Date: Sat, 30 Aug 2003 20:04:00 +1000 (EST) From: Bruce Evans <bde@zeta.org.au> To: Dag-Erling =?iso-8859-1?q?Sm=F8rgrav?= <des@des.no> Cc: cvs-all@FreeBSD.org Subject: Re: cvs commit: src/sys/boot/i386/boot2 boot1.s Message-ID: <20030830195022.D3440@gamplex.bde.org> In-Reply-To: <xzpd6eocecg.fsf@dwp.des.no> References: <200308220159.h7M1xTKi048687@repoman.freebsd.org> <xzpd6eocecg.fsf@dwp.des.no>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sat, 30 Aug 2003, Dag-Erling [iso-8859-1] Sm=F8rgrav wrote: > Warner Losh <imp@FreeBSD.org> writes: > > Log: > > Many newer CF do not handle having the entire track read from them at > > boot time. Instead, read it a sector at a time. While this sounds > > like a significant slowdown, I've not been able to measure any > > signficant difference. > > It should make a big difference on floppies. Probably not even there for suitably modern hardware. Try "dd if=3D/dev/fd0 of=3D/dev/null". It reads mostly as fast as possible, 1 sector at a time, although there is no track buffer in the driver. It only misses reading the next sector and has to wait for 1 revolution after seeking, and naively implemented track reads would have the same inefficiency. I think all hardware newer than 8-10 years old can keep up doing single-sector reads, at least with the FreeBSD driver, but my BIOSes couldn't keep up in 1993 when I implemented the track buffer in biosboot. Bruce
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20030830195022.D3440>