Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S933822AbbENPfj (ORCPT ); Thu, 14 May 2015 11:35:39 -0400 Received: from mailgw01.mediatek.com ([210.61.82.183]:44238 "EHLO mailgw01.mediatek.com" rhost-flags-OK-FAIL-OK-FAIL) by vger.kernel.org with ESMTP id S933338AbbENPfg (ORCPT ); Thu, 14 May 2015 11:35:36 -0400 X-Listener-Flag: 11101 Message-ID: <1431617729.15140.54.camel@mtksdaap41> Subject: Re: [PATCH 2/2] pwm: add Mediatek display PWM driver support From: YH Huang To: Thierry Reding CC: Matthias Brugger , Mark Rutland , Rob Herring , Pawel Moll , , , , , , , Sascha Hauer , Date: Thu, 14 May 2015 23:35:29 +0800 In-Reply-To: <20150512133202.GA24684@ulmo.nvidia.com> References: <1431336382-13167-1-git-send-email-yh.huang@mediatek.com> <1431336382-13167-3-git-send-email-yh.huang@mediatek.com> <20150512133202.GA24684@ulmo.nvidia.com> Content-Type: text/plain; charset="UTF-8" X-Mailer: Evolution 3.2.3-0ubuntu6 Content-Transfer-Encoding: 7bit MIME-Version: 1.0 X-MTK: N Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 11322 Lines: 393 On Tue, 2015-05-12 at 15:32 +0200, Thierry Reding wrote: > On Mon, May 11, 2015 at 05:26:22PM +0800, YH Huang wrote: > > Add display PWM driver support to modify backlight for MT8173/MT6595. > > > > Signed-off-by: YH Huang > > --- > > drivers/pwm/Kconfig | 9 ++ > > drivers/pwm/Makefile | 1 + > > drivers/pwm/pwm-disp-mediatek.c | 225 ++++++++++++++++++++++++++++++++++++++++ > > 3 files changed, 235 insertions(+) > > create mode 100644 drivers/pwm/pwm-disp-mediatek.c > > > > diff --git a/drivers/pwm/Kconfig b/drivers/pwm/Kconfig > > index b1541f4..9edbb5a 100644 > > --- a/drivers/pwm/Kconfig > > +++ b/drivers/pwm/Kconfig > > @@ -111,6 +111,15 @@ config PWM_CLPS711X > > To compile this driver as a module, choose M here: the module > > will be called pwm-clps711x. > > > > +config PWM_DISP_MEDIATEK > > Please make this "PWM_MEDIATEK_DISP" so we can keep everything sorted by > vendor. > Probably "PWM_MTK_DSIP" for Matthias opinion? > > + tristate "MEDIATEK display PWM driver" > > "MediaTek"? > OK. > > + depends on OF > > Technically I think you need depends on HAS_IOMEM here to avoid breakage > on randconfig builds. > OK. > > + help > > + Generic PWM framework driver for mediatek disp-pwm device. > > "MediaTek"? Also perhaps this should describe what this PWM is instead > of just the "disp-pwm" which leaves everyone guessing what it is. From > the name I'd expect it to be the PWM that is used to control the > backlight brightness for display, but I think this description should > say that explicitly. > > > + > > + To compile this driver as a module, choose M here: the module > > + will be called pwm-disp-mediatek. > > pwm-mediatek-disp pwm-mtk-disp? > > > + > > config PWM_EP93XX > > tristate "Cirrus Logic EP93xx PWM support" > > depends on ARCH_EP93XX > > diff --git a/drivers/pwm/Makefile b/drivers/pwm/Makefile > > index ec50eb5..c5ff72a 100644 > > --- a/drivers/pwm/Makefile > > +++ b/drivers/pwm/Makefile > > @@ -8,6 +8,7 @@ obj-$(CONFIG_PWM_BCM_KONA) += pwm-bcm-kona.o > > obj-$(CONFIG_PWM_BCM2835) += pwm-bcm2835.o > > obj-$(CONFIG_PWM_BFIN) += pwm-bfin.o > > obj-$(CONFIG_PWM_CLPS711X) += pwm-clps711x.o > > +obj-$(CONFIG_PWM_DISP_MEDIATEK) += pwm-disp-mediatek.o > > Also name this "pwm-mediatek-disp.o" to reflect the Kconfig name. pwm-mtk-disp.o? > > > obj-$(CONFIG_PWM_EP93XX) += pwm-ep93xx.o > > obj-$(CONFIG_PWM_FSL_FTM) += pwm-fsl-ftm.o > > obj-$(CONFIG_PWM_IMG) += pwm-img.o > > diff --git a/drivers/pwm/pwm-disp-mediatek.c b/drivers/pwm/pwm-disp-mediatek.c > > new file mode 100644 > > index 0000000..38293af > > --- /dev/null > > +++ b/drivers/pwm/pwm-disp-mediatek.c > > @@ -0,0 +1,225 @@ > > +/* > > + * Mediatek display pulse-width-modulation controller driver. > > + * Copyright (c) 2015 MediaTek Inc. > > + * Author: YH Huang > > + * > > + * This program is free software; you can redistribute it and/or modify > > + * it under the terms of the GNU General Public License version 2 as > > + * published by the Free Software Foundation. > > + * > > + * This program is distributed in the hope that it will be useful, > > + * but WITHOUT ANY WARRANTY; without even the implied warranty of > > + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the > > + * GNU General Public License for more details. > > + */ > > + > > +#include > > +#include > > +#include > > +#include > > +#include > > +#include > > +#include > > +#include > > + > > +#define DISP_PWM_EN_OFF (0x0) > > The _OFF suffix here is unfortunate because I'm reading this as being > the "OFF" value for a field named "DISP_PWM_EN". I think it'd be better > to name this one DISP_PWM_EN. The same goes for any of the following > registers. > > Also, drop the parentheses if the expression is simple. OK. It is really confusing. > > > +#define PWM_ENABLE_SHIFT (0x0) > > +#define PWM_ENABLE_MASK (0x1 << PWM_ENABLE_SHIFT) > > + > > +#define DISP_PWM_COMMIT_OFF (0x08) > > +#define PWM_COMMIT_SHIFT (0x0) > > +#define PWM_COMMIT_MASK (0x1 << PWM_COMMIT_SHIFT) > > + > > +#define DISP_PWM_CON_0_OFF (0x10) > > +#define PWM_CLKDIV_SHIFT (0x10) > > +#define PWM_CLKDIV_MASK (0x3ff << PWM_CLKDIV_SHIFT) > > +#define PWM_CLKDIV_MAX (0x000003ff) > > + > > +#define DISP_PWM_CON_1_OFF (0x14) > > +#define PWM_PERIOD_SHIFT (0x0) > > +#define PWM_PERIOD_MASK (0xfff << PWM_PERIOD_SHIFT) > > +#define PWM_PERIOD_MAX (0x00000fff) > > +/* Shift log2(PWM_PERIOD_MAX + 1) as divisor */ > > +#define PWM_PERIOD_BIT_SHIFT 12 > > This is confusing, see below. I don't really understand what does this mean. Please help me. > > > + > > +#define PWM_HIGH_WIDTH_SHIFT (0x10) > > +#define PWM_HIGH_WIDTH_MASK (0x1fff << PWM_HIGH_WIDTH_SHIFT) > > + > > +#define NUM_PWM 1 > > You can drop this, it's only used once. > OK. > > + > > +struct mtk_disp_pwm_chip { > > + struct pwm_chip chip; > > + struct device *dev; > > + struct clk *clk_main; > > + struct clk *clk_mm; > > + void __iomem *mmio_base; > > +}; > > Please don't add this artificial padding, a single space is enough as a > separator. Also, I think you can drop the _chip suffix on the structure > name. > OK. > > + > > +static void mtk_disp_pwm_setting(void __iomem *address, u32 value, u32 mask) > > This is a bad name. I think you should adopt the naming of regmap, which > this effectively copies, so this should be: > > static void mtk_disp_pwm_update_bits(void __iomem *address, u32 mask, u32 value) > It is much readable. > > +static int mtk_disp_pwm_config(struct pwm_chip *chip, struct pwm_device *pwm, > > + int duty_ns, int period_ns) > > +{ > > + struct mtk_disp_pwm_chip *mpc; > > + u64 div, rate; > > + u32 clk_div, period, high_width, rem; > > + > > + /* > > + * Find period, high_width and clk_div to suit duty_ns and period_ns. > > + * Calculate proper div value to keep period value in the bound. > > + * > > + * period_ns = 10^9 * (clk_div + 1) * (period +1) / PWM_CLK_RATE > > + * duty_ns = 10^9 * (clk_div + 1) * (high_width + 1) / PWM_CLK_RATE > > + * > > + * period = (PWM_CLK_RATE * period_ns) / (10^9 * (clk_div + 1)) - 1 > > + * high_width = (PWM_CLK_RATE * duty_ns) / (10^9 * (clk_div + 1)) - 1 > > + */ > > + mpc = container_of(chip, struct mtk_disp_pwm_chip, chip); > > Can you add a static inline function to wrap the container_of() > invocation? You need to do this quite often and the wrapper will shorten > the code significantly: > > static inline struct mtk_disp_pwm *to_mtk_disp_pwm(struct pwm_chip *chip) > { > return container_of(chip, struct mtk_disp_pwm, chip); > } > > Also it's slightly more canonical to initialize the mpc variable as part > of the declaration. > This is a good way to shorten the code. > > + rate = clk_get_rate(mpc->clk_main); > > + clk_div = div_u64_rem(rate * period_ns, NSEC_PER_SEC, &rem) >> > > + PWM_PERIOD_BIT_SHIFT; > > + if (clk_div > PWM_CLKDIV_MAX) > > + return -EINVAL; > > + > > + div = clk_div + 1; > > + period = div64_u64(rate * period_ns, NSEC_PER_SEC * div); > > + if (period > 0) > > + period--; > > + high_width = div64_u64(rate * duty_ns, NSEC_PER_SEC * div); > > + if (high_width > 0) > > + high_width--; > > + > > + mtk_disp_pwm_setting(mpc->mmio_base + DISP_PWM_CON_0_OFF, > > + clk_div << PWM_CLKDIV_SHIFT, PWM_CLKDIV_MASK); > > + mtk_disp_pwm_setting(mpc->mmio_base + DISP_PWM_CON_1_OFF, > > + (period << PWM_PERIOD_SHIFT) | > > + (high_width << PWM_HIGH_WIDTH_SHIFT), > > + PWM_PERIOD_MASK | PWM_HIGH_WIDTH_MASK); > > Can you align these differently to make it more readable? > > mtk_display_pwm_update_bits(mpc->mmio_base + DISP_PWM_CON_1, > PWM_PERIOD_MASK | PWM_HIGH_WIDTH_MASK, > (period << PWM_PERIOD_SHIFT) | > (high_width << PWM_HIGH_WIDTH_SHIFT)); > > Or perhaps split it into multiple steps to make it even more readable: > > value = (high_width << PWM_HIGH_WIDTH_SHIFT) | > (period << PWM_PERIOD_SHIFT); > mtk_display_pwm_update_bits(mpc->mmio_base + DISP_PWM_CON_1, > PWM_HIGH_WIDTH_MASK | PWM_PERIOD_MASK, > value); > OK, it is really hard to read. > > +static int mtk_disp_pwm_probe(struct platform_device *pdev) > > +{ > > + struct mtk_disp_pwm_chip *pwm; > > + struct resource *r; > > + int ret; > > + > > + pwm = devm_kzalloc(&pdev->dev, sizeof(*pwm), GFP_KERNEL); > > + if (!pwm) > > + return -ENOMEM; > > + > > + pwm->dev = &pdev->dev; > > + > > + r = platform_get_resource(pdev, IORESOURCE_MEM, 0); > > + pwm->mmio_base = devm_ioremap_resource(&pdev->dev, r); > > + if (IS_ERR(pwm->mmio_base)) > > + return PTR_ERR(pwm->mmio_base); > > + > > + pwm->clk_main = devm_clk_get(&pdev->dev, "main"); > > + if (IS_ERR(pwm->clk_main)) > > + return PTR_ERR(pwm->clk_main); > > + pwm->clk_mm = devm_clk_get(&pdev->dev, "mm"); > > I think it's more readable if a blank line separates the above two > lines. OK. > > > + if (IS_ERR(pwm->clk_mm)) > > + return PTR_ERR(pwm->clk_mm); > > + > > + ret = clk_prepare_enable(pwm->clk_main); > > + if (ret < 0) > > + return ret; > > + ret = clk_prepare_enable(pwm->clk_mm); > > Same here. OK. > > > + if (ret < 0) { > > + clk_disable_unprepare(pwm->clk_main); > > + return ret; > > + } > > + > > + platform_set_drvdata(pdev, pwm); > > + > > + pwm->chip.dev = &pdev->dev; > > + pwm->chip.ops = &mtk_disp_pwm_ops; > > + pwm->chip.base = -1; > > + pwm->chip.npwm = NUM_PWM; > > + > > + ret = pwmchip_add(&pwm->chip); > > + if (ret < 0) { > > + dev_err(&pdev->dev, "pwmchip_add() failed: %d\n", ret); > > Shouldn't you disable the clocks here? I will fix it. > > > + return ret; > > + } > > + > > + return 0; > > +} > > + > > +static int mtk_disp_pwm_remove(struct platform_device *pdev) > > +{ > > + struct mtk_disp_pwm_chip *pc = platform_get_drvdata(pdev); > > + > > + if (WARN_ON(!pc)) > > + return -ENODEV; > > You can drop this. Better to let it crash so that you have to fix it if > you ever encounter it. I will think about it. > > > + > > + clk_disable_unprepare(pc->clk_main); > > + clk_disable_unprepare(pc->clk_mm); > > + > > + return pwmchip_remove(&pc->chip); > > +} > > + > > +static const struct of_device_id mtk_disp_pwm_of_match[] = { > > + { .compatible = "mediatek,mt6595-disp-pwm" }, > > + { } > > +}; > > + > > +MODULE_DEVICE_TABLE(of, mtk_disp_pwm_of_match); > > No blank line between the above two lines. OK. > > > + > > +static struct platform_driver mtk_disp_pwm_driver = { > > + .driver = { > > + .name = "mediatek-disp-pwm", > > + .owner = THIS_MODULE, > > + .of_match_table = mtk_disp_pwm_of_match, > > + }, > > + .probe = mtk_disp_pwm_probe, > > + .remove = mtk_disp_pwm_remove, > > +}; > > + > > +module_platform_driver(mtk_disp_pwm_driver); > > Same here. Also, no need to initialize .driver.owner, > module_platform_driver() does that for you. Got it. > > Thierry Thank for your suggestion. Regards, YH Huang -- 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/