9bad8ab13f064e49d2a01d8fe0343712322ba44e: Bug 1362154: Part 6: Fall back to OOL call for named captures in Ion r=jandem
Iain Ireland <iireland@mozilla.com> - Thu, 21 May 2020 06:37:19 +0000 - rev 531651
Push 37441 by apavel@mozilla.com at Fri, 22 May 2020 21:38:53 +0000
Bug 1362154: Part 6: Fall back to OOL call for named captures in Ion r=jandem Named captures add a `groups` property to the match result. The shape of that group depends on the regexp. Allocating an object in masm requires monomorphizing on a particular template object, but all of our existing Ion regexp code is designed to handle arbitrary regexps. This means we can't allocate the `groups` object in jitcode without a lot of work. Given that we have to call into C++ to allocate, we might as well just use the existing OOL fallback. This performs a VM call to RegExpMatcherRaw, which will simply call CreateRegExpMatchResult. Differential Revision: https://phabricator.services.mozilla.com/D76038
00d2e220a57535825b59e92df5201119109f55e4: Bug 1362154: Part 5: Create groups property on result object r=mgaudet
Iain Ireland <iireland@mozilla.com> - Wed, 20 May 2020 21:04:47 +0000 - rev 531650
Push 37441 by apavel@mozilla.com at Fri, 22 May 2020 21:38:53 +0000
Bug 1362154: Part 5: Create groups property on result object r=mgaudet When a regexp has named captures, the result object has an extra `groups` property, which is an object that in turn has a string property per named capture. This patch creates the `groups` object and attaches it to the result object. The changes here correspond to this section of the spec proposal, which conveniently highlights the changes: https://tc39.es/proposal-regexp-named-groups/#sec-regexpbuiltinexec Differential Revision: https://phabricator.services.mozilla.com/D76037
1322042730b68d46674cc39feea638a22efb81f3: Bug 1362154: Part 4: Store named capture information in RegExpShared r=mgaudet
Iain Ireland <iireland@mozilla.com> - Wed, 20 May 2020 21:04:01 +0000 - rev 531649
Push 37441 by apavel@mozilla.com at Fri, 22 May 2020 21:38:53 +0000
Bug 1362154: Part 4: Store named capture information in RegExpShared r=mgaudet If a regexp contains any named captures, the irregexp parser returns information about those captures in the form of an ArrayObject, where even elements store the capture name and odd elements store the corresponding capture index. We split this information into two parts. The names are used to create a template object with a property for each capture name. The capture indices are stored as a heap-allocated array. Both of these are stored on the RegExpShared. In the next patch, we use the template object to create a `groups` object for the regexp match result, and use the array of capture indices to populate that object. Differential Revision: https://phabricator.services.mozilla.com/D76036
7c315167d74aa8afe314d5491a3cdfc105229686: Bug 1362154: Part 3: Implement FixedArray r=mgaudet
Iain Ireland <iireland@mozilla.com> - Wed, 20 May 2020 21:03:33 +0000 - rev 531648
Push 37441 by apavel@mozilla.com at Fri, 22 May 2020 21:38:53 +0000
Bug 1362154: Part 3: Implement FixedArray r=mgaudet The irregexp parser uses FixedArray internally to return information about named captures. Now that we can allocate GC things during regexp parsing, it's trivial to implement FixedArray using the elements of a regular dense ArrayObject. Differential Revision: https://phabricator.services.mozilla.com/D76035
e7d2300295143bf951d8bdecd7619ea78722bb80: Bug 1362154: Part 2: Avoid rooting hazard in CheckPatternSyntax r=mgaudet
Iain Ireland <iireland@mozilla.com> - Wed, 20 May 2020 21:09:36 +0000 - rev 531647
Push 37441 by apavel@mozilla.com at Fri, 22 May 2020 21:38:53 +0000
Bug 1362154: Part 2: Avoid rooting hazard in CheckPatternSyntax r=mgaudet In debug, CheckPatternSyntax holds an unrooted RegExpShared across a call to the regexp parser. This is the only remaining obstacle to allocating GC things while parsing regexps. Differential Revision: https://phabricator.services.mozilla.com/D76034
7f7a204fa0246f1aa85bc58137bfc023ee2b8d43: Bug 1362154: Part 1: Rewrite FlatStringReader r=mgaudet
Iain Ireland <iireland@mozilla.com> - Wed, 20 May 2020 17:30:56 +0000 - rev 531646
Push 37441 by apavel@mozilla.com at Fri, 22 May 2020 21:38:53 +0000
Bug 1362154: Part 1: Rewrite FlatStringReader r=mgaudet My initial shim of FlatStringReader used AutoCheckCannotGC, but the API we need to support doesn't actually require that. We only need to be able to get characters by index, which at most requires a HandleLinearString. Getting rid of this AutoCheckCannotGC helps pave the way for allocating GC things while parsing regexps, which is a huge simplification for the FixedArray that irregexp uses to return named capture information. Differential Revision: https://phabricator.services.mozilla.com/D76033
d0e52aa0a9f0af8fcafddacdcc2851ac375dc250: Bug 1639615 - Update Summary to reflect if protections are turned on/off. r=prathiksha,fluent-reviewers,flod
Erica Wright <ewright@mozilla.com> - Fri, 22 May 2020 06:29:03 +0000 - rev 531645
Push 37441 by apavel@mozilla.com at Fri, 22 May 2020 21:38:53 +0000
Bug 1639615 - Update Summary to reflect if protections are turned on/off. r=prathiksha,fluent-reviewers,flod Differential Revision: https://phabricator.services.mozilla.com/D76170
d5255379819c0c4577eb4cbe46c8bb141a3d21d2: Bug 1639988 - Disable the blocking from bug 1525084. r=aosmond
Jeff Muizelaar <jmuizelaar@mozilla.com> - Fri, 22 May 2020 12:52:52 +0000 - rev 531644
Push 37441 by apavel@mozilla.com at Fri, 22 May 2020 21:38:53 +0000
Bug 1639988 - Disable the blocking from bug 1525084. r=aosmond This blocking is preventing a lot of users from getting WebRender and it doesn't seem like the problem is actually reproducible in a lot of situations. By disabling the blocking I'd like to see if we can get more reports so that we can properly tighten the list. Differential Revision: https://phabricator.services.mozilla.com/D76390
049bdaa3e410558ddcb1a71bbc26f3a16b13d0ed: Bug 1636158 - Document how to use attachment dumps in RemoteSettings r=leplatrem
Rob Wu <rob@robwu.nl> - Fri, 22 May 2020 13:09:50 +0000 - rev 531643
Push 37441 by apavel@mozilla.com at Fri, 22 May 2020 21:38:53 +0000
Bug 1636158 - Document how to use attachment dumps in RemoteSettings r=leplatrem Differential Revision: https://phabricator.services.mozilla.com/D76159
d3770d94e7c0dc8a5b06b526a01cde67de83104b: Bug 1636158 - Update the addons-mlbf.bin attachment via periodic_file_updates.sh r=sfraser
Rob Wu <rob@robwu.nl> - Fri, 22 May 2020 01:37:37 +0000 - rev 531642
Push 37441 by apavel@mozilla.com at Fri, 22 May 2020 21:38:53 +0000
Bug 1636158 - Update the addons-mlbf.bin attachment via periodic_file_updates.sh r=sfraser Tested by running the following code and verifying that the script runs successfully and generates a reasonable diff with the latest metadata and matching attachment content: ``` $ cd taskcluster/docker/periodic-updates/ $ tar -czf- -s "#$(git rev-parse --show-cdup)#topsrcdir/#" . $(awk -v d="$(git rev-parse --show-cdup)" '/^# %include/{print d$3}' Dockerfile) | docker build - -t hsts-local $ docker run -e DO_REMOTE_SETTINGS=1 -e PRODUCT="firefox" -e BRANCH="mozilla-central" -e USE_MOZILLA_CENTRAL=1 hsts-local ``` After that, I opened a shell in the container, built Firefox and ran the following test, to confirm that the updated records are used correctly: ``` mach test toolkit/mozapps/extensions/test/xpcshell/rs-blocklist/test_blocklist_mlbf_dump.js ``` Differential Revision: https://phabricator.services.mozilla.com/D76158
d07f8d9d1a01068fdad27c09966326b7109574a5: Bug 1636158 - Remove addons-bloomfilters from Android package r=agi,geckoview-reviewers,snorp
Rob Wu <rob@robwu.nl> - Wed, 20 May 2020 16:39:05 +0000 - rev 531641
Push 37441 by apavel@mozilla.com at Fri, 22 May 2020 21:38:53 +0000
Bug 1636158 - Remove addons-bloomfilters from Android package r=agi,geckoview-reviewers,snorp Those will be restored in bug 1639050 Differential Revision: https://phabricator.services.mozilla.com/D76157
171e43a075c2f8aa0fb666b4b55ae9eed4bacab5: Bug 1617369 - Reformat recent rust changes with rustfmt r=emilio DONTBUILD
Sylvestre Ledru <sledru@mozilla.com> - Fri, 22 May 2020 11:50:08 +0000 - rev 531640
Push 37441 by apavel@mozilla.com at Fri, 22 May 2020 21:38:53 +0000
Bug 1617369 - Reformat recent rust changes with rustfmt r=emilio DONTBUILD # ignore-this-changeset Depends on D76451 Differential Revision: https://phabricator.services.mozilla.com/D76452
4ad31ea203ffd9835af1245d6a9868c9878ed21f: Bug 1638173 - Setup WNP for users coming from <77.0 and receiving the 77.0 release r=RyanVM
Rail Aliiev <rail@mozilla.com> - Fri, 22 May 2020 16:21:04 +0000 - rev 531639
Push 37441 by apavel@mozilla.com at Fri, 22 May 2020 21:38:53 +0000
Bug 1638173 - Setup WNP for users coming from <77.0 and receiving the 77.0 release r=RyanVM Setup WNP for users coming from <77.0 and receiving the 77.0 release Differential Revision: https://phabricator.services.mozilla.com/D76498
879a7f7e9542f827e016c83ce1c22ec6e0acfed0: Bug 1632535 - Let Marionette choose ports for Telemetry integration tests r=grumpy,ahal
Chris H-C <chutten@mozilla.com> - Wed, 06 May 2020 20:32:16 +0000 - rev 531638
Push 37441 by apavel@mozilla.com at Fri, 22 May 2020 21:38:53 +0000
Bug 1632535 - Let Marionette choose ports for Telemetry integration tests r=grumpy,ahal Depends on D73897 Differential Revision: https://phabricator.services.mozilla.com/D73898
03b8a2c7a370571c47ceba393bfe5a80270db5d5: Backed out changeset 8999f97966d2 (bug 1239292) for causing rs bustages.
Mihai Alexandru Michis <malexandru@mozilla.com> - Fri, 22 May 2020 20:05:24 +0300 - rev 531637
Push 37441 by apavel@mozilla.com at Fri, 22 May 2020 21:38:53 +0000
Backed out changeset 8999f97966d2 (bug 1239292) for causing rs bustages. CLOSED TREE
128a2c474755e8e49a2ea3e24a1e71037e0de8b0: Backed out 4 changesets (bug 1637869) for Geckoview failures in org.mozilla.geckoview.test.NavigationDelegateTest.extensionProcessSwitching. CLOSED TREE
Dorel Luca <dluca@mozilla.com> - Fri, 22 May 2020 19:38:15 +0300 - rev 531636
Push 37441 by apavel@mozilla.com at Fri, 22 May 2020 21:38:53 +0000
Backed out 4 changesets (bug 1637869) for Geckoview failures in org.mozilla.geckoview.test.NavigationDelegateTest.extensionProcessSwitching. CLOSED TREE Backed out changeset a0c0203ee8c9 (bug 1637869) Backed out changeset 75359ba23865 (bug 1637869) Backed out changeset d300b61ed89f (bug 1637869) Backed out changeset 34389f9c86e4 (bug 1637869)
5d62e7f9ed7b605d6899a928908f5c2c652cd16d: Backed out 3 changesets (bug 1636583) for Browser-chrome failures in browser/components/urlbar/tests/browser/browser_handleCommand_fallback.js. CLOSED TREE
Dorel Luca <dluca@mozilla.com> - Fri, 22 May 2020 19:07:07 +0300 - rev 531635
Push 37441 by apavel@mozilla.com at Fri, 22 May 2020 21:38:53 +0000
Backed out 3 changesets (bug 1636583) for Browser-chrome failures in browser/components/urlbar/tests/browser/browser_handleCommand_fallback.js. CLOSED TREE Backed out changeset a2e636ff03c2 (bug 1636583) Backed out changeset 660b7de89215 (bug 1636583) Backed out changeset 9fd142817919 (bug 1636583)
0c7f55b96d599ccdf605abd903ed3b6a62187ae3: Backed out changeset 9133eeb54db8 (bug 1529203) for Crashtests failures in dom/base/crashtests/1529203-2.html. CLOSED TREE
Dorel Luca <dluca@mozilla.com> - Fri, 22 May 2020 18:54:44 +0300 - rev 531634
Push 37441 by apavel@mozilla.com at Fri, 22 May 2020 21:38:53 +0000
Backed out changeset 9133eeb54db8 (bug 1529203) for Crashtests failures in dom/base/crashtests/1529203-2.html. CLOSED TREE
a0c0203ee8c91f6897f5c1d1435d6d7ee7b1c5a8: Bug 1637869 - P4. Access pref via staticpref. r=mattwoodrow,necko-reviewers
Jean-Yves Avenard <jyavenard@mozilla.com> - Fri, 22 May 2020 04:12:42 +0000 - rev 531633
Push 37441 by apavel@mozilla.com at Fri, 22 May 2020 21:38:53 +0000
Bug 1637869 - P4. Access pref via staticpref. r=mattwoodrow,necko-reviewers It would otherwise triggers browser/base/content/test/performance/browser_preferences_usage.js That pref would have been checked with every single load, a staticpref is O-1 . Differential Revision: https://phabricator.services.mozilla.com/D76317
75359ba23865c590d0a84b94a0073ba548c3962e: Bug 1637869 - P3. Force about:credits to load in the content process. r=nika
Jean-Yves Avenard <jyavenard@mozilla.com> - Fri, 22 May 2020 04:12:34 +0000 - rev 531632
Push 37441 by apavel@mozilla.com at Fri, 22 May 2020 21:38:53 +0000
Bug 1637869 - P3. Force about:credits to load in the content process. r=nika It was apparently overlooked. Yours truly doesn't even appear on that list :( Differential Revision: https://phabricator.services.mozilla.com/D76316
(0) -300000 -100000 -30000 -10000 -3000 -1000 -300 -100 -50 -20 +20 +50 +100 +300 +1000 +3000 +10000 +30000 tip