On Tue, 16 Oct 2007 16:54:49 +0300, Vitaliy Ivanov <vitalivanov@gmail.com> wrote:

quoted text
> Again, comments are welcomed.

It looks like you misunderstood why a static lock protects open counts.
This is done so you do not need to worry about in-structure lock which
can be freed together with the structure. Look at this:

quoted text
> +static int adu_release_internal(struct adu_device *dev)
> +{
> +	/* lock this device */
> +	down(&dev->sem);
> +	/* decrement our usage count for the device */
> +	--dev->open_count;
> +	if (dev->open_count <= 0) {
> +		adu_abort_transfers(dev);
> +		dev->open_count = 0;
> +	}
> +	/* unlock this device */
> +	up(&dev->sem);

The dev->sem is entirely unnecessary here. Every time you use
open_count, it's protected by minor_table_mutex. The name is a litte
unfortunate, feel free to rename it.

This is probably a problem in 2.6 as well. I don't know why people keep
writing these things. Someone at Ontrak needs to look into it.

-- Pete
-
unsubscribe notice
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/