Re: [PATCH 5/6] toshiba_bluetooth: Change BT status message to debug

["Bjørn Mork" <bjorn@mork.no>]

Azael Avalos <coproscefalo@gmail.com> writes:

> The function toshiba_bluetooth_status s currently printing the status
> of the device whenever it is queried, but since the introduction of
> the rfkill poll code, this value will get printed everytime the poll
> occurs.
>
> This patch changes the level of the printed message from info to
> debug, and also adds a few more debug messages printing the
> killswitch, plug and power status of the device as well.
>
> Signed-off-by: Azael Avalos <coproscefalo@gmail.com>
> ---
>  drivers/platform/x86/toshiba_bluetooth.c | 6 +++++-
>  1 file changed, 5 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/platform/x86/toshiba_bluetooth.c b/drivers/platform/x86/toshiba_bluetooth.c
> index 9867ccd..93b9688 100644
> --- a/drivers/platform/x86/toshiba_bluetooth.c
> +++ b/drivers/platform/x86/toshiba_bluetooth.c
> @@ -99,7 +99,7 @@ static int toshiba_bluetooth_status(acpi_handle handle)
>  		return -ENXIO;
>  	}
>  
> -	pr_info("Bluetooth status %llu\n", status);
> +	pr_debug("Bluetooth status %llu\n", status);
>  
>  	return status;
>  }
> @@ -157,6 +157,10 @@ static int toshiba_bluetooth_sync_status(struct toshiba_bluetooth_dev *bt_dev)
>  	bt_dev->plugged = (status & BT_PLUGGED_MASK) ? true : false;
>  	bt_dev->powered = (status & BT_POWER_MASK) ? true : false;
>  
> +	pr_debug("killswitch %d\n", bt_dev->killswitch);
> +	pr_debug("plugged %d\n", bt_dev->plugged);
> +	pr_debug("powered %d\n", bt_dev->powered);


Those are terribly generic messages.  I don't think I would have guessed
which device was trying to tell me "powered 1" if I found it in the
logs...

How about using e.g dev_dbg() to get a bit more context here?

You might also want to put all three into a single call, so that they
make a single dynamic debug entry when dynamic debugging is enabled.

And looking at toshiba_bluetooth_status() I see that all callers have a
device. How about propagating the device to be able to use the dev_*
printk's there as well?  Let the device identify itself instead of
having to guess.


Bjørn