Re: [Libreoffice] Need help with failing cppunit tests

2011-03-07 Thread Francois Tigeot
On Tue, Mar 01, 2011 at 12:07:22AM +0100, Francois Tigeot wrote:
 On Mon, Feb 28, 2011 at 09:26:46PM +, Caolán McNamara wrote:
  
  This confirms the theory anyway. Doesn't help exactly pin down the
  problem, but it does point to some global dtor, or explicit
  atexit/cxa_finalize handler, so, as a guess see rtl_memory_fini in
  sal/source/alloc_global.c and the MACOSX entry in the corresponding
  makefile.mk. If you tweak those to do like macosx does it make a
  difference ?
 
 Unfortunately, it doesn't.

DragonFly's libc doesn't implement some of the needed C++ features.

Work is under way but in the meantime, the cppunit tests will continue to
fail.

-- 
Francois Tigeot
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice


Re: [Libreoffice] Need help with failing cppunit tests

2011-03-07 Thread Michael Meeks
Hi there,

On Mon, 2011-03-07 at 09:04 +0100, Francois Tigeot wrote:
 DragonFly's libc doesn't implement some of the needed C++ features.

Ho hum.

 Work is under way but in the meantime, the cppunit tests will
 continue to fail.

If the unit test fail, then almost certainly you'll fail in the same
way at run-time too.

HTH,

Michael.

-- 
 michael.me...@novell.com  , Pseudo Engineer, itinerant idiot


___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice


Re: [Libreoffice] Need help with failing cppunit tests

2011-03-07 Thread Francois Tigeot
Hiya,

On Mon, Mar 07, 2011 at 12:39:23PM +, Michael Meeks wrote:
 
   If the unit test fail, then almost certainly you'll fail in the same
 way at run-time too.

Maybe, maybe not. The cppunit tests build long chains of destructors
which may not be present in a real-world situation.

The only issue I've found at runtime so far is I'm unable to save normal
.odt or .ods files but this may not be related.

I've not yet encountered a single LibreOffice crash.

-- 
Francois Tigeot
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice


Re: [Libreoffice] Need help with failing cppunit tests

2011-03-07 Thread Caolán McNamara
On Mon, 2011-03-07 at 13:49 +0100, Francois Tigeot wrote:
 Maybe, maybe not. The cppunit tests build long chains of destructors
 which may not be present in a real-world situation.

Nah, the real-world list is staggering long :-)

 I've not yet encountered a single LibreOffice crash.

FWIW they'd only happen at exit or dlclose time of course. Mileage
varies with these things however.

C.

___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice


Re: [Libreoffice] Need help with failing cppunit tests

2011-02-28 Thread Francois Tigeot
On Thu, Feb 24, 2011 at 11:56:56AM +0100, Francois Tigeot wrote:
 On Thu, Feb 24, 2011 at 10:11:46AM +, Caolán McNamara wrote:
  On Thu, 2011-02-24 at 09:30 +0100, Francois Tigeot wrote:
  
 With STAR_RESOURCEPATH set, gdb ends up in an eternal wait state, exactly
 like when I ran it from build.

I am now using gcc 4.4.2 (was 4.1.2) and this time gdb is a bit more useful.

The cppunit test is taken from sal/qa/OStringBuffer and is a bit simpler
than my original choice.

The command is:

LD_LIBRARY_PATH=${LD_LIBRARY_PATH+${LD_LIBRARY_PATH}:}/home/ftigeot/Projects/LibreOffice/bootstrap/clone/ure/sal/unxdflyx3.pro/lib:/home/ftigeot/Projects/LibreOffice/bootstrap/solver/330/unxdflyx3.pro/lib
 ../../unxdflyx3.pro/bin/cppunittester 
../../unxdflyx3.pro/lib/librtl_OStringBuffer.so


(gdb) run
Starting program:
/home/ftigeot/Projects/LibreOffice/bootstrap/clone/ure/sal/unxdflyx3.pro/bin/cppunittester
 ../../unxdflyx3.pro/lib/librtl_OStringBuffer.so
OK (0)

Program received signal SIGSEGV, Segmentation fault.
0x000801535510 in ?? ()
(gdb) bt
#0  0x000801535510 in ?? ()
#1  0x0008010b7db9 in __cxa_finalize (dso=0x0) at 
/usr/src/lib/libc/../libc/stdlib/atexit.c:178
#2  0x0008010b7a4a in exit (status=0) at 
/usr/src/lib/libc/../libc/stdlib/exit.c:64
#3  0x0040223a in _start (ap=0x7fffcd60, cleanup=0x80050ef0d 
rtld_exit) at /usr/src/lib/csu/x86_64/crt1.c:101
(gdb) quit


-- 
Francois Tigeot
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice


Re: [Libreoffice] Need help with failing cppunit tests

2011-02-28 Thread Caolán McNamara
On Mon, 2011-02-28 at 20:07 +0100, Francois Tigeot wrote:
 Program received signal SIGSEGV, Segmentation fault.
 0x000801535510 in ?? ()
 (gdb) bt
 #0  0x000801535510 in ?? ()
 #1  0x0008010b7db9 in __cxa_finalize (dso=0x0) at 
 /usr/src/lib/libc/../libc/stdlib/atexit.c:178
 #2  0x0008010b7a4a in exit (status=0) at 
 /usr/src/lib/libc/../libc/stdlib/exit.c:64
 #3  0x0040223a in _start (ap=0x7fffcd60, cleanup=0x80050ef0d 
 rtld_exit) at /usr/src/lib/csu/x86_64/crt1.c:101
 (gdb) quit

This confirms the theory anyway. Doesn't help exactly pin down the
problem, but it does point to some global dtor, or explicit
atexit/cxa_finalize handler, so, as a guess see rtl_memory_fini in
sal/source/alloc_global.c and the MACOSX entry in the corresponding
makefile.mk. If you tweak those to do like macosx does it make a
difference ?

C.

___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice


Re: [Libreoffice] Need help with failing cppunit tests

2011-02-28 Thread Francois Tigeot
On Mon, Feb 28, 2011 at 09:26:46PM +, Caolán McNamara wrote:
 On Mon, 2011-02-28 at 20:07 +0100, Francois Tigeot wrote:
  Program received signal SIGSEGV, Segmentation fault.
  0x000801535510 in ?? ()
  (gdb) bt
  #0  0x000801535510 in ?? ()
  #1  0x0008010b7db9 in __cxa_finalize (dso=0x0) at 
  /usr/src/lib/libc/../libc/stdlib/atexit.c:178
 
 This confirms the theory anyway. Doesn't help exactly pin down the
 problem, but it does point to some global dtor, or explicit
 atexit/cxa_finalize handler, so, as a guess see rtl_memory_fini in
 sal/source/alloc_global.c and the MACOSX entry in the corresponding
 makefile.mk. If you tweak those to do like macosx does it make a
 difference ?

Unfortunately, it doesn't.

All is not lost, howewer. Thanks to you, I know at least where to look for
further investigations :-)

-- 
Francois Tigeot
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice


Re: [Libreoffice] Need help with failing cppunit tests

2011-02-24 Thread Caolán McNamara
On Thu, 2011-02-24 at 09:30 +0100, Francois Tigeot wrote:

 Next, I ran the command by hand and 
...
  ResId without ResMgr created terminate called

Running it manually in this scenario you don't have STAR_RESOURCEPATH
set, see the makefile.mk

C.

___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice


Re: [Libreoffice] Need help with failing cppunit tests

2011-02-23 Thread Caolán McNamara
On Wed, 2011-02-23 at 09:46 +0100, Francois Tigeot wrote:
 For now, I'm focussing on sc/qa/unit.

Because your log shows OK (10) before it falls over and dies, I
believe that the tests themselves are passing, and you are dying on
exit. Could be a few reasons for that, I suppose the first one to check
is what's your autogen/configure line ?

To find out if you are using system cppunit or the internal one, the
internal one has a fix for when building with debug-mode stl

Another possibility is some nasty atexit/__cxa_atexit problem. A hack to
experiment for that possibility can be found at
http://lists.freedesktop.org/archives/libreoffice/2011-January/005372.html

C.

___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice


Re: [Libreoffice] Need help with failing cppunit tests

2011-02-23 Thread Francois Tigeot
On Wed, Feb 23, 2011 at 12:36:14PM +, Caolán McNamara wrote:
 On Wed, 2011-02-23 at 09:46 +0100, Francois Tigeot wrote:
  For now, I'm focussing on sc/qa/unit.
 
 Because your log shows OK (10) before it falls over and dies, I
 believe that the tests themselves are passing, and you are dying on
 exit.

This means the other errors such as
  UniversalContentBroker::configureUcb(): No configuration
  ContentBroker::initialize - already initialized!
  PropertySetRegistry::getConfigProvider - No config provider!
  etc...
are only caused by the first issue ?

 Could be a few reasons for that, I suppose the first one to check
 is what's your autogen/configure line ?

I use these values:

export CFLAGS=-O -g -pipe
export CPPFLAGS=-I/usr/pkg/include
export LD_LIBRARY_PATH=/usr/pkg/lib

./autogen.sh \
--with-max-jobs=2 --with-num-cpus=2 \
--prefix=/usr/local/LibreOffice \
--enable-symbols --enable-debug \
--x-includes=/usr/pkg/include --x-libraries=/usr/pkg/lib \
--with-lang=en \
--without-java \
--disable-mozilla --disable-nss-module \
--disable-neon --disable-Xaw \
--with-system-headers --with-system-libs \
--with-system-python --with-system-libtextcat \
--without-system-jars \
--without-system-db \
--without-system-mdds \
--without-system-vigra \
--without-system-altlinuxhyph \
--without-system-mythes \
--without-system-odbc \
--without-system-cppunit \

The environment variables are needed, otherwise some components do not find
the correct path for include files or libraries.

 To find out if you are using system cppunit or the internal one, the
 internal one has a fix for when building with debug-mode stl

I tried both. I ported locally one of the LibreOffice patches which fixed
a similar crash to the system package but it had no effect.
Currently I'm using the internal fork.
 
 Another possibility is some nasty atexit/__cxa_atexit problem. A hack to
 experiment for that possibility can be found at
 http://lists.freedesktop.org/archives/libreoffice/2011-January/005372.html

Hmm. This could be a bit ugly.
I have applied the patch and I'm doing a partial build right now to see if it
changes anything.

-- 
Francois Tigeot
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice


Re: [Libreoffice] Need help with failing cppunit tests

2011-02-23 Thread Francois Tigeot
On Wed, Feb 23, 2011 at 02:13:34PM +0100, Francois Tigeot wrote:
 On Wed, Feb 23, 2011 at 12:36:14PM +, Caolán McNamara wrote:
  
  Another possibility is some nasty atexit/__cxa_atexit problem. A hack to
  experiment for that possibility can be found at
  http://lists.freedesktop.org/archives/libreoffice/2011-January/005372.html

Sadly, the first patch doesn't help. 
I still get the same errors, complete with segmentation faults.

-- 
Francois Tigeot
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice