linux-kernel - Re: [PATCH 2/2] mm, vmstat: reduce zone->lock holding time by /proc/pagetypeinfo

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]

Date:   Fri, 25 Oct 2019 13:43:50 +0200
From:   Michal Hocko <mhocko@...nel.org>
To:     Qian Cai <cai@....pw>
Cc:     Andrew Morton <akpm@...ux-foundation.org>,
        Mel Gorman <mgorman@...e.de>, Waiman Long <longman@...hat.com>,
        Johannes Weiner <hannes@...xchg.org>,
        Roman Gushchin <guro@...com>, Vlastimil Babka <vbabka@...e.cz>,
        Konstantin Khlebnikov <khlebnikov@...dex-team.ru>,
        Jann Horn <jannh@...gle.com>, Song Liu <songliubraving@...com>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        Rafael Aquini <aquini@...hat.com>, linux-mm@...ck.org,
        LKML <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 2/2] mm, vmstat: reduce zone->lock holding time by
 /proc/pagetypeinfo

On Fri 25-10-19 07:18:37, Qian Cai wrote:
> 
> 
> > On Oct 25, 2019, at 3:26 AM, Michal Hocko <mhocko@...nel.org> wrote:
> > 
> > Considering the pagetypeinfo is a debugging tool we do not really need
> > exact numbers here. The primary reason to look at the outuput is to see
> > how pageblocks are spread among different migratetypes and low number of
> > pages is much more interesting therefore putting a bound on the number
> > of pages on the free_list sounds like a reasonable tradeoff.
> > 
> > The new output will simply tell
> > [...]
> > Node    6, zone   Normal, type      Movable >100000 >100000 >100000 >100000  41019  31560  23996  10054   3229    983    648
> 
> It was mentioned that developers could use this file is to see the
> movement of those numbers for debugging, so this supposed to introduce
> regressions as there is no movement anymore for those 100k+ items?

Can you provide an explicit example please? As the changelog mentions
it is the "low numbers" that is really interesting when debugging
fragmentation issues. Because we are running out of a respective migrate
type and it is interesting to see why and what compaction is doing etc.

Having more than 100k pages on the respective migrate type list is a
good sign that the migrate type is not in problems. Comparing multiple
migrate types with >100k pages doesn't really need a large precision
AFAIU.

Now, I do agree that some debugging tools might get confused by > in the
output but can we wait for reports and see whether anybody actually
cares? It is more likely that fixing one off debugging tools wouldn't be
a big deal.

-- 
Michal Hocko
SUSE Labs