public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Trent Piepho <tpiepho@impinj.com>
To: "kyle.roeschley@ni.com" <kyle.roeschley@ni.com>,
	"broonie@kernel.org" <broonie@kernel.org>
Cc: "linux-spi@vger.kernel.org" <linux-spi@vger.kernel.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"geert@linux-m68k.org" <geert@linux-m68k.org>
Subject: Re: [PATCH] spi: Add a sysfs interface to instantiate devices
Date: Thu, 21 Dec 2017 21:05:43 +0000	[thread overview]
Message-ID: <1513890342.26695.4.camel@impinj.com> (raw)
In-Reply-To: <20171221200309.17967-1-kyle.roeschley@ni.com>

On Thu, 2017-12-21 at 14:03 -0600, Kyle Roeschley wrote:
> Add a sysfs interface to instantiate and delete SPI devices using the
> spidev driver. This can be used when developing a driver on a
> self-soldered board which doesn't yet have proper SPI device declaration
> at the platform level, and presumably for various debugging situations.
> 
> Inspired by 99cd8e25875a ("i2c: Add a sysfs interface to instantiate
> devices").

The i2c interface allows one to specify the type of device to create. 
Why must this interface be linked to spidev and only capable of
creating spidev devices?  

> 
> Signed-off-by: Kyle Roeschley <kyle.roeschley@ni.com>
> ---
>  Documentation/spi/spi-summary | 14 ++++++++
>  drivers/spi/spi.c             | 78 +++++++++++++++++++++++++++++++++++++++++++
>  include/linux/spi/spi.h       |  3 ++
>  3 files changed, 95 insertions(+)
> 
> diff --git a/Documentation/spi/spi-summary b/Documentation/spi/spi-summary
> index 1721c1b570c3..51d9747c4426 100644
> --- a/Documentation/spi/spi-summary
> +++ b/Documentation/spi/spi-summary
> @@ -339,6 +339,20 @@ up the spi bus master, and will likely need spi_new_device() to provide the
>  board info based on the board that was hotplugged.  Of course, you'd later
>  call at least spi_unregister_device() when that board is removed.
>  
> +Alternatively, a sysfs interface was added to let the user create devices which
> +using the spidev driver. This interface is made of 2 attribute files which are
> +created in every SPI master directory: new_device and delete_device. Both files
> +are write only and you must write the decimal SPI chip select number to them in
> +order to properly instantiate or delete a SPI device. As no two devices can be
> +attached to the same master with the same chip select line, the chip select
> +number is sufficient to uniquely identify the device to be deleted.
> +
> +Example:
> +# echo 1 > /sys/class/spi_master/spi0/new_device
> +
> +In general, this interface should only be used when in-kernel device
> +declaration can't be done.
> +
>  When Linux includes support for MMC/SD/SDIO/DataFlash cards through SPI, those
>  configurations will also be dynamic.  Fortunately, such devices all support
>  basic device identification probes, so they should hotplug normally.
> diff --git a/drivers/spi/spi.c b/drivers/spi/spi.c
> index b33a727a0158..648ccdf359f9 100644
> --- a/drivers/spi/spi.c
> +++ b/drivers/spi/spi.c
> @@ -242,8 +242,85 @@ static const struct attribute_group spi_controller_statistics_group = {
>  	.attrs  = spi_controller_statistics_attrs,
>  };
>  
> +static ssize_t
> +new_device_store(struct device *dev, struct device_attribute *attr,
> +		 const char *buf, size_t count)
> +{
> +	struct spi_controller *ctlr = container_of(dev, struct spi_controller,
> +						   dev);
> +	struct spi_device *spi;
> +	struct spi_board_info bi = {
> +		.modalias = "spidev",
> +		.max_speed_hz = ctlr->max_speed_hz,
> +	};
> +
> +	if (kstrtou16(buf, 0, &bi.chip_select) < 0)
> +		return -EINVAL;
> +
> +	spi = spi_new_device(ctlr, &bi);
> +	if (!spi) {
> +		dev_err(dev, "can't create new device\n");
> +		return -ENXIO;

I2C returns -EINVAL

> +	}
> +
> +	mutex_lock(&ctlr->bus_lock_mutex);
> +	list_add_tail(&spi->userspace_device, &ctlr->userspace_devices);
> +	mutex_unlock(&ctlr->bus_lock_mutex);
> +
> +	dev_info(dev, "created spidev device %s\n", dev_name(&spi->dev));
> +
> +	return count;
> +}
> +static DEVICE_ATTR_WO(new_device);
> +
> +static ssize_t
> +delete_device_store(struct device *dev, struct device_attribute *attr,
> +		    const char *buf, size_t count)
> +{
> +	struct spi_controller *ctlr = container_of(dev, struct spi_controller,
> +						   dev);
> +	struct spi_device *spi, *next;
> +	int ret = -ENXIO;
> +	u16 cs;
> +
> +	if (kstrtou16(buf, 0, &cs) < 0)
> +		return -EINVAL;
> +
> +	mutex_lock(&ctlr->bus_lock_mutex);
> +	list_for_each_entry_safe(spi, next, &ctlr->userspace_devices,
> +				 userspace_device) {
> +		if (spi->chip_select != cs)
> +			continue;
> +
> +		dev_info(dev, "deleting spidev device %s\n",
> +			 dev_name(&spi->dev));
> +		list_del(&spi->userspace_device);
> +		spi_unregister_device(spi);
> +		ret = count;
> +		break;
> +	}
> +	mutex_unlock(&ctlr->bus_lock_mutex);
> +
> +	if (ret == -ENXIO)
> +		dev_err(dev, "can't find spidev device %u in list\n", cs);
> +
> +	return ret;
> +}
> +static DEVICE_ATTR_WO(delete_device);
> +
> +static struct attribute *spi_controller_userspace_attrs[] = {
> +	&dev_attr_new_device.attr,
> +	&dev_attr_delete_device.attr,
> +	NULL,
> +};
> +
> +static const struct attribute_group spi_controller_userspace_group = {
> +	.attrs  = spi_controller_userspace_attrs,
> +};
> +
>  static const struct attribute_group *spi_master_groups[] = {
>  	&spi_controller_statistics_group,
> +	&spi_controller_userspace_group,
>  	NULL,
>  };
>  
> @@ -2129,6 +2206,7 @@ int spi_register_controller(struct spi_controller *ctlr)
>  			return id;
>  		ctlr->bus_num = id;
>  	}
> +	INIT_LIST_HEAD(&ctlr->userspace_devices);
>  	INIT_LIST_HEAD(&ctlr->queue);
>  	spin_lock_init(&ctlr->queue_lock);
>  	spin_lock_init(&ctlr->bus_lock_spinlock);
> diff --git a/include/linux/spi/spi.h b/include/linux/spi/spi.h
> index bc6bb325d1bf..f7255745326d 100644
> --- a/include/linux/spi/spi.h
> +++ b/include/linux/spi/spi.h
> @@ -172,6 +172,8 @@ struct spi_device {
>  	/* the statistics */
>  	struct spi_statistics	statistics;
>  
> +	struct list_head	userspace_device;
> +
>  	/*
>  	 * likely need more hooks for more protocol options affecting how
>  	 * the controller talks to each chip, like:
> @@ -410,6 +412,7 @@ struct spi_controller {
>  	struct device	dev;
>  
>  	struct list_head list;
> +	struct list_head userspace_devices;
>  
>  	/* other than negative (== assign one dynamically), bus_num is fully
>  	 * board-specific.  usually that simplifies to being SOC-specific.

  parent reply	other threads:[~2017-12-21 21:05 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-12-21 20:03 [PATCH] spi: Add a sysfs interface to instantiate devices Kyle Roeschley
2017-12-21 20:11 ` Randy Dunlap
2017-12-21 21:05 ` Trent Piepho [this message]
2017-12-22 15:56   ` Mark Brown
2017-12-22 17:11     ` Kyle Roeschley
2017-12-23  8:58       ` Geert Uytterhoeven
2017-12-27 10:31         ` Mark Brown

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1513890342.26695.4.camel@impinj.com \
    --to=tpiepho@impinj.com \
    --cc=broonie@kernel.org \
    --cc=geert@linux-m68k.org \
    --cc=kyle.roeschley@ni.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-spi@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox