........
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
The PRI and SS7 link control threads are not stopped correctly when the
chan_dahdi.so module is unloaded. The link control threads pri_dchannel()
and ss7_linkset() are not awakened from a poll() to cancel the thread.
* Added a SIGURG signal after requesting the thread cancel to break the
link control thread poll() immediately.
For SS7 it was slightly worse, the link poll() timeout would always be
whatever was the last libss7 scheduled event time used. If no libss7
scheduled event was pending, the thread could run more often than
necessary.
* Set nextms to 60 seconds for the ss7_linkset() poll() if there is no
other libss7 scheduled event.
........
Merged revisions 399818 from http://svn.asterisk.org/svn/asterisk/branches/1.8
........
Merged revisions 399834 from http://svn.asterisk.org/svn/asterisk/branches/11
........
Merged revisions 399842 from http://svn.asterisk.org/svn/asterisk/branches/12
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@399844 65c4cc65-6c06-0410-ace0-fbb531ad65f3
This change protects accesses of res_parking such that it can unload
safely once transient uses of its registered functions are complete.
The parking API has been restructured such that its consumers do not
have access to the vtable exposed by the parking provider, but instead
route through stubs to prevent consumers from holding on to function
pointers.
This adds calls to all the parking unload functions and moves
application loading and unloading into functions in
parking_applications.c similar to the rest of the parts of res_parking.
Review: https://reviewboard.asterisk.org/r/2763/
(closes issue ASTERISK-22142)
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@396890 65c4cc65-6c06-0410-ace0-fbb531ad65f3
This removes unused code, event types, IE pltypes, and event IE types
where possible and makes several functions private that were once
public. This includes a renumbering of the remaining event and IE types
which breaks binary compatibility with previous versions. The last
remaining consumers of the old event system (or parts thereof) are
main/security_events.c, res/res_security_log.c, tests/test_cel.c,
tests/test_event.c, main/cel.c, and the CEL backends.
Review: https://reviewboard.asterisk.org/r/2703/
(closes issue ASTERISK-22139)
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@396887 65c4cc65-6c06-0410-ace0-fbb531ad65f3
This code adds chan_dahdi the command 'dahdi create channels <range>'
(where <range> is a single <n>-<m> or 'new') and updates 'dahdi destroy
channel' with a similar 'dahdi destroy channels'. It allows DAHDI
channels and spans to be added after the initial channel load
(without destroying all other channels as in 'dahdi restart').
It also includes some fixes to the D-Channel / span destruction code
(r394552).
This change is intended to provide a hook for a script running from
udev once a span has been assigned ("registered") / unassigned
("unregistered") for its channels. The udev hook configures the span's
channels with dahdi_cfg -S, and can then ask Asterisk to create ethe
channels. See the scripts added to DAHDI-tools in 2.7.0.
Review: https://reviewboard.asterisk.org/r/1598/
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@396474 65c4cc65-6c06-0410-ace0-fbb531ad65f3
This commit is smaller than the initial review placed on review board. This is because
a change to allow for channel drivers to access parking functionality externally was
committed and invalidated quite a few of the changes initially made.
(closes issue ASTERISK-22039)
reported by Matt Jordan
Review: https://reviewboard.asterisk.org/r/2717
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@396103 65c4cc65-6c06-0410-ace0-fbb531ad65f3
This patch does the following:
* It adds support for externally initiated parking requests. In particular,
chan_skinny has a protocol level message that initiates a call park.
This patch now supports that option, as well as the protocol specific
mechanisms in chan_dahdi/sig_analog and chan_mgcp.
* A parking bridge features virtual table has been added that provides
access to the parking functionality that the Bridging API needs. This
includes requests to park an entire 'call' (with little or no additional
information, thank you chan_skinny), perform a blind transfer to a parking
extension, determine if an extension is a parking extension, as well as the
actual "do the parking" request from the Bridging API.
* Refactoring in chan_mgcp, chan_skinny, and chan_dahdi to make use of the new
functions
* The removal of some - but not all - dead parking code from features.c
This also fixed blind transferring a multi-party bridge to a parking lot (which
was implemented, but had at least one code path where using the parking features
kK might not have worked)
Review: https://reviewboard.asterisk.org/r/2710
(closes issue ASTERISK-22134)
Reported by: Matt Jordan
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@396028 65c4cc65-6c06-0410-ace0-fbb531ad65f3
This prevents XML documentation duplication by expanding channel and
bridge snapshot tags into channel and bridge snapshot parameter sets
with a given prefix or defaulting to no prefix. This also prevents
documentation from becoming fractured and out of date by keeping all
variations of the documentation in template form such that it only
needs to be updated once and keeps maintenance to a minimum.
Review: https://reviewboard.asterisk.org/r/2708/
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@395985 65c4cc65-6c06-0410-ace0-fbb531ad65f3
In working with res_stasis, I discovered a significant limitation to
the current structure of stasis_caching_topics: you cannot subscribe
to cache updates for a single channel/bridge/endpoint/etc.
To address this, this patch splits the cache away from the
stasis_caching_topic, making it a first class object. The stasis_cache
object is shared amongst individual stasis_caching_topics that are
created per channel/endpoint/etc. These are still forwarded to global
whatever_all_cached topics, so their use from most of the code does
not change.
In making these changes, I noticed that we frequently used a similar
pattern for bridges, endpoints and channels:
single_topic ----------------> all_topic
^
|
single_topic_cached ----+----> all_topic_cached
|
+----> cache
This pattern was extracted as the 'Stasis Caching Pattern', defined in
stasis_caching_pattern.h. This avoids a lot of duplicate code between
the different domain objects.
Since the cache is now disassociated from its upstream caching topics,
this also necessitated a change to how the 'guaranteed' flag worked
for retrieving from a cache. The code for handling the caching
guarantee was extracted into a 'stasis_topic_wait' function, which
works for any stasis_topic.
(closes issue ASTERISK-22002)
Review: https://reviewboard.asterisk.org/r/2672/
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@395954 65c4cc65-6c06-0410-ace0-fbb531ad65f3
This patch renames the bridging* files to bridge*. This may seem pedantic
and silly, but it fits better in line with current Asterisk naming conventions:
* channel is not "channeling"
* monitor is not "monitoring"
etc.
A bridge is an object. It is a first class citizen in Asterisk. "Bridging" is
the act of using a bridge on a set of channels - and the API that fulfills that
role is more than just the action.
(closes issue ASTERISK-22130)
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@395378 65c4cc65-6c06-0410-ace0-fbb531ad65f3
Created a native_dahdi bridging technology for use with the new bridging
API.
The new bridging technology is part of the chan_dahdi channel driver
because it is very specific to that driver. Rather than include the new
code directly into chan_dahdi.c the new bridge technology is in its own
file and linked into chan_dahdi.so. A large part of this change is the
mechanical process of moving declarations around so chan_dahdi.c can be
split up into more files later.
* Changed the bridging core to pass NULL frames into the channel
technologies instead of discarding them. The channel technologies may
need the proding to determine if their configuration is still valid.
(closes issue ASTERISK-21886)
Reported by: Matt Jordan
Review: https://reviewboard.asterisk.org/r/2681/
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@395154 65c4cc65-6c06-0410-ace0-fbb531ad65f3
This adds CEL support for blind and attended transfers and call pickup.
During the course of adding this functionality I noticed that
CONF_ENTER, CONF_EXIT, and BRIDGE_TO_CONF events are particularly
useless without a bridge identifier, so I added that as well.
This adds tests for blind transfers, several types of attended
transfers, and call pickup.
The extra field in CEL records now consists of a JSON blob whose fields
are defined on a per-event basis.
Review: https://reviewboard.asterisk.org/r/2658/
(closes issue ASTERISK-21565)
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@394858 65c4cc65-6c06-0410-ace0-fbb531ad65f3
When a DAHDI device is removed at run-time it sends the event
DAHDI_EVENT_REMOVED on each channel. This is intended to signal the
userspace program to close the respective file handle, as the driver of
the device will need all of them closed to properly clean-up.
This event has long since been handled in chan_dahdi (chan_zap at the
time). However the event that is sent on a D-Channel of a "PRI" (ISDN)
span simply gets ignored.
This commit adds handling for closing the file descriptor (and shutting
down the span, while we're at it).
It also adds a CLI command 'pri destroy span <N>' to destroy the span
and its DAHDI channels.
Review: https://reviewboard.asterisk.org/r/726/
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@394552 65c4cc65-6c06-0410-ace0-fbb531ad65f3
from mixing different variants or general MFC-R2 settings within the same E1 line.
Most users do not have a problem with this since MFC-R2 lines are usually fractional E1s, or
the whole E1 has the same country variant and R2 settings.
In Venezuela however is common to have inbound MFC-R2 and outbound DTMF-R2 within the same E1.
This fix now properly parses the chan_dahdi.conf file to generate a new openr2 context every
time a new channel => section is found and the configuration was changed.
(closes issue ASTERISK-21117)
Reported by: Rafael Angulo
Related Elastix issue: http://bugs.elastix.org/view.php?id=1612
........
Merged revisions 394106 from http://svn.asterisk.org/svn/asterisk/branches/1.8
........
Merged revisions 394173 from http://svn.asterisk.org/svn/asterisk/branches/11
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@394263 65c4cc65-6c06-0410-ace0-fbb531ad65f3
This patch is the initial push to update Asterisk's CDR engine for the new
bridging framework. This patch guts the existing CDR engine and builds the new
on top of messages coming across Stasis. As changes in channel state and bridge
state are detected, CDRs are built and dispatched accordingly. This
fundamentally changes CDRs in a few ways.
(1) CDRs are now *very* reflective of the actual state of channels and bridges.
This means CDRs track well with what an actual channel is doing - which
is useful in transfer scenarios (which were previously difficult to pin
down). It does, however, mean that CDRs cannot be 'fooled'. Previous
behavior in Asterisk allowed for CDR applications, channels, and other
properties to be spoofed in parts of the code - this no longer works.
(2) CDRs have defined behavior in multi-party scenarios. This behavior will not
be what everyone wants, but it is a defined behavior and as such, it is
predictable.
(3) The CDR manipulation functions and applications have been overhauled. Major
changes have been made to ResetCDR and ForkCDR in particular. Many of the
options for these two applications no longer made any sense with the new
framework and the (slightly) more immutable nature of CDRs.
There are a plethora of other changes. For a full description of CDR behavior,
see the CDR specification on the Asterisk wiki.
(closes issue ASTERISK-21196)
Review: https://reviewboard.asterisk.org/r/2486/
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@391947 65c4cc65-6c06-0410-ace0-fbb531ad65f3
Features configuration is handled in its own API in
features_config.h and features_config.c. This way, features
configuration is accessible to anything that needs it.
In addition, features configuration has been altered to
be more channel-oriented. Most callers of features API
code will be supplying a channel so that the individual
channel's settings will be acquired rather than the global
setting.
Missing from this commit is XML documentation for the
features configuration. That will be handled in a separate
commit.
Review: https://reviewboard.asterisk.org/r/2578/
(issue ASTERISK-21542)
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@390751 65c4cc65-6c06-0410-ace0-fbb531ad65f3
This patch moves a number of AMI events over to the Stasis-Core message bus.
This includes:
* ChanSpyStart/Stop
* MonitorStart/Stop
* MusicOnHoldStart/Stop
* FullyBooted/Reload
* All Voicemail/MWI related events
In addition, it adds some Stasis-Core and AMI support for generic AMI messages,
refactors the message router in AMI to use a single router with topic
forwarding for the topics that AMI cares about, and refactors MWI message
types and topics to be more name compliant.
Review: https://reviewboard.asterisk.org/r/2532
(closes issue ASTERISK-21462)
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@389733 65c4cc65-6c06-0410-ace0-fbb531ad65f3
Breaks many things until they can be reworked. A partial list:
chan_agent
chan_dahdi, chan_misdn, chan_iax2 native bridging
app_queue
COLP updates
DTMF attended transfers
Protocol attended transfers
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@389378 65c4cc65-6c06-0410-ace0-fbb531ad65f3
The new inband_on_proceeding option causes Asterisk to assume inband audio
may be present when a PROCEEDING message is received.
Q.931 Section 5.1.2 says the network cannot assume that the CPE side has
attached to the B channel at this time without explicitly sending the
progress indicator ie informing the CPE side to attach to the B channel
for audio. However, some non-compliant ISDN switches send a PROCEEDING
without the progress indicator ie indicating inband audio is available and
assume that the CPE device has connected the media path for listening to
ringback and other messages.
ASTERISK-17834 which causes this issue was dealing with a non-compliant
network switch.
(closes issue ASTERISK-21151)
Reported by: Gianluca Merlo
Tested by: rmudgett
........
Merged revisions 384685 from http://svn.asterisk.org/svn/asterisk/branches/1.8
........
Merged revisions 384689 from http://svn.asterisk.org/svn/asterisk/branches/11
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@384696 65c4cc65-6c06-0410-ace0-fbb531ad65f3
Remove MWI's dependency on the event system by moving it to
Stasis-core. This also introduces forwarding topic pools in Stasis-core
which aggregate many dynamically allocated topics into a single primary
topic.
Review: https://reviewboard.asterisk.org/r/2368/
(closes issue ASTERISK-21097)
Patch-by: Kinsey Moore
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@383284 65c4cc65-6c06-0410-ace0-fbb531ad65f3
Asterisk maintains an internal cache for devices in the event subsystem. The
device state cache holds the state of each device known to Asterisk, such that
consumers of device state information can query for the last known state for
a particular device, even if it is not part of an active call. The concept of
a device in Asterisk can include entities that do not have a physical
representation. One way that this occurred was when anonymous calls are allowed
in Asterisk. A device was automatically created and stored in the cache for
each anonymous call that occurred; this was possible in the SIP and IAX2
channel drivers and through channel drivers that utilized the
res_jabber/res_xmpp resource modules (Gtalk, Jingle, and Motif). These devices
are never removed from the system, allowing anonymous calls to potentially
exhaust a system's resources.
This patch changes the event cache subsystem and device state management to
no longer cache devices that are not associated with a physical entity.
(issue ASTERISK-20175)
Reported by: Russell Bryant, Leif Madsen, Joshua Colp
Tested by: kmoore
patches:
event-cachability-3.diff uploaded by jcolp (license 5000)
........
Merged revisions 378303 from http://svn.asterisk.org/svn/asterisk/branches/1.8
........
Merged revisions 378320 from http://svn.asterisk.org/svn/asterisk/branches/10
........
Merged revisions 378321 from http://svn.asterisk.org/svn/asterisk/branches/11
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@378322 65c4cc65-6c06-0410-ace0-fbb531ad65f3
........
r375993 | mmichelson | 2012-11-07 11:01:13 -0600 (Wed, 07 Nov 2012) | 30 lines
Fix misuses of timeouts throughout the code.
Prior to this change, a common method for determining if a timeout
was reached was to call a function such as ast_waitfor_n() and inspect
the out parameter that told how many milliseconds were left, then use
that as the input to ast_waitfor_n() on the next go-around.
The problem with this is that in some cases, submillisecond timeouts
can occur, resulting in the out parameter not decreasing any. When this
happens thousands of times, the result is that the timeout takes much
longer than intended to be reached. As an example, I had a situation where
a 3 second timeout took multiple days to finally end since most wakeups
from ast_waitfor_n() were under a millisecond.
This patch seeks to fix this pattern throughout the code. Now we log the
time when an operation began and find the difference in wall clock time
between now and when the event started. This means that sub-millisecond timeouts
now cannot play havoc when trying to determine if something has timed out.
Part of this fix also includes changing the function ast_waitfor() so that it
is possible for it to return less than zero when a negative timeout is given
to it. This makes it actually possible to detect errors in ast_waitfor() when
there is no timeout.
(closes issue ASTERISK-20414)
reported by David M. Lee
Review: https://reviewboard.asterisk.org/r/2135/
........
r375994 | mmichelson | 2012-11-07 11:08:44 -0600 (Wed, 07 Nov 2012) | 3 lines
Remove some debugging that accidentally made it in the last commit.
........
Merged revisions 375993-375994 from http://svn.asterisk.org/svn/asterisk/branches/1.8
........
Merged revisions 375995 from http://svn.asterisk.org/svn/asterisk/branches/10
........
Merged revisions 376014 from http://svn.asterisk.org/svn/asterisk/branches/11
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@376015 65c4cc65-6c06-0410-ace0-fbb531ad65f3
Update and extend the configuration_file group and enable linking. Update title that was left behind many years ago.
(issue ASTERISK-20259)
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@375006 65c4cc65-6c06-0410-ace0-fbb531ad65f3
* The following dialplan applications now recognize 'W' to pause sending
DTMF for one second in addition to the previously existing 'w' that paused
sending DTMF for half a second. Dial, ExternalIVR, and SendDTMF.
* The chan_dahdi analog port dialing and deferred DTMF dialing for PRI now
distinguishes between 'w' and 'W'. The 'w' pauses dialing for half a
second. The 'W' pauses dialing for one second.
* Created dahdi_dial_str() in chan_dahdi that eliminated a lot of
duplicated dialing code and diagnostic messages for the channel driver.
(closes issue ASTERISK-20039)
Reported by: Jeremiah Gowdy
Patches:
jgowdy-wait-6-22-2012.diff (license #5621) patch uploaded by Jeremiah Gowdy
Expanded patch to add support in chan_dahdi.
Tested by: rmudgett
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@374030 65c4cc65-6c06-0410-ace0-fbb531ad65f3
This adds HANGUPCAUSE information to called channels so that hangup
handlers can, in conjunction with predial dialplan execution, access
the hangupcause information when the dialed channel hangs up on a
one-to-one basis instead of a many-to-one basis as with HANGUPCAUSE
usage on the caller channel.
Review: https://reviewboard.asterisk.org/r/2069/
(closes issue ASTERISK-20198)
........
Merged revisions 371258 from http://svn.asterisk.org/svn/asterisk/branches/11
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@371259 65c4cc65-6c06-0410-ace0-fbb531ad65f3
