Re: [PATCH] irq: Request and release resources for chained IRQs

[Linus Walleij]

On Thu, Jun 18, 2020 at 8:42 PM Thomas Gleixner <tglx@xxxxxxxxxxxxx> wrote:
> David Heidelberg <david@xxxxxxx> writes:
> > is there chance to get this patch included or could be this issue
> > solved with different approach?
>
> Included into what? This patch is incorrect as I pointed out in review
> here:
>
>   https://lore.kernel.org/lkml/alpine.DEB.2.21.1812071143480.14498@xxxxxxxxxxxxxxxxxxxxxxx/
>
> So why are you even asking?
>
> I recommended to switch this away from chained handler and then the
> whole story ended with this mail:
>
>   https://lore.kernel.org/lkml/alpine.DEB.2.21.1812071404140.14498@xxxxxxxxxxxxxxxxxxxxxxx/
>
> I have no idea how the GPIO people solved that problem, but certainly
> not by applying this.

We didn't solve that problem.

The reason is that changing the platform from chained to regular handler
involves of course changing all the other drivers in the irqchip hierarchy
since chained is an all-or-nothing approach, and that needs to happen
over the whole set of Qualcomm drivers.

Sooner or later I guess I will get to it unless Bjorn or Brian or someone
else beats me to it.

Yours,
Linus Walleij