From owner-freebsd-hackers Thu Aug 2 7:28:28 2001 Delivered-To: freebsd-hackers@freebsd.org Received: from perninha.conectiva.com.br (perninha.conectiva.com.br [200.250.58.156]) by hub.freebsd.org (Postfix) with ESMTP id D461837B406 for ; Thu, 2 Aug 2001 07:28:18 -0700 (PDT) (envelope-from riel@conectiva.com.br) Received: from burns.conectiva (burns.conectiva [10.0.0.4]) by perninha.conectiva.com.br (Postfix) with SMTP id DFBCF38D11 for ; Thu, 2 Aug 2001 11:28:14 -0300 (EST) Received: (qmail 31003 invoked by uid 0); 2 Aug 2001 14:27:19 -0000 Received: from duckman.distro.conectiva (HELO duckman.conectiva.com.br) (root@10.0.17.2) by burns.conectiva with SMTP; 2 Aug 2001 14:27:19 -0000 Received: from localhost (riel@localhost) by duckman.conectiva.com.br (8.11.4/8.11.3) with ESMTP id f72ESDl24314; Thu, 2 Aug 2001 11:28:13 -0300 X-Authentication-Warning: duckman.distro.conectiva: riel owned process doing -bs Date: Thu, 2 Aug 2001 11:28:13 -0300 (BRST) From: Rik van Riel X-X-Sender: To: mark tinguely Cc: , , , Subject: Re: How to visit physical memory above 4G? In-Reply-To: <200108021414.f72EEBS78044@web.cs.ndsu.nodak.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG On Thu, 2 Aug 2001, mark tinguely wrote: > The addressing use 64 bits for a memory pointer and the additional > page indirection add to the overhead. The stickler is the MMU is > still 32 bits. This means the PAE must segment the 64GB space into > 4GB segments or 4 1GB segments. The OS must manage which pages are > viewable to the process at this time. There's a better option. The application does this stuff itself, by mapping and unmapping SHM segments. This needs to be done anyway, since the application wouldn't be happy if the OS randomly remapped its address space to different pages ;)) > There is a third mode of addressing using 2MB pages (simular to the > 4MB page addressing mode for the 32 bit addressing scheme) that will > only give a process access to 4GB of memory (not segmentable to > a larger space, but can address physical memory located above the 4GB > address). You can do that with 4kB pages too. Both the middle page directory and page table have 512 64-bit entries, so you can simply map 4kB pages from above the 4GB physical boundary... regards, Rik -- Executive summary of a recent Microsoft press release: "we are concerned about the GNU General Public License (GPL)" http://www.surriel.com/ http://www.conectiva.com/ http://distro.conectiva.com/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message