Date: Thu, 10 Jul 1997 13:10:40 +0200 From: Stefan Esser <se@FreeBSD.ORG> To: "John W. DeBoskey" <jwd@unx.sas.com> Cc: freebsd-current@FreeBSD.ORG Subject: Re: 3.0-970621-SNAP & fxp0 failure Message-ID: <19970710131040.40020@mi.uni-koeln.de> In-Reply-To: <199707092205.PAA24480@implode.root.com>; from David Greenman on Wed, Jul 09, 1997 at 03:05:11PM -0700 References: <199707092020.AA26603@iluvatar.unx.sas.com> <199707092205.PAA24480@implode.root.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Jul 9, David Greenman <dg@root.com> wrote: > >fxp0: <Intel EtherExpress Pro 10/100B Ethernet> > > rev 0x01 int a irq 15 on pci0.17.0 > >vx0: <3COM 3C905 Fast Etherlink XL PCI> rev 0x00 int a irq 14 on pci1.9.0 > Looks like an interrupt problem. Many systems have the secondary IDE > controller mapped to using irq15, so perhaps that is the problem. Try moving > the Intel card to a physically different slot - this should cause the PCI > BIOS to assign a different interrupt. True! The PCI/ISA code in FreeBSD does not check for conflicts between bus types, currently. (I have fixed this in my development sources, but wait for a consensus on how conflicts checking is generally approached before I'll commit my code.) It is generally good to avoid using IRQ14 and IRQ15 for PCI devices. Your BIOS may allow to reserve them for ISA devices, even if you don't actually use any IDE drives. In case you *do* have IDE drives connected, the system will most probably crash, if there is an IRQ conflict. But I actually don't know enough about IDE to understand, when IRQ14 is used for a disk drive, and when IRQ15 gets used ... You may want to do a "dmesg | grep 'irq 14'" (and the same for 'irq 15') and look for more than one line matching the conditions. Regards, STefan
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19970710131040.40020>