Le 20/08/2019 à 17:47, Richard Genoud a écrit :
> Le 14/08/2019 à 13:08, Uwe Kleine-König a écrit :
>> On Wed, Aug 14, 2019 at 12:20:33PM +0200, Geert Uytterhoeven wrote:
>>> Hi Uwe,
>>>
>>> On Wed, Aug 14, 2019 at 11:36 AM Uwe Kleine-König
>>> <u.kleine-koenig@xxxxxxxxxxxxxx> wrote:
>>>> On Wed, Aug 14, 2019 at 11:29:22AM +0200, Geert Uytterhoeven wrote:
>>>>> Since commit 1d267ea6539f2663 ("serial: mctrl-gpio: simplify init
>>>>> routine"), mctrl_gpio_init() returns failure if the assignment to any
>>>>> member of the gpio array results in an error pointer.
>>>>> Since commit c359522194593815 ("serial: mctrl_gpio: Avoid probe failures
>>>>> in case of missing gpiolib"), mctrl_gpio_to_gpiod() returns NULL in the
>>>>> !CONFIG_GPIOLIB case.
>>>>> Hence there is no longer a need to check for mctrl_gpio_to_gpiod()
>>>>> returning an error value. A simple NULL check is sufficient.
>>>>>
>>>>> This follows the spirit of commit 445df7ff3fd1a0a9 ("serial: mctrl-gpio:
>>>>> drop usages of IS_ERR_OR_NULL") in the mctrl-gpio core.
>>>>>
>>>>> Signed-off-by: Geert Uytterhoeven <geert+renesas@xxxxxxxxx>
>>>>> ---
>>>>> drivers/tty/serial/atmel_serial.c | 12 ++++--------
>>>>> 1 file changed, 4 insertions(+), 8 deletions(-)
>>>>>
>>>>> diff --git a/drivers/tty/serial/atmel_serial.c b/drivers/tty/serial/atmel_serial.c
>>>>> index 19a85d6fe3d20541..e9620a81166b7dc1 100644
>>>>> --- a/drivers/tty/serial/atmel_serial.c
>>>>> +++ b/drivers/tty/serial/atmel_serial.c
>>>>> @@ -303,32 +303,28 @@ static unsigned int atmel_get_lines_status(struct uart_port *port)
>>>>>
>>>>> mctrl_gpio_get(atmel_port->gpios, &ret);
>>>>>
>>>>> - if (!IS_ERR_OR_NULL(mctrl_gpio_to_gpiod(atmel_port->gpios,
>>>>> - UART_GPIO_CTS))) {
>>>>> + if (mctrl_gpio_to_gpiod(atmel_port->gpios, UART_GPIO_CTS)) {
>>>>> if (ret & TIOCM_CTS)
>>>>> status &= ~ATMEL_US_CTS;
>>>>> else
>>>>> status |= ATMEL_US_CTS;
>>>>> }
>>>>
>>>> The change is fine, but it seems the atmel driver doesn't use mctrl_gpio
>>>> as expected (at least as expected by me). IMHO driving the hardware
>>>> function of the CTS pin shouldn't be conditional on the presence of a
>>>> cts-gpio. Is there a reason not to just drop the if completely?
>>>
>>> The above code returns the hardware status if CTS is not a GPIO, and
>>> returns (overrides with) the GPIO status if CTS is a GPIO.
>>> Isn't that correct, or am I missing something?
>>
>> I took a deeper look into this driver now. The task for
>> atmel_get_lines_status() isn't to implement the get_mctrl() callback.
>>
>> Instead this is called in the irqhandler to set ATMEL_US_RI in a
>> "pending" value that then later in atmel_handle_status() is translated
>> to a "ring" event that is handled there.
>>
>> So the right cleanup would be to let atmel_get_lines_status() just be
>>
>> return atmel_uart_readl(port, ATMEL_US_CSR);
>>
>> . If something happend on the lines implemented as gpio the driver's irq
>> function isn't called anyhow.
>
> I'd like to give it a good test to be sure, and I'll get back to you.
So, Uwe is right.
Since commit ce59e48fdbad ("serial: mctrl_gpio: implement interrupt handling"),
atmel_get_lines_status() can be completly killed and replaced by :
atmel_uart_readl(port, ATMEL_US_CSR);
Geert, do you want to send a patch for that, or should I do it ?
Thanks,
Richard
