by Hi all!
I am trying to track down the source of the heavy-io-load panics/slow downs on
my PowerBook G3/300 (the same problem was reported by several other people on
other machines too). Unfortunately I was not able to reproduce the panics with
recent kernels, but the slow-downs remain. However, I got LOTS of these
messages from dmesg after running an io intensive test program:
Bogus interrupt -1 from from PC = c0005520
The source of these messages is do_IRQ in arch/ppc/kernel/irq.c (I think line
676 ff. in ver. 2.1.130). I #defined SHOW_GATWICK_IRQS, added KERN_DEBUG to the
printk command (line 712) and changed "i" to "irq >> 5", but the printk was
never executed (so at least the comment in line 805 is not correct?). "cat
/proc/interrupts" tells me about lots of pmac-pic/ide0, pmac-pic/via-pmu and
pmac-pic/bmac-misc interrupts and a few ones from pmac-pic/ide1 and
pmac-pic/bmac-rxdma plus lots of bad ones, but nothing from gatwick.
Maybe someone can tell me
o where the bogus interrupts come from and
o wether they might be the source for the slow downs/panics?
Yours, Albrecht.
--
+--------------------------------------------------------------------------
| Dr.-Ing. Albrecht Dre\3 ---- |
| Max-Planck-Institut f"ur Radioastronomie |\ / /o o\ |
| Technische Abt. Optische Interferometrie | \ / | / | |
| Auf dem H"ugel 69 | \ | \ ---/ |
| D-53121 Bonn (Germany) ------------+------+------------------- |
| | / | |
| Phone (+49) 228 525 319 | / / |
| Fax (+49) 228 525 411 |/ / |
+-------------- electrical engineers do it with less resistance --------------+