Re: [patch] mm, compaction: periodically schedule when freeing pages

[Mel Gorman]

On Wed, Sep 04, 2013 at 04:25:59PM -0700, David Rientjes wrote:
> We've been getting warnings about an excessive amount of time spent
> allocating pages for migration during memory compaction without
> scheduling.  isolate_freepages_block() already periodically checks for
> contended locks or the need to schedule, but isolate_freepages() never
> does.
> 
> When a zone is massively long and no suitable targets can be found, this
> iteration can be quite expensive without ever doing cond_resched().
> 
> Check periodically for the need to reschedule while the compaction free
> scanner iterates.
> 
> Signed-off-by: David Rientjes <rientjes@xxxxxxxxxx>

Ok, fair enough.

Acked-by: Mel Gorman <mgorman@xxxxxxx>

However I'm curious. Do you know why the combined use of
compact_cached_free_pfn and pageblock skip bits is not enough for the scanner
to quickly find a pageblock that is suitable for isolate_freepages_block()?
Is the pageblock skip information getting cleared frequently by kswapd
or something?

-- 
Mel Gorman
SUSE Labs
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/