bug#31312: Segmentation fault with doom-emacs, NeoTree and Zoom

[martin rudalics]

> Reproduced with emacs-26, seems there is a window without a buffer.
>
> ../../src/buffer.h:914: Emacs fatal error: assertion failed: BUFFERP (a)
>
> Thread 1 "emacs" hit Breakpoint 1, terminate_due_to_signal (sig=6, 
backtrace_limit=2147483647)
>      at ../../src/emacs.c:364
> 364          signal (sig, SIG_DFL);
> (gdb) bt
> #0  terminate_due_to_signal (sig=6, backtrace_limit=2147483647) at 
../../src/emacs.c:364
> #1  0x0000000000614655 in die (msg=0x74ecf9 "BUFFERP (a)", file=0x74ece6 
"../../src/buffer.h",
>      line=914) at ../../src/alloc.c:7423
> #2  0x000000000057a1a8 in XBUFFER (a=XIL(0)) at ../../src/buffer.h:914
> #3  0x0000000000464716 in reconsider_clip_changes (w=0x60537f0) at 
../../src/xdisp.c:13727
> #4  0x000000000046514d in redisplay_internal () at ../../src/xdisp.c:13939
> #5  0x0000000000463c16 in redisplay () at ../../src/xdisp.c:13518
> #6  0x00000000005851ae in read_char (commandflag=1, map=XIL(0x5cf7503), 
prev_event=XIL(0),
>      used_mouse_menu=0x7fffffffe41f, end_time=0x0) at 
../../src/keyboard.c:2480
> #7  0x00000000005957c5 in read_key_sequence (keybuf=0x7fffffffe570, 
bufsize=30, prompt=XIL(0),
>      dont_downcase_last=false, can_return_switch_frame=true, 
fix_current_buffer=true,
>      prevent_redisplay=false) at ../../src/keyboard.c:9147
> #8  0x0000000000581d55 in command_loop_1 () at ../../src/keyboard.c:1368
> #9  0x0000000000638529 in internal_condition_case (bfun=0x581909 
<command_loop_1>,
>      handlers=XIL(0x5250), hfun=0x580f4c <cmd_error>) at ../../src/eval.c:1332
> #10 0x0000000000581524 in command_loop_2 (ignore=XIL(0)) at 
../../src/keyboard.c:1110
> #11 0x0000000000637a32 in internal_catch (tag=XIL(0xc6f0), func=0x5814f7 
<command_loop_2>, arg=XIL(0))
>      at ../../src/eval.c:1097
> #12 0x00000000005814c2 in command_loop () at ../../src/keyboard.c:1089
> #13 0x0000000000580a36 in recursive_edit_1 () at ../../src/keyboard.c:695
> #14 0x0000000000580c2b in Frecursive_edit () at ../../src/keyboard.c:766
> #15 0x000000000057e7c6 in main (argc=1, argv=0x7fffffffea38) at 
../../src/emacs.c:1713
> (gdb) frame 3

Is w in reconsider_clip_changes the same as XWINDOW (selected_window)
when it fails?  If so, could you instrument select_window_1 (or use a
watchpoint) to check whether selected_window = window ever assigns a
window with XWINDOW (window)->contents equal Qnil?

Thanks, martin