bug#60356: 30.0.50; individual frame loses keyboard focus, focus cannot be restored

[Po Lu via "Bug reports for GNU Emacs, the Swiss army knife of text editors" <bug-gnu-emacs@gnu.org>]

git@sphalerite.org writes:

> The exact action leading to this bug is not clear to me. It seems to be
> related to lsp-mode though -- I've not observed it happening without
> lsp-mode.
>
> The symptom is that a single frame will no longer accept keyboard input,
> and will visually indicate that it is not focused (outline rectangular
> cursor where I'd usually expect a solid rectangular cursor). Mouse input
> still works, and other frames still accept keyboard input as normal.
>
> I'd be glad if anyone could point out ways to further debug this issue.
> I'm also in #emacs:matrix.org (@linus:schreibt.jetzt) in case
> fast-turnaround communication would be helpful :)

Can you reliably reproduce this issue?

If you can, please add some instrumentation to x_new_focus_frame, like
this:

diff --git a/src/xterm.c b/src/xterm.c
index 1eef8e7a724..cda2897e013 100644
--- a/src/xterm.c
+++ b/src/xterm.c
@@ -11695,6 +11695,9 @@ x_new_focus_frame (struct x_display_info *dpyinfo, struct frame *frame)
     }
 #endif
 
+  fprintf (stderr, "x_new_focus_frame: switching focus to"
+	   " %s\n", SSDATA (frame->name));
+
   if (frame != dpyinfo->x_focus_frame)
     {
       /* Set this before calling other routines, so that they see

and see what is printed out when you try to focus the frame that does
not accept input.  Thanks.