diff options
Diffstat (limited to 'drivers/iommu')
| -rw-r--r-- | drivers/iommu/Kconfig | 19 | ||||
| -rw-r--r-- | drivers/iommu/Makefile | 1 | ||||
| -rw-r--r-- | drivers/iommu/amd_iommu.c | 1032 | ||||
| -rw-r--r-- | drivers/iommu/amd_iommu_init.c | 16 | ||||
| -rw-r--r-- | drivers/iommu/amd_iommu_types.h | 1 | ||||
| -rw-r--r-- | drivers/iommu/amd_iommu_v2.c | 2 | ||||
| -rw-r--r-- | drivers/iommu/arm-smmu-v3.c | 2 | ||||
| -rw-r--r-- | drivers/iommu/arm-smmu.c | 28 | ||||
| -rw-r--r-- | drivers/iommu/dmar.c | 21 | ||||
| -rw-r--r-- | drivers/iommu/exynos-iommu.c | 107 | ||||
| -rw-r--r-- | drivers/iommu/intel-iommu.c | 12 | ||||
| -rw-r--r-- | drivers/iommu/io-pgtable-arm.c | 2 | ||||
| -rw-r--r-- | drivers/iommu/iommu.c | 32 | ||||
| -rw-r--r-- | drivers/iommu/iova.c | 8 | ||||
| -rw-r--r-- | drivers/iommu/mtk_iommu.c | 49 | ||||
| -rw-r--r-- | drivers/iommu/mtk_iommu.h | 77 | ||||
| -rw-r--r-- | drivers/iommu/mtk_iommu_v1.c | 727 | ||||
| -rw-r--r-- | drivers/iommu/rockchip-iommu.c | 181 |
18 files changed, 1505 insertions, 812 deletions
diff --git a/drivers/iommu/Kconfig b/drivers/iommu/Kconfig index f5f4d1472771..d432ca828472 100644 --- a/drivers/iommu/Kconfig +++ b/drivers/iommu/Kconfig @@ -111,6 +111,7 @@ config AMD_IOMMU select PCI_PRI select PCI_PASID select IOMMU_API + select IOMMU_IOVA depends on X86_64 && PCI && ACPI ---help--- With this option you can enable support for AMD IOMMU hardware in @@ -343,4 +344,22 @@ config MTK_IOMMU If unsure, say N here. +config MTK_IOMMU_V1 + bool "MTK IOMMU Version 1 (M4U gen1) Support" + depends on ARM + depends on ARCH_MEDIATEK || COMPILE_TEST + select ARM_DMA_USE_IOMMU + select IOMMU_API + select MEMORY + select MTK_SMI + select COMMON_CLK_MT2701_MMSYS + select COMMON_CLK_MT2701_IMGSYS + select COMMON_CLK_MT2701_VDECSYS + help + Support for the M4U on certain Mediatek SoCs. M4U generation 1 HW is + Multimedia Memory Managememt Unit. This option enables remapping of + DMA memory accesses for the multimedia subsystem. + + if unsure, say N here. + endif # IOMMU_SUPPORT diff --git a/drivers/iommu/Makefile b/drivers/iommu/Makefile index 7fe479fb2770..195f7b997d8e 100644 --- a/drivers/iommu/Makefile +++ b/drivers/iommu/Makefile @@ -18,6 +18,7 @@ obj-$(CONFIG_INTEL_IOMMU_SVM) += intel-svm.o obj-$(CONFIG_IPMMU_VMSA) += ipmmu-vmsa.o obj-$(CONFIG_IRQ_REMAP) += intel_irq_remapping.o irq_remapping.o obj-$(CONFIG_MTK_IOMMU) += mtk_iommu.o +obj-$(CONFIG_MTK_IOMMU_V1) += mtk_iommu_v1.o obj-$(CONFIG_OMAP_IOMMU) += omap-iommu.o obj-$(CONFIG_OMAP_IOMMU_DEBUG) += omap-iommu-debug.o obj-$(CONFIG_ROCKCHIP_IOMMU) += rockchip-iommu.o diff --git a/drivers/iommu/amd_iommu.c b/drivers/iommu/amd_iommu.c index 634f636393d5..33c177ba93be 100644 --- a/drivers/iommu/amd_iommu.c +++ b/drivers/iommu/amd_iommu.c @@ -21,6 +21,7 @@ #include <linux/pci.h> #include <linux/acpi.h> #include <linux/amba/bus.h> +#include <linux/platform_device.h> #include <linux/pci-ats.h> #include <linux/bitmap.h> #include <linux/slab.h> @@ -38,6 +39,7 @@ #include <linux/dma-contiguous.h> #include <linux/irqdomain.h> #include <linux/percpu.h> +#include <linux/iova.h> #include <asm/irq_remapping.h> #include <asm/io_apic.h> #include <asm/apic.h> @@ -56,6 +58,17 @@ #define LOOP_TIMEOUT 100000 +/* IO virtual address start page frame number */ +#define IOVA_START_PFN (1) +#define IOVA_PFN(addr) ((addr) >> PAGE_SHIFT) +#define DMA_32BIT_PFN IOVA_PFN(DMA_BIT_MASK(32)) + +/* Reserved IOVA ranges */ +#define MSI_RANGE_START (0xfee00000) +#define MSI_RANGE_END (0xfeefffff) +#define HT_RANGE_START (0xfd00000000ULL) +#define HT_RANGE_END (0xffffffffffULL) + /* * This bitmap is used to advertise the page sizes our hardware support * to the IOMMU core, which will then use this information to split @@ -76,6 +89,25 @@ LIST_HEAD(ioapic_map); LIST_HEAD(hpet_map); LIST_HEAD(acpihid_map); +#define FLUSH_QUEUE_SIZE 256 + +struct flush_queue_entry { + unsigned long iova_pfn; + unsigned long pages; + struct dma_ops_domain *dma_dom; +}; + +struct flush_queue { + spinlock_t lock; + unsigned next; + struct flush_queue_entry *entries; +}; + +DEFINE_PER_CPU(struct flush_queue, flush_queue); + +static atomic_t queue_timer_on; +static struct timer_list queue_timer; + /* * Domain for untranslated devices - only allocated * if iommu=pt passed on kernel cmd line. @@ -121,44 +153,19 @@ static int protection_domain_init(struct protection_domain *domain); static void detach_device(struct device *dev); /* - * For dynamic growth the aperture size is split into ranges of 128MB of - * DMA address space each. This struct represents one such range. - */ -struct aperture_range { - - spinlock_t bitmap_lock; - - /* address allocation bitmap */ - unsigned long *bitmap; - unsigned long offset; - unsigned long next_bit; - - /* - * Array of PTE pages for the aperture. In this array we save all the - * leaf pages of the domain page table used for the aperture. This way - * we don't need to walk the page table to find a specific PTE. We can - * just calculate its address in constant time. - */ - u64 *pte_pages[64]; -}; - -/* * Data container for a dma_ops specific protection domain */ struct dma_ops_domain { /* generic protection domain information */ struct protection_domain domain; - /* size of the aperture for the mappings */ - unsigned long aperture_size; - - /* aperture index we start searching for free addresses */ - u32 __percpu *next_index; - - /* address space relevant data */ - struct aperture_range *aperture[APERTURE_MAX_RANGES]; + /* IOVA RB-Tree */ + struct iova_domain iovad; }; +static struct iova_domain reserved_iova_ranges; +static struct lock_class_key reserved_rbtree_key; + /**************************************************************************** * * Helper functions @@ -224,6 +231,12 @@ static struct protection_domain *to_pdomain(struct iommu_domain *dom) return container_of(dom, struct protection_domain, domain); } +static struct dma_ops_domain* to_dma_ops_domain(struct protection_domain *domain) +{ + BUG_ON(domain->flags != PD_DMA_OPS_MASK); + return container_of(domain, struct dma_ops_domain, domain); +} + static struct iommu_dev_data *alloc_dev_data(u16 devid) { struct iommu_dev_data *dev_data; @@ -391,43 +404,6 @@ static bool pdev_pri_erratum(struct pci_dev *pdev, u32 erratum) } /* - * This function actually applies the mapping to the page table of the - * dma_ops domain. - */ -static void alloc_unity_mapping(struct dma_ops_domain *dma_dom, - struct unity_map_entry *e) -{ - u64 addr; - - 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; - int devid; - - devid = get_device_id(dev); - if (devid < 0) - return; - - 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); - } -} - -/* * This function checks if the driver got a valid device from the caller to * avoid dereferencing invalid pointers. */ @@ -454,22 +430,12 @@ 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)) 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); } @@ -1220,7 +1186,7 @@ static void domain_flush_complete(struct protection_domain *domain) int i; for (i = 0; i < amd_iommus_present; ++i) { - if (!domain->dev_iommu[i]) + if (domain && !domain->dev_iommu[i]) continue; /* @@ -1397,8 +1363,9 @@ static u64 *fetch_pte(struct protection_domain *domain, static int iommu_map_page(struct protection_domain *dom, unsigned long bus_addr, unsigned long phys_addr, + unsigned long page_size, int prot, - unsigned long page_size) + gfp_t gfp) { u64 __pte, *pte; int i, count; @@ -1410,7 +1377,7 @@ static int iommu_map_page(struct protection_domain *dom, return -EINVAL; count = PAGE_SIZE_PTE_COUNT(page_size); - pte = alloc_pte(dom, bus_addr, page_size, NULL, GFP_KERNEL); + pte = alloc_pte(dom, bus_addr, page_size, NULL, gfp); if (!pte) return -ENOMEM; @@ -1474,320 +1441,37 @@ static unsigned long iommu_unmap_page(struct protection_domain *dom, /**************************************************************************** * * The next functions belong to the address allocator for the dma_ops - * interface functions. They work like the allocators in the other IOMMU - * drivers. Its basically a bitmap which marks the allocated pages in - * the aperture. Maybe it could be enhanced in the future to a more - * efficient allocator. + * interface functions. * ****************************************************************************/ -/* - * The address allocator core functions. - * - * called with domain->lock held - */ -/* - * Used to reserve address ranges in the aperture (e.g. for exclusion - * ranges. - */ -static void dma_ops_reserve_addresses(struct dma_ops_domain *dom, - unsigned long start_page, - unsigned int pages) +static unsigned long dma_ops_alloc_iova(struct device *dev, + struct dma_ops_domain *dma_dom, + unsigned int pages, u64 dma_mask) { - unsigned int i, last_page = dom->aperture_size >> PAGE_SHIFT; - - if (start_page + pages > last_page) - pages = last_page - start_page; - - for (i = start_page; i < start_page + pages; ++i) { - int index = i / APERTURE_RANGE_PAGES; - int page = i % APERTURE_RANGE_PAGES; - __set_bit(page, dom->aperture[index]->bitmap); - } -} + unsigned long pfn = 0; -/* - * This function is used to add a new aperture range to an existing - * aperture in case of dma_ops domain allocation or address allocation - * failure. - */ -static int alloc_new_range(struct dma_ops_domain *dma_dom, - bool populate, gfp_t gfp) -{ - int index = dma_dom->aperture_size >> APERTURE_RANGE_SHIFT; - unsigned long i, old_size, pte_pgsize; - struct aperture_range *range; - struct amd_iommu *iommu; - unsigned long flags; + pages = __roundup_pow_of_two(pages); -#ifdef CONFIG_IOMMU_STRESS - populate = false; -#endif + if (dma_mask > DMA_BIT_MASK(32)) + pfn = alloc_iova_fast(&dma_dom->iovad, pages, + IOVA_PFN(DMA_BIT_MASK(32))); - if (index >= APERTURE_MAX_RANGES) - return -ENOMEM; - - range = kzalloc(sizeof(struct aperture_range), gfp); - if (!range) - return -ENOMEM; - - range->bitmap = (void *)get_zeroed_page(gfp); - if (!range->bitmap) - goto out_free; - - range->offset = dma_dom->aperture_size; - - spin_lock_init(&range->bitmap_lock); - - if (populate) { - unsigned long address = dma_dom->aperture_size; - int i, num_ptes = APERTURE_RANGE_PAGES / 512; - u64 *pte, *pte_page; - - for (i = 0; i < num_ptes; ++i) { - pte = alloc_pte(&dma_dom->domain, address, PAGE_SIZE, - &pte_page, gfp); - if (!pte) - goto out_free; - - range->pte_pages[i] = pte_page; - - address += APERTURE_RANGE_SIZE / 64; - } - } + if (!pfn) + pfn = alloc_iova_fast(&dma_dom->iovad, pages, IOVA_PFN(dma_mask)); - spin_lock_irqsave(&dma_dom->domain.lock, flags); - - /* First take the bitmap_lock and then publish the range */ - spin_lock(&range->bitmap_lock); - - old_size = dma_dom->aperture_size; - dma_dom->aperture[index] = range; - dma_dom->aperture_size += APERTURE_RANGE_SIZE; - - /* Reserve address range used for MSI messages */ - if (old_size < MSI_ADDR_BASE_LO && - dma_dom->aperture_size > MSI_ADDR_BASE_LO) { - unsigned long spage; - int pages; - - pages = iommu_num_pages(MSI_ADDR_BASE_LO, 0x10000, PAGE_SIZE); - spage = MSI_ADDR_BASE_LO >> PAGE_SHIFT; - - dma_ops_reserve_addresses(dma_dom, spage, pages); - } - - /* Initialize the exclusion range if necessary */ - for_each_iommu(iommu) { - if (iommu->exclusion_start && - iommu->exclusion_start >= dma_dom->aperture[index]->offset - && iommu->exclusion_start < dma_dom->aperture_size) { - unsigned long startpage; - int pages = iommu_num_pages(iommu->exclusion_start, - iommu->exclusion_length, - PAGE_SIZE); - startpage = iommu->exclusion_start >> PAGE_SHIFT; - dma_ops_reserve_addresses(dma_dom, startpage, pages); - } - } - - /* - * Check for areas already mapped as present in the new aperture - * range and mark those pages as reserved in the allocator. Such - * mappings may already exist as a result of requested unity - * mappings for devices. - */ - for (i = dma_dom->aperture[index]->offset; - i < dma_dom->aperture_size; - i += pte_pgsize) { - u64 *pte = fetch_pte(&dma_dom->domain, i, &pte_pgsize); - if (!pte || !IOMMU_PTE_PRESENT(*pte)) - continue; - - dma_ops_reserve_addresses(dma_dom, i >> PAGE_SHIFT, - pte_pgsize >> 12); - } - - update_domain(&dma_dom->domain); - - spin_unlock(&range->bitmap_lock); - - spin_unlock_irqrestore(&dma_dom->domain.lock, flags); - - return 0; - -out_free: - update_domain(&dma_dom->domain); - - free_page((unsigned long)range->bitmap); - - kfree(range); - - return -ENOMEM; + return (pfn << PAGE_SHIFT); } -static dma_addr_t dma_ops_aperture_alloc(struct dma_ops_domain *dom, - struct aperture_range *range, - unsigned long pages, - unsigned long dma_mask, - unsigned long boundary_size, - unsigned long align_mask, - bool trylock) +static void dma_ops_free_iova(struct dma_ops_domain *dma_dom, + unsigned long address, + unsigned int pages) { - unsigned long offset, limit, flags; - dma_addr_t address; - bool flush = false; - - offset = range->offset >> PAGE_SHIFT; - limit = iommu_device_max_index(APERTURE_RANGE_PAGES, offset, - dma_mask >> PAGE_SHIFT); - - if (trylock) { - if (!spin_trylock_irqsave(&range->bitmap_lock, flags)) - return -1; - } else { - spin_lock_irqsave(&range->bitmap_lock, flags); - } - - address = iommu_area_alloc(range->bitmap, limit, range->next_bit, - pages, offset, boundary_size, align_mask); - if (address == -1) { - /* Nothing found, retry one time */ - address = iommu_area_alloc(range->bitmap, limit, - 0, pages, offset, boundary_size, - align_mask); - flush = true; - } - - if (address != -1) - range->next_bit = address + pages; - - spin_unlock_irqrestore(&range->bitmap_lock, flags); - - if (flush) { - domain_flush_tlb(&dom->domain); - domain_flush_complete(&dom->domain); - } - - return address; -} - -static unsigned long dma_ops_area_alloc(struct device *dev, - struct dma_ops_domain *dom, - unsigned int pages, - unsigned long align_mask, - u64 dma_mask) -{ - unsigned long boundary_size, mask; - unsigned long address = -1; - bool first = true; - u32 start, i; - - preempt_disable(); - - mask = dma_get_seg_boundary(dev); - -again: - start = this_cpu_read(*dom->next_index); - - /* Sanity check - is it really necessary? */ - if (unlikely(start > APERTURE_MAX_RANGES)) { - start = 0; - this_cpu_write(*dom->next_index, 0); - } - - boundary_size = mask + 1 ? ALIGN(mask + 1, PAGE_SIZE) >> PAGE_SHIFT : - 1UL << (BITS_PER_LONG - PAGE_SHIFT); - - for (i = 0; i < APERTURE_MAX_RANGES; ++i) { - struct aperture_range *range; - int index; - - index = (start + i) % APERTURE_MAX_RANGES; - - range = dom->aperture[index]; - - if (!range || range->offset >= dma_mask) - continue; - - address = dma_ops_aperture_alloc(dom, range, pages, - dma_mask, boundary_size, - align_mask, first); - if (address != -1) { - address = range->offset + (address << PAGE_SHIFT); - this_cpu_write(*dom->next_index, index); - break; - } - } - - if (address == -1 && first) { - first = false; - goto again; - } - - preempt_enable(); - - return address; -} - -static unsigned long dma_ops_alloc_addresses(struct device *dev, - struct dma_ops_domain *dom, - unsigned int pages, - unsigned long align_mask, - u64 dma_mask) -{ - unsigned long address = -1; - - while (address == -1) { - address = dma_ops_area_alloc(dev, dom, pages, - align_mask, dma_mask); - - if (address == -1 && alloc_new_range(dom, false, GFP_ATOMIC)) - break; - } - - if (unlikely(address == -1)) - address = DMA_ERROR_CODE; - - WARN_ON((address + (PAGE_SIZE*pages)) > dom->aperture_size); - - return address; -} - -/* - * The address free function. - * - * called with domain->lock held - */ -static void dma_ops_free_addresses(struct dma_ops_domain *dom, - unsigned long address, - unsigned int pages) -{ - unsigned i = address >> APERTURE_RANGE_SHIFT; - struct aperture_range *range = dom->aperture[i]; - unsigned long flags; - - BUG_ON(i >= APERTURE_MAX_RANGES || range == NULL); - -#ifdef CONFIG_IOMMU_STRESS - if (i < 4) - return; -#endif - - if (amd_iommu_unmap_flush) { - domain_flush_tlb(&dom->domain); - domain_flush_complete(&dom->domain); - } - - address = (address % APERTURE_RANGE_SIZE) >> PAGE_SHIFT; - - spin_lock_irqsave(&range->bitmap_lock, flags); - if (address + pages > range->next_bit) - range->next_bit = address + pages; - bitmap_clear(range->bitmap, address, pages); - spin_unlock_irqrestore(&range->bitmap_lock, flags); + pages = __roundup_pow_of_two(pages); + address >>= PAGE_SHIFT; + free_iova_fast(&dma_dom->iovad, address, pages); } /**************************************************************************** @@ -1961,44 +1645,18 @@ static void free_gcr3_table(struct protection_domain *domain) */ static void dma_ops_domain_free(struct dma_ops_domain *dom) { - int i; - if (!dom) return; - free_percpu(dom->next_index); - del_domain_from_list(&dom->domain); - free_pagetable(&dom->domain); + put_iova_domain(&dom->iovad); - for (i = 0; i < APERTURE_MAX_RANGES; ++i) { - if (!dom->aperture[i]) - continue; - free_page((unsigned long)dom->aperture[i]->bitmap); - kfree(dom->aperture[i]); - } + free_pagetable(&dom->domain); kfree(dom); } -static int dma_ops_domain_alloc_apertures(struct dma_ops_domain *dma_dom, - int max_apertures) -{ - int ret, i, apertures; - - apertures = dma_dom->aperture_size >> APERTURE_RANGE_SHIFT; - ret = 0; - - for (i = apertures; i < max_apertures; ++i) { - ret = alloc_new_range(dma_dom, false, GFP_KERNEL); - if (ret) - break; - } - - return ret; -} - /* * Allocates a new protection domain usable for the dma_ops functions. * It also initializes the page table and the address allocator data @@ -2007,7 +1665,6 @@ static int dma_ops_domain_alloc_apertures(struct dma_ops_domain *dma_dom, static struct dma_ops_domain *dma_ops_domain_alloc(void) { struct dma_ops_domain *dma_dom; - int cpu; dma_dom = kzalloc(sizeof(struct dma_ops_domain), GFP_KERNEL); if (!dma_dom) @@ -2016,30 +1673,19 @@ static struct dma_ops_domain *dma_ops_domain_alloc(void) if (protection_domain_init(&dma_dom->domain)) goto free_dma_dom; - dma_dom->next_index = alloc_percpu(u32); - if (!dma_dom->next_index) - goto free_dma_dom; - - dma_dom->domain.mode = PAGE_MODE_2_LEVEL; + dma_dom->domain.mode = PAGE_MODE_3_LEVEL; dma_dom->domain.pt_root = (void *)get_zeroed_page(GFP_KERNEL); dma_dom->domain.flags = PD_DMA_OPS_MASK; - dma_dom->domain.priv = dma_dom; if (!dma_dom->domain.pt_root) goto free_dma_dom; - add_domain_to_list(&dma_dom->domain); + init_iova_domain(&dma_dom->iovad, PAGE_SIZE, + IOVA_START_PFN, DMA_32BIT_PFN); - if (alloc_new_range(dma_dom, true, GFP_KERNEL)) - goto free_dma_dom; - - /* - * mark the first page as allocated so we never return 0 as - * a valid dma-address. So we can use 0 as error value - */ - dma_dom->aperture[0]->bitmap[0] = 1; + /* Initialize reserved ranges */ + copy_reserved_iova(&reserved_iova_ranges, &dma_dom->iovad); - for_each_possible_cpu(cpu) - *per_cpu_ptr(dma_dom->next_index, cpu) = 0; + add_domain_to_list(&dma_dom->domain); return dma_dom; @@ -2482,6 +2128,92 @@ static struct iommu_group *amd_iommu_device_group(struct device *dev) * *****************************************************************************/ +static void __queue_flush(struct flush_queue *queue) +{ + struct protection_domain *domain; + unsigned long flags; + int idx; + + /* First flush TLB of all known domains */ + spin_lock_irqsave(&amd_iommu_pd_lock, flags); + list_for_each_entry(domain, &amd_iommu_pd_list, list) + domain_flush_tlb(domain); + spin_unlock_irqrestore(&amd_iommu_pd_lock, flags); + + /* Wait until flushes have completed */ + domain_flush_complete(NULL); + + for (idx = 0; idx < queue->next; ++idx) { + struct flush_queue_entry *entry; + + entry = queue->entries + idx; + + free_iova_fast(&entry->dma_dom->iovad, + entry->iova_pfn, + entry->pages); + + /* Not really necessary, just to make sure we catch any bugs */ + entry->dma_dom = NULL; + } + + queue->next = 0; +} + +static void queue_flush_all(void) +{ + int cpu; + + for_each_possible_cpu(cpu) { + struct flush_queue *queue; + unsigned long flags; + + queue = per_cpu_ptr(&flush_queue, cpu); + spin_lock_irqsave(&queue->lock, flags); + if (queue->next > 0) + __queue_flush(queue); + spin_unlock_irqrestore(&queue->lock, flags); + } +} + +static void queue_flush_timeout(unsigned long unsused) +{ + atomic_set(&queue_timer_on, 0); + queue_flush_all(); +} + +static void queue_add(struct dma_ops_domain *dma_dom, + unsigned long address, unsigned long pages) +{ + struct flush_queue_entry *entry; + struct flush_queue *queue; + unsigned long flags; + int idx; + + pages = __roundup_pow_of_two(pages); + address >>= PAGE_SHIFT; + + queue = get_cpu_ptr(&flush_queue); + spin_lock_irqsave(&queue->lock, flags); + + if (queue->next == FLUSH_QUEUE_SIZE) + __queue_flush(queue); + + idx = queue->next++; + entry = queue->entries + idx; + + entry->iova_pfn = address; + entry->pages = pages; + entry->dma_dom = dma_dom; + + spin_unlock_irqrestore(&queue->lock, flags); + + if (atomic_cmpxchg(&queue_timer_on, 0, 1) == 0) + mod_timer(&queue_timer, jiffies + msecs_to_jiffies(10)); + + put_cpu_ptr(&flush_queue); +} + + /* * In the dma_ops path we only have the struct device. This function * finds the corresponding IOMMU, the protection domain and the @@ -2492,16 +2224,11 @@ static struct iommu_group *amd_iommu_device_group(struct device *dev) static struct protection_domain *get_domain(struct device *dev) { struct protection_domain *domain; - struct iommu_domain *io_domain; if (!check_device(dev)) return ERR_PTR(-EINVAL); - io_domain = iommu_get_domain_for_dev(dev); - if (!io_domain) - return NULL; - - domain = to_pdomain(io_domain); + domain = get_dev_data(dev)->domain; if (!dma_ops_domain(domain)) return ERR_PTR(-EBUSY); @@ -2512,8 +2239,15 @@ static void update_device_table(struct protection_domain *domain) { struct iommu_dev_data *dev_data; - list_for_each_entry(dev_data, &domain->dev_list, list) + list_for_each_entry(dev_data, &domain->dev_list, list) { set_dte_entry(dev_data->devid, domain, dev_data->ats.enabled); + + if (dev_data->devid == dev_data->alias) + continue; + + /* There is an alias, update device table entry for it */ + set_dte_entry(dev_data->alias, domain, dev_data->ats.enabled); + } } static void update_domain(struct protection_domain *domain) @@ -2529,94 +2263,17 @@ static void update_domain(struct protection_domain *domain) domain->updated = false; } -/* - * This function fetches the PTE for a given address in the aperture - */ -static u64* dma_ops_get_pte(struct dma_ops_domain *dom, - unsigned long address) -{ - struct aperture_range *aperture; - u64 *pte, *pte_page; - - aperture = dom->aperture[APERTURE_RANGE_INDEX(address)]; - if (!aperture) - return NULL; - - pte = aperture->pte_pages[APERTURE_PAGE_INDEX(address)]; - if (!pte) { - pte = alloc_pte(&dom->domain, address, PAGE_SIZE, &pte_page, - GFP_ATOMIC); - aperture->pte_pages[APERTURE_PAGE_INDEX(address)] = pte_page; - } else - pte += PM_LEVEL_INDEX(0, address); - - update_domain(&dom->domain); - - return pte; -} - -/* - * This is the generic map function. It maps one 4kb page at paddr to - * the given address in the DMA address space for the domain. - */ -static dma_addr_t dma_ops_domain_map(struct dma_ops_domain *dom, - unsigned long address, - phys_addr_t paddr, - int direction) +static int dir2prot(enum dma_data_direction direction) { - u64 *pte, __pte; - - WARN_ON(address > dom->aperture_size); - - paddr &= PAGE_MASK; - - pte = dma_ops_get_pte(dom, address); - if (!pte) - return DMA_ERROR_CODE; - - __pte = paddr | IOMMU_PTE_P | IOMMU_PTE_FC; - if (direction == DMA_TO_DEVICE) - __pte |= IOMMU_PTE_IR; + return IOMMU_PROT_IR; else if (direction == DMA_FROM_DEVICE) - __pte |= IOMMU_PTE_IW; + return IOMMU_PROT_IW; else if (direction == DMA_BIDIRECTIONAL) - __pte |= IOMMU_PTE_IR | IOMMU_PTE_IW; - - WARN_ON_ONCE(*pte); - - *pte = __pte; - - return (dma_addr_t)address; -} - -/* - * The generic unmapping function for on page in the DMA address space. - */ -static void dma_ops_domain_unmap(struct dma_ops_domain *dom, - unsigned long address) -{ - struct aperture_range *aperture; - u64 *pte; - - if (address >= dom->aperture_size) - return; - - aperture = dom->aperture[APERTURE_RANGE_INDEX(address)]; - if (!aperture) - return; - - pte = aperture->pte_pages[APERTURE_PAGE_INDEX(address)]; - if (!pte) - return; - - pte += PM_LEVEL_INDEX(0, address); - - WARN_ON_ONCE(!*pte); - - *pte = 0ULL; + return IOMMU_PROT_IW | IOMMU_PROT_IR; + else + return 0; } - /* * This function contains common code for mapping of a physically * contiguous memory region into DMA address space. It is used by all @@ -2627,32 +2284,29 @@ static dma_addr_t __map_single(struct device *dev, struct dma_ops_domain *dma_dom, phys_addr_t paddr, size_t size, - int dir, - bool align, + enum dma_data_direction direction, u64 dma_mask) { dma_addr_t offset = paddr & ~PAGE_MASK; dma_addr_t address, start, ret; unsigned int pages; - unsigned long align_mask = 0; + int prot = 0; int i; pages = iommu_num_pages(paddr, size, PAGE_SIZE); paddr &= PAGE_MASK; - if (align) - align_mask = (1UL << get_order(size)) - 1; - - address = dma_ops_alloc_addresses(dev, dma_dom, pages, align_mask, - dma_mask); - + address = dma_ops_alloc_iova(dev, dma_dom, pages, dma_mask); if (address == DMA_ERROR_CODE) goto out; + prot = dir2prot(direction); + start = address; for (i = 0; i < pages; ++i) { - ret = dma_ops_domain_map(dma_dom, start, paddr, dir); - if (ret == DMA_ERROR_CODE) + ret = iommu_map_page(&dma_dom->domain, start, paddr, + PAGE_SIZE, prot, GFP_ATOMIC); + if (ret) goto out_unmap; paddr += PAGE_SIZE; @@ -2672,10 +2326,13 @@ out_unmap: for (--i; i >= 0; --i) { start -= PAGE_SIZE; - dma_ops_domain_unmap(dma_dom, start); + iommu_unmap_page(&dma_dom->domain, start, PAGE_SIZE); } - dma_ops_free_addresses(dma_dom, address, pages); + domain_flush_tlb(&dma_dom->domain); + domain_flush_complete(&dma_dom->domain); + + dma_ops_free_iova(dma_dom, address, pages); return DMA_ERROR_CODE; } @@ -2693,21 +2350,23 @@ static void __unmap_single(struct dma_ops_domain *dma_dom, dma_addr_t i, start; unsigned int pages; - if ((dma_addr == DMA_ERROR_CODE) || - (dma_addr + size > dma_dom->aperture_size)) - return; - flush_addr = dma_addr; pages = iommu_num_pages(dma_addr, size, PAGE_SIZE); dma_addr &= PAGE_MASK; start = dma_addr; for (i = 0; i < pages; ++i) { - dma_ops_domain_unmap(dma_dom, start); + iommu_unmap_page(&dma_dom->domain, start, PAGE_SIZE); start += PAGE_SIZE; } - dma_ops_free_addresses(dma_dom, dma_addr, pages); + if (amd_iommu_unmap_flush) { + dma_ops_free_iova(dma_dom, dma_addr, pages); + domain_flush_tlb(&dma_dom->domain); + domain_flush_complete(&dma_dom->domain); + } else { + queue_add(dma_dom, dma_addr, pages); + } } /* @@ -2720,6 +2379,7 @@ static dma_addr_t map_page(struct device *dev, struct page *page, { phys_addr_t paddr = page_to_phys(page) + offset; struct protection_domain *domain; + struct dma_ops_domain *dma_dom; u64 dma_mask; domain = get_domain(dev); @@ -2729,9 +2389,9 @@ static dma_addr_t map_page(struct device *dev, struct page *page, return DMA_ERROR_CODE; dma_mask = *dev->dma_mask; + dma_dom = to_dma_ops_domain(domain); - return __map_single(dev, domain->priv, paddr, size, dir, false, - dma_mask); + return __map_single(dev, dma_dom, paddr, size, dir, dma_mask); } /* @@ -2741,12 +2401,41 @@ static void unmap_page(struct device *dev, dma_addr_t dma_addr, size_t size, enum dma_data_direction dir, struct dma_attrs *attrs) { struct protection_domain *domain; + struct dma_ops_domain *dma_dom; domain = get_domain(dev); if (IS_ERR(domain)) return; - __unmap_single(domain->priv, dma_addr, size, dir); + dma_dom = to_dma_ops_domain(domain); + + __unmap_single(dma_dom, dma_addr, size, dir); +} + +static int sg_num_pages(struct device *dev, + struct scatterlist *sglist, + int nelems) +{ + unsigned long mask, boundary_size; + struct scatterlist *s; + int i, npages = 0; + + mask = dma_get_seg_boundary(dev); + boundary_size = mask + 1 ? ALIGN(mask + 1, PAGE_SIZE) >> PAGE_SHIFT : + 1UL << (BITS_PER_LONG - PAGE_SHIFT); + + for_each_sg(sglist, s, nelems, i) { + int p, n; + + s->dma_address = npages << PAGE_SHIFT; + p = npages % boundary_size; + n = iommu_num_pages(sg_phys(s), s->length, PAGE_SIZE); + if (p + n > boundary_size) + npages += boundary_size - p; + npages += n; + } + + return npages; } /* @@ -2754,46 +2443,79 @@ static void unmap_page(struct device *dev, dma_addr_t dma_addr, size_t size, * lists). */ static int map_sg(struct device *dev, struct scatterlist *sglist, - int nelems, enum dma_data_direction dir, + int nelems, enum dma_data_direction direction, struct dma_attrs *attrs) { + int mapped_pages = 0, npages = 0, prot = 0, i; struct protection_domain *domain; - int i; + struct dma_ops_domain *dma_dom; struct scatterlist *s; - phys_addr_t paddr; - int mapped_elems = 0; + unsigned long address; u64 dma_mask; domain = get_domain(dev); if (IS_ERR(domain)) return 0; + dma_dom = to_dma_ops_domain(domain); dma_mask = *dev->dma_mask; + npages = sg_num_pages(dev, sglist, nelems); + + address = dma_ops_alloc_iova(dev, dma_dom, npages, dma_mask); + if (address == DMA_ERROR_CODE) + goto out_err; + + prot = dir2prot(direction); + + /* Map all sg entries */ for_each_sg(sglist, s, nelems, i) { - paddr = sg_phys(s); + int j, pages = iommu_num_pages(sg_phys(s), s->length, PAGE_SIZE); - s->dma_address = __map_single(dev, domain->priv, - paddr, s->length, dir, false, - dma_mask); + for (j = 0; j < pages; ++j) { + unsigned long bus_addr, phys_addr; + int ret; - if (s->dma_address) { - s->dma_length = s->length; - mapped_elems++; - } else - goto unmap; + bus_addr = address + s->dma_address + (j << PAGE_SHIFT); + phys_addr = (sg_phys(s) & PAGE_MASK) + (j << PAGE_SHIFT); + ret = iommu_map_page(domain, bus_addr, phys_addr, PAGE_SIZE, prot, GFP_ATOMIC); + if (ret) + goto out_unmap; + + mapped_pages += 1; + } } - return mapped_elems; + /* Everything is mapped - write the right values into s->dma_address */ + for_each_sg(sglist, s, nelems, i) { + s->dma_address += address + s->offset; + s->dma_length = s->length; + } + + return nelems; + +out_unmap: + pr_err("%s: IOMMU mapping error in map_sg (io-pages: %d)\n", + dev_name(dev), npages); -unmap: - for_each_sg(sglist, s, mapped_elems, i) { - if (s->dma_address) - __unmap_single(domain->priv, s->dma_address, - s->dma_length, dir); - s->dma_address = s->dma_length = 0; + for_each_sg(sglist, s, nelems, i) { + int j, pages = iommu_num_pages(sg_phys(s), s->length, PAGE_SIZE); + + for (j = 0; j < pages; ++j) { + unsigned long bus_addr; + + bus_addr = address + s->dma_address + (j << PAGE_SHIFT); + iommu_unmap_page(domain, bus_addr, PAGE_SIZE); + + if (--mapped_pages) + goto out_free_iova; + } } +out_free_iova: + free_iova_fast(&dma_dom->iovad, address, npages); + +out_err: return 0; } @@ -2806,18 +2528,19 @@ static void unmap_sg(struct device *dev, struct scatterlist *sglist, struct dma_attrs *attrs) { struct protection_domain *domain; - struct scatterlist *s; - int i; + struct dma_ops_domain *dma_dom; + unsigned long startaddr; + int npages = 2; domain = get_domain(dev); if (IS_ERR(domain)) return; - for_each_sg(sglist, s, nelems, i) { - __unmap_single(domain->priv, s->dma_address, - s->dma_length, dir); - s->dma_address = s->dma_length = 0; - } + startaddr = sg_dma_address(sglist) & PAGE_MASK; + dma_dom = to_dma_ops_domain(domain); + npages = sg_num_pages(dev, sglist, nelems); + + __unmap_single(dma_dom, startaddr, npages << PAGE_SHIFT, dir); } /* @@ -2829,6 +2552,7 @@ static void *alloc_coherent(struct device *dev, size_t size, { u64 dma_mask = dev->coherent_dma_mask; struct protection_domain *domain; + struct dma_ops_domain *dma_dom; struct page *page; domain = get_domain(dev); @@ -2839,6 +2563,7 @@ static void *alloc_coherent(struct device *dev, size_t size, } else if (IS_ERR(domain)) return NULL; + dma_dom = to_dma_ops_domain(domain); size = PAGE_ALIGN(size); dma_mask = dev->coherent_dma_mask; flag &= ~(__GFP_DMA | __GFP_HIGHMEM | __GFP_DMA32); @@ -2858,8 +2583,8 @@ static void *alloc_coherent(struct device *dev, size_t size, if (!dma_mask) dma_mask = *dev->dma_mask; - *dma_addr = __map_single(dev, domain->priv, page_to_phys(page), - size, DMA_BIDIRECTIONAL, true, dma_mask); + *dma_addr = __map_single(dev, dma_dom, page_to_phys(page), + size, DMA_BIDIRECTIONAL, dma_mask); if (*dma_addr == DMA_ERROR_CODE) goto out_free; @@ -2882,6 +2607,7 @@ static void free_coherent(struct device *dev, size_t size, struct dma_attrs *attrs) { struct protection_domain *domain; + struct dma_ops_domain *dma_dom; struct page *page; page = virt_to_page(virt_addr); @@ -2891,7 +2617,9 @@ static void free_coherent(struct device *dev, size_t size, if (IS_ERR(domain)) goto free_mem; - __unmap_single(domain->priv, dma_addr, size, DMA_BIDIRECTIONAL); + dma_dom = to_dma_ops_domain(domain); + + __unmap_single(dma_dom, dma_addr, size, DMA_BIDIRECTIONAL); free_mem: if (!dma_release_from_contiguous(dev, page, size >> PAGE_SHIFT)) @@ -2907,48 +2635,92 @@ static int amd_iommu_dma_supported(struct device *dev, u64 mask) return check_device(dev); } -static int set_dma_mask(struct device *dev, u64 mask) +static struct dma_map_ops amd_iommu_dma_ops = { + .alloc = alloc_coherent, + .free = free_coherent, + .map_page = map_page, + .unmap_page = unmap_page, + .map_sg = map_sg, + .unmap_sg = unmap_sg, + .dma_supported = amd_iommu_dma_supported, +}; + +static int init_reserved_iova_ranges(void) { - struct protection_domain *domain; - int max_apertures = 1; + struct pci_dev *pdev = NULL; + struct iova *val; - domain = get_domain(dev); - if (IS_ERR(domain)) - return PTR_ERR(domain); + init_iova_domain(&reserved_iova_ranges, PAGE_SIZE, + IOVA_START_PFN, DMA_32BIT_PFN); - if (mask == DMA_BIT_MASK(64)) - max_apertures = 8; - else if (mask > DMA_BIT_MASK(32)) - max_apertures = 4; + lockdep_set_class(&reserved_iova_ranges.iova_rbtree_lock, + &reserved_rbtree_key); + + /* MSI memory range */ + val = reserve_iova(&reserved_iova_ranges, + IOVA_PFN(MSI_RANGE_START), IOVA_PFN(MSI_RANGE_END)); + if (!val) { + pr_err("Reserving MSI range failed\n"); + return -ENOMEM; + } + + /* HT memory range */ + val = reserve_iova(&reserved_iova_ranges, + IOVA_PFN(HT_RANGE_START), IOVA_PFN(HT_RANGE_END)); + if (!val) { + pr_err("Reserving HT range failed\n"); + return -ENOMEM; + } /* - * To prevent lock contention it doesn't make sense to allocate more - * apertures than online cpus + * Memory used for PCI resources + * FIXME: Check whether we can reserve the PCI-hole completly */ - if (max_apertures > num_online_cpus()) - max_apertures = num_online_cpus(); + for_each_pci_dev(pdev) { + int i; + + for (i = 0; i < PCI_NUM_RESOURCES; ++i) { + struct resource *r = &pdev->resource[i]; + + if (!(r->flags & IORESOURCE_MEM)) + continue; - if (dma_ops_domain_alloc_apertures(domain->priv, max_apertures)) - dev_err(dev, "Can't allocate %d iommu apertures\n", - max_apertures); + val = reserve_iova(&reserved_iova_ranges, + IOVA_PFN(r->start), + IOVA_PFN(r->end)); + if (!val) { + pr_err("Reserve pci-resource range failed\n"); + return -ENOMEM; + } + } + } return 0; } -static struct dma_map_ops amd_iommu_dma_ops = { - .alloc = alloc_coherent, - .free = free_coherent, - .map_page = map_page, - .unmap_page = unmap_page, - .map_sg = map_sg, - .unmap_sg = unmap_sg, - .dma_supported = amd_iommu_dma_supported, - .set_dma_mask = set_dma_mask, -}; - int __init amd_iommu_init_api(void) { - int err = 0; + int ret, cpu, err = 0; + + ret = iova_cache_get(); + if (ret) + return ret; + + ret = init_reserved_iova_ranges(); + if (ret) + return ret; + + for_each_possible_cpu(cpu) { + struct flush_queue *queue = per_cpu_ptr(&flush_queue, cpu); + + queue->entries = kzalloc(FLUSH_QUEUE_SIZE * + sizeof(*queue->entries), + GFP_KERNEL); + if (!queue->entries) + goto out_put_iova; + + spin_lock_init(&queue->lock); + } err = bus_set_iommu(&pci_bus_type, &amd_iommu_ops); if (err) @@ -2958,11 +2730,26 @@ int __init amd_iommu_init_api(void) if (err) return err; #endif + err = bus_set_iommu(&platform_bus_type, &amd_iommu_ops); + if (err) + return err; return 0; + +out_put_iova: + for_each_possible_cpu(cpu) { + struct flush_queue *queue = per_cpu_ptr(&flush_queue, cpu); + + kfree(queue->entries); + } + + return -ENOMEM; } int __init amd_iommu_init_dma_ops(void) { + setup_timer(&queue_timer, queue_flush_timeout, 0); + atomic_set(&queue_timer_on, 0); + swiotlb = iommu_pass_through ? 1 : 0; iommu_detected = 1; @@ -2981,6 +2768,7 @@ int __init amd_iommu_init_dma_ops(void) pr_info("AMD-Vi: Lazy IO/TLB flushing enabled\n"); return 0; + } /***************************************************************************** @@ -3103,9 +2891,7 @@ static struct iommu_domain *amd_iommu_domain_alloc(unsigned type) static void amd_iommu_domain_free(struct iommu_domain *dom) { struct protection_domain *domain; - - if (!dom) - return; + struct dma_ops_domain *dma_dom; domain = to_pdomain(dom); @@ -3114,13 +2900,31 @@ static void amd_iommu_domain_free(struct iommu_domain *dom) BUG_ON(domain->dev_cnt != 0); - if (domain->mode != PAGE_MODE_NONE) - free_pagetable(domain); + if (!dom) + return; + + switch (dom->type) { + case IOMMU_DOMAIN_DMA: + /* + * First make sure the domain is no longer referenced from the + * flush queue + */ + queue_flush_all(); - if (domain->flags & PD_IOMMUV2_MASK) - free_gcr3_table(domain); + /* Now release the domain */ + dma_dom = to_dma_ops_domain(domain); + dma_ops_domain_free(dma_dom); + break; + default: + if (domain->mode != PAGE_MODE_NONE) + free_pagetable(domain); + + if (domain->flags & PD_IOMMUV2_MASK) + free_gcr3_table(domain); - protection_domain_free(domain); + protection_domain_free(domain); + break; + } } static void amd_iommu_detach_device(struct iommu_domain *dom, @@ -3190,7 +2994,7 @@ static int amd_iommu_map(struct iommu_domain *dom, unsigned long iova, prot |= IOMMU_PROT_IW; mutex_lock(&domain->api_lock); - ret = iommu_map_page(domain, iova, paddr, prot, page_size); + ret = iommu_map_page(domain, iova, paddr, page_size, prot, GFP_KERNEL); mutex_unlock(&domain->api_lock); return ret; @@ -3292,6 +3096,19 @@ static void amd_iommu_put_dm_regions(struct device *dev, kfree(entry); } +static void amd_iommu_apply_dm_region(struct device *dev, + struct iommu_domain *domain, + struct iommu_dm_region *region) +{ + struct dma_ops_domain *dma_dom = to_dma_ops_domain(to_pdomain(domain)); + unsigned long start, end; + + start = IOVA_PFN(region->start); + end = IOVA_PFN(region->start + region->length); + + WARN_ON_ONCE(reserve_iova(&dma_dom->iovad, start, end) == NULL); +} + static const struct iommu_ops amd_iommu_ops = { .capable = amd_iommu_capable, .domain_alloc = amd_iommu_domain_alloc, @@ -3307,6 +3124,7 @@ static const struct iommu_ops amd_iommu_ops = { .device_group = amd_iommu_device_group, .get_dm_regions = amd_iommu_get_dm_regions, .put_dm_regions = amd_iommu_put_dm_regions, + .apply_dm_region = amd_iommu_apply_dm_region, .pgsize_bitmap = AMD_IOMMU_PGSIZES, }; diff --git a/drivers/iommu/amd_iommu_init.c b/drivers/iommu/amd_iommu_init.c index 9e0034196e10..59741ead7e15 100644 --- a/drivers/iommu/amd_iommu_init.c +++ b/drivers/iommu/amd_iommu_init.c @@ -1107,13 +1107,13 @@ static int __init init_iommu_from_acpi(struct amd_iommu *iommu, break; } + devid = e->devid; DUMP_printk(" DEV_ACPI_HID(%s[%s])\t\tdevid: %02x:%02x.%x\n", hid, uid, PCI_BUS_NUM(devid), PCI_SLOT(devid), PCI_FUNC(devid)); - devid = e->devid; flags = e->flags; ret = add_acpi_hid_device(hid, uid, &devid, false); @@ -1568,13 +1568,23 @@ static int __init amd_iommu_init_pci(void) break; } + /* + * Order is important here to make sure any unity map requirements are + * fulfilled. The unity mappings are created and written to the device + * table during the amd_iommu_init_api() call. + * + * After that we call init_device_table_dma() to make sure any + * uninitialized DTE will block DMA, and in the end we flush the caches + * of all IOMMUs to make sure the changes to the device table are + * active. + */ + ret = amd_iommu_init_api(); + init_device_table_dma(); for_each_iommu(iommu) iommu_flush_all_caches(iommu); - ret = amd_iommu_init_api(); - if (!ret) print_iommu_info(); diff --git a/drivers/iommu/amd_iommu_types.h b/drivers/iommu/amd_iommu_types.h index 590956ac704e..caf5e3822715 100644 --- a/drivers/iommu/amd_iommu_types.h +++ b/drivers/iommu/amd_iommu_types.h @@ -421,7 +421,6 @@ struct protection_domain { bool updated; /* complete domain flush required */ unsigned dev_cnt; /* devices assigned to this domain */ unsigned dev_iommu[MAX_IOMMUS]; /* per-IOMMU reference count */ - void *priv; /* private data */ }; /* diff --git a/drivers/iommu/amd_iommu_v2.c b/drivers/iommu/amd_iommu_v2.c index 56999d2fac07..0d52ceb0846d 100644 --- a/drivers/iommu/amd_iommu_v2.c +++ b/drivers/iommu/amd_iommu_v2.c @@ -961,7 +961,7 @@ static int __init amd_iommu_v2_init(void) spin_lock_init(&state_lock); ret = -ENOMEM; - iommu_wq = create_workqueue("amd_iommu_v2"); + iommu_wq = alloc_workqueue("amd_iommu_v2", WQ_MEM_RECLAIM, 0); if (iommu_wq == NULL) goto out; diff --git a/drivers/iommu/arm-smmu-v3.c b/drivers/iommu/arm-smmu-v3.c index 5f6b3bcab078..ce801170d5f2 100644 --- a/drivers/iommu/arm-smmu-v3.c +++ b/drivers/iommu/arm-smmu-v3.c @@ -2687,6 +2687,8 @@ static int __init arm_smmu_init(void) if (ret) return ret; + pci_request_acs(); + return bus_set_iommu(&pci_bus_type, &arm_smmu_ops); } diff --git a/drivers/iommu/arm-smmu.c b/drivers/iommu/arm-smmu.c index 9345a3fcb706..4f49fe29f202 100644 --- a/drivers/iommu/arm-smmu.c +++ b/drivers/iommu/arm-smmu.c @@ -987,8 +987,8 @@ static int arm_smmu_init_domain_context(struct iommu_domain *domain, * handler seeing a half-initialised domain state. */ irq = smmu->irqs[smmu->num_global_irqs + cfg->irptndx]; - ret = request_irq(irq, arm_smmu_context_fault, IRQF_SHARED, - "arm-smmu-context-fault", domain); + ret = devm_request_irq(smmu->dev, irq, arm_smmu_context_fault, + IRQF_SHARED, "arm-smmu-context-fault", domain); if (ret < 0) { dev_err(smmu->dev, "failed to request context IRQ %d (%u)\n", cfg->irptndx, irq); @@ -1028,7 +1028,7 @@ static void arm_smmu_destroy_domain_context(struct iommu_domain *domain) if (cfg->irptndx != INVALID_IRPTNDX) { irq = smmu->irqs[smmu->num_global_irqs + cfg->irptndx]; - free_irq(irq, domain); + devm_free_irq(smmu->dev, irq, domain); } free_io_pgtable_ops(smmu_domain->pgtbl_ops); @@ -1986,15 +1986,15 @@ static int arm_smmu_device_dt_probe(struct platform_device *pdev) } for (i = 0; i < smmu->num_global_irqs; ++i) { - err = request_irq(smmu->irqs[i], - arm_smmu_global_fault, - IRQF_SHARED, - "arm-smmu global fault", - smmu); + err = devm_request_irq(smmu->dev, smmu->irqs[i], + arm_smmu_global_fault, + IRQF_SHARED, + "arm-smmu global fault", + smmu); if (err) { dev_err(dev, "failed to request global IRQ %d (%u)\n", i, smmu->irqs[i]); - goto out_free_irqs; + goto out_put_masters; } } @@ -2006,10 +2006,6 @@ static int arm_smmu_device_dt_probe(struct platform_device *pdev) arm_smmu_device_reset(smmu); return 0; -out_free_irqs: - while (i--) - free_irq(smmu->irqs[i], smmu); - out_put_masters: for (node = rb_first(&smmu->masters); node; node = rb_next(node)) { struct arm_smmu_master *master @@ -2050,7 +2046,7 @@ static int arm_smmu_device_remove(struct platform_device *pdev) dev_err(dev, "removing device with active domains!\n"); for (i = 0; i < smmu->num_global_irqs; ++i) - free_irq(smmu->irqs[i], smmu); + devm_free_irq(smmu->dev, smmu->irqs[i], smmu); /* Turn the thing off */ writel(sCR0_CLIENTPD, ARM_SMMU_GR0_NS(smmu) + ARM_SMMU_GR0_sCR0); @@ -2096,8 +2092,10 @@ static int __init arm_smmu_init(void) #endif #ifdef CONFIG_PCI - if (!iommu_present(&pci_bus_type)) + if (!iommu_present(&pci_bus_type)) { + pci_request_acs(); bus_set_iommu(&pci_bus_type, &arm_smmu_ops); + } #endif return 0; diff --git a/drivers/iommu/dmar.c b/drivers/iommu/dmar.c index 6a86b5d1defa..1344e29acd86 100644 --- a/drivers/iommu/dmar.c +++ b/drivers/iommu/dmar.c @@ -241,8 +241,20 @@ int dmar_insert_dev_scope(struct dmar_pci_notify_info *info, if (!dmar_match_pci_path(info, scope->bus, path, level)) continue; - if ((scope->entry_type == ACPI_DMAR_SCOPE_TYPE_ENDPOINT) ^ - (info->dev->hdr_type == PCI_HEADER_TYPE_NORMAL)) { + /* + * We expect devices with endpoint scope to have normal PCI + * headers, and devices with bridge scope to have bridge PCI + * headers. However PCI NTB devices may be listed in the + * DMAR table with bridge scope, even though they have a + * normal PCI header. NTB devices are identified by class + * "BRIDGE_OTHER" (0680h) - we don't declare a socpe mismatch + * for this special case. + */ + if ((scope->entry_type == ACPI_DMAR_SCOPE_TYPE_ENDPOINT && + info->dev->hdr_type != PCI_HEADER_TYPE_NORMAL) || + (scope->entry_type == ACPI_DMAR_SCOPE_TYPE_BRIDGE && + (info->dev->hdr_type == PCI_HEADER_TYPE_NORMAL && + info->dev->class >> 8 != PCI_CLASS_BRIDGE_OTHER))) { pr_warn("Device scope type does not match for %s\n", pci_name(info->dev)); return -EINVAL; @@ -1155,8 +1167,6 @@ static int qi_check_fault(struct intel_iommu *iommu, int index) (unsigned long long)qi->desc[index].high); memcpy(&qi->desc[index], &qi->desc[wait_index], sizeof(struct qi_desc)); - __iommu_flush_cache(iommu, &qi->desc[index], - sizeof(struct qi_desc)); writel(DMA_FSTS_IQE, iommu->reg + DMAR_FSTS_REG); return -EINVAL; } @@ -1231,9 +1241,6 @@ restart: hw[wait_index] = wait_desc; - __iommu_flush_cache(iommu, &hw[index], sizeof(struct qi_desc)); - __iommu_flush_cache(iommu, &hw[wait_index], sizeof(struct qi_desc)); - qi->free_head = (qi->free_head + 2) % QI_LENGTH; qi->free_cnt -= 2; diff --git a/drivers/iommu/exynos-iommu.c b/drivers/iommu/exynos-iommu.c index 5ecc86cb74c8..33dcc29ec200 100644 --- a/drivers/iommu/exynos-iommu.c +++ b/drivers/iommu/exynos-iommu.c @@ -54,6 +54,10 @@ typedef u32 sysmmu_pte_t; #define lv2ent_small(pent) ((*(pent) & 2) == 2) #define lv2ent_large(pent) ((*(pent) & 3) == 1) +#ifdef CONFIG_BIG_ENDIAN +#warning "revisit driver if we can enable big-endian ptes" +#endif + /* * v1.x - v3.x SYSMMU supports 32bit physical and 32bit virtual address spaces * v5.0 introduced support for 36bit physical address space by shifting @@ -322,14 +326,27 @@ static void __sysmmu_set_ptbase(struct sysmmu_drvdata *data, phys_addr_t pgd) __sysmmu_tlb_invalidate(data); } +static void __sysmmu_enable_clocks(struct sysmmu_drvdata *data) +{ + BUG_ON(clk_prepare_enable(data->clk_master)); + BUG_ON(clk_prepare_enable(data->clk)); + BUG_ON(clk_prepare_enable(data->pclk)); + BUG_ON(clk_prepare_enable(data->aclk)); +} + +static void __sysmmu_disable_clocks(struct sysmmu_drvdata *data) +{ + clk_disable_unprepare(data->aclk); + clk_disable_unprepare(data->pclk); + clk_disable_unprepare(data->clk); + clk_disable_unprepare(data->clk_master); +} + static void __sysmmu_get_version(struct sysmmu_drvdata *data) { u32 ver; - clk_enable(data->clk_master); - clk_enable(data->clk); - clk_enable(data->pclk); - clk_enable(data->aclk); + __sysmmu_enable_clocks(data); ver = readl(data->sfrbase + REG_MMU_VERSION); @@ -342,10 +359,7 @@ static void __sysmmu_get_version(struct sysmmu_drvdata *data) dev_dbg(data->sysmmu, "hardware version: %d.%d\n", MMU_MAJ_VER(data->version), MMU_MIN_VER(data->version)); - clk_disable(data->aclk); - clk_disable(data->pclk); - clk_disable(data->clk); - clk_disable(data->clk_master); + __sysmmu_disable_clocks(data); } static void show_fault_information(struct sysmmu_drvdata *data, @@ -427,10 +441,7 @@ static void __sysmmu_disable_nocount(struct sysmmu_drvdata *data) writel(CTRL_DISABLE, data->sfrbase + REG_MMU_CTRL); writel(0, data->sfrbase + REG_MMU_CFG); - clk_disable(data->aclk); - clk_disable(data->pclk); - clk_disable(data->clk); - clk_disable(data->clk_master); + __sysmmu_disable_clocks(data); } static bool __sysmmu_disable(struct sysmmu_drvdata *data) @@ -475,10 +486,7 @@ static void __sysmmu_init_config(struct sysmmu_drvdata *data) static void __sysmmu_enable_nocount(struct sysmmu_drvdata *data) { - clk_enable(data->clk_master); - clk_enable(data->clk); - clk_enable(data->pclk); - clk_enable(data->aclk); + __sysmmu_enable_clocks(data); writel(CTRL_BLOCK, data->sfrbase + REG_MMU_CTRL); @@ -488,6 +496,12 @@ static void __sysmmu_enable_nocount(struct sysmmu_drvdata *data) writel(CTRL_ENABLE, data->sfrbase + REG_MMU_CTRL); + /* + * SYSMMU driver keeps master's clock enabled only for the short + * time, while accessing the registers. For performing address + * translation during DMA transaction it relies on the client + * driver to enable it. + */ clk_disable(data->clk_master); } @@ -524,16 +538,15 @@ static void sysmmu_tlb_invalidate_flpdcache(struct sysmmu_drvdata *data, { unsigned long flags; - clk_enable(data->clk_master); spin_lock_irqsave(&data->lock, flags); - if (is_sysmmu_active(data)) { - if (data->version >= MAKE_MMU_VER(3, 3)) - __sysmmu_tlb_invalidate_entry(data, iova, 1); + if (is_sysmmu_active(data) && data->version >= MAKE_MMU_VER(3, 3)) { + clk_enable(data->clk_master); + __sysmmu_tlb_invalidate_entry(data, iova, 1); + clk_disable(data->clk_master); } spin_unlock_irqrestore(&data->lock, flags); - clk_disable(data->clk_master); } static void sysmmu_tlb_invalidate_entry(struct sysmmu_drvdata *data, @@ -572,6 +585,8 @@ static void sysmmu_tlb_invalidate_entry(struct sysmmu_drvdata *data, spin_unlock_irqrestore(&data->lock, flags); } +static struct iommu_ops exynos_iommu_ops; + static int __init exynos_sysmmu_probe(struct platform_device *pdev) { int irq, ret; @@ -602,37 +617,22 @@ static int __init exynos_sysmmu_probe(struct platform_device *pdev) } data->clk = devm_clk_get(dev, "sysmmu"); - if (!IS_ERR(data->clk)) { - ret = clk_prepare(data->clk); - if (ret) { - dev_err(dev, "Failed to prepare clk\n"); - return ret; - } - } else { + if (PTR_ERR(data->clk) == -ENOENT) data->clk = NULL; - } + else if (IS_ERR(data->clk)) + return PTR_ERR(data->clk); data->aclk = devm_clk_get(dev, "aclk"); - if (!IS_ERR(data->aclk)) { - ret = clk_prepare(data->aclk); - if (ret) { - dev_err(dev, "Failed to prepare aclk\n"); - return ret; - } - } else { + if (PTR_ERR(data->aclk) == -ENOENT) data->aclk = NULL; - } + else if (IS_ERR(data->aclk)) + return PTR_ERR(data->aclk); data->pclk = devm_clk_get(dev, "pclk"); - if (!IS_ERR(data->pclk)) { - ret = clk_prepare(data->pclk); - if (ret) { - dev_err(dev, "Failed to prepare pclk\n"); - return ret; - } - } else { + if (PTR_ERR(data->pclk) == -ENOENT) data->pclk = NULL; - } + else if (IS_ERR(data->pclk)) + return PTR_ERR(data->pclk); if (!data->clk && (!data->aclk || !data->pclk)) { dev_err(dev, "Failed to get device clock(s)!\n"); @@ -640,15 +640,10 @@ static int __init exynos_sysmmu_probe(struct platform_device *pdev) } data->clk_master = devm_clk_get(dev, "master"); - if (!IS_ERR(data->clk_master)) { - ret = clk_prepare(data->clk_master); - if (ret) { - dev_err(dev, "Failed to prepare master's clk\n"); - return ret; - } - } else { + if (PTR_ERR(data->clk_master) == -ENOENT) data->clk_master = NULL; - } + else if (IS_ERR(data->clk_master)) + return PTR_ERR(data->clk_master); data->sysmmu = dev; spin_lock_init(&data->lock); @@ -665,6 +660,8 @@ static int __init exynos_sysmmu_probe(struct platform_device *pdev) pm_runtime_enable(dev); + of_iommu_set_ops(dev->of_node, &exynos_iommu_ops); + return 0; } @@ -709,6 +706,7 @@ static struct platform_driver exynos_sysmmu_driver __refdata = { .name = "exynos-sysmmu", .of_match_table = sysmmu_of_match, .pm = &sysmmu_pm_ops, + .suppress_bind_attrs = true, } }; @@ -716,7 +714,7 @@ static inline void update_pte(sysmmu_pte_t *ent, sysmmu_pte_t val) { dma_sync_single_for_cpu(dma_dev, virt_to_phys(ent), sizeof(*ent), DMA_TO_DEVICE); - *ent = val; + *ent = cpu_to_le32(val); dma_sync_single_for_device(dma_dev, virt_to_phys(ent), sizeof(*ent), DMA_TO_DEVICE); } @@ -1357,7 +1355,6 @@ static int __init exynos_iommu_of_setup(struct device_node *np) if (!dma_dev) dma_dev = &pdev->dev; - of_iommu_set_ops(np, &exynos_iommu_ops); return 0; } diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c index 10700945994e..9d2938e12aa3 100644 --- a/drivers/iommu/intel-iommu.c +++ b/drivers/iommu/intel-iommu.c @@ -1672,7 +1672,7 @@ static int iommu_init_domains(struct intel_iommu *iommu) return -ENOMEM; } - size = ((ndomains >> 8) + 1) * sizeof(struct dmar_domain **); + size = (ALIGN(ndomains, 256) >> 8) * sizeof(struct dmar_domain **); iommu->domains = kzalloc(size, GFP_KERNEL); if (iommu->domains) { @@ -1737,7 +1737,7 @@ static void disable_dmar_iommu(struct intel_iommu *iommu) static void free_dmar_iommu(struct intel_iommu *iommu) { if ((iommu->domains) && (iommu->domain_ids)) { - int elems = (cap_ndoms(iommu->cap) >> 8) + 1; + int elems = ALIGN(cap_ndoms(iommu->cap), 256) >> 8; int i; for (i = 0; i < elems; i++) @@ -2076,7 +2076,7 @@ out_unlock: spin_unlock(&iommu->lock); spin_unlock_irqrestore(&device_domain_lock, flags); - return 0; + return ret; } struct domain_context_mapping_data { @@ -4602,13 +4602,13 @@ static void free_all_cpu_cached_iovas(unsigned int cpu) for (i = 0; i < g_num_of_iommus; i++) { struct intel_iommu *iommu = g_iommus[i]; struct dmar_domain *domain; - u16 did; + int did; if (!iommu) continue; - for (did = 0; did < 0xffff; did++) { - domain = get_iommu_domain(iommu, did); + for (did = 0; did < cap_ndoms(iommu->cap); did++) { + domain = get_iommu_domain(iommu, (u16)did); if (!domain) continue; diff --git a/drivers/iommu/io-pgtable-arm.c b/drivers/iommu/io-pgtable-arm.c index a1ed1b73fed4..f5c90e1366ce 100644 --- a/drivers/iommu/io-pgtable-arm.c +++ b/drivers/iommu/io-pgtable-arm.c @@ -576,7 +576,7 @@ static phys_addr_t arm_lpae_iova_to_phys(struct io_pgtable_ops *ops, return 0; found_translation: - iova &= (ARM_LPAE_GRANULE(data) - 1); + iova &= (ARM_LPAE_BLOCK_SIZE(lvl, data) - 1); return ((phys_addr_t)iopte_to_pfn(pte,data) << data->pg_shift) | iova; } diff --git a/drivers/iommu/iommu.c b/drivers/iommu/iommu.c index 3000051f48b4..b06d93594436 100644 --- a/drivers/iommu/iommu.c +++ b/drivers/iommu/iommu.c @@ -34,8 +34,7 @@ #include <trace/events/iommu.h> static struct kset *iommu_group_kset; -static struct ida iommu_group_ida; -static struct mutex iommu_group_mutex; +static DEFINE_IDA(iommu_group_ida); struct iommu_callback_data { const struct iommu_ops *ops; @@ -144,9 +143,7 @@ static void iommu_group_release(struct kobject *kobj) if (group->iommu_data_release) group->iommu_data_release(group->iommu_data); - mutex_lock(&iommu_group_mutex); - ida_remove(&iommu_group_ida, group->id); - mutex_unlock(&iommu_group_mutex); + ida_simple_remove(&iommu_group_ida, group->id); if (group->default_domain) iommu_domain_free(group->default_domain); @@ -186,26 +183,17 @@ struct iommu_group *iommu_group_alloc(void) INIT_LIST_HEAD(&group->devices); BLOCKING_INIT_NOTIFIER_HEAD(&group->notifier); - mutex_lock(&iommu_group_mutex); - -again: - if (unlikely(0 == ida_pre_get(&iommu_group_ida, GFP_KERNEL))) { + ret = ida_simple_get(&iommu_group_ida, 0, 0, GFP_KERNEL); + if (ret < 0) { kfree(group); - mutex_unlock(&iommu_group_mutex); - return ERR_PTR(-ENOMEM); + return ERR_PTR(ret); } - - if (-EAGAIN == ida_get_new(&iommu_group_ida, &group->id)) - goto again; - - mutex_unlock(&iommu_group_mutex); + group->id = ret; ret = kobject_init_and_add(&group->kobj, &iommu_group_ktype, NULL, "%d", group->id); if (ret) { - mutex_lock(&iommu_group_mutex); - ida_remove(&iommu_group_ida, group->id); - mutex_unlock(&iommu_group_mutex); + ida_simple_remove(&iommu_group_ida, group->id); kfree(group); return ERR_PTR(ret); } @@ -348,6 +336,9 @@ static int iommu_group_create_direct_mappings(struct iommu_group *group, list_for_each_entry(entry, &mappings, list) { dma_addr_t start, end, addr; + if (domain->ops->apply_dm_region) + domain->ops->apply_dm_region(dev, domain, entry); + start = ALIGN(entry->start, pg_size); end = ALIGN(entry->start + entry->length, pg_size); @@ -1483,9 +1474,6 @@ static int __init iommu_init(void) { iommu_group_kset = kset_create_and_add("iommu_groups", NULL, kernel_kobj); - ida_init(&iommu_group_ida); - mutex_init(&iommu_group_mutex); - BUG_ON(!iommu_group_kset); return 0; diff --git a/drivers/iommu/iova.c b/drivers/iommu/iova.c index ba764a0835d3..e23001bfcfee 100644 --- a/drivers/iommu/iova.c +++ b/drivers/iommu/iova.c @@ -420,8 +420,10 @@ retry: /* Try replenishing IOVAs by flushing rcache. */ flushed_rcache = true; + preempt_disable(); for_each_online_cpu(cpu) free_cpu_cached_iovas(cpu, iovad); + preempt_enable(); goto retry; } @@ -749,7 +751,7 @@ static bool __iova_rcache_insert(struct iova_domain *iovad, bool can_insert = false; unsigned long flags; - cpu_rcache = this_cpu_ptr(rcache->cpu_rcaches); + cpu_rcache = get_cpu_ptr(rcache->cpu_rcaches); spin_lock_irqsave(&cpu_rcache->lock, flags); if (!iova_magazine_full(cpu_rcache->loaded)) { @@ -779,6 +781,7 @@ static bool __iova_rcache_insert(struct iova_domain *iovad, iova_magazine_push(cpu_rcache->loaded, iova_pfn); spin_unlock_irqrestore(&cpu_rcache->lock, flags); + put_cpu_ptr(rcache->cpu_rcaches); if (mag_to_free) { iova_magazine_free_pfns(mag_to_free, iovad); @@ -812,7 +815,7 @@ static unsigned long __iova_rcache_get(struct iova_rcache *rcache, bool has_pfn = false; unsigned long flags; - cpu_rcache = this_cpu_ptr(rcache->cpu_rcaches); + cpu_rcache = get_cpu_ptr(rcache->cpu_rcaches); spin_lock_irqsave(&cpu_rcache->lock, flags); if (!iova_magazine_empty(cpu_rcache->loaded)) { @@ -834,6 +837,7 @@ static unsigned long __iova_rcache_get(struct iova_rcache *rcache, iova_pfn = iova_magazine_pop(cpu_rcache->loaded, limit_pfn); spin_unlock_irqrestore(&cpu_rcache->lock, flags); + put_cpu_ptr(rcache->cpu_rcaches); return iova_pfn; } diff --git a/drivers/iommu/mtk_iommu.c b/drivers/iommu/mtk_iommu.c index c3043d8754e3..b12c12d74c33 100644 --- a/drivers/iommu/mtk_iommu.c +++ b/drivers/iommu/mtk_iommu.c @@ -34,7 +34,7 @@ #include <dt-bindings/memory/mt8173-larb-port.h> #include <soc/mediatek/smi.h> -#include "io-pgtable.h" +#include "mtk_iommu.h" #define REG_MMU_PT_BASE_ADDR 0x000 @@ -93,20 +93,6 @@ #define MTK_PROTECT_PA_ALIGN 128 -struct mtk_iommu_suspend_reg { - u32 standard_axi_mode; - u32 dcm_dis; - u32 ctrl_reg; - u32 int_control0; - u32 int_main_control; -}; - -struct mtk_iommu_client_priv { - struct list_head client; - unsigned int mtk_m4u_id; - struct device *m4udev; -}; - struct mtk_iommu_domain { spinlock_t pgtlock; /* lock for page table */ @@ -116,19 +102,6 @@ struct mtk_iommu_domain { struct iommu_domain domain; }; -struct mtk_iommu_data { - void __iomem *base; - int irq; - struct device *dev; - struct clk *bclk; - phys_addr_t protect_base; /* protect memory base */ - struct mtk_iommu_suspend_reg reg; - struct mtk_iommu_domain *m4u_dom; - struct iommu_group *m4u_group; - struct mtk_smi_iommu smi_imu; /* SMI larb iommu info */ - bool enable_4GB; -}; - static struct iommu_ops mtk_iommu_ops; static struct mtk_iommu_domain *to_mtk_domain(struct iommu_domain *dom) @@ -455,7 +428,6 @@ static int mtk_iommu_of_xlate(struct device *dev, struct of_phandle_args *args) if (!dev->archdata.iommu) { /* Get the m4u device */ m4updev = of_find_device_by_node(args->np); - of_node_put(args->np); if (WARN_ON(!m4updev)) return -EINVAL; @@ -552,25 +524,6 @@ static int mtk_iommu_hw_init(const struct mtk_iommu_data *data) return 0; } -static int compare_of(struct device *dev, void *data) -{ - return dev->of_node == data; -} - -static int mtk_iommu_bind(struct device *dev) -{ - struct mtk_iommu_data *data = dev_get_drvdata(dev); - - return component_bind_all(dev, &data->smi_imu); -} - -static void mtk_iommu_unbind(struct device *dev) -{ - struct mtk_iommu_data *data = dev_get_drvdata(dev); - - component_unbind_all(dev, &data->smi_imu); -} - static const struct component_master_ops mtk_iommu_com_ops = { .bind = mtk_iommu_bind, .unbind = mtk_iommu_unbind, diff --git a/drivers/iommu/mtk_iommu.h b/drivers/iommu/mtk_iommu.h new file mode 100644 index 000000000000..9ed0a8462ccf --- /dev/null +++ b/drivers/iommu/mtk_iommu.h @@ -0,0 +1,77 @@ +/* + * Copyright (c) 2015-2016 MediaTek Inc. + * Author: Honghui Zhang <honghui.zhang@mediatek.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#ifndef _MTK_IOMMU_H_ +#define _MTK_IOMMU_H_ + +#include <linux/clk.h> +#include <linux/component.h> +#include <linux/device.h> +#include <linux/io.h> +#include <linux/iommu.h> +#include <linux/list.h> +#include <linux/spinlock.h> +#include <soc/mediatek/smi.h> + +#include "io-pgtable.h" + +struct mtk_iommu_suspend_reg { + u32 standard_axi_mode; + u32 dcm_dis; + u32 ctrl_reg; + u32 int_control0; + u32 int_main_control; +}; + +struct mtk_iommu_client_priv { + struct list_head client; + unsigned int mtk_m4u_id; + struct device *m4udev; +}; + +struct mtk_iommu_domain; + +struct mtk_iommu_data { + void __iomem *base; + int irq; + struct device *dev; + struct clk *bclk; + phys_addr_t protect_base; /* protect memory base */ + struct mtk_iommu_suspend_reg reg; + struct mtk_iommu_domain *m4u_dom; + struct iommu_group *m4u_group; + struct mtk_smi_iommu smi_imu; /* SMI larb iommu info */ + bool enable_4GB; +}; + +static int compare_of(struct device *dev, void *data) +{ + return dev->of_node == data; +} + +static int mtk_iommu_bind(struct device *dev) +{ + struct mtk_iommu_data *data = dev_get_drvdata(dev); + + return component_bind_all(dev, &data->smi_imu); +} + +static void mtk_iommu_unbind(struct device *dev) +{ + struct mtk_iommu_data *data = dev_get_drvdata(dev); + + component_unbind_all(dev, &data->smi_imu); +} + +#endif diff --git a/drivers/iommu/mtk_iommu_v1.c b/drivers/iommu/mtk_iommu_v1.c new file mode 100644 index 000000000000..b8aeb0768483 --- /dev/null +++ b/drivers/iommu/mtk_iommu_v1.c @@ -0,0 +1,727 @@ +/* + * Copyright (c) 2015-2016 MediaTek Inc. + * Author: Honghui Zhang <honghui.zhang@mediatek.com> + * + * Based on driver/iommu/mtk_iommu.c + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ +#include <linux/bootmem.h> +#include <linux/bug.h> +#include <linux/clk.h> +#include <linux/component.h> +#include <linux/device.h> +#include <linux/dma-iommu.h> +#include <linux/err.h> +#include <linux/interrupt.h> +#include <linux/io.h> +#include <linux/iommu.h> +#include <linux/iopoll.h> +#include <linux/kmemleak.h> +#include <linux/list.h> +#include <linux/of_address.h> +#include <linux/of_iommu.h> +#include <linux/of_irq.h> +#include <linux/of_platform.h> +#include <linux/platform_device.h> +#include <linux/slab.h> +#include <linux/spinlock.h> +#include <asm/barrier.h> +#include <asm/dma-iommu.h> +#include <linux/module.h> +#include <dt-bindings/memory/mt2701-larb-port.h> +#include <soc/mediatek/smi.h> +#include "mtk_iommu.h" + +#define REG_MMU_PT_BASE_ADDR 0x000 + +#define F_ALL_INVLD 0x2 +#define F_MMU_INV_RANGE 0x1 +#define F_INVLD_EN0 BIT(0) +#define F_INVLD_EN1 BIT(1) + +#define F_MMU_FAULT_VA_MSK 0xfffff000 +#define MTK_PROTECT_PA_ALIGN 128 + +#define REG_MMU_CTRL_REG 0x210 +#define F_MMU_CTRL_COHERENT_EN BIT(8) +#define REG_MMU_IVRP_PADDR 0x214 +#define REG_MMU_INT_CONTROL 0x220 +#define F_INT_TRANSLATION_FAULT BIT(0) +#define F_INT_MAIN_MULTI_HIT_FAULT BIT(1) +#define F_INT_INVALID_PA_FAULT BIT(2) +#define F_INT_ENTRY_REPLACEMENT_FAULT BIT(3) +#define F_INT_TABLE_WALK_FAULT BIT(4) +#define F_INT_TLB_MISS_FAULT BIT(5) +#define F_INT_PFH_DMA_FIFO_OVERFLOW BIT(6) +#define F_INT_MISS_DMA_FIFO_OVERFLOW BIT(7) + +#define F_MMU_TF_PROTECT_SEL(prot) (((prot) & 0x3) << 5) +#define F_INT_CLR_BIT BIT(12) + +#define REG_MMU_FAULT_ST 0x224 +#define REG_MMU_FAULT_VA 0x228 +#define REG_MMU_INVLD_PA 0x22C +#define REG_MMU_INT_ID 0x388 +#define REG_MMU_INVALIDATE 0x5c0 +#define REG_MMU_INVLD_START_A 0x5c4 +#define REG_MMU_INVLD_END_A 0x5c8 + +#define REG_MMU_INV_SEL 0x5d8 +#define REG_MMU_STANDARD_AXI_MODE 0x5e8 + +#define REG_MMU_DCM 0x5f0 +#define F_MMU_DCM_ON BIT(1) +#define REG_MMU_CPE_DONE 0x60c +#define F_DESC_VALID 0x2 +#define F_DESC_NONSEC BIT(3) +#define MT2701_M4U_TF_LARB(TF) (6 - (((TF) >> 13) & 0x7)) +#define MT2701_M4U_TF_PORT(TF) (((TF) >> 8) & 0xF) +/* MTK generation one iommu HW only support 4K size mapping */ +#define MT2701_IOMMU_PAGE_SHIFT 12 +#define MT2701_IOMMU_PAGE_SIZE (1UL << MT2701_IOMMU_PAGE_SHIFT) + +/* + * MTK m4u support 4GB iova address space, and only support 4K page + * mapping. So the pagetable size should be exactly as 4M. + */ +#define M2701_IOMMU_PGT_SIZE SZ_4M + +struct mtk_iommu_domain { + spinlock_t pgtlock; /* lock for page table */ + struct iommu_domain domain; + u32 *pgt_va; + dma_addr_t pgt_pa; + struct mtk_iommu_data *data; +}; + +static struct mtk_iommu_domain *to_mtk_domain(struct iommu_domain *dom) +{ + return container_of(dom, struct mtk_iommu_domain, domain); +} + +static const int mt2701_m4u_in_larb[] = { + LARB0_PORT_OFFSET, LARB1_PORT_OFFSET, + LARB2_PORT_OFFSET, LARB3_PORT_OFFSET +}; + +static inline int mt2701_m4u_to_larb(int id) +{ + int i; + + for (i = ARRAY_SIZE(mt2701_m4u_in_larb) - 1; i >= 0; i--) + if ((id) >= mt2701_m4u_in_larb[i]) + return i; + + return 0; +} + +static inline int mt2701_m4u_to_port(int id) +{ + int larb = mt2701_m4u_to_larb(id); + + return id - mt2701_m4u_in_larb[larb]; +} + +static void mtk_iommu_tlb_flush_all(struct mtk_iommu_data *data) +{ + writel_relaxed(F_INVLD_EN1 | F_INVLD_EN0, + data->base + REG_MMU_INV_SEL); + writel_relaxed(F_ALL_INVLD, data->base + REG_MMU_INVALIDATE); + wmb(); /* Make sure the tlb flush all done */ +} + +static void mtk_iommu_tlb_flush_range(struct mtk_iommu_data *data, + unsigned long iova, size_t size) +{ + int ret; + u32 tmp; + + writel_relaxed(F_INVLD_EN1 | F_INVLD_EN0, + data->base + REG_MMU_INV_SEL); + writel_relaxed(iova & F_MMU_FAULT_VA_MSK, + data->base + REG_MMU_INVLD_START_A); + writel_relaxed((iova + size - 1) & F_MMU_FAULT_VA_MSK, + data->base + REG_MMU_INVLD_END_A); + writel_relaxed(F_MMU_INV_RANGE, data->base + REG_MMU_INVALIDATE); + + ret = readl_poll_timeout_atomic(data->base + REG_MMU_CPE_DONE, + tmp, tmp != 0, 10, 100000); + if (ret) { + dev_warn(data->dev, + "Partial TLB flush timed out, falling back to full flush\n"); + mtk_iommu_tlb_flush_all(data); + } + /* Clear the CPE status */ + writel_relaxed(0, data->base + REG_MMU_CPE_DONE); +} + +static irqreturn_t mtk_iommu_isr(int irq, void *dev_id) +{ + struct mtk_iommu_data *data = dev_id; + struct mtk_iommu_domain *dom = data->m4u_dom; + u32 int_state, regval, fault_iova, fault_pa; + unsigned int fault_larb, fault_port; + + /* Read error information from registers */ + int_state = readl_relaxed(data->base + REG_MMU_FAULT_ST); + fault_iova = readl_relaxed(data->base + REG_MMU_FAULT_VA); + + fault_iova &= F_MMU_FAULT_VA_MSK; + fault_pa = readl_relaxed(data->base + REG_MMU_INVLD_PA); + regval = readl_relaxed(data->base + REG_MMU_INT_ID); + fault_larb = MT2701_M4U_TF_LARB(regval); + fault_port = MT2701_M4U_TF_PORT(regval); + + /* + * MTK v1 iommu HW could not determine whether the fault is read or + * write fault, report as read fault. + */ + if (report_iommu_fault(&dom->domain, data->dev, fault_iova, + IOMMU_FAULT_READ)) + dev_err_ratelimited(data->dev, + "fault type=0x%x iova=0x%x pa=0x%x larb=%d port=%d\n", + int_state, fault_iova, fault_pa, + fault_larb, fault_port); + + /* Interrupt clear */ + regval = readl_relaxed(data->base + REG_MMU_INT_CONTROL); + regval |= F_INT_CLR_BIT; + writel_relaxed(regval, data->base + REG_MMU_INT_CONTROL); + + mtk_iommu_tlb_flush_all(data); + + return IRQ_HANDLED; +} + +static void mtk_iommu_config(struct mtk_iommu_data *data, + struct device *dev, bool enable) +{ + struct mtk_iommu_client_priv *head, *cur, *next; + struct mtk_smi_larb_iommu *larb_mmu; + unsigned int larbid, portid; + + head = dev->archdata.iommu; + list_for_each_entry_safe(cur, next, &head->client, client) { + larbid = mt2701_m4u_to_larb(cur->mtk_m4u_id); + portid = mt2701_m4u_to_port(cur->mtk_m4u_id); + larb_mmu = &data->smi_imu.larb_imu[larbid]; + + dev_dbg(dev, "%s iommu port: %d\n", + enable ? "enable" : "disable", portid); + + if (enable) + larb_mmu->mmu |= MTK_SMI_MMU_EN(portid); + else + larb_mmu->mmu &= ~MTK_SMI_MMU_EN(portid); + } +} + +static int mtk_iommu_domain_finalise(struct mtk_iommu_data *data) +{ + struct mtk_iommu_domain *dom = data->m4u_dom; + + spin_lock_init(&dom->pgtlock); + + dom->pgt_va = dma_zalloc_coherent(data->dev, + M2701_IOMMU_PGT_SIZE, + &dom->pgt_pa, GFP_KERNEL); + if (!dom->pgt_va) + return -ENOMEM; + + writel(dom->pgt_pa, data->base + REG_MMU_PT_BASE_ADDR); + + dom->data = data; + + return 0; +} + +static struct iommu_domain *mtk_iommu_domain_alloc(unsigned type) +{ + struct mtk_iommu_domain *dom; + + if (type != IOMMU_DOMAIN_UNMANAGED) + return NULL; + + dom = kzalloc(sizeof(*dom), GFP_KERNEL); + if (!dom) + return NULL; + + return &dom->domain; +} + +static void mtk_iommu_domain_free(struct iommu_domain *domain) +{ + struct mtk_iommu_domain *dom = to_mtk_domain(domain); + struct mtk_iommu_data *data = dom->data; + + dma_free_coherent(data->dev, M2701_IOMMU_PGT_SIZE, + dom->pgt_va, dom->pgt_pa); + kfree(to_mtk_domain(domain)); +} + +static int mtk_iommu_attach_device(struct iommu_domain *domain, + struct device *dev) +{ + struct mtk_iommu_domain *dom = to_mtk_domain(domain); + struct mtk_iommu_client_priv *priv = dev->archdata.iommu; + struct mtk_iommu_data *data; + int ret; + + if (!priv) + return -ENODEV; + + data = dev_get_drvdata(priv->m4udev); + if (!data->m4u_dom) { + data->m4u_dom = dom; + ret = mtk_iommu_domain_finalise(data); + if (ret) { + data->m4u_dom = NULL; + return ret; + } + } + + mtk_iommu_config(data, dev, true); + return 0; +} + +static void mtk_iommu_detach_device(struct iommu_domain *domain, + struct device *dev) +{ + struct mtk_iommu_client_priv *priv = dev->archdata.iommu; + struct mtk_iommu_data *data; + + if (!priv) + return; + + data = dev_get_drvdata(priv->m4udev); + mtk_iommu_config(data, dev, false); +} + +static int mtk_iommu_map(struct iommu_domain *domain, unsigned long iova, + phys_addr_t paddr, size_t size, int prot) +{ + struct mtk_iommu_domain *dom = to_mtk_domain(domain); + unsigned int page_num = size >> MT2701_IOMMU_PAGE_SHIFT; + unsigned long flags; + unsigned int i; + u32 *pgt_base_iova = dom->pgt_va + (iova >> MT2701_IOMMU_PAGE_SHIFT); + u32 pabase = (u32)paddr; + int map_size = 0; + + spin_lock_irqsave(&dom->pgtlock, flags); + for (i = 0; i < page_num; i++) { + if (pgt_base_iova[i]) { + memset(pgt_base_iova, 0, i * sizeof(u32)); + break; + } + pgt_base_iova[i] = pabase | F_DESC_VALID | F_DESC_NONSEC; + pabase += MT2701_IOMMU_PAGE_SIZE; + map_size += MT2701_IOMMU_PAGE_SIZE; + } + + spin_unlock_irqrestore(&dom->pgtlock, flags); + + mtk_iommu_tlb_flush_range(dom->data, iova, size); + + return map_size == size ? 0 : -EEXIST; +} + +static size_t mtk_iommu_unmap(struct iommu_domain *domain, + unsigned long iova, size_t size) +{ + struct mtk_iommu_domain *dom = to_mtk_domain(domain); + unsigned long flags; + u32 *pgt_base_iova = dom->pgt_va + (iova >> MT2701_IOMMU_PAGE_SHIFT); + unsigned int page_num = size >> MT2701_IOMMU_PAGE_SHIFT; + + spin_lock_irqsave(&dom->pgtlock, flags); + memset(pgt_base_iova, 0, page_num * sizeof(u32)); + spin_unlock_irqrestore(&dom->pgtlock, flags); + + mtk_iommu_tlb_flush_range(dom->data, iova, size); + + return size; +} + +static phys_addr_t mtk_iommu_iova_to_phys(struct iommu_domain *domain, + dma_addr_t iova) +{ + struct mtk_iommu_domain *dom = to_mtk_domain(domain); + unsigned long flags; + phys_addr_t pa; + + spin_lock_irqsave(&dom->pgtlock, flags); + pa = *(dom->pgt_va + (iova >> MT2701_IOMMU_PAGE_SHIFT)); + pa = pa & (~(MT2701_IOMMU_PAGE_SIZE - 1)); + spin_unlock_irqrestore(&dom->pgtlock, flags); + + return pa; +} + +/* + * MTK generation one iommu HW only support one iommu domain, and all the client + * sharing the same iova address space. + */ +static int mtk_iommu_create_mapping(struct device *dev, + struct of_phandle_args *args) +{ + struct mtk_iommu_client_priv *head, *priv, *next; + struct platform_device *m4updev; + struct dma_iommu_mapping *mtk_mapping; + struct device *m4udev; + int ret; + + if (args->args_count != 1) { + dev_err(dev, "invalid #iommu-cells(%d) property for IOMMU\n", + args->args_count); + return -EINVAL; + } + + if (!dev->archdata.iommu) { + /* Get the m4u device */ + m4updev = of_find_device_by_node(args->np); + if (WARN_ON(!m4updev)) + return -EINVAL; + + head = kzalloc(sizeof(*head), GFP_KERNEL); + if (!head) + return -ENOMEM; + + dev->archdata.iommu = head; + INIT_LIST_HEAD(&head->client); + head->m4udev = &m4updev->dev; + } else { + head = dev->archdata.iommu; + } + + priv = kzalloc(sizeof(*priv), GFP_KERNEL); + if (!priv) { + ret = -ENOMEM; + goto err_free_mem; + } + priv->mtk_m4u_id = args->args[0]; + list_add_tail(&priv->client, &head->client); + + m4udev = head->m4udev; + mtk_mapping = m4udev->archdata.iommu; + if (!mtk_mapping) { + /* MTK iommu support 4GB iova address space. */ + mtk_mapping = arm_iommu_create_mapping(&platform_bus_type, + 0, 1ULL << 32); + if (IS_ERR(mtk_mapping)) { + ret = PTR_ERR(mtk_mapping); + goto err_free_mem; + } + m4udev->archdata.iommu = mtk_mapping; + } + + ret = arm_iommu_attach_device(dev, mtk_mapping); + if (ret) + goto err_release_mapping; + + return 0; + +err_release_mapping: + arm_iommu_release_mapping(mtk_mapping); + m4udev->archdata.iommu = NULL; +err_free_mem: + list_for_each_entry_safe(priv, next, &head->client, client) + kfree(priv); + kfree(head); + dev->archdata.iommu = NULL; + return ret; +} + +static int mtk_iommu_add_device(struct device *dev) +{ + struct iommu_group *group; + struct of_phandle_args iommu_spec; + struct of_phandle_iterator it; + int err; + + of_for_each_phandle(&it, err, dev->of_node, "iommus", + "#iommu-cells", 0) { + int count = of_phandle_iterator_args(&it, iommu_spec.args, + MAX_PHANDLE_ARGS); + iommu_spec.np = of_node_get(it.node); + iommu_spec.args_count = count; + + mtk_iommu_create_mapping(dev, &iommu_spec); + of_node_put(iommu_spec.np); + } + + if (!dev->archdata.iommu) /* Not a iommu client device */ + return -ENODEV; + + group = iommu_group_get_for_dev(dev); + if (IS_ERR(group)) + return PTR_ERR(group); + + iommu_group_put(group); + return 0; +} + +static void mtk_iommu_remove_device(struct device *dev) +{ + struct mtk_iommu_client_priv *head, *cur, *next; + + head = dev->archdata.iommu; + if (!head) + return; + + list_for_each_entry_safe(cur, next, &head->client, client) { + list_del(&cur->client); + kfree(cur); + } + kfree(head); + dev->archdata.iommu = NULL; + + iommu_group_remove_device(dev); +} + +static struct iommu_group *mtk_iommu_device_group(struct device *dev) +{ + struct mtk_iommu_data *data; + struct mtk_iommu_client_priv *priv; + + priv = dev->archdata.iommu; + if (!priv) + return ERR_PTR(-ENODEV); + + /* All the client devices are in the same m4u iommu-group */ + data = dev_get_drvdata(priv->m4udev); + if (!data->m4u_group) { + data->m4u_group = iommu_group_alloc(); + if (IS_ERR(data->m4u_group)) + dev_err(dev, "Failed to allocate M4U IOMMU group\n"); + } + return data->m4u_group; +} + +static int mtk_iommu_hw_init(const struct mtk_iommu_data *data) +{ + u32 regval; + int ret; + + ret = clk_prepare_enable(data->bclk); + if (ret) { + dev_err(data->dev, "Failed to enable iommu bclk(%d)\n", ret); + return ret; + } + + regval = F_MMU_CTRL_COHERENT_EN | F_MMU_TF_PROTECT_SEL(2); + writel_relaxed(regval, data->base + REG_MMU_CTRL_REG); + + regval = F_INT_TRANSLATION_FAULT | + F_INT_MAIN_MULTI_HIT_FAULT | + F_INT_INVALID_PA_FAULT | + F_INT_ENTRY_REPLACEMENT_FAULT | + F_INT_TABLE_WALK_FAULT | + F_INT_TLB_MISS_FAULT | + F_INT_PFH_DMA_FIFO_OVERFLOW | + F_INT_MISS_DMA_FIFO_OVERFLOW; + writel_relaxed(regval, data->base + REG_MMU_INT_CONTROL); + + /* protect memory,hw will write here while translation fault */ + writel_relaxed(data->protect_base, + data->base + REG_MMU_IVRP_PADDR); + + writel_relaxed(F_MMU_DCM_ON, data->base + REG_MMU_DCM); + + if (devm_request_irq(data->dev, data->irq, mtk_iommu_isr, 0, + dev_name(data->dev), (void *)data)) { + writel_relaxed(0, data->base + REG_MMU_PT_BASE_ADDR); + clk_disable_unprepare(data->bclk); + dev_err(data->dev, "Failed @ IRQ-%d Request\n", data->irq); + return -ENODEV; + } + + return 0; +} + +static struct iommu_ops mtk_iommu_ops = { + .domain_alloc = mtk_iommu_domain_alloc, + .domain_free = mtk_iommu_domain_free, + .attach_dev = mtk_iommu_attach_device, + .detach_dev = mtk_iommu_detach_device, + .map = mtk_iommu_map, + .unmap = mtk_iommu_unmap, + .map_sg = default_iommu_map_sg, + .iova_to_phys = mtk_iommu_iova_to_phys, + .add_device = mtk_iommu_add_device, + .remove_device = mtk_iommu_remove_device, + .device_group = mtk_iommu_device_group, + .pgsize_bitmap = ~0UL << MT2701_IOMMU_PAGE_SHIFT, +}; + +static const struct of_device_id mtk_iommu_of_ids[] = { + { .compatible = "mediatek,mt2701-m4u", }, + {} +}; + +static const struct component_master_ops mtk_iommu_com_ops = { + .bind = mtk_iommu_bind, + .unbind = mtk_iommu_unbind, +}; + +static int mtk_iommu_probe(struct platform_device *pdev) +{ + struct mtk_iommu_data *data; + struct device *dev = &pdev->dev; + struct resource *res; + struct component_match *match = NULL; + struct of_phandle_args larb_spec; + struct of_phandle_iterator it; + void *protect; + int larb_nr, ret, err; + + data = devm_kzalloc(dev, sizeof(*data), GFP_KERNEL); + if (!data) + return -ENOMEM; + + data->dev = dev; + + /* Protect memory. HW will access here while translation fault.*/ + protect = devm_kzalloc(dev, MTK_PROTECT_PA_ALIGN * 2, + GFP_KERNEL | GFP_DMA); + if (!protect) + return -ENOMEM; + data->protect_base = ALIGN(virt_to_phys(protect), MTK_PROTECT_PA_ALIGN); + + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + data->base = devm_ioremap_resource(dev, res); + if (IS_ERR(data->base)) + return PTR_ERR(data->base); + + data->irq = platform_get_irq(pdev, 0); + if (data->irq < 0) + return data->irq; + + data->bclk = devm_clk_get(dev, "bclk"); + if (IS_ERR(data->bclk)) + return PTR_ERR(data->bclk); + + larb_nr = 0; + of_for_each_phandle(&it, err, dev->of_node, + "mediatek,larbs", NULL, 0) { + struct platform_device *plarbdev; + int count = of_phandle_iterator_args(&it, larb_spec.args, + MAX_PHANDLE_ARGS); + + if (count) + continue; + + larb_spec.np = of_node_get(it.node); + if (!of_device_is_available(larb_spec.np)) + continue; + + plarbdev = of_find_device_by_node(larb_spec.np); + of_node_put(larb_spec.np); + if (!plarbdev) { + plarbdev = of_platform_device_create( + larb_spec.np, NULL, + platform_bus_type.dev_root); + if (!plarbdev) + return -EPROBE_DEFER; + } + + data->smi_imu.larb_imu[larb_nr].dev = &plarbdev->dev; + component_match_add(dev, &match, compare_of, larb_spec.np); + larb_nr++; + } + + data->smi_imu.larb_nr = larb_nr; + + platform_set_drvdata(pdev, data); + + ret = mtk_iommu_hw_init(data); + if (ret) + return ret; + + if (!iommu_present(&platform_bus_type)) + bus_set_iommu(&platform_bus_type, &mtk_iommu_ops); + + return component_master_add_with_match(dev, &mtk_iommu_com_ops, match); +} + +static int mtk_iommu_remove(struct platform_device *pdev) +{ + struct mtk_iommu_data *data = platform_get_drvdata(pdev); + + if (iommu_present(&platform_bus_type)) + bus_set_iommu(&platform_bus_type, NULL); + + clk_disable_unprepare(data->bclk); + devm_free_irq(&pdev->dev, data->irq, data); + component_master_del(&pdev->dev, &mtk_iommu_com_ops); + return 0; +} + +static int __maybe_unused mtk_iommu_suspend(struct device *dev) +{ + struct mtk_iommu_data *data = dev_get_drvdata(dev); + struct mtk_iommu_suspend_reg *reg = &data->reg; + void __iomem *base = data->base; + + reg->standard_axi_mode = readl_relaxed(base + + REG_MMU_STANDARD_AXI_MODE); + reg->dcm_dis = readl_relaxed(base + REG_MMU_DCM); + reg->ctrl_reg = readl_relaxed(base + REG_MMU_CTRL_REG); + reg->int_control0 = readl_relaxed(base + REG_MMU_INT_CONTROL); + return 0; +} + +static int __maybe_unused mtk_iommu_resume(struct device *dev) +{ + struct mtk_iommu_data *data = dev_get_drvdata(dev); + struct mtk_iommu_suspend_reg *reg = &data->reg; + void __iomem *base = data->base; + + writel_relaxed(data->m4u_dom->pgt_pa, base + REG_MMU_PT_BASE_ADDR); + writel_relaxed(reg->standard_axi_mode, + base + REG_MMU_STANDARD_AXI_MODE); + writel_relaxed(reg->dcm_dis, base + REG_MMU_DCM); + writel_relaxed(reg->ctrl_reg, base + REG_MMU_CTRL_REG); + writel_relaxed(reg->int_control0, base + REG_MMU_INT_CONTROL); + writel_relaxed(data->protect_base, base + REG_MMU_IVRP_PADDR); + return 0; +} + +static const struct dev_pm_ops mtk_iommu_pm_ops = { + SET_SYSTEM_SLEEP_PM_OPS(mtk_iommu_suspend, mtk_iommu_resume) +}; + +static struct platform_driver mtk_iommu_driver = { + .probe = mtk_iommu_probe, + .remove = mtk_iommu_remove, + .driver = { + .name = "mtk-iommu", + .of_match_table = mtk_iommu_of_ids, + .pm = &mtk_iommu_pm_ops, + } +}; + +static int __init m4u_init(void) +{ + return platform_driver_register(&mtk_iommu_driver); +} + +static void __exit m4u_exit(void) +{ + return platform_driver_unregister(&mtk_iommu_driver); +} + +subsys_initcall(m4u_init); +module_exit(m4u_exit); + +MODULE_DESCRIPTION("IOMMU API for MTK architected m4u v1 implementations"); +MODULE_AUTHOR("Honghui Zhang <honghui.zhang@mediatek.com>"); +MODULE_LICENSE("GPL v2"); diff --git a/drivers/iommu/rockchip-iommu.c b/drivers/iommu/rockchip-iommu.c index 25b4627cb57f..9afcbf79f0b0 100644 --- a/drivers/iommu/rockchip-iommu.c +++ b/drivers/iommu/rockchip-iommu.c @@ -4,11 +4,10 @@ * published by the Free Software Foundation. */ -#include <asm/cacheflush.h> -#include <asm/pgtable.h> #include <linux/compiler.h> #include <linux/delay.h> #include <linux/device.h> +#include <linux/dma-iommu.h> #include <linux/errno.h> #include <linux/interrupt.h> #include <linux/io.h> @@ -77,7 +76,9 @@ struct rk_iommu_domain { struct list_head iommus; + struct platform_device *pdev; u32 *dt; /* page directory table */ + dma_addr_t dt_dma; spinlock_t iommus_lock; /* lock for iommus list */ spinlock_t dt_lock; /* lock for modifying page directory table */ @@ -93,14 +94,12 @@ struct rk_iommu { struct iommu_domain *domain; /* domain to which iommu is attached */ }; -static inline void rk_table_flush(u32 *va, unsigned int count) +static inline void rk_table_flush(struct rk_iommu_domain *dom, dma_addr_t dma, + unsigned int count) { - phys_addr_t pa_start = virt_to_phys(va); - phys_addr_t pa_end = virt_to_phys(va + count); - size_t size = pa_end - pa_start; + size_t size = count * sizeof(u32); /* count of u32 entry */ - __cpuc_flush_dcache_area(va, size); - outer_flush_range(pa_start, pa_end); + dma_sync_single_for_device(&dom->pdev->dev, dma, size, DMA_TO_DEVICE); } static struct rk_iommu_domain *to_rk_domain(struct iommu_domain *dom) @@ -183,10 +182,9 @@ static inline bool rk_dte_is_pt_valid(u32 dte) return dte & RK_DTE_PT_VALID; } -static u32 rk_mk_dte(u32 *pt) +static inline u32 rk_mk_dte(dma_addr_t pt_dma) { - phys_addr_t pt_phys = virt_to_phys(pt); - return (pt_phys & RK_DTE_PT_ADDRESS_MASK) | RK_DTE_PT_VALID; + return (pt_dma & RK_DTE_PT_ADDRESS_MASK) | RK_DTE_PT_VALID; } /* @@ -603,13 +601,16 @@ static void rk_iommu_zap_iova_first_last(struct rk_iommu_domain *rk_domain, static u32 *rk_dte_get_page_table(struct rk_iommu_domain *rk_domain, dma_addr_t iova) { + struct device *dev = &rk_domain->pdev->dev; u32 *page_table, *dte_addr; - u32 dte; + u32 dte_index, dte; phys_addr_t pt_phys; + dma_addr_t pt_dma; assert_spin_locked(&rk_domain->dt_lock); - dte_addr = &rk_domain->dt[rk_iova_dte_index(iova)]; + dte_index = rk_iova_dte_index(iova); + dte_addr = &rk_domain->dt[dte_index]; dte = *dte_addr; if (rk_dte_is_pt_valid(dte)) goto done; @@ -618,19 +619,27 @@ static u32 *rk_dte_get_page_table(struct rk_iommu_domain *rk_domain, if (!page_table) return ERR_PTR(-ENOMEM); - dte = rk_mk_dte(page_table); - *dte_addr = dte; + pt_dma = dma_map_single(dev, page_table, SPAGE_SIZE, DMA_TO_DEVICE); + if (dma_mapping_error(dev, pt_dma)) { + dev_err(dev, "DMA mapping error while allocating page table\n"); + free_page((unsigned long)page_table); + return ERR_PTR(-ENOMEM); + } - rk_table_flush(page_table, NUM_PT_ENTRIES); - rk_table_flush(dte_addr, 1); + dte = rk_mk_dte(pt_dma); + *dte_addr = dte; + rk_table_flush(rk_domain, pt_dma, NUM_PT_ENTRIES); + rk_table_flush(rk_domain, + rk_domain->dt_dma + dte_index * sizeof(u32), 1); done: pt_phys = rk_dte_pt_address(dte); return (u32 *)phys_to_virt(pt_phys); } static size_t rk_iommu_unmap_iova(struct rk_iommu_domain *rk_domain, - u32 *pte_addr, dma_addr_t iova, size_t size) + u32 *pte_addr, dma_addr_t pte_dma, + size_t size) { unsigned int pte_count; unsigned int pte_total = size / SPAGE_SIZE; @@ -645,14 +654,14 @@ static size_t rk_iommu_unmap_iova(struct rk_iommu_domain *rk_domain, pte_addr[pte_count] = rk_mk_pte_invalid(pte); } - rk_table_flush(pte_addr, pte_count); + rk_table_flush(rk_domain, pte_dma, pte_count); return pte_count * SPAGE_SIZE; } static int rk_iommu_map_iova(struct rk_iommu_domain *rk_domain, u32 *pte_addr, - dma_addr_t iova, phys_addr_t paddr, size_t size, - int prot) + dma_addr_t pte_dma, dma_addr_t iova, + phys_addr_t paddr, size_t size, int prot) { unsigned int pte_count; unsigned int pte_total = size / SPAGE_SIZE; @@ -671,7 +680,7 @@ static int rk_iommu_map_iova(struct rk_iommu_domain *rk_domain, u32 *pte_addr, paddr += SPAGE_SIZE; } - rk_table_flush(pte_addr, pte_count); + rk_table_flush(rk_domain, pte_dma, pte_total); /* * Zap the first and last iova to evict from iotlb any previously @@ -684,7 +693,8 @@ static int rk_iommu_map_iova(struct rk_iommu_domain *rk_domain, u32 *pte_addr, return 0; unwind: /* Unmap the range of iovas that we just mapped */ - rk_iommu_unmap_iova(rk_domain, pte_addr, iova, pte_count * SPAGE_SIZE); + rk_iommu_unmap_iova(rk_domain, pte_addr, pte_dma, + pte_count * SPAGE_SIZE); iova += pte_count * SPAGE_SIZE; page_phys = rk_pte_page_address(pte_addr[pte_count]); @@ -699,8 +709,9 @@ static int rk_iommu_map(struct iommu_domain *domain, unsigned long _iova, { struct rk_iommu_domain *rk_domain = to_rk_domain(domain); unsigned long flags; - dma_addr_t iova = (dma_addr_t)_iova; + dma_addr_t pte_dma, iova = (dma_addr_t)_iova; u32 *page_table, *pte_addr; + u32 dte_index, pte_index; int ret; spin_lock_irqsave(&rk_domain->dt_lock, flags); @@ -718,8 +729,13 @@ static int rk_iommu_map(struct iommu_domain *domain, unsigned long _iova, return PTR_ERR(page_table); } - pte_addr = &page_table[rk_iova_pte_index(iova)]; - ret = rk_iommu_map_iova(rk_domain, pte_addr, iova, paddr, size, prot); + dte_index = rk_domain->dt[rk_iova_dte_index(iova)]; + pte_index = rk_iova_pte_index(iova); + pte_addr = &page_table[pte_index]; + pte_dma = rk_dte_pt_address(dte_index) + pte_index * sizeof(u32); + ret = rk_iommu_map_iova(rk_domain, pte_addr, pte_dma, iova, + paddr, size, prot); + spin_unlock_irqrestore(&rk_domain->dt_lock, flags); return ret; @@ -730,7 +746,7 @@ static size_t rk_iommu_unmap(struct iommu_domain *domain, unsigned long _iova, { struct rk_iommu_domain *rk_domain = to_rk_domain(domain); unsigned long flags; - dma_addr_t iova = (dma_addr_t)_iova; + dma_addr_t pte_dma, iova = (dma_addr_t)_iova; phys_addr_t pt_phys; u32 dte; u32 *pte_addr; @@ -754,7 +770,8 @@ static size_t rk_iommu_unmap(struct iommu_domain *domain, unsigned long _iova, pt_phys = rk_dte_pt_address(dte); pte_addr = (u32 *)phys_to_virt(pt_phys) + rk_iova_pte_index(iova); - unmap_size = rk_iommu_unmap_iova(rk_domain, pte_addr, iova, size); + pte_dma = pt_phys + rk_iova_pte_index(iova) * sizeof(u32); + unmap_size = rk_iommu_unmap_iova(rk_domain, pte_addr, pte_dma, size); spin_unlock_irqrestore(&rk_domain->dt_lock, flags); @@ -787,7 +804,6 @@ static int rk_iommu_attach_device(struct iommu_domain *domain, struct rk_iommu_domain *rk_domain = to_rk_domain(domain); unsigned long flags; int ret, i; - phys_addr_t dte_addr; /* * Allow 'virtual devices' (e.g., drm) to attach to domain. @@ -807,14 +823,14 @@ static int rk_iommu_attach_device(struct iommu_domain *domain, iommu->domain = domain; - ret = devm_request_irq(dev, iommu->irq, rk_iommu_irq, + ret = devm_request_irq(iommu->dev, iommu->irq, rk_iommu_irq, IRQF_SHARED, dev_name(dev), iommu); if (ret) return ret; - dte_addr = virt_to_phys(rk_domain->dt); for (i = 0; i < iommu->num_mmu; i++) { - rk_iommu_write(iommu->bases[i], RK_MMU_DTE_ADDR, dte_addr); + rk_iommu_write(iommu->bases[i], RK_MMU_DTE_ADDR, + rk_domain->dt_dma); rk_iommu_base_command(iommu->bases[i], RK_MMU_CMD_ZAP_CACHE); rk_iommu_write(iommu->bases[i], RK_MMU_INT_MASK, RK_MMU_IRQ_MASK); } @@ -860,7 +876,7 @@ static void rk_iommu_detach_device(struct iommu_domain *domain, } rk_iommu_disable_stall(iommu); - devm_free_irq(dev, iommu->irq, iommu); + devm_free_irq(iommu->dev, iommu->irq, iommu); iommu->domain = NULL; @@ -870,14 +886,30 @@ static void rk_iommu_detach_device(struct iommu_domain *domain, static struct iommu_domain *rk_iommu_domain_alloc(unsigned type) { struct rk_iommu_domain *rk_domain; + struct platform_device *pdev; + struct device *iommu_dev; - if (type != IOMMU_DOMAIN_UNMANAGED) + if (type != IOMMU_DOMAIN_UNMANAGED && type != IOMMU_DOMAIN_DMA) return NULL; - rk_domain = kzalloc(sizeof(*rk_domain), GFP_KERNEL); - if (!rk_domain) + /* Register a pdev per domain, so DMA API can base on this *dev + * even some virtual master doesn't have an iommu slave + */ + pdev = platform_device_register_simple("rk_iommu_domain", + PLATFORM_DEVID_AUTO, NULL, 0); + if (IS_ERR(pdev)) return NULL; + rk_domain = devm_kzalloc(&pdev->dev, sizeof(*rk_domain), GFP_KERNEL); + if (!rk_domain) + goto err_unreg_pdev; + + rk_domain->pdev = pdev; + + if (type == IOMMU_DOMAIN_DMA && + iommu_get_dma_cookie(&rk_domain->domain)) + goto err_unreg_pdev; + /* * rk32xx iommus use a 2 level pagetable. * Each level1 (dt) and level2 (pt) table has 1024 4-byte entries. @@ -885,18 +917,36 @@ static struct iommu_domain *rk_iommu_domain_alloc(unsigned type) */ rk_domain->dt = (u32 *)get_zeroed_page(GFP_KERNEL | GFP_DMA32); if (!rk_domain->dt) - goto err_dt; + goto err_put_cookie; + + iommu_dev = &pdev->dev; + rk_domain->dt_dma = dma_map_single(iommu_dev, rk_domain->dt, + SPAGE_SIZE, DMA_TO_DEVICE); + if (dma_mapping_error(iommu_dev, rk_domain->dt_dma)) { + dev_err(iommu_dev, "DMA map error for DT\n"); + goto err_free_dt; + } - rk_table_flush(rk_domain->dt, NUM_DT_ENTRIES); + rk_table_flush(rk_domain, rk_domain->dt_dma, NUM_DT_ENTRIES); spin_lock_init(&rk_domain->iommus_lock); spin_lock_init(&rk_domain->dt_lock); INIT_LIST_HEAD(&rk_domain->iommus); + rk_domain->domain.geometry.aperture_start = 0; + rk_domain->domain.geometry.aperture_end = DMA_BIT_MASK(32); + rk_domain->domain.geometry.force_aperture = true; + return &rk_domain->domain; -err_dt: - kfree(rk_domain); +err_free_dt: + free_page((unsigned long)rk_domain->dt); +err_put_cookie: + if (type == IOMMU_DOMAIN_DMA) + iommu_put_dma_cookie(&rk_domain->domain); +err_unreg_pdev: + platform_device_unregister(pdev); + return NULL; } @@ -912,12 +962,20 @@ static void rk_iommu_domain_free(struct iommu_domain *domain) if (rk_dte_is_pt_valid(dte)) { phys_addr_t pt_phys = rk_dte_pt_address(dte); u32 *page_table = phys_to_virt(pt_phys); + dma_unmap_single(&rk_domain->pdev->dev, pt_phys, + SPAGE_SIZE, DMA_TO_DEVICE); free_page((unsigned long)page_table); } } + dma_unmap_single(&rk_domain->pdev->dev, rk_domain->dt_dma, + SPAGE_SIZE, DMA_TO_DEVICE); free_page((unsigned long)rk_domain->dt); - kfree(rk_domain); + + if (domain->type == IOMMU_DOMAIN_DMA) + iommu_put_dma_cookie(&rk_domain->domain); + + platform_device_unregister(rk_domain->pdev); } static bool rk_iommu_is_dev_iommu_master(struct device *dev) @@ -1022,17 +1080,43 @@ static const struct iommu_ops rk_iommu_ops = { .detach_dev = rk_iommu_detach_device, .map = rk_iommu_map, .unmap = rk_iommu_unmap, + .map_sg = default_iommu_map_sg, .add_device = rk_iommu_add_device, .remove_device = rk_iommu_remove_device, .iova_to_phys = rk_iommu_iova_to_phys, .pgsize_bitmap = RK_IOMMU_PGSIZE_BITMAP, }; +static int rk_iommu_domain_probe(struct platform_device *pdev) +{ + struct device *dev = &pdev->dev; + + dev->dma_parms = devm_kzalloc(dev, sizeof(*dev->dma_parms), GFP_KERNEL); + if (!dev->dma_parms) + return -ENOMEM; + + /* Set dma_ops for dev, otherwise it would be dummy_dma_ops */ + arch_setup_dma_ops(dev, 0, DMA_BIT_MASK(32), NULL, false); + + dma_set_max_seg_size(dev, DMA_BIT_MASK(32)); + dma_coerce_mask_and_coherent(dev, DMA_BIT_MASK(32)); + + return 0; +} + +static struct platform_driver rk_iommu_domain_driver = { + .probe = rk_iommu_domain_probe, + .driver = { + .name = "rk_iommu_domain", + }, +}; + static int rk_iommu_probe(struct platform_device *pdev) { struct device *dev = &pdev->dev; struct rk_iommu *iommu; struct resource *res; + int num_res = pdev->num_resources; int i; iommu = devm_kzalloc(dev, sizeof(*iommu), GFP_KERNEL); @@ -1042,12 +1126,13 @@ static int rk_iommu_probe(struct platform_device *pdev) platform_set_drvdata(pdev, iommu); iommu->dev = dev; iommu->num_mmu = 0; - iommu->bases = devm_kzalloc(dev, sizeof(*iommu->bases) * iommu->num_mmu, + + iommu->bases = devm_kzalloc(dev, sizeof(*iommu->bases) * num_res, GFP_KERNEL); if (!iommu->bases) return -ENOMEM; - for (i = 0; i < pdev->num_resources; i++) { + for (i = 0; i < num_res; i++) { res = platform_get_resource(pdev, IORESOURCE_MEM, i); if (!res) continue; @@ -1103,11 +1188,19 @@ static int __init rk_iommu_init(void) if (ret) return ret; - return platform_driver_register(&rk_iommu_driver); + ret = platform_driver_register(&rk_iommu_domain_driver); + if (ret) + return ret; + + ret = platform_driver_register(&rk_iommu_driver); + if (ret) + platform_driver_unregister(&rk_iommu_domain_driver); + return ret; } static void __exit rk_iommu_exit(void) { platform_driver_unregister(&rk_iommu_driver); + platform_driver_unregister(&rk_iommu_domain_driver); } subsys_initcall(rk_iommu_init); |