This patch contains three changes:
1. a local version of glb.iglob in vendor_manifest.py that always
matches hidden files for '**'. I've tested it on Python 3.5 up to
3.12 and it works as expected.
2. a cleanup of many moz.yaml files that got convluted because of the
previous behavior of '**'
3. removal of a few useless files that are now caught by the '**'
globbing.
Differential Revision: https://phabricator.services.mozilla.com/D190195
We hit some limitations of the globbing system in exclude file, which
relies on glob.iglob and skips dotfiles...
As a side effect update the license to the one actually shipped with
libfuzzer.
Differential Revision: https://phabricator.services.mozilla.com/D190068
We hit some limitations of the globbing system in exclude file, which
relies on glob.iglob and skips dotfiles...
As a side effect update the license to the one actually shipped with
libfuzzer.
Differential Revision: https://phabricator.services.mozilla.com/D190068
Added my notes from when I was reproducing an IPC test failure.
* fix deprecation warning in fuzzfetch
* fix one spelling mistake
* add example command for IPC test case reproduction
Differential Revision: https://phabricator.services.mozilla.com/D186708
psutil is a complex requirements and we don't want to rebuild it,
because it's a native requirements. So use the same requirement as the
one we use to build python.
Differential Revision: https://phabricator.services.mozilla.com/D177487
This method always returned GetMainThreadSerialEventTarget(). This patch
switches all callers over to use that method instead.
We can't easily switch all calls to be calls to NS_GetMainThread(), as there is
no version of that method returning a bare nsIThread* instance.
I didn't introduce one, as we may want to add a lock around mMainThread in the
future, which would require removing nsThreadManager::GetMainThreadWeak. As
this method only returns nsISerialEventTarget, it method could remain
implemented, however, by returning a statically allocated fake event target
which forwards dispatches (and QIs to nsIThread) to the real main thread.
Differential Revision: https://phabricator.services.mozilla.com/D166608
This method always returned GetMainThreadSerialEventTarget(). This patch
switches all callers over to use that method instead.
We can't easily switch all calls to be calls to NS_GetMainThread(), as there is
no version of that method returning a bare nsIThread* instance.
I didn't introduce one, as we may want to add a lock around mMainThread in the
future, which would require removing nsThreadManager::GetMainThreadWeak. As
this method only returns nsISerialEventTarget, it method could remain
implemented, however, by returning a statically allocated fake event target
which forwards dispatches (and QIs to nsIThread) to the real main thread.
Differential Revision: https://phabricator.services.mozilla.com/D166608
`SyncRunnable`'s helper functions take an `nsIRunnable *`; but the most
common way of building nsIRunnables, `NS_NewRunnableFunction`, returns
an `already_AddRefed<nsIRunnable>` instead. Add two new overloads of the
helper functions to eliminate the impedance mismatch.
(This does result in an uncomfortable amount of code duplication. While
we could eliminate that with appropriate use of SFINAE, it'll be simpler
if we wait for C++20 and its `requires` keyword.)
Additionally, add two explicitly-deleted overloads to catch and prevent
a previously-common antipattern that presumably resulted from this type
mismatch: accidentally wrapping the actual runnable in two layers of
`SyncRunnable`. Fix the former use-sites appropriately. (This was
probably harmless, but is also probably best avoided.)
No functional changes. This is in some sense a continuation of bug
1281626.
(This is no longer actually relevant to bug 1772908 due to a different
approach being taken. It remains in the patchset anyway, for
simplicity's sake.)
Differential Revision: https://phabricator.services.mozilla.com/D157131