Date: Mon, 20 Apr 1998 09:26:38 -0700 From: Mike Smith <mike@smith.net.au> To: Bruce Evans <bde@zeta.org.au> Cc: mike@smith.net.au, dburr@POBoxes.com, hardware@FreeBSD.ORG, stable@FreeBSD.ORG, Studded@san.rr.com Subject: Re: best wdc0 flags ? Message-ID: <199804201626.JAA00670@dingo.cdrom.com> In-Reply-To: Your message of "Tue, 21 Apr 1998 02:05:03 %2B1000." <199804201605.CAA04114@godzilla.zeta.org.au>
next in thread | previous in thread | raw e-mail | index | archive | help
> >> > I tried those flags and noticed that it turned on 32-bit transfers for > >> >my wd0, resulting in a 20% increase in throughput. Out of curiosity, why > >> >aren't these flags included in GENERIC? > >> > >> Becuase they break operation of drives that don't support them. > > > >Do we have any examples of controllers that don't? > > I thought I did, but my oldest accessible drive (all 400MB of it from > 4 years ago) supports them. The probe seems to handle any that don't. OK. Should we make it the default then? > Setting the multi-block flag is not such a good optimization, since it > pessimizes throughput on some drives and it increases interrupt latency. Can you qualify "some drives" again? The overall performance improvement in general use is marked, and it decreases interrupt load in the DMA case. -- \\ Sometimes you're ahead, \\ Mike Smith \\ sometimes you're behind. \\ mike@smith.net.au \\ The race is long, and in the \\ msmith@freebsd.org \\ end it's only with yourself. \\ msmith@cdrom.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199804201626.JAA00670>