linux-kernel - Re: [PATCH REBASED 1/4] mm: Move recent_rotated pages calculation to shrink_inactive

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]

Date:   Tue, 28 May 2019 19:07:40 +0300
From:   Kirill Tkhai <ktkhai@...tuozzo.com>
To:     Johannes Weiner <hannes@...xchg.org>
Cc:     akpm@...ux-foundation.org, daniel.m.jordan@...cle.com,
        mhocko@...e.com, linux-mm@...ck.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH REBASED 1/4] mm: Move recent_rotated pages calculation to
 shrink_inactive_list()

On 28.05.2019 18:51, Johannes Weiner wrote:
> On Mon, Mar 18, 2019 at 12:27:59PM +0300, Kirill Tkhai wrote:
>> @@ -1945,6 +1942,8 @@ shrink_inactive_list(unsigned long nr_to_scan, struct lruvec *lruvec,
>>  		count_memcg_events(lruvec_memcg(lruvec), PGSTEAL_DIRECT,
>>  				   nr_reclaimed);
>>  	}
>> +	reclaim_stat->recent_rotated[0] = stat.nr_activate[0];
>> +	reclaim_stat->recent_rotated[1] = stat.nr_activate[1];
> 
> Surely this should be +=, right?
> 
> Otherwise we maintain essentially no history of page rotations and
> that wreaks havoc on the page cache vs. swapping reclaim balance.

Sure, thanks.

Kirill