TQM5200 problems (kernel 2.4)

[Kimmo Surakka]

> This is probably a hardware bug as described in:
> http://ozlabs.org/pipermail/linuxppc-embedded/2005-July/019038.html

Yes, that seems to be the case. The patch basically disables and again
enables the bus so that the missing clock pulse gets generated. I also
changes the code a bit so that an extra byte is not sent at the end of
transmissions.

The message http://ozlabs.org/pipermail/linuxppc-embedded/2005-July/019076.html
suggests to use a bit-banging driver instead of the buggy MPC5200 i2c
hardware. Maybe I'll look into that.

> > Warning: kfree_skb on hard IRQ c008ecfc
> > Warning: kfree_skb on hard IRQ c008ecfc
> > Warning: kfree_skb on hard IRQ c008ecfc
>
> I'm not seeing these (2.6-git from kernel.org)

Maybe it's indeed time to start using the 2.6 series, then. Thanks!

-- 
Kimmo Surakka <kusti at iki.fi>
http://www.iki.fi/kusti