Date: Tue, 20 Mar 2001 12:14:00 +1100 (EST) From: Bruce Evans <bde@zeta.org.au> To: John Baldwin <jhb@FreeBSD.ORG> Cc: Dag-Erling Smorgrav <des@ofug.org>, current@FreeBSD.ORG Subject: RE: Here's another one for you... Message-ID: <Pine.BSF.4.21.0103201206560.40617-100000@besplex.bde.org> In-Reply-To: <Pine.BSF.4.21.0103201153510.40498-100000@besplex.bde.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, 20 Mar 2001, Bruce Evans wrote: > On Mon, 19 Mar 2001, John Baldwin wrote: > > > Hmmm. An eip of 0 is bad. This could be just another instance of the bzero > > bug just in another place. You probably want to change the code that actually > > sets *bzero to i586_bzero (and same for any other ops that use floating point). > > The code in question for this lies in i386/isa/npx.c. It seems we use the fp > > regs for copyin/copyout and bcopy as well. I would just change line 458 of > > npx.c to say '#ifdef I586_CPU_XXX' for now as your temporary patch (then you > > don't need to patch pmap_zero_page() anymore.) > > There is no need to change anything. Just disable the fp optimizations > using the npx flags. Actually, there may be. The bandwidth test gets run on 586's even if the flags say not to use the result. This is to provide a "free" bandwidth test. It was harmless when the fp code wasn't broken. The flags are mainly for disabling using the fp code for accesses to broken device memory (bcopy and/or bzero were (are?) abuses to access device memory, and some device memory doesn't like 64-bit accesses). Bruce To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.4.21.0103201206560.40617-100000>