The patch titled
Subject: mm: convert folio_estimated_sharers() to folio_likely_mapped_shared()
has been added to the -mm mm-unstable branch. Its filename is
mm-convert-folio_estimated_sharers-to-folio_likely_mapped_shared.patch
This patch will shortly appear at
https://git.kernel.org/pub/scm/linux/kernel/git/akpm/25-new.git/tree/patches/mm-convert-folio_estimated_sharers-to-folio_likely_mapped_shared.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@xxxxxxxxxx>
Subject: mm: convert folio_estimated_sharers() to folio_likely_mapped_shared()
Date: Tue, 27 Feb 2024 21:15:48 +0100
Callers of folio_estimated_sharers() only care about "mapped shared vs.
mapped exclusively", not the exact estimate of sharers. Let's consolidate
and unify the condition users are checking. While at it clarify the
semantics and extend the discussion on the fuzziness.
Use the "likely mapped shared" terminology to better express what the
(adjusted) function actually checks.
Whether a partially-mappable folio is more likely to not be partially
mapped than partially mapped is debatable. In the future, we might be
able to improve our estimate for partially-mappable folios, though.
Note that we will now consistently detect "mapped shared" only if the
first subpage is actually mapped multiple times. When the first subpage
is not mapped, we will consistently detect it as "mapped exclusively".
This change should currently only affect the usage in
madvise_free_pte_range() and queue_folios_pte_range() for large folios: if
the first page was already unmapped, we would have skipped the folio.
Link: https://lkml.kernel.org/r/20240227201548.857831-1-david@xxxxxxxxxx
Signed-off-by: David Hildenbrand <david@xxxxxxxxxx>
Reviewed-by: Khalid Aziz <khalid.aziz@xxxxxxxxxx>
Acked-by: Barry Song <v-songbaohua@xxxxxxxx>
Reviewed-by: Vishal Moola (Oracle) <vishal.moola@xxxxxxxxx>
Reviewed-by: Ryan Roberts <ryan.roberts@xxxxxxx>
Reviewed-by: Zi Yan <ziy@xxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
include/linux/mm.h | 46 +++++++++++++++++++++++++++++++++----------
mm/huge_memory.c | 2 -
mm/madvise.c | 6 ++---
mm/memory.c | 2 -
mm/mempolicy.c | 14 +++++--------
mm/migrate.c | 8 +++----
6 files changed, 51 insertions(+), 27 deletions(-)
--- a/include/linux/mm.h~mm-convert-folio_estimated_sharers-to-folio_likely_mapped_shared
+++ a/include/linux/mm.h
@@ -2165,21 +2165,47 @@ static inline size_t folio_size(struct f
}
/**
- * folio_estimated_sharers - Estimate the number of sharers of a folio.
+ * folio_likely_mapped_shared - Estimate if the folio is mapped into the page
+ * tables of more than one MM
* @folio: The folio.
*
- * folio_estimated_sharers() aims to serve as a function to efficiently
- * estimate the number of processes sharing a folio. This is done by
- * looking at the precise mapcount of the first subpage in the folio, and
- * assuming the other subpages are the same. This may not be true for large
- * folios. If you want exact mapcounts for exact calculations, look at
- * page_mapcount() or folio_total_mapcount().
+ * This function checks if the folio is currently mapped into more than one
+ * MM ("mapped shared"), or if the folio is only mapped into a single MM
+ * ("mapped exclusively").
*
- * Return: The estimated number of processes sharing a folio.
+ * As precise information is not easily available for all folios, this function
+ * estimates the number of MMs ("sharers") that are currently mapping a folio
+ * using the number of times the first page of the folio is currently mapped
+ * into page tables.
+ *
+ * For small anonymous folios (except KSM folios) and anonymous hugetlb folios,
+ * the return value will be exactly correct, because they can only be mapped
+ * at most once into an MM, and they cannot be partially mapped.
+ *
+ * For other folios, the result can be fuzzy:
+ * (a) For partially-mappable large folios (THP), the return value can wrongly
+ * indicate "mapped exclusively" (false negative) when the folio is
+ * only partially mapped into at least one MM.
+ * (b) For pagecache folios (including hugetlb), the return value can wrongly
+ * indicate "mapped shared" (false positive) when two VMAs in the same MM
+ * cover the same file range.
+ * (c) For (small) KSM folios, the return value can wrongly indicate "mapped
+ * shared" (false negative), when the folio is mapped multiple times into
+ * the same MM.
+ *
+ * Further, this function only considers current page table mappings that
+ * are tracked using the folio mapcount(s). It does not consider:
+ * (1) If the folio might get mapped in the (near) future (e.g., swapcache,
+ * pagecache, temporary unmapping for migration).
+ * (2) If the folio is mapped differently (VM_PFNMAP).
+ * (3) If hugetlb page table sharing applies. Callers might want to check
+ * hugetlb_pmd_shared().
+ *
+ * Return: Whether the folio is estimated to be mapped into more than one MM.
*/
-static inline int folio_estimated_sharers(struct folio *folio)
+static inline bool folio_likely_mapped_shared(struct folio *folio)
{
- return page_mapcount(folio_page(folio, 0));
+ return page_mapcount(folio_page(folio, 0)) > 1;
}
#ifndef HAVE_ARCH_MAKE_PAGE_ACCESSIBLE
--- a/mm/huge_memory.c~mm-convert-folio_estimated_sharers-to-folio_likely_mapped_shared
+++ a/mm/huge_memory.c
@@ -1696,7 +1696,7 @@ bool madvise_free_huge_pmd(struct mmu_ga
* If other processes are mapping this folio, we couldn't discard
* the folio unless they all do MADV_FREE so let's skip the folio.
*/
- if (folio_estimated_sharers(folio) != 1)
+ if (folio_likely_mapped_shared(folio))
goto out;
if (!folio_trylock(folio))
--- a/mm/madvise.c~mm-convert-folio_estimated_sharers-to-folio_likely_mapped_shared
+++ a/mm/madvise.c
@@ -366,7 +366,7 @@ static int madvise_cold_or_pageout_pte_r
folio = pfn_folio(pmd_pfn(orig_pmd));
/* Do not interfere with other mappings of this folio */
- if (folio_estimated_sharers(folio) != 1)
+ if (folio_likely_mapped_shared(folio))
goto huge_unlock;
if (pageout_anon_only_filter && !folio_test_anon(folio))
@@ -453,7 +453,7 @@ restart:
if (folio_test_large(folio)) {
int err;
- if (folio_estimated_sharers(folio) > 1)
+ if (folio_likely_mapped_shared(folio))
break;
if (pageout_anon_only_filter && !folio_test_anon(folio))
break;
@@ -677,7 +677,7 @@ static int madvise_free_pte_range(pmd_t
if (folio_test_large(folio)) {
int err;
- if (folio_estimated_sharers(folio) != 1)
+ if (folio_likely_mapped_shared(folio))
break;
if (!folio_trylock(folio))
break;
--- a/mm/memory.c~mm-convert-folio_estimated_sharers-to-folio_likely_mapped_shared
+++ a/mm/memory.c
@@ -5112,7 +5112,7 @@ static vm_fault_t do_numa_page(struct vm
* Flag if the folio is shared between multiple address spaces. This
* is later used when determining whether to group tasks together
*/
- if (folio_estimated_sharers(folio) > 1 && (vma->vm_flags & VM_SHARED))
+ if (folio_likely_mapped_shared(folio) && (vma->vm_flags & VM_SHARED))
flags |= TNF_SHARED;
nid = folio_nid(folio);
--- a/mm/mempolicy.c~mm-convert-folio_estimated_sharers-to-folio_likely_mapped_shared
+++ a/mm/mempolicy.c
@@ -642,12 +642,11 @@ static int queue_folios_hugetlb(pte_t *p
* Unless MPOL_MF_MOVE_ALL, we try to avoid migrating a shared folio.
* Choosing not to migrate a shared folio is not counted as a failure.
*
- * To check if the folio is shared, ideally we want to make sure
- * every page is mapped to the same process. Doing that is very
- * expensive, so check the estimated sharers of the folio instead.
+ * See folio_likely_mapped_shared() on possible imprecision when we
+ * cannot easily detect if a folio is shared.
*/
if ((flags & MPOL_MF_MOVE_ALL) ||
- (folio_estimated_sharers(folio) == 1 && !hugetlb_pmd_shared(pte)))
+ (!folio_likely_mapped_shared(folio) && !hugetlb_pmd_shared(pte)))
if (!isolate_hugetlb(folio, qp->pagelist))
qp->nr_failed++;
unlock:
@@ -1032,11 +1031,10 @@ static bool migrate_folio_add(struct fol
* Unless MPOL_MF_MOVE_ALL, we try to avoid migrating a shared folio.
* Choosing not to migrate a shared folio is not counted as a failure.
*
- * To check if the folio is shared, ideally we want to make sure
- * every page is mapped to the same process. Doing that is very
- * expensive, so check the estimated sharers of the folio instead.
+ * See folio_likely_mapped_shared() on possible imprecision when we
+ * cannot easily detect if a folio is shared.
*/
- if ((flags & MPOL_MF_MOVE_ALL) || folio_estimated_sharers(folio) == 1) {
+ if ((flags & MPOL_MF_MOVE_ALL) || !folio_likely_mapped_shared(folio)) {
if (folio_isolate_lru(folio)) {
list_add_tail(&folio->lru, foliolist);
node_stat_mod_folio(folio,
--- a/mm/migrate.c~mm-convert-folio_estimated_sharers-to-folio_likely_mapped_shared
+++ a/mm/migrate.c
@@ -2593,11 +2593,11 @@ int migrate_misplaced_folio(struct folio
/*
* Don't migrate file folios that are mapped in multiple processes
* with execute permissions as they are probably shared libraries.
- * To check if the folio is shared, ideally we want to make sure
- * every page is mapped to the same process. Doing that is very
- * expensive, so check the estimated mapcount of the folio instead.
+ *
+ * See folio_likely_mapped_shared() on possible imprecision when we
+ * cannot easily detect if a folio is shared.
*/
- if (folio_estimated_sharers(folio) != 1 && folio_is_file_lru(folio) &&
+ if (folio_likely_mapped_shared(folio) && folio_is_file_lru(folio) &&
(vma->vm_flags & VM_EXEC))
goto out;
_
Patches currently in -mm which might be from david@xxxxxxxxxx are
mm-madvise-make-madv_populate_readwrite-handle-vm_fault_retry-properly.patch
mm-madvise-dont-perform-madvise-vma-walk-for-madv_populate_readwrite.patch
mm-userfaultfd-dont-place-zeropages-when-zeropages-are-disallowed.patch
s390-mm-re-enable-the-shared-zeropage-for-pv-and-skeys-kvm-guests.patch
mm-convert-folio_estimated_sharers-to-folio_likely_mapped_shared.patch
