#496 GOP-related crash with rrad.split~

[mseta]

Some patches crash pd immediately upon opening.

I found a small patch that exhibits such
behaviour. Loading individual externs and
abstractions from that patch dos not cause problems, only
opening this patch as is. Backtrace is below.

This is on Karmic and 0.42.5-extended-20100110 (this is the last
version I tried but had exactly the same issue with some november and
december build).

Hans confirmed the behaviour on Pd-0.42.5-extended-20091217 on Mac OS X 10.5.8/Intel.

./MiS

Program received signal SIGSEGV, Segmentation fault.
text_xpix (x=0x8520918, glist=0x0) at g_graph.c:619
619 g_graph.c: No such file or directory.
in g_graph.c
bt
#0 rtext_senditup (x=0x0, action=<value optimized out>,
widthp=<value optimized out>, heightp=0xbfffe818, indexp=0xbfffe814)
at g_rtext.c:164
#1 0x0807d5b1 in rtext_width (x=0x0) at g_rtext.c:357
#2 0x08077d8f in text_getrect (z=0x86d2798, glist=0x86dde50, xp1=0xbfffe8bc,
yp1=0xbfffe8b8, xp2=0xbfffe8b4, yp2=0xbfffe8b0) at g_text.c:1014
#3 0x08074d9e in graph_getrect (z=0x86dde50, glist=0x8680e78, xp1=0xbfffe91c,
yp1=0xbfffe918, xp2=0xbfffe914, yp2=0xbfffe910) at g_graph.c:921
#4 0x080966e9 in gobj_shouldvis (x=0x86dde50, glist=0x8680e78)
at g_editor.c:80
#5 0x080743b4 in glist_getcanvas (x=0x86dde50) at g_graph.c:188
#6 0x0807e1db in rtext_new (glist=0x86dde50, who=0x86de078) at g_rtext.c:62
#7 0x080992a0 in canvas_create_editor (x=0x86dde50) at g_editor.c:902
#8 0x0807dbb4 in glist_findrtext (gl=0x86dde50, who=0x86d2798)
at g_rtext.c:347
#9 0x08079dfc in text_vis (z=0x86d2798, glist=0x86dde50, vis=1)
at g_text.c:1080
#10 0x016e775a in text_visfn_hax0r (o=0x86d2798, can=0x86dde50, vis=1)
at src/classes2.cxx.fcs:1076
#11 0x08075b74 in graph_vis (gr=0x86dde50, parent_glist=0x8680e78, vis=1)
at g_graph.c:853
#12 0x08075b74 in graph_vis (gr=0x8680e78, parent_glist=0x8680d00, vis=1)
at g_graph.c:853
#13 0x080722f4 in canvas_map (x=0x8680d00, f=1) at g_canvas.c:657
#14 0x080b6556 in pd_typedmess (x=0x8680d00, s=0x851fbc0, argc=1,
argv=0xbfffef3c) at m_class.c:792
#15 0x080b613c in pd_typedmess (x=0x88161b0, s=0x851fbc0, argc=1,
argv=0xbfffef3c) at m_class.c:813
#16 0x080bb47a in binbuf_eval (x=0x8532ac0, target=<value optimized out>,
argc=0, argv=0x0) at m_binbuf.c:726
#17 0x080c7447 in socketreceiver_read (x=0x85327e8, fd=6) at s_inter.c:558
#18 0x080c8549 in sys_domicrosleep (microsec=<value optimized out>,
pollem=<value optimized out>) at s_inter.c:196
#19 0x080c3760 in m_pollingscheduler () at m_sched.c:510
#20 m_mainloop () at m_sched.c:560
#21 0x080c6509 in sys_main (argc=2, argv=0xbffff334) at s_main.c:304
#22 0x080ce1ab in main (argc=2, argv=0xbffff334) at s_entry.c:32

[mseta]

mystery crash example

[Hans-Christoph Steiner]

This still happens on Mac OS X 10.5.8/Intel using Pd-extended 0.42.5-rc2

[Hans-Christoph Steiner]

abstraction for use with matrix-ex.pd

[Hans-Christoph Steiner]

It seems that the bug is related to this message coming form glist_xtopixels() in g_graph.c:

consistency check failed: glist_xtopixels