From mboxrd@z Thu Jan 1 00:00:00 1970 Return-path: Received: from mail.kapsi.fi ([217.30.184.167]:50873 "EHLO mail.kapsi.fi" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751654AbbBQTOl (ORCPT ); Tue, 17 Feb 2015 14:14:41 -0500 Message-ID: <54E3931C.9060508@iki.fi> Date: Tue, 17 Feb 2015 21:14:36 +0200 From: Antti Palosaari MIME-Version: 1.0 To: Philip Downer , Mauro Carvalho Chehab CC: linux-media@vger.kernel.org Subject: Re: [RFC PATCH 0/1] [media] pci: Add support for DVB PCIe cards from Prospero Technologies Ltd. References: <1424116126-14052-1-git-send-email-pdowner@prospero-tech.com> <54E24C83.7090309@iki.fi> <20150216214743.6a9180a6@recife.lan> In-Reply-To: Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-media-owner@vger.kernel.org List-ID: On 02/17/2015 08:22 PM, Philip Downer wrote: > Hi Mauro, > > On Mon, Feb 16, 2015 at 11:47 PM, Mauro Carvalho Chehab > wrote: >> Em Mon, 16 Feb 2015 22:01:07 +0200 >> Antti Palosaari escreveu: >> >>> Moikka! >>> >>> On 02/16/2015 09:48 PM, Philip Downer wrote: >>>> The Vortex PCIe card by Prospero Technologies Ltd is a modular DVB card >>>> with a hardware demux, the card can support up to 8 modules which are >>>> fixed to the board at assembly time. Currently we only offer one >>>> configuration, 8 x Dibcom 7090p DVB-t tuners, but we will soon be releasing >>>> other configurations. There is also a connector for an infra-red receiver >>>> dongle on the board which supports RAW IR. >>>> >>>> The driver has been in testing on our systems (ARM Cortex-A9, Marvell Sheva, >>>> x86, x86-64) for longer than 6 months, so I'm confident that it works. >>>> However as this is the first Linux driver I've written, I'm sure there are >>>> some things that I've got wrong. One thing in particular which has been >>>> raised by one of our early testers is that we currently register all of >>>> our frontends as being attached to one adapter. This means the device is >>>> enumerated in /dev like this: >>>> >>>> /dev/dvb/adapter0/frontend0 >>>> /dev/dvb/adapter0/dvr0 >>>> /dev/dvb/adapter0/demux0 >>>> >>>> /dev/dvb/adapter0/frontend1 >>>> /dev/dvb/adapter0/dvr1 >>>> /dev/dvb/adapter0/demux1 >>>> >>>> /dev/dvb/adapter0/frontend2 >>>> /dev/dvb/adapter0/dvr2 >>>> /dev/dvb/adapter0/demux2 >>>> >>>> etc. >>>> >>>> Whilst I think this is ok according to the spec, our tester has complained >>>> that it's incompatible with their software which expects to find just one >>>> frontend per adapter. So I'm wondering if someone could confirm if what >>>> I've done with regards to this is correct. >>> >>> As I understand all those tuners are independent (could be used same >>> time) you should register those as a 8 adapters, each having single >>> frontend, dvr and demux. >> >> Yeah, creating one adapter per device is the best solution, if you >> can't do things like: >> >> frontend0 -> demux2 -> dvr5 > > Thanks for confirming what Antti said, I'll change the driver and resubmit it. Also, take care to fix issues to meet Kernel coding style and checkpatch.pl requirements where possible. Read file Documentation/CodingStyle from kernel tree. There is script to check some common style issues and more, also in kernel tree ./scripts/checkpatch.pl --file drivers/media/pci/your_driver_file.c regards Antti -- http://palosaari.fi/