| Age | Commit message (Collapse) | Author |
|---|
|
brad/merge-maint-a-to-dev
|
|
* #1354 Make coroutines use LLCoros::Mutex instead of LLMutex
* #1354 Fix some more unsafe coroutine executions.
* #1354 Implement changes requested by Nat
|
|
LF, and trim trailing whitespaces as needed
|
|
|
|
We actively use event pumps's connections in threads, make sure nothing
modifies list of connections during reset.
And in case this doesn't fix the issue list affected pump before it
crashes to have a better idea of what is going on.
|
|
LazyEventAPI is a registrar that implicitly instantiates some particular
LLEventAPI subclass on demand: that is, when LLEventPumps::obtain() tries to
find an LLEventPump by the registered name.
This leverages the new LLEventPumps::registerPumpFactory() machinery. Fix
registerPumpFactory() to adapt the passed PumpFactory to accept TypeFactory
parameters (two of which it ignores). Supplement it with
unregisterPumpFactory() to support LazyEventAPI instances with lifespans
shorter than the process -- which may be mostly test programs, but still a
hole worth closing. Similarly, add unregisterTypeFactory().
A LazyEventAPI subclass takes over responsibility for specifying the
LLEventAPI's name, desc, field, plus whatever add() calls will be needed to
register the LLEventAPI's operations. This is so we can (later) enhance
LLLeapListener to consult LazyEventAPI instances for not-yet-instantiated
LLEventAPI metadata, as well as enumerating existing LLEventAPI instances.
The trickiest part of this is capturing calls to the various
LLEventDispatcher::add() overloads in such a way that, when the LLEventAPI
subclass is eventually instantiated, we can replay them in the new instance.
LLEventAPI acquires a new protected constructor specifically for use by a
subclass registered by a companion LazyEventAPI. It accepts a const reference
to LazyEventAPIParams, intended to be opaque to the LLEventAPI subclass; the
subclass must declare a constructor that accepts and forwards the parameter
block to the new LLEventAPI constructor. The implementation delegates to the
existing LLEventAPI constructor, plus it runs deferred add() calls.
LLDispatchListener now derives from LLEventStream instead of containing it as
a data member. The reason is that if LLEventPumps::obtain() implicitly
instantiates it, LLEventPumps's destructor will try to destroy it by deleting
the LLEventPump*. If the LLEventPump returned by the factory function is a
data member of an outer class, that won't work so well. But if
LLDispatchListener (and by implication, LLEventAPI and any subclass) is
derived from LLEventPump, then the virtual destructor will Do The Right Thing.
Change LLDispatchListener to *not* allow tweaking the LLEventPump name. Since
the overwhelming use case for LLDispatchListener is LLEventAPI, accepting but
silently renaming an LLEventAPI subclass would ensure nobody could reach it.
Change LLEventDispatcher's use of std::enable_if to control the set of add()
overloads available for the intended use cases. Apparently this formulation is
just as functional at the method declaration point, while avoiding the need to
restate the whole enable_if expression at the method definition point.
Add lazyeventapi_test.cpp to exercise.
|
|
and registerTypeFactory().
Untested.
This will support registering just-in-time LLEventAPI instances, instantiated
on demand.
|
|
Mostly converted some boost pointers to std ones
Made ~LLNotificationChannelBase() more explicit to get a bit more data on location of another crash that likely happens when cleaning mItems
|
|
Superficially crash happens in disconnect() inside signal's deconstructor. Manual cleanup should help figuring out if crash happens due to named or anonymous listeners
|
|
Having a map from std::string to a factory function returning LLEventPump* is
a cool idea, especially since you could statically populate such a map with
string literals and little lambdas.
Unfortunately, static initialization of any data is a bad idea when control
can reach consuming code before that module's static data are constructed.
Since LLEventPumps is already an LLSingleton, it's simple enough to make its
map non-static and initialize it in the constructor.
But another recent static factory-function map was introduced in
llleaplistener.cpp to support the LLLeapListener::newpump() operation. That
involves no LLSingletons.
Introduce LLEventPumps::make(name, tweak, type) to instantiate an LLEventPump
subclass of the specified type with specified (name, tweak) parameters.
Instances returned by make() are owned by LLEventPumps, as with obtain().
Introduce LLEventPumps::BadType exception for when the type string isn't
recognized.
LLEventPumps::obtain() can then simply call make() when the specified instance
name doesn't already exist. The configuration data used internally by obtain()
becomes { string instance name, string subclass name }. Although this too is
currently initialized in the LLEventPumps constructor, migrating it to a
configuration file would now be far more straightforward than before.
LLLeapListener::newpump(), too, can call LLEventPumps::make() with the
caller-specified type string. This eliminates that static factory map.
newpump() must catch BadType and report the error back to its invoker.
Given that the LLEventPump subclass instances returned by make() are owned by
LLEventPumps rather than LLLeapListener, there is no further need for the
LLLeapListener::mEventPumps ptr_map, which was used only to manage lifetime.
Also remove LLLeapListener's "killpump" operation since LLEventPumps provides
no corresponding functionality.
|
|
No one uses LLEventQueue to defer posted events until the next mainloop tick
-- and with LLCoros moving to Boost.Fiber, cross-coroutine event posting works
that way anyway, making LLEventQueue pretty unnecessary.
The static RegisterFlush instance in llevents.cpp was used to call
LLEventPumps::flush() once per mainloop tick, which in turn called flush() on
every registered LLEventPump. But the only reason for that mechanism was to
support LLEventQueue. In fact, when LLEventMailDrop overrode its flush()
method for something quite different, it was startling to find that the new
flush() override was being called once per frame -- which caused at least one
fairly mysterious bug. Remove RegisterFlush. Both LLEventPumps::flush() and
LLEventPump::flush() remain for now, though intended usage is unclear.
Eliminating LLEventQueue means we must at least repurpose
LLEventPumps::mQueueNames, a map intended to make LLEventPumps::obtain()
instantiate an LLEventQueue rather than the default LLEventPump. Replace it
with mFactories, a map from desired instance name to a callable returning
LLEventPump*. New map initialization syntax plus lambda support allows us to
populate that map at compile time with little lambdas returning the correct
subclass instance.
Similarly, LLLeapListener::newpump() used to check the ["type"] entry in the
LLSD request specifically for "LLEventQueue". Introduce another such map in
llleaplistener.cpp for potential future extensibility.
Eliminate the LLEventQueue-specific test.
|
|
Overriding virtual LLEventPump::flush() for the semantic of discarding
LLEventMailDrop's queued events turns out not to be such a great idea, because
LLEventPumps::flush(), which calls every registered LLEventPump's flush()
method, is called every mainloop tick. The first time we hit a use case in
which we expected LLEventMailDrop to hold queued events across a mainloop tick,
we were baffled that they were never delivered.
Moving that logic to a separate method specific to LLEventMailDrop resolves
that problem. Naming it discard() clarifies its intended functionality.
|
|
LLEventDetail::visit_and_connect() promised special treatment for the
specific case when an LLEventPump::listen() listener was composed of (possibly
nested) boost::bind() objects storing boost::weak_ptr values -- specifically
boost::bind() rather than std::bind or lambdas, specifically boost::weak_ptr
rather than std::weak_ptr.
Outside of self-tests, it does not appear that anyone actually uses that
support.
There is good reason not to: it's a silent side effect of a complicated
compile-time inspection that could be silently derailed by use of std::bind()
or a lambda or a std::weak_ptr. Can you be sure you've engaged that promise?
How?
A more robust guarantee can be achieved by storing an LLTempBoundConnection in
the transient object itself. When the object is destroyed, the listener is
disconnected. Normal C++ rules around object destruction guarantee it. This
idiom is widely used.
There are a couple good reasons to remove the visit_and_connect() machinery:
* boost::bind() and boost::weak_ptr do not constitute the wave of the future.
Preferring those constructs to lambdas and std::weak_ptr penalizes new code,
whether by silently failing or by discouraging use of modern idioms.
* The visit_and_connect() machinery was always complicated, and apparently
never very robust. Most of its promised features have been commented out
over the years. Making the code base simpler, clearer and more maintainable
is always a useful effect.
LLEventDetail::visit_and_connect() was also used by the four
LLNotificationChannelBase::connectMumble() methods. Streamline those as well.
Of course, remove related test code.
|
|
The only usage of any of this was in test code.
|
|
This is like the existing reset() method, except that reset() is specifically
intended for shutdown: it disables every existing LLEventPump in such a way
that it cannot be subsequently reused. (The original idea was to disconnect
listeners in DLLs unloaded at shutdown.)
clear() forcibly disconnects all existing listeners, but leaves LLEventPumps
ready for reuse. This is useful (e.g.) for test programs to reset the state of
LLEventPumps between individual test functions.
|
|
|
|
mail drop does not have any outstanding events.
|
|
Previously, LLEventMailDrop would send only the first queued event to a
newly-connected listener. If you wanted to flush all queued events, you'd have
to "pump" the queue by repeatedly disconnecting and reconnecting -- with no
good way to know when you'd caught up.
The new behavior makes LLEventMailDrop resemble a multi-valued future: a
rendezvous between producer and consumer that, once connected, pushes values
rather than requiring them to be pulled (as with a simple queue) -- regardless
of the relative order in which post() and listen() are called.
|
|
LLEventPump's destructor was using LLEventPumps::instance() to unregister the
LLEventPump instance from LLEventPumps. Evidently, though, there are lingering
LLEventPump instances that persist even after the LLSingletonBase::deleteAll()
call destroys the LLEventPumps LLSingleton instance. These were resurrecting
LLEventPumps -- pointlessly, since a newly-resurrected LLEventPumps instance
can have no knowledge of the LLEventPump instance! Unregistering is
unnecessary!
What we want is a reference we can bind into each LLEventPump instance that
allows us to safely test whether the LLEventPumps instance still exists.
LLHandle is exactly that. Make LLEventPumps an LLHandleProvider and bind its
LLHandle in each LLEventPump's constructor; then the destructor can unregister
only when LLEventPumps still exists.
|
|
|
|
|
|
|
|
A shocking number of LLSingleton subclasses had public constructors -- and in
several instances, were being explicitly instantiated independently of the
LLSingleton machinery. This breaks the new LLSingleton dependency-tracking
machinery. It seems only fair that if you say you want an LLSingleton, there
should only be ONE INSTANCE!
Introduce LLSINGLETON() and LLSINGLETON_EMPTY_CTOR() macros. These handle the
friend class LLSingleton<whatevah>;
and explicitly declare a private nullary constructor.
To try to enforce the LLSINGLETON() convention, introduce a new pure virtual
LLSingleton method you_must_use_LLSINGLETON_macro() which is, as you might
suspect, defined by the macro. If you declare an LLSingleton subclass without
using LLSINGLETON() or LLSINGLETON_EMPTY_CTOR() in the class body, you can't
instantiate the subclass for lack of a you_must_use_LLSINGLETON_macro()
implementation -- which will hopefully remind the coder.
Trawl through ALL LLSingleton subclass definitions, sprinkling in
LLSINGLETON() or LLSINGLETON_EMPTY_CTOR() as appropriate. Remove all explicit
constructor declarations, public or private, along with relevant 'friend class
LLSingleton<myself>' declarations. Where destructors are declared, move them
into private section as well. Where the constructor was inline but nontrivial,
move out of class body.
Fix several LLSingleton abuses revealed by making ctors/dtors private:
LLGlobalEconomy was both an LLSingleton and the base class for
LLRegionEconomy, a non-LLSingleton. (Therefore every LLRegionEconomy instance
contained another instance of the LLGlobalEconomy "singleton.") Extract
LLBaseEconomy; LLGlobalEconomy is now a trivial subclass of that.
LLRegionEconomy, as you might suspect, now derives from LLBaseEconomy.
LLToolGrab, an LLSingleton, was also explicitly instantiated by
LLToolCompGun's constructor. Extract LLToolGrabBase, explicitly instantiated,
with trivial subclass LLToolGrab, the LLSingleton instance.
(WARNING: LLToolGrabBase methods have an unnerving tendency to go after
LLToolGrab::getInstance(). I DO NOT KNOW what should be the relationship
between the instance in LLToolCompGun and the LLToolGrab singleton instance.)
LLGridManager declared a variant constructor accepting (const std::string&),
with the comment:
// initialize with an explicity grid file for testing.
As there is no evidence of this being called from anywhere, delete it.
LLChicletBar's constructor accepted an optional (const LLSD&). As the LLSD
parameter wasn't used, and as there is no evidence of it being passed from
anywhere, delete the parameter.
LLViewerWindow::shutdownViews() was checking LLNavigationBar::
instanceExists(), then deleting its getInstance() pointer -- leaving a
dangling LLSingleton instance pointer, a land mine if any subsequent code
should attempt to reference it. Use deleteSingleton() instead.
~LLAppViewer() was calling LLViewerEventRecorder::instance() and then
explicitly calling ~LLViewerEventRecorder() on that instance -- leaving the
LLSingleton instance pointer pointing to an allocated-but-destroyed instance.
Use deleteSingleton() instead.
|
|
This means that an exception derived from LLContinueError thrown in an
LLEventPump listener won't prevent other listeners on the same LLEventPump
from receiving that event.
|
|
This also introduces LLContinueError for exceptions which should interrupt
some part of viewer processing (e.g. the current coroutine) but should attempt
to let the viewer session proceed.
Derive all existing viewer exception classes from LLException rather than from
std::runtime_error or std::logic_error.
Use BOOST_THROW_EXCEPTION() rather than plain 'throw' to enrich the thrown
exception with source file, line number and containing function.
|
|
tracking.
|
|
|
|
timeout it. Also some cleanup on LLSD construction in vivox.
|
|
|
|
delivery
|
|
|
|
conditional compile switches. Begin switch from statemachine to coroutine.
|
|
|
|
consolidated most indra-specific constants in llcommon under indra_constants.h
fixed issues with operations on mixed unit types (implicit and explicit)
made LL_INFOS() style macros variadic in order to subsume other logging methods
such as ll_infos
added optional tag output to error recorders
|
|
|
|
Each LLEventAPI method that generates a reply needs to extract the name of the
reply LLEventPump from the request, typically from a ["reply"] key, copy the
["reqid"] value from request to reply, locate the reply LLEventPump and send
the enriched reply object. Encapsulate in sendReply() function before we
proliferate doing all that by hand too many more times.
|
|
|
|
|
|
|
|
|
|
Replace LLEventPump's boost::scoped_ptr<LLStandardSignal> with
boost::shared_ptr. Take a local stack copy of that shared_ptr in post()
methods, and invoke the signal through that copy. This guards against scenario
in which LLEventPump gets destroyed during signal invocation. (See Jira for
details.) Re-enable Mani's test case that used to crash.
Introduce ll_template_cast<> to allow a template function to recognize a
parameter of a particular type.
Introduce LLListenerWrapper mechanism to support wrapper objects for
LLEventPump listeners. You instantiate an LLListenerWrapper subclass object
inline in the listen() call (typically with llwrap<>), passing it the real
listener, trusting it to forward the eventual call.
Introduce prototypical LLCoutListener and LLLogListener subclasses for
illustrative and diagnostic purposes. Test that LLLogListener doesn't block
recognizing LLEventTrackable base class bound into wrapped listener.
|
|
|
|
--HG--
branch : avatar-pipeline
|
|
The fix deletes all LLEventPumps boost::signal objects prior to unloading any dlls.
reviewed by Nat.
|
|
I'll need to rebuild that, plus a couple other minor clenaups.
|
|
Moves libllcommon.so to a staging dir for unit tests to work
and gets rid of LL_COMMON_API in forward declarations
|
|
merged all changes, post copy, from the following branches:
linden/brachnes/enable-o-v
user/cg/qar-1538
user/mani/viewer2-enable-o-v
|
|
up to viewer-2.0.0-3
|
|
|
|
ignore-dead-branch
|
