objdir/dist/include/mozilla/DeferredFinalize.h:30:1: error: unknown type name 'MOZ_NEVER_INLINE'
objdir/dist/include/mozilla/dom/BindingUtils.h:2776:5: error: no matching function for call to 'DeferredFinalize'
xpcom/base/AvailableMemoryTracker.cpp:55,40:: error: expected class name
xpcom/base/AvailableMemoryTracker.cpp:59,3:: error: 'AddRef' marked 'override' but does not override any member functions
xpcom/base/AvailableMemoryTracker.cpp:59,3:: error: 'QueryInterface' marked 'override' but does not override any member functions
xpcom/base/AvailableMemoryTracker.cpp:59,3:: error: 'Release' marked 'override' but does not override any member functions
xpcom/base/AvailableMemoryTracker.cpp:61,29:: error: unknown type name 'nsIHandleReportCallback'
xpcom/base/AvailableMemoryTracker.cpp:65,37:: error: use of undeclared identifier 'KIND_OTHER'
xpcom/base/AvailableMemoryTracker.cpp:65,49:: error: use of undeclared identifier 'UNITS_COUNT_CUMULATIVE'
xpcom/base/AvailableMemoryTracker.cpp:77,1:: error: invalid application of 'sizeof' to an incomplete type 'const QITableEntry []'
xpcom/base/AvailableMemoryTracker.cpp:77,1:: error: static_cast from '(anonymous namespace)::LowEventsReporter *' to 'nsISupports *', which are not related by inheritance, is not allowed
xpcom/base/AvailableMemoryTracker.cpp:77,38:: error: unknown type name 'nsIMemoryReporter'
xpcom/base/AvailableMemoryTracker.cpp:77,38:: error: use of undeclared identifier 'nsIMemoryReporter'
xpcom/base/AvailableMemoryTracker.cpp:187,3:: error: use of undeclared identifier 'RegisterLowMemoryEventsPhysicalDistinguishedAmount'
xpcom/base/AvailableMemoryWatcherWin.cpp:31,47:: error: base class has incomplete type
xpcom/base/AvailableMemoryWatcherWin.cpp:37,3:: error: unknown type name 'NS_DECL_NSIOBSERVER'
xpcom/base/AvailableMemoryWatcherWin.cpp:38,3:: error: 'virtual' can only appear on non-static member functions
xpcom/base/AvailableMemoryWatcherWin.cpp:38,3:: error: expected ';' at end of declaration list
xpcom/base/AvailableMemoryWatcherWin.cpp:38,3:: error: static_assert failed due to requirement 'std::is_convertible<mozilla::nsAvailableMemoryWatcher *, nsINamed *>::value' "nsITimerCallback implementations must also implement nsINamed"
xpcom/base/AvailableMemoryWatcherWin.cpp:39,3:: error: must use 'enum' tag to refer to type 'nsresult' in this scope
xpcom/base/AvailableMemoryWatcherWin.cpp:42,3:: error: must use 'enum' tag to refer to type 'nsresult' in this scope
xpcom/base/AvailableMemoryWatcherWin.cpp:102,1:: error: cannot initialize object parameter of type 'mozilla::nsAvailableMemoryWatcherBase' with an expression of type 'mozilla::nsAvailableMemoryWatcher'
xpcom/base/AvailableMemoryWatcherWin.cpp:102,1:: error: invalid application of 'sizeof' to an incomplete type 'const QITableEntry []'
xpcom/base/AvailableMemoryWatcherWin.cpp:102,1:: error: must use 'enum' tag to refer to type 'nsresult' in this scope
xpcom/base/AvailableMemoryWatcherWin.cpp:102,1:: error: static_cast from 'mozilla::nsAvailableMemoryWatcher *' to 'nsINamed *', which are not related by inheritance, is not allowed
xpcom/base/AvailableMemoryWatcherWin.cpp:102,1:: error: static_cast from 'mozilla::nsAvailableMemoryWatcher *' to 'nsIObserver *', which are not related by inheritance, is not allowed
xpcom/base/AvailableMemoryWatcherWin.cpp:102,1:: error: static_cast from 'mozilla::nsAvailableMemoryWatcher *' to 'nsITimerCallback *', which are not related by inheritance, is not allowed
xpcom/base/AvailableMemoryWatcherWin.cpp:103,59:: error: 'nsIObserver' does not refer to a value
xpcom/base/AvailableMemoryWatcherWin.cpp:103,59:: error: incomplete type 'nsIObserver' named in nested name specifier
xpcom/base/AvailableMemoryWatcherWin.cpp:140,45:: error: cannot initialize a parameter of type 'nsIObserver *' with an rvalue of type 'mozilla::nsAvailableMemoryWatcher *'
xpcom/base/AvailableMemoryWatcherWin.cpp:140,5:: error: must use 'enum' tag to refer to type 'nsresult' in this scope
xpcom/base/CycleCollectedJSRuntime.cpp:1605,7:: error: use of undeclared identifier 'PR_GetEnv'
xpcom/base/CycleCollectedJSRuntime.cpp:1605:7: error: use of undeclared identifier 'PR_GetEnv'
xpcom/base/DeferredFinalize.cpp:12,15:: error: out-of-line definition of 'DeferredFinalize' does not match any declaration in namespace 'mozilla'
xpcom/base/DeferredFinalize.cpp:12:15: error: out-of-line definition of 'DeferredFinalize' does not match any declaration in namespace 'mozilla'
xpcom/base/MemoryTelemetry.cpp:48,27:: error: unused variable 'kTotalMemoryCollectorTimeout' [-Werror,-Wunused-const-variable]
xpcom/base/MemoryTelemetry.cpp:48:27: error: unused variable 'kTotalMemoryCollectorTimeout' [-Werror,-Wunused-const-variable]
xpcom/base/nsErrorService.cpp:42,21:: error: use of function template name with no prior declaration in function call with explicit template arguments is a C++20 extension [-Werror,-Wc++20-extensions]
xpcom/base/nsErrorService.cpp:42,21:: error: use of undeclared identifier 'MakeUnique'; did you mean 'mozilla::MakeUnique'?
xpcom/base/nsErrorService.cpp:42:21: error: use of function template name with no prior declaration in function call with explicit template arguments is a C++20 extension [-Werror,-Wc++20-extensions]
xpcom/base/nsErrorService.cpp:42:21: error: use of undeclared identifier 'MakeUnique'; did you mean 'mozilla::MakeUnique'?
xpcom/base/nsMacPreferencesReader.mm:20:12: error: implicit instantiation of undefined template 'nsTSubstring<char16_t>'
xpcom/base/nsMacPreferencesReader.mm:24:60: error: use of undeclared identifier 'NSString'
xpcom/base/nsMacPreferencesReader.mm:24:69: error: expected expression
xpcom/base/nsMacPreferencesReader.mm:26:48: error: unknown type name 'NSArray'
xpcom/base/nsMacPreferencesReader.mm:28:30: error: use of undeclared identifier 'NSString'
xpcom/base/nsMacPreferencesReader.mm:29:30: error: no matching function for call to 'MakeStringSpan'
xpcom/base/nsMacPreferencesReader.mm:29:51: error: instance method '-UTF8String' not found (return type defaults to 'id') [-Werror,-Wobjc-method-access]
xpcom/base/nsMacPreferencesReader.mm:30:22: error: instance method '-isKindOfClass:' not found (return type defaults to 'id') [-Werror,-Wobjc-method-access]
xpcom/base/nsMacPreferencesReader.mm:30:37: error: receiver type 'int (*)(int)' is not 'id' or interface pointer, consider casting it to 'id' [-Werror,-Wreceiver-expr]
xpcom/base/nsMacPreferencesReader.mm:30:37: error: use of undeclared identifier 'NSNumber'; did you mean 'isnumber'?
xpcom/base/nsMacPreferencesReader.mm:30:46: error: instance method '-class' not found (return type defaults to 'id') [-Werror,-Wobjc-method-access]
xpcom/base/nsMacPreferencesReader.mm:31:27: error: cannot initialize a parameter of type 'int64_t' (aka 'long long') with an rvalue of type 'id'
xpcom/base/nsMacPreferencesReader.mm:31:33: error: instance method '-longLongValue' not found (return type defaults to 'id') [-Werror,-Wobjc-method-access]
xpcom/base/nsMacPreferencesReader.mm:32:37: error: use of undeclared identifier 'NSArray'
xpcom/base/nsMacPreferencesReader.mm:34:7: error: no matching function for call to 'EvaluateArray'
xpcom/base/nsMacPreferencesReader.mm:36:37: error: use of undeclared identifier 'NSDictionary'
xpcom/base/nsMacPreferencesReader.mm:38:7: error: no matching function for call to 'EvaluateDict'
xpcom/base/nsMacPreferencesReader.mm:44:60: error: use of undeclared identifier 'NSString'
xpcom/base/nsMacPreferencesReader.mm:44:69: error: expected expression
xpcom/base/nsSystemInfo.cpp:415,35:: error: no template named 'NotNull'; did you mean 'NonNull'?
xpcom/base/nsSystemInfo.cpp:419,25:: error: member reference type 'NonNull<IWSCProductList *>' is not a pointer; did you mean to use '.'?
xpcom/base/nsSystemInfo.cpp:419,27:: error: no member named 'get_Count' in 'mozilla::dom::NonNull<IWSCProductList *>'
xpcom/base/nsSystemInfo.cpp:426,19:: error: member reference type 'NonNull<IWSCProductList *>' is not a pointer; did you mean to use '.'?
xpcom/base/nsSystemInfo.cpp:426,21:: error: no member named 'get_Item' in 'mozilla::dom::NonNull<IWSCProductList *>'
xpcom/base/nsSystemInfo.cpp:484,17:: error: use of undeclared identifier 'ArrayLength'; did you mean 'mozilla::ArrayLength'?
xpcom/base/nsSystemInfo.cpp:484,47:: error: use of undeclared identifier 'ArrayLength'; did you mean 'mozilla::ArrayLength'?
xpcom/base/nsSystemInfo.cpp:487,36:: error: use of undeclared identifier 'ArrayLength'; did you mean 'mozilla::ArrayLength'?
xpcom/base/nsSystemInfo.cpp:500,10:: error: no matching function for call to 'EnumWSCProductList'
xpcom/base/nsSystemInfo.cpp:500,46:: error: use of undeclared identifier 'WrapNotNull'; did you mean 'mozilla::WrapNotNull'?
xpcom/base/nsTraceRefcnt.cpp:313,32:: error: use of undeclared identifier 'MakeUnique'
xpcom/base/nsTraceRefcnt.cpp:313:32: error: use of undeclared identifier 'MakeUnique'
xpcom/base/nsTraceRefcnt.cpp:461,19:: error: no matching member function for call to 'WithEntryHandle'
xpcom/base/nsTraceRefcnt.cpp:461:19: error: no matching member function for call to 'WithEntryHandle'
xpcom/base/nsTraceRefcnt.cpp:468,33:: error: use of function template name with no prior declaration in function call with explicit template arguments is a C++20 extension [-Werror,-Wc++20-extensions]
xpcom/base/nsTraceRefcnt.cpp:468,33:: error: use of undeclared identifier 'MakeUnique'
xpcom/base/nsTraceRefcnt.cpp:468:33: error: use of undeclared identifier 'MakeUnique'
xpcom/base/nsTraceRefcnt.cpp:736,20:: error: use of undeclared identifier '_write'
Differential Revision: https://phabricator.services.mozilla.com/D127036
This patch introduces an XPCOM object which is represented by the single instance of
`nsAvailableMemoryWatcherBase` so that `nsAvailableMemoryWatcher` can synchronously
access `TabUnloader`.
We currently implement a watcher class for Windows only. For other platforms, what
we need to do is to define a class inherinting `nsAvailableMemoryWatcherBase` and
a simple factory method `CreateAvailableMemoryWatcher()` returning an instance of
that class.
Differential Revision: https://phabricator.services.mozilla.com/D118393
This patch splits `nsAvailableMemoryWatcher` into 1) an nsISupports-derived class
`nsAvailableMemoryWatcherBase` and 2) a platform-specific class `nsAvailableMemoryWatcher`,
taking out the 2) part as a new file AvailableMemoryWatcherWin.cpp without any change.
Test cases for `nsAvailableMemoryWatcher` will be added by a subsequent patch.
Differential Revision: https://phabricator.services.mozilla.com/D117669
This patch introduces an XPCOM object which is represented by the single instance of
`nsAvailableMemoryWatcherBase` so that `nsAvailableMemoryWatcher` can synchronously
access `TabUnloader`.
We currently implement a watcher class for Windows only. For other platforms, what
we need to do is to define a class inherinting `nsAvailableMemoryWatcherBase` and
a simple factory method `CreateAvailableMemoryWatcher()` returning an instance of
that class.
Differential Revision: https://phabricator.services.mozilla.com/D118393
This patch splits `nsAvailableMemoryWatcher` into 1) an nsISupports-derived class
`nsAvailableMemoryWatcherBase` and 2) a platform-specific class `nsAvailableMemoryWatcher`,
taking out the 2) part as a new file AvailableMemoryWatcherWin.cpp without any change.
Test cases for `nsAvailableMemoryWatcher` will be added by a subsequent patch.
Differential Revision: https://phabricator.services.mozilla.com/D117669
`nsAvailableMemoryWatcher::mTimer` was nullptr when `LowMemoryCallback()` tried to
initialize a timer via `OnLowMemory()`. There are two possible reasons.
The first case is that NS_NewTimer() returned nullptr because the available memory was
already low when initializing `nsAvailableMemoryWatcher`. In this case, we should not
register the callback.
The second case is the low-memory callback was queued while shutting down
`nsAvailableMemoryWatcher` (just before we unregiter the callback). We do refcount
the object and use the mutex correctly, but we touch the nulled out member. We should
make an early return if the object was shut down.
Differential Revision: https://phabricator.services.mozilla.com/D118745
This patch reinstate the code to call `SaveMemoryReport` when we enter a low-memory
situation from a normal state, which was removed by bug 1586236.
Differential Revision: https://phabricator.services.mozilla.com/D115960
After bug 1586236, we use the memory resource notification object to detect a low
memory situation on Windows, which is signaled when the available physical memory
is low. If the available physical memory is low, however, it's possible that there
is still commit space enough for the application to run. In such a situation, we
don't want to make aggressive efforts to reduce memory usage.
This patch makes sure we send the memory pressure event (both New and Ongoing) only
when the available commit space is lower than the threshold value defined by the pref
"browser.low_commit_space_threshold_mb". Its default value is set to 200MB
based on our telemetry data indicating ~60% of OOM crashes with <100MB, ~75% with <300MB.
To use the pref in `nsAvailableMemoryWatcher`, this patch moves the call to
`AvailableMemoryTracker::Init()` to `XRE_mainRun()`. It was in `NS_InitXPCOM`
because the old initialization code hooked APIs and needed to be done while
the process has only one thread (bug 741540). The current `AvailableMemoryTracker`
does not use hooks, so it doesn't have to be initialized that early.
Differential Revision: https://phabricator.services.mozilla.com/D115605
This patch uses the low memory resource notification facility to detect
scenarios where physical memory is running low without polling. This is a
significant change compared to the previous behavior which measured both
available virtual memory (only on 32-bit builds) and available commit space.
Since we're not trying to avoid OOMs anymore we don't save memory reports
anymore when hitting a low-memory condition.
Differential Revision: https://phabricator.services.mozilla.com/D110354
This patch uses the low memory resource notification facility to detect
scenarios where physical memory is running low without polling. This is a
significant change compared to the previous behavior which measured both
available virtual memory (only on 32-bit builds) and available commit space.
Since we're not trying to avoid OOMs anymore we don't save memory reports
anymore when hitting a low-memory condition.
Differential Revision: https://phabricator.services.mozilla.com/D50471
This patch uses the low memory resource notification facility to detect
scenarios where physical memory is running low without polling. This is a
significant change compared to the previous behavior which measured both
available virtual memory (only on 32-bit builds) and available commit space.
Since we're not trying to avoid OOMs anymore we don't save memory reports
anymore when hitting a low-memory condition.
Differential Revision: https://phabricator.services.mozilla.com/D50471
This reverts the available memory tracker back to its old implementation that
relied on GlobalMemoryStatusEx() instead of the potentially expensive
K32GetPerformanceInfo(). In spite of its name GlobalMemoryStatusEx() may return
per-process values instead of global ones which might lead to underestimating
the actual memory pressure. Because of this limitation this change is a
stop-gap while we replace this code with a non-polling implementation based on
QueryMemoryResourceNotification().
Differential Revision: https://phabricator.services.mozilla.com/D45338
I'd like to rely on always getting a stop event when the system recovers from a memory pressure situation. AIUI this already happens on Android platforms but not on Windows. This adds the stop event for Windows.
Differential Revision: https://phabricator.services.mozilla.com/D35679
This uses GetPerformanceInformation() to measure how much commit space is
available globally. Previously we relied on GlobalMemoryStatusEx() but in
spite of the function name the commit-space values it returns are
*per process* and thus unreliable. Available physical memory measurement has
also been removed since it was never used and we don't plan on using it
anytime soon.
Differential Revision: https://phabricator.services.mozilla.com/D35018
After analyzing crash ping data we've established that the current low-memory
detection threshold is too low, there are still a fair number of crashes
happening above it. A 50% increase to 384 MiB should be just about right in
the light of recent telemetry data.
Differential Revision: https://phabricator.services.mozilla.com/D23826
nsCRTGlue.cpp has a few data tables that get broken up badly by
clang-format, so just disable the formatting. There are some more
tables in this file, but the arrangement doesn't seem to matter for
them.
The bloat log header is just a little bit of ASCII art, so just ignore
it in formatting so it is easier to read.
MOZ_COLLECT_REPORT uses very wide string constants that get rewrapped
in an ugly way by clang-format, so just disable the formatting for
them.
Depends on D13185
Differential Revision: https://phabricator.services.mozilla.com/D13186
This moves the code that detects very low memory scenarios and grabs memory
reports from the main thread event-loop to the available memory tracker.
Besides removing the overhead of the check from the event-loop code this
increases the likeliness of the reports being gathered by sampling at a
higher frequency but only when we already detected a low-memory scenario. Last
but not least this add checks for low commit-space detection alongside low
virtual-memory detection.
Differential Revision: https://phabricator.services.mozilla.com/D3669
This introduces the machinery needed to generate crash annotations from a YAML
file. The relevant C++ functions are updated to take a typed enum. JavaScript
calls are unaffected but they will throw if the string argument does not
correspond to one of the known entries in the C++ enum. The existing whitelists
and blacklists of annotations are also generated from the YAML file and all
duplicate code related to them has been consolidated. Once written out to the
.extra file the annotations are converted in string form and are no different
than the existing ones.
All existing annotations have been included in the list (and some obsolete ones
have been removed) and all call sites have been updated including tests where
appropriate.
This patch introduces a new polling mechanism to detect low-memory scenarios.
The timer fires at a relatively slow pace and stops whenever the user stops
interacting with Firefox to avoid consuming power needlessly. The polling rate
is up to 3 orders of magnitude slower than the current tracker and is
throttled when memory is running low. It also doesn't suffer from data races
that were possible with existing tracker.
Contrary to the old available memory tracker which relied on a
Windows-specific mechanism, this one could be made to work on other platforms
too. The current implementation only supports Windows 64-bit builds though.
MozReview-Commit-ID: CFHuTDqjPbL
There's no good reason why these should't just be constants. The patch also
appends "MiB" to some of the C++ values to make their meaning clearer.
This patch fixes one outright bug, and one inconsistency.
The bug is due to a prefname mismatch:
- ContentPrefs.cpp and AvailableMemoryTracker.cpp use
"memory.low_virtual_mem_threshold_mb".
- all.js uses "memory.low_virtual_memory_threshold_mb".
Which means that "memory.low_virtual_memory_threshold_mb" showed up in
about:config, but if you changed it nothing would happen because the callback
listened for changes to to "memory.low_virtual_mem_threshold_mb"!
Now for the inconsistency. The above means we actually use a value of 256 for
the virtual memory threshold, even though all.js says 128. But we *do* use a
value of 128 for the commit space threshold, because that's what all.js says
and that prefname is used correctly everywhere. The patch changes the commit
space threshold to 256 for consistency with the virtual memory threshold.
What a mess!
Free dirty pages appear as "page-cache" in about:memory reports, but
when minimizing memory, and more generally, on memory pressure, they're
currently not force-free()ed on desktop, while they are on mobile.
There doesn't seem to be much reason not to do it on desktop as well,
and it should help reduce the noise in "explicit" allocations measurements
on CI, too.
Masatoshi Kimura