Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10 5/3/83; site kpno.UUCP
Path: utzoo!linus!philabs!seismo!hao!kpno!downey
From: downey@kpno.UUCP
Newsgroups: net.unix-wizards
Subject: 750/4.1 interrupt overhead
Message-ID: <207@kpno.UUCP>
Date: Tue, 28-Jun-83 12:00:47 EDT
Article-I.D.: kpno.207
Posted: Tue Jun 28 12:00:47 1983
Date-Received: Wed, 29-Jun-83 05:31:57 EDT
Organization: Kitt Peak Natl. Obs.  Tucson, AZ
Lines: 13

Anyone know roughly how long the interrupt latency is on a 750 running 4.1?
What I mean is, how long does it take once a unibus interrupt hits before
the driver's interrupt function gets going? Assume any conditions you
want, so long as you state what they are. An ideal "quiet" system is ok.

We are thinking about writing a driver for a Dicomed image recorder but
at two interrupts per point (one to send a random x, one for y; yech!)
I'm worried about all those interrupts.

Thanks for any info or pointers,
Elwood Downey, Kitt Peak National Observatory, 950 N. Cherry Ave, Tucson, AZ
               (602) 325-9221
               {{{ucbvax|utah-cs|purdue|mcnc}!arizona}|unc|hao}!kpno!downey