mm: convert free_transhuge_folio() to folio_undo_large_rmappable()

Indirect calls are expensive, thanks to Spectre.  Test for
TRANSHUGE_PAGE_DTOR and destroy the folio appropriately.  Move the
free_compound_page() call into destroy_large_folio() to simplify later
patches.

Link: https://lkml.kernel.org/r/20230816151201.3655946-5-willy@infradead.orgSigned-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: David Hildenbrand <david@redhat.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Sidhartha Kumar <sidhartha.kumar@oracle.com>
Cc: Yanteng Si <siyanteng@loongson.cn>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

include/linux/huge_mm.h

@@ -141,8 +141,6 @@ unsigned long thp_get_unmapped_area(struct file *filp, unsigned long addr,
 		unsigned long len, unsigned long pgoff, unsigned long flags);
 
 void prep_transhuge_page(struct page *page);
-void free_transhuge_page(struct page *page);
-
 bool can_split_folio(struct folio *folio, int *pextra_pins);
 int split_huge_page_to_list(struct page *page, struct list_head *list);
 static inline int split_huge_page(struct page *page)

include/linux/mm.h

@@ -1253,9 +1253,7 @@ enum compound_dtor_id {
 #ifdef CONFIG_HUGETLB_PAGE
 	HUGETLB_PAGE_DTOR,
 #endif
-#ifdef CONFIG_TRANSPARENT_HUGEPAGE
 	TRANSHUGE_PAGE_DTOR,
-#endif
 	NR_COMPOUND_DTORS,
 };
 

mm/huge_memory.c

@@ -2776,10 +2776,9 @@ int split_huge_page_to_list(struct page *page, struct list_head *list)
 	return ret;
 }
 
-void free_transhuge_page(struct page *page)
+void folio_undo_large_rmappable(struct folio *folio)
 {
-	struct folio *folio = (struct folio *)page;
-	struct deferred_split *ds_queue = get_deferred_split_queue(folio);
+	struct deferred_split *ds_queue;
 	unsigned long flags;
 
 	/*
@@ -2787,15 +2786,16 @@ void free_transhuge_page(struct page *page)
 	 * deferred_list. If folio is not in deferred_list, it's safe
 	 * to check without acquiring the split_queue_lock.
 	 */
-	if (data_race(!list_empty(&folio->_deferred_list))) {
-		spin_lock_irqsave(&ds_queue->split_queue_lock, flags);
-		if (!list_empty(&folio->_deferred_list)) {
-			ds_queue->split_queue_len--;
-			list_del(&folio->_deferred_list);
-		}
-		spin_unlock_irqrestore(&ds_queue->split_queue_lock, flags);
+	if (data_race(list_empty(&folio->_deferred_list)))
+		return;
+
+	ds_queue = get_deferred_split_queue(folio);
+	spin_lock_irqsave(&ds_queue->split_queue_lock, flags);
+	if (!list_empty(&folio->_deferred_list)) {
+		ds_queue->split_queue_len--;
+		list_del(&folio->_deferred_list);
 	}
-	free_compound_page(page);
+	spin_unlock_irqrestore(&ds_queue->split_queue_lock, flags);
 }
 
 void deferred_split_folio(struct folio *folio)

mm/internal.h

@@ -413,6 +413,8 @@ static inline void folio_set_order(struct folio *folio, unsigned int order)
 #endif
 }
 
+void folio_undo_large_rmappable(struct folio *folio);
+
 static inline void prep_compound_head(struct page *page, unsigned int order)
 {
 	struct folio *folio = (struct folio *)page;

mm/page_alloc.c

@@ -287,9 +287,6 @@ const char * const migratetype_names[MIGRATE_TYPES] = {
 static compound_page_dtor * const compound_page_dtors[NR_COMPOUND_DTORS] = {
 	[NULL_COMPOUND_DTOR] = NULL,
 	[COMPOUND_PAGE_DTOR] = free_compound_page,
-#ifdef CONFIG_TRANSPARENT_HUGEPAGE
-	[TRANSHUGE_PAGE_DTOR] = free_transhuge_page,
-#endif
 };
 
 int min_free_kbytes = 1024;
@@ -614,6 +611,12 @@ void destroy_large_folio(struct folio *folio)
 		return;
 	}
 
+	if (folio_test_transhuge(folio) && dtor == TRANSHUGE_PAGE_DTOR) {
+		folio_undo_large_rmappable(folio);
+		free_compound_page(&folio->page);
+		return;
+	}
+
 	VM_BUG_ON_FOLIO(dtor >= NR_COMPOUND_DTORS, folio);
 	compound_page_dtors[dtor](&folio->page);
 }