Re: [PATCH v3 09/14] cpuset: Don't use the cpu_possible_mask as a last resort for cgroup v1

[Will Deacon]

On Thu, Nov 19, 2020 at 09:29:22AM +0000, Quentin Perret wrote:
> On Friday 13 Nov 2020 at 09:37:14 (+0000), Will Deacon wrote:
> > If the scheduler cannot find an allowed CPU for a task,
> > cpuset_cpus_allowed_fallback() will widen the affinity to cpu_possible_mask
> > if cgroup v1 is in use.
> > 
> > In preparation for allowing architectures to provide their own fallback
> > mask, just return early if we're not using cgroup v2 and allow
> > select_fallback_rq() to figure out the mask by itself.
> > 
> > Cc: Li Zefan <lizefan@xxxxxxxxxx>
> > Cc: Tejun Heo <tj@xxxxxxxxxx>
> > Cc: Johannes Weiner <hannes@xxxxxxxxxxx>
> > Signed-off-by: Will Deacon <will@xxxxxxxxxx>
> 
> That makes select_fallback_rq() slightly more expensive if you're using
> cgroup v1, but I don't expect that be really measurable in real-world
> workloads, so:
> 
> Reviewed-by: Quentin Perret <qperret@xxxxxxxxxx>

Cheers!

Will