The patch titled
Subject: mm: page_vma_mapped_walk(): use pmd_read_atomic()
has been added to the -mm tree. Its filename is
mm-page_vma_mapped_walk-use-pmd_read_atomic.patch
This patch should soon appear at
https://ozlabs.org/~akpm/mmots/broken-out/mm-page_vma_mapped_walk-use-pmd_read_atomic.patch
and later at
https://ozlabs.org/~akpm/mmotm/broken-out/mm-page_vma_mapped_walk-use-pmd_read_atomic.patch
Before you just go and hit "reply", please:
a) Consider who else should be cc'ed
b) Prefer to cc a suitable mailing list as well
c) Ideally: find the original patch on the mailing list and do a
reply-to-all to that, adding suitable additional cc's
*** Remember to use Documentation/process/submit-checklist.rst when testing your code ***
The -mm tree is included into linux-next and is updated
there every 3-4 working days
------------------------------------------------------
From: Hugh Dickins <hughd@xxxxxxxxxx>
Subject: mm: page_vma_mapped_walk(): use pmd_read_atomic()
page_vma_mapped_walk() cleanup: use pmd_read_atomic() with barrier()
instead of READ_ONCE() for pmde: some architectures (e.g. i386 with PAE)
have a multi-word pmd entry, for which READ_ONCE() is not good enough.
Link: https://lkml.kernel.org/r/594c1f0-d396-5346-1f36-606872cddb18@xxxxxxxxxx
Signed-off-by: Hugh Dickins <hughd@xxxxxxxxxx>
Cc: Alistair Popple <apopple@xxxxxxxxxx>
Cc: "Kirill A. Shutemov" <kirill.shutemov@xxxxxxxxxxxxxxx>
Cc: Matthew Wilcox <willy@xxxxxxxxxxxxx>
Cc: Peter Xu <peterx@xxxxxxxxxx>
Cc: Ralph Campbell <rcampbell@xxxxxxxxxx>
Cc: Wang Yugui <wangyugui@xxxxxxxxxxxx>
Cc: Will Deacon <will@xxxxxxxxxx>
Cc: Yang Shi <shy828301@xxxxxxxxx>
Cc: Zi Yan <ziy@xxxxxxxxxx>
Cc: <stable@xxxxxxxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
mm/page_vma_mapped.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
--- a/mm/page_vma_mapped.c~mm-page_vma_mapped_walk-use-pmd_read_atomic
+++ a/mm/page_vma_mapped.c
@@ -182,13 +182,16 @@ restart:
pud = pud_offset(p4d, pvmw->address);
if (!pud_present(*pud))
return false;
+
pvmw->pmd = pmd_offset(pud, pvmw->address);
/*
* Make sure the pmd value isn't cached in a register by the
* compiler and used as a stale value after we've observed a
* subsequent update.
*/
- pmde = READ_ONCE(*pvmw->pmd);
+ pmde = pmd_read_atomic(pvmw->pmd);
+ barrier();
+
if (pmd_trans_huge(pmde) || is_pmd_migration_entry(pmde)) {
pvmw->ptl = pmd_lock(mm, pvmw->pmd);
if (likely(pmd_trans_huge(*pvmw->pmd))) {
_
Patches currently in -mm which might be from hughd@xxxxxxxxxx are
mm-thp-fix-__split_huge_pmd_locked-on-shmem-migration-entry.patch
mm-thp-make-is_huge_zero_pmd-safe-and-quicker.patch
mm-thp-try_to_unmap-use-ttu_sync-for-safe-splitting.patch
mm-thp-fix-vma_address-if-virtual-address-below-file-offset.patch
mm-thp-unmap_mapping_page-to-fix-thp-truncate_cleanup_page.patch
mm-page_vma_mapped_walk-use-page-for-pvmw-page.patch
mm-page_vma_mapped_walk-settle-pagehuge-on-entry.patch
mm-page_vma_mapped_walk-use-pmd_read_atomic.patch
mm-page_vma_mapped_walk-use-pmde-for-pvmw-pmd.patch
mm-page_vma_mapped_walk-prettify-pvmw_migration-block.patch
mm-page_vma_mapped_walk-crossing-page-table-boundary.patch
mm-page_vma_mapped_walk-add-a-level-of-indentation.patch
mm-page_vma_mapped_walk-use-goto-instead-of-while-1.patch
mm-page_vma_mapped_walk-get-vma_address_end-earlier.patch
mm-thp-fix-page_vma_mapped_walk-if-thp-mapped-by-ptes.patch
mm-thp-another-pvmw_sync-fix-in-page_vma_mapped_walk.patch
mm-thp-remap_page-is-only-needed-on-anonymous-thp.patch
mm-hwpoison_user_mappings-try_to_unmap-with-ttu_sync.patch
