Re: [PATCH 7/9] serdev: Introduce new bus for serial attached devices
From: Rob Herring
Date: Thu Jan 12 2017 - 15:14:16 EST
On Sat, Jan 7, 2017 at 8:02 AM, Andy Shevchenko
<andriy.shevchenko@xxxxxxxxxxxxxxx> wrote:
> On Fri, 2017-01-06 at 10:26 -0600, Rob Herring wrote:
>> The serdev bus is designed for devices such as Bluetooth, WiFi, GPS
>> and NFC connected to UARTs on host processors. Tradionally these have
>> been handled with tty line disciplines, rfkill, and userspace glue
>> such
>> as hciattach. This approach has many drawbacks since it doesn't fit
>> into the Linux driver model. Handling of sideband signals, power
>> control
>> and firmware loading are the main issues.
>>
>> This creates a serdev bus with controllers (i.e. host serial ports)
>> and
>> attached devices. Typically, these are point to point connections, but
>> some devices have muxing protocols or a h/w mux is conceivable. Any
>> muxing is not yet supported with the serdev bus.
[...]
>> +static int of_serdev_register_devices(struct serdev_controller *ctrl)
>> +{
>> + struct device_node *node;
>> + struct serdev_device *serdev = NULL;
>> + int err;
>> + bool found = false;
>> +
>> + for_each_available_child_of_node(ctrl->dev.of_node, node) {
>> + if (!of_get_property(node, "compatible", NULL))
>> + continue;
>> +
>> + dev_dbg(&ctrl->dev, "adding child %s\n", node-
>> >full_name);
>> +
>> + serdev = serdev_device_alloc(ctrl);
>> + if (!serdev)
>> + continue;
>> +
>> + serdev->dev.of_node = node;
>> +
>> + err = serdev_device_add(serdev);
>> + if (err) {
>> + dev_err(&serdev->dev,
>> + "failure adding device. status %d\n",
>> err);
>> + serdev_device_put(serdev);
>> + }
>>
>
>> + found = true;
>
> Perhaps
>
> } else if (!found)
> found = true;
>
> Otherwise if we end up with all devices not being added, called will not
> know about it.
At least for now, we really only support 1 device attached. I'm sure
someone will come up with h/w with more than one device. RS-485 allows
it I think or someone could have muxed access.
I just did "else found = true;" as there's no need to check the condition.
>
>
>> + }
>> + if (!found)
>> + return -ENODEV;
>> +
>> + return 0;
>> +}
>>
>
>
> +/**
>> + * serdev_controller_remove(): remove an serdev controller
>> + * @ctrl: controller to remove
>> + *
>> + * Remove a serdev controller. Caller is responsible for calling
>> + * serdev_controller_put() to discard the allocated controller.
>> + */
>> +void serdev_controller_remove(struct serdev_controller *ctrl)
>> +{
>> + int dummy;
>> +
>>
>
>> + if (!ctrl)
>> + return;
>
> By the way, should we take care or caller? What is the best practice
> here?
If the caller, then every caller has to check. Better to check in one place.
Rob