[Freeciv-Dev] (PR#6406) gtk client crash on moving a caravel

["Jason Short" <jdorje@xxxxxxxxxxxxxxxxxxxxx>]

[chrisk@xxxxxxxxx - Sun Oct  5 22:14:32 2003]:

> CVS 05 OCT 2003 S1_14 BETA GTK 1.2
> 
> gtk client crash on moving a caravel
> 
> 386               Impr_Status stat = equiv_list[i][*ept];
> (gdb) bt
> #0  improvement_redundant (pplayer=0x81b9e40, pcity=0x8346078, id=7,
> want_to_build=0)
>     at improvement.c:386
> #1  0x080b383e in update_all_effects () at game.c:1082
> #2  0x0806e811 in try_update_effects (need_update=1) at packhand.c:268
> #3  0x0806f3cb in handle_short_city (packet=0x83eba78) at packhand.c:643
> #4  0x08062569 in handle_packet_input (packet=0x83eba78, type=83) at
> civclient.c:303
> #5  0x080659d3 in input_from_server (fd=6) at clinet.c:326
> #6  0x080937c2 in get_net_input (data=0x0, fid=6,
condition=GDK_INPUT_READ)
>     at gui_main.c:1061
> #7  0x40194934 in gdk_io_invoke () from /usr/lib/libgdk-1.2.so.0
> #8  0x401c63b0 in g_io_unix_dispatch () from /usr/lib/libglib-1.2.so.0
> #9  0x401c7c46 in g_main_dispatch () from /usr/lib/libglib-1.2.so.0
> #10 0x401c8273 in g_main_iterate () from /usr/lib/libglib-1.2.so.0
> #11 0x401c843c in g_main_run () from /usr/lib/libglib-1.2.so.0
> #12 0x400e076c in gtk_main () from /usr/lib/libgtk-1.2.so.0
> #13 0x0809333e in ui_main (argc=5, argv=0xbffff424) at gui_main.c:895
> #14 0x08062438 in main (argc=5, argv=0xbffff424) at civclient.c:228
> #15 0x4034e7ee in __libc_start_main () from /lib/libc.so.6
> 
> In the savegame, the actual crash happened while moving the caravel
north of
> Ireland ->S->SE.
> 
> In some cases, I was able to reproduce this bug from the savegame, in
others
> I was not (using the very same sequence of input actions to the
client, and
> restarted server and client). So I'd think this crash is due to a side
> effect of memory corruption.

I can't reproduce this (with current CVS, that is) - nor does Valgrind
give any memory warnings.  The core files aren't of any use without the
original binary.

It looks like a gen-effects problem, but could be memory corruption
elsewhere.

jason