From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: with ECARTIS (v1.0.0; list xfs); Thu, 20 Sep 2007 16:34:25 -0700 (PDT) Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130]) by oss.sgi.com (8.12.10/8.12.10/SuSE Linux 0.7) with SMTP id l8KNYEgf007644 for ; Thu, 20 Sep 2007 16:34:18 -0700 Date: Fri, 21 Sep 2007 09:34:05 +1000 From: David Chinner Subject: Re: xfs and Linux Linux 2.6.22 and Memory Message-ID: <20070920233405.GA995458@sgi.com> References: <435577.15025.qm@web35711.mail.mud.yahoo.com> <46F2E001.4040107@sandeen.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <46F2E001.4040107@sandeen.net> Sender: xfs-bounce@oss.sgi.com Errors-to: xfs-bounce@oss.sgi.com List-Id: xfs To: Eric Sandeen Cc: Mariella Petrini , xfs@oss.sgi.com On Thu, Sep 20, 2007 at 04:02:57PM -0500, Eric Sandeen wrote: > Mariella Petrini wrote: > > > QUESTION: > > > > Is there any way to release that amount of memory > > without unmounting the file systems ? > > Is that caused to some caching mechanism ? > > Or could that be caused by something else ? > > It's most likely the linux VFS caching the dentries & inodes, and > therefore caching the xfs inodes as well. Not just inodes, but file data in the page cache as well. If you want to blow it all away without unmountin filesystems, then you should read up on /proc/sys/vm sysctls (i.e. /proc/sys/vm/drop_caches).... Cheers, Dave. -- Dave Chinner Principal Engineer SGI Australian Software Group