Skip to content

Commit

Permalink
Fix unintended change
Browse files Browse the repository at this point in the history
  • Loading branch information
matthewdouglas committed Oct 8, 2024
1 parent 0cc5c95 commit 0f2dc34
Showing 1 changed file with 2 additions and 2 deletions.
4 changes: 2 additions & 2 deletions csrc/ops.cu
Original file line number Diff line number Diff line change
Expand Up @@ -824,8 +824,8 @@ template <typename T> void gemm_4bit_inference(int m, int n, int k, T * A, unsi
template <typename T, int BITS> void gemm_4bit_inference_naive(int m, int n, int k, T * A, unsigned char* B, float *absmax, float *datatype, T * out, int lda, int ldb, int ldc, int blocksize, cudaStream_t stream)
{

int num_blocks = (m+7)/8;
kgemm_4bit_inference_naive<T, 256, BITS><<< num_blocks, 256, 0, stream>>>(m, n, k, A, B, absmax, datatype, out, lda, ldb, ldc, blocksize);
int num_blocks = (m+3)/4;
kgemm_4bit_inference_naive<T, 128, BITS><<< num_blocks, 128, 0, stream>>>(m, n, k, A, B, absmax, datatype, out, lda, ldb, ldc, blocksize);
CUDA_CHECK_RETURN(cudaPeekAtLastError());
}

Expand Down

0 comments on commit 0f2dc34

Please sign in to comment.