tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-07-28 13:20:27 -04:00
Code Issues Packages Projects Releases Wiki Activity

cuda : refactor into multiple files (#6269)

Browse Source
This commit is contained in:
slaren
2024-03-25 13:50:23 +01:00
committed by GitHub
parent ad3a0505e3
commit ae1f211ce2
59 changed files with 9154 additions and 8987 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
ggml-cuda/dmmv.cuh Normal file
View File

@@ -0,0 +1,7 @@
#include "common.cuh"
void ggml_cuda_op_dequantize_mul_mat_vec(
ggml_backend_cuda_context & ctx,
const ggml_tensor * src0, const ggml_tensor * src1, ggml_tensor * dst, const char * src0_dd_i, const float * src1_ddf_i,
const char * src1_ddq_i, float * dst_dd_i, const int64_t row_low, const int64_t row_high, const int64_t src1_ncols,
const int64_t src1_padded_row_size, cudaStream_t stream);