Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site lsuc.UUCP
Path: utzoo!lsuc!dave
From: dave@lsuc.UUCP (David Sherman)
Newsgroups: net.bugs.v7,pe.cust.bugs
Subject: suggested schedule change to v7
Message-ID: <215@lsuc.UUCP>
Date: Mon, 17-Dec-84 09:56:20 EST
Article-I.D.: lsuc.215
Posted: Mon Dec 17 09:56:20 1984
Date-Received: Mon, 17-Dec-84 10:42:41 EST
Distribution: net
Organization: Law Society of Upper Canada, Toronto
Lines: 28

We have a Perkin-Elmer 3220 which is running "Edition VII
Workbench", which is basically v7.

The P-E service people here in Toronto were looking at scheduler
problems for another customer. It seemed that when many processes
were running, the scheduler would lock some out for quite a long time.
They found that the following change solved the problem:

in slp.c, routine setrun():

	if(p->p_pri < curpri)
		runrun++;

changing the test from < to <=, so it reads
	if(p->p_pri <= curpri)

This code goes back to v7 (it's identical to the version we had
on our 11/23 last year). Does anyone know whether this is a correct
fix? Are there any problems likely to result from making this change?
I think I understand the rationale behind it, but I don't know enough
about the v7 scheduler to be sure.

Dave Sherman
The Law Society of Upper Canada
Toronto
-- 
{utzoo pesnta nrcaero utcs}!lsuc!dave
{allegra decvax ihnp4 linus}!utcsrgv!lsuc!dave