Hi all,
is there any way to get more useful information about the compilation, as the CUDA --ptxas-options=-v.
I’m particulary interested in getting information about private mem allocation (how much registers were alocated per block).
Thanks
Hi all,
is there any way to get more useful information about the compilation, as the CUDA --ptxas-options=-v.
I’m particulary interested in getting information about private mem allocation (how much registers were alocated per block).
Thanks
bump.
Actually, is there a way to get information regarding this during the compile time into the cl_kernel object or along with it?
Hi all,
I’ve found a NVIDIA OpenCL compiler extension that provides the verbose option.
It’s: -cl-nv-verbose