ediff-revision miss-displays when 1<frames are present

[Daniel Ortmann]

This bug report will be sent to the Free Software Foundation,
 not to your local site managers!!
Please write in English, because the Emacs maintainers do not have
translators to read other languages for them.

In GNU Emacs 20.7.1 (rs6000-ibm-aix4.3.3.0, Motif)
 of Mon Nov 20 2000 on neon.rchland.ibm.com
configured using `configure
--prefix=/afs/rch/rel/common/prod/ct/local/200011/rs_aix43 --with-x
--with-x-toolkit=motif'

Please describe exactly what actions triggered the bug
and the precise symptoms of the bug:

-- I had 3 frames open, each in its own fvwm2 virtual display on the same
   machine, with the current one editting an RCS-controlled file.

-- I invoked ediff-revision via a key binding.

-- But only one file appeared in my frame!  (Plus the small control
window.)

-- I checked my other fvwm2 virtual displays and found the missing buffer
   displayed in one of the other frames.

-- I performed bury-buffer and returned to the original frame and pressed
'!',
   with the result that the small control window jumped to the bottom of
the
   display.

-- I exited the diff process with 'q' and restarted the process in several
   ways.  Each time something strange happened.

-- Finally, I had to *remove* the other frames in order to perform
   ediff-revision correctly.

-----------------------

I suspect that the design intent of the ediff-revision code is:

    *If* a user has multiple frames open, *then* prefer using two frames
(plus
    the control frame) for displaying the buffers (else display them in
    windows).

Please consider modifying this to:

    *If* a user has multiple frames open *and* at least two of them are in
the
    same virtual display, *then* prefer using two frames (plus the control)
    for displaying the buffers (else display them in windows).

Thank you for you consideration.

Recent input:
C-n C-n C-n C-n C-n C-n C-p return C-n C-n C-n M-f
M-f M-f down-mouse-1 mouse-1 down-mouse-1 mouse-1 down-mouse-1
mouse-1 C-x C-b switch-frame C-u C-x C-b C-n C-n return
M-> S-f11 return return n n switch-frame f12 switch-frame
! q y switch-frame C-x 2 C-x o f12 C-u C-x C-b C-n
C-n return M-< f11 return return n n q y switch-frame
f11 return return n n C-g q y C-x 5 0 C-x 5 0 C-x k
return S-f11 return return n n q y C-x 1 M-x r e p
o r t - e m tab return

Recent messages:
Note: file is write protected
Computing differences between and2_bp_p.ap.~1.7~ and and2_bp_p.ap ...

Buffer A: Processing difference region 0 of 7
Buffer B: Processing difference region 0 of 7
Processing difference regions ... done
Refining difference region 1 ...
 [2 times]
Refining difference region 2 ...
 [4 times]


--
Daniel Ortmann, IBM Circuit Technology, Rochester, MN 55901-7829
ortmann@us.ibm.com / internal 8.553.6795 / external 507.253.6795
ortmann@isl.net home 507.288.7732

"The answers are so simple, and we all know where to look,
but it's easier just to avoid the question." -- Kansas