| Message ID | 20181011192930.2844-1-josef@toxicpanda.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | blk-wbt: wake up all when we scale up, not down | expand |
On 10/11/18 1:29 PM, Josef Bacik wrote: > Tetsuo brought to my attention that I screwed up the scale_up/scale_down > helpers when I factored out the rq-qos code. We need to wake up all the > waiters when we add slots for requests to make, not when we shrink the > slots. Otherwise we'll end up things waiting forever. This was a > mistake and simply puts everything back the way it was. I'll apply this for 4.19. I haven't hit this in testing, I'm guessing the jitter when going up and down hides the issue.
diff --git a/block/blk-wbt.c b/block/blk-wbt.c index 8e20a0677dcf..8ac93fcbaa2e 100644 --- a/block/blk-wbt.c +++ b/block/blk-wbt.c @@ -310,6 +310,7 @@ static void scale_up(struct rq_wb *rwb) rq_depth_scale_up(&rwb->rq_depth); calc_wb_limits(rwb); rwb->unknown_cnt = 0; + rwb_wake_all(rwb); rwb_trace_step(rwb, "scale up"); } @@ -318,7 +319,6 @@ static void scale_down(struct rq_wb *rwb, bool hard_throttle) rq_depth_scale_down(&rwb->rq_depth, hard_throttle); calc_wb_limits(rwb); rwb->unknown_cnt = 0; - rwb_wake_all(rwb); rwb_trace_step(rwb, "scale down"); }
Tetsuo brought to my attention that I screwed up the scale_up/scale_down helpers when I factored out the rq-qos code. We need to wake up all the waiters when we add slots for requests to make, not when we shrink the slots. Otherwise we'll end up things waiting forever. This was a mistake and simply puts everything back the way it was. cc: stable@vger.kernel.org Fixes: a79050434b45 ("blk-rq-qos: refactor out common elements of blk-wbt") eported-by: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp> Signed-off-by: Josef Bacik <josef@toxicpanda.com> --- JENS! I did this on for-4.20/block FYI block/blk-wbt.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)