From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20) id 19adRL-0008If-56 for qemu-devel@nongnu.org; Thu, 10 Jul 2003 11:34:35 -0400 Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20) id 19adNn-0006Tx-4M for qemu-devel@nongnu.org; Thu, 10 Jul 2003 11:30:55 -0400 Received: from smtp5.wanadoo.fr ([193.252.22.27] helo=mwinf0402.wanadoo.fr) by monty-python.gnu.org with esmtp (Exim 4.20) id 19adMF-000636-TU for qemu-devel@nongnu.org; Thu, 10 Jul 2003 11:29:20 -0400 Received: from free.fr (unknown [81.49.237.20]) by mwinf0402.wanadoo.fr (SMTP Server) with ESMTP id C92E2800063 for ; Thu, 10 Jul 2003 17:29:18 +0200 (CEST) Message-ID: <3F0D867D.3060800@free.fr> Date: Thu, 10 Jul 2003 17:30:05 +0200 From: Fabrice Bellard MIME-Version: 1.0 References: <20030710093529.975872C085@lists.samba.org> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Subject: [Qemu-devel] 4GB/4GB split Reply-To: qemu-devel@nongnu.org List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: qemu-devel@nongnu.org Hi, I just saw the announce of Ingo Molnar about his patch for 4GB/4GB kernel split. This patch is mostly what I was dreaming in order to have optimal performances in QEMU : no guest kernel patches will be necessary and no software MMU will be necessary. I hope that some of the kernel gurus reading this list will tell Ingo that his patch may be useful also for people having less than 4GB of RAM :-) Ideally, he should add a feature to activate the 4GB/4GB on demand (if the user process maps data above 0xc0000000) and he should also be able to move dynamically the trampoline page if the user process tries to map something at its address. Fabrice.