Hello,

whenever I edit some math formula (in an otherwise empty doc) and close the document afterwards, the following valgrind report (and many more) appears:

==28940== Invalid read of size 4
==28940== at 0x8192E8E: MathCursor::info() const (math_cursor.C:1298)
==28940== by 0x80D54F7: bv_funcs::currentState(BufferView*) (bufferview_funcs.C:295)
==28940== by 0x811815D: LyXFunc::view_status_message() (lyxfunc.C:1966)
==28940== by 0x8282ADB: QtView::update_view_state_qt() (QtView.C:129)
==28940== by 0x8282A84: QtView::clearMessage() (QtView.C:116)
==28940== by 0x8067B51: BufferView::Pimpl::resizeCurrentBuffer() (BufferView_pimpl.C:484)
==28940== by 0x8066FA9: BufferView::Pimpl::buffer(Buffer*) (BufferView_pimpl.C:324)
==28940== by 0x8062122: BufferView::buffer(Buffer*) (BufferView.C:107)
==28940== by 0x8117FCD: LyXFunc::closeBuffer() (lyxfunc.C:1919)
==28940== by 0x8112C1B: LyXFunc::dispatch(FuncRequest const&, bool) (lyxfunc.C:1087)
==28940== Address 0x437F5A00 is 0 bytes inside a block of size 100 free'd
==28940== at 0x4004BDA1: __builtin_delete (vg_replace_malloc.c:233)
==28940== by 0x4004BDBF: operator delete(void*) (vg_replace_malloc.c:242)
==28940== by 0x81B1E3F: MathHullInset::~MathHullInset() (math_hullinset.C:476)
==28940== by 0x818EC8A: MathAtom::~MathAtom() (math_atom.C:49)
==28940== by 0x8189C0F: InsetFormula::~InsetFormula() (formula.C:110)
==28940== by 0x8082803: InsetList::~InsetList() (InsetList.C:52)
==28940== by 0x81389BB: Paragraph::~Paragraph() (paragraph.C:126)
==28940== by 0x80754F2: void std::_Destroy<Paragraph>(Paragraph*) (/usr/include/g++/bits/stl_construct.h:125)
==28940== by 0x807547E: std::_List_base<Paragraph, std::allocator<Paragraph> >::__clear() (/usr/include/g++/bits/list.tcc:77)
==28940== by 0x80C5314: std::list<Paragraph, std::allocator<Paragraph> >::clear() (/usr/include/g++/bits/stl_list.h:919)


==28940==    at 0x8192E8E: MathCursor::info() const (math_cursor.C:1298)
1298                    Cursor_[i].inset_->infoize(os);

Maybe you have more ideas on what is going on than me :-)

Michael




Reply via email to