From: Riana Tauro <riana.tauro@intel.com>
To: intel-xe@lists.freedesktop.org
Cc: riana.tauro@intel.com, anshuman.gupta@intel.com,
rodrigo.vivi@intel.com, aravind.iddamsetty@linux.intel.com,
badal.nilawar@intel.com, raag.jadav@intel.com,
ravi.kishore.koppuravuri@intel.com, mallesh.koujalagi@intel.com,
soham.purkait@intel.com, Matthew Brost <matthew.brost@intel.com>,
Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Subject: [PATCH v4 04/13] drm/xe: Skip device access during PCI error recovery
Date: Fri, 17 Apr 2026 14:28:16 +0530 [thread overview]
Message-ID: <20260417085812.4013309-19-riana.tauro@intel.com> (raw)
In-Reply-To: <20260417085812.4013309-15-riana.tauro@intel.com>
When a fatal error occurs and the error_detected callback is
invoked the device is inaccessible. The error_detected callback
wedges the device causing the jobs to timeout.
The timedout handler acquires forcewake to dump devcoredump and
triggers a GT reset. Since the device is inacessible this causes
errors. Skip all mmio accesses and gt reset when the device
is in recovery.
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Signed-off-by: Riana Tauro <riana.tauro@intel.com>
---
v2: add check in worker (Mallesh)
---
drivers/gpu/drm/xe/xe_gt.c | 14 +++++++++++---
drivers/gpu/drm/xe/xe_guc_submit.c | 9 +++++----
2 files changed, 16 insertions(+), 7 deletions(-)
diff --git a/drivers/gpu/drm/xe/xe_gt.c b/drivers/gpu/drm/xe/xe_gt.c
index 8a31c963c372..5ea5524d83af 100644
--- a/drivers/gpu/drm/xe/xe_gt.c
+++ b/drivers/gpu/drm/xe/xe_gt.c
@@ -917,6 +917,9 @@ static void gt_reset_worker(struct work_struct *w)
if (xe_device_wedged(gt_to_xe(gt)))
goto err_pm_put;
+ if (xe_device_is_in_recovery(gt_to_xe(gt)))
+ goto err_pm_put;
+
/* We only support GT resets with GuC submission */
if (!xe_device_uc_enabled(gt_to_xe(gt)))
goto err_pm_put;
@@ -977,18 +980,23 @@ static void gt_reset_worker(struct work_struct *w)
void xe_gt_reset_async(struct xe_gt *gt)
{
- xe_gt_info(gt, "trying reset from %ps\n", __builtin_return_address(0));
+ struct xe_device *xe = gt_to_xe(gt);
+
+ if (xe_device_is_in_recovery(xe))
+ return;
/* Don't do a reset while one is already in flight */
if (!xe_fault_inject_gt_reset() && xe_uc_reset_prepare(>->uc))
return;
+ xe_gt_info(gt, "trying reset from %ps\n", __builtin_return_address(0));
+
xe_gt_info(gt, "reset queued\n");
/* Pair with put in gt_reset_worker() if work is enqueued */
- xe_pm_runtime_get_noresume(gt_to_xe(gt));
+ xe_pm_runtime_get_noresume(xe);
if (!queue_work(gt->ordered_wq, >->reset.worker))
- xe_pm_runtime_put(gt_to_xe(gt));
+ xe_pm_runtime_put(xe);
}
void xe_gt_suspend_prepare(struct xe_gt *gt)
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
index 10556156eaad..1f32fb14a5c1 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.c
+++ b/drivers/gpu/drm/xe/xe_guc_submit.c
@@ -1522,7 +1522,7 @@ guc_exec_queue_timedout_job(struct drm_sched_job *drm_job)
* If devcoredump not captured and GuC capture for the job is not ready
* do manual capture first and decide later if we need to use it
*/
- if (!exec_queue_killed(q) && !xe->devcoredump.captured &&
+ if (!xe_device_is_in_recovery(xe) && !exec_queue_killed(q) && !xe->devcoredump.captured &&
!xe_guc_capture_get_matching_and_lock(q)) {
/* take force wake before engine register manual capture */
CLASS(xe_force_wake, fw_ref)(gt_to_fw(q->gt), XE_FORCEWAKE_ALL);
@@ -1544,8 +1544,8 @@ guc_exec_queue_timedout_job(struct drm_sched_job *drm_job)
set_exec_queue_banned(q);
/* Kick job / queue off hardware */
- if (!wedged && (exec_queue_enabled(primary) ||
- exec_queue_pending_disable(primary))) {
+ if (!xe_device_is_in_recovery(xe) && !wedged &&
+ (exec_queue_enabled(primary) || exec_queue_pending_disable(primary))) {
int ret;
if (exec_queue_reset(primary))
@@ -1613,7 +1613,8 @@ guc_exec_queue_timedout_job(struct drm_sched_job *drm_job)
trace_xe_sched_job_timedout(job);
- if (!exec_queue_killed(q))
+ /* Do not access device if in recovery */
+ if (!xe_device_is_in_recovery(xe) && !exec_queue_killed(q))
xe_devcoredump(q, job,
"Timedout job - seqno=%u, lrc_seqno=%u, guc_id=%d, flags=0x%lx",
xe_sched_job_seqno(job), xe_sched_job_lrc_seqno(job),
--
2.47.1
next prev parent reply other threads:[~2026-04-17 8:25 UTC|newest]
Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-04-17 8:58 [PATCH v4 00/13] Introduce Xe Uncorrectable Error Handling Riana Tauro
2026-04-17 8:58 ` [PATCH v4 01/13] drm/xe/xe_survivability: Decouple survivability info from boot survivability Riana Tauro
2026-04-17 8:58 ` [PATCH v4 02/13] drm/xe/xe_pci_error: Implement PCI error recovery callbacks Riana Tauro
2026-04-27 6:35 ` Raag Jadav
2026-05-06 13:59 ` Tauro, Riana
2026-04-17 8:58 ` [PATCH v4 03/13] drm/xe/xe_pci_error: Group all devres to release them on PCIe slot reset Riana Tauro
2026-04-17 8:58 ` Riana Tauro [this message]
2026-04-30 12:58 ` [PATCH v4 04/13] drm/xe: Skip device access during PCI error recovery Anshuman Gupta
2026-05-06 15:41 ` Tauro, Riana
2026-04-17 8:58 ` [PATCH v4 05/13] drm/xe/xe_ras: Initialize Uncorrectable AER Registers Riana Tauro
2026-04-27 7:56 ` Raag Jadav
2026-05-05 5:22 ` Tauro, Riana
2026-04-17 8:58 ` [PATCH v4 06/13] drm/xe/xe_ras: Add basic structures and commands for uncorrectable errors Riana Tauro
2026-04-17 17:38 ` Matt Roper
2026-04-17 21:25 ` Jadav, Raag
2026-04-17 21:32 ` Matt Roper
2026-04-20 5:34 ` Tauro, Riana
2026-04-20 7:49 ` Raag Jadav
2026-04-17 8:58 ` [PATCH v4 07/13] drm/xe/xe_ras: Add support for uncorrectable core-compute errors Riana Tauro
2026-04-27 8:24 ` Raag Jadav
2026-05-05 5:28 ` Tauro, Riana
2026-04-17 8:58 ` [PATCH v4 08/13] drm/xe/xe_ras: Handle uncorrectable SoC Internal errors Riana Tauro
2026-04-17 8:58 ` [PATCH v4 09/13] drm/xe/xe_ras: Handle uncorrectable device memory errors Riana Tauro
2026-04-21 6:08 ` Upadhyay, Tejas
2026-05-05 5:03 ` Tauro, Riana
2026-04-17 8:58 ` [PATCH v4 10/13] drm/xe/xe_ras: Add support to offline/decline a page Riana Tauro
2026-04-21 6:21 ` Upadhyay, Tejas
2026-05-05 5:16 ` Tauro, Riana
2026-04-17 8:58 ` [PATCH v4 11/13] drm/xe/xe_ras: Add support for page offline list and queue commands Riana Tauro
2026-04-21 6:19 ` Upadhyay, Tejas
2026-05-05 5:08 ` Tauro, Riana
2026-04-21 9:10 ` Upadhyay, Tejas
2026-05-05 5:17 ` Tauro, Riana
2026-04-17 8:58 ` [PATCH v4 12/13] drm/xe/xe_ras: Query errors from system controller on probe Riana Tauro
2026-04-28 11:46 ` Raag Jadav
2026-05-05 13:50 ` Tauro, Riana
2026-04-17 8:58 ` [PATCH v4 13/13] drm/xe/xe_pci_error: Process errors in mmio_enabled Riana Tauro
2026-04-28 11:39 ` Raag Jadav
2026-05-05 5:31 ` Tauro, Riana
2026-04-30 11:15 ` Gupta, Anshuman
2026-05-02 17:55 ` Raag Jadav
2026-04-20 13:33 ` ✗ CI.checkpatch: warning for Introduce Xe Uncorrectable Error Handling (rev4) Patchwork
2026-04-20 13:35 ` ✓ CI.KUnit: success " Patchwork
2026-04-20 14:42 ` ✓ Xe.CI.BAT: " Patchwork
2026-04-20 17:14 ` ✗ Xe.CI.FULL: failure " 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=20260417085812.4013309-19-riana.tauro@intel.com \
--to=riana.tauro@intel.com \
--cc=anshuman.gupta@intel.com \
--cc=aravind.iddamsetty@linux.intel.com \
--cc=badal.nilawar@intel.com \
--cc=himal.prasad.ghimiray@intel.com \
--cc=intel-xe@lists.freedesktop.org \
--cc=mallesh.koujalagi@intel.com \
--cc=matthew.brost@intel.com \
--cc=raag.jadav@intel.com \
--cc=ravi.kishore.koppuravuri@intel.com \
--cc=rodrigo.vivi@intel.com \
--cc=soham.purkait@intel.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;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).