Fix fptrunc Float64 -> Float16 rounding through Float32 #57809
Merged
Conversation
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
Widening from Float32 to Float64 and then rounding (by any method) to Float16 will not introduce any error, but going from Float64 -> Float32 -> Float16 will round incorrectly if the intermediate Float32 is halfway between two Float16s. Fixes JuliaLang#57805. Co-authored-by: Jameson Nash <jameson@juliacomputing.com>
giordano
added
bugfix
oscardssmith
approved these changes
Mar 18, 2025
Keno
reviewed
Mar 18, 2025
| #define fintrinsic_write_float16(p, x) *(uint16_t *)p = float_to_half(x) | ||
| #define fintrinsic_write_bfloat16(p, x) *(uint16_t *)p = float_to_bfloat(x) | ||
| #define fintrinsic_write_float16(p, x) *(uint16_t *)p = double_to_half(x) | ||
| #define fintrinsic_write_bfloat16(p, x) *(uint16_t *)p = double_to_bfloat(x) |
There was a problem hiding this comment.
I don't think the same logic applies to bfloat16. Since bfloat16 is just a truncated float32, shouldn't it always be legal to go through Float32? If not, it'd be good to have a test.
There was a problem hiding this comment.
According to the double_to_bfloat code, the same correction to the rounding direction is required whenever the float32 (after truncation from float64) is exactly halfway between 2 bfloat16 values, but wasn't required for the subnormal case (as was demonstrated in the original issue for float16)
KristofferC
pushed a commit
that referenced
this pull request
Mar 20, 2025
Widening from Float32 to Float64 and then rounding to Float16 will not introduce any error, but going from Float64 -> Float32 -> Float16 will round incorrectly if the intermediate Float32 is halfway between two Float16s. Fixes #57805. Thanks to @vtjnash for suggesting the fix. Co-authored-by: Jameson Nash <jameson@juliacomputing.com> (cherry picked from commit a676b12)
KristofferC
removed
the
backport 1.12
71 tasks
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.
Widening from Float32 to Float64 and then rounding to Float16 will not introduce any error, but going from Float64 -> Float32 -> Float16 will round incorrectly if the intermediate Float32 is halfway between two Float16s.
Fixes #57805.
Thanks to @vtjnash for suggesting the fix.