From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Fri, 28 Sep 2001 11:28:28 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Fri, 28 Sep 2001 11:28:19 -0400 Received: from [172.16.44.254] ([172.16.44.254]:47366 "EHLO int-mx1.corp.redhat.com") by vger.kernel.org with ESMTP id ; Fri, 28 Sep 2001 11:28:11 -0400 Date: Fri, 28 Sep 2001 10:28:36 -0500 From: Tommy Reynolds To: Norbert Roos Cc: linux-kernel@vger.kernel.org Subject: Re: inverse mmap() available? Message-Id: <20010928102836.02d0450a.reynolds@redhat.com> In-Reply-To: <3BB49438.1F2C59B3@berlin.de> In-Reply-To: <3BB49438.1F2C59B3@berlin.de> Organization: Red Hat Software, Inc. / Embedded Development X-Mailer: Sylpheed version 0.6.2cvs5 (GTK+ 1.2.9; ) X-Face: Nr)Jjr was pleased to say: > > Hello! > > Is there a way to map user space memory into kernel address space, e.g. > that i don't have to call get_user(var,...) but simply use var = > *user_space_ptr? > > What i intend to do (as the next step) is to DMA-transfer data directly > between a PCI device and user space memory. The buffer in user space > should be allocated with malloc(), so allocating a buffer in kernel and > mmap()-ping it to user space is not the solution.. > > I guess this has been asked before, any links to further information > would be great. Look at the KIOBUF stuff in for how to do this. ---------------------------------------------+----------------------------- Tommy Reynolds | mailto: Red Hat, Inc., Embedded Development Services | Phone: +1.256.704.9286 307 Wynn Drive NW, Huntsville, AL 35805 USA | FAX: +1.236.837.3839 Senior Software Developer | Mobile: +1.919.641.2923