From owner-cvs-src@FreeBSD.ORG Wed Apr 16 17:56:27 2008 Return-Path: Delivered-To: cvs-src@FreeBSD.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id C1CE51065670; Wed, 16 Apr 2008 17:56:27 +0000 (UTC) (envelope-from sos@FreeBSD.org) Received: from spider.deepcore.dk (cpe.atm2-0-70484.0x50a6c9a6.abnxx16.customer.tele.dk [80.166.201.166]) by mx1.freebsd.org (Postfix) with ESMTP id 4EE7F8FC17; Wed, 16 Apr 2008 17:56:27 +0000 (UTC) (envelope-from sos@FreeBSD.org) Received: from laptop.deepcore.dk (laptop.deepcore.dk [194.192.25.138]) by spider.deepcore.dk (8.13.8/8.13.8) with ESMTP id m3GHuPHp066994; Wed, 16 Apr 2008 19:56:25 +0200 (CEST) (envelope-from sos@FreeBSD.org) Message-Id: From: =?ISO-8859-1?Q?S=F8ren_Schmidt?= To: John Baldwin In-Reply-To: <200804161008.01661.jhb@freebsd.org> Content-Type: text/plain; charset=ISO-8859-1; format=flowed; delsp=yes Content-Transfer-Encoding: quoted-printable Mime-Version: 1.0 (Apple Message framework v919.2) Date: Wed, 16 Apr 2008 19:56:25 +0200 References: <200804131605.m3DG5Z9P041647@repoman.freebsd.org> <86wsn0zasa.fsf@ds4.des.no> <200804161008.01661.jhb@freebsd.org> X-Mailer: Apple Mail (2.919.2) Cc: =?ISO-8859-1?Q?Dag-Erling_Sm=F8rgrav?= , src-committers@FreeBSD.org, cvs-all@FreeBSD.org, cvs-src@FreeBSD.org Subject: Re: cvs commit: src/sys/dev/ata ata-all.h ata-raid.c X-BeenThere: cvs-src@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: CVS commit messages for the src tree List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 16 Apr 2008 17:56:27 -0000 Hi ATA has done this for a long time for AHCI, if needed resources are =20 there it will go for AHCI and try to use it for known chipsets. What is relatively new in ATA is that it will go for chipsets that =20 claim to be AHCI compliant via the PCI "progif" register, however the =20= AHCI BAR still needs to be populated for it to be able to grap =20 resources needed. However some BIOS's effectively disables AHCI BAR's etc when its not =20 selected, not much we can do about that, or at least it will be a =20 nightmare to support. -S=F8ren On 16Apr, 2008, at 16:08 , John Baldwin wrote: > On Monday 14 April 2008 01:09:23 pm S=F8ren Schmidt wrote: >> Hi >> >> Yes, the registers used for setting up events when hotswap happens =20= >> are >> not accessible when in legacy compat mode. > > If all you need is for the AHCI BAR to have resources assigned newer =20= > FreeBSD > kernels (at least 6.x and later) are smart enough to alloc resources =20= > for a > BAR that the BIOS doesn't assign resources for when you do a > bus_alloc_resource(). > >> -S=F8ren >> >> On 14Apr, 2008, at 18:00 , Dag-Erling Sm=F8rgrav wrote: >>> S=F8ren Schmidt writes: >>>> You have to set AHCI mode in the BIOS or the resources needed to =20= >>>> use >>>> it wont be available, so yes its software but not ours :) >>> >>> Thanks S=F8ren, I will make sure the BIOS is correctly configured. >>> >>> Does this also affect hotswap? I assume that the ICH8 supports it, >>> but >>> I haven't been able to make it work. >>> >>> DES >>> -- >>> Dag-Erling Sm=F8rgrav - des@des.no > > > > --=20 > John Baldwin > >