On Fri, Dec 17, 2010 at 3:59 PM, <s.marek at avm.de> wrote: > No, that doesn't fix it. Okay, PJ_UNUSED_ARG() would probably be optimized out by the compiler (and that's the objective of PJ_UNUSED_ARG in the first place actually). > What does, without being too much of a hack, is > dynamically allocating the memory for tmp_hdr, instead of putting it on > the stack. > It's still too "heavy". As you know we don't use malloc, and changing it to use dlg->pool would add to the memory usage of the dialog. Can we fool gcc with something like this: pjsip_generic_array_hdr tmp_hdr_buf; pjsip_generic_array_hdr *tmp_hdr = &tmp_hdr_buf; > Still, seems unnecessary, because that tmp_hdr on the stack looks pretty > ok to me, from an application programmers point of view. Maybe one should > ask the GCC folks, what they think about it? > You can try, but I'unless you can simplify the problem into a simple compilable program, I would imagine nobody will be interested in looking into that. Besides, even if this is fixed in newer gcc version, it will still take months before the fix gets to Apple, and even then somebody would still face the problem until they upgrade the compiler. > I would still be interested, if that is related to the Apple GCC version, > or if someone who is building PJSIP on Linux/Unix is able to reproduce > this issue? > It's fine here. gcc version 4.4.3 (Ubuntu 4.4.3-4ubuntu5), x86_64, -Os, made/received call with pjsua. But I've kind of guessed that. >From past experience, this usually is version/platform specific thing. Here are two more weird gcc issues in the past: https://trac.pjsip.org/repos/ticket/766 https://trac.pjsip.org/repos/ticket/683 Thanks Benny > Sebastian. > >
