[PATCH 10/18] arm64: Support hugetlb vmemmap optimization

[James Houghton]

On Tue, Aug 18, 2026 at 7:16 AM Catalin Marinas <catalin.marinas at arm.com> wrote:
>
> On Wed, Jul 08, 2026 at 03:11:20AM +0000, James Houghton wrote:
> > diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
> > index b3afe0688919..cf4bd6bd7f11 100644
> > --- a/arch/arm64/Kconfig
> > +++ b/arch/arm64/Kconfig
> > @@ -259,6 +259,7 @@ config ARM64
> >       select USER_STACKTRACE_SUPPORT
> >       select VDSO_GETRANDOM
> >       select VMAP_STACK
> > +     select ARCH_WANT_OPTIMIZE_HUGETLB_VMEMMAP if ARM64_WANT_OPTIMIZE_HUGETLB_VMEMMAP
>
> Nit: we try to keep this list alphabetically ordered.
>
> >       help
> >         ARM 64-bit (AArch64) Linux support.
> >
> > @@ -1607,6 +1608,17 @@ config ARCH_SPARSEMEM_ENABLE
> >       def_bool y
> >       select SPARSEMEM_VMEMMAP_ENABLE
> >
> > +config ARM64_WANT_OPTIMIZE_HUGETLB_VMEMMAP
> > +     bool "Enable support for HugeTLB Vmemmap Optimization (HVO)"
> > +     default n
>
> Nit: 'n' is the default already.
>
> > +     depends on HUGETLB_PAGE && SPARSEMEM_VMEMMAP && ARM64_HW_AFDBM
> > +     help
> > +       Enable support for HVO. When enabled, late-onlining of CPUs that do
>
> I'd drop 'of', not sure how it reads better. Maybe just 'late CPUs ...
> will not be onlined'.

Thanks for the feedback; I've incorporated it. But I think this patch
is going away in v2 (with patches 14-18).

> > +       not support features required by HVO will not be onlined if all boot
> > +       CPUs do have such support.
> > +
> > +       If unsure, say N
> > +
> >  config HW_PERF_EVENTS
> >       def_bool y
> >       depends on ARM_PMU
>
> --
> Catalin