Date: Thu, 21 Jun 2001 23:51:44 -0400 From: Brian Dean <bsd@bsdhome.com> To: Jonathan Lemon <jlemon@flugsvamp.com> Cc: mikulas@artax.karlin.mff.cuni.cz, hackers@FreeBSD.ORG Subject: Re: Patented algorithm in FreeBSD Message-ID: <20010621235144.A367@neutrino.bsdhome.com> In-Reply-To: <200106112237.f5BMbuh06004@prism.flugsvamp.com>; from jlemon@flugsvamp.com on Mon, Jun 11, 2001 at 05:37:56PM -0500 References: <local.mail.freebsd-hackers/Pine.LNX.3.96.1010611235853.12928A-100000@artax.karlin.mff.cuni.cz> <200106112237.f5BMbuh06004@prism.flugsvamp.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, Jun 11, 2001 at 05:37:56PM -0500, Jonathan Lemon wrote: > Hmm, let's see: > > Assignee: Digital Equipment Corporation (Maynard, MA) > Appl. No.: 646734 > Filed: May 3, 1996 > > Versus: > > * Derived from hp300 version by Mike Hibler, this version by William > * Jolitz uses a recursive map [a pde points to the page directory] to > * map the page tables using the pagetables themselves. This is done to > * reduce the impact on kernel virtual memory for lots of sparse address > * space, and to reduce the cost of memory to each process. > * > * from: hp300: @(#)pmap.h 7.2 (Berkeley) 12/16/90 > * from: @(#)pmap.h 7.4 (Berkeley) 5/12/91 > > So it looks like we have prior art by around 6 years, which would > invalidate the patent iff it was the same thing. [rummaging through my old box of manuals... Aha!] This also appears to be _very_ similar to the scheme described by Intel themselves in a skinny little book entitled "80386 System Software Writer's Guide", ISBN 1-55512-023-7, Intel order number 231499-001. This book is dated 1987 and describes the handy recursive page table mapping starting at the bottom of page 2-18. -Brian To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20010621235144.A367>