irq wrong in kernel 2.6.10?

Submitted by diederick76

on December 28, 2004 - 6:22am

Linux general

Hello,

I'm having problems booting with kernel 2.6.10. The kernel compiles okay, but when I try to boot, it stops after this line:

ACPI: PCI interrupt 0000:00:08:0[A] -> GSI 11 (level, low) -> IRQ 11

Booting kernel 2.6.9, that line differs slightly:

ACPI: PCI interrupt 0000:00:09.0[A] -> GSI 11 (level, low) -> IRQ 11

When I boot using acpi=off, it stops after these lines:

PCI: Found IRQ for device 0000:00:08.0 PCI: Sharing IRQ 11 with 0000:00:09.0 PCI: Sharing IRQ 11 with 0000:00:09.1

I have also tried compiling 2.6.10 without ACPI support, either with and without APM instead. This made no difference at all.

When booting stops, the computer only responds to its reset button, so it's hard to give more info on this, except by typing everything what's on the screen over using another computer. Not even CTRL+PgUP works.

When I boot using 2.6.9 (unpatched), everything is fine, including ACPI. IRQ is shared by uhci_hcd, ESS Maestro (the soundcard) and eth0, but not by ACPI. That one is at IRQ 9! Could it be that something's mixed up here? The BIOS config program (accessible at the beginning of booting) is very limited and has no options regarding IRQs.

I'm completely clueless as to what is the problem here, so I don't exactly know how much information would be needed to resolve this. My computer is a Compaq Armada E500 laptop with a Pentium III (Coppermine) CPU. I used my .config of the 2.6.9 to compile the 2.6.10.

I'm aware of the http://ftp.kernel.org/ pub/ linux/ kernel/ people/ lenb/ acpi/ patches/ release/ 2.6.10/ acpi-20041210-2.6.10-rc3.diff.bz2 patch. I've tried it, but it didn't make any difference.

I've attached my .config as config.txt

Any help is very welcome,
Diederick de Vries.

Attachment	Size
config.txt	28.89 KB

This is what I get at startup

Ano Nymous

December 28, 2004 - 8:44am

This is what I get at startup:

PCI: Using ACPI for IRQ routing
** PCI interrupts are no longer routed automatically. If this
** causes a device to stop working, it is probably because the
** driver failed to call pci_enable_device(). As a temporary
** workaround, the "pci=routeirq" argument restores the old
** behavior. If this argument makes the device work again,
** please email the output of "lspci" to bjorn.helgaas@hp.com
** so I can fix the driver.

Though unlikely, it may work if you use pci=routeirq. Other than that, I don't know. You could also try enabling/disabling apic at startup.

maybe, or maybe not...

Anonymous (not verified)

December 28, 2004 - 10:46am

http://lkml.org/lkml/2004/12/28/84 <--- ?

not

diederick76

December 29, 2004 - 4:39am

The kernel doesn't even compile with this patch::

kernel/irq/handle.c: In function `__do_IRQ': kernel/irq/handle.c:133: error: too many arguments to function `note_interrupt' kernel/irq/handle.c:187: error: too many arguments to function `note_interrupt' make[2]: *** [kernel/irq/handle.o] Error 1 make[1]: *** [kernel/irq] Error 2 make: *** [kernel] Error 2

Diederick

alas

diederick76

December 29, 2004 - 6:49am

No, it didn't make a difference. Thanks anyway.

I have exacly the same proble

Anonymous (not verified)

April 15, 2006 - 10:02am

I have exacly the same problem. Havn't tried the patch though... I'm stuck at using my old 2.6.15 kernel, hope they find a fix soon

lol please ignore my last com

Anonymous (not verified)

April 15, 2006 - 10:19am

lol please ignore my last comment as it doesn't make any sence:P
rather check this out: http://forums.amd.com/lofiversion/index.php/t58795.html