Date: Wed, 8 Mar 2006 22:27:01 +0100 From: Pawel Jakub Dawidek <pjd@FreeBSD.org> To: Mike Tancsa <mike@sentex.net> Cc: freebsd-stable@FreeBSD.org Subject: Re: gmirror(8) and graid3(8) changes. Message-ID: <20060308212701.GI737@garage.freebsd.pl> In-Reply-To: <6.2.3.4.0.20060308144518.08c4de08@64.7.153.2> References: <20060306222844.GC56506@garage.freebsd.pl> <6.2.3.4.0.20060308134932.0633a5e8@64.7.153.2> <20060308193925.GH737@garage.freebsd.pl> <6.2.3.4.0.20060308144518.08c4de08@64.7.153.2>
next in thread | previous in thread | raw e-mail | index | archive | help
--wjoFZxbW4tu+iR6v Content-Type: text/plain; charset=iso-8859-2 Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Wed, Mar 08, 2006 at 03:40:24PM -0500, Mike Tancsa wrote: +> At 02:39 PM 08/03/2006, Pawel Jakub Dawidek wrote: +> >On Wed, Mar 08, 2006 at 01:50:42PM -0500, Mike Tancsa wrote: +> >+> At 05:28 PM 06/03/2006, Pawel Jakub Dawidek wrote: +> >+> >Hi. +> >+> > +> >+> >Here you can find patches with changes to gmirror(8) and graid3(8): +> >+> > +> >+> > http://people.freebsd.org/~pjd/patches/gmirror.7.patch +> >+> > http://people.freebsd.org/~pjd/patches/graid3.patch +> >+> +> >+> Hi, +> >+> Against stable, this patch fails in places. This is against RELENG= _6 as of today +> > +> >graid3 from HEAD and RELENG_6 differ. Can you apply the patches to HEAD +> >version and just copy entire sys/geom/raid3 from HEAD to RELENG_6? +>=20 +> Yes, done. +>=20 +> I had saved the test results from the last set of tests and added the ne= w results. +> http://www.tancsa.com/raid3.html +>=20 +> There is both a speed improvement and regression in reading depending on= the test size. I repeated the 2G and 3G test twice and there was very lit= tle change between tests,=20 +> but it doesnt seem to make sense why 4G would be faster than 2G. Also, = this test if you recall, used to lock up the box with preemption in the ker= nel. The version in=20 +> HEAD, with the above patch works just fine in RELENG_6 using these simpl= e tests and does not lockup. Ok, speed looks acceptable. Good to hear there are no more lockups. Are you able to mensure synchronization speed between stock RELENG_6 and patched patched HEAD version? For patched HEAD version it would be great if you could experiment with different values of kern.geom.raid3.sync_requests (you need to set it in loader.conf) (the default is 4, it'll be good to see how it performs with 1, 8, 16 and maybe 32). Thank you! --=20 Pawel Jakub Dawidek http://www.wheel.pl pjd@FreeBSD.org http://www.FreeBSD.org FreeBSD committer Am I Evil? Yes, I Am! --wjoFZxbW4tu+iR6v Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (FreeBSD) iD8DBQFED0wlForvXbEpPzQRAkDaAJ9P0cRJkviPMLl48Hc0GaN74QyF1wCfbrIO Hdat4qWX2VkqS0IU+Ce+Q6Q= =G8vg -----END PGP SIGNATURE----- --wjoFZxbW4tu+iR6v--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20060308212701.GI737>