"The launch timed out and was terminated" error with Bumblebee on Linux

Question

When running a long kernel (especially in debug mode with some memory checking) on a CUDA-enabled GeForce GPU with Bumblebee, I get the following error:

CUDA error 6: the launch timed out and was terminated

This seems to be caused by the NVIDIA driver's watchdog. A solution is available here. However, why is this happening while using Bumblebee and optirun to run a simple CUDA kernel (i.e. I do not use my NVIDIA GPU for display)?

The command I used to launch the program is:

optirun [cuda-memcheck] ./my_program program_options

BenC · Accepted Answer · 2013-06-27T06:31:09.197

4

The solution (found here) was to use the --no-xorg option for optirun, i.e.:

optirun --no-xorg [cuda-memcheck or cuda-gdb] ./my_program program_options

Indeed, the default behavior of optirun is to create a secondary X server which will then be subject to the driver's watchdog. By using the --no-xorg option, we can avoid the unnecessary consequences of this extra X server. This new option is available since Bumblebee 3.2.

It also provides a way to use cuda-gdb and avoid the following error:

fatal: All CUDA devices are used for display and cannot be used while debugging. (error code = 24)

edited Jun 27 '13 at 06:31

answered Jun 27 '13 at 06:16

BenC

8,729
3
49
68

Even with the `--no-xorg` flag I still get timeout errors for a large enough (slow enough) program. Any ideas? – FreelanceConsultant Oct 25 '15 at 15:16
I never had the issue again after that, so I really don't know what that could be this time. – BenC Oct 26 '15 at 01:30

"The launch timed out and was terminated" error with Bumblebee on Linux

1 Answers1

Linked