[Linux-stm32] [PATCH RFC net-next 6/7] net: stmmac: add helpers to indicate WoL enable status

[Florian Fainelli]

On 7/29/25 11:19, Russell King (Oracle) wrote:
> On Tue, Jul 29, 2025 at 07:27:11PM +0200, Andrew Lunn wrote:
>> And i did notice that the Broadcom code is the only one doing anything
>> with enable_irq_wake()/disable_irq_wake(). We need to scatter these
>> into the drivers.
> 
> It's better to use devm_pm_set_wake_irq() in the probe function, and
> then let the core code (drivers/base/power/wakeup.c and
> drivers/base/power/wakeirq.c) handle it. This is what I'm doing for
> the rtl8211f.
> 
> IRQ wake gets enabled/disabled at suspend/resume time, rather than
> when the device wakeup state changes, which I believe is what is
> preferred.
> 

Sounds reasonable, I will go test that instead of doing the 
enable_irq_wake()/disable_irq_wake() dance. Thanks!
-- 
Florian