About "undisposed bridge, potential deadlock ahead"

Hello,

On pc Debian x86-64 with master sources updated 3 days ago + enable-dbgutil,
I runned "make check" and had a problem on starmath

I tried to understand and noticed this:
warn:binaryurp:28548:13:binaryurp/source/bridge.cxx:846: undisposed bridge,
potential deadlock ahead
when launching:
1) make debugrun then run on first console
2) make gb_JunitTest_DEBUGRUN=T JunitTest_starmath_unoapi on second console

Putting a breakpoint, I retrieved this:
(gdb) bt
#0  binaryurp::Bridge::~Bridge (this=0x2e1bf90, __in_chrg=<optimized out>)
at
/home/julien/compile-libreoffice/libreoffice/binaryurp/source/bridge.cxx:846
#1  0x00002aaac647900e in binaryurp::Bridge::~Bridge (this=0x2e1bf90,
__in_chrg=<optimized out>)
    at
/home/julien/compile-libreoffice/libreoffice/binaryurp/source/bridge.cxx:850
#2  0x00002aaaad72afca in cppu::OWeakObject::release (this=0x2e1bf90) at
/home/julien/compile-libreoffice/libreoffice/cppuhelper/source/weak.cxx:205
#3  0x00002aaac648203a in
cppu::WeakImplHelper2<com::sun::star::bridge::XBridge,
com::sun::star::lang::XComponent>::release (this=0x2e1bf90)
    at
/home/julien/compile-libreoffice/libreoffice/include/cppuhelper/implbase2.hxx:106
#4  0x00002aaac64bafdf in rtl::Reference<binaryurp::Bridge>::clear
(this=0x2e3ac88) at
/home/julien/compile-libreoffice/libreoffice/include/rtl/ref.hxx:124
#5  0x00002aaac64b7c4b in binaryurp::Reader::execute (this=0x2e3ac60) at
/home/julien/compile-libreoffice/libreoffice/binaryurp/source/reader.cxx:127
#6  0x00002aaaadec197c in salhelper::Thread::run (this=0x2e3ac60) at
/home/julien/compile-libreoffice/libreoffice/salhelper/source/thread.cxx:40
#7  0x00002aaaadec1d79 in osl::threadFunc (param=0x2e3ac70) at
/home/julien/compile-libreoffice/libreoffice/include/osl/thread.hxx:184
#8  0x00002aaaaad39889 in osl_thread_start_Impl (pData=0x2e2d1e0) at
/home/julien/compile-libreoffice/libreoffice/sal/osl/unx/thread.cxx:240
#9  0x00002aaaab6380a4 in start_thread (arg=0x2aaad45d0700) at
pthread_create.c:309
#10 0x00002aaaab36d04d in clone () at
../sysdeps/unix/sysv/linux/x86_64/clone.S:111

Indeed a Writer window is still opened at the end of the test

Also, when relaunching "make check" (without having updated local repo), I
have now this:
Failures that appeared during scenario execution:
         sc.ScAccessiblePreviewTable
         sc.ScCellRangesObj
         sc.ScDataPilotTablesObj
3 of 92 tests failed
instead of the starmath test problem.

Any thoughts?

Julien



--
View this message in context: 
http://nabble.documentfoundation.org/About-undisposed-bridge-potential-deadlock-ahead-tp4142562.html
Sent from the Dev mailing list archive at Nabble.com.