Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 23 Aug 1996 00:54:45 -0700
From:      "Jordan K. Hubbard" <jkh@time.cdrom.com>
To:        joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch)
Cc:        freebsd-hackers@FreeBSD.org
Subject:   Re: Curio 
Message-ID:  <6321.840786885@time.cdrom.com>
In-Reply-To: Your message of "Fri, 23 Aug 1996 08:59:57 %2B0200." <199608230659.IAA29250@uriah.heep.sax.de> 

next in thread | previous in thread | raw e-mail | index | archive | help
> But that's certainly due to the ISA RAM.  My 386BSD 0.1 system on a
> 386sx16 with 6 MB RAM built the entire X11R5 tree in slightly more
> than 24 hours. ;-)

Speaking of which, I've been looking at how the tree is built during a
make world and I notice that we do a make depend pass, then we do the
make all, but by doing it in two passes this way it would seem that we
lose what would be a free buffer-cache-load from the "make depend" for
the "make all" since make depend reads through all the same files.
Doncha think?

I've tried making some simple mods to bsd.port.mk, bsd.lib.mk and
bsd.sgml.mk which fold depend into all, then removing all the explicit
make depends in /usr/src/Makefile.  I'm timing a make world on it now
to see if there's any difference. :-)

						Jordan



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?6321.840786885>