memory-hotplug: skip HWPoisoned page when offlining pages

hwpoisoned may be set when we offline a page by the sysfs interface
/sys/devices/system/memory/soft_offline_page or
/sys/devices/system/memory/hard_offline_page. If we don't clear
this flag when onlining pages, this page can't be freed, and will
not in free list. So we can't offline these pages again. So we
should skip such page when offlining pages.

Signed-off-by: Wen Congyang <wency@cn.fujitsu.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Jiang Liu <liuj97@gmail.com>
Cc: Len Brown <len.brown@intel.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Christoph Lameter <cl@linux.com>
Cc: Minchan Kim <minchan.kim@gmail.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Mel Gorman <mel@csn.ul.ie>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

4 files changed, 47 insertions, 14 deletions

diff --git a/mm/memory-failure.c b/mm/memory-failure.c
index 8b20278be6a6..2c9fc7340b12 100644
--- a/mm/memory-failure.c
+++ b/mm/memory-failure.c
@@ -1385,7 +1385,7 @@ static int get_any_page(struct page *p, unsigned long pfn, int flags)
 	 * Isolate the page, so that it doesn't get reallocated if it
 	 * was free.
 	 */
-	set_migratetype_isolate(p);
+	set_migratetype_isolate(p, true);
 	/*
 	 * When the target page is a free hugepage, just remove it
 	 * from free hugepage list.
diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c
index e4eeacae2b91..0095d156324a 100644
--- a/mm/memory_hotplug.c
+++ b/mm/memory_hotplug.c
@@ -847,7 +847,7 @@ check_pages_isolated_cb(unsigned long start_pfn, unsigned long nr_pages,
 {
 	int ret;
 	long offlined = *(long *)data;
-	ret = test_pages_isolated(start_pfn, start_pfn + nr_pages);
+	ret = test_pages_isolated(start_pfn, start_pfn + nr_pages, true);
 	offlined = nr_pages;
 	if (!ret)
 		*(long *)data += offlined;
@@ -894,7 +894,8 @@ static int __ref __offline_pages(unsigned long start_pfn,
 	nr_pages = end_pfn - start_pfn;
 
 	/* set above range as isolated */
-	ret = start_isolate_page_range(start_pfn, end_pfn, MIGRATE_MOVABLE);
+	ret = start_isolate_page_range(start_pfn, end_pfn,
+				       MIGRATE_MOVABLE, true);
 	if (ret)
 		goto out;
 
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index a49b0ea3cc2f..6f50cfe98a7b 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -5616,7 +5616,8 @@ void set_pageblock_flags_group(struct page *page, unsigned long flags,
  * MIGRATE_MOVABLE block might include unmovable pages. It means you can't
  * expect this function should be exact.
  */
-bool has_unmovable_pages(struct zone *zone, struct page *page, int count)
+bool has_unmovable_pages(struct zone *zone, struct page *page, int count,
+			 bool skip_hwpoisoned_pages)
 {
 	unsigned long pfn, iter, found;
 	int mt;
@@ -5651,6 +5652,13 @@ bool has_unmovable_pages(struct zone *zone, struct page *page, int count)
 			continue;
 		}
 
+		/*
+		 * The HWPoisoned page may be not in buddy system, and
+		 * page_count() is not 0.
+		 */
+		if (skip_hwpoisoned_pages && PageHWPoison(page))
+			continue;
+
 		if (!PageLRU(page))
 			found++;
 		/*
@@ -5693,7 +5701,7 @@ bool is_pageblock_removable_nolock(struct page *page)
 			zone->zone_start_pfn + zone->spanned_pages <= pfn)
 		return false;
 
-	return !has_unmovable_pages(zone, page, 0);
+	return !has_unmovable_pages(zone, page, 0, true);
 }
 
 #ifdef CONFIG_CMA
@@ -5864,7 +5872,8 @@ int alloc_contig_range(unsigned long start, unsigned long end,
 	 */
 
 	ret = start_isolate_page_range(pfn_max_align_down(start),
-				       pfn_max_align_up(end), migratetype);
+				       pfn_max_align_up(end), migratetype,
+				       false);
 	if (ret)
 		return ret;
 
@@ -5903,7 +5912,7 @@ int alloc_contig_range(unsigned long start, unsigned long end,
 	}
 
 	/* Make sure the range is really isolated. */
-	if (test_pages_isolated(outer_start, end)) {
+	if (test_pages_isolated(outer_start, end, false)) {
 		pr_warn("alloc_contig_range test_pages_isolated(%lx, %lx) failed\n",
 		       outer_start, end);
 		ret = -EBUSY;
@@ -6018,6 +6027,16 @@ __offline_isolated_pages(unsigned long start_pfn, unsigned long end_pfn)
 			continue;
 		}
 		page = pfn_to_page(pfn);
+		/*
+		 * The HWPoisoned page may be not in buddy system, and
+		 * page_count() is not 0.
+		 */
+		if (unlikely(!PageBuddy(page) && PageHWPoison(page))) {
+			pfn++;
+			SetPageReserved(page);
+			continue;
+		}
+
 		BUG_ON(page_count(page));
 		BUG_ON(!PageBuddy(page));
 		order = page_order(page);
diff --git a/mm/page_isolation.c b/mm/page_isolation.c
index f2f5b4818e94..9d2264ea4606 100644
--- a/mm/page_isolation.c
+++ b/mm/page_isolation.c
@@ -30,7 +30,7 @@ static void restore_pageblock_isolate(struct page *page, int migratetype)
 	zone->nr_pageblock_isolate--;
 }
 
-int set_migratetype_isolate(struct page *page)
+int set_migratetype_isolate(struct page *page, bool skip_hwpoisoned_pages)
 {
 	struct zone *zone;
 	unsigned long flags, pfn;
@@ -66,7 +66,8 @@ int set_migratetype_isolate(struct page *page)
 	 * FIXME: Now, memory hotplug doesn't call shrink_slab() by itself.
 	 * We just check MOVABLE pages.
 	 */
-	if (!has_unmovable_pages(zone, page, arg.pages_found))
+	if (!has_unmovable_pages(zone, page, arg.pages_found,
+				 skip_hwpoisoned_pages))
 		ret = 0;
 
 	/*
@@ -134,7 +135,7 @@ __first_valid_page(unsigned long pfn, unsigned long nr_pages)
  * Returns 0 on success and -EBUSY if any part of range cannot be isolated.
  */
 int start_isolate_page_range(unsigned long start_pfn, unsigned long end_pfn,
-			     unsigned migratetype)
+			     unsigned migratetype, bool skip_hwpoisoned_pages)
 {
 	unsigned long pfn;
 	unsigned long undo_pfn;
@@ -147,7 +148,8 @@ int start_isolate_page_range(unsigned long start_pfn, unsigned long end_pfn,
 	     pfn < end_pfn;
 	     pfn += pageblock_nr_pages) {
 		page = __first_valid_page(pfn, pageblock_nr_pages);
-		if (page && set_migratetype_isolate(page)) {
+		if (page &&
+		    set_migratetype_isolate(page, skip_hwpoisoned_pages)) {
 			undo_pfn = pfn;
 			goto undo;
 		}
@@ -190,7 +192,8 @@ int undo_isolate_page_range(unsigned long start_pfn, unsigned long end_pfn,
  * Returns 1 if all pages in the range are isolated.
  */
 static int
-__test_page_isolated_in_pageblock(unsigned long pfn, unsigned long end_pfn)
+__test_page_isolated_in_pageblock(unsigned long pfn, unsigned long end_pfn,
+				  bool skip_hwpoisoned_pages)
 {
 	struct page *page;
 
@@ -220,6 +223,14 @@ __test_page_isolated_in_pageblock(unsigned long pfn, unsigned long end_pfn)
 		else if (page_count(page) == 0 &&
 			get_freepage_migratetype(page) == MIGRATE_ISOLATE)
 			pfn += 1;
+		else if (skip_hwpoisoned_pages && PageHWPoison(page)) {
+			/*
+			 * The HWPoisoned page may be not in buddy
+			 * system, and page_count() is not 0.
+			 */
+			pfn++;
+			continue;
+		}
 		else
 			break;
 	}
@@ -228,7 +239,8 @@ __test_page_isolated_in_pageblock(unsigned long pfn, unsigned long end_pfn)
 	return 1;
 }
 
-int test_pages_isolated(unsigned long start_pfn, unsigned long end_pfn)
+int test_pages_isolated(unsigned long start_pfn, unsigned long end_pfn,
+			bool skip_hwpoisoned_pages)
 {
 	unsigned long pfn, flags;
 	struct page *page;
@@ -251,7 +263,8 @@ int test_pages_isolated(unsigned long start_pfn, unsigned long end_pfn)
 	/* Check all pages are free or Marked as ISOLATED */
 	zone = page_zone(page);
 	spin_lock_irqsave(&zone->lock, flags);
-	ret = __test_page_isolated_in_pageblock(start_pfn, end_pfn);
+	ret = __test_page_isolated_in_pageblock(start_pfn, end_pfn,
+						skip_hwpoisoned_pages);
 	spin_unlock_irqrestore(&zone->lock, flags);
 	return ret ? 0 : -EBUSY;
 }