From: Yunxiang Li <Yunxiang.Li@amd.com>
To: <amd-gfx@lists.freedesktop.org>, <christian.koenig@amd.com>,
<tvrtko.ursulin@igalia.com>
Cc: <Alexander.Deucher@amd.com>, Yunxiang Li <Yunxiang.Li@amd.com>
Subject: [PATCH v7 0/4] rework bo mem stats tracking
Date: Sun, 10 Nov 2024 10:41:48 -0500 [thread overview]
Message-ID: <20241110154152.592-1-Yunxiang.Li@amd.com> (raw)
Right now every time the fdinfo is read, we go through the vm lists and
lock all the BOs to calcuate the statistics. This causes a lot of lock
contention when the VM is actively used. It gets worse if there is a lot
of shared BOs or if there's a lot of submissions. We have seen
submissions lock-up for seconds due to fdinfo for some workload.
Therefore, rework the implementation to track the BOs as they get moved
around.
The amd-only visible memory stat is removed to simplify implementation,
it's unclear how useful this stat is since kernel map/unmap BOs whenever
it wants to and on a modern system all of VRAM can be mapped if needed.
v5: rebase on top of the drm_print_memory_stats refactor
v6: split the drm changes into a seperate patch for drm-devel review,
fix handling of drm-total- vs drm-resident- and handle drm-purgable-.
v7: make drm-active- optional
Yunxiang Li (4):
drm: add drm_memory_stats_is_zero
drm: make drm-active- stats optional
drm/amdgpu: remove unused function parameter
drm/amdgpu: track bo memory stats at runtime
drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 4 +-
drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c | 17 +-
drivers/gpu/drm/amd/amdgpu/amdgpu_fdinfo.c | 10 +-
drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c | 3 +-
drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 99 +++-------
drivers/gpu/drm/amd/amdgpu/amdgpu_object.h | 5 +-
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h | 2 +-
drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 197 +++++++++++++++-----
drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h | 18 +-
drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c | 1 +
drivers/gpu/drm/drm_file.c | 23 ++-
drivers/gpu/drm/i915/i915_drm_client.c | 1 +
drivers/gpu/drm/xe/xe_drm_client.c | 1 +
include/drm/drm_file.h | 1 +
include/drm/drm_gem.h | 14 +-
15 files changed, 241 insertions(+), 155 deletions(-)
--
2.34.1
next reply other threads:[~2024-11-10 15:42 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-11-10 15:41 Yunxiang Li [this message]
2024-11-10 15:41 ` [PATCH v7 1/4] drm: add drm_memory_stats_is_zero Yunxiang Li
2024-11-10 15:41 ` [PATCH v7 2/4] drm: make drm-active- stats optional Yunxiang Li
2024-11-11 9:14 ` Jani Nikula
2024-11-11 14:28 ` [PATCH v8 " Yunxiang Li
2024-11-11 10:29 ` [PATCH v7 " Tvrtko Ursulin
2024-11-18 15:17 ` Li, Yunxiang (Teddy)
2024-11-18 15:42 ` Tvrtko Ursulin
2024-11-10 15:41 ` [PATCH v7 3/4] drm/amdgpu: remove unused function parameter Yunxiang Li
2024-11-10 15:41 ` [PATCH v7 4/4] drm/amdgpu: track bo memory stats at runtime Yunxiang Li
2024-11-12 10:54 ` Christian König
2024-11-12 18:16 ` Li, Yunxiang (Teddy)
2024-11-13 8:49 ` Christian König
2024-11-13 10:25 ` Tvrtko Ursulin
2024-11-13 11:38 ` Christian König
2024-11-13 13:53 ` Li, Yunxiang (Teddy)
2024-11-13 14:22 ` Christian König
2024-11-13 17:01 ` Li, Yunxiang (Teddy)
2024-11-13 17:30 ` Tvrtko Ursulin
2024-11-14 9:06 ` Matthew Auld
2024-11-14 15:52 ` Li, Yunxiang (Teddy)
2024-11-13 14:09 ` Li, Yunxiang (Teddy)
2024-11-13 14:19 ` 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=20241110154152.592-1-Yunxiang.Li@amd.com \
--to=yunxiang.li@amd.com \
--cc=Alexander.Deucher@amd.com \
--cc=amd-gfx@lists.freedesktop.org \
--cc=christian.koenig@amd.com \
--cc=tvrtko.ursulin@igalia.com \
/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