searching for reviewer(sfink)
e692dc0153e26734035bc0792d7336fee14ccb92: Bug 1640844 - Part 5: Take account of how much time we spend collecting when calculating nursery size r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Mon, 13 Jul 2020 07:49:32 +0000 - rev 540491
Push 37603 by ccoroiu@mozilla.com at Wed, 15 Jul 2020 16:06:28 +0000
Bug 1640844 - Part 5: Take account of how much time we spend collecting when calculating nursery size r=sfink This adds a goal of using less then 1% of total time collecthing the nursery, which has the effect of increasing the nursery size where a lot of nursery garbage is created but little is tenured. In local testing this is a win on ARES6 and octane, but perfherder paints a more ambiguous picture. I'd like to land this anyway and we can back it out if it causes problems. Differential Revision: https://phabricator.services.mozilla.com/D80003
e2647b165bc726430f815c3700d3b663c5bfe568: Bug 1652650 - Fix build bustage for --enable-more-deterministic builds r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Tue, 14 Jul 2020 22:35:48 +0000 - rev 540457
Push 37602 by dluca@mozilla.com at Wed, 15 Jul 2020 09:37:18 +0000
Bug 1652650 - Fix build bustage for --enable-more-deterministic builds r=sfink Differential Revision: https://phabricator.services.mozilla.com/D83467
3dfec4ecf5a5a93c53541beeb70b52a9a23403e5: Bug 1652447 - Remove IsAboutToBeFinalizedDuringSweep and replaces uses with IsAboutToBeFinalized r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Mon, 13 Jul 2020 15:05:03 +0000 - rev 540210
Push 37595 by dluca@mozilla.com at Mon, 13 Jul 2020 21:48:13 +0000
Bug 1652447 - Remove IsAboutToBeFinalizedDuringSweep and replaces uses with IsAboutToBeFinalized r=sfink I added an overload of IsInsideNursery for TenuredCell that always returns false so that IsAboutToBeFinalizedInternal should optimise out the nursery check for things that are never allocated in the nursery. Differential Revision: https://phabricator.services.mozilla.com/D83308
a624af0525ccfe823be783bfddc51ffd683d7e80: Bug 1652425 - Make WeakCache::sweep conservatively take the store buffer lock r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Mon, 13 Jul 2020 15:00:28 +0000 - rev 540209
Push 37595 by dluca@mozilla.com at Mon, 13 Jul 2020 21:48:13 +0000
Bug 1652425 - Make WeakCache::sweep conservatively take the store buffer lock r=sfink Differential Revision: https://phabricator.services.mozilla.com/D83301
981dac52307ecb57fc43a29bd3c55df97d450fe2: Bug 1652281 - Remove assertion about empty chunks pool that no longer always holds r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Mon, 13 Jul 2020 14:58:57 +0000 - rev 540208
Push 37595 by dluca@mozilla.com at Mon, 13 Jul 2020 21:48:13 +0000
Bug 1652281 - Remove assertion about empty chunks pool that no longer always holds r=sfink GCRuntime::decommitFreeArenasWithoutUnlocking can now get called in situations when we do have empty chunks, e.g. when handling OOM in GCRuntime::decommitFreeArenas. We can just remove this assertion. Differential Revision: https://phabricator.services.mozilla.com/D83295
7df94b4f2b9f8d66120a754920f74e256418214a: Bug 1640844 - Part 4: Only used smoothed growth factor for recent collections r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Mon, 13 Jul 2020 07:48:39 +0000 - rev 540147
Push 37594 by btara@mozilla.com at Mon, 13 Jul 2020 15:59:48 +0000
Bug 1640844 - Part 4: Only used smoothed growth factor for recent collections r=sfink It only makes sense to use the smoothing if the nursery has been collected recently. If not it's unlikely to correlate with recent usage. This change makes us ramp up more quickly after a period of inactivity, as we did before the earlier patches in this series landed. Differential Revision: https://phabricator.services.mozilla.com/D82970
5d0c327a36a8a1384bd5ee16f456b8f5570908c8: Bug 1640844 - Part 3: Minimize nursery size in shrinking GCs r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Mon, 13 Jul 2020 07:49:32 +0000 - rev 540146
Push 37594 by btara@mozilla.com at Mon, 13 Jul 2020 15:59:48 +0000
Bug 1640844 - Part 3: Minimize nursery size in shrinking GCs r=sfink The main purpose of this is so that that worker nurseries are minimised when workers go idle, since that uses a shrinking GC. Testing indicates that this is sufficient to fix the memory regressions. Differential Revision: https://phabricator.services.mozilla.com/D82969
77a3b9246b5b7af30be9e331fd19ea9e8d18bbd6: Bug 1652019 - Move more of decommit logic off-thread r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Fri, 10 Jul 2020 17:02:42 +0000 - rev 539927
Push 37587 by btara@mozilla.com at Fri, 10 Jul 2020 21:52:04 +0000
Bug 1652019 - Move more of decommit logic off-thread r=sfink We can sort the available chunks list and prepare the list of chunks to decommit when we're already runnin off-thread. Depends on D83107 Differential Revision: https://phabricator.services.mozilla.com/D83108
c3399c3f38314b970cea6a14da8a23714a4fcc69: Bug 1652019 - Factor out the logic to poll/wait on a background task r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Fri, 10 Jul 2020 16:59:39 +0000 - rev 539926
Push 37587 by btara@mozilla.com at Fri, 10 Jul 2020 21:52:04 +0000
Bug 1652019 - Factor out the logic to poll/wait on a background task r=sfink This happens in two places in the incrementalSlice() and we can factor it out. Differential Revision: https://phabricator.services.mozilla.com/D83107
5de1f21de997eb6f680a8e47e4d3a634b10184bd: Bug 1649757 - don't upload artifacts for various SM builds; r=sfink
Nathan Froyd <froydnj@mozilla.com> - Fri, 10 Jul 2020 15:53:38 +0000 - rev 539913
Push 37587 by btara@mozilla.com at Fri, 10 Jul 2020 21:52:04 +0000
Bug 1649757 - don't upload artifacts for various SM builds; r=sfink The artifacts (`js`, `jsapi-tests`) generated by these builds are rather large, and we don't need them for anything. Differential Revision: https://phabricator.services.mozilla.com/D81880
36245fe59dce3f813410de18a26b2033bc3ab066: Bug 1651390 - Part 2: Remove near duplicate NewString[DontDeflate] functions. r=sfink
André Bargull <andre.bargull@gmail.com> - Fri, 10 Jul 2020 12:41:32 +0000 - rev 539877
Push 37586 by ccoroiu@mozilla.com at Fri, 10 Jul 2020 16:06:24 +0000
Bug 1651390 - Part 2: Remove near duplicate NewString[DontDeflate] functions. r=sfink Depends on D82732 Differential Revision: https://phabricator.services.mozilla.com/D82733
a7dafb31359257ffb18782757fae34cdd5571384: Bug 1651390 - Part 1: Copy comments from NewString[DontDeflate]. r=sfink
André Bargull <andre.bargull@gmail.com> - Thu, 09 Jul 2020 18:12:31 +0000 - rev 539876
Push 37586 by ccoroiu@mozilla.com at Fri, 10 Jul 2020 16:06:24 +0000
Bug 1651390 - Part 1: Copy comments from NewString[DontDeflate]. r=sfink Copy the comments to the `NewString<allowGC>` alternatives before removing the functions in part 2. Differential Revision: https://phabricator.services.mozilla.com/D82732
b7a8d9f438600a3b0d73e55157ccd52ec33cc91c: Bug 1651683 - Created an abstraction for sequential iteration of multiple iterators r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Thu, 09 Jul 2020 18:18:52 +0000 - rev 539654
Push 37585 by nbeleuzu@mozilla.com at Fri, 10 Jul 2020 09:48:19 +0000
Bug 1651683 - Created an abstraction for sequential iteration of multiple iterators r=sfink Differential Revision: https://phabricator.services.mozilla.com/D82901
2cded7ee7fb200083079ec3b7c01035f07663b44: Bug 1651345 - Make addMarkObservers evict the nursery for every nursery-allocated argument it finds r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Thu, 09 Jul 2020 18:09:43 +0000 - rev 539653
Push 37585 by nbeleuzu@mozilla.com at Fri, 10 Jul 2020 09:48:19 +0000
Bug 1651345 - Make addMarkObservers evict the nursery for every nursery-allocated argument it finds r=sfink The problem is that it's not enough to evit the nursery once at the start of addMarkObservers since JS_GetElement can end up running a getter that creates new objects. Differential Revision: https://phabricator.services.mozilla.com/D82873
9a4f3aa0719c7e10000f7764eeef27a91f0b8213: Bug 1642476 - Add Basic version of getRealmConfiguration r=sfink
Matthew Gaudet <mgaudet@mozilla.com> - Tue, 07 Jul 2020 18:51:43 +0000 - rev 539181
Push 37578 by btara@mozilla.com at Wed, 08 Jul 2020 09:42:17 +0000
Bug 1642476 - Add Basic version of getRealmConfiguration r=sfink Differential Revision: https://phabricator.services.mozilla.com/D77922
ac6d26908599e15b1c129cbb0f252ca9b82836c6: Bug 1648901 - Fix crash tracing partially constructed FinalizationRegistrationsObject r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Tue, 30 Jun 2020 22:31:29 +0000 - rev 538500
Push 37564 by nbeleuzu@mozilla.com at Fri, 03 Jul 2020 03:56:55 +0000
Bug 1648901 - Fix crash tracing partially constructed FinalizationRegistrationsObject r=sfink The shell allocation metadata hook means we can see these before they've been fully initialised. Differential Revision: https://phabricator.services.mozilla.com/D81521
425a08e7e999516a02bc0aeaf3f31d6ca32569c9: Bug 1640844 - Part 2: Don't use smoothed growth factor for initial collection r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Thu, 02 Jul 2020 10:27:47 +0000 - rev 538404
Push 37562 by nerli@mozilla.com at Thu, 02 Jul 2020 15:21:09 +0000
Bug 1640844 - Part 2: Don't use smoothed growth factor for initial collection r=sfink This only uses the smoothed growth factor if it contains data from previous collections. This should fix the awsy regressions from when this landed before. Fast ramp up is important in startup to avoid growing the tenured heap which can lead to larger overall memory usage. Differential Revision: https://phabricator.services.mozilla.com/D81841
26ee57e3a7b000acf6f55f39725888a6cd8e5b69: Bug 1640844 - Part 1: Improve nursery size calculation to reduce size changes r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Thu, 02 Jul 2020 10:27:47 +0000 - rev 538403
Push 37562 by nerli@mozilla.com at Thu, 02 Jul 2020 15:21:09 +0000
Bug 1640844 - Part 1: Improve nursery size calculation to reduce size changes r=sfink When running benchmarks the nursery size changes frequently, often at every collection. This patch uses a moving average of the calculated growth factor. This takes into account information from previous collections as well when calculating the nursery size and producees much more stable results. Differential Revision: https://phabricator.services.mozilla.com/D80002
6fe50c94d4824c6d1d63fa3844489311b09464e0: Bug 1649716 - Don't resize the nursery during shutdown r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Wed, 01 Jul 2020 16:33:36 +0000 - rev 538269
Push 37559 by rmaries@mozilla.com at Wed, 01 Jul 2020 21:32:12 +0000
Bug 1649716 - Don't resize the nursery during shutdown r=sfink Differential Revision: https://phabricator.services.mozilla.com/D81833
93e2d28d05555d3f0591cdbf2acd85d7eba88531: Bug 1649716 - Factor out common code from Nursery::init and Nursery::enable r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Wed, 01 Jul 2020 16:33:23 +0000 - rev 538268
Push 37559 by rmaries@mozilla.com at Wed, 01 Jul 2020 21:32:12 +0000
Bug 1649716 - Factor out common code from Nursery::init and Nursery::enable r=sfink Differential Revision: https://phabricator.services.mozilla.com/D81832
9872086a16f9be8b4bbcd9ff459acc13af96a4d1: Bug 1647702 - Broaden the race supression for compacting race between updating ObjectGroups and Shapes r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Fri, 26 Jun 2020 16:25:03 +0000 - rev 537611
Push 37545 by nerli@mozilla.com at Sat, 27 Jun 2020 09:38:32 +0000
Bug 1647702 - Broaden the race supression for compacting race between updating ObjectGroups and Shapes r=sfink This is a data race, but I don't think it is causing problems at the moment. It will go away when we get rid of ObjectGroups and is not trivial to fix, hence the decision to suppress it. Differential Revision: https://phabricator.services.mozilla.com/D81327
5152aae02ecb2da140b59d4b244151dea88063a5: Bug 1648272 - Change store buffer mutex order so as not to conflict with wasm mutexes r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Thu, 25 Jun 2020 21:57:54 +0000 - rev 537562
Push 37545 by nerli@mozilla.com at Sat, 27 Jun 2020 09:38:32 +0000
Bug 1648272 - Change store buffer mutex order so as not to conflict with wasm mutexes r=sfink Differential Revision: https://phabricator.services.mozilla.com/D81135
27a7ed0c155ae353d86865c00c9b959edd0d2cf7: Bug 1638764 - Make the incremental marking validator check that all chunks are found in its map r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Wed, 24 Jun 2020 14:04:39 +0000 - rev 537179
Push 37537 by ncsoregi@mozilla.com at Wed, 24 Jun 2020 21:50:10 +0000
Bug 1638764 - Make the incremental marking validator check that all chunks are found in its map r=sfink I still don't know why this is happening, but this should make the problem a release mode assertion failure rather than an illegal memory access. Differential Revision: https://phabricator.services.mozilla.com/D80866
6f212eb26c1abd5a9a1de48d97a32548ae010a4f: Bug 1647325 - Trace weak edges of FinalizationRegistrationsObjects when required r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Tue, 23 Jun 2020 15:22:58 +0000 - rev 536954
Push 37533 by dluca@mozilla.com at Tue, 23 Jun 2020 21:38:40 +0000
Bug 1647325 - Trace weak edges of FinalizationRegistrationsObjects when required r=sfink FinalizationRegistrationsObject only holds a vector of weak pointers to FinalizationRecordObjects, but it still needs a trace method so that those weak pointers can get updated by a moving GC. Differential Revision: https://phabricator.services.mozilla.com/D80630
9ba80352f5c39a42e81c0ed8a68891fc449c9c56: Bug 1647319 - Use NestedIterator to implement ArenasToUnmark r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Tue, 23 Jun 2020 10:55:33 +0000 - rev 536875
Push 37533 by dluca@mozilla.com at Tue, 23 Jun 2020 21:38:40 +0000
Bug 1647319 - Use NestedIterator to implement ArenasToUnmark r=sfink Differential Revision: https://phabricator.services.mozilla.com/D80487
17c17458be2e20326f8850c2782c75ce6e6886d0: Bug 1647319 - Use NestedIterator to implement ZoneAllCellIter r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Tue, 23 Jun 2020 11:18:15 +0000 - rev 536874
Push 37533 by dluca@mozilla.com at Tue, 23 Jun 2020 21:38:40 +0000
Bug 1647319 - Use NestedIterator to implement ZoneAllCellIter r=sfink This also means we can simplify ArenaCellIter is it doesn't need to support reset() any more. I had to rename the getCell/get methods returning TenuredCell*/T* to get/as to make this work. I also changed use of |ArenaCellIter i| to |ArenaCellIter cell|, like we do for ZonesIter. Differential Revision: https://phabricator.services.mozilla.com/D80486
fe6e1d55aa8fd11ed3da2248d5e2f7e1c31ec3f9: Bug 1647319 - GCZones constructor doesn't need to take ZoneSelector parameter r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Tue, 23 Jun 2020 11:11:14 +0000 - rev 536873
Push 37533 by dluca@mozilla.com at Tue, 23 Jun 2020 21:38:40 +0000
Bug 1647319 - GCZones constructor doesn't need to take ZoneSelector parameter r=sfink There's only one place this is called with SkipAtoms and it's fine to include the atoms zone there. Differential Revision: https://phabricator.services.mozilla.com/D80485
791dfca7e74d51d3679a1aebef6d7c490f0131ce: Bug 1647319 - Use NestedIterator to implement public iterators for realms and compartments r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Tue, 23 Jun 2020 10:54:50 +0000 - rev 536872
Push 37533 by dluca@mozilla.com at Tue, 23 Jun 2020 21:38:40 +0000
Bug 1647319 - Use NestedIterator to implement public iterators for realms and compartments r=sfink Differential Revision: https://phabricator.services.mozilla.com/D80484
8624539378e168eba093ebe43b32953b0ef14689: Bug 1647319 - Add NestedIterator template class r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Tue, 23 Jun 2020 10:54:43 +0000 - rev 536871
Push 37533 by dluca@mozilla.com at Tue, 23 Jun 2020 21:38:40 +0000
Bug 1647319 - Add NestedIterator template class r=sfink Differential Revision: https://phabricator.services.mozilla.com/D80483
4a59ad84db1b0f7129b2e28c239b1035f277f3af: Bug 1647319 - Split off NonAtomZonesIter from ZonesIter and add AllZonesIter r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Tue, 23 Jun 2020 10:54:35 +0000 - rev 536870
Push 37533 by dluca@mozilla.com at Tue, 23 Jun 2020 21:38:40 +0000
Bug 1647319 - Split off NonAtomZonesIter from ZonesIter and add AllZonesIter r=sfink It's useful in later patches to have zone iterators whose constructors take a single parameter. Differential Revision: https://phabricator.services.mozilla.com/D80482
cf488a06921ec04de214013170daf0aca6a83ff6: Bug 1647319 - Use NestedIterator to implement ArenasToUnmark r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Tue, 23 Jun 2020 08:32:14 +0000 - rev 536757
Push 37533 by dluca@mozilla.com at Tue, 23 Jun 2020 21:38:40 +0000
Bug 1647319 - Use NestedIterator to implement ArenasToUnmark r=sfink Differential Revision: https://phabricator.services.mozilla.com/D80487
61916f49f2b3c38a26a8a48b597c7434e6b3cc59: Bug 1647319 - Use NestedIterator to implement ZoneAllCellIter r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Tue, 23 Jun 2020 08:31:59 +0000 - rev 536756
Push 37533 by dluca@mozilla.com at Tue, 23 Jun 2020 21:38:40 +0000
Bug 1647319 - Use NestedIterator to implement ZoneAllCellIter r=sfink This also means we can simplify ArenaCellIter is it doesn't need to support reset() any more. I had to rename the getCell/get methods returning TenuredCell*/T* to get/as to make this work. I also changed use of |ArenaCellIter i| to |ArenaCellIter cell|, like we do for ZonesIter. Differential Revision: https://phabricator.services.mozilla.com/D80486
32b62f886374909e479a29c345ff4b238da09192: Bug 1647319 - GCZones constructor doesn't need to take ZoneSelector parameter r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Tue, 23 Jun 2020 08:31:28 +0000 - rev 536755
Push 37533 by dluca@mozilla.com at Tue, 23 Jun 2020 21:38:40 +0000
Bug 1647319 - GCZones constructor doesn't need to take ZoneSelector parameter r=sfink There's only one place this is called with SkipAtoms and it's fine to include the atoms zone there. Differential Revision: https://phabricator.services.mozilla.com/D80485
0e900e18ead7ea87acc2cdc68ddb5e7b1054da6a: Bug 1647319 - Use NestedIterator to implement public iterators for realms and compartments r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Tue, 23 Jun 2020 08:30:57 +0000 - rev 536754
Push 37533 by dluca@mozilla.com at Tue, 23 Jun 2020 21:38:40 +0000
Bug 1647319 - Use NestedIterator to implement public iterators for realms and compartments r=sfink Differential Revision: https://phabricator.services.mozilla.com/D80484
bd0b3614efcf062451ed9443fdae06e5cf43b808: Bug 1647319 - Add NestedIterator template class r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Tue, 23 Jun 2020 08:30:34 +0000 - rev 536753
Push 37533 by dluca@mozilla.com at Tue, 23 Jun 2020 21:38:40 +0000
Bug 1647319 - Add NestedIterator template class r=sfink Differential Revision: https://phabricator.services.mozilla.com/D80483
1adf40ab95b9e04460a4b4e644a1490ed8a86f46: Bug 1647319 - Split off NonAtomZonesIter from ZonesIter and add AllZonesIter r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Tue, 23 Jun 2020 08:30:21 +0000 - rev 536752
Push 37533 by dluca@mozilla.com at Tue, 23 Jun 2020 21:38:40 +0000
Bug 1647319 - Split off NonAtomZonesIter from ZonesIter and add AllZonesIter r=sfink It's useful in later patches to have zone iterators whose constructors take a single parameter. Differential Revision: https://phabricator.services.mozilla.com/D80482
14c68ad90aaa1d41e0179accee9d244b62997aec: Bug 1646708 - Reset sweepMarkTaskStarted after joining the sweep task r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Mon, 22 Jun 2020 17:26:00 +0000 - rev 536742
Push 37533 by dluca@mozilla.com at Tue, 23 Jun 2020 21:38:40 +0000
Bug 1646708 - Reset sweepMarkTaskStarted after joining the sweep task r=sfink The problem here is that if we don't reset sweepMarkTaskStarted then we won't start the task next time there is work for it to do. This was missing from endSweepingSweepGroup but I've refectored so this happens in joinSweepMarkTask now. Differential Revision: https://phabricator.services.mozilla.com/D80441
2d6665fa9b438d8808c2bb8748cf1d8859327a78: Bug 1640844 - Improve nursery size calculation to reduce size changes r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Wed, 17 Jun 2020 17:47:59 +0000 - rev 536255
Push 37519 by ccoroiu@mozilla.com at Thu, 18 Jun 2020 21:28:28 +0000
Bug 1640844 - Improve nursery size calculation to reduce size changes r=sfink When running benchmarks the nursery size changes frequently, often at every collection. This patch uses a moving average of the calculated growth factor. This takes into account information from previous collections as well when calculating the nursery size and producees much more stable results. Differential Revision: https://phabricator.services.mozilla.com/D80002
6ea7e854f8c7e95f8a031e5ee19b1adbdc9a95e5: Bug 1642708 - Stop running BinAST tests in automation r=perftest-reviewers,Bebe,sfink
Ted Campbell <tcampbell@mozilla.com> - Wed, 17 Jun 2020 13:57:04 +0000 - rev 536104
Push 37516 by cbrindusan@mozilla.com at Wed, 17 Jun 2020 21:52:06 +0000
Bug 1642708 - Stop running BinAST tests in automation r=perftest-reviewers,Bebe,sfink Remove the BinAST tests from following jobs: - SM(p) - SM(cgc) Remove the BinAST performance tests: - binast-instagram Mark the WPT BinAST tests as backlog. Differential Revision: https://phabricator.services.mozilla.com/D77919
ad613fa94a83e98d0ed476ef7e5c6ead0b6d2a03: Bug 1642708 - Stop running BinAST tests in automation r=perftest-reviewers,Bebe,sfink
Ted Campbell <tcampbell@mozilla.com> - Wed, 17 Jun 2020 08:06:08 +0000 - rev 536086
Push 37516 by cbrindusan@mozilla.com at Wed, 17 Jun 2020 21:52:06 +0000
Bug 1642708 - Stop running BinAST tests in automation r=perftest-reviewers,Bebe,sfink Remove the BinAST tests from following jobs: - SM(p) - SM(cgc) Remove the BinAST performance tests: - binast-instagram Mark the WPT BinAST tests as backlog. Differential Revision: https://phabricator.services.mozilla.com/D77919
934e959205abe817c23015c326cff2413e1f040c: Bug 1645377 - Simplify the store buffer access invariant to always require taking the lock from within the GC r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Tue, 16 Jun 2020 18:40:17 +0000 - rev 535910
Push 37513 by rmaries@mozilla.com at Wed, 17 Jun 2020 03:41:56 +0000
Bug 1645377 - Simplify the store buffer access invariant to always require taking the lock from within the GC r=sfink Previously this requiring no lock for main thread access, but I'd forgotten about the fact that helper thread tasks can get run on the main thread. This patch simplifies the invariant so that we always require the lock when called from within the GC and fixes the places this didn't happen. This adds some extra locking but this will always be uncontended and shouldn't have a performance impact. Depends on D79862 Differential Revision: https://phabricator.services.mozilla.com/D79863
74f73190afad2bbde97bd6009430b87445718a01: Bug 1645530 - Allow minor GC to be requested while performing major GC r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Tue, 16 Jun 2020 18:42:37 +0000 - rev 535909
Push 37513 by rmaries@mozilla.com at Wed, 17 Jun 2020 03:41:56 +0000
Bug 1645530 - Allow minor GC to be requested while performing major GC r=sfink Since we stopped doing minor GC at the start of every sweep slice we can now add store buffer entries while sweeping has tables as table entries get moved around. This can lead to requesting (but not performing) a minor GC inside a major GC slice. Depends on D79702 Differential Revision: https://phabricator.services.mozilla.com/D79703
4df57a37fd3cbebbf47ea8ed2ecbac60983f77c8: Bug 1645379 - Disable GC zeal when running testcase for bug 1643913 r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Tue, 16 Jun 2020 18:42:37 +0000 - rev 535908
Push 37513 by rmaries@mozilla.com at Wed, 17 Jun 2020 03:41:56 +0000
Bug 1645379 - Disable GC zeal when running testcase for bug 1643913 r=sfink Use of startgc() will throw is zeal has already started an incremental GC. Differential Revision: https://phabricator.services.mozilla.com/D79702
75b4198a731db7dcbfc6280bf25639b083e36cca: Bug 1644985 - Clean up WeakRef and FinalizationRegistry data when cross compartment wrappers are nuked r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Sat, 13 Jun 2020 23:05:15 +0000 - rev 535616
Push 37505 by nerli@mozilla.com at Sun, 14 Jun 2020 21:53:00 +0000
Bug 1644985 - Clean up WeakRef and FinalizationRegistry data when cross compartment wrappers are nuked r=sfink This crash happens because we try and clean up the map from target to WeakRef in the WeakRef finalizer, and the target can be dead by this point if it's a nuked CCW (before it is nuked the CCW ensures this sweep order does not happen). The fix is to fix up the map when CCWs to WeakRefs are nuked. Fortunately there's already a hook where the GC is told about this. The same issue applies to FinalizationRecordObjects. This fix is slightly different because they don't have a target pointer so we can't find the map entry. Instead we clear the record and cleanup happens later when it gets swept. Differential Revision: https://phabricator.services.mozilla.com/D79533
5742bc131451a5af275ebe5bf99ea77e493f1a94: Bug 1645113 - Don't sweep arenas that were allocated during marking asa they cannot contain any dead cells r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Fri, 12 Jun 2020 08:03:39 +0000 - rev 535276
Push 37501 by nbeleuzu@mozilla.com at Sat, 13 Jun 2020 03:21:52 +0000
Bug 1645113 - Don't sweep arenas that were allocated during marking asa they cannot contain any dead cells r=sfink The patch adds areanas allocated during marking into a separate set of arenas lists, which are not swept but are merged back into the main arena lists at the end of sweeping. We do need to do some sweeping for newly allocated arenas on account of type inference data. I haven't looked too hard into why this is. Differential Revision: https://phabricator.services.mozilla.com/D79334
f0b43814725e210292bb970650dc079b9e03b97a: Bug 1641517 - Fix logic in ShouldIgnorePropertyDefinition to really ignore cleanupSome if it's not enabled r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Thu, 11 Jun 2020 18:02:39 +0000 - rev 535163
Push 37498 by apavel@mozilla.com at Fri, 12 Jun 2020 03:05:25 +0000
Bug 1641517 - Fix logic in ShouldIgnorePropertyDefinition to really ignore cleanupSome if it's not enabled r=sfink Disabling toSource also disabled checking for whether to ignore cleanupSome. Differential Revision: https://phabricator.services.mozilla.com/D79298
1ab3acda37b63503e98cb467431716f92cae3e69: Bug 1644763 - Enable weak references by default in the JS shell r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Thu, 11 Jun 2020 14:40:15 +0000 - rev 535116
Push 37498 by apavel@mozilla.com at Fri, 12 Jun 2020 03:05:25 +0000
Bug 1644763 - Enable weak references by default in the JS shell r=sfink This negates the existing option and removes references to the original option from the tests. Differential Revision: https://phabricator.services.mozilla.com/D79146
cbd70b0fa84946f1159c8bedd5aee923bb7b45f2: Bug 1644393 - Fix test case to work in beta/release builds r=sfink
Jon Coppeard <jcoppeard@mozilla.com> - Thu, 11 Jun 2020 09:31:07 +0000 - rev 535055
Push 37498 by apavel@mozilla.com at Fri, 12 Jun 2020 03:05:25 +0000
Bug 1644393 - Fix test case to work in beta/release builds r=sfink This changes the test to be much more specific. It adds a way of making gcslice only trigger a slice during an incremental collection, without which the collection loop can loop forever. Differential Revision: https://phabricator.services.mozilla.com/D79136
df775aa813675d71bf534c217ca5d327d6ea4686: Bug 1641369 - Part 5: Make sure the workspace is clean in SM package build. r=sfink
Tooru Fujisawa <arai_a@mac.com> - Wed, 03 Jun 2020 05:34:20 +0000 - rev 534942
Push 37496 by btara@mozilla.com at Thu, 11 Jun 2020 04:04:33 +0000
Bug 1641369 - Part 5: Make sure the workspace is clean in SM package build. r=sfink Differential Revision: https://phabricator.services.mozilla.com/D77643
84fac4666ef130014c9064197c2112092dd5a4fb: Bug 1641369 - Part 4: Make clean command more robust. r=sfink
Tooru Fujisawa <arai_a@mac.com> - Wed, 03 Jun 2020 23:03:09 +0000 - rev 534941
Push 37496 by btara@mozilla.com at Thu, 11 Jun 2020 04:04:33 +0000
Bug 1641369 - Part 4: Make clean command more robust. r=sfink Depends on D77991 Differential Revision: https://phabricator.services.mozilla.com/D77992