From owner-freebsd-multimedia Fri Mar 21 11:57:50 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id LAA08992 for multimedia-outgoing; Fri, 21 Mar 1997 11:57:50 -0800 (PST) Received: from rah.star-gate.com (rah.star-gate.com [204.188.121.18]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id LAA08982 for ; Fri, 21 Mar 1997 11:57:47 -0800 (PST) Received: from rah.star-gate.com (localhost.star-gate.com [127.0.0.1]) by rah.star-gate.com (8.8.5/8.7.3) with ESMTP id LAA01154; Fri, 21 Mar 1997 11:57:39 -0800 (PST) Message-Id: <199703211957.LAA01154@rah.star-gate.com> X-Mailer: exmh version 1.6.9 8/22/96 To: "Louis A. Mamakos" cc: Steve Passe , Michael Petry , multimedia@freebsd.org Subject: Continquous Memory vs Virtual Memory In-reply-to: Your message of "Fri, 21 Mar 1997 14:44:20 EST." <199703211944.OAA07986@whizzo.transsys.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 21 Mar 1997 11:57:38 -0800 From: Amancio Hasty Sender: owner-multimedia@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >From The Desk Of "Louis A. Mamakos" : > At least for the Bt848 class cards, continguous memory isn't strictly > required. From what I could tell reading the data sheet, you should > easily be able to do scatter DMA to multiple pages when you build the > RISC program. Once you've got that working, you ought to then be > able to accomodate any old buffer that the user specifies, perhaps even > specified with a uio struct. Yes, I can do the above and to a limited extent is being done in the driver right now. The even dma program is done in one block and the odd is done in another block. Your idea is a good one. I will look into it . My only reservation is the added complexity to the routines which build the risc programs. They are simple and easy to understand by design for the purpose of being easily maintainable by others. Tnks! Amancio