PySide Bugzilla Closed for New Bugs

PySide is now a Qt Add-on and uses the Qt Project's JIRA Bug Tracker instead of this Bugzilla instance. This Bugzilla is left for reference purposes.

Bug 381 - apiextractor segfaults when building on MeeGo (Gcc-4.5.0-3.8)
: apiextractor segfaults when building on MeeGo (Gcc-4.5.0-3.8)
Status: CLOSED FIXED
Product: PySide
Classification: Unclassified
Component: ApiExtractor
: 0.4.0
: All All
: P3 normal
Assigned To: renato filho
:
:
:
  Show dependency treegraph
 
Reported: 2010-09-23 09:21 EEST by Benoît HERVIER
Modified: 2011-05-26 17:03 EEST (History)
8 users (show)

See Also:


Attachments
Raw log of target meego_current (14.98 KB, text/plain)
2010-09-23 09:22 EEST, Benoît HERVIER
Details
Rawlog meego 1.0 (31.02 KB, text/plain)
2010-09-23 09:23 EEST, Benoît HERVIER
Details

Note You need to log in before you can comment on or make changes to this bug.
Description Benoît HERVIER 2010-09-23 09:21:26 EEST
GCC segfault on
/home/abuild/rpmbuild/BUILD/apiextractor-0.8.0/abstractmetalang.cpp:150:6:
internal compiler error: Segmentation fault

This didn't happen with installing libgcc-4.4.2-12.13.

So building apiextractor is working in the meego community obs only for target
meego 1.0 and not meego current.
Comment 1 Benoît HERVIER 2010-09-23 09:22:02 EEST
Created attachment 96 [details]
Raw log of target meego_current
Comment 2 Benoît HERVIER 2010-09-23 09:23:56 EEST
Created attachment 97 [details]
Rawlog meego 1.0
Comment 3 Hugo Parente Lima 2010-09-23 14:05:45 EEST
Maybe a meego problem, a GCC bug caused by patches applied by meego or
something else, I use GCC 4.5.1 without any additional patches on my desktop
and everything works.

A compiler would never segfault due to a bad code being compiled, the same
applies to python bindings and python code being executed, hehe, so I guess
this is a GCC or meego bug, not a PySide one.
Comment 4 Matti Airas 2010-09-23 14:54:55 EEST
Hugo, I agree, it might be a MeeGo problem. However, these are the exact same
issues I've faced a few weeks ago on the Nokia internal OBS, and since we need
to begin providing MeeGo packages pretty soon, it's still a valid one.
Comment 5 Hugo Parente Lima 2010-09-23 15:22:15 EEST
Yes, I agree, this must be fixed, but I don't know what we can do to avoid GCC
from segfaulting, so I guess is better to report this bug on meego.
Comment 6 Matti Airas 2010-09-23 15:34:15 EEST
(In reply to comment #5)
> Yes, I agree, this must be fixed, but I don't know what we can do to avoid GCC
> from segfaulting, so I guess is better to report this bug on meego.

True, agree with that. However, since I don't expect rapid fixes to toolchain
issues, we might still need to try to find a workaround. Best first try to
extract a testcase and report to meego.
Comment 7 Benoît HERVIER 2010-09-23 15:40:02 EEST
It s clearly not a PySide bug. But will be better if a work arround can be
found as i think a change in the toolchain will be really long to come.
Comment 8 renato filho 2010-09-23 16:02:32 EEST
I will try fix this ASAP, and work to get all meego related bugs fixed. I will
try this inside of meego chroot SDK. Any advices?
Comment 9 Matti Airas 2010-09-27 05:08:25 EEST
I can reproduce the bug when building from source on a MeeGo netbook install.
If you've got a suitable netbook (probably notebooks with Intel graphics chips
would work fine) lying around, you could install a MeeGo netbook UX on it and
try to compile it yourself.
Comment 10 renato filho 2010-09-27 12:20:36 EEST
You can use -DENABLE_GCC_OPTIMIZATION=On on cmake flags, to avoid this bug for
now, until the gcc get fixed.
Comment 11 Benoît HERVIER 2010-09-27 12:44:50 EEST
I confirm that it s build with the flag set on the community Meego OBS.
Comment 12 Matti Airas 2010-10-04 08:42:37 EEST
File an upstream bug against the issue:

http://bugs.meego.com/show_bug.cgi?id=7819

Changing bug status to WAITING.
Comment 13 renato filho 2011-05-05 22:02:42 EEST
fixed on meego 1.2 target.
Comment 14 renato filho 2011-05-26 17:03:50 EEST
PySide release 1.0.3