diff mbox

spi: spidev: add exclusive bus access lock via ioctls

Message ID 1502538843-3283-1-git-send-email-vicky773@gmail.com (mailing list archive)
State New, archived
Headers show

Commit Message

Vikram N Aug. 12, 2017, 11:54 a.m. UTC
There are use-cases wherein one or more applications wants to communicate
with the same spi slave chip and they wish to have exclusive bus access.
May to serialize their command,response patterns.
This can be realized in two ways,
 - Identify different applications using same spi channel and implement
   synchronization mechanisms to get exclusive bus access
 - Provide some mechanism through spidev driver to get exclusive spi bus
   access like via ioctls

This patch implements the second approach which eliminates the need for
implementing complex IPC mechanisms in the applications which shares the
spi bus. This becomes even harder if we don't have access to the source
code of the application which is using the spi channel.

Signed-off-by: Vikram N <vicky773@gmail.com>
---
 drivers/spi/spidev.c            |   14 +++++++++++++-
 include/uapi/linux/spi/spidev.h |    4 +++-
 2 files changed, 16 insertions(+), 2 deletions(-)

Comments

Mark Brown Aug. 14, 2017, 6:38 p.m. UTC | #1
On Sat, Aug 12, 2017 at 05:24:03PM +0530, Vikram N wrote:

>  	else
> -		status = spi_sync(spi, message);
> +		status = spidev->bus_locked ? spi_sync_locked(spi, message) :
> +				spi_sync(spi, message);

Please don't abuse the ternery operator, people need to be able to read
the code.

> +	case SPI_IOC_BUS_LOCK:
> +		spi_bus_lock(spi->master);
> +		spidev->bus_locked = true;
> +		break;
> +
> +	case SPI_IOC_BUS_UNLOCK:
> +		spi_bus_unlock(spi->master);
> +		spidev->bus_locked = false;
> +		break;

I'm not super convinced that this API is a good idea in general - it
seems extremely niche to be using multiple userspace programs that don't
need to coordinate at all except for a single lock (which they will all
need to use to avoid just bouncing off with errors).  That all seems
very narrow.

I'm also worried that even if there is such a use case this code is very
fragile as it stands.  If an application crashes then nothing will free
a lock it holds and any application can through simple error drop locks
that are supposed to be held by other applications.  This isn't going to
be terribly robust.
diff mbox

Patch

diff --git a/drivers/spi/spidev.c b/drivers/spi/spidev.c
index cda1071..d14a8c1 100644
--- a/drivers/spi/spidev.c
+++ b/drivers/spi/spidev.c
@@ -85,6 +85,7 @@  struct spidev_data {
 	u8			*tx_buffer;
 	u8			*rx_buffer;
 	u32			speed_hz;
+	bool			bus_locked;
 };
 
 static LIST_HEAD(device_list);
@@ -109,7 +110,8 @@  struct spidev_data {
 	if (spi == NULL)
 		status = -ESHUTDOWN;
 	else
-		status = spi_sync(spi, message);
+		status = spidev->bus_locked ? spi_sync_locked(spi, message) :
+				spi_sync(spi, message);
 
 	if (status == 0)
 		status = message->actual_length;
@@ -462,6 +464,16 @@  static int spidev_message(struct spidev_data *spidev,
 		}
 		break;
 
+	case SPI_IOC_BUS_LOCK:
+		spi_bus_lock(spi->master);
+		spidev->bus_locked = true;
+		break;
+
+	case SPI_IOC_BUS_UNLOCK:
+		spi_bus_unlock(spi->master);
+		spidev->bus_locked = false;
+		break;
+
 	default:
 		/* segmented and/or full-duplex I/O request */
 		/* Check message and copy into scratch area */
diff --git a/include/uapi/linux/spi/spidev.h b/include/uapi/linux/spi/spidev.h
index dd5f21e..1f776fe 100644
--- a/include/uapi/linux/spi/spidev.h
+++ b/include/uapi/linux/spi/spidev.h
@@ -136,6 +136,8 @@  struct spi_ioc_transfer {
 #define SPI_IOC_RD_MODE32		_IOR(SPI_IOC_MAGIC, 5, __u32)
 #define SPI_IOC_WR_MODE32		_IOW(SPI_IOC_MAGIC, 5, __u32)
 
-
+/* Exclusive bus access lock */
+#define SPI_IOC_BUS_LOCK		_IO(SPI_IOC_MAGIC, 6)
+#define SPI_IOC_BUS_UNLOCK		_IO(SPI_IOC_MAGIC, 7)
 
 #endif /* SPIDEV_H */