Messages in this thread | | | Date | Wed, 9 May 2018 17:34:10 +0100 | From | Mel Gorman <> | Subject | Re: [tip:sched/core] sched/numa: Delay retrying placement for automatic NUMA balance after wake_affine() |
| |
On Wed, May 09, 2018 at 03:58:14AM -0700, Srikar Dronamraju wrote: > In the other scenario, where wake_affine set p->numa_migrate_preferred to a > bigger value, the task calls numa_migrate_preferred(), > numa_migrate_preferred could be before p->numa_migrate_preferred. In > such a case, we should have stopped the task from migration. > However we overwrite p->numa_migrate_preferred and do the > task_numa_migrate(). Somehow this doesn't seem to achieve what the > commit intended. > > Or did I misunderstand? >
Nope, the logic is broken. While it can be "fixed", the end result adds complexity for very little gain. I believe the right path for the moment is a revert and retry at a future date.
Thanks.
-- Mel Gorman SUSE Labs
| |