This patch detects inband change for the HEVC playback so that we can
recreate decoder if the new config is different from the previous one.
Differential Revision: https://phabricator.services.mozilla.com/D187805
`READ_IN_RANGE_OR_RETURN` causes incorrectly reading bits multiple
times,so we remove it and modify the other macros to ensure that reading
bits should only be triggered once.
Differential Revision: https://phabricator.services.mozilla.com/D189871
Tor Browser (TB) currently ships Android builds built from the ESR 115 branch. This RFP special case to spoof the ESR 115 version on Android was kept in case TB decided to switch to non-ESR Android rapid releases and keep the same 115 version.
This special case is no longer necessary:
* If Android TB sticks with ESR next year, Android TB's version will already be the new ESR version.
* If TB switches from ESR to the rapid releases, Android TB should use the rapid release version because that's the version nearly all other Firefox Android users' UAs report.
Differential Revision: https://phabricator.services.mozilla.com/D190394
Using a static std::unique_ptr introduces atexit destructors which can
cause the thread pool to exit in a non-clean state.
Use StaticAutoPtr to avoid this. Clean-up a bit while at it.
We could move the thread-joining into the TaskController destructor too,
that might be worth doing.
Differential Revision: https://phabricator.services.mozilla.com/D190552
This expands D190144 so we check for actions as well as URLs. I also modified
tests for the affected row types so they test the visibility of the separator.
Differential Revision: https://phabricator.services.mozilla.com/D190494
The refactored assertions were actually slightly more strict that the original
ones and it turns out that we can pass start == end here.
Differential Revision: https://phabricator.services.mozilla.com/D190447
Our current algorithm gives a lot of test failures and test-verfiy failures.
After some logging and debugging, my best guess is that we are just getting
unlucky.
I would see test failures (failed to detect a randomization was applied) when
we e.g. fliped a mere 16 bytes from 0 -> 1. When we flipped some values from
0 -> 2 that would more reliably cause a change.
Differential Revision: https://phabricator.services.mozilla.com/D190283
This patch runs RegisterClassW and CreateWindowW a second time when they
fail, so that we can trace their execution and collect single-step data.
We collect RIP values as we step through user32 code. We also collect
changes in the error state while in user32. We store the collected data
in the stack so that it appears in the crash dumps.
Because this patch only affects behavior after a first failure (which
would normally lead to an immediate crash), it does not alter behavior
for non-crashing users.
Differential Revision: https://phabricator.services.mozilla.com/D189240
This patch changes the mach-try-perf commit message (first line) to be much shorter, and limited to 200 characters. Instead of displaying both the queries, and the actual selections on the same line, this patch splits it into two lines. This provides better formatting options for the queries displayed as well.
Differential Revision: https://phabricator.services.mozilla.com/D190090
QuickAction buttons have their opacity and font-weight hard-coded. We should use our standard variables for that instead. We should also specify a margin for the icon within buttons, as `--urlbarView-icon-margin-start` and `--urlbarView-icon-margin-end` aren't meant for this purpose and add too much space around the icon especially with rich suggestions enabled.
Differential Revision: https://phabricator.services.mozilla.com/D190027
Prepend the mochitest directory listing with `<!DOCTYPE html>` to prevent the document to be loaded in quirks mode and flood the log with console messages.
Differential Revision: https://phabricator.services.mozilla.com/D190544