cfq: set workload as expired if it doesn't have any slice left

When a group is resumed, if it doesn't have workload slice left,
we should set workload_expires as expired. Otherwise, we might
start from where we left in previous group by error.
Thanks the idea from Corrado.

Signed-off-by: Gui Jianfeng <guijianfeng@cn.fujitsu.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>

diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c
index f3f6239..e2f8046 100644 (file)
--- a/block/cfq-iosched.c
+++ b/block/cfq-iosched.c
@@ -2113,7 +2113,9 @@ static void cfq_choose_cfqg(struct cfq_data *cfqd)
                cfqd->workload_expires = jiffies + cfqg->saved_workload_slice;
                cfqd->serving_type = cfqg->saved_workload;
                cfqd->serving_prio = cfqg->saved_serving_prio;
-       }
+       } else
+               cfqd->workload_expires = jiffies - 1;
+
        choose_service_tree(cfqd, cfqg);
 }