Summary:ASTERISK-07803: [patch] Update the maximum packetization values in frame.c
Reporter:dea (dea)Labels:
Date Opened:2006-09-22 11:57:37Date Closed:2006-09-26 13:10:26
Versions:11.20.0 13.6.0 Frequency of
Environment:Attachments:( 0) iLBC_Packetization_for_Asterisk_11.patch
( 1) iLBC_Packetization_for_Asterisk_13.patch
( 2) packetization-update.diff
( 3) packetization-update-v2.diff
Description:A discussion on -dev pointed out the the maximum framing values for
many codecs was set too low.  This patch increases the limits, while
attempting to maintain headroom, for cases such as RTP over VPN or
other encapsulation.

VBR codecs iLBC and SpeeX are still using fairly low limits, as I do not
have any way to test these codecs.  Testing and feedback about any codec
is desired, but Speex and iLBC need the most love.


This is a trivial patch, so it should apply to both branches/1.5 and trunk
Comments:By: Serge Vecher (serge-v) 2006-09-22 12:24:25

btw, branches/1.5 == trunk. You probably meant to say branches/1.4

By: dea (dea) 2006-09-22 12:50:09

Fat fingers and not paying attention.  You are absolutely correct....

By: Jason Parker (jparker) 2006-09-22 14:24:54

Could you also update the rtp-packetization.txt documentation you did?

By: dea (dea) 2006-09-22 16:47:44

Sure thing.  V2 attached with the original patch plus documentation update.
Let me know if it needs to be split out.


By: Jason Parker (jparker) 2006-09-26 13:10:26

Committed to 1.4 in revision 43674, and trunk in revision 43675

By: Alexander Traud (traud) 2015-10-21 08:06:55.918-0500

??iLBC is still using a fairly low limit, as I do not have any way to test this codec.??

Here, this issue was fixed only partly with SVN [revision 43675|http://svnview.digium.com/svn/asterisk?view=revision&revision=43675]. It was never fixed for iLBC 30. The issue was introduced with SVN [revision 43243|http://svnview.digium.com/svn/asterisk?view=revision&revision=43243] and [Mantis-5162|http://bugs.digium.com/view.php?id=5162]. The mentioned asterisk-dev discussion can be found [here|http://lists.digium.com/pipermail/asterisk-dev/2006-September/023371.html]. Although, the discussion started in [Mantis-7978|http://bugs.digium.com/view.php?id=7978] actually.

9 years later, I was able to test higher packetization times (SIP/SDP ptime) with iLBC 30 in chan_sip. Please, see that attached patches.