Conversation
Rewrite float construction to be more legible and properly account for the sticky bit. Previous implementation was truncating it, giving trouble on some values.
There is an implicit leading bit in the `precisionBits`, which we must account for when specifying the max output digits of `dragon4`. Since all problems with float parsing _and_ rendering are now resolved, we re-enable tho float fuzztests.
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Both parsing and rendering of
@r*floats had some precision issues. For some values, the parsed bigint would be off-by-one. For some values, the rendered literal would be off-by-one-fraction. This broke round-tripping, and was mismatching with hoon behavior.Here, we rewrite
construct()wholesale. This gains us legibility and lets us fix the root cause elegantly, rather than patching over the problem with a spot-the-mistake branch in the code.(Arguably legibility gains don't matter since this is essentially write-once code. I haven't tested for perf, but never did that that for the old implementation either.)
We also tweak the "output size" argument we pass to
dragon4()so that we render at the correct, expected precision.We add specific regression test cases for this, and re-enable fuzzing for float auras.