niansa-misc/linux
1
0
Fork 0
mirror of synced 2025-03-06 20:59:54 +01:00
Code Issues Projects Releases Packages Wiki Activity

drm/ttm: replace busy placement with flags v6

Browse source 
Instead of a list of separate busy placement add flags which indicate
that a placement should only be used when there is room or if we need to
evict.

v2: add missing TTM_PL_FLAG_IDLE for i915
v3: fix auto build test ERROR on drm-tip/drm-tip
v4: fix some typos pointed out by checkpatch
v5: cleanup some rebase problems with VMWGFX
v6: implement some missing VMWGFX functionality pointed out by Zack,
    rename the flags as suggested by Michel, rebase on drm-tip and
    adjust XE as well

Signed-off-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Somalapuram Amaranath <Amaranath.Somalapuram@amd.com>
Reviewed-by: Zack Rusin <zack.rusin@broadcom.com>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240112125158.2748-4-christian.koenig@amd.com
This commit is contained in:
Somalapuram Amaranath 2023-11-13 16:18:44 +05:30 • committed by Christian König
parent 28e5126718
commit a78a8da51b
18 changed files with 104 additions and 180 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

6
drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
View file

@ -220,9 +220,6 @@ void amdgpu_bo_placement_from_domain(struct amdgpu_bo *abo, u32 domain)
placement->num_placement = c; placement->num_placement = c;
placement->placement = places; placement->placement = places;
placement->num_busy_placement = c;
placement->busy_placement = places;
} }
/** /**
@ -1406,8 +1403,7 @@ vm_fault_t amdgpu_bo_fault_reserve_notify(struct ttm_buffer_object *bo)
AMDGPU_GEM_DOMAIN_GTT); AMDGPU_GEM_DOMAIN_GTT);
/* Avoid costly evictions; only set GTT as a busy placement */ /* Avoid costly evictions; only set GTT as a busy placement */
abo->placement.num_busy_placement = 1; abo->placements[0].flags |= TTM_PL_FLAG_DESIRED;
abo->placement.busy_placement = &abo->placements[1];
r = ttm_bo_validate(bo, &abo->placement, &ctx); r = ttm_bo_validate(bo, &abo->placement, &ctx);
if (unlikely(r == -EBUSY || r == -ERESTARTSYS)) if (unlikely(r == -EBUSY || r == -ERESTARTSYS))

11
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
View file

@ -102,23 +102,19 @@ static void amdgpu_evict_flags(struct ttm_buffer_object *bo,
/* Don't handle scatter gather BOs */ /* Don't handle scatter gather BOs */
if (bo->type == ttm_bo_type_sg) { if (bo->type == ttm_bo_type_sg) {
placement->num_placement = 0; placement->num_placement = 0;
placement->num_busy_placement = 0;
return; return;
} }
/* Object isn't an AMDGPU object so ignore */ /* Object isn't an AMDGPU object so ignore */
if (!amdgpu_bo_is_amdgpu_bo(bo)) { if (!amdgpu_bo_is_amdgpu_bo(bo)) {
placement->placement = &placements; placement->placement = &placements;
placement->busy_placement = &placements;
placement->num_placement = 1; placement->num_placement = 1;
placement->num_busy_placement = 1;
return; return;
} }
abo = ttm_to_amdgpu_bo(bo); abo = ttm_to_amdgpu_bo(bo);
if (abo->flags & AMDGPU_GEM_CREATE_DISCARDABLE) { if (abo->flags & AMDGPU_GEM_CREATE_DISCARDABLE) {
placement->num_placement = 0; placement->num_placement = 0;
placement->num_busy_placement = 0;
return; return;
} }
@ -128,13 +124,13 @@ static void amdgpu_evict_flags(struct ttm_buffer_object *bo,
case AMDGPU_PL_OA: case AMDGPU_PL_OA:
case AMDGPU_PL_DOORBELL: case AMDGPU_PL_DOORBELL:
placement->num_placement = 0; placement->num_placement = 0;
placement->num_busy_placement = 0;
return; return;
case TTM_PL_VRAM: case TTM_PL_VRAM:
if (!adev->mman.buffer_funcs_enabled) { if (!adev->mman.buffer_funcs_enabled) {
/* Move to system memory */ /* Move to system memory */
amdgpu_bo_placement_from_domain(abo, AMDGPU_GEM_DOMAIN_CPU); amdgpu_bo_placement_from_domain(abo, AMDGPU_GEM_DOMAIN_CPU);
} else if (!amdgpu_gmc_vram_full_visible(&adev->gmc) && } else if (!amdgpu_gmc_vram_full_visible(&adev->gmc) &&
!(abo->flags & AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED) && !(abo->flags & AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED) &&
amdgpu_bo_in_cpu_visible_vram(abo)) { amdgpu_bo_in_cpu_visible_vram(abo)) {
@ -149,8 +145,7 @@ static void amdgpu_evict_flags(struct ttm_buffer_object *bo,
AMDGPU_GEM_DOMAIN_CPU); AMDGPU_GEM_DOMAIN_CPU);
abo->placements[0].fpfn = adev->gmc.visible_vram_size >> PAGE_SHIFT; abo->placements[0].fpfn = adev->gmc.visible_vram_size >> PAGE_SHIFT;
abo->placements[0].lpfn = 0; abo->placements[0].lpfn = 0;
abo->placement.busy_placement = &abo->placements[1]; abo->placements[0].flags |= TTM_PL_FLAG_DESIRED;
abo->placement.num_busy_placement = 1;
} else { } else {
/* Move to GTT memory */ /* Move to GTT memory */
amdgpu_bo_placement_from_domain(abo, AMDGPU_GEM_DOMAIN_GTT | amdgpu_bo_placement_from_domain(abo, AMDGPU_GEM_DOMAIN_GTT |
@ -967,8 +962,6 @@ int amdgpu_ttm_alloc_gart(struct ttm_buffer_object *bo)
/* allocate GART space */ /* allocate GART space */
placement.num_placement = 1; placement.num_placement = 1;
placement.placement = &placements; placement.placement = &placements;
placement.num_busy_placement = 1;
placement.busy_placement = &placements;
placements.fpfn = 0; placements.fpfn = 0;
placements.lpfn = adev->gmc.gart_size >> PAGE_SHIFT; placements.lpfn = adev->gmc.gart_size >> PAGE_SHIFT;
placements.mem_type = TTM_PL_TT; placements.mem_type = TTM_PL_TT;

2
drivers/gpu/drm/drm_gem_vram_helper.c
View file

@ -147,7 +147,6 @@ static void drm_gem_vram_placement(struct drm_gem_vram_object *gbo,
invariant_flags = TTM_PL_FLAG_TOPDOWN; invariant_flags = TTM_PL_FLAG_TOPDOWN;
gbo->placement.placement = gbo->placements; gbo->placement.placement = gbo->placements;
gbo->placement.busy_placement = gbo->placements;
if (pl_flag & DRM_GEM_VRAM_PL_FLAG_VRAM) { if (pl_flag & DRM_GEM_VRAM_PL_FLAG_VRAM) {
gbo->placements[c].mem_type = TTM_PL_VRAM; gbo->placements[c].mem_type = TTM_PL_VRAM;
@ -160,7 +159,6 @@ static void drm_gem_vram_placement(struct drm_gem_vram_object *gbo,
} }
gbo->placement.num_placement = c; gbo->placement.num_placement = c;
gbo->placement.num_busy_placement = c;
for (i = 0; i < c; ++i) { for (i = 0; i < c; ++i) {
gbo->placements[i].fpfn = 0; gbo->placements[i].fpfn = 0;

37
drivers/gpu/drm/i915/gem/i915_gem_ttm.c
View file

@ -65,8 +65,6 @@ static const struct ttm_place sys_placement_flags = {
static struct ttm_placement i915_sys_placement = { static struct ttm_placement i915_sys_placement = {
.num_placement = 1, .num_placement = 1,
.placement = &sys_placement_flags, .placement = &sys_placement_flags,
.num_busy_placement = 1,
.busy_placement = &sys_placement_flags,
}; };
/** /**
@ -157,32 +155,28 @@ i915_ttm_place_from_region(const struct intel_memory_region *mr,
static void static void
i915_ttm_placement_from_obj(const struct drm_i915_gem_object *obj, i915_ttm_placement_from_obj(const struct drm_i915_gem_object *obj,
struct ttm_place *requested, struct ttm_place *places,
struct ttm_place *busy,
struct ttm_placement *placement) struct ttm_placement *placement)
{ {
unsigned int num_allowed = obj->mm.n_placements; unsigned int num_allowed = obj->mm.n_placements;
unsigned int flags = obj->flags; unsigned int flags = obj->flags;
unsigned int i; unsigned int i;
placement->num_placement = 1; places[0].flags |= TTM_PL_FLAG_DESIRED;
i915_ttm_place_from_region(num_allowed ? obj->mm.placements[0] : i915_ttm_place_from_region(num_allowed ? obj->mm.placements[0] :
obj->mm.region, requested, obj->bo_offset, obj->mm.region, &places[0], obj->bo_offset,
obj->base.size, flags); obj->base.size, flags);
/* Cache this on object? */ /* Cache this on object? */
placement->num_busy_placement = num_allowed; for (i = 0; i < num_allowed; ++i) {
for (i = 0; i < placement->num_busy_placement; ++i) i915_ttm_place_from_region(obj->mm.placements[i],
i915_ttm_place_from_region(obj->mm.placements[i], busy + i, &places[i + 1], obj->bo_offset,
obj->bo_offset, obj->base.size, flags); obj->base.size, flags);
places[i + 1].flags |= TTM_PL_FLAG_FALLBACK;
if (num_allowed == 0) {
*busy = *requested;
placement->num_busy_placement = 1;
} }
placement->placement = requested; placement->num_placement = num_allowed + 1;
placement->busy_placement = busy; placement->placement = places;
} }
static int i915_ttm_tt_shmem_populate(struct ttm_device *bdev, static int i915_ttm_tt_shmem_populate(struct ttm_device *bdev,
@ -789,7 +783,8 @@ static int __i915_ttm_get_pages(struct drm_i915_gem_object *obj,
int ret; int ret;
/* First try only the requested placement. No eviction. */ /* First try only the requested placement. No eviction. */
real_num_busy = fetch_and_zero(&placement->num_busy_placement); real_num_busy = placement->num_placement;
placement->num_placement = 1;
ret = ttm_bo_validate(bo, placement, &ctx); ret = ttm_bo_validate(bo, placement, &ctx);
if (ret) { if (ret) {
ret = i915_ttm_err_to_gem(ret); ret = i915_ttm_err_to_gem(ret);
@ -805,7 +800,7 @@ static int __i915_ttm_get_pages(struct drm_i915_gem_object *obj,
* If the initial attempt fails, allow all accepted placements, * If the initial attempt fails, allow all accepted placements,
* evicting if necessary. * evicting if necessary.
*/ */
placement->num_busy_placement = real_num_busy; placement->num_placement = real_num_busy;
ret = ttm_bo_validate(bo, placement, &ctx); ret = ttm_bo_validate(bo, placement, &ctx);
if (ret) if (ret)
return i915_ttm_err_to_gem(ret); return i915_ttm_err_to_gem(ret);
@ -839,7 +834,7 @@ static int __i915_ttm_get_pages(struct drm_i915_gem_object *obj,
static int i915_ttm_get_pages(struct drm_i915_gem_object *obj) static int i915_ttm_get_pages(struct drm_i915_gem_object *obj)
{ {
struct ttm_place requested, busy[I915_TTM_MAX_PLACEMENTS]; struct ttm_place places[I915_TTM_MAX_PLACEMENTS + 1];
struct ttm_placement placement; struct ttm_placement placement;
/* restricted by sg_alloc_table */ /* restricted by sg_alloc_table */
@ -849,7 +844,7 @@ static int i915_ttm_get_pages(struct drm_i915_gem_object *obj)
GEM_BUG_ON(obj->mm.n_placements > I915_TTM_MAX_PLACEMENTS); GEM_BUG_ON(obj->mm.n_placements > I915_TTM_MAX_PLACEMENTS);
/* Move to the requested placement. */ /* Move to the requested placement. */
i915_ttm_placement_from_obj(obj, &requested, busy, &placement); i915_ttm_placement_from_obj(obj, places, &placement);
return __i915_ttm_get_pages(obj, &placement); return __i915_ttm_get_pages(obj, &placement);
} }
@ -879,9 +874,7 @@ static int __i915_ttm_migrate(struct drm_i915_gem_object *obj,
i915_ttm_place_from_region(mr, &requested, obj->bo_offset, i915_ttm_place_from_region(mr, &requested, obj->bo_offset,
obj->base.size, flags); obj->base.size, flags);
placement.num_placement = 1; placement.num_placement = 1;
placement.num_busy_placement = 1;
placement.placement = &requested; placement.placement = &requested;
placement.busy_placement = &requested;
ret = __i915_ttm_get_pages(obj, &placement); ret = __i915_ttm_get_pages(obj, &placement);
if (ret) if (ret)

2
drivers/gpu/drm/loongson/lsdc_ttm.c
View file

@ -54,7 +54,6 @@ static void lsdc_bo_set_placement(struct lsdc_bo *lbo, u32 domain)
pflags |= TTM_PL_FLAG_TOPDOWN; pflags |= TTM_PL_FLAG_TOPDOWN;
lbo->placement.placement = lbo->placements; lbo->placement.placement = lbo->placements;
lbo->placement.busy_placement = lbo->placements;
if (domain & LSDC_GEM_DOMAIN_VRAM) { if (domain & LSDC_GEM_DOMAIN_VRAM) {
lbo->placements[c].mem_type = TTM_PL_VRAM; lbo->placements[c].mem_type = TTM_PL_VRAM;
@ -77,7 +76,6 @@ static void lsdc_bo_set_placement(struct lsdc_bo *lbo, u32 domain)
} }
lbo->placement.num_placement = c; lbo->placement.num_placement = c;
lbo->placement.num_busy_placement = c;
for (i = 0; i < c; ++i) { for (i = 0; i < c; ++i) {
lbo->placements[i].fpfn = 0; lbo->placements[i].fpfn = 0;

59
drivers/gpu/drm/nouveau/nouveau_bo.c
View file

@ -403,27 +403,6 @@ nouveau_bo_new(struct nouveau_cli *cli, u64 size, int align,
return 0; return 0;
} }
static void
set_placement_list(struct ttm_place *pl, unsigned *n, uint32_t domain)
{
*n = 0;
if (domain & NOUVEAU_GEM_DOMAIN_VRAM) {
pl[*n].mem_type = TTM_PL_VRAM;
pl[*n].flags = 0;
(*n)++;
}
if (domain & NOUVEAU_GEM_DOMAIN_GART) {
pl[*n].mem_type = TTM_PL_TT;
pl[*n].flags = 0;
(*n)++;
}
if (domain & NOUVEAU_GEM_DOMAIN_CPU) {
pl[*n].mem_type = TTM_PL_SYSTEM;
pl[(*n)++].flags = 0;
}
}
static void static void
set_placement_range(struct nouveau_bo *nvbo, uint32_t domain) set_placement_range(struct nouveau_bo *nvbo, uint32_t domain)
{ {
@ -451,10 +430,6 @@ set_placement_range(struct nouveau_bo *nvbo, uint32_t domain)
nvbo->placements[i].fpfn = fpfn; nvbo->placements[i].fpfn = fpfn;
nvbo->placements[i].lpfn = lpfn; nvbo->placements[i].lpfn = lpfn;
} }
for (i = 0; i < nvbo->placement.num_busy_placement; ++i) {
nvbo->busy_placements[i].fpfn = fpfn;
nvbo->busy_placements[i].lpfn = lpfn;
}
} }
} }
@ -462,15 +437,32 @@ void
nouveau_bo_placement_set(struct nouveau_bo *nvbo, uint32_t domain, nouveau_bo_placement_set(struct nouveau_bo *nvbo, uint32_t domain,
uint32_t busy) uint32_t busy)
{ {
struct ttm_placement *pl = &nvbo->placement; unsigned int *n = &nvbo->placement.num_placement;
struct ttm_place *pl = nvbo->placements;
pl->placement = nvbo->placements; domain |= busy;
set_placement_list(nvbo->placements, &pl->num_placement, domain);
pl->busy_placement = nvbo->busy_placements; *n = 0;
set_placement_list(nvbo->busy_placements, &pl->num_busy_placement, if (domain & NOUVEAU_GEM_DOMAIN_VRAM) {
domain | busy); pl[*n].mem_type = TTM_PL_VRAM;
pl[*n].flags = busy & NOUVEAU_GEM_DOMAIN_VRAM ?
TTM_PL_FLAG_FALLBACK : 0;
(*n)++;
}
if (domain & NOUVEAU_GEM_DOMAIN_GART) {
pl[*n].mem_type = TTM_PL_TT;
pl[*n].flags = busy & NOUVEAU_GEM_DOMAIN_GART ?
TTM_PL_FLAG_FALLBACK : 0;
(*n)++;
}
if (domain & NOUVEAU_GEM_DOMAIN_CPU) {
pl[*n].mem_type = TTM_PL_SYSTEM;
pl[*n].flags = busy & NOUVEAU_GEM_DOMAIN_CPU ?
TTM_PL_FLAG_FALLBACK : 0;
(*n)++;
}
nvbo->placement.placement = nvbo->placements;
set_placement_range(nvbo, domain); set_placement_range(nvbo, domain);
} }
@ -1313,11 +1305,6 @@ vm_fault_t nouveau_ttm_fault_reserve_notify(struct ttm_buffer_object *bo)
nvbo->placements[i].lpfn = mappable; nvbo->placements[i].lpfn = mappable;
} }
for (i = 0; i < nvbo->placement.num_busy_placement; ++i) {
nvbo->busy_placements[i].fpfn = 0;
nvbo->busy_placements[i].lpfn = mappable;
}
nouveau_bo_placement_set(nvbo, NOUVEAU_GEM_DOMAIN_VRAM, 0); nouveau_bo_placement_set(nvbo, NOUVEAU_GEM_DOMAIN_VRAM, 0);
} }

1
drivers/gpu/drm/nouveau/nouveau_bo.h
View file

@ -15,7 +15,6 @@ struct nouveau_bo {
struct ttm_placement placement; struct ttm_placement placement;
u32 valid_domains; u32 valid_domains;
struct ttm_place placements[3]; struct ttm_place placements[3];
struct ttm_place busy_placements[3];
bool force_coherent; bool force_coherent;
struct ttm_bo_kmap_obj kmap; struct ttm_bo_kmap_obj kmap;
struct list_head head; struct list_head head;

2
drivers/gpu/drm/qxl/qxl_object.c
View file

@ -66,7 +66,6 @@ void qxl_ttm_placement_from_domain(struct qxl_bo *qbo, u32 domain)
pflag |= TTM_PL_FLAG_TOPDOWN; pflag |= TTM_PL_FLAG_TOPDOWN;
qbo->placement.placement = qbo->placements; qbo->placement.placement = qbo->placements;
qbo->placement.busy_placement = qbo->placements;
if (domain == QXL_GEM_DOMAIN_VRAM) { if (domain == QXL_GEM_DOMAIN_VRAM) {
qbo->placements[c].mem_type = TTM_PL_VRAM; qbo->placements[c].mem_type = TTM_PL_VRAM;
qbo->placements[c++].flags = pflag; qbo->placements[c++].flags = pflag;
@ -86,7 +85,6 @@ void qxl_ttm_placement_from_domain(struct qxl_bo *qbo, u32 domain)
qbo->placements[c++].flags = 0; qbo->placements[c++].flags = 0;
} }
qbo->placement.num_placement = c; qbo->placement.num_placement = c;
qbo->placement.num_busy_placement = c;
for (i = 0; i < c; ++i) { for (i = 0; i < c; ++i) {
qbo->placements[i].fpfn = 0; qbo->placements[i].fpfn = 0;
qbo->placements[i].lpfn = 0; qbo->placements[i].lpfn = 0;

2
drivers/gpu/drm/qxl/qxl_ttm.c
View file

@ -60,9 +60,7 @@ static void qxl_evict_flags(struct ttm_buffer_object *bo,
if (!qxl_ttm_bo_is_qxl_bo(bo)) { if (!qxl_ttm_bo_is_qxl_bo(bo)) {
placement->placement = &placements; placement->placement = &placements;
placement->busy_placement = &placements;
placement->num_placement = 1; placement->num_placement = 1;
placement->num_busy_placement = 1;
return; return;
} }
qbo = to_qxl_bo(bo); qbo = to_qxl_bo(bo);

2
drivers/gpu/drm/radeon/radeon_object.c
View file

@ -78,7 +78,6 @@ void radeon_ttm_placement_from_domain(struct radeon_bo *rbo, u32 domain)
u32 c = 0, i; u32 c = 0, i;
rbo->placement.placement = rbo->placements; rbo->placement.placement = rbo->placements;
rbo->placement.busy_placement = rbo->placements;
if (domain & RADEON_GEM_DOMAIN_VRAM) { if (domain & RADEON_GEM_DOMAIN_VRAM) {
/* Try placing BOs which don't need CPU access outside of the /* Try placing BOs which don't need CPU access outside of the
* CPU accessible part of VRAM * CPU accessible part of VRAM
@ -114,7 +113,6 @@ void radeon_ttm_placement_from_domain(struct radeon_bo *rbo, u32 domain)
} }
rbo->placement.num_placement = c; rbo->placement.num_placement = c;
rbo->placement.num_busy_placement = c;
for (i = 0; i < c; ++i) { for (i = 0; i < c; ++i) {
if ((rbo->flags & RADEON_GEM_CPU_ACCESS) && if ((rbo->flags & RADEON_GEM_CPU_ACCESS) &&

8
drivers/gpu/drm/radeon/radeon_ttm.c
View file

@ -92,9 +92,7 @@ static void radeon_evict_flags(struct ttm_buffer_object *bo,
if (!radeon_ttm_bo_is_radeon_bo(bo)) { if (!radeon_ttm_bo_is_radeon_bo(bo)) {
placement->placement = &placements; placement->placement = &placements;
placement->busy_placement = &placements;
placement->num_placement = 1; placement->num_placement = 1;
placement->num_busy_placement = 1;
return; return;
} }
rbo = container_of(bo, struct radeon_bo, tbo); rbo = container_of(bo, struct radeon_bo, tbo);
@ -114,15 +112,11 @@ static void radeon_evict_flags(struct ttm_buffer_object *bo,
*/ */
radeon_ttm_placement_from_domain(rbo, RADEON_GEM_DOMAIN_VRAM | radeon_ttm_placement_from_domain(rbo, RADEON_GEM_DOMAIN_VRAM |
RADEON_GEM_DOMAIN_GTT); RADEON_GEM_DOMAIN_GTT);
rbo->placement.num_busy_placement = 0;
for (i = 0; i < rbo->placement.num_placement; i++) { for (i = 0; i < rbo->placement.num_placement; i++) {
if (rbo->placements[i].mem_type == TTM_PL_VRAM) { if (rbo->placements[i].mem_type == TTM_PL_VRAM) {
if (rbo->placements[i].fpfn < fpfn) if (rbo->placements[i].fpfn < fpfn)
rbo->placements[i].fpfn = fpfn; rbo->placements[i].fpfn = fpfn;
} else { rbo->placements[0].flags |= TTM_PL_FLAG_DESIRED;
rbo->placement.busy_placement =
&rbo->placements[i];
rbo->placement.num_busy_placement = 1;
} }
} }
} else } else

1
drivers/gpu/drm/radeon/radeon_uvd.c
View file

@ -324,7 +324,6 @@ void radeon_uvd_force_into_uvd_segment(struct radeon_bo *rbo,
rbo->placements[1].fpfn += (256 * 1024 * 1024) >> PAGE_SHIFT; rbo->placements[1].fpfn += (256 * 1024 * 1024) >> PAGE_SHIFT;
rbo->placements[1].lpfn += (256 * 1024 * 1024) >> PAGE_SHIFT; rbo->placements[1].lpfn += (256 * 1024 * 1024) >> PAGE_SHIFT;
rbo->placement.num_placement++; rbo->placement.num_placement++;
rbo->placement.num_busy_placement++;
} }
void radeon_uvd_free_handles(struct radeon_device *rdev, struct drm_file *filp) void radeon_uvd_free_handles(struct radeon_device *rdev, struct drm_file *filp)

21
drivers/gpu/drm/ttm/ttm_bo.c
View file

@ -410,8 +410,8 @@ static int ttm_bo_bounce_temp_buffer(struct ttm_buffer_object *bo,
struct ttm_resource *hop_mem; struct ttm_resource *hop_mem;
int ret; int ret;
hop_placement.num_placement = hop_placement.num_busy_placement = 1; hop_placement.num_placement = 1;
hop_placement.placement = hop_placement.busy_placement = hop; hop_placement.placement = hop;
/* find space in the bounce domain */ /* find space in the bounce domain */
ret = ttm_bo_mem_space(bo, &hop_placement, &hop_mem, ctx); ret = ttm_bo_mem_space(bo, &hop_placement, &hop_mem, ctx);
@ -440,10 +440,9 @@ static int ttm_bo_evict(struct ttm_buffer_object *bo,
dma_resv_assert_held(bo->base.resv); dma_resv_assert_held(bo->base.resv);
placement.num_placement = 0; placement.num_placement = 0;
placement.num_busy_placement = 0;
bdev->funcs->evict_flags(bo, &placement); bdev->funcs->evict_flags(bo, &placement);
if (!placement.num_placement && !placement.num_busy_placement) { if (!placement.num_placement) {
ret = ttm_bo_wait_ctx(bo, ctx); ret = ttm_bo_wait_ctx(bo, ctx);
if (ret) if (ret)
return ret; return ret;
@ -791,6 +790,9 @@ int ttm_bo_mem_space(struct ttm_buffer_object *bo,
const struct ttm_place *place = &placement->placement[i]; const struct ttm_place *place = &placement->placement[i];
struct ttm_resource_manager *man; struct ttm_resource_manager *man;
if (place->flags & TTM_PL_FLAG_FALLBACK)
continue;
man = ttm_manager_type(bdev, place->mem_type); man = ttm_manager_type(bdev, place->mem_type);
if (!man || !ttm_resource_manager_used(man)) if (!man || !ttm_resource_manager_used(man))
continue; continue;
@ -813,10 +815,13 @@ int ttm_bo_mem_space(struct ttm_buffer_object *bo,
return 0; return 0;
} }
for (i = 0; i < placement->num_busy_placement; ++i) { for (i = 0; i < placement->num_placement; ++i) {
const struct ttm_place *place = &placement->busy_placement[i]; const struct ttm_place *place = &placement->placement[i];
struct ttm_resource_manager *man; struct ttm_resource_manager *man;
if (place->flags & TTM_PL_FLAG_DESIRED)
continue;
man = ttm_manager_type(bdev, place->mem_type); man = ttm_manager_type(bdev, place->mem_type);
if (!man || !ttm_resource_manager_used(man)) if (!man || !ttm_resource_manager_used(man))
continue; continue;
@ -904,11 +909,11 @@ int ttm_bo_validate(struct ttm_buffer_object *bo,
/* /*
* Remove the backing store if no placement is given. * Remove the backing store if no placement is given.
*/ */
if (!placement->num_placement && !placement->num_busy_placement) if (!placement->num_placement)
return ttm_bo_pipeline_gutting(bo); return ttm_bo_pipeline_gutting(bo);
/* Check whether we need to move buffer. */ /* Check whether we need to move buffer. */
if (bo->resource && ttm_resource_compat(bo->resource, placement)) if (bo->resource && ttm_resource_compatible(bo->resource, placement))
return 0; return 0;
/* Moving of pinned BOs is forbidden */ /* Moving of pinned BOs is forbidden */

73
drivers/gpu/drm/ttm/ttm_resource.c
View file

@ -291,37 +291,15 @@ bool ttm_resource_intersects(struct ttm_device *bdev,
} }
/** /**
* ttm_resource_compatible - test for compatibility * ttm_resource_compatible - check if resource is compatible with placement
* *
* @bdev: TTM device structure * @res: the resource to check
* @res: The resource to test * @placement: the placement to check against
* @place: The placement to test
* @size: How many bytes the new allocation needs.
* *
* Test if @res compatible with @place and @size. * Returns true if the placement is compatible.
*
* Returns true if the res placement compatible with @place and @size.
*/ */
bool ttm_resource_compatible(struct ttm_device *bdev, bool ttm_resource_compatible(struct ttm_resource *res,
struct ttm_resource *res, struct ttm_placement *placement)
const struct ttm_place *place,
size_t size)
{
struct ttm_resource_manager *man;
if (!res || !place)
return false;
man = ttm_manager_type(bdev, res->mem_type);
if (!man->func->compatible)
return true;
return man->func->compatible(man, res, place, size);
}
static bool ttm_resource_places_compat(struct ttm_resource *res,
const struct ttm_place *places,
unsigned num_placement)
{ {
struct ttm_buffer_object *bo = res->bo; struct ttm_buffer_object *bo = res->bo;
struct ttm_device *bdev = bo->bdev; struct ttm_device *bdev = bo->bdev;
@ -330,44 +308,25 @@ static bool ttm_resource_places_compat(struct ttm_resource *res,
if (res->placement & TTM_PL_FLAG_TEMPORARY) if (res->placement & TTM_PL_FLAG_TEMPORARY)
return false; return false;
for (i = 0; i < num_placement; i++) { for (i = 0; i < placement->num_placement; i++) {
const struct ttm_place *heap = &places[i]; const struct ttm_place *place = &placement->placement[i];
struct ttm_resource_manager *man;
if (!ttm_resource_compatible(bdev, res, heap, bo->base.size)) if (res->mem_type != place->mem_type)
continue; continue;
if ((res->mem_type == heap->mem_type) && man = ttm_manager_type(bdev, res->mem_type);
(!(heap->flags & TTM_PL_FLAG_CONTIGUOUS) || if (man->func->compatible &&
!man->func->compatible(man, res, place, bo->base.size))
continue;
if ((!(place->flags & TTM_PL_FLAG_CONTIGUOUS) ||
(res->placement & TTM_PL_FLAG_CONTIGUOUS))) (res->placement & TTM_PL_FLAG_CONTIGUOUS)))
return true; return true;
} }
return false; return false;
} }
/**
* ttm_resource_compat - check if resource is compatible with placement
*
* @res: the resource to check
* @placement: the placement to check against
*
* Returns true if the placement is compatible.
*/
bool ttm_resource_compat(struct ttm_resource *res,
struct ttm_placement *placement)
{
if (ttm_resource_places_compat(res, placement->placement,
placement->num_placement))
return true;
if ((placement->busy_placement != placement->placement ||
placement->num_busy_placement > placement->num_placement) &&
ttm_resource_places_compat(res, placement->busy_placement,
placement->num_busy_placement))
return true;
return false;
}
void ttm_resource_set_bo(struct ttm_resource *res, void ttm_resource_set_bo(struct ttm_resource *res,
struct ttm_buffer_object *bo) struct ttm_buffer_object *bo)
{ {

35
drivers/gpu/drm/vmwgfx/vmwgfx_bo.c
View file

@ -742,9 +742,21 @@ void vmw_bo_move_notify(struct ttm_buffer_object *bo,
vmw_resource_unbind_list(vbo); vmw_resource_unbind_list(vbo);
} }
static u32 static u32 placement_flags(u32 domain, u32 desired, u32 fallback)
set_placement_list(struct ttm_place *pl, u32 domain)
{ {
if (desired & fallback & domain)
return 0;
if (desired & domain)
return TTM_PL_FLAG_DESIRED;
return TTM_PL_FLAG_FALLBACK;
}
static u32
set_placement_list(struct ttm_place *pl, u32 desired, u32 fallback)
{
u32 domain = desired | fallback;
u32 n = 0; u32 n = 0;
/* /*
@ -752,35 +764,40 @@ set_placement_list(struct ttm_place *pl, u32 domain)
*/ */
if (domain & VMW_BO_DOMAIN_MOB) { if (domain & VMW_BO_DOMAIN_MOB) {
pl[n].mem_type = VMW_PL_MOB; pl[n].mem_type = VMW_PL_MOB;
pl[n].flags = 0; pl[n].flags = placement_flags(VMW_BO_DOMAIN_MOB, desired,
fallback);
pl[n].fpfn = 0; pl[n].fpfn = 0;
pl[n].lpfn = 0; pl[n].lpfn = 0;
n++; n++;
} }
if (domain & VMW_BO_DOMAIN_GMR) { if (domain & VMW_BO_DOMAIN_GMR) {
pl[n].mem_type = VMW_PL_GMR; pl[n].mem_type = VMW_PL_GMR;
pl[n].flags = 0; pl[n].flags = placement_flags(VMW_BO_DOMAIN_GMR, desired,
fallback);
pl[n].fpfn = 0; pl[n].fpfn = 0;
pl[n].lpfn = 0; pl[n].lpfn = 0;
n++; n++;
} }
if (domain & VMW_BO_DOMAIN_VRAM) { if (domain & VMW_BO_DOMAIN_VRAM) {
pl[n].mem_type = TTM_PL_VRAM; pl[n].mem_type = TTM_PL_VRAM;
pl[n].flags = 0; pl[n].flags = placement_flags(VMW_BO_DOMAIN_VRAM, desired,
fallback);
pl[n].fpfn = 0; pl[n].fpfn = 0;
pl[n].lpfn = 0; pl[n].lpfn = 0;
n++; n++;
} }
if (domain & VMW_BO_DOMAIN_WAITABLE_SYS) { if (domain & VMW_BO_DOMAIN_WAITABLE_SYS) {
pl[n].mem_type = VMW_PL_SYSTEM; pl[n].mem_type = VMW_PL_SYSTEM;
pl[n].flags = 0; pl[n].flags = placement_flags(VMW_BO_DOMAIN_WAITABLE_SYS,
desired, fallback);
pl[n].fpfn = 0; pl[n].fpfn = 0;
pl[n].lpfn = 0; pl[n].lpfn = 0;
n++; n++;
} }
if (domain & VMW_BO_DOMAIN_SYS) { if (domain & VMW_BO_DOMAIN_SYS) {
pl[n].mem_type = TTM_PL_SYSTEM; pl[n].mem_type = TTM_PL_SYSTEM;
pl[n].flags = 0; pl[n].flags = placement_flags(VMW_BO_DOMAIN_SYS, desired,
fallback);
pl[n].fpfn = 0; pl[n].fpfn = 0;
pl[n].lpfn = 0; pl[n].lpfn = 0;
n++; n++;
@ -806,7 +823,7 @@ void vmw_bo_placement_set(struct vmw_bo *bo, u32 domain, u32 busy_domain)
u32 i; u32 i;
pl->placement = bo->places; pl->placement = bo->places;
pl->num_placement = set_placement_list(bo->places, domain); pl->num_placement = set_placement_list(bo->places, domain, busy_domain);
if (drm_debug_enabled(DRM_UT_DRIVER) && bo->tbo.resource) { if (drm_debug_enabled(DRM_UT_DRIVER) && bo->tbo.resource) {
for (i = 0; i < pl->num_placement; ++i) { for (i = 0; i < pl->num_placement; ++i) {
@ -821,8 +838,6 @@ void vmw_bo_placement_set(struct vmw_bo *bo, u32 domain, u32 busy_domain)
__func__, bo->tbo.resource->mem_type, domain); __func__, bo->tbo.resource->mem_type, domain);
} }
pl->busy_placement = bo->busy_places;
pl->num_busy_placement = set_placement_list(bo->busy_places, busy_domain);
} }
void vmw_bo_placement_set_default_accelerated(struct vmw_bo *bo) void vmw_bo_placement_set_default_accelerated(struct vmw_bo *bo)

4
drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c
View file

@ -46,15 +46,11 @@ static const struct ttm_place sys_placement_flags = {
struct ttm_placement vmw_vram_placement = { struct ttm_placement vmw_vram_placement = {
.num_placement = 1, .num_placement = 1,
.placement = &vram_placement_flags, .placement = &vram_placement_flags,
.num_busy_placement = 1,
.busy_placement = &vram_placement_flags
}; };
struct ttm_placement vmw_sys_placement = { struct ttm_placement vmw_sys_placement = {
.num_placement = 1, .num_placement = 1,
.placement = &sys_placement_flags, .placement = &sys_placement_flags,
.num_busy_placement = 1,
.busy_placement = &sys_placement_flags
}; };
const size_t vmw_tt_size = sizeof(struct vmw_ttm_tt); const size_t vmw_tt_size = sizeof(struct vmw_ttm_tt);

10
include/drm/ttm/ttm_placement.h
View file

@ -64,6 +64,12 @@
/* For multihop handling */ /* For multihop handling */
#define TTM_PL_FLAG_TEMPORARY (1 << 2) #define TTM_PL_FLAG_TEMPORARY (1 << 2)
/* Placement is never used during eviction */
#define TTM_PL_FLAG_DESIRED (1 << 3)
/* Placement is only used during eviction */
#define TTM_PL_FLAG_FALLBACK (1 << 4)
/** /**
* struct ttm_place * struct ttm_place
* *
@ -86,16 +92,12 @@ struct ttm_place {
* *
* @num_placement: number of preferred placements * @num_placement: number of preferred placements
* @placement: preferred placements * @placement: preferred placements
* @num_busy_placement: number of preferred placements when need to evict buffer
* @busy_placement: preferred placements when need to evict buffer
* *
* Structure indicating the placement you request for an object. * Structure indicating the placement you request for an object.
*/ */
struct ttm_placement { struct ttm_placement {
unsigned num_placement; unsigned num_placement;
const struct ttm_place *placement; const struct ttm_place *placement;
unsigned num_busy_placement;
const struct ttm_place *busy_placement;
}; };
#endif #endif

6
include/drm/ttm/ttm_resource.h
View file

@ -365,11 +365,7 @@ bool ttm_resource_intersects(struct ttm_device *bdev,
struct ttm_resource *res, struct ttm_resource *res,
const struct ttm_place *place, const struct ttm_place *place,
size_t size); size_t size);
bool ttm_resource_compatible(struct ttm_device *bdev, bool ttm_resource_compatible(struct ttm_resource *res,
struct ttm_resource *res,
const struct ttm_place *place,
size_t size);
bool ttm_resource_compat(struct ttm_resource *res,
struct ttm_placement *placement); struct ttm_placement *placement);
void ttm_resource_set_bo(struct ttm_resource *res, void ttm_resource_set_bo(struct ttm_resource *res,
struct ttm_buffer_object *bo); struct ttm_buffer_object *bo);