Re: [PATCH] sched/numa: Avoid trapping faults and attempting migration of file-backed dirty pages

From: Rik van Riel
Date: Mon Mar 26 2018 - 17:59:31 EST


On Mon, 2018-03-26 at 10:43 +0100, Mel Gorman wrote:
> change_pte_range is called from task work context to mark PTEs for
> receiving
> NUMA faulting hints. If the marked pages are dirty then migration may
> fail.
> Some filesystems cannot migrate dirty pages without blocking so are
> skipped
> in MIGRATE_ASYNC mode which just wastes CPU. Even when they can, it
> can
> be a waste of cycles when the pages are shared forcing higher scan
> rates.

> Signed-off-by: Mel Gorman <mgorman@xxxxxxxxxxxxxxxxxxx>

Reviewed-by: Rik van Riel <riel@xxxxxxxxxxx>

--
All Rights Reversed.

Attachment: signature.asc
Description: This is a digitally signed message part