]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 4b20d8b)
ggml: aarch64: Implement SVE F16 kernels for vector functions (#15115)
authorPrashant Vithule <redacted>
Mon, 1 Sep 2025 18:13:16 +0000 (23:43 +0530)
committerGitHub <redacted>
Mon, 1 Sep 2025 18:13:16 +0000 (20:13 +0200)
commita0c2b207c596d1092a08615de61ab56a7d63515f
tree8e99039de33bf9baaa5effcc2f9a3849dd28aff5tree
parent4b20d8b7e31718d3fe8b4f12d220a9d19e4f1997commit | diff
ggml: aarch64: Implement SVE F16 kernels for vector functions  (#15115)

* Added sve implementation for vec_dot_fp16 Kernel

* removed white spaces

* Added comment

* removed white spaces

* changed GGML_F16x_VEC_FMA for code consistency

* Update vec.h

---------

Co-authored-by: vithulep <redacted>
ggml/src/ggml-cpu/simd-mappings.h diff | blob | history
ggml/src/ggml-cpu/vec.cpp diff | blob | history
ggml/src/ggml-cpu/vec.h diff | blob | history
Packaging of ggml-org/llama.cpp