Date: Mon, 06 Feb 2006 17:32:40 -0500 From: Dave Feustel <dfeustel@mindspring.com> To: freebsd-amd64@freebsd.org Subject: Re: Difference in stepping numbers? Message-ID: <200602061732.41302.dfeustel@mindspring.com> In-Reply-To: <20060206204419.GA51097@troutmask.apl.washington.edu> References: <20060206204419.GA51097@troutmask.apl.washington.edu>
next in thread | previous in thread | raw e-mail | index | archive | help
On Monday 06 February 2006 15:44, Steve Kargl wrote: > Anyone know what the difference between stepping=8 and a stepping=10 > opteron processors? When I purchased the processors I never noticed > the different values because dmesg reports only the cpu info for the > first processor, cpu0, which was stepping=8 in my case. > > I've recently experienced segfaults and signal 10s during builds of > the GCC source tree. I narrowed the problem down to a possible problem > with one of the cpus. Well, I yanked cpu0 with stepping=8 and moved > cpu1 with stepping=10 into the location of cpu0. Booting a UP kernel > revealed: > > CPU: AMD Opteron(tm) Processor 248 (2194.48-MHz K8-class CPU) > Origin = "AuthenticAMD" Id = 0xf5a Stepping = 10 > Features=0x78bfbff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR, > PGE,MCA,CMOV,PAT,PSE36,CLFLUSH,MMX,FXSR,SSE,SSE2> > AMD Features=0xe0500800<SYSCALL,NX,MMX+,LM,3DNow+,3DNow> > > At the moment, I have not been able to generate a signal 10 or 11 > under fairly heavy load on cpu with stepping=10. I thought that in dual-cpu systems both cpus were supposed to be of the same stepping. -- Lose, v., experience a loss, get rid of, "lose the weight" Loose, adj., not tight, let go, free, "loose clothing"
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200602061732.41302.dfeustel>