Skip to content

Conversation

@xctan
Copy link
Collaborator

@xctan xctan commented Jun 16, 2025

Fixes #14134.

Due to the symbol resolution behavior of linkers with static archives (notably bfd), the order of object files is critical. The previous order caused accelerated kernels to be "shadowed" by their generic implementations, as the generic versions were encountered first.

This change reorders the object files in CMakeLists.txt to ensure generic implementations are listed last, guaranteeing that the accelerated kernels are correctly linked.

@github-actions github-actions bot added the ggml changes relating to the ggml tensor library for machine learning label Jun 16, 2025
@xctan xctan marked this pull request as draft June 16, 2025 06:57
@jeffzhou2000
Copy link

jeffzhou2000 commented Jun 16, 2025

works perfectly in my local dev envs(aarch64 based Android phone).

@xctan xctan closed this Jun 16, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

ggml changes relating to the ggml tensor library for machine learning

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Misc. bug: Performance regression on aarch64 q4_0

2 participants