Fix double-rounding bug in double -> (b)float16 casts#8906
Merged
Conversation
Contributor
|
Interesting find! Funny to see how much code changes were required to fix one bit. |
Also add and fix some comments
Member
Author
|
I refactored a bit to share more code between coming from 32-bit and 64-bit, and added some more explanatory comments (the constants were a bit mysterious to me) |
Member
Author
|
Failure unrelated. ptal |
alexreinking
approved these changes
Jan 16, 2026
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
3 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Narrowing from a double to a (b)float16 was done via float, which is incorrect due to double-rounding. The case where it's wrong is captured in the new test case.
This PR fixes it by adding new routines to go directly from double to (b)float16. It required adding a strict_cast strict float intrinsic, to prevent bit-inexact fusion of narrowing casts.
This bug was discovered while working on emulating float16 fmas. You can do them as a double fma, but only if the narrowing cast is correct!