home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.os.linux
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!rpi!ghost.dsi.unimi.it!univ-lyon1.fr!chx400!bernina!almesber
- From: almesber@nessie.cs.id.ethz.ch (Werner Almesberger)
- Subject: Re: Serial communications problems
- Message-ID: <1992Dec13.043423.5075@bernina.ethz.ch>
- Sender: news@bernina.ethz.ch (USENET News System)
- Organization: Swiss Federal Institute of Technology (ETH), Zurich, CH
- References: <1992Dec13.021248.3074@athena.mit.edu>
- Date: Sun, 13 Dec 1992 04:34:23 GMT
- Lines: 21
-
- In article <1992Dec13.021248.3074@athena.mit.edu> tytso@ATHENA.MIT.EDU (Theodore Ts'o) writes:
- > It would be possible to write a serial driver that did not use
- > interrupts --- by hooking something to the timer interrupt and having it
- > check 100 times a second to see if there's anything on the serial port;
- > but it would ruin your system performance, and it wouldn't really work
- > unless you had a 16550A UART, or if you aren't doing anything faster
- > than 600 bps. (And even if you had a 16550A, you wouldn't be able to do
- > anything faster than 9600.)
-
- You're a bit too pessimistic here. I'm using such a driver to run all
- four COM ports and it does 2400 bps with a 16450 and 19200 bps with a
- 16550A. 19200 bps gets a bit unreliable under load, but it's still
- acceptable for Zmodem. I'm polling at 300 Hz and it doesn't lower
- system performance very noticeably.
-
- - Werner
- --
- _________________________________________________________________________
- / Werner Almesberger, ETH Zuerich, CH almesber@nessie.cs.id.ethz.ch /
- / IFW A44 Tel. +41 1 254 7213 almesberger@rzvax.ethz.ch /
- /_BITNET:_ALMESBER@CZHETH5A__HEPNET/CHADNET:_[20579::]57414::ALMESBERGER_/
-