[Bf-cycles] Cycles criteria to set the default CUDA computing device
dfelinto at gmail.com
Mon Jan 14 23:26:20 CET 2013
You explained me about the CU_DEVICE_ATTRIBUTE_KERNEL_EXEC_TIMEOUT over
IRC, but I wanted to double-check if I got it right.
I have a system with three GPUs, all of them setup as "Display GPU".
In this case, should we have the CUDA_MULTI* as the automatic option if we
set CUDA as device?
Or to have CUDA_0 as the default (as it is now) is the expected behaviour?
The available computing devices:
('Intel Xeon CPU', 'Tesla M2070', 'Tesla M2070', 'Tesla M2070', 'Tesla
(‘CUDA_0′, ‘CUDA_1′, ‘CUDA_2′, ‘CUDA_MULTI_2′)
And with this patch  I get the following info:
*** Begin CUDA Debug
CUDA device 0 is a Display GPU
CUDA device 0 has "14"
CUDA device 1 is a Display GPU
CUDA device 1 has "14"
CUDA device 2 is a Display GPU
CUDA device 2 has "14"
*** End CUDA Debug
Is this the expected behaviour?
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Bf-cycles