linux-kernel - Re: [PATCH v2 2/2] mm/pageblock: remove false sharing in pageblock

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <11d6c4ae-6eae-aef9-5652-73a8680cd6e4@suse.cz>
Date:   Tue, 1 Sep 2020 19:04:06 +0200
From:   Vlastimil Babka <vbabka@...e.cz>
To:     Alex Shi <alex.shi@...ux.alibaba.com>,
        Anshuman Khandual <anshuman.khandual@....com>,
        Matthew Wilcox <willy@...radead.org>,
        David Hildenbrand <david@...hat.com>
Cc:     Andrew Morton <akpm@...ux-foundation.org>,
        Hugh Dickins <hughd@...gle.com>,
        Alexander Duyck <alexander.h.duyck@...ux.intel.com>,
        linux-kernel@...r.kernel.org, linux-mm@...ck.org,
        Mel Gorman <mgorman@...hsingularity.net>
Subject: Re: [PATCH v2 2/2] mm/pageblock: remove false sharing in
 pageblock_flags

On 8/19/20 10:09 AM, Alex Shi wrote:
> 
> 
> 在 2020/8/19 下午3:57, Anshuman Khandual 写道:
>> 
>> 
>> On 08/19/2020 11:17 AM, Alex Shi wrote:
>>> Current pageblock_flags is only 4 bits, so it has to share a char size
>>> in cmpxchg when get set, the false sharing cause perf drop.
>>>
>>> If we incrase the bits up to 8, false sharing would gone in cmpxchg. and
>>> the only cost is half char per pageblock, which is half char per 128MB
>>> on x86, 4 chars in 1 GB.
>> 
>> Agreed that increase in memory utilization is negligible here but does
>> this really improve performance ?
>> 
> 
> It's no doubt in theory. and it would had a bad impact according to 
> commit e380bebe4771548  mm, compaction: keep migration source private to a single 

I don't think that commit is doing the test_and_set_skip() under lock to avoid
false sharing. I think it's done to simply make the test and set protected
against races without relying on e.g. a truly atomic test_and_set_bit(). It's
still noted that it's just a hint so it's not protected to others calling
set_pageblock_skip() from other contexts not under a lock.

> but I do have some problem in running thpscale/mmtest. I'd like to see if anyone
> could give a try.
> 
> BTW, I naturally hate the false sharing even it's in theory. Anyone who doesn't? :)
> 
> Thanks
> Alex 
>