gecko/media/mtransport
2014-08-18 19:14:00 -04:00
..
build Bug 1041864 - Remove LIBRARY_NAMEs that aren't used. r=mshal 2014-07-23 08:39:56 +09:00
standalone
test Bug 1052610 - Removing cipher suite profiling. r=ekr 2014-08-13 14:08:00 -05:00
third_party Bug 1045231 - Namespace nrAppKit's htonll. r=abr 2014-08-14 10:05:00 -07:00
databuffer.h Bug 1028582 - Modify TransportLayerDtls to force a fresh ECDHE server key. r=mt 2014-08-05 09:06:48 -07:00
dtlsidentity.cpp
dtlsidentity.h Bug 1027251 - Address review comments on mtransport - r=abr 2014-06-21 00:09:48 -04:00
gonk_addrs.cpp
logging.h
m_cpp_utils.h
nr_socket_prsock.cpp Bug 1054908 - Ensure socket_child_ is released on main thread. r=mt 2014-08-18 19:14:00 -04:00
nr_socket_prsock.h Bug 1054908 - Ensure socket_child_ is released on main thread. r=mt 2014-08-18 19:14:00 -04:00
nr_timer.cpp
nricectx.cpp Bug 1037100 - Remove all use of ScopedDeleteArray from media/. r=ekr 2014-07-10 20:03:12 -07:00
nricectx.h Bug 1053792 - Fix some bad implicit constructors in the mtransport code; r=jesup 2014-08-15 17:41:29 -04:00
nricemediastream.cpp
nricemediastream.h Bug 1027251 - Fix or whitelist dangerous public destructors in media/mtransport - r=abr 2014-06-20 07:08:24 -04:00
nriceresolver.cpp
nriceresolver.h Bug 1028588 - Fix dangerous public destructors in media/mtransport - r=abr 2014-06-24 12:36:44 -04:00
nriceresolverfake.cpp
nriceresolverfake.h
nrinterfaceprioritizer.cpp
nrinterfaceprioritizer.h
objs.mozbuild
README
rlogringbuffer.cpp
rlogringbuffer.h
runnable_utils_generated.h Bug 1053792 - Fix some bad implicit constructors in the mtransport code; r=jesup 2014-08-15 17:41:29 -04:00
runnable_utils.h
runnable_utils.py Bug 1053792 - Fix some bad implicit constructors in the mtransport code; r=jesup 2014-08-15 17:41:29 -04:00
sigslot.h Bug 1053792 - Fix some bad implicit constructors in the mtransport code; r=jesup 2014-08-15 17:41:29 -04:00
simpletokenbucket.cpp
simpletokenbucket.h
stun_udp_socket_filter.cpp Bug 1053792 - Fix some bad implicit constructors in the mtransport code; r=jesup 2014-08-15 17:41:29 -04:00
stun_udp_socket_filter.h Bug 1028588 - Fix dangerous public destructors in media/mtransport - r=abr 2014-06-24 12:36:44 -04:00
transportflow.cpp
transportflow.h Bug 1053792 - Fix some bad implicit constructors in the mtransport code; r=jesup 2014-08-15 17:41:29 -04:00
transportlayer.cpp
transportlayer.h Bug 1006207 - Remove unused mode_ code. r=ekr 2014-08-04 15:34:00 -04:00
transportlayerdtls.cpp Bug 1052610 - Fix an unused variable warning. r=martin.thomson 2014-08-14 10:02:00 -07:00
transportlayerdtls.h Bug 1053792 - Fix some bad implicit constructors in the mtransport code; r=jesup 2014-08-15 17:41:29 -04:00
transportlayerice.cpp
transportlayerice.h
transportlayerlog.cpp
transportlayerlog.h
transportlayerloopback.cpp
transportlayerloopback.h Bug 1053792 - Fix some bad implicit constructors in the mtransport code; r=jesup 2014-08-15 17:41:29 -04:00
transportlayerprsock.cpp
transportlayerprsock.h Bug 1028588 - Fix dangerous public destructors in media/mtransport - r=abr 2014-06-24 12:36:44 -04:00

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.