Date: Sat, 25 Oct 1997 21:55:35 +0000 From: =?iso-8859-1?Q?=DEor=F0ur?= Ivarsson <totii@est.is> To: Chuck Robey <chuckr@glue.umd.edu> Cc: "Jamil J. Weatherbee" <jamil@trojanhorse.ml.org>, freebsd-hackers@FreeBSD.ORG Subject: Re: Parity Ram Message-ID: <34526AD7.794BDF32@est.is> References: <Pine.BSF.3.96.971025151431.21591L-100000@picnic.mat.net>
next in thread | previous in thread | raw e-mail | index | archive | help
Chuck Robey wrote: > > On Sat, 25 Oct 1997, Žoršur Ivarsson wrote: > > > Jamil J. Weatherbee wrote: > > > > > > Can someone fill me in on when you would want to use parity ram as opposed > > > to non-parity ram these days? If there was some anomaly in memory how > > > would freebsd handle it (is there a trap for parity error?) > > > > As far as I know, the 'parity check fail' is connected to NMI of CPU. > > In most cases the BIOS rutines accept this and halt the computer with no > > information on where or why , only something like 'NMI detected, system > > halted' or > > 'Memory parity fail - NMI generated , system halted'. > > Huh ? BIOS routines? What's that got to do with FreeBSD? We don't use > the BIOS routines, they don't get called at all, right? If there's a > parity violation, if that's wired to NMI, then the NMI get's called, but > what that does is determined by FreeBSD, not your BIOS. > > > > > The only reason for this might be giving you some warning of failed > > memory rather > > than failed software. > > > > This has helped me several times when I was suspecting broken memory in > > the old days (90-93) :-) > > > > Thordur Ivarsson Ok, most of the old software and OSes did not fiddle with the NMI entry point so you did always get to the BIOS, but I don't know what happen in FreeBSD it self. Thordur Ivarsson
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?34526AD7.794BDF32>