tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-07-28 13:20:27 -04:00
Code Issues Packages Projects Releases Wiki Activity

ggml-cpu: bring back fp32->fp16 store nnpa

Browse Source 
Signed-off-by: Aaron Teo <aaron.teo1@ibm.com>
This commit is contained in:
Aaron Teo
2025-06-21 17:49:36 +08:00
parent 987d1690e4
commit 8ef51b9055
1 changed files with 12 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

14
ggml/src/ggml-cpu/simd-mappings.h
View File

@@ -963,7 +963,7 @@ static inline void __lsx_f16x4_store(ggml_fp16_t * x, __m128 y) {
#define GGML_F16_EPR GGML_F32_EPR
static inline float32x4_t __lzs_f16cx4_load(const ggml_fp16_t * x) {
#ifdef __NNPA__
#if defined(__NNPA__)
uint16x8_t v_x = vec_xl(0, (const ggml_fp16_t *)x);
uint16x8_t nnpa_dlf16 = vec_convert_from_fp16(v_x, 0);
return vec_extend_to_fp32_hi(nnpa_dlf16, 0);
@@ -980,8 +980,17 @@ static inline float32x4_t __lzs_f16cx4_load(const ggml_fp16_t * x) {
#endif
}
// TODO: check why this function is not being hit at all
static inline void __lzs_f16cx4_store(ggml_fp16_t * x, float32x4_t v_y) {
#if defined(__NNPA__)
float32x4_t v_zero = vec_splats(0.0f);
uint16x8_t v_xd = vec_round_from_fp32(v_y, v_zero, 0);
uint16x8_t v_x = vec_convert_to_fp16(v_xd, 0);
x[0] = vec_extract(v_x, 0);
x[1] = vec_extract(v_x, 1);
x[2] = vec_extract(v_x, 2);
x[3] = vec_extract(v_x, 3);
#else
float arr[4];
// note: keep type-cast here to prevent compiler bugs
@@ -991,6 +1000,7 @@ static inline void __lzs_f16cx4_store(ggml_fp16_t * x, float32x4_t v_y) {
for (int i = 0; i < 4; i++) {
x[i] = GGML_FP32_TO_FP16(arr[i]);
}
#endif
}
#define GGML_F16_VEC GGML_F32x4