[PATCH 2/2] block: Protect elevator from too early release

[Jarkko Lavinen]

When a disk is being removed, blk_put_queue calls elevator exit function
which will release the elevator while the elevator may still be in use.

Signed-off-by: Jarkko Lavinen <jarkko.lavinen@xxxxxxxxx>
---
 block/blk-core.c |    9 +++++++++
 1 files changed, 9 insertions(+), 0 deletions(-)

diff --git a/block/blk-core.c b/block/blk-core.c
index ac0fa10..da56586 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -1173,6 +1173,13 @@ static int __make_request(struct request_queue *q, struct bio *bio)
 	 */
 	blk_queue_bounce(q, &bio);
 
+	mutex_lock(&q->elevator->sysfs_lock);
+	if (test_bit(QUEUE_FLAG_DEAD, &q->queue_flags)) {
+		mutex_unlock(&q->elevator->sysfs_lock);
+		bio_endio(bio, -EIO);
+		return 0;
+	}
+
 	spin_lock_irq(q->queue_lock);
 
 	if (unlikely(bio_rw_flagged(bio, BIO_RW_BARRIER)) || elv_queue_empty(q))
@@ -1275,6 +1282,8 @@ out:
 	if (unplug || !queue_should_plug(q))
 		__generic_unplug_device(q);
 	spin_unlock_irq(q->queue_lock);
+	mutex_unlock(&q->elevator->sysfs_lock);
+
 	return 0;
 }
 
-- 
1.6.3.3

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/