Date: Wed, 19 Jul 95 22:21:25 MDT From: terry@cs.weber.edu (Terry Lambert) To: msmith@atrad.adelaide.edu.au (Michael Smith) Cc: john@starfire.mn.org, hackers@FreeBSD.org Subject: Re: FreeBSD 2.x may not run on Cyrix 486DL Message-ID: <9507200421.AA01058@cs.weber.edu> In-Reply-To: <199507200013.JAA09410@genesis.atrad.adelaide.edu.au> from "Michael Smith" at Jul 20, 95 09:43:30 am
next in thread | previous in thread | raw e-mail | index | archive | help
> Terry Lambert stands accused of saying: > > The correct behaviour would probably be to disable the cache on Cyrix > > chips because they do not honor the non-cacheable bit. > > Unless someone can overcorrect me, I'd ask that this be read as "some" > Cyrix chips; certainly their DX40 and DX2/66 parts _appear_ to behave > appropriately. Yes; the problems are with Cyrix parts built with the Cyrix and TI chip masks (Cyrix and TI used to use the same chip masks and do co-developement). The problem does not exist with the Cyrix masks derived from the IBM chip masks (of which the DX40 and DX2/66 are examples). I don't have the details of stepping numbers, etc. because I simply don't buy Cyrix or TI parts (mail order salesmen will assure you of generally anything with regard to revisions just to make the sale). We know at least two that work here, but it would be nice to have a list, or even better, to have an auto-detect and correct + message. Terry Lambert terry@cs.weber.edu --- Any opinions in this posting are my own and not those of my present or previous employers.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?9507200421.AA01058>