Initializing...
GPU Device 0: "Hopper" with compute capability 9.0

M: 8192 (8 x 1024)
N: 8192 (8 x 1024)
K: 4096 (4 x 1024)
Preparing data for GPU...
Required shared memory size: 68 Kb
Computing using high performance kernel = 0 - compute_dgemm_async_copy
Time: 30.856800 ms
FP64 TFLOPS: 17.82