CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-10-27 08:21:30 +00:00
Code Issues Packages Projects Releases Wiki Activity

cuda : add missing F32<->I32 entries in ggml_cuda_cpy_fn (#16060)

Browse Source
This commit is contained in:
Sigbjørn Skjæret
2025-09-18 13:28:22 +02:00
committed by GitHub
parent 2b6b55a59f
commit ad6bd9083b
1 changed files with 4 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
ggml/src/ggml-cuda/cpy.cu
View File

@@ -441,6 +441,10 @@ void* ggml_cuda_cpy_fn(const ggml_tensor * src0, ggml_tensor * src1) {
return (void*) cpy_flt<cpy_1_flt<nv_bfloat16, nv_bfloat16>>;
} else if (src0->type == GGML_TYPE_BF16 && src1->type == GGML_TYPE_F32) {
return (void*) cpy_flt<cpy_1_flt<nv_bfloat16, float>>;
} else if (src0->type == GGML_TYPE_F32 && src1->type == GGML_TYPE_I32) {
return (void*) cpy_flt<cpy_1_flt<float, int32_t>>;
} else if (src0->type == GGML_TYPE_I32 && src1->type == GGML_TYPE_F32) {
return (void*) cpy_flt<cpy_1_flt<int32_t, float>>;
} else {
GGML_ABORT("%s: unsupported type combination (%s to %s)\n", __func__,
ggml_type_name(src0->type), ggml_type_name(src1->type));