From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: with ECARTIS (v1.0.0; list xfs); Sun, 02 Sep 2007 07:44:48 -0700 (PDT) Received: from amanpulo.fs3.ph (amanpulo.fs3.ph [72.51.42.241]) by oss.sgi.com (8.12.10/8.12.10/SuSE Linux 0.7) with ESMTP id l82Eih4p012910 for ; Sun, 2 Sep 2007 07:44:45 -0700 Received: from localhost (localhost [127.0.0.1]) by amanpulo.fs3.ph (Postfix) with ESMTP id 8E7E61E0D5953 for ; Sun, 2 Sep 2007 21:46:50 +0800 (PHT) Received: from amanpulo.fs3.ph ([127.0.0.1]) by localhost (amanpulo.fs3.ph [127.0.0.1]) (amavisd-new, port 10024) with LMTP id dCbeMq+GSUE3 for ; Sun, 2 Sep 2007 21:46:41 +0800 (PHT) Received: from auctoritas.local (unknown [222.127.47.132]) (using SSLv3 with cipher RC4-MD5 (128/128 bits)) (No client certificate requested) by amanpulo.fs3.ph (Postfix) with ESMTP id D832D1E0E08B5 for ; Sun, 2 Sep 2007 21:46:40 +0800 (PHT) Subject: Re: ZFS, XFS, and EXT4 compared From: Federico Sevilla III In-Reply-To: <1188513751.24970.109.camel@edge.yarra.acx> References: <1188454611.23311.13.camel@toonses.gghcwest.com> <1188457666.24970.94.camel@edge.yarra.acx> <20070830132002.GA4086@infradead.org> <1188513751.24970.109.camel@edge.yarra.acx> Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="=-WuH1W4y6xpJyStAd+cd0" Date: Sat, 01 Sep 2007 23:07:23 +0800 Message-Id: <1188659243.4430.8.camel@auctoritas.fs3.ph> Mime-Version: 1.0 Sender: xfs-bounce@oss.sgi.com Errors-to: xfs-bounce@oss.sgi.com List-Id: xfs To: xfs@oss.sgi.com --=-WuH1W4y6xpJyStAd+cd0 Content-Type: text/plain Content-Transfer-Encoding: quoted-printable On Fri, 2007-08-31 at 08:42 +1000, Nathan Scott wrote: > Possibly. Far more importantly for XFS, there really needs to be some > way for RAID drivers to say "even though I support write barriers, its > not a good idea for filesystems to enable write barriers by default on > me". Enabling write barriers everywhere, by default, seems to have a > far worse impact than any mkfs/mount option tweaking. On all my systems with software RAID or dm-crypt (or both), mounting XFS gives me a message about barriers being disabled because the underlying device doesn't support it. For good measure I disable write caching on all my systems with either software RAID or dm-crypt (or both). Am I reading the thread correctly that even with this message showing up, I still need to mount with nobarrier explicitly to improve performance? I also think it would be nice to add something like a modern-day tuning FAQ for XFS. I know there's a bit on the FAQ already but perhaps it needs an update? Thanks. --=20 Federico Sevilla III F S 3 Consulting Inc. http://www.fs3.ph --=-WuH1W4y6xpJyStAd+cd0 Content-Type: application/pgp-signature; name=signature.asc Content-Description: This is a digitally signed message part -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQBG2YAr5rCBSJO3Rr4RAkH6AJ9sUGdQkS8qnaoY/Cb6i51vRtVTvgCZATVY l9LKoNVPbMi16yxXZ7Nx8uI= =Ymg2 -----END PGP SIGNATURE----- --=-WuH1W4y6xpJyStAd+cd0--