Date: 24 Oct 2001 21:00:33 +0200 From: Dag-Erling Smorgrav <des@ofug.org> To: Mike Silbersack <silby@silby.com> Cc: <cvs-committers@FreeBSD.org>, <cvs-all@FreeBSD.org> Subject: Re: cvs commit: src/sys/dev/ata ata-disk.c Message-ID: <xzpr8rsyib2.fsf@flood.ping.uio.no> In-Reply-To: <20011024134810.C77798-100000@achilles.silby.com> References: <20011024134810.C77798-100000@achilles.silby.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Mike Silbersack <silby@silby.com> writes: > Why are we changing this again? It was turned back on because without > write caching many people reported huge slowdowns. Has there since been > an influx of reports of huge losses of data on all versions other than > 4.3? Please look at the diff. It does not change the default value of the hw.ata.wc tunable (which, BTW, is still 0 in -CURRENT); it just fixes a bug where write cacheing would sometimes be incorrectly enabled even when the hw.ata.wc was set to 0. This is the correct behaviour for devices that support tagged queuing, but not for those that do not. (I wish people would make a rule of checking the diff before hitting "reply" to criticize a commit...) > (Should I bring up the irony of the tagged-queueing IBM drives having a > high failure rate? <g>) Only 75GXPs, AFAIK. They were pulled from the market in Norway, though not, I've been told, in many other countries. DES -- Dag-Erling Smorgrav - des@ofug.org 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?xzpr8rsyib2.fsf>