This patch makes the size of inner windows will be automatically rounded for
either window.open() with window features or setting window size through
innerWidth/Height and outerWidth/Height when fingerprinting resistance is
enabled. If the given value is greater the maximum available rounded size, then
it will be set to the maximum value. Otherwise, the size will be set to the
nearest upper 200x100.
This patch also adds one helper function in nsContentUtils for calculating the
rounded window dimensions.
MozReview-Commit-ID: J2r3951vuNN
--HG--
extra : rebase_source : a44b19bdf2ce7e90fc831ddc2b85a86d594cb0c3
This patch fixes two related issues.
1. The AutoStopwatch uses a stack-allocated `mozilla::Vector` to
communicate with its callback during each compartment switch. This
vector was designed to allow its contents to be stack-allocated but
they turned out to be accidentally heap-allocated.
2. During each tick, the stopwatch fills a vector
`recentGroups_`. This vector always started with minimal capacity and
had to grow repeatedly as groups were added, causing repeated
reallocations. This patch preallocates `recentGroups_` to have the
same capacity as the previous tick. We expect that this should
eventually reach a stable size that closely matches the actual needs
of the process.
MozReview-Commit-ID: A7e3HNdSuML
--HG--
extra : rebase_source : bb205970b4186d9325dded7ddae066bcad18fb0e
To improve build times, we've decided to only build libxul-gtest when
it's specifically requested, i.e. via `mach gtest` or similar. However,
our build setup specifies that gkrust-gtest, the Rust code specifically
for libxul-gtest, is built during normal builds. This library takes a
significant amount of time to compile and is generally not needed.
This patch changes the gkrust-gtest library to only be visible under the
same conditions that libxul-gtest is built, thus speeding up normal
builds.
Enable safebrowsing v4 completion but ignore the result by checking preference in
nsUrlClassifierLookupCallback::Completion may cause telemetry measure incorrect match
result since v4 completions will always be ignored.
So in this patch we move the preference check after telemetry is measured and then we
ignore the result.
MozReview-Commit-ID: J29JitvW3Lc
--HG--
extra : rebase_source : 0cf8797de476980ce25d83099e6526321308327c
This changeset changes tests using ForgetAboutSite.removeDataFromDomain
to yield on it, since now it is a Task
MozReview-Commit-ID: 72OEYoO1avd
--HG--
extra : rebase_source : 9ea8cc06493c3e965d260dc9377461ff29fe572a
Amended to fix review changes
(stylistic + other)
Turns all cleaners into promises so they run asyc
MozReview-Commit-ID: DV5ug6vNXkS
--HG--
extra : rebase_source : df9d9cd98f25e2a899c0a74f836f217d3ad52426
CLOSED TREE
Backed out changeset 4ac559eea9ec (bug 1348803)
Backed out changeset 2ab6e0b8aec6 (bug 1348803)
Backed out changeset f966aef934b1 (bug 1348803)
We could register media control related event after the tab has active media.
But we still need to register "audioFocusChange" in the beginning, because it
affect every tab even the tab has no active media.
MozReview-Commit-ID: 4pBKIR8F5tV
--HG--
extra : rebase_source : fc26c98ed7b33552b4eba5b20168394b1b1a4390
Also, avoid two entirely unused imports.
This avoids loading the various osfile jsms at startup.
The additional shim method for sysinfo is needed in
test_TelemetryEnvironment.js because delaying the intialization of
osfile results in loading osfile_shared_allthreads.jsm after the shim
sysinfo has been added, which initializes
Ci.nsIOSFileConstantsService, which calls into sysinfo in
InitOSFileConstants().
MozReview-Commit-ID: D7IWk20TRt1
--HG--
extra : rebase_source : cf820b1ed72a78ff34e99d2129401c4806a68175
There are two sets of three probes here.
The first set of probes reports the maximum contiguous frames dropped during
compositor animations. This gives us some idea of the volatility of the
compositor framerate.
The second set of probes reports the frame throughput, which is the number of
actual frames composited divided by the expected number of frames (based on the
vsync interval and wall-clock time) during compositor animations. The ratio is
scaled up to 1000, so values of 1000 should indicate a "perfect score". The
probes use linear bucketing with a max value of 1500 in case we get abnormal
cases where we composite more frames than expected.
Within each set, there is one probe for compositor animations triggered by the
chrome process, one for compositor animations triggered by a content process,
and one for compositor animations driven by APZ.
MozReview-Commit-ID: LFJsEPnQhja
--HG--
extra : rebase_source : d943b4612c2a84b0d4b85c5eafb0d0025ffba80f
We used to have this rule in shared stylesheet for both mobile and
desktop, but the namespace mismatched after visual refresh (De-XUL).
Now we should add this rule back to hide cursor while fading out.
MozReview-Commit-ID: HDZKn8CrQ5X
--HG--
extra : rebase_source : 7a63650b9f048dd5b0b0b79ce289bc762cc6a07b
Since a LayoutHistoryState is basically just a collection of PresStates, we just save each PresState we can find and then later restore it.
MozReview-Commit-ID: A6WpdelseHn
--HG--
extra : rebase_source : 21c2929ed64ff5ef046ea65664af78cdad517786
This patch updates the description for the probes SSL_CIPHER_SUITE_FULL and
SSL_CIPHER_SUITE_RESUMED found in Histograms.json to point to
AccumulateCipherSuite which contains the key exchanges instead of the previous
HandshakeCallback.
MozReview-Commit-ID: Bf3xKGdK4Sd
--HG--
extra : rebase_source : e278f28e6abf4044d0e554dd28bd2d550c576ef2
This patch adds an accumulation limit to Scalars IPC messages,
in a similar way as this limit was already implemented for Histograms.
After a discussion in the bug entry, 10000 was chosen as limit.
MozReview-Commit-ID: ARBUOFnfDBr
--HG--
extra : rebase_source : 353e35fe1cf37d1ebad1e54ac9d329a958f95dc3
Add a new attribute "rawValue" in each of the numeric fields. We store the
non-formatted number in this attribute, and display formatted number using
Intl.NumberFormat.
MozReview-Commit-ID: JkcBObFoYQ3
--HG--
extra : rebase_source : 05918ba57513f9c816273a758ab2aa7198722135
nsIX509Cert.getAllTokenNames() is only used (improperly) to determine if a
certificate is a built-in. nsIX509Cert.isBuiltInRoot should be used instead.
MozReview-Commit-ID: LBwI8nTc05C
--HG--
extra : rebase_source : 9494cd1243395b0d293022e981f64be560a54dec
It is unnecessary to use Vista/7+ API via GetProcAddress.
MozReview-Commit-ID: AT6OJhhf4r2
--HG--
extra : rebase_source : 0231fad656de11c8a8d6d8fa7374a8365778ea43
Gian-Carlo Pascutto
Tim Huang
Ryan VanderMeulen
David Teller
Paavini
Sebastian Hengst
Michal Novotny
Gregory Moore
Nathan Froyd
Wes Kocher
Johann Hofmann
Zibi Braniecki
Dão Gottwald
Iris Hsiao
dimi
milindl
Robert Helmer
Phil Ringnalda
Alastor Wu
Andrew McCreight
Kartikaya Gupta
Ray Lin
Jonathan Watt
David Major
Carsten "Tomcat" Book
Ehsan Akhgari
Alessio Placitelli
Hrishikesh Barman
Ting-Yu Chou
Jan Henning
Andrew Swan
Federico Padua
Jessica Jong
Cykesiopka
Kaku Kuo
Makoto Kato