Dave Hansen <haveblue@us.ibm.com> wrote:
>
> I've been beating on various versions of 2.5.59 all day long with no
> problems that I didn't cause. I started testing 2.5.61-mm1 and rand
> into a couple problems right away.
>
> The first I really doubt is -mm specific. I gets _loads_ of these, and
> the e1000 isn't working:
> NETDEV WATCHDOG: eth0: transmit timed out
> e1000: eth0 NIC Link is Up 1000 Mbps Full Duplex
>
> The e1000 driver hasn't been touched in weeks.
Don't know.
> Here's my /proc/interrupts:
> http://www.sr71.net/linux/interrupts
> I'm pretty sure we can see the problem here. Almost all interrupts are
> going to CPU0. Is this a summit thing?
No, that's the new irq balancing code. It only starts distributing
interrupts to other CPUs when the load gets higher. See how it
spread the ethernet interrupts. Apparently this is as-designed.
> The other looks a bit more insidious.
>
> Unable to handle kernel NULL pointer dereference at virtual address 0000003d
It might be best to test 2.5.61 first.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Sat Feb 15 2003 - 22:01:01 EST