recent security updates to 10.3-RELEASE broke xdm?

[Anton Shterenlikht]

>From freebsd at edvax.de Sun Oct  2 15:53:23 2016
>
>> Anyway, this is what I get:
>> 
>> ####> cat .xsession
>> #!/bin/tcsh
>> source ~/.tcshrc
>> exec ~/.xinitrc
>> ####> cat .xinitrc
>> xautolock -locker slock &
>> xconsole &
>> exec mwm
>
>Yes: This should let the session live as long as mwm lives.
>You can now easily verify this with the "non-xdm approach"
>by using "startx" - you should get your regular desktop, and
>when you exit mwm, the session terminates, and X shuts down.
>
>This is what you should definitely verify, to make sure the
>"session flow" works as expected.

Yes, this works exactly as you describe.

>> xdm info (pid 917): Starting X server on :0                                     
>> xdm info (pid 4040): sourcing /usr/local/lib/X11/xdm/Xsetup_0                   
>> xdm info (pid 4040): sourcing /usr/local/lib/X11/xdm/GiveConsole                
>> xdm info (pid 4057): executing session /usr/local/lib/X11/xdm/Xsession          
>> xdm info (pid 4040): sourcing /usr/local/lib/X11/xdm/TakeConsole                
>> XIO:  fatal IO error 35 (Resource temporarily unavailable) on X server ":0"     
>>       after 169 requests (162 known processed) with 0 events remaining.
>
>This error also looks familiar to me. Do you have any suspicious
>entries in the /var/log/Xorg.0.log file?

No, all seems fine:
http://eis.bris.ac.uk/~mexas/Xorg.0.log

>https://lists.freebsd.org/pipermail/freebsd-questions/2011-March/228663.html
>
>No solution there, just a reference to the X11 list...

Yes, not much help.

Perhaps my nvidia driver (and the card) are becoming too old:

nvidia-driver-340-340.93       x11/nvidia-driver-340

Thanks for all your help!

Anton