CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-11-14 11:07:10 +00:00
Code Issues Packages Projects Releases Wiki Activity

vDSP_vsmsa

Browse Source
This commit is contained in:
Xuan Son Nguyen
2025-07-09 11:55:56 +02:00
parent 265cb43538
commit 563aca0b56
1 changed files with 1 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
ggml/src/ggml-cpu/vec.h
View File

@@ -353,8 +353,7 @@ inline static void ggml_vec_mad_f32_unroll(const int n, const int xs, const int
inline static void ggml_vec_mad1_f32(const int n, float * y, const float s, const float b) { inline static void ggml_vec_mad1_f32(const int n, float * y, const float s, const float b) {
#if defined(GGML_USE_ACCELERATE) #if defined(GGML_USE_ACCELERATE)
vDSP_vsmul(y, 1, &s, y, 1, n); vDSP_vsmsa(y, 1, &s, &b, y, 1, n);
vDSP_vsadd(y, 1, &b, y, 1, n);
#elif defined(GGML_SIMD) #elif defined(GGML_SIMD)
#if defined(__ARM_FEATURE_SVE) #if defined(__ARM_FEATURE_SVE)
const int sve_register_length = ggml_cpu_get_sve_cnt() * 8; const int sve_register_length = ggml_cpu_get_sve_cnt() * 8;