Re: [PATCH] mfd: as3722: Handle interrupts on suspend

[Jon Hunter <jonathanh@nvidia.com>]

On 19/11/15 14:23, Jon Hunter wrote:
> The as3722 device is registered as an irqchip and the as3722-rtc interrupt
> is one of it's interrupt sources. When using the as3722-rtc as a wake-up
> device from suspend, the following is seen:
> 
>   PM: Syncing filesystems ... done.
>   Freezing user space processes ... (elapsed 0.001 seconds) done.
>   Freezing remaining freezable tasks ... (elapsed 0.001 seconds) done.
>   Suspending console(s) (use no_console_suspend to debug)
>   PM: suspend of devices complete after 161.119 msecs
>   PM: late suspend of devices complete after 1.048 msecs
>   PM: noirq suspend of devices complete after 0.756 msecs
>   Disabling non-boot CPUs ...
>   CPU1: shutdown
>   CPU2: shutdown
>   CPU3: shutdown
>   Entering suspend state LP1
>   Enabling non-boot CPUs ...
>   CPU1 is up
>   CPU2 is up
>   CPU3 is up
>   PM: noirq resume of devices complete after 0.487 msecs
>   as3722 4-0040: Failed to read IRQ status: -16
>   as3722 4-0040: Failed to read IRQ status: -16
>   as3722 4-0040: Failed to read IRQ status: -16
>   as3722 4-0040: Failed to read IRQ status: -16
>   ...
> 
> The reason why the as3722 interrupt status cannot be read is because the
> as3722 interrupt is not masked during suspend and when the as3722-rtc
> interrupt occurs, to wake-up the device, the interrupt is seen before the
> i2c controller has been resumed in order to read the as3722 interrupt
> status.
> 
> The as3722-rtc driver sets it's interrupt as a wake-up source during
> suspend, which gets propagated to the parent as3722 interrupt. However,
> the as3722-rtc driver cannot disable it's interrupt during suspend
> otherwise we would never be woken up and so the as3722 must disable it's
> interrupt instead.
> 
> Fix this by disabling the as3722 interrupt during suspend. To ensure that
> a wake-up event from the as3722 is not missing, enable the as3722 interrupt
> as a wake-up source before disabling the interrupt on entering suspend.
> 
> Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
> ---
>  drivers/mfd/as3722.c | 29 +++++++++++++++++++++++++++++
>  1 file changed, 29 insertions(+)
> 
> diff --git a/drivers/mfd/as3722.c b/drivers/mfd/as3722.c
> index 924ea90494ae..8ee6e4008276 100644
> --- a/drivers/mfd/as3722.c
> +++ b/drivers/mfd/as3722.c
> @@ -405,6 +405,8 @@ static int as3722_i2c_probe(struct i2c_client *i2c,
>  		goto scrub;
>  	}
>  
> +	device_init_wakeup(as3722->dev, true);
> +
>  	dev_dbg(as3722->dev, "AS3722 core driver initialized successfully\n");
>  	return 0;
>  
> @@ -422,6 +424,28 @@ static int as3722_i2c_remove(struct i2c_client *i2c)
>  	return 0;
>  }
>  
> +static int as3722_i2c_suspend(struct device *dev)
> +{
> +	struct as3722 *as3722 = dev_get_drvdata(dev);
> +
> +	if (device_may_wakeup(dev))
> +		enable_irq_wake(as3722->chip_irq);
> +	disable_irq(as3722->chip_irq);
> +
> +	return 0;
> +}
> +
> +static int as3722_i2c_resume(struct device *dev)
> +{
> +	struct as3722 *as3722 = dev_get_drvdata(dev);
> +
> +	if (device_may_wakeup(dev))
> +		disable_irq_wake(as3722->chip_irq);
> +	enable_irq(as3722->chip_irq);

I just realised that I should enable the irq before disabling the wakeup
here. I will fix that.

I will wait to send out a V2 for a couple days in case there are any
other comments.

Jon