Date: Fri, 19 May 1995 09:12:44 -0400 (EDT) From: Peter Dufault <dufault@hda.com> To: rgrimes@gndrsh.aac.dev.com (Rodney W. Grimes) Cc: steve2@genesis.tiac.net, hackers@FreeBSD.org, robin@interlabs.com Subject: Re: Adaptec 2940? Message-ID: <199505191312.JAA11265@hda.com> In-Reply-To: <199505190555.WAA15849@gndrsh.aac.dev.com> from "Rodney W. Grimes" at May 18, 95 10:55:52 pm
next in thread | previous in thread | raw e-mail | index | archive | help
Rodney W. Grimes writes: > > > > > Here's another one. Gateway 2000 486DX4/100 EISA, Anigma mb. > > > > [10:12pm]/usr/local/src/ram-speed% cc -O2 -o ram-speed ram-speed.c > > [10:13pm]/usr/local/src/ram-speed% ./ram-speed > > 49005fb0 0.332 uS/op 3.01e+06 op/S 11.480 Mb/S > > 8938c0df 0.130 uS/op 7.70e+06 op/S 29.371 Mb/S > > System was idle. > > > > Btw- There was some discussion a while back about Intel "Stepping 0" > > Intel CPU's. When I last booted 2.0 it reported stepping 0 for the CPU. > > What does this mean anyway? My system originally was a DX2/50 but I > > switched to the 33Mhz clock and got a clock tripled DX4 overdrive. > > If I recall correctly someone out there had a DX4 stepping 0 and where > having some problems. They called Intel who denied that a stepping 0 > of the DX4 chip was ever done. > > Now we have a second person with a stepping 0 486DX4 chip, care to > send me the complete CPUID output and if you can get the serial number > and the SX number off the chip I will see what Intel says about this. That was me. I sent the CPUID output before; I can send it again but I'll have to take a floppy to that machine and boot it. Let me know via e-mail if you want me to do that. At any rate, here is what it says on the chip: OverDRIVE DX40DPR100 C4430635 SZ926 V1.0 INTEL (M)(C) '89'93 I don't see an SX number, just that SZ number. This chip has not worked properly running FreeBSD in either of the two motherboards we tried it in, a Micronics "GEMINI" 486VL and a Compaq Presario. On the Micronics it core dumps during big compiles (that is the only evident problem; I ran with it for a few days) and on the Compaq it reproducibly dies with a "privileged instruction fault in kernel mode". The neat thing about the Compaq failure is that it is instruction pattern dependent: when you are lucky you wind up with a kernel that boots and runs fine multiuser at 100Mhz, and when you are unlucky you wind up with a kernel that reproducibly fails at the same instruction, sometimes during boot up and sometimes when you run a certain utility. The reported instruction, btw, is not privileged. The chip works OK in the Presario running FreeBSD when you turn the motherboard down to 25Mhz, and of course it "works fine with Windows" at 33Mhz. Since this was chip functions fine as a "Word 6 accellerator" it isn't crucial that I figure out why it fails with FreeBSD, though it has made me leery of buying any other 100Mhz 486's. Intel asked me to send them a bunch of info about the Compaq, which I scrounged around and then sent them, and it has vanished into the black hole of bad customer relations. Peter -- Peter Dufault Real Time Machine Control and Simulation HD Associates, Inc. Voice: 508 433 6936 dufault@hda.com Fax: 508 433 5267
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199505191312.JAA11265>