[PATCH] (for testing) revert change around removing MF_COUNT_INCREASED

[Naoya Horiguchi]

revert the following patches
- mm,madvise: call soft_offline_page() without MF_COUNT_INCREASED
- mm,madvise: Refactor madvise_inject_error
- mm,hwpoison: remove MF_COUNT_INCREASED
- mm,hwpoison: remove flag argument from soft offline functions

Signed-off-by: Naoya Horiguchi <naoya.horiguchi@xxxxxxx>
---
 drivers/base/memory.c |  2 +-
 include/linux/mm.h    |  9 +++++----
 mm/madvise.c          | 36 +++++++++++++++++++-----------------
 mm/memory-failure.c   | 31 ++++++++++++++++++++-----------
 4 files changed, 45 insertions(+), 33 deletions(-)

diff --git a/drivers/base/memory.c b/drivers/base/memory.c
index 3e6d27c9dff6..4db3c660de83 100644
--- a/drivers/base/memory.c
+++ b/drivers/base/memory.c
@@ -463,7 +463,7 @@ static ssize_t soft_offline_page_store(struct device *d=
ev,
 	if (kstrtoull(buf, 0, &pfn) < 0)
 		return -EINVAL;
 	pfn >>=3D PAGE_SHIFT;
-	ret =3D soft_offline_page(pfn);
+	ret =3D soft_offline_page(pfn, 0);
 	return ret =3D=3D 0 ? count : ret;
 }
=20
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 4f12b2465e80..442921a004a2 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -2976,9 +2976,10 @@ void register_page_bootmem_memmap(unsigned long sect=
ion_nr, struct page *map,
 				  unsigned long nr_pages);
=20
 enum mf_flags {
-	MF_ACTION_REQUIRED =3D 1 << 0,
-	MF_MUST_KILL =3D 1 << 1,
-	MF_SOFT_OFFLINE =3D 1 << 2,
+	MF_COUNT_INCREASED =3D 1 << 0,
+	MF_ACTION_REQUIRED =3D 1 << 1,
+	MF_MUST_KILL =3D 1 << 2,
+	MF_SOFT_OFFLINE =3D 1 << 3,
 };
 extern int memory_failure(unsigned long pfn, int flags);
 extern void memory_failure_queue(unsigned long pfn, int flags);
@@ -2988,7 +2989,7 @@ extern int sysctl_memory_failure_early_kill;
 extern int sysctl_memory_failure_recovery;
 extern void shake_page(struct page *p, int access);
 extern atomic_long_t num_poisoned_pages __read_mostly;
-extern int soft_offline_page(unsigned long pfn);
+extern int soft_offline_page(unsigned long pfn, int flags);
=20
=20
 /*
diff --git a/mm/madvise.c b/mm/madvise.c
index 843f6fad3b89..5fa5f66468b3 100644
--- a/mm/madvise.c
+++ b/mm/madvise.c
@@ -886,15 +886,16 @@ static long madvise_remove(struct vm_area_struct *vma=
,
 static int madvise_inject_error(int behavior,
 		unsigned long start, unsigned long end)
 {
+	struct page *page;
 	struct zone *zone;
 	unsigned long size;
=20
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
=20
+
 	for (; start < end; start +=3D size) {
 		unsigned long pfn;
-		struct page *page;
 		int ret;
=20
 		ret =3D get_user_pages_fast(start, 1, 0, &page);
@@ -909,26 +910,27 @@ static int madvise_inject_error(int behavior,
 		 */
 		size =3D page_size(compound_head(page));
=20
-		/*
-		 * The get_user_pages_fast() is just to get the pfn of the
-		 * given address, and the refcount has nothing to do with
-		 * what we try to test, so it should be released immediately.
-		 * This is racy but it's intended because the real hardware
-		 * errors could happen at any moment and memory error handlers
-		 * must properly handle the race.
-		 */
-		put_page(page);
-
 		if (behavior =3D=3D MADV_SOFT_OFFLINE) {
 			pr_info("Soft offlining pfn %#lx at process virtual address %#lx\n",
-				pfn, start);
-			ret =3D soft_offline_page(pfn);
-		} else {
-			pr_info("Injecting memory failure for pfn %#lx at process virtual addre=
ss %#lx\n",
-				pfn, start);
-			ret =3D memory_failure(pfn, 0);
+					pfn, start);
+
+			ret =3D soft_offline_page(pfn, MF_COUNT_INCREASED);
+			if (ret)
+				return ret;
+			continue;
 		}
=20
+		pr_info("Injecting memory failure for pfn %#lx at process virtual addres=
s %#lx\n",
+				pfn, start);
+
+		/*
+		 * Drop the page reference taken by get_user_pages_fast(). In
+		 * the absence of MF_COUNT_INCREASED the memory_failure()
+		 * routine is responsible for pinning the page to prevent it
+		 * from being released back to the page allocator.
+		 */
+		put_page(page);
+		ret =3D memory_failure(pfn, 0);
 		if (ret)
 			return ret;
 	}
diff --git a/mm/memory-failure.c b/mm/memory-failure.c
index a229d4694954..fd50a6f9a60d 100644
--- a/mm/memory-failure.c
+++ b/mm/memory-failure.c
@@ -1167,7 +1167,7 @@ static int memory_failure_hugetlb(unsigned long pfn, =
int flags)
=20
 	num_poisoned_pages_inc();
=20
-	if (!get_hwpoison_page(p)) {
+	if (!(flags & MF_COUNT_INCREASED) && !get_hwpoison_page(p)) {
 		/*
 		 * Check "filter hit" and "race with other subpage."
 		 */
@@ -1363,7 +1363,7 @@ int memory_failure(unsigned long pfn, int flags)
 	 * In fact it's dangerous to directly bump up page count from 0,
 	 * that may make page_ref_freeze()/page_ref_unfreeze() mismatch.
 	 */
-	if (!get_hwpoison_page(p)) {
+	if (!(flags & MF_COUNT_INCREASED) && !get_hwpoison_page(p)) {
 		if (is_free_buddy_page(p)) {
 			action_result(pfn, MF_MSG_BUDDY, MF_DELAYED);
 			return 0;
@@ -1392,7 +1392,10 @@ int memory_failure(unsigned long pfn, int flags)
 	shake_page(p, 0);
 	/* shake_page could have turned it free. */
 	if (!PageLRU(p) && is_free_buddy_page(p)) {
-		action_result(pfn, MF_MSG_BUDDY_2ND, MF_DELAYED);
+		if (flags & MF_COUNT_INCREASED)
+			action_result(pfn, MF_MSG_BUDDY, MF_DELAYED);
+		else
+			action_result(pfn, MF_MSG_BUDDY_2ND, MF_DELAYED);
 		return 0;
 	}
=20
@@ -1540,7 +1543,7 @@ static void memory_failure_work_func(struct work_stru=
ct *work)
 		if (!gotten)
 			break;
 		if (entry.flags & MF_SOFT_OFFLINE)
-			soft_offline_page(entry.pfn);
+			soft_offline_page(entry.pfn, entry.flags);
 		else
 			memory_failure(entry.pfn, entry.flags);
 	}
@@ -1679,10 +1682,13 @@ EXPORT_SYMBOL(unpoison_memory);
  * that is not free, and 1 for any other page type.
  * For 1 the page is returned with increased page count, otherwise not.
  */
-static int __get_any_page(struct page *p, unsigned long pfn)
+static int __get_any_page(struct page *p, unsigned long pfn, int flags)
 {
 	int ret;
=20
+	if (flags & MF_COUNT_INCREASED)
+		return 1;
+
 	/*
 	 * When the target page is a free hugepage, just remove it
 	 * from free hugepage list.
@@ -1709,12 +1715,12 @@ static int __get_any_page(struct page *p, unsigned =
long pfn)
 	return ret;
 }
=20
-static int get_any_page(struct page *page, unsigned long pfn)
+static int get_any_page(struct page *page, unsigned long pfn, int flags)
 {
-	int ret =3D __get_any_page(page, pfn);
+	int ret =3D __get_any_page(page, pfn, flags);
=20
 	if (ret =3D=3D -EBUSY)
-		ret =3D __get_any_page(page, pfn);
+		ret =3D __get_any_page(page, pfn, flags);
=20
 	if (ret =3D=3D 1 && !PageHuge(page) &&
 	    !PageLRU(page) && !__PageMovable(page)) {
@@ -1727,7 +1733,7 @@ static int get_any_page(struct page *page, unsigned l=
ong pfn)
 		/*
 		 * Did it turn free?
 		 */
-		ret =3D __get_any_page(page, pfn);
+		ret =3D __get_any_page(page, pfn, 0);
 		if (ret =3D=3D 1 && !PageLRU(page)) {
 			/* Drop page reference which is from __get_any_page() */
 			put_page(page);
@@ -1869,6 +1875,7 @@ static int soft_offline_free_page(struct page *page)
 /**
  * soft_offline_page - Soft offline a page.
  * @pfn: pfn to soft-offline
+ * @flags: flags. Same as memory_failure().
  *
  * Returns 0 on success, otherwise negated errno.
  *
@@ -1887,7 +1894,7 @@ static int soft_offline_free_page(struct page *page)
  * This is not a 100% solution for all memory, but tries to be
  * ``good enough'' for the majority of memory.
  */
-int soft_offline_page(unsigned long pfn)
+int soft_offline_page(unsigned long pfn, int flags)
 {
 	int ret;
 	struct page *page;
@@ -1901,11 +1908,13 @@ int soft_offline_page(unsigned long pfn)
=20
 	if (PageHWPoison(page)) {
 		pr_info("soft offline: %#lx page already poisoned\n", pfn);
+		if (flags & MF_COUNT_INCREASED)
+			put_page(page);
 		return 0;
 	}
=20
 	get_online_mems();
-	ret =3D get_any_page(page, pfn);
+	ret =3D get_any_page(page, pfn, flags);
 	put_online_mems();
=20
 	if (ret > 0)
--=20
2.25.1