From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932519AbcFBKMc (ORCPT ); Thu, 2 Jun 2016 06:12:32 -0400 Received: from mail-wm0-f41.google.com ([74.125.82.41]:36083 "EHLO mail-wm0-f41.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751628AbcFBKMb (ORCPT ); Thu, 2 Jun 2016 06:12:31 -0400 Date: Thu, 2 Jun 2016 11:12:59 +0100 From: Lee Jones To: p.zabel@pengutronix.de Cc: linux-kernel@vger.kernel.org, hdegoede@redhat.com Subject: Shared Resets Message-ID: <20160602101259.GA1525@dell> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline Content-Transfer-Encoding: 8bit User-Agent: Mutt/1.5.24 (2015-08-30) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Philipp, Hans, After rebasing to the most recent upstream kernel tag (v4.6-rc1), I witnessed some issues where a couple of USB PHY drivers where having trouble requesting their reset lines. After some poking around, I stumbled across Hans' shared reset patch. It looks reasonable at first, but when I came to make the necessary driver changes, I realised that your effectively insisting that drivers have platform specific knowledge. The way I see it, drivers should request, deassert and assert their reset lines. Just as clock users, request, enable and disable their clocks. Whether the reset lines (or clocks) are left on after an assert (disable) because there are other users, should be of no concern to the consumer, right? I would like to see (and am happy to author) a different approach, where knowledge of whether a reset line is shared or not is contained in the framework, as it is in the clock case. What are your thoughts? Kind regards, Lee -- Lee Jones Linaro STMicroelectronics Landing Team Lead Linaro.org │ Open source software for ARM SoCs Follow Linaro: Facebook | Twitter | Blog