#111 syncterm hang if start with X11 interface

[Ragnarok]

I trying to start syncterm with X11 on debian bullseye, It's hang and wont start.
Using -iS (for SDL) works fine
Running via gdb i cannot find more useful information. =|

(gdb) set args -iX
(gdb) r
Starting program: /usr/local/bin/syncterm -iX
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
[New Thread 0x7ffff7bbb700 (LWP 3123975)]
[Thread 0x7ffff7bbb700 (LWP 3123975) exited]
[New Thread 0x7ffff7fcb700 (LWP 3123976)]

^C
Thread 1 "Main Thread" received signal SIGINT, Interrupt.
futex_wait_cancelable (private=0, expected=0, futex_word=0x55555dc81038)
    at ../sysdeps/nptl/futex-internal.h:186
186 ../sysdeps/nptl/futex-internal.h: No existe el fichero o el directorio.
(gdb) bt
#0  futex_wait_cancelable (private=0, expected=0, futex_word=0x55555dc81038)
    at ../sysdeps/nptl/futex-internal.h:186
#1  __pthread_cond_wait_common (abstime=0x0, clockid=0, mutex=0x55555dc80fe8, 
    cond=0x55555dc81010) at pthread_cond_wait.c:508
#2  __pthread_cond_wait (cond=0x55555dc81010, mutex=0x55555dc80fe8)
    at pthread_cond_wait.c:638
#3  0x000055555574868d in xp_sem_wait (sem=0x55555dc6ad20 <init_complete>)
    at xpsem.c:164
#4  0x0000555555714b8a in x_init () at x_cio.c:567
#5  0x00005555556f3e14 in try_x_init (mode=5) at ciolib.c:281
#6  0x00005555556f45ed in initciolib (mode=5) at ciolib.c:503
#7  0x00005555555ae117 in main (argc=2, argv=0x7fffffffdf18) at syncterm.c:1716
(gdb) 

[Ragnarok]

Sorry! I forgot to mention that I using the lastest git code.

[Stephen James Hurd]

If you could grab the new changes from the last day or two and provide the backtrace for all threads (using thread apply all bt in gdb) it should make it clear what's happening. I expect it's a deadlock somewhere, so should be easily fixed once it's clear what's happening.

[Stephen James Hurd]

This was likely the same issue as #113 was (which is now fixed). Please test and let me know.

[Ragnarok]

just pull the git and test:

(gdb) thread apply all bt 

Thread 3 (Thread 0x7ffff7fcb700 (LWP 10419) "X11 Events"):
#0  0x00007ffff7cba96f in __GI___poll (fds=0x7ffff7fca7e8, nfds=1, timeout=-1) at ../sysdeps/unix/sysv/linux/poll.c:29
#1  0x00007ffff723fd02 in ?? () from /lib/x86_64-linux-gnu/libxcb.so.1
#2  0x00007ffff7240381 in xcb_writev () from /lib/x86_64-linux-gnu/libxcb.so.1
#3  0x00007ffff72a1eae in _XSend () from /lib/x86_64-linux-gnu/libX11.so
#4  0x00007ffff72a247c in _XReply () from /lib/x86_64-linux-gnu/libX11.so
#5  0x00007ffff7287fb4 in XGetWindowProperty () from /lib/x86_64-linux-gnu/libX11.so
#6  0x000055555572be03 in x11_get_maxsize (w=0x7ffff7fcac34, h=0x7ffff7fcac30) at x_events.c:242
#7  0x000055555572d221 in init_window () at x_events.c:572
#8  0x000055555572dfe6 in video_init () at x_events.c:747
#9  0x00005555557307ec in x11_event_thread (args=0x0) at x_events.c:1564
#10 0x00007ffff7da8ea7 in start_thread (arg=<optimized out>) at pthread_create.c:477
#11 0x00007ffff7cc6a2f in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:95

Thread 1 (Thread 0x7ffff7bc8740 (LWP 10417) "syncterm"):
#0  futex_wait_cancelable (private=0, expected=0, futex_word=0x55555dc81038) at ../sysdeps/nptl/futex-internal.h:186
#1  __pthread_cond_wait_common (abstime=0x0, clockid=0, mutex=0x55555dc80fe8, cond=0x55555dc81010) at pthread_cond_wait.c:508
#2  __pthread_cond_wait (cond=0x55555dc81010, mutex=0x55555dc80fe8) at pthread_cond_wait.c:638
#3  0x00005555557491ce in xp_sem_wait (sem=0x55555dc6ad60 <init_complete>) at xpsem.c:164
#4  0x0000555555714cff in x_init () at x_cio.c:583
#5  0x00005555556f3e44 in try_x_init (mode=5) at ciolib.c:285
#6  0x00005555556f461d in initciolib (mode=5) at ciolib.c:507
#7  0x00005555555ae139 in main (argc=2, argv=0x7fffffffdf18) at syncterm.c:1724
(gdb) 

[Stephen James Hurd]

That is very weird... what does xprop -root _NET_WORKAREA print on your system (from any terminal with X running)?

[Stephen James Hurd]

I think I found and fixed the issue, so it should be resolved now. I'm still curious what your _NET_WORKAREA is though.

[Ragnarok]

yes! last 2 commits fix the startup hang,

then I notice that maximize the windows (just doble click on windows titlebar) get a weird screen . But this is another issue.

[Ragnarok]

ragnarok@ragnarok:~$ xprop -root _NET_WORKAREA
_NET_WORKAREA(CARDINAL) = 0, 24, 2440, 1824, 0, 24, 2440, 1824, 0, 24, 2440, 1824, 0, 24, 2440, 1824

[Stephen James Hurd]

Thanks for getting back to me, don't forget to file a new bug... in the new bug, mention what WM you're using because you shouldn't even be able to maximize the window in X11 mode yet.