Messages in this thread | | | Date | Thu, 3 Nov 2022 11:11:24 +0100 | From | Michal Koutný <> | Subject | Re: [PATCH v2] sched: async unthrottling for cfs bandwidth |
| |
On Wed, Nov 02, 2022 at 05:10:08PM -0700, Josh Don <joshdon@google.com> wrote: > Without my patch, distribute_cfs_runtime() will unthrottle the > cfs_rq's, and as you point out, it doesn't actually give them any real > quota, it lets assign_cfs_rq_runtime() take care of that. But this > happens asynchronously on those cpus. If they are idle, they wait for > an IPI from the resched_curr() in unthrottled_cfs_rq(), otherwise they > simply wait until potentially the next rescheduling point. So we are > currently far from ever being guaranteed that the order the cpus pull > actual quota via assign_cfs_rq_runtime() matches the order they were > unthrottled from the list.
Thanks for breaking it down, I agree your change won't affect the fairness substantially and my other points are clarified too.
Michal
[unhandled content-type:application/pgp-signature] | |