Hello,
>> Any suggestions how to debug this?
Tuomo> Try attaching gdb (--pid=`pidof ion3`) to the running process from
Tuomo> this xterm and do a backtrace to check where Ion is and if it really
Tuomo> is hung. It could be waiting for events from X... I think such an
Tuomo> X or Xlib bug was diagnosed on the IRC channel some time ago.
Here is the backtrace
,-----
| #0 0xb7e2c9f8 in select () from /lib/tls/libc.so.6
| #1 0xb7f25ea2 in _XPollfdCacheDel () from /usr/X11R6/lib/libX11.so.6
| #2 0xb7f26e11 in _XRead () from /usr/X11R6/lib/libX11.so.6
| #3 0xb7f27973 in _XReply () from /usr/X11R6/lib/libX11.so.6
| #4 0xb7f22db4 in XSync () from /usr/X11R6/lib/libX11.so.6
| #5 0x08058acf in clientwin_get_transient_for ()
| #6 0x08058ed2 in ioncore_manage_clientwin ()
| #7 0x0805c646 in ioncore_handle_map_request ()
| #8 0x0805c54f in ioncore_handle_event ()
| #9 0x0807615f in hook_remove_extl ()
| #10 0x0807622c in hook_call_alt ()
| #11 0x08076348 in hook_call_alt_p ()
| #12 0x0805697c in ioncore_x_connection_handler ()
| #13 0x080569de in ioncore_mainloop ()
| #14 0x08056479 in main ()
`-----
What does it translate to? Some subtle timing problem that did not
surface with earlier ion snapshots? Should I start ion3 earlier or
later in the session? Currently I do:
,-----
| LANG=en_US.UTF-8; export LANG;
| xmodmap ~/.keymaps/R40
|
| CVS_RSH=/home/upf/bin/sshS PATH=~/bin:$PATH emacs -geometry 80x49+0+0&
| xautolock -time 3 -corners 0+0- &
| xset r rate 200 40
| unclutter &
| fetchmail -d 300
| while ( true) ; do
| sleep 5;
| ion3
| done
`-----
Clumsy, I agree.
Ulrich Pfeifer
