After upgrading to Tumbleweed 20220415 and emacs-28.1-393.10.x86_64,
when I open xterm and unset DISPLAY, emacs almost always crashes.
Even with -Q.  Sometimes I can get something that looks like a
potentially useful backtrace in gdb, sometimes not.  The potentially
useful one is:

  Thread 1 "emacs-gtk" received signal SIGSEGV, Segmentation fault.
  (gdb) bt
  #0  0x00000000004707a9 in  ()
  #1  0x00000000004748d3 in update_frame ()
  #2  0x00000000004a0fd5 in  ()
  #3  0x00000000004976a9 in message3_nolog ()
  #4  0x0000000000497978 in message3 ()
  #5  0x00000000005df970 in Fmessage ()
  #6  0x00000000005e3d81 in funcall_subr ()
  #7  0x00000000005e2723 in Ffuncall ()
  #8  0x0000000000629c70 in exec_byte_code ()
  #9  0x00000000005e22e8 in  ()
  #10 0x00000000005e25bd in Ffuncall ()
  #11 0x0000000000611308 in Fload ()
  #12 0x00000000005e3e76 in funcall_subr ()
  #13 0x00000000005e2723 in Ffuncall ()
  #14 0x0000000000629c70 in exec_byte_code ()
  #15 0x00000000005e22e8 in  ()
  #16 0x00000000005e7e31 in  ()
  #17 0x00000000005e8161 in eval_sub ()
  #18 0x00000000006137af in  ()
  #19 0x000000000061600d in  ()
  #20 0x0000000000616371 in Feval_buffer ()
  #21 0x00000000005e3e76 in funcall_subr ()
  #22 0x00000000005e2723 in Ffuncall ()
  #23 0x0000000000629c70 in exec_byte_code ()
  #24 0x00000000005e22e8 in  ()
  #25 0x00000000005e25bd in Ffuncall ()
  #26 0x0000000000611308 in Fload ()
  #27 0x00000000005e3e76 in funcall_subr ()
  #28 0x00000000005e2723 in Ffuncall ()
  #29 0x0000000000629c70 in exec_byte_code ()
  #30 0x00000000005e22e8 in  ()
  #31 0x00000000005e25bd in Ffuncall ()
  #32 0x0000000000629c70 in exec_byte_code ()
  #33 0x00000000005e22e8 in  ()
  #34 0x00000000005e25bd in Ffuncall ()
  #35 0x0000000000629c70 in exec_byte_code ()
  #36 0x00000000005e22e8 in  ()
  #37 0x00000000005e7e31 in  ()
  #38 0x00000000005e8161 in eval_sub ()
  #39 0x000000000055f346 in  ()
  #40 0x00000000005e0cf7 in internal_condition_case ()
  #41 0x000000000055f7b2 in  ()
  #42 0x00000000005e0c51 in internal_catch ()
  #43 0x000000000055fb5b in  ()
  #44 0x000000000055fc5f in recursive_edit_1 ()
  #45 0x000000000055fde3 in Frecursive_edit ()
  #46 0x000000000046ae2c in main ()
  (gdb) info threads
  Id   Target Id                                     Frame 
* 1    Thread 0x7fffec13b480 (LWP 18954) "emacs-gtk" 0x00000000004707a9 in ??
    ()
  2    Thread 0x7fffeb347640 (LWP 18960) "gmain"     0x00007fffef35752f in
__GI___poll (fds=0xb48020, nfds=1, timeout=-1)
    at ../sysdeps/unix/sysv/linux/poll.c:29
  (gdb) thread 2
  [Switching to thread 2 (Thread 0x7fffeb347640 (LWP 18960))]
  #0  0x00007fffef35752f in __GI___poll (fds=0xb48020, nfds=1, timeout=-1) at
../sysdeps/unix/sysv/linux/poll.c:29
  29      return SYSCALL_CANCEL (poll, fds, nfds, timeout);
  (gdb) bt
  #0  0x00007fffef35752f in __GI___poll (fds=0xb48020, nfds=1, timeout=-1) at
../sysdeps/unix/sysv/linux/poll.c:29
  #1  0x00007ffff6ef144e in  () at /lib64/libglib-2.0.so.0
  #2  0x00007ffff6ef156f in g_main_context_iteration () at
/lib64/libglib-2.0.so.0
  #3  0x00007ffff6ef15c1 in  () at /lib64/libglib-2.0.so.0
  #4  0x00007ffff6f1bac5 in  () at /lib64/libglib-2.0.so.0
  #5  0x00007fffef2db2ba in start_thread (arg=<optimized out>) at
pthread_create.c:442
  #6  0x00007fffef365460 in clone3 () at
../sysdeps/unix/sysv/linux/x86_64/clone3.S:81

Emacs seems to run fine in GNU screen (so this is my workaround) or in
a virtual terminal.

emacs-x11 and emacs-nox are also affected, except that both seem not
crash under gdb (at least not now) and after that one successful start
they also work in the same xterm window.  But not in new ones.