First stab at multiscalar speedup for computing group commitments #345
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
|
This is failing only for ed448 but I'm not sure why yet |
Oh right our vartime multiscalar mul only supports 256bit scalar fields 🙃 |
dconnolly
commented
May 9, 2023
conradoplg
reviewed
May 9, 2023
There was a problem hiding this comment.
Looks good, we just need to fix the 448 support probably by copying the refactored code from reddsa
Results on my machine:
FROST Signing ristretto255/Round 2/2
time: [94.596 µs 94.901 µs 95.243 µs]
change: [-16.059% -15.648% -15.236%] (p = 0.00 < 0.05)
Performance has improved.
FROST Signing ristretto255/Aggregate/2
time: [204.35 µs 204.96 µs 205.62 µs]
change: [-8.1068% -7.9390% -7.7735%] (p = 0.00 < 0.05)
Performance has improved.
FROST Signing ristretto255/Round 2/7
time: [182.08 µs 182.27 µs 182.52 µs]
change: [-48.358% -48.293% -48.225%] (p = 0.00 < 0.05)
Performance has improved.
FROST Signing ristretto255/Aggregate/7
time: [289.16 µs 289.31 µs 289.49 µs]
change: [-37.323% -37.154% -37.004%] (p = 0.00 < 0.05)
Performance has improved.
FROST Signing ristretto255/Round 2/67
time: [1.2781 ms 1.2819 ms 1.2868 ms]
change: [-60.825% -60.739% -60.650%] (p = 0.00 < 0.05)
Performance has improved.
FROST Signing ristretto255/Aggregate/67
time: [1.3150 ms 1.3219 ms 1.3318 ms]
change: [-60.843% -60.580% -60.303%] (p = 0.00 < 0.05)
Performance has improved.
FROST Signing ristretto255/Round 2/667
time: [12.757 ms 12.763 ms 12.770 ms]
change: [-61.301% -61.273% -61.246%] (p = 0.00 < 0.05)
Performance has improved.
FROST Signing ristretto255/Aggregate/667
time: [12.455 ms 12.473 ms 12.495 ms]
change: [-61.931% -61.868% -61.793%] (p = 0.00 < 0.05)
Performance has improved.
conradoplg
approved these changes
May 16, 2023
Codecov ReportPatch coverage:
Additional details and impacted files@@ Coverage Diff @@
## main #345 +/- ##
==========================================
+ Coverage 68.58% 68.79% +0.20%
==========================================
Files 30 30
Lines 2394 2410 +16
==========================================
+ Hits 1642 1658 +16
Misses 752 752
☔ View full report in Codecov by Sentry. |
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.
The group commitment is accumulated for all the hiding commitments, saving the per-participant binding factor and their respective binding commitment for one big vartime multiscalar multiplication at the end, which is finally added to the accumulated commitment as the final group commitment.
This leads to nearly a 50% speedup at the 667/1000 end.