home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.unix.internals
- Path: sparky!uunet!decwrl!pa.dec.com!decprl!decprl!boyd
- From: boyd@prl.dec.com (Boyd Roberts)
- Subject: Re: Poll(), sleeping on several addresses
- Message-ID: <1992Aug21.114445.19513@prl.dec.com>
- Sender: news@prl.dec.com (USENET News System)
- Nntp-Posting-Host: prl313.prl.dec.com
- Organization: Digital Equipment Corporation - Paris Research Laboratory
- References: <1992Aug20.191904.4567@candle.uucp>
- Date: Fri, 21 Aug 1992 11:44:45 GMT
- Lines: 20
-
- In article <1992Aug20.191904.4567@candle.uucp>, root@candle.uucp (Bruce Momjian) writes:
- > My AT&T Svr3.1 poll() only works on streams devices, which doesn't do me
- > much good, so I am attempting to write my own poll() as a device driver.
- >
-
- I think you're going to lose big here. If you have the source to the
- tty drivers then you'd stand a chance. Poking around in other process's
- u-areas and other driver's data structures is just not a good idea.
-
- Perhaps you should just move the whole thing out into user mode?
- Have a process polling the various tty's and get it to send up
- that data to your other process over a socket which you can select on.
-
- Solving your problems by kernel hacking is not always a _suitable_ solution.
-
-
- Boyd Roberts boyd@prl.dec.com
-
- ``When the going gets wierd, the weird turn pro...''
-
-