Re: [PATCH 2/9] mfd: Add char devices for the ST-Ericsson CG2900.

From: Arnd Bergmann
Date: Fri Oct 22 2010 - 11:49:48 EST


On Friday 22 October 2010 12:36:09 Par-Gunnar Hjalmdahl wrote:
> This patch adds char devices to the ST-Ericsson CG2900 driver.
> The reason for this is to allow users of CG2900, such as GPS, to
> be placed in user space.

Can you be more specific how you expect this to be used?

I guess you have hardware units that then each correspond to
one character device and can be talked to over a pseudo socket
interface, right?

For most devices (radio, audio, bluetooth, gps, ...), we already
have existing user interfaces, so how do they interact with this
one?

> +#define NAME "CharDev "

?

> +/* Ioctls */
> +#define CG2900_CHAR_DEV_IOCTL_RESET _IOW('U', 210, int)
> +#define CG2900_CHAR_DEV_IOCTL_CHECK4RESET _IOR('U', 212, int)
> +#define CG2900_CHAR_DEV_IOCTL_GET_REVISION _IOR('U', 213, int)
> +#define CG2900_CHAR_DEV_IOCTL_GET_SUB_VER _IOR('U', 214, int)

These definitions look wrong -- you never use the ioctl argument...

> + *
> + * Returns:
> + * Bytes successfully read (could be 0).
> + * -EBADF if NULL pointer was supplied in private data.
> + * -EFAULT if copy_to_user fails.
> + * Error codes from wait_event_interruptible.
> + */
> +static ssize_t char_dev_read(struct file *filp, char __user *buf, size_t count,
> + loff_t *f_pos)

The same comment applies here that I made for the test interface:
Why is this not an AF_BLUETOOTH socket instead of a chardev?

Unless I'm mistaken, you actually send bluetooth frames after all.

> + case CG2900_CHAR_DEV_IOCTL_RESET:
> + if (!dev) {
> + err = -EBADF;
> + goto error_handling;
> + }
> + CG2900_INFO("ioctl reset command for device %s", dev->name);
> + err = cg2900_reset(dev->dev);
> + break;
> +
> + case CG2900_CHAR_DEV_IOCTL_CHECK4RESET:
> + if (!dev) {
> + CG2900_INFO("ioctl check for reset command for device");
> + /* Return positive value if closed */
> + err = CG2900_CHAR_DEV_IOCTL_EVENT_CLOSED;
> + } else if (dev->reset_state == CG2900_CHAR_RESET) {
> + CG2900_INFO("ioctl check for reset command for device "
> + "%s", dev->name);
> + /* Return positive value if reset */
> + err = CG2900_CHAR_DEV_IOCTL_EVENT_RESET;
> + }
> + break;

Strange interface. Why do you need to check for the reset?

> + case CG2900_CHAR_DEV_IOCTL_GET_REVISION:
> + CG2900_INFO("ioctl check for local revision info");
> + if (cg2900_get_local_revision(&rev_data)) {
> + CG2900_DBG("Read revision data revision %d "
> + "sub_version %d",
> + rev_data.revision, rev_data.sub_version);
> + err = rev_data.revision;
> + } else {
> + CG2900_DBG("No revision data available");
> + err = -EIO;
> + }
> + break;
> +
> + case CG2900_CHAR_DEV_IOCTL_GET_SUB_VER:
> + CG2900_INFO("ioctl check for local sub-version info");
> + if (cg2900_get_local_revision(&rev_data)) {
> + CG2900_DBG("Read revision data revision %d "
> + "sub_version %d",
> + rev_data.revision, rev_data.sub_version);
> + err = rev_data.sub_version;
> + } else {
> + CG2900_DBG("No revision data available");
> + err = -EIO;
> + }
> + break;

These look like could better live in a sysfs attribute of the platform device.

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