[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Building GNUstep from source - failed tests

From: edwin ancaer
Subject: Re: Building GNUstep from source - failed tests
Date: Fri, 18 Jan 2013 06:34:01 +0100


After a rebuild of lobobjc2, GNUstep make and GNUstep Base, I keep having segfaults in every Objective C program I'm running.

I made a backtrace from a sample program:

address@hidden obj]$ gdb LogTest /tmp/core-e-11-1000-1000-1396-1358482218
GNU gdb (GDB) Fedora (
Copyright (C) 2012 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.  Type "show copying"
and "show warranty" for details.
This GDB was configured as "i686-redhat-linux-gnu".
For bug reporting instructions, please see:
Reading symbols from /home/edwin/FirstObjc/obj/LogTest...done.
[New LWP 1396]
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/libthread_db.so.1".
Core was generated by `./LogTest'.
Program terminated with signal 11, Segmentation fault.
#0  0xb71f171e in objc_msgSend () from /usr/GNUstep/Local/Library/Libraries/libobjc.so.4
Missing separate debuginfos, use: debuginfo-install glibc-2.15-57.fc17.i686 gmp-5.0.2-6.fc17.i686 gnutls-2.12.20-4.fc17.i686 libffi-3.0.10-2.fc17.i686 libgcc-4.7.2-2.fc17.i686 libgcrypt-1.5.0-3.fc17.i686 libgpg-error-1.10-2.fc17.i686 libicu- libstdc++-4.7.2-2.fc17.i686 libtasn1-2.12-1.fc17.i686 libxml2-2.7.8-9.fc17.i686 libxslt-1.1.26-10.fc17.i686 p11-kit-0.12-1.fc17.i686 zlib-1.2.5-7.fc17.i686
(gdb) bt
#0  0xb71f171e in objc_msgSend () from /usr/GNUstep/Local/Library/Libraries/libobjc.so.4
#1  0xb73251f0 in GSIMapAddPair (map=0x8e23f60, key=..., value=...) at ../Headers/GNUstepBase/GSIMap.h:1087
#2  0xb7325013 in -[GSDictionary initWithObjects:forKeys:count:] (self=0x8e23f5c, _cmd=0xb776b2e0, objs=0xbf959610, keys=0xbf959710, c=2)
    at GSDictionary.m:220
#3  0xb7418b6c in +[NSDictionary dictionaryWithObjectsAndKeys:] (self=0xb776b0e0, _cmd=0xb77a37d8, firstObject=0x8dfdb24) at NSDictionary.m:539
#4  0xb75663e7 in lockPath (defaultsDatabase=0x8e240cc, verbose=0 '\000') at NSUserDefaults.m:160
#5  0xb7565d1e in -[NSUserDefaults initWithContentsOfFile:] (self=0x8dfda2c, _cmd=0xb77a3530, path=0x8e23f14) at NSUserDefaults.m:1137
#6  0xb7566787 in -[NSUserDefaults initWithUser:] (self=0x8dfda2c, _cmd=0xb77a3578, userName=0x8dfdb24) at NSUserDefaults.m:1182
#7  0xb75659a3 in -[NSUserDefaults init] (self=0x8dfda2c, _cmd=0xb77a3680) at NSUserDefaults.m:1094
#8  0xb7563bdd in +[NSUserDefaults standardUserDefaults] (self=0xb77a3160, _cmd=0xb777aa10) at NSUserDefaults.m:779
#9  0xb7473916 in +[NSLocale currentLocale] (self=0xb777a770, _cmd=0xb7769310) at NSLocale.m:443
#10 0xb73fb348 in -[NSDateFormatter init] (self=0x8dd652c, _cmd=0x804ba28) at NSDateFormatter.m:112
#11 0x08048ddf in main () at source.m:12

I attached the config.log files from both make and base: they ended with config: exit 0, what makes me suppose everything was OK there. 

Then all files in make check failed, for the reason above.

I admit I've got little experience in the unix environment, but it is a littlediscouraging that after 2 weeks of trying to build gnustep/étoilé have lead to nothing. Allthough I must admit I learned asome things about unix on the way.  Maybe this building is expert stuff...

Thanks for any answers,

2013/1/12 David Chisnall <address@hidden>
I think that these failures are things I fixed in the svn version of libobjc2 last week - please can you make sure that you have the latest version?


On 12 Jan 2013, at 09:23, edwin ancaer wrote:

> Hello,
> I was looking for an interesting platform to refresh my programming skills. After looking around a while, it found GNUstep/Etoilé looking interesting.
> I've been building GNUstep following the instructions in 'Downloads: Installation instruction'  on the Etoilé website.
> Everything went fine I sucessfully built make, base, gui and back.
> However, when I ran the 'make check' for the base package, I got the following result:
>    7552 Passed tests
>      25 Failed tests
>      19 Dashed hopes
>       5 Skipped sets
>       1 Failed file
> Given the fact that ProjectCenter seems to be working fine, do I have to worry about this, or is it normal that some tasts fail?
> The failed tests are of the form:
> Testing propertyAttrs.m...
> Running base/Functions/propertyAttrs.m...
> Passed test:     Lookup succeeded for property charDefault
> Passed test:     Proprety name should be 'charDefault' was 'charDefault'
> Failed test:     Property attributes for charDefault should be 'Tc,VcharDefault' was 'TcVcharDefault'
> Passed test:     Lookup succeeded for property doubleDefault
> Passed test:     Proprety name should be 'doubleDefault' was 'doubleDefault'
> Failed test:     Property attributes for doubleDefault should be 'Td,VdoubleDefault' was 'TdVdoubleDefault'
> Passed test:     Lookup succeeded for property enumDefault
> Passed test:     Proprety name should be 'enumDefault' was 'enumDefault'
> Failed test:     Property attributes for enumDefault should be 'Ti,VenumDefault' was 'TiVenumDefault'
> Passed test:     Lookup succeeded for property floatDefault
> To me it seems like they are coding, so I wonder if I've been using the sources of a developmant version?
> Further, how do I have to interpret these errors from the tests: are they really pointing to serous problems, given the fact that ProjectCenter (seems) to work correctly?
> Thanks in advance for helping out,,
> Edwin Ancaer   .
> _______________________________________________
> Discuss-gnustep mailing list
> address@hidden
> https://lists.gnu.org/mailman/listinfo/discuss-gnustep

-- Sent from my IBM 1620

Attachment: config.log
Description: Binary data

Attachment: config.log
Description: Binary data

reply via email to

[Prev in Thread] Current Thread [Next in Thread]