Added playout-delay RTP header extension for video to JsepSessionImpl::SetupDefaultRtpExtensions.
This ensures that this extension will be preserved when generating an answer to an offer (by a sending peer) containing it.
Also updated JsepSessionTest to include a test verifying that the expected default audio and video extensions are set.
Differential Revision: https://phabricator.services.mozilla.com/D47689
--HG--
extra : moz-landing-system : lando
In many cases, we call enterTaggedTuple in situations in which we already know that there is only one possible value, which ends up encoded on 0 bits in Context. This patch lets us entirely skip the lookup.
Differential Revision: https://phabricator.services.mozilla.com/D48524
--HG--
extra : moz-landing-system : lando
There is a window inside of SendPacket where the critical section is released
which means that other code could set paused_. This would lead to us hitting
the assertion at the top of SendPacket. Checking paused_ in the while loop
will avoid this. Upstream has fixed this problem in a similar way, but the
code has changed there enough that it doesn't make sense to try to bring in
their fix directly.
Differential Revision: https://phabricator.services.mozilla.com/D48540
--HG--
extra : moz-landing-system : lando
This patch does several things. Because Workers aren't on the main thread,
many of the things done are in the name of off main thread access.
1) Changes a parameter in IsEvalAllowed from a nsIPrincipal to a bool.
We only used the principal to determined if it was the System Principal.
Principals aren't thread safe and can only be accessed on Main Thread, so
if we passed a Principal in, we would be in error. Instead only pass in
the bool which - for workers - comes from a thread-safe location.
2) Separates out the Telemetry Event Recording and sending a message to the
console into a new function nsContentSecurityUtils::NotifyEvalUsage. (And
creates a runnable that calls it.)
We do this because we will need to only call this method on the main thread.
Telemetry Event Recording has only ever been called on the Main Thread.
While I possibly-successfully cut it over to happen Off Main Thread (OMT)
by porting preferences to StaticPrefs, I don't know if there were other
threading assumptions in the Telemetry Code. So it would be much safer to
just continue recording Event Telemetry on the main thread.
Sending a message to the console requires calling GetStringBundleService()
which requires main thread. I didn't investigate if this could be made
thread-safe, I just threw it onto the main thread too.
If, in IsEvalAllowed, we are on the main thread - we call NotifyEvalUsage
directly. If we are not, we create a runnable which will then call
NotifyEvalUsage for us on the main thread.
3) Ports allow_eval_with_system_principal and allow_eval_in_parent_process
from bools to RelaxedAtomicBool - because we now check these prefs OMT.
4) In RuntimeService.cpp, adds the call to IsEvalAllowed.
5) Add resource://gre/modules/workers/require.js to the allowlist of eval
usage. This was the script that identified this gap in the first place.
It uses eval (twice) for structural reasons (scope and line number
massaging.) The contents of the eval are the result of a request to a
uri (which may be internal, like resource://). The whole point of this
is to implement a CommonJS require() api.
This usage of eval is safe because the only way an attacker can inject
into it is by either controlling the response of the uri request or
controlling (or appending to) the argument. If they can do that, they
are able to inject script into Firefox even if we cut this usage of eval
over to some other type of safe(r) script loader.
Bug 1584564 tracks making sure calls to require.js are safe.
6) Adds cld-worker.js to the allowlist. Bug 1584605 is for refactoring that
eval usage, which is decidedly non-trivial.
7) Does _not_ enforce the eval restrictions for workers. While I've gotten
try to be green and not throw up any instances of eval-usage by workers,
it is much safer to deploy this is Telemetry-only mode for Workers for
a little bit to see if anything pops up from the Nightly population.
Bug 1584602 is for enforcing the checks.
Differential Revision: https://phabricator.services.mozilla.com/D47480
--HG--
extra : moz-landing-system : lando
Also rename existing utility functions to reflect that they are for VP8 only.
Differential Revision: https://phabricator.services.mozilla.com/D47369
--HG--
extra : moz-landing-system : lando
What we actually care about here is whether itemRect is empty bceause that's
the what we'll use for the actual surface size.
Differential Revision: https://phabricator.services.mozilla.com/D48548
--HG--
extra : moz-landing-system : lando
This should fix the doc builds on Windows, as sphinx-js added Windows support in 2.3.1 and 2.4. We also now get support for variadic args, @deprecated, and @see, along with other features.
sphinx-js 2.7.1 changed the default cwd to be the one containing conf.py, so I also had to twiddle `jsdoc_config_path`.
Let some other pipenv pinnings update themselves as well, as, if I don't, they'll just update themselves the next time somebody runs `mach doc`, dirtying their tree.
I suspect this also fixes bug 1556460, whose equivalent bug in sphinx-js is https://github.com/mozilla/sphinx-js/issues/106. IOW, it should no longer break with versions of jsdoc >= 3.6.
Differential Revision: https://phabricator.services.mozilla.com/D48122
--HG--
extra : moz-landing-system : lando
With fission, we most likely have a process switch and the existing worker
target isn't properly detached. We should ensure releasing the SW whenever
the connection to the server drops
Differential Revision: https://phabricator.services.mozilla.com/D48061
--HG--
extra : moz-landing-system : lando
The original plan for the node-picker to work with multiple targets was introduced in D41598 (in bug 1568825). The idea was that, because we can have multiple independent inspectable targets, and because the client is the one doing the orchestration between them, to let the client start the node picker in all targets at once.
At the time of this first change, the code was create with this in mind, but there was really just one target (the top-level one).
So, this revision introduces the real code for this. First of all, I removed the now obsolete `getAllInspectorFronts` in `node-picker.js` because we now have a similar function on the inspector front directly.
Then the main code changes to look for are on the actor side, in the `HighlighterActor`. This is where the picking actually happens.
You have to remember that several targets will be picking at the same time, and therefore several `HighlighterActor` instances will be in pick mode at the same time.
The way they allow users to pick is by listening to mouse events (mousemove and clicks essentially).
Because these actors can't see or talk to each other, one can't tell the others that the mouse is now over its content and the other pickers should pause somewhat.
So, when one of them sees that the mouse event is happening on a remote frame, then it bails out and lets events through without handling them. This is so that the embedded document (which also has a picker running) can get a chance to receive the mouse events too.
The other aspect is that each `HighlighterActor`, when picking, does its own highlighting. So if there are 3 remote frames, then there really are 3 highlighters.
So the trick is to make sure only one of them ever appears at any given time. Again, these actors can't talk to each other directly, so the client is responsible for doing this when receiving events that a node was hovered.
This is not perfect, but should normally get far better when the new fission-compatible highlighter is in place. Indeed, when that happens, we won't have to care about this anymore, there will be only one `HighlighterRenderer` for the entire tab. So even if there are multiple `HighlighterActor` instances picking, they will all be sending events to the same renderer.
The only exception is in the browser toolbox where you can inspect both the browser UI and the content UI. In this case, there will be 2 renderers: one over the entire browser window, and one over the <browser> area. So we'll still have to do the dance of hiding one when the other is shown.
Differential Revision: https://phabricator.services.mozilla.com/D42640
--HG--
extra : moz-landing-system : lando
Harry Twyford
Emma Malysz
Noemi Erli
Andreas Schuler
Dylan Roeh
Daosheng Mu
Matthew Gregan
Ting-Yu Lin
Michael Kaply
David Teller
Doug Thayer
Dan Minor
Cosmin Sabou
Gabriel Luong
Brian Grinstead
Kevin Jacobs
Tom Ritter
Alexander Surkov
John Lin
Andrew Swan
Oana Pop Rus
Edouard Oger
Jeff Muizelaar
Junior Hsu
James Graham
Erik Rose
Andreea Pavel
Nicolas Silva
Razvan Maries
Alexandre Poirot
Patrick Brosset
Tom Tung
Robert Strong