Only enable sgemm for prompt processing, not for inference #9330
Conversation
This file contains 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
slaren
approved these changes
Sep 7, 2024
4 tasks
dsx1986
pushed a commit
to dsx1986/llama.cpp
that referenced
this pull request
Oct 29, 2024
arthw
pushed a commit
to arthw/llama.cpp
that referenced
this pull request
Nov 15, 2024
arthw
pushed a commit
to arthw/llama.cpp
that referenced
this pull request
Nov 18, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
While sgemm/tinyblas was designed to speed up prompt processing using tiled matrix multiplications, llama.cpp also calls it for inference as a 1x1 computation. Personally I think it makes more sense for us to use our dedicated
ggml_vec_dotfunctions for the inference dot products and leave sgemm for prompt processing only. We can optimize each one for its respective purpose and so forth.See my PR #8049 for an example where sgemm has faster prompt processing while
ggml_vec_dothas faster inference.