From owner-freebsd-current  Sun Aug 10 10:43:47 1997
Return-Path: <owner-freebsd-current>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.5/8.8.5) id KAA02220
          for current-outgoing; Sun, 10 Aug 1997 10:43:47 -0700 (PDT)
Received: from Ilsa.StevesCafe.com (Ilsa.StevesCafe.com [205.168.119.129])
          by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id KAA02214
          for <current@FreeBSD.ORG>; Sun, 10 Aug 1997 10:43:43 -0700 (PDT)
Received: from Ilsa.StevesCafe.com (localhost [127.0.0.1])
	by Ilsa.StevesCafe.com (8.8.6/8.8.5) with ESMTP id LAA11578;
	Sun, 10 Aug 1997 11:43:37 -0600 (MDT)
Message-Id: <199708101743.LAA11578@Ilsa.StevesCafe.com>
X-Mailer: exmh version 2.0gamma 1/27/96
From: Steve Passe <smp@csn.net>
To: nnd@itfs.nsk.su
cc: current@FreeBSD.ORG
Subject: Re: Make and SMP - what can be done ? 
In-reply-to: Your message of "10 Aug 1997 11:30:41 GMT."
             <5sk8p1$ld3@news.itfs.nsk.su> 
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date: Sun, 10 Aug 1997 11:43:37 -0600
Sender: owner-freebsd-current@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

Hi,

> 	Now when we can use SMP-FreeBSD and there is
> a possibility to speedup building various components of
> the system (making the kernel with 'make -j12' proves this
> very definitely) there is a question - HOW can we use it ?

You've touched on a topic dear to my heart!  I took a stab at this once
and gave up for lack of knowledge of the .mk system and its subltlies.

There is a BIG win to be had here, my experiments suggest "make world"
could be cut by around 25% if the -j option were added properly, maybe more.

---
> 	More stepwise process may consists in adding global
> make flag (say, JFLAG which is set to "-j12" for SMP case
> and to "" for traditional systems) and using it in various
> parts of word-building alongside with modifying appropriate
> Makefiles.

I added JMAKEFLAGS= -j12 to /etc/make.conf, then added JMAKEFLAGS
to some of the lower level Makefiles, like the ones for libraries.
This worked nicely.  I found other Makefiles that just needed a little
more work on the dependancies, they sometimes need to be more explicit
in a parallel world. Things with YACC & LEX passes die horribly.

for non SMP kernels the -j option still is a big win.  It filters out the
time spent spinning while waiting for disk IO, try -j4 when building a kernel
on a UP system.

---
> 	Is there any sense in my speculations ?
> Does anybody have some propositions for solving this problem ?
> And is there REAL problem at all ;-) ?

Bottom line is that this is something that REALLY needs doing.  You seem to
have a good handle on the concept, go for it!!!

--
Steve Passe	| powered by 
smp@csn.net	|            Symmetric MultiProcessor FreeBSD