happyz
/
llama.cpp
mirror of https://github.com/ggerganov/llama.cpp.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

fix cuda

This commit is contained in:
Xuan Son Nguyen 2025-07-09 00:00:44 +02:00
parent 4d0195324e
commit b22708fd90
1 changed files with 4 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
ggml/src/ggml-cuda/scale.cu
View File

@ -24,8 +24,10 @@ void ggml_cuda_op_scale(ggml_backend_cuda_context & ctx, ggml_tensor * dst) {
GGML_ASSERT(src0->type == GGML_TYPE_F32); GGML_ASSERT(src0->type == GGML_TYPE_F32);
GGML_ASSERT( dst->type == GGML_TYPE_F32); GGML_ASSERT( dst->type == GGML_TYPE_F32);
float scale = ((const float *)(dst->op_params))[0]; float scale;
float bias = ((const float *)(dst->op_params))[1]; float bias;
memcpy(&scale, dst->op_params, sizeof(float));
memcpy(&bias, (float *) dst->op_params + 1, sizeof(float));
scale_f32_cuda(src0_d, dst_d, scale, bias, ggml_nelements(src0), stream); scale_f32_cuda(src0_d, dst_d, scale, bias, ggml_nelements(src0), stream);
} }