Commit Graph

54 Commits (f0329105f35b53967a44b424ceae3d197d77d16d)

Author SHA1 Message Date
Joshua Colp f0329105f3 Correct a copy/pasted error message line for RTCP.
19 years ago
Joshua Colp 28dacda047 When going on hold have the side that was put on hold reinvite back to Asterisk. When going off hold have the side that was taken off hold reinvited back to the other party.
19 years ago
Joshua Colp 7229c2ba48 Add more frame types to forward in the RTP bridge loops.
19 years ago
Russell Bryant 0bc7ef3b06 Merged revisions 53039 via svnmerge from
19 years ago
Russell Bryant 06a3110ed8 Fix a problem with packet-to-packet bridging and DTMF mode translation. P2P
19 years ago
Russell Bryant 33235b40d6 Merge the changes from the /team/group/vldtmf_fixup branch.
19 years ago
Joshua Colp 9461aa0027 Pass data as well for hold/unhold/vidupdate frames. (issue #8840 reported by mdu113)
19 years ago
Joshua Colp 9f0ad6093e Return the correct result when directly writing out a packet so that the core doesn't then decide to handle it the regular way again. (issue #8833 reported by rcourtna)
19 years ago
Jason Parker 1825fc3ee2 Fix issue with dtmf continuation packets when the dtmf digit is 0...
19 years ago
Joshua Colp 240ca25bea Add support to see whether NAT was detected (yay symmetric RTP) and also add a check in chan_sip so that if NAT has been detected and the reinvite behind nat option has been turned off, then just do partial bridge. (issue #8655 reported by mnicholson)
19 years ago
Joshua Colp 46700b62af Disable the more intense packet2packet bridging until the bugs can be worked out.
19 years ago
Joshua Colp 156a6cf5db If the Packet2Packet bridge is being broken because of a masquerade then attempt to read a frame in so the masquerade actually happens. Otherwise weirdness will occur. (issue #8696 reported by kjotte)
19 years ago
Kevin P. Fleming a299485952 since these variables all have static duration, none of them need initializers (they default to zero anyway)
19 years ago
Joshua Colp 9cc04e026d Add an API call that initializes an RTP structure. We need this because chan_sip is cheeky and uses a temporary RTP structure for codec purposes, and the API calls that are used rely on the lock. (Pointed out on asterisk-dev by Andy Wang)
19 years ago
Joshua Colp 8df2c7228f Turn payload_lock into bridge_lock and make it encompass all RTP structure contents that may relate to bridge information, including who we are bridged to.
19 years ago
Joshua Colp 0995fb8aeb Payload values on the RTP structure can change AFTER a bridge has started. This comes from the packet handling of the SIP response when indication that it was answered has been sent. Therefore we need to protect this data with a lock when we read/write. (issue #8232 reported by tgrman)
19 years ago
Joshua Colp 87a03da072 Remove direct RTCP bridging. I've come to the conclusion that we should handle this through the core and not just forward it on. Should solve a few bugs.
19 years ago
Joshua Colp c866263f50 Merge in my latest RTP changes. Break out RTP and RTCP callback functions so they no longer share a common one.
19 years ago
Olle Johansson f89143bd13 - Disable RTP hold timers while T.38 fax transmission happens
19 years ago
Joshua Colp b2b70adede Do not do a partial bridge for Google Talk since we need to handle STUN. (issue #8448 reported by phsultan)
19 years ago
Joshua Colp 558639ff6a Merged revisions 48106 via svnmerge from
19 years ago
Joshua Colp 090c412c27 Video will never reach Packet2Packet bridging and can do more harm then good.
19 years ago
Joshua Colp 8c0f4c8711 If we have the non standard G726-32 setting turned on we want to return G726-32 to the SDP, not our AAL2 string. (issue #8330 reported by voipgate)
19 years ago
Joshua Colp b42279c142 Only remove/destroy the RTCP I/O item if it exists.
19 years ago
Joshua Colp bc6cff37bc If NAT detection is turned on or already detected then say NAT is active when setting the remote RTP peer when doing early bridging. (issue #8365 reported by marcelbarbulescu)
19 years ago
Joshua Colp 708b84da2f Turn notice about unknown RTCP packet type into a debug message instead.
19 years ago
Tilghman Lesher 56a1e60c54 More changes making the CLI more consistent with "category verb arguments" (continuation of issue 8236)
19 years ago
Olle Johansson e93e64a005 Bind RTCP to the same IP as RTP
19 years ago
Kevin P. Fleming 6c17f1e07e add passthrough and file format support for G.722 16KHz audio (issue #5084, original patch by andrew, updated by mithraen)
19 years ago
Joshua Colp 64a875951a Don't segfault if you're using a channel driver that doesn't turn RTCP on
19 years ago
Joshua Colp faf38b71ef Remove the seqno check for RFC2833, the handler is smart enough to not need it.
19 years ago
Joshua Colp 2d3da5b500 When the sequence number rolls over then reset the recorded sequence number for DTMF (issue #8106 reported by bungalow)
19 years ago
Paul Cadach b0bb62a7e5 Allow one-way RTP streams (device->Asterisk)
19 years ago
Joshua Colp f149686e8b Compensate for out of order packets better if RFC2833 compensation is turned on.
19 years ago
Tilghman Lesher 70af28270d Constify the result of a config retrieval function, to avoid mutilation (issue 7983).
19 years ago
Joshua Colp 3c6d5053ba Totally break a P2P bridge upon going on hold, and re-establish it upon going off hold.
19 years ago
Joshua Colp 659d467720 Expand codec check so that raw formats must be equal for a Packet2Packet bridge to occur
19 years ago
Matt O'Gorman 465adf2bf1 allow for packetization on rtp channel drivers, need to add
19 years ago
Kevin P. Fleming fcb999c01c merge qwell's CLI verbification work
19 years ago
Joshua Colp d2a359e57f Optimize a bit
19 years ago
Joshua Colp 10e361763c It's another round of RTP updates!
19 years ago
Joshua Colp f912c9e69f Unbridge the RTP streams at the correct place
19 years ago
Joshua Colp 0be2884d80 If we are doing video and we can't reinvite, then resort to generic bridging instead of Packet2Packet since video isn't supported there yet. (reported by PCadach in #asterisk-bugs)
19 years ago
Joshua Colp f2b836ff4f Tweak the DTMF muting stuff a bit to take into account VLDTMF and compensation.
19 years ago
Joshua Colp 29ee02bfce Only write a received packet out if we are actually bridged to something
19 years ago
Joshua Colp c6977b9983 Merge in VLDTMF support with Zaptel/Core done by the ever great Darumkilla Russell Bryant and the RTP portion done by myself, Muffinlicious Joshua Colp. This has gone through so many discussions/revisions it's not funny but we finally have it!
19 years ago
Joshua Colp 0855df6a5a Only feed a DTMF frame into the core if we need to
19 years ago
Joshua Colp 245aa1a62d Only switch the second alert fd (which is RTCP) to callback mode if it is in use
19 years ago
Joshua Colp 12b6ec4e11 Use an API call (ast_rtp_get_bridged) to return the RTP stream we are bridged to, and also use it in chan_sip so we know to ignore the no RTP activity checking
19 years ago
Joshua Colp da7d969ae1 If the RTP stack is already being operated in callback mode, then suspend it upon switching to P2P callback bridging. Once P2P callback bridging has ended, then restore callback mode.
19 years ago