[PATCH 0/2] Fix broken cpuset affinity handling on heterogeneous systems

[Will Deacon <will-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>]

Hi folks,

These two patches fix a couple of CPU affinity issues involving cpusets
on heterogeneous systems. A concrete example of this is running 32-bit
tasks on recent arm64 SoCs, where some of the cores are only capable of
64-bit execution.

The first patch (from Peter) fixes a regression introduced during the
recent merge window which is causing test failures in Android where the
problematic patches have been backported. The second patch fixes a
longer-standing issue, which I noticed while testing fixes for the
initial regression.

Ideally, both of these would land together, but fixing the regression
for 6.2 is my main concern.

Anyway, I don't think either Peter or I would call ourselves cpuset
experts (far from it!), so please have a look.

Cheers,

Will

Cc: Peter Zijlstra <peterz-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>
Cc: Waiman Long <longman-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
Cc: Zefan Li <lizefan.x-EC8Uxl6Npydl57MIdRCFDg@public.gmane.org>
Cc: Tejun Heo <tj-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
Cc: Johannes Weiner <hannes-druUgvl0LCNAfugRpC6u6w@public.gmane.org>
Cc: cgroups-u79uwXL29TY76Z2rM5mHXA@public.gmane.org

--->8

Peter Zijlstra (1):
  cpuset: Fix cpuset_cpus_allowed() to not filter offline CPUs

Will Deacon (1):
  cpuset: Call set_cpus_allowed_ptr() with appropriate mask for task

 kernel/cgroup/cpuset.c | 57 +++++++++++++++++++++++++++++++++---------
 1 file changed, 45 insertions(+), 12 deletions(-)

-- 
2.39.1.456.gfc5497dd1b-goog