Re: [PATCH] irqchip: Select downstream irqchip drivers for LoongArch CPU

From: Huacai Chen
Date: Mon Aug 08 2022 - 07:24:44 EST


Hi, Marc,

On Mon, Aug 8, 2022 at 7:16 PM Marc Zyngier <maz@xxxxxxxxxx> wrote:
>
> On Mon, 08 Aug 2022 12:06:23 +0100,
> Huacai Chen <chenhuacai@xxxxxxxxx> wrote:
> >
> > Hi, Marc,
> >
> > On Mon, Aug 8, 2022 at 6:05 PM Marc Zyngier <maz@xxxxxxxxxx> wrote:
> > >
> > > On Mon, 08 Aug 2022 09:53:19 +0100,
> > > Huacai Chen <chenhuacai@xxxxxxxxxxx> wrote:
> > > >
> > > > LoongArch irqchips have a fixed hierarchy which currently can't be
> > > > described by ACPI tables, so upstream irqchip drivers call downstream
> > > > irqchip drivers' initialization directly. As a result, the top level
> > > > (CPU-level) irqchip driver should explicitly select downstream drivers
> > > > to avoid build errors.
> > > >
> > > > Reported-by: kernel test robot <lkp@xxxxxxxxx>
> > > > Signed-off-by: Huacai Chen <chenhuacai@xxxxxxxxxxx>
> > > > ---
> > > > drivers/irqchip/Kconfig | 5 +++++
> > > > 1 file changed, 5 insertions(+)
> > > >
> > > > diff --git a/drivers/irqchip/Kconfig b/drivers/irqchip/Kconfig
> > > > index 66b9fa408bf2..2549daa859d6 100644
> > > > --- a/drivers/irqchip/Kconfig
> > > > +++ b/drivers/irqchip/Kconfig
> > > > @@ -561,6 +561,11 @@ config IRQ_LOONGARCH_CPU
> > > > select GENERIC_IRQ_CHIP
> > > > select IRQ_DOMAIN
> > > > select GENERIC_IRQ_EFFECTIVE_AFF_MASK
> > > > + select LOONGSON_LIOINTC
> > > > + select LOONGSON_EIOINTC
> > > > + select LOONGSON_PCH_PIC
> > > > + select LOONGSON_PCH_MSI
> > > > + select LOONGSON_PCH_LPC
> > >
> > > This triggers tons of pretty bad compilation and configuration issues,
> > > as PCI still isn't selectable, even in Linus' tree (see below). I'm
> > > guessing you still have local patches that hide this issue.
> > >
> > > Please test your patches on an upstream tree in the future.
> > I'm very sorry for that. I think this patch should be delayed after
> > 5.20-rc1, then PCI code will have been merged.
>
> Sure, I'm fine with that. However, I have seen the PCI pull request
> going in without any mention of the LoongArch support. Is it going via
> another tree?
PCI drivers have merged, and PCI enablement will go via loongarch
tree. Once I wanted to send PR to Linus today, but Steven said that
the loongarch-next branch breaks some build so I should fix that
first.

Huacai
>
> Thanks,
>
> M.
>
> --
> Without deviation from the norm, progress is not possible.