[PATCH v11 1/5] staging: vc04_services: vchiq_arm: Add new bus type and device type

Dan Carpenter dan.carpenter at linaro.org
Wed Sep 13 23:55:20 PDT 2023


On Thu, Sep 14, 2023 at 01:23:50AM +0530, Umang Jain wrote:
> +static int vchiq_bus_type_match(struct device *dev, struct device_driver *drv)
> +{
> +	if (dev->bus == &vchiq_bus_type &&
> +	    strcmp(dev_name(dev), drv->name) == 0)
> +		return 1;
> +
> +	return 0;
> +}

I was not going to comment on this, because it's unfair to nitpick a
v11 patch...  But since you're going to have to redo it anyway, could
you make this function bool and change it to return true/false.

static bool vchiq_bus_type_match(struct device *dev, struct device_driver *drv)
{
	if (dev->bus == &vchiq_bus_type &&
	    strcmp(dev_name(dev), drv->name) == 0)
		return true;

	return false;
}

> +static int vchiq_bus_probe(struct device *dev)
> +{
> +	struct vchiq_device *device = to_vchiq_device(dev);
> +	struct vchiq_driver *driver = to_vchiq_driver(dev->driver);
> +	int ret;
> +
> +	ret = driver->probe(device);
> +	if (ret == 0)
> +		return 0;
> +
> +	return ret;

Ugh...  I was going to ignore this as well, but later there is a
checkpatch warning so then I decided nitpicking was ok.  Always do
error handling.  if (ret).  Never do success handling.  if (!ret).  But
here it can be done on one line.

	return driver->probe(device);

> +}
> +
> +struct bus_type vchiq_bus_type = {
> +	.name   = "vchiq-bus",
> +	.match  = vchiq_bus_type_match,
> +	.uevent = vchiq_bus_uevent,
> +	.probe  = vchiq_bus_probe,
> +};
> +
> +static void vchiq_device_release(struct device *dev)
> +{
> +	struct vchiq_device *device = to_vchiq_device(dev);
> +
> +	kfree(device);
> +}
> +
> +struct vchiq_device *
> +vchiq_device_register(struct device *parent, const char *name)
> +{
> +	struct vchiq_device *device;
> +	int ret;
> +
> +	device = kzalloc(sizeof(*device), GFP_KERNEL);
> +	if (!device) {
> +		dev_err(parent, "Cannot register %s: Insufficient memory\n",
> +			name);

Run checkpatch.pl -f on your files.

> +		return NULL;
> +	}

Stefan already commented on the other stuff I was going to say.

regards,
dan carpenter




More information about the linux-arm-kernel mailing list