NVidia, Tyan k8w, xorg and acpi.

[drumz]

zander: and that's why I've been hoping someone from nvidia would speak up and point me in the right direction for troubleshooting this issue since the logs aren't giving anything worthwhile.

I've even run strace against it to no avail. I have found an alternative method for using strace that I'm going to try tonight that hopefully might provide a clue as to where the issue is. I'm technical, not afraid to get my fingers dirty, but need someone who's more versed in dealing with this type of lower level issue to guide me and tell me what it is I have to do to get this issue resolved once and for all (and hopefully help out a lot of other people too boot.

[ScofieldKid]

OK. After a day, 202beta didn't hack it.

Tried to launch a GL application, whamo. Hard reset. Damn.

Oh well... at least the 2D stuff is quick. Really makes you wonder what the reviewers do to get this stuff to work right.

[ sorry for these off topic messages. looks like either the K8W or the NVIDIA linux driver are just not ready for prime time. Damn. ]

[drumz]

Can you try a different card? For several people it works fine if you try a different card (although that was always with the black screen of death on startx).

If you missed it, a friend of mine has the 5700 ultra working fine on the k8w under suse 9.1 in 32bit mode.

[drumz]

Ok, I ran strace against the out of control X process. You can see the disappointing results below, absolutely no clue:

Process 6598 attached - interrupt to quit
Process 6598 detached

% time seconds usecs/call calls errors syscall
------ ----------- ----------- --------- --------- ----------------
------ ----------- ----------- --------- --------- ----------------
100.00 0.000000

The following is what /proc/interrupts looks like under a black screen of death:

CPU0
0: 237110 IO-APIC-edge timer
1: 8 IO-APIC-edge i8042
8: 0 IO-APIC-edge rtc
9: 2 IO-APIC-level acpi
12: 50 IO-APIC-edge i8042
14: 3653 IO-APIC-edge ide0
15: 19 IO-APIC-edge ide1
16: 189 IO-APIC-level nvidia
17: 7666 IO-APIC-level libata, AMD AMD8111
19: 3 IO-APIC-level ohci1394, ohci_hcd, ohci_hcd
24: 2732 IO-APIC-level eth0
NMI: 199
LOC: 237008
ERR: 0
MIS: 0


Here's when it's working in 'pci mode' (no black screen of death):
CPU0
0: 89826 IO-APIC-edge timer
1: 11 IO-APIC-edge i8042
8: 0 IO-APIC-edge rtc
9: 2 IO-APIC-level acpi
12: 50 IO-APIC-edge i8042
14: 3653 IO-APIC-edge ide0
15: 19 IO-APIC-edge ide1
16: 3157 IO-APIC-level nvidia
17: 7824 IO-APIC-level libata, AMD AMD8111
19: 3 IO-APIC-level ohci1394, ohci_hcd, ohci_hcd
24: 715 IO-APIC-level eth0
NMI: 12
LOC: 89743
ERR: 0
MIS: 0


There are many differences. So anyone have a clue what to do?

[volkher]

I have a also problem with X, Thunder K8W, Nvidia FX 5200 and Gentoo 64bit Linux: If I turn off acpi, whether in the kernel, at boot time or in bios, my X is very slow; it works actually, but you cannot really use it.
Maybe that's the opposite of your problem? Do you have any suggestions, because if I turn on acpi, I have major boot problems: the system hangs on the first boot, only if I boot a second time, it works!

regards, Volkher