From: Jeff Moyer <jmoyer@redhat.com>
Date: Wed, 7 Jul 2010 13:48:30 -0400
Subject: [block] cfq-iosched: fix bad locking in changed_ioprio
Message-id: <1278510510-27466-3-git-send-email-jmoyer@redhat.com>
Patchwork-id: 26741
O-Subject: [RHEL5 PATCH 2/2][v2] CFQ: bad locking in changed_ioprio()
Bugzilla: 582435
RH-Acked-by: Vivek Goyal <vgoyal@redhat.com>

Hi,

The previous fix included an improper assumption about irq state in the
changed_ioprio function.  This upstream patch fixes it.

  commit c1b707d253fe918b92882cff1dbd926b47e14fd2
  Author: Jens Axboe <jens.axboe@oracle.com>
  Date:   Mon Oct 30 19:54:23 2006 +0100

    [PATCH] CFQ: bad locking in changed_ioprio()

    When the ioprio code recently got juggled a bit, a bug was introduced.
    changed_ioprio() is no longer called with interrupts disabled, so using
    plain spin_lock() on the queue_lock is a bug.

    Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
    Signed-off-by: Linus Torvalds <torvalds@osdl.org>

This is related to bug 582435.  Comments, as always, are welcome.

Cheers,
Jeff
Signed-off-by: Jeff Moyer <jmoyer@redhat.com>

diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c
index 7f609df..822e8cf 100644
--- a/block/cfq-iosched.c
+++ b/block/cfq-iosched.c
@@ -1660,11 +1660,12 @@ static inline void changed_ioprio(struct cfq_io_context *cic)
 {
 	struct cfq_data *cfqd = cic->key;
 	struct cfq_queue *cfqq;
+	unsigned long flags;
 
 	if (unlikely(!cfqd))
 		return;
 
-	spin_lock(cfqd->queue->queue_lock);
+	spin_lock_irqsave(cfqd->queue->queue_lock, flags);
 
 	cfqq = cic->cfqq[ASYNC];
 	if (cfqq) {
@@ -1681,7 +1682,7 @@ static inline void changed_ioprio(struct cfq_io_context *cic)
 	if (cfqq)
 		cfq_mark_cfqq_prio_changed(cfqq);
 
-	spin_unlock(cfqd->queue->queue_lock);
+	spin_unlock_irqrestore(cfqd->queue->queue_lock, flags);
 }
 
 static void cfq_ioc_set_ioprio(struct io_context *ioc)