Angus Leeming wrote: > You can run valgrind with a --connect-gdb option (or suchlike), so > that when the program crashes gdb fires up and you can extract a > backtrace. I'd do it for you, but valgrind doesn't work on Alpha > machines...
Thank you, I missed that option (probably because it is called --db-attach=yes and db is almost automatically translated to database somewhere on the way from the eyes to the brain...) Unfortunately it does not help very much: ==29942== Process terminating with default action of signal 11 (SIGSEGV) ==29942== at 0x1B8E967E: _dl_map_object_from_fd (in /lib/ld-2.3.3.so) ==29942== by 0x1B8EACC9: _dl_map_object (in /lib/ld-2.3.3.so) ==29942== by 0x1B8F09CD: openaux (in /lib/ld-2.3.3.so) ==29942== by 0x1B8F0D55: _dl_catch_error (in /lib/ld-2.3.3.so) ==29942== ==29942== ---- Attach to debugger ? --- [Return/N/n/Y/y/C/c] ---- y ==29942== starting debugger with cmd: /usr/bin/gdb -nw /proc/29944/fd/821 29944 GNU gdb 6.2.1 Copyright 2004 Free Software Foundation, Inc. GDB is free software, covered by the GNU General Public License, and you are welcome to change it and/or distribute copies of it under certain conditions. Type "show copying" to see the conditions. There is absolutely no warranty for GDB. Type "show warranty" for details. This GDB was configured as "i586-suse-linux"...Using host libthread_db library "/lib/tls/libthread_db.so.1". Attaching to program: /proc/29944/fd/821, process 29944 0x1b8e967e in ?? () (gdb) bt #0 0x1b8e967e in ?? () Cannot access memory at address 0x52bfcfb4 (gdb) Georg
