On Fri, Nov 20, 2015 at 11:06:46AM +0100, Vlastimil Babka wrote:
On 11/20/2015 10:33 AM, Aaron Lu wrote:
>On 11/20/2015 04:55 PM, Aaron Lu wrote:
>>On 11/19/2015 09:29 PM, Vlastimil Babka wrote:
>>>+CC Andrea, David, Joonsoo
>>>On 11/19/2015 10:29 AM, Aaron Lu wrote:
>>>>The vmstat and perf-profile are also attached, please let me know if you
>>>>need any more information, thanks.
>>>Output from vmstat (the tool) isn't much useful here, a periodic
>>>/proc/vmstat" would be much better.
>>>The perf profiles are somewhat weirdly sorted by children cost (?), but
>>>I noticed a very high cost (46%) in pageblock_pfn_to_page(). This could
>>>be due to a very large but sparsely populated zone. Could you provide
>>Is a one time /proc/zoneinfo enough or also a periodic one?
>Please see attached, note that this is a new run so the perf profile is
>a little different.
DMA32 is a bit sparse:
Node 0, zone DMA32
pages free 62829
Since the other zones are much larger, probably this is not the
culprit. But tracepoints should tell us more. I have a theory that
updating free scanner's cached pfn doesn't happen if it aborts due
to need_resched() during isolate_freepages(), before hitting a valid
pageblock, if the zone has a large hole in it. But zoneinfo doesn't
Today, I revisit this issue and yes, I think that your theory is
right. isolate_freepages() will not update cached pfn until call
isolate_freepages_block(). So, if there are many holes or many
unmovable pageblocks or !isolation_suitable() pageblocks, cached pfn
will not updated if compaction aborts due to need_resched(). zoneinfo
shows that there is not much holes so I guess that this problem is caused
by latter two cases.
It is better to update cached pfn in these cases. Although I don't see
your solution yet, I guess it will help here.