From: Shuicheng Lin <shuicheng.lin@intel.com>
To: intel-xe@lists.freedesktop.org
Cc: Shuicheng Lin <shuicheng.lin@intel.com>,
	Sashiko <sashiko-bot@kernel.org>,
	stable@vger.kernel.org,
	=?UTF-8?q?Thomas=20Hellstr=C3=B6m?= <thomas.hellstrom@linux.intel.com>,
	Matthew Brost <matthew.brost@intel.com>
Subject: [PATCH v4 1/2] drm/xe/shrinker: Do not abort the walk on unshrinkable bos
Date: Tue,  1 Sep 2026 03:31:52 +0000
Message-Id: <20260901033153.219481-2-shuicheng.lin@intel.com>
In-Reply-To: <20260901033153.219481-1-shuicheng.lin@intel.com>
References: <20260901033153.219481-1-shuicheng.lin@intel.com>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

xe_bo_shrink() returns -EBUSY for a bo that is simply not a candidate:
a kernel bo, a non-purgeable bo during the purge pass, or one that is
not eviction valuable.  ttm_bo_shrink_suitable() does not filter those
out, so __xe_shrinker_walk() ends the walk at the first one it meets and
returns the error in place of the pages it had already freed.

xe_shrinker_walk() re-enters the walk for each escalation level, so the
remaining bos are usually still reached and the effect is repeated
restarts rather than lost reclaim.

Skip the bo on -EBUSY and keep walking, and on a real error report what
was already freed rather than dropping it.

Reported-by: Sashiko <sashiko-bot@kernel.org>
Fixes: 00c8efc3180f ("drm/xe: Add a shrinker for xe bos")
Cc: stable@vger.kernel.org
Assisted-by: Claude:claude-opus-5
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Shuicheng Lin <shuicheng.lin@intel.com>
---
 drivers/gpu/drm/xe/xe_shrinker.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_shrinker.c b/drivers/gpu/drm/xe/xe_shrinker.c
index 83374cd57660..953a911b51b3 100644
--- a/drivers/gpu/drm/xe/xe_shrinker.c
+++ b/drivers/gpu/drm/xe/xe_shrinker.c
@@ -79,8 +79,10 @@ static s64 __xe_shrinker_walk(struct xe_device *xe,
 				continue;
 
 			lret = xe_bo_shrink(ctx, ttm_bo, flags, scanned);
+			if (lret == -EBUSY)
+				continue;
 			if (lret < 0)
-				return lret;
+				return freed ? freed : lret;
 
 			freed += lret;
 			if (*scanned >= to_scan)
-- 
2.43.0