From: Matthew Brost <matthew.brost@intel.com>
To: intel-xe@lists.freedesktop.org
Subject: [PATCH 1/2] drm/xe: Avoid memory allocations in xe_device_declare_wedged()
Date: Thu, 26 Mar 2026 14:01:15 -0700 [thread overview]
Message-ID: <20260326210116.202585-2-matthew.brost@intel.com> (raw)
In-Reply-To: <20260326210116.202585-1-matthew.brost@intel.com>
xe_device_declare_wedged() runs in the DMA-fence signaling path, where
GFP_KERNEL memory allocations are not allowed. However, registering
xe_device_wedged_fini via drmm_add_action_or_reset() triggers a
GFP_KERNEL allocation.
Fix this by deferring the registration of xe_device_wedged_fini until
late in the driver load sequence. Additionally, drop the wedged PM
reference only if the device is actually wedged in
xe_device_wedged_fini.
Fixes: ("452bca0edbd0 drm/xe: Don't suspend device upon wedge")
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
---
drivers/gpu/drm/xe/xe_device.c | 27 +++++++++++++--------------
1 file changed, 13 insertions(+), 14 deletions(-)
diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c
index 9b0540e3e289..ffea4a453c01 100644
--- a/drivers/gpu/drm/xe/xe_device.c
+++ b/drivers/gpu/drm/xe/xe_device.c
@@ -833,6 +833,14 @@ static void detect_preproduction_hw(struct xe_device *xe)
}
}
+static void xe_device_wedged_fini(struct drm_device *drm, void *arg)
+{
+ struct xe_device *xe = arg;
+
+ if (atomic_read(&xe->wedged.flag))
+ xe_pm_runtime_put(xe);
+}
+
int xe_device_probe(struct xe_device *xe)
{
struct xe_tile *tile;
@@ -1009,6 +1017,10 @@ int xe_device_probe(struct xe_device *xe)
detect_preproduction_hw(xe);
+ err = drmm_add_action_or_reset(&xe->drm, xe_device_wedged_fini, xe);
+ if (err)
+ goto err_unregister_display;
+
return devm_add_action_or_reset(xe->drm.dev, xe_device_sanitize, xe);
err_unregister_display:
@@ -1240,13 +1252,6 @@ u64 xe_device_uncanonicalize_addr(struct xe_device *xe, u64 address)
return address & GENMASK_ULL(xe->info.va_bits - 1, 0);
}
-static void xe_device_wedged_fini(struct drm_device *drm, void *arg)
-{
- struct xe_device *xe = arg;
-
- xe_pm_runtime_put(xe);
-}
-
/**
* DOC: Xe Device Wedging
*
@@ -1324,15 +1329,9 @@ void xe_device_declare_wedged(struct xe_device *xe)
return;
}
- xe_pm_runtime_get_noresume(xe);
-
- if (drmm_add_action_or_reset(&xe->drm, xe_device_wedged_fini, xe)) {
- drm_err(&xe->drm, "Failed to register xe_device_wedged_fini clean-up. Although device is wedged.\n");
- return;
- }
-
if (!atomic_xchg(&xe->wedged.flag, 1)) {
xe->needs_flr_on_fini = true;
+ xe_pm_runtime_get_noresume(xe);
drm_err(&xe->drm,
"CRITICAL: Xe has declared device %s as wedged.\n"
"IOCTLs and executions are blocked.\n"
--
2.34.1
next prev parent reply other threads:[~2026-03-26 21:01 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-03-26 21:01 [PATCH 0/2] Wedged memory reclaim fixes Matthew Brost
2026-03-26 21:01 ` Matthew Brost [this message]
2026-03-27 13:15 ` [PATCH 1/2] drm/xe: Avoid memory allocations in xe_device_declare_wedged() Rodrigo Vivi
2026-03-26 21:01 ` [PATCH 2/2] drm/xe: Drop registration of guc_submit_wedged_fini from xe_guc_submit_wedge() Matthew Brost
2026-03-27 13:19 ` Rodrigo Vivi
2026-03-26 21:11 ` ✓ CI.KUnit: success for Wedged memory reclaim fixes Patchwork
2026-03-26 22:01 ` ✓ Xe.CI.BAT: " Patchwork
2026-03-27 14:45 ` ✓ 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=20260326210116.202585-2-matthew.brost@intel.com \
--to=matthew.brost@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