You mentioned on IRC that a "radeon_cs:0 process" prevents the system from rebooting. I assume that's the radeonsi command stream submission thread of the Xorg process being stuck somewhere in the kernel due to the GPU hang. If you try rebooting and wait for a few minutes, do backtraces of the stuck task show up in dmesg? If so, please attach those here. (It may be possible to get backtraces without waiting via some Sysrq key combination)

While bisecting an unrelated kernel bug I found that some kernels between 4.9.0 and 4.10-rc1 allowed:

service lxdm stop

to work as expected and cleanly stop the X server.

Otherwise either a kill -9 of the lxdm and Xorg process or a reboot is needed to start the X server again.

If it would be helpful I could try bisecting to find the commit that stopped X shutting down cleanly.

Created attachment 128698 [details]
backtrace of Xorg after attempting service lxdm stop

# ps -ef|grep lxdm
root      7524     1  0 13:52 ?        00:00:00 /usr/sbin/lxdm-binary -d
root      7528  7524  0 13:52 tty7     00:00:00 /usr/lib/xorg/Xorg :0 vt07 -nolisten tcp -novtswitch -auth /var/run/lxdm/lxdm-:0.auth
root      7588  5817  0 13:52 pts/0    00:00:00 grep lxdm
am64:/home/amarsh04# gdb -p 7528
GNU gdb (Debian 7.12-5) 7.12
Copyright (C) 2016 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.  Type "show copying"
and "show warranty" for details.
This GDB was configured as "x86_64-linux-gnu".
Type "show configuration" for configuration details.
For bug reporting instructions, please see:
<http://www.gnu.org/software/gdb/bugs/>.
Find the GDB manual and other documentation resources online at:
<http://www.gnu.org/software/gdb/documentation/>.
For help, type "help".
Type "apropos word" to search for commands related to "word".
Attaching to process 7528
[New LWP 7550]
[New LWP 7553]
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
0x00007fc610ece67d in pthread_join (threadid=140488490870528,
    thread_return=0x0) at pthread_join.c:90
90      pthread_join.c: No such file or directory.
(gdb) thread apply all bt full

Thread 3 (Thread 0x7fc60697e700 (LWP 7553)):
#0  __lll_lock_wait () at ../sysdeps/unix/sysv/linux/x86_64/lowlevellock.S:135
No locals.
#1  0x00007fc610ecfbf6 in __GI___pthread_mutex_lock (mutex=0x556029709d20)
    at ../nptl/pthread_mutex_lock.c:115
        id = 283991900
        __PRETTY_FUNCTION__ = "__pthread_mutex_lock"
        type = 1
        id = <optimized out>
#2  0x00005560294752a0 in input_lock ()
No symbol table info available.
#3  0x00005560294753f5 in ?? ()
No symbol table info available.
#4  0x00007fc610ecd424 in start_thread (arg=0x7fc60697e700)
    at pthread_create.c:333
        __res = <optimized out>
        pd = 0x7fc60697e700
        now = <optimized out>
        unwind_buf = {cancel_jmp_buf = {{jmp_buf = {140488490870528,
                -2498737612137141548, 0, 140731836161135, 0, 140488697360448,
                2512654887706716884, 2512606964688775892},
              mask_was_saved = 0}}, priv = {pad = {0x0, 0x0, 0x0, 0x0},
            data = {prev = 0x0, cleanup = 0x0, canceltype = 0}}}
---Type <return> to continue, or q <return> to quit---
        not_first_call = <optimized out>
        pagesize_m1 = <optimized out>
        sp = <optimized out>
        freesize = <optimized out>
        __PRETTY_FUNCTION__ = "start_thread"
#5  0x00007fc610c109bf in clone ()
    at ../sysdeps/unix/sysv/linux/x86_64/clone.S:105
No locals.

Thread 2 (Thread 0x7fc6079e6700 (LWP 7550)):
#0  pthread_cond_wait@@GLIBC_2.3.2 ()
    at ../sysdeps/unix/sysv/linux/x86_64/pthread_cond_wait.S:185
No locals.
#1  0x00007fc60c03025b in ?? () from /usr/lib/x86_64-linux-gnu/dri/r600_dri.so
No symbol table info available.
#2  0x00007fc60c0300b7 in ?? () from /usr/lib/x86_64-linux-gnu/dri/r600_dri.so
No symbol table info available.
#3  0x00007fc610ecd424 in start_thread (arg=0x7fc6079e6700)
    at pthread_create.c:333
        __res = <optimized out>
        pd = 0x7fc6079e6700
        now = <optimized out>
        unwind_buf = {cancel_jmp_buf = {{jmp_buf = {140488508073728,
                -2498737612137141548, 0, 140731836158591, 0, 140488697360448,
---Type <return> to continue, or q <return> to quit---
                2512652753107970772, 2512606964688775892},
              mask_was_saved = 0}}, priv = {pad = {0x0, 0x0, 0x0, 0x0},
            data = {prev = 0x0, cleanup = 0x0, canceltype = 0}}}
        not_first_call = <optimized out>
        pagesize_m1 = <optimized out>
        sp = <optimized out>
        freesize = <optimized out>
        __PRETTY_FUNCTION__ = "start_thread"
#4  0x00007fc610c109bf in clone ()
    at ../sysdeps/unix/sysv/linux/x86_64/clone.S:105
No locals.

Thread 1 (Thread 0x7fc612dfc140 (LWP 7528)):
#0  0x00007fc610ece67d in pthread_join (threadid=140488490870528,
    thread_return=0x0) at pthread_join.c:90
        __tid = 7553
        _buffer = {__routine = 0x7fc610ece5a0 <cleanup>,
          __arg = 0x7fc60697ed28, __canceltype = 0, __prev = 0x0}
        oldtype = 0
        pd = 0x7fc60697e700
        self = 0x7fc612dfc140
        result = 0
#1  0x00005560294759b0 in ?? ()
No symbol table info available.
---Type <return> to continue, or q <return> to quit---
#2  0x0000556029313c06 in ?? ()
No symbol table info available.
#3  0x00007fc610b482b1 in __libc_start_main (main=0x5560292fd840, argc=8,
    argv=0x7ffeaf1a5e98, init=<optimized out>, fini=<optimized out>,
    rtld_fini=<optimized out>, stack_end=0x7ffeaf1a5e88)
    at ../csu/libc-start.c:291
        result = <optimized out>
        unwind_buf = {cancel_jmp_buf = {{jmp_buf = {0, 8616197459934735060,
                93871496222800, 140731836161680, 0, 0, 2499477270933564116,
                2512606207147666132}, mask_was_saved = 0}}, priv = {pad = {
              0x0, 0x0, 0x7ffeaf1a5ee0, 0x7fc612e6c168}, data = {prev = 0x0,
              cleanup = 0x0, canceltype = -1357226272}}}
        not_first_call = <optimized out>
#4  0x00005560292fd87a in _start ()
No symbol table info available.

(In reply to Arthur Marsh from comment #0)
> I experience the lockup of the video playback window running mpv (requires
> kill -HUP from another window to terminate mpv), chromium windows not
> refreshing unless I switched to vt1 and back to vt7,

Sounds like bug 99333, fixed in xserver 1.19.1, available now in Debian sid.


> and 
> 
> service lxdm restart
> 
> failing. I needed to kill -9 the Xorg process to avoid having to reboot to
> restart Xorg.

Does this still happen with the above fixed?

If yes, please attach (as opposed to paste) another full backtrace of all threads with the xserver-xorg-core-dbgsym package installed.

Created attachment 129062 [details]
debug log as requested

debug log after running "service lxdm stop".

-- GitLab Migration Automatic Message --

This bug has been migrated to freedesktop.org's GitLab instance and has been closed from further activity.

You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.freedesktop.org/xorg/xserver/issues/512.