hangs with nvidia-driver 100.14.06

[RobotMarvin]

Hi everyone.

I have been giving the 100.14.06 driver a whirl the last few days and regularily ran into long hangs of my entire system... within seconds, the machine slowed down to an almost halt and (until now) always recovers after a minute or so... a few seconds before everything is back to normal, the screen flashes shortly (current screen -> black -> current screen).

I have attached the output of nvidia-bugreport.

Hope that helps.
matthew.

[whig]

dmesg is barfing. Both disk and video going down? Could be PSU.

...
ata5.01: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen
ata5.01: (BMDMA stat 0x64)
ata5.01: cmd a0/01:00:00:00:00/00:00:00:00:00/b0 tag 0 cdb 0x4a data 8 in
         res 58/00:02:00:00:00/00:00:00:00:00/b0 Emask 0x2 (HSM violation)
ata5: soft resetting port
ata5.00: configured for UDMA/33
ata5.01: configured for UDMA/33
ata5: EH complete
NVRM: Xid (0005:00): 8, Channel 00000000
saa7146 (0) vpeirq: used 1 times >80% of buffer (1271820 bytes now)
NVRM: Xid (0005:00): 9, Channel 00000020 Instance 00000000 Intr 00000080
...

[RobotMarvin]

Originally Posted by whig

Both disk and video going down? Could be PSU.

First if of all, my hard disks did not "go down". Those msgs are from the ATAPI drives. There are still some probs with ATAPI and libata which are not so uncommon. (take a look at google)

The hangs are not related to the 100.14.06 nvidia driver, so much I already figured because I can still reproduce them with 1.0.9755. So this leaves two possible culprits: either its the bleeding edge kernel (2.6.22-rc3-git7) or the even more bleeding edge scheduler (cfs). I'll have to investigate this even further...

Sorry for the trouble, seems I blamed the wrong guys. :-(
matthew.