Intel-XE Archive on lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH v2 1/4] drm/xe/ct: prevent UAF in send_recv()
@ 2024-10-01  8:43 Matthew Auld
  2024-10-01  8:43 ` [PATCH v2 2/4] drm/xe/ct: fix xa_store() error checking Matthew Auld
                   ` (11 more replies)
  0 siblings, 12 replies; 16+ messages in thread
From: Matthew Auld @ 2024-10-01  8:43 UTC (permalink / raw)
  To: intel-xe; +Cc: Matthew Brost, Badal Nilawar, stable

Ensure we serialize with completion side to prevent UAF with fence going
out of scope on the stack, since we have no clue if it will fire after
the timeout before we can erase from the xa. Also we have some dependent
loads and stores for which we need the correct ordering, and we lack the
needed barriers. Fix this by grabbing the ct->lock after the wait, which
is also held by the completion side.

v2 (Badal):
 - Also print done after acquiring the lock and seeing timeout.

Fixes: dd08ebf6c352 ("drm/xe: Introduce a new DRM driver for Intel GPUs")
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Badal Nilawar <badal.nilawar@intel.com>
Cc: <stable@vger.kernel.org> # v6.8+
---
 drivers/gpu/drm/xe/xe_guc_ct.c | 21 ++++++++++++++++++---
 1 file changed, 18 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_guc_ct.c b/drivers/gpu/drm/xe/xe_guc_ct.c
index 4b95f75b1546..44263b3cd8c7 100644
--- a/drivers/gpu/drm/xe/xe_guc_ct.c
+++ b/drivers/gpu/drm/xe/xe_guc_ct.c
@@ -903,16 +903,26 @@ static int guc_ct_send_recv(struct xe_guc_ct *ct, const u32 *action, u32 len,
 	}
 
 	ret = wait_event_timeout(ct->g2h_fence_wq, g2h_fence.done, HZ);
+
+	/*
+	 * Ensure we serialize with completion side to prevent UAF with fence going out of scope on
+	 * the stack, since we have no clue if it will fire after the timeout before we can erase
+	 * from the xa. Also we have some dependent loads and stores below for which we need the
+	 * correct ordering, and we lack the needed barriers.
+	 */
+	mutex_lock(&ct->lock);
 	if (!ret) {
-		xe_gt_err(gt, "Timed out wait for G2H, fence %u, action %04x",
-			  g2h_fence.seqno, action[0]);
+		xe_gt_err(gt, "Timed out wait for G2H, fence %u, action %04x, done %s",
+			  g2h_fence.seqno, action[0], str_yes_no(g2h_fence.done));
 		xa_erase_irq(&ct->fence_lookup, g2h_fence.seqno);
+		mutex_unlock(&ct->lock);
 		return -ETIME;
 	}
 
 	if (g2h_fence.retry) {
 		xe_gt_dbg(gt, "H2G action %#x retrying: reason %#x\n",
 			  action[0], g2h_fence.reason);
+		mutex_unlock(&ct->lock);
 		goto retry;
 	}
 	if (g2h_fence.fail) {
@@ -921,7 +931,12 @@ static int guc_ct_send_recv(struct xe_guc_ct *ct, const u32 *action, u32 len,
 		ret = -EIO;
 	}
 
-	return ret > 0 ? response_buffer ? g2h_fence.response_len : g2h_fence.response_data : ret;
+	if (ret > 0)
+		ret = response_buffer ? g2h_fence.response_len : g2h_fence.response_data;
+
+	mutex_unlock(&ct->lock);
+
+	return ret;
 }
 
 /**
-- 
2.46.2


^ permalink raw reply related	[flat|nested] 16+ messages in thread

end of thread, other threads:[~2024-10-01 16:43 UTC | newest]

Thread overview: 16+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2024-10-01  8:43 [PATCH v2 1/4] drm/xe/ct: prevent UAF in send_recv() Matthew Auld
2024-10-01  8:43 ` [PATCH v2 2/4] drm/xe/ct: fix xa_store() error checking Matthew Auld
2024-10-01 13:48   ` Nilawar, Badal
2024-10-01  8:43 ` [PATCH v2 3/4] drm/xe/guc_submit: " Matthew Auld
2024-10-01 13:50   ` Nilawar, Badal
2024-10-01  8:43 ` [PATCH v2 4/4] drm/xe/ct: drop irq usage of xa_erase() Matthew Auld
2024-10-01 14:55   ` Nilawar, Badal
2024-10-01  8:51 ` ✓ CI.Patch_applied: success for series starting with [v2,1/4] drm/xe/ct: prevent UAF in send_recv() Patchwork
2024-10-01  8:51 ` ✓ CI.checkpatch: " Patchwork
2024-10-01  8:52 ` ✓ CI.KUnit: " Patchwork
2024-10-01  9:09 ` ✓ CI.Build: " Patchwork
2024-10-01  9:12 ` ✓ CI.Hooks: " Patchwork
2024-10-01  9:14 ` ✓ CI.checksparse: " Patchwork
2024-10-01  9:40 ` ✓ CI.BAT: " Patchwork
2024-10-01 13:22 ` [PATCH v2 1/4] " Nilawar, Badal
2024-10-01 16:43 ` ✗ CI.FULL: failure for series starting with [v2,1/4] " Patchwork

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox