From patchwork Sat Jun 3 23:26:52 2017 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Alexander Sverdlin X-Patchwork-Id: 9764557 Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork.web.codeaurora.org (Postfix) with ESMTP id 32E0360363 for ; Sat, 3 Jun 2017 23:28:43 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 22A7A280DE for ; Sat, 3 Jun 2017 23:28:43 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id 16BA928464; Sat, 3 Jun 2017 23:28:43 +0000 (UTC) X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on pdx-wl-mail.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-1.9 required=2.0 tests=BAYES_00, DKIM_ADSP_CUSTOM_MED, DKIM_SIGNED,DKIM_VALID,FREEMAIL_FROM autolearn=ham version=3.3.1 Received: from bombadil.infradead.org (bombadil.infradead.org [65.50.211.133]) (using TLSv1.2 with cipher AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.wl.linuxfoundation.org (Postfix) with ESMTPS id 28BA7280DE for ; Sat, 3 Jun 2017 23:28:42 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=lists.infradead.org; s=bombadil.20170209; h=Sender: Content-Transfer-Encoding:Content-Type:MIME-Version:Cc:List-Subscribe: List-Help:List-Post:List-Archive:List-Unsubscribe:List-Id:References: In-Reply-To:Message-Id:Date:Subject:To:From:Reply-To:Content-ID: Content-Description:Resent-Date:Resent-From:Resent-Sender:Resent-To:Resent-Cc :Resent-Message-ID:List-Owner; bh=vD/YFFr/LmVx49fJT7Byd3H1pnK/B/HshpH40ksokEs=; b=tdca8Yq2Sd8FlUBv5r09UcmYxZ ZApntrS3sf8h9uknPR+EhSv3y9zUyRCRMlJWerEaZkPHKXfgsziT6D/S/RqYei92WwdBhE1rVRWAw 1cLPso3rprL6mfc6KrCKi6t8a/vSNT1htSqV1e7Z/XBseVjJP/llxWde9hWoYfzRUF92SVVk8hBDC dr29cIlq9lqpUsnUbG3LE/ZgyL3O0ep/q5v1yEUT7r1DV7pzTSOQZuPxHAX0lIrePiP9asJlZ3JSe SlfMv6WvgQyr/lTcyiBiJFoq/lFpsePNUu8cWUAfKyY9d5XdUfZ9iD6it+3FQm6gBc9GwXyXP+lJY ahMSE4/A==; Received: from localhost ([127.0.0.1] helo=bombadil.infradead.org) by bombadil.infradead.org with esmtp (Exim 4.87 #1 (Red Hat Linux)) id 1dHITI-0001pO-0h; Sat, 03 Jun 2017 23:28:40 +0000 Received: from mail-lf0-x244.google.com ([2a00:1450:4010:c07::244]) by bombadil.infradead.org with esmtps (Exim 4.87 #1 (Red Hat Linux)) id 1dHISg-0000cD-QX for linux-arm-kernel@lists.infradead.org; Sat, 03 Jun 2017 23:28:09 +0000 Received: by mail-lf0-x244.google.com with SMTP id v20so4149156lfa.2 for ; Sat, 03 Jun 2017 16:27:41 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=from:to:cc:subject:date:message-id:in-reply-to:references; bh=Khw07o/yJjtRBZHvigA/tliVHSh+NDkKZlsq1fz1gn4=; b=urnqIQDn4EhBnTeJYFmZpdxIyfjvjHLe6FBi+a49+pDTP8XmLkAZZqvazO0AZaKoPd MenfHcv0Nf3Hla6AbObyny7TmCl1TvbulLXosccU5xe0OOZy73X+HdnUFJQK5/EQJd1Y iwb0U3KeZ/svIPr0sR6cYkwcZxSCNlM0zv+SFs7uxcUMxlqaAmpu8YWSpGeleijHS4o6 fhX1OTVm287PMtHWe50WHXgJA3e1+9JvromLXwt7VmADVIVlu/X640skbcL6zoCYsMgW iS/y+x3Gb1Wa/TYflnesBRMd4XHO4I93leWNNukY8fHysVF3Y5AxAH17d5JfIv6hofLR 7wWw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:to:cc:subject:date:message-id:in-reply-to :references; bh=Khw07o/yJjtRBZHvigA/tliVHSh+NDkKZlsq1fz1gn4=; b=dGRnIlwGjW/qLc/XxAbql2QR1sCS0HCuUdNoxKodl0o+eBRz1+vIw+3TCcggct4lT2 lytQY3uJHXo76wS+rqx49JCVtum9KOwnZVtT82r3+Ku4wCfUxvA7SVlGA4qQ0zz5R4tW 3/DDLhM2gxKHjyErTlZznOC6pKiksw+xjEPFfAlLCOwtCMcQrCFiLpW+k48uzXDN/NUl Qih29vP2DtVI1NdbbIRIWc6SSm2tCa5rakH1gVD2yqYKjf1CnVdE8YgnakK+2FTQmdlb R/7LOuTM7X84opDl9tlpA906kiEnM439/8yL38B+jND2Hi5m8ZpSebK0iwglZXADPP7l RRww== X-Gm-Message-State: AODbwcA12KrMBq6ycm7vfcRRB+7axbXKQCemcARzC/e9DsEAbFw25gAk yYQfcBMSgD3izrtT X-Received: by 10.25.181.86 with SMTP id e83mr3667287lff.98.1496532460291; Sat, 03 Jun 2017 16:27:40 -0700 (PDT) Received: from giga1.localdomain ([195.245.52.203]) by smtp.gmail.com with ESMTPSA id 10sm1078869ljf.36.2017.06.03.16.27.38 (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Sat, 03 Jun 2017 16:27:39 -0700 (PDT) From: Alexander Sverdlin To: linux-iio@vger.kernel.org, linux-arm-kernel@lists.infradead.org Subject: [PATCH v5 5/5] iio: adc: New driver for Cirrus Logic EP93xx ADC Date: Sun, 4 Jun 2017 01:26:52 +0200 Message-Id: <20170603232652.21401-6-alexander.sverdlin@gmail.com> X-Mailer: git-send-email 2.12.2 In-Reply-To: <20170603232652.21401-1-alexander.sverdlin@gmail.com> References: <20170603232652.21401-1-alexander.sverdlin@gmail.com> X-CRM114-Version: 20100106-BlameMichelson ( TRE 0.8.0 (BSD) ) MR-646709E3 X-CRM114-CacheID: sfid-20170603_162804_305420_88F1B730 X-CRM114-Status: GOOD ( 25.88 ) X-BeenThere: linux-arm-kernel@lists.infradead.org X-Mailman-Version: 2.1.21 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: Lars-Peter Clausen , Ryan Mallon , Hartley Sweeten , Peter Meerwald-Stadler , Hartmut Knaack , Alexander Sverdlin , Jonathan Cameron MIME-Version: 1.0 Sender: "linux-arm-kernel" Errors-To: linux-arm-kernel-bounces+patchwork-linux-arm=patchwork.kernel.org@lists.infradead.org X-Virus-Scanned: ClamAV using ClamSMTP New driver adding support for ADC found on Cirrus Logic EP93xx series of SoCs. Board specific code must take care to create plaform device with all necessary resources. Signed-off-by: Alexander Sverdlin Acked-by: Jonathan Cameron --- Documentation/iio/ep93xx_adc.txt | 29 +++++ drivers/iio/adc/Kconfig | 11 ++ drivers/iio/adc/Makefile | 1 + drivers/iio/adc/ep93xx_adc.c | 255 +++++++++++++++++++++++++++++++++++++++ 4 files changed, 296 insertions(+) create mode 100644 Documentation/iio/ep93xx_adc.txt create mode 100644 drivers/iio/adc/ep93xx_adc.c diff --git a/Documentation/iio/ep93xx_adc.txt b/Documentation/iio/ep93xx_adc.txt new file mode 100644 index 000000000000..23053e7817bd --- /dev/null +++ b/Documentation/iio/ep93xx_adc.txt @@ -0,0 +1,29 @@ +Cirrus Logic EP93xx ADC driver. + +1. Overview + +The driver is intended to work on both low-end (EP9301, EP9302) devices with +5-channel ADC and high-end (EP9307, EP9312, EP9315) devices with 10-channel +touchscreen/ADC module. + +2. Channel numbering + +Numbering scheme for channels 0..4 is defined in EP9301 and EP9302 datasheets. +EP9307, EP9312 and EP9312 have 3 channels more (total 8), but the numbering is +not defined. So the last three are numbered randomly, let's say. + +Assuming ep93xx_adc is IIO device0, you'd find the following entries under +/sys/bus/iio/devices/iio:device0/: + + +-----------------+---------------+ + | sysfs entry | ball/pin name | + +-----------------+---------------+ + | in_voltage0_raw | YM | + | in_voltage1_raw | SXP | + | in_voltage2_raw | SXM | + | in_voltage3_raw | SYP | + | in_voltage4_raw | SYM | + | in_voltage5_raw | XP | + | in_voltage6_raw | XM | + | in_voltage7_raw | YP | + +-----------------+---------------+ diff --git a/drivers/iio/adc/Kconfig b/drivers/iio/adc/Kconfig index 401f47b51d83..a3eba810292a 100644 --- a/drivers/iio/adc/Kconfig +++ b/drivers/iio/adc/Kconfig @@ -249,6 +249,17 @@ config ENVELOPE_DETECTOR To compile this driver as a module, choose M here: the module will be called envelope-detector. +config EP93XX_ADC + tristate "Cirrus Logic EP93XX ADC driver" + depends on ARCH_EP93XX + help + Driver for the ADC module on the EP93XX series of SoC from Cirrus Logic. + It's recommended to switch on CONFIG_HIGH_RES_TIMERS option, in this + case driver will reduce its CPU usage by 90% in some use cases. + + To compile this driver as a module, choose M here: the module will be + called ep93xx_adc. + config EXYNOS_ADC tristate "Exynos ADC driver support" depends on ARCH_EXYNOS || ARCH_S3C24XX || ARCH_S3C64XX || (OF && COMPILE_TEST) diff --git a/drivers/iio/adc/Makefile b/drivers/iio/adc/Makefile index 9339bec4babe..d1a036d38486 100644 --- a/drivers/iio/adc/Makefile +++ b/drivers/iio/adc/Makefile @@ -25,6 +25,7 @@ obj-$(CONFIG_CC10001_ADC) += cc10001_adc.o obj-$(CONFIG_CPCAP_ADC) += cpcap-adc.o obj-$(CONFIG_DA9150_GPADC) += da9150-gpadc.o obj-$(CONFIG_ENVELOPE_DETECTOR) += envelope-detector.o +obj-$(CONFIG_EP93XX_ADC) += ep93xx_adc.o obj-$(CONFIG_EXYNOS_ADC) += exynos_adc.o obj-$(CONFIG_FSL_MX25_ADC) += fsl-imx25-gcq.o obj-$(CONFIG_HI8435) += hi8435.o diff --git a/drivers/iio/adc/ep93xx_adc.c b/drivers/iio/adc/ep93xx_adc.c new file mode 100644 index 000000000000..a179ac476c6d --- /dev/null +++ b/drivers/iio/adc/ep93xx_adc.c @@ -0,0 +1,255 @@ +/* + * Driver for ADC module on the Cirrus Logic EP93xx series of SoCs + * + * Copyright (C) 2015 Alexander Sverdlin + * + * 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. + * + * The driver uses polling to get the conversion status. According to EP93xx + * datasheets, reading ADCResult register starts the conversion, but user is also + * responsible for ensuring that delay between adjacent conversion triggers is + * long enough so that maximum allowed conversion rate is not exceeded. This + * basically renders IRQ mode unusable. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +/* + * This code could benefit from real HR Timers, but jiffy granularity would + * lower ADC conversion rate down to CONFIG_HZ, so we fallback to busy wait + * in such case. + * + * HR Timers-based version loads CPU only up to 10% during back to back ADC + * conversion, while busy wait-based version consumes whole CPU power. + */ +#ifdef CONFIG_HIGH_RES_TIMERS +#define ep93xx_adc_delay(usmin, usmax) usleep_range(usmin, usmax) +#else +#define ep93xx_adc_delay(usmin, usmax) udelay(usmin) +#endif + +#define EP93XX_ADC_RESULT 0x08 +#define EP93XX_ADC_SDR BIT(31) +#define EP93XX_ADC_SWITCH 0x18 +#define EP93XX_ADC_SW_LOCK 0x20 + +struct ep93xx_adc_priv { + struct clk *clk; + void __iomem *base; + int lastch; + struct mutex lock; +}; + +#define EP93XX_ADC_CH(index, dname, swcfg) { \ + .type = IIO_VOLTAGE, \ + .indexed = 1, \ + .channel = index, \ + .address = swcfg, \ + .datasheet_name = dname, \ + .info_mask_separate = BIT(IIO_CHAN_INFO_RAW), \ + .info_mask_shared_by_all = BIT(IIO_CHAN_INFO_SCALE) | \ + BIT(IIO_CHAN_INFO_OFFSET), \ +} + +/* + * Numbering scheme for channels 0..4 is defined in EP9301 and EP9302 datasheets. + * EP9307, EP9312 and EP9312 have 3 channels more (total 8), but the numbering is + * not defined. So the last three are numbered randomly, let's say. + */ +static const struct iio_chan_spec ep93xx_adc_channels[8] = { + EP93XX_ADC_CH(0, "YM", 0x608), + EP93XX_ADC_CH(1, "SXP", 0x680), + EP93XX_ADC_CH(2, "SXM", 0x640), + EP93XX_ADC_CH(3, "SYP", 0x620), + EP93XX_ADC_CH(4, "SYM", 0x610), + EP93XX_ADC_CH(5, "XP", 0x601), + EP93XX_ADC_CH(6, "XM", 0x602), + EP93XX_ADC_CH(7, "YP", 0x604), +}; + +static int ep93xx_read_raw(struct iio_dev *iiodev, + struct iio_chan_spec const *channel, int *value, + int *shift, long mask) +{ + struct ep93xx_adc_priv *priv = iio_priv(iiodev); + unsigned long timeout; + int ret; + + switch (mask) { + case IIO_CHAN_INFO_RAW: + mutex_lock(&priv->lock); + if (priv->lastch != channel->channel) { + priv->lastch = channel->channel; + /* + * Switch register is software-locked, unlocking must be + * immediately followed by write + */ + local_irq_disable(); + writel_relaxed(0xAA, priv->base + EP93XX_ADC_SW_LOCK); + writel_relaxed(channel->address, + priv->base + EP93XX_ADC_SWITCH); + local_irq_enable(); + /* + * Settling delay depends on module clock and could be + * 2ms or 500us + */ + ep93xx_adc_delay(2000, 2000); + } + /* Start the conversion, eventually discarding old result */ + readl_relaxed(priv->base + EP93XX_ADC_RESULT); + /* Ensure maximum conversion rate is not exceeded */ + ep93xx_adc_delay(DIV_ROUND_UP(1000000, 925), + DIV_ROUND_UP(1000000, 925)); + /* At this point conversion must be completed, but anyway... */ + ret = IIO_VAL_INT; + timeout = jiffies + msecs_to_jiffies(1) + 1; + while (1) { + u32 t; + + t = readl_relaxed(priv->base + EP93XX_ADC_RESULT); + if (t & EP93XX_ADC_SDR) { + *value = sign_extend32(t, 15); + break; + } + + if (time_after(jiffies, timeout)) { + dev_err(&iiodev->dev, "Conversion timeout\n"); + ret = -ETIMEDOUT; + break; + } + + cpu_relax(); + } + mutex_unlock(&priv->lock); + return ret; + + case IIO_CHAN_INFO_OFFSET: + /* According to datasheet, range is -25000..25000 */ + *value = 25000; + return IIO_VAL_INT; + + case IIO_CHAN_INFO_SCALE: + /* Typical supply voltage is 3.3v */ + *value = (1ULL << 32) * 3300 / 50000; + *shift = 32; + return IIO_VAL_FRACTIONAL_LOG2; + } + + return -EINVAL; +} + +static const struct iio_info ep93xx_adc_info = { + .driver_module = THIS_MODULE, + .read_raw = ep93xx_read_raw, +}; + +static int ep93xx_adc_probe(struct platform_device *pdev) +{ + int ret; + struct iio_dev *iiodev; + struct ep93xx_adc_priv *priv; + struct clk *pclk; + struct resource *res; + + iiodev = devm_iio_device_alloc(&pdev->dev, sizeof(*priv)); + if (!iiodev) + return -ENOMEM; + priv = iio_priv(iiodev); + + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + if (!res) { + dev_err(&pdev->dev, "Cannot obtain memory resource\n"); + return -ENXIO; + } + priv->base = devm_ioremap_resource(&pdev->dev, res); + if (IS_ERR(priv->base)) { + dev_err(&pdev->dev, "Cannot map memory resource\n"); + return PTR_ERR(priv->base); + } + + iiodev->dev.parent = &pdev->dev; + iiodev->name = dev_name(&pdev->dev); + iiodev->modes = INDIO_DIRECT_MODE; + iiodev->info = &ep93xx_adc_info; + iiodev->num_channels = ARRAY_SIZE(ep93xx_adc_channels); + iiodev->channels = ep93xx_adc_channels; + + priv->lastch = -1; + mutex_init(&priv->lock); + + platform_set_drvdata(pdev, iiodev); + + priv->clk = devm_clk_get(&pdev->dev, NULL); + if (IS_ERR(priv->clk)) { + dev_err(&pdev->dev, "Cannot obtain clock\n"); + return PTR_ERR(priv->clk); + } + + pclk = clk_get_parent(priv->clk); + if (!pclk) { + dev_warn(&pdev->dev, "Cannot obtain parent clock\n"); + } else { + /* + * This is actually a place for improvement: + * EP93xx ADC supports two clock divisors -- 4 and 16, + * resulting in conversion rates 3750 and 925 samples per second + * with 500us or 2ms settling time respectively. + * One might find this interesting enough to be configurable. + */ + ret = clk_set_rate(priv->clk, clk_get_rate(pclk) / 16); + if (ret) + dev_warn(&pdev->dev, "Cannot set clock rate\n"); + /* + * We can tolerate rate setting failure because the module should + * work in any case. + */ + } + + ret = clk_enable(priv->clk); + if (ret) { + dev_err(&pdev->dev, "Cannot enable clock\n"); + return ret; + } + + ret = iio_device_register(iiodev); + if (ret) + clk_disable(priv->clk); + + return ret; +} + +static int ep93xx_adc_remove(struct platform_device *pdev) +{ + struct iio_dev *iiodev = platform_get_drvdata(pdev); + struct ep93xx_adc_priv *priv = iio_priv(iiodev); + + iio_device_unregister(iiodev); + clk_disable(priv->clk); + + return 0; +} + +static struct platform_driver ep93xx_adc_driver = { + .driver = { + .name = "ep93xx-adc", + }, + .probe = ep93xx_adc_probe, + .remove = ep93xx_adc_remove, +}; +module_platform_driver(ep93xx_adc_driver); + +MODULE_AUTHOR("Alexander Sverdlin "); +MODULE_DESCRIPTION("Cirrus Logic EP93XX ADC driver"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS("platform:ep93xx-adc");