mm: page_isolation: use compound_nr() correctly in isolate_single_pageblock()

When compound_nr(page) was used, page was not guaranteed to be the head of
the compound page and it could cause an infinite loop.  Fix it by calling
it on the head page.

Link: https://lkml.kernel.org/r/20220531024450.2498431-1-zi.yan@sent.com
Fixes: b2c9e2fbba32 ("mm: make alloc_contig_range work at pageblock granularity")
Signed-off-by: Zi Yan <ziy@nvidia.com>
Reported-by: Anshuman Khandual <anshuman.khandual@arm.com>
Link: https://lore.kernel.org/linux-mm/20220530115027.123341-1-anshuman.khandual@arm.com/
Reviewed-by: Anshuman Khandual <anshuman.khandual@arm.com>
Reviewed-by: Miaohe Lin <linmiaohe@huawei.com>
Reviewed-by: Oscar Salvador <osalvador@suse.de>
Acked-by: Muchun Song <songmuchun@bytedance.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Qian Cai <quic_qiancai@quicinc.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Eric Ren <renzhengeek@gmail.com>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: Christophe Leroy <christophe.leroy@csgroup.eu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

diff --git a/mm/page_isolation.c b/mm/page_isolation.c
index 6021f84..d200d41 100644 (file)
--- a/mm/page_isolation.c
+++ b/mm/page_isolation.c
@@ -385,9 +385,9 @@ static int isolate_single_pageblock(unsigned long boundary_pfn, int flags,
                 * above do the rest. If migration is not possible, just fail.
                 */
                if (PageCompound(page)) {
-                       unsigned long nr_pages = compound_nr(page);
                        struct page *head = compound_head(page);
                        unsigned long head_pfn = page_to_pfn(head);
+                       unsigned long nr_pages = compound_nr(head);
 
                        if (head_pfn + nr_pages <= boundary_pfn) {
                                pfn = head_pfn + nr_pages;