immaTensorCoreGemm
-maxrregcount=255
cudaMallocManaged
cudaDeviceSynchronize
cudaFuncSetAttribute
cudaEventCreate
cudaEventRecord
cudaEventSynchronize
cudaEventElapsedTime
cudaFree
whole
./
../
../../common/inc
Matrix Multiply
WMMA
Tensor Cores
true
immaTensorCoreGemm.cu
1:CUDA Basic Topics
sm72
sm75
sm80
x86_64
linux
aarch64
windows7
ppc64le
linux
7.2
Tensor Core GEMM Integer MMA
exe