From owner-freebsd-amd64@FreeBSD.ORG Wed Oct 26 23:29:48 2005 Return-Path: X-Original-To: amd64@freebsd.org Delivered-To: freebsd-amd64@FreeBSD.ORG Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id E8C2E16A421 for ; Wed, 26 Oct 2005 23:29:48 +0000 (GMT) (envelope-from pete@altadena.net) Received: from puffin.altadena.net (puffin.altadena.net [207.151.161.12]) by mx1.FreeBSD.org (Postfix) with ESMTP id 69AF843D45 for ; Wed, 26 Oct 2005 23:29:48 +0000 (GMT) (envelope-from pete@altadena.net) Received: from nat-gw.home.altadena.net ([66.127.158.99] helo=[192.168.169.28]) by puffin.altadena.net with esmtps (TLSv1:AES256-SHA:256) (Exim 4.54) id 1EUuiJ-0000Sq-Kw for amd64@freebsd.org; Wed, 26 Oct 2005 16:29:47 -0700 Message-ID: <4360116A.5080202@altadena.net> Date: Wed, 26 Oct 2005 16:29:46 -0700 From: Peter Carah User-Agent: Mozilla Thunderbird 1.0.7 (X11/20051024) X-Accept-Language: en-us, en MIME-Version: 1.0 To: amd64@freebsd.org X-Enigmail-Version: 0.93.0.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Cc: Subject: Compaq V2310 (about TI cardbus) X-BeenThere: freebsd-amd64@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Porting FreeBSD to the AMD64 platform List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 26 Oct 2005 23:29:49 -0000 The Compaq V2310 has a TI combination bus controller chip that runs firewire (works), SD/SM/MS/MM card (doesn't work in either 32 or 64 bit FreeBSD) and the PCCard slot (works fine for 16 bit cards but doesn't recognize Cardbus insertions at all). The problem with flash cards is just that the driver doesn't exist - there are 2 PCI vendor/product pairs involved, both of which aren't recognized. TI claims in the datasheet (which, miraculously, is public) for their chipset that both of these interfaces are "standard", requiring one to buy $$$$$ a book from the sd card consortium. The problem with cardbus is stranger since 16 bit cards work fine. Since HP in their infinite wisdom messed up the internal wireless card for anything but windoze (uses Broadcom 4318, the supplied driver doesn't work in 32-bit FBSD, and neither the Acer 64-bit driver nor the one at Linuxant work in 64-bit FBSD) *then* they fixed the bios up so that if one supplies e.g. an Atheros minipci (like a CM9; this is a very nice ABG card), the machine won't boot, with no apparent bypass. Best solution is to somehow bypass this... There are 2 workarounds - use a modern wireless pccard (needs cardbus) or a USB (needs ralink; the manufacturers seem to shift chipsets at a whim and so one can't necessarily count on which usb dongles contain ralink chips.) I did see a nice Atheros usb dongle but our Atheros driver doesn't cover that interface :-( Anyhow, I wondered about the cardbus problem... (hits both my 802.11 card and my 5220 card, so is pretty generic). I enabled debug for the cardbus driver and get NO output whatever on cardbus insertion (lots of output for 16-bit, and the card even works). At the moment, I can't get a dmesg because someone messed up the kldloader to put out several thousand error messages on boot (bad relocation type 10). I'll try another cvsup tonight to see if that is fixed and resend this with an attached dmesg whenever possible, if it would help whoever works on this stuff. (or I *could* boot in single user, dmesg, then go to multi...) -- Pete