Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
From:   Asahi Lina <lina@asahilina.net>
Date:   Thu, 06 Apr 2023 01:37:39 +0900
Subject: [PATCH] drm/scheduler: Fix UAF race in drm_sched_entity_push_job()
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
Message-Id: <20230406-scheduler-uaf-2-v1-1-972531cf0a81@asahilina.net>
To:     Luben Tuikov <luben.tuikov@amd.com>,
        David Airlie <airlied@gmail.com>,
        Daniel Vetter <daniel@ffwll.ch>
Cc:     dri-devel@lists.freedesktop.org, linux-kernel@vger.kernel.org,
        asahi@lists.linux.dev, Asahi Lina <lina@asahilina.net>
Precedence: bulk

After a job is pushed into the queue, it is owned by the scheduler core
and may be freed at any time, so we can't write nor read the submit
timestamp after that point.

Fixes oopses observed with the drm/asahi driver, found with kASAN.

Signed-off-by: Asahi Lina <lina@asahilina.net>
---
 drivers/gpu/drm/scheduler/sched_entity.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/scheduler/sched_entity.c b/drivers/gpu/drm/scheduler/sched_entity.c
index 15d04a0ec623..e0a8890a62e2 100644
--- a/drivers/gpu/drm/scheduler/sched_entity.c
+++ b/drivers/gpu/drm/scheduler/sched_entity.c
@@ -507,12 +507,19 @@ void drm_sched_entity_push_job(struct drm_sched_job *sched_job)
 {
 	struct drm_sched_entity *entity = sched_job->entity;
 	bool first;
+	ktime_t submit_ts;
 
 	trace_drm_sched_job(sched_job, entity);
 	atomic_inc(entity->rq->sched->score);
 	WRITE_ONCE(entity->last_user, current->group_leader);
+
+	/*
+	 * After the sched_job is pushed into the entity queue, it may be
+	 * completed and freed up at any time. We can no longer access it.
+	 * Make sure to set the submit_ts first, to avoid a race.
+	 */
+	sched_job->submit_ts = submit_ts = ktime_get();
 	first = spsc_queue_push(&entity->job_queue, &sched_job->queue_node);
-	sched_job->submit_ts = ktime_get();
 
 	/* first job wakes up scheduler */
 	if (first) {
@@ -529,7 +536,7 @@ void drm_sched_entity_push_job(struct drm_sched_job *sched_job)
 		spin_unlock(&entity->rq_lock);
 
 		if (drm_sched_policy == DRM_SCHED_POLICY_FIFO)
-			drm_sched_rq_update_fifo(entity, sched_job->submit_ts);
+			drm_sched_rq_update_fifo(entity, submit_ts);
 
 		drm_sched_wakeup(entity->rq->sched);
 	}

---
base-commit: fe15c26ee26efa11741a7b632e9f23b01aca4cc6
change-id: 20230406-scheduler-uaf-2-44cf8faed245

Thank you,
~~ Lina