Bug#921178: Re: Bug#940914: uno-libs3: Segfault on launch in cppu::_copyConstructAnyFromData

2020-11-03 Thread Witold Baryluk
Ok. Good to know.

I will try to construct more exact instructions with stable reproduction method.

Thanks Rene, for testing.

On Tue, 3 Nov 2020 at 05:52, Rene Engelhard  wrote:
>
> Hi,
>
> Am 03.11.20 um 06:35 schrieb Rene Engelhard:
> >> As for reproduction, I believe the easiest is to grab the Debian live
> >> CD image and run it in qemu -kvm, and it will reproduce. It does for
> >> me. If you can't, please let me know.
> > Ah, Live?
> >
> > Does it only happen in Live and not in "real" systems?
> >
> > Will try
>
> Just works fine on debian-live-10.6.0-amd64-gnome.iso
>
>
> Regards,
>
>
> Rene
>



Bug#921178: Re: Bug#940914: uno-libs3: Segfault on launch in cppu::_copyConstructAnyFromData

2020-11-02 Thread Rene Engelhard
Hi,

Am 03.11.20 um 06:35 schrieb Rene Engelhard:
>> As for reproduction, I believe the easiest is to grab the Debian live
>> CD image and run it in qemu -kvm, and it will reproduce. It does for
>> me. If you can't, please let me know.
> Ah, Live?
>
> Does it only happen in Live and not in "real" systems?
>
> Will try

Just works fine on debian-live-10.6.0-amd64-gnome.iso


Regards,


Rene



Bug#921178: Re: Bug#940914: uno-libs3: Segfault on launch in cppu::_copyConstructAnyFromData

2020-11-02 Thread Rene Engelhard
Hi,

Am 02.11.20 um 23:30 schrieb Witold Baryluk:
> I did provide stack traces for example in February 2019.
>
> https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=921178
> https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=940914

True, after which I said I simply can't reproduce it.

> There was zero response from the maintainer, i.e. what other
> information is required.

That is demonstratablly completely untrue.

I did reply to the bug, didn't I? And tagged it + unreproducible.


What is needed: Obviously telling circumstances when it happens.

Backtraces are good, but to make upstream debug it one probably needs to
tell them when it might happen, too

(and to check whether it does not happen anymore.)


And that is totally unclear.

> As for reproduction, I believe the easiest is to grab the Debian live
> CD image and run it in qemu -kvm, and it will reproduce. It does for
> me. If you can't, please let me know.

Ah, Live?

Does it only happen in Live and not in "real" systems?

Will try


Regards,


Rene



Bug#921178: Re: Bug#940914: uno-libs3: Segfault on launch in cppu::_copyConstructAnyFromData

2020-11-02 Thread Witold Baryluk
I did provide stack traces for example in February 2019.

https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=921178
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=940914

There was zero response from the maintainer, i.e. what other
information is required.

I am attaching a new stack trace to this email just in case.

As for reproduction, I believe the easiest is to grab the Debian live
CD image and run it in qemu -kvm, and it will reproduce. It does for
me. If you can't, please let me know.

I appreciate your help Rene.

Regards,
Witold
[2062261.823403] audit: type=1400 audit(1604356080.583:1375): 
apparmor="ALLOWED" operation="open" info="Failed name lookup - disconnected 
path" error=-13 profile="libreoffice-soffice" name="rw/usr/share/drirc.d" 
pid=275558 comm="soffice.bin" requested_mask="r" denied_mask="r" fsuid=1000 
ouid=0
[2062261.824432] audit: type=1400 audit(1604356080.587:1376): 
apparmor="ALLOWED" operation="open" info="Failed name lookup - disconnected 
path" error=-13 profile="libreoffice-soffice" name="rw/usr/share/drirc.d" 
pid=275558 comm="soffice.bin" requested_mask="r" denied_mask="r" fsuid=1000 
ouid=0
[2062261.917652] audit: type=1400 audit(1604356080.679:1377): 
apparmor="ALLOWED" operation="open" info="Failed name lookup - disconnected 
path" error=-13 profile="libreoffice-soffice" name="rw/usr/share/drirc.d" 
pid=275558 comm="soffice.bin" requested_mask="r" denied_mask="r" fsuid=1000 
ouid=0
[2062261.918059] audit: type=1400 audit(1604356080.679:1378): 
apparmor="ALLOWED" operation="open" info="Failed name lookup - disconnected 
path" error=-13 profile="libreoffice-soffice" name="rw/usr/share/drirc.d" 
pid=275558 comm="soffice.bin" requested_mask="r" denied_mask="r" fsuid=1000 
ouid=0
[2062261.918601] audit: type=1400 audit(1604356080.679:1379): 
apparmor="ALLOWED" operation="open" info="Failed name lookup - disconnected 
path" error=-13 profile="libreoffice-soffice" name="rw/usr/share/drirc.d" 
pid=275558 comm="soffice.bin" requested_mask="r" denied_mask="r" fsuid=1000 
ouid=0
[2062261.918729] audit: type=1400 audit(1604356080.679:1380): 
apparmor="ALLOWED" operation="open" profile="libreoffice-soffice" 
name="/usr/share/libdrm/amdgpu.ids" pid=275558 comm="soffice.bin" 
requested_mask="r" denied_mask="r" fsuid=1000 ouid=0
[2062262.548519] audit: type=1400 audit(1604356081.311:1381): 
apparmor="ALLOWED" operation="open" info="Failed name lookup - disconnected 
path" error=-13 profile="libreoffice-soffice" 
name="rw/home/user/.config/libreoffice/4/user/extensions" pid=275527 
comm="soffice.bin" requested_mask="r" denied_mask="r" fsuid=1000 ouid=1000
[2062262.549217] audit: type=1400 audit(1604356081.311:1382): 
apparmor="ALLOWED" operation="open" info="Failed name lookup - disconnected 
path" error=-13 profile="libreoffice-soffice" 
name="rw/usr/lib/libreoffice/program/services" pid=275527 comm="soffice.bin" 
requested_mask="r" denied_mask="r" fsuid=1000 ouid=0
warning: Currently logging to gdbtrace.log.  Turn the logging off and on to make the new setting effective.
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
[Detaching after fork from child process 275558]
[New Thread 0x7fb599ea0700 (LWP 275599)]

Thread 1 "soffice.bin" received signal SIGSEGV, Segmentation fault.
cppu::_copyConstructAnyFromData (pDestAny=pDestAny@entry=0x56064a0541c8, pSource=0x7ffd7ef0dc40, pType=, pTypeDescr=, acquire=0x7fb5a2afba20 , mapping=0x0) at ./cppu/source/uno/copy.hxx:212
212	./cppu/source/uno/copy.hxx: No such file or directory.
#0  cppu::_copyConstructAnyFromData(_uno_Any*, void*, _typelib_TypeDescriptionReference*, _typelib_TypeDescription*, void (*)(void*), _uno_Mapping*) (pDestAny=pDestAny@entry=0x56064a0541c8, pSource=0x7ffd7ef0dc40, pType=, pTypeDescr=, acquire=0x7fb5a2afba20 , mapping=0x0) at ./cppu/source/uno/copy.hxx:212
#1  0x7fb59f39cc64 in cppu::_copyConstructAny(_uno_Any*, void*, _typelib_TypeDescriptionReference*, _typelib_TypeDescription*, void (*)(void*), _uno_Mapping*) (pDestAny=0x56064a0541c8, pSource=, pSource@entry=0x7ffd7ef0dc40, pType=, pTypeDescr=pTypeDescr@entry=0x0, acquire=acquire@entry=0x7fb5a2afba20 , mapping=mapping@entry=0x0) at ./cppu/source/uno/copy.hxx:289
#2  0x7fb59f39c5b3 in uno_type_any_assign(uno_Any*, void*, typelib_TypeDescriptionReference*, uno_AcquireFunc, uno_ReleaseFunc) (pDest=, pSource=pSource@entry=0x7ffd7ef0dc40, pType=, acquire=acquire@entry=0x7fb5a2afba20 , release=release@entry=0x7fb5a2afba30 ) at ./cppu/source/uno/any.cxx:39
#3  0x7fb5a47860b2 in com::sun::star::uno::operator<<=(com::sun::star::uno::Any&, com::sun::star::beans::NamedValue const&) (value=..., rAny=...) at ./include/com/sun/star/uno/Type.h:155
#4  utl::ConfigManager::acquireTree(utl::ConfigItem const&) (item=...) at ./unotools/source/config/configmgr.cxx:123
#5  0x7fb5a4786c06 in utl::ConfigManager::addConfigItem(utl::ConfigItem&) (this=0x7fb5a61aa900 ::get()::instance>, item=...) at