Messages in this thread | | | From | Xuewen Yan <> | Date | Wed, 20 Jul 2022 15:45:27 +0800 | Subject | Re: [PATCH] sched/schedutil: Fix deadlock between cpuset and cpu hotplug when using schedutil |
| |
Dear all
On Wed, Jul 13, 2022 at 11:20 AM Tejun Heo <tj@kernel.org> wrote: > > On Tue, Jul 12, 2022 at 10:49:57PM -0400, Waiman Long wrote: > > > Well, the only thing I can think of is always grabbing cpus_read_lock() > > > before grabbing threadgroup_rwsem. Waiman, what do you think? > > > > That is a possible solution as cpus_read_lock() is rather lightweight. It is > > a good practice to acquire it first. > > On a similar note, I think we probably should re-enable percpu operations on > threadgroup_rwsem too by default and allow users who are affected by slower > write path operations to opt-in. After the new CLONE_INTO_CGROUP which > doesn't need the rwsem, we have far fewer write lockers after all. >
If there's any patch for me to try? I would be very grateful.
Thanks!
--- xw.yan
| |