stylo: Avoid using InterpolateMatrix as a fallback for matched transform function pair
In the current implementation, if there is any interpolation error in a matched
transform function pair, we fall-back to use InterpolateMatrix unconditionally.
However, the error could be caused by:
1. mismatched transform function pair
2. matched transform function pair within at least one undecomposable matrix.
Using InterpolateMatrix for case 1 makes sense, however, using InterpolateMatrix
for case 2 does not. According to the spec, we should just report error for
case 2, and let the caller do the fallback procedure. Using InterpolateMatrix
for case 2 will go through more unnecessary code path, and produce more memory
usage and calculation cost, which should be avoidable.
In this patch, we add an extra pass to check if a transform function pair have
matched operations in advance. With this information, we can easily tell whether
the interpolation error in a equal-length transform function pair is caused by
case 1 or case 2. So, we can avoid the unnecessary cost.
---
<!-- Thank you for contributing to Servo! Please replace each `[ ]` by `[X]` when the step is complete, and replace `__` with appropriate data: -->
- [X] `./mach build -d` does not report any errors
- [X] `./mach test-tidy` does not report any errors
- [X] These changes fix [Bug 1399049](https://bugzilla.mozilla.org/show_bug.cgi?id=1399049)
- [X] These changes do not require tests because the change is for some performance gain, and we have tests to ensure that we won't regress the existing behavior already.
<!-- Reviewable:start -->
---
This change is [<img src="https://reviewable.io/review_button.svg" height="34" align="absmiddle" alt="Reviewable"/>](https://reviewable.io/reviews/servo/servo/18966)
<!-- Reviewable:end -->
Add iterpath and iterdir to CSS test runner.
This allows running a subset of the CSS tests again, by specifying
a directory or single test.
Fixes#18931.
<!-- Reviewable:start -->
---
This change is [<img src="https://reviewable.io/review_button.svg" height="34" align="absmiddle" alt="Reviewable"/>](https://reviewable.io/reviews/servo/servo/18959)
<!-- Reviewable:end -->
Made fire_mouse_event take an enum for event_name
Added an enum with the mouse event options for `fire_mouse_event` and refactored the `event_name` parameter to take the enum as argument.
I also added two options (Leave and Enter) which are noted as todo in the comments.
- [x] `./mach build -d` does not report any errors
- [X] `./mach test-tidy` does not report any errors
- [X] These Changes fix#18943.
- [X] These changes do not require tests because the issue said a clean build suffices.
<!-- Reviewable:start -->
---
This change is [<img src="https://reviewable.io/review_button.svg" height="34" align="absmiddle" alt="Reviewable"/>](https://reviewable.io/reviews/servo/servo/18957)
<!-- Reviewable:end -->
Do not trace Rust values when thread is shutting down.
This addresses a paint point when using debug-mozjs builds. jonco says that it is considered a leak when objects stored in side tables in a SpiderMonkey embedding are traced right before shutting down. This PR adds a per-thread flag that controls whether to run the Rust-side trace hooks, which is automatically toggled before the final GC occurs when destroying a JS runtime.
---
- [x] `./mach build -d` does not report any errors
- [x] `./mach test-tidy` does not report any errors
- [x] These changes fix#18948 and fix#18947.
- [x] These changes do not require tests because we don't use debug-mozjs on CI
<!-- Reviewable:start -->
---
This change is [<img src="https://reviewable.io/review_button.svg" height="34" align="absmiddle" alt="Reviewable"/>](https://reviewable.io/reviews/servo/servo/18952)
<!-- Reviewable:end -->
Include map length in diagnostics
If this doesn't match with the size it's more likely for anything caught
t be hardware corruption. If it does the situation is more interesting.
<!-- Reviewable:start -->
---
This change is [<img src="https://reviewable.io/review_button.svg" height="34" align="absmiddle" alt="Reviewable"/>](https://reviewable.io/reviews/servo/servo/18977)
<!-- Reviewable:end -->
Update h5e version
<!-- Please describe your changes on the following line: -->
---
<!-- Thank you for contributing to Servo! Please replace each `[ ]` by `[X]` when the step is complete, and replace `__` with appropriate data: -->
- [ ] `./mach build -d` does not report any errors
- [ ] `./mach test-tidy` does not report any errors
- [ ] These changes fix #__ (github issue number if applicable).
<!-- Either: -->
- [ ] There are tests for these changes OR
- [ ] These changes do not require tests because _____
<!-- Also, please make sure that "Allow edits from maintainers" checkbox is checked, so that we can help you if you get stuck somewhere along the way.-->
<!-- Pull requests that do not address these steps are welcome, but they will require additional verification as part of the review process. -->
<!-- Reviewable:start -->
---
This change is [<img src="https://reviewable.io/review_button.svg" height="34" align="absmiddle" alt="Reviewable"/>](https://reviewable.io/reviews/servo/servo/18946)
<!-- Reviewable:end -->
In the current implementation, if there is any interpolation error in a matched
transform function pair, we fall-back to use InterpolateMatrix unconditionally.
However, the error could be caused by:
1. mismatched transform function pair
2. matched transform function pair within at least one undecomposable matrix.
Using InterpolateMatrix for case 1 makes sense, however, using InterpolateMatrix
for case 2 does not. According to the spec, we should just report error for
case 2, and let the caller do the fallback procedure. Using InterpolateMatrix
for case 2 will go through more unnecessary code path, and produce more memory
usage and calculation cost, which should be avoidable.
In this patch, we add an extra pass to check if a transform function pair have
matched operations in advance. With this information, we can easily tell whether
the interpolation error in a equal-length transform function pair is caused by
case 1 or case 2. So, we can avoid the unnecessary cost.
Gecko bug: Bug 1399049
Update mozlog dependency
wptrunner repends on mozlog 3.5, but we only require 3.3. Somehow some of the builders have 3.5 installed, but this is causing lots of problems for us.
<!-- Reviewable:start -->
---
This change is [<img src="https://reviewable.io/review_button.svg" height="34" align="absmiddle" alt="Reviewable"/>](https://reviewable.io/reviews/servo/servo/18961)
<!-- Reviewable:end -->
Backed out changeset e64e659c077d: servo PR #18809 and revendor for r…
…eftest failures, e.g. in layout/reftests/bugs/392435-1.html. r=backout on a CLOSED TREE
Backs out https://github.com/servo/servo/pull/18809
Update bitflags to 1.0 in every servo crate
It still needs dependencies update to remove all the other bitflags
versions.
- [x] `./mach build -d` does not report any errors
- [x] `./mach test-tidy` does not report any errors
- [x] These changes do not require tests because it's a dependency update
<!-- Reviewable:start -->
---
This change is [<img src="https://reviewable.io/review_button.svg" height="34" align="absmiddle" alt="Reviewable"/>](https://reviewable.io/reviews/servo/servo/18809)
<!-- Reviewable:end -->
