From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S933311Ab3GDVmi (ORCPT ); Thu, 4 Jul 2013 17:42:38 -0400 Received: from www.linutronix.de ([62.245.132.108]:42589 "EHLO Galois.linutronix.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756909Ab3GDVmh (ORCPT ); Thu, 4 Jul 2013 17:42:37 -0400 Date: Thu, 4 Jul 2013 23:42:36 +0200 (CEST) From: Thomas Gleixner To: Jonas Jensen cc: linux-arm-kernel@lists.infradead.org, linux-kernel@vger.kernel.org, arm@kernel.org, john.stultz@linaro.org, u.kleine-koenig@pengutronix.de, tomasz.figa@gmail.com, linus.walleij@linaro.org, thomas.petazzoni@free-electrons.com, arnd@arndb.de Subject: Re: [PATCH v5] ARM: clocksource: add support for MOXA ART SoCs In-Reply-To: <1372940383-5957-1-git-send-email-jonas.jensen@gmail.com> Message-ID: References: <1372687359-18235-1-git-send-email-jonas.jensen@gmail.com> <1372940383-5957-1-git-send-email-jonas.jensen@gmail.com> User-Agent: Alpine 2.02 (DEB 1266 2009-07-14) MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Linutronix-Spam-Score: -1.0 X-Linutronix-Spam-Level: - X-Linutronix-Spam-Status: No , -1.0 points, 5.0 required, ALL_TRUSTED=-1,SHORTCIRCUIT=-0.0001 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, 4 Jul 2013, Jonas Jensen wrote: > This patch adds an clocksource driver for the main timer(s) > found on MOXA ART SoCs. > > Signed-off-by: Jonas Jensen > --- > > Notes: > Applies to next-20130703 > > Changes since v4: > > 1. add general cache for TIMER_CR register What you implemented is not a register cache. A register cache is caching the current value and not some initial constant. > +static void moxart_clkevt_mode(enum clock_event_mode mode, > + struct clock_event_device *clk) > +{ > + switch (mode) { > + case CLOCK_EVT_MODE_RESUME: > + case CLOCK_EVT_MODE_ONESHOT: > + writel(timereg_cache & ~TIMEREG_CR_1_ENABLE, base + TIMER_CR); You just modify bits on the "cache" variable. though you are not caching it. As it seems to work it looks like this register simply can be written with constants. > + timereg_cache = readl(base + TIMER_CR) | TIMEREG_CR_2_ENABLE; Why are you reading that back? You know excactly which of the timers you are using and none of those should be enabled before you reach that code. If it one of them is enabled by the boot loader you better disable it in this init function. Now if you disable all of those timers and just use a known set, then you can do without a pseudo cache variable and just write constants into the control register, right ? Thanks, tglx From mboxrd@z Thu Jan 1 00:00:00 1970 From: tglx@linutronix.de (Thomas Gleixner) Date: Thu, 4 Jul 2013 23:42:36 +0200 (CEST) Subject: [PATCH v5] ARM: clocksource: add support for MOXA ART SoCs In-Reply-To: <1372940383-5957-1-git-send-email-jonas.jensen@gmail.com> References: <1372687359-18235-1-git-send-email-jonas.jensen@gmail.com> <1372940383-5957-1-git-send-email-jonas.jensen@gmail.com> Message-ID: To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On Thu, 4 Jul 2013, Jonas Jensen wrote: > This patch adds an clocksource driver for the main timer(s) > found on MOXA ART SoCs. > > Signed-off-by: Jonas Jensen > --- > > Notes: > Applies to next-20130703 > > Changes since v4: > > 1. add general cache for TIMER_CR register What you implemented is not a register cache. A register cache is caching the current value and not some initial constant. > +static void moxart_clkevt_mode(enum clock_event_mode mode, > + struct clock_event_device *clk) > +{ > + switch (mode) { > + case CLOCK_EVT_MODE_RESUME: > + case CLOCK_EVT_MODE_ONESHOT: > + writel(timereg_cache & ~TIMEREG_CR_1_ENABLE, base + TIMER_CR); You just modify bits on the "cache" variable. though you are not caching it. As it seems to work it looks like this register simply can be written with constants. > + timereg_cache = readl(base + TIMER_CR) | TIMEREG_CR_2_ENABLE; Why are you reading that back? You know excactly which of the timers you are using and none of those should be enabled before you reach that code. If it one of them is enabled by the boot loader you better disable it in this init function. Now if you disable all of those timers and just use a known set, then you can do without a pseudo cache variable and just write constants into the control register, right ? Thanks, tglx