diff options
Diffstat (limited to 'drivers/iommu/amd_iommu.c')
-rw-r--r-- | drivers/iommu/amd_iommu.c | 589 |
1 files changed, 184 insertions, 405 deletions
diff --git a/drivers/iommu/amd_iommu.c b/drivers/iommu/amd_iommu.c index fffea87a014f..d3e5e9abe3b6 100644 --- a/drivers/iommu/amd_iommu.c +++ b/drivers/iommu/amd_iommu.c @@ -65,10 +65,6 @@ static DEFINE_RWLOCK(amd_iommu_devtable_lock); -/* A list of preallocated protection domains */ -static LIST_HEAD(iommu_pd_list); -static DEFINE_SPINLOCK(iommu_pd_list_lock); - /* List of all available dev_data structures */ static LIST_HEAD(dev_data_list); static DEFINE_SPINLOCK(dev_data_list_lock); @@ -120,7 +116,7 @@ struct iommu_cmd { struct kmem_cache *amd_iommu_irq_cache; static void update_domain(struct protection_domain *domain); -static int __init alloc_passthrough_domain(void); +static int alloc_passthrough_domain(void); /**************************************************************************** * @@ -235,31 +231,38 @@ static bool pdev_pri_erratum(struct pci_dev *pdev, u32 erratum) } /* - * In this function the list of preallocated protection domains is traversed to - * find the domain for a specific device + * This function actually applies the mapping to the page table of the + * dma_ops domain. */ -static struct dma_ops_domain *find_protection_domain(u16 devid) +static void alloc_unity_mapping(struct dma_ops_domain *dma_dom, + struct unity_map_entry *e) { - struct dma_ops_domain *entry, *ret = NULL; - unsigned long flags; - u16 alias = amd_iommu_alias_table[devid]; - - if (list_empty(&iommu_pd_list)) - return NULL; - - spin_lock_irqsave(&iommu_pd_list_lock, flags); + u64 addr; - list_for_each_entry(entry, &iommu_pd_list, list) { - if (entry->target_dev == devid || - entry->target_dev == alias) { - ret = entry; - break; - } + for (addr = e->address_start; addr < e->address_end; + addr += PAGE_SIZE) { + if (addr < dma_dom->aperture_size) + __set_bit(addr >> PAGE_SHIFT, + dma_dom->aperture[0]->bitmap); } +} + +/* + * Inits the unity mappings required for a specific device + */ +static void init_unity_mappings_for_device(struct device *dev, + struct dma_ops_domain *dma_dom) +{ + struct unity_map_entry *e; + u16 devid; - spin_unlock_irqrestore(&iommu_pd_list_lock, flags); + devid = get_device_id(dev); - return ret; + list_for_each_entry(e, &amd_iommu_unity_map, list) { + if (!(devid >= e->devid_start && devid <= e->devid_end)) + continue; + alloc_unity_mapping(dma_dom, e); + } } /* @@ -291,11 +294,23 @@ static bool check_device(struct device *dev) static void init_iommu_group(struct device *dev) { + struct dma_ops_domain *dma_domain; + struct iommu_domain *domain; struct iommu_group *group; group = iommu_group_get_for_dev(dev); - if (!IS_ERR(group)) - iommu_group_put(group); + if (IS_ERR(group)) + return; + + domain = iommu_group_default_domain(group); + if (!domain) + goto out; + + dma_domain = to_pdomain(domain)->priv; + + init_unity_mappings_for_device(dev, dma_domain); +out: + iommu_group_put(group); } static int __last_alias(struct pci_dev *pdev, u16 alias, void *data) @@ -435,64 +450,15 @@ static void iommu_uninit_device(struct device *dev) /* Unlink from alias, it may change if another device is re-plugged */ dev_data->alias_data = NULL; + /* Remove dma-ops */ + dev->archdata.dma_ops = NULL; + /* * We keep dev_data around for unplugged devices and reuse it when the * device is re-plugged - not doing so would introduce a ton of races. */ } -void __init amd_iommu_uninit_devices(void) -{ - struct iommu_dev_data *dev_data, *n; - struct pci_dev *pdev = NULL; - - for_each_pci_dev(pdev) { - - if (!check_device(&pdev->dev)) - continue; - - iommu_uninit_device(&pdev->dev); - } - - /* Free all of our dev_data structures */ - list_for_each_entry_safe(dev_data, n, &dev_data_list, dev_data_list) - free_dev_data(dev_data); -} - -int __init amd_iommu_init_devices(void) -{ - struct pci_dev *pdev = NULL; - int ret = 0; - - for_each_pci_dev(pdev) { - - if (!check_device(&pdev->dev)) - continue; - - ret = iommu_init_device(&pdev->dev); - if (ret == -ENOTSUPP) - iommu_ignore_device(&pdev->dev); - else if (ret) - goto out_free; - } - - /* - * Initialize IOMMU groups only after iommu_init_device() has - * had a chance to populate any IVRS defined aliases. - */ - for_each_pci_dev(pdev) { - if (check_device(&pdev->dev)) - init_iommu_group(&pdev->dev); - } - - return 0; - -out_free: - - amd_iommu_uninit_devices(); - - return ret; -} #ifdef CONFIG_AMD_IOMMU_STATS /* @@ -1464,94 +1430,6 @@ static unsigned long iommu_unmap_page(struct protection_domain *dom, return unmapped; } -/* - * This function checks if a specific unity mapping entry is needed for - * this specific IOMMU. - */ -static int iommu_for_unity_map(struct amd_iommu *iommu, - struct unity_map_entry *entry) -{ - u16 bdf, i; - - for (i = entry->devid_start; i <= entry->devid_end; ++i) { - bdf = amd_iommu_alias_table[i]; - if (amd_iommu_rlookup_table[bdf] == iommu) - return 1; - } - - return 0; -} - -/* - * This function actually applies the mapping to the page table of the - * dma_ops domain. - */ -static int dma_ops_unity_map(struct dma_ops_domain *dma_dom, - struct unity_map_entry *e) -{ - u64 addr; - int ret; - - for (addr = e->address_start; addr < e->address_end; - addr += PAGE_SIZE) { - ret = iommu_map_page(&dma_dom->domain, addr, addr, e->prot, - PAGE_SIZE); - if (ret) - return ret; - /* - * if unity mapping is in aperture range mark the page - * as allocated in the aperture - */ - if (addr < dma_dom->aperture_size) - __set_bit(addr >> PAGE_SHIFT, - dma_dom->aperture[0]->bitmap); - } - - return 0; -} - -/* - * Init the unity mappings for a specific IOMMU in the system - * - * Basically iterates over all unity mapping entries and applies them to - * the default domain DMA of that IOMMU if necessary. - */ -static int iommu_init_unity_mappings(struct amd_iommu *iommu) -{ - struct unity_map_entry *entry; - int ret; - - list_for_each_entry(entry, &amd_iommu_unity_map, list) { - if (!iommu_for_unity_map(iommu, entry)) - continue; - ret = dma_ops_unity_map(iommu->default_dom, entry); - if (ret) - return ret; - } - - return 0; -} - -/* - * Inits the unity mappings required for a specific device - */ -static int init_unity_mappings_for_device(struct dma_ops_domain *dma_dom, - u16 devid) -{ - struct unity_map_entry *e; - int ret; - - list_for_each_entry(e, &amd_iommu_unity_map, list) { - if (!(devid >= e->devid_start && devid <= e->devid_end)) - continue; - ret = dma_ops_unity_map(dma_dom, e); - if (ret) - return ret; - } - - return 0; -} - /**************************************************************************** * * The next functions belong to the address allocator for the dma_ops @@ -1705,14 +1583,16 @@ static unsigned long dma_ops_area_alloc(struct device *dev, unsigned long next_bit = dom->next_address % APERTURE_RANGE_SIZE; int max_index = dom->aperture_size >> APERTURE_RANGE_SHIFT; int i = start >> APERTURE_RANGE_SHIFT; - unsigned long boundary_size; + unsigned long boundary_size, mask; unsigned long address = -1; unsigned long limit; next_bit >>= PAGE_SHIFT; - boundary_size = ALIGN(dma_get_seg_boundary(dev) + 1, - PAGE_SIZE) >> PAGE_SHIFT; + mask = dma_get_seg_boundary(dev); + + boundary_size = mask + 1 ? ALIGN(mask + 1, PAGE_SIZE) >> PAGE_SHIFT : + 1UL << (BITS_PER_LONG - PAGE_SHIFT); for (;i < max_index; ++i) { unsigned long offset = dom->aperture[i]->offset >> PAGE_SHIFT; @@ -1870,9 +1750,15 @@ static void free_pt_##LVL (unsigned long __pt) \ pt = (u64 *)__pt; \ \ for (i = 0; i < 512; ++i) { \ + /* PTE present? */ \ if (!IOMMU_PTE_PRESENT(pt[i])) \ continue; \ \ + /* Large PTE? */ \ + if (PM_PTE_LEVEL(pt[i]) == 0 || \ + PM_PTE_LEVEL(pt[i]) == 7) \ + continue; \ + \ p = (unsigned long)IOMMU_PTE_PAGE(pt[i]); \ FN(p); \ } \ @@ -2009,7 +1895,6 @@ static struct dma_ops_domain *dma_ops_domain_alloc(void) goto free_dma_dom; dma_dom->need_flush = false; - dma_dom->target_dev = 0xffff; add_domain_to_list(&dma_dom->domain); @@ -2374,110 +2259,67 @@ static void detach_device(struct device *dev) dev_data->ats.enabled = false; } -/* - * Find out the protection domain structure for a given PCI device. This - * will give us the pointer to the page table root for example. - */ -static struct protection_domain *domain_for_device(struct device *dev) -{ - struct iommu_dev_data *dev_data; - struct protection_domain *dom = NULL; - unsigned long flags; - - dev_data = get_dev_data(dev); - - if (dev_data->domain) - return dev_data->domain; - - if (dev_data->alias_data != NULL) { - struct iommu_dev_data *alias_data = dev_data->alias_data; - - read_lock_irqsave(&amd_iommu_devtable_lock, flags); - if (alias_data->domain != NULL) { - __attach_device(dev_data, alias_data->domain); - dom = alias_data->domain; - } - read_unlock_irqrestore(&amd_iommu_devtable_lock, flags); - } - - return dom; -} - -static int device_change_notifier(struct notifier_block *nb, - unsigned long action, void *data) +static int amd_iommu_add_device(struct device *dev) { - struct dma_ops_domain *dma_domain; - struct protection_domain *domain; struct iommu_dev_data *dev_data; - struct device *dev = data; + struct iommu_domain *domain; struct amd_iommu *iommu; - unsigned long flags; u16 devid; + int ret; - if (!check_device(dev)) + if (!check_device(dev) || get_dev_data(dev)) return 0; - devid = get_device_id(dev); - iommu = amd_iommu_rlookup_table[devid]; - dev_data = get_dev_data(dev); - - switch (action) { - case BUS_NOTIFY_ADD_DEVICE: + devid = get_device_id(dev); + iommu = amd_iommu_rlookup_table[devid]; - iommu_init_device(dev); - init_iommu_group(dev); + ret = iommu_init_device(dev); + if (ret) { + if (ret != -ENOTSUPP) + pr_err("Failed to initialize device %s - trying to proceed anyway\n", + dev_name(dev)); - /* - * dev_data is still NULL and - * got initialized in iommu_init_device - */ - dev_data = get_dev_data(dev); + iommu_ignore_device(dev); + dev->archdata.dma_ops = &nommu_dma_ops; + goto out; + } + init_iommu_group(dev); - if (iommu_pass_through || dev_data->iommu_v2) { - dev_data->passthrough = true; - attach_device(dev, pt_domain); - break; - } + dev_data = get_dev_data(dev); - domain = domain_for_device(dev); + BUG_ON(!dev_data); - /* allocate a protection domain if a device is added */ - dma_domain = find_protection_domain(devid); - if (!dma_domain) { - dma_domain = dma_ops_domain_alloc(); - if (!dma_domain) - goto out; - dma_domain->target_dev = devid; - - spin_lock_irqsave(&iommu_pd_list_lock, flags); - list_add_tail(&dma_domain->list, &iommu_pd_list); - spin_unlock_irqrestore(&iommu_pd_list_lock, flags); - } + if (dev_data->iommu_v2) + iommu_request_dm_for_dev(dev); + /* Domains are initialized for this device - have a look what we ended up with */ + domain = iommu_get_domain_for_dev(dev); + if (domain->type == IOMMU_DOMAIN_IDENTITY) { + dev_data->passthrough = true; + dev->archdata.dma_ops = &nommu_dma_ops; + } else { dev->archdata.dma_ops = &amd_iommu_dma_ops; - - break; - case BUS_NOTIFY_REMOVED_DEVICE: - - iommu_uninit_device(dev); - - default: - goto out; } +out: iommu_completion_wait(iommu); -out: return 0; } -static struct notifier_block device_nb = { - .notifier_call = device_change_notifier, -}; - -void amd_iommu_init_notifier(void) +static void amd_iommu_remove_device(struct device *dev) { - bus_register_notifier(&pci_bus_type, &device_nb); + struct amd_iommu *iommu; + u16 devid; + + if (!check_device(dev)) + return; + + devid = get_device_id(dev); + iommu = amd_iommu_rlookup_table[devid]; + + iommu_uninit_device(dev); + iommu_completion_wait(iommu); } /***************************************************************************** @@ -2496,28 +2338,20 @@ void amd_iommu_init_notifier(void) static struct protection_domain *get_domain(struct device *dev) { struct protection_domain *domain; - struct dma_ops_domain *dma_dom; - u16 devid = get_device_id(dev); + struct iommu_domain *io_domain; if (!check_device(dev)) return ERR_PTR(-EINVAL); - domain = domain_for_device(dev); - if (domain != NULL && !dma_ops_domain(domain)) - return ERR_PTR(-EBUSY); - - if (domain != NULL) - return domain; + io_domain = iommu_get_domain_for_dev(dev); + if (!io_domain) + return NULL; - /* Device not bound yet - bind it */ - dma_dom = find_protection_domain(devid); - if (!dma_dom) - dma_dom = amd_iommu_rlookup_table[devid]->default_dom; - attach_device(dev, &dma_dom->domain); - DUMP_printk("Using protection domain %d for device %s\n", - dma_dom->domain.id, dev_name(dev)); + domain = to_pdomain(io_domain); + if (!dma_ops_domain(domain)) + return ERR_PTR(-EBUSY); - return &dma_dom->domain; + return domain; } static void update_device_table(struct protection_domain *domain) @@ -3013,54 +2847,6 @@ static int amd_iommu_dma_supported(struct device *dev, u64 mask) return check_device(dev); } -/* - * The function for pre-allocating protection domains. - * - * If the driver core informs the DMA layer if a driver grabs a device - * we don't need to preallocate the protection domains anymore. - * For now we have to. - */ -static void __init prealloc_protection_domains(void) -{ - struct iommu_dev_data *dev_data; - struct dma_ops_domain *dma_dom; - struct pci_dev *dev = NULL; - u16 devid; - - for_each_pci_dev(dev) { - - /* Do we handle this device? */ - if (!check_device(&dev->dev)) - continue; - - dev_data = get_dev_data(&dev->dev); - if (!amd_iommu_force_isolation && dev_data->iommu_v2) { - /* Make sure passthrough domain is allocated */ - alloc_passthrough_domain(); - dev_data->passthrough = true; - attach_device(&dev->dev, pt_domain); - pr_info("AMD-Vi: Using passthrough domain for device %s\n", - dev_name(&dev->dev)); - } - - /* Is there already any domain for it? */ - if (domain_for_device(&dev->dev)) - continue; - - devid = get_device_id(&dev->dev); - - dma_dom = dma_ops_domain_alloc(); - if (!dma_dom) - continue; - init_unity_mappings_for_device(dma_dom, devid); - dma_dom->target_dev = devid; - - attach_device(&dev->dev, &dma_dom->domain); - - list_add_tail(&dma_dom->list, &iommu_pd_list); - } -} - static struct dma_map_ops amd_iommu_dma_ops = { .alloc = alloc_coherent, .free = free_coherent, @@ -3071,76 +2857,16 @@ static struct dma_map_ops amd_iommu_dma_ops = { .dma_supported = amd_iommu_dma_supported, }; -static unsigned device_dma_ops_init(void) -{ - struct iommu_dev_data *dev_data; - struct pci_dev *pdev = NULL; - unsigned unhandled = 0; - - for_each_pci_dev(pdev) { - if (!check_device(&pdev->dev)) { - - iommu_ignore_device(&pdev->dev); - - unhandled += 1; - continue; - } - - dev_data = get_dev_data(&pdev->dev); - - if (!dev_data->passthrough) - pdev->dev.archdata.dma_ops = &amd_iommu_dma_ops; - else - pdev->dev.archdata.dma_ops = &nommu_dma_ops; - } - - return unhandled; -} - -/* - * The function which clues the AMD IOMMU driver into dma_ops. - */ - -void __init amd_iommu_init_api(void) +int __init amd_iommu_init_api(void) { - bus_set_iommu(&pci_bus_type, &amd_iommu_ops); + return bus_set_iommu(&pci_bus_type, &amd_iommu_ops); } int __init amd_iommu_init_dma_ops(void) { - struct amd_iommu *iommu; - int ret, unhandled; - - /* - * first allocate a default protection domain for every IOMMU we - * found in the system. Devices not assigned to any other - * protection domain will be assigned to the default one. - */ - for_each_iommu(iommu) { - iommu->default_dom = dma_ops_domain_alloc(); - if (iommu->default_dom == NULL) - return -ENOMEM; - iommu->default_dom->domain.flags |= PD_DEFAULT_MASK; - ret = iommu_init_unity_mappings(iommu); - if (ret) - goto free_domains; - } - - /* - * Pre-allocate the protection domains for each device. - */ - prealloc_protection_domains(); - iommu_detected = 1; swiotlb = 0; - /* Make the driver finally visible to the drivers */ - unhandled = device_dma_ops_init(); - if (unhandled && max_pfn > MAX_DMA32_PFN) { - /* There are unhandled devices - initialize swiotlb for them */ - swiotlb = 1; - } - amd_iommu_stats_init(); if (amd_iommu_unmap_flush) @@ -3149,14 +2875,6 @@ int __init amd_iommu_init_dma_ops(void) pr_info("AMD-Vi: Lazy IO/TLB flushing enabled\n"); return 0; - -free_domains: - - for_each_iommu(iommu) { - dma_ops_domain_free(iommu->default_dom); - } - - return ret; } /***************************************************************************** @@ -3223,7 +2941,7 @@ out_err: return NULL; } -static int __init alloc_passthrough_domain(void) +static int alloc_passthrough_domain(void) { if (pt_domain != NULL) return 0; @@ -3241,30 +2959,46 @@ static int __init alloc_passthrough_domain(void) static struct iommu_domain *amd_iommu_domain_alloc(unsigned type) { struct protection_domain *pdomain; + struct dma_ops_domain *dma_domain; - /* We only support unmanaged domains for now */ - if (type != IOMMU_DOMAIN_UNMANAGED) - return NULL; - - pdomain = protection_domain_alloc(); - if (!pdomain) - goto out_free; + switch (type) { + case IOMMU_DOMAIN_UNMANAGED: + pdomain = protection_domain_alloc(); + if (!pdomain) + return NULL; - pdomain->mode = PAGE_MODE_3_LEVEL; - pdomain->pt_root = (void *)get_zeroed_page(GFP_KERNEL); - if (!pdomain->pt_root) - goto out_free; + pdomain->mode = PAGE_MODE_3_LEVEL; + pdomain->pt_root = (void *)get_zeroed_page(GFP_KERNEL); + if (!pdomain->pt_root) { + protection_domain_free(pdomain); + return NULL; + } - pdomain->domain.geometry.aperture_start = 0; - pdomain->domain.geometry.aperture_end = ~0ULL; - pdomain->domain.geometry.force_aperture = true; + pdomain->domain.geometry.aperture_start = 0; + pdomain->domain.geometry.aperture_end = ~0ULL; + pdomain->domain.geometry.force_aperture = true; - return &pdomain->domain; + break; + case IOMMU_DOMAIN_DMA: + dma_domain = dma_ops_domain_alloc(); + if (!dma_domain) { + pr_err("AMD-Vi: Failed to allocate\n"); + return NULL; + } + pdomain = &dma_domain->domain; + break; + case IOMMU_DOMAIN_IDENTITY: + pdomain = protection_domain_alloc(); + if (!pdomain) + return NULL; -out_free: - protection_domain_free(pdomain); + pdomain->mode = PAGE_MODE_NONE; + break; + default: + return NULL; + } - return NULL; + return &pdomain->domain; } static void amd_iommu_domain_free(struct iommu_domain *dom) @@ -3414,6 +3148,47 @@ static bool amd_iommu_capable(enum iommu_cap cap) return false; } +static void amd_iommu_get_dm_regions(struct device *dev, + struct list_head *head) +{ + struct unity_map_entry *entry; + u16 devid; + + devid = get_device_id(dev); + + list_for_each_entry(entry, &amd_iommu_unity_map, list) { + struct iommu_dm_region *region; + + if (devid < entry->devid_start || devid > entry->devid_end) + continue; + + region = kzalloc(sizeof(*region), GFP_KERNEL); + if (!region) { + pr_err("Out of memory allocating dm-regions for %s\n", + dev_name(dev)); + return; + } + + region->start = entry->address_start; + region->length = entry->address_end - entry->address_start; + if (entry->prot & IOMMU_PROT_IR) + region->prot |= IOMMU_READ; + if (entry->prot & IOMMU_PROT_IW) + region->prot |= IOMMU_WRITE; + + list_add_tail(®ion->list, head); + } +} + +static void amd_iommu_put_dm_regions(struct device *dev, + struct list_head *head) +{ + struct iommu_dm_region *entry, *next; + + list_for_each_entry_safe(entry, next, head, list) + kfree(entry); +} + static const struct iommu_ops amd_iommu_ops = { .capable = amd_iommu_capable, .domain_alloc = amd_iommu_domain_alloc, @@ -3424,6 +3199,10 @@ static const struct iommu_ops amd_iommu_ops = { .unmap = amd_iommu_unmap, .map_sg = default_iommu_map_sg, .iova_to_phys = amd_iommu_iova_to_phys, + .add_device = amd_iommu_add_device, + .remove_device = amd_iommu_remove_device, + .get_dm_regions = amd_iommu_get_dm_regions, + .put_dm_regions = amd_iommu_put_dm_regions, .pgsize_bitmap = AMD_IOMMU_PGSIZES, }; |