From: "Aneesh Kumar K.V" <aneesh.kumar@xxxxxxxxxxxxxxxxxx> Subject: mm/gup: check for protnone only if it is a PTE entry Do the prot_none/FOLL_NUMA check after we are sure this is a THP pte. Archs can implement prot_none such that it can return true for regular pmd entries. Link: http://lkml.kernel.org/r/1487498326-8734-1-git-send-email-aneesh.kumar@xxxxxxxxxxxxxxxxxx Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@xxxxxxxxxxxxxxxxxx> Cc: Rik van Riel <riel@xxxxxxxxxxx> Cc: Mel Gorman <mgorman@xxxxxxxxxxxxxxxxxxx> Cc: Hillf Danton <hillf.zj@xxxxxxxxxxxxxxx> Cc: Kirill A. Shutemov <kirill.shutemov@xxxxxxxxxxxxxxx> Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> --- mm/gup.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff -puN mm/gup.c~mm-gup-check-for-protnone-only-if-it-is-a-pte-entry mm/gup.c --- a/mm/gup.c~mm-gup-check-for-protnone-only-if-it-is-a-pte-entry +++ a/mm/gup.c @@ -272,8 +272,6 @@ struct page *follow_page_mask(struct vm_ return page; return no_page_table(vma, flags); } - if ((flags & FOLL_NUMA) && pmd_protnone(*pmd)) - return no_page_table(vma, flags); if (pmd_devmap(*pmd)) { ptl = pmd_lock(mm, pmd); page = follow_devmap_pmd(vma, address, pmd, flags); @@ -284,6 +282,9 @@ struct page *follow_page_mask(struct vm_ if (likely(!pmd_trans_huge(*pmd))) return follow_page_pte(vma, address, pmd, flags); + if ((flags & FOLL_NUMA) && pmd_protnone(*pmd)) + return no_page_table(vma, flags); + ptl = pmd_lock(mm, pmd); if (unlikely(!pmd_trans_huge(*pmd))) { spin_unlock(ptl); _ -- To unsubscribe from this list: send the line "unsubscribe mm-commits" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
