2.6.9: io conflict between w83627hf_wdt and parport_pc

2.6.9: io conflict between w83627hf_wdt and parport_pc

[Joerg Sommrey]

Hi,

the w83627hf WDT is no longer working on 2.6.9. The module refuses to
load with an error message:
|WDT driver for the Winbond(TM) W83627HF Super I/O chip initialising.
|w83627hf WDT: I/O address 0x002e already in use

/proc/ioports reports:
|002e-0030 : winbond_check

I found this function in parport_pc.  After unloading parport_pc,
w83627hf_wdt still refuses to load:
|jo@bear:/usr/src/linux-2.6.9:0$ sudo modprobe w83627hf_wdt
|FATAL: Error inserting w83627hf_wdt (/lib/modules/2.6.9/kernel/drivers/char/watchdog/w83627hf_wdt.ko): Input/output error

And I'm even unable to inspect /proc/ioports now:
|jo@bear:/usr/src/linux-2.6.9:0$ cat /proc/ioports
|Segmentation fault

Everything worked perfect with 2.6.8.1 and 2.6.9-rc4-mm1
The box is a Tyan MPX S2466 w/ 2x Athlon MP 2000+.

-jo

-- 
-rw-r--r--  1 jo users 63 2004-10-20 18:32 /home/jo/.signature

Re: 2.6.9: io conflict between w83627hf_wdt and parport_pc

[Lee Revell]

On Wed, 2004-10-20 at 12:54, Joerg Sommrey wrote:
> /proc/ioports reports:
> |002e-0030 : winbond_check

Ugh.  Looks like this bug made it into 2.6.9:

http://lkml.org/lkml/2004/10/18/58

2.6.9.1?  :-)

Lee

Re: 2.6.9: io conflict between w83627hf_wdt and parport_pc

[Joerg Sommrey]

On Wed, Oct 20, 2004 at 01:21:20PM -0400, Lee Revell wrote:
> On Wed, 2004-10-20 at 12:54, Joerg Sommrey wrote:
> > /proc/ioports reports:
> > |002e-0030 : winbond_check
> 
> Ugh.  Looks like this bug made it into 2.6.9:
> 
> http://lkml.org/lkml/2004/10/18/58
Thanks for the pointer.  Andrea's parport_pc-superio-chip-fixes.patch
fixed this bug for me.

-jo

-- 
-rw-r--r--  1 jo users 63 2004-10-20 20:11 /home/jo/.signature