The aic7xxx driver is definitely unstable in my experience on SMP when
shared interrupts are used. It will usually hang on boot in this case.
This is an especially annoying problem because many SMP motherboards will
insist on assigning interrupts automatically; Intel and Supermicro are
particularly notorious for this. However, if you have IO-APIC support
enabled in the CMOS setup and in the Linux kernel, there should be no need
for assignment of shared interrupts. If you get this right, you should
see something like this:
colossus:~$ cat /proc/interrupts
CPU0 CPU1
0: 65033341 0 XT-PIC timer
1: 8 0 IO-APIC-edge keyboard
2: 0 0 XT-PIC cascade
8: 1 1 IO-APIC-edge rtc
13: 1 0 XT-PIC fpu
14: 4 3 IO-APIC-edge ide0
17: 1027421 1027589 IO-APIC-level aic7xxx
18: 1307813 1308648 IO-APIC-level Intel EtherExpress Pro 10/100 Ethernet
NMI: 0
ERR: 0
On this particular system, an Intel PR440FX "Providence" motherboard, both
SCSI and Ethernet are built-in.
It is also very strongly advised to have MTRR support enabled in the
kernel if you are playing around with systems of this caliber.
--Mike
On Tue, 28 Mar 2000, Peter Pregler wrote:
> I had (or have) similar hangs (all frozen, no syslog-entry, kernel still
> running since ping works, but all user-level stuff hangs). The hardware
> in my case is a onboard AIC-7890, an additional AHA-394X, a 3com 3c905B.
> All on a Dual-PIII-450 on a Asus motherboard running raidutils 0.9 on a
> SMP-2.2.14. I am not completely sure yet if I was able to solve the
> problem since it hang at uncertain intervalls. But what I did is make
> sure that no interrupt is shared by swapping cards in PCI-slots. The
> silly bios allocated the same interrupt to the ethernet card and to the
> onboard scsi-controller. Since I swapped the cards around to get a
> better interrupt assignment I have no freezesanymore. But since this is
> only 10 days now I am not yet convinced thatit was the actual cause.
