diff options
Diffstat (limited to 'common/recipes-kernel/linux/files/0508-drm-amdgpu-group-BOs-by-log2-of-the-size-on-the-LRU-.patch')
| -rw-r--r-- | common/recipes-kernel/linux/files/0508-drm-amdgpu-group-BOs-by-log2-of-the-size-on-the-LRU-.patch | 144 |
1 files changed, 144 insertions, 0 deletions
diff --git a/common/recipes-kernel/linux/files/0508-drm-amdgpu-group-BOs-by-log2-of-the-size-on-the-LRU-.patch b/common/recipes-kernel/linux/files/0508-drm-amdgpu-group-BOs-by-log2-of-the-size-on-the-LRU-.patch new file mode 100644 index 00000000..3ef61152 --- /dev/null +++ b/common/recipes-kernel/linux/files/0508-drm-amdgpu-group-BOs-by-log2-of-the-size-on-the-LRU-.patch @@ -0,0 +1,144 @@ +From 9cc9a4472afca6cc46f79dfafe53787a200c6121 Mon Sep 17 00:00:00 2001 +From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com> +Date: Fri, 15 Apr 2016 17:19:16 +0200 +Subject: [PATCH 0508/1110] drm/amdgpu: group BOs by log2 of the size on the + LRU v2 +MIME-Version: 1.0 +Content-Type: text/plain; charset=UTF-8 +Content-Transfer-Encoding: 8bit + +This allows us to have small BOs on the LRU before big ones. + +v2: fix of by one and list corruption bug + +Signed-off-by: Christian König <christian.koenig@amd.com> +Reviewed-by: Alex Deucher <alexander.deucher@amd.com> +Signed-off-by: Alex Deucher <alexander.deucher@amd.com> +Signed-off-by: Kalyan Alle <kalyan.alle@amd.com> +--- + drivers/gpu/drm/amd/amdgpu/amdgpu.h | 11 ++++++ + drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 61 +++++++++++++++++++++++++++++++-- + 2 files changed, 70 insertions(+), 2 deletions(-) + +diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h +index 0214003..06cf1eb 100644 +--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h ++++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h +@@ -393,6 +393,14 @@ unsigned amdgpu_fence_count_emitted(struct amdgpu_ring *ring); + /* + * TTM. + */ ++ ++#define AMDGPU_TTM_LRU_SIZE 20 ++ ++struct amdgpu_mman_lru { ++ struct list_head *lru[TTM_NUM_MEM_TYPES]; ++ struct list_head *swap_lru; ++}; ++ + struct amdgpu_mman { + struct ttm_bo_global_ref bo_global_ref; + struct drm_global_reference mem_global_ref; +@@ -410,6 +418,9 @@ struct amdgpu_mman { + struct amdgpu_ring *buffer_funcs_ring; + /* Scheduler entity for buffer moves */ + struct amd_sched_entity entity; ++ ++ /* custom LRU management */ ++ struct amdgpu_mman_lru log2_size[AMDGPU_TTM_LRU_SIZE]; + }; + + int amdgpu_copy_buffer(struct amdgpu_ring *ring, +diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +index 1ec9491..70f005d 100644 +--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c ++++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +@@ -910,6 +910,52 @@ uint32_t amdgpu_ttm_tt_pte_flags(struct amdgpu_device *adev, struct ttm_tt *ttm, + return flags; + } + ++static void amdgpu_ttm_lru_removal(struct ttm_buffer_object *tbo) ++{ ++ struct amdgpu_device *adev = amdgpu_get_adev(tbo->bdev); ++ unsigned i, j; ++ ++ for (i = 0; i < AMDGPU_TTM_LRU_SIZE; ++i) { ++ struct amdgpu_mman_lru *lru = &adev->mman.log2_size[i]; ++ ++ for (j = 0; j < TTM_NUM_MEM_TYPES; ++j) ++ if (&tbo->lru == lru->lru[j]) ++ lru->lru[j] = tbo->lru.prev; ++ ++ if (&tbo->swap == lru->swap_lru) ++ lru->swap_lru = tbo->swap.prev; ++ } ++} ++ ++static struct amdgpu_mman_lru *amdgpu_ttm_lru(struct ttm_buffer_object *tbo) ++{ ++ struct amdgpu_device *adev = amdgpu_get_adev(tbo->bdev); ++ unsigned log2_size = min(ilog2(tbo->num_pages), ++ AMDGPU_TTM_LRU_SIZE - 1); ++ ++ return &adev->mman.log2_size[log2_size]; ++} ++ ++static struct list_head *amdgpu_ttm_lru_tail(struct ttm_buffer_object *tbo) ++{ ++ struct amdgpu_mman_lru *lru = amdgpu_ttm_lru(tbo); ++ struct list_head *res = lru->lru[tbo->mem.mem_type]; ++ ++ lru->lru[tbo->mem.mem_type] = &tbo->lru; ++ ++ return res; ++} ++ ++static struct list_head *amdgpu_ttm_swap_lru_tail(struct ttm_buffer_object *tbo) ++{ ++ struct amdgpu_mman_lru *lru = amdgpu_ttm_lru(tbo); ++ struct list_head *res = lru->swap_lru; ++ ++ lru->swap_lru = &tbo->swap; ++ ++ return res; ++} ++ + static struct ttm_bo_driver amdgpu_bo_driver = { + .ttm_tt_create = &amdgpu_ttm_tt_create, + .ttm_tt_populate = &amdgpu_ttm_tt_populate, +@@ -923,12 +969,14 @@ static struct ttm_bo_driver amdgpu_bo_driver = { + .fault_reserve_notify = &amdgpu_bo_fault_reserve_notify, + .io_mem_reserve = &amdgpu_ttm_io_mem_reserve, + .io_mem_free = &amdgpu_ttm_io_mem_free, +- .lru_tail = &ttm_bo_default_lru_tail, +- .swap_lru_tail = &ttm_bo_default_swap_lru_tail, ++ .lru_removal = &amdgpu_ttm_lru_removal, ++ .lru_tail = &amdgpu_ttm_lru_tail, ++ .swap_lru_tail = &amdgpu_ttm_swap_lru_tail, + }; + + int amdgpu_ttm_init(struct amdgpu_device *adev) + { ++ unsigned i, j; + int r; + + r = amdgpu_ttm_global_init(adev); +@@ -946,6 +994,15 @@ int amdgpu_ttm_init(struct amdgpu_device *adev) + DRM_ERROR("failed initializing buffer object driver(%d).\n", r); + return r; + } ++ ++ for (i = 0; i < AMDGPU_TTM_LRU_SIZE; ++i) { ++ struct amdgpu_mman_lru *lru = &adev->mman.log2_size[i]; ++ ++ for (j = 0; j < TTM_NUM_MEM_TYPES; ++j) ++ lru->lru[j] = &adev->mman.bdev.man[j].lru; ++ lru->swap_lru = &adev->mman.bdev.glob->swap_lru; ++ } ++ + adev->mman.initialized = true; + r = ttm_bo_init_mm(&adev->mman.bdev, TTM_PL_VRAM, + adev->mc.real_vram_size >> PAGE_SHIFT); +-- +2.7.4 + |