mm: Use multi-index entries in the page cache

We currently store large folios as 2^N consecutive entries.  While this
consumes rather more memory than necessary, it also turns out to be buggy.
A writeback operation which starts within a tail page of a dirty folio will
not write back the folio as the xarray's dirty bit is only set on the
head index.  With multi-index entries, the dirty bit will be found no
matter where in the folio the operation starts.

This does end up simplifying the page cache slightly, although not as
much as I had hoped.

Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Reviewed-by: William Kucharski <william.kucharski@oracle.com>

1 files changed, 14 insertions, 4 deletions

diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index e5483347291c..f58524394dc1 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -2614,6 +2614,7 @@ int split_huge_page_to_list(struct page *page, struct list_head *list)
 {
 	struct page *head = compound_head(page);
 	struct deferred_split *ds_queue = get_deferred_split_queue(head);
+	XA_STATE(xas, &head->mapping->i_pages, head->index);
 	struct anon_vma *anon_vma = NULL;
 	struct address_space *mapping = NULL;
 	int extra_pins, ret;
@@ -2652,6 +2653,13 @@ int split_huge_page_to_list(struct page *page, struct list_head *list)
 			goto out;
 		}
 
+		xas_split_alloc(&xas, head, compound_order(head),
+				mapping_gfp_mask(mapping) & GFP_RECLAIM_MASK);
+		if (xas_error(&xas)) {
+			ret = xas_error(&xas);
+			goto out;
+		}
+
 		anon_vma = NULL;
 		i_mmap_lock_read(mapping);
 
@@ -2681,13 +2689,12 @@ int split_huge_page_to_list(struct page *page, struct list_head *list)
 	/* block interrupt reentry in xa_lock and spinlock */
 	local_irq_disable();
 	if (mapping) {
-		XA_STATE(xas, &mapping->i_pages, page_index(head));
-
 		/*
 		 * Check if the head page is present in page cache.
 		 * We assume all tail are present too, if head is there.
 		 */
-		xa_lock(&mapping->i_pages);
+		xas_lock(&xas);
+		xas_reset(&xas);
 		if (xas_load(&xas) != head)
 			goto fail;
 	}
@@ -2703,6 +2710,7 @@ int split_huge_page_to_list(struct page *page, struct list_head *list)
 		if (mapping) {
 			int nr = thp_nr_pages(head);
 
+			xas_split(&xas, head, thp_order(head));
 			if (PageSwapBacked(head)) {
 				__mod_lruvec_page_state(head, NR_SHMEM_THPS,
 							-nr);
@@ -2719,7 +2727,7 @@ int split_huge_page_to_list(struct page *page, struct list_head *list)
 		spin_unlock(&ds_queue->split_queue_lock);
 fail:
 		if (mapping)
-			xa_unlock(&mapping->i_pages);
+			xas_unlock(&xas);
 		local_irq_enable();
 		remap_page(head, thp_nr_pages(head));
 		ret = -EBUSY;
@@ -2733,6 +2741,8 @@ out_unlock:
 	if (mapping)
 		i_mmap_unlock_read(mapping);
 out:
+	/* Free any memory we didn't use */
+	xas_nomem(&xas, 0);
 	count_vm_event(!ret ? THP_SPLIT_PAGE : THP_SPLIT_PAGE_FAILED);
 	return ret;
 }