| Age | Commit message (Collapse) | Author |
|---|
|
This code comes from svn/asterisk/team/russell/event_performance/.
Here is a summary of the changes that have been made, in order of both
invasiveness and performance impact, from smallest to largest.
1) Asterisk 1.6.1 introduces some additional logic to be able to handle
distributed device state. This functionality comes at a cost.
One relatively minor change in this patch is that the extra processing
required for distributed device state is now completely bypassed if
it's not needed.
2) One of the things that I noticed when profiling this code was that a
_lot_ of time was spent doing string comparisons. I changed the way
strings are represented in an event to include a hash value at the front.
So, before doing a string comparison, we do an integer comparison on the
hash.
3) Finally, the code that handles the event cache has been re-written.
I tried to do this in a such a way that it had minimal impact on the API.
I did have to change one API call, though - ast_event_queue_and_cache().
However, the way it works now is nicer, IMO. Each type of event that
can be cached (MWI, device state) has its own hash table and rules for
hashing and comparing objects. This by far made the biggest impact on
performance.
For additional details regarding this code and how it was tested, please see the
review request.
(closes issue #14738)
Reported by: russell
Review: http://reviewboard.digium.com/r/205/
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@184339 65c4cc65-6c06-0410-ace0-fbb531ad65f3
|
|
------------------------------------------------------------------------
r159808 | kpfleming | 2008-11-29 10:58:29 -0600 (Sat, 29 Nov 2008) | 7 lines
update dev-mode compiler flags to match the ones used by default on Ubuntu Intrepid, so all developers will see the same warnings and errors
since this branch already had some printf format attributes, enable checking for them and tag functions that didn't have them
format attributes in a consistent way
------------------------------------------------------------------------
in addition:
move some format attributes from main/utils.c to the header files they belong in, and fix up references to the relevant functions based on new compiler warnings
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@159818 65c4cc65-6c06-0410-ace0-fbb531ad65f3
|
|
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@137082 65c4cc65-6c06-0410-ace0-fbb531ad65f3
|
|
Asterisk boxes with mismatched ie_maps.
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@121649 65c4cc65-6c06-0410-ace0-fbb531ad65f3
|
|
This commit pulls in a batch of improvements and additions to the event API.
Changes include:
- the ability to dynamically build a subscription. This is useful if you're
building a subscription based on something you receive from the network,
or from options in a configuration file.
- Add tables of event types and IE types and the corresponding string
representation for implementing text based protocols that use these
events, for showing events on the CLI, reading configuration that
references event information, among other things.
- Add a table that maps IE types and the corresponding payload type.
- an API call to get the total size of an event
- an API call to get all events from the cache that match a subscription
- a new IE payload type, raw, which I used for transporting the Entity ID in
my code for handling distributed device state.
- Code improvements to reduce code duplication
- Include the Entity ID of the server that originated the event in every event
- an additional event type, DEVICE_STATE_CHANGE, to help facilitate distributed
device state. DEVICE_STATE is a state change on one server, DEVICE_STATE_CHANGE
is the aggregate device state change across all servers.
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@121555 65c4cc65-6c06-0410-ace0-fbb531ad65f3
|
|
Asterisk core.
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@115324 65c4cc65-6c06-0410-ace0-fbb531ad65f3
|
|
Doubly-Linked list for events, gives fast deletions, for the sake of channel driver mwi events. From team/murf/bug11210.
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@114183 65c4cc65-6c06-0410-ace0-fbb531ad65f3
|
|
allowed for the iterator
to indicate there was more data than there really was, causing the iterator read beyond the end
of the event structure. This led to invalid memory reads and potential crashes.
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@103559 65c4cc65-6c06-0410-ace0-fbb531ad65f3
|
|
in the event API.
ast_event_iterator_get_ie_type should return the ie type in
host byte order. Furthermore, ast_event_get_ie_raw should already
have its ie type argument in host byte order since it could be called
externally (and it in fact is called in this way by ast_event_get_cached).
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@103282 65c4cc65-6c06-0410-ace0-fbb531ad65f3
|
|
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@98435 65c4cc65-6c06-0410-ace0-fbb531ad65f3
|
|
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@98434 65c4cc65-6c06-0410-ace0-fbb531ad65f3
|
|
get_unaligned functions as const
* In event.c, use get_unaligned_uint32() in a couple of places to fix issues on
architectures that don't allow unaligned access
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@92305 65c4cc65-6c06-0410-ace0-fbb531ad65f3
|
|
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@89465 65c4cc65-6c06-0410-ace0-fbb531ad65f3
|
|
build times - tested, there is no measureable difference before and
after this commit.
In this change:
use asterisk/compat.h to include a small set of system headers:
inttypes.h, unistd.h, stddef.h, stddint.h, sys/types.h, stdarg.h,
stdlib.h, alloca.h, stdio.h
Where available, the inclusion is conditional on HAVE_FOO_H as determined
by autoconf.
Normally, source files should not include any of the above system headers,
and instead use either "asterisk.h" or "asterisk/compat.h" which does it
better.
For the time being I have left alone second-level directories
(main/db1-ast, etc.).
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@89333 65c4cc65-6c06-0410-ace0-fbb531ad65f3
|
|
- the *_CURRENT macros no longer need the list head pointer argument
- add AST_LIST_MOVE_CURRENT to encapsulate the remove/add operation when moving entries between lists
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@89106 65c4cc65-6c06-0410-ace0-fbb531ad65f3
|
|
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@81356 65c4cc65-6c06-0410-ace0-fbb531ad65f3
|
|
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@81355 65c4cc65-6c06-0410-ace0-fbb531ad65f3
|
|
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@81354 65c4cc65-6c06-0410-ace0-fbb531ad65f3
|
|
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@81353 65c4cc65-6c06-0410-ace0-fbb531ad65f3
|
|
ast_event_iterator_next.
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@81348 65c4cc65-6c06-0410-ace0-fbb531ad65f3
|
|
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@81345 65c4cc65-6c06-0410-ace0-fbb531ad65f3
|
|
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@81344 65c4cc65-6c06-0410-ace0-fbb531ad65f3
|
|
have multiple information elements (while there was nothing preventing it before you could not actually access any except the first one).
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@81334 65c4cc65-6c06-0410-ace0-fbb531ad65f3
|
|
The way a device state change propagates is kind of silly, in my opinion. A
device state provider calls a function that indicates that the state of a
device has changed. Then, another thread goes back and calls a callback for
the device state provider to find out what the new state is before it can go
send it off to whoever cares.
I have changed it so that you can include the state that the device has changed
to in the first function call from the device state provider. This removes the
need to have to call the callback, which locks up critical containers to go find
out what the state changed to.
This change set changes the "simple" device state providers to use the new method.
This includes parking, meetme, and SLA.
I have also mostly converted chan_agent in my branch, but still have some more
things to think through before presenting the plan for converting channel drivers
to ensure all of the right events get generated ...
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@79027 65c4cc65-6c06-0410-ace0-fbb531ad65f3
|
|
guidelines changes
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@67864 65c4cc65-6c06-0410-ace0-fbb531ad65f3
|
|
When we implement new API's - please include a small general overview in Doxygen
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@64496 65c4cc65-6c06-0410-ace0-fbb531ad65f3
|
|
This set of changes introduces a new generic event API for use within Asterisk.
I am still working on a way for events to be shared between servers, but this
part is ready and can already be used inside of Asterisk.
This set of changes introduces the first use of the API, as well. I have
restructured the way that MWI (message waiting indication) is handled. It is
now event based instead of polling based. For example, if there are a bunch
of SIP phones subscribed to mailboxes, then chan_sip will not have to
constantly poll the mailboxes for changes. app_voicemail will generate events
when changes occur.
See UPGRADE.txt and CHANGES for some more information on the effects of these
changes from the user perspective. For developer information, see the text in
include/asterisk/event.h.
As always, additional feedback is welcome on the asterisk-dev mailing list.
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@62292 65c4cc65-6c06-0410-ace0-fbb531ad65f3
|
