Re: Test on Ubuntu 12.04 64bits

[David Chisnall]

I've updated the analyser results here:

http://theravensnest.org/gscorebase/

Stef - please make sure that the functions in the headers have the correct 
ownership annotations, or CoreBase will behave incorrectly when used with ARC, 
which will make porting iOS / OS X code difficult.

David

On 27 May 2012, at 17:45, Benoît Garrigues wrote:

> Updated corebase to rev 35167.
> 
> - Two remaining compilation warnings :
> 
> Compiling file CFLocale.c ...
> CFLocale.c:335:15: warning: result of comparison against a string
> literal is unspecified (use strncmp instead) [-Wstring-compare]
>  if (context == (const void*)ICU_CALENDAR_KEY)
>              ^  ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> 1 warning generated.
> 
> Compiling file NSCFError.m ...
> NSCFError.m:48:10: warning: incompatible pointer types returning
> 'CFErrorRef' (aka 'NSError *') from a function with result type
> 'NSCFError *' [-Wincompatible-pointer-types]
>  return CFErrorCreate (NULL, domain, code, userInfo);
>         ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> 1 warning generated.
> 
> 
> - Problem in configure.ac and Source/GNUmakefile.in when using gnustep
> libobjc2 and gcc libobjc is installed :
> I have to edit manually Source/GNUmakefile
> 
>     => 1st try : replace -lobjc with -l:libobjc.so.4
> 
> $ make check
>    ...
>    198 Passed tests
>      8 Failed builds
>      1 Failed test
>      1 Failed file
>      1 Dashed hope
> 
> Error in tests.log :
> ...
> Building in CFLocale
> Making all for test_tool create...
> Making all for test_tool displayvalues...
> Making all for test_tool identifier...
> Making all for test_tool values...
> Compiling file create.m ...
> Compiling file displayvalues.m ...
> Compiling file identifier.m ...
> Compiling file values.m ...
> Linking test_tool identifier ...
> Linking test_tool displayvalues ...
> Linking test_tool create ...
> Linking test_tool values ...
> /home/benoit/Projets/OpenSource/objc/GNUstep-anonymous/dev-libs/corebase/Source/./obj/libgnustep-corebase.so:
> undefined reference to `__objc_class_name_NSObject'
> /home/benoit/Projets/OpenSource/objc/GNUstep-anonymous/dev-libs/corebase/Source/./obj/libgnustep-corebase.so:
> undefined reference to `__objc_class_name_NSObject'
> /home/benoit/Projets/OpenSource/objc/GNUstep-anonymous/dev-libs/corebase/Source/./obj/libgnustep-corebase.so:
> undefined reference to `__objc_class_name_NSMutableString'
> /home/benoit/Projets/OpenSource/objc/GNUstep-anonymous/dev-libs/corebase/Source/./obj/libgnustep-corebase.so:
> undefined reference to `__objc_class_name_NSMutableString'
> /home/benoit/Projets/OpenSource/objc/GNUstep-anonymous/dev-libs/corebase/Source/./obj/libgnustep-corebase.so:
> undefined reference to `__objc_class_name_NSBundle'
> ...
> 
> 
>    => 2nd try : replace -lobjc with `gnustep-config --base-libs`
> 
> `gnustep-config --base-libs` = -rdynamic -pthread -fobjc-nonfragile-abi
> -fgnu-runtime -L/home/benoit/GNUstep/Library/Libraries
> -L/usr/GNUstep/Local/Library/Libraries
> -L/usr/GNUstep/System/Library/Libraries -lgnustep-base -l:libobjc.so.4 -lm
> 
> $ make check
>    ...
>    251 Passed tests
>      2 Failed tests
>      1 Failed file
>      1 Dashed hope
> 
> tests.log attached.
> 
> - Trace for error in ./Tests/CFAttributedString/obj/mutable :
> 
> Program received signal SIGSEGV, Segmentation fault.
> 0x00007ffff7086efe in objc_test_class_flag (aClass=0x74007300200061,
> flag=objc_class_flag_hidden_class) at
> /home/benoit/Projets/OpenSource/objc/GNUstep-anonymous/dev-libs/libobjc2/./class.h:248
> 248        return (aClass->info & (unsigned long)flag) == (unsigned
> long)flag;
> (gdb) bt
> #0  0x00007ffff7086efe in objc_test_class_flag (aClass=0x74007300200061,
> flag=objc_class_flag_hidden_class) at
> /home/benoit/Projets/OpenSource/objc/GNUstep-anonymous/dev-libs/libobjc2/./class.h:248
> #1  0x00007ffff708791f in object_getClass (obj=0x687350) at runtime.c:757
> #2  0x00007ffff72c2939 in CFGetTypeID (cf=<optimized out>) at
> CFRuntime.c:308
> #3  0x00007ffff72c2ca2 in CFHash (cf=<error reading variable: Unhandled
> dwarf expression opcode 0xf1>) at CFRuntime.c:318
> #4  0x00007ffff72cf6f9 in GSHashTableFindBucket (table=<error reading
> variable: Unhandled dwarf expression opcode 0x0>, key=<error reading
> variable: Unhandled dwarf expression opcode 0x0>) at GSHashTable.c:187
> #5  0x00007ffff72d02e4 in GSHashTableRemoveValue (table=<optimized out>,
> key=<optimized out>) at GSHashTable.c:653
> #6  0x00007ffff72b8fdb in CFAttributedStringUncacheAttribute
> (attribs=<optimized out>, attribs=<optimized out>, attribs=<optimized
> out>) at CFAttributedString.c:137
> #7  RemoveAttributesAtIndex (str=<error reading variable: Unhandled
> dwarf expression opcode 0x0>, range=...) at CFAttributedString.c:523
> #8  0x00007ffff72b8eff in CFAttributedStringCreateMutableCopy
> (alloc=<error reading variable: Unhandled dwarf expression opcode 0x0>,
> maxLength=<optimized out>, str=<optimized out>) at CFAttributedString.c:625
> #9  0x000000000040122f in main () at mutable.m:24
> 
> 
> Benoit
> 
> 
> 
> Le 27/05/2012 17:52, Stefan Bidi a écrit :
>> On Sun, May 27, 2012 at 10:11 AM, Fred Kiefer <fredkiefer@gmx.de> wrote:
>>> David's and your changes help a lot. The only remaining compiler warning I
>>> get is this:
>>> 
>>> Compiling file CFTimeZone.c ...
>>> CFTimeZone.c: In function ‘CFTimeZoneCreate’:
>>> CFTimeZone.c:163:3: warning: dereferencing type-punned pointer will break
>>> strict-aliasing rules [-Wstrict-aliasing]
>>> CFTimeZone.c:164:3: warning: dereferencing type-punned pointer will break
>>> strict-aliasing rules [-Wstrict-aliasing]
>>> CFTimeZone.c:165:3: warning: dereferencing type-punned pointer will break
>>> strict-aliasing rules [-Wstrict-aliasing]
>> I've known about this for sometime now, but not sure how to tell the
>> compiler "trust me, I know what I'm doing this time!"
>> 
>>> And the tests look much better as well:
>>> 
>>> 
>>> CFString/format.m:
>>> Failed test:     format.m:28 ... Float/Doubles are formatted correctly
>> Could I get your tests.log file?  Not sure why this would fail.  The
>> only explanation I've got is rounding error.
>> 
>>> --- Running tests in CFTimeZone ---
>>> 
>>> CFTimeZone/general.m:
>>> Failed test:     general.m:54 ... Daylight Saving time offset at 0 second
>>> from absolute epoch is '3600.000000'.
>> I'll eventually fix this.
>> 
>>> --- Running tests in CFTree ---
>>> --- Running tests in CFURL ---
>>> --- Running tests in CFURLAccess ---
>>> 
>>> CFURLAccess/basic.m:
>>> Failed test:     basic.m:24 ... Directory was successfully created.
>> This is giving you an error because the test is trying to create a
>> directory that already exists.  It exists because the last time you
>> ran the test it crashed before cleaning up.  Running them again will
>> yield a different result now that it ran to the end and deleted
>> everything correctly.
>> 
>>> --- Running tests in CFUUID ---
>>> find: `TestDir': Datei oder Verzeichnis nicht gefunden
>>> 
>>>   259 Passed tests
>>>     3 Failed tests
>>>     1 Dashed hope
>>> 
>>> 
>>> _______________________________________________
>>> Discuss-gnustep mailing list
>>> Discuss-gnustep@gnu.org
>>> https://lists.gnu.org/mailman/listinfo/discuss-gnustep
>> _______________________________________________
>> Discuss-gnustep mailing list
>> Discuss-gnustep@gnu.org
>> https://lists.gnu.org/mailman/listinfo/discuss-gnustep
> <tests.log>_______________________________________________
> Discuss-gnustep mailing list
> Discuss-gnustep@gnu.org
> https://lists.gnu.org/mailman/listinfo/discuss-gnustep

--
This email complies with ISO 3103