From owner-freebsd-current@FreeBSD.ORG Tue Jan 14 20:15:58 2014 Return-Path: Delivered-To: freebsd-current@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [8.8.178.115]) (using TLSv1 with cipher ADH-AES256-SHA (256/256 bits)) (No client certificate requested) by hub.freebsd.org (Postfix) with ESMTPS id DD92FEEC; Tue, 14 Jan 2014 20:15:58 +0000 (UTC) Received: from bigwig.baldwin.cx (bigwig.baldwin.cx [IPv6:2001:470:1f11:75::1]) (using TLSv1 with cipher ADH-CAMELLIA256-SHA (256/256 bits)) (No client certificate requested) by mx1.freebsd.org (Postfix) with ESMTPS id B25D710B4; Tue, 14 Jan 2014 20:15:58 +0000 (UTC) Received: from jhbbsd.localnet (unknown [209.249.190.124]) by bigwig.baldwin.cx (Postfix) with ESMTPSA id A52AFB946; Tue, 14 Jan 2014 15:15:57 -0500 (EST) From: John Baldwin To: freebsd-stable@freebsd.org Subject: Re: Shared IRQ on ATOM based Soekris net6501 board Date: Tue, 14 Jan 2014 11:09:18 -0500 User-Agent: KMail/1.13.5 (FreeBSD/8.4-CBSD-20130906; KDE/4.5.5; amd64; ; ) References: <20140114160346.2c34db92@fusen> In-Reply-To: <20140114160346.2c34db92@fusen> MIME-Version: 1.0 Content-Type: Text/Plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Message-Id: <201401141109.18137.jhb@freebsd.org> X-Greylist: Sender succeeded SMTP AUTH, not delayed by milter-greylist-4.2.7 (bigwig.baldwin.cx); Tue, 14 Jan 2014 15:15:57 -0500 (EST) Cc: Christophe Prevotaux , freebsd-current@freebsd.org X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.17 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, 14 Jan 2014 20:15:58 -0000 On Tuesday, January 14, 2014 10:03:46 am Christophe Prevotaux wrote: > > I have found out that when I use 2 AR9580 mini PCIe board > some irq (namely in that case irq18 is shared over 3 devices) > > I wonder why it is done like this and if this could be something to fix ? You can't fix it. This is how the interrupts are connected on your motherboard. It's true that in PCI-e there are no longer physical interrupt lines, but the connections may still be hardcoded in silicon. The only chance you might possibly have is to use ACPI (which doesn't work on your board) as if the PCI interrupts are hooked up to an interrupt router prior to the I/O APIC you might be able to shuffle things. A far better idea would be to use a board that supports MSI interrupts (and/or fix the driver to use MSI if it is not doing so already). -- John Baldwin