........
  r399887 | dlee | 2013-09-26 10:41:47 -0500 (Thu, 26 Sep 2013) | 1 line
  
  Minor performance bump by not allocate manager variable struct if we don't need it
........
  r400138 | dlee | 2013-09-30 10:24:00 -0500 (Mon, 30 Sep 2013) | 23 lines
  
  Stasis performance improvements
  
  This patch addresses several performance problems that were found in
  the initial performance testing of Asterisk 12.
  
  The Stasis dispatch object was allocated as an AO2 object, even though
  it has a very confined lifecycle. This was replaced with a straight
  ast_malloc().
  
  The Stasis message router was spending an inordinate amount of time
  searching hash tables. In this case, most of our routers had 6 or
  fewer routes in them to begin with. This was replaced with an array
  that's searched linearly for the route.
  
  We more heavily rely on AO2 objects in Asterisk 12, and the memset()
  in ao2_ref() actually became noticeable on the profile. This was
  #ifdef'ed to only run when AO2_DEBUG was enabled.
  
  After being misled by an erroneous comment in taskprocessor.c during
  profiling, the wrong comment was removed.
  
  Review: https://reviewboard.asterisk.org/r/2873/
........
  r400178 | dlee | 2013-09-30 13:26:27 -0500 (Mon, 30 Sep 2013) | 24 lines
  
  Taskprocessor optimization; switch Stasis to use taskprocessors
  
  This patch optimizes taskprocessor to use a semaphore for signaling,
  which the OS can do a better job at managing contention and waiting
  that we can with a mutex and condition.
  
  The taskprocessor execution was also slightly optimized to reduce the
  number of locks taken.
  
  The only observable difference in the taskprocessor implementation is
  that when the final reference to the taskprocessor goes away, it will
  execute all tasks to completion instead of discarding the unexecuted
  tasks.
  
  For systems where unnamed semaphores are not supported, a really
  simple semaphore implementation is provided. (Which gives identical
  performance as the original taskprocessor implementation).
  
  The way we ended up implementing Stasis caused the threadpool to be a
  burden instead of a boost to performance. This was switched to just
  use taskprocessors directly for subscriptions.
  
  Review: https://reviewboard.asterisk.org/r/2881/
........
  r400180 | dlee | 2013-09-30 13:39:34 -0500 (Mon, 30 Sep 2013) | 28 lines
  
  Optimize how Stasis forwards are dispatched
  
  This patch optimizes how forwards are dispatched in Stasis.
  
  Originally, forwards were dispatched as subscriptions that are invoked
  on the publishing thread. This did not account for the vast number of
  forwards we would end up having in the system, and the amount of work it
  would take to walk though the forward subscriptions.
  
  This patch modifies Stasis so that rather than walking the tree of
  forwards on every dispatch, when forwards and subscriptions are changed,
  the subscriber list for every topic in the tree is changed.
  
  This has a couple of benefits. First, this reduces the workload of
  dispatching messages. It also reduces contention when dispatching to
  different topics that happen to forward to the same aggregation topic
  (as happens with all of the channel, bridge and endpoint topics).
  
  Since forwards are no longer subscriptions, the bulk of this patch is
  simply changing stasis_subscription objects to stasis_forward objects
  (which, admittedly, I should have done in the first place.)
  
  Since this required me to yet again put in a growing array, I finally
  abstracted that out into a set of ast_vector macros in
  asterisk/vector.h.
  
  Review: https://reviewboard.asterisk.org/r/2883/
........
  r400181 | dlee | 2013-09-30 13:48:57 -0500 (Mon, 30 Sep 2013) | 28 lines
  
  Remove dispatch object allocation from Stasis publishing
  
  While looking for areas for performance improvement, I realized that an
  unused feature in Stasis was negatively impacting performance.
  
  When a message is sent to a subscriber, a dispatch object is allocated
  for the dispatch, containing the topic the message was published to, the
  subscriber the message is being sent to, and the message itself.
  
  The topic is actually unused by any subscriber in Asterisk today. And
  the subscriber is associated with the taskprocessor the message is being
  dispatched to.
  
  First, this patch removes the unused topic parameter from Stasis
  subscription callbacks.
  
  Second, this patch introduces the concept of taskprocessor local data,
  data that may be set on a taskprocessor and provided along with the data
  pointer when a task is pushed using the ast_taskprocessor_push_local()
  call. This allows the task to have both data specific to that
  taskprocessor, in addition to data specific to that invocation.
  
  With those two changes, the dispatch object can be removed completely,
  and the message is simply refcounted and sent directly to the
  taskprocessor.
  
  Review: https://reviewboard.asterisk.org/r/2884/
........

Merged revisions 399887,400138,400178,400180-400181 from http://svn.asterisk.org/svn/asterisk/branches/12


git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@400186 65c4cc65-6c06-0410-ace0-fbb531ad65f3

This patch adds the ability to create a serializer from a thread pool. A
serializer is a ast_taskprocessor with the same contract as a default
taskprocessor (tasks execute serially) except instead of executing out
of a dedicated thread, execution occurs in a thread from a
ast_threadpool. Think of it as a lightweight thread.

While it guarantees that each task will complete before executing the
next, there is no guarantee as to which thread from the pool individual
tasks will execute. This normally only matters if your code relys on
thread specific information, such as thread locals.

This patch also fixes a bug in how the 'was_empty' parameter is computed
for the push callback, and gets rid of the unused 'shutting_down' field.

Review: https://reviewboard.asterisk.org/r/2323/


git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@381326 65c4cc65-6c06-0410-ace0-fbb531ad65f3

* Add a max_size option for threadpools. Also added a test for this option.
* Fixed comments to be more accurate and have fewer typos.
* Updated copyright dates on new files.



git-svn-id: https://origsvn.digium.com/svn/asterisk/team/mmichelson/threadpool@379375 65c4cc65-6c06-0410-ace0-fbb531ad65f3

git-svn-id: https://origsvn.digium.com/svn/asterisk/team/mmichelson/threadpool@379125 65c4cc65-6c06-0410-ace0-fbb531ad65f3

Now user data is allocated by the creator of the taskprocessor
listener and that user data is passed into ast_taskprocessor_listener_alloc().
Similarly, freeing of the user data is left up to the user himself. He can
free the data when the taskprocessor shuts down, or he can choose to hold
onto it if it makes sense to do so.

This, unsurprisingly, makes threadpool allocation a LOT cleaner now.



git-svn-id: https://origsvn.digium.com/svn/asterisk/team/mmichelson/threadpool@379120 65c4cc65-6c06-0410-ace0-fbb531ad65f3

The new thread creation test fails because Asterisk locks up
while trying to lock a taskprocessor.

While trying to debug that, I found a race condition during taskprocessor
creation where a default taskprocessor listener could try to operate on
a partially started taskprocessor. This was fixed by adding a new callback
to taskprocessor listeners.

Then while testing that change, I found some bugs in the taskprocessor
tests where I was not properly unlocking when done with a lock. Scoped
locks have spoiled me a bit.

I still have not figured out why the threadpool thread creation test
is locking up.



git-svn-id: https://origsvn.digium.com/svn/asterisk/team/mmichelson/threadpool@377368 65c4cc65-6c06-0410-ace0-fbb531ad65f3

git-svn-id: https://origsvn.digium.com/svn/asterisk/team/mmichelson/threadpool@376498 65c4cc65-6c06-0410-ace0-fbb531ad65f3

This throws a random number to each task and stores a copy locally. After
all tasks have executed, the data is checked to be sure tasks executed
in the correct order.



git-svn-id: https://origsvn.digium.com/svn/asterisk/team/mmichelson/threadpool@376413 65c4cc65-6c06-0410-ace0-fbb531ad65f3

git-svn-id: https://origsvn.digium.com/svn/asterisk/team/mmichelson/threadpool@376412 65c4cc65-6c06-0410-ace0-fbb531ad65f3

Changes the ast_cond_wait() to an ast_cond_timedwait() so that
if there is an issue, we'll never wait forever for the task to
finish execution.



git-svn-id: https://origsvn.digium.com/svn/asterisk/team/mmichelson/threadpool@376411 65c4cc65-6c06-0410-ace0-fbb531ad65f3

git-svn-id: https://origsvn.digium.com/svn/asterisk/team/mmichelson/threadpool@376383 65c4cc65-6c06-0410-ace0-fbb531ad65f3

This helps account for the fact that it is unknown just
how many references may exist for a given taskprocessor
listener, so simply unreffing it from the taskprocessor
shutdown function is not enough to convey the gravity
of the situation.

By putting in a shutdown callback, it now becomes clear
to the listener not to try to do any further operations
on the taskprocessor.



git-svn-id: https://origsvn.digium.com/svn/asterisk/team/mmichelson/threadpool@376381 65c4cc65-6c06-0410-ace0-fbb531ad65f3

This makes it easier to follow and tests more thoroughly.



git-svn-id: https://origsvn.digium.com/svn/asterisk/team/mmichelson/threadpool@376140 65c4cc65-6c06-0410-ace0-fbb531ad65f3

are properly called.

The new test is a bit raw in that it uses magic numbers, but that
can be improved upon easily.



git-svn-id: https://origsvn.digium.com/svn/asterisk/team/mmichelson/threadpool@376124 65c4cc65-6c06-0410-ace0-fbb531ad65f3

This just queues a task and ensures that the task
gets executed by the default taskprocessor listener.



git-svn-id: https://origsvn.digium.com/svn/asterisk/team/mmichelson/threadpool@376120 65c4cc65-6c06-0410-ace0-fbb531ad65f3