BN254 block multiplier (assembly version)#30
Merged
Conversation
Merged
xrvdg
changed the title
Round towards Zero
Support for more architectures and rounding modes.
recmo
approved these changes
May 6, 2025
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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 PR adds inline assembly for bn254 block multipliers. This is a partial replacement for the Rust version as this PR doesn’t include an optimised squaring routine. Even with this the Rust version is useful as a fallback option on non-NEON architectures and as a reference implementation for testing.
Motivation
The motivation for an assembly version is to have predictable performance. With the Rust version we’ve seen that small changes to the source can have big performance differences. Which makes it likely to be susceptible to changes in Rust/LLVM as well.
Concurrent multiplications
Three concurrent multiplication has been the best performing on Raspberry Pi 5 (130ns for 3 multiplication vs 97ns for a single multiplication. On Apple Silicon (M3) we saw an improvement of ~6% running four concurrent multiplication compared to a single multiplication.
Depends on #28
Tracking: #35