[Bug bootstrap/43900] ICE in dbxout.c
--- Comment #8 from ktietz at gcc dot gnu dot org 2010-08-21 09:26 --- Ok, let us close it. -- ktietz at gcc dot gnu dot org changed: What|Removed |Added Status|UNCONFIRMED |RESOLVED Resolution||WORKSFORME http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43900
[Bug bootstrap/43900] ICE in dbxout.c
--- Comment #7 from rainer at emrich-ebersheim dot de 2010-08-21 09:08 --- (In reply to comment #6) > (In reply to comment #5) > > (In reply to comment #4) > > > (In reply to comment #2) > > > > As it turns out, the ICE only manifests in a parallel build. I tried > > > > make -j 8, > > > > my default and make -j 3. > > > > For an ordinary make there is no issue. So, I'm curious how to handle > > > > this. > > > > Any thoughts? > > > > > > > > Rainer > > > > > > > > > > Rainer, > > > > > > you are not accidentially are using rxvt terminal? > > > > > > Kai > > > > > > > I did, but I haven't had any issues in the past. So , I try without rxvt. > > > > Rainer > > > > Have you made your tests here without rxvt? I tested it in cygwin standard cmd > shell for -j3, and I can bootstrap here without flaw. > I haven't had much time for gcc lately, sorry. I think this issue was caused by my MSYS setup. So, closing as WORKSFORME seems to be the right thing. Rainer -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43900
[Bug bootstrap/43900] ICE in dbxout.c
--- Comment #6 from ktietz at gcc dot gnu dot org 2010-08-21 08:19 --- (In reply to comment #5) > (In reply to comment #4) > > (In reply to comment #2) > > > As it turns out, the ICE only manifests in a parallel build. I tried make > > > -j 8, > > > my default and make -j 3. > > > For an ordinary make there is no issue. So, I'm curious how to handle > > > this. > > > Any thoughts? > > > > > > Rainer > > > > > > > Rainer, > > > > you are not accidentially are using rxvt terminal? > > > > Kai > > > > I did, but I haven't had any issues in the past. So , I try without rxvt. > > Rainer > Have you made your tests here without rxvt? I tested it in cygwin standard cmd shell for -j3, and I can bootstrap here without flaw. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43900
[Bug bootstrap/43900] ICE in dbxout.c
--- Comment #5 from rainer at emrich-ebersheim dot de 2010-04-27 13:15 --- (In reply to comment #4) > (In reply to comment #2) > > As it turns out, the ICE only manifests in a parallel build. I tried make > > -j 8, > > my default and make -j 3. > > For an ordinary make there is no issue. So, I'm curious how to handle this. > > Any thoughts? > > > > Rainer > > > > Rainer, > > you are not accidentially are using rxvt terminal? > > Kai > I did, but I haven't had any issues in the past. So , I try without rxvt. Rainer -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43900
[Bug bootstrap/43900] ICE in dbxout.c
--- Comment #4 from ktietz at gcc dot gnu dot org 2010-04-27 07:52 --- (In reply to comment #2) > As it turns out, the ICE only manifests in a parallel build. I tried make -j > 8, > my default and make -j 3. > For an ordinary make there is no issue. So, I'm curious how to handle this. > Any thoughts? > > Rainer > Rainer, you are not accidentially are using rxvt terminal? Kai -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43900
[Bug bootstrap/43900] ICE in dbxout.c
--- Comment #3 from jakub at gcc dot gnu dot org 2010-04-27 06:16 --- It depends on what the platform allows, I'm not familiar with it at all. If you can force dumping core and getting backtrace from it, that would be interesting info, if you can preload some library to print backtrace upon crash similar to LD_PRELOAD=libSegFault.so or something similar to abrt to get the backtrace, that would be useful too. Or at least binary searching which 4.4 branch change caused it. Getting a backtrace would be at least a hint where to look for. Not sure if this could be related to address space randomization etc. -- jakub at gcc dot gnu dot org changed: What|Removed |Added CC||ktietz at gcc dot gnu dot ||org, dannysmith at gcc dot ||gnu dot org http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43900
[Bug bootstrap/43900] ICE in dbxout.c
--- Comment #2 from rainer at emrich-ebersheim dot de 2010-04-26 19:53 --- As it turns out, the ICE only manifests in a parallel build. I tried make -j 8, my default and make -j 3. For an ordinary make there is no issue. So, I'm curious how to handle this. Any thoughts? Rainer -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43900
[Bug bootstrap/43900] ICE in dbxout.c
--- Comment #1 from rainer at emrich-ebersheim dot de 2010-04-26 18:27 --- I'm still analyzing, what's going wrong here, it's a strange thing. I tried the build three times with a parallel make resulting in the above ICE. But can't reproduce it by compiling dbxout.c on the command line. Investigating Rainer -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43900
