Bug 1414789: Backout Stylo DOM APIs from beta. r=xidorn a=gchang MozReview-Commit-ID: BW5BZpC63B8

Bug 1431282. Work around layout violating its own invariants and causing stylo code to crash, _again_. r=emilio, a=gchang MozReview-Commit-ID: 3ggJI0qmOJV

bug 1429666 cubeb_resampler_speex: don't call data callback while draining r=padenot a=gchang MozReview-Commit-ID: 1XEzZjPGai9

Bug 1421345 - Check the canary during allocations. r=jet a=gchang

Bug 1401459 - always run HttpChannelChild::Release on main thread. r=mayhemer a=gchang

Bug 1430522 - [Form Autofill] Enable address autofill by default on release build. r=seanlee a=gchang MozReview-Commit-ID: 9Pf8l7PqSMi

Bug 1425612 - Better error messages for invalid structured clone data. r=sfink, a=gchang.

Bug 1426783 - Fix error handling in deserialization of invalid typed arrays. r=sfink, a=gchang.

Bug 1421345 - Check the canary during allocations. r=jet a=gchang

Bug 1429885 Support a rounding value of 0 for reduceTimerPrecision r=bkelly,timhuang a=gchang MozReview-Commit-ID: F96EUfXgK9F

Bug 1429859 - Check mDoNotTryEarlyData in Do0RTT(). r=mcmanus a=gchang

Bug 1427221 - Do nsMathMLmfencedFrame cleanup in DestroyFrom, not in the dtor. r=mattwoodrow a=gchang

Bug 1426081 - Migrate legacy search engines to WebExtensions. r=florian a=gchang MozReview-Commit-ID: JfSwJwMC46F

Bug 1416879 - Part 0d: Move browser_multie10s_update.js into its own directory. r=bkelly a=gchang This test is unfortunately a victim of a multi-e10s unregister() race from browser_force_refresh and given the imminent multi-e10s cleanup that's going to happen, the more complicated alternatives aren't worth the effort versus just moving this test to its own directory. ..and now it's disabled too. Bug 1429794 tracks re-enabling.

Bug 1416879 - Part 0c: browser_multie10s_update.js should not use setTimeout. r=bkelly a=gchang This test used a fixed setTimeout of 3secs for serving the SW. This lower-bounded the test runtime at 6 seconds, plus it was not safe in the event of a slow test runner. This set of changes, although a little ugly, improves the logic so that the SW's transmission is driven by a separate "release" fetch that is only triggered when both updates have been issued and the first update failure has been reported. This ensures we are observing the desired situation. There's also a sanity check on the number of times the SW script is fetched.

Bug 1416879 - Part 0b: browser_multie10s_update.js needs to protect its invariants, clean-up after itself. r=bkelly a=gchang

Bug 1416879 - Part 0a: Make browser_force_refresh.js clean up after itself. r=bkelly a=gchang

Bug 1416879 - Part 6: Test cancellation of diverted client-intercepted streams. r=bkelly a=gchang This adds a test where we have a ServiceWorker return 2 different types of streams that Firefox recognizes as downloads which are handled by diversion of the channel to the parent. The diverted downloads are then cancelled and we verify that cancellation actually results in the underlying connections being closed and/or the ServiceWorker notified. Our 2 types of streams are: 1. A pass-through stream that is incrementally delivered through use of an .sjs file that delivers data using setInterval. 2. A SW-authored ReadableStream (which is not enabled by default, so we set a pref.) Determining when the .sjs's stream is canceled is accomplished by opening a second "monitor" connection that only completes when the streaming connection is closed. In all cases we differentiate between cancelation and timeouts firing.

Bug 1416879 - Part 5: FetchStreamReader needs to cancel its reader when it encounters write errors. r=baku, f=bkelly a=gchang Currently, FetchStreamReader never signals to the JS stream code that the reader has been closed. This means that when a ServiceWorker passes a ReadableStream to respondWith and the HTTP Channel gets canceled, the JS code will keep generating the stream without ever realizing the data's not going anywhere. It's necessary to cancel the reader. Or do something like that, this seems to work!

Bug 1416879 - Part 4: FetchDriver needs to propagate write failures. r=baku a=gchang In the scenario where a ServiceWorker returns a pass-through fetch via `evt.respondWith(fetch("underlying"))`, in order for the "underlying" HTTP channel to be canceled when the outer HTTP channel is canceled, FetchDriver's OnDataAvailable method needs to return an error when the output pipe experiences an error. Unfortunately, the contract for ReadSegments is effectively that it returns NS_OK regardless of what the rv of the write handler returned, so relying on the returned rv is insufficient. And because various Write*() methods will all fast-path to returning NS_OK if a count of 0 is passed, it's necessary to infer a closed/broken pipe by noticing that we tried to write more than 0 bytes of data but 0 bytes were written. (This is safe because the pipe we write into was created by FetchDriver::OnStartRequest which explicitly creates an infinite pipe, so it's not possible for the write to fail due to lack of space in the pipe.)

Bug 1416879 - Part 3: (Also Bug 1418795) SyntheticDiversionListener should handle !mIPCOpen. r=bkelly a=gchang The SyntheticDiversionListener needs to handle the case where the IPC connection is gone. This patch avoids calling Send* methods which will crash the content process if the actor has already been destroyed. Additionally, OnDataAvailable will return an error in such a case so that the caller can properly handle the error rather than continuing to attempt to send data to a listener that doesn't care. This latter change is an artifact of a previous hack attempt to fix a related diversion issue that is probably not required for this stack, but makes sense as a fix, so I've left it in.

Bug 1416879 - Part 2: Allow for diversion cancellation and trigger for intercepted channels. r=bkelly, r=mayhemer a=gchang The diversion mechanism never expected to be dealing with data sourced from the content process, but that's exactly what happens with ServiceWorker-intercepted channels with the current child-intercept situation (which is being fixed). In order to allow timely cancellation of diverted intercepted channels, there needs to be a way to relay to the HttpChannelChild that it needs to be canceled so that the synthesized pump can be canceled and diversion can be marked as complete. This patch adds such a mechanism to ADivertableParentChannel and PHttpChannel for the exclusive use of InterceptedHttpChannel and then uses it.

Bug 1416879 - Part 1: DivertComplete should only be sent at OnStopRequest for synthesized responses. r=bkelly a=gchang Diversion for intercepted channels with a synthesized response is a special case. It is not appropriate to send DivertComplete when mEventQ has been drained, because we are not dealing with the usual mEventQ-enqueued OnDataAvailable payloads that had been received over the network and sent down to the child. In this case, all the data originates in the child and does not go through mEventQ. As such, the correct place to send DivertComplete is at OnStopComplete for the synthesized response.

Bug 1429885 Support a rounding value of 0 for reduceTimerPrecision r=bkelly,timhuang a=gchang MozReview-Commit-ID: F96EUfXgK9F

Bug 1429859 - Check mDoNotTryEarlyData in Do0RTT(). r=mcmanus a=gchang

Bug 1426081 - Migrate legacy search engines to WebExtensions. r=florian a=gchang MozReview-Commit-ID: JfSwJwMC46F

Bug 1424341 Add privacy.reduceTimerPrecision and privacy.reduceTimerPrecision.microseconds prefs and tests r=bkelly,timhuang,mrbkap a=gchang This pref does not override privacy.resistFingerprinting, but when it is set (and privacy.resistFingerprinting is not) we will still adjust the precision of almost all timers. The adjustment amount is the second pref, which is defaulted to 20us but now dynamically adjustable (in the scale of microseconds.) We are landing this preffed off at the current value we clamp performance.now() at which is 20us. This commit is the combination of the multiple commits listed in Bug 1424341

Bug 1429133 - Some FontExplorer managed fonts are not rendered. r=Alex_Gaynor a=gchang MozReview-Commit-ID: L5x3GNb3HGU

Bug 1426500 - Part 2: Update test_interfaces mochitest to expect WebVR interfaces to be disabled on release builds for macOS. r=smaug, a=gchang

Bug 1426500 - Disable WebVR for macOS on Beta and Release r=daoshengmu a=gchang - WebVR will continue to be enabled on macOS for Nightly and Dev Edition MozReview-Commit-ID: LpEX13yZVbb

Bug 1425509 - Initialize PerformanceService in RuntimeService, r=bkelly a=gchang

Bug 1417741 Add support of Atomic<> for Preferences::Add*VarCache r=baku,njn a=gchang on a CLOSED TREE This is a partial backport of Bug 1417741 that only adds AddAtomicBoolVarCache because that is all we need.

Bug 1417431 - secureConnectionStart should be 0 for pages with HTTP scheme. r=bkelly a=gchang

Bug 1416878 - Move the implementation of nsIWidgetListener from nsWebBrowser / nsWebShellWindow to a separate object. r=bz a=gchang MozReview-Commit-ID: 5QV6lkCCGW5

Bug 1261175. r=bobowen a=gchang MozReview-Commit-ID: B9yAN0F01rj

Bug 1429219 - Ensure VP8 simulcast with temporal layers won't fail if there aren't enough bits. r=drno, a=gchang

Bug 1428589: Don't read a destroyed cookie list if the last cookie in the entry expired. r=jdm a=gchang Consider the following situation, which is what causes the failure: * `cookies` is an array of length 1 (either from the beginning, or because other cookies have been expired while in this loop). * the only cookie that remains is expired. We append the cookie to removedList, and then we call gCookieService->RemoveCookieFromList, which has the following code: if (aIter.entry->GetCookies().Length() == 1) { mDBState->hostTable.RawRemoveEntry(aIter.entry); } else { aIter.entry->GetCookies().RemoveElementAt(aIter.index); } If we enter the first branch, as it's the case, that will destroy the array. We're effectively removing stuff from the array while mutating it, which is scary. It's fine if we don't delete the array, since we iterate through it using indices, but still it's dangerous as heck. If we're the last element in the array though, we're doomed, because `cookies` is now destroyed. We not only try to access the array length again, but also we try to index on it the next time because we never stopped the loop (`i` is still zero, and the length may very well be garbage). Fix it by keeping the length in sync from the stack and breaking out from the loop if appropriately. MozReview-Commit-ID: 6qaC9yclvP2

Bug 1396399 - Clarify rules for applying activeTab permission to content scripts. r=kmag, r=bz a=gchang MozReview-Commit-ID: 9xPDX8Qk2iR

Bug 1426449: set the correct number of temporal layers for screencasts r=drno a=gchang

Bug 1429542 Make HttpChannelChild::OverrideRunnable check for nullptr before using mNewChannel. r=asuth a=gchang

Bug 1426500 - Disable WebVR for macOS on Beta and Release r=daoshengmu a=gchang - WebVR will continue to be enabled on macOS for Nightly and Dev Edition MozReview-Commit-ID: LpEX13yZVbb

bug 1422735 - Change symbol upload URL from Socorro to Tecken. r=nalexander a=gchang MozReview-Commit-ID: BBs6G8dFfPz

Bug 1421214 - Try GoToAnchor() with unescaped string before using document's charset. r=smaug a=gchang

Bug 1427126 - Use setElementWithType instead of setElement in AppendToList. r=bhackett, a=gchang

Bug 1426801 - Clear gFirstIdleTask if we fail to dispatch the runnable. r=mrbkap, a=gchang MozReview-Commit-ID: 23f9VDTyyqg

Bug 1428165 - Part 2: Fix history expiration query. r=nalexander, a=gchang 'modified' value might be missing; in SQLite, max(123, null) is null, and so we must coalesce fields which might be missing values. MozReview-Commit-ID: Bn1P0kdaHHT

Bug 1428165 - Part 1: Ensure that 'modified' and 'created' timestamps are set when inserting history from sync. r=nalexander, a=gchang This fixes a regression introduced in Bug 1291821. History records would be bulk-inserted from sync, and our ContentProvider would erroneously forget to set these two timestamp fields. MozReview-Commit-ID: 2k0afijN62H

Bug 1426042 - Ensure we don't forget to set bidi data when required on a zero-length trailing frame. r=dholbert, a=gchang

Bug 1426042 - Reftest for bidi reordering where line has an empty trailing ::after pseudo-element. r=dholbert, a=gchang

Bug 1427089 - Wait for a previous empty transaction to complete before doing another empty transaction. r=dvander, a=gchang