igt_vm_isolation(void *arg) u32 value = 0xc5c5c5c5; u64 offset; - div64_u64_rem(i915_prandom_u64_state(&prng), - vm_total, &offset); - offset = round_down(offset, alignof_dword); - offset += I915_GTT_PAGE_SIZE; + /* Leave enough space at offset 0 for the batch */ + offset = igt_random_offset(&prng, + I915_GTT_PAGE_SIZE, vm_total, + sizeof(u32), alignof_dword); err = write_to_scratch(ctx_a, engine, offset, 0xdeadbeef); diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c index ef7c74cff28a..43912e9b683d 100644 --- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c +++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c @@ -570,7 +570,7 @@ static bool assert_mmap_offset(struct drm_i915_private *i915, obj = i915_gem_object_create_internal(i915, size); if (IS_ERR(obj)) - return PTR_ERR(obj); + return false; mmo = mmap_offset_attach(obj, I915_MMAP_OFFSET_GTT, NULL); i915_gem_object_put(obj); diff --git a/drivers/gpu/drm/i915/gem/selftests/igt_gem_utils.c b/drivers/gpu/drm/i915/gem/selftests/igt_gem_utils.c index 6718da20f35d..772d8cba7da9 100644 --- a/drivers/gpu/drm/i915/gem/selftests/igt_gem_utils.c +++ b/drivers/gpu/drm/i915/gem/selftests/igt_gem_utils.c @@ -159,7 +159,7 @@ int igt_gpu_fill_dw(struct intel_context *ce, return 0; skip_request: - i915_request_skip(rq, err); + i915_request_set_error_once(rq, err); err_request: i915_request_add(rq); err_batch: diff --git a/drivers/gpu/drm/i915/gem/selftests/mock_context.c b/drivers/gpu/drm/i915/gem/selftests/mock_context.c index b12ea1daa29d..e7e3c620f542 100644 --- a/drivers/gpu/drm/i915/gem/selftests/mock_context.c +++ b/drivers/gpu/drm/i915/gem/selftests/mock_context.c @@ -23,6 +23,9 @@ mock_context(struct drm_i915_private *i915, INIT_LIST_HEAD(&ctx->link); ctx->i915 = i915; + spin_lock_init(&ctx->stale.lock); + INIT_LIST_HEAD(&ctx->stale.engines); + i915_gem_context_set_persistence(ctx); mutex_init(&ctx->engines_mutex); diff --git a/drivers/gpu/drm/i915/gt/gen7_renderclear.c b/drivers/gpu/drm/i915/gt/gen7_renderclear.c new file mode 100644 index 000000000000..de595b66a746 --- /dev/null +++ b/drivers/gpu/drm/i915/gt/gen7_renderclear.c @@ -0,0 +1,402 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright © 2019 Intel Corporation + */ + +#include "gen7_renderclear.h" +#include "i915_drv.h" +#include "intel_gpu_commands.h" + +#define MAX_URB_ENTRIES 64 +#define STATE_SIZE (4 * 1024) +#define GT3_INLINE_DATA_DELAYS 0x1E00 +#define batch_advance(Y, CS) GEM_BUG_ON((Y)->end != (CS)) + +struct cb_kernel { + const void *data; + u32 size; +}; + +#define CB_KERNEL(name) { .data = (name), .size = sizeof(name) } + +#include "ivb_clear_kernel.c" +static const struct cb_kernel cb_kernel_ivb = CB_KERNEL(ivb_clear_kernel); + +#include "hsw_clear_kernel.c" +static const struct cb_kernel cb_kernel_hsw = CB_KERNEL(hsw_clear_kernel); + +struct batch_chunk { + struct i915_vma *vma; + u32 offset; + u32 *start; + u32 *end; + u32 max_items; +}; + +struct batch_vals { + u32 max_primitives; + u32 max_urb_entries; + u32 cmd_size; + u32 state_size; + u32 state_start; + u32 batch_size; + u32 surface_height; + u32 surface_width; + u32 scratch_size; + u32 max_size; +}; + +static void +batch_get_defaults(struct drm_i915_private *i915, struct batch_vals *bv) +{ + if (IS_HASWELL(i915)) { + bv->max_primitives = 280; + bv->max_urb_entries = MAX_URB_ENTRIES; + bv->surface_height = 16 * 16; + bv->surface_width = 32 * 2 * 16; + } else { + bv->max_primitives = 128; + bv->max_urb_entries = MAX_URB_ENTRIES / 2; + bv->surface_height = 16 * 8; + bv->surface_width = 32 * 16; + } + bv->cmd_size = bv->max_primitives * 4096; + bv->state_size = STATE_SIZE; + bv->state_start = bv->cmd_size; + bv->batch_size = bv->cmd_size + bv->state_size; + bv->scratch_size = bv->surface_height * bv->surface_width; + bv->max_size = bv->batch_size + bv->scratch_size; +} + +static void batch_init(struct batch_chunk *bc, + struct i915_vma *vma, + u32 *start, u32 offset, u32 max_bytes) +{ + bc->vma = vma; + bc->offset = offset; + bc->start = start + bc->offset / sizeof(*bc->start); + bc->end = bc->start; + bc->max_items = max_bytes / sizeof(*bc->start); +} + +static u32 batch_offset(const struct batch_chunk *bc, u32 *cs) +{ + return (cs - bc->start) * sizeof(*bc->start) + bc->offset; +} + +static u32 batch_addr(const struct batch_chunk *bc) +{ + return bc->vma->node.start; +} + +static void batch_add(struct batch_chunk *bc, const u32 d) +{ + GEM_BUG_ON((bc->end - bc->start) >= bc->max_items); + *bc->end++ = d; +} + +static u32 *batch_alloc_items(struct batch_chunk *bc, u32 align, u32 items) +{ + u32 *map; + + if (align) { + u32 *end = PTR_ALIGN(bc->end, align); + + memset32(bc->end, 0, end - bc->end); + bc->end = end; + } + + map = bc->end; + bc->end += items; + + return map; +} + +static u32 *batch_alloc_bytes(struct batch_chunk *bc, u32 align, u32 bytes) +{ + GEM_BUG_ON(!IS_ALIGNED(bytes, sizeof(*bc->start))); + return batch_alloc_items(bc, align, bytes / sizeof(*bc->start)); +} + +static u32 +gen7_fill_surface_state(struct batch_chunk *state, + const u32 dst_offset, + const struct batch_vals *bv) +{ + u32 surface_h = bv->surface_height; + u32 surface_w = bv->surface_width; + u32 *cs = batch_alloc_items(state, 32, 8); + u32 offset = batch_offset(state, cs); + +#define SURFACE_2D 1 +#define SURFACEFORMAT_B8G8R8A8_UNORM 0x0C0 +#define RENDER_CACHE_READ_WRITE 1 + + *cs++ = SURFACE_2D << 29 | + (SURFACEFORMAT_B8G8R8A8_UNORM << 18) | + (RENDER_CACHE_READ_WRITE << 8); + + *cs++ = batch_addr(state) + dst_offset; + + *cs++ = ((surface_h / 4 - 1) << 16) | (surface_w / 4 - 1); + *cs++ = surface_w; + *cs++ = 0; + *cs++ = 0; + *cs++ = 0; +#define SHADER_CHANNELS(r, g, b, a) \ + (((r) << 25) | ((g) << 22) | ((b) << 19) | ((a) << 16)) + *cs++ = SHADER_CHANNELS(4, 5, 6, 7); + batch_advance(state, cs); + + return offset; +} + +static u32 +gen7_fill_binding_table(struct batch_chunk *state, + const struct batch_vals *bv) +{ + u32 surface_start = gen7_fill_surface_state(state, bv->batch_size, bv); + u32 *cs = batch_alloc_items(state, 32, 8); + u32 offset = batch_offset(state, cs); + + *cs++ = surface_start - state->offset; + *cs++ = 0; + *cs++ = 0; + *cs++ = 0; + *cs++ = 0; + *cs++ = 0; + *cs++ = 0; + *cs++ = 0; + batch_advance(state, cs); + + return offset; +} + +static u32 +gen7_fill_kernel_data(struct batch_chunk *state, + const u32 *data, + const u32 size) +{ + return batch_offset(state, + memcpy(batch_alloc_bytes(state, 64, size), + data, size)); +} + +static u32 +gen7_fill_interface_descriptor(struct batch_chunk *state, + const struct batch_vals *bv, + const struct cb_kernel *kernel, + unsigned int count) +{ + u32 kernel_offset = + gen7_fill_kernel_data(state, kernel->data, kernel->size); + u32 binding_table = gen7_fill_binding_table(state, bv); + u32 *cs = batch_alloc_items(state, 32, 8 * count); + u32 offset = batch_offset(state, cs); + + *cs++ = kernel_offset; + *cs++ = (1 << 7) | (1 << 13); + *cs++ = 0; + *cs++ = (binding_table - state->offset) | 1; + *cs++ = 0; + *cs++ = 0; + *cs++ = 0; + *cs++ = 0; + + /* 1 - 63dummy idds */ + memset32(cs, 0x00, (count - 1) * 8); + batch_advance(state, cs + (count - 1) * 8); + + return offset; +} + +static void +gen7_emit_state_base_address(struct batch_chunk *batch, + u32 surface_state_base) +{ + u32 *cs = batch_alloc_items(batch, 0, 12); + + *cs++ = STATE_BASE_ADDRESS | (12 - 2); + /* general */ + *cs++ = batch_addr(batch) | BASE_ADDRESS_MODIFY; + /* surface */ + *cs++ = batch_addr(batch) | surface_state_base | BASE_ADDRESS_MODIFY; + /* dynamic */ + *cs++ = batch_addr(batch) | BASE_ADDRESS_MODIFY; + /* indirect */ + *cs++ = batch_addr(batch) | BASE_ADDRESS_MODIFY; + /* instruction */ + *cs++ = batch_addr(batch) | BASE_ADDRESS_MODIFY; + + /* general/dynamic/indirect/instruction access Bound */ + *cs++ = 0; + *cs++ = BASE_ADDRESS_MODIFY; + *cs++ = 0; + *cs++ = BASE_ADDRESS_MODIFY; + *cs++ = 0; + *cs++ = 0; + batch_advance(batch, cs); +} + +static void +gen7_emit_vfe_state(struct batch_chunk *batch, + const struct batch_vals *bv, + u32 urb_size, u32 curbe_size, + u32 mode) +{ + u32 urb_entries = bv->max_urb_entries; + u32 threads = bv->max_primitives - 1; + u32 *cs = batch_alloc_items(batch, 32, 8); + + *cs++ = MEDIA_VFE_STATE | (8 - 2); + + /* scratch buffer */ + *cs++ = 0; + + /* number of threads & urb entries for GPGPU vs Media Mode */ + *cs++ = threads << 16 | urb_entries << 8 | mode << 2; + + *cs++ = 0; + + /* urb entry size & curbe size in 256 bits unit */ + *cs++ = urb_size << 16 | curbe_size; + + /* scoreboard */ + *cs++ = 0; + *cs++ = 0; + *cs++ = 0; + batch_advance(batch, cs); +} + +static void +gen7_emit_interface_descriptor_load(struct batch_chunk *batch, + const u32 interface_descriptor, + unsigned int count) +{ + u32 *cs = batch_alloc_items(batch, 8, 4); + + *cs++ = MEDIA_INTERFACE_DESCRIPTOR_LOAD | (4 - 2); + *cs++ = 0; + *cs++ = count * 8 * sizeof(*cs); + + /* + * interface descriptor address - it is relative to the dynamics base + * address + */ + *cs++ = interface_descriptor; + batch_advance(batch, cs); +} + +static void +gen7_emit_media_object(struct batch_chunk *batch, + unsigned int media_object_index) +{ + unsigned int x_offset = (media_object_index % 16) * 64; + unsigned int y_offset = (media_object_index / 16) * 16; + unsigned int inline_data_size; + unsigned int media_batch_size; + unsigned int i; + u32 *cs; + + inline_data_size = 112 * 8; + media_batch_size = inline_data_size + 6; + + cs = batch_alloc_items(batch, 8, media_batch_size); + + *cs++ = MEDIA_OBJECT | (media_batch_size - 2); + + /* interface descriptor offset */ + *cs++ = 0; + + /* without indirect data */ + *cs++ = 0; + *cs++ = 0; + + /* scoreboard */ + *cs++ = 0; + *cs++ = 0; + + /* inline */ + *cs++ = (y_offset << 16) | (x_offset); + *cs++ = 0; + *cs++ = GT3_INLINE_DATA_DELAYS; + for (i = 3; i < inline_data_size; i++) + *cs++ = 0; + + batch_advance(batch, cs); +} + +static void gen7_emit_pipeline_flush(struct batch_chunk *batch) +{ + u32 *cs = batch_alloc_items(batch, 0, 5); + + *cs++ = GFX_OP_PIPE_CONTROL(5); + *cs++ = PIPE_CONTROL_STATE_CACHE_INVALIDATE | + PIPE_CONTROL_GLOBAL_GTT_IVB; + *cs++ = 0; + *cs++ = 0; + *cs++ = 0; + batch_advance(batch, cs); +} + +static void emit_batch(struct i915_vma * const vma, + u32 *start, + const struct batch_vals *bv) +{ + struct drm_i915_private *i915 = vma->vm->i915; + unsigned int desc_count = 64; + const u32 urb_size = 112; + struct batch_chunk cmds, state; + u32 interface_descriptor; + unsigned int i; + + batch_init(&cmds, vma, start, 0, bv->cmd_size); + batch_init(&state, vma, start, bv->state_start, bv->state_size); + + interface_descriptor = + gen7_fill_interface_descriptor(&state, bv, + IS_HASWELL(i915) ? + &cb_kernel_hsw : + &cb_kernel_ivb, + desc_count); + gen7_emit_pipeline_flush(&cmds); + batch_add(&cmds, PIPELINE_SELECT | PIPELINE_SELECT_MEDIA); + batch_add(&cmds, MI_NOOP); + gen7_emit_state_base_address(&cmds, interface_descriptor); + gen7_emit_pipeline_flush(&cmds); + + gen7_emit_vfe_state(&cmds, bv, urb_size - 1, 0, 0); + + gen7_emit_interface_descriptor_load(&cmds, + interface_descriptor, + desc_count); + + for (i = 0; i < bv->max_primitives; i++) + gen7_emit_media_object(&cmds, i); + + batch_add(&cmds, MI_BATCH_BUFFER_END); +} + +int gen7_setup_clear_gpr_bb(struct intel_engine_cs * const engine, + struct i915_vma * const vma) +{ + struct batch_vals bv; + u32 *batch; + + batch_get_defaults(engine->i915, &bv); + if (!vma) + return bv.max_size; + + GEM_BUG_ON(vma->obj->base.size < bv.max_size); + + batch = i915_gem_object_pin_map(vma->obj, I915_MAP_WC); + if (IS_ERR(batch)) + return PTR_ERR(batch); + + emit_batch(vma, memset(batch, 0, bv.max_size), &bv); + + i915_gem_object_flush_map(vma->obj); + i915_gem_object_unpin_map(vma->obj); + + return 0; +} diff --git a/drivers/gpu/drm/i915/gt/gen7_renderclear.h b/drivers/gpu/drm/i915/gt/gen7_renderclear.h new file mode 100644 index 000000000000..bb100748e2c6 --- /dev/null +++ b/drivers/gpu/drm/i915/gt/gen7_renderclear.h @@ -0,0 +1,15 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2019 Intel Corporation + */ + +#ifndef __GEN7_RENDERCLEAR_H__ +#define __GEN7_RENDERCLEAR_H__ + +struct intel_engine_cs; +struct i915_vma; + +int gen7_setup_clear_gpr_bb(struct intel_engine_cs * const engine, + struct i915_vma * const vma); + +#endif /* __GEN7_RENDERCLEAR_H__ */ diff --git a/drivers/gpu/drm/i915/gt/gen8_ppgtt.c b/drivers/gpu/drm/i915/gt/gen8_ppgtt.c index 4d1de2d97d5c..94e746af8926 100644 --- a/drivers/gpu/drm/i915/gt/gen8_ppgtt.c +++ b/drivers/gpu/drm/i915/gt/gen8_ppgtt.c @@ -8,6 +8,7 @@ #include "gen8_ppgtt.h" #include "i915_scatterlist.h" #include "i915_trace.h" +#include "i915_pvinfo.h" #include "i915_vgpu.h" #include "intel_gt.h" #include "intel_gtt.h" @@ -25,6 +26,30 @@ static u64 gen8_pde_encode(const dma_addr_t addr, return pde; } +static u64 gen8_pte_encode(dma_addr_t addr, + enum i915_cache_level level, + u32 flags) +{ + gen8_pte_t pte = addr | _PAGE_PRESENT | _PAGE_RW; + + if (unlikely(flags & PTE_READ_ONLY)) + pte &= ~_PAGE_RW; + + switch (level) { + case I915_CACHE_NONE: + pte |= PPAT_UNCACHED; + break; + case I915_CACHE_WT: + pte |= PPAT_DISPLAY_ELLC; + break; + default: + pte |= PPAT_CACHED; + break; + } + + return pte; +} + static void gen8_ppgtt_notify_vgt(struct i915_ppgtt *ppgtt, bool create) { struct drm_i915_private *i915 = ppgtt->vm.i915; @@ -706,6 +731,8 @@ struct i915_ppgtt *gen8_ppgtt_create(struct intel_gt *gt) ppgtt->vm.allocate_va_range = gen8_ppgtt_alloc; ppgtt->vm.clear_range = gen8_ppgtt_clear; + ppgtt->vm.pte_encode = gen8_pte_encode; + if (intel_vgpu_active(gt->i915)) gen8_ppgtt_notify_vgt(ppgtt, true); diff --git a/drivers/gpu/drm/i915/gt/hsw_clear_kernel.c b/drivers/gpu/drm/i915/gt/hsw_clear_kernel.c new file mode 100644 index 000000000000..b47f9d4a0848 --- /dev/null +++ b/drivers/gpu/drm/i915/gt/hsw_clear_kernel.c @@ -0,0 +1,61 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright © 2020 Intel Corporation + * + * Generated by: IGT Gpu Tools on Fri 21 Feb 2020 05:30:13 AM UTC + */ + +static const u32 hsw_clear_kernel[] = { + 0x00000001, 0x26020128, 0x00000024, 0x00000000, + 0x00000040, 0x20280c21, 0x00000028, 0x00000001, + 0x01000010, 0x20000c20, 0x0000002c, 0x00000000, + 0x00010220, 0x34001c00, 0x00001400, 0x00000160, + 0x00600001, 0x20600061, 0x00000000, 0x00000000, + 0x00000008, 0x20601c85, 0x00000e00, 0x0000000c, + 0x00000005, 0x20601ca5, 0x00000060, 0x00000001, + 0x00000008, 0x20641c85, 0x00000e00, 0x0000000d, + 0x00000005, 0x20641ca5, 0x00000064, 0x00000003, + 0x00000041, 0x207424a5, 0x00000064, 0x00000034, + 0x00000040, 0x206014a5, 0x00000060, 0x00000074, + 0x00000008, 0x20681c85, 0x00000e00, 0x00000008, + 0x00000005, 0x20681ca5, 0x00000068, 0x0000000f, + 0x00000041, 0x20701ca5, 0x00000060, 0x00000010, + 0x00000040, 0x206814a5, 0x00000068, 0x00000070, + 0x00600001, 0x20a00061, 0x00000000, 0x00000000, + 0x00000005, 0x206c1c85, 0x00000e00, 0x00000007, + 0x00000041, 0x206c1ca5, 0x0000006c, 0x00000004, + 0x00600001, 0x20800021, 0x008d0000, 0x00000000, + 0x00000001, 0x20800021, 0x0000006c, 0x00000000, + 0x00000001, 0x20840021, 0x00000068, 0x00000000, + 0x00000001, 0x20880061, 0x00000000, 0x00000003, + 0x00000005, 0x208c0d21, 0x00000086, 0xffffffff, + 0x05600032, 0x20a00fa1, 0x008d0080, 0x02190001, + 0x00000040, 0x20a01ca5, 0x000000a0, 0x00000001, + 0x05600032, 0x20a00fa1, 0x008d0080, 0x040a8001, + 0x02000040, 0x20281c21, 0x00000028, 0xffffffff, + 0x00010220, 0x34001c00, 0x00001400, 0xffffffe0, + 0x00000001, 0x26020128, 0x00000024, 0x00000000, + 0x00000001, 0x220010e4, 0x00000000, 0x00000000, + 0x00000001, 0x220831ec, 0x00000000, 0x007f007f, + 0x00600001, 0x20400021, 0x008d0000, 0x00000000, + 0x00600001, 0x2fe00021, 0x008d0000, 0x00000000, + 0x00200001, 0x20400121, 0x00450020, 0x00000000, + 0x00000001, 0x20480061, 0x00000000, 0x000f000f, + 0x00000005, 0x204c0d21, 0x00000046, 0xffffffef, + 0x00800001, 0x20600061, 0x00000000, 0x00000000, + 0x00800001, 0x20800061, 0x00000000, 0x00000000, + 0x00800001, 0x20a00061, 0x00000000, 0x00000000, + 0x00800001, 0x20c00061, 0x00000000, 0x00000000, + 0x00800001, 0x20e00061, 0x00000000, 0x00000000, + 0x00800001, 0x21000061, 0x00000000, 0x00000000, + 0x00800001, 0x21200061, 0x00000000, 0x00000000, + 0x00800001, 0x21400061, 0x00000000, 0x00000000, + 0x05600032, 0x20000fa0, 0x008d0040, 0x120a8000, + 0x00000040, 0x20402d21, 0x00000020, 0x00100010, + 0x05600032, 0x20000fa0, 0x008d0040, 0x120a8000, + 0x02000040, 0x22083d8c, 0x00000208, 0xffffffff, + 0x00800001, 0xa0000109, 0x00000602, 0x00000000, + 0x00000040, 0x22001c84, 0x00000200, 0x00000020, + 0x00010220, 0x34001c00, 0x00001400, 0xffffffc0, + 0x07600032, 0x20000fa0, 0x008d0fe0, 0x82000010, +}; diff --git a/drivers/gpu/drm/i915/gt/intel_context.c b/drivers/gpu/drm/i915/gt/intel_context.c index 8bb444cda14f..aea992e46c42 100644 --- a/drivers/gpu/drm/i915/gt/intel_context.c +++ b/drivers/gpu/drm/i915/gt/intel_context.c @@ -51,6 +51,11 @@ int intel_context_alloc_state(struct intel_context *ce) return -EINTR; if (!test_bit(CONTEXT_ALLOC_BIT, &ce->flags)) { + if (intel_context_is_banned(ce)) { + err = -EIO; + goto unlock; + } + err = ce->ops->alloc(ce); if (unlikely(err)) goto unlock; @@ -92,6 +97,8 @@ int __intel_context_do_pin(struct intel_context *ce) { int err; + GEM_BUG_ON(intel_context_is_closed(ce)); + if (unlikely(!test_bit(CONTEXT_ALLOC_BIT, &ce->flags))) { err = intel_context_alloc_state(ce); if (err) diff --git a/drivers/gpu/drm/i915/gt/intel_context.h b/drivers/gpu/drm/i915/gt/intel_context.h index 18efad255124..07be021882cc 100644 --- a/drivers/gpu/drm/i915/gt/intel_context.h +++ b/drivers/gpu/drm/i915/gt/intel_context.h @@ -173,6 +173,11 @@ static inline bool intel_context_is_barrier(const struct intel_context *ce) return test_bit(CONTEXT_BARRIER_BIT, &ce->flags); } +static inline bool intel_context_is_closed(const struct intel_context *ce) +{ + return test_bit(CONTEXT_CLOSED_BIT, &ce->flags); +} + static inline bool intel_context_use_semaphores(const struct intel_context *ce) { return test_bit(CONTEXT_USE_SEMAPHORES, &ce->flags); diff --git a/drivers/gpu/drm/i915/gt/intel_context_param.c b/drivers/gpu/drm/i915/gt/intel_context_param.c new file mode 100644 index 000000000000..65dcd090245d --- /dev/null +++ b/drivers/gpu/drm/i915/gt/intel_context_param.c @@ -0,0 +1,63 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright © 2019 Intel Corporation + */ + +#include "i915_active.h" +#include "intel_context.h" +#include "intel_context_param.h" +#include "intel_ring.h" + +int intel_context_set_ring_size(struct intel_context *ce, long sz) +{ + int err; + + if (intel_context_lock_pinned(ce)) + return -EINTR; + + err = i915_active_wait(&ce->active); + if (err < 0) + goto unlock; + + if (intel_context_is_pinned(ce)) { + err = -EBUSY; /* In active use, come back later! */ + goto unlock; + } + + if (test_bit(CONTEXT_ALLOC_BIT, &ce->flags)) { + struct intel_ring *ring; + + /* Replace the existing ringbuffer */ + ring = intel_engine_create_ring(ce->engine, sz); + if (IS_ERR(ring)) { + err = PTR_ERR(ring); + goto unlock; + } + + intel_ring_put(ce->ring); + ce->ring = ring; + + /* Context image will be updated on next pin */ + } else { + ce->ring = __intel_context_ring_size(sz); + } + +unlock: + intel_context_unlock_pinned(ce); + return err; +} + +long intel_context_get_ring_size(struct intel_context *ce) +{ + long sz = (unsigned long)READ_ONCE(ce->ring); + + if (test_bit(CONTEXT_ALLOC_BIT, &ce->flags)) { + if (intel_context_lock_pinned(ce)) + return -EINTR; + + sz = ce->ring->size; + intel_context_unlock_pinned(ce); + } + + return sz; +} diff --git a/drivers/gpu/drm/i915/gt/intel_context_param.h b/drivers/gpu/drm/i915/gt/intel_context_param.h new file mode 100644 index 000000000000..f053d8633fe2 --- /dev/null +++ b/drivers/gpu/drm/i915/gt/intel_context_param.h @@ -0,0 +1,14 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2019 Intel Corporation + */ + +#ifndef INTEL_CONTEXT_PARAM_H +#define INTEL_CONTEXT_PARAM_H + +struct intel_context; + +int intel_context_set_ring_size(struct intel_context *ce, long sz); +long intel_context_get_ring_size(struct intel_context *ce); + +#endif /* INTEL_CONTEXT_PARAM_H */ diff --git a/drivers/gpu/drm/i915/gt/intel_context_types.h b/drivers/gpu/drm/i915/gt/intel_context_types.h index 11278343b9b5..07cb83a0d017 100644 --- a/drivers/gpu/drm/i915/gt/intel_context_types.h +++ b/drivers/gpu/drm/i915/gt/intel_context_types.h @@ -45,8 +45,8 @@ struct intel_context { struct intel_engine_cs *engine; struct intel_engine_cs *inflight; -#define intel_context_inflight(ce) ptr_mask_bits((ce)->inflight, 2) -#define intel_context_inflight_count(ce) ptr_unmask_bits((ce)->inflight, 2) +#define intel_context_inflight(ce) ptr_mask_bits(READ_ONCE((ce)->inflight), 2) +#define intel_context_inflight_count(ce) ptr_unmask_bits(READ_ONCE((ce)->inflight), 2) struct i915_address_space *vm; struct i915_gem_context __rcu *gem_context; @@ -62,10 +62,11 @@ struct intel_context { #define CONTEXT_BARRIER_BIT 0 #define CONTEXT_ALLOC_BIT 1 #define CONTEXT_VALID_BIT 2 -#define CONTEXT_USE_SEMAPHORES 3 -#define CONTEXT_BANNED 4 -#define CONTEXT_FORCE_SINGLE_SUBMISSION 5 -#define CONTEXT_NOPREEMPT 6 +#define CONTEXT_CLOSED_BIT 3 +#define CONTEXT_USE_SEMAPHORES 4 +#define CONTEXT_BANNED 5 +#define CONTEXT_FORCE_SINGLE_SUBMISSION 6 +#define CONTEXT_NOPREEMPT 7 u32 *lrc_reg_state; u64 lrc_desc; diff --git a/drivers/gpu/drm/i915/gt/intel_engine.h b/drivers/gpu/drm/i915/gt/intel_engine.h index 29c8c03c5caa..b469de0dd9b6 100644 --- a/drivers/gpu/drm/i915/gt/intel_engine.h +++ b/drivers/gpu/drm/i915/gt/intel_engine.h @@ -107,7 +107,20 @@ execlists_num_ports(const struct intel_engine_execlists * const execlists) static inline struct i915_request * execlists_active(const struct intel_engine_execlists *execlists) { - return *READ_ONCE(execlists->active); + struct i915_request * const *cur, * const *old, *active; + + cur = READ_ONCE(execlists->active); + smp_rmb(); /* pairs with overwrite protection in process_csb() */ + do { + old = cur; + + active = READ_ONCE(*cur); + cur = READ_ONCE(execlists->active); + + smp_rmb(); /* and complete the seqlock retry */ + } while (unlikely(cur != old)); + + return active; } static inline void diff --git a/drivers/gpu/drm/i915/gt/intel_engine_cs.c b/drivers/gpu/drm/i915/gt/intel_engine_cs.c index 119c9cb24fd4..3aa8a652c16d 100644 --- a/drivers/gpu/drm/i915/gt/intel_engine_cs.c +++ b/drivers/gpu/drm/i915/gt/intel_engine_cs.c @@ -275,6 +275,7 @@ static void intel_engine_sanitize_mmio(struct intel_engine_cs *engine) static int intel_engine_setup(struct intel_gt *gt, enum intel_engine_id id) { const struct engine_info *info = &intel_engines[id]; + struct drm_i915_private *i915 = gt->i915; struct intel_engine_cs *engine; BUILD_BUG_ON(MAX_ENGINE_CLASS >= BIT(GEN11_ENGINE_CLASS_WIDTH)); @@ -301,11 +302,11 @@ static int intel_engine_setup(struct intel_gt *gt, enum intel_engine_id id) engine->id = id; engine->legacy_idx = INVALID_ENGINE; engine->mask = BIT(id); - engine->i915 = gt->i915; + engine->i915 = i915; engine->gt = gt; engine->uncore = gt->uncore; engine->hw_id = engine->guc_id = info->hw_id; - engine->mmio_base = __engine_mmio_base(gt->i915, info->mmio_bases); + engine->mmio_base = __engine_mmio_base(i915, info->mmio_bases); engine->class = info->class; engine->instance = info->instance; @@ -313,6 +314,8 @@ static int intel_engine_setup(struct intel_gt *gt, enum intel_engine_id id) engine->props.heartbeat_interval_ms = CONFIG_DRM_I915_HEARTBEAT_INTERVAL; + engine->props.max_busywait_duration_ns = + CONFIG_DRM_I915_MAX_REQUEST_BUSYWAIT; engine->props.preempt_timeout_ms = CONFIG_DRM_I915_PREEMPT_TIMEOUT; engine->props.stop_timeout_ms = @@ -320,11 +323,15 @@ static int intel_engine_setup(struct intel_gt *gt, enum intel_engine_id id) engine->props.timeslice_duration_ms = CONFIG_DRM_I915_TIMESLICE_DURATION; + /* Override to uninterruptible for OpenCL workloads. */ + if (INTEL_GEN(i915) == 12 && engine->class == RENDER_CLASS) + engine->props.preempt_timeout_ms = 0; + engine->context_size = intel_engine_context_size(gt, engine->class); if (WARN_ON(engine->context_size > BIT(20))) engine->context_size = 0; if (engine->context_size) - DRIVER_CAPS(gt->i915)->has_logical_contexts = true; + DRIVER_CAPS(i915)->has_logical_contexts = true; /* Nothing to do here, execute in order of dependencies */ engine->schedule = NULL; @@ -340,7 +347,7 @@ static int intel_engine_setup(struct intel_gt *gt, enum intel_engine_id id) gt->engine_class[info->class][info->instance] = engine; gt->engine[id] = engine; - gt->i915->engine[id] = engine; + i915->engine[id] = engine; return 0; } @@ -1379,24 +1386,24 @@ static void intel_engine_print_registers(struct intel_engine_cs *engine, char hdr[160]; int len; - len = snprintf(hdr, sizeof(hdr), - "\t\tActive[%d]: ", - (int)(port - execlists->active)); + len = scnprintf(hdr, sizeof(hdr), + "\t\tActive[%d]: ", + (int)(port - execlists->active)); if (!i915_request_signaled(rq)) { struct intel_timeline *tl = get_timeline(rq); - len += snprintf(hdr + len, sizeof(hdr) - len, - "ring:{start:%08x, hwsp:%08x, seqno:%08x, runtime:%llums}, ", - i915_ggtt_offset(rq->ring->vma), - tl ? tl->hwsp_offset : 0, - hwsp_seqno(rq), - DIV_ROUND_CLOSEST_ULL(intel_context_get_total_runtime_ns(rq->context), - 1000 * 1000)); + len += scnprintf(hdr + len, sizeof(hdr) - len, + "ring:{start:%08x, hwsp:%08x, seqno:%08x, runtime:%llums}, ", + i915_ggtt_offset(rq->ring->vma), + tl ? tl->hwsp_offset : 0, + hwsp_seqno(rq), + DIV_ROUND_CLOSEST_ULL(intel_context_get_total_runtime_ns(rq->context), + 1000 * 1000)); if (tl) intel_timeline_put(tl); } - snprintf(hdr + len, sizeof(hdr) - len, "rq: "); + scnprintf(hdr + len, sizeof(hdr) - len, "rq: "); print_request(m, rq, hdr); } for (port = execlists->pending; (rq = *port); port++) { diff --git a/drivers/gpu/drm/i915/gt/intel_engine_types.h b/drivers/gpu/drm/i915/gt/intel_engine_types.h index b23366a81048..80cdde712842 100644 --- a/drivers/gpu/drm/i915/gt/intel_engine_types.h +++ b/drivers/gpu/drm/i915/gt/intel_engine_types.h @@ -547,6 +547,7 @@ struct intel_engine_cs { struct { unsigned long heartbeat_interval_ms; + unsigned long max_busywait_duration_ns; unsigned long preempt_timeout_ms; unsigned long stop_timeout_ms; unsigned long timeslice_duration_ms; diff --git a/drivers/gpu/drm/i915/gt/intel_ggtt.c b/drivers/gpu/drm/i915/gt/intel_ggtt.c index 41a00281f364..aed498a0d032 100644 --- a/drivers/gpu/drm/i915/gt/intel_ggtt.c +++ b/drivers/gpu/drm/i915/gt/intel_ggtt.c @@ -8,6 +8,8 @@ #include <asm/set_memory.h> #include <asm/smp.h> +#include <drm/i915_drm.h> + #include "intel_gt.h" #include "i915_drv.h" #include "i915_scatterlist.h" @@ -157,6 +159,13 @@ static void gmch_ggtt_invalidate(struct i915_ggtt *ggtt) intel_gtt_chipset_flush(); } +static u64 gen8_ggtt_pte_encode(dma_addr_t addr, + enum i915_cache_level level, + u32 flags) +{ + return addr | _PAGE_PRESENT; +} + static void gen8_set_pte(void __iomem *addr, gen8_pte_t pte) { writeq(pte, addr); @@ -172,7 +181,7 @@ static void gen8_ggtt_insert_page(struct i915_address_space *vm, gen8_pte_t __iomem *pte = (gen8_pte_t __iomem *)ggtt->gsm + offset / I915_GTT_PAGE_SIZE; - gen8_set_pte(pte, gen8_pte_encode(addr, level, 0)); + gen8_set_pte(pte, gen8_ggtt_pte_encode(addr, level, 0)); ggtt->invalidate(ggtt); } @@ -185,7 +194,7 @@ static void gen8_ggtt_insert_entries(struct i915_address_space *vm, struct i915_ggtt *ggtt = i915_vm_to_ggtt(vm); struct sgt_iter sgt_iter; gen8_pte_t __iomem *gtt_entries; - const gen8_pte_t pte_encode = gen8_pte_encode(0, level, 0); + const gen8_pte_t pte_encode = gen8_ggtt_pte_encode(0, level, 0); dma_addr_t addr; /* @@ -857,7 +866,7 @@ static int gen8_gmch_probe(struct i915_ggtt *ggtt) ggtt->vm.vma_ops.set_pages = ggtt_set_pages; ggtt->vm.vma_ops.clear_pages = clear_pages; - ggtt->vm.pte_encode = gen8_pte_encode; + ggtt->vm.pte_encode = gen8_ggtt_pte_encode; setup_private_pat(ggtt->vm.gt->uncore); diff --git a/drivers/gpu/drm/i915/gt/intel_gpu_commands.h b/drivers/gpu/drm/i915/gt/intel_gpu_commands.h index 51b8718513bc..f04214a54f75 100644 --- a/drivers/gpu/drm/i915/gt/intel_gpu_commands.h +++ b/drivers/gpu/drm/i915/gt/intel_gpu_commands.h @@ -292,10 +292,21 @@ #define MI_STORE_URB_MEM MI_INSTR(0x2D, 0) #define MI_CONDITIONAL_BATCH_BUFFER_END MI_INSTR(0x36, 0) -#define PIPELINE_SELECT ((0x3<<29)|(0x1<<27)|(0x1<<24)|(0x4<<16)) -#define GFX_OP_3DSTATE_VF_STATISTICS ((0x3<<29)|(0x1<<27)|(0x0<<24)|(0xB<<16)) -#define MEDIA_VFE_STATE ((0x3<<29)|(0x2<<27)|(0x0<<24)|(0x0<<16)) +#define STATE_BASE_ADDRESS \ + ((0x3 << 29) | (0x0 << 27) | (0x1 << 24) | (0x1 << 16)) +#define BASE_ADDRESS_MODIFY REG_BIT(0) +#define PIPELINE_SELECT \ + ((0x3 << 29) | (0x1 << 27) | (0x1 << 24) | (0x4 << 16)) +#define PIPELINE_SELECT_MEDIA REG_BIT(0) +#define GFX_OP_3DSTATE_VF_STATISTICS \ + ((0x3 << 29) | (0x1 << 27) | (0x0 << 24) | (0xB << 16)) +#define MEDIA_VFE_STATE \ + ((0x3 << 29) | (0x2 << 27) | (0x0 << 24) | (0x0 << 16)) #define MEDIA_VFE_STATE_MMIO_ACCESS_MASK (0x18) +#define MEDIA_INTERFACE_DESCRIPTOR_LOAD \ + ((0x3 << 29) | (0x2 << 27) | (0x0 << 24) | (0x2 << 16)) +#define MEDIA_OBJECT \ + ((0x3 << 29) | (0x2 << 27) | (0x1 << 24) | (0x0 << 16)) #define GPGPU_OBJECT ((0x3<<29)|(0x2<<27)|(0x1<<24)|(0x4<<16)) #define GPGPU_WALKER ((0x3<<29)|(0x2<<27)|(0x1<<24)|(0x5<<16)) #define GFX_OP_3DSTATE_DX9_CONSTANTF_VS \ diff --git a/drivers/gpu/drm/i915/gt/intel_gt.c b/drivers/gpu/drm/i915/gt/intel_gt.c index 3dea8881e915..d09f7596cb98 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt.c +++ b/drivers/gpu/drm/i915/gt/intel_gt.c @@ -667,6 +667,9 @@ void intel_gt_driver_release(struct intel_gt *gt) void intel_gt_driver_late_release(struct intel_gt *gt) { + /* We need to wait for inflight RCU frees to release their grip */ + rcu_barrier(); + intel_uc_driver_late_release(>->uc); intel_gt_fini_requests(gt); intel_gt_fini_reset(gt); diff --git a/drivers/gpu/drm/i915/gt/intel_gt_requests.c b/drivers/gpu/drm/i915/gt/intel_gt_requests.c index 8a5054f21bf8..24c99d0838af 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt_requests.c +++ b/drivers/gpu/drm/i915/gt/intel_gt_requests.c @@ -147,24 +147,32 @@ long intel_gt_retire_requests_timeout(struct intel_gt *gt, long timeout) fence = i915_active_fence_get(&tl->last_request); if (fence) { + mutex_unlock(&tl->mutex); + timeout = dma_fence_wait_timeout(fence, interruptible, timeout); dma_fence_put(fence); + + /* Retirement is best effort */ + if (!mutex_trylock(&tl->mutex)) { + active_count++; + goto out_active; + } } } if (!retire_requests(tl) || flush_submission(gt)) active_count++; + mutex_unlock(&tl->mutex); - spin_lock(&timelines->lock); +out_active: spin_lock(&timelines->lock); - /* Resume iteration after dropping lock */ + /* Resume list iteration after reacquiring spinlock */ list_safe_reset_next(tl, tn, link); if (atomic_dec_and_test(&tl->active_count)) list_del(&tl->link); - mutex_unlock(&tl->mutex); /* Defer the final release to after the spinlock */ if (refcount_dec_and_test(&tl->kref.refcount)) { diff --git a/drivers/gpu/drm/i915/gt/intel_gtt.c b/drivers/gpu/drm/i915/gt/intel_gtt.c index bb9a6e638175..2a72cce63fd9 100644 --- a/drivers/gpu/drm/i915/gt/intel_gtt.c +++ b/drivers/gpu/drm/i915/gt/intel_gtt.c @@ -171,7 +171,9 @@ void __i915_vm_close(struct i915_address_space *vm) { struct i915_vma *vma, *vn; - mutex_lock(&vm->mutex); + if (!atomic_dec_and_mutex_lock(&vm->open, &vm->mutex)) + return; + list_for_each_entry_safe(vma, vn, &vm->bound_list, vm_link) { struct drm_i915_gem_object *obj = vma->obj; @@ -186,6 +188,7 @@ void __i915_vm_close(struct i915_address_space *vm) i915_gem_object_put(obj); } GEM_BUG_ON(!list_empty(&vm->bound_list)); + mutex_unlock(&vm->mutex); } @@ -484,30 +487,6 @@ void gtt_write_workarounds(struct intel_gt *gt) } } -u64 gen8_pte_encode(dma_addr_t addr, - enum i915_cache_level level, - u32 flags) -{ - gen8_pte_t pte = addr | _PAGE_PRESENT | _PAGE_RW; - - if (unlikely(flags & PTE_READ_ONLY)) - pte &= ~_PAGE_RW; - - switch (level) { - case I915_CACHE_NONE: - pte |= PPAT_UNCACHED; - break; - case I915_CACHE_WT: - pte |= PPAT_DISPLAY_ELLC; - break; - default: - pte |= PPAT_CACHED; - break; - } - - return pte; -} - static void tgl_setup_private_ppat(struct intel_uncore *uncore) { /* TGL doesn't support LLC or AGE settings */ diff --git a/drivers/gpu/drm/i915/gt/intel_gtt.h b/drivers/gpu/drm/i915/gt/intel_gtt.h index 23004445806a..b3116fe8d180 100644 --- a/drivers/gpu/drm/i915/gt/intel_gtt.h +++ b/drivers/gpu/drm/i915/gt/intel_gtt.h @@ -429,8 +429,7 @@ static inline void i915_vm_close(struct i915_address_space *vm) { GEM_BUG_ON(!atomic_read(&vm->open)); - if (atomic_dec_and_test(&vm->open)) - __i915_vm_close(vm); + __i915_vm_close(vm); i915_vm_put(vm); } @@ -515,10 +514,6 @@ struct i915_ppgtt *i915_ppgtt_create(struct intel_gt *gt); void i915_ggtt_suspend(struct i915_ggtt *gtt); void i915_ggtt_resume(struct i915_ggtt *ggtt); -u64 gen8_pte_encode(dma_addr_t addr, - enum i915_cache_level level, - u32 flags); - int setup_page_dma(struct i915_address_space *vm, struct i915_page_dma *p); void cleanup_page_dma(struct i915_address_space *vm, struct i915_page_dma *p); diff --git a/drivers/gpu/drm/i915/gt/intel_lrc.c b/drivers/gpu/drm/i915/gt/intel_lrc.c index 47561dc29304..683014e7bc51 100644 --- a/drivers/gpu/drm/i915/gt/intel_lrc.c +++ b/drivers/gpu/drm/i915/gt/intel_lrc.c @@ -245,7 +245,7 @@ static void mark_eio(struct i915_request *rq) GEM_BUG_ON(i915_request_signaled(rq)); - dma_fence_set_error(&rq->fence, -EIO); + i915_request_set_error_once(rq, -EIO); i915_request_mark_complete(rq); } @@ -293,7 +293,7 @@ static inline struct i915_priolist *to_priolist(struct rb_node *rb) static inline int rq_prio(const struct i915_request *rq) { - return rq->sched.attr.priority; + return READ_ONCE(rq->sched.attr.priority); } static int effective_prio(const struct i915_request *rq) @@ -1004,7 +1004,7 @@ __unwind_incomplete_requests(struct intel_engine_cs *engine) i915_request_cancel_breadcrumb(rq); spin_unlock(&rq->lock); } - rq->engine = owner; + WRITE_ONCE(rq->engine, owner); owner->submit_request(rq); active = NULL; } @@ -1316,7 +1316,7 @@ __execlists_schedule_out(struct i915_request *rq, * If we have just completed this context, the engine may now be * idle and we want to re-enter powersaving. */ - if (list_is_last(&rq->link, &ce->timeline->requests) && + if (list_is_last_rcu(&rq->link, &ce->timeline->requests) && i915_request_completed(rq)) intel_engine_add_retire(engine, ce->timeline); @@ -1448,6 +1448,7 @@ assert_pending_valid(const struct intel_engine_execlists *execlists, { struct i915_request * const *port, *rq; struct intel_context *ce = NULL; + bool sentinel = false; trace_ports(execlists, msg, execlists->pending); @@ -1481,6 +1482,26 @@ assert_pending_valid(const struct intel_engine_execlists *execlists, } ce = rq->context; + /* + * Sentinels are supposed to be lonely so they flush the + * current exection off the HW. Check that they are the + * only request in the pending submission. + */ + if (sentinel) { + GEM_TRACE_ERR("context:%llx after sentinel in pending[%zd]\n", + ce->timeline->fence_context, + port - execlists->pending); + return false; + } + + sentinel = i915_request_has_sentinel(rq); + if (sentinel && port != execlists->pending) { + GEM_TRACE_ERR("sentinel context:%llx not in prime position[%zd]\n", + ce->timeline->fence_context, + port - execlists->pending); + return false; + } + /* Hold tightly onto the lock to prevent concurrent retires! */ if (!spin_trylock_irqsave(&rq->lock, flags)) continue; @@ -1576,6 +1597,11 @@ static bool can_merge_ctx(const struct intel_context *prev, return true; } +static unsigned long i915_request_flags(const struct i915_request *rq) +{ + return READ_ONCE(rq->fence.flags); +} + static bool can_merge_rq(const struct i915_request *prev, const struct i915_request *next) { @@ -1593,7 +1619,7 @@ static bool can_merge_rq(const struct i915_request *prev, if (i915_request_completed(next)) return true; - if (unlikely((prev->fence.flags ^ next->fence.flags) & + if (unlikely((i915_request_flags(prev) ^ i915_request_flags(next)) & (BIT(I915_FENCE_FLAG_NOPREEMPT) | BIT(I915_FENCE_FLAG_SENTINEL)))) return false; @@ -1601,6 +1627,7 @@ static bool can_merge_rq(const struct i915_request *prev, if (!can_merge_ctx(prev->context, next->context)) return false; + GEM_BUG_ON(i915_seqno_passed(prev->fence.seqno, next->fence.seqno)); return true; } @@ -1636,7 +1663,7 @@ static bool virtual_matches(const struct virtual_engine *ve, } static void virtual_xfer_breadcrumbs(struct virtual_engine *ve, - struct intel_engine_cs *engine) + struct i915_request *rq) { struct intel_engine_cs *old = ve->siblings[0]; @@ -1644,22 +1671,21 @@ static void virtual_xfer_breadcrumbs(struct virtual_engine *ve, spin_lock(&old->breadcrumbs.irq_lock); if (!list_empty(&ve->context.signal_link)) { - list_move_tail(&ve->context.signal_link, - &engine->breadcrumbs.signalers); - intel_engine_signal_breadcrumbs(engine); - } - spin_unlock(&old->breadcrumbs.irq_lock); -} + list_del_init(&ve->context.signal_link); -static struct i915_request * -last_active(const struct intel_engine_execlists *execlists) -{ - struct i915_request * const *last = READ_ONCE(execlists->active); - - while (*last && i915_request_completed(*last)) - last++; + /* + * We cannot acquire the new engine->breadcrumbs.irq_lock + * (as we are holding a breadcrumbs.irq_lock already), + * so attach this request to the signaler on submission. + * The queued irq_work will occur when we finally drop + * the engine->active.lock after dequeue. + */ + set_bit(DMA_FENCE_FLAG_ENABLE_SIGNAL_BIT, &rq->fence.flags); - return *last; + /* Also transfer the pending irq_work for the old breadcrumb. */ + intel_engine_signal_breadcrumbs(rq->engine); + } + spin_unlock(&old->breadcrumbs.irq_lock); } #define for_each_waiter(p__, rq__) \ @@ -1735,11 +1761,9 @@ need_timeslice(struct intel_engine_cs *engine, const struct i915_request *rq) if (!intel_engine_has_timeslices(engine)) return false; - if (list_is_last(&rq->sched.link, &engine->active.requests)) - return false; - - hint = max(rq_prio(list_next_entry(rq, sched.link)), - engine->execlists.queue_priority_hint); + hint = engine->execlists.queue_priority_hint; + if (!list_is_last(&rq->sched.link, &engine->active.requests)) + hint = max(hint, rq_prio(list_next_entry(rq, sched.link))); return hint >= effective_prio(rq); } @@ -1762,12 +1786,13 @@ timeslice(const struct intel_engine_cs *engine) static unsigned long active_timeslice(const struct intel_engine_cs *engine) { - const struct i915_request *rq = *engine->execlists.active; + const struct intel_engine_execlists *execlists = &engine->execlists; + const struct i915_request *rq = *execlists->active; if (!rq || i915_request_completed(rq)) return 0; - if (engine->execlists.switch_priority_hint < effective_prio(rq)) + if (READ_ONCE(execlists->switch_priority_hint) < effective_prio(rq)) return 0; return timeslice(engine); @@ -1781,16 +1806,29 @@ static void set_timeslice(struct intel_engine_cs *engine) set_timer_ms(&engine->execlists.timer, active_timeslice(engine)); } +static void start_timeslice(struct intel_engine_cs *engine) +{ + struct intel_engine_execlists *execlists = &engine->execlists; + int prio = queue_prio(execlists); + + WRITE_ONCE(execlists->switch_priority_hint, prio); + if (prio == INT_MIN) + return; + + if (timer_pending(&execlists->timer)) + return; + + set_timer_ms(&execlists->timer, timeslice(engine)); +} + static void record_preemption(struct intel_engine_execlists *execlists) { (void)I915_SELFTEST_ONLY(execlists->preempt_hang.count++); } -static unsigned long active_preempt_timeout(struct intel_engine_cs *engine) +static unsigned long active_preempt_timeout(struct intel_engine_cs *engine, + const struct i915_request *rq) { - struct i915_request *rq; - - rq = last_active(&engine->execlists); if (!rq) return 0; @@ -1801,13 +1839,14 @@ static unsigned long active_preempt_timeout(struct intel_engine_cs *engine) return READ_ONCE(engine->props.preempt_timeout_ms); } -static void set_preempt_timeout(struct intel_engine_cs *engine) +static void set_preempt_timeout(struct intel_engine_cs *engine, + const struct i915_request *rq) { if (!intel_engine_has_preempt_reset(engine)) return; set_timer_ms(&engine->execlists.preempt, - active_preempt_timeout(engine)); + active_preempt_timeout(engine, rq)); } static inline void clear_ports(struct i915_request **ports, int count) @@ -1820,6 +1859,7 @@ static void execlists_dequeue(struct intel_engine_cs *engine) struct intel_engine_execlists * const execlists = &engine->execlists; struct i915_request **port = execlists->pending; struct i915_request ** const last_port = port + execlists->port_mask; + struct i915_request * const *active; struct i915_request *last; struct rb_node *rb; bool submit = false; @@ -1874,7 +1914,10 @@ static void execlists_dequeue(struct intel_engine_cs *engine) * i.e. we will retrigger preemption following the ack in case * of trouble. */ - last = last_active(execlists); + active = READ_ONCE(execlists->active); + while ((last = *active) && i915_request_completed(last)) + active++; + if (last) { if (need_preempt(engine, last, rb)) { ENGINE_TRACE(engine, @@ -1944,11 +1987,7 @@ static void execlists_dequeue(struct intel_engine_cs *engine) * Even if ELSP[1] is occupied and not worthy * of timeslices, our queue might be. */ - if (!execlists->timer.expires && - need_timeslice(engine, last)) - set_timer_ms(&execlists->timer, - timeslice(engine)); - + start_timeslice(engine); return; } } @@ -1983,7 +2022,8 @@ static void execlists_dequeue(struct intel_engine_cs *engine) if (last && !can_merge_rq(last, rq)) { spin_unlock(&ve->base.active.lock); - return; /* leave this for another */ + start_timeslice(engine); + return; /* leave this for another sibling */ } ENGINE_TRACE(engine, @@ -1995,13 +2035,14 @@ static void execlists_dequeue(struct intel_engine_cs *engine) "", yesno(engine != ve->siblings[0])); - ve->request = NULL; - ve->base.execlists.queue_priority_hint = INT_MIN; + WRITE_ONCE(ve->request, NULL); + WRITE_ONCE(ve->base.execlists.queue_priority_hint, + INT_MIN); rb_erase_cached(rb, &execlists->virtual); RB_CLEAR_NODE(rb); GEM_BUG_ON(!(rq->execution_mask & engine->mask)); - rq->engine = engine; + WRITE_ONCE(rq->engine, engine); if (engine != ve->siblings[0]) { u32 *regs = ve->context.lrc_reg_state; @@ -2014,7 +2055,7 @@ static void execlists_dequeue(struct intel_engine_cs *engine) engine); if (!list_empty(&ve->context.signals)) - virtual_xfer_breadcrumbs(ve, engine); + virtual_xfer_breadcrumbs(ve, rq); /* * Move the bound engine to the top of the list @@ -2121,6 +2162,9 @@ static void execlists_dequeue(struct intel_engine_cs *engine) GEM_BUG_ON(last && !can_merge_ctx(last->context, rq->context)); + GEM_BUG_ON(last && + i915_seqno_passed(last->fence.seqno, + rq->fence.seqno)); submit = true; last = rq; @@ -2159,7 +2203,7 @@ done: * Skip if we ended up with exactly the same set of requests, * e.g. trying to timeslice a pair of ordered contexts */ - if (!memcmp(execlists->active, execlists->pending, + if (!memcmp(active, execlists->pending, (port - execlists->pending + 1) * sizeof(*port))) { do execlists_schedule_out(fetch_and_zero(port)); @@ -2170,7 +2214,7 @@ done: clear_ports(port + 1, last_port - port); execlists_submit_ports(engine); - set_preempt_timeout(engine); + set_preempt_timeout(engine, *active); } else { skip_submit: ring_set_paused(engine, 0); @@ -2191,6 +2235,7 @@ cancel_port_requests(struct intel_engine_execlists * const execlists) execlists_schedule_out(*port); clear_ports(execlists->inflight, ARRAY_SIZE(execlists->inflight)); + smp_wmb(); /* complete the seqlock for execlists_active() */ WRITE_ONCE(execlists->active, execlists->inflight); } @@ -2345,6 +2390,7 @@ static void process_csb(struct intel_engine_cs *engine) /* Point active to the new ELSP; prevent overwriting */ WRITE_ONCE(execlists->active, execlists->pending); + smp_wmb(); /* notify execlists_active() */ /* cancel old inflight, prepare for switch */ trace_ports(execlists, "preempted", old); @@ -2352,11 +2398,12 @@ static void process_csb(struct intel_engine_cs *engine) execlists_schedule_out(*old++); /* switch pending to inflight */ - WRITE_ONCE(execlists->active, - memcpy(execlists->inflight, - execlists->pending, - execlists_num_ports(execlists) * - sizeof(*execlists->pending))); + memcpy(execlists->inflight, + execlists->pending, + execlists_num_ports(execlists) * + sizeof(*execlists->pending)); + smp_wmb(); /* complete the seqlock */ + WRITE_ONCE(execlists->active, execlists->inflight); WRITE_ONCE(execlists->pending[0], NULL); } else { @@ -2579,6 +2626,10 @@ static void __execlists_unhold(struct i915_request *rq) struct i915_request *w = container_of(p->waiter, typeof(*w), sched); + /* Propagate any change in error status */ + if (rq->fence.error) + i915_request_set_error_once(w, rq->fence.error); + if (w->engine != rq->engine) continue; @@ -2966,6 +3017,7 @@ __execlists_update_reg_state(const struct intel_context *ce, regs[CTX_RING_START] = i915_ggtt_offset(ring->vma); regs[CTX_RING_HEAD] = head; regs[CTX_RING_TAIL] = ring->tail; + regs[CTX_RING_CTL] = RING_CTL_SIZE(ring->size) | RING_VALID; /* RPCS */ if (engine->class == RENDER_CLASS) { @@ -3636,9 +3688,6 @@ static void __execlists_reset(struct intel_engine_cs *engine, bool stalled) if (!rq) goto unwind; - /* We still have requests in-flight; the engine should be active */ - GEM_BUG_ON(!intel_engine_pm_is_awake(engine)); - ce = rq->context; GEM_BUG_ON(!i915_vma_is_pinned(ce->state)); @@ -3648,8 +3697,12 @@ static void __execlists_reset(struct intel_engine_cs *engine, bool stalled) goto out_replay; } + /* We still have requests in-flight; the engine should be active */ + GEM_BUG_ON(!intel_engine_pm_is_awake(engine)); + /* Context has requests still in-flight; it should not be idle! */ GEM_BUG_ON(i915_active_is_idle(&ce->active)); + rq = active_request(ce->timeline, rq); head = intel_ring_wrap(ce->ring, rq->head); GEM_BUG_ON(head == ce->ring->tail); @@ -3723,7 +3776,10 @@ static void execlists_reset_rewind(struct intel_engine_cs *engine, bool stalled) static void nop_submission_tasklet(unsigned long data) { + struct intel_engine_cs * const engine = (struct intel_engine_cs *)data; + /* The driver is wedged; don't process any more events. */ + WRITE_ONCE(engine->execlists.queue_priority_hint, INT_MIN); } static void execlists_reset_cancel(struct intel_engine_cs *engine) @@ -4119,26 +4175,6 @@ static int gen12_emit_flush_render(struct i915_request *request, *cs++ = preparser_disable(false); intel_ring_advance(request, cs); - - /* - * Wa_1604544889:tgl - */ - if (IS_TGL_REVID(request->i915, TGL_REVID_A0, TGL_REVID_A0)) { - flags = 0; - flags |= PIPE_CONTROL_CS_STALL; - flags |= PIPE_CONTROL_HDC_PIPELINE_FLUSH; - - flags |= PIPE_CONTROL_STORE_DATA_INDEX; - flags |= PIPE_CONTROL_QW_WRITE; - - cs = intel_ring_begin(request, 6); - if (IS_ERR(cs)) - return PTR_ERR(cs); - - cs = gen8_emit_pipe_control(cs, flags, - LRC_PPHWSP_SCRATCH_ADDR); - intel_ring_advance(request, cs); - } } return 0; @@ -4877,7 +4913,7 @@ static intel_engine_mask_t virtual_submission_mask(struct virtual_engine *ve) mask = rq->execution_mask; if (unlikely(!mask)) { /* Invalid selection, submit to a random engine in error */ - i915_request_skip(rq, -ENODEV); + i915_request_set_error_once(rq, -ENODEV); mask = ve->siblings[0]->mask; } @@ -4891,7 +4927,7 @@ static intel_engine_mask_t virtual_submission_mask(struct virtual_engine *ve) static void virtual_submission_tasklet(unsigned long data) { struct virtual_engine * const ve = (struct virtual_engine *)data; - const int prio = ve->base.execlists.queue_priority_hint; + const int prio = READ_ONCE(ve->base.execlists.queue_priority_hint); intel_engine_mask_t mask; unsigned int n; @@ -5287,11 +5323,15 @@ void intel_execlists_show_requests(struct intel_engine_cs *engine, show_request(m, last, "\t\tE "); } - last = NULL; - count = 0; + if (execlists->switch_priority_hint != INT_MIN) + drm_printf(m, "\t\tSwitch priority hint: %d\n", + READ_ONCE(execlists->switch_priority_hint)); if (execlists->queue_priority_hint != INT_MIN) drm_printf(m, "\t\tQueue priority hint: %d\n", - execlists->queue_priority_hint); + READ_ONCE(execlists->queue_priority_hint)); + + last = NULL; + count = 0; for (rb = rb_first_cached(&execlists->queue); rb; rb = rb_next(rb)) { struct i915_priolist *p = rb_entry(rb, typeof(*p), node); int i; diff --git a/drivers/gpu/drm/i915/gt/intel_rc6.c b/drivers/gpu/drm/i915/gt/intel_rc6.c index bef132709854..3847ee44b181 100644 --- a/drivers/gpu/drm/i915/gt/intel_rc6.c +++ b/drivers/gpu/drm/i915/gt/intel_rc6.c @@ -7,6 +7,7 @@ #include <linux/pm_runtime.h> #include "i915_drv.h" +#include "i915_vgpu.h" #include "intel_gt.h" #include "intel_gt_pm.h" #include "intel_rc6.h" @@ -319,10 +320,10 @@ static int vlv_rc6_init(struct intel_rc6 *rc6) return PTR_ERR(pctx); } - GEM_BUG_ON(range_overflows_t(u64, - i915->dsm.start, - pctx->stolen->start, - U32_MAX)); + GEM_BUG_ON(range_overflows_end_t(u64, + i915->dsm.start, + pctx->stolen->start, + U32_MAX)); pctx_paddr = i915->dsm.start + pctx->stolen->start; intel_uncore_write(uncore, VLV_PCBR, pctx_paddr); @@ -602,6 +603,7 @@ void intel_rc6_unpark(struct intel_rc6 *rc6) void intel_rc6_park(struct intel_rc6 *rc6) { struct intel_uncore *uncore = rc6_to_uncore(rc6); + unsigned int target; if (!rc6->enabled) return; @@ -616,7 +618,14 @@ void intel_rc6_park(struct intel_rc6 *rc6) /* Turn off the HW timers and go directly to rc6 */ set(uncore, GEN6_RC_CONTROL, GEN6_RC_CTL_RC6_ENABLE); - set(uncore, GEN6_RC_STATE, 0x4 << RC_SW_TARGET_STATE_SHIFT); + + if (HAS_RC6pp(rc6_to_i915(rc6))) + target = 0x6; /* deepest rc6 */ + else if (HAS_RC6p(rc6_to_i915(rc6))) + target = 0x5; /* deep rc6 */ + else + target = 0x4; /* normal rc6 */ + set(uncore, GEN6_RC_STATE, target << RC_SW_TARGET_STATE_SHIFT); } void intel_rc6_disable(struct intel_rc6 *rc6) diff --git a/drivers/gpu/drm/i915/gt/intel_reset.c b/drivers/gpu/drm/i915/gt/intel_reset.c index aef6ab58d7d9..80db3c9d785e 100644 --- a/drivers/gpu/drm/i915/gt/intel_reset.c +++ b/drivers/gpu/drm/i915/gt/intel_reset.c @@ -48,8 +48,10 @@ static void engine_skip_context(struct i915_request *rq) lockdep_assert_held(&engine->active.lock); list_for_each_entry_continue(rq, &engine->active.requests, sched.link) - if (rq->context == hung_ctx) - i915_request_skip(rq, -EIO); + if (rq->context == hung_ctx) { + i915_request_set_error_once(rq, -EIO); + __i915_request_skip(rq); + } } static void client_mark_guilty(struct i915_gem_context *ctx, bool banned) @@ -86,19 +88,18 @@ static bool mark_guilty(struct i915_request *rq) bool banned; int i; + if (intel_context_is_closed(rq->context)) { + intel_context_set_banned(rq->context); + return true; + } + rcu_read_lock(); ctx = rcu_dereference(rq->context->gem_context); if (ctx && !kref_get_unless_zero(&ctx->ref)) ctx = NULL; rcu_read_unlock(); if (!ctx) - return false; - - if (i915_gem_context_is_closed(ctx)) { - intel_context_set_banned(rq->context); - banned = true; - goto out; - } + return intel_context_is_banned(rq->context); atomic_inc(&ctx->guilty_count); @@ -154,11 +155,12 @@ void __i915_request_reset(struct i915_request *rq, bool guilty) rcu_read_lock(); /* protect the GEM context */ if (guilty) { - i915_request_skip(rq, -EIO); + i915_request_set_error_once(rq, -EIO); + __i915_request_skip(rq); if (mark_guilty(rq)) engine_skip_context(rq); } else { - dma_fence_set_error(&rq->fence, -EAGAIN); + i915_request_set_error_once(rq, -EAGAIN); mark_innocent(rq); } rcu_read_unlock(); @@ -785,7 +787,7 @@ static void nop_submit_request(struct i915_request *request) unsigned long flags; RQ_TRACE(request, "-EIO\n"); - dma_fence_set_error(&request->fence, -EIO); + i915_request_set_error_once(request, -EIO); spin_lock_irqsave(&engine->active.lock, flags); __i915_request_submit(request); diff --git a/drivers/gpu/drm/i915/gt/intel_ring_submission.c b/drivers/gpu/drm/i915/gt/intel_ring_submission.c index f70b903a98bc..fdc3f10e12aa 100644 --- a/drivers/gpu/drm/i915/gt/intel_ring_submission.c +++ b/drivers/gpu/drm/i915/gt/intel_ring_submission.c @@ -29,11 +29,10 @@ #include <linux/log2.h> -#include <drm/i915_drm.h> - #include "gem/i915_gem_context.h" #include "gen6_ppgtt.h" +#include "gen7_renderclear.h" #include "i915_drv.h" #include "i915_trace.h" #include "intel_context.h" @@ -897,9 +896,7 @@ static void reset_cancel(struct intel_engine_cs *engine) /* Mark all submitted requests as skipped. */ list_for_each_entry(request, &engine->active.requests, sched.link) { - if (!i915_request_signaled(request)) - dma_fence_set_error(&request->fence, -EIO); - + i915_request_set_error_once(request, -EIO); i915_request_mark_complete(request); } @@ -1360,7 +1357,9 @@ static int load_pd_dir(struct i915_request *rq, return rq->engine->emit_flush(rq, EMIT_FLUSH); } -static inline int mi_set_context(struct i915_request *rq, u32 flags) +static inline int mi_set_context(struct i915_request *rq, + struct intel_context *ce, + u32 flags) { struct drm_i915_private *i915 = rq->i915; struct intel_engine_cs *engine = rq->engine; @@ -1435,7 +1434,7 @@ static inline int mi_set_context(struct i915_request *rq, u32 flags) *cs++ = MI_NOOP; *cs++ = MI_SET_CONTEXT; - *cs++ = i915_ggtt_offset(rq->context->state) | flags; + *cs++ = i915_ggtt_offset(ce->state) | flags; /* * w/a: MI_SET_CONTEXT must always be followed by MI_NOOP * WaMiSetContext_Hang:snb,ivb,vlv @@ -1550,13 +1549,56 @@ static int switch_mm(struct i915_request *rq, struct i915_address_space *vm) return rq->engine->emit_flush(rq, EMIT_INVALIDATE); } +static int clear_residuals(struct i915_request *rq) +{ + struct intel_engine_cs *engine = rq->engine; + int ret; + + ret = switch_mm(rq, vm_alias(engine->kernel_context->vm)); + if (ret) + return ret; + + if (engine->kernel_context->state) { + ret = mi_set_context(rq, + engine->kernel_context, + MI_MM_SPACE_GTT | MI_RESTORE_INHIBIT); + if (ret) + return ret; + } + + ret = engine->emit_bb_start(rq, + engine->wa_ctx.vma->node.start, 0, + 0); + if (ret) + return ret; + + ret = engine->emit_flush(rq, EMIT_FLUSH); + if (ret) + return ret; + + /* Always invalidate before the next switch_mm() */ + return engine->emit_flush(rq, EMIT_INVALIDATE); +} + static int switch_context(struct i915_request *rq) { + struct intel_engine_cs *engine = rq->engine; struct intel_context *ce = rq->context; + void **residuals = NULL; int ret; GEM_BUG_ON(HAS_EXECLISTS(rq->i915)); + if (engine->wa_ctx.vma && ce != engine->kernel_context) { + if (engine->wa_ctx.vma->private != ce) { + ret = clear_residuals(rq); + if (ret) + return ret; + + residuals = &engine->wa_ctx.vma->private; + } + } + ret = switch_mm(rq, vm_alias(ce->vm)); if (ret) return ret; @@ -1564,7 +1606,7 @@ static int switch_context(struct i915_request *rq) if (ce->state) { u32 flags; - GEM_BUG_ON(rq->engine->id != RCS0); + GEM_BUG_ON(engine->id != RCS0); /* For resource streamer on HSW+ and power context elsewhere */ BUILD_BUG_ON(HSW_MI_RS_SAVE_STATE_EN != MI_SAVE_EXT_STATE_EN); @@ -1576,7 +1618,7 @@ static int switch_context(struct i915_request *rq) else flags |= MI_RESTORE_INHIBIT; - ret = mi_set_context(rq, flags); + ret = mi_set_context(rq, ce, flags); if (ret) return ret; } @@ -1585,6 +1627,20 @@ static int switch_context(struct i915_request *rq) if (ret) return ret; + /* + * Now past the point of no return, this request _will_ be emitted. + * + * Or at least this preamble will be emitted, the request may be + * interrupted prior to submitting the user payload. If so, we + * still submit the "empty" request in order to preserve global + * state tracking such as this, our tracking of the current + * dirty context. + */ + if (residuals) { + intel_context_put(*residuals); + *residuals = intel_context_get(ce); + } + return 0; } @@ -1769,6 +1825,11 @@ static void ring_release(struct intel_engine_cs *engine) intel_engine_cleanup_common(engine); + if (engine->wa_ctx.vma) { + intel_context_put(engine->wa_ctx.vma->private); + i915_vma_unpin_and_release(&engine->wa_ctx.vma, 0); + } + intel_ring_unpin(engine->legacy.ring); intel_ring_put(engine->legacy.ring); @@ -1916,6 +1977,64 @@ static void setup_vecs(struct intel_engine_cs *engine) engine->emit_fini_breadcrumb = gen7_xcs_emit_breadcrumb; } +static int gen7_ctx_switch_bb_setup(struct intel_engine_cs * const engine, + struct i915_vma * const vma) +{ + return gen7_setup_clear_gpr_bb(engine, vma); +} + +static int gen7_ctx_switch_bb_init(struct intel_engine_cs *engine) +{ + struct drm_i915_gem_object *obj; + struct i915_vma *vma; + int size; + int err; + + size = gen7_ctx_switch_bb_setup(engine, NULL /* probe size */); + if (size <= 0) + return size; + + size = ALIGN(size, PAGE_SIZE); + obj = i915_gem_object_create_internal(engine->i915, size); + if (IS_ERR(obj)) + return PTR_ERR(obj); + + vma = i915_vma_instance(obj, engine->gt->vm, NULL); + if (IS_ERR(vma)) { + err = PTR_ERR(vma); + goto err_obj; + } + + vma->private = intel_context_create(engine); /* dummy residuals */ + if (IS_ERR(vma->private)) { + err = PTR_ERR(vma->private); + goto err_obj; + } + + err = i915_vma_pin(vma, 0, 0, PIN_USER | PIN_HIGH); + if (err) + goto err_private; + + err = i915_vma_sync(vma); + if (err) + goto err_unpin; + + err = gen7_ctx_switch_bb_setup(engine, vma); + if (err) + goto err_unpin; + + engine->wa_ctx.vma = vma; + return 0; + +err_unpin: + i915_vma_unpin(vma); +err_private: + intel_context_put(vma->private); +err_obj: + i915_gem_object_put(obj); + return err; +} + int intel_ring_submission_setup(struct intel_engine_cs *engine) { struct intel_timeline *timeline; @@ -1969,11 +2088,19 @@ int intel_ring_submission_setup(struct intel_engine_cs *engine) GEM_BUG_ON(timeline->hwsp_ggtt != engine->status_page.vma); + if (IS_HASWELL(engine->i915) && engine->class == RENDER_CLASS) { + err = gen7_ctx_switch_bb_init(engine); + if (err) + goto err_ring_unpin; + } + /* Finally, take ownership and responsibility for cleanup! */ engine->release = ring_release; return 0; +err_ring_unpin: + intel_ring_unpin(ring); err_ring: intel_ring_put(ring); err_timeline_unpin: @@ -1984,3 +2111,7 @@ err: intel_engine_cleanup_common(engine); return err; } + +#if IS_ENABLED(CONFIG_DRM_I915_SELFTEST) +#include "selftest_ring_submission.c" +#endif diff --git a/drivers/gpu/drm/i915/gt/intel_rps.c b/drivers/gpu/drm/i915/gt/intel_rps.c index 30ae29b30f11..cfaf141bac4d 100644 --- a/drivers/gpu/drm/i915/gt/intel_rps.c +++ b/drivers/gpu/drm/i915/gt/intel_rps.c @@ -4,6 +4,8 @@ * Copyright © 2019 Intel Corporation */ +#include <drm/i915_drm.h> + #include "i915_drv.h" #include "intel_gt.h" #include "intel_gt_irq.h" @@ -55,7 +57,7 @@ static u32 rps_pm_mask(struct intel_rps *rps, u8 val) if (val < rps->max_freq_softlimit) mask |= GEN6_PM_RP_UP_EI_EXPIRED | GEN6_PM_RP_UP_THRESHOLD; - mask &= rps->pm_events; + mask &= READ_ONCE(rps->pm_events); return rps_pm_sanitize_mask(rps, ~mask); } @@ -68,17 +70,19 @@ static void rps_reset_ei(struct intel_rps *rps) static void rps_enable_interrupts(struct intel_rps *rps) { struct intel_gt *gt = rps_to_gt(rps); + u32 events; rps_reset_ei(rps); if (IS_VALLEYVIEW(gt->i915)) /* WaGsvRC0ResidencyMethod:vlv */ - rps->pm_events = GEN6_PM_RP_UP_EI_EXPIRED; + events = GEN6_PM_RP_UP_EI_EXPIRED; else - rps->pm_events = (GEN6_PM_RP_UP_THRESHOLD | - GEN6_PM_RP_DOWN_THRESHOLD | - GEN6_PM_RP_DOWN_TIMEOUT); + events = (GEN6_PM_RP_UP_THRESHOLD | + GEN6_PM_RP_DOWN_THRESHOLD | + GEN6_PM_RP_DOWN_TIMEOUT); + WRITE_ONCE(rps->pm_events, events); spin_lock_irq(>->irq_lock); gen6_gt_pm_enable_irq(gt, rps->pm_events); spin_unlock_irq(>->irq_lock); @@ -115,8 +119,7 @@ static void rps_disable_interrupts(struct intel_rps *rps) { struct intel_gt *gt = rps_to_gt(rps); - rps->pm_events = 0; - + WRITE_ONCE(rps->pm_events, 0); set(gt->uncore, GEN6_PMINTRMSK, rps_pm_sanitize_mask(rps, ~0u)); spin_lock_irq(>->irq_lock); @@ -642,7 +645,7 @@ void intel_rps_mark_interactive(struct intel_rps *rps, bool interactive) { mutex_lock(&rps->power.mutex); if (interactive) { - if (!rps->power.interactive++ && rps->active) + if (!rps->power.interactive++ && READ_ONCE(rps->active)) rps_set_power(rps, HIGH_POWER); } else { GEM_BUG_ON(!rps->power.interactive); @@ -719,11 +722,15 @@ void intel_rps_unpark(struct intel_rps *rps) * performance, jump directly to RPe as our starting frequency. */ mutex_lock(&rps->lock); - rps->active = true; + + WRITE_ONCE(rps->active, true); + freq = max(rps->cur_freq, rps->efficient_freq), freq = clamp(freq, rps->min_freq_softlimit, rps->max_freq_softlimit); intel_rps_set(rps, freq); + rps->last_adj = 0; + mutex_unlock(&rps->lock); if (INTEL_GEN(rps_to_i915(rps)) >= 6) @@ -743,7 +750,7 @@ void intel_rps_park(struct intel_rps *rps) if (INTEL_GEN(i915) >= 6) rps_disable_interrupts(rps); - rps->active = false; + WRITE_ONCE(rps->active, false); if (rps->last_freq <= rps->idle_freq) return; @@ -763,14 +770,27 @@ void intel_rps_park(struct intel_rps *rps) intel_uncore_forcewake_get(rps_to_uncore(rps), FORCEWAKE_MEDIA); rps_set(rps, rps->idle_freq, false); intel_uncore_forcewake_put(rps_to_uncore(rps), FORCEWAKE_MEDIA); + + /* + * Since we will try and restart from the previously requested + * frequency on unparking, treat this idle point as a downclock + * interrupt and reduce the frequency for resume. If we park/unpark + * more frequently than the rps worker can run, we will not respond + * to any EI and never see a change in frequency. + * + * (Note we accommodate Cherryview's limitation of only using an + * even bin by applying it to all.) + */ + rps->cur_freq = + max_t(int, round_down(rps->cur_freq - 1, 2), rps->min_freq); } void intel_rps_boost(struct i915_request *rq) { - struct intel_rps *rps = &rq->engine->gt->rps; + struct intel_rps *rps = &READ_ONCE(rq->engine)->gt->rps; unsigned long flags; - if (i915_request_signaled(rq) || !rps->active) + if (i915_request_signaled(rq) || !READ_ONCE(rps->active)) return; /* Serializes with i915_request_retire() */ @@ -1453,12 +1473,12 @@ static void rps_work(struct work_struct *work) u32 pm_iir = 0; spin_lock_irq(>->irq_lock); - pm_iir = fetch_and_zero(&rps->pm_iir); + pm_iir = fetch_and_zero(&rps->pm_iir) & READ_ONCE(rps->pm_events); client_boost = atomic_read(&rps->num_waiters); spin_unlock_irq(>->irq_lock); /* Make sure we didn't queue anything we're not going to process. */ - if ((pm_iir & rps->pm_events) == 0 && !client_boost) + if (!pm_iir && !client_boost) goto out; mutex_lock(&rps->lock); @@ -1554,11 +1574,15 @@ void gen11_rps_irq_handler(struct intel_rps *rps, u32 pm_iir) void gen6_rps_irq_handler(struct intel_rps *rps, u32 pm_iir) { struct intel_gt *gt = rps_to_gt(rps); + u32 events; - if (pm_iir & rps->pm_events) { + events = pm_iir & READ_ONCE(rps->pm_events); + if (events) { spin_lock(>->irq_lock); - gen6_gt_pm_mask_irq(gt, pm_iir & rps->pm_events); - rps->pm_iir |= pm_iir & rps->pm_events; + + gen6_gt_pm_mask_irq(gt, events); + rps->pm_iir |= events; + schedule_work(&rps->work); spin_unlock(>->irq_lock); } diff --git a/drivers/gpu/drm/i915/gt/intel_timeline.c b/drivers/gpu/drm/i915/gt/intel_timeline.c index 54e1e55f3c81..91debbc97c9a 100644 --- a/drivers/gpu/drm/i915/gt/intel_timeline.c +++ b/drivers/gpu/drm/i915/gt/intel_timeline.c @@ -192,11 +192,15 @@ static void cacheline_release(struct intel_timeline_cacheline *cl) static void cacheline_free(struct intel_timeline_cacheline *cl) { + if (!i915_active_acquire_if_busy(&cl->active)) { + __idle_cacheline_free(cl); + return; + } + GEM_BUG_ON(ptr_test_bit(cl->vaddr, CACHELINE_FREE)); cl->vaddr = ptr_set_bit(cl->vaddr, CACHELINE_FREE); - if (i915_active_is_idle(&cl->active)) - __idle_cacheline_free(cl); + i915_active_release(&cl->active); } int intel_timeline_init(struct intel_timeline *timeline, diff --git a/drivers/gpu/drm/i915/gt/intel_workarounds.c b/drivers/gpu/drm/i915/gt/intel_workarounds.c index 887e0dc701f7..5176ad1a3976 100644 --- a/drivers/gpu/drm/i915/gt/intel_workarounds.c +++ b/drivers/gpu/drm/i915/gt/intel_workarounds.c @@ -575,29 +575,46 @@ static void icl_ctx_workarounds_init(struct intel_engine_cs *engine, /* allow headerless messages for preemptible GPGPU context */ WA_SET_BIT_MASKED(GEN10_SAMPLER_MODE, GEN11_SAMPLER_ENABLE_HEADLESS_MSG); + + /* Wa_1604278689:icl,ehl */ + wa_write(wal, IVB_FBC_RT_BASE, 0xFFFFFFFF & ~ILK_FBC_RT_VALID); + wa_write_masked_or(wal, IVB_FBC_RT_BASE_UPPER, + 0, /* write-only register; skip validation */ + 0xFFFFFFFF); + + /* Wa_1406306137:icl,ehl */ + wa_masked_en(wal, GEN9_ROW_CHICKEN4, GEN11_DIS_PICK_2ND_EU); } static void tgl_ctx_workarounds_init(struct intel_engine_cs *engine, struct i915_wa_list *wal) { - u32 val; - - /* Wa_1409142259:tgl */ + /* + * Wa_1409142259:tgl + * Wa_1409347922:tgl + * Wa_1409252684:tgl + * Wa_1409217633:tgl + * Wa_1409207793:tgl + * Wa_1409178076:tgl + * Wa_1408979724:tgl + */ WA_SET_BIT_MASKED(GEN11_COMMON_SLICE_CHICKEN3, GEN12_DISABLE_CPS_AWARE_COLOR_PIPE); - /* Wa_1604555607:tgl */ - val = intel_uncore_read(engine->uncore, FF_MODE2); - val &= ~FF_MODE2_TDS_TIMER_MASK; - val |= FF_MODE2_TDS_TIMER_128; /* - * FIXME: FF_MODE2 register is not readable till TGL B0. We can - * enable verification of WA from the later steppings, which enables - * the read of FF_MODE2. + * Wa_1604555607:gen12 and Wa_1608008084:gen12 + * FF_MODE2 register will return the wrong value when read. The default + * value for this register is zero for all fields and there are no bit + * masks. So instead of doing a RMW we should just write the TDS timer + * value for Wa_1604555607. */ - wa_add(wal, FF_MODE2, FF_MODE2_TDS_TIMER_MASK, val, - IS_TGL_REVID(engine->i915, TGL_REVID_A0, TGL_REVID_A0) ? 0 : - FF_MODE2_TDS_TIMER_MASK); + wa_add(wal, FF_MODE2, FF_MODE2_TDS_TIMER_MASK, + FF_MODE2_TDS_TIMER_128, 0); + + /* WaDisableGPGPUMidThreadPreemption:tgl */ + WA_SET_FIELD_MASKED(GEN8_CS_CHICKEN1, + GEN9_PREEMPT_GPGPU_LEVEL_MASK, + GEN9_PREEMPT_GPGPU_THREAD_GROUP_LEVEL); } static void @@ -903,11 +920,6 @@ icl_gt_workarounds_init(struct drm_i915_private *i915, struct i915_wa_list *wal) SLICE_UNIT_LEVEL_CLKGATE, MSCUNIT_CLKGATE_DIS); - /* Wa_1406680159:icl */ - wa_write_or(wal, - SUBSLICE_UNIT_LEVEL_CLKGATE, - GWUNIT_CLKGATE_DIS); - /* Wa_1406838659:icl (pre-prod) */ if (IS_ICL_REVID(i915, ICL_REVID_A0, ICL_REVID_B0)) wa_write_or(wal, @@ -936,7 +948,7 @@ tgl_gt_workarounds_init(struct drm_i915_private *i915, struct i915_wa_list *wal) SUBSLICE_UNIT_LEVEL_CLKGATE2, CPSSUNIT_CLKGATE_DIS); - /* Wa_1409180338:tgl */ + /* Wa_1607087056:tgl also know as BUG:1409180338 */ if (IS_TGL_REVID(i915, TGL_REVID_A0, TGL_REVID_A0)) wa_write_or(wal, SLICE_UNIT_LEVEL_CLKGATE, @@ -1251,6 +1263,7 @@ static void tgl_whitelist_build(struct intel_engine_cs *engine) case RENDER_CLASS: /* * WaAllowPMDepthAndInvocationCountAccessFromUMD:tgl + * Wa_1408556865:tgl * * This covers 4 registers which are next to one another : * - PS_INVOCATION_COUNT @@ -1264,6 +1277,9 @@ static void tgl_whitelist_build(struct intel_engine_cs *engine) /* Wa_1808121037:tgl */ whitelist_reg(w, GEN7_COMMON_SLICE_CHICKEN1); + + /* Wa_1806527549:tgl */ + whitelist_reg(w, HIZ_CHICKEN); break; default: break; @@ -1330,19 +1346,21 @@ rcs_engine_wa_init(struct intel_engine_cs *engine, struct i915_wa_list *wal) struct drm_i915_private *i915 = engine->i915; if (IS_TGL_REVID(i915, TGL_REVID_A0, TGL_REVID_A0)) { - /* Wa_1606700617:tgl */ - wa_masked_en(wal, - GEN9_CS_DEBUG_MODE1, - FF_DOP_CLOCK_GATE_DISABLE); - - /* Wa_1607138336:tgl */ + /* + * Wa_1607138336:tgl + * Wa_1607063988:tgl + */ wa_write_or(wal, GEN9_CTX_PREEMPT_REG, GEN12_DISABLE_POSH_BUSY_FF_DOP_CG); - /* Wa_1607030317:tgl */ - /* Wa_1607186500:tgl */ - /* Wa_1607297627:tgl */ + /* + * Wa_1607030317:tgl + * Wa_1607186500:tgl + * Wa_1607297627:tgl there is 3 entries for this WA on BSpec, 2 + * of then says it is fixed on B0 the other one says it is + * permanent + */ wa_masked_en(wal, GEN6_RC_SLEEP_PSMI_CONTROL, GEN12_WAIT_FOR_EVENT_POWER_DOWN_DISABLE | @@ -1361,10 +1379,29 @@ rcs_engine_wa_init(struct intel_engine_cs *engine, struct i915_wa_list *wal) GEN7_FF_THREAD_MODE, GEN12_FF_TESSELATION_DOP_GATE_DISABLE); + /* + * Wa_1409085225:tgl + * Wa_14010229206:tgl + */ + wa_masked_en(wal, GEN9_ROW_CHICKEN4, GEN12_DISABLE_TDL_PUSH); + + /* Wa_1408615072:tgl */ + wa_write_or(wal, UNSLICE_UNIT_LEVEL_CLKGATE2, + VSUNIT_CLKGATE_DIS_TGL); + } + + if (IS_TIGERLAKE(i915)) { /* Wa_1606931601:tgl */ + wa_masked_en(wal, GEN7_ROW_CHICKEN2, GEN12_DISABLE_EARLY_READ); + + /* Wa_1409804808:tgl */ + wa_masked_en(wal, GEN7_ROW_CHICKEN2, + GEN12_PUSH_CONST_DEREF_HOLD_DIS); + + /* Wa_1606700617:tgl */ wa_masked_en(wal, - GEN7_ROW_CHICKEN2, - GEN12_DISABLE_EARLY_READ); + GEN9_CS_DEBUG_MODE1, + FF_DOP_CLOCK_GATE_DISABLE); } if (IS_GEN(i915, 11)) { @@ -1430,10 +1467,38 @@ rcs_engine_wa_init(struct intel_engine_cs *engine, struct i915_wa_list *wal) GEN11_SCRATCH2, GEN11_COHERENT_PARTIAL_WRITE_MERGE_ENABLE, 0); + + /* WaEnable32PlaneMode:icl */ + wa_masked_en(wal, GEN9_CSFE_CHICKEN1_RCS, + GEN11_ENABLE_32_PLANE_MODE); + + /* + * Wa_1408615072:icl,ehl (vsunit) + * Wa_1407596294:icl,ehl (hsunit) + */ + wa_write_or(wal, UNSLICE_UNIT_LEVEL_CLKGATE, + VSUNIT_CLKGATE_DIS | HSUNIT_CLKGATE_DIS); + + /* Wa_1407352427:icl,ehl */ + wa_write_or(wal, UNSLICE_UNIT_LEVEL_CLKGATE2, + PSDUNIT_CLKGATE_DIS); + + /* Wa_1406680159:icl,ehl */ + wa_write_or(wal, + SUBSLICE_UNIT_LEVEL_CLKGATE, + GWUNIT_CLKGATE_DIS); + + /* + * Wa_1408767742:icl[a2..forever],ehl[all] + * Wa_1605460711:icl[a0..c0] + */ + wa_write_or(wal, + GEN7_FF_THREAD_MODE, + GEN12_FF_TESSELATION_DOP_GATE_DISABLE); } - if (IS_GEN_RANGE(i915, 9, 11)) { - /* FtrPerCtxtPreemptionGranularityControl:skl,bxt,kbl,cfl,cnl,icl */ + if (IS_GEN_RANGE(i915, 9, 12)) { + /* FtrPerCtxtPreemptionGranularityControl:skl,bxt,kbl,cfl,cnl,icl,tgl */ wa_masked_en(wal, GEN7_FF_SLICE_CS_CHICKEN1, GEN9_FFSC_PERCTX_PREEMPT_CTRL); @@ -1600,15 +1665,34 @@ err_obj: return ERR_PTR(err); } +static const struct { + u32 start; + u32 end; +} mcr_ranges_gen8[] = { + { .start = 0x5500, .end = 0x55ff }, + { .start = 0x7000, .end = 0x7fff }, + { .start = 0x9400, .end = 0x97ff }, + { .start = 0xb000, .end = 0xb3ff }, + { .start = 0xe000, .end = 0xe7ff }, + {}, +}; + static bool mcr_range(struct drm_i915_private *i915, u32 offset) { + int i; + + if (INTEL_GEN(i915) < 8) + return false; + /* - * Registers in this range are affected by the MCR selector + * Registers in these ranges are affected by the MCR selector * which only controls CPU initiated MMIO. Routing does not * work for CS access so we cannot verify them on this path. */ - if (INTEL_GEN(i915) >= 8 && (offset >= 0xb000 && offset <= 0xb4ff)) - return true; + for (i = 0; mcr_ranges_gen8[i].start; i++) + if (offset >= mcr_ranges_gen8[i].start && + offset <= mcr_ranges_gen8[i].end) + return true; return false; } diff --git a/drivers/gpu/drm/i915/gt/ivb_clear_kernel.c b/drivers/gpu/drm/i915/gt/ivb_clear_kernel.c new file mode 100644 index 000000000000..610ca7687735 --- /dev/null +++ b/drivers/gpu/drm/i915/gt/ivb_clear_kernel.c @@ -0,0 +1,61 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright © 2020 Intel Corporation + * + * Generated by: IGT Gpu Tools on Fri 21 Feb 2020 05:29:32 AM UTC + */ + +static const u32 ivb_clear_kernel[] = { + 0x00000001, 0x26020128, 0x00000024, 0x00000000, + 0x00000040, 0x20280c21, 0x00000028, 0x00000001, + 0x01000010, 0x20000c20, 0x0000002c, 0x00000000, + 0x00010220, 0x34001c00, 0x00001400, 0x0000002c, + 0x00600001, 0x20600061, 0x00000000, 0x00000000, + 0x00000008, 0x20601c85, 0x00000e00, 0x0000000c, + 0x00000005, 0x20601ca5, 0x00000060, 0x00000001, + 0x00000008, 0x20641c85, 0x00000e00, 0x0000000d, + 0x00000005, 0x20641ca5, 0x00000064, 0x00000003, + 0x00000041, 0x207424a5, 0x00000064, 0x00000034, + 0x00000040, 0x206014a5, 0x00000060, 0x00000074, + 0x00000008, 0x20681c85, 0x00000e00, 0x00000008, + 0x00000005, 0x20681ca5, 0x00000068, 0x0000000f, + 0x00000041, 0x20701ca5, 0x00000060, 0x00000010, + 0x00000040, 0x206814a5, 0x00000068, 0x00000070, + 0x00600001, 0x20a00061, 0x00000000, 0x00000000, + 0x00000005, 0x206c1c85, 0x00000e00, 0x00000007, + 0x00000041, 0x206c1ca5, 0x0000006c, 0x00000004, + 0x00600001, 0x20800021, 0x008d0000, 0x00000000, + 0x00000001, 0x20800021, 0x0000006c, 0x00000000, + 0x00000001, 0x20840021, 0x00000068, 0x00000000, + 0x00000001, 0x20880061, 0x00000000, 0x00000003, + 0x00000005, 0x208c0d21, 0x00000086, 0xffffffff, + 0x05600032, 0x20a00fa1, 0x008d0080, 0x02190001, + 0x00000040, 0x20a01ca5, 0x000000a0, 0x00000001, + 0x05600032, 0x20a00fa1, 0x008d0080, 0x040a8001, + 0x02000040, 0x20281c21, 0x00000028, 0xffffffff, + 0x00010220, 0x34001c00, 0x00001400, 0xfffffffc, + 0x00000001, 0x26020128, 0x00000024, 0x00000000, + 0x00000001, 0x220010e4, 0x00000000, 0x00000000, + 0x00000001, 0x220831ec, 0x00000000, 0x007f007f, + 0x00600001, 0x20400021, 0x008d0000, 0x00000000, + 0x00600001, 0x2fe00021, 0x008d0000, 0x00000000, + 0x00200001, 0x20400121, 0x00450020, 0x00000000, + 0x00000001, 0x20480061, 0x00000000, 0x000f000f, + 0x00000005, 0x204c0d21, 0x00000046, 0xffffffef, + 0x00800001, 0x20600061, 0x00000000, 0x00000000, + 0x00800001, 0x20800061, 0x00000000, 0x00000000, + 0x00800001, 0x20a00061, 0x00000000, 0x00000000, + 0x00800001, 0x20c00061, 0x00000000, 0x00000000, + 0x00800001, 0x20e00061, 0x00000000, 0x00000000, + 0x00800001, 0x21000061, 0x00000000, 0x00000000, + 0x00800001, 0x21200061, 0x00000000, 0x00000000, + 0x00800001, 0x21400061, 0x00000000, 0x00000000, + 0x05600032, 0x20000fa0, 0x008d0040, 0x120a8000, + 0x00000040, 0x20402d21, 0x00000020, 0x00100010, + 0x05600032, 0x20000fa0, 0x008d0040, 0x120a8000, + 0x02000040, 0x22083d8c, 0x00000208, 0xffffffff, + 0x00800001, 0xa0000109, 0x00000602, 0x00000000, + 0x00000040, 0x22001c84, 0x00000200, 0x00000020, + 0x00010220, 0x34001c00, 0x00001400, 0xfffffff8, + 0x07600032, 0x20000fa0, 0x008d0fe0, 0x82000010, +}; diff --git a/drivers/gpu/drm/i915/gt/mock_engine.c b/drivers/gpu/drm/i915/gt/mock_engine.c index 5633515c12e9..4a53ded7c2dd 100644 --- a/drivers/gpu/drm/i915/gt/mock_engine.c +++ b/drivers/gpu/drm/i915/gt/mock_engine.c @@ -244,9 +244,7 @@ static void mock_reset_cancel(struct intel_engine_cs *engine) /* Mark all submitted requests as skipped. */ list_for_each_entry(request, &engine->active.requests, sched.link) { - if (!i915_request_signaled(request)) - dma_fence_set_error(&request->fence, -EIO); - + i915_request_set_error_once(request, -EIO); i915_request_mark_complete(request); } diff --git a/drivers/gpu/drm/i915/gt/selftest_engine_heartbeat.c b/drivers/gpu/drm/i915/gt/selftest_engine_heartbeat.c index 43d4d589749f..697114dd1f47 100644 --- a/drivers/gpu/drm/i915/gt/selftest_engine_heartbeat.c +++ b/drivers/gpu/drm/i915/gt/selftest_engine_heartbeat.c @@ -142,6 +142,24 @@ out: return err; } +static void engine_heartbeat_disable(struct intel_engine_cs *engine, + unsigned long *saved) +{ + *saved = engine->props.heartbeat_interval_ms; + engine->props.heartbeat_interval_ms = 0; + + intel_engine_pm_get(engine); + intel_engine_park_heartbeat(engine); +} + +static void engine_heartbeat_enable(struct intel_engine_cs *engine, + unsigned long saved) +{ + intel_engine_pm_put(engine); + + engine->props.heartbeat_interval_ms = saved; +} + static int live_idle_flush(void *arg) { struct intel_gt *gt = arg; @@ -152,9 +170,11 @@ static int live_idle_flush(void *arg) /* Check that we can flush the idle barriers */ for_each_engine(engine, gt, id) { - intel_engine_pm_get(engine); + unsigned long heartbeat; + + engine_heartbeat_disable(engine, &heartbeat); err = __live_idle_pulse(engine, intel_engine_flush_barriers); - intel_engine_pm_put(engine); + engine_heartbeat_enable(engine, heartbeat); if (err) break; } @@ -172,9 +192,11 @@ static int live_idle_pulse(void *arg) /* Check that heartbeat pulses flush the idle barriers */ for_each_engine(engine, gt, id) { - intel_engine_pm_get(engine); + unsigned long heartbeat; + + engine_heartbeat_disable(engine, &heartbeat); err = __live_idle_pulse(engine, intel_engine_pulse); - intel_engine_pm_put(engine); + engine_heartbeat_enable(engine, heartbeat); if (err && err != -ENODEV) break; diff --git a/drivers/gpu/drm/i915/gt/selftest_hangcheck.c b/drivers/gpu/drm/i915/gt/selftest_hangcheck.c index c3514ec7b8db..2b2efff6e19d 100644 --- a/drivers/gpu/drm/i915/gt/selftest_hangcheck.c +++ b/drivers/gpu/drm/i915/gt/selftest_hangcheck.c @@ -268,7 +268,7 @@ hang_create_request(struct hang *h, struct intel_engine_cs *engine) cancel_rq: if (err) { - i915_request_skip(rq, err); + i915_request_set_error_once(rq, err); i915_request_add(rq); } unpin_hws: diff --git a/drivers/gpu/drm/i915/gt/selftest_lrc.c b/drivers/gpu/drm/i915/gt/selftest_lrc.c index febd608c23a7..6f06ba750a0a 100644 --- a/drivers/gpu/drm/i915/gt/selftest_lrc.c +++ b/drivers/gpu/drm/i915/gt/selftest_lrc.c @@ -90,6 +90,49 @@ static int wait_for_submit(struct intel_engine_cs *engine, return -ETIME; } +static int wait_for_reset(struct intel_engine_cs *engine, + struct i915_request *rq, + unsigned long timeout) +{ + timeout += jiffies; + + do { + cond_resched(); + intel_engine_flush_submission(engine); + + if (READ_ONCE(engine->execlists.pending[0])) + continue; + + if (i915_request_completed(rq)) + break; + + if (READ_ONCE(rq->fence.error)) + break; + } while (time_before(jiffies, timeout)); + + flush_scheduled_work(); + + if (rq->fence.error != -EIO) { + pr_err("%s: hanging request %llx:%lld not reset\n", + engine->name, + rq->fence.context, + rq->fence.seqno); + return -EINVAL; + } + + /* Give the request a jiffie to complete after flushing the worker */ + if (i915_request_wait(rq, 0, + max(0l, (long)(timeout - jiffies)) + 1) < 0) { + pr_err("%s: hanging request %llx:%lld did not complete\n", + engine->name, + rq->fence.context, + rq->fence.seqno); + return -ETIME; + } + + return 0; +} + static int live_sanitycheck(void *arg) { struct intel_gt *gt = arg; @@ -1805,14 +1848,9 @@ static int __cancel_active0(struct live_preempt_cancel *arg) if (err) goto out; - if (i915_request_wait(rq, 0, HZ / 5) < 0) { - err = -EIO; - goto out; - } - - if (rq->fence.error != -EIO) { - pr_err("Cancelled inflight0 request did not report -EIO\n"); - err = -EINVAL; + err = wait_for_reset(arg->engine, rq, HZ / 2); + if (err) { + pr_err("Cancelled inflight0 request did not reset\n"); goto out; } @@ -1870,10 +1908,9 @@ static int __cancel_active1(struct live_preempt_cancel *arg) goto out; igt_spinner_end(&arg->a.spin); - if (i915_request_wait(rq[1], 0, HZ / 5) < 0) { - err = -EIO; + err = wait_for_reset(arg->engine, rq[1], HZ / 2); + if (err) goto out; - } if (rq[0]->fence.error != 0) { pr_err("Normal inflight0 request did not complete\n"); @@ -1953,10 +1990,9 @@ static int __cancel_queued(struct live_preempt_cancel *arg) if (err) goto out; - if (i915_request_wait(rq[2], 0, HZ / 5) < 0) { - err = -EIO; + err = wait_for_reset(arg->engine, rq[2], HZ / 2); + if (err) goto out; - } if (rq[0]->fence.error != -EIO) { pr_err("Cancelled inflight0 request did not report -EIO\n"); @@ -2014,14 +2050,9 @@ static int __cancel_hostile(struct live_preempt_cancel *arg) if (err) goto out; - if (i915_request_wait(rq, 0, HZ / 5) < 0) { - err = -EIO; - goto out; - } - - if (rq->fence.error != -EIO) { - pr_err("Cancelled inflight0 request did not report -EIO\n"); - err = -EINVAL; + err = wait_for_reset(arg->engine, rq, HZ / 2); + if (err) { + pr_err("Cancelled inflight0 request did not reset\n"); goto out; } @@ -4015,6 +4046,32 @@ static int emit_semaphore_signal(struct intel_context *ce, void *slot) return 0; } +static int context_flush(struct intel_context *ce, long timeout) +{ + struct i915_request *rq; + struct dma_fence *fence; + int err = 0; + + rq = intel_engine_create_kernel_request(ce->engine); + if (IS_ERR(rq)) + return PTR_ERR(rq); + + fence = i915_active_fence_get(&ce->timeline->last_request); + if (fence) { + i915_request_await_dma_fence(rq, fence); + dma_fence_put(fence); + } + + rq = i915_request_get(rq); + i915_request_add(rq); + if (i915_request_wait(rq, 0, timeout) < 0) + err = -ETIME; + i915_request_put(rq); + + rmb(); /* We know the request is written, make sure all state is too! */ + return err; +} + static int live_lrc_layout(void *arg) { struct intel_gt *gt = arg; @@ -4638,18 +4695,10 @@ static int __lrc_timestamp(const struct lrc_timestamp *arg, bool preempt) wmb(); } - if (i915_request_wait(rq, 0, HZ / 2) < 0) { - err = -ETIME; - goto err; - } - - /* and wait for switch to kernel */ - if (igt_flush_test(arg->engine->i915)) { - err = -EIO; + /* And wait for switch to kernel (to save our context to memory) */ + err = context_flush(arg->ce[0], HZ / 2); + if (err) goto err; - } - - rmb(); if (!timestamp_advanced(arg->poison, slot[1])) { pr_err("%s(%s): invalid timestamp on restore, context:%x, request:%x\n", @@ -4674,9 +4723,9 @@ err: static int live_lrc_timestamp(void *arg) { + struct lrc_timestamp data = {}; struct intel_gt *gt = arg; enum intel_engine_id id; - struct lrc_timestamp data; const u32 poison[] = { 0, S32_MAX, @@ -4748,6 +4797,677 @@ err: return 0; } +static struct i915_vma * +create_user_vma(struct i915_address_space *vm, unsigned long size) +{ + struct drm_i915_gem_object *obj; + struct i915_vma *vma; + int err; + + obj = i915_gem_object_create_internal(vm->i915, size); + if (IS_ERR(obj)) + return ERR_CAST(obj); + + vma = i915_vma_instance(obj, vm, NULL); + if (IS_ERR(vma)) { + i915_gem_object_put(obj); + return vma; + } + + err = i915_vma_pin(vma, 0, 0, PIN_USER); + if (err) { + i915_gem_object_put(obj); + return ERR_PTR(err); + } + + return vma; +} + +static struct i915_vma * +store_context(struct intel_context *ce, struct i915_vma *scratch) +{ + struct i915_vma *batch; + u32 dw, x, *cs, *hw; + + batch = create_user_vma(ce->vm, SZ_64K); + if (IS_ERR(batch)) + return batch; + + cs = i915_gem_object_pin_map(batch->obj, I915_MAP_WC); + if (IS_ERR(cs)) { + i915_vma_put(batch); + return ERR_CAST(cs); + } + + x = 0; + dw = 0; + hw = ce->engine->pinned_default_state; + hw += LRC_STATE_PN * PAGE_SIZE / sizeof(*hw); + do { + u32 len = hw[dw] & 0x7f; + + if (hw[dw] == 0) { + dw++; + continue; + } + + if ((hw[dw] & GENMASK(31, 23)) != MI_INSTR(0x22, 0)) { + dw += len + 2; + continue; + } + + dw++; + len = (len + 1) / 2; + while (len--) { + *cs++ = MI_STORE_REGISTER_MEM_GEN8; + *cs++ = hw[dw]; + *cs++ = lower_32_bits(scratch->node.start + x); + *cs++ = upper_32_bits(scratch->node.start + x); + + dw += 2; + x += 4; + } + } while (dw < PAGE_SIZE / sizeof(u32) && + (hw[dw] & ~BIT(0)) != MI_BATCH_BUFFER_END); + + *cs++ = MI_BATCH_BUFFER_END; + + i915_gem_object_flush_map(batch->obj); + i915_gem_object_unpin_map(batch->obj); + + return batch; +} + +static int move_to_active(struct i915_request *rq, + struct i915_vma *vma, + unsigned int flags) +{ + int err; + + i915_vma_lock(vma); + err = i915_request_await_object(rq, vma->obj, flags); + if (!err) + err = i915_vma_move_to_active(vma, rq, flags); + i915_vma_unlock(vma); + + return err; +} + +static struct i915_request * +record_registers(struct intel_context *ce, + struct i915_vma *before, + struct i915_vma *after, + u32 *sema) +{ + struct i915_vma *b_before, *b_after; + struct i915_request *rq; + u32 *cs; + int err; + + b_before = store_context(ce, before); + if (IS_ERR(b_before)) + return ERR_CAST(b_before); + + b_after = store_context(ce, after); + if (IS_ERR(b_after)) { + rq = ERR_CAST(b_after); + goto err_before; + } + + rq = intel_context_create_request(ce); + if (IS_ERR(rq)) + goto err_after; + + err = move_to_active(rq, before, EXEC_OBJECT_WRITE); + if (err) + goto err_rq; + + err = move_to_active(rq, b_before, 0); + if (err) + goto err_rq; + + err = move_to_active(rq, after, EXEC_OBJECT_WRITE); + if (err) + goto err_rq; + + err = move_to_active(rq, b_after, 0); + if (err) + goto err_rq; + + cs = intel_ring_begin(rq, 14); + if (IS_ERR(cs)) { + err = PTR_ERR(cs); + goto err_rq; + } + + *cs++ = MI_ARB_ON_OFF | MI_ARB_DISABLE; + *cs++ = MI_BATCH_BUFFER_START_GEN8 | BIT(8); + *cs++ = lower_32_bits(b_before->node.start); + *cs++ = upper_32_bits(b_before->node.start); + + *cs++ = MI_ARB_ON_OFF | MI_ARB_ENABLE; + *cs++ = MI_SEMAPHORE_WAIT | + MI_SEMAPHORE_GLOBAL_GTT | + MI_SEMAPHORE_POLL | + MI_SEMAPHORE_SAD_NEQ_SDD; + *cs++ = 0; + *cs++ = i915_ggtt_offset(ce->engine->status_page.vma) + + offset_in_page(sema); + *cs++ = 0; + *cs++ = MI_NOOP; + + *cs++ = MI_ARB_ON_OFF | MI_ARB_DISABLE; + *cs++ = MI_BATCH_BUFFER_START_GEN8 | BIT(8); + *cs++ = lower_32_bits(b_after->node.start); + *cs++ = upper_32_bits(b_after->node.start); + + intel_ring_advance(rq, cs); + + WRITE_ONCE(*sema, 0); + i915_request_get(rq); + i915_request_add(rq); +err_after: + i915_vma_put(b_after); +err_before: + i915_vma_put(b_before); + return rq; + +err_rq: + i915_request_add(rq); + rq = ERR_PTR(err); + goto err_after; +} + +static struct i915_vma *load_context(struct intel_context *ce, u32 poison) +{ + struct i915_vma *batch; + u32 dw, *cs, *hw; + + batch = create_user_vma(ce->vm, SZ_64K); + if (IS_ERR(batch)) + return batch; + + cs = i915_gem_object_pin_map(batch->obj, I915_MAP_WC); + if (IS_ERR(cs)) { + i915_vma_put(batch); + return ERR_CAST(cs); + } + + dw = 0; + hw = ce->engine->pinned_default_state; + hw += LRC_STATE_PN * PAGE_SIZE / sizeof(*hw); + do { + u32 len = hw[dw] & 0x7f; + + if (hw[dw] == 0) { + dw++; + continue; + } + + if ((hw[dw] & GENMASK(31, 23)) != MI_INSTR(0x22, 0)) { + dw += len + 2; + continue; + } + + dw++; + len = (len + 1) / 2; + *cs++ = MI_LOAD_REGISTER_IMM(len); + while (len--) { + *cs++ = hw[dw]; + *cs++ = poison; + dw += 2; + } + } while (dw < PAGE_SIZE / sizeof(u32) && + (hw[dw] & ~BIT(0)) != MI_BATCH_BUFFER_END); + + *cs++ = MI_BATCH_BUFFER_END; + + i915_gem_object_flush_map(batch->obj); + i915_gem_object_unpin_map(batch->obj); + + return batch; +} + +static int poison_registers(struct intel_context *ce, u32 poison, u32 *sema) +{ + struct i915_request *rq; + struct i915_vma *batch; + u32 *cs; + int err; + + batch = load_context(ce, poison); + if (IS_ERR(batch)) + return PTR_ERR(batch); + + rq = intel_context_create_request(ce); + if (IS_ERR(rq)) { + err = PTR_ERR(rq); + goto err_batch; + } + + err = move_to_active(rq, batch, 0); + if (err) + goto err_rq; + + cs = intel_ring_begin(rq, 8); + if (IS_ERR(cs)) { + err = PTR_ERR(cs); + goto err_rq; + } + + *cs++ = MI_ARB_ON_OFF | MI_ARB_DISABLE; + *cs++ = MI_BATCH_BUFFER_START_GEN8 | BIT(8); + *cs++ = lower_32_bits(batch->node.start); + *cs++ = upper_32_bits(batch->node.start); + + *cs++ = MI_STORE_DWORD_IMM_GEN4 | MI_USE_GGTT; + *cs++ = i915_ggtt_offset(ce->engine->status_page.vma) + + offset_in_page(sema); + *cs++ = 0; + *cs++ = 1; + + intel_ring_advance(rq, cs); + + rq->sched.attr.priority = I915_PRIORITY_BARRIER; +err_rq: + i915_request_add(rq); +err_batch: + i915_vma_put(batch); + return err; +} + +static bool is_moving(u32 a, u32 b) +{ + return a != b; +} + +static int compare_isolation(struct intel_engine_cs *engine, + struct i915_vma *ref[2], + struct i915_vma *result[2], + struct intel_context *ce, + u32 poison) +{ + u32 x, dw, *hw, *lrc; + u32 *A[2], *B[2]; + int err = 0; + + A[0] = i915_gem_object_pin_map(ref[0]->obj, I915_MAP_WC); + if (IS_ERR(A[0])) + return PTR_ERR(A[0]); + + A[1] = i915_gem_object_pin_map(ref[1]->obj, I915_MAP_WC); + if (IS_ERR(A[1])) { + err = PTR_ERR(A[1]); + goto err_A0; + } + + B[0] = i915_gem_object_pin_map(result[0]->obj, I915_MAP_WC); + if (IS_ERR(B[0])) { + err = PTR_ERR(B[0]); + goto err_A1; + } + + B[1] = i915_gem_object_pin_map(result[1]->obj, I915_MAP_WC); + if (IS_ERR(B[1])) { + err = PTR_ERR(B[1]); + goto err_B0; + } + + lrc = i915_gem_object_pin_map(ce->state->obj, + i915_coherent_map_type(engine->i915)); + if (IS_ERR(lrc)) { + err = PTR_ERR(lrc); + goto err_B1; + } + lrc += LRC_STATE_PN * PAGE_SIZE / sizeof(*hw); + + x = 0; + dw = 0; + hw = engine->pinned_default_state; + hw += LRC_STATE_PN * PAGE_SIZE / sizeof(*hw); + do { + u32 len = hw[dw] & 0x7f; + + if (hw[dw] == 0) { + dw++; + continue; + } + + if ((hw[dw] & GENMASK(31, 23)) != MI_INSTR(0x22, 0)) { + dw += len + 2; + continue; + } + + dw++; + len = (len + 1) / 2; + while (len--) { + if (!is_moving(A[0][x], A[1][x]) && + (A[0][x] != B[0][x] || A[1][x] != B[1][x])) { + switch (hw[dw] & 4095) { + case 0x30: /* RING_HEAD */ + case 0x34: /* RING_TAIL */ + break; + + default: + pr_err("%s[%d]: Mismatch for register %4x, default %08x, reference %08x, result (%08x, %08x), poison %08x, context %08x\n", + engine->name, dw, + hw[dw], hw[dw + 1], + A[0][x], B[0][x], B[1][x], + poison, lrc[dw + 1]); + err = -EINVAL; + break; + } + } + dw += 2; + x++; + } + } while (dw < PAGE_SIZE / sizeof(u32) && + (hw[dw] & ~BIT(0)) != MI_BATCH_BUFFER_END); + + i915_gem_object_unpin_map(ce->state->obj); +err_B1: + i915_gem_object_unpin_map(result[1]->obj); +err_B0: + i915_gem_object_unpin_map(result[0]->obj); +err_A1: + i915_gem_object_unpin_map(ref[1]->obj); +err_A0: + i915_gem_object_unpin_map(ref[0]->obj); + return err; +} + +static int __lrc_isolation(struct intel_engine_cs *engine, u32 poison) +{ + u32 *sema = memset32(engine->status_page.addr + 1000, 0, 1); + struct i915_vma *ref[2], *result[2]; + struct intel_context *A, *B; + struct i915_request *rq; + int err; + + A = intel_context_create(engine); + if (IS_ERR(A)) + return PTR_ERR(A); + + B = intel_context_create(engine); + if (IS_ERR(B)) { + err = PTR_ERR(B); + goto err_A; + } + + ref[0] = create_user_vma(A->vm, SZ_64K); + if (IS_ERR(ref[0])) { + err = PTR_ERR(ref[0]); + goto err_B; + } + + ref[1] = create_user_vma(A->vm, SZ_64K); + if (IS_ERR(ref[1])) { + err = PTR_ERR(ref[1]); + goto err_ref0; + } + + rq = record_registers(A, ref[0], ref[1], sema); + if (IS_ERR(rq)) { + err = PTR_ERR(rq); + goto err_ref1; + } + + WRITE_ONCE(*sema, 1); + wmb(); + + if (i915_request_wait(rq, 0, HZ / 2) < 0) { + i915_request_put(rq); + err = -ETIME; + goto err_ref1; + } + i915_request_put(rq); + + result[0] = create_user_vma(A->vm, SZ_64K); + if (IS_ERR(result[0])) { + err = PTR_ERR(result[0]); + goto err_ref1; + } + + result[1] = create_user_vma(A->vm, SZ_64K); + if (IS_ERR(result[1])) { + err = PTR_ERR(result[1]); + goto err_result0; + } + + rq = record_registers(A, result[0], result[1], sema); + if (IS_ERR(rq)) { + err = PTR_ERR(rq); + goto err_result1; + } + + err = poison_registers(B, poison, sema); + if (err) { + WRITE_ONCE(*sema, -1); + i915_request_put(rq); + goto err_result1; + } + + if (i915_request_wait(rq, 0, HZ / 2) < 0) { + i915_request_put(rq); + err = -ETIME; + goto err_result1; + } + i915_request_put(rq); + + err = compare_isolation(engine, ref, result, A, poison); + +err_result1: + i915_vma_put(result[1]); +err_result0: + i915_vma_put(result[0]); +err_ref1: + i915_vma_put(ref[1]); +err_ref0: + i915_vma_put(ref[0]); +err_B: + intel_context_put(B); +err_A: + intel_context_put(A); + return err; +} + +static bool skip_isolation(const struct intel_engine_cs *engine) +{ + if (engine->class == COPY_ENGINE_CLASS && INTEL_GEN(engine->i915) == 9) + return true; + + if (engine->class == RENDER_CLASS && INTEL_GEN(engine->i915) == 11) + return true; + + return false; +} + +static int live_lrc_isolation(void *arg) +{ + struct intel_gt *gt = arg; + struct intel_engine_cs *engine; + enum intel_engine_id id; + const u32 poison[] = { + STACK_MAGIC, + 0x3a3a3a3a, + 0x5c5c5c5c, + 0xffffffff, + 0xffff0000, + }; + + /* + * Our goal is try and verify that per-context state cannot be + * tampered with by another non-privileged client. + * + * We take the list of context registers from the LRI in the default + * context image and attempt to modify that list from a remote context. + */ + + for_each_engine(engine, gt, id) { + int err = 0; + int i; + + /* Just don't even ask */ + if (!IS_ENABLED(CONFIG_DRM_I915_SELFTEST_BROKEN) && + skip_isolation(engine)) + continue; + + intel_engine_pm_get(engine); + if (engine->pinned_default_state) { + for (i = 0; i < ARRAY_SIZE(poison); i++) { + err = __lrc_isolation(engine, poison[i]); + if (err) + break; + + err = __lrc_isolation(engine, ~poison[i]); + if (err) + break; + } + } + intel_engine_pm_put(engine); + if (igt_flush_test(gt->i915)) + err = -EIO; + if (err) + return err; + } + + return 0; +} + +static void garbage_reset(struct intel_engine_cs *engine, + struct i915_request *rq) +{ + const unsigned int bit = I915_RESET_ENGINE + engine->id; + unsigned long *lock = &engine->gt->reset.flags; + + if (test_and_set_bit(bit, lock)) + return; + + tasklet_disable(&engine->execlists.tasklet); + + if (!rq->fence.error) + intel_engine_reset(engine, NULL); + + tasklet_enable(&engine->execlists.tasklet); + clear_and_wake_up_bit(bit, lock); +} + +static struct i915_request *garbage(struct intel_context *ce, + struct rnd_state *prng) +{ + struct i915_request *rq; + int err; + + err = intel_context_pin(ce); + if (err) + return ERR_PTR(err); + + prandom_bytes_state(prng, + ce->lrc_reg_state, + ce->engine->context_size - + LRC_STATE_PN * PAGE_SIZE); + + rq = intel_context_create_request(ce); + if (IS_ERR(rq)) { + err = PTR_ERR(rq); + goto err_unpin; + } + + i915_request_get(rq); + i915_request_add(rq); + return rq; + +err_unpin: + intel_context_unpin(ce); + return ERR_PTR(err); +} + +static int __lrc_garbage(struct intel_engine_cs *engine, struct rnd_state *prng) +{ + struct intel_context *ce; + struct i915_request *hang; + int err = 0; + + ce = intel_context_create(engine); + if (IS_ERR(ce)) + return PTR_ERR(ce); + + hang = garbage(ce, prng); + if (IS_ERR(hang)) { + err = PTR_ERR(hang); + goto err_ce; + } + + if (wait_for_submit(engine, hang, HZ / 2)) { + i915_request_put(hang); + err = -ETIME; + goto err_ce; + } + + intel_context_set_banned(ce); + garbage_reset(engine, hang); + + intel_engine_flush_submission(engine); + if (!hang->fence.error) { + i915_request_put(hang); + pr_err("%s: corrupted context was not reset\n", + engine->name); + err = -EINVAL; + goto err_ce; + } + + if (i915_request_wait(hang, 0, HZ / 2) < 0) { + pr_err("%s: corrupted context did not recover\n", + engine->name); + i915_request_put(hang); + err = -EIO; + goto err_ce; + } + i915_request_put(hang); + +err_ce: + intel_context_put(ce); + return err; +} + +static int live_lrc_garbage(void *arg) +{ + struct intel_gt *gt = arg; + struct intel_engine_cs *engine; + enum intel_engine_id id; + + /* + * Verify that we can recover if one context state is completely + * corrupted. + */ + + if (!IS_ENABLED(CONFIG_DRM_I915_SELFTEST_BROKEN)) + return 0; + + for_each_engine(engine, gt, id) { + I915_RND_STATE(prng); + int err = 0, i; + + if (!intel_has_reset_engine(engine->gt)) + continue; + + intel_engine_pm_get(engine); + for (i = 0; i < 3; i++) { + err = __lrc_garbage(engine, &prng); + if (err) + break; + } + intel_engine_pm_put(engine); + + if (igt_flush_test(gt->i915)) + err = -EIO; + if (err) + return err; + } + + return 0; +} + static int __live_pphwsp_runtime(struct intel_engine_cs *engine) { struct intel_context *ce; @@ -4845,7 +5565,9 @@ int intel_lrc_live_selftests(struct drm_i915_private *i915) SUBTEST(live_lrc_fixed), SUBTEST(live_lrc_state), SUBTEST(live_lrc_gpr), + SUBTEST(live_lrc_isolation), SUBTEST(live_lrc_timestamp), + SUBTEST(live_lrc_garbage), SUBTEST(live_pphwsp_runtime), }; diff --git a/drivers/gpu/drm/i915/gt/selftest_rc6.c b/drivers/gpu/drm/i915/gt/selftest_rc6.c index 5f7e2dcf5686..95b165faeba7 100644 --- a/drivers/gpu/drm/i915/gt/selftest_rc6.c +++ b/drivers/gpu/drm/i915/gt/selftest_rc6.c @@ -12,6 +12,21 @@ #include "selftests/i915_random.h" +static u64 rc6_residency(struct intel_rc6 *rc6) +{ + u64 result; + + /* XXX VLV_GT_MEDIA_RC6? */ + + result = intel_rc6_residency_ns(rc6, GEN6_GT_GFX_RC6); + if (HAS_RC6p(rc6_to_i915(rc6))) + result += intel_rc6_residency_ns(rc6, GEN6_GT_GFX_RC6p); + if (HAS_RC6pp(rc6_to_i915(rc6))) + result += intel_rc6_residency_ns(rc6, GEN6_GT_GFX_RC6pp); + + return result; +} + int live_rc6_manual(void *arg) { struct intel_gt *gt = arg; @@ -38,9 +53,9 @@ int live_rc6_manual(void *arg) __intel_rc6_disable(rc6); msleep(1); /* wakeup is not immediate, takes about 100us on icl */ - res[0] = intel_rc6_residency_ns(rc6, GEN6_GT_GFX_RC6); + res[0] = rc6_residency(rc6); msleep(250); - res[1] = intel_rc6_residency_ns(rc6, GEN6_GT_GFX_RC6); + res[1] = rc6_residency(rc6); if ((res[1] - res[0]) >> 10) { pr_err("RC6 residency increased by %lldus while disabled for 250ms!\n", (res[1] - res[0]) >> 10); @@ -51,9 +66,9 @@ int live_rc6_manual(void *arg) /* Manually enter RC6 */ intel_rc6_park(rc6); - res[0] = intel_rc6_residency_ns(rc6, GEN6_GT_GFX_RC6); + res[0] = rc6_residency(rc6); msleep(100); - res[1] = intel_rc6_residency_ns(rc6, GEN6_GT_GFX_RC6); + res[1] = rc6_residency(rc6); if (res[1] == res[0]) { pr_err("Did not enter RC6! RC6_STATE=%08x, RC6_CONTROL=%08x, residency=%lld\n", diff --git a/drivers/gpu/drm/i915/gt/selftest_ring_submission.c b/drivers/gpu/drm/i915/gt/selftest_ring_submission.c new file mode 100644 index 000000000000..9995faadd7e8 --- /dev/null +++ b/drivers/gpu/drm/i915/gt/selftest_ring_submission.c @@ -0,0 +1,296 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright © 2020 Intel Corporation + */ + +#include "intel_engine_pm.h" +#include "selftests/igt_flush_test.h" + +static struct i915_vma *create_wally(struct intel_engine_cs *engine) +{ + struct drm_i915_gem_object *obj; + struct i915_vma *vma; + u32 *cs; + int err; + + obj = i915_gem_object_create_internal(engine->i915, 4096); + if (IS_ERR(obj)) + return ERR_CAST(obj); + + vma = i915_vma_instance(obj, engine->gt->vm, NULL); + if (IS_ERR(vma)) { + i915_gem_object_put(obj); + return vma; + } + + err = i915_vma_pin(vma, 0, 0, PIN_USER | PIN_HIGH); + if (err) { + i915_gem_object_put(obj); + return ERR_PTR(err); + } + + err = i915_vma_sync(vma); + if (err) { + i915_gem_object_put(obj); + return ERR_PTR(err); + } + + cs = i915_gem_object_pin_map(obj, I915_MAP_WC); + if (IS_ERR(cs)) { + i915_gem_object_put(obj); + return ERR_CAST(cs); + } + + if (INTEL_GEN(engine->i915) >= 6) { + *cs++ = MI_STORE_DWORD_IMM_GEN4; + *cs++ = 0; + } else if (INTEL_GEN(engine->i915) >= 4) { + *cs++ = MI_STORE_DWORD_IMM_GEN4 | MI_USE_GGTT; + *cs++ = 0; + } else { + *cs++ = MI_STORE_DWORD_IMM | MI_MEM_VIRTUAL; + } + *cs++ = vma->node.start + 4000; + *cs++ = STACK_MAGIC; + + *cs++ = MI_BATCH_BUFFER_END; + i915_gem_object_unpin_map(obj); + + vma->private = intel_context_create(engine); /* dummy residuals */ + if (IS_ERR(vma->private)) { + vma = ERR_CAST(vma->private); + i915_gem_object_put(obj); + } + + return vma; +} + +static int context_sync(struct intel_context *ce) +{ + struct i915_request *rq; + int err = 0; + + rq = intel_context_create_request(ce); + if (IS_ERR(rq)) + return PTR_ERR(rq); + + i915_request_get(rq); + i915_request_add(rq); + + if (i915_request_wait(rq, 0, HZ / 5) < 0) + err = -ETIME; + i915_request_put(rq); + + return err; +} + +static int new_context_sync(struct intel_engine_cs *engine) +{ + struct intel_context *ce; + int err; + + ce = intel_context_create(engine); + if (IS_ERR(ce)) + return PTR_ERR(ce); + + err = context_sync(ce); + intel_context_put(ce); + + return err; +} + +static int mixed_contexts_sync(struct intel_engine_cs *engine, u32 *result) +{ + int pass; + int err; + + for (pass = 0; pass < 2; pass++) { + WRITE_ONCE(*result, 0); + err = context_sync(engine->kernel_context); + if (err || READ_ONCE(*result)) { + if (!err) { + pr_err("pass[%d] wa_bb emitted for the kernel context\n", + pass); + err = -EINVAL; + } + return err; + } + + WRITE_ONCE(*result, 0); + err = new_context_sync(engine); + if (READ_ONCE(*result) != STACK_MAGIC) { + if (!err) { + pr_err("pass[%d] wa_bb *NOT* emitted after the kernel context\n", + pass); + err = -EINVAL; + } + return err; + } + + WRITE_ONCE(*result, 0); + err = new_context_sync(engine); + if (READ_ONCE(*result) != STACK_MAGIC) { + if (!err) { + pr_err("pass[%d] wa_bb *NOT* emitted for the user context switch\n", + pass); + err = -EINVAL; + } + return err; + } + } + + return 0; +} + +static int double_context_sync_00(struct intel_engine_cs *engine, u32 *result) +{ + struct intel_context *ce; + int err, i; + + ce = intel_context_create(engine); + if (IS_ERR(ce)) + return PTR_ERR(ce); + + for (i = 0; i < 2; i++) { + WRITE_ONCE(*result, 0); + err = context_sync(ce); + if (err) + break; + } + intel_context_put(ce); + if (err) + return err; + + if (READ_ONCE(*result)) { + pr_err("wa_bb emitted between the same user context\n"); + return -EINVAL; + } + + return 0; +} + +static int kernel_context_sync_00(struct intel_engine_cs *engine, u32 *result) +{ + struct intel_context *ce; + int err, i; + + ce = intel_context_create(engine); + if (IS_ERR(ce)) + return PTR_ERR(ce); + + for (i = 0; i < 2; i++) { + WRITE_ONCE(*result, 0); + err = context_sync(ce); + if (err) + break; + + err = context_sync(engine->kernel_context); + if (err) + break; + } + intel_context_put(ce); + if (err) + return err; + + if (READ_ONCE(*result)) { + pr_err("wa_bb emitted between the same user context [with intervening kernel]\n"); + return -EINVAL; + } + + return 0; +} + +static int __live_ctx_switch_wa(struct intel_engine_cs *engine) +{ + struct i915_vma *bb; + u32 *result; + int err; + + bb = create_wally(engine); + if (IS_ERR(bb)) + return PTR_ERR(bb); + + result = i915_gem_object_pin_map(bb->obj, I915_MAP_WC); + if (IS_ERR(result)) { + intel_context_put(bb->private); + i915_vma_unpin_and_release(&bb, 0); + return PTR_ERR(result); + } + result += 1000; + + engine->wa_ctx.vma = bb; + + err = mixed_contexts_sync(engine, result); + if (err) + goto out; + + err = double_context_sync_00(engine, result); + if (err) + goto out; + + err = kernel_context_sync_00(engine, result); + if (err) + goto out; + +out: + intel_context_put(engine->wa_ctx.vma->private); + i915_vma_unpin_and_release(&engine->wa_ctx.vma, I915_VMA_RELEASE_MAP); + return err; +} + +static int live_ctx_switch_wa(void *arg) +{ + struct intel_gt *gt = arg; + struct intel_engine_cs *engine; + enum intel_engine_id id; + + /* + * Exercise the inter-context wa batch. + * + * Between each user context we run a wa batch, and since it may + * have implications for user visible state, we have to check that + * we do actually execute it. + * + * The trick we use is to replace the normal wa batch with a custom + * one that writes to a marker within it, and we can then look for + * that marker to confirm if the batch was run when we expect it, + * and equally important it was wasn't run when we don't! + */ + + for_each_engine(engine, gt, id) { + struct i915_vma *saved_wa; + int err; + + if (!intel_engine_can_store_dword(engine)) + continue; + + if (IS_GEN_RANGE(gt->i915, 4, 5)) + continue; /* MI_STORE_DWORD is privileged! */ + + saved_wa = fetch_and_zero(&engine->wa_ctx.vma); + + intel_engine_pm_get(engine); + err = __live_ctx_switch_wa(engine); + intel_engine_pm_put(engine); + if (igt_flush_test(gt->i915)) + err = -EIO; + + engine->wa_ctx.vma = saved_wa; + if (err) + return err; + } + + return 0; +} + +int intel_ring_submission_live_selftests(struct drm_i915_private *i915) +{ + static const struct i915_subtest tests[] = { + SUBTEST(live_ctx_switch_wa), + }; + + if (HAS_EXECLISTS(i915)) + return 0; + + return intel_gt_live_subtests(tests, &i915->gt); +} diff --git a/drivers/gpu/drm/i915/gt/sysfs_engines.c b/drivers/gpu/drm/i915/gt/sysfs_engines.c new file mode 100644 index 000000000000..8f9b2f33dbaf --- /dev/null +++ b/drivers/gpu/drm/i915/gt/sysfs_engines.c @@ -0,0 +1,445 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright © 2019 Intel Corporation + */ + +#include <linux/kobject.h> +#include <linux/sysfs.h> + +#include "i915_drv.h" +#include "intel_engine.h" +#include "intel_engine_heartbeat.h" +#include "sysfs_engines.h" + +struct kobj_engine { + struct kobject base; + struct intel_engine_cs *engine; +}; + +static struct intel_engine_cs *kobj_to_engine(struct kobject *kobj) +{ + return container_of(kobj, struct kobj_engine, base)->engine; +} + +static ssize_t +name_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) +{ + return sprintf(buf, "%s\n", kobj_to_engine(kobj)->name); +} + +static struct kobj_attribute name_attr = +__ATTR(name, 0444, name_show, NULL); + +static ssize_t +class_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) +{ + return sprintf(buf, "%d\n", kobj_to_engine(kobj)->uabi_class); +} + +static struct kobj_attribute class_attr = +__ATTR(class, 0444, class_show, NULL); + +static ssize_t +inst_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) +{ + return sprintf(buf, "%d\n", kobj_to_engine(kobj)->uabi_instance); +} + +static struct kobj_attribute inst_attr = +__ATTR(instance, 0444, inst_show, NULL); + +static ssize_t +mmio_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) +{ + return sprintf(buf, "0x%x\n", kobj_to_engine(kobj)->mmio_base); +} + +static struct kobj_attribute mmio_attr = +__ATTR(mmio_base, 0444, mmio_show, NULL); + +static const char * const vcs_caps[] = { + [ilog2(I915_VIDEO_CLASS_CAPABILITY_HEVC)] = "hevc", + [ilog2(I915_VIDEO_AND_ENHANCE_CLASS_CAPABILITY_SFC)] = "sfc", +}; + +static const char * const vecs_caps[] = { + [ilog2(I915_VIDEO_AND_ENHANCE_CLASS_CAPABILITY_SFC)] = "sfc", +}; + +static ssize_t repr_trim(char *buf, ssize_t len) +{ + /* Trim off the trailing space and replace with a newline */ + if (len > PAGE_SIZE) + len = PAGE_SIZE; + if (len > 0) + buf[len - 1] = '\n'; + + return len; +} + +static ssize_t +__caps_show(struct intel_engine_cs *engine, + u32 caps, char *buf, bool show_unknown) +{ + const char * const *repr; + int count, n; + ssize_t len; + + BUILD_BUG_ON(!typecheck(typeof(caps), engine->uabi_capabilities)); + + switch (engine->class) { + case VIDEO_DECODE_CLASS: + repr = vcs_caps; + count = ARRAY_SIZE(vcs_caps); + break; + + case VIDEO_ENHANCEMENT_CLASS: + repr = vecs_caps; + count = ARRAY_SIZE(vecs_caps); + break; + + default: + repr = NULL; + count = 0; + break; + } + GEM_BUG_ON(count > BITS_PER_TYPE(typeof(caps))); + + len = 0; + for_each_set_bit(n, + (unsigned long *)&caps, + show_unknown ? BITS_PER_TYPE(typeof(caps)) : count) { + if (n >= count || !repr[n]) { + if (GEM_WARN_ON(show_unknown)) + len += snprintf(buf + len, PAGE_SIZE - len, + "[%x] ", n); + } else { + len += snprintf(buf + len, PAGE_SIZE - len, + "%s ", repr[n]); + } + if (GEM_WARN_ON(len >= PAGE_SIZE)) + break; + } + return repr_trim(buf, len); +} + +static ssize_t +caps_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) +{ + struct intel_engine_cs *engine = kobj_to_engine(kobj); + + return __caps_show(engine, engine->uabi_capabilities, buf, true); +} + +static struct kobj_attribute caps_attr = +__ATTR(capabilities, 0444, caps_show, NULL); + +static ssize_t +all_caps_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) +{ + return __caps_show(kobj_to_engine(kobj), -1, buf, false); +} + +static struct kobj_attribute all_caps_attr = +__ATTR(known_capabilities, 0444, all_caps_show, NULL); + +static ssize_t +max_spin_store(struct kobject *kobj, struct kobj_attribute *attr, + const char *buf, size_t count) +{ + struct intel_engine_cs *engine = kobj_to_engine(kobj); + unsigned long long duration; + int err; + + /* + * When waiting for a request, if is it currently being executed + * on the GPU, we busywait for a short while before sleeping. The + * premise is that most requests are short, and if it is already + * executing then there is a good chance that it will complete + * before we can setup the interrupt handler and go to sleep. + * We try to offset the cost of going to sleep, by first spinning + * on the request -- if it completed in less time than it would take + * to go sleep, process the interrupt and return back to the client, + * then we have saved the client some latency, albeit at the cost + * of spinning on an expensive CPU core. + * + * While we try to avoid waiting at all for a request that is unlikely + * to complete, deciding how long it is worth spinning is for is an + * arbitrary decision: trading off power vs latency. + */ + + err = kstrtoull(buf, 0, &duration); + if (err) + return err; + + if (duration > jiffies_to_nsecs(2)) + return -EINVAL; + + WRITE_ONCE(engine->props.max_busywait_duration_ns, duration); + + return count; +} + +static ssize_t +max_spin_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) +{ + struct intel_engine_cs *engine = kobj_to_engine(kobj); + + return sprintf(buf, "%lu\n", engine->props.max_busywait_duration_ns); +} + +static struct kobj_attribute max_spin_attr = +__ATTR(max_busywait_duration_ns, 0644, max_spin_show, max_spin_store); + +static ssize_t +timeslice_store(struct kobject *kobj, struct kobj_attribute *attr, + const char *buf, size_t count) +{ + struct intel_engine_cs *engine = kobj_to_engine(kobj); + unsigned long long duration; + int err; + + /* + * Execlists uses a scheduling quantum (a timeslice) to alternate + * execution between ready-to-run contexts of equal priority. This + * ensures that all users (though only if they of equal importance) + * have the opportunity to run and prevents livelocks where contexts + * may have implicit ordering due to userspace semaphores. + */ + + err = kstrtoull(buf, 0, &duration); + if (err) + return err; + + if (duration > jiffies_to_msecs(MAX_SCHEDULE_TIMEOUT)) + return -EINVAL; + + WRITE_ONCE(engine->props.timeslice_duration_ms, duration); + + if (execlists_active(&engine->execlists)) + set_timer_ms(&engine->execlists.timer, duration); + + return count; +} + +static ssize_t +timeslice_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) +{ + struct intel_engine_cs *engine = kobj_to_engine(kobj); + + return sprintf(buf, "%lu\n", engine->props.timeslice_duration_ms); +} + +static struct kobj_attribute timeslice_duration_attr = +__ATTR(timeslice_duration_ms, 0644, timeslice_show, timeslice_store); + +static ssize_t +stop_store(struct kobject *kobj, struct kobj_attribute *attr, + const char *buf, size_t count) +{ + struct intel_engine_cs *engine = kobj_to_engine(kobj); + unsigned long long duration; + int err; + + /* + * When we allow ourselves to sleep before a GPU reset after disabling + * submission, even for a few milliseconds, gives an innocent context + * the opportunity to clear the GPU before the reset occurs. However, + * how long to sleep depends on the typical non-preemptible duration + * (a similar problem to determining the ideal preempt-reset timeout + * or even the heartbeat interval). + */ + + err = kstrtoull(buf, 0, &duration); + if (err) + return err; + + if (duration > jiffies_to_msecs(MAX_SCHEDULE_TIMEOUT)) + return -EINVAL; + + WRITE_ONCE(engine->props.stop_timeout_ms, duration); + return count; +} + +static ssize_t +stop_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) +{ + struct intel_engine_cs *engine = kobj_to_engine(kobj); + + return sprintf(buf, "%lu\n", engine->props.stop_timeout_ms); +} + +static struct kobj_attribute stop_timeout_attr = +__ATTR(stop_timeout_ms, 0644, stop_show, stop_store); + +static ssize_t +preempt_timeout_store(struct kobject *kobj, struct kobj_attribute *attr, + const char *buf, size_t count) +{ + struct intel_engine_cs *engine = kobj_to_engine(kobj); + unsigned long long timeout; + int err; + + /* + * After initialising a preemption request, we give the current + * resident a small amount of time to vacate the GPU. The preemption + * request is for a higher priority context and should be immediate to + * maintain high quality of service (and avoid priority inversion). + * However, the preemption granularity of the GPU can be quite coarse + * and so we need a compromise. + */ + + err = kstrtoull(buf, 0, &timeout); + if (err) + return err; + + if (timeout > jiffies_to_msecs(MAX_SCHEDULE_TIMEOUT)) + return -EINVAL; + + WRITE_ONCE(engine->props.preempt_timeout_ms, timeout); + + if (READ_ONCE(engine->execlists.pending[0])) + set_timer_ms(&engine->execlists.preempt, timeout); + + return count; +} + +static ssize_t +preempt_timeout_show(struct kobject *kobj, struct kobj_attribute *attr, + char *buf) +{ + struct intel_engine_cs *engine = kobj_to_engine(kobj); + + return sprintf(buf, "%lu\n", engine->props.preempt_timeout_ms); +} + +static struct kobj_attribute preempt_timeout_attr = +__ATTR(preempt_timeout_ms, 0644, preempt_timeout_show, preempt_timeout_store); + +static ssize_t +heartbeat_store(struct kobject *kobj, struct kobj_attribute *attr, + const char *buf, size_t count) +{ + struct intel_engine_cs *engine = kobj_to_engine(kobj); + unsigned long long delay; + int err; + + /* + * We monitor the health of the system via periodic heartbeat pulses. + * The pulses also provide the opportunity to perform garbage + * collection. However, we interpret an incomplete pulse (a missed + * heartbeat) as an indication that the system is no longer responsive, + * i.e. hung, and perform an engine or full GPU reset. Given that the + * preemption granularity can be very coarse on a system, the optimal + * value for any workload is unknowable! + */ + + err = kstrtoull(buf, 0, &delay); + if (err) + return err; + + if (delay >= jiffies_to_msecs(MAX_SCHEDULE_TIMEOUT)) + return -EINVAL; + + err = intel_engine_set_heartbeat(engine, delay); + if (err) + return err; + + return count; +} + +static ssize_t +heartbeat_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) +{ + struct intel_engine_cs *engine = kobj_to_engine(kobj); + + return sprintf(buf, "%lu\n", engine->props.heartbeat_interval_ms); +} + +static struct kobj_attribute heartbeat_interval_attr = +__ATTR(heartbeat_interval_ms, 0644, heartbeat_show, heartbeat_store); + +static void kobj_engine_release(struct kobject *kobj) +{ + kfree(kobj); +} + +static struct kobj_type kobj_engine_type = { + .release = kobj_engine_release, + .sysfs_ops = &kobj_sysfs_ops +}; + +static struct kobject * +kobj_engine(struct kobject *dir, struct intel_engine_cs *engine) +{ + struct kobj_engine *ke; + + ke = kzalloc(sizeof(*ke), GFP_KERNEL); + if (!ke) + return NULL; + + kobject_init(&ke->base, &kobj_engine_type); + ke->engine = engine; + + if (kobject_add(&ke->base, dir, "%s", engine->name)) { + kobject_put(&ke->base); + return NULL; + } + + /* xfer ownership to sysfs tree */ + return &ke->base; +} + +void intel_engines_add_sysfs(struct drm_i915_private *i915) +{ + static const struct attribute *files[] = { + &name_attr.attr, + &class_attr.attr, + &inst_attr.attr, + &mmio_attr.attr, + &caps_attr.attr, + &all_caps_attr.attr, + &max_spin_attr.attr, + &stop_timeout_attr.attr, +#if CONFIG_DRM_I915_HEARTBEAT_INTERVAL + &heartbeat_interval_attr.attr, +#endif + NULL + }; + + struct device *kdev = i915->drm.primary->kdev; + struct intel_engine_cs *engine; + struct kobject *dir; + + dir = kobject_create_and_add("engine", &kdev->kobj); + if (!dir) + return; + + for_each_uabi_engine(engine, i915) { + struct kobject *kobj; + + kobj = kobj_engine(dir, engine); + if (!kobj) + goto err_engine; + + if (sysfs_create_files(kobj, files)) + goto err_object; + + if (intel_engine_has_timeslices(engine) && + sysfs_create_file(kobj, ×lice_duration_attr.attr)) + goto err_engine; + + if (intel_engine_has_preempt_reset(engine) && + sysfs_create_file(kobj, &preempt_timeout_attr.attr)) + goto err_engine; + + if (0) { +err_object: + kobject_put(kobj); +err_engine: + dev_err(kdev, "Failed to add sysfs engine '%s'\n", + engine->name); + break; + } + } +} diff --git a/drivers/gpu/drm/i915/gt/sysfs_engines.h b/drivers/gpu/drm/i915/gt/sysfs_engines.h new file mode 100644 index 000000000000..9546fffe03a7 --- /dev/null +++ b/drivers/gpu/drm/i915/gt/sysfs_engines.h @@ -0,0 +1,13 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2019 Intel Corporation + */ + +#ifndef INTEL_ENGINE_SYSFS_H +#define INTEL_ENGINE_SYSFS_H + +struct drm_i915_private; + +void intel_engines_add_sysfs(struct drm_i915_private *i915); + +#endif /* INTEL_ENGINE_SYSFS_H */ diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c index 1beaa77f9bb6..fe7778c28d2d 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c @@ -456,9 +456,7 @@ static void guc_reset_cancel(struct intel_engine_cs *engine) /* Mark all executing requests as skipped. */ list_for_each_entry(rq, &engine->active.requests, sched.link) { - if (!i915_request_signaled(rq)) - dma_fence_set_error(&rq->fence, -EIO); - + i915_request_set_error_once(rq, -EIO); i915_request_mark_complete(rq); } diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c index 5434c07aefa1..18c755203688 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c @@ -43,7 +43,7 @@ void intel_uc_fw_change_status(struct intel_uc_fw *uc_fw, * features. */ #define INTEL_UC_FIRMWARE_DEFS(fw_def, guc_def, huc_def) \ - fw_def(TIGERLAKE, 0, guc_def(tgl, 35, 2, 0), huc_def(tgl, 7, 0, 3)) \ + fw_def(TIGERLAKE, 0, guc_def(tgl, 35, 2, 0), huc_def(tgl, 7, 0, 12)) \ fw_def(ELKHARTLAKE, 0, guc_def(ehl, 33, 0, 4), huc_def(ehl, 9, 0, 0)) \ fw_def(ICELAKE, 0, guc_def(icl, 33, 0, 0), huc_def(icl, 9, 0, 0)) \ fw_def(COFFEELAKE, 5, guc_def(cml, 33, 0, 0), huc_def(cml, 4, 0, 0)) \ diff --git a/drivers/gpu/drm/i915/gvt/aperture_gm.c b/drivers/gpu/drm/i915/gvt/aperture_gm.c index 771420453f82..8b13f091cee2 100644 --- a/drivers/gpu/drm/i915/gvt/aperture_gm.c +++ b/drivers/gpu/drm/i915/gvt/aperture_gm.c @@ -41,7 +41,7 @@ static int alloc_gm(struct intel_vgpu *vgpu, bool high_gm) { struct intel_gvt *gvt = vgpu->gvt; - struct drm_i915_private *dev_priv = gvt->dev_priv; + struct intel_gt *gt = gvt->gt; unsigned int flags; u64 start, end, size; struct drm_mm_node *node; @@ -61,14 +61,14 @@ static int alloc_gm(struct intel_vgpu *vgpu, bool high_gm) flags = PIN_MAPPABLE; } - mutex_lock(&dev_priv->ggtt.vm.mutex); - mmio_hw_access_pre(dev_priv); - ret = i915_gem_gtt_insert(&dev_priv->ggtt.vm, node, + mutex_lock(>->ggtt->vm.mutex); + mmio_hw_access_pre(gt); + ret = i915_gem_gtt_insert(>->ggtt->vm, node, size, I915_GTT_PAGE_SIZE, I915_COLOR_UNEVICTABLE, start, end, flags); - mmio_hw_access_post(dev_priv); - mutex_unlock(&dev_priv->ggtt.vm.mutex); + mmio_hw_access_post(gt); + mutex_unlock(>->ggtt->vm.mutex); if (ret) gvt_err("fail to alloc %s gm space from host\n", high_gm ? "high" : "low"); @@ -79,7 +79,7 @@ static int alloc_gm(struct intel_vgpu *vgpu, bool high_gm) static int alloc_vgpu_gm(struct intel_vgpu *vgpu) { struct intel_gvt *gvt = vgpu->gvt; - struct drm_i915_private *dev_priv = gvt->dev_priv; + struct intel_gt *gt = gvt->gt; int ret; ret = alloc_gm(vgpu, false); @@ -98,20 +98,21 @@ static int alloc_vgpu_gm(struct intel_vgpu *vgpu) return 0; out_free_aperture: - mutex_lock(&dev_priv->ggtt.vm.mutex); + mutex_lock(>->ggtt->vm.mutex); drm_mm_remove_node(&vgpu->gm.low_gm_node); - mutex_unlock(&dev_priv->ggtt.vm.mutex); + mutex_unlock(>->ggtt->vm.mutex); return ret; } static void free_vgpu_gm(struct intel_vgpu *vgpu) { - struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv; + struct intel_gvt *gvt = vgpu->gvt; + struct intel_gt *gt = gvt->gt; - mutex_lock(&dev_priv->ggtt.vm.mutex); + mutex_lock(>->ggtt->vm.mutex); drm_mm_remove_node(&vgpu->gm.low_gm_node); drm_mm_remove_node(&vgpu->gm.high_gm_node); - mutex_unlock(&dev_priv->ggtt.vm.mutex); + mutex_unlock(>->ggtt->vm.mutex); } /** @@ -128,28 +129,29 @@ void intel_vgpu_write_fence(struct intel_vgpu *vgpu, u32 fence, u64 value) { struct intel_gvt *gvt = vgpu->gvt; - struct drm_i915_private *dev_priv = gvt->dev_priv; + struct drm_i915_private *i915 = gvt->gt->i915; + struct intel_uncore *uncore = gvt->gt->uncore; struct i915_fence_reg *reg; i915_reg_t fence_reg_lo, fence_reg_hi; - assert_rpm_wakelock_held(&dev_priv->runtime_pm); + assert_rpm_wakelock_held(uncore->rpm); - if (WARN_ON(fence >= vgpu_fence_sz(vgpu))) + if (drm_WARN_ON(&i915->drm, fence >= vgpu_fence_sz(vgpu))) return; reg = vgpu->fence.regs[fence]; - if (WARN_ON(!reg)) + if (drm_WARN_ON(&i915->drm, !reg)) return; fence_reg_lo = FENCE_REG_GEN6_LO(reg->id); fence_reg_hi = FENCE_REG_GEN6_HI(reg->id); - I915_WRITE(fence_reg_lo, 0); - POSTING_READ(fence_reg_lo); + intel_uncore_write(uncore, fence_reg_lo, 0); + intel_uncore_posting_read(uncore, fence_reg_lo); - I915_WRITE(fence_reg_hi, upper_32_bits(value)); - I915_WRITE(fence_reg_lo, lower_32_bits(value)); - POSTING_READ(fence_reg_lo); + intel_uncore_write(uncore, fence_reg_hi, upper_32_bits(value)); + intel_uncore_write(uncore, fence_reg_lo, lower_32_bits(value)); + intel_uncore_posting_read(uncore, fence_reg_lo); } static void _clear_vgpu_fence(struct intel_vgpu *vgpu) @@ -163,42 +165,43 @@ static void _clear_vgpu_fence(struct intel_vgpu *vgpu) static void free_vgpu_fence(struct intel_vgpu *vgpu) { struct intel_gvt *gvt = vgpu->gvt; - struct drm_i915_private *dev_priv = gvt->dev_priv; + struct intel_uncore *uncore = gvt->gt->uncore; struct i915_fence_reg *reg; + intel_wakeref_t wakeref; u32 i; - if (WARN_ON(!vgpu_fence_sz(vgpu))) + if (drm_WARN_ON(&gvt->gt->i915->drm, !vgpu_fence_sz(vgpu))) return; - intel_runtime_pm_get(&dev_priv->runtime_pm); + wakeref = intel_runtime_pm_get(uncore->rpm); - mutex_lock(&dev_priv->ggtt.vm.mutex); + mutex_lock(&gvt->gt->ggtt->vm.mutex); _clear_vgpu_fence(vgpu); for (i = 0; i < vgpu_fence_sz(vgpu); i++) { reg = vgpu->fence.regs[i]; i915_unreserve_fence(reg); vgpu->fence.regs[i] = NULL; } - mutex_unlock(&dev_priv->ggtt.vm.mutex); + mutex_unlock(&gvt->gt->ggtt->vm.mutex); - intel_runtime_pm_put_unchecked(&dev_priv->runtime_pm); + intel_runtime_pm_put(uncore->rpm, wakeref); } static int alloc_vgpu_fence(struct intel_vgpu *vgpu) { struct intel_gvt *gvt = vgpu->gvt; - struct drm_i915_private *dev_priv = gvt->dev_priv; - struct intel_runtime_pm *rpm = &dev_priv->runtime_pm; + struct intel_uncore *uncore = gvt->gt->uncore; struct i915_fence_reg *reg; + intel_wakeref_t wakeref; int i; - intel_runtime_pm_get(rpm); + wakeref = intel_runtime_pm_get(uncore->rpm); /* Request fences from host */ - mutex_lock(&dev_priv->ggtt.vm.mutex); + mutex_lock(&gvt->gt->ggtt->vm.mutex); for (i = 0; i < vgpu_fence_sz(vgpu); i++) { - reg = i915_reserve_fence(&dev_priv->ggtt); + reg = i915_reserve_fence(gvt->gt->ggtt); if (IS_ERR(reg)) goto out_free_fence; @@ -207,9 +210,10 @@ static int alloc_vgpu_fence(struct intel_vgpu *vgpu) _clear_vgpu_fence(vgpu); - mutex_unlock(&dev_priv->ggtt.vm.mutex); - intel_runtime_pm_put_unchecked(rpm); + mutex_unlock(&gvt->gt->ggtt->vm.mutex); + intel_runtime_pm_put(uncore->rpm, wakeref); return 0; + out_free_fence: gvt_vgpu_err("Failed to alloc fences\n"); /* Return fences to host, if fail */ @@ -220,8 +224,8 @@ out_free_fence: i915_unreserve_fence(reg); vgpu->fence.regs[i] = NULL; } - mutex_unlock(&dev_priv->ggtt.vm.mutex); - intel_runtime_pm_put_unchecked(rpm); + mutex_unlock(&gvt->gt->ggtt->vm.mutex); + intel_runtime_pm_put_unchecked(uncore->rpm); return -ENOSPC; } @@ -315,11 +319,11 @@ void intel_vgpu_free_resource(struct intel_vgpu *vgpu) */ void intel_vgpu_reset_resource(struct intel_vgpu *vgpu) { - struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv; + struct intel_gvt *gvt = vgpu->gvt; + intel_wakeref_t wakeref; - intel_runtime_pm_get(&dev_priv->runtime_pm); - _clear_vgpu_fence(vgpu); - intel_runtime_pm_put_unchecked(&dev_priv->runtime_pm); + with_intel_runtime_pm(gvt->gt->uncore->rpm, wakeref) + _clear_vgpu_fence(vgpu); } /** diff --git a/drivers/gpu/drm/i915/gvt/cfg_space.c b/drivers/gpu/drm/i915/gvt/cfg_space.c index 19cf1bbe059d..072725a448db 100644 --- a/drivers/gpu/drm/i915/gvt/cfg_space.c +++ b/drivers/gpu/drm/i915/gvt/cfg_space.c @@ -106,10 +106,13 @@ static void vgpu_pci_cfg_mem_write(struct intel_vgpu *vgpu, unsigned int off, int intel_vgpu_emulate_cfg_read(struct intel_vgpu *vgpu, unsigned int offset, void *p_data, unsigned int bytes) { - if (WARN_ON(bytes > 4)) + struct drm_i915_private *i915 = vgpu->gvt->gt->i915; + + if (drm_WARN_ON(&i915->drm, bytes > 4)) return -EINVAL; - if (WARN_ON(offset + bytes > vgpu->gvt->device_info.cfg_space_size)) + if (drm_WARN_ON(&i915->drm, + offset + bytes > vgpu->gvt->device_info.cfg_space_size)) return -EINVAL; memcpy(p_data, vgpu_cfg_space(vgpu) + offset, bytes); @@ -297,34 +300,36 @@ static int emulate_pci_bar_write(struct intel_vgpu *vgpu, unsigned int offset, int intel_vgpu_emulate_cfg_write(struct intel_vgpu *vgpu, unsigned int offset, void *p_data, unsigned int bytes) { + struct drm_i915_private *i915 = vgpu->gvt->gt->i915; int ret; - if (WARN_ON(bytes > 4)) + if (drm_WARN_ON(&i915->drm, bytes > 4)) return -EINVAL; - if (WARN_ON(offset + bytes > vgpu->gvt->device_info.cfg_space_size)) + if (drm_WARN_ON(&i915->drm, + offset + bytes > vgpu->gvt->device_info.cfg_space_size)) return -EINVAL; /* First check if it's PCI_COMMAND */ if (IS_ALIGNED(offset, 2) && offset == PCI_COMMAND) { - if (WARN_ON(bytes > 2)) + if (drm_WARN_ON(&i915->drm, bytes > 2)) return -EINVAL; return emulate_pci_command_write(vgpu, offset, p_data, bytes); } switch (rounddown(offset, 4)) { case PCI_ROM_ADDRESS: - if (WARN_ON(!IS_ALIGNED(offset, 4))) + if (drm_WARN_ON(&i915->drm, !IS_ALIGNED(offset, 4))) return -EINVAL; return emulate_pci_rom_bar_write(vgpu, offset, p_data, bytes); case PCI_BASE_ADDRESS_0 ... PCI_BASE_ADDRESS_5: - if (WARN_ON(!IS_ALIGNED(offset, 4))) + if (drm_WARN_ON(&i915->drm, !IS_ALIGNED(offset, 4))) return -EINVAL; return emulate_pci_bar_write(vgpu, offset, p_data, bytes); case INTEL_GVT_PCI_SWSCI: - if (WARN_ON(!IS_ALIGNED(offset, 4))) + if (drm_WARN_ON(&i915->drm, !IS_ALIGNED(offset, 4))) return -EINVAL; ret = intel_vgpu_emulate_opregion_request(vgpu, *(u32 *)p_data); if (ret) @@ -332,7 +337,7 @@ int intel_vgpu_emulate_cfg_write(struct intel_vgpu *vgpu, unsigned int offset, break; case INTEL_GVT_PCI_OPREGION: - if (WARN_ON(!IS_ALIGNED(offset, 4))) + if (drm_WARN_ON(&i915->drm, !IS_ALIGNED(offset, 4))) return -EINVAL; ret = intel_vgpu_opregion_base_write_handler(vgpu, *(u32 *)p_data); @@ -391,9 +396,9 @@ void intel_vgpu_init_cfg_space(struct intel_vgpu *vgpu, memset(vgpu_cfg_space(vgpu) + INTEL_GVT_PCI_OPREGION, 0, 4); vgpu->cfg_space.bar[INTEL_GVT_PCI_BAR_GTTMMIO].size = - pci_resource_len(gvt->dev_priv->drm.pdev, 0); + pci_resource_len(gvt->gt->i915->drm.pdev, 0); vgpu->cfg_space.bar[INTEL_GVT_PCI_BAR_APERTURE].size = - pci_resource_len(gvt->dev_priv->drm.pdev, 2); + pci_resource_len(gvt->gt->i915->drm.pdev, 2); memset(vgpu_cfg_space(vgpu) + PCI_ROM_ADDRESS, 0, 4); } diff --git a/drivers/gpu/drm/i915/gvt/cmd_parser.c b/drivers/gpu/drm/i915/gvt/cmd_parser.c index 21a176cd8acc..9e065ad0658f 100644 --- a/drivers/gpu/drm/i915/gvt/cmd_parser.c +++ b/drivers/gpu/drm/i915/gvt/cmd_parser.c @@ -462,7 +462,7 @@ enum { struct parser_exec_state { struct intel_vgpu *vgpu; - int ring_id; + const struct intel_engine_cs *engine; int buf_type; @@ -635,39 +635,42 @@ static const struct decode_info *ring_decode_info[I915_NUM_ENGINES][8] = { }, }; -static inline u32 get_opcode(u32 cmd, int ring_id) +static inline u32 get_opcode(u32 cmd, const struct intel_engine_cs *engine) { const struct decode_info *d_info; - d_info = ring_decode_info[ring_id][CMD_TYPE(cmd)]; + d_info = ring_decode_info[engine->id][CMD_TYPE(cmd)]; if (d_info == NULL) return INVALID_OP; return cmd >> (32 - d_info->op_len); } -static inline const struct cmd_info *find_cmd_entry(struct intel_gvt *gvt, - unsigned int opcode, int ring_id) +static inline const struct cmd_info * +find_cmd_entry(struct intel_gvt *gvt, unsigned int opcode, + const struct intel_engine_cs *engine) { struct cmd_entry *e; hash_for_each_possible(gvt->cmd_table, e, hlist, opcode) { - if (opcode == e->info->opcode && e->info->rings & BIT(ring_id)) + if (opcode == e->info->opcode && + e->info->rings & engine->mask) return e->info; } return NULL; } -static inline const struct cmd_info *get_cmd_info(struct intel_gvt *gvt, - u32 cmd, int ring_id) +static inline const struct cmd_info * +get_cmd_info(struct intel_gvt *gvt, u32 cmd, + const struct intel_engine_cs *engine) { u32 opcode; - opcode = get_opcode(cmd, ring_id); + opcode = get_opcode(cmd, engine); if (opcode == INVALID_OP) return NULL; - return find_cmd_entry(gvt, opcode, ring_id); + return find_cmd_entry(gvt, opcode, engine); } static inline u32 sub_op_val(u32 cmd, u32 hi, u32 low) @@ -675,12 +678,12 @@ static inline u32 sub_op_val(u32 cmd, u32 hi, u32 low) return (cmd >> low) & ((1U << (hi - low + 1)) - 1); } -static inline void print_opcode(u32 cmd, int ring_id) +static inline void print_opcode(u32 cmd, const struct intel_engine_cs *engine) { const struct decode_info *d_info; int i; - d_info = ring_decode_info[ring_id][CMD_TYPE(cmd)]; + d_info = ring_decode_info[engine->id][CMD_TYPE(cmd)]; if (d_info == NULL) return; @@ -709,10 +712,11 @@ static void parser_exec_state_dump(struct parser_exec_state *s) int cnt = 0; int i; - gvt_dbg_cmd(" vgpu%d RING%d: ring_start(%08lx) ring_end(%08lx)" - " ring_head(%08lx) ring_tail(%08lx)\n", s->vgpu->id, - s->ring_id, s->ring_start, s->ring_start + s->ring_size, - s->ring_head, s->ring_tail); + gvt_dbg_cmd(" vgpu%d RING%s: ring_start(%08lx) ring_end(%08lx)" + " ring_head(%08lx) ring_tail(%08lx)\n", + s->vgpu->id, s->engine->name, + s->ring_start, s->ring_start + s->ring_size, + s->ring_head, s->ring_tail); gvt_dbg_cmd(" %s %s ip_gma(%08lx) ", s->buf_type == RING_BUFFER_INSTRUCTION ? @@ -729,7 +733,7 @@ static void parser_exec_state_dump(struct parser_exec_state *s) s->ip_va, cmd_val(s, 0), cmd_val(s, 1), cmd_val(s, 2), cmd_val(s, 3)); - print_opcode(cmd_val(s, 0), s->ring_id); + print_opcode(cmd_val(s, 0), s->engine); s->ip_va = (u32 *)((((u64)s->ip_va) >> 12) << 12); @@ -840,7 +844,6 @@ static int force_nonpriv_reg_handler(struct parser_exec_state *s, unsigned int data; u32 ring_base; u32 nopid; - struct drm_i915_private *dev_priv = s->vgpu->gvt->dev_priv; if (!strcmp(cmd, "lri")) data = cmd_val(s, index + 1); @@ -850,7 +853,7 @@ static int force_nonpriv_reg_handler(struct parser_exec_state *s, return -EINVAL; } - ring_base = dev_priv->engine[s->ring_id]->mmio_base; + ring_base = s->engine->mmio_base; nopid = i915_mmio_reg_offset(RING_NOPID(ring_base)); if (!intel_gvt_in_force_nonpriv_whitelist(gvt, data) && @@ -926,9 +929,9 @@ static int cmd_reg_handler(struct parser_exec_state *s, * update reg values in it into vregs, so LRIs in workload with * inhibit context will restore with correct values */ - if (IS_GEN(gvt->dev_priv, 9) && - intel_gvt_mmio_is_in_ctx(gvt, offset) && - !strncmp(cmd, "lri", 3)) { + if (IS_GEN(s->engine->i915, 9) && + intel_gvt_mmio_is_in_ctx(gvt, offset) && + !strncmp(cmd, "lri", 3)) { intel_gvt_hypervisor_read_gpa(s->vgpu, s->workload->ring_context_gpa + 12, &ctx_sr_ctl, 4); /* check inhibit context */ @@ -964,7 +967,6 @@ static int cmd_handler_lri(struct parser_exec_state *s) { int i, ret = 0; int cmd_len = cmd_length(s); - struct intel_gvt *gvt = s->vgpu->gvt; u32 valid_len = CMD_LEN(1); /* @@ -979,8 +981,8 @@ static int cmd_handler_lri(struct parser_exec_state *s) } for (i = 1; i < cmd_len; i += 2) { - if (IS_BROADWELL(gvt->dev_priv) && s->ring_id != RCS0) { - if (s->ring_id == BCS0 && + if (IS_BROADWELL(s->engine->i915) && s->engine->id != RCS0) { + if (s->engine->id == BCS0 && cmd_reg(s, i) == i915_mmio_reg_offset(DERRMR)) ret |= 0; else @@ -1001,9 +1003,9 @@ static int cmd_handler_lrr(struct parser_exec_state *s) int cmd_len = cmd_length(s); for (i = 1; i < cmd_len; i += 2) { - if (IS_BROADWELL(s->vgpu->gvt->dev_priv)) + if (IS_BROADWELL(s->engine->i915)) ret |= ((cmd_reg_inhibit(s, i) || - (cmd_reg_inhibit(s, i + 1)))) ? + (cmd_reg_inhibit(s, i + 1)))) ? -EBADRQC : 0; if (ret) break; @@ -1029,7 +1031,7 @@ static int cmd_handler_lrm(struct parser_exec_state *s) int cmd_len = cmd_length(s); for (i = 1; i < cmd_len;) { - if (IS_BROADWELL(gvt->dev_priv)) + if (IS_BROADWELL(s->engine->i915)) ret |= (cmd_reg_inhibit(s, i)) ? -EBADRQC : 0; if (ret) break; @@ -1141,7 +1143,7 @@ static int cmd_handler_pipe_control(struct parser_exec_state *s) if (ret) return ret; if (index_mode) { - hws_pga = s->vgpu->hws_pga[s->ring_id]; + hws_pga = s->vgpu->hws_pga[s->engine->id]; gma = hws_pga + gma; patch_value(s, cmd_ptr(s, 2), gma); val = cmd_val(s, 1) & (~(1 << 21)); @@ -1155,15 +1157,15 @@ static int cmd_handler_pipe_control(struct parser_exec_state *s) return ret; if (cmd_val(s, 1) & PIPE_CONTROL_NOTIFY) - set_bit(cmd_interrupt_events[s->ring_id].pipe_control_notify, - s->workload->pending_events); + set_bit(cmd_interrupt_events[s->engine->id].pipe_control_notify, + s->workload->pending_events); return 0; } static int cmd_handler_mi_user_interrupt(struct parser_exec_state *s) { - set_bit(cmd_interrupt_events[s->ring_id].mi_user_interrupt, - s->workload->pending_events); + set_bit(cmd_interrupt_events[s->engine->id].mi_user_interrupt, + s->workload->pending_events); patch_value(s, cmd_ptr(s, 0), MI_NOOP); return 0; } @@ -1213,7 +1215,7 @@ struct plane_code_mapping { static int gen8_decode_mi_display_flip(struct parser_exec_state *s, struct mi_display_flip_command_info *info) { - struct drm_i915_private *dev_priv = s->vgpu->gvt->dev_priv; + struct drm_i915_private *dev_priv = s->engine->i915; struct plane_code_mapping gen8_plane_code[] = { [0] = {PIPE_A, PLANE_A, PRIMARY_A_FLIP_DONE}, [1] = {PIPE_B, PLANE_A, PRIMARY_B_FLIP_DONE}, @@ -1230,7 +1232,7 @@ static int gen8_decode_mi_display_flip(struct parser_exec_state *s, dword2 = cmd_val(s, 2); v = (dword0 & GENMASK(21, 19)) >> 19; - if (WARN_ON(v >= ARRAY_SIZE(gen8_plane_code))) + if (drm_WARN_ON(&dev_priv->drm, v >= ARRAY_SIZE(gen8_plane_code))) return -EBADRQC; info->pipe = gen8_plane_code[v].pipe; @@ -1250,7 +1252,7 @@ static int gen8_decode_mi_display_flip(struct parser_exec_state *s, info->stride_reg = SPRSTRIDE(info->pipe); info->surf_reg = SPRSURF(info->pipe); } else { - WARN_ON(1); + drm_WARN_ON(&dev_priv->drm, 1); return -EBADRQC; } return 0; @@ -1259,7 +1261,7 @@ static int gen8_decode_mi_display_flip(struct parser_exec_state *s, static int skl_decode_mi_display_flip(struct parser_exec_state *s, struct mi_display_flip_command_info *info) { - struct drm_i915_private *dev_priv = s->vgpu->gvt->dev_priv; + struct drm_i915_private *dev_priv = s->engine->i915; struct intel_vgpu *vgpu = s->vgpu; u32 dword0 = cmd_val(s, 0); u32 dword1 = cmd_val(s, 1); @@ -1318,13 +1320,12 @@ static int skl_decode_mi_display_flip(struct parser_exec_state *s, static int gen8_check_mi_display_flip(struct parser_exec_state *s, struct mi_display_flip_command_info *info) { - struct drm_i915_private *dev_priv = s->vgpu->gvt->dev_priv; u32 stride, tile; if (!info->async_flip) return 0; - if (INTEL_GEN(dev_priv) >= 9) { + if (INTEL_GEN(s->engine->i915) >= 9) { stride = vgpu_vreg_t(s->vgpu, info->stride_reg) & GENMASK(9, 0); tile = (vgpu_vreg_t(s->vgpu, info->ctrl_reg) & GENMASK(12, 10)) >> 10; @@ -1347,7 +1348,7 @@ static int gen8_update_plane_mmio_from_mi_display_flip( struct parser_exec_state *s, struct mi_display_flip_command_info *info) { - struct drm_i915_private *dev_priv = s->vgpu->gvt->dev_priv; + struct drm_i915_private *dev_priv = s->engine->i915; struct intel_vgpu *vgpu = s->vgpu; set_mask_bits(&vgpu_vreg_t(vgpu, info->surf_reg), GENMASK(31, 12), @@ -1378,11 +1379,9 @@ static int gen8_update_plane_mmio_from_mi_display_flip( static int decode_mi_display_flip(struct parser_exec_state *s, struct mi_display_flip_command_info *info) { - struct drm_i915_private *dev_priv = s->vgpu->gvt->dev_priv; - - if (IS_BROADWELL(dev_priv)) + if (IS_BROADWELL(s->engine->i915)) return gen8_decode_mi_display_flip(s, info); - if (INTEL_GEN(dev_priv) >= 9) + if (INTEL_GEN(s->engine->i915) >= 9) return skl_decode_mi_display_flip(s, info); return -ENODEV; @@ -1667,7 +1666,7 @@ static int cmd_handler_mi_flush_dw(struct parser_exec_state *s) if (ret) return ret; if (index_mode) { - hws_pga = s->vgpu->hws_pga[s->ring_id]; + hws_pga = s->vgpu->hws_pga[s->engine->id]; gma = hws_pga + gma; patch_value(s, cmd_ptr(s, 1), gma); val = cmd_val(s, 0) & (~(1 << 21)); @@ -1676,8 +1675,8 @@ static int cmd_handler_mi_flush_dw(struct parser_exec_state *s) } /* Check notify bit */ if ((cmd_val(s, 0) & (1 << 8))) - set_bit(cmd_interrupt_events[s->ring_id].mi_flush_dw, - s->workload->pending_events); + set_bit(cmd_interrupt_events[s->engine->id].mi_flush_dw, + s->workload->pending_events); return ret; } @@ -1725,12 +1724,18 @@ static int copy_gma_to_hva(struct intel_vgpu *vgpu, struct intel_vgpu_mm *mm, static int batch_buffer_needs_scan(struct parser_exec_state *s) { /* Decide privilege based on address space */ - if (cmd_val(s, 0) & (1 << 8) && - !(s->vgpu->scan_nonprivbb & (1 << s->ring_id))) + if (cmd_val(s, 0) & BIT(8) && + !(s->vgpu->scan_nonprivbb & s->engine->mask)) return 0; + return 1; } +static const char *repr_addr_type(unsigned int type) +{ + return type == PPGTT_BUFFER ? "ppgtt" : "ggtt"; +} + static int find_bb_size(struct parser_exec_state *s, unsigned long *bb_size, unsigned long *bb_end_cmd_offset) @@ -1753,24 +1758,24 @@ static int find_bb_size(struct parser_exec_state *s, return -EFAULT; cmd = cmd_val(s, 0); - info = get_cmd_info(s->vgpu->gvt, cmd, s->ring_id); + info = get_cmd_info(s->vgpu->gvt, cmd, s->engine); if (info == NULL) { - gvt_vgpu_err("unknown cmd 0x%x, opcode=0x%x, addr_type=%s, ring %d, workload=%p\n", - cmd, get_opcode(cmd, s->ring_id), - (s->buf_addr_type == PPGTT_BUFFER) ? - "ppgtt" : "ggtt", s->ring_id, s->workload); + gvt_vgpu_err("unknown cmd 0x%x, opcode=0x%x, addr_type=%s, ring %s, workload=%p\n", + cmd, get_opcode(cmd, s->engine), + repr_addr_type(s->buf_addr_type), + s->engine->name, s->workload); return -EBADRQC; } do { if (copy_gma_to_hva(s->vgpu, mm, - gma, gma + 4, &cmd) < 0) + gma, gma + 4, &cmd) < 0) return -EFAULT; - info = get_cmd_info(s->vgpu->gvt, cmd, s->ring_id); + info = get_cmd_info(s->vgpu->gvt, cmd, s->engine); if (info == NULL) { - gvt_vgpu_err("unknown cmd 0x%x, opcode=0x%x, addr_type=%s, ring %d, workload=%p\n", - cmd, get_opcode(cmd, s->ring_id), - (s->buf_addr_type == PPGTT_BUFFER) ? - "ppgtt" : "ggtt", s->ring_id, s->workload); + gvt_vgpu_err("unknown cmd 0x%x, opcode=0x%x, addr_type=%s, ring %s, workload=%p\n", + cmd, get_opcode(cmd, s->engine), + repr_addr_type(s->buf_addr_type), + s->engine->name, s->workload); return -EBADRQC; } @@ -1799,12 +1804,12 @@ static int audit_bb_end(struct parser_exec_state *s, void *va) u32 cmd = *(u32 *)va; const struct cmd_info *info; - info = get_cmd_info(s->vgpu->gvt, cmd, s->ring_id); + info = get_cmd_info(s->vgpu->gvt, cmd, s->engine); if (info == NULL) { - gvt_vgpu_err("unknown cmd 0x%x, opcode=0x%x, addr_type=%s, ring %d, workload=%p\n", - cmd, get_opcode(cmd, s->ring_id), - (s->buf_addr_type == PPGTT_BUFFER) ? - "ppgtt" : "ggtt", s->ring_id, s->workload); + gvt_vgpu_err("unknown cmd 0x%x, opcode=0x%x, addr_type=%s, ring %s, workload=%p\n", + cmd, get_opcode(cmd, s->engine), + repr_addr_type(s->buf_addr_type), + s->engine->name, s->workload); return -EBADRQC; } @@ -1857,7 +1862,7 @@ static int perform_bb_shadow(struct parser_exec_state *s) if (bb->ppgtt) start_offset = gma & ~I915_GTT_PAGE_MASK; - bb->obj = i915_gem_object_create_shmem(s->vgpu->gvt->dev_priv, + bb->obj = i915_gem_object_create_shmem(s->engine->i915, round_up(bb_size + start_offset, PAGE_SIZE)); if (IS_ERR(bb->obj)) { @@ -2666,25 +2671,25 @@ static int cmd_parser_exec(struct parser_exec_state *s) if (cmd == MI_NOOP) info = &cmd_info[mi_noop_index]; else - info = get_cmd_info(s->vgpu->gvt, cmd, s->ring_id); + info = get_cmd_info(s->vgpu->gvt, cmd, s->engine); if (info == NULL) { - gvt_vgpu_err("unknown cmd 0x%x, opcode=0x%x, addr_type=%s, ring %d, workload=%p\n", - cmd, get_opcode(cmd, s->ring_id), - (s->buf_addr_type == PPGTT_BUFFER) ? - "ppgtt" : "ggtt", s->ring_id, s->workload); + gvt_vgpu_err("unknown cmd 0x%x, opcode=0x%x, addr_type=%s, ring %s, workload=%p\n", + cmd, get_opcode(cmd, s->engine), + repr_addr_type(s->buf_addr_type), + s->engine->name, s->workload); return -EBADRQC; } s->info = info; - trace_gvt_command(vgpu->id, s->ring_id, s->ip_gma, s->ip_va, + trace_gvt_command(vgpu->id, s->engine->id, s->ip_gma, s->ip_va, cmd_length(s), s->buf_type, s->buf_addr_type, s->workload, info->name); if ((info->flag & F_LEN_MASK) == F_LEN_VAR_FIXED) { ret = gvt_check_valid_cmd_length(cmd_length(s), - info->valid_len); + info->valid_len); if (ret) return ret; } @@ -2781,7 +2786,7 @@ static int scan_workload(struct intel_vgpu_workload *workload) s.buf_type = RING_BUFFER_INSTRUCTION; s.buf_addr_type = GTT_BUFFER; s.vgpu = workload->vgpu; - s.ring_id = workload->ring_id; + s.engine = workload->engine; s.ring_start = workload->rb_start; s.ring_size = _RING_CTL_BUF_SIZE(workload->rb_ctl); s.ring_head = gma_head; @@ -2790,8 +2795,7 @@ static int scan_workload(struct intel_vgpu_workload *workload) s.workload = workload; s.is_ctx_wa = false; - if ((bypass_scan_mask & (1 << workload->ring_id)) || - gma_head == gma_tail) + if (bypass_scan_mask & workload->engine->mask || gma_head == gma_tail) return 0; ret = ip_gma_set(&s, gma_head); @@ -2830,7 +2834,7 @@ static int scan_wa_ctx(struct intel_shadow_wa_ctx *wa_ctx) s.buf_type = RING_BUFFER_INSTRUCTION; s.buf_addr_type = GTT_BUFFER; s.vgpu = workload->vgpu; - s.ring_id = workload->ring_id; + s.engine = workload->engine; s.ring_start = wa_ctx->indirect_ctx.guest_gma; s.ring_size = ring_size; s.ring_head = gma_head; @@ -2855,7 +2859,6 @@ static int shadow_workload_ring_buffer(struct intel_vgpu_workload *workload) struct intel_vgpu_submission *s = &vgpu->submission; unsigned long gma_head, gma_tail, gma_top, guest_rb_size; void *shadow_ring_buffer_va; - int ring_id = workload->ring_id; int ret; guest_rb_size = _RING_CTL_BUF_SIZE(workload->rb_ctl); @@ -2868,21 +2871,21 @@ static int shadow_workload_ring_buffer(struct intel_vgpu_workload *workload) gma_tail = workload->rb_start + workload->rb_tail; gma_top = workload->rb_start + guest_rb_size; - if (workload->rb_len > s->ring_scan_buffer_size[ring_id]) { + if (workload->rb_len > s->ring_scan_buffer_size[workload->engine->id]) { void *p; /* realloc the new ring buffer if needed */ - p = krealloc(s->ring_scan_buffer[ring_id], workload->rb_len, - GFP_KERNEL); + p = krealloc(s->ring_scan_buffer[workload->engine->id], + workload->rb_len, GFP_KERNEL); if (!p) { gvt_vgpu_err("fail to re-alloc ring scan buffer\n"); return -ENOMEM; } - s->ring_scan_buffer[ring_id] = p; - s->ring_scan_buffer_size[ring_id] = workload->rb_len; + s->ring_scan_buffer[workload->engine->id] = p; + s->ring_scan_buffer_size[workload->engine->id] = workload->rb_len; } - shadow_ring_buffer_va = s->ring_scan_buffer[ring_id]; + shadow_ring_buffer_va = s->ring_scan_buffer[workload->engine->id]; /* get shadow ring buffer va */ workload->shadow_ring_buffer_va = shadow_ring_buffer_va; @@ -2940,7 +2943,7 @@ static int shadow_indirect_ctx(struct intel_shadow_wa_ctx *wa_ctx) int ret = 0; void *map; - obj = i915_gem_object_create_shmem(workload->vgpu->gvt->dev_priv, + obj = i915_gem_object_create_shmem(workload->engine->i915, roundup(ctx_size + CACHELINE_BYTES, PAGE_SIZE)); if (IS_ERR(obj)) @@ -3029,30 +3032,14 @@ int intel_gvt_scan_and_shadow_wa_ctx(struct intel_shadow_wa_ctx *wa_ctx) return 0; } -static const struct cmd_info *find_cmd_entry_any_ring(struct intel_gvt *gvt, - unsigned int opcode, unsigned long rings) -{ - const struct cmd_info *info = NULL; - unsigned int ring; - - for_each_set_bit(ring, &rings, I915_NUM_ENGINES) { - info = find_cmd_entry(gvt, opcode, ring); - if (info) - break; - } - return info; -} - static int init_cmd_table(struct intel_gvt *gvt) { + unsigned int gen_type = intel_gvt_get_device_type(gvt); int i; - struct cmd_entry *e; - const struct cmd_info *info; - unsigned int gen_type; - - gen_type = intel_gvt_get_device_type(gvt); for (i = 0; i < ARRAY_SIZE(cmd_info); i++) { + struct cmd_entry *e; + if (!(cmd_info[i].devices & gen_type)) continue; @@ -3061,23 +3048,16 @@ static int init_cmd_table(struct intel_gvt *gvt) return -ENOMEM; e->info = &cmd_info[i]; - info = find_cmd_entry_any_ring(gvt, - e->info->opcode, e->info->rings); - if (info) { - gvt_err("%s %s duplicated\n", e->info->name, - info->name); - kfree(e); - return -EEXIST; - } if (cmd_info[i].opcode == OP_MI_NOOP) mi_noop_index = i; INIT_HLIST_NODE(&e->hlist); add_cmd_entry(gvt, e); gvt_dbg_cmd("add %-30s op %04x flag %x devs %02x rings %02x\n", - e->info->name, e->info->opcode, e->info->flag, - e->info->devices, e->info->rings); + e->info->name, e->info->opcode, e->info->flag, + e->info->devices, e->info->rings); } + return 0; } diff --git a/drivers/gpu/drm/i915/gvt/debugfs.c b/drivers/gpu/drm/i915/gvt/debugfs.c index 285f6011a537..ec47d4114554 100644 --- a/drivers/gpu/drm/i915/gvt/debugfs.c +++ b/drivers/gpu/drm/i915/gvt/debugfs.c @@ -58,12 +58,11 @@ static int mmio_offset_compare(void *priv, static inline int mmio_diff_handler(struct intel_gvt *gvt, u32 offset, void *data) { - struct drm_i915_private *i915 = gvt->dev_priv; struct mmio_diff_param *param = data; struct diff_mmio *node; u32 preg, vreg; - preg = intel_uncore_read_notrace(&i915->uncore, _MMIO(offset)); + preg = intel_uncore_read_notrace(gvt->gt->uncore, _MMIO(offset)); vreg = vgpu_vreg(param->vgpu, offset); if (preg != vreg) { @@ -98,10 +97,10 @@ static int vgpu_mmio_diff_show(struct seq_file *s, void *unused) mutex_lock(&gvt->lock); spin_lock_bh(&gvt->scheduler.mmio_context_lock); - mmio_hw_access_pre(gvt->dev_priv); + mmio_hw_access_pre(gvt->gt); /* Recognize all the diff mmios to list. */ intel_gvt_for_each_tracked_mmio(gvt, mmio_diff_handler, ¶m); - mmio_hw_access_post(gvt->dev_priv); + mmio_hw_access_post(gvt->gt); spin_unlock_bh(&gvt->scheduler.mmio_context_lock); mutex_unlock(&gvt->lock); @@ -128,6 +127,7 @@ static int vgpu_scan_nonprivbb_get(void *data, u64 *val) { struct intel_vgpu *vgpu = (struct intel_vgpu *)data; + *val = vgpu->scan_nonprivbb; return 0; } @@ -142,42 +142,7 @@ static int vgpu_scan_nonprivbb_set(void *data, u64 val) { struct intel_vgpu *vgpu = (struct intel_vgpu *)data; - struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv; - enum intel_engine_id id; - char buf[128], *s; - int len; - - val &= (1 << I915_NUM_ENGINES) - 1; - - if (vgpu->scan_nonprivbb == val) - return 0; - - if (!val) - goto done; - - len = sprintf(buf, - "gvt: vgpu %d turns on non-privileged batch buffers scanning on Engines:", - vgpu->id); - - s = buf + len; - - for (id = 0; id < I915_NUM_ENGINES; id++) { - struct intel_engine_cs *engine; - - engine = dev_priv->engine[id]; - if (engine && (val & (1 << id))) { - len = snprintf(s, 4, "%d, ", engine->id); - s += len; - } else - val &= ~(1 << id); - } - - if (val) - sprintf(s, "low performance expected."); - - pr_warn("%s\n", buf); -done: vgpu->scan_nonprivbb = val; return 0; } @@ -220,7 +185,7 @@ void intel_gvt_debugfs_remove_vgpu(struct intel_vgpu *vgpu) */ void intel_gvt_debugfs_init(struct intel_gvt *gvt) { - struct drm_minor *minor = gvt->dev_priv->drm.primary; + struct drm_minor *minor = gvt->gt->i915->drm.primary; gvt->debugfs_root = debugfs_create_dir("gvt", minor->debugfs_root); diff --git a/drivers/gpu/drm/i915/gvt/display.c b/drivers/gpu/drm/i915/gvt/display.c index e1c313da6c00..6e5c9885d9fe 100644 --- a/drivers/gpu/drm/i915/gvt/display.c +++ b/drivers/gpu/drm/i915/gvt/display.c @@ -57,7 +57,7 @@ static int get_edp_pipe(struct intel_vgpu *vgpu) static int edp_pipe_is_enabled(struct intel_vgpu *vgpu) { - struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv; + struct drm_i915_private *dev_priv = vgpu->gvt->gt->i915; if (!(vgpu_vreg_t(vgpu, PIPECONF(_PIPE_EDP)) & PIPECONF_ENABLE)) return 0; @@ -69,9 +69,10 @@ static int edp_pipe_is_enabled(struct intel_vgpu *vgpu) int pipe_is_enabled(struct intel_vgpu *vgpu, int pipe) { - struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv; + struct drm_i915_private *dev_priv = vgpu->gvt->gt->i915; - if (WARN_ON(pipe < PIPE_A || pipe >= I915_MAX_PIPES)) + if (drm_WARN_ON(&dev_priv->drm, + pipe < PIPE_A || pipe >= I915_MAX_PIPES)) return -EINVAL; if (vgpu_vreg_t(vgpu, PIPECONF(pipe)) & PIPECONF_ENABLE) @@ -168,7 +169,7 @@ static u8 dpcd_fix_data[DPCD_HEADER_SIZE] = { static void emulate_monitor_status_change(struct intel_vgpu *vgpu) { - struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv; + struct drm_i915_private *dev_priv = vgpu->gvt->gt->i915; int pipe; if (IS_BROXTON(dev_priv)) { @@ -319,9 +320,10 @@ static void clean_virtual_dp_monitor(struct intel_vgpu *vgpu, int port_num) static int setup_virtual_dp_monitor(struct intel_vgpu *vgpu, int port_num, int type, unsigned int resolution) { + struct drm_i915_private *i915 = vgpu->gvt->gt->i915; struct intel_vgpu_port *port = intel_vgpu_port(vgpu, port_num); - if (WARN_ON(resolution >= GVT_EDID_NUM)) + if (drm_WARN_ON(&i915->drm, resolution >= GVT_EDID_NUM)) return -EINVAL; port->edid = kzalloc(sizeof(*(port->edid)), GFP_KERNEL); @@ -389,7 +391,7 @@ void intel_gvt_check_vblank_emulation(struct intel_gvt *gvt) static void emulate_vblank_on_pipe(struct intel_vgpu *vgpu, int pipe) { - struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv; + struct drm_i915_private *dev_priv = vgpu->gvt->gt->i915; struct intel_vgpu_irq *irq = &vgpu->irq; int vblank_event[] = { [PIPE_A] = PIPE_A_VBLANK, @@ -421,7 +423,7 @@ static void emulate_vblank(struct intel_vgpu *vgpu) int pipe; mutex_lock(&vgpu->vgpu_lock); - for_each_pipe(vgpu->gvt->dev_priv, pipe) + for_each_pipe(vgpu->gvt->gt->i915, pipe) emulate_vblank_on_pipe(vgpu, pipe); mutex_unlock(&vgpu->vgpu_lock); } @@ -454,10 +456,11 @@ void intel_gvt_emulate_vblank(struct intel_gvt *gvt) */ void intel_vgpu_emulate_hotplug(struct intel_vgpu *vgpu, bool connected) { - struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv; + struct drm_i915_private *i915 = vgpu->gvt->gt->i915; /* TODO: add more platforms support */ - if (IS_SKYLAKE(dev_priv) || IS_KABYLAKE(dev_priv)) { + if (IS_SKYLAKE(i915) || IS_KABYLAKE(i915) || + IS_COFFEELAKE(i915)) { if (connected) { vgpu_vreg_t(vgpu, SFUSE_STRAP) |= SFUSE_STRAP_DDID_DETECTED; @@ -483,7 +486,7 @@ void intel_vgpu_emulate_hotplug(struct intel_vgpu *vgpu, bool connected) */ void intel_vgpu_clean_display(struct intel_vgpu *vgpu) { - struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv; + struct drm_i915_private *dev_priv = vgpu->gvt->gt->i915; if (IS_SKYLAKE(dev_priv) || IS_KABYLAKE(dev_priv) || IS_COFFEELAKE(dev_priv)) @@ -505,7 +508,7 @@ void intel_vgpu_clean_display(struct intel_vgpu *vgpu) */ int intel_vgpu_init_display(struct intel_vgpu *vgpu, u64 resolution) { - struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv; + struct drm_i915_private *dev_priv = vgpu->gvt->gt->i915; intel_vgpu_init_i2c_edid(vgpu); diff --git a/drivers/gpu/drm/i915/gvt/dmabuf.c b/drivers/gpu/drm/i915/gvt/dmabuf.c index 2477a1e5a166..37fc460414a8 100644 --- a/drivers/gpu/drm/i915/gvt/dmabuf.c +++ b/drivers/gpu/drm/i915/gvt/dmabuf.c @@ -67,11 +67,11 @@ static int vgpu_gem_get_pages( u32 page_num; fb_info = (struct intel_vgpu_fb_info *)obj->gvt_info; - if (WARN_ON(!fb_info)) + if (drm_WARN_ON(&dev_priv->drm, !fb_info)) return -ENODEV; vgpu = fb_info->obj->vgpu; - if (WARN_ON(!vgpu)) + if (drm_WARN_ON(&dev_priv->drm, !vgpu)) return -ENODEV; st = kmalloc(sizeof(*st), GFP_KERNEL); @@ -151,12 +151,12 @@ static void dmabuf_gem_object_free(struct kref *kref) dmabuf_obj = container_of(pos, struct intel_vgpu_dmabuf_obj, list); if (dmabuf_obj == obj) { + list_del(pos); intel_gvt_hypervisor_put_vfio_device(vgpu); idr_remove(&vgpu->object_idr, dmabuf_obj->dmabuf_id); kfree(dmabuf_obj->info); kfree(dmabuf_obj); - list_del(pos); break; } } @@ -417,7 +417,7 @@ static void update_fb_info(struct vfio_device_gfx_plane_info *gvt_dmabuf, int intel_vgpu_query_plane(struct intel_vgpu *vgpu, void *args) { - struct drm_device *dev = &vgpu->gvt->dev_priv->drm; + struct drm_device *dev = &vgpu->gvt->gt->i915->drm; struct vfio_device_gfx_plane_info *gfx_plane_info = args; struct intel_vgpu_dmabuf_obj *dmabuf_obj; struct intel_vgpu_fb_info fb_info; @@ -523,7 +523,7 @@ out: /* To associate an exposed dmabuf with the dmabuf_obj */ int intel_vgpu_get_dmabuf(struct intel_vgpu *vgpu, unsigned int dmabuf_id) { - struct drm_device *dev = &vgpu->gvt->dev_priv->drm; + struct drm_device *dev = &vgpu->gvt->gt->i915->drm; struct intel_vgpu_dmabuf_obj *dmabuf_obj; struct drm_i915_gem_object *obj; struct dma_buf *dmabuf; diff --git a/drivers/gpu/drm/i915/gvt/edid.c b/drivers/gpu/drm/i915/gvt/edid.c index 1fe6124918f1..190651df5db1 100644 --- a/drivers/gpu/drm/i915/gvt/edid.c +++ b/drivers/gpu/drm/i915/gvt/edid.c @@ -135,7 +135,7 @@ static void reset_gmbus_controller(struct intel_vgpu *vgpu) static int gmbus0_mmio_write(struct intel_vgpu *vgpu, unsigned int offset, void *p_data, unsigned int bytes) { - struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv; + struct drm_i915_private *i915 = vgpu->gvt->gt->i915; int port, pin_select; memcpy(&vgpu_vreg(vgpu, offset), p_data, bytes); @@ -147,13 +147,13 @@ static int gmbus0_mmio_write(struct intel_vgpu *vgpu, if (pin_select == 0) return 0; - if (IS_BROXTON(dev_priv)) + if (IS_BROXTON(i915)) port = bxt_get_port_from_gmbus0(pin_select); - else if (IS_COFFEELAKE(dev_priv)) + else if (IS_COFFEELAKE(i915)) port = cnp_get_port_from_gmbus0(pin_select); else port = get_port_from_gmbus0(pin_select); - if (WARN_ON(port < 0)) + if (drm_WARN_ON(&i915->drm, port < 0)) return 0; vgpu->display.i2c_edid.state = I2C_GMBUS; @@ -276,7 +276,9 @@ static int gmbus1_mmio_write(struct intel_vgpu *vgpu, unsigned int offset, static int gmbus3_mmio_write(struct intel_vgpu *vgpu, unsigned int offset, void *p_data, unsigned int bytes) { - WARN_ON(1); + struct drm_i915_private *i915 = vgpu->gvt->gt->i915; + + drm_WARN_ON(&i915->drm, 1); return 0; } @@ -371,7 +373,9 @@ static int gmbus2_mmio_write(struct intel_vgpu *vgpu, unsigned int offset, int intel_gvt_i2c_handle_gmbus_read(struct intel_vgpu *vgpu, unsigned int offset, void *p_data, unsigned int bytes) { - if (WARN_ON(bytes > 8 && (offset & (bytes - 1)))) + struct drm_i915_private *i915 = vgpu->gvt->gt->i915; + + if (drm_WARN_ON(&i915->drm, bytes > 8 && (offset & (bytes - 1)))) return -EINVAL; if (offset == i915_mmio_reg_offset(PCH_GMBUS2)) @@ -399,7 +403,9 @@ int intel_gvt_i2c_handle_gmbus_read(struct intel_vgpu *vgpu, int intel_gvt_i2c_handle_gmbus_write(struct intel_vgpu *vgpu, unsigned int offset, void *p_data, unsigned int bytes) { - if (WARN_ON(bytes > 8 && (offset & (bytes - 1)))) + struct drm_i915_private *i915 = vgpu->gvt->gt->i915; + + if (drm_WARN_ON(&i915->drm, bytes > 8 && (offset & (bytes - 1)))) return -EINVAL; if (offset == i915_mmio_reg_offset(PCH_GMBUS0)) @@ -473,6 +479,7 @@ void intel_gvt_i2c_handle_aux_ch_write(struct intel_vgpu *vgpu, unsigned int offset, void *p_data) { + struct drm_i915_private *i915 = vgpu->gvt->gt->i915; struct intel_vgpu_i2c_edid *i2c_edid = &vgpu->display.i2c_edid; int msg_length, ret_msg_size; int msg, addr, ctrl, op; @@ -532,9 +539,9 @@ void intel_gvt_i2c_handle_aux_ch_write(struct intel_vgpu *vgpu, * support the gfx driver to do EDID access. */ } else { - if (WARN_ON((op & 0x1) != GVT_AUX_I2C_READ)) + if (drm_WARN_ON(&i915->drm, (op & 0x1) != GVT_AUX_I2C_READ)) return; - if (WARN_ON(msg_length != 4)) + if (drm_WARN_ON(&i915->drm, msg_length != 4)) return; if (i2c_edid->edid_available && i2c_edid->slave_selected) { unsigned char val = edid_get_byte(vgpu); diff --git a/drivers/gpu/drm/i915/gvt/execlist.c b/drivers/gpu/drm/i915/gvt/execlist.c index d6e7a1189bad..dd25c3024370 100644 --- a/drivers/gpu/drm/i915/gvt/execlist.c +++ b/drivers/gpu/drm/i915/gvt/execlist.c @@ -39,8 +39,7 @@ #define _EL_OFFSET_STATUS_BUF 0x370 #define _EL_OFFSET_STATUS_PTR 0x3A0 -#define execlist_ring_mmio(gvt, ring_id, offset) \ - (gvt->dev_priv->engine[ring_id]->mmio_base + (offset)) +#define execlist_ring_mmio(e, offset) ((e)->mmio_base + (offset)) #define valid_context(ctx) ((ctx)->valid) #define same_context(a, b) (((a)->context_id == (b)->context_id) && \ @@ -54,12 +53,12 @@ static int context_switch_events[] = { [VECS0] = VECS_AS_CONTEXT_SWITCH, }; -static int ring_id_to_context_switch_event(unsigned int ring_id) +static int to_context_switch_event(const struct intel_engine_cs *engine) { - if (WARN_ON(ring_id >= ARRAY_SIZE(context_switch_events))) + if (WARN_ON(engine->id >= ARRAY_SIZE(context_switch_events))) return -EINVAL; - return context_switch_events[ring_id]; + return context_switch_events[engine->id]; } static void switch_virtual_execlist_slot(struct intel_vgpu_execlist *execlist) @@ -93,9 +92,8 @@ static void emulate_execlist_status(struct intel_vgpu_execlist *execlist) struct execlist_ctx_descriptor_format *desc = execlist->running_context; struct intel_vgpu *vgpu = execlist->vgpu; struct execlist_status_format status; - int ring_id = execlist->ring_id; - u32 status_reg = execlist_ring_mmio(vgpu->gvt, - ring_id, _EL_OFFSET_STATUS); + u32 status_reg = + execlist_ring_mmio(execlist->engine, _EL_OFFSET_STATUS); status.ldw = vgpu_vreg(vgpu, status_reg); status.udw = vgpu_vreg(vgpu, status_reg + 4); @@ -124,21 +122,19 @@ static void emulate_execlist_status(struct intel_vgpu_execlist *execlist) } static void emulate_csb_update(struct intel_vgpu_execlist *execlist, - struct execlist_context_status_format *status, - bool trigger_interrupt_later) + struct execlist_context_status_format *status, + bool trigger_interrupt_later) { struct intel_vgpu *vgpu = execlist->vgpu; - int ring_id = execlist->ring_id; struct execlist_context_status_pointer_format ctx_status_ptr; u32 write_pointer; u32 ctx_status_ptr_reg, ctx_status_buf_reg, offset; unsigned long hwsp_gpa; - struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv; - ctx_status_ptr_reg = execlist_ring_mmio(vgpu->gvt, ring_id, - _EL_OFFSET_STATUS_PTR); - ctx_status_buf_reg = execlist_ring_mmio(vgpu->gvt, ring_id, - _EL_OFFSET_STATUS_BUF); + ctx_status_ptr_reg = + execlist_ring_mmio(execlist->engine, _EL_OFFSET_STATUS_PTR); + ctx_status_buf_reg = + execlist_ring_mmio(execlist->engine, _EL_OFFSET_STATUS_BUF); ctx_status_ptr.dw = vgpu_vreg(vgpu, ctx_status_ptr_reg); @@ -161,26 +157,24 @@ static void emulate_csb_update(struct intel_vgpu_execlist *execlist, /* Update the CSB and CSB write pointer in HWSP */ hwsp_gpa = intel_vgpu_gma_to_gpa(vgpu->gtt.ggtt_mm, - vgpu->hws_pga[ring_id]); + vgpu->hws_pga[execlist->engine->id]); if (hwsp_gpa != INTEL_GVT_INVALID_ADDR) { intel_gvt_hypervisor_write_gpa(vgpu, - hwsp_gpa + I915_HWS_CSB_BUF0_INDEX * 4 + - write_pointer * 8, - status, 8); + hwsp_gpa + I915_HWS_CSB_BUF0_INDEX * 4 + write_pointer * 8, + status, 8); intel_gvt_hypervisor_write_gpa(vgpu, - hwsp_gpa + - intel_hws_csb_write_index(dev_priv) * 4, - &write_pointer, 4); + hwsp_gpa + intel_hws_csb_write_index(execlist->engine->i915) * 4, + &write_pointer, 4); } gvt_dbg_el("vgpu%d: w pointer %u reg %x csb l %x csb h %x\n", - vgpu->id, write_pointer, offset, status->ldw, status->udw); + vgpu->id, write_pointer, offset, status->ldw, status->udw); if (trigger_interrupt_later) return; intel_vgpu_trigger_virtual_event(vgpu, - ring_id_to_context_switch_event(execlist->ring_id)); + to_context_switch_event(execlist->engine)); } static int emulate_execlist_ctx_schedule_out( @@ -261,9 +255,8 @@ static struct intel_vgpu_execlist_slot *get_next_execlist_slot( struct intel_vgpu_execlist *execlist) { struct intel_vgpu *vgpu = execlist->vgpu; - int ring_id = execlist->ring_id; - u32 status_reg = execlist_ring_mmio(vgpu->gvt, ring_id, - _EL_OFFSET_STATUS); + u32 status_reg = + execlist_ring_mmio(execlist->engine, _EL_OFFSET_STATUS); struct execlist_status_format status; status.ldw = vgpu_vreg(vgpu, status_reg); @@ -379,7 +372,6 @@ static int prepare_execlist_workload(struct intel_vgpu_workload *workload) struct intel_vgpu *vgpu = workload->vgpu; struct intel_vgpu_submission *s = &vgpu->submission; struct execlist_ctx_descriptor_format ctx[2]; - int ring_id = workload->ring_id; int ret; if (!workload->emulate_schedule_in) @@ -388,7 +380,8 @@ static int prepare_execlist_workload(struct intel_vgpu_workload *workload) ctx[0] = *get_desc_from_elsp_dwords(&workload->elsp_dwords, 0); ctx[1] = *get_desc_from_elsp_dwords(&workload->elsp_dwords, 1); - ret = emulate_execlist_schedule_in(&s->execlist[ring_id], ctx); + ret = emulate_execlist_schedule_in(&s->execlist[workload->engine->id], + ctx); if (ret) { gvt_vgpu_err("fail to emulate execlist schedule in\n"); return ret; @@ -399,21 +392,21 @@ static int prepare_execlist_workload(struct intel_vgpu_workload *workload) static int complete_execlist_workload(struct intel_vgpu_workload *workload) { struct intel_vgpu *vgpu = workload->vgpu; - int ring_id = workload->ring_id; struct intel_vgpu_submission *s = &vgpu->submission; - struct intel_vgpu_execlist *execlist = &s->execlist[ring_id]; + struct intel_vgpu_execlist *execlist = + &s->execlist[workload->engine->id]; struct intel_vgpu_workload *next_workload; - struct list_head *next = workload_q_head(vgpu, ring_id)->next; + struct list_head *next = workload_q_head(vgpu, workload->engine)->next; bool lite_restore = false; int ret = 0; - gvt_dbg_el("complete workload %p status %d\n", workload, - workload->status); + gvt_dbg_el("complete workload %p status %d\n", + workload, workload->status); - if (workload->status || (vgpu->resetting_eng & BIT(ring_id))) + if (workload->status || vgpu->resetting_eng & workload->engine->mask) goto out; - if (!list_empty(workload_q_head(vgpu, ring_id))) { + if (!list_empty(workload_q_head(vgpu, workload->engine))) { struct execlist_ctx_descriptor_format *this_desc, *next_desc; next_workload = container_of(next, @@ -436,14 +429,15 @@ out: return ret; } -static int submit_context(struct intel_vgpu *vgpu, int ring_id, - struct execlist_ctx_descriptor_format *desc, - bool emulate_schedule_in) +static int submit_context(struct intel_vgpu *vgpu, + const struct intel_engine_cs *engine, + struct execlist_ctx_descriptor_format *desc, + bool emulate_schedule_in) { struct intel_vgpu_submission *s = &vgpu->submission; struct intel_vgpu_workload *workload = NULL; - workload = intel_vgpu_create_workload(vgpu, ring_id, desc); + workload = intel_vgpu_create_workload(vgpu, engine, desc); if (IS_ERR(workload)) return PTR_ERR(workload); @@ -452,19 +446,20 @@ static int submit_context(struct intel_vgpu *vgpu, int ring_id, workload->emulate_schedule_in = emulate_schedule_in; if (emulate_schedule_in) - workload->elsp_dwords = s->execlist[ring_id].elsp_dwords; + workload->elsp_dwords = s->execlist[engine->id].elsp_dwords; gvt_dbg_el("workload %p emulate schedule_in %d\n", workload, - emulate_schedule_in); + emulate_schedule_in); intel_vgpu_queue_workload(workload); return 0; } -int intel_vgpu_submit_execlist(struct intel_vgpu *vgpu, int ring_id) +int intel_vgpu_submit_execlist(struct intel_vgpu *vgpu, + const struct intel_engine_cs *engine) { struct intel_vgpu_submission *s = &vgpu->submission; - struct intel_vgpu_execlist *execlist = &s->execlist[ring_id]; + struct intel_vgpu_execlist *execlist = &s->execlist[engine->id]; struct execlist_ctx_descriptor_format *desc[2]; int i, ret; @@ -489,7 +484,7 @@ int intel_vgpu_submit_execlist(struct intel_vgpu *vgpu, int ring_id) for (i = 0; i < ARRAY_SIZE(desc); i++) { if (!desc[i]->valid) continue; - ret = submit_context(vgpu, ring_id, desc[i], i == 0); + ret = submit_context(vgpu, engine, desc[i], i == 0); if (ret) { gvt_vgpu_err("failed to submit desc %d\n", i); return ret; @@ -504,22 +499,22 @@ inv_desc: return -EINVAL; } -static void init_vgpu_execlist(struct intel_vgpu *vgpu, int ring_id) +static void init_vgpu_execlist(struct intel_vgpu *vgpu, + const struct intel_engine_cs *engine) { struct intel_vgpu_submission *s = &vgpu->submission; - struct intel_vgpu_execlist *execlist = &s->execlist[ring_id]; + struct intel_vgpu_execlist *execlist = &s->execlist[engine->id]; struct execlist_context_status_pointer_format ctx_status_ptr; u32 ctx_status_ptr_reg; memset(execlist, 0, sizeof(*execlist)); execlist->vgpu = vgpu; - execlist->ring_id = ring_id; + execlist->engine = engine; execlist->slot[0].index = 0; execlist->slot[1].index = 1; - ctx_status_ptr_reg = execlist_ring_mmio(vgpu->gvt, ring_id, - _EL_OFFSET_STATUS_PTR); + ctx_status_ptr_reg = execlist_ring_mmio(engine, _EL_OFFSET_STATUS_PTR); ctx_status_ptr.dw = vgpu_vreg(vgpu, ctx_status_ptr_reg); ctx_status_ptr.read_ptr = 0; ctx_status_ptr.write_ptr = 0x7; @@ -529,7 +524,7 @@ static void init_vgpu_execlist(struct intel_vgpu *vgpu, int ring_id) static void clean_execlist(struct intel_vgpu *vgpu, intel_engine_mask_t engine_mask) { - struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv; + struct drm_i915_private *dev_priv = vgpu->gvt->gt->i915; struct intel_engine_cs *engine; struct intel_vgpu_submission *s = &vgpu->submission; intel_engine_mask_t tmp; @@ -544,12 +539,12 @@ static void clean_execlist(struct intel_vgpu *vgpu, static void reset_execlist(struct intel_vgpu *vgpu, intel_engine_mask_t engine_mask) { - struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv; + struct drm_i915_private *dev_priv = vgpu->gvt->gt->i915; struct intel_engine_cs *engine; intel_engine_mask_t tmp; for_each_engine_masked(engine, &dev_priv->gt, engine_mask, tmp) - init_vgpu_execlist(vgpu, engine->id); + init_vgpu_execlist(vgpu, engine); } static int init_execlist(struct intel_vgpu *vgpu, diff --git a/drivers/gpu/drm/i915/gvt/execlist.h b/drivers/gpu/drm/i915/gvt/execlist.h index 5c0c1fd30c83..d62cd14605a3 100644 --- a/drivers/gpu/drm/i915/gvt/execlist.h +++ b/drivers/gpu/drm/i915/gvt/execlist.h @@ -170,16 +170,17 @@ struct intel_vgpu_execlist { struct intel_vgpu_execlist_slot *running_slot; struct intel_vgpu_execlist_slot *pending_slot; struct execlist_ctx_descriptor_format *running_context; - int ring_id; struct intel_vgpu *vgpu; struct intel_vgpu_elsp_dwords elsp_dwords; + const struct intel_engine_cs *engine; }; void intel_vgpu_clean_execlist(struct intel_vgpu *vgpu); int intel_vgpu_init_execlist(struct intel_vgpu *vgpu); -int intel_vgpu_submit_execlist(struct intel_vgpu *vgpu, int ring_id); +int intel_vgpu_submit_execlist(struct intel_vgpu *vgpu, + const struct intel_engine_cs *engine); void intel_vgpu_reset_execlist(struct intel_vgpu *vgpu, intel_engine_mask_t engine_mask); diff --git a/drivers/gpu/drm/i915/gvt/fb_decoder.c b/drivers/gpu/drm/i915/gvt/fb_decoder.c index 8bb292b01271..0889ad8291b0 100644 --- a/drivers/gpu/drm/i915/gvt/fb_decoder.c +++ b/drivers/gpu/drm/i915/gvt/fb_decoder.c @@ -146,7 +146,7 @@ static int skl_format_to_drm(int format, bool rgb_order, bool alpha, static u32 intel_vgpu_get_stride(struct intel_vgpu *vgpu, int pipe, u32 tiled, int stride_mask, int bpp) { - struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv; + struct drm_i915_private *dev_priv = vgpu->gvt->gt->i915; u32 stride_reg = vgpu_vreg_t(vgpu, DSPSTRIDE(pipe)) & stride_mask; u32 stride = stride_reg; @@ -202,8 +202,8 @@ static int get_active_pipe(struct intel_vgpu *vgpu) int intel_vgpu_decode_primary_plane(struct intel_vgpu *vgpu, struct intel_vgpu_primary_plane_format *plane) { + struct drm_i915_private *dev_priv = vgpu->gvt->gt->i915; u32 val, fmt; - struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv; int pipe; pipe = get_active_pipe(vgpu); @@ -332,9 +332,9 @@ static int cursor_mode_to_drm(int mode) int intel_vgpu_decode_cursor_plane(struct intel_vgpu *vgpu, struct intel_vgpu_cursor_plane_format *plane) { + struct drm_i915_private *dev_priv = vgpu->gvt->gt->i915; u32 val, mode, index; u32 alpha_plane, alpha_force; - struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv; int pipe; pipe = get_active_pipe(vgpu); diff --git a/drivers/gpu/drm/i915/gvt/firmware.c b/drivers/gpu/drm/i915/gvt/firmware.c index b0c1fda32977..990a181094e3 100644 --- a/drivers/gpu/drm/i915/gvt/firmware.c +++ b/drivers/gpu/drm/i915/gvt/firmware.c @@ -68,9 +68,7 @@ static struct bin_attribute firmware_attr = { static int mmio_snapshot_handler(struct intel_gvt *gvt, u32 offset, void *data) { - struct drm_i915_private *i915 = gvt->dev_priv; - - *(u32 *)(data + offset) = intel_uncore_read_notrace(&i915->uncore, + *(u32 *)(data + offset) = intel_uncore_read_notrace(gvt->gt->uncore, _MMIO(offset)); return 0; } @@ -78,7 +76,7 @@ static int mmio_snapshot_handler(struct intel_gvt *gvt, u32 offset, void *data) static int expose_firmware_sysfs(struct intel_gvt *gvt) { struct intel_gvt_device_info *info = &gvt->device_info; - struct pci_dev *pdev = gvt->dev_priv->drm.pdev; + struct pci_dev *pdev = gvt->gt->i915->drm.pdev; struct gvt_firmware_header *h; void *firmware; void *p; @@ -129,7 +127,7 @@ static int expose_firmware_sysfs(struct intel_gvt *gvt) static void clean_firmware_sysfs(struct intel_gvt *gvt) { - struct pci_dev *pdev = gvt->dev_priv->drm.pdev; + struct pci_dev *pdev = gvt->gt->i915->drm.pdev; device_remove_bin_file(&pdev->dev, &firmware_attr); vfree(firmware_attr.private); @@ -153,8 +151,7 @@ static int verify_firmware(struct intel_gvt *gvt, const struct firmware *fw) { struct intel_gvt_device_info *info = &gvt->device_info; - struct drm_i915_private *dev_priv = gvt->dev_priv; - struct pci_dev *pdev = dev_priv->drm.pdev; + struct pci_dev *pdev = gvt->gt->i915->drm.pdev; struct gvt_firmware_header *h; unsigned long id, crc32_start; const void *mem; @@ -208,8 +205,7 @@ invalid_firmware: int intel_gvt_load_firmware(struct intel_gvt *gvt) { struct intel_gvt_device_info *info = &gvt->device_info; - struct drm_i915_private *dev_priv = gvt->dev_priv; - struct pci_dev *pdev = dev_priv->drm.pdev; + struct pci_dev *pdev = gvt->gt->i915->drm.pdev; struct intel_gvt_firmware *firmware = &gvt->firmware; struct gvt_firmware_header *h; const struct firmware *fw; @@ -244,7 +240,7 @@ int intel_gvt_load_firmware(struct intel_gvt *gvt) gvt_dbg_core("request hw state firmware %s...\n", path); - ret = request_firmware(&fw, path, &dev_priv->drm.pdev->dev); + ret = request_firmware(&fw, path, &gvt->gt->i915->drm.pdev->dev); kfree(path); if (ret) diff --git a/drivers/gpu/drm/i915/gvt/gtt.c b/drivers/gpu/drm/i915/gvt/gtt.c index 4a4828074cb7..2a4b23f8aa74 100644 --- a/drivers/gpu/drm/i915/gvt/gtt.c +++ b/drivers/gpu/drm/i915/gvt/gtt.c @@ -71,8 +71,10 @@ bool intel_gvt_ggtt_validate_range(struct intel_vgpu *vgpu, u64 addr, u32 size) /* translate a guest gmadr to host gmadr */ int intel_gvt_ggtt_gmadr_g2h(struct intel_vgpu *vgpu, u64 g_addr, u64 *h_addr) { - if (WARN(!vgpu_gmadr_is_valid(vgpu, g_addr), - "invalid guest gmadr %llx\n", g_addr)) + struct drm_i915_private *i915 = vgpu->gvt->gt->i915; + + if (drm_WARN(&i915->drm, !vgpu_gmadr_is_valid(vgpu, g_addr), + "invalid guest gmadr %llx\n", g_addr)) return -EACCES; if (vgpu_gmadr_is_aperture(vgpu, g_addr)) @@ -87,8 +89,10 @@ int intel_gvt_ggtt_gmadr_g2h(struct intel_vgpu *vgpu, u64 g_addr, u64 *h_addr) /* translate a host gmadr to guest gmadr */ int intel_gvt_ggtt_gmadr_h2g(struct intel_vgpu *vgpu, u64 h_addr, u64 *g_addr) { - if (WARN(!gvt_gmadr_is_valid(vgpu->gvt, h_addr), - "invalid host gmadr %llx\n", h_addr)) + struct drm_i915_private *i915 = vgpu->gvt->gt->i915; + + if (drm_WARN(&i915->drm, !gvt_gmadr_is_valid(vgpu->gvt, h_addr), + "invalid host gmadr %llx\n", h_addr)) return -EACCES; if (gvt_gmadr_is_aperture(vgpu->gvt, h_addr)) @@ -275,24 +279,23 @@ static inline int get_pse_type(int type) return gtt_type_table[type].pse_entry_type; } -static u64 read_pte64(struct drm_i915_private *dev_priv, unsigned long index) +static u64 read_pte64(struct i915_ggtt *ggtt, unsigned long index) { - void __iomem *addr = (gen8_pte_t __iomem *)dev_priv->ggtt.gsm + index; + void __iomem *addr = (gen8_pte_t __iomem *)ggtt->gsm + index; return readq(addr); } -static void ggtt_invalidate(struct drm_i915_private *dev_priv) +static void ggtt_invalidate(struct intel_gt *gt) { - mmio_hw_access_pre(dev_priv); - I915_WRITE(GFX_FLSH_CNTL_GEN6, GFX_FLSH_CNTL_EN); - mmio_hw_access_post(dev_priv); + mmio_hw_access_pre(gt); + intel_uncore_write(gt->uncore, GFX_FLSH_CNTL_GEN6, GFX_FLSH_CNTL_EN); + mmio_hw_access_post(gt); } -static void write_pte64(struct drm_i915_private *dev_priv, - unsigned long index, u64 pte) +static void write_pte64(struct i915_ggtt *ggtt, unsigned long index, u64 pte) { - void __iomem *addr = (gen8_pte_t __iomem *)dev_priv->ggtt.gsm + index; + void __iomem *addr = (gen8_pte_t __iomem *)ggtt->gsm + index; writeq(pte, addr); } @@ -315,7 +318,7 @@ static inline int gtt_get_entry64(void *pt, if (WARN_ON(ret)) return ret; } else if (!pt) { - e->val64 = read_pte64(vgpu->gvt->dev_priv, index); + e->val64 = read_pte64(vgpu->gvt->gt->ggtt, index); } else { e->val64 = *((u64 *)pt + index); } @@ -340,7 +343,7 @@ static inline int gtt_set_entry64(void *pt, if (WARN_ON(ret)) return ret; } else if (!pt) { - write_pte64(vgpu->gvt->dev_priv, index, e->val64); + write_pte64(vgpu->gvt->gt->ggtt, index, e->val64); } else { *((u64 *)pt + index) = e->val64; } @@ -734,7 +737,7 @@ static int detach_oos_page(struct intel_vgpu *vgpu, static void ppgtt_free_spt(struct intel_vgpu_ppgtt_spt *spt) { - struct device *kdev = &spt->vgpu->gvt->dev_priv->drm.pdev->dev; + struct device *kdev = &spt->vgpu->gvt->gt->i915->drm.pdev->dev; trace_spt_free(spt->vgpu->id, spt, spt->guest_page.type); @@ -819,7 +822,7 @@ static int reclaim_one_ppgtt_mm(struct intel_gvt *gvt); static struct intel_vgpu_ppgtt_spt *ppgtt_alloc_spt( struct intel_vgpu *vgpu, enum intel_gvt_gtt_type type) { - struct device *kdev = &vgpu->gvt->dev_priv->drm.pdev->dev; + struct device *kdev = &vgpu->gvt->gt->i915->drm.pdev->dev; struct intel_vgpu_ppgtt_spt *spt = NULL; dma_addr_t daddr; int ret; @@ -940,6 +943,7 @@ static int ppgtt_invalidate_spt(struct intel_vgpu_ppgtt_spt *spt); static int ppgtt_invalidate_spt_by_shadow_entry(struct intel_vgpu *vgpu, struct intel_gvt_gtt_entry *e) { + struct drm_i915_private *i915 = vgpu->gvt->gt->i915; struct intel_gvt_gtt_pte_ops *ops = vgpu->gvt->gtt.pte_ops; struct intel_vgpu_ppgtt_spt *s; enum intel_gvt_gtt_type cur_pt_type; @@ -952,7 +956,9 @@ static int ppgtt_invalidate_spt_by_shadow_entry(struct intel_vgpu *vgpu, if (!gtt_type_is_pt(cur_pt_type) || !gtt_type_is_pt(cur_pt_type + 1)) { - WARN(1, "Invalid page table type, cur_pt_type is: %d\n", cur_pt_type); + drm_WARN(&i915->drm, 1, + "Invalid page table type, cur_pt_type is: %d\n", + cur_pt_type); return -EINVAL; } @@ -1044,7 +1050,7 @@ fail: static bool vgpu_ips_enabled(struct intel_vgpu *vgpu) { - struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv; + struct drm_i915_private *dev_priv = vgpu->gvt->gt->i915; if (INTEL_GEN(dev_priv) == 9 || INTEL_GEN(dev_priv) == 10) { u32 ips = vgpu_vreg_t(vgpu, GEN8_GAMW_ECO_DEV_RW_IA) & @@ -1153,7 +1159,7 @@ static int is_2MB_gtt_possible(struct intel_vgpu *vgpu, struct intel_gvt_gtt_pte_ops *ops = vgpu->gvt->gtt.pte_ops; unsigned long pfn; - if (!HAS_PAGE_SIZES(vgpu->gvt->dev_priv, I915_GTT_PAGE_SIZE_2M)) + if (!HAS_PAGE_SIZES(vgpu->gvt->gt->i915, I915_GTT_PAGE_SIZE_2M)) return 0; pfn = intel_gvt_hypervisor_gfn_to_mfn(vgpu, ops->get_pfn(entry)); @@ -2314,7 +2320,7 @@ out: ggtt_invalidate_pte(vgpu, &e); ggtt_set_host_entry(ggtt_mm, &m, g_gtt_index); - ggtt_invalidate(gvt->dev_priv); + ggtt_invalidate(gvt->gt); return 0; } @@ -2347,16 +2353,18 @@ int intel_vgpu_emulate_ggtt_mmio_write(struct intel_vgpu *vgpu, static int alloc_scratch_pages(struct intel_vgpu *vgpu, enum intel_gvt_gtt_type type) { + struct drm_i915_private *i915 = vgpu->gvt->gt->i915; struct intel_vgpu_gtt *gtt = &vgpu->gtt; struct intel_gvt_gtt_pte_ops *ops = vgpu->gvt->gtt.pte_ops; int page_entry_num = I915_GTT_PAGE_SIZE >> vgpu->gvt->device_info.gtt_entry_size_shift; void *scratch_pt; int i; - struct device *dev = &vgpu->gvt->dev_priv->drm.pdev->dev; + struct device *dev = &vgpu->gvt->gt->i915->drm.pdev->dev; dma_addr_t daddr; - if (WARN_ON(type < GTT_TYPE_PPGTT_PTE_PT || type >= GTT_TYPE_MAX)) + if (drm_WARN_ON(&i915->drm, + type < GTT_TYPE_PPGTT_PTE_PT || type >= GTT_TYPE_MAX)) return -EINVAL; scratch_pt = (void *)get_zeroed_page(GFP_KERNEL); @@ -2410,7 +2418,7 @@ static int alloc_scratch_pages(struct intel_vgpu *vgpu, static int release_scratch_page_tree(struct intel_vgpu *vgpu) { int i; - struct device *dev = &vgpu->gvt->dev_priv->drm.pdev->dev; + struct device *dev = &vgpu->gvt->gt->i915->drm.pdev->dev; dma_addr_t daddr; for (i = GTT_TYPE_PPGTT_PTE_PT; i < GTT_TYPE_MAX; i++) { @@ -2682,7 +2690,7 @@ int intel_gvt_init_gtt(struct intel_gvt *gvt) { int ret; void *page; - struct device *dev = &gvt->dev_priv->drm.pdev->dev; + struct device *dev = &gvt->gt->i915->drm.pdev->dev; dma_addr_t daddr; gvt_dbg_core("init gtt\n"); @@ -2731,7 +2739,7 @@ int intel_gvt_init_gtt(struct intel_gvt *gvt) */ void intel_gvt_clean_gtt(struct intel_gvt *gvt) { - struct device *dev = &gvt->dev_priv->drm.pdev->dev; + struct device *dev = &gvt->gt->i915->drm.pdev->dev; dma_addr_t daddr = (dma_addr_t)(gvt->gtt.scratch_mfn << I915_GTT_PAGE_SHIFT); @@ -2779,7 +2787,6 @@ void intel_vgpu_invalidate_ppgtt(struct intel_vgpu *vgpu) void intel_vgpu_reset_ggtt(struct intel_vgpu *vgpu, bool invalidate_old) { struct intel_gvt *gvt = vgpu->gvt; - struct drm_i915_private *dev_priv = gvt->dev_priv; struct intel_gvt_gtt_pte_ops *pte_ops = vgpu->gvt->gtt.pte_ops; struct intel_gvt_gtt_entry entry = {.type = GTT_TYPE_GGTT_PTE}; struct intel_gvt_gtt_entry old_entry; @@ -2809,7 +2816,7 @@ void intel_vgpu_reset_ggtt(struct intel_vgpu *vgpu, bool invalidate_old) ggtt_set_host_entry(vgpu->gtt.ggtt_mm, &entry, index++); } - ggtt_invalidate(dev_priv); + ggtt_invalidate(gvt->gt); } /** diff --git a/drivers/gpu/drm/i915/gvt/gvt.c b/drivers/gpu/drm/i915/gvt/gvt.c index 8f37eefa0a02..9e1787867894 100644 --- a/drivers/gpu/drm/i915/gvt/gvt.c +++ b/drivers/gpu/drm/i915/gvt/gvt.c @@ -35,6 +35,7 @@ #include <linux/kthread.h> #include "i915_drv.h" +#include "intel_gvt.h" #include "gvt.h" #include <linux/vfio.h> #include <linux/mdev.h> @@ -49,15 +50,15 @@ static const char * const supported_hypervisors[] = { static struct intel_vgpu_type *intel_gvt_find_vgpu_type(struct intel_gvt *gvt, const char *name) { + const char *driver_name = + dev_driver_string(&gvt->gt->i915->drm.pdev->dev); int i; - struct intel_vgpu_type *t; - const char *driver_name = dev_driver_string( - &gvt->dev_priv->drm.pdev->dev); + name += strlen(driver_name) + 1; for (i = 0; i < gvt->num_types; i++) { - t = &gvt->types[i]; - if (!strncmp(t->name, name + strlen(driver_name) + 1, - sizeof(t->name))) + struct intel_vgpu_type *t = &gvt->types[i]; + + if (!strncmp(t->name, name, sizeof(t->name))) return t; } @@ -120,10 +121,8 @@ static struct attribute_group *gvt_vgpu_type_groups[] = { [0 ... NR_MAX_INTEL_VGPU_TYPES - 1] = NULL, }; -static bool intel_get_gvt_attrs(struct attribute ***type_attrs, - struct attribute_group ***intel_vgpu_type_groups) +static bool intel_get_gvt_attrs(struct attribute_group ***intel_vgpu_type_groups) { - *type_attrs = gvt_type_attrs; *intel_vgpu_type_groups = gvt_vgpu_type_groups; return true; } @@ -191,7 +190,7 @@ static const struct intel_gvt_ops intel_gvt_ops = { static void init_device_info(struct intel_gvt *gvt) { struct intel_gvt_device_info *info = &gvt->device_info; - struct pci_dev *pdev = gvt->dev_priv->drm.pdev; + struct pci_dev *pdev = gvt->gt->i915->drm.pdev; info->max_support_vgpus = 8; info->cfg_space_size = PCI_CFG_SPACE_EXP_SIZE; @@ -257,17 +256,17 @@ static int init_service_thread(struct intel_gvt *gvt) /** * intel_gvt_clean_device - clean a GVT device - * @dev_priv: i915 private + * @i915: i915 private * * This function is called at the driver unloading stage, to free the * resources owned by a GVT device. * */ -void intel_gvt_clean_device(struct drm_i915_private *dev_priv) +void intel_gvt_clean_device(struct drm_i915_private *i915) { - struct intel_gvt *gvt = to_gvt(dev_priv); + struct intel_gvt *gvt = fetch_and_zero(&i915->gvt); - if (WARN_ON(!gvt)) + if (drm_WARN_ON(&i915->drm, !gvt)) return; intel_gvt_destroy_idle_vgpu(gvt->idle_vgpu); @@ -285,13 +284,12 @@ void intel_gvt_clean_device(struct drm_i915_private *dev_priv) intel_gvt_clean_mmio_info(gvt); idr_destroy(&gvt->vgpu_idr); - kfree(dev_priv->gvt); - dev_priv->gvt = NULL; + kfree(i915->gvt); } /** * intel_gvt_init_device - initialize a GVT device - * @dev_priv: drm i915 private data + * @i915: drm i915 private data * * This function is called at the initialization stage, to initialize * necessary GVT components. @@ -300,13 +298,13 @@ void intel_gvt_clean_device(struct drm_i915_private *dev_priv) * Zero on success, negative error code if failed. * */ -int intel_gvt_init_device(struct drm_i915_private *dev_priv) +int intel_gvt_init_device(struct drm_i915_private *i915) { struct intel_gvt *gvt; struct intel_vgpu *vgpu; int ret; - if (WARN_ON(dev_priv->gvt)) + if (drm_WARN_ON(&i915->drm, i915->gvt)) return -EEXIST; gvt = kzalloc(sizeof(struct intel_gvt), GFP_KERNEL); @@ -319,7 +317,8 @@ int intel_gvt_init_device(struct drm_i915_private *dev_priv) spin_lock_init(&gvt->scheduler.mmio_context_lock); mutex_init(&gvt->lock); mutex_init(&gvt->sched_lock); - gvt->dev_priv = dev_priv; + gvt->gt = &i915->gt; + i915->gvt = gvt; init_device_info(gvt); @@ -378,8 +377,7 @@ int intel_gvt_init_device(struct drm_i915_private *dev_priv) intel_gvt_debugfs_init(gvt); gvt_dbg_core("gvt device initialization is done\n"); - dev_priv->gvt = gvt; - intel_gvt_host.dev = &dev_priv->drm.pdev->dev; + intel_gvt_host.dev = &i915->drm.pdev->dev; intel_gvt_host.initialized = true; return 0; @@ -404,6 +402,7 @@ out_clean_mmio_info: out_clean_idr: idr_destroy(&gvt->vgpu_idr); kfree(gvt); + i915->gvt = NULL; return ret; } diff --git a/drivers/gpu/drm/i915/gvt/gvt.h b/drivers/gpu/drm/i915/gvt/gvt.h index b47c6acaf9c0..58c2c7932e3f 100644 --- a/drivers/gpu/drm/i915/gvt/gvt.h +++ b/drivers/gpu/drm/i915/gvt/gvt.h @@ -196,41 +196,21 @@ struct intel_vgpu { struct dentry *debugfs; -#if IS_ENABLED(CONFIG_DRM_I915_GVT_KVMGT) - struct { - struct mdev_device *mdev; - struct vfio_region *region; - int num_regions; - struct eventfd_ctx *intx_trigger; - struct eventfd_ctx *msi_trigger; - - /* - * Two caches are used to avoid mapping duplicated pages (eg. - * scratch pages). This help to reduce dma setup overhead. - */ - struct rb_root gfn_cache; - struct rb_root dma_addr_cache; - unsigned long nr_cache_entries; - struct mutex cache_lock; - - struct notifier_block iommu_notifier; - struct notifier_block group_notifier; - struct kvm *kvm; - struct work_struct release_work; - atomic_t released; - struct vfio_device *vfio_device; - } vdev; -#endif + /* Hypervisor-specific device state. */ + void *vdev; struct list_head dmabuf_obj_list_head; struct mutex dmabuf_lock; struct idr object_idr; - struct completion vblank_done; - u32 scan_nonprivbb; }; +static inline void *intel_vgpu_vdev(struct intel_vgpu *vgpu) +{ + return vgpu->vdev; +} + /* validating GM healthy status*/ #define vgpu_is_vm_unhealthy(ret_val) \ (((ret_val) == -EBADRQC) || ((ret_val) == -EFAULT)) @@ -306,7 +286,7 @@ struct intel_gvt { /* scheduler scope lock, protect gvt and vgpu schedule related data */ struct mutex sched_lock; - struct drm_i915_private *dev_priv; + struct intel_gt *gt; struct idr vgpu_idr; /* vGPU IDR pool */ struct intel_gvt_device_info device_info; @@ -376,14 +356,15 @@ int intel_gvt_load_firmware(struct intel_gvt *gvt); #define HOST_HIGH_GM_SIZE MB_TO_BYTES(384) #define HOST_FENCE 4 +#define gvt_to_ggtt(gvt) ((gvt)->gt->ggtt) + /* Aperture/GM space definitions for GVT device */ -#define gvt_aperture_sz(gvt) (gvt->dev_priv->ggtt.mappable_end) -#define gvt_aperture_pa_base(gvt) (gvt->dev_priv->ggtt.gmadr.start) +#define gvt_aperture_sz(gvt) gvt_to_ggtt(gvt)->mappable_end +#define gvt_aperture_pa_base(gvt) gvt_to_ggtt(gvt)->gmadr.start -#define gvt_ggtt_gm_sz(gvt) (gvt->dev_priv->ggtt.vm.total) -#define gvt_ggtt_sz(gvt) \ - ((gvt->dev_priv->ggtt.vm.total >> PAGE_SHIFT) << 3) -#define gvt_hidden_sz(gvt) (gvt_ggtt_gm_sz(gvt) - gvt_aperture_sz(gvt)) +#define gvt_ggtt_gm_sz(gvt) gvt_to_ggtt(gvt)->vm.total +#define gvt_ggtt_sz(gvt) (gvt_to_ggtt(gvt)->vm.total >> PAGE_SHIFT << 3) +#define gvt_hidden_sz(gvt) (gvt_ggtt_gm_sz(gvt) - gvt_aperture_sz(gvt)) #define gvt_aperture_gmadr_base(gvt) (0) #define gvt_aperture_gmadr_end(gvt) (gvt_aperture_gmadr_base(gvt) \ @@ -394,7 +375,7 @@ int intel_gvt_load_firmware(struct intel_gvt *gvt); #define gvt_hidden_gmadr_end(gvt) (gvt_hidden_gmadr_base(gvt) \ + gvt_hidden_sz(gvt) - 1) -#define gvt_fence_sz(gvt) ((gvt)->dev_priv->ggtt.num_fences) +#define gvt_fence_sz(gvt) (gvt_to_ggtt(gvt)->num_fences) /* Aperture/GM space definitions for vGPU */ #define vgpu_aperture_offset(vgpu) ((vgpu)->gm.low_gm_node.start) @@ -570,8 +551,7 @@ struct intel_gvt_ops { void (*vgpu_deactivate)(struct intel_vgpu *); struct intel_vgpu_type *(*gvt_find_vgpu_type)(struct intel_gvt *gvt, const char *name); - bool (*get_gvt_attrs)(struct attribute ***type_attrs, - struct attribute_group ***intel_vgpu_type_groups); + bool (*get_gvt_attrs)(struct attribute_group ***intel_vgpu_type_groups); int (*vgpu_query_plane)(struct intel_vgpu *vgpu, void *); int (*vgpu_get_dmabuf)(struct intel_vgpu *vgpu, unsigned int); int (*write_protect_handler)(struct intel_vgpu *, u64, void *, @@ -586,14 +566,14 @@ enum { GVT_FAILSAFE_GUEST_ERR, }; -static inline void mmio_hw_access_pre(struct drm_i915_private *dev_priv) +static inline void mmio_hw_access_pre(struct intel_gt *gt) { - intel_runtime_pm_get(&dev_priv->runtime_pm); + intel_runtime_pm_get(gt->uncore->rpm); } -static inline void mmio_hw_access_post(struct drm_i915_private *dev_priv) +static inline void mmio_hw_access_post(struct intel_gt *gt) { - intel_runtime_pm_put_unchecked(&dev_priv->runtime_pm); + intel_runtime_pm_put_unchecked(gt->uncore->rpm); } /** diff --git a/drivers/gpu/drm/i915/gvt/handlers.c b/drivers/gpu/drm/i915/gvt/handlers.c index ae6700dc9d73..0182e2a5acff 100644 --- a/drivers/gpu/drm/i915/gvt/handlers.c +++ b/drivers/gpu/drm/i915/gvt/handlers.c @@ -49,15 +49,17 @@ unsigned long intel_gvt_get_device_type(struct intel_gvt *gvt) { - if (IS_BROADWELL(gvt->dev_priv)) + struct drm_i915_private *i915 = gvt->gt->i915; + + if (IS_BROADWELL(i915)) return D_BDW; - else if (IS_SKYLAKE(gvt->dev_priv)) + else if (IS_SKYLAKE(i915)) return D_SKL; - else if (IS_KABYLAKE(gvt->dev_priv)) + else if (IS_KABYLAKE(i915)) return D_KBL; - else if (IS_BROXTON(gvt->dev_priv)) + else if (IS_BROXTON(i915)) return D_BXT; - else if (IS_COFFEELAKE(gvt->dev_priv)) + else if (IS_COFFEELAKE(i915)) return D_CFL; return 0; @@ -142,25 +144,25 @@ static int new_mmio_info(struct intel_gvt *gvt, } /** - * intel_gvt_render_mmio_to_ring_id - convert a mmio offset into ring id + * intel_gvt_render_mmio_to_engine - convert a mmio offset into the engine * @gvt: a GVT device * @offset: register offset * * Returns: - * Ring ID on success, negative error code if failed. + * The engine containing the offset within its mmio page. */ -int intel_gvt_render_mmio_to_ring_id(struct intel_gvt *gvt, - unsigned int offset) +const struct intel_engine_cs * +intel_gvt_render_mmio_to_engine(struct intel_gvt *gvt, unsigned int offset) { - enum intel_engine_id id; struct intel_engine_cs *engine; + enum intel_engine_id id; offset &= ~GENMASK(11, 0); - for_each_engine(engine, gvt->dev_priv, id) { + for_each_engine(engine, gvt->gt, id) if (engine->mmio_base == offset) - return id; - } - return -ENODEV; + return engine; + + return NULL; } #define offset_to_fence_num(offset) \ @@ -217,7 +219,7 @@ static int gamw_echo_dev_rw_ia_write(struct intel_vgpu *vgpu, { u32 ips = (*(u32 *)p_data) & GAMW_ECO_ENABLE_64K_IPS_FIELD; - if (INTEL_GEN(vgpu->gvt->dev_priv) <= 10) { + if (INTEL_GEN(vgpu->gvt->gt->i915) <= 10) { if (ips == GAMW_ECO_ENABLE_64K_IPS_FIELD) gvt_dbg_core("vgpu%d: ips enabled\n", vgpu->id); else if (!ips) @@ -253,7 +255,7 @@ static int fence_mmio_read(struct intel_vgpu *vgpu, unsigned int off, static int fence_mmio_write(struct intel_vgpu *vgpu, unsigned int off, void *p_data, unsigned int bytes) { - struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv; + struct intel_gvt *gvt = vgpu->gvt; unsigned int fence_num = offset_to_fence_num(off); int ret; @@ -262,10 +264,10 @@ static int fence_mmio_write(struct intel_vgpu *vgpu, unsigned int off, return ret; write_vreg(vgpu, off, p_data, bytes); - mmio_hw_access_pre(dev_priv); + mmio_hw_access_pre(gvt->gt); intel_vgpu_write_fence(vgpu, fence_num, vgpu_vreg64(vgpu, fence_num_to_offset(fence_num))); - mmio_hw_access_post(dev_priv); + mmio_hw_access_post(gvt->gt); return 0; } @@ -283,7 +285,7 @@ static int mul_force_wake_write(struct intel_vgpu *vgpu, old = vgpu_vreg(vgpu, offset); new = CALC_MODE_MASK_REG(old, *(u32 *)p_data); - if (INTEL_GEN(vgpu->gvt->dev_priv) >= 9) { + if (INTEL_GEN(vgpu->gvt->gt->i915) >= 9) { switch (offset) { case FORCEWAKE_RENDER_GEN9_REG: ack_reg_offset = FORCEWAKE_ACK_RENDER_GEN9_REG; @@ -345,7 +347,7 @@ static int gdrst_mmio_write(struct intel_vgpu *vgpu, unsigned int offset, gvt_dbg_mmio("vgpu%d: request GUC Reset\n", vgpu->id); vgpu_vreg_t(vgpu, GUC_STATUS) |= GS_MIA_IN_RESET; } - engine_mask &= INTEL_INFO(vgpu->gvt->dev_priv)->engine_mask; + engine_mask &= INTEL_INFO(vgpu->gvt->gt->i915)->engine_mask; } /* vgpu_lock already hold by emulate mmio r/w */ @@ -492,7 +494,7 @@ static i915_reg_t force_nonpriv_white_list[] = { }; /* a simple bsearch */ -static inline bool in_whitelist(unsigned int reg) +static inline bool in_whitelist(u32 reg) { int left = 0, right = ARRAY_SIZE(force_nonpriv_white_list); i915_reg_t *array = force_nonpriv_white_list; @@ -514,26 +516,21 @@ static int force_nonpriv_write(struct intel_vgpu *vgpu, unsigned int offset, void *p_data, unsigned int bytes) { u32 reg_nonpriv = (*(u32 *)p_data) & REG_GENMASK(25, 2); - int ring_id = intel_gvt_render_mmio_to_ring_id(vgpu->gvt, offset); - u32 ring_base; - struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv; - int ret = -EINVAL; - - if ((bytes != 4) || ((offset & (bytes - 1)) != 0) || ring_id < 0) { - gvt_err("vgpu(%d) ring %d Invalid FORCE_NONPRIV offset %x(%dB)\n", - vgpu->id, ring_id, offset, bytes); - return ret; - } + const struct intel_engine_cs *engine = + intel_gvt_render_mmio_to_engine(vgpu->gvt, offset); - ring_base = dev_priv->engine[ring_id]->mmio_base; + if (bytes != 4 || !IS_ALIGNED(offset, bytes) || !engine) { + gvt_err("vgpu(%d) Invalid FORCE_NONPRIV offset %x(%dB)\n", + vgpu->id, offset, bytes); + return -EINVAL; + } - if (in_whitelist(reg_nonpriv) || - reg_nonpriv == i915_mmio_reg_offset(RING_NOPID(ring_base))) { - ret = intel_vgpu_default_mmio_write(vgpu, offset, p_data, - bytes); - } else + if (!in_whitelist(reg_nonpriv) && + reg_nonpriv != i915_mmio_reg_offset(RING_NOPID(engine->mmio_base))) { gvt_err("vgpu(%d) Invalid FORCE_NONPRIV write %x at offset %x\n", - vgpu->id, *(u32 *)p_data, offset); + vgpu->id, reg_nonpriv, offset); + } else + intel_vgpu_default_mmio_write(vgpu, offset, p_data, bytes); return 0; } @@ -756,7 +753,7 @@ static int south_chicken2_mmio_write(struct intel_vgpu *vgpu, static int pri_surf_mmio_write(struct intel_vgpu *vgpu, unsigned int offset, void *p_data, unsigned int bytes) { - struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv; + struct drm_i915_private *dev_priv = vgpu->gvt->gt->i915; u32 pipe = DSPSURF_TO_PIPE(offset); int event = SKL_FLIP_EVENT(pipe, PLANE_PRIMARY); @@ -797,7 +794,7 @@ static int reg50080_mmio_write(struct intel_vgpu *vgpu, unsigned int offset, void *p_data, unsigned int bytes) { - struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv; + struct drm_i915_private *dev_priv = vgpu->gvt->gt->i915; enum pipe pipe = REG_50080_TO_PIPE(offset); enum plane_id plane = REG_50080_TO_PLANE(offset); int event = SKL_FLIP_EVENT(pipe, plane); @@ -821,7 +818,7 @@ static int reg50080_mmio_write(struct intel_vgpu *vgpu, static int trigger_aux_channel_interrupt(struct intel_vgpu *vgpu, unsigned int reg) { - struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv; + struct drm_i915_private *dev_priv = vgpu->gvt->gt->i915; enum intel_gvt_event_type event; if (reg == i915_mmio_reg_offset(DP_AUX_CH_CTL(AUX_CH_A))) @@ -836,7 +833,7 @@ static int trigger_aux_channel_interrupt(struct intel_vgpu *vgpu, reg == i915_mmio_reg_offset(DP_AUX_CH_CTL(AUX_CH_D))) event = AUX_CHANNEL_D; else { - WARN_ON(true); + drm_WARN_ON(&dev_priv->drm, true); return -EINVAL; } @@ -924,11 +921,11 @@ static int dp_aux_ch_ctl_mmio_write(struct intel_vgpu *vgpu, write_vreg(vgpu, offset, p_data, bytes); data = vgpu_vreg(vgpu, offset); - if ((INTEL_GEN(vgpu->gvt->dev_priv) >= 9) + if ((INTEL_GEN(vgpu->gvt->gt->i915) >= 9) && offset != _REG_SKL_DP_AUX_CH_CTL(port_index)) { /* SKL DPB/C/D aux ctl register changed */ return 0; - } else if (IS_BROADWELL(vgpu->gvt->dev_priv) && + } else if (IS_BROADWELL(vgpu->gvt->gt->i915) && offset != _REG_HSW_DP_AUX_CH_CTL(port_index)) { /* write to the data registers */ return 0; @@ -1244,8 +1241,7 @@ static int handle_g2v_notification(struct intel_vgpu *vgpu, int notification) static int send_display_ready_uevent(struct intel_vgpu *vgpu, int ready) { - struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv; - struct kobject *kobj = &dev_priv->drm.primary->kdev->kobj; + struct kobject *kobj = &vgpu->gvt->gt->i915->drm.primary->kdev->kobj; char *env[3] = {NULL, NULL, NULL}; char vmid_str[20]; char display_ready_str[20]; @@ -1306,13 +1302,15 @@ static int pvinfo_mmio_write(struct intel_vgpu *vgpu, unsigned int offset, static int pf_write(struct intel_vgpu *vgpu, unsigned int offset, void *p_data, unsigned int bytes) { + struct drm_i915_private *i915 = vgpu->gvt->gt->i915; u32 val = *(u32 *)p_data; if ((offset == _PS_1A_CTRL || offset == _PS_2A_CTRL || offset == _PS_1B_CTRL || offset == _PS_2B_CTRL || offset == _PS_1C_CTRL) && (val & PS_PLANE_SEL_MASK) != 0) { - WARN_ONCE(true, "VM(%d): guest is trying to scaling a plane\n", - vgpu->id); + drm_WARN_ONCE(&i915->drm, true, + "VM(%d): guest is trying to scaling a plane\n", + vgpu->id); return 0; } @@ -1360,13 +1358,15 @@ static int fpga_dbg_mmio_write(struct intel_vgpu *vgpu, static int dma_ctrl_write(struct intel_vgpu *vgpu, unsigned int offset, void *p_data, unsigned int bytes) { + struct drm_i915_private *i915 = vgpu->gvt->gt->i915; u32 mode; write_vreg(vgpu, offset, p_data, bytes); mode = vgpu_vreg(vgpu, offset); if (GFX_MODE_BIT_SET_IN_MASK(mode, START_DMA)) { - WARN_ONCE(1, "VM(%d): iGVT-g doesn't support GuC\n", + drm_WARN_ONCE(&i915->drm, 1, + "VM(%d): iGVT-g doesn't support GuC\n", vgpu->id); return 0; } @@ -1377,10 +1377,12 @@ static int dma_ctrl_write(struct intel_vgpu *vgpu, unsigned int offset, static int gen9_trtte_write(struct intel_vgpu *vgpu, unsigned int offset, void *p_data, unsigned int bytes) { + struct drm_i915_private *i915 = vgpu->gvt->gt->i915; u32 trtte = *(u32 *)p_data; if ((trtte & 1) && (trtte & (1 << 1)) == 0) { - WARN(1, "VM(%d): Use physical address for TRTT!\n", + drm_WARN(&i915->drm, 1, + "VM(%d): Use physical address for TRTT!\n", vgpu->id); return -EINVAL; } @@ -1427,9 +1429,9 @@ static int mailbox_write(struct intel_vgpu *vgpu, unsigned int offset, switch (cmd) { case GEN9_PCODE_READ_MEM_LATENCY: - if (IS_SKYLAKE(vgpu->gvt->dev_priv) - || IS_KABYLAKE(vgpu->gvt->dev_priv) - || IS_COFFEELAKE(vgpu->gvt->dev_priv)) { + if (IS_SKYLAKE(vgpu->gvt->gt->i915) || + IS_KABYLAKE(vgpu->gvt->gt->i915) || + IS_COFFEELAKE(vgpu->gvt->gt->i915)) { /** * "Read memory latency" command on gen9. * Below memory latency values are read @@ -1439,7 +1441,7 @@ static int mailbox_write(struct intel_vgpu *vgpu, unsigned int offset, *data0 = 0x1e1a1100; else *data0 = 0x61514b3d; - } else if (IS_BROXTON(vgpu->gvt->dev_priv)) { + } else if (IS_BROXTON(vgpu->gvt->gt->i915)) { /** * "Read memory latency" command on gen9. * Below memory latency values are read @@ -1452,9 +1454,9 @@ static int mailbox_write(struct intel_vgpu *vgpu, unsigned int offset, } break; case SKL_PCODE_CDCLK_CONTROL: - if (IS_SKYLAKE(vgpu->gvt->dev_priv) - || IS_KABYLAKE(vgpu->gvt->dev_priv) - || IS_COFFEELAKE(vgpu->gvt->dev_priv)) + if (IS_SKYLAKE(vgpu->gvt->gt->i915) || + IS_KABYLAKE(vgpu->gvt->gt->i915) || + IS_COFFEELAKE(vgpu->gvt->gt->i915)) *data0 = SKL_CDCLK_READY_FOR_CHANGE; break; case GEN6_PCODE_READ_RC6VIDS: @@ -1478,24 +1480,26 @@ static int hws_pga_write(struct intel_vgpu *vgpu, unsigned int offset, void *p_data, unsigned int bytes) { u32 value = *(u32 *)p_data; - int ring_id = intel_gvt_render_mmio_to_ring_id(vgpu->gvt, offset); + const struct intel_engine_cs *engine = + intel_gvt_render_mmio_to_engine(vgpu->gvt, offset); if (!intel_gvt_ggtt_validate_range(vgpu, value, I915_GTT_PAGE_SIZE)) { gvt_vgpu_err("write invalid HWSP address, reg:0x%x, value:0x%x\n", offset, value); return -EINVAL; } + /* * Need to emulate all the HWSP register write to ensure host can * update the VM CSB status correctly. Here listed registers can * support BDW, SKL or other platforms with same HWSP registers. */ - if (unlikely(ring_id < 0 || ring_id >= I915_NUM_ENGINES)) { + if (unlikely(!engine)) { gvt_vgpu_err("access unknown hardware status page register:0x%x\n", offset); return -EINVAL; } - vgpu->hws_pga[ring_id] = value; + vgpu->hws_pga[engine->id] = value; gvt_dbg_mmio("VM(%d) write: 0x%x to HWSP: 0x%x\n", vgpu->id, value, offset); @@ -1507,7 +1511,7 @@ static int skl_power_well_ctl_write(struct intel_vgpu *vgpu, { u32 v = *(u32 *)p_data; - if (IS_BROXTON(vgpu->gvt->dev_priv)) + if (IS_BROXTON(vgpu->gvt->gt->i915)) v &= (1 << 31) | (1 << 29); else v &= (1 << 31) | (1 << 29) | (1 << 9) | @@ -1654,26 +1658,24 @@ static int mmio_read_from_hw(struct intel_vgpu *vgpu, unsigned int offset, void *p_data, unsigned int bytes) { struct intel_gvt *gvt = vgpu->gvt; - struct drm_i915_private *dev_priv = gvt->dev_priv; - int ring_id; - u32 ring_base; + const struct intel_engine_cs *engine = + intel_gvt_render_mmio_to_engine(gvt, offset); - ring_id = intel_gvt_render_mmio_to_ring_id(gvt, offset); /** * Read HW reg in following case * a. the offset isn't a ring mmio * b. the offset's ring is running on hw. * c. the offset is ring time stamp mmio */ - if (ring_id >= 0) - ring_base = dev_priv->engine[ring_id]->mmio_base; - - if (ring_id < 0 || vgpu == gvt->scheduler.engine_owner[ring_id] || - offset == i915_mmio_reg_offset(RING_TIMESTAMP(ring_base)) || - offset == i915_mmio_reg_offset(RING_TIMESTAMP_UDW(ring_base))) { - mmio_hw_access_pre(dev_priv); - vgpu_vreg(vgpu, offset) = I915_READ(_MMIO(offset)); - mmio_hw_access_post(dev_priv); + + if (!engine || + vgpu == gvt->scheduler.engine_owner[engine->id] || + offset == i915_mmio_reg_offset(RING_TIMESTAMP(engine->mmio_base)) || + offset == i915_mmio_reg_offset(RING_TIMESTAMP_UDW(engine->mmio_base))) { + mmio_hw_access_pre(gvt->gt); + vgpu_vreg(vgpu, offset) = + intel_uncore_read(gvt->gt->uncore, _MMIO(offset)); + mmio_hw_access_post(gvt->gt); } return intel_vgpu_default_mmio_read(vgpu, offset, p_data, bytes); @@ -1682,22 +1684,23 @@ static int mmio_read_from_hw(struct intel_vgpu *vgpu, static int elsp_mmio_write(struct intel_vgpu *vgpu, unsigned int offset, void *p_data, unsigned int bytes) { - int ring_id = intel_gvt_render_mmio_to_ring_id(vgpu->gvt, offset); + struct drm_i915_private *i915 = vgpu->gvt->gt->i915; + const struct intel_engine_cs *engine = intel_gvt_render_mmio_to_engine(vgpu->gvt, offset); struct intel_vgpu_execlist *execlist; u32 data = *(u32 *)p_data; int ret = 0; - if (WARN_ON(ring_id < 0 || ring_id >= I915_NUM_ENGINES)) + if (drm_WARN_ON(&i915->drm, !engine)) return -EINVAL; - execlist = &vgpu->submission.execlist[ring_id]; + execlist = &vgpu->submission.execlist[engine->id]; execlist->elsp_dwords.data[3 - execlist->elsp_dwords.index] = data; if (execlist->elsp_dwords.index == 3) { - ret = intel_vgpu_submit_execlist(vgpu, ring_id); + ret = intel_vgpu_submit_execlist(vgpu, engine); if(ret) - gvt_vgpu_err("fail submit workload on ring %d\n", - ring_id); + gvt_vgpu_err("fail submit workload on ring %s\n", + engine->name); } ++execlist->elsp_dwords.index; @@ -1709,12 +1712,13 @@ static int ring_mode_mmio_write(struct intel_vgpu *vgpu, unsigned int offset, void *p_data, unsigned int bytes) { u32 data = *(u32 *)p_data; - int ring_id = intel_gvt_render_mmio_to_ring_id(vgpu->gvt, offset); + const struct intel_engine_cs *engine = + intel_gvt_render_mmio_to_engine(vgpu->gvt, offset); bool enable_execlist; int ret; (*(u32 *)p_data) &= ~_MASKED_BIT_ENABLE(1); - if (IS_COFFEELAKE(vgpu->gvt->dev_priv)) + if (IS_COFFEELAKE(vgpu->gvt->gt->i915)) (*(u32 *)p_data) &= ~_MASKED_BIT_ENABLE(2); write_vreg(vgpu, offset, p_data, bytes); @@ -1723,7 +1727,7 @@ static int ring_mode_mmio_write(struct intel_vgpu *vgpu, unsigned int offset, return 0; } - if (IS_COFFEELAKE(vgpu->gvt->dev_priv) && + if (IS_COFFEELAKE(vgpu->gvt->gt->i915) && data & _MASKED_BIT_ENABLE(2)) { enter_failsafe_mode(vgpu, GVT_FAILSAFE_UNSUPPORTED_GUEST); return 0; @@ -1743,16 +1747,16 @@ static int ring_mode_mmio_write(struct intel_vgpu *vgpu, unsigned int offset, || (data & _MASKED_BIT_DISABLE(GFX_RUN_LIST_ENABLE))) { enable_execlist = !!(data & GFX_RUN_LIST_ENABLE); - gvt_dbg_core("EXECLIST %s on ring %d\n", - (enable_execlist ? "enabling" : "disabling"), - ring_id); + gvt_dbg_core("EXECLIST %s on ring %s\n", + (enable_execlist ? "enabling" : "disabling"), + engine->name); if (!enable_execlist) return 0; ret = intel_vgpu_select_submission_ops(vgpu, - BIT(ring_id), - INTEL_VGPU_EXECLIST_SUBMISSION); + engine->mask, + INTEL_VGPU_EXECLIST_SUBMISSION); if (ret) return ret; @@ -1876,7 +1880,7 @@ static int csfe_chicken1_mmio_write(struct intel_vgpu *vgpu, static int init_generic_mmio_info(struct intel_gvt *gvt) { - struct drm_i915_private *dev_priv = gvt->dev_priv; + struct drm_i915_private *dev_priv = gvt->gt->i915; int ret; MMIO_RING_DFH(RING_IMR, D_ALL, F_CMD_ACCESS, NULL, @@ -2693,7 +2697,7 @@ static int init_generic_mmio_info(struct intel_gvt *gvt) static int init_bdw_mmio_info(struct intel_gvt *gvt) { - struct drm_i915_private *dev_priv = gvt->dev_priv; + struct drm_i915_private *dev_priv = gvt->gt->i915; int ret; MMIO_DH(GEN8_GT_IMR(0), D_BDW_PLUS, NULL, intel_vgpu_reg_imr_handler); @@ -2882,7 +2886,7 @@ static int init_bdw_mmio_info(struct intel_gvt *gvt) static int init_skl_mmio_info(struct intel_gvt *gvt) { - struct drm_i915_private *dev_priv = gvt->dev_priv; + struct drm_i915_private *dev_priv = gvt->gt->i915; int ret; MMIO_DH(FORCEWAKE_RENDER_GEN9, D_SKL_PLUS, NULL, mul_force_wake_write); @@ -3131,7 +3135,7 @@ static int init_skl_mmio_info(struct intel_gvt *gvt) static int init_bxt_mmio_info(struct intel_gvt *gvt) { - struct drm_i915_private *dev_priv = gvt->dev_priv; + struct drm_i915_private *dev_priv = gvt->gt->i915; int ret; MMIO_F(_MMIO(0x80000), 0x3000, 0, 0, 0, D_BXT, NULL, NULL); @@ -3367,7 +3371,7 @@ static struct gvt_mmio_block mmio_blocks[] = { int intel_gvt_setup_mmio_info(struct intel_gvt *gvt) { struct intel_gvt_device_info *info = &gvt->device_info; - struct drm_i915_private *dev_priv = gvt->dev_priv; + struct drm_i915_private *i915 = gvt->gt->i915; int size = info->mmio_size / 4 * sizeof(*gvt->mmio.mmio_attribute); int ret; @@ -3379,20 +3383,20 @@ int intel_gvt_setup_mmio_info(struct intel_gvt *gvt) if (ret) goto err; - if (IS_BROADWELL(dev_priv)) { + if (IS_BROADWELL(i915)) { ret = init_bdw_mmio_info(gvt); if (ret) goto err; - } else if (IS_SKYLAKE(dev_priv) - || IS_KABYLAKE(dev_priv) - || IS_COFFEELAKE(dev_priv)) { + } else if (IS_SKYLAKE(i915) || + IS_KABYLAKE(i915) || + IS_COFFEELAKE(i915)) { ret = init_bdw_mmio_info(gvt); if (ret) goto err; ret = init_skl_mmio_info(gvt); if (ret) goto err; - } else if (IS_BROXTON(dev_priv)) { + } else if (IS_BROXTON(i915)) { ret = init_bdw_mmio_info(gvt); if (ret) goto err; @@ -3541,13 +3545,14 @@ bool intel_gvt_in_force_nonpriv_whitelist(struct intel_gvt *gvt, int intel_vgpu_mmio_reg_rw(struct intel_vgpu *vgpu, unsigned int offset, void *pdata, unsigned int bytes, bool is_read) { + struct drm_i915_private *i915 = vgpu->gvt->gt->i915; struct intel_gvt *gvt = vgpu->gvt; struct intel_gvt_mmio_info *mmio_info; struct gvt_mmio_block *mmio_block; gvt_mmio_func func; int ret; - if (WARN_ON(bytes > 8)) + if (drm_WARN_ON(&i915->drm, bytes > 8)) return -EINVAL; /* diff --git a/drivers/gpu/drm/i915/gvt/interrupt.c b/drivers/gpu/drm/i915/gvt/interrupt.c index 11accd3e1023..540017fed908 100644 --- a/drivers/gpu/drm/i915/gvt/interrupt.c +++ b/drivers/gpu/drm/i915/gvt/interrupt.c @@ -245,6 +245,7 @@ int intel_vgpu_reg_ier_handler(struct intel_vgpu *vgpu, unsigned int reg, void *p_data, unsigned int bytes) { struct intel_gvt *gvt = vgpu->gvt; + struct drm_i915_private *i915 = gvt->gt->i915; struct intel_gvt_irq_ops *ops = gvt->irq.ops; struct intel_gvt_irq_info *info; u32 ier = *(u32 *)p_data; @@ -255,7 +256,7 @@ int intel_vgpu_reg_ier_handler(struct intel_vgpu *vgpu, vgpu_vreg(vgpu, reg) = ier; info = regbase_to_irq_info(gvt, ier_to_regbase(reg)); - if (WARN_ON(!info)) + if (drm_WARN_ON(&i915->drm, !info)) return -EINVAL; if (info->has_upstream_irq) @@ -282,6 +283,7 @@ int intel_vgpu_reg_ier_handler(struct intel_vgpu *vgpu, int intel_vgpu_reg_iir_handler(struct intel_vgpu *vgpu, unsigned int reg, void *p_data, unsigned int bytes) { + struct drm_i915_private *i915 = vgpu->gvt->gt->i915; struct intel_gvt_irq_info *info = regbase_to_irq_info(vgpu->gvt, iir_to_regbase(reg)); u32 iir = *(u32 *)p_data; @@ -289,7 +291,7 @@ int intel_vgpu_reg_iir_handler(struct intel_vgpu *vgpu, unsigned int reg, trace_write_ir(vgpu->id, "IIR", reg, iir, vgpu_vreg(vgpu, reg), (vgpu_vreg(vgpu, reg) ^ iir)); - if (WARN_ON(!info)) + if (drm_WARN_ON(&i915->drm, !info)) return -EINVAL; vgpu_vreg(vgpu, reg) &= ~iir; @@ -319,6 +321,7 @@ static struct intel_gvt_irq_map gen8_irq_map[] = { static void update_upstream_irq(struct intel_vgpu *vgpu, struct intel_gvt_irq_info *info) { + struct drm_i915_private *i915 = vgpu->gvt->gt->i915; struct intel_gvt_irq *irq = &vgpu->gvt->irq; struct intel_gvt_irq_map *map = irq->irq_map; struct intel_gvt_irq_info *up_irq_info = NULL; @@ -340,7 +343,8 @@ static void update_upstream_irq(struct intel_vgpu *vgpu, if (!up_irq_info) up_irq_info = irq->info[map->up_irq_group]; else - WARN_ON(up_irq_info != irq->info[map->up_irq_group]); + drm_WARN_ON(&i915->drm, up_irq_info != + irq->info[map->up_irq_group]); bit = map->up_irq_bit; @@ -350,7 +354,7 @@ static void update_upstream_irq(struct intel_vgpu *vgpu, clear_bits |= (1 << bit); } - if (WARN_ON(!up_irq_info)) + if (drm_WARN_ON(&i915->drm, !up_irq_info)) return; if (up_irq_info->group == INTEL_GVT_IRQ_INFO_MASTER) { @@ -536,7 +540,7 @@ static void gen8_init_irq( SET_BIT_INFO(irq, 4, VCS_MI_FLUSH_DW, INTEL_GVT_IRQ_INFO_GT1); SET_BIT_INFO(irq, 8, VCS_AS_CONTEXT_SWITCH, INTEL_GVT_IRQ_INFO_GT1); - if (HAS_ENGINE(gvt->dev_priv, VCS1)) { + if (HAS_ENGINE(gvt->gt->i915, VCS1)) { SET_BIT_INFO(irq, 16, VCS2_MI_USER_INTERRUPT, INTEL_GVT_IRQ_INFO_GT1); SET_BIT_INFO(irq, 20, VCS2_MI_FLUSH_DW, @@ -568,7 +572,7 @@ static void gen8_init_irq( SET_BIT_INFO(irq, 22, DP_C_HOTPLUG, INTEL_GVT_IRQ_INFO_PCH); SET_BIT_INFO(irq, 23, DP_D_HOTPLUG, INTEL_GVT_IRQ_INFO_PCH); - if (IS_BROADWELL(gvt->dev_priv)) { + if (IS_BROADWELL(gvt->gt->i915)) { SET_BIT_INFO(irq, 25, AUX_CHANNEL_B, INTEL_GVT_IRQ_INFO_PCH); SET_BIT_INFO(irq, 26, AUX_CHANNEL_C, INTEL_GVT_IRQ_INFO_PCH); SET_BIT_INFO(irq, 27, AUX_CHANNEL_D, INTEL_GVT_IRQ_INFO_PCH); @@ -581,7 +585,7 @@ static void gen8_init_irq( SET_BIT_INFO(irq, 4, PRIMARY_C_FLIP_DONE, INTEL_GVT_IRQ_INFO_DE_PIPE_C); SET_BIT_INFO(irq, 5, SPRITE_C_FLIP_DONE, INTEL_GVT_IRQ_INFO_DE_PIPE_C); - } else if (INTEL_GEN(gvt->dev_priv) >= 9) { + } else if (INTEL_GEN(gvt->gt->i915) >= 9) { SET_BIT_INFO(irq, 25, AUX_CHANNEL_B, INTEL_GVT_IRQ_INFO_DE_PORT); SET_BIT_INFO(irq, 26, AUX_CHANNEL_C, INTEL_GVT_IRQ_INFO_DE_PORT); SET_BIT_INFO(irq, 27, AUX_CHANNEL_D, INTEL_GVT_IRQ_INFO_DE_PORT); @@ -618,13 +622,14 @@ static struct intel_gvt_irq_ops gen8_irq_ops = { void intel_vgpu_trigger_virtual_event(struct intel_vgpu *vgpu, enum intel_gvt_event_type event) { + struct drm_i915_private *i915 = vgpu->gvt->gt->i915; struct intel_gvt *gvt = vgpu->gvt; struct intel_gvt_irq *irq = &gvt->irq; gvt_event_virt_handler_t handler; struct intel_gvt_irq_ops *ops = gvt->irq.ops; handler = get_event_virt_handler(irq, event); - WARN_ON(!handler); + drm_WARN_ON(&i915->drm, !handler); handler(irq, event, vgpu); diff --git a/drivers/gpu/drm/i915/gvt/kvmgt.c b/drivers/gpu/drm/i915/gvt/kvmgt.c index 3259a1fa69e1..074c4efb58eb 100644 --- a/drivers/gpu/drm/i915/gvt/kvmgt.c +++ b/drivers/gpu/drm/i915/gvt/kvmgt.c @@ -108,6 +108,36 @@ struct gvt_dma { struct kref ref; }; +struct kvmgt_vdev { + struct intel_vgpu *vgpu; + struct mdev_device *mdev; + struct vfio_region *region; + int num_regions; + struct eventfd_ctx *intx_trigger; + struct eventfd_ctx *msi_trigger; + + /* + * Two caches are used to avoid mapping duplicated pages (eg. + * scratch pages). This help to reduce dma setup overhead. + */ + struct rb_root gfn_cache; + struct rb_root dma_addr_cache; + unsigned long nr_cache_entries; + struct mutex cache_lock; + + struct notifier_block iommu_notifier; + struct notifier_block group_notifier; + struct kvm *kvm; + struct work_struct release_work; + atomic_t released; + struct vfio_device *vfio_device; +}; + +static inline struct kvmgt_vdev *kvmgt_vdev(struct intel_vgpu *vgpu) +{ + return intel_vgpu_vdev(vgpu); +} + static inline bool handle_valid(unsigned long handle) { return !!(handle & ~0xff); @@ -120,6 +150,7 @@ static bool kvmgt_guest_exit(struct kvmgt_guest_info *info); static void gvt_unpin_guest_page(struct intel_vgpu *vgpu, unsigned long gfn, unsigned long size) { + struct drm_i915_private *i915 = vgpu->gvt->gt->i915; int total_pages; int npage; int ret; @@ -129,8 +160,8 @@ static void gvt_unpin_guest_page(struct intel_vgpu *vgpu, unsigned long gfn, for (npage = 0; npage < total_pages; npage++) { unsigned long cur_gfn = gfn + npage; - ret = vfio_unpin_pages(mdev_dev(vgpu->vdev.mdev), &cur_gfn, 1); - WARN_ON(ret != 1); + ret = vfio_unpin_pages(mdev_dev(kvmgt_vdev(vgpu)->mdev), &cur_gfn, 1); + drm_WARN_ON(&i915->drm, ret != 1); } } @@ -152,7 +183,7 @@ static int gvt_pin_guest_page(struct intel_vgpu *vgpu, unsigned long gfn, unsigned long cur_gfn = gfn + npage; unsigned long pfn; - ret = vfio_pin_pages(mdev_dev(vgpu->vdev.mdev), &cur_gfn, 1, + ret = vfio_pin_pages(mdev_dev(kvmgt_vdev(vgpu)->mdev), &cur_gfn, 1, IOMMU_READ | IOMMU_WRITE, &pfn); if (ret != 1) { gvt_vgpu_err("vfio_pin_pages failed for gfn 0x%lx, ret %d\n", @@ -187,7 +218,7 @@ err: static int gvt_dma_map_page(struct intel_vgpu *vgpu, unsigned long gfn, dma_addr_t *dma_addr, unsigned long size) { - struct device *dev = &vgpu->gvt->dev_priv->drm.pdev->dev; + struct device *dev = &vgpu->gvt->gt->i915->drm.pdev->dev; struct page *page = NULL; int ret; @@ -210,7 +241,7 @@ static int gvt_dma_map_page(struct intel_vgpu *vgpu, unsigned long gfn, static void gvt_dma_unmap_page(struct intel_vgpu *vgpu, unsigned long gfn, dma_addr_t dma_addr, unsigned long size) { - struct device *dev = &vgpu->gvt->dev_priv->drm.pdev->dev; + struct device *dev = &vgpu->gvt->gt->i915->drm.pdev->dev; dma_unmap_page(dev, dma_addr, size, PCI_DMA_BIDIRECTIONAL); gvt_unpin_guest_page(vgpu, gfn, size); @@ -219,7 +250,7 @@ static void gvt_dma_unmap_page(struct intel_vgpu *vgpu, unsigned long gfn, static struct gvt_dma *__gvt_cache_find_dma_addr(struct intel_vgpu *vgpu, dma_addr_t dma_addr) { - struct rb_node *node = vgpu->vdev.dma_addr_cache.rb_node; + struct rb_node *node = kvmgt_vdev(vgpu)->dma_addr_cache.rb_node; struct gvt_dma *itr; while (node) { @@ -237,7 +268,7 @@ static struct gvt_dma *__gvt_cache_find_dma_addr(struct intel_vgpu *vgpu, static struct gvt_dma *__gvt_cache_find_gfn(struct intel_vgpu *vgpu, gfn_t gfn) { - struct rb_node *node = vgpu->vdev.gfn_cache.rb_node; + struct rb_node *node = kvmgt_vdev(vgpu)->gfn_cache.rb_node; struct gvt_dma *itr; while (node) { @@ -258,6 +289,7 @@ static int __gvt_cache_add(struct intel_vgpu *vgpu, gfn_t gfn, { struct gvt_dma *new, *itr; struct rb_node **link, *parent = NULL; + struct kvmgt_vdev *vdev = kvmgt_vdev(vgpu); new = kzalloc(sizeof(struct gvt_dma), GFP_KERNEL); if (!new) @@ -270,7 +302,7 @@ static int __gvt_cache_add(struct intel_vgpu *vgpu, gfn_t gfn, kref_init(&new->ref); /* gfn_cache maps gfn to struct gvt_dma. */ - link = &vgpu->vdev.gfn_cache.rb_node; + link = &vdev->gfn_cache.rb_node; while (*link) { parent = *link; itr = rb_entry(parent, struct gvt_dma, gfn_node); @@ -281,11 +313,11 @@ static int __gvt_cache_add(struct intel_vgpu *vgpu, gfn_t gfn, link = &parent->rb_right; } rb_link_node(&new->gfn_node, parent, link); - rb_insert_color(&new->gfn_node, &vgpu->vdev.gfn_cache); + rb_insert_color(&new->gfn_node, &vdev->gfn_cache); /* dma_addr_cache maps dma addr to struct gvt_dma. */ parent = NULL; - link = &vgpu->vdev.dma_addr_cache.rb_node; + link = &vdev->dma_addr_cache.rb_node; while (*link) { parent = *link; itr = rb_entry(parent, struct gvt_dma, dma_addr_node); @@ -296,46 +328,51 @@ static int __gvt_cache_add(struct intel_vgpu *vgpu, gfn_t gfn, link = &parent->rb_right; } rb_link_node(&new->dma_addr_node, parent, link); - rb_insert_color(&new->dma_addr_node, &vgpu->vdev.dma_addr_cache); + rb_insert_color(&new->dma_addr_node, &vdev->dma_addr_cache); - vgpu->vdev.nr_cache_entries++; + vdev->nr_cache_entries++; return 0; } static void __gvt_cache_remove_entry(struct intel_vgpu *vgpu, struct gvt_dma *entry) { - rb_erase(&entry->gfn_node, &vgpu->vdev.gfn_cache); - rb_erase(&entry->dma_addr_node, &vgpu->vdev.dma_addr_cache); + struct kvmgt_vdev *vdev = kvmgt_vdev(vgpu); + + rb_erase(&entry->gfn_node, &vdev->gfn_cache); + rb_erase(&entry->dma_addr_node, &vdev->dma_addr_cache); kfree(entry); - vgpu->vdev.nr_cache_entries--; + vdev->nr_cache_entries--; } static void gvt_cache_destroy(struct intel_vgpu *vgpu) { struct gvt_dma *dma; struct rb_node *node = NULL; + struct kvmgt_vdev *vdev = kvmgt_vdev(vgpu); for (;;) { - mutex_lock(&vgpu->vdev.cache_lock); - node = rb_first(&vgpu->vdev.gfn_cache); + mutex_lock(&vdev->cache_lock); + node = rb_first(&vdev->gfn_cache); if (!node) { - mutex_unlock(&vgpu->vdev.cache_lock); + mutex_unlock(&vdev->cache_lock); break; } dma = rb_entry(node, struct gvt_dma, gfn_node); gvt_dma_unmap_page(vgpu, dma->gfn, dma->dma_addr, dma->size); __gvt_cache_remove_entry(vgpu, dma); - mutex_unlock(&vgpu->vdev.cache_lock); + mutex_unlock(&vdev->cache_lock); } } static void gvt_cache_init(struct intel_vgpu *vgpu) { - vgpu->vdev.gfn_cache = RB_ROOT; - vgpu->vdev.dma_addr_cache = RB_ROOT; - vgpu->vdev.nr_cache_entries = 0; - mutex_init(&vgpu->vdev.cache_lock); + struct kvmgt_vdev *vdev = kvmgt_vdev(vgpu); + + vdev->gfn_cache = RB_ROOT; + vdev->dma_addr_cache = RB_ROOT; + vdev->nr_cache_entries = 0; + mutex_init(&vdev->cache_lock); } static void kvmgt_protect_table_init(struct kvmgt_guest_info *info) @@ -409,16 +446,18 @@ static void kvmgt_protect_table_del(struct kvmgt_guest_info *info, static size_t intel_vgpu_reg_rw_opregion(struct intel_vgpu *vgpu, char *buf, size_t count, loff_t *ppos, bool iswrite) { + struct kvmgt_vdev *vdev = kvmgt_vdev(vgpu); unsigned int i = VFIO_PCI_OFFSET_TO_INDEX(*ppos) - VFIO_PCI_NUM_REGIONS; - void *base = vgpu->vdev.region[i].data; + void *base = vdev->region[i].data; loff_t pos = *ppos & VFIO_PCI_OFFSET_MASK; - if (pos >= vgpu->vdev.region[i].size || iswrite) { + + if (pos >= vdev->region[i].size || iswrite) { gvt_vgpu_err("invalid op or offset for Intel vgpu OpRegion\n"); return -EINVAL; } - count = min(count, (size_t)(vgpu->vdev.region[i].size - pos)); + count = min(count, (size_t)(vdev->region[i].size - pos)); memcpy(buf, base + pos, count); return count; @@ -512,7 +551,7 @@ static size_t intel_vgpu_reg_rw_edid(struct intel_vgpu *vgpu, char *buf, unsigned int i = VFIO_PCI_OFFSET_TO_INDEX(*ppos) - VFIO_PCI_NUM_REGIONS; struct vfio_edid_region *region = - (struct vfio_edid_region *)vgpu->vdev.region[i].data; + (struct vfio_edid_region *)kvmgt_vdev(vgpu)->region[i].data; loff_t pos = *ppos & VFIO_PCI_OFFSET_MASK; if (pos < region->vfio_edid_regs.edid_offset) { @@ -544,32 +583,34 @@ static int intel_vgpu_register_reg(struct intel_vgpu *vgpu, const struct intel_vgpu_regops *ops, size_t size, u32 flags, void *data) { + struct kvmgt_vdev *vdev = kvmgt_vdev(vgpu); struct vfio_region *region; - region = krealloc(vgpu->vdev.region, - (vgpu->vdev.num_regions + 1) * sizeof(*region), + region = krealloc(vdev->region, + (vdev->num_regions + 1) * sizeof(*region), GFP_KERNEL); if (!region) return -ENOMEM; - vgpu->vdev.region = region; - vgpu->vdev.region[vgpu->vdev.num_regions].type = type; - vgpu->vdev.region[vgpu->vdev.num_regions].subtype = subtype; - vgpu->vdev.region[vgpu->vdev.num_regions].ops = ops; - vgpu->vdev.region[vgpu->vdev.num_regions].size = size; - vgpu->vdev.region[vgpu->vdev.num_regions].flags = flags; - vgpu->vdev.region[vgpu->vdev.num_regions].data = data; - vgpu->vdev.num_regions++; + vdev->region = region; + vdev->region[vdev->num_regions].type = type; + vdev->region[vdev->num_regions].subtype = subtype; + vdev->region[vdev->num_regions].ops = ops; + vdev->region[vdev->num_regions].size = size; + vdev->region[vdev->num_regions].flags = flags; + vdev->region[vdev->num_regions].data = data; + vdev->num_regions++; return 0; } static int kvmgt_get_vfio_device(void *p_vgpu) { struct intel_vgpu *vgpu = (struct intel_vgpu *)p_vgpu; + struct kvmgt_vdev *vdev = kvmgt_vdev(vgpu); - vgpu->vdev.vfio_device = vfio_device_get_from_dev( - mdev_dev(vgpu->vdev.mdev)); - if (!vgpu->vdev.vfio_device) { + vdev->vfio_device = vfio_device_get_from_dev( + mdev_dev(vdev->mdev)); + if (!vdev->vfio_device) { gvt_vgpu_err("failed to get vfio device\n"); return -ENODEV; } @@ -637,10 +678,12 @@ static int kvmgt_set_edid(void *p_vgpu, int port_num) static void kvmgt_put_vfio_device(void *vgpu) { - if (WARN_ON(!((struct intel_vgpu *)vgpu)->vdev.vfio_device)) + struct kvmgt_vdev *vdev = kvmgt_vdev((struct intel_vgpu *)vgpu); + + if (WARN_ON(!vdev->vfio_device)) return; - vfio_device_put(((struct intel_vgpu *)vgpu)->vdev.vfio_device); + vfio_device_put(vdev->vfio_device); } static int intel_vgpu_create(struct kobject *kobj, struct mdev_device *mdev) @@ -669,9 +712,9 @@ static int intel_vgpu_create(struct kobject *kobj, struct mdev_device *mdev) goto out; } - INIT_WORK(&vgpu->vdev.release_work, intel_vgpu_release_work); + INIT_WORK(&kvmgt_vdev(vgpu)->release_work, intel_vgpu_release_work); - vgpu->vdev.mdev = mdev; + kvmgt_vdev(vgpu)->mdev = mdev; mdev_set_drvdata(mdev, vgpu); gvt_dbg_core("intel_vgpu_create succeeded for mdev: %s\n", @@ -696,9 +739,10 @@ static int intel_vgpu_remove(struct mdev_device *mdev) static int intel_vgpu_iommu_notifier(struct notifier_block *nb, unsigned long action, void *data) { - struct intel_vgpu *vgpu = container_of(nb, - struct intel_vgpu, - vdev.iommu_notifier); + struct kvmgt_vdev *vdev = container_of(nb, + struct kvmgt_vdev, + iommu_notifier); + struct intel_vgpu *vgpu = vdev->vgpu; if (action == VFIO_IOMMU_NOTIFY_DMA_UNMAP) { struct vfio_iommu_type1_dma_unmap *unmap = data; @@ -708,7 +752,7 @@ static int intel_vgpu_iommu_notifier(struct notifier_block *nb, iov_pfn = unmap->iova >> PAGE_SHIFT; end_iov_pfn = iov_pfn + unmap->size / PAGE_SIZE; - mutex_lock(&vgpu->vdev.cache_lock); + mutex_lock(&vdev->cache_lock); for (; iov_pfn < end_iov_pfn; iov_pfn++) { entry = __gvt_cache_find_gfn(vgpu, iov_pfn); if (!entry) @@ -718,7 +762,7 @@ static int intel_vgpu_iommu_notifier(struct notifier_block *nb, entry->size); __gvt_cache_remove_entry(vgpu, entry); } - mutex_unlock(&vgpu->vdev.cache_lock); + mutex_unlock(&vdev->cache_lock); } return NOTIFY_OK; @@ -727,16 +771,16 @@ static int intel_vgpu_iommu_notifier(struct notifier_block *nb, static int intel_vgpu_group_notifier(struct notifier_block *nb, unsigned long action, void *data) { - struct intel_vgpu *vgpu = container_of(nb, - struct intel_vgpu, - vdev.group_notifier); + struct kvmgt_vdev *vdev = container_of(nb, + struct kvmgt_vdev, + group_notifier); /* the only action we care about */ if (action == VFIO_GROUP_NOTIFY_SET_KVM) { - vgpu->vdev.kvm = data; + vdev->kvm = data; if (!data) - schedule_work(&vgpu->vdev.release_work); + schedule_work(&vdev->release_work); } return NOTIFY_OK; @@ -745,15 +789,16 @@ static int intel_vgpu_group_notifier(struct notifier_block *nb, static int intel_vgpu_open(struct mdev_device *mdev) { struct intel_vgpu *vgpu = mdev_get_drvdata(mdev); + struct kvmgt_vdev *vdev = kvmgt_vdev(vgpu); unsigned long events; int ret; - vgpu->vdev.iommu_notifier.notifier_call = intel_vgpu_iommu_notifier; - vgpu->vdev.group_notifier.notifier_call = intel_vgpu_group_notifier; + vdev->iommu_notifier.notifier_call = intel_vgpu_iommu_notifier; + vdev->group_notifier.notifier_call = intel_vgpu_group_notifier; events = VFIO_IOMMU_NOTIFY_DMA_UNMAP; ret = vfio_register_notifier(mdev_dev(mdev), VFIO_IOMMU_NOTIFY, &events, - &vgpu->vdev.iommu_notifier); + &vdev->iommu_notifier); if (ret != 0) { gvt_vgpu_err("vfio_register_notifier for iommu failed: %d\n", ret); @@ -762,7 +807,7 @@ static int intel_vgpu_open(struct mdev_device *mdev) events = VFIO_GROUP_NOTIFY_SET_KVM; ret = vfio_register_notifier(mdev_dev(mdev), VFIO_GROUP_NOTIFY, &events, - &vgpu->vdev.group_notifier); + &vdev->group_notifier); if (ret != 0) { gvt_vgpu_err("vfio_register_notifier for group failed: %d\n", ret); @@ -781,51 +826,56 @@ static int intel_vgpu_open(struct mdev_device *mdev) intel_gvt_ops->vgpu_activate(vgpu); - atomic_set(&vgpu->vdev.released, 0); + atomic_set(&vdev->released, 0); return ret; undo_group: vfio_unregister_notifier(mdev_dev(mdev), VFIO_GROUP_NOTIFY, - &vgpu->vdev.group_notifier); + &vdev->group_notifier); undo_iommu: vfio_unregister_notifier(mdev_dev(mdev), VFIO_IOMMU_NOTIFY, - &vgpu->vdev.iommu_notifier); + &vdev->iommu_notifier); out: return ret; } static void intel_vgpu_release_msi_eventfd_ctx(struct intel_vgpu *vgpu) { + struct kvmgt_vdev *vdev = kvmgt_vdev(vgpu); struct eventfd_ctx *trigger; - trigger = vgpu->vdev.msi_trigger; + trigger = vdev->msi_trigger; if (trigger) { eventfd_ctx_put(trigger); - vgpu->vdev.msi_trigger = NULL; + vdev->msi_trigger = NULL; } } static void __intel_vgpu_release(struct intel_vgpu *vgpu) { + struct kvmgt_vdev *vdev = kvmgt_vdev(vgpu); + struct drm_i915_private *i915 = vgpu->gvt->gt->i915; struct kvmgt_guest_info *info; int ret; if (!handle_valid(vgpu->handle)) return; - if (atomic_cmpxchg(&vgpu->vdev.released, 0, 1)) + if (atomic_cmpxchg(&vdev->released, 0, 1)) return; intel_gvt_ops->vgpu_release(vgpu); - ret = vfio_unregister_notifier(mdev_dev(vgpu->vdev.mdev), VFIO_IOMMU_NOTIFY, - &vgpu->vdev.iommu_notifier); - WARN(ret, "vfio_unregister_notifier for iommu failed: %d\n", ret); + ret = vfio_unregister_notifier(mdev_dev(vdev->mdev), VFIO_IOMMU_NOTIFY, + &vdev->iommu_notifier); + drm_WARN(&i915->drm, ret, + "vfio_unregister_notifier for iommu failed: %d\n", ret); - ret = vfio_unregister_notifier(mdev_dev(vgpu->vdev.mdev), VFIO_GROUP_NOTIFY, - &vgpu->vdev.group_notifier); - WARN(ret, "vfio_unregister_notifier for group failed: %d\n", ret); + ret = vfio_unregister_notifier(mdev_dev(vdev->mdev), VFIO_GROUP_NOTIFY, + &vdev->group_notifier); + drm_WARN(&i915->drm, ret, + "vfio_unregister_notifier for group failed: %d\n", ret); /* dereference module reference taken at open */ module_put(THIS_MODULE); @@ -835,7 +885,7 @@ static void __intel_vgpu_release(struct intel_vgpu *vgpu) intel_vgpu_release_msi_eventfd_ctx(vgpu); - vgpu->vdev.kvm = NULL; + vdev->kvm = NULL; vgpu->handle = 0; } @@ -848,10 +898,10 @@ static void intel_vgpu_release(struct mdev_device *mdev) static void intel_vgpu_release_work(struct work_struct *work) { - struct intel_vgpu *vgpu = container_of(work, struct intel_vgpu, - vdev.release_work); + struct kvmgt_vdev *vdev = container_of(work, struct kvmgt_vdev, + release_work); - __intel_vgpu_release(vgpu); + __intel_vgpu_release(vdev->vgpu); } static u64 intel_vgpu_get_bar_addr(struct intel_vgpu *vgpu, int bar) @@ -913,7 +963,7 @@ static int intel_vgpu_aperture_rw(struct intel_vgpu *vgpu, u64 off, return -EINVAL; } - aperture_va = io_mapping_map_wc(&vgpu->gvt->dev_priv->ggtt.iomap, + aperture_va = io_mapping_map_wc(&vgpu->gvt->gt->ggtt->iomap, ALIGN_DOWN(off, PAGE_SIZE), count + offset_in_page(off)); if (!aperture_va) @@ -933,12 +983,13 @@ static ssize_t intel_vgpu_rw(struct mdev_device *mdev, char *buf, size_t count, loff_t *ppos, bool is_write) { struct intel_vgpu *vgpu = mdev_get_drvdata(mdev); + struct kvmgt_vdev *vdev = kvmgt_vdev(vgpu); unsigned int index = VFIO_PCI_OFFSET_TO_INDEX(*ppos); u64 pos = *ppos & VFIO_PCI_OFFSET_MASK; int ret = -EINVAL; - if (index >= VFIO_PCI_NUM_REGIONS + vgpu->vdev.num_regions) { + if (index >= VFIO_PCI_NUM_REGIONS + vdev->num_regions) { gvt_vgpu_err("invalid index: %u\n", index); return -EINVAL; } @@ -967,11 +1018,11 @@ static ssize_t intel_vgpu_rw(struct mdev_device *mdev, char *buf, case VFIO_PCI_ROM_REGION_INDEX: break; default: - if (index >= VFIO_PCI_NUM_REGIONS + vgpu->vdev.num_regions) + if (index >= VFIO_PCI_NUM_REGIONS + vdev->num_regions) return -EINVAL; index -= VFIO_PCI_NUM_REGIONS; - return vgpu->vdev.region[index].ops->rw(vgpu, buf, count, + return vdev->region[index].ops->rw(vgpu, buf, count, ppos, is_write); } @@ -1224,7 +1275,7 @@ static int intel_vgpu_set_msi_trigger(struct intel_vgpu *vgpu, gvt_vgpu_err("eventfd_ctx_fdget failed\n"); return PTR_ERR(trigger); } - vgpu->vdev.msi_trigger = trigger; + kvmgt_vdev(vgpu)->msi_trigger = trigger; } else if ((flags & VFIO_IRQ_SET_DATA_NONE) && !count) intel_vgpu_release_msi_eventfd_ctx(vgpu); @@ -1276,6 +1327,7 @@ static long intel_vgpu_ioctl(struct mdev_device *mdev, unsigned int cmd, unsigned long arg) { struct intel_vgpu *vgpu = mdev_get_drvdata(mdev); + struct kvmgt_vdev *vdev = kvmgt_vdev(vgpu); unsigned long minsz; gvt_dbg_core("vgpu%d ioctl, cmd: %d\n", vgpu->id, cmd); @@ -1294,7 +1346,7 @@ static long intel_vgpu_ioctl(struct mdev_device *mdev, unsigned int cmd, info.flags = VFIO_DEVICE_FLAGS_PCI; info.flags |= VFIO_DEVICE_FLAGS_RESET; info.num_regions = VFIO_PCI_NUM_REGIONS + - vgpu->vdev.num_regions; + vdev->num_regions; info.num_irqs = VFIO_PCI_NUM_IRQS; return copy_to_user((void __user *)arg, &info, minsz) ? @@ -1385,22 +1437,22 @@ static long intel_vgpu_ioctl(struct mdev_device *mdev, unsigned int cmd, .header.version = 1 }; if (info.index >= VFIO_PCI_NUM_REGIONS + - vgpu->vdev.num_regions) + vdev->num_regions) return -EINVAL; info.index = array_index_nospec(info.index, VFIO_PCI_NUM_REGIONS + - vgpu->vdev.num_regions); + vdev->num_regions); i = info.index - VFIO_PCI_NUM_REGIONS; info.offset = VFIO_PCI_INDEX_TO_OFFSET(info.index); - info.size = vgpu->vdev.region[i].size; - info.flags = vgpu->vdev.region[i].flags; + info.size = vdev->region[i].size; + info.flags = vdev->region[i].flags; - cap_type.type = vgpu->vdev.region[i].type; - cap_type.subtype = vgpu->vdev.region[i].subtype; + cap_type.type = vdev->region[i].type; + cap_type.subtype = vdev->region[i].subtype; ret = vfio_info_add_capability(&caps, &cap_type.header, @@ -1597,12 +1649,10 @@ static struct mdev_parent_ops intel_vgpu_ops = { static int kvmgt_host_init(struct device *dev, void *gvt, const void *ops) { - struct attribute **kvm_type_attrs; struct attribute_group **kvm_vgpu_type_groups; intel_gvt_ops = ops; - if (!intel_gvt_ops->get_gvt_attrs(&kvm_type_attrs, - &kvm_vgpu_type_groups)) + if (!intel_gvt_ops->get_gvt_attrs(&kvm_vgpu_type_groups)) return -EFAULT; intel_vgpu_ops.supported_type_groups = kvm_vgpu_type_groups; @@ -1742,13 +1792,15 @@ static int kvmgt_guest_init(struct mdev_device *mdev) { struct kvmgt_guest_info *info; struct intel_vgpu *vgpu; + struct kvmgt_vdev *vdev; struct kvm *kvm; vgpu = mdev_get_drvdata(mdev); if (handle_valid(vgpu->handle)) return -EEXIST; - kvm = vgpu->vdev.kvm; + vdev = kvmgt_vdev(vgpu); + kvm = vdev->kvm; if (!kvm || kvm->mm != current->mm) { gvt_vgpu_err("KVM is required to use Intel vGPU\n"); return -ESRCH; @@ -1769,8 +1821,6 @@ static int kvmgt_guest_init(struct mdev_device *mdev) kvmgt_protect_table_init(info); gvt_cache_init(vgpu); - init_completion(&vgpu->vblank_done); - info->track_node.track_write = kvmgt_page_track_write; info->track_node.track_flush_slot = kvmgt_page_track_flush_slot; kvm_page_track_register_notifier(kvm, &info->track_node); @@ -1778,7 +1828,7 @@ static int kvmgt_guest_init(struct mdev_device *mdev) info->debugfs_cache_entries = debugfs_create_ulong( "kvmgt_nr_cache_entries", 0444, vgpu->debugfs, - &vgpu->vdev.nr_cache_entries); + &vdev->nr_cache_entries); return 0; } @@ -1795,9 +1845,17 @@ static bool kvmgt_guest_exit(struct kvmgt_guest_info *info) return true; } -static int kvmgt_attach_vgpu(void *vgpu, unsigned long *handle) +static int kvmgt_attach_vgpu(void *p_vgpu, unsigned long *handle) { - /* nothing to do here */ + struct intel_vgpu *vgpu = (struct intel_vgpu *)p_vgpu; + + vgpu->vdev = kzalloc(sizeof(struct kvmgt_vdev), GFP_KERNEL); + + if (!vgpu->vdev) + return -ENOMEM; + + kvmgt_vdev(vgpu)->vgpu = vgpu; + return 0; } @@ -1805,29 +1863,34 @@ static void kvmgt_detach_vgpu(void *p_vgpu) { int i; struct intel_vgpu *vgpu = (struct intel_vgpu *)p_vgpu; + struct kvmgt_vdev *vdev = kvmgt_vdev(vgpu); - if (!vgpu->vdev.region) + if (!vdev->region) return; - for (i = 0; i < vgpu->vdev.num_regions; i++) - if (vgpu->vdev.region[i].ops->release) - vgpu->vdev.region[i].ops->release(vgpu, - &vgpu->vdev.region[i]); - vgpu->vdev.num_regions = 0; - kfree(vgpu->vdev.region); - vgpu->vdev.region = NULL; + for (i = 0; i < vdev->num_regions; i++) + if (vdev->region[i].ops->release) + vdev->region[i].ops->release(vgpu, + &vdev->region[i]); + vdev->num_regions = 0; + kfree(vdev->region); + vdev->region = NULL; + + kfree(vdev); } static int kvmgt_inject_msi(unsigned long handle, u32 addr, u16 data) { struct kvmgt_guest_info *info; struct intel_vgpu *vgpu; + struct kvmgt_vdev *vdev; if (!handle_valid(handle)) return -ESRCH; info = (struct kvmgt_guest_info *)handle; vgpu = info->vgpu; + vdev = kvmgt_vdev(vgpu); /* * When guest is poweroff, msi_trigger is set to NULL, but vgpu's @@ -1838,10 +1901,10 @@ static int kvmgt_inject_msi(unsigned long handle, u32 addr, u16 data) * enabled by guest. so if msi_trigger is null, success is still * returned and don't inject interrupt into guest. */ - if (vgpu->vdev.msi_trigger == NULL) + if (vdev->msi_trigger == NULL) return 0; - if (eventfd_signal(vgpu->vdev.msi_trigger, 1) == 1) + if (eventfd_signal(vdev->msi_trigger, 1) == 1) return 0; return -EFAULT; @@ -1867,26 +1930,26 @@ static unsigned long kvmgt_gfn_to_pfn(unsigned long handle, unsigned long gfn) static int kvmgt_dma_map_guest_page(unsigned long handle, unsigned long gfn, unsigned long size, dma_addr_t *dma_addr) { - struct kvmgt_guest_info *info; struct intel_vgpu *vgpu; + struct kvmgt_vdev *vdev; struct gvt_dma *entry; int ret; if (!handle_valid(handle)) return -EINVAL; - info = (struct kvmgt_guest_info *)handle; - vgpu = info->vgpu; + vgpu = ((struct kvmgt_guest_info *)handle)->vgpu; + vdev = kvmgt_vdev(vgpu); - mutex_lock(&info->vgpu->vdev.cache_lock); + mutex_lock(&vdev->cache_lock); - entry = __gvt_cache_find_gfn(info->vgpu, gfn); + entry = __gvt_cache_find_gfn(vgpu, gfn); if (!entry) { ret = gvt_dma_map_page(vgpu, gfn, dma_addr, size); if (ret) goto err_unlock; - ret = __gvt_cache_add(info->vgpu, gfn, *dma_addr, size); + ret = __gvt_cache_add(vgpu, gfn, *dma_addr, size); if (ret) goto err_unmap; } else if (entry->size != size) { @@ -1898,7 +1961,7 @@ static int kvmgt_dma_map_guest_page(unsigned long handle, unsigned long gfn, if (ret) goto err_unlock; - ret = __gvt_cache_add(info->vgpu, gfn, *dma_addr, size); + ret = __gvt_cache_add(vgpu, gfn, *dma_addr, size); if (ret) goto err_unmap; } else { @@ -1906,19 +1969,20 @@ static int kvmgt_dma_map_guest_page(unsigned long handle, unsigned long gfn, *dma_addr = entry->dma_addr; } - mutex_unlock(&info->vgpu->vdev.cache_lock); + mutex_unlock(&vdev->cache_lock); return 0; err_unmap: gvt_dma_unmap_page(vgpu, gfn, *dma_addr, size); err_unlock: - mutex_unlock(&info->vgpu->vdev.cache_lock); + mutex_unlock(&vdev->cache_lock); return ret; } static int kvmgt_dma_pin_guest_page(unsigned long handle, dma_addr_t dma_addr) { struct kvmgt_guest_info *info; + struct kvmgt_vdev *vdev; struct gvt_dma *entry; int ret = 0; @@ -1926,14 +1990,15 @@ static int kvmgt_dma_pin_guest_page(unsigned long handle, dma_addr_t dma_addr) return -ENODEV; info = (struct kvmgt_guest_info *)handle; + vdev = kvmgt_vdev(info->vgpu); - mutex_lock(&info->vgpu->vdev.cache_lock); + mutex_lock(&vdev->cache_lock); entry = __gvt_cache_find_dma_addr(info->vgpu, dma_addr); if (entry) kref_get(&entry->ref); else ret = -ENOMEM; - mutex_unlock(&info->vgpu->vdev.cache_lock); + mutex_unlock(&vdev->cache_lock); return ret; } @@ -1949,19 +2014,21 @@ static void __gvt_dma_release(struct kref *ref) static void kvmgt_dma_unmap_guest_page(unsigned long handle, dma_addr_t dma_addr) { - struct kvmgt_guest_info *info; + struct intel_vgpu *vgpu; + struct kvmgt_vdev *vdev; struct gvt_dma *entry; if (!handle_valid(handle)) return; - info = (struct kvmgt_guest_info *)handle; + vgpu = ((struct kvmgt_guest_info *)handle)->vgpu; + vdev = kvmgt_vdev(vgpu); - mutex_lock(&info->vgpu->vdev.cache_lock); - entry = __gvt_cache_find_dma_addr(info->vgpu, dma_addr); + mutex_lock(&vdev->cache_lock); + entry = __gvt_cache_find_dma_addr(vgpu, dma_addr); if (entry) kref_put(&entry->ref, __gvt_dma_release); - mutex_unlock(&info->vgpu->vdev.cache_lock); + mutex_unlock(&vdev->cache_lock); } static int kvmgt_rw_gpa(unsigned long handle, unsigned long gpa, diff --git a/drivers/gpu/drm/i915/gvt/mmio.c b/drivers/gpu/drm/i915/gvt/mmio.c index a55178884d67..291993615af9 100644 --- a/drivers/gpu/drm/i915/gvt/mmio.c +++ b/drivers/gpu/drm/i915/gvt/mmio.c @@ -103,6 +103,7 @@ int intel_vgpu_emulate_mmio_read(struct intel_vgpu *vgpu, u64 pa, void *p_data, unsigned int bytes) { struct intel_gvt *gvt = vgpu->gvt; + struct drm_i915_private *i915 = gvt->gt->i915; unsigned int offset = 0; int ret = -EINVAL; @@ -114,15 +115,17 @@ int intel_vgpu_emulate_mmio_read(struct intel_vgpu *vgpu, u64 pa, offset = intel_vgpu_gpa_to_mmio_offset(vgpu, pa); - if (WARN_ON(bytes > 8)) + if (drm_WARN_ON(&i915->drm, bytes > 8)) goto err; if (reg_is_gtt(gvt, offset)) { - if (WARN_ON(!IS_ALIGNED(offset, 4) && !IS_ALIGNED(offset, 8))) + if (drm_WARN_ON(&i915->drm, !IS_ALIGNED(offset, 4) && + !IS_ALIGNED(offset, 8))) goto err; - if (WARN_ON(bytes != 4 && bytes != 8)) + if (drm_WARN_ON(&i915->drm, bytes != 4 && bytes != 8)) goto err; - if (WARN_ON(!reg_is_gtt(gvt, offset + bytes - 1))) + if (drm_WARN_ON(&i915->drm, + !reg_is_gtt(gvt, offset + bytes - 1))) goto err; ret = intel_vgpu_emulate_ggtt_mmio_read(vgpu, offset, @@ -132,16 +135,16 @@ int intel_vgpu_emulate_mmio_read(struct intel_vgpu *vgpu, u64 pa, goto out; } - if (WARN_ON_ONCE(!reg_is_mmio(gvt, offset))) { + if (drm_WARN_ON_ONCE(&i915->drm, !reg_is_mmio(gvt, offset))) { ret = intel_gvt_hypervisor_read_gpa(vgpu, pa, p_data, bytes); goto out; } - if (WARN_ON(!reg_is_mmio(gvt, offset + bytes - 1))) + if (drm_WARN_ON(&i915->drm, !reg_is_mmio(gvt, offset + bytes - 1))) goto err; if (!intel_gvt_mmio_is_unalign(gvt, offset)) { - if (WARN_ON(!IS_ALIGNED(offset, bytes))) + if (drm_WARN_ON(&i915->drm, !IS_ALIGNED(offset, bytes))) goto err; } @@ -175,6 +178,7 @@ int intel_vgpu_emulate_mmio_write(struct intel_vgpu *vgpu, u64 pa, void *p_data, unsigned int bytes) { struct intel_gvt *gvt = vgpu->gvt; + struct drm_i915_private *i915 = gvt->gt->i915; unsigned int offset = 0; int ret = -EINVAL; @@ -187,15 +191,17 @@ int intel_vgpu_emulate_mmio_write(struct intel_vgpu *vgpu, u64 pa, offset = intel_vgpu_gpa_to_mmio_offset(vgpu, pa); - if (WARN_ON(bytes > 8)) + if (drm_WARN_ON(&i915->drm, bytes > 8)) goto err; if (reg_is_gtt(gvt, offset)) { - if (WARN_ON(!IS_ALIGNED(offset, 4) && !IS_ALIGNED(offset, 8))) + if (drm_WARN_ON(&i915->drm, !IS_ALIGNED(offset, 4) && + !IS_ALIGNED(offset, 8))) goto err; - if (WARN_ON(bytes != 4 && bytes != 8)) + if (drm_WARN_ON(&i915->drm, bytes != 4 && bytes != 8)) goto err; - if (WARN_ON(!reg_is_gtt(gvt, offset + bytes - 1))) + if (drm_WARN_ON(&i915->drm, + !reg_is_gtt(gvt, offset + bytes - 1))) goto err; ret = intel_vgpu_emulate_ggtt_mmio_write(vgpu, offset, @@ -205,7 +211,7 @@ int intel_vgpu_emulate_mmio_write(struct intel_vgpu *vgpu, u64 pa, goto out; } - if (WARN_ON_ONCE(!reg_is_mmio(gvt, offset))) { + if (drm_WARN_ON_ONCE(&i915->drm, !reg_is_mmio(gvt, offset))) { ret = intel_gvt_hypervisor_write_gpa(vgpu, pa, p_data, bytes); goto out; } @@ -245,7 +251,7 @@ void intel_vgpu_reset_mmio(struct intel_vgpu *vgpu, bool dmlr) /* set the bit 0:2(Core C-State ) to C0 */ vgpu_vreg_t(vgpu, GEN6_GT_CORE_STATUS) = 0; - if (IS_BROXTON(vgpu->gvt->dev_priv)) { + if (IS_BROXTON(vgpu->gvt->gt->i915)) { vgpu_vreg_t(vgpu, BXT_P_CR_GT_DISP_PWRON) &= ~(BIT(0) | BIT(1)); vgpu_vreg_t(vgpu, BXT_PORT_CL1CM_DW0(DPIO_PHY0)) &= diff --git a/drivers/gpu/drm/i915/gvt/mmio.h b/drivers/gpu/drm/i915/gvt/mmio.h index 2e68f4b02c94..cc4812648bf4 100644 --- a/drivers/gpu/drm/i915/gvt/mmio.h +++ b/drivers/gpu/drm/i915/gvt/mmio.h @@ -69,8 +69,8 @@ struct intel_gvt_mmio_info { struct hlist_node node; }; -int intel_gvt_render_mmio_to_ring_id(struct intel_gvt *gvt, - unsigned int reg); +const struct intel_engine_cs * +intel_gvt_render_mmio_to_engine(struct intel_gvt *gvt, unsigned int reg); unsigned long intel_gvt_get_device_type(struct intel_gvt *gvt); bool intel_gvt_match_device(struct intel_gvt *gvt, unsigned long device); diff --git a/drivers/gpu/drm/i915/gvt/mmio_context.c b/drivers/gpu/drm/i915/gvt/mmio_context.c index aaf15916d29a..2ccaf78f96e8 100644 --- a/drivers/gpu/drm/i915/gvt/mmio_context.c +++ b/drivers/gpu/drm/i915/gvt/mmio_context.c @@ -157,12 +157,13 @@ static u32 gen9_mocs_mmio_offset_list[] = { [VECS0] = 0xcb00, }; -static void load_render_mocs(struct drm_i915_private *dev_priv) +static void load_render_mocs(const struct intel_engine_cs *engine) { - struct intel_gvt *gvt = dev_priv->gvt; - i915_reg_t offset; + struct intel_gvt *gvt = engine->i915->gvt; + struct intel_uncore *uncore = engine->uncore; u32 cnt = gvt->engine_mmio_list.mocs_mmio_offset_list_cnt; u32 *regs = gvt->engine_mmio_list.mocs_mmio_offset_list; + i915_reg_t offset; int ring_id, i; /* Platform doesn't have mocs mmios. */ @@ -170,12 +171,13 @@ static void load_render_mocs(struct drm_i915_private *dev_priv) return; for (ring_id = 0; ring_id < cnt; ring_id++) { - if (!HAS_ENGINE(dev_priv, ring_id)) + if (!HAS_ENGINE(engine->i915, ring_id)) continue; + offset.reg = regs[ring_id]; for (i = 0; i < GEN9_MOCS_SIZE; i++) { gen9_render_mocs.control_table[ring_id][i] = - I915_READ_FW(offset); + intel_uncore_read_fw(uncore, offset); offset.reg += 4; } } @@ -183,7 +185,7 @@ static void load_render_mocs(struct drm_i915_private *dev_priv) offset.reg = 0xb020; for (i = 0; i < GEN9_MOCS_SIZE / 2; i++) { gen9_render_mocs.l3cc_table[i] = - I915_READ_FW(offset); + intel_uncore_read_fw(uncore, offset); offset.reg += 4; } gen9_render_mocs.initialized = true; @@ -214,13 +216,11 @@ restore_context_mmio_for_inhibit(struct intel_vgpu *vgpu, *cs++ = MI_LOAD_REGISTER_IMM(count); for (mmio = gvt->engine_mmio_list.mmio; i915_mmio_reg_valid(mmio->reg); mmio++) { - if (mmio->ring_id != ring_id || - !mmio->in_context) + if (mmio->id != ring_id || !mmio->in_context) continue; *cs++ = i915_mmio_reg_offset(mmio->reg); - *cs++ = vgpu_vreg_t(vgpu, mmio->reg) | - (mmio->mask << 16); + *cs++ = vgpu_vreg_t(vgpu, mmio->reg) | (mmio->mask << 16); gvt_dbg_core("add lri reg pair 0x%x:0x%x in inhibit ctx, vgpu:%d, rind_id:%d\n", *(cs-2), *(cs-1), vgpu->id, ring_id); } @@ -344,10 +344,10 @@ static u32 gen8_tlb_mmio_offset_list[] = { [VECS0] = 0x4270, }; -static void handle_tlb_pending_event(struct intel_vgpu *vgpu, int ring_id) +static void handle_tlb_pending_event(struct intel_vgpu *vgpu, + const struct intel_engine_cs *engine) { - struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv; - struct intel_uncore *uncore = &dev_priv->uncore; + struct intel_uncore *uncore = engine->uncore; struct intel_vgpu_submission *s = &vgpu->submission; u32 *regs = vgpu->gvt->engine_mmio_list.tlb_mmio_offset_list; u32 cnt = vgpu->gvt->engine_mmio_list.tlb_mmio_offset_list_cnt; @@ -357,13 +357,13 @@ static void handle_tlb_pending_event(struct intel_vgpu *vgpu, int ring_id) if (!regs) return; - if (WARN_ON(ring_id >= cnt)) + if (drm_WARN_ON(&engine->i915->drm, engine->id >= cnt)) return; - if (!test_and_clear_bit(ring_id, (void *)s->tlb_handle_pending)) + if (!test_and_clear_bit(engine->id, (void *)s->tlb_handle_pending)) return; - reg = _MMIO(regs[ring_id]); + reg = _MMIO(regs[engine->id]); /* WaForceWakeRenderDuringMmioTLBInvalidate:skl * we need to put a forcewake when invalidating RCS TLB caches, @@ -372,30 +372,27 @@ static void handle_tlb_pending_event(struct intel_vgpu *vgpu, int ring_id) */ fw = intel_uncore_forcewake_for_reg(uncore, reg, FW_REG_READ | FW_REG_WRITE); - if (ring_id == RCS0 && INTEL_GEN(dev_priv) >= 9) + if (engine->id == RCS0 && INTEL_GEN(engine->i915) >= 9) fw |= FORCEWAKE_RENDER; intel_uncore_forcewake_get(uncore, fw); intel_uncore_write_fw(uncore, reg, 0x1); - if (wait_for_atomic((intel_uncore_read_fw(uncore, reg) == 0), 50)) - gvt_vgpu_err("timeout in invalidate ring (%d) tlb\n", ring_id); + if (wait_for_atomic(intel_uncore_read_fw(uncore, reg) == 0, 50)) + gvt_vgpu_err("timeout in invalidate ring %s tlb\n", + engine->name); else vgpu_vreg_t(vgpu, reg) = 0; intel_uncore_forcewake_put(uncore, fw); - gvt_dbg_core("invalidate TLB for ring %d\n", ring_id); + gvt_dbg_core("invalidate TLB for ring %s\n", engine->name); } static void switch_mocs(struct intel_vgpu *pre, struct intel_vgpu *next, - int ring_id) + const struct intel_engine_cs *engine) { - struct drm_i915_private *dev_priv; - i915_reg_t offset, l3_offset; - u32 old_v, new_v; - u32 regs[] = { [RCS0] = 0xc800, [VCS0] = 0xc900, @@ -403,36 +400,38 @@ static void switch_mocs(struct intel_vgpu *pre, struct intel_vgpu *next, [BCS0] = 0xcc00, [VECS0] = 0xcb00, }; + struct intel_uncore *uncore = engine->uncore; + i915_reg_t offset, l3_offset; + u32 old_v, new_v; int i; - dev_priv = pre ? pre->gvt->dev_priv : next->gvt->dev_priv; - if (WARN_ON(ring_id >= ARRAY_SIZE(regs))) + if (drm_WARN_ON(&engine->i915->drm, engine->id >= ARRAY_SIZE(regs))) return; - if (ring_id == RCS0 && IS_GEN(dev_priv, 9)) + if (engine->id == RCS0 && IS_GEN(engine->i915, 9)) return; if (!pre && !gen9_render_mocs.initialized) - load_render_mocs(dev_priv); + load_render_mocs(engine); - offset.reg = regs[ring_id]; + offset.reg = regs[engine->id]; for (i = 0; i < GEN9_MOCS_SIZE; i++) { if (pre) old_v = vgpu_vreg_t(pre, offset); else - old_v = gen9_render_mocs.control_table[ring_id][i]; + old_v = gen9_render_mocs.control_table[engine->id][i]; if (next) new_v = vgpu_vreg_t(next, offset); else - new_v = gen9_render_mocs.control_table[ring_id][i]; + new_v = gen9_render_mocs.control_table[engine->id][i]; if (old_v != new_v) - I915_WRITE_FW(offset, new_v); + intel_uncore_write_fw(uncore, offset, new_v); offset.reg += 4; } - if (ring_id == RCS0) { + if (engine->id == RCS0) { l3_offset.reg = 0xb020; for (i = 0; i < GEN9_MOCS_SIZE / 2; i++) { if (pre) @@ -445,7 +444,7 @@ static void switch_mocs(struct intel_vgpu *pre, struct intel_vgpu *next, new_v = gen9_render_mocs.l3cc_table[i]; if (old_v != new_v) - I915_WRITE_FW(l3_offset, new_v); + intel_uncore_write_fw(uncore, l3_offset, new_v); l3_offset.reg += 4; } @@ -467,38 +466,40 @@ bool is_inhibit_context(struct intel_context *ce) /* Switch ring mmio values (context). */ static void switch_mmio(struct intel_vgpu *pre, struct intel_vgpu *next, - int ring_id) + const struct intel_engine_cs *engine) { - struct drm_i915_private *dev_priv; + struct intel_uncore *uncore = engine->uncore; struct intel_vgpu_submission *s; struct engine_mmio *mmio; u32 old_v, new_v; - dev_priv = pre ? pre->gvt->dev_priv : next->gvt->dev_priv; - if (INTEL_GEN(dev_priv) >= 9) - switch_mocs(pre, next, ring_id); + if (INTEL_GEN(engine->i915) >= 9) + switch_mocs(pre, next, engine); - for (mmio = dev_priv->gvt->engine_mmio_list.mmio; + for (mmio = engine->i915->gvt->engine_mmio_list.mmio; i915_mmio_reg_valid(mmio->reg); mmio++) { - if (mmio->ring_id != ring_id) + if (mmio->id != engine->id) continue; /* * No need to do save or restore of the mmio which is in context * state image on gen9, it's initialized by lri command and * save or restore with context together. */ - if (IS_GEN(dev_priv, 9) && mmio->in_context) + if (IS_GEN(engine->i915, 9) && mmio->in_context) continue; // save if (pre) { - vgpu_vreg_t(pre, mmio->reg) = I915_READ_FW(mmio->reg); + vgpu_vreg_t(pre, mmio->reg) = + intel_uncore_read_fw(uncore, mmio->reg); if (mmio->mask) vgpu_vreg_t(pre, mmio->reg) &= - ~(mmio->mask << 16); + ~(mmio->mask << 16); old_v = vgpu_vreg_t(pre, mmio->reg); - } else - old_v = mmio->value = I915_READ_FW(mmio->reg); + } else { + old_v = mmio->value = + intel_uncore_read_fw(uncore, mmio->reg); + } // restore if (next) { @@ -509,12 +510,12 @@ static void switch_mmio(struct intel_vgpu *pre, * itself. */ if (mmio->in_context && - !is_inhibit_context(s->shadow[ring_id])) + !is_inhibit_context(s->shadow[engine->id])) continue; if (mmio->mask) new_v = vgpu_vreg_t(next, mmio->reg) | - (mmio->mask << 16); + (mmio->mask << 16); else new_v = vgpu_vreg_t(next, mmio->reg); } else { @@ -526,7 +527,7 @@ static void switch_mmio(struct intel_vgpu *pre, new_v = mmio->value; } - I915_WRITE_FW(mmio->reg, new_v); + intel_uncore_write_fw(uncore, mmio->reg, new_v); trace_render_mmio(pre ? pre->id : 0, next ? next->id : 0, @@ -536,39 +537,37 @@ static void switch_mmio(struct intel_vgpu *pre, } if (next) - handle_tlb_pending_event(next, ring_id); + handle_tlb_pending_event(next, engine); } /** * intel_gvt_switch_render_mmio - switch mmio context of specific engine * @pre: the last vGPU that own the engine * @next: the vGPU to switch to - * @ring_id: specify the engine + * @engine: the engine * * If pre is null indicates that host own the engine. If next is null * indicates that we are switching to host workload. */ void intel_gvt_switch_mmio(struct intel_vgpu *pre, - struct intel_vgpu *next, int ring_id) + struct intel_vgpu *next, + const struct intel_engine_cs *engine) { - struct drm_i915_private *dev_priv; - - if (WARN_ON(!pre && !next)) + if (WARN(!pre && !next, "switch ring %s from host to HOST\n", + engine->name)) return; - gvt_dbg_render("switch ring %d from %s to %s\n", ring_id, + gvt_dbg_render("switch ring %s from %s to %s\n", engine->name, pre ? "vGPU" : "host", next ? "vGPU" : "HOST"); - dev_priv = pre ? pre->gvt->dev_priv : next->gvt->dev_priv; - /** * We are using raw mmio access wrapper to improve the * performace for batch mmio read/write, so we need * handle forcewake mannually. */ - intel_uncore_forcewake_get(&dev_priv->uncore, FORCEWAKE_ALL); - switch_mmio(pre, next, ring_id); - intel_uncore_forcewake_put(&dev_priv->uncore, FORCEWAKE_ALL); + intel_uncore_forcewake_get(engine->uncore, FORCEWAKE_ALL); + switch_mmio(pre, next, engine); + intel_uncore_forcewake_put(engine->uncore, FORCEWAKE_ALL); } /** @@ -580,7 +579,7 @@ void intel_gvt_init_engine_mmio_context(struct intel_gvt *gvt) { struct engine_mmio *mmio; - if (INTEL_GEN(gvt->dev_priv) >= 9) { + if (INTEL_GEN(gvt->gt->i915) >= 9) { gvt->engine_mmio_list.mmio = gen9_engine_mmio_list; gvt->engine_mmio_list.tlb_mmio_offset_list = gen8_tlb_mmio_offset_list; gvt->engine_mmio_list.tlb_mmio_offset_list_cnt = ARRAY_SIZE(gen8_tlb_mmio_offset_list); @@ -595,7 +594,7 @@ void intel_gvt_init_engine_mmio_context(struct intel_gvt *gvt) for (mmio = gvt->engine_mmio_list.mmio; i915_mmio_reg_valid(mmio->reg); mmio++) { if (mmio->in_context) { - gvt->engine_mmio_list.ctx_mmio_count[mmio->ring_id]++; + gvt->engine_mmio_list.ctx_mmio_count[mmio->id]++; intel_gvt_mmio_set_in_ctx(gvt, mmio->reg.reg); } } diff --git a/drivers/gpu/drm/i915/gvt/mmio_context.h b/drivers/gpu/drm/i915/gvt/mmio_context.h index f7eaa442403f..970704b18f23 100644 --- a/drivers/gpu/drm/i915/gvt/mmio_context.h +++ b/drivers/gpu/drm/i915/gvt/mmio_context.h @@ -37,7 +37,7 @@ #define __GVT_RENDER_H__ struct engine_mmio { - int ring_id; + enum intel_engine_id id; i915_reg_t reg; u32 mask; bool in_context; @@ -45,7 +45,8 @@ struct engine_mmio { }; void intel_gvt_switch_mmio(struct intel_vgpu *pre, - struct intel_vgpu *next, int ring_id); + struct intel_vgpu *next, + const struct intel_engine_cs *engine); void intel_gvt_init_engine_mmio_context(struct intel_gvt *gvt); diff --git a/drivers/gpu/drm/i915/gvt/opregion.c b/drivers/gpu/drm/i915/gvt/opregion.c index 867e7629025b..33569b910ed5 100644 --- a/drivers/gpu/drm/i915/gvt/opregion.c +++ b/drivers/gpu/drm/i915/gvt/opregion.c @@ -147,15 +147,14 @@ static void virt_vbt_generation(struct vbt *v) /* there's features depending on version! */ v->header.version = 155; v->header.header_size = sizeof(v->header); - v->header.vbt_size = sizeof(struct vbt) - sizeof(v->header); + v->header.vbt_size = sizeof(struct vbt); v->header.bdb_offset = offsetof(struct vbt, bdb_header); strcpy(&v->bdb_header.signature[0], "BIOS_DATA_BLOCK"); v->bdb_header.version = 186; /* child_dev_size = 33 */ v->bdb_header.header_size = sizeof(v->bdb_header); - v->bdb_header.bdb_size = sizeof(struct vbt) - sizeof(struct vbt_header) - - sizeof(struct bdb_header); + v->bdb_header.bdb_size = sizeof(struct vbt) - sizeof(struct vbt_header); /* general features */ v->general_features_header.id = BDB_GENERAL_FEATURES; diff --git a/drivers/gpu/drm/i915/gvt/sched_policy.c b/drivers/gpu/drm/i915/gvt/sched_policy.c index 2369d4a9af94..036b74fe9298 100644 --- a/drivers/gpu/drm/i915/gvt/sched_policy.c +++ b/drivers/gpu/drm/i915/gvt/sched_policy.c @@ -39,8 +39,8 @@ static bool vgpu_has_pending_workload(struct intel_vgpu *vgpu) enum intel_engine_id i; struct intel_engine_cs *engine; - for_each_engine(engine, vgpu->gvt->dev_priv, i) { - if (!list_empty(workload_q_head(vgpu, i))) + for_each_engine(engine, vgpu->gvt->gt, i) { + if (!list_empty(workload_q_head(vgpu, engine))) return true; } @@ -152,8 +152,8 @@ static void try_to_schedule_next_vgpu(struct intel_gvt *gvt) scheduler->need_reschedule = true; /* still have uncompleted workload? */ - for_each_engine(engine, gvt->dev_priv, i) { - if (scheduler->current_workload[i]) + for_each_engine(engine, gvt->gt, i) { + if (scheduler->current_workload[engine->id]) return; } @@ -169,8 +169,8 @@ static void try_to_schedule_next_vgpu(struct intel_gvt *gvt) scheduler->need_reschedule = false; /* wake up workload dispatch thread */ - for_each_engine(engine, gvt->dev_priv, i) - wake_up(&scheduler->waitq[i]); + for_each_engine(engine, gvt->gt, i) + wake_up(&scheduler->waitq[engine->id]); } static struct intel_vgpu *find_busy_vgpu(struct gvt_sched_data *sched_data) @@ -444,9 +444,10 @@ void intel_vgpu_stop_schedule(struct intel_vgpu *vgpu) { struct intel_gvt_workload_scheduler *scheduler = &vgpu->gvt->scheduler; - int ring_id; struct vgpu_sched_data *vgpu_data = vgpu->sched_data; - struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv; + struct drm_i915_private *dev_priv = vgpu->gvt->gt->i915; + struct intel_engine_cs *engine; + enum intel_engine_id id; if (!vgpu_data->active) return; @@ -467,10 +468,10 @@ void intel_vgpu_stop_schedule(struct intel_vgpu *vgpu) intel_runtime_pm_get(&dev_priv->runtime_pm); spin_lock_bh(&scheduler->mmio_context_lock); - for (ring_id = 0; ring_id < I915_NUM_ENGINES; ring_id++) { - if (scheduler->engine_owner[ring_id] == vgpu) { - intel_gvt_switch_mmio(vgpu, NULL, ring_id); - scheduler->engine_owner[ring_id] = NULL; + for_each_engine(engine, vgpu->gvt->gt, id) { + if (scheduler->engine_owner[engine->id] == vgpu) { + intel_gvt_switch_mmio(vgpu, NULL, engine); + scheduler->engine_owner[engine->id] = NULL; } } spin_unlock_bh(&scheduler->mmio_context_lock); diff --git a/drivers/gpu/drm/i915/gvt/scheduler.c b/drivers/gpu/drm/i915/gvt/scheduler.c index e8c0885df978..1c95bf8cbed0 100644 --- a/drivers/gpu/drm/i915/gvt/scheduler.c +++ b/drivers/gpu/drm/i915/gvt/scheduler.c @@ -84,7 +84,7 @@ static void update_shadow_pdps(struct intel_vgpu_workload *workload) static void sr_oa_regs(struct intel_vgpu_workload *workload, u32 *reg_state, bool save) { - struct drm_i915_private *dev_priv = workload->vgpu->gvt->dev_priv; + struct drm_i915_private *dev_priv = workload->vgpu->gvt->gt->i915; u32 ctx_oactxctrl = dev_priv->perf.ctx_oactxctrl_offset; u32 ctx_flexeu0 = dev_priv->perf.ctx_flexeu0_offset; int i = 0; @@ -98,7 +98,7 @@ static void sr_oa_regs(struct intel_vgpu_workload *workload, i915_mmio_reg_offset(EU_PERF_CNTL6), }; - if (workload->ring_id != RCS0) + if (workload->engine->id != RCS0) return; if (save) { @@ -128,7 +128,6 @@ static int populate_shadow_context(struct intel_vgpu_workload *workload) { struct intel_vgpu *vgpu = workload->vgpu; struct intel_gvt *gvt = vgpu->gvt; - int ring_id = workload->ring_id; struct drm_i915_gem_object *ctx_obj = workload->req->context->state->obj; struct execlist_ring_context *shadow_ring_context; @@ -154,7 +153,7 @@ static int populate_shadow_context(struct intel_vgpu_workload *workload) COPY_REG_MASKED(ctx_ctrl); COPY_REG(ctx_timestamp); - if (ring_id == RCS0) { + if (workload->engine->id == RCS0) { COPY_REG(bb_per_ctx_ptr); COPY_REG(rcs_indirect_ctx); COPY_REG(rcs_indirect_ctx_offset); @@ -175,14 +174,14 @@ static int populate_shadow_context(struct intel_vgpu_workload *workload) if (IS_RESTORE_INHIBIT(shadow_ring_context->ctx_ctrl.val)) return 0; - gvt_dbg_sched("ring id %d workload lrca %x", ring_id, - workload->ctx_desc.lrca); - - context_page_num = gvt->dev_priv->engine[ring_id]->context_size; + gvt_dbg_sched("ring %s workload lrca %x", + workload->engine->name, + workload->ctx_desc.lrca); + context_page_num = workload->engine->context_size; context_page_num = context_page_num >> PAGE_SHIFT; - if (IS_BROADWELL(gvt->dev_priv) && ring_id == RCS0) + if (IS_BROADWELL(gvt->gt->i915) && workload->engine->id == RCS0) context_page_num = 19; i = 2; @@ -210,38 +209,43 @@ static inline bool is_gvt_request(struct i915_request *rq) return intel_context_force_single_submission(rq->context); } -static void save_ring_hw_state(struct intel_vgpu *vgpu, int ring_id) +static void save_ring_hw_state(struct intel_vgpu *vgpu, + const struct intel_engine_cs *engine) { - struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv; - u32 ring_base = dev_priv->engine[ring_id]->mmio_base; + struct intel_uncore *uncore = engine->uncore; i915_reg_t reg; - reg = RING_INSTDONE(ring_base); - vgpu_vreg(vgpu, i915_mmio_reg_offset(reg)) = I915_READ_FW(reg); - reg = RING_ACTHD(ring_base); - vgpu_vreg(vgpu, i915_mmio_reg_offset(reg)) = I915_READ_FW(reg); - reg = RING_ACTHD_UDW(ring_base); - vgpu_vreg(vgpu, i915_mmio_reg_offset(reg)) = I915_READ_FW(reg); + reg = RING_INSTDONE(engine->mmio_base); + vgpu_vreg(vgpu, i915_mmio_reg_offset(reg)) = + intel_uncore_read(uncore, reg); + + reg = RING_ACTHD(engine->mmio_base); + vgpu_vreg(vgpu, i915_mmio_reg_offset(reg)) = + intel_uncore_read(uncore, reg); + + reg = RING_ACTHD_UDW(engine->mmio_base); + vgpu_vreg(vgpu, i915_mmio_reg_offset(reg)) = + intel_uncore_read(uncore, reg); } static int shadow_context_status_change(struct notifier_block *nb, unsigned long action, void *data) { - struct i915_request *req = data; + struct i915_request *rq = data; struct intel_gvt *gvt = container_of(nb, struct intel_gvt, - shadow_ctx_notifier_block[req->engine->id]); + shadow_ctx_notifier_block[rq->engine->id]); struct intel_gvt_workload_scheduler *scheduler = &gvt->scheduler; - enum intel_engine_id ring_id = req->engine->id; + enum intel_engine_id ring_id = rq->engine->id; struct intel_vgpu_workload *workload; unsigned long flags; - if (!is_gvt_request(req)) { + if (!is_gvt_request(rq)) { spin_lock_irqsave(&scheduler->mmio_context_lock, flags); if (action == INTEL_CONTEXT_SCHEDULE_IN && scheduler->engine_owner[ring_id]) { /* Switch ring from vGPU to host. */ intel_gvt_switch_mmio(scheduler->engine_owner[ring_id], - NULL, ring_id); + NULL, rq->engine); scheduler->engine_owner[ring_id] = NULL; } spin_unlock_irqrestore(&scheduler->mmio_context_lock, flags); @@ -259,7 +263,7 @@ static int shadow_context_status_change(struct notifier_block *nb, if (workload->vgpu != scheduler->engine_owner[ring_id]) { /* Switch ring from host to vGPU or vGPU to vGPU. */ intel_gvt_switch_mmio(scheduler->engine_owner[ring_id], - workload->vgpu, ring_id); + workload->vgpu, rq->engine); scheduler->engine_owner[ring_id] = workload->vgpu; } else gvt_dbg_sched("skip ring %d mmio switch for vgpu%d\n", @@ -268,11 +272,11 @@ static int shadow_context_status_change(struct notifier_block *nb, atomic_set(&workload->shadow_ctx_active, 1); break; case INTEL_CONTEXT_SCHEDULE_OUT: - save_ring_hw_state(workload->vgpu, ring_id); + save_ring_hw_state(workload->vgpu, rq->engine); atomic_set(&workload->shadow_ctx_active, 0); break; case INTEL_CONTEXT_SCHEDULE_PREEMPTED: - save_ring_hw_state(workload->vgpu, ring_id); + save_ring_hw_state(workload->vgpu, rq->engine); break; default: WARN_ON(1); @@ -391,7 +395,7 @@ intel_gvt_workload_req_alloc(struct intel_vgpu_workload *workload) if (workload->req) return 0; - rq = i915_request_create(s->shadow[workload->ring_id]); + rq = i915_request_create(s->shadow[workload->engine->id]); if (IS_ERR(rq)) { gvt_vgpu_err("fail to allocate gem request\n"); return PTR_ERR(rq); @@ -420,15 +424,16 @@ int intel_gvt_scan_and_shadow_workload(struct intel_vgpu_workload *workload) if (workload->shadow) return 0; - if (!test_and_set_bit(workload->ring_id, s->shadow_ctx_desc_updated)) - shadow_context_descriptor_update(s->shadow[workload->ring_id], + if (!test_and_set_bit(workload->engine->id, s->shadow_ctx_desc_updated)) + shadow_context_descriptor_update(s->shadow[workload->engine->id], workload); ret = intel_gvt_scan_and_shadow_ringbuffer(workload); if (ret) return ret; - if (workload->ring_id == RCS0 && workload->wa_ctx.indirect_ctx.size) { + if (workload->engine->id == RCS0 && + workload->wa_ctx.indirect_ctx.size) { ret = intel_gvt_scan_and_shadow_wa_ctx(&workload->wa_ctx); if (ret) goto err_shadow; @@ -436,6 +441,7 @@ int intel_gvt_scan_and_shadow_workload(struct intel_vgpu_workload *workload) workload->shadow = true; return 0; + err_shadow: release_shadow_wa_ctx(&workload->wa_ctx); return ret; @@ -567,12 +573,8 @@ static int prepare_shadow_wa_ctx(struct intel_shadow_wa_ctx *wa_ctx) static void update_vreg_in_ctx(struct intel_vgpu_workload *workload) { - struct intel_vgpu *vgpu = workload->vgpu; - struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv; - u32 ring_base; - - ring_base = dev_priv->engine[workload->ring_id]->mmio_base; - vgpu_vreg_t(vgpu, RING_START(ring_base)) = workload->rb_start; + vgpu_vreg_t(workload->vgpu, RING_START(workload->engine->mmio_base)) = + workload->rb_start; } static void release_shadow_batch_buffer(struct intel_vgpu_workload *workload) @@ -608,7 +610,6 @@ static int prepare_workload(struct intel_vgpu_workload *workload) { struct intel_vgpu *vgpu = workload->vgpu; struct intel_vgpu_submission *s = &vgpu->submission; - int ring = workload->ring_id; int ret = 0; ret = intel_vgpu_pin_mm(workload->shadow_mm); @@ -625,7 +626,7 @@ static int prepare_workload(struct intel_vgpu_workload *workload) update_shadow_pdps(workload); - set_context_ppgtt_from_shadow(workload, s->shadow[ring]); + set_context_ppgtt_from_shadow(workload, s->shadow[workload->engine->id]); ret = intel_vgpu_sync_oos_pages(workload->vgpu); if (ret) { @@ -677,11 +678,10 @@ static int dispatch_workload(struct intel_vgpu_workload *workload) { struct intel_vgpu *vgpu = workload->vgpu; struct i915_request *rq; - int ring_id = workload->ring_id; int ret; - gvt_dbg_sched("ring id %d prepare to dispatch workload %p\n", - ring_id, workload); + gvt_dbg_sched("ring id %s prepare to dispatch workload %p\n", + workload->engine->name, workload); mutex_lock(&vgpu->vgpu_lock); @@ -710,8 +710,8 @@ out: } if (!IS_ERR_OR_NULL(workload->req)) { - gvt_dbg_sched("ring id %d submit workload to i915 %p\n", - ring_id, workload->req); + gvt_dbg_sched("ring id %s submit workload to i915 %p\n", + workload->engine->name, workload->req); i915_request_add(workload->req); workload->dispatched = true; } @@ -722,8 +722,8 @@ err_req: return ret; } -static struct intel_vgpu_workload *pick_next_workload( - struct intel_gvt *gvt, int ring_id) +static struct intel_vgpu_workload * +pick_next_workload(struct intel_gvt *gvt, struct intel_engine_cs *engine) { struct intel_gvt_workload_scheduler *scheduler = &gvt->scheduler; struct intel_vgpu_workload *workload = NULL; @@ -735,27 +735,27 @@ static struct intel_vgpu_workload *pick_next_workload( * bail out */ if (!scheduler->current_vgpu) { - gvt_dbg_sched("ring id %d stop - no current vgpu\n", ring_id); + gvt_dbg_sched("ring %s stop - no current vgpu\n", engine->name); goto out; } if (scheduler->need_reschedule) { - gvt_dbg_sched("ring id %d stop - will reschedule\n", ring_id); + gvt_dbg_sched("ring %s stop - will reschedule\n", engine->name); goto out; } if (!scheduler->current_vgpu->active || - list_empty(workload_q_head(scheduler->current_vgpu, ring_id))) + list_empty(workload_q_head(scheduler->current_vgpu, engine))) goto out; /* * still have current workload, maybe the workload disptacher * fail to submit it for some reason, resubmit it. */ - if (scheduler->current_workload[ring_id]) { - workload = scheduler->current_workload[ring_id]; - gvt_dbg_sched("ring id %d still have current workload %p\n", - ring_id, workload); + if (scheduler->current_workload[engine->id]) { + workload = scheduler->current_workload[engine->id]; + gvt_dbg_sched("ring %s still have current workload %p\n", + engine->name, workload); goto out; } @@ -765,13 +765,14 @@ static struct intel_vgpu_workload *pick_next_workload( * will wait the current workload is finished when trying to * schedule out a vgpu. */ - scheduler->current_workload[ring_id] = container_of( - workload_q_head(scheduler->current_vgpu, ring_id)->next, - struct intel_vgpu_workload, list); + scheduler->current_workload[engine->id] = + list_first_entry(workload_q_head(scheduler->current_vgpu, + engine), + struct intel_vgpu_workload, list); - workload = scheduler->current_workload[ring_id]; + workload = scheduler->current_workload[engine->id]; - gvt_dbg_sched("ring id %d pick new workload %p\n", ring_id, workload); + gvt_dbg_sched("ring %s pick new workload %p\n", engine->name, workload); atomic_inc(&workload->vgpu->submission.running_workload_num); out: @@ -783,14 +784,12 @@ static void update_guest_context(struct intel_vgpu_workload *workload) { struct i915_request *rq = workload->req; struct intel_vgpu *vgpu = workload->vgpu; - struct intel_gvt *gvt = vgpu->gvt; struct drm_i915_gem_object *ctx_obj = rq->context->state->obj; struct execlist_ring_context *shadow_ring_context; struct page *page; void *src; unsigned long context_gpa, context_page_num; int i; - struct drm_i915_private *dev_priv = gvt->dev_priv; u32 ring_base; u32 head, tail; u16 wrap_count; @@ -811,14 +810,14 @@ static void update_guest_context(struct intel_vgpu_workload *workload) head = (wrap_count << RB_HEAD_WRAP_CNT_OFF) | tail; - ring_base = dev_priv->engine[workload->ring_id]->mmio_base; + ring_base = rq->engine->mmio_base; vgpu_vreg_t(vgpu, RING_TAIL(ring_base)) = tail; vgpu_vreg_t(vgpu, RING_HEAD(ring_base)) = head; context_page_num = rq->engine->context_size; context_page_num = context_page_num >> PAGE_SHIFT; - if (IS_BROADWELL(gvt->dev_priv) && rq->engine->id == RCS0) + if (IS_BROADWELL(rq->i915) && rq->engine->id == RCS0) context_page_num = 19; i = 2; @@ -869,7 +868,7 @@ void intel_vgpu_clean_workloads(struct intel_vgpu *vgpu, intel_engine_mask_t engine_mask) { struct intel_vgpu_submission *s = &vgpu->submission; - struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv; + struct drm_i915_private *dev_priv = vgpu->gvt->gt->i915; struct intel_engine_cs *engine; struct intel_vgpu_workload *pos, *n; intel_engine_mask_t tmp; @@ -966,54 +965,47 @@ static void complete_current_workload(struct intel_gvt *gvt, int ring_id) mutex_unlock(&vgpu->vgpu_lock); } -struct workload_thread_param { - struct intel_gvt *gvt; - int ring_id; -}; - -static int workload_thread(void *priv) +static int workload_thread(void *arg) { - struct workload_thread_param *p = (struct workload_thread_param *)priv; - struct intel_gvt *gvt = p->gvt; - int ring_id = p->ring_id; + struct intel_engine_cs *engine = arg; + const bool need_force_wake = INTEL_GEN(engine->i915) >= 9; + struct intel_gvt *gvt = engine->i915->gvt; struct intel_gvt_workload_scheduler *scheduler = &gvt->scheduler; struct intel_vgpu_workload *workload = NULL; struct intel_vgpu *vgpu = NULL; int ret; - bool need_force_wake = (INTEL_GEN(gvt->dev_priv) >= 9); DEFINE_WAIT_FUNC(wait, woken_wake_function); - struct intel_runtime_pm *rpm = &gvt->dev_priv->runtime_pm; - - kfree(p); - gvt_dbg_core("workload thread for ring %d started\n", ring_id); + gvt_dbg_core("workload thread for ring %s started\n", engine->name); while (!kthread_should_stop()) { - add_wait_queue(&scheduler->waitq[ring_id], &wait); + intel_wakeref_t wakeref; + + add_wait_queue(&scheduler->waitq[engine->id], &wait); do { - workload = pick_next_workload(gvt, ring_id); + workload = pick_next_workload(gvt, engine); if (workload) break; wait_woken(&wait, TASK_INTERRUPTIBLE, MAX_SCHEDULE_TIMEOUT); } while (!kthread_should_stop()); - remove_wait_queue(&scheduler->waitq[ring_id], &wait); + remove_wait_queue(&scheduler->waitq[engine->id], &wait); if (!workload) break; - gvt_dbg_sched("ring id %d next workload %p vgpu %d\n", - workload->ring_id, workload, - workload->vgpu->id); + gvt_dbg_sched("ring %s next workload %p vgpu %d\n", + engine->name, workload, + workload->vgpu->id); - intel_runtime_pm_get(rpm); + wakeref = intel_runtime_pm_get(engine->uncore->rpm); - gvt_dbg_sched("ring id %d will dispatch workload %p\n", - workload->ring_id, workload); + gvt_dbg_sched("ring %s will dispatch workload %p\n", + engine->name, workload); if (need_force_wake) - intel_uncore_forcewake_get(&gvt->dev_priv->uncore, - FORCEWAKE_ALL); + intel_uncore_forcewake_get(engine->uncore, + FORCEWAKE_ALL); /* * Update the vReg of the vGPU which submitted this * workload. The vGPU may use these registers for checking @@ -1030,21 +1022,21 @@ static int workload_thread(void *priv) goto complete; } - gvt_dbg_sched("ring id %d wait workload %p\n", - workload->ring_id, workload); + gvt_dbg_sched("ring %s wait workload %p\n", + engine->name, workload); i915_request_wait(workload->req, 0, MAX_SCHEDULE_TIMEOUT); complete: gvt_dbg_sched("will complete workload %p, status: %d\n", - workload, workload->status); + workload, workload->status); - complete_current_workload(gvt, ring_id); + complete_current_workload(gvt, engine->id); if (need_force_wake) - intel_uncore_forcewake_put(&gvt->dev_priv->uncore, - FORCEWAKE_ALL); + intel_uncore_forcewake_put(engine->uncore, + FORCEWAKE_ALL); - intel_runtime_pm_put_unchecked(rpm); + intel_runtime_pm_put(engine->uncore->rpm, wakeref); if (ret && (vgpu_is_vm_unhealthy(ret))) enter_failsafe_mode(vgpu, GVT_FAILSAFE_GUEST_ERR); } @@ -1073,7 +1065,7 @@ void intel_gvt_clean_workload_scheduler(struct intel_gvt *gvt) gvt_dbg_core("clean workload scheduler\n"); - for_each_engine(engine, gvt->dev_priv, i) { + for_each_engine(engine, gvt->gt, i) { atomic_notifier_chain_unregister( &engine->context_status_notifier, &gvt->shadow_ctx_notifier_block[i]); @@ -1084,7 +1076,6 @@ void intel_gvt_clean_workload_scheduler(struct intel_gvt *gvt) int intel_gvt_init_workload_scheduler(struct intel_gvt *gvt) { struct intel_gvt_workload_scheduler *scheduler = &gvt->scheduler; - struct workload_thread_param *param = NULL; struct intel_engine_cs *engine; enum intel_engine_id i; int ret; @@ -1093,20 +1084,11 @@ int intel_gvt_init_workload_scheduler(struct intel_gvt *gvt) init_waitqueue_head(&scheduler->workload_complete_wq); - for_each_engine(engine, gvt->dev_priv, i) { + for_each_engine(engine, gvt->gt, i) { init_waitqueue_head(&scheduler->waitq[i]); - param = kzalloc(sizeof(*param), GFP_KERNEL); - if (!param) { - ret = -ENOMEM; - goto err; - } - - param->gvt = gvt; - param->ring_id = i; - - scheduler->thread[i] = kthread_run(workload_thread, param, - "gvt workload %d", i); + scheduler->thread[i] = kthread_run(workload_thread, engine, + "gvt:%s", engine->name); if (IS_ERR(scheduler->thread[i])) { gvt_err("fail to create workload thread\n"); ret = PTR_ERR(scheduler->thread[i]); @@ -1118,11 +1100,11 @@ int intel_gvt_init_workload_scheduler(struct intel_gvt *gvt) atomic_notifier_chain_register(&engine->context_status_notifier, &gvt->shadow_ctx_notifier_block[i]); } + return 0; + err: intel_gvt_clean_workload_scheduler(gvt); - kfree(param); - param = NULL; return ret; } @@ -1160,7 +1142,7 @@ void intel_vgpu_clean_submission(struct intel_vgpu *vgpu) intel_vgpu_select_submission_ops(vgpu, ALL_ENGINES, 0); i915_context_ppgtt_root_restore(s, i915_vm_to_ppgtt(s->shadow[0]->vm)); - for_each_engine(engine, vgpu->gvt->dev_priv, id) + for_each_engine(engine, vgpu->gvt->gt, id) intel_context_unpin(s->shadow[id]); kmem_cache_destroy(s->workloads); @@ -1217,7 +1199,7 @@ i915_context_ppgtt_root_save(struct intel_vgpu_submission *s, */ int intel_vgpu_setup_submission(struct intel_vgpu *vgpu) { - struct drm_i915_private *i915 = vgpu->gvt->dev_priv; + struct drm_i915_private *i915 = vgpu->gvt->gt->i915; struct intel_vgpu_submission *s = &vgpu->submission; struct intel_engine_cs *engine; struct i915_ppgtt *ppgtt; @@ -1230,7 +1212,7 @@ int intel_vgpu_setup_submission(struct intel_vgpu *vgpu) i915_context_ppgtt_root_save(s, ppgtt); - for_each_engine(engine, i915, i) { + for_each_engine(engine, vgpu->gvt->gt, i) { struct intel_context *ce; INIT_LIST_HEAD(&s->workload_q_head[i]); @@ -1283,7 +1265,7 @@ int intel_vgpu_setup_submission(struct intel_vgpu *vgpu) out_shadow_ctx: i915_context_ppgtt_root_restore(s, ppgtt); - for_each_engine(engine, i915, i) { + for_each_engine(engine, vgpu->gvt->gt, i) { if (IS_ERR(s->shadow[i])) break; @@ -1310,6 +1292,7 @@ int intel_vgpu_select_submission_ops(struct intel_vgpu *vgpu, intel_engine_mask_t engine_mask, unsigned int interface) { + struct drm_i915_private *i915 = vgpu->gvt->gt->i915; struct intel_vgpu_submission *s = &vgpu->submission; const struct intel_vgpu_submission_ops *ops[] = { [INTEL_VGPU_EXECLIST_SUBMISSION] = @@ -1317,10 +1300,11 @@ int intel_vgpu_select_submission_ops(struct intel_vgpu *vgpu, }; int ret; - if (WARN_ON(interface >= ARRAY_SIZE(ops))) + if (drm_WARN_ON(&i915->drm, interface >= ARRAY_SIZE(ops))) return -EINVAL; - if (WARN_ON(interface == 0 && engine_mask != ALL_ENGINES)) + if (drm_WARN_ON(&i915->drm, + interface == 0 && engine_mask != ALL_ENGINES)) return -EINVAL; if (s->active) @@ -1442,7 +1426,7 @@ static int prepare_mm(struct intel_vgpu_workload *workload) /** * intel_vgpu_create_workload - create a vGPU workload * @vgpu: a vGPU - * @ring_id: ring index + * @engine: the engine * @desc: a guest context descriptor * * This function is called when creating a vGPU workload. @@ -1453,14 +1437,14 @@ static int prepare_mm(struct intel_vgpu_workload *workload) * */ struct intel_vgpu_workload * -intel_vgpu_create_workload(struct intel_vgpu *vgpu, int ring_id, +intel_vgpu_create_workload(struct intel_vgpu *vgpu, + const struct intel_engine_cs *engine, struct execlist_ctx_descriptor_format *desc) { struct intel_vgpu_submission *s = &vgpu->submission; - struct list_head *q = workload_q_head(vgpu, ring_id); + struct list_head *q = workload_q_head(vgpu, engine); struct intel_vgpu_workload *last_workload = NULL; struct intel_vgpu_workload *workload = NULL; - struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv; u64 ring_context_gpa; u32 head, tail, start, ctl, ctx_ctl, per_ctx, indirect_ctx; u32 guest_head; @@ -1487,10 +1471,10 @@ intel_vgpu_create_workload(struct intel_vgpu *vgpu, int ring_id, list_for_each_entry_reverse(last_workload, q, list) { if (same_context(&last_workload->ctx_desc, desc)) { - gvt_dbg_el("ring id %d cur workload == last\n", - ring_id); + gvt_dbg_el("ring %s cur workload == last\n", + engine->name); gvt_dbg_el("ctx head %x real head %lx\n", head, - last_workload->rb_tail); + last_workload->rb_tail); /* * cannot use guest context head pointer here, * as it might not be updated at this time @@ -1500,7 +1484,7 @@ intel_vgpu_create_workload(struct intel_vgpu *vgpu, int ring_id, } } - gvt_dbg_el("ring id %d begin a new workload\n", ring_id); + gvt_dbg_el("ring %s begin a new workload\n", engine->name); /* record some ring buffer register values for scan and shadow */ intel_gvt_hypervisor_read_gpa(vgpu, ring_context_gpa + @@ -1520,7 +1504,7 @@ intel_vgpu_create_workload(struct intel_vgpu *vgpu, int ring_id, if (IS_ERR(workload)) return workload; - workload->ring_id = ring_id; + workload->engine = engine; workload->ctx_desc = *desc; workload->ring_context_gpa = ring_context_gpa; workload->rb_head = head; @@ -1529,7 +1513,7 @@ intel_vgpu_create_workload(struct intel_vgpu *vgpu, int ring_id, workload->rb_start = start; workload->rb_ctl = ctl; - if (ring_id == RCS0) { + if (engine->id == RCS0) { intel_gvt_hypervisor_read_gpa(vgpu, ring_context_gpa + RING_CTX_OFF(bb_per_ctx_ptr.val), &per_ctx, 4); intel_gvt_hypervisor_read_gpa(vgpu, ring_context_gpa + @@ -1567,8 +1551,8 @@ intel_vgpu_create_workload(struct intel_vgpu *vgpu, int ring_id, } } - gvt_dbg_el("workload %p ring id %d head %x tail %x start %x ctl %x\n", - workload, ring_id, head, tail, start, ctl); + gvt_dbg_el("workload %p ring %s head %x tail %x start %x ctl %x\n", + workload, engine->name, head, tail, start, ctl); ret = prepare_mm(workload); if (ret) { @@ -1579,10 +1563,11 @@ intel_vgpu_create_workload(struct intel_vgpu *vgpu, int ring_id, /* Only scan and shadow the first workload in the queue * as there is only one pre-allocated buf-obj for shadow. */ - if (list_empty(workload_q_head(vgpu, ring_id))) { - intel_runtime_pm_get(&dev_priv->runtime_pm); - ret = intel_gvt_scan_and_shadow_workload(workload); - intel_runtime_pm_put_unchecked(&dev_priv->runtime_pm); + if (list_empty(q)) { + intel_wakeref_t wakeref; + + with_intel_runtime_pm(engine->gt->uncore->rpm, wakeref) + ret = intel_gvt_scan_and_shadow_workload(workload); } if (ret) { @@ -1602,7 +1587,7 @@ intel_vgpu_create_workload(struct intel_vgpu *vgpu, int ring_id, void intel_vgpu_queue_workload(struct intel_vgpu_workload *workload) { list_add_tail(&workload->list, - workload_q_head(workload->vgpu, workload->ring_id)); + workload_q_head(workload->vgpu, workload->engine)); intel_gvt_kick_schedule(workload->vgpu->gvt); - wake_up(&workload->vgpu->gvt->scheduler.waitq[workload->ring_id]); + wake_up(&workload->vgpu->gvt->scheduler.waitq[workload->engine->id]); } diff --git a/drivers/gpu/drm/i915/gvt/scheduler.h b/drivers/gpu/drm/i915/gvt/scheduler.h index c50d14a9ce85..bf7fc0ca4cb1 100644 --- a/drivers/gpu/drm/i915/gvt/scheduler.h +++ b/drivers/gpu/drm/i915/gvt/scheduler.h @@ -79,7 +79,7 @@ struct intel_shadow_wa_ctx { struct intel_vgpu_workload { struct intel_vgpu *vgpu; - int ring_id; + const struct intel_engine_cs *engine; struct i915_request *req; /* if this workload has been dispatched to i915? */ bool dispatched; @@ -129,8 +129,8 @@ struct intel_vgpu_shadow_bb { bool ppgtt; }; -#define workload_q_head(vgpu, ring_id) \ - (&(vgpu->submission.workload_q_head[ring_id])) +#define workload_q_head(vgpu, e) \ + (&(vgpu)->submission.workload_q_head[(e)->id]) void intel_vgpu_queue_workload(struct intel_vgpu_workload *workload); @@ -155,7 +155,8 @@ extern const struct intel_vgpu_submission_ops intel_vgpu_execlist_submission_ops; struct intel_vgpu_workload * -intel_vgpu_create_workload(struct intel_vgpu *vgpu, int ring_id, +intel_vgpu_create_workload(struct intel_vgpu *vgpu, + const struct intel_engine_cs *engine, struct execlist_ctx_descriptor_format *desc); void intel_vgpu_destroy_workload(struct intel_vgpu_workload *workload); diff --git a/drivers/gpu/drm/i915/gvt/vgpu.c b/drivers/gpu/drm/i915/gvt/vgpu.c index 85bd9bf4f6ee..1d5ff88078bd 100644 --- a/drivers/gpu/drm/i915/gvt/vgpu.c +++ b/drivers/gpu/drm/i915/gvt/vgpu.c @@ -37,6 +37,7 @@ void populate_pvinfo_page(struct intel_vgpu *vgpu) { + struct drm_i915_private *i915 = vgpu->gvt->gt->i915; /* setup the ballooning information */ vgpu_vreg64_t(vgpu, vgtif_reg(magic)) = VGT_MAGIC; vgpu_vreg_t(vgpu, vgtif_reg(version_major)) = 1; @@ -69,7 +70,7 @@ void populate_pvinfo_page(struct intel_vgpu *vgpu) vgpu_hidden_gmadr_base(vgpu), vgpu_hidden_sz(vgpu)); gvt_dbg_core("fence size %d\n", vgpu_fence_sz(vgpu)); - WARN_ON(sizeof(struct vgt_if) != VGT_PVINFO_SIZE); + drm_WARN_ON(&i915->drm, sizeof(struct vgt_if) != VGT_PVINFO_SIZE); } #define VGPU_MAX_WEIGHT 16 @@ -148,12 +149,12 @@ int intel_gvt_init_vgpu_types(struct intel_gvt *gvt) gvt->types[i].avail_instance = min(low_avail / vgpu_types[i].low_mm, high_avail / vgpu_types[i].high_mm); - if (IS_GEN(gvt->dev_priv, 8)) + if (IS_GEN(gvt->gt->i915, 8)) sprintf(gvt->types[i].name, "GVTg_V4_%s", - vgpu_types[i].name); - else if (IS_GEN(gvt->dev_priv, 9)) + vgpu_types[i].name); + else if (IS_GEN(gvt->gt->i915, 9)) sprintf(gvt->types[i].name, "GVTg_V5_%s", - vgpu_types[i].name); + vgpu_types[i].name); gvt_dbg_core("type[%d]: %s avail %u low %u high %u fence %u weight %u res %s\n", i, gvt->types[i].name, @@ -271,11 +272,19 @@ void intel_gvt_release_vgpu(struct intel_vgpu *vgpu) void intel_gvt_destroy_vgpu(struct intel_vgpu *vgpu) { struct intel_gvt *gvt = vgpu->gvt; + struct drm_i915_private *i915 = gvt->gt->i915; - mutex_lock(&vgpu->vgpu_lock); + drm_WARN(&i915->drm, vgpu->active, "vGPU is still active!\n"); - WARN(vgpu->active, "vGPU is still active!\n"); + /* + * remove idr first so later clean can judge if need to stop + * service if no active vgpu. + */ + mutex_lock(&gvt->lock); + idr_remove(&gvt->vgpu_idr, vgpu->id); + mutex_unlock(&gvt->lock); + mutex_lock(&vgpu->vgpu_lock); intel_gvt_debugfs_remove_vgpu(vgpu); intel_vgpu_clean_sched_policy(vgpu); intel_vgpu_clean_submission(vgpu); @@ -290,7 +299,6 @@ void intel_gvt_destroy_vgpu(struct intel_vgpu *vgpu) mutex_unlock(&vgpu->vgpu_lock); mutex_lock(&gvt->lock); - idr_remove(&gvt->vgpu_idr, vgpu->id); if (idr_is_empty(&gvt->vgpu_idr)) intel_gvt_clean_irq(gvt); intel_gvt_update_vgpu_types(gvt); @@ -426,9 +434,7 @@ static struct intel_vgpu *__intel_gvt_create_vgpu(struct intel_gvt *gvt, if (ret) goto out_clean_sched_policy; - /*TODO: add more platforms support */ - if (IS_SKYLAKE(gvt->dev_priv) || IS_KABYLAKE(gvt->dev_priv)) - ret = intel_gvt_hypervisor_set_edid(vgpu, PORT_D); + ret = intel_gvt_hypervisor_set_edid(vgpu, PORT_D); if (ret) goto out_clean_sched_policy; @@ -560,9 +566,9 @@ void intel_gvt_reset_vgpu_locked(struct intel_vgpu *vgpu, bool dmlr, intel_vgpu_reset_mmio(vgpu, dmlr); populate_pvinfo_page(vgpu); - intel_vgpu_reset_display(vgpu); if (dmlr) { + intel_vgpu_reset_display(vgpu); intel_vgpu_reset_cfg_space(vgpu); /* only reset the failsafe mode when dmlr reset */ vgpu->failsafe = false; diff --git a/drivers/gpu/drm/i915/i915_active.c b/drivers/gpu/drm/i915/i915_active.c index 992b00fc5745..c4048628188a 100644 --- a/drivers/gpu/drm/i915/i915_active.c +++ b/drivers/gpu/drm/i915/i915_active.c @@ -7,6 +7,7 @@ #include <linux/debugobjects.h> #include "gt/intel_context.h" +#include "gt/intel_engine_heartbeat.h" #include "gt/intel_engine_pm.h" #include "gt/intel_ring.h" @@ -452,6 +453,9 @@ static void enable_signaling(struct i915_active_fence *active) { struct dma_fence *fence; + if (unlikely(is_barrier(active))) + return; + fence = i915_active_fence_get(active); if (!fence) return; @@ -460,26 +464,49 @@ static void enable_signaling(struct i915_active_fence *active) dma_fence_put(fence); } -int i915_active_wait(struct i915_active *ref) +static int flush_barrier(struct active_node *it) { - struct active_node *it, *n; - int err = 0; + struct intel_engine_cs *engine; - might_sleep(); + if (likely(!is_barrier(&it->base))) + return 0; - if (!i915_active_acquire_if_busy(ref)) + engine = __barrier_to_engine(it); + smp_rmb(); /* serialise with add_active_barriers */ + if (!is_barrier(&it->base)) return 0; - /* Flush lazy signals */ + return intel_engine_flush_barriers(engine); +} + +static int flush_lazy_signals(struct i915_active *ref) +{ + struct active_node *it, *n; + int err = 0; + enable_signaling(&ref->excl); rbtree_postorder_for_each_entry_safe(it, n, &ref->tree, node) { - if (is_barrier(&it->base)) /* unconnected idle barrier */ - continue; + err = flush_barrier(it); /* unconnected idle barrier? */ + if (err) + break; enable_signaling(&it->base); } - /* Any fence added after the wait begins will not be auto-signaled */ + return err; +} + +int i915_active_wait(struct i915_active *ref) +{ + int err; + + might_sleep(); + + if (!i915_active_acquire_if_busy(ref)) + return 0; + + /* Any fence added after the wait begins will not be auto-signaled */ + err = flush_lazy_signals(ref); i915_active_release(ref); if (err) return err; @@ -491,25 +518,81 @@ int i915_active_wait(struct i915_active *ref) return 0; } -int i915_request_await_active(struct i915_request *rq, struct i915_active *ref) +static int __await_active(struct i915_active_fence *active, + int (*fn)(void *arg, struct dma_fence *fence), + void *arg) +{ + struct dma_fence *fence; + + if (is_barrier(active)) /* XXX flush the barrier? */ + return 0; + + fence = i915_active_fence_get(active); + if (fence) { + int err; + + err = fn(arg, fence); + dma_fence_put(fence); + if (err < 0) + return err; + } + + return 0; +} + +static int await_active(struct i915_active *ref, + unsigned int flags, + int (*fn)(void *arg, struct dma_fence *fence), + void *arg) { int err = 0; + /* We must always wait for the exclusive fence! */ if (rcu_access_pointer(ref->excl.fence)) { - struct dma_fence *fence; - - rcu_read_lock(); - fence = dma_fence_get_rcu_safe(&ref->excl.fence); - rcu_read_unlock(); - if (fence) { - err = i915_request_await_dma_fence(rq, fence); - dma_fence_put(fence); + err = __await_active(&ref->excl, fn, arg); + if (err) + return err; + } + + if (flags & I915_ACTIVE_AWAIT_ALL && i915_active_acquire_if_busy(ref)) { + struct active_node *it, *n; + + rbtree_postorder_for_each_entry_safe(it, n, &ref->tree, node) { + err = __await_active(&it->base, fn, arg); + if (err) + break; } + i915_active_release(ref); + if (err) + return err; } - /* In the future we may choose to await on all fences */ + return 0; +} + +static int rq_await_fence(void *arg, struct dma_fence *fence) +{ + return i915_request_await_dma_fence(arg, fence); +} - return err; +int i915_request_await_active(struct i915_request *rq, + struct i915_active *ref, + unsigned int flags) +{ + return await_active(ref, flags, rq_await_fence, rq); +} + +static int sw_await_fence(void *arg, struct dma_fence *fence) +{ + return i915_sw_fence_await_dma_fence(arg, fence, 0, + GFP_NOWAIT | __GFP_NOWARN); +} + +int i915_sw_fence_await_active(struct i915_sw_fence *fence, + struct i915_active *ref, + unsigned int flags) +{ + return await_active(ref, flags, sw_await_fence, fence); } #if IS_ENABLED(CONFIG_DRM_I915_DEBUG_GEM) diff --git a/drivers/gpu/drm/i915/i915_active.h b/drivers/gpu/drm/i915/i915_active.h index 973ff0447c6c..b3282ae7913c 100644 --- a/drivers/gpu/drm/i915/i915_active.h +++ b/drivers/gpu/drm/i915/i915_active.h @@ -183,7 +183,13 @@ static inline bool i915_active_has_exclusive(struct i915_active *ref) int i915_active_wait(struct i915_active *ref); -int i915_request_await_active(struct i915_request *rq, struct i915_active *ref); +int i915_sw_fence_await_active(struct i915_sw_fence *fence, + struct i915_active *ref, + unsigned int flags); +int i915_request_await_active(struct i915_request *rq, + struct i915_active *ref, + unsigned int flags); +#define I915_ACTIVE_AWAIT_ALL BIT(0) int i915_active_acquire(struct i915_active *ref); bool i915_active_acquire_if_busy(struct i915_active *ref); diff --git a/drivers/gpu/drm/i915/i915_buddy.c b/drivers/gpu/drm/i915/i915_buddy.c index 66883af64ca1..20babbdb297d 100644 --- a/drivers/gpu/drm/i915/i915_buddy.c +++ b/drivers/gpu/drm/i915/i915_buddy.c @@ -312,7 +312,8 @@ i915_buddy_alloc(struct i915_buddy_mm *mm, unsigned int order) return block; out_free: - __i915_buddy_free(mm, block); + if (i != order) + __i915_buddy_free(mm, block); return ERR_PTR(err); } diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c index 8f2525e4ce0f..6ca797128aa1 100644 --- a/drivers/gpu/drm/i915/i915_debugfs.c +++ b/drivers/gpu/drm/i915/i915_debugfs.c @@ -996,220 +996,6 @@ static int i915_frequency_info(struct seq_file *m, void *unused) return ret; } -static int ilk_drpc_info(struct seq_file *m) -{ - struct drm_i915_private *i915 = node_to_i915(m->private); - struct intel_uncore *uncore = &i915->uncore; - u32 rgvmodectl, rstdbyctl; - u16 crstandvid; - - rgvmodectl = intel_uncore_read(uncore, MEMMODECTL); - rstdbyctl = intel_uncore_read(uncore, RSTDBYCTL); - crstandvid = intel_uncore_read16(uncore, CRSTANDVID); - - seq_printf(m, "HD boost: %s\n", yesno(rgvmodectl & MEMMODE_BOOST_EN)); - seq_printf(m, "Boost freq: %d\n", - (rgvmodectl & MEMMODE_BOOST_FREQ_MASK) >> - MEMMODE_BOOST_FREQ_SHIFT); - seq_printf(m, "HW control enabled: %s\n", - yesno(rgvmodectl & MEMMODE_HWIDLE_EN)); - seq_printf(m, "SW control enabled: %s\n", - yesno(rgvmodectl & MEMMODE_SWMODE_EN)); - seq_printf(m, "Gated voltage change: %s\n", - yesno(rgvmodectl & MEMMODE_RCLK_GATE)); - seq_printf(m, "Starting frequency: P%d\n", - (rgvmodectl & MEMMODE_FSTART_MASK) >> MEMMODE_FSTART_SHIFT); - seq_printf(m, "Max P-state: P%d\n", - (rgvmodectl & MEMMODE_FMAX_MASK) >> MEMMODE_FMAX_SHIFT); - seq_printf(m, "Min P-state: P%d\n", (rgvmodectl & MEMMODE_FMIN_MASK)); - seq_printf(m, "RS1 VID: %d\n", (crstandvid & 0x3f)); - seq_printf(m, "RS2 VID: %d\n", ((crstandvid >> 8) & 0x3f)); - seq_printf(m, "Render standby enabled: %s\n", - yesno(!(rstdbyctl & RCX_SW_EXIT))); - seq_puts(m, "Current RS state: "); - switch (rstdbyctl & RSX_STATUS_MASK) { - case RSX_STATUS_ON: - seq_puts(m, "on\n"); - break; - case RSX_STATUS_RC1: - seq_puts(m, "RC1\n"); - break; - case RSX_STATUS_RC1E: - seq_puts(m, "RC1E\n"); - break; - case RSX_STATUS_RS1: - seq_puts(m, "RS1\n"); - break; - case RSX_STATUS_RS2: - seq_puts(m, "RS2 (RC6)\n"); - break; - case RSX_STATUS_RS3: - seq_puts(m, "RC3 (RC6+)\n"); - break; - default: - seq_puts(m, "unknown\n"); - break; - } - - return 0; -} - -static int i915_forcewake_domains(struct seq_file *m, void *data) -{ - struct drm_i915_private *i915 = node_to_i915(m->private); - struct intel_uncore *uncore = &i915->uncore; - struct intel_uncore_forcewake_domain *fw_domain; - unsigned int tmp; - - seq_printf(m, "user.bypass_count = %u\n", - uncore->user_forcewake_count); - - for_each_fw_domain(fw_domain, uncore, tmp) - seq_printf(m, "%s.wake_count = %u\n", - intel_uncore_forcewake_domain_to_str(fw_domain->id), - READ_ONCE(fw_domain->wake_count)); - - return 0; -} - -static void print_rc6_res(struct seq_file *m, - const char *title, - const i915_reg_t reg) -{ - struct drm_i915_private *i915 = node_to_i915(m->private); - intel_wakeref_t wakeref; - - with_intel_runtime_pm(&i915->runtime_pm, wakeref) - seq_printf(m, "%s %u (%llu us)\n", title, - intel_uncore_read(&i915->uncore, reg), - intel_rc6_residency_us(&i915->gt.rc6, reg)); -} - -static int vlv_drpc_info(struct seq_file *m) -{ - struct drm_i915_private *dev_priv = node_to_i915(m->private); - u32 rcctl1, pw_status; - - pw_status = I915_READ(VLV_GTLC_PW_STATUS); - rcctl1 = I915_READ(GEN6_RC_CONTROL); - - seq_printf(m, "RC6 Enabled: %s\n", - yesno(rcctl1 & (GEN7_RC_CTL_TO_MODE | - GEN6_RC_CTL_EI_MODE(1)))); - seq_printf(m, "Render Power Well: %s\n", - (pw_status & VLV_GTLC_PW_RENDER_STATUS_MASK) ? "Up" : "Down"); - seq_printf(m, "Media Power Well: %s\n", - (pw_status & VLV_GTLC_PW_MEDIA_STATUS_MASK) ? "Up" : "Down"); - - print_rc6_res(m, "Render RC6 residency since boot:", VLV_GT_RENDER_RC6); - print_rc6_res(m, "Media RC6 residency since boot:", VLV_GT_MEDIA_RC6); - - return i915_forcewake_domains(m, NULL); -} - -static int gen6_drpc_info(struct seq_file *m) -{ - struct drm_i915_private *dev_priv = node_to_i915(m->private); - u32 gt_core_status, rcctl1, rc6vids = 0; - u32 gen9_powergate_enable = 0, gen9_powergate_status = 0; - - gt_core_status = I915_READ_FW(GEN6_GT_CORE_STATUS); - trace_i915_reg_rw(false, GEN6_GT_CORE_STATUS, gt_core_status, 4, true); - - rcctl1 = I915_READ(GEN6_RC_CONTROL); - if (INTEL_GEN(dev_priv) >= 9) { - gen9_powergate_enable = I915_READ(GEN9_PG_ENABLE); - gen9_powergate_status = I915_READ(GEN9_PWRGT_DOMAIN_STATUS); - } - - if (INTEL_GEN(dev_priv) <= 7) - sandybridge_pcode_read(dev_priv, GEN6_PCODE_READ_RC6VIDS, - &rc6vids, NULL); - - seq_printf(m, "RC1e Enabled: %s\n", - yesno(rcctl1 & GEN6_RC_CTL_RC1e_ENABLE)); - seq_printf(m, "RC6 Enabled: %s\n", - yesno(rcctl1 & GEN6_RC_CTL_RC6_ENABLE)); - if (INTEL_GEN(dev_priv) >= 9) { - seq_printf(m, "Render Well Gating Enabled: %s\n", - yesno(gen9_powergate_enable & GEN9_RENDER_PG_ENABLE)); - seq_printf(m, "Media Well Gating Enabled: %s\n", - yesno(gen9_powergate_enable & GEN9_MEDIA_PG_ENABLE)); - } - seq_printf(m, "Deep RC6 Enabled: %s\n", - yesno(rcctl1 & GEN6_RC_CTL_RC6p_ENABLE)); - seq_printf(m, "Deepest RC6 Enabled: %s\n", - yesno(rcctl1 & GEN6_RC_CTL_RC6pp_ENABLE)); - seq_puts(m, "Current RC state: "); - switch (gt_core_status & GEN6_RCn_MASK) { - case GEN6_RC0: - if (gt_core_status & GEN6_CORE_CPD_STATE_MASK) - seq_puts(m, "Core Power Down\n"); - else - seq_puts(m, "on\n"); - break; - case GEN6_RC3: - seq_puts(m, "RC3\n"); - break; - case GEN6_RC6: - seq_puts(m, "RC6\n"); - break; - case GEN6_RC7: - seq_puts(m, "RC7\n"); - break; - default: - seq_puts(m, "Unknown\n"); - break; - } - - seq_printf(m, "Core Power Down: %s\n", - yesno(gt_core_status & GEN6_CORE_CPD_STATE_MASK)); - if (INTEL_GEN(dev_priv) >= 9) { - seq_printf(m, "Render Power Well: %s\n", - (gen9_powergate_status & - GEN9_PWRGT_RENDER_STATUS_MASK) ? "Up" : "Down"); - seq_printf(m, "Media Power Well: %s\n", - (gen9_powergate_status & - GEN9_PWRGT_MEDIA_STATUS_MASK) ? "Up" : "Down"); - } - - /* Not exactly sure what this is */ - print_rc6_res(m, "RC6 \"Locked to RPn\" residency since boot:", - GEN6_GT_GFX_RC6_LOCKED); - print_rc6_res(m, "RC6 residency since boot:", GEN6_GT_GFX_RC6); - print_rc6_res(m, "RC6+ residency since boot:", GEN6_GT_GFX_RC6p); - print_rc6_res(m, "RC6++ residency since boot:", GEN6_GT_GFX_RC6pp); - - if (INTEL_GEN(dev_priv) <= 7) { - seq_printf(m, "RC6 voltage: %dmV\n", - GEN6_DECODE_RC6_VID(((rc6vids >> 0) & 0xff))); - seq_printf(m, "RC6+ voltage: %dmV\n", - GEN6_DECODE_RC6_VID(((rc6vids >> 8) & 0xff))); - seq_printf(m, "RC6++ voltage: %dmV\n", - GEN6_DECODE_RC6_VID(((rc6vids >> 16) & 0xff))); - } - - return i915_forcewake_domains(m, NULL); -} - -static int i915_drpc_info(struct seq_file *m, void *unused) -{ - struct drm_i915_private *dev_priv = node_to_i915(m->private); - intel_wakeref_t wakeref; - int err = -ENODEV; - - with_intel_runtime_pm(&dev_priv->runtime_pm, wakeref) { - if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) - err = vlv_drpc_info(m); - else if (INTEL_GEN(dev_priv) >= 6) - err = gen6_drpc_info(m); - else - err = ilk_drpc_info(m); - } - - return err; -} - static int i915_ring_freq_table(struct seq_file *m, void *unused) { struct drm_i915_private *dev_priv = node_to_i915(m->private); @@ -2360,10 +2146,8 @@ static const struct drm_info_list i915_debugfs_list[] = { {"i915_guc_stage_pool", i915_guc_stage_pool, 0}, {"i915_huc_load_status", i915_huc_load_status_info, 0}, {"i915_frequency_info", i915_frequency_info, 0}, - {"i915_drpc_info", i915_drpc_info, 0}, {"i915_ring_freq_table", i915_ring_freq_table, 0}, {"i915_context_status", i915_context_status, 0}, - {"i915_forcewake_domains", i915_forcewake_domains, 0}, {"i915_swizzle_info", i915_swizzle_info, 0}, {"i915_llc", i915_llc, 0}, {"i915_runtime_pm_status", i915_runtime_pm_status, 0}, diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c index dba5fe1391e8..82d9df15b22b 100644 --- a/drivers/gpu/drm/i915/i915_drv.c +++ b/drivers/gpu/drm/i915/i915_drv.c @@ -44,7 +44,6 @@ #include <drm/drm_ioctl.h> #include <drm/drm_irq.h> #include <drm/drm_probe_helper.h> -#include <drm/i915_drm.h> #include "display/intel_acpi.h" #include "display/intel_audio.h" @@ -71,6 +70,7 @@ #include "i915_debugfs.h" #include "i915_drv.h" +#include "i915_ioc32.h" #include "i915_irq.h" #include "i915_memcpy.h" #include "i915_perf.h" @@ -80,6 +80,8 @@ #include "i915_sysfs.h" #include "i915_trace.h" #include "i915_vgpu.h" +#include "intel_dram.h" +#include "intel_gvt.h" #include "intel_memory_region.h" #include "intel_pm.h" #include "vlv_suspend.h" @@ -238,8 +240,14 @@ static int i915_driver_modeset_probe_noirq(struct drm_i915_private *i915) intel_csr_ucode_init(i915); + ret = intel_modeset_init_noirq(i915); + if (ret) + goto cleanup_vga_client; + return 0; +cleanup_vga_client: + intel_vga_unregister(i915); out: return ret; } @@ -381,6 +389,7 @@ static void intel_detect_preproduction_hw(struct drm_i915_private *dev_priv) pre |= IS_SKL_REVID(dev_priv, 0, SKL_REVID_F0); pre |= IS_BXT_REVID(dev_priv, 0, BXT_REVID_B_LAST); pre |= IS_KBL_REVID(dev_priv, 0, KBL_REVID_A0); + pre |= IS_GLK_REVID(dev_priv, 0, GLK_REVID_A2); if (pre) { drm_err(&dev_priv->drm, "This is a pre-production stepping. " @@ -459,7 +468,6 @@ static int i915_driver_early_probe(struct drm_i915_private *dev_priv) intel_init_display_hooks(dev_priv); intel_init_clock_gating_hooks(dev_priv); intel_init_audio_hooks(dev_priv); - intel_display_crc_init(dev_priv); intel_detect_preproduction_hw(dev_priv); @@ -558,494 +566,6 @@ static void intel_sanitize_options(struct drm_i915_private *dev_priv) intel_gvt_sanitize_options(dev_priv); } -#define DRAM_TYPE_STR(type) [INTEL_DRAM_ ## type] = #type - -static const char *intel_dram_type_str(enum intel_dram_type type) -{ - static const char * const str[] = { - DRAM_TYPE_STR(UNKNOWN), - DRAM_TYPE_STR(DDR3), - DRAM_TYPE_STR(DDR4), - DRAM_TYPE_STR(LPDDR3), - DRAM_TYPE_STR(LPDDR4), - }; - - if (type >= ARRAY_SIZE(str)) - type = INTEL_DRAM_UNKNOWN; - - return str[type]; -} - -#undef DRAM_TYPE_STR - -static int intel_dimm_num_devices(const struct dram_dimm_info *dimm) -{ - return dimm->ranks * 64 / (dimm->width ?: 1); -} - -/* Returns total GB for the whole DIMM */ -static int skl_get_dimm_size(u16 val) -{ - return val & SKL_DRAM_SIZE_MASK; -} - -static int skl_get_dimm_width(u16 val) -{ - if (skl_get_dimm_size(val) == 0) - return 0; - - switch (val & SKL_DRAM_WIDTH_MASK) { - case SKL_DRAM_WIDTH_X8: - case SKL_DRAM_WIDTH_X16: - case SKL_DRAM_WIDTH_X32: - val = (val & SKL_DRAM_WIDTH_MASK) >> SKL_DRAM_WIDTH_SHIFT; - return 8 << val; - default: - MISSING_CASE(val); - return 0; - } -} - -static int skl_get_dimm_ranks(u16 val) -{ - if (skl_get_dimm_size(val) == 0) - return 0; - - val = (val & SKL_DRAM_RANK_MASK) >> SKL_DRAM_RANK_SHIFT; - - return val + 1; -} - -/* Returns total GB for the whole DIMM */ -static int cnl_get_dimm_size(u16 val) -{ - return (val & CNL_DRAM_SIZE_MASK) / 2; -} - -static int cnl_get_dimm_width(u16 val) -{ - if (cnl_get_dimm_size(val) == 0) - return 0; - - switch (val & CNL_DRAM_WIDTH_MASK) { - case CNL_DRAM_WIDTH_X8: - case CNL_DRAM_WIDTH_X16: - case CNL_DRAM_WIDTH_X32: - val = (val & CNL_DRAM_WIDTH_MASK) >> CNL_DRAM_WIDTH_SHIFT; - return 8 << val; - default: - MISSING_CASE(val); - return 0; - } -} - -static int cnl_get_dimm_ranks(u16 val) -{ - if (cnl_get_dimm_size(val) == 0) - return 0; - - val = (val & CNL_DRAM_RANK_MASK) >> CNL_DRAM_RANK_SHIFT; - - return val + 1; -} - -static bool -skl_is_16gb_dimm(const struct dram_dimm_info *dimm) -{ - /* Convert total GB to Gb per DRAM device */ - return 8 * dimm->size / (intel_dimm_num_devices(dimm) ?: 1) == 16; -} - -static void -skl_dram_get_dimm_info(struct drm_i915_private *dev_priv, - struct dram_dimm_info *dimm, - int channel, char dimm_name, u16 val) -{ - if (INTEL_GEN(dev_priv) >= 10) { - dimm->size = cnl_get_dimm_size(val); - dimm->width = cnl_get_dimm_width(val); - dimm->ranks = cnl_get_dimm_ranks(val); - } else { - dimm->size = skl_get_dimm_size(val); - dimm->width = skl_get_dimm_width(val); - dimm->ranks = skl_get_dimm_ranks(val); - } - - drm_dbg_kms(&dev_priv->drm, - "CH%u DIMM %c size: %u GB, width: X%u, ranks: %u, 16Gb DIMMs: %s\n", - channel, dimm_name, dimm->size, dimm->width, dimm->ranks, - yesno(skl_is_16gb_dimm(dimm))); -} - -static int -skl_dram_get_channel_info(struct drm_i915_private *dev_priv, - struct dram_channel_info *ch, - int channel, u32 val) -{ - skl_dram_get_dimm_info(dev_priv, &ch->dimm_l, - channel, 'L', val & 0xffff); - skl_dram_get_dimm_info(dev_priv, &ch->dimm_s, - channel, 'S', val >> 16); - - if (ch->dimm_l.size == 0 && ch->dimm_s.size == 0) { - drm_dbg_kms(&dev_priv->drm, "CH%u not populated\n", channel); - return -EINVAL; - } - - if (ch->dimm_l.ranks == 2 || ch->dimm_s.ranks == 2) - ch->ranks = 2; - else if (ch->dimm_l.ranks == 1 && ch->dimm_s.ranks == 1) - ch->ranks = 2; - else - ch->ranks = 1; - - ch->is_16gb_dimm = - skl_is_16gb_dimm(&ch->dimm_l) || - skl_is_16gb_dimm(&ch->dimm_s); - - drm_dbg_kms(&dev_priv->drm, "CH%u ranks: %u, 16Gb DIMMs: %s\n", - channel, ch->ranks, yesno(ch->is_16gb_dimm)); - - return 0; -} - -static bool -intel_is_dram_symmetric(const struct dram_channel_info *ch0, - const struct dram_channel_info *ch1) -{ - return !memcmp(ch0, ch1, sizeof(*ch0)) && - (ch0->dimm_s.size == 0 || - !memcmp(&ch0->dimm_l, &ch0->dimm_s, sizeof(ch0->dimm_l))); -} - -static int -skl_dram_get_channels_info(struct drm_i915_private *dev_priv) -{ - struct dram_info *dram_info = &dev_priv->dram_info; - struct dram_channel_info ch0 = {}, ch1 = {}; - u32 val; - int ret; - - val = I915_READ(SKL_MAD_DIMM_CH0_0_0_0_MCHBAR_MCMAIN); - ret = skl_dram_get_channel_info(dev_priv, &ch0, 0, val); - if (ret == 0) - dram_info->num_channels++; - - val = I915_READ(SKL_MAD_DIMM_CH1_0_0_0_MCHBAR_MCMAIN); - ret = skl_dram_get_channel_info(dev_priv, &ch1, 1, val); - if (ret == 0) - dram_info->num_channels++; - - if (dram_info->num_channels == 0) { - drm_info(&dev_priv->drm, - "Number of memory channels is zero\n"); - return -EINVAL; - } - - /* - * If any of the channel is single rank channel, worst case output - * will be same as if single rank memory, so consider single rank - * memory. - */ - if (ch0.ranks == 1 || ch1.ranks == 1) - dram_info->ranks = 1; - else - dram_info->ranks = max(ch0.ranks, ch1.ranks); - - if (dram_info->ranks == 0) { - drm_info(&dev_priv->drm, - "couldn't get memory rank information\n"); - return -EINVAL; - } - - dram_info->is_16gb_dimm = ch0.is_16gb_dimm || ch1.is_16gb_dimm; - - dram_info->symmetric_memory = intel_is_dram_symmetric(&ch0, &ch1); - - drm_dbg_kms(&dev_priv->drm, "Memory configuration is symmetric? %s\n", - yesno(dram_info->symmetric_memory)); - return 0; -} - -static enum intel_dram_type -skl_get_dram_type(struct drm_i915_private *dev_priv) -{ - u32 val; - - val = I915_READ(SKL_MAD_INTER_CHANNEL_0_0_0_MCHBAR_MCMAIN); - - switch (val & SKL_DRAM_DDR_TYPE_MASK) { - case SKL_DRAM_DDR_TYPE_DDR3: - return INTEL_DRAM_DDR3; - case SKL_DRAM_DDR_TYPE_DDR4: - return INTEL_DRAM_DDR4; - case SKL_DRAM_DDR_TYPE_LPDDR3: - return INTEL_DRAM_LPDDR3; - case SKL_DRAM_DDR_TYPE_LPDDR4: - return INTEL_DRAM_LPDDR4; - default: - MISSING_CASE(val); - return INTEL_DRAM_UNKNOWN; - } -} - -static int -skl_get_dram_info(struct drm_i915_private *dev_priv) -{ - struct dram_info *dram_info = &dev_priv->dram_info; - u32 mem_freq_khz, val; - int ret; - - dram_info->type = skl_get_dram_type(dev_priv); - drm_dbg_kms(&dev_priv->drm, "DRAM type: %s\n", - intel_dram_type_str(dram_info->type)); - - ret = skl_dram_get_channels_info(dev_priv); - if (ret) - return ret; - - val = I915_READ(SKL_MC_BIOS_DATA_0_0_0_MCHBAR_PCU); - mem_freq_khz = DIV_ROUND_UP((val & SKL_REQ_DATA_MASK) * - SKL_MEMORY_FREQ_MULTIPLIER_HZ, 1000); - - dram_info->bandwidth_kbps = dram_info->num_channels * - mem_freq_khz * 8; - - if (dram_info->bandwidth_kbps == 0) { - drm_info(&dev_priv->drm, - "Couldn't get system memory bandwidth\n"); - return -EINVAL; - } - - dram_info->valid = true; - return 0; -} - -/* Returns Gb per DRAM device */ -static int bxt_get_dimm_size(u32 val) -{ - switch (val & BXT_DRAM_SIZE_MASK) { - case BXT_DRAM_SIZE_4GBIT: - return 4; - case BXT_DRAM_SIZE_6GBIT: - return 6; - case BXT_DRAM_SIZE_8GBIT: - return 8; - case BXT_DRAM_SIZE_12GBIT: - return 12; - case BXT_DRAM_SIZE_16GBIT: - return 16; - default: - MISSING_CASE(val); - return 0; - } -} - -static int bxt_get_dimm_width(u32 val) -{ - if (!bxt_get_dimm_size(val)) - return 0; - - val = (val & BXT_DRAM_WIDTH_MASK) >> BXT_DRAM_WIDTH_SHIFT; - - return 8 << val; -} - -static int bxt_get_dimm_ranks(u32 val) -{ - if (!bxt_get_dimm_size(val)) - return 0; - - switch (val & BXT_DRAM_RANK_MASK) { - case BXT_DRAM_RANK_SINGLE: - return 1; - case BXT_DRAM_RANK_DUAL: - return 2; - default: - MISSING_CASE(val); - return 0; - } -} - -static enum intel_dram_type bxt_get_dimm_type(u32 val) -{ - if (!bxt_get_dimm_size(val)) - return INTEL_DRAM_UNKNOWN; - - switch (val & BXT_DRAM_TYPE_MASK) { - case BXT_DRAM_TYPE_DDR3: - return INTEL_DRAM_DDR3; - case BXT_DRAM_TYPE_LPDDR3: - return INTEL_DRAM_LPDDR3; - case BXT_DRAM_TYPE_DDR4: - return INTEL_DRAM_DDR4; - case BXT_DRAM_TYPE_LPDDR4: - return INTEL_DRAM_LPDDR4; - default: - MISSING_CASE(val); - return INTEL_DRAM_UNKNOWN; - } -} - -static void bxt_get_dimm_info(struct dram_dimm_info *dimm, - u32 val) -{ - dimm->width = bxt_get_dimm_width(val); - dimm->ranks = bxt_get_dimm_ranks(val); - - /* - * Size in register is Gb per DRAM device. Convert to total - * GB to match the way we report this for non-LP platforms. - */ - dimm->size = bxt_get_dimm_size(val) * intel_dimm_num_devices(dimm) / 8; -} - -static int -bxt_get_dram_info(struct drm_i915_private *dev_priv) -{ - struct dram_info *dram_info = &dev_priv->dram_info; - u32 dram_channels; - u32 mem_freq_khz, val; - u8 num_active_channels; - int i; - - val = I915_READ(BXT_P_CR_MC_BIOS_REQ_0_0_0); - mem_freq_khz = DIV_ROUND_UP((val & BXT_REQ_DATA_MASK) * - BXT_MEMORY_FREQ_MULTIPLIER_HZ, 1000); - - dram_channels = val & BXT_DRAM_CHANNEL_ACTIVE_MASK; - num_active_channels = hweight32(dram_channels); - - /* Each active bit represents 4-byte channel */ - dram_info->bandwidth_kbps = (mem_freq_khz * num_active_channels * 4); - - if (dram_info->bandwidth_kbps == 0) { - drm_info(&dev_priv->drm, - "Couldn't get system memory bandwidth\n"); - return -EINVAL; - } - - /* - * Now read each DUNIT8/9/10/11 to check the rank of each dimms. - */ - for (i = BXT_D_CR_DRP0_DUNIT_START; i <= BXT_D_CR_DRP0_DUNIT_END; i++) { - struct dram_dimm_info dimm; - enum intel_dram_type type; - - val = I915_READ(BXT_D_CR_DRP0_DUNIT(i)); - if (val == 0xFFFFFFFF) - continue; - - dram_info->num_channels++; - - bxt_get_dimm_info(&dimm, val); - type = bxt_get_dimm_type(val); - - drm_WARN_ON(&dev_priv->drm, type != INTEL_DRAM_UNKNOWN && - dram_info->type != INTEL_DRAM_UNKNOWN && - dram_info->type != type); - - drm_dbg_kms(&dev_priv->drm, - "CH%u DIMM size: %u GB, width: X%u, ranks: %u, type: %s\n", - i - BXT_D_CR_DRP0_DUNIT_START, - dimm.size, dimm.width, dimm.ranks, - intel_dram_type_str(type)); - - /* - * If any of the channel is single rank channel, - * worst case output will be same as if single rank - * memory, so consider single rank memory. - */ - if (dram_info->ranks == 0) - dram_info->ranks = dimm.ranks; - else if (dimm.ranks == 1) - dram_info->ranks = 1; - - if (type != INTEL_DRAM_UNKNOWN) - dram_info->type = type; - } - - if (dram_info->type == INTEL_DRAM_UNKNOWN || - dram_info->ranks == 0) { - drm_info(&dev_priv->drm, "couldn't get memory information\n"); - return -EINVAL; - } - - dram_info->valid = true; - return 0; -} - -static void -intel_get_dram_info(struct drm_i915_private *dev_priv) -{ - struct dram_info *dram_info = &dev_priv->dram_info; - int ret; - - /* - * Assume 16Gb DIMMs are present until proven otherwise. - * This is only used for the level 0 watermark latency - * w/a which does not apply to bxt/glk. - */ - dram_info->is_16gb_dimm = !IS_GEN9_LP(dev_priv); - - if (INTEL_GEN(dev_priv) < 9 || !HAS_DISPLAY(dev_priv)) - return; - - if (IS_GEN9_LP(dev_priv)) - ret = bxt_get_dram_info(dev_priv); - else - ret = skl_get_dram_info(dev_priv); - if (ret) - return; - - drm_dbg_kms(&dev_priv->drm, "DRAM bandwidth: %u kBps, channels: %u\n", - dram_info->bandwidth_kbps, - dram_info->num_channels); - - drm_dbg_kms(&dev_priv->drm, "DRAM ranks: %u, 16Gb DIMMs: %s\n", - dram_info->ranks, yesno(dram_info->is_16gb_dimm)); -} - -static u32 gen9_edram_size_mb(struct drm_i915_private *dev_priv, u32 cap) -{ - static const u8 ways[8] = { 4, 8, 12, 16, 16, 16, 16, 16 }; - static const u8 sets[4] = { 1, 1, 2, 2 }; - - return EDRAM_NUM_BANKS(cap) * - ways[EDRAM_WAYS_IDX(cap)] * - sets[EDRAM_SETS_IDX(cap)]; -} - -static void edram_detect(struct drm_i915_private *dev_priv) -{ - u32 edram_cap = 0; - - if (!(IS_HASWELL(dev_priv) || - IS_BROADWELL(dev_priv) || - INTEL_GEN(dev_priv) >= 9)) - return; - - edram_cap = __raw_uncore_read32(&dev_priv->uncore, HSW_EDRAM_CAP); - - /* NB: We can't write IDICR yet because we don't have gt funcs set up */ - - if (!(edram_cap & EDRAM_ENABLED)) - return; - - /* - * The needed capability bits for size calculation are not there with - * pre gen9 so return 128MB always. - */ - if (INTEL_GEN(dev_priv) < 9) - dev_priv->edram_size_mb = 128; - else - dev_priv->edram_size_mb = - gen9_edram_size_mb(dev_priv, edram_cap); - - dev_info(dev_priv->drm.dev, - "Found %uMB of eDRAM\n", dev_priv->edram_size_mb); -} - /** * i915_driver_hw_probe - setup state requiring device access * @dev_priv: device private @@ -1089,7 +609,7 @@ static int i915_driver_hw_probe(struct drm_i915_private *dev_priv) intel_sanitize_options(dev_priv); /* needs to be done before ggtt probe */ - edram_detect(dev_priv); + intel_dram_edram_detect(dev_priv); i915_perf_init(dev_priv); @@ -1191,7 +711,7 @@ static int i915_driver_hw_probe(struct drm_i915_private *dev_priv) * Fill the dram structure to get the system raw bandwidth and * dram info. This will be used for memory latency calculation. */ - intel_get_dram_info(dev_priv); + intel_dram_detect(dev_priv); intel_bw_init_hw(dev_priv); @@ -1240,12 +760,7 @@ static void i915_driver_register(struct drm_i915_private *dev_priv) i915_gem_driver_register(dev_priv); i915_pmu_register(dev_priv); - /* - * Notify a valid surface after modesetting, - * when running inside a VM. - */ - if (intel_vgpu_active(dev_priv)) - I915_WRITE(vgtif_reg(display_ready), VGT_DRV_DISPLAY_READY); + intel_vgpu_register(dev_priv); /* Reveal our presence to userspace */ if (drm_dev_register(dev, 0) == 0) { @@ -1375,8 +890,6 @@ i915_driver_create(struct pci_dev *pdev, const struct pci_device_id *ent) return ERR_PTR(err); } - i915->drm.dev_private = i915; - i915->drm.pdev = pdev; pci_set_drvdata(pdev, i915); @@ -1454,7 +967,7 @@ int i915_driver_probe(struct pci_dev *pdev, const struct pci_device_id *ent) disable_rpm_wakeref_asserts(&i915->runtime_pm); - i915_detect_vgpu(i915); + intel_vgpu_detect(i915); ret = i915_driver_mmio_probe(i915); if (ret < 0) @@ -2245,7 +1758,7 @@ static const struct file_operations i915_driver_fops = { .mmap = i915_gem_mmap, .poll = drm_poll, .read = drm_read, - .compat_ioctl = i915_compat_ioctl, + .compat_ioctl = i915_ioc32_compat_ioctl, .llseek = noop_llseek, }; diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index ea13fc0b409b..1f5b9a584f71 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -59,7 +59,6 @@ #include <drm/drm_connector.h> #include <drm/i915_mei_hdcp_interface.h> -#include "i915_fixed.h" #include "i915_params.h" #include "i915_reg.h" #include "i915_utils.h" @@ -105,18 +104,23 @@ #include "intel_region_lmem.h" -#include "intel_gvt.h" - /* General customization: */ #define DRIVER_NAME "i915" #define DRIVER_DESC "Intel Graphics" -#define DRIVER_DATE "20200225" -#define DRIVER_TIMESTAMP 1582656081 +#define DRIVER_DATE "20200313" +#define DRIVER_TIMESTAMP 1584144591 struct drm_i915_gem_object; +/* + * The code assumes that the hpd_pins below have consecutive values and + * starting with HPD_PORT_A, the HPD pin associated with any port can be + * retrieved by adding the corresponding port (or phy) enum value to + * HPD_PORT_A in most cases. For example: + * HPD_PORT_C = HPD_PORT_A + PHY_C - PHY_A + */ enum hpd_pin { HPD_NONE = 0, HPD_TV = HPD_NONE, /* TV is known to be unreliable */ @@ -732,14 +736,6 @@ enum intel_ddb_partitioning { INTEL_DDB_PART_5_6, /* IVB+ */ }; -struct intel_wm_level { - bool enable; - u32 pri_val; - u32 spr_val; - u32 cur_val; - u32 fbc_val; -}; - struct ilk_wm_values { u32 wm_pipe[3]; u32 wm_lp[3]; @@ -798,56 +794,6 @@ static inline bool skl_ddb_entry_equal(const struct skl_ddb_entry *e1, return false; } -struct skl_wm_level { - u16 min_ddb_alloc; - u16 plane_res_b; - u8 plane_res_l; - bool plane_en; - bool ignore_lines; -}; - -/* Stores plane specific WM parameters */ -struct skl_wm_params { - bool x_tiled, y_tiled; - bool rc_surface; - bool is_planar; - u32 width; - u8 cpp; - u32 plane_pixel_rate; - u32 y_min_scanlines; - u32 plane_bytes_per_line; - uint_fixed_16_16_t plane_blocks_per_line; - uint_fixed_16_16_t y_tile_minimum; - u32 linetime_us; - u32 dbuf_block_size; -}; - -enum intel_pipe_crc_source { - INTEL_PIPE_CRC_SOURCE_NONE, - INTEL_PIPE_CRC_SOURCE_PLANE1, - INTEL_PIPE_CRC_SOURCE_PLANE2, - INTEL_PIPE_CRC_SOURCE_PLANE3, - INTEL_PIPE_CRC_SOURCE_PLANE4, - INTEL_PIPE_CRC_SOURCE_PLANE5, - INTEL_PIPE_CRC_SOURCE_PLANE6, - INTEL_PIPE_CRC_SOURCE_PLANE7, - INTEL_PIPE_CRC_SOURCE_PIPE, - /* TV/DP on pre-gen5/vlv can't use the pipe source. */ - INTEL_PIPE_CRC_SOURCE_TV, - INTEL_PIPE_CRC_SOURCE_DP_B, - INTEL_PIPE_CRC_SOURCE_DP_C, - INTEL_PIPE_CRC_SOURCE_DP_D, - INTEL_PIPE_CRC_SOURCE_AUTO, - INTEL_PIPE_CRC_SOURCE_MAX, -}; - -#define INTEL_PIPE_CRC_ENTRIES_NR 128 -struct intel_pipe_crc { - spinlock_t lock; - int skipped; - enum intel_pipe_crc_source source; -}; - struct i915_frontbuffer_tracking { spinlock_t lock; @@ -865,13 +811,6 @@ struct i915_virtual_gpu { u32 caps; }; -/* used in computing the new watermarks state */ -struct intel_wm_config { - unsigned int num_pipes_active; - bool sprites_enabled; - bool sprites_scaled; -}; - struct intel_cdclk_config { unsigned int cdclk, vco, ref, bypass; u8 voltage_level; @@ -1043,21 +982,24 @@ struct drm_i915_private { struct intel_crtc *plane_to_crtc_mapping[I915_MAX_PIPES]; struct intel_crtc *pipe_to_crtc_mapping[I915_MAX_PIPES]; -#ifdef CONFIG_DEBUG_FS - struct intel_pipe_crc pipe_crc[I915_MAX_PIPES]; -#endif + /** + * dpll and cdclk state is protected by connection_mutex + * dpll.lock serializes intel_{prepare,enable,disable}_shared_dpll. + * Must be global rather than per dpll, because on some platforms plls + * share registers. + */ + struct { + struct mutex lock; - /* dpll and cdclk state is protected by connection_mutex */ - int num_shared_dpll; - struct intel_shared_dpll shared_dplls[I915_NUM_PLLS]; - const struct intel_dpll_mgr *dpll_mgr; + int num_shared_dpll; + struct intel_shared_dpll shared_dplls[I915_NUM_PLLS]; + const struct intel_dpll_mgr *mgr; - /* - * dpll_lock serializes intel_{prepare,enable,disable}_shared_dpll. - * Must be global rather than per dpll, because on some platforms - * plls share registers. - */ - struct mutex dpll_lock; + struct { + int nssc; + int ssc; + } ref_clks; + } dpll; struct list_head global_obj_list; @@ -1078,8 +1020,6 @@ struct drm_i915_private { struct work_struct free_work; } atomic_helper; - u16 orig_clock; - bool mchbar_need_disable; struct intel_l3_parity l3_parity; @@ -1274,16 +1214,6 @@ struct drm_i915_private { */ }; -struct dram_dimm_info { - u8 size, width, ranks; -}; - -struct dram_channel_info { - struct dram_dimm_info dimm_l, dimm_s; - u8 ranks; - bool is_16gb_dimm; -}; - static inline struct drm_i915_private *to_i915(const struct drm_device *dev) { return container_of(dev, struct drm_i915_private, drm); @@ -1554,6 +1484,8 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915, #define GLK_REVID_A0 0x0 #define GLK_REVID_A1 0x1 +#define GLK_REVID_A2 0x2 +#define GLK_REVID_B0 0x3 #define IS_GLK_REVID(dev_priv, since, until) \ (IS_GEMINILAKE(dev_priv) && IS_REVID(dev_priv, since, until)) @@ -1737,11 +1669,6 @@ intel_ggtt_update_needs_vtd_wa(struct drm_i915_private *dev_priv) } /* i915_drv.c */ -#ifdef CONFIG_COMPAT -long i915_compat_ioctl(struct file *filp, unsigned int cmd, unsigned long arg); -#else -#define i915_compat_ioctl NULL -#endif extern const struct dev_pm_ops i915_pm_ops; int i915_driver_probe(struct pci_dev *pdev, const struct pci_device_id *ent); @@ -1750,16 +1677,6 @@ void i915_driver_remove(struct drm_i915_private *i915); int i915_resume_switcheroo(struct drm_i915_private *i915); int i915_suspend_switcheroo(struct drm_i915_private *i915, pm_message_t state); -static inline bool intel_gvt_active(struct drm_i915_private *dev_priv) -{ - return dev_priv->gvt; -} - -static inline bool intel_vgpu_active(struct drm_i915_private *dev_priv) -{ - return dev_priv->vgpu.active; -} - int i915_getparam_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv); @@ -1824,12 +1741,6 @@ int i915_gem_object_unbind(struct drm_i915_gem_object *obj, void i915_gem_runtime_suspend(struct drm_i915_private *dev_priv); -static inline int __must_check -i915_mutex_lock_interruptible(struct drm_device *dev) -{ - return mutex_lock_interruptible(&dev->struct_mutex); -} - int i915_gem_dumb_create(struct drm_file *file_priv, struct drm_device *dev, struct drm_mode_create_dumb *args); diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index d92cf966fa3f..ca5420012a22 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -26,7 +26,6 @@ */ #include <drm/drm_vma_manager.h> -#include <drm/i915_drm.h> #include <linux/dma-fence-array.h> #include <linux/kthread.h> #include <linux/dma-resv.h> diff --git a/drivers/gpu/drm/i915/i915_gem_evict.c b/drivers/gpu/drm/i915/i915_gem_evict.c index 0697bedebeef..4518b9b35c3d 100644 --- a/drivers/gpu/drm/i915/i915_gem_evict.c +++ b/drivers/gpu/drm/i915/i915_gem_evict.c @@ -26,8 +26,6 @@ * */ -#include <drm/i915_drm.h> - #include "gem/i915_gem_context.h" #include "gt/intel_gt_requests.h" @@ -292,7 +290,8 @@ int i915_gem_evict_for_node(struct i915_address_space *vm, GEM_BUG_ON(!drm_mm_node_allocated(node)); vma = container_of(node, typeof(*vma), node); - /* If we are using coloring to insert guard pages between + /* + * If we are using coloring to insert guard pages between * different cache domains within the address space, we have * to check whether the objects on either side of our range * abutt and conflict. If they are in conflict, then we evict @@ -309,22 +308,18 @@ int i915_gem_evict_for_node(struct i915_address_space *vm, } } - if (flags & PIN_NONBLOCK && - (i915_vma_is_pinned(vma) || i915_vma_is_active(vma))) { + if (i915_vma_is_pinned(vma)) { ret = -ENOSPC; break; } - /* Overlap of objects in the same batch? */ - if (i915_vma_is_pinned(vma)) { + if (flags & PIN_NONBLOCK && i915_vma_is_active(vma)) { ret = -ENOSPC; - if (vma->exec_flags && - *vma->exec_flags & EXEC_OBJECT_PINNED) - ret = -EINVAL; break; } - /* Never show fear in the face of dragons! + /* + * Never show fear in the face of dragons! * * We cannot directly remove this node from within this * iterator and as with i915_gem_evict_something() we employ diff --git a/drivers/gpu/drm/i915/i915_gem_fence_reg.c b/drivers/gpu/drm/i915/i915_gem_fence_reg.c index 049cd3785347..d152b648c73c 100644 --- a/drivers/gpu/drm/i915/i915_gem_fence_reg.c +++ b/drivers/gpu/drm/i915/i915_gem_fence_reg.c @@ -21,10 +21,9 @@ * IN THE SOFTWARE. */ -#include <drm/i915_drm.h> - #include "i915_drv.h" #include "i915_scatterlist.h" +#include "i915_pvinfo.h" #include "i915_vgpu.h" /** diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c index e7834fa1e0ac..cb43381b0d37 100644 --- a/drivers/gpu/drm/i915/i915_gem_gtt.c +++ b/drivers/gpu/drm/i915/i915_gem_gtt.c @@ -15,8 +15,6 @@ #include <asm/set_memory.h> #include <asm/smp.h> -#include <drm/i915_drm.h> - #include "display/intel_frontbuffer.h" #include "gt/intel_gt.h" #include "gt/intel_gt_requests.h" diff --git a/drivers/gpu/drm/i915/i915_ioc32.c b/drivers/gpu/drm/i915/i915_ioc32.c index c1007245f46d..8e45ca3d2ede 100644 --- a/drivers/gpu/drm/i915/i915_ioc32.c +++ b/drivers/gpu/drm/i915/i915_ioc32.c @@ -28,9 +28,10 @@ */ #include <linux/compat.h> -#include <drm/i915_drm.h> #include <drm/drm_ioctl.h> + #include "i915_drv.h" +#include "i915_ioc32.h" struct drm_i915_getparam32 { s32 param; @@ -67,7 +68,7 @@ static drm_ioctl_compat_t *i915_compat_ioctls[] = { }; /** - * i915_compat_ioctl - handle the mistakes of the past + * i915_ioc32_compat_ioctl - handle the mistakes of the past * @filp: the file pointer * @cmd: the ioctl command (and encoded flags) * @arg: the ioctl argument (from userspace) @@ -75,7 +76,7 @@ static drm_ioctl_compat_t *i915_compat_ioctls[] = { * Called whenever a 32-bit process running under a 64-bit kernel * performs an ioctl on /dev/dri/card<n>. */ -long i915_compat_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) +long i915_ioc32_compat_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) { unsigned int nr = DRM_IOCTL_NR(cmd); drm_ioctl_compat_t *fn = NULL; diff --git a/drivers/gpu/drm/i915/i915_ioc32.h b/drivers/gpu/drm/i915/i915_ioc32.h new file mode 100644 index 000000000000..40dcd55ca213 --- /dev/null +++ b/drivers/gpu/drm/i915/i915_ioc32.h @@ -0,0 +1,17 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2020 Intel Corporation + */ + +#ifndef __I915_IOC32_H__ +#define __I915_IOC32_H__ + +#ifdef CONFIG_COMPAT +struct file; +long i915_ioc32_compat_ioctl(struct file *filp, unsigned int cmd, + unsigned long arg); +#else +#define i915_ioc32_compat_ioctl NULL +#endif + +#endif /* __I915_IOC32_H__ */ diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c index 5d1b11346145..9f0653cf0510 100644 --- a/drivers/gpu/drm/i915/i915_irq.c +++ b/drivers/gpu/drm/i915/i915_irq.c @@ -34,7 +34,6 @@ #include <drm/drm_drv.h> #include <drm/drm_irq.h> -#include <drm/i915_drm.h> #include "display/intel_display_types.h" #include "display/intel_fifo_underrun.h" @@ -169,6 +168,14 @@ static const u32 hpd_tgp[HPD_NUM_PINS] = { [HPD_PORT_I] = SDE_TC_HOTPLUG_ICP(PORT_TC6), }; +static void +intel_handle_vblank(struct drm_i915_private *dev_priv, enum pipe pipe) +{ + struct intel_crtc *crtc = intel_get_crtc_for_pipe(dev_priv, pipe); + + drm_crtc_handle_vblank(&crtc->base); +} + void gen3_irq_reset(struct intel_uncore *uncore, i915_reg_t imr, i915_reg_t iir, i915_reg_t ier) { @@ -1210,8 +1217,8 @@ static void display_pipe_crc_irq_handler(struct drm_i915_private *dev_priv, u32 crc2, u32 crc3, u32 crc4) { - struct intel_pipe_crc *pipe_crc = &dev_priv->pipe_crc[pipe]; struct intel_crtc *crtc = intel_get_crtc_for_pipe(dev_priv, pipe); + struct intel_pipe_crc *pipe_crc = &crtc->pipe_crc; u32 crcs[5] = { crc0, crc1, crc2, crc3, crc4 }; trace_intel_pipe_crc(crtc, crcs); @@ -1374,7 +1381,7 @@ static void i8xx_pipestat_irq_handler(struct drm_i915_private *dev_priv, for_each_pipe(dev_priv, pipe) { if (pipe_stats[pipe] & PIPE_VBLANK_INTERRUPT_STATUS) - drm_handle_vblank(&dev_priv->drm, pipe); + intel_handle_vblank(dev_priv, pipe); if (pipe_stats[pipe] & PIPE_CRC_DONE_INTERRUPT_STATUS) i9xx_pipe_crc_irq_handler(dev_priv, pipe); @@ -1392,7 +1399,7 @@ static void i915_pipestat_irq_handler(struct drm_i915_private *dev_priv, for_each_pipe(dev_priv, pipe) { if (pipe_stats[pipe] & PIPE_VBLANK_INTERRUPT_STATUS) - drm_handle_vblank(&dev_priv->drm, pipe); + intel_handle_vblank(dev_priv, pipe); if (pipe_stats[pipe] & PIPE_LEGACY_BLC_EVENT_STATUS) blc_event = true; @@ -1416,7 +1423,7 @@ static void i965_pipestat_irq_handler(struct drm_i915_private *dev_priv, for_each_pipe(dev_priv, pipe) { if (pipe_stats[pipe] & PIPE_START_VBLANK_INTERRUPT_STATUS) - drm_handle_vblank(&dev_priv->drm, pipe); + intel_handle_vblank(dev_priv, pipe); if (pipe_stats[pipe] & PIPE_LEGACY_BLC_EVENT_STATUS) blc_event = true; @@ -1442,7 +1449,7 @@ static void valleyview_pipestat_irq_handler(struct drm_i915_private *dev_priv, for_each_pipe(dev_priv, pipe) { if (pipe_stats[pipe] & PIPE_START_VBLANK_INTERRUPT_STATUS) - drm_handle_vblank(&dev_priv->drm, pipe); + intel_handle_vblank(dev_priv, pipe); if (pipe_stats[pipe] & PIPE_CRC_DONE_INTERRUPT_STATUS) i9xx_pipe_crc_irq_handler(dev_priv, pipe); @@ -1749,11 +1756,12 @@ static void ibx_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir) if (pch_iir & SDE_POISON) drm_err(&dev_priv->drm, "PCH poison interrupt\n"); - if (pch_iir & SDE_FDI_MASK) + if (pch_iir & SDE_FDI_MASK) { for_each_pipe(dev_priv, pipe) drm_dbg(&dev_priv->drm, " pipe %c FDI IIR: 0x%08x\n", pipe_name(pipe), I915_READ(FDI_RX_IIR(pipe))); + } if (pch_iir & (SDE_TRANSB_CRC_DONE | SDE_TRANSA_CRC_DONE)) drm_dbg(&dev_priv->drm, "PCH transcoder CRC done interrupt\n"); @@ -1833,11 +1841,12 @@ static void cpt_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir) if (pch_iir & SDE_AUDIO_CP_CHG_CPT) drm_dbg(&dev_priv->drm, "Audio CP change interrupt\n"); - if (pch_iir & SDE_FDI_MASK_CPT) + if (pch_iir & SDE_FDI_MASK_CPT) { for_each_pipe(dev_priv, pipe) drm_dbg(&dev_priv->drm, " pipe %c FDI IIR: 0x%08x\n", pipe_name(pipe), I915_READ(FDI_RX_IIR(pipe))); + } if (pch_iir & SDE_ERROR_CPT) cpt_serr_int_handler(dev_priv); @@ -1978,7 +1987,7 @@ static void ilk_display_irq_handler(struct drm_i915_private *dev_priv, for_each_pipe(dev_priv, pipe) { if (de_iir & DE_PIPE_VBLANK(pipe)) - drm_handle_vblank(&dev_priv->drm, pipe); + intel_handle_vblank(dev_priv, pipe); if (de_iir & DE_PIPE_FIFO_UNDERRUN(pipe)) intel_cpu_fifo_underrun_irq_handler(dev_priv, pipe); @@ -2031,7 +2040,7 @@ static void ivb_display_irq_handler(struct drm_i915_private *dev_priv, for_each_pipe(dev_priv, pipe) { if (de_iir & (DE_PIPE_VBLANK_IVB(pipe))) - drm_handle_vblank(&dev_priv->drm, pipe); + intel_handle_vblank(dev_priv, pipe); } /* check event from PCH */ @@ -2344,7 +2353,7 @@ gen8_de_irq_handler(struct drm_i915_private *dev_priv, u32 master_ctl) I915_WRITE(GEN8_DE_PIPE_IIR(pipe), iir); if (iir & GEN8_PIPE_VBLANK) - drm_handle_vblank(&dev_priv->drm, pipe); + intel_handle_vblank(dev_priv, pipe); if (iir & GEN8_PIPE_CDCLK_CRC_DONE) hsw_pipe_crc_irq_handler(dev_priv, pipe); diff --git a/drivers/gpu/drm/i915/i915_pci.c b/drivers/gpu/drm/i915/i915_pci.c index 1d678aa7d420..2c80a0194c80 100644 --- a/drivers/gpu/drm/i915/i915_pci.c +++ b/drivers/gpu/drm/i915/i915_pci.c @@ -26,6 +26,7 @@ #include <linux/vga_switcheroo.h> #include <drm/drm_drv.h> +#include <drm/i915_pciids.h> #include "display/intel_fbdev.h" @@ -822,7 +823,6 @@ static const struct intel_device_info tgl_info = { GEN12_FEATURES, PLATFORM(INTEL_TIGERLAKE), .pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C) | BIT(PIPE_D), - .require_force_probe = 1, .display.has_modular_fia = 1, .engine_mask = BIT(RCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS0) | BIT(VCS2), diff --git a/drivers/gpu/drm/i915/i915_perf.c b/drivers/gpu/drm/i915/i915_perf.c index e34c79df6ebc..551be589d6f4 100644 --- a/drivers/gpu/drm/i915/i915_perf.c +++ b/drivers/gpu/drm/i915/i915_perf.c @@ -1405,8 +1405,10 @@ static void i915_oa_stream_destroy(struct i915_perf_stream *stream) /* * Unset exclusive_stream first, it will be checked while disabling * the metric set on gen8+. + * + * See i915_oa_init_reg_state() and lrc_configure_all_contexts() */ - perf->exclusive_stream = NULL; + WRITE_ONCE(perf->exclusive_stream, NULL); perf->ops.disable_metric_set(stream); free_oa_buffer(stream); @@ -1970,9 +1972,10 @@ out: return i915_vma_get(oa_bo->vma); } -static int emit_oa_config(struct i915_perf_stream *stream, - struct i915_oa_config *oa_config, - struct intel_context *ce) +static struct i915_request * +emit_oa_config(struct i915_perf_stream *stream, + struct i915_oa_config *oa_config, + struct intel_context *ce) { struct i915_request *rq; struct i915_vma *vma; @@ -1980,7 +1983,7 @@ static int emit_oa_config(struct i915_perf_stream *stream, vma = get_oa_vma(stream, oa_config); if (IS_ERR(vma)) - return PTR_ERR(vma); + return ERR_CAST(vma); err = i915_vma_pin(vma, 0, 0, PIN_GLOBAL | PIN_HIGH); if (err) @@ -2005,13 +2008,17 @@ static int emit_oa_config(struct i915_perf_stream *stream, err = rq->engine->emit_bb_start(rq, vma->node.start, 0, I915_DISPATCH_SECURE); + if (err) + goto err_add_request; + + i915_request_get(rq); err_add_request: i915_request_add(rq); err_vma_unpin: i915_vma_unpin(vma); err_vma_put: i915_vma_put(vma); - return err; + return err ? ERR_PTR(err) : rq; } static struct intel_context *oa_context(struct i915_perf_stream *stream) @@ -2019,7 +2026,8 @@ static struct intel_context *oa_context(struct i915_perf_stream *stream) return stream->pinned_ctx ?: stream->engine->kernel_context; } -static int hsw_enable_metric_set(struct i915_perf_stream *stream) +static struct i915_request * +hsw_enable_metric_set(struct i915_perf_stream *stream) { struct intel_uncore *uncore = stream->uncore; @@ -2194,7 +2202,9 @@ static int gen8_modify_self(struct intel_context *ce, struct i915_request *rq; int err; + intel_engine_pm_get(ce->engine); rq = i915_request_create(ce); + intel_engine_pm_put(ce->engine); if (IS_ERR(rq)) return PTR_ERR(rq); @@ -2422,7 +2432,8 @@ static int lrc_configure_all_contexts(struct i915_perf_stream *stream, return oa_configure_all_contexts(stream, regs, ARRAY_SIZE(regs)); } -static int gen8_enable_metric_set(struct i915_perf_stream *stream) +static struct i915_request * +gen8_enable_metric_set(struct i915_perf_stream *stream) { struct intel_uncore *uncore = stream->uncore; struct i915_oa_config *oa_config = stream->oa_config; @@ -2464,7 +2475,7 @@ static int gen8_enable_metric_set(struct i915_perf_stream *stream) */ ret = lrc_configure_all_contexts(stream, oa_config); if (ret) - return ret; + return ERR_PTR(ret); return emit_oa_config(stream, oa_config, oa_context(stream)); } @@ -2476,7 +2487,8 @@ static u32 oag_report_ctx_switches(const struct i915_perf_stream *stream) 0 : GEN12_OAG_OA_DEBUG_DISABLE_CTX_SWITCH_REPORTS); } -static int gen12_enable_metric_set(struct i915_perf_stream *stream) +static struct i915_request * +gen12_enable_metric_set(struct i915_perf_stream *stream) { struct intel_uncore *uncore = stream->uncore; struct i915_oa_config *oa_config = stream->oa_config; @@ -2507,7 +2519,7 @@ static int gen12_enable_metric_set(struct i915_perf_stream *stream) */ ret = gen12_configure_all_contexts(stream, oa_config); if (ret) - return ret; + return ERR_PTR(ret); /* * For Gen12, performance counters are context @@ -2517,7 +2529,7 @@ static int gen12_enable_metric_set(struct i915_perf_stream *stream) if (stream->ctx) { ret = gen12_configure_oar_context(stream, true); if (ret) - return ret; + return ERR_PTR(ret); } return emit_oa_config(stream, oa_config, oa_context(stream)); @@ -2688,6 +2700,14 @@ static void gen12_oa_disable(struct i915_perf_stream *stream) 50)) drm_err(&stream->perf->i915->drm, "wait for OA to be disabled timed out\n"); + + intel_uncore_write(uncore, GEN12_OA_TLB_INV_CR, 1); + if (intel_wait_for_register(uncore, + GEN12_OA_TLB_INV_CR, + 1, 0, + 50)) + drm_err(&stream->perf->i915->drm, + "wait for OA tlb invalidate timed out\n"); } /** @@ -2715,6 +2735,20 @@ static const struct i915_perf_stream_ops i915_oa_stream_ops = { .read = i915_oa_read, }; +static int i915_perf_stream_enable_sync(struct i915_perf_stream *stream) +{ + struct i915_request *rq; + + rq = stream->perf->ops.enable_metric_set(stream); + if (IS_ERR(rq)) + return PTR_ERR(rq); + + i915_request_wait(rq, 0, MAX_SCHEDULE_TIMEOUT); + i915_request_put(rq); + + return 0; +} + /** * i915_oa_stream_init - validate combined props for OA stream and init * @stream: An i915 perf stream @@ -2847,9 +2881,9 @@ static int i915_oa_stream_init(struct i915_perf_stream *stream, goto err_oa_buf_alloc; stream->ops = &i915_oa_stream_ops; - perf->exclusive_stream = stream; + WRITE_ONCE(perf->exclusive_stream, stream); - ret = perf->ops.enable_metric_set(stream); + ret = i915_perf_stream_enable_sync(stream); if (ret) { DRM_DEBUG("Unable to enable metric set\n"); goto err_enable; @@ -2867,7 +2901,7 @@ static int i915_oa_stream_init(struct i915_perf_stream *stream, return 0; err_enable: - perf->exclusive_stream = NULL; + WRITE_ONCE(perf->exclusive_stream, NULL); perf->ops.disable_metric_set(stream); free_oa_buffer(stream); @@ -2893,12 +2927,11 @@ void i915_oa_init_reg_state(const struct intel_context *ce, { struct i915_perf_stream *stream; - /* perf.exclusive_stream serialised by lrc_configure_all_contexts() */ - if (engine->class != RENDER_CLASS) return; - stream = engine->i915->perf.exclusive_stream; + /* perf.exclusive_stream serialised by lrc_configure_all_contexts() */ + stream = READ_ONCE(engine->i915->perf.exclusive_stream); /* * For gen12, only CTX_R_PWR_CLK_STATE needs update, but the caller * is already doing that, so nothing to be done for gen12 here. @@ -3167,7 +3200,7 @@ static long i915_perf_config_locked(struct i915_perf_stream *stream, return -EINVAL; if (config != stream->oa_config) { - int err; + struct i915_request *rq; /* * If OA is bound to a specific context, emit the @@ -3178,11 +3211,13 @@ static long i915_perf_config_locked(struct i915_perf_stream *stream, * When set globally, we use a low priority kernel context, * so it will effectively take effect when idle. */ - err = emit_oa_config(stream, config, oa_context(stream)); - if (err == 0) + rq = emit_oa_config(stream, config, oa_context(stream)); + if (!IS_ERR(rq)) { config = xchg(&stream->oa_config, config); - else - ret = err; + i915_request_put(rq); + } else { + ret = PTR_ERR(rq); + } } i915_oa_config_put(config); diff --git a/drivers/gpu/drm/i915/i915_perf_types.h b/drivers/gpu/drm/i915/i915_perf_types.h index 45e581455f5d..a0e22f00f6cf 100644 --- a/drivers/gpu/drm/i915/i915_perf_types.h +++ b/drivers/gpu/drm/i915/i915_perf_types.h @@ -339,7 +339,8 @@ struct i915_oa_ops { * counter reports being sampled. May apply system constraints such as * disabling EU clock gating as required. */ - int (*enable_metric_set)(struct i915_perf_stream *stream); + struct i915_request * + (*enable_metric_set)(struct i915_perf_stream *stream); /** * @disable_metric_set: Remove system constraints associated with using diff --git a/drivers/gpu/drm/i915/i915_pmu.h b/drivers/gpu/drm/i915/i915_pmu.h index f1d6cad0d7d5..941f0c14037c 100644 --- a/drivers/gpu/drm/i915/i915_pmu.h +++ b/drivers/gpu/drm/i915/i915_pmu.h @@ -10,7 +10,7 @@ #include <linux/hrtimer.h> #include <linux/perf_event.h> #include <linux/spinlock_types.h> -#include <drm/i915_drm.h> +#include <uapi/drm/i915_drm.h> struct drm_i915_private; diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index f45b5e86ec63..59e64acc2c56 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -693,6 +693,8 @@ static inline bool i915_mmio_reg_valid(i915_reg_t reg) #define OABUFFER_SIZE_8M (6 << 3) #define OABUFFER_SIZE_16M (7 << 3) +#define GEN12_OA_TLB_INV_CR _MMIO(0xceec) + /* Gen12 OAR unit */ #define GEN12_OAR_OACONTROL _MMIO(0x2960) #define GEN12_OAR_OACONTROL_COUNTER_FORMAT_SHIFT 1 @@ -3285,6 +3287,7 @@ static inline bool i915_mmio_reg_valid(i915_reg_t reg) /* Framebuffer compression for Ivybridge */ #define IVB_FBC_RT_BASE _MMIO(0x7020) +#define IVB_FBC_RT_BASE_UPPER _MMIO(0x7024) #define IPS_CTL _MMIO(0x43408) #define IPS_ENABLE (1 << 31) @@ -4860,16 +4863,6 @@ enum { #define _PP_STATUS 0x61200 #define PP_STATUS(pps_idx) _MMIO_PPS(pps_idx, _PP_STATUS) #define PP_ON REG_BIT(31) - -#define _PP_CONTROL_1 0xc7204 -#define _PP_CONTROL_2 0xc7304 -#define ICP_PP_CONTROL(x) _MMIO(((x) == 1) ? _PP_CONTROL_1 : \ - _PP_CONTROL_2) -#define POWER_CYCLE_DELAY_MASK REG_GENMASK(8, 4) -#define VDD_OVERRIDE_FORCE REG_BIT(3) -#define BACKLIGHT_ENABLE REG_BIT(2) -#define PWR_DOWN_ON_RESET REG_BIT(1) -#define PWR_STATE_TARGET REG_BIT(0) /* * Indicates that all dependencies of the panel are on: * @@ -5880,7 +5873,6 @@ enum { #define _PIPEAGCMAX 0x70010 #define _PIPEBGCMAX 0x71010 -#define PIPEGCMAX_RGB_MASK REG_GENMASK(15, 0) #define PIPEGCMAX(pipe, i) _MMIO_PIPE2(pipe, _PIPEAGCMAX + (i) * 4) #define _PIPE_MISC_A 0x70030 @@ -5889,6 +5881,7 @@ enum { #define PIPEMISC_YUV420_MODE_FULL_BLEND (1 << 26) /* glk+ */ #define PIPEMISC_HDR_MODE_PRECISION (1 << 23) /* icl+ */ #define PIPEMISC_OUTPUT_COLORSPACE_YUV (1 << 11) +#define PIPEMISC_PIXEL_ROUNDING_TRUNC REG_BIT(8) /* tgl+ */ #define PIPEMISC_DITHER_BPC_MASK (7 << 5) #define PIPEMISC_DITHER_8_BPC (0 << 5) #define PIPEMISC_DITHER_10_BPC (1 << 5) @@ -9149,14 +9142,19 @@ enum { #define THROTTLE_12_5 (7 << 2) #define DISABLE_EARLY_EOT (1 << 1) -#define GEN7_ROW_CHICKEN2 _MMIO(0xe4f4) -#define GEN12_DISABLE_EARLY_READ BIT(14) +#define GEN7_ROW_CHICKEN2 _MMIO(0xe4f4) +#define GEN12_DISABLE_EARLY_READ REG_BIT(14) +#define GEN12_PUSH_CONST_DEREF_HOLD_DIS REG_BIT(8) #define GEN7_ROW_CHICKEN2_GT2 _MMIO(0xf4f4) #define DOP_CLOCK_GATING_DISABLE (1 << 0) #define PUSH_CONSTANT_DEREF_DISABLE (1 << 8) #define GEN11_TDL_CLOCK_GATING_FIX_DISABLE (1 << 1) +#define GEN9_ROW_CHICKEN4 _MMIO(0xe48c) +#define GEN12_DISABLE_TDL_PUSH REG_BIT(9) +#define GEN11_DIS_PICK_2ND_EU REG_BIT(7) + #define HSW_ROW_CHICKEN3 _MMIO(0xe49c) #define HSW_ROW_CHICKEN3_L3_GLOBAL_ATOMICS_DISABLE (1 << 6) @@ -9257,6 +9255,10 @@ enum { #define AUD_CONFIG_PIXEL_CLOCK_HDMI_74250 (7 << 16) #define AUD_CONFIG_PIXEL_CLOCK_HDMI_148352 (8 << 16) #define AUD_CONFIG_PIXEL_CLOCK_HDMI_148500 (9 << 16) +#define AUD_CONFIG_PIXEL_CLOCK_HDMI_296703 (10 << 16) +#define AUD_CONFIG_PIXEL_CLOCK_HDMI_297000 (11 << 16) +#define AUD_CONFIG_PIXEL_CLOCK_HDMI_593407 (12 << 16) +#define AUD_CONFIG_PIXEL_CLOCK_HDMI_594000 (13 << 16) #define AUD_CONFIG_DISABLE_NCTS (1 << 3) /* HSW Audio */ diff --git a/drivers/gpu/drm/i915/i915_request.c b/drivers/gpu/drm/i915/i915_request.c index d53af93b919b..c0df71d7d0ff 100644 --- a/drivers/gpu/drm/i915/i915_request.c +++ b/drivers/gpu/drm/i915/i915_request.c @@ -290,7 +290,7 @@ bool i915_request_retire(struct i915_request *rq) spin_unlock_irq(&rq->lock); remove_from_client(rq); - list_del(&rq->link); + __list_del_entry(&rq->link); /* poison neither prev/next (RCU walks) */ intel_context_exit(rq->context); intel_context_unpin(rq->context); @@ -363,6 +363,50 @@ __await_execution(struct i915_request *rq, return 0; } +static bool fatal_error(int error) +{ + switch (error) { + case 0: /* not an error! */ + case -EAGAIN: /* innocent victim of a GT reset (__i915_request_reset) */ + case -ETIMEDOUT: /* waiting for Godot (timer_i915_sw_fence_wake) */ + return false; + default: + return true; + } +} + +void __i915_request_skip(struct i915_request *rq) +{ + GEM_BUG_ON(!fatal_error(rq->fence.error)); + + if (rq->infix == rq->postfix) + return; + + /* + * As this request likely depends on state from the lost + * context, clear out all the user operations leaving the + * breadcrumb at the end (so we get the fence notifications). + */ + __i915_request_fill(rq, 0); + rq->infix = rq->postfix; +} + +void i915_request_set_error_once(struct i915_request *rq, int error) +{ + int old; + + GEM_BUG_ON(!IS_ERR_VALUE((long)error)); + + if (i915_request_signaled(rq)) + return; + + old = READ_ONCE(rq->fence.error); + do { + if (fatal_error(old)) + return; + } while (!try_cmpxchg(&rq->fence.error, &old, error)); +} + bool __i915_request_submit(struct i915_request *request) { struct intel_engine_cs *engine = request->engine; @@ -392,8 +436,10 @@ bool __i915_request_submit(struct i915_request *request) if (i915_request_completed(request)) goto xfer; - if (intel_context_is_banned(request->context)) - i915_request_skip(request, -EIO); + if (unlikely(intel_context_is_banned(request->context))) + i915_request_set_error_once(request, -EIO); + if (unlikely(fatal_error(request->fence.error))) + __i915_request_skip(request); /* * Are we using semaphores when the gpu is already saturated? @@ -519,7 +565,7 @@ submit_notify(struct i915_sw_fence *fence, enum i915_sw_fence_notify state) trace_i915_request_submit(request); if (unlikely(fence->error)) - i915_request_skip(request, fence->error); + i915_request_set_error_once(request, fence->error); /* * We need to serialize use of the submit_request() callback @@ -542,19 +588,31 @@ submit_notify(struct i915_sw_fence *fence, enum i915_sw_fence_notify state) return NOTIFY_DONE; } +static void irq_semaphore_cb(struct irq_work *wrk) +{ + struct i915_request *rq = + container_of(wrk, typeof(*rq), semaphore_work); + + i915_schedule_bump_priority(rq, I915_PRIORITY_NOSEMAPHORE); + i915_request_put(rq); +} + static int __i915_sw_fence_call semaphore_notify(struct i915_sw_fence *fence, enum i915_sw_fence_notify state) { - struct i915_request *request = - container_of(fence, typeof(*request), semaphore); + struct i915_request *rq = container_of(fence, typeof(*rq), semaphore); switch (state) { case FENCE_COMPLETE: - i915_schedule_bump_priority(request, I915_PRIORITY_NOSEMAPHORE); + if (!(READ_ONCE(rq->sched.attr.priority) & I915_PRIORITY_NOSEMAPHORE)) { + i915_request_get(rq); + init_irq_work(&rq->semaphore_work, irq_semaphore_cb); + irq_work_queue(&rq->semaphore_work); + } break; case FENCE_FREE: - i915_request_put(request); + i915_request_put(rq); break; } @@ -691,6 +749,7 @@ __i915_request_create(struct intel_context *ce, gfp_t gfp) RCU_INIT_POINTER(rq->timeline, tl); RCU_INIT_POINTER(rq->hwsp_cacheline, tl->hwsp_cacheline); rq->hwsp_seqno = tl->hwsp_seqno; + GEM_BUG_ON(i915_request_completed(rq)); rq->rcustate = get_state_synchronize_rcu(); /* acts as smp_mb() */ @@ -736,6 +795,8 @@ __i915_request_create(struct intel_context *ce, gfp_t gfp) rq->infix = rq->ring->emit; /* end of header; start of user payload */ intel_context_mark_active(ce); + list_add_tail_rcu(&rq->link, &tl->requests); + return rq; err_unwind: @@ -789,16 +850,26 @@ i915_request_await_start(struct i915_request *rq, struct i915_request *signal) struct dma_fence *fence; int err; - GEM_BUG_ON(i915_request_timeline(rq) == - rcu_access_pointer(signal->timeline)); + if (i915_request_timeline(rq) == rcu_access_pointer(signal->timeline)) + return 0; + + if (i915_request_started(signal)) + return 0; fence = NULL; rcu_read_lock(); spin_lock_irq(&signal->lock); - if (!i915_request_started(signal) && - !list_is_first(&signal->link, - &rcu_dereference(signal->timeline)->requests)) { - struct i915_request *prev = list_prev_entry(signal, link); + do { + struct list_head *pos = READ_ONCE(signal->link.prev); + struct i915_request *prev; + + /* Confirm signal has not been retired, the link is valid */ + if (unlikely(i915_request_started(signal))) + break; + + /* Is signal the earliest request on its timeline? */ + if (pos == &rcu_dereference(signal->timeline)->requests) + break; /* * Peek at the request before us in the timeline. That @@ -806,20 +877,25 @@ i915_request_await_start(struct i915_request *rq, struct i915_request *signal) * after acquiring a reference to it, confirm that it is * still part of the signaler's timeline. */ - if (i915_request_get_rcu(prev)) { - if (list_next_entry(prev, link) == signal) - fence = &prev->fence; - else - i915_request_put(prev); + prev = list_entry(pos, typeof(*prev), link); + if (!i915_request_get_rcu(prev)) + break; + + /* After the strong barrier, confirm prev is still attached */ + if (unlikely(READ_ONCE(prev->link.next) != &signal->link)) { + i915_request_put(prev); + break; } - } + + fence = &prev->fence; + } while (0); spin_unlock_irq(&signal->lock); rcu_read_unlock(); if (!fence) return 0; err = 0; - if (intel_timeline_sync_is_later(i915_request_timeline(rq), fence)) + if (!intel_timeline_sync_is_later(i915_request_timeline(rq), fence)) err = i915_sw_fence_await_dma_fence(&rq->submit, fence, 0, I915_FENCE_GFP); @@ -843,7 +919,7 @@ already_busywaiting(struct i915_request *rq) * * See the are-we-too-late? check in __i915_request_submit(). */ - return rq->sched.semaphores | rq->engine->saturated; + return rq->sched.semaphores | READ_ONCE(rq->engine->saturated); } static int @@ -901,6 +977,8 @@ emit_semaphore_wait(struct i915_request *to, struct i915_request *from, gfp_t gfp) { + const intel_engine_mask_t mask = READ_ONCE(from->engine)->mask; + if (!intel_context_use_semaphores(to->context)) goto await_fence; @@ -908,7 +986,7 @@ emit_semaphore_wait(struct i915_request *to, goto await_fence; /* Just emit the first semaphore we see as request space is limited. */ - if (already_busywaiting(to) & from->engine->mask) + if (already_busywaiting(to) & mask) goto await_fence; if (i915_request_await_start(to, from) < 0) @@ -921,7 +999,7 @@ emit_semaphore_wait(struct i915_request *to, if (__emit_semaphore_wait(to, from, from->fence.seqno)) goto await_fence; - to->sched.semaphores |= from->engine->mask; + to->sched.semaphores |= mask; to->sched.flags |= I915_SCHED_HAS_SEMAPHORE_CHAIN; return 0; @@ -1064,14 +1142,45 @@ __i915_request_await_execution(struct i915_request *to, &from->fence)) return 0; - /* Ensure both start together [after all semaphores in signal] */ - if (intel_engine_has_semaphores(to->engine)) - err = __emit_semaphore_wait(to, from, from->fence.seqno - 1); - else - err = i915_request_await_start(to, from); + /* + * Wait until the start of this request. + * + * The execution cb fires when we submit the request to HW. But in + * many cases this may be long before the request itself is ready to + * run (consider that we submit 2 requests for the same context, where + * the request of interest is behind an indefinite spinner). So we hook + * up to both to reduce our queues and keep the execution lag minimised + * in the worst case, though we hope that the await_start is elided. + */ + err = i915_request_await_start(to, from); if (err < 0) return err; + /* + * Ensure both start together [after all semaphores in signal] + * + * Now that we are queued to the HW at roughly the same time (thanks + * to the execute cb) and are ready to run at roughly the same time + * (thanks to the await start), our signaler may still be indefinitely + * delayed by waiting on a semaphore from a remote engine. If our + * signaler depends on a semaphore, so indirectly do we, and we do not + * want to start our payload until our signaler also starts theirs. + * So we wait. + * + * However, there is also a second condition for which we need to wait + * for the precise start of the signaler. Consider that the signaler + * was submitted in a chain of requests following another context + * (with just an ordinary intra-engine fence dependency between the + * two). In this case the signaler is queued to HW, but not for + * immediate execution, and so we must wait until it reaches the + * active slot. + */ + if (intel_engine_has_semaphores(to->engine)) { + err = __emit_semaphore_wait(to, from, from->fence.seqno - 1); + if (err < 0) + return err; + } + /* Couple the dependency tree for PI on this exposed to->fence */ if (to->engine->schedule) { err = i915_sched_node_add_dependency(&to->sched, &from->sched); @@ -1192,23 +1301,6 @@ i915_request_await_object(struct i915_request *to, return ret; } -void i915_request_skip(struct i915_request *rq, int error) -{ - GEM_BUG_ON(!IS_ERR_VALUE((long)error)); - dma_fence_set_error(&rq->fence, error); - - if (rq->infix == rq->postfix) - return; - - /* - * As this request likely depends on state from the lost - * context, clear out all the user operations leaving the - * breadcrumb at the end (so we get the fence notifications). - */ - __i915_request_fill(rq, 0); - rq->infix = rq->postfix; -} - static struct i915_request * __i915_request_add_to_timeline(struct i915_request *rq) { @@ -1238,7 +1330,17 @@ __i915_request_add_to_timeline(struct i915_request *rq) prev = to_request(__i915_active_fence_set(&timeline->last_request, &rq->fence)); if (prev && !i915_request_completed(prev)) { - if (is_power_of_2(prev->engine->mask | rq->engine->mask)) + /* + * The requests are supposed to be kept in order. However, + * we need to be wary in case the timeline->last_request + * is used as a barrier for external modification to this + * context. + */ + GEM_BUG_ON(prev->context == rq->context && + i915_seqno_passed(prev->fence.seqno, + rq->fence.seqno)); + + if (is_power_of_2(READ_ONCE(prev->engine)->mask | rq->engine->mask)) i915_sw_fence_await_sw_fence(&rq->submit, &prev->submit, &rq->submitq); @@ -1253,8 +1355,6 @@ __i915_request_add_to_timeline(struct i915_request *rq) 0); } - list_add_tail(&rq->link, &timeline->requests); - /* * Make sure that no request gazumped us - if it was allocated after * our i915_request_alloc() and called __i915_request_add() before @@ -1314,9 +1414,9 @@ void __i915_request_queue(struct i915_request *rq, * decide whether to preempt the entire chain so that it is ready to * run at the earliest possible convenience. */ - i915_sw_fence_commit(&rq->semaphore); if (attr && rq->engine->schedule) rq->engine->schedule(rq, attr); + i915_sw_fence_commit(&rq->semaphore); i915_sw_fence_commit(&rq->submit); } @@ -1324,39 +1424,23 @@ void i915_request_add(struct i915_request *rq) { struct intel_timeline * const tl = i915_request_timeline(rq); struct i915_sched_attr attr = {}; - struct i915_request *prev; + struct i915_gem_context *ctx; lockdep_assert_held(&tl->mutex); lockdep_unpin_lock(&tl->mutex, rq->cookie); trace_i915_request_add(rq); + __i915_request_commit(rq); - prev = __i915_request_commit(rq); - - if (rcu_access_pointer(rq->context->gem_context)) - attr = i915_request_gem_context(rq)->sched; + /* XXX placeholder for selftests */ + rcu_read_lock(); + ctx = rcu_dereference(rq->context->gem_context); + if (ctx) + attr = ctx->sched; + rcu_read_unlock(); - /* - * Boost actual workloads past semaphores! - * - * With semaphores we spin on one engine waiting for another, - * simply to reduce the latency of starting our work when - * the signaler completes. However, if there is any other - * work that we could be doing on this engine instead, that - * is better utilisation and will reduce the overall duration - * of the current work. To avoid PI boosting a semaphore - * far in the distance past over useful work, we keep a history - * of any semaphore use along our dependency chain. - */ if (!(rq->sched.flags & I915_SCHED_HAS_SEMAPHORE_CHAIN)) attr.priority |= I915_PRIORITY_NOSEMAPHORE; - - /* - * Boost priorities to new clients (new request flows). - * - * Allow interactive/synchronous clients to jump ahead of - * the bulk clients. (FQ_CODEL) - */ if (list_empty(&rq->sched.signalers_list)) attr.priority |= I915_PRIORITY_WAIT; @@ -1364,32 +1448,10 @@ void i915_request_add(struct i915_request *rq) __i915_request_queue(rq, &attr); local_bh_enable(); /* Kick the execlists tasklet if just scheduled */ - /* - * In typical scenarios, we do not expect the previous request on - * the timeline to be still tracked by timeline->last_request if it - * has been completed. If the completed request is still here, that - * implies that request retirement is a long way behind submission, - * suggesting that we haven't been retiring frequently enough from - * the combination of retire-before-alloc, waiters and the background - * retirement worker. So if the last request on this timeline was - * already completed, do a catch up pass, flushing the retirement queue - * up to this client. Since we have now moved the heaviest operations - * during retirement onto secondary workers, such as freeing objects - * or contexts, retiring a bunch of requests is mostly list management - * (and cache misses), and so we should not be overly penalizing this - * client by performing excess work, though we may still performing - * work on behalf of others -- but instead we should benefit from - * improved resource management. (Well, that's the theory at least.) - */ - if (prev && - i915_request_completed(prev) && - rcu_access_pointer(prev->timeline) == tl) - i915_request_retire_upto(prev); - mutex_unlock(&tl->mutex); } -static unsigned long local_clock_us(unsigned int *cpu) +static unsigned long local_clock_ns(unsigned int *cpu) { unsigned long t; @@ -1406,7 +1468,7 @@ static unsigned long local_clock_us(unsigned int *cpu) * stop busywaiting, see busywait_stop(). */ *cpu = get_cpu(); - t = local_clock() >> 10; + t = local_clock(); put_cpu(); return t; @@ -1416,15 +1478,15 @@ static bool busywait_stop(unsigned long timeout, unsigned int cpu) { unsigned int this_cpu; - if (time_after(local_clock_us(&this_cpu), timeout)) + if (time_after(local_clock_ns(&this_cpu), timeout)) return true; return this_cpu != cpu; } -static bool __i915_spin_request(const struct i915_request * const rq, - int state, unsigned long timeout_us) +static bool __i915_spin_request(const struct i915_request * const rq, int state) { + unsigned long timeout_ns; unsigned int cpu; /* @@ -1452,7 +1514,8 @@ static bool __i915_spin_request(const struct i915_request * const rq, * takes to sleep on a request, on the order of a microsecond. */ - timeout_us += local_clock_us(&cpu); + timeout_ns = READ_ONCE(rq->engine->props.max_busywait_duration_ns); + timeout_ns += local_clock_ns(&cpu); do { if (i915_request_completed(rq)) return true; @@ -1460,7 +1523,7 @@ static bool __i915_spin_request(const struct i915_request * const rq, if (signal_pending_state(state, current)) break; - if (busywait_stop(timeout_us, cpu)) + if (busywait_stop(timeout_ns, cpu)) break; cpu_relax(); @@ -1546,8 +1609,8 @@ long i915_request_wait(struct i915_request *rq, * completion. That requires having a good predictor for the request * duration, which we currently lack. */ - if (IS_ACTIVE(CONFIG_DRM_I915_SPIN_REQUEST) && - __i915_spin_request(rq, state, CONFIG_DRM_I915_SPIN_REQUEST)) { + if (IS_ACTIVE(CONFIG_DRM_I915_MAX_REQUEST_BUSYWAIT) && + __i915_spin_request(rq, state)) { dma_fence_signal(&rq->fence); goto out; } diff --git a/drivers/gpu/drm/i915/i915_request.h b/drivers/gpu/drm/i915/i915_request.h index da8420f03232..3c552bfea67a 100644 --- a/drivers/gpu/drm/i915/i915_request.h +++ b/drivers/gpu/drm/i915/i915_request.h @@ -26,6 +26,7 @@ #define I915_REQUEST_H #include <linux/dma-fence.h> +#include <linux/irq_work.h> #include <linux/lockdep.h> #include "gem/i915_gem_context_types.h" @@ -208,6 +209,7 @@ struct i915_request { }; struct list_head execute_cb; struct i915_sw_fence semaphore; + struct irq_work semaphore_work; /* * A list of everyone we wait upon, and everyone who waits upon us. @@ -303,6 +305,9 @@ __i915_request_create(struct intel_context *ce, gfp_t gfp); struct i915_request * __must_check i915_request_create(struct intel_context *ce); +void i915_request_set_error_once(struct i915_request *rq, int error); +void __i915_request_skip(struct i915_request *rq); + struct i915_request *__i915_request_commit(struct i915_request *request); void __i915_request_queue(struct i915_request *rq, const struct i915_sched_attr *attr); @@ -352,8 +357,6 @@ void i915_request_add(struct i915_request *rq); bool __i915_request_submit(struct i915_request *request); void i915_request_submit(struct i915_request *request); -void i915_request_skip(struct i915_request *request, int error); - void __i915_request_unsubmit(struct i915_request *request); void i915_request_unsubmit(struct i915_request *request); @@ -395,7 +398,9 @@ static inline bool i915_seqno_passed(u32 seq1, u32 seq2) static inline u32 __hwsp_seqno(const struct i915_request *rq) { - return READ_ONCE(*rq->hwsp_seqno); + const u32 *hwsp = READ_ONCE(rq->hwsp_seqno); + + return READ_ONCE(*hwsp); } /** @@ -509,7 +514,8 @@ static inline bool i915_request_completed(const struct i915_request *rq) static inline void i915_request_mark_complete(struct i915_request *rq) { - rq->hwsp_seqno = (u32 *)&rq->fence.seqno; /* decouple from HWSP */ + WRITE_ONCE(rq->hwsp_seqno, /* decouple from HWSP */ + (u32 *)&rq->fence.seqno); } static inline bool i915_request_has_waitboost(const struct i915_request *rq) diff --git a/drivers/gpu/drm/i915/i915_scheduler.c b/drivers/gpu/drm/i915/i915_scheduler.c index be770f2419b1..68b06a7ba667 100644 --- a/drivers/gpu/drm/i915/i915_scheduler.c +++ b/drivers/gpu/drm/i915/i915_scheduler.c @@ -209,6 +209,8 @@ static void kick_submission(struct intel_engine_cs *engine, if (!inflight) goto unlock; + engine->execlists.queue_priority_hint = prio; + /* * If we are already the currently executing context, don't * bother evaluating if we should preempt ourselves. @@ -216,7 +218,6 @@ static void kick_submission(struct intel_engine_cs *engine, if (inflight->context == rq->context) goto unlock; - engine->execlists.queue_priority_hint = prio; if (need_preempt(prio, rq_prio(inflight))) tasklet_hi_schedule(&engine->execlists.tasklet); @@ -227,10 +228,10 @@ unlock: static void __i915_schedule(struct i915_sched_node *node, const struct i915_sched_attr *attr) { + const int prio = max(attr->priority, node->attr.priority); struct intel_engine_cs *engine; struct i915_dependency *dep, *p; struct i915_dependency stack; - const int prio = attr->priority; struct sched_cache cache; LIST_HEAD(dfs); @@ -238,9 +239,6 @@ static void __i915_schedule(struct i915_sched_node *node, lockdep_assert_held(&schedule_lock); GEM_BUG_ON(prio == I915_PRIORITY_INVALID); - if (prio <= READ_ONCE(node->attr.priority)) - return; - if (node_signaled(node)) return; @@ -324,7 +322,7 @@ static void __i915_schedule(struct i915_sched_node *node, GEM_BUG_ON(node_to_request(node)->engine != engine); - node->attr.priority = prio; + WRITE_ONCE(node->attr.priority, prio); /* * Once the request is ready, it will be placed into the diff --git a/drivers/gpu/drm/i915/i915_suspend.c b/drivers/gpu/drm/i915/i915_suspend.c index 8812cdd9007f..ed2be3489f8e 100644 --- a/drivers/gpu/drm/i915/i915_suspend.c +++ b/drivers/gpu/drm/i915/i915_suspend.c @@ -24,8 +24,6 @@ * SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ -#include <drm/i915_drm.h> - #include "display/intel_fbc.h" #include "display/intel_gmbus.h" #include "display/intel_vga.h" diff --git a/drivers/gpu/drm/i915/i915_sysfs.c b/drivers/gpu/drm/i915/i915_sysfs.c index c14d762bd652..45d32ef42787 100644 --- a/drivers/gpu/drm/i915/i915_sysfs.c +++ b/drivers/gpu/drm/i915/i915_sysfs.c @@ -32,6 +32,7 @@ #include "gt/intel_rc6.h" #include "gt/intel_rps.h" +#include "gt/sysfs_engines.h" #include "i915_drv.h" #include "i915_sysfs.h" @@ -606,6 +607,8 @@ void i915_setup_sysfs(struct drm_i915_private *dev_priv) drm_err(&dev_priv->drm, "RPS sysfs setup failed\n"); i915_setup_error_capture(kdev); + + intel_engines_add_sysfs(dev_priv); } void i915_teardown_sysfs(struct drm_i915_private *dev_priv) diff --git a/drivers/gpu/drm/i915/i915_utils.c b/drivers/gpu/drm/i915/i915_utils.c index 632d6953c78d..029854ae65fc 100644 --- a/drivers/gpu/drm/i915/i915_utils.c +++ b/drivers/gpu/drm/i915/i915_utils.c @@ -8,7 +8,6 @@ #include "i915_drv.h" #include "i915_utils.h" -#define FDO_BUG_URL "https://gitlab.freedesktop.org/drm/intel/-/wikis/How-to-file-i915-bugs" #define FDO_BUG_MSG "Please file a bug on drm/i915; see " FDO_BUG_URL " for details." void diff --git a/drivers/gpu/drm/i915/i915_utils.h b/drivers/gpu/drm/i915/i915_utils.h index b0ade76bec90..03a73d2bd50d 100644 --- a/drivers/gpu/drm/i915/i915_utils.h +++ b/drivers/gpu/drm/i915/i915_utils.h @@ -34,6 +34,8 @@ struct drm_i915_private; struct timer_list; +#define FDO_BUG_URL "https://gitlab.freedesktop.org/drm/intel/-/wikis/How-to-file-i915-bugs" + #undef WARN_ON /* Many gcc seem to no see through this and fall over :( */ #if 0 @@ -100,12 +102,24 @@ bool i915_error_injected(void); typeof(max) max__ = (max); \ (void)(&start__ == &size__); \ (void)(&start__ == &max__); \ - start__ > max__ || size__ > max__ - start__; \ + start__ >= max__ || size__ > max__ - start__; \ }) #define range_overflows_t(type, start, size, max) \ range_overflows((type)(start), (type)(size), (type)(max)) +#define range_overflows_end(start, size, max) ({ \ + typeof(start) start__ = (start); \ + typeof(size) size__ = (size); \ + typeof(max) max__ = (max); \ + (void)(&start__ == &size__); \ + (void)(&start__ == &max__); \ + start__ > max__ || size__ > max__ - start__; \ +}) + +#define range_overflows_end_t(type, start, size, max) \ + range_overflows_end((type)(start), (type)(size), (type)(max)) + /* Note we don't consider signbits :| */ #define overflows_type(x, T) \ (sizeof(x) > sizeof(T) && (x) >> BITS_PER_TYPE(T)) @@ -234,6 +248,11 @@ static inline u64 ptr_to_u64(const void *ptr) __idx; \ }) +static inline bool is_power_of_2_u64(u64 n) +{ + return (n != 0 && ((n & (n - 1)) == 0)); +} + static inline void __list_del_many(struct list_head *head, struct list_head *first) { @@ -241,6 +260,12 @@ static inline void __list_del_many(struct list_head *head, WRITE_ONCE(head->next, first); } +static inline int list_is_last_rcu(const struct list_head *list, + const struct list_head *head) +{ + return READ_ONCE(list->next) == head; +} + /* * Wait until the work is finally complete, even if it tries to postpone * by requeueing itself. Note, that if the worker never cancels itself, diff --git a/drivers/gpu/drm/i915/i915_vgpu.c b/drivers/gpu/drm/i915/i915_vgpu.c index 4afe21662266..70fca72f5162 100644 --- a/drivers/gpu/drm/i915/i915_vgpu.c +++ b/drivers/gpu/drm/i915/i915_vgpu.c @@ -21,6 +21,8 @@ * SOFTWARE. */ +#include "i915_drv.h" +#include "i915_pvinfo.h" #include "i915_vgpu.h" /** @@ -51,13 +53,13 @@ */ /** - * i915_detect_vgpu - detect virtual GPU + * intel_vgpu_detect - detect virtual GPU * @dev_priv: i915 device private * * This function is called at the initialization stage, to detect whether * running on a vGPU. */ -void i915_detect_vgpu(struct drm_i915_private *dev_priv) +void intel_vgpu_detect(struct drm_i915_private *dev_priv) { struct pci_dev *pdev = dev_priv->drm.pdev; u64 magic; @@ -102,11 +104,36 @@ out: pci_iounmap(pdev, shared_area); } +void intel_vgpu_register(struct drm_i915_private *i915) +{ + /* + * Notify a valid surface after modesetting, when running inside a VM. + */ + if (intel_vgpu_active(i915)) + intel_uncore_write(&i915->uncore, vgtif_reg(display_ready), + VGT_DRV_DISPLAY_READY); +} + +bool intel_vgpu_active(struct drm_i915_private *dev_priv) +{ + return dev_priv->vgpu.active; +} + bool intel_vgpu_has_full_ppgtt(struct drm_i915_private *dev_priv) { return dev_priv->vgpu.caps & VGT_CAPS_FULL_PPGTT; } +bool intel_vgpu_has_hwsp_emulation(struct drm_i915_private *dev_priv) +{ + return dev_priv->vgpu.caps & VGT_CAPS_HWSP_EMULATION; +} + +bool intel_vgpu_has_huge_gtt(struct drm_i915_private *dev_priv) +{ + return dev_priv->vgpu.caps & VGT_CAPS_HUGE_GTT; +} + struct _balloon_info_ { /* * There are up to 2 regions per mappable/unmappable graphic diff --git a/drivers/gpu/drm/i915/i915_vgpu.h b/drivers/gpu/drm/i915/i915_vgpu.h index 8b3663dad193..ffbb77d08048 100644 --- a/drivers/gpu/drm/i915/i915_vgpu.h +++ b/drivers/gpu/drm/i915/i915_vgpu.h @@ -24,24 +24,17 @@ #ifndef _I915_VGPU_H_ #define _I915_VGPU_H_ -#include "i915_drv.h" -#include "i915_pvinfo.h" +#include <linux/types.h> -void i915_detect_vgpu(struct drm_i915_private *dev_priv); +struct drm_i915_private; +struct i915_ggtt; -bool intel_vgpu_has_full_ppgtt(struct drm_i915_private *dev_priv); - -static inline bool -intel_vgpu_has_hwsp_emulation(struct drm_i915_private *dev_priv) -{ - return dev_priv->vgpu.caps & VGT_CAPS_HWSP_EMULATION; -} - -static inline bool -intel_vgpu_has_huge_gtt(struct drm_i915_private *dev_priv) -{ - return dev_priv->vgpu.caps & VGT_CAPS_HUGE_GTT; -} +void intel_vgpu_detect(struct drm_i915_private *i915); +bool intel_vgpu_active(struct drm_i915_private *i915); +void intel_vgpu_register(struct drm_i915_private *i915); +bool intel_vgpu_has_full_ppgtt(struct drm_i915_private *i915); +bool intel_vgpu_has_hwsp_emulation(struct drm_i915_private *i915); +bool intel_vgpu_has_huge_gtt(struct drm_i915_private *i915); int intel_vgt_balloon(struct i915_ggtt *ggtt); void intel_vgt_deballoon(struct i915_ggtt *ggtt); diff --git a/drivers/gpu/drm/i915/i915_vma.c b/drivers/gpu/drm/i915/i915_vma.c index 298ca4316e65..08699fa069aa 100644 --- a/drivers/gpu/drm/i915/i915_vma.c +++ b/drivers/gpu/drm/i915/i915_vma.c @@ -641,7 +641,6 @@ i915_vma_insert(struct i915_vma *vma, u64 size, u64 alignment, u64 flags) u64 start, end; int ret; - GEM_BUG_ON(i915_vma_is_closed(vma)); GEM_BUG_ON(i915_vma_is_bound(vma, I915_VMA_GLOBAL_BIND | I915_VMA_LOCAL_BIND)); GEM_BUG_ON(drm_mm_node_allocated(&vma->node)); @@ -1098,6 +1097,7 @@ void i915_vma_release(struct kref *ref) void i915_vma_parked(struct intel_gt *gt) { struct i915_vma *vma, *next; + LIST_HEAD(closed); spin_lock_irq(>->closed_lock); list_for_each_entry_safe(vma, next, >->closed_vma, closed_link) { @@ -1109,28 +1109,26 @@ void i915_vma_parked(struct intel_gt *gt) if (!kref_get_unless_zero(&obj->base.refcount)) continue; - if (i915_vm_tryopen(vm)) { - list_del_init(&vma->closed_link); - } else { + if (!i915_vm_tryopen(vm)) { i915_gem_object_put(obj); - obj = NULL; + continue; } - spin_unlock_irq(>->closed_lock); + list_move(&vma->closed_link, &closed); + } + spin_unlock_irq(>->closed_lock); - if (obj) { - __i915_vma_put(vma); - i915_gem_object_put(obj); - } + /* As the GT is held idle, no vma can be reopened as we destroy them */ + list_for_each_entry_safe(vma, next, &closed, closed_link) { + struct drm_i915_gem_object *obj = vma->obj; + struct i915_address_space *vm = vma->vm; - i915_vm_close(vm); + INIT_LIST_HEAD(&vma->closed_link); + __i915_vma_put(vma); - /* Restart after dropping lock */ - spin_lock_irq(>->closed_lock); - next = list_first_entry(>->closed_vma, - typeof(*next), closed_link); + i915_gem_object_put(obj); + i915_vm_close(vm); } - spin_unlock_irq(>->closed_lock); } static void __i915_vma_iounmap(struct i915_vma *vma) @@ -1174,7 +1172,7 @@ int __i915_vma_move_to_active(struct i915_vma *vma, struct i915_request *rq) GEM_BUG_ON(!i915_vma_is_pinned(vma)); /* Wait for the vma to be bound before we start! */ - err = i915_request_await_active(rq, &vma->active); + err = i915_request_await_active(rq, &vma->active, 0); if (err) return err; diff --git a/drivers/gpu/drm/i915/i915_vma_types.h b/drivers/gpu/drm/i915/i915_vma_types.h index e0942efd5236..63831cdb7402 100644 --- a/drivers/gpu/drm/i915/i915_vma_types.h +++ b/drivers/gpu/drm/i915/i915_vma_types.h @@ -273,21 +273,10 @@ struct i915_vma { struct rb_node obj_node; struct hlist_node obj_hash; - /** This vma's place in the execbuf reservation list */ - struct list_head exec_link; - struct list_head reloc_link; - /** This vma's place in the eviction list */ struct list_head evict_link; struct list_head closed_link; - - /** - * Used for performing relocations during execbuffer insertion. - */ - unsigned int *exec_flags; - struct hlist_node exec_node; - u32 exec_handle; }; #endif diff --git a/drivers/gpu/drm/i915/intel_device_info.c b/drivers/gpu/drm/i915/intel_device_info.c index 8e99ad097830..d7fe12734db8 100644 --- a/drivers/gpu/drm/i915/intel_device_info.c +++ b/drivers/gpu/drm/i915/intel_device_info.c @@ -23,6 +23,7 @@ */ #include <drm/drm_print.h> +#include <drm/i915_pciids.h> #include "display/intel_cdclk.h" #include "intel_device_info.h" diff --git a/drivers/gpu/drm/i915/intel_dram.c b/drivers/gpu/drm/i915/intel_dram.c new file mode 100644 index 000000000000..6b922efb1d7c --- /dev/null +++ b/drivers/gpu/drm/i915/intel_dram.c @@ -0,0 +1,500 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright © 2020 Intel Corporation + */ + +#include "i915_drv.h" +#include "intel_dram.h" + +struct dram_dimm_info { + u8 size, width, ranks; +}; + +struct dram_channel_info { + struct dram_dimm_info dimm_l, dimm_s; + u8 ranks; + bool is_16gb_dimm; +}; + +#define DRAM_TYPE_STR(type) [INTEL_DRAM_ ## type] = #type + +static const char *intel_dram_type_str(enum intel_dram_type type) +{ + static const char * const str[] = { + DRAM_TYPE_STR(UNKNOWN), + DRAM_TYPE_STR(DDR3), + DRAM_TYPE_STR(DDR4), + DRAM_TYPE_STR(LPDDR3), + DRAM_TYPE_STR(LPDDR4), + }; + + if (type >= ARRAY_SIZE(str)) + type = INTEL_DRAM_UNKNOWN; + + return str[type]; +} + +#undef DRAM_TYPE_STR + +static int intel_dimm_num_devices(const struct dram_dimm_info *dimm) +{ + return dimm->ranks * 64 / (dimm->width ?: 1); +} + +/* Returns total GB for the whole DIMM */ +static int skl_get_dimm_size(u16 val) +{ + return val & SKL_DRAM_SIZE_MASK; +} + +static int skl_get_dimm_width(u16 val) +{ + if (skl_get_dimm_size(val) == 0) + return 0; + + switch (val & SKL_DRAM_WIDTH_MASK) { + case SKL_DRAM_WIDTH_X8: + case SKL_DRAM_WIDTH_X16: + case SKL_DRAM_WIDTH_X32: + val = (val & SKL_DRAM_WIDTH_MASK) >> SKL_DRAM_WIDTH_SHIFT; + return 8 << val; + default: + MISSING_CASE(val); + return 0; + } +} + +static int skl_get_dimm_ranks(u16 val) +{ + if (skl_get_dimm_size(val) == 0) + return 0; + + val = (val & SKL_DRAM_RANK_MASK) >> SKL_DRAM_RANK_SHIFT; + + return val + 1; +} + +/* Returns total GB for the whole DIMM */ +static int cnl_get_dimm_size(u16 val) +{ + return (val & CNL_DRAM_SIZE_MASK) / 2; +} + +static int cnl_get_dimm_width(u16 val) +{ + if (cnl_get_dimm_size(val) == 0) + return 0; + + switch (val & CNL_DRAM_WIDTH_MASK) { + case CNL_DRAM_WIDTH_X8: + case CNL_DRAM_WIDTH_X16: + case CNL_DRAM_WIDTH_X32: + val = (val & CNL_DRAM_WIDTH_MASK) >> CNL_DRAM_WIDTH_SHIFT; + return 8 << val; + default: + MISSING_CASE(val); + return 0; + } +} + +static int cnl_get_dimm_ranks(u16 val) +{ + if (cnl_get_dimm_size(val) == 0) + return 0; + + val = (val & CNL_DRAM_RANK_MASK) >> CNL_DRAM_RANK_SHIFT; + + return val + 1; +} + +static bool +skl_is_16gb_dimm(const struct dram_dimm_info *dimm) +{ + /* Convert total GB to Gb per DRAM device */ + return 8 * dimm->size / (intel_dimm_num_devices(dimm) ?: 1) == 16; +} + +static void +skl_dram_get_dimm_info(struct drm_i915_private *i915, + struct dram_dimm_info *dimm, + int channel, char dimm_name, u16 val) +{ + if (INTEL_GEN(i915) >= 10) { + dimm->size = cnl_get_dimm_size(val); + dimm->width = cnl_get_dimm_width(val); + dimm->ranks = cnl_get_dimm_ranks(val); + } else { + dimm->size = skl_get_dimm_size(val); + dimm->width = skl_get_dimm_width(val); + dimm->ranks = skl_get_dimm_ranks(val); + } + + drm_dbg_kms(&i915->drm, + "CH%u DIMM %c size: %u GB, width: X%u, ranks: %u, 16Gb DIMMs: %s\n", + channel, dimm_name, dimm->size, dimm->width, dimm->ranks, + yesno(skl_is_16gb_dimm(dimm))); +} + +static int +skl_dram_get_channel_info(struct drm_i915_private *i915, + struct dram_channel_info *ch, + int channel, u32 val) +{ + skl_dram_get_dimm_info(i915, &ch->dimm_l, + channel, 'L', val & 0xffff); + skl_dram_get_dimm_info(i915, &ch->dimm_s, + channel, 'S', val >> 16); + + if (ch->dimm_l.size == 0 && ch->dimm_s.size == 0) { + drm_dbg_kms(&i915->drm, "CH%u not populated\n", channel); + return -EINVAL; + } + + if (ch->dimm_l.ranks == 2 || ch->dimm_s.ranks == 2) + ch->ranks = 2; + else if (ch->dimm_l.ranks == 1 && ch->dimm_s.ranks == 1) + ch->ranks = 2; + else + ch->ranks = 1; + + ch->is_16gb_dimm = skl_is_16gb_dimm(&ch->dimm_l) || + skl_is_16gb_dimm(&ch->dimm_s); + + drm_dbg_kms(&i915->drm, "CH%u ranks: %u, 16Gb DIMMs: %s\n", + channel, ch->ranks, yesno(ch->is_16gb_dimm)); + + return 0; +} + +static bool +intel_is_dram_symmetric(const struct dram_channel_info *ch0, + const struct dram_channel_info *ch1) +{ + return !memcmp(ch0, ch1, sizeof(*ch0)) && + (ch0->dimm_s.size == 0 || + !memcmp(&ch0->dimm_l, &ch0->dimm_s, sizeof(ch0->dimm_l))); +} + +static int +skl_dram_get_channels_info(struct drm_i915_private *i915) +{ + struct dram_info *dram_info = &i915->dram_info; + struct dram_channel_info ch0 = {}, ch1 = {}; + u32 val; + int ret; + + val = intel_uncore_read(&i915->uncore, + SKL_MAD_DIMM_CH0_0_0_0_MCHBAR_MCMAIN); + ret = skl_dram_get_channel_info(i915, &ch0, 0, val); + if (ret == 0) + dram_info->num_channels++; + + val = intel_uncore_read(&i915->uncore, + SKL_MAD_DIMM_CH1_0_0_0_MCHBAR_MCMAIN); + ret = skl_dram_get_channel_info(i915, &ch1, 1, val); + if (ret == 0) + dram_info->num_channels++; + + if (dram_info->num_channels == 0) { + drm_info(&i915->drm, "Number of memory channels is zero\n"); + return -EINVAL; + } + + /* + * If any of the channel is single rank channel, worst case output + * will be same as if single rank memory, so consider single rank + * memory. + */ + if (ch0.ranks == 1 || ch1.ranks == 1) + dram_info->ranks = 1; + else + dram_info->ranks = max(ch0.ranks, ch1.ranks); + + if (dram_info->ranks == 0) { + drm_info(&i915->drm, "couldn't get memory rank information\n"); + return -EINVAL; + } + + dram_info->is_16gb_dimm = ch0.is_16gb_dimm || ch1.is_16gb_dimm; + + dram_info->symmetric_memory = intel_is_dram_symmetric(&ch0, &ch1); + + drm_dbg_kms(&i915->drm, "Memory configuration is symmetric? %s\n", + yesno(dram_info->symmetric_memory)); + + return 0; +} + +static enum intel_dram_type +skl_get_dram_type(struct drm_i915_private *i915) +{ + u32 val; + + val = intel_uncore_read(&i915->uncore, + SKL_MAD_INTER_CHANNEL_0_0_0_MCHBAR_MCMAIN); + + switch (val & SKL_DRAM_DDR_TYPE_MASK) { + case SKL_DRAM_DDR_TYPE_DDR3: + return INTEL_DRAM_DDR3; + case SKL_DRAM_DDR_TYPE_DDR4: + return INTEL_DRAM_DDR4; + case SKL_DRAM_DDR_TYPE_LPDDR3: + return INTEL_DRAM_LPDDR3; + case SKL_DRAM_DDR_TYPE_LPDDR4: + return INTEL_DRAM_LPDDR4; + default: + MISSING_CASE(val); + return INTEL_DRAM_UNKNOWN; + } +} + +static int +skl_get_dram_info(struct drm_i915_private *i915) +{ + struct dram_info *dram_info = &i915->dram_info; + u32 mem_freq_khz, val; + int ret; + + dram_info->type = skl_get_dram_type(i915); + drm_dbg_kms(&i915->drm, "DRAM type: %s\n", + intel_dram_type_str(dram_info->type)); + + ret = skl_dram_get_channels_info(i915); + if (ret) + return ret; + + val = intel_uncore_read(&i915->uncore, + SKL_MC_BIOS_DATA_0_0_0_MCHBAR_PCU); + mem_freq_khz = DIV_ROUND_UP((val & SKL_REQ_DATA_MASK) * + SKL_MEMORY_FREQ_MULTIPLIER_HZ, 1000); + + dram_info->bandwidth_kbps = dram_info->num_channels * + mem_freq_khz * 8; + + if (dram_info->bandwidth_kbps == 0) { + drm_info(&i915->drm, + "Couldn't get system memory bandwidth\n"); + return -EINVAL; + } + + dram_info->valid = true; + return 0; +} + +/* Returns Gb per DRAM device */ +static int bxt_get_dimm_size(u32 val) +{ + switch (val & BXT_DRAM_SIZE_MASK) { + case BXT_DRAM_SIZE_4GBIT: + return 4; + case BXT_DRAM_SIZE_6GBIT: + return 6; + case BXT_DRAM_SIZE_8GBIT: + return 8; + case BXT_DRAM_SIZE_12GBIT: + return 12; + case BXT_DRAM_SIZE_16GBIT: + return 16; + default: + MISSING_CASE(val); + return 0; + } +} + +static int bxt_get_dimm_width(u32 val) +{ + if (!bxt_get_dimm_size(val)) + return 0; + + val = (val & BXT_DRAM_WIDTH_MASK) >> BXT_DRAM_WIDTH_SHIFT; + + return 8 << val; +} + +static int bxt_get_dimm_ranks(u32 val) +{ + if (!bxt_get_dimm_size(val)) + return 0; + + switch (val & BXT_DRAM_RANK_MASK) { + case BXT_DRAM_RANK_SINGLE: + return 1; + case BXT_DRAM_RANK_DUAL: + return 2; + default: + MISSING_CASE(val); + return 0; + } +} + +static enum intel_dram_type bxt_get_dimm_type(u32 val) +{ + if (!bxt_get_dimm_size(val)) + return INTEL_DRAM_UNKNOWN; + + switch (val & BXT_DRAM_TYPE_MASK) { + case BXT_DRAM_TYPE_DDR3: + return INTEL_DRAM_DDR3; + case BXT_DRAM_TYPE_LPDDR3: + return INTEL_DRAM_LPDDR3; + case BXT_DRAM_TYPE_DDR4: + return INTEL_DRAM_DDR4; + case BXT_DRAM_TYPE_LPDDR4: + return INTEL_DRAM_LPDDR4; + default: + MISSING_CASE(val); + return INTEL_DRAM_UNKNOWN; + } +} + +static void bxt_get_dimm_info(struct dram_dimm_info *dimm, u32 val) +{ + dimm->width = bxt_get_dimm_width(val); + dimm->ranks = bxt_get_dimm_ranks(val); + + /* + * Size in register is Gb per DRAM device. Convert to total + * GB to match the way we report this for non-LP platforms. + */ + dimm->size = bxt_get_dimm_size(val) * intel_dimm_num_devices(dimm) / 8; +} + +static int bxt_get_dram_info(struct drm_i915_private *i915) +{ + struct dram_info *dram_info = &i915->dram_info; + u32 dram_channels; + u32 mem_freq_khz, val; + u8 num_active_channels; + int i; + + val = intel_uncore_read(&i915->uncore, BXT_P_CR_MC_BIOS_REQ_0_0_0); + mem_freq_khz = DIV_ROUND_UP((val & BXT_REQ_DATA_MASK) * + BXT_MEMORY_FREQ_MULTIPLIER_HZ, 1000); + + dram_channels = val & BXT_DRAM_CHANNEL_ACTIVE_MASK; + num_active_channels = hweight32(dram_channels); + + /* Each active bit represents 4-byte channel */ + dram_info->bandwidth_kbps = (mem_freq_khz * num_active_channels * 4); + + if (dram_info->bandwidth_kbps == 0) { + drm_info(&i915->drm, + "Couldn't get system memory bandwidth\n"); + return -EINVAL; + } + + /* + * Now read each DUNIT8/9/10/11 to check the rank of each dimms. + */ + for (i = BXT_D_CR_DRP0_DUNIT_START; i <= BXT_D_CR_DRP0_DUNIT_END; i++) { + struct dram_dimm_info dimm; + enum intel_dram_type type; + + val = intel_uncore_read(&i915->uncore, BXT_D_CR_DRP0_DUNIT(i)); + if (val == 0xFFFFFFFF) + continue; + + dram_info->num_channels++; + + bxt_get_dimm_info(&dimm, val); + type = bxt_get_dimm_type(val); + + drm_WARN_ON(&i915->drm, type != INTEL_DRAM_UNKNOWN && + dram_info->type != INTEL_DRAM_UNKNOWN && + dram_info->type != type); + + drm_dbg_kms(&i915->drm, + "CH%u DIMM size: %u GB, width: X%u, ranks: %u, type: %s\n", + i - BXT_D_CR_DRP0_DUNIT_START, + dimm.size, dimm.width, dimm.ranks, + intel_dram_type_str(type)); + + /* + * If any of the channel is single rank channel, + * worst case output will be same as if single rank + * memory, so consider single rank memory. + */ + if (dram_info->ranks == 0) + dram_info->ranks = dimm.ranks; + else if (dimm.ranks == 1) + dram_info->ranks = 1; + + if (type != INTEL_DRAM_UNKNOWN) + dram_info->type = type; + } + + if (dram_info->type == INTEL_DRAM_UNKNOWN || dram_info->ranks == 0) { + drm_info(&i915->drm, "couldn't get memory information\n"); + return -EINVAL; + } + + dram_info->valid = true; + + return 0; +} + +void intel_dram_detect(struct drm_i915_private *i915) +{ + struct dram_info *dram_info = &i915->dram_info; + int ret; + + /* + * Assume 16Gb DIMMs are present until proven otherwise. + * This is only used for the level 0 watermark latency + * w/a which does not apply to bxt/glk. + */ + dram_info->is_16gb_dimm = !IS_GEN9_LP(i915); + + if (INTEL_GEN(i915) < 9 || !HAS_DISPLAY(i915)) + return; + + if (IS_GEN9_LP(i915)) + ret = bxt_get_dram_info(i915); + else + ret = skl_get_dram_info(i915); + if (ret) + return; + + drm_dbg_kms(&i915->drm, "DRAM bandwidth: %u kBps, channels: %u\n", + dram_info->bandwidth_kbps, dram_info->num_channels); + + drm_dbg_kms(&i915->drm, "DRAM ranks: %u, 16Gb DIMMs: %s\n", + dram_info->ranks, yesno(dram_info->is_16gb_dimm)); +} + +static u32 gen9_edram_size_mb(struct drm_i915_private *i915, u32 cap) +{ + static const u8 ways[8] = { 4, 8, 12, 16, 16, 16, 16, 16 }; + static const u8 sets[4] = { 1, 1, 2, 2 }; + + return EDRAM_NUM_BANKS(cap) * + ways[EDRAM_WAYS_IDX(cap)] * + sets[EDRAM_SETS_IDX(cap)]; +} + +void intel_dram_edram_detect(struct drm_i915_private *i915) +{ + u32 edram_cap = 0; + + if (!(IS_HASWELL(i915) || IS_BROADWELL(i915) || INTEL_GEN(i915) >= 9)) + return; + + edram_cap = __raw_uncore_read32(&i915->uncore, HSW_EDRAM_CAP); + + /* NB: We can't write IDICR yet because we don't have gt funcs set up */ + + if (!(edram_cap & EDRAM_ENABLED)) + return; + + /* + * The needed capability bits for size calculation are not there with + * pre gen9 so return 128MB always. + */ + if (INTEL_GEN(i915) < 9) + i915->edram_size_mb = 128; + else + i915->edram_size_mb = gen9_edram_size_mb(i915, edram_cap); + + dev_info(i915->drm.dev, + "Found %uMB of eDRAM\n", i915->edram_size_mb); +} diff --git a/drivers/gpu/drm/i915/intel_dram.h b/drivers/gpu/drm/i915/intel_dram.h new file mode 100644 index 000000000000..4ba13c13162c --- /dev/null +++ b/drivers/gpu/drm/i915/intel_dram.h @@ -0,0 +1,14 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2020 Intel Corporation + */ + +#ifndef __INTEL_DRAM_H__ +#define __INTEL_DRAM_H__ + +struct drm_i915_private; + +void intel_dram_edram_detect(struct drm_i915_private *i915); +void intel_dram_detect(struct drm_i915_private *i915); + +#endif /* __INTEL_DRAM_H__ */ diff --git a/drivers/gpu/drm/i915/intel_gvt.c b/drivers/gpu/drm/i915/intel_gvt.c index e73fd752adef..21b91313cc5d 100644 --- a/drivers/gpu/drm/i915/intel_gvt.c +++ b/drivers/gpu/drm/i915/intel_gvt.c @@ -22,6 +22,7 @@ */ #include "i915_drv.h" +#include "i915_vgpu.h" #include "intel_gvt.h" /** @@ -124,6 +125,11 @@ bail: return 0; } +static inline bool intel_gvt_active(struct drm_i915_private *dev_priv) +{ + return dev_priv->gvt; +} + /** * intel_gvt_driver_remove - cleanup GVT components when i915 driver is * unbinding diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c index 22aa205793e5..8375054ba27d 100644 --- a/drivers/gpu/drm/i915/intel_pm.c +++ b/drivers/gpu/drm/i915/intel_pm.c @@ -40,12 +40,36 @@ #include "gt/intel_llc.h" #include "i915_drv.h" +#include "i915_fixed.h" #include "i915_irq.h" #include "i915_trace.h" #include "intel_pm.h" #include "intel_sideband.h" #include "../../../platform/x86/intel_ips.h" +/* Stores plane specific WM parameters */ +struct skl_wm_params { + bool x_tiled, y_tiled; + bool rc_surface; + bool is_planar; + u32 width; + u8 cpp; + u32 plane_pixel_rate; + u32 y_min_scanlines; + u32 plane_bytes_per_line; + uint_fixed_16_16_t plane_blocks_per_line; + uint_fixed_16_16_t y_tile_minimum; + u32 linetime_us; + u32 dbuf_block_size; +}; + +/* used in computing the new watermarks state */ +struct intel_wm_config { + unsigned int num_pipes_active; + bool sprites_enabled; + bool sprites_scaled; +}; + static void gen9_init_clock_gating(struct drm_i915_private *dev_priv) { if (HAS_LLC(dev_priv)) { @@ -128,16 +152,6 @@ static void glk_init_clock_gating(struct drm_i915_private *dev_priv) */ I915_WRITE(GEN9_CLKGATE_DIS_0, I915_READ(GEN9_CLKGATE_DIS_0) | PWM1_GATING_DIS | PWM2_GATING_DIS); - - /* WaDDIIOTimeout:glk */ - if (IS_GLK_REVID(dev_priv, 0, GLK_REVID_A1)) { - u32 val = I915_READ(CHICKEN_MISC_2); - val &= ~(GLK_CL0_PWR_DOWN | - GLK_CL1_PWR_DOWN | - GLK_CL2_PWR_DOWN); - I915_WRITE(CHICKEN_MISC_2, val); - } - } static void pnv_get_mem_freq(struct drm_i915_private *dev_priv) @@ -2776,7 +2790,7 @@ static bool ilk_validate_wm_level(int level, } static void ilk_compute_wm_level(const struct drm_i915_private *dev_priv, - const struct intel_crtc *intel_crtc, + const struct intel_crtc *crtc, int level, struct intel_crtc_state *crtc_state, const struct intel_plane_state *pristate, @@ -3107,7 +3121,7 @@ static bool ilk_validate_pipe_wm(const struct drm_i915_private *dev_priv, static int ilk_compute_pipe_wm(struct intel_crtc_state *crtc_state) { struct drm_i915_private *dev_priv = to_i915(crtc_state->uapi.crtc->dev); - struct intel_crtc *intel_crtc = to_intel_crtc(crtc_state->uapi.crtc); + struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); struct intel_pipe_wm *pipe_wm; struct intel_plane *plane; const struct intel_plane_state *plane_state; @@ -3147,7 +3161,7 @@ static int ilk_compute_pipe_wm(struct intel_crtc_state *crtc_state) usable_level = 0; memset(&pipe_wm->wm, 0, sizeof(pipe_wm->wm)); - ilk_compute_wm_level(dev_priv, intel_crtc, 0, crtc_state, + ilk_compute_wm_level(dev_priv, crtc, 0, crtc_state, pristate, sprstate, curstate, &pipe_wm->wm[0]); if (!ilk_validate_pipe_wm(dev_priv, pipe_wm)) @@ -3158,7 +3172,7 @@ static int ilk_compute_pipe_wm(struct intel_crtc_state *crtc_state) for (level = 1; level <= usable_level; level++) { struct intel_wm_level *wm = &pipe_wm->wm[level]; - ilk_compute_wm_level(dev_priv, intel_crtc, level, crtc_state, + ilk_compute_wm_level(dev_priv, crtc, level, crtc_state, pristate, sprstate, curstate, wm); /* @@ -3843,7 +3857,7 @@ static u16 intel_get_ddb_size(struct drm_i915_private *dev_priv) } static u8 skl_compute_dbuf_slices(const struct intel_crtc_state *crtc_state, - u32 active_pipes); + u8 active_pipes); static void skl_ddb_get_pipe_allocation_limits(struct drm_i915_private *dev_priv, @@ -4184,50 +4198,51 @@ static const struct dbuf_slice_conf_entry icl_allowed_dbufs[] = { .active_pipes = BIT(PIPE_A), .dbuf_mask = { - [PIPE_A] = BIT(DBUF_S1) - } + [PIPE_A] = BIT(DBUF_S1), + }, }, { .active_pipes = BIT(PIPE_B), .dbuf_mask = { - [PIPE_B] = BIT(DBUF_S1) - } + [PIPE_B] = BIT(DBUF_S1), + }, }, { .active_pipes = BIT(PIPE_A) | BIT(PIPE_B), .dbuf_mask = { [PIPE_A] = BIT(DBUF_S1), - [PIPE_B] = BIT(DBUF_S2) - } + [PIPE_B] = BIT(DBUF_S2), + }, }, { .active_pipes = BIT(PIPE_C), .dbuf_mask = { - [PIPE_C] = BIT(DBUF_S2) - } + [PIPE_C] = BIT(DBUF_S2), + }, }, { .active_pipes = BIT(PIPE_A) | BIT(PIPE_C), .dbuf_mask = { [PIPE_A] = BIT(DBUF_S1), - [PIPE_C] = BIT(DBUF_S2) - } + [PIPE_C] = BIT(DBUF_S2), + }, }, { .active_pipes = BIT(PIPE_B) | BIT(PIPE_C), .dbuf_mask = { [PIPE_B] = BIT(DBUF_S1), - [PIPE_C] = BIT(DBUF_S2) - } + [PIPE_C] = BIT(DBUF_S2), + }, }, { .active_pipes = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C), .dbuf_mask = { [PIPE_A] = BIT(DBUF_S1), [PIPE_B] = BIT(DBUF_S1), - [PIPE_C] = BIT(DBUF_S2) - } + [PIPE_C] = BIT(DBUF_S2), + }, }, + {} }; /* @@ -4246,100 +4261,100 @@ static const struct dbuf_slice_conf_entry tgl_allowed_dbufs[] = { .active_pipes = BIT(PIPE_A), .dbuf_mask = { - [PIPE_A] = BIT(DBUF_S1) | BIT(DBUF_S2) - } + [PIPE_A] = BIT(DBUF_S1) | BIT(DBUF_S2), + }, }, { .active_pipes = BIT(PIPE_B), .dbuf_mask = { - [PIPE_B] = BIT(DBUF_S1) | BIT(DBUF_S2) - } + [PIPE_B] = BIT(DBUF_S1) | BIT(DBUF_S2), + }, }, { .active_pipes = BIT(PIPE_A) | BIT(PIPE_B), .dbuf_mask = { [PIPE_A] = BIT(DBUF_S2), - [PIPE_B] = BIT(DBUF_S1) - } + [PIPE_B] = BIT(DBUF_S1), + }, }, { .active_pipes = BIT(PIPE_C), .dbuf_mask = { - [PIPE_C] = BIT(DBUF_S2) | BIT(DBUF_S1) - } + [PIPE_C] = BIT(DBUF_S2) | BIT(DBUF_S1), + }, }, { .active_pipes = BIT(PIPE_A) | BIT(PIPE_C), .dbuf_mask = { [PIPE_A] = BIT(DBUF_S1), - [PIPE_C] = BIT(DBUF_S2) - } + [PIPE_C] = BIT(DBUF_S2), + }, }, { .active_pipes = BIT(PIPE_B) | BIT(PIPE_C), .dbuf_mask = { [PIPE_B] = BIT(DBUF_S1), - [PIPE_C] = BIT(DBUF_S2) - } + [PIPE_C] = BIT(DBUF_S2), + }, }, { .active_pipes = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C), .dbuf_mask = { [PIPE_A] = BIT(DBUF_S1), [PIPE_B] = BIT(DBUF_S1), - [PIPE_C] = BIT(DBUF_S2) - } + [PIPE_C] = BIT(DBUF_S2), + }, }, { .active_pipes = BIT(PIPE_D), .dbuf_mask = { - [PIPE_D] = BIT(DBUF_S2) | BIT(DBUF_S1) - } + [PIPE_D] = BIT(DBUF_S2) | BIT(DBUF_S1), + }, }, { .active_pipes = BIT(PIPE_A) | BIT(PIPE_D), .dbuf_mask = { [PIPE_A] = BIT(DBUF_S1), - [PIPE_D] = BIT(DBUF_S2) - } + [PIPE_D] = BIT(DBUF_S2), + }, }, { .active_pipes = BIT(PIPE_B) | BIT(PIPE_D), .dbuf_mask = { [PIPE_B] = BIT(DBUF_S1), - [PIPE_D] = BIT(DBUF_S2) - } + [PIPE_D] = BIT(DBUF_S2), + }, }, { .active_pipes = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_D), .dbuf_mask = { [PIPE_A] = BIT(DBUF_S1), [PIPE_B] = BIT(DBUF_S1), - [PIPE_D] = BIT(DBUF_S2) - } + [PIPE_D] = BIT(DBUF_S2), + }, }, { .active_pipes = BIT(PIPE_C) | BIT(PIPE_D), .dbuf_mask = { [PIPE_C] = BIT(DBUF_S1), - [PIPE_D] = BIT(DBUF_S2) - } + [PIPE_D] = BIT(DBUF_S2), + }, }, { .active_pipes = BIT(PIPE_A) | BIT(PIPE_C) | BIT(PIPE_D), .dbuf_mask = { [PIPE_A] = BIT(DBUF_S1), [PIPE_C] = BIT(DBUF_S2), - [PIPE_D] = BIT(DBUF_S2) - } + [PIPE_D] = BIT(DBUF_S2), + }, }, { .active_pipes = BIT(PIPE_B) | BIT(PIPE_C) | BIT(PIPE_D), .dbuf_mask = { [PIPE_B] = BIT(DBUF_S1), [PIPE_C] = BIT(DBUF_S2), - [PIPE_D] = BIT(DBUF_S2) - } + [PIPE_D] = BIT(DBUF_S2), + }, }, { .active_pipes = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C) | BIT(PIPE_D), @@ -4347,19 +4362,18 @@ static const struct dbuf_slice_conf_entry tgl_allowed_dbufs[] = [PIPE_A] = BIT(DBUF_S1), [PIPE_B] = BIT(DBUF_S1), [PIPE_C] = BIT(DBUF_S2), - [PIPE_D] = BIT(DBUF_S2) - } + [PIPE_D] = BIT(DBUF_S2), + }, }, + {} }; -static u8 compute_dbuf_slices(enum pipe pipe, - u32 active_pipes, - const struct dbuf_slice_conf_entry *dbuf_slices, - int size) +static u8 compute_dbuf_slices(enum pipe pipe, u8 active_pipes, + const struct dbuf_slice_conf_entry *dbuf_slices) { int i; - for (i = 0; i < size; i++) { + for (i = 0; i < dbuf_slices[i].active_pipes; i++) { if (dbuf_slices[i].active_pipes == active_pipes) return dbuf_slices[i].dbuf_mask[pipe]; } @@ -4371,8 +4385,7 @@ static u8 compute_dbuf_slices(enum pipe pipe, * returns correspondent DBuf slice mask as stated in BSpec for particular * platform. */ -static u32 icl_compute_dbuf_slices(enum pipe pipe, - u32 active_pipes) +static u8 icl_compute_dbuf_slices(enum pipe pipe, u8 active_pipes) { /* * FIXME: For ICL this is still a bit unclear as prev BSpec revision @@ -4386,32 +4399,25 @@ static u32 icl_compute_dbuf_slices(enum pipe pipe, * still here - we will need it once those additional constraints * pop up. */ - return compute_dbuf_slices(pipe, active_pipes, - icl_allowed_dbufs, - ARRAY_SIZE(icl_allowed_dbufs)); + return compute_dbuf_slices(pipe, active_pipes, icl_allowed_dbufs); } -static u32 tgl_compute_dbuf_slices(enum pipe pipe, - u32 active_pipes) +static u8 tgl_compute_dbuf_slices(enum pipe pipe, u8 active_pipes) { - return compute_dbuf_slices(pipe, active_pipes, - tgl_allowed_dbufs, - ARRAY_SIZE(tgl_allowed_dbufs)); + return compute_dbuf_slices(pipe, active_pipes, tgl_allowed_dbufs); } static u8 skl_compute_dbuf_slices(const struct intel_crtc_state *crtc_state, - u32 active_pipes) + u8 active_pipes) { struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); enum pipe pipe = crtc->pipe; if (IS_GEN(dev_priv, 12)) - return tgl_compute_dbuf_slices(pipe, - active_pipes); + return tgl_compute_dbuf_slices(pipe, active_pipes); else if (IS_GEN(dev_priv, 11)) - return icl_compute_dbuf_slices(pipe, - active_pipes); + return icl_compute_dbuf_slices(pipe, active_pipes); /* * For anything else just return one slice yet. * Should be extended for other platforms. @@ -4470,14 +4476,10 @@ skl_get_total_relative_data_rate(struct intel_crtc_state *crtc_state, u64 *plane_data_rate, u64 *uv_plane_data_rate) { - struct drm_atomic_state *state = crtc_state->uapi.state; struct intel_plane *plane; const struct intel_plane_state *plane_state; u64 total_data_rate = 0; - if (WARN_ON(!state)) - return 0; - /* Calculate and cache data rate for each plane */ intel_atomic_crtc_state_for_each_plane_state(plane, plane_state, crtc_state) { enum plane_id plane_id = plane->id; @@ -4505,9 +4507,6 @@ icl_get_total_relative_data_rate(struct intel_crtc_state *crtc_state, const struct intel_plane_state *plane_state; u64 total_data_rate = 0; - if (WARN_ON(!crtc_state->uapi.state)) - return 0; - /* Calculate and cache data rate for each plane */ intel_atomic_crtc_state_for_each_plane_state(plane, plane_state, crtc_state) { enum plane_id plane_id = plane->id; @@ -4548,10 +4547,8 @@ icl_get_total_relative_data_rate(struct intel_crtc_state *crtc_state, static int skl_allocate_pipe_ddb(struct intel_crtc_state *crtc_state) { - struct drm_atomic_state *state = crtc_state->uapi.state; - struct drm_crtc *crtc = crtc_state->uapi.crtc; - struct drm_i915_private *dev_priv = to_i915(crtc->dev); - struct intel_crtc *intel_crtc = to_intel_crtc(crtc); + struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); + struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); struct skl_ddb_entry *alloc = &crtc_state->wm.skl.ddb; u16 alloc_size, start = 0; u16 total[I915_MAX_PLANES] = {}; @@ -4568,9 +4565,6 @@ skl_allocate_pipe_ddb(struct intel_crtc_state *crtc_state) memset(crtc_state->wm.skl.plane_ddb_y, 0, sizeof(crtc_state->wm.skl.plane_ddb_y)); memset(crtc_state->wm.skl.plane_ddb_uv, 0, sizeof(crtc_state->wm.skl.plane_ddb_uv)); - if (drm_WARN_ON(&dev_priv->drm, !state)) - return 0; - if (!crtc_state->hw.active) { alloc->start = alloc->end = 0; return 0; @@ -4609,7 +4603,7 @@ skl_allocate_pipe_ddb(struct intel_crtc_state *crtc_state) */ for (level = ilk_wm_max_level(dev_priv); level >= 0; level--) { blocks = 0; - for_each_plane_id_on_crtc(intel_crtc, plane_id) { + for_each_plane_id_on_crtc(crtc, plane_id) { const struct skl_plane_wm *wm = &crtc_state->wm.skl.optimal.planes[plane_id]; @@ -4646,7 +4640,7 @@ skl_allocate_pipe_ddb(struct intel_crtc_state *crtc_state) * watermark level, plus an extra share of the leftover blocks * proportional to its relative data rate. */ - for_each_plane_id_on_crtc(intel_crtc, plane_id) { + for_each_plane_id_on_crtc(crtc, plane_id) { const struct skl_plane_wm *wm = &crtc_state->wm.skl.optimal.planes[plane_id]; u64 rate; @@ -4685,7 +4679,7 @@ skl_allocate_pipe_ddb(struct intel_crtc_state *crtc_state) /* Set the actual DDB start/end points for each plane */ start = alloc->start; - for_each_plane_id_on_crtc(intel_crtc, plane_id) { + for_each_plane_id_on_crtc(crtc, plane_id) { struct skl_ddb_entry *plane_alloc = &crtc_state->wm.skl.plane_ddb_y[plane_id]; struct skl_ddb_entry *uv_plane_alloc = @@ -4719,7 +4713,7 @@ skl_allocate_pipe_ddb(struct intel_crtc_state *crtc_state) * that aren't actually possible. */ for (level++; level <= ilk_wm_max_level(dev_priv); level++) { - for_each_plane_id_on_crtc(intel_crtc, plane_id) { + for_each_plane_id_on_crtc(crtc, plane_id) { struct skl_plane_wm *wm = &crtc_state->wm.skl.optimal.planes[plane_id]; @@ -4756,7 +4750,7 @@ skl_allocate_pipe_ddb(struct intel_crtc_state *crtc_state) * Go back and disable the transition watermark if it turns out we * don't have enough DDB blocks for it. */ - for_each_plane_id_on_crtc(intel_crtc, plane_id) { + for_each_plane_id_on_crtc(crtc, plane_id) { struct skl_plane_wm *wm = &crtc_state->wm.skl.optimal.planes[plane_id]; @@ -5126,21 +5120,30 @@ static void skl_compute_transition_wm(const struct intel_crtc_state *crtc_state, { struct drm_device *dev = crtc_state->uapi.crtc->dev; const struct drm_i915_private *dev_priv = to_i915(dev); - u16 trans_min, trans_y_tile_min; - const u16 trans_amount = 10; /* This is configurable amount */ + u16 trans_min, trans_amount, trans_y_tile_min; u16 wm0_sel_res_b, trans_offset_b, res_blocks; - /* Transition WM are not recommended by HW team for GEN9 */ - if (INTEL_GEN(dev_priv) <= 9) - return; - /* Transition WM don't make any sense if ipc is disabled */ if (!dev_priv->ipc_enabled) return; - trans_min = 14; + /* + * WaDisableTWM:skl,kbl,cfl,bxt + * Transition WM are not recommended by HW team for GEN9 + */ + if (IS_GEN9_BC(dev_priv) || IS_BROXTON(dev_priv)) + return; + if (INTEL_GEN(dev_priv) >= 11) trans_min = 4; + else + trans_min = 14; + + /* Display WA #1140: glk,cnl */ + if (IS_CANNONLAKE(dev_priv) || IS_GEMINILAKE(dev_priv)) + trans_amount = 0; + else + trans_amount = 10; /* This is configurable amount */ trans_offset_b = trans_min + trans_amount; @@ -5167,7 +5170,6 @@ static void skl_compute_transition_wm(const struct intel_crtc_state *crtc_state, /* WA BUG:1938466 add one block for non y-tile planes */ if (IS_CNL_REVID(dev_priv, CNL_REVID_A0, CNL_REVID_A0)) res_blocks += 1; - } /* @@ -5410,8 +5412,12 @@ static bool skl_plane_wm_equals(struct drm_i915_private *dev_priv, int level, max_level = ilk_wm_max_level(dev_priv); for (level = 0; level <= max_level; level++) { - if (!skl_wm_level_equals(&wm1->wm[level], &wm2->wm[level]) || - !skl_wm_level_equals(&wm1->uv_wm[level], &wm2->uv_wm[level])) + /* + * We don't check uv_wm as the hardware doesn't actually + * use it. It only gets used for calculating the required + * ddb allocation. + */ + if (!skl_wm_level_equals(&wm1->wm[level], &wm2->wm[level])) return false; } @@ -5768,16 +5774,24 @@ skl_compute_wm(struct intel_atomic_state *state) ret = skl_build_pipe_wm(new_crtc_state); if (ret) return ret; - - ret = skl_wm_add_affected_planes(state, crtc); - if (ret) - return ret; } ret = skl_compute_ddb(state); if (ret) return ret; + /* + * skl_compute_ddb() will have adjusted the final watermarks + * based on how much ddb is available. Now we can actually + * check if the final watermarks changed. + */ + for_each_oldnew_intel_crtc_in_state(state, crtc, old_crtc_state, + new_crtc_state, i) { + ret = skl_wm_add_affected_planes(state, crtc); + if (ret) + return ret; + } + skl_print_wm_changes(state); return 0; @@ -6812,21 +6826,6 @@ static void icl_init_clock_gating(struct drm_i915_private *dev_priv) I915_WRITE(GEN10_DFR_RATIO_EN_AND_CHICKEN, I915_READ(GEN10_DFR_RATIO_EN_AND_CHICKEN) & ~DFR_DISABLE); - /* WaEnable32PlaneMode:icl */ - I915_WRITE(GEN9_CSFE_CHICKEN1_RCS, - _MASKED_BIT_ENABLE(GEN11_ENABLE_32_PLANE_MODE)); - - /* - * Wa_1408615072:icl,ehl (vsunit) - * Wa_1407596294:icl,ehl (hsunit) - */ - intel_uncore_rmw(&dev_priv->uncore, UNSLICE_UNIT_LEVEL_CLKGATE, - 0, VSUNIT_CLKGATE_DIS | HSUNIT_CLKGATE_DIS); - - /* Wa_1407352427:icl,ehl */ - intel_uncore_rmw(&dev_priv->uncore, UNSLICE_UNIT_LEVEL_CLKGATE2, - 0, PSDUNIT_CLKGATE_DIS); - /*Wa_14010594013:icl, ehl */ intel_uncore_rmw(&dev_priv->uncore, GEN8_CHICKEN_DCPR_1, 0, CNL_DELAY_PMRSP); @@ -6837,10 +6836,6 @@ static void tgl_init_clock_gating(struct drm_i915_private *dev_priv) u32 vd_pg_enable = 0; unsigned int i; - /* Wa_1408615072:tgl */ - intel_uncore_rmw(&dev_priv->uncore, UNSLICE_UNIT_LEVEL_CLKGATE2, - 0, VSUNIT_CLKGATE_DIS_TGL); - /* This is not a WA. Enable VD HCP & MFX_ENC powergate */ for (i = 0; i < I915_MAX_VCS; i++) { if (HAS_ENGINE(dev_priv, _VCS(i))) diff --git a/drivers/gpu/drm/i915/selftests/i915_active.c b/drivers/gpu/drm/i915/selftests/i915_active.c index ef572a0c2566..68bbb1580162 100644 --- a/drivers/gpu/drm/i915/selftests/i915_active.c +++ b/drivers/gpu/drm/i915/selftests/i915_active.c @@ -201,11 +201,57 @@ static int live_active_retire(void *arg) return err; } +static int live_active_barrier(void *arg) +{ + struct drm_i915_private *i915 = arg; + struct intel_engine_cs *engine; + struct live_active *active; + int err = 0; + + /* Check that we get a callback when requests retire upon waiting */ + + active = __live_alloc(i915); + if (!active) + return -ENOMEM; + + err = i915_active_acquire(&active->base); + if (err) + goto out; + + for_each_uabi_engine(engine, i915) { + err = i915_active_acquire_preallocate_barrier(&active->base, + engine); + if (err) + break; + + i915_active_acquire_barrier(&active->base); + } + + i915_active_release(&active->base); + + if (err == 0) + err = i915_active_wait(&active->base); + + if (err == 0 && !READ_ONCE(active->retired)) { + pr_err("i915_active not retired after flushing barriers!\n"); + err = -EINVAL; + } + +out: + __live_put(active); + + if (igt_flush_test(i915)) + err = -EIO; + + return err; +} + int i915_active_live_selftests(struct drm_i915_private *i915) { static const struct i915_subtest tests[] = { SUBTEST(live_active_wait), SUBTEST(live_active_retire), + SUBTEST(live_active_barrier), }; if (intel_gt_is_wedged(&i915->gt)) @@ -265,28 +311,40 @@ static void spin_unlock_wait(spinlock_t *lock) spin_unlock_irq(lock); } +static void active_flush(struct i915_active *ref, + struct i915_active_fence *active) +{ + struct dma_fence *fence; + + fence = xchg(__active_fence_slot(active), NULL); + if (!fence) + return; + + spin_lock_irq(fence->lock); + __list_del_entry(&active->cb.node); + spin_unlock_irq(fence->lock); /* serialise with fence->cb_list */ + atomic_dec(&ref->count); + + GEM_BUG_ON(!test_bit(DMA_FENCE_FLAG_SIGNALED_BIT, &fence->flags)); +} + void i915_active_unlock_wait(struct i915_active *ref) { if (i915_active_acquire_if_busy(ref)) { struct active_node *it, *n; + /* Wait for all active callbacks */ rcu_read_lock(); - rbtree_postorder_for_each_entry_safe(it, n, &ref->tree, node) { - struct dma_fence *f; - - /* Wait for all active callbacks */ - f = rcu_dereference(it->base.fence); - if (f) - spin_unlock_wait(f->lock); - } + active_flush(ref, &ref->excl); + rbtree_postorder_for_each_entry_safe(it, n, &ref->tree, node) + active_flush(ref, &it->base); rcu_read_unlock(); i915_active_release(ref); } /* And wait for the retire callback */ - spin_lock_irq(&ref->tree_lock); - spin_unlock_irq(&ref->tree_lock); + spin_unlock_wait(&ref->tree_lock); /* ... which may have been on a thread instead */ flush_work(&ref->work); diff --git a/drivers/gpu/drm/i915/selftests/i915_buddy.c b/drivers/gpu/drm/i915/selftests/i915_buddy.c index 1b856bae67b5..939a6caebb03 100644 --- a/drivers/gpu/drm/i915/selftests/i915_buddy.c +++ b/drivers/gpu/drm/i915/selftests/i915_buddy.c @@ -298,10 +298,12 @@ static void igt_mm_config(u64 *size, u64 *chunk_size) static int igt_buddy_alloc_smoke(void *arg) { struct i915_buddy_mm mm; - int max_order; + IGT_TIMEOUT(end_time); + I915_RND_STATE(prng); u64 chunk_size; u64 mm_size; - int err; + int *order; + int err, i; igt_mm_config(&mm_size, &chunk_size); @@ -313,10 +315,16 @@ static int igt_buddy_alloc_smoke(void *arg) return err; } - for (max_order = mm.max_order; max_order >= 0; max_order--) { + order = i915_random_order(mm.max_order + 1, &prng); + if (!order) + goto out_fini; + + for (i = 0; i <= mm.max_order; ++i) { struct i915_buddy_block *block; - int order; + int max_order = order[i]; + bool timeout = false; LIST_HEAD(blocks); + int order; u64 total; err = igt_check_mm(&mm); @@ -360,6 +368,11 @@ retry: } total += i915_buddy_block_size(&mm, block); + + if (__igt_timeout(end_time, NULL)) { + timeout = true; + break; + } } while (total < mm.size); if (!err) @@ -373,7 +386,7 @@ retry: pr_err("post-mm check failed\n"); } - if (err) + if (err || timeout) break; cond_resched(); @@ -382,6 +395,8 @@ retry: if (err == -ENOMEM) err = 0; + kfree(order); +out_fini: i915_buddy_fini(&mm); return err; diff --git a/drivers/gpu/drm/i915/selftests/i915_live_selftests.h b/drivers/gpu/drm/i915/selftests/i915_live_selftests.h index 34138c7bdd15..0a953bfc0585 100644 --- a/drivers/gpu/drm/i915/selftests/i915_live_selftests.h +++ b/drivers/gpu/drm/i915/selftests/i915_live_selftests.h @@ -43,6 +43,7 @@ selftest(reset, intel_reset_live_selftests) selftest(memory_region, intel_memory_region_live_selftests) selftest(hangcheck, intel_hangcheck_live_selftests) selftest(execlists, intel_execlists_live_selftests) +selftest(ring_submission, intel_ring_submission_live_selftests) selftest(perf, i915_perf_live_selftests) /* Here be dragons: keep last to run last! */ selftest(late_gt_pm, intel_gt_pm_late_selftests) diff --git a/drivers/gpu/drm/i915/selftests/igt_spinner.c b/drivers/gpu/drm/i915/selftests/igt_spinner.c index e8a58fe49c39..9ad4ab088466 100644 --- a/drivers/gpu/drm/i915/selftests/igt_spinner.c +++ b/drivers/gpu/drm/i915/selftests/igt_spinner.c @@ -183,7 +183,7 @@ igt_spinner_create_request(struct igt_spinner *spin, cancel_rq: if (err) { - i915_request_skip(rq, err); + i915_request_set_error_once(rq, err); i915_request_add(rq); } unpin_hws: diff --git a/drivers/gpu/drm/i915/selftests/mock_gem_device.c b/drivers/gpu/drm/i915/selftests/mock_gem_device.c index 3b8986983afc..754d0eb6beaa 100644 --- a/drivers/gpu/drm/i915/selftests/mock_gem_device.c +++ b/drivers/gpu/drm/i915/selftests/mock_gem_device.c @@ -144,7 +144,6 @@ struct drm_i915_private *mock_gem_device(void) goto put_device; } i915->drm.pdev = pdev; - i915->drm.dev_private = i915; intel_runtime_pm_init_early(&i915->runtime_pm); diff --git a/drivers/gpu/drm/mediatek/mtk_dpi.c b/drivers/gpu/drm/mediatek/mtk_dpi.c index 14fbe1c09ce9..4f0ce4cd5b8c 100644 --- a/drivers/gpu/drm/mediatek/mtk_dpi.c +++ b/drivers/gpu/drm/mediatek/mtk_dpi.c @@ -664,6 +664,16 @@ static unsigned int mt2701_calculate_factor(int clock) return 1; } +static unsigned int mt8183_calculate_factor(int clock) +{ + if (clock <= 27000) + return 8; + else if (clock <= 167000) + return 4; + else + return 2; +} + static const struct mtk_dpi_conf mt8173_conf = { .cal_factor = mt8173_calculate_factor, .reg_h_fre_con = 0xe0, @@ -675,6 +685,11 @@ static const struct mtk_dpi_conf mt2701_conf = { .edge_sel_en = true, }; +static const struct mtk_dpi_conf mt8183_conf = { + .cal_factor = mt8183_calculate_factor, + .reg_h_fre_con = 0xe0, +}; + static int mtk_dpi_probe(struct platform_device *pdev) { struct device *dev = &pdev->dev; @@ -770,6 +785,9 @@ static const struct of_device_id mtk_dpi_of_ids[] = { { .compatible = "mediatek,mt8173-dpi", .data = &mt8173_conf, }, + { .compatible = "mediatek,mt8183-dpi", + .data = &mt8183_conf, + }, { }, }; diff --git a/drivers/gpu/drm/mediatek/mtk_drm_crtc.c b/drivers/gpu/drm/mediatek/mtk_drm_crtc.c index 0dfcd1787e65..fe85e487e477 100644 --- a/drivers/gpu/drm/mediatek/mtk_drm_crtc.c +++ b/drivers/gpu/drm/mediatek/mtk_drm_crtc.c @@ -486,6 +486,7 @@ static void mtk_drm_crtc_hw_config(struct mtk_drm_crtc *mtk_crtc) } #if IS_REACHABLE(CONFIG_MTK_CMDQ) if (mtk_crtc->cmdq_client) { + mbox_flush(mtk_crtc->cmdq_client->chan, 2000); cmdq_handle = cmdq_pkt_create(mtk_crtc->cmdq_client, PAGE_SIZE); cmdq_pkt_clear_event(cmdq_handle, mtk_crtc->cmdq_event); cmdq_pkt_wfe(cmdq_handle, mtk_crtc->cmdq_event); @@ -636,10 +637,18 @@ static const struct drm_crtc_helper_funcs mtk_crtc_helper_funcs = { static int mtk_drm_crtc_init(struct drm_device *drm, struct mtk_drm_crtc *mtk_crtc, - struct drm_plane *primary, - struct drm_plane *cursor, unsigned int pipe) + unsigned int pipe) { - int ret; + struct drm_plane *primary = NULL; + struct drm_plane *cursor = NULL; + int i, ret; + + for (i = 0; i < mtk_crtc->layer_nr; i++) { + if (mtk_crtc->planes[i].type == DRM_PLANE_TYPE_PRIMARY) + primary = &mtk_crtc->planes[i]; + else if (mtk_crtc->planes[i].type == DRM_PLANE_TYPE_CURSOR) + cursor = &mtk_crtc->planes[i]; + } ret = drm_crtc_init_with_planes(drm, &mtk_crtc->base, primary, cursor, &mtk_crtc_funcs, NULL); @@ -689,11 +698,12 @@ static int mtk_drm_crtc_num_comp_planes(struct mtk_drm_crtc *mtk_crtc, } static inline -enum drm_plane_type mtk_drm_crtc_plane_type(unsigned int plane_idx) +enum drm_plane_type mtk_drm_crtc_plane_type(unsigned int plane_idx, + unsigned int num_planes) { if (plane_idx == 0) return DRM_PLANE_TYPE_PRIMARY; - else if (plane_idx == 1) + else if (plane_idx == (num_planes - 1)) return DRM_PLANE_TYPE_CURSOR; else return DRM_PLANE_TYPE_OVERLAY; @@ -712,7 +722,8 @@ static int mtk_drm_crtc_init_comp_planes(struct drm_device *drm_dev, ret = mtk_plane_init(drm_dev, &mtk_crtc->planes[mtk_crtc->layer_nr], BIT(pipe), - mtk_drm_crtc_plane_type(mtk_crtc->layer_nr), + mtk_drm_crtc_plane_type(mtk_crtc->layer_nr, + num_planes), mtk_ddp_comp_supported_rotations(comp)); if (ret) return ret; @@ -807,9 +818,7 @@ int mtk_drm_crtc_create(struct drm_device *drm_dev, return ret; } - ret = mtk_drm_crtc_init(drm_dev, mtk_crtc, &mtk_crtc->planes[0], - mtk_crtc->layer_nr > 1 ? &mtk_crtc->planes[1] : - NULL, pipe); + ret = mtk_drm_crtc_init(drm_dev, mtk_crtc, pipe); if (ret < 0) return ret; @@ -828,7 +837,8 @@ int mtk_drm_crtc_create(struct drm_device *drm_dev, drm_crtc_index(&mtk_crtc->base)); mtk_crtc->cmdq_client = NULL; } - ret = of_property_read_u32_index(dev->of_node, "mediatek,gce-events", + ret = of_property_read_u32_index(priv->mutex_node, + "mediatek,gce-events", drm_crtc_index(&mtk_crtc->base), &mtk_crtc->cmdq_event); if (ret) diff --git a/drivers/gpu/drm/mediatek/mtk_drm_ddp_comp.c b/drivers/gpu/drm/mediatek/mtk_drm_ddp_comp.c index 1f5a112bb034..57c88de9a329 100644 --- a/drivers/gpu/drm/mediatek/mtk_drm_ddp_comp.c +++ b/drivers/gpu/drm/mediatek/mtk_drm_ddp_comp.c @@ -471,6 +471,7 @@ int mtk_ddp_comp_init(struct device *dev, struct device_node *node, /* Only DMA capable components need the LARB property */ comp->larb_dev = NULL; if (type != MTK_DISP_OVL && + type != MTK_DISP_OVL_2L && type != MTK_DISP_RDMA && type != MTK_DISP_WDMA) return 0; diff --git a/drivers/gpu/drm/mediatek/mtk_drm_plane.c b/drivers/gpu/drm/mediatek/mtk_drm_plane.c index 914cc7619cd7..c2bd683a87c8 100644 --- a/drivers/gpu/drm/mediatek/mtk_drm_plane.c +++ b/drivers/gpu/drm/mediatek/mtk_drm_plane.c @@ -80,6 +80,7 @@ static int mtk_plane_atomic_async_check(struct drm_plane *plane, struct drm_plane_state *state) { struct drm_crtc_state *crtc_state; + int ret; if (plane != state->crtc->cursor) return -EINVAL; @@ -90,6 +91,11 @@ static int mtk_plane_atomic_async_check(struct drm_plane *plane, if (!plane->state->fb) return -EINVAL; + ret = mtk_drm_crtc_plane_check(state->crtc, plane, + to_mtk_plane_state(state)); + if (ret) + return ret; + if (state->state) crtc_state = drm_atomic_get_existing_crtc_state(state->state, state->crtc); @@ -115,6 +121,7 @@ static void mtk_plane_atomic_async_update(struct drm_plane *plane, plane->state->src_y = new_state->src_y; plane->state->src_h = new_state->src_h; plane->state->src_w = new_state->src_w; + swap(plane->state->fb, new_state->fb); state->pending.async_dirty = true; mtk_drm_crtc_async_update(new_state->crtc, plane, new_state); diff --git a/drivers/gpu/drm/msm/adreno/a5xx_gpu.c b/drivers/gpu/drm/msm/adreno/a5xx_gpu.c index 7d9e63e20ded..724024a2243a 100644 --- a/drivers/gpu/drm/msm/adreno/a5xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a5xx_gpu.c @@ -1446,18 +1446,31 @@ static const struct adreno_gpu_funcs funcs = { static void check_speed_bin(struct device *dev) { struct nvmem_cell *cell; - u32 bin, val; + u32 val; + + /* + * If the OPP table specifies a opp-supported-hw property then we have + * to set something with dev_pm_opp_set_supported_hw() or the table + * doesn't get populated so pick an arbitrary value that should + * ensure the default frequencies are selected but not conflict with any + * actual bins + */ + val = 0x80; cell = nvmem_cell_get(dev, "speed_bin"); - /* If a nvmem cell isn't defined, nothing to do */ - if (IS_ERR(cell)) - return; + if (!IS_ERR(cell)) { + void *buf = nvmem_cell_read(cell, NULL); + + if (!IS_ERR(buf)) { + u8 bin = *((u8 *) buf); - bin = *((u32 *) nvmem_cell_read(cell, NULL)); - nvmem_cell_put(cell); + val = (1 << bin); + kfree(buf); + } - val = (1 << bin); + nvmem_cell_put(cell); + } dev_pm_opp_set_supported_hw(dev, &val, 1); } diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c index 983afeaee737..c4e71abbdd53 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c +++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c @@ -2,6 +2,7 @@ /* Copyright (c) 2017-2019 The Linux Foundation. All rights reserved. */ #include <linux/clk.h> +#include <linux/dma-mapping.h> #include <linux/interconnect.h> #include <linux/pm_domain.h> #include <linux/pm_opp.h> @@ -796,12 +797,41 @@ bool a6xx_gmu_isidle(struct a6xx_gmu *gmu) return true; } +#define GBIF_CLIENT_HALT_MASK BIT(0) +#define GBIF_ARB_HALT_MASK BIT(1) + +static void a6xx_bus_clear_pending_transactions(struct adreno_gpu *adreno_gpu) +{ + struct msm_gpu *gpu = &adreno_gpu->base; + + if (!a6xx_has_gbif(adreno_gpu)) { + gpu_write(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL0, 0xf); + spin_until((gpu_read(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL1) & + 0xf) == 0xf); + gpu_write(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL0, 0); + + return; + } + + /* Halt new client requests on GBIF */ + gpu_write(gpu, REG_A6XX_GBIF_HALT, GBIF_CLIENT_HALT_MASK); + spin_until((gpu_read(gpu, REG_A6XX_GBIF_HALT_ACK) & + (GBIF_CLIENT_HALT_MASK)) == GBIF_CLIENT_HALT_MASK); + + /* Halt all AXI requests on GBIF */ + gpu_write(gpu, REG_A6XX_GBIF_HALT, GBIF_ARB_HALT_MASK); + spin_until((gpu_read(gpu, REG_A6XX_GBIF_HALT_ACK) & + (GBIF_ARB_HALT_MASK)) == GBIF_ARB_HALT_MASK); + + /* The GBIF halt needs to be explicitly cleared */ + gpu_write(gpu, REG_A6XX_GBIF_HALT, 0x0); +} + /* Gracefully try to shut down the GMU and by extension the GPU */ static void a6xx_gmu_shutdown(struct a6xx_gmu *gmu) { struct a6xx_gpu *a6xx_gpu = container_of(gmu, struct a6xx_gpu, gmu); struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; - struct msm_gpu *gpu = &adreno_gpu->base; u32 val; /* @@ -819,11 +849,7 @@ static void a6xx_gmu_shutdown(struct a6xx_gmu *gmu) return; } - /* Clear the VBIF pipe before shutting down */ - gpu_write(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL0, 0xf); - spin_until((gpu_read(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL1) & 0xf) - == 0xf); - gpu_write(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL0, 0); + a6xx_bus_clear_pending_transactions(adreno_gpu); /* tell the GMU we want to slumber */ a6xx_gmu_notify_slumber(gmu); @@ -895,21 +921,10 @@ int a6xx_gmu_stop(struct a6xx_gpu *a6xx_gpu) static void a6xx_gmu_memory_free(struct a6xx_gmu *gmu, struct a6xx_gmu_bo *bo) { - int count, i; - u64 iova; - if (IS_ERR_OR_NULL(bo)) return; - count = bo->size >> PAGE_SHIFT; - iova = bo->iova; - - for (i = 0; i < count; i++, iova += PAGE_SIZE) { - iommu_unmap(gmu->domain, iova, PAGE_SIZE); - __free_pages(bo->pages[i], 0); - } - - kfree(bo->pages); + dma_free_wc(gmu->dev, bo->size, bo->virt, bo->iova); kfree(bo); } @@ -917,7 +932,6 @@ static struct a6xx_gmu_bo *a6xx_gmu_memory_alloc(struct a6xx_gmu *gmu, size_t size) { struct a6xx_gmu_bo *bo; - int ret, count, i; bo = kzalloc(sizeof(*bo), GFP_KERNEL); if (!bo) @@ -925,86 +939,14 @@ static struct a6xx_gmu_bo *a6xx_gmu_memory_alloc(struct a6xx_gmu *gmu, bo->size = PAGE_ALIGN(size); - count = bo->size >> PAGE_SHIFT; + bo->virt = dma_alloc_wc(gmu->dev, bo->size, &bo->iova, GFP_KERNEL); - bo->pages = kcalloc(count, sizeof(struct page *), GFP_KERNEL); - if (!bo->pages) { + if (!bo->virt) { kfree(bo); return ERR_PTR(-ENOMEM); } - for (i = 0; i < count; i++) { - bo->pages[i] = alloc_page(GFP_KERNEL); - if (!bo->pages[i]) - goto err; - } - - bo->iova = gmu->uncached_iova_base; - - for (i = 0; i < count; i++) { - ret = iommu_map(gmu->domain, - bo->iova + (PAGE_SIZE * i), - page_to_phys(bo->pages[i]), PAGE_SIZE, - IOMMU_READ | IOMMU_WRITE); - - if (ret) { - DRM_DEV_ERROR(gmu->dev, "Unable to map GMU buffer object\n"); - - for (i = i - 1 ; i >= 0; i--) - iommu_unmap(gmu->domain, - bo->iova + (PAGE_SIZE * i), - PAGE_SIZE); - - goto err; - } - } - - bo->virt = vmap(bo->pages, count, VM_IOREMAP, - pgprot_writecombine(PAGE_KERNEL)); - if (!bo->virt) - goto err; - - /* Align future IOVA addresses on 1MB boundaries */ - gmu->uncached_iova_base += ALIGN(size, SZ_1M); - return bo; - -err: - for (i = 0; i < count; i++) { - if (bo->pages[i]) - __free_pages(bo->pages[i], 0); - } - - kfree(bo->pages); - kfree(bo); - - return ERR_PTR(-ENOMEM); -} - -static int a6xx_gmu_memory_probe(struct a6xx_gmu *gmu) -{ - int ret; - - /* - * The GMU address space is hardcoded to treat the range - * 0x60000000 - 0x80000000 as un-cached memory. All buffers shared - * between the GMU and the CPU will live in this space - */ - gmu->uncached_iova_base = 0x60000000; - - - gmu->domain = iommu_domain_alloc(&platform_bus_type); - if (!gmu->domain) - return -ENODEV; - - ret = iommu_attach_device(gmu->domain, gmu->dev); - - if (ret) { - iommu_domain_free(gmu->domain); - gmu->domain = NULL; - } - - return ret; } /* Return the 'arc-level' for the given frequency */ @@ -1264,10 +1206,6 @@ void a6xx_gmu_remove(struct a6xx_gpu *a6xx_gpu) a6xx_gmu_memory_free(gmu, gmu->hfi); - iommu_detach_device(gmu->domain, gmu->dev); - - iommu_domain_free(gmu->domain); - free_irq(gmu->gmu_irq, gmu); free_irq(gmu->hfi_irq, gmu); @@ -1288,7 +1226,15 @@ int a6xx_gmu_init(struct a6xx_gpu *a6xx_gpu, struct device_node *node) gmu->dev = &pdev->dev; - of_dma_configure(gmu->dev, node, true); + /* Pass force_dma false to require the DT to set the dma region */ + ret = of_dma_configure(gmu->dev, node, false); + if (ret) + return ret; + + /* Set the mask after the of_dma_configure() */ + ret = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(31)); + if (ret) + return ret; /* Fow now, don't do anything fancy until we get our feet under us */ gmu->idle_level = GMU_IDLE_STATE_ACTIVE; @@ -1300,11 +1246,6 @@ int a6xx_gmu_init(struct a6xx_gpu *a6xx_gpu, struct device_node *node) if (ret) goto err_put_device; - /* Set up the IOMMU context bank */ - ret = a6xx_gmu_memory_probe(gmu); - if (ret) - goto err_put_device; - /* Allocate memory for for the HFI queues */ gmu->hfi = a6xx_gmu_memory_alloc(gmu, SZ_16K); if (IS_ERR(gmu->hfi)) @@ -1350,11 +1291,6 @@ err_mmio: err_memory: a6xx_gmu_memory_free(gmu, gmu->hfi); - if (gmu->domain) { - iommu_detach_device(gmu->domain, gmu->dev); - - iommu_domain_free(gmu->domain); - } ret = -ENODEV; err_put_device: diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.h b/drivers/gpu/drm/msm/adreno/a6xx_gmu.h index 2af91ed7ed0c..4af65a36d5ca 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.h +++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.h @@ -12,8 +12,7 @@ struct a6xx_gmu_bo { void *virt; size_t size; - u64 iova; - struct page **pages; + dma_addr_t iova; }; /* @@ -49,9 +48,6 @@ struct a6xx_gmu { int hfi_irq; int gmu_irq; - struct iommu_domain *domain; - u64 uncached_iova_base; - struct device *gxpd; int idle_level; diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c index daf07800cde0..68af24150de5 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c @@ -378,18 +378,6 @@ static int a6xx_hw_init(struct msm_gpu *gpu) struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); int ret; - /* - * During a previous slumber, GBIF halt is asserted to ensure - * no further transaction can go through GPU before GPU - * headswitch is turned off. - * - * This halt is deasserted once headswitch goes off but - * incase headswitch doesn't goes off clear GBIF halt - * here to ensure GPU wake-up doesn't fail because of - * halted GPU transactions. - */ - gpu_write(gpu, REG_A6XX_GBIF_HALT, 0x0); - /* Make sure the GMU keeps the GPU on while we set it up */ a6xx_gmu_set_oob(&a6xx_gpu->gmu, GMU_OOB_GPU_SET); @@ -470,10 +458,12 @@ static int a6xx_hw_init(struct msm_gpu *gpu) /* Select CP0 to always count cycles */ gpu_write(gpu, REG_A6XX_CP_PERFCTR_CP_SEL_0, PERF_CP_ALWAYS_COUNT); - gpu_write(gpu, REG_A6XX_RB_NC_MODE_CNTL, 2 << 1); - gpu_write(gpu, REG_A6XX_TPL1_NC_MODE_CNTL, 2 << 1); - gpu_write(gpu, REG_A6XX_SP_NC_MODE_CNTL, 2 << 1); - gpu_write(gpu, REG_A6XX_UCHE_MODE_CNTL, 2 << 21); + if (adreno_is_a630(adreno_gpu)) { + gpu_write(gpu, REG_A6XX_RB_NC_MODE_CNTL, 2 << 1); + gpu_write(gpu, REG_A6XX_TPL1_NC_MODE_CNTL, 2 << 1); + gpu_write(gpu, REG_A6XX_SP_NC_MODE_CNTL, 2 << 1); + gpu_write(gpu, REG_A6XX_UCHE_MODE_CNTL, 2 << 21); + } /* Enable fault detection */ gpu_write(gpu, REG_A6XX_RBBM_INTERFACE_HANG_INT_CNTL, @@ -748,39 +738,6 @@ static const u32 a6xx_register_offsets[REG_ADRENO_REGISTER_MAX] = { REG_ADRENO_DEFINE(REG_ADRENO_CP_RB_CNTL, REG_A6XX_CP_RB_CNTL), }; -#define GBIF_CLIENT_HALT_MASK BIT(0) -#define GBIF_ARB_HALT_MASK BIT(1) - -static void a6xx_bus_clear_pending_transactions(struct adreno_gpu *adreno_gpu) -{ - struct msm_gpu *gpu = &adreno_gpu->base; - - if(!a6xx_has_gbif(adreno_gpu)){ - gpu_write(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL0, 0xf); - spin_until((gpu_read(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL1) & - 0xf) == 0xf); - gpu_write(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL0, 0); - - return; - } - - /* Halt new client requests on GBIF */ - gpu_write(gpu, REG_A6XX_GBIF_HALT, GBIF_CLIENT_HALT_MASK); - spin_until((gpu_read(gpu, REG_A6XX_GBIF_HALT_ACK) & - (GBIF_CLIENT_HALT_MASK)) == GBIF_CLIENT_HALT_MASK); - - /* Halt all AXI requests on GBIF */ - gpu_write(gpu, REG_A6XX_GBIF_HALT, GBIF_ARB_HALT_MASK); - spin_until((gpu_read(gpu, REG_A6XX_GBIF_HALT_ACK) & - (GBIF_ARB_HALT_MASK)) == GBIF_ARB_HALT_MASK); - - /* - * GMU needs DDR access in slumber path. Deassert GBIF halt now - * to allow for GMU to access system memory. - */ - gpu_write(gpu, REG_A6XX_GBIF_HALT, 0x0); -} - static int a6xx_pm_resume(struct msm_gpu *gpu) { struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); @@ -805,16 +762,6 @@ static int a6xx_pm_suspend(struct msm_gpu *gpu) devfreq_suspend_device(gpu->devfreq.devfreq); - /* - * Make sure the GMU is idle before continuing (because some transitions - * may use VBIF - */ - a6xx_gmu_wait_for_idle(&a6xx_gpu->gmu); - - /* Clear the VBIF pipe before shutting down */ - /* FIXME: This accesses the GPU - do we need to make sure it is on? */ - a6xx_bus_clear_pending_transactions(adreno_gpu); - return a6xx_gmu_stop(a6xx_gpu); } diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.h b/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.h index e67c20c415af..24c974c293e5 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.h +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.h @@ -379,7 +379,7 @@ static const struct a6xx_indexed_registers { }; static const struct a6xx_indexed_registers a6xx_cp_mempool_indexed = { - "CP_MEMPOOOL", REG_A6XX_CP_MEM_POOL_DBG_ADDR, + "CP_MEMPOOL", REG_A6XX_CP_MEM_POOL_DBG_ADDR, REG_A6XX_CP_MEM_POOL_DBG_DATA, 0x2060, }; diff --git a/drivers/gpu/drm/msm/adreno/a6xx_hfi.c b/drivers/gpu/drm/msm/adreno/a6xx_hfi.c index eda11abc5f01..e450e0b97211 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_hfi.c +++ b/drivers/gpu/drm/msm/adreno/a6xx_hfi.c @@ -7,6 +7,7 @@ #include "a6xx_gmu.h" #include "a6xx_gmu.xml.h" +#include "a6xx_gpu.h" #define HFI_MSG_ID(val) [val] = #val @@ -216,48 +217,82 @@ static int a6xx_hfi_send_perf_table(struct a6xx_gmu *gmu) NULL, 0); } -static int a6xx_hfi_send_bw_table(struct a6xx_gmu *gmu) +static void a618_build_bw_table(struct a6xx_hfi_msg_bw_table *msg) { - struct a6xx_hfi_msg_bw_table msg = { 0 }; + /* Send a single "off" entry since the 618 GMU doesn't do bus scaling */ + msg->bw_level_num = 1; + + msg->ddr_cmds_num = 3; + msg->ddr_wait_bitmask = 0x01; + + msg->ddr_cmds_addrs[0] = 0x50000; + msg->ddr_cmds_addrs[1] = 0x5003c; + msg->ddr_cmds_addrs[2] = 0x5000c; + + msg->ddr_cmds_data[0][0] = 0x40000000; + msg->ddr_cmds_data[0][1] = 0x40000000; + msg->ddr_cmds_data[0][2] = 0x40000000; /* - * The sdm845 GMU doesn't do bus frequency scaling on its own but it - * does need at least one entry in the list because it might be accessed - * when the GMU is shutting down. Send a single "off" entry. + * These are the CX (CNOC) votes - these are used by the GMU but the + * votes are known and fixed for the target */ + msg->cnoc_cmds_num = 1; + msg->cnoc_wait_bitmask = 0x01; + + msg->cnoc_cmds_addrs[0] = 0x5007c; + msg->cnoc_cmds_data[0][0] = 0x40000000; + msg->cnoc_cmds_data[1][0] = 0x60000001; +} - msg.bw_level_num = 1; +static void a6xx_build_bw_table(struct a6xx_hfi_msg_bw_table *msg) +{ + /* Send a single "off" entry since the 630 GMU doesn't do bus scaling */ + msg->bw_level_num = 1; - msg.ddr_cmds_num = 3; - msg.ddr_wait_bitmask = 0x07; + msg->ddr_cmds_num = 3; + msg->ddr_wait_bitmask = 0x07; - msg.ddr_cmds_addrs[0] = 0x50000; - msg.ddr_cmds_addrs[1] = 0x5005c; - msg.ddr_cmds_addrs[2] = 0x5000c; + msg->ddr_cmds_addrs[0] = 0x50000; + msg->ddr_cmds_addrs[1] = 0x5005c; + msg->ddr_cmds_addrs[2] = 0x5000c; - msg.ddr_cmds_data[0][0] = 0x40000000; - msg.ddr_cmds_data[0][1] = 0x40000000; - msg.ddr_cmds_data[0][2] = 0x40000000; + msg->ddr_cmds_data[0][0] = 0x40000000; + msg->ddr_cmds_data[0][1] = 0x40000000; + msg->ddr_cmds_data[0][2] = 0x40000000; /* * These are the CX (CNOC) votes. This is used but the values for the * sdm845 GMU are known and fixed so we can hard code them. */ - msg.cnoc_cmds_num = 3; - msg.cnoc_wait_bitmask = 0x05; + msg->cnoc_cmds_num = 3; + msg->cnoc_wait_bitmask = 0x05; - msg.cnoc_cmds_addrs[0] = 0x50034; - msg.cnoc_cmds_addrs[1] = 0x5007c; - msg.cnoc_cmds_addrs[2] = 0x5004c; + msg->cnoc_cmds_addrs[0] = 0x50034; + msg->cnoc_cmds_addrs[1] = 0x5007c; + msg->cnoc_cmds_addrs[2] = 0x5004c; - msg.cnoc_cmds_data[0][0] = 0x40000000; - msg.cnoc_cmds_data[0][1] = 0x00000000; - msg.cnoc_cmds_data[0][2] = 0x40000000; + msg->cnoc_cmds_data[0][0] = 0x40000000; + msg->cnoc_cmds_data[0][1] = 0x00000000; + msg->cnoc_cmds_data[0][2] = 0x40000000; + + msg->cnoc_cmds_data[1][0] = 0x60000001; + msg->cnoc_cmds_data[1][1] = 0x20000001; + msg->cnoc_cmds_data[1][2] = 0x60000001; +} + + +static int a6xx_hfi_send_bw_table(struct a6xx_gmu *gmu) +{ + struct a6xx_hfi_msg_bw_table msg = { 0 }; + struct a6xx_gpu *a6xx_gpu = container_of(gmu, struct a6xx_gpu, gmu); + struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; - msg.cnoc_cmds_data[1][0] = 0x60000001; - msg.cnoc_cmds_data[1][1] = 0x20000001; - msg.cnoc_cmds_data[1][2] = 0x60000001; + if (adreno_is_a618(adreno_gpu)) + a618_build_bw_table(&msg); + else + a6xx_build_bw_table(&msg); return a6xx_hfi_send_msg(gmu, HFI_H2F_MSG_BW_TABLE, &msg, sizeof(msg), NULL, 0); diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.c b/drivers/gpu/drm/msm/adreno/adreno_gpu.c index 7fd29829b2fa..1d5c43c22269 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_gpu.c +++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.c @@ -673,7 +673,7 @@ static char *adreno_gpu_ascii85_encode(u32 *src, size_t len) return NULL; for (i = 0; i < l; i++) - buf_itr += snprintf(buf + buf_itr, buffer_size - buf_itr, "%s", + buf_itr += scnprintf(buf + buf_itr, buffer_size - buf_itr, "%s", ascii85_encode(src[i], out)); return buf; diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c index 58d3400668f5..a1b79ee2bd9d 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c @@ -164,7 +164,6 @@ enum dpu_enc_rc_states { * clks and resources after IDLE_TIMEOUT time. * @vsync_event_work: worker to handle vsync event for autorefresh * @topology: topology of the display - * @mode_set_complete: flag to indicate modeset completion * @idle_timeout: idle timeout duration in milliseconds */ struct dpu_encoder_virt { @@ -202,7 +201,6 @@ struct dpu_encoder_virt { struct delayed_work delayed_off_work; struct kthread_work vsync_event_work; struct msm_display_topology topology; - bool mode_set_complete; u32 idle_timeout; }; @@ -461,7 +459,7 @@ void dpu_encoder_helper_split_config( struct msm_display_info *disp_info; if (!phys_enc->hw_mdptop || !phys_enc->parent) { - DPU_ERROR("invalid arg(s), encoder %d\n", phys_enc != 0); + DPU_ERROR("invalid arg(s), encoder %d\n", phys_enc != NULL); return; } @@ -562,12 +560,13 @@ static int dpu_encoder_virt_atomic_check( const struct drm_display_mode *mode; struct drm_display_mode *adj_mode; struct msm_display_topology topology; + struct dpu_global_state *global_state; int i = 0; int ret = 0; if (!drm_enc || !crtc_state || !conn_state) { DPU_ERROR("invalid arg(s), drm_enc %d, crtc/conn state %d/%d\n", - drm_enc != 0, crtc_state != 0, conn_state != 0); + drm_enc != NULL, crtc_state != NULL, conn_state != NULL); return -EINVAL; } @@ -578,6 +577,7 @@ static int dpu_encoder_virt_atomic_check( dpu_kms = to_dpu_kms(priv->kms); mode = &crtc_state->mode; adj_mode = &crtc_state->adjusted_mode; + global_state = dpu_kms_get_existing_global_state(dpu_kms); trace_dpu_enc_atomic_check(DRMID(drm_enc)); /* @@ -609,17 +609,15 @@ static int dpu_encoder_virt_atomic_check( topology = dpu_encoder_get_topology(dpu_enc, dpu_kms, adj_mode); - /* Reserve dynamic resources now. Indicating AtomicTest phase */ + /* Reserve dynamic resources now. */ if (!ret) { /* * Avoid reserving resources when mode set is pending. Topology * info may not be available to complete reservation. */ - if (drm_atomic_crtc_needs_modeset(crtc_state) - && dpu_enc->mode_set_complete) { - ret = dpu_rm_reserve(&dpu_kms->rm, drm_enc, crtc_state, - topology, true); - dpu_enc->mode_set_complete = false; + if (drm_atomic_crtc_needs_modeset(crtc_state)) { + ret = dpu_rm_reserve(&dpu_kms->rm, global_state, + drm_enc, crtc_state, topology); } } @@ -956,12 +954,13 @@ static void dpu_encoder_virt_mode_set(struct drm_encoder *drm_enc, struct drm_connector *conn = NULL, *conn_iter; struct drm_crtc *drm_crtc; struct dpu_crtc_state *cstate; - struct dpu_rm_hw_iter hw_iter; + struct dpu_global_state *global_state; struct msm_display_topology topology; - struct dpu_hw_ctl *hw_ctl[MAX_CHANNELS_PER_ENC] = { NULL }; - struct dpu_hw_mixer *hw_lm[MAX_CHANNELS_PER_ENC] = { NULL }; - int num_lm = 0, num_ctl = 0; - int i, j, ret; + struct dpu_hw_blk *hw_pp[MAX_CHANNELS_PER_ENC]; + struct dpu_hw_blk *hw_ctl[MAX_CHANNELS_PER_ENC]; + struct dpu_hw_blk *hw_lm[MAX_CHANNELS_PER_ENC]; + int num_lm, num_ctl, num_pp; + int i, j; if (!drm_enc) { DPU_ERROR("invalid encoder\n"); @@ -975,6 +974,12 @@ static void dpu_encoder_virt_mode_set(struct drm_encoder *drm_enc, dpu_kms = to_dpu_kms(priv->kms); connector_list = &dpu_kms->dev->mode_config.connector_list; + global_state = dpu_kms_get_existing_global_state(dpu_kms); + if (IS_ERR_OR_NULL(global_state)) { + DPU_ERROR("Failed to get global state"); + return; + } + trace_dpu_enc_mode_set(DRMID(drm_enc)); list_for_each_entry(conn_iter, connector_list, head) @@ -995,77 +1000,57 @@ static void dpu_encoder_virt_mode_set(struct drm_encoder *drm_enc, topology = dpu_encoder_get_topology(dpu_enc, dpu_kms, adj_mode); - /* Reserve dynamic resources now. Indicating non-AtomicTest phase */ - ret = dpu_rm_reserve(&dpu_kms->rm, drm_enc, drm_crtc->state, - topology, false); - if (ret) { - DPU_ERROR_ENC(dpu_enc, - "failed to reserve hw resources, %d\n", ret); - return; - } - - dpu_rm_init_hw_iter(&hw_iter, drm_enc->base.id, DPU_HW_BLK_PINGPONG); - for (i = 0; i < MAX_CHANNELS_PER_ENC; i++) { - dpu_enc->hw_pp[i] = NULL; - if (!dpu_rm_get_hw(&dpu_kms->rm, &hw_iter)) - break; - dpu_enc->hw_pp[i] = (struct dpu_hw_pingpong *) hw_iter.hw; - } - - dpu_rm_init_hw_iter(&hw_iter, drm_enc->base.id, DPU_HW_BLK_CTL); - for (i = 0; i < MAX_CHANNELS_PER_ENC; i++) { - if (!dpu_rm_get_hw(&dpu_kms->rm, &hw_iter)) - break; - hw_ctl[i] = (struct dpu_hw_ctl *)hw_iter.hw; - num_ctl++; - } + /* Query resource that have been reserved in atomic check step. */ + num_pp = dpu_rm_get_assigned_resources(&dpu_kms->rm, global_state, + drm_enc->base.id, DPU_HW_BLK_PINGPONG, hw_pp, + ARRAY_SIZE(hw_pp)); + num_ctl = dpu_rm_get_assigned_resources(&dpu_kms->rm, global_state, + drm_enc->base.id, DPU_HW_BLK_CTL, hw_ctl, ARRAY_SIZE(hw_ctl)); + num_lm = dpu_rm_get_assigned_resources(&dpu_kms->rm, global_state, + drm_enc->base.id, DPU_HW_BLK_LM, hw_lm, ARRAY_SIZE(hw_lm)); - dpu_rm_init_hw_iter(&hw_iter, drm_enc->base.id, DPU_HW_BLK_LM); - for (i = 0; i < MAX_CHANNELS_PER_ENC; i++) { - if (!dpu_rm_get_hw(&dpu_kms->rm, &hw_iter)) - break; - hw_lm[i] = (struct dpu_hw_mixer *)hw_iter.hw; - num_lm++; - } + for (i = 0; i < MAX_CHANNELS_PER_ENC; i++) + dpu_enc->hw_pp[i] = i < num_pp ? to_dpu_hw_pingpong(hw_pp[i]) + : NULL; cstate = to_dpu_crtc_state(drm_crtc->state); for (i = 0; i < num_lm; i++) { int ctl_idx = (i < num_ctl) ? i : (num_ctl-1); - cstate->mixers[i].hw_lm = hw_lm[i]; - cstate->mixers[i].lm_ctl = hw_ctl[ctl_idx]; + cstate->mixers[i].hw_lm = to_dpu_hw_mixer(hw_lm[i]); + cstate->mixers[i].lm_ctl = to_dpu_hw_ctl(hw_ctl[ctl_idx]); } cstate->num_mixers = num_lm; for (i = 0; i < dpu_enc->num_phys_encs; i++) { + int num_blk; + struct dpu_hw_blk *hw_blk[MAX_CHANNELS_PER_ENC]; struct dpu_encoder_phys *phys = dpu_enc->phys_encs[i]; if (!dpu_enc->hw_pp[i]) { DPU_ERROR_ENC(dpu_enc, "no pp block assigned at idx: %d\n", i); - goto error; + return; } if (!hw_ctl[i]) { DPU_ERROR_ENC(dpu_enc, "no ctl block assigned at idx: %d\n", i); - goto error; + return; } phys->hw_pp = dpu_enc->hw_pp[i]; - phys->hw_ctl = hw_ctl[i]; + phys->hw_ctl = to_dpu_hw_ctl(hw_ctl[i]); - dpu_rm_init_hw_iter(&hw_iter, drm_enc->base.id, - DPU_HW_BLK_INTF); - for (j = 0; j < MAX_CHANNELS_PER_ENC; j++) { + num_blk = dpu_rm_get_assigned_resources(&dpu_kms->rm, + global_state, drm_enc->base.id, DPU_HW_BLK_INTF, + hw_blk, ARRAY_SIZE(hw_blk)); + for (j = 0; j < num_blk; j++) { struct dpu_hw_intf *hw_intf; - if (!dpu_rm_get_hw(&dpu_kms->rm, &hw_iter)) - break; - - hw_intf = (struct dpu_hw_intf *)hw_iter.hw; + hw_intf = to_dpu_hw_intf(hw_blk[i]); if (hw_intf->idx == phys->intf_idx) phys->hw_intf = hw_intf; } @@ -1073,18 +1058,13 @@ static void dpu_encoder_virt_mode_set(struct drm_encoder *drm_enc, if (!phys->hw_intf) { DPU_ERROR_ENC(dpu_enc, "no intf block assigned at idx: %d\n", i); - goto error; + return; } phys->connector = conn->state->connector; if (phys->ops.mode_set) phys->ops.mode_set(phys, mode, adj_mode); } - - dpu_enc->mode_set_complete = true; - -error: - dpu_rm_release(&dpu_kms->rm, drm_enc); } static void _dpu_encoder_virt_enable_helper(struct drm_encoder *drm_enc) @@ -1181,6 +1161,7 @@ static void dpu_encoder_virt_disable(struct drm_encoder *drm_enc) struct dpu_encoder_virt *dpu_enc = NULL; struct msm_drm_private *priv; struct dpu_kms *dpu_kms; + struct dpu_global_state *global_state; int i = 0; if (!drm_enc) { @@ -1199,6 +1180,7 @@ static void dpu_encoder_virt_disable(struct drm_encoder *drm_enc) priv = drm_enc->dev->dev_private; dpu_kms = to_dpu_kms(priv->kms); + global_state = dpu_kms_get_existing_global_state(dpu_kms); trace_dpu_enc_disable(DRMID(drm_enc)); @@ -1228,7 +1210,7 @@ static void dpu_encoder_virt_disable(struct drm_encoder *drm_enc) DPU_DEBUG_ENC(dpu_enc, "encoder disabled\n"); - dpu_rm_release(&dpu_kms->rm, drm_enc); + dpu_rm_release(global_state, drm_enc); mutex_unlock(&dpu_enc->enc_lock); } @@ -1964,7 +1946,7 @@ static int dpu_encoder_virt_add_phys_encs( if (IS_ERR_OR_NULL(enc)) { DPU_ERROR_ENC(dpu_enc, "failed to init vid enc: %ld\n", PTR_ERR(enc)); - return enc == 0 ? -EINVAL : PTR_ERR(enc); + return enc == NULL ? -EINVAL : PTR_ERR(enc); } dpu_enc->phys_encs[dpu_enc->num_phys_encs] = enc; @@ -1977,7 +1959,7 @@ static int dpu_encoder_virt_add_phys_encs( if (IS_ERR_OR_NULL(enc)) { DPU_ERROR_ENC(dpu_enc, "failed to init cmd enc: %ld\n", PTR_ERR(enc)); - return enc == 0 ? -EINVAL : PTR_ERR(enc); + return enc == NULL ? -EINVAL : PTR_ERR(enc); } dpu_enc->phys_encs[dpu_enc->num_phys_encs] = enc; @@ -2008,7 +1990,7 @@ static int dpu_encoder_setup_display(struct dpu_encoder_virt *dpu_enc, struct dpu_enc_phys_init_params phys_params; if (!dpu_enc) { - DPU_ERROR("invalid arg(s), enc %d\n", dpu_enc != 0); + DPU_ERROR("invalid arg(s), enc %d\n", dpu_enc != NULL); return -EINVAL; } diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c index 39e1e280ba44..8493d68ad841 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c @@ -411,7 +411,7 @@ static void _dpu_encoder_phys_cmd_pingpong_config( to_dpu_encoder_phys_cmd(phys_enc); if (!phys_enc->hw_pp || !phys_enc->hw_ctl->ops.setup_intf_cfg) { - DPU_ERROR("invalid arg(s), enc %d\n", phys_enc != 0); + DPU_ERROR("invalid arg(s), enc %d\n", phys_enc != NULL); return; } @@ -440,7 +440,7 @@ static void dpu_encoder_phys_cmd_enable_helper( u32 flush_mask = 0; if (!phys_enc->hw_pp) { - DPU_ERROR("invalid arg(s), encoder %d\n", phys_enc != 0); + DPU_ERROR("invalid arg(s), encoder %d\n", phys_enc != NULL); return; } diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c index c71c18de5966..b5a49050d131 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c @@ -239,7 +239,7 @@ static void dpu_encoder_phys_vid_setup_timing_engine( struct dpu_hw_intf_cfg intf_cfg = { 0 }; if (!phys_enc->hw_ctl->ops.setup_intf_cfg) { - DPU_ERROR("invalid encoder %d\n", phys_enc != 0); + DPU_ERROR("invalid encoder %d\n", phys_enc != NULL); return; } @@ -559,7 +559,7 @@ static void dpu_encoder_phys_vid_disable(struct dpu_encoder_phys *phys_enc) if (!phys_enc->hw_intf) { DPU_ERROR("invalid hw_intf %d hw_ctl %d\n", - phys_enc->hw_intf != 0, phys_enc->hw_ctl != 0); + phys_enc->hw_intf != NULL, phys_enc->hw_ctl != NULL); return; } diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c index 528632690f1e..a05282dede91 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c @@ -255,13 +255,13 @@ static const struct dpu_format dpu_format_map[] = { INTERLEAVED_RGB_FMT(RGB565, 0, COLOR_5BIT, COLOR_6BIT, COLOR_5BIT, - C2_R_Cr, C0_G_Y, C1_B_Cb, 0, 3, + C1_B_Cb, C0_G_Y, C2_R_Cr, 0, 3, false, 2, 0, DPU_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(BGR565, 0, COLOR_5BIT, COLOR_6BIT, COLOR_5BIT, - C1_B_Cb, C0_G_Y, C2_R_Cr, 0, 3, + C2_R_Cr, C0_G_Y, C1_B_Cb, 0, 3, false, 2, 0, DPU_FETCH_LINEAR, 1), diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h index 85468981632d..0ead64d3f63d 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h @@ -90,6 +90,16 @@ struct dpu_hw_intf { }; /** + * to_dpu_hw_intf - convert base object dpu_hw_base to container + * @hw: Pointer to base hardware block + * return: Pointer to hardware block container + */ +static inline struct dpu_hw_intf *to_dpu_hw_intf(struct dpu_hw_blk *hw) +{ + return container_of(hw, struct dpu_hw_intf, base); +} + +/** * dpu_hw_intf_init(): Initializes the intf driver for the passed * interface idx. * @idx: interface index for which driver object is required diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h index 3d6f46b1db30..d73cb73e938b 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h @@ -97,6 +97,16 @@ struct dpu_hw_pingpong { }; /** + * to_dpu_hw_pingpong - convert base object dpu_hw_base to container + * @hw: Pointer to base hardware block + * return: Pointer to hardware block container + */ +static inline struct dpu_hw_pingpong *to_dpu_hw_pingpong(struct dpu_hw_blk *hw) +{ + return container_of(hw, struct dpu_hw_pingpong, base); +} + +/** * dpu_hw_pingpong_init - initializes the pingpong driver for the passed * pingpong idx. * @idx: Pingpong index for which driver object is required diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c index cb08fafb1dc1..ce19f1d39367 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c @@ -138,16 +138,12 @@ static int _dpu_debugfs_show_regset32(struct seq_file *s, void *data) { struct dpu_debugfs_regset32 *regset = s->private; struct dpu_kms *dpu_kms = regset->dpu_kms; - struct drm_device *dev; - struct msm_drm_private *priv; void __iomem *base; uint32_t i, addr; if (!dpu_kms->mmio) return 0; - dev = dpu_kms->dev; - priv = dev->dev_private; base = dpu_kms->mmio + regset->offset; /* insert padding spaces, if needed */ @@ -228,6 +224,85 @@ static int dpu_kms_debugfs_init(struct msm_kms *kms, struct drm_minor *minor) } #endif +/* Global/shared object state funcs */ + +/* + * This is a helper that returns the private state currently in operation. + * Note that this would return the "old_state" if called in the atomic check + * path, and the "new_state" after the atomic swap has been done. + */ +struct dpu_global_state * +dpu_kms_get_existing_global_state(struct dpu_kms *dpu_kms) +{ + return to_dpu_global_state(dpu_kms->global_state.state); +} + +/* + * This acquires the modeset lock set aside for global state, creates + * a new duplicated private object state. + */ +struct dpu_global_state *dpu_kms_get_global_state(struct drm_atomic_state *s) +{ + struct msm_drm_private *priv = s->dev->dev_private; + struct dpu_kms *dpu_kms = to_dpu_kms(priv->kms); + struct drm_private_state *priv_state; + int ret; + + ret = drm_modeset_lock(&dpu_kms->global_state_lock, s->acquire_ctx); + if (ret) + return ERR_PTR(ret); + + priv_state = drm_atomic_get_private_obj_state(s, + &dpu_kms->global_state); + if (IS_ERR(priv_state)) + return ERR_CAST(priv_state); + + return to_dpu_global_state(priv_state); +} + +static struct drm_private_state * +dpu_kms_global_duplicate_state(struct drm_private_obj *obj) +{ + struct dpu_global_state *state; + + state = kmemdup(obj->state, sizeof(*state), GFP_KERNEL); + if (!state) + return NULL; + + __drm_atomic_helper_private_obj_duplicate_state(obj, &state->base); + + return &state->base; +} + +static void dpu_kms_global_destroy_state(struct drm_private_obj *obj, + struct drm_private_state *state) +{ + struct dpu_global_state *dpu_state = to_dpu_global_state(state); + + kfree(dpu_state); +} + +static const struct drm_private_state_funcs dpu_kms_global_state_funcs = { + .atomic_duplicate_state = dpu_kms_global_duplicate_state, + .atomic_destroy_state = dpu_kms_global_destroy_state, +}; + +static int dpu_kms_global_obj_init(struct dpu_kms *dpu_kms) +{ + struct dpu_global_state *state; + + drm_modeset_lock_init(&dpu_kms->global_state_lock); + + state = kzalloc(sizeof(*state), GFP_KERNEL); + if (!state) + return -ENOMEM; + + drm_atomic_private_obj_init(dpu_kms->dev, &dpu_kms->global_state, + &state->base, + &dpu_kms_global_state_funcs); + return 0; +} + static int dpu_kms_enable_vblank(struct msm_kms *kms, struct drm_crtc *crtc) { return dpu_crtc_vblank(crtc, true); @@ -267,8 +342,6 @@ static ktime_t dpu_kms_vsync_time(struct msm_kms *kms, struct drm_crtc *crtc) static void dpu_kms_prepare_commit(struct msm_kms *kms, struct drm_atomic_state *state) { - struct dpu_kms *dpu_kms; - struct drm_device *dev; struct drm_crtc *crtc; struct drm_crtc_state *crtc_state; struct drm_encoder *encoder; @@ -276,8 +349,6 @@ static void dpu_kms_prepare_commit(struct msm_kms *kms, if (!kms) return; - dpu_kms = to_dpu_kms(kms); - dev = dpu_kms->dev; /* Call prepare_commit for all affected encoders */ for_each_new_crtc_in_state(state, crtc, crtc_state, i) { @@ -552,11 +623,8 @@ static long dpu_kms_round_pixclk(struct msm_kms *kms, unsigned long rate, static void _dpu_kms_hw_destroy(struct dpu_kms *dpu_kms) { - struct drm_device *dev; int i; - dev = dpu_kms->dev; - if (dpu_kms->hw_intr) dpu_hw_intr_destroy(dpu_kms->hw_intr); dpu_kms->hw_intr = NULL; @@ -760,7 +828,6 @@ static int dpu_kms_hw_init(struct msm_kms *kms) { struct dpu_kms *dpu_kms; struct drm_device *dev; - struct msm_drm_private *priv; int i, rc = -EINVAL; if (!kms) { @@ -770,7 +837,10 @@ static int dpu_kms_hw_init(struct msm_kms *kms) dpu_kms = to_dpu_kms(kms); dev = dpu_kms->dev; - priv = dev->dev_private; + + rc = dpu_kms_global_obj_init(dpu_kms); + if (rc) + return rc; atomic_set(&dpu_kms->bandwidth_ref, 0); @@ -1018,10 +1088,8 @@ static int __maybe_unused dpu_runtime_suspend(struct device *dev) int rc = -1; struct platform_device *pdev = to_platform_device(dev); struct dpu_kms *dpu_kms = platform_get_drvdata(pdev); - struct drm_device *ddev; struct dss_module_power *mp = &dpu_kms->mp; - ddev = dpu_kms->dev; rc = msm_dss_enable_clk(mp->clk_config, mp->num_clk, false); if (rc) DPU_ERROR("clock disable failed rc:%d\n", rc); diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h index c6169e7df19d..211f5de99a44 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h @@ -111,6 +111,13 @@ struct dpu_kms { struct dpu_core_perf perf; + /* + * Global private object state, Do not access directly, use + * dpu_kms_global_get_state() + */ + struct drm_modeset_lock global_state_lock; + struct drm_private_obj global_state; + struct dpu_rm rm; bool rm_init; @@ -139,6 +146,25 @@ struct vsync_info { #define to_dpu_kms(x) container_of(x, struct dpu_kms, base) +#define to_dpu_global_state(x) container_of(x, struct dpu_global_state, base) + +/* Global private object state for tracking resources that are shared across + * multiple kms objects (planes/crtcs/etc). + */ +struct dpu_global_state { + struct drm_private_state base; + + uint32_t pingpong_to_enc_id[PINGPONG_MAX - PINGPONG_0]; + uint32_t mixer_to_enc_id[LM_MAX - LM_0]; + uint32_t ctl_to_enc_id[CTL_MAX - CTL_0]; + uint32_t intf_to_enc_id[INTF_MAX - INTF_0]; +}; + +struct dpu_global_state + *dpu_kms_get_existing_global_state(struct dpu_kms *dpu_kms); +struct dpu_global_state + *__must_check dpu_kms_get_global_state(struct drm_atomic_state *s); + /** * Debugfs functions - extra helper functions for debugfs support * diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_mdss.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_mdss.c index 29705e773a4b..80d3cfc14007 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_mdss.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_mdss.c @@ -12,6 +12,7 @@ #define to_dpu_mdss(x) container_of(x, struct dpu_mdss, base) +#define HW_REV 0x0 #define HW_INTR_STATUS 0x0010 /* Max BW defined in KBps */ @@ -22,6 +23,17 @@ struct dpu_irq_controller { struct irq_domain *domain; }; +struct dpu_hw_cfg { + u32 val; + u32 offset; +}; + +struct dpu_mdss_hw_init_handler { + u32 hw_rev; + u32 hw_reg_count; + struct dpu_hw_cfg* hw_cfg; +}; + struct dpu_mdss { struct msm_mdss base; void __iomem *mmio; @@ -32,6 +44,44 @@ struct dpu_mdss { u32 num_paths; }; +static struct dpu_hw_cfg hw_cfg[] = { + { + /* UBWC global settings */ + .val = 0x1E, + .offset = 0x144, + } +}; + +static struct dpu_mdss_hw_init_handler cfg_handler[] = { + { .hw_rev = DPU_HW_VER_620, + .hw_reg_count = ARRAY_SIZE(hw_cfg), + .hw_cfg = hw_cfg + }, +}; + +static void dpu_mdss_hw_init(struct dpu_mdss *dpu_mdss, u32 hw_rev) +{ + int i; + u32 count = 0; + struct dpu_hw_cfg *hw_cfg = NULL; + + for (i = 0; i < ARRAY_SIZE(cfg_handler); i++) { + if (cfg_handler[i].hw_rev == hw_rev) { + hw_cfg = cfg_handler[i].hw_cfg; + count = cfg_handler[i].hw_reg_count; + break; + } + } + + for (i = 0; i < count; i++ ) { + writel_relaxed(hw_cfg->val, + dpu_mdss->mmio + hw_cfg->offset); + hw_cfg++; + } + + return; +} + static int dpu_mdss_parse_data_bus_icc_path(struct drm_device *dev, struct dpu_mdss *dpu_mdss) { @@ -174,12 +224,18 @@ static int dpu_mdss_enable(struct msm_mdss *mdss) struct dpu_mdss *dpu_mdss = to_dpu_mdss(mdss); struct dss_module_power *mp = &dpu_mdss->mp; int ret; + u32 mdss_rev; dpu_mdss_icc_request_bw(mdss); ret = msm_dss_enable_clk(mp->clk_config, mp->num_clk, true); - if (ret) + if (ret) { DPU_ERROR("clock enable failed, ret:%d\n", ret); + return ret; + } + + mdss_rev = readl_relaxed(dpu_mdss->mmio + HW_REV); + dpu_mdss_hw_init(dpu_mdss, mdss_rev); return ret; } diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c index 23f5b1433b35..9b62451b01ee 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c @@ -12,8 +12,12 @@ #include "dpu_encoder.h" #include "dpu_trace.h" -#define RESERVED_BY_OTHER(h, r) \ - ((h)->enc_id && (h)->enc_id != r) + +static inline bool reserved_by_other(uint32_t *res_map, int idx, + uint32_t enc_id) +{ + return res_map[idx] && res_map[idx] != enc_id; +} /** * struct dpu_rm_requirements - Reservation requirements parameter bundle @@ -25,171 +29,43 @@ struct dpu_rm_requirements { struct dpu_encoder_hw_resources hw_res; }; - -/** - * struct dpu_rm_hw_blk - hardware block tracking list member - * @list: List head for list of all hardware blocks tracking items - * @id: Hardware ID number, within it's own space, ie. LM_X - * @enc_id: Encoder id to which this blk is binded - * @hw: Pointer to the hardware register access object for this block - */ -struct dpu_rm_hw_blk { - struct list_head list; - uint32_t id; - uint32_t enc_id; - struct dpu_hw_blk *hw; -}; - -void dpu_rm_init_hw_iter( - struct dpu_rm_hw_iter *iter, - uint32_t enc_id, - enum dpu_hw_blk_type type) -{ - memset(iter, 0, sizeof(*iter)); - iter->enc_id = enc_id; - iter->type = type; -} - -static bool _dpu_rm_get_hw_locked(struct dpu_rm *rm, struct dpu_rm_hw_iter *i) +int dpu_rm_destroy(struct dpu_rm *rm) { - struct list_head *blk_list; - - if (!rm || !i || i->type >= DPU_HW_BLK_MAX) { - DPU_ERROR("invalid rm\n"); - return false; - } + int i; - i->hw = NULL; - blk_list = &rm->hw_blks[i->type]; + for (i = 0; i < ARRAY_SIZE(rm->pingpong_blks); i++) { + struct dpu_hw_pingpong *hw; - if (i->blk && (&i->blk->list == blk_list)) { - DPU_DEBUG("attempt resume iteration past last\n"); - return false; - } - - i->blk = list_prepare_entry(i->blk, blk_list, list); - - list_for_each_entry_continue(i->blk, blk_list, list) { - if (i->enc_id == i->blk->enc_id) { - i->hw = i->blk->hw; - DPU_DEBUG("found type %d id %d for enc %d\n", - i->type, i->blk->id, i->enc_id); - return true; + if (rm->pingpong_blks[i]) { + hw = to_dpu_hw_pingpong(rm->pingpong_blks[i]); + dpu_hw_pingpong_destroy(hw); } } + for (i = 0; i < ARRAY_SIZE(rm->mixer_blks); i++) { + struct dpu_hw_mixer *hw; - DPU_DEBUG("no match, type %d for enc %d\n", i->type, i->enc_id); - - return false; -} - -bool dpu_rm_get_hw(struct dpu_rm *rm, struct dpu_rm_hw_iter *i) -{ - bool ret; - - mutex_lock(&rm->rm_lock); - ret = _dpu_rm_get_hw_locked(rm, i); - mutex_unlock(&rm->rm_lock); - - return ret; -} - -static void _dpu_rm_hw_destroy(enum dpu_hw_blk_type type, void *hw) -{ - switch (type) { - case DPU_HW_BLK_LM: - dpu_hw_lm_destroy(hw); - break; - case DPU_HW_BLK_CTL: - dpu_hw_ctl_destroy(hw); - break; - case DPU_HW_BLK_PINGPONG: - dpu_hw_pingpong_destroy(hw); - break; - case DPU_HW_BLK_INTF: - dpu_hw_intf_destroy(hw); - break; - case DPU_HW_BLK_SSPP: - /* SSPPs are not managed by the resource manager */ - case DPU_HW_BLK_TOP: - /* Top is a singleton, not managed in hw_blks list */ - case DPU_HW_BLK_MAX: - default: - DPU_ERROR("unsupported block type %d\n", type); - break; - } -} - -int dpu_rm_destroy(struct dpu_rm *rm) -{ - struct dpu_rm_hw_blk *hw_cur, *hw_nxt; - enum dpu_hw_blk_type type; - - for (type = 0; type < DPU_HW_BLK_MAX; type++) { - list_for_each_entry_safe(hw_cur, hw_nxt, &rm->hw_blks[type], - list) { - list_del(&hw_cur->list); - _dpu_rm_hw_destroy(type, hw_cur->hw); - kfree(hw_cur); + if (rm->mixer_blks[i]) { + hw = to_dpu_hw_mixer(rm->mixer_blks[i]); + dpu_hw_lm_destroy(hw); } } + for (i = 0; i < ARRAY_SIZE(rm->ctl_blks); i++) { + struct dpu_hw_ctl *hw; - mutex_destroy(&rm->rm_lock); - - return 0; -} - -static int _dpu_rm_hw_blk_create( - struct dpu_rm *rm, - const struct dpu_mdss_cfg *cat, - void __iomem *mmio, - enum dpu_hw_blk_type type, - uint32_t id, - const void *hw_catalog_info) -{ - struct dpu_rm_hw_blk *blk; - void *hw; - - switch (type) { - case DPU_HW_BLK_LM: - hw = dpu_hw_lm_init(id, mmio, cat); - break; - case DPU_HW_BLK_CTL: - hw = dpu_hw_ctl_init(id, mmio, cat); - break; - case DPU_HW_BLK_PINGPONG: - hw = dpu_hw_pingpong_init(id, mmio, cat); - break; - case DPU_HW_BLK_INTF: - hw = dpu_hw_intf_init(id, mmio, cat); - break; - case DPU_HW_BLK_SSPP: - /* SSPPs are not managed by the resource manager */ - case DPU_HW_BLK_TOP: - /* Top is a singleton, not managed in hw_blks list */ - case DPU_HW_BLK_MAX: - default: - DPU_ERROR("unsupported block type %d\n", type); - return -EINVAL; - } - - if (IS_ERR_OR_NULL(hw)) { - DPU_ERROR("failed hw object creation: type %d, err %ld\n", - type, PTR_ERR(hw)); - return -EFAULT; + if (rm->ctl_blks[i]) { + hw = to_dpu_hw_ctl(rm->ctl_blks[i]); + dpu_hw_ctl_destroy(hw); + } } + for (i = 0; i < ARRAY_SIZE(rm->intf_blks); i++) { + struct dpu_hw_intf *hw; - blk = kzalloc(sizeof(*blk), GFP_KERNEL); - if (!blk) { - _dpu_rm_hw_destroy(type, hw); - return -ENOMEM; + if (rm->intf_blks[i]) { + hw = to_dpu_hw_intf(rm->intf_blks[i]); + dpu_hw_intf_destroy(hw); + } } - blk->id = id; - blk->hw = hw; - blk->enc_id = 0; - list_add_tail(&blk->list, &rm->hw_blks[type]); - return 0; } @@ -198,7 +74,6 @@ int dpu_rm_init(struct dpu_rm *rm, void __iomem *mmio) { int rc, i; - enum dpu_hw_blk_type type; if (!rm || !cat || !mmio) { DPU_ERROR("invalid kms\n"); @@ -208,13 +83,9 @@ int dpu_rm_init(struct dpu_rm *rm, /* Clear, setup lists */ memset(rm, 0, sizeof(*rm)); - mutex_init(&rm->rm_lock); - - for (type = 0; type < DPU_HW_BLK_MAX; type++) - INIT_LIST_HEAD(&rm->hw_blks[type]); - /* Interrogate HW catalog and create tracking items for hw blocks */ for (i = 0; i < cat->mixer_count; i++) { + struct dpu_hw_mixer *hw; const struct dpu_lm_cfg *lm = &cat->mixer[i]; if (lm->pingpong == PINGPONG_MAX) { @@ -222,12 +93,17 @@ int dpu_rm_init(struct dpu_rm *rm, continue; } - rc = _dpu_rm_hw_blk_create(rm, cat, mmio, DPU_HW_BLK_LM, - cat->mixer[i].id, &cat->mixer[i]); - if (rc) { - DPU_ERROR("failed: lm hw not available\n"); + if (lm->id < LM_0 || lm->id >= LM_MAX) { + DPU_ERROR("skip mixer %d with invalid id\n", lm->id); + continue; + } + hw = dpu_hw_lm_init(lm->id, mmio, cat); + if (IS_ERR_OR_NULL(hw)) { + rc = PTR_ERR(hw); + DPU_ERROR("failed lm object creation: err %d\n", rc); goto fail; } + rm->mixer_blks[lm->id - LM_0] = &hw->base; if (!rm->lm_max_width) { rm->lm_max_width = lm->sblk->maxwidth; @@ -243,35 +119,59 @@ int dpu_rm_init(struct dpu_rm *rm, } for (i = 0; i < cat->pingpong_count; i++) { - rc = _dpu_rm_hw_blk_create(rm, cat, mmio, DPU_HW_BLK_PINGPONG, - cat->pingpong[i].id, &cat->pingpong[i]); - if (rc) { - DPU_ERROR("failed: pp hw not available\n"); + struct dpu_hw_pingpong *hw; + const struct dpu_pingpong_cfg *pp = &cat->pingpong[i]; + + if (pp->id < PINGPONG_0 || pp->id >= PINGPONG_MAX) { + DPU_ERROR("skip pingpong %d with invalid id\n", pp->id); + continue; + } + hw = dpu_hw_pingpong_init(pp->id, mmio, cat); + if (IS_ERR_OR_NULL(hw)) { + rc = PTR_ERR(hw); + DPU_ERROR("failed pingpong object creation: err %d\n", + rc); goto fail; } + rm->pingpong_blks[pp->id - PINGPONG_0] = &hw->base; } for (i = 0; i < cat->intf_count; i++) { - if (cat->intf[i].type == INTF_NONE) { + struct dpu_hw_intf *hw; + const struct dpu_intf_cfg *intf = &cat->intf[i]; + + if (intf->type == INTF_NONE) { DPU_DEBUG("skip intf %d with type none\n", i); continue; } - - rc = _dpu_rm_hw_blk_create(rm, cat, mmio, DPU_HW_BLK_INTF, - cat->intf[i].id, &cat->intf[i]); - if (rc) { - DPU_ERROR("failed: intf hw not available\n"); + if (intf->id < INTF_0 || intf->id >= INTF_MAX) { + DPU_ERROR("skip intf %d with invalid id\n", intf->id); + continue; + } + hw = dpu_hw_intf_init(intf->id, mmio, cat); + if (IS_ERR_OR_NULL(hw)) { + rc = PTR_ERR(hw); + DPU_ERROR("failed intf object creation: err %d\n", rc); goto fail; } + rm->intf_blks[intf->id - INTF_0] = &hw->base; } for (i = 0; i < cat->ctl_count; i++) { - rc = _dpu_rm_hw_blk_create(rm, cat, mmio, DPU_HW_BLK_CTL, - cat->ctl[i].id, &cat->ctl[i]); - if (rc) { - DPU_ERROR("failed: ctl hw not available\n"); + struct dpu_hw_ctl *hw; + const struct dpu_ctl_cfg *ctl = &cat->ctl[i]; + + if (ctl->id < CTL_0 || ctl->id >= CTL_MAX) { + DPU_ERROR("skip ctl %d with invalid id\n", ctl->id); + continue; + } + hw = dpu_hw_ctl_init(ctl->id, mmio, cat); + if (IS_ERR_OR_NULL(hw)) { + rc = PTR_ERR(hw); + DPU_ERROR("failed ctl object creation: err %d\n", rc); goto fail; } + rm->ctl_blks[ctl->id - CTL_0] = &hw->base; } return 0; @@ -279,7 +179,7 @@ int dpu_rm_init(struct dpu_rm *rm, fail: dpu_rm_destroy(rm); - return rc; + return rc ? rc : -EFAULT; } static bool _dpu_rm_needs_split_display(const struct msm_display_topology *top) @@ -288,85 +188,81 @@ static bool _dpu_rm_needs_split_display(const struct msm_display_topology *top) } /** + * _dpu_rm_check_lm_peer - check if a mixer is a peer of the primary + * @rm: dpu resource manager handle + * @primary_idx: index of primary mixer in rm->mixer_blks[] + * @peer_idx: index of other mixer in rm->mixer_blks[] + * @Return: true if rm->mixer_blks[peer_idx] is a peer of + * rm->mixer_blks[primary_idx] + */ +static bool _dpu_rm_check_lm_peer(struct dpu_rm *rm, int primary_idx, + int peer_idx) +{ + const struct dpu_lm_cfg *prim_lm_cfg; + const struct dpu_lm_cfg *peer_cfg; + + prim_lm_cfg = to_dpu_hw_mixer(rm->mixer_blks[primary_idx])->cap; + peer_cfg = to_dpu_hw_mixer(rm->mixer_blks[peer_idx])->cap; + + if (!test_bit(peer_cfg->id, &prim_lm_cfg->lm_pair_mask)) { + DPU_DEBUG("lm %d not peer of lm %d\n", peer_cfg->id, + peer_cfg->id); + return false; + } + return true; +} + +/** * _dpu_rm_check_lm_and_get_connected_blks - check if proposed layer mixer meets * proposed use case requirements, incl. hardwired dependent blocks like * pingpong * @rm: dpu resource manager handle * @enc_id: encoder id requesting for allocation - * @reqs: proposed use case requirements - * @lm: proposed layer mixer, function checks if lm, and all other hardwired - * blocks connected to the lm (pp) is available and appropriate - * @pp: output parameter, pingpong block attached to the layer mixer. - * NULL if pp was not available, or not matching requirements. - * @primary_lm: if non-null, this function check if lm is compatible primary_lm - * as well as satisfying all other requirements + * @lm_idx: index of proposed layer mixer in rm->mixer_blks[], function checks + * if lm, and all other hardwired blocks connected to the lm (pp) is + * available and appropriate + * @pp_idx: output parameter, index of pingpong block attached to the layer + * mixer in rm->pongpong_blks[]. * @Return: true if lm matches all requirements, false otherwise */ -static bool _dpu_rm_check_lm_and_get_connected_blks( - struct dpu_rm *rm, - uint32_t enc_id, - struct dpu_rm_requirements *reqs, - struct dpu_rm_hw_blk *lm, - struct dpu_rm_hw_blk **pp, - struct dpu_rm_hw_blk *primary_lm) +static bool _dpu_rm_check_lm_and_get_connected_blks(struct dpu_rm *rm, + struct dpu_global_state *global_state, + uint32_t enc_id, int lm_idx, int *pp_idx) { - const struct dpu_lm_cfg *lm_cfg = to_dpu_hw_mixer(lm->hw)->cap; - struct dpu_rm_hw_iter iter; - - *pp = NULL; - - DPU_DEBUG("check lm %d pp %d\n", - lm_cfg->id, lm_cfg->pingpong); - - /* Check if this layer mixer is a peer of the proposed primary LM */ - if (primary_lm) { - const struct dpu_lm_cfg *prim_lm_cfg = - to_dpu_hw_mixer(primary_lm->hw)->cap; - - if (!test_bit(lm_cfg->id, &prim_lm_cfg->lm_pair_mask)) { - DPU_DEBUG("lm %d not peer of lm %d\n", lm_cfg->id, - prim_lm_cfg->id); - return false; - } - } + const struct dpu_lm_cfg *lm_cfg; + int idx; /* Already reserved? */ - if (RESERVED_BY_OTHER(lm, enc_id)) { - DPU_DEBUG("lm %d already reserved\n", lm_cfg->id); + if (reserved_by_other(global_state->mixer_to_enc_id, lm_idx, enc_id)) { + DPU_DEBUG("lm %d already reserved\n", lm_idx + LM_0); return false; } - dpu_rm_init_hw_iter(&iter, 0, DPU_HW_BLK_PINGPONG); - while (_dpu_rm_get_hw_locked(rm, &iter)) { - if (iter.blk->id == lm_cfg->pingpong) { - *pp = iter.blk; - break; - } - } - - if (!*pp) { + lm_cfg = to_dpu_hw_mixer(rm->mixer_blks[lm_idx])->cap; + idx = lm_cfg->pingpong - PINGPONG_0; + if (idx < 0 || idx >= ARRAY_SIZE(rm->pingpong_blks)) { DPU_ERROR("failed to get pp on lm %d\n", lm_cfg->pingpong); return false; } - if (RESERVED_BY_OTHER(*pp, enc_id)) { - DPU_DEBUG("lm %d pp %d already reserved\n", lm->id, - (*pp)->id); + if (reserved_by_other(global_state->pingpong_to_enc_id, idx, enc_id)) { + DPU_DEBUG("lm %d pp %d already reserved\n", lm_cfg->id, + lm_cfg->pingpong); return false; } - + *pp_idx = idx; return true; } -static int _dpu_rm_reserve_lms(struct dpu_rm *rm, uint32_t enc_id, +static int _dpu_rm_reserve_lms(struct dpu_rm *rm, + struct dpu_global_state *global_state, + uint32_t enc_id, struct dpu_rm_requirements *reqs) { - struct dpu_rm_hw_blk *lm[MAX_BLOCKS]; - struct dpu_rm_hw_blk *pp[MAX_BLOCKS]; - struct dpu_rm_hw_iter iter_i, iter_j; - int lm_count = 0; - int i, rc = 0; + int lm_idx[MAX_BLOCKS]; + int pp_idx[MAX_BLOCKS]; + int i, j, lm_count = 0; if (!reqs->topology.num_lm) { DPU_ERROR("invalid number of lm: %d\n", reqs->topology.num_lm); @@ -374,36 +270,40 @@ static int _dpu_rm_reserve_lms(struct dpu_rm *rm, uint32_t enc_id, } /* Find a primary mixer */ - dpu_rm_init_hw_iter(&iter_i, 0, DPU_HW_BLK_LM); - while (lm_count != reqs->topology.num_lm && - _dpu_rm_get_hw_locked(rm, &iter_i)) { - memset(&lm, 0, sizeof(lm)); - memset(&pp, 0, sizeof(pp)); + for (i = 0; i < ARRAY_SIZE(rm->mixer_blks) && + lm_count < reqs->topology.num_lm; i++) { + if (!rm->mixer_blks[i]) + continue; lm_count = 0; - lm[lm_count] = iter_i.blk; + lm_idx[lm_count] = i; - if (!_dpu_rm_check_lm_and_get_connected_blks( - rm, enc_id, reqs, lm[lm_count], - &pp[lm_count], NULL)) + if (!_dpu_rm_check_lm_and_get_connected_blks(rm, global_state, + enc_id, i, &pp_idx[lm_count])) { continue; + } ++lm_count; /* Valid primary mixer found, find matching peers */ - dpu_rm_init_hw_iter(&iter_j, 0, DPU_HW_BLK_LM); + for (j = i + 1; j < ARRAY_SIZE(rm->mixer_blks) && + lm_count < reqs->topology.num_lm; j++) { + if (!rm->mixer_blks[j]) + continue; - while (lm_count != reqs->topology.num_lm && - _dpu_rm_get_hw_locked(rm, &iter_j)) { - if (iter_i.blk == iter_j.blk) + if (!_dpu_rm_check_lm_peer(rm, i, j)) { + DPU_DEBUG("lm %d not peer of lm %d\n", LM_0 + j, + LM_0 + i); continue; + } - if (!_dpu_rm_check_lm_and_get_connected_blks( - rm, enc_id, reqs, iter_j.blk, - &pp[lm_count], iter_i.blk)) + if (!_dpu_rm_check_lm_and_get_connected_blks(rm, + global_state, enc_id, j, + &pp_idx[lm_count])) { continue; + } - lm[lm_count] = iter_j.blk; + lm_idx[lm_count] = j; ++lm_count; } } @@ -413,65 +313,65 @@ static int _dpu_rm_reserve_lms(struct dpu_rm *rm, uint32_t enc_id, return -ENAVAIL; } - for (i = 0; i < ARRAY_SIZE(lm); i++) { - if (!lm[i]) - break; + for (i = 0; i < lm_count; i++) { + global_state->mixer_to_enc_id[lm_idx[i]] = enc_id; + global_state->pingpong_to_enc_id[pp_idx[i]] = enc_id; - lm[i]->enc_id = enc_id; - pp[i]->enc_id = enc_id; - - trace_dpu_rm_reserve_lms(lm[i]->id, enc_id, pp[i]->id); + trace_dpu_rm_reserve_lms(lm_idx[i] + LM_0, enc_id, + pp_idx[i] + PINGPONG_0); } - return rc; + return 0; } static int _dpu_rm_reserve_ctls( struct dpu_rm *rm, + struct dpu_global_state *global_state, uint32_t enc_id, const struct msm_display_topology *top) { - struct dpu_rm_hw_blk *ctls[MAX_BLOCKS]; - struct dpu_rm_hw_iter iter; - int i = 0, num_ctls = 0; - bool needs_split_display = false; - - memset(&ctls, 0, sizeof(ctls)); + int ctl_idx[MAX_BLOCKS]; + int i = 0, j, num_ctls; + bool needs_split_display; /* each hw_intf needs its own hw_ctrl to program its control path */ num_ctls = top->num_intf; needs_split_display = _dpu_rm_needs_split_display(top); - dpu_rm_init_hw_iter(&iter, 0, DPU_HW_BLK_CTL); - while (_dpu_rm_get_hw_locked(rm, &iter)) { - const struct dpu_hw_ctl *ctl = to_dpu_hw_ctl(iter.blk->hw); - unsigned long features = ctl->caps->features; + for (j = 0; j < ARRAY_SIZE(rm->ctl_blks); j++) { + const struct dpu_hw_ctl *ctl; + unsigned long features; bool has_split_display; - if (RESERVED_BY_OTHER(iter.blk, enc_id)) + if (!rm->ctl_blks[j]) + continue; + if (reserved_by_other(global_state->ctl_to_enc_id, j, enc_id)) continue; + ctl = to_dpu_hw_ctl(rm->ctl_blks[j]); + features = ctl->caps->features; has_split_display = BIT(DPU_CTL_SPLIT_DISPLAY) & features; - DPU_DEBUG("ctl %d caps 0x%lX\n", iter.blk->id, features); + DPU_DEBUG("ctl %d caps 0x%lX\n", rm->ctl_blks[j]->id, features); if (needs_split_display != has_split_display) continue; - ctls[i] = iter.blk; - DPU_DEBUG("ctl %d match\n", iter.blk->id); + ctl_idx[i] = j; + DPU_DEBUG("ctl %d match\n", j + CTL_0); if (++i == num_ctls) break; + } if (i != num_ctls) return -ENAVAIL; - for (i = 0; i < ARRAY_SIZE(ctls) && i < num_ctls; i++) { - ctls[i]->enc_id = enc_id; - trace_dpu_rm_reserve_ctls(ctls[i]->id, enc_id); + for (i = 0; i < ARRAY_SIZE(ctl_idx) && i < num_ctls; i++) { + global_state->ctl_to_enc_id[ctl_idx[i]] = enc_id; + trace_dpu_rm_reserve_ctls(i + CTL_0, enc_id); } return 0; @@ -479,40 +379,34 @@ static int _dpu_rm_reserve_ctls( static int _dpu_rm_reserve_intf( struct dpu_rm *rm, + struct dpu_global_state *global_state, uint32_t enc_id, - uint32_t id, - enum dpu_hw_blk_type type) + uint32_t id) { - struct dpu_rm_hw_iter iter; - int ret = 0; - - /* Find the block entry in the rm, and note the reservation */ - dpu_rm_init_hw_iter(&iter, 0, type); - while (_dpu_rm_get_hw_locked(rm, &iter)) { - if (iter.blk->id != id) - continue; + int idx = id - INTF_0; - if (RESERVED_BY_OTHER(iter.blk, enc_id)) { - DPU_ERROR("type %d id %d already reserved\n", type, id); - return -ENAVAIL; - } - - iter.blk->enc_id = enc_id; - trace_dpu_rm_reserve_intf(iter.blk->id, enc_id); - break; + if (idx < 0 || idx >= ARRAY_SIZE(rm->intf_blks)) { + DPU_ERROR("invalid intf id: %d", id); + return -EINVAL; } - /* Shouldn't happen since intfs are fixed at probe */ - if (!iter.hw) { - DPU_ERROR("couldn't find type %d id %d\n", type, id); + if (!rm->intf_blks[idx]) { + DPU_ERROR("couldn't find intf id %d\n", id); return -EINVAL; } - return ret; + if (reserved_by_other(global_state->intf_to_enc_id, idx, enc_id)) { + DPU_ERROR("intf id %d already reserved\n", id); + return -ENAVAIL; + } + + global_state->intf_to_enc_id[idx] = enc_id; + return 0; } static int _dpu_rm_reserve_intf_related_hw( struct dpu_rm *rm, + struct dpu_global_state *global_state, uint32_t enc_id, struct dpu_encoder_hw_resources *hw_res) { @@ -523,8 +417,7 @@ static int _dpu_rm_reserve_intf_related_hw( if (hw_res->intfs[i] == INTF_MODE_NONE) continue; id = i + INTF_0; - ret = _dpu_rm_reserve_intf(rm, enc_id, id, - DPU_HW_BLK_INTF); + ret = _dpu_rm_reserve_intf(rm, global_state, enc_id, id); if (ret) return ret; } @@ -534,25 +427,27 @@ static int _dpu_rm_reserve_intf_related_hw( static int _dpu_rm_make_reservation( struct dpu_rm *rm, + struct dpu_global_state *global_state, struct drm_encoder *enc, - struct drm_crtc_state *crtc_state, struct dpu_rm_requirements *reqs) { int ret; - ret = _dpu_rm_reserve_lms(rm, enc->base.id, reqs); + ret = _dpu_rm_reserve_lms(rm, global_state, enc->base.id, reqs); if (ret) { DPU_ERROR("unable to find appropriate mixers\n"); return ret; } - ret = _dpu_rm_reserve_ctls(rm, enc->base.id, &reqs->topology); + ret = _dpu_rm_reserve_ctls(rm, global_state, enc->base.id, + &reqs->topology); if (ret) { DPU_ERROR("unable to find appropriate CTL\n"); return ret; } - ret = _dpu_rm_reserve_intf_related_hw(rm, enc->base.id, &reqs->hw_res); + ret = _dpu_rm_reserve_intf_related_hw(rm, global_state, enc->base.id, + &reqs->hw_res); if (ret) return ret; @@ -560,9 +455,7 @@ static int _dpu_rm_make_reservation( } static int _dpu_rm_populate_requirements( - struct dpu_rm *rm, struct drm_encoder *enc, - struct drm_crtc_state *crtc_state, struct dpu_rm_requirements *reqs, struct msm_display_topology req_topology) { @@ -577,37 +470,36 @@ static int _dpu_rm_populate_requirements( return 0; } -static void _dpu_rm_release_reservation(struct dpu_rm *rm, uint32_t enc_id) +static void _dpu_rm_clear_mapping(uint32_t *res_mapping, int cnt, + uint32_t enc_id) { - struct dpu_rm_hw_blk *blk; - enum dpu_hw_blk_type type; - - for (type = 0; type < DPU_HW_BLK_MAX; type++) { - list_for_each_entry(blk, &rm->hw_blks[type], list) { - if (blk->enc_id == enc_id) { - blk->enc_id = 0; - DPU_DEBUG("rel enc %d %d %d\n", enc_id, - type, blk->id); - } - } + int i; + + for (i = 0; i < cnt; i++) { + if (res_mapping[i] == enc_id) + res_mapping[i] = 0; } } -void dpu_rm_release(struct dpu_rm *rm, struct drm_encoder *enc) +void dpu_rm_release(struct dpu_global_state *global_state, + struct drm_encoder *enc) { - mutex_lock(&rm->rm_lock); - - _dpu_rm_release_reservation(rm, enc->base.id); - - mutex_unlock(&rm->rm_lock); + _dpu_rm_clear_mapping(global_state->pingpong_to_enc_id, + ARRAY_SIZE(global_state->pingpong_to_enc_id), enc->base.id); + _dpu_rm_clear_mapping(global_state->mixer_to_enc_id, + ARRAY_SIZE(global_state->mixer_to_enc_id), enc->base.id); + _dpu_rm_clear_mapping(global_state->ctl_to_enc_id, + ARRAY_SIZE(global_state->ctl_to_enc_id), enc->base.id); + _dpu_rm_clear_mapping(global_state->intf_to_enc_id, + ARRAY_SIZE(global_state->intf_to_enc_id), enc->base.id); } int dpu_rm_reserve( struct dpu_rm *rm, + struct dpu_global_state *global_state, struct drm_encoder *enc, struct drm_crtc_state *crtc_state, - struct msm_display_topology topology, - bool test_only) + struct msm_display_topology topology) { struct dpu_rm_requirements reqs; int ret; @@ -616,31 +508,75 @@ int dpu_rm_reserve( if (!drm_atomic_crtc_needs_modeset(crtc_state)) return 0; - DRM_DEBUG_KMS("reserving hw for enc %d crtc %d test_only %d\n", - enc->base.id, crtc_state->crtc->base.id, test_only); + if (IS_ERR(global_state)) { + DPU_ERROR("failed to global state\n"); + return PTR_ERR(global_state); + } - mutex_lock(&rm->rm_lock); + DRM_DEBUG_KMS("reserving hw for enc %d crtc %d\n", + enc->base.id, crtc_state->crtc->base.id); - ret = _dpu_rm_populate_requirements(rm, enc, crtc_state, &reqs, - topology); + ret = _dpu_rm_populate_requirements(enc, &reqs, topology); if (ret) { DPU_ERROR("failed to populate hw requirements\n"); - goto end; + return ret; } - ret = _dpu_rm_make_reservation(rm, enc, crtc_state, &reqs); - if (ret) { + ret = _dpu_rm_make_reservation(rm, global_state, enc, &reqs); + if (ret) DPU_ERROR("failed to reserve hw resources: %d\n", ret); - _dpu_rm_release_reservation(rm, enc->base.id); - } else if (test_only) { - /* test_only: test the reservation and then undo */ - DPU_DEBUG("test_only: discard test [enc: %d]\n", - enc->base.id); - _dpu_rm_release_reservation(rm, enc->base.id); - } -end: - mutex_unlock(&rm->rm_lock); + return ret; } + +int dpu_rm_get_assigned_resources(struct dpu_rm *rm, + struct dpu_global_state *global_state, uint32_t enc_id, + enum dpu_hw_blk_type type, struct dpu_hw_blk **blks, int blks_size) +{ + struct dpu_hw_blk **hw_blks; + uint32_t *hw_to_enc_id; + int i, num_blks, max_blks; + + switch (type) { + case DPU_HW_BLK_PINGPONG: + hw_blks = rm->pingpong_blks; + hw_to_enc_id = global_state->pingpong_to_enc_id; + max_blks = ARRAY_SIZE(rm->pingpong_blks); + break; + case DPU_HW_BLK_LM: + hw_blks = rm->mixer_blks; + hw_to_enc_id = global_state->mixer_to_enc_id; + max_blks = ARRAY_SIZE(rm->mixer_blks); + break; + case DPU_HW_BLK_CTL: + hw_blks = rm->ctl_blks; + hw_to_enc_id = global_state->ctl_to_enc_id; + max_blks = ARRAY_SIZE(rm->ctl_blks); + break; + case DPU_HW_BLK_INTF: + hw_blks = rm->intf_blks; + hw_to_enc_id = global_state->intf_to_enc_id; + max_blks = ARRAY_SIZE(rm->intf_blks); + break; + default: + DPU_ERROR("blk type %d not managed by rm\n", type); + return 0; + } + + num_blks = 0; + for (i = 0; i < max_blks; i++) { + if (hw_to_enc_id[i] != enc_id) + continue; + + if (num_blks == blks_size) { + DPU_ERROR("More than %d resources assigned to enc %d\n", + blks_size, enc_id); + break; + } + blks[num_blks++] = hw_blks[i]; + } + + return num_blks; +} diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.h index 9c580a017094..6d2b04f306f0 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.h @@ -11,37 +11,24 @@ #include "msm_kms.h" #include "dpu_hw_top.h" +struct dpu_global_state; + /** * struct dpu_rm - DPU dynamic hardware resource manager - * @hw_blks: array of lists of hardware resources present in the system, one - * list per type of hardware block + * @pingpong_blks: array of pingpong hardware resources + * @mixer_blks: array of layer mixer hardware resources + * @ctl_blks: array of ctl hardware resources + * @intf_blks: array of intf hardware resources * @lm_max_width: cached layer mixer maximum width * @rm_lock: resource manager mutex */ struct dpu_rm { - struct list_head hw_blks[DPU_HW_BLK_MAX]; - uint32_t lm_max_width; - struct mutex rm_lock; -}; + struct dpu_hw_blk *pingpong_blks[PINGPONG_MAX - PINGPONG_0]; + struct dpu_hw_blk *mixer_blks[LM_MAX - LM_0]; + struct dpu_hw_blk *ctl_blks[CTL_MAX - CTL_0]; + struct dpu_hw_blk *intf_blks[INTF_MAX - INTF_0]; -/** - * struct dpu_rm_hw_blk - resource manager internal structure - * forward declaration for single iterator definition without void pointer - */ -struct dpu_rm_hw_blk; - -/** - * struct dpu_rm_hw_iter - iterator for use with dpu_rm - * @hw: dpu_hw object requested, or NULL on failure - * @blk: dpu_rm internal block representation. Clients ignore. Used as iterator. - * @enc_id: DRM ID of Encoder client wishes to search for, or 0 for Any Encoder - * @type: Hardware Block Type client wishes to search for. - */ -struct dpu_rm_hw_iter { - void *hw; - struct dpu_rm_hw_blk *blk; - uint32_t enc_id; - enum dpu_hw_blk_type type; + uint32_t lm_max_width; }; /** @@ -74,14 +61,13 @@ int dpu_rm_destroy(struct dpu_rm *rm); * @drm_enc: DRM Encoder handle * @crtc_state: Proposed Atomic DRM CRTC State handle * @topology: Pointer to topology info for the display - * @test_only: Atomic-Test phase, discard results (unless property overrides) * @Return: 0 on Success otherwise -ERROR */ int dpu_rm_reserve(struct dpu_rm *rm, + struct dpu_global_state *global_state, struct drm_encoder *drm_enc, struct drm_crtc_state *crtc_state, - struct msm_display_topology topology, - bool test_only); + struct msm_display_topology topology); /** * dpu_rm_reserve - Given the encoder for the display chain, release any @@ -90,31 +76,14 @@ int dpu_rm_reserve(struct dpu_rm *rm, * @enc: DRM Encoder handle * @Return: 0 on Success otherwise -ERROR */ -void dpu_rm_release(struct dpu_rm *rm, struct drm_encoder *enc); +void dpu_rm_release(struct dpu_global_state *global_state, + struct drm_encoder *enc); /** - * dpu_rm_init_hw_iter - setup given iterator for new iteration over hw list - * using dpu_rm_get_hw - * @iter: iter object to initialize - * @enc_id: DRM ID of Encoder client wishes to search for, or 0 for Any Encoder - * @type: Hardware Block Type client wishes to search for. - */ -void dpu_rm_init_hw_iter( - struct dpu_rm_hw_iter *iter, - uint32_t enc_id, - enum dpu_hw_blk_type type); -/** - * dpu_rm_get_hw - retrieve reserved hw object given encoder and hw type - * Meant to do a single pass through the hardware list to iteratively - * retrieve hardware blocks of a given type for a given encoder. - * Initialize an iterator object. - * Set hw block type of interest. Set encoder id of interest, 0 for any. - * Function returns first hw of type for that encoder. - * Subsequent calls will return the next reserved hw of that type in-order. - * Iterator HW pointer will be null on failure to find hw. - * @rm: DPU Resource Manager handle - * @iter: iterator object - * @Return: true on match found, false on no match found + * Get hw resources of the given type that are assigned to this encoder. */ -bool dpu_rm_get_hw(struct dpu_rm *rm, struct dpu_rm_hw_iter *iter); +int dpu_rm_get_assigned_resources(struct dpu_rm *rm, + struct dpu_global_state *global_state, uint32_t enc_id, + enum dpu_hw_blk_type type, struct dpu_hw_blk **blks, int blks_size); #endif /* __DPU_RM_H__ */ + diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_vbif.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_vbif.c index 93ab36bd8df3..5e8c3f3e6625 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_vbif.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_vbif.c @@ -24,7 +24,7 @@ static int _dpu_vbif_wait_for_xin_halt(struct dpu_hw_vbif *vbif, u32 xin_id) int rc; if (!vbif || !vbif->cap || !vbif->ops.get_halt_ctrl) { - DPU_ERROR("invalid arguments vbif %d\n", vbif != 0); + DPU_ERROR("invalid arguments vbif %d\n", vbif != NULL); return -EINVAL; } @@ -106,7 +106,7 @@ static u32 _dpu_vbif_get_ot_limit(struct dpu_hw_vbif *vbif, u32 val; if (!vbif || !vbif->cap) { - DPU_ERROR("invalid arguments vbif %d\n", vbif != 0); + DPU_ERROR("invalid arguments vbif %d\n", vbif != NULL); return -EINVAL; } @@ -164,7 +164,7 @@ void dpu_vbif_set_ot_limit(struct dpu_kms *dpu_kms, if (!vbif || !mdp) { DPU_DEBUG("invalid arguments vbif %d mdp %d\n", - vbif != 0, mdp != 0); + vbif != NULL, mdp != NULL); return; } diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_crtc.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_crtc.c index b3d0a0fe76b9..998bef1190a3 100644 --- a/drivers/gpu/drm/msm/disp/mdp5/mdp5_crtc.c +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_crtc.c @@ -1191,8 +1191,8 @@ static void mdp5_crtc_wait_for_pp_done(struct drm_crtc *crtc) ret = wait_for_completion_timeout(&mdp5_crtc->pp_completion, msecs_to_jiffies(50)); if (ret == 0) - dev_warn(dev->dev, "pp done time out, lm=%d\n", - mdp5_cstate->pipeline.mixer->lm); + dev_warn_ratelimited(dev->dev, "pp done time out, lm=%d\n", + mdp5_cstate->pipeline.mixer->lm); } static void mdp5_crtc_wait_for_flush_done(struct drm_crtc *crtc) diff --git a/drivers/gpu/drm/msm/dsi/dsi_manager.c b/drivers/gpu/drm/msm/dsi/dsi_manager.c index 6af26ab5b09d..4b363bd7ddff 100644 --- a/drivers/gpu/drm/msm/dsi/dsi_manager.c +++ b/drivers/gpu/drm/msm/dsi/dsi_manager.c @@ -336,7 +336,7 @@ static int dsi_mgr_connector_get_modes(struct drm_connector *connector) return num; } -static int dsi_mgr_connector_mode_valid(struct drm_connector *connector, +static enum drm_mode_status dsi_mgr_connector_mode_valid(struct drm_connector *connector, struct drm_display_mode *mode) { int id = dsi_mgr_connector_get_id(connector); @@ -506,6 +506,7 @@ static void dsi_mgr_bridge_post_disable(struct drm_bridge *bridge) struct msm_dsi *msm_dsi1 = dsi_mgr_get_dsi(DSI_1); struct mipi_dsi_host *host = msm_dsi->host; struct drm_panel *panel = msm_dsi->panel; + struct msm_dsi_pll *src_pll; bool is_dual_dsi = IS_DUAL_DSI(); int ret; @@ -539,6 +540,10 @@ static void dsi_mgr_bridge_post_disable(struct drm_bridge *bridge) id, ret); } + /* Save PLL status if it is a clock source */ + src_pll = msm_dsi_phy_get_pll(msm_dsi->phy); + msm_dsi_pll_save_state(src_pll); + ret = msm_dsi_host_power_off(host); if (ret) pr_err("%s: host %d power off failed,%d\n", __func__, id, ret); diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c index b0cfa67d2a57..f509ebd77500 100644 --- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c +++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c @@ -724,10 +724,6 @@ void msm_dsi_phy_disable(struct msm_dsi_phy *phy) if (!phy || !phy->cfg->ops.disable) return; - /* Save PLL status if it is a clock source */ - if (phy->usecase != MSM_DSI_PHY_SLAVE) - msm_dsi_pll_save_state(phy->pll); - phy->cfg->ops.disable(phy); dsi_phy_regulator_disable(phy); diff --git a/drivers/gpu/drm/msm/dsi/pll/dsi_pll_10nm.c b/drivers/gpu/drm/msm/dsi/pll/dsi_pll_10nm.c index 1c894548dd72..6ac04fc303f5 100644 --- a/drivers/gpu/drm/msm/dsi/pll/dsi_pll_10nm.c +++ b/drivers/gpu/drm/msm/dsi/pll/dsi_pll_10nm.c @@ -411,6 +411,12 @@ static int dsi_pll_10nm_vco_prepare(struct clk_hw *hw) if (pll_10nm->slave) dsi_pll_enable_pll_bias(pll_10nm->slave); + rc = dsi_pll_10nm_vco_set_rate(hw,pll_10nm->vco_current_rate, 0); + if (rc) { + pr_err("vco_set_rate failed, rc=%d\n", rc); + return rc; + } + /* Start PLL */ pll_write(pll_10nm->phy_cmn_mmio + REG_DSI_10nm_PHY_CMN_PLL_CNTRL, 0x01); diff --git a/drivers/gpu/drm/msm/edp/edp.c b/drivers/gpu/drm/msm/edp/edp.c index a78d6077802b..106a67473af5 100644 --- a/drivers/gpu/drm/msm/edp/edp.c +++ b/drivers/gpu/drm/msm/edp/edp.c @@ -178,10 +178,6 @@ int msm_edp_modeset_init(struct msm_edp *edp, struct drm_device *dev, goto fail; } - ret = drm_bridge_attach(encoder, edp->bridge, NULL, 0); - if (ret) - goto fail; - priv->bridges[priv->num_bridges++] = edp->bridge; priv->connectors[priv->num_connectors++] = edp->connector; diff --git a/drivers/gpu/drm/msm/hdmi/hdmi.c b/drivers/gpu/drm/msm/hdmi/hdmi.c index 3a8646535c14..737453b6e596 100644 --- a/drivers/gpu/drm/msm/hdmi/hdmi.c +++ b/drivers/gpu/drm/msm/hdmi/hdmi.c @@ -327,10 +327,6 @@ int msm_hdmi_modeset_init(struct hdmi *hdmi, goto fail; } - ret = drm_bridge_attach(encoder, hdmi->bridge, NULL, 0); - if (ret) - goto fail; - priv->bridges[priv->num_bridges++] = hdmi->bridge; priv->connectors[priv->num_connectors++] = hdmi->connector; diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c index 2a82c23a6e4d..29295dee2a2e 100644 --- a/drivers/gpu/drm/msm/msm_drv.c +++ b/drivers/gpu/drm/msm/msm_drv.c @@ -444,8 +444,10 @@ static int msm_drm_init(struct device *dev, struct drm_driver *drv) if (!dev->dma_parms) { dev->dma_parms = devm_kzalloc(dev, sizeof(*dev->dma_parms), GFP_KERNEL); - if (!dev->dma_parms) - return -ENOMEM; + if (!dev->dma_parms) { + ret = -ENOMEM; + goto err_msm_uninit; + } } dma_set_max_seg_size(dev, DMA_BIT_MASK(32)); diff --git a/drivers/gpu/drm/msm/msm_gem.h b/drivers/gpu/drm/msm/msm_gem.h index 9e0953c2b7ce..30584eaf8cc8 100644 --- a/drivers/gpu/drm/msm/msm_gem.h +++ b/drivers/gpu/drm/msm/msm_gem.h @@ -157,7 +157,17 @@ struct msm_gem_submit { uint32_t handle; }; uint64_t iova; - } bos[0]; + } bos[]; }; +/* helper to determine of a buffer in submit should be dumped, used for both + * devcoredump and debugfs cmdstream dumping: + */ +static inline bool +should_dump(struct msm_gem_submit *submit, int idx) +{ + extern bool rd_full; + return rd_full || (submit->bos[idx].flags & MSM_SUBMIT_BO_DUMP); +} + #endif /* __MSM_GEM_H__ */ diff --git a/drivers/gpu/drm/msm/msm_gpu.c b/drivers/gpu/drm/msm/msm_gpu.c index 18f3a5c53ffb..615c5cda5389 100644 --- a/drivers/gpu/drm/msm/msm_gpu.c +++ b/drivers/gpu/drm/msm/msm_gpu.c @@ -355,16 +355,34 @@ static void msm_gpu_crashstate_capture(struct msm_gpu *gpu, state->cmd = kstrdup(cmd, GFP_KERNEL); if (submit) { - int i; - - state->bos = kcalloc(submit->nr_cmds, + int i, nr = 0; + + /* count # of buffers to dump: */ + for (i = 0; i < submit->nr_bos; i++) + if (should_dump(submit, i)) + nr++; + /* always dump cmd bo's, but don't double count them: */ + for (i = 0; i < submit->nr_cmds; i++) + if (!should_dump(submit, submit->cmd[i].idx)) + nr++; + + state->bos = kcalloc(nr, sizeof(struct msm_gpu_state_bo), GFP_KERNEL); + for (i = 0; i < submit->nr_bos; i++) { + if (should_dump(submit, i)) { + msm_gpu_crashstate_get_bo(state, submit->bos[i].obj, + submit->bos[i].iova, submit->bos[i].flags); + } + } + for (i = 0; state->bos && i < submit->nr_cmds; i++) { int idx = submit->cmd[i].idx; - msm_gpu_crashstate_get_bo(state, submit->bos[idx].obj, - submit->bos[idx].iova, submit->bos[idx].flags); + if (!should_dump(submit, submit->cmd[i].idx)) { + msm_gpu_crashstate_get_bo(state, submit->bos[idx].obj, + submit->bos[idx].iova, submit->bos[idx].flags); + } } } diff --git a/drivers/gpu/drm/msm/msm_rd.c b/drivers/gpu/drm/msm/msm_rd.c index af7ceb246c7c..732f65df5c4f 100644 --- a/drivers/gpu/drm/msm/msm_rd.c +++ b/drivers/gpu/drm/msm/msm_rd.c @@ -43,7 +43,7 @@ #include "msm_gpu.h" #include "msm_gem.h" -static bool rd_full = false; +bool rd_full = false; MODULE_PARM_DESC(rd_full, "If true, $debugfs/.../rd will snapshot all buffer contents"); module_param_named(rd_full, rd_full, bool, 0600); @@ -336,12 +336,6 @@ static void snapshot_buf(struct msm_rd_state *rd, msm_gem_put_vaddr(&obj->base); } -static bool -should_dump(struct msm_gem_submit *submit, int idx) -{ - return rd_full || (submit->bos[idx].flags & MSM_SUBMIT_BO_DUMP); -} - /* called under struct_mutex */ void msm_rd_dump_submit(struct msm_rd_state *rd, struct msm_gem_submit *submit, const char *fmt, ...) diff --git a/drivers/gpu/drm/nouveau/dispnv50/wndw.c b/drivers/gpu/drm/nouveau/dispnv50/wndw.c index 890315291b01..bb737f9281e6 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/wndw.c +++ b/drivers/gpu/drm/nouveau/dispnv50/wndw.c @@ -458,6 +458,8 @@ nv50_wndw_atomic_check(struct drm_plane *plane, struct drm_plane_state *state) asyw->clr.ntfy = armw->ntfy.handle != 0; asyw->clr.sema = armw->sema.handle != 0; asyw->clr.xlut = armw->xlut.handle != 0; + if (asyw->clr.xlut && asyw->visible) + asyw->set.xlut = asyw->xlut.handle != 0; asyw->clr.csc = armw->csc.valid; if (wndw->func->image_clr) asyw->clr.image = armw->image.handle[0] != 0; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c index c7d700916eae..8ebbe1656008 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c @@ -2579,6 +2579,7 @@ nv166_chipset = { static const struct nvkm_device_chip nv167_chipset = { .name = "TU117", + .acr = tu102_acr_new, .bar = tu102_bar_new, .bios = nvkm_bios_new, .bus = gf100_bus_new, @@ -2607,6 +2608,7 @@ nv167_chipset = { .disp = tu102_disp_new, .dma = gv100_dma_new, .fifo = tu102_fifo_new, + .gr = tu102_gr_new, .nvdec[0] = gm107_nvdec_new, .nvenc[0] = gm107_nvenc_new, .sec2 = tu102_sec2_new, @@ -2615,6 +2617,7 @@ nv167_chipset = { static const struct nvkm_device_chip nv168_chipset = { .name = "TU116", + .acr = tu102_acr_new, .bar = tu102_bar_new, .bios = nvkm_bios_new, .bus = gf100_bus_new, @@ -2643,6 +2646,7 @@ nv168_chipset = { .disp = tu102_disp_new, .dma = gv100_dma_new, .fifo = tu102_fifo_new, + .gr = tu102_gr_new, .nvdec[0] = gm107_nvdec_new, .nvenc[0] = gm107_nvenc_new, .sec2 = tu102_sec2_new, diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/tu102.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/tu102.c index 454668b1cf54..a9efa4d78be9 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/tu102.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/tu102.c @@ -164,6 +164,32 @@ MODULE_FIRMWARE("nvidia/tu106/gr/sw_nonctx.bin"); MODULE_FIRMWARE("nvidia/tu106/gr/sw_bundle_init.bin"); MODULE_FIRMWARE("nvidia/tu106/gr/sw_method_init.bin"); +MODULE_FIRMWARE("nvidia/tu117/gr/fecs_bl.bin"); +MODULE_FIRMWARE("nvidia/tu117/gr/fecs_inst.bin"); +MODULE_FIRMWARE("nvidia/tu117/gr/fecs_data.bin"); +MODULE_FIRMWARE("nvidia/tu117/gr/fecs_sig.bin"); +MODULE_FIRMWARE("nvidia/tu117/gr/gpccs_bl.bin"); +MODULE_FIRMWARE("nvidia/tu117/gr/gpccs_inst.bin"); +MODULE_FIRMWARE("nvidia/tu117/gr/gpccs_data.bin"); +MODULE_FIRMWARE("nvidia/tu117/gr/gpccs_sig.bin"); +MODULE_FIRMWARE("nvidia/tu117/gr/sw_ctx.bin"); +MODULE_FIRMWARE("nvidia/tu117/gr/sw_nonctx.bin"); +MODULE_FIRMWARE("nvidia/tu117/gr/sw_bundle_init.bin"); +MODULE_FIRMWARE("nvidia/tu117/gr/sw_method_init.bin"); + +MODULE_FIRMWARE("nvidia/tu116/gr/fecs_bl.bin"); +MODULE_FIRMWARE("nvidia/tu116/gr/fecs_inst.bin"); +MODULE_FIRMWARE("nvidia/tu116/gr/fecs_data.bin"); +MODULE_FIRMWARE("nvidia/tu116/gr/fecs_sig.bin"); +MODULE_FIRMWARE("nvidia/tu116/gr/gpccs_bl.bin"); +MODULE_FIRMWARE("nvidia/tu116/gr/gpccs_inst.bin"); +MODULE_FIRMWARE("nvidia/tu116/gr/gpccs_data.bin"); +MODULE_FIRMWARE("nvidia/tu116/gr/gpccs_sig.bin"); +MODULE_FIRMWARE("nvidia/tu116/gr/sw_ctx.bin"); +MODULE_FIRMWARE("nvidia/tu116/gr/sw_nonctx.bin"); +MODULE_FIRMWARE("nvidia/tu116/gr/sw_bundle_init.bin"); +MODULE_FIRMWARE("nvidia/tu116/gr/sw_method_init.bin"); + static const struct gf100_gr_fwif tu102_gr_fwif[] = { { 0, gm200_gr_load, &tu102_gr, &gp108_gr_fecs_acr, &gp108_gr_gpccs_acr }, diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/acr/tu102.c b/drivers/gpu/drm/nouveau/nvkm/subdev/acr/tu102.c index 7f4b89d82d32..d28d8f36ae24 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/acr/tu102.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/acr/tu102.c @@ -107,6 +107,12 @@ MODULE_FIRMWARE("nvidia/tu104/acr/ucode_unload.bin"); MODULE_FIRMWARE("nvidia/tu106/acr/unload_bl.bin"); MODULE_FIRMWARE("nvidia/tu106/acr/ucode_unload.bin"); +MODULE_FIRMWARE("nvidia/tu116/acr/unload_bl.bin"); +MODULE_FIRMWARE("nvidia/tu116/acr/ucode_unload.bin"); + +MODULE_FIRMWARE("nvidia/tu117/acr/unload_bl.bin"); +MODULE_FIRMWARE("nvidia/tu117/acr/ucode_unload.bin"); + static const struct nvkm_acr_hsf_fwif tu102_acr_unload_fwif[] = { { 0, nvkm_acr_hsfw_load, &gp108_acr_unload_0 }, @@ -130,6 +136,8 @@ tu102_acr_asb_0 = { MODULE_FIRMWARE("nvidia/tu102/acr/ucode_asb.bin"); MODULE_FIRMWARE("nvidia/tu104/acr/ucode_asb.bin"); MODULE_FIRMWARE("nvidia/tu106/acr/ucode_asb.bin"); +MODULE_FIRMWARE("nvidia/tu116/acr/ucode_asb.bin"); +MODULE_FIRMWARE("nvidia/tu117/acr/ucode_asb.bin"); static const struct nvkm_acr_hsf_fwif tu102_acr_asb_fwif[] = { @@ -154,6 +162,12 @@ MODULE_FIRMWARE("nvidia/tu104/acr/ucode_ahesasc.bin"); MODULE_FIRMWARE("nvidia/tu106/acr/bl.bin"); MODULE_FIRMWARE("nvidia/tu106/acr/ucode_ahesasc.bin"); +MODULE_FIRMWARE("nvidia/tu116/acr/bl.bin"); +MODULE_FIRMWARE("nvidia/tu116/acr/ucode_ahesasc.bin"); + +MODULE_FIRMWARE("nvidia/tu117/acr/bl.bin"); +MODULE_FIRMWARE("nvidia/tu117/acr/ucode_ahesasc.bin"); + static const struct nvkm_acr_hsf_fwif tu102_acr_ahesasc_fwif[] = { { 0, nvkm_acr_hsfw_load, &tu102_acr_ahesasc_0 }, diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gv100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gv100.c index 389bad312bf2..10ff5d053f7e 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gv100.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gv100.c @@ -51,3 +51,5 @@ MODULE_FIRMWARE("nvidia/gv100/nvdec/scrubber.bin"); MODULE_FIRMWARE("nvidia/tu102/nvdec/scrubber.bin"); MODULE_FIRMWARE("nvidia/tu104/nvdec/scrubber.bin"); MODULE_FIRMWARE("nvidia/tu106/nvdec/scrubber.bin"); +MODULE_FIRMWARE("nvidia/tu116/nvdec/scrubber.bin"); +MODULE_FIRMWARE("nvidia/tu117/nvdec/scrubber.bin"); diff --git a/drivers/gpu/drm/panfrost/panfrost_job.c b/drivers/gpu/drm/panfrost/panfrost_job.c index 8fd8726313ae..7914b1570841 100644 --- a/drivers/gpu/drm/panfrost/panfrost_job.c +++ b/drivers/gpu/drm/panfrost/panfrost_job.c @@ -280,12 +280,8 @@ static void panfrost_job_cleanup(struct kref *ref) } if (job->bos) { - struct panfrost_gem_object *bo; - - for (i = 0; i < job->bo_count; i++) { - bo = to_panfrost_bo(job->bos[i]); + for (i = 0; i < job->bo_count; i++) drm_gem_object_put_unlocked(job->bos[i]); - } kvfree(job->bos); } diff --git a/drivers/gpu/drm/panfrost/panfrost_mmu.c b/drivers/gpu/drm/panfrost/panfrost_mmu.c index 23314f41717b..ed28aeba6d59 100644 --- a/drivers/gpu/drm/panfrost/panfrost_mmu.c +++ b/drivers/gpu/drm/panfrost/panfrost_mmu.c @@ -151,7 +151,12 @@ u32 panfrost_mmu_as_get(struct panfrost_device *pfdev, struct panfrost_mmu *mmu) as = mmu->as; if (as >= 0) { int en = atomic_inc_return(&mmu->as_count); - WARN_ON(en >= NUM_JOB_SLOTS); + + /* + * AS can be retained by active jobs or a perfcnt context, + * hence the '+ 1' here. + */ + WARN_ON(en >= (NUM_JOB_SLOTS + 1)); list_move(&mmu->list, &pfdev->as_lru_list); goto out; @@ -596,33 +601,27 @@ static irqreturn_t panfrost_mmu_irq_handler_thread(int irq, void *data) source_id = (fault_status >> 16); /* Page fault only */ - if ((status & mask) == BIT(i)) { - WARN_ON(exception_type < 0xC1 || exception_type > 0xC4); - + ret = -1; + if ((status & mask) == BIT(i) && (exception_type & 0xF8) == 0xC0) ret = panfrost_mmu_map_fault_addr(pfdev, i, addr); - if (!ret) { - mmu_write(pfdev, MMU_INT_CLEAR, BIT(i)); - status &= ~mask; - continue; - } - } - /* terminal fault, print info about the fault */ - dev_err(pfdev->dev, - "Unhandled Page fault in AS%d at VA 0x%016llX\n" - "Reason: %s\n" - "raw fault status: 0x%X\n" - "decoded fault status: %s\n" - "exception type 0x%X: %s\n" - "access type 0x%X: %s\n" - "source id 0x%X\n", - i, addr, - "TODO", - fault_status, - (fault_status & (1 << 10) ? "DECODER FAULT" : "SLAVE FAULT"), - exception_type, panfrost_exception_name(pfdev, exception_type), - access_type, access_type_name(pfdev, fault_status), - source_id); + if (ret) + /* terminal fault, print info about the fault */ + dev_err(pfdev->dev, + "Unhandled Page fault in AS%d at VA 0x%016llX\n" + "Reason: %s\n" + "raw fault status: 0x%X\n" + "decoded fault status: %s\n" + "exception type 0x%X: %s\n" + "access type 0x%X: %s\n" + "source id 0x%X\n", + i, addr, + "TODO", + fault_status, + (fault_status & (1 << 10) ? "DECODER FAULT" : "SLAVE FAULT"), + exception_type, panfrost_exception_name(pfdev, exception_type), + access_type, access_type_name(pfdev, fault_status), + source_id); mmu_write(pfdev, MMU_INT_CLEAR, mask); diff --git a/drivers/gpu/drm/panfrost/panfrost_perfcnt.c b/drivers/gpu/drm/panfrost/panfrost_perfcnt.c index 684820448be3..6913578d5aa7 100644 --- a/drivers/gpu/drm/panfrost/panfrost_perfcnt.c +++ b/drivers/gpu/drm/panfrost/panfrost_perfcnt.c @@ -73,7 +73,7 @@ static int panfrost_perfcnt_enable_locked(struct panfrost_device *pfdev, struct panfrost_file_priv *user = file_priv->driver_priv; struct panfrost_perfcnt *perfcnt = pfdev->perfcnt; struct drm_gem_shmem_object *bo; - u32 cfg; + u32 cfg, as; int ret; if (user == perfcnt->user) @@ -126,12 +126,8 @@ static int panfrost_perfcnt_enable_locked(struct panfrost_device *pfdev, perfcnt->user = user; - /* - * Always use address space 0 for now. - * FIXME: this needs to be updated when we start using different - * address space. - */ - cfg = GPU_PERFCNT_CFG_AS(0) | + as = panfrost_mmu_as_get(pfdev, perfcnt->mapping->mmu); + cfg = GPU_PERFCNT_CFG_AS(as) | GPU_PERFCNT_CFG_MODE(GPU_PERFCNT_CFG_MODE_MANUAL); /* @@ -195,6 +191,7 @@ static int panfrost_perfcnt_disable_locked(struct panfrost_device *pfdev, drm_gem_shmem_vunmap(&perfcnt->mapping->obj->base.base, perfcnt->buf); perfcnt->buf = NULL; panfrost_gem_close(&perfcnt->mapping->obj->base.base, file_priv); + panfrost_mmu_as_put(pfdev, perfcnt->mapping->mmu); panfrost_gem_mapping_put(perfcnt->mapping); perfcnt->mapping = NULL; pm_runtime_mark_last_busy(pfdev->dev); diff --git a/drivers/gpu/drm/radeon/radeon_ttm.c b/drivers/gpu/drm/radeon/radeon_ttm.c index badf1b6d1549..5d50c9edbe80 100644 --- a/drivers/gpu/drm/radeon/radeon_ttm.c +++ b/drivers/gpu/drm/radeon/radeon_ttm.c @@ -523,7 +523,7 @@ static int radeon_ttm_tt_pin_userptr(struct ttm_tt *ttm) r = -ENOMEM; nents = dma_map_sg(rdev->dev, ttm->sg->sgl, ttm->sg->nents, direction); - if (nents != ttm->sg->nents) + if (nents == 0) goto release_sg; drm_prime_sg_to_page_addr_arrays(ttm->sg, ttm->pages, diff --git a/drivers/gpu/drm/radeon/si_dpm.c b/drivers/gpu/drm/radeon/si_dpm.c index 05e8b4d0af3f..2cb85dbe728f 100644 --- a/drivers/gpu/drm/radeon/si_dpm.c +++ b/drivers/gpu/drm/radeon/si_dpm.c @@ -2979,7 +2979,6 @@ static void si_apply_state_adjust_rules(struct radeon_device *rdev, if (rdev->family == CHIP_HAINAN) { if ((rdev->pdev->revision == 0x81) || - (rdev->pdev->revision == 0x83) || (rdev->pdev->revision == 0xC3) || (rdev->pdev->device == 0x6664) || (rdev->pdev->device == 0x6665) || diff --git a/drivers/gpu/drm/scheduler/gpu_scheduler_trace.h b/drivers/gpu/drm/scheduler/gpu_scheduler_trace.h index d79086498aff..877ce9b127f1 100644 --- a/drivers/gpu/drm/scheduler/gpu_scheduler_trace.h +++ b/drivers/gpu/drm/scheduler/gpu_scheduler_trace.h @@ -59,6 +59,33 @@ TRACE_EVENT(drm_sched_job, __entry->job_count, __entry->hw_job_count) ); +TRACE_EVENT(drm_run_job, + TP_PROTO(struct drm_sched_job *sched_job, struct drm_sched_entity *entity), + TP_ARGS(sched_job, entity), + TP_STRUCT__entry( + __field(struct drm_sched_entity *, entity) + __field(struct dma_fence *, fence) + __field(const char *, name) + __field(uint64_t, id) + __field(u32, job_count) + __field(int, hw_job_count) + ), + + TP_fast_assign( + __entry->entity = entity; + __entry->id = sched_job->id; + __entry->fence = &sched_job->s_fence->finished; + __entry->name = sched_job->sched->name; + __entry->job_count = spsc_queue_count(&entity->job_queue); + __entry->hw_job_count = atomic_read( + &sched_job->sched->hw_rq_count); + ), + TP_printk("entity=%p, id=%llu, fence=%p, ring=%s, job count:%u, hw job count:%d", + __entry->entity, __entry->id, + __entry->fence, __entry->name, + __entry->job_count, __entry->hw_job_count) +); + TRACE_EVENT(drm_sched_process_job, TP_PROTO(struct drm_sched_fence *fence), TP_ARGS(fence), diff --git a/drivers/gpu/drm/scheduler/sched_entity.c b/drivers/gpu/drm/scheduler/sched_entity.c index 63bccd201b97..c803e14eed91 100644 --- a/drivers/gpu/drm/scheduler/sched_entity.c +++ b/drivers/gpu/drm/scheduler/sched_entity.c @@ -84,6 +84,24 @@ int drm_sched_entity_init(struct drm_sched_entity *entity, EXPORT_SYMBOL(drm_sched_entity_init); /** + * drm_sched_entity_modify_sched - Modify sched of an entity + * @entity: scheduler entity to init + * @sched_list: the list of new drm scheds which will replace + * existing entity->sched_list + * @num_sched_list: number of drm sched in sched_list + */ +void drm_sched_entity_modify_sched(struct drm_sched_entity *entity, + struct drm_gpu_scheduler **sched_list, + unsigned int num_sched_list) +{ + WARN_ON(!num_sched_list || !sched_list); + + entity->sched_list = sched_list; + entity->num_sched_list = num_sched_list; +} +EXPORT_SYMBOL(drm_sched_entity_modify_sched); + +/** * drm_sched_entity_is_idle - Check if entity is idle * * @entity: scheduler entity @@ -120,38 +138,6 @@ bool drm_sched_entity_is_ready(struct drm_sched_entity *entity) } /** - * drm_sched_entity_get_free_sched - Get the rq from rq_list with least load - * - * @entity: scheduler entity - * - * Return the pointer to the rq with least load. - */ -static struct drm_sched_rq * -drm_sched_entity_get_free_sched(struct drm_sched_entity *entity) -{ - struct drm_sched_rq *rq = NULL; - unsigned int min_score = UINT_MAX, num_score; - int i; - - for (i = 0; i < entity->num_sched_list; ++i) { - struct drm_gpu_scheduler *sched = entity->sched_list[i]; - - if (!entity->sched_list[i]->ready) { - DRM_WARN("sched%s is not ready, skipping", sched->name); - continue; - } - - num_score = atomic_read(&sched->score); - if (num_score < min_score) { - min_score = num_score; - rq = &entity->sched_list[i]->sched_rq[entity->priority]; - } - } - - return rq; -} - -/** * drm_sched_entity_flush - Flush a context entity * * @entity: scheduler entity @@ -461,6 +447,7 @@ struct drm_sched_job *drm_sched_entity_pop_job(struct drm_sched_entity *entity) void drm_sched_entity_select_rq(struct drm_sched_entity *entity) { struct dma_fence *fence; + struct drm_gpu_scheduler *sched; struct drm_sched_rq *rq; if (spsc_queue_count(&entity->job_queue) || entity->num_sched_list <= 1) @@ -471,7 +458,8 @@ void drm_sched_entity_select_rq(struct drm_sched_entity *entity) return; spin_lock(&entity->rq_lock); - rq = drm_sched_entity_get_free_sched(entity); + sched = drm_sched_pick_best(entity->sched_list, entity->num_sched_list); + rq = sched ? &sched->sched_rq[entity->priority] : NULL; if (rq != entity->rq) { drm_sched_rq_remove_entity(entity->rq, entity); entity->rq = rq; @@ -498,7 +486,7 @@ void drm_sched_entity_push_job(struct drm_sched_job *sched_job, bool first; trace_drm_sched_job(sched_job, entity); - atomic_inc(&entity->rq->sched->score); + atomic_inc(&entity->rq->sched->num_jobs); WRITE_ONCE(entity->last_user, current->group_leader); first = spsc_queue_push(&entity->job_queue, &sched_job->queue_node); diff --git a/drivers/gpu/drm/scheduler/sched_main.c b/drivers/gpu/drm/scheduler/sched_main.c index 71ce6215956f..8e731ed0d9d9 100644 --- a/drivers/gpu/drm/scheduler/sched_main.c +++ b/drivers/gpu/drm/scheduler/sched_main.c @@ -92,7 +92,6 @@ void drm_sched_rq_add_entity(struct drm_sched_rq *rq, if (!list_empty(&entity->list)) return; spin_lock(&rq->lock); - atomic_inc(&rq->sched->score); list_add_tail(&entity->list, &rq->entities); spin_unlock(&rq->lock); } @@ -111,7 +110,6 @@ void drm_sched_rq_remove_entity(struct drm_sched_rq *rq, if (list_empty(&entity->list)) return; spin_lock(&rq->lock); - atomic_dec(&rq->sched->score); list_del_init(&entity->list); if (rq->current_entity == entity) rq->current_entity = NULL; @@ -222,8 +220,7 @@ EXPORT_SYMBOL(drm_sched_fault); * * Suspend the delayed work timeout for the scheduler. This is done by * modifying the delayed work timeout to an arbitrary large value, - * MAX_SCHEDULE_TIMEOUT in this case. Note that this function can be - * called from an IRQ context. + * MAX_SCHEDULE_TIMEOUT in this case. * * Returns the timeout remaining * @@ -252,46 +249,41 @@ EXPORT_SYMBOL(drm_sched_suspend_timeout); * @sched: scheduler instance for which to resume the timeout * @remaining: remaining timeout * - * Resume the delayed work timeout for the scheduler. Note that - * this function can be called from an IRQ context. + * Resume the delayed work timeout for the scheduler. */ void drm_sched_resume_timeout(struct drm_gpu_scheduler *sched, unsigned long remaining) { - unsigned long flags; - - spin_lock_irqsave(&sched->job_list_lock, flags); + spin_lock(&sched->job_list_lock); if (list_empty(&sched->ring_mirror_list)) cancel_delayed_work(&sched->work_tdr); else mod_delayed_work(system_wq, &sched->work_tdr, remaining); - spin_unlock_irqrestore(&sched->job_list_lock, flags); + spin_unlock(&sched->job_list_lock); } EXPORT_SYMBOL(drm_sched_resume_timeout); static void drm_sched_job_begin(struct drm_sched_job *s_job) { struct drm_gpu_scheduler *sched = s_job->sched; - unsigned long flags; - spin_lock_irqsave(&sched->job_list_lock, flags); + spin_lock(&sched->job_list_lock); list_add_tail(&s_job->node, &sched->ring_mirror_list); drm_sched_start_timeout(sched); - spin_unlock_irqrestore(&sched->job_list_lock, flags); + spin_unlock(&sched->job_list_lock); } static void drm_sched_job_timedout(struct work_struct *work) { struct drm_gpu_scheduler *sched; struct drm_sched_job *job; - unsigned long flags; sched = container_of(work, struct drm_gpu_scheduler, work_tdr.work); /* Protects against concurrent deletion in drm_sched_get_cleanup_job */ - spin_lock_irqsave(&sched->job_list_lock, flags); + spin_lock(&sched->job_list_lock); job = list_first_entry_or_null(&sched->ring_mirror_list, struct drm_sched_job, node); @@ -302,7 +294,7 @@ static void drm_sched_job_timedout(struct work_struct *work) * is parked at which point it's safe. */ list_del_init(&job->node); - spin_unlock_irqrestore(&sched->job_list_lock, flags); + spin_unlock(&sched->job_list_lock); job->sched->ops->timedout_job(job); @@ -315,12 +307,12 @@ static void drm_sched_job_timedout(struct work_struct *work) sched->free_guilty = false; } } else { - spin_unlock_irqrestore(&sched->job_list_lock, flags); + spin_unlock(&sched->job_list_lock); } - spin_lock_irqsave(&sched->job_list_lock, flags); + spin_lock(&sched->job_list_lock); drm_sched_start_timeout(sched); - spin_unlock_irqrestore(&sched->job_list_lock, flags); + spin_unlock(&sched->job_list_lock); } /** @@ -383,7 +375,6 @@ EXPORT_SYMBOL(drm_sched_increase_karma); void drm_sched_stop(struct drm_gpu_scheduler *sched, struct drm_sched_job *bad) { struct drm_sched_job *s_job, *tmp; - unsigned long flags; kthread_park(sched->thread); @@ -417,9 +408,9 @@ void drm_sched_stop(struct drm_gpu_scheduler *sched, struct drm_sched_job *bad) * remove job from ring_mirror_list. * Locking here is for concurrent resume timeout */ - spin_lock_irqsave(&sched->job_list_lock, flags); + spin_lock(&sched->job_list_lock); list_del_init(&s_job->node); - spin_unlock_irqrestore(&sched->job_list_lock, flags); + spin_unlock(&sched->job_list_lock); /* * Wait for job's HW fence callback to finish using s_job @@ -462,7 +453,6 @@ EXPORT_SYMBOL(drm_sched_stop); void drm_sched_start(struct drm_gpu_scheduler *sched, bool full_recovery) { struct drm_sched_job *s_job, *tmp; - unsigned long flags; int r; /* @@ -491,9 +481,9 @@ void drm_sched_start(struct drm_gpu_scheduler *sched, bool full_recovery) } if (full_recovery) { - spin_lock_irqsave(&sched->job_list_lock, flags); + spin_lock(&sched->job_list_lock); drm_sched_start_timeout(sched); - spin_unlock_irqrestore(&sched->job_list_lock, flags); + spin_unlock(&sched->job_list_lock); } kthread_unpark(sched->thread); @@ -657,11 +647,13 @@ static void drm_sched_process_job(struct dma_fence *f, struct dma_fence_cb *cb) struct drm_gpu_scheduler *sched = s_fence->sched; atomic_dec(&sched->hw_rq_count); - atomic_dec(&sched->score); + atomic_dec(&sched->num_jobs); trace_drm_sched_process_job(s_fence); + dma_fence_get(&s_fence->finished); drm_sched_fence_finished(s_fence); + dma_fence_put(&s_fence->finished); wake_up_interruptible(&sched->wake_up_worker); } @@ -677,7 +669,6 @@ static struct drm_sched_job * drm_sched_get_cleanup_job(struct drm_gpu_scheduler *sched) { struct drm_sched_job *job; - unsigned long flags; /* * Don't destroy jobs while the timeout worker is running OR thread @@ -688,7 +679,7 @@ drm_sched_get_cleanup_job(struct drm_gpu_scheduler *sched) __kthread_should_park(sched->thread)) return NULL; - spin_lock_irqsave(&sched->job_list_lock, flags); + spin_lock(&sched->job_list_lock); job = list_first_entry_or_null(&sched->ring_mirror_list, struct drm_sched_job, node); @@ -702,12 +693,48 @@ drm_sched_get_cleanup_job(struct drm_gpu_scheduler *sched) drm_sched_start_timeout(sched); } - spin_unlock_irqrestore(&sched->job_list_lock, flags); + spin_unlock(&sched->job_list_lock); return job; } /** + * drm_sched_pick_best - Get a drm sched from a sched_list with the least load + * @sched_list: list of drm_gpu_schedulers + * @num_sched_list: number of drm_gpu_schedulers in the sched_list + * + * Returns pointer of the sched with the least load or NULL if none of the + * drm_gpu_schedulers are ready + */ +struct drm_gpu_scheduler * +drm_sched_pick_best(struct drm_gpu_scheduler **sched_list, + unsigned int num_sched_list) +{ + struct drm_gpu_scheduler *sched, *picked_sched = NULL; + int i; + unsigned int min_jobs = UINT_MAX, num_jobs; + + for (i = 0; i < num_sched_list; ++i) { + sched = sched_list[i]; + + if (!sched->ready) { + DRM_WARN("scheduler %s is not ready, skipping", + sched->name); + continue; + } + + num_jobs = atomic_read(&sched->num_jobs); + if (num_jobs < min_jobs) { + min_jobs = num_jobs; + picked_sched = sched; + } + } + + return picked_sched; +} +EXPORT_SYMBOL(drm_sched_pick_best); + +/** * drm_sched_blocked - check if the scheduler is blocked * * @sched: scheduler instance @@ -773,6 +800,7 @@ static int drm_sched_main(void *param) atomic_inc(&sched->hw_rq_count); drm_sched_job_begin(sched_job); + trace_drm_run_job(sched_job, entity); fence = sched->ops->run_job(sched_job); drm_sched_fence_scheduled(s_fence); @@ -832,7 +860,7 @@ int drm_sched_init(struct drm_gpu_scheduler *sched, spin_lock_init(&sched->job_list_lock); atomic_set(&sched->hw_rq_count, 0); INIT_DELAYED_WORK(&sched->work_tdr, drm_sched_job_timedout); - atomic_set(&sched->score, 0); + atomic_set(&sched->num_jobs, 0); atomic64_set(&sched->job_id_count, 0); /* Each scheduler will run on a seperate kernel thread */ diff --git a/drivers/gpu/drm/selftests/drm_cmdline_selftests.h b/drivers/gpu/drm/selftests/drm_cmdline_selftests.h index ceac7af9a172..29e367db6118 100644 --- a/drivers/gpu/drm/selftests/drm_cmdline_selftests.h +++ b/drivers/gpu/drm/selftests/drm_cmdline_selftests.h @@ -53,6 +53,7 @@ cmdline_test(drm_cmdline_test_rotate_0) cmdline_test(drm_cmdline_test_rotate_90) cmdline_test(drm_cmdline_test_rotate_180) cmdline_test(drm_cmdline_test_rotate_270) +cmdline_test(drm_cmdline_test_rotate_multiple) cmdline_test(drm_cmdline_test_rotate_invalid_val) cmdline_test(drm_cmdline_test_rotate_truncated) cmdline_test(drm_cmdline_test_hmirror) diff --git a/drivers/gpu/drm/selftests/test-drm_cmdline_parser.c b/drivers/gpu/drm/selftests/test-drm_cmdline_parser.c index 520f3e66a384..d96cd890def6 100644 --- a/drivers/gpu/drm/selftests/test-drm_cmdline_parser.c +++ b/drivers/gpu/drm/selftests/test-drm_cmdline_parser.c @@ -856,6 +856,17 @@ static int drm_cmdline_test_rotate_270(void *ignored) return 0; } +static int drm_cmdline_test_rotate_multiple(void *ignored) +{ + struct drm_cmdline_mode mode = { }; + + FAIL_ON(drm_mode_parse_command_line_for_connector("720x480,rotate=0,rotate=90", + &no_connector, + &mode)); + + return 0; +} + static int drm_cmdline_test_rotate_invalid_val(void *ignored) { struct drm_cmdline_mode mode = { }; @@ -888,7 +899,7 @@ static int drm_cmdline_test_hmirror(void *ignored) FAIL_ON(!mode.specified); FAIL_ON(mode.xres != 720); FAIL_ON(mode.yres != 480); - FAIL_ON(mode.rotation_reflection != DRM_MODE_REFLECT_X); + FAIL_ON(mode.rotation_reflection != (DRM_MODE_ROTATE_0 | DRM_MODE_REFLECT_X)); FAIL_ON(mode.refresh_specified); @@ -913,7 +924,7 @@ static int drm_cmdline_test_vmirror(void *ignored) FAIL_ON(!mode.specified); FAIL_ON(mode.xres != 720); FAIL_ON(mode.yres != 480); - FAIL_ON(mode.rotation_reflection != DRM_MODE_REFLECT_Y); + FAIL_ON(mode.rotation_reflection != (DRM_MODE_ROTATE_0 | DRM_MODE_REFLECT_Y)); FAIL_ON(mode.refresh_specified); diff --git a/drivers/gpu/drm/sun4i/sun8i_mixer.c b/drivers/gpu/drm/sun4i/sun8i_mixer.c index 7c24f8f832a5..4a64f7ae437a 100644 --- a/drivers/gpu/drm/sun4i/sun8i_mixer.c +++ b/drivers/gpu/drm/sun4i/sun8i_mixer.c @@ -107,48 +107,128 @@ static const struct de2_fmt_info de2_formats[] = { .csc = SUN8I_CSC_MODE_OFF, }, { + /* for DE2 VI layer which ignores alpha */ + .drm_fmt = DRM_FORMAT_XRGB4444, + .de2_fmt = SUN8I_MIXER_FBFMT_ARGB4444, + .rgb = true, + .csc = SUN8I_CSC_MODE_OFF, + }, + { .drm_fmt = DRM_FORMAT_ABGR4444, .de2_fmt = SUN8I_MIXER_FBFMT_ABGR4444, .rgb = true, .csc = SUN8I_CSC_MODE_OFF, }, { + /* for DE2 VI layer which ignores alpha */ + .drm_fmt = DRM_FORMAT_XBGR4444, + .de2_fmt = SUN8I_MIXER_FBFMT_ABGR4444, + .rgb = true, + .csc = SUN8I_CSC_MODE_OFF, + }, + { .drm_fmt = DRM_FORMAT_RGBA4444, .de2_fmt = SUN8I_MIXER_FBFMT_RGBA4444, .rgb = true, .csc = SUN8I_CSC_MODE_OFF, }, { + /* for DE2 VI layer which ignores alpha */ + .drm_fmt = DRM_FORMAT_RGBX4444, + .de2_fmt = SUN8I_MIXER_FBFMT_RGBA4444, + .rgb = true, + .csc = SUN8I_CSC_MODE_OFF, + }, + { .drm_fmt = DRM_FORMAT_BGRA4444, .de2_fmt = SUN8I_MIXER_FBFMT_BGRA4444, .rgb = true, .csc = SUN8I_CSC_MODE_OFF, }, { + /* for DE2 VI layer which ignores alpha */ + .drm_fmt = DRM_FORMAT_BGRX4444, + .de2_fmt = SUN8I_MIXER_FBFMT_BGRA4444, + .rgb = true, + .csc = SUN8I_CSC_MODE_OFF, + }, + { .drm_fmt = DRM_FORMAT_ARGB1555, .de2_fmt = SUN8I_MIXER_FBFMT_ARGB1555, .rgb = true, .csc = SUN8I_CSC_MODE_OFF, }, { + /* for DE2 VI layer which ignores alpha */ + .drm_fmt = DRM_FORMAT_XRGB1555, + .de2_fmt = SUN8I_MIXER_FBFMT_ARGB1555, + .rgb = true, + .csc = SUN8I_CSC_MODE_OFF, + }, + { .drm_fmt = DRM_FORMAT_ABGR1555, .de2_fmt = SUN8I_MIXER_FBFMT_ABGR1555, .rgb = true, .csc = SUN8I_CSC_MODE_OFF, }, { + /* for DE2 VI layer which ignores alpha */ + .drm_fmt = DRM_FORMAT_XBGR1555, + .de2_fmt = SUN8I_MIXER_FBFMT_ABGR1555, + .rgb = true, + .csc = SUN8I_CSC_MODE_OFF, + }, + { .drm_fmt = DRM_FORMAT_RGBA5551, .de2_fmt = SUN8I_MIXER_FBFMT_RGBA5551, .rgb = true, .csc = SUN8I_CSC_MODE_OFF, }, { + /* for DE2 VI layer which ignores alpha */ + .drm_fmt = DRM_FORMAT_RGBX5551, + .de2_fmt = SUN8I_MIXER_FBFMT_RGBA5551, + .rgb = true, + .csc = SUN8I_CSC_MODE_OFF, + }, + { .drm_fmt = DRM_FORMAT_BGRA5551, .de2_fmt = SUN8I_MIXER_FBFMT_BGRA5551, .rgb = true, .csc = SUN8I_CSC_MODE_OFF, }, { + /* for DE2 VI layer which ignores alpha */ + .drm_fmt = DRM_FORMAT_BGRX5551, + .de2_fmt = SUN8I_MIXER_FBFMT_BGRA5551, + .rgb = true, + .csc = SUN8I_CSC_MODE_OFF, + }, + { + .drm_fmt = DRM_FORMAT_ARGB2101010, + .de2_fmt = SUN8I_MIXER_FBFMT_ARGB2101010, + .rgb = true, + .csc = SUN8I_CSC_MODE_OFF, + }, + { + .drm_fmt = DRM_FORMAT_ABGR2101010, + .de2_fmt = SUN8I_MIXER_FBFMT_ABGR2101010, + .rgb = true, + .csc = SUN8I_CSC_MODE_OFF, + }, + { + .drm_fmt = DRM_FORMAT_RGBA1010102, + .de2_fmt = SUN8I_MIXER_FBFMT_RGBA1010102, + .rgb = true, + .csc = SUN8I_CSC_MODE_OFF, + }, + { + .drm_fmt = DRM_FORMAT_BGRA1010102, + .de2_fmt = SUN8I_MIXER_FBFMT_BGRA1010102, + .rgb = true, + .csc = SUN8I_CSC_MODE_OFF, + }, + { .drm_fmt = DRM_FORMAT_UYVY, .de2_fmt = SUN8I_MIXER_FBFMT_UYVY, .rgb = false, @@ -197,12 +277,6 @@ static const struct de2_fmt_info de2_formats[] = { .csc = SUN8I_CSC_MODE_YUV2RGB, }, { - .drm_fmt = DRM_FORMAT_YUV444, - .de2_fmt = SUN8I_MIXER_FBFMT_RGB888, - .rgb = true, - .csc = SUN8I_CSC_MODE_YUV2RGB, - }, - { .drm_fmt = DRM_FORMAT_YUV422, .de2_fmt = SUN8I_MIXER_FBFMT_YUV422, .rgb = false, @@ -221,12 +295,6 @@ static const struct de2_fmt_info de2_formats[] = { .csc = SUN8I_CSC_MODE_YUV2RGB, }, { - .drm_fmt = DRM_FORMAT_YVU444, - .de2_fmt = SUN8I_MIXER_FBFMT_RGB888, - .rgb = true, - .csc = SUN8I_CSC_MODE_YVU2RGB, - }, - { .drm_fmt = DRM_FORMAT_YVU422, .de2_fmt = SUN8I_MIXER_FBFMT_YUV422, .rgb = false, @@ -244,6 +312,18 @@ static const struct de2_fmt_info de2_formats[] = { .rgb = false, .csc = SUN8I_CSC_MODE_YVU2RGB, }, + { + .drm_fmt = DRM_FORMAT_P010, + .de2_fmt = SUN8I_MIXER_FBFMT_P010_YUV, + .rgb = false, + .csc = SUN8I_CSC_MODE_YUV2RGB, + }, + { + .drm_fmt = DRM_FORMAT_P210, + .de2_fmt = SUN8I_MIXER_FBFMT_P210_YUV, + .rgb = false, + .csc = SUN8I_CSC_MODE_YUV2RGB, + }, }; const struct de2_fmt_info *sun8i_mixer_format_info(u32 format) diff --git a/drivers/gpu/drm/sun4i/sun8i_mixer.h b/drivers/gpu/drm/sun4i/sun8i_mixer.h index c6cc94057faf..345b28b0a80a 100644 --- a/drivers/gpu/drm/sun4i/sun8i_mixer.h +++ b/drivers/gpu/drm/sun4i/sun8i_mixer.h @@ -93,6 +93,10 @@ #define SUN8I_MIXER_FBFMT_ABGR1555 17 #define SUN8I_MIXER_FBFMT_RGBA5551 18 #define SUN8I_MIXER_FBFMT_BGRA5551 19 +#define SUN8I_MIXER_FBFMT_ARGB2101010 20 +#define SUN8I_MIXER_FBFMT_ABGR2101010 21 +#define SUN8I_MIXER_FBFMT_RGBA1010102 22 +#define SUN8I_MIXER_FBFMT_BGRA1010102 23 #define SUN8I_MIXER_FBFMT_YUYV 0 #define SUN8I_MIXER_FBFMT_UYVY 1 @@ -109,6 +113,13 @@ /* format 12 is semi-planar YUV411 UVUV */ /* format 13 is semi-planar YUV411 VUVU */ #define SUN8I_MIXER_FBFMT_YUV411 14 +/* format 15 doesn't exist */ +/* format 16 is P010 YVU */ +#define SUN8I_MIXER_FBFMT_P010_YUV 17 +/* format 18 is P210 YVU */ +#define SUN8I_MIXER_FBFMT_P210_YUV 19 +/* format 20 is packed YVU444 10-bit */ +/* format 21 is packed YUV444 10-bit */ /* * Sub-engines listed bellow are unused for now. The EN registers are here only diff --git a/drivers/gpu/drm/sun4i/sun8i_vi_layer.c b/drivers/gpu/drm/sun4i/sun8i_vi_layer.c index 42d445d23773..b8398ca18b0f 100644 --- a/drivers/gpu/drm/sun4i/sun8i_vi_layer.c +++ b/drivers/gpu/drm/sun4i/sun8i_vi_layer.c @@ -398,24 +398,66 @@ static const struct drm_plane_funcs sun8i_vi_layer_funcs = { }; /* - * While all RGB formats are supported, VI planes don't support - * alpha blending, so there is no point having formats with alpha - * channel if their opaque analog exist. + * While DE2 VI layer supports same RGB formats as UI layer, alpha + * channel is ignored. This structure lists all unique variants + * where alpha channel is replaced with "don't care" (X) channel. */ static const u32 sun8i_vi_layer_formats[] = { + DRM_FORMAT_BGR565, + DRM_FORMAT_BGR888, + DRM_FORMAT_BGRX4444, + DRM_FORMAT_BGRX5551, + DRM_FORMAT_BGRX8888, + DRM_FORMAT_RGB565, + DRM_FORMAT_RGB888, + DRM_FORMAT_RGBX4444, + DRM_FORMAT_RGBX5551, + DRM_FORMAT_RGBX8888, + DRM_FORMAT_XBGR1555, + DRM_FORMAT_XBGR4444, + DRM_FORMAT_XBGR8888, + DRM_FORMAT_XRGB1555, + DRM_FORMAT_XRGB4444, + DRM_FORMAT_XRGB8888, + + DRM_FORMAT_NV16, + DRM_FORMAT_NV12, + DRM_FORMAT_NV21, + DRM_FORMAT_NV61, + DRM_FORMAT_UYVY, + DRM_FORMAT_VYUY, + DRM_FORMAT_YUYV, + DRM_FORMAT_YVYU, + DRM_FORMAT_YUV411, + DRM_FORMAT_YUV420, + DRM_FORMAT_YUV422, + DRM_FORMAT_YVU411, + DRM_FORMAT_YVU420, + DRM_FORMAT_YVU422, +}; + +static const u32 sun8i_vi_layer_de3_formats[] = { DRM_FORMAT_ABGR1555, + DRM_FORMAT_ABGR2101010, DRM_FORMAT_ABGR4444, + DRM_FORMAT_ABGR8888, DRM_FORMAT_ARGB1555, + DRM_FORMAT_ARGB2101010, DRM_FORMAT_ARGB4444, + DRM_FORMAT_ARGB8888, DRM_FORMAT_BGR565, DRM_FORMAT_BGR888, + DRM_FORMAT_BGRA1010102, DRM_FORMAT_BGRA5551, DRM_FORMAT_BGRA4444, + DRM_FORMAT_BGRA8888, DRM_FORMAT_BGRX8888, DRM_FORMAT_RGB565, DRM_FORMAT_RGB888, + DRM_FORMAT_RGBA1010102, DRM_FORMAT_RGBA4444, DRM_FORMAT_RGBA5551, + DRM_FORMAT_RGBA8888, DRM_FORMAT_RGBX8888, DRM_FORMAT_XBGR8888, DRM_FORMAT_XRGB8888, @@ -424,6 +466,8 @@ static const u32 sun8i_vi_layer_formats[] = { DRM_FORMAT_NV12, DRM_FORMAT_NV21, DRM_FORMAT_NV61, + DRM_FORMAT_P010, + DRM_FORMAT_P210, DRM_FORMAT_UYVY, DRM_FORMAT_VYUY, DRM_FORMAT_YUYV, @@ -431,11 +475,9 @@ static const u32 sun8i_vi_layer_formats[] = { DRM_FORMAT_YUV411, DRM_FORMAT_YUV420, DRM_FORMAT_YUV422, - DRM_FORMAT_YUV444, DRM_FORMAT_YVU411, DRM_FORMAT_YVU420, DRM_FORMAT_YVU422, - DRM_FORMAT_YVU444, }; struct sun8i_vi_layer *sun8i_vi_layer_init_one(struct drm_device *drm, @@ -443,19 +485,27 @@ struct sun8i_vi_layer *sun8i_vi_layer_init_one(struct drm_device *drm, int index) { u32 supported_encodings, supported_ranges; + unsigned int plane_cnt, format_count; struct sun8i_vi_layer *layer; - unsigned int plane_cnt; + const u32 *formats; int ret; layer = devm_kzalloc(drm->dev, sizeof(*layer), GFP_KERNEL); if (!layer) return ERR_PTR(-ENOMEM); + if (mixer->cfg->is_de3) { + formats = sun8i_vi_layer_de3_formats; + format_count = ARRAY_SIZE(sun8i_vi_layer_de3_formats); + } else { + formats = sun8i_vi_layer_formats; + format_count = ARRAY_SIZE(sun8i_vi_layer_formats); + } + /* possible crtcs are set later */ ret = drm_universal_plane_init(drm, &layer->plane, 0, &sun8i_vi_layer_funcs, - sun8i_vi_layer_formats, - ARRAY_SIZE(sun8i_vi_layer_formats), + formats, format_count, NULL, DRM_PLANE_TYPE_OVERLAY, NULL); if (ret) { dev_err(drm->dev, "Couldn't initialize layer\n"); diff --git a/drivers/gpu/drm/tegra/dc.c b/drivers/gpu/drm/tegra/dc.c index 7c70fd31a4c2..1a7b08f35776 100644 --- a/drivers/gpu/drm/tegra/dc.c +++ b/drivers/gpu/drm/tegra/dc.c @@ -2503,7 +2503,6 @@ static int tegra_dc_couple(struct tegra_dc *dc) static int tegra_dc_probe(struct platform_device *pdev) { - struct resource *regs; struct tegra_dc *dc; int err; @@ -2560,8 +2559,7 @@ static int tegra_dc_probe(struct platform_device *pdev) tegra_powergate_power_off(dc->powergate); } - regs = platform_get_resource(pdev, IORESOURCE_MEM, 0); - dc->regs = devm_ioremap_resource(&pdev->dev, regs); + dc->regs = devm_platform_ioremap_resource(pdev, 0); if (IS_ERR(dc->regs)) return PTR_ERR(dc->regs); @@ -2573,7 +2571,13 @@ static int tegra_dc_probe(struct platform_device *pdev) err = tegra_dc_rgb_probe(dc); if (err < 0 && err != -ENODEV) { - dev_err(&pdev->dev, "failed to probe RGB output: %d\n", err); + const char *level = KERN_ERR; + + if (err == -EPROBE_DEFER) + level = KERN_DEBUG; + + dev_printk(level, dc->dev, "failed to probe RGB output: %d\n", + err); return err; } @@ -2588,10 +2592,16 @@ static int tegra_dc_probe(struct platform_device *pdev) if (err < 0) { dev_err(&pdev->dev, "failed to register host1x client: %d\n", err); - return err; + goto disable_pm; } return 0; + +disable_pm: + pm_runtime_disable(&pdev->dev); + tegra_dc_rgb_remove(dc); + + return err; } static int tegra_dc_remove(struct platform_device *pdev) diff --git a/drivers/gpu/drm/tegra/hdmi.c b/drivers/gpu/drm/tegra/hdmi.c index 6f117628f257..38252c0f068d 100644 --- a/drivers/gpu/drm/tegra/hdmi.c +++ b/drivers/gpu/drm/tegra/hdmi.c @@ -1648,6 +1648,7 @@ static irqreturn_t tegra_hdmi_irq(int irq, void *data) static int tegra_hdmi_probe(struct platform_device *pdev) { + const char *level = KERN_ERR; struct tegra_hdmi *hdmi; struct resource *regs; int err; @@ -1686,21 +1687,36 @@ static int tegra_hdmi_probe(struct platform_device *pdev) } hdmi->hdmi = devm_regulator_get(&pdev->dev, "hdmi"); - if (IS_ERR(hdmi->hdmi)) { - dev_err(&pdev->dev, "failed to get HDMI regulator\n"); - return PTR_ERR(hdmi->hdmi); + err = PTR_ERR_OR_ZERO(hdmi->hdmi); + if (err) { + if (err == -EPROBE_DEFER) + level = KERN_DEBUG; + + dev_printk(level, &pdev->dev, + "failed to get HDMI regulator: %d\n", err); + return err; } hdmi->pll = devm_regulator_get(&pdev->dev, "pll"); - if (IS_ERR(hdmi->pll)) { - dev_err(&pdev->dev, "failed to get PLL regulator\n"); - return PTR_ERR(hdmi->pll); + err = PTR_ERR_OR_ZERO(hdmi->pll); + if (err) { + if (err == -EPROBE_DEFER) + level = KERN_DEBUG; + + dev_printk(level, &pdev->dev, + "failed to get PLL regulator: %d\n", err); + return err; } hdmi->vdd = devm_regulator_get(&pdev->dev, "vdd"); - if (IS_ERR(hdmi->vdd)) { - dev_err(&pdev->dev, "failed to get VDD regulator\n"); - return PTR_ERR(hdmi->vdd); + err = PTR_ERR_OR_ZERO(hdmi->vdd); + if (err) { + if (err == -EPROBE_DEFER) + level = KERN_DEBUG; + + dev_printk(level, &pdev->dev, + "failed to get VDD regulator: %d\n", err); + return err; } hdmi->output.dev = &pdev->dev; diff --git a/drivers/gpu/drm/ttm/ttm_bo_util.c b/drivers/gpu/drm/ttm/ttm_bo_util.c index f1f670642c97..52d2b71f1588 100644 --- a/drivers/gpu/drm/ttm/ttm_bo_util.c +++ b/drivers/gpu/drm/ttm/ttm_bo_util.c @@ -514,6 +514,7 @@ static int ttm_buffer_object_transfer(struct ttm_buffer_object *bo, fbo->base.base.resv = &fbo->base.base._resv; dma_resv_init(&fbo->base.base._resv); + fbo->base.base.dev = NULL; ret = dma_resv_trylock(&fbo->base.base._resv); WARN_ON(!ret); diff --git a/drivers/gpu/drm/ttm/ttm_bo_vm.c b/drivers/gpu/drm/ttm/ttm_bo_vm.c index 389128b8c4dd..0af14835504c 100644 --- a/drivers/gpu/drm/ttm/ttm_bo_vm.c +++ b/drivers/gpu/drm/ttm/ttm_bo_vm.c @@ -156,6 +156,89 @@ vm_fault_t ttm_bo_vm_reserve(struct ttm_buffer_object *bo, } EXPORT_SYMBOL(ttm_bo_vm_reserve); +#ifdef CONFIG_TRANSPARENT_HUGEPAGE +/** + * ttm_bo_vm_insert_huge - Insert a pfn for PUD or PMD faults + * @vmf: Fault data + * @bo: The buffer object + * @page_offset: Page offset from bo start + * @fault_page_size: The size of the fault in pages. + * @pgprot: The page protections. + * Does additional checking whether it's possible to insert a PUD or PMD + * pfn and performs the insertion. + * + * Return: VM_FAULT_NOPAGE on successful insertion, VM_FAULT_FALLBACK if + * a huge fault was not possible, or on insertion error. + */ +static vm_fault_t ttm_bo_vm_insert_huge(struct vm_fault *vmf, + struct ttm_buffer_object *bo, + pgoff_t page_offset, + pgoff_t fault_page_size, + pgprot_t pgprot) +{ + pgoff_t i; + vm_fault_t ret; + unsigned long pfn; + pfn_t pfnt; + struct ttm_tt *ttm = bo->ttm; + bool write = vmf->flags & FAULT_FLAG_WRITE; + + /* Fault should not cross bo boundary. */ + page_offset &= ~(fault_page_size - 1); + if (page_offset + fault_page_size > bo->num_pages) + goto out_fallback; + + if (bo->mem.bus.is_iomem) + pfn = ttm_bo_io_mem_pfn(bo, page_offset); + else + pfn = page_to_pfn(ttm->pages[page_offset]); + + /* pfn must be fault_page_size aligned. */ + if ((pfn & (fault_page_size - 1)) != 0) + goto out_fallback; + + /* Check that memory is contiguous. */ + if (!bo->mem.bus.is_iomem) { + for (i = 1; i < fault_page_size; ++i) { + if (page_to_pfn(ttm->pages[page_offset + i]) != pfn + i) + goto out_fallback; + } + } else if (bo->bdev->driver->io_mem_pfn) { + for (i = 1; i < fault_page_size; ++i) { + if (ttm_bo_io_mem_pfn(bo, page_offset + i) != pfn + i) + goto out_fallback; + } + } + + pfnt = __pfn_to_pfn_t(pfn, PFN_DEV); + if (fault_page_size == (HPAGE_PMD_SIZE >> PAGE_SHIFT)) + ret = vmf_insert_pfn_pmd_prot(vmf, pfnt, pgprot, write); +#ifdef CONFIG_HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD + else if (fault_page_size == (HPAGE_PUD_SIZE >> PAGE_SHIFT)) + ret = vmf_insert_pfn_pud_prot(vmf, pfnt, pgprot, write); +#endif + else + WARN_ON_ONCE(ret = VM_FAULT_FALLBACK); + + if (ret != VM_FAULT_NOPAGE) + goto out_fallback; + + return VM_FAULT_NOPAGE; +out_fallback: + count_vm_event(THP_FAULT_FALLBACK); + return VM_FAULT_FALLBACK; +} +#else +static vm_fault_t ttm_bo_vm_insert_huge(struct vm_fault *vmf, + struct ttm_buffer_object *bo, + pgoff_t page_offset, + pgoff_t fault_page_size, + pgprot_t pgprot) +{ + return VM_FAULT_FALLBACK; +} +#endif + /** * ttm_bo_vm_fault_reserved - TTM fault helper * @vmf: The struct vm_fault given as argument to the fault callback @@ -163,6 +246,7 @@ EXPORT_SYMBOL(ttm_bo_vm_reserve); * @num_prefault: Maximum number of prefault pages. The caller may want to * specify this based on madvice settings and the size of the GPU object * backed by the memory. + * @fault_page_size: The size of the fault in pages. * * This function inserts one or more page table entries pointing to the * memory backing the buffer object, and then returns a return code @@ -176,7 +260,8 @@ EXPORT_SYMBOL(ttm_bo_vm_reserve); */ vm_fault_t ttm_bo_vm_fault_reserved(struct vm_fault *vmf, pgprot_t prot, - pgoff_t num_prefault) + pgoff_t num_prefault, + pgoff_t fault_page_size) { struct vm_area_struct *vma = vmf->vma; struct ttm_buffer_object *bo = vma->vm_private_data; @@ -268,6 +353,13 @@ vm_fault_t ttm_bo_vm_fault_reserved(struct vm_fault *vmf, prot = pgprot_decrypted(prot); } + /* We don't prefault on huge faults. Yet. */ + if (IS_ENABLED(CONFIG_TRANSPARENT_HUGEPAGE) && fault_page_size != 1) { + ret = ttm_bo_vm_insert_huge(vmf, bo, page_offset, + fault_page_size, prot); + goto out_io_unlock; + } + /* * Speculatively prefault a number of pages. Only error on * first page. @@ -334,7 +426,7 @@ vm_fault_t ttm_bo_vm_fault(struct vm_fault *vmf) return ret; prot = vma->vm_page_prot; - ret = ttm_bo_vm_fault_reserved(vmf, prot, TTM_BO_VM_NUM_PREFAULT); + ret = ttm_bo_vm_fault_reserved(vmf, prot, TTM_BO_VM_NUM_PREFAULT, 1); if (ret == VM_FAULT_RETRY && !(vmf->flags & FAULT_FLAG_RETRY_NOWAIT)) return ret; @@ -344,6 +436,66 @@ vm_fault_t ttm_bo_vm_fault(struct vm_fault *vmf) } EXPORT_SYMBOL(ttm_bo_vm_fault); +#ifdef CONFIG_TRANSPARENT_HUGEPAGE +/** + * ttm_pgprot_is_wrprotecting - Is a page protection value write-protecting? + * @prot: The page protection value + * + * Return: true if @prot is write-protecting. false otherwise. + */ +static bool ttm_pgprot_is_wrprotecting(pgprot_t prot) +{ + /* + * This is meant to say "pgprot_wrprotect(prot) == prot" in a generic + * way. Unfortunately there is no generic pgprot_wrprotect. + */ + return pte_val(pte_wrprotect(__pte(pgprot_val(prot)))) == + pgprot_val(prot); +} + +static vm_fault_t ttm_bo_vm_huge_fault(struct vm_fault *vmf, + enum page_entry_size pe_size) +{ + struct vm_area_struct *vma = vmf->vma; + pgprot_t prot; + struct ttm_buffer_object *bo = vma->vm_private_data; + vm_fault_t ret; + pgoff_t fault_page_size = 0; + bool write = vmf->flags & FAULT_FLAG_WRITE; + + switch (pe_size) { + case PE_SIZE_PMD: + fault_page_size = HPAGE_PMD_SIZE >> PAGE_SHIFT; + break; +#ifdef CONFIG_HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD + case PE_SIZE_PUD: + fault_page_size = HPAGE_PUD_SIZE >> PAGE_SHIFT; + break; +#endif + default: + WARN_ON_ONCE(1); + return VM_FAULT_FALLBACK; + } + + /* Fallback on write dirty-tracking or COW */ + if (write && ttm_pgprot_is_wrprotecting(vma->vm_page_prot)) + return VM_FAULT_FALLBACK; + + ret = ttm_bo_vm_reserve(bo, vmf); + if (ret) + return ret; + + prot = vm_get_page_prot(vma->vm_flags); + ret = ttm_bo_vm_fault_reserved(vmf, prot, 1, fault_page_size); + if (ret == VM_FAULT_RETRY && !(vmf->flags & FAULT_FLAG_RETRY_NOWAIT)) + return ret; + + dma_resv_unlock(bo->base.resv); + + return ret; +} +#endif + void ttm_bo_vm_open(struct vm_area_struct *vma) { struct ttm_buffer_object *bo = vma->vm_private_data; @@ -445,7 +597,10 @@ static const struct vm_operations_struct ttm_bo_vm_ops = { .fault = ttm_bo_vm_fault, .open = ttm_bo_vm_open, .close = ttm_bo_vm_close, - .access = ttm_bo_vm_access + .access = ttm_bo_vm_access, +#ifdef CONFIG_TRANSPARENT_HUGEPAGE + .huge_fault = ttm_bo_vm_huge_fault, +#endif }; static struct ttm_buffer_object *ttm_bo_vm_lookup(struct ttm_bo_device *bdev, diff --git a/drivers/gpu/drm/virtio/virtgpu_object.c b/drivers/gpu/drm/virtio/virtgpu_object.c index cc65e8b3ec8b..2bfb13d1932e 100644 --- a/drivers/gpu/drm/virtio/virtgpu_object.c +++ b/drivers/gpu/drm/virtio/virtgpu_object.c @@ -43,8 +43,8 @@ static int virtio_gpu_resource_id_get(struct virtio_gpu_device *vgdev, * "f91a9dd35715 Fix unlinking resources from hash * table." (Feb 2019) fixes the bug. */ - static int handle; - handle++; + static atomic_t seqno = ATOMIC_INIT(0); + int handle = atomic_inc_return(&seqno); *resid = handle + 1; } else { int handle = ida_alloc(&vgdev->resource_ida, GFP_KERNEL); @@ -130,6 +130,7 @@ struct drm_gem_object *virtio_gpu_create_object(struct drm_device *dev, return NULL; bo->base.base.funcs = &virtio_gpu_shmem_funcs; + bo->base.map_cached = true; return &bo->base.base; } diff --git a/drivers/gpu/drm/vmwgfx/Makefile b/drivers/gpu/drm/vmwgfx/Makefile index c877a21a0739..31f85f09f1fc 100644 --- a/drivers/gpu/drm/vmwgfx/Makefile +++ b/drivers/gpu/drm/vmwgfx/Makefile @@ -8,7 +8,8 @@ vmwgfx-y := vmwgfx_execbuf.o vmwgfx_gmr.o vmwgfx_kms.o vmwgfx_drv.o \ vmwgfx_cmdbuf_res.o vmwgfx_cmdbuf.o vmwgfx_stdu.o \ vmwgfx_cotable.o vmwgfx_so.o vmwgfx_binding.o vmwgfx_msg.o \ vmwgfx_simple_resource.o vmwgfx_va.o vmwgfx_blit.o \ - vmwgfx_validation.o vmwgfx_page_dirty.o \ + vmwgfx_validation.o vmwgfx_page_dirty.o vmwgfx_streamoutput.o \ ttm_object.o ttm_lock.o +vmwgfx-$(CONFIG_TRANSPARENT_HUGEPAGE) += vmwgfx_thp.o obj-$(CONFIG_DRM_VMWGFX) := vmwgfx.o diff --git a/drivers/gpu/drm/vmwgfx/device_include/svga3d_cmd.h b/drivers/gpu/drm/vmwgfx/device_include/svga3d_cmd.h index 9cbba0e8ce6a..799bc0963f7a 100644 --- a/drivers/gpu/drm/vmwgfx/device_include/svga3d_cmd.h +++ b/drivers/gpu/drm/vmwgfx/device_include/svga3d_cmd.h @@ -1,6 +1,6 @@ /* SPDX-License-Identifier: GPL-2.0 OR MIT */ /********************************************************** - * Copyright 1998-2015 VMware, Inc. + * Copyright 1998-2020 VMware, Inc. * * Permission is hereby granted, free of charge, to any person * obtaining a copy of this software and associated documentation @@ -104,12 +104,12 @@ typedef enum { SVGA_3D_CMD_DEAD1 = 1083, SVGA_3D_CMD_DEAD2 = 1084, - SVGA_3D_CMD_LOGICOPS_BITBLT = 1085, - SVGA_3D_CMD_LOGICOPS_TRANSBLT = 1086, - SVGA_3D_CMD_LOGICOPS_STRETCHBLT = 1087, - SVGA_3D_CMD_LOGICOPS_COLORFILL = 1088, - SVGA_3D_CMD_LOGICOPS_ALPHABLEND = 1089, - SVGA_3D_CMD_LOGICOPS_CLEARTYPEBLEND = 1090, + SVGA_3D_CMD_DEAD12 = 1085, + SVGA_3D_CMD_DEAD13 = 1086, + SVGA_3D_CMD_DEAD14 = 1087, + SVGA_3D_CMD_DEAD15 = 1088, + SVGA_3D_CMD_DEAD16 = 1089, + SVGA_3D_CMD_DEAD17 = 1090, SVGA_3D_CMD_SET_OTABLE_BASE = 1091, SVGA_3D_CMD_READBACK_OTABLE = 1092, @@ -261,30 +261,23 @@ typedef enum { SVGA_3D_CMD_DX_SET_VS_CONSTANT_BUFFER_OFFSET = 1220, SVGA_3D_CMD_DX_SET_PS_CONSTANT_BUFFER_OFFSET = 1221, SVGA_3D_CMD_DX_SET_GS_CONSTANT_BUFFER_OFFSET = 1222, - - /* - * Reserve some IDs to be used for the SM5 shader types. - */ - SVGA_3D_CMD_DX_RESERVED1 = 1223, - SVGA_3D_CMD_DX_RESERVED2 = 1224, - SVGA_3D_CMD_DX_RESERVED3 = 1225, + SVGA_3D_CMD_DX_SET_HS_CONSTANT_BUFFER_OFFSET = 1223, + SVGA_3D_CMD_DX_SET_DS_CONSTANT_BUFFER_OFFSET = 1224, + SVGA_3D_CMD_DX_SET_CS_CONSTANT_BUFFER_OFFSET = 1225, SVGA_3D_CMD_DX_COND_BIND_ALL_SHADER = 1226, SVGA_3D_CMD_DX_MAX = 1227, SVGA_3D_CMD_SCREEN_COPY = 1227, - /* - * Reserve some IDs to be used for video. - */ - SVGA_3D_CMD_VIDEO_RESERVED1 = 1228, - SVGA_3D_CMD_VIDEO_RESERVED2 = 1229, - SVGA_3D_CMD_VIDEO_RESERVED3 = 1230, - SVGA_3D_CMD_VIDEO_RESERVED4 = 1231, - SVGA_3D_CMD_VIDEO_RESERVED5 = 1232, - SVGA_3D_CMD_VIDEO_RESERVED6 = 1233, - SVGA_3D_CMD_VIDEO_RESERVED7 = 1234, - SVGA_3D_CMD_VIDEO_RESERVED8 = 1235, + SVGA_3D_CMD_RESERVED1 = 1228, + SVGA_3D_CMD_RESERVED2 = 1229, + SVGA_3D_CMD_RESERVED3 = 1230, + SVGA_3D_CMD_RESERVED4 = 1231, + SVGA_3D_CMD_RESERVED5 = 1232, + SVGA_3D_CMD_RESERVED6 = 1233, + SVGA_3D_CMD_RESERVED7 = 1234, + SVGA_3D_CMD_RESERVED8 = 1235, SVGA_3D_CMD_GROW_OTABLE = 1236, SVGA_3D_CMD_DX_GROW_COTABLE = 1237, @@ -298,7 +291,46 @@ typedef enum { SVGA_3D_CMD_DX_PRED_CONVERT = 1243, SVGA_3D_CMD_WHOLE_SURFACE_COPY = 1244, - SVGA_3D_CMD_MAX = 1245, + SVGA_3D_CMD_DX_DEFINE_UA_VIEW = 1245, + SVGA_3D_CMD_DX_DESTROY_UA_VIEW = 1246, + SVGA_3D_CMD_DX_CLEAR_UA_VIEW_UINT = 1247, + SVGA_3D_CMD_DX_CLEAR_UA_VIEW_FLOAT = 1248, + SVGA_3D_CMD_DX_COPY_STRUCTURE_COUNT = 1249, + SVGA_3D_CMD_DX_SET_UA_VIEWS = 1250, + + SVGA_3D_CMD_DX_DRAW_INDEXED_INSTANCED_INDIRECT = 1251, + SVGA_3D_CMD_DX_DRAW_INSTANCED_INDIRECT = 1252, + SVGA_3D_CMD_DX_DISPATCH = 1253, + SVGA_3D_CMD_DX_DISPATCH_INDIRECT = 1254, + + SVGA_3D_CMD_WRITE_ZERO_SURFACE = 1255, + SVGA_3D_CMD_HINT_ZERO_SURFACE = 1256, + SVGA_3D_CMD_DX_TRANSFER_TO_BUFFER = 1257, + SVGA_3D_CMD_DX_SET_STRUCTURE_COUNT = 1258, + + SVGA_3D_CMD_LOGICOPS_BITBLT = 1259, + SVGA_3D_CMD_LOGICOPS_TRANSBLT = 1260, + SVGA_3D_CMD_LOGICOPS_STRETCHBLT = 1261, + SVGA_3D_CMD_LOGICOPS_COLORFILL = 1262, + SVGA_3D_CMD_LOGICOPS_ALPHABLEND = 1263, + SVGA_3D_CMD_LOGICOPS_CLEARTYPEBLEND = 1264, + + SVGA_3D_CMD_RESERVED2_1 = 1265, + + SVGA_3D_CMD_RESERVED2_2 = 1266, + SVGA_3D_CMD_DEFINE_GB_SURFACE_V4 = 1267, + SVGA_3D_CMD_DX_SET_CS_UA_VIEWS = 1268, + SVGA_3D_CMD_DX_SET_MIN_LOD = 1269, + SVGA_3D_CMD_RESERVED2_3 = 1270, + SVGA_3D_CMD_RESERVED2_4 = 1271, + SVGA_3D_CMD_DX_DEFINE_DEPTHSTENCIL_VIEW_V2 = 1272, + SVGA_3D_CMD_DX_DEFINE_STREAMOUTPUT_WITH_MOB = 1273, + SVGA_3D_CMD_DX_SET_SHADER_IFACE = 1274, + SVGA_3D_CMD_DX_BIND_STREAMOUTPUT = 1275, + SVGA_3D_CMD_SURFACE_STRETCHBLT_NON_MS_TO_MS = 1276, + SVGA_3D_CMD_DX_BIND_SHADER_IFACE = 1277, + + SVGA_3D_CMD_MAX = 1278, SVGA_3D_CMD_FUTURE_MAX = 3000 } SVGAFifo3dCmdId; @@ -334,6 +366,7 @@ struct { uint32 sid; SVGA3dSurface1Flags surfaceFlags; SVGA3dSurfaceFormat format; + /* * If surfaceFlags has SVGA3D_SURFACE_CUBEMAP bit set, all SVGA3dSurfaceFace * structures must have the same value of numMipLevels field. @@ -341,6 +374,7 @@ struct { * numMipLevels set to 0. */ SVGA3dSurfaceFace face[SVGA3D_MAX_SURFACE_FACES]; + /* * Followed by an SVGA3dSize structure for each mip level in each face. * @@ -360,6 +394,7 @@ struct { uint32 sid; SVGA3dSurface1Flags surfaceFlags; SVGA3dSurfaceFormat format; + /* * If surfaceFlags has SVGA3D_SURFACE_CUBEMAP bit set, all SVGA3dSurfaceFace * structures must have the same value of numMipLevels field. @@ -369,6 +404,7 @@ struct { SVGA3dSurfaceFace face[SVGA3D_MAX_SURFACE_FACES]; uint32 multisampleCount; SVGA3dTextureFilter autogenFilter; + /* * Followed by an SVGA3dSize structure for each mip level in each face. * @@ -517,6 +553,18 @@ typedef struct { SVGA3dSurfaceImageId src; SVGA3dSurfaceImageId dest; + SVGA3dBox boxSrc; + SVGA3dBox boxDest; +} +#include "vmware_pack_end.h" +SVGA3dCmdSurfaceStretchBltNonMSToMS; +/* SVGA_3D_CMD_SURFACE_STRETCHBLT_NON_MS_TO_MS */ + +typedef +#include "vmware_pack_begin.h" +struct { + SVGA3dSurfaceImageId src; + SVGA3dSurfaceImageId dest; SVGA3dBox boxSrc; SVGA3dBox boxDest; SVGA3dStretchBltMode mode; @@ -555,6 +603,7 @@ struct { SVGAGuestImage guest; SVGA3dSurfaceImageId host; SVGA3dTransferType transfer; + /* * Followed by variable number of SVGA3dCopyBox structures. For consistency * in all clipping logic and coordinate translation, we define the @@ -789,7 +838,7 @@ struct { uint32 indexBufferSid; /* Valid index buffer sid. */ uint32 indexBufferOffset; /* Byte offset into the vertex buffer, almost */ - /* always 0 for DX9 guests, non-zero for OpenGL */ + /* always 0 for pre SM guests, non-zero for OpenGL */ /* guests. We can't represent non-multiple of */ /* stride offsets in D3D9Renderer... */ uint8 indexBufferStride; /* Allowable values = 1, 2, or 4 */ @@ -1228,6 +1277,7 @@ struct SVGA3dCmdLogicOpsBitBlt { SVGA3dSurfaceImageId src; SVGA3dSurfaceImageId dst; SVGA3dLogicOp logicOp; + SVGA3dLogicOpRop3 logicOpRop3; /* Followed by variable number of SVGA3dCopyBox structures */ } #include "vmware_pack_end.h" @@ -1247,7 +1297,8 @@ struct SVGA3dCmdLogicOpsTransBlt { uint32 color; uint32 flags; SVGA3dBox srcBox; - SVGA3dBox dstBox; + SVGA3dSignedBox dstBox; + SVGA3dBox clipBox; } #include "vmware_pack_end.h" SVGA3dCmdLogicOpsTransBlt; /* SVGA_3D_CMD_LOGICOPS_TRANSBLT */ @@ -1266,7 +1317,8 @@ struct SVGA3dCmdLogicOpsStretchBlt { uint16 mode; uint16 flags; SVGA3dBox srcBox; - SVGA3dBox dstBox; + SVGA3dSignedBox dstBox; + SVGA3dBox clipBox; } #include "vmware_pack_end.h" SVGA3dCmdLogicOpsStretchBlt; /* SVGA_3D_CMD_LOGICOPS_STRETCHBLT */ @@ -1283,6 +1335,7 @@ struct SVGA3dCmdLogicOpsColorFill { SVGA3dSurfaceImageId dst; uint32 color; SVGA3dLogicOp logicOp; + SVGA3dLogicOpRop3 logicOpRop3; /* Followed by variable number of SVGA3dRect structures. */ } #include "vmware_pack_end.h" @@ -1302,7 +1355,8 @@ struct SVGA3dCmdLogicOpsAlphaBlend { uint32 alphaVal; uint32 flags; SVGA3dBox srcBox; - SVGA3dBox dstBox; + SVGA3dSignedBox dstBox; + SVGA3dBox clipBox; } #include "vmware_pack_end.h" SVGA3dCmdLogicOpsAlphaBlend; /* SVGA_3D_CMD_LOGICOPS_ALPHABLEND */ @@ -1365,8 +1419,9 @@ struct { SVGA3dSurface2Flags surface2Flags; uint8 multisamplePattern; uint8 qualityLevel; - uint8 pad0[2]; - uint32 pad1[3]; + uint16 bufferByteStride; + float minLOD; + uint32 pad0[2]; } #include "vmware_pack_end.h" SVGAOTableSurfaceEntry; @@ -1543,7 +1598,7 @@ typedef #include "vmware_pack_begin.h" struct { SVGAOTableType type; - PPN baseAddress; + PPN32 baseAddress; uint32 sizeInBytes; uint32 validSizeInBytes; SVGAMobFormat ptDepth; @@ -1599,7 +1654,7 @@ typedef struct SVGA3dCmdDefineGBMob { SVGAMobId mobid; SVGAMobFormat ptDepth; - PPN base; + PPN32 base; uint32 sizeInBytes; } #include "vmware_pack_end.h" @@ -1618,7 +1673,6 @@ struct SVGA3dCmdDestroyGBMob { #include "vmware_pack_end.h" SVGA3dCmdDestroyGBMob; /* SVGA_3D_CMD_DESTROY_GB_MOB */ - /* * Define a memory object (Mob) in the OTable with a PPN64 base. */ @@ -1719,6 +1773,27 @@ struct SVGA3dCmdDefineGBSurface_v3 { SVGA3dCmdDefineGBSurface_v3; /* SVGA_3D_CMD_DEFINE_GB_SURFACE_V3 */ /* + * Defines a guest-backed surface, adding buffer byte stride. + */ +typedef +#include "vmware_pack_begin.h" +struct SVGA3dCmdDefineGBSurface_v4 { + uint32 sid; + SVGA3dSurfaceAllFlags surfaceFlags; + SVGA3dSurfaceFormat format; + uint32 numMipLevels; + uint32 multisampleCount; + SVGA3dMSPattern multisamplePattern; + SVGA3dMSQualityLevel qualityLevel; + SVGA3dTextureFilter autogenFilter; + SVGA3dSize size; + uint32 arraySize; + uint32 bufferByteStride; +} +#include "vmware_pack_end.h" +SVGA3dCmdDefineGBSurface_v4; /* SVGA_3D_CMD_DEFINE_GB_SURFACE_V4 */ + +/* * Destroy a guest-backed surface. */ @@ -2181,4 +2256,20 @@ SVGA3dCmdScreenCopy; /* SVGA_3D_CMD_SCREEN_COPY */ #define SVGA_SCREEN_COPY_STATUS_SUCCESS 0x01 #define SVGA_SCREEN_COPY_STATUS_INVALID 0xFFFFFFFF +typedef +#include "vmware_pack_begin.h" +struct { + uint32 sid; +} +#include "vmware_pack_end.h" +SVGA3dCmdWriteZeroSurface; /* SVGA_3D_CMD_WRITE_ZERO_SURFACE */ + +typedef +#include "vmware_pack_begin.h" +struct { + uint32 sid; +} +#include "vmware_pack_end.h" +SVGA3dCmdHintZeroSurface; /* SVGA_3D_CMD_HINT_ZERO_SURFACE */ + #endif /* _SVGA3D_CMD_H_ */ diff --git a/drivers/gpu/drm/vmwgfx/device_include/svga3d_devcaps.h b/drivers/gpu/drm/vmwgfx/device_include/svga3d_devcaps.h index f256560049bf..617b468c626c 100644 --- a/drivers/gpu/drm/vmwgfx/device_include/svga3d_devcaps.h +++ b/drivers/gpu/drm/vmwgfx/device_include/svga3d_devcaps.h @@ -1,6 +1,6 @@ /* SPDX-License-Identifier: GPL-2.0 OR MIT */ /********************************************************** - * Copyright 1998-2015 VMware, Inc. + * Copyright 1998-2019 VMware, Inc. * * Permission is hereby granted, free of charge, to any person * obtaining a copy of this software and associated documentation @@ -39,6 +39,8 @@ #include "includeCheck.h" +#include "svga3d_types.h" + /* * 3D Hardware Version * @@ -69,381 +71,408 @@ typedef enum { * DevCap indexes. */ -typedef enum { - SVGA3D_DEVCAP_INVALID = ((uint32)-1), - SVGA3D_DEVCAP_3D = 0, - SVGA3D_DEVCAP_MAX_LIGHTS = 1, - - /* - * SVGA3D_DEVCAP_MAX_TEXTURES reflects the maximum number of - * fixed-function texture units available. Each of these units - * work in both FFP and Shader modes, and they support texture - * transforms and texture coordinates. The host may have additional - * texture image units that are only usable with shaders. - */ - SVGA3D_DEVCAP_MAX_TEXTURES = 2, - SVGA3D_DEVCAP_MAX_CLIP_PLANES = 3, - SVGA3D_DEVCAP_VERTEX_SHADER_VERSION = 4, - SVGA3D_DEVCAP_VERTEX_SHADER = 5, - SVGA3D_DEVCAP_FRAGMENT_SHADER_VERSION = 6, - SVGA3D_DEVCAP_FRAGMENT_SHADER = 7, - SVGA3D_DEVCAP_MAX_RENDER_TARGETS = 8, - SVGA3D_DEVCAP_S23E8_TEXTURES = 9, - SVGA3D_DEVCAP_S10E5_TEXTURES = 10, - SVGA3D_DEVCAP_MAX_FIXED_VERTEXBLEND = 11, - SVGA3D_DEVCAP_D16_BUFFER_FORMAT = 12, - SVGA3D_DEVCAP_D24S8_BUFFER_FORMAT = 13, - SVGA3D_DEVCAP_D24X8_BUFFER_FORMAT = 14, - SVGA3D_DEVCAP_QUERY_TYPES = 15, - SVGA3D_DEVCAP_TEXTURE_GRADIENT_SAMPLING = 16, - SVGA3D_DEVCAP_MAX_POINT_SIZE = 17, - SVGA3D_DEVCAP_MAX_SHADER_TEXTURES = 18, - SVGA3D_DEVCAP_MAX_TEXTURE_WIDTH = 19, - SVGA3D_DEVCAP_MAX_TEXTURE_HEIGHT = 20, - SVGA3D_DEVCAP_MAX_VOLUME_EXTENT = 21, - SVGA3D_DEVCAP_MAX_TEXTURE_REPEAT = 22, - SVGA3D_DEVCAP_MAX_TEXTURE_ASPECT_RATIO = 23, - SVGA3D_DEVCAP_MAX_TEXTURE_ANISOTROPY = 24, - SVGA3D_DEVCAP_MAX_PRIMITIVE_COUNT = 25, - SVGA3D_DEVCAP_MAX_VERTEX_INDEX = 26, - SVGA3D_DEVCAP_MAX_VERTEX_SHADER_INSTRUCTIONS = 27, - SVGA3D_DEVCAP_MAX_FRAGMENT_SHADER_INSTRUCTIONS = 28, - SVGA3D_DEVCAP_MAX_VERTEX_SHADER_TEMPS = 29, - SVGA3D_DEVCAP_MAX_FRAGMENT_SHADER_TEMPS = 30, - SVGA3D_DEVCAP_TEXTURE_OPS = 31, - SVGA3D_DEVCAP_SURFACEFMT_X8R8G8B8 = 32, - SVGA3D_DEVCAP_SURFACEFMT_A8R8G8B8 = 33, - SVGA3D_DEVCAP_SURFACEFMT_A2R10G10B10 = 34, - SVGA3D_DEVCAP_SURFACEFMT_X1R5G5B5 = 35, - SVGA3D_DEVCAP_SURFACEFMT_A1R5G5B5 = 36, - SVGA3D_DEVCAP_SURFACEFMT_A4R4G4B4 = 37, - SVGA3D_DEVCAP_SURFACEFMT_R5G6B5 = 38, - SVGA3D_DEVCAP_SURFACEFMT_LUMINANCE16 = 39, - SVGA3D_DEVCAP_SURFACEFMT_LUMINANCE8_ALPHA8 = 40, - SVGA3D_DEVCAP_SURFACEFMT_ALPHA8 = 41, - SVGA3D_DEVCAP_SURFACEFMT_LUMINANCE8 = 42, - SVGA3D_DEVCAP_SURFACEFMT_Z_D16 = 43, - SVGA3D_DEVCAP_SURFACEFMT_Z_D24S8 = 44, - SVGA3D_DEVCAP_SURFACEFMT_Z_D24X8 = 45, - SVGA3D_DEVCAP_SURFACEFMT_DXT1 = 46, - SVGA3D_DEVCAP_SURFACEFMT_DXT2 = 47, - SVGA3D_DEVCAP_SURFACEFMT_DXT3 = 48, - SVGA3D_DEVCAP_SURFACEFMT_DXT4 = 49, - SVGA3D_DEVCAP_SURFACEFMT_DXT5 = 50, - SVGA3D_DEVCAP_SURFACEFMT_BUMPX8L8V8U8 = 51, - SVGA3D_DEVCAP_SURFACEFMT_A2W10V10U10 = 52, - SVGA3D_DEVCAP_SURFACEFMT_BUMPU8V8 = 53, - SVGA3D_DEVCAP_SURFACEFMT_Q8W8V8U8 = 54, - SVGA3D_DEVCAP_SURFACEFMT_CxV8U8 = 55, - SVGA3D_DEVCAP_SURFACEFMT_R_S10E5 = 56, - SVGA3D_DEVCAP_SURFACEFMT_R_S23E8 = 57, - SVGA3D_DEVCAP_SURFACEFMT_RG_S10E5 = 58, - SVGA3D_DEVCAP_SURFACEFMT_RG_S23E8 = 59, - SVGA3D_DEVCAP_SURFACEFMT_ARGB_S10E5 = 60, - SVGA3D_DEVCAP_SURFACEFMT_ARGB_S23E8 = 61, - - /* - * There is a hole in our devcap definitions for - * historical reasons. - * - * Define a constant just for completeness. - */ - SVGA3D_DEVCAP_MISSING62 = 62, - - SVGA3D_DEVCAP_MAX_VERTEX_SHADER_TEXTURES = 63, - - /* - * Note that MAX_SIMULTANEOUS_RENDER_TARGETS is a maximum count of color - * render targets. This does not include the depth or stencil targets. - */ - SVGA3D_DEVCAP_MAX_SIMULTANEOUS_RENDER_TARGETS = 64, - - SVGA3D_DEVCAP_SURFACEFMT_V16U16 = 65, - SVGA3D_DEVCAP_SURFACEFMT_G16R16 = 66, - SVGA3D_DEVCAP_SURFACEFMT_A16B16G16R16 = 67, - SVGA3D_DEVCAP_SURFACEFMT_UYVY = 68, - SVGA3D_DEVCAP_SURFACEFMT_YUY2 = 69, - SVGA3D_DEVCAP_MULTISAMPLE_NONMASKABLESAMPLES = 70, - SVGA3D_DEVCAP_MULTISAMPLE_MASKABLESAMPLES = 71, - SVGA3D_DEVCAP_ALPHATOCOVERAGE = 72, - SVGA3D_DEVCAP_SUPERSAMPLE = 73, - SVGA3D_DEVCAP_AUTOGENMIPMAPS = 74, - SVGA3D_DEVCAP_SURFACEFMT_NV12 = 75, - SVGA3D_DEVCAP_SURFACEFMT_AYUV = 76, - - /* - * This is the maximum number of SVGA context IDs that the guest - * can define using SVGA_3D_CMD_CONTEXT_DEFINE. - */ - SVGA3D_DEVCAP_MAX_CONTEXT_IDS = 77, - - /* - * This is the maximum number of SVGA surface IDs that the guest - * can define using SVGA_3D_CMD_SURFACE_DEFINE*. - */ - SVGA3D_DEVCAP_MAX_SURFACE_IDS = 78, - - SVGA3D_DEVCAP_SURFACEFMT_Z_DF16 = 79, - SVGA3D_DEVCAP_SURFACEFMT_Z_DF24 = 80, - SVGA3D_DEVCAP_SURFACEFMT_Z_D24S8_INT = 81, - - SVGA3D_DEVCAP_SURFACEFMT_ATI1 = 82, - SVGA3D_DEVCAP_SURFACEFMT_ATI2 = 83, - - /* - * Deprecated. - */ - SVGA3D_DEVCAP_DEAD1 = 84, - - /* - * This contains several SVGA_3D_CAPS_VIDEO_DECODE elements - * ored together, one for every type of video decoding supported. - */ - SVGA3D_DEVCAP_VIDEO_DECODE = 85, - - /* - * This contains several SVGA_3D_CAPS_VIDEO_PROCESS elements - * ored together, one for every type of video processing supported. - */ - SVGA3D_DEVCAP_VIDEO_PROCESS = 86, - - SVGA3D_DEVCAP_LINE_AA = 87, /* boolean */ - SVGA3D_DEVCAP_LINE_STIPPLE = 88, /* boolean */ - SVGA3D_DEVCAP_MAX_LINE_WIDTH = 89, /* float */ - SVGA3D_DEVCAP_MAX_AA_LINE_WIDTH = 90, /* float */ - - SVGA3D_DEVCAP_SURFACEFMT_YV12 = 91, - - /* - * Does the host support the SVGA logic ops commands? - */ - SVGA3D_DEVCAP_LOGICOPS = 92, - - /* - * Are TS_CONSTANT, TS_COLOR_KEY, and TS_COLOR_KEY_ENABLE supported? - */ - SVGA3D_DEVCAP_TS_COLOR_KEY = 93, /* boolean */ - - /* - * Deprecated. - */ - SVGA3D_DEVCAP_DEAD2 = 94, - - /* - * Does the device support DXContexts? - */ - SVGA3D_DEVCAP_DXCONTEXT = 95, - - /* - * What is the maximum size of a texture array? - * - * (Even if this cap is zero, cubemaps are still allowed.) - */ - SVGA3D_DEVCAP_MAX_TEXTURE_ARRAY_SIZE = 96, - - /* - * What is the maximum number of vertex buffers or vertex input registers - * that can be expected to work correctly with a DXContext? - * - * The guest is allowed to set up to SVGA3D_DX_MAX_VERTEXBUFFERS, but - * anything in excess of this cap is not guaranteed to render correctly. - * - * Similarly, the guest can set up to SVGA3D_DX_MAX_VERTEXINPUTREGISTERS - * input registers without the SVGA3D_DEVCAP_SM4_1 cap, or - * SVGA3D_DX_SM41_MAX_VERTEXINPUTREGISTERS with the SVGA3D_DEVCAP_SM4_1, - * but only the registers up to this cap value are guaranteed to render - * correctly. - * - * If guest-drivers are able to expose a lower-limit, it's recommended - * that they clamp to this value. Otherwise, the host will make a - * best-effort on case-by-case basis if guests exceed this. - */ - SVGA3D_DEVCAP_DX_MAX_VERTEXBUFFERS = 97, - - /* - * What is the maximum number of constant buffers that can be expected to - * work correctly with a DX context? - * - * The guest is allowed to set up to SVGA3D_DX_MAX_CONSTBUFFERS, but - * anything in excess of this cap is not guaranteed to render correctly. - * - * If guest-drivers are able to expose a lower-limit, it's recommended - * that they clamp to this value. Otherwise, the host will make a - * best-effort on case-by-case basis if guests exceed this. - */ - SVGA3D_DEVCAP_DX_MAX_CONSTANT_BUFFERS = 98, - - /* - * Does the device support provoking vertex control? - * - * If this cap is present, the provokingVertexLast field in the - * rasterizer state is enabled. (Guests can then set it to FALSE, - * meaning that the first vertex is the provoking vertex, or TRUE, - * meaning that the last verteix is the provoking vertex.) - * - * If this cap is FALSE, then guests should set the provokingVertexLast - * to FALSE, otherwise rendering behavior is undefined. - */ - SVGA3D_DEVCAP_DX_PROVOKING_VERTEX = 99, - - SVGA3D_DEVCAP_DXFMT_X8R8G8B8 = 100, - SVGA3D_DEVCAP_DXFMT_A8R8G8B8 = 101, - SVGA3D_DEVCAP_DXFMT_R5G6B5 = 102, - SVGA3D_DEVCAP_DXFMT_X1R5G5B5 = 103, - SVGA3D_DEVCAP_DXFMT_A1R5G5B5 = 104, - SVGA3D_DEVCAP_DXFMT_A4R4G4B4 = 105, - SVGA3D_DEVCAP_DXFMT_Z_D32 = 106, - SVGA3D_DEVCAP_DXFMT_Z_D16 = 107, - SVGA3D_DEVCAP_DXFMT_Z_D24S8 = 108, - SVGA3D_DEVCAP_DXFMT_Z_D15S1 = 109, - SVGA3D_DEVCAP_DXFMT_LUMINANCE8 = 110, - SVGA3D_DEVCAP_DXFMT_LUMINANCE4_ALPHA4 = 111, - SVGA3D_DEVCAP_DXFMT_LUMINANCE16 = 112, - SVGA3D_DEVCAP_DXFMT_LUMINANCE8_ALPHA8 = 113, - SVGA3D_DEVCAP_DXFMT_DXT1 = 114, - SVGA3D_DEVCAP_DXFMT_DXT2 = 115, - SVGA3D_DEVCAP_DXFMT_DXT3 = 116, - SVGA3D_DEVCAP_DXFMT_DXT4 = 117, - SVGA3D_DEVCAP_DXFMT_DXT5 = 118, - SVGA3D_DEVCAP_DXFMT_BUMPU8V8 = 119, - SVGA3D_DEVCAP_DXFMT_BUMPL6V5U5 = 120, - SVGA3D_DEVCAP_DXFMT_BUMPX8L8V8U8 = 121, - SVGA3D_DEVCAP_DXFMT_FORMAT_DEAD1 = 122, - SVGA3D_DEVCAP_DXFMT_ARGB_S10E5 = 123, - SVGA3D_DEVCAP_DXFMT_ARGB_S23E8 = 124, - SVGA3D_DEVCAP_DXFMT_A2R10G10B10 = 125, - SVGA3D_DEVCAP_DXFMT_V8U8 = 126, - SVGA3D_DEVCAP_DXFMT_Q8W8V8U8 = 127, - SVGA3D_DEVCAP_DXFMT_CxV8U8 = 128, - SVGA3D_DEVCAP_DXFMT_X8L8V8U8 = 129, - SVGA3D_DEVCAP_DXFMT_A2W10V10U10 = 130, - SVGA3D_DEVCAP_DXFMT_ALPHA8 = 131, - SVGA3D_DEVCAP_DXFMT_R_S10E5 = 132, - SVGA3D_DEVCAP_DXFMT_R_S23E8 = 133, - SVGA3D_DEVCAP_DXFMT_RG_S10E5 = 134, - SVGA3D_DEVCAP_DXFMT_RG_S23E8 = 135, - SVGA3D_DEVCAP_DXFMT_BUFFER = 136, - SVGA3D_DEVCAP_DXFMT_Z_D24X8 = 137, - SVGA3D_DEVCAP_DXFMT_V16U16 = 138, - SVGA3D_DEVCAP_DXFMT_G16R16 = 139, - SVGA3D_DEVCAP_DXFMT_A16B16G16R16 = 140, - SVGA3D_DEVCAP_DXFMT_UYVY = 141, - SVGA3D_DEVCAP_DXFMT_YUY2 = 142, - SVGA3D_DEVCAP_DXFMT_NV12 = 143, - SVGA3D_DEVCAP_DXFMT_AYUV = 144, - SVGA3D_DEVCAP_DXFMT_R32G32B32A32_TYPELESS = 145, - SVGA3D_DEVCAP_DXFMT_R32G32B32A32_UINT = 146, - SVGA3D_DEVCAP_DXFMT_R32G32B32A32_SINT = 147, - SVGA3D_DEVCAP_DXFMT_R32G32B32_TYPELESS = 148, - SVGA3D_DEVCAP_DXFMT_R32G32B32_FLOAT = 149, - SVGA3D_DEVCAP_DXFMT_R32G32B32_UINT = 150, - SVGA3D_DEVCAP_DXFMT_R32G32B32_SINT = 151, - SVGA3D_DEVCAP_DXFMT_R16G16B16A16_TYPELESS = 152, - SVGA3D_DEVCAP_DXFMT_R16G16B16A16_UINT = 153, - SVGA3D_DEVCAP_DXFMT_R16G16B16A16_SNORM = 154, - SVGA3D_DEVCAP_DXFMT_R16G16B16A16_SINT = 155, - SVGA3D_DEVCAP_DXFMT_R32G32_TYPELESS = 156, - SVGA3D_DEVCAP_DXFMT_R32G32_UINT = 157, - SVGA3D_DEVCAP_DXFMT_R32G32_SINT = 158, - SVGA3D_DEVCAP_DXFMT_R32G8X24_TYPELESS = 159, - SVGA3D_DEVCAP_DXFMT_D32_FLOAT_S8X24_UINT = 160, - SVGA3D_DEVCAP_DXFMT_R32_FLOAT_X8X24 = 161, - SVGA3D_DEVCAP_DXFMT_X32_G8X24_UINT = 162, - SVGA3D_DEVCAP_DXFMT_R10G10B10A2_TYPELESS = 163, - SVGA3D_DEVCAP_DXFMT_R10G10B10A2_UINT = 164, - SVGA3D_DEVCAP_DXFMT_R11G11B10_FLOAT = 165, - SVGA3D_DEVCAP_DXFMT_R8G8B8A8_TYPELESS = 166, - SVGA3D_DEVCAP_DXFMT_R8G8B8A8_UNORM = 167, - SVGA3D_DEVCAP_DXFMT_R8G8B8A8_UNORM_SRGB = 168, - SVGA3D_DEVCAP_DXFMT_R8G8B8A8_UINT = 169, - SVGA3D_DEVCAP_DXFMT_R8G8B8A8_SINT = 170, - SVGA3D_DEVCAP_DXFMT_R16G16_TYPELESS = 171, - SVGA3D_DEVCAP_DXFMT_R16G16_UINT = 172, - SVGA3D_DEVCAP_DXFMT_R16G16_SINT = 173, - SVGA3D_DEVCAP_DXFMT_R32_TYPELESS = 174, - SVGA3D_DEVCAP_DXFMT_D32_FLOAT = 175, - SVGA3D_DEVCAP_DXFMT_R32_UINT = 176, - SVGA3D_DEVCAP_DXFMT_R32_SINT = 177, - SVGA3D_DEVCAP_DXFMT_R24G8_TYPELESS = 178, - SVGA3D_DEVCAP_DXFMT_D24_UNORM_S8_UINT = 179, - SVGA3D_DEVCAP_DXFMT_R24_UNORM_X8 = 180, - SVGA3D_DEVCAP_DXFMT_X24_G8_UINT = 181, - SVGA3D_DEVCAP_DXFMT_R8G8_TYPELESS = 182, - SVGA3D_DEVCAP_DXFMT_R8G8_UNORM = 183, - SVGA3D_DEVCAP_DXFMT_R8G8_UINT = 184, - SVGA3D_DEVCAP_DXFMT_R8G8_SINT = 185, - SVGA3D_DEVCAP_DXFMT_R16_TYPELESS = 186, - SVGA3D_DEVCAP_DXFMT_R16_UNORM = 187, - SVGA3D_DEVCAP_DXFMT_R16_UINT = 188, - SVGA3D_DEVCAP_DXFMT_R16_SNORM = 189, - SVGA3D_DEVCAP_DXFMT_R16_SINT = 190, - SVGA3D_DEVCAP_DXFMT_R8_TYPELESS = 191, - SVGA3D_DEVCAP_DXFMT_R8_UNORM = 192, - SVGA3D_DEVCAP_DXFMT_R8_UINT = 193, - SVGA3D_DEVCAP_DXFMT_R8_SNORM = 194, - SVGA3D_DEVCAP_DXFMT_R8_SINT = 195, - SVGA3D_DEVCAP_DXFMT_P8 = 196, - SVGA3D_DEVCAP_DXFMT_R9G9B9E5_SHAREDEXP = 197, - SVGA3D_DEVCAP_DXFMT_R8G8_B8G8_UNORM = 198, - SVGA3D_DEVCAP_DXFMT_G8R8_G8B8_UNORM = 199, - SVGA3D_DEVCAP_DXFMT_BC1_TYPELESS = 200, - SVGA3D_DEVCAP_DXFMT_BC1_UNORM_SRGB = 201, - SVGA3D_DEVCAP_DXFMT_BC2_TYPELESS = 202, - SVGA3D_DEVCAP_DXFMT_BC2_UNORM_SRGB = 203, - SVGA3D_DEVCAP_DXFMT_BC3_TYPELESS = 204, - SVGA3D_DEVCAP_DXFMT_BC3_UNORM_SRGB = 205, - SVGA3D_DEVCAP_DXFMT_BC4_TYPELESS = 206, - SVGA3D_DEVCAP_DXFMT_ATI1 = 207, - SVGA3D_DEVCAP_DXFMT_BC4_SNORM = 208, - SVGA3D_DEVCAP_DXFMT_BC5_TYPELESS = 209, - SVGA3D_DEVCAP_DXFMT_ATI2 = 210, - SVGA3D_DEVCAP_DXFMT_BC5_SNORM = 211, - SVGA3D_DEVCAP_DXFMT_R10G10B10_XR_BIAS_A2_UNORM = 212, - SVGA3D_DEVCAP_DXFMT_B8G8R8A8_TYPELESS = 213, - SVGA3D_DEVCAP_DXFMT_B8G8R8A8_UNORM_SRGB = 214, - SVGA3D_DEVCAP_DXFMT_B8G8R8X8_TYPELESS = 215, - SVGA3D_DEVCAP_DXFMT_B8G8R8X8_UNORM_SRGB = 216, - SVGA3D_DEVCAP_DXFMT_Z_DF16 = 217, - SVGA3D_DEVCAP_DXFMT_Z_DF24 = 218, - SVGA3D_DEVCAP_DXFMT_Z_D24S8_INT = 219, - SVGA3D_DEVCAP_DXFMT_YV12 = 220, - SVGA3D_DEVCAP_DXFMT_R32G32B32A32_FLOAT = 221, - SVGA3D_DEVCAP_DXFMT_R16G16B16A16_FLOAT = 222, - SVGA3D_DEVCAP_DXFMT_R16G16B16A16_UNORM = 223, - SVGA3D_DEVCAP_DXFMT_R32G32_FLOAT = 224, - SVGA3D_DEVCAP_DXFMT_R10G10B10A2_UNORM = 225, - SVGA3D_DEVCAP_DXFMT_R8G8B8A8_SNORM = 226, - SVGA3D_DEVCAP_DXFMT_R16G16_FLOAT = 227, - SVGA3D_DEVCAP_DXFMT_R16G16_UNORM = 228, - SVGA3D_DEVCAP_DXFMT_R16G16_SNORM = 229, - SVGA3D_DEVCAP_DXFMT_R32_FLOAT = 230, - SVGA3D_DEVCAP_DXFMT_R8G8_SNORM = 231, - SVGA3D_DEVCAP_DXFMT_R16_FLOAT = 232, - SVGA3D_DEVCAP_DXFMT_D16_UNORM = 233, - SVGA3D_DEVCAP_DXFMT_A8_UNORM = 234, - SVGA3D_DEVCAP_DXFMT_BC1_UNORM = 235, - SVGA3D_DEVCAP_DXFMT_BC2_UNORM = 236, - SVGA3D_DEVCAP_DXFMT_BC3_UNORM = 237, - SVGA3D_DEVCAP_DXFMT_B5G6R5_UNORM = 238, - SVGA3D_DEVCAP_DXFMT_B5G5R5A1_UNORM = 239, - SVGA3D_DEVCAP_DXFMT_B8G8R8A8_UNORM = 240, - SVGA3D_DEVCAP_DXFMT_B8G8R8X8_UNORM = 241, - SVGA3D_DEVCAP_DXFMT_BC4_UNORM = 242, - SVGA3D_DEVCAP_DXFMT_BC5_UNORM = 243, - - /* - * Advertises shaderModel 4.1 support, independent blend-states, - * cube-map arrays, and a higher vertex input registers limit. - * - * (See documentation on SVGA3D_DEVCAP_DX_MAX_VERTEXBUFFERS.) - */ - SVGA3D_DEVCAP_SM41 = 244, - - SVGA3D_DEVCAP_MULTISAMPLE_2X = 245, - SVGA3D_DEVCAP_MULTISAMPLE_4X = 246, - - SVGA3D_DEVCAP_MAX /* This must be the last index. */ -} SVGA3dDevCapIndex; +typedef uint32 SVGA3dDevCapIndex; + +#define SVGA3D_DEVCAP_INVALID ((uint32)-1) +#define SVGA3D_DEVCAP_3D 0 +#define SVGA3D_DEVCAP_MAX_LIGHTS 1 + +/* + * SVGA3D_DEVCAP_MAX_TEXTURES reflects the maximum number of + * fixed-function texture units available. Each of these units + * work in both FFP and Shader modes, and they support texture + * transforms and texture coordinates. The host may have additional + * texture image units that are only usable with shaders. + */ +#define SVGA3D_DEVCAP_MAX_TEXTURES 2 +#define SVGA3D_DEVCAP_MAX_CLIP_PLANES 3 +#define SVGA3D_DEVCAP_VERTEX_SHADER_VERSION 4 +#define SVGA3D_DEVCAP_VERTEX_SHADER 5 +#define SVGA3D_DEVCAP_FRAGMENT_SHADER_VERSION 6 +#define SVGA3D_DEVCAP_FRAGMENT_SHADER 7 +#define SVGA3D_DEVCAP_MAX_RENDER_TARGETS 8 +#define SVGA3D_DEVCAP_S23E8_TEXTURES 9 +#define SVGA3D_DEVCAP_S10E5_TEXTURES 10 +#define SVGA3D_DEVCAP_MAX_FIXED_VERTEXBLEND 11 +#define SVGA3D_DEVCAP_D16_BUFFER_FORMAT 12 +#define SVGA3D_DEVCAP_D24S8_BUFFER_FORMAT 13 +#define SVGA3D_DEVCAP_D24X8_BUFFER_FORMAT 14 +#define SVGA3D_DEVCAP_QUERY_TYPES 15 +#define SVGA3D_DEVCAP_TEXTURE_GRADIENT_SAMPLING 16 +#define SVGA3D_DEVCAP_MAX_POINT_SIZE 17 +#define SVGA3D_DEVCAP_MAX_SHADER_TEXTURES 18 +#define SVGA3D_DEVCAP_MAX_TEXTURE_WIDTH 19 +#define SVGA3D_DEVCAP_MAX_TEXTURE_HEIGHT 20 +#define SVGA3D_DEVCAP_MAX_VOLUME_EXTENT 21 +#define SVGA3D_DEVCAP_MAX_TEXTURE_REPEAT 22 +#define SVGA3D_DEVCAP_MAX_TEXTURE_ASPECT_RATIO 23 +#define SVGA3D_DEVCAP_MAX_TEXTURE_ANISOTROPY 24 +#define SVGA3D_DEVCAP_MAX_PRIMITIVE_COUNT 25 +#define SVGA3D_DEVCAP_MAX_VERTEX_INDEX 26 +#define SVGA3D_DEVCAP_MAX_VERTEX_SHADER_INSTRUCTIONS 27 +#define SVGA3D_DEVCAP_MAX_FRAGMENT_SHADER_INSTRUCTIONS 28 +#define SVGA3D_DEVCAP_MAX_VERTEX_SHADER_TEMPS 29 +#define SVGA3D_DEVCAP_MAX_FRAGMENT_SHADER_TEMPS 30 +#define SVGA3D_DEVCAP_TEXTURE_OPS 31 +#define SVGA3D_DEVCAP_SURFACEFMT_X8R8G8B8 32 +#define SVGA3D_DEVCAP_SURFACEFMT_A8R8G8B8 33 +#define SVGA3D_DEVCAP_SURFACEFMT_A2R10G10B10 34 +#define SVGA3D_DEVCAP_SURFACEFMT_X1R5G5B5 35 +#define SVGA3D_DEVCAP_SURFACEFMT_A1R5G5B5 36 +#define SVGA3D_DEVCAP_SURFACEFMT_A4R4G4B4 37 +#define SVGA3D_DEVCAP_SURFACEFMT_R5G6B5 38 +#define SVGA3D_DEVCAP_SURFACEFMT_LUMINANCE16 39 +#define SVGA3D_DEVCAP_SURFACEFMT_LUMINANCE8_ALPHA8 40 +#define SVGA3D_DEVCAP_SURFACEFMT_ALPHA8 41 +#define SVGA3D_DEVCAP_SURFACEFMT_LUMINANCE8 42 +#define SVGA3D_DEVCAP_SURFACEFMT_Z_D16 43 +#define SVGA3D_DEVCAP_SURFACEFMT_Z_D24S8 44 +#define SVGA3D_DEVCAP_SURFACEFMT_Z_D24X8 45 +#define SVGA3D_DEVCAP_SURFACEFMT_DXT1 46 +#define SVGA3D_DEVCAP_SURFACEFMT_DXT2 47 +#define SVGA3D_DEVCAP_SURFACEFMT_DXT3 48 +#define SVGA3D_DEVCAP_SURFACEFMT_DXT4 49 +#define SVGA3D_DEVCAP_SURFACEFMT_DXT5 50 +#define SVGA3D_DEVCAP_SURFACEFMT_BUMPX8L8V8U8 51 +#define SVGA3D_DEVCAP_SURFACEFMT_A2W10V10U10 52 +#define SVGA3D_DEVCAP_SURFACEFMT_BUMPU8V8 53 +#define SVGA3D_DEVCAP_SURFACEFMT_Q8W8V8U8 54 +#define SVGA3D_DEVCAP_SURFACEFMT_CxV8U8 55 +#define SVGA3D_DEVCAP_SURFACEFMT_R_S10E5 56 +#define SVGA3D_DEVCAP_SURFACEFMT_R_S23E8 57 +#define SVGA3D_DEVCAP_SURFACEFMT_RG_S10E5 58 +#define SVGA3D_DEVCAP_SURFACEFMT_RG_S23E8 59 +#define SVGA3D_DEVCAP_SURFACEFMT_ARGB_S10E5 60 +#define SVGA3D_DEVCAP_SURFACEFMT_ARGB_S23E8 61 + +/* + * There is a hole in our devcap definitions for + * historical reasons. + * + * Define a constant just for completeness. + */ +#define SVGA3D_DEVCAP_MISSING62 62 + +#define SVGA3D_DEVCAP_MAX_VERTEX_SHADER_TEXTURES 63 + +/* + * Note that MAX_SIMULTANEOUS_RENDER_TARGETS is a maximum count of color + * render targets. This does not include the depth or stencil targets. + */ +#define SVGA3D_DEVCAP_MAX_SIMULTANEOUS_RENDER_TARGETS 64 + +#define SVGA3D_DEVCAP_SURFACEFMT_V16U16 65 +#define SVGA3D_DEVCAP_SURFACEFMT_G16R16 66 +#define SVGA3D_DEVCAP_SURFACEFMT_A16B16G16R16 67 +#define SVGA3D_DEVCAP_SURFACEFMT_UYVY 68 +#define SVGA3D_DEVCAP_SURFACEFMT_YUY2 69 + +/* + * Deprecated. + */ +#define SVGA3D_DEVCAP_DEAD4 70 +#define SVGA3D_DEVCAP_DEAD5 71 +#define SVGA3D_DEVCAP_DEAD7 72 +#define SVGA3D_DEVCAP_DEAD6 73 + +#define SVGA3D_DEVCAP_AUTOGENMIPMAPS 74 +#define SVGA3D_DEVCAP_SURFACEFMT_NV12 75 +#define SVGA3D_DEVCAP_DEAD10 76 + +/* + * This is the maximum number of SVGA context IDs that the guest + * can define using SVGA_3D_CMD_CONTEXT_DEFINE. + */ +#define SVGA3D_DEVCAP_MAX_CONTEXT_IDS 77 + +/* + * This is the maximum number of SVGA surface IDs that the guest + * can define using SVGA_3D_CMD_SURFACE_DEFINE*. + */ +#define SVGA3D_DEVCAP_MAX_SURFACE_IDS 78 + +#define SVGA3D_DEVCAP_SURFACEFMT_Z_DF16 79 +#define SVGA3D_DEVCAP_SURFACEFMT_Z_DF24 80 +#define SVGA3D_DEVCAP_SURFACEFMT_Z_D24S8_INT 81 + +#define SVGA3D_DEVCAP_SURFACEFMT_ATI1 82 +#define SVGA3D_DEVCAP_SURFACEFMT_ATI2 83 + +/* + * Deprecated. + */ +#define SVGA3D_DEVCAP_DEAD1 84 +#define SVGA3D_DEVCAP_DEAD8 85 +#define SVGA3D_DEVCAP_DEAD9 86 + +#define SVGA3D_DEVCAP_LINE_AA 87 /* boolean */ +#define SVGA3D_DEVCAP_LINE_STIPPLE 88 /* boolean */ +#define SVGA3D_DEVCAP_MAX_LINE_WIDTH 89 /* float */ +#define SVGA3D_DEVCAP_MAX_AA_LINE_WIDTH 90 /* float */ + +#define SVGA3D_DEVCAP_SURFACEFMT_YV12 91 + +/* + * Deprecated. + */ +#define SVGA3D_DEVCAP_DEAD3 92 + +/* + * Are TS_CONSTANT, TS_COLOR_KEY, and TS_COLOR_KEY_ENABLE supported? + */ +#define SVGA3D_DEVCAP_TS_COLOR_KEY 93 /* boolean */ + +/* + * Deprecated. + */ +#define SVGA3D_DEVCAP_DEAD2 94 + +/* + * Does the device support DXContexts? + */ +#define SVGA3D_DEVCAP_DXCONTEXT 95 + +/* + * Deprecated. + */ +#define SVGA3D_DEVCAP_DEAD11 96 + +/* + * What is the maximum number of vertex buffers or vertex input registers + * that can be expected to work correctly with a DXContext? + * + * The guest is allowed to set up to SVGA3D_DX_MAX_VERTEXBUFFERS, but + * anything in excess of this cap is not guaranteed to render correctly. + * + * Similarly, the guest can set up to SVGA3D_DX_MAX_VERTEXINPUTREGISTERS + * input registers without the SVGA3D_DEVCAP_SM4_1 cap, or + * SVGA3D_DX_SM41_MAX_VERTEXINPUTREGISTERS with the SVGA3D_DEVCAP_SM4_1, + * but only the registers up to this cap value are guaranteed to render + * correctly. + * + * If guest-drivers are able to expose a lower-limit, it's recommended + * that they clamp to this value. Otherwise, the host will make a + * best-effort on case-by-case basis if guests exceed this. + */ +#define SVGA3D_DEVCAP_DX_MAX_VERTEXBUFFERS 97 + +/* + * What is the maximum number of constant buffers that can be expected to + * work correctly with a DX context? + * + * The guest is allowed to set up to SVGA3D_DX_MAX_CONSTBUFFERS, but + * anything in excess of this cap is not guaranteed to render correctly. + * + * If guest-drivers are able to expose a lower-limit, it's recommended + * that they clamp to this value. Otherwise, the host will make a + * best-effort on case-by-case basis if guests exceed this. + */ +#define SVGA3D_DEVCAP_DX_MAX_CONSTANT_BUFFERS 98 + +/* + * Does the device support provoking vertex control? + * + * If this cap is present, the provokingVertexLast field in the + * rasterizer state is enabled. (Guests can then set it to FALSE, + * meaning that the first vertex is the provoking vertex, or TRUE, + * meaning that the last verteix is the provoking vertex.) + * + * If this cap is FALSE, then guests should set the provokingVertexLast + * to FALSE, otherwise rendering behavior is undefined. + */ +#define SVGA3D_DEVCAP_DX_PROVOKING_VERTEX 99 + +#define SVGA3D_DEVCAP_DXFMT_X8R8G8B8 100 +#define SVGA3D_DEVCAP_DXFMT_A8R8G8B8 101 +#define SVGA3D_DEVCAP_DXFMT_R5G6B5 102 +#define SVGA3D_DEVCAP_DXFMT_X1R5G5B5 103 +#define SVGA3D_DEVCAP_DXFMT_A1R5G5B5 104 +#define SVGA3D_DEVCAP_DXFMT_A4R4G4B4 105 +#define SVGA3D_DEVCAP_DXFMT_Z_D32 106 +#define SVGA3D_DEVCAP_DXFMT_Z_D16 107 +#define SVGA3D_DEVCAP_DXFMT_Z_D24S8 108 +#define SVGA3D_DEVCAP_DXFMT_Z_D15S1 109 +#define SVGA3D_DEVCAP_DXFMT_LUMINANCE8 110 +#define SVGA3D_DEVCAP_DXFMT_LUMINANCE4_ALPHA4 111 +#define SVGA3D_DEVCAP_DXFMT_LUMINANCE16 112 +#define SVGA3D_DEVCAP_DXFMT_LUMINANCE8_ALPHA8 113 +#define SVGA3D_DEVCAP_DXFMT_DXT1 114 +#define SVGA3D_DEVCAP_DXFMT_DXT2 115 +#define SVGA3D_DEVCAP_DXFMT_DXT3 116 +#define SVGA3D_DEVCAP_DXFMT_DXT4 117 +#define SVGA3D_DEVCAP_DXFMT_DXT5 118 +#define SVGA3D_DEVCAP_DXFMT_BUMPU8V8 119 +#define SVGA3D_DEVCAP_DXFMT_BUMPL6V5U5 120 +#define SVGA3D_DEVCAP_DXFMT_BUMPX8L8V8U8 121 +#define SVGA3D_DEVCAP_DXFMT_FORMAT_DEAD1 122 +#define SVGA3D_DEVCAP_DXFMT_ARGB_S10E5 123 +#define SVGA3D_DEVCAP_DXFMT_ARGB_S23E8 124 +#define SVGA3D_DEVCAP_DXFMT_A2R10G10B10 125 +#define SVGA3D_DEVCAP_DXFMT_V8U8 126 +#define SVGA3D_DEVCAP_DXFMT_Q8W8V8U8 127 +#define SVGA3D_DEVCAP_DXFMT_CxV8U8 128 +#define SVGA3D_DEVCAP_DXFMT_X8L8V8U8 129 +#define SVGA3D_DEVCAP_DXFMT_A2W10V10U10 130 +#define SVGA3D_DEVCAP_DXFMT_ALPHA8 131 +#define SVGA3D_DEVCAP_DXFMT_R_S10E5 132 +#define SVGA3D_DEVCAP_DXFMT_R_S23E8 133 +#define SVGA3D_DEVCAP_DXFMT_RG_S10E5 134 +#define SVGA3D_DEVCAP_DXFMT_RG_S23E8 135 +#define SVGA3D_DEVCAP_DXFMT_BUFFER 136 +#define SVGA3D_DEVCAP_DXFMT_Z_D24X8 137 +#define SVGA3D_DEVCAP_DXFMT_V16U16 138 +#define SVGA3D_DEVCAP_DXFMT_G16R16 139 +#define SVGA3D_DEVCAP_DXFMT_A16B16G16R16 140 +#define SVGA3D_DEVCAP_DXFMT_UYVY 141 +#define SVGA3D_DEVCAP_DXFMT_YUY2 142 +#define SVGA3D_DEVCAP_DXFMT_NV12 143 +#define SVGA3D_DEVCAP_DXFMT_FORMAT_DEAD2 144 +#define SVGA3D_DEVCAP_DXFMT_R32G32B32A32_TYPELESS 145 +#define SVGA3D_DEVCAP_DXFMT_R32G32B32A32_UINT 146 +#define SVGA3D_DEVCAP_DXFMT_R32G32B32A32_SINT 147 +#define SVGA3D_DEVCAP_DXFMT_R32G32B32_TYPELESS 148 +#define SVGA3D_DEVCAP_DXFMT_R32G32B32_FLOAT 149 +#define SVGA3D_DEVCAP_DXFMT_R32G32B32_UINT 150 +#define SVGA3D_DEVCAP_DXFMT_R32G32B32_SINT 151 +#define SVGA3D_DEVCAP_DXFMT_R16G16B16A16_TYPELESS 152 +#define SVGA3D_DEVCAP_DXFMT_R16G16B16A16_UINT 153 +#define SVGA3D_DEVCAP_DXFMT_R16G16B16A16_SNORM 154 +#define SVGA3D_DEVCAP_DXFMT_R16G16B16A16_SINT 155 +#define SVGA3D_DEVCAP_DXFMT_R32G32_TYPELESS 156 +#define SVGA3D_DEVCAP_DXFMT_R32G32_UINT 157 +#define SVGA3D_DEVCAP_DXFMT_R32G32_SINT 158 +#define SVGA3D_DEVCAP_DXFMT_R32G8X24_TYPELESS 159 +#define SVGA3D_DEVCAP_DXFMT_D32_FLOAT_S8X24_UINT 160 +#define SVGA3D_DEVCAP_DXFMT_R32_FLOAT_X8X24 161 +#define SVGA3D_DEVCAP_DXFMT_X32_G8X24_UINT 162 +#define SVGA3D_DEVCAP_DXFMT_R10G10B10A2_TYPELESS 163 +#define SVGA3D_DEVCAP_DXFMT_R10G10B10A2_UINT 164 +#define SVGA3D_DEVCAP_DXFMT_R11G11B10_FLOAT 165 +#define SVGA3D_DEVCAP_DXFMT_R8G8B8A8_TYPELESS 166 +#define SVGA3D_DEVCAP_DXFMT_R8G8B8A8_UNORM 167 +#define SVGA3D_DEVCAP_DXFMT_R8G8B8A8_UNORM_SRGB 168 +#define SVGA3D_DEVCAP_DXFMT_R8G8B8A8_UINT 169 +#define SVGA3D_DEVCAP_DXFMT_R8G8B8A8_SINT 170 +#define SVGA3D_DEVCAP_DXFMT_R16G16_TYPELESS 171 +#define SVGA3D_DEVCAP_DXFMT_R16G16_UINT 172 +#define SVGA3D_DEVCAP_DXFMT_R16G16_SINT 173 +#define SVGA3D_DEVCAP_DXFMT_R32_TYPELESS 174 +#define SVGA3D_DEVCAP_DXFMT_D32_FLOAT 175 +#define SVGA3D_DEVCAP_DXFMT_R32_UINT 176 +#define SVGA3D_DEVCAP_DXFMT_R32_SINT 177 +#define SVGA3D_DEVCAP_DXFMT_R24G8_TYPELESS 178 +#define SVGA3D_DEVCAP_DXFMT_D24_UNORM_S8_UINT 179 +#define SVGA3D_DEVCAP_DXFMT_R24_UNORM_X8 180 +#define SVGA3D_DEVCAP_DXFMT_X24_G8_UINT 181 +#define SVGA3D_DEVCAP_DXFMT_R8G8_TYPELESS 182 +#define SVGA3D_DEVCAP_DXFMT_R8G8_UNORM 183 +#define SVGA3D_DEVCAP_DXFMT_R8G8_UINT 184 +#define SVGA3D_DEVCAP_DXFMT_R8G8_SINT 185 +#define SVGA3D_DEVCAP_DXFMT_R16_TYPELESS 186 +#define SVGA3D_DEVCAP_DXFMT_R16_UNORM 187 +#define SVGA3D_DEVCAP_DXFMT_R16_UINT 188 +#define SVGA3D_DEVCAP_DXFMT_R16_SNORM 189 +#define SVGA3D_DEVCAP_DXFMT_R16_SINT 190 +#define SVGA3D_DEVCAP_DXFMT_R8_TYPELESS 191 +#define SVGA3D_DEVCAP_DXFMT_R8_UNORM 192 +#define SVGA3D_DEVCAP_DXFMT_R8_UINT 193 +#define SVGA3D_DEVCAP_DXFMT_R8_SNORM 194 +#define SVGA3D_DEVCAP_DXFMT_R8_SINT 195 +#define SVGA3D_DEVCAP_DXFMT_P8 196 +#define SVGA3D_DEVCAP_DXFMT_R9G9B9E5_SHAREDEXP 197 +#define SVGA3D_DEVCAP_DXFMT_R8G8_B8G8_UNORM 198 +#define SVGA3D_DEVCAP_DXFMT_G8R8_G8B8_UNORM 199 +#define SVGA3D_DEVCAP_DXFMT_BC1_TYPELESS 200 +#define SVGA3D_DEVCAP_DXFMT_BC1_UNORM_SRGB 201 +#define SVGA3D_DEVCAP_DXFMT_BC2_TYPELESS 202 +#define SVGA3D_DEVCAP_DXFMT_BC2_UNORM_SRGB 203 +#define SVGA3D_DEVCAP_DXFMT_BC3_TYPELESS 204 +#define SVGA3D_DEVCAP_DXFMT_BC3_UNORM_SRGB 205 +#define SVGA3D_DEVCAP_DXFMT_BC4_TYPELESS 206 +#define SVGA3D_DEVCAP_DXFMT_ATI1 207 +#define SVGA3D_DEVCAP_DXFMT_BC4_SNORM 208 +#define SVGA3D_DEVCAP_DXFMT_BC5_TYPELESS 209 +#define SVGA3D_DEVCAP_DXFMT_ATI2 210 +#define SVGA3D_DEVCAP_DXFMT_BC5_SNORM 211 +#define SVGA3D_DEVCAP_DXFMT_R10G10B10_XR_BIAS_A2_UNORM 212 +#define SVGA3D_DEVCAP_DXFMT_B8G8R8A8_TYPELESS 213 +#define SVGA3D_DEVCAP_DXFMT_B8G8R8A8_UNORM_SRGB 214 +#define SVGA3D_DEVCAP_DXFMT_B8G8R8X8_TYPELESS 215 +#define SVGA3D_DEVCAP_DXFMT_B8G8R8X8_UNORM_SRGB 216 +#define SVGA3D_DEVCAP_DXFMT_Z_DF16 217 +#define SVGA3D_DEVCAP_DXFMT_Z_DF24 218 +#define SVGA3D_DEVCAP_DXFMT_Z_D24S8_INT 219 +#define SVGA3D_DEVCAP_DXFMT_YV12 220 +#define SVGA3D_DEVCAP_DXFMT_R32G32B32A32_FLOAT 221 +#define SVGA3D_DEVCAP_DXFMT_R16G16B16A16_FLOAT 222 +#define SVGA3D_DEVCAP_DXFMT_R16G16B16A16_UNORM 223 +#define SVGA3D_DEVCAP_DXFMT_R32G32_FLOAT 224 +#define SVGA3D_DEVCAP_DXFMT_R10G10B10A2_UNORM 225 +#define SVGA3D_DEVCAP_DXFMT_R8G8B8A8_SNORM 226 +#define SVGA3D_DEVCAP_DXFMT_R16G16_FLOAT 227 +#define SVGA3D_DEVCAP_DXFMT_R16G16_UNORM 228 +#define SVGA3D_DEVCAP_DXFMT_R16G16_SNORM 229 +#define SVGA3D_DEVCAP_DXFMT_R32_FLOAT 230 +#define SVGA3D_DEVCAP_DXFMT_R8G8_SNORM 231 +#define SVGA3D_DEVCAP_DXFMT_R16_FLOAT 232 +#define SVGA3D_DEVCAP_DXFMT_D16_UNORM 233 +#define SVGA3D_DEVCAP_DXFMT_A8_UNORM 234 +#define SVGA3D_DEVCAP_DXFMT_BC1_UNORM 235 +#define SVGA3D_DEVCAP_DXFMT_BC2_UNORM 236 +#define SVGA3D_DEVCAP_DXFMT_BC3_UNORM 237 +#define SVGA3D_DEVCAP_DXFMT_B5G6R5_UNORM 238 +#define SVGA3D_DEVCAP_DXFMT_B5G5R5A1_UNORM 239 +#define SVGA3D_DEVCAP_DXFMT_B8G8R8A8_UNORM 240 +#define SVGA3D_DEVCAP_DXFMT_B8G8R8X8_UNORM 241 +#define SVGA3D_DEVCAP_DXFMT_BC4_UNORM 242 +#define SVGA3D_DEVCAP_DXFMT_BC5_UNORM 243 + +/* + * Advertises shaderModel 4.1 support, independent blend-states, + * cube-map arrays, and a higher vertex input registers limit. + * + * (See documentation on SVGA3D_DEVCAP_DX_MAX_VERTEXBUFFERS.) + */ +#define SVGA3D_DEVCAP_SM41 244 +#define SVGA3D_DEVCAP_MULTISAMPLE_2X 245 +#define SVGA3D_DEVCAP_MULTISAMPLE_4X 246 + +/* + * Indicates that the device has rendering support for + * the full multisample quality. If this cap is not present, + * the host may or may not support full quality rendering. + * + * See also SVGA_REG_MS_HINT_RESOLVED. + */ +#define SVGA3D_DEVCAP_MS_FULL_QUALITY 247 + +/* + * Advertises support for the SVGA3D LogicOps commands. + */ +#define SVGA3D_DEVCAP_LOGICOPS 248 + +/* + * Advertises support for using logicOps in the DXBlendStates. + */ +#define SVGA3D_DEVCAP_LOGIC_BLENDOPS 249 + +/* +* Note DXFMT range is now non-contiguous. +*/ +#define SVGA3D_DEVCAP_RESERVED_1 250 +#define SVGA3D_DEVCAP_DXFMT_BC6H_TYPELESS 251 +#define SVGA3D_DEVCAP_DXFMT_BC6H_UF16 252 +#define SVGA3D_DEVCAP_DXFMT_BC6H_SF16 253 +#define SVGA3D_DEVCAP_DXFMT_BC7_TYPELESS 254 +#define SVGA3D_DEVCAP_DXFMT_BC7_UNORM 255 +#define SVGA3D_DEVCAP_DXFMT_BC7_UNORM_SRGB 256 +#define SVGA3D_DEVCAP_RESERVED_2 257 + +#define SVGA3D_DEVCAP_SM5 258 +#define SVGA3D_DEVCAP_MULTISAMPLE_8X 259 + +/* This must be the last index. */ +#define SVGA3D_DEVCAP_MAX 260 /* * Bit definitions for DXFMT devcaps @@ -472,10 +501,10 @@ typedef enum { #define SVGA3D_DXFMT_MAX (1 << 10) typedef union { - Bool b; + SVGA3dBool b; uint32 u; - int32 i; - float f; + int32 i; + float f; } SVGA3dDevCapResult; #endif /* _SVGA3D_DEVCAPS_H_ */ diff --git a/drivers/gpu/drm/vmwgfx/device_include/svga3d_dx.h b/drivers/gpu/drm/vmwgfx/device_include/svga3d_dx.h index 7a49c94df221..f703ac2b1768 100644 --- a/drivers/gpu/drm/vmwgfx/device_include/svga3d_dx.h +++ b/drivers/gpu/drm/vmwgfx/device_include/svga3d_dx.h @@ -1,6 +1,6 @@ /* SPDX-License-Identifier: GPL-2.0 OR MIT */ /********************************************************** - * Copyright 2012-2015 VMware, Inc. + * Copyright 2012-2019 VMware, Inc. * * Permission is hereby granted, free of charge, to any person * obtaining a copy of this software and associated documentation @@ -118,12 +118,14 @@ typedef uint8 SVGA3dMultisampleRastEnable; #define SVGA3D_DX_MAX_SRVIEWS 128 #define SVGA3D_DX_MAX_CONSTBUFFERS 16 #define SVGA3D_DX_MAX_SAMPLERS 16 +#define SVGA3D_DX_MAX_CLASS_INSTANCES 253 #define SVGA3D_DX_MAX_CONSTBUF_BINDING_SIZE (4096 * 4 * (uint32)sizeof(uint32)) typedef uint32 SVGA3dShaderResourceViewId; typedef uint32 SVGA3dRenderTargetViewId; typedef uint32 SVGA3dDepthStencilViewId; +typedef uint32 SVGA3dUAViewId; typedef uint32 SVGA3dShaderId; typedef uint32 SVGA3dElementLayoutId; @@ -145,6 +147,17 @@ typedef union { float value[4]; } SVGA3dRGBAFloat; +typedef union { + struct { + uint32 r; + uint32 g; + uint32 b; + uint32 a; + }; + + uint32 value[4]; +} SVGA3dRGBAUint32; + typedef #include "vmware_pack_begin.h" struct { @@ -249,6 +262,39 @@ struct SVGA3dCmdDXSetShader { #include "vmware_pack_end.h" SVGA3dCmdDXSetShader; /* SVGA_3D_CMD_DX_SET_SHADER */ +typedef union { + struct { + uint32 cbOffset : 12; + uint32 cbId : 4; + uint32 baseSamp : 4; + uint32 baseTex : 7; + uint32 reserved : 5; + }; + uint32 value; +} SVGA3dIfaceData; + +typedef +#include "vmware_pack_begin.h" +struct SVGA3dCmdDXSetShaderIface { + SVGA3dShaderType type; + uint32 numClassInstances; + uint32 index; + uint32 iface; + SVGA3dIfaceData data; +} +#include "vmware_pack_end.h" +SVGA3dCmdDXSetShaderIface; /* SVGA_3D_CMD_DX_SET_SHADER_IFACE */ + +typedef +#include "vmware_pack_begin.h" +struct SVGA3dCmdDXBindShaderIface { + uint32 cid; + SVGAMobId mobid; + uint32 offsetInBytes; +} +#include "vmware_pack_end.h" +SVGA3dCmdDXBindShaderIface; /* SVGA_3D_CMD_DX_BIND_SHADER_IFACE */ + typedef #include "vmware_pack_begin.h" struct SVGA3dCmdDXSetSamplers { @@ -306,6 +352,26 @@ SVGA3dCmdDXDrawIndexedInstanced; /* SVGA_3D_CMD_DX_DRAW_INDEXED_INSTANCED */ typedef #include "vmware_pack_begin.h" +struct SVGA3dCmdDXDrawIndexedInstancedIndirect { + SVGA3dSurfaceId argsBufferSid; + uint32 byteOffsetForArgs; +} +#include "vmware_pack_end.h" +SVGA3dCmdDXDrawIndexedInstancedIndirect; +/* SVGA_3D_CMD_DX_DRAW_INDEXED_INSTANCED_INDIRECT */ + +typedef +#include "vmware_pack_begin.h" +struct SVGA3dCmdDXDrawInstancedIndirect { + SVGA3dSurfaceId argsBufferSid; + uint32 byteOffsetForArgs; +} +#include "vmware_pack_end.h" +SVGA3dCmdDXDrawInstancedIndirect; +/* SVGA_3D_CMD_DX_DRAW_INSTANCED_INDIRECT */ + +typedef +#include "vmware_pack_begin.h" struct SVGA3dCmdDXDrawAuto { uint32 pad0; } @@ -314,6 +380,27 @@ SVGA3dCmdDXDrawAuto; /* SVGA_3D_CMD_DX_DRAW_AUTO */ typedef #include "vmware_pack_begin.h" +struct SVGA3dCmdDXDispatch { + uint32 threadGroupCountX; + uint32 threadGroupCountY; + uint32 threadGroupCountZ; +} +#include "vmware_pack_end.h" +SVGA3dCmdDXDispatch; +/* SVGA_3D_CMD_DX_DISPATCH */ + +typedef +#include "vmware_pack_begin.h" +struct SVGA3dCmdDXDispatchIndirect { + SVGA3dSurfaceId argsBufferSid; + uint32 byteOffsetForArgs; +} +#include "vmware_pack_end.h" +SVGA3dCmdDXDispatchIndirect; +/* SVGA_3D_CMD_DX_DISPATCH_INDIRECT */ + +typedef +#include "vmware_pack_begin.h" struct SVGA3dCmdDXSetInputLayout { SVGA3dElementLayoutId elementLayoutId; } @@ -525,7 +612,7 @@ struct MKS3dDXSOState { uint32 offset; /* Starting offset */ uint32 intOffset; /* Internal offset */ uint32 vertexCount; /* vertices written */ - uint32 sizeInBytes; /* max bytes to write */ + uint32 dead; } #include "vmware_pack_end.h" SVGA3dDXSOState; @@ -786,6 +873,31 @@ struct SVGA3dCmdDXTransferFromBuffer { SVGA3dCmdDXTransferFromBuffer; /* SVGA_3D_CMD_DX_TRANSFER_FROM_BUFFER */ +#define SVGA3D_TRANSFER_TO_BUFFER_READBACK (1 << 0) +#define SVGA3D_TRANSFER_TO_BUFFER_FLAGS_MASK (1 << 0) +typedef uint32 SVGA3dTransferToBufferFlags; + +/* + * Raw byte wise transfer to a buffer surface from another surface + * of the requested box. Supported if SVGA_CAP_DX2 is set. This + * command does not take a context. + */ +typedef +#include "vmware_pack_begin.h" +struct SVGA3dCmdDXTransferToBuffer { + SVGA3dSurfaceId srcSid; + uint32 srcSubResource; + SVGA3dBox srcBox; + SVGA3dSurfaceId destSid; + uint32 destOffset; + uint32 destPitch; + uint32 destSlicePitch; + SVGA3dTransferToBufferFlags flags; +} +#include "vmware_pack_end.h" +SVGA3dCmdDXTransferToBuffer; /* SVGA_3D_CMD_DX_TRANSFER_TO_BUFFER */ + + /* * Raw byte wise transfer from a buffer surface into another surface * of the requested box. Supported if SVGA3D_DEVCAP_DXCONTEXT is set. @@ -905,6 +1017,20 @@ typedef SVGA3dCmdDXSetConstantBufferOffset SVGA3dCmdDXSetPSConstantBufferOffset; typedef SVGA3dCmdDXSetConstantBufferOffset SVGA3dCmdDXSetGSConstantBufferOffset; /* SVGA_3D_CMD_DX_SET_GS_CONSTANT_BUFFER_OFFSET */ +typedef SVGA3dCmdDXSetConstantBufferOffset SVGA3dCmdDXSetHSConstantBufferOffset; +/* SVGA_3D_CMD_DX_SET_HS_CONSTANT_BUFFER_OFFSET */ + +typedef SVGA3dCmdDXSetConstantBufferOffset SVGA3dCmdDXSetDSConstantBufferOffset; +/* SVGA_3D_CMD_DX_SET_DS_CONSTANT_BUFFER_OFFSET */ + +typedef SVGA3dCmdDXSetConstantBufferOffset SVGA3dCmdDXSetCSConstantBufferOffset; +/* SVGA_3D_CMD_DX_SET_CS_CONSTANT_BUFFER_OFFSET */ + + +#define SVGA3D_BUFFEREX_SRV_RAW (1 << 0) +#define SVGA3D_BUFFEREX_SRV_FLAGS_MAX (1 << 1) +#define SVGA3D_BUFFEREX_SRV_FLAGS_MASK (SVGA3D_BUFFEREX_SRV_FLAGS_MAX - 1) +typedef uint32 SVGA3dBufferExFlags; typedef #include "vmware_pack_begin.h" @@ -925,7 +1051,7 @@ struct { struct { uint32 firstElement; uint32 numElements; - uint32 flags; + SVGA3dBufferExFlags flags; uint32 pad0; } bufferex; }; @@ -1072,6 +1198,32 @@ struct SVGA3dCmdDXDefineDepthStencilView { SVGA3dCmdDXDefineDepthStencilView; /* SVGA_3D_CMD_DX_DEFINE_DEPTHSTENCIL_VIEW */ +/* + * Version 2 needed in order to start validating and using the flags + * field. Unfortunately the device wasn't validating or using the + * flags field and the driver wasn't initializing it in shipped code, + * so a new version of the command is needed to allow that code to + * continue to work. + */ +typedef +#include "vmware_pack_begin.h" +struct SVGA3dCmdDXDefineDepthStencilView_v2 { + SVGA3dDepthStencilViewId depthStencilViewId; + + SVGA3dSurfaceId sid; + SVGA3dSurfaceFormat format; + SVGA3dResourceType resourceDimension; + uint32 mipSlice; + uint32 firstArraySlice; + uint32 arraySize; + SVGA3DCreateDSViewFlags flags; + uint8 pad0; + uint16 pad1; +} +#include "vmware_pack_end.h" +SVGA3dCmdDXDefineDepthStencilView_v2; +/* SVGA_3D_CMD_DX_DEFINE_DEPTHSTENCIL_VIEW_V2 */ + typedef #include "vmware_pack_begin.h" struct SVGA3dCmdDXDestroyDepthStencilView { @@ -1081,6 +1233,138 @@ struct SVGA3dCmdDXDestroyDepthStencilView { SVGA3dCmdDXDestroyDepthStencilView; /* SVGA_3D_CMD_DX_DESTROY_DEPTHSTENCIL_VIEW */ + +#define SVGA3D_UABUFFER_RAW (1 << 0) +#define SVGA3D_UABUFFER_APPEND (1 << 1) +#define SVGA3D_UABUFFER_COUNTER (1 << 2) +typedef uint32 SVGA3dUABufferFlags; + +typedef +#include "vmware_pack_begin.h" +struct { + union { + struct { + uint32 firstElement; + uint32 numElements; + SVGA3dUABufferFlags flags; + uint32 padding0; + uint32 padding1; + } buffer; + struct { + uint32 mipSlice; + uint32 firstArraySlice; + uint32 arraySize; + uint32 padding0; + uint32 padding1; + } tex; /* 1d, 2d */ + struct { + uint32 mipSlice; + uint32 firstW; + uint32 wSize; + uint32 padding0; + uint32 padding1; + } tex3D; + }; +} +#include "vmware_pack_end.h" +SVGA3dUAViewDesc; + +typedef +#include "vmware_pack_begin.h" +struct { + SVGA3dSurfaceId sid; + SVGA3dSurfaceFormat format; + SVGA3dResourceType resourceDimension; + SVGA3dUAViewDesc desc; + uint32 structureCount; + uint32 pad[7]; +} +#include "vmware_pack_end.h" +SVGACOTableDXUAViewEntry; + +typedef +#include "vmware_pack_begin.h" +struct SVGA3dCmdDXDefineUAView { + SVGA3dUAViewId uaViewId; + + SVGA3dSurfaceId sid; + SVGA3dSurfaceFormat format; + SVGA3dResourceType resourceDimension; + + SVGA3dUAViewDesc desc; +} +#include "vmware_pack_end.h" +SVGA3dCmdDXDefineUAView; +/* SVGA_3D_CMD_DX_DEFINE_UA_VIEW */ + +typedef +#include "vmware_pack_begin.h" +struct SVGA3dCmdDXDestroyUAView { + SVGA3dUAViewId uaViewId; +} +#include "vmware_pack_end.h" +SVGA3dCmdDXDestroyUAView; +/* SVGA_3D_CMD_DX_DESTROY_UA_VIEW */ + +typedef +#include "vmware_pack_begin.h" +struct SVGA3dCmdDXClearUAViewUint { + SVGA3dUAViewId uaViewId; + SVGA3dRGBAUint32 value; +} +#include "vmware_pack_end.h" +SVGA3dCmdDXClearUAViewUint; +/* SVGA_3D_CMD_DX_CLEAR_UA_VIEW_UINT */ + +typedef +#include "vmware_pack_begin.h" +struct SVGA3dCmdDXClearUAViewFloat { + SVGA3dUAViewId uaViewId; + SVGA3dRGBAFloat value; +} +#include "vmware_pack_end.h" +SVGA3dCmdDXClearUAViewFloat; +/* SVGA_3D_CMD_DX_CLEAR_UA_VIEW_FLOAT */ + +typedef +#include "vmware_pack_begin.h" +struct SVGA3dCmdDXCopyStructureCount { + SVGA3dUAViewId srcUAViewId; + SVGA3dSurfaceId destSid; + uint32 destByteOffset; +} +#include "vmware_pack_end.h" +SVGA3dCmdDXCopyStructureCount; +/* SVGA_3D_CMD_DX_COPY_STRUCTURE_COUNT */ + +typedef +#include "vmware_pack_begin.h" +struct SVGA3dCmdDXSetStructureCount { + SVGA3dUAViewId uaViewId; + uint32 structureCount; +} +#include "vmware_pack_end.h" +SVGA3dCmdDXSetStructureCount; +/* SVGA_3D_CMD_DX_SET_STRUCTURE_COUNT */ + +typedef +#include "vmware_pack_begin.h" +struct SVGA3dCmdDXSetUAViews { + uint32 uavSpliceIndex; + /* Followed by a variable number of SVGA3dUAViewId's. */ +} +#include "vmware_pack_end.h" +SVGA3dCmdDXSetUAViews; /* SVGA_3D_CMD_DX_SET_UA_VIEWS */ + +typedef +#include "vmware_pack_begin.h" +struct SVGA3dCmdDXSetCSUAViews { + uint32 startIndex; + /* Followed by a variable number of SVGA3dUAViewId's. */ +} +#include "vmware_pack_end.h" +SVGA3dCmdDXSetCSUAViews; /* SVGA_3D_CMD_DX_SET_CS_UA_VIEWS */ + typedef #include "vmware_pack_begin.h" struct SVGA3dInputElementDesc { @@ -1099,7 +1383,7 @@ typedef struct { uint32 elid; uint32 numDescs; - SVGA3dInputElementDesc desc[32]; + SVGA3dInputElementDesc descs[32]; uint32 pad[62]; } #include "vmware_pack_end.h" @@ -1261,7 +1545,8 @@ struct { uint8 lineStippleEnable; uint8 lineStippleFactor; uint16 lineStipplePattern; - uint32 forcedSampleCount; + uint8 forcedSampleCount; + uint8 mustBeZero[3]; } #include "vmware_pack_end.h" SVGACOTableDXRasterizerStateEntry; @@ -1352,6 +1637,71 @@ struct SVGA3dCmdDXDestroySamplerState { #include "vmware_pack_end.h" SVGA3dCmdDXDestroySamplerState; /* SVGA_3D_CMD_DX_DESTROY_SAMPLER_STATE */ + +#define SVGADX_SIGNATURE_SEMANTIC_NAME_UNDEFINED 0 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_POSITION 1 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_CLIP_DISTANCE 2 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_CULL_DISTANCE 3 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_RENDER_TARGET_ARRAY_INDEX 4 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_VIEWPORT_ARRAY_INDEX 5 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_VERTEX_ID 6 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_PRIMITIVE_ID 7 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_INSTANCE_ID 8 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_IS_FRONT_FACE 9 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_SAMPLE_INDEX 10 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_FINAL_QUAD_U_EQ_0_EDGE_TESSFACTOR 11 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_FINAL_QUAD_V_EQ_0_EDGE_TESSFACTOR 12 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_FINAL_QUAD_U_EQ_1_EDGE_TESSFACTOR 13 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_FINAL_QUAD_V_EQ_1_EDGE_TESSFACTOR 14 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_FINAL_QUAD_U_INSIDE_TESSFACTOR 15 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_FINAL_QUAD_V_INSIDE_TESSFACTOR 16 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_FINAL_TRI_U_EQ_0_EDGE_TESSFACTOR 17 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_FINAL_TRI_V_EQ_0_EDGE_TESSFACTOR 18 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_FINAL_TRI_W_EQ_0_EDGE_TESSFACTOR 19 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_FINAL_TRI_INSIDE_TESSFACTOR 20 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_FINAL_LINE_DETAIL_TESSFACTOR 21 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_FINAL_LINE_DENSITY_TESSFACTOR 22 +#define SVGADX_SIGNATURE_SEMANTIC_NAME_MAX 23 +typedef uint32 SVGA3dDXSignatureSemanticName; + +#define SVGADX_SIGNATURE_REGISTER_COMPONENT_UNKNOWN 0 +typedef uint32 SVGA3dDXSignatureRegisterComponentType; + +#define SVGADX_SIGNATURE_MIN_PRECISION_DEFAULT 0 +typedef uint32 SVGA3dDXSignatureMinPrecision; + +typedef +#include "vmware_pack_begin.h" +struct SVGA3dDXSignatureEntry { + uint32 registerIndex; + SVGA3dDXSignatureSemanticName semanticName; + uint32 mask; /* Lower 4 bits represent X, Y, Z, W channels */ + SVGA3dDXSignatureRegisterComponentType componentType; + SVGA3dDXSignatureMinPrecision minPrecision; +} +#include "vmware_pack_end.h" +SVGA3dDXShaderSignatureEntry; + +#define SVGADX_SIGNATURE_HEADER_VERSION_0 0x08a92d12 + +/* + * The SVGA3dDXSignatureHeader structure is added after the shader + * body in the mob that is bound to the shader. It is followed by the + * specified number of SVGA3dDXSignatureEntry structures for each of + * the three types of signatures in the order (input, output, patch + * constants). + */ +typedef +#include "vmware_pack_begin.h" +struct SVGA3dDXSignatureHeader { + uint32 headerVersion; + uint32 numInputSignatures; + uint32 numOutputSignatures; + uint32 numPatchConstantSignatures; +} +#include "vmware_pack_end.h" +SVGA3dDXShaderSignatureHeader; + typedef #include "vmware_pack_begin.h" struct SVGA3dCmdDXDefineShader { @@ -1415,7 +1765,8 @@ SVGA3dCmdDXCondBindAllShader; /* SVGA_3D_CMD_DX_COND_BIND_ALL_SHADER */ /* * The maximum number of streamout decl's in each streamout entry. */ -#define SVGA3D_MAX_STREAMOUT_DECLS 64 +#define SVGA3D_MAX_DX10_STREAMOUT_DECLS 64 +#define SVGA3D_MAX_STREAMOUT_DECLS 512 typedef #include "vmware_pack_begin.h" @@ -1434,10 +1785,16 @@ typedef #include "vmware_pack_begin.h" struct SVGAOTableStreamOutputEntry { uint32 numOutputStreamEntries; - SVGA3dStreamOutputDeclarationEntry decl[SVGA3D_MAX_STREAMOUT_DECLS]; + SVGA3dStreamOutputDeclarationEntry decl[SVGA3D_MAX_DX10_STREAMOUT_DECLS]; uint32 streamOutputStrideInBytes[SVGA3D_DX_MAX_SOTARGETS]; uint32 rasterizedStream; - uint32 pad[250]; + uint32 numOutputStreamStrides; + uint32 mobid; + uint32 offsetInBytes; + uint8 usesMob; + uint8 pad0; + uint16 pad1; + uint32 pad2[246]; } #include "vmware_pack_end.h" SVGACOTableDXStreamOutputEntry; @@ -1447,13 +1804,47 @@ typedef struct SVGA3dCmdDXDefineStreamOutput { SVGA3dStreamOutputId soid; uint32 numOutputStreamEntries; - SVGA3dStreamOutputDeclarationEntry decl[SVGA3D_MAX_STREAMOUT_DECLS]; + SVGA3dStreamOutputDeclarationEntry decl[SVGA3D_MAX_DX10_STREAMOUT_DECLS]; uint32 streamOutputStrideInBytes[SVGA3D_DX_MAX_SOTARGETS]; uint32 rasterizedStream; } #include "vmware_pack_end.h" SVGA3dCmdDXDefineStreamOutput; /* SVGA_3D_CMD_DX_DEFINE_STREAMOUTPUT */ +/* + * Version 2 needed in order to start validating and using the + * rasterizedStream field. Unfortunately the device wasn't validating + * or using this field and the driver wasn't initializing it in shipped + * code, so a new version of the command is needed to allow that code + * to continue to work. Also added new numOutputStreamStrides field. + */ + +#define SVGA3D_DX_SO_NO_RASTERIZED_STREAM 0xFFFFFFFF + +typedef +#include "vmware_pack_begin.h" +struct SVGA3dCmdDXDefineStreamOutputWithMob { + SVGA3dStreamOutputId soid; + uint32 numOutputStreamEntries; + uint32 numOutputStreamStrides; + uint32 streamOutputStrideInBytes[SVGA3D_DX_MAX_SOTARGETS]; + uint32 rasterizedStream; +} +#include "vmware_pack_end.h" +SVGA3dCmdDXDefineStreamOutputWithMob; +/* SVGA_3D_CMD_DX_DEFINE_STREAMOUTPUT_WITH_MOB */ + +typedef +#include "vmware_pack_begin.h" +struct SVGA3dCmdDXBindStreamOutput { + SVGA3dStreamOutputId soid; + uint32 mobid; + uint32 offsetInBytes; + uint32 sizeInBytes; +} +#include "vmware_pack_end.h" +SVGA3dCmdDXBindStreamOutput; /* SVGA_3D_CMD_DX_BIND_STREAMOUTPUT */ + typedef #include "vmware_pack_begin.h" struct SVGA3dCmdDXDestroyStreamOutput { @@ -1472,6 +1863,15 @@ SVGA3dCmdDXSetStreamOutput; /* SVGA_3D_CMD_DX_SET_STREAMOUTPUT */ typedef #include "vmware_pack_begin.h" +struct SVGA3dCmdDXSetMinLOD { + SVGA3dSurfaceId sid; + float minLOD; +} +#include "vmware_pack_end.h" +SVGA3dCmdDXSetMinLOD; /* SVGA_3D_CMD_DX_SET_MIN_LOD */ + +typedef +#include "vmware_pack_begin.h" struct { uint64 value; uint32 mobId; @@ -1581,33 +1981,38 @@ struct SVGADXContextMobFormat { uint32 rasterizerStateId; uint32 depthStencilViewId; uint32 renderTargetViewIds[SVGA3D_MAX_SIMULTANEOUS_RENDER_TARGETS]; - uint32 unorderedAccessViewIds[SVGA3D_MAX_UAVIEWS]; } renderState; + uint32 pad0[8]; + struct { uint32 targets[SVGA3D_DX_MAX_SOTARGETS]; uint32 soid; } streamOut; - uint32 pad0[11]; + + uint32 pad1[10]; + + uint32 uavSpliceIndex; uint8 numViewports; uint8 numScissorRects; - uint16 pad1[1]; + uint16 pad2[1]; - uint32 pad2[3]; + uint32 pad3[3]; SVGA3dViewport viewports[SVGA3D_DX_MAX_VIEWPORTS]; - uint32 pad3[32]; + uint32 pad4[32]; SVGASignedRect scissorRects[SVGA3D_DX_MAX_SCISSORRECTS]; - uint32 pad4[64]; + uint32 pad5[64]; struct { uint32 queryID; uint32 value; } predication; - uint32 pad5[2]; + SVGAMobId shaderIfaceMobid; + uint32 shaderIfaceOffset; struct { uint32 shaderId; SVGA3dConstantBufferBinding constantBuffers[SVGA3D_DX_MAX_CONSTBUFFERS]; @@ -1619,11 +2024,38 @@ struct SVGADXContextMobFormat { SVGA3dQueryId queryID[SVGA3D_MAX_QUERY]; SVGA3dCOTableData cotables[SVGA_COTABLE_MAX]; - uint32 pad7[380]; + + uint32 pad7[64]; + + uint32 uaViewIds[SVGA3D_DX11_1_MAX_UAVIEWS]; + uint32 csuaViewIds[SVGA3D_DX11_1_MAX_UAVIEWS]; + + uint32 pad8[188]; } #include "vmware_pack_end.h" SVGADXContextMobFormat; +/* + * There is conflicting documentation on max class instances (253 vs 256). The + * lower value is the one used throughout the device, but since mob format is + * more involved to increase if needed, conservatively use the higher one here. + */ +#define SVGA3D_DX_MAX_CLASS_INSTANCES_PADDED 256 + +typedef +#include "vmware_pack_begin.h" +struct SVGADXShaderIfaceMobFormat { + struct { + uint32 numClassInstances; + uint32 iface[SVGA3D_DX_MAX_CLASS_INSTANCES_PADDED]; + SVGA3dIfaceData data[SVGA3D_DX_MAX_CLASS_INSTANCES_PADDED]; + } shaderIfaceState[SVGA3D_NUM_SHADERTYPE]; + + uint32 pad0[1018]; +} +#include "vmware_pack_end.h" +SVGADXShaderIfaceMobFormat; + typedef #include "vmware_pack_begin.h" struct SVGA3dCmdDXTempSetContext { diff --git a/drivers/gpu/drm/vmwgfx/device_include/svga3d_limits.h b/drivers/gpu/drm/vmwgfx/device_include/svga3d_limits.h index b22a67f15660..f4375a41b3aa 100644 --- a/drivers/gpu/drm/vmwgfx/device_include/svga3d_limits.h +++ b/drivers/gpu/drm/vmwgfx/device_include/svga3d_limits.h @@ -1,6 +1,6 @@ /* SPDX-License-Identifier: GPL-2.0 OR MIT */ /********************************************************** - * Copyright 2007-2015 VMware, Inc. + * Copyright 2007-2019 VMware, Inc. * * Permission is hereby granted, free of charge, to any person * obtaining a copy of this software and associated documentation @@ -40,11 +40,25 @@ #include "includeCheck.h" #define SVGA3D_NUM_CLIPPLANES 6 +#define SVGA3D_MAX_CONTEXT_IDS 256 +#define SVGA3D_MAX_SURFACE_IDS (32 * 1024) + +/* + * While there are separate bind-points for RenderTargetViews and + * UnorderedAccessViews in a DXContext, there is in fact one shared + * semantic space that the guest-driver can use on any given draw call. + * So there are really only 8 slots that can be spilt up between them, with the + * spliceIndex controlling where the UAV's sit in the collapsed array. + */ #define SVGA3D_MAX_RENDER_TARGETS 8 #define SVGA3D_MAX_SIMULTANEOUS_RENDER_TARGETS (SVGA3D_MAX_RENDER_TARGETS) #define SVGA3D_MAX_UAVIEWS 8 -#define SVGA3D_MAX_CONTEXT_IDS 256 -#define SVGA3D_MAX_SURFACE_IDS (32 * 1024) +#define SVGA3D_DX11_1_MAX_UAVIEWS 64 + +/* + * Maximum canonical size of a surface in host-backed mode (pre-GBObjects). + */ +#define SVGA3D_HB_MAX_SURFACE_SIZE MBYTES_2_BYTES(128) /* * Maximum ID a shader can be assigned on a given context. @@ -59,6 +73,8 @@ #define SVGA3D_NUM_TEXTURE_UNITS 32 #define SVGA3D_NUM_LIGHTS 8 +#define SVGA3D_MAX_VIDEOPROCESSOR_SAMPLERS 32 + /* * Maximum size in dwords of shader text the SVGA device will allow. * Currently 8 MB. @@ -67,6 +83,11 @@ #define SVGA3D_MAX_SHADER_MEMORY (SVGA3D_MAX_SHADER_MEMORY_BYTES / \ sizeof(uint32)) +/* + * The maximum value of threadGroupCount in each dimension + */ +#define SVGA3D_MAX_SHADER_THREAD_GROUPS 65535 + #define SVGA3D_MAX_CLIP_PLANES 6 /* @@ -85,7 +106,9 @@ /* * Maximum number of array indexes in a GB surface (with DX enabled). */ -#define SVGA3D_MAX_SURFACE_ARRAYSIZE 512 +#define SVGA3D_SM4_MAX_SURFACE_ARRAYSIZE 512 +#define SVGA3D_SM5_MAX_SURFACE_ARRAYSIZE 2048 +#define SVGA3D_MAX_SURFACE_ARRAYSIZE SVGA3D_SM5_MAX_SURFACE_ARRAYSIZE /* * The maximum number of vertex arrays we're guaranteed to support in @@ -99,4 +122,9 @@ */ #define SVGA3D_MAX_DRAW_PRIMITIVE_RANGES 32 +/* + * The maximum number of samples that can be contained in a surface. + */ +#define SVGA3D_MAX_SAMPLES 8 + #endif /* _SVGA3D_LIMITS_H_ */ diff --git a/drivers/gpu/drm/vmwgfx/device_include/svga3d_surfacedefs.h b/drivers/gpu/drm/vmwgfx/device_include/svga3d_surfacedefs.h index 61414f105c67..4db25bd9fa22 100644 --- a/drivers/gpu/drm/vmwgfx/device_include/svga3d_surfacedefs.h +++ b/drivers/gpu/drm/vmwgfx/device_include/svga3d_surfacedefs.h @@ -131,6 +131,8 @@ enum svga3d_block_desc { SVGA3DBLOCKDESC_BC3 = 1 << 26, SVGA3DBLOCKDESC_BC4 = 1 << 27, SVGA3DBLOCKDESC_BC5 = 1 << 28, + SVGA3DBLOCKDESC_BC6H = 1 << 29, + SVGA3DBLOCKDESC_BC7 = 1 << 30, SVGA3DBLOCKDESC_A_UINT = SVGA3DBLOCKDESC_ALPHA | SVGA3DBLOCKDESC_UINT | @@ -290,6 +292,18 @@ enum svga3d_block_desc { SVGA3DBLOCKDESC_COMP_UNORM, SVGA3DBLOCKDESC_BC5_COMP_SNORM = SVGA3DBLOCKDESC_BC5 | SVGA3DBLOCKDESC_COMP_SNORM, + SVGA3DBLOCKDESC_BC6H_COMP_TYPELESS = SVGA3DBLOCKDESC_BC6H | + SVGA3DBLOCKDESC_COMP_TYPELESS, + SVGA3DBLOCKDESC_BC6H_COMP_UF16 = SVGA3DBLOCKDESC_BC6H | + SVGA3DBLOCKDESC_COMPRESSED, + SVGA3DBLOCKDESC_BC6H_COMP_SF16 = SVGA3DBLOCKDESC_BC6H | + SVGA3DBLOCKDESC_COMPRESSED, + SVGA3DBLOCKDESC_BC7_COMP_TYPELESS = SVGA3DBLOCKDESC_BC7 | + SVGA3DBLOCKDESC_COMP_TYPELESS, + SVGA3DBLOCKDESC_BC7_COMP_UNORM = SVGA3DBLOCKDESC_BC7 | + SVGA3DBLOCKDESC_COMP_UNORM, + SVGA3DBLOCKDESC_BC7_COMP_UNORM_SRGB = SVGA3DBLOCKDESC_BC7_COMP_UNORM | + SVGA3DBLOCKDESC_SRGB, SVGA3DBLOCKDESC_NV12 = SVGA3DBLOCKDESC_YUV_VIDEO | SVGA3DBLOCKDESC_PLANAR_YUV | @@ -494,7 +508,7 @@ static const struct svga3d_surface_desc svga3d_surface_descs[] = { {{8}, {8}, {8}, {0}}, {{16}, {8}, {0}, {0}}}, - {SVGA3D_FORMAT_DEAD1, SVGA3DBLOCKDESC_UVL, + {SVGA3D_FORMAT_DEAD1, SVGA3DBLOCKDESC_NONE, {1, 1, 1}, 3, 3, {{8}, {8}, {8}, {0}}, {{16}, {8}, {0}, {0}}}, @@ -604,7 +618,7 @@ static const struct svga3d_surface_desc svga3d_surface_descs[] = { {{0}, {0}, {48}, {0}}, {{0}, {0}, {0}, {0}}}, - {SVGA3D_AYUV, SVGA3DBLOCKDESC_AYUV, + {SVGA3D_FORMAT_DEAD2, SVGA3DBLOCKDESC_NONE, {1, 1, 1}, 4, 4, {{8}, {8}, {8}, {8}}, {{0}, {8}, {16}, {24}}}, @@ -1103,6 +1117,46 @@ static const struct svga3d_surface_desc svga3d_surface_descs[] = { {4, 4, 1}, 16, 16, {{0}, {0}, {128}, {0}}, {{0}, {0}, {0}, {0}}}, + + {SVGA3D_B4G4R4A4_UNORM, SVGA3DBLOCKDESC_RGBA_UNORM, + {1, 1, 1}, 2, 2, + {{4}, {4}, {4}, {4}}, + {{0}, {4}, {8}, {12}}}, + + {SVGA3D_BC6H_TYPELESS, SVGA3DBLOCKDESC_BC6H_COMP_TYPELESS, + {4, 4, 1}, 16, 16, + {{0}, {0}, {128}, {0}}, + {{0}, {0}, {0}, {0}}}, + + {SVGA3D_BC6H_UF16, SVGA3DBLOCKDESC_BC6H_COMP_UF16, + {4, 4, 1}, 16, 16, + {{0}, {0}, {128}, {0}}, + {{0}, {0}, {0}, {0}}}, + + {SVGA3D_BC6H_SF16, SVGA3DBLOCKDESC_BC6H_COMP_SF16, + {4, 4, 1}, 16, 16, + {{0}, {0}, {128}, {0}}, + {{0}, {0}, {0}, {0}}}, + + {SVGA3D_BC7_TYPELESS, SVGA3DBLOCKDESC_BC7_COMP_TYPELESS, + {4, 4, 1}, 16, 16, + {{0}, {0}, {128}, {0}}, + {{0}, {0}, {0}, {0}}}, + + {SVGA3D_BC7_UNORM, SVGA3DBLOCKDESC_BC7_COMP_UNORM, + {4, 4, 1}, 16, 16, + {{0}, {0}, {128}, {0}}, + {{0}, {0}, {0}, {0}}}, + + {SVGA3D_BC7_UNORM_SRGB, SVGA3DBLOCKDESC_BC7_COMP_UNORM_SRGB, + {4, 4, 1}, 16, 16, + {{0}, {0}, {128}, {0}}, + {{0}, {0}, {0}, {0}}}, + + {SVGA3D_AYUV, SVGA3DBLOCKDESC_AYUV, + {1, 1, 1}, 4, 4, + {{8}, {8}, {8}, {8}}, + {{0}, {8}, {16}, {24}}}, }; static inline u32 clamped_umul32(u32 a, u32 b) diff --git a/drivers/gpu/drm/vmwgfx/device_include/svga3d_types.h b/drivers/gpu/drm/vmwgfx/device_include/svga3d_types.h index 308370665a8e..77e338a65791 100644 --- a/drivers/gpu/drm/vmwgfx/device_include/svga3d_types.h +++ b/drivers/gpu/drm/vmwgfx/device_include/svga3d_types.h @@ -116,6 +116,19 @@ SVGA3dBox; typedef #include "vmware_pack_begin.h" struct { + int32 x; + int32 y; + int32 z; + int32 w; + int32 h; + int32 d; +} +#include "vmware_pack_end.h" +SVGA3dSignedBox; + +typedef +#include "vmware_pack_begin.h" +struct { uint32 x; uint32 y; uint32 z; @@ -198,8 +211,7 @@ typedef enum SVGA3dSurfaceFormat { /* Planar video formats */ SVGA3D_NV12 = 44, - /* Video format with alpha */ - SVGA3D_AYUV = 45, + SVGA3D_FORMAT_DEAD2 = 45, SVGA3D_R32G32B32A32_TYPELESS = 46, SVGA3D_R32G32B32A32_UINT = 47, @@ -305,6 +317,18 @@ typedef enum SVGA3dSurfaceFormat { SVGA3D_B8G8R8X8_UNORM = 142, SVGA3D_BC4_UNORM = 143, SVGA3D_BC5_UNORM = 144, + SVGA3D_B4G4R4A4_UNORM = 145, + + /* DX11 compressed formats */ + SVGA3D_BC6H_TYPELESS = 146, + SVGA3D_BC6H_UF16 = 147, + SVGA3D_BC6H_SF16 = 148, + SVGA3D_BC7_TYPELESS = 149, + SVGA3D_BC7_UNORM = 150, + SVGA3D_BC7_UNORM_SRGB = 151, + + /* Video format with alpha */ + SVGA3D_AYUV = 152, SVGA3D_FORMAT_MAX } SVGA3dSurfaceFormat; @@ -326,10 +350,10 @@ typedef enum SVGA3dSurfaceFormat { #define SVGA3D_SURFACE_HINT_RENDERTARGET (CONST64U(1) << 6) #define SVGA3D_SURFACE_HINT_DEPTHSTENCIL (CONST64U(1) << 7) #define SVGA3D_SURFACE_HINT_WRITEONLY (CONST64U(1) << 8) -#define SVGA3D_SURFACE_MASKABLE_ANTIALIAS (CONST64U(1) << 9) +#define SVGA3D_SURFACE_DEAD2 (CONST64U(1) << 9) #define SVGA3D_SURFACE_AUTOGENMIPMAPS (CONST64U(1) << 10) -#define SVGA3D_SURFACE_DECODE_RENDERTARGET (CONST64U(1) << 11) +#define SVGA3D_SURFACE_DEAD1 (CONST64U(1) << 11) /* * Is this surface using a base-level pitch for it's mob backing? @@ -387,7 +411,7 @@ typedef enum SVGA3dSurfaceFormat { * Setting this flag allow this surface to be used with the * SVGA_3D_CMD_DX_TRANSFER_FROM_BUFFER command. It is only valid for * buffer surfaces, and no bind flags are allowed to be set on surfaces - * with this flag. + * with this flag except SVGA3D_SURFACE_TRANSFER_TO_BUFFER. */ #define SVGA3D_SURFACE_TRANSFER_FROM_BUFFER (CONST64U(1) << 30) @@ -402,7 +426,31 @@ typedef enum SVGA3dSurfaceFormat { */ #define SVGA3D_SURFACE_MULTISAMPLE (CONST64U(1) << 32) -#define SVGA3D_SURFACE_FLAG_MAX (CONST64U(1) << 33) +/* + * Specified that the surface is allowed to be bound to a UAView. + */ +#define SVGA3D_SURFACE_BIND_UAVIEW (CONST64U(1) << 33) + +/* + * Setting this flag allow this surface to be used with the + * SVGA_3D_CMD_DX_TRANSFER_TO_BUFFER command. It is only valid for + * buffer surfaces, and no bind flags are allowed to be set on surfaces + * with this flag except SVGA3D_SURFACE_TRANSFER_FROM_BUFFER. + */ +#define SVGA3D_SURFACE_TRANSFER_TO_BUFFER (CONST64U(1) << 34) + +#define SVGA3D_SURFACE_BIND_LOGICOPS (CONST64U(1) << 35) + +/* + * Optional flags for use with SVGA3D_SURFACE_BIND_UAVIEW + */ +#define SVGA3D_SURFACE_BIND_RAW_VIEWS (CONST64U(1) << 36) +#define SVGA3D_SURFACE_BUFFER_STRUCTURED (CONST64U(1) << 37) + +#define SVGA3D_SURFACE_DRAWINDIRECT_ARGS (CONST64U(1) << 38) +#define SVGA3D_SURFACE_RESOURCE_CLAMP (CONST64U(1) << 39) + +#define SVGA3D_SURFACE_FLAG_MAX (CONST64U(1) << 40) /* * Surface flags types: @@ -428,17 +476,25 @@ typedef uint64 SVGA3dSurfaceAllFlags; SVGA3D_SURFACE_STAGING_DOWNLOAD | \ SVGA3D_SURFACE_HINT_INDIRECT_UPDATE | \ SVGA3D_SURFACE_TRANSFER_FROM_BUFFER | \ - SVGA3D_SURFACE_MULTISAMPLE \ + SVGA3D_SURFACE_RESERVED1 | \ + SVGA3D_SURFACE_MULTISAMPLE | \ + SVGA3D_SURFACE_BIND_UAVIEW | \ + SVGA3D_SURFACE_TRANSFER_TO_BUFFER | \ + SVGA3D_SURFACE_BIND_LOGICOPS | \ + SVGA3D_SURFACE_BIND_RAW_VIEWS | \ + SVGA3D_SURFACE_BUFFER_STRUCTURED | \ + SVGA3D_SURFACE_DRAWINDIRECT_ARGS | \ + SVGA3D_SURFACE_RESOURCE_CLAMP \ ) #define SVGA3D_SURFACE_HB_PRESENT_DISALLOWED_MASK \ ( SVGA3D_SURFACE_1D | \ + SVGA3D_SURFACE_RESERVED1 | \ SVGA3D_SURFACE_MULTISAMPLE \ ) #define SVGA3D_SURFACE_2D_DISALLOWED_MASK \ ( SVGA3D_SURFACE_CUBEMAP | \ - SVGA3D_SURFACE_MASKABLE_ANTIALIAS | \ SVGA3D_SURFACE_AUTOGENMIPMAPS | \ SVGA3D_SURFACE_VOLUME | \ SVGA3D_SURFACE_1D | \ @@ -448,7 +504,14 @@ typedef uint64 SVGA3dSurfaceAllFlags; SVGA3D_SURFACE_BIND_DEPTH_STENCIL | \ SVGA3D_SURFACE_BIND_STREAM_OUTPUT | \ SVGA3D_SURFACE_TRANSFER_FROM_BUFFER | \ - SVGA3D_SURFACE_MULTISAMPLE \ + SVGA3D_SURFACE_RESERVED1 | \ + SVGA3D_SURFACE_MULTISAMPLE | \ + SVGA3D_SURFACE_BIND_UAVIEW | \ + SVGA3D_SURFACE_TRANSFER_TO_BUFFER | \ + SVGA3D_SURFACE_BIND_RAW_VIEWS | \ + SVGA3D_SURFACE_BUFFER_STRUCTURED | \ + SVGA3D_SURFACE_DRAWINDIRECT_ARGS | \ + SVGA3D_SURFACE_RESOURCE_CLAMP \ ) #define SVGA3D_SURFACE_BASICOPS_DISALLOWED_MASK \ @@ -456,6 +519,7 @@ typedef uint64 SVGA3dSurfaceAllFlags; SVGA3D_SURFACE_AUTOGENMIPMAPS | \ SVGA3D_SURFACE_VOLUME | \ SVGA3D_SURFACE_1D | \ + SVGA3D_SURFACE_RESERVED1 | \ SVGA3D_SURFACE_MULTISAMPLE \ ) @@ -474,7 +538,14 @@ typedef uint64 SVGA3dSurfaceAllFlags; SVGA3D_SURFACE_STAGING_DOWNLOAD | \ SVGA3D_SURFACE_HINT_INDIRECT_UPDATE | \ SVGA3D_SURFACE_TRANSFER_FROM_BUFFER | \ - SVGA3D_SURFACE_MULTISAMPLE \ + SVGA3D_SURFACE_RESERVED1 | \ + SVGA3D_SURFACE_MULTISAMPLE | \ + SVGA3D_SURFACE_BIND_UAVIEW | \ + SVGA3D_SURFACE_TRANSFER_TO_BUFFER | \ + SVGA3D_SURFACE_BIND_RAW_VIEWS | \ + SVGA3D_SURFACE_BUFFER_STRUCTURED | \ + SVGA3D_SURFACE_DRAWINDIRECT_ARGS | \ + SVGA3D_SURFACE_RESOURCE_CLAMP \ ) #define SVGA3D_SURFACE_BUFFER_DISALLOWED_MASK \ @@ -482,10 +553,11 @@ typedef uint64 SVGA3dSurfaceAllFlags; SVGA3D_SURFACE_AUTOGENMIPMAPS | \ SVGA3D_SURFACE_VOLUME | \ SVGA3D_SURFACE_1D | \ - SVGA3D_SURFACE_MASKABLE_ANTIALIAS | \ + SVGA3D_SURFACE_DEAD2 | \ SVGA3D_SURFACE_ARRAY | \ SVGA3D_SURFACE_MULTISAMPLE | \ - SVGA3D_SURFACE_MOB_PITCH \ + SVGA3D_SURFACE_MOB_PITCH | \ + SVGA3D_SURFACE_RESOURCE_CLAMP \ ) #define SVGA3D_SURFACE_MULTISAMPLE_DISALLOWED_MASK \ @@ -494,14 +566,23 @@ typedef uint64 SVGA3dSurfaceAllFlags; SVGA3D_SURFACE_VOLUME | \ SVGA3D_SURFACE_1D | \ SVGA3D_SURFACE_SCREENTARGET | \ - SVGA3D_SURFACE_MOB_PITCH \ + SVGA3D_SURFACE_MOB_PITCH | \ + SVGA3D_SURFACE_TRANSFER_FROM_BUFFER | \ + SVGA3D_SURFACE_RESERVED1 | \ + SVGA3D_SURFACE_BIND_UAVIEW | \ + SVGA3D_SURFACE_TRANSFER_TO_BUFFER | \ + SVGA3D_SURFACE_BIND_LOGICOPS | \ + SVGA3D_SURFACE_BIND_RAW_VIEWS | \ + SVGA3D_SURFACE_BUFFER_STRUCTURED | \ + SVGA3D_SURFACE_DRAWINDIRECT_ARGS \ ) -#define SVGA3D_SURFACE_DX_ONLY_MASK \ - ( SVGA3D_SURFACE_BIND_STREAM_OUTPUT | \ - SVGA3D_SURFACE_STAGING_UPLOAD | \ - SVGA3D_SURFACE_STAGING_DOWNLOAD | \ - SVGA3D_SURFACE_TRANSFER_FROM_BUFFER \ +#define SVGA3D_SURFACE_DX_ONLY_MASK \ + ( SVGA3D_SURFACE_BIND_STREAM_OUTPUT | \ + SVGA3D_SURFACE_STAGING_UPLOAD | \ + SVGA3D_SURFACE_STAGING_DOWNLOAD | \ + SVGA3D_SURFACE_TRANSFER_FROM_BUFFER | \ + SVGA3D_SURFACE_TRANSFER_TO_BUFFER \ ) #define SVGA3D_SURFACE_STAGING_MASK \ @@ -516,9 +597,135 @@ typedef uint64 SVGA3dSurfaceAllFlags; SVGA3D_SURFACE_BIND_SHADER_RESOURCE | \ SVGA3D_SURFACE_BIND_RENDER_TARGET | \ SVGA3D_SURFACE_BIND_DEPTH_STENCIL | \ - SVGA3D_SURFACE_BIND_STREAM_OUTPUT \ + SVGA3D_SURFACE_BIND_STREAM_OUTPUT | \ + SVGA3D_SURFACE_BIND_UAVIEW | \ + SVGA3D_SURFACE_BIND_LOGICOPS | \ + SVGA3D_SURFACE_BIND_RAW_VIEWS \ + ) + +#define SVGA3D_SURFACE_VADECODE_DISALLOWED_MASK \ + ( SVGA3D_SURFACE_CUBEMAP | \ + SVGA3D_SURFACE_HINT_STATIC | \ + SVGA3D_SURFACE_HINT_DYNAMIC | \ + SVGA3D_SURFACE_HINT_INDEXBUFFER | \ + SVGA3D_SURFACE_HINT_VERTEXBUFFER | \ + SVGA3D_SURFACE_HINT_TEXTURE | \ + SVGA3D_SURFACE_HINT_RENDERTARGET | \ + SVGA3D_SURFACE_HINT_DEPTHSTENCIL | \ + SVGA3D_SURFACE_HINT_WRITEONLY | \ + SVGA3D_SURFACE_DEAD2 | \ + SVGA3D_SURFACE_AUTOGENMIPMAPS | \ + SVGA3D_SURFACE_HINT_RT_LOCKABLE | \ + SVGA3D_SURFACE_VOLUME | \ + SVGA3D_SURFACE_SCREENTARGET | \ + SVGA3D_SURFACE_1D | \ + SVGA3D_SURFACE_BIND_VERTEX_BUFFER | \ + SVGA3D_SURFACE_BIND_INDEX_BUFFER | \ + SVGA3D_SURFACE_BIND_CONSTANT_BUFFER | \ + SVGA3D_SURFACE_BIND_RENDER_TARGET | \ + SVGA3D_SURFACE_BIND_SHADER_RESOURCE | \ + SVGA3D_SURFACE_BIND_DEPTH_STENCIL | \ + SVGA3D_SURFACE_BIND_STREAM_OUTPUT | \ + SVGA3D_SURFACE_INACTIVE | \ + SVGA3D_SURFACE_STAGING_UPLOAD | \ + SVGA3D_SURFACE_STAGING_DOWNLOAD | \ + SVGA3D_SURFACE_HINT_INDIRECT_UPDATE | \ + SVGA3D_SURFACE_TRANSFER_FROM_BUFFER | \ + SVGA3D_SURFACE_MULTISAMPLE | \ + SVGA3D_SURFACE_BIND_UAVIEW | \ + SVGA3D_SURFACE_TRANSFER_TO_BUFFER | \ + SVGA3D_SURFACE_BIND_LOGICOPS | \ + SVGA3D_SURFACE_BIND_RAW_VIEWS | \ + SVGA3D_SURFACE_BUFFER_STRUCTURED | \ + SVGA3D_SURFACE_DRAWINDIRECT_ARGS | \ + SVGA3D_SURFACE_RESOURCE_CLAMP \ + ) + +#define SVGA3D_SURFACE_VAPROCESSFRAME_OUTPUT_DISALLOWED_MASK \ + ( SVGA3D_SURFACE_HINT_INDEXBUFFER | \ + SVGA3D_SURFACE_HINT_VERTEXBUFFER | \ + SVGA3D_SURFACE_HINT_DEPTHSTENCIL | \ + SVGA3D_SURFACE_DEAD2 | \ + SVGA3D_SURFACE_VOLUME | \ + SVGA3D_SURFACE_1D | \ + SVGA3D_SURFACE_BIND_VERTEX_BUFFER | \ + SVGA3D_SURFACE_BIND_INDEX_BUFFER | \ + SVGA3D_SURFACE_BIND_CONSTANT_BUFFER | \ + SVGA3D_SURFACE_BIND_DEPTH_STENCIL | \ + SVGA3D_SURFACE_BIND_STREAM_OUTPUT | \ + SVGA3D_SURFACE_INACTIVE | \ + SVGA3D_SURFACE_STAGING_UPLOAD | \ + SVGA3D_SURFACE_STAGING_DOWNLOAD | \ + SVGA3D_SURFACE_TRANSFER_FROM_BUFFER | \ + SVGA3D_SURFACE_VADECODE | \ + SVGA3D_SURFACE_MULTISAMPLE | \ + SVGA3D_SURFACE_BIND_UAVIEW | \ + SVGA3D_SURFACE_TRANSFER_TO_BUFFER | \ + SVGA3D_SURFACE_BIND_LOGICOPS | \ + SVGA3D_SURFACE_BIND_RAW_VIEWS | \ + SVGA3D_SURFACE_BUFFER_STRUCTURED | \ + SVGA3D_SURFACE_DRAWINDIRECT_ARGS | \ + SVGA3D_SURFACE_RESOURCE_CLAMP \ + ) + +#define SVGA3D_SURFACE_VAPROCESSFRAME_INPUT_DISALLOWED_MASK \ + ( SVGA3D_SURFACE_CUBEMAP | \ + SVGA3D_SURFACE_HINT_INDEXBUFFER | \ + SVGA3D_SURFACE_HINT_VERTEXBUFFER | \ + SVGA3D_SURFACE_HINT_DEPTHSTENCIL | \ + SVGA3D_SURFACE_DEAD2 | \ + SVGA3D_SURFACE_VOLUME | \ + SVGA3D_SURFACE_SCREENTARGET | \ + SVGA3D_SURFACE_1D | \ + SVGA3D_SURFACE_BIND_VERTEX_BUFFER | \ + SVGA3D_SURFACE_BIND_INDEX_BUFFER | \ + SVGA3D_SURFACE_BIND_CONSTANT_BUFFER | \ + SVGA3D_SURFACE_BIND_DEPTH_STENCIL | \ + SVGA3D_SURFACE_BIND_STREAM_OUTPUT | \ + SVGA3D_SURFACE_STAGING_UPLOAD | \ + SVGA3D_SURFACE_STAGING_DOWNLOAD | \ + SVGA3D_SURFACE_TRANSFER_FROM_BUFFER | \ + SVGA3D_SURFACE_MULTISAMPLE | \ + SVGA3D_SURFACE_BIND_UAVIEW | \ + SVGA3D_SURFACE_TRANSFER_TO_BUFFER | \ + SVGA3D_SURFACE_BIND_LOGICOPS | \ + SVGA3D_SURFACE_BIND_RAW_VIEWS | \ + SVGA3D_SURFACE_BUFFER_STRUCTURED | \ + SVGA3D_SURFACE_DRAWINDIRECT_ARGS | \ + SVGA3D_SURFACE_RESOURCE_CLAMP \ + ) + +#define SVGA3D_SURFACE_LOGICOPS_DISALLOWED_MASK \ + ( SVGA3D_SURFACE_CUBEMAP | \ + SVGA3D_SURFACE_DEAD2 | \ + SVGA3D_SURFACE_AUTOGENMIPMAPS | \ + SVGA3D_SURFACE_VOLUME | \ + SVGA3D_SURFACE_1D | \ + SVGA3D_SURFACE_BIND_VERTEX_BUFFER | \ + SVGA3D_SURFACE_BIND_INDEX_BUFFER | \ + SVGA3D_SURFACE_BIND_CONSTANT_BUFFER | \ + SVGA3D_SURFACE_BIND_DEPTH_STENCIL | \ + SVGA3D_SURFACE_BIND_STREAM_OUTPUT | \ + SVGA3D_SURFACE_TRANSFER_FROM_BUFFER | \ + SVGA3D_SURFACE_VADECODE | \ + SVGA3D_SURFACE_MULTISAMPLE | \ + SVGA3D_SURFACE_BIND_UAVIEW | \ + SVGA3D_SURFACE_TRANSFER_TO_BUFFER | \ + SVGA3D_SURFACE_BIND_RAW_VIEWS | \ + SVGA3D_SURFACE_BUFFER_STRUCTURED | \ + SVGA3D_SURFACE_DRAWINDIRECT_ARGS | \ + SVGA3D_SURFACE_RESOURCE_CLAMP \ ) +#define SVGA3D_BUFFER_STRUCTURED_STRIDE_MAX 2048 + + +/* + * These are really the D3DFORMAT_OP defines from the wdk. We need + * them so that we can query the host for what the supported surface + * operations are (when we're using the D3D backend, in particular), + * and so we can send those operations to the guest. + */ typedef enum { SVGA3DFORMAT_OP_TEXTURE = 0x00000001, SVGA3DFORMAT_OP_VOLUMETEXTURE = 0x00000002, @@ -1338,7 +1545,40 @@ typedef enum { SVGA3D_PRIMITIVE_LINESTRIP_ADJ = 8, SVGA3D_PRIMITIVE_TRIANGLELIST_ADJ = 9, SVGA3D_PRIMITIVE_TRIANGLESTRIP_ADJ = 10, - SVGA3D_PRIMITIVE_MAX + SVGA3D_PRIMITIVE_DX10_MAX = 11, + SVGA3D_PRIMITIVE_1_CONTROL_POINT_PATCH = 11, + SVGA3D_PRIMITIVE_2_CONTROL_POINT_PATCH = 12, + SVGA3D_PRIMITIVE_3_CONTROL_POINT_PATCH = 13, + SVGA3D_PRIMITIVE_4_CONTROL_POINT_PATCH = 14, + SVGA3D_PRIMITIVE_5_CONTROL_POINT_PATCH = 15, + SVGA3D_PRIMITIVE_6_CONTROL_POINT_PATCH = 16, + SVGA3D_PRIMITIVE_7_CONTROL_POINT_PATCH = 17, + SVGA3D_PRIMITIVE_8_CONTROL_POINT_PATCH = 18, + SVGA3D_PRIMITIVE_9_CONTROL_POINT_PATCH = 19, + SVGA3D_PRIMITIVE_10_CONTROL_POINT_PATCH = 20, + SVGA3D_PRIMITIVE_11_CONTROL_POINT_PATCH = 21, + SVGA3D_PRIMITIVE_12_CONTROL_POINT_PATCH = 22, + SVGA3D_PRIMITIVE_13_CONTROL_POINT_PATCH = 23, + SVGA3D_PRIMITIVE_14_CONTROL_POINT_PATCH = 24, + SVGA3D_PRIMITIVE_15_CONTROL_POINT_PATCH = 25, + SVGA3D_PRIMITIVE_16_CONTROL_POINT_PATCH = 26, + SVGA3D_PRIMITIVE_17_CONTROL_POINT_PATCH = 27, + SVGA3D_PRIMITIVE_18_CONTROL_POINT_PATCH = 28, + SVGA3D_PRIMITIVE_19_CONTROL_POINT_PATCH = 29, + SVGA3D_PRIMITIVE_20_CONTROL_POINT_PATCH = 30, + SVGA3D_PRIMITIVE_21_CONTROL_POINT_PATCH = 31, + SVGA3D_PRIMITIVE_22_CONTROL_POINT_PATCH = 32, + SVGA3D_PRIMITIVE_23_CONTROL_POINT_PATCH = 33, + SVGA3D_PRIMITIVE_24_CONTROL_POINT_PATCH = 34, + SVGA3D_PRIMITIVE_25_CONTROL_POINT_PATCH = 35, + SVGA3D_PRIMITIVE_26_CONTROL_POINT_PATCH = 36, + SVGA3D_PRIMITIVE_27_CONTROL_POINT_PATCH = 37, + SVGA3D_PRIMITIVE_28_CONTROL_POINT_PATCH = 38, + SVGA3D_PRIMITIVE_29_CONTROL_POINT_PATCH = 39, + SVGA3D_PRIMITIVE_30_CONTROL_POINT_PATCH = 40, + SVGA3D_PRIMITIVE_31_CONTROL_POINT_PATCH = 41, + SVGA3D_PRIMITIVE_32_CONTROL_POINT_PATCH = 42, + SVGA3D_PRIMITIVE_MAX = 43 } SVGA3dPrimitiveType; typedef enum { @@ -1442,16 +1682,15 @@ typedef enum { SVGA3D_QUERYTYPE_STREAMOUTPUTSTATS = 5, SVGA3D_QUERYTYPE_STREAMOVERFLOWPREDICATE = 6, SVGA3D_QUERYTYPE_OCCLUSION64 = 7, - SVGA3D_QUERYTYPE_EVENT = 8, - SVGA3D_QUERYTYPE_DX10_MAX = 9, - SVGA3D_QUERYTYPE_SOSTATS_STREAM0 = 9, - SVGA3D_QUERYTYPE_SOSTATS_STREAM1 = 10, - SVGA3D_QUERYTYPE_SOSTATS_STREAM2 = 11, - SVGA3D_QUERYTYPE_SOSTATS_STREAM3 = 12, - SVGA3D_QUERYTYPE_SOP_STREAM0 = 13, - SVGA3D_QUERYTYPE_SOP_STREAM1 = 14, - SVGA3D_QUERYTYPE_SOP_STREAM2 = 15, - SVGA3D_QUERYTYPE_SOP_STREAM3 = 16, + SVGA3D_QUERYTYPE_DX10_MAX = 8, + SVGA3D_QUERYTYPE_SOSTATS_STREAM0 = 8, + SVGA3D_QUERYTYPE_SOSTATS_STREAM1 = 9, + SVGA3D_QUERYTYPE_SOSTATS_STREAM2 = 10, + SVGA3D_QUERYTYPE_SOSTATS_STREAM3 = 11, + SVGA3D_QUERYTYPE_SOP_STREAM0 = 12, + SVGA3D_QUERYTYPE_SOP_STREAM1 = 13, + SVGA3D_QUERYTYPE_SOP_STREAM2 = 14, + SVGA3D_QUERYTYPE_SOP_STREAM3 = 15, SVGA3D_QUERYTYPE_MAX } SVGA3dQueryType; @@ -1584,28 +1823,33 @@ typedef enum { SVGA3D_READ_HOST_VRAM = 2, } SVGA3dTransferType; -typedef enum { - SVGA3D_LOGICOP_INVALID = 0, - SVGA3D_LOGICOP_MIN = 1, - SVGA3D_LOGICOP_COPY = 1, - SVGA3D_LOGICOP_NOT = 2, - SVGA3D_LOGICOP_AND = 3, - SVGA3D_LOGICOP_OR = 4, - SVGA3D_LOGICOP_XOR = 5, - SVGA3D_LOGICOP_NXOR = 6, - SVGA3D_LOGICOP_ROP3MIN = 30, /* 7-29 are reserved for future logic ops. */ - SVGA3D_LOGICOP_ROP3MAX = (SVGA3D_LOGICOP_ROP3MIN + 255), - SVGA3D_LOGICOP_MAX = (SVGA3D_LOGICOP_ROP3MAX + 1), -} SVGA3dLogicOp; +#define SVGA3D_LOGICOP_INVALID 0 +#define SVGA3D_LOGICOP_MIN 1 +#define SVGA3D_LOGICOP_COPY 1 +#define SVGA3D_LOGICOP_NOT 2 +#define SVGA3D_LOGICOP_AND 3 +#define SVGA3D_LOGICOP_OR 4 +#define SVGA3D_LOGICOP_XOR 5 +#define SVGA3D_LOGICOP_NXOR 6 +#define SVGA3D_LOGICOP_ROP3 7 +#define SVGA3D_LOGICOP_MAX 8 + +typedef uint16 SVGA3dLogicOp; + +#define SVGA3D_LOGICOP_ROP3_INVALID ((uint16) -1) +#define SVGA3D_LOGICOP_ROP3_MIN 0 +#define SVGA3D_LOGICOP_ROP3_MAX 256 + +typedef uint16 SVGA3dLogicOpRop3; typedef #include "vmware_pack_begin.h" struct { union { struct { - uint16 function; /* SVGA3dFogFunction */ - uint8 type; /* SVGA3dFogType */ - uint8 base; /* SVGA3dFogBase */ + uint16 function; // SVGA3dFogFunction + uint8 type; // SVGA3dFogType + uint8 base; // SVGA3dFogBase }; uint32 uintValue; }; @@ -1742,4 +1986,15 @@ typedef enum SVGA3dMSQualityLevel { SVGA3D_MS_QUALITY_MAX = 2, } SVGA3dMSQualityLevel; +/* + * Screen Target Update Flags + */ + +typedef enum SVGA3dFrameUpdateType { + SVGA3D_FRAME_END = 0, + SVGA3D_FRAME_PARTIAL = 1, + SVGA3D_FRAME_UNKNOWN = 2, + SVGA3D_FRAME_MAX = 3, +} SVGA3dFrameUpdateType; + #endif /* _SVGA3D_TYPES_H_ */ diff --git a/drivers/gpu/drm/vmwgfx/device_include/svga_reg.h b/drivers/gpu/drm/vmwgfx/device_include/svga_reg.h index 056f54b35d73..19fb9e3299e7 100644 --- a/drivers/gpu/drm/vmwgfx/device_include/svga_reg.h +++ b/drivers/gpu/drm/vmwgfx/device_include/svga_reg.h @@ -70,8 +70,7 @@ typedef uint32 SVGAMobId; /* * Legal values for the SVGA_REG_CURSOR_ON register in old-fashioned - * cursor bypass mode. This is still supported, but no new guest - * drivers should use it. + * cursor bypass mode. */ #define SVGA_CURSOR_ON_HIDE 0x0 #define SVGA_CURSOR_ON_SHOW 0x1 @@ -137,6 +136,17 @@ typedef uint32 SVGAMobId; #define SVGA_IRQFLAG_ERROR 0x10 /* Error while processing commands */ /* + * The byte-size is the size of the actual cursor data, + * possibly after expanding it to the current bit depth. + * + * 40K is sufficient memory for two 32-bit planes for a 64 x 64 cursor. + * + * The dimension limit is a bound on the maximum width or height. + */ +#define SVGA_MAX_CURSOR_CMD_BYTES (40 * 1024) +#define SVGA_MAX_CURSOR_CMD_DIMENSION 1024 + +/* * Registers */ @@ -169,7 +179,7 @@ enum { SVGA_REG_SYNC = 21, /* See "FIFO Synchronization Registers" */ SVGA_REG_BUSY = 22, /* See "FIFO Synchronization Registers" */ SVGA_REG_GUEST_ID = 23, /* (Deprecated) */ - SVGA_REG_CURSOR_ID = 24, /* (Deprecated) */ + SVGA_REG_DEAD = 24, /* Drivers should never write this. */ SVGA_REG_CURSOR_X = 25, /* (Deprecated) */ SVGA_REG_CURSOR_Y = 26, /* (Deprecated) */ SVGA_REG_CURSOR_ON = 27, /* (Deprecated) */ @@ -208,7 +218,13 @@ enum { SVGA_REG_MAX_PRIMARY_MEM = 50, SVGA_REG_MAX_PRIMARY_BOUNDING_BOX_MEM = 50, - SVGA_REG_SUGGESTED_GBOBJECT_MEM_SIZE_KB = 51, /* Sugested limit on mob mem */ + /* + * Legacy version of SVGA_REG_GBOBJECT_MEM_SIZE_KB for drivers that + * don't know how to convert to a 64-bit byte value without overflowing. + * (See SVGA_REG_GBOBJECT_MEM_SIZE_KB). + */ + SVGA_REG_SUGGESTED_GBOBJECT_MEM_SIZE_KB = 51, + SVGA_REG_DEV_CAP = 52, /* Write dev cap index, read value */ SVGA_REG_CMD_PREPEND_LOW = 53, SVGA_REG_CMD_PREPEND_HIGH = 54, @@ -218,7 +234,59 @@ enum { SVGA_REG_BLANK_SCREEN_TARGETS = 58, SVGA_REG_CAP2 = 59, SVGA_REG_DEVEL_CAP = 60, - SVGA_REG_TOP = 61, /* Must be 1 more than the last register */ + + /* + * Allow the guest to hint to the device which driver is running. + * + * This should not generally change device behavior, but might be + * convenient to work-around specific bugs in guest drivers. + * + * Drivers should first write their id value into SVGA_REG_GUEST_DRIVER_ID, + * and then fill out all of the version registers that they have defined. + * + * After the driver has written all of the registers, they should + * then write the value SVGA_REG_GUEST_DRIVER_ID_SUBMIT to the + * SVGA_REG_GUEST_DRIVER_ID register, to signal that they have finished. + * + * The SVGA_REG_GUEST_DRIVER_ID values are defined below by the + * SVGARegGuestDriverId enum. + * + * The SVGA_REG_GUEST_DRIVER_VERSION fields are driver-specific, + * but ideally should encode a monotonically increasing number that allows + * the device to perform inequality checks against ranges of driver versions. + */ + SVGA_REG_GUEST_DRIVER_ID = 61, + SVGA_REG_GUEST_DRIVER_VERSION1 = 62, + SVGA_REG_GUEST_DRIVER_VERSION2 = 63, + SVGA_REG_GUEST_DRIVER_VERSION3 = 64, + SVGA_REG_CURSOR_MOBID = 65, + SVGA_REG_CURSOR_MAX_BYTE_SIZE = 66, + SVGA_REG_CURSOR_MAX_DIMENSION = 67, + + SVGA_REG_FIFO_CAPS = 68, + SVGA_REG_FENCE = 69, + + SVGA_REG_RESERVED1 = 70, + SVGA_REG_RESERVED2 = 71, + SVGA_REG_RESERVED3 = 72, + SVGA_REG_RESERVED4 = 73, + SVGA_REG_RESERVED5 = 74, + SVGA_REG_SCREENDMA = 75, + + /* + * The maximum amount of guest-backed objects that the device can have + * resident at a time. Guest-drivers should keep their working set size + * below this limit for best performance. + * + * Note that this value is in kilobytes, and not bytes, because the actual + * number of bytes might be larger than can fit in a 32-bit register. + * + * PLEASE USE A 64-BIT VALUE WHEN CONVERTING THIS INTO BYTES. + * (See SVGA_REG_SUGGESTED_GBOBJECT_MEM_SIZE_KB). + */ + SVGA_REG_GBOBJECT_MEM_SIZE_KB = 76, + + SVGA_REG_TOP = 77, /* Must be 1 more than the last register */ SVGA_PALETTE_BASE = 1024, /* Base of SVGA color map */ /* Next 768 (== 256*3) registers exist for colormap */ @@ -229,6 +297,20 @@ enum { the use of the current SVGA driver. */ }; + +/* + * Values for SVGA_REG_GUEST_DRIVER_ID. + */ +typedef enum SVGARegGuestDriverId { + SVGA_REG_GUEST_DRIVER_ID_UNKNOWN = 0, + SVGA_REG_GUEST_DRIVER_ID_WDDM = 1, + SVGA_REG_GUEST_DRIVER_ID_LINUX = 2, + SVGA_REG_GUEST_DRIVER_ID_MAX, + + SVGA_REG_GUEST_DRIVER_ID_SUBMIT = MAX_UINT32, +} SVGARegGuestDriverId; + + /* * Guest memory regions (GMRs): * @@ -416,7 +498,6 @@ typedef enum { SVGA_CB_CONTEXT_0 = 0x0, SVGA_CB_CONTEXT_1 = 0x1, /* Supported with SVGA_CAP_HP_CMD_QUEUE */ SVGA_CB_CONTEXT_MAX = 0x2, - SVGA_CB_CONTEXT_HP_MAX = 0x2, } SVGACBContext; @@ -733,9 +814,6 @@ SVGASignedPoint; * and must not be reused. Those capabilities will never be reported * by new versions of the SVGA device. * - * XXX: Add longer descriptions for each capability, including a list - * of the new features that each capability provides. - * * SVGA_CAP_IRQMASK -- * Provides device interrupts. Adds device register SVGA_REG_IRQMASK * to set interrupt mask and direct I/O port SVGA_IRQSTATUS_PORT to @@ -842,17 +920,51 @@ SVGASignedPoint; * Allow the IntraSurfaceCopy command. * * SVGA_CAP2_DX2 -- - * Allow the DefineGBSurface_v3, WholeSurfaceCopy. + * Allow the DefineGBSurface_v3, WholeSurfaceCopy, WriteZeroSurface, and + * HintZeroSurface commands, and the SVGA_REG_GUEST_DRIVER_ID register. + * + * SVGA_CAP2_GB_MEMSIZE_2 -- + * Allow the SVGA_REG_GBOBJECT_MEM_SIZE_KB register. + * + * SVGA_CAP2_SCREENDMA_REG -- + * Allow the SVGA_REG_SCREENDMA register. + * + * SVGA_CAP2_OTABLE_PTDEPTH_2 -- + * Allow 2 level page tables for OTable commands. + * + * SVGA_CAP2_NON_MS_TO_MS_STRETCHBLT -- + * Allow a stretch blt from a non-multisampled surface to a multisampled + * surface. + * + * SVGA_CAP2_CURSOR_MOB -- + * Allow the SVGA_REG_CURSOR_MOBID register. + * + * SVGA_CAP2_MSHINT -- + * Allow the SVGA_REG_MSHINT register. + * + * SVGA_CAP2_DX3 -- + * Allows the DefineGBSurface_v4 command. + * Allows the DXDefineDepthStencilView_v2, DXDefineStreamOutputWithMob, + * and DXBindStreamOutput commands if 3D is also available. + * Allows the DXPredStagingCopy and DXStagingCopy commands if SM41 + * is also available. * * SVGA_CAP2_RESERVED -- * Reserve the last bit for extending the SVGA capabilities to some * future mechanisms. */ -#define SVGA_CAP2_NONE 0x00000000 -#define SVGA_CAP2_GROW_OTABLE 0x00000001 -#define SVGA_CAP2_INTRA_SURFACE_COPY 0x00000002 -#define SVGA_CAP2_DX2 0x00000004 -#define SVGA_CAP2_RESERVED 0x80000000 +#define SVGA_CAP2_NONE 0x00000000 +#define SVGA_CAP2_GROW_OTABLE 0x00000001 +#define SVGA_CAP2_INTRA_SURFACE_COPY 0x00000002 +#define SVGA_CAP2_DX2 0x00000004 +#define SVGA_CAP2_GB_MEMSIZE_2 0x00000008 +#define SVGA_CAP2_SCREENDMA_REG 0x00000010 +#define SVGA_CAP2_OTABLE_PTDEPTH_2 0x00000020 +#define SVGA_CAP2_NON_MS_TO_MS_STRETCHBLT 0x00000040 +#define SVGA_CAP2_CURSOR_MOB 0x00000080 +#define SVGA_CAP2_MSHINT 0x00000100 +#define SVGA_CAP2_DX3 0x00000400 +#define SVGA_CAP2_RESERVED 0x80000000 /* @@ -875,7 +987,9 @@ typedef enum { SVGABackdoorCapFifoCaps = 1, SVGABackdoorCap3dHWVersion = 2, SVGABackdoorCapDeviceCaps2 = 3, - SVGABackdoorCapMax = 4, + SVGABackdoorCapDevelCaps = 4, + SVGABackdoorDevelRenderer = 5, + SVGABackdoorCapMax = 6, } SVGABackdoorCapType; @@ -1055,103 +1169,80 @@ enum { /* * FIFO Synchronization Registers * - * This explains the relationship between the various FIFO - * sync-related registers in IOSpace and in FIFO space. - * * SVGA_REG_SYNC -- * - * The SYNC register can be used in two different ways by the guest: - * - * 1. If the guest wishes to fully sync (drain) the FIFO, - * it will write once to SYNC then poll on the BUSY - * register. The FIFO is sync'ed once BUSY is zero. - * - * 2. If the guest wants to asynchronously wake up the host, - * it will write once to SYNC without polling on BUSY. - * Ideally it will do this after some new commands have - * been placed in the FIFO, and after reading a zero - * from SVGA_FIFO_BUSY. - * - * (1) is the original behaviour that SYNC was designed to - * support. Originally, a write to SYNC would implicitly - * trigger a read from BUSY. This causes us to synchronously - * process the FIFO. - * - * This behaviour has since been changed so that writing SYNC - * will *not* implicitly cause a read from BUSY. Instead, it - * makes a channel call which asynchronously wakes up the MKS - * thread. - * - * New guests can use this new behaviour to implement (2) - * efficiently. This lets guests get the host's attention - * without waiting for the MKS to poll, which gives us much - * better CPU utilization on SMP hosts and on UP hosts while - * we're blocked on the host GPU. - * - * Old guests shouldn't notice the behaviour change. SYNC was - * never guaranteed to process the entire FIFO, since it was - * bounded to a particular number of CPU cycles. Old guests will - * still loop on the BUSY register until the FIFO is empty. - * - * Writing to SYNC currently has the following side-effects: - * - * - Sets SVGA_REG_BUSY to TRUE (in the monitor) - * - Asynchronously wakes up the MKS thread for FIFO processing - * - The value written to SYNC is recorded as a "reason", for - * stats purposes. - * - * If SVGA_FIFO_BUSY is available, drivers are advised to only - * write to SYNC if SVGA_FIFO_BUSY is FALSE. Drivers should set - * SVGA_FIFO_BUSY to TRUE after writing to SYNC. The MKS will - * eventually set SVGA_FIFO_BUSY on its own, but this approach - * lets the driver avoid sending multiple asynchronous wakeup - * messages to the MKS thread. + * The SYNC register can be used by the guest driver to signal to the + * device that the guest driver is waiting for previously submitted + * commands to complete. + * + * When the guest driver writes to the SYNC register, the device sets + * the BUSY register to TRUE, and starts processing the submitted commands + * (if it was not already doing so). When all previously submitted + * commands are finished and the device is idle again, it sets the BUSY + * register back to FALSE. (If the guest driver submits new commands + * after writing the SYNC register, the new commands are not guaranteed + * to have been procesesd.) + * + * When guest drivers are submitting commands using the FIFO, the device + * periodically polls to check for new FIFO commands when idle, which may + * introduce a delay in command processing. If the guest-driver wants + * the commands to be processed quickly (which it typically does), it + * should write SYNC after each batch of commands is committed to the + * FIFO to immediately wake up the device. For even better performance, + * the guest can use the SVGA_FIFO_BUSY register to avoid these extra + * SYNC writes if the device is already active, using the technique known + * as "Ringing the Doorbell" (described below). (Note that command + * buffer submission implicitly wakes up the device, and so doesn't + * suffer from this problem.) + * + * The SYNC register can also be used in combination with BUSY to + * synchronously ensure that all SVGA commands are processed (with both + * the FIFO and command-buffers). To do this, the guest driver should + * write to SYNC, and then loop reading BUSY until BUSY returns FALSE. + * This technique is known as a "Legacy Sync". * * SVGA_REG_BUSY -- * * This register is set to TRUE when SVGA_REG_SYNC is written, - * and it reads as FALSE when the FIFO has been completely - * drained. - * - * Every read from this register causes us to synchronously - * process FIFO commands. There is no guarantee as to how many - * commands each read will process. + * and is set back to FALSE when the device has finished processing + * all commands and is idle again. * - * CPU time spent processing FIFO commands will be billed to - * the guest. + * Every read from the BUSY reigster will block for an undefined + * amount of time (normally until the device finishes some interesting + * work unit), or the device is idle. * - * New drivers should avoid using this register unless they - * need to guarantee that the FIFO is completely drained. It - * is overkill for performing a sync-to-fence. Older drivers - * will use this register for any type of synchronization. + * Guest drivers can also do a partial Legacy Sync to check for some + * particular condition, for instance by stopping early when a fence + * passes before BUSY has been set back to FALSE. This is particularly + * useful if the guest-driver knows that it is blocked waiting on the + * device, because it will yield CPU time back to the host. * * SVGA_FIFO_BUSY -- * - * This register is a fast way for the guest driver to check - * whether the FIFO is already being processed. It reads and - * writes at normal RAM speeds, with no monitor intervention. - * - * If this register reads as TRUE, the host is guaranteeing that - * any new commands written into the FIFO will be noticed before - * the MKS goes back to sleep. + * The SVGA_FIFO_BUSY register is a fast way for the guest driver to check + * whether the device is actively processing FIFO commands before writing + * the more expensive SYNC register. * - * If this register reads as FALSE, no such guarantee can be - * made. + * If this register reads as TRUE, the device is actively processing + * FIFO commands. * - * The guest should use this register to quickly determine - * whether or not it needs to wake up the host. If the guest - * just wrote a command or group of commands that it would like - * the host to begin processing, it should: + * If this register reads as FALSE, the device may not be actively + * processing commands, and the guest driver should try + * "Ringing the Doorbell". * - * 1. Read SVGA_FIFO_BUSY. If it reads as TRUE, no further - * action is necessary. + * To Ring the Doorbell, the guest should: * - * 2. Write TRUE to SVGA_FIFO_BUSY. This informs future guest - * code that we've already sent a SYNC to the host and we - * don't need to send a duplicate. + * 1. Have already written their batch of commands into the FIFO. + * 2. Check if the SVGA_FIFO_BUSY register is available by reading + * SVGA_FIFO_MIN. + * 3. Read SVGA_FIFO_BUSY. If it reads as TRUE, the device is actively + * processing FIFO commands, and no further action is necessary. + * 4. If SVGA_FIFO_BUSY was FALSE, write TRUE to SVGA_REG_SYNC. * - * 3. Write a reason to SVGA_REG_SYNC. This will send an - * asynchronous wakeup to the MKS thread. + * For maximum performance, this procedure should be followed after + * every meaningful batch of commands has been written into the FIFO. + * (Normally when the underlying application signals it's finished a + * meaningful work unit by calling Flush.) */ @@ -1164,9 +1255,6 @@ enum { * Video -- SVGA Video overlay units are supported * Escape -- Escape command is supported * - * XXX: Add longer descriptions for each capability, including a list - * of the new features that each capability provides. - * * SVGA_FIFO_CAP_SCREEN_OBJECT -- * * Provides dynamic multi-screen rendering, for improved Unity and @@ -1279,6 +1367,15 @@ enum { /* + * ScreenDMA Register Values + */ + +#define SVGA_SCREENDMA_REG_UNDEFINED 0 +#define SVGA_SCREENDMA_REG_NOT_PRESENT 1 +#define SVGA_SCREENDMA_REG_PRESENT 2 +#define SVGA_SCREENDMA_REG_MAX 3 + +/* * Video overlay support */ @@ -1665,6 +1762,80 @@ SVGAFifoCmdDefineAlphaCursor; /* + * Provide a new large cursor image, as an AND/XOR mask. + * + * Should only be used for CursorMob functionality + */ + +typedef +#include "vmware_pack_begin.h" +struct { + uint32 hotspotX; + uint32 hotspotY; + uint32 width; + uint32 height; + uint32 andMaskDepth; + uint32 xorMaskDepth; + /* + * Followed by scanline data for AND mask, then XOR mask. + * Each scanline is padded to a 32-bit boundary. + */ +} +#include "vmware_pack_end.h" +SVGAGBColorCursorHeader; + + +/* + * Provide a new large cursor image, in 32-bit BGRA format. + * + * Should only be used for CursorMob functionality + */ + +typedef +#include "vmware_pack_begin.h" +struct { + uint32 hotspotX; + uint32 hotspotY; + uint32 width; + uint32 height; + /* Followed by scanline data */ +} +#include "vmware_pack_end.h" +SVGAGBAlphaCursorHeader; + + /* + * Define the SVGA guest backed cursor types + */ + +typedef enum { + SVGA_COLOR_CURSOR = 0, + SVGA_ALPHA_CURSOR = 1, +} SVGAGBCursorType; + +/* + * Provide a new large cursor image. + * + * Should only be used for CursorMob functionality + */ + +typedef +#include "vmware_pack_begin.h" +struct { + SVGAGBCursorType type; + union { + SVGAGBColorCursorHeader colorHeader; + SVGAGBAlphaCursorHeader alphaHeader; + } header; + uint32 sizeInBytes; + /* + * Followed by the cursor data + */ +} +#include "vmware_pack_end.h" +SVGAGBCursorHeader; + + +/* * SVGA_CMD_UPDATE_VERBOSE -- * * Just like SVGA_CMD_UPDATE, but also provide a per-rectangle @@ -2061,9 +2232,12 @@ SVGAFifoCmdRemapGMR2; #define SVGA_VRAM_MAX_SIZE (128 * 1024 * 1024) #define SVGA_MEMORY_SIZE_MAX (1024 * 1024 * 1024) #define SVGA_FIFO_SIZE_MAX (2 * 1024 * 1024) -#define SVGA_GRAPHICS_MEMORY_KB_MIN (32 * 1024) -#define SVGA_GRAPHICS_MEMORY_KB_MAX (2 * 1024 * 1024) -#define SVGA_GRAPHICS_MEMORY_KB_DEFAULT (256 * 1024) +#define SVGA_GRAPHICS_MEMORY_KB_MIN (32 * 1024) +#define SVGA_GRAPHICS_MEMORY_KB_MAX_2GB (2 * 1024 * 1024) +#define SVGA_GRAPHICS_MEMORY_KB_MAX_3GB (3 * 1024 * 1024) +#define SVGA_GRAPHICS_MEMORY_KB_MAX_4GB (4 * 1024 * 1024) +#define SVGA_GRAPHICS_MEMORY_KB_MAX_8GB (8 * 1024 * 1024) +#define SVGA_GRAPHICS_MEMORY_KB_DEFAULT (256 * 1024) #define SVGA_VRAM_SIZE_W2K (64 * 1024 * 1024) /* 64 MB */ @@ -2086,4 +2260,6 @@ SVGAFifoCmdRemapGMR2; #define SVGA_FIFO_SIZE_GBOBJECTS (256 * 1024) #define SVGA_VRAM_SIZE_GBOBJECTS (4 * 1024 * 1024) +#define SVGA_PCI_REGS_PAGES (1) + #endif diff --git a/drivers/gpu/drm/vmwgfx/device_include/svga_types.h b/drivers/gpu/drm/vmwgfx/device_include/svga_types.h index 350bbc6fab02..beddccee40f6 100644 --- a/drivers/gpu/drm/vmwgfx/device_include/svga_types.h +++ b/drivers/gpu/drm/vmwgfx/device_include/svga_types.h @@ -37,6 +37,7 @@ typedef s8 int8; typedef uint64 PA; typedef uint32 PPN; +typedef uint32 PPN32; typedef uint64 PPN64; typedef bool Bool; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_binding.c b/drivers/gpu/drm/vmwgfx/vmwgfx_binding.c index 66e14e38d5e8..f41550797970 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_binding.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_binding.c @@ -57,9 +57,11 @@ #define VMW_BINDING_RT_BIT 0 #define VMW_BINDING_PS_BIT 1 -#define VMW_BINDING_SO_BIT 2 +#define VMW_BINDING_SO_T_BIT 2 #define VMW_BINDING_VB_BIT 3 -#define VMW_BINDING_NUM_BITS 4 +#define VMW_BINDING_UAV_BIT 4 +#define VMW_BINDING_CS_UAV_BIT 5 +#define VMW_BINDING_NUM_BITS 6 #define VMW_BINDING_PS_SR_BIT 0 @@ -75,6 +77,8 @@ * @vertex_buffers: Vertex buffer bindings. * @index_buffer: Index buffer binding. * @per_shader: Per shader-type bindings. + * @ua_views: UAV bindings. + * @so_state: StreamOutput bindings. * @dirty: Bitmap tracking per binding-type changes that have not yet * been emitted to the device. * @dirty_vb: Bitmap tracking individual vertex buffer binding changes that @@ -95,10 +99,12 @@ struct vmw_ctx_binding_state { struct vmw_ctx_bindinfo_view render_targets[SVGA3D_RT_MAX]; struct vmw_ctx_bindinfo_tex texture_units[SVGA3D_NUM_TEXTURE_UNITS]; struct vmw_ctx_bindinfo_view ds_view; - struct vmw_ctx_bindinfo_so so_targets[SVGA3D_DX_MAX_SOTARGETS]; + struct vmw_ctx_bindinfo_so_target so_targets[SVGA3D_DX_MAX_SOTARGETS]; struct vmw_ctx_bindinfo_vb vertex_buffers[SVGA3D_DX_MAX_VERTEXBUFFERS]; struct vmw_ctx_bindinfo_ib index_buffer; - struct vmw_dx_shader_bindings per_shader[SVGA3D_NUM_SHADERTYPE_DX10]; + struct vmw_dx_shader_bindings per_shader[SVGA3D_NUM_SHADERTYPE]; + struct vmw_ctx_bindinfo_uav ua_views[VMW_MAX_UAV_BIND_TYPE]; + struct vmw_ctx_bindinfo_so so_state; unsigned long dirty; DECLARE_BITMAP(dirty_vb, SVGA3D_DX_MAX_VERTEXBUFFERS); @@ -115,12 +121,16 @@ static int vmw_binding_scrub_texture(struct vmw_ctx_bindinfo *bi, bool rebind); static int vmw_binding_scrub_cb(struct vmw_ctx_bindinfo *bi, bool rebind); static int vmw_binding_scrub_dx_rt(struct vmw_ctx_bindinfo *bi, bool rebind); static int vmw_binding_scrub_sr(struct vmw_ctx_bindinfo *bi, bool rebind); -static int vmw_binding_scrub_so(struct vmw_ctx_bindinfo *bi, bool rebind); +static int vmw_binding_scrub_so_target(struct vmw_ctx_bindinfo *bi, bool rebind); static int vmw_binding_emit_dirty(struct vmw_ctx_binding_state *cbs); static int vmw_binding_scrub_dx_shader(struct vmw_ctx_bindinfo *bi, bool rebind); static int vmw_binding_scrub_ib(struct vmw_ctx_bindinfo *bi, bool rebind); static int vmw_binding_scrub_vb(struct vmw_ctx_bindinfo *bi, bool rebind); +static int vmw_binding_scrub_uav(struct vmw_ctx_bindinfo *bi, bool rebind); +static int vmw_binding_scrub_cs_uav(struct vmw_ctx_bindinfo *bi, bool rebind); +static int vmw_binding_scrub_so(struct vmw_ctx_bindinfo *bi, bool rebind); + static void vmw_binding_build_asserts(void) __attribute__ ((unused)); typedef int (*vmw_scrub_func)(struct vmw_ctx_bindinfo *, bool); @@ -151,6 +161,9 @@ static const size_t vmw_binding_shader_offsets[] = { offsetof(struct vmw_ctx_binding_state, per_shader[0].shader), offsetof(struct vmw_ctx_binding_state, per_shader[1].shader), offsetof(struct vmw_ctx_binding_state, per_shader[2].shader), + offsetof(struct vmw_ctx_binding_state, per_shader[3].shader), + offsetof(struct vmw_ctx_binding_state, per_shader[4].shader), + offsetof(struct vmw_ctx_binding_state, per_shader[5].shader), }; static const size_t vmw_binding_rt_offsets[] = { offsetof(struct vmw_ctx_binding_state, render_targets), @@ -162,6 +175,9 @@ static const size_t vmw_binding_cb_offsets[] = { offsetof(struct vmw_ctx_binding_state, per_shader[0].const_buffers), offsetof(struct vmw_ctx_binding_state, per_shader[1].const_buffers), offsetof(struct vmw_ctx_binding_state, per_shader[2].const_buffers), + offsetof(struct vmw_ctx_binding_state, per_shader[3].const_buffers), + offsetof(struct vmw_ctx_binding_state, per_shader[4].const_buffers), + offsetof(struct vmw_ctx_binding_state, per_shader[5].const_buffers), }; static const size_t vmw_binding_dx_ds_offsets[] = { offsetof(struct vmw_ctx_binding_state, ds_view), @@ -170,8 +186,11 @@ static const size_t vmw_binding_sr_offsets[] = { offsetof(struct vmw_ctx_binding_state, per_shader[0].shader_res), offsetof(struct vmw_ctx_binding_state, per_shader[1].shader_res), offsetof(struct vmw_ctx_binding_state, per_shader[2].shader_res), + offsetof(struct vmw_ctx_binding_state, per_shader[3].shader_res), + offsetof(struct vmw_ctx_binding_state, per_shader[4].shader_res), + offsetof(struct vmw_ctx_binding_state, per_shader[5].shader_res), }; -static const size_t vmw_binding_so_offsets[] = { +static const size_t vmw_binding_so_target_offsets[] = { offsetof(struct vmw_ctx_binding_state, so_targets), }; static const size_t vmw_binding_vb_offsets[] = { @@ -180,6 +199,15 @@ static const size_t vmw_binding_vb_offsets[] = { static const size_t vmw_binding_ib_offsets[] = { offsetof(struct vmw_ctx_binding_state, index_buffer), }; +static const size_t vmw_binding_uav_offsets[] = { + offsetof(struct vmw_ctx_binding_state, ua_views[0].views), +}; +static const size_t vmw_binding_cs_uav_offsets[] = { + offsetof(struct vmw_ctx_binding_state, ua_views[1].views), +}; +static const size_t vmw_binding_so_offsets[] = { + offsetof(struct vmw_ctx_binding_state, so_state), +}; static const struct vmw_binding_info vmw_binding_infos[] = { [vmw_ctx_binding_shader] = { @@ -214,10 +242,10 @@ static const struct vmw_binding_info vmw_binding_infos[] = { .size = sizeof(struct vmw_ctx_bindinfo_view), .offsets = vmw_binding_dx_ds_offsets, .scrub_func = vmw_binding_scrub_dx_rt}, - [vmw_ctx_binding_so] = { - .size = sizeof(struct vmw_ctx_bindinfo_so), - .offsets = vmw_binding_so_offsets, - .scrub_func = vmw_binding_scrub_so}, + [vmw_ctx_binding_so_target] = { + .size = sizeof(struct vmw_ctx_bindinfo_so_target), + .offsets = vmw_binding_so_target_offsets, + .scrub_func = vmw_binding_scrub_so_target}, [vmw_ctx_binding_vb] = { .size = sizeof(struct vmw_ctx_bindinfo_vb), .offsets = vmw_binding_vb_offsets, @@ -226,6 +254,18 @@ static const struct vmw_binding_info vmw_binding_infos[] = { .size = sizeof(struct vmw_ctx_bindinfo_ib), .offsets = vmw_binding_ib_offsets, .scrub_func = vmw_binding_scrub_ib}, + [vmw_ctx_binding_uav] = { + .size = sizeof(struct vmw_ctx_bindinfo_view), + .offsets = vmw_binding_uav_offsets, + .scrub_func = vmw_binding_scrub_uav}, + [vmw_ctx_binding_cs_uav] = { + .size = sizeof(struct vmw_ctx_bindinfo_view), + .offsets = vmw_binding_cs_uav_offsets, + .scrub_func = vmw_binding_scrub_cs_uav}, + [vmw_ctx_binding_so] = { + .size = sizeof(struct vmw_ctx_bindinfo_so), + .offsets = vmw_binding_so_offsets, + .scrub_func = vmw_binding_scrub_so}, }; /** @@ -312,6 +352,18 @@ void vmw_binding_add(struct vmw_ctx_binding_state *cbs, } /** + * vmw_binding_add_uav_index - Add UAV index for tracking. + * @cbs: Pointer to the context binding state tracker. + * @slot: UAV type to which bind this index. + * @index: The splice index to track. + */ +void vmw_binding_add_uav_index(struct vmw_ctx_binding_state *cbs, uint32 slot, + uint32 index) +{ + cbs->ua_views[slot].index = index; +} + +/** * vmw_binding_transfer: Transfer a context binding tracking entry. * * @cbs: Pointer to the persistent context binding state tracker. @@ -450,6 +502,10 @@ void vmw_binding_state_commit(struct vmw_ctx_binding_state *to, vmw_binding_transfer(to, from, entry); vmw_binding_drop(entry); } + + /* Also transfer uav splice indices */ + to->ua_views[0].index = from->ua_views[0].index; + to->ua_views[1].index = from->ua_views[1].index; } /** @@ -828,8 +884,8 @@ static void vmw_collect_so_targets(struct vmw_ctx_binding_state *cbs, const struct vmw_ctx_bindinfo *bi, u32 max_num) { - const struct vmw_ctx_bindinfo_so *biso = - container_of(bi, struct vmw_ctx_bindinfo_so, bi); + const struct vmw_ctx_bindinfo_so_target *biso = + container_of(bi, struct vmw_ctx_bindinfo_so_target, bi); unsigned long i; SVGA3dSoTarget *so_buffer = (SVGA3dSoTarget *) cbs->bind_cmd_buffer; @@ -854,11 +910,11 @@ static void vmw_collect_so_targets(struct vmw_ctx_binding_state *cbs, } /** - * vmw_binding_emit_set_so - Issue delayed streamout binding commands + * vmw_emit_set_so_target - Issue delayed streamout binding commands * * @cbs: Pointer to the context's struct vmw_ctx_binding_state */ -static int vmw_emit_set_so(struct vmw_ctx_binding_state *cbs) +static int vmw_emit_set_so_target(struct vmw_ctx_binding_state *cbs) { const struct vmw_ctx_bindinfo *loc = &cbs->so_targets[0].bi; struct { @@ -1005,6 +1061,66 @@ static int vmw_emit_set_vb(struct vmw_ctx_binding_state *cbs) return 0; } +static int vmw_emit_set_uav(struct vmw_ctx_binding_state *cbs) +{ + const struct vmw_ctx_bindinfo *loc = &cbs->ua_views[0].views[0].bi; + struct { + SVGA3dCmdHeader header; + SVGA3dCmdDXSetUAViews body; + } *cmd; + size_t cmd_size, view_id_size; + const struct vmw_resource *ctx = vmw_cbs_context(cbs); + + vmw_collect_view_ids(cbs, loc, SVGA3D_MAX_UAVIEWS); + view_id_size = cbs->bind_cmd_count*sizeof(uint32); + cmd_size = sizeof(*cmd) + view_id_size; + cmd = VMW_FIFO_RESERVE_DX(ctx->dev_priv, cmd_size, ctx->id); + if (!cmd) + return -ENOMEM; + + cmd->header.id = SVGA_3D_CMD_DX_SET_UA_VIEWS; + cmd->header.size = sizeof(cmd->body) + view_id_size; + + /* Splice index is specified user-space */ + cmd->body.uavSpliceIndex = cbs->ua_views[0].index; + + memcpy(&cmd[1], cbs->bind_cmd_buffer, view_id_size); + + vmw_fifo_commit(ctx->dev_priv, cmd_size); + + return 0; +} + +static int vmw_emit_set_cs_uav(struct vmw_ctx_binding_state *cbs) +{ + const struct vmw_ctx_bindinfo *loc = &cbs->ua_views[1].views[0].bi; + struct { + SVGA3dCmdHeader header; + SVGA3dCmdDXSetCSUAViews body; + } *cmd; + size_t cmd_size, view_id_size; + const struct vmw_resource *ctx = vmw_cbs_context(cbs); + + vmw_collect_view_ids(cbs, loc, SVGA3D_MAX_UAVIEWS); + view_id_size = cbs->bind_cmd_count*sizeof(uint32); + cmd_size = sizeof(*cmd) + view_id_size; + cmd = VMW_FIFO_RESERVE_DX(ctx->dev_priv, cmd_size, ctx->id); + if (!cmd) + return -ENOMEM; + + cmd->header.id = SVGA_3D_CMD_DX_SET_CS_UA_VIEWS; + cmd->header.size = sizeof(cmd->body) + view_id_size; + + /* Start index is specified user-space */ + cmd->body.startIndex = cbs->ua_views[1].index; + + memcpy(&cmd[1], cbs->bind_cmd_buffer, view_id_size); + + vmw_fifo_commit(ctx->dev_priv, cmd_size); + + return 0; +} + /** * vmw_binding_emit_dirty - Issue delayed binding commands * @@ -1030,12 +1146,18 @@ static int vmw_binding_emit_dirty(struct vmw_ctx_binding_state *cbs) case VMW_BINDING_PS_BIT: ret = vmw_binding_emit_dirty_ps(cbs); break; - case VMW_BINDING_SO_BIT: - ret = vmw_emit_set_so(cbs); + case VMW_BINDING_SO_T_BIT: + ret = vmw_emit_set_so_target(cbs); break; case VMW_BINDING_VB_BIT: ret = vmw_emit_set_vb(cbs); break; + case VMW_BINDING_UAV_BIT: + ret = vmw_emit_set_uav(cbs); + break; + case VMW_BINDING_CS_UAV_BIT: + ret = vmw_emit_set_cs_uav(cbs); + break; default: BUG(); } @@ -1089,18 +1211,18 @@ static int vmw_binding_scrub_dx_rt(struct vmw_ctx_bindinfo *bi, bool rebind) } /** - * vmw_binding_scrub_so - Schedule a dx streamoutput buffer binding + * vmw_binding_scrub_so_target - Schedule a dx streamoutput buffer binding * scrub from a context * * @bi: single binding information. * @rebind: Whether to issue a bind instead of scrub command. */ -static int vmw_binding_scrub_so(struct vmw_ctx_bindinfo *bi, bool rebind) +static int vmw_binding_scrub_so_target(struct vmw_ctx_bindinfo *bi, bool rebind) { struct vmw_ctx_binding_state *cbs = vmw_context_binding_state(bi->ctx); - __set_bit(VMW_BINDING_SO_BIT, &cbs->dirty); + __set_bit(VMW_BINDING_SO_T_BIT, &cbs->dirty); return 0; } @@ -1162,6 +1284,49 @@ static int vmw_binding_scrub_ib(struct vmw_ctx_bindinfo *bi, bool rebind) return 0; } +static int vmw_binding_scrub_uav(struct vmw_ctx_bindinfo *bi, bool rebind) +{ + struct vmw_ctx_binding_state *cbs = vmw_context_binding_state(bi->ctx); + + __set_bit(VMW_BINDING_UAV_BIT, &cbs->dirty); + return 0; +} + +static int vmw_binding_scrub_cs_uav(struct vmw_ctx_bindinfo *bi, bool rebind) +{ + struct vmw_ctx_binding_state *cbs = vmw_context_binding_state(bi->ctx); + + __set_bit(VMW_BINDING_CS_UAV_BIT, &cbs->dirty); + return 0; +} + +/** + * vmw_binding_scrub_so - Scrub a streamoutput binding from context. + * @bi: Single binding information. + * @rebind: Whether to issue a bind instead of scrub command. + */ +static int vmw_binding_scrub_so(struct vmw_ctx_bindinfo *bi, bool rebind) +{ + struct vmw_ctx_bindinfo_so *binding = + container_of(bi, typeof(*binding), bi); + struct vmw_private *dev_priv = bi->ctx->dev_priv; + struct { + SVGA3dCmdHeader header; + SVGA3dCmdDXSetStreamOutput body; + } *cmd; + + cmd = VMW_FIFO_RESERVE_DX(dev_priv, sizeof(*cmd), bi->ctx->id); + if (!cmd) + return -ENOMEM; + + cmd->header.id = SVGA_3D_CMD_DX_SET_STREAMOUTPUT; + cmd->header.size = sizeof(cmd->body); + cmd->body.soid = rebind ? bi->res->id : SVGA3D_INVALID_ID; + vmw_fifo_commit(dev_priv, sizeof(*cmd)); + + return 0; +} + /** * vmw_binding_state_alloc - Allocate a struct vmw_ctx_binding_state with * memory accounting. @@ -1248,8 +1413,8 @@ void vmw_binding_state_reset(struct vmw_ctx_binding_state *cbs) * Each time a resource is put on the validation list as the result of a * context binding referencing it, we need to determine whether that resource * will be dirtied (written to by the GPU) as a result of the corresponding - * GPU operation. Currently rendertarget-, depth-stencil-, and - * stream-output-target bindings are capable of dirtying its resource. + * GPU operation. Currently rendertarget-, depth-stencil-, stream-output-target + * and unordered access view bindings are capable of dirtying its resource. * * Return: Whether the binding type dirties the resource its binding points to. */ @@ -1259,11 +1424,13 @@ u32 vmw_binding_dirtying(enum vmw_ctx_binding_type binding_type) [vmw_ctx_binding_rt] = VMW_RES_DIRTY_SET, [vmw_ctx_binding_dx_rt] = VMW_RES_DIRTY_SET, [vmw_ctx_binding_ds] = VMW_RES_DIRTY_SET, - [vmw_ctx_binding_so] = VMW_RES_DIRTY_SET, + [vmw_ctx_binding_so_target] = VMW_RES_DIRTY_SET, + [vmw_ctx_binding_uav] = VMW_RES_DIRTY_SET, + [vmw_ctx_binding_cs_uav] = VMW_RES_DIRTY_SET, }; /* Review this function as new bindings are added. */ - BUILD_BUG_ON(vmw_ctx_binding_max != 11); + BUILD_BUG_ON(vmw_ctx_binding_max != 14); return is_binding_dirtying[binding_type]; } diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_binding.h b/drivers/gpu/drm/vmwgfx/vmwgfx_binding.h index cd9805c045cb..dcb71fd0bb3b 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_binding.h +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_binding.h @@ -33,6 +33,8 @@ #define VMW_MAX_VIEW_BINDINGS 128 +#define VMW_MAX_UAV_BIND_TYPE 2 + struct vmw_private; struct vmw_ctx_binding_state; @@ -48,9 +50,12 @@ enum vmw_ctx_binding_type { vmw_ctx_binding_dx_rt, vmw_ctx_binding_sr, vmw_ctx_binding_ds, - vmw_ctx_binding_so, + vmw_ctx_binding_so_target, vmw_ctx_binding_vb, vmw_ctx_binding_ib, + vmw_ctx_binding_uav, + vmw_ctx_binding_cs_uav, + vmw_ctx_binding_so, vmw_ctx_binding_max }; @@ -128,14 +133,14 @@ struct vmw_ctx_bindinfo_view { }; /** - * struct vmw_ctx_bindinfo_so - StreamOutput binding metadata + * struct vmw_ctx_bindinfo_so_target - StreamOutput binding metadata * * @bi: struct vmw_ctx_bindinfo we derive from. * @offset: Device data used to reconstruct binding command. * @size: Device data used to reconstruct binding command. * @slot: Device data used to reconstruct binding command. */ -struct vmw_ctx_bindinfo_so { +struct vmw_ctx_bindinfo_so_target { struct vmw_ctx_bindinfo bi; uint32 offset; uint32 size; @@ -189,9 +194,31 @@ struct vmw_dx_shader_bindings { unsigned long dirty; }; +/** + * struct vmw_ctx_bindinfo_uav - UAV context binding state. + * @views: UAV view bindings. + * @splice_index: The device splice index set by user-space. + */ +struct vmw_ctx_bindinfo_uav { + struct vmw_ctx_bindinfo_view views[SVGA3D_MAX_UAVIEWS]; + uint32 index; +}; + +/** + * struct vmw_ctx_bindinfo_so - Stream output binding metadata. + * @bi: struct vmw_ctx_bindinfo we derive from. + * @slot: Device data used to reconstruct binding command. + */ +struct vmw_ctx_bindinfo_so { + struct vmw_ctx_bindinfo bi; + uint32 slot; +}; + extern void vmw_binding_add(struct vmw_ctx_binding_state *cbs, const struct vmw_ctx_bindinfo *ci, u32 shader_slot, u32 slot); +extern void vmw_binding_add_uav_index(struct vmw_ctx_binding_state *cbs, + uint32 slot, uint32 splice_index); extern void vmw_binding_state_commit(struct vmw_ctx_binding_state *to, struct vmw_ctx_binding_state *from); diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf.c b/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf.c index 065015d2a8f6..3b41cf63110a 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf.c @@ -1241,7 +1241,8 @@ int vmw_cmdbuf_set_pool_size(struct vmw_cmdbuf_man *man, * actually call into the already enabled manager, when * binding the MOB. */ - if (!(dev_priv->capabilities & SVGA_CAP_DX)) + if (!(dev_priv->capabilities & SVGA_CAP_DX) || + !dev_priv->has_mob) return -ENOMEM; ret = ttm_bo_create(&dev_priv->bdev, size, ttm_bo_type_device, diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_context.c b/drivers/gpu/drm/vmwgfx/vmwgfx_context.c index a56c9d802382..61c246335e66 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_context.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_context.c @@ -36,7 +36,7 @@ struct vmw_user_context { struct vmw_resource res; struct vmw_ctx_binding_state *cbs; struct vmw_cmdbuf_res_manager *man; - struct vmw_resource *cotables[SVGA_COTABLE_DX10_MAX]; + struct vmw_resource *cotables[SVGA_COTABLE_MAX]; spinlock_t cotable_lock; struct vmw_buffer_object *dx_query_mob; }; @@ -116,12 +116,15 @@ static const struct vmw_res_func vmw_dx_context_func = { * Context management: */ -static void vmw_context_cotables_unref(struct vmw_user_context *uctx) +static void vmw_context_cotables_unref(struct vmw_private *dev_priv, + struct vmw_user_context *uctx) { struct vmw_resource *res; int i; + u32 cotable_max = has_sm5_context(dev_priv) ? + SVGA_COTABLE_MAX : SVGA_COTABLE_DX10_MAX; - for (i = 0; i < SVGA_COTABLE_DX10_MAX; ++i) { + for (i = 0; i < cotable_max; ++i) { spin_lock(&uctx->cotable_lock); res = uctx->cotables[i]; uctx->cotables[i] = NULL; @@ -155,7 +158,7 @@ static void vmw_hw_context_destroy(struct vmw_resource *res) !dev_priv->query_cid_valid) __vmw_execbuf_release_pinned_bo(dev_priv, NULL); mutex_unlock(&dev_priv->cmdbuf_mutex); - vmw_context_cotables_unref(uctx); + vmw_context_cotables_unref(dev_priv, uctx); return; } @@ -208,7 +211,9 @@ static int vmw_gb_context_init(struct vmw_private *dev_priv, spin_lock_init(&uctx->cotable_lock); if (dx) { - for (i = 0; i < SVGA_COTABLE_DX10_MAX; ++i) { + u32 cotable_max = has_sm5_context(dev_priv) ? + SVGA_COTABLE_MAX : SVGA_COTABLE_DX10_MAX; + for (i = 0; i < cotable_max; ++i) { uctx->cotables[i] = vmw_cotable_alloc(dev_priv, &uctx->res, i); if (IS_ERR(uctx->cotables[i])) { @@ -222,7 +227,7 @@ static int vmw_gb_context_init(struct vmw_private *dev_priv, return 0; out_cotables: - vmw_context_cotables_unref(uctx); + vmw_context_cotables_unref(dev_priv, uctx); out_err: if (res_free) res_free(res); @@ -545,10 +550,12 @@ void vmw_dx_context_scrub_cotables(struct vmw_resource *ctx, { struct vmw_user_context *uctx = container_of(ctx, struct vmw_user_context, res); + u32 cotable_max = has_sm5_context(ctx->dev_priv) ? + SVGA_COTABLE_MAX : SVGA_COTABLE_DX10_MAX; int i; vmw_binding_state_scrub(uctx->cbs); - for (i = 0; i < SVGA_COTABLE_DX10_MAX; ++i) { + for (i = 0; i < cotable_max; ++i) { struct vmw_resource *res; /* Avoid racing with ongoing cotable destruction. */ @@ -731,7 +738,7 @@ static int vmw_context_define(struct drm_device *dev, void *data, }; int ret; - if (!dev_priv->has_dx && dx) { + if (!has_sm4_context(dev_priv) && dx) { VMW_DEBUG_USER("DX contexts not supported by device.\n"); return -EINVAL; } @@ -839,7 +846,10 @@ struct vmw_cmdbuf_res_manager *vmw_context_res_man(struct vmw_resource *ctx) struct vmw_resource *vmw_context_cotable(struct vmw_resource *ctx, SVGACOTableType cotable_type) { - if (cotable_type >= SVGA_COTABLE_DX10_MAX) + u32 cotable_max = has_sm5_context(ctx->dev_priv) ? + SVGA_COTABLE_MAX : SVGA_COTABLE_DX10_MAX; + + if (cotable_type >= cotable_max) return ERR_PTR(-EINVAL); return container_of(ctx, struct vmw_user_context, res)-> diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_cotable.c b/drivers/gpu/drm/vmwgfx/vmwgfx_cotable.c index 3ca5cf375b01..65e8e7a97724 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_cotable.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_cotable.c @@ -80,9 +80,10 @@ static const struct vmw_cotable_info co_info[] = { {1, sizeof(SVGACOTableDXDepthStencilEntry), NULL}, {1, sizeof(SVGACOTableDXRasterizerStateEntry), NULL}, {1, sizeof(SVGACOTableDXSamplerEntry), NULL}, - {1, sizeof(SVGACOTableDXStreamOutputEntry), NULL}, + {1, sizeof(SVGACOTableDXStreamOutputEntry), &vmw_dx_streamoutput_cotable_list_scrub}, {1, sizeof(SVGACOTableDXQueryEntry), NULL}, - {1, sizeof(SVGACOTableDXShaderEntry), &vmw_dx_shader_cotable_list_scrub} + {1, sizeof(SVGACOTableDXShaderEntry), &vmw_dx_shader_cotable_list_scrub}, + {1, sizeof(SVGACOTableDXUAViewEntry), &vmw_view_cotable_list_destroy} }; /* @@ -102,6 +103,7 @@ const SVGACOTableType vmw_cotable_scrub_order[] = { SVGA_COTABLE_SAMPLER, SVGA_COTABLE_STREAMOUTPUT, SVGA_COTABLE_DXQUERY, + SVGA_COTABLE_UAVIEW, }; static int vmw_cotable_bind(struct vmw_resource *res, diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c index cf3dc56d7cf4..c2247a893ed4 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c @@ -29,6 +29,7 @@ #include <linux/dma-mapping.h> #include <linux/module.h> #include <linux/pci.h> +#include <linux/mem_encrypt.h> #include <drm/drm_drv.h> #include <drm/drm_ioctl.h> @@ -289,6 +290,8 @@ static void vmw_print_capabilities2(uint32_t capabilities2) DRM_INFO(" Grow oTable.\n"); if (capabilities2 & SVGA_CAP2_INTRA_SURFACE_COPY) DRM_INFO(" IntraSurface copy.\n"); + if (capabilities2 & SVGA_CAP2_DX3) + DRM_INFO(" DX3.\n"); } static void vmw_print_capabilities(uint32_t capabilities) @@ -448,7 +451,7 @@ static int vmw_request_device(struct vmw_private *dev_priv) dev_priv->cman = vmw_cmdbuf_man_create(dev_priv); if (IS_ERR(dev_priv->cman)) { dev_priv->cman = NULL; - dev_priv->has_dx = false; + dev_priv->sm_type = VMW_SM_LEGACY; } ret = vmw_request_device_late(dev_priv); @@ -575,6 +578,10 @@ static int vmw_dma_select_mode(struct vmw_private *dev_priv) [vmw_dma_map_populate] = "Caching DMA mappings.", [vmw_dma_map_bind] = "Giving up DMA mappings early."}; + /* TTM currently doesn't fully support SEV encryption. */ + if (mem_encrypt_active()) + return -EINVAL; + if (vmw_force_coherent) dev_priv->map_mode = vmw_dma_alloc_coherent; else if (vmw_restrict_iommu) @@ -682,8 +689,10 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset) ret = vmw_dma_select_mode(dev_priv); if (unlikely(ret != 0)) { - DRM_INFO("Restricting capabilities due to IOMMU setup.\n"); + DRM_INFO("Restricting capabilities since DMA not available.\n"); refuse_dma = true; + if (dev_priv->capabilities & SVGA_CAP_GBOBJECTS) + DRM_INFO("Disabling 3D acceleration.\n"); } dev_priv->vram_size = vmw_read(dev_priv, SVGA_REG_VRAM_SIZE); @@ -711,9 +720,15 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset) dev_priv->max_mob_pages = 0; dev_priv->max_mob_size = 0; if (dev_priv->capabilities & SVGA_CAP_GBOBJECTS) { - uint64_t mem_size = - vmw_read(dev_priv, - SVGA_REG_SUGGESTED_GBOBJECT_MEM_SIZE_KB); + uint64_t mem_size; + + if (dev_priv->capabilities2 & SVGA_CAP2_GB_MEMSIZE_2) + mem_size = vmw_read(dev_priv, + SVGA_REG_GBOBJECT_MEM_SIZE_KB); + else + mem_size = + vmw_read(dev_priv, + SVGA_REG_SUGGESTED_GBOBJECT_MEM_SIZE_KB); /* * Workaround for low memory 2D VMs to compensate for the @@ -866,7 +881,7 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset) dev_priv->has_gmr = false; } - if (dev_priv->capabilities & SVGA_CAP_GBOBJECTS) { + if (dev_priv->capabilities & SVGA_CAP_GBOBJECTS && !refuse_dma) { dev_priv->has_mob = true; if (ttm_bo_init_mm(&dev_priv->bdev, VMW_PL_MOB, VMW_PL_MOB) != 0) { @@ -876,14 +891,32 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset) } } - if (dev_priv->has_mob) { + if (dev_priv->has_mob && (dev_priv->capabilities & SVGA_CAP_DX)) { spin_lock(&dev_priv->cap_lock); vmw_write(dev_priv, SVGA_REG_DEV_CAP, SVGA3D_DEVCAP_DXCONTEXT); - dev_priv->has_dx = !!vmw_read(dev_priv, SVGA_REG_DEV_CAP); + if (vmw_read(dev_priv, SVGA_REG_DEV_CAP)) + dev_priv->sm_type = VMW_SM_4; spin_unlock(&dev_priv->cap_lock); } vmw_validation_mem_init_ttm(dev_priv, VMWGFX_VALIDATION_MEM_GRAN); + + /* SVGA_CAP2_DX2 (DefineGBSurface_v3) is needed for SM4_1 support */ + if (has_sm4_context(dev_priv) && + (dev_priv->capabilities2 & SVGA_CAP2_DX2)) { + vmw_write(dev_priv, SVGA_REG_DEV_CAP, SVGA3D_DEVCAP_SM41); + + if (vmw_read(dev_priv, SVGA_REG_DEV_CAP)) + dev_priv->sm_type = VMW_SM_4_1; + + if (has_sm4_1_context(dev_priv) && + (dev_priv->capabilities2 & SVGA_CAP2_DX3)) { + vmw_write(dev_priv, SVGA_REG_DEV_CAP, SVGA3D_DEVCAP_SM5); + if (vmw_read(dev_priv, SVGA_REG_DEV_CAP)) + dev_priv->sm_type = VMW_SM_5; + } + } + ret = vmw_kms_init(dev_priv); if (unlikely(ret != 0)) goto out_no_kms; @@ -893,23 +926,14 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset) if (ret) goto out_no_fifo; - if (dev_priv->has_dx) { - /* - * SVGA_CAP2_DX2 (DefineGBSurface_v3) is needed for SM4_1 - * support - */ - if ((dev_priv->capabilities2 & SVGA_CAP2_DX2) != 0) { - vmw_write(dev_priv, SVGA_REG_DEV_CAP, - SVGA3D_DEVCAP_SM41); - dev_priv->has_sm4_1 = vmw_read(dev_priv, - SVGA_REG_DEV_CAP); - } - } - - DRM_INFO("DX: %s\n", dev_priv->has_dx ? "yes." : "no."); DRM_INFO("Atomic: %s\n", (dev->driver->driver_features & DRIVER_ATOMIC) ? "yes." : "no."); - DRM_INFO("SM4_1: %s\n", dev_priv->has_sm4_1 ? "yes." : "no."); + if (dev_priv->sm_type == VMW_SM_5) + DRM_INFO("SM5 support available.\n"); + if (dev_priv->sm_type == VMW_SM_4_1) + DRM_INFO("SM4_1 support available.\n"); + if (dev_priv->sm_type == VMW_SM_4) + DRM_INFO("SM4 support available.\n"); snprintf(host_log, sizeof(host_log), "vmwgfx: %s-%s", VMWGFX_REPO, VMWGFX_GIT_VERSION); @@ -1223,6 +1247,18 @@ static void vmw_remove(struct pci_dev *pdev) pci_disable_device(pdev); } +static unsigned long +vmw_get_unmapped_area(struct file *file, unsigned long uaddr, + unsigned long len, unsigned long pgoff, + unsigned long flags) +{ + struct drm_file *file_priv = file->private_data; + struct vmw_private *dev_priv = vmw_priv(file_priv->minor->dev); + + return drm_get_unmapped_area(file, uaddr, len, pgoff, flags, + &dev_priv->vma_manager); +} + static int vmwgfx_pm_notifier(struct notifier_block *nb, unsigned long val, void *ptr) { @@ -1394,6 +1430,7 @@ static const struct file_operations vmwgfx_driver_fops = { .compat_ioctl = vmw_compat_ioctl, #endif .llseek = noop_llseek, + .get_unmapped_area = vmw_get_unmapped_area, }; static struct drm_driver driver = { diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h index 4ce39c94c0eb..8cdcd6e5f9e1 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h @@ -58,7 +58,7 @@ #define VMWGFX_DRIVER_NAME "vmwgfx" #define VMWGFX_DRIVER_DATE "20200114" #define VMWGFX_DRIVER_MAJOR 2 -#define VMWGFX_DRIVER_MINOR 17 +#define VMWGFX_DRIVER_MINOR 18 #define VMWGFX_DRIVER_PATCHLEVEL 0 #define VMWGFX_FIFO_STATIC_SIZE (1024*1024) #define VMWGFX_MAX_RELOCATIONS 2048 @@ -202,6 +202,7 @@ enum vmw_res_type { vmw_res_dx_context, vmw_res_cotable, vmw_res_view, + vmw_res_streamoutput, vmw_res_max }; @@ -210,7 +211,8 @@ enum vmw_res_type { */ enum vmw_cmdbuf_res_type { vmw_cmdbuf_res_shader, - vmw_cmdbuf_res_view + vmw_cmdbuf_res_view, + vmw_cmdbuf_res_streamoutput }; struct vmw_cmdbuf_res_manager; @@ -223,24 +225,58 @@ struct vmw_cursor_snooper { struct vmw_framebuffer; struct vmw_surface_offset; -struct vmw_surface { - struct vmw_resource res; - SVGA3dSurfaceAllFlags flags; - uint32_t format; - uint32_t mip_levels[DRM_VMW_MAX_SURFACE_FACES]; +/** + * struct vmw_surface_metadata - Metadata describing a surface. + * + * @flags: Device flags. + * @format: Surface SVGA3D_x format. + * @mip_levels: Mip level for each face. For GB first index is used only. + * @multisample_count: Sample count. + * @multisample_pattern: Sample patterns. + * @quality_level: Quality level. + * @autogen_filter: Filter for automatically generated mipmaps. + * @array_size: Number of array elements for a 1D/2D texture. For cubemap + texture number of faces * array_size. This should be 0 for pre + SM4 device. + * @buffer_byte_stride: Buffer byte stride. + * @num_sizes: Size of @sizes. For GB surface this should always be 1. + * @base_size: Surface dimension. + * @sizes: Array representing mip sizes. Legacy only. + * @scanout: Whether this surface will be used for scanout. + * + * This tracks metadata for both legacy and guest backed surface. + */ +struct vmw_surface_metadata { + u64 flags; + u32 format; + u32 mip_levels[DRM_VMW_MAX_SURFACE_FACES]; + u32 multisample_count; + u32 multisample_pattern; + u32 quality_level; + u32 autogen_filter; + u32 array_size; + u32 num_sizes; + u32 buffer_byte_stride; struct drm_vmw_size base_size; struct drm_vmw_size *sizes; - uint32_t num_sizes; bool scanout; - uint32_t array_size; - /* TODO so far just a extra pointer */ +}; + +/** + * struct vmw_surface: Resource structure for a surface. + * + * @res: The base resource for this surface. + * @metadata: Metadata for this surface resource. + * @snooper: Cursor data. Legacy surface only. + * @offsets: Legacy surface only. + * @view_list: List of views bound to this surface. + */ +struct vmw_surface { + struct vmw_resource res; + struct vmw_surface_metadata metadata; struct vmw_cursor_snooper snooper; struct vmw_surface_offset *offsets; - SVGA3dTextureFilter autogen_filter; - uint32_t multisample_count; struct list_head view_list; - SVGA3dMSPattern multisample_pattern; - SVGA3dMSQualityLevel quality_level; }; struct vmw_marker_queue { @@ -441,6 +477,22 @@ enum { VMW_IRQTHREAD_MAX }; +/** + * enum vmw_sm_type - Graphics context capability supported by device. + * @VMW_SM_LEGACY: Pre DX context. + * @VMW_SM_4: Context support upto SM4. + * @VMW_SM_4_1: Context support upto SM4_1. + * @VMW_SM_5: Context support up to SM5. + * @VMW_SM_MAX: Should be the last. + */ +enum vmw_sm_type { + VMW_SM_LEGACY = 0, + VMW_SM_4, + VMW_SM_4_1, + VMW_SM_5, + VMW_SM_MAX +}; + struct vmw_private { struct ttm_bo_device bdev; @@ -475,22 +527,9 @@ struct vmw_private { bool has_mob; spinlock_t hw_lock; spinlock_t cap_lock; - bool has_dx; bool assume_16bpp; - bool has_sm4_1; - - /* - * VGA registers. - */ - - struct vmw_vga_topology_state vga_save[VMWGFX_MAX_DISPLAYS]; - uint32_t vga_width; - uint32_t vga_height; - uint32_t vga_bpp; - uint32_t vga_bpl; - uint32_t vga_pitchlock; - uint32_t num_displays; + enum vmw_sm_type sm_type; /* * Framebuffer info. @@ -661,6 +700,39 @@ static inline uint32_t vmw_read(struct vmw_private *dev_priv, return val; } +/** + * has_sm4_context - Does the device support SM4 context. + * @dev_priv: Device private. + * + * Return: Bool value if device support SM4 context or not. + */ +static inline bool has_sm4_context(const struct vmw_private *dev_priv) +{ + return (dev_priv->sm_type >= VMW_SM_4); +} + +/** + * has_sm4_1_context - Does the device support SM4_1 context. + * @dev_priv: Device private. + * + * Return: Bool value if device support SM4_1 context or not. + */ +static inline bool has_sm4_1_context(const struct vmw_private *dev_priv) +{ + return (dev_priv->sm_type >= VMW_SM_4_1); +} + +/** + * has_sm5_context - Does the device support SM5 context. + * @dev_priv: Device private. + * + * Return: Bool value if device support SM5 context or not. + */ +static inline bool has_sm5_context(const struct vmw_private *dev_priv) +{ + return (dev_priv->sm_type >= VMW_SM_5); +} + extern void vmw_svga_enable(struct vmw_private *dev_priv); extern void vmw_svga_disable(struct vmw_private *dev_priv); @@ -900,7 +972,6 @@ extern void vmw_fifo_commit(struct vmw_private *dev_priv, uint32_t bytes); extern void vmw_fifo_commit_flush(struct vmw_private *dev_priv, uint32_t bytes); extern int vmw_fifo_send_fence(struct vmw_private *dev_priv, uint32_t *seqno); -extern void vmw_fifo_ping_host_locked(struct vmw_private *, uint32_t reason); extern void vmw_fifo_ping_host(struct vmw_private *dev_priv, uint32_t reason); extern bool vmw_fifo_have_3d(struct vmw_private *dev_priv); extern bool vmw_fifo_have_pitchlock(struct vmw_private *dev_priv); @@ -929,6 +1000,7 @@ extern int vmw_mmap(struct file *filp, struct vm_area_struct *vma); extern void vmw_validation_mem_init_ttm(struct vmw_private *dev_priv, size_t gran); + /** * TTM buffer object driver - vmwgfx_ttm_buffer.c */ @@ -947,7 +1019,6 @@ extern struct ttm_placement vmw_mob_placement; extern struct ttm_placement vmw_mob_ne_placement; extern struct ttm_placement vmw_nonfixed_placement; extern struct ttm_bo_driver vmw_bo_driver; -extern int vmw_dma_quiescent(struct drm_device *dev); extern int vmw_bo_map_dma(struct ttm_buffer_object *bo); extern void vmw_bo_unmap_dma(struct ttm_buffer_object *bo); extern const struct vmw_sg_table * @@ -1085,8 +1156,6 @@ int vmw_fb_on(struct vmw_private *vmw_priv); int vmw_kms_init(struct vmw_private *dev_priv); int vmw_kms_close(struct vmw_private *dev_priv); -int vmw_kms_save_vga(struct vmw_private *vmw_priv); -int vmw_kms_restore_vga(struct vmw_private *vmw_priv); int vmw_kms_cursor_bypass_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv); void vmw_kms_cursor_post_execbuf(struct vmw_private *dev_priv); @@ -1139,7 +1208,6 @@ int vmw_overlay_init(struct vmw_private *dev_priv); int vmw_overlay_close(struct vmw_private *dev_priv); int vmw_overlay_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv); -int vmw_overlay_stop_all(struct vmw_private *dev_priv); int vmw_overlay_resume_all(struct vmw_private *dev_priv); int vmw_overlay_pause_all(struct vmw_private *dev_priv); int vmw_overlay_claim(struct vmw_private *dev_priv, uint32_t *out); @@ -1186,10 +1254,6 @@ extern void vmw_otables_takedown(struct vmw_private *dev_priv); extern const struct vmw_user_resource_conv *user_context_converter; -extern int vmw_context_check(struct vmw_private *dev_priv, - struct ttm_object_file *tfile, - int id, - struct vmw_resource **p_res); extern int vmw_context_define_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv); extern int vmw_extended_context_define_ioctl(struct drm_device *dev, void *data, @@ -1219,7 +1283,6 @@ vmw_context_get_dx_query_mob(struct vmw_resource *ctx_res); extern const struct vmw_user_resource_conv *user_surface_converter; -extern void vmw_surface_res_free(struct vmw_resource *res); extern int vmw_surface_destroy_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv); extern int vmw_surface_define_ioctl(struct drm_device *dev, void *data, @@ -1230,11 +1293,6 @@ extern int vmw_gb_surface_define_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv); extern int vmw_gb_surface_reference_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv); -extern int vmw_surface_check(struct vmw_private *dev_priv, - struct ttm_object_file *tfile, - uint32_t handle, int *id); -extern int vmw_surface_validate(struct vmw_private *dev_priv, - struct vmw_surface *srf); int vmw_surface_gb_priv_define(struct drm_device *dev, uint32_t user_accounting_size, SVGA3dSurfaceAllFlags svga3d_flags, @@ -1254,6 +1312,11 @@ extern int vmw_gb_surface_reference_ext_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv); +int vmw_gb_surface_define(struct vmw_private *dev_priv, + uint32_t user_accounting_size, + const struct vmw_surface_metadata *req, + struct vmw_surface **srf_out); + /* * Shader management - vmwgfx_shader.c */ @@ -1287,6 +1350,24 @@ vmw_shader_lookup(struct vmw_cmdbuf_res_manager *man, u32 user_key, SVGA3dShaderType shader_type); /* + * Streamoutput management + */ +struct vmw_resource * +vmw_dx_streamoutput_lookup(struct vmw_cmdbuf_res_manager *man, + u32 user_key); +int vmw_dx_streamoutput_add(struct vmw_cmdbuf_res_manager *man, + struct vmw_resource *ctx, + SVGA3dStreamOutputId user_key, + struct list_head *list); +void vmw_dx_streamoutput_set_size(struct vmw_resource *res, u32 size); +int vmw_dx_streamoutput_remove(struct vmw_cmdbuf_res_manager *man, + SVGA3dStreamOutputId user_key, + struct list_head *list); +void vmw_dx_streamoutput_cotable_list_scrub(struct vmw_private *dev_priv, + struct list_head *list, + bool readback); + +/* * Command buffer managed resources - vmwgfx_cmdbuf_res.c */ @@ -1430,6 +1511,17 @@ void vmw_bo_dirty_unmap(struct vmw_buffer_object *vbo, pgoff_t start, pgoff_t end); vm_fault_t vmw_bo_vm_fault(struct vm_fault *vmf); vm_fault_t vmw_bo_vm_mkwrite(struct vm_fault *vmf); +#ifdef CONFIG_TRANSPARENT_HUGEPAGE +vm_fault_t vmw_bo_vm_huge_fault(struct vm_fault *vmf, + enum page_entry_size pe_size); +#endif + +/* Transparent hugepage support - vmwgfx_thp.c */ +#ifdef CONFIG_TRANSPARENT_HUGEPAGE +extern const struct ttm_mem_type_manager_func vmw_thp_func; +#else +#define vmw_thp_func ttm_bo_manager_func +#endif /** * VMW_DEBUG_KMS - Debug output for kernel mode-setting diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c index 73489a45decb..367d5b87ee6a 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c @@ -459,10 +459,13 @@ static int vmw_resource_context_res_add(struct vmw_private *dev_priv, int ret = 0; struct vmw_resource *res; u32 i; + u32 cotable_max = has_sm5_context(ctx->dev_priv) ? + SVGA_COTABLE_MAX : SVGA_COTABLE_DX10_MAX; /* Add all cotables to the validation list. */ - if (dev_priv->has_dx && vmw_res_type(ctx) == vmw_res_dx_context) { - for (i = 0; i < SVGA_COTABLE_DX10_MAX; ++i) { + if (has_sm4_context(dev_priv) && + vmw_res_type(ctx) == vmw_res_dx_context) { + for (i = 0; i < cotable_max; ++i) { res = vmw_context_cotable(ctx, i); if (IS_ERR(res)) continue; @@ -489,7 +492,8 @@ static int vmw_resource_context_res_add(struct vmw_private *dev_priv, break; } - if (dev_priv->has_dx && vmw_res_type(ctx) == vmw_res_dx_context) { + if (has_sm4_context(dev_priv) && + vmw_res_type(ctx) == vmw_res_dx_context) { struct vmw_buffer_object *dx_query_mob; dx_query_mob = vmw_context_get_dx_query_mob(ctx); @@ -2116,6 +2120,9 @@ vmw_cmd_dx_set_single_constant_buffer(struct vmw_private *dev_priv, SVGA3dCmdHeader *header) { VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXSetSingleConstantBuffer); + SVGA3dShaderType max_shader_num = has_sm5_context(dev_priv) ? + SVGA3D_NUM_SHADERTYPE : SVGA3D_NUM_SHADERTYPE_DX10; + struct vmw_resource *res = NULL; struct vmw_ctx_validation_info *ctx_node = VMW_GET_CTX_NODE(sw_context); struct vmw_ctx_bindinfo_cb binding; @@ -2139,7 +2146,7 @@ vmw_cmd_dx_set_single_constant_buffer(struct vmw_private *dev_priv, binding.size = cmd->body.sizeInBytes; binding.slot = cmd->body.slot; - if (binding.shader_slot >= SVGA3D_NUM_SHADERTYPE_DX10 || + if (binding.shader_slot >= max_shader_num || binding.slot >= SVGA3D_DX_MAX_CONSTBUFFERS) { VMW_DEBUG_USER("Illegal const buffer shader %u slot %u.\n", (unsigned int) cmd->body.type, @@ -2167,12 +2174,15 @@ static int vmw_cmd_dx_set_shader_res(struct vmw_private *dev_priv, { VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXSetShaderResources) = container_of(header, typeof(*cmd), header); + SVGA3dShaderType max_allowed = has_sm5_context(dev_priv) ? + SVGA3D_SHADERTYPE_MAX : SVGA3D_SHADERTYPE_DX10_MAX; + u32 num_sr_view = (cmd->header.size - sizeof(cmd->body)) / sizeof(SVGA3dShaderResourceViewId); if ((u64) cmd->body.startView + (u64) num_sr_view > (u64) SVGA3D_DX_MAX_SRVIEWS || - cmd->body.type >= SVGA3D_SHADERTYPE_DX10_MAX) { + cmd->body.type >= max_allowed) { VMW_DEBUG_USER("Invalid shader binding.\n"); return -EINVAL; } @@ -2196,6 +2206,8 @@ static int vmw_cmd_dx_set_shader(struct vmw_private *dev_priv, SVGA3dCmdHeader *header) { VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXSetShader); + SVGA3dShaderType max_allowed = has_sm5_context(dev_priv) ? + SVGA3D_SHADERTYPE_MAX : SVGA3D_SHADERTYPE_DX10_MAX; struct vmw_resource *res = NULL; struct vmw_ctx_validation_info *ctx_node = VMW_GET_CTX_NODE(sw_context); struct vmw_ctx_bindinfo_shader binding; @@ -2206,7 +2218,7 @@ static int vmw_cmd_dx_set_shader(struct vmw_private *dev_priv, cmd = container_of(header, typeof(*cmd), header); - if (cmd->body.type >= SVGA3D_SHADERTYPE_DX10_MAX || + if (cmd->body.type >= max_allowed || cmd->body.type < SVGA3D_SHADERTYPE_MIN) { VMW_DEBUG_USER("Illegal shader type %u.\n", (unsigned int) cmd->body.type); @@ -2467,7 +2479,7 @@ static int vmw_cmd_dx_set_so_targets(struct vmw_private *dev_priv, SVGA3dCmdHeader *header) { struct vmw_ctx_validation_info *ctx_node = VMW_GET_CTX_NODE(sw_context); - struct vmw_ctx_bindinfo_so binding; + struct vmw_ctx_bindinfo_so_target binding; struct vmw_resource *res; struct { SVGA3dCmdHeader header; @@ -2497,7 +2509,7 @@ static int vmw_cmd_dx_set_so_targets(struct vmw_private *dev_priv, binding.bi.ctx = ctx_node->ctx; binding.bi.res = res; - binding.bi.bt = vmw_ctx_binding_so, + binding.bi.bt = vmw_ctx_binding_so_target, binding.offset = cmd->targets[i].offset; binding.size = cmd->targets[i].sizeInBytes; binding.slot = i; @@ -2804,6 +2816,352 @@ static int vmw_cmd_intra_surface_copy(struct vmw_private *dev_priv, &cmd->body.surface.sid, NULL); } +static int vmw_cmd_sm5(struct vmw_private *dev_priv, + struct vmw_sw_context *sw_context, + SVGA3dCmdHeader *header) +{ + if (!has_sm5_context(dev_priv)) + return -EINVAL; + + return 0; +} + +static int vmw_cmd_sm5_view_define(struct vmw_private *dev_priv, + struct vmw_sw_context *sw_context, + SVGA3dCmdHeader *header) +{ + if (!has_sm5_context(dev_priv)) + return -EINVAL; + + return vmw_cmd_dx_view_define(dev_priv, sw_context, header); +} + +static int vmw_cmd_sm5_view_remove(struct vmw_private *dev_priv, + struct vmw_sw_context *sw_context, + SVGA3dCmdHeader *header) +{ + if (!has_sm5_context(dev_priv)) + return -EINVAL; + + return vmw_cmd_dx_view_remove(dev_priv, sw_context, header); +} + +static int vmw_cmd_clear_uav_uint(struct vmw_private *dev_priv, + struct vmw_sw_context *sw_context, + SVGA3dCmdHeader *header) +{ + struct { + SVGA3dCmdHeader header; + SVGA3dCmdDXClearUAViewUint body; + } *cmd = container_of(header, typeof(*cmd), header); + struct vmw_resource *ret; + + if (!has_sm5_context(dev_priv)) + return -EINVAL; + + ret = vmw_view_id_val_add(sw_context, vmw_view_ua, + cmd->body.uaViewId); + + return PTR_ERR_OR_ZERO(ret); +} + +static int vmw_cmd_clear_uav_float(struct vmw_private *dev_priv, + struct vmw_sw_context *sw_context, + SVGA3dCmdHeader *header) +{ + struct { + SVGA3dCmdHeader header; + SVGA3dCmdDXClearUAViewFloat body; + } *cmd = container_of(header, typeof(*cmd), header); + struct vmw_resource *ret; + + if (!has_sm5_context(dev_priv)) + return -EINVAL; + + ret = vmw_view_id_val_add(sw_context, vmw_view_ua, + cmd->body.uaViewId); + + return PTR_ERR_OR_ZERO(ret); +} + +static int vmw_cmd_set_uav(struct vmw_private *dev_priv, + struct vmw_sw_context *sw_context, + SVGA3dCmdHeader *header) +{ + struct { + SVGA3dCmdHeader header; + SVGA3dCmdDXSetUAViews body; + } *cmd = container_of(header, typeof(*cmd), header); + u32 num_uav = (cmd->header.size - sizeof(cmd->body)) / + sizeof(SVGA3dUAViewId); + int ret; + + if (!has_sm5_context(dev_priv)) + return -EINVAL; + + if (num_uav > SVGA3D_MAX_UAVIEWS) { + VMW_DEBUG_USER("Invalid UAV binding.\n"); + return -EINVAL; + } + + ret = vmw_view_bindings_add(sw_context, vmw_view_ua, + vmw_ctx_binding_uav, 0, (void *)&cmd[1], + num_uav, 0); + if (ret) + return ret; + + vmw_binding_add_uav_index(sw_context->dx_ctx_node->staged, 0, + cmd->body.uavSpliceIndex); + + return ret; +} + +static int vmw_cmd_set_cs_uav(struct vmw_private *dev_priv, + struct vmw_sw_context *sw_context, + SVGA3dCmdHeader *header) +{ + struct { + SVGA3dCmdHeader header; + SVGA3dCmdDXSetCSUAViews body; + } *cmd = container_of(header, typeof(*cmd), header); + u32 num_uav = (cmd->header.size - sizeof(cmd->body)) / + sizeof(SVGA3dUAViewId); + int ret; + + if (!has_sm5_context(dev_priv)) + return -EINVAL; + + if (num_uav > SVGA3D_MAX_UAVIEWS) { + VMW_DEBUG_USER("Invalid UAV binding.\n"); + return -EINVAL; + } + + ret = vmw_view_bindings_add(sw_context, vmw_view_ua, + vmw_ctx_binding_cs_uav, 0, (void *)&cmd[1], + num_uav, 0); + if (ret) + return ret; + + vmw_binding_add_uav_index(sw_context->dx_ctx_node->staged, 1, + cmd->body.startIndex); + + return ret; +} + +static int vmw_cmd_dx_define_streamoutput(struct vmw_private *dev_priv, + struct vmw_sw_context *sw_context, + SVGA3dCmdHeader *header) +{ + struct vmw_ctx_validation_info *ctx_node = sw_context->dx_ctx_node; + struct vmw_resource *res; + struct { + SVGA3dCmdHeader header; + SVGA3dCmdDXDefineStreamOutputWithMob body; + } *cmd = container_of(header, typeof(*cmd), header); + int ret; + + if (!has_sm5_context(dev_priv)) + return -EINVAL; + + if (!ctx_node) { + DRM_ERROR("DX Context not set.\n"); + return -EINVAL; + } + + res = vmw_context_cotable(ctx_node->ctx, SVGA_COTABLE_STREAMOUTPUT); + ret = vmw_cotable_notify(res, cmd->body.soid); + if (ret) + return ret; + + return vmw_dx_streamoutput_add(sw_context->man, ctx_node->ctx, + cmd->body.soid, + &sw_context->staged_cmd_res); +} + +static int vmw_cmd_dx_destroy_streamoutput(struct vmw_private *dev_priv, + struct vmw_sw_context *sw_context, + SVGA3dCmdHeader *header) +{ + struct vmw_ctx_validation_info *ctx_node = sw_context->dx_ctx_node; + struct vmw_resource *res; + struct { + SVGA3dCmdHeader header; + SVGA3dCmdDXDestroyStreamOutput body; + } *cmd = container_of(header, typeof(*cmd), header); + + if (!ctx_node) { + DRM_ERROR("DX Context not set.\n"); + return -EINVAL; + } + + /* + * When device does not support SM5 then streamoutput with mob command is + * not available to user-space. Simply return in this case. + */ + if (!has_sm5_context(dev_priv)) + return 0; + + /* + * With SM5 capable device if lookup fails then user-space probably used + * old streamoutput define command. Return without an error. + */ + res = vmw_dx_streamoutput_lookup(vmw_context_res_man(ctx_node->ctx), + cmd->body.soid); + if (IS_ERR(res)) + return 0; + + return vmw_dx_streamoutput_remove(sw_context->man, cmd->body.soid, + &sw_context->staged_cmd_res); +} + +static int vmw_cmd_dx_bind_streamoutput(struct vmw_private *dev_priv, + struct vmw_sw_context *sw_context, + SVGA3dCmdHeader *header) +{ + struct vmw_ctx_validation_info *ctx_node = sw_context->dx_ctx_node; + struct vmw_resource *res; + struct { + SVGA3dCmdHeader header; + SVGA3dCmdDXBindStreamOutput body; + } *cmd = container_of(header, typeof(*cmd), header); + int ret; + + if (!has_sm5_context(dev_priv)) + return -EINVAL; + + if (!ctx_node) { + DRM_ERROR("DX Context not set.\n"); + return -EINVAL; + } + + res = vmw_dx_streamoutput_lookup(vmw_context_res_man(ctx_node->ctx), + cmd->body.soid); + if (IS_ERR(res)) { + DRM_ERROR("Cound not find streamoutput to bind.\n"); + return PTR_ERR(res); + } + + vmw_dx_streamoutput_set_size(res, cmd->body.sizeInBytes); + + ret = vmw_execbuf_res_noctx_val_add(sw_context, res, + VMW_RES_DIRTY_NONE); + if (ret) { + DRM_ERROR("Error creating resource validation node.\n"); + return ret; + } + + return vmw_cmd_res_switch_backup(dev_priv, sw_context, res, + &cmd->body.mobid, + cmd->body.offsetInBytes); +} + +static int vmw_cmd_dx_set_streamoutput(struct vmw_private *dev_priv, + struct vmw_sw_context *sw_context, + SVGA3dCmdHeader *header) +{ + struct vmw_ctx_validation_info *ctx_node = sw_context->dx_ctx_node; + struct vmw_resource *res; + struct vmw_ctx_bindinfo_so binding; + struct { + SVGA3dCmdHeader header; + SVGA3dCmdDXSetStreamOutput body; + } *cmd = container_of(header, typeof(*cmd), header); + int ret; + + if (!ctx_node) { + DRM_ERROR("DX Context not set.\n"); + return -EINVAL; + } + + if (cmd->body.soid == SVGA3D_INVALID_ID) + return 0; + + /* + * When device does not support SM5 then streamoutput with mob command is + * not available to user-space. Simply return in this case. + */ + if (!has_sm5_context(dev_priv)) + return 0; + + /* + * With SM5 capable device if lookup fails then user-space probably used + * old streamoutput define command. Return without an error. + */ + res = vmw_dx_streamoutput_lookup(vmw_context_res_man(ctx_node->ctx), + cmd->body.soid); + if (IS_ERR(res)) { + return 0; + } + + ret = vmw_execbuf_res_noctx_val_add(sw_context, res, + VMW_RES_DIRTY_NONE); + if (ret) { + DRM_ERROR("Error creating resource validation node.\n"); + return ret; + } + + binding.bi.ctx = ctx_node->ctx; + binding.bi.res = res; + binding.bi.bt = vmw_ctx_binding_so; + binding.slot = 0; /* Only one SO set to context at a time. */ + + vmw_binding_add(sw_context->dx_ctx_node->staged, &binding.bi, 0, + binding.slot); + + return ret; +} + +static int vmw_cmd_indexed_instanced_indirect(struct vmw_private *dev_priv, + struct vmw_sw_context *sw_context, + SVGA3dCmdHeader *header) +{ + struct vmw_draw_indexed_instanced_indirect_cmd { + SVGA3dCmdHeader header; + SVGA3dCmdDXDrawIndexedInstancedIndirect body; + } *cmd = container_of(header, typeof(*cmd), header); + + if (!has_sm5_context(dev_priv)) + return -EINVAL; + + return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface, + VMW_RES_DIRTY_NONE, user_surface_converter, + &cmd->body.argsBufferSid, NULL); +} + +static int vmw_cmd_instanced_indirect(struct vmw_private *dev_priv, + struct vmw_sw_context *sw_context, + SVGA3dCmdHeader *header) +{ + struct vmw_draw_instanced_indirect_cmd { + SVGA3dCmdHeader header; + SVGA3dCmdDXDrawInstancedIndirect body; + } *cmd = container_of(header, typeof(*cmd), header); + + if (!has_sm5_context(dev_priv)) + return -EINVAL; + + return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface, + VMW_RES_DIRTY_NONE, user_surface_converter, + &cmd->body.argsBufferSid, NULL); +} + +static int vmw_cmd_dispatch_indirect(struct vmw_private *dev_priv, + struct vmw_sw_context *sw_context, + SVGA3dCmdHeader *header) +{ + struct vmw_dispatch_indirect_cmd { + SVGA3dCmdHeader header; + SVGA3dCmdDXDispatchIndirect body; + } *cmd = container_of(header, typeof(*cmd), header); + + if (!has_sm5_context(dev_priv)) + return -EINVAL; + + return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface, + VMW_RES_DIRTY_NONE, user_surface_converter, + &cmd->body.argsBufferSid, NULL); +} + static int vmw_cmd_check_not_3d(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, void *buf, uint32_t *size) @@ -2922,18 +3280,12 @@ static const struct vmw_cmd_entry vmw_cmd_entries[SVGA_3D_CMD_MAX] = { false, false, false), VMW_CMD_DEF(SVGA_3D_CMD_DEAD2, &vmw_cmd_invalid, false, false, false), - VMW_CMD_DEF(SVGA_3D_CMD_LOGICOPS_BITBLT, &vmw_cmd_invalid, - false, false, false), - VMW_CMD_DEF(SVGA_3D_CMD_LOGICOPS_TRANSBLT, &vmw_cmd_invalid, - false, false, false), - VMW_CMD_DEF(SVGA_3D_CMD_LOGICOPS_STRETCHBLT, &vmw_cmd_invalid, - false, false, false), - VMW_CMD_DEF(SVGA_3D_CMD_LOGICOPS_COLORFILL, &vmw_cmd_invalid, - false, false, false), - VMW_CMD_DEF(SVGA_3D_CMD_LOGICOPS_ALPHABLEND, &vmw_cmd_invalid, - false, false, false), - VMW_CMD_DEF(SVGA_3D_CMD_LOGICOPS_CLEARTYPEBLEND, &vmw_cmd_invalid, - false, false, false), + VMW_CMD_DEF(SVGA_3D_CMD_DEAD12, &vmw_cmd_invalid, false, false, false), + VMW_CMD_DEF(SVGA_3D_CMD_DEAD13, &vmw_cmd_invalid, false, false, false), + VMW_CMD_DEF(SVGA_3D_CMD_DEAD14, &vmw_cmd_invalid, false, false, false), + VMW_CMD_DEF(SVGA_3D_CMD_DEAD15, &vmw_cmd_invalid, false, false, false), + VMW_CMD_DEF(SVGA_3D_CMD_DEAD16, &vmw_cmd_invalid, false, false, false), + VMW_CMD_DEF(SVGA_3D_CMD_DEAD17, &vmw_cmd_invalid, false, false, false), VMW_CMD_DEF(SVGA_3D_CMD_SET_OTABLE_BASE, &vmw_cmd_invalid, false, false, true), VMW_CMD_DEF(SVGA_3D_CMD_READBACK_OTABLE, &vmw_cmd_invalid, @@ -3141,9 +3493,9 @@ static const struct vmw_cmd_entry vmw_cmd_entries[SVGA_3D_CMD_MAX] = { VMW_CMD_DEF(SVGA_3D_CMD_DX_DEFINE_STREAMOUTPUT, &vmw_cmd_dx_so_define, true, false, true), VMW_CMD_DEF(SVGA_3D_CMD_DX_DESTROY_STREAMOUTPUT, - &vmw_cmd_dx_cid_check, true, false, true), - VMW_CMD_DEF(SVGA_3D_CMD_DX_SET_STREAMOUTPUT, &vmw_cmd_dx_cid_check, - true, false, true), + &vmw_cmd_dx_destroy_streamoutput, true, false, true), + VMW_CMD_DEF(SVGA_3D_CMD_DX_SET_STREAMOUTPUT, + &vmw_cmd_dx_set_streamoutput, true, false, true), VMW_CMD_DEF(SVGA_3D_CMD_DX_SET_SOTARGETS, &vmw_cmd_dx_set_so_targets, true, false, true), VMW_CMD_DEF(SVGA_3D_CMD_DX_SET_INPUT_LAYOUT, @@ -3159,6 +3511,37 @@ static const struct vmw_cmd_entry vmw_cmd_entries[SVGA_3D_CMD_MAX] = { true, false, true), VMW_CMD_DEF(SVGA_3D_CMD_INTRA_SURFACE_COPY, &vmw_cmd_intra_surface_copy, true, false, true), + + /* + * SM5 commands + */ + VMW_CMD_DEF(SVGA_3D_CMD_DX_DEFINE_UA_VIEW, &vmw_cmd_sm5_view_define, + true, false, true), + VMW_CMD_DEF(SVGA_3D_CMD_DX_DESTROY_UA_VIEW, &vmw_cmd_sm5_view_remove, + true, false, true), + VMW_CMD_DEF(SVGA_3D_CMD_DX_CLEAR_UA_VIEW_UINT, &vmw_cmd_clear_uav_uint, + true, false, true), + VMW_CMD_DEF(SVGA_3D_CMD_DX_CLEAR_UA_VIEW_FLOAT, + &vmw_cmd_clear_uav_float, true, false, true), + VMW_CMD_DEF(SVGA_3D_CMD_DX_COPY_STRUCTURE_COUNT, &vmw_cmd_invalid, true, + false, true), + VMW_CMD_DEF(SVGA_3D_CMD_DX_SET_UA_VIEWS, &vmw_cmd_set_uav, true, false, + true), + VMW_CMD_DEF(SVGA_3D_CMD_DX_DRAW_INDEXED_INSTANCED_INDIRECT, + &vmw_cmd_indexed_instanced_indirect, true, false, true), + VMW_CMD_DEF(SVGA_3D_CMD_DX_DRAW_INSTANCED_INDIRECT, + &vmw_cmd_instanced_indirect, true, false, true), + VMW_CMD_DEF(SVGA_3D_CMD_DX_DISPATCH, &vmw_cmd_sm5, true, false, true), + VMW_CMD_DEF(SVGA_3D_CMD_DX_DISPATCH_INDIRECT, + &vmw_cmd_dispatch_indirect, true, false, true), + VMW_CMD_DEF(SVGA_3D_CMD_DX_SET_CS_UA_VIEWS, &vmw_cmd_set_cs_uav, true, + false, true), + VMW_CMD_DEF(SVGA_3D_CMD_DX_DEFINE_DEPTHSTENCIL_VIEW_V2, + &vmw_cmd_sm5_view_define, true, false, true), + VMW_CMD_DEF(SVGA_3D_CMD_DX_DEFINE_STREAMOUTPUT_WITH_MOB, + &vmw_cmd_dx_define_streamoutput, true, false, true), + VMW_CMD_DEF(SVGA_3D_CMD_DX_BIND_STREAMOUTPUT, + &vmw_cmd_dx_bind_streamoutput, true, false, true), }; bool vmw_cmd_describe(const void *buf, u32 *size, char const **cmd) diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c b/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c index e5252ef3812f..6941689085ed 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c @@ -169,10 +169,8 @@ void vmw_fifo_ping_host(struct vmw_private *dev_priv, uint32_t reason) { u32 *fifo_mem = dev_priv->mmio_virt; - preempt_disable(); if (cmpxchg(fifo_mem + SVGA_FIFO_BUSY, 0, 1) == 0) vmw_write(dev_priv, SVGA_REG_SYNC, reason); - preempt_enable(); } void vmw_fifo_release(struct vmw_private *dev_priv, struct vmw_fifo_state *fifo) diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c index a15375eb476e..f681b7b4df1b 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c @@ -114,10 +114,13 @@ int vmw_getparam_ioctl(struct drm_device *dev, void *data, (dev_priv->active_display_unit == vmw_du_screen_target); break; case DRM_VMW_PARAM_DX: - param->value = dev_priv->has_dx; + param->value = has_sm4_context(dev_priv); break; case DRM_VMW_PARAM_SM4_1: - param->value = dev_priv->has_sm4_1; + param->value = has_sm4_1_context(dev_priv); + break; + case DRM_VMW_PARAM_SM5: + param->value = has_sm5_context(dev_priv); break; default: return -EINVAL; @@ -126,14 +129,17 @@ int vmw_getparam_ioctl(struct drm_device *dev, void *data, return 0; } -static u32 vmw_mask_multisample(unsigned int cap, u32 fmt_value) +static u32 vmw_mask_legacy_multisample(unsigned int cap, u32 fmt_value) { /* * A version of user-space exists which use MULTISAMPLE_MASKABLESAMPLES * to check the sample count supported by virtual device. Since there * never was support for multisample count for backing MOB return 0. + * + * MULTISAMPLE_MASKABLESAMPLES devcap is marked as deprecated by virtual + * device. */ - if (cap == SVGA3D_DEVCAP_MULTISAMPLE_MASKABLESAMPLES) + if (cap == SVGA3D_DEVCAP_DEAD5) return 0; return fmt_value; @@ -164,7 +170,7 @@ static int vmw_fill_compat_cap(struct vmw_private *dev_priv, void *bounce, for (i = 0; i < max_size; ++i) { vmw_write(dev_priv, SVGA_REG_DEV_CAP, i); compat_cap->pairs[i][0] = i; - compat_cap->pairs[i][1] = vmw_mask_multisample + compat_cap->pairs[i][1] = vmw_mask_legacy_multisample (i, vmw_read(dev_priv, SVGA_REG_DEV_CAP)); } spin_unlock(&dev_priv->cap_lock); @@ -220,7 +226,7 @@ int vmw_get_cap_3d_ioctl(struct drm_device *dev, void *data, spin_lock(&dev_priv->cap_lock); for (i = 0; i < num; ++i) { vmw_write(dev_priv, SVGA_REG_DEV_CAP, i); - *bounce32++ = vmw_mask_multisample + *bounce32++ = vmw_mask_legacy_multisample (i, vmw_read(dev_priv, SVGA_REG_DEV_CAP)); } spin_unlock(&dev_priv->cap_lock); diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c index eb6e23e8d8ef..04d66592f605 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c @@ -905,14 +905,14 @@ static int vmw_kms_new_framebuffer_surface(struct vmw_private *dev_priv, */ /* Surface must be marked as a scanout. */ - if (unlikely(!surface->scanout)) + if (unlikely(!surface->metadata.scanout)) return -EINVAL; - if (unlikely(surface->mip_levels[0] != 1 || - surface->num_sizes != 1 || - surface->base_size.width < mode_cmd->width || - surface->base_size.height < mode_cmd->height || - surface->base_size.depth != 1)) { + if (unlikely(surface->metadata.mip_levels[0] != 1 || + surface->metadata.num_sizes != 1 || + surface->metadata.base_size.width < mode_cmd->width || + surface->metadata.base_size.height < mode_cmd->height || + surface->metadata.base_size.depth != 1)) { DRM_ERROR("Incompatible surface dimensions " "for requested mode.\n"); return -EINVAL; @@ -941,7 +941,7 @@ static int vmw_kms_new_framebuffer_surface(struct vmw_private *dev_priv, * For DX, surface format validation is done when surface->scanout * is set. */ - if (!dev_priv->has_dx && format != surface->format) { + if (!has_sm4_context(dev_priv) && format != surface->metadata.format) { DRM_ERROR("Invalid surface format for requested mode.\n"); return -EINVAL; } @@ -1144,8 +1144,8 @@ static int vmw_create_bo_proxy(struct drm_device *dev, struct vmw_buffer_object *bo_mob, struct vmw_surface **srf_out) { + struct vmw_surface_metadata metadata = {0}; uint32_t format; - struct drm_vmw_size content_base_size = {0}; struct vmw_resource *res; unsigned int bytes_pp; struct drm_format_name_buf format_name; @@ -1175,22 +1175,15 @@ static int vmw_create_bo_proxy(struct drm_device *dev, return -EINVAL; } - content_base_size.width = mode_cmd->pitches[0] / bytes_pp; - content_base_size.height = mode_cmd->height; - content_base_size.depth = 1; - - ret = vmw_surface_gb_priv_define(dev, - 0, /* kernel visible only */ - 0, /* flags */ - format, - true, /* can be a scanout buffer */ - 1, /* num of mip levels */ - 0, - 0, - content_base_size, - SVGA3D_MS_PATTERN_NONE, - SVGA3D_MS_QUALITY_NONE, - srf_out); + metadata.format = format; + metadata.mip_levels[0] = 1; + metadata.num_sizes = 1; + metadata.base_size.width = mode_cmd->pitches[0] / bytes_pp; + metadata.base_size.height = mode_cmd->height; + metadata.base_size.depth = 1; + metadata.scanout = true; + + ret = vmw_gb_surface_define(vmw_priv(dev), 0, &metadata, srf_out); if (ret) { DRM_ERROR("Failed to allocate proxy content buffer\n"); return ret; @@ -1897,87 +1890,6 @@ int vmw_kms_write_svga(struct vmw_private *vmw_priv, return 0; } -int vmw_kms_save_vga(struct vmw_private *vmw_priv) -{ - struct vmw_vga_topology_state *save; - uint32_t i; - - vmw_priv->vga_width = vmw_read(vmw_priv, SVGA_REG_WIDTH); - vmw_priv->vga_height = vmw_read(vmw_priv, SVGA_REG_HEIGHT); - vmw_priv->vga_bpp = vmw_read(vmw_priv, SVGA_REG_BITS_PER_PIXEL); - if (vmw_priv->capabilities & SVGA_CAP_PITCHLOCK) - vmw_priv->vga_pitchlock = - vmw_read(vmw_priv, SVGA_REG_PITCHLOCK); - else if (vmw_fifo_have_pitchlock(vmw_priv)) - vmw_priv->vga_pitchlock = vmw_mmio_read(vmw_priv->mmio_virt + - SVGA_FIFO_PITCHLOCK); - - if (!(vmw_priv->capabilities & SVGA_CAP_DISPLAY_TOPOLOGY)) - return 0; - - vmw_priv->num_displays = vmw_read(vmw_priv, - SVGA_REG_NUM_GUEST_DISPLAYS); - - if (vmw_priv->num_displays == 0) - vmw_priv->num_displays = 1; - - for (i = 0; i < vmw_priv->num_displays; ++i) { - save = &vmw_priv->vga_save[i]; - vmw_write(vmw_priv, SVGA_REG_DISPLAY_ID, i); - save->primary = vmw_read(vmw_priv, SVGA_REG_DISPLAY_IS_PRIMARY); - save->pos_x = vmw_read(vmw_priv, SVGA_REG_DISPLAY_POSITION_X); - save->pos_y = vmw_read(vmw_priv, SVGA_REG_DISPLAY_POSITION_Y); - save->width = vmw_read(vmw_priv, SVGA_REG_DISPLAY_WIDTH); - save->height = vmw_read(vmw_priv, SVGA_REG_DISPLAY_HEIGHT); - vmw_write(vmw_priv, SVGA_REG_DISPLAY_ID, SVGA_ID_INVALID); - if (i == 0 && vmw_priv->num_displays == 1 && - save->width == 0 && save->height == 0) { - - /* - * It should be fairly safe to assume that these - * values are uninitialized. - */ - - save->width = vmw_priv->vga_width - save->pos_x; - save->height = vmw_priv->vga_height - save->pos_y; - } - } - - return 0; -} - -int vmw_kms_restore_vga(struct vmw_private *vmw_priv) -{ - struct vmw_vga_topology_state *save; - uint32_t i; - - vmw_write(vmw_priv, SVGA_REG_WIDTH, vmw_priv->vga_width); - vmw_write(vmw_priv, SVGA_REG_HEIGHT, vmw_priv->vga_height); - vmw_write(vmw_priv, SVGA_REG_BITS_PER_PIXEL, vmw_priv->vga_bpp); - if (vmw_priv->capabilities & SVGA_CAP_PITCHLOCK) - vmw_write(vmw_priv, SVGA_REG_PITCHLOCK, - vmw_priv->vga_pitchlock); - else if (vmw_fifo_have_pitchlock(vmw_priv)) - vmw_mmio_write(vmw_priv->vga_pitchlock, - vmw_priv->mmio_virt + SVGA_FIFO_PITCHLOCK); - - if (!(vmw_priv->capabilities & SVGA_CAP_DISPLAY_TOPOLOGY)) - return 0; - - for (i = 0; i < vmw_priv->num_displays; ++i) { - save = &vmw_priv->vga_save[i]; - vmw_write(vmw_priv, SVGA_REG_DISPLAY_ID, i); - vmw_write(vmw_priv, SVGA_REG_DISPLAY_IS_PRIMARY, save->primary); - vmw_write(vmw_priv, SVGA_REG_DISPLAY_POSITION_X, save->pos_x); - vmw_write(vmw_priv, SVGA_REG_DISPLAY_POSITION_Y, save->pos_y); - vmw_write(vmw_priv, SVGA_REG_DISPLAY_WIDTH, save->width); - vmw_write(vmw_priv, SVGA_REG_DISPLAY_HEIGHT, save->height); - vmw_write(vmw_priv, SVGA_REG_DISPLAY_ID, SVGA_ID_INVALID); - } - - return 0; -} - bool vmw_kms_validate_mode_vram(struct vmw_private *dev_priv, uint32_t pitch, uint32_t height) @@ -2597,7 +2509,7 @@ int vmw_kms_update_proxy(struct vmw_resource *res, int increment) { struct vmw_private *dev_priv = res->dev_priv; - struct drm_vmw_size *size = &vmw_res_to_srf(res)->base_size; + struct drm_vmw_size *size = &vmw_res_to_srf(res)->metadata.base_size; struct { SVGA3dCmdHeader header; SVGA3dCmdUpdateGBImage body; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_mob.c b/drivers/gpu/drm/vmwgfx/vmwgfx_mob.c index 0a6bbac00896..e8eb42933ca2 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_mob.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_mob.c @@ -320,7 +320,7 @@ int vmw_otables_setup(struct vmw_private *dev_priv) struct vmw_otable **otables = &dev_priv->otable_batch.otables; int ret; - if (dev_priv->has_dx) { + if (has_sm4_context(dev_priv)) { *otables = kmemdup(dx_tables, sizeof(dx_tables), GFP_KERNEL); if (!(*otables)) return -ENOMEM; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_overlay.c b/drivers/gpu/drm/vmwgfx/vmwgfx_overlay.c index fdb52f6d29fb..cd7ed1650d60 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_overlay.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_overlay.c @@ -354,37 +354,6 @@ static int vmw_overlay_update_stream(struct vmw_private *dev_priv, } /** - * Stop all streams. - * - * Used by the fb code when starting. - * - * Takes the overlay lock. - */ -int vmw_overlay_stop_all(struct vmw_private *dev_priv) -{ - struct vmw_overlay *overlay = dev_priv->overlay_priv; - int i, ret; - - if (!overlay) - return 0; - - mutex_lock(&overlay->mutex); - - for (i = 0; i < VMW_MAX_NUM_STREAMS; i++) { - struct vmw_stream *stream = &overlay->stream[i]; - if (!stream->buf) - continue; - - ret = vmw_overlay_stop(dev_priv, i, false, false); - WARN_ON(ret != 0); - } - - mutex_unlock(&overlay->mutex); - - return 0; -} - -/** * Try to resume all paused streams. * * Used by the kms code after moving a new scanout buffer to vram. diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_page_dirty.c b/drivers/gpu/drm/vmwgfx/vmwgfx_page_dirty.c index f07aa857587c..d4d66532f9c9 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_page_dirty.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_page_dirty.c @@ -69,7 +69,7 @@ struct vmw_bo_dirty { unsigned int ref_count; unsigned long bitmap_size; size_t size; - unsigned long bitmap[0]; + unsigned long bitmap[]; }; /** @@ -473,11 +473,11 @@ vm_fault_t vmw_bo_vm_fault(struct vm_fault *vmf) * a lot of unnecessary write faults. */ if (vbo->dirty && vbo->dirty->method == VMW_BO_DIRTY_MKWRITE) - prot = vma->vm_page_prot; + prot = vm_get_page_prot(vma->vm_flags & ~VM_SHARED); else prot = vm_get_page_prot(vma->vm_flags); - ret = ttm_bo_vm_fault_reserved(vmf, prot, num_prefault); + ret = ttm_bo_vm_fault_reserved(vmf, prot, num_prefault, 1); if (ret == VM_FAULT_RETRY && !(vmf->flags & FAULT_FLAG_RETRY_NOWAIT)) return ret; @@ -486,3 +486,75 @@ out_unlock: return ret; } + +#ifdef CONFIG_TRANSPARENT_HUGEPAGE +vm_fault_t vmw_bo_vm_huge_fault(struct vm_fault *vmf, + enum page_entry_size pe_size) +{ + struct vm_area_struct *vma = vmf->vma; + struct ttm_buffer_object *bo = (struct ttm_buffer_object *) + vma->vm_private_data; + struct vmw_buffer_object *vbo = + container_of(bo, struct vmw_buffer_object, base); + pgprot_t prot; + vm_fault_t ret; + pgoff_t fault_page_size; + bool write = vmf->flags & FAULT_FLAG_WRITE; + bool is_cow_mapping = + (vma->vm_flags & (VM_SHARED | VM_MAYWRITE)) == VM_MAYWRITE; + + switch (pe_size) { + case PE_SIZE_PMD: + fault_page_size = HPAGE_PMD_SIZE >> PAGE_SHIFT; + break; +#ifdef CONFIG_HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD + case PE_SIZE_PUD: + fault_page_size = HPAGE_PUD_SIZE >> PAGE_SHIFT; + break; +#endif + default: + WARN_ON_ONCE(1); + return VM_FAULT_FALLBACK; + } + + /* Always do write dirty-tracking and COW on PTE level. */ + if (write && (READ_ONCE(vbo->dirty) || is_cow_mapping)) + return VM_FAULT_FALLBACK; + + ret = ttm_bo_vm_reserve(bo, vmf); + if (ret) + return ret; + + if (vbo->dirty) { + pgoff_t allowed_prefault; + unsigned long page_offset; + + page_offset = vmf->pgoff - + drm_vma_node_start(&bo->base.vma_node); + if (page_offset >= bo->num_pages || + vmw_resources_clean(vbo, page_offset, + page_offset + PAGE_SIZE, + &allowed_prefault)) { + ret = VM_FAULT_SIGBUS; + goto out_unlock; + } + + /* + * Write protect, so we get a new fault on write, and can + * split. + */ + prot = vm_get_page_prot(vma->vm_flags & ~VM_SHARED); + } else { + prot = vm_get_page_prot(vma->vm_flags); + } + + ret = ttm_bo_vm_fault_reserved(vmf, prot, 1, fault_page_size); + if (ret == VM_FAULT_RETRY && !(vmf->flags & FAULT_FLAG_RETRY_NOWAIT)) + return ret; + +out_unlock: + dma_resv_unlock(bo->base.resv); + + return ret; +} +#endif diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_so.c b/drivers/gpu/drm/vmwgfx/vmwgfx_so.c index 63807361e16f..3f97b61dd5d8 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_so.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_so.c @@ -319,7 +319,8 @@ int vmw_view_add(struct vmw_cmdbuf_res_manager *man, static const size_t vmw_view_define_sizes[] = { [vmw_view_sr] = sizeof(SVGA3dCmdDXDefineShaderResourceView), [vmw_view_rt] = sizeof(SVGA3dCmdDXDefineRenderTargetView), - [vmw_view_ds] = sizeof(SVGA3dCmdDXDefineDepthStencilView) + [vmw_view_ds] = sizeof(SVGA3dCmdDXDefineDepthStencilView), + [vmw_view_ua] = sizeof(SVGA3dCmdDXDefineUAView) }; struct vmw_private *dev_priv = ctx->dev_priv; @@ -499,8 +500,8 @@ struct vmw_resource *vmw_view_lookup(struct vmw_cmdbuf_res_manager *man, * Each time a resource is put on the validation list as the result of a * view pointing to it, we need to determine whether that resource will * be dirtied (written to by the GPU) as a result of the corresponding - * GPU operation. Currently only rendertarget- and depth-stencil views are - * capable of dirtying its resource. + * GPU operation. Currently only rendertarget-, depth-stencil and unordered + * access views are capable of dirtying its resource. * * Return: Whether the view type of @res dirties the resource it points to. */ @@ -509,10 +510,11 @@ u32 vmw_view_dirtying(struct vmw_resource *res) static u32 view_is_dirtying[vmw_view_max] = { [vmw_view_rt] = VMW_RES_DIRTY_SET, [vmw_view_ds] = VMW_RES_DIRTY_SET, + [vmw_view_ua] = VMW_RES_DIRTY_SET, }; /* Update this function as we add more view types */ - BUILD_BUG_ON(vmw_view_max != 3); + BUILD_BUG_ON(vmw_view_max != 4); return view_is_dirtying[vmw_view(res)->view_type]; } @@ -520,12 +522,14 @@ const u32 vmw_view_destroy_cmds[] = { [vmw_view_sr] = SVGA_3D_CMD_DX_DESTROY_SHADERRESOURCE_VIEW, [vmw_view_rt] = SVGA_3D_CMD_DX_DESTROY_RENDERTARGET_VIEW, [vmw_view_ds] = SVGA_3D_CMD_DX_DESTROY_DEPTHSTENCIL_VIEW, + [vmw_view_ua] = SVGA_3D_CMD_DX_DESTROY_UA_VIEW, }; const SVGACOTableType vmw_view_cotables[] = { [vmw_view_sr] = SVGA_COTABLE_SRVIEW, [vmw_view_rt] = SVGA_COTABLE_RTVIEW, [vmw_view_ds] = SVGA_COTABLE_DSVIEW, + [vmw_view_ua] = SVGA_COTABLE_UAVIEW, }; const SVGACOTableType vmw_so_cotables[] = { diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_so.h b/drivers/gpu/drm/vmwgfx/vmwgfx_so.h index 12565047bc55..f48b84bfeeac 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_so.h +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_so.h @@ -30,6 +30,7 @@ enum vmw_view_type { vmw_view_sr, vmw_view_rt, vmw_view_ds, + vmw_view_ua, vmw_view_max, }; @@ -61,6 +62,7 @@ union vmw_view_destroy { struct SVGA3dCmdDXDestroyRenderTargetView rtv; struct SVGA3dCmdDXDestroyShaderResourceView srv; struct SVGA3dCmdDXDestroyDepthStencilView dsv; + struct SVGA3dCmdDXDestroyUAView uav; u32 view_id; }; @@ -87,6 +89,10 @@ static inline enum vmw_view_type vmw_view_cmd_to_type(u32 id) { u32 tmp = (id - SVGA_3D_CMD_DX_DEFINE_SHADERRESOURCE_VIEW) / 2; + if (id == SVGA_3D_CMD_DX_DEFINE_UA_VIEW || + id == SVGA_3D_CMD_DX_DESTROY_UA_VIEW) + return vmw_view_ua; + if (tmp > (u32)vmw_view_max) return vmw_view_max; @@ -123,6 +129,7 @@ static inline enum vmw_so_type vmw_so_cmd_to_type(u32 id) case SVGA_3D_CMD_DX_DESTROY_SAMPLER_STATE: return vmw_so_ss; case SVGA_3D_CMD_DX_DEFINE_STREAMOUTPUT: + case SVGA_3D_CMD_DX_DEFINE_STREAMOUTPUT_WITH_MOB: case SVGA_3D_CMD_DX_DESTROY_STREAMOUTPUT: return vmw_so_so; default: diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c b/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c index 570687a1a327..9ffa9c75a5da 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c @@ -590,7 +590,7 @@ static void vmw_stdu_bo_cpu_commit(struct vmw_kms_dirty *dirty) return; /* Assume we are blitting from Guest (bo) to Host (display_srf) */ - dst_pitch = stdu->display_srf->base_size.width * stdu->cpp; + dst_pitch = stdu->display_srf->metadata.base_size.width * stdu->cpp; dst_bo = &stdu->display_srf->res.backup->base; dst_offset = ddirty->top * dst_pitch + ddirty->left * stdu->cpp; @@ -1041,7 +1041,6 @@ vmw_stdu_primary_plane_prepare_fb(struct drm_plane *plane, struct vmw_plane_state *vps = vmw_plane_state_to_vps(new_state); enum stdu_content_type new_content_type; struct vmw_framebuffer_surface *new_vfbs; - struct drm_crtc *crtc = new_state->crtc; uint32_t hdisplay = new_state->crtc_w, vdisplay = new_state->crtc_h; int ret; @@ -1058,8 +1057,9 @@ vmw_stdu_primary_plane_prepare_fb(struct drm_plane *plane, vfb = vmw_framebuffer_to_vfb(new_fb); new_vfbs = (vfb->bo) ? NULL : vmw_framebuffer_to_vfbs(new_fb); - if (new_vfbs && new_vfbs->surface->base_size.width == hdisplay && - new_vfbs->surface->base_size.height == vdisplay) + if (new_vfbs && + new_vfbs->surface->metadata.base_size.width == hdisplay && + new_vfbs->surface->metadata.base_size.height == vdisplay) new_content_type = SAME_AS_DISPLAY; else if (vfb->bo) new_content_type = SEPARATE_BO; @@ -1067,12 +1067,11 @@ vmw_stdu_primary_plane_prepare_fb(struct drm_plane *plane, new_content_type = SEPARATE_SURFACE; if (new_content_type != SAME_AS_DISPLAY) { - struct vmw_surface content_srf; - struct drm_vmw_size display_base_size = {0}; + struct vmw_surface_metadata metadata = {0}; - display_base_size.width = hdisplay; - display_base_size.height = vdisplay; - display_base_size.depth = 1; + metadata.base_size.width = hdisplay; + metadata.base_size.height = vdisplay; + metadata.base_size.depth = 1; /* * If content buffer is a buffer object, then we have to @@ -1082,15 +1081,15 @@ vmw_stdu_primary_plane_prepare_fb(struct drm_plane *plane, switch (new_fb->format->cpp[0]*8) { case 32: - content_srf.format = SVGA3D_X8R8G8B8; + metadata.format = SVGA3D_X8R8G8B8; break; case 16: - content_srf.format = SVGA3D_R5G6B5; + metadata.format = SVGA3D_R5G6B5; break; case 8: - content_srf.format = SVGA3D_P8; + metadata.format = SVGA3D_P8; break; default: @@ -1098,22 +1097,20 @@ vmw_stdu_primary_plane_prepare_fb(struct drm_plane *plane, return -EINVAL; } - content_srf.flags = 0; - content_srf.mip_levels[0] = 1; - content_srf.multisample_count = 0; - content_srf.multisample_pattern = - SVGA3D_MS_PATTERN_NONE; - content_srf.quality_level = SVGA3D_MS_QUALITY_NONE; + metadata.mip_levels[0] = 1; + metadata.num_sizes = 1; + metadata.scanout = true; } else { - content_srf = *new_vfbs->surface; + metadata = new_vfbs->surface->metadata; } if (vps->surf) { - struct drm_vmw_size cur_base_size = vps->surf->base_size; + struct drm_vmw_size cur_base_size = + vps->surf->metadata.base_size; - if (cur_base_size.width != display_base_size.width || - cur_base_size.height != display_base_size.height || - vps->surf->format != content_srf.format) { + if (cur_base_size.width != metadata.base_size.width || + cur_base_size.height != metadata.base_size.height || + vps->surf->metadata.format != metadata.format) { WARN_ON(vps->pinned != 0); vmw_surface_unreference(&vps->surf); } @@ -1121,20 +1118,8 @@ vmw_stdu_primary_plane_prepare_fb(struct drm_plane *plane, } if (!vps->surf) { - ret = vmw_surface_gb_priv_define - (crtc->dev, - /* Kernel visible only */ - 0, - content_srf.flags, - content_srf.format, - true, /* a scanout buffer */ - content_srf.mip_levels[0], - content_srf.multisample_count, - 0, - display_base_size, - content_srf.multisample_pattern, - content_srf.quality_level, - &vps->surf); + ret = vmw_gb_surface_define(dev_priv, 0, &metadata, + &vps->surf); if (ret != 0) { DRM_ERROR("Couldn't allocate STDU surface.\n"); return ret; @@ -1311,7 +1296,7 @@ vmw_stdu_bo_populate_update_cpu(struct vmw_du_update_plane *update, void *cmd, diff.cpp = stdu->cpp; dst_bo = &stdu->display_srf->res.backup->base; - dst_pitch = stdu->display_srf->base_size.width * stdu->cpp; + dst_pitch = stdu->display_srf->metadata.base_size.width * stdu->cpp; dst_offset = bb->y1 * dst_pitch + bb->x1 * stdu->cpp; src_bo = &vfbbo->buffer->base; @@ -1888,7 +1873,7 @@ int vmw_kms_stdu_init_display(struct vmw_private *dev_priv) /* Do nothing if Screen Target support is turned off */ - if (!VMWGFX_ENABLE_SCREEN_TARGET_OTABLE) + if (!VMWGFX_ENABLE_SCREEN_TARGET_OTABLE || !dev_priv->has_mob) return -ENOSYS; if (!(dev_priv->capabilities & SVGA_CAP_GBOBJECTS)) diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_streamoutput.c b/drivers/gpu/drm/vmwgfx/vmwgfx_streamoutput.c new file mode 100644 index 000000000000..193192456663 --- /dev/null +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_streamoutput.c @@ -0,0 +1,387 @@ +// SPDX-License-Identifier: GPL-2.0 OR MIT +/************************************************************************** + * + * Copyright © 2018-2019 VMware, Inc., Palo Alto, CA., USA + * All Rights Reserved. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the + * "Software"), to deal in the Software without restriction, including + * without limitation the rights to use, copy, modify, merge, publish, + * distribute, sub license, and/or sell copies of the Software, and to + * permit persons to whom the Software is furnished to do so, subject to + * the following conditions: + * + * The above copyright notice and this permission notice (including the + * next paragraph) shall be included in all copies or substantial portions + * of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDERS, AUTHORS AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM, + * DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR + * OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE + * USE OR OTHER DEALINGS IN THE SOFTWARE. + * + **************************************************************************/ + +#include <drm/ttm/ttm_placement.h> + +#include "vmwgfx_drv.h" +#include "vmwgfx_resource_priv.h" +#include "vmwgfx_binding.h" + +/** + * struct vmw_dx_streamoutput - Streamoutput resource metadata. + * @res: Base resource struct. + * @ctx: Non-refcounted context to which @res belong. + * @cotable: Refcounted cotable holding this Streamoutput. + * @cotable_head: List head for cotable-so_res list. + * @id: User-space provided identifier. + * @size: User-space provided mob size. + * @committed: Whether streamoutput is actually created or pending creation. + */ +struct vmw_dx_streamoutput { + struct vmw_resource res; + struct vmw_resource *ctx; + struct vmw_resource *cotable; + struct list_head cotable_head; + u32 id; + u32 size; + bool committed; +}; + +static int vmw_dx_streamoutput_create(struct vmw_resource *res); +static int vmw_dx_streamoutput_bind(struct vmw_resource *res, + struct ttm_validate_buffer *val_buf); +static int vmw_dx_streamoutput_unbind(struct vmw_resource *res, bool readback, + struct ttm_validate_buffer *val_buf); +static void vmw_dx_streamoutput_commit_notify(struct vmw_resource *res, + enum vmw_cmdbuf_res_state state); + +static size_t vmw_streamoutput_size; + +static const struct vmw_res_func vmw_dx_streamoutput_func = { + .res_type = vmw_res_streamoutput, + .needs_backup = true, + .may_evict = false, + .type_name = "DX streamoutput", + .backup_placement = &vmw_mob_placement, + .create = vmw_dx_streamoutput_create, + .destroy = NULL, /* Command buffer managed resource. */ + .bind = vmw_dx_streamoutput_bind, + .unbind = vmw_dx_streamoutput_unbind, + .commit_notify = vmw_dx_streamoutput_commit_notify, +}; + +static inline struct vmw_dx_streamoutput * +vmw_res_to_dx_streamoutput(struct vmw_resource *res) +{ + return container_of(res, struct vmw_dx_streamoutput, res); +} + +/** + * vmw_dx_streamoutput_unscrub - Reattach the MOB to streamoutput. + * @res: The streamoutput resource. + * + * Return: 0 on success, negative error code on failure. + */ +static int vmw_dx_streamoutput_unscrub(struct vmw_resource *res) +{ + struct vmw_dx_streamoutput *so = vmw_res_to_dx_streamoutput(res); + struct vmw_private *dev_priv = res->dev_priv; + struct { + SVGA3dCmdHeader header; + SVGA3dCmdDXBindStreamOutput body; + } *cmd; + + if (!list_empty(&so->cotable_head) || !so->committed ) + return 0; + + cmd = VMW_FIFO_RESERVE_DX(dev_priv, sizeof(*cmd), so->ctx->id); + if (!cmd) + return -ENOMEM; + + cmd->header.id = SVGA_3D_CMD_DX_BIND_STREAMOUTPUT; + cmd->header.size = sizeof(cmd->body); + cmd->body.soid = so->id; + cmd->body.mobid = res->backup->base.mem.start; + cmd->body.offsetInBytes = res->backup_offset; + cmd->body.sizeInBytes = so->size; + vmw_fifo_commit(dev_priv, sizeof(*cmd)); + + vmw_cotable_add_resource(so->cotable, &so->cotable_head); + + return 0; +} + +static int vmw_dx_streamoutput_create(struct vmw_resource *res) +{ + struct vmw_private *dev_priv = res->dev_priv; + struct vmw_dx_streamoutput *so = vmw_res_to_dx_streamoutput(res); + int ret = 0; + + WARN_ON_ONCE(!so->committed); + + if (vmw_resource_mob_attached(res)) { + mutex_lock(&dev_priv->binding_mutex); + ret = vmw_dx_streamoutput_unscrub(res); + mutex_unlock(&dev_priv->binding_mutex); + } + + res->id = so->id; + + return ret; +} + +static int vmw_dx_streamoutput_bind(struct vmw_resource *res, + struct ttm_validate_buffer *val_buf) +{ + struct vmw_private *dev_priv = res->dev_priv; + struct ttm_buffer_object *bo = val_buf->bo; + int ret; + + if (WARN_ON(bo->mem.mem_type != VMW_PL_MOB)) + return -EINVAL; + + mutex_lock(&dev_priv->binding_mutex); + ret = vmw_dx_streamoutput_unscrub(res); + mutex_unlock(&dev_priv->binding_mutex); + + return ret; +} + +/** + * vmw_dx_streamoutput_scrub - Unbind the MOB from streamoutput. + * @res: The streamoutput resource. + * + * Return: 0 on success, negative error code on failure. + */ +static int vmw_dx_streamoutput_scrub(struct vmw_resource *res) +{ + struct vmw_private *dev_priv = res->dev_priv; + struct vmw_dx_streamoutput *so = vmw_res_to_dx_streamoutput(res); + struct { + SVGA3dCmdHeader header; + SVGA3dCmdDXBindStreamOutput body; + } *cmd; + + if (list_empty(&so->cotable_head)) + return 0; + + WARN_ON_ONCE(!so->committed); + + cmd = VMW_FIFO_RESERVE_DX(dev_priv, sizeof(*cmd), so->ctx->id); + if (!cmd) + return -ENOMEM; + + cmd->header.id = SVGA_3D_CMD_DX_BIND_STREAMOUTPUT; + cmd->header.size = sizeof(cmd->body); + cmd->body.soid = res->id; + cmd->body.mobid = SVGA3D_INVALID_ID; + cmd->body.offsetInBytes = 0; + cmd->body.sizeInBytes = so->size; + vmw_fifo_commit(dev_priv, sizeof(*cmd)); + + res->id = -1; + list_del_init(&so->cotable_head); + + return 0; +} + +static int vmw_dx_streamoutput_unbind(struct vmw_resource *res, bool readback, + struct ttm_validate_buffer *val_buf) +{ + struct vmw_private *dev_priv = res->dev_priv; + struct vmw_fence_obj *fence; + int ret; + + if (WARN_ON(res->backup->base.mem.mem_type != VMW_PL_MOB)) + return -EINVAL; + + mutex_lock(&dev_priv->binding_mutex); + ret = vmw_dx_streamoutput_scrub(res); + mutex_unlock(&dev_priv->binding_mutex); + + if (ret) + return ret; + + (void) vmw_execbuf_fence_commands(NULL, dev_priv, &fence, NULL); + vmw_bo_fence_single(val_buf->bo, fence); + + if (fence != NULL) + vmw_fence_obj_unreference(&fence); + + return 0; +} + +static void vmw_dx_streamoutput_commit_notify(struct vmw_resource *res, + enum vmw_cmdbuf_res_state state) +{ + struct vmw_private *dev_priv = res->dev_priv; + struct vmw_dx_streamoutput *so = vmw_res_to_dx_streamoutput(res); + + if (state == VMW_CMDBUF_RES_ADD) { + mutex_lock(&dev_priv->binding_mutex); + vmw_cotable_add_resource(so->cotable, &so->cotable_head); + so->committed = true; + res->id = so->id; + mutex_unlock(&dev_priv->binding_mutex); + } else { + mutex_lock(&dev_priv->binding_mutex); + list_del_init(&so->cotable_head); + so->committed = false; + res->id = -1; + mutex_unlock(&dev_priv->binding_mutex); + } +} + +/** + * vmw_dx_streamoutput_lookup - Do a streamoutput resource lookup by user key. + * @man: Command buffer managed resource manager for current context. + * @user_key: User-space identifier for lookup. + * + * Return: Valid refcounted vmw_resource on success, error pointer on failure. + */ +struct vmw_resource * +vmw_dx_streamoutput_lookup(struct vmw_cmdbuf_res_manager *man, + u32 user_key) +{ + return vmw_cmdbuf_res_lookup(man, vmw_cmdbuf_res_streamoutput, + user_key); +} + +static void vmw_dx_streamoutput_res_free(struct vmw_resource *res) +{ + struct vmw_private *dev_priv = res->dev_priv; + struct vmw_dx_streamoutput *so = vmw_res_to_dx_streamoutput(res); + + vmw_resource_unreference(&so->cotable); + kfree(so); + ttm_mem_global_free(vmw_mem_glob(dev_priv), vmw_streamoutput_size); +} + +static void vmw_dx_streamoutput_hw_destroy(struct vmw_resource *res) +{ + /* Destroyed by user-space cmd buf or as part of context takedown. */ + res->id = -1; +} + +/** + * vmw_dx_streamoutput_add - Add a streamoutput as a cmd buf managed resource. + * @man: Command buffer managed resource manager for current context. + * @ctx: Pointer to context resource. + * @user_key: The identifier for this streamoutput. + * @list: The list of staged command buffer managed resources. + * + * Return: 0 on success, negative error code on failure. + */ +int vmw_dx_streamoutput_add(struct vmw_cmdbuf_res_manager *man, + struct vmw_resource *ctx, u32 user_key, + struct list_head *list) +{ + struct vmw_dx_streamoutput *so; + struct vmw_resource *res; + struct vmw_private *dev_priv = ctx->dev_priv; + struct ttm_operation_ctx ttm_opt_ctx = { + .interruptible = true, + .no_wait_gpu = false + }; + int ret; + + if (!vmw_streamoutput_size) + vmw_streamoutput_size = ttm_round_pot(sizeof(*so)); + + ret = ttm_mem_global_alloc(vmw_mem_glob(dev_priv), + vmw_streamoutput_size, &ttm_opt_ctx); + if (ret) { + if (ret != -ERESTARTSYS) + DRM_ERROR("Out of graphics memory for streamout.\n"); + return ret; + } + + so = kmalloc(sizeof(*so), GFP_KERNEL); + if (!so) { + ttm_mem_global_free(vmw_mem_glob(dev_priv), + vmw_streamoutput_size); + return -ENOMEM; + } + + res = &so->res; + so->ctx = ctx; + so->cotable = vmw_resource_reference + (vmw_context_cotable(ctx, SVGA_COTABLE_STREAMOUTPUT)); + so->id = user_key; + so->committed = false; + INIT_LIST_HEAD(&so->cotable_head); + ret = vmw_resource_init(dev_priv, res, true, + vmw_dx_streamoutput_res_free, + &vmw_dx_streamoutput_func); + if (ret) + goto out_resource_init; + + ret = vmw_cmdbuf_res_add(man, vmw_cmdbuf_res_streamoutput, user_key, + res, list); + if (ret) + goto out_resource_init; + + res->id = so->id; + res->hw_destroy = vmw_dx_streamoutput_hw_destroy; + +out_resource_init: + vmw_resource_unreference(&res); + + return ret; +} + +/** + * vmw_dx_streamoutput_set_size - Sets streamoutput mob size in res struct. + * @res: The streamoutput res for which need to set size. + * @size: The size provided by user-space to set. + */ +void vmw_dx_streamoutput_set_size(struct vmw_resource *res, u32 size) +{ + struct vmw_dx_streamoutput *so = vmw_res_to_dx_streamoutput(res); + + so->size = size; +} + +/** + * vmw_dx_streamoutput_remove - Stage streamoutput for removal. + * @man: Command buffer managed resource manager for current context. + * @user_key: The identifier for this streamoutput. + * @list: The list of staged command buffer managed resources. + * + * Return: 0 on success, negative error code on failure. + */ +int vmw_dx_streamoutput_remove(struct vmw_cmdbuf_res_manager *man, + u32 user_key, + struct list_head *list) +{ + struct vmw_resource *r; + + return vmw_cmdbuf_res_remove(man, vmw_cmdbuf_res_streamoutput, + (u32)user_key, list, &r); +} + +/** + * vmw_dx_streamoutput_cotable_list_scrub - cotable unbind_func callback. + * @dev_priv: Device private. + * @list: The list of cotable resources. + * @readback: Whether the call was part of a readback unbind. + */ +void vmw_dx_streamoutput_cotable_list_scrub(struct vmw_private *dev_priv, + struct list_head *list, + bool readback) +{ + struct vmw_dx_streamoutput *entry, *next; + + lockdep_assert_held_once(&dev_priv->binding_mutex); + + list_for_each_entry_safe(entry, next, list, cotable_head) { + WARN_ON(vmw_dx_streamoutput_scrub(&entry->res)); + if (!readback) + entry->committed =false; + } +} diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c b/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c index 3ce630aa4fde..7ef51fa84b01 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c @@ -79,7 +79,7 @@ struct vmw_surface_dirty { struct svga3dsurface_cache cache; size_t size; u32 num_subres; - SVGA3dBox boxes[0]; + SVGA3dBox boxes[]; }; static void vmw_user_surface_free(struct vmw_resource *res); @@ -199,7 +199,7 @@ struct vmw_surface_destroy { */ static inline uint32_t vmw_surface_dma_size(const struct vmw_surface *srf) { - return srf->num_sizes * sizeof(struct vmw_surface_dma); + return srf->metadata.num_sizes * sizeof(struct vmw_surface_dma); } @@ -213,7 +213,7 @@ static inline uint32_t vmw_surface_dma_size(const struct vmw_surface *srf) */ static inline uint32_t vmw_surface_define_size(const struct vmw_surface *srf) { - return sizeof(struct vmw_surface_define) + srf->num_sizes * + return sizeof(struct vmw_surface_define) + srf->metadata.num_sizes * sizeof(SVGA3dSize); } @@ -262,7 +262,8 @@ static void vmw_surface_define_encode(const struct vmw_surface *srf, uint32_t cmd_len; int i; - cmd_len = sizeof(cmd->body) + srf->num_sizes * sizeof(SVGA3dSize); + cmd_len = sizeof(cmd->body) + srf->metadata.num_sizes * + sizeof(SVGA3dSize); cmd->header.id = SVGA_3D_CMD_SURFACE_DEFINE; cmd->header.size = cmd_len; @@ -272,16 +273,16 @@ static void vmw_surface_define_encode(const struct vmw_surface *srf, * since driver internally stores as 64 bit. * For legacy surface define only 32 bit flag is supported. */ - cmd->body.surfaceFlags = (SVGA3dSurface1Flags)srf->flags; - cmd->body.format = srf->format; + cmd->body.surfaceFlags = (SVGA3dSurface1Flags)srf->metadata.flags; + cmd->body.format = srf->metadata.format; for (i = 0; i < DRM_VMW_MAX_SURFACE_FACES; ++i) - cmd->body.face[i].numMipLevels = srf->mip_levels[i]; + cmd->body.face[i].numMipLevels = srf->metadata.mip_levels[i]; cmd += 1; cmd_size = (SVGA3dSize *) cmd; - src_size = srf->sizes; + src_size = srf->metadata.sizes; - for (i = 0; i < srf->num_sizes; ++i, cmd_size++, src_size++) { + for (i = 0; i < srf->metadata.num_sizes; ++i, cmd_size++, src_size++) { cmd_size->width = src_size->width; cmd_size->height = src_size->height; cmd_size->depth = src_size->depth; @@ -305,15 +306,15 @@ static void vmw_surface_dma_encode(struct vmw_surface *srf, uint32_t i; struct vmw_surface_dma *cmd = (struct vmw_surface_dma *)cmd_space; const struct svga3d_surface_desc *desc = - svga3dsurface_get_desc(srf->format); + svga3dsurface_get_desc(srf->metadata.format); - for (i = 0; i < srf->num_sizes; ++i) { + for (i = 0; i < srf->metadata.num_sizes; ++i) { SVGA3dCmdHeader *header = &cmd->header; SVGA3dCmdSurfaceDMA *body = &cmd->body; SVGA3dCopyBox *cb = &cmd->cb; SVGA3dCmdSurfaceDMASuffix *suffix = &cmd->suffix; const struct vmw_surface_offset *cur_offset = &srf->offsets[i]; - const struct drm_vmw_size *cur_size = &srf->sizes[i]; + const struct drm_vmw_size *cur_size = &srf->metadata.sizes[i]; header->id = SVGA_3D_CMD_SURFACE_DMA; header->size = sizeof(*body) + sizeof(*cb) + sizeof(*suffix); @@ -669,7 +670,7 @@ static void vmw_user_surface_free(struct vmw_resource *res) if (user_srf->master) drm_master_put(&user_srf->master); kfree(srf->offsets); - kfree(srf->sizes); + kfree(srf->metadata.sizes); kfree(srf->snooper.image); ttm_prime_object_kfree(user_srf, prime); ttm_mem_global_free(vmw_mem_glob(dev_priv), size); @@ -728,6 +729,7 @@ int vmw_surface_define_ioctl(struct drm_device *dev, void *data, struct vmw_private *dev_priv = vmw_priv(dev); struct vmw_user_surface *user_srf; struct vmw_surface *srf; + struct vmw_surface_metadata *metadata; struct vmw_resource *res; struct vmw_resource *tmp; union drm_vmw_surface_create_arg *arg = @@ -793,43 +795,45 @@ int vmw_surface_define_ioctl(struct drm_device *dev, void *data, } srf = &user_srf->srf; + metadata = &srf->metadata; res = &srf->res; /* Driver internally stores as 64-bit flags */ - srf->flags = (SVGA3dSurfaceAllFlags)req->flags; - srf->format = req->format; - srf->scanout = req->scanout; + metadata->flags = (SVGA3dSurfaceAllFlags)req->flags; + metadata->format = req->format; + metadata->scanout = req->scanout; - memcpy(srf->mip_levels, req->mip_levels, sizeof(srf->mip_levels)); - srf->num_sizes = num_sizes; + memcpy(metadata->mip_levels, req->mip_levels, + sizeof(metadata->mip_levels)); + metadata->num_sizes = num_sizes; user_srf->size = size; - srf->sizes = memdup_user((struct drm_vmw_size __user *)(unsigned long) - req->size_addr, - sizeof(*srf->sizes) * srf->num_sizes); - if (IS_ERR(srf->sizes)) { - ret = PTR_ERR(srf->sizes); + metadata->sizes = + memdup_user((struct drm_vmw_size __user *)(unsigned long) + req->size_addr, + sizeof(*metadata->sizes) * metadata->num_sizes); + if (IS_ERR(metadata->sizes)) { + ret = PTR_ERR(metadata->sizes); goto out_no_sizes; } - srf->offsets = kmalloc_array(srf->num_sizes, - sizeof(*srf->offsets), + srf->offsets = kmalloc_array(metadata->num_sizes, sizeof(*srf->offsets), GFP_KERNEL); if (unlikely(!srf->offsets)) { ret = -ENOMEM; goto out_no_offsets; } - srf->base_size = *srf->sizes; - srf->autogen_filter = SVGA3D_TEX_FILTER_NONE; - srf->multisample_count = 0; - srf->multisample_pattern = SVGA3D_MS_PATTERN_NONE; - srf->quality_level = SVGA3D_MS_QUALITY_NONE; + metadata->base_size = *srf->metadata.sizes; + metadata->autogen_filter = SVGA3D_TEX_FILTER_NONE; + metadata->multisample_count = 0; + metadata->multisample_pattern = SVGA3D_MS_PATTERN_NONE; + metadata->quality_level = SVGA3D_MS_QUALITY_NONE; cur_bo_offset = 0; cur_offset = srf->offsets; - cur_size = srf->sizes; + cur_size = metadata->sizes; for (i = 0; i < DRM_VMW_MAX_SURFACE_FACES; ++i) { - for (j = 0; j < srf->mip_levels[i]; ++j) { + for (j = 0; j < metadata->mip_levels[i]; ++j) { uint32_t stride = svga3dsurface_calculate_pitch (desc, cur_size); @@ -843,11 +847,11 @@ int vmw_surface_define_ioctl(struct drm_device *dev, void *data, } } res->backup_size = cur_bo_offset; - if (srf->scanout && - srf->num_sizes == 1 && - srf->sizes[0].width == 64 && - srf->sizes[0].height == 64 && - srf->format == SVGA3D_A8R8G8B8) { + if (metadata->scanout && + metadata->num_sizes == 1 && + metadata->sizes[0].width == 64 && + metadata->sizes[0].height == 64 && + metadata->format == SVGA3D_A8R8G8B8) { srf->snooper.image = kzalloc(64 * 64 * 4, GFP_KERNEL); if (!srf->snooper.image) { @@ -911,7 +915,7 @@ int vmw_surface_define_ioctl(struct drm_device *dev, void *data, out_no_copy: kfree(srf->offsets); out_no_offsets: - kfree(srf->sizes); + kfree(metadata->sizes); out_no_sizes: ttm_prime_object_kfree(user_srf, prime); out_no_user_srf: @@ -1031,18 +1035,19 @@ int vmw_surface_reference_ioctl(struct drm_device *dev, void *data, srf = &user_srf->srf; /* Downcast of flags when sending back to user space */ - rep->flags = (uint32_t)srf->flags; - rep->format = srf->format; - memcpy(rep->mip_levels, srf->mip_levels, sizeof(srf->mip_levels)); + rep->flags = (uint32_t)srf->metadata.flags; + rep->format = srf->metadata.format; + memcpy(rep->mip_levels, srf->metadata.mip_levels, + sizeof(srf->metadata.mip_levels)); user_sizes = (struct drm_vmw_size __user *)(unsigned long) rep->size_addr; if (user_sizes) - ret = copy_to_user(user_sizes, &srf->base_size, - sizeof(srf->base_size)); + ret = copy_to_user(user_sizes, &srf->metadata.base_size, + sizeof(srf->metadata.base_size)); if (unlikely(ret != 0)) { VMW_DEBUG_USER("copy_to_user failed %p %u\n", user_sizes, - srf->num_sizes); + srf->metadata.num_sizes); ttm_ref_object_base_unref(tfile, base->handle, TTM_REF_USAGE); ret = -EFAULT; } @@ -1062,6 +1067,7 @@ static int vmw_gb_surface_create(struct vmw_resource *res) { struct vmw_private *dev_priv = res->dev_priv; struct vmw_surface *srf = vmw_res_to_srf(res); + struct vmw_surface_metadata *metadata = &srf->metadata; uint32_t cmd_len, cmd_id, submit_len; int ret; struct { @@ -1076,6 +1082,10 @@ static int vmw_gb_surface_create(struct vmw_resource *res) SVGA3dCmdHeader header; SVGA3dCmdDefineGBSurface_v3 body; } *cmd3; + struct { + SVGA3dCmdHeader header; + SVGA3dCmdDefineGBSurface_v4 body; + } *cmd4; if (likely(res->id != -1)) return 0; @@ -1092,12 +1102,16 @@ static int vmw_gb_surface_create(struct vmw_resource *res) goto out_no_fifo; } - if (dev_priv->has_sm4_1 && srf->array_size > 0) { + if (has_sm5_context(dev_priv) && metadata->array_size > 0) { + cmd_id = SVGA_3D_CMD_DEFINE_GB_SURFACE_V4; + cmd_len = sizeof(cmd4->body); + submit_len = sizeof(*cmd4); + } else if (has_sm4_1_context(dev_priv) && metadata->array_size > 0) { cmd_id = SVGA_3D_CMD_DEFINE_GB_SURFACE_V3; cmd_len = sizeof(cmd3->body); submit_len = sizeof(*cmd3); - } else if (srf->array_size > 0) { - /* has_dx checked on creation time. */ + } else if (metadata->array_size > 0) { + /* VMW_SM_4 support verified at creation time. */ cmd_id = SVGA_3D_CMD_DEFINE_GB_SURFACE_V2; cmd_len = sizeof(cmd2->body); submit_len = sizeof(*cmd2); @@ -1110,51 +1124,68 @@ static int vmw_gb_surface_create(struct vmw_resource *res) cmd = VMW_FIFO_RESERVE(dev_priv, submit_len); cmd2 = (typeof(cmd2))cmd; cmd3 = (typeof(cmd3))cmd; + cmd4 = (typeof(cmd4))cmd; if (unlikely(!cmd)) { ret = -ENOMEM; goto out_no_fifo; } - if (dev_priv->has_sm4_1 && srf->array_size > 0) { + if (has_sm5_context(dev_priv) && metadata->array_size > 0) { + cmd4->header.id = cmd_id; + cmd4->header.size = cmd_len; + cmd4->body.sid = srf->res.id; + cmd4->body.surfaceFlags = metadata->flags; + cmd4->body.format = metadata->format; + cmd4->body.numMipLevels = metadata->mip_levels[0]; + cmd4->body.multisampleCount = metadata->multisample_count; + cmd4->body.multisamplePattern = metadata->multisample_pattern; + cmd4->body.qualityLevel = metadata->quality_level; + cmd4->body.autogenFilter = metadata->autogen_filter; + cmd4->body.size.width = metadata->base_size.width; + cmd4->body.size.height = metadata->base_size.height; + cmd4->body.size.depth = metadata->base_size.depth; + cmd4->body.arraySize = metadata->array_size; + cmd4->body.bufferByteStride = metadata->buffer_byte_stride; + } else if (has_sm4_1_context(dev_priv) && metadata->array_size > 0) { cmd3->header.id = cmd_id; cmd3->header.size = cmd_len; cmd3->body.sid = srf->res.id; - cmd3->body.surfaceFlags = srf->flags; - cmd3->body.format = srf->format; - cmd3->body.numMipLevels = srf->mip_levels[0]; - cmd3->body.multisampleCount = srf->multisample_count; - cmd3->body.multisamplePattern = srf->multisample_pattern; - cmd3->body.qualityLevel = srf->quality_level; - cmd3->body.autogenFilter = srf->autogen_filter; - cmd3->body.size.width = srf->base_size.width; - cmd3->body.size.height = srf->base_size.height; - cmd3->body.size.depth = srf->base_size.depth; - cmd3->body.arraySize = srf->array_size; - } else if (srf->array_size > 0) { + cmd3->body.surfaceFlags = metadata->flags; + cmd3->body.format = metadata->format; + cmd3->body.numMipLevels = metadata->mip_levels[0]; + cmd3->body.multisampleCount = metadata->multisample_count; + cmd3->body.multisamplePattern = metadata->multisample_pattern; + cmd3->body.qualityLevel = metadata->quality_level; + cmd3->body.autogenFilter = metadata->autogen_filter; + cmd3->body.size.width = metadata->base_size.width; + cmd3->body.size.height = metadata->base_size.height; + cmd3->body.size.depth = metadata->base_size.depth; + cmd3->body.arraySize = metadata->array_size; + } else if (metadata->array_size > 0) { cmd2->header.id = cmd_id; cmd2->header.size = cmd_len; cmd2->body.sid = srf->res.id; - cmd2->body.surfaceFlags = srf->flags; - cmd2->body.format = srf->format; - cmd2->body.numMipLevels = srf->mip_levels[0]; - cmd2->body.multisampleCount = srf->multisample_count; - cmd2->body.autogenFilter = srf->autogen_filter; - cmd2->body.size.width = srf->base_size.width; - cmd2->body.size.height = srf->base_size.height; - cmd2->body.size.depth = srf->base_size.depth; - cmd2->body.arraySize = srf->array_size; + cmd2->body.surfaceFlags = metadata->flags; + cmd2->body.format = metadata->format; + cmd2->body.numMipLevels = metadata->mip_levels[0]; + cmd2->body.multisampleCount = metadata->multisample_count; + cmd2->body.autogenFilter = metadata->autogen_filter; + cmd2->body.size.width = metadata->base_size.width; + cmd2->body.size.height = metadata->base_size.height; + cmd2->body.size.depth = metadata->base_size.depth; + cmd2->body.arraySize = metadata->array_size; } else { cmd->header.id = cmd_id; cmd->header.size = cmd_len; cmd->body.sid = srf->res.id; - cmd->body.surfaceFlags = srf->flags; - cmd->body.format = srf->format; - cmd->body.numMipLevels = srf->mip_levels[0]; - cmd->body.multisampleCount = srf->multisample_count; - cmd->body.autogenFilter = srf->autogen_filter; - cmd->body.size.width = srf->base_size.width; - cmd->body.size.height = srf->base_size.height; - cmd->body.size.depth = srf->base_size.depth; + cmd->body.surfaceFlags = metadata->flags; + cmd->body.format = metadata->format; + cmd->body.numMipLevels = metadata->mip_levels[0]; + cmd->body.multisampleCount = metadata->multisample_count; + cmd->body.autogenFilter = metadata->autogen_filter; + cmd->body.size.width = metadata->base_size.width; + cmd->body.size.height = metadata->base_size.height; + cmd->body.size.depth = metadata->base_size.depth; } vmw_fifo_commit(dev_priv, submit_len); @@ -1314,7 +1345,6 @@ static int vmw_gb_surface_destroy(struct vmw_resource *res) return 0; } - /** * vmw_gb_surface_define_ioctl - Ioctl function implementing * the user surface define functionality. @@ -1336,6 +1366,7 @@ int vmw_gb_surface_define_ioctl(struct drm_device *dev, void *data, req_ext.svga3d_flags_upper_32_bits = 0; req_ext.multisample_pattern = SVGA3D_MS_PATTERN_NONE; req_ext.quality_level = SVGA3D_MS_QUALITY_NONE; + req_ext.buffer_byte_stride = 0; req_ext.must_be_zero = 0; return vmw_gb_surface_define_internal(dev, &req_ext, rep, file_priv); @@ -1371,171 +1402,6 @@ int vmw_gb_surface_reference_ioctl(struct drm_device *dev, void *data, } /** - * vmw_surface_gb_priv_define - Define a private GB surface - * - * @dev: Pointer to a struct drm_device - * @user_accounting_size: Used to track user-space memory usage, set - * to 0 for kernel mode only memory - * @svga3d_flags: SVGA3d surface flags for the device - * @format: requested surface format - * @for_scanout: true if inteded to be used for scanout buffer - * @num_mip_levels: number of MIP levels - * @multisample_count: - * @array_size: Surface array size. - * @size: width, heigh, depth of the surface requested - * @multisample_pattern: Multisampling pattern when msaa is supported - * @quality_level: Precision settings - * @user_srf_out: allocated user_srf. Set to NULL on failure. - * - * GB surfaces allocated by this function will not have a user mode handle, and - * thus will only be visible to vmwgfx. For optimization reasons the - * surface may later be given a user mode handle by another function to make - * it available to user mode drivers. - */ -int vmw_surface_gb_priv_define(struct drm_device *dev, - uint32_t user_accounting_size, - SVGA3dSurfaceAllFlags svga3d_flags, - SVGA3dSurfaceFormat format, - bool for_scanout, - uint32_t num_mip_levels, - uint32_t multisample_count, - uint32_t array_size, - struct drm_vmw_size size, - SVGA3dMSPattern multisample_pattern, - SVGA3dMSQualityLevel quality_level, - struct vmw_surface **srf_out) -{ - struct vmw_private *dev_priv = vmw_priv(dev); - struct vmw_user_surface *user_srf; - struct ttm_operation_ctx ctx = { - .interruptible = true, - .no_wait_gpu = false - }; - struct vmw_surface *srf; - int ret; - u32 num_layers = 1; - u32 sample_count = 1; - - *srf_out = NULL; - - if (for_scanout) { - if (!svga3dsurface_is_screen_target_format(format)) { - VMW_DEBUG_USER("Invalid Screen Target surface format."); - return -EINVAL; - } - - if (size.width > dev_priv->texture_max_width || - size.height > dev_priv->texture_max_height) { - VMW_DEBUG_USER("%ux%u\n, exceeds max surface size %ux%u", - size.width, size.height, - dev_priv->texture_max_width, - dev_priv->texture_max_height); - return -EINVAL; - } - } else { - const struct svga3d_surface_desc *desc; - - desc = svga3dsurface_get_desc(format); - if (unlikely(desc->block_desc == SVGA3DBLOCKDESC_NONE)) { - VMW_DEBUG_USER("Invalid surface format.\n"); - return -EINVAL; - } - } - - /* array_size must be null for non-GL3 host. */ - if (array_size > 0 && !dev_priv->has_dx) { - VMW_DEBUG_USER("Tried to create DX surface on non-DX host.\n"); - return -EINVAL; - } - - ret = ttm_read_lock(&dev_priv->reservation_sem, true); - if (unlikely(ret != 0)) - return ret; - - ret = ttm_mem_global_alloc(vmw_mem_glob(dev_priv), - user_accounting_size, &ctx); - if (unlikely(ret != 0)) { - if (ret != -ERESTARTSYS) - DRM_ERROR("Out of graphics memory for surface" - " creation.\n"); - goto out_unlock; - } - - user_srf = kzalloc(sizeof(*user_srf), GFP_KERNEL); - if (unlikely(!user_srf)) { - ret = -ENOMEM; - goto out_no_user_srf; - } - - *srf_out = &user_srf->srf; - user_srf->size = user_accounting_size; - user_srf->prime.base.shareable = false; - user_srf->prime.base.tfile = NULL; - - srf = &user_srf->srf; - srf->flags = svga3d_flags; - srf->format = format; - srf->scanout = for_scanout; - srf->mip_levels[0] = num_mip_levels; - srf->num_sizes = 1; - srf->sizes = NULL; - srf->offsets = NULL; - srf->base_size = size; - srf->autogen_filter = SVGA3D_TEX_FILTER_NONE; - srf->array_size = array_size; - srf->multisample_count = multisample_count; - srf->multisample_pattern = multisample_pattern; - srf->quality_level = quality_level; - - if (array_size) - num_layers = array_size; - else if (svga3d_flags & SVGA3D_SURFACE_CUBEMAP) - num_layers = SVGA3D_MAX_SURFACE_FACES; - - if (srf->flags & SVGA3D_SURFACE_MULTISAMPLE) - sample_count = srf->multisample_count; - - srf->res.backup_size = - svga3dsurface_get_serialized_size_extended(srf->format, - srf->base_size, - srf->mip_levels[0], - num_layers, - sample_count); - - if (srf->flags & SVGA3D_SURFACE_BIND_STREAM_OUTPUT) - srf->res.backup_size += sizeof(SVGA3dDXSOState); - - /* - * Don't set SVGA3D_SURFACE_SCREENTARGET flag for a scanout surface with - * size greater than STDU max width/height. This is really a workaround - * to support creation of big framebuffer requested by some user-space - * for whole topology. That big framebuffer won't really be used for - * binding with screen target as during prepare_fb a separate surface is - * created so it's safe to ignore SVGA3D_SURFACE_SCREENTARGET flag. - */ - if (dev_priv->active_display_unit == vmw_du_screen_target && - for_scanout && size.width <= dev_priv->stdu_max_width && - size.height <= dev_priv->stdu_max_height) - srf->flags |= SVGA3D_SURFACE_SCREENTARGET; - - /* - * From this point, the generic resource management functions - * destroy the object on failure. - */ - ret = vmw_surface_init(dev_priv, srf, vmw_user_surface_free); - - ttm_read_unlock(&dev_priv->reservation_sem); - return ret; - -out_no_user_srf: - ttm_mem_global_free(vmw_mem_glob(dev_priv), user_accounting_size); - -out_unlock: - ttm_read_unlock(&dev_priv->reservation_sem); - return ret; -} - -/** * vmw_gb_surface_define_ext_ioctl - Ioctl function implementing * the user surface define functionality. * @@ -1588,43 +1454,60 @@ vmw_gb_surface_define_internal(struct drm_device *dev, struct drm_vmw_gb_surface_create_rep *rep, struct drm_file *file_priv) { + struct ttm_object_file *tfile = vmw_fpriv(file_priv)->tfile; struct vmw_private *dev_priv = vmw_priv(dev); struct vmw_user_surface *user_srf; + struct vmw_surface_metadata metadata = {0}; struct vmw_surface *srf; struct vmw_resource *res; struct vmw_resource *tmp; - struct ttm_object_file *tfile = vmw_fpriv(file_priv)->tfile; - int ret; + int ret = 0; uint32_t size; uint32_t backup_handle = 0; SVGA3dSurfaceAllFlags svga3d_flags_64 = SVGA3D_FLAGS_64(req->svga3d_flags_upper_32_bits, req->base.svga3d_flags); - if (!dev_priv->has_sm4_1) { - /* - * If SM4_1 is not support then cannot send 64-bit flag to - * device. - */ + /* array_size must be null for non-GL3 host. */ + if (req->base.array_size > 0 && !has_sm4_context(dev_priv)) { + VMW_DEBUG_USER("SM4 surface not supported.\n"); + return -EINVAL; + } + + if (!has_sm4_1_context(dev_priv)) { if (req->svga3d_flags_upper_32_bits != 0) - return -EINVAL; + ret = -EINVAL; if (req->base.multisample_count != 0) - return -EINVAL; + ret = -EINVAL; if (req->multisample_pattern != SVGA3D_MS_PATTERN_NONE) - return -EINVAL; + ret = -EINVAL; if (req->quality_level != SVGA3D_MS_QUALITY_NONE) - return -EINVAL; + ret = -EINVAL; + + if (ret) { + VMW_DEBUG_USER("SM4.1 surface not supported.\n"); + return ret; + } + } + + if (req->buffer_byte_stride > 0 && !has_sm5_context(dev_priv)) { + VMW_DEBUG_USER("SM5 surface not supported.\n"); + return -EINVAL; } if ((svga3d_flags_64 & SVGA3D_SURFACE_MULTISAMPLE) && - req->base.multisample_count == 0) + req->base.multisample_count == 0) { + VMW_DEBUG_USER("Invalid sample count.\n"); return -EINVAL; + } - if (req->base.mip_levels > DRM_VMW_MAX_MIP_LEVELS) + if (req->base.mip_levels > DRM_VMW_MAX_MIP_LEVELS) { + VMW_DEBUG_USER("Invalid mip level.\n"); return -EINVAL; + } if (unlikely(vmw_user_surface_size == 0)) vmw_user_surface_size = ttm_round_pot(sizeof(*user_srf)) + @@ -1632,22 +1515,25 @@ vmw_gb_surface_define_internal(struct drm_device *dev, size = vmw_user_surface_size; + metadata.flags = svga3d_flags_64; + metadata.format = req->base.format; + metadata.mip_levels[0] = req->base.mip_levels; + metadata.multisample_count = req->base.multisample_count; + metadata.multisample_pattern = req->multisample_pattern; + metadata.quality_level = req->quality_level; + metadata.array_size = req->base.array_size; + metadata.buffer_byte_stride = req->buffer_byte_stride; + metadata.num_sizes = 1; + metadata.base_size = req->base.base_size; + metadata.scanout = req->base.drm_surface_flags & + drm_vmw_surface_flag_scanout; + /* Define a surface based on the parameters. */ - ret = vmw_surface_gb_priv_define(dev, - size, - svga3d_flags_64, - req->base.format, - req->base.drm_surface_flags & - drm_vmw_surface_flag_scanout, - req->base.mip_levels, - req->base.multisample_count, - req->base.array_size, - req->base.base_size, - req->multisample_pattern, - req->quality_level, - &srf); - if (unlikely(ret != 0)) + ret = vmw_gb_surface_define(dev_priv, size, &metadata, &srf); + if (ret != 0) { + VMW_DEBUG_USER("Failed to define surface.\n"); return ret; + } user_srf = container_of(srf, struct vmw_user_surface, srf); if (drm_is_primary_client(file_priv)) @@ -1762,6 +1648,7 @@ vmw_gb_surface_reference_internal(struct drm_device *dev, struct ttm_object_file *tfile = vmw_fpriv(file_priv)->tfile; struct vmw_surface *srf; struct vmw_user_surface *user_srf; + struct vmw_surface_metadata *metadata; struct ttm_base_object *base; uint32_t backup_handle; int ret = -EINVAL; @@ -1777,6 +1664,7 @@ vmw_gb_surface_reference_internal(struct drm_device *dev, DRM_ERROR("Shared GB surface is missing a backup buffer.\n"); goto out_bad_resource; } + metadata = &srf->metadata; mutex_lock(&dev_priv->cmdbuf_mutex); /* Protect res->backup */ ret = vmw_user_bo_reference(tfile, srf->res.backup, &backup_handle); @@ -1790,15 +1678,15 @@ vmw_gb_surface_reference_internal(struct drm_device *dev, goto out_bad_resource; } - rep->creq.base.svga3d_flags = SVGA3D_FLAGS_LOWER_32(srf->flags); - rep->creq.base.format = srf->format; - rep->creq.base.mip_levels = srf->mip_levels[0]; + rep->creq.base.svga3d_flags = SVGA3D_FLAGS_LOWER_32(metadata->flags); + rep->creq.base.format = metadata->format; + rep->creq.base.mip_levels = metadata->mip_levels[0]; rep->creq.base.drm_surface_flags = 0; - rep->creq.base.multisample_count = srf->multisample_count; - rep->creq.base.autogen_filter = srf->autogen_filter; - rep->creq.base.array_size = srf->array_size; + rep->creq.base.multisample_count = metadata->multisample_count; + rep->creq.base.autogen_filter = metadata->autogen_filter; + rep->creq.base.array_size = metadata->array_size; rep->creq.base.buffer_handle = backup_handle; - rep->creq.base.base_size = srf->base_size; + rep->creq.base.base_size = metadata->base_size; rep->crep.handle = user_srf->prime.base.handle; rep->crep.backup_size = srf->res.backup_size; rep->crep.buffer_handle = backup_handle; @@ -1808,9 +1696,9 @@ vmw_gb_surface_reference_internal(struct drm_device *dev, rep->creq.version = drm_vmw_gb_surface_v1; rep->creq.svga3d_flags_upper_32_bits = - SVGA3D_FLAGS_UPPER_32(srf->flags); - rep->creq.multisample_pattern = srf->multisample_pattern; - rep->creq.quality_level = srf->quality_level; + SVGA3D_FLAGS_UPPER_32(metadata->flags); + rep->creq.multisample_pattern = metadata->multisample_pattern; + rep->creq.quality_level = metadata->quality_level; rep->creq.must_be_zero = 0; out_bad_resource: @@ -1968,7 +1856,7 @@ static void vmw_surface_dirty_range_add(struct vmw_resource *res, size_t start, start >= res->backup_offset + res->backup_size)) return; - if (srf->format == SVGA3D_BUFFER) + if (srf->metadata.format == SVGA3D_BUFFER) vmw_surface_buf_dirty_range_add(res, start, end); else vmw_surface_tex_dirty_range_add(res, start, end); @@ -2058,6 +1946,7 @@ static int vmw_surface_dirty_sync(struct vmw_resource *res) static int vmw_surface_dirty_alloc(struct vmw_resource *res) { struct vmw_surface *srf = vmw_res_to_srf(res); + const struct vmw_surface_metadata *metadata = &srf->metadata; struct vmw_surface_dirty *dirty; u32 num_layers = 1; u32 num_mip; @@ -2070,12 +1959,12 @@ static int vmw_surface_dirty_alloc(struct vmw_resource *res) }; int ret; - if (srf->array_size) - num_layers = srf->array_size; - else if (srf->flags & SVGA3D_SURFACE_CUBEMAP) + if (metadata->array_size) + num_layers = metadata->array_size; + else if (metadata->flags & SVGA3D_SURFACE_CUBEMAP) num_layers *= SVGA3D_MAX_SURFACE_FACES; - num_mip = srf->mip_levels[0]; + num_mip = metadata->mip_levels[0]; if (!num_mip) num_mip = 1; @@ -2096,9 +1985,10 @@ static int vmw_surface_dirty_alloc(struct vmw_resource *res) goto out_no_dirty; } - num_samples = max_t(u32, 1, srf->multisample_count); - ret = svga3dsurface_setup_cache(&srf->base_size, srf->format, num_mip, - num_layers, num_samples, &dirty->cache); + num_samples = max_t(u32, 1, metadata->multisample_count); + ret = svga3dsurface_setup_cache(&metadata->base_size, metadata->format, + num_mip, num_layers, num_samples, + &dirty->cache); if (ret) goto out_no_cache; @@ -2153,3 +2043,147 @@ static int vmw_surface_clean(struct vmw_resource *res) return 0; } + +/* + * vmw_gb_surface_define - Define a private GB surface + * + * @dev_priv: Pointer to a device private. + * @user_accounting_size: Used to track user-space memory usage, set + * to 0 for kernel mode only memory + * @metadata: Metadata representing the surface to create. + * @user_srf_out: allocated user_srf. Set to NULL on failure. + * + * GB surfaces allocated by this function will not have a user mode handle, and + * thus will only be visible to vmwgfx. For optimization reasons the + * surface may later be given a user mode handle by another function to make + * it available to user mode drivers. + */ +int vmw_gb_surface_define(struct vmw_private *dev_priv, + uint32_t user_accounting_size, + const struct vmw_surface_metadata *req, + struct vmw_surface **srf_out) +{ + struct vmw_surface_metadata *metadata; + struct vmw_user_surface *user_srf; + struct vmw_surface *srf; + struct ttm_operation_ctx ctx = { + .interruptible = true, + .no_wait_gpu = false + }; + u32 sample_count = 1; + u32 num_layers = 1; + int ret; + + *srf_out = NULL; + + if (req->scanout) { + if (!svga3dsurface_is_screen_target_format(req->format)) { + VMW_DEBUG_USER("Invalid Screen Target surface format."); + return -EINVAL; + } + + if (req->base_size.width > dev_priv->texture_max_width || + req->base_size.height > dev_priv->texture_max_height) { + VMW_DEBUG_USER("%ux%u\n, exceed max surface size %ux%u", + req->base_size.width, + req->base_size.height, + dev_priv->texture_max_width, + dev_priv->texture_max_height); + return -EINVAL; + } + } else { + const struct svga3d_surface_desc *desc = + svga3dsurface_get_desc(req->format); + + if (desc->block_desc == SVGA3DBLOCKDESC_NONE) { + VMW_DEBUG_USER("Invalid surface format.\n"); + return -EINVAL; + } + } + + if (req->autogen_filter != SVGA3D_TEX_FILTER_NONE) + return -EINVAL; + + if (req->num_sizes != 1) + return -EINVAL; + + if (req->sizes != NULL) + return -EINVAL; + + ret = ttm_read_lock(&dev_priv->reservation_sem, true); + if (unlikely(ret != 0)) + return ret; + + ret = ttm_mem_global_alloc(vmw_mem_glob(dev_priv), + user_accounting_size, &ctx); + if (ret != 0) { + if (ret != -ERESTARTSYS) + DRM_ERROR("Out of graphics memory for surface.\n"); + goto out_unlock; + } + + user_srf = kzalloc(sizeof(*user_srf), GFP_KERNEL); + if (unlikely(!user_srf)) { + ret = -ENOMEM; + goto out_no_user_srf; + } + + *srf_out = &user_srf->srf; + user_srf->size = user_accounting_size; + user_srf->prime.base.shareable = false; + user_srf->prime.base.tfile = NULL; + + srf = &user_srf->srf; + srf->metadata = *req; + srf->offsets = NULL; + + metadata = &srf->metadata; + + if (metadata->array_size) + num_layers = req->array_size; + else if (metadata->flags & SVGA3D_SURFACE_CUBEMAP) + num_layers = SVGA3D_MAX_SURFACE_FACES; + + if (metadata->flags & SVGA3D_SURFACE_MULTISAMPLE) + sample_count = metadata->multisample_count; + + srf->res.backup_size = + svga3dsurface_get_serialized_size_extended(metadata->format, + metadata->base_size, + metadata->mip_levels[0], + num_layers, + sample_count); + + if (metadata->flags & SVGA3D_SURFACE_BIND_STREAM_OUTPUT) + srf->res.backup_size += sizeof(SVGA3dDXSOState); + + /* + * Don't set SVGA3D_SURFACE_SCREENTARGET flag for a scanout surface with + * size greater than STDU max width/height. This is really a workaround + * to support creation of big framebuffer requested by some user-space + * for whole topology. That big framebuffer won't really be used for + * binding with screen target as during prepare_fb a separate surface is + * created so it's safe to ignore SVGA3D_SURFACE_SCREENTARGET flag. + */ + if (dev_priv->active_display_unit == vmw_du_screen_target && + metadata->scanout && + metadata->base_size.width <= dev_priv->stdu_max_width && + metadata->base_size.height <= dev_priv->stdu_max_height) + metadata->flags |= SVGA3D_SURFACE_SCREENTARGET; + + /* + * From this point, the generic resource management functions + * destroy the object on failure. + */ + ret = vmw_surface_init(dev_priv, srf, vmw_user_surface_free); + + ttm_read_unlock(&dev_priv->reservation_sem); + return ret; + +out_no_user_srf: + ttm_mem_global_free(vmw_mem_glob(dev_priv), user_accounting_size); + +out_unlock: + ttm_read_unlock(&dev_priv->reservation_sem); + return ret; +} diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_thp.c b/drivers/gpu/drm/vmwgfx/vmwgfx_thp.c new file mode 100644 index 000000000000..b7c816ba7166 --- /dev/null +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_thp.c @@ -0,0 +1,166 @@ +// SPDX-License-Identifier: GPL-2.0 OR MIT +/* + * Huge page-table-entry support for IO memory. + * + * Copyright (C) 2007-2019 Vmware, Inc. All rights reservedd. + */ +#include "vmwgfx_drv.h" +#include <drm/ttm/ttm_module.h> +#include <drm/ttm/ttm_bo_driver.h> +#include <drm/ttm/ttm_placement.h> + +/** + * struct vmw_thp_manager - Range manager implementing huge page alignment + * + * @mm: The underlying range manager. Protected by @lock. + * @lock: Manager lock. + */ +struct vmw_thp_manager { + struct drm_mm mm; + spinlock_t lock; +}; + +static int vmw_thp_insert_aligned(struct drm_mm *mm, struct drm_mm_node *node, + unsigned long align_pages, + const struct ttm_place *place, + struct ttm_mem_reg *mem, + unsigned long lpfn, + enum drm_mm_insert_mode mode) +{ + if (align_pages >= mem->page_alignment && + (!mem->page_alignment || align_pages % mem->page_alignment == 0)) { + return drm_mm_insert_node_in_range(mm, node, + mem->num_pages, + align_pages, 0, + place->fpfn, lpfn, mode); + } + + return -ENOSPC; +} + +static int vmw_thp_get_node(struct ttm_mem_type_manager *man, + struct ttm_buffer_object *bo, + const struct ttm_place *place, + struct ttm_mem_reg *mem) +{ + struct vmw_thp_manager *rman = (struct vmw_thp_manager *) man->priv; + struct drm_mm *mm = &rman->mm; + struct drm_mm_node *node; + unsigned long align_pages; + unsigned long lpfn; + enum drm_mm_insert_mode mode = DRM_MM_INSERT_BEST; + int ret; + + node = kzalloc(sizeof(*node), GFP_KERNEL); + if (!node) + return -ENOMEM; + + lpfn = place->lpfn; + if (!lpfn) + lpfn = man->size; + + mode = DRM_MM_INSERT_BEST; + if (place->flags & TTM_PL_FLAG_TOPDOWN) + mode = DRM_MM_INSERT_HIGH; + + spin_lock(&rman->lock); + if (IS_ENABLED(CONFIG_HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD)) { + align_pages = (HPAGE_PUD_SIZE >> PAGE_SHIFT); + if (mem->num_pages >= align_pages) { + ret = vmw_thp_insert_aligned(mm, node, align_pages, + place, mem, lpfn, mode); + if (!ret) + goto found_unlock; + } + } + + align_pages = (HPAGE_PMD_SIZE >> PAGE_SHIFT); + if (mem->num_pages >= align_pages) { + ret = vmw_thp_insert_aligned(mm, node, align_pages, place, mem, + lpfn, mode); + if (!ret) + goto found_unlock; + } + + ret = drm_mm_insert_node_in_range(mm, node, mem->num_pages, + mem->page_alignment, 0, + place->fpfn, lpfn, mode); +found_unlock: + spin_unlock(&rman->lock); + + if (unlikely(ret)) { + kfree(node); + } else { + mem->mm_node = node; + mem->start = node->start; + } + + return 0; +} + + + +static void vmw_thp_put_node(struct ttm_mem_type_manager *man, + struct ttm_mem_reg *mem) +{ + struct vmw_thp_manager *rman = (struct vmw_thp_manager *) man->priv; + + if (mem->mm_node) { + spin_lock(&rman->lock); + drm_mm_remove_node(mem->mm_node); + spin_unlock(&rman->lock); + + kfree(mem->mm_node); + mem->mm_node = NULL; + } +} + +static int vmw_thp_init(struct ttm_mem_type_manager *man, + unsigned long p_size) +{ + struct vmw_thp_manager *rman; + + rman = kzalloc(sizeof(*rman), GFP_KERNEL); + if (!rman) + return -ENOMEM; + + drm_mm_init(&rman->mm, 0, p_size); + spin_lock_init(&rman->lock); + man->priv = rman; + return 0; +} + +static int vmw_thp_takedown(struct ttm_mem_type_manager *man) +{ + struct vmw_thp_manager *rman = (struct vmw_thp_manager *) man->priv; + struct drm_mm *mm = &rman->mm; + + spin_lock(&rman->lock); + if (drm_mm_clean(mm)) { + drm_mm_takedown(mm); + spin_unlock(&rman->lock); + kfree(rman); + man->priv = NULL; + return 0; + } + spin_unlock(&rman->lock); + return -EBUSY; +} + +static void vmw_thp_debug(struct ttm_mem_type_manager *man, + struct drm_printer *printer) +{ + struct vmw_thp_manager *rman = (struct vmw_thp_manager *) man->priv; + + spin_lock(&rman->lock); + drm_mm_print(&rman->mm, printer); + spin_unlock(&rman->lock); +} + +const struct ttm_mem_type_manager_func vmw_thp_func = { + .init = vmw_thp_init, + .takedown = vmw_thp_takedown, + .get_node = vmw_thp_get_node, + .put_node = vmw_thp_put_node, + .debug = vmw_thp_debug +}; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c index 3f3b2c7a208a..bf0bc4697959 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c @@ -749,7 +749,7 @@ static int vmw_init_mem_type(struct ttm_bo_device *bdev, uint32_t type, break; case TTM_PL_VRAM: /* "On-card" video ram */ - man->func = &ttm_bo_manager_func; + man->func = &vmw_thp_func; man->gpu_offset = 0; man->flags = TTM_MEMTYPE_FLAG_FIXED | TTM_MEMTYPE_FLAG_MAPPABLE; man->available_caching = TTM_PL_FLAG_CACHED; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_glue.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_glue.c index aa7e50f63b94..3c03b1746661 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_glue.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_glue.c @@ -34,7 +34,10 @@ int vmw_mmap(struct file *filp, struct vm_area_struct *vma) .page_mkwrite = vmw_bo_vm_mkwrite, .fault = vmw_bo_vm_fault, .open = ttm_bo_vm_open, - .close = ttm_bo_vm_close + .close = ttm_bo_vm_close, +#ifdef CONFIG_TRANSPARENT_HUGEPAGE + .huge_fault = vmw_bo_vm_huge_fault, +#endif }; struct drm_file *file_priv = filp->private_data; struct vmw_private *dev_priv = vmw_priv(file_priv->minor->dev); diff --git a/drivers/hid/hid-alps.c b/drivers/hid/hid-alps.c index ae79a7c66737..fa704153cb00 100644 --- a/drivers/hid/hid-alps.c +++ b/drivers/hid/hid-alps.c @@ -730,7 +730,7 @@ static int alps_input_configured(struct hid_device *hdev, struct hid_input *hi) if (data->has_sp) { input2 = input_allocate_device(); if (!input2) { - input_free_device(input2); + ret = -ENOMEM; goto exit; } diff --git a/drivers/hid/hid-apple.c b/drivers/hid/hid-apple.c index 6ac8becc2372..d732d1d10caf 100644 --- a/drivers/hid/hid-apple.c +++ b/drivers/hid/hid-apple.c @@ -340,7 +340,8 @@ static int apple_input_mapping(struct hid_device *hdev, struct hid_input *hi, unsigned long **bit, int *max) { if (usage->hid == (HID_UP_CUSTOM | 0x0003) || - usage->hid == (HID_UP_MSVENDOR | 0x0003)) { + usage->hid == (HID_UP_MSVENDOR | 0x0003) || + usage->hid == (HID_UP_HPVENDOR2 | 0x0003)) { /* The fn key on Apple USB keyboards */ set_bit(EV_REP, hi->input->evbit); hid_map_usage_clear(hi, usage, bit, max, EV_KEY, KEY_FN); diff --git a/drivers/hid/hid-bigbenff.c b/drivers/hid/hid-bigbenff.c index 3f6abd190df4..db6da21ade06 100644 --- a/drivers/hid/hid-bigbenff.c +++ b/drivers/hid/hid-bigbenff.c @@ -174,6 +174,7 @@ static __u8 pid0902_rdesc_fixed[] = { struct bigben_device { struct hid_device *hid; struct hid_report *report; + bool removed; u8 led_state; /* LED1 = 1 .. LED4 = 8 */ u8 right_motor_on; /* right motor off/on 0/1 */ u8 left_motor_force; /* left motor force 0-255 */ @@ -190,6 +191,9 @@ static void bigben_worker(struct work_struct *work) struct bigben_device, worker); struct hid_field *report_field = bigben->report->field[0]; + if (bigben->removed) + return; + if (bigben->work_led) { bigben->work_led = false; report_field->value[0] = 0x01; /* 1 = led message */ @@ -220,10 +224,16 @@ static void bigben_worker(struct work_struct *work) static int hid_bigben_play_effect(struct input_dev *dev, void *data, struct ff_effect *effect) { - struct bigben_device *bigben = data; + struct hid_device *hid = input_get_drvdata(dev); + struct bigben_device *bigben = hid_get_drvdata(hid); u8 right_motor_on; u8 left_motor_force; + if (!bigben) { + hid_err(hid, "no device data\n"); + return 0; + } + if (effect->type != FF_RUMBLE) return 0; @@ -298,8 +308,8 @@ static void bigben_remove(struct hid_device *hid) { struct bigben_device *bigben = hid_get_drvdata(hid); + bigben->removed = true; cancel_work_sync(&bigben->worker); - hid_hw_close(hid); hid_hw_stop(hid); } @@ -319,6 +329,7 @@ static int bigben_probe(struct hid_device *hid, return -ENOMEM; hid_set_drvdata(hid, bigben); bigben->hid = hid; + bigben->removed = false; error = hid_parse(hid); if (error) { @@ -341,10 +352,10 @@ static int bigben_probe(struct hid_device *hid, INIT_WORK(&bigben->worker, bigben_worker); - error = input_ff_create_memless(hidinput->input, bigben, + error = input_ff_create_memless(hidinput->input, NULL, hid_bigben_play_effect); if (error) - return error; + goto error_hw_stop; name_sz = strlen(dev_name(&hid->dev)) + strlen(":red:bigben#") + 1; @@ -354,8 +365,10 @@ static int bigben_probe(struct hid_device *hid, sizeof(struct led_classdev) + name_sz, GFP_KERNEL ); - if (!led) - return -ENOMEM; + if (!led) { + error = -ENOMEM; + goto error_hw_stop; + } name = (void *)(&led[1]); snprintf(name, name_sz, "%s:red:bigben%d", @@ -369,7 +382,7 @@ static int bigben_probe(struct hid_device *hid, bigben->leds[n] = led; error = devm_led_classdev_register(&hid->dev, led); if (error) - return error; + goto error_hw_stop; } /* initial state: LED1 is on, no rumble effect */ @@ -383,6 +396,10 @@ static int bigben_probe(struct hid_device *hid, hid_info(hid, "LED and force feedback support for BigBen gamepad\n"); return 0; + +error_hw_stop: + hid_hw_stop(hid); + return error; } static __u8 *bigben_report_fixup(struct hid_device *hid, __u8 *rdesc, diff --git a/drivers/hid/hid-core.c b/drivers/hid/hid-core.c index 851fe54ea59e..359616e3efbb 100644 --- a/drivers/hid/hid-core.c +++ b/drivers/hid/hid-core.c @@ -1741,7 +1741,9 @@ int hid_report_raw_event(struct hid_device *hid, int type, u8 *data, u32 size, rsize = ((report->size - 1) >> 3) + 1; - if (rsize > HID_MAX_BUFFER_SIZE) + if (report_enum->numbered && rsize >= HID_MAX_BUFFER_SIZE) + rsize = HID_MAX_BUFFER_SIZE - 1; + else if (rsize > HID_MAX_BUFFER_SIZE) rsize = HID_MAX_BUFFER_SIZE; if (csize < rsize) { diff --git a/drivers/hid/hid-google-hammer.c b/drivers/hid/hid-google-hammer.c index 2aa4ed157aec..85a054f1ce38 100644 --- a/drivers/hid/hid-google-hammer.c +++ b/drivers/hid/hid-google-hammer.c @@ -533,6 +533,8 @@ static const struct hid_device_id hammer_devices[] = { { HID_DEVICE(BUS_USB, HID_GROUP_GENERIC, USB_VENDOR_ID_GOOGLE, USB_DEVICE_ID_GOOGLE_MASTERBALL) }, { HID_DEVICE(BUS_USB, HID_GROUP_GENERIC, + USB_VENDOR_ID_GOOGLE, USB_DEVICE_ID_GOOGLE_MOONBALL) }, + { HID_DEVICE(BUS_USB, HID_GROUP_GENERIC, USB_VENDOR_ID_GOOGLE, USB_DEVICE_ID_GOOGLE_STAFF) }, { HID_DEVICE(BUS_USB, HID_GROUP_GENERIC, USB_VENDOR_ID_GOOGLE, USB_DEVICE_ID_GOOGLE_WAND) }, diff --git a/drivers/hid/hid-hyperv.c b/drivers/hid/hid-hyperv.c index dddfca555df9..0b6ee1dee625 100644 --- a/drivers/hid/hid-hyperv.c +++ b/drivers/hid/hid-hyperv.c @@ -193,8 +193,7 @@ static void mousevsc_on_receive_device_info(struct mousevsc_dev *input_device, goto cleanup; /* The pointer is not NULL when we resume from hibernation */ - if (input_device->hid_desc != NULL) - kfree(input_device->hid_desc); + kfree(input_device->hid_desc); input_device->hid_desc = kmemdup(desc, desc->bLength, GFP_ATOMIC); if (!input_device->hid_desc) @@ -207,8 +206,7 @@ static void mousevsc_on_receive_device_info(struct mousevsc_dev *input_device, } /* The pointer is not NULL when we resume from hibernation */ - if (input_device->report_desc != NULL) - kfree(input_device->report_desc); + kfree(input_device->report_desc); input_device->report_desc = kzalloc(input_device->report_desc_size, GFP_ATOMIC); diff --git a/drivers/hid/hid-ids.h b/drivers/hid/hid-ids.h index 3a400ce603c4..9f2213426556 100644 --- a/drivers/hid/hid-ids.h +++ b/drivers/hid/hid-ids.h @@ -478,6 +478,7 @@ #define USB_DEVICE_ID_GOOGLE_WHISKERS 0x5030 #define USB_DEVICE_ID_GOOGLE_MASTERBALL 0x503c #define USB_DEVICE_ID_GOOGLE_MAGNEMITE 0x503d +#define USB_DEVICE_ID_GOOGLE_MOONBALL 0x5044 #define USB_VENDOR_ID_GOTOP 0x08f2 #define USB_DEVICE_ID_SUPER_Q2 0x007f @@ -726,6 +727,7 @@ #define USB_DEVICE_ID_LENOVO_X1_COVER 0x6085 #define USB_DEVICE_ID_LENOVO_X1_TAB 0x60a3 #define USB_DEVICE_ID_LENOVO_X1_TAB3 0x60b5 +#define USB_DEVICE_ID_LENOVO_PIXART_USB_MOUSE_608D 0x608d #define USB_VENDOR_ID_LG 0x1fd2 #define USB_DEVICE_ID_LG_MULTITOUCH 0x0064 diff --git a/drivers/hid/hid-ite.c b/drivers/hid/hid-ite.c index c436e12feb23..6c55682c5974 100644 --- a/drivers/hid/hid-ite.c +++ b/drivers/hid/hid-ite.c @@ -41,8 +41,9 @@ static const struct hid_device_id ite_devices[] = { { HID_USB_DEVICE(USB_VENDOR_ID_ITE, USB_DEVICE_ID_ITE8595) }, { HID_USB_DEVICE(USB_VENDOR_ID_258A, USB_DEVICE_ID_258A_6A88) }, /* ITE8595 USB kbd ctlr, with Synaptics touchpad connected to it. */ - { HID_USB_DEVICE(USB_VENDOR_ID_SYNAPTICS, - USB_DEVICE_ID_SYNAPTICS_ACER_SWITCH5_012) }, + { HID_DEVICE(BUS_USB, HID_GROUP_GENERIC, + USB_VENDOR_ID_SYNAPTICS, + USB_DEVICE_ID_SYNAPTICS_ACER_SWITCH5_012) }, { } }; MODULE_DEVICE_TABLE(hid, ite_devices); diff --git a/drivers/hid/hid-logitech-hidpp.c b/drivers/hid/hid-logitech-hidpp.c index 70e1cb928bf0..094f4f1b6555 100644 --- a/drivers/hid/hid-logitech-hidpp.c +++ b/drivers/hid/hid-logitech-hidpp.c @@ -1256,36 +1256,35 @@ static int hidpp20_battery_map_status_voltage(u8 data[3], int *voltage, { int status; - long charge_sts = (long)data[2]; + long flags = (long) data[2]; - *level = POWER_SUPPLY_CAPACITY_LEVEL_UNKNOWN; - switch (data[2] & 0xe0) { - case 0x00: - status = POWER_SUPPLY_STATUS_CHARGING; - break; - case 0x20: - status = POWER_SUPPLY_STATUS_FULL; - *level = POWER_SUPPLY_CAPACITY_LEVEL_FULL; - break; - case 0x40: + if (flags & 0x80) + switch (flags & 0x07) { + case 0: + status = POWER_SUPPLY_STATUS_CHARGING; + break; + case 1: + status = POWER_SUPPLY_STATUS_FULL; + *level = POWER_SUPPLY_CAPACITY_LEVEL_FULL; + break; + case 2: + status = POWER_SUPPLY_STATUS_NOT_CHARGING; + break; + default: + status = POWER_SUPPLY_STATUS_UNKNOWN; + break; + } + else status = POWER_SUPPLY_STATUS_DISCHARGING; - break; - case 0xe0: - status = POWER_SUPPLY_STATUS_NOT_CHARGING; - break; - default: - status = POWER_SUPPLY_STATUS_UNKNOWN; - } *charge_type = POWER_SUPPLY_CHARGE_TYPE_STANDARD; - if (test_bit(3, &charge_sts)) { + if (test_bit(3, &flags)) { *charge_type = POWER_SUPPLY_CHARGE_TYPE_FAST; } - if (test_bit(4, &charge_sts)) { + if (test_bit(4, &flags)) { *charge_type = POWER_SUPPLY_CHARGE_TYPE_TRICKLE; } - - if (test_bit(5, &charge_sts)) { + if (test_bit(5, &flags)) { *level = POWER_SUPPLY_CAPACITY_LEVEL_CRITICAL; } diff --git a/drivers/hid/hid-picolcd_fb.c b/drivers/hid/hid-picolcd_fb.c index a549c42e8c90..33c102a60992 100644 --- a/drivers/hid/hid-picolcd_fb.c +++ b/drivers/hid/hid-picolcd_fb.c @@ -458,9 +458,9 @@ static ssize_t picolcd_fb_update_rate_show(struct device *dev, if (ret >= PAGE_SIZE) break; else if (i == fb_update_rate) - ret += snprintf(buf+ret, PAGE_SIZE-ret, "[%u] ", i); + ret += scnprintf(buf+ret, PAGE_SIZE-ret, "[%u] ", i); else - ret += snprintf(buf+ret, PAGE_SIZE-ret, "%u ", i); + ret += scnprintf(buf+ret, PAGE_SIZE-ret, "%u ", i); if (ret > 0) buf[min(ret, (size_t)PAGE_SIZE)-1] = '\n'; return ret; diff --git a/drivers/hid/hid-quirks.c b/drivers/hid/hid-quirks.c index 0e7b2d998395..3735546bb524 100644 --- a/drivers/hid/hid-quirks.c +++ b/drivers/hid/hid-quirks.c @@ -103,6 +103,7 @@ static const struct hid_device_id hid_quirks[] = { { HID_USB_DEVICE(USB_VENDOR_ID_KYE, USB_DEVICE_ID_KYE_PENSKETCH_M912), HID_QUIRK_MULTI_INPUT }, { HID_USB_DEVICE(USB_VENDOR_ID_KYE, USB_DEVICE_ID_KYE_EASYPEN_M406XE), HID_QUIRK_MULTI_INPUT }, { HID_USB_DEVICE(USB_VENDOR_ID_KYE, USB_DEVICE_ID_PIXART_USB_OPTICAL_MOUSE_ID2), HID_QUIRK_ALWAYS_POLL }, + { HID_USB_DEVICE(USB_VENDOR_ID_LENOVO, USB_DEVICE_ID_LENOVO_PIXART_USB_MOUSE_608D), HID_QUIRK_ALWAYS_POLL }, { HID_USB_DEVICE(USB_VENDOR_ID_LOGITECH, USB_DEVICE_ID_LOGITECH_C007), HID_QUIRK_ALWAYS_POLL }, { HID_USB_DEVICE(USB_VENDOR_ID_LOGITECH, USB_DEVICE_ID_LOGITECH_C077), HID_QUIRK_ALWAYS_POLL }, { HID_USB_DEVICE(USB_VENDOR_ID_LOGITECH, USB_DEVICE_ID_LOGITECH_KEYBOARD_G710_PLUS), HID_QUIRK_NOGET }, diff --git a/drivers/hid/hid-sensor-custom.c b/drivers/hid/hid-sensor-custom.c index fb827c295842..4d25577a8573 100644 --- a/drivers/hid/hid-sensor-custom.c +++ b/drivers/hid/hid-sensor-custom.c @@ -313,7 +313,7 @@ static ssize_t show_value(struct device *dev, struct device_attribute *attr, while (i < ret) { if (i + attribute->size > ret) { - len += snprintf(&buf[len], + len += scnprintf(&buf[len], PAGE_SIZE - len, "%d ", values[i]); break; @@ -336,10 +336,10 @@ static ssize_t show_value(struct device *dev, struct device_attribute *attr, ++i; break; } - len += snprintf(&buf[len], PAGE_SIZE - len, + len += scnprintf(&buf[len], PAGE_SIZE - len, "%lld ", value); } - len += snprintf(&buf[len], PAGE_SIZE - len, "\n"); + len += scnprintf(&buf[len], PAGE_SIZE - len, "\n"); return len; } else if (input) diff --git a/drivers/hid/i2c-hid/i2c-hid-dmi-quirks.c b/drivers/hid/i2c-hid/i2c-hid-dmi-quirks.c index d31ea82b84c1..a66f08041a1a 100644 --- a/drivers/hid/i2c-hid/i2c-hid-dmi-quirks.c +++ b/drivers/hid/i2c-hid/i2c-hid-dmi-quirks.c @@ -342,6 +342,14 @@ static const struct dmi_system_id i2c_hid_dmi_desc_override_table[] = { .driver_data = (void *)&sipodev_desc }, { + .ident = "Trekstor SURFBOOK E11B", + .matches = { + DMI_EXACT_MATCH(DMI_SYS_VENDOR, "TREKSTOR"), + DMI_EXACT_MATCH(DMI_PRODUCT_NAME, "SURFBOOK E11B"), + }, + .driver_data = (void *)&sipodev_desc + }, + { .ident = "Direkt-Tek DTLAPY116-2", .matches = { DMI_EXACT_MATCH(DMI_SYS_VENDOR, "Direkt-Tek"), diff --git a/drivers/hid/usbhid/hiddev.c b/drivers/hid/usbhid/hiddev.c index a970b809d778..4140dea693e9 100644 --- a/drivers/hid/usbhid/hiddev.c +++ b/drivers/hid/usbhid/hiddev.c @@ -932,9 +932,9 @@ void hiddev_disconnect(struct hid_device *hid) hiddev->exist = 0; if (hiddev->open) { - mutex_unlock(&hiddev->existancelock); hid_hw_close(hiddev->hid); wake_up_interruptible(&hiddev->wait); + mutex_unlock(&hiddev->existancelock); } else { mutex_unlock(&hiddev->existancelock); kfree(hiddev); diff --git a/drivers/hwmon/acpi_power_meter.c b/drivers/hwmon/acpi_power_meter.c index 4cf25458f0b9..0db8ef4fd6e1 100644 --- a/drivers/hwmon/acpi_power_meter.c +++ b/drivers/hwmon/acpi_power_meter.c @@ -355,7 +355,9 @@ static ssize_t show_str(struct device *dev, struct acpi_device *acpi_dev = to_acpi_device(dev); struct acpi_power_meter_resource *resource = acpi_dev->driver_data; acpi_string val; + int ret; + mutex_lock(&resource->lock); switch (attr->index) { case 0: val = resource->model_number; @@ -372,8 +374,9 @@ static ssize_t show_str(struct device *dev, val = ""; break; } - - return sprintf(buf, "%s\n", val); + ret = sprintf(buf, "%s\n", val); + mutex_unlock(&resource->lock); + return ret; } static ssize_t show_val(struct device *dev, @@ -817,11 +820,12 @@ static void acpi_power_meter_notify(struct acpi_device *device, u32 event) resource = acpi_driver_data(device); - mutex_lock(&resource->lock); switch (event) { case METER_NOTIFY_CONFIG: + mutex_lock(&resource->lock); free_capabilities(resource); res = read_capabilities(resource); + mutex_unlock(&resource->lock); if (res) break; @@ -830,15 +834,12 @@ static void acpi_power_meter_notify(struct acpi_device *device, u32 event) break; case METER_NOTIFY_TRIP: sysfs_notify(&device->dev.kobj, NULL, POWER_AVERAGE_NAME); - update_meter(resource); break; case METER_NOTIFY_CAP: sysfs_notify(&device->dev.kobj, NULL, POWER_CAP_NAME); - update_cap(resource); break; case METER_NOTIFY_INTERVAL: sysfs_notify(&device->dev.kobj, NULL, POWER_AVG_INTERVAL_NAME); - update_avg_interval(resource); break; case METER_NOTIFY_CAPPING: sysfs_notify(&device->dev.kobj, NULL, POWER_ALARM_NAME); @@ -848,7 +849,6 @@ static void acpi_power_meter_notify(struct acpi_device *device, u32 event) WARN(1, "Unexpected event %d\n", event); break; } - mutex_unlock(&resource->lock); acpi_bus_generate_netlink_event(ACPI_POWER_METER_CLASS, dev_name(&device->dev), event, 0); @@ -912,8 +912,8 @@ static int acpi_power_meter_remove(struct acpi_device *device) resource = acpi_driver_data(device); hwmon_device_unregister(resource->hwmon_dev); - free_capabilities(resource); remove_attrs(resource); + free_capabilities(resource); kfree(resource); return 0; diff --git a/drivers/hwmon/adt7462.c b/drivers/hwmon/adt7462.c index 9632e2e3c4bb..319a0519ebdb 100644 --- a/drivers/hwmon/adt7462.c +++ b/drivers/hwmon/adt7462.c @@ -413,7 +413,7 @@ static int ADT7462_REG_VOLT(struct adt7462_data *data, int which) return 0x95; break; } - return -ENODEV; + return 0; } /* Provide labels for sysfs */ diff --git a/drivers/hwmon/pmbus/xdpe12284.c b/drivers/hwmon/pmbus/xdpe12284.c index ecd9b65627ec..660556b89e9f 100644 --- a/drivers/hwmon/pmbus/xdpe12284.c +++ b/drivers/hwmon/pmbus/xdpe12284.c @@ -18,6 +18,59 @@ #define XDPE122_AMD_625MV 0x10 /* AMD mode 6.25mV */ #define XDPE122_PAGE_NUM 2 +static int xdpe122_read_word_data(struct i2c_client *client, int page, int reg) +{ + const struct pmbus_driver_info *info = pmbus_get_driver_info(client); + long val; + s16 exponent; + s32 mantissa; + int ret; + + switch (reg) { + case PMBUS_VOUT_OV_FAULT_LIMIT: + case PMBUS_VOUT_UV_FAULT_LIMIT: + ret = pmbus_read_word_data(client, page, reg); + if (ret < 0) + return ret; + + /* Convert register value to LINEAR11 data. */ + exponent = ((s16)ret) >> 11; + mantissa = ((s16)((ret & GENMASK(10, 0)) << 5)) >> 5; + val = mantissa * 1000L; + if (exponent >= 0) + val <<= exponent; + else + val >>= -exponent; + + /* Convert data to VID register. */ + switch (info->vrm_version[page]) { + case vr13: + if (val >= 500) + return 1 + DIV_ROUND_CLOSEST(val - 500, 10); + return 0; + case vr12: + if (val >= 250) + return 1 + DIV_ROUND_CLOSEST(val - 250, 5); + return 0; + case imvp9: + if (val >= 200) + return 1 + DIV_ROUND_CLOSEST(val - 200, 10); + return 0; + case amd625mv: + if (val >= 200 && val <= 1550) + return DIV_ROUND_CLOSEST((1550 - val) * 100, + 625); + return 0; + default: + return -EINVAL; + } + default: + return -ENODATA; + } + + return 0; +} + static int xdpe122_identify(struct i2c_client *client, struct pmbus_driver_info *info) { @@ -70,6 +123,7 @@ static struct pmbus_driver_info xdpe122_info = { PMBUS_HAVE_TEMP | PMBUS_HAVE_STATUS_TEMP | PMBUS_HAVE_POUT | PMBUS_HAVE_PIN | PMBUS_HAVE_STATUS_INPUT, .identify = xdpe122_identify, + .read_word_data = xdpe122_read_word_data, }; static int xdpe122_probe(struct i2c_client *client, diff --git a/drivers/hwmon/w83627ehf.c b/drivers/hwmon/w83627ehf.c index 7ffadc2da57b..5a5120121e50 100644 --- a/drivers/hwmon/w83627ehf.c +++ b/drivers/hwmon/w83627ehf.c @@ -1346,8 +1346,13 @@ w83627ehf_is_visible(const void *drvdata, enum hwmon_sensor_types type, /* channel 0.., name 1.. */ if (!(data->have_temp & (1 << channel))) return 0; - if (attr == hwmon_temp_input || attr == hwmon_temp_label) + if (attr == hwmon_temp_input) return 0444; + if (attr == hwmon_temp_label) { + if (data->temp_label) + return 0444; + return 0; + } if (channel == 2 && data->temp3_val_only) return 0; if (attr == hwmon_temp_max) { diff --git a/drivers/hwtracing/intel_th/msu.c b/drivers/hwtracing/intel_th/msu.c index 8e48c7458aa3..255f8f41c8ff 100644 --- a/drivers/hwtracing/intel_th/msu.c +++ b/drivers/hwtracing/intel_th/msu.c @@ -718,9 +718,6 @@ static int msc_win_set_lockout(struct msc_window *win, if (old != expect) { ret = -EINVAL; - dev_warn_ratelimited(msc_dev(win->msc), - "expected lockout state %d, got %d\n", - expect, old); goto unlock; } @@ -741,6 +738,10 @@ unlock: /* from intel_th_msc_window_unlock(), don't warn if not locked */ if (expect == WIN_LOCKED && old == new) return 0; + + dev_warn_ratelimited(msc_dev(win->msc), + "expected lockout state %d, got %d\n", + expect, old); } return ret; @@ -760,7 +761,7 @@ static int msc_configure(struct msc *msc) lockdep_assert_held(&msc->buf_mutex); if (msc->mode > MSC_MODE_MULTI) - return -ENOTSUPP; + return -EINVAL; if (msc->mode == MSC_MODE_MULTI) { if (msc_win_set_lockout(msc->cur_win, WIN_READY, WIN_INUSE)) @@ -1294,7 +1295,7 @@ static int msc_buffer_alloc(struct msc *msc, unsigned long *nr_pages, } else if (msc->mode == MSC_MODE_MULTI) { ret = msc_buffer_multi_alloc(msc, nr_pages, nr_wins); } else { - ret = -ENOTSUPP; + ret = -EINVAL; } if (!ret) { @@ -1530,7 +1531,7 @@ static ssize_t intel_th_msc_read(struct file *file, char __user *buf, if (ret >= 0) *ppos = iter->offset; } else { - ret = -ENOTSUPP; + ret = -EINVAL; } put_count: diff --git a/drivers/hwtracing/intel_th/pci.c b/drivers/hwtracing/intel_th/pci.c index e9d90b53bbc4..86aa6a46bcba 100644 --- a/drivers/hwtracing/intel_th/pci.c +++ b/drivers/hwtracing/intel_th/pci.c @@ -235,6 +235,11 @@ static const struct pci_device_id intel_th_pci_id_table[] = { .driver_data = (kernel_ulong_t)&intel_th_2x, }, { + /* Elkhart Lake CPU */ + PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x4529), + .driver_data = (kernel_ulong_t)&intel_th_2x, + }, + { /* Elkhart Lake */ PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x4b26), .driver_data = (kernel_ulong_t)&intel_th_2x, diff --git a/drivers/hwtracing/stm/p_sys-t.c b/drivers/hwtracing/stm/p_sys-t.c index b178a5495b67..360b5c03df95 100644 --- a/drivers/hwtracing/stm/p_sys-t.c +++ b/drivers/hwtracing/stm/p_sys-t.c @@ -238,7 +238,7 @@ static struct configfs_attribute *sys_t_policy_attrs[] = { static inline bool sys_t_need_ts(struct sys_t_output *op) { if (op->node.ts_interval && - time_after(op->ts_jiffies + op->node.ts_interval, jiffies)) { + time_after(jiffies, op->ts_jiffies + op->node.ts_interval)) { op->ts_jiffies = jiffies; return true; @@ -250,8 +250,8 @@ static inline bool sys_t_need_ts(struct sys_t_output *op) static bool sys_t_need_clock_sync(struct sys_t_output *op) { if (op->node.clocksync_interval && - time_after(op->clocksync_jiffies + op->node.clocksync_interval, - jiffies)) { + time_after(jiffies, + op->clocksync_jiffies + op->node.clocksync_interval)) { op->clocksync_jiffies = jiffies; return true; diff --git a/drivers/i2c/busses/i2c-altera.c b/drivers/i2c/busses/i2c-altera.c index 5255d3755411..1de23b4f3809 100644 --- a/drivers/i2c/busses/i2c-altera.c +++ b/drivers/i2c/busses/i2c-altera.c @@ -171,7 +171,7 @@ static void altr_i2c_init(struct altr_i2c_dev *idev) /* SCL Low Time */ writel(t_low, idev->base + ALTR_I2C_SCL_LOW); /* SDA Hold Time, 300ns */ - writel(div_u64(300 * clk_mhz, 1000), idev->base + ALTR_I2C_SDA_HOLD); + writel(3 * clk_mhz / 10, idev->base + ALTR_I2C_SDA_HOLD); /* Mask all master interrupt bits */ altr_i2c_int_enable(idev, ALTR_I2C_ALL_IRQ, false); diff --git a/drivers/i2c/busses/i2c-designware-pcidrv.c b/drivers/i2c/busses/i2c-designware-pcidrv.c index 050adda7c1bd..05b35ac33ce3 100644 --- a/drivers/i2c/busses/i2c-designware-pcidrv.c +++ b/drivers/i2c/busses/i2c-designware-pcidrv.c @@ -313,6 +313,7 @@ static void i2c_dw_pci_remove(struct pci_dev *pdev) pm_runtime_get_noresume(&pdev->dev); i2c_del_adapter(&dev->adapter); + devm_free_irq(&pdev->dev, dev->irq, dev); pci_free_irq_vectors(pdev); } diff --git a/drivers/i2c/busses/i2c-gpio.c b/drivers/i2c/busses/i2c-gpio.c index 3a9e840a3546..a4a6825c8758 100644 --- a/drivers/i2c/busses/i2c-gpio.c +++ b/drivers/i2c/busses/i2c-gpio.c @@ -348,7 +348,7 @@ static struct gpio_desc *i2c_gpio_get_desc(struct device *dev, if (ret == -ENOENT) retdesc = ERR_PTR(-EPROBE_DEFER); - if (ret != -EPROBE_DEFER) + if (PTR_ERR(retdesc) != -EPROBE_DEFER) dev_err(dev, "error trying to get descriptor: %d\n", ret); return retdesc; diff --git a/drivers/i2c/busses/i2c-hix5hd2.c b/drivers/i2c/busses/i2c-hix5hd2.c index 8497c7a95dd4..224f830f77f9 100644 --- a/drivers/i2c/busses/i2c-hix5hd2.c +++ b/drivers/i2c/busses/i2c-hix5hd2.c @@ -477,6 +477,7 @@ static int hix5hd2_i2c_remove(struct platform_device *pdev) i2c_del_adapter(&priv->adap); pm_runtime_disable(priv->dev); pm_runtime_set_suspended(priv->dev); + clk_disable_unprepare(priv->clk); return 0; } diff --git a/drivers/i2c/busses/i2c-i801.c b/drivers/i2c/busses/i2c-i801.c index ca4f096fef74..a9c03f5c3482 100644 --- a/drivers/i2c/busses/i2c-i801.c +++ b/drivers/i2c/busses/i2c-i801.c @@ -132,11 +132,6 @@ #define TCOBASE 0x050 #define TCOCTL 0x054 -#define ACPIBASE 0x040 -#define ACPIBASE_SMI_OFF 0x030 -#define ACPICTRL 0x044 -#define ACPICTRL_EN 0x080 - #define SBREG_BAR 0x10 #define SBREG_SMBCTRL 0xc6000c #define SBREG_SMBCTRL_DNV 0xcf000c @@ -1553,7 +1548,7 @@ i801_add_tco_spt(struct i801_priv *priv, struct pci_dev *pci_dev, pci_bus_write_config_byte(pci_dev->bus, devfn, 0xe1, hidden); spin_unlock(&p2sb_spinlock); - res = &tco_res[ICH_RES_MEM_OFF]; + res = &tco_res[1]; if (pci_dev->device == PCI_DEVICE_ID_INTEL_DNV_SMBUS) res->start = (resource_size_t)base64_addr + SBREG_SMBCTRL_DNV; else @@ -1563,7 +1558,7 @@ i801_add_tco_spt(struct i801_priv *priv, struct pci_dev *pci_dev, res->flags = IORESOURCE_MEM; return platform_device_register_resndata(&pci_dev->dev, "iTCO_wdt", -1, - tco_res, 3, &spt_tco_platform_data, + tco_res, 2, &spt_tco_platform_data, sizeof(spt_tco_platform_data)); } @@ -1576,17 +1571,16 @@ static struct platform_device * i801_add_tco_cnl(struct i801_priv *priv, struct pci_dev *pci_dev, struct resource *tco_res) { - return platform_device_register_resndata(&pci_dev->dev, "iTCO_wdt", -1, - tco_res, 2, &cnl_tco_platform_data, - sizeof(cnl_tco_platform_data)); + return platform_device_register_resndata(&pci_dev->dev, + "iTCO_wdt", -1, tco_res, 1, &cnl_tco_platform_data, + sizeof(cnl_tco_platform_data)); } static void i801_add_tco(struct i801_priv *priv) { - u32 base_addr, tco_base, tco_ctl, ctrl_val; struct pci_dev *pci_dev = priv->pci_dev; - struct resource tco_res[3], *res; - unsigned int devfn; + struct resource tco_res[2], *res; + u32 tco_base, tco_ctl; /* If we have ACPI based watchdog use that instead */ if (acpi_has_watchdog()) @@ -1601,30 +1595,15 @@ static void i801_add_tco(struct i801_priv *priv) return; memset(tco_res, 0, sizeof(tco_res)); - - res = &tco_res[ICH_RES_IO_TCO]; - res->start = tco_base & ~1; - res->end = res->start + 32 - 1; - res->flags = IORESOURCE_IO; - /* - * Power Management registers. + * Always populate the main iTCO IO resource here. The second entry + * for NO_REBOOT MMIO is filled by the SPT specific function. */ - devfn = PCI_DEVFN(PCI_SLOT(pci_dev->devfn), 2); - pci_bus_read_config_dword(pci_dev->bus, devfn, ACPIBASE, &base_addr); - - res = &tco_res[ICH_RES_IO_SMI]; - res->start = (base_addr & ~1) + ACPIBASE_SMI_OFF; - res->end = res->start + 3; + res = &tco_res[0]; + res->start = tco_base & ~1; + res->end = res->start + 32 - 1; res->flags = IORESOURCE_IO; - /* - * Enable the ACPI I/O space. - */ - pci_bus_read_config_dword(pci_dev->bus, devfn, ACPICTRL, &ctrl_val); - ctrl_val |= ACPICTRL_EN; - pci_bus_write_config_dword(pci_dev->bus, devfn, ACPICTRL, ctrl_val); - if (priv->features & FEATURE_TCO_CNL) priv->tco_pdev = i801_add_tco_cnl(priv, pci_dev, tco_res); else diff --git a/drivers/i2c/busses/i2c-jz4780.c b/drivers/i2c/busses/i2c-jz4780.c index 16a67a64284a..b426fc956938 100644 --- a/drivers/i2c/busses/i2c-jz4780.c +++ b/drivers/i2c/busses/i2c-jz4780.c @@ -78,25 +78,6 @@ #define X1000_I2C_DC_STOP BIT(9) -static const char * const jz4780_i2c_abrt_src[] = { - "ABRT_7B_ADDR_NOACK", - "ABRT_10ADDR1_NOACK", - "ABRT_10ADDR2_NOACK", - "ABRT_XDATA_NOACK", - "ABRT_GCALL_NOACK", - "ABRT_GCALL_READ", - "ABRT_HS_ACKD", - "SBYTE_ACKDET", - "ABRT_HS_NORSTRT", - "SBYTE_NORSTRT", - "ABRT_10B_RD_NORSTRT", - "ABRT_MASTER_DIS", - "ARB_LOST", - "SLVFLUSH_TXFIFO", - "SLV_ARBLOST", - "SLVRD_INTX", -}; - #define JZ4780_I2C_INTST_IGC BIT(11) #define JZ4780_I2C_INTST_ISTT BIT(10) #define JZ4780_I2C_INTST_ISTP BIT(9) @@ -576,21 +557,8 @@ done: static void jz4780_i2c_txabrt(struct jz4780_i2c *i2c, int src) { - int i; - - dev_err(&i2c->adap.dev, "txabrt: 0x%08x\n", src); - dev_err(&i2c->adap.dev, "device addr=%x\n", - jz4780_i2c_readw(i2c, JZ4780_I2C_TAR)); - dev_err(&i2c->adap.dev, "send cmd count:%d %d\n", - i2c->cmd, i2c->cmd_buf[i2c->cmd]); - dev_err(&i2c->adap.dev, "receive data count:%d %d\n", - i2c->cmd, i2c->data_buf[i2c->cmd]); - - for (i = 0; i < 16; i++) { - if (src & BIT(i)) - dev_dbg(&i2c->adap.dev, "I2C TXABRT[%d]=%s\n", - i, jz4780_i2c_abrt_src[i]); - } + dev_dbg(&i2c->adap.dev, "txabrt: 0x%08x, cmd: %d, send: %d, recv: %d\n", + src, i2c->cmd, i2c->cmd_buf[i2c->cmd], i2c->data_buf[i2c->cmd]); } static inline int jz4780_i2c_xfer_read(struct jz4780_i2c *i2c, diff --git a/drivers/i2c/busses/i2c-nvidia-gpu.c b/drivers/i2c/busses/i2c-nvidia-gpu.c index 62e18b4db0ed..f5d25ce00f03 100644 --- a/drivers/i2c/busses/i2c-nvidia-gpu.c +++ b/drivers/i2c/busses/i2c-nvidia-gpu.c @@ -8,6 +8,7 @@ #include <linux/delay.h> #include <linux/i2c.h> #include <linux/interrupt.h> +#include <linux/iopoll.h> #include <linux/module.h> #include <linux/pci.h> #include <linux/platform_device.h> @@ -75,20 +76,15 @@ static void gpu_enable_i2c_bus(struct gpu_i2c_dev *i2cd) static int gpu_i2c_check_status(struct gpu_i2c_dev *i2cd) { - unsigned long target = jiffies + msecs_to_jiffies(1000); u32 val; + int ret; - do { - val = readl(i2cd->regs + I2C_MST_CNTL); - if (!(val & I2C_MST_CNTL_CYCLE_TRIGGER)) - break; - if ((val & I2C_MST_CNTL_STATUS) != - I2C_MST_CNTL_STATUS_BUS_BUSY) - break; - usleep_range(500, 600); - } while (time_is_after_jiffies(target)); - - if (time_is_before_jiffies(target)) { + ret = readl_poll_timeout(i2cd->regs + I2C_MST_CNTL, val, + !(val & I2C_MST_CNTL_CYCLE_TRIGGER) || + (val & I2C_MST_CNTL_STATUS) != I2C_MST_CNTL_STATUS_BUS_BUSY, + 500, 1000 * USEC_PER_MSEC); + + if (ret) { dev_err(i2cd->dev, "i2c timeout error %x\n", val); return -ETIMEDOUT; } diff --git a/drivers/i2c/busses/i2c-pca-platform.c b/drivers/i2c/busses/i2c-pca-platform.c index a7a81846d5b1..635dd697ac0b 100644 --- a/drivers/i2c/busses/i2c-pca-platform.c +++ b/drivers/i2c/busses/i2c-pca-platform.c @@ -140,7 +140,7 @@ static int i2c_pca_pf_probe(struct platform_device *pdev) int ret = 0; int irq; - irq = platform_get_irq(pdev, 0); + irq = platform_get_irq_optional(pdev, 0); /* If irq is 0, we do polling. */ if (irq < 0) irq = 0; diff --git a/drivers/i2c/busses/i2c-st.c b/drivers/i2c/busses/i2c-st.c index 54e1fc8a495e..f7f7b5b64720 100644 --- a/drivers/i2c/busses/i2c-st.c +++ b/drivers/i2c/busses/i2c-st.c @@ -434,6 +434,7 @@ static void st_i2c_wr_fill_tx_fifo(struct st_i2c_dev *i2c_dev) /** * st_i2c_rd_fill_tx_fifo() - Fill the Tx FIFO in read mode * @i2c_dev: Controller's private data + * @max: Maximum amount of data to fill into the Tx FIFO * * This functions fills the Tx FIFO with fixed pattern when * in read mode to trigger clock. diff --git a/drivers/i2c/i2c-core-acpi.c b/drivers/i2c/i2c-core-acpi.c index 8f3dbc97a057..8b0ff780919b 100644 --- a/drivers/i2c/i2c-core-acpi.c +++ b/drivers/i2c/i2c-core-acpi.c @@ -394,9 +394,17 @@ EXPORT_SYMBOL_GPL(i2c_acpi_find_adapter_by_handle); static struct i2c_client *i2c_acpi_find_client_by_adev(struct acpi_device *adev) { struct device *dev; + struct i2c_client *client; dev = bus_find_device_by_acpi_dev(&i2c_bus_type, adev); - return dev ? i2c_verify_client(dev) : NULL; + if (!dev) + return NULL; + + client = i2c_verify_client(dev); + if (!client) + put_device(dev); + + return client; } static int i2c_acpi_notify(struct notifier_block *nb, unsigned long value, diff --git a/drivers/ide/ide-gd.c b/drivers/ide/ide-gd.c index 1bb99b556393..05c26986637b 100644 --- a/drivers/ide/ide-gd.c +++ b/drivers/ide/ide-gd.c @@ -361,7 +361,7 @@ static const struct block_device_operations ide_gd_ops = { .release = ide_gd_release, .ioctl = ide_gd_ioctl, #ifdef CONFIG_COMPAT - .ioctl = ide_gd_compat_ioctl, + .compat_ioctl = ide_gd_compat_ioctl, #endif .getgeo = ide_gd_getgeo, .check_events = ide_gd_check_events, diff --git a/drivers/iio/accel/adxl372.c b/drivers/iio/accel/adxl372.c index 67b8817995c0..60daf04ce188 100644 --- a/drivers/iio/accel/adxl372.c +++ b/drivers/iio/accel/adxl372.c @@ -237,6 +237,7 @@ static const struct adxl372_axis_lookup adxl372_axis_lookup_table[] = { .realbits = 12, \ .storagebits = 16, \ .shift = 4, \ + .endianness = IIO_BE, \ }, \ } diff --git a/drivers/iio/accel/st_accel_i2c.c b/drivers/iio/accel/st_accel_i2c.c index 633955d764cc..849cf74153c4 100644 --- a/drivers/iio/accel/st_accel_i2c.c +++ b/drivers/iio/accel/st_accel_i2c.c @@ -110,7 +110,7 @@ MODULE_DEVICE_TABLE(of, st_accel_of_match); #ifdef CONFIG_ACPI static const struct acpi_device_id st_accel_acpi_match[] = { - {"SMO8840", (kernel_ulong_t)LNG2DM_ACCEL_DEV_NAME}, + {"SMO8840", (kernel_ulong_t)LIS2DH12_ACCEL_DEV_NAME}, {"SMO8A90", (kernel_ulong_t)LNG2DM_ACCEL_DEV_NAME}, { }, }; diff --git a/drivers/iio/adc/at91-sama5d2_adc.c b/drivers/iio/adc/at91-sama5d2_adc.c index a5c7771227d5..9d96f7d08b95 100644 --- a/drivers/iio/adc/at91-sama5d2_adc.c +++ b/drivers/iio/adc/at91-sama5d2_adc.c @@ -723,6 +723,7 @@ static int at91_adc_configure_trigger(struct iio_trigger *trig, bool state) for_each_set_bit(bit, indio->active_scan_mask, indio->num_channels) { struct iio_chan_spec const *chan = at91_adc_chan_get(indio, bit); + u32 cor; if (!chan) continue; @@ -732,6 +733,20 @@ static int at91_adc_configure_trigger(struct iio_trigger *trig, bool state) continue; if (state) { + cor = at91_adc_readl(st, AT91_SAMA5D2_COR); + + if (chan->differential) + cor |= (BIT(chan->channel) | + BIT(chan->channel2)) << + AT91_SAMA5D2_COR_DIFF_OFFSET; + else + cor &= ~(BIT(chan->channel) << + AT91_SAMA5D2_COR_DIFF_OFFSET); + + at91_adc_writel(st, AT91_SAMA5D2_COR, cor); + } + + if (state) { at91_adc_writel(st, AT91_SAMA5D2_CHER, BIT(chan->channel)); /* enable irq only if not using DMA */ diff --git a/drivers/iio/adc/stm32-dfsdm-adc.c b/drivers/iio/adc/stm32-dfsdm-adc.c index 2aad2cda6943..76a60d93fe23 100644 --- a/drivers/iio/adc/stm32-dfsdm-adc.c +++ b/drivers/iio/adc/stm32-dfsdm-adc.c @@ -842,31 +842,6 @@ static inline void stm32_dfsdm_process_data(struct stm32_dfsdm_adc *adc, } } -static irqreturn_t stm32_dfsdm_adc_trigger_handler(int irq, void *p) -{ - struct iio_poll_func *pf = p; - struct iio_dev *indio_dev = pf->indio_dev; - struct stm32_dfsdm_adc *adc = iio_priv(indio_dev); - int available = stm32_dfsdm_adc_dma_residue(adc); - - while (available >= indio_dev->scan_bytes) { - s32 *buffer = (s32 *)&adc->rx_buf[adc->bufi]; - - stm32_dfsdm_process_data(adc, buffer); - - iio_push_to_buffers_with_timestamp(indio_dev, buffer, - pf->timestamp); - available -= indio_dev->scan_bytes; - adc->bufi += indio_dev->scan_bytes; - if (adc->bufi >= adc->buf_sz) - adc->bufi = 0; - } - - iio_trigger_notify_done(indio_dev->trig); - - return IRQ_HANDLED; -} - static void stm32_dfsdm_dma_buffer_done(void *data) { struct iio_dev *indio_dev = data; @@ -874,11 +849,6 @@ static void stm32_dfsdm_dma_buffer_done(void *data) int available = stm32_dfsdm_adc_dma_residue(adc); size_t old_pos; - if (indio_dev->currentmode & INDIO_BUFFER_TRIGGERED) { - iio_trigger_poll_chained(indio_dev->trig); - return; - } - /* * FIXME: In Kernel interface does not support cyclic DMA buffer,and * offers only an interface to push data samples per samples. @@ -906,7 +876,15 @@ static void stm32_dfsdm_dma_buffer_done(void *data) adc->bufi = 0; old_pos = 0; } - /* regular iio buffer without trigger */ + /* + * In DMA mode the trigger services of IIO are not used + * (e.g. no call to iio_trigger_poll). + * Calling irq handler associated to the hardware trigger is not + * relevant as the conversions have already been done. Data + * transfers are performed directly in DMA callback instead. + * This implementation avoids to call trigger irq handler that + * may sleep, in an atomic context (DMA irq handler context). + */ if (adc->dev_data->type == DFSDM_IIO) iio_push_to_buffers(indio_dev, buffer); } @@ -1536,8 +1514,7 @@ static int stm32_dfsdm_adc_init(struct iio_dev *indio_dev) } ret = iio_triggered_buffer_setup(indio_dev, - &iio_pollfunc_store_time, - &stm32_dfsdm_adc_trigger_handler, + &iio_pollfunc_store_time, NULL, &stm32_dfsdm_buffer_setup_ops); if (ret) { stm32_dfsdm_dma_release(indio_dev); diff --git a/drivers/iio/chemical/Kconfig b/drivers/iio/chemical/Kconfig index 0b91de4df8f4..a7e65a59bf42 100644 --- a/drivers/iio/chemical/Kconfig +++ b/drivers/iio/chemical/Kconfig @@ -91,6 +91,8 @@ config SPS30 tristate "SPS30 particulate matter sensor" depends on I2C select CRC8 + select IIO_BUFFER + select IIO_TRIGGERED_BUFFER help Say Y here to build support for the Sensirion SPS30 particulate matter sensor. diff --git a/drivers/iio/light/vcnl4000.c b/drivers/iio/light/vcnl4000.c index b0e241aaefb4..e5b00a6611ac 100644 --- a/drivers/iio/light/vcnl4000.c +++ b/drivers/iio/light/vcnl4000.c @@ -167,16 +167,17 @@ static int vcnl4200_init(struct vcnl4000_data *data) data->vcnl4200_ps.reg = VCNL4200_PS_DATA; switch (id) { case VCNL4200_PROD_ID: - /* Integration time is 50ms, but the experiments */ - /* show 54ms in total. */ - data->vcnl4200_al.sampling_rate = ktime_set(0, 54000 * 1000); - data->vcnl4200_ps.sampling_rate = ktime_set(0, 4200 * 1000); + /* Default wait time is 50ms, add 20% tolerance. */ + data->vcnl4200_al.sampling_rate = ktime_set(0, 60000 * 1000); + /* Default wait time is 4.8ms, add 20% tolerance. */ + data->vcnl4200_ps.sampling_rate = ktime_set(0, 5760 * 1000); data->al_scale = 24000; break; case VCNL4040_PROD_ID: - /* Integration time is 80ms, add 10ms. */ - data->vcnl4200_al.sampling_rate = ktime_set(0, 100000 * 1000); - data->vcnl4200_ps.sampling_rate = ktime_set(0, 100000 * 1000); + /* Default wait time is 80ms, add 20% tolerance. */ + data->vcnl4200_al.sampling_rate = ktime_set(0, 96000 * 1000); + /* Default wait time is 5ms, add 20% tolerance. */ + data->vcnl4200_ps.sampling_rate = ktime_set(0, 6000 * 1000); data->al_scale = 120000; break; } diff --git a/drivers/iio/magnetometer/ak8974.c b/drivers/iio/magnetometer/ak8974.c index fc7e910f8e8b..d32996702110 100644 --- a/drivers/iio/magnetometer/ak8974.c +++ b/drivers/iio/magnetometer/ak8974.c @@ -564,7 +564,7 @@ static int ak8974_read_raw(struct iio_dev *indio_dev, * We read all axes and discard all but one, for optimized * reading, use the triggered buffer. */ - *val = le16_to_cpu(hw_values[chan->address]); + *val = (s16)le16_to_cpu(hw_values[chan->address]); ret = IIO_VAL_INT; } diff --git a/drivers/iio/proximity/ping.c b/drivers/iio/proximity/ping.c index 34aff108dff5..12b893c5b0ee 100644 --- a/drivers/iio/proximity/ping.c +++ b/drivers/iio/proximity/ping.c @@ -269,7 +269,7 @@ static const struct iio_chan_spec ping_chan_spec[] = { static const struct of_device_id of_ping_match[] = { { .compatible = "parallax,ping", .data = &pa_ping_cfg}, - { .compatible = "parallax,laserping", .data = &pa_ping_cfg}, + { .compatible = "parallax,laserping", .data = &pa_laser_ping_cfg}, {}, }; diff --git a/drivers/iio/trigger/stm32-timer-trigger.c b/drivers/iio/trigger/stm32-timer-trigger.c index 2e0d32aa8436..2f82e8c32186 100644 --- a/drivers/iio/trigger/stm32-timer-trigger.c +++ b/drivers/iio/trigger/stm32-timer-trigger.c @@ -161,7 +161,8 @@ static int stm32_timer_start(struct stm32_timer_trigger *priv, return 0; } -static void stm32_timer_stop(struct stm32_timer_trigger *priv) +static void stm32_timer_stop(struct stm32_timer_trigger *priv, + struct iio_trigger *trig) { u32 ccer, cr1; @@ -179,6 +180,12 @@ static void stm32_timer_stop(struct stm32_timer_trigger *priv) regmap_write(priv->regmap, TIM_PSC, 0); regmap_write(priv->regmap, TIM_ARR, 0); + /* Force disable master mode */ + if (stm32_timer_is_trgo2_name(trig->name)) + regmap_update_bits(priv->regmap, TIM_CR2, TIM_CR2_MMS2, 0); + else + regmap_update_bits(priv->regmap, TIM_CR2, TIM_CR2_MMS, 0); + /* Make sure that registers are updated */ regmap_update_bits(priv->regmap, TIM_EGR, TIM_EGR_UG, TIM_EGR_UG); } @@ -197,7 +204,7 @@ static ssize_t stm32_tt_store_frequency(struct device *dev, return ret; if (freq == 0) { - stm32_timer_stop(priv); + stm32_timer_stop(priv, trig); } else { ret = stm32_timer_start(priv, trig, freq); if (ret) diff --git a/drivers/infiniband/core/cm.c b/drivers/infiniband/core/cm.c index 68cc1b2d6824..15e99a888427 100644 --- a/drivers/infiniband/core/cm.c +++ b/drivers/infiniband/core/cm.c @@ -1191,6 +1191,7 @@ struct ib_cm_id *ib_cm_insert_listen(struct ib_device *device, /* Sharing an ib_cm_id with different handlers is not * supported */ spin_unlock_irqrestore(&cm.lock, flags); + ib_destroy_cm_id(cm_id); return ERR_PTR(-EINVAL); } refcount_inc(&cm_id_priv->refcount); diff --git a/drivers/infiniband/core/cma.c b/drivers/infiniband/core/cma.c index 72f032160c4b..2dec3a02ab9f 100644 --- a/drivers/infiniband/core/cma.c +++ b/drivers/infiniband/core/cma.c @@ -3212,19 +3212,26 @@ int rdma_resolve_addr(struct rdma_cm_id *id, struct sockaddr *src_addr, int ret; id_priv = container_of(id, struct rdma_id_private, id); + memcpy(cma_dst_addr(id_priv), dst_addr, rdma_addr_size(dst_addr)); if (id_priv->state == RDMA_CM_IDLE) { ret = cma_bind_addr(id, src_addr, dst_addr); - if (ret) + if (ret) { + memset(cma_dst_addr(id_priv), 0, + rdma_addr_size(dst_addr)); return ret; + } } - if (cma_family(id_priv) != dst_addr->sa_family) + if (cma_family(id_priv) != dst_addr->sa_family) { + memset(cma_dst_addr(id_priv), 0, rdma_addr_size(dst_addr)); return -EINVAL; + } - if (!cma_comp_exch(id_priv, RDMA_CM_ADDR_BOUND, RDMA_CM_ADDR_QUERY)) + if (!cma_comp_exch(id_priv, RDMA_CM_ADDR_BOUND, RDMA_CM_ADDR_QUERY)) { + memset(cma_dst_addr(id_priv), 0, rdma_addr_size(dst_addr)); return -EINVAL; + } - memcpy(cma_dst_addr(id_priv), dst_addr, rdma_addr_size(dst_addr)); if (cma_any_addr(dst_addr)) { ret = cma_resolve_loopback(id_priv); } else { diff --git a/drivers/infiniband/core/core_priv.h b/drivers/infiniband/core/core_priv.h index b1457b3464d3..cf42acca4a3a 100644 --- a/drivers/infiniband/core/core_priv.h +++ b/drivers/infiniband/core/core_priv.h @@ -338,6 +338,20 @@ static inline struct ib_qp *_ib_create_qp(struct ib_device *dev, qp->pd = pd; qp->uobject = uobj; qp->real_qp = qp; + + qp->qp_type = attr->qp_type; + qp->rwq_ind_tbl = attr->rwq_ind_tbl; + qp->send_cq = attr->send_cq; + qp->recv_cq = attr->recv_cq; + qp->srq = attr->srq; + qp->rwq_ind_tbl = attr->rwq_ind_tbl; + qp->event_handler = attr->event_handler; + + atomic_set(&qp->usecnt, 0); + spin_lock_init(&qp->mr_lock); + INIT_LIST_HEAD(&qp->rdma_mrs); + INIT_LIST_HEAD(&qp->sig_mrs); + /* * We don't track XRC QPs for now, because they don't have PD * and more importantly they are created internaly by driver, diff --git a/drivers/infiniband/core/device.c b/drivers/infiniband/core/device.c index f6c255202d7f..d0b3d35ad3e4 100644 --- a/drivers/infiniband/core/device.c +++ b/drivers/infiniband/core/device.c @@ -896,7 +896,9 @@ static int add_one_compat_dev(struct ib_device *device, cdev->dev.parent = device->dev.parent; rdma_init_coredev(cdev, device, read_pnet(&rnet->net)); cdev->dev.release = compatdev_release; - dev_set_name(&cdev->dev, "%s", dev_name(&device->dev)); + ret = dev_set_name(&cdev->dev, "%s", dev_name(&device->dev)); + if (ret) + goto add_err; ret = device_add(&cdev->dev); if (ret) diff --git a/drivers/infiniband/core/iwcm.c b/drivers/infiniband/core/iwcm.c index ade71823370f..da8adadf4755 100644 --- a/drivers/infiniband/core/iwcm.c +++ b/drivers/infiniband/core/iwcm.c @@ -159,8 +159,10 @@ static void dealloc_work_entries(struct iwcm_id_private *cm_id_priv) { struct list_head *e, *tmp; - list_for_each_safe(e, tmp, &cm_id_priv->work_free_list) + list_for_each_safe(e, tmp, &cm_id_priv->work_free_list) { + list_del(e); kfree(list_entry(e, struct iwcm_work, free_list)); + } } static int alloc_work_entries(struct iwcm_id_private *cm_id_priv, int count) diff --git a/drivers/infiniband/core/nldev.c b/drivers/infiniband/core/nldev.c index 37b433aa7306..9eec26d10d7b 100644 --- a/drivers/infiniband/core/nldev.c +++ b/drivers/infiniband/core/nldev.c @@ -918,6 +918,10 @@ static int nldev_set_doit(struct sk_buff *skb, struct nlmsghdr *nlh, nla_strlcpy(name, tb[RDMA_NLDEV_ATTR_DEV_NAME], IB_DEVICE_NAME_MAX); + if (strlen(name) == 0) { + err = -EINVAL; + goto done; + } err = ib_device_rename(device, name); goto done; } @@ -1514,7 +1518,7 @@ static int nldev_newlink(struct sk_buff *skb, struct nlmsghdr *nlh, nla_strlcpy(ibdev_name, tb[RDMA_NLDEV_ATTR_DEV_NAME], sizeof(ibdev_name)); - if (strchr(ibdev_name, '%')) + if (strchr(ibdev_name, '%') || strlen(ibdev_name) == 0) return -EINVAL; nla_strlcpy(type, tb[RDMA_NLDEV_ATTR_LINK_TYPE], sizeof(type)); @@ -1757,6 +1761,8 @@ static int nldev_stat_set_doit(struct sk_buff *skb, struct nlmsghdr *nlh, if (ret) goto err_msg; } else { + if (!tb[RDMA_NLDEV_ATTR_RES_LQPN]) + goto err_msg; qpn = nla_get_u32(tb[RDMA_NLDEV_ATTR_RES_LQPN]); if (tb[RDMA_NLDEV_ATTR_STAT_COUNTER_ID]) { cntn = nla_get_u32(tb[RDMA_NLDEV_ATTR_STAT_COUNTER_ID]); diff --git a/drivers/infiniband/core/rw.c b/drivers/infiniband/core/rw.c index 4fad732f9b3c..06e5b6787443 100644 --- a/drivers/infiniband/core/rw.c +++ b/drivers/infiniband/core/rw.c @@ -273,6 +273,23 @@ static int rdma_rw_init_single_wr(struct rdma_rw_ctx *ctx, struct ib_qp *qp, return 1; } +static void rdma_rw_unmap_sg(struct ib_device *dev, struct scatterlist *sg, + u32 sg_cnt, enum dma_data_direction dir) +{ + if (is_pci_p2pdma_page(sg_page(sg))) + pci_p2pdma_unmap_sg(dev->dma_device, sg, sg_cnt, dir); + else + ib_dma_unmap_sg(dev, sg, sg_cnt, dir); +} + +static int rdma_rw_map_sg(struct ib_device *dev, struct scatterlist *sg, + u32 sg_cnt, enum dma_data_direction dir) +{ + if (is_pci_p2pdma_page(sg_page(sg))) + return pci_p2pdma_map_sg(dev->dma_device, sg, sg_cnt, dir); + return ib_dma_map_sg(dev, sg, sg_cnt, dir); +} + /** * rdma_rw_ctx_init - initialize a RDMA READ/WRITE context * @ctx: context to initialize @@ -295,11 +312,7 @@ int rdma_rw_ctx_init(struct rdma_rw_ctx *ctx, struct ib_qp *qp, u8 port_num, struct ib_device *dev = qp->pd->device; int ret; - if (is_pci_p2pdma_page(sg_page(sg))) - ret = pci_p2pdma_map_sg(dev->dma_device, sg, sg_cnt, dir); - else - ret = ib_dma_map_sg(dev, sg, sg_cnt, dir); - + ret = rdma_rw_map_sg(dev, sg, sg_cnt, dir); if (!ret) return -ENOMEM; sg_cnt = ret; @@ -338,7 +351,7 @@ int rdma_rw_ctx_init(struct rdma_rw_ctx *ctx, struct ib_qp *qp, u8 port_num, return ret; out_unmap_sg: - ib_dma_unmap_sg(dev, sg, sg_cnt, dir); + rdma_rw_unmap_sg(dev, sg, sg_cnt, dir); return ret; } EXPORT_SYMBOL(rdma_rw_ctx_init); @@ -588,11 +601,7 @@ void rdma_rw_ctx_destroy(struct rdma_rw_ctx *ctx, struct ib_qp *qp, u8 port_num, break; } - if (is_pci_p2pdma_page(sg_page(sg))) - pci_p2pdma_unmap_sg(qp->pd->device->dma_device, sg, - sg_cnt, dir); - else - ib_dma_unmap_sg(qp->pd->device, sg, sg_cnt, dir); + rdma_rw_unmap_sg(qp->pd->device, sg, sg_cnt, dir); } EXPORT_SYMBOL(rdma_rw_ctx_destroy); diff --git a/drivers/infiniband/core/security.c b/drivers/infiniband/core/security.c index 2b4d80393bd0..75e7ec017836 100644 --- a/drivers/infiniband/core/security.c +++ b/drivers/infiniband/core/security.c @@ -340,20 +340,19 @@ static struct ib_ports_pkeys *get_new_pps(const struct ib_qp *qp, return NULL; if (qp_attr_mask & IB_QP_PORT) - new_pps->main.port_num = - (qp_pps) ? qp_pps->main.port_num : qp_attr->port_num; - if (qp_attr_mask & IB_QP_PKEY_INDEX) - new_pps->main.pkey_index = (qp_pps) ? qp_pps->main.pkey_index : - qp_attr->pkey_index; - if ((qp_attr_mask & IB_QP_PKEY_INDEX) && (qp_attr_mask & IB_QP_PORT)) - new_pps->main.state = IB_PORT_PKEY_VALID; - - if (!(qp_attr_mask & (IB_QP_PKEY_INDEX || IB_QP_PORT)) && qp_pps) { + new_pps->main.port_num = qp_attr->port_num; + else if (qp_pps) new_pps->main.port_num = qp_pps->main.port_num; + + if (qp_attr_mask & IB_QP_PKEY_INDEX) + new_pps->main.pkey_index = qp_attr->pkey_index; + else if (qp_pps) new_pps->main.pkey_index = qp_pps->main.pkey_index; - if (qp_pps->main.state != IB_PORT_PKEY_NOT_VALID) - new_pps->main.state = IB_PORT_PKEY_VALID; - } + + if (((qp_attr_mask & IB_QP_PKEY_INDEX) && + (qp_attr_mask & IB_QP_PORT)) || + (qp_pps && qp_pps->main.state != IB_PORT_PKEY_NOT_VALID)) + new_pps->main.state = IB_PORT_PKEY_VALID; if (qp_attr_mask & IB_QP_ALT_PATH) { new_pps->alt.port_num = qp_attr->alt_port_num; diff --git a/drivers/infiniband/core/umem_odp.c b/drivers/infiniband/core/umem_odp.c index b8c657b28380..3b1e627d9a8d 100644 --- a/drivers/infiniband/core/umem_odp.c +++ b/drivers/infiniband/core/umem_odp.c @@ -181,14 +181,28 @@ ib_umem_odp_alloc_child(struct ib_umem_odp *root, unsigned long addr, odp_data->page_shift = PAGE_SHIFT; odp_data->notifier.ops = ops; + /* + * A mmget must be held when registering a notifier, the owming_mm only + * has a mm_grab at this point. + */ + if (!mmget_not_zero(umem->owning_mm)) { + ret = -EFAULT; + goto out_free; + } + odp_data->tgid = get_pid(root->tgid); ret = ib_init_umem_odp(odp_data, ops); - if (ret) { - put_pid(odp_data->tgid); - kfree(odp_data); - return ERR_PTR(ret); - } + if (ret) + goto out_tgid; + mmput(umem->owning_mm); return odp_data; + +out_tgid: + put_pid(odp_data->tgid); + mmput(umem->owning_mm); +out_free: + kfree(odp_data); + return ERR_PTR(ret); } EXPORT_SYMBOL(ib_umem_odp_alloc_child); @@ -261,8 +275,8 @@ void ib_umem_odp_release(struct ib_umem_odp *umem_odp) mmu_interval_notifier_remove(&umem_odp->notifier); kvfree(umem_odp->dma_list); kvfree(umem_odp->page_list); - put_pid(umem_odp->tgid); } + put_pid(umem_odp->tgid); kfree(umem_odp); } EXPORT_SYMBOL(ib_umem_odp_release); diff --git a/drivers/infiniband/core/user_mad.c b/drivers/infiniband/core/user_mad.c index 1235ffb2389b..da229eab5903 100644 --- a/drivers/infiniband/core/user_mad.c +++ b/drivers/infiniband/core/user_mad.c @@ -1129,17 +1129,30 @@ static const struct file_operations umad_sm_fops = { .llseek = no_llseek, }; +static struct ib_umad_port *get_port(struct ib_device *ibdev, + struct ib_umad_device *umad_dev, + unsigned int port) +{ + if (!umad_dev) + return ERR_PTR(-EOPNOTSUPP); + if (!rdma_is_port_valid(ibdev, port)) + return ERR_PTR(-EINVAL); + if (!rdma_cap_ib_mad(ibdev, port)) + return ERR_PTR(-EOPNOTSUPP); + + return &umad_dev->ports[port - rdma_start_port(ibdev)]; +} + static int ib_umad_get_nl_info(struct ib_device *ibdev, void *client_data, struct ib_client_nl_info *res) { - struct ib_umad_device *umad_dev = client_data; + struct ib_umad_port *port = get_port(ibdev, client_data, res->port); - if (!rdma_is_port_valid(ibdev, res->port)) - return -EINVAL; + if (IS_ERR(port)) + return PTR_ERR(port); res->abi = IB_USER_MAD_ABI_VERSION; - res->cdev = &umad_dev->ports[res->port - rdma_start_port(ibdev)].dev; - + res->cdev = &port->dev; return 0; } @@ -1154,15 +1167,13 @@ MODULE_ALIAS_RDMA_CLIENT("umad"); static int ib_issm_get_nl_info(struct ib_device *ibdev, void *client_data, struct ib_client_nl_info *res) { - struct ib_umad_device *umad_dev = - ib_get_client_data(ibdev, &umad_client); + struct ib_umad_port *port = get_port(ibdev, client_data, res->port); - if (!rdma_is_port_valid(ibdev, res->port)) - return -EINVAL; + if (IS_ERR(port)) + return PTR_ERR(port); res->abi = IB_USER_MAD_ABI_VERSION; - res->cdev = &umad_dev->ports[res->port - rdma_start_port(ibdev)].sm_dev; - + res->cdev = &port->sm_dev; return 0; } diff --git a/drivers/infiniband/core/uverbs_cmd.c b/drivers/infiniband/core/uverbs_cmd.c index 025933752e1d..060b4ebbd2ba 100644 --- a/drivers/infiniband/core/uverbs_cmd.c +++ b/drivers/infiniband/core/uverbs_cmd.c @@ -1445,16 +1445,7 @@ static int create_qp(struct uverbs_attr_bundle *attrs, if (ret) goto err_cb; - qp->pd = pd; - qp->send_cq = attr.send_cq; - qp->recv_cq = attr.recv_cq; - qp->srq = attr.srq; - qp->rwq_ind_tbl = ind_tbl; - qp->event_handler = attr.event_handler; - qp->qp_type = attr.qp_type; - atomic_set(&qp->usecnt, 0); atomic_inc(&pd->usecnt); - qp->port = 0; if (attr.send_cq) atomic_inc(&attr.send_cq->usecnt); if (attr.recv_cq) diff --git a/drivers/infiniband/core/verbs.c b/drivers/infiniband/core/verbs.c index 3ebae3b65c28..e62c9dfc7837 100644 --- a/drivers/infiniband/core/verbs.c +++ b/drivers/infiniband/core/verbs.c @@ -1185,16 +1185,6 @@ struct ib_qp *ib_create_qp_user(struct ib_pd *pd, if (ret) goto err; - qp->qp_type = qp_init_attr->qp_type; - qp->rwq_ind_tbl = qp_init_attr->rwq_ind_tbl; - - atomic_set(&qp->usecnt, 0); - qp->mrs_used = 0; - spin_lock_init(&qp->mr_lock); - INIT_LIST_HEAD(&qp->rdma_mrs); - INIT_LIST_HEAD(&qp->sig_mrs); - qp->port = 0; - if (qp_init_attr->qp_type == IB_QPT_XRC_TGT) { struct ib_qp *xrc_qp = create_xrc_qp_user(qp, qp_init_attr, udata); diff --git a/drivers/infiniband/hw/hfi1/user_sdma.c b/drivers/infiniband/hw/hfi1/user_sdma.c index c2f0d9ba93de..13e4203497b3 100644 --- a/drivers/infiniband/hw/hfi1/user_sdma.c +++ b/drivers/infiniband/hw/hfi1/user_sdma.c @@ -141,6 +141,7 @@ static int defer_packet_queue( */ xchg(&pq->state, SDMA_PKT_Q_DEFERRED); if (list_empty(&pq->busy.list)) { + pq->busy.lock = &sde->waitlock; iowait_get_priority(&pq->busy); iowait_queue(pkts_sent, &pq->busy, &sde->dmawait); } @@ -155,6 +156,7 @@ static void activate_packet_queue(struct iowait *wait, int reason) { struct hfi1_user_sdma_pkt_q *pq = container_of(wait, struct hfi1_user_sdma_pkt_q, busy); + pq->busy.lock = NULL; xchg(&pq->state, SDMA_PKT_Q_ACTIVE); wake_up(&wait->wait_dma); }; @@ -256,6 +258,21 @@ pq_reqs_nomem: return ret; } +static void flush_pq_iowait(struct hfi1_user_sdma_pkt_q *pq) +{ + unsigned long flags; + seqlock_t *lock = pq->busy.lock; + + if (!lock) + return; + write_seqlock_irqsave(lock, flags); + if (!list_empty(&pq->busy.list)) { + list_del_init(&pq->busy.list); + pq->busy.lock = NULL; + } + write_sequnlock_irqrestore(lock, flags); +} + int hfi1_user_sdma_free_queues(struct hfi1_filedata *fd, struct hfi1_ctxtdata *uctxt) { @@ -281,6 +298,7 @@ int hfi1_user_sdma_free_queues(struct hfi1_filedata *fd, kfree(pq->reqs); kfree(pq->req_in_use); kmem_cache_destroy(pq->txreq_cache); + flush_pq_iowait(pq); kfree(pq); } else { spin_unlock(&fd->pq_rcu_lock); @@ -587,11 +605,12 @@ int hfi1_user_sdma_process_request(struct hfi1_filedata *fd, if (ret < 0) { if (ret != -EBUSY) goto free_req; - wait_event_interruptible_timeout( + if (wait_event_interruptible_timeout( pq->busy.wait_dma, - (pq->state == SDMA_PKT_Q_ACTIVE), + pq->state == SDMA_PKT_Q_ACTIVE, msecs_to_jiffies( - SDMA_IOWAIT_TIMEOUT)); + SDMA_IOWAIT_TIMEOUT)) <= 0) + flush_pq_iowait(pq); } } *count += idx; diff --git a/drivers/infiniband/hw/hfi1/verbs.c b/drivers/infiniband/hw/hfi1/verbs.c index 089e201d7550..2f6323ad9c59 100644 --- a/drivers/infiniband/hw/hfi1/verbs.c +++ b/drivers/infiniband/hw/hfi1/verbs.c @@ -515,10 +515,11 @@ static inline void hfi1_handle_packet(struct hfi1_packet *packet, opa_get_lid(packet->dlid, 9B)); if (!mcast) goto drop; + rcu_read_lock(); list_for_each_entry_rcu(p, &mcast->qp_list, list) { packet->qp = p->qp; if (hfi1_do_pkey_check(packet)) - goto drop; + goto unlock_drop; spin_lock_irqsave(&packet->qp->r_lock, flags); packet_handler = qp_ok(packet); if (likely(packet_handler)) @@ -527,6 +528,7 @@ static inline void hfi1_handle_packet(struct hfi1_packet *packet, ibp->rvp.n_pkt_drops++; spin_unlock_irqrestore(&packet->qp->r_lock, flags); } + rcu_read_unlock(); /* * Notify rvt_multicast_detach() if it is waiting for us * to finish. diff --git a/drivers/infiniband/hw/mlx5/cq.c b/drivers/infiniband/hw/mlx5/cq.c index 367a71bc5f4b..3dec3de903b7 100644 --- a/drivers/infiniband/hw/mlx5/cq.c +++ b/drivers/infiniband/hw/mlx5/cq.c @@ -330,6 +330,22 @@ static void mlx5_handle_error_cqe(struct mlx5_ib_dev *dev, dump_cqe(dev, cqe); } +static void handle_atomics(struct mlx5_ib_qp *qp, struct mlx5_cqe64 *cqe64, + u16 tail, u16 head) +{ + u16 idx; + + do { + idx = tail & (qp->sq.wqe_cnt - 1); + if (idx == head) + break; + + tail = qp->sq.w_list[idx].next; + } while (1); + tail = qp->sq.w_list[idx].next; + qp->sq.last_poll = tail; +} + static void free_cq_buf(struct mlx5_ib_dev *dev, struct mlx5_ib_cq_buf *buf) { mlx5_frag_buf_free(dev->mdev, &buf->frag_buf); @@ -368,7 +384,7 @@ static void get_sig_err_item(struct mlx5_sig_err_cqe *cqe, } static void sw_comp(struct mlx5_ib_qp *qp, int num_entries, struct ib_wc *wc, - int *npolled, int is_send) + int *npolled, bool is_send) { struct mlx5_ib_wq *wq; unsigned int cur; @@ -383,10 +399,16 @@ static void sw_comp(struct mlx5_ib_qp *qp, int num_entries, struct ib_wc *wc, return; for (i = 0; i < cur && np < num_entries; i++) { - wc->wr_id = wq->wrid[wq->tail & (wq->wqe_cnt - 1)]; + unsigned int idx; + + idx = (is_send) ? wq->last_poll : wq->tail; + idx &= (wq->wqe_cnt - 1); + wc->wr_id = wq->wrid[idx]; wc->status = IB_WC_WR_FLUSH_ERR; wc->vendor_err = MLX5_CQE_SYNDROME_WR_FLUSH_ERR; wq->tail++; + if (is_send) + wq->last_poll = wq->w_list[idx].next; np++; wc->qp = &qp->ibqp; wc++; @@ -473,6 +495,7 @@ repoll: wqe_ctr = be16_to_cpu(cqe64->wqe_counter); idx = wqe_ctr & (wq->wqe_cnt - 1); handle_good_req(wc, cqe64, wq, idx); + handle_atomics(*cur_qp, cqe64, wq->last_poll, idx); wc->wr_id = wq->wrid[idx]; wq->tail = wq->wqe_head[idx] + 1; wc->status = IB_WC_SUCCESS; diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c index e4bcfa81b70a..ffa7c2100edb 100644 --- a/drivers/infiniband/hw/mlx5/main.c +++ b/drivers/infiniband/hw/mlx5/main.c @@ -5722,9 +5722,10 @@ mlx5_ib_counter_alloc_stats(struct rdma_counter *counter) const struct mlx5_ib_counters *cnts = get_counters(dev, counter->port - 1); - /* Q counters are in the beginning of all counters */ return rdma_alloc_hw_stats_struct(cnts->names, - cnts->num_q_counters, + cnts->num_q_counters + + cnts->num_cong_counters + + cnts->num_ext_ppcnt_counters, RDMA_HW_STATS_DEFAULT_LIFESPAN); } diff --git a/drivers/infiniband/hw/mlx5/mlx5_ib.h b/drivers/infiniband/hw/mlx5/mlx5_ib.h index d9bffcc93587..f3bdbd5e5096 100644 --- a/drivers/infiniband/hw/mlx5/mlx5_ib.h +++ b/drivers/infiniband/hw/mlx5/mlx5_ib.h @@ -288,6 +288,7 @@ struct mlx5_ib_wq { unsigned head; unsigned tail; u16 cur_post; + u16 last_poll; void *cur_edge; }; @@ -636,6 +637,7 @@ struct mlx5_ib_mr { /* For ODP and implicit */ atomic_t num_deferred_work; + wait_queue_head_t q_deferred_work; struct xarray implicit_children; union { struct rcu_head rcu; diff --git a/drivers/infiniband/hw/mlx5/odp.c b/drivers/infiniband/hw/mlx5/odp.c index 4216814ba871..bf50cd91f472 100644 --- a/drivers/infiniband/hw/mlx5/odp.c +++ b/drivers/infiniband/hw/mlx5/odp.c @@ -235,7 +235,8 @@ static void free_implicit_child_mr(struct mlx5_ib_mr *mr, bool need_imr_xlt) mr->parent = NULL; mlx5_mr_cache_free(mr->dev, mr); ib_umem_odp_release(odp); - atomic_dec(&imr->num_deferred_work); + if (atomic_dec_and_test(&imr->num_deferred_work)) + wake_up(&imr->q_deferred_work); } static void free_implicit_child_mr_work(struct work_struct *work) @@ -554,6 +555,7 @@ struct mlx5_ib_mr *mlx5_ib_alloc_implicit_mr(struct mlx5_ib_pd *pd, imr->umem = &umem_odp->umem; imr->is_odp_implicit = true; atomic_set(&imr->num_deferred_work, 0); + init_waitqueue_head(&imr->q_deferred_work); xa_init(&imr->implicit_children); err = mlx5_ib_update_xlt(imr, 0, @@ -611,10 +613,7 @@ void mlx5_ib_free_implicit_mr(struct mlx5_ib_mr *imr) * under xa_lock while the child is in the xarray. Thus at this point * it is only decreasing, and all work holding it is now on the wq. */ - if (atomic_read(&imr->num_deferred_work)) { - flush_workqueue(system_unbound_wq); - WARN_ON(atomic_read(&imr->num_deferred_work)); - } + wait_event(imr->q_deferred_work, !atomic_read(&imr->num_deferred_work)); /* * Fence the imr before we destroy the children. This allows us to @@ -645,10 +644,7 @@ void mlx5_ib_fence_odp_mr(struct mlx5_ib_mr *mr) /* Wait for all running page-fault handlers to finish. */ synchronize_srcu(&mr->dev->odp_srcu); - if (atomic_read(&mr->num_deferred_work)) { - flush_workqueue(system_unbound_wq); - WARN_ON(atomic_read(&mr->num_deferred_work)); - } + wait_event(mr->q_deferred_work, !atomic_read(&mr->num_deferred_work)); dma_fence_odp_mr(mr); } @@ -1720,7 +1716,8 @@ static void destroy_prefetch_work(struct prefetch_mr_work *work) u32 i; for (i = 0; i < work->num_sge; ++i) - atomic_dec(&work->frags[i].mr->num_deferred_work); + if (atomic_dec_and_test(&work->frags[i].mr->num_deferred_work)) + wake_up(&work->frags[i].mr->q_deferred_work); kvfree(work); } diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5/qp.c index 957f3a52589b..8fe149e808af 100644 --- a/drivers/infiniband/hw/mlx5/qp.c +++ b/drivers/infiniband/hw/mlx5/qp.c @@ -3775,6 +3775,7 @@ static int __mlx5_ib_modify_qp(struct ib_qp *ibqp, qp->sq.cur_post = 0; if (qp->sq.wqe_cnt) qp->sq.cur_edge = get_sq_edge(&qp->sq, 0); + qp->sq.last_poll = 0; qp->db.db[MLX5_RCV_DBR] = 0; qp->db.db[MLX5_SND_DBR] = 0; } @@ -6204,6 +6205,10 @@ struct ib_wq *mlx5_ib_create_wq(struct ib_pd *pd, if (udata->outlen && udata->outlen < min_resp_len) return ERR_PTR(-EINVAL); + if (!capable(CAP_SYS_RAWIO) && + init_attr->create_flags & IB_WQ_FLAGS_DELAY_DROP) + return ERR_PTR(-EPERM); + dev = to_mdev(pd->device); switch (init_attr->wq_type) { case IB_WQT_RQ: diff --git a/drivers/infiniband/hw/qib/qib_verbs.c b/drivers/infiniband/hw/qib/qib_verbs.c index 33778d451b82..5ef93f8f17a1 100644 --- a/drivers/infiniband/hw/qib/qib_verbs.c +++ b/drivers/infiniband/hw/qib/qib_verbs.c @@ -329,8 +329,10 @@ void qib_ib_rcv(struct qib_ctxtdata *rcd, void *rhdr, void *data, u32 tlen) if (mcast == NULL) goto drop; this_cpu_inc(ibp->pmastats->n_multicast_rcv); + rcu_read_lock(); list_for_each_entry_rcu(p, &mcast->qp_list, list) qib_qp_rcv(rcd, hdr, 1, data, tlen, p->qp); + rcu_read_unlock(); /* * Notify rvt_multicast_detach() if it is waiting for us * to finish. diff --git a/drivers/infiniband/sw/rdmavt/cq.c b/drivers/infiniband/sw/rdmavt/cq.c index 13d7f66eadab..5724cbbe38b1 100644 --- a/drivers/infiniband/sw/rdmavt/cq.c +++ b/drivers/infiniband/sw/rdmavt/cq.c @@ -327,7 +327,7 @@ void rvt_destroy_cq(struct ib_cq *ibcq, struct ib_udata *udata) if (cq->ip) kref_put(&cq->ip->ref, rvt_release_mmap_info); else - vfree(cq->queue); + vfree(cq->kqueue); } /** diff --git a/drivers/infiniband/sw/siw/siw_main.c b/drivers/infiniband/sw/siw/siw_main.c index 96ed349c0939..5cd40fb9e20c 100644 --- a/drivers/infiniband/sw/siw/siw_main.c +++ b/drivers/infiniband/sw/siw/siw_main.c @@ -388,6 +388,9 @@ static struct siw_device *siw_device_create(struct net_device *netdev) { .max_segment_size = SZ_2G }; base_dev->num_comp_vectors = num_possible_cpus(); + xa_init_flags(&sdev->qp_xa, XA_FLAGS_ALLOC1); + xa_init_flags(&sdev->mem_xa, XA_FLAGS_ALLOC1); + ib_set_device_ops(base_dev, &siw_device_ops); rv = ib_device_set_netdev(base_dev, netdev, 1); if (rv) @@ -415,9 +418,6 @@ static struct siw_device *siw_device_create(struct net_device *netdev) sdev->attrs.max_srq_wr = SIW_MAX_SRQ_WR; sdev->attrs.max_srq_sge = SIW_MAX_SGE; - xa_init_flags(&sdev->qp_xa, XA_FLAGS_ALLOC1); - xa_init_flags(&sdev->mem_xa, XA_FLAGS_ALLOC1); - INIT_LIST_HEAD(&sdev->cep_list); INIT_LIST_HEAD(&sdev->qp_list); diff --git a/drivers/infiniband/ulp/isert/ib_isert.c b/drivers/infiniband/ulp/isert/ib_isert.c index b273e421e910..a1a035270cab 100644 --- a/drivers/infiniband/ulp/isert/ib_isert.c +++ b/drivers/infiniband/ulp/isert/ib_isert.c @@ -2575,6 +2575,17 @@ isert_wait4logout(struct isert_conn *isert_conn) } } +static void +isert_wait4cmds(struct iscsi_conn *conn) +{ + isert_info("iscsi_conn %p\n", conn); + + if (conn->sess) { + target_sess_cmd_list_set_waiting(conn->sess->se_sess); + target_wait_for_sess_cmds(conn->sess->se_sess); + } +} + /** * isert_put_unsol_pending_cmds() - Drop commands waiting for * unsolicitate dataout @@ -2622,6 +2633,7 @@ static void isert_wait_conn(struct iscsi_conn *conn) ib_drain_qp(isert_conn->qp); isert_put_unsol_pending_cmds(conn); + isert_wait4cmds(conn); isert_wait4logout(isert_conn); queue_work(isert_release_wq, &isert_conn->release_work); diff --git a/drivers/input/input.c b/drivers/input/input.c index fce43e62dd45..3cfd2c18eebd 100644 --- a/drivers/input/input.c +++ b/drivers/input/input.c @@ -190,6 +190,7 @@ static void input_repeat_key(struct timer_list *t) input_value_sync }; + input_set_timestamp(dev, ktime_get()); input_pass_values(dev, vals, ARRAY_SIZE(vals)); if (dev->rep[REP_PERIOD]) diff --git a/drivers/input/keyboard/tm2-touchkey.c b/drivers/input/keyboard/tm2-touchkey.c index 14b55bacdd0f..fb078e049413 100644 --- a/drivers/input/keyboard/tm2-touchkey.c +++ b/drivers/input/keyboard/tm2-touchkey.c @@ -75,6 +75,14 @@ static struct touchkey_variant aries_touchkey_variant = { .cmd_led_off = ARIES_TOUCHKEY_CMD_LED_OFF, }; +static const struct touchkey_variant tc360_touchkey_variant = { + .keycode_reg = 0x00, + .base_reg = 0x00, + .fixed_regulator = true, + .cmd_led_on = TM2_TOUCHKEY_CMD_LED_ON, + .cmd_led_off = TM2_TOUCHKEY_CMD_LED_OFF, +}; + static int tm2_touchkey_led_brightness_set(struct led_classdev *led_dev, enum led_brightness brightness) { @@ -327,6 +335,9 @@ static const struct of_device_id tm2_touchkey_of_match[] = { }, { .compatible = "cypress,aries-touchkey", .data = &aries_touchkey_variant, + }, { + .compatible = "coreriver,tc360-touchkey", + .data = &tc360_touchkey_variant, }, { }, }; diff --git a/drivers/input/mouse/synaptics.c b/drivers/input/mouse/synaptics.c index 2c666fb34625..4d2036209b45 100644 --- a/drivers/input/mouse/synaptics.c +++ b/drivers/input/mouse/synaptics.c @@ -186,6 +186,7 @@ static const char * const smbus_pnp_ids[] = { "SYN3052", /* HP EliteBook 840 G4 */ "SYN3221", /* HP 15-ay000 */ "SYN323d", /* HP Spectre X360 13-w013dx */ + "SYN3257", /* HP Envy 13-ad105ng */ NULL }; diff --git a/drivers/input/rmi4/rmi_f11.c b/drivers/input/rmi4/rmi_f11.c index 6adea8a3e8fb..ffa39ab153f2 100644 --- a/drivers/input/rmi4/rmi_f11.c +++ b/drivers/input/rmi4/rmi_f11.c @@ -1203,8 +1203,8 @@ static int rmi_f11_initialize(struct rmi_function *fn) * If distance threshold values are set, switch to reduced reporting * mode so they actually get used by the controller. */ - if (ctrl->ctrl0_11[RMI_F11_DELTA_X_THRESHOLD] || - ctrl->ctrl0_11[RMI_F11_DELTA_Y_THRESHOLD]) { + if (sensor->axis_align.delta_x_threshold || + sensor->axis_align.delta_y_threshold) { ctrl->ctrl0_11[0] &= ~RMI_F11_REPORT_MODE_MASK; ctrl->ctrl0_11[0] |= RMI_F11_REPORT_MODE_REDUCED; } diff --git a/drivers/input/touchscreen/raydium_i2c_ts.c b/drivers/input/touchscreen/raydium_i2c_ts.c index 6ed9f22e6401..fe245439adee 100644 --- a/drivers/input/touchscreen/raydium_i2c_ts.c +++ b/drivers/input/touchscreen/raydium_i2c_ts.c @@ -432,7 +432,7 @@ static int raydium_i2c_write_object(struct i2c_client *client, return 0; } -static bool raydium_i2c_boot_trigger(struct i2c_client *client) +static int raydium_i2c_boot_trigger(struct i2c_client *client) { static const u8 cmd[7][6] = { { 0x08, 0x0C, 0x09, 0x00, 0x50, 0xD7 }, @@ -457,10 +457,10 @@ static bool raydium_i2c_boot_trigger(struct i2c_client *client) } } - return false; + return 0; } -static bool raydium_i2c_fw_trigger(struct i2c_client *client) +static int raydium_i2c_fw_trigger(struct i2c_client *client) { static const u8 cmd[5][11] = { { 0, 0x09, 0x71, 0x0C, 0x09, 0x00, 0x50, 0xD7, 0, 0, 0 }, @@ -483,7 +483,7 @@ static bool raydium_i2c_fw_trigger(struct i2c_client *client) } } - return false; + return 0; } static int raydium_i2c_check_path(struct i2c_client *client) diff --git a/drivers/interconnect/core.c b/drivers/interconnect/core.c index f277e467156f..2c6515e3ecf1 100644 --- a/drivers/interconnect/core.c +++ b/drivers/interconnect/core.c @@ -445,6 +445,11 @@ struct icc_path *of_icc_get(struct device *dev, const char *name) path->name = kasprintf(GFP_KERNEL, "%s-%s", src_node->name, dst_node->name); + if (!path->name) { + kfree(path); + return ERR_PTR(-ENOMEM); + } + return path; } EXPORT_SYMBOL_GPL(of_icc_get); @@ -579,6 +584,10 @@ struct icc_path *icc_get(struct device *dev, const int src_id, const int dst_id) } path->name = kasprintf(GFP_KERNEL, "%s-%s", src->name, dst->name); + if (!path->name) { + kfree(path); + path = ERR_PTR(-ENOMEM); + } out: mutex_unlock(&icc_lock); return path; diff --git a/drivers/iommu/Makefile b/drivers/iommu/Makefile index 2104fb8afc06..9f33fdb3bb05 100644 --- a/drivers/iommu/Makefile +++ b/drivers/iommu/Makefile @@ -14,8 +14,8 @@ obj-$(CONFIG_MSM_IOMMU) += msm_iommu.o obj-$(CONFIG_AMD_IOMMU) += amd_iommu.o amd_iommu_init.o amd_iommu_quirks.o obj-$(CONFIG_AMD_IOMMU_DEBUGFS) += amd_iommu_debugfs.o obj-$(CONFIG_AMD_IOMMU_V2) += amd_iommu_v2.o -obj-$(CONFIG_ARM_SMMU) += arm-smmu-mod.o -arm-smmu-mod-objs += arm-smmu.o arm-smmu-impl.o arm-smmu-qcom.o +obj-$(CONFIG_ARM_SMMU) += arm_smmu.o +arm_smmu-objs += arm-smmu.o arm-smmu-impl.o arm-smmu-qcom.o obj-$(CONFIG_ARM_SMMU_V3) += arm-smmu-v3.o obj-$(CONFIG_DMAR_TABLE) += dmar.o obj-$(CONFIG_INTEL_IOMMU) += intel-iommu.o intel-pasid.o diff --git a/drivers/iommu/amd_iommu.c b/drivers/iommu/amd_iommu.c index aac132bd1ef0..20cce366e951 100644 --- a/drivers/iommu/amd_iommu.c +++ b/drivers/iommu/amd_iommu.c @@ -3826,7 +3826,7 @@ int amd_iommu_activate_guest_mode(void *data) entry->lo.fields_vapic.ga_tag = ir_data->ga_tag; return modify_irte_ga(ir_data->irq_2_irte.devid, - ir_data->irq_2_irte.index, entry, NULL); + ir_data->irq_2_irte.index, entry, ir_data); } EXPORT_SYMBOL(amd_iommu_activate_guest_mode); @@ -3852,7 +3852,7 @@ int amd_iommu_deactivate_guest_mode(void *data) APICID_TO_IRTE_DEST_HI(cfg->dest_apicid); return modify_irte_ga(ir_data->irq_2_irte.devid, - ir_data->irq_2_irte.index, entry, NULL); + ir_data->irq_2_irte.index, entry, ir_data); } EXPORT_SYMBOL(amd_iommu_deactivate_guest_mode); diff --git a/drivers/iommu/amd_iommu_init.c b/drivers/iommu/amd_iommu_init.c index 2759a8d57b7f..6be3853a5d97 100644 --- a/drivers/iommu/amd_iommu_init.c +++ b/drivers/iommu/amd_iommu_init.c @@ -2523,6 +2523,7 @@ static int __init early_amd_iommu_init(void) struct acpi_table_header *ivrs_base; acpi_status status; int i, remap_cache_sz, ret = 0; + u32 pci_id; if (!amd_iommu_detected) return -ENODEV; @@ -2610,6 +2611,16 @@ static int __init early_amd_iommu_init(void) if (ret) goto out; + /* Disable IOMMU if there's Stoney Ridge graphics */ + for (i = 0; i < 32; i++) { + pci_id = read_pci_config(0, i, 0, 0); + if ((pci_id & 0xffff) == 0x1002 && (pci_id >> 16) == 0x98e4) { + pr_info("Disable IOMMU on Stoney Ridge\n"); + amd_iommu_disabled = true; + break; + } + } + /* Disable any previously enabled IOMMUs */ if (!is_kdump_kernel() || amd_iommu_disabled) disable_iommus(); @@ -2718,7 +2729,7 @@ static int __init state_next(void) ret = early_amd_iommu_init(); init_state = ret ? IOMMU_INIT_ERROR : IOMMU_ACPI_FINISHED; if (init_state == IOMMU_ACPI_FINISHED && amd_iommu_disabled) { - pr_info("AMD IOMMU disabled on kernel command-line\n"); + pr_info("AMD IOMMU disabled\n"); init_state = IOMMU_CMDLINE_DISABLED; ret = -EINVAL; } diff --git a/drivers/iommu/dma-iommu.c b/drivers/iommu/dma-iommu.c index a2e96a5fd9a7..ba128d1cdaee 100644 --- a/drivers/iommu/dma-iommu.c +++ b/drivers/iommu/dma-iommu.c @@ -177,15 +177,15 @@ static int cookie_init_hw_msi_region(struct iommu_dma_cookie *cookie, start -= iova_offset(iovad, start); num_pages = iova_align(iovad, end - start) >> iova_shift(iovad); - msi_page = kcalloc(num_pages, sizeof(*msi_page), GFP_KERNEL); - if (!msi_page) - return -ENOMEM; - for (i = 0; i < num_pages; i++) { - msi_page[i].phys = start; - msi_page[i].iova = start; - INIT_LIST_HEAD(&msi_page[i].list); - list_add(&msi_page[i].list, &cookie->msi_page_list); + msi_page = kmalloc(sizeof(*msi_page), GFP_KERNEL); + if (!msi_page) + return -ENOMEM; + + msi_page->phys = start; + msi_page->iova = start; + INIT_LIST_HEAD(&msi_page->list); + list_add(&msi_page->list, &cookie->msi_page_list); start += iovad->granule; } diff --git a/drivers/iommu/dmar.c b/drivers/iommu/dmar.c index 071bb42bbbc5..f77dae7ba7d4 100644 --- a/drivers/iommu/dmar.c +++ b/drivers/iommu/dmar.c @@ -28,6 +28,7 @@ #include <linux/slab.h> #include <linux/iommu.h> #include <linux/numa.h> +#include <linux/limits.h> #include <asm/irq_remapping.h> #include <asm/iommu_table.h> @@ -128,6 +129,13 @@ dmar_alloc_pci_notify_info(struct pci_dev *dev, unsigned long event) BUG_ON(dev->is_virtfn); + /* + * Ignore devices that have a domain number higher than what can + * be looked up in DMAR, e.g. VMD subdevices with domain 0x10000 + */ + if (pci_domain_nr(dev->bus) > U16_MAX) + return NULL; + /* Only generate path[] for device addition event */ if (event == BUS_NOTIFY_ADD_DEVICE) for (tmp = dev; tmp; tmp = tmp->bus->self) @@ -363,7 +371,8 @@ dmar_find_dmaru(struct acpi_dmar_hardware_unit *drhd) { struct dmar_drhd_unit *dmaru; - list_for_each_entry_rcu(dmaru, &dmar_drhd_units, list) + list_for_each_entry_rcu(dmaru, &dmar_drhd_units, list, + dmar_rcu_check()) if (dmaru->segment == drhd->segment && dmaru->reg_base_addr == drhd->address) return dmaru; @@ -440,12 +449,13 @@ static int __init dmar_parse_one_andd(struct acpi_dmar_header *header, /* Check for NUL termination within the designated length */ if (strnlen(andd->device_name, header->length - 8) == header->length - 8) { - WARN_TAINT(1, TAINT_FIRMWARE_WORKAROUND, + pr_warn(FW_BUG "Your BIOS is broken; ANDD object name is not NUL-terminated\n" "BIOS vendor: %s; Ver: %s; Product Version: %s\n", dmi_get_system_info(DMI_BIOS_VENDOR), dmi_get_system_info(DMI_BIOS_VERSION), dmi_get_system_info(DMI_PRODUCT_VERSION)); + add_taint(TAINT_FIRMWARE_WORKAROUND, LOCKDEP_STILL_OK); return -EINVAL; } pr_info("ANDD device: %x name: %s\n", andd->device_number, @@ -471,14 +481,14 @@ static int dmar_parse_one_rhsa(struct acpi_dmar_header *header, void *arg) return 0; } } - WARN_TAINT( - 1, TAINT_FIRMWARE_WORKAROUND, + pr_warn(FW_BUG "Your BIOS is broken; RHSA refers to non-existent DMAR unit at %llx\n" "BIOS vendor: %s; Ver: %s; Product Version: %s\n", - drhd->reg_base_addr, + rhsa->base_address, dmi_get_system_info(DMI_BIOS_VENDOR), dmi_get_system_info(DMI_BIOS_VERSION), dmi_get_system_info(DMI_PRODUCT_VERSION)); + add_taint(TAINT_FIRMWARE_WORKAROUND, LOCKDEP_STILL_OK); return 0; } @@ -827,14 +837,14 @@ int __init dmar_table_init(void) static void warn_invalid_dmar(u64 addr, const char *message) { - WARN_TAINT_ONCE( - 1, TAINT_FIRMWARE_WORKAROUND, + pr_warn_once(FW_BUG "Your BIOS is broken; DMAR reported at address %llx%s!\n" "BIOS vendor: %s; Ver: %s; Product Version: %s\n", addr, message, dmi_get_system_info(DMI_BIOS_VENDOR), dmi_get_system_info(DMI_BIOS_VERSION), dmi_get_system_info(DMI_PRODUCT_VERSION)); + add_taint(TAINT_FIRMWARE_WORKAROUND, LOCKDEP_STILL_OK); } static int __ref diff --git a/drivers/iommu/intel-iommu-debugfs.c b/drivers/iommu/intel-iommu-debugfs.c index c1257bef553c..3eb1fe240fb0 100644 --- a/drivers/iommu/intel-iommu-debugfs.c +++ b/drivers/iommu/intel-iommu-debugfs.c @@ -33,38 +33,42 @@ struct iommu_regset { #define IOMMU_REGSET_ENTRY(_reg_) \ { DMAR_##_reg_##_REG, __stringify(_reg_) } -static const struct iommu_regset iommu_regs[] = { + +static const struct iommu_regset iommu_regs_32[] = { IOMMU_REGSET_ENTRY(VER), - IOMMU_REGSET_ENTRY(CAP), - IOMMU_REGSET_ENTRY(ECAP), IOMMU_REGSET_ENTRY(GCMD), IOMMU_REGSET_ENTRY(GSTS), - IOMMU_REGSET_ENTRY(RTADDR), - IOMMU_REGSET_ENTRY(CCMD), IOMMU_REGSET_ENTRY(FSTS), IOMMU_REGSET_ENTRY(FECTL), IOMMU_REGSET_ENTRY(FEDATA), IOMMU_REGSET_ENTRY(FEADDR), IOMMU_REGSET_ENTRY(FEUADDR), - IOMMU_REGSET_ENTRY(AFLOG), IOMMU_REGSET_ENTRY(PMEN), IOMMU_REGSET_ENTRY(PLMBASE), IOMMU_REGSET_ENTRY(PLMLIMIT), + IOMMU_REGSET_ENTRY(ICS), + IOMMU_REGSET_ENTRY(PRS), + IOMMU_REGSET_ENTRY(PECTL), + IOMMU_REGSET_ENTRY(PEDATA), + IOMMU_REGSET_ENTRY(PEADDR), + IOMMU_REGSET_ENTRY(PEUADDR), +}; + +static const struct iommu_regset iommu_regs_64[] = { + IOMMU_REGSET_ENTRY(CAP), + IOMMU_REGSET_ENTRY(ECAP), + IOMMU_REGSET_ENTRY(RTADDR), + IOMMU_REGSET_ENTRY(CCMD), + IOMMU_REGSET_ENTRY(AFLOG), IOMMU_REGSET_ENTRY(PHMBASE), IOMMU_REGSET_ENTRY(PHMLIMIT), IOMMU_REGSET_ENTRY(IQH), IOMMU_REGSET_ENTRY(IQT), IOMMU_REGSET_ENTRY(IQA), - IOMMU_REGSET_ENTRY(ICS), IOMMU_REGSET_ENTRY(IRTA), IOMMU_REGSET_ENTRY(PQH), IOMMU_REGSET_ENTRY(PQT), IOMMU_REGSET_ENTRY(PQA), - IOMMU_REGSET_ENTRY(PRS), - IOMMU_REGSET_ENTRY(PECTL), - IOMMU_REGSET_ENTRY(PEDATA), - IOMMU_REGSET_ENTRY(PEADDR), - IOMMU_REGSET_ENTRY(PEUADDR), IOMMU_REGSET_ENTRY(MTRRCAP), IOMMU_REGSET_ENTRY(MTRRDEF), IOMMU_REGSET_ENTRY(MTRR_FIX64K_00000), @@ -127,10 +131,16 @@ static int iommu_regset_show(struct seq_file *m, void *unused) * by adding the offset to the pointer (virtual address). */ raw_spin_lock_irqsave(&iommu->register_lock, flag); - for (i = 0 ; i < ARRAY_SIZE(iommu_regs); i++) { - value = dmar_readq(iommu->reg + iommu_regs[i].offset); + for (i = 0 ; i < ARRAY_SIZE(iommu_regs_32); i++) { + value = dmar_readl(iommu->reg + iommu_regs_32[i].offset); + seq_printf(m, "%-16s\t0x%02x\t\t0x%016llx\n", + iommu_regs_32[i].regs, iommu_regs_32[i].offset, + value); + } + for (i = 0 ; i < ARRAY_SIZE(iommu_regs_64); i++) { + value = dmar_readq(iommu->reg + iommu_regs_64[i].offset); seq_printf(m, "%-16s\t0x%02x\t\t0x%016llx\n", - iommu_regs[i].regs, iommu_regs[i].offset, + iommu_regs_64[i].regs, iommu_regs_64[i].offset, value); } raw_spin_unlock_irqrestore(&iommu->register_lock, flag); @@ -272,9 +282,16 @@ static int dmar_translation_struct_show(struct seq_file *m, void *unused) { struct dmar_drhd_unit *drhd; struct intel_iommu *iommu; + u32 sts; rcu_read_lock(); for_each_active_iommu(iommu, drhd) { + sts = dmar_readl(iommu->reg + DMAR_GSTS_REG); + if (!(sts & DMA_GSTS_TES)) { + seq_printf(m, "DMA Remapping is not enabled on %s\n", + iommu->name); + continue; + } root_tbl_walk(m, iommu); seq_putc(m, '\n'); } @@ -415,6 +432,7 @@ static int ir_translation_struct_show(struct seq_file *m, void *unused) struct dmar_drhd_unit *drhd; struct intel_iommu *iommu; u64 irta; + u32 sts; rcu_read_lock(); for_each_active_iommu(iommu, drhd) { @@ -424,7 +442,8 @@ static int ir_translation_struct_show(struct seq_file *m, void *unused) seq_printf(m, "Remapped Interrupt supported on IOMMU: %s\n", iommu->name); - if (iommu->ir_table) { + sts = dmar_readl(iommu->reg + DMAR_GSTS_REG); + if (iommu->ir_table && (sts & DMA_GSTS_IRES)) { irta = virt_to_phys(iommu->ir_table->base); seq_printf(m, " IR table address:%llx\n", irta); ir_tbl_remap_entry_show(m, iommu); diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c index 9dc37672bf89..4be549478691 100644 --- a/drivers/iommu/intel-iommu.c +++ b/drivers/iommu/intel-iommu.c @@ -762,6 +762,11 @@ static int iommu_dummy(struct device *dev) return dev->archdata.iommu == DUMMY_DEVICE_DOMAIN_INFO; } +static bool attach_deferred(struct device *dev) +{ + return dev->archdata.iommu == DEFER_DEVICE_DOMAIN_INFO; +} + /** * is_downstream_to_pci_bridge - test if a device belongs to the PCI * sub-hierarchy of a candidate PCI-PCI bridge @@ -2510,8 +2515,7 @@ struct dmar_domain *find_domain(struct device *dev) { struct device_domain_info *info; - if (unlikely(dev->archdata.iommu == DEFER_DEVICE_DOMAIN_INFO || - dev->archdata.iommu == DUMMY_DEVICE_DOMAIN_INFO)) + if (unlikely(attach_deferred(dev) || iommu_dummy(dev))) return NULL; if (dev_is_pci(dev)) @@ -2525,18 +2529,14 @@ struct dmar_domain *find_domain(struct device *dev) return NULL; } -static struct dmar_domain *deferred_attach_domain(struct device *dev) +static void do_deferred_attach(struct device *dev) { - if (unlikely(dev->archdata.iommu == DEFER_DEVICE_DOMAIN_INFO)) { - struct iommu_domain *domain; - - dev->archdata.iommu = NULL; - domain = iommu_get_domain_for_dev(dev); - if (domain) - intel_iommu_attach_device(domain, dev); - } + struct iommu_domain *domain; - return find_domain(dev); + dev->archdata.iommu = NULL; + domain = iommu_get_domain_for_dev(dev); + if (domain) + intel_iommu_attach_device(domain, dev); } static inline struct device_domain_info * @@ -2916,7 +2916,7 @@ static int identity_mapping(struct device *dev) struct device_domain_info *info; info = dev->archdata.iommu; - if (info && info != DUMMY_DEVICE_DOMAIN_INFO && info != DEFER_DEVICE_DOMAIN_INFO) + if (info) return (info->domain == si_domain); return 0; @@ -3587,6 +3587,9 @@ static bool iommu_need_mapping(struct device *dev) if (iommu_dummy(dev)) return false; + if (unlikely(attach_deferred(dev))) + do_deferred_attach(dev); + ret = identity_mapping(dev); if (ret) { u64 dma_mask = *dev->dma_mask; @@ -3635,7 +3638,7 @@ static dma_addr_t __intel_map_single(struct device *dev, phys_addr_t paddr, BUG_ON(dir == DMA_NONE); - domain = deferred_attach_domain(dev); + domain = find_domain(dev); if (!domain) return DMA_MAPPING_ERROR; @@ -3855,7 +3858,7 @@ static int intel_map_sg(struct device *dev, struct scatterlist *sglist, int nele if (!iommu_need_mapping(dev)) return dma_direct_map_sg(dev, sglist, nelems, dir, attrs); - domain = deferred_attach_domain(dev); + domain = find_domain(dev); if (!domain) return 0; @@ -3950,7 +3953,11 @@ bounce_map_single(struct device *dev, phys_addr_t paddr, size_t size, int prot = 0; int ret; - domain = deferred_attach_domain(dev); + if (unlikely(attach_deferred(dev))) + do_deferred_attach(dev); + + domain = find_domain(dev); + if (WARN_ON(dir == DMA_NONE || !domain)) return DMA_MAPPING_ERROR; @@ -4254,10 +4261,11 @@ static void quirk_ioat_snb_local_iommu(struct pci_dev *pdev) /* we know that the this iommu should be at offset 0xa000 from vtbar */ drhd = dmar_find_matched_drhd_unit(pdev); - if (WARN_TAINT_ONCE(!drhd || drhd->reg_base_addr - vtbar != 0xa000, - TAINT_FIRMWARE_WORKAROUND, - "BIOS assigned incorrect VT-d unit for Intel(R) QuickData Technology device\n")) + if (!drhd || drhd->reg_base_addr - vtbar != 0xa000) { + pr_warn_once(FW_BUG "BIOS assigned incorrect VT-d unit for Intel(R) QuickData Technology device\n"); + add_taint(TAINT_FIRMWARE_WORKAROUND, LOCKDEP_STILL_OK); pdev->dev.archdata.iommu = DUMMY_DEVICE_DOMAIN_INFO; + } } DECLARE_PCI_FIXUP_ENABLE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_IOAT_SNB, quirk_ioat_snb_local_iommu); @@ -4453,14 +4461,16 @@ int __init dmar_parse_one_rmrr(struct acpi_dmar_header *header, void *arg) struct dmar_rmrr_unit *rmrru; rmrr = (struct acpi_dmar_reserved_memory *)header; - if (rmrr_sanity_check(rmrr)) - WARN_TAINT(1, TAINT_FIRMWARE_WORKAROUND, + if (rmrr_sanity_check(rmrr)) { + pr_warn(FW_BUG "Your BIOS is broken; bad RMRR [%#018Lx-%#018Lx]\n" "BIOS vendor: %s; Ver: %s; Product Version: %s\n", rmrr->base_address, rmrr->end_address, dmi_get_system_info(DMI_BIOS_VENDOR), dmi_get_system_info(DMI_BIOS_VERSION), dmi_get_system_info(DMI_PRODUCT_VERSION)); + add_taint(TAINT_FIRMWARE_WORKAROUND, LOCKDEP_STILL_OK); + } rmrru = kzalloc(sizeof(*rmrru), GFP_KERNEL); if (!rmrru) @@ -5123,6 +5133,9 @@ int __init intel_iommu_init(void) down_write(&dmar_global_lock); + if (!no_iommu) + intel_iommu_debugfs_init(); + if (no_iommu || dmar_disabled) { /* * We exit the function here to ensure IOMMU's remapping and @@ -5186,6 +5199,7 @@ int __init intel_iommu_init(void) init_iommu_pm_ops(); + down_read(&dmar_global_lock); for_each_active_iommu(iommu, drhd) { iommu_device_sysfs_add(&iommu->iommu, NULL, intel_iommu_groups, @@ -5193,6 +5207,7 @@ int __init intel_iommu_init(void) iommu_device_set_ops(&iommu->iommu, &intel_iommu_ops); iommu_device_register(&iommu->iommu); } + up_read(&dmar_global_lock); bus_set_iommu(&pci_bus_type, &intel_iommu_ops); if (si_domain && !hw_pass_through) @@ -5203,7 +5218,6 @@ int __init intel_iommu_init(void) down_read(&dmar_global_lock); if (probe_acpi_namespace_devices()) pr_warn("ACPI name space devices didn't probe correctly\n"); - up_read(&dmar_global_lock); /* Finally, we enable the DMA remapping hardware. */ for_each_iommu(iommu, drhd) { @@ -5212,10 +5226,11 @@ int __init intel_iommu_init(void) iommu_disable_protect_mem_regions(iommu); } + up_read(&dmar_global_lock); + pr_info("Intel(R) Virtualization Technology for Directed I/O\n"); intel_iommu_enabled = 1; - intel_iommu_debugfs_init(); return 0; @@ -5693,8 +5708,10 @@ static phys_addr_t intel_iommu_iova_to_phys(struct iommu_domain *domain, u64 phys = 0; pte = pfn_to_dma_pte(dmar_domain, iova >> VTD_PAGE_SHIFT, &level); - if (pte) - phys = dma_pte_addr(pte); + if (pte && dma_pte_present(pte)) + phys = dma_pte_addr(pte) + + (iova & (BIT_MASK(level_to_offset_bits(level) + + VTD_PAGE_SHIFT) - 1)); return phys; } @@ -6133,7 +6150,7 @@ intel_iommu_aux_get_pasid(struct iommu_domain *domain, struct device *dev) static bool intel_iommu_is_attach_deferred(struct iommu_domain *domain, struct device *dev) { - return dev->archdata.iommu == DEFER_DEVICE_DOMAIN_INFO; + return attach_deferred(dev); } static int diff --git a/drivers/iommu/io-pgtable-arm.c b/drivers/iommu/io-pgtable-arm.c index 983b08477e64..04fbd4bf0ff9 100644 --- a/drivers/iommu/io-pgtable-arm.c +++ b/drivers/iommu/io-pgtable-arm.c @@ -468,7 +468,7 @@ static int arm_lpae_map(struct io_pgtable_ops *ops, unsigned long iova, arm_lpae_iopte *ptep = data->pgd; int ret, lvl = data->start_level; arm_lpae_iopte prot; - long iaext = (long)iova >> cfg->ias; + long iaext = (s64)iova >> cfg->ias; /* If no access, then nothing to do */ if (!(iommu_prot & (IOMMU_READ | IOMMU_WRITE))) @@ -645,7 +645,7 @@ static size_t arm_lpae_unmap(struct io_pgtable_ops *ops, unsigned long iova, struct arm_lpae_io_pgtable *data = io_pgtable_ops_to_data(ops); struct io_pgtable_cfg *cfg = &data->iop.cfg; arm_lpae_iopte *ptep = data->pgd; - long iaext = (long)iova >> cfg->ias; + long iaext = (s64)iova >> cfg->ias; if (WARN_ON(!size || (size & cfg->pgsize_bitmap) != size)) return 0; diff --git a/drivers/iommu/qcom_iommu.c b/drivers/iommu/qcom_iommu.c index 39759db4f003..4328da0b0a9f 100644 --- a/drivers/iommu/qcom_iommu.c +++ b/drivers/iommu/qcom_iommu.c @@ -344,21 +344,19 @@ static void qcom_iommu_domain_free(struct iommu_domain *domain) { struct qcom_iommu_domain *qcom_domain = to_qcom_iommu_domain(domain); - if (WARN_ON(qcom_domain->iommu)) /* forgot to detach? */ - return; - iommu_put_dma_cookie(domain); - /* NOTE: unmap can be called after client device is powered off, - * for example, with GPUs or anything involving dma-buf. So we - * cannot rely on the device_link. Make sure the IOMMU is on to - * avoid unclocked accesses in the TLB inv path: - */ - pm_runtime_get_sync(qcom_domain->iommu->dev); - - free_io_pgtable_ops(qcom_domain->pgtbl_ops); - - pm_runtime_put_sync(qcom_domain->iommu->dev); + if (qcom_domain->iommu) { + /* + * NOTE: unmap can be called after client device is powered + * off, for example, with GPUs or anything involving dma-buf. + * So we cannot rely on the device_link. Make sure the IOMMU + * is on to avoid unclocked accesses in the TLB inv path: + */ + pm_runtime_get_sync(qcom_domain->iommu->dev); + free_io_pgtable_ops(qcom_domain->pgtbl_ops); + pm_runtime_put_sync(qcom_domain->iommu->dev); + } kfree(qcom_domain); } @@ -404,7 +402,7 @@ static void qcom_iommu_detach_dev(struct iommu_domain *domain, struct device *de struct qcom_iommu_domain *qcom_domain = to_qcom_iommu_domain(domain); unsigned i; - if (!qcom_domain->iommu) + if (WARN_ON(!qcom_domain->iommu)) return; pm_runtime_get_sync(qcom_iommu->dev); @@ -417,8 +415,6 @@ static void qcom_iommu_detach_dev(struct iommu_domain *domain, struct device *de ctx->domain = NULL; } pm_runtime_put_sync(qcom_iommu->dev); - - qcom_domain->iommu = NULL; } static int qcom_iommu_map(struct iommu_domain *domain, unsigned long iova, diff --git a/drivers/irqchip/irq-gic-v3.c b/drivers/irqchip/irq-gic-v3.c index c1f7af9d9ae7..1eec9d4649d5 100644 --- a/drivers/irqchip/irq-gic-v3.c +++ b/drivers/irqchip/irq-gic-v3.c @@ -34,6 +34,7 @@ #define GICD_INT_NMI_PRI (GICD_INT_DEF_PRI & ~0x80) #define FLAGS_WORKAROUND_GICR_WAKER_MSM8996 (1ULL << 0) +#define FLAGS_WORKAROUND_CAVIUM_ERRATUM_38539 (1ULL << 1) struct redist_region { void __iomem *redist_base; @@ -1464,6 +1465,15 @@ static bool gic_enable_quirk_msm8996(void *data) return true; } +static bool gic_enable_quirk_cavium_38539(void *data) +{ + struct gic_chip_data *d = data; + + d->flags |= FLAGS_WORKAROUND_CAVIUM_ERRATUM_38539; + + return true; +} + static bool gic_enable_quirk_hip06_07(void *data) { struct gic_chip_data *d = data; @@ -1503,6 +1513,19 @@ static const struct gic_quirk gic_quirks[] = { .init = gic_enable_quirk_hip06_07, }, { + /* + * Reserved register accesses generate a Synchronous + * External Abort. This erratum applies to: + * - ThunderX: CN88xx + * - OCTEON TX: CN83xx, CN81xx + * - OCTEON TX2: CN93xx, CN96xx, CN98xx, CNF95xx* + */ + .desc = "GICv3: Cavium erratum 38539", + .iidr = 0xa000034c, + .mask = 0xe8f00fff, + .init = gic_enable_quirk_cavium_38539, + }, + { } }; @@ -1577,7 +1600,12 @@ static int __init gic_init_bases(void __iomem *dist_base, pr_info("%d SPIs implemented\n", GIC_LINE_NR - 32); pr_info("%d Extended SPIs implemented\n", GIC_ESPI_NR); - gic_data.rdists.gicd_typer2 = readl_relaxed(gic_data.dist_base + GICD_TYPER2); + /* + * ThunderX1 explodes on reading GICD_TYPER2, in violation of the + * architecture spec (which says that reserved registers are RES0). + */ + if (!(gic_data.flags & FLAGS_WORKAROUND_CAVIUM_ERRATUM_38539)) + gic_data.rdists.gicd_typer2 = readl_relaxed(gic_data.dist_base + GICD_TYPER2); gic_data.domain = irq_domain_create_tree(handle, &gic_irq_domain_ops, &gic_data); diff --git a/drivers/macintosh/therm_windtunnel.c b/drivers/macintosh/therm_windtunnel.c index 8c744578122a..a0d87ed9da69 100644 --- a/drivers/macintosh/therm_windtunnel.c +++ b/drivers/macintosh/therm_windtunnel.c @@ -300,9 +300,11 @@ static int control_loop(void *dummy) /* i2c probing and setup */ /************************************************************************/ -static int -do_attach( struct i2c_adapter *adapter ) +static void do_attach(struct i2c_adapter *adapter) { + struct i2c_board_info info = { }; + struct device_node *np; + /* scan 0x48-0x4f (DS1775) and 0x2c-2x2f (ADM1030) */ static const unsigned short scan_ds1775[] = { 0x48, 0x49, 0x4a, 0x4b, 0x4c, 0x4d, 0x4e, 0x4f, @@ -313,25 +315,24 @@ do_attach( struct i2c_adapter *adapter ) I2C_CLIENT_END }; - if( strncmp(adapter->name, "uni-n", 5) ) - return 0; - - if( !x.running ) { - struct i2c_board_info info; + if (x.running || strncmp(adapter->name, "uni-n", 5)) + return; - memset(&info, 0, sizeof(struct i2c_board_info)); - strlcpy(info.type, "therm_ds1775", I2C_NAME_SIZE); + np = of_find_compatible_node(adapter->dev.of_node, NULL, "MAC,ds1775"); + if (np) { + of_node_put(np); + } else { + strlcpy(info.type, "MAC,ds1775", I2C_NAME_SIZE); i2c_new_probed_device(adapter, &info, scan_ds1775, NULL); + } - strlcpy(info.type, "therm_adm1030", I2C_NAME_SIZE); + np = of_find_compatible_node(adapter->dev.of_node, NULL, "MAC,adm1030"); + if (np) { + of_node_put(np); + } else { + strlcpy(info.type, "MAC,adm1030", I2C_NAME_SIZE); i2c_new_probed_device(adapter, &info, scan_adm1030, NULL); - - if( x.thermostat && x.fan ) { - x.running = 1; - x.poll_task = kthread_run(control_loop, NULL, "g4fand"); - } } - return 0; } static int @@ -404,8 +405,8 @@ out: enum chip { ds1775, adm1030 }; static const struct i2c_device_id therm_windtunnel_id[] = { - { "therm_ds1775", ds1775 }, - { "therm_adm1030", adm1030 }, + { "MAC,ds1775", ds1775 }, + { "MAC,adm1030", adm1030 }, { } }; MODULE_DEVICE_TABLE(i2c, therm_windtunnel_id); @@ -414,6 +415,7 @@ static int do_probe(struct i2c_client *cl, const struct i2c_device_id *id) { struct i2c_adapter *adapter = cl->adapter; + int ret = 0; if( !i2c_check_functionality(adapter, I2C_FUNC_SMBUS_WORD_DATA | I2C_FUNC_SMBUS_WRITE_BYTE) ) @@ -421,11 +423,19 @@ do_probe(struct i2c_client *cl, const struct i2c_device_id *id) switch (id->driver_data) { case adm1030: - return attach_fan( cl ); + ret = attach_fan(cl); + break; case ds1775: - return attach_thermostat(cl); + ret = attach_thermostat(cl); + break; } - return 0; + + if (!x.running && x.thermostat && x.fan) { + x.running = 1; + x.poll_task = kthread_run(control_loop, NULL, "g4fand"); + } + + return ret; } static struct i2c_driver g4fan_driver = { diff --git a/drivers/macintosh/windfarm_ad7417_sensor.c b/drivers/macintosh/windfarm_ad7417_sensor.c index 125605987b44..e7dec328c7cf 100644 --- a/drivers/macintosh/windfarm_ad7417_sensor.c +++ b/drivers/macintosh/windfarm_ad7417_sensor.c @@ -312,9 +312,16 @@ static const struct i2c_device_id wf_ad7417_id[] = { }; MODULE_DEVICE_TABLE(i2c, wf_ad7417_id); +static const struct of_device_id wf_ad7417_of_id[] = { + { .compatible = "ad7417", }, + { } +}; +MODULE_DEVICE_TABLE(of, wf_ad7417_of_id); + static struct i2c_driver wf_ad7417_driver = { .driver = { .name = "wf_ad7417", + .of_match_table = wf_ad7417_of_id, }, .probe = wf_ad7417_probe, .remove = wf_ad7417_remove, diff --git a/drivers/macintosh/windfarm_fcu_controls.c b/drivers/macintosh/windfarm_fcu_controls.c index 67daeec94b44..2470e5a725c8 100644 --- a/drivers/macintosh/windfarm_fcu_controls.c +++ b/drivers/macintosh/windfarm_fcu_controls.c @@ -580,9 +580,16 @@ static const struct i2c_device_id wf_fcu_id[] = { }; MODULE_DEVICE_TABLE(i2c, wf_fcu_id); +static const struct of_device_id wf_fcu_of_id[] = { + { .compatible = "fcu", }, + { } +}; +MODULE_DEVICE_TABLE(of, wf_fcu_of_id); + static struct i2c_driver wf_fcu_driver = { .driver = { .name = "wf_fcu", + .of_match_table = wf_fcu_of_id, }, .probe = wf_fcu_probe, .remove = wf_fcu_remove, diff --git a/drivers/macintosh/windfarm_lm75_sensor.c b/drivers/macintosh/windfarm_lm75_sensor.c index 282c28a17ea1..1e5fa09845e7 100644 --- a/drivers/macintosh/windfarm_lm75_sensor.c +++ b/drivers/macintosh/windfarm_lm75_sensor.c @@ -14,6 +14,7 @@ #include <linux/init.h> #include <linux/wait.h> #include <linux/i2c.h> +#include <linux/of_device.h> #include <asm/prom.h> #include <asm/machdep.h> #include <asm/io.h> @@ -91,9 +92,14 @@ static int wf_lm75_probe(struct i2c_client *client, const struct i2c_device_id *id) { struct wf_lm75_sensor *lm; - int rc, ds1775 = id->driver_data; + int rc, ds1775; const char *name, *loc; + if (id) + ds1775 = id->driver_data; + else + ds1775 = !!of_device_get_match_data(&client->dev); + DBG("wf_lm75: creating %s device at address 0x%02x\n", ds1775 ? "ds1775" : "lm75", client->addr); @@ -164,9 +170,17 @@ static const struct i2c_device_id wf_lm75_id[] = { }; MODULE_DEVICE_TABLE(i2c, wf_lm75_id); +static const struct of_device_id wf_lm75_of_id[] = { + { .compatible = "lm75", .data = (void *)0}, + { .compatible = "ds1775", .data = (void *)1 }, + { } +}; +MODULE_DEVICE_TABLE(of, wf_lm75_of_id); + static struct i2c_driver wf_lm75_driver = { .driver = { .name = "wf_lm75", + .of_match_table = wf_lm75_of_id, }, .probe = wf_lm75_probe, .remove = wf_lm75_remove, diff --git a/drivers/macintosh/windfarm_lm87_sensor.c b/drivers/macintosh/windfarm_lm87_sensor.c index b03a33b803b7..d011899c0a8a 100644 --- a/drivers/macintosh/windfarm_lm87_sensor.c +++ b/drivers/macintosh/windfarm_lm87_sensor.c @@ -166,9 +166,16 @@ static const struct i2c_device_id wf_lm87_id[] = { }; MODULE_DEVICE_TABLE(i2c, wf_lm87_id); +static const struct of_device_id wf_lm87_of_id[] = { + { .compatible = "lm87cimt", }, + { } +}; +MODULE_DEVICE_TABLE(of, wf_lm87_of_id); + static struct i2c_driver wf_lm87_driver = { .driver = { .name = "wf_lm87", + .of_match_table = wf_lm87_of_id, }, .probe = wf_lm87_probe, .remove = wf_lm87_remove, diff --git a/drivers/macintosh/windfarm_max6690_sensor.c b/drivers/macintosh/windfarm_max6690_sensor.c index e666cc020683..1e7b03d44ad9 100644 --- a/drivers/macintosh/windfarm_max6690_sensor.c +++ b/drivers/macintosh/windfarm_max6690_sensor.c @@ -120,9 +120,16 @@ static const struct i2c_device_id wf_max6690_id[] = { }; MODULE_DEVICE_TABLE(i2c, wf_max6690_id); +static const struct of_device_id wf_max6690_of_id[] = { + { .compatible = "max6690", }, + { } +}; +MODULE_DEVICE_TABLE(of, wf_max6690_of_id); + static struct i2c_driver wf_max6690_driver = { .driver = { .name = "wf_max6690", + .of_match_table = wf_max6690_of_id, }, .probe = wf_max6690_probe, .remove = wf_max6690_remove, diff --git a/drivers/macintosh/windfarm_smu_sat.c b/drivers/macintosh/windfarm_smu_sat.c index c84ec49c3741..cb75dc035616 100644 --- a/drivers/macintosh/windfarm_smu_sat.c +++ b/drivers/macintosh/windfarm_smu_sat.c @@ -341,9 +341,16 @@ static const struct i2c_device_id wf_sat_id[] = { }; MODULE_DEVICE_TABLE(i2c, wf_sat_id); +static const struct of_device_id wf_sat_of_id[] = { + { .compatible = "smu-sat", }, + { } +}; +MODULE_DEVICE_TABLE(of, wf_sat_of_id); + static struct i2c_driver wf_sat_driver = { .driver = { .name = "wf_smu_sat", + .of_match_table = wf_sat_of_id, }, .probe = wf_sat_probe, .remove = wf_sat_remove, diff --git a/drivers/md/bcache/alloc.c b/drivers/md/bcache/alloc.c index 8bc1faf71ff2..a1df0d95151c 100644 --- a/drivers/md/bcache/alloc.c +++ b/drivers/md/bcache/alloc.c @@ -67,7 +67,6 @@ #include <linux/blkdev.h> #include <linux/kthread.h> #include <linux/random.h> -#include <linux/sched/signal.h> #include <trace/events/bcache.h> #define MAX_OPEN_BUCKETS 128 @@ -734,21 +733,8 @@ int bch_open_buckets_alloc(struct cache_set *c) int bch_cache_allocator_start(struct cache *ca) { - struct task_struct *k; - - /* - * In case previous btree check operation occupies too many - * system memory for bcache btree node cache, and the - * registering process is selected by OOM killer. Here just - * ignore the SIGKILL sent by OOM killer if there is, to - * avoid kthread_run() being failed by pending signals. The - * bcache registering process will exit after the registration - * done. - */ - if (signal_pending(current)) - flush_signals(current); - - k = kthread_run(bch_allocator_thread, ca, "bcache_allocator"); + struct task_struct *k = kthread_run(bch_allocator_thread, + ca, "bcache_allocator"); if (IS_ERR(k)) return PTR_ERR(k); diff --git a/drivers/md/bcache/btree.c b/drivers/md/bcache/btree.c index b12186c87f52..fa872df4e770 100644 --- a/drivers/md/bcache/btree.c +++ b/drivers/md/bcache/btree.c @@ -34,7 +34,6 @@ #include <linux/random.h> #include <linux/rcupdate.h> #include <linux/sched/clock.h> -#include <linux/sched/signal.h> #include <linux/rculist.h> #include <linux/delay.h> #include <trace/events/bcache.h> @@ -1914,18 +1913,6 @@ static int bch_gc_thread(void *arg) int bch_gc_thread_start(struct cache_set *c) { - /* - * In case previous btree check operation occupies too many - * system memory for bcache btree node cache, and the - * registering process is selected by OOM killer. Here just - * ignore the SIGKILL sent by OOM killer if there is, to - * avoid kthread_run() being failed by pending signals. The - * bcache registering process will exit after the registration - * done. - */ - if (signal_pending(current)) - flush_signals(current); - c->gc_thread = kthread_run(bch_gc_thread, c, "bcache_gc"); return PTR_ERR_OR_ZERO(c->gc_thread); } diff --git a/drivers/md/dm-bio-record.h b/drivers/md/dm-bio-record.h index c82578af56a5..2ea0360108e1 100644 --- a/drivers/md/dm-bio-record.h +++ b/drivers/md/dm-bio-record.h @@ -20,8 +20,13 @@ struct dm_bio_details { struct gendisk *bi_disk; u8 bi_partno; + int __bi_remaining; unsigned long bi_flags; struct bvec_iter bi_iter; + bio_end_io_t *bi_end_io; +#if defined(CONFIG_BLK_DEV_INTEGRITY) + struct bio_integrity_payload *bi_integrity; +#endif }; static inline void dm_bio_record(struct dm_bio_details *bd, struct bio *bio) @@ -30,6 +35,11 @@ static inline void dm_bio_record(struct dm_bio_details *bd, struct bio *bio) bd->bi_partno = bio->bi_partno; bd->bi_flags = bio->bi_flags; bd->bi_iter = bio->bi_iter; + bd->__bi_remaining = atomic_read(&bio->__bi_remaining); + bd->bi_end_io = bio->bi_end_io; +#if defined(CONFIG_BLK_DEV_INTEGRITY) + bd->bi_integrity = bio_integrity(bio); +#endif } static inline void dm_bio_restore(struct dm_bio_details *bd, struct bio *bio) @@ -38,6 +48,11 @@ static inline void dm_bio_restore(struct dm_bio_details *bd, struct bio *bio) bio->bi_partno = bd->bi_partno; bio->bi_flags = bd->bi_flags; bio->bi_iter = bd->bi_iter; + atomic_set(&bio->__bi_remaining, bd->__bi_remaining); + bio->bi_end_io = bd->bi_end_io; +#if defined(CONFIG_BLK_DEV_INTEGRITY) + bio->bi_integrity = bd->bi_integrity; +#endif } #endif diff --git a/drivers/md/dm-cache-target.c b/drivers/md/dm-cache-target.c index 2d32821b3a5b..d3bb355819a4 100644 --- a/drivers/md/dm-cache-target.c +++ b/drivers/md/dm-cache-target.c @@ -2846,8 +2846,8 @@ static void cache_postsuspend(struct dm_target *ti) prevent_background_work(cache); BUG_ON(atomic_read(&cache->nr_io_migrations)); - cancel_delayed_work(&cache->waker); - flush_workqueue(cache->wq); + cancel_delayed_work_sync(&cache->waker); + drain_workqueue(cache->wq); WARN_ON(cache->tracker.in_flight); /* @@ -3492,7 +3492,7 @@ static void cache_io_hints(struct dm_target *ti, struct queue_limits *limits) static struct target_type cache_target = { .name = "cache", - .version = {2, 1, 0}, + .version = {2, 2, 0}, .module = THIS_MODULE, .ctr = cache_ctr, .dtr = cache_dtr, diff --git a/drivers/md/dm-integrity.c b/drivers/md/dm-integrity.c index b225b3e445fa..2f03fecd312d 100644 --- a/drivers/md/dm-integrity.c +++ b/drivers/md/dm-integrity.c @@ -6,6 +6,8 @@ * This file is released under the GPL. */ +#include "dm-bio-record.h" + #include <linux/compiler.h> #include <linux/module.h> #include <linux/device-mapper.h> @@ -201,17 +203,19 @@ struct dm_integrity_c { __u8 log2_blocks_per_bitmap_bit; unsigned char mode; - int suspending; int failed; struct crypto_shash *internal_hash; + struct dm_target *ti; + /* these variables are locked with endio_wait.lock */ struct rb_root in_progress; struct list_head wait_list; wait_queue_head_t endio_wait; struct workqueue_struct *wait_wq; + struct workqueue_struct *offload_wq; unsigned char commit_seq; commit_id_t commit_ids[N_COMMIT_IDS]; @@ -293,11 +297,7 @@ struct dm_integrity_io { struct completion *completion; - struct gendisk *orig_bi_disk; - u8 orig_bi_partno; - bio_end_io_t *orig_bi_end_io; - struct bio_integrity_payload *orig_bi_integrity; - struct bvec_iter orig_bi_iter; + struct dm_bio_details bio_details; }; struct journal_completion { @@ -1439,7 +1439,7 @@ static void dec_in_flight(struct dm_integrity_io *dio) dio->range.logical_sector += dio->range.n_sectors; bio_advance(bio, dio->range.n_sectors << SECTOR_SHIFT); INIT_WORK(&dio->work, integrity_bio_wait); - queue_work(ic->wait_wq, &dio->work); + queue_work(ic->offload_wq, &dio->work); return; } do_endio_flush(ic, dio); @@ -1450,14 +1450,9 @@ static void integrity_end_io(struct bio *bio) { struct dm_integrity_io *dio = dm_per_bio_data(bio, sizeof(struct dm_integrity_io)); - bio->bi_iter = dio->orig_bi_iter; - bio->bi_disk = dio->orig_bi_disk; - bio->bi_partno = dio->orig_bi_partno; - if (dio->orig_bi_integrity) { - bio->bi_integrity = dio->orig_bi_integrity; + dm_bio_restore(&dio->bio_details, bio); + if (bio->bi_integrity) bio->bi_opf |= REQ_INTEGRITY; - } - bio->bi_end_io = dio->orig_bi_end_io; if (dio->completion) complete(dio->completion); @@ -1542,7 +1537,7 @@ static void integrity_metadata(struct work_struct *w) } } - __bio_for_each_segment(bv, bio, iter, dio->orig_bi_iter) { + __bio_for_each_segment(bv, bio, iter, dio->bio_details.bi_iter) { unsigned pos; char *mem, *checksums_ptr; @@ -1586,7 +1581,7 @@ again: if (likely(checksums != checksums_onstack)) kfree(checksums); } else { - struct bio_integrity_payload *bip = dio->orig_bi_integrity; + struct bio_integrity_payload *bip = dio->bio_details.bi_integrity; if (bip) { struct bio_vec biv; @@ -1865,7 +1860,7 @@ static void dm_integrity_map_continue(struct dm_integrity_io *dio, bool from_map if (need_sync_io && from_map) { INIT_WORK(&dio->work, integrity_bio_wait); - queue_work(ic->metadata_wq, &dio->work); + queue_work(ic->offload_wq, &dio->work); return; } @@ -2005,20 +2000,13 @@ offload_to_thread: } else dio->completion = NULL; - dio->orig_bi_iter = bio->bi_iter; - - dio->orig_bi_disk = bio->bi_disk; - dio->orig_bi_partno = bio->bi_partno; + dm_bio_record(&dio->bio_details, bio); bio_set_dev(bio, ic->dev->bdev); - - dio->orig_bi_integrity = bio_integrity(bio); bio->bi_integrity = NULL; bio->bi_opf &= ~REQ_INTEGRITY; - - dio->orig_bi_end_io = bio->bi_end_io; bio->bi_end_io = integrity_end_io; - bio->bi_iter.bi_size = dio->range.n_sectors << SECTOR_SHIFT; + generic_make_request(bio); if (need_sync_io) { @@ -2315,7 +2303,7 @@ static void integrity_writer(struct work_struct *w) unsigned prev_free_sectors; /* the following test is not needed, but it tests the replay code */ - if (READ_ONCE(ic->suspending) && !ic->meta_dev) + if (unlikely(dm_suspended(ic->ti)) && !ic->meta_dev) return; spin_lock_irq(&ic->endio_wait.lock); @@ -2376,7 +2364,7 @@ static void integrity_recalc(struct work_struct *w) next_chunk: - if (unlikely(READ_ONCE(ic->suspending))) + if (unlikely(dm_suspended(ic->ti))) goto unlock_ret; range.logical_sector = le64_to_cpu(ic->sb->recalc_sector); @@ -2501,7 +2489,7 @@ static void bitmap_block_work(struct work_struct *w) dio->range.n_sectors, BITMAP_OP_TEST_ALL_SET)) { remove_range(ic, &dio->range); INIT_WORK(&dio->work, integrity_bio_wait); - queue_work(ic->wait_wq, &dio->work); + queue_work(ic->offload_wq, &dio->work); } else { block_bitmap_op(ic, ic->journal, dio->range.logical_sector, dio->range.n_sectors, BITMAP_OP_SET); @@ -2524,7 +2512,7 @@ static void bitmap_block_work(struct work_struct *w) remove_range(ic, &dio->range); INIT_WORK(&dio->work, integrity_bio_wait); - queue_work(ic->wait_wq, &dio->work); + queue_work(ic->offload_wq, &dio->work); } queue_delayed_work(ic->commit_wq, &ic->bitmap_flush_work, ic->bitmap_flush_interval); @@ -2804,8 +2792,6 @@ static void dm_integrity_postsuspend(struct dm_target *ti) del_timer_sync(&ic->autocommit_timer); - WRITE_ONCE(ic->suspending, 1); - if (ic->recalc_wq) drain_workqueue(ic->recalc_wq); @@ -2834,8 +2820,6 @@ static void dm_integrity_postsuspend(struct dm_target *ti) #endif } - WRITE_ONCE(ic->suspending, 0); - BUG_ON(!RB_EMPTY_ROOT(&ic->in_progress)); ic->journal_uptodate = true; @@ -2888,17 +2872,24 @@ static void dm_integrity_resume(struct dm_target *ti) } else { replay_journal(ic); if (ic->mode == 'B') { - int mode; ic->sb->flags |= cpu_to_le32(SB_FLAG_DIRTY_BITMAP); ic->sb->log2_blocks_per_bitmap_bit = ic->log2_blocks_per_bitmap_bit; r = sync_rw_sb(ic, REQ_OP_WRITE, REQ_FUA); if (unlikely(r)) dm_integrity_io_error(ic, "writing superblock", r); - mode = ic->recalculate_flag ? BITMAP_OP_SET : BITMAP_OP_CLEAR; - block_bitmap_op(ic, ic->journal, 0, ic->provided_data_sectors, mode); - block_bitmap_op(ic, ic->recalc_bitmap, 0, ic->provided_data_sectors, mode); - block_bitmap_op(ic, ic->may_write_bitmap, 0, ic->provided_data_sectors, mode); + block_bitmap_op(ic, ic->journal, 0, ic->provided_data_sectors, BITMAP_OP_CLEAR); + block_bitmap_op(ic, ic->recalc_bitmap, 0, ic->provided_data_sectors, BITMAP_OP_CLEAR); + block_bitmap_op(ic, ic->may_write_bitmap, 0, ic->provided_data_sectors, BITMAP_OP_CLEAR); + if (ic->sb->flags & cpu_to_le32(SB_FLAG_RECALCULATING) && + le64_to_cpu(ic->sb->recalc_sector) < ic->provided_data_sectors) { + block_bitmap_op(ic, ic->journal, le64_to_cpu(ic->sb->recalc_sector), + ic->provided_data_sectors - le64_to_cpu(ic->sb->recalc_sector), BITMAP_OP_SET); + block_bitmap_op(ic, ic->recalc_bitmap, le64_to_cpu(ic->sb->recalc_sector), + ic->provided_data_sectors - le64_to_cpu(ic->sb->recalc_sector), BITMAP_OP_SET); + block_bitmap_op(ic, ic->may_write_bitmap, le64_to_cpu(ic->sb->recalc_sector), + ic->provided_data_sectors - le64_to_cpu(ic->sb->recalc_sector), BITMAP_OP_SET); + } rw_journal_sectors(ic, REQ_OP_WRITE, REQ_FUA | REQ_SYNC, 0, ic->n_bitmap_blocks * (BITMAP_BLOCK_SIZE >> SECTOR_SHIFT), NULL); } @@ -2967,7 +2958,7 @@ static void dm_integrity_status(struct dm_target *ti, status_type_t type, DMEMIT(" meta_device:%s", ic->meta_dev->name); if (ic->sectors_per_block != 1) DMEMIT(" block_size:%u", ic->sectors_per_block << SECTOR_SHIFT); - if (ic->recalculate_flag) + if (ic->sb->flags & cpu_to_le32(SB_FLAG_RECALCULATING)) DMEMIT(" recalculate"); DMEMIT(" journal_sectors:%u", ic->initial_sectors - SB_SECTORS); DMEMIT(" interleave_sectors:%u", 1U << ic->sb->log2_interleave_sectors); @@ -3623,6 +3614,7 @@ static int dm_integrity_ctr(struct dm_target *ti, unsigned argc, char **argv) } ti->private = ic; ti->per_io_data_size = sizeof(struct dm_integrity_io); + ic->ti = ti; ic->in_progress = RB_ROOT; INIT_LIST_HEAD(&ic->wait_list); @@ -3836,6 +3828,14 @@ static int dm_integrity_ctr(struct dm_target *ti, unsigned argc, char **argv) goto bad; } + ic->offload_wq = alloc_workqueue("dm-integrity-offload", WQ_MEM_RECLAIM, + METADATA_WORKQUEUE_MAX_ACTIVE); + if (!ic->offload_wq) { + ti->error = "Cannot allocate workqueue"; + r = -ENOMEM; + goto bad; + } + ic->commit_wq = alloc_workqueue("dm-integrity-commit", WQ_MEM_RECLAIM, 1); if (!ic->commit_wq) { ti->error = "Cannot allocate workqueue"; @@ -4140,6 +4140,8 @@ static void dm_integrity_dtr(struct dm_target *ti) destroy_workqueue(ic->metadata_wq); if (ic->wait_wq) destroy_workqueue(ic->wait_wq); + if (ic->offload_wq) + destroy_workqueue(ic->offload_wq); if (ic->commit_wq) destroy_workqueue(ic->commit_wq); if (ic->writer_wq) @@ -4200,7 +4202,7 @@ static void dm_integrity_dtr(struct dm_target *ti) static struct target_type integrity_target = { .name = "integrity", - .version = {1, 4, 0}, + .version = {1, 5, 0}, .module = THIS_MODULE, .features = DM_TARGET_SINGLETON | DM_TARGET_INTEGRITY, .ctr = dm_integrity_ctr, diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c index 2bc18c9c3abc..58fd137b6ae1 100644 --- a/drivers/md/dm-mpath.c +++ b/drivers/md/dm-mpath.c @@ -2053,7 +2053,7 @@ static int multipath_busy(struct dm_target *ti) *---------------------------------------------------------------*/ static struct target_type multipath_target = { .name = "multipath", - .version = {1, 13, 0}, + .version = {1, 14, 0}, .features = DM_TARGET_SINGLETON | DM_TARGET_IMMUTABLE | DM_TARGET_PASSES_INTEGRITY, .module = THIS_MODULE, diff --git a/drivers/md/dm-thin-metadata.c b/drivers/md/dm-thin-metadata.c index fc9947d6210c..76b6b323bf4b 100644 --- a/drivers/md/dm-thin-metadata.c +++ b/drivers/md/dm-thin-metadata.c @@ -960,9 +960,9 @@ int dm_pool_metadata_close(struct dm_pool_metadata *pmd) DMWARN("%s: __commit_transaction() failed, error = %d", __func__, r); } + pmd_write_unlock(pmd); if (!pmd->fail_io) __destroy_persistent_data_objects(pmd); - pmd_write_unlock(pmd); kfree(pmd); return 0; diff --git a/drivers/md/dm-verity-target.c b/drivers/md/dm-verity-target.c index 0d61e9c67986..eec9f252e935 100644 --- a/drivers/md/dm-verity-target.c +++ b/drivers/md/dm-verity-target.c @@ -1221,7 +1221,7 @@ bad: static struct target_type verity_target = { .name = "verity", - .version = {1, 5, 0}, + .version = {1, 6, 0}, .module = THIS_MODULE, .ctr = verity_ctr, .dtr = verity_dtr, diff --git a/drivers/md/dm-writecache.c b/drivers/md/dm-writecache.c index b9e27e37a943..a09bdc000e64 100644 --- a/drivers/md/dm-writecache.c +++ b/drivers/md/dm-writecache.c @@ -625,6 +625,12 @@ static void writecache_add_to_freelist(struct dm_writecache *wc, struct wc_entry wc->freelist_size++; } +static inline void writecache_verify_watermark(struct dm_writecache *wc) +{ + if (unlikely(wc->freelist_size + wc->writeback_size <= wc->freelist_high_watermark)) + queue_work(wc->writeback_wq, &wc->writeback_work); +} + static struct wc_entry *writecache_pop_from_freelist(struct dm_writecache *wc, sector_t expected_sector) { struct wc_entry *e; @@ -650,8 +656,8 @@ static struct wc_entry *writecache_pop_from_freelist(struct dm_writecache *wc, s list_del(&e->lru); } wc->freelist_size--; - if (unlikely(wc->freelist_size + wc->writeback_size <= wc->freelist_high_watermark)) - queue_work(wc->writeback_wq, &wc->writeback_work); + + writecache_verify_watermark(wc); return e; } @@ -842,7 +848,7 @@ static void writecache_suspend(struct dm_target *ti) } wc_unlock(wc); - flush_workqueue(wc->writeback_wq); + drain_workqueue(wc->writeback_wq); wc_lock(wc); if (flush_on_suspend) @@ -965,6 +971,8 @@ erase_this: writecache_commit_flushed(wc, false); } + writecache_verify_watermark(wc); + wc_unlock(wc); } @@ -2312,7 +2320,7 @@ static void writecache_status(struct dm_target *ti, status_type_t type, static struct target_type writecache_target = { .name = "writecache", - .version = {1, 1, 1}, + .version = {1, 2, 0}, .module = THIS_MODULE, .ctr = writecache_ctr, .dtr = writecache_dtr, diff --git a/drivers/md/dm-zoned-target.c b/drivers/md/dm-zoned-target.c index 70a1063161c0..f4f83d39b3dc 100644 --- a/drivers/md/dm-zoned-target.c +++ b/drivers/md/dm-zoned-target.c @@ -533,8 +533,9 @@ static int dmz_queue_chunk_work(struct dmz_target *dmz, struct bio *bio) /* Get the BIO chunk work. If one is not active yet, create one */ cw = radix_tree_lookup(&dmz->chunk_rxtree, chunk); - if (!cw) { - + if (cw) { + dmz_get_chunk_work(cw); + } else { /* Create a new chunk work */ cw = kmalloc(sizeof(struct dm_chunk_work), GFP_NOIO); if (unlikely(!cw)) { @@ -543,7 +544,7 @@ static int dmz_queue_chunk_work(struct dmz_target *dmz, struct bio *bio) } INIT_WORK(&cw->work, dmz_chunk_work); - refcount_set(&cw->refcount, 0); + refcount_set(&cw->refcount, 1); cw->target = dmz; cw->chunk = chunk; bio_list_init(&cw->bio_list); @@ -556,7 +557,6 @@ static int dmz_queue_chunk_work(struct dmz_target *dmz, struct bio *bio) } bio_list_add(&cw->bio_list, bio); - dmz_get_chunk_work(cw); dmz_reclaim_bio_acc(dmz->reclaim); if (queue_work(dmz->chunk_wq, &cw->work)) @@ -967,7 +967,7 @@ static int dmz_iterate_devices(struct dm_target *ti, static struct target_type dmz_type = { .name = "zoned", - .version = {1, 0, 0}, + .version = {1, 1, 0}, .features = DM_TARGET_SINGLETON | DM_TARGET_ZONED_HM, .module = THIS_MODULE, .ctr = dmz_ctr, diff --git a/drivers/md/dm.c b/drivers/md/dm.c index b89f07ee2eff..0413018c8305 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -1788,7 +1788,8 @@ static int dm_any_congested(void *congested_data, int bdi_bits) * With request-based DM we only need to check the * top-level queue for congestion. */ - r = md->queue->backing_dev_info->wb.state & bdi_bits; + struct backing_dev_info *bdi = md->queue->backing_dev_info; + r = bdi->wb.congested->state & bdi_bits; } else { map = dm_get_live_table_fast(md); if (map) @@ -1854,15 +1855,6 @@ static const struct dax_operations dm_dax_ops; static void dm_wq_work(struct work_struct *work); -static void dm_init_normal_md_queue(struct mapped_device *md) -{ - /* - * Initialize aspects of queue that aren't relevant for blk-mq - */ - md->queue->backing_dev_info->congested_data = md; - md->queue->backing_dev_info->congested_fn = dm_any_congested; -} - static void cleanup_mapped_device(struct mapped_device *md) { if (md->wq) @@ -2249,6 +2241,12 @@ struct queue_limits *dm_get_queue_limits(struct mapped_device *md) } EXPORT_SYMBOL_GPL(dm_get_queue_limits); +static void dm_init_congested_fn(struct mapped_device *md) +{ + md->queue->backing_dev_info->congested_data = md; + md->queue->backing_dev_info->congested_fn = dm_any_congested; +} + /* * Setup the DM device's queue based on md's type */ @@ -2265,11 +2263,12 @@ int dm_setup_md_queue(struct mapped_device *md, struct dm_table *t) DMERR("Cannot initialize queue for request-based dm-mq mapped device"); return r; } + dm_init_congested_fn(md); break; case DM_TYPE_BIO_BASED: case DM_TYPE_DAX_BIO_BASED: case DM_TYPE_NVME_BIO_BASED: - dm_init_normal_md_queue(md); + dm_init_congested_fn(md); break; case DM_TYPE_NONE: WARN_ON_ONCE(true); @@ -2368,6 +2367,7 @@ static void __dm_destroy(struct mapped_device *md, bool wait) map = dm_get_live_table(md, &srcu_idx); if (!dm_suspended_md(md)) { dm_table_presuspend_targets(map); + set_bit(DMF_SUSPENDED, &md->flags); dm_table_postsuspend_targets(map); } /* dm_put_live_table must be before msleep, otherwise deadlock is possible */ diff --git a/drivers/media/mc/mc-entity.c b/drivers/media/mc/mc-entity.c index 7c429ce98bae..668770e9f609 100644 --- a/drivers/media/mc/mc-entity.c +++ b/drivers/media/mc/mc-entity.c @@ -639,9 +639,9 @@ int media_get_pad_index(struct media_entity *entity, bool is_sink, return -EINVAL; for (i = 0; i < entity->num_pads; i++) { - if (entity->pads[i].flags == MEDIA_PAD_FL_SINK) + if (entity->pads[i].flags & MEDIA_PAD_FL_SINK) pad_is_sink = true; - else if (entity->pads[i].flags == MEDIA_PAD_FL_SOURCE) + else if (entity->pads[i].flags & MEDIA_PAD_FL_SOURCE) pad_is_sink = false; else continue; /* This is an error! */ diff --git a/drivers/media/platform/vicodec/codec-v4l2-fwht.c b/drivers/media/platform/vicodec/codec-v4l2-fwht.c index 3c93d9232c3c..b6e39fbd8ad5 100644 --- a/drivers/media/platform/vicodec/codec-v4l2-fwht.c +++ b/drivers/media/platform/vicodec/codec-v4l2-fwht.c @@ -27,17 +27,17 @@ static const struct v4l2_fwht_pixfmt_info v4l2_fwht_pixfmts[] = { { V4L2_PIX_FMT_BGR24, 3, 3, 1, 3, 3, 1, 1, 3, 1, FWHT_FL_PIXENC_RGB}, { V4L2_PIX_FMT_RGB24, 3, 3, 1, 3, 3, 1, 1, 3, 1, FWHT_FL_PIXENC_RGB}, { V4L2_PIX_FMT_HSV24, 3, 3, 1, 3, 3, 1, 1, 3, 1, FWHT_FL_PIXENC_HSV}, - { V4L2_PIX_FMT_BGR32, 4, 4, 1, 4, 4, 1, 1, 3, 1, FWHT_FL_PIXENC_RGB}, - { V4L2_PIX_FMT_XBGR32, 4, 4, 1, 4, 4, 1, 1, 3, 1, FWHT_FL_PIXENC_RGB}, + { V4L2_PIX_FMT_BGR32, 4, 4, 1, 4, 4, 1, 1, 4, 1, FWHT_FL_PIXENC_RGB}, + { V4L2_PIX_FMT_XBGR32, 4, 4, 1, 4, 4, 1, 1, 4, 1, FWHT_FL_PIXENC_RGB}, { V4L2_PIX_FMT_ABGR32, 4, 4, 1, 4, 4, 1, 1, 4, 1, FWHT_FL_PIXENC_RGB}, - { V4L2_PIX_FMT_RGB32, 4, 4, 1, 4, 4, 1, 1, 3, 1, FWHT_FL_PIXENC_RGB}, - { V4L2_PIX_FMT_XRGB32, 4, 4, 1, 4, 4, 1, 1, 3, 1, FWHT_FL_PIXENC_RGB}, + { V4L2_PIX_FMT_RGB32, 4, 4, 1, 4, 4, 1, 1, 4, 1, FWHT_FL_PIXENC_RGB}, + { V4L2_PIX_FMT_XRGB32, 4, 4, 1, 4, 4, 1, 1, 4, 1, FWHT_FL_PIXENC_RGB}, { V4L2_PIX_FMT_ARGB32, 4, 4, 1, 4, 4, 1, 1, 4, 1, FWHT_FL_PIXENC_RGB}, - { V4L2_PIX_FMT_BGRX32, 4, 4, 1, 4, 4, 1, 1, 3, 1, FWHT_FL_PIXENC_RGB}, + { V4L2_PIX_FMT_BGRX32, 4, 4, 1, 4, 4, 1, 1, 4, 1, FWHT_FL_PIXENC_RGB}, { V4L2_PIX_FMT_BGRA32, 4, 4, 1, 4, 4, 1, 1, 4, 1, FWHT_FL_PIXENC_RGB}, - { V4L2_PIX_FMT_RGBX32, 4, 4, 1, 4, 4, 1, 1, 3, 1, FWHT_FL_PIXENC_RGB}, + { V4L2_PIX_FMT_RGBX32, 4, 4, 1, 4, 4, 1, 1, 4, 1, FWHT_FL_PIXENC_RGB}, { V4L2_PIX_FMT_RGBA32, 4, 4, 1, 4, 4, 1, 1, 4, 1, FWHT_FL_PIXENC_RGB}, - { V4L2_PIX_FMT_HSV32, 4, 4, 1, 4, 4, 1, 1, 3, 1, FWHT_FL_PIXENC_HSV}, + { V4L2_PIX_FMT_HSV32, 4, 4, 1, 4, 4, 1, 1, 4, 1, FWHT_FL_PIXENC_HSV}, { V4L2_PIX_FMT_GREY, 1, 1, 1, 1, 0, 1, 1, 1, 1, FWHT_FL_PIXENC_RGB}, }; @@ -175,22 +175,14 @@ static int prepare_raw_frame(struct fwht_raw_frame *rf, case V4L2_PIX_FMT_RGB32: case V4L2_PIX_FMT_XRGB32: case V4L2_PIX_FMT_HSV32: - rf->cr = rf->luma + 1; - rf->cb = rf->cr + 2; - rf->luma += 2; - break; - case V4L2_PIX_FMT_BGR32: - case V4L2_PIX_FMT_XBGR32: - rf->cb = rf->luma; - rf->cr = rf->cb + 2; - rf->luma++; - break; case V4L2_PIX_FMT_ARGB32: rf->alpha = rf->luma; rf->cr = rf->luma + 1; rf->cb = rf->cr + 2; rf->luma += 2; break; + case V4L2_PIX_FMT_BGR32: + case V4L2_PIX_FMT_XBGR32: case V4L2_PIX_FMT_ABGR32: rf->cb = rf->luma; rf->cr = rf->cb + 2; @@ -198,10 +190,6 @@ static int prepare_raw_frame(struct fwht_raw_frame *rf, rf->alpha = rf->cr + 1; break; case V4L2_PIX_FMT_BGRX32: - rf->cb = rf->luma + 1; - rf->cr = rf->cb + 2; - rf->luma += 2; - break; case V4L2_PIX_FMT_BGRA32: rf->alpha = rf->luma; rf->cb = rf->luma + 1; @@ -209,10 +197,6 @@ static int prepare_raw_frame(struct fwht_raw_frame *rf, rf->luma += 2; break; case V4L2_PIX_FMT_RGBX32: - rf->cr = rf->luma; - rf->cb = rf->cr + 2; - rf->luma++; - break; case V4L2_PIX_FMT_RGBA32: rf->alpha = rf->luma + 3; rf->cr = rf->luma; diff --git a/drivers/media/usb/pulse8-cec/pulse8-cec.c b/drivers/media/usb/pulse8-cec/pulse8-cec.c index afda438d4e0a..0655aa9ecf28 100644 --- a/drivers/media/usb/pulse8-cec/pulse8-cec.c +++ b/drivers/media/usb/pulse8-cec/pulse8-cec.c @@ -635,8 +635,6 @@ static void pulse8_cec_adap_free(struct cec_adapter *adap) cancel_delayed_work_sync(&pulse8->ping_eeprom_work); cancel_work_sync(&pulse8->irq_work); cancel_work_sync(&pulse8->tx_work); - serio_close(pulse8->serio); - serio_set_drvdata(pulse8->serio, NULL); kfree(pulse8); } @@ -652,6 +650,9 @@ static void pulse8_disconnect(struct serio *serio) struct pulse8 *pulse8 = serio_get_drvdata(serio); cec_unregister_adapter(pulse8->adap); + pulse8->serio = NULL; + serio_set_drvdata(serio, NULL); + serio_close(serio); } static int pulse8_setup(struct pulse8 *pulse8, struct serio *serio, @@ -840,6 +841,8 @@ static int pulse8_connect(struct serio *serio, struct serio_driver *drv) serio_set_drvdata(serio, pulse8); INIT_WORK(&pulse8->irq_work, pulse8_irq_work_handler); INIT_WORK(&pulse8->tx_work, pulse8_tx_work_handler); + INIT_DELAYED_WORK(&pulse8->ping_eeprom_work, + pulse8_ping_eeprom_work_handler); mutex_init(&pulse8->lock); spin_lock_init(&pulse8->msg_lock); pulse8->config_pending = false; @@ -865,17 +868,16 @@ static int pulse8_connect(struct serio *serio, struct serio_driver *drv) pulse8->restoring_config = true; } - INIT_DELAYED_WORK(&pulse8->ping_eeprom_work, - pulse8_ping_eeprom_work_handler); schedule_delayed_work(&pulse8->ping_eeprom_work, PING_PERIOD); return 0; close_serio: + pulse8->serio = NULL; + serio_set_drvdata(serio, NULL); serio_close(serio); delete_adap: cec_delete_adapter(pulse8->adap); - serio_set_drvdata(serio, NULL); free_device: kfree(pulse8); return err; diff --git a/drivers/media/v4l2-core/v4l2-mem2mem.c b/drivers/media/v4l2-core/v4l2-mem2mem.c index 1afd9c6ad908..cc34c5ab7009 100644 --- a/drivers/media/v4l2-core/v4l2-mem2mem.c +++ b/drivers/media/v4l2-core/v4l2-mem2mem.c @@ -880,12 +880,12 @@ int v4l2_m2m_register_media_controller(struct v4l2_m2m_dev *m2m_dev, goto err_rel_entity1; /* Connect the three entities */ - ret = media_create_pad_link(m2m_dev->source, 0, &m2m_dev->proc, 1, + ret = media_create_pad_link(m2m_dev->source, 0, &m2m_dev->proc, 0, MEDIA_LNK_FL_IMMUTABLE | MEDIA_LNK_FL_ENABLED); if (ret) goto err_rel_entity2; - ret = media_create_pad_link(&m2m_dev->proc, 0, &m2m_dev->sink, 0, + ret = media_create_pad_link(&m2m_dev->proc, 1, &m2m_dev->sink, 0, MEDIA_LNK_FL_IMMUTABLE | MEDIA_LNK_FL_ENABLED); if (ret) goto err_rm_links0; diff --git a/drivers/misc/altera-stapl/altera.c b/drivers/misc/altera-stapl/altera.c index 25e5f24b3fec..5bdf57472314 100644 --- a/drivers/misc/altera-stapl/altera.c +++ b/drivers/misc/altera-stapl/altera.c @@ -2112,8 +2112,8 @@ exit_done: return status; } -static int altera_get_note(u8 *p, s32 program_size, - s32 *offset, char *key, char *value, int length) +static int altera_get_note(u8 *p, s32 program_size, s32 *offset, + char *key, char *value, int keylen, int vallen) /* * Gets key and value of NOTE fields in the JBC file. * Can be called in two modes: if offset pointer is NULL, @@ -2170,7 +2170,7 @@ static int altera_get_note(u8 *p, s32 program_size, &p[note_table + (8 * i) + 4])]; if (value != NULL) - strlcpy(value, value_ptr, length); + strlcpy(value, value_ptr, vallen); } } @@ -2189,13 +2189,13 @@ static int altera_get_note(u8 *p, s32 program_size, strlcpy(key, &p[note_strings + get_unaligned_be32( &p[note_table + (8 * i)])], - length); + keylen); if (value != NULL) strlcpy(value, &p[note_strings + get_unaligned_be32( &p[note_table + (8 * i) + 4])], - length); + vallen); *offset = i + 1; } @@ -2449,7 +2449,7 @@ int altera_init(struct altera_config *config, const struct firmware *fw) __func__, (format_version == 2) ? "Jam STAPL" : "pre-standardized Jam 1.1"); while (altera_get_note((u8 *)fw->data, fw->size, - &offset, key, value, 256) == 0) + &offset, key, value, 32, 256) == 0) printk(KERN_INFO "%s: NOTE \"%s\" = \"%s\"\n", __func__, key, value); } diff --git a/drivers/misc/cardreader/rts5227.c b/drivers/misc/cardreader/rts5227.c index 4feed296a327..423fecc19fc4 100644 --- a/drivers/misc/cardreader/rts5227.c +++ b/drivers/misc/cardreader/rts5227.c @@ -394,7 +394,7 @@ static const struct pcr_ops rts522a_pcr_ops = { void rts522a_init_params(struct rtsx_pcr *pcr) { rts5227_init_params(pcr); - + pcr->tx_initial_phase = SET_CLOCK_PHASE(20, 20, 11); pcr->reg_pm_ctrl3 = RTS522A_PM_CTRL3; pcr->option.ocp_en = 1; diff --git a/drivers/misc/cardreader/rts5249.c b/drivers/misc/cardreader/rts5249.c index db936e4d6e56..1a81cda948c1 100644 --- a/drivers/misc/cardreader/rts5249.c +++ b/drivers/misc/cardreader/rts5249.c @@ -618,6 +618,7 @@ static const struct pcr_ops rts524a_pcr_ops = { void rts524a_init_params(struct rtsx_pcr *pcr) { rts5249_init_params(pcr); + pcr->tx_initial_phase = SET_CLOCK_PHASE(27, 29, 11); pcr->option.ltr_l1off_sspwrgate = LTR_L1OFF_SSPWRGATE_5250_DEF; pcr->option.ltr_l1off_snooze_sspwrgate = LTR_L1OFF_SNOOZE_SSPWRGATE_5250_DEF; @@ -733,6 +734,7 @@ static const struct pcr_ops rts525a_pcr_ops = { void rts525a_init_params(struct rtsx_pcr *pcr) { rts5249_init_params(pcr); + pcr->tx_initial_phase = SET_CLOCK_PHASE(25, 29, 11); pcr->option.ltr_l1off_sspwrgate = LTR_L1OFF_SSPWRGATE_5250_DEF; pcr->option.ltr_l1off_snooze_sspwrgate = LTR_L1OFF_SNOOZE_SSPWRGATE_5250_DEF; diff --git a/drivers/misc/cardreader/rts5260.c b/drivers/misc/cardreader/rts5260.c index 4214f02a17fd..711054ebad74 100644 --- a/drivers/misc/cardreader/rts5260.c +++ b/drivers/misc/cardreader/rts5260.c @@ -662,7 +662,7 @@ void rts5260_init_params(struct rtsx_pcr *pcr) pcr->sd30_drive_sel_1v8 = CFG_DRIVER_TYPE_B; pcr->sd30_drive_sel_3v3 = CFG_DRIVER_TYPE_B; pcr->aspm_en = ASPM_L1_EN; - pcr->tx_initial_phase = SET_CLOCK_PHASE(1, 29, 16); + pcr->tx_initial_phase = SET_CLOCK_PHASE(27, 29, 11); pcr->rx_initial_phase = SET_CLOCK_PHASE(24, 6, 5); pcr->ic_version = rts5260_get_ic_version(pcr); diff --git a/drivers/misc/cardreader/rts5261.c b/drivers/misc/cardreader/rts5261.c index bc4967a6efa1..78c3b1d424c3 100644 --- a/drivers/misc/cardreader/rts5261.c +++ b/drivers/misc/cardreader/rts5261.c @@ -764,7 +764,7 @@ void rts5261_init_params(struct rtsx_pcr *pcr) pcr->sd30_drive_sel_1v8 = CFG_DRIVER_TYPE_B; pcr->sd30_drive_sel_3v3 = CFG_DRIVER_TYPE_B; pcr->aspm_en = ASPM_L1_EN; - pcr->tx_initial_phase = SET_CLOCK_PHASE(20, 27, 16); + pcr->tx_initial_phase = SET_CLOCK_PHASE(27, 27, 11); pcr->rx_initial_phase = SET_CLOCK_PHASE(24, 6, 5); pcr->ic_version = rts5261_get_ic_version(pcr); diff --git a/drivers/misc/eeprom/at24.c b/drivers/misc/eeprom/at24.c index 031eb64549af..282c9ef68ed2 100644 --- a/drivers/misc/eeprom/at24.c +++ b/drivers/misc/eeprom/at24.c @@ -712,13 +712,14 @@ static int at24_probe(struct i2c_client *client) * chip is functional. */ err = at24_read(at24, 0, &test_byte, 1); - pm_runtime_idle(dev); if (err) { pm_runtime_disable(dev); regulator_disable(at24->vcc_reg); return -ENODEV; } + pm_runtime_idle(dev); + if (writable) dev_info(dev, "%u byte %s EEPROM, writable, %u bytes/write\n", byte_len, client->name, at24->write_max); diff --git a/drivers/misc/habanalabs/device.c b/drivers/misc/habanalabs/device.c index b155e9549076..b680b0caa69b 100644 --- a/drivers/misc/habanalabs/device.c +++ b/drivers/misc/habanalabs/device.c @@ -598,7 +598,9 @@ int hl_device_set_debug_mode(struct hl_device *hdev, bool enable) goto out; } - hdev->asic_funcs->halt_coresight(hdev); + if (!hdev->hard_reset_pending) + hdev->asic_funcs->halt_coresight(hdev); + hdev->in_debug = 0; goto out; @@ -1189,6 +1191,7 @@ int hl_device_init(struct hl_device *hdev, struct class *hclass) if (hdev->asic_funcs->get_hw_state(hdev) == HL_DEVICE_HW_STATE_DIRTY) { dev_info(hdev->dev, "H/W state is dirty, must reset before initializing\n"); + hdev->asic_funcs->halt_engines(hdev, true); hdev->asic_funcs->hw_fini(hdev, true); } diff --git a/drivers/misc/habanalabs/goya/goya.c b/drivers/misc/habanalabs/goya/goya.c index 7344e8a222ae..b8a8de24aaf7 100644 --- a/drivers/misc/habanalabs/goya/goya.c +++ b/drivers/misc/habanalabs/goya/goya.c @@ -895,6 +895,11 @@ void goya_init_dma_qmans(struct hl_device *hdev) */ static void goya_disable_external_queues(struct hl_device *hdev) { + struct goya_device *goya = hdev->asic_specific; + + if (!(goya->hw_cap_initialized & HW_CAP_DMA)) + return; + WREG32(mmDMA_QM_0_GLBL_CFG0, 0); WREG32(mmDMA_QM_1_GLBL_CFG0, 0); WREG32(mmDMA_QM_2_GLBL_CFG0, 0); @@ -956,6 +961,11 @@ static int goya_stop_external_queues(struct hl_device *hdev) { int rc, retval = 0; + struct goya_device *goya = hdev->asic_specific; + + if (!(goya->hw_cap_initialized & HW_CAP_DMA)) + return retval; + rc = goya_stop_queue(hdev, mmDMA_QM_0_GLBL_CFG1, mmDMA_QM_0_CP_STS, @@ -1744,9 +1754,18 @@ void goya_init_tpc_qmans(struct hl_device *hdev) */ static void goya_disable_internal_queues(struct hl_device *hdev) { + struct goya_device *goya = hdev->asic_specific; + + if (!(goya->hw_cap_initialized & HW_CAP_MME)) + goto disable_tpc; + WREG32(mmMME_QM_GLBL_CFG0, 0); WREG32(mmMME_CMDQ_GLBL_CFG0, 0); +disable_tpc: + if (!(goya->hw_cap_initialized & HW_CAP_TPC)) + return; + WREG32(mmTPC0_QM_GLBL_CFG0, 0); WREG32(mmTPC0_CMDQ_GLBL_CFG0, 0); @@ -1782,8 +1801,12 @@ static void goya_disable_internal_queues(struct hl_device *hdev) */ static int goya_stop_internal_queues(struct hl_device *hdev) { + struct goya_device *goya = hdev->asic_specific; int rc, retval = 0; + if (!(goya->hw_cap_initialized & HW_CAP_MME)) + goto stop_tpc; + /* * Each queue (QMAN) is a separate H/W logic. That means that each * QMAN can be stopped independently and failure to stop one does NOT @@ -1810,6 +1833,10 @@ static int goya_stop_internal_queues(struct hl_device *hdev) retval = -EIO; } +stop_tpc: + if (!(goya->hw_cap_initialized & HW_CAP_TPC)) + return retval; + rc = goya_stop_queue(hdev, mmTPC0_QM_GLBL_CFG1, mmTPC0_QM_CP_STS, @@ -1975,6 +2002,11 @@ static int goya_stop_internal_queues(struct hl_device *hdev) static void goya_dma_stall(struct hl_device *hdev) { + struct goya_device *goya = hdev->asic_specific; + + if (!(goya->hw_cap_initialized & HW_CAP_DMA)) + return; + WREG32(mmDMA_QM_0_GLBL_CFG1, 1 << DMA_QM_0_GLBL_CFG1_DMA_STOP_SHIFT); WREG32(mmDMA_QM_1_GLBL_CFG1, 1 << DMA_QM_1_GLBL_CFG1_DMA_STOP_SHIFT); WREG32(mmDMA_QM_2_GLBL_CFG1, 1 << DMA_QM_2_GLBL_CFG1_DMA_STOP_SHIFT); @@ -1984,6 +2016,11 @@ static void goya_dma_stall(struct hl_device *hdev) static void goya_tpc_stall(struct hl_device *hdev) { + struct goya_device *goya = hdev->asic_specific; + + if (!(goya->hw_cap_initialized & HW_CAP_TPC)) + return; + WREG32(mmTPC0_CFG_TPC_STALL, 1 << TPC0_CFG_TPC_STALL_V_SHIFT); WREG32(mmTPC1_CFG_TPC_STALL, 1 << TPC1_CFG_TPC_STALL_V_SHIFT); WREG32(mmTPC2_CFG_TPC_STALL, 1 << TPC2_CFG_TPC_STALL_V_SHIFT); @@ -1996,6 +2033,11 @@ static void goya_tpc_stall(struct hl_device *hdev) static void goya_mme_stall(struct hl_device *hdev) { + struct goya_device *goya = hdev->asic_specific; + + if (!(goya->hw_cap_initialized & HW_CAP_MME)) + return; + WREG32(mmMME_STALL, 0xFFFFFFFF); } @@ -4648,8 +4690,6 @@ static int goya_memset_device_memory(struct hl_device *hdev, u64 addr, u64 size, rc = goya_send_job_on_qman0(hdev, job); - hl_cb_put(job->patched_cb); - hl_debugfs_remove_job(hdev, job); kfree(job); cb->cs_cnt--; diff --git a/drivers/mmc/core/core.c b/drivers/mmc/core/core.c index aa54d359dab7..a971c4bcc442 100644 --- a/drivers/mmc/core/core.c +++ b/drivers/mmc/core/core.c @@ -1732,8 +1732,11 @@ static int mmc_do_erase(struct mmc_card *card, unsigned int from, * the erase operation does not exceed the max_busy_timeout, we should * use R1B response. Or we need to prevent the host from doing hw busy * detection, which is done by converting to a R1 response instead. + * Note, some hosts requires R1B, which also means they are on their own + * when it comes to deal with the busy timeout. */ - if (card->host->max_busy_timeout && + if (!(card->host->caps & MMC_CAP_NEED_RSP_BUSY) && + card->host->max_busy_timeout && busy_timeout > card->host->max_busy_timeout) { cmd.flags = MMC_RSP_SPI_R1 | MMC_RSP_R1 | MMC_CMD_AC; } else { diff --git a/drivers/mmc/core/mmc.c b/drivers/mmc/core/mmc.c index f6912ded652d..de14b5845f52 100644 --- a/drivers/mmc/core/mmc.c +++ b/drivers/mmc/core/mmc.c @@ -1910,9 +1910,12 @@ static int mmc_sleep(struct mmc_host *host) * If the max_busy_timeout of the host is specified, validate it against * the sleep cmd timeout. A failure means we need to prevent the host * from doing hw busy detection, which is done by converting to a R1 - * response instead of a R1B. + * response instead of a R1B. Note, some hosts requires R1B, which also + * means they are on their own when it comes to deal with the busy + * timeout. */ - if (host->max_busy_timeout && (timeout_ms > host->max_busy_timeout)) { + if (!(host->caps & MMC_CAP_NEED_RSP_BUSY) && host->max_busy_timeout && + (timeout_ms > host->max_busy_timeout)) { cmd.flags = MMC_RSP_R1 | MMC_CMD_AC; } else { cmd.flags = MMC_RSP_R1B | MMC_CMD_AC; diff --git a/drivers/mmc/core/mmc_ops.c b/drivers/mmc/core/mmc_ops.c index da425ee2d9bf..e025604e17d4 100644 --- a/drivers/mmc/core/mmc_ops.c +++ b/drivers/mmc/core/mmc_ops.c @@ -542,9 +542,11 @@ int __mmc_switch(struct mmc_card *card, u8 set, u8 index, u8 value, * If the max_busy_timeout of the host is specified, make sure it's * enough to fit the used timeout_ms. In case it's not, let's instruct * the host to avoid HW busy detection, by converting to a R1 response - * instead of a R1B. + * instead of a R1B. Note, some hosts requires R1B, which also means + * they are on their own when it comes to deal with the busy timeout. */ - if (host->max_busy_timeout && (timeout_ms > host->max_busy_timeout)) + if (!(host->caps & MMC_CAP_NEED_RSP_BUSY) && host->max_busy_timeout && + (timeout_ms > host->max_busy_timeout)) use_r1b_resp = false; cmd.opcode = MMC_SWITCH; diff --git a/drivers/mmc/host/rtsx_pci_sdmmc.c b/drivers/mmc/host/rtsx_pci_sdmmc.c index bd50935dc37d..11087976ab19 100644 --- a/drivers/mmc/host/rtsx_pci_sdmmc.c +++ b/drivers/mmc/host/rtsx_pci_sdmmc.c @@ -606,19 +606,22 @@ static int sd_change_phase(struct realtek_pci_sdmmc *host, u8 sample_point, bool rx) { struct rtsx_pcr *pcr = host->pcr; - + u16 SD_VP_CTL = 0; dev_dbg(sdmmc_dev(host), "%s(%s): sample_point = %d\n", __func__, rx ? "RX" : "TX", sample_point); rtsx_pci_write_register(pcr, CLK_CTL, CHANGE_CLK, CHANGE_CLK); - if (rx) + if (rx) { + SD_VP_CTL = SD_VPRX_CTL; rtsx_pci_write_register(pcr, SD_VPRX_CTL, PHASE_SELECT_MASK, sample_point); - else + } else { + SD_VP_CTL = SD_VPTX_CTL; rtsx_pci_write_register(pcr, SD_VPTX_CTL, PHASE_SELECT_MASK, sample_point); - rtsx_pci_write_register(pcr, SD_VPCLK0_CTL, PHASE_NOT_RESET, 0); - rtsx_pci_write_register(pcr, SD_VPCLK0_CTL, PHASE_NOT_RESET, + } + rtsx_pci_write_register(pcr, SD_VP_CTL, PHASE_NOT_RESET, 0); + rtsx_pci_write_register(pcr, SD_VP_CTL, PHASE_NOT_RESET, PHASE_NOT_RESET); rtsx_pci_write_register(pcr, CLK_CTL, CHANGE_CLK, 0); rtsx_pci_write_register(pcr, SD_CFG1, SD_ASYNC_FIFO_NOT_RST, 0); diff --git a/drivers/mmc/host/sdhci-acpi.c b/drivers/mmc/host/sdhci-acpi.c index 9651dca6863e..2a2173d953f5 100644 --- a/drivers/mmc/host/sdhci-acpi.c +++ b/drivers/mmc/host/sdhci-acpi.c @@ -23,6 +23,7 @@ #include <linux/pm.h> #include <linux/pm_runtime.h> #include <linux/delay.h> +#include <linux/dmi.h> #include <linux/mmc/host.h> #include <linux/mmc/pm.h> @@ -72,9 +73,16 @@ struct sdhci_acpi_host { const struct sdhci_acpi_slot *slot; struct platform_device *pdev; bool use_runtime_pm; + bool is_intel; + bool reset_signal_volt_on_suspend; unsigned long private[0] ____cacheline_aligned; }; +enum { + DMI_QUIRK_RESET_SD_SIGNAL_VOLT_ON_SUSP = BIT(0), + DMI_QUIRK_SD_NO_WRITE_PROTECT = BIT(1), +}; + static inline void *sdhci_acpi_priv(struct sdhci_acpi_host *c) { return (void *)c->private; @@ -391,6 +399,8 @@ static int intel_probe_slot(struct platform_device *pdev, struct acpi_device *ad host->mmc_host_ops.start_signal_voltage_switch = intel_start_signal_voltage_switch; + c->is_intel = true; + return 0; } @@ -647,6 +657,36 @@ static const struct acpi_device_id sdhci_acpi_ids[] = { }; MODULE_DEVICE_TABLE(acpi, sdhci_acpi_ids); +static const struct dmi_system_id sdhci_acpi_quirks[] = { + { + /* + * The Lenovo Miix 320-10ICR has a bug in the _PS0 method of + * the SHC1 ACPI device, this bug causes it to reprogram the + * wrong LDO (DLDO3) to 1.8V if 1.8V modes are used and the + * card is (runtime) suspended + resumed. DLDO3 is used for + * the LCD and setting it to 1.8V causes the LCD to go black. + */ + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"), + DMI_MATCH(DMI_PRODUCT_VERSION, "Lenovo MIIX 320-10ICR"), + }, + .driver_data = (void *)DMI_QUIRK_RESET_SD_SIGNAL_VOLT_ON_SUSP, + }, + { + /* + * The Acer Aspire Switch 10 (SW5-012) microSD slot always + * reports the card being write-protected even though microSD + * cards do not have a write-protect switch at all. + */ + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Acer"), + DMI_MATCH(DMI_PRODUCT_NAME, "Aspire SW5-012"), + }, + .driver_data = (void *)DMI_QUIRK_SD_NO_WRITE_PROTECT, + }, + {} /* Terminating entry */ +}; + static const struct sdhci_acpi_slot *sdhci_acpi_get_slot(struct acpi_device *adev) { const struct sdhci_acpi_uid_slot *u; @@ -663,17 +703,23 @@ static int sdhci_acpi_probe(struct platform_device *pdev) struct device *dev = &pdev->dev; const struct sdhci_acpi_slot *slot; struct acpi_device *device, *child; + const struct dmi_system_id *id; struct sdhci_acpi_host *c; struct sdhci_host *host; struct resource *iomem; resource_size_t len; size_t priv_size; + int quirks = 0; int err; device = ACPI_COMPANION(dev); if (!device) return -ENODEV; + id = dmi_first_match(sdhci_acpi_quirks); + if (id) + quirks = (long)id->driver_data; + slot = sdhci_acpi_get_slot(device); /* Power on the SDHCI controller and its children */ @@ -759,6 +805,12 @@ static int sdhci_acpi_probe(struct platform_device *pdev) dev_warn(dev, "failed to setup card detect gpio\n"); c->use_runtime_pm = false; } + + if (quirks & DMI_QUIRK_RESET_SD_SIGNAL_VOLT_ON_SUSP) + c->reset_signal_volt_on_suspend = true; + + if (quirks & DMI_QUIRK_SD_NO_WRITE_PROTECT) + host->mmc->caps2 |= MMC_CAP2_NO_WRITE_PROTECT; } err = sdhci_setup_host(host); @@ -823,17 +875,39 @@ static int sdhci_acpi_remove(struct platform_device *pdev) return 0; } +static void __maybe_unused sdhci_acpi_reset_signal_voltage_if_needed( + struct device *dev) +{ + struct sdhci_acpi_host *c = dev_get_drvdata(dev); + struct sdhci_host *host = c->host; + + if (c->is_intel && c->reset_signal_volt_on_suspend && + host->mmc->ios.signal_voltage != MMC_SIGNAL_VOLTAGE_330) { + struct intel_host *intel_host = sdhci_acpi_priv(c); + unsigned int fn = INTEL_DSM_V33_SWITCH; + u32 result = 0; + + intel_dsm(intel_host, dev, fn, &result); + } +} + #ifdef CONFIG_PM_SLEEP static int sdhci_acpi_suspend(struct device *dev) { struct sdhci_acpi_host *c = dev_get_drvdata(dev); struct sdhci_host *host = c->host; + int ret; if (host->tuning_mode != SDHCI_TUNING_MODE_3) mmc_retune_needed(host->mmc); - return sdhci_suspend_host(host); + ret = sdhci_suspend_host(host); + if (ret) + return ret; + + sdhci_acpi_reset_signal_voltage_if_needed(dev); + return 0; } static int sdhci_acpi_resume(struct device *dev) @@ -853,11 +927,17 @@ static int sdhci_acpi_runtime_suspend(struct device *dev) { struct sdhci_acpi_host *c = dev_get_drvdata(dev); struct sdhci_host *host = c->host; + int ret; if (host->tuning_mode != SDHCI_TUNING_MODE_3) mmc_retune_needed(host->mmc); - return sdhci_runtime_suspend_host(host); + ret = sdhci_runtime_suspend_host(host); + if (ret) + return ret; + + sdhci_acpi_reset_signal_voltage_if_needed(dev); + return 0; } static int sdhci_acpi_runtime_resume(struct device *dev) diff --git a/drivers/mmc/host/sdhci-cadence.c b/drivers/mmc/host/sdhci-cadence.c index 5827d3751b81..e573495f8726 100644 --- a/drivers/mmc/host/sdhci-cadence.c +++ b/drivers/mmc/host/sdhci-cadence.c @@ -11,6 +11,7 @@ #include <linux/mmc/host.h> #include <linux/mmc/mmc.h> #include <linux/of.h> +#include <linux/of_device.h> #include "sdhci-pltfm.h" @@ -235,6 +236,11 @@ static const struct sdhci_ops sdhci_cdns_ops = { .set_uhs_signaling = sdhci_cdns_set_uhs_signaling, }; +static const struct sdhci_pltfm_data sdhci_cdns_uniphier_pltfm_data = { + .ops = &sdhci_cdns_ops, + .quirks2 = SDHCI_QUIRK2_PRESET_VALUE_BROKEN, +}; + static const struct sdhci_pltfm_data sdhci_cdns_pltfm_data = { .ops = &sdhci_cdns_ops, }; @@ -334,6 +340,7 @@ static void sdhci_cdns_hs400_enhanced_strobe(struct mmc_host *mmc, static int sdhci_cdns_probe(struct platform_device *pdev) { struct sdhci_host *host; + const struct sdhci_pltfm_data *data; struct sdhci_pltfm_host *pltfm_host; struct sdhci_cdns_priv *priv; struct clk *clk; @@ -350,8 +357,12 @@ static int sdhci_cdns_probe(struct platform_device *pdev) if (ret) return ret; + data = of_device_get_match_data(dev); + if (!data) + data = &sdhci_cdns_pltfm_data; + nr_phy_params = sdhci_cdns_phy_param_count(dev->of_node); - host = sdhci_pltfm_init(pdev, &sdhci_cdns_pltfm_data, + host = sdhci_pltfm_init(pdev, data, struct_size(priv, phy_params, nr_phy_params)); if (IS_ERR(host)) { ret = PTR_ERR(host); @@ -431,7 +442,10 @@ static const struct dev_pm_ops sdhci_cdns_pm_ops = { }; static const struct of_device_id sdhci_cdns_match[] = { - { .compatible = "socionext,uniphier-sd4hc" }, + { + .compatible = "socionext,uniphier-sd4hc", + .data = &sdhci_cdns_uniphier_pltfm_data, + }, { .compatible = "cdns,sd4hc" }, { /* sentinel */ } }; diff --git a/drivers/mmc/host/sdhci-msm.c b/drivers/mmc/host/sdhci-msm.c index c3a160c18047..3955fa5db43c 100644 --- a/drivers/mmc/host/sdhci-msm.c +++ b/drivers/mmc/host/sdhci-msm.c @@ -1590,7 +1590,7 @@ static u32 sdhci_msm_cqe_irq(struct sdhci_host *host, u32 intmask) return 0; } -void sdhci_msm_cqe_disable(struct mmc_host *mmc, bool recovery) +static void sdhci_msm_cqe_disable(struct mmc_host *mmc, bool recovery) { struct sdhci_host *host = mmc_priv(mmc); unsigned long flags; diff --git a/drivers/mmc/host/sdhci-of-at91.c b/drivers/mmc/host/sdhci-of-at91.c index ab2bd314a390..fcef5c0d0908 100644 --- a/drivers/mmc/host/sdhci-of-at91.c +++ b/drivers/mmc/host/sdhci-of-at91.c @@ -132,7 +132,8 @@ static void sdhci_at91_reset(struct sdhci_host *host, u8 mask) sdhci_reset(host, mask); - if (host->mmc->caps & MMC_CAP_NONREMOVABLE) + if ((host->mmc->caps & MMC_CAP_NONREMOVABLE) + || mmc_gpio_get_cd(host->mmc) >= 0) sdhci_at91_set_force_card_detect(host); if (priv->cal_always_on && (mask & SDHCI_RESET_ALL)) @@ -427,8 +428,11 @@ static int sdhci_at91_probe(struct platform_device *pdev) * detection procedure using the SDMCC_CD signal is bypassed. * This bit is reset when a software reset for all command is performed * so we need to implement our own reset function to set back this bit. + * + * WA: SAMA5D2 doesn't drive CMD if using CD GPIO line. */ - if (host->mmc->caps & MMC_CAP_NONREMOVABLE) + if ((host->mmc->caps & MMC_CAP_NONREMOVABLE) + || mmc_gpio_get_cd(host->mmc) >= 0) sdhci_at91_set_force_card_detect(host); pm_runtime_put_autosuspend(&pdev->dev); diff --git a/drivers/mmc/host/sdhci-omap.c b/drivers/mmc/host/sdhci-omap.c index 882053151a47..c4978177ef88 100644 --- a/drivers/mmc/host/sdhci-omap.c +++ b/drivers/mmc/host/sdhci-omap.c @@ -1192,6 +1192,9 @@ static int sdhci_omap_probe(struct platform_device *pdev) if (of_find_property(dev->of_node, "dmas", NULL)) sdhci_switch_external_dma(host, true); + /* R1B responses is required to properly manage HW busy detection. */ + mmc->caps |= MMC_CAP_NEED_RSP_BUSY; + ret = sdhci_setup_host(host); if (ret) goto err_put_sync; diff --git a/drivers/mmc/host/sdhci-pci-gli.c b/drivers/mmc/host/sdhci-pci-gli.c index 5eea8d70a85d..ce15a05f23d4 100644 --- a/drivers/mmc/host/sdhci-pci-gli.c +++ b/drivers/mmc/host/sdhci-pci-gli.c @@ -262,10 +262,26 @@ static int gl9750_execute_tuning(struct sdhci_host *host, u32 opcode) return 0; } +static void gli_pcie_enable_msi(struct sdhci_pci_slot *slot) +{ + int ret; + + ret = pci_alloc_irq_vectors(slot->chip->pdev, 1, 1, + PCI_IRQ_MSI | PCI_IRQ_MSIX); + if (ret < 0) { + pr_warn("%s: enable PCI MSI failed, error=%d\n", + mmc_hostname(slot->host->mmc), ret); + return; + } + + slot->host->irq = pci_irq_vector(slot->chip->pdev, 0); +} + static int gli_probe_slot_gl9750(struct sdhci_pci_slot *slot) { struct sdhci_host *host = slot->host; + gli_pcie_enable_msi(slot); slot->host->mmc->caps2 |= MMC_CAP2_NO_SDIO; sdhci_enable_v4_mode(host); @@ -276,6 +292,7 @@ static int gli_probe_slot_gl9755(struct sdhci_pci_slot *slot) { struct sdhci_host *host = slot->host; + gli_pcie_enable_msi(slot); slot->host->mmc->caps2 |= MMC_CAP2_NO_SDIO; sdhci_enable_v4_mode(host); diff --git a/drivers/mmc/host/sdhci-tegra.c b/drivers/mmc/host/sdhci-tegra.c index 403ac44a7378..a25c3a4d3f6c 100644 --- a/drivers/mmc/host/sdhci-tegra.c +++ b/drivers/mmc/host/sdhci-tegra.c @@ -1552,6 +1552,9 @@ static int sdhci_tegra_probe(struct platform_device *pdev) if (tegra_host->soc_data->nvquirks & NVQUIRK_ENABLE_DDR50) host->mmc->caps |= MMC_CAP_1_8V_DDR; + /* R1B responses is required to properly manage HW busy detection. */ + host->mmc->caps |= MMC_CAP_NEED_RSP_BUSY; + tegra_sdhci_parse_dt(host); tegra_host->power_gpio = devm_gpiod_get_optional(&pdev->dev, "power", diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig index 25a8f9387d5a..db8884ad6d40 100644 --- a/drivers/net/Kconfig +++ b/drivers/net/Kconfig @@ -149,6 +149,7 @@ config NET_FC config IFB tristate "Intermediate Functional Block support" depends on NET_CLS_ACT + select NET_REDIRECT ---help--- This is an intermediate driver that allows sharing of resources. diff --git a/drivers/net/bonding/bond_alb.c b/drivers/net/bonding/bond_alb.c index 1cc2cd894f87..c81698550e5a 100644 --- a/drivers/net/bonding/bond_alb.c +++ b/drivers/net/bonding/bond_alb.c @@ -50,11 +50,6 @@ struct arp_pkt { }; #pragma pack() -static inline struct arp_pkt *arp_pkt(const struct sk_buff *skb) -{ - return (struct arp_pkt *)skb_network_header(skb); -} - /* Forward declaration */ static void alb_send_learning_packets(struct slave *slave, u8 mac_addr[], bool strict_match); @@ -553,10 +548,11 @@ static void rlb_req_update_subnet_clients(struct bonding *bond, __be32 src_ip) spin_unlock(&bond->mode_lock); } -static struct slave *rlb_choose_channel(struct sk_buff *skb, struct bonding *bond) +static struct slave *rlb_choose_channel(struct sk_buff *skb, + struct bonding *bond, + const struct arp_pkt *arp) { struct alb_bond_info *bond_info = &(BOND_ALB_INFO(bond)); - struct arp_pkt *arp = arp_pkt(skb); struct slave *assigned_slave, *curr_active_slave; struct rlb_client_info *client_info; u32 hash_index = 0; @@ -653,8 +649,12 @@ static struct slave *rlb_choose_channel(struct sk_buff *skb, struct bonding *bon */ static struct slave *rlb_arp_xmit(struct sk_buff *skb, struct bonding *bond) { - struct arp_pkt *arp = arp_pkt(skb); struct slave *tx_slave = NULL; + struct arp_pkt *arp; + + if (!pskb_network_may_pull(skb, sizeof(*arp))) + return NULL; + arp = (struct arp_pkt *)skb_network_header(skb); /* Don't modify or load balance ARPs that do not originate locally * (e.g.,arrive via a bridge). @@ -664,7 +664,7 @@ static struct slave *rlb_arp_xmit(struct sk_buff *skb, struct bonding *bond) if (arp->op_code == htons(ARPOP_REPLY)) { /* the arp must be sent on the selected rx channel */ - tx_slave = rlb_choose_channel(skb, bond); + tx_slave = rlb_choose_channel(skb, bond, arp); if (tx_slave) bond_hw_addr_copy(arp->mac_src, tx_slave->dev->dev_addr, tx_slave->dev->addr_len); @@ -676,7 +676,7 @@ static struct slave *rlb_arp_xmit(struct sk_buff *skb, struct bonding *bond) * When the arp reply is received the entry will be updated * with the correct unicast address of the client. */ - tx_slave = rlb_choose_channel(skb, bond); + tx_slave = rlb_choose_channel(skb, bond, arp); /* The ARP reply packets must be delayed so that * they can cancel out the influence of the ARP request. diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 48d5ec770b94..d10805e5e623 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -3526,6 +3526,47 @@ static void bond_fold_stats(struct rtnl_link_stats64 *_res, } } +#ifdef CONFIG_LOCKDEP +static int bond_get_lowest_level_rcu(struct net_device *dev) +{ + struct net_device *ldev, *next, *now, *dev_stack[MAX_NEST_DEV + 1]; + struct list_head *niter, *iter, *iter_stack[MAX_NEST_DEV + 1]; + int cur = 0, max = 0; + + now = dev; + iter = &dev->adj_list.lower; + + while (1) { + next = NULL; + while (1) { + ldev = netdev_next_lower_dev_rcu(now, &iter); + if (!ldev) + break; + + next = ldev; + niter = &ldev->adj_list.lower; + dev_stack[cur] = now; + iter_stack[cur++] = iter; + if (max <= cur) + max = cur; + break; + } + + if (!next) { + if (!cur) + return max; + next = dev_stack[--cur]; + niter = iter_stack[cur]; + } + + now = next; + iter = niter; + } + + return max; +} +#endif + static void bond_get_stats(struct net_device *bond_dev, struct rtnl_link_stats64 *stats) { @@ -3533,11 +3574,17 @@ static void bond_get_stats(struct net_device *bond_dev, struct rtnl_link_stats64 temp; struct list_head *iter; struct slave *slave; + int nest_level = 0; - spin_lock(&bond->stats_lock); - memcpy(stats, &bond->bond_stats, sizeof(*stats)); rcu_read_lock(); +#ifdef CONFIG_LOCKDEP + nest_level = bond_get_lowest_level_rcu(bond_dev); +#endif + + spin_lock_nested(&bond->stats_lock, nest_level); + memcpy(stats, &bond->bond_stats, sizeof(*stats)); + bond_for_each_slave_rcu(bond, slave, iter) { const struct rtnl_link_stats64 *new = dev_get_stats(slave->dev, &temp); @@ -3547,10 +3594,10 @@ static void bond_get_stats(struct net_device *bond_dev, /* save off the slave stats for the next run */ memcpy(&slave->slave_stats, new, sizeof(*new)); } - rcu_read_unlock(); memcpy(&bond->bond_stats, stats, sizeof(*stats)); spin_unlock(&bond->stats_lock); + rcu_read_unlock(); } static int bond_do_ioctl(struct net_device *bond_dev, struct ifreq *ifr, int cmd) @@ -3640,6 +3687,8 @@ static int bond_do_ioctl(struct net_device *bond_dev, struct ifreq *ifr, int cmd case BOND_RELEASE_OLD: case SIOCBONDRELEASE: res = bond_release(bond_dev, slave_dev); + if (!res) + netdev_update_lockdep_key(slave_dev); break; case BOND_SETHWADDR_OLD: case SIOCBONDSETHWADDR: diff --git a/drivers/net/bonding/bond_options.c b/drivers/net/bonding/bond_options.c index ddb3916d3506..215c10923289 100644 --- a/drivers/net/bonding/bond_options.c +++ b/drivers/net/bonding/bond_options.c @@ -1398,6 +1398,8 @@ static int bond_option_slaves_set(struct bonding *bond, case '-': slave_dbg(bond->dev, dev, "Releasing interface\n"); ret = bond_release(bond->dev, dev); + if (!ret) + netdev_update_lockdep_key(dev); break; default: diff --git a/drivers/net/caif/caif_spi.c b/drivers/net/caif/caif_spi.c index 8e81bdf98ac6..63f2548f5b1b 100644 --- a/drivers/net/caif/caif_spi.c +++ b/drivers/net/caif/caif_spi.c @@ -141,29 +141,29 @@ static ssize_t dbgfs_state(struct file *file, char __user *user_buf, return 0; /* Print out debug information. */ - len += snprintf((buf + len), (DEBUGFS_BUF_SIZE - len), - "CAIF SPI debug information:\n"); - - len += snprintf((buf + len), (DEBUGFS_BUF_SIZE - len), FLAVOR); - - len += snprintf((buf + len), (DEBUGFS_BUF_SIZE - len), - "STATE: %d\n", cfspi->dbg_state); - len += snprintf((buf + len), (DEBUGFS_BUF_SIZE - len), - "Previous CMD: 0x%x\n", cfspi->pcmd); - len += snprintf((buf + len), (DEBUGFS_BUF_SIZE - len), - "Current CMD: 0x%x\n", cfspi->cmd); - len += snprintf((buf + len), (DEBUGFS_BUF_SIZE - len), - "Previous TX len: %d\n", cfspi->tx_ppck_len); - len += snprintf((buf + len), (DEBUGFS_BUF_SIZE - len), - "Previous RX len: %d\n", cfspi->rx_ppck_len); - len += snprintf((buf + len), (DEBUGFS_BUF_SIZE - len), - "Current TX len: %d\n", cfspi->tx_cpck_len); - len += snprintf((buf + len), (DEBUGFS_BUF_SIZE - len), - "Current RX len: %d\n", cfspi->rx_cpck_len); - len += snprintf((buf + len), (DEBUGFS_BUF_SIZE - len), - "Next TX len: %d\n", cfspi->tx_npck_len); - len += snprintf((buf + len), (DEBUGFS_BUF_SIZE - len), - "Next RX len: %d\n", cfspi->rx_npck_len); + len += scnprintf((buf + len), (DEBUGFS_BUF_SIZE - len), + "CAIF SPI debug information:\n"); + + len += scnprintf((buf + len), (DEBUGFS_BUF_SIZE - len), FLAVOR); + + len += scnprintf((buf + len), (DEBUGFS_BUF_SIZE - len), + "STATE: %d\n", cfspi->dbg_state); + len += scnprintf((buf + len), (DEBUGFS_BUF_SIZE - len), + "Previous CMD: 0x%x\n", cfspi->pcmd); + len += scnprintf((buf + len), (DEBUGFS_BUF_SIZE - len), + "Current CMD: 0x%x\n", cfspi->cmd); + len += scnprintf((buf + len), (DEBUGFS_BUF_SIZE - len), + "Previous TX len: %d\n", cfspi->tx_ppck_len); + len += scnprintf((buf + len), (DEBUGFS_BUF_SIZE - len), + "Previous RX len: %d\n", cfspi->rx_ppck_len); + len += scnprintf((buf + len), (DEBUGFS_BUF_SIZE - len), + "Current TX len: %d\n", cfspi->tx_cpck_len); + len += scnprintf((buf + len), (DEBUGFS_BUF_SIZE - len), + "Current RX len: %d\n", cfspi->rx_cpck_len); + len += scnprintf((buf + len), (DEBUGFS_BUF_SIZE - len), + "Next TX len: %d\n", cfspi->tx_npck_len); + len += scnprintf((buf + len), (DEBUGFS_BUF_SIZE - len), + "Next RX len: %d\n", cfspi->rx_npck_len); if (len > DEBUGFS_BUF_SIZE) len = DEBUGFS_BUF_SIZE; @@ -180,23 +180,23 @@ static ssize_t print_frame(char *buf, size_t size, char *frm, int len = 0; int i; for (i = 0; i < count; i++) { - len += snprintf((buf + len), (size - len), + len += scnprintf((buf + len), (size - len), "[0x" BYTE_HEX_FMT "]", frm[i]); if ((i == cut) && (count > (cut * 2))) { /* Fast forward. */ i = count - cut; - len += snprintf((buf + len), (size - len), - "--- %zu bytes skipped ---\n", - count - (cut * 2)); + len += scnprintf((buf + len), (size - len), + "--- %zu bytes skipped ---\n", + count - (cut * 2)); } if ((!(i % 10)) && i) { - len += snprintf((buf + len), (DEBUGFS_BUF_SIZE - len), - "\n"); + len += scnprintf((buf + len), (DEBUGFS_BUF_SIZE - len), + "\n"); } } - len += snprintf((buf + len), (DEBUGFS_BUF_SIZE - len), "\n"); + len += scnprintf((buf + len), (DEBUGFS_BUF_SIZE - len), "\n"); return len; } @@ -214,18 +214,18 @@ static ssize_t dbgfs_frame(struct file *file, char __user *user_buf, return 0; /* Print out debug information. */ - len += snprintf((buf + len), (DEBUGFS_BUF_SIZE - len), - "Current frame:\n"); + len += scnprintf((buf + len), (DEBUGFS_BUF_SIZE - len), + "Current frame:\n"); - len += snprintf((buf + len), (DEBUGFS_BUF_SIZE - len), - "Tx data (Len: %d):\n", cfspi->tx_cpck_len); + len += scnprintf((buf + len), (DEBUGFS_BUF_SIZE - len), + "Tx data (Len: %d):\n", cfspi->tx_cpck_len); len += print_frame((buf + len), (DEBUGFS_BUF_SIZE - len), cfspi->xfer.va_tx[0], (cfspi->tx_cpck_len + SPI_CMD_SZ), 100); - len += snprintf((buf + len), (DEBUGFS_BUF_SIZE - len), - "Rx data (Len: %d):\n", cfspi->rx_cpck_len); + len += scnprintf((buf + len), (DEBUGFS_BUF_SIZE - len), + "Rx data (Len: %d):\n", cfspi->rx_cpck_len); len += print_frame((buf + len), (DEBUGFS_BUF_SIZE - len), cfspi->xfer.va_rx, diff --git a/drivers/net/can/dev.c b/drivers/net/can/dev.c index 6ee06a49fb4c..68834a2853c9 100644 --- a/drivers/net/can/dev.c +++ b/drivers/net/can/dev.c @@ -883,6 +883,7 @@ static const struct nla_policy can_policy[IFLA_CAN_MAX + 1] = { = { .len = sizeof(struct can_bittiming) }, [IFLA_CAN_DATA_BITTIMING_CONST] = { .len = sizeof(struct can_bittiming_const) }, + [IFLA_CAN_TERMINATION] = { .type = NLA_U16 }, }; static int can_validate(struct nlattr *tb[], struct nlattr *data[], diff --git a/drivers/net/can/slcan.c b/drivers/net/can/slcan.c index 2f5c287eac95..a3664281a33f 100644 --- a/drivers/net/can/slcan.c +++ b/drivers/net/can/slcan.c @@ -625,7 +625,10 @@ err_free_chan: tty->disc_data = NULL; clear_bit(SLF_INUSE, &sl->flags); slc_free_netdev(sl->dev); + /* do not call free_netdev before rtnl_unlock */ + rtnl_unlock(); free_netdev(sl->dev); + return err; err_exit: rtnl_unlock(); diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c index 449a22172e07..1a69286daa8d 100644 --- a/drivers/net/dsa/b53/b53_common.c +++ b/drivers/net/dsa/b53/b53_common.c @@ -1366,6 +1366,9 @@ void b53_vlan_add(struct dsa_switch *ds, int port, b53_get_vlan_entry(dev, vid, vl); + if (vid == 0 && vid == b53_default_pvid(dev)) + untagged = true; + vl->members |= BIT(port); if (untagged && !dsa_is_cpu_port(ds, port)) vl->untag |= BIT(port); diff --git a/drivers/net/dsa/bcm_sf2.c b/drivers/net/dsa/bcm_sf2.c index d1955543acd1..b0f5280a83cb 100644 --- a/drivers/net/dsa/bcm_sf2.c +++ b/drivers/net/dsa/bcm_sf2.c @@ -69,8 +69,7 @@ static void bcm_sf2_imp_setup(struct dsa_switch *ds, int port) /* Force link status for IMP port */ reg = core_readl(priv, offset); reg |= (MII_SW_OR | LINK_STS); - if (priv->type == BCM7278_DEVICE_ID) - reg |= GMII_SPEED_UP_2G; + reg &= ~GMII_SPEED_UP_2G; core_writel(priv, reg, offset); /* Enable Broadcast, Multicast, Unicast forwarding to IMP port */ diff --git a/drivers/net/dsa/mt7530.c b/drivers/net/dsa/mt7530.c index 022466ca1c19..7cbd1bd4c5a6 100644 --- a/drivers/net/dsa/mt7530.c +++ b/drivers/net/dsa/mt7530.c @@ -566,7 +566,7 @@ mt7530_mib_reset(struct dsa_switch *ds) static void mt7530_port_set_status(struct mt7530_priv *priv, int port, int enable) { - u32 mask = PMCR_TX_EN | PMCR_RX_EN; + u32 mask = PMCR_TX_EN | PMCR_RX_EN | PMCR_FORCE_LNK; if (enable) mt7530_set(priv, MT7530_PMCR_P(port), mask); @@ -1444,7 +1444,7 @@ static void mt7530_phylink_mac_config(struct dsa_switch *ds, int port, mcr_new &= ~(PMCR_FORCE_SPEED_1000 | PMCR_FORCE_SPEED_100 | PMCR_FORCE_FDX | PMCR_TX_FC_EN | PMCR_RX_FC_EN); mcr_new |= PMCR_IFG_XMIT(1) | PMCR_MAC_MODE | PMCR_BACKOFF_EN | - PMCR_BACKPR_EN | PMCR_FORCE_MODE | PMCR_FORCE_LNK; + PMCR_BACKPR_EN | PMCR_FORCE_MODE; /* Are we connected to external phy */ if (port == 5 && dsa_is_user_port(ds, 5)) diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c index 8c9289549688..2f993e673ec7 100644 --- a/drivers/net/dsa/mv88e6xxx/chip.c +++ b/drivers/net/dsa/mv88e6xxx/chip.c @@ -2769,6 +2769,8 @@ static u64 mv88e6xxx_devlink_atu_bin_get(struct mv88e6xxx_chip *chip, goto unlock; } + occupancy &= MV88E6XXX_G2_ATU_STATS_MASK; + unlock: mv88e6xxx_reg_unlock(chip); diff --git a/drivers/net/dsa/mv88e6xxx/global1.c b/drivers/net/dsa/mv88e6xxx/global1.c index b016cc205f81..ca3a7a7a73c3 100644 --- a/drivers/net/dsa/mv88e6xxx/global1.c +++ b/drivers/net/dsa/mv88e6xxx/global1.c @@ -278,13 +278,13 @@ int mv88e6095_g1_set_egress_port(struct mv88e6xxx_chip *chip, switch (direction) { case MV88E6XXX_EGRESS_DIR_INGRESS: dest_port_chip = &chip->ingress_dest_port; - reg &= MV88E6185_G1_MONITOR_CTL_INGRESS_DEST_MASK; + reg &= ~MV88E6185_G1_MONITOR_CTL_INGRESS_DEST_MASK; reg |= port << __bf_shf(MV88E6185_G1_MONITOR_CTL_INGRESS_DEST_MASK); break; case MV88E6XXX_EGRESS_DIR_EGRESS: dest_port_chip = &chip->egress_dest_port; - reg &= MV88E6185_G1_MONITOR_CTL_EGRESS_DEST_MASK; + reg &= ~MV88E6185_G1_MONITOR_CTL_EGRESS_DEST_MASK; reg |= port << __bf_shf(MV88E6185_G1_MONITOR_CTL_EGRESS_DEST_MASK); break; diff --git a/drivers/net/dsa/mv88e6xxx/global2.c b/drivers/net/dsa/mv88e6xxx/global2.c index 01503014b1ee..8fd483020c5b 100644 --- a/drivers/net/dsa/mv88e6xxx/global2.c +++ b/drivers/net/dsa/mv88e6xxx/global2.c @@ -1099,6 +1099,13 @@ int mv88e6xxx_g2_irq_setup(struct mv88e6xxx_chip *chip) { int err, irq, virq; + chip->g2_irq.masked = ~0; + mv88e6xxx_reg_lock(chip); + err = mv88e6xxx_g2_int_mask(chip, ~chip->g2_irq.masked); + mv88e6xxx_reg_unlock(chip); + if (err) + return err; + chip->g2_irq.domain = irq_domain_add_simple( chip->dev->of_node, 16, 0, &mv88e6xxx_g2_irq_domain_ops, chip); if (!chip->g2_irq.domain) @@ -1108,7 +1115,6 @@ int mv88e6xxx_g2_irq_setup(struct mv88e6xxx_chip *chip) irq_create_mapping(chip->g2_irq.domain, irq); chip->g2_irq.chip = mv88e6xxx_g2_irq_chip; - chip->g2_irq.masked = ~0; chip->device_irq = irq_find_mapping(chip->g1_irq.domain, MV88E6XXX_G1_STS_IRQ_DEVICE); diff --git a/drivers/net/dsa/sja1105/sja1105_main.c b/drivers/net/dsa/sja1105/sja1105_main.c index 03ba6d25f7fe..7edea5741a5f 100644 --- a/drivers/net/dsa/sja1105/sja1105_main.c +++ b/drivers/net/dsa/sja1105/sja1105_main.c @@ -1741,7 +1741,8 @@ static void sja1105_teardown(struct dsa_switch *ds) if (!dsa_is_user_port(ds, port)) continue; - kthread_destroy_worker(sp->xmit_worker); + if (sp->xmit_worker) + kthread_destroy_worker(sp->xmit_worker); } sja1105_tas_teardown(ds); diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.c b/drivers/net/ethernet/amazon/ena/ena_netdev.c index 0b2fd96b93d7..cada6e7e30f4 100644 --- a/drivers/net/ethernet/amazon/ena/ena_netdev.c +++ b/drivers/net/ethernet/amazon/ena/ena_netdev.c @@ -1018,13 +1018,9 @@ static int ena_refill_rx_bufs(struct ena_ring *rx_ring, u32 num) struct ena_rx_buffer *rx_info; req_id = rx_ring->free_ids[next_to_use]; - rc = validate_rx_req_id(rx_ring, req_id); - if (unlikely(rc < 0)) - break; rx_info = &rx_ring->rx_buffer_info[req_id]; - rc = ena_alloc_rx_page(rx_ring, rx_info, GFP_ATOMIC | __GFP_COMP); if (unlikely(rc < 0)) { @@ -1379,9 +1375,15 @@ static struct sk_buff *ena_rx_skb(struct ena_ring *rx_ring, struct ena_rx_buffer *rx_info; u16 len, req_id, buf = 0; void *va; + int rc; len = ena_bufs[buf].len; req_id = ena_bufs[buf].req_id; + + rc = validate_rx_req_id(rx_ring, req_id); + if (unlikely(rc < 0)) + return NULL; + rx_info = &rx_ring->rx_buffer_info[req_id]; if (unlikely(!rx_info->page)) { @@ -1454,6 +1456,11 @@ static struct sk_buff *ena_rx_skb(struct ena_ring *rx_ring, buf++; len = ena_bufs[buf].len; req_id = ena_bufs[buf].req_id; + + rc = validate_rx_req_id(rx_ring, req_id); + if (unlikely(rc < 0)) + return NULL; + rx_info = &rx_ring->rx_buffer_info[req_id]; } while (1); @@ -1968,7 +1975,7 @@ static int ena_enable_msix(struct ena_adapter *adapter) } /* Reserved the max msix vectors we might need */ - msix_vecs = ENA_MAX_MSIX_VEC(adapter->num_io_queues); + msix_vecs = ENA_MAX_MSIX_VEC(adapter->max_num_io_queues); netif_dbg(adapter, probe, adapter->netdev, "trying to enable MSI-X, vectors %d\n", msix_vecs); @@ -2068,6 +2075,7 @@ static int ena_request_mgmnt_irq(struct ena_adapter *adapter) static int ena_request_io_irq(struct ena_adapter *adapter) { + u32 io_queue_count = adapter->num_io_queues + adapter->xdp_num_queues; unsigned long flags = 0; struct ena_irq *irq; int rc = 0, i, k; @@ -2078,7 +2086,7 @@ static int ena_request_io_irq(struct ena_adapter *adapter) return -EINVAL; } - for (i = ENA_IO_IRQ_FIRST_IDX; i < adapter->msix_vecs; i++) { + for (i = ENA_IO_IRQ_FIRST_IDX; i < ENA_MAX_MSIX_VEC(io_queue_count); i++) { irq = &adapter->irq_tbl[i]; rc = request_irq(irq->vector, irq->handler, flags, irq->name, irq->data); @@ -2119,6 +2127,7 @@ static void ena_free_mgmnt_irq(struct ena_adapter *adapter) static void ena_free_io_irq(struct ena_adapter *adapter) { + u32 io_queue_count = adapter->num_io_queues + adapter->xdp_num_queues; struct ena_irq *irq; int i; @@ -2129,7 +2138,7 @@ static void ena_free_io_irq(struct ena_adapter *adapter) } #endif /* CONFIG_RFS_ACCEL */ - for (i = ENA_IO_IRQ_FIRST_IDX; i < adapter->msix_vecs; i++) { + for (i = ENA_IO_IRQ_FIRST_IDX; i < ENA_MAX_MSIX_VEC(io_queue_count); i++) { irq = &adapter->irq_tbl[i]; irq_set_affinity_hint(irq->vector, NULL); free_irq(irq->vector, irq->data); @@ -2144,12 +2153,13 @@ static void ena_disable_msix(struct ena_adapter *adapter) static void ena_disable_io_intr_sync(struct ena_adapter *adapter) { + u32 io_queue_count = adapter->num_io_queues + adapter->xdp_num_queues; int i; if (!netif_running(adapter->netdev)) return; - for (i = ENA_IO_IRQ_FIRST_IDX; i < adapter->msix_vecs; i++) + for (i = ENA_IO_IRQ_FIRST_IDX; i < ENA_MAX_MSIX_VEC(io_queue_count); i++) synchronize_irq(adapter->irq_tbl[i].vector); } @@ -3476,6 +3486,7 @@ static int ena_restore_device(struct ena_adapter *adapter) netif_carrier_on(adapter->netdev); mod_timer(&adapter->timer_service, round_jiffies(jiffies + HZ)); + adapter->last_keep_alive_jiffies = jiffies; dev_err(&pdev->dev, "Device reset completed successfully, Driver info: %s\n", version); @@ -4325,13 +4336,15 @@ err_disable_device: /*****************************************************************************/ -/* ena_remove - Device Removal Routine +/* __ena_shutoff - Helper used in both PCI remove/shutdown routines * @pdev: PCI device information struct + * @shutdown: Is it a shutdown operation? If false, means it is a removal * - * ena_remove is called by the PCI subsystem to alert the driver - * that it should release a PCI device. + * __ena_shutoff is a helper routine that does the real work on shutdown and + * removal paths; the difference between those paths is with regards to whether + * dettach or unregister the netdevice. */ -static void ena_remove(struct pci_dev *pdev) +static void __ena_shutoff(struct pci_dev *pdev, bool shutdown) { struct ena_adapter *adapter = pci_get_drvdata(pdev); struct ena_com_dev *ena_dev; @@ -4350,13 +4363,17 @@ static void ena_remove(struct pci_dev *pdev) cancel_work_sync(&adapter->reset_task); - rtnl_lock(); + rtnl_lock(); /* lock released inside the below if-else block */ ena_destroy_device(adapter, true); - rtnl_unlock(); - - unregister_netdev(netdev); - - free_netdev(netdev); + if (shutdown) { + netif_device_detach(netdev); + dev_close(netdev); + rtnl_unlock(); + } else { + rtnl_unlock(); + unregister_netdev(netdev); + free_netdev(netdev); + } ena_com_rss_destroy(ena_dev); @@ -4371,6 +4388,30 @@ static void ena_remove(struct pci_dev *pdev) vfree(ena_dev); } +/* ena_remove - Device Removal Routine + * @pdev: PCI device information struct + * + * ena_remove is called by the PCI subsystem to alert the driver + * that it should release a PCI device. + */ + +static void ena_remove(struct pci_dev *pdev) +{ + __ena_shutoff(pdev, false); +} + +/* ena_shutdown - Device Shutdown Routine + * @pdev: PCI device information struct + * + * ena_shutdown is called by the PCI subsystem to alert the driver that + * a shutdown/reboot (or kexec) is happening and device must be disabled. + */ + +static void ena_shutdown(struct pci_dev *pdev) +{ + __ena_shutoff(pdev, true); +} + #ifdef CONFIG_PM /* ena_suspend - PM suspend callback * @pdev: PCI device information struct @@ -4420,6 +4461,7 @@ static struct pci_driver ena_pci_driver = { .id_table = ena_pci_tbl, .probe = ena_probe, .remove = ena_remove, + .shutdown = ena_shutdown, #ifdef CONFIG_PM .suspend = ena_suspend, .resume = ena_resume, diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c b/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c index a1f99bef4a68..7b55633d2cb9 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c @@ -722,6 +722,11 @@ static int aq_ethtool_set_priv_flags(struct net_device *ndev, u32 flags) if (flags & ~AQ_PRIV_FLAGS_MASK) return -EOPNOTSUPP; + if (hweight32((flags | priv_flags) & AQ_HW_LOOPBACK_MASK) > 1) { + netdev_info(ndev, "Can't enable more than one loopback simultaneously\n"); + return -EINVAL; + } + cfg->priv_flags = flags; if ((priv_flags ^ flags) & BIT(AQ_HW_LOOPBACK_DMA_NET)) { diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_filters.c b/drivers/net/ethernet/aquantia/atlantic/aq_filters.c index 6102251bb909..03ff92bc4a7f 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_filters.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_filters.c @@ -163,7 +163,7 @@ aq_check_approve_fvlan(struct aq_nic_s *aq_nic, } if ((aq_nic->ndev->features & NETIF_F_HW_VLAN_CTAG_FILTER) && - (!test_bit(be16_to_cpu(fsp->h_ext.vlan_tci), + (!test_bit(be16_to_cpu(fsp->h_ext.vlan_tci) & VLAN_VID_MASK, aq_nic->active_vlans))) { netdev_err(aq_nic->ndev, "ethtool: unknown vlan-id specified"); diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_hw.h b/drivers/net/ethernet/aquantia/atlantic/aq_hw.h index cc70c606b6ef..251767c31f7e 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_hw.h +++ b/drivers/net/ethernet/aquantia/atlantic/aq_hw.h @@ -337,6 +337,8 @@ struct aq_fw_ops { void (*enable_ptp)(struct aq_hw_s *self, int enable); + void (*adjust_ptp)(struct aq_hw_s *self, uint64_t adj); + int (*set_eee_rate)(struct aq_hw_s *self, u32 speed); int (*get_eee_rate)(struct aq_hw_s *self, u32 *rate, diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_nic.c b/drivers/net/ethernet/aquantia/atlantic/aq_nic.c index c85e3e29012c..e95f6a6bef73 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_nic.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_nic.c @@ -533,8 +533,10 @@ unsigned int aq_nic_map_skb(struct aq_nic_s *self, struct sk_buff *skb, dx_buff->len, DMA_TO_DEVICE); - if (unlikely(dma_mapping_error(aq_nic_get_dev(self), dx_buff->pa))) + if (unlikely(dma_mapping_error(aq_nic_get_dev(self), dx_buff->pa))) { + ret = 0; goto exit; + } first = dx_buff; dx_buff->len_pkt = skb->len; @@ -655,10 +657,6 @@ int aq_nic_xmit(struct aq_nic_s *self, struct sk_buff *skb) if (likely(frags)) { err = self->aq_hw_ops->hw_ring_tx_xmit(self->aq_hw, ring, frags); - if (err >= 0) { - ++ring->stats.tx.packets; - ring->stats.tx.bytes += skb->len; - } } else { err = NETDEV_TX_BUSY; } diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_pci_func.c b/drivers/net/ethernet/aquantia/atlantic/aq_pci_func.c index 6b27af0db499..78b6f3248756 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_pci_func.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_pci_func.c @@ -359,7 +359,8 @@ static int aq_suspend_common(struct device *dev, bool deep) netif_device_detach(nic->ndev); netif_tx_stop_all_queues(nic->ndev); - aq_nic_stop(nic); + if (netif_running(nic->ndev)) + aq_nic_stop(nic); if (deep) { aq_nic_deinit(nic, !nic->aq_hw->aq_nic_cfg->wol); @@ -375,7 +376,7 @@ static int atl_resume_common(struct device *dev, bool deep) { struct pci_dev *pdev = to_pci_dev(dev); struct aq_nic_s *nic; - int ret; + int ret = 0; nic = pci_get_drvdata(pdev); @@ -390,9 +391,11 @@ static int atl_resume_common(struct device *dev, bool deep) goto err_exit; } - ret = aq_nic_start(nic); - if (ret) - goto err_exit; + if (netif_running(nic->ndev)) { + ret = aq_nic_start(nic); + if (ret) + goto err_exit; + } netif_device_attach(nic->ndev); netif_tx_start_all_queues(nic->ndev); diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_ring.c b/drivers/net/ethernet/aquantia/atlantic/aq_ring.c index 951d86f8b66e..bae95a618560 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_ring.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_ring.c @@ -272,9 +272,12 @@ bool aq_ring_tx_clean(struct aq_ring_s *self) } } - if (unlikely(buff->is_eop)) - dev_kfree_skb_any(buff->skb); + if (unlikely(buff->is_eop)) { + ++self->stats.rx.packets; + self->stats.tx.bytes += buff->skb->len; + dev_kfree_skb_any(buff->skb); + } buff->pa = 0U; buff->eop_index = 0xffffU; self->sw_head = aq_ring_next_dx(self, self->sw_head); @@ -351,7 +354,8 @@ int aq_ring_rx_clean(struct aq_ring_s *self, err = 0; goto err_exit; } - if (buff->is_error || buff->is_cso_err) { + if (buff->is_error || + (buff->is_lro && buff->is_cso_err)) { buff_ = buff; do { next_ = buff_->next, diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_ring.h b/drivers/net/ethernet/aquantia/atlantic/aq_ring.h index 991e4d31b094..2c96f20f6289 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_ring.h +++ b/drivers/net/ethernet/aquantia/atlantic/aq_ring.h @@ -78,7 +78,8 @@ struct __packed aq_ring_buff_s { u32 is_cleaned:1; u32 is_error:1; u32 is_vlan:1; - u32 rsvd3:4; + u32 is_lro:1; + u32 rsvd3:3; u16 eop_index; u16 rsvd4; }; diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c index ec041f78d063..d20d91cdece8 100644 --- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c @@ -823,6 +823,8 @@ static int hw_atl_b0_hw_ring_rx_receive(struct aq_hw_s *self, } } + buff->is_lro = !!(HW_ATL_B0_RXD_WB_STAT2_RSCCNT & + rxd_wb->status); if (HW_ATL_B0_RXD_WB_STAT2_EOP & rxd_wb->status) { buff->len = rxd_wb->pkt_len % AQ_CFG_RX_FRAME_MAX; @@ -835,8 +837,7 @@ static int hw_atl_b0_hw_ring_rx_receive(struct aq_hw_s *self, rxd_wb->pkt_len > AQ_CFG_RX_FRAME_MAX ? AQ_CFG_RX_FRAME_MAX : rxd_wb->pkt_len; - if (HW_ATL_B0_RXD_WB_STAT2_RSCCNT & - rxd_wb->status) { + if (buff->is_lro) { /* LRO */ buff->next = rxd_wb->next_desc_ptr; ++ring->stats.rx.lro_packets; @@ -884,13 +885,16 @@ static int hw_atl_b0_hw_packet_filter_set(struct aq_hw_s *self, { struct aq_nic_cfg_s *cfg = self->aq_nic_cfg; unsigned int i = 0U; + u32 vlan_promisc; + u32 l2_promisc; - hw_atl_rpfl2promiscuous_mode_en_set(self, - IS_FILTER_ENABLED(IFF_PROMISC)); + l2_promisc = IS_FILTER_ENABLED(IFF_PROMISC) || + !!(cfg->priv_flags & BIT(AQ_HW_LOOPBACK_DMA_NET)); + vlan_promisc = l2_promisc || cfg->is_vlan_force_promisc; - hw_atl_rpf_vlan_prom_mode_en_set(self, - IS_FILTER_ENABLED(IFF_PROMISC) || - cfg->is_vlan_force_promisc); + hw_atl_rpfl2promiscuous_mode_en_set(self, l2_promisc); + + hw_atl_rpf_vlan_prom_mode_en_set(self, vlan_promisc); hw_atl_rpfl2multicast_flr_en_set(self, IS_FILTER_ENABLED(IFF_ALLMULTI) && @@ -1161,6 +1165,8 @@ static int hw_atl_b0_adj_sys_clock(struct aq_hw_s *self, s64 delta) { self->ptp_clk_offset += delta; + self->aq_fw_ops->adjust_ptp(self, self->ptp_clk_offset); + return 0; } @@ -1211,7 +1217,7 @@ static int hw_atl_b0_gpio_pulse(struct aq_hw_s *self, u32 index, fwreq.ptp_gpio_ctrl.index = index; fwreq.ptp_gpio_ctrl.period = period; /* Apply time offset */ - fwreq.ptp_gpio_ctrl.start = start - self->ptp_clk_offset; + fwreq.ptp_gpio_ctrl.start = start; size = sizeof(fwreq.msg_id) + sizeof(fwreq.ptp_gpio_ctrl); return self->aq_fw_ops->send_fw_request(self, &fwreq, size); diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils.c index f547baa6c954..354705f9bc49 100644 --- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils.c +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils.c @@ -22,6 +22,7 @@ #define HW_ATL_MIF_ADDR 0x0208U #define HW_ATL_MIF_VAL 0x020CU +#define HW_ATL_MPI_RPC_ADDR 0x0334U #define HW_ATL_RPC_CONTROL_ADR 0x0338U #define HW_ATL_RPC_STATE_ADR 0x033CU @@ -53,15 +54,14 @@ enum mcp_area { }; static int hw_atl_utils_ver_match(u32 ver_expected, u32 ver_actual); - static int hw_atl_utils_mpi_set_state(struct aq_hw_s *self, enum hal_atl_utils_fw_state_e state); - static u32 hw_atl_utils_get_mpi_mbox_tid(struct aq_hw_s *self); static u32 hw_atl_utils_mpi_get_state(struct aq_hw_s *self); static u32 hw_atl_utils_mif_cmd_get(struct aq_hw_s *self); static u32 hw_atl_utils_mif_addr_get(struct aq_hw_s *self); static u32 hw_atl_utils_rpc_state_get(struct aq_hw_s *self); +static u32 aq_fw1x_rpc_get(struct aq_hw_s *self); int hw_atl_utils_initfw(struct aq_hw_s *self, const struct aq_fw_ops **fw_ops) { @@ -476,6 +476,10 @@ static int hw_atl_utils_init_ucp(struct aq_hw_s *self, self, self->mbox_addr, self->mbox_addr != 0U, 1000U, 10000U); + err = readx_poll_timeout_atomic(aq_fw1x_rpc_get, self, + self->rpc_addr, + self->rpc_addr != 0U, + 1000U, 100000U); return err; } @@ -531,6 +535,12 @@ int hw_atl_utils_fw_rpc_wait(struct aq_hw_s *self, self, fw.val, sw.tid == fw.tid, 1000U, 100000U); + if (err < 0) + goto err_exit; + + err = aq_hw_err_from_flags(self); + if (err < 0) + goto err_exit; if (fw.len == 0xFFFFU) { err = hw_atl_utils_fw_rpc_call(self, sw.len); @@ -1025,6 +1035,11 @@ static u32 hw_atl_utils_rpc_state_get(struct aq_hw_s *self) return aq_hw_read_reg(self, HW_ATL_RPC_STATE_ADR); } +static u32 aq_fw1x_rpc_get(struct aq_hw_s *self) +{ + return aq_hw_read_reg(self, HW_ATL_MPI_RPC_ADDR); +} + const struct aq_fw_ops aq_fw_1x_ops = { .init = hw_atl_utils_mpi_create, .deinit = hw_atl_fw1x_deinit, diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils_fw2x.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils_fw2x.c index 97ebf849695f..77a4ed64830f 100644 --- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils_fw2x.c +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils_fw2x.c @@ -30,6 +30,9 @@ #define HW_ATL_FW3X_EXT_CONTROL_ADDR 0x378 #define HW_ATL_FW3X_EXT_STATE_ADDR 0x37c +#define HW_ATL_FW3X_PTP_ADJ_LSW_ADDR 0x50a0 +#define HW_ATL_FW3X_PTP_ADJ_MSW_ADDR 0x50a4 + #define HW_ATL_FW2X_CAP_PAUSE BIT(CAPS_HI_PAUSE) #define HW_ATL_FW2X_CAP_ASYM_PAUSE BIT(CAPS_HI_ASYMMETRIC_PAUSE) #define HW_ATL_FW2X_CAP_SLEEP_PROXY BIT(CAPS_HI_SLEEP_PROXY) @@ -475,6 +478,14 @@ static void aq_fw3x_enable_ptp(struct aq_hw_s *self, int enable) aq_hw_write_reg(self, HW_ATL_FW3X_EXT_CONTROL_ADDR, ptp_opts); } +static void aq_fw3x_adjust_ptp(struct aq_hw_s *self, uint64_t adj) +{ + aq_hw_write_reg(self, HW_ATL_FW3X_PTP_ADJ_LSW_ADDR, + (adj >> 0) & 0xffffffff); + aq_hw_write_reg(self, HW_ATL_FW3X_PTP_ADJ_MSW_ADDR, + (adj >> 32) & 0xffffffff); +} + static int aq_fw2x_led_control(struct aq_hw_s *self, u32 mode) { if (self->fw_ver_actual < HW_ATL_FW_VER_LED) @@ -633,4 +644,5 @@ const struct aq_fw_ops aq_fw_2x_ops = { .enable_ptp = aq_fw3x_enable_ptp, .led_control = aq_fw2x_led_control, .set_phyloopback = aq_fw2x_set_phyloopback, + .adjust_ptp = aq_fw3x_adjust_ptp, }; diff --git a/drivers/net/ethernet/broadcom/bcmsysport.c b/drivers/net/ethernet/broadcom/bcmsysport.c index e0611cba87f9..15b31cddc054 100644 --- a/drivers/net/ethernet/broadcom/bcmsysport.c +++ b/drivers/net/ethernet/broadcom/bcmsysport.c @@ -2135,7 +2135,7 @@ static int bcm_sysport_rule_set(struct bcm_sysport_priv *priv, return -ENOSPC; index = find_first_zero_bit(priv->filters, RXCHK_BRCM_TAG_MAX); - if (index > RXCHK_BRCM_TAG_MAX) + if (index >= RXCHK_BRCM_TAG_MAX) return -ENOSPC; /* Location is the classification ID, and index is the position diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 597e6fd5bfea..d28b406a26b1 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -6880,12 +6880,12 @@ skip_rdma: } ena |= FUNC_BACKING_STORE_CFG_REQ_DFLT_ENABLES; rc = bnxt_hwrm_func_backing_store_cfg(bp, ena); - if (rc) + if (rc) { netdev_err(bp->dev, "Failed configuring context mem, rc = %d.\n", rc); - else - ctx->flags |= BNXT_CTX_FLAG_INITED; - + return rc; + } + ctx->flags |= BNXT_CTX_FLAG_INITED; return 0; } @@ -7406,14 +7406,22 @@ static int bnxt_hwrm_port_qstats_ext(struct bnxt *bp) pri2cos = &resp2->pri0_cos_queue_id; for (i = 0; i < 8; i++) { u8 queue_id = pri2cos[i]; + u8 queue_idx; + /* Per port queue IDs start from 0, 10, 20, etc */ + queue_idx = queue_id % 10; + if (queue_idx > BNXT_MAX_QUEUE) { + bp->pri2cos_valid = false; + goto qstats_done; + } for (j = 0; j < bp->max_q; j++) { if (bp->q_ids[j] == queue_id) - bp->pri2cos[i] = j; + bp->pri2cos_idx[i] = queue_idx; } } bp->pri2cos_valid = 1; } +qstats_done: mutex_unlock(&bp->hwrm_cmd_lock); return rc; } @@ -10982,13 +10990,13 @@ static int bnxt_change_mtu(struct net_device *dev, int new_mtu) struct bnxt *bp = netdev_priv(dev); if (netif_running(dev)) - bnxt_close_nic(bp, false, false); + bnxt_close_nic(bp, true, false); dev->mtu = new_mtu; bnxt_set_ring_params(bp); if (netif_running(dev)) - return bnxt_open_nic(bp, false, false); + return bnxt_open_nic(bp, true, false); return 0; } @@ -11252,7 +11260,7 @@ static void bnxt_cfg_ntp_filters(struct bnxt *bp) } } if (test_and_clear_bit(BNXT_HWRM_PF_UNLOAD_SP_EVENT, &bp->sp_event)) - netdev_info(bp->dev, "Receive PF driver unload event!"); + netdev_info(bp->dev, "Receive PF driver unload event!\n"); } #else @@ -11669,6 +11677,10 @@ static int bnxt_set_dflt_rings(struct bnxt *bp, bool sh) bp->rx_nr_rings++; bp->cp_nr_rings++; } + if (rc) { + bp->tx_nr_rings = 0; + bp->rx_nr_rings = 0; + } return rc; } @@ -11759,7 +11771,7 @@ static int bnxt_pcie_dsn_get(struct bnxt *bp, u8 dsn[]) u32 dw; if (!pos) { - netdev_info(bp->dev, "Unable do read adapter's DSN"); + netdev_info(bp->dev, "Unable do read adapter's DSN\n"); return -EOPNOTSUPP; } @@ -11786,6 +11798,14 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) if (version_printed++ == 0) pr_info("%s", version); + /* Clear any pending DMA transactions from crash kernel + * while loading driver in capture kernel. + */ + if (is_kdump_kernel()) { + pci_clear_master(pdev); + pcie_flr(pdev); + } + max_irqs = bnxt_get_max_irq(pdev); dev = alloc_etherdev_mq(sizeof(*bp), max_irqs); if (!dev) @@ -11954,12 +11974,12 @@ init_err_pci_clean: bnxt_hwrm_func_drv_unrgtr(bp); bnxt_free_hwrm_short_cmd_req(bp); bnxt_free_hwrm_resources(bp); - bnxt_free_ctx_mem(bp); - kfree(bp->ctx); - bp->ctx = NULL; kfree(bp->fw_health); bp->fw_health = NULL; bnxt_cleanup_pci(bp); + bnxt_free_ctx_mem(bp); + kfree(bp->ctx); + bp->ctx = NULL; init_err_free: free_netdev(dev); @@ -11983,10 +12003,10 @@ static void bnxt_shutdown(struct pci_dev *pdev) dev_close(dev); bnxt_ulp_shutdown(bp); + bnxt_clear_int_mode(bp); + pci_disable_device(pdev); if (system_state == SYSTEM_POWER_OFF) { - bnxt_clear_int_mode(bp); - pci_disable_device(pdev); pci_wake_from_d3(pdev, bp->wol); pci_set_power_state(pdev, PCI_D3hot); } diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h index cabef0b4f5fb..63b170658532 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h @@ -1716,7 +1716,7 @@ struct bnxt { u16 fw_rx_stats_ext_size; u16 fw_tx_stats_ext_size; u16 hw_ring_stats_size; - u8 pri2cos[8]; + u8 pri2cos_idx[8]; u8 pri2cos_valid; u16 hwrm_max_req_len; diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_dcb.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_dcb.c index fb6f30d0d1d0..b1511bcffb1b 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_dcb.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_dcb.c @@ -479,24 +479,26 @@ static int bnxt_dcbnl_ieee_getets(struct net_device *dev, struct ieee_ets *ets) { struct bnxt *bp = netdev_priv(dev); struct ieee_ets *my_ets = bp->ieee_ets; + int rc; ets->ets_cap = bp->max_tc; if (!my_ets) { - int rc; - if (bp->dcbx_cap & DCB_CAP_DCBX_HOST) return 0; my_ets = kzalloc(sizeof(*my_ets), GFP_KERNEL); if (!my_ets) - return 0; + return -ENOMEM; rc = bnxt_hwrm_queue_cos2bw_qcfg(bp, my_ets); if (rc) - return 0; + goto error; rc = bnxt_hwrm_queue_pri2cos_qcfg(bp, my_ets); if (rc) - return 0; + goto error; + + /* cache result */ + bp->ieee_ets = my_ets; } ets->cbs = my_ets->cbs; @@ -505,6 +507,9 @@ static int bnxt_dcbnl_ieee_getets(struct net_device *dev, struct ieee_ets *ets) memcpy(ets->tc_tsa, my_ets->tc_tsa, sizeof(ets->tc_tsa)); memcpy(ets->prio_tc, my_ets->prio_tc, sizeof(ets->prio_tc)); return 0; +error: + kfree(my_ets); + return rc; } static int bnxt_dcbnl_ieee_setets(struct net_device *dev, struct ieee_ets *ets) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c index eec0168330b7..d3c93ccee86a 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c @@ -641,14 +641,14 @@ static int bnxt_dl_params_register(struct bnxt *bp) rc = devlink_params_register(bp->dl, bnxt_dl_params, ARRAY_SIZE(bnxt_dl_params)); if (rc) { - netdev_warn(bp->dev, "devlink_params_register failed. rc=%d", + netdev_warn(bp->dev, "devlink_params_register failed. rc=%d\n", rc); return rc; } rc = devlink_port_params_register(&bp->dl_port, bnxt_dl_port_params, ARRAY_SIZE(bnxt_dl_port_params)); if (rc) { - netdev_err(bp->dev, "devlink_port_params_register failed"); + netdev_err(bp->dev, "devlink_port_params_register failed\n"); devlink_params_unregister(bp->dl, bnxt_dl_params, ARRAY_SIZE(bnxt_dl_params)); return rc; @@ -679,7 +679,7 @@ int bnxt_dl_register(struct bnxt *bp) else dl = devlink_alloc(&bnxt_vf_dl_ops, sizeof(struct bnxt_dl)); if (!dl) { - netdev_warn(bp->dev, "devlink_alloc failed"); + netdev_warn(bp->dev, "devlink_alloc failed\n"); return -ENOMEM; } @@ -692,7 +692,7 @@ int bnxt_dl_register(struct bnxt *bp) rc = devlink_register(dl, &bp->pdev->dev); if (rc) { - netdev_warn(bp->dev, "devlink_register failed. rc=%d", rc); + netdev_warn(bp->dev, "devlink_register failed. rc=%d\n", rc); goto err_dl_free; } @@ -704,7 +704,7 @@ int bnxt_dl_register(struct bnxt *bp) sizeof(bp->dsn)); rc = devlink_port_register(dl, &bp->dl_port, bp->pf.port_id); if (rc) { - netdev_err(bp->dev, "devlink_port_register failed"); + netdev_err(bp->dev, "devlink_port_register failed\n"); goto err_dl_unreg; } diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c index 6171fa8b3677..3f8a1ded662a 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c @@ -589,25 +589,25 @@ skip_ring_stats: if (bp->pri2cos_valid) { for (i = 0; i < 8; i++, j++) { long n = bnxt_rx_bytes_pri_arr[i].base_off + - bp->pri2cos[i]; + bp->pri2cos_idx[i]; buf[j] = le64_to_cpu(*(rx_port_stats_ext + n)); } for (i = 0; i < 8; i++, j++) { long n = bnxt_rx_pkts_pri_arr[i].base_off + - bp->pri2cos[i]; + bp->pri2cos_idx[i]; buf[j] = le64_to_cpu(*(rx_port_stats_ext + n)); } for (i = 0; i < 8; i++, j++) { long n = bnxt_tx_bytes_pri_arr[i].base_off + - bp->pri2cos[i]; + bp->pri2cos_idx[i]; buf[j] = le64_to_cpu(*(tx_port_stats_ext + n)); } for (i = 0; i < 8; i++, j++) { long n = bnxt_tx_pkts_pri_arr[i].base_off + - bp->pri2cos[i]; + bp->pri2cos_idx[i]; buf[j] = le64_to_cpu(*(tx_port_stats_ext + n)); } @@ -2007,8 +2007,8 @@ int bnxt_flash_package_from_file(struct net_device *dev, const char *filename, struct hwrm_nvm_install_update_output *resp = bp->hwrm_cmd_resp_addr; struct hwrm_nvm_install_update_input install = {0}; const struct firmware *fw; - int rc, hwrm_err = 0; u32 item_len; + int rc = 0; u16 index; bnxt_hwrm_fw_set_time(bp); @@ -2028,7 +2028,7 @@ int bnxt_flash_package_from_file(struct net_device *dev, const char *filename, } if (fw->size > item_len) { - netdev_err(dev, "PKG insufficient update area in nvram: %lu", + netdev_err(dev, "PKG insufficient update area in nvram: %lu\n", (unsigned long)fw->size); rc = -EFBIG; } else { @@ -2052,15 +2052,14 @@ int bnxt_flash_package_from_file(struct net_device *dev, const char *filename, memcpy(kmem, fw->data, fw->size); modify.host_src_addr = cpu_to_le64(dma_handle); - hwrm_err = hwrm_send_message(bp, &modify, - sizeof(modify), - FLASH_PACKAGE_TIMEOUT); + rc = hwrm_send_message(bp, &modify, sizeof(modify), + FLASH_PACKAGE_TIMEOUT); dma_free_coherent(&bp->pdev->dev, fw->size, kmem, dma_handle); } } release_firmware(fw); - if (rc || hwrm_err) + if (rc) goto err_exit; if ((install_type & 0xffff) == 0) @@ -2069,20 +2068,19 @@ int bnxt_flash_package_from_file(struct net_device *dev, const char *filename, install.install_type = cpu_to_le32(install_type); mutex_lock(&bp->hwrm_cmd_lock); - hwrm_err = _hwrm_send_message(bp, &install, sizeof(install), - INSTALL_PACKAGE_TIMEOUT); - if (hwrm_err) { + rc = _hwrm_send_message(bp, &install, sizeof(install), + INSTALL_PACKAGE_TIMEOUT); + if (rc) { u8 error_code = ((struct hwrm_err_output *)resp)->cmd_err; if (resp->error_code && error_code == NVM_INSTALL_UPDATE_CMD_ERR_CODE_FRAG_ERR) { install.flags |= cpu_to_le16( NVM_INSTALL_UPDATE_REQ_FLAGS_ALLOWED_TO_DEFRAG); - hwrm_err = _hwrm_send_message(bp, &install, - sizeof(install), - INSTALL_PACKAGE_TIMEOUT); + rc = _hwrm_send_message(bp, &install, sizeof(install), + INSTALL_PACKAGE_TIMEOUT); } - if (hwrm_err) + if (rc) goto flash_pkg_exit; } @@ -2094,7 +2092,7 @@ int bnxt_flash_package_from_file(struct net_device *dev, const char *filename, flash_pkg_exit: mutex_unlock(&bp->hwrm_cmd_lock); err_exit: - if (hwrm_err == -EACCES) + if (rc == -EACCES) bnxt_print_admin_err(bp); return rc; } @@ -3338,7 +3336,7 @@ err: kfree(coredump.data); *dump_len += sizeof(struct bnxt_coredump_record); if (rc == -ENOBUFS) - netdev_err(bp->dev, "Firmware returned large coredump buffer"); + netdev_err(bp->dev, "Firmware returned large coredump buffer\n"); return rc; } diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c index 0cc6ec51f45f..9bec256b0934 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c @@ -50,7 +50,7 @@ static u16 bnxt_flow_get_dst_fid(struct bnxt *pf_bp, struct net_device *dev) /* check if dev belongs to the same switch */ if (!netdev_port_same_parent_id(pf_bp->dev, dev)) { - netdev_info(pf_bp->dev, "dev(ifindex=%d) not on same switch", + netdev_info(pf_bp->dev, "dev(ifindex=%d) not on same switch\n", dev->ifindex); return BNXT_FID_INVALID; } @@ -70,7 +70,7 @@ static int bnxt_tc_parse_redir(struct bnxt *bp, struct net_device *dev = act->dev; if (!dev) { - netdev_info(bp->dev, "no dev in mirred action"); + netdev_info(bp->dev, "no dev in mirred action\n"); return -EINVAL; } @@ -106,7 +106,7 @@ static int bnxt_tc_parse_tunnel_set(struct bnxt *bp, const struct ip_tunnel_key *tun_key = &tun_info->key; if (ip_tunnel_info_af(tun_info) != AF_INET) { - netdev_info(bp->dev, "only IPv4 tunnel-encap is supported"); + netdev_info(bp->dev, "only IPv4 tunnel-encap is supported\n"); return -EOPNOTSUPP; } @@ -295,7 +295,7 @@ static int bnxt_tc_parse_actions(struct bnxt *bp, int i, rc; if (!flow_action_has_entries(flow_action)) { - netdev_info(bp->dev, "no actions"); + netdev_info(bp->dev, "no actions\n"); return -EINVAL; } @@ -370,7 +370,7 @@ static int bnxt_tc_parse_flow(struct bnxt *bp, /* KEY_CONTROL and KEY_BASIC are needed for forming a meaningful key */ if ((dissector->used_keys & BIT(FLOW_DISSECTOR_KEY_CONTROL)) == 0 || (dissector->used_keys & BIT(FLOW_DISSECTOR_KEY_BASIC)) == 0) { - netdev_info(bp->dev, "cannot form TC key: used_keys = 0x%x", + netdev_info(bp->dev, "cannot form TC key: used_keys = 0x%x\n", dissector->used_keys); return -EOPNOTSUPP; } @@ -508,7 +508,7 @@ static int bnxt_hwrm_cfa_flow_free(struct bnxt *bp, rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); if (rc) - netdev_info(bp->dev, "%s: Error rc=%d", __func__, rc); + netdev_info(bp->dev, "%s: Error rc=%d\n", __func__, rc); return rc; } @@ -841,7 +841,7 @@ static int hwrm_cfa_decap_filter_alloc(struct bnxt *bp, resp = bnxt_get_hwrm_resp_addr(bp, &req); *decap_filter_handle = resp->decap_filter_id; } else { - netdev_info(bp->dev, "%s: Error rc=%d", __func__, rc); + netdev_info(bp->dev, "%s: Error rc=%d\n", __func__, rc); } mutex_unlock(&bp->hwrm_cmd_lock); @@ -859,7 +859,7 @@ static int hwrm_cfa_decap_filter_free(struct bnxt *bp, rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); if (rc) - netdev_info(bp->dev, "%s: Error rc=%d", __func__, rc); + netdev_info(bp->dev, "%s: Error rc=%d\n", __func__, rc); return rc; } @@ -906,7 +906,7 @@ static int hwrm_cfa_encap_record_alloc(struct bnxt *bp, resp = bnxt_get_hwrm_resp_addr(bp, &req); *encap_record_handle = resp->encap_record_id; } else { - netdev_info(bp->dev, "%s: Error rc=%d", __func__, rc); + netdev_info(bp->dev, "%s: Error rc=%d\n", __func__, rc); } mutex_unlock(&bp->hwrm_cmd_lock); @@ -924,7 +924,7 @@ static int hwrm_cfa_encap_record_free(struct bnxt *bp, rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); if (rc) - netdev_info(bp->dev, "%s: Error rc=%d", __func__, rc); + netdev_info(bp->dev, "%s: Error rc=%d\n", __func__, rc); return rc; } @@ -943,7 +943,7 @@ static int bnxt_tc_put_l2_node(struct bnxt *bp, tc_info->l2_ht_params); if (rc) netdev_err(bp->dev, - "Error: %s: rhashtable_remove_fast: %d", + "Error: %s: rhashtable_remove_fast: %d\n", __func__, rc); kfree_rcu(l2_node, rcu); } @@ -972,7 +972,7 @@ bnxt_tc_get_l2_node(struct bnxt *bp, struct rhashtable *l2_table, if (rc) { kfree_rcu(l2_node, rcu); netdev_err(bp->dev, - "Error: %s: rhashtable_insert_fast: %d", + "Error: %s: rhashtable_insert_fast: %d\n", __func__, rc); return NULL; } @@ -1031,7 +1031,7 @@ static bool bnxt_tc_can_offload(struct bnxt *bp, struct bnxt_tc_flow *flow) if ((flow->flags & BNXT_TC_FLOW_FLAGS_PORTS) && (flow->l4_key.ip_proto != IPPROTO_TCP && flow->l4_key.ip_proto != IPPROTO_UDP)) { - netdev_info(bp->dev, "Cannot offload non-TCP/UDP (%d) ports", + netdev_info(bp->dev, "Cannot offload non-TCP/UDP (%d) ports\n", flow->l4_key.ip_proto); return false; } @@ -1088,7 +1088,7 @@ static int bnxt_tc_put_tunnel_node(struct bnxt *bp, rc = rhashtable_remove_fast(tunnel_table, &tunnel_node->node, *ht_params); if (rc) { - netdev_err(bp->dev, "rhashtable_remove_fast rc=%d", rc); + netdev_err(bp->dev, "rhashtable_remove_fast rc=%d\n", rc); rc = -1; } kfree_rcu(tunnel_node, rcu); @@ -1129,7 +1129,7 @@ bnxt_tc_get_tunnel_node(struct bnxt *bp, struct rhashtable *tunnel_table, tunnel_node->refcount++; return tunnel_node; err: - netdev_info(bp->dev, "error rc=%d", rc); + netdev_info(bp->dev, "error rc=%d\n", rc); return NULL; } @@ -1187,7 +1187,7 @@ static void bnxt_tc_put_decap_l2_node(struct bnxt *bp, &decap_l2_node->node, tc_info->decap_l2_ht_params); if (rc) - netdev_err(bp->dev, "rhashtable_remove_fast rc=%d", rc); + netdev_err(bp->dev, "rhashtable_remove_fast rc=%d\n", rc); kfree_rcu(decap_l2_node, rcu); } } @@ -1227,7 +1227,7 @@ static int bnxt_tc_resolve_tunnel_hdrs(struct bnxt *bp, rt = ip_route_output_key(dev_net(real_dst_dev), &flow); if (IS_ERR(rt)) { - netdev_info(bp->dev, "no route to %pI4b", &flow.daddr); + netdev_info(bp->dev, "no route to %pI4b\n", &flow.daddr); return -EOPNOTSUPP; } @@ -1241,7 +1241,7 @@ static int bnxt_tc_resolve_tunnel_hdrs(struct bnxt *bp, if (vlan->real_dev != real_dst_dev) { netdev_info(bp->dev, - "dst_dev(%s) doesn't use PF-if(%s)", + "dst_dev(%s) doesn't use PF-if(%s)\n", netdev_name(dst_dev), netdev_name(real_dst_dev)); rc = -EOPNOTSUPP; @@ -1253,7 +1253,7 @@ static int bnxt_tc_resolve_tunnel_hdrs(struct bnxt *bp, #endif } else if (dst_dev != real_dst_dev) { netdev_info(bp->dev, - "dst_dev(%s) for %pI4b is not PF-if(%s)", + "dst_dev(%s) for %pI4b is not PF-if(%s)\n", netdev_name(dst_dev), &flow.daddr, netdev_name(real_dst_dev)); rc = -EOPNOTSUPP; @@ -1262,7 +1262,7 @@ static int bnxt_tc_resolve_tunnel_hdrs(struct bnxt *bp, nbr = dst_neigh_lookup(&rt->dst, &flow.daddr); if (!nbr) { - netdev_info(bp->dev, "can't lookup neighbor for %pI4b", + netdev_info(bp->dev, "can't lookup neighbor for %pI4b\n", &flow.daddr); rc = -EOPNOTSUPP; goto put_rt; @@ -1472,7 +1472,7 @@ static int __bnxt_tc_del_flow(struct bnxt *bp, rc = rhashtable_remove_fast(&tc_info->flow_table, &flow_node->node, tc_info->flow_ht_params); if (rc) - netdev_err(bp->dev, "Error: %s: rhashtable_remove_fast rc=%d", + netdev_err(bp->dev, "Error: %s: rhashtable_remove_fast rc=%d\n", __func__, rc); kfree_rcu(flow_node, rcu); @@ -1587,7 +1587,7 @@ unlock: free_node: kfree_rcu(new_node, rcu); done: - netdev_err(bp->dev, "Error: %s: cookie=0x%lx error=%d", + netdev_err(bp->dev, "Error: %s: cookie=0x%lx error=%d\n", __func__, tc_flow_cmd->cookie, rc); return rc; } @@ -1700,7 +1700,7 @@ bnxt_hwrm_cfa_flow_stats_get(struct bnxt *bp, int num_flows, le64_to_cpu(resp_bytes[i]); } } else { - netdev_info(bp->dev, "error rc=%d", rc); + netdev_info(bp->dev, "error rc=%d\n", rc); } mutex_unlock(&bp->hwrm_cmd_lock); @@ -1970,7 +1970,7 @@ static int bnxt_tc_indr_block_event(struct notifier_block *nb, bp); if (rc) netdev_info(bp->dev, - "Failed to register indirect blk: dev: %s", + "Failed to register indirect blk: dev: %s\n", netdev->name); break; case NETDEV_UNREGISTER: diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_vfr.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_vfr.c index b010b34cdaf8..6f2faf81c1ae 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_vfr.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_vfr.c @@ -43,7 +43,7 @@ static int hwrm_cfa_vfr_alloc(struct bnxt *bp, u16 vf_idx, netdev_dbg(bp->dev, "tx_cfa_action=0x%x, rx_cfa_code=0x%x", *tx_cfa_action, *rx_cfa_code); } else { - netdev_info(bp->dev, "%s error rc=%d", __func__, rc); + netdev_info(bp->dev, "%s error rc=%d\n", __func__, rc); } mutex_unlock(&bp->hwrm_cmd_lock); @@ -60,7 +60,7 @@ static int hwrm_cfa_vfr_free(struct bnxt *bp, u16 vf_idx) rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); if (rc) - netdev_info(bp->dev, "%s error rc=%d", __func__, rc); + netdev_info(bp->dev, "%s error rc=%d\n", __func__, rc); return rc; } @@ -465,7 +465,7 @@ static int bnxt_vf_reps_create(struct bnxt *bp) return 0; err: - netdev_info(bp->dev, "%s error=%d", __func__, rc); + netdev_info(bp->dev, "%s error=%d\n", __func__, rc); kfree(cfa_code_map); __bnxt_vf_reps_destroy(bp); return rc; @@ -488,7 +488,7 @@ int bnxt_dl_eswitch_mode_set(struct devlink *devlink, u16 mode, mutex_lock(&bp->sriov_lock); if (bp->eswitch_mode == mode) { - netdev_info(bp->dev, "already in %s eswitch mode", + netdev_info(bp->dev, "already in %s eswitch mode\n", mode == DEVLINK_ESWITCH_MODE_LEGACY ? "legacy" : "switchdev"); rc = -EINVAL; @@ -508,7 +508,7 @@ int bnxt_dl_eswitch_mode_set(struct devlink *devlink, u16 mode, } if (pci_num_vf(bp->pdev) == 0) { - netdev_info(bp->dev, "Enable VFs before setting switchdev mode"); + netdev_info(bp->dev, "Enable VFs before setting switchdev mode\n"); rc = -EPERM; goto done; } diff --git a/drivers/net/ethernet/broadcom/cnic_defs.h b/drivers/net/ethernet/broadcom/cnic_defs.h index b38499774071..99e2c6d4d8c3 100644 --- a/drivers/net/ethernet/broadcom/cnic_defs.h +++ b/drivers/net/ethernet/broadcom/cnic_defs.h @@ -543,13 +543,13 @@ struct l4_kwq_update_pg { #define L4_KWQ_UPDATE_PG_RESERVERD2_SHIFT 2 #endif #if defined(__BIG_ENDIAN) - u16 reserverd3; + u16 reserved3; u8 da0; u8 da1; #elif defined(__LITTLE_ENDIAN) u8 da1; u8 da0; - u16 reserverd3; + u16 reserved3; #endif #if defined(__BIG_ENDIAN) u8 da2; diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c index e50a15397e11..1d678bee2cc9 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c +++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c @@ -94,12 +94,6 @@ static inline void dmadesc_set_length_status(struct bcmgenet_priv *priv, bcmgenet_writel(value, d + DMA_DESC_LENGTH_STATUS); } -static inline u32 dmadesc_get_length_status(struct bcmgenet_priv *priv, - void __iomem *d) -{ - return bcmgenet_readl(d + DMA_DESC_LENGTH_STATUS); -} - static inline void dmadesc_set_addr(struct bcmgenet_priv *priv, void __iomem *d, dma_addr_t addr) @@ -508,61 +502,6 @@ static int bcmgenet_set_link_ksettings(struct net_device *dev, return phy_ethtool_ksettings_set(dev->phydev, cmd); } -static void bcmgenet_set_rx_csum(struct net_device *dev, - netdev_features_t wanted) -{ - struct bcmgenet_priv *priv = netdev_priv(dev); - u32 rbuf_chk_ctrl; - bool rx_csum_en; - - rx_csum_en = !!(wanted & NETIF_F_RXCSUM); - - rbuf_chk_ctrl = bcmgenet_rbuf_readl(priv, RBUF_CHK_CTRL); - - /* enable rx checksumming */ - if (rx_csum_en) - rbuf_chk_ctrl |= RBUF_RXCHK_EN | RBUF_L3_PARSE_DIS; - else - rbuf_chk_ctrl &= ~RBUF_RXCHK_EN; - priv->desc_rxchk_en = rx_csum_en; - - /* If UniMAC forwards CRC, we need to skip over it to get - * a valid CHK bit to be set in the per-packet status word - */ - if (rx_csum_en && priv->crc_fwd_en) - rbuf_chk_ctrl |= RBUF_SKIP_FCS; - else - rbuf_chk_ctrl &= ~RBUF_SKIP_FCS; - - bcmgenet_rbuf_writel(priv, rbuf_chk_ctrl, RBUF_CHK_CTRL); -} - -static void bcmgenet_set_tx_csum(struct net_device *dev, - netdev_features_t wanted) -{ - struct bcmgenet_priv *priv = netdev_priv(dev); - bool desc_64b_en; - u32 tbuf_ctrl, rbuf_ctrl; - - tbuf_ctrl = bcmgenet_tbuf_ctrl_get(priv); - rbuf_ctrl = bcmgenet_rbuf_readl(priv, RBUF_CTRL); - - desc_64b_en = !!(wanted & NETIF_F_HW_CSUM); - - /* enable 64 bytes descriptor in both directions (RBUF and TBUF) */ - if (desc_64b_en) { - tbuf_ctrl |= RBUF_64B_EN; - rbuf_ctrl |= RBUF_64B_EN; - } else { - tbuf_ctrl &= ~RBUF_64B_EN; - rbuf_ctrl &= ~RBUF_64B_EN; - } - priv->desc_64b_en = desc_64b_en; - - bcmgenet_tbuf_ctrl_set(priv, tbuf_ctrl); - bcmgenet_rbuf_writel(priv, rbuf_ctrl, RBUF_CTRL); -} - static int bcmgenet_set_features(struct net_device *dev, netdev_features_t features) { @@ -578,9 +517,6 @@ static int bcmgenet_set_features(struct net_device *dev, reg = bcmgenet_umac_readl(priv, UMAC_CMD); priv->crc_fwd_en = !!(reg & CMD_CRC_FWD); - bcmgenet_set_tx_csum(dev, features); - bcmgenet_set_rx_csum(dev, features); - clk_disable_unprepare(priv->clk); return ret; @@ -1475,8 +1411,8 @@ static void bcmgenet_tx_reclaim_all(struct net_device *dev) /* Reallocate the SKB to put enough headroom in front of it and insert * the transmit checksum offsets in the descriptors */ -static struct sk_buff *bcmgenet_put_tx_csum(struct net_device *dev, - struct sk_buff *skb) +static struct sk_buff *bcmgenet_add_tsb(struct net_device *dev, + struct sk_buff *skb) { struct bcmgenet_priv *priv = netdev_priv(dev); struct status_64 *status = NULL; @@ -1590,13 +1526,11 @@ static netdev_tx_t bcmgenet_xmit(struct sk_buff *skb, struct net_device *dev) */ GENET_CB(skb)->bytes_sent = skb->len; - /* set the SKB transmit checksum */ - if (priv->desc_64b_en) { - skb = bcmgenet_put_tx_csum(dev, skb); - if (!skb) { - ret = NETDEV_TX_OK; - goto out; - } + /* add the Transmit Status Block */ + skb = bcmgenet_add_tsb(dev, skb); + if (!skb) { + ret = NETDEV_TX_OK; + goto out; } for (i = 0; i <= nr_frags; i++) { @@ -1775,6 +1709,9 @@ static unsigned int bcmgenet_desc_rx(struct bcmgenet_rx_ring *ring, while ((rxpktprocessed < rxpkttoprocess) && (rxpktprocessed < budget)) { + struct status_64 *status; + __be16 rx_csum; + cb = &priv->rx_cbs[ring->read_ptr]; skb = bcmgenet_rx_refill(priv, cb); @@ -1783,20 +1720,12 @@ static unsigned int bcmgenet_desc_rx(struct bcmgenet_rx_ring *ring, goto next; } - if (!priv->desc_64b_en) { - dma_length_status = - dmadesc_get_length_status(priv, cb->bd_addr); - } else { - struct status_64 *status; - __be16 rx_csum; - - status = (struct status_64 *)skb->data; - dma_length_status = status->length_status; + status = (struct status_64 *)skb->data; + dma_length_status = status->length_status; + if (dev->features & NETIF_F_RXCSUM) { rx_csum = (__force __be16)(status->rx_csum & 0xffff); - if (priv->desc_rxchk_en) { - skb->csum = (__force __wsum)ntohs(rx_csum); - skb->ip_summed = CHECKSUM_COMPLETE; - } + skb->csum = (__force __wsum)ntohs(rx_csum); + skb->ip_summed = CHECKSUM_COMPLETE; } /* DMA flags and length are still valid no matter how @@ -1840,14 +1769,10 @@ static unsigned int bcmgenet_desc_rx(struct bcmgenet_rx_ring *ring, } /* error packet */ skb_put(skb, len); - if (priv->desc_64b_en) { - skb_pull(skb, 64); - len -= 64; - } - /* remove hardware 2bytes added for IP alignment */ - skb_pull(skb, 2); - len -= 2; + /* remove RSB and hardware 2bytes added for IP alignment */ + skb_pull(skb, 66); + len -= 66; if (priv->crc_fwd_en) { skb_trim(skb, len - ETH_FCS_LEN); @@ -1965,6 +1890,8 @@ static void umac_enable_set(struct bcmgenet_priv *priv, u32 mask, bool enable) u32 reg; reg = bcmgenet_umac_readl(priv, UMAC_CMD); + if (reg & CMD_SW_RESET) + return; if (enable) reg |= mask; else @@ -1984,11 +1911,9 @@ static void reset_umac(struct bcmgenet_priv *priv) bcmgenet_rbuf_ctrl_set(priv, 0); udelay(10); - /* disable MAC while updating its registers */ - bcmgenet_umac_writel(priv, 0, UMAC_CMD); - - /* issue soft reset with (rg)mii loopback to ensure a stable rxclk */ - bcmgenet_umac_writel(priv, CMD_SW_RESET | CMD_LCL_LOOP_EN, UMAC_CMD); + /* issue soft reset and disable MAC while updating its registers */ + bcmgenet_umac_writel(priv, CMD_SW_RESET, UMAC_CMD); + udelay(2); } static void bcmgenet_intr_disable(struct bcmgenet_priv *priv) @@ -2038,11 +1963,28 @@ static void init_umac(struct bcmgenet_priv *priv) bcmgenet_umac_writel(priv, ENET_MAX_MTU_SIZE, UMAC_MAX_FRAME_LEN); - /* init rx registers, enable ip header optimization */ + /* init tx registers, enable TSB */ + reg = bcmgenet_tbuf_ctrl_get(priv); + reg |= TBUF_64B_EN; + bcmgenet_tbuf_ctrl_set(priv, reg); + + /* init rx registers, enable ip header optimization and RSB */ reg = bcmgenet_rbuf_readl(priv, RBUF_CTRL); - reg |= RBUF_ALIGN_2B; + reg |= RBUF_ALIGN_2B | RBUF_64B_EN; bcmgenet_rbuf_writel(priv, reg, RBUF_CTRL); + /* enable rx checksumming */ + reg = bcmgenet_rbuf_readl(priv, RBUF_CHK_CTRL); + reg |= RBUF_RXCHK_EN | RBUF_L3_PARSE_DIS; + /* If UniMAC forwards CRC, we need to skip over it to get + * a valid CHK bit to be set in the per-packet status word + */ + if (priv->crc_fwd_en) + reg |= RBUF_SKIP_FCS; + else + reg &= ~RBUF_SKIP_FCS; + bcmgenet_rbuf_writel(priv, reg, RBUF_CHK_CTRL); + if (!GENET_IS_V1(priv) && !GENET_IS_V2(priv)) bcmgenet_rbuf_writel(priv, 1, RBUF_TBUF_SIZE_CTRL); diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.h b/drivers/net/ethernet/broadcom/genet/bcmgenet.h index 61a6fe9f4cec..daf8fb2c39b6 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmgenet.h +++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.h @@ -273,6 +273,7 @@ struct bcmgenet_mib_counters { #define RBUF_FLTR_LEN_SHIFT 8 #define TBUF_CTRL 0x00 +#define TBUF_64B_EN (1 << 0) #define TBUF_BP_MC 0x0C #define TBUF_ENERGY_CTRL 0x14 #define TBUF_EEE_EN (1 << 0) @@ -662,8 +663,6 @@ struct bcmgenet_priv { unsigned int irq0_stat; /* HW descriptors/checksum variables */ - bool desc_64b_en; - bool desc_rxchk_en; bool crc_fwd_en; u32 dma_max_burst_length; diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet_wol.c b/drivers/net/ethernet/broadcom/genet/bcmgenet_wol.c index ea20d94bd050..c9a43695b182 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmgenet_wol.c +++ b/drivers/net/ethernet/broadcom/genet/bcmgenet_wol.c @@ -132,8 +132,12 @@ int bcmgenet_wol_power_down_cfg(struct bcmgenet_priv *priv, return -EINVAL; } - /* disable RX */ + /* Can't suspend with WoL if MAC is still in reset */ reg = bcmgenet_umac_readl(priv, UMAC_CMD); + if (reg & CMD_SW_RESET) + reg &= ~CMD_SW_RESET; + + /* disable RX */ reg &= ~CMD_RX_EN; bcmgenet_umac_writel(priv, reg, UMAC_CMD); mdelay(10); diff --git a/drivers/net/ethernet/broadcom/genet/bcmmii.c b/drivers/net/ethernet/broadcom/genet/bcmmii.c index 6392a2530183..b5930f80039d 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmmii.c +++ b/drivers/net/ethernet/broadcom/genet/bcmmii.c @@ -95,6 +95,12 @@ void bcmgenet_mii_setup(struct net_device *dev) CMD_HD_EN | CMD_RX_PAUSE_IGNORE | CMD_TX_PAUSE_IGNORE); reg |= cmd_bits; + if (reg & CMD_SW_RESET) { + reg &= ~CMD_SW_RESET; + bcmgenet_umac_writel(priv, reg, UMAC_CMD); + udelay(2); + reg |= CMD_TX_EN | CMD_RX_EN; + } bcmgenet_umac_writel(priv, reg, UMAC_CMD); } else { /* done if nothing has changed */ @@ -181,38 +187,8 @@ int bcmgenet_mii_config(struct net_device *dev, bool init) const char *phy_name = NULL; u32 id_mode_dis = 0; u32 port_ctrl; - int bmcr = -1; - int ret; u32 reg; - /* MAC clocking workaround during reset of umac state machines */ - reg = bcmgenet_umac_readl(priv, UMAC_CMD); - if (reg & CMD_SW_RESET) { - /* An MII PHY must be isolated to prevent TXC contention */ - if (priv->phy_interface == PHY_INTERFACE_MODE_MII) { - ret = phy_read(phydev, MII_BMCR); - if (ret >= 0) { - bmcr = ret; - ret = phy_write(phydev, MII_BMCR, - bmcr | BMCR_ISOLATE); - } - if (ret) { - netdev_err(dev, "failed to isolate PHY\n"); - return ret; - } - } - /* Switch MAC clocking to RGMII generated clock */ - bcmgenet_sys_writel(priv, PORT_MODE_EXT_GPHY, SYS_PORT_CTRL); - /* Ensure 5 clks with Rx disabled - * followed by 5 clks with Reset asserted - */ - udelay(4); - reg &= ~(CMD_SW_RESET | CMD_LCL_LOOP_EN); - bcmgenet_umac_writel(priv, reg, UMAC_CMD); - /* Ensure 5 more clocks before Rx is enabled */ - udelay(2); - } - switch (priv->phy_interface) { case PHY_INTERFACE_MODE_INTERNAL: phy_name = "internal PHY"; @@ -282,10 +258,6 @@ int bcmgenet_mii_config(struct net_device *dev, bool init) bcmgenet_sys_writel(priv, port_ctrl, SYS_PORT_CTRL); - /* Restore the MII PHY after isolation */ - if (bmcr >= 0) - phy_write(phydev, MII_BMCR, bmcr); - priv->ext_phy = !priv->internal_phy && (priv->phy_interface != PHY_INTERFACE_MODE_MOCA); @@ -294,6 +266,7 @@ int bcmgenet_mii_config(struct net_device *dev, bool init) */ if (priv->ext_phy) { reg = bcmgenet_ext_readl(priv, EXT_RGMII_OOB_CTRL); + reg &= ~ID_MODE_DIS; reg |= id_mode_dis; if (GENET_IS_V1(priv) || GENET_IS_V2(priv) || GENET_IS_V3(priv)) reg |= RGMII_MODE_EN_V123; diff --git a/drivers/net/ethernet/cadence/macb.h b/drivers/net/ethernet/cadence/macb.h index dbf7070fcdba..a3f0f27fc79a 100644 --- a/drivers/net/ethernet/cadence/macb.h +++ b/drivers/net/ethernet/cadence/macb.h @@ -652,6 +652,7 @@ #define MACB_CAPS_GEM_HAS_PTP 0x00000040 #define MACB_CAPS_BD_RD_PREFETCH 0x00000080 #define MACB_CAPS_NEEDS_RSTONUBR 0x00000100 +#define MACB_CAPS_MACB_IS_EMAC 0x08000000 #define MACB_CAPS_FIFO_MODE 0x10000000 #define MACB_CAPS_GIGABIT_MODE_AVAILABLE 0x20000000 #define MACB_CAPS_SG_DISABLED 0x40000000 diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c index 4508f0d150da..2c28da1737fe 100644 --- a/drivers/net/ethernet/cadence/macb_main.c +++ b/drivers/net/ethernet/cadence/macb_main.c @@ -572,8 +572,21 @@ static void macb_mac_config(struct phylink_config *config, unsigned int mode, old_ctrl = ctrl = macb_or_gem_readl(bp, NCFGR); /* Clear all the bits we might set later */ - ctrl &= ~(GEM_BIT(GBE) | MACB_BIT(SPD) | MACB_BIT(FD) | MACB_BIT(PAE) | - GEM_BIT(SGMIIEN) | GEM_BIT(PCSSEL)); + ctrl &= ~(MACB_BIT(SPD) | MACB_BIT(FD) | MACB_BIT(PAE)); + + if (bp->caps & MACB_CAPS_MACB_IS_EMAC) { + if (state->interface == PHY_INTERFACE_MODE_RMII) + ctrl |= MACB_BIT(RM9200_RMII); + } else { + ctrl &= ~(GEM_BIT(GBE) | GEM_BIT(SGMIIEN) | GEM_BIT(PCSSEL)); + + /* We do not support MLO_PAUSE_RX yet */ + if (state->pause & MLO_PAUSE_TX) + ctrl |= MACB_BIT(PAE); + + if (state->interface == PHY_INTERFACE_MODE_SGMII) + ctrl |= GEM_BIT(SGMIIEN) | GEM_BIT(PCSSEL); + } if (state->speed == SPEED_1000) ctrl |= GEM_BIT(GBE); @@ -583,13 +596,6 @@ static void macb_mac_config(struct phylink_config *config, unsigned int mode, if (state->duplex) ctrl |= MACB_BIT(FD); - /* We do not support MLO_PAUSE_RX yet */ - if (state->pause & MLO_PAUSE_TX) - ctrl |= MACB_BIT(PAE); - - if (state->interface == PHY_INTERFACE_MODE_SGMII) - ctrl |= GEM_BIT(SGMIIEN) | GEM_BIT(PCSSEL); - /* Apply the new configuration, if any */ if (old_ctrl ^ ctrl) macb_or_gem_writel(bp, NCFGR, ctrl); @@ -608,9 +614,10 @@ static void macb_mac_link_down(struct phylink_config *config, unsigned int mode, unsigned int q; u32 ctrl; - for (q = 0, queue = bp->queues; q < bp->num_queues; ++q, ++queue) - queue_writel(queue, IDR, - bp->rx_intr_mask | MACB_TX_INT_FLAGS | MACB_BIT(HRESP)); + if (!(bp->caps & MACB_CAPS_MACB_IS_EMAC)) + for (q = 0, queue = bp->queues; q < bp->num_queues; ++q, ++queue) + queue_writel(queue, IDR, + bp->rx_intr_mask | MACB_TX_INT_FLAGS | MACB_BIT(HRESP)); /* Disable Rx and Tx */ ctrl = macb_readl(bp, NCR) & ~(MACB_BIT(RE) | MACB_BIT(TE)); @@ -627,17 +634,19 @@ static void macb_mac_link_up(struct phylink_config *config, unsigned int mode, struct macb_queue *queue; unsigned int q; - macb_set_tx_clk(bp->tx_clk, bp->speed, ndev); + if (!(bp->caps & MACB_CAPS_MACB_IS_EMAC)) { + macb_set_tx_clk(bp->tx_clk, bp->speed, ndev); - /* Initialize rings & buffers as clearing MACB_BIT(TE) in link down - * cleared the pipeline and control registers. - */ - bp->macbgem_ops.mog_init_rings(bp); - macb_init_buffers(bp); + /* Initialize rings & buffers as clearing MACB_BIT(TE) in link down + * cleared the pipeline and control registers. + */ + bp->macbgem_ops.mog_init_rings(bp); + macb_init_buffers(bp); - for (q = 0, queue = bp->queues; q < bp->num_queues; ++q, ++queue) - queue_writel(queue, IER, - bp->rx_intr_mask | MACB_TX_INT_FLAGS | MACB_BIT(HRESP)); + for (q = 0, queue = bp->queues; q < bp->num_queues; ++q, ++queue) + queue_writel(queue, IER, + bp->rx_intr_mask | MACB_TX_INT_FLAGS | MACB_BIT(HRESP)); + } /* Enable Rx and Tx */ macb_writel(bp, NCR, macb_readl(bp, NCR) | MACB_BIT(RE) | MACB_BIT(TE)); @@ -3790,6 +3799,10 @@ static int at91ether_open(struct net_device *dev) u32 ctl; int ret; + ret = pm_runtime_get_sync(&lp->pdev->dev); + if (ret < 0) + return ret; + /* Clear internal statistics */ ctl = macb_readl(lp, NCR); macb_writel(lp, NCR, ctl | MACB_BIT(CLRSTAT)); @@ -3854,7 +3867,7 @@ static int at91ether_close(struct net_device *dev) q->rx_buffers, q->rx_buffers_dma); q->rx_buffers = NULL; - return 0; + return pm_runtime_put(&lp->pdev->dev); } /* Transmit packet */ @@ -4037,7 +4050,6 @@ static int at91ether_init(struct platform_device *pdev) struct net_device *dev = platform_get_drvdata(pdev); struct macb *bp = netdev_priv(dev); int err; - u32 reg; bp->queues[0].bp = bp; @@ -4051,11 +4063,7 @@ static int at91ether_init(struct platform_device *pdev) macb_writel(bp, NCR, 0); - reg = MACB_BF(CLK, MACB_CLK_DIV32) | MACB_BIT(BIG); - if (bp->phy_interface == PHY_INTERFACE_MODE_RMII) - reg |= MACB_BIT(RM9200_RMII); - - macb_writel(bp, NCFGR, reg); + macb_writel(bp, NCFGR, MACB_BF(CLK, MACB_CLK_DIV32) | MACB_BIT(BIG)); return 0; } @@ -4214,7 +4222,7 @@ static const struct macb_config sama5d4_config = { }; static const struct macb_config emac_config = { - .caps = MACB_CAPS_NEEDS_RSTONUBR, + .caps = MACB_CAPS_NEEDS_RSTONUBR | MACB_CAPS_MACB_IS_EMAC, .clk_init = at91ether_clk_init, .init = at91ether_init, }; diff --git a/drivers/net/ethernet/cavium/thunder/thunder_bgx.c b/drivers/net/ethernet/cavium/thunder/thunder_bgx.c index 17a4110c2e49..8ff28ed04b7f 100644 --- a/drivers/net/ethernet/cavium/thunder/thunder_bgx.c +++ b/drivers/net/ethernet/cavium/thunder/thunder_bgx.c @@ -410,10 +410,19 @@ void bgx_lmac_rx_tx_enable(int node, int bgx_idx, int lmacid, bool enable) lmac = &bgx->lmac[lmacid]; cfg = bgx_reg_read(bgx, lmacid, BGX_CMRX_CFG); - if (enable) + if (enable) { cfg |= CMR_PKT_RX_EN | CMR_PKT_TX_EN; - else + + /* enable TX FIFO Underflow interrupt */ + bgx_reg_modify(bgx, lmacid, BGX_GMP_GMI_TXX_INT_ENA_W1S, + GMI_TXX_INT_UNDFLW); + } else { cfg &= ~(CMR_PKT_RX_EN | CMR_PKT_TX_EN); + + /* Disable TX FIFO Underflow interrupt */ + bgx_reg_modify(bgx, lmacid, BGX_GMP_GMI_TXX_INT_ENA_W1C, + GMI_TXX_INT_UNDFLW); + } bgx_reg_write(bgx, lmacid, BGX_CMRX_CFG, cfg); if (bgx->is_rgx) @@ -1535,6 +1544,48 @@ static int bgx_init_phy(struct bgx *bgx) return bgx_init_of_phy(bgx); } +static irqreturn_t bgx_intr_handler(int irq, void *data) +{ + struct bgx *bgx = (struct bgx *)data; + u64 status, val; + int lmac; + + for (lmac = 0; lmac < bgx->lmac_count; lmac++) { + status = bgx_reg_read(bgx, lmac, BGX_GMP_GMI_TXX_INT); + if (status & GMI_TXX_INT_UNDFLW) { + pci_err(bgx->pdev, "BGX%d lmac%d UNDFLW\n", + bgx->bgx_id, lmac); + val = bgx_reg_read(bgx, lmac, BGX_CMRX_CFG); + val &= ~CMR_EN; + bgx_reg_write(bgx, lmac, BGX_CMRX_CFG, val); + val |= CMR_EN; + bgx_reg_write(bgx, lmac, BGX_CMRX_CFG, val); + } + /* clear interrupts */ + bgx_reg_write(bgx, lmac, BGX_GMP_GMI_TXX_INT, status); + } + + return IRQ_HANDLED; +} + +static void bgx_register_intr(struct pci_dev *pdev) +{ + struct bgx *bgx = pci_get_drvdata(pdev); + int ret; + + ret = pci_alloc_irq_vectors(pdev, BGX_LMAC_VEC_OFFSET, + BGX_LMAC_VEC_OFFSET, PCI_IRQ_ALL_TYPES); + if (ret < 0) { + pci_err(pdev, "Req for #%d msix vectors failed\n", + BGX_LMAC_VEC_OFFSET); + return; + } + ret = pci_request_irq(pdev, GMPX_GMI_TX_INT, bgx_intr_handler, NULL, + bgx, "BGX%d", bgx->bgx_id); + if (ret) + pci_free_irq(pdev, GMPX_GMI_TX_INT, bgx); +} + static int bgx_probe(struct pci_dev *pdev, const struct pci_device_id *ent) { int err; @@ -1550,7 +1601,7 @@ static int bgx_probe(struct pci_dev *pdev, const struct pci_device_id *ent) pci_set_drvdata(pdev, bgx); - err = pci_enable_device(pdev); + err = pcim_enable_device(pdev); if (err) { dev_err(dev, "Failed to enable PCI device\n"); pci_set_drvdata(pdev, NULL); @@ -1604,6 +1655,8 @@ static int bgx_probe(struct pci_dev *pdev, const struct pci_device_id *ent) bgx_init_hw(bgx); + bgx_register_intr(pdev); + /* Enable all LMACs */ for (lmac = 0; lmac < bgx->lmac_count; lmac++) { err = bgx_lmac_enable(bgx, lmac); @@ -1620,6 +1673,7 @@ static int bgx_probe(struct pci_dev *pdev, const struct pci_device_id *ent) err_enable: bgx_vnic[bgx->bgx_id] = NULL; + pci_free_irq(pdev, GMPX_GMI_TX_INT, bgx); err_release_regions: pci_release_regions(pdev); err_disable_device: @@ -1637,6 +1691,8 @@ static void bgx_remove(struct pci_dev *pdev) for (lmac = 0; lmac < bgx->lmac_count; lmac++) bgx_lmac_disable(bgx, lmac); + pci_free_irq(pdev, GMPX_GMI_TX_INT, bgx); + bgx_vnic[bgx->bgx_id] = NULL; pci_release_regions(pdev); pci_disable_device(pdev); diff --git a/drivers/net/ethernet/cavium/thunder/thunder_bgx.h b/drivers/net/ethernet/cavium/thunder/thunder_bgx.h index 25888706bdcd..cdea49392185 100644 --- a/drivers/net/ethernet/cavium/thunder/thunder_bgx.h +++ b/drivers/net/ethernet/cavium/thunder/thunder_bgx.h @@ -180,6 +180,15 @@ #define BGX_GMP_GMI_TXX_BURST 0x38228 #define BGX_GMP_GMI_TXX_MIN_PKT 0x38240 #define BGX_GMP_GMI_TXX_SGMII_CTL 0x38300 +#define BGX_GMP_GMI_TXX_INT 0x38500 +#define BGX_GMP_GMI_TXX_INT_W1S 0x38508 +#define BGX_GMP_GMI_TXX_INT_ENA_W1C 0x38510 +#define BGX_GMP_GMI_TXX_INT_ENA_W1S 0x38518 +#define GMI_TXX_INT_PTP_LOST BIT_ULL(4) +#define GMI_TXX_INT_LATE_COL BIT_ULL(3) +#define GMI_TXX_INT_XSDEF BIT_ULL(2) +#define GMI_TXX_INT_XSCOL BIT_ULL(1) +#define GMI_TXX_INT_UNDFLW BIT_ULL(0) #define BGX_MSIX_VEC_0_29_ADDR 0x400000 /* +(0..29) << 4 */ #define BGX_MSIX_VEC_0_29_CTL 0x400008 diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c index 2a2938bbb93a..fc05248984fc 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c @@ -902,7 +902,7 @@ void clear_all_filters(struct adapter *adapter) adapter->tids.tid_tab[i]; if (f && (f->valid || f->pending)) - cxgb4_del_filter(dev, i, &f->fs); + cxgb4_del_filter(dev, f->tid, &f->fs); } sb = t4_read_reg(adapter, LE_DB_SRVR_START_INDEX_A); @@ -910,7 +910,7 @@ void clear_all_filters(struct adapter *adapter) f = (struct filter_entry *)adapter->tids.tid_tab[i]; if (f && (f->valid || f->pending)) - cxgb4_del_filter(dev, i, &f->fs); + cxgb4_del_filter(dev, f->tid, &f->fs); } } } diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c index 649842a8aa28..97f90edbc068 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c @@ -5381,12 +5381,11 @@ static inline bool is_x_10g_port(const struct link_config *lc) static int cfg_queues(struct adapter *adap) { u32 avail_qsets, avail_eth_qsets, avail_uld_qsets; + u32 i, n10g = 0, qidx = 0, n1g = 0; + u32 ncpus = num_online_cpus(); u32 niqflint, neq, num_ulds; struct sge *s = &adap->sge; - u32 i, n10g = 0, qidx = 0; -#ifndef CONFIG_CHELSIO_T4_DCB - int q10g = 0; -#endif + u32 q10g = 0, q1g; /* Reduce memory usage in kdump environment, disable all offload. */ if (is_kdump_kernel() || (is_uld(adap) && t4_uld_mem_alloc(adap))) { @@ -5424,44 +5423,50 @@ static int cfg_queues(struct adapter *adap) n10g += is_x_10g_port(&adap2pinfo(adap, i)->link_cfg); avail_eth_qsets = min_t(u32, avail_qsets, MAX_ETH_QSETS); + + /* We default to 1 queue per non-10G port and up to # of cores queues + * per 10G port. + */ + if (n10g) + q10g = (avail_eth_qsets - (adap->params.nports - n10g)) / n10g; + + n1g = adap->params.nports - n10g; #ifdef CONFIG_CHELSIO_T4_DCB /* For Data Center Bridging support we need to be able to support up * to 8 Traffic Priorities; each of which will be assigned to its * own TX Queue in order to prevent Head-Of-Line Blocking. */ + q1g = 8; if (adap->params.nports * 8 > avail_eth_qsets) { dev_err(adap->pdev_dev, "DCB avail_eth_qsets=%d < %d!\n", avail_eth_qsets, adap->params.nports * 8); return -ENOMEM; } - for_each_port(adap, i) { - struct port_info *pi = adap2pinfo(adap, i); + if (adap->params.nports * ncpus < avail_eth_qsets) + q10g = max(8U, ncpus); + else + q10g = max(8U, q10g); - pi->first_qset = qidx; - pi->nqsets = is_kdump_kernel() ? 1 : 8; - qidx += pi->nqsets; - } -#else /* !CONFIG_CHELSIO_T4_DCB */ - /* We default to 1 queue per non-10G port and up to # of cores queues - * per 10G port. - */ - if (n10g) - q10g = (avail_eth_qsets - (adap->params.nports - n10g)) / n10g; - if (q10g > netif_get_num_default_rss_queues()) - q10g = netif_get_num_default_rss_queues(); + while ((q10g * n10g) > (avail_eth_qsets - n1g * q1g)) + q10g--; - if (is_kdump_kernel()) +#else /* !CONFIG_CHELSIO_T4_DCB */ + q1g = 1; + q10g = min(q10g, ncpus); +#endif /* !CONFIG_CHELSIO_T4_DCB */ + if (is_kdump_kernel()) { q10g = 1; + q1g = 1; + } for_each_port(adap, i) { struct port_info *pi = adap2pinfo(adap, i); pi->first_qset = qidx; - pi->nqsets = is_x_10g_port(&pi->link_cfg) ? q10g : 1; + pi->nqsets = is_x_10g_port(&pi->link_cfg) ? q10g : q1g; qidx += pi->nqsets; } -#endif /* !CONFIG_CHELSIO_T4_DCB */ s->ethqsets = qidx; s->max_ethqsets = qidx; /* MSI-X may lower it later */ @@ -5473,7 +5478,7 @@ static int cfg_queues(struct adapter *adap) * capped by the number of available cores. */ num_ulds = adap->num_uld + adap->num_ofld_uld; - i = min_t(u32, MAX_OFLD_QSETS, num_online_cpus()); + i = min_t(u32, MAX_OFLD_QSETS, ncpus); avail_uld_qsets = roundup(i, adap->params.nports); if (avail_qsets < num_ulds * adap->params.nports) { adap->params.offload = 0; diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ptp.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ptp.c index 58a039c3224a..af1f40cbccc8 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ptp.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ptp.c @@ -246,6 +246,9 @@ static int cxgb4_ptp_fineadjtime(struct adapter *adapter, s64 delta) FW_PTP_CMD_PORTID_V(0)); c.retval_len16 = cpu_to_be32(FW_CMD_LEN16_V(sizeof(c) / 16)); c.u.ts.sc = FW_PTP_SC_ADJ_FTIME; + c.u.ts.sign = (delta < 0) ? 1 : 0; + if (delta < 0) + delta = -delta; c.u.ts.tm = cpu_to_be64(delta); err = t4_wr_mbox(adapter, adapter->mbox, &c, sizeof(c), NULL); diff --git a/drivers/net/ethernet/chelsio/cxgb4/sge.c b/drivers/net/ethernet/chelsio/cxgb4/sge.c index 97cda501e7e8..cab3d17e0e1a 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/sge.c +++ b/drivers/net/ethernet/chelsio/cxgb4/sge.c @@ -1307,8 +1307,9 @@ static inline void *write_tso_wr(struct adapter *adap, struct sk_buff *skb, int t4_sge_eth_txq_egress_update(struct adapter *adap, struct sge_eth_txq *eq, int maxreclaim) { + unsigned int reclaimed, hw_cidx; struct sge_txq *q = &eq->q; - unsigned int reclaimed; + int hw_in_use; if (!q->in_use || !__netif_tx_trylock(eq->txq)) return 0; @@ -1316,12 +1317,17 @@ int t4_sge_eth_txq_egress_update(struct adapter *adap, struct sge_eth_txq *eq, /* Reclaim pending completed TX Descriptors. */ reclaimed = reclaim_completed_tx(adap, &eq->q, maxreclaim, true); + hw_cidx = ntohs(READ_ONCE(q->stat->cidx)); + hw_in_use = q->pidx - hw_cidx; + if (hw_in_use < 0) + hw_in_use += q->size; + /* If the TX Queue is currently stopped and there's now more than half * the queue available, restart it. Otherwise bail out since the rest * of what we want do here is with the possibility of shipping any * currently buffered Coalesced TX Work Request. */ - if (netif_tx_queue_stopped(eq->txq) && txq_avail(q) > (q->size / 2)) { + if (netif_tx_queue_stopped(eq->txq) && hw_in_use < (q->size / 2)) { netif_tx_wake_queue(eq->txq); eq->q.restarts++; } @@ -1486,16 +1492,7 @@ static netdev_tx_t cxgb4_eth_xmit(struct sk_buff *skb, struct net_device *dev) * has opened up. */ eth_txq_stop(q); - - /* If we're using the SGE Doorbell Queue Timer facility, we - * don't need to ask the Firmware to send us Egress Queue CIDX - * Updates: the Hardware will do this automatically. And - * since we send the Ingress Queue CIDX Updates to the - * corresponding Ethernet Response Queue, we'll get them very - * quickly. - */ - if (!q->dbqt) - wr_mid |= FW_WR_EQUEQ_F | FW_WR_EQUIQ_F; + wr_mid |= FW_WR_EQUEQ_F | FW_WR_EQUIQ_F; } wr = (void *)&q->q.desc[q->q.pidx]; @@ -1805,16 +1802,7 @@ static netdev_tx_t cxgb4_vf_eth_xmit(struct sk_buff *skb, * has opened up. */ eth_txq_stop(txq); - - /* If we're using the SGE Doorbell Queue Timer facility, we - * don't need to ask the Firmware to send us Egress Queue CIDX - * Updates: the Hardware will do this automatically. And - * since we send the Ingress Queue CIDX Updates to the - * corresponding Ethernet Response Queue, we'll get them very - * quickly. - */ - if (!txq->dbqt) - wr_mid |= FW_WR_EQUEQ_F | FW_WR_EQUIQ_F; + wr_mid |= FW_WR_EQUEQ_F | FW_WR_EQUIQ_F; } /* Start filling in our Work Request. Note that we do _not_ handle @@ -3370,26 +3358,6 @@ static void t4_tx_completion_handler(struct sge_rspq *rspq, } txq = &s->ethtxq[pi->first_qset + rspq->idx]; - - /* We've got the Hardware Consumer Index Update in the Egress Update - * message. If we're using the SGE Doorbell Queue Timer mechanism, - * these Egress Update messages will be our sole CIDX Updates we get - * since we don't want to chew up PCIe bandwidth for both Ingress - * Messages and Status Page writes. However, The code which manages - * reclaiming successfully DMA'ed TX Work Requests uses the CIDX value - * stored in the Status Page at the end of the TX Queue. It's easiest - * to simply copy the CIDX Update value from the Egress Update message - * to the Status Page. Also note that no Endian issues need to be - * considered here since both are Big Endian and we're just copying - * bytes consistently ... - */ - if (txq->dbqt) { - struct cpl_sge_egr_update *egr; - - egr = (struct cpl_sge_egr_update *)rsp; - WRITE_ONCE(txq->q.stat->cidx, egr->cidx); - } - t4_sge_eth_txq_egress_update(adapter, txq, -1); } diff --git a/drivers/net/ethernet/davicom/dm9000.c b/drivers/net/ethernet/davicom/dm9000.c index 1ea3372775e6..e94ae9b94dbf 100644 --- a/drivers/net/ethernet/davicom/dm9000.c +++ b/drivers/net/ethernet/davicom/dm9000.c @@ -1405,6 +1405,8 @@ static struct dm9000_plat_data *dm9000_parse_dt(struct device *dev) mac_addr = of_get_mac_address(np); if (!IS_ERR(mac_addr)) ether_addr_copy(pdata->dev_addr, mac_addr); + else if (PTR_ERR(mac_addr) == -EPROBE_DEFER) + return ERR_CAST(mac_addr); return pdata; } diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c index fd93d542f497..ca74a684a904 100644 --- a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c +++ b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c @@ -1,4 +1,5 @@ /* Copyright 2008 - 2016 Freescale Semiconductor Inc. + * Copyright 2020 NXP * * Redistribution and use in source and binary forms, with or without * modification, are permitted provided that the following conditions are met: @@ -123,7 +124,22 @@ MODULE_PARM_DESC(tx_timeout, "The Tx timeout in ms"); #define FSL_QMAN_MAX_OAL 127 /* Default alignment for start of data in an Rx FD */ +#ifdef CONFIG_DPAA_ERRATUM_A050385 +/* aligning data start to 64 avoids DMA transaction splits, unless the buffer + * is crossing a 4k page boundary + */ +#define DPAA_FD_DATA_ALIGNMENT (fman_has_errata_a050385() ? 64 : 16) +/* aligning to 256 avoids DMA transaction splits caused by 4k page boundary + * crossings; also, all SG fragments except the last must have a size multiple + * of 256 to avoid DMA transaction splits + */ +#define DPAA_A050385_ALIGN 256 +#define DPAA_FD_RX_DATA_ALIGNMENT (fman_has_errata_a050385() ? \ + DPAA_A050385_ALIGN : 16) +#else #define DPAA_FD_DATA_ALIGNMENT 16 +#define DPAA_FD_RX_DATA_ALIGNMENT DPAA_FD_DATA_ALIGNMENT +#endif /* The DPAA requires 256 bytes reserved and mapped for the SGT */ #define DPAA_SGT_SIZE 256 @@ -158,8 +174,13 @@ MODULE_PARM_DESC(tx_timeout, "The Tx timeout in ms"); #define DPAA_PARSE_RESULTS_SIZE sizeof(struct fman_prs_result) #define DPAA_TIME_STAMP_SIZE 8 #define DPAA_HASH_RESULTS_SIZE 8 +#ifdef CONFIG_DPAA_ERRATUM_A050385 +#define DPAA_RX_PRIV_DATA_SIZE (DPAA_A050385_ALIGN - (DPAA_PARSE_RESULTS_SIZE\ + + DPAA_TIME_STAMP_SIZE + DPAA_HASH_RESULTS_SIZE)) +#else #define DPAA_RX_PRIV_DATA_SIZE (u16)(DPAA_TX_PRIV_DATA_SIZE + \ dpaa_rx_extra_headroom) +#endif #define DPAA_ETH_PCD_RXQ_NUM 128 @@ -180,7 +201,12 @@ static struct dpaa_bp *dpaa_bp_array[BM_MAX_NUM_OF_POOLS]; #define DPAA_BP_RAW_SIZE 4096 +#ifdef CONFIG_DPAA_ERRATUM_A050385 +#define dpaa_bp_size(raw_size) (SKB_WITH_OVERHEAD(raw_size) & \ + ~(DPAA_A050385_ALIGN - 1)) +#else #define dpaa_bp_size(raw_size) SKB_WITH_OVERHEAD(raw_size) +#endif static int dpaa_max_frm; @@ -1192,7 +1218,7 @@ static int dpaa_eth_init_rx_port(struct fman_port *port, struct dpaa_bp *bp, buf_prefix_content.pass_prs_result = true; buf_prefix_content.pass_hash_result = true; buf_prefix_content.pass_time_stamp = true; - buf_prefix_content.data_align = DPAA_FD_DATA_ALIGNMENT; + buf_prefix_content.data_align = DPAA_FD_RX_DATA_ALIGNMENT; rx_p = ¶ms.specific_params.rx_params; rx_p->err_fqid = errq->fqid; @@ -1662,6 +1688,8 @@ static u8 rx_csum_offload(const struct dpaa_priv *priv, const struct qm_fd *fd) return CHECKSUM_NONE; } +#define PTR_IS_ALIGNED(x, a) (IS_ALIGNED((unsigned long)(x), (a))) + /* Build a linear skb around the received buffer. * We are guaranteed there is enough room at the end of the data buffer to * accommodate the shared info area of the skb. @@ -1733,8 +1761,7 @@ static struct sk_buff *sg_fd_to_skb(const struct dpaa_priv *priv, sg_addr = qm_sg_addr(&sgt[i]); sg_vaddr = phys_to_virt(sg_addr); - WARN_ON(!IS_ALIGNED((unsigned long)sg_vaddr, - SMP_CACHE_BYTES)); + WARN_ON(!PTR_IS_ALIGNED(sg_vaddr, SMP_CACHE_BYTES)); dma_unmap_page(priv->rx_dma_dev, sg_addr, DPAA_BP_RAW_SIZE, DMA_FROM_DEVICE); @@ -2022,6 +2049,75 @@ static inline int dpaa_xmit(struct dpaa_priv *priv, return 0; } +#ifdef CONFIG_DPAA_ERRATUM_A050385 +int dpaa_a050385_wa(struct net_device *net_dev, struct sk_buff **s) +{ + struct dpaa_priv *priv = netdev_priv(net_dev); + struct sk_buff *new_skb, *skb = *s; + unsigned char *start, i; + + /* check linear buffer alignment */ + if (!PTR_IS_ALIGNED(skb->data, DPAA_A050385_ALIGN)) + goto workaround; + + /* linear buffers just need to have an aligned start */ + if (!skb_is_nonlinear(skb)) + return 0; + + /* linear data size for nonlinear skbs needs to be aligned */ + if (!IS_ALIGNED(skb_headlen(skb), DPAA_A050385_ALIGN)) + goto workaround; + + for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) { + skb_frag_t *frag = &skb_shinfo(skb)->frags[i]; + + /* all fragments need to have aligned start addresses */ + if (!IS_ALIGNED(skb_frag_off(frag), DPAA_A050385_ALIGN)) + goto workaround; + + /* all but last fragment need to have aligned sizes */ + if (!IS_ALIGNED(skb_frag_size(frag), DPAA_A050385_ALIGN) && + (i < skb_shinfo(skb)->nr_frags - 1)) + goto workaround; + } + + return 0; + +workaround: + /* copy all the skb content into a new linear buffer */ + new_skb = netdev_alloc_skb(net_dev, skb->len + DPAA_A050385_ALIGN - 1 + + priv->tx_headroom); + if (!new_skb) + return -ENOMEM; + + /* NET_SKB_PAD bytes already reserved, adding up to tx_headroom */ + skb_reserve(new_skb, priv->tx_headroom - NET_SKB_PAD); + + /* Workaround for DPAA_A050385 requires data start to be aligned */ + start = PTR_ALIGN(new_skb->data, DPAA_A050385_ALIGN); + if (start - new_skb->data != 0) + skb_reserve(new_skb, start - new_skb->data); + + skb_put(new_skb, skb->len); + skb_copy_bits(skb, 0, new_skb->data, skb->len); + skb_copy_header(new_skb, skb); + new_skb->dev = skb->dev; + + /* We move the headroom when we align it so we have to reset the + * network and transport header offsets relative to the new data + * pointer. The checksum offload relies on these offsets. + */ + skb_set_network_header(new_skb, skb_network_offset(skb)); + skb_set_transport_header(new_skb, skb_transport_offset(skb)); + + /* TODO: does timestamping need the result in the old skb? */ + dev_kfree_skb(skb); + *s = new_skb; + + return 0; +} +#endif + static netdev_tx_t dpaa_start_xmit(struct sk_buff *skb, struct net_device *net_dev) { @@ -2068,6 +2164,14 @@ dpaa_start_xmit(struct sk_buff *skb, struct net_device *net_dev) nonlinear = skb_is_nonlinear(skb); } +#ifdef CONFIG_DPAA_ERRATUM_A050385 + if (unlikely(fman_has_errata_a050385())) { + if (dpaa_a050385_wa(net_dev, &skb)) + goto enomem; + nonlinear = skb_is_nonlinear(skb); + } +#endif + if (nonlinear) { /* Just create a S/G fd based on the skb */ err = skb_to_sg_fd(priv, skb, &fd); @@ -2741,9 +2845,7 @@ static inline u16 dpaa_get_headroom(struct dpaa_buffer_layout *bl) headroom = (u16)(bl->priv_data_size + DPAA_PARSE_RESULTS_SIZE + DPAA_TIME_STAMP_SIZE + DPAA_HASH_RESULTS_SIZE); - return DPAA_FD_DATA_ALIGNMENT ? ALIGN(headroom, - DPAA_FD_DATA_ALIGNMENT) : - headroom; + return ALIGN(headroom, DPAA_FD_DATA_ALIGNMENT); } static int dpaa_eth_probe(struct platform_device *pdev) diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c index 4432a59904c7..23c5fef2f1ad 100644 --- a/drivers/net/ethernet/freescale/fec_main.c +++ b/drivers/net/ethernet/freescale/fec_main.c @@ -2529,15 +2529,15 @@ fec_enet_set_coalesce(struct net_device *ndev, struct ethtool_coalesce *ec) return -EINVAL; } - cycle = fec_enet_us_to_itr_clock(ndev, fep->rx_time_itr); + cycle = fec_enet_us_to_itr_clock(ndev, ec->rx_coalesce_usecs); if (cycle > 0xFFFF) { dev_err(dev, "Rx coalesced usec exceed hardware limitation\n"); return -EINVAL; } - cycle = fec_enet_us_to_itr_clock(ndev, fep->tx_time_itr); + cycle = fec_enet_us_to_itr_clock(ndev, ec->tx_coalesce_usecs); if (cycle > 0xFFFF) { - dev_err(dev, "Rx coalesced usec exceed hardware limitation\n"); + dev_err(dev, "Tx coalesced usec exceed hardware limitation\n"); return -EINVAL; } diff --git a/drivers/net/ethernet/freescale/fman/Kconfig b/drivers/net/ethernet/freescale/fman/Kconfig index 0139cb9042ec..34150182cc35 100644 --- a/drivers/net/ethernet/freescale/fman/Kconfig +++ b/drivers/net/ethernet/freescale/fman/Kconfig @@ -8,3 +8,31 @@ config FSL_FMAN help Freescale Data-Path Acceleration Architecture Frame Manager (FMan) support + +config DPAA_ERRATUM_A050385 + bool + depends on ARM64 && FSL_DPAA + default y + help + DPAA FMan erratum A050385 software workaround implementation: + align buffers, data start, SG fragment length to avoid FMan DMA + splits. + FMAN DMA read or writes under heavy traffic load may cause FMAN + internal resource leak thus stopping further packet processing. + The FMAN internal queue can overflow when FMAN splits single + read or write transactions into multiple smaller transactions + such that more than 17 AXI transactions are in flight from FMAN + to interconnect. When the FMAN internal queue overflows, it can + stall further packet processing. The issue can occur with any + one of the following three conditions: + 1. FMAN AXI transaction crosses 4K address boundary (Errata + A010022) + 2. FMAN DMA address for an AXI transaction is not 16 byte + aligned, i.e. the last 4 bits of an address are non-zero + 3. Scatter Gather (SG) frames have more than one SG buffer in + the SG list and any one of the buffers, except the last + buffer in the SG list has data size that is not a multiple + of 16 bytes, i.e., other than 16, 32, 48, 64, etc. + With any one of the above three conditions present, there is + likelihood of stalled FMAN packet processing, especially under + stress with multiple ports injecting line-rate traffic. diff --git a/drivers/net/ethernet/freescale/fman/fman.c b/drivers/net/ethernet/freescale/fman/fman.c index 934111def0be..f151d6e111dd 100644 --- a/drivers/net/ethernet/freescale/fman/fman.c +++ b/drivers/net/ethernet/freescale/fman/fman.c @@ -1,5 +1,6 @@ /* * Copyright 2008-2015 Freescale Semiconductor Inc. + * Copyright 2020 NXP * * Redistribution and use in source and binary forms, with or without * modification, are permitted provided that the following conditions are met: @@ -566,6 +567,10 @@ struct fman_cfg { u32 qmi_def_tnums_thresh; }; +#ifdef CONFIG_DPAA_ERRATUM_A050385 +static bool fman_has_err_a050385; +#endif + static irqreturn_t fman_exceptions(struct fman *fman, enum fman_exceptions exception) { @@ -2518,6 +2523,14 @@ struct fman *fman_bind(struct device *fm_dev) } EXPORT_SYMBOL(fman_bind); +#ifdef CONFIG_DPAA_ERRATUM_A050385 +bool fman_has_errata_a050385(void) +{ + return fman_has_err_a050385; +} +EXPORT_SYMBOL(fman_has_errata_a050385); +#endif + static irqreturn_t fman_err_irq(int irq, void *handle) { struct fman *fman = (struct fman *)handle; @@ -2845,6 +2858,11 @@ static struct fman *read_dts_node(struct platform_device *of_dev) goto fman_free; } +#ifdef CONFIG_DPAA_ERRATUM_A050385 + fman_has_err_a050385 = + of_property_read_bool(fm_node, "fsl,erratum-a050385"); +#endif + return fman; fman_node_put: diff --git a/drivers/net/ethernet/freescale/fman/fman.h b/drivers/net/ethernet/freescale/fman/fman.h index 935c317fa696..f2ede1360f03 100644 --- a/drivers/net/ethernet/freescale/fman/fman.h +++ b/drivers/net/ethernet/freescale/fman/fman.h @@ -1,5 +1,6 @@ /* * Copyright 2008-2015 Freescale Semiconductor Inc. + * Copyright 2020 NXP * * Redistribution and use in source and binary forms, with or without * modification, are permitted provided that the following conditions are met: @@ -398,6 +399,10 @@ u16 fman_get_max_frm(void); int fman_get_rx_extra_headroom(void); +#ifdef CONFIG_DPAA_ERRATUM_A050385 +bool fman_has_errata_a050385(void); +#endif + struct fman *fman_bind(struct device *dev); #endif /* __FM_H */ diff --git a/drivers/net/ethernet/freescale/fman/fman_memac.c b/drivers/net/ethernet/freescale/fman/fman_memac.c index e1901874c19f..0d2b4ab01f24 100644 --- a/drivers/net/ethernet/freescale/fman/fman_memac.c +++ b/drivers/net/ethernet/freescale/fman/fman_memac.c @@ -782,7 +782,7 @@ int memac_adjust_link(struct fman_mac *memac, u16 speed) /* Set full duplex */ tmp &= ~IF_MODE_HD; - if (memac->phy_if == PHY_INTERFACE_MODE_RGMII) { + if (phy_interface_mode_is_rgmii(memac->phy_if)) { /* Configure RGMII in manual mode */ tmp &= ~IF_MODE_RGMII_AUTO; tmp &= ~IF_MODE_RGMII_SP_MASK; diff --git a/drivers/net/ethernet/hisilicon/hns3/hclge_mbx.h b/drivers/net/ethernet/hisilicon/hns3/hclge_mbx.h index 1b0313900f98..d87158acdf6f 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hclge_mbx.h +++ b/drivers/net/ethernet/hisilicon/hns3/hclge_mbx.h @@ -46,6 +46,7 @@ enum HCLGE_MBX_OPCODE { HCLGE_MBX_PUSH_VLAN_INFO, /* (PF -> VF) push port base vlan */ HCLGE_MBX_GET_MEDIA_TYPE, /* (VF -> PF) get media type */ HCLGE_MBX_PUSH_PROMISC_INFO, /* (PF -> VF) push vf promisc info */ + HCLGE_MBX_VF_UNINIT, /* (VF -> PF) vf is unintializing */ HCLGE_MBX_GET_VF_FLR_STATUS = 200, /* (M7 -> PF) get vf flr status */ HCLGE_MBX_PUSH_LINK_STATUS, /* (M7 -> PF) get port link status */ diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c index acb796cc10d0..a7f40aa1a0ea 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c @@ -1711,7 +1711,7 @@ static int hns3_setup_tc(struct net_device *netdev, void *type_data) netif_dbg(h, drv, netdev, "setup tc: num_tc=%u\n", tc); return (kinfo->dcb_ops && kinfo->dcb_ops->setup_tc) ? - kinfo->dcb_ops->setup_tc(h, tc, prio_tc) : -EOPNOTSUPP; + kinfo->dcb_ops->setup_tc(h, tc ? tc : 1, prio_tc) : -EOPNOTSUPP; } static int hns3_nic_setup_tc(struct net_device *dev, enum tc_setup_type type, diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c index 492bc9446463..d3b0cd74ecd2 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c @@ -2446,10 +2446,12 @@ static int hclge_cfg_mac_speed_dup_hw(struct hclge_dev *hdev, int speed, int hclge_cfg_mac_speed_dup(struct hclge_dev *hdev, int speed, u8 duplex) { + struct hclge_mac *mac = &hdev->hw.mac; int ret; duplex = hclge_check_speed_dup(duplex, speed); - if (hdev->hw.mac.speed == speed && hdev->hw.mac.duplex == duplex) + if (!mac->support_autoneg && mac->speed == speed && + mac->duplex == duplex) return 0; ret = hclge_cfg_mac_speed_dup_hw(hdev, speed, duplex); @@ -7743,16 +7745,27 @@ static int hclge_set_vlan_filter_ctrl(struct hclge_dev *hdev, u8 vlan_type, struct hclge_desc desc; int ret; - hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_VLAN_FILTER_CTRL, false); - + /* read current vlan filter parameter */ + hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_VLAN_FILTER_CTRL, true); req = (struct hclge_vlan_filter_ctrl_cmd *)desc.data; req->vlan_type = vlan_type; - req->vlan_fe = filter_en ? fe_type : 0; req->vf_id = vf_id; ret = hclge_cmd_send(&hdev->hw, &desc, 1); + if (ret) { + dev_err(&hdev->pdev->dev, + "failed to get vlan filter config, ret = %d.\n", ret); + return ret; + } + + /* modify and write new config parameter */ + hclge_cmd_reuse_desc(&desc, false); + req->vlan_fe = filter_en ? + (req->vlan_fe | fe_type) : (req->vlan_fe & ~fe_type); + + ret = hclge_cmd_send(&hdev->hw, &desc, 1); if (ret) - dev_err(&hdev->pdev->dev, "set vlan filter fail, ret =%d.\n", + dev_err(&hdev->pdev->dev, "failed to set vlan filter, ret = %d.\n", ret); return ret; @@ -8270,6 +8283,7 @@ void hclge_rm_vport_all_vlan_table(struct hclge_vport *vport, bool is_del_list) kfree(vlan); } } + clear_bit(vport->vport_id, hdev->vf_vlan_full); } void hclge_uninit_vport_vlan_table(struct hclge_dev *hdev) @@ -8486,6 +8500,28 @@ static int hclge_set_vf_vlan_filter(struct hnae3_handle *handle, int vfid, } } +static void hclge_clear_vf_vlan(struct hclge_dev *hdev) +{ + struct hclge_vlan_info *vlan_info; + struct hclge_vport *vport; + int ret; + int vf; + + /* clear port base vlan for all vf */ + for (vf = HCLGE_VF_VPORT_START_NUM; vf < hdev->num_alloc_vport; vf++) { + vport = &hdev->vport[vf]; + vlan_info = &vport->port_base_vlan_cfg.vlan_info; + + ret = hclge_set_vlan_filter_hw(hdev, htons(ETH_P_8021Q), + vport->vport_id, + vlan_info->vlan_tag, true); + if (ret) + dev_err(&hdev->pdev->dev, + "failed to clear vf vlan for vf%d, ret = %d\n", + vf - HCLGE_VF_VPORT_START_NUM, ret); + } +} + int hclge_set_vlan_filter(struct hnae3_handle *handle, __be16 proto, u16 vlan_id, bool is_kill) { @@ -9895,6 +9931,7 @@ static void hclge_uninit_ae_dev(struct hnae3_ae_dev *ae_dev) struct hclge_mac *mac = &hdev->hw.mac; hclge_reset_vf_rate(hdev); + hclge_clear_vf_vlan(hdev); hclge_misc_affinity_teardown(hdev); hclge_state_uninit(hdev); diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c index a3c0822191a9..3d850f6b1e37 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c @@ -799,6 +799,7 @@ void hclge_mbx_handler(struct hclge_dev *hdev) hclge_get_link_mode(vport, req); break; case HCLGE_MBX_GET_VF_FLR_STATUS: + case HCLGE_MBX_VF_UNINIT: hclge_rm_vport_all_mac_table(vport, true, HCLGE_MAC_ADDR_UC); hclge_rm_vport_all_mac_table(vport, true, diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c index d6597206e692..0510d85a7f6a 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c @@ -2803,6 +2803,9 @@ static void hclgevf_uninit_hdev(struct hclgevf_dev *hdev) { hclgevf_state_uninit(hdev); + hclgevf_send_mbx_msg(hdev, HCLGE_MBX_VF_UNINIT, 0, NULL, 0, + false, NULL, 0); + if (test_bit(HCLGEVF_STATE_IRQ_INITED, &hdev->state)) { hclgevf_misc_irq_uninit(hdev); hclgevf_uninit_msi(hdev); diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_cmdq.c b/drivers/net/ethernet/huawei/hinic/hinic_hw_cmdq.c index eb53c15b13f3..5f2d57d1b2d3 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hw_cmdq.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_cmdq.c @@ -389,7 +389,8 @@ static int cmdq_sync_cmd_direct_resp(struct hinic_cmdq *cmdq, spin_unlock_bh(&cmdq->cmdq_lock); - if (!wait_for_completion_timeout(&done, CMDQ_TIMEOUT)) { + if (!wait_for_completion_timeout(&done, + msecs_to_jiffies(CMDQ_TIMEOUT))) { spin_lock_bh(&cmdq->cmdq_lock); if (cmdq->errcode[curr_prod_idx] == &errcode) @@ -623,6 +624,8 @@ static int cmdq_cmd_ceq_handler(struct hinic_cmdq *cmdq, u16 ci, if (!CMDQ_WQE_COMPLETED(be32_to_cpu(ctrl->ctrl_info))) return -EBUSY; + dma_rmb(); + errcode = CMDQ_WQE_ERRCODE_GET(be32_to_cpu(status->status_info), VAL); cmdq_sync_cmd_handler(cmdq, ci, errcode); diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_dev.c b/drivers/net/ethernet/huawei/hinic/hinic_hw_dev.c index 6f2cf569a283..c7c75b772a86 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hw_dev.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_dev.c @@ -297,6 +297,7 @@ static int set_hw_ioctxt(struct hinic_hwdev *hwdev, unsigned int rq_depth, } hw_ioctxt.func_idx = HINIC_HWIF_FUNC_IDX(hwif); + hw_ioctxt.ppf_idx = HINIC_HWIF_PPF_IDX(hwif); hw_ioctxt.set_cmdq_depth = HW_IOCTXT_SET_CMDQ_DEPTH_DEFAULT; hw_ioctxt.cmdq_depth = 0; @@ -359,50 +360,6 @@ static int wait_for_db_state(struct hinic_hwdev *hwdev) return -EFAULT; } -static int wait_for_io_stopped(struct hinic_hwdev *hwdev) -{ - struct hinic_cmd_io_status cmd_io_status; - struct hinic_hwif *hwif = hwdev->hwif; - struct pci_dev *pdev = hwif->pdev; - struct hinic_pfhwdev *pfhwdev; - unsigned long end; - u16 out_size; - int err; - - if (!HINIC_IS_PF(hwif) && !HINIC_IS_PPF(hwif)) { - dev_err(&pdev->dev, "Unsupported PCI Function type\n"); - return -EINVAL; - } - - pfhwdev = container_of(hwdev, struct hinic_pfhwdev, hwdev); - - cmd_io_status.func_idx = HINIC_HWIF_FUNC_IDX(hwif); - - end = jiffies + msecs_to_jiffies(IO_STATUS_TIMEOUT); - do { - err = hinic_msg_to_mgmt(&pfhwdev->pf_to_mgmt, HINIC_MOD_COMM, - HINIC_COMM_CMD_IO_STATUS_GET, - &cmd_io_status, sizeof(cmd_io_status), - &cmd_io_status, &out_size, - HINIC_MGMT_MSG_SYNC); - if ((err) || (out_size != sizeof(cmd_io_status))) { - dev_err(&pdev->dev, "Failed to get IO status, ret = %d\n", - err); - return err; - } - - if (cmd_io_status.status == IO_STOPPED) { - dev_info(&pdev->dev, "IO stopped\n"); - return 0; - } - - msleep(20); - } while (time_before(jiffies, end)); - - dev_err(&pdev->dev, "Wait for IO stopped - Timeout\n"); - return -ETIMEDOUT; -} - /** * clear_io_resource - set the IO resources as not active in the NIC * @hwdev: the NIC HW device @@ -422,11 +379,8 @@ static int clear_io_resources(struct hinic_hwdev *hwdev) return -EINVAL; } - err = wait_for_io_stopped(hwdev); - if (err) { - dev_err(&pdev->dev, "IO has not stopped yet\n"); - return err; - } + /* sleep 100ms to wait for firmware stopping I/O */ + msleep(100); cmd_clear_io_res.func_idx = HINIC_HWIF_FUNC_IDX(hwif); diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_dev.h b/drivers/net/ethernet/huawei/hinic/hinic_hw_dev.h index b069045de416..66fd2340d447 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hw_dev.h +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_dev.h @@ -151,8 +151,8 @@ struct hinic_cmd_hw_ioctxt { u8 lro_en; u8 rsvd3; + u8 ppf_idx; u8 rsvd4; - u8 rsvd5; u16 rq_depth; u16 rx_buf_sz_idx; diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_eqs.c b/drivers/net/ethernet/huawei/hinic/hinic_hw_eqs.c index 79243b626ddb..c0b6bcb067cd 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hw_eqs.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_eqs.c @@ -188,7 +188,7 @@ static u8 eq_cons_idx_checksum_set(u32 val) * eq_update_ci - update the HW cons idx of event queue * @eq: the event queue to update the cons idx for **/ -static void eq_update_ci(struct hinic_eq *eq) +static void eq_update_ci(struct hinic_eq *eq, u32 arm_state) { u32 val, addr = EQ_CONS_IDX_REG_ADDR(eq); @@ -202,7 +202,7 @@ static void eq_update_ci(struct hinic_eq *eq) val |= HINIC_EQ_CI_SET(eq->cons_idx, IDX) | HINIC_EQ_CI_SET(eq->wrapped, WRAPPED) | - HINIC_EQ_CI_SET(EQ_ARMED, INT_ARMED); + HINIC_EQ_CI_SET(arm_state, INT_ARMED); val |= HINIC_EQ_CI_SET(eq_cons_idx_checksum_set(val), XOR_CHKSUM); @@ -235,6 +235,8 @@ static void aeq_irq_handler(struct hinic_eq *eq) if (HINIC_EQ_ELEM_DESC_GET(aeqe_desc, WRAPPED) == eq->wrapped) break; + dma_rmb(); + event = HINIC_EQ_ELEM_DESC_GET(aeqe_desc, TYPE); if (event >= HINIC_MAX_AEQ_EVENTS) { dev_err(&pdev->dev, "Unknown AEQ Event %d\n", event); @@ -347,7 +349,7 @@ static void eq_irq_handler(void *data) else if (eq->type == HINIC_CEQ) ceq_irq_handler(eq); - eq_update_ci(eq); + eq_update_ci(eq, EQ_ARMED); } /** @@ -702,7 +704,7 @@ static int init_eq(struct hinic_eq *eq, struct hinic_hwif *hwif, } set_eq_ctrls(eq); - eq_update_ci(eq); + eq_update_ci(eq, EQ_ARMED); err = alloc_eq_pages(eq); if (err) { @@ -752,18 +754,28 @@ err_req_irq: **/ static void remove_eq(struct hinic_eq *eq) { - struct msix_entry *entry = &eq->msix_entry; - - free_irq(entry->vector, eq); + hinic_set_msix_state(eq->hwif, eq->msix_entry.entry, + HINIC_MSIX_DISABLE); + free_irq(eq->msix_entry.vector, eq); if (eq->type == HINIC_AEQ) { struct hinic_eq_work *aeq_work = &eq->aeq_work; cancel_work_sync(&aeq_work->work); + /* clear aeq_len to avoid hw access host memory */ + hinic_hwif_write_reg(eq->hwif, + HINIC_CSR_AEQ_CTRL_1_ADDR(eq->q_id), 0); } else if (eq->type == HINIC_CEQ) { tasklet_kill(&eq->ceq_tasklet); + /* clear ceq_len to avoid hw access host memory */ + hinic_hwif_write_reg(eq->hwif, + HINIC_CSR_CEQ_CTRL_1_ADDR(eq->q_id), 0); } + /* update cons_idx to avoid invalid interrupt */ + eq->cons_idx = hinic_hwif_read_reg(eq->hwif, EQ_PROD_IDX_REG_ADDR(eq)); + eq_update_ci(eq, EQ_NOT_ARMED); + free_eq_pages(eq); } diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_if.h b/drivers/net/ethernet/huawei/hinic/hinic_hw_if.h index 517794509eb2..c7bb9ceca72c 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hw_if.h +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_if.h @@ -137,6 +137,7 @@ #define HINIC_HWIF_FUNC_IDX(hwif) ((hwif)->attr.func_idx) #define HINIC_HWIF_PCI_INTF(hwif) ((hwif)->attr.pci_intf_idx) #define HINIC_HWIF_PF_IDX(hwif) ((hwif)->attr.pf_idx) +#define HINIC_HWIF_PPF_IDX(hwif) ((hwif)->attr.ppf_idx) #define HINIC_FUNC_TYPE(hwif) ((hwif)->attr.func_type) #define HINIC_IS_PF(hwif) (HINIC_FUNC_TYPE(hwif) == HINIC_PF) diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_mgmt.c b/drivers/net/ethernet/huawei/hinic/hinic_hw_mgmt.c index c1a6be6bf6a8..8995e32dd1c0 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hw_mgmt.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_mgmt.c @@ -43,7 +43,7 @@ #define MSG_NOT_RESP 0xFFFF -#define MGMT_MSG_TIMEOUT 1000 +#define MGMT_MSG_TIMEOUT 5000 #define mgmt_to_pfhwdev(pf_mgmt) \ container_of(pf_mgmt, struct hinic_pfhwdev, pf_to_mgmt) @@ -267,7 +267,8 @@ static int msg_to_mgmt_sync(struct hinic_pf_to_mgmt *pf_to_mgmt, goto unlock_sync_msg; } - if (!wait_for_completion_timeout(recv_done, MGMT_MSG_TIMEOUT)) { + if (!wait_for_completion_timeout(recv_done, + msecs_to_jiffies(MGMT_MSG_TIMEOUT))) { dev_err(&pdev->dev, "MGMT timeout, MSG id = %d\n", msg_id); err = -ETIMEDOUT; goto unlock_sync_msg; diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_qp.h b/drivers/net/ethernet/huawei/hinic/hinic_hw_qp.h index f4a339b10b10..79091e131418 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hw_qp.h +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_qp.h @@ -94,6 +94,7 @@ struct hinic_rq { struct hinic_wq *wq; + struct cpumask affinity_mask; u32 irq; u16 msix_entry; diff --git a/drivers/net/ethernet/huawei/hinic/hinic_main.c b/drivers/net/ethernet/huawei/hinic/hinic_main.c index 02a14f5e7fe3..13560975c103 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_main.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_main.c @@ -356,7 +356,8 @@ static void hinic_enable_rss(struct hinic_dev *nic_dev) if (!num_cpus) num_cpus = num_online_cpus(); - nic_dev->num_qps = min_t(u16, nic_dev->max_qps, num_cpus); + nic_dev->num_qps = hinic_hwdev_num_qps(hwdev); + nic_dev->num_qps = min_t(u16, nic_dev->num_qps, num_cpus); nic_dev->rss_limit = nic_dev->num_qps; nic_dev->num_rss = nic_dev->num_qps; diff --git a/drivers/net/ethernet/huawei/hinic/hinic_rx.c b/drivers/net/ethernet/huawei/hinic/hinic_rx.c index 56ea6d692f1c..815649e37cb1 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_rx.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_rx.c @@ -350,6 +350,9 @@ static int rxq_recv(struct hinic_rxq *rxq, int budget) if (!rq_wqe) break; + /* make sure we read rx_done before packet length */ + dma_rmb(); + cqe = rq->cqe[ci]; status = be32_to_cpu(cqe->status); hinic_rq_get_sge(rxq->rq, rq_wqe, ci, &sge); @@ -475,7 +478,6 @@ static int rx_request_irq(struct hinic_rxq *rxq) struct hinic_hwdev *hwdev = nic_dev->hwdev; struct hinic_rq *rq = rxq->rq; struct hinic_qp *qp; - struct cpumask mask; int err; rx_add_napi(rxq); @@ -492,8 +494,8 @@ static int rx_request_irq(struct hinic_rxq *rxq) } qp = container_of(rq, struct hinic_qp, rq); - cpumask_set_cpu(qp->q_id % num_online_cpus(), &mask); - return irq_set_affinity_hint(rq->irq, &mask); + cpumask_set_cpu(qp->q_id % num_online_cpus(), &rq->affinity_mask); + return irq_set_affinity_hint(rq->irq, &rq->affinity_mask); } static void rx_free_irq(struct hinic_rxq *rxq) diff --git a/drivers/net/ethernet/huawei/hinic/hinic_tx.c b/drivers/net/ethernet/huawei/hinic/hinic_tx.c index 0e13d1c7e474..365016450bdb 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_tx.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_tx.c @@ -45,7 +45,7 @@ #define HW_CONS_IDX(sq) be16_to_cpu(*(u16 *)((sq)->hw_ci_addr)) -#define MIN_SKB_LEN 17 +#define MIN_SKB_LEN 32 #define MAX_PAYLOAD_OFFSET 221 #define TRANSPORT_OFFSET(l4_hdr, skb) ((u32)((l4_hdr) - (skb)->data)) @@ -622,6 +622,8 @@ static int free_tx_poll(struct napi_struct *napi, int budget) do { hw_ci = HW_CONS_IDX(sq) & wq->mask; + dma_rmb(); + /* Reading a WQEBB to get real WQE size and consumer index. */ sq_wqe = hinic_sq_read_wqebb(sq, &skb, &wqe_size, &sw_ci); if ((!sq_wqe) || diff --git a/drivers/net/ethernet/ibm/ibmvnic.c b/drivers/net/ethernet/ibm/ibmvnic.c index c75239d8820f..4bd33245bad6 100644 --- a/drivers/net/ethernet/ibm/ibmvnic.c +++ b/drivers/net/ethernet/ibm/ibmvnic.c @@ -2142,6 +2142,8 @@ static void __ibmvnic_reset(struct work_struct *work) { struct ibmvnic_rwi *rwi; struct ibmvnic_adapter *adapter; + bool saved_state = false; + unsigned long flags; u32 reset_state; int rc = 0; @@ -2153,17 +2155,25 @@ static void __ibmvnic_reset(struct work_struct *work) return; } - reset_state = adapter->state; - rwi = get_next_rwi(adapter); while (rwi) { + spin_lock_irqsave(&adapter->state_lock, flags); + if (adapter->state == VNIC_REMOVING || adapter->state == VNIC_REMOVED) { + spin_unlock_irqrestore(&adapter->state_lock, flags); kfree(rwi); rc = EBUSY; break; } + if (!saved_state) { + reset_state = adapter->state; + adapter->state = VNIC_RESETTING; + saved_state = true; + } + spin_unlock_irqrestore(&adapter->state_lock, flags); + if (rwi->reset_reason == VNIC_RESET_CHANGE_PARAM) { /* CHANGE_PARAM requestor holds rtnl_lock */ rc = do_change_param_reset(adapter, rwi, reset_state); @@ -5091,6 +5101,7 @@ static int ibmvnic_probe(struct vio_dev *dev, const struct vio_device_id *id) __ibmvnic_delayed_reset); INIT_LIST_HEAD(&adapter->rwi_list); spin_lock_init(&adapter->rwi_lock); + spin_lock_init(&adapter->state_lock); mutex_init(&adapter->fw_lock); init_completion(&adapter->init_done); init_completion(&adapter->fw_done); @@ -5163,8 +5174,17 @@ static int ibmvnic_remove(struct vio_dev *dev) { struct net_device *netdev = dev_get_drvdata(&dev->dev); struct ibmvnic_adapter *adapter = netdev_priv(netdev); + unsigned long flags; + + spin_lock_irqsave(&adapter->state_lock, flags); + if (adapter->state == VNIC_RESETTING) { + spin_unlock_irqrestore(&adapter->state_lock, flags); + return -EBUSY; + } adapter->state = VNIC_REMOVING; + spin_unlock_irqrestore(&adapter->state_lock, flags); + rtnl_lock(); unregister_netdevice(netdev); diff --git a/drivers/net/ethernet/ibm/ibmvnic.h b/drivers/net/ethernet/ibm/ibmvnic.h index 60eccaf91b12..f8416e1d4cf0 100644 --- a/drivers/net/ethernet/ibm/ibmvnic.h +++ b/drivers/net/ethernet/ibm/ibmvnic.h @@ -941,7 +941,8 @@ enum vnic_state {VNIC_PROBING = 1, VNIC_CLOSING, VNIC_CLOSED, VNIC_REMOVING, - VNIC_REMOVED}; + VNIC_REMOVED, + VNIC_RESETTING}; enum ibmvnic_reset_reason {VNIC_RESET_FAILOVER = 1, VNIC_RESET_MOBILITY, @@ -1090,4 +1091,7 @@ struct ibmvnic_adapter { struct ibmvnic_tunables desired; struct ibmvnic_tunables fallback; + + /* Used for serializatin of state field */ + spinlock_t state_lock; }; diff --git a/drivers/net/ethernet/intel/ice/ice_ethtool.c b/drivers/net/ethernet/intel/ice/ice_ethtool.c index b002ab4e5838..77c412a7e7a4 100644 --- a/drivers/net/ethernet/intel/ice/ice_ethtool.c +++ b/drivers/net/ethernet/intel/ice/ice_ethtool.c @@ -2936,13 +2936,6 @@ ice_set_pauseparam(struct net_device *netdev, struct ethtool_pauseparam *pause) else return -EINVAL; - /* Tell the OS link is going down, the link will go back up when fw - * says it is ready asynchronously - */ - ice_print_link_msg(vsi, false); - netif_carrier_off(netdev); - netif_tx_stop_all_queues(netdev); - /* Set the FC mode and only restart AN if link is up */ status = ice_set_fc(pi, &aq_failures, link_up); @@ -3489,21 +3482,13 @@ ice_set_rc_coalesce(enum ice_container_type c_type, struct ethtool_coalesce *ec, return -EINVAL; } - /* hardware only supports an ITR granularity of 2us */ - if (coalesce_usecs % 2 != 0) { - netdev_info(vsi->netdev, "Invalid value, %s-usecs must be even\n", - c_type_str); - return -EINVAL; - } - if (use_adaptive_coalesce) { rc->itr_setting |= ICE_ITR_DYNAMIC; } else { - /* store user facing value how it was set */ + /* save the user set usecs */ rc->itr_setting = coalesce_usecs; - /* set to static and convert to value HW understands */ - rc->target_itr = - ITR_TO_REG(ITR_REG_ALIGN(rc->itr_setting)); + /* device ITR granularity is in 2 usec increments */ + rc->target_itr = ITR_REG_ALIGN(rc->itr_setting); } return 0; @@ -3597,6 +3582,30 @@ ice_is_coalesce_param_invalid(struct net_device *netdev, } /** + * ice_print_if_odd_usecs - print message if user tries to set odd [tx|rx]-usecs + * @netdev: netdev used for print + * @itr_setting: previous user setting + * @use_adaptive_coalesce: if adaptive coalesce is enabled or being enabled + * @coalesce_usecs: requested value of [tx|rx]-usecs + * @c_type_str: either "rx" or "tx" to match user set field of [tx|rx]-usecs + */ +static void +ice_print_if_odd_usecs(struct net_device *netdev, u16 itr_setting, + u32 use_adaptive_coalesce, u32 coalesce_usecs, + const char *c_type_str) +{ + if (use_adaptive_coalesce) + return; + + itr_setting = ITR_TO_REG(itr_setting); + + if (itr_setting != coalesce_usecs && (coalesce_usecs % 2)) + netdev_info(netdev, "User set %s-usecs to %d, device only supports even values. Rounding down and attempting to set %s-usecs to %d\n", + c_type_str, coalesce_usecs, c_type_str, + ITR_REG_ALIGN(coalesce_usecs)); +} + +/** * __ice_set_coalesce - set ITR/INTRL values for the device * @netdev: pointer to the netdev associated with this query * @ec: ethtool structure to fill with driver's coalesce settings @@ -3616,8 +3625,19 @@ __ice_set_coalesce(struct net_device *netdev, struct ethtool_coalesce *ec, return -EINVAL; if (q_num < 0) { + struct ice_q_vector *q_vector = vsi->q_vectors[0]; int v_idx; + if (q_vector) { + ice_print_if_odd_usecs(netdev, q_vector->rx.itr_setting, + ec->use_adaptive_rx_coalesce, + ec->rx_coalesce_usecs, "rx"); + + ice_print_if_odd_usecs(netdev, q_vector->tx.itr_setting, + ec->use_adaptive_tx_coalesce, + ec->tx_coalesce_usecs, "tx"); + } + ice_for_each_q_vector(vsi, v_idx) { /* In some cases if DCB is configured the num_[rx|tx]q * can be less than vsi->num_q_vectors. This check diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.h b/drivers/net/ethernet/intel/ice/ice_txrx.h index 14a1bf445889..7ee00a128663 100644 --- a/drivers/net/ethernet/intel/ice/ice_txrx.h +++ b/drivers/net/ethernet/intel/ice/ice_txrx.h @@ -222,7 +222,7 @@ enum ice_rx_dtype { #define ICE_ITR_GRAN_S 1 /* ITR granularity is always 2us */ #define ICE_ITR_GRAN_US BIT(ICE_ITR_GRAN_S) #define ICE_ITR_MASK 0x1FFE /* ITR register value alignment mask */ -#define ITR_REG_ALIGN(setting) __ALIGN_MASK(setting, ~ICE_ITR_MASK) +#define ITR_REG_ALIGN(setting) ((setting) & ICE_ITR_MASK) #define ICE_ITR_ADAPTIVE_MIN_INC 0x0002 #define ICE_ITR_ADAPTIVE_MIN_USECS 0x0002 diff --git a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c index 262714d5f54a..75c70d432c72 100644 --- a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c +++ b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c @@ -1874,6 +1874,48 @@ error_param: } /** + * ice_wait_on_vf_reset - poll to make sure a given VF is ready after reset + * @vf: The VF being resseting + * + * The max poll time is about ~800ms, which is about the maximum time it takes + * for a VF to be reset and/or a VF driver to be removed. + */ +static void ice_wait_on_vf_reset(struct ice_vf *vf) +{ + int i; + + for (i = 0; i < ICE_MAX_VF_RESET_TRIES; i++) { + if (test_bit(ICE_VF_STATE_INIT, vf->vf_states)) + break; + msleep(ICE_MAX_VF_RESET_SLEEP_MS); + } +} + +/** + * ice_check_vf_ready_for_cfg - check if VF is ready to be configured/queried + * @vf: VF to check if it's ready to be configured/queried + * + * The purpose of this function is to make sure the VF is not in reset, not + * disabled, and initialized so it can be configured and/or queried by a host + * administrator. + */ +static int ice_check_vf_ready_for_cfg(struct ice_vf *vf) +{ + struct ice_pf *pf; + + ice_wait_on_vf_reset(vf); + + if (ice_is_vf_disabled(vf)) + return -EINVAL; + + pf = vf->pf; + if (ice_check_vf_init(pf, vf)) + return -EBUSY; + + return 0; +} + +/** * ice_set_vf_spoofchk * @netdev: network interface device structure * @vf_id: VF identifier @@ -1890,16 +1932,16 @@ int ice_set_vf_spoofchk(struct net_device *netdev, int vf_id, bool ena) enum ice_status status; struct device *dev; struct ice_vf *vf; - int ret = 0; + int ret; dev = ice_pf_to_dev(pf); if (ice_validate_vf_id(pf, vf_id)) return -EINVAL; vf = &pf->vf[vf_id]; - - if (ice_check_vf_init(pf, vf)) - return -EBUSY; + ret = ice_check_vf_ready_for_cfg(vf); + if (ret) + return ret; vf_vsi = pf->vsi[vf->lan_vsi_idx]; if (!vf_vsi) { @@ -2696,7 +2738,7 @@ ice_set_vf_port_vlan(struct net_device *netdev, int vf_id, u16 vlan_id, u8 qos, struct ice_vsi *vsi; struct device *dev; struct ice_vf *vf; - int ret = 0; + int ret; dev = ice_pf_to_dev(pf); if (ice_validate_vf_id(pf, vf_id)) @@ -2714,13 +2756,15 @@ ice_set_vf_port_vlan(struct net_device *netdev, int vf_id, u16 vlan_id, u8 qos, vf = &pf->vf[vf_id]; vsi = pf->vsi[vf->lan_vsi_idx]; - if (ice_check_vf_init(pf, vf)) - return -EBUSY; + + ret = ice_check_vf_ready_for_cfg(vf); + if (ret) + return ret; if (le16_to_cpu(vsi->info.pvid) == vlanprio) { /* duplicate request, so just return success */ dev_dbg(dev, "Duplicate pvid %d request\n", vlanprio); - return ret; + return 0; } /* If PVID, then remove all filters on the old VLAN */ @@ -2731,7 +2775,7 @@ ice_set_vf_port_vlan(struct net_device *netdev, int vf_id, u16 vlan_id, u8 qos, if (vlan_id || qos) { ret = ice_vsi_manage_pvid(vsi, vlanprio, true); if (ret) - goto error_set_pvid; + return ret; } else { ice_vsi_manage_pvid(vsi, 0, false); vsi->info.pvid = 0; @@ -2744,7 +2788,7 @@ ice_set_vf_port_vlan(struct net_device *netdev, int vf_id, u16 vlan_id, u8 qos, /* add new VLAN filter for each MAC */ ret = ice_vsi_add_vlan(vsi, vlan_id); if (ret) - goto error_set_pvid; + return ret; } /* The Port VLAN needs to be saved across resets the same as the @@ -2752,8 +2796,7 @@ ice_set_vf_port_vlan(struct net_device *netdev, int vf_id, u16 vlan_id, u8 qos, */ vf->port_vlan_id = le16_to_cpu(vsi->info.pvid); -error_set_pvid: - return ret; + return 0; } /** @@ -3237,23 +3280,6 @@ ice_get_vf_cfg(struct net_device *netdev, int vf_id, struct ifla_vf_info *ivi) } /** - * ice_wait_on_vf_reset - * @vf: The VF being resseting - * - * Poll to make sure a given VF is ready after reset - */ -static void ice_wait_on_vf_reset(struct ice_vf *vf) -{ - int i; - - for (i = 0; i < ICE_MAX_VF_RESET_WAIT; i++) { - if (test_bit(ICE_VF_STATE_INIT, vf->vf_states)) - break; - msleep(20); - } -} - -/** * ice_set_vf_mac * @netdev: network interface device structure * @vf_id: VF identifier @@ -3265,29 +3291,21 @@ int ice_set_vf_mac(struct net_device *netdev, int vf_id, u8 *mac) { struct ice_pf *pf = ice_netdev_to_pf(netdev); struct ice_vf *vf; - int ret = 0; + int ret; if (ice_validate_vf_id(pf, vf_id)) return -EINVAL; - vf = &pf->vf[vf_id]; - /* Don't set MAC on disabled VF */ - if (ice_is_vf_disabled(vf)) - return -EINVAL; - - /* In case VF is in reset mode, wait until it is completed. Depending - * on factors like queue disabling routine, this could take ~250ms - */ - ice_wait_on_vf_reset(vf); - - if (ice_check_vf_init(pf, vf)) - return -EBUSY; - if (is_zero_ether_addr(mac) || is_multicast_ether_addr(mac)) { netdev_err(netdev, "%pM not a valid unicast address\n", mac); return -EINVAL; } + vf = &pf->vf[vf_id]; + ret = ice_check_vf_ready_for_cfg(vf); + if (ret) + return ret; + /* copy MAC into dflt_lan_addr and trigger a VF reset. The reset * flow will use the updated dflt_lan_addr and add a MAC filter * using ice_add_mac. Also set pf_set_mac to indicate that the PF has @@ -3299,7 +3317,7 @@ int ice_set_vf_mac(struct net_device *netdev, int vf_id, u8 *mac) vf_id, mac); ice_vc_reset_vf(vf); - return ret; + return 0; } /** @@ -3314,22 +3332,15 @@ int ice_set_vf_trust(struct net_device *netdev, int vf_id, bool trusted) { struct ice_pf *pf = ice_netdev_to_pf(netdev); struct ice_vf *vf; + int ret; if (ice_validate_vf_id(pf, vf_id)) return -EINVAL; vf = &pf->vf[vf_id]; - /* Don't set Trusted Mode on disabled VF */ - if (ice_is_vf_disabled(vf)) - return -EINVAL; - - /* In case VF is in reset mode, wait until it is completed. Depending - * on factors like queue disabling routine, this could take ~250ms - */ - ice_wait_on_vf_reset(vf); - - if (ice_check_vf_init(pf, vf)) - return -EBUSY; + ret = ice_check_vf_ready_for_cfg(vf); + if (ret) + return ret; /* Check if already trusted */ if (trusted == vf->trusted) @@ -3355,13 +3366,15 @@ int ice_set_vf_link_state(struct net_device *netdev, int vf_id, int link_state) { struct ice_pf *pf = ice_netdev_to_pf(netdev); struct ice_vf *vf; + int ret; if (ice_validate_vf_id(pf, vf_id)) return -EINVAL; vf = &pf->vf[vf_id]; - if (ice_check_vf_init(pf, vf)) - return -EBUSY; + ret = ice_check_vf_ready_for_cfg(vf); + if (ret) + return ret; switch (link_state) { case IFLA_VF_LINK_STATE_AUTO: @@ -3397,14 +3410,15 @@ int ice_get_vf_stats(struct net_device *netdev, int vf_id, struct ice_eth_stats *stats; struct ice_vsi *vsi; struct ice_vf *vf; + int ret; if (ice_validate_vf_id(pf, vf_id)) return -EINVAL; vf = &pf->vf[vf_id]; - - if (ice_check_vf_init(pf, vf)) - return -EBUSY; + ret = ice_check_vf_ready_for_cfg(vf); + if (ret) + return ret; vsi = pf->vsi[vf->lan_vsi_idx]; if (!vsi) diff --git a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.h b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.h index 4647d636ed36..ac67982751df 100644 --- a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.h +++ b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.h @@ -38,7 +38,8 @@ #define ICE_MAX_POLICY_INTR_PER_VF 33 #define ICE_MIN_INTR_PER_VF (ICE_MIN_QS_PER_VF + 1) #define ICE_DFLT_INTR_PER_VF (ICE_DFLT_QS_PER_VF + 1) -#define ICE_MAX_VF_RESET_WAIT 15 +#define ICE_MAX_VF_RESET_TRIES 40 +#define ICE_MAX_VF_RESET_SLEEP_MS 20 #define ice_for_each_vf(pf, i) \ for ((i) = 0; (i) < (pf)->num_alloc_vfs; (i)++) diff --git a/drivers/net/ethernet/marvell/mvmdio.c b/drivers/net/ethernet/marvell/mvmdio.c index 0b9e851f3da4..d14762d93640 100644 --- a/drivers/net/ethernet/marvell/mvmdio.c +++ b/drivers/net/ethernet/marvell/mvmdio.c @@ -347,7 +347,7 @@ static int orion_mdio_probe(struct platform_device *pdev) } - dev->err_interrupt = platform_get_irq(pdev, 0); + dev->err_interrupt = platform_get_irq_optional(pdev, 0); if (dev->err_interrupt > 0 && resource_size(r) < MVMDIO_ERR_INT_MASK + 4) { dev_err(&pdev->dev, diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c index 98017e7d5dd0..11babc79dc6c 100644 --- a/drivers/net/ethernet/marvell/mvneta.c +++ b/drivers/net/ethernet/marvell/mvneta.c @@ -3036,11 +3036,10 @@ static int mvneta_poll(struct napi_struct *napi, int budget) /* For the case where the last mvneta_poll did not process all * RX packets */ - rx_queue = fls(((cause_rx_tx >> 8) & 0xff)); - cause_rx_tx |= pp->neta_armada3700 ? pp->cause_rx_tx : port->cause_rx_tx; + rx_queue = fls(((cause_rx_tx >> 8) & 0xff)); if (rx_queue) { rx_queue = rx_queue - 1; if (pp->bm_priv) diff --git a/drivers/net/ethernet/mellanox/mlx4/mcg.c b/drivers/net/ethernet/mellanox/mlx4/mcg.c index 9c481823b3e8..9486caecfbdc 100644 --- a/drivers/net/ethernet/mellanox/mlx4/mcg.c +++ b/drivers/net/ethernet/mellanox/mlx4/mcg.c @@ -906,59 +906,59 @@ static void mlx4_err_rule(struct mlx4_dev *dev, char *str, int len = 0; mlx4_err(dev, "%s", str); - len += snprintf(buf + len, BUF_SIZE - len, - "port = %d prio = 0x%x qp = 0x%x ", - rule->port, rule->priority, rule->qpn); + len += scnprintf(buf + len, BUF_SIZE - len, + "port = %d prio = 0x%x qp = 0x%x ", + rule->port, rule->priority, rule->qpn); list_for_each_entry(cur, &rule->list, list) { switch (cur->id) { case MLX4_NET_TRANS_RULE_ID_ETH: - len += snprintf(buf + len, BUF_SIZE - len, - "dmac = %pM ", &cur->eth.dst_mac); + len += scnprintf(buf + len, BUF_SIZE - len, + "dmac = %pM ", &cur->eth.dst_mac); if (cur->eth.ether_type) - len += snprintf(buf + len, BUF_SIZE - len, - "ethertype = 0x%x ", - be16_to_cpu(cur->eth.ether_type)); + len += scnprintf(buf + len, BUF_SIZE - len, + "ethertype = 0x%x ", + be16_to_cpu(cur->eth.ether_type)); if (cur->eth.vlan_id) - len += snprintf(buf + len, BUF_SIZE - len, - "vlan-id = %d ", - be16_to_cpu(cur->eth.vlan_id)); + len += scnprintf(buf + len, BUF_SIZE - len, + "vlan-id = %d ", + be16_to_cpu(cur->eth.vlan_id)); break; case MLX4_NET_TRANS_RULE_ID_IPV4: if (cur->ipv4.src_ip) - len += snprintf(buf + len, BUF_SIZE - len, - "src-ip = %pI4 ", - &cur->ipv4.src_ip); + len += scnprintf(buf + len, BUF_SIZE - len, + "src-ip = %pI4 ", + &cur->ipv4.src_ip); if (cur->ipv4.dst_ip) - len += snprintf(buf + len, BUF_SIZE - len, - "dst-ip = %pI4 ", - &cur->ipv4.dst_ip); + len += scnprintf(buf + len, BUF_SIZE - len, + "dst-ip = %pI4 ", + &cur->ipv4.dst_ip); break; case MLX4_NET_TRANS_RULE_ID_TCP: case MLX4_NET_TRANS_RULE_ID_UDP: if (cur->tcp_udp.src_port) - len += snprintf(buf + len, BUF_SIZE - len, - "src-port = %d ", - be16_to_cpu(cur->tcp_udp.src_port)); + len += scnprintf(buf + len, BUF_SIZE - len, + "src-port = %d ", + be16_to_cpu(cur->tcp_udp.src_port)); if (cur->tcp_udp.dst_port) - len += snprintf(buf + len, BUF_SIZE - len, - "dst-port = %d ", - be16_to_cpu(cur->tcp_udp.dst_port)); + len += scnprintf(buf + len, BUF_SIZE - len, + "dst-port = %d ", + be16_to_cpu(cur->tcp_udp.dst_port)); break; case MLX4_NET_TRANS_RULE_ID_IB: - len += snprintf(buf + len, BUF_SIZE - len, - "dst-gid = %pI6\n", cur->ib.dst_gid); - len += snprintf(buf + len, BUF_SIZE - len, - "dst-gid-mask = %pI6\n", - cur->ib.dst_gid_msk); + len += scnprintf(buf + len, BUF_SIZE - len, + "dst-gid = %pI6\n", cur->ib.dst_gid); + len += scnprintf(buf + len, BUF_SIZE - len, + "dst-gid-mask = %pI6\n", + cur->ib.dst_gid_msk); break; case MLX4_NET_TRANS_RULE_ID_VXLAN: - len += snprintf(buf + len, BUF_SIZE - len, - "VNID = %d ", be32_to_cpu(cur->vxlan.vni)); + len += scnprintf(buf + len, BUF_SIZE - len, + "VNID = %d ", be32_to_cpu(cur->vxlan.vni)); break; case MLX4_NET_TRANS_RULE_ID_IPV6: break; @@ -967,7 +967,7 @@ static void mlx4_err_rule(struct mlx4_dev *dev, char *str, break; } } - len += snprintf(buf + len, BUF_SIZE - len, "\n"); + len += scnprintf(buf + len, BUF_SIZE - len, "\n"); mlx4_err(dev, "%s", buf); if (len >= BUF_SIZE) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index 220ef9f06f84..c9606b8ab6ef 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -371,6 +371,7 @@ enum { struct mlx5e_sq_wqe_info { u8 opcode; + u8 num_wqebbs; /* Auxiliary data for different opcodes. */ union { @@ -1059,6 +1060,7 @@ int mlx5e_modify_rq_state(struct mlx5e_rq *rq, int curr_state, int next_state); void mlx5e_activate_rq(struct mlx5e_rq *rq); void mlx5e_deactivate_rq(struct mlx5e_rq *rq); void mlx5e_free_rx_descs(struct mlx5e_rq *rq); +void mlx5e_free_rx_in_progress_descs(struct mlx5e_rq *rq); void mlx5e_activate_icosq(struct mlx5e_icosq *icosq); void mlx5e_deactivate_icosq(struct mlx5e_icosq *icosq); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/health.c b/drivers/net/ethernet/mellanox/mlx5/core/en/health.c index 3a975641f902..20b907dc1e29 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/health.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/health.c @@ -200,7 +200,7 @@ int mlx5e_health_report(struct mlx5e_priv *priv, netdev_err(priv->netdev, err_str); if (!reporter) - return err_ctx->recover(&err_ctx->ctx); + return err_ctx->recover(err_ctx->ctx); return devlink_health_report(reporter, err_str, err_ctx); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/health.h b/drivers/net/ethernet/mellanox/mlx5/core/en/health.h index d3693fa547ac..e54f70d9af22 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/health.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/health.h @@ -10,8 +10,7 @@ static inline bool cqe_syndrome_needs_recover(u8 syndrome) { - return syndrome == MLX5_CQE_SYNDROME_LOCAL_LENGTH_ERR || - syndrome == MLX5_CQE_SYNDROME_LOCAL_QP_OP_ERR || + return syndrome == MLX5_CQE_SYNDROME_LOCAL_QP_OP_ERR || syndrome == MLX5_CQE_SYNDROME_LOCAL_PROT_ERR || syndrome == MLX5_CQE_SYNDROME_WR_FLUSH_ERR; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c index 6c72b592315b..a01e2de2488f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c @@ -90,7 +90,7 @@ static int mlx5e_rx_reporter_err_icosq_cqe_recover(void *ctx) goto out; mlx5e_reset_icosq_cc_pc(icosq); - mlx5e_free_rx_descs(rq); + mlx5e_free_rx_in_progress_descs(rq); clear_bit(MLX5E_SQ_STATE_RECOVERING, &icosq->state); mlx5e_activate_icosq(icosq); mlx5e_activate_rq(rq); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h index 7c8796d9743f..f07b1399744e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h @@ -179,6 +179,16 @@ mlx5e_tx_dma_unmap(struct device *pdev, struct mlx5e_sq_dma *dma) } } +static inline void mlx5e_rqwq_reset(struct mlx5e_rq *rq) +{ + if (rq->wq_type == MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ) { + mlx5_wq_ll_reset(&rq->mpwqe.wq); + rq->mpwqe.actual_wq_head = 0; + } else { + mlx5_wq_cyc_reset(&rq->wqe.wq); + } +} + /* SW parser related functions */ struct mlx5e_swp_spec { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls.h index a3efa29a4629..63116be6b1d6 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls.h @@ -38,8 +38,8 @@ enum { enum { MLX5E_TLS_PROGRESS_PARAMS_RECORD_TRACKER_STATE_START = 0, - MLX5E_TLS_PROGRESS_PARAMS_RECORD_TRACKER_STATE_SEARCHING = 1, - MLX5E_TLS_PROGRESS_PARAMS_RECORD_TRACKER_STATE_TRACKING = 2, + MLX5E_TLS_PROGRESS_PARAMS_RECORD_TRACKER_STATE_TRACKING = 1, + MLX5E_TLS_PROGRESS_PARAMS_RECORD_TRACKER_STATE_SEARCHING = 2, }; struct mlx5e_ktls_offload_context_tx { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_tx.c index f260dd96873b..52a56622034a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_tx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_tx.c @@ -218,7 +218,7 @@ tx_sync_info_get(struct mlx5e_ktls_offload_context_tx *priv_tx, * this packet was already acknowledged and its record info * was released. */ - ends_before = before(tcp_seq + datalen, tls_record_start_seq(record)); + ends_before = before(tcp_seq + datalen - 1, tls_record_start_seq(record)); if (unlikely(tls_record_is_start_marker(record))) { ret = ends_before ? MLX5E_KTLS_SYNC_SKIP_NO_DATA : MLX5E_KTLS_SYNC_FAIL; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index 454d3459bd8b..4ef3dc79f73c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -712,6 +712,9 @@ int mlx5e_modify_rq_state(struct mlx5e_rq *rq, int curr_state, int next_state) if (!in) return -ENOMEM; + if (curr_state == MLX5_RQC_STATE_RST && next_state == MLX5_RQC_STATE_RDY) + mlx5e_rqwq_reset(rq); + rqc = MLX5_ADDR_OF(modify_rq_in, in, ctx); MLX5_SET(modify_rq_in, in, rq_state, curr_state); @@ -810,6 +813,29 @@ int mlx5e_wait_for_min_rx_wqes(struct mlx5e_rq *rq, int wait_time) return -ETIMEDOUT; } +void mlx5e_free_rx_in_progress_descs(struct mlx5e_rq *rq) +{ + struct mlx5_wq_ll *wq; + u16 head; + int i; + + if (rq->wq_type != MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ) + return; + + wq = &rq->mpwqe.wq; + head = wq->head; + + /* Outstanding UMR WQEs (in progress) start at wq->head */ + for (i = 0; i < rq->mpwqe.umr_in_progress; i++) { + rq->dealloc_wqe(rq, head); + head = mlx5_wq_ll_get_wqe_next_ix(wq, head); + } + + rq->mpwqe.actual_wq_head = wq->head; + rq->mpwqe.umr_in_progress = 0; + rq->mpwqe.umr_completed = 0; +} + void mlx5e_free_rx_descs(struct mlx5e_rq *rq) { __be16 wqe_ix_be; @@ -817,14 +843,8 @@ void mlx5e_free_rx_descs(struct mlx5e_rq *rq) if (rq->wq_type == MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ) { struct mlx5_wq_ll *wq = &rq->mpwqe.wq; - u16 head = wq->head; - int i; - /* Outstanding UMR WQEs (in progress) start at wq->head */ - for (i = 0; i < rq->mpwqe.umr_in_progress; i++) { - rq->dealloc_wqe(rq, head); - head = mlx5_wq_ll_get_wqe_next_ix(wq, head); - } + mlx5e_free_rx_in_progress_descs(rq); while (!mlx5_wq_ll_is_empty(wq)) { struct mlx5e_rx_wqe_ll *wqe; @@ -5144,7 +5164,6 @@ static void mlx5e_nic_enable(struct mlx5e_priv *priv) static void mlx5e_nic_disable(struct mlx5e_priv *priv) { - struct net_device *netdev = priv->netdev; struct mlx5_core_dev *mdev = priv->mdev; #ifdef CONFIG_MLX5_CORE_EN_DCB @@ -5165,7 +5184,7 @@ static void mlx5e_nic_disable(struct mlx5e_priv *priv) mlx5e_monitor_counter_cleanup(priv); mlx5e_disable_async_events(priv); - mlx5_lag_remove(mdev, netdev); + mlx5_lag_remove(mdev); } int mlx5e_update_nic_rx(struct mlx5e_priv *priv) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c index 7b48ccacebe2..6ed307d7f191 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c @@ -1861,7 +1861,6 @@ static void mlx5e_uplink_rep_enable(struct mlx5e_priv *priv) static void mlx5e_uplink_rep_disable(struct mlx5e_priv *priv) { - struct net_device *netdev = priv->netdev; struct mlx5_core_dev *mdev = priv->mdev; struct mlx5e_rep_priv *rpriv = priv->ppriv; @@ -1870,7 +1869,7 @@ static void mlx5e_uplink_rep_disable(struct mlx5e_priv *priv) #endif mlx5_notifier_unregister(mdev, &priv->events_nb); cancel_work_sync(&rpriv->uplink_priv.reoffload_flows_work); - mlx5_lag_remove(mdev, netdev); + mlx5_lag_remove(mdev); } static MLX5E_DEFINE_STATS_GRP(sw_rep, 0); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c index 1c3ab69cbd96..312d4692425b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c @@ -477,6 +477,7 @@ static inline void mlx5e_fill_icosq_frag_edge(struct mlx5e_icosq *sq, /* fill sq frag edge with nops to avoid wqe wrapping two pages */ for (; wi < edge_wi; wi++) { wi->opcode = MLX5_OPCODE_NOP; + wi->num_wqebbs = 1; mlx5e_post_nop(wq, sq->sqn, &sq->pc); } } @@ -525,6 +526,7 @@ static int mlx5e_alloc_rx_mpwqe(struct mlx5e_rq *rq, u16 ix) umr_wqe->uctrl.xlt_offset = cpu_to_be16(xlt_offset); sq->db.ico_wqe[pi].opcode = MLX5_OPCODE_UMR; + sq->db.ico_wqe[pi].num_wqebbs = MLX5E_UMR_WQEBBS; sq->db.ico_wqe[pi].umr.rq = rq; sq->pc += MLX5E_UMR_WQEBBS; @@ -621,6 +623,7 @@ void mlx5e_poll_ico_cq(struct mlx5e_cq *cq) ci = mlx5_wq_cyc_ctr2ix(&sq->wq, sqcc); wi = &sq->db.ico_wqe[ci]; + sqcc += wi->num_wqebbs; if (last_wqe && unlikely(get_cqe_opcode(cqe) != MLX5_CQE_REQ)) { netdev_WARN_ONCE(cq->channel->netdev, @@ -631,16 +634,12 @@ void mlx5e_poll_ico_cq(struct mlx5e_cq *cq) break; } - if (likely(wi->opcode == MLX5_OPCODE_UMR)) { - sqcc += MLX5E_UMR_WQEBBS; + if (likely(wi->opcode == MLX5_OPCODE_UMR)) wi->umr.rq->mpwqe.umr_completed++; - } else if (likely(wi->opcode == MLX5_OPCODE_NOP)) { - sqcc++; - } else { + else if (unlikely(wi->opcode != MLX5_OPCODE_NOP)) netdev_WARN_ONCE(cq->channel->netdev, "Bad OPCODE in ICOSQ WQE info: 0x%x\n", wi->opcode); - } } while (!last_wqe); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c index 74091f72c9a8..ec5fc52bf572 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c @@ -2476,10 +2476,11 @@ static int offload_pedit_fields(struct pedit_headers_action *hdrs, continue; if (f->field_bsize == 32) { - mask_be32 = *(__be32 *)&mask; + mask_be32 = (__be32)mask; mask = (__force unsigned long)cpu_to_le32(be32_to_cpu(mask_be32)); } else if (f->field_bsize == 16) { - mask_be16 = *(__be16 *)&mask; + mask_be32 = (__be32)mask; + mask_be16 = *(__be16 *)&mask_be32; mask = (__force unsigned long)cpu_to_le16(be16_to_cpu(mask_be16)); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c index 257a7c9f7a14..800d34ed8a96 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c @@ -78,6 +78,7 @@ void mlx5e_trigger_irq(struct mlx5e_icosq *sq) u16 pi = mlx5_wq_cyc_ctr2ix(wq, sq->pc); sq->db.ico_wqe[pi].opcode = MLX5_OPCODE_NOP; + sq->db.ico_wqe[pi].num_wqebbs = 1; nopwqe = mlx5e_post_nop(wq, sq->sqn, &sq->pc); mlx5e_notify_hw(wq, sq->pc, sq->uar_map, &nopwqe->ctrl); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c index 5acf60b1bbfe..e49acd0c5da5 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c @@ -459,12 +459,16 @@ static void esw_destroy_legacy_table(struct mlx5_eswitch *esw) static int esw_legacy_enable(struct mlx5_eswitch *esw) { - int ret; + struct mlx5_vport *vport; + int ret, i; ret = esw_create_legacy_table(esw); if (ret) return ret; + mlx5_esw_for_each_vf_vport(esw, i, vport, esw->esw_funcs.num_vfs) + vport->info.link_state = MLX5_VPORT_ADMIN_STATE_AUTO; + ret = mlx5_eswitch_enable_pf_vf_vports(esw, MLX5_LEGACY_SRIOV_VPORT_EVENTS); if (ret) esw_destroy_legacy_table(esw); @@ -2452,25 +2456,17 @@ out: int mlx5_eswitch_get_vepa(struct mlx5_eswitch *esw, u8 *setting) { - int err = 0; - if (!esw) return -EOPNOTSUPP; if (!ESW_ALLOWED(esw)) return -EPERM; - mutex_lock(&esw->state_lock); - if (esw->mode != MLX5_ESWITCH_LEGACY) { - err = -EOPNOTSUPP; - goto out; - } + if (esw->mode != MLX5_ESWITCH_LEGACY) + return -EOPNOTSUPP; *setting = esw->fdb_table.legacy.vepa_uplink_rule ? 1 : 0; - -out: - mutex_unlock(&esw->state_lock); - return err; + return 0; } int mlx5_eswitch_set_vport_trust(struct mlx5_eswitch *esw, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c index 979f13bdc203..1a57b2bd74b8 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -1172,7 +1172,7 @@ static int esw_offloads_start(struct mlx5_eswitch *esw, return -EINVAL; } - mlx5_eswitch_disable(esw, true); + mlx5_eswitch_disable(esw, false); mlx5_eswitch_update_num_of_vfs(esw, esw->dev->priv.sriov.num_vfs); err = mlx5_eswitch_enable(esw, MLX5_ESWITCH_OFFLOADS); if (err) { @@ -2065,7 +2065,7 @@ static int esw_offloads_stop(struct mlx5_eswitch *esw, { int err, err1; - mlx5_eswitch_disable(esw, true); + mlx5_eswitch_disable(esw, false); err = mlx5_eswitch_enable(esw, MLX5_ESWITCH_LEGACY); if (err) { NL_SET_ERR_MSG_MOD(extack, "Failed setting eswitch to legacy"); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_chains.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_chains.c index c5a446e295aa..4276194b633f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_chains.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_chains.c @@ -35,7 +35,7 @@ static const unsigned int ESW_POOLS[] = { 4 * 1024 * 1024, 1 * 1024 * 1024, 64 * 1024, - 4 * 1024, }; + 128 }; struct mlx5_esw_chains_priv { struct rhashtable chains_ht; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lag.c b/drivers/net/ethernet/mellanox/mlx5/core/lag.c index b91eabc09fbc..93052b07c76c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lag.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/lag.c @@ -464,9 +464,6 @@ static int mlx5_lag_netdev_event(struct notifier_block *this, struct mlx5_lag *ldev; int changed = 0; - if (!net_eq(dev_net(ndev), &init_net)) - return NOTIFY_DONE; - if ((event != NETDEV_CHANGEUPPER) && (event != NETDEV_CHANGELOWERSTATE)) return NOTIFY_DONE; @@ -586,8 +583,7 @@ void mlx5_lag_add(struct mlx5_core_dev *dev, struct net_device *netdev) if (!ldev->nb.notifier_call) { ldev->nb.notifier_call = mlx5_lag_netdev_event; - if (register_netdevice_notifier_dev_net(netdev, &ldev->nb, - &ldev->nn)) { + if (register_netdevice_notifier_net(&init_net, &ldev->nb)) { ldev->nb.notifier_call = NULL; mlx5_core_err(dev, "Failed to register LAG netdev notifier\n"); } @@ -600,7 +596,7 @@ void mlx5_lag_add(struct mlx5_core_dev *dev, struct net_device *netdev) } /* Must be called with intf_mutex held */ -void mlx5_lag_remove(struct mlx5_core_dev *dev, struct net_device *netdev) +void mlx5_lag_remove(struct mlx5_core_dev *dev) { struct mlx5_lag *ldev; int i; @@ -619,9 +615,10 @@ void mlx5_lag_remove(struct mlx5_core_dev *dev, struct net_device *netdev) break; if (i == MLX5_MAX_PORTS) { - if (ldev->nb.notifier_call) - unregister_netdevice_notifier_dev_net(netdev, &ldev->nb, - &ldev->nn); + if (ldev->nb.notifier_call) { + unregister_netdevice_notifier_net(&init_net, &ldev->nb); + ldev->nb.notifier_call = NULL; + } mlx5_lag_mp_cleanup(ldev); cancel_delayed_work_sync(&ldev->bond_work); mlx5_lag_dev_free(ldev); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lag.h b/drivers/net/ethernet/mellanox/mlx5/core/lag.h index 316ab09e2664..f1068aac6406 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lag.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/lag.h @@ -44,7 +44,6 @@ struct mlx5_lag { struct workqueue_struct *wq; struct delayed_work bond_work; struct notifier_block nb; - struct netdev_net_notifier nn; struct lag_mp lag_mp; }; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h index fcce9e0fc82c..da67b28d6e23 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h @@ -157,7 +157,7 @@ int mlx5_query_qcam_reg(struct mlx5_core_dev *mdev, u32 *qcam, u8 feature_group, u8 access_reg_group); void mlx5_lag_add(struct mlx5_core_dev *dev, struct net_device *netdev); -void mlx5_lag_remove(struct mlx5_core_dev *dev, struct net_device *netdev); +void mlx5_lag_remove(struct mlx5_core_dev *dev); int mlx5_irq_table_init(struct mlx5_core_dev *dev); void mlx5_irq_table_cleanup(struct mlx5_core_dev *dev); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c index 6dec2a550a10..2d93228ff633 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c @@ -933,7 +933,6 @@ static int dr_actions_l2_rewrite(struct mlx5dr_domain *dmn, action->rewrite.data = (void *)ops; action->rewrite.num_of_actions = i; - action->rewrite.chunk->byte_size = i * sizeof(*ops); ret = mlx5dr_send_postsend_action(dmn, action); if (ret) { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c index c7f10d4f8f8d..095ec7b1399d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c @@ -558,7 +558,8 @@ int mlx5dr_send_postsend_action(struct mlx5dr_domain *dmn, int ret; send_info.write.addr = (uintptr_t)action->rewrite.data; - send_info.write.length = action->rewrite.chunk->byte_size; + send_info.write.length = action->rewrite.num_of_actions * + DR_MODIFY_ACTION_SIZE; send_info.write.lkey = 0; send_info.remote_addr = action->rewrite.chunk->mr_addr; send_info.rkey = action->rewrite.chunk->rkey; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.c index c6c7d1defbd7..aade62a9ee5c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.c @@ -2307,7 +2307,9 @@ static int dr_ste_build_src_gvmi_qpn_tag(struct mlx5dr_match_param *value, struct mlx5dr_cmd_vport_cap *vport_cap; struct mlx5dr_domain *dmn = sb->dmn; struct mlx5dr_cmd_caps *caps; + u8 *bit_mask = sb->bit_mask; u8 *tag = hw_ste->tag; + bool source_gvmi_set; DR_STE_SET_TAG(src_gvmi_qp, tag, source_qp, misc, source_sqn); @@ -2328,7 +2330,8 @@ static int dr_ste_build_src_gvmi_qpn_tag(struct mlx5dr_match_param *value, if (!vport_cap) return -EINVAL; - if (vport_cap->vport_gvmi) + source_gvmi_set = MLX5_GET(ste_src_gvmi_qp, bit_mask, source_gvmi); + if (vport_cap->vport_gvmi && source_gvmi_set) MLX5_SET(ste_src_gvmi_qp, tag, source_gvmi, vport_cap->vport_gvmi); misc->source_eswitch_owner_vhca_id = 0; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c index 3abfc8125926..c2027192e21e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c @@ -66,15 +66,20 @@ static int mlx5_cmd_dr_create_flow_table(struct mlx5_flow_root_namespace *ns, struct mlx5_flow_table *next_ft) { struct mlx5dr_table *tbl; + u32 flags; int err; if (mlx5_dr_is_fw_table(ft->flags)) return mlx5_fs_cmd_get_fw_cmds()->create_flow_table(ns, ft, log_size, next_ft); + flags = ft->flags; + /* turn off encap/decap if not supported for sw-str by fw */ + if (!MLX5_CAP_FLOWTABLE(ns->dev, sw_owner_reformat_supported)) + flags = ft->flags & ~(MLX5_FLOW_TABLE_TUNNEL_EN_REFORMAT | + MLX5_FLOW_TABLE_TUNNEL_EN_DECAP); - tbl = mlx5dr_table_create(ns->fs_dr_domain.dr_domain, - ft->level, ft->flags); + tbl = mlx5dr_table_create(ns->fs_dr_domain.dr_domain, ft->level, flags); if (!tbl) { mlx5_core_err(ns->dev, "Failed creating dr flow_table\n"); return -EINVAL; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/vport.c b/drivers/net/ethernet/mellanox/mlx5/core/vport.c index 1faac31f74d0..23f879da9104 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/vport.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/vport.c @@ -1071,6 +1071,9 @@ int mlx5_core_modify_hca_vport_context(struct mlx5_core_dev *dev, MLX5_SET64(hca_vport_context, ctx, port_guid, req->port_guid); if (req->field_select & MLX5_HCA_VPORT_SEL_NODE_GUID) MLX5_SET64(hca_vport_context, ctx, node_guid, req->node_guid); + MLX5_SET(hca_vport_context, ctx, cap_mask1, req->cap_mask1); + MLX5_SET(hca_vport_context, ctx, cap_mask1_field_select, + req->cap_mask1_perm); err = mlx5_cmd_exec(dev, in, in_sz, out, sizeof(out)); ex: kfree(in); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/wq.c b/drivers/net/ethernet/mellanox/mlx5/core/wq.c index 02f7e4a39578..01f075fac276 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/wq.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/wq.c @@ -94,6 +94,13 @@ void mlx5_wq_cyc_wqe_dump(struct mlx5_wq_cyc *wq, u16 ix, u8 nstrides) print_hex_dump(KERN_WARNING, "", DUMP_PREFIX_OFFSET, 16, 1, wqe, len, false); } +void mlx5_wq_cyc_reset(struct mlx5_wq_cyc *wq) +{ + wq->wqe_ctr = 0; + wq->cur_sz = 0; + mlx5_wq_cyc_update_db_record(wq); +} + int mlx5_wq_qp_create(struct mlx5_core_dev *mdev, struct mlx5_wq_param *param, void *qpc, struct mlx5_wq_qp *wq, struct mlx5_wq_ctrl *wq_ctrl) @@ -192,6 +199,19 @@ err_db_free: return err; } +static void mlx5_wq_ll_init_list(struct mlx5_wq_ll *wq) +{ + struct mlx5_wqe_srq_next_seg *next_seg; + int i; + + for (i = 0; i < wq->fbc.sz_m1; i++) { + next_seg = mlx5_wq_ll_get_wqe(wq, i); + next_seg->next_wqe_index = cpu_to_be16(i + 1); + } + next_seg = mlx5_wq_ll_get_wqe(wq, i); + wq->tail_next = &next_seg->next_wqe_index; +} + int mlx5_wq_ll_create(struct mlx5_core_dev *mdev, struct mlx5_wq_param *param, void *wqc, struct mlx5_wq_ll *wq, struct mlx5_wq_ctrl *wq_ctrl) @@ -199,9 +219,7 @@ int mlx5_wq_ll_create(struct mlx5_core_dev *mdev, struct mlx5_wq_param *param, u8 log_wq_stride = MLX5_GET(wq, wqc, log_wq_stride); u8 log_wq_sz = MLX5_GET(wq, wqc, log_wq_sz); struct mlx5_frag_buf_ctrl *fbc = &wq->fbc; - struct mlx5_wqe_srq_next_seg *next_seg; int err; - int i; err = mlx5_db_alloc_node(mdev, &wq_ctrl->db, param->db_numa_node); if (err) { @@ -220,13 +238,7 @@ int mlx5_wq_ll_create(struct mlx5_core_dev *mdev, struct mlx5_wq_param *param, mlx5_init_fbc(wq_ctrl->buf.frags, log_wq_stride, log_wq_sz, fbc); - for (i = 0; i < fbc->sz_m1; i++) { - next_seg = mlx5_wq_ll_get_wqe(wq, i); - next_seg->next_wqe_index = cpu_to_be16(i + 1); - } - next_seg = mlx5_wq_ll_get_wqe(wq, i); - wq->tail_next = &next_seg->next_wqe_index; - + mlx5_wq_ll_init_list(wq); wq_ctrl->mdev = mdev; return 0; @@ -237,6 +249,15 @@ err_db_free: return err; } +void mlx5_wq_ll_reset(struct mlx5_wq_ll *wq) +{ + wq->head = 0; + wq->wqe_ctr = 0; + wq->cur_sz = 0; + mlx5_wq_ll_init_list(wq); + mlx5_wq_ll_update_db_record(wq); +} + void mlx5_wq_destroy(struct mlx5_wq_ctrl *wq_ctrl) { mlx5_frag_buf_free(wq_ctrl->mdev, &wq_ctrl->buf); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/wq.h b/drivers/net/ethernet/mellanox/mlx5/core/wq.h index d9a94bc223c0..4cadc336593f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/wq.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/wq.h @@ -80,6 +80,7 @@ int mlx5_wq_cyc_create(struct mlx5_core_dev *mdev, struct mlx5_wq_param *param, void *wqc, struct mlx5_wq_cyc *wq, struct mlx5_wq_ctrl *wq_ctrl); void mlx5_wq_cyc_wqe_dump(struct mlx5_wq_cyc *wq, u16 ix, u8 nstrides); +void mlx5_wq_cyc_reset(struct mlx5_wq_cyc *wq); int mlx5_wq_qp_create(struct mlx5_core_dev *mdev, struct mlx5_wq_param *param, void *qpc, struct mlx5_wq_qp *wq, @@ -92,6 +93,7 @@ int mlx5_cqwq_create(struct mlx5_core_dev *mdev, struct mlx5_wq_param *param, int mlx5_wq_ll_create(struct mlx5_core_dev *mdev, struct mlx5_wq_param *param, void *wqc, struct mlx5_wq_ll *wq, struct mlx5_wq_ctrl *wq_ctrl); +void mlx5_wq_ll_reset(struct mlx5_wq_ll *wq); void mlx5_wq_destroy(struct mlx5_wq_ctrl *wq_ctrl); diff --git a/drivers/net/ethernet/mellanox/mlxsw/pci.c b/drivers/net/ethernet/mellanox/mlxsw/pci.c index 914c33e46fb4..e9ded1a6e131 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/pci.c +++ b/drivers/net/ethernet/mellanox/mlxsw/pci.c @@ -1322,36 +1322,64 @@ static void mlxsw_pci_mbox_free(struct mlxsw_pci *mlxsw_pci, mbox->mapaddr); } -static int mlxsw_pci_sw_reset(struct mlxsw_pci *mlxsw_pci, - const struct pci_device_id *id) +static int mlxsw_pci_sys_ready_wait(struct mlxsw_pci *mlxsw_pci, + const struct pci_device_id *id, + u32 *p_sys_status) { unsigned long end; - char mrsr_pl[MLXSW_REG_MRSR_LEN]; - int err; + u32 val; - mlxsw_reg_mrsr_pack(mrsr_pl); - err = mlxsw_reg_write(mlxsw_pci->core, MLXSW_REG(mrsr), mrsr_pl); - if (err) - return err; if (id->device == PCI_DEVICE_ID_MELLANOX_SWITCHX2) { msleep(MLXSW_PCI_SW_RESET_TIMEOUT_MSECS); return 0; } - /* We must wait for the HW to become responsive once again. */ + /* We must wait for the HW to become responsive. */ msleep(MLXSW_PCI_SW_RESET_WAIT_MSECS); end = jiffies + msecs_to_jiffies(MLXSW_PCI_SW_RESET_TIMEOUT_MSECS); do { - u32 val = mlxsw_pci_read32(mlxsw_pci, FW_READY); - + val = mlxsw_pci_read32(mlxsw_pci, FW_READY); if ((val & MLXSW_PCI_FW_READY_MASK) == MLXSW_PCI_FW_READY_MAGIC) return 0; cond_resched(); } while (time_before(jiffies, end)); + + *p_sys_status = val & MLXSW_PCI_FW_READY_MASK; + return -EBUSY; } +static int mlxsw_pci_sw_reset(struct mlxsw_pci *mlxsw_pci, + const struct pci_device_id *id) +{ + struct pci_dev *pdev = mlxsw_pci->pdev; + char mrsr_pl[MLXSW_REG_MRSR_LEN]; + u32 sys_status; + int err; + + err = mlxsw_pci_sys_ready_wait(mlxsw_pci, id, &sys_status); + if (err) { + dev_err(&pdev->dev, "Failed to reach system ready status before reset. Status is 0x%x\n", + sys_status); + return err; + } + + mlxsw_reg_mrsr_pack(mrsr_pl); + err = mlxsw_reg_write(mlxsw_pci->core, MLXSW_REG(mrsr), mrsr_pl); + if (err) + return err; + + err = mlxsw_pci_sys_ready_wait(mlxsw_pci, id, &sys_status); + if (err) { + dev_err(&pdev->dev, "Failed to reach system ready status after reset. Status is 0x%x\n", + sys_status); + return err; + } + + return 0; +} + static int mlxsw_pci_alloc_irq_vectors(struct mlxsw_pci *mlxsw_pci) { int err; diff --git a/drivers/net/ethernet/mellanox/mlxsw/pci_hw.h b/drivers/net/ethernet/mellanox/mlxsw/pci_hw.h index e0d7d2d9a0c8..43fa8c85b5d9 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/pci_hw.h +++ b/drivers/net/ethernet/mellanox/mlxsw/pci_hw.h @@ -28,7 +28,7 @@ #define MLXSW_PCI_SW_RESET 0xF0010 #define MLXSW_PCI_SW_RESET_RST_BIT BIT(0) #define MLXSW_PCI_SW_RESET_TIMEOUT_MSECS 900000 -#define MLXSW_PCI_SW_RESET_WAIT_MSECS 100 +#define MLXSW_PCI_SW_RESET_WAIT_MSECS 200 #define MLXSW_PCI_FW_READY 0xA1844 #define MLXSW_PCI_FW_READY_MASK 0xFFFF #define MLXSW_PCI_FW_READY_MAGIC 0x5E diff --git a/drivers/net/ethernet/mellanox/mlxsw/reg.h b/drivers/net/ethernet/mellanox/mlxsw/reg.h index dd6685156396..e05d1d1be2fd 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/reg.h +++ b/drivers/net/ethernet/mellanox/mlxsw/reg.h @@ -3572,7 +3572,7 @@ MLXSW_ITEM32(reg, qeec, mase, 0x10, 31, 1); * When in bytes mode, value is specified in units of 1000bps. * Access: RW */ -MLXSW_ITEM32(reg, qeec, max_shaper_rate, 0x10, 0, 28); +MLXSW_ITEM32(reg, qeec, max_shaper_rate, 0x10, 0, 31); /* reg_qeec_de * DWRR configuration enable. Enables configuration of the dwrr and diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_mr.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_mr.c index 54275624718b..336e5ecc68f8 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_mr.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_mr.c @@ -637,12 +637,12 @@ static int mlxsw_sp_mr_vif_resolve(struct mlxsw_sp_mr_table *mr_table, return 0; err_erif_unresolve: - list_for_each_entry_from_reverse(erve, &mr_vif->route_evif_list, - vif_node) + list_for_each_entry_continue_reverse(erve, &mr_vif->route_evif_list, + vif_node) mlxsw_sp_mr_route_evif_unresolve(mr_table, erve); err_irif_unresolve: - list_for_each_entry_from_reverse(irve, &mr_vif->route_ivif_list, - vif_node) + list_for_each_entry_continue_reverse(irve, &mr_vif->route_ivif_list, + vif_node) mlxsw_sp_mr_route_ivif_unresolve(mr_table, irve); mr_vif->rif = NULL; return err; diff --git a/drivers/net/ethernet/micrel/ks8851_mll.c b/drivers/net/ethernet/micrel/ks8851_mll.c index a41a90c589db..45cc840d8e2e 100644 --- a/drivers/net/ethernet/micrel/ks8851_mll.c +++ b/drivers/net/ethernet/micrel/ks8851_mll.c @@ -157,21 +157,47 @@ static int msg_enable; */ /** - * ks_rdreg8 - read 8 bit register from device + * ks_check_endian - Check whether endianness of the bus is correct * @ks : The chip information - * @offset: The register address * - * Read a 8bit register from the chip, returning the result + * The KS8851-16MLL EESK pin allows selecting the endianness of the 16bit + * bus. To maintain optimum performance, the bus endianness should be set + * such that it matches the endianness of the CPU. */ -static u8 ks_rdreg8(struct ks_net *ks, int offset) + +static int ks_check_endian(struct ks_net *ks) { - u16 data; - u8 shift_bit = offset & 0x03; - u8 shift_data = (offset & 1) << 3; - ks->cmd_reg_cache = (u16) offset | (u16)(BE0 << shift_bit); - iowrite16(ks->cmd_reg_cache, ks->hw_addr_cmd); - data = ioread16(ks->hw_addr); - return (u8)(data >> shift_data); + u16 cider; + + /* + * Read CIDER register first, however read it the "wrong" way around. + * If the endian strap on the KS8851-16MLL in incorrect and the chip + * is operating in different endianness than the CPU, then the meaning + * of BE[3:0] byte-enable bits is also swapped such that: + * BE[3,2,1,0] becomes BE[1,0,3,2] + * + * Luckily for us, the byte-enable bits are the top four MSbits of + * the address register and the CIDER register is at offset 0xc0. + * Hence, by reading address 0xc0c0, which is not impacted by endian + * swapping, we assert either BE[3:2] or BE[1:0] while reading the + * CIDER register. + * + * If the bus configuration is correct, reading 0xc0c0 asserts + * BE[3:2] and this read returns 0x0000, because to read register + * with bottom two LSbits of address set to 0, BE[1:0] must be + * asserted. + * + * If the bus configuration is NOT correct, reading 0xc0c0 asserts + * BE[1:0] and this read returns non-zero 0x8872 value. + */ + iowrite16(BE3 | BE2 | KS_CIDER, ks->hw_addr_cmd); + cider = ioread16(ks->hw_addr); + if (!cider) + return 0; + + netdev_err(ks->netdev, "incorrect EESK endian strap setting\n"); + + return -EINVAL; } /** @@ -190,22 +216,6 @@ static u16 ks_rdreg16(struct ks_net *ks, int offset) } /** - * ks_wrreg8 - write 8bit register value to chip - * @ks: The chip information - * @offset: The register address - * @value: The value to write - * - */ -static void ks_wrreg8(struct ks_net *ks, int offset, u8 value) -{ - u8 shift_bit = (offset & 0x03); - u16 value_write = (u16)(value << ((offset & 1) << 3)); - ks->cmd_reg_cache = (u16)offset | (BE0 << shift_bit); - iowrite16(ks->cmd_reg_cache, ks->hw_addr_cmd); - iowrite16(value_write, ks->hw_addr); -} - -/** * ks_wrreg16 - write 16bit register value to chip * @ks: The chip information * @offset: The register address @@ -324,8 +334,7 @@ static void ks_read_config(struct ks_net *ks) u16 reg_data = 0; /* Regardless of bus width, 8 bit read should always work.*/ - reg_data = ks_rdreg8(ks, KS_CCR) & 0x00FF; - reg_data |= ks_rdreg8(ks, KS_CCR+1) << 8; + reg_data = ks_rdreg16(ks, KS_CCR); /* addr/data bus are multiplexed */ ks->sharedbus = (reg_data & CCR_SHARED) == CCR_SHARED; @@ -429,7 +438,7 @@ static inline void ks_read_qmu(struct ks_net *ks, u16 *buf, u32 len) /* 1. set sudo DMA mode */ ks_wrreg16(ks, KS_RXFDPR, RXFDPR_RXFPAI); - ks_wrreg8(ks, KS_RXQCR, (ks->rc_rxqcr | RXQCR_SDA) & 0xff); + ks_wrreg16(ks, KS_RXQCR, ks->rc_rxqcr | RXQCR_SDA); /* 2. read prepend data */ /** @@ -446,7 +455,7 @@ static inline void ks_read_qmu(struct ks_net *ks, u16 *buf, u32 len) ks_inblk(ks, buf, ALIGN(len, 4)); /* 4. reset sudo DMA Mode */ - ks_wrreg8(ks, KS_RXQCR, ks->rc_rxqcr); + ks_wrreg16(ks, KS_RXQCR, ks->rc_rxqcr); } /** @@ -548,14 +557,17 @@ static irqreturn_t ks_irq(int irq, void *pw) { struct net_device *netdev = pw; struct ks_net *ks = netdev_priv(netdev); + unsigned long flags; u16 status; + spin_lock_irqsave(&ks->statelock, flags); /*this should be the first in IRQ handler */ ks_save_cmd_reg(ks); status = ks_rdreg16(ks, KS_ISR); if (unlikely(!status)) { ks_restore_cmd_reg(ks); + spin_unlock_irqrestore(&ks->statelock, flags); return IRQ_NONE; } @@ -581,6 +593,7 @@ static irqreturn_t ks_irq(int irq, void *pw) ks->netdev->stats.rx_over_errors++; /* this should be the last in IRQ handler*/ ks_restore_cmd_reg(ks); + spin_unlock_irqrestore(&ks->statelock, flags); return IRQ_HANDLED; } @@ -650,6 +663,7 @@ static int ks_net_stop(struct net_device *netdev) /* shutdown RX/TX QMU */ ks_disable_qmu(ks); + ks_disable_int(ks); /* set powermode to soft power down to save power */ ks_set_powermode(ks, PMECR_PM_SOFTDOWN); @@ -679,13 +693,13 @@ static void ks_write_qmu(struct ks_net *ks, u8 *pdata, u16 len) ks->txh.txw[1] = cpu_to_le16(len); /* 1. set sudo-DMA mode */ - ks_wrreg8(ks, KS_RXQCR, (ks->rc_rxqcr | RXQCR_SDA) & 0xff); + ks_wrreg16(ks, KS_RXQCR, ks->rc_rxqcr | RXQCR_SDA); /* 2. write status/lenth info */ ks_outblk(ks, ks->txh.txw, 4); /* 3. write pkt data */ ks_outblk(ks, (u16 *)pdata, ALIGN(len, 4)); /* 4. reset sudo-DMA mode */ - ks_wrreg8(ks, KS_RXQCR, ks->rc_rxqcr); + ks_wrreg16(ks, KS_RXQCR, ks->rc_rxqcr); /* 5. Enqueue Tx(move the pkt from TX buffer into TXQ) */ ks_wrreg16(ks, KS_TXQCR, TXQCR_METFE); /* 6. wait until TXQCR_METFE is auto-cleared */ @@ -706,10 +720,9 @@ static netdev_tx_t ks_start_xmit(struct sk_buff *skb, struct net_device *netdev) { netdev_tx_t retv = NETDEV_TX_OK; struct ks_net *ks = netdev_priv(netdev); + unsigned long flags; - disable_irq(netdev->irq); - ks_disable_int(ks); - spin_lock(&ks->statelock); + spin_lock_irqsave(&ks->statelock, flags); /* Extra space are required: * 4 byte for alignment, 4 for status/length, 4 for CRC @@ -723,9 +736,7 @@ static netdev_tx_t ks_start_xmit(struct sk_buff *skb, struct net_device *netdev) dev_kfree_skb(skb); } else retv = NETDEV_TX_BUSY; - spin_unlock(&ks->statelock); - ks_enable_int(ks); - enable_irq(netdev->irq); + spin_unlock_irqrestore(&ks->statelock, flags); return retv; } @@ -1251,6 +1262,10 @@ static int ks8851_probe(struct platform_device *pdev) goto err_free; } + err = ks_check_endian(ks); + if (err) + goto err_free; + netdev->irq = platform_get_irq(pdev, 0); if ((int)netdev->irq < 0) { diff --git a/drivers/net/ethernet/mscc/ocelot.c b/drivers/net/ethernet/mscc/ocelot.c index 86d543ab1ab9..d3b7373c5961 100644 --- a/drivers/net/ethernet/mscc/ocelot.c +++ b/drivers/net/ethernet/mscc/ocelot.c @@ -2176,24 +2176,29 @@ static int ocelot_init_timestamp(struct ocelot *ocelot) return 0; } -static void ocelot_port_set_mtu(struct ocelot *ocelot, int port, size_t mtu) +/* Configure the maximum SDU (L2 payload) on RX to the value specified in @sdu. + * The length of VLAN tags is accounted for automatically via DEV_MAC_TAGS_CFG. + */ +static void ocelot_port_set_maxlen(struct ocelot *ocelot, int port, size_t sdu) { struct ocelot_port *ocelot_port = ocelot->ports[port]; + int maxlen = sdu + ETH_HLEN + ETH_FCS_LEN; int atop_wm; - ocelot_port_writel(ocelot_port, mtu, DEV_MAC_MAXLEN_CFG); + ocelot_port_writel(ocelot_port, maxlen, DEV_MAC_MAXLEN_CFG); /* Set Pause WM hysteresis - * 152 = 6 * mtu / OCELOT_BUFFER_CELL_SZ - * 101 = 4 * mtu / OCELOT_BUFFER_CELL_SZ + * 152 = 6 * maxlen / OCELOT_BUFFER_CELL_SZ + * 101 = 4 * maxlen / OCELOT_BUFFER_CELL_SZ */ ocelot_write_rix(ocelot, SYS_PAUSE_CFG_PAUSE_ENA | SYS_PAUSE_CFG_PAUSE_STOP(101) | SYS_PAUSE_CFG_PAUSE_START(152), SYS_PAUSE_CFG, port); /* Tail dropping watermark */ - atop_wm = (ocelot->shared_queue_sz - 9 * mtu) / OCELOT_BUFFER_CELL_SZ; - ocelot_write_rix(ocelot, ocelot_wm_enc(9 * mtu), + atop_wm = (ocelot->shared_queue_sz - 9 * maxlen) / + OCELOT_BUFFER_CELL_SZ; + ocelot_write_rix(ocelot, ocelot_wm_enc(9 * maxlen), SYS_ATOP, port); ocelot_write(ocelot, ocelot_wm_enc(atop_wm), SYS_ATOP_TOT_CFG); } @@ -2222,9 +2227,10 @@ void ocelot_init_port(struct ocelot *ocelot, int port) DEV_MAC_HDX_CFG); /* Set Max Length and maximum tags allowed */ - ocelot_port_set_mtu(ocelot, port, VLAN_ETH_FRAME_LEN); + ocelot_port_set_maxlen(ocelot, port, ETH_DATA_LEN); ocelot_port_writel(ocelot_port, DEV_MAC_TAGS_CFG_TAG_ID(ETH_P_8021AD) | DEV_MAC_TAGS_CFG_VLAN_AWR_ENA | + DEV_MAC_TAGS_CFG_VLAN_DBL_AWR_ENA | DEV_MAC_TAGS_CFG_VLAN_LEN_AWR_ENA, DEV_MAC_TAGS_CFG); @@ -2310,18 +2316,18 @@ void ocelot_set_cpu_port(struct ocelot *ocelot, int cpu, * Only one port can be an NPI at the same time. */ if (cpu < ocelot->num_phys_ports) { - int mtu = VLAN_ETH_FRAME_LEN + OCELOT_TAG_LEN; + int sdu = ETH_DATA_LEN + OCELOT_TAG_LEN; ocelot_write(ocelot, QSYS_EXT_CPU_CFG_EXT_CPUQ_MSK_M | QSYS_EXT_CPU_CFG_EXT_CPU_PORT(cpu), QSYS_EXT_CPU_CFG); if (injection == OCELOT_TAG_PREFIX_SHORT) - mtu += OCELOT_SHORT_PREFIX_LEN; + sdu += OCELOT_SHORT_PREFIX_LEN; else if (injection == OCELOT_TAG_PREFIX_LONG) - mtu += OCELOT_LONG_PREFIX_LEN; + sdu += OCELOT_LONG_PREFIX_LEN; - ocelot_port_set_mtu(ocelot, cpu, mtu); + ocelot_port_set_maxlen(ocelot, cpu, sdu); } /* CPU port Injection/Extraction configuration */ diff --git a/drivers/net/ethernet/mscc/ocelot_board.c b/drivers/net/ethernet/mscc/ocelot_board.c index b38820849faa..1135a18019c7 100644 --- a/drivers/net/ethernet/mscc/ocelot_board.c +++ b/drivers/net/ethernet/mscc/ocelot_board.c @@ -114,6 +114,14 @@ static irqreturn_t ocelot_xtr_irq_handler(int irq, void *arg) if (err != 4) break; + /* At this point the IFH was read correctly, so it is safe to + * presume that there is no error. The err needs to be reset + * otherwise a frame could come in CPU queue between the while + * condition and the check for error later on. And in that case + * the new frame is just removed and not processed. + */ + err = 0; + ocelot_parse_ifh(ifh, &info); ocelot_port = ocelot->ports[info.port]; diff --git a/drivers/net/ethernet/neterion/vxge/vxge-config.h b/drivers/net/ethernet/neterion/vxge/vxge-config.h index e678ba379598..628fa9b2f741 100644 --- a/drivers/net/ethernet/neterion/vxge/vxge-config.h +++ b/drivers/net/ethernet/neterion/vxge/vxge-config.h @@ -2045,7 +2045,7 @@ vxge_hw_vpath_strip_fcs_check(struct __vxge_hw_device *hldev, u64 vpath_mask); if ((level >= VXGE_ERR && VXGE_COMPONENT_LL & VXGE_DEBUG_ERR_MASK) || \ (level >= VXGE_TRACE && VXGE_COMPONENT_LL & VXGE_DEBUG_TRACE_MASK))\ if ((mask & VXGE_DEBUG_MASK) == mask) \ - printk(fmt "\n", __VA_ARGS__); \ + printk(fmt "\n", ##__VA_ARGS__); \ } while (0) #else #define vxge_debug_ll(level, mask, fmt, ...) diff --git a/drivers/net/ethernet/neterion/vxge/vxge-main.h b/drivers/net/ethernet/neterion/vxge/vxge-main.h index 59a57ff5e96a..9c86f4f9cd42 100644 --- a/drivers/net/ethernet/neterion/vxge/vxge-main.h +++ b/drivers/net/ethernet/neterion/vxge/vxge-main.h @@ -452,49 +452,49 @@ int vxge_fw_upgrade(struct vxgedev *vdev, char *fw_name, int override); #if (VXGE_DEBUG_LL_CONFIG & VXGE_DEBUG_MASK) #define vxge_debug_ll_config(level, fmt, ...) \ - vxge_debug_ll(level, VXGE_DEBUG_LL_CONFIG, fmt, __VA_ARGS__) + vxge_debug_ll(level, VXGE_DEBUG_LL_CONFIG, fmt, ##__VA_ARGS__) #else #define vxge_debug_ll_config(level, fmt, ...) #endif #if (VXGE_DEBUG_INIT & VXGE_DEBUG_MASK) #define vxge_debug_init(level, fmt, ...) \ - vxge_debug_ll(level, VXGE_DEBUG_INIT, fmt, __VA_ARGS__) + vxge_debug_ll(level, VXGE_DEBUG_INIT, fmt, ##__VA_ARGS__) #else #define vxge_debug_init(level, fmt, ...) #endif #if (VXGE_DEBUG_TX & VXGE_DEBUG_MASK) #define vxge_debug_tx(level, fmt, ...) \ - vxge_debug_ll(level, VXGE_DEBUG_TX, fmt, __VA_ARGS__) + vxge_debug_ll(level, VXGE_DEBUG_TX, fmt, ##__VA_ARGS__) #else #define vxge_debug_tx(level, fmt, ...) #endif #if (VXGE_DEBUG_RX & VXGE_DEBUG_MASK) #define vxge_debug_rx(level, fmt, ...) \ - vxge_debug_ll(level, VXGE_DEBUG_RX, fmt, __VA_ARGS__) + vxge_debug_ll(level, VXGE_DEBUG_RX, fmt, ##__VA_ARGS__) #else #define vxge_debug_rx(level, fmt, ...) #endif #if (VXGE_DEBUG_MEM & VXGE_DEBUG_MASK) #define vxge_debug_mem(level, fmt, ...) \ - vxge_debug_ll(level, VXGE_DEBUG_MEM, fmt, __VA_ARGS__) + vxge_debug_ll(level, VXGE_DEBUG_MEM, fmt, ##__VA_ARGS__) #else #define vxge_debug_mem(level, fmt, ...) #endif #if (VXGE_DEBUG_ENTRYEXIT & VXGE_DEBUG_MASK) #define vxge_debug_entryexit(level, fmt, ...) \ - vxge_debug_ll(level, VXGE_DEBUG_ENTRYEXIT, fmt, __VA_ARGS__) + vxge_debug_ll(level, VXGE_DEBUG_ENTRYEXIT, fmt, ##__VA_ARGS__) #else #define vxge_debug_entryexit(level, fmt, ...) #endif #if (VXGE_DEBUG_INTR & VXGE_DEBUG_MASK) #define vxge_debug_intr(level, fmt, ...) \ - vxge_debug_ll(level, VXGE_DEBUG_INTR, fmt, __VA_ARGS__) + vxge_debug_ll(level, VXGE_DEBUG_INTR, fmt, ##__VA_ARGS__) #else #define vxge_debug_intr(level, fmt, ...) #endif diff --git a/drivers/net/ethernet/netronome/nfp/nfpcore/nfp6000_pcie.c b/drivers/net/ethernet/netronome/nfp/nfpcore/nfp6000_pcie.c index b454db283aef..684e4e036c55 100644 --- a/drivers/net/ethernet/netronome/nfp/nfpcore/nfp6000_pcie.c +++ b/drivers/net/ethernet/netronome/nfp/nfpcore/nfp6000_pcie.c @@ -616,7 +616,7 @@ static int enable_bars(struct nfp6000_pcie *nfp, u16 interface) if (bar->iomem) { int pf; - msg += snprintf(msg, end - msg, "0.0: General/MSI-X SRAM, "); + msg += scnprintf(msg, end - msg, "0.0: General/MSI-X SRAM, "); atomic_inc(&bar->refcnt); bars_free--; @@ -661,7 +661,7 @@ static int enable_bars(struct nfp6000_pcie *nfp, u16 interface) /* Configure, and lock, BAR0.1 for PCIe XPB (MSI-X PBA) */ bar = &nfp->bar[1]; - msg += snprintf(msg, end - msg, "0.1: PCIe XPB/MSI-X PBA, "); + msg += scnprintf(msg, end - msg, "0.1: PCIe XPB/MSI-X PBA, "); atomic_inc(&bar->refcnt); bars_free--; @@ -680,8 +680,8 @@ static int enable_bars(struct nfp6000_pcie *nfp, u16 interface) bar->iomem = ioremap(nfp_bar_resource_start(bar), nfp_bar_resource_len(bar)); if (bar->iomem) { - msg += snprintf(msg, end - msg, - "0.%d: Explicit%d, ", 4 + i, i); + msg += scnprintf(msg, end - msg, + "0.%d: Explicit%d, ", 4 + i, i); atomic_inc(&bar->refcnt); bars_free--; diff --git a/drivers/net/ethernet/pensando/ionic/ionic_dev.c b/drivers/net/ethernet/pensando/ionic/ionic_dev.c index 87f82f36812f..46107de5e6c3 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_dev.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_dev.c @@ -103,7 +103,7 @@ int ionic_heartbeat_check(struct ionic *ionic) { struct ionic_dev *idev = &ionic->idev; unsigned long hb_time; - u32 fw_status; + u8 fw_status; u32 hb; /* wait a little more than one second before testing again */ @@ -111,9 +111,12 @@ int ionic_heartbeat_check(struct ionic *ionic) if (time_before(hb_time, (idev->last_hb_time + ionic->watchdog_period))) return 0; - /* firmware is useful only if fw_status is non-zero */ - fw_status = ioread32(&idev->dev_info_regs->fw_status); - if (!fw_status) + /* firmware is useful only if the running bit is set and + * fw_status != 0xff (bad PCI read) + */ + fw_status = ioread8(&idev->dev_info_regs->fw_status); + if (fw_status == 0xff || + !(fw_status & IONIC_FW_STS_F_RUNNING)) return -ENXIO; /* early FW has no heartbeat, else FW will return non-zero */ diff --git a/drivers/net/ethernet/pensando/ionic/ionic_if.h b/drivers/net/ethernet/pensando/ionic/ionic_if.h index ce07c2931a72..51adf5059834 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_if.h +++ b/drivers/net/ethernet/pensando/ionic/ionic_if.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB OR BSD-2-Clause */ +/* SPDX-License-Identifier: (GPL-2.0 OR Linux-OpenIB) OR BSD-2-Clause */ /* Copyright (c) 2017-2019 Pensando Systems, Inc. All rights reserved. */ #ifndef _IONIC_IF_H_ @@ -2445,6 +2445,7 @@ union ionic_dev_info_regs { u8 version; u8 asic_type; u8 asic_rev; +#define IONIC_FW_STS_F_RUNNING 0x1 u8 fw_status; u32 fw_heartbeat; char fw_version[IONIC_DEVINFO_FWVERS_BUFLEN]; diff --git a/drivers/net/ethernet/pensando/ionic/ionic_lif.c b/drivers/net/ethernet/pensando/ionic/ionic_lif.c index 191271f6260d..938e19ee0bcd 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_lif.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_lif.c @@ -948,18 +948,18 @@ static void ionic_lif_rx_mode(struct ionic_lif *lif, unsigned int rx_mode) int i; #define REMAIN(__x) (sizeof(buf) - (__x)) - i = snprintf(buf, sizeof(buf), "rx_mode 0x%04x -> 0x%04x:", - lif->rx_mode, rx_mode); + i = scnprintf(buf, sizeof(buf), "rx_mode 0x%04x -> 0x%04x:", + lif->rx_mode, rx_mode); if (rx_mode & IONIC_RX_MODE_F_UNICAST) - i += snprintf(&buf[i], REMAIN(i), " RX_MODE_F_UNICAST"); + i += scnprintf(&buf[i], REMAIN(i), " RX_MODE_F_UNICAST"); if (rx_mode & IONIC_RX_MODE_F_MULTICAST) - i += snprintf(&buf[i], REMAIN(i), " RX_MODE_F_MULTICAST"); + i += scnprintf(&buf[i], REMAIN(i), " RX_MODE_F_MULTICAST"); if (rx_mode & IONIC_RX_MODE_F_BROADCAST) - i += snprintf(&buf[i], REMAIN(i), " RX_MODE_F_BROADCAST"); + i += scnprintf(&buf[i], REMAIN(i), " RX_MODE_F_BROADCAST"); if (rx_mode & IONIC_RX_MODE_F_PROMISC) - i += snprintf(&buf[i], REMAIN(i), " RX_MODE_F_PROMISC"); + i += scnprintf(&buf[i], REMAIN(i), " RX_MODE_F_PROMISC"); if (rx_mode & IONIC_RX_MODE_F_ALLMULTI) - i += snprintf(&buf[i], REMAIN(i), " RX_MODE_F_ALLMULTI"); + i += scnprintf(&buf[i], REMAIN(i), " RX_MODE_F_ALLMULTI"); netdev_dbg(lif->netdev, "lif%d %s\n", lif->index, buf); err = ionic_adminq_post_wait(lif, &ctx); @@ -1688,7 +1688,7 @@ static int ionic_set_vf_mac(struct net_device *netdev, int vf, u8 *mac) if (!(is_zero_ether_addr(mac) || is_valid_ether_addr(mac))) return -EINVAL; - down_read(&ionic->vf_op_lock); + down_write(&ionic->vf_op_lock); if (vf >= pci_num_vf(ionic->pdev) || !ionic->vfs) { ret = -EINVAL; @@ -1698,7 +1698,7 @@ static int ionic_set_vf_mac(struct net_device *netdev, int vf, u8 *mac) ether_addr_copy(ionic->vfs[vf].macaddr, mac); } - up_read(&ionic->vf_op_lock); + up_write(&ionic->vf_op_lock); return ret; } @@ -1719,7 +1719,7 @@ static int ionic_set_vf_vlan(struct net_device *netdev, int vf, u16 vlan, if (proto != htons(ETH_P_8021Q)) return -EPROTONOSUPPORT; - down_read(&ionic->vf_op_lock); + down_write(&ionic->vf_op_lock); if (vf >= pci_num_vf(ionic->pdev) || !ionic->vfs) { ret = -EINVAL; @@ -1730,7 +1730,7 @@ static int ionic_set_vf_vlan(struct net_device *netdev, int vf, u16 vlan, ionic->vfs[vf].vlanid = vlan; } - up_read(&ionic->vf_op_lock); + up_write(&ionic->vf_op_lock); return ret; } diff --git a/drivers/net/ethernet/pensando/ionic/ionic_regs.h b/drivers/net/ethernet/pensando/ionic/ionic_regs.h index 03ee5a36472b..2e174f45c030 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_regs.h +++ b/drivers/net/ethernet/pensando/ionic/ionic_regs.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB OR BSD-2-Clause */ +/* SPDX-License-Identifier: (GPL-2.0 OR Linux-OpenIB) OR BSD-2-Clause */ /* Copyright (c) 2018-2019 Pensando Systems, Inc. All rights reserved. */ #ifndef IONIC_REGS_H diff --git a/drivers/net/ethernet/qlogic/qede/qede.h b/drivers/net/ethernet/qlogic/qede/qede.h index e8a1b27db84d..234c6f30effb 100644 --- a/drivers/net/ethernet/qlogic/qede/qede.h +++ b/drivers/net/ethernet/qlogic/qede/qede.h @@ -163,6 +163,8 @@ struct qede_rdma_dev { struct list_head entry; struct list_head rdma_event_list; struct workqueue_struct *rdma_wq; + struct kref refcnt; + struct completion event_comp; bool exp_recovery; }; diff --git a/drivers/net/ethernet/qlogic/qede/qede_rdma.c b/drivers/net/ethernet/qlogic/qede/qede_rdma.c index ffabc2d2f082..2d873ae8a234 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_rdma.c +++ b/drivers/net/ethernet/qlogic/qede/qede_rdma.c @@ -59,6 +59,9 @@ static void _qede_rdma_dev_add(struct qede_dev *edev) static int qede_rdma_create_wq(struct qede_dev *edev) { INIT_LIST_HEAD(&edev->rdma_info.rdma_event_list); + kref_init(&edev->rdma_info.refcnt); + init_completion(&edev->rdma_info.event_comp); + edev->rdma_info.rdma_wq = create_singlethread_workqueue("rdma_wq"); if (!edev->rdma_info.rdma_wq) { DP_NOTICE(edev, "qedr: Could not create workqueue\n"); @@ -83,8 +86,23 @@ static void qede_rdma_cleanup_event(struct qede_dev *edev) } } +static void qede_rdma_complete_event(struct kref *ref) +{ + struct qede_rdma_dev *rdma_dev = + container_of(ref, struct qede_rdma_dev, refcnt); + + /* no more events will be added after this */ + complete(&rdma_dev->event_comp); +} + static void qede_rdma_destroy_wq(struct qede_dev *edev) { + /* Avoid race with add_event flow, make sure it finishes before + * we start accessing the list and cleaning up the work + */ + kref_put(&edev->rdma_info.refcnt, qede_rdma_complete_event); + wait_for_completion(&edev->rdma_info.event_comp); + qede_rdma_cleanup_event(edev); destroy_workqueue(edev->rdma_info.rdma_wq); } @@ -310,15 +328,24 @@ static void qede_rdma_add_event(struct qede_dev *edev, if (!edev->rdma_info.qedr_dev) return; + /* We don't want the cleanup flow to start while we're allocating and + * scheduling the work + */ + if (!kref_get_unless_zero(&edev->rdma_info.refcnt)) + return; /* already being destroyed */ + event_node = qede_rdma_get_free_event_node(edev); if (!event_node) - return; + goto out; event_node->event = event; event_node->ptr = edev; INIT_WORK(&event_node->work, qede_rdma_handle_event); queue_work(edev->rdma_info.rdma_wq, &event_node->work); + +out: + kref_put(&edev->rdma_info.refcnt, qede_rdma_complete_event); } void qede_rdma_dev_event_open(struct qede_dev *edev) diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_init.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_init.c index 07f9067affc6..cda5b0a9e948 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_init.c +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_init.c @@ -1720,7 +1720,7 @@ static int qlcnic_83xx_get_reset_instruction_template(struct qlcnic_adapter *p_d ahw->reset.seq_error = 0; ahw->reset.buff = kzalloc(QLC_83XX_RESTART_TEMPLATE_SIZE, GFP_KERNEL); - if (p_dev->ahw->reset.buff == NULL) + if (ahw->reset.buff == NULL) return -ENOMEM; p_buff = p_dev->ahw->reset.buff; diff --git a/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c b/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c index 06de59521fc4..fbf4cbcf1a65 100644 --- a/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c +++ b/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c @@ -13,25 +13,6 @@ #include "rmnet_vnd.h" #include "rmnet_private.h" -/* Locking scheme - - * The shared resource which needs to be protected is realdev->rx_handler_data. - * For the writer path, this is using rtnl_lock(). The writer paths are - * rmnet_newlink(), rmnet_dellink() and rmnet_force_unassociate_device(). These - * paths are already called with rtnl_lock() acquired in. There is also an - * ASSERT_RTNL() to ensure that we are calling with rtnl acquired. For - * dereference here, we will need to use rtnl_dereference(). Dev list writing - * needs to happen with rtnl_lock() acquired for netdev_master_upper_dev_link(). - * For the reader path, the real_dev->rx_handler_data is called in the TX / RX - * path. We only need rcu_read_lock() for these scenarios. In these cases, - * the rcu_read_lock() is held in __dev_queue_xmit() and - * netif_receive_skb_internal(), so readers need to use rcu_dereference_rtnl() - * to get the relevant information. For dev list reading, we again acquire - * rcu_read_lock() in rmnet_dellink() for netdev_master_upper_dev_get_rcu(). - * We also use unregister_netdevice_many() to free all rmnet devices in - * rmnet_force_unassociate_device() so we dont lose the rtnl_lock() and free in - * same context. - */ - /* Local Definitions and Declarations */ static const struct nla_policy rmnet_policy[IFLA_RMNET_MAX + 1] = { @@ -51,9 +32,10 @@ rmnet_get_port_rtnl(const struct net_device *real_dev) return rtnl_dereference(real_dev->rx_handler_data); } -static int rmnet_unregister_real_device(struct net_device *real_dev, - struct rmnet_port *port) +static int rmnet_unregister_real_device(struct net_device *real_dev) { + struct rmnet_port *port = rmnet_get_port_rtnl(real_dev); + if (port->nr_rmnet_devs) return -EINVAL; @@ -61,9 +43,6 @@ static int rmnet_unregister_real_device(struct net_device *real_dev, kfree(port); - /* release reference on real_dev */ - dev_put(real_dev); - netdev_dbg(real_dev, "Removed from rmnet\n"); return 0; } @@ -89,9 +68,6 @@ static int rmnet_register_real_device(struct net_device *real_dev) return -EBUSY; } - /* hold on to real dev for MAP data */ - dev_hold(real_dev); - for (entry = 0; entry < RMNET_MAX_LOGICAL_EP; entry++) INIT_HLIST_HEAD(&port->muxed_ep[entry]); @@ -99,28 +75,33 @@ static int rmnet_register_real_device(struct net_device *real_dev) return 0; } -static void rmnet_unregister_bridge(struct net_device *dev, - struct rmnet_port *port) +static void rmnet_unregister_bridge(struct rmnet_port *port) { - struct rmnet_port *bridge_port; - struct net_device *bridge_dev; + struct net_device *bridge_dev, *real_dev, *rmnet_dev; + struct rmnet_port *real_port; if (port->rmnet_mode != RMNET_EPMODE_BRIDGE) return; - /* bridge slave handling */ + rmnet_dev = port->rmnet_dev; if (!port->nr_rmnet_devs) { - bridge_dev = port->bridge_ep; + /* bridge device */ + real_dev = port->bridge_ep; + bridge_dev = port->dev; - bridge_port = rmnet_get_port_rtnl(bridge_dev); - bridge_port->bridge_ep = NULL; - bridge_port->rmnet_mode = RMNET_EPMODE_VND; + real_port = rmnet_get_port_rtnl(real_dev); + real_port->bridge_ep = NULL; + real_port->rmnet_mode = RMNET_EPMODE_VND; } else { + /* real device */ bridge_dev = port->bridge_ep; - bridge_port = rmnet_get_port_rtnl(bridge_dev); - rmnet_unregister_real_device(bridge_dev, bridge_port); + port->bridge_ep = NULL; + port->rmnet_mode = RMNET_EPMODE_VND; } + + netdev_upper_dev_unlink(bridge_dev, rmnet_dev); + rmnet_unregister_real_device(bridge_dev); } static int rmnet_newlink(struct net *src_net, struct net_device *dev, @@ -135,6 +116,11 @@ static int rmnet_newlink(struct net *src_net, struct net_device *dev, int err = 0; u16 mux_id; + if (!tb[IFLA_LINK]) { + NL_SET_ERR_MSG_MOD(extack, "link not specified"); + return -EINVAL; + } + real_dev = __dev_get_by_index(src_net, nla_get_u32(tb[IFLA_LINK])); if (!real_dev || !dev) return -ENODEV; @@ -157,7 +143,12 @@ static int rmnet_newlink(struct net *src_net, struct net_device *dev, if (err) goto err1; + err = netdev_upper_dev_link(real_dev, dev, extack); + if (err < 0) + goto err2; + port->rmnet_mode = mode; + port->rmnet_dev = dev; hlist_add_head_rcu(&ep->hlnode, &port->muxed_ep[mux_id]); @@ -173,8 +164,11 @@ static int rmnet_newlink(struct net *src_net, struct net_device *dev, return 0; +err2: + unregister_netdevice(dev); + rmnet_vnd_dellink(mux_id, port, ep); err1: - rmnet_unregister_real_device(real_dev, port); + rmnet_unregister_real_device(real_dev); err0: kfree(ep); return err; @@ -183,77 +177,74 @@ err0: static void rmnet_dellink(struct net_device *dev, struct list_head *head) { struct rmnet_priv *priv = netdev_priv(dev); - struct net_device *real_dev; + struct net_device *real_dev, *bridge_dev; + struct rmnet_port *real_port, *bridge_port; struct rmnet_endpoint *ep; - struct rmnet_port *port; - u8 mux_id; + u8 mux_id = priv->mux_id; real_dev = priv->real_dev; - if (!real_dev || !rmnet_is_real_dev_registered(real_dev)) + if (!rmnet_is_real_dev_registered(real_dev)) return; - port = rmnet_get_port_rtnl(real_dev); - - mux_id = rmnet_vnd_get_mux(dev); + real_port = rmnet_get_port_rtnl(real_dev); + bridge_dev = real_port->bridge_ep; + if (bridge_dev) { + bridge_port = rmnet_get_port_rtnl(bridge_dev); + rmnet_unregister_bridge(bridge_port); + } - ep = rmnet_get_endpoint(port, mux_id); + ep = rmnet_get_endpoint(real_port, mux_id); if (ep) { hlist_del_init_rcu(&ep->hlnode); - rmnet_unregister_bridge(dev, port); - rmnet_vnd_dellink(mux_id, port, ep); + rmnet_vnd_dellink(mux_id, real_port, ep); kfree(ep); } - rmnet_unregister_real_device(real_dev, port); + netdev_upper_dev_unlink(real_dev, dev); + rmnet_unregister_real_device(real_dev); unregister_netdevice_queue(dev, head); } -static void rmnet_force_unassociate_device(struct net_device *dev) +static void rmnet_force_unassociate_device(struct net_device *real_dev) { - struct net_device *real_dev = dev; struct hlist_node *tmp_ep; struct rmnet_endpoint *ep; struct rmnet_port *port; unsigned long bkt_ep; LIST_HEAD(list); - if (!rmnet_is_real_dev_registered(real_dev)) - return; - - ASSERT_RTNL(); - - port = rmnet_get_port_rtnl(dev); - - rcu_read_lock(); - rmnet_unregister_bridge(dev, port); - - hash_for_each_safe(port->muxed_ep, bkt_ep, tmp_ep, ep, hlnode) { - unregister_netdevice_queue(ep->egress_dev, &list); - rmnet_vnd_dellink(ep->mux_id, port, ep); + port = rmnet_get_port_rtnl(real_dev); - hlist_del_init_rcu(&ep->hlnode); - kfree(ep); + if (port->nr_rmnet_devs) { + /* real device */ + rmnet_unregister_bridge(port); + hash_for_each_safe(port->muxed_ep, bkt_ep, tmp_ep, ep, hlnode) { + unregister_netdevice_queue(ep->egress_dev, &list); + netdev_upper_dev_unlink(real_dev, ep->egress_dev); + rmnet_vnd_dellink(ep->mux_id, port, ep); + hlist_del_init_rcu(&ep->hlnode); + kfree(ep); + } + rmnet_unregister_real_device(real_dev); + unregister_netdevice_many(&list); + } else { + rmnet_unregister_bridge(port); } - - rcu_read_unlock(); - unregister_netdevice_many(&list); - - rmnet_unregister_real_device(real_dev, port); } static int rmnet_config_notify_cb(struct notifier_block *nb, unsigned long event, void *data) { - struct net_device *dev = netdev_notifier_info_to_dev(data); + struct net_device *real_dev = netdev_notifier_info_to_dev(data); - if (!dev) + if (!rmnet_is_real_dev_registered(real_dev)) return NOTIFY_DONE; switch (event) { case NETDEV_UNREGISTER: - netdev_dbg(dev, "Kernel unregister\n"); - rmnet_force_unassociate_device(dev); + netdev_dbg(real_dev, "Kernel unregister\n"); + rmnet_force_unassociate_device(real_dev); break; default: @@ -295,16 +286,18 @@ static int rmnet_changelink(struct net_device *dev, struct nlattr *tb[], if (!dev) return -ENODEV; - real_dev = __dev_get_by_index(dev_net(dev), - nla_get_u32(tb[IFLA_LINK])); - - if (!real_dev || !rmnet_is_real_dev_registered(real_dev)) + real_dev = priv->real_dev; + if (!rmnet_is_real_dev_registered(real_dev)) return -ENODEV; port = rmnet_get_port_rtnl(real_dev); if (data[IFLA_RMNET_MUX_ID]) { mux_id = nla_get_u16(data[IFLA_RMNET_MUX_ID]); + if (rmnet_get_endpoint(port, mux_id)) { + NL_SET_ERR_MSG_MOD(extack, "MUX ID already exists"); + return -EINVAL; + } ep = rmnet_get_endpoint(port, priv->mux_id); if (!ep) return -ENODEV; @@ -379,11 +372,10 @@ struct rtnl_link_ops rmnet_link_ops __read_mostly = { .fill_info = rmnet_fill_info, }; -/* Needs either rcu_read_lock() or rtnl lock */ -struct rmnet_port *rmnet_get_port(struct net_device *real_dev) +struct rmnet_port *rmnet_get_port_rcu(struct net_device *real_dev) { if (rmnet_is_real_dev_registered(real_dev)) - return rcu_dereference_rtnl(real_dev->rx_handler_data); + return rcu_dereference_bh(real_dev->rx_handler_data); else return NULL; } @@ -409,7 +401,7 @@ int rmnet_add_bridge(struct net_device *rmnet_dev, struct rmnet_port *port, *slave_port; int err; - port = rmnet_get_port(real_dev); + port = rmnet_get_port_rtnl(real_dev); /* If there is more than one rmnet dev attached, its probably being * used for muxing. Skip the briding in that case @@ -417,6 +409,9 @@ int rmnet_add_bridge(struct net_device *rmnet_dev, if (port->nr_rmnet_devs > 1) return -EINVAL; + if (port->rmnet_mode != RMNET_EPMODE_VND) + return -EINVAL; + if (rmnet_is_real_dev_registered(slave_dev)) return -EBUSY; @@ -424,9 +419,17 @@ int rmnet_add_bridge(struct net_device *rmnet_dev, if (err) return -EBUSY; - slave_port = rmnet_get_port(slave_dev); + err = netdev_master_upper_dev_link(slave_dev, rmnet_dev, NULL, NULL, + extack); + if (err) { + rmnet_unregister_real_device(slave_dev); + return err; + } + + slave_port = rmnet_get_port_rtnl(slave_dev); slave_port->rmnet_mode = RMNET_EPMODE_BRIDGE; slave_port->bridge_ep = real_dev; + slave_port->rmnet_dev = rmnet_dev; port->rmnet_mode = RMNET_EPMODE_BRIDGE; port->bridge_ep = slave_dev; @@ -438,16 +441,9 @@ int rmnet_add_bridge(struct net_device *rmnet_dev, int rmnet_del_bridge(struct net_device *rmnet_dev, struct net_device *slave_dev) { - struct rmnet_priv *priv = netdev_priv(rmnet_dev); - struct net_device *real_dev = priv->real_dev; - struct rmnet_port *port, *slave_port; + struct rmnet_port *port = rmnet_get_port_rtnl(slave_dev); - port = rmnet_get_port(real_dev); - port->rmnet_mode = RMNET_EPMODE_VND; - port->bridge_ep = NULL; - - slave_port = rmnet_get_port(slave_dev); - rmnet_unregister_real_device(slave_dev, slave_port); + rmnet_unregister_bridge(port); netdev_dbg(slave_dev, "removed from rmnet as slave\n"); return 0; @@ -473,8 +469,8 @@ static int __init rmnet_init(void) static void __exit rmnet_exit(void) { - unregister_netdevice_notifier(&rmnet_dev_notifier); rtnl_link_unregister(&rmnet_link_ops); + unregister_netdevice_notifier(&rmnet_dev_notifier); } module_init(rmnet_init) diff --git a/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.h b/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.h index cd0a6bcbe74a..be515982d628 100644 --- a/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.h +++ b/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.h @@ -28,6 +28,7 @@ struct rmnet_port { u8 rmnet_mode; struct hlist_head muxed_ep[RMNET_MAX_LOGICAL_EP]; struct net_device *bridge_ep; + struct net_device *rmnet_dev; }; extern struct rtnl_link_ops rmnet_link_ops; @@ -65,7 +66,7 @@ struct rmnet_priv { struct rmnet_priv_stats stats; }; -struct rmnet_port *rmnet_get_port(struct net_device *real_dev); +struct rmnet_port *rmnet_get_port_rcu(struct net_device *real_dev); struct rmnet_endpoint *rmnet_get_endpoint(struct rmnet_port *port, u8 mux_id); int rmnet_add_bridge(struct net_device *rmnet_dev, struct net_device *slave_dev, diff --git a/drivers/net/ethernet/qualcomm/rmnet/rmnet_handlers.c b/drivers/net/ethernet/qualcomm/rmnet/rmnet_handlers.c index 1b74bc160402..29a7bfa2584d 100644 --- a/drivers/net/ethernet/qualcomm/rmnet/rmnet_handlers.c +++ b/drivers/net/ethernet/qualcomm/rmnet/rmnet_handlers.c @@ -159,6 +159,9 @@ static int rmnet_map_egress_handler(struct sk_buff *skb, static void rmnet_bridge_handler(struct sk_buff *skb, struct net_device *bridge_dev) { + if (skb_mac_header_was_set(skb)) + skb_push(skb, skb->mac_len); + if (bridge_dev) { skb->dev = bridge_dev; dev_queue_xmit(skb); @@ -184,7 +187,7 @@ rx_handler_result_t rmnet_rx_handler(struct sk_buff **pskb) return RX_HANDLER_PASS; dev = skb->dev; - port = rmnet_get_port(dev); + port = rmnet_get_port_rcu(dev); switch (port->rmnet_mode) { case RMNET_EPMODE_VND: @@ -217,7 +220,7 @@ void rmnet_egress_handler(struct sk_buff *skb) skb->dev = priv->real_dev; mux_id = priv->mux_id; - port = rmnet_get_port(skb->dev); + port = rmnet_get_port_rcu(skb->dev); if (!port) goto drop; diff --git a/drivers/net/ethernet/qualcomm/rmnet/rmnet_vnd.c b/drivers/net/ethernet/qualcomm/rmnet/rmnet_vnd.c index 509dfc895a33..26ad40f19c64 100644 --- a/drivers/net/ethernet/qualcomm/rmnet/rmnet_vnd.c +++ b/drivers/net/ethernet/qualcomm/rmnet/rmnet_vnd.c @@ -266,14 +266,6 @@ int rmnet_vnd_dellink(u8 id, struct rmnet_port *port, return 0; } -u8 rmnet_vnd_get_mux(struct net_device *rmnet_dev) -{ - struct rmnet_priv *priv; - - priv = netdev_priv(rmnet_dev); - return priv->mux_id; -} - int rmnet_vnd_do_flow_control(struct net_device *rmnet_dev, int enable) { netdev_dbg(rmnet_dev, "Setting VND TX queue state to %d\n", enable); diff --git a/drivers/net/ethernet/qualcomm/rmnet/rmnet_vnd.h b/drivers/net/ethernet/qualcomm/rmnet/rmnet_vnd.h index 54cbaf3c3bc4..14d77c709d4a 100644 --- a/drivers/net/ethernet/qualcomm/rmnet/rmnet_vnd.h +++ b/drivers/net/ethernet/qualcomm/rmnet/rmnet_vnd.h @@ -16,6 +16,5 @@ int rmnet_vnd_dellink(u8 id, struct rmnet_port *port, struct rmnet_endpoint *ep); void rmnet_vnd_rx_fixup(struct sk_buff *skb, struct net_device *dev); void rmnet_vnd_tx_fixup(struct sk_buff *skb, struct net_device *dev); -u8 rmnet_vnd_get_mux(struct net_device *rmnet_dev); void rmnet_vnd_setup(struct net_device *dev); #endif /* _RMNET_VND_H_ */ diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c index a2168a14794c..791d99b9e1cf 100644 --- a/drivers/net/ethernet/realtek/r8169_main.c +++ b/drivers/net/ethernet/realtek/r8169_main.c @@ -5194,7 +5194,7 @@ static int rtl_alloc_irq(struct rtl8169_private *tp) RTL_W8(tp, Config2, RTL_R8(tp, Config2) & ~MSIEnable); rtl_lock_config_regs(tp); /* fall through */ - case RTL_GIGA_MAC_VER_07 ... RTL_GIGA_MAC_VER_24: + case RTL_GIGA_MAC_VER_07 ... RTL_GIGA_MAC_VER_17: flags = PCI_IRQ_LEGACY; break; default: @@ -5285,6 +5285,13 @@ static int r8169_mdio_register(struct rtl8169_private *tp) if (!tp->phydev) { mdiobus_unregister(new_bus); return -ENODEV; + } else if (!tp->phydev->drv) { + /* Most chip versions fail with the genphy driver. + * Therefore ensure that the dedicated PHY driver is loaded. + */ + dev_err(&pdev->dev, "realtek.ko not loaded, maybe it needs to be added to initramfs?\n"); + mdiobus_unregister(new_bus); + return -EUNATCH; } /* PHY will be woken up in rtl_open() */ @@ -5446,15 +5453,6 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) int chipset, region; int jumbo_max, rc; - /* Some tools for creating an initramfs don't consider softdeps, then - * r8169.ko may be in initramfs, but realtek.ko not. Then the generic - * PHY driver is used that doesn't work with most chip versions. - */ - if (!driver_find("RTL8201CP Ethernet", &mdio_bus_type)) { - dev_err(&pdev->dev, "realtek.ko not loaded, maybe it needs to be added to initramfs?\n"); - return -ENOENT; - } - dev = devm_alloc_etherdev(&pdev->dev, sizeof (*tp)); if (!dev) return -ENOMEM; diff --git a/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c b/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c index c705743d69f7..2cc8184b7e6b 100644 --- a/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c +++ b/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c @@ -2277,7 +2277,7 @@ static int __init sxgbe_cmdline_opt(char *str) if (!str || !*str) return -EINVAL; while ((opt = strsep(&str, ",")) != NULL) { - if (!strncmp(opt, "eee_timer:", 6)) { + if (!strncmp(opt, "eee_timer:", 10)) { if (kstrtoint(opt + 10, 0, &eee_timer)) goto err; } diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c index 52113b7529d6..3f16bd807c6e 100644 --- a/drivers/net/ethernet/sfc/ef10.c +++ b/drivers/net/ethernet/sfc/ef10.c @@ -2853,11 +2853,24 @@ efx_ef10_handle_tx_event(struct efx_channel *channel, efx_qword_t *event) } /* Transmit timestamps are only available for 8XXX series. They result - * in three events per packet. These occur in order, and are: - * - the normal completion event + * in up to three events per packet. These occur in order, and are: + * - the normal completion event (may be omitted) * - the low part of the timestamp * - the high part of the timestamp * + * It's possible for multiple completion events to appear before the + * corresponding timestamps. So we can for example get: + * COMP N + * COMP N+1 + * TS_LO N + * TS_HI N + * TS_LO N+1 + * TS_HI N+1 + * + * In addition it's also possible for the adjacent completions to be + * merged, so we may not see COMP N above. As such, the completion + * events are not very useful here. + * * Each part of the timestamp is itself split across two 16 bit * fields in the event. */ @@ -2865,17 +2878,7 @@ efx_ef10_handle_tx_event(struct efx_channel *channel, efx_qword_t *event) switch (tx_ev_type) { case TX_TIMESTAMP_EVENT_TX_EV_COMPLETION: - /* In case of Queue flush or FLR, we might have received - * the previous TX completion event but not the Timestamp - * events. - */ - if (tx_queue->completed_desc_ptr != tx_queue->ptr_mask) - efx_xmit_done(tx_queue, tx_queue->completed_desc_ptr); - - tx_ev_desc_ptr = EFX_QWORD_FIELD(*event, - ESF_DZ_TX_DESCR_INDX); - tx_queue->completed_desc_ptr = - tx_ev_desc_ptr & tx_queue->ptr_mask; + /* Ignore this event - see above. */ break; case TX_TIMESTAMP_EVENT_TX_EV_TSTAMP_LO: @@ -2887,8 +2890,7 @@ efx_ef10_handle_tx_event(struct efx_channel *channel, efx_qword_t *event) ts_part = efx_ef10_extract_event_ts(event); tx_queue->completed_timestamp_major = ts_part; - efx_xmit_done(tx_queue, tx_queue->completed_desc_ptr); - tx_queue->completed_desc_ptr = tx_queue->ptr_mask; + efx_xmit_done_single(tx_queue); break; default: diff --git a/drivers/net/ethernet/sfc/efx.h b/drivers/net/ethernet/sfc/efx.h index f1bdb04efbe4..95395d67ea2d 100644 --- a/drivers/net/ethernet/sfc/efx.h +++ b/drivers/net/ethernet/sfc/efx.h @@ -20,6 +20,7 @@ netdev_tx_t efx_hard_start_xmit(struct sk_buff *skb, struct net_device *net_dev); netdev_tx_t efx_enqueue_skb(struct efx_tx_queue *tx_queue, struct sk_buff *skb); void efx_xmit_done(struct efx_tx_queue *tx_queue, unsigned int index); +void efx_xmit_done_single(struct efx_tx_queue *tx_queue); int efx_setup_tc(struct net_device *net_dev, enum tc_setup_type type, void *type_data); extern unsigned int efx_piobuf_size; diff --git a/drivers/net/ethernet/sfc/efx_channels.c b/drivers/net/ethernet/sfc/efx_channels.c index aeb5e8aa2f2a..73d4e39b5b16 100644 --- a/drivers/net/ethernet/sfc/efx_channels.c +++ b/drivers/net/ethernet/sfc/efx_channels.c @@ -583,6 +583,7 @@ struct efx_channel *efx_copy_channel(const struct efx_channel *old_channel) if (tx_queue->channel) tx_queue->channel = channel; tx_queue->buffer = NULL; + tx_queue->cb_page = NULL; memset(&tx_queue->txd, 0, sizeof(tx_queue->txd)); } diff --git a/drivers/net/ethernet/sfc/mcdi.c b/drivers/net/ethernet/sfc/mcdi.c index 2713300343c7..15c731d04065 100644 --- a/drivers/net/ethernet/sfc/mcdi.c +++ b/drivers/net/ethernet/sfc/mcdi.c @@ -212,12 +212,14 @@ static void efx_mcdi_send_request(struct efx_nic *efx, unsigned cmd, * progress on a NIC at any one time. So no need for locking. */ for (i = 0; i < hdr_len / 4 && bytes < PAGE_SIZE; i++) - bytes += snprintf(buf + bytes, PAGE_SIZE - bytes, - " %08x", le32_to_cpu(hdr[i].u32[0])); + bytes += scnprintf(buf + bytes, PAGE_SIZE - bytes, + " %08x", + le32_to_cpu(hdr[i].u32[0])); for (i = 0; i < inlen / 4 && bytes < PAGE_SIZE; i++) - bytes += snprintf(buf + bytes, PAGE_SIZE - bytes, - " %08x", le32_to_cpu(inbuf[i].u32[0])); + bytes += scnprintf(buf + bytes, PAGE_SIZE - bytes, + " %08x", + le32_to_cpu(inbuf[i].u32[0])); netif_info(efx, hw, efx->net_dev, "MCDI RPC REQ:%s\n", buf); } @@ -302,15 +304,15 @@ static void efx_mcdi_read_response_header(struct efx_nic *efx) */ for (i = 0; i < hdr_len && bytes < PAGE_SIZE; i++) { efx->type->mcdi_read_response(efx, &hdr, (i * 4), 4); - bytes += snprintf(buf + bytes, PAGE_SIZE - bytes, - " %08x", le32_to_cpu(hdr.u32[0])); + bytes += scnprintf(buf + bytes, PAGE_SIZE - bytes, + " %08x", le32_to_cpu(hdr.u32[0])); } for (i = 0; i < data_len && bytes < PAGE_SIZE; i++) { efx->type->mcdi_read_response(efx, &hdr, mcdi->resp_hdr_len + (i * 4), 4); - bytes += snprintf(buf + bytes, PAGE_SIZE - bytes, - " %08x", le32_to_cpu(hdr.u32[0])); + bytes += scnprintf(buf + bytes, PAGE_SIZE - bytes, + " %08x", le32_to_cpu(hdr.u32[0])); } netif_info(efx, hw, efx->net_dev, "MCDI RPC RESP:%s\n", buf); @@ -1417,9 +1419,11 @@ void efx_mcdi_print_fwver(struct efx_nic *efx, char *buf, size_t len) } ver_words = (__le16 *)MCDI_PTR(outbuf, GET_VERSION_OUT_VERSION); - offset = snprintf(buf, len, "%u.%u.%u.%u", - le16_to_cpu(ver_words[0]), le16_to_cpu(ver_words[1]), - le16_to_cpu(ver_words[2]), le16_to_cpu(ver_words[3])); + offset = scnprintf(buf, len, "%u.%u.%u.%u", + le16_to_cpu(ver_words[0]), + le16_to_cpu(ver_words[1]), + le16_to_cpu(ver_words[2]), + le16_to_cpu(ver_words[3])); /* EF10 may have multiple datapath firmware variants within a * single version. Report which variants are running. @@ -1427,9 +1431,9 @@ void efx_mcdi_print_fwver(struct efx_nic *efx, char *buf, size_t len) if (efx_nic_rev(efx) >= EFX_REV_HUNT_A0) { struct efx_ef10_nic_data *nic_data = efx->nic_data; - offset += snprintf(buf + offset, len - offset, " rx%x tx%x", - nic_data->rx_dpcpu_fw_id, - nic_data->tx_dpcpu_fw_id); + offset += scnprintf(buf + offset, len - offset, " rx%x tx%x", + nic_data->rx_dpcpu_fw_id, + nic_data->tx_dpcpu_fw_id); /* It's theoretically possible for the string to exceed 31 * characters, though in practice the first three version diff --git a/drivers/net/ethernet/sfc/net_driver.h b/drivers/net/ethernet/sfc/net_driver.h index 9f9886f222c8..8164f0edcbf0 100644 --- a/drivers/net/ethernet/sfc/net_driver.h +++ b/drivers/net/ethernet/sfc/net_driver.h @@ -208,8 +208,6 @@ struct efx_tx_buffer { * avoid cache-line ping-pong between the xmit path and the * completion path. * @merge_events: Number of TX merged completion events - * @completed_desc_ptr: Most recent completed pointer - only used with - * timestamping. * @completed_timestamp_major: Top part of the most recent tx timestamp. * @completed_timestamp_minor: Low part of the most recent tx timestamp. * @insert_count: Current insert pointer @@ -269,7 +267,6 @@ struct efx_tx_queue { unsigned int merge_events; unsigned int bytes_compl; unsigned int pkts_compl; - unsigned int completed_desc_ptr; u32 completed_timestamp_major; u32 completed_timestamp_minor; diff --git a/drivers/net/ethernet/sfc/ptp.c b/drivers/net/ethernet/sfc/ptp.c index af15a737c675..59b4f16896a8 100644 --- a/drivers/net/ethernet/sfc/ptp.c +++ b/drivers/net/ethernet/sfc/ptp.c @@ -560,13 +560,45 @@ efx_ptp_mac_nic_to_ktime_correction(struct efx_nic *efx, u32 nic_major, u32 nic_minor, s32 correction) { + u32 sync_timestamp; ktime_t kt = { 0 }; + s16 delta; if (!(nic_major & 0x80000000)) { WARN_ON_ONCE(nic_major >> 16); - /* Use the top bits from the latest sync event. */ - nic_major &= 0xffff; - nic_major |= (last_sync_timestamp_major(efx) & 0xffff0000); + + /* Medford provides 48 bits of timestamp, so we must get the top + * 16 bits from the timesync event state. + * + * We only have the lower 16 bits of the time now, but we do + * have a full resolution timestamp at some point in past. As + * long as the difference between the (real) now and the sync + * is less than 2^15, then we can reconstruct the difference + * between those two numbers using only the lower 16 bits of + * each. + * + * Put another way + * + * a - b = ((a mod k) - b) mod k + * + * when -k/2 < (a-b) < k/2. In our case k is 2^16. We know + * (a mod k) and b, so can calculate the delta, a - b. + * + */ + sync_timestamp = last_sync_timestamp_major(efx); + + /* Because delta is s16 this does an implicit mask down to + * 16 bits which is what we need, assuming + * MEDFORD_TX_SECS_EVENT_BITS is 16. delta is signed so that + * we can deal with the (unlikely) case of sync timestamps + * arriving from the future. + */ + delta = nic_major - sync_timestamp; + + /* Recover the fully specified time now, by applying the offset + * to the (fully specified) sync time. + */ + nic_major = sync_timestamp + delta; kt = ptp->nic_to_kernel_time(nic_major, nic_minor, correction); diff --git a/drivers/net/ethernet/sfc/tx.c b/drivers/net/ethernet/sfc/tx.c index 04d7f41d7ed9..8aafc54a4684 100644 --- a/drivers/net/ethernet/sfc/tx.c +++ b/drivers/net/ethernet/sfc/tx.c @@ -535,6 +535,44 @@ netdev_tx_t efx_hard_start_xmit(struct sk_buff *skb, return efx_enqueue_skb(tx_queue, skb); } +void efx_xmit_done_single(struct efx_tx_queue *tx_queue) +{ + unsigned int pkts_compl = 0, bytes_compl = 0; + unsigned int read_ptr; + bool finished = false; + + read_ptr = tx_queue->read_count & tx_queue->ptr_mask; + + while (!finished) { + struct efx_tx_buffer *buffer = &tx_queue->buffer[read_ptr]; + + if (!efx_tx_buffer_in_use(buffer)) { + struct efx_nic *efx = tx_queue->efx; + + netif_err(efx, hw, efx->net_dev, + "TX queue %d spurious single TX completion\n", + tx_queue->queue); + efx_schedule_reset(efx, RESET_TYPE_TX_SKIP); + return; + } + + /* Need to check the flag before dequeueing. */ + if (buffer->flags & EFX_TX_BUF_SKB) + finished = true; + efx_dequeue_buffer(tx_queue, buffer, &pkts_compl, &bytes_compl); + + ++tx_queue->read_count; + read_ptr = tx_queue->read_count & tx_queue->ptr_mask; + } + + tx_queue->pkts_compl += pkts_compl; + tx_queue->bytes_compl += bytes_compl; + + EFX_WARN_ON_PARANOID(pkts_compl != 1); + + efx_xmit_done_check_empty(tx_queue); +} + void efx_init_tx_queue_core_txq(struct efx_tx_queue *tx_queue) { struct efx_nic *efx = tx_queue->efx; diff --git a/drivers/net/ethernet/sfc/tx_common.c b/drivers/net/ethernet/sfc/tx_common.c index b1571e9789d0..70876df1da69 100644 --- a/drivers/net/ethernet/sfc/tx_common.c +++ b/drivers/net/ethernet/sfc/tx_common.c @@ -80,7 +80,6 @@ void efx_init_tx_queue(struct efx_tx_queue *tx_queue) tx_queue->xmit_more_available = false; tx_queue->timestamping = (efx_ptp_use_mac_tx_timestamps(efx) && tx_queue->channel == efx_ptp_channel(efx)); - tx_queue->completed_desc_ptr = tx_queue->ptr_mask; tx_queue->completed_timestamp_major = 0; tx_queue->completed_timestamp_minor = 0; @@ -210,10 +209,9 @@ static void efx_dequeue_buffers(struct efx_tx_queue *tx_queue, while (read_ptr != stop_index) { struct efx_tx_buffer *buffer = &tx_queue->buffer[read_ptr]; - if (!(buffer->flags & EFX_TX_BUF_OPTION) && - unlikely(buffer->len == 0)) { + if (!efx_tx_buffer_in_use(buffer)) { netif_err(efx, tx_err, efx->net_dev, - "TX queue %d spurious TX completion id %x\n", + "TX queue %d spurious TX completion id %d\n", tx_queue->queue, read_ptr); efx_schedule_reset(efx, RESET_TYPE_TX_SKIP); return; @@ -226,6 +224,19 @@ static void efx_dequeue_buffers(struct efx_tx_queue *tx_queue, } } +void efx_xmit_done_check_empty(struct efx_tx_queue *tx_queue) +{ + if ((int)(tx_queue->read_count - tx_queue->old_write_count) >= 0) { + tx_queue->old_write_count = READ_ONCE(tx_queue->write_count); + if (tx_queue->read_count == tx_queue->old_write_count) { + /* Ensure that read_count is flushed. */ + smp_mb(); + tx_queue->empty_read_count = + tx_queue->read_count | EFX_EMPTY_COUNT_VALID; + } + } +} + void efx_xmit_done(struct efx_tx_queue *tx_queue, unsigned int index) { unsigned int fill_level, pkts_compl = 0, bytes_compl = 0; @@ -256,15 +267,7 @@ void efx_xmit_done(struct efx_tx_queue *tx_queue, unsigned int index) netif_tx_wake_queue(tx_queue->core_txq); } - /* Check whether the hardware queue is now empty */ - if ((int)(tx_queue->read_count - tx_queue->old_write_count) >= 0) { - tx_queue->old_write_count = READ_ONCE(tx_queue->write_count); - if (tx_queue->read_count == tx_queue->old_write_count) { - smp_mb(); - tx_queue->empty_read_count = - tx_queue->read_count | EFX_EMPTY_COUNT_VALID; - } - } + efx_xmit_done_check_empty(tx_queue); } /* Remove buffers put into a tx_queue for the current packet. diff --git a/drivers/net/ethernet/sfc/tx_common.h b/drivers/net/ethernet/sfc/tx_common.h index f92f1fe3a87f..99cf7ce2f36c 100644 --- a/drivers/net/ethernet/sfc/tx_common.h +++ b/drivers/net/ethernet/sfc/tx_common.h @@ -21,6 +21,12 @@ void efx_dequeue_buffer(struct efx_tx_queue *tx_queue, unsigned int *pkts_compl, unsigned int *bytes_compl); +static inline bool efx_tx_buffer_in_use(struct efx_tx_buffer *buffer) +{ + return buffer->len || (buffer->flags & EFX_TX_BUF_OPTION); +} + +void efx_xmit_done_check_empty(struct efx_tx_queue *tx_queue); void efx_xmit_done(struct efx_tx_queue *tx_queue, unsigned int index); void efx_enqueue_unwind(struct efx_tx_queue *tx_queue, diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c index dc50ba13a746..2d5573b3dee1 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c @@ -1411,7 +1411,7 @@ static int rk_gmac_probe(struct platform_device *pdev) ret = rk_gmac_clk_init(plat_dat); if (ret) - return ret; + goto err_remove_config_dt; ret = rk_gmac_powerup(plat_dat->bsp_priv); if (ret) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c index d0356fbd1e43..542784300620 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c @@ -24,6 +24,7 @@ static void dwmac1000_core_init(struct mac_device_info *hw, struct net_device *dev) { + struct stmmac_priv *priv = netdev_priv(dev); void __iomem *ioaddr = hw->pcsr; u32 value = readl(ioaddr + GMAC_CONTROL); int mtu = dev->mtu; @@ -35,7 +36,7 @@ static void dwmac1000_core_init(struct mac_device_info *hw, * Broadcom tags can look like invalid LLC/SNAP packets and cause the * hardware to truncate packets on reception. */ - if (netdev_uses_dsa(dev)) + if (netdev_uses_dsa(dev) || !priv->plat->enh_desc) value &= ~GMAC_CONTROL_ACS; if (mtu > 1500) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index 5836b21edd7e..7da18c9afa01 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -4405,6 +4405,8 @@ static void stmmac_init_fs(struct net_device *dev) { struct stmmac_priv *priv = netdev_priv(dev); + rtnl_lock(); + /* Create per netdev entries */ priv->dbgfs_dir = debugfs_create_dir(dev->name, stmmac_fs_dir); @@ -4416,14 +4418,13 @@ static void stmmac_init_fs(struct net_device *dev) debugfs_create_file("dma_cap", 0444, priv->dbgfs_dir, dev, &stmmac_dma_cap_fops); - register_netdevice_notifier(&stmmac_notifier); + rtnl_unlock(); } static void stmmac_exit_fs(struct net_device *dev) { struct stmmac_priv *priv = netdev_priv(dev); - unregister_netdevice_notifier(&stmmac_notifier); debugfs_remove_recursive(priv->dbgfs_dir); } #endif /* CONFIG_DEBUG_FS */ @@ -4940,14 +4941,14 @@ int stmmac_dvr_remove(struct device *dev) netdev_info(priv->dev, "%s: removing driver", __func__); -#ifdef CONFIG_DEBUG_FS - stmmac_exit_fs(ndev); -#endif stmmac_stop_all_dma(priv); stmmac_mac_set(priv, priv->ioaddr, false); netif_carrier_off(ndev); unregister_netdev(ndev); +#ifdef CONFIG_DEBUG_FS + stmmac_exit_fs(ndev); +#endif phylink_destroy(priv->phylink); if (priv->plat->stmmac_rst) reset_control_assert(priv->plat->stmmac_rst); @@ -5166,6 +5167,7 @@ static int __init stmmac_init(void) /* Create debugfs main directory if it doesn't exist yet */ if (!stmmac_fs_dir) stmmac_fs_dir = debugfs_create_dir(STMMAC_RESOURCE_NAME, NULL); + register_netdevice_notifier(&stmmac_notifier); #endif return 0; @@ -5174,6 +5176,7 @@ static int __init stmmac_init(void) static void __exit stmmac_exit(void) { #ifdef CONFIG_DEBUG_FS + unregister_netdevice_notifier(&stmmac_notifier); debugfs_remove_recursive(stmmac_fs_dir); #endif } diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c index d10ac54bf385..13fafd905db8 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c @@ -663,16 +663,22 @@ int stmmac_get_platform_resources(struct platform_device *pdev, * In case the wake up interrupt is not passed from the platform * so the driver will continue to use the mac irq (ndev->irq) */ - stmmac_res->wol_irq = platform_get_irq_byname(pdev, "eth_wake_irq"); + stmmac_res->wol_irq = + platform_get_irq_byname_optional(pdev, "eth_wake_irq"); if (stmmac_res->wol_irq < 0) { if (stmmac_res->wol_irq == -EPROBE_DEFER) return -EPROBE_DEFER; + dev_info(&pdev->dev, "IRQ eth_wake_irq not found\n"); stmmac_res->wol_irq = stmmac_res->irq; } - stmmac_res->lpi_irq = platform_get_irq_byname(pdev, "eth_lpi"); - if (stmmac_res->lpi_irq == -EPROBE_DEFER) - return -EPROBE_DEFER; + stmmac_res->lpi_irq = + platform_get_irq_byname_optional(pdev, "eth_lpi"); + if (stmmac_res->lpi_irq < 0) { + if (stmmac_res->lpi_irq == -EPROBE_DEFER) + return -EPROBE_DEFER; + dev_info(&pdev->dev, "IRQ eth_lpi not found\n"); + } res = platform_get_resource(pdev, IORESOURCE_MEM, 0); stmmac_res->addr = devm_ioremap_resource(&pdev->dev, res); diff --git a/drivers/net/ethernet/xilinx/ll_temac.h b/drivers/net/ethernet/xilinx/ll_temac.h index 276292bca334..53fb8141f1a6 100644 --- a/drivers/net/ethernet/xilinx/ll_temac.h +++ b/drivers/net/ethernet/xilinx/ll_temac.h @@ -375,10 +375,14 @@ struct temac_local { int tx_bd_next; int tx_bd_tail; int rx_bd_ci; + int rx_bd_tail; /* DMA channel control setup */ u32 tx_chnl_ctrl; u32 rx_chnl_ctrl; + u8 coalesce_count_rx; + + struct delayed_work restart_work; }; /* Wrappers for temac_ior()/temac_iow() function pointers above */ diff --git a/drivers/net/ethernet/xilinx/ll_temac_main.c b/drivers/net/ethernet/xilinx/ll_temac_main.c index 6f11f52c9a9e..9461acec6f70 100644 --- a/drivers/net/ethernet/xilinx/ll_temac_main.c +++ b/drivers/net/ethernet/xilinx/ll_temac_main.c @@ -51,6 +51,7 @@ #include <linux/ip.h> #include <linux/slab.h> #include <linux/interrupt.h> +#include <linux/workqueue.h> #include <linux/dma-mapping.h> #include <linux/processor.h> #include <linux/platform_data/xilinx-ll-temac.h> @@ -367,6 +368,8 @@ static int temac_dma_bd_init(struct net_device *ndev) skb_dma_addr = dma_map_single(ndev->dev.parent, skb->data, XTE_MAX_JUMBO_FRAME_SIZE, DMA_FROM_DEVICE); + if (dma_mapping_error(ndev->dev.parent, skb_dma_addr)) + goto out; lp->rx_bd_v[i].phys = cpu_to_be32(skb_dma_addr); lp->rx_bd_v[i].len = cpu_to_be32(XTE_MAX_JUMBO_FRAME_SIZE); lp->rx_bd_v[i].app0 = cpu_to_be32(STS_CTRL_APP0_IRQONEND); @@ -387,12 +390,13 @@ static int temac_dma_bd_init(struct net_device *ndev) lp->tx_bd_next = 0; lp->tx_bd_tail = 0; lp->rx_bd_ci = 0; + lp->rx_bd_tail = RX_BD_NUM - 1; /* Enable RX DMA transfers */ wmb(); lp->dma_out(lp, RX_CURDESC_PTR, lp->rx_bd_p); lp->dma_out(lp, RX_TAILDESC_PTR, - lp->rx_bd_p + (sizeof(*lp->rx_bd_v) * (RX_BD_NUM - 1))); + lp->rx_bd_p + (sizeof(*lp->rx_bd_v) * lp->rx_bd_tail)); /* Prepare for TX DMA transfer */ lp->dma_out(lp, TX_CURDESC_PTR, lp->tx_bd_p); @@ -788,6 +792,9 @@ static void temac_start_xmit_done(struct net_device *ndev) stat = be32_to_cpu(cur_p->app0); } + /* Matches barrier in temac_start_xmit */ + smp_mb(); + netif_wake_queue(ndev); } @@ -830,9 +837,19 @@ temac_start_xmit(struct sk_buff *skb, struct net_device *ndev) cur_p = &lp->tx_bd_v[lp->tx_bd_tail]; if (temac_check_tx_bd_space(lp, num_frag + 1)) { - if (!netif_queue_stopped(ndev)) - netif_stop_queue(ndev); - return NETDEV_TX_BUSY; + if (netif_queue_stopped(ndev)) + return NETDEV_TX_BUSY; + + netif_stop_queue(ndev); + + /* Matches barrier in temac_start_xmit_done */ + smp_mb(); + + /* Space might have just been freed - check again */ + if (temac_check_tx_bd_space(lp, num_frag)) + return NETDEV_TX_BUSY; + + netif_wake_queue(ndev); } cur_p->app0 = 0; @@ -850,12 +867,16 @@ temac_start_xmit(struct sk_buff *skb, struct net_device *ndev) skb_dma_addr = dma_map_single(ndev->dev.parent, skb->data, skb_headlen(skb), DMA_TO_DEVICE); cur_p->len = cpu_to_be32(skb_headlen(skb)); + if (WARN_ON_ONCE(dma_mapping_error(ndev->dev.parent, skb_dma_addr))) { + dev_kfree_skb_any(skb); + ndev->stats.tx_dropped++; + return NETDEV_TX_OK; + } cur_p->phys = cpu_to_be32(skb_dma_addr); ptr_to_txbd((void *)skb, cur_p); for (ii = 0; ii < num_frag; ii++) { - lp->tx_bd_tail++; - if (lp->tx_bd_tail >= TX_BD_NUM) + if (++lp->tx_bd_tail >= TX_BD_NUM) lp->tx_bd_tail = 0; cur_p = &lp->tx_bd_v[lp->tx_bd_tail]; @@ -863,6 +884,27 @@ temac_start_xmit(struct sk_buff *skb, struct net_device *ndev) skb_frag_address(frag), skb_frag_size(frag), DMA_TO_DEVICE); + if (dma_mapping_error(ndev->dev.parent, skb_dma_addr)) { + if (--lp->tx_bd_tail < 0) + lp->tx_bd_tail = TX_BD_NUM - 1; + cur_p = &lp->tx_bd_v[lp->tx_bd_tail]; + while (--ii >= 0) { + --frag; + dma_unmap_single(ndev->dev.parent, + be32_to_cpu(cur_p->phys), + skb_frag_size(frag), + DMA_TO_DEVICE); + if (--lp->tx_bd_tail < 0) + lp->tx_bd_tail = TX_BD_NUM - 1; + cur_p = &lp->tx_bd_v[lp->tx_bd_tail]; + } + dma_unmap_single(ndev->dev.parent, + be32_to_cpu(cur_p->phys), + skb_headlen(skb), DMA_TO_DEVICE); + dev_kfree_skb_any(skb); + ndev->stats.tx_dropped++; + return NETDEV_TX_OK; + } cur_p->phys = cpu_to_be32(skb_dma_addr); cur_p->len = cpu_to_be32(skb_frag_size(frag)); cur_p->app0 = 0; @@ -884,31 +926,56 @@ temac_start_xmit(struct sk_buff *skb, struct net_device *ndev) return NETDEV_TX_OK; } +static int ll_temac_recv_buffers_available(struct temac_local *lp) +{ + int available; + + if (!lp->rx_skb[lp->rx_bd_ci]) + return 0; + available = 1 + lp->rx_bd_tail - lp->rx_bd_ci; + if (available <= 0) + available += RX_BD_NUM; + return available; +} static void ll_temac_recv(struct net_device *ndev) { struct temac_local *lp = netdev_priv(ndev); - struct sk_buff *skb, *new_skb; - unsigned int bdstat; - struct cdmac_bd *cur_p; - dma_addr_t tail_p, skb_dma_addr; - int length; unsigned long flags; + int rx_bd; + bool update_tail = false; spin_lock_irqsave(&lp->rx_lock, flags); - tail_p = lp->rx_bd_p + sizeof(*lp->rx_bd_v) * lp->rx_bd_ci; - cur_p = &lp->rx_bd_v[lp->rx_bd_ci]; - - bdstat = be32_to_cpu(cur_p->app0); - while ((bdstat & STS_CTRL_APP0_CMPLT)) { + /* Process all received buffers, passing them on network + * stack. After this, the buffer descriptors will be in an + * un-allocated stage, where no skb is allocated for it, and + * they are therefore not available for TEMAC/DMA. + */ + do { + struct cdmac_bd *bd = &lp->rx_bd_v[lp->rx_bd_ci]; + struct sk_buff *skb = lp->rx_skb[lp->rx_bd_ci]; + unsigned int bdstat = be32_to_cpu(bd->app0); + int length; + + /* While this should not normally happen, we can end + * here when GFP_ATOMIC allocations fail, and we + * therefore have un-allocated buffers. + */ + if (!skb) + break; - skb = lp->rx_skb[lp->rx_bd_ci]; - length = be32_to_cpu(cur_p->app4) & 0x3FFF; + /* Loop over all completed buffer descriptors */ + if (!(bdstat & STS_CTRL_APP0_CMPLT)) + break; - dma_unmap_single(ndev->dev.parent, be32_to_cpu(cur_p->phys), + dma_unmap_single(ndev->dev.parent, be32_to_cpu(bd->phys), XTE_MAX_JUMBO_FRAME_SIZE, DMA_FROM_DEVICE); + /* The buffer is not valid for DMA anymore */ + bd->phys = 0; + bd->len = 0; + length = be32_to_cpu(bd->app4) & 0x3FFF; skb_put(skb, length); skb->protocol = eth_type_trans(skb, ndev); skb_checksum_none_assert(skb); @@ -923,43 +990,102 @@ static void ll_temac_recv(struct net_device *ndev) * (back) for proper IP checksum byte order * (be16). */ - skb->csum = htons(be32_to_cpu(cur_p->app3) & 0xFFFF); + skb->csum = htons(be32_to_cpu(bd->app3) & 0xFFFF); skb->ip_summed = CHECKSUM_COMPLETE; } if (!skb_defer_rx_timestamp(skb)) netif_rx(skb); + /* The skb buffer is now owned by network stack above */ + lp->rx_skb[lp->rx_bd_ci] = NULL; ndev->stats.rx_packets++; ndev->stats.rx_bytes += length; - new_skb = netdev_alloc_skb_ip_align(ndev, - XTE_MAX_JUMBO_FRAME_SIZE); - if (!new_skb) { - spin_unlock_irqrestore(&lp->rx_lock, flags); - return; + rx_bd = lp->rx_bd_ci; + if (++lp->rx_bd_ci >= RX_BD_NUM) + lp->rx_bd_ci = 0; + } while (rx_bd != lp->rx_bd_tail); + + /* DMA operations will halt when the last buffer descriptor is + * processed (ie. the one pointed to by RX_TAILDESC_PTR). + * When that happens, no more interrupt events will be + * generated. No IRQ_COAL or IRQ_DLY, and not even an + * IRQ_ERR. To avoid stalling, we schedule a delayed work + * when there is a potential risk of that happening. The work + * will call this function, and thus re-schedule itself until + * enough buffers are available again. + */ + if (ll_temac_recv_buffers_available(lp) < lp->coalesce_count_rx) + schedule_delayed_work(&lp->restart_work, HZ / 1000); + + /* Allocate new buffers for those buffer descriptors that were + * passed to network stack. Note that GFP_ATOMIC allocations + * can fail (e.g. when a larger burst of GFP_ATOMIC + * allocations occurs), so while we try to allocate all + * buffers in the same interrupt where they were processed, we + * continue with what we could get in case of allocation + * failure. Allocation of remaining buffers will be retried + * in following calls. + */ + while (1) { + struct sk_buff *skb; + struct cdmac_bd *bd; + dma_addr_t skb_dma_addr; + + rx_bd = lp->rx_bd_tail + 1; + if (rx_bd >= RX_BD_NUM) + rx_bd = 0; + bd = &lp->rx_bd_v[rx_bd]; + + if (bd->phys) + break; /* All skb's allocated */ + + skb = netdev_alloc_skb_ip_align(ndev, XTE_MAX_JUMBO_FRAME_SIZE); + if (!skb) { + dev_warn(&ndev->dev, "skb alloc failed\n"); + break; } - cur_p->app0 = cpu_to_be32(STS_CTRL_APP0_IRQONEND); - skb_dma_addr = dma_map_single(ndev->dev.parent, new_skb->data, + skb_dma_addr = dma_map_single(ndev->dev.parent, skb->data, XTE_MAX_JUMBO_FRAME_SIZE, DMA_FROM_DEVICE); - cur_p->phys = cpu_to_be32(skb_dma_addr); - cur_p->len = cpu_to_be32(XTE_MAX_JUMBO_FRAME_SIZE); - lp->rx_skb[lp->rx_bd_ci] = new_skb; + if (WARN_ON_ONCE(dma_mapping_error(ndev->dev.parent, + skb_dma_addr))) { + dev_kfree_skb_any(skb); + break; + } - lp->rx_bd_ci++; - if (lp->rx_bd_ci >= RX_BD_NUM) - lp->rx_bd_ci = 0; + bd->phys = cpu_to_be32(skb_dma_addr); + bd->len = cpu_to_be32(XTE_MAX_JUMBO_FRAME_SIZE); + bd->app0 = cpu_to_be32(STS_CTRL_APP0_IRQONEND); + lp->rx_skb[rx_bd] = skb; - cur_p = &lp->rx_bd_v[lp->rx_bd_ci]; - bdstat = be32_to_cpu(cur_p->app0); + lp->rx_bd_tail = rx_bd; + update_tail = true; + } + + /* Move tail pointer when buffers have been allocated */ + if (update_tail) { + lp->dma_out(lp, RX_TAILDESC_PTR, + lp->rx_bd_p + sizeof(*lp->rx_bd_v) * lp->rx_bd_tail); } - lp->dma_out(lp, RX_TAILDESC_PTR, tail_p); spin_unlock_irqrestore(&lp->rx_lock, flags); } +/* Function scheduled to ensure a restart in case of DMA halt + * condition caused by running out of buffer descriptors. + */ +static void ll_temac_restart_work_func(struct work_struct *work) +{ + struct temac_local *lp = container_of(work, struct temac_local, + restart_work.work); + struct net_device *ndev = lp->ndev; + + ll_temac_recv(ndev); +} + static irqreturn_t ll_temac_tx_irq(int irq, void *_ndev) { struct net_device *ndev = _ndev; @@ -1052,6 +1178,8 @@ static int temac_stop(struct net_device *ndev) dev_dbg(&ndev->dev, "temac_close()\n"); + cancel_delayed_work_sync(&lp->restart_work); + free_irq(lp->tx_irq, ndev); free_irq(lp->rx_irq, ndev); @@ -1173,6 +1301,7 @@ static int temac_probe(struct platform_device *pdev) lp->dev = &pdev->dev; lp->options = XTE_OPTION_DEFAULTS; spin_lock_init(&lp->rx_lock); + INIT_DELAYED_WORK(&lp->restart_work, ll_temac_restart_work_func); /* Setup mutex for synchronization of indirect register access */ if (pdata) { @@ -1279,6 +1408,7 @@ static int temac_probe(struct platform_device *pdev) */ lp->tx_chnl_ctrl = 0x10220000; lp->rx_chnl_ctrl = 0xff070000; + lp->coalesce_count_rx = 0x07; /* Finished with the DMA node; drop the reference */ of_node_put(dma_np); @@ -1310,11 +1440,14 @@ static int temac_probe(struct platform_device *pdev) (pdata->tx_irq_count << 16); else lp->tx_chnl_ctrl = 0x10220000; - if (pdata->rx_irq_timeout || pdata->rx_irq_count) + if (pdata->rx_irq_timeout || pdata->rx_irq_count) { lp->rx_chnl_ctrl = (pdata->rx_irq_timeout << 24) | (pdata->rx_irq_count << 16); - else + lp->coalesce_count_rx = pdata->rx_irq_count; + } else { lp->rx_chnl_ctrl = 0xff070000; + lp->coalesce_count_rx = 0x07; + } } /* Error handle returned DMA RX and TX interrupts */ diff --git a/drivers/net/geneve.c b/drivers/net/geneve.c index 75757e9954ba..09f279c0182b 100644 --- a/drivers/net/geneve.c +++ b/drivers/net/geneve.c @@ -1845,8 +1845,6 @@ static void geneve_destroy_tunnels(struct net *net, struct list_head *head) if (!net_eq(dev_net(geneve->dev), net)) unregister_netdevice_queue(geneve->dev, head); } - - WARN_ON_ONCE(!list_empty(&gn->sock_list)); } static void __net_exit geneve_exit_batch_net(struct list_head *net_list) @@ -1861,6 +1859,12 @@ static void __net_exit geneve_exit_batch_net(struct list_head *net_list) /* unregister the devices gathered above */ unregister_netdevice_many(&list); rtnl_unlock(); + + list_for_each_entry(net, net_list, exit_list) { + const struct geneve_net *gn = net_generic(net, geneve_net_id); + + WARN_ON_ONCE(!list_empty(&gn->sock_list)); + } } static struct pernet_operations geneve_net_ops = { diff --git a/drivers/net/hyperv/netvsc.c b/drivers/net/hyperv/netvsc.c index ae3f3084c2ed..1b320bcf150a 100644 --- a/drivers/net/hyperv/netvsc.c +++ b/drivers/net/hyperv/netvsc.c @@ -99,7 +99,7 @@ static struct netvsc_device *alloc_net_device(void) init_waitqueue_head(&net_device->wait_drain); net_device->destroy = false; - net_device->tx_disable = false; + net_device->tx_disable = true; net_device->max_pkt = RNDIS_MAX_PKT_DEFAULT; net_device->pkt_align = RNDIS_PKT_ALIGN_DEFAULT; diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c index 65e12cb07f45..2c0a24c606fc 100644 --- a/drivers/net/hyperv/netvsc_drv.c +++ b/drivers/net/hyperv/netvsc_drv.c @@ -1068,6 +1068,7 @@ static int netvsc_attach(struct net_device *ndev, } /* In any case device is now ready */ + nvdev->tx_disable = false; netif_device_attach(ndev); /* Note: enable and attach happen when sub-channels setup */ @@ -2476,6 +2477,8 @@ static int netvsc_probe(struct hv_device *dev, else net->max_mtu = ETH_DATA_LEN; + nvdev->tx_disable = false; + ret = register_netdevice(net); if (ret != 0) { pr_err("Unable to register netdev.\n"); diff --git a/drivers/net/ifb.c b/drivers/net/ifb.c index 242b9b0943f8..7fe306e76281 100644 --- a/drivers/net/ifb.c +++ b/drivers/net/ifb.c @@ -75,7 +75,7 @@ static void ifb_ri_tasklet(unsigned long _txp) } while ((skb = __skb_dequeue(&txp->tq)) != NULL) { - skb->tc_redirected = 0; + skb->redirected = 0; skb->tc_skip_classify = 1; u64_stats_update_begin(&txp->tsync); @@ -96,7 +96,7 @@ static void ifb_ri_tasklet(unsigned long _txp) rcu_read_unlock(); skb->skb_iif = txp->dev->ifindex; - if (!skb->tc_from_ingress) { + if (!skb->from_ingress) { dev_queue_xmit(skb); } else { skb_pull_rcsum(skb, skb->mac_len); @@ -243,7 +243,7 @@ static netdev_tx_t ifb_xmit(struct sk_buff *skb, struct net_device *dev) txp->rx_bytes += skb->len; u64_stats_update_end(&txp->rsync); - if (!skb->tc_redirected || !skb->skb_iif) { + if (!skb->redirected || !skb->skb_iif) { dev_kfree_skb(skb); dev->stats.rx_dropped++; return NETDEV_TX_OK; diff --git a/drivers/net/ipvlan/ipvlan_core.c b/drivers/net/ipvlan/ipvlan_core.c index 30cd0c4f0be0..8801d093135c 100644 --- a/drivers/net/ipvlan/ipvlan_core.c +++ b/drivers/net/ipvlan/ipvlan_core.c @@ -293,6 +293,7 @@ void ipvlan_process_multicast(struct work_struct *work) } if (dev) dev_put(dev); + cond_resched(); } } @@ -498,19 +499,21 @@ static int ipvlan_process_outbound(struct sk_buff *skb) struct ethhdr *ethh = eth_hdr(skb); int ret = NET_XMIT_DROP; - /* In this mode we dont care about multicast and broadcast traffic */ - if (is_multicast_ether_addr(ethh->h_dest)) { - pr_debug_ratelimited("Dropped {multi|broad}cast of type=[%x]\n", - ntohs(skb->protocol)); - kfree_skb(skb); - goto out; - } - /* The ipvlan is a pseudo-L2 device, so the packets that we receive * will have L2; which need to discarded and processed further * in the net-ns of the main-device. */ if (skb_mac_header_was_set(skb)) { + /* In this mode we dont care about + * multicast and broadcast traffic */ + if (is_multicast_ether_addr(ethh->h_dest)) { + pr_debug_ratelimited( + "Dropped {multi|broad}cast of type=[%x]\n", + ntohs(skb->protocol)); + kfree_skb(skb); + goto out; + } + skb_pull(skb, sizeof(*ethh)); skb->mac_header = (typeof(skb->mac_header))~0U; skb_reset_network_header(skb); diff --git a/drivers/net/ipvlan/ipvlan_main.c b/drivers/net/ipvlan/ipvlan_main.c index a70662261a5a..f195f278a83a 100644 --- a/drivers/net/ipvlan/ipvlan_main.c +++ b/drivers/net/ipvlan/ipvlan_main.c @@ -164,7 +164,6 @@ static void ipvlan_uninit(struct net_device *dev) static int ipvlan_open(struct net_device *dev) { struct ipvl_dev *ipvlan = netdev_priv(dev); - struct net_device *phy_dev = ipvlan->phy_dev; struct ipvl_addr *addr; if (ipvlan->port->mode == IPVLAN_MODE_L3 || @@ -178,7 +177,7 @@ static int ipvlan_open(struct net_device *dev) ipvlan_ht_addr_add(ipvlan, addr); rcu_read_unlock(); - return dev_uc_add(phy_dev, phy_dev->dev_addr); + return 0; } static int ipvlan_stop(struct net_device *dev) @@ -190,8 +189,6 @@ static int ipvlan_stop(struct net_device *dev) dev_uc_unsync(phy_dev, dev); dev_mc_unsync(phy_dev, dev); - dev_uc_del(phy_dev, phy_dev->dev_addr); - rcu_read_lock(); list_for_each_entry_rcu(addr, &ipvlan->addrs, anode) ipvlan_ht_addr_del(addr); diff --git a/drivers/net/macsec.c b/drivers/net/macsec.c index 45bfd99f17fa..92bc2b2df660 100644 --- a/drivers/net/macsec.c +++ b/drivers/net/macsec.c @@ -19,6 +19,7 @@ #include <net/gro_cells.h> #include <net/macsec.h> #include <linux/phy.h> +#include <linux/if_arp.h> #include <uapi/linux/if_macsec.h> @@ -424,6 +425,11 @@ static struct macsec_eth_header *macsec_ethhdr(struct sk_buff *skb) return (struct macsec_eth_header *)skb_mac_header(skb); } +static sci_t dev_to_sci(struct net_device *dev, __be16 port) +{ + return make_sci(dev->dev_addr, port); +} + static void __macsec_pn_wrapped(struct macsec_secy *secy, struct macsec_tx_sa *tx_sa) { @@ -3268,6 +3274,20 @@ static int macsec_set_mac_address(struct net_device *dev, void *p) out: ether_addr_copy(dev->dev_addr, addr->sa_data); + macsec->secy.sci = dev_to_sci(dev, MACSEC_PORT_ES); + + /* If h/w offloading is available, propagate to the device */ + if (macsec_is_offloaded(macsec)) { + const struct macsec_ops *ops; + struct macsec_context ctx; + + ops = macsec_get_ops(macsec, &ctx); + if (ops) { + ctx.secy = &macsec->secy; + macsec_offload(ops->mdo_upd_secy, &ctx); + } + } + return 0; } @@ -3342,6 +3362,7 @@ static const struct device_type macsec_type = { static const struct nla_policy macsec_rtnl_policy[IFLA_MACSEC_MAX + 1] = { [IFLA_MACSEC_SCI] = { .type = NLA_U64 }, + [IFLA_MACSEC_PORT] = { .type = NLA_U16 }, [IFLA_MACSEC_ICV_LEN] = { .type = NLA_U8 }, [IFLA_MACSEC_CIPHER_SUITE] = { .type = NLA_U64 }, [IFLA_MACSEC_WINDOW] = { .type = NLA_U32 }, @@ -3592,11 +3613,6 @@ static bool sci_exists(struct net_device *dev, sci_t sci) return false; } -static sci_t dev_to_sci(struct net_device *dev, __be16 port) -{ - return make_sci(dev->dev_addr, port); -} - static int macsec_add_dev(struct net_device *dev, sci_t sci, u8 icv_len) { struct macsec_dev *macsec = macsec_priv(dev); @@ -3650,6 +3666,8 @@ static int macsec_newlink(struct net *net, struct net_device *dev, real_dev = __dev_get_by_index(net, nla_get_u32(tb[IFLA_LINK])); if (!real_dev) return -ENODEV; + if (real_dev->type != ARPHRD_ETHER) + return -EINVAL; dev->priv_flags |= IFF_MACSEC; diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c index 81aa7adf4801..e7289d67268f 100644 --- a/drivers/net/macvlan.c +++ b/drivers/net/macvlan.c @@ -334,6 +334,8 @@ static void macvlan_process_broadcast(struct work_struct *w) if (src) dev_put(src->dev); consume_skb(skb); + + cond_resched(); } } diff --git a/drivers/net/netdevsim/ipsec.c b/drivers/net/netdevsim/ipsec.c index e27fc1a4516d..3811f1bde84e 100644 --- a/drivers/net/netdevsim/ipsec.c +++ b/drivers/net/netdevsim/ipsec.c @@ -29,9 +29,9 @@ static ssize_t nsim_dbg_netdev_ops_read(struct file *filp, return -ENOMEM; p = buf; - p += snprintf(p, bufsize - (p - buf), - "SA count=%u tx=%u\n", - ipsec->count, ipsec->tx); + p += scnprintf(p, bufsize - (p - buf), + "SA count=%u tx=%u\n", + ipsec->count, ipsec->tx); for (i = 0; i < NSIM_IPSEC_MAX_SA_COUNT; i++) { struct nsim_sa *sap = &ipsec->sa[i]; @@ -39,18 +39,18 @@ static ssize_t nsim_dbg_netdev_ops_read(struct file *filp, if (!sap->used) continue; - p += snprintf(p, bufsize - (p - buf), - "sa[%i] %cx ipaddr=0x%08x %08x %08x %08x\n", - i, (sap->rx ? 'r' : 't'), sap->ipaddr[0], - sap->ipaddr[1], sap->ipaddr[2], sap->ipaddr[3]); - p += snprintf(p, bufsize - (p - buf), - "sa[%i] spi=0x%08x proto=0x%x salt=0x%08x crypt=%d\n", - i, be32_to_cpu(sap->xs->id.spi), - sap->xs->id.proto, sap->salt, sap->crypt); - p += snprintf(p, bufsize - (p - buf), - "sa[%i] key=0x%08x %08x %08x %08x\n", - i, sap->key[0], sap->key[1], - sap->key[2], sap->key[3]); + p += scnprintf(p, bufsize - (p - buf), + "sa[%i] %cx ipaddr=0x%08x %08x %08x %08x\n", + i, (sap->rx ? 'r' : 't'), sap->ipaddr[0], + sap->ipaddr[1], sap->ipaddr[2], sap->ipaddr[3]); + p += scnprintf(p, bufsize - (p - buf), + "sa[%i] spi=0x%08x proto=0x%x salt=0x%08x crypt=%d\n", + i, be32_to_cpu(sap->xs->id.spi), + sap->xs->id.proto, sap->salt, sap->crypt); + p += scnprintf(p, bufsize - (p - buf), + "sa[%i] key=0x%08x %08x %08x %08x\n", + i, sap->key[0], sap->key[1], + sap->key[2], sap->key[3]); } len = simple_read_from_buffer(buffer, count, ppos, buf, p - buf); diff --git a/drivers/net/phy/bcm63xx.c b/drivers/net/phy/bcm63xx.c index 23f1958ba6ad..459fb2069c7e 100644 --- a/drivers/net/phy/bcm63xx.c +++ b/drivers/net/phy/bcm63xx.c @@ -73,6 +73,7 @@ static struct phy_driver bcm63xx_driver[] = { /* same phy as above, with just a different OUI */ .phy_id = 0x002bdc00, .phy_id_mask = 0xfffffc00, + .name = "Broadcom BCM63XX (2)", /* PHY_BASIC_FEATURES */ .flags = PHY_IS_INTERNAL, .config_init = bcm63xx_config_init, diff --git a/drivers/net/phy/broadcom.c b/drivers/net/phy/broadcom.c index 7d68b28bb893..a62229a8b1a4 100644 --- a/drivers/net/phy/broadcom.c +++ b/drivers/net/phy/broadcom.c @@ -410,7 +410,7 @@ static int bcm5481_config_aneg(struct phy_device *phydev) struct device_node *np = phydev->mdio.dev.of_node; int ret; - /* Aneg firsly. */ + /* Aneg firstly. */ ret = genphy_config_aneg(phydev); /* Then we can set up the delay. */ @@ -463,7 +463,7 @@ static int bcm54616s_config_aneg(struct phy_device *phydev) { int ret; - /* Aneg firsly. */ + /* Aneg firstly. */ if (phydev->dev_flags & PHY_BCM_FLAGS_MODE_1000BX) ret = genphy_c37_config_aneg(phydev); else diff --git a/drivers/net/phy/dp83867.c b/drivers/net/phy/dp83867.c index 967f57ed0b65..9a07ad137c2e 100644 --- a/drivers/net/phy/dp83867.c +++ b/drivers/net/phy/dp83867.c @@ -28,7 +28,8 @@ #define DP83867_CTRL 0x1f /* Extended Registers */ -#define DP83867_CFG4 0x0031 +#define DP83867_FLD_THR_CFG 0x002e +#define DP83867_CFG4 0x0031 #define DP83867_CFG4_SGMII_ANEG_MASK (BIT(5) | BIT(6)) #define DP83867_CFG4_SGMII_ANEG_TIMER_11MS (3 << 5) #define DP83867_CFG4_SGMII_ANEG_TIMER_800US (2 << 5) @@ -91,6 +92,7 @@ #define DP83867_STRAP_STS2_CLK_SKEW_RX_MASK GENMASK(2, 0) #define DP83867_STRAP_STS2_CLK_SKEW_RX_SHIFT 0 #define DP83867_STRAP_STS2_CLK_SKEW_NONE BIT(2) +#define DP83867_STRAP_STS2_STRAP_FLD BIT(10) /* PHY CTRL bits */ #define DP83867_PHYCR_TX_FIFO_DEPTH_SHIFT 14 @@ -125,6 +127,9 @@ /* CFG4 bits */ #define DP83867_CFG4_PORT_MIRROR_EN BIT(0) +/* FLD_THR_CFG */ +#define DP83867_FLD_THR_CFG_ENERGY_LOST_THR_MASK 0x7 + enum { DP83867_PORT_MIRROING_KEEP, DP83867_PORT_MIRROING_EN, @@ -476,6 +481,20 @@ static int dp83867_config_init(struct phy_device *phydev) phy_clear_bits_mmd(phydev, DP83867_DEVADDR, DP83867_CFG4, BIT(7)); + bs = phy_read_mmd(phydev, DP83867_DEVADDR, DP83867_STRAP_STS2); + if (bs & DP83867_STRAP_STS2_STRAP_FLD) { + /* When using strap to enable FLD, the ENERGY_LOST_FLD_THR will + * be set to 0x2. This may causes the PHY link to be unstable - + * the default value 0x1 need to be restored. + */ + ret = phy_modify_mmd(phydev, DP83867_DEVADDR, + DP83867_FLD_THR_CFG, + DP83867_FLD_THR_CFG_ENERGY_LOST_THR_MASK, + 0x1); + if (ret) + return ret; + } + if (phy_interface_is_rgmii(phydev) || phydev->interface == PHY_INTERFACE_MODE_SGMII) { val = phy_read(phydev, MII_DP83867_PHYCTRL); diff --git a/drivers/net/phy/marvell.c b/drivers/net/phy/marvell.c index 28e33ece4ce1..9a8badafea8a 100644 --- a/drivers/net/phy/marvell.c +++ b/drivers/net/phy/marvell.c @@ -1306,6 +1306,9 @@ static int marvell_read_status_page_an(struct phy_device *phydev, } } + if (!(status & MII_M1011_PHY_STATUS_RESOLVED)) + return 0; + if (status & MII_M1011_PHY_STATUS_FULLDUPLEX) phydev->duplex = DUPLEX_FULL; else @@ -1365,6 +1368,8 @@ static int marvell_read_status_page(struct phy_device *phydev, int page) linkmode_zero(phydev->lp_advertising); phydev->pause = 0; phydev->asym_pause = 0; + phydev->speed = SPEED_UNKNOWN; + phydev->duplex = DUPLEX_UNKNOWN; if (phydev->autoneg == AUTONEG_ENABLE) err = marvell_read_status_page_an(phydev, fiber, status); diff --git a/drivers/net/phy/mdio-bcm-iproc.c b/drivers/net/phy/mdio-bcm-iproc.c index 7e9975d25066..f1ded03f0229 100644 --- a/drivers/net/phy/mdio-bcm-iproc.c +++ b/drivers/net/phy/mdio-bcm-iproc.c @@ -178,6 +178,23 @@ static int iproc_mdio_remove(struct platform_device *pdev) return 0; } +#ifdef CONFIG_PM_SLEEP +int iproc_mdio_resume(struct device *dev) +{ + struct platform_device *pdev = to_platform_device(dev); + struct iproc_mdio_priv *priv = platform_get_drvdata(pdev); + + /* restore the mii clock configuration */ + iproc_mdio_config_clk(priv->base); + + return 0; +} + +static const struct dev_pm_ops iproc_mdio_pm_ops = { + .resume = iproc_mdio_resume +}; +#endif /* CONFIG_PM_SLEEP */ + static const struct of_device_id iproc_mdio_of_match[] = { { .compatible = "brcm,iproc-mdio", }, { /* sentinel */ }, @@ -188,6 +205,9 @@ static struct platform_driver iproc_mdio_driver = { .driver = { .name = "iproc-mdio", .of_match_table = iproc_mdio_of_match, +#ifdef CONFIG_PM_SLEEP + .pm = &iproc_mdio_pm_ops, +#endif }, .probe = iproc_mdio_probe, .remove = iproc_mdio_remove, diff --git a/drivers/net/phy/mdio-bcm-unimac.c b/drivers/net/phy/mdio-bcm-unimac.c index 4a28fb29adaa..fbd36891ee64 100644 --- a/drivers/net/phy/mdio-bcm-unimac.c +++ b/drivers/net/phy/mdio-bcm-unimac.c @@ -242,11 +242,9 @@ static int unimac_mdio_probe(struct platform_device *pdev) return -ENOMEM; } - priv->clk = devm_clk_get(&pdev->dev, NULL); - if (PTR_ERR(priv->clk) == -EPROBE_DEFER) + priv->clk = devm_clk_get_optional(&pdev->dev, NULL); + if (IS_ERR(priv->clk)) return PTR_ERR(priv->clk); - else - priv->clk = NULL; ret = clk_prepare_enable(priv->clk); if (ret) diff --git a/drivers/net/phy/mdio-mux-bcm-iproc.c b/drivers/net/phy/mdio-mux-bcm-iproc.c index 88d409e48c1f..aad6809ebe39 100644 --- a/drivers/net/phy/mdio-mux-bcm-iproc.c +++ b/drivers/net/phy/mdio-mux-bcm-iproc.c @@ -288,8 +288,13 @@ static int mdio_mux_iproc_suspend(struct device *dev) static int mdio_mux_iproc_resume(struct device *dev) { struct iproc_mdiomux_desc *md = dev_get_drvdata(dev); + int rc; - clk_prepare_enable(md->core_clk); + rc = clk_prepare_enable(md->core_clk); + if (rc) { + dev_err(md->dev, "failed to enable core clk\n"); + return rc; + } mdio_mux_iproc_config(md); return 0; diff --git a/drivers/net/phy/mscc.c b/drivers/net/phy/mscc.c index 937ac7da2789..f686f40f6bdc 100644 --- a/drivers/net/phy/mscc.c +++ b/drivers/net/phy/mscc.c @@ -345,11 +345,11 @@ enum macsec_bank { BIT(VSC8531_FORCE_LED_OFF) | \ BIT(VSC8531_FORCE_LED_ON)) -#define MSCC_VSC8584_REVB_INT8051_FW "mscc_vsc8584_revb_int8051_fb48.bin" +#define MSCC_VSC8584_REVB_INT8051_FW "microchip/mscc_vsc8584_revb_int8051_fb48.bin" #define MSCC_VSC8584_REVB_INT8051_FW_START_ADDR 0xe800 #define MSCC_VSC8584_REVB_INT8051_FW_CRC 0xfb48 -#define MSCC_VSC8574_REVB_INT8051_FW "mscc_vsc8574_revb_int8051_29e8.bin" +#define MSCC_VSC8574_REVB_INT8051_FW "microchip/mscc_vsc8574_revb_int8051_29e8.bin" #define MSCC_VSC8574_REVB_INT8051_FW_START_ADDR 0x4000 #define MSCC_VSC8574_REVB_INT8051_FW_CRC 0x29e8 diff --git a/drivers/net/phy/phy-c45.c b/drivers/net/phy/phy-c45.c index a1caeee12236..dd2e23fb67c0 100644 --- a/drivers/net/phy/phy-c45.c +++ b/drivers/net/phy/phy-c45.c @@ -167,7 +167,7 @@ EXPORT_SYMBOL_GPL(genphy_c45_restart_aneg); */ int genphy_c45_check_and_restart_aneg(struct phy_device *phydev, bool restart) { - int ret = 0; + int ret; if (!restart) { /* Configure and restart aneg if it wasn't set before */ @@ -180,9 +180,9 @@ int genphy_c45_check_and_restart_aneg(struct phy_device *phydev, bool restart) } if (restart) - ret = genphy_c45_restart_aneg(phydev); + return genphy_c45_restart_aneg(phydev); - return ret; + return 0; } EXPORT_SYMBOL_GPL(genphy_c45_check_and_restart_aneg); diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c index d76e038cf2cb..355bfdef48d2 100644 --- a/drivers/net/phy/phy.c +++ b/drivers/net/phy/phy.c @@ -727,7 +727,8 @@ static irqreturn_t phy_interrupt(int irq, void *phy_dat) phy_trigger_machine(phydev); } - if (phy_clear_interrupt(phydev)) + /* did_interrupt() may have cleared the interrupt already */ + if (!phydev->drv->did_interrupt && phy_clear_interrupt(phydev)) goto phy_err; return IRQ_HANDLED; diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c index 6a5056e0ae77..28e3c5c0e3c3 100644 --- a/drivers/net/phy/phy_device.c +++ b/drivers/net/phy/phy_device.c @@ -247,7 +247,7 @@ static bool mdio_bus_phy_may_suspend(struct phy_device *phydev) * MDIO bus driver and clock gated at this point. */ if (!netdev) - return !phydev->suspended; + goto out; if (netdev->wol_enabled) return false; @@ -267,7 +267,8 @@ static bool mdio_bus_phy_may_suspend(struct phy_device *phydev) if (device_may_wakeup(&netdev->dev)) return false; - return true; +out: + return !phydev->suspended; } static int mdio_bus_phy_suspend(struct device *dev) @@ -285,6 +286,8 @@ static int mdio_bus_phy_suspend(struct device *dev) if (!mdio_bus_phy_may_suspend(phydev)) return 0; + phydev->suspended_by_mdio_bus = 1; + return phy_suspend(phydev); } @@ -293,9 +296,11 @@ static int mdio_bus_phy_resume(struct device *dev) struct phy_device *phydev = to_phy_device(dev); int ret; - if (!mdio_bus_phy_may_suspend(phydev)) + if (!phydev->suspended_by_mdio_bus) goto no_resume; + phydev->suspended_by_mdio_bus = 0; + ret = phy_resume(phydev); if (ret < 0) return ret; @@ -1792,7 +1797,7 @@ EXPORT_SYMBOL(genphy_restart_aneg); */ int genphy_check_and_restart_aneg(struct phy_device *phydev, bool restart) { - int ret = 0; + int ret; if (!restart) { /* Advertisement hasn't changed, but maybe aneg was never on to @@ -1807,9 +1812,9 @@ int genphy_check_and_restart_aneg(struct phy_device *phydev, bool restart) } if (restart) - ret = genphy_restart_aneg(phydev); + return genphy_restart_aneg(phydev); - return ret; + return 0; } EXPORT_SYMBOL(genphy_check_and_restart_aneg); diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c index 70b9a143db84..6e66b8e77ec7 100644 --- a/drivers/net/phy/phylink.c +++ b/drivers/net/phy/phylink.c @@ -761,8 +761,14 @@ static int phylink_bringup_phy(struct phylink *pl, struct phy_device *phy, config.interface = interface; ret = phylink_validate(pl, supported, &config); - if (ret) + if (ret) { + phylink_warn(pl, "validation of %s with support %*pb and advertisement %*pb failed: %d\n", + phy_modes(config.interface), + __ETHTOOL_LINK_MODE_MASK_NBITS, phy->supported, + __ETHTOOL_LINK_MODE_MASK_NBITS, config.advertising, + ret); return ret; + } phy->phylink = pl; phy->phy_link_change = phylink_phy_change; diff --git a/drivers/net/phy/sfp-bus.c b/drivers/net/phy/sfp-bus.c index d949ea7b4f8c..6900c68260e0 100644 --- a/drivers/net/phy/sfp-bus.c +++ b/drivers/net/phy/sfp-bus.c @@ -572,13 +572,15 @@ static void sfp_upstream_clear(struct sfp_bus *bus) * the sfp_bus structure, incrementing its reference count. This must * be put via sfp_bus_put() when done. * - * Returns: on success, a pointer to the sfp_bus structure, - * %NULL if no SFP is specified, - * on failure, an error pointer value: - * corresponding to the errors detailed for - * fwnode_property_get_reference_args(). - * %-ENOMEM if we failed to allocate the bus. - * an error from the upstream's connect_phy() method. + * Returns: + * - on success, a pointer to the sfp_bus structure, + * - %NULL if no SFP is specified, + * - on failure, an error pointer value: + * + * - corresponding to the errors detailed for + * fwnode_property_get_reference_args(). + * - %-ENOMEM if we failed to allocate the bus. + * - an error from the upstream's connect_phy() method. */ struct sfp_bus *sfp_bus_find_fwnode(struct fwnode_handle *fwnode) { @@ -612,13 +614,15 @@ EXPORT_SYMBOL_GPL(sfp_bus_find_fwnode); * the SFP bus using sfp_register_upstream(). This takes a reference on the * bus, so it is safe to put the bus after this call. * - * Returns: on success, a pointer to the sfp_bus structure, - * %NULL if no SFP is specified, - * on failure, an error pointer value: - * corresponding to the errors detailed for - * fwnode_property_get_reference_args(). - * %-ENOMEM if we failed to allocate the bus. - * an error from the upstream's connect_phy() method. + * Returns: + * - on success, a pointer to the sfp_bus structure, + * - %NULL if no SFP is specified, + * - on failure, an error pointer value: + * + * - corresponding to the errors detailed for + * fwnode_property_get_reference_args(). + * - %-ENOMEM if we failed to allocate the bus. + * - an error from the upstream's connect_phy() method. */ int sfp_bus_add_upstream(struct sfp_bus *bus, void *upstream, const struct sfp_upstream_ops *ops) diff --git a/drivers/net/slip/slhc.c b/drivers/net/slip/slhc.c index 58a69f830d29..f78ceba42e57 100644 --- a/drivers/net/slip/slhc.c +++ b/drivers/net/slip/slhc.c @@ -232,7 +232,7 @@ slhc_compress(struct slcompress *comp, unsigned char *icp, int isize, struct cstate *cs = lcs->next; unsigned long deltaS, deltaA; short changes = 0; - int hlen; + int nlen, hlen; unsigned char new_seq[16]; unsigned char *cp = new_seq; struct iphdr *ip; @@ -248,6 +248,8 @@ slhc_compress(struct slcompress *comp, unsigned char *icp, int isize, return isize; ip = (struct iphdr *) icp; + if (ip->version != 4 || ip->ihl < 5) + return isize; /* Bail if this packet isn't TCP, or is an IP fragment */ if (ip->protocol != IPPROTO_TCP || (ntohs(ip->frag_off) & 0x3fff)) { @@ -258,10 +260,14 @@ slhc_compress(struct slcompress *comp, unsigned char *icp, int isize, comp->sls_o_tcp++; return isize; } - /* Extract TCP header */ + nlen = ip->ihl * 4; + if (isize < nlen + sizeof(*th)) + return isize; - th = (struct tcphdr *)(((unsigned char *)ip) + ip->ihl*4); - hlen = ip->ihl*4 + th->doff*4; + th = (struct tcphdr *)(icp + nlen); + if (th->doff < sizeof(struct tcphdr) / 4) + return isize; + hlen = nlen + th->doff * 4; /* Bail if the TCP packet isn't `compressible' (i.e., ACK isn't set or * some other control bit is set). Also uncompressible if diff --git a/drivers/net/slip/slip.c b/drivers/net/slip/slip.c index 6f4d7ba8b109..babb01888b78 100644 --- a/drivers/net/slip/slip.c +++ b/drivers/net/slip/slip.c @@ -863,7 +863,10 @@ err_free_chan: tty->disc_data = NULL; clear_bit(SLF_INUSE, &sl->flags); sl_free_netdev(sl->dev); + /* do not call free_netdev before rtnl_unlock */ + rtnl_unlock(); free_netdev(sl->dev); + return err; err_exit: rtnl_unlock(); diff --git a/drivers/net/team/team.c b/drivers/net/team/team.c index ca70a1d840eb..4004f98e50d9 100644 --- a/drivers/net/team/team.c +++ b/drivers/net/team/team.c @@ -2240,6 +2240,8 @@ team_nl_option_policy[TEAM_ATTR_OPTION_MAX + 1] = { [TEAM_ATTR_OPTION_CHANGED] = { .type = NLA_FLAG }, [TEAM_ATTR_OPTION_TYPE] = { .type = NLA_U8 }, [TEAM_ATTR_OPTION_DATA] = { .type = NLA_BINARY }, + [TEAM_ATTR_OPTION_PORT_IFINDEX] = { .type = NLA_U32 }, + [TEAM_ATTR_OPTION_ARRAY_INDEX] = { .type = NLA_U32 }, }; static int team_nl_cmd_noop(struct sk_buff *skb, struct genl_info *info) diff --git a/drivers/net/usb/qmi_wwan.c b/drivers/net/usb/qmi_wwan.c index 3b7a3b8a5e06..6c738a271257 100644 --- a/drivers/net/usb/qmi_wwan.c +++ b/drivers/net/usb/qmi_wwan.c @@ -337,6 +337,9 @@ static void qmi_wwan_netdev_setup(struct net_device *net) netdev_dbg(net, "mode: raw IP\n"); } else if (!net->header_ops) { /* don't bother if already set */ ether_setup(net); + /* Restoring min/max mtu values set originally by usbnet */ + net->min_mtu = 0; + net->max_mtu = ETH_MAX_MTU; clear_bit(EVENT_NO_IP_ALIGN, &dev->flags); netdev_dbg(net, "mode: Ethernet\n"); } @@ -1207,6 +1210,7 @@ static const struct usb_device_id products[] = { {QMI_FIXED_INTF(0x1435, 0xd182, 5)}, /* Wistron NeWeb D18 */ {QMI_FIXED_INTF(0x1435, 0xd191, 4)}, /* Wistron NeWeb D19Q1 */ {QMI_QUIRK_SET_DTR(0x1508, 0x1001, 4)}, /* Fibocom NL668 series */ + {QMI_FIXED_INTF(0x1690, 0x7588, 4)}, /* ASKEY WWHC050 */ {QMI_FIXED_INTF(0x16d8, 0x6003, 0)}, /* CMOTech 6003 */ {QMI_FIXED_INTF(0x16d8, 0x6007, 0)}, /* CMOTech CHE-628S */ {QMI_FIXED_INTF(0x16d8, 0x6008, 0)}, /* CMOTech CMU-301 */ diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c index 78ddbaf6401b..95b19ce96513 100644 --- a/drivers/net/usb/r8152.c +++ b/drivers/net/usb/r8152.c @@ -3221,6 +3221,8 @@ static u16 r8153_phy_status(struct r8152 *tp, u16 desired) } msleep(20); + if (test_bit(RTL8152_UNPLUG, &tp->flags)) + break; } return data; @@ -5402,7 +5404,10 @@ static void r8153_init(struct r8152 *tp) if (ocp_read_word(tp, MCU_TYPE_PLA, PLA_BOOT_CTRL) & AUTOLOAD_DONE) break; + msleep(20); + if (test_bit(RTL8152_UNPLUG, &tp->flags)) + break; } data = r8153_phy_status(tp, 0); @@ -5539,7 +5544,10 @@ static void r8153b_init(struct r8152 *tp) if (ocp_read_word(tp, MCU_TYPE_PLA, PLA_BOOT_CTRL) & AUTOLOAD_DONE) break; + msleep(20); + if (test_bit(RTL8152_UNPLUG, &tp->flags)) + break; } data = r8153_phy_status(tp, 0); diff --git a/drivers/net/veth.c b/drivers/net/veth.c index 8cdc4415fa70..d4cbb9e8c63f 100644 --- a/drivers/net/veth.c +++ b/drivers/net/veth.c @@ -328,7 +328,7 @@ static void veth_get_stats64(struct net_device *dev, rcu_read_lock(); peer = rcu_dereference(priv->peer); if (peer) { - tot->rx_dropped += veth_stats_tx(peer, &packets, &bytes); + veth_stats_tx(peer, &packets, &bytes); tot->rx_bytes += bytes; tot->rx_packets += packets; diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c index d3b08b76b1ec..45308b3350cf 100644 --- a/drivers/net/vxlan.c +++ b/drivers/net/vxlan.c @@ -2779,10 +2779,19 @@ static void vxlan_vs_add_dev(struct vxlan_sock *vs, struct vxlan_dev *vxlan, /* Setup stats when device is created */ static int vxlan_init(struct net_device *dev) { + struct vxlan_dev *vxlan = netdev_priv(dev); + int err; + dev->tstats = netdev_alloc_pcpu_stats(struct pcpu_sw_netstats); if (!dev->tstats) return -ENOMEM; + err = gro_cells_init(&vxlan->gro_cells, dev); + if (err) { + free_percpu(dev->tstats); + return err; + } + return 0; } @@ -3043,8 +3052,6 @@ static void vxlan_setup(struct net_device *dev) vxlan->dev = dev; - gro_cells_init(&vxlan->gro_cells, dev); - for (h = 0; h < FDB_HASH_SIZE; ++h) { spin_lock_init(&vxlan->hash_lock[h]); INIT_HLIST_HEAD(&vxlan->fdb_head[h]); diff --git a/drivers/net/wireguard/device.c b/drivers/net/wireguard/device.c index 43db442b1373..3ac3f8570ca1 100644 --- a/drivers/net/wireguard/device.c +++ b/drivers/net/wireguard/device.c @@ -122,7 +122,7 @@ static netdev_tx_t wg_xmit(struct sk_buff *skb, struct net_device *dev) u32 mtu; int ret; - if (unlikely(wg_skb_examine_untrusted_ip_hdr(skb) != skb->protocol)) { + if (unlikely(!wg_check_packet_protocol(skb))) { ret = -EPROTONOSUPPORT; net_dbg_ratelimited("%s: Invalid IP packet\n", dev->name); goto err; @@ -258,6 +258,8 @@ static void wg_setup(struct net_device *dev) enum { WG_NETDEV_FEATURES = NETIF_F_HW_CSUM | NETIF_F_RXCSUM | NETIF_F_SG | NETIF_F_GSO | NETIF_F_GSO_SOFTWARE | NETIF_F_HIGHDMA }; + const int overhead = MESSAGE_MINIMUM_LENGTH + sizeof(struct udphdr) + + max(sizeof(struct ipv6hdr), sizeof(struct iphdr)); dev->netdev_ops = &netdev_ops; dev->hard_header_len = 0; @@ -271,9 +273,8 @@ static void wg_setup(struct net_device *dev) dev->features |= WG_NETDEV_FEATURES; dev->hw_features |= WG_NETDEV_FEATURES; dev->hw_enc_features |= WG_NETDEV_FEATURES; - dev->mtu = ETH_DATA_LEN - MESSAGE_MINIMUM_LENGTH - - sizeof(struct udphdr) - - max(sizeof(struct ipv6hdr), sizeof(struct iphdr)); + dev->mtu = ETH_DATA_LEN - overhead; + dev->max_mtu = round_down(INT_MAX, MESSAGE_PADDING_MULTIPLE) - overhead; SET_NETDEV_DEVTYPE(dev, &device_type); diff --git a/drivers/net/wireguard/netlink.c b/drivers/net/wireguard/netlink.c index bda26405497c..802099c8828a 100644 --- a/drivers/net/wireguard/netlink.c +++ b/drivers/net/wireguard/netlink.c @@ -411,11 +411,7 @@ static int set_peer(struct wg_device *wg, struct nlattr **attrs) peer = wg_peer_create(wg, public_key, preshared_key); if (IS_ERR(peer)) { - /* Similar to the above, if the key is invalid, we skip - * it without fanfare, so that services don't need to - * worry about doing key validation themselves. - */ - ret = PTR_ERR(peer) == -EKEYREJECTED ? 0 : PTR_ERR(peer); + ret = PTR_ERR(peer); peer = NULL; goto out; } @@ -569,7 +565,7 @@ static int wg_set_device(struct sk_buff *skb, struct genl_info *info) private_key); list_for_each_entry_safe(peer, temp, &wg->peer_list, peer_list) { - BUG_ON(!wg_noise_precompute_static_static(peer)); + wg_noise_precompute_static_static(peer); wg_noise_expire_current_peer_keypairs(peer); } wg_cookie_checker_precompute_device_keys(&wg->cookie_checker); diff --git a/drivers/net/wireguard/noise.c b/drivers/net/wireguard/noise.c index 919d9d866446..708dc61c974f 100644 --- a/drivers/net/wireguard/noise.c +++ b/drivers/net/wireguard/noise.c @@ -44,32 +44,23 @@ void __init wg_noise_init(void) } /* Must hold peer->handshake.static_identity->lock */ -bool wg_noise_precompute_static_static(struct wg_peer *peer) +void wg_noise_precompute_static_static(struct wg_peer *peer) { - bool ret; - down_write(&peer->handshake.lock); - if (peer->handshake.static_identity->has_identity) { - ret = curve25519( - peer->handshake.precomputed_static_static, + if (!peer->handshake.static_identity->has_identity || + !curve25519(peer->handshake.precomputed_static_static, peer->handshake.static_identity->static_private, - peer->handshake.remote_static); - } else { - u8 empty[NOISE_PUBLIC_KEY_LEN] = { 0 }; - - ret = curve25519(empty, empty, peer->handshake.remote_static); + peer->handshake.remote_static)) memset(peer->handshake.precomputed_static_static, 0, NOISE_PUBLIC_KEY_LEN); - } up_write(&peer->handshake.lock); - return ret; } -bool wg_noise_handshake_init(struct noise_handshake *handshake, - struct noise_static_identity *static_identity, - const u8 peer_public_key[NOISE_PUBLIC_KEY_LEN], - const u8 peer_preshared_key[NOISE_SYMMETRIC_KEY_LEN], - struct wg_peer *peer) +void wg_noise_handshake_init(struct noise_handshake *handshake, + struct noise_static_identity *static_identity, + const u8 peer_public_key[NOISE_PUBLIC_KEY_LEN], + const u8 peer_preshared_key[NOISE_SYMMETRIC_KEY_LEN], + struct wg_peer *peer) { memset(handshake, 0, sizeof(*handshake)); init_rwsem(&handshake->lock); @@ -81,7 +72,7 @@ bool wg_noise_handshake_init(struct noise_handshake *handshake, NOISE_SYMMETRIC_KEY_LEN); handshake->static_identity = static_identity; handshake->state = HANDSHAKE_ZEROED; - return wg_noise_precompute_static_static(peer); + wg_noise_precompute_static_static(peer); } static void handshake_zero(struct noise_handshake *handshake) @@ -403,6 +394,19 @@ static bool __must_check mix_dh(u8 chaining_key[NOISE_HASH_LEN], return true; } +static bool __must_check mix_precomputed_dh(u8 chaining_key[NOISE_HASH_LEN], + u8 key[NOISE_SYMMETRIC_KEY_LEN], + const u8 precomputed[NOISE_PUBLIC_KEY_LEN]) +{ + static u8 zero_point[NOISE_PUBLIC_KEY_LEN]; + if (unlikely(!crypto_memneq(precomputed, zero_point, NOISE_PUBLIC_KEY_LEN))) + return false; + kdf(chaining_key, key, NULL, precomputed, NOISE_HASH_LEN, + NOISE_SYMMETRIC_KEY_LEN, 0, NOISE_PUBLIC_KEY_LEN, + chaining_key); + return true; +} + static void mix_hash(u8 hash[NOISE_HASH_LEN], const u8 *src, size_t src_len) { struct blake2s_state blake; @@ -531,10 +535,9 @@ wg_noise_handshake_create_initiation(struct message_handshake_initiation *dst, NOISE_PUBLIC_KEY_LEN, key, handshake->hash); /* ss */ - kdf(handshake->chaining_key, key, NULL, - handshake->precomputed_static_static, NOISE_HASH_LEN, - NOISE_SYMMETRIC_KEY_LEN, 0, NOISE_PUBLIC_KEY_LEN, - handshake->chaining_key); + if (!mix_precomputed_dh(handshake->chaining_key, key, + handshake->precomputed_static_static)) + goto out; /* {t} */ tai64n_now(timestamp); @@ -595,9 +598,9 @@ wg_noise_handshake_consume_initiation(struct message_handshake_initiation *src, handshake = &peer->handshake; /* ss */ - kdf(chaining_key, key, NULL, handshake->precomputed_static_static, - NOISE_HASH_LEN, NOISE_SYMMETRIC_KEY_LEN, 0, NOISE_PUBLIC_KEY_LEN, - chaining_key); + if (!mix_precomputed_dh(chaining_key, key, + handshake->precomputed_static_static)) + goto out; /* {t} */ if (!message_decrypt(t, src->encrypted_timestamp, diff --git a/drivers/net/wireguard/noise.h b/drivers/net/wireguard/noise.h index 138a07bb817c..f532d59d3f19 100644 --- a/drivers/net/wireguard/noise.h +++ b/drivers/net/wireguard/noise.h @@ -94,11 +94,11 @@ struct noise_handshake { struct wg_device; void wg_noise_init(void); -bool wg_noise_handshake_init(struct noise_handshake *handshake, - struct noise_static_identity *static_identity, - const u8 peer_public_key[NOISE_PUBLIC_KEY_LEN], - const u8 peer_preshared_key[NOISE_SYMMETRIC_KEY_LEN], - struct wg_peer *peer); +void wg_noise_handshake_init(struct noise_handshake *handshake, + struct noise_static_identity *static_identity, + const u8 peer_public_key[NOISE_PUBLIC_KEY_LEN], + const u8 peer_preshared_key[NOISE_SYMMETRIC_KEY_LEN], + struct wg_peer *peer); void wg_noise_handshake_clear(struct noise_handshake *handshake); static inline void wg_noise_reset_last_sent_handshake(atomic64_t *handshake_ns) { @@ -116,7 +116,7 @@ void wg_noise_expire_current_peer_keypairs(struct wg_peer *peer); void wg_noise_set_static_identity_private_key( struct noise_static_identity *static_identity, const u8 private_key[NOISE_PUBLIC_KEY_LEN]); -bool wg_noise_precompute_static_static(struct wg_peer *peer); +void wg_noise_precompute_static_static(struct wg_peer *peer); bool wg_noise_handshake_create_initiation(struct message_handshake_initiation *dst, diff --git a/drivers/net/wireguard/peer.c b/drivers/net/wireguard/peer.c index 071eedf33f5a..1d634bd3038f 100644 --- a/drivers/net/wireguard/peer.c +++ b/drivers/net/wireguard/peer.c @@ -34,11 +34,8 @@ struct wg_peer *wg_peer_create(struct wg_device *wg, return ERR_PTR(ret); peer->device = wg; - if (!wg_noise_handshake_init(&peer->handshake, &wg->static_identity, - public_key, preshared_key, peer)) { - ret = -EKEYREJECTED; - goto err_1; - } + wg_noise_handshake_init(&peer->handshake, &wg->static_identity, + public_key, preshared_key, peer); if (dst_cache_init(&peer->endpoint_cache, GFP_KERNEL)) goto err_1; if (wg_packet_queue_init(&peer->tx_queue, wg_packet_tx_worker, false, diff --git a/drivers/net/wireguard/queueing.h b/drivers/net/wireguard/queueing.h index fecb559cbdb6..3432232afe06 100644 --- a/drivers/net/wireguard/queueing.h +++ b/drivers/net/wireguard/queueing.h @@ -66,7 +66,7 @@ struct packet_cb { #define PACKET_PEER(skb) (PACKET_CB(skb)->keypair->entry.peer) /* Returns either the correct skb->protocol value, or 0 if invalid. */ -static inline __be16 wg_skb_examine_untrusted_ip_hdr(struct sk_buff *skb) +static inline __be16 wg_examine_packet_protocol(struct sk_buff *skb) { if (skb_network_header(skb) >= skb->head && (skb_network_header(skb) + sizeof(struct iphdr)) <= @@ -81,6 +81,12 @@ static inline __be16 wg_skb_examine_untrusted_ip_hdr(struct sk_buff *skb) return 0; } +static inline bool wg_check_packet_protocol(struct sk_buff *skb) +{ + __be16 real_protocol = wg_examine_packet_protocol(skb); + return real_protocol && skb->protocol == real_protocol; +} + static inline void wg_reset_packet(struct sk_buff *skb) { skb_scrub_packet(skb, true); @@ -94,8 +100,8 @@ static inline void wg_reset_packet(struct sk_buff *skb) skb->dev = NULL; #ifdef CONFIG_NET_SCHED skb->tc_index = 0; - skb_reset_tc(skb); #endif + skb_reset_redirect(skb); skb->hdr_len = skb_headroom(skb); skb_reset_mac_header(skb); skb_reset_network_header(skb); diff --git a/drivers/net/wireguard/receive.c b/drivers/net/wireguard/receive.c index 9c6bab9c981f..da3b782ab7d3 100644 --- a/drivers/net/wireguard/receive.c +++ b/drivers/net/wireguard/receive.c @@ -56,7 +56,7 @@ static int prepare_skb_header(struct sk_buff *skb, struct wg_device *wg) size_t data_offset, data_len, header_len; struct udphdr *udp; - if (unlikely(wg_skb_examine_untrusted_ip_hdr(skb) != skb->protocol || + if (unlikely(!wg_check_packet_protocol(skb) || skb_transport_header(skb) < skb->head || (skb_transport_header(skb) + sizeof(struct udphdr)) > skb_tail_pointer(skb))) @@ -118,10 +118,13 @@ static void wg_receive_handshake_packet(struct wg_device *wg, under_load = skb_queue_len(&wg->incoming_handshakes) >= MAX_QUEUED_INCOMING_HANDSHAKES / 8; - if (under_load) + if (under_load) { last_under_load = ktime_get_coarse_boottime_ns(); - else if (last_under_load) + } else if (last_under_load) { under_load = !wg_birthdate_has_expired(last_under_load, 1); + if (!under_load) + last_under_load = 0; + } mac_state = wg_cookie_validate_packet(&wg->cookie_checker, skb, under_load); if ((under_load && mac_state == VALID_MAC_WITH_COOKIE) || @@ -385,7 +388,7 @@ static void wg_packet_consume_data_done(struct wg_peer *peer, */ skb->ip_summed = CHECKSUM_UNNECESSARY; skb->csum_level = ~0; /* All levels */ - skb->protocol = wg_skb_examine_untrusted_ip_hdr(skb); + skb->protocol = wg_examine_packet_protocol(skb); if (skb->protocol == htons(ETH_P_IP)) { len = ntohs(ip_hdr(skb)->tot_len); if (unlikely(len < sizeof(struct iphdr))) @@ -584,8 +587,7 @@ void wg_packet_receive(struct wg_device *wg, struct sk_buff *skb) wg_packet_consume_data(wg, skb); break; default: - net_dbg_skb_ratelimited("%s: Invalid packet from %pISpfsc\n", - wg->dev->name, skb); + WARN(1, "Non-exhaustive parsing of packet header lead to unknown packet type!\n"); goto err; } return; diff --git a/drivers/net/wireguard/send.c b/drivers/net/wireguard/send.c index c13260563446..7348c10cbae3 100644 --- a/drivers/net/wireguard/send.c +++ b/drivers/net/wireguard/send.c @@ -143,16 +143,22 @@ static void keep_key_fresh(struct wg_peer *peer) static unsigned int calculate_skb_padding(struct sk_buff *skb) { + unsigned int padded_size, last_unit = skb->len; + + if (unlikely(!PACKET_CB(skb)->mtu)) + return ALIGN(last_unit, MESSAGE_PADDING_MULTIPLE) - last_unit; + /* We do this modulo business with the MTU, just in case the networking * layer gives us a packet that's bigger than the MTU. In that case, we * wouldn't want the final subtraction to overflow in the case of the - * padded_size being clamped. + * padded_size being clamped. Fortunately, that's very rarely the case, + * so we optimize for that not happening. */ - unsigned int last_unit = skb->len % PACKET_CB(skb)->mtu; - unsigned int padded_size = ALIGN(last_unit, MESSAGE_PADDING_MULTIPLE); + if (unlikely(last_unit > PACKET_CB(skb)->mtu)) + last_unit %= PACKET_CB(skb)->mtu; - if (padded_size > PACKET_CB(skb)->mtu) - padded_size = PACKET_CB(skb)->mtu; + padded_size = min(PACKET_CB(skb)->mtu, + ALIGN(last_unit, MESSAGE_PADDING_MULTIPLE)); return padded_size - last_unit; } diff --git a/drivers/net/wireguard/socket.c b/drivers/net/wireguard/socket.c index 262f3b5c819d..b0d6541582d3 100644 --- a/drivers/net/wireguard/socket.c +++ b/drivers/net/wireguard/socket.c @@ -432,7 +432,6 @@ void wg_socket_reinit(struct wg_device *wg, struct sock *new4, wg->incoming_port = ntohs(inet_sk(new4)->inet_sport); mutex_unlock(&wg->socket_update_lock); synchronize_rcu(); - synchronize_net(); sock_free(old4); sock_free(old6); } diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/22000.c b/drivers/net/wireless/intel/iwlwifi/cfg/22000.c index a22a830019c0..355af47c5f73 100644 --- a/drivers/net/wireless/intel/iwlwifi/cfg/22000.c +++ b/drivers/net/wireless/intel/iwlwifi/cfg/22000.c @@ -283,6 +283,7 @@ const struct iwl_cfg iwl_ax101_cfg_qu_c0_hr_b0 = { * HT size; mac80211 would otherwise pick the HE max (256) by default. */ .max_tx_agg_size = IEEE80211_MAX_AMPDU_BUF_HT, + .tx_with_siso_diversity = true, .num_rbds = IWL_NUM_RBDS_22000_HE, }; @@ -309,6 +310,7 @@ const struct iwl_cfg iwl_ax101_cfg_quz_hr = { * HT size; mac80211 would otherwise pick the HE max (256) by default. */ .max_tx_agg_size = IEEE80211_MAX_AMPDU_BUF_HT, + .tx_with_siso_diversity = true, .num_rbds = IWL_NUM_RBDS_22000_HE, }; diff --git a/drivers/net/wireless/intel/iwlwifi/fw/acpi.c b/drivers/net/wireless/intel/iwlwifi/fw/acpi.c index 48d375a86d86..ba2aff3af0fe 100644 --- a/drivers/net/wireless/intel/iwlwifi/fw/acpi.c +++ b/drivers/net/wireless/intel/iwlwifi/fw/acpi.c @@ -6,7 +6,7 @@ * GPL LICENSE SUMMARY * * Copyright(c) 2017 Intel Deutschland GmbH - * Copyright (C) 2019 Intel Corporation + * Copyright (C) 2019 - 2020 Intel Corporation * * This program is free software; you can redistribute it and/or modify * it under the terms of version 2 of the GNU General Public License as @@ -27,7 +27,7 @@ * BSD LICENSE * * Copyright(c) 2017 Intel Deutschland GmbH - * Copyright (C) 2019 Intel Corporation + * Copyright (C) 2019 - 2020 Intel Corporation * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -491,13 +491,13 @@ int iwl_validate_sar_geo_profile(struct iwl_fw_runtime *fwrt, } IWL_EXPORT_SYMBOL(iwl_validate_sar_geo_profile); -void iwl_sar_geo_init(struct iwl_fw_runtime *fwrt, - struct iwl_per_chain_offset_group *table) +int iwl_sar_geo_init(struct iwl_fw_runtime *fwrt, + struct iwl_per_chain_offset_group *table) { int ret, i, j; if (!iwl_sar_geo_support(fwrt)) - return; + return -EOPNOTSUPP; ret = iwl_sar_get_wgds_table(fwrt); if (ret < 0) { @@ -505,7 +505,7 @@ void iwl_sar_geo_init(struct iwl_fw_runtime *fwrt, "Geo SAR BIOS table invalid or unavailable. (%d)\n", ret); /* we don't fail if the table is not available */ - return; + return -ENOENT; } BUILD_BUG_ON(ACPI_NUM_GEO_PROFILES * ACPI_WGDS_NUM_BANDS * @@ -530,5 +530,7 @@ void iwl_sar_geo_init(struct iwl_fw_runtime *fwrt, i, j, value[1], value[2], value[0]); } } + + return 0; } IWL_EXPORT_SYMBOL(iwl_sar_geo_init); diff --git a/drivers/net/wireless/intel/iwlwifi/fw/acpi.h b/drivers/net/wireless/intel/iwlwifi/fw/acpi.h index 4a6e8262974b..5590e5cc8fbb 100644 --- a/drivers/net/wireless/intel/iwlwifi/fw/acpi.h +++ b/drivers/net/wireless/intel/iwlwifi/fw/acpi.h @@ -6,7 +6,7 @@ * GPL LICENSE SUMMARY * * Copyright(c) 2017 Intel Deutschland GmbH - * Copyright(c) 2018 - 2019 Intel Corporation + * Copyright(c) 2018 - 2020 Intel Corporation * * This program is free software; you can redistribute it and/or modify * it under the terms of version 2 of the GNU General Public License as @@ -27,7 +27,7 @@ * BSD LICENSE * * Copyright(c) 2017 Intel Deutschland GmbH - * Copyright(c) 2018 - 2019 Intel Corporation + * Copyright(c) 2018 - 2020 Intel Corporation * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -171,8 +171,9 @@ bool iwl_sar_geo_support(struct iwl_fw_runtime *fwrt); int iwl_validate_sar_geo_profile(struct iwl_fw_runtime *fwrt, struct iwl_host_cmd *cmd); -void iwl_sar_geo_init(struct iwl_fw_runtime *fwrt, - struct iwl_per_chain_offset_group *table); +int iwl_sar_geo_init(struct iwl_fw_runtime *fwrt, + struct iwl_per_chain_offset_group *table); + #else /* CONFIG_ACPI */ static inline void *iwl_acpi_get_object(struct device *dev, acpi_string method) @@ -243,9 +244,10 @@ static inline int iwl_validate_sar_geo_profile(struct iwl_fw_runtime *fwrt, return -ENOENT; } -static inline void iwl_sar_geo_init(struct iwl_fw_runtime *fwrt, - struct iwl_per_chain_offset_group *table) +static inline int iwl_sar_geo_init(struct iwl_fw_runtime *fwrt, + struct iwl_per_chain_offset_group *table) { + return -ENOENT; } #endif /* CONFIG_ACPI */ diff --git a/drivers/net/wireless/intel/iwlwifi/fw/dbg.c b/drivers/net/wireless/intel/iwlwifi/fw/dbg.c index 91df1ee25dd0..8796ab8f2a5f 100644 --- a/drivers/net/wireless/intel/iwlwifi/fw/dbg.c +++ b/drivers/net/wireless/intel/iwlwifi/fw/dbg.c @@ -8,7 +8,7 @@ * Copyright(c) 2008 - 2014 Intel Corporation. All rights reserved. * Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH * Copyright(c) 2015 - 2017 Intel Deutschland GmbH - * Copyright(c) 2018 - 2019 Intel Corporation + * Copyright(c) 2018 - 2020 Intel Corporation * * This program is free software; you can redistribute it and/or modify * it under the terms of version 2 of the GNU General Public License as @@ -31,7 +31,7 @@ * Copyright(c) 2005 - 2014 Intel Corporation. All rights reserved. * Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH * Copyright(c) 2015 - 2017 Intel Deutschland GmbH - * Copyright(c) 2018 - 2019 Intel Corporation + * Copyright(c) 2018 - 2020 Intel Corporation * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -1409,11 +1409,7 @@ static int iwl_dump_ini_rxf_iter(struct iwl_fw_runtime *fwrt, goto out; } - /* - * region register have absolute value so apply rxf offset after - * reading the registers - */ - offs += rxf_data.offset; + offs = rxf_data.offset; /* Lock fence */ iwl_write_prph_no_grab(fwrt->trans, RXF_SET_FENCE_MODE + offs, 0x1); @@ -2494,10 +2490,7 @@ static void iwl_fw_dbg_collect_sync(struct iwl_fw_runtime *fwrt, u8 wk_idx) goto out; } - if (iwl_fw_dbg_stop_restart_recording(fwrt, ¶ms, true)) { - IWL_ERR(fwrt, "Failed to stop DBGC recording, aborting dump\n"); - goto out; - } + iwl_fw_dbg_stop_restart_recording(fwrt, ¶ms, true); IWL_DEBUG_FW_INFO(fwrt, "WRT: Data collection start\n"); if (iwl_trans_dbg_ini_valid(fwrt->trans)) @@ -2662,14 +2655,14 @@ static int iwl_fw_dbg_restart_recording(struct iwl_trans *trans, return 0; } -int iwl_fw_dbg_stop_restart_recording(struct iwl_fw_runtime *fwrt, - struct iwl_fw_dbg_params *params, - bool stop) +void iwl_fw_dbg_stop_restart_recording(struct iwl_fw_runtime *fwrt, + struct iwl_fw_dbg_params *params, + bool stop) { int ret = 0; if (test_bit(STATUS_FW_ERROR, &fwrt->trans->status)) - return 0; + return; if (fw_has_capa(&fwrt->fw->ucode_capa, IWL_UCODE_TLV_CAPA_DBG_SUSPEND_RESUME_CMD_SUPP)) @@ -2686,7 +2679,5 @@ int iwl_fw_dbg_stop_restart_recording(struct iwl_fw_runtime *fwrt, iwl_fw_set_dbg_rec_on(fwrt); } #endif - - return ret; } IWL_EXPORT_SYMBOL(iwl_fw_dbg_stop_restart_recording); diff --git a/drivers/net/wireless/intel/iwlwifi/fw/dbg.h b/drivers/net/wireless/intel/iwlwifi/fw/dbg.h index 179f2905d56b..9d3513213f5f 100644 --- a/drivers/net/wireless/intel/iwlwifi/fw/dbg.h +++ b/drivers/net/wireless/intel/iwlwifi/fw/dbg.h @@ -239,9 +239,9 @@ _iwl_fw_dbg_trigger_simple_stop(struct iwl_fw_runtime *fwrt, _iwl_fw_dbg_trigger_simple_stop((fwrt), (wdev), \ iwl_fw_dbg_get_trigger((fwrt)->fw,\ (trig))) -int iwl_fw_dbg_stop_restart_recording(struct iwl_fw_runtime *fwrt, - struct iwl_fw_dbg_params *params, - bool stop); +void iwl_fw_dbg_stop_restart_recording(struct iwl_fw_runtime *fwrt, + struct iwl_fw_dbg_params *params, + bool stop); #ifdef CONFIG_IWLWIFI_DEBUGFS static inline void iwl_fw_set_dbg_rec_on(struct iwl_fw_runtime *fwrt) diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c index 2d1cb4647c3b..0481796f75bc 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c +++ b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c @@ -1467,7 +1467,7 @@ static void iwl_req_fw_callback(const struct firmware *ucode_raw, void *context) kmemdup(pieces->dbg_conf_tlv[i], pieces->dbg_conf_tlv_len[i], GFP_KERNEL); - if (!pieces->dbg_conf_tlv_len[i]) + if (!pieces->dbg_conf_tlv[i]) goto out_free_fw; } } diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c index 54c094e88474..98263cd37944 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c @@ -762,10 +762,17 @@ static int iwl_mvm_sar_geo_init(struct iwl_mvm *mvm) u16 cmd_wide_id = WIDE_ID(PHY_OPS_GROUP, GEO_TX_POWER_LIMIT); union geo_tx_power_profiles_cmd cmd; u16 len; + int ret; cmd.geo_cmd.ops = cpu_to_le32(IWL_PER_CHAIN_OFFSET_SET_TABLES); - iwl_sar_geo_init(&mvm->fwrt, cmd.geo_cmd.table); + ret = iwl_sar_geo_init(&mvm->fwrt, cmd.geo_cmd.table); + /* + * It is a valid scenario to not support SAR, or miss wgds table, + * but in that case there is no need to send the command. + */ + if (ret) + return 0; cmd.geo_cmd.table_revision = cpu_to_le32(mvm->fwrt.geo_rev); diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c b/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c index 70b29bf16bb9..60296a754af2 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c @@ -308,7 +308,8 @@ iwl_parse_nvm_sections(struct iwl_mvm *mvm) } /* PHY_SKU section is mandatory in B0 */ - if (!mvm->nvm_sections[NVM_SECTION_TYPE_PHY_SKU].data) { + if (mvm->trans->cfg->nvm_type == IWL_NVM_EXT && + !mvm->nvm_sections[NVM_SECTION_TYPE_PHY_SKU].data) { IWL_ERR(mvm, "Can't parse phy_sku in B0, empty sections\n"); return NULL; diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c index e2cf9e015ef8..ca99a9c4f70e 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c @@ -6,7 +6,7 @@ * GPL LICENSE SUMMARY * * Copyright(c) 2017 Intel Deutschland GmbH - * Copyright(c) 2018 - 2019 Intel Corporation + * Copyright(c) 2018 - 2020 Intel Corporation * * This program is free software; you can redistribute it and/or modify * it under the terms of version 2 of the GNU General Public License as @@ -27,7 +27,7 @@ * BSD LICENSE * * Copyright(c) 2017 Intel Deutschland GmbH - * Copyright(c) 2018 - 2019 Intel Corporation + * Copyright(c) 2018 - 2020 Intel Corporation * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -147,7 +147,11 @@ static u16 rs_fw_get_config_flags(struct iwl_mvm *mvm, (vht_ena && (vht_cap->cap & IEEE80211_VHT_CAP_RXLDPC)))) flags |= IWL_TLC_MNG_CFG_FLAGS_LDPC_MSK; - /* consider our LDPC support in case of HE */ + /* consider LDPC support in case of HE */ + if (he_cap->has_he && (he_cap->he_cap_elem.phy_cap_info[1] & + IEEE80211_HE_PHY_CAP1_LDPC_CODING_IN_PAYLOAD)) + flags |= IWL_TLC_MNG_CFG_FLAGS_LDPC_MSK; + if (sband->iftype_data && sband->iftype_data->he_cap.has_he && !(sband->iftype_data->he_cap.he_cap_elem.phy_cap_info[1] & IEEE80211_HE_PHY_CAP1_LDPC_CODING_IN_PAYLOAD)) @@ -191,11 +195,13 @@ rs_fw_vht_set_enabled_rates(const struct ieee80211_sta *sta, { u16 supp; int i, highest_mcs; + u8 nss = sta->rx_nss; - for (i = 0; i < sta->rx_nss; i++) { - if (i == IWL_TLC_NSS_MAX) - break; + /* the station support only a single receive chain */ + if (sta->smps_mode == IEEE80211_SMPS_STATIC) + nss = 1; + for (i = 0; i < nss && i < IWL_TLC_NSS_MAX; i++) { highest_mcs = rs_fw_vht_highest_rx_mcs_index(vht_cap, i + 1); if (!highest_mcs) continue; @@ -241,8 +247,13 @@ rs_fw_he_set_enabled_rates(const struct ieee80211_sta *sta, u16 tx_mcs_160 = le16_to_cpu(sband->iftype_data->he_cap.he_mcs_nss_supp.tx_mcs_160); int i; + u8 nss = sta->rx_nss; + + /* the station support only a single receive chain */ + if (sta->smps_mode == IEEE80211_SMPS_STATIC) + nss = 1; - for (i = 0; i < sta->rx_nss && i < IWL_TLC_NSS_MAX; i++) { + for (i = 0; i < nss && i < IWL_TLC_NSS_MAX; i++) { u16 _mcs_160 = (mcs_160 >> (2 * i)) & 0x3; u16 _mcs_80 = (mcs_80 >> (2 * i)) & 0x3; u16 _tx_mcs_160 = (tx_mcs_160 >> (2 * i)) & 0x3; @@ -303,8 +314,14 @@ static void rs_fw_set_supp_rates(struct ieee80211_sta *sta, cmd->mode = IWL_TLC_MNG_MODE_HT; cmd->ht_rates[IWL_TLC_NSS_1][IWL_TLC_HT_BW_NONE_160] = cpu_to_le16(ht_cap->mcs.rx_mask[0]); - cmd->ht_rates[IWL_TLC_NSS_2][IWL_TLC_HT_BW_NONE_160] = - cpu_to_le16(ht_cap->mcs.rx_mask[1]); + + /* the station support only a single receive chain */ + if (sta->smps_mode == IEEE80211_SMPS_STATIC) + cmd->ht_rates[IWL_TLC_NSS_2][IWL_TLC_HT_BW_NONE_160] = + 0; + else + cmd->ht_rates[IWL_TLC_NSS_2][IWL_TLC_HT_BW_NONE_160] = + cpu_to_le16(ht_cap->mcs.rx_mask[1]); } } diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/time-event.c b/drivers/net/wireless/intel/iwlwifi/mvm/time-event.c index c0b420fe5e48..1babc4bb5194 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/time-event.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/time-event.c @@ -785,7 +785,9 @@ void iwl_mvm_rx_session_protect_notif(struct iwl_mvm *mvm, if (!le32_to_cpu(notif->status)) { iwl_mvm_te_check_disconnect(mvm, vif, "Session protection failure"); + spin_lock_bh(&mvm->time_event_lock); iwl_mvm_te_clear_data(mvm, te_data); + spin_unlock_bh(&mvm->time_event_lock); } if (le32_to_cpu(notif->start)) { @@ -801,7 +803,9 @@ void iwl_mvm_rx_session_protect_notif(struct iwl_mvm *mvm, */ iwl_mvm_te_check_disconnect(mvm, vif, "No beacon heard and the session protection is over already..."); + spin_lock_bh(&mvm->time_event_lock); iwl_mvm_te_clear_data(mvm, te_data); + spin_unlock_bh(&mvm->time_event_lock); } goto out_unlock; diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c index 97f227f3cbc3..f441b20e1642 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c @@ -981,6 +981,9 @@ static const struct iwl_dev_info iwl_dev_info_table[] = { IWL_DEV_INFO(0x2526, 0x0014, iwl9260_2ac_160_cfg, iwl9260_160_name), IWL_DEV_INFO(0x2526, 0x0018, iwl9260_2ac_160_cfg, iwl9260_160_name), IWL_DEV_INFO(0x2526, 0x001C, iwl9260_2ac_160_cfg, iwl9260_160_name), + IWL_DEV_INFO(0x2526, 0x4010, iwl9260_2ac_160_cfg, iwl9260_160_name), + IWL_DEV_INFO(0x2526, 0x4018, iwl9260_2ac_160_cfg, iwl9260_160_name), + IWL_DEV_INFO(0x2526, 0x401C, iwl9260_2ac_160_cfg, iwl9260_160_name), IWL_DEV_INFO(0x2526, 0x6010, iwl9260_2ac_160_cfg, iwl9260_160_name), IWL_DEV_INFO(0x2526, 0x6014, iwl9260_2ac_160_cfg, iwl9260_160_name), IWL_DEV_INFO(0x2526, 0x8014, iwl9260_2ac_160_cfg, iwl9260_160_name), diff --git a/drivers/net/wireless/mediatek/mt76/dma.c b/drivers/net/wireless/mediatek/mt76/dma.c index 6173c80189ba..1847f55e199b 100644 --- a/drivers/net/wireless/mediatek/mt76/dma.c +++ b/drivers/net/wireless/mediatek/mt76/dma.c @@ -447,10 +447,13 @@ mt76_add_fragment(struct mt76_dev *dev, struct mt76_queue *q, void *data, struct page *page = virt_to_head_page(data); int offset = data - page_address(page); struct sk_buff *skb = q->rx_head; + struct skb_shared_info *shinfo = skb_shinfo(skb); - offset += q->buf_offset; - skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, page, offset, len, - q->buf_size); + if (shinfo->nr_frags < ARRAY_SIZE(shinfo->frags)) { + offset += q->buf_offset; + skb_add_rx_frag(skb, shinfo->nr_frags, page, offset, len, + q->buf_size); + } if (more) return; diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/trx.h b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/trx.h index 917729807514..e17f70b4d199 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/trx.h +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/trx.h @@ -561,6 +561,7 @@ static inline void clear_pci_tx_desc_content(__le32 *__pdesc, int _size) rxmcs == DESC92C_RATE11M) struct phy_status_rpt { + u8 padding[2]; u8 ch_corr[2]; u8 cck_sig_qual_ofdm_pwdb_all; u8 cck_agc_rpt_ofdm_cfosho_a; diff --git a/drivers/net/wireless/ti/wlcore/main.c b/drivers/net/wireless/ti/wlcore/main.c index ed049c9f7e29..f140f7d7f553 100644 --- a/drivers/net/wireless/ti/wlcore/main.c +++ b/drivers/net/wireless/ti/wlcore/main.c @@ -6274,7 +6274,7 @@ static int wl1271_init_ieee80211(struct wl1271 *wl) wl->hw->wiphy->flags |= WIPHY_FLAG_AP_UAPSD | WIPHY_FLAG_HAS_REMAIN_ON_CHANNEL | WIPHY_FLAG_HAS_CHANNEL_SWITCH | -+ WIPHY_FLAG_IBSS_RSN; + WIPHY_FLAG_IBSS_RSN; wl->hw->wiphy->features |= NL80211_FEATURE_AP_SCAN; diff --git a/drivers/nfc/fdp/fdp.c b/drivers/nfc/fdp/fdp.c index 0cc9ac856fe2..ed2123129e0e 100644 --- a/drivers/nfc/fdp/fdp.c +++ b/drivers/nfc/fdp/fdp.c @@ -184,7 +184,7 @@ static int fdp_nci_send_patch(struct nci_dev *ndev, u8 conn_id, u8 type) const struct firmware *fw; struct sk_buff *skb; unsigned long len; - u8 max_size, payload_size; + int max_size, payload_size; int rc = 0; if ((type == NCI_PATCH_TYPE_OTP && !info->otp_patch) || @@ -207,8 +207,7 @@ static int fdp_nci_send_patch(struct nci_dev *ndev, u8 conn_id, u8 type) while (len) { - payload_size = min_t(unsigned long, (unsigned long) max_size, - len); + payload_size = min_t(unsigned long, max_size, len); skb = nci_skb_alloc(ndev, (NCI_CTRL_HDR_SIZE + payload_size), GFP_KERNEL); diff --git a/drivers/nfc/pn544/i2c.c b/drivers/nfc/pn544/i2c.c index 720c89d6066e..4ac8cb262559 100644 --- a/drivers/nfc/pn544/i2c.c +++ b/drivers/nfc/pn544/i2c.c @@ -225,6 +225,7 @@ static void pn544_hci_i2c_platform_init(struct pn544_i2c_phy *phy) out: gpiod_set_value_cansleep(phy->gpiod_en, !phy->en_polarity); + usleep_range(10000, 15000); } static void pn544_hci_i2c_enable_mode(struct pn544_i2c_phy *phy, int run_mode) diff --git a/drivers/nfc/pn544/pn544.c b/drivers/nfc/pn544/pn544.c index 2b83156efe3f..b788870473e8 100644 --- a/drivers/nfc/pn544/pn544.c +++ b/drivers/nfc/pn544/pn544.c @@ -682,7 +682,7 @@ static int pn544_hci_tm_send(struct nfc_hci_dev *hdev, struct sk_buff *skb) static int pn544_hci_check_presence(struct nfc_hci_dev *hdev, struct nfc_target *target) { - pr_debug("supported protocol %d\b", target->supported_protocols); + pr_debug("supported protocol %d\n", target->supported_protocols); if (target->supported_protocols & (NFC_PROTO_ISO14443_MASK | NFC_PROTO_ISO14443_B_MASK)) { return nfc_hci_send_cmd(hdev, target->hci_reader_gate, diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c index ada59df642d2..a4d8c90ee7cc 100644 --- a/drivers/nvme/host/core.c +++ b/drivers/nvme/host/core.c @@ -1165,8 +1165,8 @@ static int nvme_identify_ns(struct nvme_ctrl *ctrl, static int nvme_features(struct nvme_ctrl *dev, u8 op, unsigned int fid, unsigned int dword11, void *buffer, size_t buflen, u32 *result) { + union nvme_result res = { 0 }; struct nvme_command c; - union nvme_result res; int ret; memset(&c, 0, sizeof(c)); diff --git a/drivers/nvme/host/multipath.c b/drivers/nvme/host/multipath.c index 797c18337d96..a11900cf3a36 100644 --- a/drivers/nvme/host/multipath.c +++ b/drivers/nvme/host/multipath.c @@ -715,6 +715,7 @@ int nvme_mpath_init(struct nvme_ctrl *ctrl, struct nvme_id_ctrl *id) } INIT_WORK(&ctrl->ana_work, nvme_ana_work); + kfree(ctrl->ana_log_buf); ctrl->ana_log_buf = kmalloc(ctrl->ana_log_size, GFP_KERNEL); if (!ctrl->ana_log_buf) { error = -ENOMEM; diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c index 9c80f9f08149..d3f23d6254e4 100644 --- a/drivers/nvme/host/pci.c +++ b/drivers/nvme/host/pci.c @@ -1078,9 +1078,9 @@ static int nvme_poll(struct blk_mq_hw_ctx *hctx) spin_lock(&nvmeq->cq_poll_lock); found = nvme_process_cq(nvmeq, &start, &end, -1); + nvme_complete_cqes(nvmeq, start, end); spin_unlock(&nvmeq->cq_poll_lock); - nvme_complete_cqes(nvmeq, start, end); return found; } @@ -2747,6 +2747,18 @@ static unsigned long check_vendor_combination_bug(struct pci_dev *pdev) (dmi_match(DMI_BOARD_NAME, "PRIME B350M-A") || dmi_match(DMI_BOARD_NAME, "PRIME Z370-A"))) return NVME_QUIRK_NO_APST; + } else if ((pdev->vendor == 0x144d && (pdev->device == 0xa801 || + pdev->device == 0xa808 || pdev->device == 0xa809)) || + (pdev->vendor == 0x1e0f && pdev->device == 0x0001)) { + /* + * Forcing to use host managed nvme power settings for + * lowest idle power with quick resume latency on + * Samsung and Toshiba SSDs based on suspend behavior + * on Coffee Lake board for LENOVO C640 + */ + if ((dmi_match(DMI_BOARD_VENDOR, "LENOVO")) && + dmi_match(DMI_BOARD_NAME, "LNVNB161216")) + return NVME_QUIRK_SIMPLE_SUSPEND; } return 0; @@ -3109,7 +3121,8 @@ static const struct pci_device_id nvme_id_table[] = { .driver_data = NVME_QUIRK_NO_DEEPEST_PS | NVME_QUIRK_IGNORE_DEV_SUBNQN, }, { PCI_DEVICE_CLASS(PCI_CLASS_STORAGE_EXPRESS, 0xffffff) }, - { PCI_DEVICE(PCI_VENDOR_ID_APPLE, 0x2001) }, + { PCI_DEVICE(PCI_VENDOR_ID_APPLE, 0x2001), + .driver_data = NVME_QUIRK_SINGLE_VECTOR }, { PCI_DEVICE(PCI_VENDOR_ID_APPLE, 0x2003) }, { PCI_DEVICE(PCI_VENDOR_ID_APPLE, 0x2005), .driver_data = NVME_QUIRK_SINGLE_VECTOR | diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c index 3e85c5cacefd..0fe08c4dfd2f 100644 --- a/drivers/nvme/host/rdma.c +++ b/drivers/nvme/host/rdma.c @@ -850,9 +850,11 @@ out_free_tagset: if (new) blk_mq_free_tag_set(ctrl->ctrl.admin_tagset); out_free_async_qe: - nvme_rdma_free_qe(ctrl->device->dev, &ctrl->async_event_sqe, - sizeof(struct nvme_command), DMA_TO_DEVICE); - ctrl->async_event_sqe.data = NULL; + if (ctrl->async_event_sqe.data) { + nvme_rdma_free_qe(ctrl->device->dev, &ctrl->async_event_sqe, + sizeof(struct nvme_command), DMA_TO_DEVICE); + ctrl->async_event_sqe.data = NULL; + } out_free_queue: nvme_rdma_free_queue(&ctrl->queues[0]); return error; diff --git a/drivers/nvme/target/tcp.c b/drivers/nvme/target/tcp.c index af674fc0bb1e..5bb5342b8d0c 100644 --- a/drivers/nvme/target/tcp.c +++ b/drivers/nvme/target/tcp.c @@ -515,7 +515,7 @@ static int nvmet_try_send_data_pdu(struct nvmet_tcp_cmd *cmd) return 1; } -static int nvmet_try_send_data(struct nvmet_tcp_cmd *cmd) +static int nvmet_try_send_data(struct nvmet_tcp_cmd *cmd, bool last_in_batch) { struct nvmet_tcp_queue *queue = cmd->queue; int ret; @@ -523,9 +523,15 @@ static int nvmet_try_send_data(struct nvmet_tcp_cmd *cmd) while (cmd->cur_sg) { struct page *page = sg_page(cmd->cur_sg); u32 left = cmd->cur_sg->length - cmd->offset; + int flags = MSG_DONTWAIT; + + if ((!last_in_batch && cmd->queue->send_list_len) || + cmd->wbytes_done + left < cmd->req.transfer_len || + queue->data_digest || !queue->nvme_sq.sqhd_disabled) + flags |= MSG_MORE; ret = kernel_sendpage(cmd->queue->sock, page, cmd->offset, - left, MSG_DONTWAIT | MSG_MORE); + left, flags); if (ret <= 0) return ret; @@ -660,7 +666,7 @@ static int nvmet_tcp_try_send_one(struct nvmet_tcp_queue *queue, } if (cmd->state == NVMET_TCP_SEND_DATA) { - ret = nvmet_try_send_data(cmd); + ret = nvmet_try_send_data(cmd, last_in_batch); if (ret <= 0) goto done_send; } diff --git a/drivers/of/of_mdio.c b/drivers/of/of_mdio.c index 8270bbf505fb..9f982c0627a0 100644 --- a/drivers/of/of_mdio.c +++ b/drivers/of/of_mdio.c @@ -306,6 +306,7 @@ int of_mdiobus_register(struct mii_bus *mdio, struct device_node *np) rc = of_mdiobus_register_phy(mdio, child, addr); if (rc && rc != -ENODEV) goto unregister; + break; } } } diff --git a/drivers/pci/controller/pcie-brcmstb.c b/drivers/pci/controller/pcie-brcmstb.c index d20aabc26273..3a10e678c7f4 100644 --- a/drivers/pci/controller/pcie-brcmstb.c +++ b/drivers/pci/controller/pcie-brcmstb.c @@ -670,7 +670,7 @@ static inline int brcm_pcie_get_rc_bar2_size_and_offset(struct brcm_pcie *pcie, * outbound memory @ 3GB). So instead it will start at the 1x * multiple of its size */ - if (!*rc_bar2_size || *rc_bar2_offset % *rc_bar2_size || + if (!*rc_bar2_size || (*rc_bar2_offset & (*rc_bar2_size - 1)) || (*rc_bar2_offset < SZ_4G && *rc_bar2_offset > SZ_2G)) { dev_err(dev, "Invalid rc_bar2_offset/size: size 0x%llx, off 0x%llx\n", *rc_bar2_size, *rc_bar2_offset); diff --git a/drivers/perf/arm_pmu_acpi.c b/drivers/perf/arm_pmu_acpi.c index acce8781c456..f5c7a845cd7b 100644 --- a/drivers/perf/arm_pmu_acpi.c +++ b/drivers/perf/arm_pmu_acpi.c @@ -24,8 +24,6 @@ static int arm_pmu_acpi_register_irq(int cpu) int gsi, trigger; gicc = acpi_cpu_get_madt_gicc(cpu); - if (WARN_ON(!gicc)) - return -EINVAL; gsi = gicc->performance_interrupt; @@ -64,11 +62,10 @@ static void arm_pmu_acpi_unregister_irq(int cpu) int gsi; gicc = acpi_cpu_get_madt_gicc(cpu); - if (!gicc) - return; gsi = gicc->performance_interrupt; - acpi_unregister_gsi(gsi); + if (gsi) + acpi_unregister_gsi(gsi); } #if IS_ENABLED(CONFIG_ARM_SPE_PMU) diff --git a/drivers/perf/fsl_imx8_ddr_perf.c b/drivers/perf/fsl_imx8_ddr_perf.c index 95dca2cb5265..90884d14f95f 100644 --- a/drivers/perf/fsl_imx8_ddr_perf.c +++ b/drivers/perf/fsl_imx8_ddr_perf.c @@ -388,9 +388,10 @@ static void ddr_perf_counter_enable(struct ddr_pmu *pmu, int config, if (enable) { /* - * must disable first, then enable again - * otherwise, cycle counter will not work - * if previous state is enabled. + * cycle counter is special which should firstly write 0 then + * write 1 into CLEAR bit to clear it. Other counters only + * need write 0 into CLEAR bit and it turns out to be 1 by + * hardware. Below enable flow is harmless for all counters. */ writel(0, pmu->base + reg); val = CNTL_EN | CNTL_CLEAR; @@ -398,7 +399,8 @@ static void ddr_perf_counter_enable(struct ddr_pmu *pmu, int config, writel(val, pmu->base + reg); } else { /* Disable counter */ - writel(0, pmu->base + reg); + val = readl_relaxed(pmu->base + reg) & CNTL_EN_MASK; + writel(val, pmu->base + reg); } } diff --git a/drivers/phy/allwinner/phy-sun50i-usb3.c b/drivers/phy/allwinner/phy-sun50i-usb3.c index 1169f3e83a6f..b1c04f71a31d 100644 --- a/drivers/phy/allwinner/phy-sun50i-usb3.c +++ b/drivers/phy/allwinner/phy-sun50i-usb3.c @@ -49,7 +49,7 @@ #define SUNXI_LOS_BIAS(n) ((n) << 3) #define SUNXI_LOS_BIAS_MASK GENMASK(5, 3) #define SUNXI_TXVBOOSTLVL(n) ((n) << 0) -#define SUNXI_TXVBOOSTLVL_MASK GENMASK(0, 2) +#define SUNXI_TXVBOOSTLVL_MASK GENMASK(2, 0) struct sun50i_usb3_phy { struct phy *phy; diff --git a/drivers/phy/broadcom/phy-brcm-sata.c b/drivers/phy/broadcom/phy-brcm-sata.c index 4710cfcc3037..18251f232172 100644 --- a/drivers/phy/broadcom/phy-brcm-sata.c +++ b/drivers/phy/broadcom/phy-brcm-sata.c @@ -186,29 +186,6 @@ enum sata_phy_ctrl_regs { PHY_CTRL_1_RESET = BIT(0), }; -static inline void __iomem *brcm_sata_pcb_base(struct brcm_sata_port *port) -{ - struct brcm_sata_phy *priv = port->phy_priv; - u32 size = 0; - - switch (priv->version) { - case BRCM_SATA_PHY_STB_16NM: - case BRCM_SATA_PHY_STB_28NM: - case BRCM_SATA_PHY_IPROC_NS2: - case BRCM_SATA_PHY_DSL_28NM: - size = SATA_PCB_REG_28NM_SPACE_SIZE; - break; - case BRCM_SATA_PHY_STB_40NM: - size = SATA_PCB_REG_40NM_SPACE_SIZE; - break; - default: - dev_err(priv->dev, "invalid phy version\n"); - break; - } - - return priv->phy_base + (port->portnum * size); -} - static inline void __iomem *brcm_sata_ctrl_base(struct brcm_sata_port *port) { struct brcm_sata_phy *priv = port->phy_priv; @@ -226,19 +203,34 @@ static inline void __iomem *brcm_sata_ctrl_base(struct brcm_sata_port *port) return priv->ctrl_base + (port->portnum * size); } -static void brcm_sata_phy_wr(void __iomem *pcb_base, u32 bank, +static void brcm_sata_phy_wr(struct brcm_sata_port *port, u32 bank, u32 ofs, u32 msk, u32 value) { + struct brcm_sata_phy *priv = port->phy_priv; + void __iomem *pcb_base = priv->phy_base; u32 tmp; + if (priv->version == BRCM_SATA_PHY_STB_40NM) + bank += (port->portnum * SATA_PCB_REG_40NM_SPACE_SIZE); + else + pcb_base += (port->portnum * SATA_PCB_REG_28NM_SPACE_SIZE); + writel(bank, pcb_base + SATA_PCB_BANK_OFFSET); tmp = readl(pcb_base + SATA_PCB_REG_OFFSET(ofs)); tmp = (tmp & msk) | value; writel(tmp, pcb_base + SATA_PCB_REG_OFFSET(ofs)); } -static u32 brcm_sata_phy_rd(void __iomem *pcb_base, u32 bank, u32 ofs) +static u32 brcm_sata_phy_rd(struct brcm_sata_port *port, u32 bank, u32 ofs) { + struct brcm_sata_phy *priv = port->phy_priv; + void __iomem *pcb_base = priv->phy_base; + + if (priv->version == BRCM_SATA_PHY_STB_40NM) + bank += (port->portnum * SATA_PCB_REG_40NM_SPACE_SIZE); + else + pcb_base += (port->portnum * SATA_PCB_REG_28NM_SPACE_SIZE); + writel(bank, pcb_base + SATA_PCB_BANK_OFFSET); return readl(pcb_base + SATA_PCB_REG_OFFSET(ofs)); } @@ -250,16 +242,15 @@ static u32 brcm_sata_phy_rd(void __iomem *pcb_base, u32 bank, u32 ofs) static void brcm_stb_sata_ssc_init(struct brcm_sata_port *port) { - void __iomem *base = brcm_sata_pcb_base(port); struct brcm_sata_phy *priv = port->phy_priv; u32 tmp; /* override the TX spread spectrum setting */ tmp = TXPMD_CONTROL1_TX_SSC_EN_FRC_VAL | TXPMD_CONTROL1_TX_SSC_EN_FRC; - brcm_sata_phy_wr(base, TXPMD_REG_BANK, TXPMD_CONTROL1, ~tmp, tmp); + brcm_sata_phy_wr(port, TXPMD_REG_BANK, TXPMD_CONTROL1, ~tmp, tmp); /* set fixed min freq */ - brcm_sata_phy_wr(base, TXPMD_REG_BANK, TXPMD_TX_FREQ_CTRL_CONTROL2, + brcm_sata_phy_wr(port, TXPMD_REG_BANK, TXPMD_TX_FREQ_CTRL_CONTROL2, ~TXPMD_TX_FREQ_CTRL_CONTROL2_FMIN_MASK, STB_FMIN_VAL_DEFAULT); @@ -271,7 +262,7 @@ static void brcm_stb_sata_ssc_init(struct brcm_sata_port *port) tmp = STB_FMAX_VAL_DEFAULT; } - brcm_sata_phy_wr(base, TXPMD_REG_BANK, TXPMD_TX_FREQ_CTRL_CONTROL3, + brcm_sata_phy_wr(port, TXPMD_REG_BANK, TXPMD_TX_FREQ_CTRL_CONTROL3, ~TXPMD_TX_FREQ_CTRL_CONTROL3_FMAX_MASK, tmp); } @@ -280,7 +271,6 @@ static void brcm_stb_sata_ssc_init(struct brcm_sata_port *port) static int brcm_stb_sata_rxaeq_init(struct brcm_sata_port *port) { - void __iomem *base = brcm_sata_pcb_base(port); u32 tmp = 0, reg = 0; switch (port->rxaeq_mode) { @@ -301,8 +291,8 @@ static int brcm_stb_sata_rxaeq_init(struct brcm_sata_port *port) break; } - brcm_sata_phy_wr(base, AEQRX_REG_BANK_0, reg, ~tmp, tmp); - brcm_sata_phy_wr(base, AEQRX_REG_BANK_1, reg, ~tmp, tmp); + brcm_sata_phy_wr(port, AEQRX_REG_BANK_0, reg, ~tmp, tmp); + brcm_sata_phy_wr(port, AEQRX_REG_BANK_1, reg, ~tmp, tmp); return 0; } @@ -316,18 +306,17 @@ static int brcm_stb_sata_init(struct brcm_sata_port *port) static int brcm_stb_sata_16nm_ssc_init(struct brcm_sata_port *port) { - void __iomem *base = brcm_sata_pcb_base(port); u32 tmp, value; /* Reduce CP tail current to 1/16th of its default value */ - brcm_sata_phy_wr(base, PLL1_REG_BANK, PLL1_ACTRL6, 0, 0x141); + brcm_sata_phy_wr(port, PLL1_REG_BANK, PLL1_ACTRL6, 0, 0x141); /* Turn off CP tail current boost */ - brcm_sata_phy_wr(base, PLL1_REG_BANK, PLL1_ACTRL8, 0, 0xc006); + brcm_sata_phy_wr(port, PLL1_REG_BANK, PLL1_ACTRL8, 0, 0xc006); /* Set a specific AEQ equalizer value */ tmp = AEQ_FRC_EQ_FORCE_VAL | AEQ_FRC_EQ_FORCE; - brcm_sata_phy_wr(base, AEQRX_REG_BANK_0, AEQ_FRC_EQ, + brcm_sata_phy_wr(port, AEQRX_REG_BANK_0, AEQ_FRC_EQ, ~(tmp | AEQ_RFZ_FRC_VAL | AEQ_FRC_EQ_VAL_MASK << AEQ_FRC_EQ_VAL_SHIFT), tmp | 32 << AEQ_FRC_EQ_VAL_SHIFT); @@ -337,7 +326,7 @@ static int brcm_stb_sata_16nm_ssc_init(struct brcm_sata_port *port) value = 0x52; else value = 0; - brcm_sata_phy_wr(base, RXPMD_REG_BANK, RXPMD_RX_CDR_CONTROL1, + brcm_sata_phy_wr(port, RXPMD_REG_BANK, RXPMD_RX_CDR_CONTROL1, ~RXPMD_RX_PPM_VAL_MASK, value); /* Set proportional loop bandwith Gen1/2/3 */ @@ -352,7 +341,7 @@ static int brcm_stb_sata_16nm_ssc_init(struct brcm_sata_port *port) value = 1 << RXPMD_G1_CDR_PROP_BW_SHIFT | 1 << RXPMD_G2_CDR_PROP_BW_SHIFT | 1 << RXPMD_G3_CDR_PROB_BW_SHIFT; - brcm_sata_phy_wr(base, RXPMD_REG_BANK, RXPMD_RX_CDR_CDR_PROP_BW, ~tmp, + brcm_sata_phy_wr(port, RXPMD_REG_BANK, RXPMD_RX_CDR_CDR_PROP_BW, ~tmp, value); /* Set CDR integral loop acquisition bandwidth for Gen1/2/3 */ @@ -365,7 +354,7 @@ static int brcm_stb_sata_16nm_ssc_init(struct brcm_sata_port *port) 1 << RXPMD_G3_CDR_ACQ_INT_BW_SHIFT; else value = 0; - brcm_sata_phy_wr(base, RXPMD_REG_BANK, RXPMD_RX_CDR_CDR_ACQ_INTEG_BW, + brcm_sata_phy_wr(port, RXPMD_REG_BANK, RXPMD_RX_CDR_CDR_ACQ_INTEG_BW, ~tmp, value); /* Set CDR integral loop locking bandwidth to 1 for Gen 1/2/3 */ @@ -378,7 +367,7 @@ static int brcm_stb_sata_16nm_ssc_init(struct brcm_sata_port *port) 1 << RXPMD_G3_CDR_LOCK_INT_BW_SHIFT; else value = 0; - brcm_sata_phy_wr(base, RXPMD_REG_BANK, RXPMD_RX_CDR_CDR_LOCK_INTEG_BW, + brcm_sata_phy_wr(port, RXPMD_REG_BANK, RXPMD_RX_CDR_CDR_LOCK_INTEG_BW, ~tmp, value); /* Set no guard band and clamp CDR */ @@ -387,11 +376,11 @@ static int brcm_stb_sata_16nm_ssc_init(struct brcm_sata_port *port) value = 0x51; else value = 0; - brcm_sata_phy_wr(base, RXPMD_REG_BANK, RXPMD_RX_FREQ_MON_CONTROL1, + brcm_sata_phy_wr(port, RXPMD_REG_BANK, RXPMD_RX_FREQ_MON_CONTROL1, ~tmp, RXPMD_MON_CORRECT_EN | value); /* Turn on/off SSC */ - brcm_sata_phy_wr(base, TX_REG_BANK, TX_ACTRL5, ~TX_ACTRL5_SSC_EN, + brcm_sata_phy_wr(port, TX_REG_BANK, TX_ACTRL5, ~TX_ACTRL5_SSC_EN, port->ssc_en ? TX_ACTRL5_SSC_EN : 0); return 0; @@ -411,7 +400,6 @@ static int brcm_ns2_sata_init(struct brcm_sata_port *port) { int try; unsigned int val; - void __iomem *base = brcm_sata_pcb_base(port); void __iomem *ctrl_base = brcm_sata_ctrl_base(port); struct device *dev = port->phy_priv->dev; @@ -421,24 +409,24 @@ static int brcm_ns2_sata_init(struct brcm_sata_port *port) val |= (0x4 << OOB_CTRL1_BURST_MIN_SHIFT); val |= (0x9 << OOB_CTRL1_WAKE_IDLE_MAX_SHIFT); val |= (0x3 << OOB_CTRL1_WAKE_IDLE_MIN_SHIFT); - brcm_sata_phy_wr(base, OOB_REG_BANK, OOB_CTRL1, 0x0, val); + brcm_sata_phy_wr(port, OOB_REG_BANK, OOB_CTRL1, 0x0, val); val = 0x0; val |= (0x1b << OOB_CTRL2_RESET_IDLE_MAX_SHIFT); val |= (0x2 << OOB_CTRL2_BURST_CNT_SHIFT); val |= (0x9 << OOB_CTRL2_RESET_IDLE_MIN_SHIFT); - brcm_sata_phy_wr(base, OOB_REG_BANK, OOB_CTRL2, 0x0, val); + brcm_sata_phy_wr(port, OOB_REG_BANK, OOB_CTRL2, 0x0, val); /* Configure PHY PLL register bank 1 */ val = NS2_PLL1_ACTRL2_MAGIC; - brcm_sata_phy_wr(base, PLL1_REG_BANK, PLL1_ACTRL2, 0x0, val); + brcm_sata_phy_wr(port, PLL1_REG_BANK, PLL1_ACTRL2, 0x0, val); val = NS2_PLL1_ACTRL3_MAGIC; - brcm_sata_phy_wr(base, PLL1_REG_BANK, PLL1_ACTRL3, 0x0, val); + brcm_sata_phy_wr(port, PLL1_REG_BANK, PLL1_ACTRL3, 0x0, val); val = NS2_PLL1_ACTRL4_MAGIC; - brcm_sata_phy_wr(base, PLL1_REG_BANK, PLL1_ACTRL4, 0x0, val); + brcm_sata_phy_wr(port, PLL1_REG_BANK, PLL1_ACTRL4, 0x0, val); /* Configure PHY BLOCK0 register bank */ /* Set oob_clk_sel to refclk/2 */ - brcm_sata_phy_wr(base, BLOCK0_REG_BANK, BLOCK0_SPARE, + brcm_sata_phy_wr(port, BLOCK0_REG_BANK, BLOCK0_SPARE, ~BLOCK0_SPARE_OOB_CLK_SEL_MASK, BLOCK0_SPARE_OOB_CLK_SEL_REFBY2); @@ -451,7 +439,7 @@ static int brcm_ns2_sata_init(struct brcm_sata_port *port) /* Wait for PHY PLL lock by polling pll_lock bit */ try = 50; while (try) { - val = brcm_sata_phy_rd(base, BLOCK0_REG_BANK, + val = brcm_sata_phy_rd(port, BLOCK0_REG_BANK, BLOCK0_XGXSSTATUS); if (val & BLOCK0_XGXSSTATUS_PLL_LOCK) break; @@ -471,9 +459,7 @@ static int brcm_ns2_sata_init(struct brcm_sata_port *port) static int brcm_nsp_sata_init(struct brcm_sata_port *port) { - struct brcm_sata_phy *priv = port->phy_priv; struct device *dev = port->phy_priv->dev; - void __iomem *base = priv->phy_base; unsigned int oob_bank; unsigned int val, try; @@ -490,36 +476,36 @@ static int brcm_nsp_sata_init(struct brcm_sata_port *port) val |= (0x06 << OOB_CTRL1_BURST_MIN_SHIFT); val |= (0x0f << OOB_CTRL1_WAKE_IDLE_MAX_SHIFT); val |= (0x06 << OOB_CTRL1_WAKE_IDLE_MIN_SHIFT); - brcm_sata_phy_wr(base, oob_bank, OOB_CTRL1, 0x0, val); + brcm_sata_phy_wr(port, oob_bank, OOB_CTRL1, 0x0, val); val = 0x0; val |= (0x2e << OOB_CTRL2_RESET_IDLE_MAX_SHIFT); val |= (0x02 << OOB_CTRL2_BURST_CNT_SHIFT); val |= (0x16 << OOB_CTRL2_RESET_IDLE_MIN_SHIFT); - brcm_sata_phy_wr(base, oob_bank, OOB_CTRL2, 0x0, val); + brcm_sata_phy_wr(port, oob_bank, OOB_CTRL2, 0x0, val); - brcm_sata_phy_wr(base, PLL_REG_BANK_0, PLL_ACTRL2, + brcm_sata_phy_wr(port, PLL_REG_BANK_0, PLL_ACTRL2, ~(PLL_ACTRL2_SELDIV_MASK << PLL_ACTRL2_SELDIV_SHIFT), 0x0c << PLL_ACTRL2_SELDIV_SHIFT); - brcm_sata_phy_wr(base, PLL_REG_BANK_0, PLL_CAP_CONTROL, + brcm_sata_phy_wr(port, PLL_REG_BANK_0, PLL_CAP_CONTROL, 0xff0, 0x4f0); val = PLLCONTROL_0_FREQ_DET_RESTART | PLLCONTROL_0_FREQ_MONITOR; - brcm_sata_phy_wr(base, PLL_REG_BANK_0, PLL_REG_BANK_0_PLLCONTROL_0, + brcm_sata_phy_wr(port, PLL_REG_BANK_0, PLL_REG_BANK_0_PLLCONTROL_0, ~val, val); val = PLLCONTROL_0_SEQ_START; - brcm_sata_phy_wr(base, PLL_REG_BANK_0, PLL_REG_BANK_0_PLLCONTROL_0, + brcm_sata_phy_wr(port, PLL_REG_BANK_0, PLL_REG_BANK_0_PLLCONTROL_0, ~val, 0); mdelay(10); - brcm_sata_phy_wr(base, PLL_REG_BANK_0, PLL_REG_BANK_0_PLLCONTROL_0, + brcm_sata_phy_wr(port, PLL_REG_BANK_0, PLL_REG_BANK_0_PLLCONTROL_0, ~val, val); /* Wait for pll_seq_done bit */ try = 50; while (--try) { - val = brcm_sata_phy_rd(base, BLOCK0_REG_BANK, + val = brcm_sata_phy_rd(port, BLOCK0_REG_BANK, BLOCK0_XGXSSTATUS); if (val & BLOCK0_XGXSSTATUS_PLL_LOCK) break; @@ -546,27 +532,25 @@ static int brcm_nsp_sata_init(struct brcm_sata_port *port) static int brcm_sr_sata_init(struct brcm_sata_port *port) { - struct brcm_sata_phy *priv = port->phy_priv; struct device *dev = port->phy_priv->dev; - void __iomem *base = priv->phy_base; unsigned int val, try; /* Configure PHY PLL register bank 1 */ val = SR_PLL1_ACTRL2_MAGIC; - brcm_sata_phy_wr(base, PLL1_REG_BANK, PLL1_ACTRL2, 0x0, val); + brcm_sata_phy_wr(port, PLL1_REG_BANK, PLL1_ACTRL2, 0x0, val); val = SR_PLL1_ACTRL3_MAGIC; - brcm_sata_phy_wr(base, PLL1_REG_BANK, PLL1_ACTRL3, 0x0, val); + brcm_sata_phy_wr(port, PLL1_REG_BANK, PLL1_ACTRL3, 0x0, val); val = SR_PLL1_ACTRL4_MAGIC; - brcm_sata_phy_wr(base, PLL1_REG_BANK, PLL1_ACTRL4, 0x0, val); + brcm_sata_phy_wr(port, PLL1_REG_BANK, PLL1_ACTRL4, 0x0, val); /* Configure PHY PLL register bank 0 */ val = SR_PLL0_ACTRL6_MAGIC; - brcm_sata_phy_wr(base, PLL_REG_BANK_0, PLL_ACTRL6, 0x0, val); + brcm_sata_phy_wr(port, PLL_REG_BANK_0, PLL_ACTRL6, 0x0, val); /* Wait for PHY PLL lock by polling pll_lock bit */ try = 50; do { - val = brcm_sata_phy_rd(base, BLOCK0_REG_BANK, + val = brcm_sata_phy_rd(port, BLOCK0_REG_BANK, BLOCK0_XGXSSTATUS); if (val & BLOCK0_XGXSSTATUS_PLL_LOCK) break; @@ -581,7 +565,7 @@ static int brcm_sr_sata_init(struct brcm_sata_port *port) } /* Invert Tx polarity */ - brcm_sata_phy_wr(base, TX_REG_BANK, TX_ACTRL0, + brcm_sata_phy_wr(port, TX_REG_BANK, TX_ACTRL0, ~TX_ACTRL0_TXPOL_FLIP, TX_ACTRL0_TXPOL_FLIP); /* Configure OOB control to handle 100MHz reference clock */ @@ -589,52 +573,51 @@ static int brcm_sr_sata_init(struct brcm_sata_port *port) (0x4 << OOB_CTRL1_BURST_MIN_SHIFT) | (0x8 << OOB_CTRL1_WAKE_IDLE_MAX_SHIFT) | (0x3 << OOB_CTRL1_WAKE_IDLE_MIN_SHIFT)); - brcm_sata_phy_wr(base, OOB_REG_BANK, OOB_CTRL1, 0x0, val); + brcm_sata_phy_wr(port, OOB_REG_BANK, OOB_CTRL1, 0x0, val); val = ((0x1b << OOB_CTRL2_RESET_IDLE_MAX_SHIFT) | (0x2 << OOB_CTRL2_BURST_CNT_SHIFT) | (0x9 << OOB_CTRL2_RESET_IDLE_MIN_SHIFT)); - brcm_sata_phy_wr(base, OOB_REG_BANK, OOB_CTRL2, 0x0, val); + brcm_sata_phy_wr(port, OOB_REG_BANK, OOB_CTRL2, 0x0, val); return 0; } static int brcm_dsl_sata_init(struct brcm_sata_port *port) { - void __iomem *base = brcm_sata_pcb_base(port); struct device *dev = port->phy_priv->dev; unsigned int try; u32 tmp; - brcm_sata_phy_wr(base, PLL1_REG_BANK, PLL1_ACTRL7, 0, 0x873); + brcm_sata_phy_wr(port, PLL1_REG_BANK, PLL1_ACTRL7, 0, 0x873); - brcm_sata_phy_wr(base, PLL1_REG_BANK, PLL1_ACTRL6, 0, 0xc000); + brcm_sata_phy_wr(port, PLL1_REG_BANK, PLL1_ACTRL6, 0, 0xc000); - brcm_sata_phy_wr(base, PLL_REG_BANK_0, PLL_REG_BANK_0_PLLCONTROL_0, + brcm_sata_phy_wr(port, PLL_REG_BANK_0, PLL_REG_BANK_0_PLLCONTROL_0, 0, 0x3089); usleep_range(1000, 2000); - brcm_sata_phy_wr(base, PLL_REG_BANK_0, PLL_REG_BANK_0_PLLCONTROL_0, + brcm_sata_phy_wr(port, PLL_REG_BANK_0, PLL_REG_BANK_0_PLLCONTROL_0, 0, 0x3088); usleep_range(1000, 2000); - brcm_sata_phy_wr(base, AEQRX_REG_BANK_1, AEQRX_SLCAL0_CTRL0, + brcm_sata_phy_wr(port, AEQRX_REG_BANK_1, AEQRX_SLCAL0_CTRL0, 0, 0x3000); - brcm_sata_phy_wr(base, AEQRX_REG_BANK_1, AEQRX_SLCAL1_CTRL0, + brcm_sata_phy_wr(port, AEQRX_REG_BANK_1, AEQRX_SLCAL1_CTRL0, 0, 0x3000); usleep_range(1000, 2000); - brcm_sata_phy_wr(base, PLL_REG_BANK_0, PLL_CAP_CHARGE_TIME, 0, 0x32); + brcm_sata_phy_wr(port, PLL_REG_BANK_0, PLL_CAP_CHARGE_TIME, 0, 0x32); - brcm_sata_phy_wr(base, PLL_REG_BANK_0, PLL_VCO_CAL_THRESH, 0, 0xa); + brcm_sata_phy_wr(port, PLL_REG_BANK_0, PLL_VCO_CAL_THRESH, 0, 0xa); - brcm_sata_phy_wr(base, PLL_REG_BANK_0, PLL_FREQ_DET_TIME, 0, 0x64); + brcm_sata_phy_wr(port, PLL_REG_BANK_0, PLL_FREQ_DET_TIME, 0, 0x64); usleep_range(1000, 2000); /* Acquire PLL lock */ try = 50; while (try) { - tmp = brcm_sata_phy_rd(base, BLOCK0_REG_BANK, + tmp = brcm_sata_phy_rd(port, BLOCK0_REG_BANK, BLOCK0_XGXSSTATUS); if (tmp & BLOCK0_XGXSSTATUS_PLL_LOCK) break; @@ -687,10 +670,9 @@ static int brcm_sata_phy_init(struct phy *phy) static void brcm_stb_sata_calibrate(struct brcm_sata_port *port) { - void __iomem *base = brcm_sata_pcb_base(port); u32 tmp = BIT(8); - brcm_sata_phy_wr(base, RXPMD_REG_BANK, RXPMD_RX_FREQ_MON_CONTROL1, + brcm_sata_phy_wr(port, RXPMD_REG_BANK, RXPMD_RX_FREQ_MON_CONTROL1, ~tmp, tmp); } diff --git a/drivers/phy/motorola/phy-mapphone-mdm6600.c b/drivers/phy/motorola/phy-mapphone-mdm6600.c index f20524f0c21d..94a34cf75eb3 100644 --- a/drivers/phy/motorola/phy-mapphone-mdm6600.c +++ b/drivers/phy/motorola/phy-mapphone-mdm6600.c @@ -20,6 +20,7 @@ #define PHY_MDM6600_PHY_DELAY_MS 4000 /* PHY enable 2.2s to 3.5s */ #define PHY_MDM6600_ENABLED_DELAY_MS 8000 /* 8s more total for MDM6600 */ +#define PHY_MDM6600_WAKE_KICK_MS 600 /* time on after GPIO toggle */ #define MDM6600_MODEM_IDLE_DELAY_MS 1000 /* modem after USB suspend */ #define MDM6600_MODEM_WAKE_DELAY_MS 200 /* modem response after idle */ @@ -243,10 +244,24 @@ static irqreturn_t phy_mdm6600_wakeirq_thread(int irq, void *data) { struct phy_mdm6600 *ddata = data; struct gpio_desc *mode_gpio1; + int error, wakeup; mode_gpio1 = ddata->mode_gpios->desc[PHY_MDM6600_MODE1]; - dev_dbg(ddata->dev, "OOB wake on mode_gpio1: %i\n", - gpiod_get_value(mode_gpio1)); + wakeup = gpiod_get_value(mode_gpio1); + if (!wakeup) + return IRQ_NONE; + + dev_dbg(ddata->dev, "OOB wake on mode_gpio1: %i\n", wakeup); + error = pm_runtime_get_sync(ddata->dev); + if (error < 0) { + pm_runtime_put_noidle(ddata->dev); + + return IRQ_NONE; + } + + /* Just wake-up and kick the autosuspend timer */ + pm_runtime_mark_last_busy(ddata->dev); + pm_runtime_put_autosuspend(ddata->dev); return IRQ_HANDLED; } @@ -496,8 +511,14 @@ static void phy_mdm6600_modem_wake(struct work_struct *work) ddata = container_of(work, struct phy_mdm6600, modem_wake_work.work); phy_mdm6600_wake_modem(ddata); + + /* + * The modem does not always stay awake 1.2 seconds after toggling + * the wake GPIO, and sometimes it idles after about some 600 ms + * making writes time out. + */ schedule_delayed_work(&ddata->modem_wake_work, - msecs_to_jiffies(MDM6600_MODEM_IDLE_DELAY_MS)); + msecs_to_jiffies(PHY_MDM6600_WAKE_KICK_MS)); } static int __maybe_unused phy_mdm6600_runtime_suspend(struct device *dev) diff --git a/drivers/phy/phy-core.c b/drivers/phy/phy-core.c index cd5a6c95dbdc..a27b8d578d7f 100644 --- a/drivers/phy/phy-core.c +++ b/drivers/phy/phy-core.c @@ -688,11 +688,9 @@ struct phy *phy_get(struct device *dev, const char *string) get_device(&phy->dev); link = device_link_add(dev, &phy->dev, DL_FLAG_STATELESS); - if (!link) { - dev_err(dev, "failed to create device link to %s\n", + if (!link) + dev_dbg(dev, "failed to create device link to %s\n", dev_name(phy->dev.parent)); - return ERR_PTR(-EINVAL); - } return phy; } @@ -803,11 +801,9 @@ struct phy *devm_of_phy_get(struct device *dev, struct device_node *np, } link = device_link_add(dev, &phy->dev, DL_FLAG_STATELESS); - if (!link) { - dev_err(dev, "failed to create device link to %s\n", + if (!link) + dev_dbg(dev, "failed to create device link to %s\n", dev_name(phy->dev.parent)); - return ERR_PTR(-EINVAL); - } return phy; } @@ -852,11 +848,9 @@ struct phy *devm_of_phy_get_by_index(struct device *dev, struct device_node *np, devres_add(dev, ptr); link = device_link_add(dev, &phy->dev, DL_FLAG_STATELESS); - if (!link) { - dev_err(dev, "failed to create device link to %s\n", + if (!link) + dev_dbg(dev, "failed to create device link to %s\n", dev_name(phy->dev.parent)); - return ERR_PTR(-EINVAL); - } return phy; } diff --git a/drivers/phy/ti/phy-gmii-sel.c b/drivers/phy/ti/phy-gmii-sel.c index a28bd15297f5..1c536fc03c83 100644 --- a/drivers/phy/ti/phy-gmii-sel.c +++ b/drivers/phy/ti/phy-gmii-sel.c @@ -80,20 +80,20 @@ static int phy_gmii_sel_mode(struct phy *phy, enum phy_mode mode, int submode) break; case PHY_INTERFACE_MODE_MII: - mode = AM33XX_GMII_SEL_MODE_MII; + case PHY_INTERFACE_MODE_GMII: + gmii_sel_mode = AM33XX_GMII_SEL_MODE_MII; break; default: - dev_warn(dev, - "port%u: unsupported mode: \"%s\". Defaulting to MII.\n", - if_phy->id, phy_modes(rgmii_id)); + dev_warn(dev, "port%u: unsupported mode: \"%s\"\n", + if_phy->id, phy_modes(submode)); return -EINVAL; } if_phy->phy_if_mode = submode; dev_dbg(dev, "%s id:%u mode:%u rgmii_id:%d rmii_clk_ext:%d\n", - __func__, if_phy->id, mode, rgmii_id, + __func__, if_phy->id, submode, rgmii_id, if_phy->rmii_clock_external); regfield = if_phy->fields[PHY_GMII_SEL_PORT_MODE]; diff --git a/drivers/pinctrl/cirrus/pinctrl-madera-core.c b/drivers/pinctrl/cirrus/pinctrl-madera-core.c index 7b6409ef553c..dce2626384a9 100644 --- a/drivers/pinctrl/cirrus/pinctrl-madera-core.c +++ b/drivers/pinctrl/cirrus/pinctrl-madera-core.c @@ -1073,13 +1073,26 @@ static int madera_pin_probe(struct platform_device *pdev) return ret; } + platform_set_drvdata(pdev, priv); + dev_dbg(priv->dev, "pinctrl probed ok\n"); return 0; } +static int madera_pin_remove(struct platform_device *pdev) +{ + struct madera_pin_private *priv = platform_get_drvdata(pdev); + + if (priv->madera->pdata.gpio_configs) + pinctrl_unregister_mappings(priv->madera->pdata.gpio_configs); + + return 0; +} + static struct platform_driver madera_pin_driver = { .probe = madera_pin_probe, + .remove = madera_pin_remove, .driver = { .name = "madera-pinctrl", }, diff --git a/drivers/pinctrl/core.c b/drivers/pinctrl/core.c index 446d84fe0e31..f23c55e22195 100644 --- a/drivers/pinctrl/core.c +++ b/drivers/pinctrl/core.c @@ -2021,7 +2021,6 @@ static int pinctrl_claim_hogs(struct pinctrl_dev *pctldev) return PTR_ERR(pctldev->p); } - kref_get(&pctldev->p->users); pctldev->hog_default = pinctrl_lookup_state(pctldev->p, PINCTRL_STATE_DEFAULT); if (IS_ERR(pctldev->hog_default)) { diff --git a/drivers/pinctrl/freescale/pinctrl-scu.c b/drivers/pinctrl/freescale/pinctrl-scu.c index 73bf1d9f9cc6..23cf04bdfc55 100644 --- a/drivers/pinctrl/freescale/pinctrl-scu.c +++ b/drivers/pinctrl/freescale/pinctrl-scu.c @@ -23,12 +23,12 @@ struct imx_sc_msg_req_pad_set { struct imx_sc_rpc_msg hdr; u32 val; u16 pad; -} __packed; +} __packed __aligned(4); struct imx_sc_msg_req_pad_get { struct imx_sc_rpc_msg hdr; u16 pad; -} __packed; +} __packed __aligned(4); struct imx_sc_msg_resp_pad_get { struct imx_sc_rpc_msg hdr; diff --git a/drivers/pinctrl/meson/pinctrl-meson-gxl.c b/drivers/pinctrl/meson/pinctrl-meson-gxl.c index 1b6e8646700f..2ac921c83da9 100644 --- a/drivers/pinctrl/meson/pinctrl-meson-gxl.c +++ b/drivers/pinctrl/meson/pinctrl-meson-gxl.c @@ -147,8 +147,8 @@ static const unsigned int sdio_d0_pins[] = { GPIOX_0 }; static const unsigned int sdio_d1_pins[] = { GPIOX_1 }; static const unsigned int sdio_d2_pins[] = { GPIOX_2 }; static const unsigned int sdio_d3_pins[] = { GPIOX_3 }; -static const unsigned int sdio_cmd_pins[] = { GPIOX_4 }; -static const unsigned int sdio_clk_pins[] = { GPIOX_5 }; +static const unsigned int sdio_clk_pins[] = { GPIOX_4 }; +static const unsigned int sdio_cmd_pins[] = { GPIOX_5 }; static const unsigned int sdio_irq_pins[] = { GPIOX_7 }; static const unsigned int nand_ce0_pins[] = { BOOT_8 }; diff --git a/drivers/pinctrl/pinctrl-falcon.c b/drivers/pinctrl/pinctrl-falcon.c index a454f57c264e..62c02b969327 100644 --- a/drivers/pinctrl/pinctrl-falcon.c +++ b/drivers/pinctrl/pinctrl-falcon.c @@ -451,7 +451,7 @@ static int pinctrl_falcon_probe(struct platform_device *pdev) falcon_info.clk[*bank] = clk_get(&ppdev->dev, NULL); if (IS_ERR(falcon_info.clk[*bank])) { dev_err(&ppdev->dev, "failed to get clock\n"); - of_node_put(np) + of_node_put(np); return PTR_ERR(falcon_info.clk[*bank]); } falcon_info.membase[*bank] = devm_ioremap_resource(&pdev->dev, diff --git a/drivers/pinctrl/qcom/pinctrl-msm.c b/drivers/pinctrl/qcom/pinctrl-msm.c index 9a8daa256a32..1a948c3f54b7 100644 --- a/drivers/pinctrl/qcom/pinctrl-msm.c +++ b/drivers/pinctrl/qcom/pinctrl-msm.c @@ -1104,7 +1104,6 @@ static int msm_gpio_init(struct msm_pinctrl *pctrl) pctrl->irq_chip.irq_mask = msm_gpio_irq_mask; pctrl->irq_chip.irq_unmask = msm_gpio_irq_unmask; pctrl->irq_chip.irq_ack = msm_gpio_irq_ack; - pctrl->irq_chip.irq_eoi = irq_chip_eoi_parent; pctrl->irq_chip.irq_set_type = msm_gpio_irq_set_type; pctrl->irq_chip.irq_set_wake = msm_gpio_irq_set_wake; pctrl->irq_chip.irq_request_resources = msm_gpio_irq_reqres; @@ -1118,7 +1117,7 @@ static int msm_gpio_init(struct msm_pinctrl *pctrl) if (!chip->irq.parent_domain) return -EPROBE_DEFER; chip->irq.child_to_parent_hwirq = msm_gpio_wakeirq; - + pctrl->irq_chip.irq_eoi = irq_chip_eoi_parent; /* * Let's skip handling the GPIOs, if the parent irqchip * is handling the direct connect IRQ of the GPIO. diff --git a/drivers/pinctrl/qcom/pinctrl-ssbi-gpio.c b/drivers/pinctrl/qcom/pinctrl-ssbi-gpio.c index fba1d41d20ec..338a15d08629 100644 --- a/drivers/pinctrl/qcom/pinctrl-ssbi-gpio.c +++ b/drivers/pinctrl/qcom/pinctrl-ssbi-gpio.c @@ -794,7 +794,7 @@ static int pm8xxx_gpio_probe(struct platform_device *pdev) girq->fwnode = of_node_to_fwnode(pctrl->dev->of_node); girq->parent_domain = parent_domain; girq->child_to_parent_hwirq = pm8xxx_child_to_parent_hwirq; - girq->populate_parent_alloc_arg = gpiochip_populate_parent_fwspec_fourcell; + girq->populate_parent_alloc_arg = gpiochip_populate_parent_fwspec_twocell; girq->child_offset_to_irq = pm8xxx_child_offset_to_irq; girq->child_irq_domain_ops.translate = pm8xxx_domain_translate; diff --git a/drivers/platform/chrome/wilco_ec/properties.c b/drivers/platform/chrome/wilco_ec/properties.c index e69682c95ea2..62f27610dd33 100644 --- a/drivers/platform/chrome/wilco_ec/properties.c +++ b/drivers/platform/chrome/wilco_ec/properties.c @@ -5,7 +5,7 @@ #include <linux/platform_data/wilco-ec.h> #include <linux/string.h> -#include <linux/unaligned/le_memmove.h> +#include <asm/unaligned.h> /* Operation code; what the EC should do with the property */ enum ec_property_op { diff --git a/drivers/regulator/stm32-vrefbuf.c b/drivers/regulator/stm32-vrefbuf.c index bdfaf7edb75a..992bc18101ef 100644 --- a/drivers/regulator/stm32-vrefbuf.c +++ b/drivers/regulator/stm32-vrefbuf.c @@ -88,7 +88,7 @@ static int stm32_vrefbuf_disable(struct regulator_dev *rdev) } val = readl_relaxed(priv->base + STM32_VREFBUF_CSR); - val = (val & ~STM32_ENVR) | STM32_HIZ; + val &= ~STM32_ENVR; writel_relaxed(val, priv->base + STM32_VREFBUF_CSR); pm_runtime_mark_last_busy(priv->dev); @@ -175,6 +175,7 @@ static const struct regulator_desc stm32_vrefbuf_regu = { .volt_table = stm32_vrefbuf_voltages, .n_voltages = ARRAY_SIZE(stm32_vrefbuf_voltages), .ops = &stm32_vrefbuf_volt_ops, + .off_on_delay = 1000, .type = REGULATOR_VOLTAGE, .owner = THIS_MODULE, }; diff --git a/drivers/reset/Kconfig b/drivers/reset/Kconfig index 461b0e506a26..d9efbfd29646 100644 --- a/drivers/reset/Kconfig +++ b/drivers/reset/Kconfig @@ -51,6 +51,7 @@ config RESET_BRCMSTB config RESET_BRCMSTB_RESCAL bool "Broadcom STB RESCAL reset controller" + depends on HAS_IOMEM default ARCH_BRCMSTB || COMPILE_TEST help This enables the RESCAL reset controller for SATA, PCIe0, or PCIe1 on @@ -73,7 +74,7 @@ config RESET_IMX7 config RESET_INTEL_GW bool "Intel Reset Controller Driver" - depends on OF + depends on OF && HAS_IOMEM select REGMAP_MMIO help This enables the reset controller driver for Intel Gateway SoCs. diff --git a/drivers/rtc/Kconfig b/drivers/rtc/Kconfig index 34c8b6c7e095..8e503881d9d6 100644 --- a/drivers/rtc/Kconfig +++ b/drivers/rtc/Kconfig @@ -327,6 +327,7 @@ config RTC_DRV_MAX6900 config RTC_DRV_MAX8907 tristate "Maxim MAX8907" depends on MFD_MAX8907 || COMPILE_TEST + select REGMAP_IRQ help If you say yes here you will get support for the RTC of Maxim MAX8907 PMIC. diff --git a/drivers/s390/block/dasd.c b/drivers/s390/block/dasd.c index 6cca72782af6..cf87eb27879f 100644 --- a/drivers/s390/block/dasd.c +++ b/drivers/s390/block/dasd.c @@ -178,6 +178,8 @@ struct dasd_block *dasd_alloc_block(void) (unsigned long) block); INIT_LIST_HEAD(&block->ccw_queue); spin_lock_init(&block->queue_lock); + INIT_LIST_HEAD(&block->format_list); + spin_lock_init(&block->format_lock); timer_setup(&block->timer, dasd_block_timeout, 0); spin_lock_init(&block->profile.lock); @@ -1779,20 +1781,26 @@ void dasd_int_handler(struct ccw_device *cdev, unsigned long intparm, if (dasd_ese_needs_format(cqr->block, irb)) { if (rq_data_dir((struct request *)cqr->callback_data) == READ) { - device->discipline->ese_read(cqr); + device->discipline->ese_read(cqr, irb); cqr->status = DASD_CQR_SUCCESS; cqr->stopclk = now; dasd_device_clear_timer(device); dasd_schedule_device_bh(device); return; } - fcqr = device->discipline->ese_format(device, cqr); + fcqr = device->discipline->ese_format(device, cqr, irb); if (IS_ERR(fcqr)) { + if (PTR_ERR(fcqr) == -EINVAL) { + cqr->status = DASD_CQR_ERROR; + return; + } /* * If we can't format now, let the request go * one extra round. Maybe we can format later. */ cqr->status = DASD_CQR_QUEUED; + dasd_schedule_device_bh(device); + return; } else { fcqr->status = DASD_CQR_QUEUED; cqr->status = DASD_CQR_QUEUED; @@ -2748,11 +2756,13 @@ static void __dasd_cleanup_cqr(struct dasd_ccw_req *cqr) { struct request *req; blk_status_t error = BLK_STS_OK; + unsigned int proc_bytes; int status; req = (struct request *) cqr->callback_data; dasd_profile_end(cqr->block, cqr, req); + proc_bytes = cqr->proc_bytes; status = cqr->block->base->discipline->free_cp(cqr, req); if (status < 0) error = errno_to_blk_status(status); @@ -2783,7 +2793,18 @@ static void __dasd_cleanup_cqr(struct dasd_ccw_req *cqr) blk_mq_end_request(req, error); blk_mq_run_hw_queues(req->q, true); } else { - blk_mq_complete_request(req); + /* + * Partial completed requests can happen with ESE devices. + * During read we might have gotten a NRF error and have to + * complete a request partially. + */ + if (proc_bytes) { + blk_update_request(req, BLK_STS_OK, + blk_rq_bytes(req) - proc_bytes); + blk_mq_requeue_request(req, true); + } else { + blk_mq_complete_request(req); + } } } diff --git a/drivers/s390/block/dasd_eckd.c b/drivers/s390/block/dasd_eckd.c index a28b9ff82378..ad44d22e8859 100644 --- a/drivers/s390/block/dasd_eckd.c +++ b/drivers/s390/block/dasd_eckd.c @@ -207,6 +207,45 @@ static void set_ch_t(struct ch_t *geo, __u32 cyl, __u8 head) geo->head |= head; } +/* + * calculate failing track from sense data depending if + * it is an EAV device or not + */ +static int dasd_eckd_track_from_irb(struct irb *irb, struct dasd_device *device, + sector_t *track) +{ + struct dasd_eckd_private *private = device->private; + u8 *sense = NULL; + u32 cyl; + u8 head; + + sense = dasd_get_sense(irb); + if (!sense) { + DBF_DEV_EVENT(DBF_WARNING, device, "%s", + "ESE error no sense data\n"); + return -EINVAL; + } + if (!(sense[27] & DASD_SENSE_BIT_2)) { + DBF_DEV_EVENT(DBF_WARNING, device, "%s", + "ESE error no valid track data\n"); + return -EINVAL; + } + + if (sense[27] & DASD_SENSE_BIT_3) { + /* enhanced addressing */ + cyl = sense[30] << 20; + cyl |= (sense[31] & 0xF0) << 12; + cyl |= sense[28] << 8; + cyl |= sense[29]; + } else { + cyl = sense[29] << 8; + cyl |= sense[30]; + } + head = sense[31] & 0x0F; + *track = cyl * private->rdc_data.trk_per_cyl + head; + return 0; +} + static int set_timestamp(struct ccw1 *ccw, struct DE_eckd_data *data, struct dasd_device *device) { @@ -2986,6 +3025,37 @@ static int dasd_eckd_format_device(struct dasd_device *base, 0, NULL); } +static bool test_and_set_format_track(struct dasd_format_entry *to_format, + struct dasd_block *block) +{ + struct dasd_format_entry *format; + unsigned long flags; + bool rc = false; + + spin_lock_irqsave(&block->format_lock, flags); + list_for_each_entry(format, &block->format_list, list) { + if (format->track == to_format->track) { + rc = true; + goto out; + } + } + list_add_tail(&to_format->list, &block->format_list); + +out: + spin_unlock_irqrestore(&block->format_lock, flags); + return rc; +} + +static void clear_format_track(struct dasd_format_entry *format, + struct dasd_block *block) +{ + unsigned long flags; + + spin_lock_irqsave(&block->format_lock, flags); + list_del_init(&format->list); + spin_unlock_irqrestore(&block->format_lock, flags); +} + /* * Callback function to free ESE format requests. */ @@ -2993,15 +3063,19 @@ static void dasd_eckd_ese_format_cb(struct dasd_ccw_req *cqr, void *data) { struct dasd_device *device = cqr->startdev; struct dasd_eckd_private *private = device->private; + struct dasd_format_entry *format = data; + clear_format_track(format, cqr->basedev->block); private->count--; dasd_ffree_request(cqr, device); } static struct dasd_ccw_req * -dasd_eckd_ese_format(struct dasd_device *startdev, struct dasd_ccw_req *cqr) +dasd_eckd_ese_format(struct dasd_device *startdev, struct dasd_ccw_req *cqr, + struct irb *irb) { struct dasd_eckd_private *private; + struct dasd_format_entry *format; struct format_data_t fdata; unsigned int recs_per_trk; struct dasd_ccw_req *fcqr; @@ -3011,23 +3085,39 @@ dasd_eckd_ese_format(struct dasd_device *startdev, struct dasd_ccw_req *cqr) struct request *req; sector_t first_trk; sector_t last_trk; + sector_t curr_trk; int rc; req = cqr->callback_data; - base = cqr->block->base; + block = cqr->block; + base = block->base; private = base->private; - block = base->block; blksize = block->bp_block; recs_per_trk = recs_per_track(&private->rdc_data, 0, blksize); + format = &startdev->format_entry; first_trk = blk_rq_pos(req) >> block->s2b_shift; sector_div(first_trk, recs_per_trk); last_trk = (blk_rq_pos(req) + blk_rq_sectors(req) - 1) >> block->s2b_shift; sector_div(last_trk, recs_per_trk); + rc = dasd_eckd_track_from_irb(irb, base, &curr_trk); + if (rc) + return ERR_PTR(rc); - fdata.start_unit = first_trk; - fdata.stop_unit = last_trk; + if (curr_trk < first_trk || curr_trk > last_trk) { + DBF_DEV_EVENT(DBF_WARNING, startdev, + "ESE error track %llu not within range %llu - %llu\n", + curr_trk, first_trk, last_trk); + return ERR_PTR(-EINVAL); + } + format->track = curr_trk; + /* test if track is already in formatting by another thread */ + if (test_and_set_format_track(format, block)) + return ERR_PTR(-EEXIST); + + fdata.start_unit = curr_trk; + fdata.stop_unit = curr_trk; fdata.blksize = blksize; fdata.intensity = private->uses_cdl ? DASD_FMT_INT_COMPAT : 0; @@ -3044,6 +3134,7 @@ dasd_eckd_ese_format(struct dasd_device *startdev, struct dasd_ccw_req *cqr) return fcqr; fcqr->callback = dasd_eckd_ese_format_cb; + fcqr->callback_data = (void *) format; return fcqr; } @@ -3051,29 +3142,87 @@ dasd_eckd_ese_format(struct dasd_device *startdev, struct dasd_ccw_req *cqr) /* * When data is read from an unformatted area of an ESE volume, this function * returns zeroed data and thereby mimics a read of zero data. + * + * The first unformatted track is the one that got the NRF error, the address is + * encoded in the sense data. + * + * All tracks before have returned valid data and should not be touched. + * All tracks after the unformatted track might be formatted or not. This is + * currently not known, remember the processed data and return the remainder of + * the request to the blocklayer in __dasd_cleanup_cqr(). */ -static void dasd_eckd_ese_read(struct dasd_ccw_req *cqr) +static int dasd_eckd_ese_read(struct dasd_ccw_req *cqr, struct irb *irb) { + struct dasd_eckd_private *private; + sector_t first_trk, last_trk; + sector_t first_blk, last_blk; unsigned int blksize, off; + unsigned int recs_per_trk; struct dasd_device *base; struct req_iterator iter; + struct dasd_block *block; + unsigned int skip_block; + unsigned int blk_count; struct request *req; struct bio_vec bv; + sector_t curr_trk; + sector_t end_blk; char *dst; + int rc; req = (struct request *) cqr->callback_data; base = cqr->block->base; blksize = base->block->bp_block; + block = cqr->block; + private = base->private; + skip_block = 0; + blk_count = 0; + + recs_per_trk = recs_per_track(&private->rdc_data, 0, blksize); + first_trk = first_blk = blk_rq_pos(req) >> block->s2b_shift; + sector_div(first_trk, recs_per_trk); + last_trk = last_blk = + (blk_rq_pos(req) + blk_rq_sectors(req) - 1) >> block->s2b_shift; + sector_div(last_trk, recs_per_trk); + rc = dasd_eckd_track_from_irb(irb, base, &curr_trk); + if (rc) + return rc; + + /* sanity check if the current track from sense data is valid */ + if (curr_trk < first_trk || curr_trk > last_trk) { + DBF_DEV_EVENT(DBF_WARNING, base, + "ESE error track %llu not within range %llu - %llu\n", + curr_trk, first_trk, last_trk); + return -EINVAL; + } + + /* + * if not the first track got the NRF error we have to skip over valid + * blocks + */ + if (curr_trk != first_trk) + skip_block = curr_trk * recs_per_trk - first_blk; + + /* we have no information beyond the current track */ + end_blk = (curr_trk + 1) * recs_per_trk; rq_for_each_segment(bv, req, iter) { dst = page_address(bv.bv_page) + bv.bv_offset; for (off = 0; off < bv.bv_len; off += blksize) { - if (dst && rq_data_dir(req) == READ) { + if (first_blk + blk_count >= end_blk) { + cqr->proc_bytes = blk_count * blksize; + return 0; + } + if (dst && !skip_block) { dst += off; memset(dst, 0, blksize); + } else { + skip_block--; } + blk_count++; } } + return 0; } /* diff --git a/drivers/s390/block/dasd_int.h b/drivers/s390/block/dasd_int.h index 91c9f9586e0f..fa552f9f1666 100644 --- a/drivers/s390/block/dasd_int.h +++ b/drivers/s390/block/dasd_int.h @@ -187,6 +187,7 @@ struct dasd_ccw_req { void (*callback)(struct dasd_ccw_req *, void *data); void *callback_data; + unsigned int proc_bytes; /* bytes for partial completion */ }; /* @@ -387,8 +388,9 @@ struct dasd_discipline { int (*ext_pool_warn_thrshld)(struct dasd_device *); int (*ext_pool_oos)(struct dasd_device *); int (*ext_pool_exhaust)(struct dasd_device *, struct dasd_ccw_req *); - struct dasd_ccw_req *(*ese_format)(struct dasd_device *, struct dasd_ccw_req *); - void (*ese_read)(struct dasd_ccw_req *); + struct dasd_ccw_req *(*ese_format)(struct dasd_device *, + struct dasd_ccw_req *, struct irb *); + int (*ese_read)(struct dasd_ccw_req *, struct irb *); }; extern struct dasd_discipline *dasd_diag_discipline_pointer; @@ -474,6 +476,11 @@ struct dasd_profile { spinlock_t lock; }; +struct dasd_format_entry { + struct list_head list; + sector_t track; +}; + struct dasd_device { /* Block device stuff. */ struct dasd_block *block; @@ -539,6 +546,7 @@ struct dasd_device { struct dentry *debugfs_dentry; struct dentry *hosts_dentry; struct dasd_profile profile; + struct dasd_format_entry format_entry; }; struct dasd_block { @@ -564,6 +572,9 @@ struct dasd_block { struct dentry *debugfs_dentry; struct dasd_profile profile; + + struct list_head format_list; + spinlock_t format_lock; }; struct dasd_attention_data { diff --git a/drivers/s390/cio/blacklist.c b/drivers/s390/cio/blacklist.c index da642e811f7f..4dd2eb634856 100644 --- a/drivers/s390/cio/blacklist.c +++ b/drivers/s390/cio/blacklist.c @@ -303,8 +303,10 @@ static void * cio_ignore_proc_seq_next(struct seq_file *s, void *it, loff_t *offset) { struct ccwdev_iter *iter; + loff_t p = *offset; - if (*offset >= (__MAX_SUBCHANNEL + 1) * (__MAX_SSID + 1)) + (*offset)++; + if (p >= (__MAX_SUBCHANNEL + 1) * (__MAX_SSID + 1)) return NULL; iter = it; if (iter->devno == __MAX_SUBCHANNEL) { @@ -314,7 +316,6 @@ cio_ignore_proc_seq_next(struct seq_file *s, void *it, loff_t *offset) return NULL; } else iter->devno++; - (*offset)++; return iter; } diff --git a/drivers/s390/cio/chp.c b/drivers/s390/cio/chp.c index 51038ec309c1..dfcbe54591fb 100644 --- a/drivers/s390/cio/chp.c +++ b/drivers/s390/cio/chp.c @@ -135,7 +135,7 @@ static ssize_t chp_measurement_chars_read(struct file *filp, struct channel_path *chp; struct device *device; - device = container_of(kobj, struct device, kobj); + device = kobj_to_dev(kobj); chp = to_channelpath(device); if (chp->cmg == -1) return 0; @@ -184,7 +184,7 @@ static ssize_t chp_measurement_read(struct file *filp, struct kobject *kobj, struct device *device; unsigned int size; - device = container_of(kobj, struct device, kobj); + device = kobj_to_dev(kobj); chp = to_channelpath(device); css = to_css(chp->dev.parent); diff --git a/drivers/s390/cio/qdio_setup.c b/drivers/s390/cio/qdio_setup.c index 3ab8e80d7bbc..e115623b86b2 100644 --- a/drivers/s390/cio/qdio_setup.c +++ b/drivers/s390/cio/qdio_setup.c @@ -8,6 +8,7 @@ #include <linux/kernel.h> #include <linux/slab.h> #include <linux/export.h> +#include <linux/io.h> #include <asm/qdio.h> #include "cio.h" @@ -205,7 +206,7 @@ static void setup_storage_lists(struct qdio_q *q, struct qdio_irq *irq_ptr, /* fill in sl */ for (j = 0; j < QDIO_MAX_BUFFERS_PER_Q; j++) - q->sl->element[j].sbal = (unsigned long)q->sbal[j]; + q->sl->element[j].sbal = virt_to_phys(q->sbal[j]); } static void setup_queues(struct qdio_irq *irq_ptr, diff --git a/drivers/s390/crypto/zcrypt_ep11misc.c b/drivers/s390/crypto/zcrypt_ep11misc.c index d4caf46ff9df..2afe2153b34e 100644 --- a/drivers/s390/crypto/zcrypt_ep11misc.c +++ b/drivers/s390/crypto/zcrypt_ep11misc.c @@ -887,7 +887,7 @@ static int ep11_unwrapkey(u16 card, u16 domain, /* empty pin tag */ *p++ = 0x04; *p++ = 0; - /* encrytped key value tag and bytes */ + /* encrypted key value tag and bytes */ p += asn1tag_write(p, 0x04, enckey, enckeysize); /* reply cprb and payload */ @@ -1095,7 +1095,7 @@ int ep11_clr2keyblob(u16 card, u16 domain, u32 keybitsize, u32 keygenflags, /* Step 1: generate AES 256 bit random kek key */ rc = ep11_genaeskey(card, domain, 256, - 0x00006c00, /* EN/DECRYTP, WRAP/UNWRAP */ + 0x00006c00, /* EN/DECRYPT, WRAP/UNWRAP */ kek, &keklen); if (rc) { DEBUG_ERR( diff --git a/drivers/s390/net/qeth_core.h b/drivers/s390/net/qeth_core.h index 9575a627a1e1..468cada49e72 100644 --- a/drivers/s390/net/qeth_core.h +++ b/drivers/s390/net/qeth_core.h @@ -369,7 +369,7 @@ enum qeth_qdio_info_states { struct qeth_buffer_pool_entry { struct list_head list; struct list_head init_list; - void *elements[QDIO_MAX_ELEMENTS_PER_BUFFER]; + struct page *elements[QDIO_MAX_ELEMENTS_PER_BUFFER]; }; struct qeth_qdio_buffer_pool { @@ -983,7 +983,7 @@ extern const struct attribute_group qeth_device_blkt_group; extern const struct device_type qeth_generic_devtype; const char *qeth_get_cardname_short(struct qeth_card *); -int qeth_realloc_buffer_pool(struct qeth_card *, int); +int qeth_resize_buffer_pool(struct qeth_card *card, unsigned int count); int qeth_core_load_discipline(struct qeth_card *, enum qeth_discipline_id); void qeth_core_free_discipline(struct qeth_card *); diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c index 9639938581f5..6d3f2f14b414 100644 --- a/drivers/s390/net/qeth_core_main.c +++ b/drivers/s390/net/qeth_core_main.c @@ -65,7 +65,6 @@ static struct lock_class_key qdio_out_skb_queue_key; static void qeth_issue_next_read_cb(struct qeth_card *card, struct qeth_cmd_buffer *iob, unsigned int data_length); -static void qeth_free_buffer_pool(struct qeth_card *); static int qeth_qdio_establish(struct qeth_card *); static void qeth_free_qdio_queues(struct qeth_card *card); static void qeth_notify_skbs(struct qeth_qdio_out_q *queue, @@ -212,49 +211,121 @@ void qeth_clear_working_pool_list(struct qeth_card *card) } EXPORT_SYMBOL_GPL(qeth_clear_working_pool_list); +static void qeth_free_pool_entry(struct qeth_buffer_pool_entry *entry) +{ + unsigned int i; + + for (i = 0; i < ARRAY_SIZE(entry->elements); i++) { + if (entry->elements[i]) + __free_page(entry->elements[i]); + } + + kfree(entry); +} + +static void qeth_free_buffer_pool(struct qeth_card *card) +{ + struct qeth_buffer_pool_entry *entry, *tmp; + + list_for_each_entry_safe(entry, tmp, &card->qdio.init_pool.entry_list, + init_list) { + list_del(&entry->init_list); + qeth_free_pool_entry(entry); + } +} + +static struct qeth_buffer_pool_entry *qeth_alloc_pool_entry(unsigned int pages) +{ + struct qeth_buffer_pool_entry *entry; + unsigned int i; + + entry = kzalloc(sizeof(*entry), GFP_KERNEL); + if (!entry) + return NULL; + + for (i = 0; i < pages; i++) { + entry->elements[i] = alloc_page(GFP_KERNEL); + + if (!entry->elements[i]) { + qeth_free_pool_entry(entry); + return NULL; + } + } + + return entry; +} + static int qeth_alloc_buffer_pool(struct qeth_card *card) { - struct qeth_buffer_pool_entry *pool_entry; - void *ptr; - int i, j; + unsigned int buf_elements = QETH_MAX_BUFFER_ELEMENTS(card); + unsigned int i; QETH_CARD_TEXT(card, 5, "alocpool"); for (i = 0; i < card->qdio.init_pool.buf_count; ++i) { - pool_entry = kzalloc(sizeof(*pool_entry), GFP_KERNEL); - if (!pool_entry) { + struct qeth_buffer_pool_entry *entry; + + entry = qeth_alloc_pool_entry(buf_elements); + if (!entry) { qeth_free_buffer_pool(card); return -ENOMEM; } - for (j = 0; j < QETH_MAX_BUFFER_ELEMENTS(card); ++j) { - ptr = (void *) __get_free_page(GFP_KERNEL); - if (!ptr) { - while (j > 0) - free_page((unsigned long) - pool_entry->elements[--j]); - kfree(pool_entry); - qeth_free_buffer_pool(card); - return -ENOMEM; - } - pool_entry->elements[j] = ptr; - } - list_add(&pool_entry->init_list, - &card->qdio.init_pool.entry_list); + + list_add(&entry->init_list, &card->qdio.init_pool.entry_list); } return 0; } -int qeth_realloc_buffer_pool(struct qeth_card *card, int bufcnt) +int qeth_resize_buffer_pool(struct qeth_card *card, unsigned int count) { + unsigned int buf_elements = QETH_MAX_BUFFER_ELEMENTS(card); + struct qeth_qdio_buffer_pool *pool = &card->qdio.init_pool; + struct qeth_buffer_pool_entry *entry, *tmp; + int delta = count - pool->buf_count; + LIST_HEAD(entries); + QETH_CARD_TEXT(card, 2, "realcbp"); - /* TODO: steel/add buffers from/to a running card's buffer pool (?) */ - qeth_clear_working_pool_list(card); - qeth_free_buffer_pool(card); - card->qdio.in_buf_pool.buf_count = bufcnt; - card->qdio.init_pool.buf_count = bufcnt; - return qeth_alloc_buffer_pool(card); + /* Defer until queue is allocated: */ + if (!card->qdio.in_q) + goto out; + + /* Remove entries from the pool: */ + while (delta < 0) { + entry = list_first_entry(&pool->entry_list, + struct qeth_buffer_pool_entry, + init_list); + list_del(&entry->init_list); + qeth_free_pool_entry(entry); + + delta++; + } + + /* Allocate additional entries: */ + while (delta > 0) { + entry = qeth_alloc_pool_entry(buf_elements); + if (!entry) { + list_for_each_entry_safe(entry, tmp, &entries, + init_list) { + list_del(&entry->init_list); + qeth_free_pool_entry(entry); + } + + return -ENOMEM; + } + + list_add(&entry->init_list, &entries); + + delta--; + } + + list_splice(&entries, &pool->entry_list); + +out: + card->qdio.in_buf_pool.buf_count = count; + pool->buf_count = count; + return 0; } -EXPORT_SYMBOL_GPL(qeth_realloc_buffer_pool); +EXPORT_SYMBOL_GPL(qeth_resize_buffer_pool); static void qeth_free_qdio_queue(struct qeth_qdio_q *q) { @@ -1128,9 +1199,10 @@ static void qeth_clear_output_buffer(struct qeth_qdio_out_q *queue, qeth_tx_complete_buf(buf, error, budget); for (i = 0; i < queue->max_elements; ++i) { - if (buf->buffer->element[i].addr && buf->is_header[i]) - kmem_cache_free(qeth_core_header_cache, - buf->buffer->element[i].addr); + void *data = phys_to_virt(buf->buffer->element[i].addr); + + if (data && buf->is_header[i]) + kmem_cache_free(qeth_core_header_cache, data); buf->is_header[i] = 0; } @@ -1169,19 +1241,6 @@ void qeth_drain_output_queues(struct qeth_card *card) } EXPORT_SYMBOL_GPL(qeth_drain_output_queues); -static void qeth_free_buffer_pool(struct qeth_card *card) -{ - struct qeth_buffer_pool_entry *pool_entry, *tmp; - int i = 0; - list_for_each_entry_safe(pool_entry, tmp, - &card->qdio.init_pool.entry_list, init_list){ - for (i = 0; i < QETH_MAX_BUFFER_ELEMENTS(card); ++i) - free_page((unsigned long)pool_entry->elements[i]); - list_del(&pool_entry->init_list); - kfree(pool_entry); - } -} - static int qeth_osa_set_output_queues(struct qeth_card *card, bool single) { unsigned int count = single ? 1 : card->dev->num_tx_queues; @@ -1203,7 +1262,6 @@ static int qeth_osa_set_output_queues(struct qeth_card *card, bool single) if (count == 1) dev_info(&card->gdev->dev, "Priority Queueing not supported\n"); - card->qdio.default_out_queue = single ? 0 : QETH_DEFAULT_QUEUE; card->qdio.no_out_queues = count; return 0; } @@ -2392,7 +2450,6 @@ static void qeth_free_qdio_queues(struct qeth_card *card) return; qeth_free_cq(card); - cancel_delayed_work_sync(&card->buffer_reclaim_work); for (j = 0; j < QDIO_MAX_BUFFERS_PER_Q; ++j) { if (card->qdio.in_q->bufs[j].rx_skb) dev_kfree_skb_any(card->qdio.in_q->bufs[j].rx_skb); @@ -2574,7 +2631,6 @@ static struct qeth_buffer_pool_entry *qeth_find_free_buffer_pool_entry( struct list_head *plh; struct qeth_buffer_pool_entry *entry; int i, free; - struct page *page; if (list_empty(&card->qdio.in_buf_pool.entry_list)) return NULL; @@ -2583,7 +2639,7 @@ static struct qeth_buffer_pool_entry *qeth_find_free_buffer_pool_entry( entry = list_entry(plh, struct qeth_buffer_pool_entry, list); free = 1; for (i = 0; i < QETH_MAX_BUFFER_ELEMENTS(card); ++i) { - if (page_count(virt_to_page(entry->elements[i])) > 1) { + if (page_count(entry->elements[i]) > 1) { free = 0; break; } @@ -2598,15 +2654,15 @@ static struct qeth_buffer_pool_entry *qeth_find_free_buffer_pool_entry( entry = list_entry(card->qdio.in_buf_pool.entry_list.next, struct qeth_buffer_pool_entry, list); for (i = 0; i < QETH_MAX_BUFFER_ELEMENTS(card); ++i) { - if (page_count(virt_to_page(entry->elements[i])) > 1) { - page = alloc_page(GFP_ATOMIC); - if (!page) { + if (page_count(entry->elements[i]) > 1) { + struct page *page = alloc_page(GFP_ATOMIC); + + if (!page) return NULL; - } else { - free_page((unsigned long)entry->elements[i]); - entry->elements[i] = page_address(page); - QETH_CARD_STAT_INC(card, rx_sg_alloc_page); - } + + __free_page(entry->elements[i]); + entry->elements[i] = page; + QETH_CARD_STAT_INC(card, rx_sg_alloc_page); } } list_del_init(&entry->list); @@ -2624,12 +2680,12 @@ static int qeth_init_input_buffer(struct qeth_card *card, ETH_HLEN + sizeof(struct ipv6hdr)); if (!buf->rx_skb) - return 1; + return -ENOMEM; } pool_entry = qeth_find_free_buffer_pool_entry(card); if (!pool_entry) - return 1; + return -ENOBUFS; /* * since the buffer is accessed only from the input_tasklet @@ -2641,7 +2697,8 @@ static int qeth_init_input_buffer(struct qeth_card *card, buf->pool_entry = pool_entry; for (i = 0; i < QETH_MAX_BUFFER_ELEMENTS(card); ++i) { buf->buffer->element[i].length = PAGE_SIZE; - buf->buffer->element[i].addr = pool_entry->elements[i]; + buf->buffer->element[i].addr = + page_to_phys(pool_entry->elements[i]); if (i == QETH_MAX_BUFFER_ELEMENTS(card) - 1) buf->buffer->element[i].eflags = SBAL_EFLAGS_LAST_ENTRY; else @@ -2673,10 +2730,15 @@ static int qeth_init_qdio_queues(struct qeth_card *card) /* inbound queue */ qdio_reset_buffers(card->qdio.in_q->qdio_bufs, QDIO_MAX_BUFFERS_PER_Q); memset(&card->rx, 0, sizeof(struct qeth_rx)); + qeth_initialize_working_pool_list(card); /*give only as many buffers to hardware as we have buffer pool entries*/ - for (i = 0; i < card->qdio.in_buf_pool.buf_count - 1; ++i) - qeth_init_input_buffer(card, &card->qdio.in_q->bufs[i]); + for (i = 0; i < card->qdio.in_buf_pool.buf_count - 1; i++) { + rc = qeth_init_input_buffer(card, &card->qdio.in_q->bufs[i]); + if (rc) + return rc; + } + card->qdio.in_q->next_buf_to_init = card->qdio.in_buf_pool.buf_count - 1; rc = do_QDIO(CARD_DDEV(card), QDIO_FLAG_SYNC_INPUT, 0, 0, @@ -3459,9 +3521,8 @@ static void qeth_qdio_cq_handler(struct qeth_card *card, unsigned int qdio_err, while ((e < QDIO_MAX_ELEMENTS_PER_BUFFER) && buffer->element[e].addr) { - unsigned long phys_aob_addr; + unsigned long phys_aob_addr = buffer->element[e].addr; - phys_aob_addr = (unsigned long) buffer->element[e].addr; qeth_qdio_handle_aob(card, phys_aob_addr); ++e; } @@ -3750,7 +3811,7 @@ static unsigned int __qeth_fill_buffer(struct sk_buff *skb, elem_length = min_t(unsigned int, length, PAGE_SIZE - offset_in_page(data)); - buffer->element[element].addr = data; + buffer->element[element].addr = virt_to_phys(data); buffer->element[element].length = elem_length; length -= elem_length; if (is_first_elem) { @@ -3780,7 +3841,7 @@ static unsigned int __qeth_fill_buffer(struct sk_buff *skb, elem_length = min_t(unsigned int, length, PAGE_SIZE - offset_in_page(data)); - buffer->element[element].addr = data; + buffer->element[element].addr = virt_to_phys(data); buffer->element[element].length = elem_length; buffer->element[element].eflags = SBAL_EFLAGS_MIDDLE_FRAG; @@ -3820,7 +3881,7 @@ static unsigned int qeth_fill_buffer(struct qeth_qdio_out_buffer *buf, int element = buf->next_element_to_fill; is_first_elem = false; - buffer->element[element].addr = hdr; + buffer->element[element].addr = virt_to_phys(hdr); buffer->element[element].length = hd_len; buffer->element[element].eflags = SBAL_EFLAGS_FIRST_FRAG; /* remember to free cache-allocated qeth_hdr: */ @@ -4746,10 +4807,10 @@ static void qeth_qdio_establish_cq(struct qeth_card *card, if (card->options.cq == QETH_CQ_ENABLED) { int offset = QDIO_MAX_BUFFERS_PER_Q * (card->qdio.no_in_queues - 1); - for (i = 0; i < QDIO_MAX_BUFFERS_PER_Q; ++i) { - in_sbal_ptrs[offset + i] = (struct qdio_buffer *) - virt_to_phys(card->qdio.c_q->bufs[i].buffer); - } + + for (i = 0; i < QDIO_MAX_BUFFERS_PER_Q; i++) + in_sbal_ptrs[offset + i] = + card->qdio.c_q->bufs[i].buffer; queue_start_poll[card->qdio.no_in_queues - 1] = NULL; } @@ -4783,10 +4844,9 @@ static int qeth_qdio_establish(struct qeth_card *card) rc = -ENOMEM; goto out_free_qib_param; } - for (i = 0; i < QDIO_MAX_BUFFERS_PER_Q; ++i) { - in_sbal_ptrs[i] = (struct qdio_buffer *) - virt_to_phys(card->qdio.in_q->bufs[i].buffer); - } + + for (i = 0; i < QDIO_MAX_BUFFERS_PER_Q; i++) + in_sbal_ptrs[i] = card->qdio.in_q->bufs[i].buffer; queue_start_poll = kcalloc(card->qdio.no_in_queues, sizeof(void *), GFP_KERNEL); @@ -4807,11 +4867,11 @@ static int qeth_qdio_establish(struct qeth_card *card) rc = -ENOMEM; goto out_free_queue_start_poll; } + for (i = 0, k = 0; i < card->qdio.no_out_queues; ++i) - for (j = 0; j < QDIO_MAX_BUFFERS_PER_Q; ++j, ++k) { - out_sbal_ptrs[k] = (struct qdio_buffer *)virt_to_phys( - card->qdio.out_qs[i]->bufs[j]->buffer); - } + for (j = 0; j < QDIO_MAX_BUFFERS_PER_Q; j++, k++) + out_sbal_ptrs[k] = + card->qdio.out_qs[i]->bufs[j]->buffer; memset(&init_data, 0, sizeof(struct qdio_initialize)); init_data.cdev = CARD_DDEV(card); @@ -5289,7 +5349,7 @@ next_packet: offset = 0; } - hdr = element->addr + offset; + hdr = phys_to_virt(element->addr) + offset; offset += sizeof(*hdr); skb = NULL; @@ -5344,7 +5404,7 @@ next_packet: } use_rx_sg = (card->options.cq == QETH_CQ_ENABLED) || - ((skb_len >= card->options.rx_sg_cb) && + (skb_len > card->options.rx_sg_cb && !atomic_read(&card->force_alloc_skb) && !IS_OSN(card)); @@ -5388,7 +5448,7 @@ use_skb: walk_packet: while (skb_len) { int data_len = min(skb_len, (int)(element->length - offset)); - char *data = element->addr + offset; + char *data = phys_to_virt(element->addr) + offset; skb_len -= data_len; offset += data_len; @@ -5447,7 +5507,6 @@ static int qeth_extract_skbs(struct qeth_card *card, int budget, { int work_done = 0; - WARN_ON_ONCE(!budget); *done = false; while (budget) { diff --git a/drivers/s390/net/qeth_core_sys.c b/drivers/s390/net/qeth_core_sys.c index 2bd9993aa60b..78cae61bc924 100644 --- a/drivers/s390/net/qeth_core_sys.c +++ b/drivers/s390/net/qeth_core_sys.c @@ -247,8 +247,8 @@ static ssize_t qeth_dev_bufcnt_store(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { struct qeth_card *card = dev_get_drvdata(dev); + unsigned int cnt; char *tmp; - int cnt, old_cnt; int rc = 0; mutex_lock(&card->conf_mutex); @@ -257,13 +257,12 @@ static ssize_t qeth_dev_bufcnt_store(struct device *dev, goto out; } - old_cnt = card->qdio.in_buf_pool.buf_count; cnt = simple_strtoul(buf, &tmp, 10); cnt = (cnt < QETH_IN_BUF_COUNT_MIN) ? QETH_IN_BUF_COUNT_MIN : ((cnt > QETH_IN_BUF_COUNT_MAX) ? QETH_IN_BUF_COUNT_MAX : cnt); - if (old_cnt != cnt) { - rc = qeth_realloc_buffer_pool(card, cnt); - } + + rc = qeth_resize_buffer_pool(card, cnt); + out: mutex_unlock(&card->conf_mutex); return rc ? rc : count; diff --git a/drivers/s390/net/qeth_l2_main.c b/drivers/s390/net/qeth_l2_main.c index 692bd2623401..8fb29371788b 100644 --- a/drivers/s390/net/qeth_l2_main.c +++ b/drivers/s390/net/qeth_l2_main.c @@ -284,6 +284,7 @@ static void qeth_l2_stop_card(struct qeth_card *card) if (card->state == CARD_STATE_SOFTSETUP) { qeth_clear_ipacmd_list(card); qeth_drain_output_queues(card); + cancel_delayed_work_sync(&card->buffer_reclaim_work); card->state = CARD_STATE_DOWN; } @@ -1707,15 +1708,14 @@ int qeth_l2_vnicc_set_state(struct qeth_card *card, u32 vnicc, bool state) QETH_CARD_TEXT(card, 2, "vniccsch"); - /* do not change anything if BridgePort is enabled */ - if (qeth_bridgeport_is_in_use(card)) - return -EBUSY; - /* check if characteristic and enable/disable are supported */ if (!(card->options.vnicc.sup_chars & vnicc) || !(card->options.vnicc.set_char_sup & vnicc)) return -EOPNOTSUPP; + if (qeth_bridgeport_is_in_use(card)) + return -EBUSY; + /* set enable/disable command and store wanted characteristic */ if (state) { cmd = IPA_VNICC_ENABLE; @@ -1761,14 +1761,13 @@ int qeth_l2_vnicc_get_state(struct qeth_card *card, u32 vnicc, bool *state) QETH_CARD_TEXT(card, 2, "vniccgch"); - /* do not get anything if BridgePort is enabled */ - if (qeth_bridgeport_is_in_use(card)) - return -EBUSY; - /* check if characteristic is supported */ if (!(card->options.vnicc.sup_chars & vnicc)) return -EOPNOTSUPP; + if (qeth_bridgeport_is_in_use(card)) + return -EBUSY; + /* if card is ready, query current VNICC state */ if (qeth_card_hw_is_reachable(card)) rc = qeth_l2_vnicc_query_chars(card); @@ -1786,15 +1785,14 @@ int qeth_l2_vnicc_set_timeout(struct qeth_card *card, u32 timeout) QETH_CARD_TEXT(card, 2, "vniccsto"); - /* do not change anything if BridgePort is enabled */ - if (qeth_bridgeport_is_in_use(card)) - return -EBUSY; - /* check if characteristic and set_timeout are supported */ if (!(card->options.vnicc.sup_chars & QETH_VNICC_LEARNING) || !(card->options.vnicc.getset_timeout_sup & QETH_VNICC_LEARNING)) return -EOPNOTSUPP; + if (qeth_bridgeport_is_in_use(card)) + return -EBUSY; + /* do we need to do anything? */ if (card->options.vnicc.learning_timeout == timeout) return rc; @@ -1823,14 +1821,14 @@ int qeth_l2_vnicc_get_timeout(struct qeth_card *card, u32 *timeout) QETH_CARD_TEXT(card, 2, "vniccgto"); - /* do not get anything if BridgePort is enabled */ - if (qeth_bridgeport_is_in_use(card)) - return -EBUSY; - /* check if characteristic and get_timeout are supported */ if (!(card->options.vnicc.sup_chars & QETH_VNICC_LEARNING) || !(card->options.vnicc.getset_timeout_sup & QETH_VNICC_LEARNING)) return -EOPNOTSUPP; + + if (qeth_bridgeport_is_in_use(card)) + return -EBUSY; + /* if card is ready, get timeout. Otherwise, just return stored value */ *timeout = card->options.vnicc.learning_timeout; if (qeth_card_hw_is_reachable(card)) diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c index 317d56647a4a..82f800d1d7b3 100644 --- a/drivers/s390/net/qeth_l3_main.c +++ b/drivers/s390/net/qeth_l3_main.c @@ -1178,6 +1178,7 @@ static void qeth_l3_stop_card(struct qeth_card *card) qeth_l3_clear_ip_htable(card, 1); qeth_clear_ipacmd_list(card); qeth_drain_output_queues(card); + cancel_delayed_work_sync(&card->buffer_reclaim_work); card->state = CARD_STATE_DOWN; } diff --git a/drivers/s390/net/qeth_l3_sys.c b/drivers/s390/net/qeth_l3_sys.c index 29f2517d2a31..a3d1c3bdfadb 100644 --- a/drivers/s390/net/qeth_l3_sys.c +++ b/drivers/s390/net/qeth_l3_sys.c @@ -206,12 +206,11 @@ static ssize_t qeth_l3_dev_sniffer_store(struct device *dev, qdio_get_ssqd_desc(CARD_DDEV(card), &card->ssqd); if (card->ssqd.qdioac2 & CHSC_AC2_SNIFFER_AVAILABLE) { card->options.sniffer = i; - if (card->qdio.init_pool.buf_count != - QETH_IN_BUF_COUNT_MAX) - qeth_realloc_buffer_pool(card, - QETH_IN_BUF_COUNT_MAX); - } else + qeth_resize_buffer_pool(card, QETH_IN_BUF_COUNT_MAX); + } else { rc = -EPERM; + } + break; default: rc = -EINVAL; diff --git a/drivers/s390/scsi/zfcp_fsf.c b/drivers/s390/scsi/zfcp_fsf.c index 223a805f0b0b..cae9b7ff79b0 100644 --- a/drivers/s390/scsi/zfcp_fsf.c +++ b/drivers/s390/scsi/zfcp_fsf.c @@ -2510,7 +2510,7 @@ void zfcp_fsf_reqid_check(struct zfcp_qdio *qdio, int sbal_idx) for (idx = 0; idx < QDIO_MAX_ELEMENTS_PER_BUFFER; idx++) { sbale = &sbal->element[idx]; - req_id = (unsigned long) sbale->addr; + req_id = sbale->addr; fsf_req = zfcp_reqlist_find_rm(adapter->req_list, req_id); if (!fsf_req) { diff --git a/drivers/s390/scsi/zfcp_fsf.h b/drivers/s390/scsi/zfcp_fsf.h index 2b1e4da1944f..4bfb79f20588 100644 --- a/drivers/s390/scsi/zfcp_fsf.h +++ b/drivers/s390/scsi/zfcp_fsf.h @@ -410,7 +410,7 @@ struct fsf_qtcb_bottom_port { u8 cb_util; u8 a_util; u8 res2; - u16 temperature; + s16 temperature; u16 vcc; u16 tx_bias; u16 tx_power; diff --git a/drivers/s390/scsi/zfcp_qdio.c b/drivers/s390/scsi/zfcp_qdio.c index 661436a92f8e..f0d6296e673b 100644 --- a/drivers/s390/scsi/zfcp_qdio.c +++ b/drivers/s390/scsi/zfcp_qdio.c @@ -98,7 +98,7 @@ static void zfcp_qdio_int_resp(struct ccw_device *cdev, unsigned int qdio_err, memset(pl, 0, ZFCP_QDIO_MAX_SBALS_PER_REQ * sizeof(void *)); sbale = qdio->res_q[idx]->element; - req_id = (u64) sbale->addr; + req_id = sbale->addr; scount = min(sbale->scount + 1, ZFCP_QDIO_MAX_SBALS_PER_REQ + 1); /* incl. signaling SBAL */ @@ -199,7 +199,7 @@ int zfcp_qdio_sbals_from_sg(struct zfcp_qdio *qdio, struct zfcp_qdio_req *q_req, q_req->sbal_number); return -EINVAL; } - sbale->addr = sg_virt(sg); + sbale->addr = sg_phys(sg); sbale->length = sg->length; } return 0; @@ -418,7 +418,7 @@ int zfcp_qdio_open(struct zfcp_qdio *qdio) sbale->length = 0; sbale->eflags = SBAL_EFLAGS_LAST_ENTRY; sbale->sflags = 0; - sbale->addr = NULL; + sbale->addr = 0; } if (do_QDIO(cdev, QDIO_FLAG_SYNC_INPUT, 0, 0, QDIO_MAX_BUFFERS_PER_Q)) diff --git a/drivers/s390/scsi/zfcp_qdio.h b/drivers/s390/scsi/zfcp_qdio.h index 2a816a37b3c0..6b43d6b254be 100644 --- a/drivers/s390/scsi/zfcp_qdio.h +++ b/drivers/s390/scsi/zfcp_qdio.h @@ -122,14 +122,14 @@ void zfcp_qdio_req_init(struct zfcp_qdio *qdio, struct zfcp_qdio_req *q_req, % QDIO_MAX_BUFFERS_PER_Q; sbale = zfcp_qdio_sbale_req(qdio, q_req); - sbale->addr = (void *) req_id; + sbale->addr = req_id; sbale->eflags = 0; sbale->sflags = SBAL_SFLAGS0_COMMAND | sbtype; if (unlikely(!data)) return; sbale++; - sbale->addr = data; + sbale->addr = virt_to_phys(data); sbale->length = len; } @@ -152,7 +152,7 @@ void zfcp_qdio_fill_next(struct zfcp_qdio *qdio, struct zfcp_qdio_req *q_req, BUG_ON(q_req->sbale_curr == qdio->max_sbale_per_sbal - 1); q_req->sbale_curr++; sbale = zfcp_qdio_sbale_curr(qdio, q_req); - sbale->addr = data; + sbale->addr = virt_to_phys(data); sbale->length = len; } diff --git a/drivers/s390/scsi/zfcp_sysfs.c b/drivers/s390/scsi/zfcp_sysfs.c index 494b9fe9cc94..a711a0d15100 100644 --- a/drivers/s390/scsi/zfcp_sysfs.c +++ b/drivers/s390/scsi/zfcp_sysfs.c @@ -800,7 +800,7 @@ static ZFCP_DEV_ATTR(adapter_diag, b2b_credit, 0400, static ZFCP_DEV_ATTR(adapter_diag_sfp, _name, 0400, \ zfcp_sysfs_adapter_diag_sfp_##_name##_show, NULL) -ZFCP_DEFINE_DIAG_SFP_ATTR(temperature, temperature, 5, "%hu"); +ZFCP_DEFINE_DIAG_SFP_ATTR(temperature, temperature, 6, "%hd"); ZFCP_DEFINE_DIAG_SFP_ATTR(vcc, vcc, 5, "%hu"); ZFCP_DEFINE_DIAG_SFP_ATTR(tx_bias, tx_bias, 5, "%hu"); ZFCP_DEFINE_DIAG_SFP_ATTR(tx_power, tx_power, 5, "%hu"); diff --git a/drivers/scsi/ipr.c b/drivers/scsi/ipr.c index ae45cbe98ae2..cd8db1349871 100644 --- a/drivers/scsi/ipr.c +++ b/drivers/scsi/ipr.c @@ -9950,6 +9950,7 @@ static void ipr_init_ioa_cfg(struct ipr_ioa_cfg *ioa_cfg, ioa_cfg->max_devs_supported = ipr_max_devs; if (ioa_cfg->sis64) { + host->max_channel = IPR_MAX_SIS64_BUSES; host->max_id = IPR_MAX_SIS64_TARGETS_PER_BUS; host->max_lun = IPR_MAX_SIS64_LUNS_PER_TARGET; if (ipr_max_devs > IPR_MAX_SIS64_DEVS) @@ -9958,6 +9959,7 @@ static void ipr_init_ioa_cfg(struct ipr_ioa_cfg *ioa_cfg, + ((sizeof(struct ipr_config_table_entry64) * ioa_cfg->max_devs_supported))); } else { + host->max_channel = IPR_VSET_BUS; host->max_id = IPR_MAX_NUM_TARGETS_PER_BUS; host->max_lun = IPR_MAX_NUM_LUNS_PER_TARGET; if (ipr_max_devs > IPR_MAX_PHYSICAL_DEVS) @@ -9967,7 +9969,6 @@ static void ipr_init_ioa_cfg(struct ipr_ioa_cfg *ioa_cfg, * ioa_cfg->max_devs_supported))); } - host->max_channel = IPR_VSET_BUS; host->unique_id = host->host_no; host->max_cmd_len = IPR_MAX_CDB_LEN; host->can_queue = ioa_cfg->max_cmds; diff --git a/drivers/scsi/ipr.h b/drivers/scsi/ipr.h index a67baeb36d1f..b97aa9ac2ffe 100644 --- a/drivers/scsi/ipr.h +++ b/drivers/scsi/ipr.h @@ -1300,6 +1300,7 @@ struct ipr_resource_entry { #define IPR_ARRAY_VIRTUAL_BUS 0x1 #define IPR_VSET_VIRTUAL_BUS 0x2 #define IPR_IOAFP_VIRTUAL_BUS 0x3 +#define IPR_MAX_SIS64_BUSES 0x4 #define IPR_GET_RES_PHYS_LOC(res) \ (((res)->bus << 24) | ((res)->target << 8) | (res)->lun) diff --git a/drivers/scsi/libfc/fc_disc.c b/drivers/scsi/libfc/fc_disc.c index 9c5f7c9178c6..2b865c6423e2 100644 --- a/drivers/scsi/libfc/fc_disc.c +++ b/drivers/scsi/libfc/fc_disc.c @@ -628,6 +628,8 @@ redisc: } out: kref_put(&rdata->kref, fc_rport_destroy); + if (!IS_ERR(fp)) + fc_frame_free(fp); } /** diff --git a/drivers/scsi/megaraid/megaraid_sas_fusion.c b/drivers/scsi/megaraid/megaraid_sas_fusion.c index f3b36fd0a0eb..b2ad96564484 100644 --- a/drivers/scsi/megaraid/megaraid_sas_fusion.c +++ b/drivers/scsi/megaraid/megaraid_sas_fusion.c @@ -623,7 +623,8 @@ retry_alloc: fusion->io_request_frames = dma_pool_alloc(fusion->io_request_frames_pool, - GFP_KERNEL, &fusion->io_request_frames_phys); + GFP_KERNEL | __GFP_NOWARN, + &fusion->io_request_frames_phys); if (!fusion->io_request_frames) { if (instance->max_fw_cmds >= (MEGASAS_REDUCE_QD_COUNT * 2)) { instance->max_fw_cmds -= MEGASAS_REDUCE_QD_COUNT; @@ -661,7 +662,7 @@ retry_alloc: fusion->io_request_frames = dma_pool_alloc(fusion->io_request_frames_pool, - GFP_KERNEL, + GFP_KERNEL | __GFP_NOWARN, &fusion->io_request_frames_phys); if (!fusion->io_request_frames) { diff --git a/drivers/scsi/qla2xxx/qla_os.c b/drivers/scsi/qla2xxx/qla_os.c index b520a980d1dc..7a94e1171c72 100644 --- a/drivers/scsi/qla2xxx/qla_os.c +++ b/drivers/scsi/qla2xxx/qla_os.c @@ -864,7 +864,7 @@ qla2xxx_queuecommand(struct Scsi_Host *host, struct scsi_cmnd *cmd) goto qc24_fail_command; } - if (atomic_read(&fcport->state) != FCS_ONLINE) { + if (atomic_read(&fcport->state) != FCS_ONLINE || fcport->deleted) { if (atomic_read(&fcport->state) == FCS_DEVICE_DEAD || atomic_read(&base_vha->loop_state) == LOOP_DEAD) { ql_dbg(ql_dbg_io, vha, 0x3005, @@ -946,7 +946,7 @@ qla2xxx_mqueuecommand(struct Scsi_Host *host, struct scsi_cmnd *cmd, goto qc24_fail_command; } - if (atomic_read(&fcport->state) != FCS_ONLINE) { + if (atomic_read(&fcport->state) != FCS_ONLINE || fcport->deleted) { if (atomic_read(&fcport->state) == FCS_DEVICE_DEAD || atomic_read(&base_vha->loop_state) == LOOP_DEAD) { ql_dbg(ql_dbg_io, vha, 0x3077, diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c index 8ca9299ffd36..2710a0e5ae6d 100644 --- a/drivers/scsi/sd.c +++ b/drivers/scsi/sd.c @@ -3169,9 +3169,11 @@ static int sd_revalidate_disk(struct gendisk *disk) if (sd_validate_opt_xfer_size(sdkp, dev_max)) { q->limits.io_opt = logical_to_bytes(sdp, sdkp->opt_xfer_blocks); rw_max = logical_to_sectors(sdp, sdkp->opt_xfer_blocks); - } else + } else { + q->limits.io_opt = 0; rw_max = min_not_zero(logical_to_sectors(sdp, dev_max), (sector_t)BLK_DEF_MAX_SECTORS); + } /* Do not exceed controller limit */ rw_max = min(rw_max, queue_max_hw_sectors(q)); diff --git a/drivers/scsi/sd_zbc.c b/drivers/scsi/sd_zbc.c index e4282bce5834..f45c22b09726 100644 --- a/drivers/scsi/sd_zbc.c +++ b/drivers/scsi/sd_zbc.c @@ -161,6 +161,7 @@ int sd_zbc_report_zones(struct gendisk *disk, sector_t sector, unsigned int nr_zones, report_zones_cb cb, void *data) { struct scsi_disk *sdkp = scsi_disk(disk); + sector_t capacity = logical_to_sectors(sdkp->device, sdkp->capacity); unsigned int nr, i; unsigned char *buf; size_t offset, buflen = 0; @@ -171,11 +172,15 @@ int sd_zbc_report_zones(struct gendisk *disk, sector_t sector, /* Not a zoned device */ return -EOPNOTSUPP; + if (!capacity) + /* Device gone or invalid */ + return -ENODEV; + buf = sd_zbc_alloc_report_buffer(sdkp, nr_zones, &buflen); if (!buf) return -ENOMEM; - while (zone_idx < nr_zones && sector < get_capacity(disk)) { + while (zone_idx < nr_zones && sector < capacity) { ret = sd_zbc_do_report_zones(sdkp, buf, buflen, sectors_to_logical(sdkp->device, sector), true); if (ret) diff --git a/drivers/scsi/sr.c b/drivers/scsi/sr.c index 0fbb8fe6e521..e4240e4ae8bb 100644 --- a/drivers/scsi/sr.c +++ b/drivers/scsi/sr.c @@ -688,7 +688,7 @@ static const struct block_device_operations sr_bdops = .release = sr_block_release, .ioctl = sr_block_ioctl, #ifdef CONFIG_COMPAT - .ioctl = sr_block_compat_ioctl, + .compat_ioctl = sr_block_compat_ioctl, #endif .check_events = sr_block_check_events, .revalidate_disk = sr_block_revalidate_disk, diff --git a/drivers/scsi/ufs/ufshcd.c b/drivers/scsi/ufs/ufshcd.c index abd0e6b05f79..2d705694636c 100644 --- a/drivers/scsi/ufs/ufshcd.c +++ b/drivers/scsi/ufs/ufshcd.c @@ -3884,18 +3884,25 @@ EXPORT_SYMBOL_GPL(ufshcd_uic_hibern8_exit); void ufshcd_auto_hibern8_update(struct ufs_hba *hba, u32 ahit) { unsigned long flags; + bool update = false; - if (!(hba->capabilities & MASK_AUTO_HIBERN8_SUPPORT)) + if (!ufshcd_is_auto_hibern8_supported(hba)) return; spin_lock_irqsave(hba->host->host_lock, flags); - if (hba->ahit == ahit) - goto out_unlock; - hba->ahit = ahit; - if (!pm_runtime_suspended(hba->dev)) - ufshcd_writel(hba, hba->ahit, REG_AUTO_HIBERNATE_IDLE_TIMER); -out_unlock: + if (hba->ahit != ahit) { + hba->ahit = ahit; + update = true; + } spin_unlock_irqrestore(hba->host->host_lock, flags); + + if (update && !pm_runtime_suspended(hba->dev)) { + pm_runtime_get_sync(hba->dev); + ufshcd_hold(hba, false); + ufshcd_auto_hibern8_enable(hba); + ufshcd_release(hba); + pm_runtime_put(hba->dev); + } } EXPORT_SYMBOL_GPL(ufshcd_auto_hibern8_update); diff --git a/drivers/slimbus/qcom-ngd-ctrl.c b/drivers/slimbus/qcom-ngd-ctrl.c index e3f5ebc0c05e..fc2575fef51b 100644 --- a/drivers/slimbus/qcom-ngd-ctrl.c +++ b/drivers/slimbus/qcom-ngd-ctrl.c @@ -1320,6 +1320,9 @@ static const struct of_device_id qcom_slim_ngd_dt_match[] = { { .compatible = "qcom,slim-ngd-v1.5.0", .data = &ngd_v1_5_offset_info, + },{ + .compatible = "qcom,slim-ngd-v2.1.0", + .data = &ngd_v1_5_offset_info, }, {} }; diff --git a/drivers/soc/fsl/dpio/dpio-driver.c b/drivers/soc/fsl/dpio/dpio-driver.c index 70014ecce2a7..7b642c330977 100644 --- a/drivers/soc/fsl/dpio/dpio-driver.c +++ b/drivers/soc/fsl/dpio/dpio-driver.c @@ -233,10 +233,6 @@ static int dpaa2_dpio_probe(struct fsl_mc_device *dpio_dev) goto err_allocate_irqs; } - err = register_dpio_irq_handlers(dpio_dev, desc.cpu); - if (err) - goto err_register_dpio_irq; - priv->io = dpaa2_io_create(&desc, dev); if (!priv->io) { dev_err(dev, "dpaa2_io_create failed\n"); @@ -244,6 +240,10 @@ static int dpaa2_dpio_probe(struct fsl_mc_device *dpio_dev) goto err_dpaa2_io_create; } + err = register_dpio_irq_handlers(dpio_dev, desc.cpu); + if (err) + goto err_register_dpio_irq; + dev_info(dev, "probed\n"); dev_dbg(dev, " receives_notifications = %d\n", desc.receives_notifications); diff --git a/drivers/soc/imx/soc-imx-scu.c b/drivers/soc/imx/soc-imx-scu.c index fb70b8a3f7c5..20d37eaeb5f2 100644 --- a/drivers/soc/imx/soc-imx-scu.c +++ b/drivers/soc/imx/soc-imx-scu.c @@ -25,7 +25,7 @@ struct imx_sc_msg_misc_get_soc_id { u32 id; } resp; } data; -} __packed; +} __packed __aligned(4); struct imx_sc_msg_misc_get_soc_uid { struct imx_sc_rpc_msg hdr; diff --git a/drivers/soc/samsung/exynos-chipid.c b/drivers/soc/samsung/exynos-chipid.c index 2dad4961a80b..8d4d05086906 100644 --- a/drivers/soc/samsung/exynos-chipid.c +++ b/drivers/soc/samsung/exynos-chipid.c @@ -59,7 +59,7 @@ static int __init exynos_chipid_early_init(void) syscon = of_find_compatible_node(NULL, NULL, "samsung,exynos4210-chipid"); if (!syscon) - return ENODEV; + return -ENODEV; regmap = device_node_to_regmap(syscon); of_node_put(syscon); diff --git a/drivers/spi/atmel-quadspi.c b/drivers/spi/atmel-quadspi.c index fd8007ebb145..13def7f78b9e 100644 --- a/drivers/spi/atmel-quadspi.c +++ b/drivers/spi/atmel-quadspi.c @@ -149,6 +149,7 @@ struct atmel_qspi { struct clk *qspick; struct platform_device *pdev; const struct atmel_qspi_caps *caps; + resource_size_t mmap_size; u32 pending; u32 mr; u32 scr; @@ -329,6 +330,14 @@ static int atmel_qspi_exec_op(struct spi_mem *mem, const struct spi_mem_op *op) u32 sr, offset; int err; + /* + * Check if the address exceeds the MMIO window size. An improvement + * would be to add support for regular SPI mode and fall back to it + * when the flash memories overrun the controller's memory space. + */ + if (op->addr.val + op->data.nbytes > aq->mmap_size) + return -ENOTSUPP; + err = atmel_qspi_set_cfg(aq, op, &offset); if (err) return err; @@ -480,6 +489,8 @@ static int atmel_qspi_probe(struct platform_device *pdev) goto exit; } + aq->mmap_size = resource_size(res); + /* Get the peripheral clock */ aq->pclk = devm_clk_get(&pdev->dev, "pclk"); if (IS_ERR(aq->pclk)) diff --git a/drivers/spi/spi-bcm63xx-hsspi.c b/drivers/spi/spi-bcm63xx-hsspi.c index 7327309ea3d5..6c235306c0e4 100644 --- a/drivers/spi/spi-bcm63xx-hsspi.c +++ b/drivers/spi/spi-bcm63xx-hsspi.c @@ -366,7 +366,6 @@ static int bcm63xx_hsspi_probe(struct platform_device *pdev) goto out_disable_clk; rate = clk_get_rate(pll_clk); - clk_disable_unprepare(pll_clk); if (!rate) { ret = -EINVAL; goto out_disable_pll_clk; diff --git a/drivers/spi/spi-omap2-mcspi.c b/drivers/spi/spi-omap2-mcspi.c index 7e2292c11d12..e9e256718ef4 100644 --- a/drivers/spi/spi-omap2-mcspi.c +++ b/drivers/spi/spi-omap2-mcspi.c @@ -130,6 +130,7 @@ struct omap2_mcspi { int fifo_depth; bool slave_aborted; unsigned int pin_dir:1; + size_t max_xfer_len; }; struct omap2_mcspi_cs { @@ -974,20 +975,12 @@ static int omap2_mcspi_setup_transfer(struct spi_device *spi, * Note that we currently allow DMA only if we get a channel * for both rx and tx. Otherwise we'll do PIO for both rx and tx. */ -static int omap2_mcspi_request_dma(struct spi_device *spi) +static int omap2_mcspi_request_dma(struct omap2_mcspi *mcspi, + struct omap2_mcspi_dma *mcspi_dma) { - struct spi_master *master = spi->master; - struct omap2_mcspi *mcspi; - struct omap2_mcspi_dma *mcspi_dma; int ret = 0; - mcspi = spi_master_get_devdata(master); - mcspi_dma = mcspi->dma_channels + spi->chip_select; - - init_completion(&mcspi_dma->dma_rx_completion); - init_completion(&mcspi_dma->dma_tx_completion); - - mcspi_dma->dma_rx = dma_request_chan(&master->dev, + mcspi_dma->dma_rx = dma_request_chan(mcspi->dev, mcspi_dma->dma_rx_ch_name); if (IS_ERR(mcspi_dma->dma_rx)) { ret = PTR_ERR(mcspi_dma->dma_rx); @@ -995,7 +988,7 @@ static int omap2_mcspi_request_dma(struct spi_device *spi) goto no_dma; } - mcspi_dma->dma_tx = dma_request_chan(&master->dev, + mcspi_dma->dma_tx = dma_request_chan(mcspi->dev, mcspi_dma->dma_tx_ch_name); if (IS_ERR(mcspi_dma->dma_tx)) { ret = PTR_ERR(mcspi_dma->dma_tx); @@ -1004,20 +997,40 @@ static int omap2_mcspi_request_dma(struct spi_device *spi) mcspi_dma->dma_rx = NULL; } + init_completion(&mcspi_dma->dma_rx_completion); + init_completion(&mcspi_dma->dma_tx_completion); + no_dma: return ret; } +static void omap2_mcspi_release_dma(struct spi_master *master) +{ + struct omap2_mcspi *mcspi = spi_master_get_devdata(master); + struct omap2_mcspi_dma *mcspi_dma; + int i; + + for (i = 0; i < master->num_chipselect; i++) { + mcspi_dma = &mcspi->dma_channels[i]; + + if (mcspi_dma->dma_rx) { + dma_release_channel(mcspi_dma->dma_rx); + mcspi_dma->dma_rx = NULL; + } + if (mcspi_dma->dma_tx) { + dma_release_channel(mcspi_dma->dma_tx); + mcspi_dma->dma_tx = NULL; + } + } +} + static int omap2_mcspi_setup(struct spi_device *spi) { int ret; struct omap2_mcspi *mcspi = spi_master_get_devdata(spi->master); struct omap2_mcspi_regs *ctx = &mcspi->ctx; - struct omap2_mcspi_dma *mcspi_dma; struct omap2_mcspi_cs *cs = spi->controller_state; - mcspi_dma = &mcspi->dma_channels[spi->chip_select]; - if (!cs) { cs = kzalloc(sizeof *cs, GFP_KERNEL); if (!cs) @@ -1042,13 +1055,6 @@ static int omap2_mcspi_setup(struct spi_device *spi) } } - if (!mcspi_dma->dma_rx || !mcspi_dma->dma_tx) { - ret = omap2_mcspi_request_dma(spi); - if (ret) - dev_warn(&spi->dev, "not using DMA for McSPI (%d)\n", - ret); - } - ret = pm_runtime_get_sync(mcspi->dev); if (ret < 0) { pm_runtime_put_noidle(mcspi->dev); @@ -1065,12 +1071,8 @@ static int omap2_mcspi_setup(struct spi_device *spi) static void omap2_mcspi_cleanup(struct spi_device *spi) { - struct omap2_mcspi *mcspi; - struct omap2_mcspi_dma *mcspi_dma; struct omap2_mcspi_cs *cs; - mcspi = spi_master_get_devdata(spi->master); - if (spi->controller_state) { /* Unlink controller state from context save list */ cs = spi->controller_state; @@ -1079,19 +1081,6 @@ static void omap2_mcspi_cleanup(struct spi_device *spi) kfree(cs); } - if (spi->chip_select < spi->master->num_chipselect) { - mcspi_dma = &mcspi->dma_channels[spi->chip_select]; - - if (mcspi_dma->dma_rx) { - dma_release_channel(mcspi_dma->dma_rx); - mcspi_dma->dma_rx = NULL; - } - if (mcspi_dma->dma_tx) { - dma_release_channel(mcspi_dma->dma_tx); - mcspi_dma->dma_tx = NULL; - } - } - if (gpio_is_valid(spi->cs_gpio)) gpio_free(spi->cs_gpio); } @@ -1302,9 +1291,24 @@ static bool omap2_mcspi_can_dma(struct spi_master *master, if (spi_controller_is_slave(master)) return true; + master->dma_rx = mcspi_dma->dma_rx; + master->dma_tx = mcspi_dma->dma_tx; + return (xfer->len >= DMA_MIN_BYTES); } +static size_t omap2_mcspi_max_xfer_size(struct spi_device *spi) +{ + struct omap2_mcspi *mcspi = spi_master_get_devdata(spi->master); + struct omap2_mcspi_dma *mcspi_dma = + &mcspi->dma_channels[spi->chip_select]; + + if (mcspi->max_xfer_len && mcspi_dma->dma_rx) + return mcspi->max_xfer_len; + + return SIZE_MAX; +} + static int omap2_mcspi_controller_setup(struct omap2_mcspi *mcspi) { struct spi_master *master = mcspi->master; @@ -1373,6 +1377,11 @@ static struct omap2_mcspi_platform_config omap4_pdata = { .regs_offset = OMAP4_MCSPI_REG_OFFSET, }; +static struct omap2_mcspi_platform_config am654_pdata = { + .regs_offset = OMAP4_MCSPI_REG_OFFSET, + .max_xfer_len = SZ_4K - 1, +}; + static const struct of_device_id omap_mcspi_of_match[] = { { .compatible = "ti,omap2-mcspi", @@ -1382,6 +1391,10 @@ static const struct of_device_id omap_mcspi_of_match[] = { .compatible = "ti,omap4-mcspi", .data = &omap4_pdata, }, + { + .compatible = "ti,am654-mcspi", + .data = &am654_pdata, + }, { }, }; MODULE_DEVICE_TABLE(of, omap_mcspi_of_match); @@ -1439,6 +1452,10 @@ static int omap2_mcspi_probe(struct platform_device *pdev) mcspi->pin_dir = pdata->pin_dir; } regs_offset = pdata->regs_offset; + if (pdata->max_xfer_len) { + mcspi->max_xfer_len = pdata->max_xfer_len; + master->max_transfer_size = omap2_mcspi_max_xfer_size; + } r = platform_get_resource(pdev, IORESOURCE_MEM, 0); mcspi->base = devm_ioremap_resource(&pdev->dev, r); @@ -1464,6 +1481,11 @@ static int omap2_mcspi_probe(struct platform_device *pdev) for (i = 0; i < master->num_chipselect; i++) { sprintf(mcspi->dma_channels[i].dma_rx_ch_name, "rx%d", i); sprintf(mcspi->dma_channels[i].dma_tx_ch_name, "tx%d", i); + + status = omap2_mcspi_request_dma(mcspi, + &mcspi->dma_channels[i]); + if (status == -EPROBE_DEFER) + goto free_master; } status = platform_get_irq(pdev, 0); @@ -1501,6 +1523,7 @@ disable_pm: pm_runtime_put_sync(&pdev->dev); pm_runtime_disable(&pdev->dev); free_master: + omap2_mcspi_release_dma(master); spi_master_put(master); return status; } @@ -1510,6 +1533,8 @@ static int omap2_mcspi_remove(struct platform_device *pdev) struct spi_master *master = platform_get_drvdata(pdev); struct omap2_mcspi *mcspi = spi_master_get_devdata(master); + omap2_mcspi_release_dma(master); + pm_runtime_dont_use_autosuspend(mcspi->dev); pm_runtime_put_sync(mcspi->dev); pm_runtime_disable(&pdev->dev); diff --git a/drivers/spi/spi-pxa2xx.c b/drivers/spi/spi-pxa2xx.c index 4c7a71f0fb3e..2e318158fca9 100644 --- a/drivers/spi/spi-pxa2xx.c +++ b/drivers/spi/spi-pxa2xx.c @@ -70,6 +70,10 @@ MODULE_ALIAS("platform:pxa2xx-spi"); #define LPSS_CAPS_CS_EN_SHIFT 9 #define LPSS_CAPS_CS_EN_MASK (0xf << LPSS_CAPS_CS_EN_SHIFT) +#define LPSS_PRIV_CLOCK_GATE 0x38 +#define LPSS_PRIV_CLOCK_GATE_CLK_CTL_MASK 0x3 +#define LPSS_PRIV_CLOCK_GATE_CLK_CTL_FORCE_ON 0x3 + struct lpss_config { /* LPSS offset from drv_data->ioaddr */ unsigned offset; @@ -86,6 +90,8 @@ struct lpss_config { unsigned cs_sel_shift; unsigned cs_sel_mask; unsigned cs_num; + /* Quirks */ + unsigned cs_clk_stays_gated : 1; }; /* Keep these sorted with enum pxa_ssp_type */ @@ -156,6 +162,7 @@ static const struct lpss_config lpss_platforms[] = { .tx_threshold_hi = 56, .cs_sel_shift = 8, .cs_sel_mask = 3 << 8, + .cs_clk_stays_gated = true, }, }; @@ -383,6 +390,22 @@ static void lpss_ssp_cs_control(struct spi_device *spi, bool enable) else value |= LPSS_CS_CONTROL_CS_HIGH; __lpss_ssp_write_priv(drv_data, config->reg_cs_ctrl, value); + if (config->cs_clk_stays_gated) { + u32 clkgate; + + /* + * Changing CS alone when dynamic clock gating is on won't + * actually flip CS at that time. This ruins SPI transfers + * that specify delays, or have no data. Toggle the clock mode + * to force on briefly to poke the CS pin to move. + */ + clkgate = __lpss_ssp_read_priv(drv_data, LPSS_PRIV_CLOCK_GATE); + value = (clkgate & ~LPSS_PRIV_CLOCK_GATE_CLK_CTL_MASK) | + LPSS_PRIV_CLOCK_GATE_CLK_CTL_FORCE_ON; + + __lpss_ssp_write_priv(drv_data, LPSS_PRIV_CLOCK_GATE, value); + __lpss_ssp_write_priv(drv_data, LPSS_PRIV_CLOCK_GATE, clkgate); + } } static void cs_assert(struct spi_device *spi) diff --git a/drivers/spi/spi-qup.c b/drivers/spi/spi-qup.c index dd3434a407ea..a364b99497e2 100644 --- a/drivers/spi/spi-qup.c +++ b/drivers/spi/spi-qup.c @@ -1217,6 +1217,11 @@ static int spi_qup_suspend(struct device *device) struct spi_qup *controller = spi_master_get_devdata(master); int ret; + if (pm_runtime_suspended(device)) { + ret = spi_qup_pm_resume_runtime(device); + if (ret) + return ret; + } ret = spi_master_suspend(master); if (ret) return ret; @@ -1225,10 +1230,8 @@ static int spi_qup_suspend(struct device *device) if (ret) return ret; - if (!pm_runtime_suspended(device)) { - clk_disable_unprepare(controller->cclk); - clk_disable_unprepare(controller->iclk); - } + clk_disable_unprepare(controller->cclk); + clk_disable_unprepare(controller->iclk); return 0; } diff --git a/drivers/spi/spi-zynqmp-gqspi.c b/drivers/spi/spi-zynqmp-gqspi.c index 60c4de4e4485..7412a3042a8d 100644 --- a/drivers/spi/spi-zynqmp-gqspi.c +++ b/drivers/spi/spi-zynqmp-gqspi.c @@ -401,9 +401,6 @@ static void zynqmp_qspi_chipselect(struct spi_device *qspi, bool is_high) zynqmp_gqspi_write(xqspi, GQSPI_GEN_FIFO_OFST, genfifoentry); - /* Dummy generic FIFO entry */ - zynqmp_gqspi_write(xqspi, GQSPI_GEN_FIFO_OFST, 0x0); - /* Manually start the generic FIFO command */ zynqmp_gqspi_write(xqspi, GQSPI_CONFIG_OFST, zynqmp_gqspi_read(xqspi, GQSPI_CONFIG_OFST) | diff --git a/drivers/spi/spi.c b/drivers/spi/spi.c index 38b4c78df506..755221bc3745 100644 --- a/drivers/spi/spi.c +++ b/drivers/spi/spi.c @@ -2639,7 +2639,7 @@ int spi_register_controller(struct spi_controller *ctlr) if (ctlr->use_gpio_descriptors) { status = spi_get_gpio_descs(ctlr); if (status) - return status; + goto free_bus_id; /* * A controller using GPIO descriptors always * supports SPI_CS_HIGH if need be. @@ -2649,7 +2649,7 @@ int spi_register_controller(struct spi_controller *ctlr) /* Legacy code path for GPIOs from DT */ status = of_spi_get_gpio_numbers(ctlr); if (status) - return status; + goto free_bus_id; } } @@ -2657,17 +2657,14 @@ int spi_register_controller(struct spi_controller *ctlr) * Even if it's just one always-selected device, there must * be at least one chipselect. */ - if (!ctlr->num_chipselect) - return -EINVAL; + if (!ctlr->num_chipselect) { + status = -EINVAL; + goto free_bus_id; + } status = device_add(&ctlr->dev); - if (status < 0) { - /* free bus id */ - mutex_lock(&board_lock); - idr_remove(&spi_master_idr, ctlr->bus_num); - mutex_unlock(&board_lock); - goto done; - } + if (status < 0) + goto free_bus_id; dev_dbg(dev, "registered %s %s\n", spi_controller_is_slave(ctlr) ? "slave" : "master", dev_name(&ctlr->dev)); @@ -2683,11 +2680,7 @@ int spi_register_controller(struct spi_controller *ctlr) status = spi_controller_initialize_queue(ctlr); if (status) { device_del(&ctlr->dev); - /* free bus id */ - mutex_lock(&board_lock); - idr_remove(&spi_master_idr, ctlr->bus_num); - mutex_unlock(&board_lock); - goto done; + goto free_bus_id; } } /* add statistics */ @@ -2702,7 +2695,12 @@ int spi_register_controller(struct spi_controller *ctlr) /* Register devices from the device tree and ACPI */ of_register_spi_devices(ctlr); acpi_register_spi_devices(ctlr); -done: + return status; + +free_bus_id: + mutex_lock(&board_lock); + idr_remove(&spi_master_idr, ctlr->bus_num); + mutex_unlock(&board_lock); return status; } EXPORT_SYMBOL_GPL(spi_register_controller); diff --git a/drivers/spi/spidev.c b/drivers/spi/spidev.c index 1e217e3e9486..2ab6e782f14c 100644 --- a/drivers/spi/spidev.c +++ b/drivers/spi/spidev.c @@ -396,6 +396,7 @@ spidev_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) else retval = get_user(tmp, (u32 __user *)arg); if (retval == 0) { + struct spi_controller *ctlr = spi->controller; u32 save = spi->mode; if (tmp & ~SPI_MODE_MASK) { @@ -403,6 +404,10 @@ spidev_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) break; } + if (ctlr->use_gpio_descriptors && ctlr->cs_gpiods && + ctlr->cs_gpiods[spi->chip_select]) + tmp |= SPI_CS_HIGH; + tmp |= spi->mode & ~SPI_MODE_MASK; spi->mode = (u16)tmp; retval = spi_setup(spi); diff --git a/drivers/staging/android/Kconfig b/drivers/staging/android/Kconfig index d6d605d5cbde..8d8fd5c29349 100644 --- a/drivers/staging/android/Kconfig +++ b/drivers/staging/android/Kconfig @@ -14,14 +14,6 @@ config ASHMEM It is, in theory, a good memory allocator for low-memory devices, because it can discard shared memory units when under memory pressure. -config ANDROID_VSOC - tristate "Android Virtual SoC support" - depends on PCI_MSI - help - This option adds support for the Virtual SoC driver needed to boot - a 'cuttlefish' Android image inside QEmu. The driver interacts with - a QEmu ivshmem device. If built as a module, it will be called vsoc. - source "drivers/staging/android/ion/Kconfig" endif # if ANDROID diff --git a/drivers/staging/android/Makefile b/drivers/staging/android/Makefile index 14bd9c6ce10d..3b66cd0b0ec5 100644 --- a/drivers/staging/android/Makefile +++ b/drivers/staging/android/Makefile @@ -4,4 +4,3 @@ ccflags-y += -I$(src) # needed for trace events obj-y += ion/ obj-$(CONFIG_ASHMEM) += ashmem.o -obj-$(CONFIG_ANDROID_VSOC) += vsoc.o diff --git a/drivers/staging/android/TODO b/drivers/staging/android/TODO index 767dd98fd92d..80eccfaf6db5 100644 --- a/drivers/staging/android/TODO +++ b/drivers/staging/android/TODO @@ -9,14 +9,5 @@ ion/ - Split /dev/ion up into multiple nodes (e.g. /dev/ion/heap0) - Better test framework (integration with VGEM was suggested) -vsoc.c, uapi/vsoc_shm.h - - The current driver uses the same wait queue for all of the futexes in a - region. This will cause false wakeups in regions with a large number of - waiting threads. We should eventually use multiple queues and select the - queue based on the region. - - Add debugfs support for examining the permissions of regions. - - Remove VSOC_WAIT_FOR_INCOMING_INTERRUPT ioctl. This functionality has been - superseded by the futex and is there for legacy reasons. - Please send patches to Greg Kroah-Hartman <greg@kroah.com> and Cc: Arve HjønnevÃ¥g <arve@android.com> and Riley Andrews <riandrews@android.com> diff --git a/drivers/staging/android/ashmem.c b/drivers/staging/android/ashmem.c index 5891d0744a76..8044510d8ec6 100644 --- a/drivers/staging/android/ashmem.c +++ b/drivers/staging/android/ashmem.c @@ -351,8 +351,23 @@ static inline vm_flags_t calc_vm_may_flags(unsigned long prot) _calc_vm_trans(prot, PROT_EXEC, VM_MAYEXEC); } +static int ashmem_vmfile_mmap(struct file *file, struct vm_area_struct *vma) +{ + /* do not allow to mmap ashmem backing shmem file directly */ + return -EPERM; +} + +static unsigned long +ashmem_vmfile_get_unmapped_area(struct file *file, unsigned long addr, + unsigned long len, unsigned long pgoff, + unsigned long flags) +{ + return current->mm->get_unmapped_area(file, addr, len, pgoff, flags); +} + static int ashmem_mmap(struct file *file, struct vm_area_struct *vma) { + static struct file_operations vmfile_fops; struct ashmem_area *asma = file->private_data; int ret = 0; @@ -393,6 +408,19 @@ static int ashmem_mmap(struct file *file, struct vm_area_struct *vma) } vmfile->f_mode |= FMODE_LSEEK; asma->file = vmfile; + /* + * override mmap operation of the vmfile so that it can't be + * remapped which would lead to creation of a new vma with no + * asma permission checks. Have to override get_unmapped_area + * as well to prevent VM_BUG_ON check for f_ops modification. + */ + if (!vmfile_fops.mmap) { + vmfile_fops = *vmfile->f_op; + vmfile_fops.mmap = ashmem_vmfile_mmap; + vmfile_fops.get_unmapped_area = + ashmem_vmfile_get_unmapped_area; + } + vmfile->f_op = &vmfile_fops; } get_file(asma->file); diff --git a/drivers/staging/android/uapi/vsoc_shm.h b/drivers/staging/android/uapi/vsoc_shm.h deleted file mode 100644 index 6291fb24efb2..000000000000 --- a/drivers/staging/android/uapi/vsoc_shm.h +++ /dev/null @@ -1,295 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0 */ -/* - * Copyright (C) 2017 Google, Inc. - * - */ - -#ifndef _UAPI_LINUX_VSOC_SHM_H -#define _UAPI_LINUX_VSOC_SHM_H - -#include <linux/types.h> - -/** - * A permission is a token that permits a receiver to read and/or write an area - * of memory within a Vsoc region. - * - * An fd_scoped permission grants both read and write access, and can be - * attached to a file description (see open(2)). - * Ownership of the area can then be shared by passing a file descriptor - * among processes. - * - * begin_offset and end_offset define the area of memory that is controlled by - * the permission. owner_offset points to a word, also in shared memory, that - * controls ownership of the area. - * - * ownership of the region expires when the associated file description is - * released. - * - * At most one permission can be attached to each file description. - * - * This is useful when implementing HALs like gralloc that scope and pass - * ownership of shared resources via file descriptors. - * - * The caller is responsibe for doing any fencing. - * - * The calling process will normally identify a currently free area of - * memory. It will construct a proposed fd_scoped_permission_arg structure: - * - * begin_offset and end_offset describe the area being claimed - * - * owner_offset points to the location in shared memory that indicates the - * owner of the area. - * - * owned_value is the value that will be stored in owner_offset iff the - * permission can be granted. It must be different than VSOC_REGION_FREE. - * - * Two fd_scoped_permission structures are compatible if they vary only by - * their owned_value fields. - * - * The driver ensures that, for any group of simultaneous callers proposing - * compatible fd_scoped_permissions, it will accept exactly one of the - * propopsals. The other callers will get a failure with errno of EAGAIN. - * - * A process receiving a file descriptor can identify the region being - * granted using the VSOC_GET_FD_SCOPED_PERMISSION ioctl. - */ -struct fd_scoped_permission { - __u32 begin_offset; - __u32 end_offset; - __u32 owner_offset; - __u32 owned_value; -}; - -/* - * This value represents a free area of memory. The driver expects to see this - * value at owner_offset when creating a permission otherwise it will not do it, - * and will write this value back once the permission is no longer needed. - */ -#define VSOC_REGION_FREE ((__u32)0) - -/** - * ioctl argument for VSOC_CREATE_FD_SCOPE_PERMISSION - */ -struct fd_scoped_permission_arg { - struct fd_scoped_permission perm; - __s32 managed_region_fd; -}; - -#define VSOC_NODE_FREE ((__u32)0) - -/* - * Describes a signal table in shared memory. Each non-zero entry in the - * table indicates that the receiver should signal the futex at the given - * offset. Offsets are relative to the region, not the shared memory window. - * - * interrupt_signalled_offset is used to reliably signal interrupts across the - * vmm boundary. There are two roles: transmitter and receiver. For example, - * in the host_to_guest_signal_table the host is the transmitter and the - * guest is the receiver. The protocol is as follows: - * - * 1. The transmitter should convert the offset of the futex to an offset - * in the signal table [0, (1 << num_nodes_lg2)) - * The transmitter can choose any appropriate hashing algorithm, including - * hash = futex_offset & ((1 << num_nodes_lg2) - 1) - * - * 3. The transmitter should atomically compare and swap futex_offset with 0 - * at hash. There are 3 possible outcomes - * a. The swap fails because the futex_offset is already in the table. - * The transmitter should stop. - * b. Some other offset is in the table. This is a hash collision. The - * transmitter should move to another table slot and try again. One - * possible algorithm: - * hash = (hash + 1) & ((1 << num_nodes_lg2) - 1) - * c. The swap worked. Continue below. - * - * 3. The transmitter atomically swaps 1 with the value at the - * interrupt_signalled_offset. There are two outcomes: - * a. The prior value was 1. In this case an interrupt has already been - * posted. The transmitter is done. - * b. The prior value was 0, indicating that the receiver may be sleeping. - * The transmitter will issue an interrupt. - * - * 4. On waking the receiver immediately exchanges a 0 with the - * interrupt_signalled_offset. If it receives a 0 then this a spurious - * interrupt. That may occasionally happen in the current protocol, but - * should be rare. - * - * 5. The receiver scans the signal table by atomicaly exchanging 0 at each - * location. If a non-zero offset is returned from the exchange the - * receiver wakes all sleepers at the given offset: - * futex((int*)(region_base + old_value), FUTEX_WAKE, MAX_INT); - * - * 6. The receiver thread then does a conditional wait, waking immediately - * if the value at interrupt_signalled_offset is non-zero. This catches cases - * here additional signals were posted while the table was being scanned. - * On the guest the wait is handled via the VSOC_WAIT_FOR_INCOMING_INTERRUPT - * ioctl. - */ -struct vsoc_signal_table_layout { - /* log_2(Number of signal table entries) */ - __u32 num_nodes_lg2; - /* - * Offset to the first signal table entry relative to the start of the - * region - */ - __u32 futex_uaddr_table_offset; - /* - * Offset to an atomic_t / atomic uint32_t. A non-zero value indicates - * that one or more offsets are currently posted in the table. - * semi-unique access to an entry in the table - */ - __u32 interrupt_signalled_offset; -}; - -#define VSOC_REGION_WHOLE ((__s32)0) -#define VSOC_DEVICE_NAME_SZ 16 - -/** - * Each HAL would (usually) talk to a single device region - * Mulitple entities care about these regions: - * - The ivshmem_server will populate the regions in shared memory - * - The guest kernel will read the region, create minor device nodes, and - * allow interested parties to register for FUTEX_WAKE events in the region - * - HALs will access via the minor device nodes published by the guest kernel - * - Host side processes will access the region via the ivshmem_server: - * 1. Pass name to ivshmem_server at a UNIX socket - * 2. ivshmemserver will reply with 2 fds: - * - host->guest doorbell fd - * - guest->host doorbell fd - * - fd for the shared memory region - * - region offset - * 3. Start a futex receiver thread on the doorbell fd pointed at the - * signal_nodes - */ -struct vsoc_device_region { - __u16 current_version; - __u16 min_compatible_version; - __u32 region_begin_offset; - __u32 region_end_offset; - __u32 offset_of_region_data; - struct vsoc_signal_table_layout guest_to_host_signal_table; - struct vsoc_signal_table_layout host_to_guest_signal_table; - /* Name of the device. Must always be terminated with a '\0', so - * the longest supported device name is 15 characters. - */ - char device_name[VSOC_DEVICE_NAME_SZ]; - /* There are two ways that permissions to access regions are handled: - * - When subdivided_by is VSOC_REGION_WHOLE, any process that can - * open the device node for the region gains complete access to it. - * - When subdivided is set processes that open the region cannot - * access it. Access to a sub-region must be established by invoking - * the VSOC_CREATE_FD_SCOPE_PERMISSION ioctl on the region - * referenced in subdivided_by, providing a fileinstance - * (represented by a fd) opened on this region. - */ - __u32 managed_by; -}; - -/* - * The vsoc layout descriptor. - * The first 4K should be reserved for the shm header and region descriptors. - * The regions should be page aligned. - */ - -struct vsoc_shm_layout_descriptor { - __u16 major_version; - __u16 minor_version; - - /* size of the shm. This may be redundant but nice to have */ - __u32 size; - - /* number of shared memory regions */ - __u32 region_count; - - /* The offset to the start of region descriptors */ - __u32 vsoc_region_desc_offset; -}; - -/* - * This specifies the current version that should be stored in - * vsoc_shm_layout_descriptor.major_version and - * vsoc_shm_layout_descriptor.minor_version. - * It should be updated only if the vsoc_device_region and - * vsoc_shm_layout_descriptor structures have changed. - * Versioning within each region is transferred - * via the min_compatible_version and current_version fields in - * vsoc_device_region. The driver does not consult these fields: they are left - * for the HALs and host processes and will change independently of the layout - * version. - */ -#define CURRENT_VSOC_LAYOUT_MAJOR_VERSION 2 -#define CURRENT_VSOC_LAYOUT_MINOR_VERSION 0 - -#define VSOC_CREATE_FD_SCOPED_PERMISSION \ - _IOW(0xF5, 0, struct fd_scoped_permission) -#define VSOC_GET_FD_SCOPED_PERMISSION _IOR(0xF5, 1, struct fd_scoped_permission) - -/* - * This is used to signal the host to scan the guest_to_host_signal_table - * for new futexes to wake. This sends an interrupt if one is not already - * in flight. - */ -#define VSOC_MAYBE_SEND_INTERRUPT_TO_HOST _IO(0xF5, 2) - -/* - * When this returns the guest will scan host_to_guest_signal_table to - * check for new futexes to wake. - */ -/* TODO(ghartman): Consider moving this to the bottom half */ -#define VSOC_WAIT_FOR_INCOMING_INTERRUPT _IO(0xF5, 3) - -/* - * Guest HALs will use this to retrieve the region description after - * opening their device node. - */ -#define VSOC_DESCRIBE_REGION _IOR(0xF5, 4, struct vsoc_device_region) - -/* - * Wake any threads that may be waiting for a host interrupt on this region. - * This is mostly used during shutdown. - */ -#define VSOC_SELF_INTERRUPT _IO(0xF5, 5) - -/* - * This is used to signal the host to scan the guest_to_host_signal_table - * for new futexes to wake. This sends an interrupt unconditionally. - */ -#define VSOC_SEND_INTERRUPT_TO_HOST _IO(0xF5, 6) - -enum wait_types { - VSOC_WAIT_UNDEFINED = 0, - VSOC_WAIT_IF_EQUAL = 1, - VSOC_WAIT_IF_EQUAL_TIMEOUT = 2 -}; - -/* - * Wait for a condition to be true - * - * Note, this is sized and aligned so the 32 bit and 64 bit layouts are - * identical. - */ -struct vsoc_cond_wait { - /* Input: Offset of the 32 bit word to check */ - __u32 offset; - /* Input: Value that will be compared with the offset */ - __u32 value; - /* Monotonic time to wake at in seconds */ - __u64 wake_time_sec; - /* Input: Monotonic time to wait in nanoseconds */ - __u32 wake_time_nsec; - /* Input: Type of wait */ - __u32 wait_type; - /* Output: Number of times the thread woke before returning. */ - __u32 wakes; - /* Ensure that we're 8-byte aligned and 8 byte length for 32/64 bit - * compatibility. - */ - __u32 reserved_1; -}; - -#define VSOC_COND_WAIT _IOWR(0xF5, 7, struct vsoc_cond_wait) - -/* Wake any local threads waiting at the offset given in arg */ -#define VSOC_COND_WAKE _IO(0xF5, 8) - -#endif /* _UAPI_LINUX_VSOC_SHM_H */ diff --git a/drivers/staging/android/vsoc.c b/drivers/staging/android/vsoc.c deleted file mode 100644 index 1240bb0317d9..000000000000 --- a/drivers/staging/android/vsoc.c +++ /dev/null @@ -1,1149 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0 -/* - * drivers/android/staging/vsoc.c - * - * Android Virtual System on a Chip (VSoC) driver - * - * Copyright (C) 2017 Google, Inc. - * - * Author: ghartman@google.com - * - * Based on drivers/char/kvm_ivshmem.c - driver for KVM Inter-VM shared memory - * Copyright 2009 Cam Macdonell <cam@cs.ualberta.ca> - * - * Based on cirrusfb.c and 8139cp.c: - * Copyright 1999-2001 Jeff Garzik - * Copyright 2001-2004 Jeff Garzik - */ - -#include <linux/dma-mapping.h> -#include <linux/freezer.h> -#include <linux/futex.h> -#include <linux/init.h> -#include <linux/kernel.h> -#include <linux/module.h> -#include <linux/mutex.h> -#include <linux/pci.h> -#include <linux/proc_fs.h> -#include <linux/sched.h> -#include <linux/syscalls.h> -#include <linux/uaccess.h> -#include <linux/interrupt.h> -#include <linux/cdev.h> -#include <linux/file.h> -#include "uapi/vsoc_shm.h" - -#define VSOC_DEV_NAME "vsoc" - -/* - * Description of the ivshmem-doorbell PCI device used by QEmu. These - * constants follow docs/specs/ivshmem-spec.txt, which can be found in - * the QEmu repository. This was last reconciled with the version that - * came out with 2.8 - */ - -/* - * These constants are determined KVM Inter-VM shared memory device - * register offsets - */ -enum { - INTR_MASK = 0x00, /* Interrupt Mask */ - INTR_STATUS = 0x04, /* Interrupt Status */ - IV_POSITION = 0x08, /* VM ID */ - DOORBELL = 0x0c, /* Doorbell */ -}; - -static const int REGISTER_BAR; /* Equal to 0 */ -static const int MAX_REGISTER_BAR_LEN = 0x100; -/* - * The MSI-x BAR is not used directly. - * - * static const int MSI_X_BAR = 1; - */ -static const int SHARED_MEMORY_BAR = 2; - -struct vsoc_region_data { - char name[VSOC_DEVICE_NAME_SZ + 1]; - wait_queue_head_t interrupt_wait_queue; - /* TODO(b/73664181): Use multiple futex wait queues */ - wait_queue_head_t futex_wait_queue; - /* Flag indicating that an interrupt has been signalled by the host. */ - atomic_t *incoming_signalled; - /* Flag indicating the guest has signalled the host. */ - atomic_t *outgoing_signalled; - bool irq_requested; - bool device_created; -}; - -struct vsoc_device { - /* Kernel virtual address of REGISTER_BAR. */ - void __iomem *regs; - /* Physical address of SHARED_MEMORY_BAR. */ - phys_addr_t shm_phys_start; - /* Kernel virtual address of SHARED_MEMORY_BAR. */ - void __iomem *kernel_mapped_shm; - /* Size of the entire shared memory window in bytes. */ - size_t shm_size; - /* - * Pointer to the virtual address of the shared memory layout structure. - * This is probably identical to kernel_mapped_shm, but saving this - * here saves a lot of annoying casts. - */ - struct vsoc_shm_layout_descriptor *layout; - /* - * Points to a table of region descriptors in the kernel's virtual - * address space. Calculated from - * vsoc_shm_layout_descriptor.vsoc_region_desc_offset - */ - struct vsoc_device_region *regions; - /* Head of a list of permissions that have been granted. */ - struct list_head permissions; - struct pci_dev *dev; - /* Per-region (and therefore per-interrupt) information. */ - struct vsoc_region_data *regions_data; - /* - * Table of msi-x entries. This has to be separated from struct - * vsoc_region_data because the kernel deals with them as an array. - */ - struct msix_entry *msix_entries; - /* Mutex that protectes the permission list */ - struct mutex mtx; - /* Major number assigned by the kernel */ - int major; - /* Character device assigned by the kernel */ - struct cdev cdev; - /* Device class assigned by the kernel */ - struct class *class; - /* - * Flags that indicate what we've initialized. These are used to do an - * orderly cleanup of the device. - */ - bool enabled_device; - bool requested_regions; - bool cdev_added; - bool class_added; - bool msix_enabled; -}; - -static struct vsoc_device vsoc_dev; - -/* - * TODO(ghartman): Add a /sys filesystem entry that summarizes the permissions. - */ - -struct fd_scoped_permission_node { - struct fd_scoped_permission permission; - struct list_head list; -}; - -struct vsoc_private_data { - struct fd_scoped_permission_node *fd_scoped_permission_node; -}; - -static long vsoc_ioctl(struct file *, unsigned int, unsigned long); -static int vsoc_mmap(struct file *, struct vm_area_struct *); -static int vsoc_open(struct inode *, struct file *); -static int vsoc_release(struct inode *, struct file *); -static ssize_t vsoc_read(struct file *, char __user *, size_t, loff_t *); -static ssize_t vsoc_write(struct file *, const char __user *, size_t, loff_t *); -static loff_t vsoc_lseek(struct file *filp, loff_t offset, int origin); -static int -do_create_fd_scoped_permission(struct vsoc_device_region *region_p, - struct fd_scoped_permission_node *np, - struct fd_scoped_permission_arg __user *arg); -static void -do_destroy_fd_scoped_permission(struct vsoc_device_region *owner_region_p, - struct fd_scoped_permission *perm); -static long do_vsoc_describe_region(struct file *, - struct vsoc_device_region __user *); -static ssize_t vsoc_get_area(struct file *filp, __u32 *perm_off); - -/** - * Validate arguments on entry points to the driver. - */ -inline int vsoc_validate_inode(struct inode *inode) -{ - if (iminor(inode) >= vsoc_dev.layout->region_count) { - dev_err(&vsoc_dev.dev->dev, - "describe_region: invalid region %d\n", iminor(inode)); - return -ENODEV; - } - return 0; -} - -inline int vsoc_validate_filep(struct file *filp) -{ - int ret = vsoc_validate_inode(file_inode(filp)); - - if (ret) - return ret; - if (!filp->private_data) { - dev_err(&vsoc_dev.dev->dev, - "No private data on fd, region %d\n", - iminor(file_inode(filp))); - return -EBADFD; - } - return 0; -} - -/* Converts from shared memory offset to virtual address */ -static inline void *shm_off_to_virtual_addr(__u32 offset) -{ - return (void __force *)vsoc_dev.kernel_mapped_shm + offset; -} - -/* Converts from shared memory offset to physical address */ -static inline phys_addr_t shm_off_to_phys_addr(__u32 offset) -{ - return vsoc_dev.shm_phys_start + offset; -} - -/** - * Convenience functions to obtain the region from the inode or file. - * Dangerous to call before validating the inode/file. - */ -static -inline struct vsoc_device_region *vsoc_region_from_inode(struct inode *inode) -{ - return &vsoc_dev.regions[iminor(inode)]; -} - -static -inline struct vsoc_device_region *vsoc_region_from_filep(struct file *inode) -{ - return vsoc_region_from_inode(file_inode(inode)); -} - -static inline uint32_t vsoc_device_region_size(struct vsoc_device_region *r) -{ - return r->region_end_offset - r->region_begin_offset; -} - -static const struct file_operations vsoc_ops = { - .owner = THIS_MODULE, - .open = vsoc_open, - .mmap = vsoc_mmap, - .read = vsoc_read, - .unlocked_ioctl = vsoc_ioctl, - .compat_ioctl = vsoc_ioctl, - .write = vsoc_write, - .llseek = vsoc_lseek, - .release = vsoc_release, -}; - -static struct pci_device_id vsoc_id_table[] = { - {0x1af4, 0x1110, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, - {0}, -}; - -MODULE_DEVICE_TABLE(pci, vsoc_id_table); - -static void vsoc_remove_device(struct pci_dev *pdev); -static int vsoc_probe_device(struct pci_dev *pdev, - const struct pci_device_id *ent); - -static struct pci_driver vsoc_pci_driver = { - .name = "vsoc", - .id_table = vsoc_id_table, - .probe = vsoc_probe_device, - .remove = vsoc_remove_device, -}; - -static int -do_create_fd_scoped_permission(struct vsoc_device_region *region_p, - struct fd_scoped_permission_node *np, - struct fd_scoped_permission_arg __user *arg) -{ - struct file *managed_filp; - s32 managed_fd; - atomic_t *owner_ptr = NULL; - struct vsoc_device_region *managed_region_p; - - if (copy_from_user(&np->permission, - &arg->perm, sizeof(np->permission)) || - copy_from_user(&managed_fd, - &arg->managed_region_fd, sizeof(managed_fd))) { - return -EFAULT; - } - managed_filp = fdget(managed_fd).file; - /* Check that it's a valid fd, */ - if (!managed_filp || vsoc_validate_filep(managed_filp)) - return -EPERM; - /* EEXIST if the given fd already has a permission. */ - if (((struct vsoc_private_data *)managed_filp->private_data)-> - fd_scoped_permission_node) - return -EEXIST; - managed_region_p = vsoc_region_from_filep(managed_filp); - /* Check that the provided region is managed by this one */ - if (&vsoc_dev.regions[managed_region_p->managed_by] != region_p) - return -EPERM; - /* The area must be well formed and have non-zero size */ - if (np->permission.begin_offset >= np->permission.end_offset) - return -EINVAL; - /* The area must fit in the memory window */ - if (np->permission.end_offset > - vsoc_device_region_size(managed_region_p)) - return -ERANGE; - /* The area must be in the region data section */ - if (np->permission.begin_offset < - managed_region_p->offset_of_region_data) - return -ERANGE; - /* The area must be page aligned */ - if (!PAGE_ALIGNED(np->permission.begin_offset) || - !PAGE_ALIGNED(np->permission.end_offset)) - return -EINVAL; - /* Owner offset must be naturally aligned in the window */ - if (np->permission.owner_offset & - (sizeof(np->permission.owner_offset) - 1)) - return -EINVAL; - /* The owner flag must reside in the owner memory */ - if (np->permission.owner_offset + sizeof(np->permission.owner_offset) > - vsoc_device_region_size(region_p)) - return -ERANGE; - /* The owner flag must reside in the data section */ - if (np->permission.owner_offset < region_p->offset_of_region_data) - return -EINVAL; - /* The owner value must change to claim the memory */ - if (np->permission.owned_value == VSOC_REGION_FREE) - return -EINVAL; - owner_ptr = - (atomic_t *)shm_off_to_virtual_addr(region_p->region_begin_offset + - np->permission.owner_offset); - /* We've already verified that this is in the shared memory window, so - * it should be safe to write to this address. - */ - if (atomic_cmpxchg(owner_ptr, - VSOC_REGION_FREE, - np->permission.owned_value) != VSOC_REGION_FREE) { - return -EBUSY; - } - ((struct vsoc_private_data *)managed_filp->private_data)-> - fd_scoped_permission_node = np; - /* The file offset needs to be adjusted if the calling - * process did any read/write operations on the fd - * before creating the permission. - */ - if (managed_filp->f_pos) { - if (managed_filp->f_pos > np->permission.end_offset) { - /* If the offset is beyond the permission end, set it - * to the end. - */ - managed_filp->f_pos = np->permission.end_offset; - } else { - /* If the offset is within the permission interval - * keep it there otherwise reset it to zero. - */ - if (managed_filp->f_pos < np->permission.begin_offset) { - managed_filp->f_pos = 0; - } else { - managed_filp->f_pos -= - np->permission.begin_offset; - } - } - } - return 0; -} - -static void -do_destroy_fd_scoped_permission_node(struct vsoc_device_region *owner_region_p, - struct fd_scoped_permission_node *node) -{ - if (node) { - do_destroy_fd_scoped_permission(owner_region_p, - &node->permission); - mutex_lock(&vsoc_dev.mtx); - list_del(&node->list); - mutex_unlock(&vsoc_dev.mtx); - kfree(node); - } -} - -static void -do_destroy_fd_scoped_permission(struct vsoc_device_region *owner_region_p, - struct fd_scoped_permission *perm) -{ - atomic_t *owner_ptr = NULL; - int prev = 0; - - if (!perm) - return; - owner_ptr = (atomic_t *)shm_off_to_virtual_addr - (owner_region_p->region_begin_offset + perm->owner_offset); - prev = atomic_xchg(owner_ptr, VSOC_REGION_FREE); - if (prev != perm->owned_value) - dev_err(&vsoc_dev.dev->dev, - "%x-%x: owner (%s) %x: expected to be %x was %x", - perm->begin_offset, perm->end_offset, - owner_region_p->device_name, perm->owner_offset, - perm->owned_value, prev); -} - -static long do_vsoc_describe_region(struct file *filp, - struct vsoc_device_region __user *dest) -{ - struct vsoc_device_region *region_p; - int retval = vsoc_validate_filep(filp); - - if (retval) - return retval; - region_p = vsoc_region_from_filep(filp); - if (copy_to_user(dest, region_p, sizeof(*region_p))) - return -EFAULT; - return 0; -} - -/** - * Implements the inner logic of cond_wait. Copies to and from userspace are - * done in the helper function below. - */ -static int handle_vsoc_cond_wait(struct file *filp, struct vsoc_cond_wait *arg) -{ - DEFINE_WAIT(wait); - u32 region_number = iminor(file_inode(filp)); - struct vsoc_region_data *data = vsoc_dev.regions_data + region_number; - struct hrtimer_sleeper timeout, *to = NULL; - int ret = 0; - struct vsoc_device_region *region_p = vsoc_region_from_filep(filp); - atomic_t *address = NULL; - ktime_t wake_time; - - /* Ensure that the offset is aligned */ - if (arg->offset & (sizeof(uint32_t) - 1)) - return -EADDRNOTAVAIL; - /* Ensure that the offset is within shared memory */ - if (((uint64_t)arg->offset) + region_p->region_begin_offset + - sizeof(uint32_t) > region_p->region_end_offset) - return -E2BIG; - address = shm_off_to_virtual_addr(region_p->region_begin_offset + - arg->offset); - - /* Ensure that the type of wait is valid */ - switch (arg->wait_type) { - case VSOC_WAIT_IF_EQUAL: - break; - case VSOC_WAIT_IF_EQUAL_TIMEOUT: - to = &timeout; - break; - default: - return -EINVAL; - } - - if (to) { - /* Copy the user-supplied timesec into the kernel structure. - * We do things this way to flatten differences between 32 bit - * and 64 bit timespecs. - */ - if (arg->wake_time_nsec >= NSEC_PER_SEC) - return -EINVAL; - wake_time = ktime_set(arg->wake_time_sec, arg->wake_time_nsec); - - hrtimer_init_sleeper_on_stack(to, CLOCK_MONOTONIC, - HRTIMER_MODE_ABS); - hrtimer_set_expires_range_ns(&to->timer, wake_time, - current->timer_slack_ns); - } - - while (1) { - prepare_to_wait(&data->futex_wait_queue, &wait, - TASK_INTERRUPTIBLE); - /* - * Check the sentinel value after prepare_to_wait. If the value - * changes after this check the writer will call signal, - * changing the task state from INTERRUPTIBLE to RUNNING. That - * will ensure that schedule() will eventually schedule this - * task. - */ - if (atomic_read(address) != arg->value) { - ret = 0; - break; - } - if (to) { - hrtimer_sleeper_start_expires(to, HRTIMER_MODE_ABS); - if (likely(to->task)) - freezable_schedule(); - hrtimer_cancel(&to->timer); - if (!to->task) { - ret = -ETIMEDOUT; - break; - } - } else { - freezable_schedule(); - } - /* Count the number of times that we woke up. This is useful - * for unit testing. - */ - ++arg->wakes; - if (signal_pending(current)) { - ret = -EINTR; - break; - } - } - finish_wait(&data->futex_wait_queue, &wait); - if (to) - destroy_hrtimer_on_stack(&to->timer); - return ret; -} - -/** - * Handles the details of copying from/to userspace to ensure that the copies - * happen on all of the return paths of cond_wait. - */ -static int do_vsoc_cond_wait(struct file *filp, - struct vsoc_cond_wait __user *untrusted_in) -{ - struct vsoc_cond_wait arg; - int rval = 0; - - if (copy_from_user(&arg, untrusted_in, sizeof(arg))) - return -EFAULT; - /* wakes is an out parameter. Initialize it to something sensible. */ - arg.wakes = 0; - rval = handle_vsoc_cond_wait(filp, &arg); - if (copy_to_user(untrusted_in, &arg, sizeof(arg))) - return -EFAULT; - return rval; -} - -static int do_vsoc_cond_wake(struct file *filp, uint32_t offset) -{ - struct vsoc_device_region *region_p = vsoc_region_from_filep(filp); - u32 region_number = iminor(file_inode(filp)); - struct vsoc_region_data *data = vsoc_dev.regions_data + region_number; - /* Ensure that the offset is aligned */ - if (offset & (sizeof(uint32_t) - 1)) - return -EADDRNOTAVAIL; - /* Ensure that the offset is within shared memory */ - if (((uint64_t)offset) + region_p->region_begin_offset + - sizeof(uint32_t) > region_p->region_end_offset) - return -E2BIG; - /* - * TODO(b/73664181): Use multiple futex wait queues. - * We need to wake every sleeper when the condition changes. Typically - * only a single thread will be waiting on the condition, but there - * are exceptions. The worst case is about 10 threads. - */ - wake_up_interruptible_all(&data->futex_wait_queue); - return 0; -} - -static long vsoc_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) -{ - int rv = 0; - struct vsoc_device_region *region_p; - u32 reg_num; - struct vsoc_region_data *reg_data; - int retval = vsoc_validate_filep(filp); - - if (retval) - return retval; - region_p = vsoc_region_from_filep(filp); - reg_num = iminor(file_inode(filp)); - reg_data = vsoc_dev.regions_data + reg_num; - switch (cmd) { - case VSOC_CREATE_FD_SCOPED_PERMISSION: - { - struct fd_scoped_permission_node *node = NULL; - - node = kzalloc(sizeof(*node), GFP_KERNEL); - /* We can't allocate memory for the permission */ - if (!node) - return -ENOMEM; - INIT_LIST_HEAD(&node->list); - rv = do_create_fd_scoped_permission - (region_p, - node, - (struct fd_scoped_permission_arg __user *)arg); - if (!rv) { - mutex_lock(&vsoc_dev.mtx); - list_add(&node->list, &vsoc_dev.permissions); - mutex_unlock(&vsoc_dev.mtx); - } else { - kfree(node); - return rv; - } - } - break; - - case VSOC_GET_FD_SCOPED_PERMISSION: - { - struct fd_scoped_permission_node *node = - ((struct vsoc_private_data *)filp->private_data)-> - fd_scoped_permission_node; - if (!node) - return -ENOENT; - if (copy_to_user - ((struct fd_scoped_permission __user *)arg, - &node->permission, sizeof(node->permission))) - return -EFAULT; - } - break; - - case VSOC_MAYBE_SEND_INTERRUPT_TO_HOST: - if (!atomic_xchg(reg_data->outgoing_signalled, 1)) { - writel(reg_num, vsoc_dev.regs + DOORBELL); - return 0; - } else { - return -EBUSY; - } - break; - - case VSOC_SEND_INTERRUPT_TO_HOST: - writel(reg_num, vsoc_dev.regs + DOORBELL); - return 0; - case VSOC_WAIT_FOR_INCOMING_INTERRUPT: - wait_event_interruptible - (reg_data->interrupt_wait_queue, - (atomic_read(reg_data->incoming_signalled) != 0)); - break; - - case VSOC_DESCRIBE_REGION: - return do_vsoc_describe_region - (filp, - (struct vsoc_device_region __user *)arg); - - case VSOC_SELF_INTERRUPT: - atomic_set(reg_data->incoming_signalled, 1); - wake_up_interruptible(®_data->interrupt_wait_queue); - break; - - case VSOC_COND_WAIT: - return do_vsoc_cond_wait(filp, - (struct vsoc_cond_wait __user *)arg); - case VSOC_COND_WAKE: - return do_vsoc_cond_wake(filp, arg); - - default: - return -EINVAL; - } - return 0; -} - -static ssize_t vsoc_read(struct file *filp, char __user *buffer, size_t len, - loff_t *poffset) -{ - __u32 area_off; - const void *area_p; - ssize_t area_len; - int retval = vsoc_validate_filep(filp); - - if (retval) - return retval; - area_len = vsoc_get_area(filp, &area_off); - area_p = shm_off_to_virtual_addr(area_off); - area_p += *poffset; - area_len -= *poffset; - if (area_len <= 0) - return 0; - if (area_len < len) - len = area_len; - if (copy_to_user(buffer, area_p, len)) - return -EFAULT; - *poffset += len; - return len; -} - -static loff_t vsoc_lseek(struct file *filp, loff_t offset, int origin) -{ - ssize_t area_len = 0; - int retval = vsoc_validate_filep(filp); - - if (retval) - return retval; - area_len = vsoc_get_area(filp, NULL); - switch (origin) { - case SEEK_SET: - break; - - case SEEK_CUR: - if (offset > 0 && offset + filp->f_pos < 0) - return -EOVERFLOW; - offset += filp->f_pos; - break; - - case SEEK_END: - if (offset > 0 && offset + area_len < 0) - return -EOVERFLOW; - offset += area_len; - break; - - case SEEK_DATA: - if (offset >= area_len) - return -EINVAL; - if (offset < 0) - offset = 0; - break; - - case SEEK_HOLE: - /* Next hole is always the end of the region, unless offset is - * beyond that - */ - if (offset < area_len) - offset = area_len; - break; - - default: - return -EINVAL; - } - - if (offset < 0 || offset > area_len) - return -EINVAL; - filp->f_pos = offset; - - return offset; -} - -static ssize_t vsoc_write(struct file *filp, const char __user *buffer, - size_t len, loff_t *poffset) -{ - __u32 area_off; - void *area_p; - ssize_t area_len; - int retval = vsoc_validate_filep(filp); - - if (retval) - return retval; - area_len = vsoc_get_area(filp, &area_off); - area_p = shm_off_to_virtual_addr(area_off); - area_p += *poffset; - area_len -= *poffset; - if (area_len <= 0) - return 0; - if (area_len < len) - len = area_len; - if (copy_from_user(area_p, buffer, len)) - return -EFAULT; - *poffset += len; - return len; -} - -static irqreturn_t vsoc_interrupt(int irq, void *region_data_v) -{ - struct vsoc_region_data *region_data = - (struct vsoc_region_data *)region_data_v; - int reg_num = region_data - vsoc_dev.regions_data; - - if (unlikely(!region_data)) - return IRQ_NONE; - - if (unlikely(reg_num < 0 || - reg_num >= vsoc_dev.layout->region_count)) { - dev_err(&vsoc_dev.dev->dev, - "invalid irq @%p reg_num=0x%04x\n", - region_data, reg_num); - return IRQ_NONE; - } - if (unlikely(vsoc_dev.regions_data + reg_num != region_data)) { - dev_err(&vsoc_dev.dev->dev, - "irq not aligned @%p reg_num=0x%04x\n", - region_data, reg_num); - return IRQ_NONE; - } - wake_up_interruptible(®ion_data->interrupt_wait_queue); - return IRQ_HANDLED; -} - -static int vsoc_probe_device(struct pci_dev *pdev, - const struct pci_device_id *ent) -{ - int result; - int i; - resource_size_t reg_size; - dev_t devt; - - vsoc_dev.dev = pdev; - result = pci_enable_device(pdev); - if (result) { - dev_err(&pdev->dev, - "pci_enable_device failed %s: error %d\n", - pci_name(pdev), result); - return result; - } - vsoc_dev.enabled_device = true; - result = pci_request_regions(pdev, "vsoc"); - if (result < 0) { - dev_err(&pdev->dev, "pci_request_regions failed\n"); - vsoc_remove_device(pdev); - return -EBUSY; - } - vsoc_dev.requested_regions = true; - /* Set up the control registers in BAR 0 */ - reg_size = pci_resource_len(pdev, REGISTER_BAR); - if (reg_size > MAX_REGISTER_BAR_LEN) - vsoc_dev.regs = - pci_iomap(pdev, REGISTER_BAR, MAX_REGISTER_BAR_LEN); - else - vsoc_dev.regs = pci_iomap(pdev, REGISTER_BAR, reg_size); - - if (!vsoc_dev.regs) { - dev_err(&pdev->dev, - "cannot map registers of size %zu\n", - (size_t)reg_size); - vsoc_remove_device(pdev); - return -EBUSY; - } - - /* Map the shared memory in BAR 2 */ - vsoc_dev.shm_phys_start = pci_resource_start(pdev, SHARED_MEMORY_BAR); - vsoc_dev.shm_size = pci_resource_len(pdev, SHARED_MEMORY_BAR); - - dev_info(&pdev->dev, "shared memory @ DMA %pa size=0x%zx\n", - &vsoc_dev.shm_phys_start, vsoc_dev.shm_size); - vsoc_dev.kernel_mapped_shm = pci_iomap_wc(pdev, SHARED_MEMORY_BAR, 0); - if (!vsoc_dev.kernel_mapped_shm) { - dev_err(&vsoc_dev.dev->dev, "cannot iomap region\n"); - vsoc_remove_device(pdev); - return -EBUSY; - } - - vsoc_dev.layout = (struct vsoc_shm_layout_descriptor __force *) - vsoc_dev.kernel_mapped_shm; - dev_info(&pdev->dev, "major_version: %d\n", - vsoc_dev.layout->major_version); - dev_info(&pdev->dev, "minor_version: %d\n", - vsoc_dev.layout->minor_version); - dev_info(&pdev->dev, "size: 0x%x\n", vsoc_dev.layout->size); - dev_info(&pdev->dev, "regions: %d\n", vsoc_dev.layout->region_count); - if (vsoc_dev.layout->major_version != - CURRENT_VSOC_LAYOUT_MAJOR_VERSION) { - dev_err(&vsoc_dev.dev->dev, - "driver supports only major_version %d\n", - CURRENT_VSOC_LAYOUT_MAJOR_VERSION); - vsoc_remove_device(pdev); - return -EBUSY; - } - result = alloc_chrdev_region(&devt, 0, vsoc_dev.layout->region_count, - VSOC_DEV_NAME); - if (result) { - dev_err(&vsoc_dev.dev->dev, "alloc_chrdev_region failed\n"); - vsoc_remove_device(pdev); - return -EBUSY; - } - vsoc_dev.major = MAJOR(devt); - cdev_init(&vsoc_dev.cdev, &vsoc_ops); - vsoc_dev.cdev.owner = THIS_MODULE; - result = cdev_add(&vsoc_dev.cdev, devt, vsoc_dev.layout->region_count); - if (result) { - dev_err(&vsoc_dev.dev->dev, "cdev_add error\n"); - vsoc_remove_device(pdev); - return -EBUSY; - } - vsoc_dev.cdev_added = true; - vsoc_dev.class = class_create(THIS_MODULE, VSOC_DEV_NAME); - if (IS_ERR(vsoc_dev.class)) { - dev_err(&vsoc_dev.dev->dev, "class_create failed\n"); - vsoc_remove_device(pdev); - return PTR_ERR(vsoc_dev.class); - } - vsoc_dev.class_added = true; - vsoc_dev.regions = (struct vsoc_device_region __force *) - ((void *)vsoc_dev.layout + - vsoc_dev.layout->vsoc_region_desc_offset); - vsoc_dev.msix_entries = - kcalloc(vsoc_dev.layout->region_count, - sizeof(vsoc_dev.msix_entries[0]), GFP_KERNEL); - if (!vsoc_dev.msix_entries) { - dev_err(&vsoc_dev.dev->dev, - "unable to allocate msix_entries\n"); - vsoc_remove_device(pdev); - return -ENOSPC; - } - vsoc_dev.regions_data = - kcalloc(vsoc_dev.layout->region_count, - sizeof(vsoc_dev.regions_data[0]), GFP_KERNEL); - if (!vsoc_dev.regions_data) { - dev_err(&vsoc_dev.dev->dev, - "unable to allocate regions' data\n"); - vsoc_remove_device(pdev); - return -ENOSPC; - } - for (i = 0; i < vsoc_dev.layout->region_count; ++i) - vsoc_dev.msix_entries[i].entry = i; - - result = pci_enable_msix_exact(vsoc_dev.dev, vsoc_dev.msix_entries, - vsoc_dev.layout->region_count); - if (result) { - dev_info(&pdev->dev, "pci_enable_msix failed: %d\n", result); - vsoc_remove_device(pdev); - return -ENOSPC; - } - /* Check that all regions are well formed */ - for (i = 0; i < vsoc_dev.layout->region_count; ++i) { - const struct vsoc_device_region *region = vsoc_dev.regions + i; - - if (!PAGE_ALIGNED(region->region_begin_offset) || - !PAGE_ALIGNED(region->region_end_offset)) { - dev_err(&vsoc_dev.dev->dev, - "region %d not aligned (%x:%x)", i, - region->region_begin_offset, - region->region_end_offset); - vsoc_remove_device(pdev); - return -EFAULT; - } - if (region->region_begin_offset >= region->region_end_offset || - region->region_end_offset > vsoc_dev.shm_size) { - dev_err(&vsoc_dev.dev->dev, - "region %d offsets are wrong: %x %x %zx", - i, region->region_begin_offset, - region->region_end_offset, vsoc_dev.shm_size); - vsoc_remove_device(pdev); - return -EFAULT; - } - if (region->managed_by >= vsoc_dev.layout->region_count) { - dev_err(&vsoc_dev.dev->dev, - "region %d has invalid owner: %u", - i, region->managed_by); - vsoc_remove_device(pdev); - return -EFAULT; - } - } - vsoc_dev.msix_enabled = true; - for (i = 0; i < vsoc_dev.layout->region_count; ++i) { - const struct vsoc_device_region *region = vsoc_dev.regions + i; - size_t name_sz = sizeof(vsoc_dev.regions_data[i].name) - 1; - const struct vsoc_signal_table_layout *h_to_g_signal_table = - ®ion->host_to_guest_signal_table; - const struct vsoc_signal_table_layout *g_to_h_signal_table = - ®ion->guest_to_host_signal_table; - - vsoc_dev.regions_data[i].name[name_sz] = '\0'; - memcpy(vsoc_dev.regions_data[i].name, region->device_name, - name_sz); - dev_info(&pdev->dev, "region %d name=%s\n", - i, vsoc_dev.regions_data[i].name); - init_waitqueue_head - (&vsoc_dev.regions_data[i].interrupt_wait_queue); - init_waitqueue_head(&vsoc_dev.regions_data[i].futex_wait_queue); - vsoc_dev.regions_data[i].incoming_signalled = - shm_off_to_virtual_addr(region->region_begin_offset) + - h_to_g_signal_table->interrupt_signalled_offset; - vsoc_dev.regions_data[i].outgoing_signalled = - shm_off_to_virtual_addr(region->region_begin_offset) + - g_to_h_signal_table->interrupt_signalled_offset; - result = request_irq(vsoc_dev.msix_entries[i].vector, - vsoc_interrupt, 0, - vsoc_dev.regions_data[i].name, - vsoc_dev.regions_data + i); - if (result) { - dev_info(&pdev->dev, - "request_irq failed irq=%d vector=%d\n", - i, vsoc_dev.msix_entries[i].vector); - vsoc_remove_device(pdev); - return -ENOSPC; - } - vsoc_dev.regions_data[i].irq_requested = true; - if (!device_create(vsoc_dev.class, NULL, - MKDEV(vsoc_dev.major, i), - NULL, vsoc_dev.regions_data[i].name)) { - dev_err(&vsoc_dev.dev->dev, "device_create failed\n"); - vsoc_remove_device(pdev); - return -EBUSY; - } - vsoc_dev.regions_data[i].device_created = true; - } - return 0; -} - -/* - * This should undo all of the allocations in the probe function in reverse - * order. - * - * Notes: - * - * The device may have been partially initialized, so double check - * that the allocations happened. - * - * This function may be called multiple times, so mark resources as freed - * as they are deallocated. - */ -static void vsoc_remove_device(struct pci_dev *pdev) -{ - int i; - /* - * pdev is the first thing to be set on probe and the last thing - * to be cleared here. If it's NULL then there is no cleanup. - */ - if (!pdev || !vsoc_dev.dev) - return; - dev_info(&pdev->dev, "remove_device\n"); - if (vsoc_dev.regions_data) { - for (i = 0; i < vsoc_dev.layout->region_count; ++i) { - if (vsoc_dev.regions_data[i].device_created) { - device_destroy(vsoc_dev.class, - MKDEV(vsoc_dev.major, i)); - vsoc_dev.regions_data[i].device_created = false; - } - if (vsoc_dev.regions_data[i].irq_requested) - free_irq(vsoc_dev.msix_entries[i].vector, NULL); - vsoc_dev.regions_data[i].irq_requested = false; - } - kfree(vsoc_dev.regions_data); - vsoc_dev.regions_data = NULL; - } - if (vsoc_dev.msix_enabled) { - pci_disable_msix(pdev); - vsoc_dev.msix_enabled = false; - } - kfree(vsoc_dev.msix_entries); - vsoc_dev.msix_entries = NULL; - vsoc_dev.regions = NULL; - if (vsoc_dev.class_added) { - class_destroy(vsoc_dev.class); - vsoc_dev.class_added = false; - } - if (vsoc_dev.cdev_added) { - cdev_del(&vsoc_dev.cdev); - vsoc_dev.cdev_added = false; - } - if (vsoc_dev.major && vsoc_dev.layout) { - unregister_chrdev_region(MKDEV(vsoc_dev.major, 0), - vsoc_dev.layout->region_count); - vsoc_dev.major = 0; - } - vsoc_dev.layout = NULL; - if (vsoc_dev.kernel_mapped_shm) { - pci_iounmap(pdev, vsoc_dev.kernel_mapped_shm); - vsoc_dev.kernel_mapped_shm = NULL; - } - if (vsoc_dev.regs) { - pci_iounmap(pdev, vsoc_dev.regs); - vsoc_dev.regs = NULL; - } - if (vsoc_dev.requested_regions) { - pci_release_regions(pdev); - vsoc_dev.requested_regions = false; - } - if (vsoc_dev.enabled_device) { - pci_disable_device(pdev); - vsoc_dev.enabled_device = false; - } - /* Do this last: it indicates that the device is not initialized. */ - vsoc_dev.dev = NULL; -} - -static void __exit vsoc_cleanup_module(void) -{ - vsoc_remove_device(vsoc_dev.dev); - pci_unregister_driver(&vsoc_pci_driver); -} - -static int __init vsoc_init_module(void) -{ - int err = -ENOMEM; - - INIT_LIST_HEAD(&vsoc_dev.permissions); - mutex_init(&vsoc_dev.mtx); - - err = pci_register_driver(&vsoc_pci_driver); - if (err < 0) - return err; - return 0; -} - -static int vsoc_open(struct inode *inode, struct file *filp) -{ - /* Can't use vsoc_validate_filep because filp is still incomplete */ - int ret = vsoc_validate_inode(inode); - - if (ret) - return ret; - filp->private_data = - kzalloc(sizeof(struct vsoc_private_data), GFP_KERNEL); - if (!filp->private_data) - return -ENOMEM; - return 0; -} - -static int vsoc_release(struct inode *inode, struct file *filp) -{ - struct vsoc_private_data *private_data = NULL; - struct fd_scoped_permission_node *node = NULL; - struct vsoc_device_region *owner_region_p = NULL; - int retval = vsoc_validate_filep(filp); - - if (retval) - return retval; - private_data = (struct vsoc_private_data *)filp->private_data; - if (!private_data) - return 0; - - node = private_data->fd_scoped_permission_node; - if (node) { - owner_region_p = vsoc_region_from_inode(inode); - if (owner_region_p->managed_by != VSOC_REGION_WHOLE) { - owner_region_p = - &vsoc_dev.regions[owner_region_p->managed_by]; - } - do_destroy_fd_scoped_permission_node(owner_region_p, node); - private_data->fd_scoped_permission_node = NULL; - } - kfree(private_data); - filp->private_data = NULL; - - return 0; -} - -/* - * Returns the device relative offset and length of the area specified by the - * fd scoped permission. If there is no fd scoped permission set, a default - * permission covering the entire region is assumed, unless the region is owned - * by another one, in which case the default is a permission with zero size. - */ -static ssize_t vsoc_get_area(struct file *filp, __u32 *area_offset) -{ - __u32 off = 0; - ssize_t length = 0; - struct vsoc_device_region *region_p; - struct fd_scoped_permission *perm; - - region_p = vsoc_region_from_filep(filp); - off = region_p->region_begin_offset; - perm = &((struct vsoc_private_data *)filp->private_data)-> - fd_scoped_permission_node->permission; - if (perm) { - off += perm->begin_offset; - length = perm->end_offset - perm->begin_offset; - } else if (region_p->managed_by == VSOC_REGION_WHOLE) { - /* No permission set and the regions is not owned by another, - * default to full region access. - */ - length = vsoc_device_region_size(region_p); - } else { - /* return zero length, access is denied. */ - length = 0; - } - if (area_offset) - *area_offset = off; - return length; -} - -static int vsoc_mmap(struct file *filp, struct vm_area_struct *vma) -{ - unsigned long len = vma->vm_end - vma->vm_start; - __u32 area_off; - phys_addr_t mem_off; - ssize_t area_len; - int retval = vsoc_validate_filep(filp); - - if (retval) - return retval; - area_len = vsoc_get_area(filp, &area_off); - /* Add the requested offset */ - area_off += (vma->vm_pgoff << PAGE_SHIFT); - area_len -= (vma->vm_pgoff << PAGE_SHIFT); - if (area_len < len) - return -EINVAL; - vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot); - mem_off = shm_off_to_phys_addr(area_off); - if (io_remap_pfn_range(vma, vma->vm_start, mem_off >> PAGE_SHIFT, - len, vma->vm_page_prot)) - return -EAGAIN; - return 0; -} - -module_init(vsoc_init_module); -module_exit(vsoc_cleanup_module); - -MODULE_LICENSE("GPL"); -MODULE_AUTHOR("Greg Hartman <ghartman@google.com>"); -MODULE_DESCRIPTION("VSoC interpretation of QEmu's ivshmem device"); -MODULE_VERSION("1.0"); diff --git a/drivers/staging/greybus/audio_manager.c b/drivers/staging/greybus/audio_manager.c index 9b19ea9d3fa1..9a3f7c034ab4 100644 --- a/drivers/staging/greybus/audio_manager.c +++ b/drivers/staging/greybus/audio_manager.c @@ -92,8 +92,8 @@ void gb_audio_manager_remove_all(void) list_for_each_entry_safe(module, next, &modules_list, list) { list_del(&module->list); - kobject_put(&module->kobj); ida_simple_remove(&module_id, module->id); + kobject_put(&module->kobj); } is_empty = list_empty(&modules_list); diff --git a/drivers/staging/greybus/tools/loopback_test.c b/drivers/staging/greybus/tools/loopback_test.c index ba6f905f26fa..69c6dce9be31 100644 --- a/drivers/staging/greybus/tools/loopback_test.c +++ b/drivers/staging/greybus/tools/loopback_test.c @@ -19,6 +19,7 @@ #include <signal.h> #define MAX_NUM_DEVICES 10 +#define MAX_SYSFS_PREFIX 0x80 #define MAX_SYSFS_PATH 0x200 #define CSV_MAX_LINE 0x1000 #define SYSFS_MAX_INT 0x20 @@ -67,7 +68,7 @@ struct loopback_results { }; struct loopback_device { - char name[MAX_SYSFS_PATH]; + char name[MAX_STR_LEN]; char sysfs_entry[MAX_SYSFS_PATH]; char debugfs_entry[MAX_SYSFS_PATH]; struct loopback_results results; @@ -93,8 +94,8 @@ struct loopback_test { int stop_all; int poll_count; char test_name[MAX_STR_LEN]; - char sysfs_prefix[MAX_SYSFS_PATH]; - char debugfs_prefix[MAX_SYSFS_PATH]; + char sysfs_prefix[MAX_SYSFS_PREFIX]; + char debugfs_prefix[MAX_SYSFS_PREFIX]; struct timespec poll_timeout; struct loopback_device devices[MAX_NUM_DEVICES]; struct loopback_results aggregate_results; @@ -637,7 +638,7 @@ baddir: static int open_poll_files(struct loopback_test *t) { struct loopback_device *dev; - char buf[MAX_STR_LEN]; + char buf[MAX_SYSFS_PATH + MAX_STR_LEN]; char dummy; int fds_idx = 0; int i; @@ -655,7 +656,7 @@ static int open_poll_files(struct loopback_test *t) goto err; } read(t->fds[fds_idx].fd, &dummy, 1); - t->fds[fds_idx].events = EPOLLERR|EPOLLPRI; + t->fds[fds_idx].events = POLLERR | POLLPRI; t->fds[fds_idx].revents = 0; fds_idx++; } @@ -748,7 +749,7 @@ static int wait_for_complete(struct loopback_test *t) } for (i = 0; i < t->poll_count; i++) { - if (t->fds[i].revents & EPOLLPRI) { + if (t->fds[i].revents & POLLPRI) { /* Dummy read to clear the event */ read(t->fds[i].fd, &dummy, 1); number_of_events++; @@ -907,10 +908,10 @@ int main(int argc, char *argv[]) t.iteration_max = atoi(optarg); break; case 'S': - snprintf(t.sysfs_prefix, MAX_SYSFS_PATH, "%s", optarg); + snprintf(t.sysfs_prefix, MAX_SYSFS_PREFIX, "%s", optarg); break; case 'D': - snprintf(t.debugfs_prefix, MAX_SYSFS_PATH, "%s", optarg); + snprintf(t.debugfs_prefix, MAX_SYSFS_PREFIX, "%s", optarg); break; case 'm': t.mask = atol(optarg); @@ -961,10 +962,10 @@ int main(int argc, char *argv[]) } if (!strcmp(t.sysfs_prefix, "")) - snprintf(t.sysfs_prefix, MAX_SYSFS_PATH, "%s", sysfs_prefix); + snprintf(t.sysfs_prefix, MAX_SYSFS_PREFIX, "%s", sysfs_prefix); if (!strcmp(t.debugfs_prefix, "")) - snprintf(t.debugfs_prefix, MAX_SYSFS_PATH, "%s", debugfs_prefix); + snprintf(t.debugfs_prefix, MAX_SYSFS_PREFIX, "%s", debugfs_prefix); ret = find_loopback_devices(&t); if (ret) diff --git a/drivers/staging/media/hantro/hantro_drv.c b/drivers/staging/media/hantro/hantro_drv.c index 97c615a2f057..c98835326135 100644 --- a/drivers/staging/media/hantro/hantro_drv.c +++ b/drivers/staging/media/hantro/hantro_drv.c @@ -558,13 +558,13 @@ static int hantro_attach_func(struct hantro_dev *vpu, goto err_rel_entity1; /* Connect the three entities */ - ret = media_create_pad_link(&func->vdev.entity, 0, &func->proc, 1, + ret = media_create_pad_link(&func->vdev.entity, 0, &func->proc, 0, MEDIA_LNK_FL_IMMUTABLE | MEDIA_LNK_FL_ENABLED); if (ret) goto err_rel_entity2; - ret = media_create_pad_link(&func->proc, 0, &func->sink, 0, + ret = media_create_pad_link(&func->proc, 1, &func->sink, 0, MEDIA_LNK_FL_IMMUTABLE | MEDIA_LNK_FL_ENABLED); if (ret) diff --git a/drivers/staging/rtl8188eu/os_dep/ioctl_linux.c b/drivers/staging/rtl8188eu/os_dep/ioctl_linux.c index 9b6ea86d1dcf..ba53959e1303 100644 --- a/drivers/staging/rtl8188eu/os_dep/ioctl_linux.c +++ b/drivers/staging/rtl8188eu/os_dep/ioctl_linux.c @@ -2009,21 +2009,16 @@ static int wpa_supplicant_ioctl(struct net_device *dev, struct iw_point *p) struct ieee_param *param; uint ret = 0; - if (p->length < sizeof(struct ieee_param) || !p->pointer) { - ret = -EINVAL; - goto out; - } + if (!p->pointer || p->length != sizeof(struct ieee_param)) + return -EINVAL; param = (struct ieee_param *)rtw_malloc(p->length); - if (!param) { - ret = -ENOMEM; - goto out; - } + if (!param) + return -ENOMEM; if (copy_from_user(param, p->pointer, p->length)) { kfree(param); - ret = -EFAULT; - goto out; + return -EFAULT; } switch (param->cmd) { @@ -2054,9 +2049,6 @@ static int wpa_supplicant_ioctl(struct net_device *dev, struct iw_point *p) ret = -EFAULT; kfree(param); - -out: - return ret; } @@ -2791,26 +2783,19 @@ static int rtw_hostapd_ioctl(struct net_device *dev, struct iw_point *p) * so, we just check hw_init_completed */ - if (!padapter->hw_init_completed) { - ret = -EPERM; - goto out; - } + if (!padapter->hw_init_completed) + return -EPERM; - if (!p->pointer) { - ret = -EINVAL; - goto out; - } + if (!p->pointer || p->length != sizeof(struct ieee_param)) + return -EINVAL; param = (struct ieee_param *)rtw_malloc(p->length); - if (!param) { - ret = -ENOMEM; - goto out; - } + if (!param) + return -ENOMEM; if (copy_from_user(param, p->pointer, p->length)) { kfree(param); - ret = -EFAULT; - goto out; + return -EFAULT; } switch (param->cmd) { @@ -2865,7 +2850,6 @@ static int rtw_hostapd_ioctl(struct net_device *dev, struct iw_point *p) if (ret == 0 && copy_to_user(p->pointer, param, p->length)) ret = -EFAULT; kfree(param); -out: return ret; } #endif diff --git a/drivers/staging/rtl8188eu/os_dep/usb_intf.c b/drivers/staging/rtl8188eu/os_dep/usb_intf.c index b5d42f411dd8..845c8817281c 100644 --- a/drivers/staging/rtl8188eu/os_dep/usb_intf.c +++ b/drivers/staging/rtl8188eu/os_dep/usb_intf.c @@ -38,6 +38,7 @@ static const struct usb_device_id rtw_usb_id_tbl[] = { {USB_DEVICE(0x2001, 0x331B)}, /* D-Link DWA-121 rev B1 */ {USB_DEVICE(0x2357, 0x010c)}, /* TP-Link TL-WN722N v2 */ {USB_DEVICE(0x2357, 0x0111)}, /* TP-Link TL-WN727N v5.21 */ + {USB_DEVICE(0x2C4E, 0x0102)}, /* MERCUSYS MW150US v2 */ {USB_DEVICE(0x0df6, 0x0076)}, /* Sitecom N150 v2 */ {USB_DEVICE(USB_VENDER_ID_REALTEK, 0xffef)}, /* Rosewill RNX-N150NUB */ {} /* Terminating entry */ diff --git a/drivers/staging/rtl8723bs/hal/rtl8723bs_xmit.c b/drivers/staging/rtl8723bs/hal/rtl8723bs_xmit.c index b44e902ed338..b6d56cfb0a19 100644 --- a/drivers/staging/rtl8723bs/hal/rtl8723bs_xmit.c +++ b/drivers/staging/rtl8723bs/hal/rtl8723bs_xmit.c @@ -476,14 +476,13 @@ int rtl8723bs_xmit_thread(void *context) s32 ret; struct adapter *padapter; struct xmit_priv *pxmitpriv; - u8 thread_name[20] = "RTWHALXT"; - + u8 thread_name[20]; ret = _SUCCESS; padapter = context; pxmitpriv = &padapter->xmitpriv; - rtw_sprintf(thread_name, 20, "%s-"ADPT_FMT, thread_name, ADPT_ARG(padapter)); + rtw_sprintf(thread_name, 20, "RTWHALXT-" ADPT_FMT, ADPT_ARG(padapter)); thread_enter(thread_name); DBG_871X("start "FUNC_ADPT_FMT"\n", FUNC_ADPT_ARG(padapter)); diff --git a/drivers/staging/rtl8723bs/os_dep/ioctl_linux.c b/drivers/staging/rtl8723bs/os_dep/ioctl_linux.c index db6528a01229..9b9038e7deb1 100644 --- a/drivers/staging/rtl8723bs/os_dep/ioctl_linux.c +++ b/drivers/staging/rtl8723bs/os_dep/ioctl_linux.c @@ -3373,21 +3373,16 @@ static int wpa_supplicant_ioctl(struct net_device *dev, struct iw_point *p) /* down(&ieee->wx_sem); */ - if (p->length < sizeof(struct ieee_param) || !p->pointer) { - ret = -EINVAL; - goto out; - } + if (!p->pointer || p->length != sizeof(struct ieee_param)) + return -EINVAL; param = rtw_malloc(p->length); - if (param == NULL) { - ret = -ENOMEM; - goto out; - } + if (param == NULL) + return -ENOMEM; if (copy_from_user(param, p->pointer, p->length)) { kfree(param); - ret = -EFAULT; - goto out; + return -EFAULT; } switch (param->cmd) { @@ -3421,12 +3416,8 @@ static int wpa_supplicant_ioctl(struct net_device *dev, struct iw_point *p) kfree(param); -out: - /* up(&ieee->wx_sem); */ - return ret; - } static int rtw_set_encryption(struct net_device *dev, struct ieee_param *param, u32 param_len) @@ -4200,28 +4191,19 @@ static int rtw_hostapd_ioctl(struct net_device *dev, struct iw_point *p) * so, we just check hw_init_completed */ - if (!padapter->hw_init_completed) { - ret = -EPERM; - goto out; - } - + if (!padapter->hw_init_completed) + return -EPERM; - /* if (p->length < sizeof(struct ieee_param) || !p->pointer) { */ - if (!p->pointer) { - ret = -EINVAL; - goto out; - } + if (!p->pointer || p->length != sizeof(*param)) + return -EINVAL; param = rtw_malloc(p->length); - if (param == NULL) { - ret = -ENOMEM; - goto out; - } + if (param == NULL) + return -ENOMEM; if (copy_from_user(param, p->pointer, p->length)) { kfree(param); - ret = -EFAULT; - goto out; + return -EFAULT; } /* DBG_871X("%s, cmd =%d\n", __func__, param->cmd); */ @@ -4321,13 +4303,8 @@ static int rtw_hostapd_ioctl(struct net_device *dev, struct iw_point *p) if (ret == 0 && copy_to_user(p->pointer, param, p->length)) ret = -EFAULT; - kfree(param); - -out: - return ret; - } static int rtw_wx_set_priv(struct net_device *dev, diff --git a/drivers/staging/speakup/main.c b/drivers/staging/speakup/main.c index 488f2539aa9a..81ecfd1a200d 100644 --- a/drivers/staging/speakup/main.c +++ b/drivers/staging/speakup/main.c @@ -561,7 +561,7 @@ static u_long get_word(struct vc_data *vc) return 0; } else if (tmpx < vc->vc_cols - 2 && (ch == SPACE || ch == 0 || (ch < 0x100 && IS_WDLM(ch))) && - get_char(vc, (u_short *)&tmp_pos + 1, &temp) > SPACE) { + get_char(vc, (u_short *)tmp_pos + 1, &temp) > SPACE) { tmp_pos += 2; tmpx++; } else { diff --git a/drivers/staging/speakup/selection.c b/drivers/staging/speakup/selection.c index a8b4d0c5ab7e..032f3264fba1 100644 --- a/drivers/staging/speakup/selection.c +++ b/drivers/staging/speakup/selection.c @@ -51,9 +51,7 @@ static void __speakup_set_selection(struct work_struct *work) goto unref; } - console_lock(); set_selection_kernel(&sel, tty); - console_unlock(); unref: tty_kref_put(tty); diff --git a/drivers/staging/vt6656/dpc.c b/drivers/staging/vt6656/dpc.c index 821aae8ca402..a0b60e7d1086 100644 --- a/drivers/staging/vt6656/dpc.c +++ b/drivers/staging/vt6656/dpc.c @@ -98,7 +98,7 @@ int vnt_rx_data(struct vnt_private *priv, struct vnt_rcb *ptr_rcb, vnt_rf_rssi_to_dbm(priv, tail->rssi, &rx_dbm); - priv->bb_pre_ed_rssi = (u8)rx_dbm + 1; + priv->bb_pre_ed_rssi = (u8)-rx_dbm + 1; priv->current_rssi = priv->bb_pre_ed_rssi; skb_pull(skb, sizeof(*head)); diff --git a/drivers/staging/wfx/Documentation/devicetree/bindings/net/wireless/siliabs,wfx.txt b/drivers/staging/wfx/Documentation/devicetree/bindings/net/wireless/siliabs,wfx.txt index 26de6762b942..081d58abd5ac 100644 --- a/drivers/staging/wfx/Documentation/devicetree/bindings/net/wireless/siliabs,wfx.txt +++ b/drivers/staging/wfx/Documentation/devicetree/bindings/net/wireless/siliabs,wfx.txt @@ -93,5 +93,5 @@ Some properties are recognized either by SPI and SDIO versions: Must contains 64 hexadecimal digits. Not supported in current version. WFx driver also supports `mac-address` and `local-mac-address` as described in -Documentation/devicetree/binding/net/ethernet.txt +Documentation/devicetree/bindings/net/ethernet.txt diff --git a/drivers/staging/wfx/hif_tx.c b/drivers/staging/wfx/hif_tx.c index 2428363371fa..77bca43aca42 100644 --- a/drivers/staging/wfx/hif_tx.c +++ b/drivers/staging/wfx/hif_tx.c @@ -140,6 +140,7 @@ int hif_shutdown(struct wfx_dev *wdev) else control_reg_write(wdev, 0); mutex_unlock(&wdev->hif_cmd.lock); + mutex_unlock(&wdev->hif_cmd.key_renew_lock); kfree(hif); return ret; } @@ -289,7 +290,7 @@ int hif_stop_scan(struct wfx_vif *wvif) } int hif_join(struct wfx_vif *wvif, const struct ieee80211_bss_conf *conf, - const struct ieee80211_channel *channel, const u8 *ssidie) + struct ieee80211_channel *channel, const u8 *ssid, int ssidlen) { int ret; struct hif_msg *hif; @@ -307,9 +308,9 @@ int hif_join(struct wfx_vif *wvif, const struct ieee80211_bss_conf *conf, body->basic_rate_set = cpu_to_le32(wfx_rate_mask_to_hw(wvif->wdev, conf->basic_rates)); memcpy(body->bssid, conf->bssid, sizeof(body->bssid)); - if (!conf->ibss_joined && ssidie) { - body->ssid_length = cpu_to_le32(ssidie[1]); - memcpy(body->ssid, &ssidie[2], ssidie[1]); + if (!conf->ibss_joined && ssid) { + body->ssid_length = cpu_to_le32(ssidlen); + memcpy(body->ssid, ssid, ssidlen); } wfx_fill_header(hif, wvif->id, HIF_REQ_ID_JOIN, sizeof(*body)); ret = wfx_cmd_send(wvif->wdev, hif, NULL, 0, false); @@ -427,9 +428,9 @@ int hif_start(struct wfx_vif *wvif, const struct ieee80211_bss_conf *conf, struct hif_msg *hif; struct hif_req_start *body = wfx_alloc_hif(sizeof(*body), &hif); - body->dtim_period = conf->dtim_period, - body->short_preamble = conf->use_short_preamble, - body->channel_number = cpu_to_le16(channel->hw_value), + body->dtim_period = conf->dtim_period; + body->short_preamble = conf->use_short_preamble; + body->channel_number = cpu_to_le16(channel->hw_value); body->beacon_interval = cpu_to_le32(conf->beacon_int); body->basic_rate_set = cpu_to_le32(wfx_rate_mask_to_hw(wvif->wdev, conf->basic_rates)); diff --git a/drivers/staging/wfx/hif_tx.h b/drivers/staging/wfx/hif_tx.h index 20977e461718..f8520a14c14c 100644 --- a/drivers/staging/wfx/hif_tx.h +++ b/drivers/staging/wfx/hif_tx.h @@ -46,7 +46,7 @@ int hif_scan(struct wfx_vif *wvif, struct cfg80211_scan_request *req80211, int chan_start, int chan_num); int hif_stop_scan(struct wfx_vif *wvif); int hif_join(struct wfx_vif *wvif, const struct ieee80211_bss_conf *conf, - const struct ieee80211_channel *channel, const u8 *ssidie); + struct ieee80211_channel *channel, const u8 *ssid, int ssidlen); int hif_set_pm(struct wfx_vif *wvif, bool ps, int dynamic_ps_timeout); int hif_set_bss_params(struct wfx_vif *wvif, const struct hif_req_set_bss_params *arg); diff --git a/drivers/staging/wfx/hif_tx_mib.h b/drivers/staging/wfx/hif_tx_mib.h index bf3769c2a9b6..26b1406f9f6c 100644 --- a/drivers/staging/wfx/hif_tx_mib.h +++ b/drivers/staging/wfx/hif_tx_mib.h @@ -191,10 +191,10 @@ static inline int hif_set_block_ack_policy(struct wfx_vif *wvif, } static inline int hif_set_association_mode(struct wfx_vif *wvif, - struct ieee80211_bss_conf *info, - struct ieee80211_sta_ht_cap *ht_cap) + struct ieee80211_bss_conf *info) { int basic_rates = wfx_rate_mask_to_hw(wvif->wdev, info->basic_rates); + struct ieee80211_sta *sta = NULL; struct hif_mib_set_association_mode val = { .preambtype_use = 1, .mode = 1, @@ -204,12 +204,17 @@ static inline int hif_set_association_mode(struct wfx_vif *wvif, .basic_rate_set = cpu_to_le32(basic_rates) }; + rcu_read_lock(); // protect sta + if (info->bssid && !info->ibss_joined) + sta = ieee80211_find_sta(wvif->vif, info->bssid); + // FIXME: it is strange to not retrieve all information from bss_info - if (ht_cap && ht_cap->ht_supported) { - val.mpdu_start_spacing = ht_cap->ampdu_density; + if (sta && sta->ht_cap.ht_supported) { + val.mpdu_start_spacing = sta->ht_cap.ampdu_density; if (!(info->ht_operation_mode & IEEE80211_HT_OP_MODE_NON_GF_STA_PRSNT)) - val.greenfield = !!(ht_cap->cap & IEEE80211_HT_CAP_GRN_FLD); + val.greenfield = !!(sta->ht_cap.cap & IEEE80211_HT_CAP_GRN_FLD); } + rcu_read_unlock(); return hif_write_mib(wvif->wdev, wvif->id, HIF_MIB_ID_SET_ASSOCIATION_MODE, &val, sizeof(val)); diff --git a/drivers/staging/wfx/sta.c b/drivers/staging/wfx/sta.c index 03d0f224ffdb..af4f4bbd0572 100644 --- a/drivers/staging/wfx/sta.c +++ b/drivers/staging/wfx/sta.c @@ -491,9 +491,11 @@ static void wfx_set_mfp(struct wfx_vif *wvif, static void wfx_do_join(struct wfx_vif *wvif) { int ret; - const u8 *ssidie; struct ieee80211_bss_conf *conf = &wvif->vif->bss_conf; struct cfg80211_bss *bss = NULL; + u8 ssid[IEEE80211_MAX_SSID_LEN]; + const u8 *ssidie = NULL; + int ssidlen = 0; wfx_tx_lock_flush(wvif->wdev); @@ -514,11 +516,14 @@ static void wfx_do_join(struct wfx_vif *wvif) if (!wvif->beacon_int) wvif->beacon_int = 1; - rcu_read_lock(); + rcu_read_lock(); // protect ssidie if (!conf->ibss_joined) ssidie = ieee80211_bss_get_ie(bss, WLAN_EID_SSID); - else - ssidie = NULL; + if (ssidie) { + ssidlen = ssidie[1]; + memcpy(ssid, &ssidie[2], ssidie[1]); + } + rcu_read_unlock(); wfx_tx_flush(wvif->wdev); @@ -527,10 +532,8 @@ static void wfx_do_join(struct wfx_vif *wvif) wfx_set_mfp(wvif, bss); - /* Perform actual join */ wvif->wdev->tx_burst_idx = -1; - ret = hif_join(wvif, conf, wvif->channel, ssidie); - rcu_read_unlock(); + ret = hif_join(wvif, conf, wvif->channel, ssid, ssidlen); if (ret) { ieee80211_connection_loss(wvif->vif); wvif->join_complete_status = -1; @@ -605,7 +608,9 @@ int wfx_sta_remove(struct ieee80211_hw *hw, struct ieee80211_vif *vif, int i; for (i = 0; i < ARRAY_SIZE(sta_priv->buffered); i++) - WARN(sta_priv->buffered[i], "release station while Tx is in progress"); + if (sta_priv->buffered[i]) + dev_warn(wvif->wdev->dev, "release station while %d pending frame on queue %d", + sta_priv->buffered[i], i); // FIXME: see note in wfx_sta_add() if (vif->type == NL80211_IFTYPE_STATION) return 0; @@ -689,6 +694,7 @@ static void wfx_join_finalize(struct wfx_vif *wvif, wfx_rate_mask_to_hw(wvif->wdev, sta->supp_rates[wvif->channel->band]); else wvif->bss_params.operational_rate_set = -1; + rcu_read_unlock(); if (sta && info->ht_operation_mode & IEEE80211_HT_OP_MODE_NON_GF_STA_PRSNT) hif_dual_cts_protection(wvif, true); @@ -701,8 +707,7 @@ static void wfx_join_finalize(struct wfx_vif *wvif, wvif->bss_params.beacon_lost_count = 20; wvif->bss_params.aid = info->aid; - hif_set_association_mode(wvif, info, sta ? &sta->ht_cap : NULL); - rcu_read_unlock(); + hif_set_association_mode(wvif, info); if (!info->ibss_joined) { hif_keep_alive_period(wvif, 30 /* sec */); diff --git a/drivers/target/iscsi/iscsi_target.c b/drivers/target/iscsi/iscsi_target.c index b94ed4e30770..09e55ea0bf5d 100644 --- a/drivers/target/iscsi/iscsi_target.c +++ b/drivers/target/iscsi/iscsi_target.c @@ -1165,9 +1165,7 @@ int iscsit_setup_scsi_cmd(struct iscsi_conn *conn, struct iscsi_cmd *cmd, hdr->cmdsn, be32_to_cpu(hdr->data_length), payload_length, conn->cid); - if (target_get_sess_cmd(&cmd->se_cmd, true) < 0) - return iscsit_add_reject_cmd(cmd, - ISCSI_REASON_WAITING_FOR_LOGOUT, buf); + target_get_sess_cmd(&cmd->se_cmd, true); cmd->sense_reason = transport_lookup_cmd_lun(&cmd->se_cmd, scsilun_to_int(&hdr->lun)); @@ -2004,9 +2002,7 @@ iscsit_handle_task_mgt_cmd(struct iscsi_conn *conn, struct iscsi_cmd *cmd, conn->sess->se_sess, 0, DMA_NONE, TCM_SIMPLE_TAG, cmd->sense_buffer + 2); - if (target_get_sess_cmd(&cmd->se_cmd, true) < 0) - return iscsit_add_reject_cmd(cmd, - ISCSI_REASON_WAITING_FOR_LOGOUT, buf); + target_get_sess_cmd(&cmd->se_cmd, true); /* * TASK_REASSIGN for ERL=2 / connection stays inside of @@ -4149,6 +4145,9 @@ int iscsit_close_connection( iscsit_stop_nopin_response_timer(conn); iscsit_stop_nopin_timer(conn); + if (conn->conn_transport->iscsit_wait_conn) + conn->conn_transport->iscsit_wait_conn(conn); + /* * During Connection recovery drop unacknowledged out of order * commands for this connection, and prepare the other commands @@ -4231,11 +4230,6 @@ int iscsit_close_connection( * must wait until they have completed. */ iscsit_check_conn_usage_count(conn); - target_sess_cmd_list_set_waiting(sess->se_sess); - target_wait_for_sess_cmds(sess->se_sess); - - if (conn->conn_transport->iscsit_wait_conn) - conn->conn_transport->iscsit_wait_conn(conn); ahash_request_free(conn->conn_tx_hash); if (conn->conn_rx_hash) { diff --git a/drivers/target/target_core_transport.c b/drivers/target/target_core_transport.c index ea482d4b1f00..0ae9e60fc4d5 100644 --- a/drivers/target/target_core_transport.c +++ b/drivers/target/target_core_transport.c @@ -666,6 +666,11 @@ static int transport_cmd_check_stop_to_fabric(struct se_cmd *cmd) target_remove_from_state_list(cmd); + /* + * Clear struct se_cmd->se_lun before the handoff to FE. + */ + cmd->se_lun = NULL; + spin_lock_irqsave(&cmd->t_state_lock, flags); /* * Determine if frontend context caller is requesting the stopping of @@ -693,6 +698,17 @@ static int transport_cmd_check_stop_to_fabric(struct se_cmd *cmd) return cmd->se_tfo->check_stop_free(cmd); } +static void transport_lun_remove_cmd(struct se_cmd *cmd) +{ + struct se_lun *lun = cmd->se_lun; + + if (!lun) + return; + + if (cmpxchg(&cmd->lun_ref_active, true, false)) + percpu_ref_put(&lun->lun_ref); +} + static void target_complete_failure_work(struct work_struct *work) { struct se_cmd *cmd = container_of(work, struct se_cmd, work); @@ -783,6 +799,8 @@ static void target_handle_abort(struct se_cmd *cmd) WARN_ON_ONCE(kref_read(&cmd->cmd_kref) == 0); + transport_lun_remove_cmd(cmd); + transport_cmd_check_stop_to_fabric(cmd); } @@ -1708,6 +1726,7 @@ static void target_complete_tmr_failure(struct work_struct *work) se_cmd->se_tmr_req->response = TMR_LUN_DOES_NOT_EXIST; se_cmd->se_tfo->queue_tm_rsp(se_cmd); + transport_lun_remove_cmd(se_cmd); transport_cmd_check_stop_to_fabric(se_cmd); } @@ -1898,6 +1917,7 @@ void transport_generic_request_failure(struct se_cmd *cmd, goto queue_full; check_stop: + transport_lun_remove_cmd(cmd); transport_cmd_check_stop_to_fabric(cmd); return; @@ -2195,6 +2215,7 @@ queue_status: transport_handle_queue_full(cmd, cmd->se_dev, ret, false); return; } + transport_lun_remove_cmd(cmd); transport_cmd_check_stop_to_fabric(cmd); } @@ -2289,6 +2310,7 @@ static void target_complete_ok_work(struct work_struct *work) if (ret) goto queue_full; + transport_lun_remove_cmd(cmd); transport_cmd_check_stop_to_fabric(cmd); return; } @@ -2314,6 +2336,7 @@ static void target_complete_ok_work(struct work_struct *work) if (ret) goto queue_full; + transport_lun_remove_cmd(cmd); transport_cmd_check_stop_to_fabric(cmd); return; } @@ -2349,6 +2372,7 @@ queue_rsp: if (ret) goto queue_full; + transport_lun_remove_cmd(cmd); transport_cmd_check_stop_to_fabric(cmd); return; } @@ -2384,6 +2408,7 @@ queue_status: break; } + transport_lun_remove_cmd(cmd); transport_cmd_check_stop_to_fabric(cmd); return; @@ -2710,6 +2735,9 @@ int transport_generic_free_cmd(struct se_cmd *cmd, int wait_for_tasks) */ if (cmd->state_active) target_remove_from_state_list(cmd); + + if (cmd->se_lun) + transport_lun_remove_cmd(cmd); } if (aborted) cmd->free_compl = &compl; @@ -2781,9 +2809,6 @@ static void target_release_cmd_kref(struct kref *kref) struct completion *abrt_compl = se_cmd->abrt_compl; unsigned long flags; - if (se_cmd->lun_ref_active) - percpu_ref_put(&se_cmd->se_lun->lun_ref); - if (se_sess) { spin_lock_irqsave(&se_sess->sess_cmd_lock, flags); list_del_init(&se_cmd->se_cmd_list); diff --git a/drivers/tee/amdtee/Kconfig b/drivers/tee/amdtee/Kconfig index 4e32b6413b41..191f9715fa9a 100644 --- a/drivers/tee/amdtee/Kconfig +++ b/drivers/tee/amdtee/Kconfig @@ -3,6 +3,6 @@ config AMDTEE tristate "AMD-TEE" default m - depends on CRYPTO_DEV_SP_PSP + depends on CRYPTO_DEV_SP_PSP && CRYPTO_DEV_CCP_DD help This implements AMD's Trusted Execution Environment (TEE) driver. diff --git a/drivers/tee/amdtee/core.c b/drivers/tee/amdtee/core.c index 6370bb55f512..27b4cd77d0db 100644 --- a/drivers/tee/amdtee/core.c +++ b/drivers/tee/amdtee/core.c @@ -139,6 +139,9 @@ static struct amdtee_session *find_session(struct amdtee_context_data *ctxdata, u32 index = get_session_index(session); struct amdtee_session *sess; + if (index >= TEE_NUM_SESSIONS) + return NULL; + list_for_each_entry(sess, &ctxdata->sess_list, list_node) if (ta_handle == sess->ta_handle && test_bit(index, sess->sess_mask)) @@ -212,6 +215,19 @@ unlock: return rc; } +static void destroy_session(struct kref *ref) +{ + struct amdtee_session *sess = container_of(ref, struct amdtee_session, + refcount); + + /* Unload the TA from TEE */ + handle_unload_ta(sess->ta_handle); + mutex_lock(&session_list_mutex); + list_del(&sess->list_node); + mutex_unlock(&session_list_mutex); + kfree(sess); +} + int amdtee_open_session(struct tee_context *ctx, struct tee_ioctl_open_session_arg *arg, struct tee_param *param) @@ -236,15 +252,13 @@ int amdtee_open_session(struct tee_context *ctx, /* Load the TA binary into TEE environment */ handle_load_ta(ta, ta_size, arg); - if (arg->ret == TEEC_SUCCESS) { - mutex_lock(&session_list_mutex); - sess = alloc_session(ctxdata, arg->session); - mutex_unlock(&session_list_mutex); - } - if (arg->ret != TEEC_SUCCESS) goto out; + mutex_lock(&session_list_mutex); + sess = alloc_session(ctxdata, arg->session); + mutex_unlock(&session_list_mutex); + if (!sess) { rc = -ENOMEM; goto out; @@ -259,40 +273,29 @@ int amdtee_open_session(struct tee_context *ctx, if (i >= TEE_NUM_SESSIONS) { pr_err("reached maximum session count %d\n", TEE_NUM_SESSIONS); + kref_put(&sess->refcount, destroy_session); rc = -ENOMEM; goto out; } /* Open session with loaded TA */ handle_open_session(arg, &session_info, param); - - if (arg->ret == TEEC_SUCCESS) { - sess->session_info[i] = session_info; - set_session_id(sess->ta_handle, i, &arg->session); - } else { + if (arg->ret != TEEC_SUCCESS) { pr_err("open_session failed %d\n", arg->ret); spin_lock(&sess->lock); clear_bit(i, sess->sess_mask); spin_unlock(&sess->lock); + kref_put(&sess->refcount, destroy_session); + goto out; } + + sess->session_info[i] = session_info; + set_session_id(sess->ta_handle, i, &arg->session); out: free_pages((u64)ta, get_order(ta_size)); return rc; } -static void destroy_session(struct kref *ref) -{ - struct amdtee_session *sess = container_of(ref, struct amdtee_session, - refcount); - - /* Unload the TA from TEE */ - handle_unload_ta(sess->ta_handle); - mutex_lock(&session_list_mutex); - list_del(&sess->list_node); - mutex_unlock(&session_list_mutex); - kfree(sess); -} - int amdtee_close_session(struct tee_context *ctx, u32 session) { struct amdtee_context_data *ctxdata = ctx->data; diff --git a/drivers/thunderbolt/switch.c b/drivers/thunderbolt/switch.c index ad5479f21174..a2ce99051c51 100644 --- a/drivers/thunderbolt/switch.c +++ b/drivers/thunderbolt/switch.c @@ -348,6 +348,12 @@ out: return ret; } +static int tb_switch_nvm_no_read(void *priv, unsigned int offset, void *val, + size_t bytes) +{ + return -EPERM; +} + static int tb_switch_nvm_write(void *priv, unsigned int offset, void *val, size_t bytes) { @@ -393,6 +399,7 @@ static struct nvmem_device *register_nvmem(struct tb_switch *sw, int id, config.read_only = true; } else { config.name = "nvm_non_active"; + config.reg_read = tb_switch_nvm_no_read; config.reg_write = tb_switch_nvm_write; config.root_only = true; } @@ -947,7 +954,7 @@ static bool tb_port_is_width_supported(struct tb_port *port, int width) ret = tb_port_read(port, &phy, TB_CFG_PORT, port->cap_phy + LANE_ADP_CS_0, 1); if (ret) - return ret; + return false; widths = (phy & LANE_ADP_CS_0_SUPPORTED_WIDTH_MASK) >> LANE_ADP_CS_0_SUPPORTED_WIDTH_SHIFT; diff --git a/drivers/tty/serdev/core.c b/drivers/tty/serdev/core.c index 42345e79920c..c5f0d936b003 100644 --- a/drivers/tty/serdev/core.c +++ b/drivers/tty/serdev/core.c @@ -18,6 +18,7 @@ #include <linux/sched.h> #include <linux/serdev.h> #include <linux/slab.h> +#include <linux/platform_data/x86/apple.h> static bool is_registered; static DEFINE_IDA(ctrl_ida); @@ -631,6 +632,15 @@ static int acpi_serdev_check_resources(struct serdev_controller *ctrl, if (ret) return ret; + /* + * Apple machines provide an empty resource template, so on those + * machines just look for immediate children with a "baud" property + * (from the _DSM method) instead. + */ + if (!lookup.controller_handle && x86_apple_machine && + !acpi_dev_get_property(adev, "baud", ACPI_TYPE_BUFFER, NULL)) + acpi_get_parent(adev->handle, &lookup.controller_handle); + /* Make sure controller and ResourceSource handle match */ if (ACPI_HANDLE(ctrl->dev.parent) != lookup.controller_handle) return -ENODEV; diff --git a/drivers/tty/serdev/serdev-ttyport.c b/drivers/tty/serdev/serdev-ttyport.c index d1cdd2ab8b4c..d367803e2044 100644 --- a/drivers/tty/serdev/serdev-ttyport.c +++ b/drivers/tty/serdev/serdev-ttyport.c @@ -265,7 +265,6 @@ struct device *serdev_tty_port_register(struct tty_port *port, struct device *parent, struct tty_driver *drv, int idx) { - const struct tty_port_client_operations *old_ops; struct serdev_controller *ctrl; struct serport *serport; int ret; @@ -284,7 +283,6 @@ struct device *serdev_tty_port_register(struct tty_port *port, ctrl->ops = &ctrl_ops; - old_ops = port->client_ops; port->client_ops = &client_ops; port->client_data = ctrl; @@ -297,7 +295,7 @@ struct device *serdev_tty_port_register(struct tty_port *port, err_reset_data: port->client_data = NULL; - port->client_ops = old_ops; + port->client_ops = &tty_port_default_client_ops; serdev_controller_put(ctrl); return ERR_PTR(ret); @@ -312,8 +310,8 @@ int serdev_tty_port_unregister(struct tty_port *port) return -ENODEV; serdev_controller_remove(ctrl); - port->client_ops = NULL; port->client_data = NULL; + port->client_ops = &tty_port_default_client_ops; serdev_controller_put(ctrl); return 0; diff --git a/drivers/tty/serial/8250/8250_aspeed_vuart.c b/drivers/tty/serial/8250/8250_aspeed_vuart.c index d657aa14c3e4..c33e02cbde93 100644 --- a/drivers/tty/serial/8250/8250_aspeed_vuart.c +++ b/drivers/tty/serial/8250/8250_aspeed_vuart.c @@ -446,7 +446,6 @@ static int aspeed_vuart_probe(struct platform_device *pdev) port.port.line = rc; port.port.irq = irq_of_parse_and_map(np, 0); - port.port.irqflags = IRQF_SHARED; port.port.handle_irq = aspeed_vuart_handle_irq; port.port.iotype = UPIO_MEM; port.port.type = PORT_16550A; diff --git a/drivers/tty/serial/8250/8250_core.c b/drivers/tty/serial/8250/8250_core.c index 0894a22fd702..f2a33c9082a6 100644 --- a/drivers/tty/serial/8250/8250_core.c +++ b/drivers/tty/serial/8250/8250_core.c @@ -174,7 +174,7 @@ static int serial_link_irq_chain(struct uart_8250_port *up) struct hlist_head *h; struct hlist_node *n; struct irq_info *i; - int ret, irq_flags = up->port.flags & UPF_SHARE_IRQ ? IRQF_SHARED : 0; + int ret; mutex_lock(&hash_mutex); @@ -209,9 +209,8 @@ static int serial_link_irq_chain(struct uart_8250_port *up) INIT_LIST_HEAD(&up->list); i->head = &up->list; spin_unlock_irq(&i->lock); - irq_flags |= up->port.irqflags; ret = request_irq(up->port.irq, serial8250_interrupt, - irq_flags, up->port.name, i); + up->port.irqflags, up->port.name, i); if (ret < 0) serial_do_unlink(i, up); } diff --git a/drivers/tty/serial/8250/8250_exar.c b/drivers/tty/serial/8250/8250_exar.c index 91e9b070d36d..d330da76d6b6 100644 --- a/drivers/tty/serial/8250/8250_exar.c +++ b/drivers/tty/serial/8250/8250_exar.c @@ -25,6 +25,14 @@ #include "8250.h" +#define PCI_DEVICE_ID_ACCES_COM_2S 0x1052 +#define PCI_DEVICE_ID_ACCES_COM_4S 0x105d +#define PCI_DEVICE_ID_ACCES_COM_8S 0x106c +#define PCI_DEVICE_ID_ACCES_COM232_8 0x10a8 +#define PCI_DEVICE_ID_ACCES_COM_2SM 0x10d2 +#define PCI_DEVICE_ID_ACCES_COM_4SM 0x10db +#define PCI_DEVICE_ID_ACCES_COM_8SM 0x10ea + #define PCI_DEVICE_ID_COMMTECH_4224PCI335 0x0002 #define PCI_DEVICE_ID_COMMTECH_4222PCI335 0x0004 #define PCI_DEVICE_ID_COMMTECH_2324PCI335 0x000a @@ -677,6 +685,22 @@ static int __maybe_unused exar_resume(struct device *dev) static SIMPLE_DEV_PM_OPS(exar_pci_pm, exar_suspend, exar_resume); +static const struct exar8250_board acces_com_2x = { + .num_ports = 2, + .setup = pci_xr17c154_setup, +}; + +static const struct exar8250_board acces_com_4x = { + .num_ports = 4, + .setup = pci_xr17c154_setup, +}; + +static const struct exar8250_board acces_com_8x = { + .num_ports = 8, + .setup = pci_xr17c154_setup, +}; + + static const struct exar8250_board pbn_fastcom335_2 = { .num_ports = 2, .setup = pci_fastcom335_setup, @@ -745,6 +769,15 @@ static const struct exar8250_board pbn_exar_XR17V8358 = { } static const struct pci_device_id exar_pci_tbl[] = { + EXAR_DEVICE(ACCESSIO, ACCES_COM_2S, acces_com_2x), + EXAR_DEVICE(ACCESSIO, ACCES_COM_4S, acces_com_4x), + EXAR_DEVICE(ACCESSIO, ACCES_COM_8S, acces_com_8x), + EXAR_DEVICE(ACCESSIO, ACCES_COM232_8, acces_com_8x), + EXAR_DEVICE(ACCESSIO, ACCES_COM_2SM, acces_com_2x), + EXAR_DEVICE(ACCESSIO, ACCES_COM_4SM, acces_com_4x), + EXAR_DEVICE(ACCESSIO, ACCES_COM_8SM, acces_com_8x), + + CONNECT_DEVICE(XR17C152, UART_2_232, pbn_connect), CONNECT_DEVICE(XR17C154, UART_4_232, pbn_connect), CONNECT_DEVICE(XR17C158, UART_8_232, pbn_connect), diff --git a/drivers/tty/serial/8250/8250_of.c b/drivers/tty/serial/8250/8250_of.c index 531ad67395e0..f6687756ec5e 100644 --- a/drivers/tty/serial/8250/8250_of.c +++ b/drivers/tty/serial/8250/8250_of.c @@ -202,7 +202,6 @@ static int of_platform_serial_setup(struct platform_device *ofdev, port->type = type; port->uartclk = clk; - port->irqflags |= IRQF_SHARED; if (of_property_read_bool(np, "no-loopback-test")) port->flags |= UPF_SKIP_TEST; diff --git a/drivers/tty/serial/8250/8250_port.c b/drivers/tty/serial/8250/8250_port.c index 430e3467aff7..0325f2e53b74 100644 --- a/drivers/tty/serial/8250/8250_port.c +++ b/drivers/tty/serial/8250/8250_port.c @@ -2177,6 +2177,10 @@ int serial8250_do_startup(struct uart_port *port) } } + /* Check if we need to have shared IRQs */ + if (port->irq && (up->port.flags & UPF_SHARE_IRQ)) + up->port.irqflags |= IRQF_SHARED; + if (port->irq && !(up->port.flags & UPF_NO_THRE_TEST)) { unsigned char iir1; /* diff --git a/drivers/tty/serial/ar933x_uart.c b/drivers/tty/serial/ar933x_uart.c index 3bdd56a1021b..ea12f10610b6 100644 --- a/drivers/tty/serial/ar933x_uart.c +++ b/drivers/tty/serial/ar933x_uart.c @@ -286,6 +286,10 @@ static void ar933x_uart_set_termios(struct uart_port *port, ar933x_uart_rmw_set(up, AR933X_UART_CS_REG, AR933X_UART_CS_HOST_INT_EN); + /* enable RX and TX ready overide */ + ar933x_uart_rmw_set(up, AR933X_UART_CS_REG, + AR933X_UART_CS_TX_READY_ORIDE | AR933X_UART_CS_RX_READY_ORIDE); + /* reenable the UART */ ar933x_uart_rmw(up, AR933X_UART_CS_REG, AR933X_UART_CS_IF_MODE_M << AR933X_UART_CS_IF_MODE_S, @@ -418,6 +422,10 @@ static int ar933x_uart_startup(struct uart_port *port) ar933x_uart_rmw_set(up, AR933X_UART_CS_REG, AR933X_UART_CS_HOST_INT_EN); + /* enable RX and TX ready overide */ + ar933x_uart_rmw_set(up, AR933X_UART_CS_REG, + AR933X_UART_CS_TX_READY_ORIDE | AR933X_UART_CS_RX_READY_ORIDE); + /* Enable RX interrupts */ up->ier = AR933X_UART_INT_RX_VALID; ar933x_uart_write(up, AR933X_UART_INT_EN_REG, up->ier); diff --git a/drivers/tty/serial/atmel_serial.c b/drivers/tty/serial/atmel_serial.c index c15c398c88a9..a39c87a7c2e1 100644 --- a/drivers/tty/serial/atmel_serial.c +++ b/drivers/tty/serial/atmel_serial.c @@ -570,7 +570,8 @@ static void atmel_stop_tx(struct uart_port *port) atmel_uart_writel(port, ATMEL_US_IDR, atmel_port->tx_done_mask); if (atmel_uart_is_half_duplex(port)) - atmel_start_rx(port); + if (!atomic_read(&atmel_port->tasklet_shutdown)) + atmel_start_rx(port); } diff --git a/drivers/tty/serial/cpm_uart/cpm_uart_core.c b/drivers/tty/serial/cpm_uart/cpm_uart_core.c index 19d5a4cf29a6..d4b81b06e0cb 100644 --- a/drivers/tty/serial/cpm_uart/cpm_uart_core.c +++ b/drivers/tty/serial/cpm_uart/cpm_uart_core.c @@ -1373,6 +1373,7 @@ static struct console cpm_scc_uart_console = { static int __init cpm_uart_console_init(void) { + cpm_muram_init(); register_console(&cpm_scc_uart_console); return 0; } diff --git a/drivers/tty/serial/fsl_lpuart.c b/drivers/tty/serial/fsl_lpuart.c index 91e2805e6441..c31b8f3db6bf 100644 --- a/drivers/tty/serial/fsl_lpuart.c +++ b/drivers/tty/serial/fsl_lpuart.c @@ -264,6 +264,7 @@ struct lpuart_port { int rx_dma_rng_buf_len; unsigned int dma_tx_nents; wait_queue_head_t dma_wait; + bool id_allocated; }; struct lpuart_soc_data { @@ -2390,6 +2391,8 @@ static int __init lpuart32_imx_early_console_setup(struct earlycon_device *devic OF_EARLYCON_DECLARE(lpuart, "fsl,vf610-lpuart", lpuart_early_console_setup); OF_EARLYCON_DECLARE(lpuart32, "fsl,ls1021a-lpuart", lpuart32_early_console_setup); OF_EARLYCON_DECLARE(lpuart32, "fsl,imx7ulp-lpuart", lpuart32_imx_early_console_setup); +EARLYCON_DECLARE(lpuart, lpuart_early_console_setup); +EARLYCON_DECLARE(lpuart32, lpuart32_early_console_setup); #define LPUART_CONSOLE (&lpuart_console) #define LPUART32_CONSOLE (&lpuart32_console) @@ -2420,19 +2423,6 @@ static int lpuart_probe(struct platform_device *pdev) if (!sport) return -ENOMEM; - ret = of_alias_get_id(np, "serial"); - if (ret < 0) { - ret = ida_simple_get(&fsl_lpuart_ida, 0, UART_NR, GFP_KERNEL); - if (ret < 0) { - dev_err(&pdev->dev, "port line is full, add device failed\n"); - return ret; - } - } - if (ret >= ARRAY_SIZE(lpuart_ports)) { - dev_err(&pdev->dev, "serial%d out of range\n", ret); - return -EINVAL; - } - sport->port.line = ret; res = platform_get_resource(pdev, IORESOURCE_MEM, 0); sport->port.membase = devm_ioremap_resource(&pdev->dev, res); if (IS_ERR(sport->port.membase)) @@ -2477,9 +2467,25 @@ static int lpuart_probe(struct platform_device *pdev) } } + ret = of_alias_get_id(np, "serial"); + if (ret < 0) { + ret = ida_simple_get(&fsl_lpuart_ida, 0, UART_NR, GFP_KERNEL); + if (ret < 0) { + dev_err(&pdev->dev, "port line is full, add device failed\n"); + return ret; + } + sport->id_allocated = true; + } + if (ret >= ARRAY_SIZE(lpuart_ports)) { + dev_err(&pdev->dev, "serial%d out of range\n", ret); + ret = -EINVAL; + goto failed_out_of_range; + } + sport->port.line = ret; + ret = lpuart_enable_clks(sport); if (ret) - return ret; + goto failed_clock_enable; sport->port.uartclk = lpuart_get_baud_clk_rate(sport); lpuart_ports[sport->port.line] = sport; @@ -2529,6 +2535,10 @@ static int lpuart_probe(struct platform_device *pdev) failed_attach_port: failed_irq_request: lpuart_disable_clks(sport); +failed_clock_enable: +failed_out_of_range: + if (sport->id_allocated) + ida_simple_remove(&fsl_lpuart_ida, sport->port.line); return ret; } @@ -2538,7 +2548,8 @@ static int lpuart_remove(struct platform_device *pdev) uart_remove_one_port(&lpuart_reg, &sport->port); - ida_simple_remove(&fsl_lpuart_ida, sport->port.line); + if (sport->id_allocated) + ida_simple_remove(&fsl_lpuart_ida, sport->port.line); lpuart_disable_clks(sport); diff --git a/drivers/tty/serial/imx.c b/drivers/tty/serial/imx.c index 0c6c63166250..d337782b3648 100644 --- a/drivers/tty/serial/imx.c +++ b/drivers/tty/serial/imx.c @@ -599,7 +599,7 @@ static void imx_uart_dma_tx(struct imx_port *sport) sport->tx_bytes = uart_circ_chars_pending(xmit); - if (xmit->tail < xmit->head) { + if (xmit->tail < xmit->head || xmit->head == 0) { sport->dma_tx_nents = 1; sg_init_one(sgl, xmit->buf + xmit->tail, sport->tx_bytes); } else { diff --git a/drivers/tty/serial/mvebu-uart.c b/drivers/tty/serial/mvebu-uart.c index c12a12556339..4e9a590712cb 100644 --- a/drivers/tty/serial/mvebu-uart.c +++ b/drivers/tty/serial/mvebu-uart.c @@ -851,7 +851,7 @@ static int mvebu_uart_probe(struct platform_device *pdev) port->membase = devm_ioremap_resource(&pdev->dev, reg); if (IS_ERR(port->membase)) - return -PTR_ERR(port->membase); + return PTR_ERR(port->membase); mvuart = devm_kzalloc(&pdev->dev, sizeof(struct mvebu_uart), GFP_KERNEL); diff --git a/drivers/tty/serial/qcom_geni_serial.c b/drivers/tty/serial/qcom_geni_serial.c index 191abb18fc2a..0bd1684cabb3 100644 --- a/drivers/tty/serial/qcom_geni_serial.c +++ b/drivers/tty/serial/qcom_geni_serial.c @@ -129,6 +129,7 @@ static int handle_rx_console(struct uart_port *uport, u32 bytes, bool drop); static int handle_rx_uart(struct uart_port *uport, u32 bytes, bool drop); static unsigned int qcom_geni_serial_tx_empty(struct uart_port *port); static void qcom_geni_serial_stop_rx(struct uart_port *uport); +static void qcom_geni_serial_handle_rx(struct uart_port *uport, bool drop); static const unsigned long root_freq[] = {7372800, 14745600, 19200000, 29491200, 32000000, 48000000, 64000000, 80000000, @@ -599,7 +600,7 @@ static void qcom_geni_serial_stop_rx(struct uart_port *uport) u32 irq_en; u32 status; struct qcom_geni_serial_port *port = to_dev_port(uport, uport); - u32 irq_clear = S_CMD_DONE_EN; + u32 s_irq_status; irq_en = readl(uport->membase + SE_GENI_S_IRQ_EN); irq_en &= ~(S_RX_FIFO_WATERMARK_EN | S_RX_FIFO_LAST_EN); @@ -615,10 +616,19 @@ static void qcom_geni_serial_stop_rx(struct uart_port *uport) return; geni_se_cancel_s_cmd(&port->se); - qcom_geni_serial_poll_bit(uport, SE_GENI_S_CMD_CTRL_REG, - S_GENI_CMD_CANCEL, false); + qcom_geni_serial_poll_bit(uport, SE_GENI_S_IRQ_STATUS, + S_CMD_CANCEL_EN, true); + /* + * If timeout occurs secondary engine remains active + * and Abort sequence is executed. + */ + s_irq_status = readl(uport->membase + SE_GENI_S_IRQ_STATUS); + /* Flush the Rx buffer */ + if (s_irq_status & S_RX_FIFO_LAST_EN) + qcom_geni_serial_handle_rx(uport, true); + writel(s_irq_status, uport->membase + SE_GENI_S_IRQ_CLEAR); + status = readl(uport->membase + SE_GENI_STATUS); - writel(irq_clear, uport->membase + SE_GENI_S_IRQ_CLEAR); if (status & S_GENI_CMD_ACTIVE) qcom_geni_serial_abort_rx(uport); } diff --git a/drivers/tty/serial/serial-tegra.c b/drivers/tty/serial/serial-tegra.c index 33034b852a51..8de8bac9c6c7 100644 --- a/drivers/tty/serial/serial-tegra.c +++ b/drivers/tty/serial/serial-tegra.c @@ -692,11 +692,22 @@ static void tegra_uart_copy_rx_to_tty(struct tegra_uart_port *tup, count, DMA_TO_DEVICE); } +static void do_handle_rx_pio(struct tegra_uart_port *tup) +{ + struct tty_struct *tty = tty_port_tty_get(&tup->uport.state->port); + struct tty_port *port = &tup->uport.state->port; + + tegra_uart_handle_rx_pio(tup, port); + if (tty) { + tty_flip_buffer_push(port); + tty_kref_put(tty); + } +} + static void tegra_uart_rx_buffer_push(struct tegra_uart_port *tup, unsigned int residue) { struct tty_port *port = &tup->uport.state->port; - struct tty_struct *tty = tty_port_tty_get(port); unsigned int count; async_tx_ack(tup->rx_dma_desc); @@ -705,11 +716,7 @@ static void tegra_uart_rx_buffer_push(struct tegra_uart_port *tup, /* If we are here, DMA is stopped */ tegra_uart_copy_rx_to_tty(tup, port, count); - tegra_uart_handle_rx_pio(tup, port); - if (tty) { - tty_flip_buffer_push(port); - tty_kref_put(tty); - } + do_handle_rx_pio(tup); } static void tegra_uart_rx_dma_complete(void *args) @@ -749,8 +756,10 @@ static void tegra_uart_terminate_rx_dma(struct tegra_uart_port *tup) { struct dma_tx_state state; - if (!tup->rx_dma_active) + if (!tup->rx_dma_active) { + do_handle_rx_pio(tup); return; + } dmaengine_terminate_all(tup->rx_dma_chan); dmaengine_tx_status(tup->rx_dma_chan, tup->rx_cookie, &state); @@ -816,18 +825,6 @@ static void tegra_uart_handle_modem_signal_change(struct uart_port *u) uart_handle_cts_change(&tup->uport, msr & UART_MSR_CTS); } -static void do_handle_rx_pio(struct tegra_uart_port *tup) -{ - struct tty_struct *tty = tty_port_tty_get(&tup->uport.state->port); - struct tty_port *port = &tup->uport.state->port; - - tegra_uart_handle_rx_pio(tup, port); - if (tty) { - tty_flip_buffer_push(port); - tty_kref_put(tty); - } -} - static irqreturn_t tegra_uart_isr(int irq, void *data) { struct tegra_uart_port *tup = data; diff --git a/drivers/tty/tty_io.c b/drivers/tty/tty_io.c index a1453fe10862..5a6f36b391d9 100644 --- a/drivers/tty/tty_io.c +++ b/drivers/tty/tty_io.c @@ -1589,9 +1589,7 @@ void tty_kclose(struct tty_struct *tty) tty_debug_hangup(tty, "freeing structure\n"); /* * The release_tty function takes care of the details of clearing - * the slots and preserving the termios structure. The tty_unlock_pair - * should be safe as we keep a kref while the tty is locked (so the - * unlock never unlocks a freed tty). + * the slots and preserving the termios structure. */ mutex_lock(&tty_mutex); tty_port_set_kopened(tty->port, 0); @@ -1621,9 +1619,7 @@ void tty_release_struct(struct tty_struct *tty, int idx) tty_debug_hangup(tty, "freeing structure\n"); /* * The release_tty function takes care of the details of clearing - * the slots and preserving the termios structure. The tty_unlock_pair - * should be safe as we keep a kref while the tty is locked (so the - * unlock never unlocks a freed tty). + * the slots and preserving the termios structure. */ mutex_lock(&tty_mutex); release_tty(tty, idx); @@ -2734,9 +2730,11 @@ static int compat_tty_tiocgserial(struct tty_struct *tty, struct serial_struct32 v32; struct serial_struct v; int err; - memset(&v, 0, sizeof(struct serial_struct)); - if (!tty->ops->set_serial) + memset(&v, 0, sizeof(v)); + memset(&v32, 0, sizeof(v32)); + + if (!tty->ops->get_serial) return -ENOTTY; err = tty->ops->get_serial(tty, &v); if (!err) { diff --git a/drivers/tty/tty_port.c b/drivers/tty/tty_port.c index 044c3cbdcfa4..ea80bf872f54 100644 --- a/drivers/tty/tty_port.c +++ b/drivers/tty/tty_port.c @@ -52,10 +52,11 @@ static void tty_port_default_wakeup(struct tty_port *port) } } -static const struct tty_port_client_operations default_client_ops = { +const struct tty_port_client_operations tty_port_default_client_ops = { .receive_buf = tty_port_default_receive_buf, .write_wakeup = tty_port_default_wakeup, }; +EXPORT_SYMBOL_GPL(tty_port_default_client_ops); void tty_port_init(struct tty_port *port) { @@ -68,7 +69,7 @@ void tty_port_init(struct tty_port *port) spin_lock_init(&port->lock); port->close_delay = (50 * HZ) / 100; port->closing_wait = (3000 * HZ) / 100; - port->client_ops = &default_client_ops; + port->client_ops = &tty_port_default_client_ops; kref_init(&port->kref); } EXPORT_SYMBOL(tty_port_init); diff --git a/drivers/tty/vt/selection.c b/drivers/tty/vt/selection.c index 78732feaf65b..d7d2e4b844bc 100644 --- a/drivers/tty/vt/selection.c +++ b/drivers/tty/vt/selection.c @@ -16,6 +16,7 @@ #include <linux/tty.h> #include <linux/sched.h> #include <linux/mm.h> +#include <linux/mutex.h> #include <linux/slab.h> #include <linux/types.h> @@ -29,6 +30,8 @@ #include <linux/console.h> #include <linux/tty_flip.h> +#include <linux/sched/signal.h> + /* Don't take this from <ctype.h>: 011-015 on the screen aren't spaces */ #define isspace(c) ((c) == ' ') @@ -43,6 +46,7 @@ static volatile int sel_start = -1; /* cleared by clear_selection */ static int sel_end; static int sel_buffer_lth; static char *sel_buffer; +static DEFINE_MUTEX(sel_lock); /* clear_selection, highlight and highlight_pointer can be called from interrupt (via scrollback/front) */ @@ -177,14 +181,14 @@ int set_selection_user(const struct tiocl_selection __user *sel, return set_selection_kernel(&v, tty); } -int set_selection_kernel(struct tiocl_selection *v, struct tty_struct *tty) +static int __set_selection_kernel(struct tiocl_selection *v, struct tty_struct *tty) { struct vc_data *vc = vc_cons[fg_console].d; int new_sel_start, new_sel_end, spc; char *bp, *obp; int i, ps, pe, multiplier; u32 c; - int mode; + int mode, ret = 0; poke_blanked_console(); @@ -332,7 +336,21 @@ int set_selection_kernel(struct tiocl_selection *v, struct tty_struct *tty) } } sel_buffer_lth = bp - sel_buffer; - return 0; + + return ret; +} + +int set_selection_kernel(struct tiocl_selection *v, struct tty_struct *tty) +{ + int ret; + + mutex_lock(&sel_lock); + console_lock(); + ret = __set_selection_kernel(v, tty); + console_unlock(); + mutex_unlock(&sel_lock); + + return ret; } EXPORT_SYMBOL_GPL(set_selection_kernel); @@ -350,6 +368,7 @@ int paste_selection(struct tty_struct *tty) unsigned int count; struct tty_ldisc *ld; DECLARE_WAITQUEUE(wait, current); + int ret = 0; console_lock(); poke_blanked_console(); @@ -361,10 +380,17 @@ int paste_selection(struct tty_struct *tty) tty_buffer_lock_exclusive(&vc->port); add_wait_queue(&vc->paste_wait, &wait); + mutex_lock(&sel_lock); while (sel_buffer && sel_buffer_lth > pasted) { set_current_state(TASK_INTERRUPTIBLE); + if (signal_pending(current)) { + ret = -EINTR; + break; + } if (tty_throttled(tty)) { + mutex_unlock(&sel_lock); schedule(); + mutex_lock(&sel_lock); continue; } __set_current_state(TASK_RUNNING); @@ -373,11 +399,12 @@ int paste_selection(struct tty_struct *tty) count); pasted += count; } + mutex_unlock(&sel_lock); remove_wait_queue(&vc->paste_wait, &wait); __set_current_state(TASK_RUNNING); tty_buffer_unlock_exclusive(&vc->port); tty_ldisc_deref(ld); - return 0; + return ret; } EXPORT_SYMBOL_GPL(paste_selection); diff --git a/drivers/tty/vt/vt.c b/drivers/tty/vt/vt.c index 35d21cdb60d0..15d27698054a 100644 --- a/drivers/tty/vt/vt.c +++ b/drivers/tty/vt/vt.c @@ -936,10 +936,21 @@ static void flush_scrollback(struct vc_data *vc) WARN_CONSOLE_UNLOCKED(); set_origin(vc); - if (vc->vc_sw->con_flush_scrollback) + if (vc->vc_sw->con_flush_scrollback) { vc->vc_sw->con_flush_scrollback(vc); - else + } else if (con_is_visible(vc)) { + /* + * When no con_flush_scrollback method is provided then the + * legacy way for flushing the scrollback buffer is to use + * a side effect of the con_switch method. We do it only on + * the foreground console as background consoles have no + * scrollback buffers in that case and we obviously don't + * want to switch to them. + */ + hide_cursor(vc); vc->vc_sw->con_switch(vc); + set_cursor(vc); + } } /* @@ -3035,10 +3046,8 @@ int tioclinux(struct tty_struct *tty, unsigned long arg) switch (type) { case TIOCL_SETSEL: - console_lock(); ret = set_selection_user((struct tiocl_selection __user *)(p+1), tty); - console_unlock(); break; case TIOCL_PASTESEL: ret = paste_selection(tty); diff --git a/drivers/tty/vt/vt_ioctl.c b/drivers/tty/vt/vt_ioctl.c index 8b0ed139592f..ee6c91ef1f6c 100644 --- a/drivers/tty/vt/vt_ioctl.c +++ b/drivers/tty/vt/vt_ioctl.c @@ -876,15 +876,20 @@ int vt_ioctl(struct tty_struct *tty, return -EINVAL; for (i = 0; i < MAX_NR_CONSOLES; i++) { + struct vc_data *vcp; + if (!vc_cons[i].d) continue; console_lock(); - if (v.v_vlin) - vc_cons[i].d->vc_scan_lines = v.v_vlin; - if (v.v_clin) - vc_cons[i].d->vc_font.height = v.v_clin; - vc_cons[i].d->vc_resize_user = 1; - vc_resize(vc_cons[i].d, v.v_cols, v.v_rows); + vcp = vc_cons[i].d; + if (vcp) { + if (v.v_vlin) + vcp->vc_scan_lines = v.v_vlin; + if (v.v_clin) + vcp->vc_font.height = v.v_clin; + vcp->vc_resize_user = 1; + vc_resize(vcp, v.v_cols, v.v_rows); + } console_unlock(); } break; diff --git a/drivers/usb/cdns3/gadget.c b/drivers/usb/cdns3/gadget.c index 736b0c6e27fe..3574dbb09366 100644 --- a/drivers/usb/cdns3/gadget.c +++ b/drivers/usb/cdns3/gadget.c @@ -2550,7 +2550,7 @@ found: /* Update ring only if removed request is on pending_req_list list */ if (req_on_hw_ring) { link_trb->buffer = TRB_BUFFER(priv_ep->trb_pool_dma + - (priv_req->start_trb * TRB_SIZE)); + ((priv_req->end_trb + 1) * TRB_SIZE)); link_trb->control = (link_trb->control & TRB_CYCLE) | TRB_TYPE(TRB_LINK) | TRB_CHAIN; @@ -2595,11 +2595,21 @@ int __cdns3_gadget_ep_clear_halt(struct cdns3_endpoint *priv_ep) { struct cdns3_device *priv_dev = priv_ep->cdns3_dev; struct usb_request *request; + struct cdns3_request *priv_req; + struct cdns3_trb *trb = NULL; int ret; int val; trace_cdns3_halt(priv_ep, 0, 0); + request = cdns3_next_request(&priv_ep->pending_req_list); + if (request) { + priv_req = to_cdns3_request(request); + trb = priv_req->trb; + if (trb) + trb->control = trb->control ^ TRB_CYCLE; + } + writel(EP_CMD_CSTALL | EP_CMD_EPRST, &priv_dev->regs->ep_cmd); /* wait for EPRST cleared */ @@ -2610,10 +2620,11 @@ int __cdns3_gadget_ep_clear_halt(struct cdns3_endpoint *priv_ep) priv_ep->flags &= ~(EP_STALLED | EP_STALL_PENDING); - request = cdns3_next_request(&priv_ep->pending_req_list); - - if (request) + if (request) { + if (trb) + trb->control = trb->control ^ TRB_CYCLE; cdns3_rearm_transfer(priv_ep, 1); + } cdns3_start_all_request(priv_dev, priv_ep); return ret; diff --git a/drivers/usb/chipidea/udc.c b/drivers/usb/chipidea/udc.c index ffaf46f5d062..4c4ac30db498 100644 --- a/drivers/usb/chipidea/udc.c +++ b/drivers/usb/chipidea/udc.c @@ -1530,18 +1530,19 @@ static const struct usb_ep_ops usb_ep_ops = { static void ci_hdrc_gadget_connect(struct usb_gadget *_gadget, int is_active) { struct ci_hdrc *ci = container_of(_gadget, struct ci_hdrc, gadget); - unsigned long flags; if (is_active) { pm_runtime_get_sync(&_gadget->dev); hw_device_reset(ci); - spin_lock_irqsave(&ci->lock, flags); + spin_lock_irq(&ci->lock); if (ci->driver) { hw_device_state(ci, ci->ep0out->qh.dma); usb_gadget_set_state(_gadget, USB_STATE_POWERED); + spin_unlock_irq(&ci->lock); usb_udc_vbus_handler(_gadget, true); + } else { + spin_unlock_irq(&ci->lock); } - spin_unlock_irqrestore(&ci->lock, flags); } else { usb_udc_vbus_handler(_gadget, false); if (ci->driver) diff --git a/drivers/usb/class/cdc-acm.c b/drivers/usb/class/cdc-acm.c index 62f4fb9b362f..47f09a6ce7bd 100644 --- a/drivers/usb/class/cdc-acm.c +++ b/drivers/usb/class/cdc-acm.c @@ -896,10 +896,10 @@ static int get_serial_info(struct tty_struct *tty, struct serial_struct *ss) ss->xmit_fifo_size = acm->writesize; ss->baud_base = le32_to_cpu(acm->line.dwDTERate); - ss->close_delay = acm->port.close_delay / 10; + ss->close_delay = jiffies_to_msecs(acm->port.close_delay) / 10; ss->closing_wait = acm->port.closing_wait == ASYNC_CLOSING_WAIT_NONE ? ASYNC_CLOSING_WAIT_NONE : - acm->port.closing_wait / 10; + jiffies_to_msecs(acm->port.closing_wait) / 10; return 0; } @@ -907,24 +907,32 @@ static int set_serial_info(struct tty_struct *tty, struct serial_struct *ss) { struct acm *acm = tty->driver_data; unsigned int closing_wait, close_delay; + unsigned int old_closing_wait, old_close_delay; int retval = 0; - close_delay = ss->close_delay * 10; + close_delay = msecs_to_jiffies(ss->close_delay * 10); closing_wait = ss->closing_wait == ASYNC_CLOSING_WAIT_NONE ? - ASYNC_CLOSING_WAIT_NONE : ss->closing_wait * 10; + ASYNC_CLOSING_WAIT_NONE : + msecs_to_jiffies(ss->closing_wait * 10); + + /* we must redo the rounding here, so that the values match */ + old_close_delay = jiffies_to_msecs(acm->port.close_delay) / 10; + old_closing_wait = acm->port.closing_wait == ASYNC_CLOSING_WAIT_NONE ? + ASYNC_CLOSING_WAIT_NONE : + jiffies_to_msecs(acm->port.closing_wait) / 10; mutex_lock(&acm->port.mutex); - if (!capable(CAP_SYS_ADMIN)) { - if ((close_delay != acm->port.close_delay) || - (closing_wait != acm->port.closing_wait)) + if ((ss->close_delay != old_close_delay) || + (ss->closing_wait != old_closing_wait)) { + if (!capable(CAP_SYS_ADMIN)) retval = -EPERM; - else - retval = -EOPNOTSUPP; - } else { - acm->port.close_delay = close_delay; - acm->port.closing_wait = closing_wait; - } + else { + acm->port.close_delay = close_delay; + acm->port.closing_wait = closing_wait; + } + } else + retval = -EOPNOTSUPP; mutex_unlock(&acm->port.mutex); return retval; diff --git a/drivers/usb/core/config.c b/drivers/usb/core/config.c index 26bc05e48d8a..b7918f695434 100644 --- a/drivers/usb/core/config.c +++ b/drivers/usb/core/config.c @@ -256,6 +256,7 @@ static int usb_parse_endpoint(struct device *ddev, int cfgno, struct usb_host_interface *ifp, int num_ep, unsigned char *buffer, int size) { + struct usb_device *udev = to_usb_device(ddev); unsigned char *buffer0 = buffer; struct usb_endpoint_descriptor *d; struct usb_host_endpoint *endpoint; @@ -297,6 +298,16 @@ static int usb_parse_endpoint(struct device *ddev, int cfgno, goto skip_to_next_endpoint_or_interface_descriptor; } + /* Ignore blacklisted endpoints */ + if (udev->quirks & USB_QUIRK_ENDPOINT_BLACKLIST) { + if (usb_endpoint_is_blacklisted(udev, ifp, d)) { + dev_warn(ddev, "config %d interface %d altsetting %d has a blacklisted endpoint with address 0x%X, skipping\n", + cfgno, inum, asnum, + d->bEndpointAddress); + goto skip_to_next_endpoint_or_interface_descriptor; + } + } + endpoint = &ifp->endpoint[ifp->desc.bNumEndpoints]; ++ifp->desc.bNumEndpoints; @@ -311,7 +322,7 @@ static int usb_parse_endpoint(struct device *ddev, int cfgno, j = 255; if (usb_endpoint_xfer_int(d)) { i = 1; - switch (to_usb_device(ddev)->speed) { + switch (udev->speed) { case USB_SPEED_SUPER_PLUS: case USB_SPEED_SUPER: case USB_SPEED_HIGH: @@ -332,8 +343,7 @@ static int usb_parse_endpoint(struct device *ddev, int cfgno, /* * This quirk fixes bIntervals reported in ms. */ - if (to_usb_device(ddev)->quirks & - USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL) { + if (udev->quirks & USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL) { n = clamp(fls(d->bInterval) + 3, i, j); i = j = n; } @@ -341,8 +351,7 @@ static int usb_parse_endpoint(struct device *ddev, int cfgno, * This quirk fixes bIntervals reported in * linear microframes. */ - if (to_usb_device(ddev)->quirks & - USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL) { + if (udev->quirks & USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL) { n = clamp(fls(d->bInterval), i, j); i = j = n; } @@ -359,7 +368,7 @@ static int usb_parse_endpoint(struct device *ddev, int cfgno, } else if (usb_endpoint_xfer_isoc(d)) { i = 1; j = 16; - switch (to_usb_device(ddev)->speed) { + switch (udev->speed) { case USB_SPEED_HIGH: n = 7; /* 8 ms = 2^(7-1) uframes */ break; @@ -381,8 +390,7 @@ static int usb_parse_endpoint(struct device *ddev, int cfgno, * explicitly forbidden by the USB spec. In an attempt to make * them usable, we will try treating them as Interrupt endpoints. */ - if (to_usb_device(ddev)->speed == USB_SPEED_LOW && - usb_endpoint_xfer_bulk(d)) { + if (udev->speed == USB_SPEED_LOW && usb_endpoint_xfer_bulk(d)) { dev_warn(ddev, "config %d interface %d altsetting %d " "endpoint 0x%X is Bulk; changing to Interrupt\n", cfgno, inum, asnum, d->bEndpointAddress); @@ -406,7 +414,7 @@ static int usb_parse_endpoint(struct device *ddev, int cfgno, /* Find the highest legal maxpacket size for this endpoint */ i = 0; /* additional transactions per microframe */ - switch (to_usb_device(ddev)->speed) { + switch (udev->speed) { case USB_SPEED_LOW: maxpacket_maxes = low_speed_maxpacket_maxes; break; @@ -442,8 +450,7 @@ static int usb_parse_endpoint(struct device *ddev, int cfgno, * maxpacket sizes other than 512. High speed HCDs may not * be able to handle that particular bug, so let's warn... */ - if (to_usb_device(ddev)->speed == USB_SPEED_HIGH - && usb_endpoint_xfer_bulk(d)) { + if (udev->speed == USB_SPEED_HIGH && usb_endpoint_xfer_bulk(d)) { if (maxp != 512) dev_warn(ddev, "config %d interface %d altsetting %d " "bulk endpoint 0x%X has invalid maxpacket %d\n", @@ -452,7 +459,7 @@ static int usb_parse_endpoint(struct device *ddev, int cfgno, } /* Parse a possible SuperSpeed endpoint companion descriptor */ - if (to_usb_device(ddev)->speed >= USB_SPEED_SUPER) + if (udev->speed >= USB_SPEED_SUPER) usb_parse_ss_endpoint_companion(ddev, cfgno, inum, asnum, endpoint, buffer, size); diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c index 3405b146edc9..54cd8ef795ec 100644 --- a/drivers/usb/core/hub.c +++ b/drivers/usb/core/hub.c @@ -38,7 +38,9 @@ #include "otg_whitelist.h" #define USB_VENDOR_GENESYS_LOGIC 0x05e3 +#define USB_VENDOR_SMSC 0x0424 #define HUB_QUIRK_CHECK_PORT_AUTOSUSPEND 0x01 +#define HUB_QUIRK_DISABLE_AUTOSUSPEND 0x02 #define USB_TP_TRANSMISSION_DELAY 40 /* ns */ #define USB_TP_TRANSMISSION_DELAY_MAX 65535 /* ns */ @@ -986,13 +988,17 @@ int usb_remove_device(struct usb_device *udev) { struct usb_hub *hub; struct usb_interface *intf; + int ret; if (!udev->parent) /* Can't remove a root hub */ return -EINVAL; hub = usb_hub_to_struct_hub(udev->parent); intf = to_usb_interface(hub->intfdev); - usb_autopm_get_interface(intf); + ret = usb_autopm_get_interface(intf); + if (ret < 0) + return ret; + set_bit(udev->portnum, hub->removed_bits); hub_port_logical_disconnect(hub, udev->portnum); usb_autopm_put_interface(intf); @@ -1217,11 +1223,6 @@ static void hub_activate(struct usb_hub *hub, enum hub_activation_type type) #ifdef CONFIG_PM udev->reset_resume = 1; #endif - /* Don't set the change_bits when the device - * was powered off. - */ - if (test_bit(port1, hub->power_bits)) - set_bit(port1, hub->change_bits); } else { /* The power session is gone; tell hub_wq */ @@ -1731,6 +1732,10 @@ static void hub_disconnect(struct usb_interface *intf) kfree(hub->buffer); pm_suspend_ignore_children(&intf->dev, false); + + if (hub->quirk_disable_autosuspend) + usb_autopm_put_interface(intf); + kref_put(&hub->kref, hub_release); } @@ -1863,6 +1868,11 @@ static int hub_probe(struct usb_interface *intf, const struct usb_device_id *id) if (id->driver_info & HUB_QUIRK_CHECK_PORT_AUTOSUSPEND) hub->quirk_check_port_auto_suspend = 1; + if (id->driver_info & HUB_QUIRK_DISABLE_AUTOSUSPEND) { + hub->quirk_disable_autosuspend = 1; + usb_autopm_get_interface_no_resume(intf); + } + if (hub_configure(hub, &desc->endpoint[0].desc) >= 0) return 0; @@ -5599,6 +5609,10 @@ out_hdev_lock: } static const struct usb_device_id hub_id_table[] = { + { .match_flags = USB_DEVICE_ID_MATCH_VENDOR | USB_DEVICE_ID_MATCH_INT_CLASS, + .idVendor = USB_VENDOR_SMSC, + .bInterfaceClass = USB_CLASS_HUB, + .driver_info = HUB_QUIRK_DISABLE_AUTOSUSPEND}, { .match_flags = USB_DEVICE_ID_MATCH_VENDOR | USB_DEVICE_ID_MATCH_INT_CLASS, .idVendor = USB_VENDOR_GENESYS_LOGIC, diff --git a/drivers/usb/core/hub.h b/drivers/usb/core/hub.h index a9e24e4b8df1..a97dd1ba964e 100644 --- a/drivers/usb/core/hub.h +++ b/drivers/usb/core/hub.h @@ -61,6 +61,7 @@ struct usb_hub { unsigned quiescing:1; unsigned disconnected:1; unsigned in_reset:1; + unsigned quirk_disable_autosuspend:1; unsigned quirk_check_port_auto_suspend:1; diff --git a/drivers/usb/core/port.c b/drivers/usb/core/port.c index bbbb35fa639f..235a7c645503 100644 --- a/drivers/usb/core/port.c +++ b/drivers/usb/core/port.c @@ -213,7 +213,10 @@ static int usb_port_runtime_resume(struct device *dev) if (!port_dev->is_superspeed && peer) pm_runtime_get_sync(&peer->dev); - usb_autopm_get_interface(intf); + retval = usb_autopm_get_interface(intf); + if (retval < 0) + return retval; + retval = usb_hub_set_port_power(hdev, hub, port1, true); msleep(hub_power_on_good_delay(hub)); if (udev && !retval) { @@ -266,7 +269,10 @@ static int usb_port_runtime_suspend(struct device *dev) if (usb_port_block_power_off) return -EBUSY; - usb_autopm_get_interface(intf); + retval = usb_autopm_get_interface(intf); + if (retval < 0) + return retval; + retval = usb_hub_set_port_power(hdev, hub, port1, false); usb_clear_port_feature(hdev, port1, USB_PORT_FEAT_C_CONNECTION); if (!port_dev->is_superspeed) diff --git a/drivers/usb/core/quirks.c b/drivers/usb/core/quirks.c index 6b6413073584..da30b5664ff3 100644 --- a/drivers/usb/core/quirks.c +++ b/drivers/usb/core/quirks.c @@ -231,6 +231,9 @@ static const struct usb_device_id usb_quirk_list[] = { /* Logitech PTZ Pro Camera */ { USB_DEVICE(0x046d, 0x0853), .driver_info = USB_QUIRK_DELAY_INIT }, + /* Logitech Screen Share */ + { USB_DEVICE(0x046d, 0x086c), .driver_info = USB_QUIRK_NO_LPM }, + /* Logitech Quickcam Fusion */ { USB_DEVICE(0x046d, 0x08c1), .driver_info = USB_QUIRK_RESET_RESUME }, @@ -354,6 +357,10 @@ static const struct usb_device_id usb_quirk_list[] = { { USB_DEVICE(0x0904, 0x6103), .driver_info = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL }, + /* Sound Devices USBPre2 */ + { USB_DEVICE(0x0926, 0x0202), .driver_info = + USB_QUIRK_ENDPOINT_BLACKLIST }, + /* Keytouch QWERTY Panel keyboard */ { USB_DEVICE(0x0926, 0x3333), .driver_info = USB_QUIRK_CONFIG_INTF_STRINGS }, @@ -371,6 +378,12 @@ static const struct usb_device_id usb_quirk_list[] = { { USB_DEVICE(0x0b05, 0x17e0), .driver_info = USB_QUIRK_IGNORE_REMOTE_WAKEUP }, + /* Realtek hub in Dell WD19 (Type-C) */ + { USB_DEVICE(0x0bda, 0x0487), .driver_info = USB_QUIRK_NO_LPM }, + + /* Generic RTL8153 based ethernet adapters */ + { USB_DEVICE(0x0bda, 0x8153), .driver_info = USB_QUIRK_NO_LPM }, + /* Action Semiconductor flash disk */ { USB_DEVICE(0x10d6, 0x2200), .driver_info = USB_QUIRK_STRING_FETCH_255 }, @@ -445,6 +458,9 @@ static const struct usb_device_id usb_quirk_list[] = { /* INTEL VALUE SSD */ { USB_DEVICE(0x8086, 0xf1a5), .driver_info = USB_QUIRK_RESET_RESUME }, + /* novation SoundControl XL */ + { USB_DEVICE(0x1235, 0x0061), .driver_info = USB_QUIRK_RESET_RESUME }, + { } /* terminating entry must be last */ }; @@ -472,6 +488,39 @@ static const struct usb_device_id usb_amd_resume_quirk_list[] = { { } /* terminating entry must be last */ }; +/* + * Entries for blacklisted endpoints that should be ignored when parsing + * configuration descriptors. + * + * Matched for devices with USB_QUIRK_ENDPOINT_BLACKLIST. + */ +static const struct usb_device_id usb_endpoint_blacklist[] = { + { USB_DEVICE_INTERFACE_NUMBER(0x0926, 0x0202, 1), .driver_info = 0x85 }, + { } +}; + +bool usb_endpoint_is_blacklisted(struct usb_device *udev, + struct usb_host_interface *intf, + struct usb_endpoint_descriptor *epd) +{ + const struct usb_device_id *id; + unsigned int address; + + for (id = usb_endpoint_blacklist; id->match_flags; ++id) { + if (!usb_match_device(udev, id)) + continue; + + if (!usb_match_one_id_intf(udev, intf, id)) + continue; + + address = id->driver_info; + if (address == epd->bEndpointAddress) + return true; + } + + return false; +} + static bool usb_match_any_interface(struct usb_device *udev, const struct usb_device_id *id) { diff --git a/drivers/usb/core/usb.h b/drivers/usb/core/usb.h index cf4783cf661a..3ad0ee57e859 100644 --- a/drivers/usb/core/usb.h +++ b/drivers/usb/core/usb.h @@ -37,6 +37,9 @@ extern void usb_authorize_interface(struct usb_interface *); extern void usb_detect_quirks(struct usb_device *udev); extern void usb_detect_interface_quirks(struct usb_device *udev); extern void usb_release_quirk_list(void); +extern bool usb_endpoint_is_blacklisted(struct usb_device *udev, + struct usb_host_interface *intf, + struct usb_endpoint_descriptor *epd); extern int usb_remove_device(struct usb_device *udev); extern int usb_get_device_descriptor(struct usb_device *dev, diff --git a/drivers/usb/dwc2/gadget.c b/drivers/usb/dwc2/gadget.c index 88f7d6d4ff2d..92ed32ec1607 100644 --- a/drivers/usb/dwc2/gadget.c +++ b/drivers/usb/dwc2/gadget.c @@ -1083,11 +1083,6 @@ static void dwc2_hsotg_start_req(struct dwc2_hsotg *hsotg, else packets = 1; /* send one packet if length is zero. */ - if (hs_ep->isochronous && length > (hs_ep->mc * hs_ep->ep.maxpacket)) { - dev_err(hsotg->dev, "req length > maxpacket*mc\n"); - return; - } - if (dir_in && index != 0) if (hs_ep->isochronous) epsize = DXEPTSIZ_MC(packets); @@ -1391,6 +1386,13 @@ static int dwc2_hsotg_ep_queue(struct usb_ep *ep, struct usb_request *req, req->actual = 0; req->status = -EINPROGRESS; + /* Don't queue ISOC request if length greater than mps*mc */ + if (hs_ep->isochronous && + req->length > (hs_ep->mc * hs_ep->ep.maxpacket)) { + dev_err(hs->dev, "req length > maxpacket*mc\n"); + return -EINVAL; + } + /* In DDMA mode for ISOC's don't queue request if length greater * than descriptor limits. */ @@ -1632,6 +1634,7 @@ static int dwc2_hsotg_process_req_status(struct dwc2_hsotg *hsotg, struct dwc2_hsotg_ep *ep0 = hsotg->eps_out[0]; struct dwc2_hsotg_ep *ep; __le16 reply; + u16 status; int ret; dev_dbg(hsotg->dev, "%s: USB_REQ_GET_STATUS\n", __func__); @@ -1643,11 +1646,10 @@ static int dwc2_hsotg_process_req_status(struct dwc2_hsotg *hsotg, switch (ctrl->bRequestType & USB_RECIP_MASK) { case USB_RECIP_DEVICE: - /* - * bit 0 => self powered - * bit 1 => remote wakeup - */ - reply = cpu_to_le16(0); + status = 1 << USB_DEVICE_SELF_POWERED; + status |= hsotg->remote_wakeup_allowed << + USB_DEVICE_REMOTE_WAKEUP; + reply = cpu_to_le16(status); break; case USB_RECIP_INTERFACE: @@ -1758,7 +1760,10 @@ static int dwc2_hsotg_process_req_feature(struct dwc2_hsotg *hsotg, case USB_RECIP_DEVICE: switch (wValue) { case USB_DEVICE_REMOTE_WAKEUP: - hsotg->remote_wakeup_allowed = 1; + if (set) + hsotg->remote_wakeup_allowed = 1; + else + hsotg->remote_wakeup_allowed = 0; break; case USB_DEVICE_TEST_MODE: @@ -1768,16 +1773,17 @@ static int dwc2_hsotg_process_req_feature(struct dwc2_hsotg *hsotg, return -EINVAL; hsotg->test_mode = wIndex >> 8; - ret = dwc2_hsotg_send_reply(hsotg, ep0, NULL, 0); - if (ret) { - dev_err(hsotg->dev, - "%s: failed to send reply\n", __func__); - return ret; - } break; default: return -ENOENT; } + + ret = dwc2_hsotg_send_reply(hsotg, ep0, NULL, 0); + if (ret) { + dev_err(hsotg->dev, + "%s: failed to send reply\n", __func__); + return ret; + } break; case USB_RECIP_ENDPOINT: diff --git a/drivers/usb/dwc3/debug.h b/drivers/usb/dwc3/debug.h index e56beb9d1e36..4a13ceaf4093 100644 --- a/drivers/usb/dwc3/debug.h +++ b/drivers/usb/dwc3/debug.h @@ -256,86 +256,77 @@ static inline const char *dwc3_ep_event_string(char *str, size_t size, u8 epnum = event->endpoint_number; size_t len; int status; - int ret; - ret = snprintf(str, size, "ep%d%s: ", epnum >> 1, + len = scnprintf(str, size, "ep%d%s: ", epnum >> 1, (epnum & 1) ? "in" : "out"); - if (ret < 0) - return "UNKNOWN"; status = event->status; switch (event->endpoint_event) { case DWC3_DEPEVT_XFERCOMPLETE: - len = strlen(str); - snprintf(str + len, size - len, "Transfer Complete (%c%c%c)", + len += scnprintf(str + len, size - len, + "Transfer Complete (%c%c%c)", status & DEPEVT_STATUS_SHORT ? 'S' : 's', status & DEPEVT_STATUS_IOC ? 'I' : 'i', status & DEPEVT_STATUS_LST ? 'L' : 'l'); - len = strlen(str); - if (epnum <= 1) - snprintf(str + len, size - len, " [%s]", + scnprintf(str + len, size - len, " [%s]", dwc3_ep0_state_string(ep0state)); break; case DWC3_DEPEVT_XFERINPROGRESS: - len = strlen(str); - - snprintf(str + len, size - len, "Transfer In Progress [%d] (%c%c%c)", + scnprintf(str + len, size - len, + "Transfer In Progress [%d] (%c%c%c)", event->parameters, status & DEPEVT_STATUS_SHORT ? 'S' : 's', status & DEPEVT_STATUS_IOC ? 'I' : 'i', status & DEPEVT_STATUS_LST ? 'M' : 'm'); break; case DWC3_DEPEVT_XFERNOTREADY: - len = strlen(str); - - snprintf(str + len, size - len, "Transfer Not Ready [%d]%s", + len += scnprintf(str + len, size - len, + "Transfer Not Ready [%d]%s", event->parameters, status & DEPEVT_STATUS_TRANSFER_ACTIVE ? " (Active)" : " (Not Active)"); - len = strlen(str); - /* Control Endpoints */ if (epnum <= 1) { int phase = DEPEVT_STATUS_CONTROL_PHASE(event->status); switch (phase) { case DEPEVT_STATUS_CONTROL_DATA: - snprintf(str + ret, size - ret, + scnprintf(str + len, size - len, " [Data Phase]"); break; case DEPEVT_STATUS_CONTROL_STATUS: - snprintf(str + ret, size - ret, + scnprintf(str + len, size - len, " [Status Phase]"); } } break; case DWC3_DEPEVT_RXTXFIFOEVT: - snprintf(str + ret, size - ret, "FIFO"); + scnprintf(str + len, size - len, "FIFO"); break; case DWC3_DEPEVT_STREAMEVT: status = event->status; switch (status) { case DEPEVT_STREAMEVT_FOUND: - snprintf(str + ret, size - ret, " Stream %d Found", + scnprintf(str + len, size - len, " Stream %d Found", event->parameters); break; case DEPEVT_STREAMEVT_NOTFOUND: default: - snprintf(str + ret, size - ret, " Stream Not Found"); + scnprintf(str + len, size - len, " Stream Not Found"); break; } break; case DWC3_DEPEVT_EPCMDCMPLT: - snprintf(str + ret, size - ret, "Endpoint Command Complete"); + scnprintf(str + len, size - len, "Endpoint Command Complete"); break; default: - snprintf(str, size, "UNKNOWN"); + scnprintf(str + len, size - len, "UNKNOWN"); } return str; diff --git a/drivers/usb/dwc3/gadget.c b/drivers/usb/dwc3/gadget.c index 1b8014ab0b25..1e00bf2d65a2 100644 --- a/drivers/usb/dwc3/gadget.c +++ b/drivers/usb/dwc3/gadget.c @@ -1071,7 +1071,14 @@ static void dwc3_prepare_one_trb_sg(struct dwc3_ep *dep, unsigned int rem = length % maxp; unsigned chain = true; - if (sg_is_last(s)) + /* + * IOMMU driver is coalescing the list of sgs which shares a + * page boundary into one and giving it to USB driver. With + * this the number of sgs mapped is not equal to the number of + * sgs passed. So mark the chain bit to false if it isthe last + * mapped sg. + */ + if (i == remaining - 1) chain = false; if (rem && usb_endpoint_dir_out(dep->endpoint.desc) && !chain) { @@ -2429,7 +2436,8 @@ static int dwc3_gadget_ep_reclaim_completed_trb(struct dwc3_ep *dep, if (event->status & DEPEVT_STATUS_SHORT && !chain) return 1; - if (event->status & DEPEVT_STATUS_IOC) + if ((trb->ctrl & DWC3_TRB_CTRL_IOC) || + (trb->ctrl & DWC3_TRB_CTRL_LST)) return 1; return 0; diff --git a/drivers/usb/gadget/composite.c b/drivers/usb/gadget/composite.c index 3b4f67000315..223f72d4d9ed 100644 --- a/drivers/usb/gadget/composite.c +++ b/drivers/usb/gadget/composite.c @@ -437,12 +437,14 @@ static u8 encode_bMaxPower(enum usb_device_speed speed, val = CONFIG_USB_GADGET_VBUS_DRAW; if (!val) return 0; - switch (speed) { - case USB_SPEED_SUPER: - return DIV_ROUND_UP(val, 8); - default: - return DIV_ROUND_UP(val, 2); - } + if (speed < USB_SPEED_SUPER) + return min(val, 500U) / 2; + else + /* + * USB 3.x supports up to 900mA, but since 900 isn't divisible + * by 8 the integral division will effectively cap to 896mA. + */ + return min(val, 900U) / 8; } static int config_buf(struct usb_configuration *config, @@ -854,6 +856,10 @@ static int set_config(struct usb_composite_dev *cdev, /* when we return, be sure our power usage is valid */ power = c->MaxPower ? c->MaxPower : CONFIG_USB_GADGET_VBUS_DRAW; + if (gadget->speed < USB_SPEED_SUPER) + power = min(power, 500U); + else + power = min(power, 900U); done: usb_gadget_vbus_draw(gadget, power); if (result >= 0 && cdev->delayed_status) @@ -2280,7 +2286,7 @@ void composite_resume(struct usb_gadget *gadget) { struct usb_composite_dev *cdev = get_gadget_data(gadget); struct usb_function *f; - u16 maxpower; + unsigned maxpower; /* REVISIT: should we have config level * suspend/resume callbacks? @@ -2294,10 +2300,14 @@ void composite_resume(struct usb_gadget *gadget) f->resume(f); } - maxpower = cdev->config->MaxPower; + maxpower = cdev->config->MaxPower ? + cdev->config->MaxPower : CONFIG_USB_GADGET_VBUS_DRAW; + if (gadget->speed < USB_SPEED_SUPER) + maxpower = min(maxpower, 500U); + else + maxpower = min(maxpower, 900U); - usb_gadget_vbus_draw(gadget, maxpower ? - maxpower : CONFIG_USB_GADGET_VBUS_DRAW); + usb_gadget_vbus_draw(gadget, maxpower); } cdev->suspended = 0; diff --git a/drivers/usb/gadget/function/f_fs.c b/drivers/usb/gadget/function/f_fs.c index 6171d28331e6..571917677d35 100644 --- a/drivers/usb/gadget/function/f_fs.c +++ b/drivers/usb/gadget/function/f_fs.c @@ -1162,18 +1162,19 @@ static int ffs_aio_cancel(struct kiocb *kiocb) { struct ffs_io_data *io_data = kiocb->private; struct ffs_epfile *epfile = kiocb->ki_filp->private_data; + unsigned long flags; int value; ENTER(); - spin_lock_irq(&epfile->ffs->eps_lock); + spin_lock_irqsave(&epfile->ffs->eps_lock, flags); if (likely(io_data && io_data->ep && io_data->req)) value = usb_ep_dequeue(io_data->ep, io_data->req); else value = -EINVAL; - spin_unlock_irq(&epfile->ffs->eps_lock); + spin_unlock_irqrestore(&epfile->ffs->eps_lock, flags); return value; } diff --git a/drivers/usb/gadget/function/u_audio.c b/drivers/usb/gadget/function/u_audio.c index 6d956f190f5a..e6d32c536781 100644 --- a/drivers/usb/gadget/function/u_audio.c +++ b/drivers/usb/gadget/function/u_audio.c @@ -361,7 +361,7 @@ int u_audio_start_capture(struct g_audio *audio_dev) ep = audio_dev->out_ep; prm = &uac->c_prm; config_ep_by_speed(gadget, &audio_dev->func, ep); - req_len = prm->max_psize; + req_len = ep->maxpacket; prm->ep_enabled = true; usb_ep_enable(ep); @@ -379,7 +379,7 @@ int u_audio_start_capture(struct g_audio *audio_dev) req->context = &prm->ureq[i]; req->length = req_len; req->complete = u_audio_iso_complete; - req->buf = prm->rbuf + i * prm->max_psize; + req->buf = prm->rbuf + i * ep->maxpacket; } if (usb_ep_queue(ep, prm->ureq[i].req, GFP_ATOMIC)) @@ -430,9 +430,9 @@ int u_audio_start_playback(struct g_audio *audio_dev) uac->p_pktsize = min_t(unsigned int, uac->p_framesize * (params->p_srate / uac->p_interval), - prm->max_psize); + ep->maxpacket); - if (uac->p_pktsize < prm->max_psize) + if (uac->p_pktsize < ep->maxpacket) uac->p_pktsize_residue = uac->p_framesize * (params->p_srate % uac->p_interval); else @@ -457,7 +457,7 @@ int u_audio_start_playback(struct g_audio *audio_dev) req->context = &prm->ureq[i]; req->length = req_len; req->complete = u_audio_iso_complete; - req->buf = prm->rbuf + i * prm->max_psize; + req->buf = prm->rbuf + i * ep->maxpacket; } if (usb_ep_queue(ep, prm->ureq[i].req, GFP_ATOMIC)) diff --git a/drivers/usb/gadget/function/u_serial.c b/drivers/usb/gadget/function/u_serial.c index f986e5c55974..8167d379e115 100644 --- a/drivers/usb/gadget/function/u_serial.c +++ b/drivers/usb/gadget/function/u_serial.c @@ -561,8 +561,10 @@ static int gs_start_io(struct gs_port *port) port->n_read = 0; started = gs_start_rx(port); - /* unblock any pending writes into our circular buffer */ if (started) { + gs_start_tx(port); + /* Unblock any pending writes into our circular buffer, in case + * we didn't in gs_start_tx() */ tty_wakeup(port->port.tty); } else { gs_free_requests(ep, head, &port->read_allocated); diff --git a/drivers/usb/gadget/udc/udc-xilinx.c b/drivers/usb/gadget/udc/udc-xilinx.c index 29d8e5f8bb58..b1cfc8279c3d 100644 --- a/drivers/usb/gadget/udc/udc-xilinx.c +++ b/drivers/usb/gadget/udc/udc-xilinx.c @@ -1399,7 +1399,6 @@ err: /** * xudc_stop - stops the device. * @gadget: pointer to the usb gadget structure - * @driver: pointer to usb gadget driver structure * * Return: zero always */ diff --git a/drivers/usb/host/xhci-hub.c b/drivers/usb/host/xhci-hub.c index 7a3a29e5e9d2..af92b2576fe9 100644 --- a/drivers/usb/host/xhci-hub.c +++ b/drivers/usb/host/xhci-hub.c @@ -55,6 +55,7 @@ static u8 usb_bos_descriptor [] = { static int xhci_create_usb3_bos_desc(struct xhci_hcd *xhci, char *buf, u16 wLength) { + struct xhci_port_cap *port_cap = NULL; int i, ssa_count; u32 temp; u16 desc_size, ssp_cap_size, ssa_size = 0; @@ -64,16 +65,24 @@ static int xhci_create_usb3_bos_desc(struct xhci_hcd *xhci, char *buf, ssp_cap_size = sizeof(usb_bos_descriptor) - desc_size; /* does xhci support USB 3.1 Enhanced SuperSpeed */ - if (xhci->usb3_rhub.min_rev >= 0x01) { + for (i = 0; i < xhci->num_port_caps; i++) { + if (xhci->port_caps[i].maj_rev == 0x03 && + xhci->port_caps[i].min_rev >= 0x01) { + usb3_1 = true; + port_cap = &xhci->port_caps[i]; + break; + } + } + + if (usb3_1) { /* does xhci provide a PSI table for SSA speed attributes? */ - if (xhci->usb3_rhub.psi_count) { + if (port_cap->psi_count) { /* two SSA entries for each unique PSI ID, RX and TX */ - ssa_count = xhci->usb3_rhub.psi_uid_count * 2; + ssa_count = port_cap->psi_uid_count * 2; ssa_size = ssa_count * sizeof(u32); ssp_cap_size -= 16; /* skip copying the default SSA */ } desc_size += ssp_cap_size; - usb3_1 = true; } memcpy(buf, &usb_bos_descriptor, min(desc_size, wLength)); @@ -99,7 +108,7 @@ static int xhci_create_usb3_bos_desc(struct xhci_hcd *xhci, char *buf, } /* If PSI table exists, add the custom speed attributes from it */ - if (usb3_1 && xhci->usb3_rhub.psi_count) { + if (usb3_1 && port_cap->psi_count) { u32 ssp_cap_base, bm_attrib, psi, psi_mant, psi_exp; int offset; @@ -111,7 +120,7 @@ static int xhci_create_usb3_bos_desc(struct xhci_hcd *xhci, char *buf, /* attribute count SSAC bits 4:0 and ID count SSIC bits 8:5 */ bm_attrib = (ssa_count - 1) & 0x1f; - bm_attrib |= (xhci->usb3_rhub.psi_uid_count - 1) << 5; + bm_attrib |= (port_cap->psi_uid_count - 1) << 5; put_unaligned_le32(bm_attrib, &buf[ssp_cap_base + 4]); if (wLength < desc_size + ssa_size) @@ -124,8 +133,8 @@ static int xhci_create_usb3_bos_desc(struct xhci_hcd *xhci, char *buf, * USB 3.1 requires two SSA entries (RX and TX) for every link */ offset = desc_size; - for (i = 0; i < xhci->usb3_rhub.psi_count; i++) { - psi = xhci->usb3_rhub.psi[i]; + for (i = 0; i < port_cap->psi_count; i++) { + psi = port_cap->psi[i]; psi &= ~USB_SSP_SUBLINK_SPEED_RSVD; psi_exp = XHCI_EXT_PORT_PSIE(psi); psi_mant = XHCI_EXT_PORT_PSIM(psi); diff --git a/drivers/usb/host/xhci-mem.c b/drivers/usb/host/xhci-mem.c index 3b1388fa2f36..884c601bfa15 100644 --- a/drivers/usb/host/xhci-mem.c +++ b/drivers/usb/host/xhci-mem.c @@ -1475,9 +1475,15 @@ int xhci_endpoint_init(struct xhci_hcd *xhci, /* Allow 3 retries for everything but isoc, set CErr = 3 */ if (!usb_endpoint_xfer_isoc(&ep->desc)) err_count = 3; - /* Some devices get this wrong */ - if (usb_endpoint_xfer_bulk(&ep->desc) && udev->speed == USB_SPEED_HIGH) - max_packet = 512; + /* HS bulk max packet should be 512, FS bulk supports 8, 16, 32 or 64 */ + if (usb_endpoint_xfer_bulk(&ep->desc)) { + if (udev->speed == USB_SPEED_HIGH) + max_packet = 512; + if (udev->speed == USB_SPEED_FULL) { + max_packet = rounddown_pow_of_two(max_packet); + max_packet = clamp_val(max_packet, 8, 64); + } + } /* xHCI 1.0 and 1.1 indicates that ctrl ep avg TRB Length should be 8 */ if (usb_endpoint_xfer_control(&ep->desc) && xhci->hci_version >= 0x100) avg_trb_len = 8; @@ -1909,17 +1915,17 @@ no_bw: xhci->usb3_rhub.num_ports = 0; xhci->num_active_eps = 0; kfree(xhci->usb2_rhub.ports); - kfree(xhci->usb2_rhub.psi); kfree(xhci->usb3_rhub.ports); - kfree(xhci->usb3_rhub.psi); kfree(xhci->hw_ports); kfree(xhci->rh_bw); kfree(xhci->ext_caps); + for (i = 0; i < xhci->num_port_caps; i++) + kfree(xhci->port_caps[i].psi); + kfree(xhci->port_caps); + xhci->num_port_caps = 0; xhci->usb2_rhub.ports = NULL; - xhci->usb2_rhub.psi = NULL; xhci->usb3_rhub.ports = NULL; - xhci->usb3_rhub.psi = NULL; xhci->hw_ports = NULL; xhci->rh_bw = NULL; xhci->ext_caps = NULL; @@ -2120,6 +2126,7 @@ static void xhci_add_in_port(struct xhci_hcd *xhci, unsigned int num_ports, u8 major_revision, minor_revision; struct xhci_hub *rhub; struct device *dev = xhci_to_hcd(xhci)->self.sysdev; + struct xhci_port_cap *port_cap; temp = readl(addr); major_revision = XHCI_EXT_PORT_MAJOR(temp); @@ -2154,31 +2161,39 @@ static void xhci_add_in_port(struct xhci_hcd *xhci, unsigned int num_ports, /* WTF? "Valid values are ‘1’ to MaxPorts" */ return; - rhub->psi_count = XHCI_EXT_PORT_PSIC(temp); - if (rhub->psi_count) { - rhub->psi = kcalloc_node(rhub->psi_count, sizeof(*rhub->psi), - GFP_KERNEL, dev_to_node(dev)); - if (!rhub->psi) - rhub->psi_count = 0; + port_cap = &xhci->port_caps[xhci->num_port_caps++]; + if (xhci->num_port_caps > max_caps) + return; + + port_cap->maj_rev = major_revision; + port_cap->min_rev = minor_revision; + port_cap->psi_count = XHCI_EXT_PORT_PSIC(temp); - rhub->psi_uid_count++; - for (i = 0; i < rhub->psi_count; i++) { - rhub->psi[i] = readl(addr + 4 + i); + if (port_cap->psi_count) { + port_cap->psi = kcalloc_node(port_cap->psi_count, + sizeof(*port_cap->psi), + GFP_KERNEL, dev_to_node(dev)); + if (!port_cap->psi) + port_cap->psi_count = 0; + + port_cap->psi_uid_count++; + for (i = 0; i < port_cap->psi_count; i++) { + port_cap->psi[i] = readl(addr + 4 + i); /* count unique ID values, two consecutive entries can * have the same ID if link is assymetric */ - if (i && (XHCI_EXT_PORT_PSIV(rhub->psi[i]) != - XHCI_EXT_PORT_PSIV(rhub->psi[i - 1]))) - rhub->psi_uid_count++; + if (i && (XHCI_EXT_PORT_PSIV(port_cap->psi[i]) != + XHCI_EXT_PORT_PSIV(port_cap->psi[i - 1]))) + port_cap->psi_uid_count++; xhci_dbg(xhci, "PSIV:%d PSIE:%d PLT:%d PFD:%d LP:%d PSIM:%d\n", - XHCI_EXT_PORT_PSIV(rhub->psi[i]), - XHCI_EXT_PORT_PSIE(rhub->psi[i]), - XHCI_EXT_PORT_PLT(rhub->psi[i]), - XHCI_EXT_PORT_PFD(rhub->psi[i]), - XHCI_EXT_PORT_LP(rhub->psi[i]), - XHCI_EXT_PORT_PSIM(rhub->psi[i])); + XHCI_EXT_PORT_PSIV(port_cap->psi[i]), + XHCI_EXT_PORT_PSIE(port_cap->psi[i]), + XHCI_EXT_PORT_PLT(port_cap->psi[i]), + XHCI_EXT_PORT_PFD(port_cap->psi[i]), + XHCI_EXT_PORT_LP(port_cap->psi[i]), + XHCI_EXT_PORT_PSIM(port_cap->psi[i])); } } /* cache usb2 port capabilities */ @@ -2213,6 +2228,7 @@ static void xhci_add_in_port(struct xhci_hcd *xhci, unsigned int num_ports, continue; } hw_port->rhub = rhub; + hw_port->port_cap = port_cap; rhub->num_ports++; } /* FIXME: Should we disable ports not in the Extended Capabilities? */ @@ -2303,6 +2319,11 @@ static int xhci_setup_port_arrays(struct xhci_hcd *xhci, gfp_t flags) if (!xhci->ext_caps) return -ENOMEM; + xhci->port_caps = kcalloc_node(cap_count, sizeof(*xhci->port_caps), + flags, dev_to_node(dev)); + if (!xhci->port_caps) + return -ENOMEM; + offset = cap_start; while (offset) { diff --git a/drivers/usb/host/xhci-pci.c b/drivers/usb/host/xhci-pci.c index 4917c5b033fa..1fddc41fa1f3 100644 --- a/drivers/usb/host/xhci-pci.c +++ b/drivers/usb/host/xhci-pci.c @@ -49,6 +49,7 @@ #define PCI_DEVICE_ID_INTEL_TITAN_RIDGE_4C_XHCI 0x15ec #define PCI_DEVICE_ID_INTEL_TITAN_RIDGE_DD_XHCI 0x15f0 #define PCI_DEVICE_ID_INTEL_ICE_LAKE_XHCI 0x8a13 +#define PCI_DEVICE_ID_INTEL_CML_XHCI 0xa3af #define PCI_DEVICE_ID_AMD_PROMONTORYA_4 0x43b9 #define PCI_DEVICE_ID_AMD_PROMONTORYA_3 0x43ba @@ -135,7 +136,8 @@ static void xhci_pci_quirks(struct device *dev, struct xhci_hcd *xhci) xhci->quirks |= XHCI_AMD_PLL_FIX; if (pdev->vendor == PCI_VENDOR_ID_AMD && - (pdev->device == 0x15e0 || + (pdev->device == 0x145c || + pdev->device == 0x15e0 || pdev->device == 0x15e1 || pdev->device == 0x43bb)) xhci->quirks |= XHCI_SUSPEND_DELAY; @@ -187,7 +189,8 @@ static void xhci_pci_quirks(struct device *dev, struct xhci_hcd *xhci) pdev->device == PCI_DEVICE_ID_INTEL_BROXTON_M_XHCI || pdev->device == PCI_DEVICE_ID_INTEL_BROXTON_B_XHCI || pdev->device == PCI_DEVICE_ID_INTEL_APL_XHCI || - pdev->device == PCI_DEVICE_ID_INTEL_DNV_XHCI)) { + pdev->device == PCI_DEVICE_ID_INTEL_DNV_XHCI || + pdev->device == PCI_DEVICE_ID_INTEL_CML_XHCI)) { xhci->quirks |= XHCI_PME_STUCK_QUIRK; } if (pdev->vendor == PCI_VENDOR_ID_INTEL && @@ -302,6 +305,9 @@ static int xhci_pci_setup(struct usb_hcd *hcd) if (!usb_hcd_is_primary_hcd(hcd)) return 0; + if (xhci->quirks & XHCI_PME_STUCK_QUIRK) + xhci_pme_acpi_rtd3_enable(pdev); + xhci_dbg(xhci, "Got SBRN %u\n", (unsigned int) xhci->sbrn); /* Find any debug ports */ @@ -359,9 +365,6 @@ static int xhci_pci_probe(struct pci_dev *dev, const struct pci_device_id *id) HCC_MAX_PSA(xhci->hcc_params) >= 4) xhci->shared_hcd->can_do_streams = 1; - if (xhci->quirks & XHCI_PME_STUCK_QUIRK) - xhci_pme_acpi_rtd3_enable(dev); - /* USB-2 and USB-3 roothubs initialized, allow runtime pm suspend */ pm_runtime_put_noidle(&dev->dev); diff --git a/drivers/usb/host/xhci-plat.c b/drivers/usb/host/xhci-plat.c index d90cd5ec09cf..315b4552693c 100644 --- a/drivers/usb/host/xhci-plat.c +++ b/drivers/usb/host/xhci-plat.c @@ -445,6 +445,7 @@ MODULE_DEVICE_TABLE(acpi, usb_xhci_acpi_match); static struct platform_driver usb_xhci_driver = { .probe = xhci_plat_probe, .remove = xhci_plat_remove, + .shutdown = usb_hcd_platform_shutdown, .driver = { .name = "xhci-hcd", .pm = &xhci_plat_pm_ops, diff --git a/drivers/usb/host/xhci-trace.h b/drivers/usb/host/xhci-trace.h index 56eb867803a6..b19582b2a72c 100644 --- a/drivers/usb/host/xhci-trace.h +++ b/drivers/usb/host/xhci-trace.h @@ -289,23 +289,12 @@ DECLARE_EVENT_CLASS(xhci_log_urb, ), TP_printk("ep%d%s-%s: urb %p pipe %u slot %d length %d/%d sgs %d/%d stream %d flags %08x", __entry->epnum, __entry->dir_in ? "in" : "out", - ({ char *s; - switch (__entry->type) { - case USB_ENDPOINT_XFER_INT: - s = "intr"; - break; - case USB_ENDPOINT_XFER_CONTROL: - s = "control"; - break; - case USB_ENDPOINT_XFER_BULK: - s = "bulk"; - break; - case USB_ENDPOINT_XFER_ISOC: - s = "isoc"; - break; - default: - s = "UNKNOWN"; - } s; }), __entry->urb, __entry->pipe, __entry->slot_id, + __print_symbolic(__entry->type, + { USB_ENDPOINT_XFER_INT, "intr" }, + { USB_ENDPOINT_XFER_CONTROL, "control" }, + { USB_ENDPOINT_XFER_BULK, "bulk" }, + { USB_ENDPOINT_XFER_ISOC, "isoc" }), + __entry->urb, __entry->pipe, __entry->slot_id, __entry->actual, __entry->length, __entry->num_mapped_sgs, __entry->num_sgs, __entry->stream, __entry->flags ) diff --git a/drivers/usb/host/xhci.h b/drivers/usb/host/xhci.h index 13d8838cd552..3ecee10fdcdc 100644 --- a/drivers/usb/host/xhci.h +++ b/drivers/usb/host/xhci.h @@ -1702,12 +1702,20 @@ struct xhci_bus_state { * Intel Lynx Point LP xHCI host. */ #define XHCI_MAX_REXIT_TIMEOUT_MS 20 +struct xhci_port_cap { + u32 *psi; /* array of protocol speed ID entries */ + u8 psi_count; + u8 psi_uid_count; + u8 maj_rev; + u8 min_rev; +}; struct xhci_port { __le32 __iomem *addr; int hw_portnum; int hcd_portnum; struct xhci_hub *rhub; + struct xhci_port_cap *port_cap; }; struct xhci_hub { @@ -1719,9 +1727,6 @@ struct xhci_hub { /* supported prococol extended capabiliy values */ u8 maj_rev; u8 min_rev; - u32 *psi; /* array of protocol speed ID entries */ - u8 psi_count; - u8 psi_uid_count; }; /* There is one xhci_hcd structure per controller */ @@ -1880,6 +1885,9 @@ struct xhci_hcd { /* cached usb2 extened protocol capabilites */ u32 *ext_caps; unsigned int num_ext_caps; + /* cached extended protocol port capabilities */ + struct xhci_port_cap *port_caps; + unsigned int num_port_caps; /* Compliance Mode Recovery Data */ struct timer_list comp_mode_recovery_timer; u32 port_status_u0; diff --git a/drivers/usb/misc/iowarrior.c b/drivers/usb/misc/iowarrior.c index dce44fbf031f..dce20301e367 100644 --- a/drivers/usb/misc/iowarrior.c +++ b/drivers/usb/misc/iowarrior.c @@ -33,6 +33,14 @@ #define USB_DEVICE_ID_CODEMERCS_IOWPV2 0x1512 /* full speed iowarrior */ #define USB_DEVICE_ID_CODEMERCS_IOW56 0x1503 +/* fuller speed iowarrior */ +#define USB_DEVICE_ID_CODEMERCS_IOW28 0x1504 +#define USB_DEVICE_ID_CODEMERCS_IOW28L 0x1505 +#define USB_DEVICE_ID_CODEMERCS_IOW100 0x1506 + +/* OEMed devices */ +#define USB_DEVICE_ID_CODEMERCS_IOW24SAG 0x158a +#define USB_DEVICE_ID_CODEMERCS_IOW56AM 0x158b /* Get a minor range for your devices from the usb maintainer */ #ifdef CONFIG_USB_DYNAMIC_MINORS @@ -133,6 +141,11 @@ static const struct usb_device_id iowarrior_ids[] = { {USB_DEVICE(USB_VENDOR_ID_CODEMERCS, USB_DEVICE_ID_CODEMERCS_IOWPV1)}, {USB_DEVICE(USB_VENDOR_ID_CODEMERCS, USB_DEVICE_ID_CODEMERCS_IOWPV2)}, {USB_DEVICE(USB_VENDOR_ID_CODEMERCS, USB_DEVICE_ID_CODEMERCS_IOW56)}, + {USB_DEVICE(USB_VENDOR_ID_CODEMERCS, USB_DEVICE_ID_CODEMERCS_IOW24SAG)}, + {USB_DEVICE(USB_VENDOR_ID_CODEMERCS, USB_DEVICE_ID_CODEMERCS_IOW56AM)}, + {USB_DEVICE(USB_VENDOR_ID_CODEMERCS, USB_DEVICE_ID_CODEMERCS_IOW28)}, + {USB_DEVICE(USB_VENDOR_ID_CODEMERCS, USB_DEVICE_ID_CODEMERCS_IOW28L)}, + {USB_DEVICE(USB_VENDOR_ID_CODEMERCS, USB_DEVICE_ID_CODEMERCS_IOW100)}, {} /* Terminating entry */ }; MODULE_DEVICE_TABLE(usb, iowarrior_ids); @@ -357,6 +370,7 @@ static ssize_t iowarrior_write(struct file *file, } switch (dev->product_id) { case USB_DEVICE_ID_CODEMERCS_IOW24: + case USB_DEVICE_ID_CODEMERCS_IOW24SAG: case USB_DEVICE_ID_CODEMERCS_IOWPV1: case USB_DEVICE_ID_CODEMERCS_IOWPV2: case USB_DEVICE_ID_CODEMERCS_IOW40: @@ -371,6 +385,10 @@ static ssize_t iowarrior_write(struct file *file, goto exit; break; case USB_DEVICE_ID_CODEMERCS_IOW56: + case USB_DEVICE_ID_CODEMERCS_IOW56AM: + case USB_DEVICE_ID_CODEMERCS_IOW28: + case USB_DEVICE_ID_CODEMERCS_IOW28L: + case USB_DEVICE_ID_CODEMERCS_IOW100: /* The IOW56 uses asynchronous IO and more urbs */ if (atomic_read(&dev->write_busy) == MAX_WRITES_IN_FLIGHT) { /* Wait until we are below the limit for submitted urbs */ @@ -493,6 +511,7 @@ static long iowarrior_ioctl(struct file *file, unsigned int cmd, switch (cmd) { case IOW_WRITE: if (dev->product_id == USB_DEVICE_ID_CODEMERCS_IOW24 || + dev->product_id == USB_DEVICE_ID_CODEMERCS_IOW24SAG || dev->product_id == USB_DEVICE_ID_CODEMERCS_IOWPV1 || dev->product_id == USB_DEVICE_ID_CODEMERCS_IOWPV2 || dev->product_id == USB_DEVICE_ID_CODEMERCS_IOW40) { @@ -767,7 +786,11 @@ static int iowarrior_probe(struct usb_interface *interface, goto error; } - if (dev->product_id == USB_DEVICE_ID_CODEMERCS_IOW56) { + if ((dev->product_id == USB_DEVICE_ID_CODEMERCS_IOW56) || + (dev->product_id == USB_DEVICE_ID_CODEMERCS_IOW56AM) || + (dev->product_id == USB_DEVICE_ID_CODEMERCS_IOW28) || + (dev->product_id == USB_DEVICE_ID_CODEMERCS_IOW28L) || + (dev->product_id == USB_DEVICE_ID_CODEMERCS_IOW100)) { res = usb_find_last_int_out_endpoint(iface_desc, &dev->int_out_endpoint); if (res) { @@ -780,7 +803,11 @@ static int iowarrior_probe(struct usb_interface *interface, /* we have to check the report_size often, so remember it in the endianness suitable for our machine */ dev->report_size = usb_endpoint_maxp(dev->int_in_endpoint); if ((dev->interface->cur_altsetting->desc.bInterfaceNumber == 0) && - (dev->product_id == USB_DEVICE_ID_CODEMERCS_IOW56)) + ((dev->product_id == USB_DEVICE_ID_CODEMERCS_IOW56) || + (dev->product_id == USB_DEVICE_ID_CODEMERCS_IOW56AM) || + (dev->product_id == USB_DEVICE_ID_CODEMERCS_IOW28) || + (dev->product_id == USB_DEVICE_ID_CODEMERCS_IOW28L) || + (dev->product_id == USB_DEVICE_ID_CODEMERCS_IOW100))) /* IOWarrior56 has wMaxPacketSize different from report size */ dev->report_size = 7; diff --git a/drivers/usb/misc/usb251xb.c b/drivers/usb/misc/usb251xb.c index 10c9e7f6273e..29fe5771c21b 100644 --- a/drivers/usb/misc/usb251xb.c +++ b/drivers/usb/misc/usb251xb.c @@ -424,10 +424,6 @@ static int usb251xb_get_ofdata(struct usb251xb *hub, return err; } - hub->vdd = devm_regulator_get(dev, "vdd"); - if (IS_ERR(hub->vdd)) - return PTR_ERR(hub->vdd); - if (of_property_read_u16_array(np, "vendor-id", &hub->vendor_id, 1)) hub->vendor_id = USB251XB_DEF_VENDOR_ID; @@ -640,6 +636,13 @@ static int usb251xb_get_ofdata(struct usb251xb *hub, } #endif /* CONFIG_OF */ +static void usb251xb_regulator_disable_action(void *data) +{ + struct usb251xb *hub = data; + + regulator_disable(hub->vdd); +} + static int usb251xb_probe(struct usb251xb *hub) { struct device *dev = hub->dev; @@ -676,10 +679,19 @@ static int usb251xb_probe(struct usb251xb *hub) if (err) return err; + hub->vdd = devm_regulator_get(dev, "vdd"); + if (IS_ERR(hub->vdd)) + return PTR_ERR(hub->vdd); + err = regulator_enable(hub->vdd); if (err) return err; + err = devm_add_action_or_reset(dev, + usb251xb_regulator_disable_action, hub); + if (err) + return err; + err = usb251xb_connect(hub); if (err) { dev_err(dev, "Failed to connect hub (%d)\n", err); diff --git a/drivers/usb/phy/phy-tegra-usb.c b/drivers/usb/phy/phy-tegra-usb.c index 037e8eee737d..6153cc35aba0 100644 --- a/drivers/usb/phy/phy-tegra-usb.c +++ b/drivers/usb/phy/phy-tegra-usb.c @@ -969,6 +969,10 @@ static int utmi_phy_probe(struct tegra_usb_phy *tegra_phy, return -ENXIO; } + /* + * Note that UTMI pad registers are shared by all PHYs, therefore + * devm_platform_ioremap_resource() can't be used here. + */ tegra_phy->pad_regs = devm_ioremap(&pdev->dev, res->start, resource_size(res)); if (!tegra_phy->pad_regs) { @@ -1087,6 +1091,10 @@ static int tegra_usb_phy_probe(struct platform_device *pdev) return -ENXIO; } + /* + * Note that PHY and USB controller are using shared registers, + * therefore devm_platform_ioremap_resource() can't be used here. + */ tegra_phy->regs = devm_ioremap(&pdev->dev, res->start, resource_size(res)); if (!tegra_phy->regs) { diff --git a/drivers/usb/serial/ch341.c b/drivers/usb/serial/ch341.c index d3f420f3a083..c5ecdcd51ffc 100644 --- a/drivers/usb/serial/ch341.c +++ b/drivers/usb/serial/ch341.c @@ -205,6 +205,16 @@ static int ch341_get_divisor(speed_t speed) 16 * speed - 16 * CH341_CLKRATE / (clk_div * (div + 1))) div++; + /* + * Prefer lower base clock (fact = 0) if even divisor. + * + * Note that this makes the receiver more tolerant to errors. + */ + if (fact == 1 && div % 2 == 0) { + div /= 2; + fact = 0; + } + return (0x100 - div) << 8 | fact << 2 | ps; } diff --git a/drivers/usb/serial/ir-usb.c b/drivers/usb/serial/ir-usb.c index 79d0586e2b33..172261a908d8 100644 --- a/drivers/usb/serial/ir-usb.c +++ b/drivers/usb/serial/ir-usb.c @@ -448,7 +448,7 @@ static void ir_set_termios(struct tty_struct *tty, usb_sndbulkpipe(udev, port->bulk_out_endpointAddress), transfer_buffer, 1, &actual_length, 5000); if (ret || actual_length != 1) { - if (actual_length != 1) + if (!ret) ret = -EIO; dev_err(&port->dev, "failed to change line speed: %d\n", ret); } diff --git a/drivers/usb/serial/option.c b/drivers/usb/serial/option.c index 084cc2fff3ae..0b5dcf973d94 100644 --- a/drivers/usb/serial/option.c +++ b/drivers/usb/serial/option.c @@ -1183,6 +1183,8 @@ static const struct usb_device_id option_ids[] = { .driver_info = NCTRL(0) }, { USB_DEVICE_INTERFACE_CLASS(TELIT_VENDOR_ID, 0x110a, 0xff), /* Telit ME910G1 */ .driver_info = NCTRL(0) | RSVD(3) }, + { USB_DEVICE_INTERFACE_CLASS(TELIT_VENDOR_ID, 0x110b, 0xff), /* Telit ME910G1 (ECM) */ + .driver_info = NCTRL(0) }, { USB_DEVICE(TELIT_VENDOR_ID, TELIT_PRODUCT_LE910), .driver_info = NCTRL(0) | RSVD(1) | RSVD(2) }, { USB_DEVICE(TELIT_VENDOR_ID, TELIT_PRODUCT_LE910_USBCFG4), diff --git a/drivers/usb/serial/pl2303.c b/drivers/usb/serial/pl2303.c index aab737e1e7b6..c5a2995dfa2e 100644 --- a/drivers/usb/serial/pl2303.c +++ b/drivers/usb/serial/pl2303.c @@ -99,6 +99,7 @@ static const struct usb_device_id id_table[] = { { USB_DEVICE(SUPERIAL_VENDOR_ID, SUPERIAL_PRODUCT_ID) }, { USB_DEVICE(HP_VENDOR_ID, HP_LD220_PRODUCT_ID) }, { USB_DEVICE(HP_VENDOR_ID, HP_LD220TA_PRODUCT_ID) }, + { USB_DEVICE(HP_VENDOR_ID, HP_LD381_PRODUCT_ID) }, { USB_DEVICE(HP_VENDOR_ID, HP_LD960_PRODUCT_ID) }, { USB_DEVICE(HP_VENDOR_ID, HP_LD960TA_PRODUCT_ID) }, { USB_DEVICE(HP_VENDOR_ID, HP_LCM220_PRODUCT_ID) }, diff --git a/drivers/usb/serial/pl2303.h b/drivers/usb/serial/pl2303.h index a019ea7e6e0e..52db5519aaf0 100644 --- a/drivers/usb/serial/pl2303.h +++ b/drivers/usb/serial/pl2303.h @@ -130,6 +130,7 @@ #define HP_LM920_PRODUCT_ID 0x026b #define HP_TD620_PRODUCT_ID 0x0956 #define HP_LD960_PRODUCT_ID 0x0b39 +#define HP_LD381_PRODUCT_ID 0x0f7f #define HP_LCM220_PRODUCT_ID 0x3139 #define HP_LCM960_PRODUCT_ID 0x3239 #define HP_LD220_PRODUCT_ID 0x3524 diff --git a/drivers/usb/storage/uas.c b/drivers/usb/storage/uas.c index 95bba3ba6ac6..3670fda02c34 100644 --- a/drivers/usb/storage/uas.c +++ b/drivers/usb/storage/uas.c @@ -45,6 +45,7 @@ struct uas_dev_info { struct scsi_cmnd *cmnd[MAX_CMNDS]; spinlock_t lock; struct work_struct work; + struct work_struct scan_work; /* for async scanning */ }; enum { @@ -114,6 +115,17 @@ out: spin_unlock_irqrestore(&devinfo->lock, flags); } +static void uas_scan_work(struct work_struct *work) +{ + struct uas_dev_info *devinfo = + container_of(work, struct uas_dev_info, scan_work); + struct Scsi_Host *shost = usb_get_intfdata(devinfo->intf); + + dev_dbg(&devinfo->intf->dev, "starting scan\n"); + scsi_scan_host(shost); + dev_dbg(&devinfo->intf->dev, "scan complete\n"); +} + static void uas_add_work(struct uas_cmd_info *cmdinfo) { struct scsi_pointer *scp = (void *)cmdinfo; @@ -982,6 +994,7 @@ static int uas_probe(struct usb_interface *intf, const struct usb_device_id *id) init_usb_anchor(&devinfo->data_urbs); spin_lock_init(&devinfo->lock); INIT_WORK(&devinfo->work, uas_do_work); + INIT_WORK(&devinfo->scan_work, uas_scan_work); result = uas_configure_endpoints(devinfo); if (result) @@ -998,7 +1011,9 @@ static int uas_probe(struct usb_interface *intf, const struct usb_device_id *id) if (result) goto free_streams; - scsi_scan_host(shost); + /* Submit the delayed_work for SCSI-device scanning */ + schedule_work(&devinfo->scan_work); + return result; free_streams: @@ -1166,6 +1181,12 @@ static void uas_disconnect(struct usb_interface *intf) usb_kill_anchored_urbs(&devinfo->data_urbs); uas_zap_pending(devinfo, DID_NO_CONNECT); + /* + * Prevent SCSI scanning (if it hasn't started yet) + * or wait for the SCSI-scanning routine to stop. + */ + cancel_work_sync(&devinfo->scan_work); + scsi_remove_host(shost); uas_free_streams(devinfo); scsi_host_put(shost); diff --git a/drivers/usb/storage/unusual_devs.h b/drivers/usb/storage/unusual_devs.h index 1cd9b6305b06..1880f3e13f57 100644 --- a/drivers/usb/storage/unusual_devs.h +++ b/drivers/usb/storage/unusual_devs.h @@ -1258,6 +1258,12 @@ UNUSUAL_DEV( 0x090a, 0x1200, 0x0000, 0x9999, USB_SC_RBC, USB_PR_BULK, NULL, 0 ), +UNUSUAL_DEV(0x090c, 0x1000, 0x1100, 0x1100, + "Samsung", + "Flash Drive FIT", + USB_SC_DEVICE, USB_PR_DEVICE, NULL, + US_FL_MAX_SECTORS_64), + /* aeb */ UNUSUAL_DEV( 0x090c, 0x1132, 0x0000, 0xffff, "Feiya", diff --git a/drivers/usb/typec/ucsi/displayport.c b/drivers/usb/typec/ucsi/displayport.c index 0f1273ae086c..048381c058a5 100644 --- a/drivers/usb/typec/ucsi/displayport.c +++ b/drivers/usb/typec/ucsi/displayport.c @@ -271,6 +271,9 @@ void ucsi_displayport_remove_partner(struct typec_altmode *alt) return; dp = typec_altmode_get_drvdata(alt); + if (!dp) + return; + dp->data.conf = 0; dp->data.status = 0; dp->initialized = false; @@ -285,6 +288,8 @@ struct typec_altmode *ucsi_register_displayport(struct ucsi_connector *con, struct typec_altmode *alt; struct ucsi_dp *dp; + mutex_lock(&con->lock); + /* We can't rely on the firmware with the capabilities. */ desc->vdo |= DP_CAP_DP_SIGNALING | DP_CAP_RECEPTACLE; @@ -293,12 +298,15 @@ struct typec_altmode *ucsi_register_displayport(struct ucsi_connector *con, desc->vdo |= all_assignments << 16; alt = typec_port_register_altmode(con->port, desc); - if (IS_ERR(alt)) + if (IS_ERR(alt)) { + mutex_unlock(&con->lock); return alt; + } dp = devm_kzalloc(&alt->dev, sizeof(*dp), GFP_KERNEL); if (!dp) { typec_unregister_altmode(alt); + mutex_unlock(&con->lock); return ERR_PTR(-ENOMEM); } @@ -311,5 +319,7 @@ struct typec_altmode *ucsi_register_displayport(struct ucsi_connector *con, alt->ops = &ucsi_displayport_ops; typec_altmode_set_drvdata(alt, dp); + mutex_unlock(&con->lock); + return alt; } diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c index e158159671fa..18e205eeb9af 100644 --- a/drivers/vhost/net.c +++ b/drivers/vhost/net.c @@ -1414,10 +1414,6 @@ static int vhost_net_release(struct inode *inode, struct file *f) static struct socket *get_raw_socket(int fd) { - struct { - struct sockaddr_ll sa; - char buf[MAX_ADDR_LEN]; - } uaddr; int r; struct socket *sock = sockfd_lookup(fd, &r); @@ -1430,11 +1426,7 @@ static struct socket *get_raw_socket(int fd) goto err; } - r = sock->ops->getname(sock, (struct sockaddr *)&uaddr.sa, 0); - if (r < 0) - goto err; - - if (uaddr.sa.sll_family != AF_PACKET) { + if (sock->sk->sk_family != AF_PACKET) { r = -EPFNOSUPPORT; goto err; } diff --git a/drivers/video/backlight/Kconfig b/drivers/video/backlight/Kconfig index 95e2000c1491..7d22d7377606 100644 --- a/drivers/video/backlight/Kconfig +++ b/drivers/video/backlight/Kconfig @@ -456,6 +456,13 @@ config BACKLIGHT_RAVE_SP help Support for backlight control on RAVE SP device. +config BACKLIGHT_LED + tristate "Generic LED based Backlight Driver" + depends on LEDS_CLASS && OF + help + If you have a LCD backlight adjustable by LED class driver, say Y + to enable this driver. + endif # BACKLIGHT_CLASS_DEVICE endmenu diff --git a/drivers/video/backlight/Makefile b/drivers/video/backlight/Makefile index 6f8777037c37..0c1a1524627a 100644 --- a/drivers/video/backlight/Makefile +++ b/drivers/video/backlight/Makefile @@ -57,3 +57,4 @@ obj-$(CONFIG_BACKLIGHT_TPS65217) += tps65217_bl.o obj-$(CONFIG_BACKLIGHT_WM831X) += wm831x_bl.o obj-$(CONFIG_BACKLIGHT_ARCXCNN) += arcxcnn_bl.o obj-$(CONFIG_BACKLIGHT_RAVE_SP) += rave-sp-backlight.o +obj-$(CONFIG_BACKLIGHT_LED) += led_bl.o diff --git a/drivers/video/backlight/led_bl.c b/drivers/video/backlight/led_bl.c new file mode 100644 index 000000000000..3f66549997c8 --- /dev/null +++ b/drivers/video/backlight/led_bl.c @@ -0,0 +1,260 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (C) 2015-2019 Texas Instruments Incorporated - http://www.ti.com/ + * Author: Tomi Valkeinen <tomi.valkeinen@ti.com> + * + * Based on pwm_bl.c + */ + +#include <linux/backlight.h> +#include <linux/leds.h> +#include <linux/module.h> +#include <linux/platform_device.h> + +struct led_bl_data { + struct device *dev; + struct backlight_device *bl_dev; + struct led_classdev **leds; + bool enabled; + int nb_leds; + unsigned int *levels; + unsigned int default_brightness; + unsigned int max_brightness; +}; + +static void led_bl_set_brightness(struct led_bl_data *priv, int level) +{ + int i; + int bkl_brightness; + + if (priv->levels) + bkl_brightness = priv->levels[level]; + else + bkl_brightness = level; + + for (i = 0; i < priv->nb_leds; i++) + led_set_brightness(priv->leds[i], bkl_brightness); + + priv->enabled = true; +} + +static void led_bl_power_off(struct led_bl_data *priv) +{ + int i; + + if (!priv->enabled) + return; + + for (i = 0; i < priv->nb_leds; i++) + led_set_brightness(priv->leds[i], LED_OFF); + + priv->enabled = false; +} + +static int led_bl_update_status(struct backlight_device *bl) +{ + struct led_bl_data *priv = bl_get_data(bl); + int brightness = bl->props.brightness; + + if (bl->props.power != FB_BLANK_UNBLANK || + bl->props.fb_blank != FB_BLANK_UNBLANK || + bl->props.state & BL_CORE_FBBLANK) + brightness = 0; + + if (brightness > 0) + led_bl_set_brightness(priv, brightness); + else + led_bl_power_off(priv); + + return 0; +} + +static const struct backlight_ops led_bl_ops = { + .update_status = led_bl_update_status, +}; + +static int led_bl_get_leds(struct device *dev, + struct led_bl_data *priv) +{ + int i, nb_leds, ret; + struct device_node *node = dev->of_node; + struct led_classdev **leds; + unsigned int max_brightness; + unsigned int default_brightness; + + ret = of_count_phandle_with_args(node, "leds", NULL); + if (ret < 0) { + dev_err(dev, "Unable to get led count\n"); + return -EINVAL; + } + + nb_leds = ret; + if (nb_leds < 1) { + dev_err(dev, "At least one LED must be specified!\n"); + return -EINVAL; + } + + leds = devm_kzalloc(dev, sizeof(struct led_classdev *) * nb_leds, + GFP_KERNEL); + if (!leds) + return -ENOMEM; + + for (i = 0; i < nb_leds; i++) { + leds[i] = devm_of_led_get(dev, i); + if (IS_ERR(leds[i])) + return PTR_ERR(leds[i]); + } + + /* check that the LEDs all have the same brightness range */ + max_brightness = leds[0]->max_brightness; + for (i = 1; i < nb_leds; i++) { + if (max_brightness != leds[i]->max_brightness) { + dev_err(dev, "LEDs must have identical ranges\n"); + return -EINVAL; + } + } + + /* get the default brightness from the first LED from the list */ + default_brightness = leds[0]->brightness; + + priv->nb_leds = nb_leds; + priv->leds = leds; + priv->max_brightness = max_brightness; + priv->default_brightness = default_brightness; + + return 0; +} + +static int led_bl_parse_levels(struct device *dev, + struct led_bl_data *priv) +{ + struct device_node *node = dev->of_node; + int num_levels; + u32 value; + int ret; + + if (!node) + return -ENODEV; + + num_levels = of_property_count_u32_elems(node, "brightness-levels"); + if (num_levels > 1) { + int i; + unsigned int db; + u32 *levels = NULL; + + levels = devm_kzalloc(dev, sizeof(u32) * num_levels, + GFP_KERNEL); + if (!levels) + return -ENOMEM; + + ret = of_property_read_u32_array(node, "brightness-levels", + levels, + num_levels); + if (ret < 0) + return ret; + + /* + * Try to map actual LED brightness to backlight brightness + * level + */ + db = priv->default_brightness; + for (i = 0 ; i < num_levels; i++) { + if ((i && db > levels[i-1]) && db <= levels[i]) + break; + } + priv->default_brightness = i; + priv->max_brightness = num_levels - 1; + priv->levels = levels; + } else if (num_levels >= 0) + dev_warn(dev, "Not enough levels defined\n"); + + ret = of_property_read_u32(node, "default-brightness-level", &value); + if (!ret && value <= priv->max_brightness) + priv->default_brightness = value; + else if (!ret && value > priv->max_brightness) + dev_warn(dev, "Invalid default brightness. Ignoring it\n"); + + return 0; +} + +static int led_bl_probe(struct platform_device *pdev) +{ + struct backlight_properties props; + struct led_bl_data *priv; + int ret, i; + + priv = devm_kzalloc(&pdev->dev, sizeof(*priv), GFP_KERNEL); + if (!priv) + return -ENOMEM; + + platform_set_drvdata(pdev, priv); + + priv->dev = &pdev->dev; + + ret = led_bl_get_leds(&pdev->dev, priv); + if (ret) + return ret; + + ret = led_bl_parse_levels(&pdev->dev, priv); + if (ret < 0) { + dev_err(&pdev->dev, "Failed to parse DT data\n"); + return ret; + } + + memset(&props, 0, sizeof(struct backlight_properties)); + props.type = BACKLIGHT_RAW; + props.max_brightness = priv->max_brightness; + props.brightness = priv->default_brightness; + props.power = (priv->default_brightness > 0) ? FB_BLANK_POWERDOWN : + FB_BLANK_UNBLANK; + priv->bl_dev = backlight_device_register(dev_name(&pdev->dev), + &pdev->dev, priv, &led_bl_ops, &props); + if (IS_ERR(priv->bl_dev)) { + dev_err(&pdev->dev, "Failed to register backlight\n"); + return PTR_ERR(priv->bl_dev); + } + + for (i = 0; i < priv->nb_leds; i++) + led_sysfs_disable(priv->leds[i]); + + backlight_update_status(priv->bl_dev); + + return 0; +} + +static int led_bl_remove(struct platform_device *pdev) +{ + struct led_bl_data *priv = platform_get_drvdata(pdev); + struct backlight_device *bl = priv->bl_dev; + int i; + + backlight_device_unregister(bl); + + led_bl_power_off(priv); + for (i = 0; i < priv->nb_leds; i++) + led_sysfs_enable(priv->leds[i]); + + return 0; +} + +static const struct of_device_id led_bl_of_match[] = { + { .compatible = "led-backlight" }, + { } +}; + +MODULE_DEVICE_TABLE(of, led_bl_of_match); + +static struct platform_driver led_bl_driver = { + .driver = { + .name = "led-backlight", + .of_match_table = of_match_ptr(led_bl_of_match), + }, + .probe = led_bl_probe, + .remove = led_bl_remove, +}; + +module_platform_driver(led_bl_driver); + +MODULE_DESCRIPTION("LED based Backlight Driver"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS("platform:led-backlight"); diff --git a/drivers/video/console/vgacon.c b/drivers/video/console/vgacon.c index de7b8382aba9..998b0de1812f 100644 --- a/drivers/video/console/vgacon.c +++ b/drivers/video/console/vgacon.c @@ -1316,6 +1316,9 @@ static int vgacon_font_get(struct vc_data *c, struct console_font *font) static int vgacon_resize(struct vc_data *c, unsigned int width, unsigned int height, unsigned int user) { + if ((width << 1) * height > vga_vram_size) + return -EINVAL; + if (width % 2 || width > screen_info.orig_video_cols || height > (screen_info.orig_video_lines * vga_default_font_height)/ c->vc_font.height) diff --git a/drivers/virtio/virtio_balloon.c b/drivers/virtio/virtio_balloon.c index 7bfe365d9372..341458fd95ca 100644 --- a/drivers/virtio/virtio_balloon.c +++ b/drivers/virtio/virtio_balloon.c @@ -959,8 +959,8 @@ out_iput: iput(vb->vb_dev_info.inode); out_kern_unmount: kern_unmount(balloon_mnt); -#endif out_del_vqs: +#endif vdev->config->del_vqs(vdev); out_free_vb: kfree(vb); diff --git a/drivers/virtio/virtio_ring.c b/drivers/virtio/virtio_ring.c index 867c7ebd3f10..58b96baa8d48 100644 --- a/drivers/virtio/virtio_ring.c +++ b/drivers/virtio/virtio_ring.c @@ -2203,10 +2203,10 @@ void vring_del_virtqueue(struct virtqueue *_vq) vq->split.queue_size_in_bytes, vq->split.vring.desc, vq->split.queue_dma_addr); - - kfree(vq->split.desc_state); } } + if (!vq->packed_ring) + kfree(vq->split.desc_state); list_del(&_vq->list); kfree(vq); } diff --git a/drivers/watchdog/Kconfig b/drivers/watchdog/Kconfig index cec868f8db3f..9ea2b43d4b01 100644 --- a/drivers/watchdog/Kconfig +++ b/drivers/watchdog/Kconfig @@ -207,6 +207,7 @@ config DA9063_WATCHDOG config DA9062_WATCHDOG tristate "Dialog DA9062/61 Watchdog" depends on MFD_DA9062 || COMPILE_TEST + depends on I2C select WATCHDOG_CORE help Support for the watchdog in the DA9062 and DA9061 PMICs. @@ -841,6 +842,7 @@ config MEDIATEK_WATCHDOG tristate "Mediatek SoCs watchdog support" depends on ARCH_MEDIATEK || COMPILE_TEST select WATCHDOG_CORE + select RESET_CONTROLLER help Say Y here to include support for the watchdog timer in Mediatek SoCs. diff --git a/drivers/watchdog/da9062_wdt.c b/drivers/watchdog/da9062_wdt.c index 47eefe072b40..0ad15d55071c 100644 --- a/drivers/watchdog/da9062_wdt.c +++ b/drivers/watchdog/da9062_wdt.c @@ -16,6 +16,7 @@ #include <linux/jiffies.h> #include <linux/mfd/da9062/registers.h> #include <linux/mfd/da9062/core.h> +#include <linux/property.h> #include <linux/regmap.h> #include <linux/of.h> @@ -31,6 +32,7 @@ static const unsigned int wdt_timeout[] = { 0, 2, 4, 8, 16, 32, 65, 131 }; struct da9062_watchdog { struct da9062 *hw; struct watchdog_device wdtdev; + bool use_sw_pm; }; static unsigned int da9062_wdt_timeout_to_sel(unsigned int secs) @@ -95,13 +97,6 @@ static int da9062_wdt_stop(struct watchdog_device *wdd) struct da9062_watchdog *wdt = watchdog_get_drvdata(wdd); int ret; - ret = da9062_reset_watchdog_timer(wdt); - if (ret) { - dev_err(wdt->hw->dev, "Failed to ping the watchdog (err = %d)\n", - ret); - return ret; - } - ret = regmap_update_bits(wdt->hw->regmap, DA9062AA_CONTROL_D, DA9062AA_TWDSCALE_MASK, @@ -200,6 +195,8 @@ static int da9062_wdt_probe(struct platform_device *pdev) if (!wdt) return -ENOMEM; + wdt->use_sw_pm = device_property_present(dev, "dlg,use-sw-pm"); + wdt->hw = chip; wdt->wdtdev.info = &da9062_watchdog_info; @@ -226,6 +223,10 @@ static int da9062_wdt_probe(struct platform_device *pdev) static int __maybe_unused da9062_wdt_suspend(struct device *dev) { struct watchdog_device *wdd = dev_get_drvdata(dev); + struct da9062_watchdog *wdt = watchdog_get_drvdata(wdd); + + if (!wdt->use_sw_pm) + return 0; if (watchdog_active(wdd)) return da9062_wdt_stop(wdd); @@ -236,6 +237,10 @@ static int __maybe_unused da9062_wdt_suspend(struct device *dev) static int __maybe_unused da9062_wdt_resume(struct device *dev) { struct watchdog_device *wdd = dev_get_drvdata(dev); + struct da9062_watchdog *wdt = watchdog_get_drvdata(wdd); + + if (!wdt->use_sw_pm) + return 0; if (watchdog_active(wdd)) return da9062_wdt_start(wdd); diff --git a/drivers/watchdog/iTCO_vendor.h b/drivers/watchdog/iTCO_vendor.h index 0f7373ba10d5..69e92e692ae0 100644 --- a/drivers/watchdog/iTCO_vendor.h +++ b/drivers/watchdog/iTCO_vendor.h @@ -1,10 +1,12 @@ /* SPDX-License-Identifier: GPL-2.0 */ /* iTCO Vendor Specific Support hooks */ #ifdef CONFIG_ITCO_VENDOR_SUPPORT +extern int iTCO_vendorsupport; extern void iTCO_vendor_pre_start(struct resource *, unsigned int); extern void iTCO_vendor_pre_stop(struct resource *); extern int iTCO_vendor_check_noreboot_on(void); #else +#define iTCO_vendorsupport 0 #define iTCO_vendor_pre_start(acpibase, heartbeat) {} #define iTCO_vendor_pre_stop(acpibase) {} #define iTCO_vendor_check_noreboot_on() 1 diff --git a/drivers/watchdog/iTCO_vendor_support.c b/drivers/watchdog/iTCO_vendor_support.c index 4f1b96f59349..cf0eaa04b064 100644 --- a/drivers/watchdog/iTCO_vendor_support.c +++ b/drivers/watchdog/iTCO_vendor_support.c @@ -39,8 +39,10 @@ /* Broken BIOS */ #define BROKEN_BIOS 911 -static int vendorsupport; -module_param(vendorsupport, int, 0); +int iTCO_vendorsupport; +EXPORT_SYMBOL(iTCO_vendorsupport); + +module_param_named(vendorsupport, iTCO_vendorsupport, int, 0); MODULE_PARM_DESC(vendorsupport, "iTCO vendor specific support mode, default=" "0 (none), 1=SuperMicro Pent3, 911=Broken SMI BIOS"); @@ -152,7 +154,7 @@ static void broken_bios_stop(struct resource *smires) void iTCO_vendor_pre_start(struct resource *smires, unsigned int heartbeat) { - switch (vendorsupport) { + switch (iTCO_vendorsupport) { case SUPERMICRO_OLD_BOARD: supermicro_old_pre_start(smires); break; @@ -165,7 +167,7 @@ EXPORT_SYMBOL(iTCO_vendor_pre_start); void iTCO_vendor_pre_stop(struct resource *smires) { - switch (vendorsupport) { + switch (iTCO_vendorsupport) { case SUPERMICRO_OLD_BOARD: supermicro_old_pre_stop(smires); break; @@ -178,7 +180,7 @@ EXPORT_SYMBOL(iTCO_vendor_pre_stop); int iTCO_vendor_check_noreboot_on(void) { - switch (vendorsupport) { + switch (iTCO_vendorsupport) { case SUPERMICRO_OLD_BOARD: return 0; default: @@ -189,13 +191,13 @@ EXPORT_SYMBOL(iTCO_vendor_check_noreboot_on); static int __init iTCO_vendor_init_module(void) { - if (vendorsupport == SUPERMICRO_NEW_BOARD) { + if (iTCO_vendorsupport == SUPERMICRO_NEW_BOARD) { pr_warn("Option vendorsupport=%d is no longer supported, " "please use the w83627hf_wdt driver instead\n", SUPERMICRO_NEW_BOARD); return -EINVAL; } - pr_info("vendor-support=%d\n", vendorsupport); + pr_info("vendor-support=%d\n", iTCO_vendorsupport); return 0; } diff --git a/drivers/watchdog/iTCO_wdt.c b/drivers/watchdog/iTCO_wdt.c index 156360e37714..e707c4797f76 100644 --- a/drivers/watchdog/iTCO_wdt.c +++ b/drivers/watchdog/iTCO_wdt.c @@ -459,13 +459,25 @@ static int iTCO_wdt_probe(struct platform_device *pdev) if (!p->tco_res) return -ENODEV; - p->smi_res = platform_get_resource(pdev, IORESOURCE_IO, ICH_RES_IO_SMI); - if (!p->smi_res) - return -ENODEV; - p->iTCO_version = pdata->version; p->pci_dev = to_pci_dev(dev->parent); + p->smi_res = platform_get_resource(pdev, IORESOURCE_IO, ICH_RES_IO_SMI); + if (p->smi_res) { + /* The TCO logic uses the TCO_EN bit in the SMI_EN register */ + if (!devm_request_region(dev, p->smi_res->start, + resource_size(p->smi_res), + pdev->name)) { + pr_err("I/O address 0x%04llx already in use, device disabled\n", + (u64)SMI_EN(p)); + return -EBUSY; + } + } else if (iTCO_vendorsupport || + turn_SMI_watchdog_clear_off >= p->iTCO_version) { + pr_err("SMI I/O resource is missing\n"); + return -ENODEV; + } + iTCO_wdt_no_reboot_bit_setup(p, pdata); /* @@ -492,14 +504,6 @@ static int iTCO_wdt_probe(struct platform_device *pdev) /* Set the NO_REBOOT bit to prevent later reboots, just for sure */ p->update_no_reboot_bit(p->no_reboot_priv, true); - /* The TCO logic uses the TCO_EN bit in the SMI_EN register */ - if (!devm_request_region(dev, p->smi_res->start, - resource_size(p->smi_res), - pdev->name)) { - pr_err("I/O address 0x%04llx already in use, device disabled\n", - (u64)SMI_EN(p)); - return -EBUSY; - } if (turn_SMI_watchdog_clear_off >= p->iTCO_version) { /* * Bit 13: TCO_EN -> 0 diff --git a/drivers/watchdog/wdat_wdt.c b/drivers/watchdog/wdat_wdt.c index b069349b52f5..3065dd670a18 100644 --- a/drivers/watchdog/wdat_wdt.c +++ b/drivers/watchdog/wdat_wdt.c @@ -54,6 +54,13 @@ module_param(nowayout, bool, 0); MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started (default=" __MODULE_STRING(WATCHDOG_NOWAYOUT) ")"); +#define WDAT_DEFAULT_TIMEOUT 30 + +static int timeout = WDAT_DEFAULT_TIMEOUT; +module_param(timeout, int, 0); +MODULE_PARM_DESC(timeout, "Watchdog timeout in seconds (default=" + __MODULE_STRING(WDAT_DEFAULT_TIMEOUT) ")"); + static int wdat_wdt_read(struct wdat_wdt *wdat, const struct wdat_instruction *instr, u32 *value) { @@ -389,7 +396,7 @@ static int wdat_wdt_probe(struct platform_device *pdev) memset(&r, 0, sizeof(r)); r.start = gas->address; - r.end = r.start + gas->access_width - 1; + r.end = r.start + ACPI_ACCESS_BYTE_WIDTH(gas->access_width) - 1; if (gas->space_id == ACPI_ADR_SPACE_SYSTEM_MEMORY) { r.flags = IORESOURCE_MEM; } else if (gas->space_id == ACPI_ADR_SPACE_SYSTEM_IO) { @@ -438,6 +445,22 @@ static int wdat_wdt_probe(struct platform_device *pdev) platform_set_drvdata(pdev, wdat); + /* + * Set initial timeout so that userspace has time to configure the + * watchdog properly after it has opened the device. In some cases + * the BIOS default is too short and causes immediate reboot. + */ + if (timeout * 1000 < wdat->wdd.min_hw_heartbeat_ms || + timeout * 1000 > wdat->wdd.max_hw_heartbeat_ms) { + dev_warn(dev, "Invalid timeout %d given, using %d\n", + timeout, WDAT_DEFAULT_TIMEOUT); + timeout = WDAT_DEFAULT_TIMEOUT; + } + + ret = wdat_wdt_set_timeout(&wdat->wdd, timeout); + if (ret) + return ret; + watchdog_set_nowayout(&wdat->wdd, nowayout); return devm_watchdog_register_device(dev, &wdat->wdd); } diff --git a/drivers/xen/preempt.c b/drivers/xen/preempt.c index 70650b248de5..17240c5325a3 100644 --- a/drivers/xen/preempt.c +++ b/drivers/xen/preempt.c @@ -33,7 +33,9 @@ asmlinkage __visible void xen_maybe_preempt_hcall(void) * cpu. */ __this_cpu_write(xen_in_preemptible_hcall, false); - _cond_resched(); + local_irq_enable(); + cond_resched(); + local_irq_disable(); __this_cpu_write(xen_in_preemptible_hcall, true); } } diff --git a/drivers/xen/xen-pciback/pciback.h b/drivers/xen/xen-pciback/pciback.h index ce1077e32466..7c95516a860f 100644 --- a/drivers/xen/xen-pciback/pciback.h +++ b/drivers/xen/xen-pciback/pciback.h @@ -52,7 +52,7 @@ struct xen_pcibk_dev_data { unsigned int ack_intr:1; /* .. and ACK-ing */ unsigned long handled; unsigned int irq; /* Saved in case device transitions to MSI/MSI-X */ - char irq_name[0]; /* xen-pcibk[000:04:00.0] */ + char irq_name[]; /* xen-pcibk[000:04:00.0] */ }; /* Used by XenBus and xen_pcibk_ops.c */ diff --git a/drivers/xen/xenbus/xenbus_comms.c b/drivers/xen/xenbus/xenbus_comms.c index d239fc3c5e3d..eb5151fc8efa 100644 --- a/drivers/xen/xenbus/xenbus_comms.c +++ b/drivers/xen/xenbus/xenbus_comms.c @@ -313,6 +313,8 @@ static int process_msg(void) req->msg.type = state.msg.type; req->msg.len = state.msg.len; req->body = state.body; + /* write body, then update state */ + virt_wmb(); req->state = xb_req_state_got_reply; req->cb(req); } else @@ -395,6 +397,8 @@ static int process_writes(void) if (state.req->state == xb_req_state_aborted) kfree(state.req); else { + /* write err, then update state */ + virt_wmb(); state.req->state = xb_req_state_got_reply; wake_up(&state.req->wq); } diff --git a/drivers/xen/xenbus/xenbus_probe.c b/drivers/xen/xenbus/xenbus_probe.c index 66975da4f3b6..8c4d05b687b7 100644 --- a/drivers/xen/xenbus/xenbus_probe.c +++ b/drivers/xen/xenbus/xenbus_probe.c @@ -239,9 +239,9 @@ int xenbus_dev_probe(struct device *_dev) goto fail; } - spin_lock(&dev->reclaim_lock); + down(&dev->reclaim_sem); err = drv->probe(dev, id); - spin_unlock(&dev->reclaim_lock); + up(&dev->reclaim_sem); if (err) goto fail_put; @@ -271,9 +271,9 @@ int xenbus_dev_remove(struct device *_dev) free_otherend_watch(dev); if (drv->remove) { - spin_lock(&dev->reclaim_lock); + down(&dev->reclaim_sem); drv->remove(dev); - spin_unlock(&dev->reclaim_lock); + up(&dev->reclaim_sem); } module_put(drv->driver.owner); @@ -473,7 +473,7 @@ int xenbus_probe_node(struct xen_bus_type *bus, goto fail; dev_set_name(&xendev->dev, "%s", devname); - spin_lock_init(&xendev->reclaim_lock); + sema_init(&xendev->reclaim_sem, 1); /* Register with generic device framework. */ err = device_register(&xendev->dev); diff --git a/drivers/xen/xenbus/xenbus_probe_backend.c b/drivers/xen/xenbus/xenbus_probe_backend.c index 791f6fe01e91..9b2fbe69bccc 100644 --- a/drivers/xen/xenbus/xenbus_probe_backend.c +++ b/drivers/xen/xenbus/xenbus_probe_backend.c @@ -45,6 +45,7 @@ #include <linux/mm.h> #include <linux/notifier.h> #include <linux/export.h> +#include <linux/semaphore.h> #include <asm/page.h> #include <asm/pgtable.h> @@ -257,10 +258,10 @@ static int backend_reclaim_memory(struct device *dev, void *data) drv = to_xenbus_driver(dev->driver); if (drv && drv->reclaim_memory) { xdev = to_xenbus_device(dev); - if (!spin_trylock(&xdev->reclaim_lock)) + if (down_trylock(&xdev->reclaim_sem)) return 0; drv->reclaim_memory(xdev); - spin_unlock(&xdev->reclaim_lock); + up(&xdev->reclaim_sem); } return 0; } diff --git a/drivers/xen/xenbus/xenbus_xs.c b/drivers/xen/xenbus/xenbus_xs.c index ddc18da61834..3a06eb699f33 100644 --- a/drivers/xen/xenbus/xenbus_xs.c +++ b/drivers/xen/xenbus/xenbus_xs.c @@ -191,8 +191,11 @@ static bool xenbus_ok(void) static bool test_reply(struct xb_req_data *req) { - if (req->state == xb_req_state_got_reply || !xenbus_ok()) + if (req->state == xb_req_state_got_reply || !xenbus_ok()) { + /* read req->state before all other fields */ + virt_rmb(); return true; + } /* Make sure to reread req->state each time. */ barrier(); @@ -202,7 +205,7 @@ static bool test_reply(struct xb_req_data *req) static void *read_reply(struct xb_req_data *req) { - while (req->state != xb_req_state_got_reply) { + do { wait_event(req->wq, test_reply(req)); if (!xenbus_ok()) @@ -216,7 +219,7 @@ static void *read_reply(struct xb_req_data *req) if (req->err) return ERR_PTR(req->err); - } + } while (req->state != xb_req_state_got_reply); return req->body; } |