From: Jiwoo Kim <arpe1618@gmail.com>
To: Rob Clark <robin.clark@oss.qualcomm.com>,
Dmitry Baryshkov <lumag@kernel.org>
Cc: Akhil P Oommen <akhilpo@oss.qualcomm.com>,
Konrad Dybcio <konradybcio@kernel.org>,
Anna Maniscalco <anna.maniscalco2000@gmail.com>,
Julius Bairaktaris <julius@bairaktaris.de>,
Abhinav Kumar <abhinav.kumar@linux.dev>,
Jessica Zhang <jesszhan0024@gmail.com>,
Sean Paul <sean@poorly.run>,
Marijn Suijten <marijn.suijten@somainline.org>,
David Airlie <airlied@gmail.com>, Simona Vetter <simona@ffwll.ch>,
linux-arm-msm@vger.kernel.org, dri-devel@lists.freedesktop.org,
freedreno@lists.freedesktop.org, linux-kernel@vger.kernel.org,
stable@vger.kernel.org
Subject: [PATCH v4 0/5] drm/msm: Fix VM_BIND job, queue and entity lifetimes
Date: Wed, 18 Nov 2026 11:07:11 +0900 [thread overview]
Message-ID: <20261118020716.1698-1-arpe1618@gmail.com> (raw)
These fix what happens to VM_BIND jobs, queues and entities when the
file or a VM_BIND queue is closed while jobs are still pending, and
make the VM unusable when VM_BIND jobs of different queues apply an
unmap before the map that it undoes.
Patch 1 makes msm_gem_vm_close() wait for VM_BIND jobs, as it does for
submits. Patch 2 handles an unmap that finds part of its range not
mapped, or that comes before the map it undoes, like a failed map.
Patch 3 makes a VM_BIND job that the scheduler kills before it runs
apply its queued unmaps before it drops its references to the unmapped
objects, release its BO pins, and mark the VM unusable.
Patch 4 destroys the VM_BIND entity when its queue is closed, instead
of on the final put, which can come after the VM's scheduler has been
torn down. Patch 5 puts the queue last when a job is freed, as the
queue may hold the last reference to the VM.
Changes in v4:
- New patch 1, as Rob suggested: VM_BIND jobs add their fences to
vm->last_fence, keeping the previous fence if the merge fails. Patch
4 is still needed for killed jobs, which the scheduler frees only
after it has signalled their fences.
- New patch 2, for the ordering problem across VM_BIND queues that
Sashiko found in v3 patch 2, and that v3 patch 1 also made reachable
through killed jobs. Unmap errors are now handled like map errors, as
Rob suggested, and so is an unmap that comes before the map it undoes.
A job that fails in the ioctl after updating the VMAs now marks the
VM unusable as well.
- Patch 3 (v3 patch 1): uses the checks of patch 2, and describes the
window that Sashiko pointed out in v3, where a job that waits for a
killed job applies its updates before the killed job's unmaps.
- Patch 4 (v3 patch 2): changelog updated for patch 1, and a second
Fixes tag, for the commit that added VM_BIND jobs.
Changes in v3:
- Patch 1: only apply the queued ops of a job that the scheduler frees
after killing it, once its dependencies and the previous job of its
entity are done. The error path of msm_ioctl_vm_bind() frees the job
right away, so v2 could apply its unmaps too early and out of order
(Sashiko). That path is left as it is in mainline, and the title now
says which jobs the patch is about.
- Patch 3: rebased on patch 1, title adjusted.
Changes in v2:
- New patch 1, for an issue that Sashiko reported on v1 patch 1, which
that patch makes easier to hit.
- Patch 2: changelog refers to patch 1.
Review of v2 by Rob:
https://lore.kernel.org/r/CACSVV00ZppfBKwsHJ+NBYBB7a2f3g1O=6J4Gt1m_A74fudZQoQ@mail.gmail.com
v3: https://lore.kernel.org/r/20261114141851.134290-1-arpe1618@gmail.com
v2: https://lore.kernel.org/r/20261114081902.1669-1-arpe1618@gmail.com
v1: https://lore.kernel.org/r/20261114062541.2001-1-arpe1618@gmail.com
Jiwoo Kim (5):
drm/msm: Wait for VM_BIND jobs in msm_gem_vm_close()
drm/msm: Mark the VM unusable on out-of-order VM_BIND unmaps
drm/msm: Release the pins and ops of a killed VM_BIND job
drm/msm: Destroy the VM_BIND entity when its queue is closed
drm/msm: Put the VM_BIND queue last when freeing a VM_BIND job
drivers/gpu/drm/msm/msm_gem.h | 7 +
drivers/gpu/drm/msm/msm_gem_vma.c | 242 ++++++++++++++++++++++++--
drivers/gpu/drm/msm/msm_gpu.h | 6 +
drivers/gpu/drm/msm/msm_submitqueue.c | 50 ++++--
4 files changed, 272 insertions(+), 33 deletions(-)
base-commit: 1e59855738454b14021b64115f0090eafc7526ae
--
2.43.0
next reply other threads:[~2026-11-18 2:07 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-11-18 2:07 Jiwoo Kim [this message]
2026-11-18 2:07 ` [PATCH v4 1/5] drm/msm: Wait for VM_BIND jobs in msm_gem_vm_close() Jiwoo Kim
2026-11-18 2:07 ` [PATCH v4 2/5] drm/msm: Mark the VM unusable on out-of-order VM_BIND unmaps Jiwoo Kim
2026-11-18 2:45 ` sashiko-bot
2026-11-18 2:07 ` [PATCH v4 3/5] drm/msm: Release the pins and ops of a killed VM_BIND job Jiwoo Kim
2026-11-18 2:07 ` [PATCH v4 4/5] drm/msm: Destroy the VM_BIND entity when its queue is closed Jiwoo Kim
2026-11-18 2:07 ` [PATCH v4 5/5] drm/msm: Put the VM_BIND queue last when freeing a VM_BIND job Jiwoo Kim
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=20261118020716.1698-1-arpe1618@gmail.com \
--to=arpe1618@gmail.com \
--cc=abhinav.kumar@linux.dev \
--cc=airlied@gmail.com \
--cc=akhilpo@oss.qualcomm.com \
--cc=anna.maniscalco2000@gmail.com \
--cc=dri-devel@lists.freedesktop.org \
--cc=freedreno@lists.freedesktop.org \
--cc=jesszhan0024@gmail.com \
--cc=julius@bairaktaris.de \
--cc=konradybcio@kernel.org \
--cc=linux-arm-msm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=lumag@kernel.org \
--cc=marijn.suijten@somainline.org \
--cc=robin.clark@oss.qualcomm.com \
--cc=sean@poorly.run \
--cc=simona@ffwll.ch \
--cc=stable@vger.kernel.org \
/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 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.