Date: Mon, 22 Jan 96 14:53:47 MET From: Greg Lehey <lehey.pad@sni.de> To: mpp@mpp.minn.net (Mike Pritchard) Cc: hackers@freebsd.org (Hackers; FreeBSD) Subject: Re: stanford benchmark/usenix Message-ID: <199601221357.OAA15408@nixpbe.pdb.sni.de> In-Reply-To: <199601221355.HAA20992@mpp.minn.net>; from "Mike Pritchard" at Jan 22, 96 7:55 am
next in thread | previous in thread | raw e-mail | index | archive | help
> > Greg Lehey wrote: > > > > > > >>>> David Greenman said: > > > > > >Do we have pentium optimized bcopy and bzero ? > > > > > > > > > > > >Because some of the benchmarks could clearly benefit from them. > > > > > > > > > > After reading the Usenix paper on OS performance on Pentium machines, I'm > > > > > inclined to add optimized code to our libc. Basically, get the processor typ > > > > e > > > > > (probably via sysctl) and use this to control which versions are called - > > > > > similar to what I recently did with bzero in the kernel. > > [trimmed] > > > Wouldn't it make more sense to have separate libraries for each > > processor type, and to install the appropriate versions? > > What if you want to share your /usr with a bunch of diskless > machines of mixed cpu types? You use the generic version. If you're running dickless, you're probably not looking for blazing fast performance. Greg
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199601221357.OAA15408>