From: Andrey Grodzovsky <andrey.grodzovsky@amd.com> To: <dri-devel@lists.freedesktop.org>, <amd-gfx@lists.freedesktop.org> Cc: jingwen.chen2@amd.com, Christian.Koenig@amd.com, monk.liu@amd.com, yiqing.yao@amd.com Subject: [PATCH 0/5] Rework amdgpu HW fence refocunt and update scheduler parent fence refcount. Date: Mon, 20 Jun 2022 18:02:57 -0400 [thread overview] Message-ID: <20220620220302.86389-1-andrey.grodzovsky@amd.com> (raw) Yiqing raised a problem of negative fence refcount for resubmitted jobs in amdgpu and suggested a workaround in [1]. I took a look myself and discovered some deeper problems both in amdgpu and scheduler code. Yiqing helped with testing the new code and also drew a detailed refcount and flow tracing diagram for parent (HW) fence life cycle and refcount under various cases for the proposed patchset at [2]. [1] - https://lore.kernel.org/all/731b7ff1-3cc9-e314-df2a-7c51b76d4db0@amd.com/t/#r00c728fcc069b1276642c325bfa9d82bf8fa21a3 [2] - https://drive.google.com/file/d/1yEoeW6OQC9WnwmzFW6NBLhFP_jD0xcHm/view?usp=sharing Andrey Grodzovsky (5): drm/amdgpu: Fix possible refcount leak for release of external_hw_fence drm/amdgpu: Add put fence in amdgpu_fence_driver_clear_job_fences drm/amdgpu: Prevent race between late signaled fences and GPU reset. drm/sched: Partial revert of 'drm/sched: Keep s_fence->parent pointer' drm/amdgpu: Follow up change to previous drm scheduler change. drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c | 2 ++ drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 27 ++++++++++++---- drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c | 37 ++++++++++++++++++++-- drivers/gpu/drm/amd/amdgpu/amdgpu_job.c | 12 +++---- drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h | 1 + drivers/gpu/drm/scheduler/sched_main.c | 16 ++++++++-- 6 files changed, 78 insertions(+), 17 deletions(-) -- 2.25.1
WARNING: multiple messages have this Message-ID (diff)
From: Andrey Grodzovsky <andrey.grodzovsky@amd.com> To: <dri-devel@lists.freedesktop.org>, <amd-gfx@lists.freedesktop.org> Cc: Andrey Grodzovsky <andrey.grodzovsky@amd.com>, jingwen.chen2@amd.com, Christian.Koenig@amd.com, monk.liu@amd.com, yiqing.yao@amd.com Subject: [PATCH 0/5] Rework amdgpu HW fence refocunt and update scheduler parent fence refcount. Date: Mon, 20 Jun 2022 18:02:57 -0400 [thread overview] Message-ID: <20220620220302.86389-1-andrey.grodzovsky@amd.com> (raw) Yiqing raised a problem of negative fence refcount for resubmitted jobs in amdgpu and suggested a workaround in [1]. I took a look myself and discovered some deeper problems both in amdgpu and scheduler code. Yiqing helped with testing the new code and also drew a detailed refcount and flow tracing diagram for parent (HW) fence life cycle and refcount under various cases for the proposed patchset at [2]. [1] - https://lore.kernel.org/all/731b7ff1-3cc9-e314-df2a-7c51b76d4db0@amd.com/t/#r00c728fcc069b1276642c325bfa9d82bf8fa21a3 [2] - https://drive.google.com/file/d/1yEoeW6OQC9WnwmzFW6NBLhFP_jD0xcHm/view?usp=sharing Andrey Grodzovsky (5): drm/amdgpu: Fix possible refcount leak for release of external_hw_fence drm/amdgpu: Add put fence in amdgpu_fence_driver_clear_job_fences drm/amdgpu: Prevent race between late signaled fences and GPU reset. drm/sched: Partial revert of 'drm/sched: Keep s_fence->parent pointer' drm/amdgpu: Follow up change to previous drm scheduler change. drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c | 2 ++ drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 27 ++++++++++++---- drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c | 37 ++++++++++++++++++++-- drivers/gpu/drm/amd/amdgpu/amdgpu_job.c | 12 +++---- drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h | 1 + drivers/gpu/drm/scheduler/sched_main.c | 16 ++++++++-- 6 files changed, 78 insertions(+), 17 deletions(-) -- 2.25.1
next reply other threads:[~2022-06-20 22:03 UTC|newest] Thread overview: 34+ messages / expand[flat|nested] mbox.gz Atom feed top 2022-06-20 22:02 Andrey Grodzovsky [this message] 2022-06-20 22:02 ` [PATCH 0/5] Rework amdgpu HW fence refocunt and update scheduler parent fence refcount Andrey Grodzovsky 2022-06-20 22:02 ` [PATCH 1/5] drm/amdgpu: Fix possible refcount leak for release of external_hw_fence Andrey Grodzovsky 2022-06-20 22:02 ` Andrey Grodzovsky 2022-06-21 7:19 ` Christian König 2022-06-21 19:34 ` Andrey Grodzovsky 2022-06-22 9:00 ` Christian König 2022-06-22 15:01 ` Andrey Grodzovsky 2022-06-22 15:04 ` Christian König 2022-06-23 21:18 ` Andrey Grodzovsky 2022-06-24 6:00 ` Christian König 2022-06-20 22:02 ` [PATCH 2/5] drm/amdgpu: Add put fence in amdgpu_fence_driver_clear_job_fences Andrey Grodzovsky 2022-06-20 22:02 ` Andrey Grodzovsky 2022-06-21 7:21 ` Christian König 2022-06-20 22:03 ` [PATCH 3/5] drm/amdgpu: Prevent race between late signaled fences and GPU reset Andrey Grodzovsky 2022-06-20 22:03 ` Andrey Grodzovsky 2022-06-21 7:25 ` Christian König 2022-06-21 19:45 ` Andrey Grodzovsky 2022-06-22 1:47 ` VURDIGERENATARAJ, CHANDAN 2022-06-22 2:41 ` Andrey Grodzovsky 2022-06-22 17:31 ` Andrey Grodzovsky 2022-06-23 5:57 ` Christian König 2022-06-20 22:03 ` [PATCH 4/5] drm/sched: Partial revert of 'drm/sched: Keep s_fence->parent pointer' Andrey Grodzovsky 2022-06-20 22:03 ` Andrey Grodzovsky 2022-06-21 7:26 ` Christian König 2022-06-20 22:03 ` [PATCH 5/5] drm/amdgpu: Follow up change to previous drm scheduler change Andrey Grodzovsky 2022-06-20 22:03 ` Andrey Grodzovsky 2022-06-21 7:28 ` Christian König 2022-06-21 20:00 ` Andrey Grodzovsky 2022-06-22 7:17 ` Christian König 2022-06-22 17:19 ` Andrey Grodzovsky 2022-06-23 5:52 ` Christian König 2022-06-23 14:51 ` Andrey Grodzovsky 2022-06-23 14:54 ` 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=20220620220302.86389-1-andrey.grodzovsky@amd.com \ --to=andrey.grodzovsky@amd.com \ --cc=Christian.Koenig@amd.com \ --cc=amd-gfx@lists.freedesktop.org \ --cc=dri-devel@lists.freedesktop.org \ --cc=jingwen.chen2@amd.com \ --cc=monk.liu@amd.com \ --cc=yiqing.yao@amd.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: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.