public_git
/
whisper.cpp
mirror of https://github.com/ggml-org/whisper.cpp.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

ggml-cpu/ops: vectorize flash-attention V-cache F16 to F32 conversion (llama/26947) 

Co-authored-by: jinzihao <jinzihao.jzh@alibaba-inc.com>
This commit is contained in:
jinzihao 2026-08-13 17:04:46 +08:00 committed by Georgi Gerganov
parent ac9a74f1a9
commit 62031fed0b
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
ggml/src/ggml-cpu/ops.cpp
View File

@ -8941,7 +8941,7 @@ static void ggml_compute_forward_flash_attn_ext_tiled(
for (int tk = 0; tk < kv_tile; tk++) {
const char * v_data = (const char *)v->data + (ic + tk)*nbv1 + iv2*nbv2 + iv3*nbv3;
if (kv_type == GGML_TYPE_F16) {
ggml_fp16_to_fp32_row((const ggml_fp16_t *)v_data, V32 + tk * DV, DV);
ggml_cpu_fp16_to_fp32((const ggml_fp16_t *)v_data, V32 + tk * DV, DV);
} else {
memcpy(V32 + tk * DV, v_data, DV * sizeof(float));
}