Add more unit tests for rounding values to pixel grid #1776
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
|
The latest updates on your projects. Learn more about Vercel for Git ↗︎
|
|
This pull request was exported from Phabricator. Differential Revision: D67712673 |
rshest
added a commit
to rshest/yoga
that referenced
this pull request
Dec 30, 2024
Summary: # Changelog: [Internal] - I was looking into some profiling results and noticed that Yoga has some quirky custom logic when it comes to rounding to the pixel grid, especially for the values that are halfway through. There are already unit tests in place, but those cases weren't covered, so I am adding these extra corner cases here, to make sure they are covered in case that the actual rounding function may get modified (which it might make sense to, as it's currently can be more expensive than it could have been in some scenarios). Differential Revision: D67712673
rshest
force-pushed
the
export-D67712673
branch
from
ba3b385 to
2c2c685
Compare
|
This pull request was exported from Phabricator. Differential Revision: D67712673 |
) Summary: X-link: facebook/litho#1036 X-link: facebook/react-native#48404 ## Changelog: [Internal] - This popped up when profiling some heavy UI performance, calling `fmod` operation in Yoga's `roundLayoutResultsToPixelGrid` in `PixelGrid.cpp` can be expensive, furthermore it turns out that some of the calls were redundant. This replaces the duplicate calls to fmod with an equivalent single round operation, which for e.g. clang compiler on Windows brings the code in question from ~50 instructions (including 4 call instructions to the fmod function) down to ~30 instructions (without any external calls), and the layout operation being **~1% more efficient** for the particular benchmark I was looking into. Differential Revision: D67689065
Summary: # Changelog: [Internal] - I was looking into some profiling results and noticed that Yoga has some quirky custom logic when it comes to rounding to the pixel grid, especially for the values that are halfway through. There are already unit tests in place, but those cases weren't covered, so I am adding these extra corner cases here, to make sure they are covered in case that the actual rounding function may get modified (which it might make sense to, as it's currently can be more expensive than it could have been in some scenarios). Differential Revision: D67712673
rshest
force-pushed
the
export-D67712673
branch
from
2c2c685 to
76ab9b8
Compare
|
This pull request was exported from Phabricator. Differential Revision: D67712673 |
|
This pull request has been merged in 9591210. |
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.
Summary:
Changelog:
[Internal] -
I was looking into some profiling results and noticed that Yoga has some quirky custom logic when it comes to rounding to the pixel grid, especially for the values that are halfway through.
There are already unit tests in place, but those cases weren't covered, so I am adding these extra corner cases here, to make sure they are covered in case that the actual rounding function may get modified (which it might make sense to, as it's currently can be more expensive than it could have been in some scenarios).
Differential Revision: D67712673