[PATCH 4.13 097/160] mm, hugetlb, soft_offline: save compound page order before page migration

From: Greg Kroah-Hartman
Date: Tue Oct 10 2017 - 16:23:12 EST


4.13-stable review patch. If anyone has any objections, please let me know.

------------------

From: Alexandru Moise <00moses.alexander00@xxxxxxxxx>

commit 19bfbe22f59a207417b2679e7e83c180419c9ec5 upstream.

This fixes a bug in madvise() where if you'd try to soft offline a
hugepage via madvise(), while walking the address range you'd end up,
using the wrong page offset due to attempting to get the compound order
of a former but presently not compound page, due to dissolving the huge
page (since commit c3114a84f7f9: "mm: hugetlb: soft-offline: dissolve
source hugepage after successful migration").

As a result I ended up with all my free pages except one being offlined.

Link: http://lkml.kernel.org/r/20170912204306.GA12053@xxxxxxxxx
Fixes: c3114a84f7f9 ("mm: hugetlb: soft-offline: dissolve source hugepage after successful migration")
Signed-off-by: Alexandru Moise <00moses.alexander00@xxxxxxxxx>
Cc: Anshuman Khandual <khandual@xxxxxxxxxxxxxxxxxx>
Cc: Michal Hocko <mhocko@xxxxxxxx>
Cc: Andrea Arcangeli <aarcange@xxxxxxxxxx>
Cc: Minchan Kim <minchan@xxxxxxxxxx>
Cc: Hillf Danton <hdanton@xxxxxxxx>
Cc: Shaohua Li <shli@xxxxxx>
Cc: Mike Rapoport <rppt@xxxxxxxxxxxxxxxxxx>
Cc: "Kirill A. Shutemov" <kirill.shutemov@xxxxxxxxxxxxxxx>
Cc: Mel Gorman <mgorman@xxxxxxxxxxxxxxxxxxx>
Cc: David Rientjes <rientjes@xxxxxxxxxx>
Cc: Rik van Riel <riel@xxxxxxxxxx>
Cc: Naoya Horiguchi <n-horiguchi@xxxxxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
Signed-off-by: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>

---
mm/madvise.c | 12 ++++++++++--
1 file changed, 10 insertions(+), 2 deletions(-)

--- a/mm/madvise.c
+++ b/mm/madvise.c
@@ -614,18 +614,26 @@ static int madvise_inject_error(int beha
{
struct page *page;
struct zone *zone;
+ unsigned int order;

if (!capable(CAP_SYS_ADMIN))
return -EPERM;

- for (; start < end; start += PAGE_SIZE <<
- compound_order(compound_head(page))) {
+
+ for (; start < end; start += PAGE_SIZE << order) {
int ret;

ret = get_user_pages_fast(start, 1, 0, &page);
if (ret != 1)
return ret;

+ /*
+ * When soft offlining hugepages, after migrating the page
+ * we dissolve it, therefore in the second loop "page" will
+ * no longer be a compound page, and order will be 0.
+ */
+ order = compound_order(compound_head(page));
+
if (PageHWPoison(page)) {
put_page(page);
continue;