Re: nobody cared about IRQ 19 (firewire, on a HP 2510p notebook)

[Thomas Gleixner <tglx@linutronix.de>]

On Sun, 30 Mar 2008, Stefan Richter wrote:
> I tend to believe it is a problem to be addressed in the x86 platform support,
> not a driver problem.

Depends. It might be unfixable.

> I Cc'd some random x86 folk...  To rehash the issue:
> 
>   - Controller: Ricoh Co Ltd R5C832 IEEE 1394 Controller (rev 04)
>     Board: PM965/GM965/GL960 based
>     The R5C832 is known to work with ohci1394 according to
>     http://hardware4linux.info/component/14348/ and other reports.
> 
>   - Martin also saw it happen with Linux 2.6.22.
> 
>   - firewire-ohci + firewire-core as well as ohci1394 + ieee1394
>     appear to initialize the controller on Martin's laptop just fine.
>     Among else this means that a number of register reads and writes
>     succeed.
> 
>   - Some time later, without having actually used the FireWire
>     controller, "irq 19: nobody cared"/ "Disabling IRQ #19" occurs.
>     AFAIU the code, this is apparently because firewire-ohci's or
>     ohci1394's IRQ handler was called repeatedly but got either 0 or ~0
>     when reading the chip's interrupt event register.

Can we please have a full boot log (dmesg) and the output of /proc/interrupts and "lspci -vvv"

Thanks,
	tglx