I would like to perform a matrix operation on my kernel thank to cublas.
On my computer the program works propertly (7_CUDALibraries/simpleDevLibCUBLAS) but on the jetson k1 the program send me these messages:
GPU Device 0: “GK20A” with compute capability 3.2
simpleDevLibCUBLAS examples requires Compute Capability of SM 3.5 or higher
Is it possible to use cublas on kernel on 3.2 device, on jetson k1?