llama.cpp
Leverage the existing GGML_F32_VEC helpers to vectorize ggml_vec_set_f32 for faster fills
#16522
Merged

Leverage the existing GGML_F32_VEC helpers to vectorize ggml_vec_set_f32 for faster fills #16522

sirus20x6
Leverage the existing GGML_F32_VEC helpers to broadcast the fill valu…
33087594
sirus20x6 sirus20x6 requested a review from ggerganov ggerganov 169 days ago
sirus20x6 sirus20x6 requested a review from slaren slaren 169 days ago
github-actions github-actions added ggml
sirus20x6 sirus20x6 marked this pull request as draft 169 days ago
Vectorize additional f32 helper loops
dff11736
sirus20x6 sirus20x6 marked this pull request as ready for review 169 days ago
sirus20x6
ggerganov
ggerganov commented on 2025-10-13
Normalize f32 helper tails for ggml vec ops
e4189f75
ggerganov
ggerganov approved these changes on 2025-10-14
CISC
slaren
slaren approved these changes on 2025-10-22
slaren slaren merged 19a5a3ed into master 158 days ago
CISC
slaren

Login to write a write a comment.

Login via GitHub

Reviewers
Assignees
No one assigned
Labels
Milestone