Date: Fri, 06 Dec 1996 10:31:00 -0700 From: Steve Passe <smp@csn.net> To: Peter Wemm <peter@spinner.dialix.com>, smp@freebsd.org Subject: Re: last major problem Message-ID: <199612061731.KAA20846@clem.systemsix.com> In-Reply-To: Your message of "Thu, 05 Dec 1996 23:47:35 MST." <199612060647.XAA18003@clem.systemsix.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Hi, Erich also reports his quad P6 failing after starting additional CPUs, in the same repeatable manner as everyone else. He reports ability to recompile kernel many times with only 1 CPU running. So I think its definately a P6 problem. Nobody running P5s have reported it. So the next question is whether its strictly an APIC_IO bug, an SMP_INVLTLB bug, or a bug from combining the 2. Since you can't run SMP_INVLTLB without APIO_IO, the one test people with P6s can do is to see if its any more stable with APIC_IO alone, ie. NO SMP_INVLTLB. If they are more (completely) stable without SMP_INVLTLB then it suggests that is the cause (but is NOT proof). If however the problem continues to exist then we completely eliminate SMP_INVLTLB as the culprit and can blame APIC_IO. P6 testers? (boy I'm jealous, wish I had one of those toys!) -- Steve Passe | powered by smp@csn.net | FreeBSD
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199612061731.KAA20846>