summaryrefslogtreecommitdiff
path: root/block/blk-cgroup.c
diff options
context:
space:
mode:
authorTejun Heo <tj@kernel.org>2011-12-14 00:33:38 +0100
committerJens Axboe <axboe@kernel.dk>2011-12-14 00:33:38 +0100
commitdc86900e0a8f665122de6faadd27fb4c6d2b3e4d (patch)
treeec24aa4b076f54fcfb0558d5113f5c5e0f4bc173 /block/blk-cgroup.c
parent283287a52e3c3f7f8f9da747f4b8c5202740d776 (diff)
block, cfq: move ioc ioprio/cgroup changed handling to cic
ioprio/cgroup change was handled by marking the changed state in ioc and, on the following access to the ioc, performing RCU-protected iteration through all cic's grabbing the matching queue_lock. This patch moves the changed state to each cic. When ioprio or cgroup changes, the respective bit is set on all cic's of the ioc and when each of those cic (not ioc) is accessed, change is applied for that specific ioc-queue pair. This also fixes the following two race conditions between setting and clearing of changed states. * Missing barrier between assign/load of ioprio and ioprio_changed allowed applying old ioprio. * Change requests could happen between application of change and clearing of changed variables. Signed-off-by: Tejun Heo <tj@kernel.org> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Diffstat (limited to 'block/blk-cgroup.c')
-rw-r--r--block/blk-cgroup.c2
1 files changed, 1 insertions, 1 deletions
diff --git a/block/blk-cgroup.c b/block/blk-cgroup.c
index 4b001dcd85b..dc00835aab6 100644
--- a/block/blk-cgroup.c
+++ b/block/blk-cgroup.c
@@ -1648,7 +1648,7 @@ static void blkiocg_attach_task(struct cgroup *cgrp, struct task_struct *tsk)
/* we don't lose anything even if ioc allocation fails */
ioc = get_task_io_context(tsk, GFP_ATOMIC, NUMA_NO_NODE);
if (ioc) {
- ioc->cgroup_changed = 1;
+ ioc_cgroup_changed(ioc);
put_io_context(ioc);
}
}