Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755414AbbG1JSE (ORCPT ); Tue, 28 Jul 2015 05:18:04 -0400 Received: from mail-by2on0131.outbound.protection.outlook.com ([207.46.100.131]:22990 "EHLO na01-by2-obe.outbound.protection.outlook.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1751865AbbG1JRw convert rfc822-to-8bit (ORCPT ); Tue, 28 Jul 2015 05:17:52 -0400 From: Chen Bough To: Dmitry Torokhov CC: "robh+dt@kernel.org" , "pawel.moll@arm.com" , "mark.rutland@arm.com" , "ijc+devicetree@hellion.org.uk" , "galak@codeaurora.org" , "shawnguo@kernel.org" , "kernel@pengutronix.de" , "linux@arm.linux.org.uk" , "hans.verkuil@cisco.com" , "hadess@hadess.net" , "mchehab@osg.samsung.com" , "mamlinav@gmail.com" , "arnd@arndb.de" , "jonathar@broadcom.com" , "hdegoede@redhat.com" , "christian.gmeiner@gmail.com" , "scott.liu@emc.com.tw" , "geert@linux-m68k.org" , "benjamin.tissoires@redhat.com" , "sebastien.szymanski@armadeus.com" , "sbranden@broadcom.com" , "devicetree@vger.kernel.org" , "linux-kernel@vger.kernel.org" , "linux-arm-kernel@lists.infradead.org" , "linux-input@vger.kernel.org" Subject: RE: [PATCH 1/5] input: touchscreen: add imx6ul_tsc driver support Thread-Topic: [PATCH 1/5] input: touchscreen: add imx6ul_tsc driver support Thread-Index: AQHQyF685o+CIxTjW0isptLfNqyAiZ3v39+AgAC0cZA= Date: Tue, 28 Jul 2015 09:17:49 +0000 Message-ID: References: <1437996407-27266-1-git-send-email-haibo.chen@freescale.com> <1437996407-27266-2-git-send-email-haibo.chen@freescale.com> <20150727220511.GC3613@dtor-ws> In-Reply-To: <20150727220511.GC3613@dtor-ws> Accept-Language: zh-CN, en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: authentication-results: gmail.com; dkim=none (message not signed) header.d=none; x-originating-ip: [199.59.226.141] x-microsoft-exchange-diagnostics: 1;BY1PR03MB1388;5:CrXVEmswdzptBePi1B3I6JMlihTEjri6b+NulbrGl8YFaCUl6kyfD3gUoOdpenHLUp094cM3lCczuEht4tgattFLWAMmRS98FiX2N3kvu51wYL6HPr/7rSetLvjWMcvVyhYJ++eD4r4T8RZ+tWAa4g==;24:y+6HmnBIUzb6nkP1NfuylWLPo1pZvUQPQgPElWtawcB18PRFE+SqBmF+jqTxG6zyL4k5+IHi2Rp1CSZ4zVwxBfyRmJj0iKcteBUuKHv/DKc=;20:yMB8v1Pzn/+qoDdUFjYqA80uqefTVJtXyf1ISyY5bkrOux27hmIPPNF62tFeXkMSsuOlGMGiTeAQllzT3nIrpw== x-microsoft-antispam: UriScan:;BCL:0;PCL:0;RULEID:;SRVR:BY1PR03MB1388; by1pr03mb1388: X-MS-Exchange-Organization-RulesExecuted x-microsoft-antispam-prvs: x-exchange-antispam-report-test: UriScan:; x-exchange-antispam-report-cfa-test: BCL:0;PCL:0;RULEID:(601004)(5005006)(3002001);SRVR:BY1PR03MB1388;BCL:0;PCL:0;RULEID:;SRVR:BY1PR03MB1388; x-forefront-prvs: 06515DA04B x-forefront-antispam-report: SFV:NSPM;SFS:(10019020)(6009001)(13464003)(57704003)(24454002)(377454003)(50986999)(575784001)(2950100001)(102836002)(2900100001)(54356999)(66066001)(106116001)(74316001)(76576001)(40100003)(92566002)(77096005)(189998001)(76176999)(99286002)(86362001)(87936001)(77156002)(62966003)(19580395003)(5003600100002)(5002640100001)(46102003)(33656002)(110136002)(19580405001)(2656002)(5001960100002)(7059030)(579004);DIR:OUT;SFP:1102;SCL:1;SRVR:BY1PR03MB1388;H:BY1PR03MB1388.namprd03.prod.outlook.com;FPR:;SPF:None;MLV:sfv;LANG:en; Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 8BIT MIME-Version: 1.0 X-OriginatorOrg: freescale.com X-MS-Exchange-CrossTenant-originalarrivaltime: 28 Jul 2015 09:17:49.7585 (UTC) X-MS-Exchange-CrossTenant-fromentityheader: Hosted X-MS-Exchange-CrossTenant-id: 710a03f5-10f6-4d38-9ff4-a80b81da590d X-MS-Exchange-Transport-CrossTenantHeadersStamped: BY1PR03MB1388 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 20521 Lines: 667 Hi Dmitry, Thanks for your careful review! My comments is following. Best Regards Haibo Chen > -----Original Message----- > From: Dmitry Torokhov [mailto:dmitry.torokhov@gmail.com] > Sent: Tuesday, July 28, 2015 6:05 AM > To: Chen Haibo-B51421 > Cc: robh+dt@kernel.org; pawel.moll@arm.com; mark.rutland@arm.com; > ijc+devicetree@hellion.org.uk; galak@codeaurora.org; shawnguo@kernel.org; > kernel@pengutronix.de; linux@arm.linux.org.uk; hans.verkuil@cisco.com; > hadess@hadess.net; mchehab@osg.samsung.com; mamlinav@gmail.com; > arnd@arndb.de; jonathar@broadcom.com; hdegoede@redhat.com; > christian.gmeiner@gmail.com; scott.liu@emc.com.tw; geert@linux-m68k.org; > benjamin.tissoires@redhat.com; sebastien.szymanski@armadeus.com; > sbranden@broadcom.com; devicetree@vger.kernel.org; linux- > kernel@vger.kernel.org; linux-arm-kernel@lists.infradead.org; linux- > input@vger.kernel.org > Subject: Re: [PATCH 1/5] input: touchscreen: add imx6ul_tsc driver > support > > Hi Haibo, > > On Mon, Jul 27, 2015 at 07:26:43PM +0800, Haibo Chen wrote: > > Freescale i.MX6UL contains a internal touchscreen controller, this > > patch add a driver to support this controller. > > > > Signed-off-by: Haibo Chen > > --- > > drivers/input/touchscreen/Kconfig | 12 + > > drivers/input/touchscreen/Makefile | 1 + > > drivers/input/touchscreen/imx6ul_tsc.c | 510 > > +++++++++++++++++++++++++++++++++ > > 3 files changed, 523 insertions(+) > > create mode 100644 drivers/input/touchscreen/imx6ul_tsc.c > > > > diff --git a/drivers/input/touchscreen/Kconfig > > b/drivers/input/touchscreen/Kconfig > > index 5b272ba..32c300d 100644 > > --- a/drivers/input/touchscreen/Kconfig > > +++ b/drivers/input/touchscreen/Kconfig > > @@ -479,6 +479,18 @@ config TOUCHSCREEN_MTOUCH > > To compile this driver as a module, choose M here: the > > module will be called mtouch. > > > > +config TOUCHSCREEN_IMX6UL_TSC > > + tristate "Freescale i.MX6UL touchscreen controller" > > + depends on OF > > + help > > + Say Y here if you have a Freescale i.MX6UL, and want to > > + use the internal touchscreen controller. > > + > > + If unsure, say N. > > + > > + To compile this driver as a module, choose M here: the > > + moduel will be called imx6ul_tsc. > > + > > config TOUCHSCREEN_INEXIO > > tristate "iNexio serial touchscreens" > > select SERIO > > diff --git a/drivers/input/touchscreen/Makefile > > b/drivers/input/touchscreen/Makefile > > index c85aae2..9379b32 100644 > > --- a/drivers/input/touchscreen/Makefile > > +++ b/drivers/input/touchscreen/Makefile > > @@ -38,6 +38,7 @@ obj-$(CONFIG_TOUCHSCREEN_EGALAX) += egalax_ts.o > > obj-$(CONFIG_TOUCHSCREEN_FUJITSU) += fujitsu_ts.o > > obj-$(CONFIG_TOUCHSCREEN_GOODIX) += goodix.o > > obj-$(CONFIG_TOUCHSCREEN_ILI210X) += ili210x.o > > +obj-$(CONFIG_TOUCHSCREEN_IMX6UL_TSC) += imx6ul_tsc.o > > obj-$(CONFIG_TOUCHSCREEN_INEXIO) += inexio.o > > obj-$(CONFIG_TOUCHSCREEN_INTEL_MID) += intel-mid-touch.o > > obj-$(CONFIG_TOUCHSCREEN_IPROC) += bcm_iproc_tsc.o > > diff --git a/drivers/input/touchscreen/imx6ul_tsc.c > > b/drivers/input/touchscreen/imx6ul_tsc.c > > new file mode 100644 > > index 0000000..5a5a368 > > --- /dev/null > > +++ b/drivers/input/touchscreen/imx6ul_tsc.c > > @@ -0,0 +1,510 @@ > > +/* > > + * Freescale i.MX6UL touchscreen controller driver > > + * > > + * Copyright (C) 2015 Freescale Semiconductor, Inc. > > + * > > + * 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. > > + */ > > + > > +#include > > +#include > > +#include > > +#include > > +#include > > +#include > > +#include > > +#include > > +#include > > +#include > > +#include > > +#include > > +#include > > +#include > > +#include > > +#include > > + > > +/* ADC configuration registers field define */ > > +#define ADC_AIEN (0x1 << 7) > > +#define ADC_CONV_DISABLE 0x1F > > +#define ADC_CAL (0x1 << 7) > > +#define ADC_CALF 0x2 > > +#define ADC_12BIT_MODE (0x2 << 2) > > +#define ADC_IPG_CLK 0x00 > > +#define ADC_CLK_DIV_8 (0x03 << 5) > > +#define ADC_SHORT_SAMPLE_MODE (0x0 << 4) > > +#define ADC_HARDWARE_TRIGGER (0x1 << 13) > > +#define SELECT_CHANNEL_4 0x04 > > +#define SELECT_CHANNEL_1 0x01 > > +#define DISABLE_CONVERSION_INT (0x0 << 7) > > + > > +/* ADC registers */ > > +#define REG_ADC_HC0 0x00 > > +#define REG_ADC_HC1 0x04 > > +#define REG_ADC_HC2 0x08 > > +#define REG_ADC_HC3 0x0C > > +#define REG_ADC_HC4 0x10 > > +#define REG_ADC_HS 0x14 > > +#define REG_ADC_R0 0x18 > > +#define REG_ADC_CFG 0x2C > > +#define REG_ADC_GC 0x30 > > +#define REG_ADC_GS 0x34 > > + > > +#define ADC_TIMEOUT msecs_to_jiffies(100) > > + > > +/* TSC registers */ > > +#define REG_TSC_BASIC_SETING 0x00 > > +#define REG_TSC_PRE_CHARGE_TIME 0x10 > > +#define REG_TSC_FLOW_CONTROL 0x20 > > +#define REG_TSC_MEASURE_VALUE 0x30 > > +#define REG_TSC_INT_EN 0x40 > > +#define REG_TSC_INT_SIG_EN 0x50 > > +#define REG_TSC_INT_STATUS 0x60 > > +#define REG_TSC_DEBUG_MODE 0x70 > > +#define REG_TSC_DEBUG_MODE2 0x80 > > + > > +/* TSC configuration registers field define */ > > +#define DETECT_4_WIRE_MODE (0x0 << 4) > > +#define AUTO_MEASURE 0x1 > > +#define MEASURE_SIGNAL 0x1 > > +#define DETECT_SIGNAL (0x1 << 4) > > +#define VALID_SIGNAL (0x1 << 8) > > +#define MEASURE_INT_EN 0x1 > > +#define MEASURE_SIG_EN 0x1 > > +#define VALID_SIG_EN (0x1 << 8) > > +#define DE_GLITCH_2 (0x2 << 29) > > +#define START_SENSE (0x1 << 12) > > +#define TSC_DISABLE (0x1 << 16) > > +#define DETECT_MODE 0x2 > > + > > +struct imx6ul_tsc { > > + struct device *dev; > > + struct input_dev *input; > > + void __iomem *tsc_regs; > > + void __iomem *adc_regs; > > + struct clk *tsc_clk; > > + struct clk *adc_clk; > > + > > + int tsc_irq; > > + int adc_irq; > > You do not need to store tsc_irq and adc_irq since you do not reference > them anywhere but in probe(). > [haibo] yes, you are right, I will remove these and just add these values in probe() > > + int value; > > Why do we need this? [haibo] in tsc_irq(), I use this value to record the tsc measure value. But now I find the value just used In tsc_irq, so I will remove this value out of imx6ul_tsc. > > > + int xnur_gpio; > > + int measure_delay_time; > > + int pre_charge_time; > > + > > + struct completion completion; > > +}; > > + > > +/* > > + * TSC module need ADC to get the measure value. So > > + * before config TSC, we should initialize ADC module. > > + */ > > +static void imx6ul_adc_init(struct imx6ul_tsc *tsc) { > > + int adc_hc = 0; > > + int adc_gc; > > + int adc_gs; > > + int adc_cfg; > > + int timeout; > > + > > + adc_cfg = readl(tsc->adc_regs + REG_ADC_CFG); > > + adc_cfg |= ADC_12BIT_MODE | ADC_IPG_CLK; > > + adc_cfg |= ADC_CLK_DIV_8 | ADC_SHORT_SAMPLE_MODE; > > + adc_cfg &= ~ADC_HARDWARE_TRIGGER; > > + writel(adc_cfg, tsc->adc_regs + REG_ADC_CFG); > > + > > + /* enable calibration interrupt */ > > + adc_hc |= ADC_AIEN; > > + adc_hc |= ADC_CONV_DISABLE; > > + writel(adc_hc, tsc->adc_regs + REG_ADC_HC0); > > + > > + /* start ADC calibration */ > > + adc_gc = readl(tsc->adc_regs + REG_ADC_GC); > > + adc_gc |= ADC_CAL; > > + writel(adc_gc, tsc->adc_regs + REG_ADC_GC); > > + > > + timeout = wait_for_completion_timeout > > + (&tsc->completion, ADC_TIMEOUT); > > + if (timeout == 0) > > + dev_err(tsc->dev, "Timeout for adc calibration\n"); > > + > > + adc_gs = readl(tsc->adc_regs + REG_ADC_GS); > > + if (adc_gs & ADC_CALF) > > + dev_err(tsc->dev, "ADC calibration failed\n"); > > + > > + /* TSC need the ADC work in hardware trigger */ > > + adc_cfg = readl(tsc->adc_regs + REG_ADC_CFG); > > + adc_cfg |= ADC_HARDWARE_TRIGGER; > > + writel(adc_cfg, tsc->adc_regs + REG_ADC_CFG); } > > + > > +/* > > + * This is a TSC workaround. Currently TSC misconnect two > > + * ADC channels, this function remap channel configure for > > + * hardware trigger. > > + */ > > +static void imx6ul_tsc_channel_config(struct imx6ul_tsc *tsc) { > > + int adc_hc0, adc_hc1, adc_hc2, adc_hc3, adc_hc4; > > + > > + adc_hc0 = DISABLE_CONVERSION_INT; > > + writel(adc_hc0, tsc->adc_regs + REG_ADC_HC0); > > + > > + adc_hc1 = DISABLE_CONVERSION_INT | SELECT_CHANNEL_4; > > + writel(adc_hc1, tsc->adc_regs + REG_ADC_HC1); > > + > > + adc_hc2 = DISABLE_CONVERSION_INT; > > + writel(adc_hc2, tsc->adc_regs + REG_ADC_HC2); > > + > > + adc_hc3 = DISABLE_CONVERSION_INT | SELECT_CHANNEL_1; > > + writel(adc_hc3, tsc->adc_regs + REG_ADC_HC3); > > + > > + adc_hc4 = DISABLE_CONVERSION_INT; > > + writel(adc_hc4, tsc->adc_regs + REG_ADC_HC4); } > > + > > +/* > > + * TSC setting, confige the pre-charge time and measure delay time. > > + * different touch screen may need different pre-charge time and > > + * measure delay time. > > + */ > > +static void imx6ul_tsc_set(struct imx6ul_tsc *tsc) { > > + int basic_setting = 0; > > + int start; > > + > > + basic_setting |= tsc->measure_delay_time << 8; > > + basic_setting |= DETECT_4_WIRE_MODE | AUTO_MEASURE; > > + writel(basic_setting, tsc->tsc_regs + REG_TSC_BASIC_SETING); > > + > > + writel(DE_GLITCH_2, tsc->tsc_regs + REG_TSC_DEBUG_MODE2); > > + > > + writel(tsc->pre_charge_time, tsc->tsc_regs + > REG_TSC_PRE_CHARGE_TIME); > > + writel(MEASURE_INT_EN, tsc->tsc_regs + REG_TSC_INT_EN); > > + writel(MEASURE_SIG_EN | VALID_SIG_EN, > > + tsc->tsc_regs + REG_TSC_INT_SIG_EN); > > + > > + /* start sense detection */ > > + start = readl(tsc->tsc_regs + REG_TSC_FLOW_CONTROL); > > + start |= START_SENSE; > > + start &= ~TSC_DISABLE; > > + writel(start, tsc->tsc_regs + REG_TSC_FLOW_CONTROL); } > > + > > +static void imx6ul_tsc_init(struct imx6ul_tsc *tsc) { > > + imx6ul_adc_init(tsc); > > + imx6ul_tsc_channel_config(tsc); > > + imx6ul_tsc_set(tsc); > > +} > > + > > +static irqreturn_t tsc_irq(int irq, void *dev_id) { > > + struct imx6ul_tsc *tsc = (struct imx6ul_tsc *)dev_id; > > + int status; > > + int x, y; > > + int xnur; > > + int debug_mode2; > > + int state_machine; > > + int start; > > + unsigned long timeout; > > + > > + status = readl(tsc->tsc_regs + REG_TSC_INT_STATUS); > > + > > + /* write 1 to clear the bit measure-signal */ > > + writel(MEASURE_SIGNAL | DETECT_SIGNAL, > > + tsc->tsc_regs + REG_TSC_INT_STATUS); > > + > > + /* It's a HW self-clean bit. Set this bit and start sense detection > */ > > + start = readl(tsc->tsc_regs + REG_TSC_FLOW_CONTROL); > > + start |= START_SENSE; > > + writel(start, tsc->tsc_regs + REG_TSC_FLOW_CONTROL); > > + > > + if (status & MEASURE_SIGNAL) { > > + tsc->value = readl(tsc->tsc_regs + REG_TSC_MEASURE_VALUE); > > + x = (tsc->value >> 16) & 0x0fff; > > + y = tsc->value & 0x0fff; > > + > > + /* > > + * Delay some time(max 2ms), wait the pre-charge done. > > + * After the pre-change mode, TSC go into detect mode. > > + * And in detect mode, we can get the xnur gpio value. > > + * If xnur is low, this means the touch screen still > > + * be touched. If xnur is high, this means finger leave > > + * the touch screen. > > + */ > > + timeout = jiffies + HZ/500; > > + do { > > + if (time_after(jiffies, timeout)) { > > + xnur = 0; > > + goto touch_event; > > + } > > + usleep_range(200, 400); > > + debug_mode2 = readl(tsc->tsc_regs + > REG_TSC_DEBUG_MODE2); > > + state_machine = (debug_mode2 >> 20) & 0x7; > > + } while (state_machine != DETECT_MODE); > > + usleep_range(200, 400); > > + > > + xnur = gpio_get_value(tsc->xnur_gpio); > > +touch_event: > > + if (xnur == 0) { > > + input_report_key(tsc->input, BTN_TOUCH, 1); > > + input_report_abs(tsc->input, ABS_X, x); > > + input_report_abs(tsc->input, ABS_Y, y); > > + } else > > + input_report_key(tsc->input, BTN_TOUCH, 0); > > + > > + input_sync(tsc->input); > > + } > > + > > + return IRQ_HANDLED; > > +} > > + > > +static irqreturn_t adc_irq(int irq, void *dev_id) { > > + struct imx6ul_tsc *tsc = (struct imx6ul_tsc *)dev_id; > > + int coco; > > + int value; > > + > > + coco = readl(tsc->adc_regs + REG_ADC_HS); > > + if (coco & 0x01) { > > + value = readl(tsc->adc_regs + REG_ADC_R0); > > + complete(&tsc->completion); > > + } > > + > > + return IRQ_HANDLED; > > +} > > + > > +static const struct of_device_id imx6ul_tsc_match[] = { > > + { .compatible = "fsl,imx6ul-tsc", }, > > + { /* sentinel */ } > > +}; > > +MODULE_DEVICE_TABLE(of, imx6ul_tsc_match); > > + > > +static int imx6ul_tsc_probe(struct platform_device *pdev) { > > + struct imx6ul_tsc *tsc; > > + struct resource *tsc_mem; > > + struct resource *adc_mem; > > + struct input_dev *input_dev; > > + struct device_node *np = pdev->dev.of_node; > > + int err; > > + > > + tsc = kzalloc(sizeof(struct imx6ul_tsc), GFP_KERNEL); > > Why not devm_kzalloc()? > [haibo] It's better to use devm_kzalloc(), I will change this. > > + if (!tsc) { > > + err = -ENOMEM; > > + goto err_free_mem_tsc; > > + } > > + > > + input_dev = input_allocate_device(); > > devm_input_allocate_device(&pdev->dev)? > [haibo] will change this. > > > + if (!input_dev) { > > + err = -ENOMEM; > > + goto err_free_mem; > > + } > > + > > + tsc->dev = &pdev->dev; > > + > > + tsc->input = input_dev; > > + tsc->input->evbit[0] = BIT_MASK(EV_KEY) | BIT_MASK(EV_ABS); > > + tsc->input->keybit[BIT_WORD(BTN_TOUCH)] = BIT_MASK(BTN_TOUCH); > > + input_set_abs_params(tsc->input, ABS_X, 0, 0xFFF, 0, 0); > > + input_set_abs_params(tsc->input, ABS_Y, 0, 0xFFF, 0, 0); > > + > > + tsc->input->name = "iMX6UL TouchScreen Controller"; > > + > > + tsc->xnur_gpio = of_get_named_gpio(np, "xnur-gpio", 0); > > + err = gpio_request_one(tsc->xnur_gpio, GPIOF_IN, "tsc_X-"); > > + if (err) { > > + dev_err(&pdev->dev, "failed to request GPIO tsc_X-\n"); > > + goto err_free_mem; > > + } > > + > > + tsc_mem = platform_get_resource(pdev, IORESOURCE_MEM, 0); > > + tsc->tsc_regs = devm_ioremap_resource(&pdev->dev, tsc_mem); > > + if (IS_ERR(tsc->tsc_regs)) { > > + dev_err(&pdev->dev, "failed to remap tsc memory\n"); > > + err = PTR_ERR(tsc->tsc_regs); > > + goto err_free_mem; > > + } > > + > > + adc_mem = platform_get_resource(pdev, IORESOURCE_MEM, 1); > > + tsc->adc_regs = devm_ioremap_resource(&pdev->dev, adc_mem); > > + if (IS_ERR(tsc->adc_regs)) { > > + dev_err(&pdev->dev, "failed to remap adc memory\n"); > > + err = PTR_ERR(tsc->adc_regs); > > + goto err_free_mem; > > + } > > + > > + tsc->tsc_clk = devm_clk_get(&pdev->dev, "tsc"); > > + if (IS_ERR(tsc->tsc_clk)) { > > + dev_err(&pdev->dev, "failed getting tsc clock\n"); > > + err = PTR_ERR(tsc->tsc_clk); > > + goto err_free_mem; > > + } > > + > > + tsc->adc_clk = devm_clk_get(&pdev->dev, "adc"); > > + if (IS_ERR(tsc->adc_clk)) { > > + dev_err(&pdev->dev, "failed getting adc clock\n"); > > + err = PTR_ERR(tsc->adc_clk); > > + goto err_free_mem; > > + } > > + > > + tsc->tsc_irq = platform_get_irq(pdev, 0); > > + if (tsc->tsc_irq <= 0) { > > + dev_err(&pdev->dev, "no tsc irq resource?\n"); > > + err = -EINVAL; > > + goto err_free_mem; > > + } > > + > > + tsc->adc_irq = platform_get_irq(pdev, 1); > > + if (tsc->adc_irq <= 0) { > > + dev_err(&pdev->dev, "no adc irq resource?\n"); > > + err = -EINVAL; > > I think platform_get_irq() return negative errors, so let's check for > tsc->adc_irq < 0 and do: > > return tsc->adc_irq? [haibo] Good suggestion! I will take this. > > > + goto err_free_mem; > > + } > > + > > + err = devm_request_threaded_irq(tsc->dev, tsc->tsc_irq, > > + NULL, tsc_irq, > > + IRQF_TRIGGER_HIGH | IRQF_ONESHOT, > > Do we need to hard-code the trigger? Maybe rely on board/OF code to set > up the trigger flags properly? > [haibo] this tsc is inside the imx6ul soc, it do not relay on board. > > + dev_name(&pdev->dev), tsc); > > + if (err < 0) { > > + dev_err(&pdev->dev, > > + "failed requesting tsc irq %d\n", > > + tsc->tsc_irq); > > + goto err_free_mem; > > + } > > + > > + err = devm_request_irq(tsc->dev, tsc->adc_irq, > > + adc_irq, 0, > > + dev_name(&pdev->dev), tsc); > > + if (err < 0) { > > + dev_err(&pdev->dev, > > + "failed requesting adc irq %d\n", > > + tsc->adc_irq); > > + goto err_free_mem; > > + } > > + > > + err = of_property_read_u32(np, "measure_delay_time", > > + &tsc->measure_delay_time); > > + if (err) > > + tsc->measure_delay_time = 0xffff; > > + > > + err = of_property_read_u32(np, "pre_charge_time", > > + &tsc->pre_charge_time); > > + if (err) > > + tsc->pre_charge_time = 0xfff; > > + > > + init_completion(&tsc->completion); > > + > > + err = clk_prepare_enable(tsc->adc_clk); > > + if (err) { > > + dev_err(&pdev->dev, > > + "Could not prepare or enable the adc clock.\n"); > > + goto err_free_mem; > > + } > > + > > + err = clk_prepare_enable(tsc->tsc_clk); > > + if (err) { > > + dev_err(&pdev->dev, > > + "Could not prepare or enable the tsc clock.\n"); > > + goto error_tsc_clk_enable; > > + } > > + > > + err = input_register_device(tsc->input); > > + if (err < 0) { > > + dev_err(&pdev->dev, "failed to register input device\n"); > > + err = -EIO; > > + goto err_input_register; > > + } > > + > > + imx6ul_tsc_init(tsc); > > + > > + platform_set_drvdata(pdev, tsc); > > + return 0; > > + > > +err_input_register: > > + clk_disable_unprepare(tsc->tsc_clk); > > +error_tsc_clk_enable: > > + clk_disable_unprepare(tsc->adc_clk); > > +err_free_mem: > > + input_free_device(tsc->input); > > +err_free_mem_tsc: > > + kfree(tsc); > > + return err; > > +} > > + > > +static int imx6ul_tsc_remove(struct platform_device *pdev) { > > + struct imx6ul_tsc *tsc = platform_get_drvdata(pdev); > > + > > + clk_disable_unprepare(tsc->tsc_clk); > > + clk_disable_unprepare(tsc->adc_clk); > > Just cutting clocks is not really nice. Why don't you shut down the > device as you do in suspend, before disabling the clocks. [haibo] Yes, you are right, it's my negligence. > > > + input_unregister_device(tsc->input); > > + kfree(tsc); > > + > > + return 0; > > +} > > + > > +#ifdef CONFIG_PM_SLEEP > > Drop ifdef. > > > +static int imx6ul_tsc_suspend(struct device *dev) > > __maybe_unused. > > > +{ > > + struct platform_device *pdev = to_platform_device(dev); > > + struct imx6ul_tsc *tsc = platform_get_drvdata(pdev); > > + int tsc_flow; > > + int adc_cfg; > > + > > + /* TSC controller enters to idle status */ > > + tsc_flow = readl(tsc->tsc_regs + REG_TSC_FLOW_CONTROL); > > + tsc_flow |= TSC_DISABLE; > > + writel(tsc_flow, tsc->tsc_regs + REG_TSC_FLOW_CONTROL); > > + > > + /* ADC controller enters to stop mode */ > > + adc_cfg = readl(tsc->adc_regs + REG_ADC_HC0); > > + adc_cfg |= ADC_CONV_DISABLE; > > + writel(adc_cfg, tsc->adc_regs + REG_ADC_HC0); > > + > > + clk_disable_unprepare(tsc->tsc_clk); > > + clk_disable_unprepare(tsc->adc_clk); > > + > > + return 0; > > +} > > + > > +static int imx6ul_tsc_resume(struct device *dev) > > __maybe_unused. > [haibo] will add this . > > +{ > > + struct platform_device *pdev = to_platform_device(dev); > > + struct imx6ul_tsc *tsc = platform_get_drvdata(pdev); > > + int err; > > + > > + err = clk_prepare_enable(tsc->adc_clk); > > + if (err) > > + return err; > > + > > + err = clk_prepare_enable(tsc->tsc_clk); > > + if (err) > > Disable tsc->adc_clk? > [haibo] will add this. > > + return err; > > + > > + imx6ul_tsc_init(tsc); > > + return 0; > > +} > > +#endif > > + > > +static SIMPLE_DEV_PM_OPS(imx6ul_tsc_pm_ops, > > + imx6ul_tsc_suspend, > > + imx6ul_tsc_resume); > > + > > +static struct platform_driver imx6ul_tsc_driver = { > > + .driver = { > > + .name = "imx6ul-tsc", > > + .owner = THIS_MODULE, > > + .of_match_table = imx6ul_tsc_match, > > + .pm = &imx6ul_tsc_pm_ops, > > + }, > > + .probe = imx6ul_tsc_probe, > > + .remove = imx6ul_tsc_remove, > > +}; > > +module_platform_driver(imx6ul_tsc_driver); > > + > > +MODULE_AUTHOR("Haibo Chen "); > > +MODULE_DESCRIPTION("Freescale i.MX6UL Touchscreen controller > > +driver"); MODULE_LICENSE("GPL v2"); > > -- > > 1.9.1 > > > > -- > Dmitry -- 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/