🚀 go-pugleaf

  Hello,

  At the risk of repeating an old report (but if so I've not been
  able to find it):

  at our RS/6000 installation, we are having continual occurrences
  of the following pattern:
  - run emacs on a file
  - type any key -- echoing, arrow, control -- doesn't matter
  - in the bottom-row window, the message "Fatal error(23)." appears,
    followed by the echo of the character (on that row).
  - a number of possibilities are observed from that point:
    ^C will cause "Quit." to appear, and the cursor returns to the 
    editing window.  Any further typing appears there, but does not
    appear to go to emacs until <RETURN> is pressed.
    ^S appears to cause the incremental search to be started, but still
    requires that <RETURN> be pressed before the input is given to 
    emacs.  (This also appears true if ^S is the first character typed).
  - ^Z in either window causes emacs to suspend.  It can be killed at
    this point.  However, since emacs is documented as running in "raw"
    terminal mode, I'm not sure this is the  correct behaviour for ^Z.

  However, at another time, emacs will run just as expected (at least to
  the degree to which we can reasonably evaluate it).

  I regret I do not have printed documentation available for consultation.
  Nor do the FAQ or bugs lists I've seen mention fatal error 23.

  I have checked emacs' executable (using grep) and the files in the
  directories under /usr/local/lib/emacs/$VERSION/ -- the message "Fatal
  error" does not appear in them.

  If desired, I'll report the version of AI/X 6000 and tcsh we are 
  running under.

  This problem has been plaguing us for several weeks now, and is 
  interfering with the organisation of work, since when emacs fails
  on one machine, we have to resort to another -- and this site does
  not appear to be running NFS.


  Any assistance will be much appreciated.


  Thank you,

  Alastair Milne