Using ddd-3.2-4 to edit gdb settings causes gdb-5.0-7 to get a segmentation fault. Here's the stack: #0 0x809f5b7 in set_architecture_from_arch_mach () #1 0x80c8946 in execute_command () #2 0x809a34d in async_disable_stdin () #3 0x809a9c4 in async_disable_stdin () #4 0x8165c58 in rl_callback_read_char () #5 0x809a220 in stdin_event_handler () #6 0x80e1c4a in delete_file_handler () #7 0x80e16c9 in _initialize_regcache () #8 0x80e1719 in _initialize_regcache () #9 0x80c7b4c in catch_errors () #10 0x80e1764 in start_event_loop () #11 0x806d248 in _start () #12 0x80c7b4c in catch_errors () #13 0x806dbe3 in _start () #14 0x80c7b4c in catch_errors () #15 0x806dc17 in main () #16 0x400a2b65 in __libc_start_main (main=0x806dbf0 <main>, argc=4, ubp_av=0xbffff744, init=0x806c718 <_init>, fini=0x8175d7c <_fini>, rtld_fini=0x4000df24 <_dl_fini>, stack_end=0xbffff73c) at ../sysdeps/generic/libc-start.c:111 (gdb) Segmentation fault
Please try 5.0-11 in Rawhide.
Have you tried the Rawhide version yet? Also, exactly how did you produce this bug?
I upgraded to gdb-5.0-11 and it still happens but is not predictable. I was debugging an OpenMotif application and trying to print out a struct. I was able to use printf's to get around this. I will try to contruct a small program to isolate the problem. (Sorry, I should have done this the first time.)
I can still reproduce this (message of gdb segaulting) just by choosing "edit gb preferences." Debugging this is a pain.
Fixed in gdb-5.0rh-9 - coming soon to Rawhide, available for a limited time from http://people.redhat.com/teg/gdb/