From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752502AbbCKViO (ORCPT ); Wed, 11 Mar 2015 17:38:14 -0400 Received: from mail-gw2-out.broadcom.com ([216.31.210.63]:19857 "EHLO mail-gw2-out.broadcom.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751704AbbCKViM (ORCPT ); Wed, 11 Mar 2015 17:38:12 -0400 X-IronPort-AV: E=Sophos;i="5.11,384,1422950400"; d="scan'208";a="59200038" Message-ID: <5500B5B6.50609@broadcom.com> Date: Wed, 11 Mar 2015 14:37:58 -0700 From: Arun Ramamurthy User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Thunderbird/31.5.0 MIME-Version: 1.0 To: Arnd Bergmann CC: , Alan Stern , Mark Rutland , , Scott Branden , "Pawel Moll" , Ian Campbell , Ray Jui , , "Kishon Vijay Abraham I" , Jonathan Richardson , "Rob Herring" , , Dmitry Torokhov , Kumar Gala , Anatol Pomazau Subject: Re: [PATCH 1/2] phy: usbphy: Add dt documentation for Broadcom Cygnus USB PHY driver References: <1424200821-16368-1-git-send-email-arun.ramamurthy@broadcom.com> <54E53278.70705@broadcom.com> <54EE67A6.609@broadcom.com> <2079271.sEacVGN4vm@wuerfel> In-Reply-To: <2079271.sEacVGN4vm@wuerfel> Content-Type: text/plain; charset="windows-1252"; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 15-03-10 01:27 PM, Arnd Bergmann wrote: > On Wednesday 25 February 2015 16:24:06 Arun Ramamurthy wrote: >>>>> Should I run this by Alan Stern? >>>> >>>> I've added him to Cc here. He clearly didn't know the background about >>>> the DT binding change, and should not need to, but he may have an opinion >>>> on what names we should use. >>>> >>> >> Arnd, should I re patch the ehci-platform driver to avoid phy-names >> entirely? Alan, if not do you have an opinion on what the usb phy names >> should be? The current patch uses "usbp" + port number such as "usbp0" , >> "usbp1" etc > > I think avoiding the phy names would be best here, but it requires a > kernel API change first, because we do not have a way to get a phy > by index as we do for other subsystems (e.g. clocks or gpios). > Arnd , there is an existing api _of_phy_get that gets a phy by index. However it is not exported and is called from of_phy_get which is in turn called from devm_of_phy_get. My plan is to create a new function maybe devm_of_phy_get_by_index that calls _of_phy_get directly? What are your thoughts? > Arnd >