bug#10120: 24.0.91; frame-visible-p returns t, but frame not visible/iconified

["Drew Adams" <drew.adams@oracle.com>]

Probably not very useful as is - sorry about that.  Take this as an FYI,
in case you get further, related reports.
 
With my setup (not with emacs -Q), at some point I got into a state
where there was a buffer associated with a window in a frame, and that
frame was neither visible anywhere nor iconified.  It was not listed in
the Windows task bar, which lists each frame.
 
Yet Emacs thought that that the frame was visible (e.g. frame-visible-p
returned t).  The buffer existed and was live, but it was displayed in
no visible or iconified window/frame anywhere.
 
Emacs functions such as raise-frame and select-frame-set-input-focus
returned normally, as if they worked, and the latter even deselected the
current frame (unhighlighted the title bar and borders) as if it had
selected another frame.  But the target frame was not visible anywhere
(and not just off-screen, as determined by the Windows task-bar apps
list).
 
Here are some things I noted when in the debugger, in case it helps:
 
* display-buffer(#<buffer foo.el> t)
* pop-to-buffer(#<buffer foo.el> t nil)
* switch-to-buffer-other-frame(#<buffer foo.el>)
 
Then:
* display-buffer--maybe-same-window(#<buffer foo.el>
                                    ((inhibit-same-window . t)))
returned nil.
 
Then:
* get-buffer-window-list(#<buffer foo.el> nomini 0)
returned (#<window 50 on foo.el>)
 
Then:
* display-buffer-record-window(reuse #<window 50 on foo.el>
                               #<buffer foo.el>)
returned nil.
 
Then:
* window-parameter(#<window 50 on foo.el> quit-restore)
returned nil.  Note that this window was visible nowhere.
 
Then:
* frame-visible-p(#<frame foo.el 047B6E00>)
returned t, even though the frame is not visible anywhere
and is not in the list of frames in the Windows task bar.
* window-frame(#<window 50 on foo.el>)
returned #<frame foo.el 047B6E00>.
* window--display-buffer-1(#<window 50 on foo.el>)
 
Then:
* raise-frame(#<frame foo.el 047B6E00>)
returned nil.
 
Then:
* select-window(#<window 50 on foo.el>)
returned #<window 50 on foo.el>.
 
So display-buffer-reuse-window, and hence display-buffer,
returned #<window 50 on foo.el>.  And pop-to-buffer, and hence
switch-to-buffer-other-frame, returned buffer #<buffer foo.el>.
 
HTH in some way (ring a bell?).  Sorry I don't have more info.

In GNU Emacs 24.0.91.1 (i386-mingw-nt5.1.2600) of 2011-11-21 on MARVIN
 Windowing system distributor `Microsoft Corp.', version 5.1.2600
 configured using `configure --with-gcc (4.6) --no-opt --cflags
 -ID:/devel/emacs/libs/libXpm-3.5.8/include
 -ID:/devel/emacs/libs/libXpm-3.5.8/src
 -ID:/devel/emacs/libs/libpng-dev_1.4.3-1/include
 -ID:/devel/emacs/libs/zlib-dev_1.2.5-2/include
 -ID:/devel/emacs/libs/giflib-4.1.4-1/include
 -ID:/devel/emacs/libs/jpeg-6b-4/include
 -ID:/devel/emacs/libs/tiff-3.8.2-1/include
 -ID:/devel/emacs/libs/gnutls-2.10.1/include --ldflags
 -LD:/devel/emacs/libs/gnutls-2.10.1/lib'