[gnu.emacs] GnuEmacs 18.53 on Sequent S27

nrouquet@pollux.usc.edu (Nicolas Rouquette) (07/25/89)

Hello, I compiled GnuEmacs 18.53 on a Sequent Symmetry S27 running
Dynix 3.0.12 with Dynix' and the m-sequent.h file as is. Emacs seems
to run fine except that there is a nasty interaction with flow of
control. When I run Emacs and press ^l two times quickly, some garbage
appears on the screen and there is a fatal error (11) and core dumped.
Some other times, when I load a file from the comand line argument, it
looks as if the Seqent sent some ^S^Q sequence that Emacs grabbed and
interpreted as: I-Search ^Q. Also, pressing ^S two times consecutively
crashes Emacs no matter what else I did. 

Any suggestions appreciated.

Nick.

rjk@sawmill.uucp (Richard Kuhns) (07/27/89)

Several people have asked about getting GNU Emacs running (properly!)
on a Symmetry.  My experience (I've had everything from 18.41 to 18.54
running on an S27) indicates the following:  if you use the standard C
compiler, DON'T optimize.  I used m-symmetry.h -- if this isn't part
of the standard distribution, let me know and I'll mail you a copy.
Back in the 18.4? days it wasn't -- I put one together myself and I'm
afraid I haven't kept track in more recent distributions.

If you use gcc (1.35, anyhow), you can compile GNU emacs with
optimization, and it apparently works just fine (don't forget the
-traditional switch!) -- that's how I've compiled the last couple of
revisions, and I haven't had any problems.  You'll also get a smaller
binary.

I haven't bothered trying atscc yet -- I'm well satisfied with gcc and
see no reason to change, since I don't have much use (yet) for the
parallel processing capabilities Sequent's C compilers provide.

Rich Kuhns
the_known_world!newton.physics.purdue.edu!sawmill!rjk

PS.  I just tried to compile the recent omega distribution from
comp.sources.games using atscc -- the load failed with `undefined
symbol L1136'.  gcc (with -traditional) compiled it with nary a
complaint.  I haven't bothered to try the standard cc.

rich@sendai.sendai.ann-arbor.mi.us (K. Richard Magill) (07/30/89)

In article <RJK.89Jul27095511@sawmill.uucp> rjk@sawmill.uucp (Richard Kuhns) writes:

   Several people have asked about getting GNU Emacs running (properly!)
   on a Symmetry.

I haven't done it all on all combinations, but I've done emacs
18.48-present on dynix 2.x - present.

I've had luck with cc -O, cc, and gcc (use -traditional.  otherwise,
it compiles, but doesn't run.) in most forms.  Use m-sequent.h, and
s-bsd4-3.h.  The only trick is that under X you need to put a
"(set-input-mode nil nil)" in a default.el for your system.

   I haven't bothered trying atscc yet -- I'm well satisfied with gcc
   and see no reason to change, since I don't have much use (yet) for
   the parallel processing capabilities Sequent's C compilers provide.

same here.  Actually, I need similar facilities but all of sequent's
pplib require related processes.
--
rich.