Revert LLAMA_NATIVE to OFF in flake.nix #5066
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Collaborator
Oh yes, we surely would prefer that |
SomeoneSerge
approved these changes
Jan 21, 2024
SomeoneSerge
added
the
nix
philiptaron
reviewed
Jan 22, 2024
Collaborator
philiptaron
left a comment
philiptaron
left a comment
There was a problem hiding this comment.
Dang, that's my fault in doing the transcription. Call it an "off by on" error 😅 . Thanks for the PR; LGTM.
jordankanter
pushed a commit
to jordankanter/llama.cpp
that referenced
this pull request
Feb 3, 2024
hodlen
pushed a commit
to hodlen/llama.cpp
that referenced
this pull request
Apr 1, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
I've noticed that since PR #4605, performance (CPU-only) took a massive dive when using the Nix flake (I went from ~4 tokens/s to <0.5). It seems that the slowdown is caused by
LLAMA_NATIVE=ON. Reverting toOFF(as it was before the PR) restores the expected performance.This regression was observed on both an i7-1165G7 and a Ryzen 3800X running NixOS.
FWIW, the llama-cpp package in nixpkgs has
LLAMA_NATIVE=OFF.I'm not sure what the implications of turning off
LLAMA_NATIVEare, maybe @philiptaron and @SomeoneSerge want to chime in.