Simplifies the concurrent operations of wifi scanning and reduces the
frequency in common cases. Wifi scanning when on mobile is reduced from
every 5 seconds to every minute. Wifi scans will also happen whenever
a new listener is registered.
Differential Revision: https://phabricator.services.mozilla.com/D176200
Reorganizes the wifi-scanning code and makes more of it platform-generic
to ease the transition from polling the wifi to usually scanning only on
network changes. This is essentially just moving files/code around and
promoting nsWifiMonitor::DoScan to be platform-independent.
Differential Revision: https://phabricator.services.mozilla.com/D176199
Add a separate check for spec breaking allows of certain MIME
types. Having this separated out means that we can make the rest of
the implementation behave exactly like spec.
Some tradeoffs that we need in the current state are:
* Allowing "application/dash+xml"
* Allowing "application/vnd.apple.mpegurl"
* Allowing "text/vtt"
* Allow all MIME types beginning with "audio/mpeg"
* Allow "text/plain" when there is a no-sniff header.
Differential Revision: https://phabricator.services.mozilla.com/D176821
We make sure to not send any data to the content process in case of
fetching an opaque resource. This is way to remain more web
compatible, but is also in conflict with the ORB specification.
Differential Revision: https://phabricator.services.mozilla.com/D173454
creating streams is allowed before a transport is ready, and should fail
with InvalidStateError if the stream goes to closed or failed.
Differential Revision: https://phabricator.services.mozilla.com/D177108
After PFetch is enabled, fetch() call in workers will not create a channel in the content process anymore.
Although netmonitor also watches the channels in the parent process, the created channel still loses the BrowsingContext information for netmonitor to connect the network event and the channel.
In P1, https://phabricator.services.mozilla.com/D174249, we propagate the BrowsingContext ID through PFetch.
In this patch, we need to save it in channel's LoadInfo for netmonitor.
In order not to confuse with nsILoadInfo's BrowsingContextID, we create a new attribute WorkerAssociatedBrowsingContextID in nsILoadInfo.
Depends on D174249
Differential Revision: https://phabricator.services.mozilla.com/D174441
These registrars can hold channels alive, which can hold listeners alive,
which can hold cycle collected things alive. By clearing the registrars
before the final CC instead of after the final CC, we can avoid spurious
leaks.
It would make more logical sense to do this earlier, at net teardown,
but I've made the minimal change to try to avoid stirring up weird
shutdown issues.
Differential Revision: https://phabricator.services.mozilla.com/D177182
This test was intermittent because sometimes we'd pick up the rcode for the A request, and sometimes for the AAAA request.
This test change makes sure we always have a consistent value for the skip reason.
Differential Revision: https://phabricator.services.mozilla.com/D177150
After PFetch is enabled, fetch() call in workers will not create a channel in the content process anymore.
Although netmonitor also watches the channels in the parent process, the created channel still loses the BrowsingContext information for netmonitor to connect the network event and the channel.
In P1, https://phabricator.services.mozilla.com/D174249, we propagate the BrowsingContext ID through PFetch.
In this patch, we need to save it in channel's LoadInfo for netmonitor.
In order not to confuse with nsILoadInfo's BrowsingContextID, we create a new attribute WorkerAssociatedBrowsingContextID in nsILoadInfo.
Depends on D174249
Differential Revision: https://phabricator.services.mozilla.com/D174441
In Bug 1596576 we will disable by default the forcing of tcp send buffer sizes on http/2 uploads.
This provides a mechanism for users to test early.
Differential Revision: https://phabricator.services.mozilla.com/D176927
In Bug 1596576 we will disable by default the forcing of tcp send buffer sizes on http/2 uploads.
This provides a mechanism for users to test early.
Differential Revision: https://phabricator.services.mozilla.com/D176927
GeckoView supports resource://android/ URL to access package root. But when
using this URL, `nsDocShellLoadState::ValidateWithOriginalState` returns URI
mismatch because `SubstitutingJARURI` is serialized to other type.
So `SubstitutingJARURI` should be serialized to same type.
To serialize this via IPC, we return Mutator type. But this doesn't allow
to modify properties except for IPC serialization. But some Set* can return
`NS_OK` when having same value for `ContentPrincipal::GetSiteOriginNoSuffix`.
Differential Revision: https://phabricator.services.mozilla.com/D175765
Kershaw Chang
dylan
Cristian Tuns
David Parks
Andreas Farre
Valentin Gosu
Em Zhan
Noemi Erli
ffxbld
Randell Jesup
Eden Chuang
Andrew McCreight
Jan Varga
Iulian Moraru
Natalia Csoregi
Andrew Creskey
Robert Longson
Makoto Kato
Bilal
Narcis Beleuzu
Stanca Serban
Geoff Lankow