home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!auspex-gw!guy
- From: guy@Auspex.COM (Guy Harris)
- Newsgroups: comp.unix.programmer
- Subject: Re: Problem with Pseudo TTY of SunOS 4.1.1 for Sun3
- Message-ID: <14513@auspex-gw.auspex.com>
- Date: 9 Sep 92 06:44:26 GMT
- References: <1992Sep9.010847.1104@pasteur.Berkeley.EDU>
- Sender: news@auspex-gw.auspex.com
- Distribution: usa
- Organization: Auspex Systems, Santa Clara
- Lines: 18
- Nntp-Posting-Host: auspex.auspex.com
-
- > Is this a bug with SunOS,
-
- No, it's a botch in the BSD pseudo-tty implementation; the BSD one let
- you do certain tty "ioctl"s on the master side. It shouldn't have done
- so; pseudo-tty masters are *NOT* ttys, they're pseudo-tty masters.
-
- The SunOS 4.x pseudo-tty implementation does some old BSD "ioctl"s for
- backwards compatibility - and the code to do so is *REALLY GROSS*, as
- the STREAMS-based pty implementation doesn't make it at *all* easy to do
- that - but it doesn't do the newer "ioctl"s used by "isatty()" and by
- the standard I/O library to determine if something is a tty.
-
- >or do I need to do something special
- >in making the master pty a tty as well?
-
- It's *not* a tty. Explicitly use "setlinebuf()", or "setbuf()", or
- "setvbuf()", to control the standard I/O buffering on pseudo-tty
- masters.
-