Need help debugging hard lock on high disk activity

[smani]

You're probably right. You seem to be doing the right things to get to a root cause of this problem. Anyway, that's why I suggested before that you post this issue on the kernel.org forums. However, if it is a kernel issue, I cannot see why others haven't had the same problem.

I guess this hardware combination is pretty rare (i.e. it when AGP started being replaced by PCIE, and those who built up a new workstation would rather buy a MB with PCIE slot for future upgrades). And I also guess most of the workstations running linux are those sold from HP/Dell which use their own mainboards. Hence I am not so surprised that the issue has not been (publicly) discovered yet.

Anyway, "kernel.org forums", could you please indicate which forums (or mailing list) you mean?
Thanks so far for all your answers!

[Rubberman]

Well, they don't have a specific forum, but they have a discussion mailing list, and a bug reporting link. Here they are:

Reporting kernel bugs/problems: Reporting bugs for the Linux kernel
Kernel mailing list & archives: The linux-kernel mailing list FAQ
MARC: Mailing list ARChives

[Rubberman]

You might also check with Home | The Linux Foundation

[smani]

Great, thanks a lot!

[smani]

I just tracked down the source of the problem to the ACPI APIC support in the BIOS... You mentioning interrupts just gave me the idea to hunt down the bios for any interrupt related settings, and, here we are ...

[Rubberman]

So, that fixed it? Enquiring minds, and all that.

[smani]

Well yeah copied 1gb files for one hour now without any hang... Lost Hyperthreading though, but I guess that's pretty much irrelevant. Wenn I'll have some more time I'll evventually subscribe to the kernel mailing list and continue hunting, but for now this is more than good enough! Again, thanks for your inputs!