From: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
To: "Alex Deucher" <alexander.deucher@amd.com>,
"Christian König" <christian.koenig@amd.com>,
"David Airlie" <airlied@gmail.com>,
"Simona Vetter" <simona@ffwll.ch>
Cc: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>,
<amd-gfx@lists.freedesktop.org>,
<dri-devel@lists.freedesktop.org>, <linux-kernel@vger.kernel.org>
Subject: [PATCH v1 16/20] drm/amdgpu: give ttm entities access to all the sdma scheds
Date: Tue, 4 Nov 2025 09:35:31 +0100 [thread overview]
Message-ID: <20251104083605.13677-17-pierre-eric.pelloux-prayer@amd.com> (raw)
In-Reply-To: <20251104083605.13677-1-pierre-eric.pelloux-prayer@amd.com>
Signed-off-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
---
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 23 ++++++++++-------------
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h | 1 +
2 files changed, 11 insertions(+), 13 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index 47eaab9350ae..d88bdb2ac083 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -2187,8 +2187,8 @@ u32 amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable)
u32 num_clear_entities, num_move_entities;
u32 windows, w;
- num_clear_entities = adev->sdma.num_instances;
- num_move_entities = MIN(adev->sdma.num_instances, TTM_FENCES_MAX_SLOT_COUNT);
+ num_clear_entities = MIN(adev->mman.num_buffer_funcs_rings, TTM_FENCES_MAX_SLOT_COUNT);
+ num_move_entities = MIN(adev->mman.num_buffer_funcs_rings, TTM_FENCES_MAX_SLOT_COUNT);
man->pipelined_eviction.n_fences = num_move_entities;
windows = adev->gmc.is_app_apu ? 0 : (2 * num_move_entities + num_clear_entities);
@@ -2197,13 +2197,8 @@ u32 amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable)
return windows;
if (enable) {
- struct amdgpu_ring *ring;
- struct drm_gpu_scheduler *sched;
-
- ring = adev->mman.buffer_funcs_rings[0];
- sched = &ring->sched;
r = drm_sched_entity_init(&adev->mman.default_entity.base,
- DRM_SCHED_PRIORITY_KERNEL, &sched,
+ DRM_SCHED_PRIORITY_KERNEL, adev->mman.scheds,
1, NULL);
if (r) {
dev_err(adev->dev, "Failed setting up entity (%d)\n",
@@ -2215,8 +2210,8 @@ u32 amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable)
atomic_set(&adev->mman.next_move_entity, 0);
for (i = 0; i < num_move_entities; i++) {
r = drm_sched_entity_init(&adev->mman.move_entities[i].base,
- DRM_SCHED_PRIORITY_NORMAL, &sched,
- 1, NULL);
+ DRM_SCHED_PRIORITY_NORMAL, adev->mman.scheds,
+ adev->mman.num_buffer_funcs_rings, NULL);
if (r) {
dev_err(adev->dev,
"Failed setting up TTM BO move entities (%d)\n",
@@ -2238,8 +2233,8 @@ u32 amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable)
for (i = 0; i < num_clear_entities; i++) {
r = drm_sched_entity_init(&adev->mman.clear_entities[i].base,
- DRM_SCHED_PRIORITY_NORMAL, &sched,
- 1, NULL);
+ DRM_SCHED_PRIORITY_NORMAL, adev->mman.scheds,
+ adev->mman.num_buffer_funcs_rings, NULL);
if (r) {
for (j = 0; j < num_move_entities; j++)
drm_sched_entity_destroy(
@@ -2650,8 +2645,10 @@ void amdgpu_sdma_set_buffer_funcs_rings(struct amdgpu_device *adev)
struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
int i;
- for (i = 0; i < adev->sdma.num_instances; i++)
+ for (i = 0; i < adev->sdma.num_instances; i++) {
adev->mman.buffer_funcs_rings[i] = &adev->sdma.instance[i].ring;
+ adev->mman.scheds[i] = &adev->sdma.instance[i].ring.sched;
+ }
adev->mman.num_buffer_funcs_rings = hub->sdma_invalidation_workaround ?
1 : adev->sdma.num_instances;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
index d7fee371b814..c059a3d52b57 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
@@ -68,6 +68,7 @@ struct amdgpu_mman {
const struct amdgpu_buffer_funcs *buffer_funcs;
struct amdgpu_ring *buffer_funcs_rings[AMDGPU_MAX_RINGS];
u32 num_buffer_funcs_rings;
+ struct drm_gpu_scheduler *scheds[AMDGPU_MAX_RINGS];
bool buffer_funcs_enabled;
struct mutex gtt_window_lock;
--
2.43.0
next prev parent reply other threads:[~2025-11-04 8:37 UTC|newest]
Thread overview: 40+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-11-04 8:35 [PATCH v1 00/20] drm/amdgpu: use all SDMA instances for TTM clears and moves Pierre-Eric Pelloux-Prayer
2025-11-04 8:35 ` [PATCH v1 01/20] drm/amdgpu: give each kernel job a unique id Pierre-Eric Pelloux-Prayer
2025-11-04 13:58 ` Christian König
2025-11-04 8:35 ` [PATCH v1 02/20] drm/ttm: rework pipelined eviction fence handling Pierre-Eric Pelloux-Prayer
2025-11-04 14:12 ` Christian König
2025-11-05 10:34 ` Pierre-Eric Pelloux-Prayer
2025-11-05 13:00 ` Christian König
2025-11-04 8:35 ` [PATCH v1 03/20] drm/amdgpu: remove direct_submit arg from amdgpu_copy_buffer Pierre-Eric Pelloux-Prayer
2025-11-04 14:13 ` Christian König
2025-11-04 8:35 ` [PATCH v1 04/20] drm/amdgpu: introduce amdgpu_ttm_entity Pierre-Eric Pelloux-Prayer
2025-11-04 15:13 ` Christian König
2025-11-04 8:35 ` [PATCH v1 05/20] drm/amdgpu: pass the entity to use to ttm functions Pierre-Eric Pelloux-Prayer
2025-11-04 8:35 ` [PATCH v1 06/20] drm/amdgpu: statically assign gart windows to ttm entities Pierre-Eric Pelloux-Prayer
2025-11-04 8:35 ` [PATCH v1 07/20] drm/amdgpu: allocate multiple clear entities Pierre-Eric Pelloux-Prayer
2025-11-04 8:35 ` [PATCH v1 08/20] drm/amdgpu: allocate multiple move entities Pierre-Eric Pelloux-Prayer
2025-11-04 8:35 ` [PATCH v1 09/20] drm/amdgpu: pass optional dependency to amdgpu_fill_buffer Pierre-Eric Pelloux-Prayer
2025-11-04 8:35 ` [PATCH v1 10/20] drm/amdgpu: prepare amdgpu_fill_buffer to use N entities Pierre-Eric Pelloux-Prayer
2025-11-04 8:35 ` [PATCH v1 11/20] drm/amdgpu: use multiple entities in amdgpu_fill_buffer Pierre-Eric Pelloux-Prayer
2025-11-04 16:40 ` Christian König
2025-11-05 10:39 ` Pierre-Eric Pelloux-Prayer
2025-11-05 13:03 ` Christian König
2025-11-12 10:39 ` Pierre-Eric Pelloux-Prayer
2025-11-04 8:35 ` [PATCH v1 12/20] drm/amdgpu: use TTM_FENCES_MAX_SLOT_COUNT Pierre-Eric Pelloux-Prayer
2025-11-05 8:36 ` Christian König
2025-11-04 8:35 ` [PATCH v1 13/20] drm/amdgpu: use multiple entities in amdgpu_move_blit Pierre-Eric Pelloux-Prayer
2025-11-04 8:35 ` [PATCH v1 14/20] drm/amdgpu: pass all the sdma rings to amdgpu_mman Pierre-Eric Pelloux-Prayer
2025-11-04 8:35 ` [PATCH v1 15/20] drm/amdgpu: introduce amdgpu_sdma_set_vm_pte_scheds Pierre-Eric Pelloux-Prayer
2025-11-05 8:41 ` Christian König
2025-11-04 8:35 ` Pierre-Eric Pelloux-Prayer [this message]
2025-11-05 8:45 ` [PATCH v1 16/20] drm/amdgpu: give ttm entities access to all the sdma scheds Christian König
2025-11-05 10:17 ` Pierre-Eric Pelloux-Prayer
2025-11-04 8:35 ` [PATCH v1 17/20] drm/amdgpu: get rid of amdgpu_ttm_clear_buffer Pierre-Eric Pelloux-Prayer
2025-11-05 8:46 ` Christian König
2025-11-05 9:30 ` Pierre-Eric Pelloux-Prayer
2025-11-05 10:07 ` Christian König
2025-11-04 8:35 ` [PATCH v1 18/20] drm/amdgpu: rename amdgpu_fill_buffer as amdgpu_clear_buffer Pierre-Eric Pelloux-Prayer
2025-11-05 8:47 ` Christian König
2025-11-04 8:35 ` [PATCH v1 19/20] drm/amdgpu: use larger gart window when possible Pierre-Eric Pelloux-Prayer
2025-11-04 8:35 ` [PATCH v1 20/20] drm/amdgpu: double AMDGPU_GTT_MAX_TRANSFER_SIZE Pierre-Eric Pelloux-Prayer
2025-11-04 13:56 ` Christian König
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20251104083605.13677-17-pierre-eric.pelloux-prayer@amd.com \
--to=pierre-eric.pelloux-prayer@amd.com \
--cc=airlied@gmail.com \
--cc=alexander.deucher@amd.com \
--cc=amd-gfx@lists.freedesktop.org \
--cc=christian.koenig@amd.com \
--cc=dri-devel@lists.freedesktop.org \
--cc=linux-kernel@vger.kernel.org \
--cc=simona@ffwll.ch \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®