diff options
| author | Linus Torvalds <torvalds@linux-foundation.org> | 2012-07-26 23:18:18 +0200 |
|---|---|---|
| committer | Linus Torvalds <torvalds@linux-foundation.org> | 2012-07-26 23:18:18 +0200 |
| commit | bd22dc17e49973d3d4925970260e9e37f7580a9f (patch) | |
| tree | 581a7c7527f628aa91eb2e0680b765a9673bc974 /drivers/gpu/drm/i915 | |
| parent | Merge tag 'dlm-3.6' of git://git.kernel.org/pub/scm/linux/kernel/git/teigland... (diff) | |
| parent | Merge branch 'drm-intel-fixes' of git://people.freedesktop.org/~danvet/drm-in... (diff) | |
| download | linux-bd22dc17e49973d3d4925970260e9e37f7580a9f.tar.xz linux-bd22dc17e49973d3d4925970260e9e37f7580a9f.zip | |
Merge branch 'drm-next' of git://people.freedesktop.org/~airlied/linux
Pull drm updates from Dave Airlie:
"One of the smaller drm -next pulls in ages!
Ben (nouveau) has a rewrite in progress but we decided to leave it
stew for another cycle, so just some fixes from him.
- radeon: lots of documentation work, fixes, more ring and locking
changes, pcie gen2, more dp fixes.
- i915: haswell features, gpu reset fixes, /dev/agpgart removal on
machines that we never used it on, more VGA/HDP fix., more DP fixes
- drm core: cleanups from Daniel, sis 64-bit fixes, range allocator
colouring.
but yeah fairly quiet merge this time, probably because I missed half
of it!"
Trivial add-add conflict in include/linux/pci_regs.h
* 'drm-next' of git://people.freedesktop.org/~airlied/linux: (255 commits)
drm/nouveau: init vblank requests list
drm/nv50: extend vblank semaphore to generic dmaobj + offset pair
drm/nouveau: mark most of our ioctls as deprecated, move to compat layer
drm/nouveau: move current gpuobj code out of nouveau_object.c
drm/nouveau/gem: fix object reference leak in a failure path
drm/nv50: rename INVALID_QUERY_OR_TEXTURE error to INVALID_OPERATION
drm/nv84: decode PCRYPT errors
drm/nouveau: dcb table quirk for fdo#50830
nouveau: Fix alignment requirements on src and dst addresses
drm/i915: unbreak lastclose for failed driver init
drm/i915: Set the context before setting up regs for the context.
drm/i915: constify mode in crtc_mode_fixup
drm/i915/lvds: ditch ->prepare special case
drm/i915: dereferencing an error pointer
drm/i915: fix invalid reference handling of the default ctx obj
drm/i915: Add -EIO to the list of known errors for __wait_seqno
drm/i915: Flush the context object from the CPU caches upon switching
drm/radeon: fix dpms on/off on trinity/aruba v2
drm/radeon: on hotplug force link training to happen (v2)
drm/radeon: fix hotplug of DP to DVI|HDMI passive adapters (v2)
...
Diffstat (limited to 'drivers/gpu/drm/i915')
36 files changed, 3064 insertions, 942 deletions
diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile index 2e9268da58d8..b0bacdba6d7e 100644 --- a/drivers/gpu/drm/i915/Makefile +++ b/drivers/gpu/drm/i915/Makefile @@ -7,6 +7,7 @@ i915-y := i915_drv.o i915_dma.o i915_irq.o \ i915_debugfs.o \ i915_suspend.o \ i915_gem.o \ + i915_gem_context.o \ i915_gem_debug.o \ i915_gem_evict.o \ i915_gem_execbuffer.o \ diff --git a/drivers/gpu/drm/i915/dvo.h b/drivers/gpu/drm/i915/dvo.h index 8c2ad014c47f..58914691a77b 100644 --- a/drivers/gpu/drm/i915/dvo.h +++ b/drivers/gpu/drm/i915/dvo.h @@ -86,7 +86,7 @@ struct intel_dvo_dev_ops { * buses with clock limitations. */ bool (*mode_fixup)(struct intel_dvo_device *dvo, - struct drm_display_mode *mode, + const struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode); /* diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c index 5363e9c66c27..359f6e8b9b00 100644 --- a/drivers/gpu/drm/i915/i915_debugfs.c +++ b/drivers/gpu/drm/i915/i915_debugfs.c @@ -676,6 +676,7 @@ static void i915_ring_error_state(struct seq_file *m, seq_printf(m, " INSTPM: 0x%08x\n", error->instpm[ring]); seq_printf(m, " FADDR: 0x%08x\n", error->faddr[ring]); if (INTEL_INFO(dev)->gen >= 6) { + seq_printf(m, " RC PSMI: 0x%08x\n", error->rc_psmi[ring]); seq_printf(m, " FAULT_REG: 0x%08x\n", error->fault_reg[ring]); seq_printf(m, " SYNC_0: 0x%08x\n", error->semaphore_mboxes[ring][0]); @@ -713,6 +714,7 @@ static int i915_error_state(struct seq_file *m, void *unused) seq_printf(m, "EIR: 0x%08x\n", error->eir); seq_printf(m, "IER: 0x%08x\n", error->ier); seq_printf(m, "PGTBL_ER: 0x%08x\n", error->pgtbl_er); + seq_printf(m, "CCID: 0x%08x\n", error->ccid); for (i = 0; i < dev_priv->num_fence_regs; i++) seq_printf(m, " fence[%d] = %08llx\n", i, error->fence[i]); @@ -1765,6 +1767,64 @@ static const struct file_operations i915_max_freq_fops = { }; static ssize_t +i915_min_freq_read(struct file *filp, char __user *ubuf, size_t max, + loff_t *ppos) +{ + struct drm_device *dev = filp->private_data; + drm_i915_private_t *dev_priv = dev->dev_private; + char buf[80]; + int len; + + len = snprintf(buf, sizeof(buf), + "min freq: %d\n", dev_priv->min_delay * 50); + + if (len > sizeof(buf)) + len = sizeof(buf); + + return simple_read_from_buffer(ubuf, max, ppos, buf, len); +} + +static ssize_t +i915_min_freq_write(struct file *filp, const char __user *ubuf, size_t cnt, + loff_t *ppos) +{ + struct drm_device *dev = filp->private_data; + struct drm_i915_private *dev_priv = dev->dev_private; + char buf[20]; + int val = 1; + + if (cnt > 0) { + if (cnt > sizeof(buf) - 1) + return -EINVAL; + + if (copy_from_user(buf, ubuf, cnt)) + return -EFAULT; + buf[cnt] = 0; + + val = simple_strtoul(buf, NULL, 0); + } + + DRM_DEBUG_DRIVER("Manually setting min freq to %d\n", val); + + /* + * Turbo will still be enabled, but won't go below the set value. + */ + dev_priv->min_delay = val / 50; + + gen6_set_rps(dev, val / 50); + + return cnt; +} + +static const struct file_operations i915_min_freq_fops = { + .owner = THIS_MODULE, + .open = simple_open, + .read = i915_min_freq_read, + .write = i915_min_freq_write, + .llseek = default_llseek, +}; + +static ssize_t i915_cache_sharing_read(struct file *filp, char __user *ubuf, size_t max, @@ -1997,6 +2057,12 @@ int i915_debugfs_init(struct drm_minor *minor) return ret; ret = i915_debugfs_create(minor->debugfs_root, minor, + "i915_min_freq", + &i915_min_freq_fops); + if (ret) + return ret; + + ret = i915_debugfs_create(minor->debugfs_root, minor, "i915_cache_sharing", &i915_cache_sharing_fops); if (ret) @@ -2028,6 +2094,8 @@ void i915_debugfs_cleanup(struct drm_minor *minor) 1, minor); drm_debugfs_remove_files((struct drm_info_list *) &i915_max_freq_fops, 1, minor); + drm_debugfs_remove_files((struct drm_info_list *) &i915_min_freq_fops, + 1, minor); drm_debugfs_remove_files((struct drm_info_list *) &i915_cache_sharing_fops, 1, minor); drm_debugfs_remove_files((struct drm_info_list *) &i915_ring_stop_fops, diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c index 36822b924eb1..9cf7dfe022b9 100644 --- a/drivers/gpu/drm/i915/i915_dma.c +++ b/drivers/gpu/drm/i915/i915_dma.c @@ -1006,6 +1006,9 @@ static int i915_getparam(struct drm_device *dev, void *data, case I915_PARAM_HAS_ALIASING_PPGTT: value = dev_priv->mm.aliasing_ppgtt ? 1 : 0; break; + case I915_PARAM_HAS_WAIT_TIMEOUT: + value = 1; + break; default: DRM_DEBUG_DRIVER("Unknown parameter %d\n", param->param); @@ -1082,8 +1085,8 @@ static int i915_set_status_page(struct drm_device *dev, void *data, ring->status_page.gfx_addr = hws->addr & (0x1ffff<<12); - dev_priv->dri1.gfx_hws_cpu_addr = ioremap_wc(dev->agp->base + hws->addr, - 4096); + dev_priv->dri1.gfx_hws_cpu_addr = + ioremap_wc(dev_priv->mm.gtt_base_addr + hws->addr, 4096); if (dev_priv->dri1.gfx_hws_cpu_addr == NULL) { i915_dma_cleanup(dev); ring->status_page.gfx_addr = 0; @@ -1411,7 +1414,7 @@ static void i915_kick_out_firmware_fb(struct drm_i915_private *dev_priv) if (!ap) return; - ap->ranges[0].base = dev_priv->dev->agp->base; + ap->ranges[0].base = dev_priv->mm.gtt->gma_bus_addr; ap->ranges[0].size = dev_priv->mm.gtt->gtt_mappable_entries << PAGE_SHIFT; primary = @@ -1467,11 +1470,18 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags) goto free_priv; } + ret = intel_gmch_probe(dev_priv->bridge_dev, dev->pdev, NULL); + if (!ret) { + DRM_ERROR("failed to set up gmch\n"); + ret = -EIO; + goto put_bridge; + } + dev_priv->mm.gtt = intel_gtt_get(); if (!dev_priv->mm.gtt) { DRM_ERROR("Failed to initialize GTT\n"); ret = -ENODEV; - goto put_bridge; + goto put_gmch; } i915_kick_out_firmware_fb(dev_priv); @@ -1498,19 +1508,22 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags) if (!dev_priv->regs) { DRM_ERROR("failed to map registers\n"); ret = -EIO; - goto put_bridge; + goto put_gmch; } aperture_size = dev_priv->mm.gtt->gtt_mappable_entries << PAGE_SHIFT; + dev_priv->mm.gtt_base_addr = dev_priv->mm.gtt->gma_bus_addr; dev_priv->mm.gtt_mapping = - io_mapping_create_wc(dev->agp->base, aperture_size); + io_mapping_create_wc(dev_priv->mm.gtt_base_addr, + aperture_size); if (dev_priv->mm.gtt_mapping == NULL) { ret = -EIO; goto out_rmmap; } - i915_mtrr_setup(dev_priv, dev->agp->base, aperture_size); + i915_mtrr_setup(dev_priv, dev_priv->mm.gtt_base_addr, + aperture_size); /* The i915 workqueue is primarily used for batched retirement of * requests (and thus managing bo) once the task has been completed @@ -1534,7 +1547,11 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags) goto out_mtrrfree; } + /* This must be called before any calls to HAS_PCH_* */ + intel_detect_pch(dev); + intel_irq_init(dev); + intel_gt_init(dev); /* Try to make sure MCHBAR is enabled before poking at it */ intel_setup_mchbar(dev); @@ -1567,7 +1584,6 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags) if (!IS_I945G(dev) && !IS_I945GM(dev)) pci_enable_msi(dev->pdev); - spin_lock_init(&dev_priv->gt_lock); spin_lock_init(&dev_priv->irq_lock); spin_lock_init(&dev_priv->error_lock); spin_lock_init(&dev_priv->rps_lock); @@ -1586,8 +1602,6 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags) /* Start out suspended */ dev_priv->mm.suspended = 1; - intel_detect_pch(dev); - if (drm_core_check_feature(dev, DRIVER_MODESET)) { ret = i915_load_modeset_init(dev); if (ret < 0) { @@ -1622,13 +1636,16 @@ out_gem_unload: destroy_workqueue(dev_priv->wq); out_mtrrfree: if (dev_priv->mm.gtt_mtrr >= 0) { - mtrr_del(dev_priv->mm.gtt_mtrr, dev->agp->base, - dev->agp->agp_info.aper_size * 1024 * 1024); + mtrr_del(dev_priv->mm.gtt_mtrr, + dev_priv->mm.gtt_base_addr, + aperture_size); dev_priv->mm.gtt_mtrr = -1; } io_mapping_free(dev_priv->mm.gtt_mapping); out_rmmap: pci_iounmap(dev->pdev, dev_priv->regs); +put_gmch: + intel_gmch_remove(); put_bridge: pci_dev_put(dev_priv->bridge_dev); free_priv: @@ -1660,8 +1677,9 @@ int i915_driver_unload(struct drm_device *dev) io_mapping_free(dev_priv->mm.gtt_mapping); if (dev_priv->mm.gtt_mtrr >= 0) { - mtrr_del(dev_priv->mm.gtt_mtrr, dev->agp->base, - dev->agp->agp_info.aper_size * 1024 * 1024); + mtrr_del(dev_priv->mm.gtt_mtrr, + dev_priv->mm.gtt_base_addr, + dev_priv->mm.gtt->gtt_mappable_entries * PAGE_SIZE); dev_priv->mm.gtt_mtrr = -1; } @@ -1702,6 +1720,7 @@ int i915_driver_unload(struct drm_device *dev) mutex_lock(&dev->struct_mutex); i915_gem_free_all_phys_object(dev); i915_gem_cleanup_ringbuffer(dev); + i915_gem_context_fini(dev); mutex_unlock(&dev->struct_mutex); i915_gem_cleanup_aliasing_ppgtt(dev); i915_gem_cleanup_stolen(dev); @@ -1741,6 +1760,8 @@ int i915_driver_open(struct drm_device *dev, struct drm_file *file) spin_lock_init(&file_priv->mm.lock); INIT_LIST_HEAD(&file_priv->mm.request_list); + idr_init(&file_priv->context_idr); + return 0; } @@ -1760,7 +1781,13 @@ void i915_driver_lastclose(struct drm_device * dev) { drm_i915_private_t *dev_priv = dev->dev_private; - if (!dev_priv || drm_core_check_feature(dev, DRIVER_MODESET)) { + /* On gen6+ we refuse to init without kms enabled, but then the drm core + * goes right around and calls lastclose. Check for this and don't clean + * up anything. */ + if (!dev_priv) + return; + + if (drm_core_check_feature(dev, DRIVER_MODESET)) { intel_fb_restore_mode(dev); vga_switcheroo_process_delayed_switch(); return; @@ -1773,6 +1800,7 @@ void i915_driver_lastclose(struct drm_device * dev) void i915_driver_preclose(struct drm_device * dev, struct drm_file *file_priv) { + i915_gem_context_close(dev, file_priv); i915_gem_release(dev, file_priv); } @@ -1826,6 +1854,9 @@ struct drm_ioctl_desc i915_ioctls[] = { DRM_IOCTL_DEF_DRV(I915_OVERLAY_ATTRS, intel_overlay_attrs, DRM_MASTER|DRM_CONTROL_ALLOW|DRM_UNLOCKED), DRM_IOCTL_DEF_DRV(I915_SET_SPRITE_COLORKEY, intel_sprite_set_colorkey, DRM_MASTER|DRM_CONTROL_ALLOW|DRM_UNLOCKED), DRM_IOCTL_DEF_DRV(I915_GET_SPRITE_COLORKEY, intel_sprite_get_colorkey, DRM_MASTER|DRM_CONTROL_ALLOW|DRM_UNLOCKED), + DRM_IOCTL_DEF_DRV(I915_GEM_WAIT, i915_gem_wait_ioctl, DRM_AUTH|DRM_UNLOCKED), + DRM_IOCTL_DEF_DRV(I915_GEM_CONTEXT_CREATE, i915_gem_context_create_ioctl, DRM_UNLOCKED), + DRM_IOCTL_DEF_DRV(I915_GEM_CONTEXT_DESTROY, i915_gem_context_destroy_ioctl, DRM_UNLOCKED), }; int i915_max_ioctl = DRM_ARRAY_SIZE(i915_ioctls); diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c index 9fe9ebe52a7a..ed22612bc847 100644 --- a/drivers/gpu/drm/i915/i915_drv.c +++ b/drivers/gpu/drm/i915/i915_drv.c @@ -32,6 +32,7 @@ #include "drm.h" #include "i915_drm.h" #include "i915_drv.h" +#include "i915_trace.h" #include "intel_drv.h" #include <linux/console.h> @@ -215,7 +216,6 @@ static const struct intel_device_info intel_ironlake_d_info = { .gen = 5, .need_gfx_hws = 1, .has_hotplug = 1, .has_bsd_ring = 1, - .has_pch_split = 1, }; static const struct intel_device_info intel_ironlake_m_info = { @@ -223,7 +223,6 @@ static const struct intel_device_info intel_ironlake_m_info = { .need_gfx_hws = 1, .has_hotplug = 1, .has_fbc = 1, .has_bsd_ring = 1, - .has_pch_split = 1, }; static const struct intel_device_info intel_sandybridge_d_info = { @@ -232,7 +231,6 @@ static const struct intel_device_info intel_sandybridge_d_info = { .has_bsd_ring = 1, .has_blt_ring = 1, .has_llc = 1, - .has_pch_split = 1, .has_force_wake = 1, }; @@ -243,7 +241,6 @@ static const struct intel_device_info intel_sandybridge_m_info = { .has_bsd_ring = 1, .has_blt_ring = 1, .has_llc = 1, - .has_pch_split = 1, .has_force_wake = 1, }; @@ -253,7 +250,6 @@ static const struct intel_device_info intel_ivybridge_d_info = { .has_bsd_ring = 1, .has_blt_ring = 1, .has_llc = 1, - .has_pch_split = 1, .has_force_wake = 1, }; @@ -264,7 +260,6 @@ static const struct intel_device_info intel_ivybridge_m_info = { .has_bsd_ring = 1, .has_blt_ring = 1, .has_llc = 1, - .has_pch_split = 1, .has_force_wake = 1, }; @@ -292,7 +287,6 @@ static const struct intel_device_info intel_haswell_d_info = { .has_bsd_ring = 1, .has_blt_ring = 1, .has_llc = 1, - .has_pch_split = 1, .has_force_wake = 1, }; @@ -302,7 +296,6 @@ static const struct intel_device_info intel_haswell_m_info = { .has_bsd_ring = 1, .has_blt_ring = 1, .has_llc = 1, - .has_pch_split = 1, .has_force_wake = 1, }; @@ -358,6 +351,9 @@ static const struct pci_device_id pciidlist[] = { /* aka */ INTEL_VGA_DEVICE(0x0406, &intel_haswell_m_info), /* GT1 mobile */ INTEL_VGA_DEVICE(0x0416, &intel_haswell_m_info), /* GT2 mobile */ INTEL_VGA_DEVICE(0x0c16, &intel_haswell_d_info), /* SDV */ + INTEL_VGA_DEVICE(0x0f30, &intel_valleyview_m_info), + INTEL_VGA_DEVICE(0x0157, &intel_valleyview_m_info), + INTEL_VGA_DEVICE(0x0155, &intel_valleyview_d_info), {0, 0, 0} }; @@ -429,135 +425,6 @@ bool i915_semaphore_is_enabled(struct drm_device *dev) return 1; } -void __gen6_gt_force_wake_get(struct drm_i915_private *dev_priv) -{ - int count; - - count = 0; - while (count++ < 50 && (I915_READ_NOTRACE(FORCEWAKE_ACK) & 1)) - udelay(10); - - I915_WRITE_NOTRACE(FORCEWAKE, 1); - POSTING_READ(FORCEWAKE); - - count = 0; - while (count++ < 50 && (I915_READ_NOTRACE(FORCEWAKE_ACK) & 1) == 0) - udelay(10); -} - -void __gen6_gt_force_wake_mt_get(struct drm_i915_private *dev_priv) -{ - int count; - - count = 0; - while (count++ < 50 && (I915_READ_NOTRACE(FORCEWAKE_MT_ACK) & 1)) - udelay(10); - - I915_WRITE_NOTRACE(FORCEWAKE_MT, _MASKED_BIT_ENABLE(1)); - POSTING_READ(FORCEWAKE_MT); - - count = 0; - while (count++ < 50 && (I915_READ_NOTRACE(FORCEWAKE_MT_ACK) & 1) == 0) - udelay(10); -} - -/* - * Generally this is called implicitly by the register read function. However, - * if some sequence requires the GT to not power down then this function should - * be called at the beginning of the sequence followed by a call to - * gen6_gt_force_wake_put() at the end of the sequence. - */ -void gen6_gt_force_wake_get(struct drm_i915_private *dev_priv) -{ - unsigned long irqflags; - - spin_lock_irqsave(&dev_priv->gt_lock, irqflags); - if (dev_priv->forcewake_count++ == 0) - dev_priv->display.force_wake_get(dev_priv); - spin_unlock_irqrestore(&dev_priv->gt_lock, irqflags); -} - -static void gen6_gt_check_fifodbg(struct drm_i915_private *dev_priv) -{ - u32 gtfifodbg; - gtfifodbg = I915_READ_NOTRACE(GTFIFODBG); - if (WARN(gtfifodbg & GT_FIFO_CPU_ERROR_MASK, - "MMIO read or write has been dropped %x\n", gtfifodbg)) - I915_WRITE_NOTRACE(GTFIFODBG, GT_FIFO_CPU_ERROR_MASK); -} - -void __gen6_gt_force_wake_put(struct drm_i915_private *dev_priv) -{ - I915_WRITE_NOTRACE(FORCEWAKE, 0); - /* The below doubles as a POSTING_READ */ - gen6_gt_check_fifodbg(dev_priv); -} - -void __gen6_gt_force_wake_mt_put(struct drm_i915_private *dev_priv) -{ - I915_WRITE_NOTRACE(FORCEWAKE_MT, _MASKED_BIT_DISABLE(1)); - /* The below doubles as a POSTING_READ */ - gen6_gt_check_fifodbg(dev_priv); -} - -/* - * see gen6_gt_force_wake_get() - */ -void gen6_gt_force_wake_put(struct drm_i915_private *dev_priv) -{ - unsigned long irqflags; - - spin_lock_irqsave(&dev_priv->gt_lock, irqflags); - if (--dev_priv->forcewake_count == 0) - dev_priv->display.force_wake_put(dev_priv); - spin_unlock_irqrestore(&dev_priv->gt_lock, irqflags); -} - -int __gen6_gt_wait_for_fifo(struct drm_i915_private *dev_priv) -{ - int ret = 0; - - if (dev_priv->gt_fifo_count < GT_FIFO_NUM_RESERVED_ENTRIES) { - int loop = 500; - u32 fifo = I915_READ_NOTRACE(GT_FIFO_FREE_ENTRIES); - while (fifo <= GT_FIFO_NUM_RESERVED_ENTRIES && loop--) { - udelay(10); - fifo = I915_READ_NOTRACE(GT_FIFO_FREE_ENTRIES); - } - if (WARN_ON(loop < 0 && fifo <= GT_FIFO_NUM_RESERVED_ENTRIES)) - ++ret; - dev_priv->gt_fifo_count = fifo; - } - dev_priv->gt_fifo_count--; - - return ret; -} - -void vlv_force_wake_get(struct drm_i915_private *dev_priv) -{ - int count; - - count = 0; - - /* Already awake? */ - if ((I915_READ(0x130094) & 0xa1) == 0xa1) - return; - - I915_WRITE_NOTRACE(FORCEWAKE_VLV, 0xffffffff); - POSTING_READ(FORCEWAKE_VLV); - - count = 0; - while (count++ < 50 && (I915_READ_NOTRACE(FORCEWAKE_ACK_VLV) & 1) == 0) - udelay(10); -} - -void vlv_force_wake_put(struct drm_i915_private *dev_priv) -{ - I915_WRITE_NOTRACE(FORCEWAKE_VLV, 0xffff0000); - /* FIXME: confirm VLV behavior with Punit folks */ - POSTING_READ(FORCEWAKE_VLV); -} - static int i915_drm_freeze(struct drm_device *dev) { struct drm_i915_private *dev_priv = dev->dev_private; @@ -637,7 +504,7 @@ static int i915_drm_thaw(struct drm_device *dev) /* KMS EnterVT equivalent */ if (drm_core_check_feature(dev, DRIVER_MODESET)) { - if (HAS_PCH_SPLIT(dev)) + if (HAS_PCH_IBX(dev) || HAS_PCH_CPT(dev)) ironlake_init_pch_refclk(dev); mutex_lock(&dev->struct_mutex); @@ -794,9 +661,9 @@ static int gen6_do_reset(struct drm_device *dev) /* If reset with a user forcewake, try to restore, otherwise turn it off */ if (dev_priv->forcewake_count) - dev_priv->display.force_wake_get(dev_priv); + dev_priv->gt.force_wake_get(dev_priv); else - dev_priv->display.force_wake_put(dev_priv); + dev_priv->gt.force_wake_put(dev_priv); /* Restore fifo count */ dev_priv->gt_fifo_count = I915_READ_NOTRACE(GT_FIFO_FREE_ENTRIES); @@ -805,7 +672,7 @@ static int gen6_do_reset(struct drm_device *dev) return ret; } -static int intel_gpu_reset(struct drm_device *dev) +int intel_gpu_reset(struct drm_device *dev) { struct drm_i915_private *dev_priv = dev->dev_private; int ret = -ENODEV; @@ -863,10 +730,7 @@ int i915_reset(struct drm_device *dev) if (!i915_try_reset) return 0; - if (!mutex_trylock(&dev->struct_mutex)) - return -EBUSY; - - dev_priv->stop_rings = 0; + mutex_lock(&dev->struct_mutex); i915_gem_reset(dev); @@ -909,12 +773,16 @@ int i915_reset(struct drm_device *dev) for_each_ring(ring, dev_priv, i) ring->init(ring); + i915_gem_context_init(dev); i915_gem_init_ppgtt(dev); - mutex_unlock(&dev->struct_mutex); + /* + * It would make sense to re-init all the other hw state, at + * least the rps/rc6/emon init done within modeset_init_hw. For + * some unknown reason, this blows up my ilk, so don't. + */ - if (drm_core_check_feature(dev, DRIVER_MODESET)) - intel_modeset_init_hw(dev); + mutex_unlock(&dev->struct_mutex); drm_irq_uninstall(dev); drm_irq_install(dev); @@ -925,10 +793,12 @@ int i915_reset(struct drm_device *dev) return 0; } - static int __devinit i915_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent) { + struct intel_device_info *intel_info = + (struct intel_device_info *) ent->driver_data; + /* Only bind to function 0 of the device. Early generations * used function 1 as a placeholder for multi-head. This causes * us confusion instead, especially on the systems where both @@ -937,6 +807,18 @@ i915_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent) if (PCI_FUNC(pdev->devfn)) return -ENODEV; + /* We've managed to ship a kms-enabled ddx that shipped with an XvMC + * implementation for gen3 (and only gen3) that used legacy drm maps + * (gasp!) to share buffers between X and the client. Hence we need to + * keep around the fake agp stuff for gen3, even when kms is enabled. */ + if (intel_info->gen != 3) { + driver.driver_features &= + ~(DRIVER_USE_AGP | DRIVER_REQUIRE_AGP); + } else if (!intel_agp_enabled) { + DRM_ERROR("drm/i915 can't work without intel_agp module!\n"); + return -ENODEV; + } + return drm_get_pci_dev(pdev, ent, &driver); } @@ -1058,7 +940,6 @@ static struct drm_driver driver = { .resume = i915_resume, .device_is_agp = i915_driver_device_is_agp, - .reclaim_buffers = drm_core_reclaim_buffers, .master_create = i915_master_create, .master_destroy = i915_master_destroy, #if defined(CONFIG_DEBUG_FS) @@ -1097,11 +978,6 @@ static struct pci_driver i915_pci_driver = { static int __init i915_init(void) { - if (!intel_agp_enabled) { - DRM_ERROR("drm/i915 can't work without intel_agp module!\n"); - return -ENODEV; - } - driver.num_ioctls = i915_max_ioctl; /* @@ -1149,6 +1025,84 @@ MODULE_LICENSE("GPL and additional rights"); ((reg) < 0x40000) && \ ((reg) != FORCEWAKE)) +static bool IS_DISPLAYREG(u32 reg) +{ + /* + * This should make it easier to transition modules over to the + * new register block scheme, since we can do it incrementally. + */ + if (reg >= 0x180000) + return false; + + if (reg >= RENDER_RING_BASE && + reg < RENDER_RING_BASE + 0xff) + return false; + if (reg >= GEN6_BSD_RING_BASE && + reg < GEN6_BSD_RING_BASE + 0xff) + return false; + if (reg >= BLT_RING_BASE && + reg < BLT_RING_BASE + 0xff) + return false; + + if (reg == PGTBL_ER) + return false; + + if (reg >= IPEIR_I965 && + reg < HWSTAM) + return false; + + if (reg == MI_MODE) + return false; + + if (reg == GFX_MODE_GEN7) + return false; + + if (reg == RENDER_HWS_PGA_GEN7 || + reg == BSD_HWS_PGA_GEN7 || + reg == BLT_HWS_PGA_GEN7) + return false; + + if (reg == GEN6_BSD_SLEEP_PSMI_CONTROL || + reg == GEN6_BSD_RNCID) + return false; + + if (reg == GEN6_BLITTER_ECOSKPD) + return false; + + if (reg >= 0x4000c && + reg <= 0x4002c) + return false; + + if (reg >= 0x4f000 && + reg <= 0x4f08f) + return false; + + if (reg >= 0x4f100 && + reg <= 0x4f11f) + return false; + + if (reg >= VLV_MASTER_IER && + reg <= GEN6_PMIER) + return false; + + if (reg >= FENCE_REG_SANDYBRIDGE_0 && + reg < (FENCE_REG_SANDYBRIDGE_0 + (16*8))) + return false; + + if (reg >= VLV_IIR_RW && + reg <= VLV_ISR) + return false; + + if (reg == FORCEWAKE_VLV || + reg == FORCEWAKE_ACK_VLV) + return false; + + if (reg == GEN6_GDRST) + return false; + + return true; +} + #define __i915_read(x, y) \ u##x i915_read##x(struct drm_i915_private *dev_priv, u32 reg) { \ u##x val = 0; \ @@ -1156,11 +1110,13 @@ u##x i915_read##x(struct drm_i915_private *dev_priv, u32 reg) { \ unsigned long irqflags; \ spin_lock_irqsave(&dev_priv->gt_lock, irqflags); \ if (dev_priv->forcewake_count == 0) \ - dev_priv->display.force_wake_get(dev_priv); \ + dev_priv->gt.force_wake_get(dev_priv); \ val = read##y(dev_priv->regs + reg); \ if (dev_priv->forcewake_count == 0) \ - dev_priv->display.force_wake_put(dev_priv); \ + dev_priv->gt.force_wake_put(dev_priv); \ spin_unlock_irqrestore(&dev_priv->gt_lock, irqflags); \ + } else if (IS_VALLEYVIEW(dev_priv->dev) && IS_DISPLAYREG(reg)) { \ + val = read##y(dev_priv->regs + reg + 0x180000); \ } else { \ val = read##y(dev_priv->regs + reg); \ } \ @@ -1181,7 +1137,11 @@ void i915_write##x(struct drm_i915_private *dev_priv, u32 reg, u##x val) { \ if (NEEDS_FORCE_WAKE((dev_priv), (reg))) { \ __fifo_ret = __gen6_gt_wait_for_fifo(dev_priv); \ } \ - write##y(val, dev_priv->regs + reg); \ + if (IS_VALLEYVIEW(dev_priv->dev) && IS_DISPLAYREG(reg)) { \ + write##y(val, dev_priv->regs + reg + 0x180000); \ + } else { \ + write##y(val, dev_priv->regs + reg); \ + } \ if (unlikely(__fifo_ret)) { \ gen6_gt_check_fifodbg(dev_priv); \ } \ diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index b0b676abde0d..627fe35781b4 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -79,6 +79,10 @@ enum port { #define for_each_pipe(p) for ((p) = 0; (p) < dev_priv->num_pipe; (p)++) +#define for_each_encoder_on_crtc(dev, __crtc, intel_encoder) \ + list_for_each_entry((intel_encoder), &(dev)->mode_config.encoder_list, base.head) \ + if ((intel_encoder)->base.crtc == (__crtc)) + struct intel_pch_pll { int refcount; /* count of number of CRTCs sharing this PLL */ int active; /* count of number of active CRTCs (i.e. DPMS on) */ @@ -176,6 +180,7 @@ struct drm_i915_error_state { u32 eir; u32 pgtbl_er; u32 ier; + u32 ccid; bool waiting[I915_NUM_RINGS]; u32 pipestat[I915_MAX_PIPES]; u32 tail[I915_NUM_RINGS]; @@ -185,6 +190,7 @@ struct drm_i915_error_state { u32 instdone[I915_NUM_RINGS]; u32 acthd[I915_NUM_RINGS]; u32 semaphore_mboxes[I915_NUM_RINGS][I915_NUM_RINGS - 1]; + u32 rc_psmi[I915_NUM_RINGS]; /* sleep state */ /* our own tracking of ring head and tail */ u32 cpu_ring_head[I915_NUM_RINGS]; u32 cpu_ring_tail[I915_NUM_RINGS]; @@ -261,8 +267,6 @@ struct drm_i915_display_funcs { struct drm_i915_gem_object *obj); int (*update_plane)(struct drm_crtc *crtc, struct drm_framebuffer *fb, int x, int y); - void (*force_wake_get)(struct drm_i915_private *dev_priv); - void (*force_wake_put)(struct drm_i915_private *dev_priv); /* clock updates for mode set */ /* cursor updates */ /* render clock increase/decrease */ @@ -270,6 +274,11 @@ struct drm_i915_display_funcs { /* pll clock increase/decrease */ }; +struct drm_i915_gt_funcs { + void (*force_wake_get)(struct drm_i915_private *dev_priv); + void (*force_wake_put)(struct drm_i915_private *dev_priv); +}; + struct intel_device_info { u8 gen; u8 is_mobile:1; @@ -284,7 +293,6 @@ struct intel_device_info { u8 is_crestline:1; u8 is_ivybridge:1; u8 is_valleyview:1; - u8 has_pch_split:1; u8 has_force_wake:1; u8 is_haswell:1; u8 has_fbc:1; @@ -309,6 +317,17 @@ struct i915_hw_ppgtt { dma_addr_t scratch_page_dma_addr; }; + +/* This must match up with the value previously used for execbuf2.rsvd1. */ +#define DEFAULT_CONTEXT_ID 0 +struct i915_hw_context { + int id; + bool is_initialized; + struct drm_i915_file_private *file_priv; + struct intel_ring_buffer *ring; + struct drm_i915_gem_object *obj; +}; + enum no_fbc_reason { FBC_NO_OUTPUT, /* no outputs enabled to compress */ FBC_STOLEN_TOO_SMALL, /* not enough space to hold compressed buffers */ @@ -321,6 +340,7 @@ enum no_fbc_reason { }; enum intel_pch { + PCH_NONE = 0, /* No PCH present */ PCH_IBX, /* Ibexpeak PCH */ PCH_CPT, /* Cougarpoint PCH */ PCH_LPT, /* Lynxpoint PCH */ @@ -350,6 +370,8 @@ typedef struct drm_i915_private { int relative_constants_mode; void __iomem *regs; + + struct drm_i915_gt_funcs gt; /** gt_fifo_count and the subsequent register write are synchronized * with dev->struct_mutex. */ unsigned gt_fifo_count; @@ -652,11 +674,14 @@ typedef struct drm_i915_private { unsigned long gtt_end; struct io_mapping *gtt_mapping; + phys_addr_t gtt_base_addr; int gtt_mtrr; /** PPGTT used for aliasing the PPGTT with the GTT */ struct i915_hw_ppgtt *aliasing_ppgtt; + u32 *l3_remap_info; + struct shrinker inactive_shrinker; /** @@ -817,6 +842,10 @@ typedef struct drm_i915_private { struct drm_property *broadcast_rgb_property; struct drm_property *force_audio_property; + + struct work_struct parity_error_work; + bool hw_contexts_disabled; + uint32_t hw_context_size; } drm_i915_private_t; /* Iterate over initialised rings */ @@ -1026,6 +1055,7 @@ struct drm_i915_file_private { struct spinlock lock; struct list_head request_list; } mm; + struct idr context_idr; }; #define INTEL_INFO(dev) (((struct drm_i915_private *) (dev)->dev_private)->info) @@ -1071,7 +1101,8 @@ struct drm_i915_file_private { #define HAS_LLC(dev) (INTEL_INFO(dev)->has_llc) #define I915_NEED_GFX_HWS(dev) (INTEL_INFO(dev)->need_gfx_hws) -#define HAS_ALIASING_PPGTT(dev) (INTEL_INFO(dev)->gen >=6) +#define HAS_HW_CONTEXTS(dev) (INTEL_INFO(dev)->gen >= 6) +#define HAS_ALIASING_PPGTT(dev) (INTEL_INFO(dev)->gen >=6 && !IS_VALLEYVIEW(dev)) #define HAS_OVERLAY(dev) (INTEL_INFO(dev)->has_overlay) #define OVERLAY_NEEDS_PHYSICAL(dev) (INTEL_INFO(dev)->overlay_needs_physical) @@ -1094,13 +1125,13 @@ struct drm_i915_file_private { #define HAS_PIPE_CXSR(dev) (INTEL_INFO(dev)->has_pipe_cxsr) #define I915_HAS_FBC(dev) (INTEL_INFO(dev)->has_fbc) -#define HAS_PCH_SPLIT(dev) (INTEL_INFO(dev)->has_pch_split) #define HAS_PIPE_CONTROL(dev) (INTEL_INFO(dev)->gen >= 5) #define INTEL_PCH_TYPE(dev) (((struct drm_i915_private *)(dev)->dev_private)->pch_type) #define HAS_PCH_LPT(dev) (INTEL_PCH_TYPE(dev) == PCH_LPT) #define HAS_PCH_CPT(dev) (INTEL_PCH_TYPE(dev) == PCH_CPT) #define HAS_PCH_IBX(dev) (INTEL_PCH_TYPE(dev) == PCH_IBX) +#define HAS_PCH_SPLIT(dev) (INTEL_PCH_TYPE(dev) != PCH_NONE) #define HAS_FORCE_WAKE(dev) (INTEL_INFO(dev)->has_force_wake) @@ -1166,6 +1197,7 @@ extern long i915_compat_ioctl(struct file *filp, unsigned int cmd, extern int i915_emit_box(struct drm_device *dev, struct drm_clip_rect *box, int DR1, int DR4); +extern int intel_gpu_reset(struct drm_device *dev); extern int i915_reset(struct drm_device *dev); extern unsigned long i915_chipset_val(struct drm_i915_private *dev_priv); extern unsigned long i915_mch_val(struct drm_i915_private *dev_priv); @@ -1178,6 +1210,7 @@ void i915_hangcheck_elapsed(unsigned long data); void i915_handle_error(struct drm_device *dev, bool wedged); extern void intel_irq_init(struct drm_device *dev); +extern void intel_gt_init(struct drm_device *dev); void i915_error_state_free(struct kref *error_ref); @@ -1237,6 +1270,8 @@ int i915_gem_get_tiling(struct drm_device *dev, void *data, struct drm_file *file_priv); int i915_gem_get_aperture_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv); +int i915_gem_wait_ioctl(struct drm_device *dev, void *data, + struct drm_file *file_priv); void i915_gem_load(struct drm_device *dev); int i915_gem_init_object(struct drm_gem_object *obj); int __must_check i915_gem_flush_ring(struct intel_ring_buffer *ring, @@ -1306,6 +1341,8 @@ i915_gem_object_unpin_fence(struct drm_i915_gem_object *obj) void i915_gem_retire_requests(struct drm_device *dev); void i915_gem_retire_requests_ring(struct intel_ring_buffer *ring); +int __must_check i915_gem_check_wedge(struct drm_i915_private *dev_priv, + bool interruptible); void i915_gem_reset(struct drm_device *dev); void i915_gem_clflush_object(struct drm_i915_gem_object *obj); @@ -1315,6 +1352,7 @@ int __must_check i915_gem_object_set_domain(struct drm_i915_gem_object *obj, int __must_check i915_gem_object_finish_gpu(struct drm_i915_gem_object *obj); int __must_check i915_gem_init(struct drm_device *dev); int __must_check i915_gem_init_hw(struct drm_device *dev); +void i915_gem_l3_remap(struct drm_device *dev); void i915_gem_init_swizzling(struct drm_device *dev); void i915_gem_init_ppgtt(struct drm_device *dev); void i915_gem_cleanup_ringbuffer(struct drm_device *dev); @@ -1323,8 +1361,8 @@ int __must_check i915_gem_idle(struct drm_device *dev); int __must_check i915_add_request(struct intel_ring_buffer *ring, struct drm_file *file, struct drm_i915_gem_request *request); -int __must_check i915_wait_request(struct intel_ring_buffer *ring, - uint32_t seqno); +int __must_check i915_wait_seqno(struct intel_ring_buffer *ring, + uint32_t seqno); int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf); int __must_check i915_gem_object_set_to_gtt_domain(struct drm_i915_gem_object *obj, @@ -1358,6 +1396,16 @@ struct drm_gem_object *i915_gem_prime_import(struct drm_device *dev, struct dma_buf *i915_gem_prime_export(struct drm_device *dev, struct drm_gem_object *gem_obj, int flags); +/* i915_gem_context.c */ +void i915_gem_context_init(struct drm_device *dev); +void i915_gem_context_fini(struct drm_device *dev); +void i915_gem_context_close(struct drm_device *dev, struct drm_file *file); +int i915_switch_context(struct intel_ring_buffer *ring, + struct drm_file *file, int to_id); +int i915_gem_context_create_ioctl(struct drm_device *dev, void *data, + struct drm_file *file); +int i915_gem_context_destroy_ioctl(struct drm_device *dev, void *data, + struct drm_file *file); /* i915_gem_gtt.c */ int __must_check i915_gem_init_aliasing_ppgtt(struct drm_device *dev); @@ -1475,20 +1523,12 @@ extern bool intel_fbc_enabled(struct drm_device *dev); extern void intel_disable_fbc(struct drm_device *dev); extern bool ironlake_set_drps(struct drm_device *dev, u8 val); extern void ironlake_init_pch_refclk(struct drm_device *dev); -extern void ironlake_enable_rc6(struct drm_device *dev); extern void gen6_set_rps(struct drm_device *dev, u8 val); extern void intel_detect_pch(struct drm_device *dev); extern int intel_trans_dp_port_sel(struct drm_crtc *crtc); extern int intel_enable_rc6(const struct drm_device *dev); extern bool i915_semaphore_is_enabled(struct drm_device *dev); -extern void __gen6_gt_force_wake_get(struct drm_i915_private *dev_priv); -extern void __gen6_gt_force_wake_mt_get(struct drm_i915_private *dev_priv); -extern void __gen6_gt_force_wake_put(struct drm_i915_private *dev_priv); -extern void __gen6_gt_force_wake_mt_put(struct drm_i915_private *dev_priv); - -extern void vlv_force_wake_get(struct drm_i915_private *dev_priv); -extern void vlv_force_wake_put(struct drm_i915_private *dev_priv); /* overlay */ #ifdef CONFIG_DEBUG_FS diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index 288d7b8f49ae..5c4657a54f97 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -96,9 +96,18 @@ i915_gem_wait_for_error(struct drm_device *dev) if (!atomic_read(&dev_priv->mm.wedged)) return 0; - ret = wait_for_completion_interruptible(x); - if (ret) + /* + * Only wait 10 seconds for the gpu reset to complete to avoid hanging + * userspace. If it takes that long something really bad is going on and + * we should simply try to bail out and fail as gracefully as possible. + */ + ret = wait_for_completion_interruptible_timeout(x, 10*HZ); + if (ret == 0) { + DRM_ERROR("Timed out waiting for the gpu reset to complete\n"); + return -EIO; + } else if (ret < 0) { return ret; + } if (atomic_read(&dev_priv->mm.wedged)) { /* GPU is hung, bump the completion count to account for @@ -1122,7 +1131,7 @@ int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf) obj->fault_mappable = true; - pfn = ((dev->agp->base + obj->gtt_offset) >> PAGE_SHIFT) + + pfn = ((dev_priv->mm.gtt_base_addr + obj->gtt_offset) >> PAGE_SHIFT) + page_offset; /* Finally, remap it using the new GTT offset */ @@ -1132,6 +1141,11 @@ unlock: out: switch (ret) { case -EIO: + /* If this -EIO is due to a gpu hang, give the reset code a + * chance to clean up the mess. Otherwise return the proper + * SIGBUS. */ + if (!atomic_read(&dev_priv->mm.wedged)) + return VM_FAULT_SIGBUS; case -EAGAIN: /* Give the error handler a chance to run and move the * objects off the GPU active list. Next time we service the @@ -1568,6 +1582,21 @@ i915_add_request(struct intel_ring_buffer *ring, int was_empty; int ret; + /* + * Emit any outstanding flushes - execbuf can fail to emit the flush + * after having emitted the batchbuffer command. Hence we need to fix + * things up similar to emitting the lazy request. The difference here + * is that the flush _must_ happen before the next request, no matter + * what. + */ + if (ring->gpu_caches_dirty) { + ret = i915_gem_flush_ring(ring, 0, I915_GEM_GPU_DOMAINS); + if (ret) + return ret; + + ring->gpu_caches_dirty = false; + } + BUG_ON(request == NULL); seqno = i915_gem_next_request_seqno(ring); @@ -1613,6 +1642,9 @@ i915_add_request(struct intel_ring_buffer *ring, queue_delayed_work(dev_priv->wq, &dev_priv->mm.retire_work, HZ); } + + WARN_ON(!list_empty(&ring->gpu_write_list)); + return 0; } @@ -1827,14 +1859,11 @@ i915_gem_retire_work_handler(struct work_struct *work) */ idle = true; for_each_ring(ring, dev_priv, i) { - if (!list_empty(&ring->gpu_write_list)) { + if (ring->gpu_caches_dirty) { struct drm_i915_gem_request *request; - int ret; - ret = i915_gem_flush_ring(ring, - 0, I915_GEM_GPU_DOMAINS); request = kzalloc(sizeof(*request), GFP_KERNEL); - if (ret || request == NULL || + if (request == NULL || i915_add_request(ring, NULL, request)) kfree(request); } @@ -1848,11 +1877,10 @@ i915_gem_retire_work_handler(struct work_struct *work) mutex_unlock(&dev->struct_mutex); } -static int -i915_gem_check_wedge(struct drm_i915_private *dev_priv) +int +i915_gem_check_wedge(struct drm_i915_private *dev_priv, + bool interruptible) { - BUG_ON(!mutex_is_locked(&dev_priv->dev->struct_mutex)); - if (atomic_read(&dev_priv->mm.wedged)) { struct completion *x = &dev_priv->error_completion; bool recovery_complete; @@ -1863,7 +1891,16 @@ i915_gem_check_wedge(struct drm_i915_private *dev_priv) recovery_complete = x->done > 0; spin_unlock_irqrestore(&x->wait.lock, flags); - return recovery_complete ? -EIO : -EAGAIN; + /* Non-interruptible callers can't handle -EAGAIN, hence return + * -EIO unconditionally for these. */ + if (!interruptible) + return -EIO; + + /* Recovery complete, but still wedged means reset failure. */ + if (recovery_complete) + return -EIO; + + return -EAGAIN; } return 0; @@ -1899,34 +1936,85 @@ i915_gem_check_olr(struct intel_ring_buffer *ring, u32 seqno) return ret; } +/** + * __wait_seqno - wait until execution of seqno has finished + * @ring: the ring expected to report seqno + * @seqno: duh! + * @interruptible: do an interruptible wait (normally yes) + * @timeout: in - how long to wait (NULL forever); out - how much time remaining + * + * Returns 0 if the seqno was found within the alloted time. Else returns the + * errno with remaining time filled in timeout argument. + */ static int __wait_seqno(struct intel_ring_buffer *ring, u32 seqno, - bool interruptible) + bool interruptible, struct timespec *timeout) { drm_i915_private_t *dev_priv = ring->dev->dev_private; - int ret = 0; + struct timespec before, now, wait_time={1,0}; + unsigned long timeout_jiffies; + long end; + bool wait_forever = true; + int ret; if (i915_seqno_passed(ring->get_seqno(ring), seqno)) return 0; trace_i915_gem_request_wait_begin(ring, seqno); + + if (timeout != NULL) { + wait_time = *timeout; + wait_forever = false; + } + + timeout_jiffies = timespec_to_jiffies(&wait_time); + if (WARN_ON(!ring->irq_get(ring))) return -ENODEV; + /* Record current time in case interrupted by signal, or wedged * */ + getrawmonotonic(&before); + #define EXIT_COND \ (i915_seqno_passed(ring->get_seqno(ring), seqno) || \ atomic_read(&dev_priv->mm.wedged)) + do { + if (interruptible) + end = wait_event_interruptible_timeout(ring->irq_queue, + EXIT_COND, + timeout_jiffies); + else + end = wait_event_timeout(ring->irq_queue, EXIT_COND, + timeout_jiffies); - if (interruptible) - ret = wait_event_interruptible(ring->irq_queue, - EXIT_COND); - else - wait_event(ring->irq_queue, EXIT_COND); + ret = i915_gem_check_wedge(dev_priv, interruptible); + if (ret) + end = ret; + } while (end == 0 && wait_forever); + + getrawmonotonic(&now); ring->irq_put(ring); trace_i915_gem_request_wait_end(ring, seqno); #undef EXIT_COND - return ret; + if (timeout) { + struct timespec sleep_time = timespec_sub(now, before); + *timeout = timespec_sub(*timeout, sleep_time); + } + + switch (end) { + case -EIO: + case -EAGAIN: /* Wedged */ + case -ERESTARTSYS: /* Signal */ + return (int)end; + case 0: /* Timeout */ + if (timeout) + set_normalized_timespec(timeout, 0, 0); + return -ETIME; + default: /* Completed */ + WARN_ON(end < 0); /* We're not aware of other errors */ + return 0; + } } /** @@ -1934,15 +2022,14 @@ static int __wait_seqno(struct intel_ring_buffer *ring, u32 seqno, * request and object lists appropriately for that event. */ int -i915_wait_request(struct intel_ring_buffer *ring, - uint32_t seqno) +i915_wait_seqno(struct intel_ring_buffer *ring, uint32_t seqno) { drm_i915_private_t *dev_priv = ring->dev->dev_private; int ret = 0; BUG_ON(seqno == 0); - ret = i915_gem_check_wedge(dev_priv); + ret = i915_gem_check_wedge(dev_priv, dev_priv->mm.interruptible); if (ret) return ret; @@ -1950,9 +2037,7 @@ i915_wait_request(struct intel_ring_buffer *ring, if (ret) return ret; - ret = __wait_seqno(ring, seqno, dev_priv->mm.interruptible); - if (atomic_read(&dev_priv->mm.wedged)) - ret = -EAGAIN; + ret = __wait_seqno(ring, seqno, dev_priv->mm.interruptible, NULL); return ret; } @@ -1975,7 +2060,7 @@ i915_gem_object_wait_rendering(struct drm_i915_gem_object *obj) * it. */ if (obj->active) { - ret = i915_wait_request(obj->ring, obj->last_rendering_seqno); + ret = i915_wait_seqno(obj->ring, obj->last_rendering_seqno); if (ret) return ret; i915_gem_retire_requests_ring(obj->ring); @@ -1985,6 +2070,115 @@ i915_gem_object_wait_rendering(struct drm_i915_gem_object *obj) } /** + * Ensures that an object will eventually get non-busy by flushing any required + * write domains, emitting any outstanding lazy request and retiring and + * completed requests. + */ +static int +i915_gem_object_flush_active(struct drm_i915_gem_object *obj) +{ + int ret; + + if (obj->active) { + ret = i915_gem_object_flush_gpu_write_domain(obj); + if (ret) + return ret; + + ret = i915_gem_check_olr(obj->ring, + obj->last_rendering_seqno); + if (ret) + return ret; + i915_gem_retire_requests_ring(obj->ring); + } + + return 0; +} + +/** + * i915_gem_wait_ioctl - implements DRM_IOCTL_I915_GEM_WAIT + * @DRM_IOCTL_ARGS: standard ioctl arguments + * + * Returns 0 if successful, else an error is returned with the remaining time in + * the timeout parameter. + * -ETIME: object is still busy after timeout + * -ERESTARTSYS: signal interrupted the wait + * -ENONENT: object doesn't exist + * Also possible, but rare: + * -EAGAIN: GPU wedged + * -ENOMEM: damn + * -ENODEV: Internal IRQ fail + * -E?: The add request failed + * + * The wait ioctl with a timeout of 0 reimplements the busy ioctl. With any + * non-zero timeout parameter the wait ioctl will wait for the given number of + * nanoseconds on an object becoming unbusy. Since the wait itself does so + * without holding struct_mutex the object may become re-busied before this + * function completes. A similar but shorter * race condition exists in the busy + * ioctl + */ +int +i915_gem_wait_ioctl(struct drm_device *dev, void *data, struct drm_file *file) +{ + struct drm_i915_gem_wait *args = data; + struct drm_i915_gem_object *obj; + struct intel_ring_buffer *ring = NULL; + struct timespec timeout_stack, *timeout = NULL; + u32 seqno = 0; + int ret = 0; + + if (args->timeout_ns >= 0) { + timeout_stack = ns_to_timespec(args->timeout_ns); + timeout = &timeout_stack; + } + + ret = i915_mutex_lock_interruptible(dev); + if (ret) + return ret; + + obj = to_intel_bo(drm_gem_object_lookup(dev, file, args->bo_handle)); + if (&obj->base == NULL) { + mutex_unlock(&dev->struct_mutex); + return -ENOENT; + } + + /* Need to make sure the object gets inactive eventually. */ + ret = i915_gem_object_flush_active(obj); + if (ret) + goto out; + + if (obj->active) { + seqno = obj->last_rendering_seqno; + ring = obj->ring; + } + + if (seqno == 0) + goto out; + + /* Do this after OLR check to make sure we make forward progress polling + * on this IOCTL with a 0 timeout (like busy ioctl) + */ + if (!args->timeout_ns) { + ret = -ETIME; + goto out; + } + + drm_gem_object_unreference(&obj->base); + mutex_unlock(&dev->struct_mutex); + + ret = __wait_seqno(ring, seqno, true, timeout); + if (timeout) { + WARN_ON(!timespec_valid(timeout)); + args->timeout_ns = timespec_to_ns(timeout); + } + return ret; + +out: + drm_gem_object_unreference(&obj->base); + mutex_unlock(&dev->struct_mutex); + return ret; +} + +/** * i915_gem_object_sync - sync an object to a ring. * * @obj: object which may be in use on another ring. @@ -2160,7 +2354,7 @@ static int i915_ring_idle(struct intel_ring_buffer *ring) return ret; } - return i915_wait_request(ring, i915_gem_next_request_seqno(ring)); + return i915_wait_seqno(ring, i915_gem_next_request_seqno(ring)); } int i915_gpu_idle(struct drm_device *dev) @@ -2178,6 +2372,10 @@ int i915_gpu_idle(struct drm_device *dev) /* Is the device fubar? */ if (WARN_ON(!list_empty(&ring->gpu_write_list))) return -EBUSY; + + ret = i915_switch_context(ring, NULL, DEFAULT_CONTEXT_ID); + if (ret) + return ret; } return 0; @@ -2364,7 +2562,7 @@ i915_gem_object_flush_fence(struct drm_i915_gem_object *obj) } if (obj->last_fenced_seqno) { - ret = i915_wait_request(obj->ring, obj->last_fenced_seqno); + ret = i915_wait_seqno(obj->ring, obj->last_fenced_seqno); if (ret) return ret; @@ -2551,8 +2749,8 @@ i915_gem_object_bind_to_gtt(struct drm_i915_gem_object *obj, if (map_and_fenceable) free_space = drm_mm_search_free_in_range(&dev_priv->mm.gtt_space, - size, alignment, 0, - dev_priv->mm.gtt_mappable_end, + size, alignment, + 0, dev_priv->mm.gtt_mappable_end, 0); else free_space = drm_mm_search_free(&dev_priv->mm.gtt_space, @@ -2563,7 +2761,7 @@ i915_gem_object_bind_to_gtt(struct drm_i915_gem_object *obj, obj->gtt_space = drm_mm_get_block_range_generic(free_space, size, alignment, 0, - dev_priv->mm.gtt_mappable_end, + 0, dev_priv->mm.gtt_mappable_end, 0); else obj->gtt_space = @@ -3030,7 +3228,7 @@ i915_gem_ring_throttle(struct drm_device *dev, struct drm_file *file) if (seqno == 0) return 0; - ret = __wait_seqno(ring, seqno, true); + ret = __wait_seqno(ring, seqno, true, NULL); if (ret == 0) queue_delayed_work(dev_priv->wq, &dev_priv->mm.retire_work, 0); @@ -3199,30 +3397,9 @@ i915_gem_busy_ioctl(struct drm_device *dev, void *data, * become non-busy without any further actions, therefore emit any * necessary flushes here. */ - args->busy = obj->active; - if (args->busy) { - /* Unconditionally flush objects, even when the gpu still uses this - * object. Userspace calling this function indicates that it wants to - * use this buffer rather sooner than later, so issuing the required - * flush earlier is beneficial. - */ - if (obj->base.write_domain & I915_GEM_GPU_DOMAINS) { - ret = i915_gem_flush_ring(obj->ring, - 0, obj->base.write_domain); - } else { - ret = i915_gem_check_olr(obj->ring, - obj->last_rendering_seqno); - } + ret = i915_gem_object_flush_active(obj); - /* Update the active list for the hardware's current position. - * Otherwise this only updates on a delayed timer or when irqs - * are actually unmasked, and our working set ends up being - * larger than required. - */ - i915_gem_retire_requests_ring(obj->ring); - - args->busy = obj->active; - } + args->busy = obj->active; drm_gem_object_unreference(&obj->base); unlock: @@ -3435,6 +3612,38 @@ i915_gem_idle(struct drm_device *dev) return 0; } +void i915_gem_l3_remap(struct drm_device *dev) +{ + drm_i915_private_t *dev_priv = dev->dev_private; + u32 misccpctl; + int i; + + if (!IS_IVYBRIDGE(dev)) + return; + + if (!dev_priv->mm.l3_remap_info) + return; + + misccpctl = I915_READ(GEN7_MISCCPCTL); + I915_WRITE(GEN7_MISCCPCTL, misccpctl & ~GEN7_DOP_CLOCK_GATE_ENABLE); + POSTING_READ(GEN7_MISCCPCTL); + + for (i = 0; i < GEN7_L3LOG_SIZE; i += 4) { + u32 remap = I915_READ(GEN7_L3LOG_BASE + i); + if (remap && remap != dev_priv->mm.l3_remap_info[i/4]) + DRM_DEBUG("0x%x was already programmed to %x\n", + GEN7_L3LOG_BASE + i, remap); + if (remap && !dev_priv->mm.l3_remap_info[i/4]) + DRM_DEBUG_DRIVER("Clearing remapped register\n"); + I915_WRITE(GEN7_L3LOG_BASE + i, dev_priv->mm.l3_remap_info[i/4]); + } + + /* Make sure all the writes land before disabling dop clock gating */ + POSTING_READ(GEN7_L3LOG_BASE); + + I915_WRITE(GEN7_MISCCPCTL, misccpctl); +} + void i915_gem_init_swizzling(struct drm_device *dev) { drm_i915_private_t *dev_priv = dev->dev_private; @@ -3518,12 +3727,33 @@ void i915_gem_init_ppgtt(struct drm_device *dev) } } +static bool +intel_enable_blt(struct drm_device *dev) +{ + if (!HAS_BLT(dev)) + return false; + + /* The blitter was dysfunctional on early prototypes */ + if (IS_GEN6(dev) && dev->pdev->revision < 8) { + DRM_INFO("BLT not supported on this pre-production hardware;" + " graphics performance will be degraded.\n"); + return false; + } + + return true; +} + int i915_gem_init_hw(struct drm_device *dev) { drm_i915_private_t *dev_priv = dev->dev_private; int ret; + if (!intel_enable_gtt()) + return -EIO; + + i915_gem_l3_remap(dev); + i915_gem_init_swizzling(dev); ret = intel_init_render_ring_buffer(dev); @@ -3536,7 +3766,7 @@ i915_gem_init_hw(struct drm_device *dev) goto cleanup_render_ring; } - if (HAS_BLT(dev)) { + if (intel_enable_blt(dev)) { ret = intel_init_blt_ring_buffer(dev); if (ret) goto cleanup_bsd_ring; @@ -3544,6 +3774,11 @@ i915_gem_init_hw(struct drm_device *dev) dev_priv->next_seqno = 1; + /* + * XXX: There was some w/a described somewhere suggesting loading + * contexts before PPGTT. + */ + i915_gem_context_init(dev); i915_gem_init_ppgtt(dev); return 0; diff --git a/drivers/gpu/drm/i915/i915_gem_context.c b/drivers/gpu/drm/i915/i915_gem_context.c new file mode 100644 index 000000000000..da8b01fb1bf8 --- /dev/null +++ b/drivers/gpu/drm/i915/i915_gem_context.c @@ -0,0 +1,536 @@ +/* + * Copyright © 2011-2012 Intel Corporation + * + * 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 (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 NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS 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. + * + * Authors: + * Ben Widawsky <ben@bwidawsk.net> + * + */ + +/* + * This file implements HW context support. On gen5+ a HW context consists of an + * opaque GPU object which is referenced at times of context saves and restores. + * With RC6 enabled, the context is also referenced as the GPU enters and exists + * from RC6 (GPU has it's own internal power context, except on gen5). Though + * something like a context does exist for the media ring, the code only + * supports contexts for the render ring. + * + * In software, there is a distinction between contexts created by the user, + * and the default HW context. The default HW context is used by GPU clients + * that do not request setup of their own hardware context. The default + * context's state is never restored to help prevent programming errors. This + * would happen if a client ran and piggy-backed off another clients GPU state. + * The default context only exists to give the GPU some offset to load as the + * current to invoke a save of the context we actually care about. In fact, the + * code could likely be constructed, albeit in a more complicated fashion, to + * never use the default context, though that limits the driver's ability to + * swap out, and/or destroy other contexts. + * + * All other contexts are created as a request by the GPU client. These contexts + * store GPU state, and thus allow GPU clients to not re-emit state (and + * potentially query certain state) at any time. The kernel driver makes + * certain that the appropriate commands are inserted. + * + * The context life cycle is semi-complicated in that context BOs may live + * longer than the context itself because of the way the hardware, and object + * tracking works. Below is a very crude representation of the state machine + * describing the context life. + * refcount pincount active + * S0: initial state 0 0 0 + * S1: context created 1 0 0 + * S2: context is currently running 2 1 X + * S3: GPU referenced, but not current 2 0 1 + * S4: context is current, but destroyed 1 1 0 + * S5: like S3, but destroyed 1 0 1 + * + * The most common (but not all) transitions: + * S0->S1: client creates a context + * S1->S2: client submits execbuf with context + * S2->S3: other clients submits execbuf with context + * S3->S1: context object was retired + * S3->S2: clients submits another execbuf + * S2->S4: context destroy called with current context + * S3->S5->S0: destroy path + * S4->S5->S0: destroy path on current context + * + * There are two confusing terms used above: + * The "current context" means the context which is currently running on the + * GPU. The GPU has loaded it's state already and has stored away the gtt + * offset of the BO. The GPU is not actively referencing the data at this + * offset, but it will on the next context switch. The only way to avoid this + * is to do a GPU reset. + * + * An "active context' is one which was previously the "current context" and is + * on the active list waiting for the next context switch to occur. Until this + * happens, the object must remain at the same gtt offset. It is therefore + * possible to destroy a context, but it is still active. + * + */ + +#include "drmP.h" +#include "i915_drm.h" +#include "i915_drv.h" + +/* This is a HW constraint. The value below is the largest known requirement + * I've seen in a spec to date, and that was a workaround for a non-shipping + * part. It should be safe to decrease this, but it's more future proof as is. + */ +#define CONTEXT_ALIGN (64<<10) + +static struct i915_hw_context * +i915_gem_context_get(struct drm_i915_file_private *file_priv, u32 id); +static int do_switch(struct drm_i915_gem_object *from_obj, + struct i915_hw_context *to, u32 seqno); + +static int get_context_size(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + int ret; + u32 reg; + + switch (INTEL_INFO(dev)->gen) { + case 6: + reg = I915_READ(CXT_SIZE); + ret = GEN6_CXT_TOTAL_SIZE(reg) * 64; + break; + case 7: + reg = I915_READ(GEN7_CXT_SIZE); + ret = GEN7_CXT_TOTAL_SIZE(reg) * 64; + break; + default: + BUG(); + } + + return ret; +} + +static void do_destroy(struct i915_hw_context *ctx) +{ + struct drm_device *dev = ctx->obj->base.dev; + struct drm_i915_private *dev_priv = dev->dev_private; + + if (ctx->file_priv) + idr_remove(&ctx->file_priv->context_idr, ctx->id); + else + BUG_ON(ctx != dev_priv->ring[RCS].default_context); + + drm_gem_object_unreference(&ctx->obj->base); + kfree(ctx); +} + +static struct i915_hw_context * +create_hw_context(struct drm_device *dev, + struct drm_i915_file_private *file_priv) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + struct i915_hw_context *ctx; + int ret, id; + + ctx = kzalloc(sizeof(struct drm_i915_file_private), GFP_KERNEL); + if (ctx == NULL) + return ERR_PTR(-ENOMEM); + + ctx->obj = i915_gem_alloc_object(dev, dev_priv->hw_context_size); + if (ctx->obj == NULL) { + kfree(ctx); + DRM_DEBUG_DRIVER("Context object allocated failed\n"); + return ERR_PTR(-ENOMEM); + } + + /* The ring associated with the context object is handled by the normal + * object tracking code. We give an initial ring value simple to pass an + * assertion in the context switch code. + */ + ctx->ring = &dev_priv->ring[RCS]; + + /* Default context will never have a file_priv */ + if (file_priv == NULL) + return ctx; + + ctx->file_priv = file_priv; + +again: + if (idr_pre_get(&file_priv->context_idr, GFP_KERNEL) == 0) { + ret = -ENOMEM; + DRM_DEBUG_DRIVER("idr allocation failed\n"); + goto err_out; + } + + ret = idr_get_new_above(&file_priv->context_idr, ctx, + DEFAULT_CONTEXT_ID + 1, &id); + if (ret == 0) + ctx->id = id; + + if (ret == -EAGAIN) + goto again; + else if (ret) + goto err_out; + + return ctx; + +err_out: + do_destroy(ctx); + return ERR_PTR(ret); +} + +static inline bool is_default_context(struct i915_hw_context *ctx) +{ + return (ctx == ctx->ring->default_context); +} + +/** + * The default context needs to exist per ring that uses contexts. It stores the + * context state of the GPU for applications that don't utilize HW contexts, as + * well as an idle case. + */ +static int create_default_context(struct drm_i915_private *dev_priv) +{ + struct i915_hw_context *ctx; + int ret; + + BUG_ON(!mutex_is_locked(&dev_priv->dev->struct_mutex)); + + ctx = create_hw_context(dev_priv->dev, NULL); + if (IS_ERR(ctx)) + return PTR_ERR(ctx); + + /* We may need to do things with the shrinker which require us to + * immediately switch back to the default context. This can cause a + * problem as pinning the default context also requires GTT space which + * may not be available. To avoid this we always pin the + * default context. + */ + dev_priv->ring[RCS].default_context = ctx; + ret = i915_gem_object_pin(ctx->obj, CONTEXT_ALIGN, false); + if (ret) { + do_destroy(ctx); + return ret; + } + + ret = do_switch(NULL, ctx, 0); + if (ret) { + i915_gem_object_unpin(ctx->obj); + do_destroy(ctx); + } else { + DRM_DEBUG_DRIVER("Default HW context loaded\n"); + } + + return ret; +} + +void i915_gem_context_init(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + uint32_t ctx_size; + + if (!HAS_HW_CONTEXTS(dev)) { + dev_priv->hw_contexts_disabled = true; + return; + } + + /* If called from reset, or thaw... we've been here already */ + if (dev_priv->hw_contexts_disabled || + dev_priv->ring[RCS].default_context) + return; + + ctx_size = get_context_size(dev); + dev_priv->hw_context_size = get_context_size(dev); + dev_priv->hw_context_size = round_up(dev_priv->hw_context_size, 4096); + + if (ctx_size <= 0 || ctx_size > (1<<20)) { + dev_priv->hw_contexts_disabled = true; + return; + } + + if (create_default_context(dev_priv)) { + dev_priv->hw_contexts_disabled = true; + return; + } + + DRM_DEBUG_DRIVER("HW context support initialized\n"); +} + +void i915_gem_context_fini(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + + if (dev_priv->hw_contexts_disabled) + return; + + /* The only known way to stop the gpu from accessing the hw context is + * to reset it. Do this as the very last operation to avoid confusing + * other code, leading to spurious errors. */ + intel_gpu_reset(dev); + + i915_gem_object_unpin(dev_priv->ring[RCS].default_context->obj); + + do_destroy(dev_priv->ring[RCS].default_context); +} + +static int context_idr_cleanup(int id, void *p, void *data) +{ + struct i915_hw_context *ctx = p; + + BUG_ON(id == DEFAULT_CONTEXT_ID); + + do_destroy(ctx); + + return 0; +} + +void i915_gem_context_close(struct drm_device *dev, struct drm_file *file) +{ + struct drm_i915_file_private *file_priv = file->driver_priv; + + mutex_lock(&dev->struct_mutex); + idr_for_each(&file_priv->context_idr, context_idr_cleanup, NULL); + idr_destroy(&file_priv->context_idr); + mutex_unlock(&dev->struct_mutex); +} + +static struct i915_hw_context * +i915_gem_context_get(struct drm_i915_file_private *file_priv, u32 id) +{ + return (struct i915_hw_context *)idr_find(&file_priv->context_idr, id); +} + +static inline int +mi_set_context(struct intel_ring_buffer *ring, + struct i915_hw_context *new_context, + u32 hw_flags) +{ + int ret; + + /* w/a: If Flush TLB Invalidation Mode is enabled, driver must do a TLB + * invalidation prior to MI_SET_CONTEXT. On GEN6 we don't set the value + * explicitly, so we rely on the value at ring init, stored in + * itlb_before_ctx_switch. + */ + if (IS_GEN6(ring->dev) && ring->itlb_before_ctx_switch) { + ret = ring->flush(ring, 0, 0); + if (ret) + return ret; + } + + ret = intel_ring_begin(ring, 6); + if (ret) + return ret; + + if (IS_GEN7(ring->dev)) + intel_ring_emit(ring, MI_ARB_ON_OFF | MI_ARB_DISABLE); + else + intel_ring_emit(ring, MI_NOOP); + + intel_ring_emit(ring, MI_NOOP); + intel_ring_emit(ring, MI_SET_CONTEXT); + intel_ring_emit(ring, new_context->obj->gtt_offset | + MI_MM_SPACE_GTT | + MI_SAVE_EXT_STATE_EN | + MI_RESTORE_EXT_STATE_EN | + hw_flags); + /* w/a: MI_SET_CONTEXT must always be followed by MI_NOOP */ + intel_ring_emit(ring, MI_NOOP); + + if (IS_GEN7(ring->dev)) + intel_ring_emit(ring, MI_ARB_ON_OFF | MI_ARB_ENABLE); + else + intel_ring_emit(ring, MI_NOOP); + + intel_ring_advance(ring); + + return ret; +} + +static int do_switch(struct drm_i915_gem_object *from_obj, + struct i915_hw_context *to, + u32 seqno) +{ + struct intel_ring_buffer *ring = NULL; + u32 hw_flags = 0; + int ret; + + BUG_ON(to == NULL); + BUG_ON(from_obj != NULL && from_obj->pin_count == 0); + + ret = i915_gem_object_pin(to->obj, CONTEXT_ALIGN, false); + if (ret) + return ret; + + /* Clear this page out of any CPU caches for coherent swap-in/out. Note + * that thanks to write = false in this call and us not setting any gpu + * write domains when putting a context object onto the active list + * (when switching away from it), this won't block. + * XXX: We need a real interface to do this instead of trickery. */ + ret = i915_gem_object_set_to_gtt_domain(to->obj, false); + if (ret) { + i915_gem_object_unpin(to->obj); + return ret; + } + + if (!to->obj->has_global_gtt_mapping) + i915_gem_gtt_bind_object(to->obj, to->obj->cache_level); + + if (!to->is_initialized || is_default_context(to)) + hw_flags |= MI_RESTORE_INHIBIT; + else if (WARN_ON_ONCE(from_obj == to->obj)) /* not yet expected */ + hw_flags |= MI_FORCE_RESTORE; + + ring = to->ring; + ret = mi_set_context(ring, to, hw_flags); + if (ret) { + i915_gem_object_unpin(to->obj); + return ret; + } + + /* The backing object for the context is done after switching to the + * *next* context. Therefore we cannot retire the previous context until + * the next context has already started running. In fact, the below code + * is a bit suboptimal because the retiring can occur simply after the + * MI_SET_CONTEXT instead of when the next seqno has completed. + */ + if (from_obj != NULL) { + from_obj->base.read_domains = I915_GEM_DOMAIN_INSTRUCTION; + i915_gem_object_move_to_active(from_obj, ring, seqno); + /* As long as MI_SET_CONTEXT is serializing, ie. it flushes the + * whole damn pipeline, we don't need to explicitly mark the + * object dirty. The only exception is that the context must be + * correct in case the object gets swapped out. Ideally we'd be + * able to defer doing this until we know the object would be + * swapped, but there is no way to do that yet. + */ + from_obj->dirty = 1; + BUG_ON(from_obj->ring != to->ring); + i915_gem_object_unpin(from_obj); + + drm_gem_object_unreference(&from_obj->base); + } + + drm_gem_object_reference(&to->obj->base); + ring->last_context_obj = to->obj; + to->is_initialized = true; + + return 0; +} + +/** + * i915_switch_context() - perform a GPU context switch. + * @ring: ring for which we'll execute the context switch + * @file_priv: file_priv associated with the context, may be NULL + * @id: context id number + * @seqno: sequence number by which the new context will be switched to + * @flags: + * + * The context life cycle is simple. The context refcount is incremented and + * decremented by 1 and create and destroy. If the context is in use by the GPU, + * it will have a refoucnt > 1. This allows us to destroy the context abstract + * object while letting the normal object tracking destroy the backing BO. + */ +int i915_switch_context(struct intel_ring_buffer *ring, + struct drm_file *file, + int to_id) +{ + struct drm_i915_private *dev_priv = ring->dev->dev_private; + struct drm_i915_file_private *file_priv = NULL; + struct i915_hw_context *to; + struct drm_i915_gem_object *from_obj = ring->last_context_obj; + int ret; + + if (dev_priv->hw_contexts_disabled) + return 0; + + if (ring != &dev_priv->ring[RCS]) + return 0; + + if (file) + file_priv = file->driver_priv; + + if (to_id == DEFAULT_CONTEXT_ID) { + to = ring->default_context; + } else { + to = i915_gem_context_get(file_priv, to_id); + if (to == NULL) + return -ENOENT; + } + + if (from_obj == to->obj) + return 0; + + return do_switch(from_obj, to, i915_gem_next_request_seqno(to->ring)); +} + +int i915_gem_context_create_ioctl(struct drm_device *dev, void *data, + struct drm_file *file) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + struct drm_i915_gem_context_create *args = data; + struct drm_i915_file_private *file_priv = file->driver_priv; + struct i915_hw_context *ctx; + int ret; + + if (!(dev->driver->driver_features & DRIVER_GEM)) + return -ENODEV; + + if (dev_priv->hw_contexts_disabled) + return -ENODEV; + + ret = i915_mutex_lock_interruptible(dev); + if (ret) + return ret; + + ctx = create_hw_context(dev, file_priv); + mutex_unlock(&dev->struct_mutex); + if (IS_ERR(ctx)) + return PTR_ERR(ctx); + + args->ctx_id = ctx->id; + DRM_DEBUG_DRIVER("HW context %d created\n", args->ctx_id); + + return 0; +} + +int i915_gem_context_destroy_ioctl(struct drm_device *dev, void *data, + struct drm_file *file) +{ + struct drm_i915_gem_context_destroy *args = data; + struct drm_i915_file_private *file_priv = file->driver_priv; + struct i915_hw_context *ctx; + int ret; + + if (!(dev->driver->driver_features & DRIVER_GEM)) + return -ENODEV; + + ret = i915_mutex_lock_interruptible(dev); + if (ret) + return ret; + + ctx = i915_gem_context_get(file_priv, args->ctx_id); + if (!ctx) { + mutex_unlock(&dev->struct_mutex); + return -ENOENT; + } + + do_destroy(ctx); + + mutex_unlock(&dev->struct_mutex); + + DRM_DEBUG_DRIVER("HW context %d destroyed\n", args->ctx_id); + return 0; +} diff --git a/drivers/gpu/drm/i915/i915_gem_debug.c b/drivers/gpu/drm/i915/i915_gem_debug.c index a4f6aaabca99..bddf7bed183f 100644 --- a/drivers/gpu/drm/i915/i915_gem_debug.c +++ b/drivers/gpu/drm/i915/i915_gem_debug.c @@ -132,7 +132,8 @@ i915_gem_object_check_coherency(struct drm_i915_gem_object *obj, int handle) __func__, obj, obj->gtt_offset, handle, obj->size / 1024); - gtt_mapping = ioremap(dev->agp->base + obj->gtt_offset, obj->base.size); + gtt_mapping = ioremap(dev_priv->mm.gtt_base_addr + obj->gtt_offset, + obj->base.size); if (gtt_mapping == NULL) { DRM_ERROR("failed to map GTT space\n"); return; diff --git a/drivers/gpu/drm/i915/i915_gem_evict.c b/drivers/gpu/drm/i915/i915_gem_evict.c index ae7c24e12e52..eba0308f10e3 100644 --- a/drivers/gpu/drm/i915/i915_gem_evict.c +++ b/drivers/gpu/drm/i915/i915_gem_evict.c @@ -78,11 +78,12 @@ i915_gem_evict_something(struct drm_device *dev, int min_size, INIT_LIST_HEAD(&unwind_list); if (mappable) - drm_mm_init_scan_with_range(&dev_priv->mm.gtt_space, min_size, - alignment, 0, - dev_priv->mm.gtt_mappable_end); + drm_mm_init_scan_with_range(&dev_priv->mm.gtt_space, + min_size, alignment, 0, + 0, dev_priv->mm.gtt_mappable_end); else - drm_mm_init_scan(&dev_priv->mm.gtt_space, min_size, alignment); + drm_mm_init_scan(&dev_priv->mm.gtt_space, + min_size, alignment, 0); /* First see if there is a large enough contiguous idle region... */ list_for_each_entry(obj, &dev_priv->mm.inactive_list, mm_list) { diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c index 974a9f1068a3..5af631e788c8 100644 --- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c +++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c @@ -810,33 +810,16 @@ err: return ret; } -static int +static void i915_gem_execbuffer_flush(struct drm_device *dev, uint32_t invalidate_domains, - uint32_t flush_domains, - uint32_t flush_rings) + uint32_t flush_domains) { - drm_i915_private_t *dev_priv = dev->dev_private; - int i, ret; - if (flush_domains & I915_GEM_DOMAIN_CPU) intel_gtt_chipset_flush(); if (flush_domains & I915_GEM_DOMAIN_GTT) wmb(); - - if ((flush_domains | invalidate_domains) & I915_GEM_GPU_DOMAINS) { - for (i = 0; i < I915_NUM_RINGS; i++) - if (flush_rings & (1 << i)) { - ret = i915_gem_flush_ring(&dev_priv->ring[i], - invalidate_domains, - flush_domains); - if (ret) - return ret; - } - } - - return 0; } static int @@ -885,12 +868,9 @@ i915_gem_execbuffer_move_to_gpu(struct intel_ring_buffer *ring, i915_gem_object_set_to_gpu_domain(obj, ring, &cd); if (cd.invalidate_domains | cd.flush_domains) { - ret = i915_gem_execbuffer_flush(ring->dev, - cd.invalidate_domains, - cd.flush_domains, - cd.flush_rings); - if (ret) - return ret; + i915_gem_execbuffer_flush(ring->dev, + cd.invalidate_domains, + cd.flush_domains); } if (cd.flips) { @@ -905,6 +885,16 @@ i915_gem_execbuffer_move_to_gpu(struct intel_ring_buffer *ring, return ret; } + /* Unconditionally invalidate gpu caches and ensure that we do flush + * any residual writes from the previous batch. + */ + ret = i915_gem_flush_ring(ring, + I915_GEM_GPU_DOMAINS, + ring->gpu_caches_dirty ? I915_GEM_GPU_DOMAINS : 0); + if (ret) + return ret; + + ring->gpu_caches_dirty = false; return 0; } @@ -983,26 +973,13 @@ i915_gem_execbuffer_retire_commands(struct drm_device *dev, struct intel_ring_buffer *ring) { struct drm_i915_gem_request *request; - u32 invalidate; - /* - * Ensure that the commands in the batch buffer are - * finished before the interrupt fires. - * - * The sampler always gets flushed on i965 (sigh). - */ - invalidate = I915_GEM_DOMAIN_COMMAND; - if (INTEL_INFO(dev)->gen >= 4) - invalidate |= I915_GEM_DOMAIN_SAMPLER; - if (ring->flush(ring, invalidate, 0)) { - i915_gem_next_request_seqno(ring); - return; - } + /* Unconditionally force add_request to emit a full flush. */ + ring->gpu_caches_dirty = true; /* Add a breadcrumb for the completion of the batch buffer */ request = kzalloc(sizeof(*request), GFP_KERNEL); if (request == NULL || i915_add_request(ring, file, request)) { - i915_gem_next_request_seqno(ring); kfree(request); } } @@ -1044,6 +1021,7 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data, struct drm_i915_gem_object *batch_obj; struct drm_clip_rect *cliprects = NULL; struct intel_ring_buffer *ring; + u32 ctx_id = i915_execbuffer2_get_context_id(*args); u32 exec_start, exec_len; u32 seqno; u32 mask; @@ -1065,9 +1043,19 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data, break; case I915_EXEC_BSD: ring = &dev_priv->ring[VCS]; + if (ctx_id != 0) { + DRM_DEBUG("Ring %s doesn't support contexts\n", + ring->name); + return -EPERM; + } break; case I915_EXEC_BLT: ring = &dev_priv->ring[BCS]; + if (ctx_id != 0) { + DRM_DEBUG("Ring %s doesn't support contexts\n", + ring->name); + return -EPERM; + } break; default: DRM_DEBUG("execbuf with unknown ring: %d\n", @@ -1240,6 +1228,10 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data, } } + ret = i915_switch_context(ring, file, ctx_id); + if (ret) + goto err; + if (ring == &dev_priv->ring[RCS] && mode != dev_priv->relative_constants_mode) { ret = intel_ring_begin(ring, 4); @@ -1367,6 +1359,7 @@ i915_gem_execbuffer(struct drm_device *dev, void *data, exec2.num_cliprects = args->num_cliprects; exec2.cliprects_ptr = args->cliprects_ptr; exec2.flags = I915_EXEC_RENDER; + i915_execbuffer2_set_context_id(exec2, 0); ret = i915_gem_do_execbuffer(dev, data, file, &exec2, exec2_list); if (!ret) { diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c index ed3224c37423..8a3828528b9d 100644 --- a/drivers/gpu/drm/i915/i915_irq.c +++ b/drivers/gpu/drm/i915/i915_irq.c @@ -375,6 +375,86 @@ static void gen6_pm_rps_work(struct work_struct *work) mutex_unlock(&dev_priv->dev->struct_mutex); } + +/** + * ivybridge_parity_work - Workqueue called when a parity error interrupt + * occurred. + * @work: workqueue struct + * + * Doesn't actually do anything except notify userspace. As a consequence of + * this event, userspace should try to remap the bad rows since statistically + * it is likely the same row is more likely to go bad again. + */ +static void ivybridge_parity_work(struct work_struct *work) +{ + drm_i915_private_t *dev_priv = container_of(work, drm_i915_private_t, + parity_error_work); + u32 error_status, row, bank, subbank; + char *parity_event[5]; + uint32_t misccpctl; + unsigned long flags; + + /* We must turn off DOP level clock gating to access the L3 registers. + * In order to prevent a get/put style interface, acquire struct mutex + * any time we access those registers. + */ + mutex_lock(&dev_priv->dev->struct_mutex); + + misccpctl = I915_READ(GEN7_MISCCPCTL); + I915_WRITE(GEN7_MISCCPCTL, misccpctl & ~GEN7_DOP_CLOCK_GATE_ENABLE); + POSTING_READ(GEN7_MISCCPCTL); + + error_status = I915_READ(GEN7_L3CDERRST1); + row = GEN7_PARITY_ERROR_ROW(error_status); + bank = GEN7_PARITY_ERROR_BANK(error_status); + subbank = GEN7_PARITY_ERROR_SUBBANK(error_status); + + I915_WRITE(GEN7_L3CDERRST1, GEN7_PARITY_ERROR_VALID | + GEN7_L3CDERRST1_ENABLE); + POSTING_READ(GEN7_L3CDERRST1); + + I915_WRITE(GEN7_MISCCPCTL, misccpctl); + + spin_lock_irqsave(&dev_priv->irq_lock, flags); + dev_priv->gt_irq_mask &= ~GT_GEN7_L3_PARITY_ERROR_INTERRUPT; + I915_WRITE(GTIMR, dev_priv->gt_irq_mask); + spin_unlock_irqrestore(&dev_priv->irq_lock, flags); + + mutex_unlock(&dev_priv->dev->struct_mutex); + + parity_event[0] = "L3_PARITY_ERROR=1"; + parity_event[1] = kasprintf(GFP_KERNEL, "ROW=%d", row); + parity_event[2] = kasprintf(GFP_KERNEL, "BANK=%d", bank); + parity_event[3] = kasprintf(GFP_KERNEL, "SUBBANK=%d", subbank); + parity_event[4] = NULL; + + kobject_uevent_env(&dev_priv->dev->primary->kdev.kobj, + KOBJ_CHANGE, parity_event); + + DRM_DEBUG("Parity error: Row = %d, Bank = %d, Sub bank = %d.\n", + row, bank, subbank); + + kfree(parity_event[3]); + kfree(parity_event[2]); + kfree(parity_event[1]); +} + +static void ivybridge_handle_parity_error(struct drm_device *dev) +{ + drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private; + unsigned long flags; + + if (!IS_IVYBRIDGE(dev)) + return; + + spin_lock_irqsave(&dev_priv->irq_lock, flags); + dev_priv->gt_irq_mask |= GT_GEN7_L3_PARITY_ERROR_INTERRUPT; + I915_WRITE(GTIMR, dev_priv->gt_irq_mask); + spin_unlock_irqrestore(&dev_priv->irq_lock, flags); + + queue_work(dev_priv->wq, &dev_priv->parity_error_work); +} + static void snb_gt_irq_handler(struct drm_device *dev, struct drm_i915_private *dev_priv, u32 gt_iir) @@ -394,6 +474,9 @@ static void snb_gt_irq_handler(struct drm_device *dev, DRM_ERROR("GT error interrupt 0x%08x\n", gt_iir); i915_handle_error(dev, false); } + + if (gt_iir & GT_GEN7_L3_PARITY_ERROR_INTERRUPT) + ivybridge_handle_parity_error(dev); } static void gen6_queue_rps_work(struct drm_i915_private *dev_priv, @@ -429,15 +512,10 @@ static irqreturn_t valleyview_irq_handler(DRM_IRQ_ARGS) unsigned long irqflags; int pipe; u32 pipe_stats[I915_MAX_PIPES]; - u32 vblank_status; - int vblank = 0; bool blc_event; atomic_inc(&dev_priv->irq_received); - vblank_status = PIPE_START_VBLANK_INTERRUPT_STATUS | - PIPE_VBLANK_INTERRUPT_STATUS; - while (true) { iir = I915_READ(VLV_IIR); gt_iir = I915_READ(GTIIR); @@ -467,6 +545,16 @@ static irqreturn_t valleyview_irq_handler(DRM_IRQ_ARGS) } spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags); + for_each_pipe(pipe) { + if (pipe_stats[pipe] & PIPE_VBLANK_INTERRUPT_STATUS) + drm_handle_vblank(dev, pipe); + + if (pipe_stats[pipe] & PLANE_FLIPDONE_INT_STATUS_VLV) { + intel_prepare_page_flip(dev, pipe); + intel_finish_page_flip(dev, pipe); + } + } + /* Consume port. Then clear IIR or we'll miss events */ if (iir & I915_DISPLAY_PORT_INTERRUPT) { u32 hotplug_status = I915_READ(PORT_HOTPLUG_STAT); @@ -481,19 +569,6 @@ static irqreturn_t valleyview_irq_handler(DRM_IRQ_ARGS) I915_READ(PORT_HOTPLUG_STAT); } - - if (iir & I915_DISPLAY_PIPE_A_VBLANK_INTERRUPT) { - drm_handle_vblank(dev, 0); - vblank++; - intel_finish_page_flip(dev, 0); - } - - if (iir & I915_DISPLAY_PIPE_B_VBLANK_INTERRUPT) { - drm_handle_vblank(dev, 1); - vblank++; - intel_finish_page_flip(dev, 0); - } - if (pipe_stats[pipe] & PIPE_LEGACY_BLC_EVENT_STATUS) blc_event = true; @@ -991,6 +1066,7 @@ static void i915_record_ring_state(struct drm_device *dev, struct drm_i915_private *dev_priv = dev->dev_private; if (INTEL_INFO(dev)->gen >= 6) { + error->rc_psmi[ring->id] = I915_READ(ring->mmio_base + 0x50); error->fault_reg[ring->id] = I915_READ(RING_FAULT_REG(ring)); error->semaphore_mboxes[ring->id][0] = I915_READ(RING_SYNC_0(ring->mmio_base)); @@ -1104,6 +1180,7 @@ static void i915_capture_error_state(struct drm_device *dev) kref_init(&error->ref); error->eir = I915_READ(EIR); error->pgtbl_er = I915_READ(PGTBL_ER); + error->ccid = I915_READ(CCID); if (HAS_PCH_SPLIT(dev)) error->ier = I915_READ(DEIER) | I915_READ(GTIER); @@ -1426,23 +1503,20 @@ static int valleyview_enable_vblank(struct drm_device *dev, int pipe) { drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private; unsigned long irqflags; - u32 dpfl, imr; + u32 imr; if (!i915_pipe_enabled(dev, pipe)) return -EINVAL; spin_lock_irqsave(&dev_priv->irq_lock, irqflags); - dpfl = I915_READ(VLV_DPFLIPSTAT); imr = I915_READ(VLV_IMR); - if (pipe == 0) { - dpfl |= PIPEA_VBLANK_INT_EN; + if (pipe == 0) imr &= ~I915_DISPLAY_PIPE_A_VBLANK_INTERRUPT; - } else { - dpfl |= PIPEA_VBLANK_INT_EN; + else imr &= ~I915_DISPLAY_PIPE_B_VBLANK_INTERRUPT; - } - I915_WRITE(VLV_DPFLIPSTAT, dpfl); I915_WRITE(VLV_IMR, imr); + i915_enable_pipestat(dev_priv, pipe, + PIPE_START_VBLANK_INTERRUPT_ENABLE); spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags); return 0; @@ -1492,20 +1566,17 @@ static void valleyview_disable_vblank(struct drm_device *dev, int pipe) { drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private; unsigned long irqflags; - u32 dpfl, imr; + u32 imr; spin_lock_irqsave(&dev_priv->irq_lock, irqflags); - dpfl = I915_READ(VLV_DPFLIPSTAT); + i915_disable_pipestat(dev_priv, pipe, + PIPE_START_VBLANK_INTERRUPT_ENABLE); imr = I915_READ(VLV_IMR); - if (pipe == 0) { - dpfl &= ~PIPEA_VBLANK_INT_EN; + if (pipe == 0) imr |= I915_DISPLAY_PIPE_A_VBLANK_INTERRUPT; - } else { - dpfl &= ~PIPEB_VBLANK_INT_EN; + else imr |= I915_DISPLAY_PIPE_B_VBLANK_INTERRUPT; - } I915_WRITE(VLV_IMR, imr); - I915_WRITE(VLV_DPFLIPSTAT, dpfl); spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags); } @@ -1648,7 +1719,6 @@ static void ironlake_irq_preinstall(struct drm_device *dev) atomic_set(&dev_priv->irq_received, 0); - I915_WRITE(HWSTAM, 0xeffe); /* XXX hotplug from PCH */ @@ -1811,13 +1881,13 @@ static int ivybridge_irq_postinstall(struct drm_device *dev) DE_PIPEA_VBLANK_IVB); POSTING_READ(DEIER); - dev_priv->gt_irq_mask = ~0; + dev_priv->gt_irq_mask = ~GT_GEN7_L3_PARITY_ERROR_INTERRUPT; I915_WRITE(GTIIR, I915_READ(GTIIR)); I915_WRITE(GTIMR, dev_priv->gt_irq_mask); render_irqs = GT_USER_INTERRUPT | GEN6_BSD_USER_INTERRUPT | - GEN6_BLITTER_USER_INTERRUPT; + GEN6_BLITTER_USER_INTERRUPT | GT_GEN7_L3_PARITY_ERROR_INTERRUPT; I915_WRITE(GTIER, render_irqs); POSTING_READ(GTIER); @@ -1840,16 +1910,24 @@ static int ivybridge_irq_postinstall(struct drm_device *dev) static int valleyview_irq_postinstall(struct drm_device *dev) { drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private; - u32 render_irqs; u32 enable_mask; u32 hotplug_en = I915_READ(PORT_HOTPLUG_EN); + u32 pipestat_enable = PLANE_FLIP_DONE_INT_EN_VLV; u16 msid; enable_mask = I915_DISPLAY_PORT_INTERRUPT; - enable_mask |= I915_DISPLAY_PIPE_A_VBLANK_INTERRUPT | + enable_mask |= I915_DISPLAY_PIPE_A_EVENT_INTERRUPT | + I915_DISPLAY_PIPE_A_VBLANK_INTERRUPT | + I915_DISPLAY_PIPE_B_EVENT_INTERRUPT | I915_DISPLAY_PIPE_B_VBLANK_INTERRUPT; - dev_priv->irq_mask = ~enable_mask; + /* + *Leave vblank interrupts masked initially. enable/disable will + * toggle them based on usage. + */ + dev_priv->irq_mask = (~enable_mask) | + I915_DISPLAY_PIPE_A_VBLANK_INTERRUPT | + I915_DISPLAY_PIPE_B_VBLANK_INTERRUPT; dev_priv->pipestat[0] = 0; dev_priv->pipestat[1] = 0; @@ -1868,26 +1946,27 @@ static int valleyview_irq_postinstall(struct drm_device *dev) I915_WRITE(PIPESTAT(1), 0xffff); POSTING_READ(VLV_IER); + i915_enable_pipestat(dev_priv, 0, pipestat_enable); + i915_enable_pipestat(dev_priv, 1, pipestat_enable); + I915_WRITE(VLV_IIR, 0xffffffff); I915_WRITE(VLV_IIR, 0xffffffff); - render_irqs = GT_GEN6_BLT_FLUSHDW_NOTIFY_INTERRUPT | - GT_GEN6_BLT_CS_ERROR_INTERRUPT | - GT_GEN6_BLT_USER_INTERRUPT | - GT_GEN6_BSD_USER_INTERRUPT | - GT_GEN6_BSD_CS_ERROR_INTERRUPT | - GT_GEN7_L3_PARITY_ERROR_INTERRUPT | - GT_PIPE_NOTIFY | - GT_RENDER_CS_ERROR_INTERRUPT | - GT_SYNC_STATUS | - GT_USER_INTERRUPT; - - dev_priv->gt_irq_mask = ~render_irqs; + dev_priv->gt_irq_mask = ~0; I915_WRITE(GTIIR, I915_READ(GTIIR)); I915_WRITE(GTIIR, I915_READ(GTIIR)); - I915_WRITE(GTIMR, 0); - I915_WRITE(GTIER, render_irqs); + I915_WRITE(GTIMR, dev_priv->gt_irq_mask); + I915_WRITE(GTIER, GT_GEN6_BLT_FLUSHDW_NOTIFY_INTERRUPT | + GT_GEN6_BLT_CS_ERROR_INTERRUPT | + GT_GEN6_BLT_USER_INTERRUPT | + GT_GEN6_BSD_USER_INTERRUPT | + GT_GEN6_BSD_CS_ERROR_INTERRUPT | + GT_GEN7_L3_PARITY_ERROR_INTERRUPT | + GT_PIPE_NOTIFY | + GT_RENDER_CS_ERROR_INTERRUPT | + GT_SYNC_STATUS | + GT_USER_INTERRUPT); POSTING_READ(GTIER); /* ack & enable invalid PTE error interrupts */ @@ -2166,9 +2245,9 @@ static int i915_irq_postinstall(struct drm_device *dev) hotplug_en |= HDMIC_HOTPLUG_INT_EN; if (dev_priv->hotplug_supported_mask & HDMID_HOTPLUG_INT_STATUS) hotplug_en |= HDMID_HOTPLUG_INT_EN; - if (dev_priv->hotplug_supported_mask & SDVOC_HOTPLUG_INT_STATUS) + if (dev_priv->hotplug_supported_mask & SDVOC_HOTPLUG_INT_STATUS_I915) hotplug_en |= SDVOC_HOTPLUG_INT_EN; - if (dev_priv->hotplug_supported_mask & SDVOB_HOTPLUG_INT_STATUS) + if (dev_priv->hotplug_supported_mask & SDVOB_HOTPLUG_INT_STATUS_I915) hotplug_en |= SDVOB_HOTPLUG_INT_EN; if (dev_priv->hotplug_supported_mask & CRT_HOTPLUG_INT_STATUS) { hotplug_en |= CRT_HOTPLUG_INT_EN; @@ -2328,10 +2407,8 @@ static void i965_irq_preinstall(struct drm_device * dev) atomic_set(&dev_priv->irq_received, 0); - if (I915_HAS_HOTPLUG(dev)) { - I915_WRITE(PORT_HOTPLUG_EN, 0); - I915_WRITE(PORT_HOTPLUG_STAT, I915_READ(PORT_HOTPLUG_STAT)); - } + I915_WRITE(PORT_HOTPLUG_EN, 0); + I915_WRITE(PORT_HOTPLUG_STAT, I915_READ(PORT_HOTPLUG_STAT)); I915_WRITE(HWSTAM, 0xeffe); for_each_pipe(pipe) @@ -2344,11 +2421,13 @@ static void i965_irq_preinstall(struct drm_device * dev) static int i965_irq_postinstall(struct drm_device *dev) { drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private; + u32 hotplug_en; u32 enable_mask; u32 error_mask; /* Unmask the interrupts that we always want on. */ dev_priv->irq_mask = ~(I915_ASLE_INTERRUPT | + I915_DISPLAY_PORT_INTERRUPT | I915_DISPLAY_PIPE_A_EVENT_INTERRUPT | I915_DISPLAY_PIPE_B_EVENT_INTERRUPT | I915_DISPLAY_PLANE_A_FLIP_PENDING_INTERRUPT | @@ -2364,13 +2443,6 @@ static int i965_irq_postinstall(struct drm_device *dev) dev_priv->pipestat[0] = 0; dev_priv->pipestat[1] = 0; - if (I915_HAS_HOTPLUG(dev)) { - /* Enable in IER... */ - enable_mask |= I915_DISPLAY_PORT_INTERRUPT; - /* and unmask in IMR */ - dev_priv->irq_mask &= ~I915_DISPLAY_PORT_INTERRUPT; - } - /* * Enable some error detection, note the instruction error mask * bit is reserved, so we leave it masked. @@ -2390,36 +2462,40 @@ static int i965_irq_postinstall(struct drm_device *dev) I915_WRITE(IER, enable_mask); POSTING_READ(IER); - if (I915_HAS_HOTPLUG(dev)) { - u32 hotplug_en = I915_READ(PORT_HOTPLUG_EN); - - /* Note HDMI and DP share bits */ - if (dev_priv->hotplug_supported_mask & HDMIB_HOTPLUG_INT_STATUS) - hotplug_en |= HDMIB_HOTPLUG_INT_EN; - if (dev_priv->hotplug_supported_mask & HDMIC_HOTPLUG_INT_STATUS) - hotplug_en |= HDMIC_HOTPLUG_INT_EN; - if (dev_priv->hotplug_supported_mask & HDMID_HOTPLUG_INT_STATUS) - hotplug_en |= HDMID_HOTPLUG_INT_EN; - if (dev_priv->hotplug_supported_mask & SDVOC_HOTPLUG_INT_STATUS) + /* Note HDMI and DP share hotplug bits */ + hotplug_en = 0; + if (dev_priv->hotplug_supported_mask & HDMIB_HOTPLUG_INT_STATUS) + hotplug_en |= HDMIB_HOTPLUG_INT_EN; + if (dev_priv->hotplug_supported_mask & HDMIC_HOTPLUG_INT_STATUS) + hotplug_en |= HDMIC_HOTPLUG_INT_EN; + if (dev_priv->hotplug_supported_mask & HDMID_HOTPLUG_INT_STATUS) + hotplug_en |= HDMID_HOTPLUG_INT_EN; + if (IS_G4X(dev)) { + if (dev_priv->hotplug_supported_mask & SDVOC_HOTPLUG_INT_STATUS_G4X) hotplug_en |= SDVOC_HOTPLUG_INT_EN; - if (dev_priv->hotplug_supported_mask & SDVOB_HOTPLUG_INT_STATUS) + if (dev_priv->hotplug_supported_mask & SDVOB_HOTPLUG_INT_STATUS_G4X) hotplug_en |= SDVOB_HOTPLUG_INT_EN; - if (dev_priv->hotplug_supported_mask & CRT_HOTPLUG_INT_STATUS) { - hotplug_en |= CRT_HOTPLUG_INT_EN; + } else { + if (dev_priv->hotplug_supported_mask & SDVOC_HOTPLUG_INT_STATUS_I965) + hotplug_en |= SDVOC_HOTPLUG_INT_EN; + if (dev_priv->hotplug_supported_mask & SDVOB_HOTPLUG_INT_STATUS_I965) + hotplug_en |= SDVOB_HOTPLUG_INT_EN; + } + if (dev_priv->hotplug_supported_mask & CRT_HOTPLUG_INT_STATUS) { + hotplug_en |= CRT_HOTPLUG_INT_EN; - /* Programming the CRT detection parameters tends - to generate a spurious hotplug event about three - seconds later. So just do it once. - */ - if (IS_G4X(dev)) - hotplug_en |= CRT_HOTPLUG_ACTIVATION_PERIOD_64; - hotplug_en |= CRT_HOTPLUG_VOLTAGE_COMPARE_50; - } + /* Programming the CRT detection parameters tends + to generate a spurious hotplug event about three + seconds later. So just do it once. + */ + if (IS_G4X(dev)) + hotplug_en |= CRT_HOTPLUG_ACTIVATION_PERIOD_64; + hotplug_en |= CRT_HOTPLUG_VOLTAGE_COMPARE_50; + } - /* Ignore TV since it's buggy */ + /* Ignore TV since it's buggy */ - I915_WRITE(PORT_HOTPLUG_EN, hotplug_en); - } + I915_WRITE(PORT_HOTPLUG_EN, hotplug_en); intel_opregion_enable_asle(dev); @@ -2477,8 +2553,7 @@ static irqreturn_t i965_irq_handler(DRM_IRQ_ARGS) ret = IRQ_HANDLED; /* Consume port. Then clear IIR or we'll miss events */ - if ((I915_HAS_HOTPLUG(dev)) && - (iir & I915_DISPLAY_PORT_INTERRUPT)) { + if (iir & I915_DISPLAY_PORT_INTERRUPT) { u32 hotplug_status = I915_READ(PORT_HOTPLUG_STAT); DRM_DEBUG_DRIVER("hotplug event received, stat 0x%08x\n", @@ -2551,10 +2626,8 @@ static void i965_irq_uninstall(struct drm_device * dev) if (!dev_priv) return; - if (I915_HAS_HOTPLUG(dev)) { - I915_WRITE(PORT_HOTPLUG_EN, 0); - I915_WRITE(PORT_HOTPLUG_STAT, I915_READ(PORT_HOTPLUG_STAT)); - } + I915_WRITE(PORT_HOTPLUG_EN, 0); + I915_WRITE(PORT_HOTPLUG_STAT, I915_READ(PORT_HOTPLUG_STAT)); I915_WRITE(HWSTAM, 0xffffffff); for_each_pipe(pipe) @@ -2575,6 +2648,7 @@ void intel_irq_init(struct drm_device *dev) INIT_WORK(&dev_priv->hotplug_work, i915_hotplug_work_func); INIT_WORK(&dev_priv->error_work, i915_error_work_func); INIT_WORK(&dev_priv->rps_work, gen6_pm_rps_work); + INIT_WORK(&dev_priv->parity_error_work, ivybridge_parity_work); dev->driver->get_vblank_counter = i915_get_vblank_counter; dev->max_vblank_count = 0xffffff; /* only 24 bits of frame count */ diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index 48d5e8e051cf..acc99b21e0b6 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -217,6 +217,9 @@ #define MI_DISPLAY_FLIP_IVB_SPRITE_B (3 << 19) #define MI_DISPLAY_FLIP_IVB_PLANE_C (4 << 19) #define MI_DISPLAY_FLIP_IVB_SPRITE_C (5 << 19) +#define MI_ARB_ON_OFF MI_INSTR(0x08, 0) +#define MI_ARB_ENABLE (1<<0) +#define MI_ARB_DISABLE (0<<0) #define MI_SET_CONTEXT MI_INSTR(0x18, 0) #define MI_MM_SPACE_GTT (1<<8) @@ -299,6 +302,7 @@ #define DISPLAY_PLANE_B (1<<20) #define GFX_OP_PIPE_CONTROL(len) ((0x3<<29)|(0x3<<27)|(0x2<<24)|(len-2)) #define PIPE_CONTROL_CS_STALL (1<<20) +#define PIPE_CONTROL_TLB_INVALIDATE (1<<18) #define PIPE_CONTROL_QW_WRITE (1<<14) #define PIPE_CONTROL_DEPTH_STALL (1<<13) #define PIPE_CONTROL_WRITE_FLUSH (1<<12) @@ -686,10 +690,10 @@ #define GEN6_BLITTER_FBC_NOTIFY (1<<3) #define GEN6_BSD_SLEEP_PSMI_CONTROL 0x12050 -#define GEN6_BSD_SLEEP_PSMI_CONTROL_RC_ILDL_MESSAGE_MODIFY_MASK (1 << 16) -#define GEN6_BSD_SLEEP_PSMI_CONTROL_RC_ILDL_MESSAGE_DISABLE (1 << 0) -#define GEN6_BSD_SLEEP_PSMI_CONTROL_RC_ILDL_MESSAGE_ENABLE 0 -#define GEN6_BSD_SLEEP_PSMI_CONTROL_IDLE_INDICATOR (1 << 3) +#define GEN6_BSD_SLEEP_MSG_DISABLE (1 << 0) +#define GEN6_BSD_SLEEP_FLUSH_DISABLE (1 << 2) +#define GEN6_BSD_SLEEP_INDICATOR (1 << 3) +#define GEN6_BSD_GO_INDICATOR (1 << 4) #define GEN6_BSD_HWSTAM 0x12098 #define GEN6_BSD_IMR 0x120a8 @@ -908,6 +912,7 @@ #define DPLL_P2_CLOCK_DIV_MASK 0x03000000 /* i915 */ #define DPLL_FPA01_P1_POST_DIV_MASK 0x00ff0000 /* i915 */ #define DPLL_FPA01_P1_POST_DIV_MASK_PINEVIEW 0x00ff8000 /* Pineview */ +#define DPLL_LOCK_VLV (1<<15) #define DPLL_INTEGRATED_CLOCK_VLV (1<<13) #define SRX_INDEX 0x3c4 @@ -1453,6 +1458,10 @@ #define DDRMPLL1 0X12c20 #define PEG_BAND_GAP_DATA 0x14d68 +#define GEN6_GT_THREAD_STATUS_REG 0x13805c +#define GEN6_GT_THREAD_STATUS_CORE_MASK 0x7 +#define GEN6_GT_THREAD_STATUS_CORE_MASK_HSW (0x7 | (0x07 << 16)) + #define GEN6_GT_PERF_STATUS 0x145948 #define GEN6_RP_STATE_LIMITS 0x145994 #define GEN6_RP_STATE_CAP 0x145998 @@ -1462,6 +1471,31 @@ */ #define CCID 0x2180 #define CCID_EN (1<<0) +#define CXT_SIZE 0x21a0 +#define GEN6_CXT_POWER_SIZE(cxt_reg) ((cxt_reg >> 24) & 0x3f) +#define GEN6_CXT_RING_SIZE(cxt_reg) ((cxt_reg >> 18) & 0x3f) +#define GEN6_CXT_RENDER_SIZE(cxt_reg) ((cxt_reg >> 12) & 0x3f) +#define GEN6_CXT_EXTENDED_SIZE(cxt_reg) ((cxt_reg >> 6) & 0x3f) +#define GEN6_CXT_PIPELINE_SIZE(cxt_reg) ((cxt_reg >> 0) & 0x3f) +#define GEN6_CXT_TOTAL_SIZE(cxt_reg) (GEN6_CXT_POWER_SIZE(cxt_reg) + \ + GEN6_CXT_RING_SIZE(cxt_reg) + \ + GEN6_CXT_RENDER_SIZE(cxt_reg) + \ + GEN6_CXT_EXTENDED_SIZE(cxt_reg) + \ + GEN6_CXT_PIPELINE_SIZE(cxt_reg)) +#define GEN7_CXT_SIZE 0x21a8 +#define GEN7_CXT_POWER_SIZE(ctx_reg) ((ctx_reg >> 25) & 0x7f) +#define GEN7_CXT_RING_SIZE(ctx_reg) ((ctx_reg >> 22) & 0x7) +#define GEN7_CXT_RENDER_SIZE(ctx_reg) ((ctx_reg >> 16) & 0x3f) +#define GEN7_CXT_EXTENDED_SIZE(ctx_reg) ((ctx_reg >> 9) & 0x7f) +#define GEN7_CXT_GT1_SIZE(ctx_reg) ((ctx_reg >> 6) & 0x7) +#define GEN7_CXT_VFSTATE_SIZE(ctx_reg) ((ctx_reg >> 0) & 0x3f) +#define GEN7_CXT_TOTAL_SIZE(ctx_reg) (GEN7_CXT_POWER_SIZE(ctx_reg) + \ + GEN7_CXT_RING_SIZE(ctx_reg) + \ + GEN7_CXT_RENDER_SIZE(ctx_reg) + \ + GEN7_CXT_EXTENDED_SIZE(ctx_reg) + \ + GEN7_CXT_GT1_SIZE(ctx_reg) + \ + GEN7_CXT_VFSTATE_SIZE(ctx_reg)) + /* * Overlay regs */ @@ -1566,20 +1600,34 @@ #define CRT_HOTPLUG_DETECT_VOLTAGE_475MV (1 << 2) #define PORT_HOTPLUG_STAT 0x61114 -#define HDMIB_HOTPLUG_INT_STATUS (1 << 29) -#define DPB_HOTPLUG_INT_STATUS (1 << 29) -#define HDMIC_HOTPLUG_INT_STATUS (1 << 28) -#define DPC_HOTPLUG_INT_STATUS (1 << 28) -#define HDMID_HOTPLUG_INT_STATUS (1 << 27) -#define DPD_HOTPLUG_INT_STATUS (1 << 27) +/* HDMI/DP bits are gen4+ */ +#define DPB_HOTPLUG_LIVE_STATUS (1 << 29) +#define DPC_HOTPLUG_LIVE_STATUS (1 << 28) +#define DPD_HOTPLUG_LIVE_STATUS (1 << 27) +#define DPD_HOTPLUG_INT_STATUS (3 << 21) +#define DPC_HOTPLUG_INT_STATUS (3 << 19) +#define DPB_HOTPLUG_INT_STATUS (3 << 17) +/* HDMI bits are shared with the DP bits */ +#define HDMIB_HOTPLUG_LIVE_STATUS (1 << 29) +#define HDMIC_HOTPLUG_LIVE_STATUS (1 << 28) +#define HDMID_HOTPLUG_LIVE_STATUS (1 << 27) +#define HDMID_HOTPLUG_INT_STATUS (3 << 21) +#define HDMIC_HOTPLUG_INT_STATUS (3 << 19) +#define HDMIB_HOTPLUG_INT_STATUS (3 << 17) +/* CRT/TV common between gen3+ */ #define CRT_HOTPLUG_INT_STATUS (1 << 11) #define TV_HOTPLUG_INT_STATUS (1 << 10) #define CRT_HOTPLUG_MONITOR_MASK (3 << 8) #define CRT_HOTPLUG_MONITOR_COLOR (3 << 8) #define CRT_HOTPLUG_MONITOR_MONO (2 << 8) #define CRT_HOTPLUG_MONITOR_NONE (0 << 8) -#define SDVOC_HOTPLUG_INT_STATUS (1 << 7) -#define SDVOB_HOTPLUG_INT_STATUS (1 << 6) +/* SDVO is different across gen3/4 */ +#define SDVOC_HOTPLUG_INT_STATUS_G4X (1 << 3) +#define SDVOB_HOTPLUG_INT_STATUS_G4X (1 << 2) +#define SDVOC_HOTPLUG_INT_STATUS_I965 (3 << 4) +#define SDVOB_HOTPLUG_INT_STATUS_I965 (3 << 2) +#define SDVOC_HOTPLUG_INT_STATUS_I915 (1 << 7) +#define SDVOB_HOTPLUG_INT_STATUS_I915 (1 << 6) /* SDVO port control */ #define SDVOB 0x61140 @@ -1711,8 +1759,10 @@ #define VIDEO_DIP_PORT_C (2 << 29) #define VIDEO_DIP_PORT_D (3 << 29) #define VIDEO_DIP_PORT_MASK (3 << 29) +#define VIDEO_DIP_ENABLE_GCP (1 << 25) #define VIDEO_DIP_ENABLE_AVI (1 << 21) #define VIDEO_DIP_ENABLE_VENDOR (2 << 21) +#define VIDEO_DIP_ENABLE_GAMUT (4 << 21) #define VIDEO_DIP_ENABLE_SPD (8 << 21) #define VIDEO_DIP_SELECT_AVI (0 << 19) #define VIDEO_DIP_SELECT_VENDOR (1 << 19) @@ -1723,7 +1773,11 @@ #define VIDEO_DIP_FREQ_2VSYNC (2 << 16) #define VIDEO_DIP_FREQ_MASK (3 << 16) /* HSW and later: */ +#define VIDEO_DIP_ENABLE_VSC_HSW (1 << 20) +#define VIDEO_DIP_ENABLE_GCP_HSW (1 << 16) #define VIDEO_DIP_ENABLE_AVI_HSW (1 << 12) +#define VIDEO_DIP_ENABLE_VS_HSW (1 << 8) +#define VIDEO_DIP_ENABLE_GMP_HSW (1 << 4) #define VIDEO_DIP_ENABLE_SPD_HSW (1 << 0) /* Panel power sequencing */ @@ -1795,18 +1849,35 @@ #define PFIT_AUTO_RATIOS 0x61238 /* Backlight control */ -#define BLC_PWM_CTL 0x61254 -#define BACKLIGHT_MODULATION_FREQ_SHIFT (17) #define BLC_PWM_CTL2 0x61250 /* 965+ only */ -#define BLM_COMBINATION_MODE (1 << 30) +#define BLM_PWM_ENABLE (1 << 31) +#define BLM_COMBINATION_MODE (1 << 30) /* gen4 only */ +#define BLM_PIPE_SELECT (1 << 29) +#define BLM_PIPE_SELECT_IVB (3 << 29) +#define BLM_PIPE_A (0 << 29) +#define BLM_PIPE_B (1 << 29) +#define BLM_PIPE_C (2 << 29) /* ivb + */ +#define BLM_PIPE(pipe) ((pipe) << 29) +#define BLM_POLARITY_I965 (1 << 28) /* gen4 only */ +#define BLM_PHASE_IN_INTERUPT_STATUS (1 << 26) +#define BLM_PHASE_IN_ENABLE (1 << 25) +#define BLM_PHASE_IN_INTERUPT_ENABL (1 << 24) +#define BLM_PHASE_IN_TIME_BASE_SHIFT (16) +#define BLM_PHASE_IN_TIME_BASE_MASK (0xff << 16) +#define BLM_PHASE_IN_COUNT_SHIFT (8) +#define BLM_PHASE_IN_COUNT_MASK (0xff << 8) +#define BLM_PHASE_IN_INCR_SHIFT (0) +#define BLM_PHASE_IN_INCR_MASK (0xff << 0) +#define BLC_PWM_CTL 0x61254 /* * This is the most significant 15 bits of the number of backlight cycles in a * complete cycle of the modulated backlight control. * * The actual value is this field multiplied by two. */ -#define BACKLIGHT_MODULATION_FREQ_MASK (0x7fff << 17) -#define BLM_LEGACY_MODE (1 << 16) +#define BACKLIGHT_MODULATION_FREQ_SHIFT (17) +#define BACKLIGHT_MODULATION_FREQ_MASK (0x7fff << 17) +#define BLM_LEGACY_MODE (1 << 16) /* gen2 only */ /* * This is the number of cycles out of the backlight modulation cycle for which * the backlight is on. @@ -1816,9 +1887,24 @@ */ #define BACKLIGHT_DUTY_CYCLE_SHIFT (0) #define BACKLIGHT_DUTY_CYCLE_MASK (0xffff) +#define BACKLIGHT_DUTY_CYCLE_MASK_PNV (0xfffe) +#define BLM_POLARITY_PNV (1 << 0) /* pnv only */ #define BLC_HIST_CTL 0x61260 +/* New registers for PCH-split platforms. Safe where new bits show up, the + * register layout machtes with gen4 BLC_PWM_CTL[12]. */ +#define BLC_PWM_CPU_CTL2 0x48250 +#define BLC_PWM_CPU_CTL 0x48254 + +/* PCH CTL1 is totally different, all but the below bits are reserved. CTL2 is + * like the normal CTL from gen4 and earlier. Hooray for confusing naming. */ +#define BLC_PWM_PCH_CTL1 0xc8250 +#define BLM_PCH_PWM_ENABLE (1 << 31) +#define BLM_PCH_OVERRIDE_ENABLE (1 << 30) +#define BLM_PCH_POLARITY (1 << 29) +#define BLC_PWM_PCH_CTL2 0xc8254 + /* TV port control */ #define TV_CTL 0x68000 /** Enables the TV encoder */ @@ -2583,13 +2669,13 @@ #define PIPESTAT(pipe) _PIPE(pipe, _PIPEASTAT, _PIPEBSTAT) #define VLV_DPFLIPSTAT 0x70028 -#define PIPEB_LINE_COMPARE_STATUS (1<<29) +#define PIPEB_LINE_COMPARE_INT_EN (1<<29) #define PIPEB_HLINE_INT_EN (1<<28) #define PIPEB_VBLANK_INT_EN (1<<27) #define SPRITED_FLIPDONE_INT_EN (1<<26) #define SPRITEC_FLIPDONE_INT_EN (1<<25) #define PLANEB_FLIPDONE_INT_EN (1<<24) -#define PIPEA_LINE_COMPARE_STATUS (1<<21) +#define PIPEA_LINE_COMPARE_INT_EN (1<<21) #define PIPEA_HLINE_INT_EN (1<<20) #define PIPEA_VBLANK_INT_EN (1<<19) #define SPRITEB_FLIPDONE_INT_EN (1<<18) @@ -2897,13 +2983,14 @@ #define DSPSIZE(plane) _PIPE(plane, _DSPASIZE, _DSPBSIZE) #define DSPSURF(plane) _PIPE(plane, _DSPASURF, _DSPBSURF) #define DSPTILEOFF(plane) _PIPE(plane, _DSPATILEOFF, _DSPBTILEOFF) +#define DSPLINOFF(plane) DSPADDR(plane) /* Display/Sprite base address macros */ #define DISP_BASEADDR_MASK (0xfffff000) #define I915_LO_DISPBASE(val) (val & ~DISP_BASEADDR_MASK) #define I915_HI_DISPBASE(val) (val & DISP_BASEADDR_MASK) #define I915_MODIFY_DISPBASE(reg, gfx_addr) \ - (I915_WRITE(reg, gfx_addr | I915_LO_DISPBASE(I915_READ(reg)))) + (I915_WRITE((reg), (gfx_addr) | I915_LO_DISPBASE(I915_READ(reg)))) /* VBIOS flags */ #define SWF00 0x71410 @@ -3771,6 +3858,9 @@ #define _FDI_RXA_TUSIZE2 0xf0038 #define _FDI_RXB_TUSIZE1 0xf1030 #define _FDI_RXB_TUSIZE2 0xf1038 +#define FDI_RX_TP1_TO_TP2_48 (2<<20) +#define FDI_RX_TP1_TO_TP2_64 (3<<20) +#define FDI_RX_FDI_DELAY_90 (0x90<<0) #define FDI_RX_MISC(pipe) _PIPE(pipe, _FDI_RXA_MISC, _FDI_RXB_MISC) #define FDI_RX_TUSIZE1(pipe) _PIPE(pipe, _FDI_RXA_TUSIZE1, _FDI_RXB_TUSIZE1) #define FDI_RX_TUSIZE2(pipe) _PIPE(pipe, _FDI_RXA_TUSIZE2, _FDI_RXB_TUSIZE2) @@ -3824,7 +3914,6 @@ #define ADPA_CRT_HOTPLUG_FORCE_TRIGGER (1<<16) /* or SDVOB */ -#define VLV_HDMIB 0x61140 #define HDMIB 0xe1140 #define PORT_ENABLE (1 << 31) #define TRANSCODER(pipe) ((pipe) << 30) @@ -3855,20 +3944,18 @@ #define PCH_LVDS 0xe1180 #define LVDS_DETECTED (1 << 1) -#define BLC_PWM_CPU_CTL2 0x48250 -#define PWM_ENABLE (1 << 31) -#define PWM_PIPE_A (0 << 29) -#define PWM_PIPE_B (1 << 29) -#define BLC_PWM_CPU_CTL 0x48254 +/* vlv has 2 sets of panel control regs. */ +#define PIPEA_PP_STATUS 0x61200 +#define PIPEA_PP_CONTROL 0x61204 +#define PIPEA_PP_ON_DELAYS 0x61208 +#define PIPEA_PP_OFF_DELAYS 0x6120c +#define PIPEA_PP_DIVISOR 0x61210 -#define BLC_PWM_PCH_CTL1 0xc8250 -#define PWM_PCH_ENABLE (1 << 31) -#define PWM_POLARITY_ACTIVE_LOW (1 << 29) -#define PWM_POLARITY_ACTIVE_HIGH (0 << 29) -#define PWM_POLARITY_ACTIVE_LOW2 (1 << 28) -#define PWM_POLARITY_ACTIVE_HIGH2 (0 << 28) - -#define BLC_PWM_PCH_CTL2 0xc8254 +#define PIPEB_PP_STATUS 0x61300 +#define PIPEB_PP_CONTROL 0x61304 +#define PIPEB_PP_ON_DELAYS 0x61308 +#define PIPEB_PP_OFF_DELAYS 0x6130c +#define PIPEB_PP_DIVISOR 0x61310 #define PCH_PP_STATUS 0xc7200 #define PCH_PP_CONTROL 0xc7204 @@ -3992,6 +4079,7 @@ #define FORCEWAKE 0xA18C #define FORCEWAKE_VLV 0x1300b0 #define FORCEWAKE_ACK_VLV 0x1300b4 +#define FORCEWAKE_ACK_HSW 0x130044 #define FORCEWAKE_ACK 0x130090 #define FORCEWAKE_MT 0xa188 /* multi-threaded */ #define FORCEWAKE_MT_ACK 0x130040 @@ -4012,10 +4100,15 @@ # define GEN6_CSUNIT_CLOCK_GATE_DISABLE (1 << 7) #define GEN6_UCGCTL2 0x9404 +# define GEN7_VDSUNIT_CLOCK_GATE_DISABLE (1 << 30) +# define GEN7_TDLUNIT_CLOCK_GATE_DISABLE (1 << 22) # define GEN6_RCZUNIT_CLOCK_GATE_DISABLE (1 << 13) # define GEN6_RCPBUNIT_CLOCK_GATE_DISABLE (1 << 12) # define GEN6_RCCUNIT_CLOCK_GATE_DISABLE (1 << 11) +#define GEN7_UCGCTL4 0x940c +#define GEN7_L3BANK2X_CLOCK_GATE_DISABLE (1<<25) + #define GEN6_RPNSWREQ 0xA008 #define GEN6_TURBO_DISABLE (1<<31) #define GEN6_FREQUENCY(x) ((x)<<25) @@ -4047,6 +4140,7 @@ #define GEN6_RP_UP_IDLE_MIN (0x1<<3) #define GEN6_RP_UP_BUSY_AVG (0x2<<3) #define GEN6_RP_UP_BUSY_CONT (0x4<<3) +#define GEN7_RP_DOWN_IDLE_AVG (0x2<<0) #define GEN6_RP_DOWN_IDLE_CONT (0x1<<0) #define GEN6_RP_UP_THRESHOLD 0xA02C #define GEN6_RP_DOWN_THRESHOLD 0xA030 @@ -4111,6 +4205,26 @@ #define GEN6_RC6 3 #define GEN6_RC7 4 +#define GEN7_MISCCPCTL (0x9424) +#define GEN7_DOP_CLOCK_GATE_ENABLE (1<<0) + +/* IVYBRIDGE DPF */ +#define GEN7_L3CDERRST1 0xB008 /* L3CD Error Status 1 */ +#define GEN7_L3CDERRST1_ROW_MASK (0x7ff<<14) +#define GEN7_PARITY_ERROR_VALID (1<<13) +#define GEN7_L3CDERRST1_BANK_MASK (3<<11) +#define GEN7_L3CDERRST1_SUBBANK_MASK (7<<8) +#define GEN7_PARITY_ERROR_ROW(reg) \ + ((reg & GEN7_L3CDERRST1_ROW_MASK) >> 14) +#define GEN7_PARITY_ERROR_BANK(reg) \ + ((reg & GEN7_L3CDERRST1_BANK_MASK) >> 11) +#define GEN7_PARITY_ERROR_SUBBANK(reg) \ + ((reg & GEN7_L3CDERRST1_SUBBANK_MASK) >> 8) +#define GEN7_L3CDERRST1_ENABLE (1<<7) + +#define GEN7_L3LOG_BASE 0xB070 +#define GEN7_L3LOG_SIZE 0x80 + #define G4X_AUD_VID_DID 0x62020 #define INTEL_AUDIO_DEVCL 0x808629FB #define INTEL_AUDIO_DEVBLC 0x80862801 @@ -4177,7 +4291,7 @@ PIPE_DDI_FUNC_CTL_B) #define PIPE_DDI_FUNC_ENABLE (1<<31) /* Those bits are ignored by pipe EDP since it can only connect to DDI A */ -#define PIPE_DDI_PORT_MASK (0xf<<28) +#define PIPE_DDI_PORT_MASK (7<<28) #define PIPE_DDI_SELECT_PORT(x) ((x)<<28) #define PIPE_DDI_MODE_SELECT_HDMI (0<<24) #define PIPE_DDI_MODE_SELECT_DVI (1<<24) @@ -4335,7 +4449,7 @@ #define PIPE_WM_LINETIME_B 0x45274 #define PIPE_WM_LINETIME(pipe) _PIPE(pipe, \ PIPE_WM_LINETIME_A, \ - PIPE_WM_LINETIME_A) + PIPE_WM_LINETIME_B) #define PIPE_WM_LINETIME_MASK (0x1ff) #define PIPE_WM_LINETIME_TIME(x) ((x)) #define PIPE_WM_LINETIME_IPS_LINETIME_MASK (0x1ff<<16) @@ -4347,4 +4461,9 @@ #define SFUSE_STRAP_DDIC_DETECTED (1<<1) #define SFUSE_STRAP_DDID_DETECTED (1<<0) +#define WM_DBG 0x45280 +#define WM_DBG_DISALLOW_MULTIPLE_LP (1<<0) +#define WM_DBG_DISALLOW_MAXFIFO (1<<1) +#define WM_DBG_DISALLOW_SPRITE (1<<2) + #endif /* _I915_REG_H_ */ diff --git a/drivers/gpu/drm/i915/i915_suspend.c b/drivers/gpu/drm/i915/i915_suspend.c index a748e5cabe14..4776ccf1b3cd 100644 --- a/drivers/gpu/drm/i915/i915_suspend.c +++ b/drivers/gpu/drm/i915/i915_suspend.c @@ -828,10 +828,7 @@ int i915_save_state(struct drm_device *dev) dev_priv->saveIMR = I915_READ(IMR); } - if (IS_IRONLAKE_M(dev)) - ironlake_disable_drps(dev); - if (INTEL_INFO(dev)->gen >= 6) - gen6_disable_rps(dev); + intel_disable_gt_powersave(dev); /* Cache mode state */ dev_priv->saveCACHE_MODE_0 = I915_READ(CACHE_MODE_0); diff --git a/drivers/gpu/drm/i915/i915_sysfs.c b/drivers/gpu/drm/i915/i915_sysfs.c index 79f83445afa0..2f5388af8df9 100644 --- a/drivers/gpu/drm/i915/i915_sysfs.c +++ b/drivers/gpu/drm/i915/i915_sysfs.c @@ -29,6 +29,7 @@ #include <linux/module.h> #include <linux/stat.h> #include <linux/sysfs.h> +#include "intel_drv.h" #include "i915_drv.h" static u32 calc_residency(struct drm_device *dev, const u32 reg) @@ -92,20 +93,134 @@ static struct attribute_group rc6_attr_group = { .attrs = rc6_attrs }; -void i915_setup_sysfs(struct drm_device *dev) +static int l3_access_valid(struct drm_device *dev, loff_t offset) +{ + if (!IS_IVYBRIDGE(dev)) + return -EPERM; + + if (offset % 4 != 0) + return -EINVAL; + + if (offset >= GEN7_L3LOG_SIZE) + return -ENXIO; + + return 0; +} + +static ssize_t +i915_l3_read(struct file *filp, struct kobject *kobj, + struct bin_attribute *attr, char *buf, + loff_t offset, size_t count) +{ + struct device *dev = container_of(kobj, struct device, kobj); + struct drm_minor *dminor = container_of(dev, struct drm_minor, kdev); + struct drm_device *drm_dev = dminor->dev; + struct drm_i915_private *dev_priv = drm_dev->dev_private; + uint32_t misccpctl; + int i, ret; + + ret = l3_access_valid(drm_dev, offset); + if (ret) + return ret; + + ret = i915_mutex_lock_interruptible(drm_dev); + if (ret) + return ret; + + misccpctl = I915_READ(GEN7_MISCCPCTL); + I915_WRITE(GEN7_MISCCPCTL, misccpctl & ~GEN7_DOP_CLOCK_GATE_ENABLE); + + for (i = offset; count >= 4 && i < GEN7_L3LOG_SIZE; i += 4, count -= 4) + *((uint32_t *)(&buf[i])) = I915_READ(GEN7_L3LOG_BASE + i); + + I915_WRITE(GEN7_MISCCPCTL, misccpctl); + + mutex_unlock(&drm_dev->struct_mutex); + + return i - offset; +} + +static ssize_t +i915_l3_write(struct file *filp, struct kobject *kobj, + struct bin_attribute *attr, char *buf, + loff_t offset, size_t count) { + struct device *dev = container_of(kobj, struct device, kobj); + struct drm_minor *dminor = container_of(dev, struct drm_minor, kdev); + struct drm_device *drm_dev = dminor->dev; + struct drm_i915_private *dev_priv = drm_dev->dev_private; + u32 *temp = NULL; /* Just here to make handling failures easy */ int ret; - /* ILK doesn't have any residency information */ - if (INTEL_INFO(dev)->gen < 6) - return; + ret = l3_access_valid(drm_dev, offset); + if (ret) + return ret; - ret = sysfs_merge_group(&dev->primary->kdev.kobj, &rc6_attr_group); + ret = i915_mutex_lock_interruptible(drm_dev); if (ret) - DRM_ERROR("sysfs setup failed\n"); + return ret; + + if (!dev_priv->mm.l3_remap_info) { + temp = kzalloc(GEN7_L3LOG_SIZE, GFP_KERNEL); + if (!temp) { + mutex_unlock(&drm_dev->struct_mutex); + return -ENOMEM; + } + } + + ret = i915_gpu_idle(drm_dev); + if (ret) { + kfree(temp); + mutex_unlock(&drm_dev->struct_mutex); + return ret; + } + + /* TODO: Ideally we really want a GPU reset here to make sure errors + * aren't propagated. Since I cannot find a stable way to reset the GPU + * at this point it is left as a TODO. + */ + if (temp) + dev_priv->mm.l3_remap_info = temp; + + memcpy(dev_priv->mm.l3_remap_info + (offset/4), + buf + (offset/4), + count); + + i915_gem_l3_remap(drm_dev); + + mutex_unlock(&drm_dev->struct_mutex); + + return count; +} + +static struct bin_attribute dpf_attrs = { + .attr = {.name = "l3_parity", .mode = (S_IRUSR | S_IWUSR)}, + .size = GEN7_L3LOG_SIZE, + .read = i915_l3_read, + .write = i915_l3_write, + .mmap = NULL +}; + +void i915_setup_sysfs(struct drm_device *dev) +{ + int ret; + + if (INTEL_INFO(dev)->gen >= 6) { + ret = sysfs_merge_group(&dev->primary->kdev.kobj, + &rc6_attr_group); + if (ret) + DRM_ERROR("RC6 residency sysfs setup failed\n"); + } + + if (IS_IVYBRIDGE(dev)) { + ret = device_create_bin_file(&dev->primary->kdev, &dpf_attrs); + if (ret) + DRM_ERROR("l3 parity sysfs setup failed\n"); + } } void i915_teardown_sysfs(struct drm_device *dev) { + device_remove_bin_file(&dev->primary->kdev, &dpf_attrs); sysfs_unmerge_group(&dev->primary->kdev.kobj, &rc6_attr_group); } diff --git a/drivers/gpu/drm/i915/i915_trace.h b/drivers/gpu/drm/i915/i915_trace.h index dac7bba4d9da..fe90b3a84a6d 100644 --- a/drivers/gpu/drm/i915/i915_trace.h +++ b/drivers/gpu/drm/i915/i915_trace.h @@ -311,9 +311,33 @@ DEFINE_EVENT(i915_gem_request, i915_gem_request_retire, TP_ARGS(ring, seqno) ); -DEFINE_EVENT(i915_gem_request, i915_gem_request_wait_begin, +TRACE_EVENT(i915_gem_request_wait_begin, TP_PROTO(struct intel_ring_buffer *ring, u32 seqno), - TP_ARGS(ring, seqno) + TP_ARGS(ring, seqno), + + TP_STRUCT__entry( + __field(u32, dev) + __field(u32, ring) + __field(u32, seqno) + __field(bool, blocking) + ), + + /* NB: the blocking information is racy since mutex_is_locked + * doesn't check that the current thread holds the lock. The only + * other option would be to pass the boolean information of whether + * or not the class was blocking down through the stack which is + * less desirable. + */ + TP_fast_assign( + __entry->dev = ring->dev->primary->index; + __entry->ring = ring->id; + __entry->seqno = seqno; + __entry->blocking = mutex_is_locked(&ring->dev->struct_mutex); + ), + + TP_printk("dev=%u, ring=%u, seqno=%u, blocking=%s", + __entry->dev, __entry->ring, __entry->seqno, + __entry->blocking ? "yes (NB)" : "no") ); DEFINE_EVENT(i915_gem_request, i915_gem_request_wait_end, diff --git a/drivers/gpu/drm/i915/intel_bios.c b/drivers/gpu/drm/i915/intel_bios.c index 353459362f6f..8c6074154bf6 100644 --- a/drivers/gpu/drm/i915/intel_bios.c +++ b/drivers/gpu/drm/i915/intel_bios.c @@ -692,7 +692,7 @@ static const struct dmi_system_id intel_no_opregion_vbt[] = { * * Returns 0 on success, nonzero on failure. */ -bool +int intel_parse_bios(struct drm_device *dev) { struct drm_i915_private *dev_priv = dev->dev_private; diff --git a/drivers/gpu/drm/i915/intel_bios.h b/drivers/gpu/drm/i915/intel_bios.h index dbda6e3bdf07..31c2107e7825 100644 --- a/drivers/gpu/drm/i915/intel_bios.h +++ b/drivers/gpu/drm/i915/intel_bios.h @@ -476,7 +476,7 @@ struct bdb_edp { } __attribute__ ((packed)); void intel_setup_bios(struct drm_device *dev); -bool intel_parse_bios(struct drm_device *dev); +int intel_parse_bios(struct drm_device *dev); /* * Driver<->VBIOS interaction occurs through scratch bits in diff --git a/drivers/gpu/drm/i915/intel_crt.c b/drivers/gpu/drm/i915/intel_crt.c index 75a70c46ef1b..7ed4a41c3965 100644 --- a/drivers/gpu/drm/i915/intel_crt.c +++ b/drivers/gpu/drm/i915/intel_crt.c @@ -88,6 +88,9 @@ static void gmch_crt_dpms(struct drm_encoder *encoder, int mode) temp &= ~(ADPA_HSYNC_CNTL_DISABLE | ADPA_VSYNC_CNTL_DISABLE); temp &= ~ADPA_DAC_ENABLE; + if (IS_VALLEYVIEW(dev) && mode != DRM_MODE_DPMS_ON) + mode = DRM_MODE_DPMS_OFF; + switch (mode) { case DRM_MODE_DPMS_ON: temp |= ADPA_DAC_ENABLE; @@ -129,7 +132,7 @@ static int intel_crt_mode_valid(struct drm_connector *connector, } static bool intel_crt_mode_fixup(struct drm_encoder *encoder, - struct drm_display_mode *mode, + const struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode) { return true; @@ -230,6 +233,42 @@ static bool intel_ironlake_crt_detect_hotplug(struct drm_connector *connector) return ret; } +static bool valleyview_crt_detect_hotplug(struct drm_connector *connector) +{ + struct drm_device *dev = connector->dev; + struct drm_i915_private *dev_priv = dev->dev_private; + u32 adpa; + bool ret; + u32 save_adpa; + + save_adpa = adpa = I915_READ(ADPA); + DRM_DEBUG_KMS("trigger hotplug detect cycle: adpa=0x%x\n", adpa); + + adpa |= ADPA_CRT_HOTPLUG_FORCE_TRIGGER; + + I915_WRITE(ADPA, adpa); + + if (wait_for((I915_READ(ADPA) & ADPA_CRT_HOTPLUG_FORCE_TRIGGER) == 0, + 1000)) { + DRM_DEBUG_KMS("timed out waiting for FORCE_TRIGGER"); + I915_WRITE(ADPA, save_adpa); + } + + /* Check the status to see if both blue and green are on now */ + adpa = I915_READ(ADPA); + if ((adpa & ADPA_CRT_HOTPLUG_MONITOR_MASK) != 0) + ret = true; + else + ret = false; + + DRM_DEBUG_KMS("valleyview hotplug adpa=0x%x, result %d\n", adpa, ret); + + /* FIXME: debug force function and remove */ + ret = true; + + return ret; +} + /** * Uses CRT_HOTPLUG_EN and CRT_HOTPLUG_STAT to detect CRT presence. * @@ -249,6 +288,9 @@ static bool intel_crt_detect_hotplug(struct drm_connector *connector) if (HAS_PCH_SPLIT(dev)) return intel_ironlake_crt_detect_hotplug(connector); + if (IS_VALLEYVIEW(dev)) + return valleyview_crt_detect_hotplug(connector); + /* * On 4 series desktop, CRT detect sequence need to be done twice * to get a reliable result. @@ -288,39 +330,34 @@ static bool intel_crt_detect_ddc(struct drm_connector *connector) { struct intel_crt *crt = intel_attached_crt(connector); struct drm_i915_private *dev_priv = crt->base.base.dev->dev_private; + struct edid *edid; + struct i2c_adapter *i2c; - /* CRT should always be at 0, but check anyway */ - if (crt->base.type != INTEL_OUTPUT_ANALOG) - return false; + BUG_ON(crt->base.type != INTEL_OUTPUT_ANALOG); + + i2c = intel_gmbus_get_adapter(dev_priv, dev_priv->crt_ddc_pin); + edid = drm_get_edid(connector, i2c); - if (intel_ddc_probe(&crt->base, dev_priv->crt_ddc_pin)) { - struct edid *edid; - bool is_digital = false; - struct i2c_adapter *i2c; + if (edid) { + bool is_digital = edid->input & DRM_EDID_INPUT_DIGITAL; - i2c = intel_gmbus_get_adapter(dev_priv, dev_priv->crt_ddc_pin); - edid = drm_get_edid(connector, i2c); /* * This may be a DVI-I connector with a shared DDC * link between analog and digital outputs, so we * have to check the EDID input spec of the attached device. - * - * On the other hand, what should we do if it is a broken EDID? */ - if (edid != NULL) { - is_digital = edid->input & DRM_EDID_INPUT_DIGITAL; - connector->display_info.raw_edid = NULL; - kfree(edid); - } - if (!is_digital) { DRM_DEBUG_KMS("CRT detected via DDC:0x50 [EDID]\n"); return true; - } else { - DRM_DEBUG_KMS("CRT not detected via DDC:0x50 [EDID reports a digital panel]\n"); } + + DRM_DEBUG_KMS("CRT not detected via DDC:0x50 [EDID reports a digital panel]\n"); + } else { + DRM_DEBUG_KMS("CRT not detected via DDC:0x50 [no valid EDID found]\n"); } + kfree(edid); + return false; } @@ -453,18 +490,27 @@ intel_crt_detect(struct drm_connector *connector, bool force) struct intel_load_detect_pipe tmp; if (I915_HAS_HOTPLUG(dev)) { + /* We can not rely on the HPD pin always being correctly wired + * up, for example many KVM do not pass it through, and so + * only trust an assertion that the monitor is connected. + */ if (intel_crt_detect_hotplug(connector)) { DRM_DEBUG_KMS("CRT detected via hotplug\n"); return connector_status_connected; - } else { + } else DRM_DEBUG_KMS("CRT not detected via hotplug\n"); - return connector_status_disconnected; - } } if (intel_crt_detect_ddc(connector)) return connector_status_connected; + /* Load detection is broken on HPD capable machines. Whoever wants a + * broken monitor (without edid) to work behind a broken kvm (that fails + * to have the right resistors for HP detection) needs to fix this up. + * For now just bail out. */ + if (I915_HAS_HOTPLUG(dev)) + return connector_status_disconnected; + if (!force) return connector->status; diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c index 46d1e886c692..933c74859172 100644 --- a/drivers/gpu/drm/i915/intel_ddi.c +++ b/drivers/gpu/drm/i915/intel_ddi.c @@ -170,6 +170,15 @@ void hsw_fdi_link_train(struct drm_crtc *crtc) udelay(600); + /* We need to program FDI_RX_MISC with the default TP1 to TP2 + * values before enabling the receiver, and configure the delay + * for the FDI timing generator to 90h. Luckily, all the other + * bits are supposed to be zeroed, so we can write those values + * directly. + */ + I915_WRITE(FDI_RX_MISC(pipe), FDI_RX_TP1_TO_TP2_48 | + FDI_RX_FDI_DELAY_90); + /* Enable CPU FDI Receiver with auto-training */ reg = FDI_RX_CTL(pipe); I915_WRITE(reg, @@ -726,8 +735,7 @@ void intel_ddi_mode_set(struct drm_encoder *encoder, I915_WRITE(DDI_FUNC_CTL(pipe), temp); - intel_hdmi_set_avi_infoframe(encoder, adjusted_mode); - intel_hdmi_set_spd_infoframe(encoder); + intel_hdmi->set_infoframes(encoder, adjusted_mode); } void intel_ddi_dpms(struct drm_encoder *encoder, int mode) diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index a8538ac0299d..f6159765f1eb 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -98,6 +98,11 @@ intel_find_pll_ironlake_dp(const intel_limit_t *, struct drm_crtc *crtc, int target, int refclk, intel_clock_t *match_clock, intel_clock_t *best_clock); +static bool +intel_vlv_find_best_pll(const intel_limit_t *limit, struct drm_crtc *crtc, + int target, int refclk, intel_clock_t *match_clock, + intel_clock_t *best_clock); + static inline u32 /* units of 100MHz */ intel_fdi_link_freq(struct drm_device *dev) { @@ -359,6 +364,48 @@ static const intel_limit_t intel_limits_ironlake_display_port = { .find_pll = intel_find_pll_ironlake_dp, }; +static const intel_limit_t intel_limits_vlv_dac = { + .dot = { .min = 25000, .max = 270000 }, + .vco = { .min = 4000000, .max = 6000000 }, + .n = { .min = 1, .max = 7 }, + .m = { .min = 22, .max = 450 }, /* guess */ + .m1 = { .min = 2, .max = 3 }, + .m2 = { .min = 11, .max = 156 }, + .p = { .min = 10, .max = 30 }, + .p1 = { .min = 2, .max = 3 }, + .p2 = { .dot_limit = 270000, + .p2_slow = 2, .p2_fast = 20 }, + .find_pll = intel_vlv_find_best_pll, +}; + +static const intel_limit_t intel_limits_vlv_hdmi = { + .dot = { .min = 20000, .max = 165000 }, + .vco = { .min = 5994000, .max = 4000000 }, + .n = { .min = 1, .max = 7 }, + .m = { .min = 60, .max = 300 }, /* guess */ + .m1 = { .min = 2, .max = 3 }, + .m2 = { .min = 11, .max = 156 }, + .p = { .min = 10, .max = 30 }, + .p1 = { .min = 2, .max = 3 }, + .p2 = { .dot_limit = 270000, + .p2_slow = 2, .p2_fast = 20 }, + .find_pll = intel_vlv_find_best_pll, +}; + +static const intel_limit_t intel_limits_vlv_dp = { + .dot = { .min = 162000, .max = 270000 }, + .vco = { .min = 5994000, .max = 4000000 }, + .n = { .min = 1, .max = 7 }, + .m = { .min = 60, .max = 300 }, /* guess */ + .m1 = { .min = 2, .max = 3 }, + .m2 = { .min = 11, .max = 156 }, + .p = { .min = 10, .max = 30 }, + .p1 = { .min = 2, .max = 3 }, + .p2 = { .dot_limit = 270000, + .p2_slow = 2, .p2_fast = 20 }, + .find_pll = intel_vlv_find_best_pll, +}; + u32 intel_dpio_read(struct drm_i915_private *dev_priv, int reg) { unsigned long flags; @@ -384,6 +431,28 @@ out_unlock: return val; } +static void intel_dpio_write(struct drm_i915_private *dev_priv, int reg, + u32 val) +{ + unsigned long flags; + + spin_lock_irqsave(&dev_priv->dpio_lock, flags); + if (wait_for_atomic_us((I915_READ(DPIO_PKT) & DPIO_BUSY) == 0, 100)) { + DRM_ERROR("DPIO idle wait timed out\n"); + goto out_unlock; + } + + I915_WRITE(DPIO_DATA, val); + I915_WRITE(DPIO_REG, reg); + I915_WRITE(DPIO_PKT, DPIO_RID | DPIO_OP_WRITE | DPIO_PORTID | + DPIO_BYTE); + if (wait_for_atomic_us((I915_READ(DPIO_PKT) & DPIO_BUSY) == 0, 100)) + DRM_ERROR("DPIO write wait timed out\n"); + +out_unlock: + spin_unlock_irqrestore(&dev_priv->dpio_lock, flags); +} + static void vlv_init_dpio(struct drm_device *dev) { struct drm_i915_private *dev_priv = dev->dev_private; @@ -434,7 +503,7 @@ static bool is_dual_link_lvds(struct drm_i915_private *dev_priv, * register is uninitialized. */ val = I915_READ(reg); - if (!(val & ~LVDS_DETECTED)) + if (!(val & ~(LVDS_PIPE_MASK | LVDS_DETECTED))) val = dev_priv->bios_lvds_val; dev_priv->lvds_val = val; } @@ -510,6 +579,13 @@ static const intel_limit_t *intel_limit(struct drm_crtc *crtc, int refclk) limit = &intel_limits_pineview_lvds; else limit = &intel_limits_pineview_sdvo; + } else if (IS_VALLEYVIEW(dev)) { + if (intel_pipe_has_type(crtc, INTEL_OUTPUT_ANALOG)) + limit = &intel_limits_vlv_dac; + else if (intel_pipe_has_type(crtc, INTEL_OUTPUT_HDMI)) + limit = &intel_limits_vlv_hdmi; + else + limit = &intel_limits_vlv_dp; } else if (!IS_GEN2(dev)) { if (intel_pipe_has_type(crtc, INTEL_OUTPUT_LVDS)) limit = &intel_limits_i9xx_lvds; @@ -551,11 +627,10 @@ static void intel_clock(struct drm_device *dev, int refclk, intel_clock_t *clock bool intel_pipe_has_type(struct drm_crtc *crtc, int type) { struct drm_device *dev = crtc->dev; - struct drm_mode_config *mode_config = &dev->mode_config; struct intel_encoder *encoder; - list_for_each_entry(encoder, &mode_config->encoder_list, base.head) - if (encoder->base.crtc == crtc && encoder->type == type) + for_each_encoder_on_crtc(dev, crtc, encoder) + if (encoder->type == type) return true; return false; @@ -783,6 +858,73 @@ intel_find_pll_g4x_dp(const intel_limit_t *limit, struct drm_crtc *crtc, memcpy(best_clock, &clock, sizeof(intel_clock_t)); return true; } +static bool +intel_vlv_find_best_pll(const intel_limit_t *limit, struct drm_crtc *crtc, + int target, int refclk, intel_clock_t *match_clock, + intel_clock_t *best_clock) +{ + u32 p1, p2, m1, m2, vco, bestn, bestm1, bestm2, bestp1, bestp2; + u32 m, n, fastclk; + u32 updrate, minupdate, fracbits, p; + unsigned long bestppm, ppm, absppm; + int dotclk, flag; + + dotclk = target * 1000; + bestppm = 1000000; + ppm = absppm = 0; + fastclk = dotclk / (2*100); + updrate = 0; + minupdate = 19200; + fracbits = 1; + n = p = p1 = p2 = m = m1 = m2 = vco = bestn = 0; + bestm1 = bestm2 = bestp1 = bestp2 = 0; + + /* based on hardware requirement, prefer smaller n to precision */ + for (n = limit->n.min; n <= ((refclk) / minupdate); n++) { + updrate = refclk / n; + for (p1 = limit->p1.max; p1 > limit->p1.min; p1--) { + for (p2 = limit->p2.p2_fast+1; p2 > 0; p2--) { + if (p2 > 10) + p2 = p2 - 1; + p = p1 * p2; + /* based on hardware requirement, prefer bigger m1,m2 values */ + for (m1 = limit->m1.min; m1 <= limit->m1.max; m1++) { + m2 = (((2*(fastclk * p * n / m1 )) + + refclk) / (2*refclk)); + m = m1 * m2; + vco = updrate * m; + if (vco >= limit->vco.min && vco < limit->vco.max) { + ppm = 1000000 * ((vco / p) - fastclk) / fastclk; + absppm = (ppm > 0) ? ppm : (-ppm); + if (absppm < 100 && ((p1 * p2) > (bestp1 * bestp2))) { + bestppm = 0; + flag = 1; + } + if (absppm < bestppm - 10) { + bestppm = absppm; + flag = 1; + } + if (flag) { + bestn = n; + bestm1 = m1; + bestm2 = m2; + bestp1 = p1; + bestp2 = p2; + flag = 0; + } + } + } + } + } + } + best_clock->n = bestn; + best_clock->m1 = bestm1; + best_clock->m2 = bestm2; + best_clock->p1 = bestp1; + best_clock->p2 = bestp2; + + return true; +} static void ironlake_wait_for_vblank(struct drm_device *dev, int pipe) { @@ -1232,6 +1374,9 @@ static void assert_pch_dp_disabled(struct drm_i915_private *dev_priv, WARN(dp_pipe_enabled(dev_priv, pipe, port_sel, val), "PCH DP (0x%08x) enabled on transcoder %c, should be disabled\n", reg, pipe_name(pipe)); + + WARN(HAS_PCH_IBX(dev_priv->dev) && (val & SDVO_PIPE_B_SELECT), + "IBX PCH dp port still using transcoder B\n"); } static void assert_pch_hdmi_disabled(struct drm_i915_private *dev_priv, @@ -1241,6 +1386,9 @@ static void assert_pch_hdmi_disabled(struct drm_i915_private *dev_priv, WARN(hdmi_pipe_enabled(dev_priv, val, pipe), "PCH HDMI (0x%08x) enabled on transcoder %c, should be disabled\n", reg, pipe_name(pipe)); + + WARN(HAS_PCH_IBX(dev_priv->dev) && (val & SDVO_PIPE_B_SELECT), + "IBX PCH hdmi port still using transcoder B\n"); } static void assert_pch_ports_disabled(struct drm_i915_private *dev_priv, @@ -1287,7 +1435,7 @@ static void intel_enable_pll(struct drm_i915_private *dev_priv, enum pipe pipe) u32 val; /* No really, not for ILK+ */ - BUG_ON(dev_priv->info->gen >= 5); + BUG_ON(!IS_VALLEYVIEW(dev_priv->dev) && dev_priv->info->gen >= 5); /* PLL is protected by panel, make sure we can write it */ if (IS_MOBILE(dev_priv->dev) && !IS_I830(dev_priv->dev)) @@ -1344,7 +1492,7 @@ intel_sbi_write(struct drm_i915_private *dev_priv, u16 reg, u32 value) unsigned long flags; spin_lock_irqsave(&dev_priv->dpio_lock, flags); - if (wait_for((I915_READ(SBI_CTL_STAT) & SBI_READY) == 0, + if (wait_for((I915_READ(SBI_CTL_STAT) & SBI_BUSY) == 0, 100)) { DRM_ERROR("timeout waiting for SBI to become ready\n"); goto out_unlock; @@ -1358,7 +1506,7 @@ intel_sbi_write(struct drm_i915_private *dev_priv, u16 reg, u32 value) SBI_BUSY | SBI_CTL_OP_CRWR); - if (wait_for((I915_READ(SBI_CTL_STAT) & (SBI_READY | SBI_RESPONSE_SUCCESS)) == 0, + if (wait_for((I915_READ(SBI_CTL_STAT) & (SBI_BUSY | SBI_RESPONSE_FAIL)) == 0, 100)) { DRM_ERROR("timeout waiting for SBI to complete write transaction\n"); goto out_unlock; @@ -1372,10 +1520,10 @@ static u32 intel_sbi_read(struct drm_i915_private *dev_priv, u16 reg) { unsigned long flags; - u32 value; + u32 value = 0; spin_lock_irqsave(&dev_priv->dpio_lock, flags); - if (wait_for((I915_READ(SBI_CTL_STAT) & SBI_READY) == 0, + if (wait_for((I915_READ(SBI_CTL_STAT) & SBI_BUSY) == 0, 100)) { DRM_ERROR("timeout waiting for SBI to become ready\n"); goto out_unlock; @@ -1387,7 +1535,7 @@ intel_sbi_read(struct drm_i915_private *dev_priv, u16 reg) SBI_BUSY | SBI_CTL_OP_CRRD); - if (wait_for((I915_READ(SBI_CTL_STAT) & (SBI_READY | SBI_RESPONSE_SUCCESS)) == 0, + if (wait_for((I915_READ(SBI_CTL_STAT) & (SBI_BUSY | SBI_RESPONSE_FAIL)) == 0, 100)) { DRM_ERROR("timeout waiting for SBI to complete read transaction\n"); goto out_unlock; @@ -1824,6 +1972,22 @@ void intel_unpin_fb_obj(struct drm_i915_gem_object *obj) i915_gem_object_unpin(obj); } +/* Computes the linear offset to the base tile and adjusts x, y. bytes per pixel + * is assumed to be a power-of-two. */ +static unsigned long gen4_compute_dspaddr_offset_xtiled(int *x, int *y, + unsigned int bpp, + unsigned int pitch) +{ + int tile_rows, tiles; + + tile_rows = *y / 8; + *y %= 8; + tiles = *x / (512/bpp); + *x %= 512/bpp; + + return tile_rows * pitch * 8 + tiles * 4096; +} + static int i9xx_update_plane(struct drm_crtc *crtc, struct drm_framebuffer *fb, int x, int y) { @@ -1833,7 +1997,7 @@ static int i9xx_update_plane(struct drm_crtc *crtc, struct drm_framebuffer *fb, struct intel_framebuffer *intel_fb; struct drm_i915_gem_object *obj; int plane = intel_crtc->plane; - unsigned long Start, Offset; + unsigned long linear_offset; u32 dspcntr; u32 reg; @@ -1880,18 +2044,28 @@ static int i9xx_update_plane(struct drm_crtc *crtc, struct drm_framebuffer *fb, I915_WRITE(reg, dspcntr); - Start = obj->gtt_offset; - Offset = y * fb->pitches[0] + x * (fb->bits_per_pixel / 8); + linear_offset = y * fb->pitches[0] + x * (fb->bits_per_pixel / 8); - DRM_DEBUG_KMS("Writing base %08lX %08lX %d %d %d\n", - Start, Offset, x, y, fb->pitches[0]); + if (INTEL_INFO(dev)->gen >= 4) { + intel_crtc->dspaddr_offset = + gen4_compute_dspaddr_offset_xtiled(&x, &y, + fb->bits_per_pixel / 8, + fb->pitches[0]); + linear_offset -= intel_crtc->dspaddr_offset; + } else { + intel_crtc->dspaddr_offset = linear_offset; + } + + DRM_DEBUG_KMS("Writing base %08X %08lX %d %d %d\n", + obj->gtt_offset, linear_offset, x, y, fb->pitches[0]); I915_WRITE(DSPSTRIDE(plane), fb->pitches[0]); if (INTEL_INFO(dev)->gen >= 4) { - I915_MODIFY_DISPBASE(DSPSURF(plane), Start); + I915_MODIFY_DISPBASE(DSPSURF(plane), + obj->gtt_offset + intel_crtc->dspaddr_offset); I915_WRITE(DSPTILEOFF(plane), (y << 16) | x); - I915_WRITE(DSPADDR(plane), Offset); + I915_WRITE(DSPLINOFF(plane), linear_offset); } else - I915_WRITE(DSPADDR(plane), Start + Offset); + I915_WRITE(DSPADDR(plane), obj->gtt_offset + linear_offset); POSTING_READ(reg); return 0; @@ -1906,7 +2080,7 @@ static int ironlake_update_plane(struct drm_crtc *crtc, struct intel_framebuffer *intel_fb; struct drm_i915_gem_object *obj; int plane = intel_crtc->plane; - unsigned long Start, Offset; + unsigned long linear_offset; u32 dspcntr; u32 reg; @@ -1961,15 +2135,20 @@ static int ironlake_update_plane(struct drm_crtc *crtc, I915_WRITE(reg, dspcntr); - Start = obj->gtt_offset; - Offset = y * fb->pitches[0] + x * (fb->bits_per_pixel / 8); + linear_offset = y * fb->pitches[0] + x * (fb->bits_per_pixel / 8); + intel_crtc->dspaddr_offset = + gen4_compute_dspaddr_offset_xtiled(&x, &y, + fb->bits_per_pixel / 8, + fb->pitches[0]); + linear_offset -= intel_crtc->dspaddr_offset; - DRM_DEBUG_KMS("Writing base %08lX %08lX %d %d %d\n", - Start, Offset, x, y, fb->pitches[0]); + DRM_DEBUG_KMS("Writing base %08X %08lX %d %d %d\n", + obj->gtt_offset, linear_offset, x, y, fb->pitches[0]); I915_WRITE(DSPSTRIDE(plane), fb->pitches[0]); - I915_MODIFY_DISPBASE(DSPSURF(plane), Start); + I915_MODIFY_DISPBASE(DSPSURF(plane), + obj->gtt_offset + intel_crtc->dspaddr_offset); I915_WRITE(DSPTILEOFF(plane), (y << 16) | x); - I915_WRITE(DSPADDR(plane), Offset); + I915_WRITE(DSPLINOFF(plane), linear_offset); POSTING_READ(reg); return 0; @@ -2656,16 +2835,13 @@ static void intel_crtc_wait_for_pending_flips(struct drm_crtc *crtc) static bool intel_crtc_driving_pch(struct drm_crtc *crtc) { struct drm_device *dev = crtc->dev; - struct drm_mode_config *mode_config = &dev->mode_config; struct intel_encoder *encoder; /* * If there's a non-PCH eDP on this crtc, it must be DP_A, and that * must be driven by its own crtc; no sharing is possible. */ - list_for_each_entry(encoder, &mode_config->encoder_list, base.head) { - if (encoder->base.crtc != crtc) - continue; + for_each_encoder_on_crtc(dev, crtc, encoder) { /* On Haswell, LPT PCH handles the VGA connection via FDI, and Haswell * CPU handles all others */ @@ -3397,7 +3573,7 @@ void intel_encoder_destroy(struct drm_encoder *encoder) } static bool intel_crtc_mode_fixup(struct drm_crtc *crtc, - struct drm_display_mode *mode, + const struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode) { struct drm_device *dev = crtc->dev; @@ -3554,16 +3730,12 @@ static bool intel_choose_pipe_bpp_dither(struct drm_crtc *crtc, { struct drm_device *dev = crtc->dev; struct drm_i915_private *dev_priv = dev->dev_private; - struct drm_encoder *encoder; struct drm_connector *connector; + struct intel_encoder *intel_encoder; unsigned int display_bpc = UINT_MAX, bpc; /* Walk the encoders & connectors on this crtc, get min bpc */ - list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) { - struct intel_encoder *intel_encoder = to_intel_encoder(encoder); - - if (encoder->crtc != crtc) - continue; + for_each_encoder_on_crtc(dev, crtc, intel_encoder) { if (intel_encoder->type == INTEL_OUTPUT_LVDS) { unsigned int lvds_bpc; @@ -3595,7 +3767,7 @@ static bool intel_choose_pipe_bpp_dither(struct drm_crtc *crtc, /* Not one of the known troublemakers, check the EDID */ list_for_each_entry(connector, &dev->mode_config.connector_list, head) { - if (connector->encoder != encoder) + if (connector->encoder != &intel_encoder->base) continue; /* Don't use an invalid EDID bpc value */ @@ -3666,13 +3838,37 @@ static bool intel_choose_pipe_bpp_dither(struct drm_crtc *crtc, return display_bpc != bpc; } +static int vlv_get_refclk(struct drm_crtc *crtc) +{ + struct drm_device *dev = crtc->dev; + struct drm_i915_private *dev_priv = dev->dev_private; + int refclk = 27000; /* for DP & HDMI */ + + return 100000; /* only one validated so far */ + + if (intel_pipe_has_type(crtc, INTEL_OUTPUT_ANALOG)) { + refclk = 96000; + } else if (intel_pipe_has_type(crtc, INTEL_OUTPUT_LVDS)) { + if (intel_panel_use_ssc(dev_priv)) + refclk = 100000; + else + refclk = 96000; + } else if (intel_pipe_has_type(crtc, INTEL_OUTPUT_EDP)) { + refclk = 100000; + } + + return refclk; +} + static int i9xx_get_refclk(struct drm_crtc *crtc, int num_connectors) { struct drm_device *dev = crtc->dev; struct drm_i915_private *dev_priv = dev->dev_private; int refclk; - if (intel_pipe_has_type(crtc, INTEL_OUTPUT_LVDS) && + if (IS_VALLEYVIEW(dev)) { + refclk = vlv_get_refclk(crtc); + } else if (intel_pipe_has_type(crtc, INTEL_OUTPUT_LVDS) && intel_panel_use_ssc(dev_priv) && num_connectors < 2) { refclk = dev_priv->lvds_ssc_freq * 1000; DRM_DEBUG_KMS("using SSC reference clock of %d MHz\n", @@ -3787,6 +3983,72 @@ static void intel_update_lvds(struct drm_crtc *crtc, intel_clock_t *clock, I915_WRITE(LVDS, temp); } +static void vlv_update_pll(struct drm_crtc *crtc, + struct drm_display_mode *mode, + struct drm_display_mode *adjusted_mode, + intel_clock_t *clock, intel_clock_t *reduced_clock, + int refclk, int num_connectors) +{ + struct drm_device *dev = crtc->dev; + struct drm_i915_private *dev_priv = dev->dev_private; + struct intel_crtc *intel_crtc = to_intel_crtc(crtc); + int pipe = intel_crtc->pipe; + u32 dpll, mdiv, pdiv; + u32 bestn, bestm1, bestm2, bestp1, bestp2; + bool is_hdmi; + + is_hdmi = intel_pipe_has_type(crtc, INTEL_OUTPUT_HDMI); + + bestn = clock->n; + bestm1 = clock->m1; + bestm2 = clock->m2; + bestp1 = clock->p1; + bestp2 = clock->p2; + + /* Enable DPIO clock input */ + dpll = DPLL_EXT_BUFFER_ENABLE_VLV | DPLL_REFA_CLK_ENABLE_VLV | + DPLL_VGA_MODE_DIS | DPLL_INTEGRATED_CLOCK_VLV; + I915_WRITE(DPLL(pipe), dpll); + POSTING_READ(DPLL(pipe)); + + mdiv = ((bestm1 << DPIO_M1DIV_SHIFT) | (bestm2 & DPIO_M2DIV_MASK)); + mdiv |= ((bestp1 << DPIO_P1_SHIFT) | (bestp2 << DPIO_P2_SHIFT)); + mdiv |= ((bestn << DPIO_N_SHIFT)); + mdiv |= (1 << DPIO_POST_DIV_SHIFT); + mdiv |= (1 << DPIO_K_SHIFT); + mdiv |= DPIO_ENABLE_CALIBRATION; + intel_dpio_write(dev_priv, DPIO_DIV(pipe), mdiv); + + intel_dpio_write(dev_priv, DPIO_CORE_CLK(pipe), 0x01000000); + + pdiv = DPIO_REFSEL_OVERRIDE | (5 << DPIO_PLL_MODESEL_SHIFT) | + (3 << DPIO_BIAS_CURRENT_CTL_SHIFT) | (1<<20) | + (8 << DPIO_DRIVER_CTL_SHIFT) | (5 << DPIO_CLK_BIAS_CTL_SHIFT); + intel_dpio_write(dev_priv, DPIO_REFSFR(pipe), pdiv); + + intel_dpio_write(dev_priv, DPIO_LFP_COEFF(pipe), 0x009f0051); + + dpll |= DPLL_VCO_ENABLE; + I915_WRITE(DPLL(pipe), dpll); + POSTING_READ(DPLL(pipe)); + if (wait_for(((I915_READ(DPLL(pipe)) & DPLL_LOCK_VLV) == DPLL_LOCK_VLV), 1)) + DRM_ERROR("DPLL %d failed to lock\n", pipe); + + if (is_hdmi) { + u32 temp = intel_mode_get_pixel_multiplier(adjusted_mode); + + if (temp > 1) + temp = (temp - 1) << DPLL_MD_UDI_MULTIPLIER_SHIFT; + else + temp = 0; + + I915_WRITE(DPLL_MD(pipe), temp); + POSTING_READ(DPLL_MD(pipe)); + } + + intel_dpio_write(dev_priv, DPIO_FASTCLK_DISABLE, 0x641); /* ??? */ +} + static void i9xx_update_pll(struct drm_crtc *crtc, struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode, @@ -3974,15 +4236,11 @@ static int i9xx_crtc_mode_set(struct drm_crtc *crtc, u32 dspcntr, pipeconf, vsyncshift; bool ok, has_reduced_clock = false, is_sdvo = false; bool is_lvds = false, is_tv = false, is_dp = false; - struct drm_mode_config *mode_config = &dev->mode_config; struct intel_encoder *encoder; const intel_limit_t *limit; int ret; - list_for_each_entry(encoder, &mode_config->encoder_list, base.head) { - if (encoder->base.crtc != crtc) - continue; - + for_each_encoder_on_crtc(dev, crtc, encoder) { switch (encoder->type) { case INTEL_OUTPUT_LVDS: is_lvds = true; @@ -4044,6 +4302,9 @@ static int i9xx_crtc_mode_set(struct drm_crtc *crtc, if (IS_GEN2(dev)) i8xx_update_pll(crtc, adjusted_mode, &clock, num_connectors); + else if (IS_VALLEYVIEW(dev)) + vlv_update_pll(crtc, mode,adjusted_mode, &clock, NULL, + refclk, num_connectors); else i9xx_update_pll(crtc, mode, adjusted_mode, &clock, has_reduced_clock ? &reduced_clock : NULL, @@ -4282,15 +4543,11 @@ static int ironlake_get_refclk(struct drm_crtc *crtc) struct drm_device *dev = crtc->dev; struct drm_i915_private *dev_priv = dev->dev_private; struct intel_encoder *encoder; - struct drm_mode_config *mode_config = &dev->mode_config; struct intel_encoder *edp_encoder = NULL; int num_connectors = 0; bool is_lvds = false; - list_for_each_entry(encoder, &mode_config->encoder_list, base.head) { - if (encoder->base.crtc != crtc) - continue; - + for_each_encoder_on_crtc(dev, crtc, encoder) { switch (encoder->type) { case INTEL_OUTPUT_LVDS: is_lvds = true; @@ -4327,7 +4584,6 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc, u32 dpll, fp = 0, fp2 = 0, dspcntr, pipeconf; bool ok, has_reduced_clock = false, is_sdvo = false; bool is_crt = false, is_lvds = false, is_tv = false, is_dp = false; - struct drm_mode_config *mode_config = &dev->mode_config; struct intel_encoder *encoder, *edp_encoder = NULL; const intel_limit_t *limit; int ret; @@ -4338,10 +4594,7 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc, bool dither; bool is_cpu_edp = false, is_pch_edp = false; - list_for_each_entry(encoder, &mode_config->encoder_list, base.head) { - if (encoder->base.crtc != crtc) - continue; - + for_each_encoder_on_crtc(dev, crtc, encoder) { switch (encoder->type) { case INTEL_OUTPUT_LVDS: is_lvds = true; @@ -4405,25 +4658,10 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc, &clock, &reduced_clock); } - /* SDVO TV has fixed PLL values depend on its clock range, - this mirrors vbios setting. */ - if (is_sdvo && is_tv) { - if (adjusted_mode->clock >= 100000 - && adjusted_mode->clock < 140500) { - clock.p1 = 2; - clock.p2 = 10; - clock.n = 3; - clock.m1 = 16; - clock.m2 = 8; - } else if (adjusted_mode->clock >= 140500 - && adjusted_mode->clock <= 200000) { - clock.p1 = 1; - clock.p2 = 10; - clock.n = 6; - clock.m1 = 12; - clock.m2 = 8; - } - } + + if (is_sdvo && is_tv) + i9xx_adjust_sdvo_tv_clock(adjusted_mode, &clock); + /* FDI link */ pixel_multiplier = intel_mode_get_pixel_multiplier(adjusted_mode); @@ -4431,16 +4669,8 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc, /* CPU eDP doesn't require FDI link, so just set DP M/N according to current link config */ if (is_cpu_edp) { - target_clock = mode->clock; intel_edp_link_config(edp_encoder, &lane, &link_bw); } else { - /* [e]DP over FDI requires target mode clock - instead of link clock */ - if (is_dp) - target_clock = mode->clock; - else - target_clock = adjusted_mode->clock; - /* FDI is a binary signal running at ~2.7GHz, encoding * each output octet as 10 bits. The actual frequency * is stored as a divider into a 100MHz clock, and the @@ -4451,6 +4681,14 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc, link_bw = intel_fdi_link_freq(dev) * MHz(100)/KHz(1)/10; } + /* [e]DP over FDI requires target mode clock instead of link clock. */ + if (edp_encoder) + target_clock = intel_edp_target_clock(edp_encoder, mode); + else if (is_dp) + target_clock = mode->clock; + else + target_clock = adjusted_mode->clock; + /* determine panel color depth */ temp = I915_READ(PIPECONF(pipe)); temp &= ~PIPE_BPC_MASK; @@ -4662,16 +4900,8 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc, if (is_lvds && has_reduced_clock && i915_powersave) { I915_WRITE(intel_crtc->pch_pll->fp1_reg, fp2); intel_crtc->lowfreq_avail = true; - if (HAS_PIPE_CXSR(dev)) { - DRM_DEBUG_KMS("enabling CxSR downclocking\n"); - pipeconf |= PIPECONF_CXSR_DOWNCLOCK; - } } else { I915_WRITE(intel_crtc->pch_pll->fp1_reg, fp); - if (HAS_PIPE_CXSR(dev)) { - DRM_DEBUG_KMS("disabling CxSR downclocking\n"); - pipeconf &= ~PIPECONF_CXSR_DOWNCLOCK; - } } } @@ -5975,7 +6205,6 @@ static int intel_gen2_queue_flip(struct drm_device *dev, { struct drm_i915_private *dev_priv = dev->dev_private; struct intel_crtc *intel_crtc = to_intel_crtc(crtc); - unsigned long offset; u32 flip_mask; struct intel_ring_buffer *ring = &dev_priv->ring[RCS]; int ret; @@ -5984,9 +6213,6 @@ static int intel_gen2_queue_flip(struct drm_device *dev, if (ret) goto err; - /* Offset into the new buffer for cases of shared fbs between CRTCs */ - offset = crtc->y * fb->pitches[0] + crtc->x * fb->bits_per_pixel/8; - ret = intel_ring_begin(ring, 6); if (ret) goto err_unpin; @@ -6003,7 +6229,7 @@ static int intel_gen2_queue_flip(struct drm_device *dev, intel_ring_emit(ring, MI_DISPLAY_FLIP | MI_DISPLAY_FLIP_PLANE(intel_crtc->plane)); intel_ring_emit(ring, fb->pitches[0]); - intel_ring_emit(ring, obj->gtt_offset + offset); + intel_ring_emit(ring, obj->gtt_offset + intel_crtc->dspaddr_offset); intel_ring_emit(ring, 0); /* aux display base address, unused */ intel_ring_advance(ring); return 0; @@ -6021,7 +6247,6 @@ static int intel_gen3_queue_flip(struct drm_device *dev, { struct drm_i915_private *dev_priv = dev->dev_private; struct intel_crtc *intel_crtc = to_intel_crtc(crtc); - unsigned long offset; u32 flip_mask; struct intel_ring_buffer *ring = &dev_priv->ring[RCS]; int ret; @@ -6030,9 +6255,6 @@ static int intel_gen3_queue_flip(struct drm_device *dev, if (ret) goto err; - /* Offset into the new buffer for cases of shared fbs between CRTCs */ - offset = crtc->y * fb->pitches[0] + crtc->x * fb->bits_per_pixel/8; - ret = intel_ring_begin(ring, 6); if (ret) goto err_unpin; @@ -6046,7 +6268,7 @@ static int intel_gen3_queue_flip(struct drm_device *dev, intel_ring_emit(ring, MI_DISPLAY_FLIP_I915 | MI_DISPLAY_FLIP_PLANE(intel_crtc->plane)); intel_ring_emit(ring, fb->pitches[0]); - intel_ring_emit(ring, obj->gtt_offset + offset); + intel_ring_emit(ring, obj->gtt_offset + intel_crtc->dspaddr_offset); intel_ring_emit(ring, MI_NOOP); intel_ring_advance(ring); @@ -6084,7 +6306,9 @@ static int intel_gen4_queue_flip(struct drm_device *dev, intel_ring_emit(ring, MI_DISPLAY_FLIP | MI_DISPLAY_FLIP_PLANE(intel_crtc->plane)); intel_ring_emit(ring, fb->pitches[0]); - intel_ring_emit(ring, obj->gtt_offset | obj->tiling_mode); + intel_ring_emit(ring, + (obj->gtt_offset + intel_crtc->dspaddr_offset) | + obj->tiling_mode); /* XXX Enabling the panel-fitter across page-flip is so far * untested on non-native modes, so ignore it for now. @@ -6124,7 +6348,7 @@ static int intel_gen6_queue_flip(struct drm_device *dev, intel_ring_emit(ring, MI_DISPLAY_FLIP | MI_DISPLAY_FLIP_PLANE(intel_crtc->plane)); intel_ring_emit(ring, fb->pitches[0] | obj->tiling_mode); - intel_ring_emit(ring, obj->gtt_offset); + intel_ring_emit(ring, obj->gtt_offset + intel_crtc->dspaddr_offset); /* Contrary to the suggestions in the documentation, * "Enable Panel Fitter" does not seem to be required when page @@ -6187,7 +6411,7 @@ static int intel_gen7_queue_flip(struct drm_device *dev, intel_ring_emit(ring, MI_DISPLAY_FLIP_I915 | plane_bit); intel_ring_emit(ring, (fb->pitches[0] | obj->tiling_mode)); - intel_ring_emit(ring, (obj->gtt_offset)); + intel_ring_emit(ring, obj->gtt_offset + intel_crtc->dspaddr_offset); intel_ring_emit(ring, (MI_NOOP)); intel_ring_advance(ring); return 0; @@ -6219,6 +6443,19 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc, unsigned long flags; int ret; + /* Can't change pixel format via MI display flips. */ + if (fb->pixel_format != crtc->fb->pixel_format) + return -EINVAL; + + /* + * TILEOFF/LINOFF registers can't be changed via MI display flips. + * Note that pitch changes could also affect these register. + */ + if (INTEL_INFO(dev)->gen > 3 && + (fb->offsets[0] != crtc->fb->offsets[0] || + fb->pitches[0] != crtc->fb->pitches[0])) + return -EINVAL; + work = kzalloc(sizeof *work, GFP_KERNEL); if (work == NULL) return -ENOMEM; @@ -6249,7 +6486,9 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc, intel_fb = to_intel_framebuffer(fb); obj = intel_fb->obj; - mutex_lock(&dev->struct_mutex); + ret = i915_mutex_lock_interruptible(dev); + if (ret) + goto cleanup; /* Reference the objects for the scheduled work. */ drm_gem_object_reference(&work->old_fb_obj->base); @@ -6284,6 +6523,7 @@ cleanup_pending: drm_gem_object_unreference(&obj->base); mutex_unlock(&dev->struct_mutex); +cleanup: spin_lock_irqsave(&dev->event_lock, flags); intel_crtc->unpin_work = NULL; spin_unlock_irqrestore(&dev->event_lock, flags); @@ -6566,7 +6806,24 @@ static void intel_setup_outputs(struct drm_device *dev) if (!dpd_is_edp && (I915_READ(PCH_DP_D) & DP_DETECTED)) intel_dp_init(dev, PCH_DP_D); + } else if (IS_VALLEYVIEW(dev)) { + int found; + if (I915_READ(SDVOB) & PORT_DETECTED) { + /* SDVOB multiplex with HDMIB */ + found = intel_sdvo_init(dev, SDVOB, true); + if (!found) + intel_hdmi_init(dev, SDVOB); + if (!found && (I915_READ(DP_B) & DP_DETECTED)) + intel_dp_init(dev, DP_B); + } + + if (I915_READ(SDVOC) & PORT_DETECTED) + intel_hdmi_init(dev, SDVOC); + + /* Shares lanes with HDMI on SDVOC */ + if (I915_READ(DP_C) & DP_DETECTED) + intel_dp_init(dev, DP_C); } else if (SUPPORTS_DIGITAL_OUTPUTS(dev)) { bool found = false; @@ -6623,7 +6880,7 @@ static void intel_setup_outputs(struct drm_device *dev) /* disable all the possible outputs/crtcs before entering KMS mode */ drm_helper_disable_unused_functions(dev); - if (HAS_PCH_SPLIT(dev)) + if (HAS_PCH_IBX(dev) || HAS_PCH_CPT(dev)) ironlake_init_pch_refclk(dev); } @@ -6777,9 +7034,6 @@ static void intel_init_display(struct drm_device *dev) dev_priv->display.write_eld = ironlake_write_eld; } else dev_priv->display.update_wm = NULL; - } else if (IS_VALLEYVIEW(dev)) { - dev_priv->display.force_wake_get = vlv_force_wake_get; - dev_priv->display.force_wake_put = vlv_force_wake_put; } else if (IS_G4X(dev)) { dev_priv->display.write_eld = g4x_write_eld; } @@ -6923,20 +7177,18 @@ static void i915_disable_vga(struct drm_device *dev) void intel_modeset_init_hw(struct drm_device *dev) { - struct drm_i915_private *dev_priv = dev->dev_private; + /* We attempt to init the necessary power wells early in the initialization + * time, so the subsystems that expect power to be enabled can work. + */ + intel_init_power_wells(dev); - intel_init_clock_gating(dev); + intel_prepare_ddi(dev); - if (IS_IRONLAKE_M(dev)) { - ironlake_enable_drps(dev); - ironlake_enable_rc6(dev); - intel_init_emon(dev); - } + intel_init_clock_gating(dev); - if ((IS_GEN6(dev) || IS_GEN7(dev)) && !IS_VALLEYVIEW(dev)) { - gen6_enable_rps(dev_priv); - gen6_update_ring_freq(dev_priv); - } + mutex_lock(&dev->struct_mutex); + intel_enable_gt_powersave(dev); + mutex_unlock(&dev->struct_mutex); } void intel_modeset_init(struct drm_device *dev) @@ -6958,8 +7210,6 @@ void intel_modeset_init(struct drm_device *dev) intel_init_pm(dev); - intel_prepare_ddi(dev); - intel_init_display(dev); if (IS_GEN2(dev)) { @@ -6972,7 +7222,7 @@ void intel_modeset_init(struct drm_device *dev) dev->mode_config.max_width = 8192; dev->mode_config.max_height = 8192; } - dev->mode_config.fb_base = dev->agp->base; + dev->mode_config.fb_base = dev_priv->mm.gtt_base_addr; DRM_DEBUG_KMS("%d display pipe%s available.\n", dev_priv->num_pipe, dev_priv->num_pipe > 1 ? "s" : ""); @@ -7025,13 +7275,9 @@ void intel_modeset_cleanup(struct drm_device *dev) intel_disable_fbc(dev); - if (IS_IRONLAKE_M(dev)) - ironlake_disable_drps(dev); - if ((IS_GEN6(dev) || IS_GEN7(dev)) && !IS_VALLEYVIEW(dev)) - gen6_disable_rps(dev); + intel_disable_gt_powersave(dev); - if (IS_IRONLAKE_M(dev)) - ironlake_disable_rc6(dev); + ironlake_teardown_rc6(dev); if (IS_VALLEYVIEW(dev)) vlv_init_dpio(dev); diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c index c0449324143c..0a56b9ab0f58 100644 --- a/drivers/gpu/drm/i915/intel_dp.c +++ b/drivers/gpu/drm/i915/intel_dp.c @@ -155,6 +155,18 @@ intel_edp_link_config(struct intel_encoder *intel_encoder, *link_bw = 270000; } +int +intel_edp_target_clock(struct intel_encoder *intel_encoder, + struct drm_display_mode *mode) +{ + struct intel_dp *intel_dp = container_of(intel_encoder, struct intel_dp, base); + + if (intel_dp->panel_fixed_mode) + return intel_dp->panel_fixed_mode->clock; + else + return mode->clock; +} + static int intel_dp_max_lane_count(struct intel_dp *intel_dp) { @@ -225,7 +237,7 @@ intel_dp_max_data_rate(int max_link_clock, int max_lanes) static bool intel_dp_adjust_dithering(struct intel_dp *intel_dp, struct drm_display_mode *mode, - struct drm_display_mode *adjusted_mode) + bool adjust_mode) { int max_link_clock = intel_dp_link_clock(intel_dp_max_link_bw(intel_dp)); int max_lanes = intel_dp_max_lane_count(intel_dp); @@ -239,8 +251,8 @@ intel_dp_adjust_dithering(struct intel_dp *intel_dp, if (mode_rate > max_rate) return false; - if (adjusted_mode) - adjusted_mode->private_flags + if (adjust_mode) + mode->private_flags |= INTEL_MODE_DP_FORCE_6BPC; return true; @@ -263,7 +275,7 @@ intel_dp_mode_valid(struct drm_connector *connector, return MODE_PANEL; } - if (!intel_dp_adjust_dithering(intel_dp, mode, NULL)) + if (!intel_dp_adjust_dithering(intel_dp, mode, false)) return MODE_CLOCK_HIGH; if (mode->clock < 10000) @@ -691,7 +703,8 @@ intel_dp_i2c_init(struct intel_dp *intel_dp, } static bool -intel_dp_mode_fixup(struct drm_encoder *encoder, struct drm_display_mode *mode, +intel_dp_mode_fixup(struct drm_encoder *encoder, + const struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode) { struct drm_device *dev = encoder->dev; @@ -706,28 +719,23 @@ intel_dp_mode_fixup(struct drm_encoder *encoder, struct drm_display_mode *mode, intel_fixed_panel_mode(intel_dp->panel_fixed_mode, adjusted_mode); intel_pch_panel_fitting(dev, DRM_MODE_SCALE_FULLSCREEN, mode, adjusted_mode); - /* - * the mode->clock is used to calculate the Data&Link M/N - * of the pipe. For the eDP the fixed clock should be used. - */ - mode->clock = intel_dp->panel_fixed_mode->clock; } - if (mode->flags & DRM_MODE_FLAG_DBLCLK) + if (adjusted_mode->flags & DRM_MODE_FLAG_DBLCLK) return false; DRM_DEBUG_KMS("DP link computation with max lane count %i " "max bw %02x pixel clock %iKHz\n", - max_lane_count, bws[max_clock], mode->clock); + max_lane_count, bws[max_clock], adjusted_mode->clock); - if (!intel_dp_adjust_dithering(intel_dp, mode, adjusted_mode)) + if (!intel_dp_adjust_dithering(intel_dp, adjusted_mode, true)) return false; bpp = adjusted_mode->private_flags & INTEL_MODE_DP_FORCE_6BPC ? 18 : 24; - mode_rate = intel_dp_link_required(mode->clock, bpp); + mode_rate = intel_dp_link_required(adjusted_mode->clock, bpp); - for (lane_count = 1; lane_count <= max_lane_count; lane_count <<= 1) { - for (clock = 0; clock <= max_clock; clock++) { + for (clock = 0; clock <= max_clock; clock++) { + for (lane_count = 1; lane_count <= max_lane_count; lane_count <<= 1) { int link_avail = intel_dp_max_data_rate(intel_dp_link_clock(bws[clock]), lane_count); if (mode_rate <= link_avail) { @@ -786,8 +794,7 @@ intel_dp_set_m_n(struct drm_crtc *crtc, struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode) { struct drm_device *dev = crtc->dev; - struct drm_mode_config *mode_config = &dev->mode_config; - struct drm_encoder *encoder; + struct intel_encoder *encoder; struct drm_i915_private *dev_priv = dev->dev_private; struct intel_crtc *intel_crtc = to_intel_crtc(crtc); int lane_count = 4; @@ -797,13 +804,9 @@ intel_dp_set_m_n(struct drm_crtc *crtc, struct drm_display_mode *mode, /* * Find the lane count in the intel_encoder private */ - list_for_each_entry(encoder, &mode_config->encoder_list, head) { - struct intel_dp *intel_dp; - - if (encoder->crtc != crtc) - continue; + for_each_encoder_on_crtc(dev, crtc, encoder) { + struct intel_dp *intel_dp = enc_to_intel_dp(&encoder->base); - intel_dp = enc_to_intel_dp(encoder); if (intel_dp->base.type == INTEL_OUTPUT_DISPLAYPORT || intel_dp->base.type == INTEL_OUTPUT_EDP) { @@ -1768,7 +1771,7 @@ intel_dp_start_link_train(struct intel_dp *intel_dp) for (i = 0; i < intel_dp->lane_count; i++) if ((intel_dp->train_set[i] & DP_TRAIN_MAX_SWING_REACHED) == 0) break; - if (i == intel_dp->lane_count) { + if (i == intel_dp->lane_count && voltage_tries == 5) { ++loop_tries; if (loop_tries == 5) { DRM_DEBUG_KMS("too many full retries, give up\n"); @@ -1922,7 +1925,7 @@ intel_dp_link_down(struct intel_dp *intel_dp) DP |= DP_LINK_TRAIN_OFF; } - if (!HAS_PCH_CPT(dev) && + if (HAS_PCH_IBX(dev) && I915_READ(intel_dp->output_reg) & DP_PIPEB_SELECT) { struct drm_crtc *crtc = intel_dp->base.base.crtc; @@ -2099,25 +2102,23 @@ g4x_dp_detect(struct intel_dp *intel_dp) { struct drm_device *dev = intel_dp->base.base.dev; struct drm_i915_private *dev_priv = dev->dev_private; - uint32_t temp, bit; + uint32_t bit; switch (intel_dp->output_reg) { case DP_B: - bit = DPB_HOTPLUG_INT_STATUS; + bit = DPB_HOTPLUG_LIVE_STATUS; break; case DP_C: - bit = DPC_HOTPLUG_INT_STATUS; + bit = DPC_HOTPLUG_LIVE_STATUS; break; case DP_D: - bit = DPD_HOTPLUG_INT_STATUS; + bit = DPD_HOTPLUG_LIVE_STATUS; break; default: return connector_status_unknown; } - temp = I915_READ(PORT_HOTPLUG_STAT); - - if ((temp & bit) == 0) + if ((I915_READ(PORT_HOTPLUG_STAT) & bit) == 0) return connector_status_disconnected; return intel_dp_detect_dpcd(intel_dp); @@ -2399,16 +2400,11 @@ int intel_trans_dp_port_sel(struct drm_crtc *crtc) { struct drm_device *dev = crtc->dev; - struct drm_mode_config *mode_config = &dev->mode_config; - struct drm_encoder *encoder; - - list_for_each_entry(encoder, &mode_config->encoder_list, head) { - struct intel_dp *intel_dp; + struct intel_encoder *encoder; - if (encoder->crtc != crtc) - continue; + for_each_encoder_on_crtc(dev, crtc, encoder) { + struct intel_dp *intel_dp = enc_to_intel_dp(&encoder->base); - intel_dp = enc_to_intel_dp(encoder); if (intel_dp->base.type == INTEL_OUTPUT_DISPLAYPORT || intel_dp->base.type == INTEL_OUTPUT_EDP) return intel_dp->output_reg; @@ -2520,19 +2516,19 @@ intel_dp_init(struct drm_device *dev, int output_reg) case DP_B: case PCH_DP_B: dev_priv->hotplug_supported_mask |= - HDMIB_HOTPLUG_INT_STATUS; + DPB_HOTPLUG_INT_STATUS; name = "DPDDC-B"; break; case DP_C: case PCH_DP_C: dev_priv->hotplug_supported_mask |= - HDMIC_HOTPLUG_INT_STATUS; + DPC_HOTPLUG_INT_STATUS; name = "DPDDC-C"; break; case DP_D: case PCH_DP_D: dev_priv->hotplug_supported_mask |= - HDMID_HOTPLUG_INT_STATUS; + DPD_HOTPLUG_INT_STATUS; name = "DPDDC-D"; break; } diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h index 3e0918834e7e..84353559441c 100644 --- a/drivers/gpu/drm/i915/intel_drv.h +++ b/drivers/gpu/drm/i915/intel_drv.h @@ -169,6 +169,7 @@ struct intel_crtc { u8 lut_r[256], lut_g[256], lut_b[256]; int dpms_mode; bool active; /* is the crtc on? independent of the dpms mode */ + bool primary_disabled; /* is the crtc obscured by a plane? */ bool busy; /* is scanout buffer being updated frequently? */ struct timer_list idle_timer; bool lowfreq_avail; @@ -176,6 +177,11 @@ struct intel_crtc { struct intel_unpin_work *unpin_work; int fdi_lanes; + /* Display surface base address adjustement for pageflips. Note that on + * gen4+ this only adjusts up to a tile, offsets within a tile are + * handled in the hw itself (with the TILEOFF register). */ + unsigned long dspaddr_offset; + struct drm_i915_gem_object *cursor_bo; uint32_t cursor_addr; int16_t cursor_x, cursor_y; @@ -191,7 +197,6 @@ struct intel_plane { struct drm_plane base; enum pipe pipe; struct drm_i915_gem_object *obj; - bool primary_disabled; int max_downscale; u32 lut_r[1024], lut_g[1024], lut_b[1024]; void (*update_plane)(struct drm_plane *plane, @@ -301,6 +306,8 @@ struct intel_hdmi { enum hdmi_force_audio force_audio; void (*write_infoframe)(struct drm_encoder *encoder, struct dip_infoframe *frame); + void (*set_infoframes)(struct drm_encoder *encoder, + struct drm_display_mode *adjusted_mode); }; static inline struct drm_crtc * @@ -335,7 +342,6 @@ struct intel_fbc_work { }; int intel_ddc_get_modes(struct drm_connector *c, struct i2c_adapter *adapter); -extern bool intel_ddc_probe(struct intel_encoder *intel_encoder, int ddc_bus); extern void intel_attach_force_audio_property(struct drm_connector *connector); extern void intel_attach_broadcast_rgb_property(struct drm_connector *connector); @@ -343,9 +349,6 @@ extern void intel_attach_broadcast_rgb_property(struct drm_connector *connector) extern void intel_crt_init(struct drm_device *dev); extern void intel_hdmi_init(struct drm_device *dev, int sdvox_reg); extern struct intel_hdmi *enc_to_intel_hdmi(struct drm_encoder *encoder); -extern void intel_hdmi_set_avi_infoframe(struct drm_encoder *encoder, - struct drm_display_mode *adjusted_mode); -extern void intel_hdmi_set_spd_infoframe(struct drm_encoder *encoder); extern void intel_dip_infoframe_csum(struct dip_infoframe *avi_if); extern bool intel_sdvo_init(struct drm_device *dev, uint32_t sdvo_reg, bool is_sdvob); @@ -360,6 +363,8 @@ intel_dp_set_m_n(struct drm_crtc *crtc, struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode); extern bool intel_dpd_is_edp(struct drm_device *dev); extern void intel_edp_link_config(struct intel_encoder *, int *, int *); +extern int intel_edp_target_clock(struct intel_encoder *, + struct drm_display_mode *mode); extern bool intel_encoder_is_pch_edp(struct drm_encoder *encoder); extern int intel_plane_init(struct drm_device *dev, enum pipe pipe); extern void intel_flush_display_plane(struct drm_i915_private *dev_priv, @@ -372,13 +377,14 @@ extern void intel_fixed_panel_mode(struct drm_display_mode *fixed_mode, struct drm_display_mode *adjusted_mode); extern void intel_pch_panel_fitting(struct drm_device *dev, int fitting_mode, - struct drm_display_mode *mode, + const struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode); extern u32 intel_panel_get_max_backlight(struct drm_device *dev); extern u32 intel_panel_get_backlight(struct drm_device *dev); extern void intel_panel_set_backlight(struct drm_device *dev, u32 level); extern int intel_panel_setup_backlight(struct drm_device *dev); -extern void intel_panel_enable_backlight(struct drm_device *dev); +extern void intel_panel_enable_backlight(struct drm_device *dev, + enum pipe pipe); extern void intel_panel_disable_backlight(struct drm_device *dev); extern void intel_panel_destroy_backlight(struct drm_device *dev); extern enum drm_connector_status intel_panel_detect(struct drm_device *dev); @@ -423,9 +429,6 @@ extern void intel_crtc_fb_gamma_set(struct drm_crtc *crtc, u16 red, u16 green, extern void intel_crtc_fb_gamma_get(struct drm_crtc *crtc, u16 *red, u16 *green, u16 *blue, int regno); extern void intel_enable_clock_gating(struct drm_device *dev); -extern void ironlake_disable_rc6(struct drm_device *dev); -extern void ironlake_enable_drps(struct drm_device *dev); -extern void ironlake_disable_drps(struct drm_device *dev); extern int intel_pin_and_fence_fb_obj(struct drm_device *dev, struct drm_i915_gem_object *obj, @@ -492,10 +495,11 @@ extern void intel_update_fbc(struct drm_device *dev); extern void intel_gpu_ips_init(struct drm_i915_private *dev_priv); extern void intel_gpu_ips_teardown(void); -extern void gen6_enable_rps(struct drm_i915_private *dev_priv); -extern void gen6_update_ring_freq(struct drm_i915_private *dev_priv); -extern void gen6_disable_rps(struct drm_device *dev); -extern void intel_init_emon(struct drm_device *dev); +extern void intel_init_power_wells(struct drm_device *dev); +extern void intel_enable_gt_powersave(struct drm_device *dev); +extern void intel_disable_gt_powersave(struct drm_device *dev); +extern void gen6_gt_check_fifodbg(struct drm_i915_private *dev_priv); +extern void ironlake_teardown_rc6(struct drm_device *dev); extern void intel_ddi_dpms(struct drm_encoder *encoder, int mode); extern void intel_ddi_mode_set(struct drm_encoder *encoder, diff --git a/drivers/gpu/drm/i915/intel_dvo.c b/drivers/gpu/drm/i915/intel_dvo.c index 60ba50b956f2..36c542e5036b 100644 --- a/drivers/gpu/drm/i915/intel_dvo.c +++ b/drivers/gpu/drm/i915/intel_dvo.c @@ -136,7 +136,7 @@ static int intel_dvo_mode_valid(struct drm_connector *connector, } static bool intel_dvo_mode_fixup(struct drm_encoder *encoder, - struct drm_display_mode *mode, + const struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode) { struct intel_dvo *intel_dvo = enc_to_intel_dvo(encoder); diff --git a/drivers/gpu/drm/i915/intel_fb.c b/drivers/gpu/drm/i915/intel_fb.c index bf8690720a0c..97f673523b97 100644 --- a/drivers/gpu/drm/i915/intel_fb.c +++ b/drivers/gpu/drm/i915/intel_fb.c @@ -65,7 +65,7 @@ static int intelfb_create(struct intel_fbdev *ifbdev, struct drm_i915_private *dev_priv = dev->dev_private; struct fb_info *info; struct drm_framebuffer *fb; - struct drm_mode_fb_cmd2 mode_cmd; + struct drm_mode_fb_cmd2 mode_cmd = {}; struct drm_i915_gem_object *obj; struct device *device = &dev->pdev->dev; int size, ret; @@ -140,7 +140,9 @@ static int intelfb_create(struct intel_fbdev *ifbdev, info->fix.smem_start = dev->mode_config.fb_base + obj->gtt_offset; info->fix.smem_len = size; - info->screen_base = ioremap_wc(dev->agp->base + obj->gtt_offset, size); + info->screen_base = + ioremap_wc(dev_priv->mm.gtt_base_addr + obj->gtt_offset, + size); if (!info->screen_base) { ret = -ENOSPC; goto out_unpin; diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c index 2ead3bf7c21d..98f602427eb8 100644 --- a/drivers/gpu/drm/i915/intel_hdmi.c +++ b/drivers/gpu/drm/i915/intel_hdmi.c @@ -37,6 +37,19 @@ #include "i915_drm.h" #include "i915_drv.h" +static void +assert_hdmi_port_disabled(struct intel_hdmi *intel_hdmi) +{ + struct drm_device *dev = intel_hdmi->base.base.dev; + struct drm_i915_private *dev_priv = dev->dev_private; + uint32_t enabled_bits; + + enabled_bits = IS_HASWELL(dev) ? DDI_BUF_CTL_ENABLE : SDVO_ENABLE; + + WARN(I915_READ(intel_hdmi->sdvox_reg) & enabled_bits, + "HDMI port enabled, expecting disabled\n"); +} + struct intel_hdmi *enc_to_intel_hdmi(struct drm_encoder *encoder) { return container_of(encoder, struct intel_hdmi, base.base); @@ -121,36 +134,31 @@ static void g4x_write_infoframe(struct drm_encoder *encoder, uint32_t *data = (uint32_t *)frame; struct drm_device *dev = encoder->dev; struct drm_i915_private *dev_priv = dev->dev_private; - struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder); u32 val = I915_READ(VIDEO_DIP_CTL); unsigned i, len = DIP_HEADER_SIZE + frame->len; - val &= ~VIDEO_DIP_PORT_MASK; - if (intel_hdmi->sdvox_reg == SDVOB) - val |= VIDEO_DIP_PORT_B; - else if (intel_hdmi->sdvox_reg == SDVOC) - val |= VIDEO_DIP_PORT_C; - else - return; + WARN(!(val & VIDEO_DIP_ENABLE), "Writing DIP with CTL reg disabled\n"); val &= ~(VIDEO_DIP_SELECT_MASK | 0xf); /* clear DIP data offset */ val |= g4x_infoframe_index(frame); val &= ~g4x_infoframe_enable(frame); - val |= VIDEO_DIP_ENABLE; I915_WRITE(VIDEO_DIP_CTL, val); + mmiowb(); for (i = 0; i < len; i += 4) { I915_WRITE(VIDEO_DIP_DATA, *data); data++; } + mmiowb(); val |= g4x_infoframe_enable(frame); val &= ~VIDEO_DIP_FREQ_MASK; val |= VIDEO_DIP_FREQ_VSYNC; I915_WRITE(VIDEO_DIP_CTL, val); + POSTING_READ(VIDEO_DIP_CTL); } static void ibx_write_infoframe(struct drm_encoder *encoder, @@ -160,46 +168,32 @@ static void ibx_write_infoframe(struct drm_encoder *encoder, struct drm_device *dev = encoder->dev; struct drm_i915_private *dev_priv = dev->dev_private; struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc); - struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder); int reg = TVIDEO_DIP_CTL(intel_crtc->pipe); unsigned i, len = DIP_HEADER_SIZE + frame->len; u32 val = I915_READ(reg); - val &= ~VIDEO_DIP_PORT_MASK; - switch (intel_hdmi->sdvox_reg) { - case HDMIB: - val |= VIDEO_DIP_PORT_B; - break; - case HDMIC: - val |= VIDEO_DIP_PORT_C; - break; - case HDMID: - val |= VIDEO_DIP_PORT_D; - break; - default: - return; - } - - intel_wait_for_vblank(dev, intel_crtc->pipe); + WARN(!(val & VIDEO_DIP_ENABLE), "Writing DIP with CTL reg disabled\n"); val &= ~(VIDEO_DIP_SELECT_MASK | 0xf); /* clear DIP data offset */ val |= g4x_infoframe_index(frame); val &= ~g4x_infoframe_enable(frame); - val |= VIDEO_DIP_ENABLE; I915_WRITE(reg, val); + mmiowb(); for (i = 0; i < len; i += 4) { I915_WRITE(TVIDEO_DIP_DATA(intel_crtc->pipe), *data); data++; } + mmiowb(); val |= g4x_infoframe_enable(frame); val &= ~VIDEO_DIP_FREQ_MASK; val |= VIDEO_DIP_FREQ_VSYNC; I915_WRITE(reg, val); + POSTING_READ(reg); } static void cpt_write_infoframe(struct drm_encoder *encoder, @@ -213,32 +207,31 @@ static void cpt_write_infoframe(struct drm_encoder *encoder, unsigned i, len = DIP_HEADER_SIZE + frame->len; u32 val = I915_READ(reg); - intel_wait_for_vblank(dev, intel_crtc->pipe); + WARN(!(val & VIDEO_DIP_ENABLE), "Writing DIP with CTL reg disabled\n"); val &= ~(VIDEO_DIP_SELECT_MASK | 0xf); /* clear DIP data offset */ val |= g4x_infoframe_index(frame); /* The DIP control register spec says that we need to update the AVI * infoframe without clearing its enable bit */ - if (frame->type == DIP_TYPE_AVI) - val |= VIDEO_DIP_ENABLE_AVI; - else + if (frame->type != DIP_TYPE_AVI) val &= ~g4x_infoframe_enable(frame); - val |= VIDEO_DIP_ENABLE; - I915_WRITE(reg, val); + mmiowb(); for (i = 0; i < len; i += 4) { I915_WRITE(TVIDEO_DIP_DATA(intel_crtc->pipe), *data); data++; } + mmiowb(); val |= g4x_infoframe_enable(frame); val &= ~VIDEO_DIP_FREQ_MASK; val |= VIDEO_DIP_FREQ_VSYNC; I915_WRITE(reg, val); + POSTING_READ(reg); } static void vlv_write_infoframe(struct drm_encoder *encoder, @@ -252,26 +245,28 @@ static void vlv_write_infoframe(struct drm_encoder *encoder, unsigned i, len = DIP_HEADER_SIZE + frame->len; u32 val = I915_READ(reg); - intel_wait_for_vblank(dev, intel_crtc->pipe); + WARN(!(val & VIDEO_DIP_ENABLE), "Writing DIP with CTL reg disabled\n"); val &= ~(VIDEO_DIP_SELECT_MASK | 0xf); /* clear DIP data offset */ val |= g4x_infoframe_index(frame); val &= ~g4x_infoframe_enable(frame); - val |= VIDEO_DIP_ENABLE; I915_WRITE(reg, val); + mmiowb(); for (i = 0; i < len; i += 4) { I915_WRITE(VLV_TVIDEO_DIP_DATA(intel_crtc->pipe), *data); data++; } + mmiowb(); val |= g4x_infoframe_enable(frame); val &= ~VIDEO_DIP_FREQ_MASK; val |= VIDEO_DIP_FREQ_VSYNC; I915_WRITE(reg, val); + POSTING_READ(reg); } static void hsw_write_infoframe(struct drm_encoder *encoder, @@ -289,18 +284,19 @@ static void hsw_write_infoframe(struct drm_encoder *encoder, if (data_reg == 0) return; - intel_wait_for_vblank(dev, intel_crtc->pipe); - val &= ~hsw_infoframe_enable(frame); I915_WRITE(ctl_reg, val); + mmiowb(); for (i = 0; i < len; i += 4) { I915_WRITE(data_reg + i, *data); data++; } + mmiowb(); val |= hsw_infoframe_enable(frame); I915_WRITE(ctl_reg, val); + POSTING_READ(ctl_reg); } static void intel_set_infoframe(struct drm_encoder *encoder, @@ -308,14 +304,11 @@ static void intel_set_infoframe(struct drm_encoder *encoder, { struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder); - if (!intel_hdmi->has_hdmi_sink) - return; - intel_dip_infoframe_csum(frame); intel_hdmi->write_infoframe(encoder, frame); } -void intel_hdmi_set_avi_infoframe(struct drm_encoder *encoder, +static void intel_hdmi_set_avi_infoframe(struct drm_encoder *encoder, struct drm_display_mode *adjusted_mode) { struct dip_infoframe avi_if = { @@ -330,7 +323,7 @@ void intel_hdmi_set_avi_infoframe(struct drm_encoder *encoder, intel_set_infoframe(encoder, &avi_if); } -void intel_hdmi_set_spd_infoframe(struct drm_encoder *encoder) +static void intel_hdmi_set_spd_infoframe(struct drm_encoder *encoder) { struct dip_infoframe spd_if; @@ -345,6 +338,223 @@ void intel_hdmi_set_spd_infoframe(struct drm_encoder *encoder) intel_set_infoframe(encoder, &spd_if); } +static void g4x_set_infoframes(struct drm_encoder *encoder, + struct drm_display_mode *adjusted_mode) +{ + struct drm_i915_private *dev_priv = encoder->dev->dev_private; + struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder); + u32 reg = VIDEO_DIP_CTL; + u32 val = I915_READ(reg); + u32 port; + + assert_hdmi_port_disabled(intel_hdmi); + + /* If the registers were not initialized yet, they might be zeroes, + * which means we're selecting the AVI DIP and we're setting its + * frequency to once. This seems to really confuse the HW and make + * things stop working (the register spec says the AVI always needs to + * be sent every VSync). So here we avoid writing to the register more + * than we need and also explicitly select the AVI DIP and explicitly + * set its frequency to every VSync. Avoiding to write it twice seems to + * be enough to solve the problem, but being defensive shouldn't hurt us + * either. */ + val |= VIDEO_DIP_SELECT_AVI | VIDEO_DIP_FREQ_VSYNC; + + if (!intel_hdmi->has_hdmi_sink) { + if (!(val & VIDEO_DIP_ENABLE)) + return; + val &= ~VIDEO_DIP_ENABLE; + I915_WRITE(reg, val); + POSTING_READ(reg); + return; + } + + switch (intel_hdmi->sdvox_reg) { + case SDVOB: + port = VIDEO_DIP_PORT_B; + break; + case SDVOC: + port = VIDEO_DIP_PORT_C; + break; + default: + return; + } + + if (port != (val & VIDEO_DIP_PORT_MASK)) { + if (val & VIDEO_DIP_ENABLE) { + val &= ~VIDEO_DIP_ENABLE; + I915_WRITE(reg, val); + POSTING_READ(reg); + } + val &= ~VIDEO_DIP_PORT_MASK; + val |= port; + } + + val |= VIDEO_DIP_ENABLE; + val &= ~VIDEO_DIP_ENABLE_VENDOR; + + I915_WRITE(reg, val); + POSTING_READ(reg); + + intel_hdmi_set_avi_infoframe(encoder, adjusted_mode); + intel_hdmi_set_spd_infoframe(encoder); +} + +static void ibx_set_infoframes(struct drm_encoder *encoder, + struct drm_display_mode *adjusted_mode) +{ + struct drm_i915_private *dev_priv = encoder->dev->dev_private; + struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc); + struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder); + u32 reg = TVIDEO_DIP_CTL(intel_crtc->pipe); + u32 val = I915_READ(reg); + u32 port; + + assert_hdmi_port_disabled(intel_hdmi); + + /* See the big comment in g4x_set_infoframes() */ + val |= VIDEO_DIP_SELECT_AVI | VIDEO_DIP_FREQ_VSYNC; + + if (!intel_hdmi->has_hdmi_sink) { + if (!(val & VIDEO_DIP_ENABLE)) + return; + val &= ~VIDEO_DIP_ENABLE; + I915_WRITE(reg, val); + POSTING_READ(reg); + return; + } + + switch (intel_hdmi->sdvox_reg) { + case HDMIB: + port = VIDEO_DIP_PORT_B; + break; + case HDMIC: + port = VIDEO_DIP_PORT_C; + break; + case HDMID: + port = VIDEO_DIP_PORT_D; + break; + default: + return; + } + + if (port != (val & VIDEO_DIP_PORT_MASK)) { + if (val & VIDEO_DIP_ENABLE) { + val &= ~VIDEO_DIP_ENABLE; + I915_WRITE(reg, val); + POSTING_READ(reg); + } + val &= ~VIDEO_DIP_PORT_MASK; + val |= port; + } + + val |= VIDEO_DIP_ENABLE; + val &= ~(VIDEO_DIP_ENABLE_VENDOR | VIDEO_DIP_ENABLE_GAMUT | + VIDEO_DIP_ENABLE_GCP); + + I915_WRITE(reg, val); + POSTING_READ(reg); + + intel_hdmi_set_avi_infoframe(encoder, adjusted_mode); + intel_hdmi_set_spd_infoframe(encoder); +} + +static void cpt_set_infoframes(struct drm_encoder *encoder, + struct drm_display_mode *adjusted_mode) +{ + struct drm_i915_private *dev_priv = encoder->dev->dev_private; + struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc); + struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder); + u32 reg = TVIDEO_DIP_CTL(intel_crtc->pipe); + u32 val = I915_READ(reg); + + assert_hdmi_port_disabled(intel_hdmi); + + /* See the big comment in g4x_set_infoframes() */ + val |= VIDEO_DIP_SELECT_AVI | VIDEO_DIP_FREQ_VSYNC; + + if (!intel_hdmi->has_hdmi_sink) { + if (!(val & VIDEO_DIP_ENABLE)) + return; + val &= ~(VIDEO_DIP_ENABLE | VIDEO_DIP_ENABLE_AVI); + I915_WRITE(reg, val); + POSTING_READ(reg); + return; + } + + /* Set both together, unset both together: see the spec. */ + val |= VIDEO_DIP_ENABLE | VIDEO_DIP_ENABLE_AVI; + val &= ~(VIDEO_DIP_ENABLE_VENDOR | VIDEO_DIP_ENABLE_GAMUT | + VIDEO_DIP_ENABLE_GCP); + + I915_WRITE(reg, val); + POSTING_READ(reg); + + intel_hdmi_set_avi_infoframe(encoder, adjusted_mode); + intel_hdmi_set_spd_infoframe(encoder); +} + +static void vlv_set_infoframes(struct drm_encoder *encoder, + struct drm_display_mode *adjusted_mode) +{ + struct drm_i915_private *dev_priv = encoder->dev->dev_private; + struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc); + struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder); + u32 reg = VLV_TVIDEO_DIP_CTL(intel_crtc->pipe); + u32 val = I915_READ(reg); + + assert_hdmi_port_disabled(intel_hdmi); + + /* See the big comment in g4x_set_infoframes() */ + val |= VIDEO_DIP_SELECT_AVI | VIDEO_DIP_FREQ_VSYNC; + + if (!intel_hdmi->has_hdmi_sink) { + if (!(val & VIDEO_DIP_ENABLE)) + return; + val &= ~VIDEO_DIP_ENABLE; + I915_WRITE(reg, val); + POSTING_READ(reg); + return; + } + + val |= VIDEO_DIP_ENABLE; + val &= ~(VIDEO_DIP_ENABLE_VENDOR | VIDEO_DIP_ENABLE_GAMUT | + VIDEO_DIP_ENABLE_GCP); + + I915_WRITE(reg, val); + POSTING_READ(reg); + + intel_hdmi_set_avi_infoframe(encoder, adjusted_mode); + intel_hdmi_set_spd_infoframe(encoder); +} + +static void hsw_set_infoframes(struct drm_encoder *encoder, + struct drm_display_mode *adjusted_mode) +{ + struct drm_i915_private *dev_priv = encoder->dev->dev_private; + struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc); + struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder); + u32 reg = HSW_TVIDEO_DIP_CTL(intel_crtc->pipe); + u32 val = I915_READ(reg); + + assert_hdmi_port_disabled(intel_hdmi); + + if (!intel_hdmi->has_hdmi_sink) { + I915_WRITE(reg, 0); + POSTING_READ(reg); + return; + } + + val &= ~(VIDEO_DIP_ENABLE_VSC_HSW | VIDEO_DIP_ENABLE_GCP_HSW | + VIDEO_DIP_ENABLE_VS_HSW | VIDEO_DIP_ENABLE_GMP_HSW); + + I915_WRITE(reg, val); + POSTING_READ(reg); + + intel_hdmi_set_avi_infoframe(encoder, adjusted_mode); + intel_hdmi_set_spd_infoframe(encoder); +} + static void intel_hdmi_mode_set(struct drm_encoder *encoder, struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode) @@ -355,7 +565,7 @@ static void intel_hdmi_mode_set(struct drm_encoder *encoder, struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder); u32 sdvox; - sdvox = SDVO_ENCODING_HDMI | SDVO_BORDER_ENABLE; + sdvox = SDVO_ENCODING_HDMI; if (!HAS_PCH_SPLIT(dev)) sdvox |= intel_hdmi->color_range; if (adjusted_mode->flags & DRM_MODE_FLAG_PVSYNC) @@ -382,14 +592,13 @@ static void intel_hdmi_mode_set(struct drm_encoder *encoder, if (HAS_PCH_CPT(dev)) sdvox |= PORT_TRANS_SEL_CPT(intel_crtc->pipe); - else if (intel_crtc->pipe == 1) + else if (intel_crtc->pipe == PIPE_B) sdvox |= SDVO_PIPE_B_SELECT; I915_WRITE(intel_hdmi->sdvox_reg, sdvox); POSTING_READ(intel_hdmi->sdvox_reg); - intel_hdmi_set_avi_infoframe(encoder, adjusted_mode); - intel_hdmi_set_spd_infoframe(encoder); + intel_hdmi->set_infoframes(encoder, adjusted_mode); } static void intel_hdmi_dpms(struct drm_encoder *encoder, int mode) @@ -405,6 +614,36 @@ static void intel_hdmi_dpms(struct drm_encoder *encoder, int mode) temp = I915_READ(intel_hdmi->sdvox_reg); + /* HW workaround for IBX, we need to move the port to transcoder A + * before disabling it. */ + if (HAS_PCH_IBX(dev)) { + struct drm_crtc *crtc = encoder->crtc; + int pipe = crtc ? to_intel_crtc(crtc)->pipe : -1; + + if (mode != DRM_MODE_DPMS_ON) { + if (temp & SDVO_PIPE_B_SELECT) { + temp &= ~SDVO_PIPE_B_SELECT; + I915_WRITE(intel_hdmi->sdvox_reg, temp); + POSTING_READ(intel_hdmi->sdvox_reg); + + /* Again we need to write this twice. */ + I915_WRITE(intel_hdmi->sdvox_reg, temp); + POSTING_READ(intel_hdmi->sdvox_reg); + + /* Transcoder selection bits only update + * effectively on vblank. */ + if (crtc) + intel_wait_for_vblank(dev, pipe); + else + msleep(50); + } + } else { + /* Restore the transcoder select bit. */ + if (pipe == PIPE_B) + enable_bits |= SDVO_PIPE_B_SELECT; + } + } + /* HW workaround, need to toggle enable bit off and on for 12bpc, but * we do this anyway which shows more stable in testing. */ @@ -446,12 +685,33 @@ static int intel_hdmi_mode_valid(struct drm_connector *connector, } static bool intel_hdmi_mode_fixup(struct drm_encoder *encoder, - struct drm_display_mode *mode, + const struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode) { return true; } +static bool g4x_hdmi_connected(struct intel_hdmi *intel_hdmi) +{ + struct drm_device *dev = intel_hdmi->base.base.dev; + struct drm_i915_private *dev_priv = dev->dev_private; + uint32_t bit; + + switch (intel_hdmi->sdvox_reg) { + case SDVOB: + bit = HDMIB_HOTPLUG_LIVE_STATUS; + break; + case SDVOC: + bit = HDMIC_HOTPLUG_LIVE_STATUS; + break; + default: + bit = 0; + break; + } + + return I915_READ(PORT_HOTPLUG_STAT) & bit; +} + static enum drm_connector_status intel_hdmi_detect(struct drm_connector *connector, bool force) { @@ -460,6 +720,9 @@ intel_hdmi_detect(struct drm_connector *connector, bool force) struct edid *edid; enum drm_connector_status status = connector_status_disconnected; + if (IS_G4X(connector->dev) && !g4x_hdmi_connected(intel_hdmi)) + return status; + intel_hdmi->has_hdmi_sink = false; intel_hdmi->has_audio = false; edid = drm_get_edid(connector, @@ -633,7 +896,6 @@ void intel_hdmi_init(struct drm_device *dev, int sdvox_reg) struct intel_encoder *intel_encoder; struct intel_connector *intel_connector; struct intel_hdmi *intel_hdmi; - int i; intel_hdmi = kzalloc(sizeof(struct intel_hdmi), GFP_KERNEL); if (!intel_hdmi) @@ -710,26 +972,19 @@ void intel_hdmi_init(struct drm_device *dev, int sdvox_reg) if (!HAS_PCH_SPLIT(dev)) { intel_hdmi->write_infoframe = g4x_write_infoframe; - I915_WRITE(VIDEO_DIP_CTL, 0); + intel_hdmi->set_infoframes = g4x_set_infoframes; } else if (IS_VALLEYVIEW(dev)) { intel_hdmi->write_infoframe = vlv_write_infoframe; - for_each_pipe(i) - I915_WRITE(VLV_TVIDEO_DIP_CTL(i), 0); + intel_hdmi->set_infoframes = vlv_set_infoframes; } else if (IS_HASWELL(dev)) { - /* FIXME: Haswell has a new set of DIP frame registers, but we are - * just doing the minimal required for HDMI to work at this stage. - */ intel_hdmi->write_infoframe = hsw_write_infoframe; - for_each_pipe(i) - I915_WRITE(HSW_TVIDEO_DIP_CTL(i), 0); + intel_hdmi->set_infoframes = hsw_set_infoframes; } else if (HAS_PCH_IBX(dev)) { intel_hdmi->write_infoframe = ibx_write_infoframe; - for_each_pipe(i) - I915_WRITE(TVIDEO_DIP_CTL(i), 0); + intel_hdmi->set_infoframes = ibx_set_infoframes; } else { intel_hdmi->write_infoframe = cpt_write_infoframe; - for_each_pipe(i) - I915_WRITE(TVIDEO_DIP_CTL(i), 0); + intel_hdmi->set_infoframes = cpt_set_infoframes; } if (IS_HASWELL(dev)) diff --git a/drivers/gpu/drm/i915/intel_lvds.c b/drivers/gpu/drm/i915/intel_lvds.c index 08eb04c787e8..e05c0d3e3440 100644 --- a/drivers/gpu/drm/i915/intel_lvds.c +++ b/drivers/gpu/drm/i915/intel_lvds.c @@ -71,6 +71,7 @@ static struct intel_lvds *intel_attached_lvds(struct drm_connector *connector) static void intel_lvds_enable(struct intel_lvds *intel_lvds) { struct drm_device *dev = intel_lvds->base.base.dev; + struct intel_crtc *intel_crtc = to_intel_crtc(intel_lvds->base.base.crtc); struct drm_i915_private *dev_priv = dev->dev_private; u32 ctl_reg, lvds_reg, stat_reg; @@ -107,7 +108,7 @@ static void intel_lvds_enable(struct intel_lvds *intel_lvds) if (wait_for((I915_READ(stat_reg) & PP_ON) != 0, 1000)) DRM_ERROR("timed out waiting for panel to power on\n"); - intel_panel_enable_backlight(dev); + intel_panel_enable_backlight(dev, intel_crtc->pipe); } static void intel_lvds_disable(struct intel_lvds *intel_lvds) @@ -228,14 +229,14 @@ static inline u32 panel_fitter_scaling(u32 source, u32 target) } static bool intel_lvds_mode_fixup(struct drm_encoder *encoder, - struct drm_display_mode *mode, + const struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode) { struct drm_device *dev = encoder->dev; struct drm_i915_private *dev_priv = dev->dev_private; struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc); struct intel_lvds *intel_lvds = to_intel_lvds(encoder); - struct drm_encoder *tmp_encoder; + struct intel_encoder *tmp_encoder; u32 pfit_control = 0, pfit_pgm_ratios = 0, border = 0; int pipe; @@ -246,8 +247,8 @@ static bool intel_lvds_mode_fixup(struct drm_encoder *encoder, } /* Should never happen!! */ - list_for_each_entry(tmp_encoder, &dev->mode_config.encoder_list, head) { - if (tmp_encoder != encoder && tmp_encoder->crtc == encoder->crtc) { + for_each_encoder_on_crtc(dev, encoder->crtc, tmp_encoder) { + if (&tmp_encoder->base != encoder) { DRM_ERROR("Can't enable LVDS and another " "encoder on the same pipe\n"); return false; @@ -408,13 +409,7 @@ static void intel_lvds_prepare(struct drm_encoder *encoder) { struct intel_lvds *intel_lvds = to_intel_lvds(encoder); - /* - * Prior to Ironlake, we must disable the pipe if we want to adjust - * the panel fitter. However at all other times we can just reset - * the registers regardless. - */ - if (!HAS_PCH_SPLIT(encoder->dev) && intel_lvds->pfit_dirty) - intel_lvds_disable(intel_lvds); + intel_lvds_disable(intel_lvds); } static void intel_lvds_commit(struct drm_encoder *encoder) @@ -777,6 +772,14 @@ static const struct dmi_system_id intel_no_lvds[] = { DMI_MATCH(DMI_BOARD_NAME, "MS-7469"), }, }, + { + .callback = intel_no_lvds_dmi_callback, + .ident = "ZOTAC ZBOXSD-ID12/ID13", + .matches = { + DMI_MATCH(DMI_BOARD_VENDOR, "ZOTAC"), + DMI_MATCH(DMI_BOARD_NAME, "ZBOXSD-ID12/ID13"), + }, + }, { } /* terminating entry */ }; @@ -967,6 +970,8 @@ bool intel_lvds_init(struct drm_device *dev) intel_encoder->clone_mask = (1 << INTEL_LVDS_CLONE_BIT); if (HAS_PCH_SPLIT(dev)) intel_encoder->crtc_mask = (1 << 0) | (1 << 1) | (1 << 2); + else if (IS_GEN4(dev)) + intel_encoder->crtc_mask = (1 << 0) | (1 << 1); else intel_encoder->crtc_mask = (1 << 1); @@ -1074,35 +1079,14 @@ bool intel_lvds_init(struct drm_device *dev) goto failed; out: + /* + * Unlock registers and just + * leave them unlocked + */ if (HAS_PCH_SPLIT(dev)) { - u32 pwm; - - pipe = (I915_READ(PCH_LVDS) & LVDS_PIPEB_SELECT) ? 1 : 0; - - /* make sure PWM is enabled and locked to the LVDS pipe */ - pwm = I915_READ(BLC_PWM_CPU_CTL2); - if (pipe == 0 && (pwm & PWM_PIPE_B)) - I915_WRITE(BLC_PWM_CPU_CTL2, pwm & ~PWM_ENABLE); - if (pipe) - pwm |= PWM_PIPE_B; - else - pwm &= ~PWM_PIPE_B; - I915_WRITE(BLC_PWM_CPU_CTL2, pwm | PWM_ENABLE); - - pwm = I915_READ(BLC_PWM_PCH_CTL1); - pwm |= PWM_PCH_ENABLE; - I915_WRITE(BLC_PWM_PCH_CTL1, pwm); - /* - * Unlock registers and just - * leave them unlocked - */ I915_WRITE(PCH_PP_CONTROL, I915_READ(PCH_PP_CONTROL) | PANEL_UNLOCK_REGS); } else { - /* - * Unlock registers and just - * leave them unlocked - */ I915_WRITE(PP_CONTROL, I915_READ(PP_CONTROL) | PANEL_UNLOCK_REGS); } diff --git a/drivers/gpu/drm/i915/intel_modes.c b/drivers/gpu/drm/i915/intel_modes.c index d67ec3a51e42..45848b9b670b 100644 --- a/drivers/gpu/drm/i915/intel_modes.c +++ b/drivers/gpu/drm/i915/intel_modes.c @@ -33,34 +33,6 @@ #include "i915_drv.h" /** - * intel_ddc_probe - * - */ -bool intel_ddc_probe(struct intel_encoder *intel_encoder, int ddc_bus) -{ - struct drm_i915_private *dev_priv = intel_encoder->base.dev->dev_private; - u8 out_buf[] = { 0x0, 0x0}; - u8 buf[2]; - struct i2c_msg msgs[] = { - { - .addr = DDC_ADDR, - .flags = 0, - .len = 1, - .buf = out_buf, - }, - { - .addr = DDC_ADDR, - .flags = I2C_M_RD, - .len = 1, - .buf = buf, - } - }; - - return i2c_transfer(intel_gmbus_get_adapter(dev_priv, ddc_bus), - msgs, 2) == 2; -} - -/** * intel_ddc_get_modes - get modelist from monitor * @connector: DRM connector device to use * @adapter: i2c adapter diff --git a/drivers/gpu/drm/i915/intel_overlay.c b/drivers/gpu/drm/i915/intel_overlay.c index 458743da3774..830d0dd610e1 100644 --- a/drivers/gpu/drm/i915/intel_overlay.c +++ b/drivers/gpu/drm/i915/intel_overlay.c @@ -226,7 +226,7 @@ static int intel_overlay_do_wait_request(struct intel_overlay *overlay, } overlay->last_flip_req = request->seqno; overlay->flip_tail = tail; - ret = i915_wait_request(ring, overlay->last_flip_req); + ret = i915_wait_seqno(ring, overlay->last_flip_req); if (ret) return ret; i915_gem_retire_requests(dev); @@ -452,7 +452,7 @@ static int intel_overlay_recover_from_interrupt(struct intel_overlay *overlay) if (overlay->last_flip_req == 0) return 0; - ret = i915_wait_request(ring, overlay->last_flip_req); + ret = i915_wait_seqno(ring, overlay->last_flip_req); if (ret) return ret; i915_gem_retire_requests(dev); diff --git a/drivers/gpu/drm/i915/intel_panel.c b/drivers/gpu/drm/i915/intel_panel.c index 2a1625d84a69..10c7d39034e1 100644 --- a/drivers/gpu/drm/i915/intel_panel.c +++ b/drivers/gpu/drm/i915/intel_panel.c @@ -56,7 +56,7 @@ intel_fixed_panel_mode(struct drm_display_mode *fixed_mode, void intel_pch_panel_fitting(struct drm_device *dev, int fitting_mode, - struct drm_display_mode *mode, + const struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode) { struct drm_i915_private *dev_priv = dev->dev_private; @@ -287,9 +287,24 @@ void intel_panel_disable_backlight(struct drm_device *dev) dev_priv->backlight_enabled = false; intel_panel_actually_set_backlight(dev, 0); + + if (INTEL_INFO(dev)->gen >= 4) { + uint32_t reg, tmp; + + reg = HAS_PCH_SPLIT(dev) ? BLC_PWM_CPU_CTL2 : BLC_PWM_CTL2; + + I915_WRITE(reg, I915_READ(reg) & ~BLM_PWM_ENABLE); + + if (HAS_PCH_SPLIT(dev)) { + tmp = I915_READ(BLC_PWM_PCH_CTL1); + tmp &= ~BLM_PCH_PWM_ENABLE; + I915_WRITE(BLC_PWM_PCH_CTL1, tmp); + } + } } -void intel_panel_enable_backlight(struct drm_device *dev) +void intel_panel_enable_backlight(struct drm_device *dev, + enum pipe pipe) { struct drm_i915_private *dev_priv = dev->dev_private; @@ -298,6 +313,40 @@ void intel_panel_enable_backlight(struct drm_device *dev) dev_priv->backlight_enabled = true; intel_panel_actually_set_backlight(dev, dev_priv->backlight_level); + + if (INTEL_INFO(dev)->gen >= 4) { + uint32_t reg, tmp; + + reg = HAS_PCH_SPLIT(dev) ? BLC_PWM_CPU_CTL2 : BLC_PWM_CTL2; + + + tmp = I915_READ(reg); + + /* Note that this can also get called through dpms changes. And + * we don't track the backlight dpms state, hence check whether + * we have to do anything first. */ + if (tmp & BLM_PWM_ENABLE) + return; + + if (dev_priv->num_pipe == 3) + tmp &= ~BLM_PIPE_SELECT_IVB; + else + tmp &= ~BLM_PIPE_SELECT; + + tmp |= BLM_PIPE(pipe); + tmp &= ~BLM_PWM_ENABLE; + + I915_WRITE(reg, tmp); + POSTING_READ(reg); + I915_WRITE(reg, tmp | BLM_PWM_ENABLE); + + if (HAS_PCH_SPLIT(dev)) { + tmp = I915_READ(BLC_PWM_PCH_CTL1); + tmp |= BLM_PCH_PWM_ENABLE; + tmp &= ~BLM_PCH_OVERRIDE_ENABLE; + I915_WRITE(BLC_PWM_PCH_CTL1, tmp); + } + } } static void intel_panel_init_backlight(struct drm_device *dev) diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c index d0ce2a5b1d3f..94aabcaa3a67 100644 --- a/drivers/gpu/drm/i915/intel_pm.c +++ b/drivers/gpu/drm/i915/intel_pm.c @@ -387,8 +387,6 @@ void intel_update_fbc(struct drm_device *dev) struct drm_i915_gem_object *obj; int enable_fbc; - DRM_DEBUG_KMS("\n"); - if (!i915_powersave) return; @@ -405,7 +403,9 @@ void intel_update_fbc(struct drm_device *dev) * - going to an unsupported config (interlace, pixel multiply, etc.) */ list_for_each_entry(tmp_crtc, &dev->mode_config.crtc_list, head) { - if (tmp_crtc->enabled && tmp_crtc->fb) { + if (tmp_crtc->enabled && + !to_intel_crtc(tmp_crtc)->primary_disabled && + tmp_crtc->fb) { if (crtc) { DRM_DEBUG_KMS("more than one pipe active, disabling compression\n"); dev_priv->no_fbc_reason = FBC_MULTIPLE_PIPES; @@ -2182,7 +2182,7 @@ bool ironlake_set_drps(struct drm_device *dev, u8 val) return true; } -void ironlake_enable_drps(struct drm_device *dev) +static void ironlake_enable_drps(struct drm_device *dev) { struct drm_i915_private *dev_priv = dev->dev_private; u32 rgvmodectl = I915_READ(MEMMODECTL); @@ -2246,7 +2246,7 @@ void ironlake_enable_drps(struct drm_device *dev) getrawmonotonic(&dev_priv->last_time2); } -void ironlake_disable_drps(struct drm_device *dev) +static void ironlake_disable_drps(struct drm_device *dev) { struct drm_i915_private *dev_priv = dev->dev_private; u16 rgvswctl = I915_READ16(MEMSWCTL); @@ -2299,10 +2299,11 @@ void gen6_set_rps(struct drm_device *dev, u8 val) dev_priv->cur_delay = val; } -void gen6_disable_rps(struct drm_device *dev) +static void gen6_disable_rps(struct drm_device *dev) { struct drm_i915_private *dev_priv = dev->dev_private; + I915_WRITE(GEN6_RC_CONTROL, 0); I915_WRITE(GEN6_RPNSWREQ, 1 << 31); I915_WRITE(GEN6_PMINTRMSK, 0xffffffff); I915_WRITE(GEN6_PMIER, 0); @@ -2332,9 +2333,11 @@ int intel_enable_rc6(const struct drm_device *dev) if (INTEL_INFO(dev)->gen == 5) return 0; - /* Sorry Haswell, no RC6 for you for now. */ + /* On Haswell, only RC6 is available. So let's enable it by default to + * provide better testing and coverage since the beginning. + */ if (IS_HASWELL(dev)) - return 0; + return INTEL_RC6_ENABLE; /* * Disable rc6 on Sandybridge @@ -2347,8 +2350,9 @@ int intel_enable_rc6(const struct drm_device *dev) return (INTEL_RC6_ENABLE | INTEL_RC6p_ENABLE); } -void gen6_enable_rps(struct drm_i915_private *dev_priv) +static void gen6_enable_rps(struct drm_device *dev) { + struct drm_i915_private *dev_priv = dev->dev_private; struct intel_ring_buffer *ring; u32 rp_state_cap; u32 gt_perf_status; @@ -2357,6 +2361,8 @@ void gen6_enable_rps(struct drm_i915_private *dev_priv) int rc6_mode; int i; + WARN_ON(!mutex_is_locked(&dev->struct_mutex)); + /* Here begins a magic sequence of register writes to enable * auto-downclocking. * @@ -2364,7 +2370,6 @@ void gen6_enable_rps(struct drm_i915_private *dev_priv) * userspace... */ I915_WRITE(GEN6_RC_STATE, 0); - mutex_lock(&dev_priv->dev->struct_mutex); /* Clear the DBG now so we don't confuse earlier errors */ if ((gtfifodbg = I915_READ(GTFIFODBG))) { @@ -2400,20 +2405,24 @@ void gen6_enable_rps(struct drm_i915_private *dev_priv) I915_WRITE(GEN6_RC6p_THRESHOLD, 100000); I915_WRITE(GEN6_RC6pp_THRESHOLD, 64000); /* unused */ + /* Check if we are enabling RC6 */ rc6_mode = intel_enable_rc6(dev_priv->dev); if (rc6_mode & INTEL_RC6_ENABLE) rc6_mask |= GEN6_RC_CTL_RC6_ENABLE; - if (rc6_mode & INTEL_RC6p_ENABLE) - rc6_mask |= GEN6_RC_CTL_RC6p_ENABLE; + /* We don't use those on Haswell */ + if (!IS_HASWELL(dev)) { + if (rc6_mode & INTEL_RC6p_ENABLE) + rc6_mask |= GEN6_RC_CTL_RC6p_ENABLE; - if (rc6_mode & INTEL_RC6pp_ENABLE) - rc6_mask |= GEN6_RC_CTL_RC6pp_ENABLE; + if (rc6_mode & INTEL_RC6pp_ENABLE) + rc6_mask |= GEN6_RC_CTL_RC6pp_ENABLE; + } DRM_INFO("Enabling RC6 states: RC6 %s, RC6p %s, RC6pp %s\n", - (rc6_mode & INTEL_RC6_ENABLE) ? "on" : "off", - (rc6_mode & INTEL_RC6p_ENABLE) ? "on" : "off", - (rc6_mode & INTEL_RC6pp_ENABLE) ? "on" : "off"); + (rc6_mask & GEN6_RC_CTL_RC6_ENABLE) ? "on" : "off", + (rc6_mask & GEN6_RC_CTL_RC6p_ENABLE) ? "on" : "off", + (rc6_mask & GEN6_RC_CTL_RC6pp_ENABLE) ? "on" : "off"); I915_WRITE(GEN6_RC_CONTROL, rc6_mask | @@ -2431,10 +2440,19 @@ void gen6_enable_rps(struct drm_i915_private *dev_priv) I915_WRITE(GEN6_RP_INTERRUPT_LIMITS, dev_priv->max_delay << 24 | dev_priv->min_delay << 16); - I915_WRITE(GEN6_RP_UP_THRESHOLD, 10000); - I915_WRITE(GEN6_RP_DOWN_THRESHOLD, 1000000); - I915_WRITE(GEN6_RP_UP_EI, 100000); - I915_WRITE(GEN6_RP_DOWN_EI, 5000000); + + if (IS_HASWELL(dev)) { + I915_WRITE(GEN6_RP_UP_THRESHOLD, 59400); + I915_WRITE(GEN6_RP_DOWN_THRESHOLD, 245000); + I915_WRITE(GEN6_RP_UP_EI, 66000); + I915_WRITE(GEN6_RP_DOWN_EI, 350000); + } else { + I915_WRITE(GEN6_RP_UP_THRESHOLD, 10000); + I915_WRITE(GEN6_RP_DOWN_THRESHOLD, 1000000); + I915_WRITE(GEN6_RP_UP_EI, 100000); + I915_WRITE(GEN6_RP_DOWN_EI, 5000000); + } + I915_WRITE(GEN6_RP_IDLE_HYSTERSIS, 10); I915_WRITE(GEN6_RP_CONTROL, GEN6_RP_MEDIA_TURBO | @@ -2442,7 +2460,7 @@ void gen6_enable_rps(struct drm_i915_private *dev_priv) GEN6_RP_MEDIA_IS_GFX | GEN6_RP_ENABLE | GEN6_RP_UP_BUSY_AVG | - GEN6_RP_DOWN_IDLE_CONT); + (IS_HASWELL(dev) ? GEN7_RP_DOWN_IDLE_AVG : GEN6_RP_DOWN_IDLE_CONT)); if (wait_for((I915_READ(GEN6_PCODE_MAILBOX) & GEN6_PCODE_READY) == 0, 500)) @@ -2473,14 +2491,7 @@ void gen6_enable_rps(struct drm_i915_private *dev_priv) gen6_set_rps(dev_priv->dev, (gt_perf_status & 0xff00) >> 8); /* requires MSI enabled */ - I915_WRITE(GEN6_PMIER, - GEN6_PM_MBOX_EVENT | - GEN6_PM_THERMAL_EVENT | - GEN6_PM_RP_DOWN_TIMEOUT | - GEN6_PM_RP_UP_THRESHOLD | - GEN6_PM_RP_DOWN_THRESHOLD | - GEN6_PM_RP_UP_EI_EXPIRED | - GEN6_PM_RP_DOWN_EI_EXPIRED); + I915_WRITE(GEN6_PMIER, GEN6_PM_DEFERRED_EVENTS); spin_lock_irq(&dev_priv->rps_lock); WARN_ON(dev_priv->pm_iir != 0); I915_WRITE(GEN6_PMIMR, 0); @@ -2489,15 +2500,17 @@ void gen6_enable_rps(struct drm_i915_private *dev_priv) I915_WRITE(GEN6_PMINTRMSK, 0); gen6_gt_force_wake_put(dev_priv); - mutex_unlock(&dev_priv->dev->struct_mutex); } -void gen6_update_ring_freq(struct drm_i915_private *dev_priv) +static void gen6_update_ring_freq(struct drm_device *dev) { + struct drm_i915_private *dev_priv = dev->dev_private; int min_freq = 15; int gpu_freq, ia_freq, max_ia_freq; int scaling_factor = 180; + WARN_ON(!mutex_is_locked(&dev->struct_mutex)); + max_ia_freq = cpufreq_quick_get_max(0); /* * Default to measured freq if none found, PCU will ensure we don't go @@ -2509,8 +2522,6 @@ void gen6_update_ring_freq(struct drm_i915_private *dev_priv) /* Convert from kHz to MHz */ max_ia_freq /= 1000; - mutex_lock(&dev_priv->dev->struct_mutex); - /* * For each potential GPU frequency, load a ring frequency we'd like * to use for memory access. We do this by specifying the IA frequency @@ -2541,11 +2552,9 @@ void gen6_update_ring_freq(struct drm_i915_private *dev_priv) continue; } } - - mutex_unlock(&dev_priv->dev->struct_mutex); } -static void ironlake_teardown_rc6(struct drm_device *dev) +void ironlake_teardown_rc6(struct drm_device *dev) { struct drm_i915_private *dev_priv = dev->dev_private; @@ -2562,7 +2571,7 @@ static void ironlake_teardown_rc6(struct drm_device *dev) } } -void ironlake_disable_rc6(struct drm_device *dev) +static void ironlake_disable_rc6(struct drm_device *dev) { struct drm_i915_private *dev_priv = dev->dev_private; @@ -2578,8 +2587,6 @@ void ironlake_disable_rc6(struct drm_device *dev) I915_WRITE(RSTDBYCTL, I915_READ(RSTDBYCTL) & ~RCX_SW_EXIT); POSTING_READ(RSTDBYCTL); } - - ironlake_teardown_rc6(dev); } static int ironlake_setup_rc6(struct drm_device *dev) @@ -2601,7 +2608,7 @@ static int ironlake_setup_rc6(struct drm_device *dev) return 0; } -void ironlake_enable_rc6(struct drm_device *dev) +static void ironlake_enable_rc6(struct drm_device *dev) { struct drm_i915_private *dev_priv = dev->dev_private; struct intel_ring_buffer *ring = &dev_priv->ring[RCS]; @@ -2613,12 +2620,11 @@ void ironlake_enable_rc6(struct drm_device *dev) if (!intel_enable_rc6(dev)) return; - mutex_lock(&dev->struct_mutex); + WARN_ON(!mutex_is_locked(&dev->struct_mutex)); + ret = ironlake_setup_rc6(dev); - if (ret) { - mutex_unlock(&dev->struct_mutex); + if (ret) return; - } /* * GPU can automatically power down the render unit if given a page @@ -2627,7 +2633,6 @@ void ironlake_enable_rc6(struct drm_device *dev) ret = intel_ring_begin(ring, 6); if (ret) { ironlake_teardown_rc6(dev); - mutex_unlock(&dev->struct_mutex); return; } @@ -2652,13 +2657,11 @@ void ironlake_enable_rc6(struct drm_device *dev) if (ret) { DRM_ERROR("failed to enable ironlake power power savings\n"); ironlake_teardown_rc6(dev); - mutex_unlock(&dev->struct_mutex); return; } I915_WRITE(PWRCTXA, dev_priv->pwrctx->gtt_offset | PWRCTX_EN); I915_WRITE(RSTDBYCTL, I915_READ(RSTDBYCTL) & ~RCX_SW_EXIT); - mutex_unlock(&dev->struct_mutex); } static unsigned long intel_pxfreq(u32 vidfreq) @@ -3154,8 +3157,7 @@ void intel_gpu_ips_teardown(void) i915_mch_dev = NULL; spin_unlock(&mchdev_lock); } - -void intel_init_emon(struct drm_device *dev) +static void intel_init_emon(struct drm_device *dev) { struct drm_i915_private *dev_priv = dev->dev_private; u32 lcfuse; @@ -3226,6 +3228,28 @@ void intel_init_emon(struct drm_device *dev) dev_priv->corr = (lcfuse & LCFUSE_HIV_MASK); } +void intel_disable_gt_powersave(struct drm_device *dev) +{ + if (IS_IRONLAKE_M(dev)) { + ironlake_disable_drps(dev); + ironlake_disable_rc6(dev); + } else if (INTEL_INFO(dev)->gen >= 6 && !IS_VALLEYVIEW(dev)) { + gen6_disable_rps(dev); + } +} + +void intel_enable_gt_powersave(struct drm_device *dev) +{ + if (IS_IRONLAKE_M(dev)) { + ironlake_enable_drps(dev); + ironlake_enable_rc6(dev); + intel_init_emon(dev); + } else if ((IS_GEN6(dev) || IS_GEN7(dev)) && !IS_VALLEYVIEW(dev)) { + gen6_enable_rps(dev); + gen6_update_ring_freq(dev); + } +} + static void ironlake_init_clock_gating(struct drm_device *dev) { struct drm_i915_private *dev_priv = dev->dev_private; @@ -3328,8 +3352,12 @@ static void gen6_init_clock_gating(struct drm_device *dev) * * According to the spec, bit 11 (RCCUNIT) must also be set, * but we didn't debug actual testcases to find it out. + * + * Also apply WaDisableVDSUnitClockGating and + * WaDisableRCPBUnitClockGating. */ I915_WRITE(GEN6_UCGCTL2, + GEN7_VDSUNIT_CLOCK_GATE_DISABLE | GEN6_RCPBUNIT_CLOCK_GATE_DISABLE | GEN6_RCCUNIT_CLOCK_GATE_DISABLE); @@ -3357,6 +3385,9 @@ static void gen6_init_clock_gating(struct drm_device *dev) ILK_DPARB_CLK_GATE | ILK_DPFD_CLK_GATE); + I915_WRITE(GEN6_MBCTL, I915_READ(GEN6_MBCTL) | + GEN6_MBCTL_ENABLE_BOOT_FETCH); + for_each_pipe(pipe) { I915_WRITE(DSPCNTR(pipe), I915_READ(DSPCNTR(pipe)) | @@ -3377,7 +3408,7 @@ static void gen7_setup_fixed_func_scheduler(struct drm_i915_private *dev_priv) I915_WRITE(GEN7_FF_THREAD_MODE, reg); } -static void ivybridge_init_clock_gating(struct drm_device *dev) +static void haswell_init_clock_gating(struct drm_device *dev) { struct drm_i915_private *dev_priv = dev->dev_private; int pipe; @@ -3427,13 +3458,24 @@ static void ivybridge_init_clock_gating(struct drm_device *dev) /* WaDisable4x2SubspanOptimization */ I915_WRITE(CACHE_MODE_1, _MASKED_BIT_ENABLE(PIXEL_SUBSPAN_COLLECT_OPT_DISABLE)); + + /* XXX: This is a workaround for early silicon revisions and should be + * removed later. + */ + I915_WRITE(WM_DBG, + I915_READ(WM_DBG) | + WM_DBG_DISALLOW_MULTIPLE_LP | + WM_DBG_DISALLOW_SPRITE | + WM_DBG_DISALLOW_MAXFIFO); + } -static void valleyview_init_clock_gating(struct drm_device *dev) +static void ivybridge_init_clock_gating(struct drm_device *dev) { struct drm_i915_private *dev_priv = dev->dev_private; int pipe; uint32_t dspclk_gate = VRHUNIT_CLOCK_GATE_DISABLE; + uint32_t snpcr; I915_WRITE(PCH_DSPCLK_GATE_D, dspclk_gate); @@ -3441,10 +3483,77 @@ static void valleyview_init_clock_gating(struct drm_device *dev) I915_WRITE(WM2_LP_ILK, 0); I915_WRITE(WM1_LP_ILK, 0); - /* According to the spec, bit 13 (RCZUNIT) must be set on IVB. + I915_WRITE(ILK_DSPCLK_GATE, IVB_VRHUNIT_CLK_GATE); + + I915_WRITE(IVB_CHICKEN3, + CHICKEN3_DGMG_REQ_OUT_FIX_DISABLE | + CHICKEN3_DGMG_DONE_FIX_DISABLE); + + /* Apply the WaDisableRHWOOptimizationForRenderHang workaround. */ + I915_WRITE(GEN7_COMMON_SLICE_CHICKEN1, + GEN7_CSC1_RHWO_OPT_DISABLE_IN_RCC); + + /* WaApplyL3ControlAndL3ChickenMode requires those two on Ivy Bridge */ + I915_WRITE(GEN7_L3CNTLREG1, + GEN7_WA_FOR_GEN7_L3_CONTROL); + I915_WRITE(GEN7_L3_CHICKEN_MODE_REGISTER, + GEN7_WA_L3_CHICKEN_MODE); + + /* According to the BSpec vol1g, bit 12 (RCPBUNIT) clock + * gating disable must be set. Failure to set it results in + * flickering pixels due to Z write ordering failures after + * some amount of runtime in the Mesa "fire" demo, and Unigine + * Sanctuary and Tropics, and apparently anything else with + * alpha test or pixel discard. + * + * According to the spec, bit 11 (RCCUNIT) must also be set, + * but we didn't debug actual testcases to find it out. + * + * According to the spec, bit 13 (RCZUNIT) must be set on IVB. * This implements the WaDisableRCZUnitClockGating workaround. */ - I915_WRITE(GEN6_UCGCTL2, GEN6_RCZUNIT_CLOCK_GATE_DISABLE); + I915_WRITE(GEN6_UCGCTL2, + GEN6_RCZUNIT_CLOCK_GATE_DISABLE | + GEN6_RCCUNIT_CLOCK_GATE_DISABLE); + + /* This is required by WaCatErrorRejectionIssue */ + I915_WRITE(GEN7_SQ_CHICKEN_MBCUNIT_CONFIG, + I915_READ(GEN7_SQ_CHICKEN_MBCUNIT_CONFIG) | + GEN7_SQ_CHICKEN_MBCUNIT_SQINTMOB); + + for_each_pipe(pipe) { + I915_WRITE(DSPCNTR(pipe), + I915_READ(DSPCNTR(pipe)) | + DISPPLANE_TRICKLE_FEED_DISABLE); + intel_flush_display_plane(dev_priv, pipe); + } + + I915_WRITE(GEN6_MBCTL, I915_READ(GEN6_MBCTL) | + GEN6_MBCTL_ENABLE_BOOT_FETCH); + + gen7_setup_fixed_func_scheduler(dev_priv); + + /* WaDisable4x2SubspanOptimization */ + I915_WRITE(CACHE_MODE_1, + _MASKED_BIT_ENABLE(PIXEL_SUBSPAN_COLLECT_OPT_DISABLE)); + + snpcr = I915_READ(GEN6_MBCUNIT_SNPCR); + snpcr &= ~GEN6_MBC_SNPCR_MASK; + snpcr |= GEN6_MBC_SNPCR_MED; + I915_WRITE(GEN6_MBCUNIT_SNPCR, snpcr); +} + +static void valleyview_init_clock_gating(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + int pipe; + uint32_t dspclk_gate = VRHUNIT_CLOCK_GATE_DISABLE; + + I915_WRITE(PCH_DSPCLK_GATE_D, dspclk_gate); + + I915_WRITE(WM3_LP_ILK, 0); + I915_WRITE(WM2_LP_ILK, 0); + I915_WRITE(WM1_LP_ILK, 0); I915_WRITE(ILK_DSPCLK_GATE, IVB_VRHUNIT_CLK_GATE); @@ -3465,6 +3574,35 @@ static void valleyview_init_clock_gating(struct drm_device *dev) I915_READ(GEN7_SQ_CHICKEN_MBCUNIT_CONFIG) | GEN7_SQ_CHICKEN_MBCUNIT_SQINTMOB); + I915_WRITE(GEN6_MBCTL, I915_READ(GEN6_MBCTL) | + GEN6_MBCTL_ENABLE_BOOT_FETCH); + + + /* According to the BSpec vol1g, bit 12 (RCPBUNIT) clock + * gating disable must be set. Failure to set it results in + * flickering pixels due to Z write ordering failures after + * some amount of runtime in the Mesa "fire" demo, and Unigine + * Sanctuary and Tropics, and apparently anything else with + * alpha test or pixel discard. + * + * According to the spec, bit 11 (RCCUNIT) must also be set, + * but we didn't debug actual testcases to find it out. + * + * According to the spec, bit 13 (RCZUNIT) must be set on IVB. + * This implements the WaDisableRCZUnitClockGating workaround. + * + * Also apply WaDisableVDSUnitClockGating and + * WaDisableRCPBUnitClockGating. + */ + I915_WRITE(GEN6_UCGCTL2, + GEN7_VDSUNIT_CLOCK_GATE_DISABLE | + GEN7_TDLUNIT_CLOCK_GATE_DISABLE | + GEN6_RCZUNIT_CLOCK_GATE_DISABLE | + GEN6_RCPBUNIT_CLOCK_GATE_DISABLE | + GEN6_RCCUNIT_CLOCK_GATE_DISABLE); + + I915_WRITE(GEN7_UCGCTL4, GEN7_L3BANK2X_CLOCK_GATE_DISABLE); + for_each_pipe(pipe) { I915_WRITE(DSPCNTR(pipe), I915_READ(DSPCNTR(pipe)) | @@ -3474,6 +3612,19 @@ static void valleyview_init_clock_gating(struct drm_device *dev) I915_WRITE(CACHE_MODE_1, _MASKED_BIT_ENABLE(PIXEL_SUBSPAN_COLLECT_OPT_DISABLE)); + + /* + * On ValleyView, the GUnit needs to signal the GT + * when flip and other events complete. So enable + * all the GUnit->GT interrupts here + */ + I915_WRITE(VLV_DPFLIPSTAT, PIPEB_LINE_COMPARE_INT_EN | + PIPEB_HLINE_INT_EN | PIPEB_VBLANK_INT_EN | + SPRITED_FLIPDONE_INT_EN | SPRITEC_FLIPDONE_INT_EN | + PLANEB_FLIPDONE_INT_EN | PIPEA_LINE_COMPARE_INT_EN | + PIPEA_HLINE_INT_EN | PIPEA_VBLANK_INT_EN | + SPRITEB_FLIPDONE_INT_EN | SPRITEA_FLIPDONE_INT_EN | + PLANEA_FLIPDONE_INT_EN); } static void g4x_init_clock_gating(struct drm_device *dev) @@ -3681,34 +3832,6 @@ void intel_init_pm(struct drm_device *dev) /* For FIFO watermark updates */ if (HAS_PCH_SPLIT(dev)) { - dev_priv->display.force_wake_get = __gen6_gt_force_wake_get; - dev_priv->display.force_wake_put = __gen6_gt_force_wake_put; - - /* IVB configs may use multi-threaded forcewake */ - if (IS_IVYBRIDGE(dev) || IS_HASWELL(dev)) { - u32 ecobus; - - /* A small trick here - if the bios hasn't configured MT forcewake, - * and if the device is in RC6, then force_wake_mt_get will not wake - * the device and the ECOBUS read will return zero. Which will be - * (correctly) interpreted by the test below as MT forcewake being - * disabled. - */ - mutex_lock(&dev->struct_mutex); - __gen6_gt_force_wake_mt_get(dev_priv); - ecobus = I915_READ_NOTRACE(ECOBUS); - __gen6_gt_force_wake_mt_put(dev_priv); - mutex_unlock(&dev->struct_mutex); - - if (ecobus & FORCEWAKE_MT_ENABLE) { - DRM_DEBUG_KMS("Using MT version of forcewake\n"); - dev_priv->display.force_wake_get = - __gen6_gt_force_wake_mt_get; - dev_priv->display.force_wake_put = - __gen6_gt_force_wake_mt_put; - } - } - if (HAS_PCH_IBX(dev)) dev_priv->display.init_pch_clock_gating = ibx_init_clock_gating; else if (HAS_PCH_CPT(dev)) @@ -3756,7 +3879,7 @@ void intel_init_pm(struct drm_device *dev) "Disable CxSR\n"); dev_priv->display.update_wm = NULL; } - dev_priv->display.init_clock_gating = ivybridge_init_clock_gating; + dev_priv->display.init_clock_gating = haswell_init_clock_gating; dev_priv->display.sanitize_pm = gen6_sanitize_pm; } else dev_priv->display.update_wm = NULL; @@ -3764,8 +3887,6 @@ void intel_init_pm(struct drm_device *dev) dev_priv->display.update_wm = valleyview_update_wm; dev_priv->display.init_clock_gating = valleyview_init_clock_gating; - dev_priv->display.force_wake_get = vlv_force_wake_get; - dev_priv->display.force_wake_put = vlv_force_wake_put; } else if (IS_PINEVIEW(dev)) { if (!intel_get_cxsr_latency(IS_PINEVIEW_G(dev), dev_priv->is_ddr3, @@ -3811,10 +3932,196 @@ void intel_init_pm(struct drm_device *dev) else dev_priv->display.get_fifo_size = i830_get_fifo_size; } +} + +static void __gen6_gt_wait_for_thread_c0(struct drm_i915_private *dev_priv) +{ + u32 gt_thread_status_mask; + + if (IS_HASWELL(dev_priv->dev)) + gt_thread_status_mask = GEN6_GT_THREAD_STATUS_CORE_MASK_HSW; + else + gt_thread_status_mask = GEN6_GT_THREAD_STATUS_CORE_MASK; - /* We attempt to init the necessary power wells early in the initialization - * time, so the subsystems that expect power to be enabled can work. + /* w/a for a sporadic read returning 0 by waiting for the GT + * thread to wake up. */ - intel_init_power_wells(dev); + if (wait_for_atomic_us((I915_READ_NOTRACE(GEN6_GT_THREAD_STATUS_REG) & gt_thread_status_mask) == 0, 500)) + DRM_ERROR("GT thread status wait timed out\n"); +} + +static void __gen6_gt_force_wake_get(struct drm_i915_private *dev_priv) +{ + u32 forcewake_ack; + + if (IS_HASWELL(dev_priv->dev)) + forcewake_ack = FORCEWAKE_ACK_HSW; + else + forcewake_ack = FORCEWAKE_ACK; + + if (wait_for_atomic_us((I915_READ_NOTRACE(forcewake_ack) & 1) == 0, 500)) + DRM_ERROR("Force wake wait timed out\n"); + + I915_WRITE_NOTRACE(FORCEWAKE, 1); + + if (wait_for_atomic_us((I915_READ_NOTRACE(forcewake_ack) & 1), 500)) + DRM_ERROR("Force wake wait timed out\n"); + + __gen6_gt_wait_for_thread_c0(dev_priv); +} + +static void __gen6_gt_force_wake_mt_get(struct drm_i915_private *dev_priv) +{ + u32 forcewake_ack; + + if (IS_HASWELL(dev_priv->dev)) + forcewake_ack = FORCEWAKE_ACK_HSW; + else + forcewake_ack = FORCEWAKE_MT_ACK; + + if (wait_for_atomic_us((I915_READ_NOTRACE(forcewake_ack) & 1) == 0, 500)) + DRM_ERROR("Force wake wait timed out\n"); + + I915_WRITE_NOTRACE(FORCEWAKE_MT, _MASKED_BIT_ENABLE(1)); + + if (wait_for_atomic_us((I915_READ_NOTRACE(forcewake_ack) & 1), 500)) + DRM_ERROR("Force wake wait timed out\n"); + + __gen6_gt_wait_for_thread_c0(dev_priv); +} + +/* + * Generally this is called implicitly by the register read function. However, + * if some sequence requires the GT to not power down then this function should + * be called at the beginning of the sequence followed by a call to + * gen6_gt_force_wake_put() at the end of the sequence. + */ +void gen6_gt_force_wake_get(struct drm_i915_private *dev_priv) +{ + unsigned long irqflags; + + spin_lock_irqsave(&dev_priv->gt_lock, irqflags); + if (dev_priv->forcewake_count++ == 0) + dev_priv->gt.force_wake_get(dev_priv); + spin_unlock_irqrestore(&dev_priv->gt_lock, irqflags); +} + +void gen6_gt_check_fifodbg(struct drm_i915_private *dev_priv) +{ + u32 gtfifodbg; + gtfifodbg = I915_READ_NOTRACE(GTFIFODBG); + if (WARN(gtfifodbg & GT_FIFO_CPU_ERROR_MASK, + "MMIO read or write has been dropped %x\n", gtfifodbg)) + I915_WRITE_NOTRACE(GTFIFODBG, GT_FIFO_CPU_ERROR_MASK); +} + +static void __gen6_gt_force_wake_put(struct drm_i915_private *dev_priv) +{ + I915_WRITE_NOTRACE(FORCEWAKE, 0); + /* The below doubles as a POSTING_READ */ + gen6_gt_check_fifodbg(dev_priv); +} + +static void __gen6_gt_force_wake_mt_put(struct drm_i915_private *dev_priv) +{ + I915_WRITE_NOTRACE(FORCEWAKE_MT, _MASKED_BIT_DISABLE(1)); + /* The below doubles as a POSTING_READ */ + gen6_gt_check_fifodbg(dev_priv); +} + +/* + * see gen6_gt_force_wake_get() + */ +void gen6_gt_force_wake_put(struct drm_i915_private *dev_priv) +{ + unsigned long irqflags; + + spin_lock_irqsave(&dev_priv->gt_lock, irqflags); + if (--dev_priv->forcewake_count == 0) + dev_priv->gt.force_wake_put(dev_priv); + spin_unlock_irqrestore(&dev_priv->gt_lock, irqflags); +} + +int __gen6_gt_wait_for_fifo(struct drm_i915_private *dev_priv) +{ + int ret = 0; + + if (dev_priv->gt_fifo_count < GT_FIFO_NUM_RESERVED_ENTRIES) { + int loop = 500; + u32 fifo = I915_READ_NOTRACE(GT_FIFO_FREE_ENTRIES); + while (fifo <= GT_FIFO_NUM_RESERVED_ENTRIES && loop--) { + udelay(10); + fifo = I915_READ_NOTRACE(GT_FIFO_FREE_ENTRIES); + } + if (WARN_ON(loop < 0 && fifo <= GT_FIFO_NUM_RESERVED_ENTRIES)) + ++ret; + dev_priv->gt_fifo_count = fifo; + } + dev_priv->gt_fifo_count--; + + return ret; +} + +static void vlv_force_wake_get(struct drm_i915_private *dev_priv) +{ + /* Already awake? */ + if ((I915_READ(0x130094) & 0xa1) == 0xa1) + return; + + I915_WRITE_NOTRACE(FORCEWAKE_VLV, 0xffffffff); + POSTING_READ(FORCEWAKE_VLV); + + if (wait_for_atomic_us((I915_READ_NOTRACE(FORCEWAKE_ACK_VLV) & 1), 500)) + DRM_ERROR("Force wake wait timed out\n"); + + __gen6_gt_wait_for_thread_c0(dev_priv); +} + +static void vlv_force_wake_put(struct drm_i915_private *dev_priv) +{ + I915_WRITE_NOTRACE(FORCEWAKE_VLV, 0xffff0000); + /* FIXME: confirm VLV behavior with Punit folks */ + POSTING_READ(FORCEWAKE_VLV); +} + +void intel_gt_init(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + + spin_lock_init(&dev_priv->gt_lock); + + if (IS_VALLEYVIEW(dev)) { + dev_priv->gt.force_wake_get = vlv_force_wake_get; + dev_priv->gt.force_wake_put = vlv_force_wake_put; + } else if (INTEL_INFO(dev)->gen >= 6) { + dev_priv->gt.force_wake_get = __gen6_gt_force_wake_get; + dev_priv->gt.force_wake_put = __gen6_gt_force_wake_put; + + /* IVB configs may use multi-threaded forcewake */ + if (IS_IVYBRIDGE(dev) || IS_HASWELL(dev)) { + u32 ecobus; + + /* A small trick here - if the bios hasn't configured + * MT forcewake, and if the device is in RC6, then + * force_wake_mt_get will not wake the device and the + * ECOBUS read will return zero. Which will be + * (correctly) interpreted by the test below as MT + * forcewake being disabled. + */ + mutex_lock(&dev->struct_mutex); + __gen6_gt_force_wake_mt_get(dev_priv); + ecobus = I915_READ_NOTRACE(ECOBUS); + __gen6_gt_force_wake_mt_put(dev_priv); + mutex_unlock(&dev->struct_mutex); + + if (ecobus & FORCEWAKE_MT_ENABLE) { + DRM_DEBUG_KMS("Using MT version of forcewake\n"); + dev_priv->gt.force_wake_get = + __gen6_gt_force_wake_mt_get; + dev_priv->gt.force_wake_put = + __gen6_gt_force_wake_mt_put; + } + } + } } diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index e5b84ff89ca5..bf0195a96d53 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -219,19 +219,28 @@ gen6_render_ring_flush(struct intel_ring_buffer *ring, int ret; /* Force SNB workarounds for PIPE_CONTROL flushes */ - intel_emit_post_sync_nonzero_flush(ring); + ret = intel_emit_post_sync_nonzero_flush(ring); + if (ret) + return ret; /* Just flush everything. Experiments have shown that reducing the * number of bits based on the write domains has little performance * impact. */ flags |= PIPE_CONTROL_RENDER_TARGET_CACHE_FLUSH; + flags |= PIPE_CONTROL_TLB_INVALIDATE; flags |= PIPE_CONTROL_INSTRUCTION_CACHE_INVALIDATE; flags |= PIPE_CONTROL_TEXTURE_CACHE_INVALIDATE; flags |= PIPE_CONTROL_DEPTH_CACHE_FLUSH; flags |= PIPE_CONTROL_VF_CACHE_INVALIDATE; flags |= PIPE_CONTROL_CONST_CACHE_INVALIDATE; flags |= PIPE_CONTROL_STATE_CACHE_INVALIDATE; + /* + * Ensure that any following seqno writes only happen when the render + * cache is indeed flushed (but only if the caller actually wants that). + */ + if (flush_domains) + flags |= PIPE_CONTROL_CS_STALL; ret = intel_ring_begin(ring, 6); if (ret) @@ -433,11 +442,21 @@ static int init_render_ring(struct intel_ring_buffer *ring) */ I915_WRITE(CACHE_MODE_0, _MASKED_BIT_DISABLE(CM0_STC_EVICT_DISABLE_LRA_SNB)); + + /* This is not explicitly set for GEN6, so read the register. + * see intel_ring_mi_set_context() for why we care. + * TODO: consider explicitly setting the bit for GEN5 + */ + ring->itlb_before_ctx_switch = + !!(I915_READ(GFX_MODE) & GFX_TLB_INVALIDATE_ALWAYS); } if (INTEL_INFO(dev)->gen >= 6) I915_WRITE(INSTPM, _MASKED_BIT_ENABLE(INSTPM_FORCE_ORDERING)); + if (IS_IVYBRIDGE(dev)) + I915_WRITE_IMR(ring, ~GEN6_RENDER_L3_PARITY_ERROR); + return ret; } @@ -825,7 +844,11 @@ gen6_ring_get_irq(struct intel_ring_buffer *ring) spin_lock_irqsave(&dev_priv->irq_lock, flags); if (ring->irq_refcount++ == 0) { - I915_WRITE_IMR(ring, ~ring->irq_enable_mask); + if (IS_IVYBRIDGE(dev) && ring->id == RCS) + I915_WRITE_IMR(ring, ~(ring->irq_enable_mask | + GEN6_RENDER_L3_PARITY_ERROR)); + else + I915_WRITE_IMR(ring, ~ring->irq_enable_mask); dev_priv->gt_irq_mask &= ~ring->irq_enable_mask; I915_WRITE(GTIMR, dev_priv->gt_irq_mask); POSTING_READ(GTIMR); @@ -844,7 +867,10 @@ gen6_ring_put_irq(struct intel_ring_buffer *ring) spin_lock_irqsave(&dev_priv->irq_lock, flags); if (--ring->irq_refcount == 0) { - I915_WRITE_IMR(ring, ~0); + if (IS_IVYBRIDGE(dev) && ring->id == RCS) + I915_WRITE_IMR(ring, ~GEN6_RENDER_L3_PARITY_ERROR); + else + I915_WRITE_IMR(ring, ~0); dev_priv->gt_irq_mask |= ring->irq_enable_mask; I915_WRITE(GTIMR, dev_priv->gt_irq_mask); POSTING_READ(GTIMR); @@ -946,6 +972,7 @@ static int init_status_page(struct intel_ring_buffer *ring) ring->status_page.gfx_addr = obj->gtt_offset; ring->status_page.page_addr = kmap(obj->pages[0]); if (ring->status_page.page_addr == NULL) { + ret = -ENOMEM; goto err_unpin; } ring->status_page.obj = obj; @@ -969,6 +996,7 @@ static int intel_init_ring_buffer(struct drm_device *dev, struct intel_ring_buffer *ring) { struct drm_i915_gem_object *obj; + struct drm_i915_private *dev_priv = dev->dev_private; int ret; ring->dev = dev; @@ -1002,8 +1030,9 @@ static int intel_init_ring_buffer(struct drm_device *dev, if (ret) goto err_unpin; - ring->virtual_start = ioremap_wc(dev->agp->base + obj->gtt_offset, - ring->size); + ring->virtual_start = + ioremap_wc(dev_priv->mm.gtt->gma_bus_addr + obj->gtt_offset, + ring->size); if (ring->virtual_start == NULL) { DRM_ERROR("Failed to map ringbuffer.\n"); ret = -EINVAL; @@ -1089,20 +1118,9 @@ static int intel_wrap_ring_buffer(struct intel_ring_buffer *ring) static int intel_ring_wait_seqno(struct intel_ring_buffer *ring, u32 seqno) { - struct drm_i915_private *dev_priv = ring->dev->dev_private; - bool was_interruptible; int ret; - /* XXX As we have not yet audited all the paths to check that - * they are ready for ERESTARTSYS from intel_ring_begin, do not - * allow us to be interruptible by a signal. - */ - was_interruptible = dev_priv->mm.interruptible; - dev_priv->mm.interruptible = false; - - ret = i915_wait_request(ring, seqno); - - dev_priv->mm.interruptible = was_interruptible; + ret = i915_wait_seqno(ring, seqno); if (!ret) i915_gem_retire_requests_ring(ring); @@ -1200,8 +1218,10 @@ int intel_wait_ring_buffer(struct intel_ring_buffer *ring, int n) } msleep(1); - if (atomic_read(&dev_priv->mm.wedged)) - return -EAGAIN; + + ret = i915_gem_check_wedge(dev_priv, dev_priv->mm.interruptible); + if (ret) + return ret; } while (!time_after(jiffies, end)); trace_i915_ring_wait_end(ring); return -EBUSY; @@ -1210,12 +1230,13 @@ int intel_wait_ring_buffer(struct intel_ring_buffer *ring, int n) int intel_ring_begin(struct intel_ring_buffer *ring, int num_dwords) { - struct drm_i915_private *dev_priv = ring->dev->dev_private; + drm_i915_private_t *dev_priv = ring->dev->dev_private; int n = 4*num_dwords; int ret; - if (unlikely(atomic_read(&dev_priv->mm.wedged))) - return -EIO; + ret = i915_gem_check_wedge(dev_priv, dev_priv->mm.interruptible); + if (ret) + return ret; if (unlikely(ring->tail + n > ring->effective_size)) { ret = intel_wrap_ring_buffer(ring); @@ -1250,20 +1271,31 @@ static void gen6_bsd_ring_write_tail(struct intel_ring_buffer *ring, drm_i915_private_t *dev_priv = ring->dev->dev_private; /* Every tail move must follow the sequence below */ + + /* Disable notification that the ring is IDLE. The GT + * will then assume that it is busy and bring it out of rc6. + */ I915_WRITE(GEN6_BSD_SLEEP_PSMI_CONTROL, - GEN6_BSD_SLEEP_PSMI_CONTROL_RC_ILDL_MESSAGE_MODIFY_MASK | - GEN6_BSD_SLEEP_PSMI_CONTROL_RC_ILDL_MESSAGE_DISABLE); - I915_WRITE(GEN6_BSD_RNCID, 0x0); + _MASKED_BIT_ENABLE(GEN6_BSD_SLEEP_MSG_DISABLE)); + /* Clear the context id. Here be magic! */ + I915_WRITE64(GEN6_BSD_RNCID, 0x0); + + /* Wait for the ring not to be idle, i.e. for it to wake up. */ if (wait_for((I915_READ(GEN6_BSD_SLEEP_PSMI_CONTROL) & - GEN6_BSD_SLEEP_PSMI_CONTROL_IDLE_INDICATOR) == 0, - 50)) - DRM_ERROR("timed out waiting for IDLE Indicator\n"); + GEN6_BSD_SLEEP_INDICATOR) == 0, + 50)) + DRM_ERROR("timed out waiting for the BSD ring to wake up\n"); + /* Now that the ring is fully powered up, update the tail */ I915_WRITE_TAIL(ring, value); + POSTING_READ(RING_TAIL(ring->mmio_base)); + + /* Let the ring send IDLE messages to the GT again, + * and so let it sleep to conserve power when idle. + */ I915_WRITE(GEN6_BSD_SLEEP_PSMI_CONTROL, - GEN6_BSD_SLEEP_PSMI_CONTROL_RC_ILDL_MESSAGE_MODIFY_MASK | - GEN6_BSD_SLEEP_PSMI_CONTROL_RC_ILDL_MESSAGE_ENABLE); + _MASKED_BIT_DISABLE(GEN6_BSD_SLEEP_MSG_DISABLE)); } static int gen6_ring_flush(struct intel_ring_buffer *ring, diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h index 55d3da26bae7..1d3c81fdad92 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.h +++ b/drivers/gpu/drm/i915/intel_ringbuffer.h @@ -113,9 +113,17 @@ struct intel_ring_buffer { * Do we have some not yet emitted requests outstanding? */ u32 outstanding_lazy_request; + bool gpu_caches_dirty; wait_queue_head_t irq_queue; + /** + * Do an explicit TLB flush before MI_SET_CONTEXT + */ + bool itlb_before_ctx_switch; + struct i915_hw_context *default_context; + struct drm_i915_gem_object *last_context_obj; + void *private; }; diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c index b6a9d45fc3c6..26a6a4d0d078 100644 --- a/drivers/gpu/drm/i915/intel_sdvo.c +++ b/drivers/gpu/drm/i915/intel_sdvo.c @@ -140,9 +140,6 @@ struct intel_sdvo { /* DDC bus used by this SDVO encoder */ uint8_t ddc_bus; - - /* Input timings for adjusted_mode */ - struct intel_sdvo_dtd input_dtd; }; struct intel_sdvo_connector { @@ -938,7 +935,7 @@ static bool intel_sdvo_set_tv_format(struct intel_sdvo *intel_sdvo) static bool intel_sdvo_set_output_timings_from_mode(struct intel_sdvo *intel_sdvo, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct intel_sdvo_dtd output_dtd; @@ -953,11 +950,15 @@ intel_sdvo_set_output_timings_from_mode(struct intel_sdvo *intel_sdvo, return true; } +/* Asks the sdvo controller for the preferred input mode given the output mode. + * Unfortunately we have to set up the full output mode to do that. */ static bool -intel_sdvo_set_input_timings_for_mode(struct intel_sdvo *intel_sdvo, - struct drm_display_mode *mode, - struct drm_display_mode *adjusted_mode) +intel_sdvo_get_preferred_input_mode(struct intel_sdvo *intel_sdvo, + const struct drm_display_mode *mode, + struct drm_display_mode *adjusted_mode) { + struct intel_sdvo_dtd input_dtd; + /* Reset the input timing to the screen. Assume always input 0. */ if (!intel_sdvo_set_target_input(intel_sdvo)) return false; @@ -969,16 +970,16 @@ intel_sdvo_set_input_timings_for_mode(struct intel_sdvo *intel_sdvo, return false; if (!intel_sdvo_get_preferred_input_timing(intel_sdvo, - &intel_sdvo->input_dtd)) + &input_dtd)) return false; - intel_sdvo_get_mode_from_dtd(adjusted_mode, &intel_sdvo->input_dtd); + intel_sdvo_get_mode_from_dtd(adjusted_mode, &input_dtd); return true; } static bool intel_sdvo_mode_fixup(struct drm_encoder *encoder, - struct drm_display_mode *mode, + const struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode) { struct intel_sdvo *intel_sdvo = to_intel_sdvo(encoder); @@ -993,17 +994,17 @@ static bool intel_sdvo_mode_fixup(struct drm_encoder *encoder, if (!intel_sdvo_set_output_timings_from_mode(intel_sdvo, mode)) return false; - (void) intel_sdvo_set_input_timings_for_mode(intel_sdvo, - mode, - adjusted_mode); + (void) intel_sdvo_get_preferred_input_mode(intel_sdvo, + mode, + adjusted_mode); } else if (intel_sdvo->is_lvds) { if (!intel_sdvo_set_output_timings_from_mode(intel_sdvo, intel_sdvo->sdvo_lvds_fixed_mode)) return false; - (void) intel_sdvo_set_input_timings_for_mode(intel_sdvo, - mode, - adjusted_mode); + (void) intel_sdvo_get_preferred_input_mode(intel_sdvo, + mode, + adjusted_mode); } /* Make the CRTC code factor in the SDVO pixel multiplier. The @@ -1057,7 +1058,9 @@ static void intel_sdvo_mode_set(struct drm_encoder *encoder, intel_sdvo->sdvo_lvds_fixed_mode); else intel_sdvo_get_dtd_from_mode(&output_dtd, mode); - (void) intel_sdvo_set_output_timing(intel_sdvo, &output_dtd); + if (!intel_sdvo_set_output_timing(intel_sdvo, &output_dtd)) + DRM_INFO("Setting output timings on %s failed\n", + SDVO_NAME(intel_sdvo)); /* Set the input timing to the screen. Assume always input 0. */ if (!intel_sdvo_set_target_input(intel_sdvo)) @@ -1079,7 +1082,9 @@ static void intel_sdvo_mode_set(struct drm_encoder *encoder, * adjusted_mode. */ intel_sdvo_get_dtd_from_mode(&input_dtd, adjusted_mode); - (void) intel_sdvo_set_input_timing(intel_sdvo, &input_dtd); + if (!intel_sdvo_set_input_timing(intel_sdvo, &input_dtd)) + DRM_INFO("Setting input timings on %s failed\n", + SDVO_NAME(intel_sdvo)); switch (pixel_multiplier) { default: @@ -1376,7 +1381,7 @@ intel_sdvo_detect(struct drm_connector *connector, bool force) /* add 30ms delay when the output type might be TV */ if (intel_sdvo->caps.output_flags & SDVO_TV_MASK) - mdelay(30); + msleep(30); if (!intel_sdvo_read_response(intel_sdvo, &response, 2)) return connector_status_unknown; @@ -2521,6 +2526,7 @@ bool intel_sdvo_init(struct drm_device *dev, uint32_t sdvo_reg, bool is_sdvob) struct drm_i915_private *dev_priv = dev->dev_private; struct intel_encoder *intel_encoder; struct intel_sdvo *intel_sdvo; + u32 hotplug_mask; int i; intel_sdvo = kzalloc(sizeof(struct intel_sdvo), GFP_KERNEL); @@ -2552,10 +2558,18 @@ bool intel_sdvo_init(struct drm_device *dev, uint32_t sdvo_reg, bool is_sdvob) } } - if (intel_sdvo->is_sdvob) - dev_priv->hotplug_supported_mask |= SDVOB_HOTPLUG_INT_STATUS; - else - dev_priv->hotplug_supported_mask |= SDVOC_HOTPLUG_INT_STATUS; + hotplug_mask = 0; + if (IS_G4X(dev)) { + hotplug_mask = intel_sdvo->is_sdvob ? + SDVOB_HOTPLUG_INT_STATUS_G4X : SDVOC_HOTPLUG_INT_STATUS_G4X; + } else if (IS_GEN4(dev)) { + hotplug_mask = intel_sdvo->is_sdvob ? + SDVOB_HOTPLUG_INT_STATUS_I965 : SDVOC_HOTPLUG_INT_STATUS_I965; + } else { + hotplug_mask = intel_sdvo->is_sdvob ? + SDVOB_HOTPLUG_INT_STATUS_I915 : SDVOC_HOTPLUG_INT_STATUS_I915; + } + dev_priv->hotplug_supported_mask |= hotplug_mask; drm_encoder_helper_add(&intel_encoder->base, &intel_sdvo_helper_funcs); diff --git a/drivers/gpu/drm/i915/intel_sprite.c b/drivers/gpu/drm/i915/intel_sprite.c index 2a20fb0781d7..cc8df4de2d92 100644 --- a/drivers/gpu/drm/i915/intel_sprite.c +++ b/drivers/gpu/drm/i915/intel_sprite.c @@ -56,6 +56,7 @@ ivb_update_plane(struct drm_plane *plane, struct drm_framebuffer *fb, sprctl &= ~SPRITE_PIXFORMAT_MASK; sprctl &= ~SPRITE_RGB_ORDER_RGBX; sprctl &= ~SPRITE_YUV_BYTE_ORDER_MASK; + sprctl &= ~SPRITE_TILED; switch (fb->pixel_format) { case DRM_FORMAT_XBGR8888: @@ -84,7 +85,7 @@ ivb_update_plane(struct drm_plane *plane, struct drm_framebuffer *fb, break; default: DRM_DEBUG_DRIVER("bad pixel format, assuming RGBX888\n"); - sprctl |= DVS_FORMAT_RGBX888; + sprctl |= SPRITE_FORMAT_RGBX888; pixel_size = 4; break; } @@ -233,6 +234,7 @@ ilk_update_plane(struct drm_plane *plane, struct drm_framebuffer *fb, dvscntr &= ~DVS_PIXFORMAT_MASK; dvscntr &= ~DVS_RGB_ORDER_XBGR; dvscntr &= ~DVS_YUV_BYTE_ORDER_MASK; + dvscntr &= ~DVS_TILED; switch (fb->pixel_format) { case DRM_FORMAT_XBGR8888: @@ -326,6 +328,12 @@ intel_enable_primary(struct drm_crtc *crtc) struct intel_crtc *intel_crtc = to_intel_crtc(crtc); int reg = DSPCNTR(intel_crtc->plane); + if (!intel_crtc->primary_disabled) + return; + + intel_crtc->primary_disabled = false; + intel_update_fbc(dev); + I915_WRITE(reg, I915_READ(reg) | DISPLAY_PLANE_ENABLE); } @@ -337,7 +345,13 @@ intel_disable_primary(struct drm_crtc *crtc) struct intel_crtc *intel_crtc = to_intel_crtc(crtc); int reg = DSPCNTR(intel_crtc->plane); + if (intel_crtc->primary_disabled) + return; + I915_WRITE(reg, I915_READ(reg) & ~DISPLAY_PLANE_ENABLE); + + intel_crtc->primary_disabled = true; + intel_update_fbc(dev); } static int @@ -485,18 +499,14 @@ intel_update_plane(struct drm_plane *plane, struct drm_crtc *crtc, * Be sure to re-enable the primary before the sprite is no longer * covering it fully. */ - if (!disable_primary && intel_plane->primary_disabled) { + if (!disable_primary) intel_enable_primary(crtc); - intel_plane->primary_disabled = false; - } intel_plane->update_plane(plane, fb, obj, crtc_x, crtc_y, crtc_w, crtc_h, x, y, src_w, src_h); - if (disable_primary) { + if (disable_primary) intel_disable_primary(crtc); - intel_plane->primary_disabled = true; - } /* Unpin old obj after new one is active to avoid ugliness */ if (old_obj) { @@ -527,11 +537,8 @@ intel_disable_plane(struct drm_plane *plane) struct intel_plane *intel_plane = to_intel_plane(plane); int ret = 0; - if (intel_plane->primary_disabled) { + if (plane->crtc) intel_enable_primary(plane->crtc); - intel_plane->primary_disabled = false; - } - intel_plane->disable_plane(plane); if (!intel_plane->obj) @@ -685,6 +692,7 @@ intel_plane_init(struct drm_device *dev, enum pipe pipe) break; default: + kfree(intel_plane); return -ENODEV; } @@ -699,4 +707,3 @@ intel_plane_init(struct drm_device *dev, enum pipe pipe) return ret; } - diff --git a/drivers/gpu/drm/i915/intel_tv.c b/drivers/gpu/drm/i915/intel_tv.c index a233a51fd7e6..befce6c49704 100644 --- a/drivers/gpu/drm/i915/intel_tv.c +++ b/drivers/gpu/drm/i915/intel_tv.c @@ -891,24 +891,21 @@ intel_tv_mode_valid(struct drm_connector *connector, static bool -intel_tv_mode_fixup(struct drm_encoder *encoder, struct drm_display_mode *mode, +intel_tv_mode_fixup(struct drm_encoder *encoder, + const struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode) { struct drm_device *dev = encoder->dev; - struct drm_mode_config *drm_config = &dev->mode_config; struct intel_tv *intel_tv = enc_to_intel_tv(encoder); const struct tv_mode *tv_mode = intel_tv_mode_find(intel_tv); - struct drm_encoder *other_encoder; + struct intel_encoder *other_encoder; if (!tv_mode) return false; - /* FIXME: lock encoder list */ - list_for_each_entry(other_encoder, &drm_config->encoder_list, head) { - if (other_encoder != encoder && - other_encoder->crtc == encoder->crtc) + for_each_encoder_on_crtc(dev, encoder->crtc, other_encoder) + if (&other_encoder->base != encoder) return false; - } adjusted_mode->clock = tv_mode->clock; return true; |