[asterisk-dev] [Code Review] 3296: res_fax_spandsp: Resolve random segfaults in libspandsp when T.38 gateway is enabled

Michal Rybarik reviewboard at asterisk.org
Tue Mar 4 16:28:55 CST 2014



> On March 4, 2014, 10:15 p.m., Matt Jordan wrote:
> > Hm... looks like whatever did you made doesn't apply cleanly. You may want to make sure there aren't any conflicts or other wacky things happening.

Diff rev.2 which I uploaded, was against current revision 409629, not against 407152 as the first diff was. Probably it was not good idea to change revision during reviewing process. Diff rev.3 is against 407152 and it seems better, hopefully it'll be OK now.


- Michal


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/3296/#review11050
-----------------------------------------------------------


On March 4, 2014, 9:24 p.m., Michal Rybarik wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviewboard.asterisk.org/r/3296/
> -----------------------------------------------------------
> 
> (Updated March 4, 2014, 9:24 p.m.)
> 
> 
> Review request for Asterisk Developers.
> 
> 
> Bugs: ASTERISK-20149
>     https://issues.asterisk.org/jira/browse/ASTERISK-20149
> 
> 
> Repository: Asterisk
> 
> 
> Description
> -------
> 
> T.38 gateway was producing random segfaults in libspandsp (during fax tone detection). Segfaults could happen when alaw/ulaw frames were passed to libspandsp code, because this code expect only slinear frames. Recoding of alaw/ulaw to slinear was added here.
> 
> 
> Diffs
> -----
> 
>   /branches/11/res/res_fax_spandsp.c 409629 
> 
> Diff: https://reviewboard.asterisk.org/r/3296/diff/
> 
> 
> Testing
> -------
> 
> Used in production on 3 hosts (2x SIP/DAHDI, 1x SIP/SIP), no segfault seen after applying patch (month ago). Before patch, there were segfaults every few days.
> 
> 
> Thanks,
> 
> Michal Rybarik
> 
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.digium.com/pipermail/asterisk-dev/attachments/20140304/848f7b43/attachment.html>


More information about the asterisk-dev mailing list