mm/hugetlb: convert hugetlb_change_protection() to folios

The for loop inside hugetlb_change_protection() increments by the huge
page size:

psize = huge_page_size(h);
for (; address < end; address += psize)

so we are operating on the head page of the huge pages between address and
end. We can safely convert the struct page usage to struct folio.

Link: https://lkml.kernel.org/r/[e-mail]
Signed-off-by: Sidhartha Kumar <[e-mail]>
Reviewed-by: Matthew Wilcox (Oracle) <[e-mail]>
Reviewed-by: Oscar Salvador <[e-mail]>
Cc: Muchun Song <[e-mail]>
Cc: Sidhartha Kumar <[e-mail]>
Signed-off-by: Andrew Morton <[e-mail]>

mm/hugetlb.c

···
 			/* Nothing to do. */
 		} else if (unlikely(is_hugetlb_entry_migration(pte))) {
 			swp_entry_t entry = pte_to_swp_entry(pte);
-			struct page *page = pfn_swap_entry_to_page(entry);
+			struct folio *folio = pfn_swap_entry_folio(entry);
 			pte_t newpte = pte;
 
 			if (is_writable_migration_entry(entry)) {
-				if (PageAnon(page))
+				if (folio_test_anon(folio))
 					entry = make_readable_exclusive_migration_entry(
 								swp_offset(entry));
 				else