You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Using the example compartmental/rall.py in cuda_standalone mode on my GT 610 GPU the kernel execution fails due to CUDA error (during integration): too many resources requested for launch (when calling the *_integration kernel).
The problems seems to be the kernel register usage and the resolution is to remove (at least two) unused pointers from the kernel argument list. However, as the latter is generated in cuda_standalone/device.py and we either have to reduce the items in uses_variables within the respective template (and explicitly declare variables in kernels) or otherwise we need to have a way of explicitly excluding variables from the list %DEVICE_PARAMETERS% (and for %KERNEL_VARIABLES% similarily).
Using the example
compartmental/rall.py
in cuda_standalone mode on my GT 610 GPU the kernel execution fails due toCUDA error (during integration): too many resources requested for launch
(when calling the *_integration kernel).The problems seems to be the kernel register usage and the resolution is to remove (at least two) unused pointers from the kernel argument list. However, as the latter is generated in
cuda_standalone/device.py
and we either have to reduce the items inuses_variables
within the respective template (and explicitly declare variables in kernels) or otherwise we need to have a way of explicitly excluding variables from the list%DEVICE_PARAMETERS%
(and for%KERNEL_VARIABLES%
similarily).related with iss #4
The text was updated successfully, but these errors were encountered: