Linux - unexpected interrupt...

Mike Bilow mikebw at bilow.bilow.uu.ids.net
Sat Apr 10 00:34:03 PDT 1999



Christian Brandt wrote in a message to Mike Bilow:

 CB> Apr  4 18:39:14 lance kernel: unexpected IRQ vector 77 on CPU#1!
 CB> Apr  4 18:39:22 lance kernel: unexpected IRQ vector 70 on CPU#0!
 CB> Apr  4 18:41:35 lance kernel: Uhhuh. NMI received for unknown 
 CB> reason 2c.
 CB> Apr  4 18:41:35 lance kernel: Dazed and confused, but trying to 
 CB> continue
 CB> Apr  4 18:41:35 lance kernel: Do you have a strange power saving
 CB> mode enabled?

 CB> Those NMI and IRQ-party goes on at an rate of one event every 
 CB> two minutes.

 CB>  Will it wreck my system or can I ignore it?

Wow!  It's been a while since I saw vectors 0x70-0x77 pulled.  In the ancient
days of real mode, these were the vectors which received the high-numbered
IRQs.  That is, IRQ 8 would use vector 0x70, IRQ 9 would use vector 0x71, and
so on up to IRQ 15 which would use vector 0x77.

 CB> The System is a GigabyteDX Dual-PentiumMMX233-System with
 CB> onboard aic7880 UltraWide.

Does this have IO-APIC support?  Here's what an Intel PR440FX "Providence"
looks like on stock kernel 2.2.5, with the AIC-7880 controller on IRQ 17:

colossus:~$ cat /proc/interrupts
           CPU0       CPU1
  0:    2317540          0          XT-PIC  timer
  1:          0          2    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:      21771      21832   IO-APIC-level  aic7xxx
 18:      33455      35813   IO-APIC-level  Intel EtherExpress Pro 10/100
NMI:          0
ERR:          0

Whenever you see anything involving IRQ 8, that looks like an SMP issue.  We
have had great results running kernel 2.2.5 on this SMP motherboard.  Do you
have enhanced RTC support compiled into your kernel?  Do you have MTRR support
enabled in your kernel?
 
-- Mike




To Unsubscribe: send mail to majordomo at FreeBSD.org
with "unsubscribe aic7xxx" in the body of the message




More information about the aic7xxx mailing list