Skip invoking bindgen if no header changes
This can avoid doing bindgen when build script is called for updating other files, e.g. properties.
This uses a global modified time, so there is a chance that some of the files which can be skipped but not skipped. But given that we do all three files in parallel, that would unlikely affect the actual runtime.
Using lots of `Mutex` could be an issue, but it doesn't seem to be in practice. Since only one thread would hold the lock of `ADDED_PATHS`, there is never a competitor for the lock of `LAST_MODIFIED`.
r? @emilio
<!-- 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/14754)
<!-- Reviewable:end -->
Add asserts to make failure clearer
Gecko currently doesn't pass absolute path for `MOZ_DIST`, which leads to obscure panic when running the build script. This patch adds some assertions so that failures around this would be clearer.
r? @emilio
<!-- 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/14563)
<!-- Reviewable:end -->
I noticed that our current behavior in ContentRangeInserted is incorrect. Unlike
ContentInserted (where this code lived originally), ContentRangeInserted takes a
start and end element. I'm not sure if we ever take that path for new content that
needs style, but it seemed sketchy. And generally, it seems nice to just always
style new content the same way (though we still need to style NAC by the subtree
root, since it hasn't been attached to the parent yet).
For situations where there is indeed only one unstyled child, the traversal
overhead should be neglible, since we special-case the single-element in
parallel.rs to avoid calling into rayon.
Being more explicit about what we want here also makes us more robust against
the other handful of callpaths that can take us into
nsCSSFrameConstructor::{ContentRangeInserted,ContentAppended}. Currently we
can call StyleNewSubtree on an already-styled element via RecreateFramesForContent,
which triggers an assertion in the servo traversal.
MozReview-Commit-ID: DqCGh90deHH
Majority of build_gecko.rs is just the straightforward conversion from
regen.py. There are two differences that:
1. Side in whitelist is changed to mozilla::Side
2. std::atomic__My_base is added to opaque types for Windows
