[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <87sfbubg3j.fsf@yhuang6-desk2.ccr.corp.intel.com>
Date: Thu, 18 May 2023 09:09:20 +0800
From: "Huang, Ying" <ying.huang@...el.com>
To: David Hildenbrand <david@...hat.com>
Cc: Khalid Aziz <khalid.aziz@...cle.com>, akpm@...ux-foundation.org,
willy@...radead.org, steven.sistare@...cle.com,
mgorman@...hsingularity.net, khalid@...nel.org, linux-mm@...ck.org,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH v3] mm, compaction: Skip all non-migratable pages during
scan
David Hildenbrand <david@...hat.com> writes:
> On 17.05.23 18:15, Khalid Aziz wrote:
>> Pages pinned in memory through extra refcounts can not be migrated.
>> Currently as isolate_migratepages_block() scans pages for
>> compaction, it skips any pinned anonymous pages. All non-migratable
>> pages should be skipped and not just the anonymous pinned pages.
>> This patch adds a check for extra refcounts on a page to determine
>> if the page can be migrated. This was seen as a real issue on a
>> customer workload where a large number of pages were pinned by vfio
>> on the host and any attempts to allocate hugepages resulted in
>> significant amount of cpu time spent in either direct compaction or
>> in kcompactd scanning vfio pinned pages over and over again that can
>> not be migrated.
>
> How will this change affect alloc_contig_range(), such as used for CMA
> allocations or virtio-mem? alloc_contig_range() ends up calling
> isolate_migratepages_range() -> isolate_migratepages_block().
IIUC, cc->alloc_contig can be used to distinguish contiguous allocation
and compaction. And, from the original commit which introduced
anonymous pages skipping (commit 119d6d59dcc0 ("mm, compaction: avoid
isolating pinned pages ")) and this patch, large number of migration
failure during compaction causes real issue too. So, I suggest to use
cc->alloc_contig here.
> We don't want to fail early in case there is a short-term pin that
> might go away any moment after we isolated ... that will make the
> situation worse for these use cases, especially if MIGRATE_CMA or
> ZONE_MOVABLE is involved.
Best Regards,
Huang, Ying
Powered by blists - more mailing lists