Messages in this thread | | | Subject | Re: [PATCH block/for-linus] iocost: don't let vrate run wild while there's no saturation signal | From | Jens Axboe <> | Date | Thu, 14 May 2020 09:34:17 -0600 |
| |
On 5/14/20 8:51 AM, Tejun Heo wrote: > On Mon, Oct 14, 2019 at 05:18:11PM -0700, Tejun Heo wrote: >> When the QoS targets are met and nothing is being throttled, there's >> no way to tell how saturated the underlying device is - it could be >> almost entirely idle, at the cusp of saturation or anywhere inbetween. >> Given that there's no information, it's best to keep vrate as-is in >> this state. Before 7cd806a9a953 ("iocost: improve nr_lagging >> handling"), this was the case - if the device isn't missing QoS >> targets and nothing is being throttled, busy_level was reset to zero. > ... >> Signed-off-by: Tejun Heo <tj@kernel.org> >> Reported-by: Andy Newell <newella@fb.com> >> Fixes: 7cd806a9a953 ("iocost: improve nr_lagging handling") > > Jens, this one fell through the cracks. It still applies with only a small > offset. Can you please apply?
Looks like it did, queued up for 5.8 now.
-- Jens Axboe
| |