Re: Interrupt Counter smp and aic7xxx

Doug Ledford (dledford@dialnet.net)
Sun, 15 Feb 1998 10:20:00 -0600


Carsten Gross wrote:
> I think there's a problem with the interrupt counter in Linux 2.1.86 (SMP
> compiled) and the aic7xxx driver on a Tyan Tomcat II Dual Pentium Board with
> BIOS rev. V4.01.
>
> python:~>cat /proc/scsi/aic7xxx/0
> Adaptec AIC7xxx driver version: 4.1/3.2
> [...]
> Base IO Memory: 0xe0804000
> IRQ: 18
> SCBs: Used 16, HW 16, Page 16
> Interrupts: 4525
>
> If I 'cat' the interrupt count, it says:
> python:~>cat /proc/interrupts
> CPU0 CPU1
> [...]
> 18: 8940 8950 IO-APIC aic7xxx
>
> This was nearly at the same time. This big difference is weired.

Not really. The interrupt count frmo /proc/scsi/aic7xxx/x is the number of
times we've ran through the isr routine. The interrupt count in the
/proc/interrupts file is the number of interrupts actually received from the
card. The aic7xxx_isr is a somewhat slow routine that will try to catch
multiple interrupts with one instance if possible. So, most likely, what
you're seeing here is the result of us catching several interrupts in one
call into the isr on a regular basis.

-- 

Doug Ledford <dledford@dialnet.net> Opinions expressed are my own, but they should be everybody's.

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu