linux-kernel - Re: [PATCH -mm] vmscan: bail out of page reclaim after swap_cluster

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, 14 Nov 2008 09:18:28 -0800
From:	Andrew Morton <akpm@...ux-foundation.org>
To:	Rik van Riel <riel@...hat.com>
Cc:	linux-kernel@...r.kernel.org, linux-mm@...ck.org
Subject: Re: [PATCH -mm] vmscan: bail out of page reclaim after
 swap_cluster_max pages

On Fri, 14 Nov 2008 09:36:28 -0500 Rik van Riel <riel@...hat.com> wrote:

> Andrew Morton wrote:
> > On Thu, 13 Nov 2008 17:12:08 -0500 Rik van Riel <riel@...hat.com> wrote:
> > 
> >> Sometimes the VM spends the first few priority rounds rotating back
> >> referenced pages and submitting IO.  Once we get to a lower priority,
> >> sometimes the VM ends up freeing way too many pages.
> >>
> >> The fix is relatively simple: in shrink_zone() we can check how many
> >> pages we have already freed and break out of the loop.
> >>
> >> However, in order to do this we do need to know how many pages we already
> >> freed, so move nr_reclaimed into scan_control.
> > 
> > There was a reason for not doing this, but I forget what it was.  It might require
> > some changelog archeology.  iirc it was to do with balancing scanning rates
> > between the various things which we scan.
> 
> I've seen worse symptoms without this code, though. Pretty
> much all 2.6 kernels show bad behaviour occasionally.
> 
> Sometimes the VM gets in such a state where multiple processes
> cannot find anything readily evictable, and they all end up
> at a lower priority level.
> 
> This can cause them to evict more than half of everything from
> memory, before breaking out of the pageout loop and swapping
> things back in.  On my 2GB desktop, I've seen as much as 1200MB
> memory free due to such a swapout storm.  It is possible more is
> free at the top of the cycle, but X and gnome-terminal and top
> and everything else is stuck, so that's not actually visible :)
> 
> I am not convinced that a scanning imbalance is more serious.

I'm not as sure as you are that it was done this way to avoid scanning
imbalance.  I don't remember the reasons :(

It isn't necessarily true that this change and <whatever that was> are
mutually exclusive things.

> Of course, one thing we could do is exempt kswapd from this check.
> During light reclaim, kswapd does most of the eviction so scanning
> should remain balanced.  Having one process fall down to a lower
> priority level is also not a big problem.
> 
> As long as the direct reclaim processes do not also fall into the
> same trap, the situation should be manageable.
> 
> Does that sound reasonable to you?

I'll need to find some time to go dig through the changelogs.  
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/