From owner-freebsd-current@FreeBSD.ORG Tue Aug 10 07:24:48 2004 Return-Path: Delivered-To: freebsd-current@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 1317916A4CE for ; Tue, 10 Aug 2004 07:24:48 +0000 (GMT) Received: from mail.gmx.net (pop.gmx.de [213.165.64.20]) by mx1.FreeBSD.org (Postfix) with SMTP id 9013843D31 for ; Tue, 10 Aug 2004 07:24:44 +0000 (GMT) (envelope-from sebastian.ssmoller@gmx.net) Received: (qmail 32659 invoked by uid 65534); 10 Aug 2004 07:24:43 -0000 Received: from pD9E82618.dip.t-dialin.net (HELO tyrael.linnet) (217.232.38.24) by mail.gmx.net (mp019) with SMTP; 10 Aug 2004 09:24:43 +0200 X-Authenticated: #15005775 Date: Tue, 10 Aug 2004 09:26:48 +0200 From: sebastian ssmoller To: freebsd-current@freebsd.org Message-Id: <20040810092648.64e582c2.sebastian.ssmoller@gmx.net> In-Reply-To: <41182054.7090109@root.org> References: <41131EEC.4050909@root.org> <291946042.22742@mails.tsinghua.edu.cn> <41182054.7090109@root.org> X-Mailer: Sylpheed version 0.9.9-gtk2-20040229 (GTK+ 2.4.1; i386-portbld-freebsd5.2.1) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Subject: Re: PLEASE TEST: acpi pci irq routing X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 10 Aug 2004 07:24:48 -0000 hi, not sure what i should expect (to see). but this patch looks good here. everything works fine. no errors nor warnings. all irqs seem to be ok. thx regards, seb ps: hardware: asus m2400n notebook On Mon, 09 Aug 2004 18:09:40 -0700 Nate Lawson wrote: > Thanks for all the feedback to everyone who responded. I have > finished > reworking our PCI irq routing approach and the result is much simpler > while retaining the existing weighting algorithm from acpi_pci_link.c > (around since Oct. 2002). It removes the hack which causes the first > IRQ to be chosen from possible IRQs if the BIOS hasn't routed an > interrupt. The selection algorithm is to always use the BIOS irq, if > valid, and then select from the weighted priority list if that fails. > I've tested both approaches fully. > > Please test this patch to be sure all PCI devices continue (or begin) > to > function as expected. If something fails for you, please send me the > dmesg from boot -v. > > Thanks, > Nate > -- "Perfection is achieved, not when there is nothing left to add, but when there is nothing left to take away." --- Antoine de St. Exupery, Wind, Sand, and Stars, 1939