CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-11-08 10:07:01 +00:00
Code Issues Packages Projects Releases Wiki Activity

sycl: Remove not needed copy f16->f32 for dnnl mul mat (#14125)

Browse Source
This commit is contained in:
Anton Mitkov
2025-06-12 14:15:11 +01:00
committed by GitHub
parent a681b4ba83
commit ed52f3668e
2 changed files with 6 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
ggml/src/ggml-sycl/gemm.hpp
View File

@@ -65,6 +65,9 @@ public:
dnnl::primitive_attr primitive_attr;
primitive_attr.set_scratchpad_mode(dnnl::scratchpad_mode::user);
#ifdef GGML_SYCL_F16
primitive_attr.set_fpmath_mode(dnnl::fpmath_mode::f16);
#endif
auto a_mem = dnnl::memory(a_in_md, eng, const_cast<void*>(a));
auto b_mem = dnnl::memory(b_in_md, eng, const_cast<void*>(b));