From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932971AbbA2Bv0 (ORCPT ); Wed, 28 Jan 2015 20:51:26 -0500 Received: from mout.kundenserver.de ([212.227.17.13]:54176 "EHLO mout.kundenserver.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1758843AbbA2BvW (ORCPT ); Wed, 28 Jan 2015 20:51:22 -0500 From: Arnd Bergmann To: Zubair Lutfullah Kakakhel Cc: Guenter Roeck , wim@iguana.be, devicetree@vger.kernel.org, linux-kernel@vger.kernel.org, linux-watchdog@vger.kernel.org, paul@crapouillou.net Subject: Re: [PATCH 1/2] dt: watchdog: Add DT binding documentation for jz47xx watchdog timer Date: Wed, 28 Jan 2015 12:23:48 +0100 Message-ID: <9831838.T0biTLFe1N@wuerfel> User-Agent: KMail/4.11.5 (Linux/3.16.0-10-generic; KDE/4.11.5; x86_64; ; ) In-Reply-To: <54C8B974.8030703@imgtec.com> References: <1422371490-44402-1-git-send-email-Zubair.Kakakhel@imgtec.com> <9081741.fTg5kD1W54@wuerfel> <54C8B974.8030703@imgtec.com> MIME-Version: 1.0 Content-Transfer-Encoding: 7Bit Content-Type: text/plain; charset="us-ascii" X-Provags-ID: V03:K0:WuTvrsnsGBLG5WSxNH6mBLAjQZ1bjEovJlFREJYCIpY7q0jJhaS vQCISJV6rUE2HBhEE+lxjdBTjd/tLNjwJZvOXVrX8OX5VvOeWRh+tgUT2d7Sg/Hrx4z2ju+ RG7w5+nSQpefEjxA2D9kz5aRhpLggrqrvWliPfEN/TGpETKrcFKQKSa+zonDHegAX0bidER O96YDjT3cCYXDKdwMozkA== X-UI-Out-Filterresults: notjunk:1; Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wednesday 28 January 2015 10:27:00 Zubair Lutfullah Kakakhel wrote: > > There is on-going work to fix jz4740, add jz4780 and shake the entire clock tree as well. > > Patch 14 onwards in this series > http://patchwork.linux-mips.org/bundle/paulburton/ci20-v3.20/ Right, that looks good. > Instead of lumping things out in huge changesets, I intended to push out the minor patches that are disjoint. > > That was the purpose of sending these two patches. > > Current binding requires the clock-name to be "rtc". Hence the name at the moment. The problem with the binding is that once you've established it, it is very hard to change. One possible solution would be to try 'of_clk_get(dev->of_node, 0)' in the driver first, and only then call clk_get(dev, "rtc"). Doing that, you can have an anonymous clock in the DT without waiting for the clock driver to get merged first or breaking the existing setup. Arnd