Optimize l2_norm_f32 op with SIMD #14970

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Open

TIKki43 wants to merge 1 commit into ggml-org:master from TIKki43:perf/l2-norm-optimize

+6 −12

TIKki43 commented Jul 30, 2025

I added an optimized version of ggml_compute_forward_l2_norm_f32 op in ops.cpp.

I replaced naive scalar loop with ggml_vec* operations based on SIMD.

Passed all test-backend-ops test -o L2_NORM -b CPU
Verifed correctness
Benchmarked perfomance and got 6% more time efficiency on function

I hope this helps^)


          Optimize l2_norm_f32 op with SIMD

f51b116

github-actions bot added the ggml label

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

ggml