From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o787kead242465 for ; Sun, 8 Aug 2010 02:46:40 -0500 Received: from mailsrv14.zmi.at (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 2D7B0D76721 for ; Sun, 8 Aug 2010 00:47:00 -0700 (PDT) Received: from mailsrv14.zmi.at (mailsrv1.zmi.at [212.69.164.54]) by cuda.sgi.com with ESMTP id kADZHkndxDaxO9K7 for ; Sun, 08 Aug 2010 00:47:00 -0700 (PDT) From: Michael Monnerie Subject: Re: How to track down abysmal performance ata - raid1 - crypto - vg/lv - xfs Date: Sun, 8 Aug 2010 09:46:53 +0200 References: <20100804085039.GA11671@infradead.org> <4C5A774D.6000400@hardwarefreak.com> <20100807101349.GD7362@dastard> In-Reply-To: <20100807101349.GD7362@dastard> MIME-Version: 1.0 Message-Id: <201008080946.57635@zmi.at> List-Id: XFS Filesystem from SGI List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============3497980372135725310==" Sender: xfs-bounces@oss.sgi.com Errors-To: xfs-bounces@oss.sgi.com To: xfs@oss.sgi.com --===============3497980372135725310== Content-Type: multipart/signed; boundary="nextPart4150250.H0cggjMXBm"; protocol="application/pgp-signature"; micalg=pgp-sha1 Content-Transfer-Encoding: 7bit --nextPart4150250.H0cggjMXBm Content-Type: Text/Plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable On Samstag, 7. August 2010 Dave Chinner wrote: > > Are you referring to /sys/block/sdx/queue/read_ahead_kb? I'm > > pretty sure this works below the FS level and below the partition > > level. This read_ahead works at the block device level. At least > > for individual or JBOD. >=20 > That number is used to initialise the default readahead value for > any file descriptor opened on the filesystem. readahead is tracked > per-fd at the page cache level, so is effectively at the filesystem > level, not the block device. >=20 > > Are you saying this setting gets ignored by the kernel if/when > > mdadm, LVM, and/or crypto are used? >=20 > Only the value from the block device the filesystem sits on is used. > i.e. if you are using /dev/md0, then the filesystem uses the value > from /sys/block/md0/queue/read_ahead_kb and ignores all the ones set > on the /dev/sdX devices that make up /dev/md0. =20 That would be a wonderful piece of documentation belonging to=20 Documentation/sysctl/fs.txt or vm.txt, and /sbin/sysctl should not only=20 change /proc/sys values, but also /sys entries. Such performance knobs are very poorly documented, but would be an=20 important bit for administrators to tune systems for their needs. This=20 is something that's really missing in Linux. =2D-=20 mit freundlichen Gr=FCssen, Michael Monnerie, Ing. BSc it-management Internet Services http://proteger.at [gesprochen: Prot-e-schee] Tel: 0660 / 415 65 31 ****** Aktuelles Radiointerview! ****** http://www.it-podcast.at/aktuelle-sendung.html // Wir haben im Moment zwei H=E4user zu verkaufen: // http://zmi.at/langegg/ // http://zmi.at/haus2009/ --nextPart4150250.H0cggjMXBm Content-Type: application/pgp-signature; name=signature.asc Content-Description: This is a digitally signed message part. -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.12 (GNU/Linux) iEYEABECAAYFAkxeYPEACgkQzhSR9xwSCbTtcACg2gg83JJO+lbH3KLkgJ4fDXlU DdIAnAkrxrZk2tQe3iQ1GfstSe1GxLez =7Hw6 -----END PGP SIGNATURE----- --nextPart4150250.H0cggjMXBm-- --===============3497980372135725310== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ xfs mailing list xfs@oss.sgi.com http://oss.sgi.com/mailman/listinfo/xfs --===============3497980372135725310==--