> -----Original Message-----
> From: Sudeep Holla <[email protected]>
> Sent: Wednesday, August 5, 2020 4:48 PM
> To: Jason Liu <[email protected]>
> Cc: Marc Zyngier <[email protected]>; [email protected];
> [email protected]; [email protected]; [email protected]; Sudeep
> Holla <[email protected]>; [email protected]
> Subject: Re: [PATCH 1/1] arm64: kexec: no need to do irq_chip->irq_mask if it
> already masked
>
> On Wed, Aug 05, 2020 at 06:31:20AM +0000, Jason Liu wrote:
> > >
> > > Indeed, the runtime PM ops in that driver looks dodgy. Any calls to
> > > mask_irq from drivers or anywhere with irqchip suspended with just
> > > blows up the system.
> >
> > If you look at the chip->irq_mask implementation on different
> > platforms, almost all with directly access the register of the irqchip
> > including
> irqsteer.
> > There are fine due to driver will use the common mask_irq API.
> >
>
> That still doesn't explain how you can prevent system from blowing up if
> chip->irq_mask gets called with irqchip suspended ?
Do you mean driver, the driver will not call chip->irq_mask when irqchip was in
suspend.
if it does, it will be big BUG. I believe we don't do that in the driver.
>
> --
> Regards,
> Sudeep
