Re: [PATCH v5 07/15] cpuset: Don't use the cpu_possible_mask as a last resort for cgroup v1

[Will Deacon]

On Thu, Dec 17, 2020 at 02:44:01PM +0100, Peter Zijlstra wrote:
> On Thu, Dec 17, 2020 at 12:15:52PM +0000, Qais Yousef wrote:
> > On 12/08/20 13:28, 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>
> > > Reviewed-by: Quentin Perret <qperret@xxxxxxxxxx>
> > > Signed-off-by: Will Deacon <will@xxxxxxxxxx>
> > > ---
> > >  kernel/cgroup/cpuset.c | 6 ++++--
> > >  1 file changed, 4 insertions(+), 2 deletions(-)
> > > 
> > > diff --git a/kernel/cgroup/cpuset.c b/kernel/cgroup/cpuset.c
> > > index 57b5b5d0a5fd..e970737c3ed2 100644
> > > --- a/kernel/cgroup/cpuset.c
> > > +++ b/kernel/cgroup/cpuset.c
> > > @@ -3299,9 +3299,11 @@ void cpuset_cpus_allowed(struct task_struct *tsk, struct cpumask *pmask)
> > >  
> > >  void cpuset_cpus_allowed_fallback(struct task_struct *tsk)
> > >  {
> > > +	if (!is_in_v2_mode())
> > > +		return; /* select_fallback_rq will try harder */
> > > +
> > >  	rcu_read_lock();
> > > -	do_set_cpus_allowed(tsk, is_in_v2_mode() ?
> > > -		task_cs(tsk)->cpus_allowed : cpu_possible_mask);
> > > +	do_set_cpus_allowed(tsk, task_cs(tsk)->cpus_allowed);
> > 
> > Why is it safe to return that for cpuset v2?
> 
> v1
> 
> Because in that case it does cpu_possible_mask, which, if you look at
> select_fallback_rq(), is exactly what happens when cpuset 'fails' to
> find a candidate.
> 
> Or at least, that's how I read the patch.

I think Qais a point with v2 though: if task_cs(tsk)->cpus_allowed
contains 64-bit-only CPUs, then we're in trouble here. I should be
taking the intersection with the task_cpu_possible_mask() for the task.

Will