[asterisk-dev] [Code Review] 3800: media formats: chan_iax2 - Obey preferred codec order when selecting format for calls
Jonathan Rose
reviewboard at asterisk.org
Mon Jul 21 11:26:41 CDT 2014
-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/3800/
-----------------------------------------------------------
(Updated July 21, 2014, 11:26 a.m.)
Status
------
This change has been marked as submitted.
Review request for Asterisk Developers, Joshua Colp, Matt Jordan, and opticron.
Changes
-------
Committed in revision 419109
Bugs: ASTERISK-23958
https://issues.asterisk.org/jira/browse/ASTERISK-23958
Repository: Asterisk
Description
-------
At some point in the media_formats work, the function edited in this patch was modified in a way that left it abandoning the preference order of the codecs and just choosing from the first value in the codecs bitfield. This patch changes that function to evaluate using the preferred codecs order.
Diffs
-----
/team/group/media_formats-reviewed-trunk/channels/chan_iax2.c 418711
Diff: https://reviewboard.asterisk.org/r/3800/diff/
Testing
-------
[lappy]
type=friend
...
disallow = all
allow = ulaw
allow = gsm
allow = alaw
context = default
[deskbox]
type=friend
...
disallow=all
allow=gsm
allow=alaw
context = default
Using these two peers, I had each call the other.
lappy -> deskbox
Asterisk 11: Codec chosen for lappy was ulaw. Codec chosen for deskbox was alaw, presumably because of favorable transcoding costs.
without patch: Codec chosen for both was gsm in spite of preferred order for lappy favoring ulaw
with patch: Codec chosen for lappy was ulaw. Codec chosen for deskbox was alaw, again presumably because of favorable transcoding costs.
deskbox -> lappy
It was GSM for both endpoints in all cases because deskbox favors gsm and lappy supports it.
Thanks,
Jonathan Rose
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.digium.com/pipermail/asterisk-dev/attachments/20140721/4232762d/attachment.html>
More information about the asterisk-dev
mailing list