xfs
[Top] [All Lists]
<prev [Date] next>
[Advanced]
 <prev [Thread] next>

Re: Kernel 2.6.9 Multiple Page Allocation Failures

from [Nick Piggin] [Permanent Link][Original]
To: Lukas Hejtmanek <xhejtman@xxxxxxxxxxxxxxxxx>
Subject: Re: Kernel 2.6.9 Multiple Page Allocation Failures
From: Nick Piggin <piggin@xxxxxxxxxxxxxxx>
Date: Wed, 08 Dec 2004 09:52:42 +1100
Cc: Andrew Morton <akpm@xxxxxxxx>, zaphodb@xxxxxxxxxxx, marcelo.tosatti@xxxxxxxxxxxx, linux-kernel@xxxxxxxxxxxxxxx, linux-xfs@xxxxxxxxxxx
In-reply-to: <20041203121129.GC27716@mail.muni.cz>
References: <20041116170527.GA3525@mail.muni.cz> <20041121014350.GJ4999@zaphods.net> <20041121024226.GK4999@zaphods.net> <20041202195422.GA20771@mail.muni.cz> <20041202122546.59ff814f.akpm@osdl.org> <20041202210348.GD20771@mail.muni.cz> <20041202223146.GA31508@zaphods.net> <20041202145610.49e27b49.akpm@osdl.org> <20041202231837.GB15185@mail.muni.cz> <20041202161839.736352c2.akpm@osdl.org> <20041203121129.GC27716@mail.muni.cz>
Sender: linux-xfs-bounce@xxxxxxxxxxx
User-agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.3) Gecko/20041007 Debian/1.7.3-5 


Lukas Hejtmanek wrote:

On Thu, Dec 02, 2004 at 04:18:39PM -0800, Andrew Morton wrote:

All I can say is "experiment with it".

It might be useful to renice kswapd so that userspace processes do not
increase its latency.


Hmm, increasing the min free kb to 64MB and renicing kswapd to -8 seems to solve the issue. However, for me it seems as not so good solution mainly because
2.6.6-bk4 kernel is just ok without any tweeks.



You're using 2.6.9, right? 2.6.10 should be better in this regard.


[More with this subject...]
<Prev in Thread] Current Thread [Next in Thread>
Previous by Date:  Large file support in Linux, Mike Xu
Next by Date:  Re: Large file support in Linux, Nathan Scott
Previous by Thread:  Re: Kernel 2.6.9 Multiple Page Allocation Failures, Lukas Hejtmanek
Next by Thread:  Re: Kernel 2.6.9 Multiple Page Allocation Failures, Nathan Scott
Indexes:  [Date] [Thread] [Top] [All Lists]