On Tue, Aug 22, 2017 at 3:52 PM, Linus Torvalds
<[email protected]> wrote:
>
> The *other* memory policies look fairly sane. They basically have a
> fairly well-defined preferred node for the policy (although the
> "MPOL_INTERLEAVE" looks wrong for a hugepage). But
> MPOL_PREFERRED/MPOL_F_LOCAL really looks completely broken.
Of course, I don't know if that customer test-case actually triggers
that MPOL_PREFERRED/MPOL_F_LOCAL case at all.
So again, that issue may not even be what is going on.
Linus
