[Gambas-user] Bug during compilation of the lastest Gambas svn sources rev#4720

Benoît Minisini gambas at ...1...
Wed May 16 20:05:45 CEST 2012


Le 16/05/2012 18:53, Olivier Cruilles a écrit :
> Hi Benoit,
>
> I think this is better than the preview email:
>
> [linus at ...2073... ~]$ valgrind --tool=memcheck --num-callers=50 gbr3 /usr/bin/gambas3.gambas /local/Gambas-3/gbEditor3
> ==26488== Memcheck, a memory error detector
> ==26488== Copyright (C) 2002-2009, and GNU GPL'd, by Julian Seward et al.
> ==26488== Using Valgrind-3.5.0 and LibVEX; rerun with -h for copyright info
> ==26488== Command: gbr3 /usr/bin/gambas3.gambas /local/Gambas-3/gbEditor3
> ==26488==
> Gtk-Message: (for origin information, set GTK_DEBUG): failed to retrieve property `GtkTreeView::odd-row-color' of type `GdkColor' from rc file value "((GString*) 0x62d9480)" of type `GString'
>
>
>
> warning: unable to load 24/earth
> ==26488== Invalid read of size 4
> ==26488==    at 0x806413D: GB_Raise (gbx_api.c:603)
> ==26488==    by 0x600BAA3: post_event(void*, int) (CEditor.cpp:1416)
> ==26488==    by 0x8074D15: SIGNAL_raise_callbacks (gbx_signal.c:143)

Mmm. It is theoritically impossible that SIGNAL_raise_callbacks calls 
the post_event() function, as only one signal handler is registered by 
the interpreter at the moment, to handle child process termination. And 
this is not the post_event routine.

Can you try the last revision? I fixed some race conditions in the 
signal management routines. Maybe it could be the cause of these strange 
errors...

-- 
Benoît Minisini




More information about the User mailing list