summaryrefslogtreecommitdiff
path: root/block
diff options
context:
space:
mode:
authorPaolo Valente <paolo.valente@linaro.org>2021-06-19 17:09:48 +0300
committerJens Axboe <axboe@kernel.dk>2021-06-22 00:03:41 +0300
commit9a2ac41b13c573703d6689f51f3e27dd658324be (patch)
tree9b1815feafb49eb8c3e1404ae8bfc88b880208e3 /block
parentefc72524b3a9e4e7bc7c07f756528736409ec1b7 (diff)
downloadlinux-9a2ac41b13c573703d6689f51f3e27dd658324be.tar.xz
block, bfq: reset waker pointer with shared queues
Commit 85686d0dc194 ("block, bfq: keep shared queues out of the waker mechanism") leaves shared bfq_queues out of the waker-detection mechanism. It attains this goal by not updating the pointer last_completed_rq_bfqq, if the last request completed belongs to a shared bfq_queue (so that the pointer will not point to the shared bfq_queue). Yet this has a side effect: the pointer last_completed_rq_bfqq keeps pointing, deceptively, to a bfq_queue that actually is not the last one to have had a request completed. As a consequence, such a bfq_queue may deceptively be considered as a waker of some bfq_queue, even of some shared bfq_queue. To address this issue, reset last_completed_rq_bfqq if the last request completed belongs to a shared queue. Fixes: 85686d0dc194 ("block, bfq: keep shared queues out of the waker mechanism") Signed-off-by: Paolo Valente <paolo.valente@linaro.org> Link: https://lore.kernel.org/r/20210619140948.98712-8-paolo.valente@linaro.org Signed-off-by: Jens Axboe <axboe@kernel.dk>
Diffstat (limited to 'block')
-rw-r--r--block/bfq-iosched.c6
1 files changed, 4 insertions, 2 deletions
diff --git a/block/bfq-iosched.c b/block/bfq-iosched.c
index a273b2bcea2a..fedb0a8fd388 100644
--- a/block/bfq-iosched.c
+++ b/block/bfq-iosched.c
@@ -6165,11 +6165,13 @@ static void bfq_completed_request(struct bfq_queue *bfqq, struct bfq_data *bfqd)
* of other queues. But a false waker will unjustly steal
* bandwidth to its supposedly woken queue. So considering
* also shared queues in the waking mechanism may cause more
- * control troubles than throughput benefits. Then do not set
- * last_completed_rq_bfqq to bfqq if bfqq is a shared queue.
+ * control troubles than throughput benefits. Then reset
+ * last_completed_rq_bfqq if bfqq is a shared queue.
*/
if (!bfq_bfqq_coop(bfqq))
bfqd->last_completed_rq_bfqq = bfqq;
+ else
+ bfqd->last_completed_rq_bfqq = NULL;
/*
* If we are waiting to discover whether the request pattern