tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-07-27 03:33:46 -04:00
Code Issues Packages Projects Releases Wiki Activity

ggml-cpu: activate nnpa for ggml_cpu_fp16_to_fp32

Browse Source 
Signed-off-by: Aaron Teo <aaron.teo1@ibm.com>
This commit is contained in:
Aaron Teo
2025-06-21 16:01:55 +08:00
parent 9330454cb8
commit ebc1d19f62
1 changed files with 13 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

13
ggml/src/ggml-cpu/ggml-cpu.c
View File

@ -3168,6 +3168,19 @@ void ggml_cpu_fp16_to_fp32(const ggml_fp16_t * x, float * y, int64_t n) {
_mm_storeu_ps(y + i, y_vec); _mm_storeu_ps(y + i, y_vec);
} }
#endif #endif
#if defined(__NNPA__)
for (; i + 3 < n; i += 4) {
uint16x8_t v_x = vec_xl(0, (const ggml_fp16_t *)(x + i));
uint16x8_t nnpa_dfl16 = vec_convert_from_fp16(v_x, 0);
float32x4_t result = vec_extend_to_fp32_hi(nnpa_dfl16, 0);
y[i + 0] = result[0];
y[i + 1] = result[1];
y[i + 2] = result[2];
y[i + 3] = result[3];
}
#endif
for (; i < n; ++i) { for (; i < n; ++i) {
y[i] = GGML_FP16_TO_FP32(x[i]); y[i] = GGML_FP16_TO_FP32(x[i]);
} }