On Sun, 28 May 2000, Keith Owens wrote:
> On Sun, 28 May 2000 00:50:23 -0700 (PDT),
> Andre Hedrick <[EMAIL PROTECTED]> wrote:
> >Ingo M..........where are you.........
> >May 28 00:42:50 p6dnf kernel:
> > spurious APIC interrupt on CPU#0, should never happen.
> >Screen FLOODED had to ssh to from another box to read the logs
> >Sheesh it runs but the screen is APIC error to infinity.
>
> BP6 has done this since day 1, Igno says the messages are info only.
> Against 2.4.0-test1-ac4
There is one hook involved here........it blows interrupts is SMP mode.
The ISR's have not changed but the code is becoming unstable.
The fact that high torture under UP and SMP-NOAPIC are stable with loads
approaching 60.00 means that the code can be reasonably assumed correct.
Now more facts......the onboard or "IO-APIC-edge" are clean and do not
give "lost interrupt"; however, the offboard or add-on PCI-IDE cards or
"IO-APIC-level" give "lost interrupt". This is regardless of DMA or PIO.
This I would define as a SMP specific issue, your thoughts....
Andre Hedrick
The Linux ATA/IDE guy
-
Linux SMP list: FIRST see FAQ at http://www.irisa.fr/prive/dmentre/smp-howto/
To Unsubscribe: send "unsubscribe linux-smp" to [EMAIL PROTECTED]
