perf: faster poseidon hades permutation#170
Open
feltroidprime wants to merge 3 commits intostarkware-libs:masterfrom
Open
perf: faster poseidon hades permutation#170feltroidprime wants to merge 3 commits intostarkware-libs:masterfrom
feltroidprime wants to merge 3 commits intostarkware-libs:masterfrom
Conversation
|
Fyi a slightly different version of this will be in |
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
This gives ~7x improvement by not using numpy for vectors/matrices, using native python pow function, and using the mds matrix directly instead of matrix multiplication, taking advantage of the small constants.
This would allow faster trace generation when using cairo-run for programs that uses intensively the Poseidon builtin since the permutation is used in the builtin runner.
Quite a few interesting programs could benefit from it. :)
For ~100k hashes on a i7-12700h (one core), it takes around 13 seconds instead of 90.
The only counterpart is that it is removing generality over the Mds Matrix, but it's not really a problem in our context.
This change is