Date: Wed, 14 May 2003 17:29:27 -0600 From: Scott Long <scott_long@btc.adaptec.com> To: Nate Lawson <nate@root.org> Cc: scsi@freebsd.org Subject: Re: SCSI geometry calculation? Message-ID: <3EC2D157.5000705@btc.adaptec.com> In-Reply-To: <20030514161256.N79399@root.org> References: <20030514151752.B79363@root.org> <3EC2CBD0.3020102@btc.adaptec.com> <20030514161256.N79399@root.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Nate Lawson wrote: > On Wed, 14 May 2003, Scott Long wrote: > >>>Most drivers use >1G: 255/63, else 64/32. Exceptions are: >>>* aac - >=2G: 255/63, >=1G 128/32, else 64/32 >>>* aha - same >>>* bt - same as aha > > > Why are aha and bt using an extra 2G step instead of the normal 1G step? > They're not RAID controllers. Did you use that geometry for aac > intentionally or was it just a cut/paste? > The aac definition was taken from the aac spec. The aac spec coincidentally is similar to other adaptec products, though Justin can speak definitively on aic7xxx/aic79xx. > >>I'm really not sure what to say >>about the boundary cases other than if they are buggy, few people >>notice. > > > Would it be ok to move them all to > and not use >=? > Probably yes, though it might be useful to get build some experimental evidence first. Scott
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3EC2D157.5000705>