jeremybennett opened the ticket and wrote:

I am repeatedly getting a segmentation fault when running 0.7.5. This occurs when running my own NoAI AI (JeremyAI), and the problem has only started since I introduced code using AIGroup.StartAutoReplace.

I've tried running under GDB to capture a backtrace as follows:

(gdb) bt
# 0 std::_Rb_tree_increment (__x=0x19)
at ../../../../libstdc++-v3/src/tree.cc:63
# 1 0x0015f13d in std::_Rb_tree_increment (__x=0x8d0d760)
at ../../../../libstdc++-v3/src/tree.cc:83
# 2 0x080b7eab in ftell () at ioftell.c:37
# 3 0x08096d45 in ftell () at ioftell.c:37
# 4 0x08077998 in ftell () at ioftell.c:37
# 5 0x0807a6f1 in ftell () at ioftell.c:37
# 6 0x0804d572 in ftell () at ioftell.c:37
# 7 0x081c020a in ftell () at ioftell.c:37
# 8 0x08086ada in ftell () at ioftell.c:37
# 9 0x080811a3 in ftell () at ioftell.c:37
# 10 0x08180116 in ftell () at ioftell.c:37
# 11 0x08181060 in ftell () at ioftell.c:37
# 12 0x0821de55 in ftell () at ioftell.c:37
# 13 0x08182071 in ftell () at ioftell.c:37
# 14 0x0820ec13 in ftell () at ioftell.c:37
# 15 0x00782a86 in __libc_start_main (main=0x820ebd0 <ftell+1846508>, argc=1,
ubp_av=0xbffff104, init=0x8253790 <ftell+2128044>,
fini=0x8253780 <ftell+2128028>, rtld_fini=0x7587e0 <_dl_fini>,
stack_end=0xbffff0fc) at libc-start.c:220
# 16 0x0804bf21 in ftell () at ioftell.c:37
(gdb)

It seems the tree being processed when the fault has occurred is corrupt - 0x19 seems an unlikely valid pointer value:

(gdb) print __x
$2 = (std::_Rb_tree_node_base *) 0x19
(gdb)

I attach the most recent autosave session and the main.nut and info.nut (the only files) from JeremyAI.

Let me know if I can add more diagnostics to help track the problem.

Attachments

• autosave6.sav (553.84 KiB)
• main.nut (60.30 KiB)
• info.nut (3.22 KiB)

Reported version: 0.7.5
Operating system: All

This issue was imported from FlySpray: https://bugs.openttd.org/task/3566