From owner-freebsd-smp Wed Mar 18 13:08:14 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id NAA21212 for freebsd-smp-outgoing; Wed, 18 Mar 1998 13:08:14 -0800 (PST) (envelope-from owner-freebsd-smp@FreeBSD.ORG) Received: from danberlin.resnet.rochester.edu (danberlin.resnet.rochester.edu [128.151.91.212]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id NAA21205 for ; Wed, 18 Mar 1998 13:08:11 -0800 (PST) (envelope-from root@danberlin.resnet.rochester.edu) Received: from localhost (root@localhost) by danberlin.resnet.rochester.edu (8.8.8/8.8.8) with SMTP id QAA00255; Wed, 18 Mar 1998 16:03:05 -0500 (EST) (envelope-from root@danberlin.resnet.rochester.edu) Date: Wed, 18 Mar 1998 16:03:04 -0500 (EST) From: Daniel Berlin To: ejon@colltech.com cc: Tor Egge , freebsd-smp@FreeBSD.ORG Subject: Re: MP config table robustness In-Reply-To: <199803181409.IAA19540@psasolar.psa.pencom.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hey Tor, looks like the patch does a lot of good. Sorry i couldn't get you the other MP table earlier than I did, looks like this patch makes a lot of people happy. --Dan On Wed, 18 Mar 1998 ejon@colltech.com wrote: > Tor, > I could kiss you! I applied this patch on my recalcitrant > DEC Personal Workstation 6200 and acheived my first FreeBSD-SMP > boot! MP 1.4 is not currently an option in my BIOS (doubt if DEC > is ever going to get around to updating it again), so this > patch probably wins by better supporting us MP 1.1 holdouts. > > Anyway, thanks! I _definitely_ owe you a beer for this. Next time > you're in the Boston area... > > Eric > > Strange sunspot activity caused Tor Egge to write: > > > > I see two problems with the current APIC code > > > > 1. Interrupts on the ISA bus are assumed to be conforming. When > > PCI interrupts are redirected to the ISA bus, this assumption can > > be wrong. > > > > 2. A PCI interrupt not specified in the MP table as a PCI interrupt > > is treated as a ISA interrupt delivered to the same interrupt > > pin number on the IOAPIC as the ISA interrupt number. This is > > inconsistent with the handling if ISA devices, where the IOAPIC > > pin number doesn't need to be the same as the ISA interrupt number. > > > > With a few changes, my -SMP machine (ASUS P6UP65 motherboard) can boot > > the -SMP kernel when MP 1.4 is disabled in the BIOS (as long as it has > > enough free interrupt to use for the PCI devices). The same kernel > > also works with MP 1.4 enabled in the bios. > > > > This change also seems to help on some versions of SuperMicro motherboards. > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-smp" in the body of the message > To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message