Re: [PATCH 1/6] IIO: Core sysfs only support.

Hi,
I know that this patchset is probably a bit outdated but still...

On 07/11/2011 15:52, jic23@cam.ac.uk wrote:

From: Jonathan Cameron <redacted>

Add support for simple sysfs only interfaces.

Bulk of patch is concerned with taking struct iio_chan_spec
arrays and generating all the relevant interfaces from them.

Signed-off-by: Jonathan Cameron <redacted>

First, you can add my
Tested-by: Maxime Ripard <redacted>

I have encountered a bug during my tests though, see below

+struct iio_dev *iio_device_allocate(int sizeof_priv)
+{
+	struct iio_dev *dev;
+	size_t alloc_size;
+
+	alloc_size = sizeof(struct iio_dev);
+	if (sizeof_priv) {
+		alloc_size = ALIGN(alloc_size, IIO_ALIGN);
+		alloc_size += sizeof_priv;
+	}
+	/* ensure cacheline alignment of whole construct */
+	alloc_size += IIO_ALIGN - 1;
+
+	dev = kzalloc(alloc_size, GFP_KERNEL);
+
+	if (dev) {
+		dev->dev.groups = dev->groups;
+		dev->dev.type = &iio_dev_type;
+		dev->dev.bus = &iio_bus_type;
+		device_initialize(&dev->dev);
+		dev_set_drvdata(&dev->dev, (void *)dev);
+		mutex_init(&dev->mlock);
+	}
+
+	return dev;
+}
+EXPORT_SYMBOL_GPL(iio_device_allocate);
+
+void iio_device_free(struct iio_dev *dev)
+{
+	if (dev)
+		iio_put_device(dev);
+}
+EXPORT_SYMBOL_GPL(iio_device_free);

Here, the iio_put_device is called in free, but get_device is never
called, but seem to be called in iio_device_register, which, if you ever
encounter an error in the probe function, you will never have called the
iio_device_register function, before calling the iio_device_free
function, which leads to a segfault.

Maxime

-- 
Maxime Ripard, Free Electrons
Kernel, drivers, real-time and embedded Linux
development, consulting, training and support.
http://free-electrons.com