Re: [PATCH V1] iommu/hyperv: Create hyperv subdirectory under drivers/iommu

[Mukesh R]

On 6/1/26 21:45, Jacob Pan wrote:
> Hi Mukesh,
>
> On Mon, 1 Jun 2026 14:23:18 -0700
> Mukesh R <mrathor@xxxxxxxxxxxxxxxxxxx> wrote:
>
> > > > diff --git a/drivers/iommu/hyperv/Makefile
> > > > b/drivers/iommu/hyperv/Makefile new file mode 100644
> > > > index 000000000000..0053e00e08e6
> > > > --- /dev/null
> > > > +++ b/drivers/iommu/hyperv/Makefile
> > > > @@ -0,0 +1,2 @@
> > > > +# SPDX-License-Identifier: GPL-2.0
> > > > +obj-$(CONFIG_IRQ_REMAP) += hv-irq-remap.o
> > > Should the name be x86 specific? This file will never be built for
> > > ARM because:
> > >    CONFIG_IRQ_REMAP depends on:
> > >    depends on X86_64 && X86_IO_APIC && PCI_MSI && ACPI
> > >
> > > perhaps, hv-x86-irq-remap.c? hv-irq-remap.c sounds like a
> > > generic Hyper-V IRQ-remapping backend and may be misleading once
> > > arm64 Hyper-V IOMMU/interrupt support grows under this directory.
> >
> > we could, but looking at early version of hyperv-iommu-arm.c, it looks
> > very similar, and so if there are very few ifdefs, we could just
> > keep one file rather than replicating bunch of code.
>
> With this patch as is, how could you keep one file for both ARM and X86
> where hv-irq-remap.c does not build on ARM?

We can remove the x86 restriction, make sure it builds on arm. Then
during arm patch, I can help decide whether to fork or just add ifdefs
in the file. I'll submit V2 with x86 restriction removed.

Thanks,
-Mukesh