diff options
Diffstat (limited to 'drivers/gpu/drm/xe/xe_bo.h')
| -rw-r--r-- | drivers/gpu/drm/xe/xe_bo.h | 383 |
1 files changed, 330 insertions, 53 deletions
diff --git a/drivers/gpu/drm/xe/xe_bo.h b/drivers/gpu/drm/xe/xe_bo.h index d9386ab03140..152bfcffe399 100644 --- a/drivers/gpu/drm/xe/xe_bo.h +++ b/drivers/gpu/drm/xe/xe_bo.h @@ -6,12 +6,18 @@ #ifndef _XE_BO_H_ #define _XE_BO_H_ +#include <drm/drm_prime.h> #include <drm/ttm/ttm_tt.h> +#include <xen/xen.h> + #include "xe_bo_types.h" +#include "xe_ggtt.h" #include "xe_macros.h" +#include "xe_validation.h" #include "xe_vm_types.h" #include "xe_vm.h" +#include "xe_vram_types.h" #define XE_DEFAULT_GTT_SIZE_MB 3072ULL /* 3GB by default */ @@ -23,15 +29,16 @@ #define XE_BO_FLAG_VRAM_MASK (XE_BO_FLAG_VRAM0 | XE_BO_FLAG_VRAM1) /* -- */ #define XE_BO_FLAG_STOLEN BIT(4) +#define XE_BO_FLAG_VRAM(vram) (XE_BO_FLAG_VRAM0 << ((vram)->id)) #define XE_BO_FLAG_VRAM_IF_DGFX(tile) (IS_DGFX(tile_to_xe(tile)) ? \ - XE_BO_FLAG_VRAM0 << (tile)->id : \ + XE_BO_FLAG_VRAM((tile)->mem.vram) : \ XE_BO_FLAG_SYSTEM) #define XE_BO_FLAG_GGTT BIT(5) #define XE_BO_FLAG_IGNORE_MIN_PAGE_SIZE BIT(6) #define XE_BO_FLAG_PINNED BIT(7) #define XE_BO_FLAG_NO_RESV_EVICT BIT(8) #define XE_BO_FLAG_DEFER_BACKING BIT(9) -#define XE_BO_FLAG_SCANOUT BIT(10) +#define XE_BO_FLAG_FORCE_WC BIT(10) #define XE_BO_FLAG_FIXED_PLACEMENT BIT(11) #define XE_BO_FLAG_PAGETABLE BIT(12) #define XE_BO_FLAG_NEEDS_CPU_ACCESS BIT(13) @@ -39,19 +46,26 @@ #define XE_BO_FLAG_NEEDS_64K BIT(15) #define XE_BO_FLAG_NEEDS_2M BIT(16) #define XE_BO_FLAG_GGTT_INVALIDATE BIT(17) -#define XE_BO_FLAG_GGTT0 BIT(18) -#define XE_BO_FLAG_GGTT1 BIT(19) -#define XE_BO_FLAG_GGTT2 BIT(20) -#define XE_BO_FLAG_GGTT3 BIT(21) -#define XE_BO_FLAG_GGTT_ALL (XE_BO_FLAG_GGTT0 | \ - XE_BO_FLAG_GGTT1 | \ - XE_BO_FLAG_GGTT2 | \ - XE_BO_FLAG_GGTT3) +#define XE_BO_FLAG_PINNED_NORESTORE BIT(18) +#define XE_BO_FLAG_PINNED_LATE_RESTORE BIT(19) +#define XE_BO_FLAG_GGTT0 BIT(20) +#define XE_BO_FLAG_GGTT1 BIT(21) +#define XE_BO_FLAG_GGTT2 BIT(22) +#define XE_BO_FLAG_GGTT3 BIT(23) +#define XE_BO_FLAG_CPU_ADDR_MIRROR BIT(24) +#define XE_BO_FLAG_FORCE_USER_VRAM BIT(25) +#define XE_BO_FLAG_NO_COMPRESSION BIT(26) +#define XE_BO_FLAG_NEEDS_1G BIT(27) /* this one is trigger internally only */ #define XE_BO_FLAG_INTERNAL_TEST BIT(30) #define XE_BO_FLAG_INTERNAL_64K BIT(31) +#define XE_BO_FLAG_GGTT_ALL (XE_BO_FLAG_GGTT0 | \ + XE_BO_FLAG_GGTT1 | \ + XE_BO_FLAG_GGTT2 | \ + XE_BO_FLAG_GGTT3) + #define XE_BO_FLAG_GGTTx(tile) \ (XE_BO_FLAG_GGTT0 << (tile)->id) @@ -75,54 +89,70 @@ #define XE_BO_PROPS_INVALID (-1) +#define XE_PCI_BARRIER_MMAP_OFFSET (0x50 << XE_PTE_SHIFT) + +/** + * enum xe_madv_purgeable_state - Buffer object purgeable state enumeration + * + * This enum defines the possible purgeable states for a buffer object, + * allowing userspace to provide memory usage hints to the kernel for + * better memory management under pressure. + * + * @XE_MADV_PURGEABLE_WILLNEED: The buffer object is needed and should not be purged. + * This is the default state. + * @XE_MADV_PURGEABLE_DONTNEED: The buffer object is not currently needed and can be + * purged by the kernel under memory pressure. + * @XE_MADV_PURGEABLE_PURGED: The buffer object has been purged by the kernel. + * + * Accessing a purged buffer will result in an error. Per i915 semantics, + * once purged, a BO remains permanently invalid and must be destroyed and recreated. + */ +enum xe_madv_purgeable_state { + XE_MADV_PURGEABLE_WILLNEED, + XE_MADV_PURGEABLE_DONTNEED, + XE_MADV_PURGEABLE_PURGED, +}; + struct sg_table; struct xe_bo *xe_bo_alloc(void); void xe_bo_free(struct xe_bo *bo); -struct xe_bo *___xe_bo_create_locked(struct xe_device *xe, struct xe_bo *bo, - struct xe_tile *tile, struct dma_resv *resv, - struct ttm_lru_bulk_move *bulk, size_t size, - u16 cpu_caching, enum ttm_bo_type type, - u32 flags); -struct xe_bo * -xe_bo_create_locked_range(struct xe_device *xe, - struct xe_tile *tile, struct xe_vm *vm, - size_t size, u64 start, u64 end, - enum ttm_bo_type type, u32 flags, u64 alignment); +struct xe_bo *xe_bo_init_locked(struct xe_device *xe, struct xe_bo *bo, + struct xe_tile *tile, struct dma_resv *resv, + struct ttm_lru_bulk_move *bulk, size_t size, + u16 cpu_caching, enum ttm_bo_type type, + u32 flags, struct dma_buf *dma_buf, + struct drm_exec *exec); struct xe_bo *xe_bo_create_locked(struct xe_device *xe, struct xe_tile *tile, struct xe_vm *vm, size_t size, - enum ttm_bo_type type, u32 flags); -struct xe_bo *xe_bo_create(struct xe_device *xe, struct xe_tile *tile, - struct xe_vm *vm, size_t size, - enum ttm_bo_type type, u32 flags); -struct xe_bo *xe_bo_create_user(struct xe_device *xe, struct xe_tile *tile, - struct xe_vm *vm, size_t size, - u16 cpu_caching, - u32 flags); + enum ttm_bo_type type, u32 flags, + struct drm_exec *exec); +struct xe_bo *xe_bo_create_user(struct xe_device *xe, struct xe_vm *vm, size_t size, + u16 cpu_caching, u32 flags, struct drm_exec *exec); struct xe_bo *xe_bo_create_pin_map(struct xe_device *xe, struct xe_tile *tile, struct xe_vm *vm, size_t size, - enum ttm_bo_type type, u32 flags); -struct xe_bo *xe_bo_create_pin_map_at(struct xe_device *xe, struct xe_tile *tile, - struct xe_vm *vm, size_t size, u64 offset, - enum ttm_bo_type type, u32 flags); -struct xe_bo *xe_bo_create_pin_map_at_aligned(struct xe_device *xe, - struct xe_tile *tile, - struct xe_vm *vm, - size_t size, u64 offset, - enum ttm_bo_type type, u32 flags, - u64 alignment); -struct xe_bo *xe_bo_create_from_data(struct xe_device *xe, struct xe_tile *tile, - const void *data, size_t size, - enum ttm_bo_type type, u32 flags); + enum ttm_bo_type type, u32 flags, + struct drm_exec *exec); +struct xe_bo *xe_bo_create_pin_map_novm(struct xe_device *xe, struct xe_tile *tile, + size_t size, enum ttm_bo_type type, u32 flags, + bool intr); +struct xe_bo *xe_bo_create_pin_range_novm(struct xe_device *xe, struct xe_tile *tile, + size_t size, u64 start, u64 end, + enum ttm_bo_type type, u32 flags); +struct xe_bo * +xe_bo_create_pin_map_at_novm(struct xe_device *xe, struct xe_tile *tile, + size_t size, u64 offset, enum ttm_bo_type type, + u32 flags, u64 alignment, bool intr); struct xe_bo *xe_managed_bo_create_pin_map(struct xe_device *xe, struct xe_tile *tile, size_t size, u32 flags); +void xe_managed_bo_unpin_map_no_vm(struct xe_bo *bo); struct xe_bo *xe_managed_bo_create_from_data(struct xe_device *xe, struct xe_tile *tile, const void *data, size_t size, u32 flags); int xe_managed_bo_reinit_in_vram(struct xe_device *xe, struct xe_tile *tile, struct xe_bo **src); int xe_bo_placement_for_flags(struct xe_device *xe, struct xe_bo *bo, - u32 bo_flags); + u32 bo_flags, enum ttm_bo_type type); static inline struct xe_bo *ttm_to_xe_bo(const struct ttm_buffer_object *bo) { @@ -146,6 +176,28 @@ static inline struct xe_bo *xe_bo_get(struct xe_bo *bo) void xe_bo_put(struct xe_bo *bo); +/* + * xe_bo_get_unless_zero() - Conditionally obtain a GEM object refcount on an + * xe bo + * @bo: The bo for which we want to obtain a refcount. + * + * There is a short window between where the bo's GEM object refcount reaches + * zero and where we put the final ttm_bo reference. Code in the eviction- and + * shrinking path should therefore attempt to grab a gem object reference before + * trying to use members outside of the base class ttm object. This function is + * intended for that purpose. On successful return, this function must be paired + * with an xe_bo_put(). + * + * Return: @bo on success, NULL on failure. + */ +static inline __must_check struct xe_bo *xe_bo_get_unless_zero(struct xe_bo *bo) +{ + if (!bo || !kref_get_unless_zero(&bo->ttm.base.refcount)) + return NULL; + + return bo; +} + static inline void __xe_bo_unset_bulk_move(struct xe_bo *bo) { if (bo) @@ -173,17 +225,143 @@ static inline void xe_bo_unlock_vm_held(struct xe_bo *bo) } } -int xe_bo_pin_external(struct xe_bo *bo); -int xe_bo_pin(struct xe_bo *bo); +int xe_bo_pin_external(struct xe_bo *bo, bool in_place, struct drm_exec *exec); +int xe_bo_pin(struct xe_bo *bo, struct drm_exec *exec); void xe_bo_unpin_external(struct xe_bo *bo); void xe_bo_unpin(struct xe_bo *bo); -int xe_bo_validate(struct xe_bo *bo, struct xe_vm *vm, bool allow_res_evict); +int xe_bo_validate(struct xe_bo *bo, struct xe_vm *vm, bool allow_res_evict, + struct drm_exec *exec); static inline bool xe_bo_is_pinned(struct xe_bo *bo) { return bo->ttm.pin_count; } +static inline bool xe_bo_is_protected(const struct xe_bo *bo) +{ + return bo->pxp_key_instance; +} + +/** + * xe_bo_is_purged() - Check if buffer object has been purged + * @bo: The buffer object to check + * + * Checks if the buffer object's backing store has been discarded by the + * kernel due to memory pressure after being marked as purgeable (DONTNEED). + * Once purged, the BO cannot be restored and any attempt to use it will fail. + * + * Context: Caller must hold the BO's dma-resv lock + * Return: true if the BO has been purged, false otherwise + */ +static inline bool xe_bo_is_purged(struct xe_bo *bo) +{ + xe_bo_assert_held(bo); + return bo->purgeable.state == XE_MADV_PURGEABLE_PURGED; +} + +/** + * xe_bo_madv_is_dontneed() - Check if BO is marked as DONTNEED + * @bo: The buffer object to check + * + * Checks if userspace has marked this BO as DONTNEED (i.e., its contents + * are not currently needed and can be discarded under memory pressure). + * This is used internally to decide whether a BO is eligible for purging. + * + * Context: Caller must hold the BO's dma-resv lock + * Return: true if the BO is marked DONTNEED, false otherwise + */ +static inline bool xe_bo_madv_is_dontneed(struct xe_bo *bo) +{ + xe_bo_assert_held(bo); + return bo->purgeable.state == XE_MADV_PURGEABLE_DONTNEED; +} + +void xe_bo_set_purgeable_state(struct xe_bo *bo, enum xe_madv_purgeable_state new_state); + +/** + * xe_bo_willneed_get_locked() - Acquire a WILLNEED holder on a BO + * @bo: Buffer object + * + * Increments willneed_count and, on a 0->1 transition, promotes the BO + * from DONTNEED to WILLNEED. PURGED is terminal and is never modified. + * + * Caller must hold the BO's dma-resv lock. + */ +static inline void xe_bo_willneed_get_locked(struct xe_bo *bo) +{ + xe_bo_assert_held(bo); + + /* Imported BOs are owned externally; do not track purgeability. */ + if (drm_gem_is_imported(&bo->ttm.base)) + return; + + if (bo->purgeable.willneed_count++ == 0 && xe_bo_madv_is_dontneed(bo)) + xe_bo_set_purgeable_state(bo, XE_MADV_PURGEABLE_WILLNEED); +} + +/** + * xe_bo_willneed_put_locked() - Release a WILLNEED holder on a BO + * @bo: Buffer object + * + * Decrements willneed_count and, on a 1->0 transition, marks the BO + * DONTNEED only if it still has VMAs (implying all active VMAs are + * DONTNEED). If the last VMA is being removed, preserve the current BO + * state to match the previous VMA-walk semantics. + * + * PURGED is terminal and the BO state is never modified. + * + * Caller must hold the BO's dma-resv lock. + */ +static inline void xe_bo_willneed_put_locked(struct xe_bo *bo) +{ + xe_bo_assert_held(bo); + + if (drm_gem_is_imported(&bo->ttm.base)) + return; + + xe_assert(xe_bo_device(bo), bo->purgeable.willneed_count > 0); + if (--bo->purgeable.willneed_count == 0 && bo->purgeable.vma_count > 0 && + !xe_bo_is_purged(bo)) + xe_bo_set_purgeable_state(bo, XE_MADV_PURGEABLE_DONTNEED); +} + +/** + * xe_bo_vma_count_inc_locked() - Account a new VMA on a BO + * @bo: Buffer object + * + * Increments vma_count. + * + * Caller must hold the BO's dma-resv lock. + */ +static inline void xe_bo_vma_count_inc_locked(struct xe_bo *bo) +{ + xe_bo_assert_held(bo); + + if (drm_gem_is_imported(&bo->ttm.base)) + return; + + bo->purgeable.vma_count++; +} + +/** + * xe_bo_vma_count_dec_locked() - Account a VMA removal on a BO + * @bo: Buffer object + * + * Decrements vma_count. + * + * Caller must hold the BO's dma-resv lock. + */ +static inline void xe_bo_vma_count_dec_locked(struct xe_bo *bo) +{ + xe_bo_assert_held(bo); + + if (drm_gem_is_imported(&bo->ttm.base)) + return; + + xe_assert(xe_bo_device(bo), bo->purgeable.vma_count > 0); + bo->purgeable.vma_count--; +} + static inline void xe_bo_unpin_map_no_vm(struct xe_bo *bo) { if (likely(bo)) { @@ -205,17 +383,31 @@ xe_bo_main_addr(struct xe_bo *bo, size_t page_size) return xe_bo_addr(bo, 0, page_size); } +/** + * xe_bo_size() - Xe BO size + * @bo: The bo object. + * + * Simple helper to return Xe BO's size. + * + * Return: Xe BO's size + */ +static inline size_t xe_bo_size(struct xe_bo *bo) +{ + return bo->ttm.base.size; +} + static inline u32 __xe_bo_ggtt_addr(struct xe_bo *bo, u8 tile_id) { struct xe_ggtt_node *ggtt_node = bo->ggtt_node[tile_id]; + u64 offset; if (XE_WARN_ON(!ggtt_node)) return 0; - XE_WARN_ON(ggtt_node->base.size > bo->size); - XE_WARN_ON(ggtt_node->base.start + ggtt_node->base.size > (1ull << 32)); - return ggtt_node->base.start; + offset = xe_ggtt_node_addr(ggtt_node); + XE_WARN_ON(offset + xe_bo_size(bo) > (1ull << 32)); + return offset; } static inline u32 @@ -232,19 +424,26 @@ int xe_bo_read(struct xe_bo *bo, u64 offset, void *dst, int size); bool mem_type_is_vram(u32 mem_type); bool xe_bo_is_vram(struct xe_bo *bo); +bool xe_bo_is_visible_vram(struct xe_bo *bo); bool xe_bo_is_stolen(struct xe_bo *bo); bool xe_bo_is_stolen_devmem(struct xe_bo *bo); +bool xe_bo_is_vm_bound(struct xe_bo *bo); bool xe_bo_has_single_placement(struct xe_bo *bo); uint64_t vram_region_gpu_offset(struct ttm_resource *res); bool xe_bo_can_migrate(struct xe_bo *bo, u32 mem_type); -int xe_bo_migrate(struct xe_bo *bo, u32 mem_type); -int xe_bo_evict(struct xe_bo *bo, bool force_alloc); +int xe_bo_migrate(struct xe_bo *bo, u32 mem_type, struct ttm_operation_ctx *ctc, + struct drm_exec *exec); +int xe_bo_evict(struct xe_bo *bo, struct drm_exec *exec); int xe_bo_evict_pinned(struct xe_bo *bo); +int xe_bo_notifier_prepare_pinned(struct xe_bo *bo); +int xe_bo_notifier_unprepare_pinned(struct xe_bo *bo); int xe_bo_restore_pinned(struct xe_bo *bo); +int xe_bo_dma_unmap_pinned(struct xe_bo *bo); + extern const struct ttm_device_funcs xe_ttm_funcs; extern const char *const xe_mem_type_to_name[]; @@ -260,9 +459,26 @@ int xe_bo_dumb_create(struct drm_file *file_priv, bool xe_bo_needs_ccs_pages(struct xe_bo *bo); +int xe_bo_decompress(struct xe_bo *bo); + static inline size_t xe_bo_ccs_pages_start(struct xe_bo *bo) { - return PAGE_ALIGN(bo->ttm.base.size); + return PAGE_ALIGN(xe_bo_size(bo)); +} + +/** + * xe_bo_has_valid_ccs_bb - Check if CCS's BBs were setup for the BO. + * @bo: the &xe_bo to check + * + * The CCS's BBs should only be setup by the driver VF, but it is safe + * to call this function also by non-VF driver. + * + * Return: true iff the CCS's BBs are setup, false otherwise. + */ +static inline bool xe_bo_has_valid_ccs_bb(struct xe_bo *bo) +{ + return bo->bb_ccs[XE_SRIOV_VF_CCS_READ_CTX] && + bo->bb_ccs[XE_SRIOV_VF_CCS_WRITE_CTX]; } static inline bool xe_bo_has_pages(struct xe_bo *bo) @@ -315,8 +531,40 @@ xe_bo_put_deferred(struct xe_bo *bo, struct llist_head *deferred) void xe_bo_put_commit(struct llist_head *deferred); +/** + * xe_bo_put_async() - Put BO async + * @bo: The bo to put. + * + * Put BO async, the final put is deferred to a worker to exit an IRQ context. + */ +static inline void +xe_bo_put_async(struct xe_bo *bo) +{ + struct xe_bo_dev *bo_device = &xe_bo_device(bo)->bo_device; + + if (xe_bo_put_deferred(bo, &bo_device->async_list)) + schedule_work(&bo_device->async_free); +} + +void xe_bo_dev_init(struct xe_bo_dev *bo_device); + +void xe_bo_dev_fini(struct xe_bo_dev *bo_device); + struct sg_table *xe_bo_sg(struct xe_bo *bo); +/** + * xe_bo_sg_is_contiguous() - Check if a BO's DMA address space is contiguous. + * @bo: the BO to check (must have a valid sg table, i.e. !xe_bo_is_vram()) + * @len: required contiguous length in bytes + * + * Returns true if the first @len bytes of the BO are mapped to a contiguous + * DMA address range. + */ +static inline bool xe_bo_sg_is_contiguous(struct xe_bo *bo, size_t len) +{ + return drm_prime_get_contiguous_size(xe_bo_sg(bo)) >= len; +} + /* * xe_sg_segment_size() - Provides upper limit for sg segment size. * @dev: device pointer @@ -329,6 +577,23 @@ static inline unsigned int xe_sg_segment_size(struct device *dev) struct scatterlist __maybe_unused sg; size_t max = BIT_ULL(sizeof(sg.length) * 8) - 1; + /* + * For Xen PV guests pages aren't contiguous in DMA (machine) address + * space. The DMA API takes care of that both in dma_alloc_* (by + * calling into the hypervisor to make the pages contiguous) and in + * dma_map_* (by bounce buffering). But xe (like i915, see commit + * 78a07fe777c4) ignores the coherency aspects of the DMA API and thus + * can't cope with bounce buffering actually happening, so add a hack + * here to force small allocations and mappings when running in PV + * mode on Xen. + * + * Note this will still break if bounce buffering is required for other + * reasons, like confidential computing hypervisors or PCIe root ports + * with addressing limitations. + */ + if (xen_pv_domain()) + return PAGE_SIZE; + max = min_t(size_t, max, dma_max_mapping_size(dev)); /* @@ -341,7 +606,20 @@ static inline unsigned int xe_sg_segment_size(struct device *dev) return round_down(max / 2, PAGE_SIZE); } -#if IS_ENABLED(CONFIG_DRM_XE_KUNIT_TEST) +/** + * struct xe_bo_shrink_flags - flags governing the shrink behaviour. + * @purge: Only purging allowed. Don't shrink if bo not purgeable. + * @writeback: Attempt to immediately move content to swap. + */ +struct xe_bo_shrink_flags { + u32 purge : 1; + u32 writeback : 1; +}; + +long xe_bo_shrink(struct ttm_operation_ctx *ctx, struct ttm_buffer_object *bo, + const struct xe_bo_shrink_flags flags, + unsigned long *scanned); + /** * xe_bo_is_mem_type - Whether the bo currently resides in the given * TTM memory type @@ -356,4 +634,3 @@ static inline bool xe_bo_is_mem_type(struct xe_bo *bo, u32 mem_type) return bo->ttm.resource->mem_type == mem_type; } #endif -#endif |
