From patchwork Thu Oct 5 11:03:06 2017 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Linus Walleij X-Patchwork-Id: 821707 Return-Path: X-Original-To: incoming@patchwork.ozlabs.org Delivered-To: patchwork-incoming@bilbo.ozlabs.org Authentication-Results: ozlabs.org; spf=none (mailfrom) smtp.mailfrom=vger.kernel.org (client-ip=209.132.180.67; helo=vger.kernel.org; envelope-from=linux-i2c-owner@vger.kernel.org; receiver=) Authentication-Results: ozlabs.org; dkim=pass (1024-bit key; unprotected) header.d=linaro.org header.i=@linaro.org header.b="EJn6RLfT"; dkim-atps=neutral Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by ozlabs.org (Postfix) with ESMTP id 3y78yb1Dpzz9sRq for ; Thu, 5 Oct 2017 22:03:35 +1100 (AEDT) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751893AbdJELDe (ORCPT ); Thu, 5 Oct 2017 07:03:34 -0400 Received: from mail-wm0-f49.google.com ([74.125.82.49]:55806 "EHLO mail-wm0-f49.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751890AbdJELDd (ORCPT ); Thu, 5 Oct 2017 07:03:33 -0400 Received: by mail-wm0-f49.google.com with SMTP id u138so1404556wmu.4 for ; Thu, 05 Oct 2017 04:03:33 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linaro.org; s=google; h=from:to:cc:subject:date:message-id:in-reply-to:references; bh=au8fLN3Plt58pYHTTMVj7TGv6MUyhxMXJA+ajTAj/V0=; b=EJn6RLfTxa4DnBiN7X8p/f1Tu8S0yAuM4IcF4H8wILIhTb/irYjiu+oZHsasGJ1C5Z x34e5k1S76rcOgsRASAxilawYxjaW7eGdSD6U7EthJrc0ooFlWyvddplxeKWjSEQVUrz 5WnvwJvuqtaXeLgkroaPVyZhnA5leF3JP50Po= 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=au8fLN3Plt58pYHTTMVj7TGv6MUyhxMXJA+ajTAj/V0=; b=Y0ohYPDCx/MTwTUcQiBHLZk43GQmBqe2qO2aqwSYd56eOAmhJHofhxp3jTSzeGEz3M Pyl/BNHp2/1C1vjbMi3IBvpfkiXjOqmZvehGFoINPGIGUTDv39EllRniMegoTur3B0+B Q9pGjNrhQkuHSfTaXuNN+43OWRXUetovlROOZOm5rfnXYSD7CAlQpONIMJQyxCTrhrtX IJ8HJbl2es7MmysbVaLZV/ObX6WdJCaekBEh3pT5jiKqiPmkg+/QC73az/9PDhGvrNCn LwUgq4Y7F42tT6Plgnoht8w6DaSNSaMNcW9tbBt58f8tBKEIDmR8IFEwd5FnnzGJwQt0 2KcQ== X-Gm-Message-State: AMCzsaUp4vNojZKptCSmnhL11DQ/1j1wV7+dsajEw/RZDJ+jpPmA1Qyl kX3I1+kmlsjXjpLnBPZm6Bzm7Ajrc2Q= X-Google-Smtp-Source: AOwi7QBekFd3rK0zhSiGJv0HbSV/E+0dJzUp/l1DIjnhiqFWmZX0E+0V2gWsUez2E/mrITmeMi9VKQ== X-Received: by 10.28.129.197 with SMTP id c188mr8354190wmd.58.1507201412245; Thu, 05 Oct 2017 04:03:32 -0700 (PDT) Received: from genomnajs.local (catv-80-98-18-187.catv.broadband.hu. [80.98.18.187]) by smtp.gmail.com with ESMTPSA id e134sm17336478wma.31.2017.10.05.04.03.31 (version=TLS1_2 cipher=ECDHE-RSA-CHACHA20-POLY1305 bits=256/256); Thu, 05 Oct 2017 04:03:31 -0700 (PDT) From: Linus Walleij To: Wolfram Sang , linux-i2c@vger.kernel.org Cc: Linus Walleij Subject: [PATCH 3/7 v3] i2c: gpio: Enforce open drain through gpiolib Date: Thu, 5 Oct 2017 13:03:06 +0200 Message-Id: <20171005110310.5900-4-linus.walleij@linaro.org> X-Mailer: git-send-email 2.13.5 In-Reply-To: <20171005110310.5900-1-linus.walleij@linaro.org> References: <20171005110310.5900-1-linus.walleij@linaro.org> Sender: linux-i2c-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-i2c@vger.kernel.org The I2C GPIO bitbang driver currently emulates open drain behaviour by implementing what the gpiolib already does: not actively driving the line high, instead setting it to input. This makes no sense. Use the new facility in gpiolib to request the lines enforced into open drain mode, and let the open drain emulation already present in the gpiolib kick in and handle this. As a bonus: if the GPIO driver in the back-end actually supports open drain in hardware using the .set_config() callback, it will be utilized. That's correct: we never used that hardware feature before, instead relying on emulating open drain even if the GPIO controller could actually handle this for us. Users will sometimes get messages like this: gpio-485 (?): enforced open drain please flag it properly in DT/ACPI DSDT/board file gpio-486 (?): enforced open drain please flag it properly in DT/ACPI DSDT/board file i2c-gpio gpio-i2c: using lines 485 (SDA) and 486 (SCL) Which is completely proper: since the line is used as open drain, it should actually be flagged properly with e.g. gpios = <&gpio0 5 (GPIO_ACTIVE_HIGH|GPIO_OPEN_DRAIN)>, <&gpio0 6 (GPIO_ACTIVE_HIGH|GPIO_OPEN_DRAIN)>; Or similar facilities in board file descriptor tables or ACPI DSDT. Tested-by: Geert Uytterhoeven Signed-off-by: Linus Walleij --- ChangeLog v2->v3: - Rebased ChangeLog v1->v2: - Fix a minor typo in error path (copy-paste scl was sda) --- drivers/i2c/busses/i2c-gpio.c | 102 ++++++++++++++++-------------------------- 1 file changed, 39 insertions(+), 63 deletions(-) diff --git a/drivers/i2c/busses/i2c-gpio.c b/drivers/i2c/busses/i2c-gpio.c index b4664037eded..97b9c29e9429 100644 --- a/drivers/i2c/busses/i2c-gpio.c +++ b/drivers/i2c/busses/i2c-gpio.c @@ -25,23 +25,6 @@ struct i2c_gpio_private_data { struct i2c_gpio_platform_data pdata; }; -/* Toggle SDA by changing the direction of the pin */ -static void i2c_gpio_setsda_dir(void *data, int state) -{ - struct i2c_gpio_private_data *priv = data; - - /* - * This is a way of saying "do not drive - * me actively high" which means emulating open drain. - * The right way to do this is for gpiolib to - * handle this, by the function below. - */ - if (state) - gpiod_direction_input(priv->sda); - else - gpiod_direction_output(priv->sda, 0); -} - /* * Toggle SDA by changing the output value of the pin. This is only * valid for pins configured as open drain (i.e. setting the value @@ -54,17 +37,6 @@ static void i2c_gpio_setsda_val(void *data, int state) gpiod_set_value(priv->sda, state); } -/* Toggle SCL by changing the direction of the pin. */ -static void i2c_gpio_setscl_dir(void *data, int state) -{ - struct i2c_gpio_private_data *priv = data; - - if (state) - gpiod_direction_input(priv->scl); - else - gpiod_direction_output(priv->scl, 0); -} - /* * Toggle SCL by changing the output value of the pin. This is used * for pins that are configured as open drain and for output-only @@ -116,30 +88,13 @@ static int i2c_gpio_probe(struct platform_device *pdev) struct i2c_gpio_platform_data *pdata; struct i2c_algo_bit_data *bit_data; struct i2c_adapter *adap; + enum gpiod_flags gflags; int ret; priv = devm_kzalloc(&pdev->dev, sizeof(*priv), GFP_KERNEL); if (!priv) return -ENOMEM; - /* First get the GPIO pins; if it fails, we'll defer the probe. */ - priv->sda = devm_gpiod_get_index(&pdev->dev, NULL, 0, GPIOD_OUT_HIGH); - if (IS_ERR(priv->sda)) { - ret = PTR_ERR(priv->sda); - /* FIXME: hack in the old code, is this really necessary? */ - if (ret == -EINVAL) - ret = -EPROBE_DEFER; - return ret; - } - priv->scl = devm_gpiod_get_index(&pdev->dev, NULL, 1, GPIOD_OUT_LOW); - if (IS_ERR(priv->scl)) { - ret = PTR_ERR(priv->scl); - /* FIXME: hack in the old code, is this really necessary? */ - if (ret == -EINVAL) - ret = -EPROBE_DEFER; - return ret; - } - adap = &priv->adap; bit_data = &priv->bit_data; pdata = &priv->pdata; @@ -157,27 +112,48 @@ static int i2c_gpio_probe(struct platform_device *pdev) } /* - * FIXME: this is a hack emulating the open drain emulation - * that gpiolib can already do for us. Make all clients properly - * flag their lines as open drain and get rid of this property - * and the special callback. + * First get the GPIO pins; if it fails, we'll defer the probe. + * If the SDA line is marked from platform data or device tree as + * "open drain" it means something outside of our control is making + * this line being handled as open drain, and we should just handle + * it as any other output. Else we enforce open drain as this is + * required for an I2C bus. */ - if (pdata->sda_is_open_drain) { - gpiod_direction_output(priv->sda, 1); - bit_data->setsda = i2c_gpio_setsda_val; - } else { - gpiod_direction_input(priv->sda); - bit_data->setsda = i2c_gpio_setsda_dir; + if (pdata->sda_is_open_drain) + gflags = GPIOD_OUT_HIGH; + else + gflags = GPIOD_OUT_HIGH_OPEN_DRAIN; + priv->sda = devm_gpiod_get_index(&pdev->dev, NULL, 0, gflags); + if (IS_ERR(priv->sda)) { + ret = PTR_ERR(priv->sda); + /* FIXME: hack in the old code, is this really necessary? */ + if (ret == -EINVAL) + ret = -EPROBE_DEFER; + return ret; } - - if (pdata->scl_is_open_drain || pdata->scl_is_output_only) { - gpiod_direction_output(priv->scl, 1); - bit_data->setscl = i2c_gpio_setscl_val; - } else { - gpiod_direction_input(priv->scl); - bit_data->setscl = i2c_gpio_setscl_dir; + /* + * If the SCL line is marked from platform data or device tree as + * "open drain" it means something outside of our control is making + * this line being handled as open drain, and we should just handle + * it as any other output. Else we enforce open drain as this is + * required for an I2C bus. + */ + if (pdata->scl_is_open_drain) + gflags = GPIOD_OUT_LOW; + else + gflags = GPIOD_OUT_LOW_OPEN_DRAIN; + priv->scl = devm_gpiod_get_index(&pdev->dev, NULL, 1, gflags); + if (IS_ERR(priv->scl)) { + ret = PTR_ERR(priv->scl); + /* FIXME: hack in the old code, is this really necessary? */ + if (ret == -EINVAL) + ret = -EPROBE_DEFER; + return ret; } + bit_data->setsda = i2c_gpio_setsda_val; + bit_data->setscl = i2c_gpio_setscl_val; + if (!pdata->scl_is_output_only) bit_data->getscl = i2c_gpio_getscl; bit_data->getsda = i2c_gpio_getsda;