From: Tomasz Lis <tomasz.lis@intel.com>
To: intel-xe@lists.freedesktop.org
Cc: "Michał Winiarski" <michal.winiarski@intel.com>,
"Michał Wajdeczko" <michal.wajdeczko@intel.com>,
"Piotr Piórkowski" <piotr.piorkowski@intel.com>,
"Matthew Brost" <matthew.brost@intel.com>
Subject: [PATCH v3 4/4] drm/xe/vf: Redo LRC creation while in VF fixups
Date: Thu, 26 Feb 2026 00:54:47 +0100 [thread overview]
Message-ID: <20260225235447.2772383-5-tomasz.lis@intel.com> (raw)
In-Reply-To: <20260225235447.2772383-1-tomasz.lis@intel.com>
If the xe module within a VM was creating a new LRC during save/
restore, this LRC will be invalid. The fixups procedure may not
be able to reach it, as there will be a race to add the new LRC
reference to an exec queue.
Even if the new LRC which was being created during VM migration is
added to EQ in time for fixups, said LRC may still remain damaged.
In a small percentage of specially crafted test cases, the resulting
LRC was still damaged and caused GPU hang.
Any LRC which could be created in such a situation, have to be
re-created.
Due to VM having arbitrarily set amount of CPU cores, it is possible
to limit the amount to 1. In such case, there is a possibility that
kernel will switch CPU contexts in a way which allows to miss
VF migration recovery running in parallel (by simply not switching
to the LRC creation thread during recovery). Therefore checking
if the migration is in progress just after LRC creation, is not
enough to ensure detection.
Free the incorrectly created LRC, and trigger a re-run of the
creation, but only after waiting for default LRC to get fixups.
Use additional atomic value increased after fixups, to ensure any VF
migration that avoided detection by just checking for recovery in
progress, will be caught.
v2: Merge marker and wait for default LRC, reducing amount of calls
within xe_init_eq(). Alter the LRC creation loop to remove a race
with post-migration fixups worker.
Signed-off-by: Tomasz Lis <tomasz.lis@intel.com>
---
drivers/gpu/drm/xe/xe_exec_queue.c | 29 ++++++++++++++++-------
drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 25 +++++++++++++++++--
drivers/gpu/drm/xe/xe_gt_sriov_vf.h | 3 ++-
drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h | 2 ++
4 files changed, 47 insertions(+), 12 deletions(-)
diff --git a/drivers/gpu/drm/xe/xe_exec_queue.c b/drivers/gpu/drm/xe/xe_exec_queue.c
index 2cb37af42021..07c6d8bc3ad8 100644
--- a/drivers/gpu/drm/xe/xe_exec_queue.c
+++ b/drivers/gpu/drm/xe/xe_exec_queue.c
@@ -365,17 +365,28 @@ static int __xe_exec_queue_init(struct xe_exec_queue *q, u32 exec_queue_flags)
* from the moment vCPU resumes execution.
*/
for (i = 0; i < q->width; ++i) {
- struct xe_lrc *lrc;
+ struct xe_lrc *__lrc = NULL;
+ int marker;
- xe_gt_sriov_vf_wait_valid_ggtt(q->gt);
- lrc = xe_lrc_create(q->hwe, q->vm, q->replay_state,
- xe_lrc_ring_size(), q->msix_vec, flags);
- if (IS_ERR(lrc)) {
- err = PTR_ERR(lrc);
- goto err_lrc;
- }
+ do {
+ struct xe_lrc *lrc;
+
+ marker = xe_gt_sriov_vf_wait_valid_ggtt(q->gt);
+
+ lrc = xe_lrc_create(q->hwe, q->vm, q->replay_state,
+ xe_lrc_ring_size(), q->msix_vec, flags);
+ if (IS_ERR(lrc)) {
+ err = PTR_ERR(lrc);
+ goto err_lrc;
+ }
+
+ xe_exec_queue_set_lrc(q, lrc, i);
+
+ if (__lrc)
+ xe_lrc_put(__lrc);
+ __lrc = lrc;
- xe_exec_queue_set_lrc(q, lrc, i);
+ } while (marker != xe_vf_migration_fixups_complete_count(q->gt));
}
return 0;
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
index 7f83c0d3b099..3da38f2ee317 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
@@ -1277,6 +1277,8 @@ static int vf_post_migration_fixups(struct xe_gt *gt)
if (err)
return err;
+ atomic_inc(>->sriov.vf.migration.fixups_complete);
+
return 0;
}
@@ -1515,20 +1517,39 @@ static bool vf_valid_ggtt(struct xe_gt *gt)
return true;
}
+int xe_vf_migration_fixups_complete_count(struct xe_gt *gt)
+{
+ if (!IS_SRIOV_VF(gt_to_xe(gt)) ||
+ !xe_sriov_vf_migration_supported(gt_to_xe(gt)))
+ return 0;
+
+ /* should never match fixups_complete value */
+ if (!vf_valid_ggtt(gt))
+ return -1;
+
+ return atomic_read(>->sriov.vf.migration.fixups_complete);
+}
+
/**
* xe_gt_sriov_vf_wait_valid_ggtt() - wait for valid GGTT nodes and address refs
* @gt: the &xe_gt
*/
-void xe_gt_sriov_vf_wait_valid_ggtt(struct xe_gt *gt)
+int xe_gt_sriov_vf_wait_valid_ggtt(struct xe_gt *gt)
{
int ret;
+ /*
+ * this condition shall identical to one in
+ * xe_vf_migration_fixups_complete_count()
+ */
if (!IS_SRIOV_VF(gt_to_xe(gt)) ||
!xe_sriov_vf_migration_supported(gt_to_xe(gt)))
- return;
+ return 0;
ret = wait_event_interruptible_timeout(gt->sriov.vf.migration.wq,
vf_valid_ggtt(gt),
HZ * 5);
xe_gt_WARN_ON(gt, !ret);
+
+ return atomic_read(>->sriov.vf.migration.fixups_complete);
}
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
index 7d97189c2d3d..a6f7127521a5 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
@@ -39,6 +39,7 @@ void xe_gt_sriov_vf_print_config(struct xe_gt *gt, struct drm_printer *p);
void xe_gt_sriov_vf_print_runtime(struct xe_gt *gt, struct drm_printer *p);
void xe_gt_sriov_vf_print_version(struct xe_gt *gt, struct drm_printer *p);
-void xe_gt_sriov_vf_wait_valid_ggtt(struct xe_gt *gt);
+int xe_gt_sriov_vf_wait_valid_ggtt(struct xe_gt *gt);
+int xe_vf_migration_fixups_complete_count(struct xe_gt *gt);
#endif
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
index fca18be589db..0b397f259a26 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
@@ -54,6 +54,8 @@ struct xe_gt_sriov_vf_migration {
wait_queue_head_t wq;
/** @scratch: Scratch memory for VF recovery */
void *scratch;
+ /** @fixups_complete: Counts completed fixups stages */
+ atomic_t fixups_complete;
/** @debug: Debug hooks for delaying migration */
struct {
/**
--
2.25.1
next prev parent reply other threads:[~2026-02-25 23:50 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-02-25 23:54 [PATCH v3 0/4] drm/xe/vf: Fix exec queue creation during post-migration recovery Tomasz Lis
2026-02-25 23:54 ` [PATCH v3 1/4] drm/xe/queue: Call fini on exec queue creation fail Tomasz Lis
2026-02-25 23:54 ` [PATCH v3 2/4] drm/xe/queue: Wrappers for setting and getting LRC references Tomasz Lis
2026-02-26 1:42 ` Matthew Brost
2026-02-26 1:57 ` Matthew Brost
2026-02-26 17:26 ` Lis, Tomasz
2026-02-26 17:30 ` Matthew Brost
2026-02-25 23:54 ` [PATCH v3 3/4] drm/xe/vf: Wait for all fixups before using default LRCs Tomasz Lis
2026-02-25 23:54 ` Tomasz Lis [this message]
2026-02-26 1:49 ` [PATCH v3 4/4] drm/xe/vf: Redo LRC creation while in VF fixups Matthew Brost
2026-02-26 1:24 ` ✓ CI.KUnit: success for drm/xe/vf: Fix exec queue creation during post-migration recovery (rev3) Patchwork
2026-02-26 2:02 ` ✓ Xe.CI.BAT: " Patchwork
2026-02-26 6:08 ` ✗ 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=20260225235447.2772383-5-tomasz.lis@intel.com \
--to=tomasz.lis@intel.com \
--cc=intel-xe@lists.freedesktop.org \
--cc=matthew.brost@intel.com \
--cc=michal.wajdeczko@intel.com \
--cc=michal.winiarski@intel.com \
--cc=piotr.piorkowski@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