From: Shuicheng Lin <shuicheng.lin@intel.com>
To: intel-xe@lists.freedesktop.org
Cc: Shuicheng Lin <shuicheng.lin@intel.com>
Subject: [PATCH v6 0/2] drm/xe/shrinker: Runtime PM reference for non-system memory
Date: Fri, 4 Sep 2026 21:50:47 +0000 [thread overview]
Message-ID: <20260904215049.543535-1-shuicheng.lin@intel.com> (raw)
Patch 2 fixes the reported bug: the shrinker walks the SYSTEM and TT
LRUs without a runtime PM reference, so with the device runtime
suspended the page table zap trips an assert and the TLB invalidation
returns -ENODEV.
Patch 1 comes first because its Fixes commit is a year older and it
applies on its own to stable trees that must not take patch 2. It
separates the freed page count from the return value so an error no
longer discards the pages already freed.
Verified with igt@xe_madvise@dontneed-before-exec while the GPU is
runtime suspended: 15/15 runs hit the WARN before, 0/15 after, and a
CI-like configuration with the device active still passes 10/10.
Shuicheng Lin (2):
drm/xe/shrinker: Return the freed page count through a parameter
drm/xe/shrinker: Take a runtime PM ref before shrinking non-system
memory
drivers/gpu/drm/xe/xe_shrinker.c | 114 ++++++++++++++++++-------------
1 file changed, 66 insertions(+), 48 deletions(-)
--
2.43.0
next reply other threads:[~2026-09-04 21:51 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-09-04 21:50 Shuicheng Lin [this message]
2026-09-04 21:50 ` [PATCH v6 1/2] drm/xe/shrinker: Return the freed page count through a parameter Shuicheng Lin
2026-09-04 21:59 ` sashiko-bot
2026-09-09 12:58 ` Thomas Hellström
2026-09-04 21:50 ` [PATCH v6 2/2] drm/xe/shrinker: Take a runtime PM ref before shrinking non-system memory Shuicheng Lin
2026-09-04 22:05 ` sashiko-bot
2026-09-07 8:23 ` Thomas Hellström
2026-09-04 22:10 ` ✓ CI.KUnit: success for drm/xe/shrinker: Runtime PM reference for " Patchwork
2026-09-04 23:05 ` ✓ Xe.CI.BAT: " Patchwork
2026-09-05 4:15 ` ✓ Xe.CI.FULL: " Patchwork
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=20260904215049.543535-1-shuicheng.lin@intel.com \
--to=shuicheng.lin@intel.com \
--cc=intel-xe@lists.freedesktop.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox