Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755371Ab3G2VDO (ORCPT ); Mon, 29 Jul 2013 17:03:14 -0400 Received: from mail-bk0-f52.google.com ([209.85.214.52]:49057 "EHLO mail-bk0-f52.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752410Ab3G2VDM (ORCPT ); Mon, 29 Jul 2013 17:03:12 -0400 Message-ID: <51F6D88B.80202@gmail.com> Date: Mon, 29 Jul 2013 23:03:07 +0200 From: Sylwester Nawrocki User-Agent: Mozilla/5.0 (X11; Linux i686; rv:11.0) Gecko/20120412 Thunderbird/11.0.1 MIME-Version: 1.0 To: Julius Werner CC: Felipe Balbi , linux-kernel@vger.kernel.org, linux-samsung-soc@vger.kernel.org, Tomasz Figa , Vivek Gautam , devicetree@vger.kernel.org Subject: Re: [PATCH] usb: phy-samsung-usb: Simplify PMU register handling References: <1375130839-5523-1-git-send-email-jwerner@chromium.org> In-Reply-To: <1375130839-5523-1-git-send-email-jwerner@chromium.org> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 3818 Lines: 117 Hi, On 07/29/2013 10:47 PM, Julius Werner wrote: > This patch simplifies the way the phy-samsung-usb code finds the correct > power management register to enable PHY clock gating. Previously, the > code would calculate the register address from a device tree supplied > base address and add an offset based on the PHY type. > > Since every PHY has its own device tree entry and needs only one > register, we can just encode the address itself in the device tree and > remove the diffentiation in the code. The bitmask needed to specify the > bit within that register stays in place, allowing support for platforms > like s3c64xx that use different bits within the same register. Due to > this simplification, the whole complication of a Samsung-specific USB > PHY type can be removed from the PHY driver. > > Change-Id: Id823f04bbf1942f307bd1d24ec9d8d55a69b0e56 > Signed-off-by: Julius Werner > --- > arch/arm/boot/dts/exynos5250.dtsi | 4 ++-- > drivers/usb/phy/phy-samsung-usb.c | 46 +++++++++----------------------------- > drivers/usb/phy/phy-samsung-usb.h | 34 +++++++--------------------- > drivers/usb/phy/phy-samsung-usb2.c | 34 ++++++++-------------------- > drivers/usb/phy/phy-samsung-usb3.c | 13 +++-------- > 5 files changed, 32 insertions(+), 99 deletions(-) > > diff --git a/arch/arm/boot/dts/exynos5250.dtsi b/arch/arm/boot/dts/exynos5250.dtsi > index ef57277..5a754d7 100644 > --- a/arch/arm/boot/dts/exynos5250.dtsi > +++ b/arch/arm/boot/dts/exynos5250.dtsi > @@ -473,7 +473,7 @@ > ranges; > > usbphy-sys { > - reg =<0x10040704 0x8>; > + reg =<0x10040704 0x4>; > }; > }; > > @@ -505,7 +505,7 @@ > ranges; > > usbphy-sys { > - reg =<0x10040704 0x8>, > + reg =<0x10040708 0x4>, > <0x10050230 0x4>; > }; > }; > diff --git a/drivers/usb/phy/phy-samsung-usb.c b/drivers/usb/phy/phy-samsung-usb.c > index ac025ca..c142233 100644 > --- a/drivers/usb/phy/phy-samsung-usb.c > +++ b/drivers/usb/phy/phy-samsung-usb.c > @@ -27,7 +27,6 @@ > #include > #include > #include > -#include > > #include "phy-samsung-usb.h" > > @@ -42,9 +41,9 @@ int samsung_usbphy_parse_dt(struct samsung_usbphy *sphy) > return -ENODEV; > } > > - sphy->pmuregs = of_iomap(usbphy_sys, 0); > + sphy->pmureg = of_iomap(usbphy_sys, 0); > > - if (sphy->pmuregs == NULL) { > + if (sphy->pmureg == NULL) { > dev_err(sphy->dev, "Can't get usb-phy pmu control register\n"); > goto err0; > } > @@ -75,35 +74,26 @@ EXPORT_SYMBOL_GPL(samsung_usbphy_parse_dt); > */ > void samsung_usbphy_set_isolation_4210(struct samsung_usbphy *sphy, bool on) > { > - void __iomem *reg = NULL; > u32 reg_val; > u32 en_mask = 0; > > - if (!sphy->pmuregs) { > + if (!sphy->pmureg) { > dev_warn(sphy->dev, "Can't set pmu isolation\n"); > return; > } > > - if (sphy->phy_type == USB_PHY_TYPE_DEVICE) { > - reg = sphy->pmuregs + sphy->drv_data->devphy_reg_offset; > - en_mask = sphy->drv_data->devphy_en_mask; > - } else if (sphy->phy_type == USB_PHY_TYPE_HOST) { > - reg = sphy->pmuregs + sphy->drv_data->hostphy_reg_offset; > - en_mask = sphy->drv_data->hostphy_en_mask; > - } How is en_mask supposed to get initialized, now when this code is removed ? It's always 0 now, isn't it ? > - > - reg_val = readl(reg); > + reg_val = readl(sphy->pmureg); > > if (on) > reg_val&= ~en_mask; > else > reg_val |= en_mask; > > - writel(reg_val, reg); > + writel(reg_val, sphy->pmureg); -- Regards, Sylwester -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/