vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

update warp size

Browse source
This commit is contained in:
Henri Vasserman 2023-05-20 23:42:20 +03:00
parent b19fefef94
commit 600ace39c8
No known key found for this signature in database
GPG key ID: 2995FC0F58B1A986
1 changed files with 1 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

2
ggml-cuda.cu
View file

@ -132,7 +132,7 @@ static_assert(sizeof(block_q8_0) == sizeof(ggml_fp16_t) + QK8_0, "wrong q8_0 blo
#define CUDA_MUL_BLOCK_SIZE 256
#define CUDA_DEQUANTIZE_BLOCK_SIZE 256
#define CUDA_DMMV_BLOCK_SIZE 32 // dmmv = dequantize_mul_mat_vec
#define CUDA_DMMV_BLOCK_SIZE 64 // dmmv = dequantize_mul_mat_vec
static __global__ void mul_f32(const float * x, const float * y, float * dst, const int kx, const int ky) {
const int i = blockDim.x*blockIdx.x + threadIdx.x;