[BUG] Hopper groupgemm example fails for mnk(1638, 6144, 3584)

[zhipeng93]

Describe the bug

The example code here [1] fails to run mnk=(1638, 6144, 3584) and Got cutlass error: Invalid status at: 670.

Steps/Code to reproduce bug

cd cutlass/examples/57_hopper_grouped_gemm

nvcc -arch=sm_90a -I ../../include -I ../common/ -I ../../tools/util/include --expt-relaxed-constexpr -DNDEBUG 57_hopper_grouped_gemm.cu

./a.out --m=1638 --n=6144 --k=3584 --groups=5

Expected behavior The example code can run.

Environment details (please complete the following information): Docker, H800

[1] https://github.com/NVIDIA/cutlass/tree/main/examples/57_hopper_grouped_gemm