I know of one problem with destructed indices that I've already got a test case for. Do you have destructed indices in your case too?
I don't get a segfault at precisely that point however, so it might be that you see a different problem even if it too involves destructs. So if it isn't too much work to isolate a test case, I'd appreciate it.
/ Martin Stjernholm, Roxen IS
Previous text:
2004-06-10 14:55: Subject: segfault in red-black?
How interesting is this? Should I persue it further? (Pike 7.7 from CVS tuesday.)
Program received signal SIGSEGV, Segmentation fault. 0x080c28d0 in multiset_delete_node (l=0x82fd00c, nodepos=1) at /home/mirar/pike/src/multiset.c:2559 2559 while (RBNODE (node) != existing) LOW_RB_TRACK_PREV (rbstack, node); (gdb) bt #0 0x080c28d0 in multiset_delete_node (l=0x82fd00c, nodepos=1) at /home/mirar/pike/src/multiset.c:2559 #1 0x080bfec4 in multiset_next (l=0x82fd00c, nodepos=2) at /home/mirar/pike/src/multiset.c:1757 #2 0x08127a6f in foreach_iterate (o=0x85e8b28, do_step=1) at iterators.cmod:1944 #3 0x0807189f in test_opcode_F_FOREACH_LOOP () at interpret_functions.h:1416 #4 0x083eadf3 in ?? () #5 0x00000011 in ?? () #6 0x00000001 in ?? () #7 0x0000000f in ?? () #8 0x4020d6d0 in __after_morecore_hook () from /lib/libc.so.6 #9 0x4020cc40 in ?? () from /lib/libc.so.6 #10 0xbfffee20 in ?? () #11 0x000000d2 in ?? ()
void draw2() { Node node1=this; -> foreach (indices(links);;Node node2) { passed->line(node1->x,node1->y, node2->x,node2->y, 255,0,0); } }
/ Mirar