Change Line::nearest implementation to be branchless
#505
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
tomcur
commented
Sep 14, 2025
| // Clamp the parameter to be on the line segment. This results in `t==0` if `t==inf` above. | ||
| #[expect( | ||
| clippy::manual_clamp, | ||
| reason = "`clamp` uses slightly more instructions than chained `max` and `min` on x86 and aarch64" |
Member
Author
There was a problem hiding this comment.
aarch64: https://godbolt.org/z/Mv4Y19jrM
x86: https://godbolt.org/z/WW19hMd6v
Perhaps we should disable this lint in kurbo.
Member
There was a problem hiding this comment.
Huh. I guess clamp can't really be improved because of how its NaN behavior is specified.
Member
Author
There was a problem hiding this comment.
Seems that way! (As a separate note, we actually use some of the NaN semantics of the chained-min-max in the analytical pixel coverage calculation in the sparse strips work, so this lint can be a bit noisy.)
jneem
approved these changes
Sep 14, 2025
This is a branchless calculation of a point-onto-line-segment projection. Also marks the method `#[inline]`. I started looking at this because of the math we introduced in linebender/vello#1214. This might allow us to introduce some tighter bounds there without introducing additional branching. This should be useful here regardless of whether the above happens.
Co-authored-by: jneem <joeneeman@gmail.com>
tomcur
force-pushed
the
branchless-line-nearest
branch
from
fb967fb to
e8baf8a
Compare
github-merge-queue bot
pushed a commit
that referenced
this pull request
Sep 20, 2025
Missed in #505.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 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Outdated suggestions cannot be applied.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.
This is a branchless calculation of a point-onto-line-segment projection. Also marks the method
#[inline]and adds some testing.I started looking at this because of the math we introduced in linebender/vello#1214. This might allow us to introduce some tighter bounds there without introducing additional branching.
This should be useful here regardless of whether the above happens.