|
From: | Riccardo Mottola |
Subject: | Re: problems compiling on solaris 10 |
Date: | Mon, 15 Apr 2013 23:02:45 +0200 |
User-agent: | Mozilla/5.0 (Windows NT 5.1; rv:20.0) Gecko/20100101 Firefox/20.0 SeaMonkey/2.17 |
Hi, Sebastian Reitenbach wrote:
On Monday, April 15, 2013 19:43 CEST, Riccardo Mottola <riccardo.mottola@libero.it> wrote:You are right, it is the test program. config_thread.m As I wrote, I am able to reproduce the problem outside the configure environment and played also with the passed compilation options. The -DGNU_RUNTIME causes all the warnings and problems, but why precisely I don't know. I noticed the warnings happen also on other platforms, but not the seg-fault.Hi! does really nobody have a hint on how to continue on this? The test is minimal and the stacktrace is pretty low-level.does gcc segfault, while you are compiling, or does the program segfault, when you run it? From the backtrace I would guess the latter, but I may be wrong.
I guess that is where the building of the binary package happened. I could try playing with some links and get hold of the source + potential patches.In any case, I think, the libobjc you are using has debugging symbols in it, therefore it knows the location where it was compiled on, under home/maciej/... When you put the sources there, then the list command in gdb should show you something more useful.
R
[Prev in Thread] | Current Thread | [Next in Thread] |