Message ID | 20201008093154.2991-1-christian.koenig@amd.com (mailing list archive) |
---|---|
State | New, archived |
Headers | show |
Series | [1/4] drm/ttm: set the tt caching state at creation time | expand |
Ping? Anybody any more comments on this? Otherwise I'm going to push it to drm-misc-next by tomorrow or so. Thanks, Christian. Am 08.10.20 um 11:31 schrieb Christian König: > All drivers can determine the tt caching state at creation time, > no need to do this on the fly during every validation. > > Signed-off-by: Christian König <christian.koenig@amd.com> > Reviewed-by: Michael J. Ruhl <michael.j.ruhl@intel.com> > --- > drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c | 2 +- > drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 11 +++++-- > drivers/gpu/drm/drm_gem_vram_helper.c | 2 +- > drivers/gpu/drm/nouveau/nouveau_sgdma.c | 13 ++++++++- > drivers/gpu/drm/qxl/qxl_ttm.c | 2 +- > drivers/gpu/drm/radeon/radeon_ttm.c | 16 ++++++++-- > drivers/gpu/drm/ttm/ttm_agp_backend.c | 2 +- > drivers/gpu/drm/ttm/ttm_page_alloc.c | 26 ++++++++--------- > drivers/gpu/drm/ttm/ttm_page_alloc_dma.c | 20 ++++++------- > drivers/gpu/drm/ttm/ttm_tt.c | 33 +++++++++++---------- > drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c | 6 ++-- > include/drm/ttm/ttm_caching.h | 34 ++++++++++++++++++++++ > include/drm/ttm/ttm_tt.h | 16 ++++------ > 13 files changed, 123 insertions(+), 60 deletions(-) > create mode 100644 include/drm/ttm/ttm_caching.h > > diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c > index 213ef090bb0e..3c5ad69eff19 100644 > --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c > +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c > @@ -124,7 +124,7 @@ uint64_t amdgpu_gmc_agp_addr(struct ttm_buffer_object *bo) > struct amdgpu_device *adev = amdgpu_ttm_adev(bo->bdev); > struct ttm_dma_tt *ttm; > > - if (bo->num_pages != 1 || bo->ttm->caching_state == tt_cached) > + if (bo->num_pages != 1 || bo->ttm->caching == ttm_cached) > return AMDGPU_BO_INVALID_OFFSET; > > ttm = container_of(bo->ttm, struct ttm_dma_tt, ttm); > diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c > index 399961035ae6..7f41a47e7353 100644 > --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c > +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c > @@ -1292,7 +1292,9 @@ static void amdgpu_ttm_backend_destroy(struct ttm_bo_device *bdev, > static struct ttm_tt *amdgpu_ttm_tt_create(struct ttm_buffer_object *bo, > uint32_t page_flags) > { > + struct amdgpu_bo *abo = ttm_to_amdgpu_bo(bo); > struct amdgpu_ttm_tt *gtt; > + enum ttm_caching caching; > > gtt = kzalloc(sizeof(struct amdgpu_ttm_tt), GFP_KERNEL); > if (gtt == NULL) { > @@ -1300,8 +1302,13 @@ static struct ttm_tt *amdgpu_ttm_tt_create(struct ttm_buffer_object *bo, > } > gtt->gobj = &bo->base; > > + if (abo->flags & AMDGPU_GEM_CREATE_CPU_GTT_USWC) > + caching = ttm_write_combined; > + else > + caching = ttm_cached; > + > /* allocate space for the uninitialized page entries */ > - if (ttm_sg_tt_init(>t->ttm, bo, page_flags)) { > + if (ttm_sg_tt_init(>t->ttm, bo, page_flags, caching)) { > kfree(gtt); > return NULL; > } > @@ -1525,7 +1532,7 @@ uint64_t amdgpu_ttm_tt_pde_flags(struct ttm_tt *ttm, struct ttm_resource *mem) > if (mem && mem->mem_type == TTM_PL_TT) { > flags |= AMDGPU_PTE_SYSTEM; > > - if (ttm->caching_state == tt_cached) > + if (ttm->caching == ttm_cached) > flags |= AMDGPU_PTE_SNOOPED; > } > > diff --git a/drivers/gpu/drm/drm_gem_vram_helper.c b/drivers/gpu/drm/drm_gem_vram_helper.c > index 3213429f8444..ad58d0af5141 100644 > --- a/drivers/gpu/drm/drm_gem_vram_helper.c > +++ b/drivers/gpu/drm/drm_gem_vram_helper.c > @@ -918,7 +918,7 @@ static struct ttm_tt *bo_driver_ttm_tt_create(struct ttm_buffer_object *bo, > if (!tt) > return NULL; > > - ret = ttm_tt_init(tt, bo, page_flags); > + ret = ttm_tt_init(tt, bo, page_flags, ttm_cached); > if (ret < 0) > goto err_ttm_tt_init; > > diff --git a/drivers/gpu/drm/nouveau/nouveau_sgdma.c b/drivers/gpu/drm/nouveau/nouveau_sgdma.c > index 806d9ec310f5..cd6fdebae795 100644 > --- a/drivers/gpu/drm/nouveau/nouveau_sgdma.c > +++ b/drivers/gpu/drm/nouveau/nouveau_sgdma.c > @@ -5,6 +5,7 @@ > #include "nouveau_drv.h" > #include "nouveau_mem.h" > #include "nouveau_ttm.h" > +#include "nouveau_bo.h" > > struct nouveau_sgdma_be { > /* this has to be the first field so populate/unpopulated in > @@ -67,13 +68,23 @@ nouveau_sgdma_unbind(struct ttm_bo_device *bdev, struct ttm_tt *ttm) > struct ttm_tt * > nouveau_sgdma_create_ttm(struct ttm_buffer_object *bo, uint32_t page_flags) > { > + struct nouveau_drm *drm = nouveau_bdev(bo->bdev); > + struct nouveau_bo *nvbo = nouveau_bo(bo); > struct nouveau_sgdma_be *nvbe; > + enum ttm_caching caching; > + > + if (nvbo->force_coherent) > + caching = ttm_uncached; > + else if (drm->agp.bridge) > + caching = ttm_write_combined; > + else > + caching = ttm_cached; > > nvbe = kzalloc(sizeof(*nvbe), GFP_KERNEL); > if (!nvbe) > return NULL; > > - if (ttm_dma_tt_init(&nvbe->ttm, bo, page_flags)) { > + if (ttm_dma_tt_init(&nvbe->ttm, bo, page_flags, caching)) { > kfree(nvbe); > return NULL; > } > diff --git a/drivers/gpu/drm/qxl/qxl_ttm.c b/drivers/gpu/drm/qxl/qxl_ttm.c > index 669bceb58205..f50863493f64 100644 > --- a/drivers/gpu/drm/qxl/qxl_ttm.c > +++ b/drivers/gpu/drm/qxl/qxl_ttm.c > @@ -133,7 +133,7 @@ static struct ttm_tt *qxl_ttm_tt_create(struct ttm_buffer_object *bo, > ttm = kzalloc(sizeof(struct ttm_tt), GFP_KERNEL); > if (ttm == NULL) > return NULL; > - if (ttm_tt_init(ttm, bo, page_flags)) { > + if (ttm_tt_init(ttm, bo, page_flags, ttm_cached)) { > kfree(ttm); > return NULL; > } > diff --git a/drivers/gpu/drm/radeon/radeon_ttm.c b/drivers/gpu/drm/radeon/radeon_ttm.c > index 63e38b05a5bc..130a7cea35c3 100644 > --- a/drivers/gpu/drm/radeon/radeon_ttm.c > +++ b/drivers/gpu/drm/radeon/radeon_ttm.c > @@ -546,7 +546,7 @@ static int radeon_ttm_backend_bind(struct ttm_bo_device *bdev, > WARN(1, "nothing to bind %lu pages for mreg %p back %p!\n", > ttm->num_pages, bo_mem, ttm); > } > - if (ttm->caching_state == tt_cached) > + if (ttm->caching == ttm_cached) > flags |= RADEON_GART_PAGE_SNOOP; > r = radeon_gart_bind(rdev, gtt->offset, ttm->num_pages, > ttm->pages, gtt->ttm.dma_address, flags); > @@ -590,6 +590,10 @@ static struct ttm_tt *radeon_ttm_tt_create(struct ttm_buffer_object *bo, > { > struct radeon_device *rdev; > struct radeon_ttm_tt *gtt; > + enum ttm_caching caching; > + struct radeon_bo *rbo; > + > + rbo = container_of(bo, struct radeon_bo, tbo); > > rdev = radeon_get_rdev(bo->bdev); > #if IS_ENABLED(CONFIG_AGP) > @@ -603,7 +607,15 @@ static struct ttm_tt *radeon_ttm_tt_create(struct ttm_buffer_object *bo, > if (gtt == NULL) { > return NULL; > } > - if (ttm_dma_tt_init(>t->ttm, bo, page_flags)) { > + > + if (rbo->flags & RADEON_GEM_GTT_UC) > + caching = ttm_uncached; > + else if (rbo->flags & RADEON_GEM_GTT_WC) > + caching = ttm_write_combined; > + else > + caching = ttm_cached; > + > + if (ttm_dma_tt_init(>t->ttm, bo, page_flags, caching)) { > kfree(gtt); > return NULL; > } > diff --git a/drivers/gpu/drm/ttm/ttm_agp_backend.c b/drivers/gpu/drm/ttm/ttm_agp_backend.c > index a98fd795b752..a723062d37e7 100644 > --- a/drivers/gpu/drm/ttm/ttm_agp_backend.c > +++ b/drivers/gpu/drm/ttm/ttm_agp_backend.c > @@ -136,7 +136,7 @@ struct ttm_tt *ttm_agp_tt_create(struct ttm_buffer_object *bo, > agp_be->mem = NULL; > agp_be->bridge = bridge; > > - if (ttm_tt_init(&agp_be->ttm, bo, page_flags)) { > + if (ttm_tt_init(&agp_be->ttm, bo, page_flags, ttm_write_combined)) { > kfree(agp_be); > return NULL; > } > diff --git a/drivers/gpu/drm/ttm/ttm_page_alloc.c b/drivers/gpu/drm/ttm/ttm_page_alloc.c > index 111031cbb6df..c8f6790962b9 100644 > --- a/drivers/gpu/drm/ttm/ttm_page_alloc.c > +++ b/drivers/gpu/drm/ttm/ttm_page_alloc.c > @@ -220,14 +220,14 @@ static struct ttm_pool_manager *_manager; > /** > * Select the right pool or requested caching state and ttm flags. */ > static struct ttm_page_pool *ttm_get_pool(int flags, bool huge, > - enum ttm_caching_state cstate) > + enum ttm_caching cstate) > { > int pool_index; > > - if (cstate == tt_cached) > + if (cstate == ttm_cached) > return NULL; > > - if (cstate == tt_wc) > + if (cstate == ttm_write_combined) > pool_index = 0x0; > else > pool_index = 0x1; > @@ -441,17 +441,17 @@ static void ttm_pool_mm_shrink_fini(struct ttm_pool_manager *manager) > } > > static int ttm_set_pages_caching(struct page **pages, > - enum ttm_caching_state cstate, unsigned cpages) > + enum ttm_caching cstate, unsigned cpages) > { > int r = 0; > /* Set page caching */ > switch (cstate) { > - case tt_uncached: > + case ttm_uncached: > r = ttm_set_pages_array_uc(pages, cpages); > if (r) > pr_err("Failed to set %d pages to uc!\n", cpages); > break; > - case tt_wc: > + case ttm_write_combined: > r = ttm_set_pages_array_wc(pages, cpages); > if (r) > pr_err("Failed to set %d pages to wc!\n", cpages); > @@ -486,7 +486,7 @@ static void ttm_handle_caching_failure(struct page **failed_pages, > * pages returned in pages array. > */ > static int ttm_alloc_new_pages(struct list_head *pages, gfp_t gfp_flags, > - int ttm_flags, enum ttm_caching_state cstate, > + int ttm_flags, enum ttm_caching cstate, > unsigned count, unsigned order) > { > struct page **caching_array; > @@ -566,7 +566,7 @@ static int ttm_alloc_new_pages(struct list_head *pages, gfp_t gfp_flags, > * pages is small. > */ > static void ttm_page_pool_fill_locked(struct ttm_page_pool *pool, int ttm_flags, > - enum ttm_caching_state cstate, > + enum ttm_caching cstate, > unsigned count, unsigned long *irq_flags) > { > struct page *p; > @@ -626,7 +626,7 @@ static void ttm_page_pool_fill_locked(struct ttm_page_pool *pool, int ttm_flags, > static int ttm_page_pool_get_pages(struct ttm_page_pool *pool, > struct list_head *pages, > int ttm_flags, > - enum ttm_caching_state cstate, > + enum ttm_caching cstate, > unsigned count, unsigned order) > { > unsigned long irq_flags; > @@ -703,7 +703,7 @@ static int ttm_page_pool_get_pages(struct ttm_page_pool *pool, > > /* Put all pages in pages list to correct pool to wait for reuse */ > static void ttm_put_pages(struct page **pages, unsigned npages, int flags, > - enum ttm_caching_state cstate) > + enum ttm_caching cstate) > { > struct ttm_page_pool *pool = ttm_get_pool(flags, false, cstate); > #ifdef CONFIG_TRANSPARENT_HUGEPAGE > @@ -821,7 +821,7 @@ static void ttm_put_pages(struct page **pages, unsigned npages, int flags, > * cached pages. > */ > static int ttm_get_pages(struct page **pages, unsigned npages, int flags, > - enum ttm_caching_state cstate) > + enum ttm_caching cstate) > { > struct ttm_page_pool *pool = ttm_get_pool(flags, false, cstate); > #ifdef CONFIG_TRANSPARENT_HUGEPAGE > @@ -1040,7 +1040,7 @@ ttm_pool_unpopulate_helper(struct ttm_tt *ttm, unsigned mem_count_update) > > put_pages: > ttm_put_pages(ttm->pages, ttm->num_pages, ttm->page_flags, > - ttm->caching_state); > + ttm->caching); > ttm_tt_set_unpopulated(ttm); > } > > @@ -1057,7 +1057,7 @@ int ttm_pool_populate(struct ttm_tt *ttm, struct ttm_operation_ctx *ctx) > return -ENOMEM; > > ret = ttm_get_pages(ttm->pages, ttm->num_pages, ttm->page_flags, > - ttm->caching_state); > + ttm->caching); > if (unlikely(ret != 0)) { > ttm_pool_unpopulate_helper(ttm, 0); > return ret; > diff --git a/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c b/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c > index 1045a5c26ee3..6625b43f6256 100644 > --- a/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c > +++ b/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c > @@ -325,15 +325,15 @@ static struct dma_page *__ttm_dma_alloc_page(struct dma_pool *pool) > } > return d_page; > } > -static enum pool_type ttm_to_type(int flags, enum ttm_caching_state cstate) > +static enum pool_type ttm_to_type(int flags, enum ttm_caching cstate) > { > enum pool_type type = IS_UNDEFINED; > > if (flags & TTM_PAGE_FLAG_DMA32) > type |= IS_DMA32; > - if (cstate == tt_cached) > + if (cstate == ttm_cached) > type |= IS_CACHED; > - else if (cstate == tt_uncached) > + else if (cstate == ttm_uncached) > type |= IS_UC; > else > type |= IS_WC; > @@ -663,7 +663,7 @@ static struct dma_pool *ttm_dma_find_pool(struct device *dev, > * are pages that have changed their caching state already put them to the > * pool. > */ > -static void ttm_dma_handle_caching_state_failure(struct dma_pool *pool, > +static void ttm_dma_handle_caching_failure(struct dma_pool *pool, > struct list_head *d_pages, > struct page **failed_pages, > unsigned cpages) > @@ -734,7 +734,7 @@ static int ttm_dma_pool_alloc_new_pages(struct dma_pool *pool, > r = ttm_set_pages_caching(pool, caching_array, > cpages); > if (r) > - ttm_dma_handle_caching_state_failure( > + ttm_dma_handle_caching_failure( > pool, d_pages, caching_array, > cpages); > } > @@ -760,7 +760,7 @@ static int ttm_dma_pool_alloc_new_pages(struct dma_pool *pool, > r = ttm_set_pages_caching(pool, caching_array, > cpages); > if (r) { > - ttm_dma_handle_caching_state_failure( > + ttm_dma_handle_caching_failure( > pool, d_pages, caching_array, > cpages); > goto out; > @@ -773,7 +773,7 @@ static int ttm_dma_pool_alloc_new_pages(struct dma_pool *pool, > if (cpages) { > r = ttm_set_pages_caching(pool, caching_array, cpages); > if (r) > - ttm_dma_handle_caching_state_failure(pool, d_pages, > + ttm_dma_handle_caching_failure(pool, d_pages, > caching_array, cpages); > } > out: > @@ -904,7 +904,7 @@ int ttm_dma_populate(struct ttm_dma_tt *ttm_dma, struct device *dev, > INIT_LIST_HEAD(&ttm_dma->pages_list); > i = 0; > > - type = ttm_to_type(ttm->page_flags, ttm->caching_state); > + type = ttm_to_type(ttm->page_flags, ttm->caching); > > #ifdef CONFIG_TRANSPARENT_HUGEPAGE > if (ttm->page_flags & TTM_PAGE_FLAG_DMA32) > @@ -1000,7 +1000,7 @@ void ttm_dma_unpopulate(struct ttm_dma_tt *ttm_dma, struct device *dev) > unsigned count, i, npages = 0; > unsigned long irq_flags; > > - type = ttm_to_type(ttm->page_flags, ttm->caching_state); > + type = ttm_to_type(ttm->page_flags, ttm->caching); > > #ifdef CONFIG_TRANSPARENT_HUGEPAGE > pool = ttm_dma_find_pool(dev, type | IS_HUGE); > @@ -1032,7 +1032,7 @@ void ttm_dma_unpopulate(struct ttm_dma_tt *ttm_dma, struct device *dev) > return; > > is_cached = (ttm_dma_find_pool(pool->dev, > - ttm_to_type(ttm->page_flags, tt_cached)) == pool); > + ttm_to_type(ttm->page_flags, ttm_cached)) == pool); > > /* make sure pages array match list and count number of pages */ > count = 0; > diff --git a/drivers/gpu/drm/ttm/ttm_tt.c b/drivers/gpu/drm/ttm/ttm_tt.c > index 23e9604bc924..a465f51df027 100644 > --- a/drivers/gpu/drm/ttm/ttm_tt.c > +++ b/drivers/gpu/drm/ttm/ttm_tt.c > @@ -114,31 +114,30 @@ static int ttm_sg_tt_alloc_page_directory(struct ttm_dma_tt *ttm) > return 0; > } > > -static int ttm_tt_set_caching(struct ttm_tt *ttm, > - enum ttm_caching_state c_state) > +static int ttm_tt_set_caching(struct ttm_tt *ttm, enum ttm_caching caching) > { > - if (ttm->caching_state == c_state) > + if (ttm->caching == caching) > return 0; > > /* Can't change the caching state after TT is populated */ > if (WARN_ON_ONCE(ttm_tt_is_populated(ttm))) > return -EINVAL; > > - ttm->caching_state = c_state; > + ttm->caching = caching; > > return 0; > } > > int ttm_tt_set_placement_caching(struct ttm_tt *ttm, uint32_t placement) > { > - enum ttm_caching_state state; > + enum ttm_caching state; > > if (placement & TTM_PL_FLAG_WC) > - state = tt_wc; > + state = ttm_write_combined; > else if (placement & TTM_PL_FLAG_UNCACHED) > - state = tt_uncached; > + state = ttm_uncached; > else > - state = tt_cached; > + state = ttm_cached; > > return ttm_tt_set_caching(ttm, state); > } > @@ -162,20 +161,22 @@ void ttm_tt_destroy(struct ttm_bo_device *bdev, struct ttm_tt *ttm) > > static void ttm_tt_init_fields(struct ttm_tt *ttm, > struct ttm_buffer_object *bo, > - uint32_t page_flags) > + uint32_t page_flags, > + enum ttm_caching caching) > { > ttm->num_pages = bo->num_pages; > - ttm->caching_state = tt_cached; > + ttm->caching = ttm_cached; > ttm->page_flags = page_flags; > ttm_tt_set_unpopulated(ttm); > ttm->swap_storage = NULL; > ttm->sg = bo->sg; > + ttm->caching = caching; > } > > int ttm_tt_init(struct ttm_tt *ttm, struct ttm_buffer_object *bo, > - uint32_t page_flags) > + uint32_t page_flags, enum ttm_caching caching) > { > - ttm_tt_init_fields(ttm, bo, page_flags); > + ttm_tt_init_fields(ttm, bo, page_flags, caching); > > if (ttm_tt_alloc_page_directory(ttm)) { > pr_err("Failed allocating page table\n"); > @@ -193,11 +194,11 @@ void ttm_tt_fini(struct ttm_tt *ttm) > EXPORT_SYMBOL(ttm_tt_fini); > > int ttm_dma_tt_init(struct ttm_dma_tt *ttm_dma, struct ttm_buffer_object *bo, > - uint32_t page_flags) > + uint32_t page_flags, enum ttm_caching caching) > { > struct ttm_tt *ttm = &ttm_dma->ttm; > > - ttm_tt_init_fields(ttm, bo, page_flags); > + ttm_tt_init_fields(ttm, bo, page_flags, caching); > > INIT_LIST_HEAD(&ttm_dma->pages_list); > if (ttm_dma_tt_alloc_page_directory(ttm_dma)) { > @@ -209,12 +210,12 @@ int ttm_dma_tt_init(struct ttm_dma_tt *ttm_dma, struct ttm_buffer_object *bo, > EXPORT_SYMBOL(ttm_dma_tt_init); > > int ttm_sg_tt_init(struct ttm_dma_tt *ttm_dma, struct ttm_buffer_object *bo, > - uint32_t page_flags) > + uint32_t page_flags, enum ttm_caching caching) > { > struct ttm_tt *ttm = &ttm_dma->ttm; > int ret; > > - ttm_tt_init_fields(ttm, bo, page_flags); > + ttm_tt_init_fields(ttm, bo, page_flags, caching); > > INIT_LIST_HEAD(&ttm_dma->pages_list); > if (page_flags & TTM_PAGE_FLAG_SG) > diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c > index 7b5fd5288870..1fa7f9438ec4 100644 > --- a/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c > +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c > @@ -647,9 +647,11 @@ static struct ttm_tt *vmw_ttm_tt_create(struct ttm_buffer_object *bo, > vmw_be->mob = NULL; > > if (vmw_be->dev_priv->map_mode == vmw_dma_alloc_coherent) > - ret = ttm_dma_tt_init(&vmw_be->dma_ttm, bo, page_flags); > + ret = ttm_dma_tt_init(&vmw_be->dma_ttm, bo, page_flags, > + ttm_cached); > else > - ret = ttm_tt_init(&vmw_be->dma_ttm.ttm, bo, page_flags); > + ret = ttm_tt_init(&vmw_be->dma_ttm.ttm, bo, page_flags, > + ttm_cached); > if (unlikely(ret != 0)) > goto out_no_init; > > diff --git a/include/drm/ttm/ttm_caching.h b/include/drm/ttm/ttm_caching.h > new file mode 100644 > index 000000000000..161624dcf6be > --- /dev/null > +++ b/include/drm/ttm/ttm_caching.h > @@ -0,0 +1,34 @@ > +/* > + * Copyright 2020 Advanced Micro Devices, Inc. > + * > + * Permission is hereby granted, free of charge, to any person obtaining a > + * copy of this software and associated documentation files (the "Software"), > + * to deal in the Software without restriction, including without limitation > + * the rights to use, copy, modify, merge, publish, distribute, sublicense, > + * and/or sell copies of the Software, and to permit persons to whom the > + * Software is furnished to do so, subject to the following conditions: > + * > + * The above copyright notice and this permission notice shall be included in > + * all copies or substantial portions of the Software. > + * > + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR > + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, > + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL > + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR > + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, > + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR > + * OTHER DEALINGS IN THE SOFTWARE. > + * > + * Authors: Christian König > + */ > + > +#ifndef _TTM_CACHING_H_ > +#define _TTM_CACHING_H_ > + > +enum ttm_caching { > + ttm_uncached, > + ttm_write_combined, > + ttm_cached > +}; > + > +#endif > diff --git a/include/drm/ttm/ttm_tt.h b/include/drm/ttm/ttm_tt.h > index 5d1835d44084..c39c722d5184 100644 > --- a/include/drm/ttm/ttm_tt.h > +++ b/include/drm/ttm/ttm_tt.h > @@ -28,6 +28,7 @@ > #define _TTM_TT_H_ > > #include <linux/types.h> > +#include <drm/ttm/ttm_caching.h> > > struct ttm_tt; > struct ttm_resource; > @@ -42,12 +43,6 @@ struct ttm_operation_ctx; > > #define TTM_PAGE_FLAG_PRIV_POPULATED (1 << 31) > > -enum ttm_caching_state { > - tt_uncached, > - tt_wc, > - tt_cached > -}; > - > /** > * struct ttm_tt > * > @@ -69,7 +64,7 @@ struct ttm_tt { > unsigned long num_pages; > struct sg_table *sg; /* for SG objects via dma-buf */ > struct file *swap_storage; > - enum ttm_caching_state caching_state; > + enum ttm_caching caching; > }; > > static inline bool ttm_tt_is_populated(struct ttm_tt *tt) > @@ -121,6 +116,7 @@ int ttm_tt_create(struct ttm_buffer_object *bo, bool zero_alloc); > * @ttm: The struct ttm_tt. > * @bo: The buffer object we create the ttm for. > * @page_flags: Page flags as identified by TTM_PAGE_FLAG_XX flags. > + * @caching: the desired caching state of the pages > * > * Create a struct ttm_tt to back data with system memory pages. > * No pages are actually allocated. > @@ -128,11 +124,11 @@ int ttm_tt_create(struct ttm_buffer_object *bo, bool zero_alloc); > * NULL: Out of memory. > */ > int ttm_tt_init(struct ttm_tt *ttm, struct ttm_buffer_object *bo, > - uint32_t page_flags); > + uint32_t page_flags, enum ttm_caching caching); > int ttm_dma_tt_init(struct ttm_dma_tt *ttm_dma, struct ttm_buffer_object *bo, > - uint32_t page_flags); > + uint32_t page_flags, enum ttm_caching caching); > int ttm_sg_tt_init(struct ttm_dma_tt *ttm_dma, struct ttm_buffer_object *bo, > - uint32_t page_flags); > + uint32_t page_flags, enum ttm_caching caching); > > /** > * ttm_tt_fini
On Mon, Oct 12, 2020 at 10:57:57AM +0200, Christian König wrote: > Ping? Anybody any more comments on this? > > Otherwise I'm going to push it to drm-misc-next by tomorrow or so. tbh the entire coherency/caching topic is imo a giantic mess in drivers/gpu (mostly because we're half-fighting dma-api all the time). But I don't have clear opinion where to go, hence *shrug*. -Daniel > > Thanks, > Christian. > > Am 08.10.20 um 11:31 schrieb Christian König: > > All drivers can determine the tt caching state at creation time, > > no need to do this on the fly during every validation. > > > > Signed-off-by: Christian König <christian.koenig@amd.com> > > Reviewed-by: Michael J. Ruhl <michael.j.ruhl@intel.com> > > --- > > drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c | 2 +- > > drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 11 +++++-- > > drivers/gpu/drm/drm_gem_vram_helper.c | 2 +- > > drivers/gpu/drm/nouveau/nouveau_sgdma.c | 13 ++++++++- > > drivers/gpu/drm/qxl/qxl_ttm.c | 2 +- > > drivers/gpu/drm/radeon/radeon_ttm.c | 16 ++++++++-- > > drivers/gpu/drm/ttm/ttm_agp_backend.c | 2 +- > > drivers/gpu/drm/ttm/ttm_page_alloc.c | 26 ++++++++--------- > > drivers/gpu/drm/ttm/ttm_page_alloc_dma.c | 20 ++++++------- > > drivers/gpu/drm/ttm/ttm_tt.c | 33 +++++++++++---------- > > drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c | 6 ++-- > > include/drm/ttm/ttm_caching.h | 34 ++++++++++++++++++++++ > > include/drm/ttm/ttm_tt.h | 16 ++++------ > > 13 files changed, 123 insertions(+), 60 deletions(-) > > create mode 100644 include/drm/ttm/ttm_caching.h > > > > diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c > > index 213ef090bb0e..3c5ad69eff19 100644 > > --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c > > +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c > > @@ -124,7 +124,7 @@ uint64_t amdgpu_gmc_agp_addr(struct ttm_buffer_object *bo) > > struct amdgpu_device *adev = amdgpu_ttm_adev(bo->bdev); > > struct ttm_dma_tt *ttm; > > - if (bo->num_pages != 1 || bo->ttm->caching_state == tt_cached) > > + if (bo->num_pages != 1 || bo->ttm->caching == ttm_cached) > > return AMDGPU_BO_INVALID_OFFSET; > > ttm = container_of(bo->ttm, struct ttm_dma_tt, ttm); > > diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c > > index 399961035ae6..7f41a47e7353 100644 > > --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c > > +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c > > @@ -1292,7 +1292,9 @@ static void amdgpu_ttm_backend_destroy(struct ttm_bo_device *bdev, > > static struct ttm_tt *amdgpu_ttm_tt_create(struct ttm_buffer_object *bo, > > uint32_t page_flags) > > { > > + struct amdgpu_bo *abo = ttm_to_amdgpu_bo(bo); > > struct amdgpu_ttm_tt *gtt; > > + enum ttm_caching caching; > > gtt = kzalloc(sizeof(struct amdgpu_ttm_tt), GFP_KERNEL); > > if (gtt == NULL) { > > @@ -1300,8 +1302,13 @@ static struct ttm_tt *amdgpu_ttm_tt_create(struct ttm_buffer_object *bo, > > } > > gtt->gobj = &bo->base; > > + if (abo->flags & AMDGPU_GEM_CREATE_CPU_GTT_USWC) > > + caching = ttm_write_combined; > > + else > > + caching = ttm_cached; > > + > > /* allocate space for the uninitialized page entries */ > > - if (ttm_sg_tt_init(>t->ttm, bo, page_flags)) { > > + if (ttm_sg_tt_init(>t->ttm, bo, page_flags, caching)) { > > kfree(gtt); > > return NULL; > > } > > @@ -1525,7 +1532,7 @@ uint64_t amdgpu_ttm_tt_pde_flags(struct ttm_tt *ttm, struct ttm_resource *mem) > > if (mem && mem->mem_type == TTM_PL_TT) { > > flags |= AMDGPU_PTE_SYSTEM; > > - if (ttm->caching_state == tt_cached) > > + if (ttm->caching == ttm_cached) > > flags |= AMDGPU_PTE_SNOOPED; > > } > > diff --git a/drivers/gpu/drm/drm_gem_vram_helper.c b/drivers/gpu/drm/drm_gem_vram_helper.c > > index 3213429f8444..ad58d0af5141 100644 > > --- a/drivers/gpu/drm/drm_gem_vram_helper.c > > +++ b/drivers/gpu/drm/drm_gem_vram_helper.c > > @@ -918,7 +918,7 @@ static struct ttm_tt *bo_driver_ttm_tt_create(struct ttm_buffer_object *bo, > > if (!tt) > > return NULL; > > - ret = ttm_tt_init(tt, bo, page_flags); > > + ret = ttm_tt_init(tt, bo, page_flags, ttm_cached); > > if (ret < 0) > > goto err_ttm_tt_init; > > diff --git a/drivers/gpu/drm/nouveau/nouveau_sgdma.c b/drivers/gpu/drm/nouveau/nouveau_sgdma.c > > index 806d9ec310f5..cd6fdebae795 100644 > > --- a/drivers/gpu/drm/nouveau/nouveau_sgdma.c > > +++ b/drivers/gpu/drm/nouveau/nouveau_sgdma.c > > @@ -5,6 +5,7 @@ > > #include "nouveau_drv.h" > > #include "nouveau_mem.h" > > #include "nouveau_ttm.h" > > +#include "nouveau_bo.h" > > struct nouveau_sgdma_be { > > /* this has to be the first field so populate/unpopulated in > > @@ -67,13 +68,23 @@ nouveau_sgdma_unbind(struct ttm_bo_device *bdev, struct ttm_tt *ttm) > > struct ttm_tt * > > nouveau_sgdma_create_ttm(struct ttm_buffer_object *bo, uint32_t page_flags) > > { > > + struct nouveau_drm *drm = nouveau_bdev(bo->bdev); > > + struct nouveau_bo *nvbo = nouveau_bo(bo); > > struct nouveau_sgdma_be *nvbe; > > + enum ttm_caching caching; > > + > > + if (nvbo->force_coherent) > > + caching = ttm_uncached; > > + else if (drm->agp.bridge) > > + caching = ttm_write_combined; > > + else > > + caching = ttm_cached; > > nvbe = kzalloc(sizeof(*nvbe), GFP_KERNEL); > > if (!nvbe) > > return NULL; > > - if (ttm_dma_tt_init(&nvbe->ttm, bo, page_flags)) { > > + if (ttm_dma_tt_init(&nvbe->ttm, bo, page_flags, caching)) { > > kfree(nvbe); > > return NULL; > > } > > diff --git a/drivers/gpu/drm/qxl/qxl_ttm.c b/drivers/gpu/drm/qxl/qxl_ttm.c > > index 669bceb58205..f50863493f64 100644 > > --- a/drivers/gpu/drm/qxl/qxl_ttm.c > > +++ b/drivers/gpu/drm/qxl/qxl_ttm.c > > @@ -133,7 +133,7 @@ static struct ttm_tt *qxl_ttm_tt_create(struct ttm_buffer_object *bo, > > ttm = kzalloc(sizeof(struct ttm_tt), GFP_KERNEL); > > if (ttm == NULL) > > return NULL; > > - if (ttm_tt_init(ttm, bo, page_flags)) { > > + if (ttm_tt_init(ttm, bo, page_flags, ttm_cached)) { > > kfree(ttm); > > return NULL; > > } > > diff --git a/drivers/gpu/drm/radeon/radeon_ttm.c b/drivers/gpu/drm/radeon/radeon_ttm.c > > index 63e38b05a5bc..130a7cea35c3 100644 > > --- a/drivers/gpu/drm/radeon/radeon_ttm.c > > +++ b/drivers/gpu/drm/radeon/radeon_ttm.c > > @@ -546,7 +546,7 @@ static int radeon_ttm_backend_bind(struct ttm_bo_device *bdev, > > WARN(1, "nothing to bind %lu pages for mreg %p back %p!\n", > > ttm->num_pages, bo_mem, ttm); > > } > > - if (ttm->caching_state == tt_cached) > > + if (ttm->caching == ttm_cached) > > flags |= RADEON_GART_PAGE_SNOOP; > > r = radeon_gart_bind(rdev, gtt->offset, ttm->num_pages, > > ttm->pages, gtt->ttm.dma_address, flags); > > @@ -590,6 +590,10 @@ static struct ttm_tt *radeon_ttm_tt_create(struct ttm_buffer_object *bo, > > { > > struct radeon_device *rdev; > > struct radeon_ttm_tt *gtt; > > + enum ttm_caching caching; > > + struct radeon_bo *rbo; > > + > > + rbo = container_of(bo, struct radeon_bo, tbo); > > rdev = radeon_get_rdev(bo->bdev); > > #if IS_ENABLED(CONFIG_AGP) > > @@ -603,7 +607,15 @@ static struct ttm_tt *radeon_ttm_tt_create(struct ttm_buffer_object *bo, > > if (gtt == NULL) { > > return NULL; > > } > > - if (ttm_dma_tt_init(>t->ttm, bo, page_flags)) { > > + > > + if (rbo->flags & RADEON_GEM_GTT_UC) > > + caching = ttm_uncached; > > + else if (rbo->flags & RADEON_GEM_GTT_WC) > > + caching = ttm_write_combined; > > + else > > + caching = ttm_cached; > > + > > + if (ttm_dma_tt_init(>t->ttm, bo, page_flags, caching)) { > > kfree(gtt); > > return NULL; > > } > > diff --git a/drivers/gpu/drm/ttm/ttm_agp_backend.c b/drivers/gpu/drm/ttm/ttm_agp_backend.c > > index a98fd795b752..a723062d37e7 100644 > > --- a/drivers/gpu/drm/ttm/ttm_agp_backend.c > > +++ b/drivers/gpu/drm/ttm/ttm_agp_backend.c > > @@ -136,7 +136,7 @@ struct ttm_tt *ttm_agp_tt_create(struct ttm_buffer_object *bo, > > agp_be->mem = NULL; > > agp_be->bridge = bridge; > > - if (ttm_tt_init(&agp_be->ttm, bo, page_flags)) { > > + if (ttm_tt_init(&agp_be->ttm, bo, page_flags, ttm_write_combined)) { > > kfree(agp_be); > > return NULL; > > } > > diff --git a/drivers/gpu/drm/ttm/ttm_page_alloc.c b/drivers/gpu/drm/ttm/ttm_page_alloc.c > > index 111031cbb6df..c8f6790962b9 100644 > > --- a/drivers/gpu/drm/ttm/ttm_page_alloc.c > > +++ b/drivers/gpu/drm/ttm/ttm_page_alloc.c > > @@ -220,14 +220,14 @@ static struct ttm_pool_manager *_manager; > > /** > > * Select the right pool or requested caching state and ttm flags. */ > > static struct ttm_page_pool *ttm_get_pool(int flags, bool huge, > > - enum ttm_caching_state cstate) > > + enum ttm_caching cstate) > > { > > int pool_index; > > - if (cstate == tt_cached) > > + if (cstate == ttm_cached) > > return NULL; > > - if (cstate == tt_wc) > > + if (cstate == ttm_write_combined) > > pool_index = 0x0; > > else > > pool_index = 0x1; > > @@ -441,17 +441,17 @@ static void ttm_pool_mm_shrink_fini(struct ttm_pool_manager *manager) > > } > > static int ttm_set_pages_caching(struct page **pages, > > - enum ttm_caching_state cstate, unsigned cpages) > > + enum ttm_caching cstate, unsigned cpages) > > { > > int r = 0; > > /* Set page caching */ > > switch (cstate) { > > - case tt_uncached: > > + case ttm_uncached: > > r = ttm_set_pages_array_uc(pages, cpages); > > if (r) > > pr_err("Failed to set %d pages to uc!\n", cpages); > > break; > > - case tt_wc: > > + case ttm_write_combined: > > r = ttm_set_pages_array_wc(pages, cpages); > > if (r) > > pr_err("Failed to set %d pages to wc!\n", cpages); > > @@ -486,7 +486,7 @@ static void ttm_handle_caching_failure(struct page **failed_pages, > > * pages returned in pages array. > > */ > > static int ttm_alloc_new_pages(struct list_head *pages, gfp_t gfp_flags, > > - int ttm_flags, enum ttm_caching_state cstate, > > + int ttm_flags, enum ttm_caching cstate, > > unsigned count, unsigned order) > > { > > struct page **caching_array; > > @@ -566,7 +566,7 @@ static int ttm_alloc_new_pages(struct list_head *pages, gfp_t gfp_flags, > > * pages is small. > > */ > > static void ttm_page_pool_fill_locked(struct ttm_page_pool *pool, int ttm_flags, > > - enum ttm_caching_state cstate, > > + enum ttm_caching cstate, > > unsigned count, unsigned long *irq_flags) > > { > > struct page *p; > > @@ -626,7 +626,7 @@ static void ttm_page_pool_fill_locked(struct ttm_page_pool *pool, int ttm_flags, > > static int ttm_page_pool_get_pages(struct ttm_page_pool *pool, > > struct list_head *pages, > > int ttm_flags, > > - enum ttm_caching_state cstate, > > + enum ttm_caching cstate, > > unsigned count, unsigned order) > > { > > unsigned long irq_flags; > > @@ -703,7 +703,7 @@ static int ttm_page_pool_get_pages(struct ttm_page_pool *pool, > > /* Put all pages in pages list to correct pool to wait for reuse */ > > static void ttm_put_pages(struct page **pages, unsigned npages, int flags, > > - enum ttm_caching_state cstate) > > + enum ttm_caching cstate) > > { > > struct ttm_page_pool *pool = ttm_get_pool(flags, false, cstate); > > #ifdef CONFIG_TRANSPARENT_HUGEPAGE > > @@ -821,7 +821,7 @@ static void ttm_put_pages(struct page **pages, unsigned npages, int flags, > > * cached pages. > > */ > > static int ttm_get_pages(struct page **pages, unsigned npages, int flags, > > - enum ttm_caching_state cstate) > > + enum ttm_caching cstate) > > { > > struct ttm_page_pool *pool = ttm_get_pool(flags, false, cstate); > > #ifdef CONFIG_TRANSPARENT_HUGEPAGE > > @@ -1040,7 +1040,7 @@ ttm_pool_unpopulate_helper(struct ttm_tt *ttm, unsigned mem_count_update) > > put_pages: > > ttm_put_pages(ttm->pages, ttm->num_pages, ttm->page_flags, > > - ttm->caching_state); > > + ttm->caching); > > ttm_tt_set_unpopulated(ttm); > > } > > @@ -1057,7 +1057,7 @@ int ttm_pool_populate(struct ttm_tt *ttm, struct ttm_operation_ctx *ctx) > > return -ENOMEM; > > ret = ttm_get_pages(ttm->pages, ttm->num_pages, ttm->page_flags, > > - ttm->caching_state); > > + ttm->caching); > > if (unlikely(ret != 0)) { > > ttm_pool_unpopulate_helper(ttm, 0); > > return ret; > > diff --git a/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c b/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c > > index 1045a5c26ee3..6625b43f6256 100644 > > --- a/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c > > +++ b/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c > > @@ -325,15 +325,15 @@ static struct dma_page *__ttm_dma_alloc_page(struct dma_pool *pool) > > } > > return d_page; > > } > > -static enum pool_type ttm_to_type(int flags, enum ttm_caching_state cstate) > > +static enum pool_type ttm_to_type(int flags, enum ttm_caching cstate) > > { > > enum pool_type type = IS_UNDEFINED; > > if (flags & TTM_PAGE_FLAG_DMA32) > > type |= IS_DMA32; > > - if (cstate == tt_cached) > > + if (cstate == ttm_cached) > > type |= IS_CACHED; > > - else if (cstate == tt_uncached) > > + else if (cstate == ttm_uncached) > > type |= IS_UC; > > else > > type |= IS_WC; > > @@ -663,7 +663,7 @@ static struct dma_pool *ttm_dma_find_pool(struct device *dev, > > * are pages that have changed their caching state already put them to the > > * pool. > > */ > > -static void ttm_dma_handle_caching_state_failure(struct dma_pool *pool, > > +static void ttm_dma_handle_caching_failure(struct dma_pool *pool, > > struct list_head *d_pages, > > struct page **failed_pages, > > unsigned cpages) > > @@ -734,7 +734,7 @@ static int ttm_dma_pool_alloc_new_pages(struct dma_pool *pool, > > r = ttm_set_pages_caching(pool, caching_array, > > cpages); > > if (r) > > - ttm_dma_handle_caching_state_failure( > > + ttm_dma_handle_caching_failure( > > pool, d_pages, caching_array, > > cpages); > > } > > @@ -760,7 +760,7 @@ static int ttm_dma_pool_alloc_new_pages(struct dma_pool *pool, > > r = ttm_set_pages_caching(pool, caching_array, > > cpages); > > if (r) { > > - ttm_dma_handle_caching_state_failure( > > + ttm_dma_handle_caching_failure( > > pool, d_pages, caching_array, > > cpages); > > goto out; > > @@ -773,7 +773,7 @@ static int ttm_dma_pool_alloc_new_pages(struct dma_pool *pool, > > if (cpages) { > > r = ttm_set_pages_caching(pool, caching_array, cpages); > > if (r) > > - ttm_dma_handle_caching_state_failure(pool, d_pages, > > + ttm_dma_handle_caching_failure(pool, d_pages, > > caching_array, cpages); > > } > > out: > > @@ -904,7 +904,7 @@ int ttm_dma_populate(struct ttm_dma_tt *ttm_dma, struct device *dev, > > INIT_LIST_HEAD(&ttm_dma->pages_list); > > i = 0; > > - type = ttm_to_type(ttm->page_flags, ttm->caching_state); > > + type = ttm_to_type(ttm->page_flags, ttm->caching); > > #ifdef CONFIG_TRANSPARENT_HUGEPAGE > > if (ttm->page_flags & TTM_PAGE_FLAG_DMA32) > > @@ -1000,7 +1000,7 @@ void ttm_dma_unpopulate(struct ttm_dma_tt *ttm_dma, struct device *dev) > > unsigned count, i, npages = 0; > > unsigned long irq_flags; > > - type = ttm_to_type(ttm->page_flags, ttm->caching_state); > > + type = ttm_to_type(ttm->page_flags, ttm->caching); > > #ifdef CONFIG_TRANSPARENT_HUGEPAGE > > pool = ttm_dma_find_pool(dev, type | IS_HUGE); > > @@ -1032,7 +1032,7 @@ void ttm_dma_unpopulate(struct ttm_dma_tt *ttm_dma, struct device *dev) > > return; > > is_cached = (ttm_dma_find_pool(pool->dev, > > - ttm_to_type(ttm->page_flags, tt_cached)) == pool); > > + ttm_to_type(ttm->page_flags, ttm_cached)) == pool); > > /* make sure pages array match list and count number of pages */ > > count = 0; > > diff --git a/drivers/gpu/drm/ttm/ttm_tt.c b/drivers/gpu/drm/ttm/ttm_tt.c > > index 23e9604bc924..a465f51df027 100644 > > --- a/drivers/gpu/drm/ttm/ttm_tt.c > > +++ b/drivers/gpu/drm/ttm/ttm_tt.c > > @@ -114,31 +114,30 @@ static int ttm_sg_tt_alloc_page_directory(struct ttm_dma_tt *ttm) > > return 0; > > } > > -static int ttm_tt_set_caching(struct ttm_tt *ttm, > > - enum ttm_caching_state c_state) > > +static int ttm_tt_set_caching(struct ttm_tt *ttm, enum ttm_caching caching) > > { > > - if (ttm->caching_state == c_state) > > + if (ttm->caching == caching) > > return 0; > > /* Can't change the caching state after TT is populated */ > > if (WARN_ON_ONCE(ttm_tt_is_populated(ttm))) > > return -EINVAL; > > - ttm->caching_state = c_state; > > + ttm->caching = caching; > > return 0; > > } > > int ttm_tt_set_placement_caching(struct ttm_tt *ttm, uint32_t placement) > > { > > - enum ttm_caching_state state; > > + enum ttm_caching state; > > if (placement & TTM_PL_FLAG_WC) > > - state = tt_wc; > > + state = ttm_write_combined; > > else if (placement & TTM_PL_FLAG_UNCACHED) > > - state = tt_uncached; > > + state = ttm_uncached; > > else > > - state = tt_cached; > > + state = ttm_cached; > > return ttm_tt_set_caching(ttm, state); > > } > > @@ -162,20 +161,22 @@ void ttm_tt_destroy(struct ttm_bo_device *bdev, struct ttm_tt *ttm) > > static void ttm_tt_init_fields(struct ttm_tt *ttm, > > struct ttm_buffer_object *bo, > > - uint32_t page_flags) > > + uint32_t page_flags, > > + enum ttm_caching caching) > > { > > ttm->num_pages = bo->num_pages; > > - ttm->caching_state = tt_cached; > > + ttm->caching = ttm_cached; > > ttm->page_flags = page_flags; > > ttm_tt_set_unpopulated(ttm); > > ttm->swap_storage = NULL; > > ttm->sg = bo->sg; > > + ttm->caching = caching; > > } > > int ttm_tt_init(struct ttm_tt *ttm, struct ttm_buffer_object *bo, > > - uint32_t page_flags) > > + uint32_t page_flags, enum ttm_caching caching) > > { > > - ttm_tt_init_fields(ttm, bo, page_flags); > > + ttm_tt_init_fields(ttm, bo, page_flags, caching); > > if (ttm_tt_alloc_page_directory(ttm)) { > > pr_err("Failed allocating page table\n"); > > @@ -193,11 +194,11 @@ void ttm_tt_fini(struct ttm_tt *ttm) > > EXPORT_SYMBOL(ttm_tt_fini); > > int ttm_dma_tt_init(struct ttm_dma_tt *ttm_dma, struct ttm_buffer_object *bo, > > - uint32_t page_flags) > > + uint32_t page_flags, enum ttm_caching caching) > > { > > struct ttm_tt *ttm = &ttm_dma->ttm; > > - ttm_tt_init_fields(ttm, bo, page_flags); > > + ttm_tt_init_fields(ttm, bo, page_flags, caching); > > INIT_LIST_HEAD(&ttm_dma->pages_list); > > if (ttm_dma_tt_alloc_page_directory(ttm_dma)) { > > @@ -209,12 +210,12 @@ int ttm_dma_tt_init(struct ttm_dma_tt *ttm_dma, struct ttm_buffer_object *bo, > > EXPORT_SYMBOL(ttm_dma_tt_init); > > int ttm_sg_tt_init(struct ttm_dma_tt *ttm_dma, struct ttm_buffer_object *bo, > > - uint32_t page_flags) > > + uint32_t page_flags, enum ttm_caching caching) > > { > > struct ttm_tt *ttm = &ttm_dma->ttm; > > int ret; > > - ttm_tt_init_fields(ttm, bo, page_flags); > > + ttm_tt_init_fields(ttm, bo, page_flags, caching); > > INIT_LIST_HEAD(&ttm_dma->pages_list); > > if (page_flags & TTM_PAGE_FLAG_SG) > > diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c > > index 7b5fd5288870..1fa7f9438ec4 100644 > > --- a/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c > > +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c > > @@ -647,9 +647,11 @@ static struct ttm_tt *vmw_ttm_tt_create(struct ttm_buffer_object *bo, > > vmw_be->mob = NULL; > > if (vmw_be->dev_priv->map_mode == vmw_dma_alloc_coherent) > > - ret = ttm_dma_tt_init(&vmw_be->dma_ttm, bo, page_flags); > > + ret = ttm_dma_tt_init(&vmw_be->dma_ttm, bo, page_flags, > > + ttm_cached); > > else > > - ret = ttm_tt_init(&vmw_be->dma_ttm.ttm, bo, page_flags); > > + ret = ttm_tt_init(&vmw_be->dma_ttm.ttm, bo, page_flags, > > + ttm_cached); > > if (unlikely(ret != 0)) > > goto out_no_init; > > diff --git a/include/drm/ttm/ttm_caching.h b/include/drm/ttm/ttm_caching.h > > new file mode 100644 > > index 000000000000..161624dcf6be > > --- /dev/null > > +++ b/include/drm/ttm/ttm_caching.h > > @@ -0,0 +1,34 @@ > > +/* > > + * Copyright 2020 Advanced Micro Devices, Inc. > > + * > > + * Permission is hereby granted, free of charge, to any person obtaining a > > + * copy of this software and associated documentation files (the "Software"), > > + * to deal in the Software without restriction, including without limitation > > + * the rights to use, copy, modify, merge, publish, distribute, sublicense, > > + * and/or sell copies of the Software, and to permit persons to whom the > > + * Software is furnished to do so, subject to the following conditions: > > + * > > + * The above copyright notice and this permission notice shall be included in > > + * all copies or substantial portions of the Software. > > + * > > + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR > > + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, > > + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL > > + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR > > + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, > > + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR > > + * OTHER DEALINGS IN THE SOFTWARE. > > + * > > + * Authors: Christian König > > + */ > > + > > +#ifndef _TTM_CACHING_H_ > > +#define _TTM_CACHING_H_ > > + > > +enum ttm_caching { > > + ttm_uncached, > > + ttm_write_combined, > > + ttm_cached > > +}; > > + > > +#endif > > diff --git a/include/drm/ttm/ttm_tt.h b/include/drm/ttm/ttm_tt.h > > index 5d1835d44084..c39c722d5184 100644 > > --- a/include/drm/ttm/ttm_tt.h > > +++ b/include/drm/ttm/ttm_tt.h > > @@ -28,6 +28,7 @@ > > #define _TTM_TT_H_ > > #include <linux/types.h> > > +#include <drm/ttm/ttm_caching.h> > > struct ttm_tt; > > struct ttm_resource; > > @@ -42,12 +43,6 @@ struct ttm_operation_ctx; > > #define TTM_PAGE_FLAG_PRIV_POPULATED (1 << 31) > > -enum ttm_caching_state { > > - tt_uncached, > > - tt_wc, > > - tt_cached > > -}; > > - > > /** > > * struct ttm_tt > > * > > @@ -69,7 +64,7 @@ struct ttm_tt { > > unsigned long num_pages; > > struct sg_table *sg; /* for SG objects via dma-buf */ > > struct file *swap_storage; > > - enum ttm_caching_state caching_state; > > + enum ttm_caching caching; > > }; > > static inline bool ttm_tt_is_populated(struct ttm_tt *tt) > > @@ -121,6 +116,7 @@ int ttm_tt_create(struct ttm_buffer_object *bo, bool zero_alloc); > > * @ttm: The struct ttm_tt. > > * @bo: The buffer object we create the ttm for. > > * @page_flags: Page flags as identified by TTM_PAGE_FLAG_XX flags. > > + * @caching: the desired caching state of the pages > > * > > * Create a struct ttm_tt to back data with system memory pages. > > * No pages are actually allocated. > > @@ -128,11 +124,11 @@ int ttm_tt_create(struct ttm_buffer_object *bo, bool zero_alloc); > > * NULL: Out of memory. > > */ > > int ttm_tt_init(struct ttm_tt *ttm, struct ttm_buffer_object *bo, > > - uint32_t page_flags); > > + uint32_t page_flags, enum ttm_caching caching); > > int ttm_dma_tt_init(struct ttm_dma_tt *ttm_dma, struct ttm_buffer_object *bo, > > - uint32_t page_flags); > > + uint32_t page_flags, enum ttm_caching caching); > > int ttm_sg_tt_init(struct ttm_dma_tt *ttm_dma, struct ttm_buffer_object *bo, > > - uint32_t page_flags); > > + uint32_t page_flags, enum ttm_caching caching); > > /** > > * ttm_tt_fini > > _______________________________________________ > dri-devel mailing list > dri-devel@lists.freedesktop.org > https://lists.freedesktop.org/mailman/listinfo/dri-devel
Am 12.10.20 um 16:14 schrieb Daniel Vetter: > On Mon, Oct 12, 2020 at 10:57:57AM +0200, Christian König wrote: >> Ping? Anybody any more comments on this? >> >> Otherwise I'm going to push it to drm-misc-next by tomorrow or so. > tbh the entire coherency/caching topic is imo a giantic mess in > drivers/gpu (mostly because we're half-fighting dma-api all the time). But > I don't have clear opinion where to go, hence *shrug*. Well exactly that's why I'm doing the first step here by removing the illusion that TTM can magically changing the caching of a BO :) Christian. > -Daniel > >> Thanks, >> Christian. >> >> Am 08.10.20 um 11:31 schrieb Christian König: >>> All drivers can determine the tt caching state at creation time, >>> no need to do this on the fly during every validation. >>> >>> Signed-off-by: Christian König <christian.koenig@amd.com> >>> Reviewed-by: Michael J. Ruhl <michael.j.ruhl@intel.com> >>> --- >>> drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c | 2 +- >>> drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 11 +++++-- >>> drivers/gpu/drm/drm_gem_vram_helper.c | 2 +- >>> drivers/gpu/drm/nouveau/nouveau_sgdma.c | 13 ++++++++- >>> drivers/gpu/drm/qxl/qxl_ttm.c | 2 +- >>> drivers/gpu/drm/radeon/radeon_ttm.c | 16 ++++++++-- >>> drivers/gpu/drm/ttm/ttm_agp_backend.c | 2 +- >>> drivers/gpu/drm/ttm/ttm_page_alloc.c | 26 ++++++++--------- >>> drivers/gpu/drm/ttm/ttm_page_alloc_dma.c | 20 ++++++------- >>> drivers/gpu/drm/ttm/ttm_tt.c | 33 +++++++++++---------- >>> drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c | 6 ++-- >>> include/drm/ttm/ttm_caching.h | 34 ++++++++++++++++++++++ >>> include/drm/ttm/ttm_tt.h | 16 ++++------ >>> 13 files changed, 123 insertions(+), 60 deletions(-) >>> create mode 100644 include/drm/ttm/ttm_caching.h >>> >>> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c >>> index 213ef090bb0e..3c5ad69eff19 100644 >>> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c >>> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c >>> @@ -124,7 +124,7 @@ uint64_t amdgpu_gmc_agp_addr(struct ttm_buffer_object *bo) >>> struct amdgpu_device *adev = amdgpu_ttm_adev(bo->bdev); >>> struct ttm_dma_tt *ttm; >>> - if (bo->num_pages != 1 || bo->ttm->caching_state == tt_cached) >>> + if (bo->num_pages != 1 || bo->ttm->caching == ttm_cached) >>> return AMDGPU_BO_INVALID_OFFSET; >>> ttm = container_of(bo->ttm, struct ttm_dma_tt, ttm); >>> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c >>> index 399961035ae6..7f41a47e7353 100644 >>> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c >>> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c >>> @@ -1292,7 +1292,9 @@ static void amdgpu_ttm_backend_destroy(struct ttm_bo_device *bdev, >>> static struct ttm_tt *amdgpu_ttm_tt_create(struct ttm_buffer_object *bo, >>> uint32_t page_flags) >>> { >>> + struct amdgpu_bo *abo = ttm_to_amdgpu_bo(bo); >>> struct amdgpu_ttm_tt *gtt; >>> + enum ttm_caching caching; >>> gtt = kzalloc(sizeof(struct amdgpu_ttm_tt), GFP_KERNEL); >>> if (gtt == NULL) { >>> @@ -1300,8 +1302,13 @@ static struct ttm_tt *amdgpu_ttm_tt_create(struct ttm_buffer_object *bo, >>> } >>> gtt->gobj = &bo->base; >>> + if (abo->flags & AMDGPU_GEM_CREATE_CPU_GTT_USWC) >>> + caching = ttm_write_combined; >>> + else >>> + caching = ttm_cached; >>> + >>> /* allocate space for the uninitialized page entries */ >>> - if (ttm_sg_tt_init(>t->ttm, bo, page_flags)) { >>> + if (ttm_sg_tt_init(>t->ttm, bo, page_flags, caching)) { >>> kfree(gtt); >>> return NULL; >>> } >>> @@ -1525,7 +1532,7 @@ uint64_t amdgpu_ttm_tt_pde_flags(struct ttm_tt *ttm, struct ttm_resource *mem) >>> if (mem && mem->mem_type == TTM_PL_TT) { >>> flags |= AMDGPU_PTE_SYSTEM; >>> - if (ttm->caching_state == tt_cached) >>> + if (ttm->caching == ttm_cached) >>> flags |= AMDGPU_PTE_SNOOPED; >>> } >>> diff --git a/drivers/gpu/drm/drm_gem_vram_helper.c b/drivers/gpu/drm/drm_gem_vram_helper.c >>> index 3213429f8444..ad58d0af5141 100644 >>> --- a/drivers/gpu/drm/drm_gem_vram_helper.c >>> +++ b/drivers/gpu/drm/drm_gem_vram_helper.c >>> @@ -918,7 +918,7 @@ static struct ttm_tt *bo_driver_ttm_tt_create(struct ttm_buffer_object *bo, >>> if (!tt) >>> return NULL; >>> - ret = ttm_tt_init(tt, bo, page_flags); >>> + ret = ttm_tt_init(tt, bo, page_flags, ttm_cached); >>> if (ret < 0) >>> goto err_ttm_tt_init; >>> diff --git a/drivers/gpu/drm/nouveau/nouveau_sgdma.c b/drivers/gpu/drm/nouveau/nouveau_sgdma.c >>> index 806d9ec310f5..cd6fdebae795 100644 >>> --- a/drivers/gpu/drm/nouveau/nouveau_sgdma.c >>> +++ b/drivers/gpu/drm/nouveau/nouveau_sgdma.c >>> @@ -5,6 +5,7 @@ >>> #include "nouveau_drv.h" >>> #include "nouveau_mem.h" >>> #include "nouveau_ttm.h" >>> +#include "nouveau_bo.h" >>> struct nouveau_sgdma_be { >>> /* this has to be the first field so populate/unpopulated in >>> @@ -67,13 +68,23 @@ nouveau_sgdma_unbind(struct ttm_bo_device *bdev, struct ttm_tt *ttm) >>> struct ttm_tt * >>> nouveau_sgdma_create_ttm(struct ttm_buffer_object *bo, uint32_t page_flags) >>> { >>> + struct nouveau_drm *drm = nouveau_bdev(bo->bdev); >>> + struct nouveau_bo *nvbo = nouveau_bo(bo); >>> struct nouveau_sgdma_be *nvbe; >>> + enum ttm_caching caching; >>> + >>> + if (nvbo->force_coherent) >>> + caching = ttm_uncached; >>> + else if (drm->agp.bridge) >>> + caching = ttm_write_combined; >>> + else >>> + caching = ttm_cached; >>> nvbe = kzalloc(sizeof(*nvbe), GFP_KERNEL); >>> if (!nvbe) >>> return NULL; >>> - if (ttm_dma_tt_init(&nvbe->ttm, bo, page_flags)) { >>> + if (ttm_dma_tt_init(&nvbe->ttm, bo, page_flags, caching)) { >>> kfree(nvbe); >>> return NULL; >>> } >>> diff --git a/drivers/gpu/drm/qxl/qxl_ttm.c b/drivers/gpu/drm/qxl/qxl_ttm.c >>> index 669bceb58205..f50863493f64 100644 >>> --- a/drivers/gpu/drm/qxl/qxl_ttm.c >>> +++ b/drivers/gpu/drm/qxl/qxl_ttm.c >>> @@ -133,7 +133,7 @@ static struct ttm_tt *qxl_ttm_tt_create(struct ttm_buffer_object *bo, >>> ttm = kzalloc(sizeof(struct ttm_tt), GFP_KERNEL); >>> if (ttm == NULL) >>> return NULL; >>> - if (ttm_tt_init(ttm, bo, page_flags)) { >>> + if (ttm_tt_init(ttm, bo, page_flags, ttm_cached)) { >>> kfree(ttm); >>> return NULL; >>> } >>> diff --git a/drivers/gpu/drm/radeon/radeon_ttm.c b/drivers/gpu/drm/radeon/radeon_ttm.c >>> index 63e38b05a5bc..130a7cea35c3 100644 >>> --- a/drivers/gpu/drm/radeon/radeon_ttm.c >>> +++ b/drivers/gpu/drm/radeon/radeon_ttm.c >>> @@ -546,7 +546,7 @@ static int radeon_ttm_backend_bind(struct ttm_bo_device *bdev, >>> WARN(1, "nothing to bind %lu pages for mreg %p back %p!\n", >>> ttm->num_pages, bo_mem, ttm); >>> } >>> - if (ttm->caching_state == tt_cached) >>> + if (ttm->caching == ttm_cached) >>> flags |= RADEON_GART_PAGE_SNOOP; >>> r = radeon_gart_bind(rdev, gtt->offset, ttm->num_pages, >>> ttm->pages, gtt->ttm.dma_address, flags); >>> @@ -590,6 +590,10 @@ static struct ttm_tt *radeon_ttm_tt_create(struct ttm_buffer_object *bo, >>> { >>> struct radeon_device *rdev; >>> struct radeon_ttm_tt *gtt; >>> + enum ttm_caching caching; >>> + struct radeon_bo *rbo; >>> + >>> + rbo = container_of(bo, struct radeon_bo, tbo); >>> rdev = radeon_get_rdev(bo->bdev); >>> #if IS_ENABLED(CONFIG_AGP) >>> @@ -603,7 +607,15 @@ static struct ttm_tt *radeon_ttm_tt_create(struct ttm_buffer_object *bo, >>> if (gtt == NULL) { >>> return NULL; >>> } >>> - if (ttm_dma_tt_init(>t->ttm, bo, page_flags)) { >>> + >>> + if (rbo->flags & RADEON_GEM_GTT_UC) >>> + caching = ttm_uncached; >>> + else if (rbo->flags & RADEON_GEM_GTT_WC) >>> + caching = ttm_write_combined; >>> + else >>> + caching = ttm_cached; >>> + >>> + if (ttm_dma_tt_init(>t->ttm, bo, page_flags, caching)) { >>> kfree(gtt); >>> return NULL; >>> } >>> diff --git a/drivers/gpu/drm/ttm/ttm_agp_backend.c b/drivers/gpu/drm/ttm/ttm_agp_backend.c >>> index a98fd795b752..a723062d37e7 100644 >>> --- a/drivers/gpu/drm/ttm/ttm_agp_backend.c >>> +++ b/drivers/gpu/drm/ttm/ttm_agp_backend.c >>> @@ -136,7 +136,7 @@ struct ttm_tt *ttm_agp_tt_create(struct ttm_buffer_object *bo, >>> agp_be->mem = NULL; >>> agp_be->bridge = bridge; >>> - if (ttm_tt_init(&agp_be->ttm, bo, page_flags)) { >>> + if (ttm_tt_init(&agp_be->ttm, bo, page_flags, ttm_write_combined)) { >>> kfree(agp_be); >>> return NULL; >>> } >>> diff --git a/drivers/gpu/drm/ttm/ttm_page_alloc.c b/drivers/gpu/drm/ttm/ttm_page_alloc.c >>> index 111031cbb6df..c8f6790962b9 100644 >>> --- a/drivers/gpu/drm/ttm/ttm_page_alloc.c >>> +++ b/drivers/gpu/drm/ttm/ttm_page_alloc.c >>> @@ -220,14 +220,14 @@ static struct ttm_pool_manager *_manager; >>> /** >>> * Select the right pool or requested caching state and ttm flags. */ >>> static struct ttm_page_pool *ttm_get_pool(int flags, bool huge, >>> - enum ttm_caching_state cstate) >>> + enum ttm_caching cstate) >>> { >>> int pool_index; >>> - if (cstate == tt_cached) >>> + if (cstate == ttm_cached) >>> return NULL; >>> - if (cstate == tt_wc) >>> + if (cstate == ttm_write_combined) >>> pool_index = 0x0; >>> else >>> pool_index = 0x1; >>> @@ -441,17 +441,17 @@ static void ttm_pool_mm_shrink_fini(struct ttm_pool_manager *manager) >>> } >>> static int ttm_set_pages_caching(struct page **pages, >>> - enum ttm_caching_state cstate, unsigned cpages) >>> + enum ttm_caching cstate, unsigned cpages) >>> { >>> int r = 0; >>> /* Set page caching */ >>> switch (cstate) { >>> - case tt_uncached: >>> + case ttm_uncached: >>> r = ttm_set_pages_array_uc(pages, cpages); >>> if (r) >>> pr_err("Failed to set %d pages to uc!\n", cpages); >>> break; >>> - case tt_wc: >>> + case ttm_write_combined: >>> r = ttm_set_pages_array_wc(pages, cpages); >>> if (r) >>> pr_err("Failed to set %d pages to wc!\n", cpages); >>> @@ -486,7 +486,7 @@ static void ttm_handle_caching_failure(struct page **failed_pages, >>> * pages returned in pages array. >>> */ >>> static int ttm_alloc_new_pages(struct list_head *pages, gfp_t gfp_flags, >>> - int ttm_flags, enum ttm_caching_state cstate, >>> + int ttm_flags, enum ttm_caching cstate, >>> unsigned count, unsigned order) >>> { >>> struct page **caching_array; >>> @@ -566,7 +566,7 @@ static int ttm_alloc_new_pages(struct list_head *pages, gfp_t gfp_flags, >>> * pages is small. >>> */ >>> static void ttm_page_pool_fill_locked(struct ttm_page_pool *pool, int ttm_flags, >>> - enum ttm_caching_state cstate, >>> + enum ttm_caching cstate, >>> unsigned count, unsigned long *irq_flags) >>> { >>> struct page *p; >>> @@ -626,7 +626,7 @@ static void ttm_page_pool_fill_locked(struct ttm_page_pool *pool, int ttm_flags, >>> static int ttm_page_pool_get_pages(struct ttm_page_pool *pool, >>> struct list_head *pages, >>> int ttm_flags, >>> - enum ttm_caching_state cstate, >>> + enum ttm_caching cstate, >>> unsigned count, unsigned order) >>> { >>> unsigned long irq_flags; >>> @@ -703,7 +703,7 @@ static int ttm_page_pool_get_pages(struct ttm_page_pool *pool, >>> /* Put all pages in pages list to correct pool to wait for reuse */ >>> static void ttm_put_pages(struct page **pages, unsigned npages, int flags, >>> - enum ttm_caching_state cstate) >>> + enum ttm_caching cstate) >>> { >>> struct ttm_page_pool *pool = ttm_get_pool(flags, false, cstate); >>> #ifdef CONFIG_TRANSPARENT_HUGEPAGE >>> @@ -821,7 +821,7 @@ static void ttm_put_pages(struct page **pages, unsigned npages, int flags, >>> * cached pages. >>> */ >>> static int ttm_get_pages(struct page **pages, unsigned npages, int flags, >>> - enum ttm_caching_state cstate) >>> + enum ttm_caching cstate) >>> { >>> struct ttm_page_pool *pool = ttm_get_pool(flags, false, cstate); >>> #ifdef CONFIG_TRANSPARENT_HUGEPAGE >>> @@ -1040,7 +1040,7 @@ ttm_pool_unpopulate_helper(struct ttm_tt *ttm, unsigned mem_count_update) >>> put_pages: >>> ttm_put_pages(ttm->pages, ttm->num_pages, ttm->page_flags, >>> - ttm->caching_state); >>> + ttm->caching); >>> ttm_tt_set_unpopulated(ttm); >>> } >>> @@ -1057,7 +1057,7 @@ int ttm_pool_populate(struct ttm_tt *ttm, struct ttm_operation_ctx *ctx) >>> return -ENOMEM; >>> ret = ttm_get_pages(ttm->pages, ttm->num_pages, ttm->page_flags, >>> - ttm->caching_state); >>> + ttm->caching); >>> if (unlikely(ret != 0)) { >>> ttm_pool_unpopulate_helper(ttm, 0); >>> return ret; >>> diff --git a/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c b/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c >>> index 1045a5c26ee3..6625b43f6256 100644 >>> --- a/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c >>> +++ b/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c >>> @@ -325,15 +325,15 @@ static struct dma_page *__ttm_dma_alloc_page(struct dma_pool *pool) >>> } >>> return d_page; >>> } >>> -static enum pool_type ttm_to_type(int flags, enum ttm_caching_state cstate) >>> +static enum pool_type ttm_to_type(int flags, enum ttm_caching cstate) >>> { >>> enum pool_type type = IS_UNDEFINED; >>> if (flags & TTM_PAGE_FLAG_DMA32) >>> type |= IS_DMA32; >>> - if (cstate == tt_cached) >>> + if (cstate == ttm_cached) >>> type |= IS_CACHED; >>> - else if (cstate == tt_uncached) >>> + else if (cstate == ttm_uncached) >>> type |= IS_UC; >>> else >>> type |= IS_WC; >>> @@ -663,7 +663,7 @@ static struct dma_pool *ttm_dma_find_pool(struct device *dev, >>> * are pages that have changed their caching state already put them to the >>> * pool. >>> */ >>> -static void ttm_dma_handle_caching_state_failure(struct dma_pool *pool, >>> +static void ttm_dma_handle_caching_failure(struct dma_pool *pool, >>> struct list_head *d_pages, >>> struct page **failed_pages, >>> unsigned cpages) >>> @@ -734,7 +734,7 @@ static int ttm_dma_pool_alloc_new_pages(struct dma_pool *pool, >>> r = ttm_set_pages_caching(pool, caching_array, >>> cpages); >>> if (r) >>> - ttm_dma_handle_caching_state_failure( >>> + ttm_dma_handle_caching_failure( >>> pool, d_pages, caching_array, >>> cpages); >>> } >>> @@ -760,7 +760,7 @@ static int ttm_dma_pool_alloc_new_pages(struct dma_pool *pool, >>> r = ttm_set_pages_caching(pool, caching_array, >>> cpages); >>> if (r) { >>> - ttm_dma_handle_caching_state_failure( >>> + ttm_dma_handle_caching_failure( >>> pool, d_pages, caching_array, >>> cpages); >>> goto out; >>> @@ -773,7 +773,7 @@ static int ttm_dma_pool_alloc_new_pages(struct dma_pool *pool, >>> if (cpages) { >>> r = ttm_set_pages_caching(pool, caching_array, cpages); >>> if (r) >>> - ttm_dma_handle_caching_state_failure(pool, d_pages, >>> + ttm_dma_handle_caching_failure(pool, d_pages, >>> caching_array, cpages); >>> } >>> out: >>> @@ -904,7 +904,7 @@ int ttm_dma_populate(struct ttm_dma_tt *ttm_dma, struct device *dev, >>> INIT_LIST_HEAD(&ttm_dma->pages_list); >>> i = 0; >>> - type = ttm_to_type(ttm->page_flags, ttm->caching_state); >>> + type = ttm_to_type(ttm->page_flags, ttm->caching); >>> #ifdef CONFIG_TRANSPARENT_HUGEPAGE >>> if (ttm->page_flags & TTM_PAGE_FLAG_DMA32) >>> @@ -1000,7 +1000,7 @@ void ttm_dma_unpopulate(struct ttm_dma_tt *ttm_dma, struct device *dev) >>> unsigned count, i, npages = 0; >>> unsigned long irq_flags; >>> - type = ttm_to_type(ttm->page_flags, ttm->caching_state); >>> + type = ttm_to_type(ttm->page_flags, ttm->caching); >>> #ifdef CONFIG_TRANSPARENT_HUGEPAGE >>> pool = ttm_dma_find_pool(dev, type | IS_HUGE); >>> @@ -1032,7 +1032,7 @@ void ttm_dma_unpopulate(struct ttm_dma_tt *ttm_dma, struct device *dev) >>> return; >>> is_cached = (ttm_dma_find_pool(pool->dev, >>> - ttm_to_type(ttm->page_flags, tt_cached)) == pool); >>> + ttm_to_type(ttm->page_flags, ttm_cached)) == pool); >>> /* make sure pages array match list and count number of pages */ >>> count = 0; >>> diff --git a/drivers/gpu/drm/ttm/ttm_tt.c b/drivers/gpu/drm/ttm/ttm_tt.c >>> index 23e9604bc924..a465f51df027 100644 >>> --- a/drivers/gpu/drm/ttm/ttm_tt.c >>> +++ b/drivers/gpu/drm/ttm/ttm_tt.c >>> @@ -114,31 +114,30 @@ static int ttm_sg_tt_alloc_page_directory(struct ttm_dma_tt *ttm) >>> return 0; >>> } >>> -static int ttm_tt_set_caching(struct ttm_tt *ttm, >>> - enum ttm_caching_state c_state) >>> +static int ttm_tt_set_caching(struct ttm_tt *ttm, enum ttm_caching caching) >>> { >>> - if (ttm->caching_state == c_state) >>> + if (ttm->caching == caching) >>> return 0; >>> /* Can't change the caching state after TT is populated */ >>> if (WARN_ON_ONCE(ttm_tt_is_populated(ttm))) >>> return -EINVAL; >>> - ttm->caching_state = c_state; >>> + ttm->caching = caching; >>> return 0; >>> } >>> int ttm_tt_set_placement_caching(struct ttm_tt *ttm, uint32_t placement) >>> { >>> - enum ttm_caching_state state; >>> + enum ttm_caching state; >>> if (placement & TTM_PL_FLAG_WC) >>> - state = tt_wc; >>> + state = ttm_write_combined; >>> else if (placement & TTM_PL_FLAG_UNCACHED) >>> - state = tt_uncached; >>> + state = ttm_uncached; >>> else >>> - state = tt_cached; >>> + state = ttm_cached; >>> return ttm_tt_set_caching(ttm, state); >>> } >>> @@ -162,20 +161,22 @@ void ttm_tt_destroy(struct ttm_bo_device *bdev, struct ttm_tt *ttm) >>> static void ttm_tt_init_fields(struct ttm_tt *ttm, >>> struct ttm_buffer_object *bo, >>> - uint32_t page_flags) >>> + uint32_t page_flags, >>> + enum ttm_caching caching) >>> { >>> ttm->num_pages = bo->num_pages; >>> - ttm->caching_state = tt_cached; >>> + ttm->caching = ttm_cached; >>> ttm->page_flags = page_flags; >>> ttm_tt_set_unpopulated(ttm); >>> ttm->swap_storage = NULL; >>> ttm->sg = bo->sg; >>> + ttm->caching = caching; >>> } >>> int ttm_tt_init(struct ttm_tt *ttm, struct ttm_buffer_object *bo, >>> - uint32_t page_flags) >>> + uint32_t page_flags, enum ttm_caching caching) >>> { >>> - ttm_tt_init_fields(ttm, bo, page_flags); >>> + ttm_tt_init_fields(ttm, bo, page_flags, caching); >>> if (ttm_tt_alloc_page_directory(ttm)) { >>> pr_err("Failed allocating page table\n"); >>> @@ -193,11 +194,11 @@ void ttm_tt_fini(struct ttm_tt *ttm) >>> EXPORT_SYMBOL(ttm_tt_fini); >>> int ttm_dma_tt_init(struct ttm_dma_tt *ttm_dma, struct ttm_buffer_object *bo, >>> - uint32_t page_flags) >>> + uint32_t page_flags, enum ttm_caching caching) >>> { >>> struct ttm_tt *ttm = &ttm_dma->ttm; >>> - ttm_tt_init_fields(ttm, bo, page_flags); >>> + ttm_tt_init_fields(ttm, bo, page_flags, caching); >>> INIT_LIST_HEAD(&ttm_dma->pages_list); >>> if (ttm_dma_tt_alloc_page_directory(ttm_dma)) { >>> @@ -209,12 +210,12 @@ int ttm_dma_tt_init(struct ttm_dma_tt *ttm_dma, struct ttm_buffer_object *bo, >>> EXPORT_SYMBOL(ttm_dma_tt_init); >>> int ttm_sg_tt_init(struct ttm_dma_tt *ttm_dma, struct ttm_buffer_object *bo, >>> - uint32_t page_flags) >>> + uint32_t page_flags, enum ttm_caching caching) >>> { >>> struct ttm_tt *ttm = &ttm_dma->ttm; >>> int ret; >>> - ttm_tt_init_fields(ttm, bo, page_flags); >>> + ttm_tt_init_fields(ttm, bo, page_flags, caching); >>> INIT_LIST_HEAD(&ttm_dma->pages_list); >>> if (page_flags & TTM_PAGE_FLAG_SG) >>> diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c >>> index 7b5fd5288870..1fa7f9438ec4 100644 >>> --- a/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c >>> +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c >>> @@ -647,9 +647,11 @@ static struct ttm_tt *vmw_ttm_tt_create(struct ttm_buffer_object *bo, >>> vmw_be->mob = NULL; >>> if (vmw_be->dev_priv->map_mode == vmw_dma_alloc_coherent) >>> - ret = ttm_dma_tt_init(&vmw_be->dma_ttm, bo, page_flags); >>> + ret = ttm_dma_tt_init(&vmw_be->dma_ttm, bo, page_flags, >>> + ttm_cached); >>> else >>> - ret = ttm_tt_init(&vmw_be->dma_ttm.ttm, bo, page_flags); >>> + ret = ttm_tt_init(&vmw_be->dma_ttm.ttm, bo, page_flags, >>> + ttm_cached); >>> if (unlikely(ret != 0)) >>> goto out_no_init; >>> diff --git a/include/drm/ttm/ttm_caching.h b/include/drm/ttm/ttm_caching.h >>> new file mode 100644 >>> index 000000000000..161624dcf6be >>> --- /dev/null >>> +++ b/include/drm/ttm/ttm_caching.h >>> @@ -0,0 +1,34 @@ >>> +/* >>> + * Copyright 2020 Advanced Micro Devices, Inc. >>> + * >>> + * Permission is hereby granted, free of charge, to any person obtaining a >>> + * copy of this software and associated documentation files (the "Software"), >>> + * to deal in the Software without restriction, including without limitation >>> + * the rights to use, copy, modify, merge, publish, distribute, sublicense, >>> + * and/or sell copies of the Software, and to permit persons to whom the >>> + * Software is furnished to do so, subject to the following conditions: >>> + * >>> + * The above copyright notice and this permission notice shall be included in >>> + * all copies or substantial portions of the Software. >>> + * >>> + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR >>> + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, >>> + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL >>> + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR >>> + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, >>> + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR >>> + * OTHER DEALINGS IN THE SOFTWARE. >>> + * >>> + * Authors: Christian König >>> + */ >>> + >>> +#ifndef _TTM_CACHING_H_ >>> +#define _TTM_CACHING_H_ >>> + >>> +enum ttm_caching { >>> + ttm_uncached, >>> + ttm_write_combined, >>> + ttm_cached >>> +}; >>> + >>> +#endif >>> diff --git a/include/drm/ttm/ttm_tt.h b/include/drm/ttm/ttm_tt.h >>> index 5d1835d44084..c39c722d5184 100644 >>> --- a/include/drm/ttm/ttm_tt.h >>> +++ b/include/drm/ttm/ttm_tt.h >>> @@ -28,6 +28,7 @@ >>> #define _TTM_TT_H_ >>> #include <linux/types.h> >>> +#include <drm/ttm/ttm_caching.h> >>> struct ttm_tt; >>> struct ttm_resource; >>> @@ -42,12 +43,6 @@ struct ttm_operation_ctx; >>> #define TTM_PAGE_FLAG_PRIV_POPULATED (1 << 31) >>> -enum ttm_caching_state { >>> - tt_uncached, >>> - tt_wc, >>> - tt_cached >>> -}; >>> - >>> /** >>> * struct ttm_tt >>> * >>> @@ -69,7 +64,7 @@ struct ttm_tt { >>> unsigned long num_pages; >>> struct sg_table *sg; /* for SG objects via dma-buf */ >>> struct file *swap_storage; >>> - enum ttm_caching_state caching_state; >>> + enum ttm_caching caching; >>> }; >>> static inline bool ttm_tt_is_populated(struct ttm_tt *tt) >>> @@ -121,6 +116,7 @@ int ttm_tt_create(struct ttm_buffer_object *bo, bool zero_alloc); >>> * @ttm: The struct ttm_tt. >>> * @bo: The buffer object we create the ttm for. >>> * @page_flags: Page flags as identified by TTM_PAGE_FLAG_XX flags. >>> + * @caching: the desired caching state of the pages >>> * >>> * Create a struct ttm_tt to back data with system memory pages. >>> * No pages are actually allocated. >>> @@ -128,11 +124,11 @@ int ttm_tt_create(struct ttm_buffer_object *bo, bool zero_alloc); >>> * NULL: Out of memory. >>> */ >>> int ttm_tt_init(struct ttm_tt *ttm, struct ttm_buffer_object *bo, >>> - uint32_t page_flags); >>> + uint32_t page_flags, enum ttm_caching caching); >>> int ttm_dma_tt_init(struct ttm_dma_tt *ttm_dma, struct ttm_buffer_object *bo, >>> - uint32_t page_flags); >>> + uint32_t page_flags, enum ttm_caching caching); >>> int ttm_sg_tt_init(struct ttm_dma_tt *ttm_dma, struct ttm_buffer_object *bo, >>> - uint32_t page_flags); >>> + uint32_t page_flags, enum ttm_caching caching); >>> /** >>> * ttm_tt_fini >> _______________________________________________ >> dri-devel mailing list >> dri-devel@lists.freedesktop.org >> https://lists.freedesktop.org/mailman/listinfo/dri-devel
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c index 213ef090bb0e..3c5ad69eff19 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c @@ -124,7 +124,7 @@ uint64_t amdgpu_gmc_agp_addr(struct ttm_buffer_object *bo) struct amdgpu_device *adev = amdgpu_ttm_adev(bo->bdev); struct ttm_dma_tt *ttm; - if (bo->num_pages != 1 || bo->ttm->caching_state == tt_cached) + if (bo->num_pages != 1 || bo->ttm->caching == ttm_cached) return AMDGPU_BO_INVALID_OFFSET; ttm = container_of(bo->ttm, struct ttm_dma_tt, ttm); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c index 399961035ae6..7f41a47e7353 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c @@ -1292,7 +1292,9 @@ static void amdgpu_ttm_backend_destroy(struct ttm_bo_device *bdev, static struct ttm_tt *amdgpu_ttm_tt_create(struct ttm_buffer_object *bo, uint32_t page_flags) { + struct amdgpu_bo *abo = ttm_to_amdgpu_bo(bo); struct amdgpu_ttm_tt *gtt; + enum ttm_caching caching; gtt = kzalloc(sizeof(struct amdgpu_ttm_tt), GFP_KERNEL); if (gtt == NULL) { @@ -1300,8 +1302,13 @@ static struct ttm_tt *amdgpu_ttm_tt_create(struct ttm_buffer_object *bo, } gtt->gobj = &bo->base; + if (abo->flags & AMDGPU_GEM_CREATE_CPU_GTT_USWC) + caching = ttm_write_combined; + else + caching = ttm_cached; + /* allocate space for the uninitialized page entries */ - if (ttm_sg_tt_init(>t->ttm, bo, page_flags)) { + if (ttm_sg_tt_init(>t->ttm, bo, page_flags, caching)) { kfree(gtt); return NULL; } @@ -1525,7 +1532,7 @@ uint64_t amdgpu_ttm_tt_pde_flags(struct ttm_tt *ttm, struct ttm_resource *mem) if (mem && mem->mem_type == TTM_PL_TT) { flags |= AMDGPU_PTE_SYSTEM; - if (ttm->caching_state == tt_cached) + if (ttm->caching == ttm_cached) flags |= AMDGPU_PTE_SNOOPED; } diff --git a/drivers/gpu/drm/drm_gem_vram_helper.c b/drivers/gpu/drm/drm_gem_vram_helper.c index 3213429f8444..ad58d0af5141 100644 --- a/drivers/gpu/drm/drm_gem_vram_helper.c +++ b/drivers/gpu/drm/drm_gem_vram_helper.c @@ -918,7 +918,7 @@ static struct ttm_tt *bo_driver_ttm_tt_create(struct ttm_buffer_object *bo, if (!tt) return NULL; - ret = ttm_tt_init(tt, bo, page_flags); + ret = ttm_tt_init(tt, bo, page_flags, ttm_cached); if (ret < 0) goto err_ttm_tt_init; diff --git a/drivers/gpu/drm/nouveau/nouveau_sgdma.c b/drivers/gpu/drm/nouveau/nouveau_sgdma.c index 806d9ec310f5..cd6fdebae795 100644 --- a/drivers/gpu/drm/nouveau/nouveau_sgdma.c +++ b/drivers/gpu/drm/nouveau/nouveau_sgdma.c @@ -5,6 +5,7 @@ #include "nouveau_drv.h" #include "nouveau_mem.h" #include "nouveau_ttm.h" +#include "nouveau_bo.h" struct nouveau_sgdma_be { /* this has to be the first field so populate/unpopulated in @@ -67,13 +68,23 @@ nouveau_sgdma_unbind(struct ttm_bo_device *bdev, struct ttm_tt *ttm) struct ttm_tt * nouveau_sgdma_create_ttm(struct ttm_buffer_object *bo, uint32_t page_flags) { + struct nouveau_drm *drm = nouveau_bdev(bo->bdev); + struct nouveau_bo *nvbo = nouveau_bo(bo); struct nouveau_sgdma_be *nvbe; + enum ttm_caching caching; + + if (nvbo->force_coherent) + caching = ttm_uncached; + else if (drm->agp.bridge) + caching = ttm_write_combined; + else + caching = ttm_cached; nvbe = kzalloc(sizeof(*nvbe), GFP_KERNEL); if (!nvbe) return NULL; - if (ttm_dma_tt_init(&nvbe->ttm, bo, page_flags)) { + if (ttm_dma_tt_init(&nvbe->ttm, bo, page_flags, caching)) { kfree(nvbe); return NULL; } diff --git a/drivers/gpu/drm/qxl/qxl_ttm.c b/drivers/gpu/drm/qxl/qxl_ttm.c index 669bceb58205..f50863493f64 100644 --- a/drivers/gpu/drm/qxl/qxl_ttm.c +++ b/drivers/gpu/drm/qxl/qxl_ttm.c @@ -133,7 +133,7 @@ static struct ttm_tt *qxl_ttm_tt_create(struct ttm_buffer_object *bo, ttm = kzalloc(sizeof(struct ttm_tt), GFP_KERNEL); if (ttm == NULL) return NULL; - if (ttm_tt_init(ttm, bo, page_flags)) { + if (ttm_tt_init(ttm, bo, page_flags, ttm_cached)) { kfree(ttm); return NULL; } diff --git a/drivers/gpu/drm/radeon/radeon_ttm.c b/drivers/gpu/drm/radeon/radeon_ttm.c index 63e38b05a5bc..130a7cea35c3 100644 --- a/drivers/gpu/drm/radeon/radeon_ttm.c +++ b/drivers/gpu/drm/radeon/radeon_ttm.c @@ -546,7 +546,7 @@ static int radeon_ttm_backend_bind(struct ttm_bo_device *bdev, WARN(1, "nothing to bind %lu pages for mreg %p back %p!\n", ttm->num_pages, bo_mem, ttm); } - if (ttm->caching_state == tt_cached) + if (ttm->caching == ttm_cached) flags |= RADEON_GART_PAGE_SNOOP; r = radeon_gart_bind(rdev, gtt->offset, ttm->num_pages, ttm->pages, gtt->ttm.dma_address, flags); @@ -590,6 +590,10 @@ static struct ttm_tt *radeon_ttm_tt_create(struct ttm_buffer_object *bo, { struct radeon_device *rdev; struct radeon_ttm_tt *gtt; + enum ttm_caching caching; + struct radeon_bo *rbo; + + rbo = container_of(bo, struct radeon_bo, tbo); rdev = radeon_get_rdev(bo->bdev); #if IS_ENABLED(CONFIG_AGP) @@ -603,7 +607,15 @@ static struct ttm_tt *radeon_ttm_tt_create(struct ttm_buffer_object *bo, if (gtt == NULL) { return NULL; } - if (ttm_dma_tt_init(>t->ttm, bo, page_flags)) { + + if (rbo->flags & RADEON_GEM_GTT_UC) + caching = ttm_uncached; + else if (rbo->flags & RADEON_GEM_GTT_WC) + caching = ttm_write_combined; + else + caching = ttm_cached; + + if (ttm_dma_tt_init(>t->ttm, bo, page_flags, caching)) { kfree(gtt); return NULL; } diff --git a/drivers/gpu/drm/ttm/ttm_agp_backend.c b/drivers/gpu/drm/ttm/ttm_agp_backend.c index a98fd795b752..a723062d37e7 100644 --- a/drivers/gpu/drm/ttm/ttm_agp_backend.c +++ b/drivers/gpu/drm/ttm/ttm_agp_backend.c @@ -136,7 +136,7 @@ struct ttm_tt *ttm_agp_tt_create(struct ttm_buffer_object *bo, agp_be->mem = NULL; agp_be->bridge = bridge; - if (ttm_tt_init(&agp_be->ttm, bo, page_flags)) { + if (ttm_tt_init(&agp_be->ttm, bo, page_flags, ttm_write_combined)) { kfree(agp_be); return NULL; } diff --git a/drivers/gpu/drm/ttm/ttm_page_alloc.c b/drivers/gpu/drm/ttm/ttm_page_alloc.c index 111031cbb6df..c8f6790962b9 100644 --- a/drivers/gpu/drm/ttm/ttm_page_alloc.c +++ b/drivers/gpu/drm/ttm/ttm_page_alloc.c @@ -220,14 +220,14 @@ static struct ttm_pool_manager *_manager; /** * Select the right pool or requested caching state and ttm flags. */ static struct ttm_page_pool *ttm_get_pool(int flags, bool huge, - enum ttm_caching_state cstate) + enum ttm_caching cstate) { int pool_index; - if (cstate == tt_cached) + if (cstate == ttm_cached) return NULL; - if (cstate == tt_wc) + if (cstate == ttm_write_combined) pool_index = 0x0; else pool_index = 0x1; @@ -441,17 +441,17 @@ static void ttm_pool_mm_shrink_fini(struct ttm_pool_manager *manager) } static int ttm_set_pages_caching(struct page **pages, - enum ttm_caching_state cstate, unsigned cpages) + enum ttm_caching cstate, unsigned cpages) { int r = 0; /* Set page caching */ switch (cstate) { - case tt_uncached: + case ttm_uncached: r = ttm_set_pages_array_uc(pages, cpages); if (r) pr_err("Failed to set %d pages to uc!\n", cpages); break; - case tt_wc: + case ttm_write_combined: r = ttm_set_pages_array_wc(pages, cpages); if (r) pr_err("Failed to set %d pages to wc!\n", cpages); @@ -486,7 +486,7 @@ static void ttm_handle_caching_failure(struct page **failed_pages, * pages returned in pages array. */ static int ttm_alloc_new_pages(struct list_head *pages, gfp_t gfp_flags, - int ttm_flags, enum ttm_caching_state cstate, + int ttm_flags, enum ttm_caching cstate, unsigned count, unsigned order) { struct page **caching_array; @@ -566,7 +566,7 @@ static int ttm_alloc_new_pages(struct list_head *pages, gfp_t gfp_flags, * pages is small. */ static void ttm_page_pool_fill_locked(struct ttm_page_pool *pool, int ttm_flags, - enum ttm_caching_state cstate, + enum ttm_caching cstate, unsigned count, unsigned long *irq_flags) { struct page *p; @@ -626,7 +626,7 @@ static void ttm_page_pool_fill_locked(struct ttm_page_pool *pool, int ttm_flags, static int ttm_page_pool_get_pages(struct ttm_page_pool *pool, struct list_head *pages, int ttm_flags, - enum ttm_caching_state cstate, + enum ttm_caching cstate, unsigned count, unsigned order) { unsigned long irq_flags; @@ -703,7 +703,7 @@ static int ttm_page_pool_get_pages(struct ttm_page_pool *pool, /* Put all pages in pages list to correct pool to wait for reuse */ static void ttm_put_pages(struct page **pages, unsigned npages, int flags, - enum ttm_caching_state cstate) + enum ttm_caching cstate) { struct ttm_page_pool *pool = ttm_get_pool(flags, false, cstate); #ifdef CONFIG_TRANSPARENT_HUGEPAGE @@ -821,7 +821,7 @@ static void ttm_put_pages(struct page **pages, unsigned npages, int flags, * cached pages. */ static int ttm_get_pages(struct page **pages, unsigned npages, int flags, - enum ttm_caching_state cstate) + enum ttm_caching cstate) { struct ttm_page_pool *pool = ttm_get_pool(flags, false, cstate); #ifdef CONFIG_TRANSPARENT_HUGEPAGE @@ -1040,7 +1040,7 @@ ttm_pool_unpopulate_helper(struct ttm_tt *ttm, unsigned mem_count_update) put_pages: ttm_put_pages(ttm->pages, ttm->num_pages, ttm->page_flags, - ttm->caching_state); + ttm->caching); ttm_tt_set_unpopulated(ttm); } @@ -1057,7 +1057,7 @@ int ttm_pool_populate(struct ttm_tt *ttm, struct ttm_operation_ctx *ctx) return -ENOMEM; ret = ttm_get_pages(ttm->pages, ttm->num_pages, ttm->page_flags, - ttm->caching_state); + ttm->caching); if (unlikely(ret != 0)) { ttm_pool_unpopulate_helper(ttm, 0); return ret; diff --git a/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c b/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c index 1045a5c26ee3..6625b43f6256 100644 --- a/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c +++ b/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c @@ -325,15 +325,15 @@ static struct dma_page *__ttm_dma_alloc_page(struct dma_pool *pool) } return d_page; } -static enum pool_type ttm_to_type(int flags, enum ttm_caching_state cstate) +static enum pool_type ttm_to_type(int flags, enum ttm_caching cstate) { enum pool_type type = IS_UNDEFINED; if (flags & TTM_PAGE_FLAG_DMA32) type |= IS_DMA32; - if (cstate == tt_cached) + if (cstate == ttm_cached) type |= IS_CACHED; - else if (cstate == tt_uncached) + else if (cstate == ttm_uncached) type |= IS_UC; else type |= IS_WC; @@ -663,7 +663,7 @@ static struct dma_pool *ttm_dma_find_pool(struct device *dev, * are pages that have changed their caching state already put them to the * pool. */ -static void ttm_dma_handle_caching_state_failure(struct dma_pool *pool, +static void ttm_dma_handle_caching_failure(struct dma_pool *pool, struct list_head *d_pages, struct page **failed_pages, unsigned cpages) @@ -734,7 +734,7 @@ static int ttm_dma_pool_alloc_new_pages(struct dma_pool *pool, r = ttm_set_pages_caching(pool, caching_array, cpages); if (r) - ttm_dma_handle_caching_state_failure( + ttm_dma_handle_caching_failure( pool, d_pages, caching_array, cpages); } @@ -760,7 +760,7 @@ static int ttm_dma_pool_alloc_new_pages(struct dma_pool *pool, r = ttm_set_pages_caching(pool, caching_array, cpages); if (r) { - ttm_dma_handle_caching_state_failure( + ttm_dma_handle_caching_failure( pool, d_pages, caching_array, cpages); goto out; @@ -773,7 +773,7 @@ static int ttm_dma_pool_alloc_new_pages(struct dma_pool *pool, if (cpages) { r = ttm_set_pages_caching(pool, caching_array, cpages); if (r) - ttm_dma_handle_caching_state_failure(pool, d_pages, + ttm_dma_handle_caching_failure(pool, d_pages, caching_array, cpages); } out: @@ -904,7 +904,7 @@ int ttm_dma_populate(struct ttm_dma_tt *ttm_dma, struct device *dev, INIT_LIST_HEAD(&ttm_dma->pages_list); i = 0; - type = ttm_to_type(ttm->page_flags, ttm->caching_state); + type = ttm_to_type(ttm->page_flags, ttm->caching); #ifdef CONFIG_TRANSPARENT_HUGEPAGE if (ttm->page_flags & TTM_PAGE_FLAG_DMA32) @@ -1000,7 +1000,7 @@ void ttm_dma_unpopulate(struct ttm_dma_tt *ttm_dma, struct device *dev) unsigned count, i, npages = 0; unsigned long irq_flags; - type = ttm_to_type(ttm->page_flags, ttm->caching_state); + type = ttm_to_type(ttm->page_flags, ttm->caching); #ifdef CONFIG_TRANSPARENT_HUGEPAGE pool = ttm_dma_find_pool(dev, type | IS_HUGE); @@ -1032,7 +1032,7 @@ void ttm_dma_unpopulate(struct ttm_dma_tt *ttm_dma, struct device *dev) return; is_cached = (ttm_dma_find_pool(pool->dev, - ttm_to_type(ttm->page_flags, tt_cached)) == pool); + ttm_to_type(ttm->page_flags, ttm_cached)) == pool); /* make sure pages array match list and count number of pages */ count = 0; diff --git a/drivers/gpu/drm/ttm/ttm_tt.c b/drivers/gpu/drm/ttm/ttm_tt.c index 23e9604bc924..a465f51df027 100644 --- a/drivers/gpu/drm/ttm/ttm_tt.c +++ b/drivers/gpu/drm/ttm/ttm_tt.c @@ -114,31 +114,30 @@ static int ttm_sg_tt_alloc_page_directory(struct ttm_dma_tt *ttm) return 0; } -static int ttm_tt_set_caching(struct ttm_tt *ttm, - enum ttm_caching_state c_state) +static int ttm_tt_set_caching(struct ttm_tt *ttm, enum ttm_caching caching) { - if (ttm->caching_state == c_state) + if (ttm->caching == caching) return 0; /* Can't change the caching state after TT is populated */ if (WARN_ON_ONCE(ttm_tt_is_populated(ttm))) return -EINVAL; - ttm->caching_state = c_state; + ttm->caching = caching; return 0; } int ttm_tt_set_placement_caching(struct ttm_tt *ttm, uint32_t placement) { - enum ttm_caching_state state; + enum ttm_caching state; if (placement & TTM_PL_FLAG_WC) - state = tt_wc; + state = ttm_write_combined; else if (placement & TTM_PL_FLAG_UNCACHED) - state = tt_uncached; + state = ttm_uncached; else - state = tt_cached; + state = ttm_cached; return ttm_tt_set_caching(ttm, state); } @@ -162,20 +161,22 @@ void ttm_tt_destroy(struct ttm_bo_device *bdev, struct ttm_tt *ttm) static void ttm_tt_init_fields(struct ttm_tt *ttm, struct ttm_buffer_object *bo, - uint32_t page_flags) + uint32_t page_flags, + enum ttm_caching caching) { ttm->num_pages = bo->num_pages; - ttm->caching_state = tt_cached; + ttm->caching = ttm_cached; ttm->page_flags = page_flags; ttm_tt_set_unpopulated(ttm); ttm->swap_storage = NULL; ttm->sg = bo->sg; + ttm->caching = caching; } int ttm_tt_init(struct ttm_tt *ttm, struct ttm_buffer_object *bo, - uint32_t page_flags) + uint32_t page_flags, enum ttm_caching caching) { - ttm_tt_init_fields(ttm, bo, page_flags); + ttm_tt_init_fields(ttm, bo, page_flags, caching); if (ttm_tt_alloc_page_directory(ttm)) { pr_err("Failed allocating page table\n"); @@ -193,11 +194,11 @@ void ttm_tt_fini(struct ttm_tt *ttm) EXPORT_SYMBOL(ttm_tt_fini); int ttm_dma_tt_init(struct ttm_dma_tt *ttm_dma, struct ttm_buffer_object *bo, - uint32_t page_flags) + uint32_t page_flags, enum ttm_caching caching) { struct ttm_tt *ttm = &ttm_dma->ttm; - ttm_tt_init_fields(ttm, bo, page_flags); + ttm_tt_init_fields(ttm, bo, page_flags, caching); INIT_LIST_HEAD(&ttm_dma->pages_list); if (ttm_dma_tt_alloc_page_directory(ttm_dma)) { @@ -209,12 +210,12 @@ int ttm_dma_tt_init(struct ttm_dma_tt *ttm_dma, struct ttm_buffer_object *bo, EXPORT_SYMBOL(ttm_dma_tt_init); int ttm_sg_tt_init(struct ttm_dma_tt *ttm_dma, struct ttm_buffer_object *bo, - uint32_t page_flags) + uint32_t page_flags, enum ttm_caching caching) { struct ttm_tt *ttm = &ttm_dma->ttm; int ret; - ttm_tt_init_fields(ttm, bo, page_flags); + ttm_tt_init_fields(ttm, bo, page_flags, caching); INIT_LIST_HEAD(&ttm_dma->pages_list); if (page_flags & TTM_PAGE_FLAG_SG) diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c index 7b5fd5288870..1fa7f9438ec4 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c @@ -647,9 +647,11 @@ static struct ttm_tt *vmw_ttm_tt_create(struct ttm_buffer_object *bo, vmw_be->mob = NULL; if (vmw_be->dev_priv->map_mode == vmw_dma_alloc_coherent) - ret = ttm_dma_tt_init(&vmw_be->dma_ttm, bo, page_flags); + ret = ttm_dma_tt_init(&vmw_be->dma_ttm, bo, page_flags, + ttm_cached); else - ret = ttm_tt_init(&vmw_be->dma_ttm.ttm, bo, page_flags); + ret = ttm_tt_init(&vmw_be->dma_ttm.ttm, bo, page_flags, + ttm_cached); if (unlikely(ret != 0)) goto out_no_init; diff --git a/include/drm/ttm/ttm_caching.h b/include/drm/ttm/ttm_caching.h new file mode 100644 index 000000000000..161624dcf6be --- /dev/null +++ b/include/drm/ttm/ttm_caching.h @@ -0,0 +1,34 @@ +/* + * Copyright 2020 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: Christian König + */ + +#ifndef _TTM_CACHING_H_ +#define _TTM_CACHING_H_ + +enum ttm_caching { + ttm_uncached, + ttm_write_combined, + ttm_cached +}; + +#endif diff --git a/include/drm/ttm/ttm_tt.h b/include/drm/ttm/ttm_tt.h index 5d1835d44084..c39c722d5184 100644 --- a/include/drm/ttm/ttm_tt.h +++ b/include/drm/ttm/ttm_tt.h @@ -28,6 +28,7 @@ #define _TTM_TT_H_ #include <linux/types.h> +#include <drm/ttm/ttm_caching.h> struct ttm_tt; struct ttm_resource; @@ -42,12 +43,6 @@ struct ttm_operation_ctx; #define TTM_PAGE_FLAG_PRIV_POPULATED (1 << 31) -enum ttm_caching_state { - tt_uncached, - tt_wc, - tt_cached -}; - /** * struct ttm_tt * @@ -69,7 +64,7 @@ struct ttm_tt { unsigned long num_pages; struct sg_table *sg; /* for SG objects via dma-buf */ struct file *swap_storage; - enum ttm_caching_state caching_state; + enum ttm_caching caching; }; static inline bool ttm_tt_is_populated(struct ttm_tt *tt) @@ -121,6 +116,7 @@ int ttm_tt_create(struct ttm_buffer_object *bo, bool zero_alloc); * @ttm: The struct ttm_tt. * @bo: The buffer object we create the ttm for. * @page_flags: Page flags as identified by TTM_PAGE_FLAG_XX flags. + * @caching: the desired caching state of the pages * * Create a struct ttm_tt to back data with system memory pages. * No pages are actually allocated. @@ -128,11 +124,11 @@ int ttm_tt_create(struct ttm_buffer_object *bo, bool zero_alloc); * NULL: Out of memory. */ int ttm_tt_init(struct ttm_tt *ttm, struct ttm_buffer_object *bo, - uint32_t page_flags); + uint32_t page_flags, enum ttm_caching caching); int ttm_dma_tt_init(struct ttm_dma_tt *ttm_dma, struct ttm_buffer_object *bo, - uint32_t page_flags); + uint32_t page_flags, enum ttm_caching caching); int ttm_sg_tt_init(struct ttm_dma_tt *ttm_dma, struct ttm_buffer_object *bo, - uint32_t page_flags); + uint32_t page_flags, enum ttm_caching caching); /** * ttm_tt_fini