The devm_clk_get_enabled() helpers:
    - call devm_clk_get()
    - call clk_prepare_enable() and register what is needed in order to
     call clk_disable_unprepare() when needed, as a managed resource.

This simplifies the code and avoids calls to clk_disable_unprepare().

Signed-off-by: Lei Liu <liulei.rjpt@vivo.com>
---
 drivers/tty/serial/atmel_serial.c | 8 +-------
 1 file changed, 1 insertion(+), 7 deletions(-)

diff --git a/drivers/tty/serial/atmel_serial.c b/drivers/tty/serial/atmel_serial.c
index 09b246c9e389..209f3d41a17c 100644
--- a/drivers/tty/serial/atmel_serial.c
+++ b/drivers/tty/serial/atmel_serial.c
@@ -2910,14 +2910,11 @@ static int atmel_serial_probe(struct platform_device *pdev)
 	atomic_set(&atmel_port->tasklet_shutdown, 0);
 	spin_lock_init(&atmel_port->lock_suspended);
 
-	atmel_port->clk = devm_clk_get(&pdev->dev, "usart");
+	atmel_port->clk = devm_clk_get_enabled(&pdev->dev, "usart");
 	if (IS_ERR(atmel_port->clk)) {
 		ret = PTR_ERR(atmel_port->clk);
 		goto err;
 	}
-	ret = clk_prepare_enable(atmel_port->clk);
-	if (ret)
-		goto err;
 
 	atmel_port->gclk = devm_clk_get_optional(&pdev->dev, "gclk");
 	if (IS_ERR(atmel_port->gclk)) {
@@ -2968,15 +2965,12 @@ static int atmel_serial_probe(struct platform_device *pdev)
 	 * The peripheral clock can now safely be disabled till the port
 	 * is used
 	 */
-	clk_disable_unprepare(atmel_port->clk);
-
 	return 0;
 
 err_add_port:
 	kfree(atmel_port->rx_ring.buf);
 	atmel_port->rx_ring.buf = NULL;
 err_clk_disable_unprepare:
-	clk_disable_unprepare(atmel_port->clk);
 	clear_bit(atmel_port->uart.line, atmel_ports_in_use);
 err:
 	return ret;
-- 
2.34.1

Le 22/08/2024 à 05:39, Lei Liu a écrit :
> The devm_clk_get_enabled() helpers:
>      - call devm_clk_get()
>      - call clk_prepare_enable() and register what is needed in order to
>       call clk_disable_unprepare() when needed, as a managed resource.
> 
> This simplifies the code and avoids calls to clk_disable_unprepare().
> 
> Signed-off-by: Lei Liu <liulei.rjpt@vivo.com>
> ---
>   drivers/tty/serial/atmel_serial.c | 8 +-------
>   1 file changed, 1 insertion(+), 7 deletions(-)
> 
> diff --git a/drivers/tty/serial/atmel_serial.c b/drivers/tty/serial/atmel_serial.c
> index 09b246c9e389..209f3d41a17c 100644
> --- a/drivers/tty/serial/atmel_serial.c
> +++ b/drivers/tty/serial/atmel_serial.c
> @@ -2910,14 +2910,11 @@ static int atmel_serial_probe(struct platform_device *pdev)
>   	atomic_set(&atmel_port->tasklet_shutdown, 0);
>   	spin_lock_init(&atmel_port->lock_suspended);
>   
> -	atmel_port->clk = devm_clk_get(&pdev->dev, "usart");
> +	atmel_port->clk = devm_clk_get_enabled(&pdev->dev, "usart");
>   	if (IS_ERR(atmel_port->clk)) {
>   		ret = PTR_ERR(atmel_port->clk);
>   		goto err;
>   	}
> -	ret = clk_prepare_enable(atmel_port->clk);
> -	if (ret)
> -		goto err;
>   
>   	atmel_port->gclk = devm_clk_get_optional(&pdev->dev, "gclk");
>   	if (IS_ERR(atmel_port->gclk)) {
> @@ -2968,15 +2965,12 @@ static int atmel_serial_probe(struct platform_device *pdev)
>   	 * The peripheral clock can now safely be disabled till the port
>   	 * is used
>   	 */
> -	clk_disable_unprepare(atmel_port->clk);
> -
Why removing this ?
This is not an error path.

>   	return 0;
>   
>   err_add_port:
>   	kfree(atmel_port->rx_ring.buf);
>   	atmel_port->rx_ring.buf = NULL;
>   err_clk_disable_unprepare:
> -	clk_disable_unprepare(atmel_port->clk);
>   	clear_bit(atmel_port->uart.line, atmel_ports_in_use);
>   err:
>   	return ret;

Thanks,
Richard.

On Thu, Aug 22, 2024 at 03:28:40PM +0200, Richard GENOUD wrote:
> Le 22/08/2024 à 05:39, Lei Liu a écrit :
> > The devm_clk_get_enabled() helpers:
> >      - call devm_clk_get()
> >      - call clk_prepare_enable() and register what is needed in order to
> >       call clk_disable_unprepare() when needed, as a managed resource.
> > 
> > This simplifies the code and avoids calls to clk_disable_unprepare().

...

> >   	 * The peripheral clock can now safely be disabled till the port
> >   	 * is used
> >   	 */
> > -	clk_disable_unprepare(atmel_port->clk);
> > -
> Why removing this ?
> This is not an error path.

Good point, I wouldn't apply this patch as well as a few others in this series
due to this reason.

Instead it might make sense to add a comment on top of devm_clk_get() to
explain why _enabled() variant is *not* used.

-- 
With Best Regards,
Andy Shevchenko

On 22/08/2024 16:34:28+0300, Andy Shevchenko wrote:
> On Thu, Aug 22, 2024 at 03:28:40PM +0200, Richard GENOUD wrote:
> > Le 22/08/2024 à 05:39, Lei Liu a écrit :
> > > The devm_clk_get_enabled() helpers:
> > >      - call devm_clk_get()
> > >      - call clk_prepare_enable() and register what is needed in order to
> > >       call clk_disable_unprepare() when needed, as a managed resource.
> > > 
> > > This simplifies the code and avoids calls to clk_disable_unprepare().
> 
> ...
> 
> > >   	 * The peripheral clock can now safely be disabled till the port
> > >   	 * is used
> > >   	 */
> > > -	clk_disable_unprepare(atmel_port->clk);
> > > -
> > Why removing this ?
> > This is not an error path.
> 
> Good point, I wouldn't apply this patch as well as a few others in this series
> due to this reason.
> 
> Instead it might make sense to add a comment on top of devm_clk_get() to
> explain why _enabled() variant is *not* used.

Or maybe stop doing brainded conversions to new APIs.

> 
> -- 
> With Best Regards,
> Andy Shevchenko
> 
> 

-- 
Alexandre Belloni, co-owner and COO, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com