mirror of
https://github.com/ggerganov/llama.cpp.git
synced 2024-12-25 02:44:36 +00:00
cuda : fix bounds check for src0 rows in MMVQ kernel (whisper/2231)
* cuda : fix bounds check for src0 rows in MMVQ kernel * Update ggml-cuda/mmvq.cu Co-authored-by: Johannes Gäßler <johannesg@5d6.de> --------- Co-authored-by: Johannes Gäßler <johannesg@5d6.de>
This commit is contained in:
parent
b5fcf8ef5c
commit
19b7a836f6
@ -117,7 +117,7 @@ static __global__ void mul_mat_vec_q(
|
|||||||
tmp[j][i] = warp_reduce_sum(tmp[j][i]);
|
tmp[j][i] = warp_reduce_sum(tmp[j][i]);
|
||||||
}
|
}
|
||||||
|
|
||||||
if (threadIdx.x < rows_per_cuda_block) {
|
if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) {
|
||||||
dst[j*nrows_dst + row0 + threadIdx.x] = tmp[j][threadIdx.x];
|
dst[j*nrows_dst + row0 + threadIdx.x] = tmp[j][threadIdx.x];
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
Loading…
Reference in New Issue
Block a user