From owner-freebsd-hackers Thu Mar 13 09:39:11 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id JAA10043 for hackers-outgoing; Thu, 13 Mar 1997 09:39:11 -0800 (PST) Received: from chouette.inria.fr (chouette.inria.fr [138.96.24.103]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id JAA10034 for ; Thu, 13 Mar 1997 09:39:08 -0800 (PST) Received: by chouette.inria.fr (8.8.5/8.6.12) id SAA11564; Thu, 13 Mar 1997 18:38:56 +0100 (MET) Date: Thu, 13 Mar 1997 18:38:56 +0100 (MET) Message-Id: <199703131738.SAA11564@chouette.inria.fr> From: Emmanuel Duros To: erich@lodgenet.com CC: freebsd-hackers@FreeBSD.ORG, pplc@chouette.inria.fr In-reply-to: <199703131721.LAA29833@jake.lodgenet.com> (erich@lodgenet.com) Subject: Re: Pb using DMA - Physical addressing Reply-to: Emmanuel.Duros@sophia.inria.fr Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Eric L. Hernes writes: >>However I had to add another function in sys/i386/isa/isa.c which I >>called isa_dmastart_blk() in order to support the block transfer >>mode. Could this feature be added to future FreeBSD releases ? >>(isa_dmastart() only support single and autoinitialize transfer mode) > >I haven't looked too closely, but is it possible to add it to the >flags argument? I'd prefer that to another funciton. >Something like `isa_dmastart(B_WRITE|DMA_BLOCK, .....);', maybe? Definitely, I think it would only require at most 20 lines of code in the isa_dmastart() function... I could post the hack if necessary. I did write the isa_dmastart_blk() function in order to not interfere with other device drivers, I have been rebooting my PC probably 100 times these last days ! Emmanuel