From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:57957) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1XRNyx-0002rg-LV for qemu-devel@nongnu.org; Tue, 09 Sep 2014 12:09:32 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1XRNys-00085D-C0 for qemu-devel@nongnu.org; Tue, 09 Sep 2014 12:09:27 -0400 Received: from mail-wi0-f178.google.com ([209.85.212.178]:63049) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1XRNys-00084q-2v for qemu-devel@nongnu.org; Tue, 09 Sep 2014 12:09:22 -0400 Received: by mail-wi0-f178.google.com with SMTP id hi2so1492322wib.17 for ; Tue, 09 Sep 2014 09:09:20 -0700 (PDT) Message-ID: <540F2618.9080600@linaro.org> Date: Tue, 09 Sep 2014 18:08:56 +0200 From: Eric Auger MIME-Version: 1.0 References: <1410249273-6063-1-git-send-email-eric.auger@linaro.org> <1410249273-6063-3-git-send-email-eric.auger@linaro.org> <540EDECB.1000801@redhat.com> <540F2325.3070009@linaro.org> In-Reply-To: Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [PATCH v3 2/6] hw/arm/dyn_sysbus_devtree: helpers for sysbus device dynamic dt node generation List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Peter Maydell Cc: Joel Schopp , Alex Williamson , Kim Phillips , eric.auger@st.com, Patch Tracking , Marc Zyngier , manish.jaggi@caviumnetworks.com, Will Deacon , QEMU Developers , Alvise Rigo , Bharat Bhushan , Alexander Graf , "kvmarm@lists.cs.columbia.edu" , Antonios Motakis , Stuart Yoder , Paolo Bonzini , =?UTF-8?B?QW5kcmVhcyBGw6RyYmVy?= , Christoffer Dall On 09/09/2014 06:00 PM, Peter Maydell wrote: > On 9 September 2014 16:56, Eric Auger wrote: >> irq_start: needed because when the "interrupts" property is set for the >> leaf component the irq number is irq_start + >> object_property_get_int(obj, "irq[i]", NULL) >> irq[i] being in [0, params->platform_bus_num_irqs] >> >> intc: this was in case the leaf component would use "interrupt-parent" >> prop. I miss experience on device trees and I don't know if it make >> sense the leaf component uses a different interrupt controller than the >> parent platform bus or if such property is mandatory in some cases. >> Maybe not needed indeed. > > Somewhat tangential, but for passthrough devices how > do we decide whether the device tree node needs to > mark the interrupt as edge or level triggered? > Presumably this is going to be a "depends on what > the passed through hardware is" thing... Hi Peter, Yes I think this is one of the reasons why Alex insisted on having the device node creation specialized for each device and not common to VFIO devices. Besides, I did not looked sufficiently at Antonios' patch "VFIO: PLATFORM: Return device tree info for a platform device node" but I guess this would enable to retrieve the property values. https://www.mail-archive.com/kvm@vger.kernel.org/msg106282.html. But I remember I read somewhere such interrupts properties were not always correct in dt? Best Regards Eric > > -- PMM >