linux-kernel - Re: [mm/swap] aae466b005: vm-scalability.throughput -2.7% regression

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] [day] [month] [year] [list]

Message-ID: <20210104012356.GA5941@js1304-desktop>
Date:   Mon, 4 Jan 2021 10:24:10 +0900
From:   Joonsoo Kim <js1304@...il.com>
To:     kernel test robot <oliver.sang@...el.com>
Cc:     Linus Torvalds <torvalds@...ux-foundation.org>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Johannes Weiner <hannes@...xchg.org>,
        Vlastimil Babka <vbabka@...e.cz>,
        Hugh Dickins <hughd@...gle.com>,
        Matthew Wilcox <willy@...radead.org>,
        Mel Gorman <mgorman@...hsingularity.net>,
        Michal Hocko <mhocko@...nel.org>,
        Minchan Kim <minchan@...nel.org>,
        LKML <linux-kernel@...r.kernel.org>, lkp@...ts.01.org,
        lkp@...el.com, ying.huang@...el.com, feng.tang@...el.com,
        zhengjun.xing@...el.com
Subject: Re: [mm/swap]  aae466b005:  vm-scalability.throughput -2.7%
 regression

On Wed, Dec 30, 2020 at 02:24:12PM +0800, kernel test robot wrote:
> 
> Greeting,
> 
> FYI, we noticed a -2.7% regression of vm-scalability.throughput due to commit:
> 
> 
> commit: aae466b0052e1888edd1d7f473d4310d64936196 ("mm/swap: implement workingset detection for anonymous LRU")
> https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master

Hello,

Thanks for reporting.

This is a known issue to me. As you can see below, slight regression
is found on only the random access workload. For pure random pattern,
detecting workingset would not guarantee a good result since these
pages doesn't accessed in the near future. Even worse, it causes small
overhead to the system since we need to demote these activated page to
the inactive list to reclaim. So, below result is natural to me.

For other access pattern, this patchset would result in positive
effect.

Thanks.