Date: Tue, 18 Jun 2002 17:25:52 +0200 From: Matthias Andree <matthias.andree@gmx.de> To: freebsd-stable@freebsd.org Subject: Re: ATA Atapi 4.6 Release Message-ID: <m3n0tsr4lb.fsf@merlin.emma.line.org> In-Reply-To: <E17KJst-0003up-00@smart.eusc.inter.net> (Matthias Schuendehuette's message of "Tue, 18 Jun 2002 16:23:02 %2B0200") References: <E17KJst-0003up-00@smart.eusc.inter.net>
next in thread | previous in thread | raw e-mail | index | archive | help
Matthias Schuendehuette <msch@snafu.de> writes: > I would encourage some posters to cool down a bit. There are only a few > types of ATA-Disks that support Tagged Queuing and they all work with > Write Cache nearly as fast as with WC+TQ. Before telling others to be quiet: write cache without tagged queueing is *unsafe* and may corrupt your data, softupdates forth or back. This is not about speed alone, but about maintaining safety while going fast. There are exactly two safe bets: 1. write cache switched off or 2. tagged queueing (which implies cache on). With FreeBSD 4.5, this is: slow or fast. With FreeBSD 4.6, this is: slow or dead slow (because TQ falls back to PIO after some timeouts). Which leaves you with the recommendation: When running FreeBSD 4.6, put these lines in your /boot/loader.conf: hw.ata.wc="0" hw.ata.tags="0" hw.ata.ata_dma="1" hw.ata.atapi_dma="1" There are a couple of drives that do support tagged queueing, more than FreeBSD will even try TQ on. I have heard of no reports of broken TQ on any other drive than the original IBM DJNA, and that I could not verify myself. -- Matthias Andree To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?m3n0tsr4lb.fsf>