summaryrefslogtreecommitdiff
path: root/fs/ioprio.c
diff options
context:
space:
mode:
authorOleg Nesterov <oleg@tv-sign.ru>2006-08-21 08:34:15 +0200
committerJens Axboe <axboe@suse.de>2006-08-21 08:34:15 +0200
commit9f83e45eb54fc7198dc59fc63255341851ba4c48 (patch)
tree3874581e08052b7f37f164110b3bd44ae5eff291 /fs/ioprio.c
parent78bd4d484f81a611ef6ff02f909e576cb9aac7f2 (diff)
downloadlwn-9f83e45eb54fc7198dc59fc63255341851ba4c48.tar.gz
lwn-9f83e45eb54fc7198dc59fc63255341851ba4c48.zip
[PATCH] Fix current_io_context() vs set_task_ioprio() race
I know nothing about io scheduler, but I suspect set_task_ioprio() is not safe. current_io_context() initializes "struct io_context", then sets ->io_context. set_task_ioprio() running on another cpu may see the changes out of order, so ->set_ioprio(ioc) may use io_context which was not initialized properly. Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru> Signed-off-by: Jens Axboe <axboe@suse.de>
Diffstat (limited to 'fs/ioprio.c')
-rw-r--r--fs/ioprio.c3
1 files changed, 3 insertions, 0 deletions
diff --git a/fs/ioprio.c b/fs/ioprio.c
index 3db31038e9ab..06578311c63f 100644
--- a/fs/ioprio.c
+++ b/fs/ioprio.c
@@ -44,6 +44,9 @@ static int set_task_ioprio(struct task_struct *task, int ioprio)
task->ioprio = ioprio;
ioc = task->io_context;
+ /* see wmb() in current_io_context() */
+ smp_read_barrier_depends();
+
if (ioc && ioc->set_ioprio)
ioc->set_ioprio(ioc, ioprio);