Граф коммитов

679 Коммитов

Автор SHA1 Сообщение Дата
Chris Peterson 37efe4d0e6 Bug 1428535 - Add missing override specifiers to overridden virtual functions. r=froydnj
MozReview-Commit-ID: DCPTnyBooIe

--HG--
extra : rebase_source : cfec2d96faeb11656d86d760a34e0a04cacddb13
extra : intermediate-source : 6176724d63788b0fe8caa3f91607c2d93dbaa7ec
extra : source : eebbb0600447f9b64aae3bcd47b4be66c02a51ea
2017-11-05 19:37:28 -08:00
Gabriele Svelto 3e3a359edc Bug 1402519 - Remove MOZ_CRASHREPORTER directives from xpcom; r=froydnj
MozReview-Commit-ID: HS3M0DJQ3qc

--HG--
extra : rebase_source : c535b0760480deea1c9889c629c52857bf52ea64
2017-10-10 11:59:39 +02:00
shindli fb855aa7ba Backed out 16 changesets (bug 1402519) for conflicts during merge r=backout on a CLOSED TREE
Backed out changeset 07fcf163241a (bug 1402519)
Backed out changeset c6d2ad45d8e2 (bug 1402519)
Backed out changeset 8a3caca61294 (bug 1402519)
Backed out changeset 01425eae2c48 (bug 1402519)
Backed out changeset cf298d3815de (bug 1402519)
Backed out changeset e1964f4389cd (bug 1402519)
Backed out changeset f405337f3569 (bug 1402519)
Backed out changeset a76356fd3359 (bug 1402519)
Backed out changeset d3bb350d1c34 (bug 1402519)
Backed out changeset 9d3bfd9f932c (bug 1402519)
Backed out changeset e3dd6e5b073f (bug 1402519)
Backed out changeset e801b0c00134 (bug 1402519)
Backed out changeset 8a4139fa5dca (bug 1402519)
Backed out changeset 8d01c14ac1ca (bug 1402519)
Backed out changeset 24e0dcd01898 (bug 1402519)
Backed out changeset f8fdf450613f (bug 1402519)
2017-11-23 00:11:44 +02:00
Gabriele Svelto 13364cc36b Bug 1402519 - Remove MOZ_CRASHREPORTER directives from xpcom; r=froydnj
MozReview-Commit-ID: HS3M0DJQ3qc

--HG--
extra : rebase_source : bb09e9e3b01d39e91a1377117fa127a09c29cccd
2017-10-10 11:59:39 +02:00
Andrew McCreight e9940ac956 Bug 1277260, part 2 - Add and use method to annotate CC crashes with a class name. r=rweiss+418169,smaug
MozReview-Commit-ID: JweRzAC89NS

--HG--
extra : rebase_source : 6290588e1a4d1c8f59bd13bd975aa7c4f13903df
2017-09-15 13:00:17 -07:00
Andrew McCreight 48ee4c9a3f Bug 1277260, part 1 - Make PtrInfo into a class and mark it final. r=smaug
MozReview-Commit-ID: 3UJ89eCkBsd

--HG--
extra : rebase_source : 90cecf5db10f391e6d7eec2bb1d44810fe5fdd3f
2017-09-15 12:35:18 -07:00
Olli Pettay eb98442ad3 Bug 1400520 - VisitEntries should decrease the length of mEntries the optimal amount, r=mccr8
--HG--
extra : rebase_source : 92ab173641f816df8ff81f4910e796051031a74d
2017-09-20 13:15:19 +03:00
Andrew McCreight 7a52e95946 Bug 1400374 - Fix compiler warning in FixGrayBits. r=smaug
The compiler complains because a postfix increment is done inside of an assert.

MozReview-Commit-ID: LSRsWMn9Tlj

--HG--
extra : rebase_source : 18539a430c3d84675c602699be9a5a62dd477fec
2017-09-15 13:01:15 -07:00
Olli Pettay 11bf6502b9 Bug 1399160 - Make CCGraphBuilder::BuildGraph to check the budget more often, r=mccr8
--HG--
extra : rebase_source : 8c8593de5e261c9c53b532c0a8f8b61e4567bd3c
2017-09-12 18:59:57 +03:00
Olli Pettay 99010cec25 Bug 1398896 - Check the cycle collection budget more often, r=mccr8 2017-09-11 23:57:22 +03:00
Olli Pettay 6e1dc494d2 Bug 1391423, add a nursery for purple buffer to allow faster addref/release on the main thread, r=mccr8
--HG--
extra : rebase_source : 4dcb8af2d3f2518ab0fd00b65fbf1d0096d8e810
2017-08-22 00:01:47 +03:00
Luke Wagner 9b038447f4 Bug 1347644 - Simplify/rename JS::AsyncTask (r=till,bkelly,mccr8)
MozReview-Commit-ID: BahNzYPv0AO
2017-08-16 14:38:53 -05:00
Andrew McCreight 5eecb0df0a Bug 1385474 - Avoid QIing for NoteXPCOMRoot. r=smaug
This callback is only used in very limited ways, so just require that
the caller pass in the canonical supports pointer, plus the
participant. This probably won't affect performance much.

MozReview-Commit-ID: CsThzFsKyYx

--HG--
extra : rebase_source : 9595b1d75fc45bc5ee6d932a840e98b5d760cb78
2017-07-28 16:11:03 -07:00
Andrew McCreight 23896c49cb Bug 1385459 - Don't use QI to canonicalize nsISupports pointers in the purple buffer. r=smaug
The nsISupports objects added to the purple buffer are already
canonical, so we can avoid some overhead by not QIing them to
nsCycleCollectionISupports. This QIing overhead shows up in profiles.

MozReview-Commit-ID: CQN6wwc7MZm

--HG--
extra : rebase_source : 27e6b70f83b42b5db7af3d1e7d62e36d6f4013a0
2017-07-28 15:24:17 -07:00
Olli Pettay 15af1ecf7d Bug 1376614 - Pass budget to forgetSkippable and return early when budget has been used, r=mccr8 2017-06-30 13:44:59 +03:00
Bill McCloskey f115503a0b Bug 1372405 - Provide names for all runnables in the tree (r=froydnj)
MozReview-Commit-ID: DKR6ROiHRS7
2017-06-26 14:19:58 -07:00
Nicholas Nethercote 58786e1ea7 Bug 1375392 - Tweak the PROFILER_LABEL* macros. r=mstange.
This patch makes the following changes to the macros.

- Removes PROFILER_LABEL_FUNC. It's only suitable for use in functions outside
  classes, due to PROFILER_FUNCTION_NAME not getting class names, and it was
  mostly misused.

- Removes PROFILER_FUNCTION_NAME. It's no longer used, and __func__ is
  universally available now anyway.

- Combines the first two string literal arguments of PROFILER_LABEL and
  PROFILER_LABEL_DYNAMIC into a single argument. There was no good reason for
  them to be separate, and it forced a '::' in the label, which isn't always
  appropriate. Also, the meaning of the "name_space" argument was interpreted
  in an interesting variety of ways.

- Adds an "AUTO_" prefix to PROFILER_LABEL and PROFILER_LABEL_DYNAMIC, to make
  it clearer they construct RAII objects rather than just being function calls.
  (I myself have screwed up the scoping because of this in the past.)

- Fills in the 'js::ProfileEntry::Category::' qualifier within the macro, so
  the caller doesn't need to. This makes a *lot* more of the uses fit onto a
  single line.

The patch also makes the following changes to the macro uses (beyond those
required by the changes described above).

- Fixes a bunch of labels that had gotten out of sync with the name of the
  class and/or function that encloses them.

- Removes a useless PROFILER_LABEL use within a trivial scope in
  EventStateManager::DispatchMouseOrPointerEvent(). It clearly wasn't serving
  any useful purpose. It also serves as extra evidence that the AUTO_ prefix is
  a good idea.

- Tweaks DecodePool::SyncRunIf{Preferred,Possible} so that the labelling is
  done within them, instead of at their callsites, because that's a more
  standard way of doing things.

--HG--
extra : rebase_source : 318d1bc6fc1425a94aacbf489dd46e4f83211de4
2017-06-22 17:08:53 +10:00
Carsten "Tomcat" Book 8a1350b5a6 Backed out changeset 4f6302a98ae4 (bug 1372405)
--HG--
extra : rebase_source : 41632f3158e88e692809731394a683d065a73dfb
2017-06-21 13:59:26 +02:00
Bill McCloskey 6b3e84ed5f Bug 1372405 - Provide names for all runnables in the tree (r=froydnj)
MozReview-Commit-ID: DKR6ROiHRS7
2017-06-20 21:44:11 -07:00
Carsten "Tomcat" Book ea1b86680c Backed out changeset 9846de3bd954 (bug 1372405)
--HG--
extra : rebase_source : 5d4a48e8ec394c329994689d938d2a6e9b2752b0
2017-06-20 08:27:02 +02:00
Bill McCloskey 4592152411 Bug 1372405 - Provide names for all runnables in the tree (r=froydnj)
MozReview-Commit-ID: DKR6ROiHRS7
2017-06-19 22:25:47 -07:00
Bill McCloskey 2ab0191db6 Bug 1365102 - Convert NS_GetCurrentThread uses in xpcom/ (r=froydnj)
MozReview-Commit-ID: 69qVzf8QJMq
2017-06-12 20:21:43 -07:00
Andrew McCreight e437a7e4e5 Bug 1367496 - Add more release asserts to the cycle collector. r=smaug
A previous bug missed a few places where we could theoretically
reenter the CC.

MozReview-Commit-ID: I0otlAEwyZa

--HG--
extra : rebase_source : 064b127a2c28a52b2807cd78031de9af4f258f60
2017-05-24 10:13:02 -07:00
Olli Pettay 0096f25b51 Bug 1358761 - replace PurpleBlock with SegmentedVector to reduce indirect memory accesses when calling suspect, r=mccr8,nfroyd
--HG--
extra : rebase_source : e74be6bfb9efbba9361d2ce3c22518379a332200
2017-05-05 00:49:22 +03:00
Bill McCloskey 34a4f034bb Bug 1359245 - Remove references to context from the cycle collector (r=mccr8)
MozReview-Commit-ID: 1QoNEiZMvBf
2017-04-27 15:34:46 -07:00
Bill McCloskey 267ad1f2b0 Bug 1359245 - Initial support for cooperative contexts (r=mccr8)
This patch adds initial support for cooperatively scheduled
CycleCollectedJSContexts.

MozReview-Commit-ID: 5pfPubHUanL
2017-04-27 15:34:46 -07:00
Bill McCloskey 207af130a2 Bug 1343396 - Split CycledCollectedJSContext and XPCJSContext into separate context and runtime classes (r=mccr8)
To run JS in separate cooperative threads, we need to split up per-thread state
from per-runtime state. This patch does that for XPConnect.

MozReview-Commit-ID: 407SlJ7nR6v
2017-04-18 17:01:04 -07:00
Eric Rahm 807d2c062b Bug 1353544 - Replace usage of PR_BEGIN_MACRO in xpcom. r=froydnj 2017-04-05 18:06:44 -07:00
Jon Coppeard 2c0a710941 Bug 1335751 - Check all gray marking state before cycle collection in debug builds r=mccr8 2017-03-05 09:23:33 +00:00
Jon Coppeard e81021b4be Backed out 2 changesets (bug 1335751) for mochitest devtools failures 2017-03-05 12:37:31 +00:00
Jon Coppeard ac9f0854be Bug 1335751 - Check all gray marking state before cycle collection in debug builds r=mccr8 2017-03-05 09:23:33 +00:00
Jon Coppeard 12f4fcc022 Bug 1342181 - Run weak mapping fixup after GC before CC r=mccr8 2017-03-05 09:19:24 +00:00
Olli Pettay 0364dbc792 Bug 1326507, remove NS_IMPL_CYCLE_COLLECTION_TRAVERSE_SCRIPT_OBJECTS, r=mccr8
--HG--
extra : rebase_source : 3ae1207308de120b7299b13ecaa95dd1612b3459
2017-01-03 21:47:55 +02:00
Olli Pettay 8aa472a93c Bug 1326507, trace after traverse, r=mccr8
--HG--
extra : rebase_source : 496a186306372567fb917c2d21488fe5a87c5f75
2017-01-03 21:46:49 +02:00
Jan de Mooij 158805801d Bug 1325075 - Fix Value::isGCThing footgun, stop returning true for NullValue. r=jonco,baku 2016-12-26 16:40:21 +01:00
Andrew McCreight 75d98cae1d Bug 1322536 - Make some mScanInProgress checks fatal in release builds. r=smaug
MozReview-Commit-ID: 2hNtePfSxkF

--HG--
extra : rebase_source : 7434f50ef61e01297c87be15e34ecdb00acdb9f6
2016-12-08 11:03:42 -10:00
Jon Coppeard 4f90a79a0c Bug 1310147 - Rename AutoAssertOnGC to AutoAssertNoGC r=sfink 2016-10-27 11:03:53 +01:00
Jon Coppeard 123936bacd Bug 1306382 - Automatically ExposeToActiveJS when reading out of a TenuredHeap<T> r=sfink r=mccr8 2016-10-07 13:58:37 +02:00
Terrence Cole 0edc10380c Bug 1297558 - Use a read barrier on Heap to ExposeToActiveJS r=sfink r=mccr8 2016-02-07 09:08:55 -08:00
Andrew McCreight 5b9c57b7a2 Bug 1301301, part 1 - Unify NoteJSObject and NoteJSScript into NoteJSChild. r=smaug
This will let my next patch pass in other GC things.

This should not change behavior in any important way.

MozReview-Commit-ID: FykviKKvQzI
2016-09-23 15:42:13 -07:00
Andi-Bogdan Postelnicu cbfe4ec772 Bug 1282408 - Add ignore-initialization-check annotation to mCurrPi from CCGraphBuilder. r=mccr8
MozReview-Commit-ID: FFgmluIyvbs
2016-09-19 16:23:40 +03:00
Terrence Cole 1a8989f3c0 Bug 1004276 - Assert that we do not GC in the wrong parts of the CC; r=mccr8
--HG--
extra : rebase_source : f5a339a1c7e356e5d726b5323b860f3cbe490ec3
2014-04-30 16:06:25 -07:00
Jan de Mooij a53986bf29 Bug 1302448 part 1 - Rename CycleCollectedJSRuntime to CycleCollectedJSContext. r=mccr8
--HG--
rename : xpcom/base/CycleCollectedJSRuntime.cpp => xpcom/base/CycleCollectedJSContext.cpp
rename : xpcom/base/CycleCollectedJSRuntime.h => xpcom/base/CycleCollectedJSContext.h
extra : rebase_source : 075214b5057f151520926715b6154e99ae80a0b3
2016-09-14 15:47:32 +02:00
Andrew McCreight f9e6d19361 Bug 1299934 - Run shutdown collection in workers in opt builds. r=froydnj,smaug
We skip running shutdown collection on the main thread, because we're
exiting right after, but the process still continues when we shut down
a worker, so we should always do collections there.

MozReview-Commit-ID: IQZItm1qWXW

--HG--
extra : rebase_source : 4fd89e36db124a524c240e806a885dee3b68979e
2016-09-08 13:04:30 -07:00
Andrew McCreight d9ec3221fa Bug 1301796 - Remove racy code in nsCycleCollector::RegisterJSRuntime. r=smaug
MozReview-Commit-ID: 18MW4EYbaC8

--HG--
extra : rebase_source : a8e4b2fbf14f5ffd6acde32bf3ba96443c8ab5d4
2016-09-09 15:14:15 -07:00
Nicholas Nethercote 34dcc7b852 Bug 1299384 - Use MOZ_MUST_USE with NS_warn_if_impl(). r=erahm.
This change avoids lots of false positives for Coverity's CHECKED_RETURN
warning, caused by NS_WARN_IF's current use in both statement-style and
expression-style.

In the case where the code within the NS_WARN_IF has side-effects, I made the
following change.

> NS_WARN_IF(NS_FAILED(FunctionWithSideEffects()));
> -->
> Unused << NS_WARN_IF(NS_FAILED(FunctionWithSideEffects()));

In the case where the code within the NS_WARN_IF lacks side-effects, I made the
following change.

> NS_WARN_IF(!condWithoutSideEffects);
> -->
> NS_WARNING_ASSERTION(condWithoutSideEffects, "msg");

This has two improvements.
- The condition is not evaluated in non-debug builds.
- The sense of the condition is inverted to the familiar "this condition should
  be true" sense used in assertions.

A common variation on the side-effect-free case is the following.

> nsresult rv = Fn();
> NS_WARN_IF_(NS_FAILED(rv));
> -->
> DebugOnly<nsresult rv> = Fn();
> NS_WARNING_ASSERTION(NS_SUCCEEDED(rv), "Fn failed");

--HG--
extra : rebase_source : 58788245021096efa8372a9dc1d597a611d45611
2016-09-02 17:12:24 +10:00
Nicholas Nethercote b71747b2ac Bug 1299727 - Rename NS_WARN_IF_FALSE as NS_WARNING_ASSERTION. r=erahm.
The new name makes the sense of the condition much clearer. E.g. compare:

  NS_WARN_IF_FALSE(!rv.Failed());

with:

  NS_WARNING_ASSERTION(!rv.Failed());

The new name also makes it clearer that it only has effect in debug builds,
because that's standard for assertions.

--HG--
extra : rebase_source : 886e57a9e433e0cb6ed635cc075b34b7ebf81853
2016-09-01 15:01:16 +10:00
Nicholas Nethercote c2306345d5 Bug 1297658 - Avoid unnecessary checking in memory reporters. r=erahm.
This patch removes checking of all the callback calls in memory reporter
CollectReport() functions, because it's not useful.

The patch also does some associated clean-up.

- Replaces some uses of nsIMemoryReporterCallback with the preferred
  nsIHandleReportCallback typedef.

- Replaces aCallback/aCb/aClosure with aHandleRepor/aData for CollectReports()
  parameter names, for consistency.

- Adds MOZ_MUST_USE/[must_use] in a few places in nsIMemoryReporter.idl.

- Uses the MOZ_COLLECT_REPORT macro in all suitable places.

Overall the patch reduces code size by ~300 lines and reduces the size of
libxul by about 37 KiB on my Linux64 builds.

--HG--
extra : rebase_source : e94323614bd10463a0c5134a7276238a7ca1cf23
2016-08-24 15:23:45 +10:00
Nicholas Nethercote 8f84642d03 Bug 1296164 (part 2) - Use [must_use] on nsIFile.open{NSPR,ANSI}FileDesc(). r=froydnj.
The patch also changes RemoteOpenFileChild::OpenNSPRFileDesc() so that it
cannot succeed with a null fd, so that checking just the return value is
sufficient.

--HG--
extra : rebase_source : cc40bbcf2a9991edc9d3da3fb624d27db50b4996
2016-08-18 15:27:16 +10:00
Nathan Froyd 41ba39b3b0 Bug 1295197 - use non-null-checked operator new in xpcom/; r=erahm
The standard placement new function is declared to not throw, which
means that, per spec, a null check on its result is required. There are
a number of places throughout xpcom/ where we know that we are passing
non-null pointers to placement new (and receiving them as a return
value), and we are therefore doing useless work performing these null
checks.

Therefore, we should be using an operator new overload that doesn't
require the null check. MFBT has just such an overload, so use that.
2016-08-22 18:40:10 -04:00