[PATCH] mm: alloc_anon_folio: avoid doing vma_thp_gfp_mask in fallback cases

[Barry Song]

From: Barry Song <v-songbaohua@xxxxxxxx>

Fallback rates surpassing 90% have been observed on phones utilizing 64KiB
CONT-PTE mTHP. In these scenarios, when one out of every 16 PTEs fails
to allocate large folios, the remaining 15 PTEs fallback. Consequently,
invoking vma_thp_gfp_mask seems redundant in such cases. Furthermore,
abstaining from its use can also contribute to improved code readability.

Cc: Ryan Roberts <ryan.roberts@xxxxxxx>
Cc: Kefeng Wang <wangkefeng.wang@xxxxxxxxxx>
Cc: John Hubbard <jhubbard@xxxxxxxxxx>
Cc: David Hildenbrand <david@xxxxxxxxxx>
Cc: Alistair Popple <apopple@xxxxxxxxxx>
Cc: Anshuman Khandual <anshuman.khandual@xxxxxxx>
Cc: Catalin Marinas <catalin.marinas@xxxxxxx>
Cc: David Rientjes <rientjes@xxxxxxxxxx>
Cc: "Huang, Ying" <ying.huang@xxxxxxxxx>
Cc: Hugh Dickins <hughd@xxxxxxxxxx>
Cc: Itaru Kitayama <itaru.kitayama@xxxxxxxxx>
Cc: Kirill A. Shutemov <kirill.shutemov@xxxxxxxxxxxxxxx>
Cc: Luis Chamberlain <mcgrof@xxxxxxxxxx>
Cc: Matthew Wilcox (Oracle) <willy@xxxxxxxxxxxxx>
Cc: Vlastimil Babka <vbabka@xxxxxxx>
Cc: Yang Shi <shy828301@xxxxxxxxx>
Cc: Yin Fengwei <fengwei.yin@xxxxxxxxx>
Cc: Yu Zhao <yuzhao@xxxxxxxxxx>
Cc: Zi Yan <ziy@xxxxxxxxxx>
Signed-off-by: Barry Song <v-songbaohua@xxxxxxxx>
---
 mm/memory.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/mm/memory.c b/mm/memory.c
index c9c1031c2ecb..010e7bb20d2b 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -4353,6 +4353,9 @@ static struct folio *alloc_anon_folio(struct vm_fault *vmf)
 
 	pte_unmap(pte);
 
+	if (!orders)
+		goto fallback;
+
 	/* Try allocating the highest of the remaining orders. */
 	gfp = vma_thp_gfp_mask(vma);
 	while (orders) {
-- 
2.34.1