From owner-cvs-all Mon Jul 22 0:17: 7 2002 Delivered-To: cvs-all@freebsd.org Received: from mx1.FreeBSD.org (mx1.FreeBSD.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id C682837B400; Mon, 22 Jul 2002 00:17:00 -0700 (PDT) Received: from odin.ac.hmc.edu (Odin.AC.HMC.Edu [134.173.32.75]) by mx1.FreeBSD.org (Postfix) with ESMTP id 1ABA043E70; Mon, 22 Jul 2002 00:17:00 -0700 (PDT) (envelope-from brdavis@odin.ac.hmc.edu) Received: from odin.ac.hmc.edu (IDENT:brdavis@localhost.localdomain [127.0.0.1]) by odin.ac.hmc.edu (8.12.3/8.12.3) with ESMTP id g6M7GwpO014287; Mon, 22 Jul 2002 00:16:58 -0700 Received: (from brdavis@localhost) by odin.ac.hmc.edu (8.12.3/8.12.3/Submit) id g6M7GwH0014286; Mon, 22 Jul 2002 00:16:58 -0700 Date: Mon, 22 Jul 2002 00:16:58 -0700 From: Brooks Davis To: Peter Wemm Cc: cvs-committers@FreeBSD.org, cvs-all@FreeBSD.org Subject: Re: cvs commit: src/sys/conf options src/sys/dev/ata ata-isa.c Message-ID: <20020722001658.B32402@Odin.AC.HMC.Edu> References: <20020721214837.A32402@Odin.AC.HMC.Edu> <20020722051725.D779F3923@overcee.wemm.org> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-md5; protocol="application/pgp-signature"; boundary="VrqPEDrXMn8OVzN4" Content-Disposition: inline User-Agent: Mutt/1.2.5.1i In-Reply-To: <20020722051725.D779F3923@overcee.wemm.org>; from peter@wemm.org on Sun, Jul 21, 2002 at 10:17:25PM -0700 X-Virus-Scanned: by amavisd-milter (http://amavis.org/) on odin.ac.hmc.edu Sender: owner-cvs-all@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG --VrqPEDrXMn8OVzN4 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Sun, Jul 21, 2002 at 10:17:25PM -0700, Peter Wemm wrote: > Brooks Davis wrote: >=20 > > With Warner's recent commit making bus interface symbols always > > avaiable, you can almost just compile ata-dma.c in. The main remaining > > problem seems to be ata_find_dev(). >=20 > When I spoke with Soren about the possibility of using function pointers > based on the type of device (pci, isa, whatever), he told me that he'd > thought about it for a while and had had going much further on his TODO > list. ie: each device would have the dma* functions pointed to directly = and > thereby eliminate the big switch(). I was just pointing it out that you can almost nuke the ugly dummy functions after Warner's change. Previously you got all sorts of undefined symbols related to pci, not it's get ata_find_dev(). Actually, I wonder if you couldn't get by with a stub ata_find_dev() for the NOPCI case. One thing worth noting is that -- if I understand the comments made at dev summit III correctly -- we've got a DMA capable ATA controler that is not a PCI device coming soon. Apparently the ATA controler Apple uses on all current Macs is not attached to a PCI bus. Thus disentangling ATA-DMA and PCI is not just a purity exercise. -- Brooks --=20 Any statement of the form "X is the one, true Y" is FALSE. PGP fingerprint 655D 519C 26A7 82E7 2529 9BF0 5D8E 8BE9 F238 1AD4 --VrqPEDrXMn8OVzN4 Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.0.6 (GNU/Linux) Comment: For info see http://www.gnupg.org iD8DBQE9O7FpXY6L6fI4GtQRAtlZAJwOKuZmCgXTkRekz7WzmTZUHrrWcACgkGcY x14yPdM8GxqwzpodEbJN4lA= =Q2uP -----END PGP SIGNATURE----- --VrqPEDrXMn8OVzN4-- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message