Keep the documentation consistent with the sgemm_1.cu code. (#2285)

* Keep the documentation consistent with the sgemm_1.cu code.

* fix typo

---------

Co-authored-by: zky <zky@126.com>
This commit is contained in:
zkyue
2025-05-20 10:53:15 +08:00
committed by GitHub
parent 5e9b8e2a25
commit 9354bfd7c1

View File

@ -206,8 +206,8 @@ With the static smem layouts, the `gemm_device` kernel can allocate the required
```cpp
// Shared memory buffers
__shared__ TA smemA[cosize_v<ABlockLayout>];
__shared__ TB smemB[cosize_v<BBlockLayout>];
__shared__ TA smemA[cosize_v<ASmemLayout>];
__shared__ TB smemB[cosize_v<BSmemLayout>];
Tensor sA = make_tensor(make_smem_ptr(smemA), sA_layout); // (BLK_M,BLK_K)
Tensor sB = make_tensor(make_smem_ptr(smemB), sB_layout); // (BLK_N,BLK_K)
```