[merged mm-hotfixes-stable] mm-page_alloc-reduce-potential-fragmentation-in-make_alloc_exact.patch removed from -mm tree

[merged mm-hotfixes-stable] mm-page_alloc-reduce-potential-fragmentation-in-make_alloc_exact.patch removed from -mm tree

[Andrew Morton]

The quilt patch titled
     Subject: mm/page_alloc: reduce potential fragmentation in make_alloc_exact()
has been removed from the -mm tree.  Its filename was
     mm-page_alloc-reduce-potential-fragmentation-in-make_alloc_exact.patch

This patch was dropped because it was merged into the mm-hotfixes-stable branch
of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm

------------------------------------------------------
From: "Liam R. Howlett" <Liam.Howlett@Oracle.com>
Subject: mm/page_alloc: reduce potential fragmentation in make_alloc_exact()
Date: Tue, 31 May 2022 09:20:51 -0400

Try to avoid using the left over split page on the next request for a page
by calling __free_pages_ok() with FPI_TO_TAIL.  This increases the
potential of defragmenting memory when it's used for a short period of
time.

Link: https://lkml.kernel.org/r/20220531185626.yvlmymbxyoe5vags@revolver
Signed-off-by: Liam R. Howlett <Liam.Howlett@oracle.com>
Suggested-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---

 mm/page_alloc.c |   18 +++++++++++-------
 1 file changed, 11 insertions(+), 7 deletions(-)

--- a/mm/page_alloc.c~mm-page_alloc-reduce-potential-fragmentation-in-make_alloc_exact
+++ a/mm/page_alloc.c
@@ -5784,14 +5784,18 @@ static void *make_alloc_exact(unsigned l
 		size_t size)
 {
 	if (addr) {
-		unsigned long alloc_end = addr + (PAGE_SIZE << order);
-		unsigned long used = addr + PAGE_ALIGN(size);
+		unsigned long nr = DIV_ROUND_UP(size, PAGE_SIZE);
+		struct page *page = virt_to_page((void *)addr);
+		struct page *last = page + nr;
 
-		split_page(virt_to_page((void *)addr), order);
-		while (used < alloc_end) {
-			free_page(used);
-			used += PAGE_SIZE;
-		}
+		split_page_owner(page, 1 << order);
+		split_page_memcg(page, 1 << order);
+		while (page < --last)
+			set_page_refcounted(last);
+
+		last = page + (1UL << order);
+		for (page += nr; page < last; page++)
+			__free_pages_ok(page, 0, FPI_TO_TAIL);
 	}
 	return (void *)addr;
 }
_

Patches currently in -mm which might be from Liam.Howlett@Oracle.com are