gecko/media/mtransport
Neil Rashbrook 921625ab05 Bug 514280 Stop using nsCOMPtr for concrete types r=bsmedberg
--HG--
extra : rebase_source : f8fb50de9292320eb7589dd28dc566d0f5044da6
2014-03-18 00:23:03 +00:00
..
build
standalone
test Bug 973395 - Move the LOCAL_INCLUDES in media to moz.build; r=glandium 2014-02-18 01:15:48 -05:00
third_party Bug 966564 - Make logging more clear when processing an unwrapped indication. r=drno 2014-03-12 10:21:48 -07:00
databuffer.h
dtlsidentity.cpp Bug 946348 - Refactor fingerprint handling. r=abr 2014-01-02 10:50:14 -05:00
dtlsidentity.h Bug 946348 - Refactor fingerprint handling. r=abr 2014-01-02 10:50:14 -05:00
gonk_addrs.cpp Bug 969469: use SyncRunnable to get local interface information on gonk. r=bwc 2014-02-18 14:44:45 +08:00
logging.h
m_cpp_utils.h
nr_socket_prsock.cpp Bug 980810 - Part 3: Fix clang -Wstring-conversion warnings in UDPSocket and mtransport code. r=bwc 2014-03-07 22:57:15 -08:00
nr_socket_prsock.h
nr_timer.cpp
nricectx.cpp Bug 970569 - Fix crasher introduced by bug 928930. r=bwc 2014-02-10 14:45:42 -08:00
nricectx.h Bug 970569 - Fix crasher introduced by bug 928930. r=bwc 2014-02-10 14:45:42 -08:00
nricemediastream.cpp Bug 906990: Part 13. Get local/remote candidates separately, instead of grabbing them from candidate pairs (means we can get candidates before pairing happens). r=ekr 2013-11-13 13:49:33 -08:00
nricemediastream.h Bug 906990: Part 13. Get local/remote candidates separately, instead of grabbing them from candidate pairs (means we can get candidates before pairing happens). r=ekr 2013-11-13 13:49:33 -08:00
nriceresolver.cpp Bug 514280 Stop using nsCOMPtr for concrete types r=bsmedberg 2014-03-18 00:23:03 +00:00
nriceresolver.h
nriceresolverfake.cpp
nriceresolverfake.h
nrinterfaceprioritizer.cpp
nrinterfaceprioritizer.h
objs.mozbuild
README
rlogringbuffer.cpp
rlogringbuffer.h
runnable_utils_generated.h
runnable_utils.h
runnable_utils.py
sigslot.h
simpletokenbucket.cpp
simpletokenbucket.h
stun_udp_socket_filter.cpp
stun_udp_socket_filter.h
transportflow.cpp Bug 967364: Pass already_AddRefed by reference instead of by value. r=bsmedberg 2014-03-15 12:00:17 -07:00
transportflow.h Bug 967364: Pass already_AddRefed by reference instead of by value. r=bsmedberg 2014-03-15 12:00:17 -07:00
transportlayer.cpp Bug 930481: Improved failure instrumentation for transport layers r=ekr 2014-01-23 14:21:24 -06:00
transportlayer.h Bug 930481: Improved failure instrumentation for transport layers r=ekr 2014-01-23 14:21:24 -06:00
transportlayerdtls.cpp Bug 930481: Improved failure instrumentation for transport layers r=ekr 2014-01-23 14:21:24 -06:00
transportlayerdtls.h
transportlayerice.cpp Bug 930481: Improved failure instrumentation for transport layers r=ekr 2014-01-23 14:21:24 -06:00
transportlayerice.h
transportlayerlog.cpp Bug 930481: Improved failure instrumentation for transport layers r=ekr 2014-01-23 14:21:24 -06:00
transportlayerlog.h
transportlayerloopback.cpp Bug 930481: Improved failure instrumentation for transport layers r=ekr 2014-01-23 14:21:24 -06:00
transportlayerloopback.h
transportlayerprsock.cpp Bug 930481: Improved failure instrumentation for transport layers r=ekr 2014-01-23 14:21:24 -06:00
transportlayerprsock.h Bug 514280 Stop using nsCOMPtr for concrete types r=bsmedberg 2014-03-18 00:23:03 +00:00

This is a generic media transport system for WebRTC.

The basic model is that you have a TransportFlow which contains a
series of TransportLayers, each of which gets an opportunity to
manipulate data up and down the stack (think SysV STREAMS or a
standard networking stack). You can also address individual
sublayers to manipulate them or to bypass reading and writing
at an upper layer; WebRTC uses this to implement DTLS-SRTP.


DATAFLOW MODEL
Unlike the existing nsSocket I/O system, this is a push rather
than a pull system. Clients of the interface do writes downward
with SendPacket() and receive notification of incoming packets
via callbacks registed via sigslot.h. It is the responsibility
of the bottom layer (or any other layer which needs to reference
external events) to arrange for that somehow; typically by
using nsITimer or the SocketTansportService.

This sort of push model is a much better fit for the demands
of WebRTC, expecially because ICE contexts span multiple
network transports.


THREADING MODEL
There are no thread locks. It is the responsibility of the caller to
arrange that any given TransportLayer/TransportFlow is only
manipulated in one thread at once. One good way to do this is to run
everything on the STS thread. Many of the existing layer implementations
(TransportLayerPrsock, TransportLayerIce, TransportLayerLoopback)
already run on STS so in those cases you must run on STS, though
you can do setup on the main thread and then activate them on the
STS.


EXISTING TRANSPORT LAYERS
The following transport layers are currently implemented:

* DTLS -- a wrapper around NSS's DTLS [RFC 6347] stack
* ICE  -- a wrapper around the nICEr ICE [RFC 5245] stack.
* Prsock -- a wrapper around NSPR sockets
* Loopback -- a loopback IO mechanism
* Logging -- a passthrough that just logs its data

The last three are primarily for debugging.