xfs
[Top] [All Lists]

Re: unable to use xfs_repair

To: Steve Lord <lord@xxxxxxx>
Subject: Re: unable to use xfs_repair
From: Andi Kleen <ak@xxxxxxx>
Date: Fri, 4 Oct 2002 18:55:04 +0200
Cc: Andi Kleen <ak@xxxxxxx>, Sidik Isani <lksi@xxxxxxxxxxxxxxx>, linux-xfs@xxxxxxxxxxx
In-reply-to: <1033749897.6896.20.camel@jen.americas.sgi.com>
References: <20021003170436.A11748@cfht.hawaii.edu> <4.3.2.7.2.20021004090036.036b5728@pop.xs4all.nl> <20021003224118.A12588@cfht.hawaii.edu> <1033747014.2457.11.camel@jen.americas.sgi.com> <20021004182151.A8443@wotan.suse.de> <1033749897.6896.20.camel@jen.americas.sgi.com>
Sender: linux-xfs-bounce@xxxxxxxxxxx
User-agent: Mutt/1.3.22.1i
On Fri, Oct 04, 2002 at 11:44:57AM -0500, Steve Lord wrote:
> On Fri, 2002-10-04 at 11:21, Andi Kleen wrote:
> > > OK, I went and had a look, and we are actually as efficient as we can 
> > > be in use of the memory itself. Only problem is that we allocate 4 times
> > > what we use. I fixed this in cvs.
> > 
> > Unless you memset it (=actually using, not allocating) and it still
> > fits into the 32big address space that could be worked around by
> > echo 1 > /proc/sys/vm/overcommit_memory 
> 
> It uses all the memory, 4 bits per filesystem block, we record what
> type of information is in that part of the fs - including free blocks.

When it allocates more than it actually faults in then this tweak may help.

-Andi


<Prev in Thread] Current Thread [Next in Thread>