[PATCH v2] serial/8250_dw: use platform_get_irq() instead of platform_get_resource()

[PATCH v2] serial/8250_dw: use platform_get_irq() instead of platform_get_resource()

[Alexey Brodkin]

It is not recommened to use platform_get_resource(pdev, IORESOURCE_IRQ)
for requesting IRQ's resources any more, as they can be not ready yet in
case of DT-booting.

platform_get_irq() instead is a recommended way for getting IRQ even if
it was not retrieved earlier.

It also makes code simpler because we're getting "int" value right away
and no conversion from resource to int is required.

Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
Cc: Vineet Gupta <vgupta@synopsys.com>
Cc: Alan Cox <alan@linux.intel.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
---

 Changes in v2:
  * Suppress error message if platform_get_irq() returns -EPROBE_DEFER

---
 drivers/tty/serial/8250/8250_dw.c | 14 ++++++++++----
 1 file changed, 10 insertions(+), 4 deletions(-)

diff --git a/drivers/tty/serial/8250/8250_dw.c b/drivers/tty/serial/8250/8250_dw.c
index e601162..cda76e8 100644
--- a/drivers/tty/serial/8250/8250_dw.c
+++ b/drivers/tty/serial/8250/8250_dw.c
@@ -384,18 +384,24 @@ static int dw8250_probe(struct platform_device *pdev)
 {
 	struct uart_8250_port uart = {};
 	struct resource *regs = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	struct resource *irq = platform_get_resource(pdev, IORESOURCE_IRQ, 0);
+	int irq = platform_get_irq(pdev, 0);
 	struct dw8250_data *data;
 	int err;
 
-	if (!regs || !irq) {
-		dev_err(&pdev->dev, "no registers/irq defined\n");
+	if (!regs) {
+		dev_err(&pdev->dev, "no registers defined\n");
 		return -EINVAL;
 	}
 
+	if (irq < 0) {
+		if (irq != -EPROBE_DEFER)
+			dev_err(&pdev->dev, "cannot get irq\n");
+		return irq;
+	}
+
 	spin_lock_init(&uart.port.lock);
 	uart.port.mapbase = regs->start;
-	uart.port.irq = irq->start;
+	uart.port.irq = irq;
 	uart.port.handle_irq = dw8250_handle_irq;
 	uart.port.pm = dw8250_do_pm;
 	uart.port.type = PORT_8250;
-- 
2.1.0

Re: [PATCH v2] serial/8250_dw: use platform_get_irq() instead of platform_get_resource()

[Andy Shevchenko]

On Tue, 2015-03-03 at 18:11 +0300, Alexey Brodkin wrote:
> It is not recommened to use platform_get_resource(pdev, IORESOURCE_IRQ)
> for requesting IRQ's resources any more, as they can be not ready yet in
> case of DT-booting.
> 
> platform_get_irq() instead is a recommended way for getting IRQ even if
> it was not retrieved earlier.
> 
> It also makes code simpler because we're getting "int" value right away
> and no conversion from resource to int is required.
> 
> Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
> Cc: Vineet Gupta <vgupta@synopsys.com>
> Cc: Alan Cox <alan@linux.intel.com>
> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
> Cc: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> ---
> 
>  Changes in v2:
>   * Suppress error message if platform_get_irq() returns -EPROBE_DEFER

Do we really need that message at all?

> 
> ---
>  drivers/tty/serial/8250/8250_dw.c | 14 ++++++++++----
>  1 file changed, 10 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/tty/serial/8250/8250_dw.c b/drivers/tty/serial/8250/8250_dw.c
> index e601162..cda76e8 100644
> --- a/drivers/tty/serial/8250/8250_dw.c
> +++ b/drivers/tty/serial/8250/8250_dw.c
> @@ -384,18 +384,24 @@ static int dw8250_probe(struct platform_device *pdev)
>  {
>  	struct uart_8250_port uart = {};
>  	struct resource *regs = platform_get_resource(pdev, IORESOURCE_MEM, 0);
> -	struct resource *irq = platform_get_resource(pdev, IORESOURCE_IRQ, 0);
> +	int irq = platform_get_irq(pdev, 0);
>  	struct dw8250_data *data;
>  	int err;
>  
> -	if (!regs || !irq) {
> -		dev_err(&pdev->dev, "no registers/irq defined\n");
> +	if (!regs) {
> +		dev_err(&pdev->dev, "no registers defined\n");
>  		return -EINVAL;
>  	}
>  
> +	if (irq < 0) {
> +		if (irq != -EPROBE_DEFER)
> +			dev_err(&pdev->dev, "cannot get irq\n");
> +		return irq;
> +	}
> +
>  	spin_lock_init(&uart.port.lock);
>  	uart.port.mapbase = regs->start;
> -	uart.port.irq = irq->start;
> +	uart.port.irq = irq;
>  	uart.port.handle_irq = dw8250_handle_irq;
>  	uart.port.pm = dw8250_do_pm;
>  	uart.port.type = PORT_8250;


-- 
Andy Shevchenko <andriy.shevchenko@intel.com>
Intel Finland Oy

Re: [PATCH v2] serial/8250_dw: use platform_get_irq() instead of platform_get_resource()

[Alexey Brodkin]

Hi Andy,

On Tue, 2015-03-03 at 17:26 +0200, Andy Shevchenko wrote:
> On Tue, 2015-03-03 at 18:11 +0300, Alexey Brodkin wrote:
> > It is not recommened to use platform_get_resource(pdev, IORESOURCE_IRQ)
> > for requesting IRQ's resources any more, as they can be not ready yet in
> > case of DT-booting.
> > 
> > platform_get_irq() instead is a recommended way for getting IRQ even if
> > it was not retrieved earlier.
> > 
> > It also makes code simpler because we're getting "int" value right away
> > and no conversion from resource to int is required.
> > 
> >  Changes in v2:
> >   * Suppress error message if platform_get_irq() returns -EPROBE_DEFER
> 
> Do we really need that message at all?

IMHO it makes sense.

For example it was useful for me when debugging stacked interrupt
controllers setup - I got explicitly notified why this particular device
failed on probe.

Note that IRQ is a pretty specific resource due to the fact of INTC
stacking and situations when each and every INTC gets finally probed.

Still if you believe we may drop this message with no loss of usability
- I'm fine with that as well.

-Alexey

Re: [PATCH v2] serial/8250_dw: use platform_get_irq() instead of platform_get_resource()

[Andy Shevchenko]

On Tue, 2015-03-03 at 15:33 +0000, Alexey Brodkin wrote:
> Hi Andy,
> 
> On Tue, 2015-03-03 at 17:26 +0200, Andy Shevchenko wrote:
> > On Tue, 2015-03-03 at 18:11 +0300, Alexey Brodkin wrote:
> > > It is not recommened to use platform_get_resource(pdev, IORESOURCE_IRQ)
> > > for requesting IRQ's resources any more, as they can be not ready yet in
> > > case of DT-booting.
> > > 
> > > platform_get_irq() instead is a recommended way for getting IRQ even if
> > > it was not retrieved earlier.
> > > 
> > > It also makes code simpler because we're getting "int" value right away
> > > and no conversion from resource to int is required.
> > > 
> > >  Changes in v2:
> > >   * Suppress error message if platform_get_irq() returns -EPROBE_DEFER
> > 
> > Do we really need that message at all?
> 
> IMHO it makes sense.
> 
> For example it was useful for me when debugging stacked interrupt
> controllers setup - I got explicitly notified why this particular device
> failed on probe.

There are so many device drivers which prints similar message that you
would consider to make a separate patch to platform.c code to do that
stuff there once for all.

> Note that IRQ is a pretty specific resource due to the fact of INTC
> stacking and situations when each and every INTC gets finally probed.
> 
> Still if you believe we may drop this message with no loss of usability
> - I'm fine with that as well.

If no one else has an objection I would prefer to skip it. You always
may get the return code from probe(), though it's not exactly mapped to
the reason why it failed.

-- 
Andy Shevchenko <andriy.shevchenko@intel.com>
Intel Finland Oy

Re: [PATCH v2] serial/8250_dw: use platform_get_irq() instead of platform_get_resource()

[Alexey Brodkin]

Hi Andy,

On Tue, 2015-03-03 at 17:43 +0200, Andy Shevchenko wrote:
> On Tue, 2015-03-03 at 15:33 +0000, Alexey Brodkin wrote:
> > Hi Andy,
> > 
> > On Tue, 2015-03-03 at 17:26 +0200, Andy Shevchenko wrote:
> > > On Tue, 2015-03-03 at 18:11 +0300, Alexey Brodkin wrote:
> > > > It is not recommened to use platform_get_resource(pdev, IORESOURCE_IRQ)
> > > > for requesting IRQ's resources any more, as they can be not ready yet in
> > > > case of DT-booting.
> > > > 
> > > > platform_get_irq() instead is a recommended way for getting IRQ even if
> > > > it was not retrieved earlier.
> > > > 
> > > > It also makes code simpler because we're getting "int" value right away
> > > > and no conversion from resource to int is required.
> > > > 
> > > >  Changes in v2:
> > > >   * Suppress error message if platform_get_irq() returns -EPROBE_DEFER
> > > 
> > > Do we really need that message at all?
> > 
> > IMHO it makes sense.
> > 
> > For example it was useful for me when debugging stacked interrupt
> > controllers setup - I got explicitly notified why this particular device
> > failed on probe.
> 
> There are so many device drivers which prints similar message that you
> would consider to make a separate patch to platform.c code to do that
> stuff there once for all.
> 
> > Note that IRQ is a pretty specific resource due to the fact of INTC
> > stacking and situations when each and every INTC gets finally probed.
> > 
> > Still if you believe we may drop this message with no loss of usability
> > - I'm fine with that as well.
> 
> If no one else has an objection I would prefer to skip it. You always
> may get the return code from probe(), though it's not exactly mapped to
> the reason why it failed.

Then we may want to do another massive clean-up because there're lots of
other drivers that have this message in some form :)

-Alexey

Re: [PATCH v2] serial/8250_dw: use platform_get_irq() instead of platform_get_resource()

[Andy Shevchenko]

On Tue, 2015-03-03 at 15:47 +0000, Alexey Brodkin wrote:
> > > For example it was useful for me when debugging stacked interrupt
> > > controllers setup - I got explicitly notified why this particular device
> > > failed on probe.
> > 
> > There are so many device drivers which prints similar message that you
> > would consider to make a separate patch to platform.c code to do that
> > stuff there once for all.

> Then we may want to do another massive clean-up because there're lots of
> other drivers that have this message in some form :)

Yeah, you may do that and it will be (by my opinion) more useful than
introducing one more place with the duplicate message.

-- 
Andy Shevchenko <andriy.shevchenko@intel.com>
Intel Finland Oy