Rob Landley wrote:
> On Friday 17 October 2008 05:29:36 Bernhard Reutner-Fischer wrote:
>> On Fri, Oct 17, 2008 at 06:15:58PM +0800, Ming-Ching Tiew wrote:
>>> I have been using asterisk ( eg 1.4.20.1 ) with uclibc-0.9.28 all
>>> the while and it has been working.
>>>
>>> But recently I feel that there is a need for me to check out
>>> uclibc-0.9.29. But after I compile asterisk against uclibc-0.9.29,
>>> if
>>
>> I suggest you try 0.9.30-rc2 plus one fix, i.e.
>> $ svn co svn://uclibc.org/trunk/uClibc
>
> More easily downloaded from the nightly snapshots directory, either:
> http://uclibc.org/downloads/snapshots/uClibc-20081017.tar.bz2
>
> Or just use the link that always goes to the most recent tarball:
> http://uclibc.org/downloads/snapshots/uClibc-snapshot.tar.bz2
>
> The final 0.9.30 release is scheduled for the end of the month, so
> now's a good time to make sure it works for you...
>
> Rob
I tested it against uclibc 0.9.30 too and I had the same problem.
After much debugging ( there goes my weekend ! ) as it turns
out the problem I observed has nothing to do with 'thread'.
I am not saying the 'thread' has no problem. I am yet to come
to that part.
The problem I observerd is confirmed due to daemon() function call.
The daemon is a little function in the uclibc, if I don't do anything,
after fork(), the parent will exit and it will cause the child to
exit too. If I put a small sleep() in the parent before it calls _exit(),
then the child could survive even after the parent exit.
Regards.
_______________________________________________
uClibc mailing list
[email protected]
http://busybox.net/cgi-bin/mailman/listinfo/uclibc
