[34-longterm 089/179] cfq-iosched: fix a rcu warning

[Paul Gortmaker]

From: Shaohua Li <shaohua.li@xxxxxxxxx>

                   -------------------
    This is a commit scheduled for the next v2.6.34 longterm release.
    http://git.kernel.org/?p=linux/kernel/git/paulg/longterm-queue-2.6.34.git
    If you see a problem with using this for longterm, please comment.
                   -------------------

commit 3181faa85bda3dc3f5e630a1846526c9caaa38e3 upstream.

I got a rcu warnning at boot. the ioc->ioc_data is rcu_deferenced, but
doesn't hold rcu_read_lock.

Signed-off-by: Shaohua Li <shaohua.li@xxxxxxxxx>
Signed-off-by: Jens Axboe <jaxboe@xxxxxxxxxxxx>
Signed-off-by: Paul Gortmaker <paul.gortmaker@xxxxxxxxxxxxx>
---
 block/cfq-iosched.c |    5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c
index 510552a..d39a07a 100644
--- a/block/cfq-iosched.c
+++ b/block/cfq-iosched.c
@@ -2559,11 +2559,14 @@ static void __cfq_exit_single_io_context(struct cfq_data *cfqd,
 	cic->dead_key = (unsigned long) cic->key;
 	cic->key = NULL;
 
+	rcu_read_lock();
 	if (rcu_dereference(ioc->ioc_data) == cic) {
+		rcu_read_unlock();
 		spin_lock(&ioc->lock);
 		rcu_assign_pointer(ioc->ioc_data, NULL);
 		spin_unlock(&ioc->lock);
-	}
+	} else
+		rcu_read_unlock();
 
 	if (cic->cfqq[BLK_RW_ASYNC]) {
 		cfq_exit_cfqq(cfqd, cic->cfqq[BLK_RW_ASYNC]);
-- 
1.7.9.6

--
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/