Date: Sun, 18 Nov 2001 10:52:16 -0800 (PST) From: Matthew Jacob <mjacob@feral.com> To: Bernd Walter <ticso@cicely8.cicely.de> Cc: Ian Dowse <iedowse@maths.tcd.ie>, cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG Subject: Re: cvs commit: src/sys/cam/scsi scsi_da.c Message-ID: <Pine.BSF.4.21.0111181052010.75433-100000@beppo> In-Reply-To: <20011118182350.D80219@cicely8.cicely.de>
next in thread | previous in thread | raw e-mail | index | archive | help
When rewriting marginal sectors, you can get long pauses. On Sun, 18 Nov 2001, Bernd Walter wrote: > On Sat, Nov 17, 2001 at 09:54:14AM -0800, Matthew Jacob wrote: > > > > > > > In message <Pine.BSF.4.21.0111170856380.6462-100000@beppo>, Matthew Jacob write > > > s: > > > >Do they really really hang, or does it simply take a very very long time to > > > >flush out the cache on an MO? > > > > > > >From the PR: > > > > > > >>Description: > > > >Fujitsu MO drives M2513A hang at various operations due to the synchronize cache > > > >command (actually, after a few minutes the bus is reset and the disk can be > > > >ejected). > > > > > > That's all the information I have - the mention of the bus resetting > > > suggests that it is more than just a long delay (+ we have other > > > similar quirks for Fujitsu drives). > > > > Hmm- rather than disable SYNC cache, maybe we ought to run up the timeout on > > this command- it's only 5 minutes, which I could see as being slightly > > marginal for an MO. > > 5 * 60s * 0.1M/s = 30M > Which means the drive must be either slower than 100k/sec or have a > cache bigger than 30M. > I would say 5 min should be sane. > > -- > B.Walter COSMO-Project http://www.cosmo-project.de > ticso@cicely.de Usergroup info@cosmo-project.de > > To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.4.21.0111181052010.75433-100000>