On Wed, 2018-03-07 at 15:01 -0800, Andrew Morton wrote:
> On Wed, 7 Mar 2018 11:32:27 -0700 Toshi Kani <toshi.k...@hpe.com> wrote:
> > Implement pud_free_pmd_page() and pmd_free_pte_page() on x86, which
> > clear a given pud/pmd entry and free up lower level page table(s).
> > Address range associated with the pud/pmd entry must have been purged
> > by INVLPG.
> OK, now we have implementations which match the naming ;) Again, is a
> cc:stable warranted?
Right. This patch 2/2 fixes the memory leak on x86.
Fixes: e61ce6ade404e ("mm: change ioremap to set up huge I/O mappings")
Patch 1/2 fixes the panic on arm64.
Fixes: 324420bf91f60 ("arm64: add support for ioremap() block mappings")
> Do you have any preferences/suggestions as to which tree these should
> be merged through? You're hitting core, arm and x86.
No, I do not have any preference.