From patchwork Sun Aug 4 00:38:52 2019 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Linus Walleij X-Patchwork-Id: 1141631 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-gpio-owner@vger.kernel.org; receiver=) Authentication-Results: ozlabs.org; dmarc=pass (p=none dis=none) header.from=linaro.org Authentication-Results: ozlabs.org; dkim=pass (2048-bit key; unprotected) header.d=linaro.org header.i=@linaro.org header.b="Bh7MnHYJ"; dkim-atps=neutral Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by ozlabs.org (Postfix) with ESMTP id 461MWx68T6z9sBF for ; Sun, 4 Aug 2019 10:41:01 +1000 (AEST) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1729420AbfHDAlA (ORCPT ); Sat, 3 Aug 2019 20:41:00 -0400 Received: from mail-lj1-f193.google.com ([209.85.208.193]:42350 "EHLO mail-lj1-f193.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1729391AbfHDAlA (ORCPT ); Sat, 3 Aug 2019 20:41:00 -0400 Received: by mail-lj1-f193.google.com with SMTP id t28so76153676lje.9 for ; Sat, 03 Aug 2019 17:40:58 -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:mime-version :content-transfer-encoding; bh=83IVBUszXJBDBbbp8swTg62jU4/KejA7UpzPe0S6xm0=; b=Bh7MnHYJZl49goeYeBqiYnQPcskpwVWmg8V1GLR4XNU0Ag8CznRicpcqNiDtOVjsJX x7x3Ku5PctABAZ7hlhuiIU+dPypZ8LUIlhrxJ5VJJkVvCNrBlyFhuuO3phV273fqwwO5 D2TqK1a4qC4ktSTEjzQBpsbVbHD4l66lzyZZgzZoU/BS74shvxkBHvTUoIw9za2a3Eya h5bx8iFBpP/t0bCzyUVw1zWiRW7JCXUNWIzoFqzK1lkFAJZ9ve6G5DzmSt4NxJzpjo7i mjJvTC7ty6KDXaFMyDskhSbQ0XvVKGZUWyoVvETJ8GF5+lMRgmF8GepMDkths/ljM8dL DTGg== 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:mime-version :content-transfer-encoding; bh=83IVBUszXJBDBbbp8swTg62jU4/KejA7UpzPe0S6xm0=; b=DgtGT1cOwPrvlz1v3WJ5JPKyLoeITXocIVyvarNvBJ+V/wf9Nc2Mg95OWEL5c5MxFv sf26Zr1T23Rqv7RUbNEpAau2lU43Kh8dr1SvAIwuL82J3tVTFLoX/VF+OysbQsS3LRo7 hzGaY0ERp2zBgISFB3C2ZYIEBd1y3hZPpQBUVjbwt599zzu8D18GZDVX8NquQypo7I5V DlUOFzNoIDfoyuhLohFsK/i7+GDvX8t+8ofU1ySOAto/pe7wvwLRVKlBcHCI6c5A53Y+ t9qnP7rXBEFjL65sglEsI8S3Hc92KWBMMjF3gTxxmohL5+fBaVJmJmO4lna5XXaZoA0C WqKA== X-Gm-Message-State: APjAAAVpOVpsvDS+TcpcCkhs1uj8rgQfNGSC5YICEnBCwpTW5o1ph/4e 0mJorqWS85h52cNjD1NtR+Dviw== X-Google-Smtp-Source: APXvYqxD+Z5r+DO33M2OTj+KZ/rbnBZZiu/iWM+krMOc8IkE+sqfU8ExW8SIln5EB5HcVOEznTj6Ew== X-Received: by 2002:a2e:8696:: with SMTP id l22mr10773261lji.201.1564879257882; Sat, 03 Aug 2019 17:40:57 -0700 (PDT) Received: from localhost.localdomain (c-2ccd225c.014-348-6c756e10.bbcust.telenor.se. [92.34.205.44]) by smtp.gmail.com with ESMTPSA id k23sm12363158ljg.90.2019.08.03.17.40.56 (version=TLS1_3 cipher=AEAD-AES256-GCM-SHA384 bits=256/256); Sat, 03 Aug 2019 17:40:56 -0700 (PDT) From: Linus Walleij To: Mark Brown , linux-spi@vger.kernel.org Cc: linux-gpio@vger.kernel.org, Bartosz Golaszewski , Linus Walleij , Lukas Wunner , Stefan Wahren , Martin Sperl , Chris Boot Subject: [PATCH v4] spi: bcm2835: Convert to use CS GPIO descriptors Date: Sun, 4 Aug 2019 02:38:52 +0200 Message-Id: <20190804003852.1312-1-linus.walleij@linaro.org> X-Mailer: git-send-email 2.21.0 MIME-Version: 1.0 Sender: linux-gpio-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-gpio@vger.kernel.org This converts the BCM2835 SPI master driver to use GPIO descriptors for chip select handling. The BCM2835 driver was relying on the core to drive the CS high/low so very small changes were needed for this part. If it managed to request the CS from the device tree node, all is pretty straight forward. However for native GPIOs this driver has a quite unorthodox loopback to request some GPIOs from the SoC GPIO chip by looking it up from the device tree using gpiochip_find() and then offseting hard into its numberspace. This has been augmented a bit by using gpiochip_request_own_desc() but this code really needs to be verified. If "native CS" is actually an SoC GPIO, why is it even done this way? Should this GPIO not just be defined in the device tree like any other CS GPIO? I'm confused. Cc: Lukas Wunner Cc: Stefan Wahren Cc: Martin Sperl Cc: Chris Boot Signed-off-by: Linus Walleij --- ChangeLog v3->v4: - Fix the offset of the chipselect line to be 8 - CS as in the original code. - Use the modified gpiochip_request_own_desc() to set up line inversion semantics if need be. Look at the OF node of the SPI device for flags. ChangeLog v2->v3: - Fix unused variable "err" compile-time message. ChangeLog RFT->v2: - Rebased on v5.1-rc1 I would very much appreciate if someone took this for a ride on top of linux-next (there are some fixes in the -rcs you need) and see if all still works as expected. --- drivers/spi/spi-bcm2835.c | 58 +++++++++++++++++++++++++++------------ 1 file changed, 40 insertions(+), 18 deletions(-) diff --git a/drivers/spi/spi-bcm2835.c b/drivers/spi/spi-bcm2835.c index 6f243a90c844..a40c09c553d3 100644 --- a/drivers/spi/spi-bcm2835.c +++ b/drivers/spi/spi-bcm2835.c @@ -25,7 +25,9 @@ #include #include #include -#include +#include +#include /* FIXME: using chip internals */ +#include /* FIXME: using chip internals */ #include #include @@ -930,14 +932,19 @@ static int chip_match_name(struct gpio_chip *chip, void *data) static int bcm2835_spi_setup(struct spi_device *spi) { - int err; struct gpio_chip *chip; + enum gpio_lookup_flags lflags; + /* * sanity checking the native-chipselects */ if (spi->mode & SPI_NO_CS) return 0; - if (gpio_is_valid(spi->cs_gpio)) + /* + * The SPI core has successfully requested the CS GPIO line from the + * device tree, so we are done. + */ + if (spi->cs_gpiod) return 0; if (spi->chip_select > 1) { /* error in the case of native CS requested with CS > 1 @@ -948,29 +955,43 @@ static int bcm2835_spi_setup(struct spi_device *spi) "setup: only two native chip-selects are supported\n"); return -EINVAL; } - /* now translate native cs to GPIO */ + + /* + * Translate native CS to GPIO + * + * FIXME: poking around in the gpiolib internals like this is + * not very good practice. Find a way to locate the real problem + * and fix it. Why is the GPIO descriptor in spi->cs_gpiod + * sometimes not assigned correctly? Erroneous device trees? + */ /* get the gpio chip for the base */ chip = gpiochip_find("pinctrl-bcm2835", chip_match_name); if (!chip) return 0; - /* and calculate the real CS */ - spi->cs_gpio = chip->base + 8 - spi->chip_select; + /* + * Retrieve the corresponding GPIO line used for CS. + * The inversion semantics will be handled by the GPIO core + * code, so we pass GPIOS_OUT_LOW for "unasserted" and + * the correct flag for inversion semantics. The SPI_CS_HIGH + * on spi->mode cannot be checked for polarity in this case + * as the flag use_gpio_descriptors enforces SPI_CS_HIGH. + */ + if (of_property_read_bool(spi->dev.of_node, "spi-cs-high")) + lflags = GPIO_ACTIVE_HIGH; + else + lflags = GPIO_ACTIVE_LOW; + spi->cs_gpiod = gpiochip_request_own_desc(chip, 8 - spi->chip_select, + DRV_NAME, + lflags, + GPIOD_OUT_LOW); + if (IS_ERR(spi->cs_gpiod)) + return PTR_ERR(spi->cs_gpiod); /* and set up the "mode" and level */ - dev_info(&spi->dev, "setting up native-CS%i as GPIO %i\n", - spi->chip_select, spi->cs_gpio); - - /* set up GPIO as output and pull to the correct level */ - err = gpio_direction_output(spi->cs_gpio, - (spi->mode & SPI_CS_HIGH) ? 0 : 1); - if (err) { - dev_err(&spi->dev, - "could not set CS%i gpio %i as output: %i", - spi->chip_select, spi->cs_gpio, err); - return err; - } + dev_info(&spi->dev, "setting up native-CS%i to use GPIO\n", + spi->chip_select); return 0; } @@ -988,6 +1009,7 @@ static int bcm2835_spi_probe(struct platform_device *pdev) platform_set_drvdata(pdev, ctlr); + ctlr->use_gpio_descriptors = true; ctlr->mode_bits = BCM2835_SPI_MODE_BITS; ctlr->bits_per_word_mask = SPI_BPW_MASK(8); ctlr->num_chipselect = 3;