Commit Graph
126 Commits
Author SHA1 Message Date
Daniel Holbert 8715febbdf Bug 1733729: Fix non-unified build bustage in xpcom directory. r=mccr8
These changes were necessary for me to build successfully with this in my
mozconfig:
 ac_add_options --enable-debug
 ac_add_options --disable-unified-build
 ac_add_options --enable-profiling

While I'm here, I removed a duplicate include in InputTaskManager.h, and made
nsThreadPool.cpp include its own header first, to match Gecko common practice.

Differential Revision: https://phabricator.services.mozilla.com/D127503
2021-10-05 00:29:56 +00:00
Gerald Squelart f25e1c0093 Bug 1691589 - Reduce reliance on GeckoProfiler.h when only labels (and maybe markers) are needed - r=necko-reviewers,geckoview-reviewers,sg,agi,florian
There are no code changes, only #include changes.
It was a fairly mechanical process: Search for all "AUTO_PROFILER_LABEL", and in each file, if only labels are used, convert "GeckoProfiler.h" into "ProfilerLabels.h" (or just add that last one where needed).
In some files, there were also some marker calls but no other profiler-related calls, in these cases "GeckoProfiler.h" was replaced with both "ProfilerLabels.h" and "ProfilerMarkers.h", which still helps in reducing the use of the all-encompassing "GeckoProfiler.h".

Differential Revision: https://phabricator.services.mozilla.com/D104588
2021-02-16 04:44:19 +00:00
Florian Quèze bf1a24a932 Bug 1688300 - Add 'Runnable' profiler markers showing when runnables are executed and a 'Task' label frame showing which tasks are started by TaskController, r=bas,gerald.
Differential Revision: https://phabricator.services.mozilla.com/D102803
2021-01-26 08:57:19 +00:00
Simon Giesecke fbb9c742c8 Bug 1679272 - Include ScopeExit.h exactly where used. r=andi
Differential Revision: https://phabricator.services.mozilla.com/D98888
2020-12-07 14:25:59 +00:00
Simon Giesecke c83a083a0c Bug 1676365 - Move SpinEventLoopUntil to separate header. r=#xpcom-reviewers
Differential Revision: https://phabricator.services.mozilla.com/D96556

Depends on D96554
2020-11-23 16:10:41 +00:00
Bas Schouten a22a4d9cd8 Bug 1669256 - Part 3: Remove priority related code from ThreadEventQueue. r=smaug
Differential Revision: https://phabricator.services.mozilla.com/D92711
2020-10-09 16:57:40 +00:00
Simon Giesecke f2ebaa3c94 Bug 1642991 - Use RemoveElementsBy where easily possible. r=froydnj
Differential Revision: https://phabricator.services.mozilla.com/D78055
2020-06-08 09:01:48 +00:00
Sylvestre Ledru 17947d45e5 Bug 1519636 - Reformat recent changes to the Google coding style r=andi
# ignore-this-changeset

Differential Revision: https://phabricator.services.mozilla.com/D76451
2020-05-25 07:42:38 +00:00
Nathan Froyd b31d9c68a8 Bug 1639632 - convert xpcom/ to inherit from mozilla::Runnable where possible; r=KrisWright
A little less boilerplate is nice.

Differential Revision: https://phabricator.services.mozilla.com/D76178
2020-05-20 20:37:58 +00:00
Honza Bambas 3869f6b082 Bug 1634950 - Add event dispatch-and-run MOZ_LOG to our xpcom threads, r=froydnj
Differential Revision: https://phabricator.services.mozilla.com/D73622
2020-05-12 12:48:49 +00:00
Gerald Squelart e9549c4bd9 Bug 1609907 - Remove AUTO_PROFILER_THREAD_SLEEP before mozilla::CondVar waits - r=mstange
CondVar already calls `AUTO_PROFILER_THREAD_SLEEP`, which shouldn't be called recursively.

mozilla::Monitor also uses CondVar, so it shouldn't be surrounded by `AUTO_PROFILER_THREAD_SLEEP` either.

Depends on D72850

Differential Revision: https://phabricator.services.mozilla.com/D72851
2020-05-01 22:12:22 +00:00
Andreas Farre 4ee379404b Bug 1620594 - Part 2: Use SchedulerGroup::Dispatch instead of SystemGroup::Dispatch. r=nika
Depends on D67631

Differential Revision: https://phabricator.services.mozilla.com/D67632
2020-04-07 15:16:33 +00:00
Eric Rahm b1ae7c8937 Bug 1241518 - Part 4: Switch nsThread to use UniquePtr r=KrisWright,kmag
This switches over one usage of `nsAutoPtr` that was just used to scope an allocation, a stack variable is used instead. The shutdown contexts array is switched over to hold `UniquePtr`s which required adding a helper to find elements in the array as `UniquePtr` does not auto-convert to its pointer type.

Differential Revision: https://phabricator.services.mozilla.com/D58317
2020-01-07 00:06:08 +00:00
Eric Rahm 6b9537eb30 Bug 1241518 - Part 1: Remove unused nsAutoPtr.h includes in xpcom/ r=kmag
This removes various unused `#include "nsAutoPtr.h"` in `xpcom/`. Additionally
adds a few includes to the media stack.

Differential Revision: https://phabricator.services.mozilla.com/D58282
2020-01-07 00:06:01 +00:00
Nathan Froyd 3a348f669a Bug 1604549 - remove scriptability and classinfo from nsThreadPool; r=bzbarsky
There's no reason for this class to be exposed to script; it'd be
awkward to use from script anyway, because the runnables script
dispatches would be running off the main thread.  Thread pool listeners
likewise can't be exposed to script or implemented by script, because
they are also running off the main thread.  The classinfo stuff just
seems like a relic from a bygone era.

Differential Revision: https://phabricator.services.mozilla.com/D57485
2019-12-17 17:12:14 +00:00
Randell Jesup 61e6e4a4a0 Bug 1572337: Make GetRunningEventDelay handle threadpools r=froydnj
Threadpools run an event that then runs other events, so we need to tweak
things for GetRunningEventDelay()

Differential Revision: https://phabricator.services.mozilla.com/D44058
2019-11-08 21:07:45 +00:00
Razvan Maries 4b01b5de2c Backed out 3 changesets (bug 1572337) ford perma fails. CLOSED TREE
Backed out changeset 00da7156d3fa (bug 1572337)
Backed out changeset 4eda65e054d8 (bug 1572337)
Backed out changeset ea6d5b4b038b (bug 1572337)
2019-11-07 17:29:46 +02:00
Randell Jesup ff0890a5d2 Bug 1572337: Make GetRunningEventDelay handle threadpools r=froydnj
Threadpools run an event that then runs other events, so we need to tweak
things for GetRunningEventDelay()

Differential Revision: https://phabricator.services.mozilla.com/D44058
2019-11-07 12:53:32 +00:00
Nathan Froyd acc7da2eaa Bug 1583964 - make IsOnCurrentThread for thread pools constant-time; r=KrisWright
It's fairly common for clients to hold a pointer to some private thread
and then inquire about `IsOnCurrentThread` in debug checks.  As we
migrate these private threads into some `nsIEventTarget` implementation
that might be running on thread pools, we need to ensure that those
`IsOnCurrentThread` continues to be relatively cheap.  The current
implementation for thread pools is not particularly efficient.

The inefficiency comes from having to iterate over all the threads in
the pool.  But there's no need to do this; we can just have each thread
set a particular thread-local variable to the thread pool it's running
on, and check the value of that thread-local variable instead.

Differential Revision: https://phabricator.services.mozilla.com/D47139
2019-09-25 22:02:08 +00:00
Nathan Froyd c0301736f2 Bug 1573160 - shutdown newly-spawned threads if we're shutting down; r=mccr8
Differential Revision: https://phabricator.services.mozilla.com/D41551
2019-08-15 17:29:26 +00:00
Honza Bambas 40b47ef22d Bug 1535361 - Let stream transport service idle with 5 threads instead of only one to not create/kill threads in quick bursts; add option to thread pool to shorten the idle timeout progressively with number of idle thread to save memory, r=dragana
Differential Revision: https://phabricator.services.mozilla.com/D23845
2019-03-19 14:35:12 +00:00
Randell Jesup 97291212a0 Bug 1522150: Rename EventPriority to EventQueuePriority to avoid name conflict with MacOS r=froyd 2019-01-26 12:18:05 -05:00
Ehsan Akhgari 498e2df734 Bug 1521000 - Part 2: Adjust our clang-format rules to include spaces after the hash for nested preprocessor directives r=sylvestre
# ignore-this-changeset
2019-01-18 10:16:18 +01:00
Sylvestre Ledru e5a134f73a Bug 1511181 - Reformat everything to the Google coding style r=ehsan a=clang-format
# ignore-this-changeset
2018-11-30 11:46:48 +01:00
shindli 869e127abc Merge inbound to mozilla-central. a=merge 2018-11-03 11:40:50 +02:00
Gerald Squelart 0b695de54f Bug 1503468 - Add more AUTO_PROFILER_THREAD_SLEEP's - r=mstange
These are around wait functions that already had an IDLE marker.

Differential Revision: https://phabricator.services.mozilla.com/D10671
2018-11-02 16:26:50 +00:00
Valentin Gosu 309f2134af Bug 1503725 - Do not deallocate nsThreadShutdownContext when leaking thread. r=erahm
Sometimes when we call ShutdownWithTimeout on a thread pool, the unresponsive
thread that we leak will actually complete before the main thread is done.
In that case, the thread will dereference the thread shutdown context, so
we must intentionally leak it too.

Differential Revision: https://phabricator.services.mozilla.com/D10645
2018-11-02 17:38:37 -04:00
Valentin Gosu 807bd5e62a Bug 1500861 - Add shutdownWithTimeout method to nsIThreadPool r=froydnj,erahm
This method is necessary because some threads might be stuck making blocking
calls. This means the thread is not processing any events, and we're unable
to safely terminate it. Our solution here is to leak the stuck threads
instead of waiting for them and potentially causing a shutdown hang.

Differential Revision: https://phabricator.services.mozilla.com/D9601
2018-10-26 18:46:00 +00:00
Nathan Froyd d0ef392c36 Bug 1486400 - add task dispatch/run delays for ChaosMode; r=jesup 2018-09-19 12:02:09 -04:00
Csoregi Natalia b816eb7909 Merge mozilla-central to autoland. a=merge CLOSED TREE 2018-06-02 01:03:45 +03:00
Emilio Cobos Álvarez 4b8b5e1717 Bug 1465585: Switch from mozilla::Move to std::move. r=froydnj
This was done automatically replacing:

  s/mozilla::Move/std::move/
  s/ Move(/ std::move(/
  s/(Move(/(std::move(/

Removing the 'using mozilla::Move;' lines.

And then with a few manual fixups, see the bug for the split series..

MozReview-Commit-ID: Jxze3adipUh
2018-06-01 10:45:27 +02:00
Markus Stange 0fc7e751f0 Bug 1462784 - Annotate idle stacks in nsThreadPool. r=froydnj
MozReview-Commit-ID: GtHBHQiXpf3
2018-05-24 08:58:44 -04:00
Csoregi Natalia 27ccec9c4c Backed out 18 changesets (bug 1462784) for ESlint failure on FlameGraph.js:1297. CLOSED TREE
Backed out changeset 79556798ff9f (bug 1462784)
Backed out changeset 88321efb673b (bug 1462784)
Backed out changeset 7880f9dc7023 (bug 1462784)
Backed out changeset 71fe35fd1f7e (bug 1462784)
Backed out changeset a543b94b049a (bug 1462784)
Backed out changeset d1ca8b0f2221 (bug 1462784)
Backed out changeset 68eabfbf3c16 (bug 1462784)
Backed out changeset 34e71c789903 (bug 1462784)
Backed out changeset 6fe79d1ca1bd (bug 1462784)
Backed out changeset e5ad2e525ea9 (bug 1462784)
Backed out changeset 329645ff1e23 (bug 1462784)
Backed out changeset e09c38853172 (bug 1462784)
Backed out changeset 0663d1a6d2da (bug 1462784)
Backed out changeset 106967fc29d2 (bug 1462784)
Backed out changeset 99b4a433a8e5 (bug 1462784)
Backed out changeset 1d38a4cf5a4a (bug 1462784)
Backed out changeset 692017229de6 (bug 1462784)
Backed out changeset c2911a626671 (bug 1462784)
2018-06-01 23:42:00 +03:00
Markus Stange 93a58dd0b2 Bug 1462784 - Annotate idle stacks in nsThreadPool. r=froydnj
MozReview-Commit-ID: GtHBHQiXpf3
2018-05-24 08:58:44 -04:00
Noemi Erli 3946d17555 Backed out 18 changesets (bug 1462784) for failures in devtools/client/performance/test/unit/test_tree-model-08.js on a CLOSED TREE
Backed out changeset a74d36598442 (bug 1462784)
Backed out changeset c8192175f360 (bug 1462784)
Backed out changeset cde492240e99 (bug 1462784)
Backed out changeset 8c8d30fa406c (bug 1462784)
Backed out changeset ad3802ffb780 (bug 1462784)
Backed out changeset 2fe10732076c (bug 1462784)
Backed out changeset 268a72b7c3c4 (bug 1462784)
Backed out changeset 4055eb6c3bc6 (bug 1462784)
Backed out changeset 3901070e2e60 (bug 1462784)
Backed out changeset 2faf787fbbdf (bug 1462784)
Backed out changeset 8f06963c7c6f (bug 1462784)
Backed out changeset 036e6f64e224 (bug 1462784)
Backed out changeset e670f156a603 (bug 1462784)
Backed out changeset cd39588aece4 (bug 1462784)
Backed out changeset 2ac65d100fa2 (bug 1462784)
Backed out changeset ea05ff70a51d (bug 1462784)
Backed out changeset 8a06c0ba42f7 (bug 1462784)
Backed out changeset 52ed9a039ad2 (bug 1462784)
2018-06-01 01:06:29 +03:00
Markus Stange 90efd81b23 Bug 1462784 - Annotate idle stacks in nsThreadPool. r=froydnj
MozReview-Commit-ID: GtHBHQiXpf3
2018-05-24 08:58:44 -04:00
Nika Layzell b4c313c43d Bug 1437167 - Part 1: Stop using PRIntervalTime as the argument to CondVar::Wait and Monitor::Wait, r=mstange, r=froydnj 2018-04-10 17:49:47 -04:00
Philippe Normand c58d812a5b Bug 1404305 - Include nsCOMArray.h and SystemGroup in nsThreadPool.cpp to fix the build in non-unified mode. r=froydnj
MozReview-Commit-ID: 1bZHw4TS9L1
2017-09-28 14:32:26 +01:00
Sean V Kelley b86c624d47 Bug 1376858 - Use SystemGroup to label nsIThread::AsyncShutdown, r=billm
MozReview-Commit-ID: IDPE6zhEnVI
2017-08-27 18:58:54 -07:00
Bill McCloskey e1b7db0651 Bug 1382922 - Refactor event queue to allow multiple implementations (r=erahm)
This patch refactors the nsThread event queue to clean it up and to make it easier to restructure. The fundamental concepts are as follows:

Each nsThread will have a pointer to a refcounted SynchronizedEventQueue. A SynchronizedEQ takes care of doing the locking and condition variable work when posting and popping events. For the actual storage of events, it delegates to an AbstractEventQueue data structure. It keeps a UniquePtr to the AbstractEventQueue that it uses for storage.

Both SynchronizedEQ and AbstractEventQueue are abstract classes. There is only one concrete implementation of SynchronizedEQ in this patch, which is called ThreadEventQueue. ThreadEventQueue uses locks and condition variables to post and pop events the same way nsThread does. It also encapsulates the functionality that DOM workers need to implement their special event loops (PushEventQueue and PopEventQueue). In later Quantum DOM work, I plan to have another SynchronizedEQ implementation for the main thread, called SchedulerEventQueue. It will have special code for the cooperatively scheduling threads in Quantum DOM.

There are two concrete implementations of AbstractEventQueue in this patch: EventQueue and PrioritizedEventQueue. EventQueue replaces the old nsEventQueue. The other AbstractEventQueue implementation is PrioritizedEventQueue, which uses multiple queues for different event priorities.

The final major piece here is ThreadEventTarget, which splits some of the code for posting events out of nsThread. Eventually, my plan is for multiple cooperatively scheduled nsThreads to be able to share a ThreadEventTarget. In this patch, though, each nsThread has its own ThreadEventTarget. The class's purpose is just to collect some related code together.

One final note: I tried to avoid virtual dispatch overhead as much as possible. Calls to SynchronizedEQ methods do use virtual dispatch, since I plan to use different implementations for different threads with Quantum DOM. But all the calls to EventQueue methods should be non-virtual. Although the methods are declared virtual, all the classes used are final and the concrete classes involved should all be known through templatization.

MozReview-Commit-ID: 9Evtr9oIJvx
2017-08-16 20:55:43 -07:00
Bill McCloskey 04b1343a42 Bug 1361164 - Add infallible IsOnCurrentThread to nsIEventTarget (r=froydnj)
MozReview-Commit-ID: 12bk9hQ7Wnv
2017-06-12 20:20:08 -07:00
Nathan Froyd 3b526b2200 Bug 1359490 - add an event loop spinning abstraction function; r=gerald
This function is arguably nicer than calling NS_ProcessNextEvent
manually, is slightly more efficient, and will enable better auditing
for NS_ProcessNextEvent when we do Quantum DOM scheduling changes.
2017-05-15 09:34:19 -04:00
Bill McCloskey f79edeabc9 Bug 1339289 - Give names to a lot of common runnables (r=ehsan)
MozReview-Commit-ID: 5IdvK6kgoAW
2017-02-15 12:30:01 -08:00
Markus Stange b742926dd8 Bug 1323100 - Use nsThreadPoolNaming::GetNextThreadName and NS_NewNamedThread in nsThreadPool. r=froydnj
MozReview-Commit-ID: 6IB5yvJtAQm
2016-12-22 00:14:30 +01:00
Olli Pettay b7858f5e83 bug 1306591, add secondary event queue to let high priority messages to be processed sooner, r=billm 2016-11-08 14:05:45 +02:00
Ben Kelly a63344ac8f Bug 1300118 P0 Rename NS_DISPATCH_TAIL to NS_DISPATCH_AT_END to avoid confusion with AbstractThread::TailDispatch. r=bholley 2016-09-13 20:12:15 -07:00
Wes Kocher 38957ba9d1 Backed out 6 changesets (bug 1300118) for frequent Windows VM Xpcshell failures a=backout CLOSED TREE
Backed out changeset 24f714ad248c (bug 1300118)
Backed out changeset 6fe61ed092e8 (bug 1300118)
Backed out changeset dd18e29b64bf (bug 1300118)
Backed out changeset 50882fb40551 (bug 1300118)
Backed out changeset 60e82c1780e5 (bug 1300118)
Backed out changeset e40df52925d5 (bug 1300118)
2016-09-12 16:34:41 -07:00
Ben Kelly 948492dc89 Bug 1300118 P0 Rename NS_DISPATCH_TAIL to NS_DISPATCH_AT_END to avoid confusion with AbstractThread::TailDispatch. r=bholley 2016-09-12 12:32:20 -07:00
Ben Kelly 449d185850 Backout rev 5c7368370ff9 to 980659720b86 (bug 1300118 and bug 1300658) for incorrect bug number in commit message. r=me 2016-09-12 12:29:17 -07:00
Ben Kelly 58e677435a Bug 1300118 P0 Rename NS_DISPATCH_TAIL to NS_DISPATCH_AT_END to avoid confusion with AbstractThread::TailDispatch. r=bholley 2016-09-12 11:21:01 -07:00