Re: bluetoothd git version -> Segmentation fault

[Johan Hedberg <johan.hedberg@gmail.com>]

Hi Stuart,

On Sat, May 02, 2009, Stuart Pook wrote:
> bluetoothd seg faulted again. This time I didn't have to type control-C.
> I just ran it and it crashed. I last did a git pull a bit before
> 2009-04-29 20:17
<snip>
> bluetoothd[3300]: connect: Operation not permitted (1)
> ==3300== Invalid read of size 1
> ==3300==    at 0x4826728: strlen (mc_replace_strmem.c:242)
> ==3300==    by 0x498E677: vfprintf (in /lib/i686/cmov/libc-2.9.so)
> ==3300==    by 0x4A2A635: __vsyslog_chk (in /lib/i686/cmov/libc-2.9.so)
> ==3300==    by 0x4A2A706: vsyslog (in /lib/i686/cmov/libc-2.9.so)
> ==3300==    by 0x12D457: error (logging.c:58)
> ==3300==    by 0x4EEC176: get_record_cb (headset.c:1432)
> ==3300==    by 0x12D9DA: search_completed_cb (glib-helper.c:209)
> ==3300==    by 0x493F0B1: sdp_process (in /usr/local/stow/bluez-4.37/lib/libbluetooth.so.3.2.3)
> ==3300==    by 0x12DA46: search_process_cb (glib-helper.c:228)
> ==3300==    by 0x48B64AC: (within /usr/lib/libglib-2.0.so.0.2000.1)
> ==3300==    by 0x487F847: g_main_context_dispatch (in /usr/lib/libglib-2.0.so.0.2000.1)
> ==3300==    by 0x4882DAA: (within /usr/lib/libglib-2.0.so.0.2000.1)
> ==3300==  Address 0x5 is not stack'd, malloc'd or (recently) free'd
> ==3300== ==3300== Process terminating with default action of signal 11 
> (SIGSEGV)

Thanks a lot for this trace! It turned out to be a simple format string
bug which is now fixed in git:
http://git.kernel.org/?p=bluetooth/bluez.git;a=commitdiff;h=ca06dd9c4d4d4c055583dd4e2b4a823264c3b687

Johan