Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 8 Oct 1996 11:15:03 +0100 (BST)
From:      Doug Rabson <dfr@render.com>
To:        dyson@freebsd.org
Cc:        Guido van Rooij <guido@gvr.win.tue.nl>, dyson@freefall.freebsd.org, CVS-committers@freefall.freebsd.org, cvs-all@freefall.freebsd.org, cvs-sys@freefall.freebsd.org
Subject:   Re: cvs commit: src/sys/vm vm_page.h
Message-ID:  <Pine.BSF.3.95.961008110943.10204X-100000@minnow.render.com>
In-Reply-To: <199610072028.PAA02845@dyson.iquest.net>

next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, 7 Oct 1996, John Dyson wrote:

> > John Dyson wrote:
> > > > 
> > > PQ_L2_SIZE is the size in pages...  PQ_L2_SIZE * PAGE_SIZE = 256k.
> > > IMO, It is kind-of bogus to "hard-code" the value, and I am thinking about
> > > a good solution for that problem.  The value is controllable from the
> > > config file, but my naming conventions leave something to be desired,
> > > so I haven't documented them (yet.)  If you have any ideas, let me know!!!
> > > 
> > 
> > Perhaps we could introduce a new word in the kernel config file.
> > Something like 
> > cache	<n>
> > where <n> is the amount of L2 cache. The L1 cache is processor specific
> > and thus can be obtained via the  cpu directive. The rest can then be
> > doen with macros.
> >
> Actually, the L1 cache will be changing on P5 class machines soon.

Yup.  P55c (Pentium with MMX extensions) have 32k cache.  It actually
improves performance quite a bit.  We have a few P55c machines here.  I
bet you could write a fast bcopy using MMX instructions to feed 64 bits at
a time to the write buffers.

--
Doug Rabson, Microsoft RenderMorphics Ltd.	Mail:  dfr@render.com
						Phone: +44 171 734 3761
						FAX:   +44 171 734 6426




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.3.95.961008110943.10204X-100000>