Intel-XE Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Tomasz Lis <tomasz.lis@intel.com>
To: intel-xe@lists.freedesktop.org
Cc: "Matthew Brost" <matthew.brost@intel.com>,
	"Michał Winiarski" <michal.winiarski@intel.com>,
	"Michał Wajdeczko" <michal.wajdeczko@intel.com>,
	"Piotr Piórkowski" <piotr.piorkowski@intel.com>,
	"Satyanarayana K V P" <satyanarayana.k.v.p@intel.com>
Subject: [PATCH v1] drm/xe: Protect against unset LRC when pausing submissions
Date: Mon, 24 Nov 2025 23:28:53 +0100	[thread overview]
Message-ID: <20251124222853.1900800-1-tomasz.lis@intel.com> (raw)

While pausing submissions, it is possible to encouner an exec queue
which is during creation, and therefore doesn't have a valid xe_lrc
struct reference.

Protect agains such situation, by checking for NULL before access.

Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Fixes: c25c1010df88 ("drm/xe/vf: Replay GuC submission state on pause / unpause")
Signed-off-by: Tomasz Lis <tomasz.lis@intel.com>
---
 drivers/gpu/drm/xe/xe_guc_submit.c | 22 ++++++++++++++++------
 1 file changed, 16 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
index 7e0882074a99..f57000c2cb9d 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.c
+++ b/drivers/gpu/drm/xe/xe_guc_submit.c
@@ -2107,6 +2107,18 @@ static void guc_exec_queue_revert_pending_state_change(struct xe_guc *guc,
 	q->guc->resume_time = 0;
 }
 
+static void lrc_parallel_clear(struct xe_lrc *lrc)
+{
+	struct xe_device *xe = gt_to_xe(lrc->gt);
+	struct iosys_map map = xe_lrc_parallel_map(lrc);
+	int i;
+
+	for (i = 0; i < WQ_SIZE / sizeof(u32); ++i)
+		parallel_write(xe, map, wq[i],
+			       FIELD_PREP(WQ_TYPE_MASK, WQ_TYPE_NOOP) |
+			       FIELD_PREP(WQ_LEN_MASK, 0));
+}
+
 /*
  * This function is quite complex but only real way to ensure no state is lost
  * during VF resume flows. The function scans the queue state, make adjustments
@@ -2130,8 +2142,8 @@ static void guc_exec_queue_pause(struct xe_guc *guc, struct xe_exec_queue *q)
 	guc_exec_queue_revert_pending_state_change(guc, q);
 
 	if (xe_exec_queue_is_parallel(q)) {
-		struct xe_device *xe = guc_to_xe(guc);
-		struct iosys_map map = xe_lrc_parallel_map(q->lrc[0]);
+		/* Pairs with WRITE_ONCE in __xe_exec_queue_init  */
+		struct xe_lrc *lrc = READ_ONCE(q->lrc[0]);
 
 		/*
 		 * NOP existing WQ commands that may contain stale GGTT
@@ -2139,10 +2151,8 @@ static void guc_exec_queue_pause(struct xe_guc *guc, struct xe_exec_queue *q)
 		 * seems to get confused if the WQ head/tail pointers are
 		 * adjusted.
 		 */
-		for (i = 0; i < WQ_SIZE / sizeof(u32); ++i)
-			parallel_write(xe, map, wq[i],
-				       FIELD_PREP(WQ_TYPE_MASK, WQ_TYPE_NOOP) |
-				       FIELD_PREP(WQ_LEN_MASK, 0));
+		if (lrc)
+			lrc_parallel_clear(lrc);
 	}
 
 	job = xe_sched_first_pending_job(sched);
-- 
2.25.1


             reply	other threads:[~2025-11-24 22:26 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-11-24 22:28 Tomasz Lis [this message]
2025-11-25  7:04 ` ✓ CI.KUnit: success for drm/xe: Protect against unset LRC when pausing submissions (rev2) Patchwork
2025-11-25  8:34 ` ✓ Xe.CI.BAT: " Patchwork
2025-11-25 11:28 ` ✗ Xe.CI.Full: failure " Patchwork
2025-11-25 18:38   ` Lis, Tomasz
  -- strict thread matches above, loose matches on Subject: below --
2025-11-24 21:23 [PATCH v1] drm/xe: Protect against unset LRC when pausing submissions Tomasz Lis
2025-11-24 21:51 ` Matthew Brost

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=20251124222853.1900800-1-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 \
    --cc=satyanarayana.k.v.p@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