-
Notifications
You must be signed in to change notification settings - Fork 40
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Refactor for Increased Safety without Performance Impacts #135
Merged
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 converts the core dragonbox algorithm for the float decimal writers to be completely safe. This ensures effectively all non-trivial uses of unsafe that could cause unsoundness are removed for 99% of users.
This causes ~0.5-1% regression in performance but this is well within an acceptable range for better safety guarantees. Most of the bounds checking is ellided by the compiler.
This in general for simple cases due to slight refactoring improves performance, however, it does have a noticeable (+2-7%) penalty in some corner cases which invoke the close-to-halfway rounding cases. That said, performance of this implementation of the Dragonbox algorithm without any unsafety always outperforms Ryu, which means it's still the fastest implementation out there even with slight regressions in corner cases.
This is a relatively trivial change and causes a small performance hit, but that change should be able to be mitigated with some minor refactoring.
One specific benchmark however has significant regressions, the random:big_ints/write_f32_lexical benchmark, while the f64 version does not. The benchmark is still considerably faster than ryu or fmt, however.
This does have some performance penalties but performance is still excellent, and so a few patches will need to be made so our compiler can elide more checks.
Performance enhancements will soon follow to restore any lost performance in some cases which impacted it up to 3%.
03f2baf
to
290e8ce
Compare
290e8ce
to
7df1e74
Compare
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.
This introduces numerous different layers of security enhancements:
Closes #100.