Quote:
Originally Posted by amonakov
If, like balgo, you have "GPU has fallen off the bus" message earlier in dmesg, try running some CUDA application before optirun. You will need to power on the card and create device files prior to running a CUDA program:
sudo tee /proc/acpi/bbswitch <<<ON
sudo mknod -m 666 /dev/nvidiactl c 195 255
sudo mknod -m 666 /dev/nvidia0 c 195 0
sudo modprobe nvidia
|
This won't work for me. The "fallen off the bus" error will be gone, when I start a cuda program before. But the other error messages (RmInitAdapter failed!) are still there.
-->
bbswitch: enabling discrete graphics
pci 0000:01:00.0: power state changed by ACPI to D0
pci 0000:01:00.0: power state changed by ACPI to D0
pci 0000:01:00.0: power state changed by ACPI to D0
pci 0000:01:00.0: power state changed by ACPI to D0
pci 0000:01:00.0: enabling device (0006 -> 0007)
nvidia: module license 'NVIDIA' taints kernel.
Disabling lock debugging due to kernel taint
nvidia 0000:01:00.0: power state changed by ACPI to D0
nvidia 0000:01:00.0: power state changed by ACPI to D0
vgaarb: device changed decodes: PCI:0000:01:00.0,olddecodes=io+mem,decodes=none
w ns=none
NVRM: loading NVIDIA UNIX x86_64 Kernel Module 302.17 Tue Jun 12 16:03:22 PDT 2012
NVRM: GPU at 0000:01:00.0 has fallen off the bus.
NVRM: RmInitAdapter failed! (0x26:0xffffffff:1170)
NVRM: rm_init_adapter(0) failed
NVRM: RmInitAdapter failed! (0x23:0x2f:674)
NVRM: rm_init_adapter(0) failed
NVRM: RmInitAdapter failed! (0x23:0x2f:674)
NVRM: rm_init_adapter(0) failed
<--
After that I'm not able to reactivate the card via bbswitch anymore.
-->
bbswitch: enabling discrete graphics
nvidia 0000:01:00.0: power state changed by ACPI to D0
nvidia 0000:01:00.0: Refused to change power state, currently in D3
nvidia 0000:01:00.0: power state changed by ACPI to D0
bbswitch: enabling discrete graphics
nvidia 0000:01:00.0: power state changed by ACPI to D0
nvidia 0000:01:00.0: Refused to change power state, currently in D3
nvidia 0000:01:00.0: power state changed by ACPI to D0
<--