[PATCH 3/3] mm: Care about shadow stack guard gap when getting an unmapped area

[Lorenzo Stoakes]

On Tue, Sep 03, 2024 at 07:20:02PM GMT, Mark Brown wrote:
> On Tue, Sep 03, 2024 at 06:49:46PM +0100, Lorenzo Stoakes wrote:
> > On Mon, Sep 02, 2024 at 08:08:15PM GMT, Mark Brown wrote:
>
> > > On x86 there is a custom arch_get_unmapped_area() which was updated by the
> > > above commit to cover this case by specifying a start_gap for allocations
> > > with VM_SHADOW_STACK. Both arm64 and RISC-V have equivalent features and
> > > use the generic implementation of arch_get_unmapped_area() so let's make
> > > the equivalent change there so they also don't get shadow stack pages
> > > placed without guard pages.
>
> > Don't you need to unwind that change in x86 now you're doing it in generic code?
>
> No, x86 had a preexisting custom implementation for some other reason
> (hence the "updated by the above commit" part above) - the shadow stack
> support would most likely have been added in the core in the first place
> were it not for that.

Oh yeah missed that!

Other than comment nice-to-have this seems fine:

Acked-by: Lorenzo Stoakes <lorenzo.stoakes at oracle.com>