-
Notifications
You must be signed in to change notification settings - Fork 331
Comments
Implement Fast CIOS for Montgomery modular multiplication#1009
Open
clementjuventin wants to merge 2 commits intoipsilon:master from
Open
Implement Fast CIOS for Montgomery modular multiplication #1009clementjuventin wants to merge 2 commits intoipsilon:master from
clementjuventin wants to merge 2 commits intoipsilon:master from
Conversation
@clementjuventin
clementjuventin
force-pushed
the
fast-montgomery-cios
branch
from
September 19, 2024 15:33
b693f02 to
79ad43b
Compare
Author
After further investigation, I found another way of ordering branches (second commit) and obtained what we were looking for (~15% gain on evmmax_mul<uint256, bn254>)!
evmmax_mul<uint256, bn254>_median -0.1529 -0.1529 27 23 27 23
evmmax_mul<uint256, secp256k1>_median +0.0059 +0.0058 28 28 28 28
I still don't get why this implementation is better, even assuming branch prediction makes the checks insignificant.
I would also like to compare the assembly code in case there is important optimization under the hood but I never did so let's see if I manage to do it
@clementjuventin
clementjuventin
force-pushed
the
fast-montgomery-cios
branch
from
September 19, 2024 15:58
79ad43b to
1d3aab1
Compare
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Uh oh!
There was an error while loading. Please reload this page.
This PR implements the improvement of the CIOS variant of the Montgomery algorithm showcased in the following document: EdMSM: Multi-Scalar-Multiplication for SNARKs and Faster Montgomery multiplication and proposed in #869.
The optimization occurs when
most_significant_p_word < (word_size / 2),pis represented by the variablemodin the current implementation.Here are the benchmarks performed after applying these changes starting from commit 01eca77.
Build:
cmake --build build --parallelBenchmarks:
taskset -c 0 ./build/bin/evmone-bench-internal --benchmark_filter=evmmax* --benchmark_repetitions=10 --benchmark_format=json --benchmark_out=cios_classic.jsonComparison:
As you can see, the results do not indicate a performance improvement.