Commit Graph

613 Commits

Author SHA1 Message Date
Anthony Minessale 57a2d59ed6 revert this, probably not needed 2012-12-02 09:58:33 -06:00
Anthony Minessale 444f7924e5 not quite, one more tweak 2012-11-30 14:05:45 -06:00
Anthony Minessale 4e3bb1a1e6 rtp and signal parsing adjustments 2012-11-30 13:54:47 -06:00
Anthony Minessale 6e3c08f590 disable catchup code when jitterbuffer is there 2012-11-30 11:43:48 -06:00
Anthony Minessale 1073f70b6f spread out timer skips to smooth out socket catch-up 2012-11-29 13:43:01 -06:00
Anthony Minessale d4f8a79299 try to improve audio stream sync 2012-11-28 23:12:35 -06:00
Brian West 89f4cd75f0 Fix 2833 Debuging 2012-11-21 13:59:23 -06:00
Anthony Minessale 6c1e81a813 FS-4797 --resolve 2012-11-06 08:53:00 -06:00
Anthony Minessale f1bcc387a3 put back seq number when pulling from jb 2012-10-30 16:05:44 -05:00
Anthony Minessale 986fede6b0 make a way to get the jitter buffer down to the codecs 2012-10-18 15:29:00 -04:00
Anthony Minessale a20aec182a add rtp_manual_video_rtp_bugs 2012-10-16 18:43:00 -04:00
Anthony Minessale 4dd1591945 FS-4684 this should help the proxy situation 2012-10-05 11:49:29 -05:00
Anthony Minessale 80c2bd786a FS-4669 --resolve 2012-10-02 10:31:32 -05:00
Anthony Minessale 8d61404016 fix issue where dtmf was not being sent on an outbound leg when there was no inbound audio 2012-09-28 16:17:20 -05:00
Anthony Minessale 5259814aee number not true 2012-09-28 14:06:04 -05:00
Anthony Minessale 4bb5c93e4a use a lower level write func for cng packets 2012-09-26 12:10:37 -05:00
Anthony Minessale 679921d914 FS-4548 --resolve such a little patch for a big problem, good find 2012-08-20 12:11:17 -05:00
Stefan Knoblich 8bcf3b4fb8 FreeSWITCH: Add switch_strerror_r() to fix problems with XSI and GNU variants of strerror_r().
GNU variant of strerror_r() returns char *, while the XSI version returns int.

To make things worse, glibc ships both and added a unused result warning
in recent versions (2.16) causing the build to fail.

Add our own custom wrapper that always returns a pointer to the message buffer
and additionally make XSI versions of strerror_r() GNU compatible by
returning "Unknown error xxx" if no error message is available.

Fixes:
    src/switch_rtp.c: In function 'rtp_common_read':
    src/switch_rtp.c:3313:15: error: ignoring return value of 'strerror_r',
    declared with attribute warn_unused_result [-Werror=unused-result]
    cc1: all warnings being treated as errors

Signed-off-by: Stefan Knoblich <stkn@openisdn.net>
2012-08-14 14:11:44 +02:00
Stefan Knoblich 3b5b59e693 Revert "FreeSWITCH: Fix warn-unused-result error in rtp_common_read()."
*sigh* of course there are two versions of strerror_r():

	char *strerror_r() - GNU
	int strerror_r() - non-GNU

Revert and back the drawing board...

This reverts commit 3c82033e53.
2012-07-31 01:58:48 +02:00
Stefan Knoblich 3c82033e53 FreeSWITCH: Fix warn-unused-result error in rtp_common_read().
strerror_r() returns the error message string (and expects you to use it).

Signed-off-by: Stefan Knoblich <stkn@openisdn.net>
2012-07-30 13:56:48 +02:00
Anthony Minessale 524468be7b FS-4373 --resolve this is probably the same issue nobody would file a bug about on the mailing list even after I begged them to from last weekend 2012-07-23 11:36:19 -05:00
Anthony Minessale db0cca3bf4 FS-4437 try this revision 2012-07-19 12:40:34 -05:00
Anthony Minessale f4e1fc4bc6 FS-4382 --resolve missing \! 2012-07-19 11:30:00 -05:00
Anthony Minessale f4715781dc FS-4382 reverse this so you have to enable instead of disable, this patch has been causing srtp issues because the change in ssrc 2012-07-18 22:01:41 -05:00
Anthony Minessale d63323977f turn on send_silence_when_idle whenever we engage srtp 2012-07-18 12:19:47 -05:00
Anthony Minessale b27aad913e typo 2012-07-17 10:32:31 -05:00
Anthony Minessale 8ad5d8a6b0 FS-4317 this might help, this fixes a regression cause by another recently altered code path and it sounds similar but I never got the new logs from you so this is a guess 2012-07-16 18:47:45 -05:00
Anthony Minessale 06046e3437 FS-4427 try this revision however i think its a problem with the remote client not properly preserving the headers of the srtp packets..... 2012-07-16 12:51:13 -05:00
Anthony Minessale b6e962de1d typo 2012-07-13 18:02:09 -05:00
Anthony Minessale 4a801c8626 only decode rtp packets in srtp mode 2012-07-13 14:58:20 -05:00
Jeff Lenk fc5490f1dc FS-4415 --resolve 2012-07-11 22:31:36 -05:00
Anthony Minessale 1e7d214ae6 compiler errs from last commit 2012-07-11 15:35:49 -05:00
Anthony Minessale 088437cab1 add some rtcp stuff for sangoma 2012-07-11 15:23:12 -05:00
Anthony Minessale 2aebe2456f first pass on some new stuff 2012-07-11 15:15:50 -05:00
Anthony Minessale 9ca929218f FS-4382 --resolve 2012-07-06 11:12:14 -05:00
Anthony Minessale 59bd4b0f14 fix regression with proxy_packet flag 2012-07-03 13:17:38 -05:00
Anthony Minessale 848f05fb41 FS-3659 this fixes sub issue, this should have been a new jira ticket 2012-06-28 14:02:02 -05:00
Travis Cross 7978bd8921 Don't truncate 5-digit port numbers
The adj_port buffer was only 5 bytes wide, so we were truncating any
port numbers >= 10000 that were written to this variable as snprintf
ensures the last byte is null.

FS-4354 --resolve
2012-06-27 20:17:08 +00:00
Jeff Lenk a97ec71902 fix warnings on windows code verification 2012-06-25 10:06:41 -05:00
Travis Cross 453b42e301 Remove extraneous parenthesis
This causes a warning that breaks the build with clang.
2012-06-24 07:36:14 +00:00
Travis Cross ad53499b42
Always set the ZRTP SAS verified flag on our side
When we're terminating ZRTP to an IVR or to a leg which is not
ZRTP-capable, there's no way for us to make an intelligent decision
about whether this flag should be set.  For a client to consider the
SAS as verified, however, both sides need to set this flag.  By always
setting this flag on our side, we leave the decision completely in the
hands of the client where we hope there is a careful human.

Signed-off-by: Travis Cross <tc@traviscross.com>
2012-06-23 22:34:06 +00:00
Anthony Minessale a5021fa384 FS-4337 --resolve this actually uncovered a bug cos it was supposed to be rtcp_bytes not bytes 2012-06-23 14:03:06 -05:00
Anthony Minessale 8b4c9dc42c FS-3659 and some refactor for rtcp ice and video ice and rtcp 2012-06-22 18:15:06 -05:00
Anthony Minessale 8bb55ed4bf abstract out originate_signal_bond to a function to avoid confustion and regressions 2012-05-29 13:10:15 -05:00
Travis Cross f37b1f0c54 squelch another source of noise ZRTP protection drops 2012-05-24 03:59:09 +00:00
Travis Cross d470ae239c don't log ZRTP protection drops
At least until we can distinguish the meaningful ones from the noise.
2012-05-24 03:52:51 +00:00
Travis Cross 236b927780 lower log level on ZRTP protection drop
When a call is being setup, a couple seconds of audio may be sent
unencrypted.  This seems to trigger the log message here with every
such packet.
2012-05-24 03:50:42 +00:00
Travis Cross 0c1a95f745 fix gcc 4.7.0 warning related to enum type
This amends commit 7bd9efc7f5.

We changed away from using unsigned here because it caused a warning
on Windows.  Using the more specific type, however, is causing a
warning on gcc-4.7 (it notices that the switch statement contains
values not present in the specified enum type).  So we're switching
back to unsigned here while keeping the type casts and hoping that
keeps Windows happy.
2012-05-24 00:48:59 +00:00
Travis Cross 83e090c30d add some casts to make things work with both gcc and VS
This amends commit 7bd9efc7f5.

Using the more explicit typedef causes gcc to warn about the
incompatible types for the reasons noted in the earlier commit
message.  This is why we just used the unsigned type earlier so we
could avoid all this casting.

The things we do to make Windows happy...
2012-05-17 16:40:30 +00:00
Jeff Lenk b65c321193 FS-4228 --resolve add zrtp support to windows vs2010 build 2012-05-17 08:38:11 -05:00