Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!uunet!husc6!think!rlk
From: rlk@think.COM (Robert Krawitz)
Newsgroups: comp.emacs,comp.windows.x
Subject: Re: Gnu Emacs X interface under Sys V
Message-ID: <11596@think.UUCP>
Date: Tue, 10-Nov-87 13:52:46 EST
Article-I.D.: think.11596
Posted: Tue Nov 10 13:52:46 1987
Date-Received: Thu, 12-Nov-87 23:20:33 EST
Sender: usenet@think.UUCP
Reply-To: rlk@THINK.COM
Distribution: na
Organization: Thinking Machines Corporation, Cambridge, MA
Lines: 25
Xref: mnetor comp.emacs:2256 comp.windows.x:1934

In article <1275@mipos3.intel.com> pinkas@cadev4.intel.com writes:
]I recently brought up Gnu Emacs on a Sys V.3 box running X10R4.  I wanted
]to have the emacs X support, so I defined the proper constants in the
]appriopriate files and so on.  The compiler complained when compiling
]xterm.c and xfns.c.  The complaints were about undefined constants and
]macros that are defined on our BSD systems.

The things that are undefined are probably related to SIGIO, which
emacs uses internally to make C-g "interrupts" work under X.  Sigio is
a berkeley mechanism for asynchronous, interrupt-driven I/O which I'm
told has no equivalent in SysV.

I have been warned about this, and I'm aware that it causes problems
for SysV users, but I have no good idea about how to fix it.  You can
take out all the references to SIGIO and it might work, but C-g
certainly won't and you'll probably have redisplay problems.  The
places where emacs actually sends itself a SIGIO you might be able to
simulate by calling the SIGIO handler directly, but I haven't ever
tried this.

cca >>>>>>>>>>  |
harvard >>>>>>  |
bloom-beacon >  |think!rlk	Robert Krawitz <rlk@think.com>
rutgers >>>>>>  |
ihnp4 >>>>>>>>  .