bug#59480: 29.0.50; emacs master with --enable-checking fails assert in x_display_set_last_user_time

bug#59480: 29.0.50; emacs master with --enable-checking fails assert in x_display_set_last_user_time

[Robert Pluim]

This is master as of right now, configured with:

./configure --enable-checking=yes,glyphs

src/emacs -Q

xterm.c:7753: Emacs fatal error: assertion failed: time <= X_ULONG_MAX

Thread 1 "emacs" hit Breakpoint 1, terminate_due_to_signal (sig=6, backtrace_limit=2147483647) at emacs.c:426
426       signal (sig, SIG_DFL);
(gdb) bt
#0  terminate_due_to_signal (sig=6, backtrace_limit=2147483647) at emacs.c:426
#1  0x00005555555b3320 in die (msg=<optimized out>, file=<optimized out>, line=<optimized out>) at alloc.c:7697
#2  0x00005555556d4d47 in x_display_set_last_user_time (dpyinfo=0x5555562ec000, time=18446744071827324120, send_event=true, set_property=true) at xterm.c:7753
#3  0x00005555556ea241 in handle_one_xevent (dpyinfo=0x5555562ec000, event=0x7fffffffd490, finish=0x555555eebd24 <current_finish>, hold_quit=0x7fffffffd730) at xterm.c:18624
#4  0x00005555556e77ac in event_handler_gdk (gxev=0x7fffffffd490, ev=0x5555560a33c0, data=0x0) at xterm.c:17412

(gdb) up
#6  0x00005555556d4d47 in x_display_set_last_user_time (dpyinfo=0x5555562ec000, time=18446744071827324120, send_event=true, set_property=true) at xterm.c:7753

7753      eassert (time <= X_ULONG_MAX);
(gdb) up
#7  0x00005555556ea241 in handle_one_xevent (dpyinfo=0x5555562ec000, event=0x7fffffffd490, finish=0x555555eebd24 <current_finish>, hold_quit=0x7fffffffd730) at xterm.c:18624
18624                   x_display_set_last_user_time (dpyinfo, event->xclient.data.l[1],
(gdb) p X_ULONG_MAX
$1 = 4294967295

X_ULONG_MAX looks very 32bit-ish :-)

(why this started happening just now I donʼt know, that assert has
been in for ages)

Robert
-- 

bug#59480: 29.0.50; emacs master with --enable-checking fails assert in x_display_set_last_user_time

[Po Lu via Bug reports for GNU Emacs, the Swiss army knife of text editors]

Robert Pluim <rpluim@gmail.com> writes:

> This is master as of right now, configured with:
>
> ./configure --enable-checking=yes,glyphs
>
> src/emacs -Q
>
> xterm.c:7753: Emacs fatal error: assertion failed: time <= X_ULONG_MAX
>
> Thread 1 "emacs" hit Breakpoint 1, terminate_due_to_signal (sig=6, backtrace_limit=2147483647) at emacs.c:426
> 426       signal (sig, SIG_DFL);
> (gdb) bt
> #0  terminate_due_to_signal (sig=6, backtrace_limit=2147483647) at emacs.c:426
> #1  0x00005555555b3320 in die (msg=<optimized out>, file=<optimized out>, line=<optimized out>) at alloc.c:7697
> #2  0x00005555556d4d47 in x_display_set_last_user_time (dpyinfo=0x5555562ec000, time=18446744071827324120, send_event=true, set_property=true) at xterm.c:7753
> #3  0x00005555556ea241 in handle_one_xevent (dpyinfo=0x5555562ec000, event=0x7fffffffd490, finish=0x555555eebd24 <current_finish>, hold_quit=0x7fffffffd730) at xterm.c:18624
> #4  0x00005555556e77ac in event_handler_gdk (gxev=0x7fffffffd490, ev=0x5555560a33c0, data=0x0) at xterm.c:17412
>
> (gdb) up
> #6  0x00005555556d4d47 in x_display_set_last_user_time (dpyinfo=0x5555562ec000, time=18446744071827324120, send_event=true, set_property=true) at xterm.c:7753
>
> 7753      eassert (time <= X_ULONG_MAX);
> (gdb) up
> #7  0x00005555556ea241 in handle_one_xevent (dpyinfo=0x5555562ec000, event=0x7fffffffd490, finish=0x555555eebd24 <current_finish>, hold_quit=0x7fffffffd730) at xterm.c:18624
> 18624                   x_display_set_last_user_time (dpyinfo, event->xclient.data.l[1],
> (gdb) p X_ULONG_MAX
> $1 = 4294967295
>
> X_ULONG_MAX looks very 32bit-ish :-)
>
> (why this started happening just now I donʼt know, that assert has
> been in for ages)
>
> Robert

Thanks, should be fixed now.

bug#59480: 29.0.50; emacs master with --enable-checking fails assert in x_display_set_last_user_time

[Robert Pluim]

tags 59480 fixed
close 59480 29.1
quit

>>>>> On Tue, 22 Nov 2022 20:56:10 +0800, Po Lu via "Bug reports for GNU Emacs, the Swiss army knife of text editors" <bug-gnu-emacs@gnu.org> said:
    >> X_ULONG_MAX looks very 32bit-ish :-)
    >> 
    >> (why this started happening just now I donʼt know, that assert has
    >> been in for ages)
    >> 
    >> Robert

    Po Lu> Thanks, should be fixed now.

Confirmed. Closing

Robert
--