+ mm-huge_memory-work-on-folio-swap-instead-of-page-private-when-splitting-folio.patch added to mm-unstable branch

+ mm-huge_memory-work-on-folio-swap-instead-of-page-private-when-splitting-folio.patch added to mm-unstable branch

[Andrew Morton]

The patch titled
     Subject: mm/huge_memory: work on folio->swap instead of page->private when splitting folio
has been added to the -mm mm-unstable branch.  Its filename is
     mm-huge_memory-work-on-folio-swap-instead-of-page-private-when-splitting-folio.patch

This patch will shortly appear at
     https://git.kernel.org/pub/scm/linux/kernel/git/akpm/25-new.git/tree/patches/mm-huge_memory-work-on-folio-swap-instead-of-page-private-when-splitting-folio.patch

This patch will later appear in the mm-unstable branch at
    git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm

Before you just go and hit "reply", please:
   a) Consider who else should be cc'ed
   b) Prefer to cc a suitable mailing list as well
   c) Ideally: find the original patch on the mailing list and do a
      reply-to-all to that, adding suitable additional cc's

*** Remember to use Documentation/process/submit-checklist.rst when testing your code ***

The -mm tree is included into linux-next via the mm-everything
branch at git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
and is updated there every 2-3 working days

------------------------------------------------------
From: David Hildenbrand <david@redhat.com>
Subject: mm/huge_memory: work on folio->swap instead of page->private when splitting folio
Date: Mon, 21 Aug 2023 18:08:49 +0200

Let's work on folio->swap instead.  While at it, use folio_test_anon() and
folio_test_swapcache() -- the original folio remains valid even after
splitting (but is then an order-0 folio).

We can probably convert a lot more to folios in that code, let's focus on
folio->swap handling only for now.

Link: https://lkml.kernel.org/r/20230821160849.531668-5-david@redhat.com
Signed-off-by: David Hildenbrand <david@redhat.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Dan Streetman <ddstreet@ieee.org>
Cc: Hugh Dickins <hughd@google.com>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: Peter Xu <peterx@redhat.com>
Cc: Seth Jennings <sjenning@redhat.com>
Cc: Vitaly Wool <vitaly.wool@konsulko.com>
Cc: Will Deacon <will@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---

 mm/huge_memory.c |   29 +++++++++++++++--------------
 1 file changed, 15 insertions(+), 14 deletions(-)

--- a/mm/huge_memory.c~mm-huge_memory-work-on-folio-swap-instead-of-page-private-when-splitting-folio
+++ a/mm/huge_memory.c
@@ -2401,10 +2401,16 @@ static void lru_add_page_tail(struct pag
 	}
 }
 
-static void __split_huge_page_tail(struct page *head, int tail,
+static void __split_huge_page_tail(struct folio *folio, int tail,
 		struct lruvec *lruvec, struct list_head *list)
 {
+	struct page *head = &folio->page;
 	struct page *page_tail = head + tail;
+	/*
+	 * Careful: new_folio is not a "real" folio before we cleared PageTail.
+	 * Don't pass it around before clear_compound_head().
+	 */
+	struct folio *new_folio = (struct folio *)page_tail;
 
 	VM_BUG_ON_PAGE(atomic_read(&page_tail->_mapcount) != -1, page_tail);
 
@@ -2453,8 +2459,8 @@ static void __split_huge_page_tail(struc
 		VM_WARN_ON_ONCE_PAGE(true, page_tail);
 		page_tail->private = 0;
 	}
-	if (PageSwapCache(head))
-		set_page_private(page_tail, (unsigned long)head->private + tail);
+	if (folio_test_swapcache(folio))
+		new_folio->swap.val = folio->swap.val + tail;
 
 	/* Page flags must be visible before we make the page non-compound. */
 	smp_wmb();
@@ -2500,11 +2506,9 @@ static void __split_huge_page(struct pag
 	/* complete memcg works before add pages to LRU */
 	split_page_memcg(head, nr);
 
-	if (PageAnon(head) && PageSwapCache(head)) {
-		swp_entry_t entry = { .val = page_private(head) };
-
-		offset = swp_offset(entry);
-		swap_cache = swap_address_space(entry);
+	if (folio_test_anon(folio) && folio_test_swapcache(folio)) {
+		offset = swp_offset(folio->swap);
+		swap_cache = swap_address_space(folio->swap);
 		xa_lock(&swap_cache->i_pages);
 	}
 
@@ -2514,7 +2518,7 @@ static void __split_huge_page(struct pag
 	ClearPageHasHWPoisoned(head);
 
 	for (i = nr - 1; i >= 1; i--) {
-		__split_huge_page_tail(head, i, lruvec, list);
+		__split_huge_page_tail(folio, i, lruvec, list);
 		/* Some pages can be beyond EOF: drop them from page cache */
 		if (head[i].index >= end) {
 			struct folio *tail = page_folio(head + i);
@@ -2559,11 +2563,8 @@ static void __split_huge_page(struct pag
 
 	remap_page(folio, nr);
 
-	if (PageSwapCache(head)) {
-		swp_entry_t entry = { .val = page_private(head) };
-
-		split_swap_cluster(entry);
-	}
+	if (folio_test_swapcache(folio))
+		split_swap_cluster(folio->swap);
 
 	for (i = 0; i < nr; i++) {
 		struct page *subpage = head + i;
_

Patches currently in -mm which might be from david@redhat.com are

mm-gup-reintroduce-foll_numa-as-foll_honor_numa_fault.patch
smaps-use-vm_normal_page_pmd-instead-of-follow_trans_huge_pmd.patch
mm-gup-handle-cont-pte-hugetlb-pages-correctly-in-gup_must_unshare-via-gup-fast.patch
kvm-explicitly-set-foll_honor_numa_fault-in-hva_to_pfn_slow.patch
mm-gup-dont-implicitly-set-foll_honor_numa_fault.patch
pgtable-improve-pte_protnone-comment.patch
selftest-mm-ksm_functional_tests-test-in-mmap_and_merge_range-if-anything-got-merged.patch
selftest-mm-ksm_functional_tests-add-prot_none-test.patch
selftest-mm-ksm_functional_tests-add-prot_none-test-fix.patch
mm-swap-stop-using-page-private-on-tail-pages-for-thp_swap.patch
mm-swap-inline-folio_set_swap_entry-and-folio_swap_entry.patch
mm-huge_memory-work-on-folio-swap-instead-of-page-private-when-splitting-folio.patch