Bug 1435319 - tests, r=valentin

Bug 1435319, r=valentin

Bug 1461948 - Uncouple ReferenceType from TypedObject's ReferenceTypeDescr. r=till The coupling created some problems for a later patch where I need to reference ReferenceTypeDescr::Type in GlobalObject.h, but can't, because this creates a circular include path. As for SIMD and Scalar, there's no reason to embed the type all the way inside the TypedObject machinery, we can just have it as its own thing.

Bug 1461948 - Generalize and abstract TypedObject field layout algorithm. r=till The algorithm is used by code that builds the TypeDesc, but also needed earlier by the Wasm parser to build the compilation environment. So abstract the algorithm and expose it.

Bug 1461948 - Factor the TypedObject struct type constructor. r=till For Wasm support we'll want to bypass the bits that interpret the object coming from JS, and instead just pass in parallel arrays of IDs and type descriptors, and other parameters. This also allows us to request an opaque type, even if none of the field types are opaque.

Bug 1453795 - Widget/Cocoa - Initialize member fields in classes/ structures. r=mstange

Bug 1453795 - XPCOM - Initialize member fields in classes/ structures. r=froydnj

Bug 1423017 followup - Make the telemetry opt-out on release channel.

bug 1470005 - change default TRR blacklist duration to 60 seconds r=mcmanus MozReview-Commit-ID: FGQCmhlBBG8

Bug 1463917 part 2 - Have windows widget render scrollcorner. r=jimm MozReview-Commit-ID: K4Cu1mL6xvH

Bug 1463917 part 1 - Add scrollcorner to -moz-appearance so that widget can render it. r=heycam MozReview-Commit-ID: 1Za22ifONfG

Bug 1469095 - Fix always-optimized jobs after bug 1465181. r=dustin

bug 1463356 do not count "not started" TRR resolves as failures r=mcmanus ... when comparing against the native resolver. DNS_TRR_COMPARE is meant to compare how the actually performed name resolves fare against each other. MozReview-Commit-ID: 98NoUGPpHr6

Bug 1469257 - [H264] Only check for SPS changes on keyframe r=bryce Some invalid streams contain SPS changes and those appear to only occur on non-keyframe, this cause all frames to be dropped until the next keyframe is found. This result in apparent freezes. While it is theoretically possible to have SPS changes inband on non-keyframe those should be very rare (I've never seen one). The content would have been invalid anyway in an non-fragmented mp4. So we now only check for a SPS change on keyframe. This would cause no affect on either windows, android or ffmpeg as those decoders handle format change fine. The mac decoder could however show garbled frames temporarily. Differential Revision: https://phabricator.services.mozilla.com/D1733

Bug 1465253: use postfix as prefix when logging and log in clear. r=bwc MozReview-Commit-ID: 4bfVQZyM9qy

Bug 1466379 - fall back to currentURI in case images are blocked, r=bz,dao MozReview-Commit-ID: FW2f18pF4t5

Bug 1469766 - Update OOM hook on rustc 1.28 after rust PR 51543. r=froydnj

Bug 1469676 - Reduce max-run-time for various build tasks; r=glandium The max-run-time for tasks appears to have been mostly cargo culted. In particular, a value of 36000 (10 hours) is quite absurd: no single task takes that long to execute. This commit reduces the max-run-time of several tasks to more reasonable values. The goal here is to prevent excessively long-running tasks. There is definitely room to further tweak values to further mitigate long-running tasks. But let's bite off the biggest chunk first, since that doesn't require much mental effort. There is a possibility I overshot on some of these tasks. If we get timeouts, we can always increase the timeout again. Differential Revision: https://phabricator.services.mozilla.com/D1716

Bug 1469683 - 3. Make sure cached session is closed after crash; r=esawin If @IgnoreCrash is specified, assume there has been a crash and therefore wait for the cached session to close at the end of the test, so that the closing operation doesn't carry over to the next test. MozReview-Commit-ID: KrMfHgIiELQ

Bug 1469683 - 2. Make child crashes throw special exception; r=esawin Make child crashes throw ChildCrashException instead of the generic RuntimeException, which can mean a variety of errors such as timeout. MozReview-Commit-ID: H0E6gipkTxg