Zwane Mwaikambo írta:
> On Sat, 27 Nov 2004, Zoltan Boszormenyi wrote:
> 
> 
>>I will soon. Although I have to apply this patch also, more than one
>>person use my machine at once. My 2 and a half year old son would be
>>angry without his cartoons. ;-)
>>
>>In the meantime it turned out that downing eth1 did not solve this
>>problem. The r8169 stopped spitting its messages but I still had one
>>
>>hda: dma_timer_expiry: dma status == 0x24
>>hda: DMA interrupt recovery
>>hda: lost interrupt
> 
> 
> Or you could try booting with acpi=off

I did. After about 2 hours uptime, I got some of these again.
Once when the machine was not too i/o stressed (playing a movie
off disk and xine was reading some hundreds of KB per second)
and then when I was extracting the kernel source tree.
It seems the problem doesn't depend on I/O stress. Although
the machine is 100MB+ into swap, it doesn make constant I/O
as the HDD led shows.

dmesg is attached again.

Now I try to compile a 2.6.10-rc2-bk11 with
the before mentioned extension and report back.

Best regards,
Zoltán Böszörményi