Re: [PATCH] of/irq: introduce of_has_named_irqs helper

[Dmitry Torokhov <dmitry.torokhov@gmail.com>]

On Tue, Jul 28, 2015 at 06:53:52PM +0530, Vignesh R wrote:
> Hi Dmitry,
> 
> On 07/25/2015 01:46 AM, Dmitry Torokhov wrote:
> > On Fri, Jul 24, 2015 at 12:26:19PM -0700, Dmitry Torokhov wrote:
> >> On Fri, Jul 24, 2015 at 02:14:57PM -0500, Rob Herring wrote:
> >>> On Fri, Jul 24, 2015 at 1:26 PM, Dmitry Torokhov
> >>> <dmitry.torokhov@gmail.com> wrote:
> >>>> Sometimes drivers might wish to transition from index-based to named
> >>>> interrupt descriptions. To aid in decision-making when parsing device
> >>>> tree data let's provide a helper that will indicate the scheme that is
> >>>> being used.
> >>>
> >>> Generally, IRQs are retrieved by platform_get_irq or
> >>> platform_get_irq_byname. Drivers should not call the of_irq_*
> >>> functions directly in most cases.
> >>
> >> That would be true for platform drivers, but not all devices are
> >> platform devices.
> >>
> >>>
> >>>>
> >>>> Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
> >>>> ---
> >>>>
> >>>> The intent is to it like this:
> >>>>
> >>>>         if (of_has_named_irqs(np) {
> >>>>                 /* Wake IRQ is optional */
> >>>>                 dev->wakeirq = of_irq_get_byname(np, "wakeup");
> >>>>                 if (dev->wakeirq < 0 && dev->wakeirq != -ENODATA)
> >>>>                         return dev->wakeirq;
> >>>>         }
> >>>
> >>> of_irq_get_byname will already return an error if the property is not
> >>> present. Use that.
> >>
> >> I do not like that it returns -EINVAL when property is missing, can we
> >> change it to return -ENODATA (so it is the same as when the property is
> >> defined but such name is missing)?
> > 
> > So here is what I had in mind.. It is based on recent patch by Vignesh
> > for pixcir touchscreen, but I think it should be made available to all
> > I2C devices. Completely untested at the moment.
> > 
> 
> Thanks for the patch! I tested this on am437x-gp-evm and the
> suspend/resume worked fine (below patch + $subject patch). I was able to
> wake the system from low power state using touchscreen. Also verified
> module insertion and removal. One comment though, please see below.
> 
> > ---
> >  drivers/i2c/i2c-core.c |   43 +++++++++++++++++++++++++++++++++++++------
> >  1 file changed, 37 insertions(+), 6 deletions(-)
> > 
> > diff --git a/drivers/i2c/i2c-core.c b/drivers/i2c/i2c-core.c
> > index e6d4935..3971461 100644
> > --- a/drivers/i2c/i2c-core.c
> > +++ b/drivers/i2c/i2c-core.c
> > @@ -47,6 +47,7 @@
> >  #include <linux/rwsem.h>
> >  #include <linux/pm_runtime.h>
> >  #include <linux/pm_domain.h>
> > +#include <linux/pm_wakeirq.h>
> >  #include <linux/acpi.h>
> >  #include <linux/jump_label.h>
> >  #include <asm/uaccess.h>
> > @@ -631,6 +632,7 @@ static int i2c_device_probe(struct device *dev)
> >  {
> >  	struct i2c_client	*client = i2c_verify_client(dev);
> >  	struct i2c_driver	*driver;
> > +	int wakeirq = 0;
> >  	int status;
> >  
> >  	if (!client)
> > @@ -639,11 +641,13 @@ static int i2c_device_probe(struct device *dev)
> >  	if (!client->irq) {
> >  		int irq = -ENOENT;
> >  
> > -		if (dev->of_node)
> > -			irq = of_irq_get(dev->of_node, 0);
> > -		else if (ACPI_COMPANION(dev))
> > +		if (dev->of_node) {
> > +			irq = of_has_named_irqs(dev->of_node) ?
> > +				of_irq_get_byname(dev->of_node, "irq") :
> > +				of_irq_get(dev->of_node, 0);
> > +		} else if (ACPI_COMPANION(dev)) {
> >  			irq = acpi_dev_gpio_irq_get(ACPI_COMPANION(dev), 0);
> > -
> > +		}
> >  		if (irq == -EPROBE_DEFER)
> >  			return irq;
> >  		if (irq < 0)
> > @@ -652,6 +656,15 @@ static int i2c_device_probe(struct device *dev)
> >  		client->irq = irq;
> >  	}
> >  
> > +	if (dev->of_node && of_has_named_irqs(dev->of_node)) {
> > +		wakeirq = of_irq_get_byname(dev->of_node, "wakeup");
> > +		if (wakeirq < 0) {
> > +			if (wakeirq != -ENODATA)
> > +				return wakeirq;
> > +			wakeirq = 0;
> > +		}
> > +	}
> > +
> >  	driver = to_i2c_driver(dev->driver);
> >  	if (!driver->probe || !driver->id_table)
> >  		return -ENODEV;
> > @@ -659,20 +672,34 @@ static int i2c_device_probe(struct device *dev)
> >  	if (!device_can_wakeup(&client->dev))
> >  		device_init_wakeup(&client->dev,
> >  					client->flags & I2C_CLIENT_WAKE);
> > +
> > +	status = wakeirq > 0 ?
> > +			dev_pm_set_dedicated_wake_irq(dev, wakeirq) :
> > +			(client->irq > 0 ?
> > +				dev_pm_set_wake_irq(dev, client->irq) : 0);
> 
> Above code tries to register wakeirq irrespective of whether the device
> is specified as wakeup-source in the dt or not. Hence, I see warn
> messages from every i2c device that has irq line but hasn't declared
> itself as wakeup-source:
> For example tps is on i2c:
> [    1.961613] tps65218 0-0024: forgot to call call device_init_wakeup?
> [    1.968340] tps65218 0-0024: failed to set up wakeup irq
> 
> May be you can register wakeup-source only if I2C_CLIENT_WAKE flag is
> set in client->flags?

Makes sense, I'll prepare a new version.

Just to confirm: do you see any issues with these 2 patches if you do
not modify the drivers to drop calls enable_irq_wake() from them?

Thanks.

-- 
Dmitry