[PATCH v2 00/20] mm: MM owner tracking for large folios (!hugetlb) + CONFIG_NO_PAGE_MAPCOUNT

[PATCH v2 00/20] mm: MM owner tracking for large folios (!hugetlb) + CONFIG_NO_PAGE_MAPCOUNT

[David Hildenbrand]

This patch series is about to haunt me in my dreams, and the time spent
on this is ridiculous ... anyhow, here goes a new version that now also
supports 32bit (I wish we would have dynamically allocated "struct folio"
already ...) and always enables the new tracking with
CONFIG_TRANSPARENT_HUGEPAGE.

Let's add an "easy" way to decide -- without false positives, without
page-mapcounts and without page table/rmap scanning -- whether a large
folio is "certainly mapped exclusively" into a single MM, or whether it
"maybe mapped shared" into multiple MMs.

Use that information to implement Copy-on-Write reuse, to convert
folio_likely_mapped_shared() to folio_maybe_mapped_share(), and to
introduce a kernel config option that let's us not use+maintain
per-page mapcounts in large folios anymore.

The bigger picture was presented at LSF/MM [1].

This series is effectively a follow-up on my early work [2], which
implemented a more precise, but also more complicated, way to identify
whether a large folio is "mapped shared" into multiple MMs or
"mapped exclusively" into a single MM.


1 Patch Organization
====================

Patch #1 -> #6: make more room in order-1 folios, so we have two
                "unsigned long" available for our purposes

Patch #7 -> #11: preparations

Patch #12: MM owner tracking for large folios

Patch #13: COW reuse for PTE-mapped anon THP

Patch #14: folio_maybe_mapped_shared()

Patch #15 -> #20: introduce and implement CONFIG_NO_PAGE_MAPCOUNT


2 MM owner tracking
===================

We assign each MM a unique ID ("MM ID"), to be able to squeeze more
information in our folios. On 32bit we use 15-bit IDs, on 64bit we use
31-bit IDs.

For each large folios, we now store two MM-ID+mapcount ("slot")
combinations:
* mm0_id + mm0_mapcount
* mm1_id + mm1_mapcount

On 32bit, we use a 16-bit per-MM mapcount, on 64bit an ordinary 32bit
mapcount. This way, we require 2x "unsigned long" on 32bit and 64bit for
both slots.

Paired with the large mapcount, we can reliably identify whether one
of these MMs is the current owner (-> owns all mappings) or even holds
all folio references (-> owns all mappings, and all references are from
mappings).

As long as only two MMs map folio pages at a time, we can reliably and
precisely identify whether a large folio is "mapped shared" or
"mapped exclusively".

Any additional MM that starts mapping the folio while there are no free
slots becomes an "untracked MM". If one such "untracked MM" is the last
one mapping a folio exclusively, we will *not* detect the folio as
"mapped exclusively" but instead as "maybe mapped shared". (exception:
only a single mapping remains)

So that's where the approach gets imprecise.

For now, we use a bit-spinlock to sync the large mapcount + slots, and
make sure we do keep the machinery fast, to not degrade (un)map performance
drastically: for example, we make sure to only use a single atomic (when
grabbing the bit-spinlock), like we would already perform when updating
the large mapcount.


3 CONFIG_NO_PAGE_MAPCOUNT
=========================

patch #15 -> #20 spell out and document what exactly is affected when
not maintaining the per-page mapcounts in large folios anymore.

Most importantly, as we cannot maintain folio->_nr_pages_mapped anymore when
(un)mapping pages, we'll account a complete folio as mapped if a
single page is mapped. In addition, we'll not detect partially mapped
anonymous folios as such in all cases yet.

Likely less relevant changes include that we might now under-estimate the
USS (Unique Set Size) of a process, but never over-estimate it.

The goal is to make CONFIG_NO_PAGE_MAPCOUNT the default at some point,
to then slowly make it the only option, as we learn about real-life
impacts and possible ways to mitigate them.


4 Performance
=============

Detailed performance numbers were included in v1 [4], and not that much
changed between v1 and v2.

I did plenty of measurements on different systems in the meantime, that
all revealed slightly different results.

The pte-mapped-folio micro-benchmarks are fairly sensitive to code layout
changes on some systems. Especially the fork() benchmark started being
more-shaky-than-before on recent kernels for some reason.

In summary, with my micro-benchmarks:

* Small folios are not impacted.

* CoW performance seems to be mostly unchanged across all folios sizes.

* CoW reuse performance of large folios now matches CoW reuse performance
  of small folios, because we now actually implement the CoW reuse
  optimization. On an Intel Xeon Silver 4210R I measured a ~65% reduction
  in runtime, on an arm64 system I measured ~54% reduction.

* munmap() performance improves with CONFIG_NO_PAGE_MAPCOUNT. I saw
  double-digit % reduction (up to ~30% on an Intel Xeon Silver 4210R
  and up to ~70% on an AmpereOne A192-32X) with larger folios. The
  larger the folios, the larger the performance improvement.

* munmao() performance very slightly (couple percent) degrades without
  CONFIG_NO_PAGE_MAPCOUNT for smaller folios. For larger folios, there
  seems to be no change at all.

* fork() performance improves with CONFIG_NO_PAGE_MAPCOUNT. I saw
  double-digit % reduction (up to ~20% on an Intel Xeon Silver 4210R
  and up to ~10% on an AmpereOne A192-32X) with larger folios. The larger
  the folios, the larger the performance improvement.

* While fork() performance without CONFIG_NO_PAGE_MAPCOUNT seems to be
  almost unchanged on some systems, I saw some degradation for
  smaller folios on the AmpereOne A192-32X. I did not investigate the
  details yet, but I suspect code layout changes or suboptimal code
  placement / inlining.

I'm not to worried about the fork() micro-benchmarks for smaller folios
given how shaky the results are lately and by how much we improved fork()
performance recently.

I also ran case-anon-cow-rand and case-anon-cow-seq part of vm-scalability,
to assess the scalability and the impact of the bit-spinlock.
My measurements on a two 2-socket 10-core Intel Xeon Silver 4210R CPU
revealed no significant changes.

Similarly, running these benchmarks with 2 MiB THPs enabled on the
AmpereOne A192-32X with 192 cores, I got < 1% difference with < 1% stdev,
which is nice.

So far, I did not get my hands on a similarly large system with multiple
sockets.

I found no other fitting scalability benchmarks that seem to really hammer
on concurrent mapping/unmapping of large folio pages like case-anon-cow-seq
does.


5 Concerns
==========

5.1 Bit spinlock
----------------

I'm not quite happy about the bit-spinlock, but so far it does not seem to
affect scalability in my measurements.

If it ever becomes a problem we could either investigate improving the
locking, or simply stopping the MM tracking once there are "too many
mappings" and simply assume that the folio is "mapped shared" until it
was freed.

This would be similar (but slightly different) to the "0,1,2,stopped"
counting idea Willy had at some point. Adding that logic to "stop tracking"
adds more code to the hot path, so I avoided that for now.


5.2 folio_maybe_mapped_shared()
-------------------------------

I documented the change from folio_likely_mapped_shared() to
folio_maybe_mapped_shared() quite extensively. If we run into surprises,
I have some ideas on how to resolve them. For now, I think we should
be fine.


5.3 Added code to map/unmap hot path
------------------------------------

So far, it looks like the added code on the rmap hot path does not
really seem to matter much in the bigger picture. I'd like to further
reduce it (and possibly improve fork() performance further), but I don't
easily see how right now. Well, and I am out of puff :)

Having that said, alternatives I considered (e.g., per-MM per-folio
mapcount) would add a lot more overhead to these hot paths.


6 Future Work
=============

6.1 Large mapcount
------------------

It would be very handy if the large mapcount would count how often folio
pages are actually mapped into page tables: a PMD on x86-64 would count
512 times. Calculating the average per-page mapcount will be easy, and
remapping (PMD->PTE) folios would get even faster.

That would also remove the need for the entire mapcount (except for
PMD-sized folios for memory statistics reasons ...), and allow for mapping
folios larger than PMDs (e.g., 4 MiB) easily.

The downside is that we maybe would also have to take the same number of
folio references to make our folio_mapcount() == folio_ref_count() work. I
think it should be possible (user space could trigger many PTE mappings
already), but we be a bit more careful about possible mapcount/refcount
overflows. (e.g., fail mapping a folio if the mapcount exceeds a certain
threshold)

Maybe some day we'll have a 64bit refcount+mapcount.

6.2 hugetlb
-----------

I'd love to make use of the same tracking also for hugetlb.

The real problem is PMD table sharing: getting a page mapped by MM X and
unmapped by MM Y will not work. With mshare, that problem should not exist
(all mapping/unmapping will be routed through the mshare MM).


7 Version Updates
=================

I did a bunch of cross-compiles and quite some testing on i386, x86-64 and
arm64. The build bots were very helpful as well.

To keep the CC list short, adding only relevant subsystem maintainers
(CCed on all patches, sorry :) ).

v1 -> v2:
* 32bit support. It would all be easier if we would already allocate
  "struct folio" dynamically, but fortunately when we manage to do that,
  it will just clean that part up again. For now, we have to relocate in
  "struct folio" the _pincount and _entire_mapcount on 32bit, and the
  hugetlb data  unconditionally.
* "mm/rmap: basic MM owner tracking for large folios (!hugetlb)"
 -> Now unconditionally enabled with CONFIG_TRANSPARENT_HUGEPAGE
 -> Some changes to slot handling to handle some edge cases in a better
    way.
 -> Reworked the way flags are stored, in light of 32bit support.
* "mm: convert folio_likely_mapped_shared() to folio_maybe_mapped_shared()"
 -> Use the new logic always such that we can rename the function
* A bunch of cleanups/simplifications

Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: "Matthew Wilcox (Oracle)" <willy@infradead.org>
Cc: Tejun Heo <tj@kernel.org>
Cc: Zefan Li <lizefan.x@bytedance.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: "Michal Koutný" <mkoutny@suse.com>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Muchun Song <muchun.song@linux.dev>
Cc: "Liam R. Howlett" <Liam.Howlett@oracle.com>
Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Jann Horn <jannh@google.com>


David Hildenbrand (20):
  mm: factor out large folio handling from folio_order() into
    folio_large_order()
  mm: factor out large folio handling from folio_nr_pages() into
    folio_large_nr_pages()
  mm: let _folio_nr_pages overlay memcg_data in first tail page
  mm: move hugetlb specific things in folio to page[3]
  mm: move _pincount in folio to page[2] on 32bit
  mm: move _entire_mapcount in folio to page[2] on 32bit
  mm/rmap: pass dst_vma to folio_dup_file_rmap_pte() and friends
  mm/rmap: pass vma to __folio_add_rmap()
  mm/rmap: abstract large mapcount operations for large folios
    (!hugetlb)
  bit_spinlock: __always_inline (un)lock functions
  mm/rmap: use folio_large_nr_pages() in add/remove functions
  mm/rmap: basic MM owner tracking for large folios (!hugetlb)
  mm: Copy-on-Write (COW) reuse support for PTE-mapped THP
  mm: convert folio_likely_mapped_shared() to
    folio_maybe_mapped_shared()
  mm: CONFIG_NO_PAGE_MAPCOUNT to prepare for not maintain per-page
    mapcounts in large folios
  fs/proc/page: remove per-page mapcount dependency for /proc/kpagecount
    (CONFIG_NO_PAGE_MAPCOUNT)
  fs/proc/task_mmu: remove per-page mapcount dependency for
    PM_MMAP_EXCLUSIVE (CONFIG_NO_PAGE_MAPCOUNT)
  fs/proc/task_mmu: remove per-page mapcount dependency for "mapmax"
    (CONFIG_NO_PAGE_MAPCOUNT)
  fs/proc/task_mmu: remove per-page mapcount dependency for
    smaps/smaps_rollup (CONFIG_NO_PAGE_MAPCOUNT)
  mm: stop maintaining the per-page mapcount of large folios
    (CONFIG_NO_PAGE_MAPCOUNT)

 .../admin-guide/cgroup-v1/memory.rst          |   4 +
 Documentation/admin-guide/cgroup-v2.rst       |  10 +-
 Documentation/admin-guide/mm/pagemap.rst      |  16 +-
 Documentation/filesystems/proc.rst            |  28 +-
 Documentation/mm/transhuge.rst                |  39 ++-
 fs/proc/internal.h                            |  39 +++
 fs/proc/page.c                                |  19 +-
 fs/proc/task_mmu.c                            |  39 ++-
 include/linux/bit_spinlock.h                  |   8 +-
 include/linux/mm.h                            |  93 +++---
 include/linux/mm_types.h                      | 110 +++++--
 include/linux/page-flags.h                    |   4 +
 include/linux/rmap.h                          | 270 ++++++++++++++++--
 kernel/fork.c                                 |  36 +++
 mm/Kconfig                                    |  22 ++
 mm/debug.c                                    |  10 +-
 mm/gup.c                                      |   8 +-
 mm/huge_memory.c                              |  20 +-
 mm/hugetlb.c                                  |   1 -
 mm/internal.h                                 |  20 +-
 mm/khugepaged.c                               |   8 +-
 mm/madvise.c                                  |   6 +-
 mm/memory.c                                   |  96 ++++++-
 mm/mempolicy.c                                |   8 +-
 mm/migrate.c                                  |   7 +-
 mm/mprotect.c                                 |   2 +-
 mm/page_alloc.c                               |  50 +++-
 mm/page_owner.c                               |   2 +-
 mm/rmap.c                                     | 118 ++++++--
 29 files changed, 903 insertions(+), 190 deletions(-)


base-commit: f7ed46277aaa8f848f18959ff68469f5186ba87c
-- 
2.48.1

[PATCH v2 01/20] mm: factor out large folio handling from folio_order() into folio_large_order()

[David Hildenbrand]

Let's factor it out into a simple helper function. This helper will
also come in handy when working with code where we know that our
folio is large.

Maybe in the future we'll have the order readily available for small and
large folios; in that case, folio_large_order() would simply translate to
folio_order().

Reviewed-by: Lance Yang <ioworker0@gmail.com>
Reviewed-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
---
 include/linux/mm.h | 13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)

diff --git a/include/linux/mm.h b/include/linux/mm.h
index d82feabbe44f8..d00214e22a174 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -1194,6 +1194,11 @@ struct inode;
 
 extern void prep_compound_page(struct page *page, unsigned int order);
 
+static inline unsigned int folio_large_order(const struct folio *folio)
+{
+	return folio->_flags_1 & 0xff;
+}
+
 /*
  * compound_order() can be called without holding a reference, which means
  * that niceties like page_folio() don't work.  These callers should be
@@ -1207,7 +1212,7 @@ static inline unsigned int compound_order(struct page *page)
 
 	if (!test_bit(PG_head, &folio->flags))
 		return 0;
-	return folio->_flags_1 & 0xff;
+	return folio_large_order(folio);
 }
 
 /**
@@ -1223,7 +1228,7 @@ static inline unsigned int folio_order(const struct folio *folio)
 {
 	if (!folio_test_large(folio))
 		return 0;
-	return folio->_flags_1 & 0xff;
+	return folio_large_order(folio);
 }
 
 #include <linux/huge_mm.h>
@@ -2139,7 +2144,7 @@ static inline long folio_nr_pages(const struct folio *folio)
 #ifdef CONFIG_64BIT
 	return folio->_folio_nr_pages;
 #else
-	return 1L << (folio->_flags_1 & 0xff);
+	return 1L << folio_large_order(folio);
 #endif
 }
 
@@ -2164,7 +2169,7 @@ static inline unsigned long compound_nr(struct page *page)
 #ifdef CONFIG_64BIT
 	return folio->_folio_nr_pages;
 #else
-	return 1L << (folio->_flags_1 & 0xff);
+	return 1L << folio_large_order(folio);
 #endif
 }
 
-- 
2.48.1

[PATCH v2 02/20] mm: factor out large folio handling from folio_nr_pages() into folio_large_nr_pages()

[David Hildenbrand]

Let's factor it out into a simple helper function. This helper will
also come in handy when working with code where we know that our
folio is large.

While at it, let's consistently return a "long" value from all these
similar functions. Note that we cannot use "unsigned int" (even though
_folio_nr_pages is of that type), because it would break some callers
that do stuff like "-folio_nr_pages()". Both "int" or "unsigned long"
would work as well.

Maybe in the future we'll have the nr_pages readily available for all
large folios, maybe even for small folios, or maybe for none.

Reviewed-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
---
 include/linux/mm.h | 28 ++++++++++++++++------------
 1 file changed, 16 insertions(+), 12 deletions(-)

diff --git a/include/linux/mm.h b/include/linux/mm.h
index d00214e22a174..7c5a8fd29cfcd 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -1199,6 +1199,18 @@ static inline unsigned int folio_large_order(const struct folio *folio)
 	return folio->_flags_1 & 0xff;
 }
 
+#ifdef CONFIG_64BIT
+static inline long folio_large_nr_pages(const struct folio *folio)
+{
+	return folio->_folio_nr_pages;
+}
+#else
+static inline long folio_large_nr_pages(const struct folio *folio)
+{
+	return 1L << folio_large_order(folio);
+}
+#endif
+
 /*
  * compound_order() can be called without holding a reference, which means
  * that niceties like page_folio() don't work.  These callers should be
@@ -2141,11 +2153,7 @@ static inline long folio_nr_pages(const struct folio *folio)
 {
 	if (!folio_test_large(folio))
 		return 1;
-#ifdef CONFIG_64BIT
-	return folio->_folio_nr_pages;
-#else
-	return 1L << folio_large_order(folio);
-#endif
+	return folio_large_nr_pages(folio);
 }
 
 /* Only hugetlbfs can allocate folios larger than MAX_ORDER */
@@ -2160,24 +2168,20 @@ static inline long folio_nr_pages(const struct folio *folio)
  * page.  compound_nr() can be called on a tail page, and is defined to
  * return 1 in that case.
  */
-static inline unsigned long compound_nr(struct page *page)
+static inline long compound_nr(struct page *page)
 {
 	struct folio *folio = (struct folio *)page;
 
 	if (!test_bit(PG_head, &folio->flags))
 		return 1;
-#ifdef CONFIG_64BIT
-	return folio->_folio_nr_pages;
-#else
-	return 1L << folio_large_order(folio);
-#endif
+	return folio_large_nr_pages(folio);
 }
 
 /**
  * thp_nr_pages - The number of regular pages in this huge page.
  * @page: The head page of a huge page.
  */
-static inline int thp_nr_pages(struct page *page)
+static inline long thp_nr_pages(struct page *page)
 {
 	return folio_nr_pages((struct folio *)page);
 }
-- 
2.48.1

[PATCH v2 03/20] mm: let _folio_nr_pages overlay memcg_data in first tail page

[David Hildenbrand]

Let's free up some more of the "unconditionally available on 64BIT"
space in order-1 folios by letting _folio_nr_pages overlay memcg_data in
the first tail page (second folio page). Consequently, we have the
optimization now whenever we have CONFIG_MEMCG, independent of 64BIT.

We have to make sure that page->memcg on tail pages does not return
"surprises". page_memcg_check() already properly refuses PageTail().
Let's do that earlier in print_page_owner_memcg() to avoid printing
wrong "Slab cache page" information. No other code should touch that
field on tail pages of compound pages.

Reset the "_nr_pages" to 0 when splitting folios, or when freeing them
back to the buddy (to avoid false page->memcg_data "bad page" reports).

Note that in __split_huge_page(), folio_nr_pages() would stop working
already as soon as we start messing with the subpages.

Most kernel configs should have at least CONFIG_MEMCG enabled, even if
disabled at runtime. 64byte "struct memmap" is what we usually have
on 64BIT.

While at it, rename "_folio_nr_pages" to "_nr_pages".

Hopefully memdescs / dynamically allocating "strut folio" in the future
will further clean this up, e.g., making _nr_pages available in all
configs and maybe even in small folios. Doing that should be fairly easy
on top of this change.

Reviewed-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
---
 include/linux/mm.h       |  4 ++--
 include/linux/mm_types.h | 30 ++++++++++++++++++++++--------
 mm/huge_memory.c         | 16 +++++++++++++---
 mm/internal.h            |  4 ++--
 mm/page_alloc.c          |  6 +++++-
 mm/page_owner.c          |  2 +-
 6 files changed, 45 insertions(+), 17 deletions(-)

diff --git a/include/linux/mm.h b/include/linux/mm.h
index 7c5a8fd29cfcd..f6b6373a864dd 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -1199,10 +1199,10 @@ static inline unsigned int folio_large_order(const struct folio *folio)
 	return folio->_flags_1 & 0xff;
 }
 
-#ifdef CONFIG_64BIT
+#ifdef NR_PAGES_IN_LARGE_FOLIO
 static inline long folio_large_nr_pages(const struct folio *folio)
 {
-	return folio->_folio_nr_pages;
+	return folio->_nr_pages;
 }
 #else
 static inline long folio_large_nr_pages(const struct folio *folio)
diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index 689b2a7461892..e81be20bbabc6 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -287,6 +287,11 @@ typedef struct {
 	unsigned long val;
 } swp_entry_t;
 
+#if defined(CONFIG_MEMCG) || defined(CONFIG_SLAB_OBJ_EXT)
+/* We have some extra room after the refcount in tail pages. */
+#define NR_PAGES_IN_LARGE_FOLIO
+#endif
+
 /**
  * struct folio - Represents a contiguous set of bytes.
  * @flags: Identical to the page flags.
@@ -312,7 +317,7 @@ typedef struct {
  * @_large_mapcount: Do not use directly, call folio_mapcount().
  * @_nr_pages_mapped: Do not use outside of rmap and debug code.
  * @_pincount: Do not use directly, call folio_maybe_dma_pinned().
- * @_folio_nr_pages: Do not use directly, call folio_nr_pages().
+ * @_nr_pages: Do not use directly, call folio_nr_pages().
  * @_hugetlb_subpool: Do not use directly, use accessor in hugetlb.h.
  * @_hugetlb_cgroup: Do not use directly, use accessor in hugetlb_cgroup.h.
  * @_hugetlb_cgroup_rsvd: Do not use directly, use accessor in hugetlb_cgroup.h.
@@ -377,13 +382,20 @@ struct folio {
 			unsigned long _flags_1;
 			unsigned long _head_1;
 	/* public: */
-			atomic_t _large_mapcount;
-			atomic_t _entire_mapcount;
-			atomic_t _nr_pages_mapped;
-			atomic_t _pincount;
-#ifdef CONFIG_64BIT
-			unsigned int _folio_nr_pages;
-#endif
+			union {
+				struct {
+					atomic_t _large_mapcount;
+					atomic_t _entire_mapcount;
+					atomic_t _nr_pages_mapped;
+					atomic_t _pincount;
+				};
+				unsigned long _usable_1[4];
+			};
+			atomic_t _mapcount_1;
+			atomic_t _refcount_1;
+#ifdef NR_PAGES_IN_LARGE_FOLIO
+			unsigned int _nr_pages;
+#endif /* NR_PAGES_IN_LARGE_FOLIO */
 	/* private: the union with struct page is transitional */
 		};
 		struct page __page_1;
@@ -435,6 +447,8 @@ FOLIO_MATCH(_last_cpupid, _last_cpupid);
 			offsetof(struct page, pg) + sizeof(struct page))
 FOLIO_MATCH(flags, _flags_1);
 FOLIO_MATCH(compound_head, _head_1);
+FOLIO_MATCH(_mapcount, _mapcount_1);
+FOLIO_MATCH(_refcount, _refcount_1);
 #undef FOLIO_MATCH
 #define FOLIO_MATCH(pg, fl)						\
 	static_assert(offsetof(struct folio, fl) ==			\
diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index 1e45064046a0f..13152313e77ce 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -3307,9 +3307,10 @@ bool can_split_folio(struct folio *folio, int caller_pins, int *pextra_pins)
  * It splits @folio into @new_order folios and copies the @folio metadata to
  * all the resulting folios.
  */
-static void __split_folio_to_order(struct folio *folio, int new_order)
+static void __split_folio_to_order(struct folio *folio, int old_order,
+		int new_order)
 {
-	long nr_pages = folio_nr_pages(folio);
+	long nr_pages = 1 << old_order;
 	long new_nr_pages = 1 << new_order;
 	long index;
 
@@ -3529,12 +3530,21 @@ static int __split_unmapped_folio(struct folio *folio, int new_order,
 			}
 		}
 
+		/*
+		 * Reset any memcg data overlay in the tail pages.
+		 * folio_nr_pages() is unreliable until prep_compound_page()
+		 * was called again.
+		 */
+#ifdef NR_PAGES_IN_LARGE_FOLIO
+		folio->_nr_pages = 0;
+#endif
+
 		/* complete memcg works before add pages to LRU */
 		split_page_memcg(&folio->page, old_order, split_order);
 		split_page_owner(&folio->page, old_order, split_order);
 		pgalloc_tag_split(folio, old_order, split_order);
 
-		__split_folio_to_order(folio, split_order);
+		__split_folio_to_order(folio, old_order, split_order);
 
 after_split:
 		/*
diff --git a/mm/internal.h b/mm/internal.h
index b07550db2bfd1..7f6d5def00fa0 100644
--- a/mm/internal.h
+++ b/mm/internal.h
@@ -684,8 +684,8 @@ static inline void folio_set_order(struct folio *folio, unsigned int order)
 		return;
 
 	folio->_flags_1 = (folio->_flags_1 & ~0xffUL) | order;
-#ifdef CONFIG_64BIT
-	folio->_folio_nr_pages = 1U << order;
+#ifdef NR_PAGES_IN_LARGE_FOLIO
+	folio->_nr_pages = 1U << order;
 #endif
 }
 
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 16dfcf7ade74a..bdfc954dab9aa 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1174,8 +1174,12 @@ __always_inline bool free_pages_prepare(struct page *page,
 	if (unlikely(order)) {
 		int i;
 
-		if (compound)
+		if (compound) {
 			page[1].flags &= ~PAGE_FLAGS_SECOND;
+#ifdef NR_PAGES_IN_LARGE_FOLIO
+			folio->_nr_pages = 0;
+#endif
+		}
 		for (i = 1; i < (1 << order); i++) {
 			if (compound)
 				bad += free_tail_page_prepare(page, page + i);
diff --git a/mm/page_owner.c b/mm/page_owner.c
index 2d6360eaccbb6..a409e2561a8fd 100644
--- a/mm/page_owner.c
+++ b/mm/page_owner.c
@@ -507,7 +507,7 @@ static inline int print_page_owner_memcg(char *kbuf, size_t count, int ret,
 
 	rcu_read_lock();
 	memcg_data = READ_ONCE(page->memcg_data);
-	if (!memcg_data)
+	if (!memcg_data || PageTail(page))
 		goto out_unlock;
 
 	if (memcg_data & MEMCG_DATA_OBJEXTS)
-- 
2.48.1

[PATCH v2 04/20] mm: move hugetlb specific things in folio to page[3]

[David Hildenbrand]

Let's just move the hugetlb specific stuff to a separate page, and stop
letting it overlay other fields for now.

This frees up some space in page[2], which we will use on 32bit to free
up some space in page[1]. While we could move these things to page[3]
instead, it's cleaner to just move the hugetlb specific things out of
the way and pack the core-folio stuff as tight as possible. ... and we
can minimize the work required in dump_folio.

We can now avoid re-initializing &folio->_deferred_list in hugetlb code.

Hopefully dynamically allocating "strut folio" in the future will further
clean this up.

Signed-off-by: David Hildenbrand <david@redhat.com>
---
 include/linux/mm_types.h | 27 +++++++++++++++++----------
 mm/hugetlb.c             |  1 -
 mm/page_alloc.c          |  5 +++++
 3 files changed, 22 insertions(+), 11 deletions(-)

diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index e81be20bbabc6..1d9c68c551d42 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -405,20 +405,23 @@ struct folio {
 			unsigned long _flags_2;
 			unsigned long _head_2;
 	/* public: */
-			void *_hugetlb_subpool;
-			void *_hugetlb_cgroup;
-			void *_hugetlb_cgroup_rsvd;
-			void *_hugetlb_hwpoison;
+			struct list_head _deferred_list;
 	/* private: the union with struct page is transitional */
 		};
+		struct page __page_2;
+	};
+	union {
 		struct {
-			unsigned long _flags_2a;
-			unsigned long _head_2a;
+			unsigned long _flags_3;
+			unsigned long _head_3;
 	/* public: */
-			struct list_head _deferred_list;
+			void *_hugetlb_subpool;
+			void *_hugetlb_cgroup;
+			void *_hugetlb_cgroup_rsvd;
+			void *_hugetlb_hwpoison;
 	/* private: the union with struct page is transitional */
 		};
-		struct page __page_2;
+		struct page __page_3;
 	};
 };
 
@@ -455,8 +458,12 @@ FOLIO_MATCH(_refcount, _refcount_1);
 			offsetof(struct page, pg) + 2 * sizeof(struct page))
 FOLIO_MATCH(flags, _flags_2);
 FOLIO_MATCH(compound_head, _head_2);
-FOLIO_MATCH(flags, _flags_2a);
-FOLIO_MATCH(compound_head, _head_2a);
+#undef FOLIO_MATCH
+#define FOLIO_MATCH(pg, fl)						\
+	static_assert(offsetof(struct folio, fl) ==			\
+			offsetof(struct page, pg) + 3 * sizeof(struct page))
+FOLIO_MATCH(flags, _flags_3);
+FOLIO_MATCH(compound_head, _head_3);
 #undef FOLIO_MATCH
 
 /**
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 9faa1034704ff..2ad5c292568ab 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -1646,7 +1646,6 @@ static void __update_and_free_hugetlb_folio(struct hstate *h,
 
 	folio_ref_unfreeze(folio, 1);
 
-	INIT_LIST_HEAD(&folio->_deferred_list);
 	hugetlb_free_folio(folio);
 }
 
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index bdfc954dab9aa..05a2a9492cdb0 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -971,6 +971,11 @@ static int free_tail_page_prepare(struct page *head_page, struct page *page)
 			goto out;
 		}
 		break;
+	case 3:
+		/* the third tail page: hugetlb specifics overlap ->mappings */
+		if (IS_ENABLED(CONFIG_HUGETLB_PAGE))
+			break;
+		fallthrough;
 	default:
 		if (page->mapping != TAIL_MAPPING) {
 			bad_page(page, "corrupted mapping in tail page");
-- 
2.48.1

[PATCH v2 05/20] mm: move _pincount in folio to page[2] on 32bit

[David Hildenbrand]

Let's free up some space on 32bit in page[1] by moving the _pincount to
page[2].

For order-1 folios (never anon folios!) on 32bit, we will now also use the
GUP_PIN_COUNTING_BIAS approach. A fully-mapped order-1 folio requires
2 references. With GUP_PIN_COUNTING_BIAS being 1024, we'd detect such
folios as "maybe pinned" with 512 full mappings, instead of 1024 for
order-0. As anon folios are out of the picture (which are the most relevant
users of checking for pinnings on *mapped* pages) and we are talking about
32bit, this is not expected to cause any trouble.

In __dump_page(), copy one additional folio page if we detect a folio
with an order > 1, so we can dump the pincount on order > 1 folios
reliably.

Note that THPs on 32bit are not particularly common (and we don't care
too much about performance), but we want to keep it working reliably,
because likely we want to use large folios there as well in the future,
independent of PMD leaf support.

Once we dynamically allocate "struct folio", fortunately the 32bit
specifics will likely go away again; even small folios could then have a
pincount and folio_has_pincount() would essentially always return
"true".

Signed-off-by: David Hildenbrand <david@redhat.com>
---
 include/linux/mm.h       | 11 +++++++++--
 include/linux/mm_types.h |  5 +++++
 mm/debug.c               | 10 +++++++++-
 mm/gup.c                 |  8 ++++----
 mm/internal.h            |  3 ++-
 mm/page_alloc.c          | 14 +++++++++++---
 6 files changed, 40 insertions(+), 11 deletions(-)

diff --git a/include/linux/mm.h b/include/linux/mm.h
index f6b6373a864dd..1a4ee028a851e 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -2004,6 +2004,13 @@ static inline struct folio *pfn_folio(unsigned long pfn)
 	return page_folio(pfn_to_page(pfn));
 }
 
+static inline bool folio_has_pincount(const struct folio *folio)
+{
+	if (IS_ENABLED(CONFIG_64BIT))
+		return folio_test_large(folio);
+	return folio_order(folio) > 1;
+}
+
 /**
  * folio_maybe_dma_pinned - Report if a folio may be pinned for DMA.
  * @folio: The folio.
@@ -2020,7 +2027,7 @@ static inline struct folio *pfn_folio(unsigned long pfn)
  * get that many refcounts, and b) all the callers of this routine are
  * expected to be able to deal gracefully with a false positive.
  *
- * For large folios, the result will be exactly correct. That's because
+ * For most large folios, the result will be exactly correct. That's because
  * we have more tracking data available: the _pincount field is used
  * instead of the GUP_PIN_COUNTING_BIAS scheme.
  *
@@ -2031,7 +2038,7 @@ static inline struct folio *pfn_folio(unsigned long pfn)
  */
 static inline bool folio_maybe_dma_pinned(struct folio *folio)
 {
-	if (folio_test_large(folio))
+	if (folio_has_pincount(folio))
 		return atomic_read(&folio->_pincount) > 0;
 
 	/*
diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index 1d9c68c551d42..31f466d8485bc 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -387,7 +387,9 @@ struct folio {
 					atomic_t _large_mapcount;
 					atomic_t _entire_mapcount;
 					atomic_t _nr_pages_mapped;
+#ifdef CONFIG_64BIT
 					atomic_t _pincount;
+#endif /* CONFIG_64BIT */
 				};
 				unsigned long _usable_1[4];
 			};
@@ -406,6 +408,9 @@ struct folio {
 			unsigned long _head_2;
 	/* public: */
 			struct list_head _deferred_list;
+#ifndef CONFIG_64BIT
+			atomic_t _pincount;
+#endif /* !CONFIG_64BIT */
 	/* private: the union with struct page is transitional */
 		};
 		struct page __page_2;
diff --git a/mm/debug.c b/mm/debug.c
index 2d1bd67d957bc..83ef3bd0ccd32 100644
--- a/mm/debug.c
+++ b/mm/debug.c
@@ -79,12 +79,17 @@ static void __dump_folio(struct folio *folio, struct page *page,
 			folio_ref_count(folio), mapcount, mapping,
 			folio->index + idx, pfn);
 	if (folio_test_large(folio)) {
+		int pincount = 0;
+
+		if (folio_has_pincount(folio))
+			pincount = atomic_read(&folio->_pincount);
+
 		pr_warn("head: order:%u mapcount:%d entire_mapcount:%d nr_pages_mapped:%d pincount:%d\n",
 				folio_order(folio),
 				folio_mapcount(folio),
 				folio_entire_mapcount(folio),
 				folio_nr_pages_mapped(folio),
-				atomic_read(&folio->_pincount));
+				pincount);
 	}
 
 #ifdef CONFIG_MEMCG
@@ -146,6 +151,9 @@ static void __dump_page(const struct page *page)
 	if (idx < MAX_FOLIO_NR_PAGES) {
 		memcpy(&folio, foliop, 2 * sizeof(struct page));
 		nr_pages = folio_nr_pages(&folio);
+		if (nr_pages > 1)
+			memcpy(&folio.__page_2, &foliop->__page_2,
+			       sizeof(struct page));
 		foliop = &folio;
 	}
 
diff --git a/mm/gup.c b/mm/gup.c
index e5040657870ea..2944fe8cf3174 100644
--- a/mm/gup.c
+++ b/mm/gup.c
@@ -109,7 +109,7 @@ static void gup_put_folio(struct folio *folio, int refs, unsigned int flags)
 		if (is_zero_folio(folio))
 			return;
 		node_stat_mod_folio(folio, NR_FOLL_PIN_RELEASED, refs);
-		if (folio_test_large(folio))
+		if (folio_has_pincount(folio))
 			atomic_sub(refs, &folio->_pincount);
 		else
 			refs *= GUP_PIN_COUNTING_BIAS;
@@ -164,7 +164,7 @@ int __must_check try_grab_folio(struct folio *folio, int refs,
 		 * Increment the normal page refcount field at least once,
 		 * so that the page really is pinned.
 		 */
-		if (folio_test_large(folio)) {
+		if (folio_has_pincount(folio)) {
 			folio_ref_add(folio, refs);
 			atomic_add(refs, &folio->_pincount);
 		} else {
@@ -223,7 +223,7 @@ void folio_add_pin(struct folio *folio)
 	 * page refcount field at least once, so that the page really is
 	 * pinned.
 	 */
-	if (folio_test_large(folio)) {
+	if (folio_has_pincount(folio)) {
 		WARN_ON_ONCE(atomic_read(&folio->_pincount) < 1);
 		folio_ref_inc(folio);
 		atomic_inc(&folio->_pincount);
@@ -575,7 +575,7 @@ static struct folio *try_grab_folio_fast(struct page *page, int refs,
 	 * is pinned.  That's why the refcount from the earlier
 	 * try_get_folio() is left intact.
 	 */
-	if (folio_test_large(folio))
+	if (folio_has_pincount(folio))
 		atomic_add(refs, &folio->_pincount);
 	else
 		folio_ref_add(folio,
diff --git a/mm/internal.h b/mm/internal.h
index 7f6d5def00fa0..d33db24c8b17b 100644
--- a/mm/internal.h
+++ b/mm/internal.h
@@ -723,7 +723,8 @@ static inline void prep_compound_head(struct page *page, unsigned int order)
 	atomic_set(&folio->_large_mapcount, -1);
 	atomic_set(&folio->_entire_mapcount, -1);
 	atomic_set(&folio->_nr_pages_mapped, 0);
-	atomic_set(&folio->_pincount, 0);
+	if (IS_ENABLED(CONFIG_64BIT) || order > 1)
+		atomic_set(&folio->_pincount, 0);
 	if (order > 1)
 		INIT_LIST_HEAD(&folio->_deferred_list);
 }
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 05a2a9492cdb0..3dff99cc54161 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -959,9 +959,11 @@ static int free_tail_page_prepare(struct page *head_page, struct page *page)
 			bad_page(page, "nonzero nr_pages_mapped");
 			goto out;
 		}
-		if (unlikely(atomic_read(&folio->_pincount))) {
-			bad_page(page, "nonzero pincount");
-			goto out;
+		if (IS_ENABLED(CONFIG_64BIT)) {
+			if (unlikely(atomic_read(&folio->_pincount))) {
+				bad_page(page, "nonzero pincount");
+				goto out;
+			}
 		}
 		break;
 	case 2:
@@ -970,6 +972,12 @@ static int free_tail_page_prepare(struct page *head_page, struct page *page)
 			bad_page(page, "on deferred list");
 			goto out;
 		}
+		if (!IS_ENABLED(CONFIG_64BIT)) {
+			if (unlikely(atomic_read(&folio->_pincount))) {
+				bad_page(page, "nonzero pincount");
+				goto out;
+			}
+		}
 		break;
 	case 3:
 		/* the third tail page: hugetlb specifics overlap ->mappings */
-- 
2.48.1

[PATCH v2 06/20] mm: move _entire_mapcount in folio to page[2] on 32bit

[David Hildenbrand]

Let's free up some space on 32bit in page[1] by moving the _pincount to
page[2].

Ordinary folios only use the entire mapcount with PMD mappings, so
order-1 folios don't apply. Similarly, hugetlb folios are always larger
than order-1, turning the entire mapcount essentially unused for all
order-1 folios. Moving it to order-1 folios will not change anything.

On 32bit, simply check in folio_entire_mapcount() whether we have an
order-1 folio, and return 0 in that case.

Note that THPs on 32bit are not particularly common (and we don't care
too much about performance), but we want to keep it working reliably,
because likely we want to use large folios there as well in the future,
independent of PMD leaf support.

Once we dynamically allocate "struct folio", the 32bit specifics will go
away again; even small folios could then have a pincount.

Signed-off-by: David Hildenbrand <david@redhat.com>
---
 include/linux/mm.h       |  2 ++
 include/linux/mm_types.h |  3 ++-
 mm/internal.h            |  5 +++--
 mm/page_alloc.c          | 12 ++++++++----
 4 files changed, 15 insertions(+), 7 deletions(-)

diff --git a/include/linux/mm.h b/include/linux/mm.h
index 1a4ee028a851e..9c1290588a11e 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -1333,6 +1333,8 @@ static inline int is_vmalloc_or_module_addr(const void *x)
 static inline int folio_entire_mapcount(const struct folio *folio)
 {
 	VM_BUG_ON_FOLIO(!folio_test_large(folio), folio);
+	if (!IS_ENABLED(CONFIG_64BIT) && unlikely(folio_large_order(folio) == 1))
+		return 0;
 	return atomic_read(&folio->_entire_mapcount) + 1;
 }
 
diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index 31f466d8485bc..c83dd2f1ee25e 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -385,9 +385,9 @@ struct folio {
 			union {
 				struct {
 					atomic_t _large_mapcount;
-					atomic_t _entire_mapcount;
 					atomic_t _nr_pages_mapped;
 #ifdef CONFIG_64BIT
+					atomic_t _entire_mapcount;
 					atomic_t _pincount;
 #endif /* CONFIG_64BIT */
 				};
@@ -409,6 +409,7 @@ struct folio {
 	/* public: */
 			struct list_head _deferred_list;
 #ifndef CONFIG_64BIT
+			atomic_t _entire_mapcount;
 			atomic_t _pincount;
 #endif /* !CONFIG_64BIT */
 	/* private: the union with struct page is transitional */
diff --git a/mm/internal.h b/mm/internal.h
index d33db24c8b17b..ffdc91b19322e 100644
--- a/mm/internal.h
+++ b/mm/internal.h
@@ -721,10 +721,11 @@ static inline void prep_compound_head(struct page *page, unsigned int order)
 
 	folio_set_order(folio, order);
 	atomic_set(&folio->_large_mapcount, -1);
-	atomic_set(&folio->_entire_mapcount, -1);
 	atomic_set(&folio->_nr_pages_mapped, 0);
-	if (IS_ENABLED(CONFIG_64BIT) || order > 1)
+	if (IS_ENABLED(CONFIG_64BIT) || order > 1) {
 		atomic_set(&folio->_pincount, 0);
+		atomic_set(&folio->_entire_mapcount, -1);
+	}
 	if (order > 1)
 		INIT_LIST_HEAD(&folio->_deferred_list);
 }
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 3dff99cc54161..7036530bd1bca 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -947,10 +947,6 @@ static int free_tail_page_prepare(struct page *head_page, struct page *page)
 	switch (page - head_page) {
 	case 1:
 		/* the first tail page: these may be in place of ->mapping */
-		if (unlikely(folio_entire_mapcount(folio))) {
-			bad_page(page, "nonzero entire_mapcount");
-			goto out;
-		}
 		if (unlikely(folio_large_mapcount(folio))) {
 			bad_page(page, "nonzero large_mapcount");
 			goto out;
@@ -960,6 +956,10 @@ static int free_tail_page_prepare(struct page *head_page, struct page *page)
 			goto out;
 		}
 		if (IS_ENABLED(CONFIG_64BIT)) {
+			if (unlikely(atomic_read(&folio->_entire_mapcount) + 1)) {
+				bad_page(page, "nonzero entire_mapcount");
+				goto out;
+			}
 			if (unlikely(atomic_read(&folio->_pincount))) {
 				bad_page(page, "nonzero pincount");
 				goto out;
@@ -973,6 +973,10 @@ static int free_tail_page_prepare(struct page *head_page, struct page *page)
 			goto out;
 		}
 		if (!IS_ENABLED(CONFIG_64BIT)) {
+			if (unlikely(atomic_read(&folio->_entire_mapcount) + 1)) {
+				bad_page(page, "nonzero entire_mapcount");
+				goto out;
+			}
 			if (unlikely(atomic_read(&folio->_pincount))) {
 				bad_page(page, "nonzero pincount");
 				goto out;
-- 
2.48.1

[PATCH v2 07/20] mm/rmap: pass dst_vma to folio_dup_file_rmap_pte() and friends

[David Hildenbrand]

We'll need access to the destination MM when modifying the large mapcount
of a non-hugetlb large folios next. So pass in the destination VMA.

Signed-off-by: David Hildenbrand <david@redhat.com>
---
 include/linux/rmap.h | 42 +++++++++++++++++++++++++-----------------
 mm/huge_memory.c     |  2 +-
 mm/memory.c          | 10 +++++-----
 3 files changed, 31 insertions(+), 23 deletions(-)

diff --git a/include/linux/rmap.h b/include/linux/rmap.h
index 6abf7960077aa..e795610bade80 100644
--- a/include/linux/rmap.h
+++ b/include/linux/rmap.h
@@ -335,7 +335,8 @@ static inline void hugetlb_remove_rmap(struct folio *folio)
 }
 
 static __always_inline void __folio_dup_file_rmap(struct folio *folio,
-		struct page *page, int nr_pages, enum rmap_level level)
+		struct page *page, int nr_pages, struct vm_area_struct *dst_vma,
+		enum rmap_level level)
 {
 	const int orig_nr_pages = nr_pages;
 
@@ -366,45 +367,47 @@ static __always_inline void __folio_dup_file_rmap(struct folio *folio,
  * @folio:	The folio to duplicate the mappings of
  * @page:	The first page to duplicate the mappings of
  * @nr_pages:	The number of pages of which the mapping will be duplicated
+ * @dst_vma:	The destination vm area
  *
  * The page range of the folio is defined by [page, page + nr_pages)
  *
  * The caller needs to hold the page table lock.
  */
 static inline void folio_dup_file_rmap_ptes(struct folio *folio,
-		struct page *page, int nr_pages)
+		struct page *page, int nr_pages, struct vm_area_struct *dst_vma)
 {
-	__folio_dup_file_rmap(folio, page, nr_pages, RMAP_LEVEL_PTE);
+	__folio_dup_file_rmap(folio, page, nr_pages, dst_vma, RMAP_LEVEL_PTE);
 }
 
 static __always_inline void folio_dup_file_rmap_pte(struct folio *folio,
-		struct page *page)
+		struct page *page, struct vm_area_struct *dst_vma)
 {
-	__folio_dup_file_rmap(folio, page, 1, RMAP_LEVEL_PTE);
+	__folio_dup_file_rmap(folio, page, 1, dst_vma, RMAP_LEVEL_PTE);
 }
 
 /**
  * folio_dup_file_rmap_pmd - duplicate a PMD mapping of a page range of a folio
  * @folio:	The folio to duplicate the mapping of
  * @page:	The first page to duplicate the mapping of
+ * @dst_vma:	The destination vm area
  *
  * The page range of the folio is defined by [page, page + HPAGE_PMD_NR)
  *
  * The caller needs to hold the page table lock.
  */
 static inline void folio_dup_file_rmap_pmd(struct folio *folio,
-		struct page *page)
+		struct page *page, struct vm_area_struct *dst_vma)
 {
 #ifdef CONFIG_TRANSPARENT_HUGEPAGE
-	__folio_dup_file_rmap(folio, page, HPAGE_PMD_NR, RMAP_LEVEL_PTE);
+	__folio_dup_file_rmap(folio, page, HPAGE_PMD_NR, dst_vma, RMAP_LEVEL_PTE);
 #else
 	WARN_ON_ONCE(true);
 #endif
 }
 
 static __always_inline int __folio_try_dup_anon_rmap(struct folio *folio,
-		struct page *page, int nr_pages, struct vm_area_struct *src_vma,
-		enum rmap_level level)
+		struct page *page, int nr_pages, struct vm_area_struct *dst_vma,
+		struct vm_area_struct *src_vma, enum rmap_level level)
 {
 	const int orig_nr_pages = nr_pages;
 	bool maybe_pinned;
@@ -470,6 +473,7 @@ static __always_inline int __folio_try_dup_anon_rmap(struct folio *folio,
  * @folio:	The folio to duplicate the mappings of
  * @page:	The first page to duplicate the mappings of
  * @nr_pages:	The number of pages of which the mapping will be duplicated
+ * @dst_vma:	The destination vm area
  * @src_vma:	The vm area from which the mappings are duplicated
  *
  * The page range of the folio is defined by [page, page + nr_pages)
@@ -488,16 +492,18 @@ static __always_inline int __folio_try_dup_anon_rmap(struct folio *folio,
  * Returns 0 if duplicating the mappings succeeded. Returns -EBUSY otherwise.
  */
 static inline int folio_try_dup_anon_rmap_ptes(struct folio *folio,
-		struct page *page, int nr_pages, struct vm_area_struct *src_vma)
+		struct page *page, int nr_pages, struct vm_area_struct *dst_vma,
+		struct vm_area_struct *src_vma)
 {
-	return __folio_try_dup_anon_rmap(folio, page, nr_pages, src_vma,
-					 RMAP_LEVEL_PTE);
+	return __folio_try_dup_anon_rmap(folio, page, nr_pages, dst_vma,
+					 src_vma, RMAP_LEVEL_PTE);
 }
 
 static __always_inline int folio_try_dup_anon_rmap_pte(struct folio *folio,
-		struct page *page, struct vm_area_struct *src_vma)
+		struct page *page, struct vm_area_struct *dst_vma,
+		struct vm_area_struct *src_vma)
 {
-	return __folio_try_dup_anon_rmap(folio, page, 1, src_vma,
+	return __folio_try_dup_anon_rmap(folio, page, 1, dst_vma, src_vma,
 					 RMAP_LEVEL_PTE);
 }
 
@@ -506,6 +512,7 @@ static __always_inline int folio_try_dup_anon_rmap_pte(struct folio *folio,
  *				 of a folio
  * @folio:	The folio to duplicate the mapping of
  * @page:	The first page to duplicate the mapping of
+ * @dst_vma:	The destination vm area
  * @src_vma:	The vm area from which the mapping is duplicated
  *
  * The page range of the folio is defined by [page, page + HPAGE_PMD_NR)
@@ -524,11 +531,12 @@ static __always_inline int folio_try_dup_anon_rmap_pte(struct folio *folio,
  * Returns 0 if duplicating the mapping succeeded. Returns -EBUSY otherwise.
  */
 static inline int folio_try_dup_anon_rmap_pmd(struct folio *folio,
-		struct page *page, struct vm_area_struct *src_vma)
+		struct page *page, struct vm_area_struct *dst_vma,
+		struct vm_area_struct *src_vma)
 {
 #ifdef CONFIG_TRANSPARENT_HUGEPAGE
-	return __folio_try_dup_anon_rmap(folio, page, HPAGE_PMD_NR, src_vma,
-					 RMAP_LEVEL_PMD);
+	return __folio_try_dup_anon_rmap(folio, page, HPAGE_PMD_NR, dst_vma,
+					 src_vma, RMAP_LEVEL_PMD);
 #else
 	WARN_ON_ONCE(true);
 	return -EBUSY;
diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index 13152313e77ce..a3264d88d4b49 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -1782,7 +1782,7 @@ int copy_huge_pmd(struct mm_struct *dst_mm, struct mm_struct *src_mm,
 	src_folio = page_folio(src_page);
 
 	folio_get(src_folio);
-	if (unlikely(folio_try_dup_anon_rmap_pmd(src_folio, src_page, src_vma))) {
+	if (unlikely(folio_try_dup_anon_rmap_pmd(src_folio, src_page, dst_vma, src_vma))) {
 		/* Page maybe pinned: split and retry the fault on PTEs. */
 		folio_put(src_folio);
 		pte_free(dst_mm, pgtable);
diff --git a/mm/memory.c b/mm/memory.c
index cacf6d53bdf32..ef74c4f009f84 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -845,7 +845,7 @@ copy_nonpresent_pte(struct mm_struct *dst_mm, struct mm_struct *src_mm,
 		folio_get(folio);
 		rss[mm_counter(folio)]++;
 		/* Cannot fail as these pages cannot get pinned. */
-		folio_try_dup_anon_rmap_pte(folio, page, src_vma);
+		folio_try_dup_anon_rmap_pte(folio, page, dst_vma, src_vma);
 
 		/*
 		 * We do not preserve soft-dirty information, because so
@@ -999,14 +999,14 @@ copy_present_ptes(struct vm_area_struct *dst_vma, struct vm_area_struct *src_vma
 		folio_ref_add(folio, nr);
 		if (folio_test_anon(folio)) {
 			if (unlikely(folio_try_dup_anon_rmap_ptes(folio, page,
-								  nr, src_vma))) {
+								  nr, dst_vma, src_vma))) {
 				folio_ref_sub(folio, nr);
 				return -EAGAIN;
 			}
 			rss[MM_ANONPAGES] += nr;
 			VM_WARN_ON_FOLIO(PageAnonExclusive(page), folio);
 		} else {
-			folio_dup_file_rmap_ptes(folio, page, nr);
+			folio_dup_file_rmap_ptes(folio, page, nr, dst_vma);
 			rss[mm_counter_file(folio)] += nr;
 		}
 		if (any_writable)
@@ -1024,7 +1024,7 @@ copy_present_ptes(struct vm_area_struct *dst_vma, struct vm_area_struct *src_vma
 		 * guarantee the pinned page won't be randomly replaced in the
 		 * future.
 		 */
-		if (unlikely(folio_try_dup_anon_rmap_pte(folio, page, src_vma))) {
+		if (unlikely(folio_try_dup_anon_rmap_pte(folio, page, dst_vma, src_vma))) {
 			/* Page may be pinned, we have to copy. */
 			folio_put(folio);
 			err = copy_present_page(dst_vma, src_vma, dst_pte, src_pte,
@@ -1034,7 +1034,7 @@ copy_present_ptes(struct vm_area_struct *dst_vma, struct vm_area_struct *src_vma
 		rss[MM_ANONPAGES]++;
 		VM_WARN_ON_FOLIO(PageAnonExclusive(page), folio);
 	} else {
-		folio_dup_file_rmap_pte(folio, page);
+		folio_dup_file_rmap_pte(folio, page, dst_vma);
 		rss[mm_counter_file(folio)]++;
 	}
 
-- 
2.48.1

[PATCH v2 08/20] mm/rmap: pass vma to __folio_add_rmap()

[David Hildenbrand]

We'll need access to the destination MM when modifying the mapcount
large folios next. So pass in the VMA.

Signed-off-by: David Hildenbrand <david@redhat.com>
---
 mm/rmap.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/mm/rmap.c b/mm/rmap.c
index bcec8677f68df..8a7d023b02e0c 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -1242,8 +1242,8 @@ int pfn_mkclean_range(unsigned long pfn, unsigned long nr_pages, pgoff_t pgoff,
 }
 
 static __always_inline unsigned int __folio_add_rmap(struct folio *folio,
-		struct page *page, int nr_pages, enum rmap_level level,
-		int *nr_pmdmapped)
+		struct page *page, int nr_pages, struct vm_area_struct *vma,
+		enum rmap_level level, int *nr_pmdmapped)
 {
 	atomic_t *mapped = &folio->_nr_pages_mapped;
 	const int orig_nr_pages = nr_pages;
@@ -1411,7 +1411,7 @@ static __always_inline void __folio_add_anon_rmap(struct folio *folio,
 
 	VM_WARN_ON_FOLIO(!folio_test_anon(folio), folio);
 
-	nr = __folio_add_rmap(folio, page, nr_pages, level, &nr_pmdmapped);
+	nr = __folio_add_rmap(folio, page, nr_pages, vma, level, &nr_pmdmapped);
 
 	if (likely(!folio_test_ksm(folio)))
 		__page_check_anon_rmap(folio, page, vma, address);
@@ -1582,7 +1582,7 @@ static __always_inline void __folio_add_file_rmap(struct folio *folio,
 
 	VM_WARN_ON_FOLIO(folio_test_anon(folio), folio);
 
-	nr = __folio_add_rmap(folio, page, nr_pages, level, &nr_pmdmapped);
+	nr = __folio_add_rmap(folio, page, nr_pages, vma, level, &nr_pmdmapped);
 	__folio_mod_stat(folio, nr, nr_pmdmapped);
 
 	/* See comments in folio_add_anon_rmap_*() */
-- 
2.48.1

[PATCH v2 09/20] mm/rmap: abstract large mapcount operations for large folios (!hugetlb)

[David Hildenbrand]

Let's abstract the operations so we can extend these operations easily.

Signed-off-by: David Hildenbrand <david@redhat.com>
---
 include/linux/rmap.h | 32 ++++++++++++++++++++++++++++----
 mm/rmap.c            | 14 ++++++--------
 2 files changed, 34 insertions(+), 12 deletions(-)

diff --git a/include/linux/rmap.h b/include/linux/rmap.h
index e795610bade80..d1e888cc97a58 100644
--- a/include/linux/rmap.h
+++ b/include/linux/rmap.h
@@ -173,6 +173,30 @@ static inline void anon_vma_merge(struct vm_area_struct *vma,
 
 struct anon_vma *folio_get_anon_vma(const struct folio *folio);
 
+static inline void folio_set_large_mapcount(struct folio *folio, int mapcount,
+		struct vm_area_struct *vma)
+{
+	/* Note: mapcounts start at -1. */
+	atomic_set(&folio->_large_mapcount, mapcount - 1);
+}
+
+static inline void folio_add_large_mapcount(struct folio *folio,
+		int diff, struct vm_area_struct *vma)
+{
+	atomic_add(diff, &folio->_large_mapcount);
+}
+
+static inline void folio_sub_large_mapcount(struct folio *folio,
+		int diff, struct vm_area_struct *vma)
+{
+	atomic_sub(diff, &folio->_large_mapcount);
+}
+
+#define folio_inc_large_mapcount(folio, vma) \
+	folio_add_large_mapcount(folio, 1, vma)
+#define folio_dec_large_mapcount(folio, vma) \
+	folio_sub_large_mapcount(folio, 1, vma)
+
 /* RMAP flags, currently only relevant for some anon rmap operations. */
 typedef int __bitwise rmap_t;
 
@@ -352,12 +376,12 @@ static __always_inline void __folio_dup_file_rmap(struct folio *folio,
 		do {
 			atomic_inc(&page->_mapcount);
 		} while (page++, --nr_pages > 0);
-		atomic_add(orig_nr_pages, &folio->_large_mapcount);
+		folio_add_large_mapcount(folio, orig_nr_pages, dst_vma);
 		break;
 	case RMAP_LEVEL_PMD:
 	case RMAP_LEVEL_PUD:
 		atomic_inc(&folio->_entire_mapcount);
-		atomic_inc(&folio->_large_mapcount);
+		folio_inc_large_mapcount(folio, dst_vma);
 		break;
 	}
 }
@@ -451,7 +475,7 @@ static __always_inline int __folio_try_dup_anon_rmap(struct folio *folio,
 				ClearPageAnonExclusive(page);
 			atomic_inc(&page->_mapcount);
 		} while (page++, --nr_pages > 0);
-		atomic_add(orig_nr_pages, &folio->_large_mapcount);
+		folio_add_large_mapcount(folio, orig_nr_pages, dst_vma);
 		break;
 	case RMAP_LEVEL_PMD:
 	case RMAP_LEVEL_PUD:
@@ -461,7 +485,7 @@ static __always_inline int __folio_try_dup_anon_rmap(struct folio *folio,
 			ClearPageAnonExclusive(page);
 		}
 		atomic_inc(&folio->_entire_mapcount);
-		atomic_inc(&folio->_large_mapcount);
+		folio_inc_large_mapcount(folio, dst_vma);
 		break;
 	}
 	return 0;
diff --git a/mm/rmap.c b/mm/rmap.c
index 8a7d023b02e0c..08846b7eced60 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -1266,7 +1266,7 @@ static __always_inline unsigned int __folio_add_rmap(struct folio *folio,
 		    atomic_add_return_relaxed(first, mapped) < ENTIRELY_MAPPED)
 			nr = first;
 
-		atomic_add(orig_nr_pages, &folio->_large_mapcount);
+		folio_add_large_mapcount(folio, orig_nr_pages, vma);
 		break;
 	case RMAP_LEVEL_PMD:
 	case RMAP_LEVEL_PUD:
@@ -1290,7 +1290,7 @@ static __always_inline unsigned int __folio_add_rmap(struct folio *folio,
 				nr = 0;
 			}
 		}
-		atomic_inc(&folio->_large_mapcount);
+		folio_inc_large_mapcount(folio, vma);
 		break;
 	}
 	return nr;
@@ -1556,14 +1556,12 @@ void folio_add_new_anon_rmap(struct folio *folio, struct vm_area_struct *vma,
 				SetPageAnonExclusive(page);
 		}
 
-		/* increment count (starts at -1) */
-		atomic_set(&folio->_large_mapcount, nr - 1);
+		folio_set_large_mapcount(folio, nr, vma);
 		atomic_set(&folio->_nr_pages_mapped, nr);
 	} else {
 		/* increment count (starts at -1) */
 		atomic_set(&folio->_entire_mapcount, 0);
-		/* increment count (starts at -1) */
-		atomic_set(&folio->_large_mapcount, 0);
+		folio_set_large_mapcount(folio, 1, vma);
 		atomic_set(&folio->_nr_pages_mapped, ENTIRELY_MAPPED);
 		if (exclusive)
 			SetPageAnonExclusive(&folio->page);
@@ -1665,7 +1663,7 @@ static __always_inline void __folio_remove_rmap(struct folio *folio,
 			break;
 		}
 
-		atomic_sub(nr_pages, &folio->_large_mapcount);
+		folio_sub_large_mapcount(folio, nr_pages, vma);
 		do {
 			last += atomic_add_negative(-1, &page->_mapcount);
 		} while (page++, --nr_pages > 0);
@@ -1678,7 +1676,7 @@ static __always_inline void __folio_remove_rmap(struct folio *folio,
 		break;
 	case RMAP_LEVEL_PMD:
 	case RMAP_LEVEL_PUD:
-		atomic_dec(&folio->_large_mapcount);
+		folio_dec_large_mapcount(folio, vma);
 		last = atomic_add_negative(-1, &folio->_entire_mapcount);
 		if (last) {
 			nr = atomic_sub_return_relaxed(ENTIRELY_MAPPED, mapped);
-- 
2.48.1

[PATCH v2 10/20] bit_spinlock: __always_inline (un)lock functions

[David Hildenbrand]

The compiler might decide that it is a smart idea to not inline
bit_spin_lock(), primarily when a couple of functions in the same file end
up calling it. Especially when used in RMAP map/unmap code next, the
compiler sometimes decides to not inline, which is then observable in
some micro-benchmarks.

Let's simply flag all lock/unlock functions as __always_inline;
arch_test_and_set_bit_lock() and friends are already tagged like that
(but not test_and_set_bit_lock() for some reason).

If ever a problem, we could split it into a fast and a slow path, and
only force the fast path to be inlined. But there is nothing
particularly "big" here.

Signed-off-by: David Hildenbrand <david@redhat.com>
---
 include/linux/bit_spinlock.h | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/include/linux/bit_spinlock.h b/include/linux/bit_spinlock.h
index bbc4730a6505c..c0989b5b0407f 100644
--- a/include/linux/bit_spinlock.h
+++ b/include/linux/bit_spinlock.h
@@ -13,7 +13,7 @@
  * Don't use this unless you really need to: spin_lock() and spin_unlock()
  * are significantly faster.
  */
-static inline void bit_spin_lock(int bitnum, unsigned long *addr)
+static __always_inline void bit_spin_lock(int bitnum, unsigned long *addr)
 {
 	/*
 	 * Assuming the lock is uncontended, this never enters
@@ -38,7 +38,7 @@ static inline void bit_spin_lock(int bitnum, unsigned long *addr)
 /*
  * Return true if it was acquired
  */
-static inline int bit_spin_trylock(int bitnum, unsigned long *addr)
+static __always_inline int bit_spin_trylock(int bitnum, unsigned long *addr)
 {
 	preempt_disable();
 #if defined(CONFIG_SMP) || defined(CONFIG_DEBUG_SPINLOCK)
@@ -54,7 +54,7 @@ static inline int bit_spin_trylock(int bitnum, unsigned long *addr)
 /*
  *  bit-based spin_unlock()
  */
-static inline void bit_spin_unlock(int bitnum, unsigned long *addr)
+static __always_inline void bit_spin_unlock(int bitnum, unsigned long *addr)
 {
 #ifdef CONFIG_DEBUG_SPINLOCK
 	BUG_ON(!test_bit(bitnum, addr));
@@ -71,7 +71,7 @@ static inline void bit_spin_unlock(int bitnum, unsigned long *addr)
  *  non-atomic version, which can be used eg. if the bit lock itself is
  *  protecting the rest of the flags in the word.
  */
-static inline void __bit_spin_unlock(int bitnum, unsigned long *addr)
+static __always_inline void __bit_spin_unlock(int bitnum, unsigned long *addr)
 {
 #ifdef CONFIG_DEBUG_SPINLOCK
 	BUG_ON(!test_bit(bitnum, addr));
-- 
2.48.1

[PATCH v2 11/20] mm/rmap: use folio_large_nr_pages() in add/remove functions

[David Hildenbrand]

Let's just use the "large" variant in code where we are sure that we
have a large folio in our hands: this way we are sure that we don't
perform any unnecessary "large" checks.

While at it, convert the VM_BUG_ON_VMA to a VM_WARN_ON_ONCE.

Maybe in the future there will not be a difference in that regard
between large and small folios; in that case, unifying the handling again
will be easy. E.g., folio_large_nr_pages() will simply translate to
folio_nr_pages() until we replace all instances.

Reviewed-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
---
 mm/rmap.c | 14 ++++++++------
 1 file changed, 8 insertions(+), 6 deletions(-)

diff --git a/mm/rmap.c b/mm/rmap.c
index 08846b7eced60..c9922928616ee 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -1274,7 +1274,7 @@ static __always_inline unsigned int __folio_add_rmap(struct folio *folio,
 		if (first) {
 			nr = atomic_add_return_relaxed(ENTIRELY_MAPPED, mapped);
 			if (likely(nr < ENTIRELY_MAPPED + ENTIRELY_MAPPED)) {
-				nr_pages = folio_nr_pages(folio);
+				nr_pages = folio_large_nr_pages(folio);
 				/*
 				 * We only track PMD mappings of PMD-sized
 				 * folios separately.
@@ -1522,14 +1522,11 @@ void folio_add_anon_rmap_pmd(struct folio *folio, struct page *page,
 void folio_add_new_anon_rmap(struct folio *folio, struct vm_area_struct *vma,
 		unsigned long address, rmap_t flags)
 {
-	const int nr = folio_nr_pages(folio);
 	const bool exclusive = flags & RMAP_EXCLUSIVE;
-	int nr_pmdmapped = 0;
+	int nr = 1, nr_pmdmapped = 0;
 
 	VM_WARN_ON_FOLIO(folio_test_hugetlb(folio), folio);
 	VM_WARN_ON_FOLIO(!exclusive && !folio_test_locked(folio), folio);
-	VM_BUG_ON_VMA(address < vma->vm_start ||
-			address + (nr << PAGE_SHIFT) > vma->vm_end, vma);
 
 	/*
 	 * VM_DROPPABLE mappings don't swap; instead they're just dropped when
@@ -1547,6 +1544,7 @@ void folio_add_new_anon_rmap(struct folio *folio, struct vm_area_struct *vma,
 	} else if (!folio_test_pmd_mappable(folio)) {
 		int i;
 
+		nr = folio_large_nr_pages(folio);
 		for (i = 0; i < nr; i++) {
 			struct page *page = folio_page(folio, i);
 
@@ -1559,6 +1557,7 @@ void folio_add_new_anon_rmap(struct folio *folio, struct vm_area_struct *vma,
 		folio_set_large_mapcount(folio, nr, vma);
 		atomic_set(&folio->_nr_pages_mapped, nr);
 	} else {
+		nr = folio_large_nr_pages(folio);
 		/* increment count (starts at -1) */
 		atomic_set(&folio->_entire_mapcount, 0);
 		folio_set_large_mapcount(folio, 1, vma);
@@ -1568,6 +1567,9 @@ void folio_add_new_anon_rmap(struct folio *folio, struct vm_area_struct *vma,
 		nr_pmdmapped = nr;
 	}
 
+	VM_WARN_ON_ONCE(address < vma->vm_start ||
+			address + (nr << PAGE_SHIFT) > vma->vm_end);
+
 	__folio_mod_stat(folio, nr, nr_pmdmapped);
 	mod_mthp_stat(folio_order(folio), MTHP_STAT_NR_ANON, 1);
 }
@@ -1681,7 +1683,7 @@ static __always_inline void __folio_remove_rmap(struct folio *folio,
 		if (last) {
 			nr = atomic_sub_return_relaxed(ENTIRELY_MAPPED, mapped);
 			if (likely(nr < ENTIRELY_MAPPED)) {
-				nr_pages = folio_nr_pages(folio);
+				nr_pages = folio_large_nr_pages(folio);
 				if (level == RMAP_LEVEL_PMD)
 					nr_pmdmapped = nr_pages;
 				nr = nr_pages - (nr & FOLIO_PAGES_MAPPED);
-- 
2.48.1

[PATCH v2 12/20] mm/rmap: basic MM owner tracking for large folios (!hugetlb)

[David Hildenbrand]

For small folios, we traditionally use the mapcount to decide
whether it was "certainly mapped exclusively" by a single MM
(mapcount == 1) or whether it "maybe mapped shared" by multiple MMs
(mapcount > 1). For PMD-sized folios that were PMD-mapped, we were able
to use a similar mechanism (single PMD mapping), but for PTE-mapped folios
and in the future folios that span multiple PMDs, this does not work.

So we need a different mechanism to handle large folios. Let's add a new
mechanism to detect whether a large folio is "certainly mapped
exclusively", or whether it is "maybe mapped shared".

We'll use this information next to optimize CoW reuse for PTE-mapped
anonymous THP, and to convert folio_likely_mapped_shared() to
folio_maybe_mapped_shared(), independent of per-page mapcounts.

For each large folio, we'll have two slots, whereby a slot stores:
 (1) an MM id: unique id assigned to each MM
 (2) a per-MM mapcount

If a slot is unoccupied, it can be taken by the next MM that maps folio
page.

In addition, we'll remember the current state -- "mapped exclusively" vs.
"maybe mapped shared" -- and use a bit spinlock to sync on updates and
to reduce the total number of atomic accesses on updates. In the
future, it might be possible to squeeze a proper spinlock into "struct
folio". For now, keep it simple, as we require the whole thing with THP
only, that is incompatible with RT.

As we have to squeeze this information into the "struct folio" of even
folios of order-1 (2 pages), and we generally want to reduce the required
metadata, we'll assign each MM a unique ID that can fit into an int. In
total, we can squeeze everything into 4x int (2x long) on 64bit.

32bit support is a bit challenging, because we only have 2x long == 2x
int in order-1 folios. But we can make it work for now, because we neither
expect many MMs nor very large folios on 32bit.

We will reliably detect folios as "mapped exclusively" vs. "mapped shared"
as long as only two MMs map pages of a folio at one point in time -- for
example with fork() and short-lived child processes, or with apps that
hand over state from one instance to another.

As soon as three MMs are involved at the same time, we might detect
"maybe mapped shared" although the folio is "mapped exclusively".

Example 1:

(1) App1 faults in a (shmem/file-backed) folio page -> Tracked as MM0
(2) App2 faults in a folio page -> Tracked as MM1
(4) App1 unmaps all folio pages

 -> We will detect "mapped exclusively".

Example 2:

(1) App1 faults in a (shmem/file-backed) folio page -> Tracked as MM0
(2) App2 faults in a folio page -> Tracked as MM1
(3) App3 faults in a folio page -> No slot available, tracked as "unknown"
(4) App1 and App2 unmap all folio pages

 -> We will detect "maybe mapped shared".

Make use of __always_inline to keep possible performance degradation
when (un)mapping large folios to a minimum.

Note: by squeezing the two flags into the "unsigned long" that stores
the MM ids, we can use non-atomic __bit_spin_unlock() and
non-atomic setting/clearing of the "maybe mapped shared" bit,
effectively not adding any new atomics on the hot path when updating the
large mapcount + new metadata, which further helps reduce the runtime
overhead in micro-benchmarks.

Signed-off-by: David Hildenbrand <david@redhat.com>
---
 Documentation/mm/transhuge.rst |   8 ++
 include/linux/mm_types.h       |  49 ++++++++++
 include/linux/page-flags.h     |   4 +
 include/linux/rmap.h           | 165 +++++++++++++++++++++++++++++++++
 kernel/fork.c                  |  36 +++++++
 mm/Kconfig                     |   4 +
 mm/internal.h                  |   5 +
 mm/page_alloc.c                |  10 ++
 8 files changed, 281 insertions(+)

diff --git a/Documentation/mm/transhuge.rst b/Documentation/mm/transhuge.rst
index a2cd8800d5279..baa17d718a762 100644
--- a/Documentation/mm/transhuge.rst
+++ b/Documentation/mm/transhuge.rst
@@ -120,11 +120,19 @@ pages:
     and also increment/decrement folio->_nr_pages_mapped by ENTIRELY_MAPPED
     when _entire_mapcount goes from -1 to 0 or 0 to -1.
 
+    We also maintain the two slots for tracking MM owners (MM ID and
+    corresponding mapcount), and the current status ("maybe mapped shared" vs.
+    "mapped exclusively").
+
   - map/unmap of individual pages with PTE entry increment/decrement
     page->_mapcount, increment/decrement folio->_large_mapcount and also
     increment/decrement folio->_nr_pages_mapped when page->_mapcount goes
     from -1 to 0 or 0 to -1 as this counts the number of pages mapped by PTE.
 
+    We also maintain the two slots for tracking MM owners (MM ID and
+    corresponding mapcount), and the current status ("maybe mapped shared" vs.
+    "mapped exclusively").
+
 split_huge_page internally has to distribute the refcounts in the head
 page to the tail pages before clearing all PG_head/tail bits from the page
 structures. It can be done easily for refcounts taken by page table
diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index c83dd2f1ee25e..2d657ac8e9b0c 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -292,6 +292,44 @@ typedef struct {
 #define NR_PAGES_IN_LARGE_FOLIO
 #endif
 
+/*
+ * On 32bit, we can cut the required metadata in half, because:
+ * (a) PID_MAX_LIMIT implicitly limits the number of MMs we could ever have,
+ *     so we can limit MM IDs to 15 bit (32767).
+ * (b) We don't expect folios where even a single complete PTE mapping by
+ *     one MM would exceed 15 bits (order-15).
+ */
+#ifdef CONFIG_64BIT
+typedef int mm_id_mapcount_t;
+#define MM_ID_MAPCOUNT_MAX		INT_MAX
+typedef unsigned int mm_id_t;
+#else /* !CONFIG_64BIT */
+typedef short mm_id_mapcount_t;
+#define MM_ID_MAPCOUNT_MAX		SHRT_MAX
+typedef unsigned short mm_id_t;
+#endif /* CONFIG_64BIT */
+
+/* We implicitly use the dummy ID for init-mm etc. where we never rmap pages. */
+#define MM_ID_DUMMY			0
+#define MM_ID_MIN			(MM_ID_DUMMY + 1)
+
+/*
+ * We leave the highest bit of each MM id unused, so we can store a flag
+ * in the highest bit of each folio->_mm_id[].
+ */
+#define MM_ID_BITS			((sizeof(mm_id_t) * BITS_PER_BYTE) - 1)
+#define MM_ID_MASK			((1U << MM_ID_BITS) - 1)
+#define MM_ID_MAX			MM_ID_MASK
+
+/*
+ * In order to use bit_spin_lock(), which requires an unsigned long, we
+ * operate on folio->_mm_ids when working on flags.
+ */
+#define FOLIO_MM_IDS_LOCK_BITNUM	MM_ID_BITS
+#define FOLIO_MM_IDS_LOCK_BIT		BIT(FOLIO_MM_IDS_LOCK_BITNUM)
+#define FOLIO_MM_IDS_SHARED_BITNUM	(2 * MM_ID_BITS + 1)
+#define FOLIO_MM_IDS_SHARED_BIT		BIT(FOLIO_MM_IDS_SHARED_BITNUM)
+
 /**
  * struct folio - Represents a contiguous set of bytes.
  * @flags: Identical to the page flags.
@@ -318,6 +356,9 @@ typedef struct {
  * @_nr_pages_mapped: Do not use outside of rmap and debug code.
  * @_pincount: Do not use directly, call folio_maybe_dma_pinned().
  * @_nr_pages: Do not use directly, call folio_nr_pages().
+ * @_mm_id: Do not use outside of rmap code.
+ * @_mm_ids: Do not use outside of rmap code.
+ * @_mm_id_mapcount: Do not use outside of rmap code.
  * @_hugetlb_subpool: Do not use directly, use accessor in hugetlb.h.
  * @_hugetlb_cgroup: Do not use directly, use accessor in hugetlb_cgroup.h.
  * @_hugetlb_cgroup_rsvd: Do not use directly, use accessor in hugetlb_cgroup.h.
@@ -390,6 +431,11 @@ struct folio {
 					atomic_t _entire_mapcount;
 					atomic_t _pincount;
 #endif /* CONFIG_64BIT */
+					mm_id_mapcount_t _mm_id_mapcount[2];
+					union {
+						mm_id_t _mm_id[2];
+						unsigned long _mm_ids;
+					};
 				};
 				unsigned long _usable_1[4];
 			};
@@ -1111,6 +1157,9 @@ struct mm_struct {
 #endif
 		} lru_gen;
 #endif /* CONFIG_LRU_GEN_WALKS_MMU */
+#ifdef CONFIG_MM_ID
+		mm_id_t mm_id;
+#endif /* CONFIG_MM_ID */
 	} __randomize_layout;
 
 	/*
diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h
index 30fe3eb62b90c..01716710066df 100644
--- a/include/linux/page-flags.h
+++ b/include/linux/page-flags.h
@@ -1222,6 +1222,10 @@ static inline int folio_has_private(const struct folio *folio)
 	return !!(folio->flags & PAGE_FLAGS_PRIVATE);
 }
 
+static inline bool folio_test_large_maybe_mapped_shared(const struct folio *folio)
+{
+	return test_bit(FOLIO_MM_IDS_SHARED_BITNUM, &folio->_mm_ids);
+}
 #undef PF_ANY
 #undef PF_HEAD
 #undef PF_NO_TAIL
diff --git a/include/linux/rmap.h b/include/linux/rmap.h
index d1e888cc97a58..c131b0efff0fa 100644
--- a/include/linux/rmap.h
+++ b/include/linux/rmap.h
@@ -13,6 +13,7 @@
 #include <linux/highmem.h>
 #include <linux/pagemap.h>
 #include <linux/memremap.h>
+#include <linux/bit_spinlock.h>
 
 /*
  * The anon_vma heads a list of private "related" vmas, to scan if
@@ -173,6 +174,169 @@ static inline void anon_vma_merge(struct vm_area_struct *vma,
 
 struct anon_vma *folio_get_anon_vma(const struct folio *folio);
 
+#ifdef CONFIG_MM_ID
+static __always_inline void folio_lock_large_mapcount(struct folio *folio)
+{
+	bit_spin_lock(FOLIO_MM_IDS_LOCK_BITNUM, &folio->_mm_ids);
+}
+
+static __always_inline void folio_unlock_large_mapcount(struct folio *folio)
+{
+	__bit_spin_unlock(FOLIO_MM_IDS_LOCK_BITNUM, &folio->_mm_ids);
+}
+
+static inline unsigned int folio_mm_id(const struct folio *folio, int idx)
+{
+	VM_WARN_ON_ONCE(idx != 0 && idx != 1);
+	return folio->_mm_id[idx] & MM_ID_MASK;
+}
+
+static inline void folio_set_mm_id(struct folio *folio, int idx, mm_id_t id)
+{
+	VM_WARN_ON_ONCE(idx != 0 && idx != 1);
+	folio->_mm_id[idx] &= ~MM_ID_MASK;
+	folio->_mm_id[idx] |= id;
+}
+
+static inline void __folio_large_mapcount_sanity_checks(const struct folio *folio,
+		int diff, mm_id_t mm_id)
+{
+	VM_WARN_ON_ONCE(!folio_test_large(folio) || folio_test_hugetlb(folio));
+	VM_WARN_ON_ONCE(diff <= 0);
+	VM_WARN_ON_ONCE(mm_id < MM_ID_MIN || mm_id > MM_ID_MAX);
+
+	/*
+	 * Make sure we can detect at least one complete PTE mapping of the
+	 * folio in a single MM as "exclusively mapped". This is primarily
+	 * a check on 32bit, where we currently reduce the size of the per-MM
+	 * mapcount to a short.
+	 */
+	VM_WARN_ON_ONCE(diff > folio_large_nr_pages(folio));
+	VM_WARN_ON_ONCE(folio_large_nr_pages(folio) - 1 > MM_ID_MAPCOUNT_MAX);
+
+	VM_WARN_ON_ONCE(folio_mm_id(folio, 0) == MM_ID_DUMMY &&
+			folio->_mm_id_mapcount[0] != -1);
+	VM_WARN_ON_ONCE(folio_mm_id(folio, 0) != MM_ID_DUMMY &&
+			folio->_mm_id_mapcount[0] < 0);
+	VM_WARN_ON_ONCE(folio_mm_id(folio, 1) == MM_ID_DUMMY &&
+			folio->_mm_id_mapcount[1] != -1);
+	VM_WARN_ON_ONCE(folio_mm_id(folio, 1) != MM_ID_DUMMY &&
+			folio->_mm_id_mapcount[1] < 0);
+	VM_WARN_ON_ONCE(!folio_mapped(folio) &&
+			folio_test_large_maybe_mapped_shared(folio));
+}
+
+static __always_inline void folio_set_large_mapcount(struct folio *folio,
+		int mapcount, struct vm_area_struct *vma)
+{
+	__folio_large_mapcount_sanity_checks(folio, mapcount, vma->vm_mm->mm_id);
+
+	VM_WARN_ON_ONCE(folio_mm_id(folio, 0) != MM_ID_DUMMY);
+	VM_WARN_ON_ONCE(folio_mm_id(folio, 1) != MM_ID_DUMMY);
+
+	/* Note: mapcounts start at -1. */
+	atomic_set(&folio->_large_mapcount, mapcount - 1);
+	folio->_mm_id_mapcount[0] = mapcount - 1;
+	folio_set_mm_id(folio, 0, vma->vm_mm->mm_id);
+}
+
+static __always_inline void folio_add_large_mapcount(struct folio *folio,
+		int diff, struct vm_area_struct *vma)
+{
+	const mm_id_t mm_id = vma->vm_mm->mm_id;
+	int new_mapcount_val;
+
+	folio_lock_large_mapcount(folio);
+	__folio_large_mapcount_sanity_checks(folio, diff, mm_id);
+
+	new_mapcount_val = atomic_read(&folio->_large_mapcount) + diff;
+	atomic_set(&folio->_large_mapcount, new_mapcount_val);
+
+	/*
+	 * If a folio is mapped more than once into an MM on 32bit, we
+	 * can in theory overflow the per-MM mapcount (although only for
+	 * fairly large folios), turning it negative. In that case, just
+	 * free up the slot and mark the folio "mapped shared", otherwise
+	 * we might be in trouble when unmapping pages later.
+	 */
+	if (folio_mm_id(folio, 0) == mm_id) {
+		folio->_mm_id_mapcount[0] += diff;
+		if (!IS_ENABLED(CONFIG_64BIT) && unlikely(folio->_mm_id_mapcount[0] < 0)) {
+			folio->_mm_id_mapcount[0] = -1;
+			folio_set_mm_id(folio, 0, MM_ID_DUMMY);
+			folio->_mm_ids |= FOLIO_MM_IDS_SHARED_BIT;
+		}
+	} else if (folio_mm_id(folio, 1) == mm_id) {
+		folio->_mm_id_mapcount[1] += diff;
+		if (!IS_ENABLED(CONFIG_64BIT) && unlikely(folio->_mm_id_mapcount[1] < 0)) {
+			folio->_mm_id_mapcount[1] = -1;
+			folio_set_mm_id(folio, 1, MM_ID_DUMMY);
+			folio->_mm_ids |= FOLIO_MM_IDS_SHARED_BIT;
+		}
+	} else if (folio_mm_id(folio, 0) == MM_ID_DUMMY) {
+		folio_set_mm_id(folio, 0, mm_id);
+		folio->_mm_id_mapcount[0] = diff - 1;
+		/* We might have other mappings already. */
+		if (new_mapcount_val != diff - 1)
+			folio->_mm_ids |= FOLIO_MM_IDS_SHARED_BIT;
+	} else if (folio_mm_id(folio, 1) == MM_ID_DUMMY) {
+		folio_set_mm_id(folio, 1, mm_id);
+		folio->_mm_id_mapcount[1] = diff - 1;
+		/* Slot 0 certainly has mappings as well. */
+		folio->_mm_ids |= FOLIO_MM_IDS_SHARED_BIT;
+	}
+	folio_unlock_large_mapcount(folio);
+}
+
+static __always_inline void folio_sub_large_mapcount(struct folio *folio,
+		int diff, struct vm_area_struct *vma)
+{
+	const mm_id_t mm_id = vma->vm_mm->mm_id;
+	int new_mapcount_val;
+
+	folio_lock_large_mapcount(folio);
+	__folio_large_mapcount_sanity_checks(folio, diff, mm_id);
+
+	new_mapcount_val = atomic_read(&folio->_large_mapcount) - diff;
+	atomic_set(&folio->_large_mapcount, new_mapcount_val);
+
+	/*
+	 * There are valid corner cases where we might underflow a per-MM
+	 * mapcount (some mappings added when no slot was free, some mappings
+	 * added once a slot was free), so we always set it to -1 once we go
+	 * negative.
+	 */
+	if (folio_mm_id(folio, 0) == mm_id) {
+		folio->_mm_id_mapcount[0] -= diff;
+		if (folio->_mm_id_mapcount[0] >= 0)
+			goto out;
+		folio->_mm_id_mapcount[0] = -1;
+		folio_set_mm_id(folio, 0, MM_ID_DUMMY);
+	} else if (folio_mm_id(folio, 1) == mm_id) {
+		folio->_mm_id_mapcount[1] -= diff;
+		if (folio->_mm_id_mapcount[1] >= 0)
+			goto out;
+		folio->_mm_id_mapcount[1] = -1;
+		folio_set_mm_id(folio, 1, MM_ID_DUMMY);
+	}
+
+	/*
+	 * If one MM slot owns all mappings, the folio is mapped exclusively.
+	 * Note that if the folio is now unmapped (new_mapcount_val == -1), both
+	 * slots must be free (mapcount == -1), and we'll also mark it as
+	 * exclusive.
+	 */
+	if (folio->_mm_id_mapcount[0] == new_mapcount_val ||
+	    folio->_mm_id_mapcount[1] == new_mapcount_val)
+		folio->_mm_ids &= ~FOLIO_MM_IDS_SHARED_BIT;
+out:
+	folio_unlock_large_mapcount(folio);
+}
+#else /* !CONFIG_MM_ID */
+/*
+ * See __folio_rmap_sanity_checks(), we might map large folios even without
+ * CONFIG_TRANSPARENT_HUGEPAGE. We'll keep that working for now.
+ */
 static inline void folio_set_large_mapcount(struct folio *folio, int mapcount,
 		struct vm_area_struct *vma)
 {
@@ -191,6 +355,7 @@ static inline void folio_sub_large_mapcount(struct folio *folio,
 {
 	atomic_sub(diff, &folio->_large_mapcount);
 }
+#endif /* CONFIG_MM_ID */
 
 #define folio_inc_large_mapcount(folio, vma) \
 	folio_add_large_mapcount(folio, 1, vma)
diff --git a/kernel/fork.c b/kernel/fork.c
index 364b2d4fd3efa..f9cf0f056eb6f 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -802,6 +802,36 @@ static int dup_mmap(struct mm_struct *mm, struct mm_struct *oldmm)
 #define mm_free_pgd(mm)
 #endif /* CONFIG_MMU */
 
+#ifdef CONFIG_MM_ID
+static DEFINE_IDA(mm_ida);
+
+static inline int mm_alloc_id(struct mm_struct *mm)
+{
+	int ret;
+
+	ret = ida_alloc_range(&mm_ida, MM_ID_MIN, MM_ID_MAX, GFP_KERNEL);
+	if (ret < 0)
+		return ret;
+	mm->mm_id = ret;
+	return 0;
+}
+
+static inline void mm_free_id(struct mm_struct *mm)
+{
+	const mm_id_t id = mm->mm_id;
+
+	mm->mm_id = MM_ID_DUMMY;
+	if (id == MM_ID_DUMMY)
+		return;
+	if (WARN_ON_ONCE(id < MM_ID_MIN || id > MM_ID_MAX))
+		return;
+	ida_free(&mm_ida, id);
+}
+#else /* !CONFIG_MM_ID */
+static inline int mm_alloc_id(struct mm_struct *mm) { return 0; }
+static inline void mm_free_id(struct mm_struct *mm) {}
+#endif /* CONFIG_MM_ID */
+
 static void check_mm(struct mm_struct *mm)
 {
 	int i;
@@ -905,6 +935,7 @@ void __mmdrop(struct mm_struct *mm)
 
 	WARN_ON_ONCE(mm == current->active_mm);
 	mm_free_pgd(mm);
+	mm_free_id(mm);
 	destroy_context(mm);
 	mmu_notifier_subscriptions_destroy(mm);
 	check_mm(mm);
@@ -1289,6 +1320,9 @@ static struct mm_struct *mm_init(struct mm_struct *mm, struct task_struct *p,
 	if (mm_alloc_pgd(mm))
 		goto fail_nopgd;
 
+	if (mm_alloc_id(mm))
+		goto fail_noid;
+
 	if (init_new_context(p, mm))
 		goto fail_nocontext;
 
@@ -1308,6 +1342,8 @@ static struct mm_struct *mm_init(struct mm_struct *mm, struct task_struct *p,
 fail_cid:
 	destroy_context(mm);
 fail_nocontext:
+	mm_free_id(mm);
+fail_noid:
 	mm_free_pgd(mm);
 fail_nopgd:
 	free_mm(mm);
diff --git a/mm/Kconfig b/mm/Kconfig
index 2761098dbc1a2..af72fe92f96b0 100644
--- a/mm/Kconfig
+++ b/mm/Kconfig
@@ -817,11 +817,15 @@ config ARCH_WANT_GENERAL_HUGETLB
 config ARCH_WANTS_THP_SWAP
 	def_bool n
 
+config MM_ID
+	def_bool n
+
 menuconfig TRANSPARENT_HUGEPAGE
 	bool "Transparent Hugepage Support"
 	depends on HAVE_ARCH_TRANSPARENT_HUGEPAGE && !PREEMPT_RT
 	select COMPACTION
 	select XARRAY_MULTI
+	select MM_ID
 	help
 	  Transparent Hugepages allows the kernel to use huge pages and
 	  huge tlb transparently to the applications whenever possible.
diff --git a/mm/internal.h b/mm/internal.h
index ffdc91b19322e..7303ddd9dac04 100644
--- a/mm/internal.h
+++ b/mm/internal.h
@@ -722,6 +722,11 @@ static inline void prep_compound_head(struct page *page, unsigned int order)
 	folio_set_order(folio, order);
 	atomic_set(&folio->_large_mapcount, -1);
 	atomic_set(&folio->_nr_pages_mapped, 0);
+	if (IS_ENABLED(CONFIG_MM_ID)) {
+		folio->_mm_ids = 0;
+		folio->_mm_id_mapcount[0] = -1;
+		folio->_mm_id_mapcount[1] = -1;
+	}
 	if (IS_ENABLED(CONFIG_64BIT) || order > 1) {
 		atomic_set(&folio->_pincount, 0);
 		atomic_set(&folio->_entire_mapcount, -1);
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 7036530bd1bca..08caa92553998 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -955,6 +955,16 @@ static int free_tail_page_prepare(struct page *head_page, struct page *page)
 			bad_page(page, "nonzero nr_pages_mapped");
 			goto out;
 		}
+		if (IS_ENABLED(CONFIG_MM_ID)) {
+			if (unlikely(folio->_mm_id_mapcount[0] != -1)) {
+				bad_page(page, "nonzero mm mapcount 0");
+				goto out;
+			}
+			if (unlikely(folio->_mm_id_mapcount[1] != -1)) {
+				bad_page(page, "nonzero mm mapcount 1");
+				goto out;
+			}
+		}
 		if (IS_ENABLED(CONFIG_64BIT)) {
 			if (unlikely(atomic_read(&folio->_entire_mapcount) + 1)) {
 				bad_page(page, "nonzero entire_mapcount");
-- 
2.48.1

[PATCH v2 13/20] mm: Copy-on-Write (COW) reuse support for PTE-mapped THP

[David Hildenbrand]

Currently, we never end up reusing PTE-mapped THPs after fork. This
wasn't really a problem with PMD-sized THPs, because they would have to
be PTE-mapped first, but it's getting a problem with smaller THP
sizes that are effectively always PTE-mapped.

With our new "mapped exclusively" vs "maybe mapped shared" logic for
large folios, implementing CoW reuse for PTE-mapped THPs is straight
forward: if exclusively mapped, make sure that all references are
from these (our) mappings. Add some helpful comments to explain the
details.

CONFIG_TRANSPARENT_HUGEPAGE selects CONFIG_MM_ID. If we spot an anon
large folio without CONFIG_TRANSPARENT_HUGEPAGE in that code, something
is seriously messed up.

There are plenty of things we can optimize in the future: For example, we
could remember that the folio is fully exclusive so we could speedup
the next fault further. Also, we could try "faulting around", turning
surrounding PTEs that map the same folio writable. But especially the
latter might increase COW latency, so it would need further
investigation.

Signed-off-by: David Hildenbrand <david@redhat.com>
---
 mm/memory.c | 84 ++++++++++++++++++++++++++++++++++++++++++++++++-----
 1 file changed, 76 insertions(+), 8 deletions(-)

diff --git a/mm/memory.c b/mm/memory.c
index ef74c4f009f84..8dc241961b684 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -3706,19 +3706,87 @@ static vm_fault_t wp_page_shared(struct vm_fault *vmf, struct folio *folio)
 	return ret;
 }
 
-static bool wp_can_reuse_anon_folio(struct folio *folio,
-				    struct vm_area_struct *vma)
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+static bool __wp_can_reuse_large_anon_folio(struct folio *folio,
+		struct vm_area_struct *vma)
 {
+	bool exclusive = false;
+
+	/* Let's just free up a large folio if only a single page is mapped. */
+	if (folio_large_mapcount(folio) <= 1)
+		return false;
+
 	/*
-	 * We could currently only reuse a subpage of a large folio if no
-	 * other subpages of the large folios are still mapped. However,
-	 * let's just consistently not reuse subpages even if we could
-	 * reuse in that scenario, and give back a large folio a bit
-	 * sooner.
+	 * The assumption for anonymous folios is that each page can only get
+	 * mapped once into each MM. The only exception are KSM folios, which
+	 * are always small.
+	 *
+	 * Each taken mapcount must be paired with exactly one taken reference,
+	 * whereby the refcount must be incremented before the mapcount when
+	 * mapping a page, and the refcount must be decremented after the
+	 * mapcount when unmapping a page.
+	 *
+	 * If all folio references are from mappings, and all mappings are in
+	 * the page tables of this MM, then this folio is exclusive to this MM.
 	 */
-	if (folio_test_large(folio))
+	if (folio_test_large_maybe_mapped_shared(folio))
+		return false;
+
+	VM_WARN_ON_ONCE(folio_test_ksm(folio));
+	VM_WARN_ON_ONCE(folio_mapcount(folio) > folio_nr_pages(folio));
+	VM_WARN_ON_ONCE(folio_entire_mapcount(folio));
+
+	if (unlikely(folio_test_swapcache(folio))) {
+		/*
+		 * Note: freeing up the swapcache will fail if some PTEs are
+		 * still swap entries.
+		 */
+		if (!folio_trylock(folio))
+			return false;
+		folio_free_swap(folio);
+		folio_unlock(folio);
+	}
+
+	if (folio_large_mapcount(folio) != folio_ref_count(folio))
 		return false;
 
+	/* Stabilize the mapcount vs. refcount and recheck. */
+	folio_lock_large_mapcount(folio);
+	VM_WARN_ON_ONCE(folio_large_mapcount(folio) < folio_ref_count(folio));
+
+	if (folio_test_large_maybe_mapped_shared(folio))
+		goto unlock;
+	if (folio_large_mapcount(folio) != folio_ref_count(folio))
+		goto unlock;
+
+	VM_WARN_ON_ONCE(folio_mm_id(folio, 0) != vma->vm_mm->mm_id &&
+			folio_mm_id(folio, 1) != vma->vm_mm->mm_id);
+
+	/*
+	 * Do we need the folio lock? Likely not. If there would have been
+	 * references from page migration/swapout, we would have detected
+	 * an additional folio reference and never ended up here.
+	 */
+	exclusive = true;
+unlock:
+	folio_unlock_large_mapcount(folio);
+	return exclusive;
+}
+#else /* !CONFIG_TRANSPARENT_HUGEPAGE */
+static bool __wp_can_reuse_large_anon_folio(struct folio *folio,
+		struct vm_area_struct *vma)
+{
+	BUILD_BUG();
+}
+#endif /* CONFIG_TRANSPARENT_HUGEPAGE */
+
+static bool wp_can_reuse_anon_folio(struct folio *folio,
+				    struct vm_area_struct *vma)
+{
+	if (IS_ENABLED(CONFIG_TRANSPARENT_HUGEPAGE) && folio_test_large(folio))
+		return __wp_can_reuse_large_anon_folio(folio, vma);
+	VM_WARN_ON_ONCE(folio_test_large(folio));
+
 	/*
 	 * We have to verify under folio lock: these early checks are
 	 * just an optimization to avoid locking the folio and freeing
-- 
2.48.1

[PATCH v2 14/20] mm: convert folio_likely_mapped_shared() to folio_maybe_mapped_shared()

[David Hildenbrand]

Let's reuse our new MM ownership tracking infrastructure for large folios
to make folio_likely_mapped_shared() never return false negatives --
never indicating "not mapped shared" although the folio *is* mapped shared.
With that, we can rename it to folio_maybe_mapped_shared() and get rid of
the dependency on the mapcount of the first folio page.

The semantics are now arguably clearer: no mixture of "false negatives" and
"false positives", only the remaining possibility for "false positives".

Thoroughly document the new semantics. We might now detect that a large
folio is "maybe mapped shared" although it *no longer* is -- but once was.
Now, if more than two MMs mapped a folio at the same time, and the MM
mapping the folio exclusively at the end is not one tracked in the two
folio MM slots, we will detect the folio as "maybe mapped shared".

For anonymous folios, usually (except weird corner cases) all PTEs
that target a "maybe mapped shared" folio are R/O. As soon as a child
process would write to them (iow, actively use them), we would CoW and
effectively replace these PTEs. Most cases (below) are not expected to
really matter with large anonymous folios for this reason.

Most importantly, there will be no change at all for:
* small folios
* hugetlb folios
* PMD-mapped PMD-sized THPs (single mapping)

This change has the potential to affect existing callers of
folio_likely_mapped_shared() -> folio_maybe_mapped_shared():

(1) fs/proc/task_mmu.c: no change (hugetlb)

(2) khugepaged counts PTEs that target shared folios towards
    max_ptes_shared (default: HPAGE_PMD_NR / 2), meaning we could skip a
    collapse where we would have previously collapsed. This only applies to
    anonymous folios and is not expected to matter in practice.

    Worth noting that this change sorts out case (A) documented in
    commit 1bafe96e89f0 ("mm/khugepaged: replace page_mapcount() check by
    folio_likely_mapped_shared()") by removing the possibility
    for "false negatives".

(3) MADV_COLD / MADV_PAGEOUT / MADV_FREE will not try splitting PTE-mapped
    THPs that are considered shared but not fully covered by the
    requested range, consequently not processing them.

    PMD-mapped PMD-sized THP are not affected, or when all PTEs are
    covered. These functions are usually only called on anon/file folios
    that are exclusively mapped most of the time (no other file mappings
    or no fork()), so the "false negatives" are not expected to matter in
    practice.

(4) mbind() / migrate_pages() / move_pages() will refuse to migrate shared
    folios unless MPOL_MF_MOVE_ALL is effective (requires CAP_SYS_NICE).
    We will now reject some folios that could be migrated.

    Similar to (3), especially with MPOL_MF_MOVE_ALL, so this is not
    expected to matter in practice.

    Note that cpuset_migrate_mm_workfn() calls do_migrate_pages() with
    MPOL_MF_MOVE_ALL.

(5) NUMA hinting

    mm/migrate.c:migrate_misplaced_folio_prepare() will skip file folios
    that are probably shared libraries (-> "mapped shared" and
    executable). This check would have detected it as a shared library
    at some point (at least 3 MMs mapping it), so detecting it
    afterwards does not sound wrong (still a shared library). Not
    expected to matter.

    mm/memory.c:numa_migrate_check() will indicate TNF_SHARED in
    MAP_SHARED file mappings when encountering a shared folio. Similar
    reasoning, not expected to matter.

    mm/mprotect.c:change_pte_range() will skip folios detected as shared
    in CoW mappings. Similarly, this is not expected to matter in
    practice, but if it would ever be a problem we could relax that
    check a bit (e.g., basing it on the average page-mapcount in a folio),
    because it was only an optimization when many (e.g., 288) processes
    were mapping the same folios -- see commit 859d4adc3415 ("mm: numa: do
    not trap faults on shared data section pages.")

(6) mm/rmap.c:folio_referenced_one() will skip exclusive swapbacked folios
    in dying processes. Applies to anonymous folios only. Without "false
    negatives", we'll now skip all actually shared ones. Skipping ones
    that are actually exclusive won't really matter, it's a pure
    optimization, and is not expected to matter in practice.

In theory, one can detect the problematic scenario: folio_mapcount() > 0
and no folio MM slot is occupied ("state unknown"). One could reset the MM
slots while doing an rmap walk, which migration / folio split already do
when setting everything up. Further, when batching PTEs we might naturally
learn about a owner (e.g., folio_mapcount() == nr_ptes) and could update
the owner. However, we'll defer that until the scenarios where it would
really matter are clear.

Signed-off-by: David Hildenbrand <david@redhat.com>
---
 fs/proc/task_mmu.c |  4 ++--
 include/linux/mm.h | 43 ++++++++++++++++++++++---------------------
 mm/huge_memory.c   |  2 +-
 mm/khugepaged.c    |  8 +++-----
 mm/madvise.c       |  6 +++---
 mm/memory.c        |  2 +-
 mm/mempolicy.c     |  8 ++++----
 mm/migrate.c       |  7 +++----
 mm/mprotect.c      |  2 +-
 mm/rmap.c          |  2 +-
 10 files changed, 41 insertions(+), 43 deletions(-)

diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
index f02cd362309a0..2bddcea65cbf1 100644
--- a/fs/proc/task_mmu.c
+++ b/fs/proc/task_mmu.c
@@ -1023,7 +1023,7 @@ static int smaps_hugetlb_range(pte_t *pte, unsigned long hmask,
 
 	if (folio) {
 		/* We treat non-present entries as "maybe shared". */
-		if (!present || folio_likely_mapped_shared(folio) ||
+		if (!present || folio_maybe_mapped_shared(folio) ||
 		    hugetlb_pmd_shared(pte))
 			mss->shared_hugetlb += huge_page_size(hstate_vma(vma));
 		else
@@ -1879,7 +1879,7 @@ static int pagemap_hugetlb_range(pte_t *ptep, unsigned long hmask,
 		if (!folio_test_anon(folio))
 			flags |= PM_FILE;
 
-		if (!folio_likely_mapped_shared(folio) &&
+		if (!folio_maybe_mapped_shared(folio) &&
 		    !hugetlb_pmd_shared(ptep))
 			flags |= PM_MMAP_EXCLUSIVE;
 
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 9c1290588a11e..98a67488b5fef 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -2245,23 +2245,18 @@ static inline size_t folio_size(const struct folio *folio)
 }
 
 /**
- * folio_likely_mapped_shared - Estimate if the folio is mapped into the page
- *				tables of more than one MM
+ * folio_maybe_mapped_shared - Whether the folio is mapped into the page
+ *			       tables of more than one MM
  * @folio: The folio.
  *
- * 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").
+ * This function checks if the folio maybe currently mapped into more than one
+ * MM ("maybe mapped shared"), or if the folio is certainly mapped into a single
+ * MM ("mapped exclusively").
  *
  * For KSM folios, this function also returns "mapped shared" when a folio is
  * mapped multiple times into the same MM, because the individual page mappings
  * are independent.
  *
- * 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 and anonymous hugetlb folios, the return
  * value will be exactly correct: non-KSM folios can only be mapped at most once
  * into an MM, and they cannot be partially mapped. KSM folios are
@@ -2269,8 +2264,8 @@ static inline size_t folio_size(const struct folio *folio)
  *
  * For other folios, the result can be fuzzy:
  *    #. 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.
+ *       indicate "mapped shared" (false positive) if a folio was mapped by
+ *       more than two MMs at one point in time.
  *    #. 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.
@@ -2287,7 +2282,7 @@ static inline size_t folio_size(const struct folio *folio)
  *
  * Return: Whether the folio is estimated to be mapped into more than one MM.
  */
-static inline bool folio_likely_mapped_shared(struct folio *folio)
+static inline bool folio_maybe_mapped_shared(struct folio *folio)
 {
 	int mapcount = folio_mapcount(folio);
 
@@ -2295,16 +2290,22 @@ static inline bool folio_likely_mapped_shared(struct folio *folio)
 	if (!folio_test_large(folio) || unlikely(folio_test_hugetlb(folio)))
 		return mapcount > 1;
 
-	/* A single mapping implies "mapped exclusively". */
-	if (mapcount <= 1)
-		return false;
-
-	/* If any page is mapped more than once we treat it "mapped shared". */
-	if (folio_entire_mapcount(folio) || mapcount > folio_nr_pages(folio))
+	/*
+	 * vm_insert_page() without CONFIG_TRANSPARENT_HUGEPAGE ...
+	 * simply assume "mapped shared", nobody should really care
+	 * about this for arbitrary kernel allocations.
+	 */
+	if (!IS_ENABLED(CONFIG_MM_ID))
 		return true;
 
-	/* Let's guess based on the first subpage. */
-	return atomic_read(&folio->_mapcount) > 0;
+	/*
+	 * A single mapping implies "mapped exclusively", even if the
+	 * folio flag says something different: it's easier to handle this
+	 * case here instead of on the RMAP hot path.
+	 */
+	if (mapcount <= 1)
+		return false;
+	return folio_test_large_maybe_mapped_shared(folio);
 }
 
 #ifndef HAVE_ARCH_MAKE_FOLIO_ACCESSIBLE
diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index a3264d88d4b49..d9a7614fe739a 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -2155,7 +2155,7 @@ bool madvise_free_huge_pmd(struct mmu_gather *tlb, struct vm_area_struct *vma,
 	 * 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_likely_mapped_shared(folio))
+	if (folio_maybe_mapped_shared(folio))
 		goto out;
 
 	if (!folio_trylock(folio))
diff --git a/mm/khugepaged.c b/mm/khugepaged.c
index 5f0be134141e8..cc945c6ab3bdb 100644
--- a/mm/khugepaged.c
+++ b/mm/khugepaged.c
@@ -607,7 +607,7 @@ static int __collapse_huge_page_isolate(struct vm_area_struct *vma,
 		VM_BUG_ON_FOLIO(!folio_test_anon(folio), folio);
 
 		/* See hpage_collapse_scan_pmd(). */
-		if (folio_likely_mapped_shared(folio)) {
+		if (folio_maybe_mapped_shared(folio)) {
 			++shared;
 			if (cc->is_khugepaged &&
 			    shared > khugepaged_max_ptes_shared) {
@@ -1359,11 +1359,9 @@ static int hpage_collapse_scan_pmd(struct mm_struct *mm,
 
 		/*
 		 * We treat a single page as shared if any part of the THP
-		 * is shared. "False negatives" from
-		 * folio_likely_mapped_shared() are not expected to matter
-		 * much in practice.
+		 * is shared.
 		 */
-		if (folio_likely_mapped_shared(folio)) {
+		if (folio_maybe_mapped_shared(folio)) {
 			++shared;
 			if (cc->is_khugepaged &&
 			    shared > khugepaged_max_ptes_shared) {
diff --git a/mm/madvise.c b/mm/madvise.c
index e01e93e179a8a..388dc289b5d12 100644
--- a/mm/madvise.c
+++ b/mm/madvise.c
@@ -387,7 +387,7 @@ static int madvise_cold_or_pageout_pte_range(pmd_t *pmd,
 		folio = pmd_folio(orig_pmd);
 
 		/* Do not interfere with other mappings of this folio */
-		if (folio_likely_mapped_shared(folio))
+		if (folio_maybe_mapped_shared(folio))
 			goto huge_unlock;
 
 		if (pageout_anon_only_filter && !folio_test_anon(folio))
@@ -486,7 +486,7 @@ static int madvise_cold_or_pageout_pte_range(pmd_t *pmd,
 			if (nr < folio_nr_pages(folio)) {
 				int err;
 
-				if (folio_likely_mapped_shared(folio))
+				if (folio_maybe_mapped_shared(folio))
 					continue;
 				if (pageout_anon_only_filter && !folio_test_anon(folio))
 					continue;
@@ -721,7 +721,7 @@ static int madvise_free_pte_range(pmd_t *pmd, unsigned long addr,
 			if (nr < folio_nr_pages(folio)) {
 				int err;
 
-				if (folio_likely_mapped_shared(folio))
+				if (folio_maybe_mapped_shared(folio))
 					continue;
 				if (!folio_trylock(folio))
 					continue;
diff --git a/mm/memory.c b/mm/memory.c
index 8dc241961b684..2a1e7d9722866 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -5672,7 +5672,7 @@ int numa_migrate_check(struct folio *folio, struct vm_fault *vmf,
 	 * Flag if the folio is shared between multiple address spaces. This
 	 * is later used when determining whether to group tasks together
 	 */
-	if (folio_likely_mapped_shared(folio) && (vma->vm_flags & VM_SHARED))
+	if (folio_maybe_mapped_shared(folio) && (vma->vm_flags & VM_SHARED))
 		*flags |= TNF_SHARED;
 	/*
 	 * For memory tiering mode, cpupid of slow memory page is used
diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index bbaadbeeb2919..530e71fe91476 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -642,11 +642,11 @@ static int queue_folios_hugetlb(pte_t *pte, unsigned long hmask,
 	 * 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.
 	 *
-	 * See folio_likely_mapped_shared() on possible imprecision when we
+	 * See folio_maybe_mapped_shared() on possible imprecision when we
 	 * cannot easily detect if a folio is shared.
 	 */
 	if ((flags & MPOL_MF_MOVE_ALL) ||
-	    (!folio_likely_mapped_shared(folio) && !hugetlb_pmd_shared(pte)))
+	    (!folio_maybe_mapped_shared(folio) && !hugetlb_pmd_shared(pte)))
 		if (!folio_isolate_hugetlb(folio, qp->pagelist))
 			qp->nr_failed++;
 unlock:
@@ -1033,10 +1033,10 @@ static bool migrate_folio_add(struct folio *folio, struct list_head *foliolist,
 	 * 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.
 	 *
-	 * See folio_likely_mapped_shared() on possible imprecision when we
+	 * See folio_maybe_mapped_shared() on possible imprecision when we
 	 * cannot easily detect if a folio is shared.
 	 */
-	if ((flags & MPOL_MF_MOVE_ALL) || !folio_likely_mapped_shared(folio)) {
+	if ((flags & MPOL_MF_MOVE_ALL) || !folio_maybe_mapped_shared(folio)) {
 		if (folio_isolate_lru(folio)) {
 			list_add_tail(&folio->lru, foliolist);
 			node_stat_mod_folio(folio,
diff --git a/mm/migrate.c b/mm/migrate.c
index 365c6daa8d1b1..fb4afd31baf0c 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -2228,7 +2228,7 @@ static int __add_folio_for_migration(struct folio *folio, int node,
 	if (folio_nid(folio) == node)
 		return 0;
 
-	if (folio_likely_mapped_shared(folio) && !migrate_all)
+	if (folio_maybe_mapped_shared(folio) && !migrate_all)
 		return -EACCES;
 
 	if (folio_test_hugetlb(folio)) {
@@ -2653,11 +2653,10 @@ int migrate_misplaced_folio_prepare(struct folio *folio,
 		 * processes with execute permissions as they are probably
 		 * shared libraries.
 		 *
-		 * See folio_likely_mapped_shared() on possible imprecision
+		 * See folio_maybe_mapped_shared() on possible imprecision
 		 * when we cannot easily detect if a folio is shared.
 		 */
-		if ((vma->vm_flags & VM_EXEC) &&
-		    folio_likely_mapped_shared(folio))
+		if ((vma->vm_flags & VM_EXEC) && folio_maybe_mapped_shared(folio))
 			return -EACCES;
 
 		/*
diff --git a/mm/mprotect.c b/mm/mprotect.c
index 1444878f7aeb2..62c1f79457412 100644
--- a/mm/mprotect.c
+++ b/mm/mprotect.c
@@ -133,7 +133,7 @@ static long change_pte_range(struct mmu_gather *tlb,
 				/* Also skip shared copy-on-write pages */
 				if (is_cow_mapping(vma->vm_flags) &&
 				    (folio_maybe_dma_pinned(folio) ||
-				     folio_likely_mapped_shared(folio)))
+				     folio_maybe_mapped_shared(folio)))
 					continue;
 
 				/*
diff --git a/mm/rmap.c b/mm/rmap.c
index c9922928616ee..8de415157bc8d 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -889,7 +889,7 @@ static bool folio_referenced_one(struct folio *folio,
 		if ((!atomic_read(&vma->vm_mm->mm_users) ||
 		    check_stable_address_space(vma->vm_mm)) &&
 		    folio_test_anon(folio) && folio_test_swapbacked(folio) &&
-		    !folio_likely_mapped_shared(folio)) {
+		    !folio_maybe_mapped_shared(folio)) {
 			pra->referenced = -1;
 			page_vma_mapped_walk_done(&pvmw);
 			return false;
-- 
2.48.1

[PATCH v2 15/20] mm: CONFIG_NO_PAGE_MAPCOUNT to prepare for not maintain per-page mapcounts in large folios

[David Hildenbrand]

We're close to the finishing line: let's introduce a new
CONFIG_NO_PAGE_MAPCOUNT config option where we will incrementally remove
any dependencies on per-page mapcounts in large folios. Once that's
done, we'll stop maintaining the per-page mapcounts with this
config option enabled.

CONFIG_NO_PAGE_MAPCOUNT will be EXPERIMENTAL for now, as we'll have to
learn about some of the real world impact of some of the implications.

As writing "!CONFIG_NO_PAGE_MAPCOUNT" is really nasty, let's introduce
a helper config option "CONFIG_PAGE_MAPCOUNT" that expresses the
negation.

Signed-off-by: David Hildenbrand <david@redhat.com>
---
 mm/Kconfig | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

diff --git a/mm/Kconfig b/mm/Kconfig
index af72fe92f96b0..64bb80a281dd9 100644
--- a/mm/Kconfig
+++ b/mm/Kconfig
@@ -889,8 +889,26 @@ config READ_ONLY_THP_FOR_FS
 	  support of file THPs will be developed in the next few release
 	  cycles.
 
+config NO_PAGE_MAPCOUNT
+	bool "No per-page mapcount (EXPERIMENTAL)"
+	help
+	  Do not maintain per-page mapcounts for pages part of larger
+	  allocations, such as transparent huge pages.
+
+	  When this config option is enabled, some interfaces that relied on
+	  this information will rely on less-precise per-allocation information
+	  instead: for example, using the average per-page mapcount in such
+	  a large allocation instead of the per-page mapcount.
+
+	  EXPERIMENTAL because the severity of some of the implications first
+	  have to be understood properly.
+
 endif # TRANSPARENT_HUGEPAGE
 
+# simple helper to make the code a bit easier to read
+config PAGE_MAPCOUNT
+	def_bool !NO_PAGE_MAPCOUNT
+
 #
 # The architecture supports pgtable leaves that is larger than PAGE_SIZE
 #
-- 
2.48.1

[PATCH v2 16/20] fs/proc/page: remove per-page mapcount dependency for /proc/kpagecount (CONFIG_NO_PAGE_MAPCOUNT)

[David Hildenbrand]

Let's implement an alternative when per-page mapcounts in large folios
are no longer maintained -- soon with CONFIG_NO_PAGE_MAPCOUNT.

For large folios, we'll return the per-page average mapcount within the
folio, except when the average is 0 but the folio is mapped: then we
return 1.

For hugetlb folios and for large folios that are fully mapped
into all address spaces, there is no change.

As an alternative, we could simply return 0 for non-hugetlb large folios,
or disable this legacy interface with CONFIG_NO_PAGE_MAPCOUNT.

But the information exposed by this interface can still be valuable, and
frequently we deal with fully-mapped large folios where the average
corresponds to the actual page mapcount. So we'll leave it like this for
now and document the new behavior.

Note: this interface is likely not very relevant for performance. If
ever required, we could try doing a rather expensive rmap walk to collect
precisely how often this folio page is mapped.

Signed-off-by: David Hildenbrand <david@redhat.com>
---
 Documentation/admin-guide/mm/pagemap.rst |  7 +++++-
 fs/proc/internal.h                       | 31 ++++++++++++++++++++++++
 fs/proc/page.c                           | 19 ++++++++++++---
 3 files changed, 53 insertions(+), 4 deletions(-)

diff --git a/Documentation/admin-guide/mm/pagemap.rst b/Documentation/admin-guide/mm/pagemap.rst
index caba0f52dd36c..49590306c61a0 100644
--- a/Documentation/admin-guide/mm/pagemap.rst
+++ b/Documentation/admin-guide/mm/pagemap.rst
@@ -42,7 +42,12 @@ There are four components to pagemap:
    skip over unmapped regions.
 
  * ``/proc/kpagecount``.  This file contains a 64-bit count of the number of
-   times each page is mapped, indexed by PFN.
+   times each page is mapped, indexed by PFN. Some kernel configurations do
+   not track the precise number of times a page part of a larger allocation
+   (e.g., THP) is mapped. In these configurations, the average number of
+   mappings per page in this larger allocation is returned instead. However,
+   if any page of the large allocation is mapped, the returned value will
+   be at least 1.
 
 The page-types tool in the tools/mm directory can be used to query the
 number of times a page is mapped.
diff --git a/fs/proc/internal.h b/fs/proc/internal.h
index 1695509370b88..16aa1fd260771 100644
--- a/fs/proc/internal.h
+++ b/fs/proc/internal.h
@@ -174,6 +174,37 @@ static inline int folio_precise_page_mapcount(struct folio *folio,
 	return mapcount;
 }
 
+/**
+ * folio_average_page_mapcount() - Average number of mappings per page in this
+ *				   folio
+ * @folio: The folio.
+ *
+ * The average number of present user page table entries that reference each
+ * page in this folio as tracked via the RMAP: either referenced directly
+ * (PTE) or as part of a larger area that covers this page (e.g., PMD).
+ *
+ * Returns: The average number of mappings per page in this folio. 0 for
+ * folios that are not mapped to user space or are not tracked via the RMAP
+ * (e.g., shared zeropage).
+ */
+static inline int folio_average_page_mapcount(struct folio *folio)
+{
+	int mapcount, entire_mapcount;
+	unsigned int adjust;
+
+	if (!folio_test_large(folio))
+		return atomic_read(&folio->_mapcount) + 1;
+
+	mapcount = folio_large_mapcount(folio);
+	entire_mapcount = folio_entire_mapcount(folio);
+	if (mapcount <= entire_mapcount)
+		return entire_mapcount;
+	mapcount -= entire_mapcount;
+
+	adjust = folio_large_nr_pages(folio) / 2;
+	return ((mapcount + adjust) >> folio_large_order(folio)) +
+		entire_mapcount;
+}
 /*
  * array.c
  */
diff --git a/fs/proc/page.c b/fs/proc/page.c
index a55f5acefa974..4d3290cc69667 100644
--- a/fs/proc/page.c
+++ b/fs/proc/page.c
@@ -67,9 +67,22 @@ static ssize_t kpagecount_read(struct file *file, char __user *buf,
 		 * memmaps that were actually initialized.
 		 */
 		page = pfn_to_online_page(pfn);
-		if (page)
-			mapcount = folio_precise_page_mapcount(page_folio(page),
-							       page);
+		if (page) {
+			struct folio *folio = page_folio(page);
+
+			if (IS_ENABLED(CONFIG_PAGE_MAPCOUNT)) {
+				mapcount = folio_precise_page_mapcount(folio, page);
+			} else {
+				/*
+				 * Indicate the per-page average, but at least "1" for
+				 * mapped folios.
+				 */
+				mapcount = folio_average_page_mapcount(folio);
+				if (!mapcount && folio_test_large(folio) &&
+				    folio_mapped(folio))
+					mapcount = 1;
+			}
+		}
 
 		if (put_user(mapcount, out)) {
 			ret = -EFAULT;
-- 
2.48.1

[PATCH v2 17/20] fs/proc/task_mmu: remove per-page mapcount dependency for PM_MMAP_EXCLUSIVE (CONFIG_NO_PAGE_MAPCOUNT)

[David Hildenbrand]

Let's implement an alternative when per-page mapcounts in large folios are
no longer maintained -- soon with CONFIG_NO_PAGE_MAPCOUNT.

PM_MMAP_EXCLUSIVE will now be set if folio_likely_mapped_shared() is
true -- when the folio is considered "mapped shared", including when
it once was "mapped shared" but no longer is, as documented.

This might result in and under-indication of "exclusively mapped", which
is considered better than over-indicating it: under-estimating the USS
(Unique Set Size) is better than over-estimating it.

As an alternative, we could simply remove that flag with
CONFIG_NO_PAGE_MAPCOUNT completely, but there might be value to it. So,
let's keep it like that and document the behavior.

Signed-off-by: David Hildenbrand <david@redhat.com>
---
 Documentation/admin-guide/mm/pagemap.rst |  9 +++++++++
 fs/proc/task_mmu.c                       | 11 +++++++++--
 2 files changed, 18 insertions(+), 2 deletions(-)

diff --git a/Documentation/admin-guide/mm/pagemap.rst b/Documentation/admin-guide/mm/pagemap.rst
index 49590306c61a0..131c86574c39a 100644
--- a/Documentation/admin-guide/mm/pagemap.rst
+++ b/Documentation/admin-guide/mm/pagemap.rst
@@ -37,6 +37,15 @@ There are four components to pagemap:
    precisely which pages are mapped (or in swap) and comparing mapped
    pages between processes.
 
+   Note that in some kernel configurations, all pages part of a larger
+   allocation (e.g., THP) might be considered "mapped shared" if the large
+   allocation is considered "mapped shared": if not all pages are exclusive to
+   the same process. Further, some kernel configurations might consider larger
+   allocations "mapped shared", if they were at one point considered
+   "mapped shared", even if they would now be considered "exclusively mapped".
+   Consequently, in these kernel configurations, bit 56 might be set although
+   the page is actually "exclusively mapped"
+
    Efficient users of this interface will use ``/proc/pid/maps`` to
    determine which areas of memory are actually mapped and llseek to
    skip over unmapped regions.
diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
index 2bddcea65cbf1..80839bbf9657f 100644
--- a/fs/proc/task_mmu.c
+++ b/fs/proc/task_mmu.c
@@ -1651,6 +1651,13 @@ static int add_to_pagemap(pagemap_entry_t *pme, struct pagemapread *pm)
 	return 0;
 }
 
+static bool __folio_page_mapped_exclusively(struct folio *folio, struct page *page)
+{
+	if (IS_ENABLED(CONFIG_PAGE_MAPCOUNT))
+		return folio_precise_page_mapcount(folio, page) == 1;
+	return !folio_maybe_mapped_shared(folio);
+}
+
 static int pagemap_pte_hole(unsigned long start, unsigned long end,
 			    __always_unused int depth, struct mm_walk *walk)
 {
@@ -1739,7 +1746,7 @@ static pagemap_entry_t pte_to_pagemap_entry(struct pagemapread *pm,
 		if (!folio_test_anon(folio))
 			flags |= PM_FILE;
 		if ((flags & PM_PRESENT) &&
-		    folio_precise_page_mapcount(folio, page) == 1)
+		    __folio_page_mapped_exclusively(folio, page))
 			flags |= PM_MMAP_EXCLUSIVE;
 	}
 	if (vma->vm_flags & VM_SOFTDIRTY)
@@ -1814,7 +1821,7 @@ static int pagemap_pmd_range(pmd_t *pmdp, unsigned long addr, unsigned long end,
 			pagemap_entry_t pme;
 
 			if (folio && (flags & PM_PRESENT) &&
-			    folio_precise_page_mapcount(folio, page + idx) == 1)
+			    __folio_page_mapped_exclusively(folio, page))
 				cur_flags |= PM_MMAP_EXCLUSIVE;
 
 			pme = make_pme(frame, cur_flags);
-- 
2.48.1

[PATCH v2 18/20] fs/proc/task_mmu: remove per-page mapcount dependency for "mapmax" (CONFIG_NO_PAGE_MAPCOUNT)

[David Hildenbrand]

Let's implement an alternative when per-page mapcounts in large folios are
no longer maintained -- soon with CONFIG_NO_PAGE_MAPCOUNT.

For calculating "mapmax", we now use the average per-page mapcount in
a large folio instead of the per-page mapcount.

For hugetlb folios and folios that are not partially mapped into MMs,
there is no change.

Likely, this change will not matter much in practice, and an alternative
might be to simple remove this stat with CONFIG_NO_PAGE_MAPCOUNT.
However, there might be value to it, so let's keep it like that and
document the behavior.

Signed-off-by: David Hildenbrand <david@redhat.com>
---
 Documentation/filesystems/proc.rst | 5 +++++
 fs/proc/task_mmu.c                 | 7 ++++++-
 2 files changed, 11 insertions(+), 1 deletion(-)

diff --git a/Documentation/filesystems/proc.rst b/Documentation/filesystems/proc.rst
index 09f0aed5a08ba..1aa190017f796 100644
--- a/Documentation/filesystems/proc.rst
+++ b/Documentation/filesystems/proc.rst
@@ -686,6 +686,11 @@ Where:
 node locality page counters (N0 == node0, N1 == node1, ...) and the kernel page
 size, in KB, that is backing the mapping up.
 
+Note that some kernel configurations do not track the precise number of times
+a page part of a larger allocation (e.g., THP) is mapped. In these
+configurations, "mapmax" might corresponds to the average number of mappings
+per page in such a larger allocation instead.
+
 1.2 Kernel data
 ---------------
 
diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
index 80839bbf9657f..d7ee842367f0f 100644
--- a/fs/proc/task_mmu.c
+++ b/fs/proc/task_mmu.c
@@ -2862,7 +2862,12 @@ static void gather_stats(struct page *page, struct numa_maps *md, int pte_dirty,
 			unsigned long nr_pages)
 {
 	struct folio *folio = page_folio(page);
-	int count = folio_precise_page_mapcount(folio, page);
+	int count;
+
+	if (IS_ENABLED(CONFIG_PAGE_MAPCOUNT))
+		count = folio_precise_page_mapcount(folio, page);
+	else
+		count = min_t(int, folio_average_page_mapcount(folio), 1);
 
 	md->pages += nr_pages;
 	if (pte_dirty || folio_test_dirty(folio))
-- 
2.48.1

[PATCH v2 19/20] fs/proc/task_mmu: remove per-page mapcount dependency for smaps/smaps_rollup (CONFIG_NO_PAGE_MAPCOUNT)

[David Hildenbrand]

Let's implement an alternative when per-page mapcounts in large folios are
no longer maintained -- soon with CONFIG_NO_PAGE_MAPCOUNT.

When computing the output for smaps / smaps_rollups, in particular when
calculating the USS (Unique Set Size) and the PSS (Proportional Set Size),
we still rely on per-page mapcounts.

To determine private vs. shared, we'll use folio_likely_mapped_shared(),
similar to how we handle PM_MMAP_EXCLUSIVE. Similarly, we might now
under-estimate the USS and count pages towards "shared" that are
actually "private" ("exclusively mapped").

When calculating the PSS, we'll now also use the average per-page
mapcount for large folios: this can result in both, an over-estimation
and an under-estimation of the PSS. The difference is not expected to
matter much in practice, but we'll have to learn as we go.

We can now provide folio_precise_page_mapcount() only with
CONFIG_PAGE_MAPCOUNT, and remove one of the last users of per-page
mapcounts when CONFIG_NO_PAGE_MAPCOUNT is enabled.

Document the new behavior.

Signed-off-by: David Hildenbrand <david@redhat.com>
---
 Documentation/filesystems/proc.rst | 13 +++++++++++++
 fs/proc/internal.h                 |  8 ++++++++
 fs/proc/task_mmu.c                 | 17 +++++++++++++++--
 3 files changed, 36 insertions(+), 2 deletions(-)

diff --git a/Documentation/filesystems/proc.rst b/Documentation/filesystems/proc.rst
index 1aa190017f796..57d55274a1f42 100644
--- a/Documentation/filesystems/proc.rst
+++ b/Documentation/filesystems/proc.rst
@@ -506,6 +506,19 @@ Note that even a page which is part of a MAP_SHARED mapping, but has only
 a single pte mapped, i.e.  is currently used by only one process, is accounted
 as private and not as shared.
 
+Note that in some kernel configurations, all pages part of a larger allocation
+(e.g., THP) might be considered "shared" if the large allocation is
+considered "shared": if not all pages are exclusive to the same process.
+Further, some kernel configurations might consider larger allocations "shared",
+if they were at one point considered "shared", even if they would now be
+considered "exclusive".
+
+Some kernel configurations do not track the precise number of times a page part
+of a larger allocation is mapped. In this case, when calculating the PSS, the
+average number of mappings per page in this larger allocation might be used
+as an approximation for the number of mappings of a page. The PSS calculation
+will be imprecise in this case.
+
 "Referenced" indicates the amount of memory currently marked as referenced or
 accessed.
 
diff --git a/fs/proc/internal.h b/fs/proc/internal.h
index 16aa1fd260771..70205425a2daa 100644
--- a/fs/proc/internal.h
+++ b/fs/proc/internal.h
@@ -143,6 +143,7 @@ unsigned name_to_int(const struct qstr *qstr);
 /* Worst case buffer size needed for holding an integer. */
 #define PROC_NUMBUF 13
 
+#ifdef CONFIG_PAGE_MAPCOUNT
 /**
  * folio_precise_page_mapcount() - Number of mappings of this folio page.
  * @folio: The folio.
@@ -173,6 +174,13 @@ static inline int folio_precise_page_mapcount(struct folio *folio,
 
 	return mapcount;
 }
+#else /* !CONFIG_PAGE_MAPCOUNT */
+static inline int folio_precise_page_mapcount(struct folio *folio,
+		struct page *page)
+{
+	BUILD_BUG();
+}
+#endif /* CONFIG_PAGE_MAPCOUNT */
 
 /**
  * folio_average_page_mapcount() - Average number of mappings per page in this
diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
index d7ee842367f0f..7ca0bc3bf417d 100644
--- a/fs/proc/task_mmu.c
+++ b/fs/proc/task_mmu.c
@@ -707,6 +707,8 @@ static void smaps_account(struct mem_size_stats *mss, struct page *page,
 	struct folio *folio = page_folio(page);
 	int i, nr = compound ? compound_nr(page) : 1;
 	unsigned long size = nr * PAGE_SIZE;
+	bool exclusive;
+	int mapcount;
 
 	/*
 	 * First accumulate quantities that depend only on |size| and the type
@@ -747,18 +749,29 @@ static void smaps_account(struct mem_size_stats *mss, struct page *page,
 				      dirty, locked, present);
 		return;
 	}
+
+	if (IS_ENABLED(CONFIG_NO_PAGE_MAPCOUNT)) {
+		mapcount = folio_average_page_mapcount(folio);
+		exclusive = !folio_maybe_mapped_shared(folio);
+	}
+
 	/*
 	 * We obtain a snapshot of the mapcount. Without holding the folio lock
 	 * this snapshot can be slightly wrong as we cannot always read the
 	 * mapcount atomically.
 	 */
 	for (i = 0; i < nr; i++, page++) {
-		int mapcount = folio_precise_page_mapcount(folio, page);
 		unsigned long pss = PAGE_SIZE << PSS_SHIFT;
+
+		if (IS_ENABLED(CONFIG_PAGE_MAPCOUNT)) {
+			mapcount = folio_precise_page_mapcount(folio, page);
+			exclusive = mapcount < 2;
+		}
+
 		if (mapcount >= 2)
 			pss /= mapcount;
 		smaps_page_accumulate(mss, folio, PAGE_SIZE, pss,
-				dirty, locked, mapcount < 2);
+				dirty, locked, exclusive);
 	}
 }
 
-- 
2.48.1

[PATCH v2 20/20] mm: stop maintaining the per-page mapcount of large folios (CONFIG_NO_PAGE_MAPCOUNT)

[David Hildenbrand]

Everything is in place to stop using the per-page mapcounts in large
folios: the mapcount of tail pages will always be logically 0 (-1 value),
just like it currently is for hugetlb folios already, and the page
mapcount of the head page is either 0 (-1 value) or contains a page type
(e.g., hugetlb).

Maintaining _nr_pages_mapped without per-page mapcounts is impossible,
so that one also has to go with CONFIG_NO_PAGE_MAPCOUNT.

There are two remaining implications:

(1) Per-node, per-cgroup and per-lruvec stats of "NR_ANON_MAPPED"
    ("mapped anonymous memory") and "NR_FILE_MAPPED"
    ("mapped file memory"):

    As soon as any page of the folio is mapped -- folio_mapped() -- we
    now account the complete folio as mapped. Once the last page is
    unmapped -- !folio_mapped() -- we account the complete folio as
    unmapped.

    This implies that ...

    * "AnonPages" and "Mapped" in /proc/meminfo and
      /sys/devices/system/node/*/meminfo
    * cgroup v2: "anon" and "file_mapped" in "memory.stat" and
      "memory.numa_stat"
    * cgroup v1: "rss" and "mapped_file" in "memory.stat" and
      "memory.numa_stat

    ... can now appear higher than before. But note that these folios do
    consume that memory, simply not all pages are actually currently
    mapped.

    It's worth nothing that other accounting in the kernel (esp. cgroup
    charging on allocation) is not affected by this change.

    [why oh why is "anon" called "rss" in cgroup v1]

 (2) Detecting partial mappings

     Detecting whether anon THPs are partially mapped gets a bit more
     unreliable. As long as a single MM maps such a large folio
     ("exclusively mapped"), we can reliably detect it. Especially before
     fork() / after a short-lived child process quit, we will detect
     partial mappings reliably, which is the common case.

     In essence, if the average per-page mapcount in an anon THP is < 1,
     we know for sure that we have a partial mapping.

     However, as soon as multiple MMs are involved, we might miss detecting
     partial mappings: this might be relevant with long-lived child
     processes. If we have a fully-mapped anon folio before fork(), once
     our child processes and our parent all unmap (zap/COW) the same pages
     (but not the complete folio), we might not detect the partial mapping.
     However, once the child processes quit we would detect the partial
     mapping.

     How relevant this case is in practice remains to be seen.
     Swapout/migration will likely mitigate this.

     In the future, RMAP walkers could check for that for that case
     (e.g., when collecting access bits during reclaim) and simply flag
     them for deferred-splitting.

Signed-off-by: David Hildenbrand <david@redhat.com>
---
 .../admin-guide/cgroup-v1/memory.rst          |  4 +
 Documentation/admin-guide/cgroup-v2.rst       | 10 ++-
 Documentation/filesystems/proc.rst            | 10 ++-
 Documentation/mm/transhuge.rst                | 31 +++++--
 include/linux/rmap.h                          | 35 ++++++--
 mm/internal.h                                 |  5 +-
 mm/page_alloc.c                               |  3 +-
 mm/rmap.c                                     | 80 +++++++++++++++++--
 8 files changed, 150 insertions(+), 28 deletions(-)

diff --git a/Documentation/admin-guide/cgroup-v1/memory.rst b/Documentation/admin-guide/cgroup-v1/memory.rst
index 286d16fc22ebb..53cf081b22e81 100644
--- a/Documentation/admin-guide/cgroup-v1/memory.rst
+++ b/Documentation/admin-guide/cgroup-v1/memory.rst
@@ -609,6 +609,10 @@ memory.stat file includes following statistics:
 
 	'rss + mapped_file" will give you resident set size of cgroup.
 
+	Note that some kernel configurations might account complete larger
+	allocations (e.g., THP) towards 'rss' and 'mapped_file', even if
+	only some, but not all that memory is mapped.
+
 	(Note: file and shmem may be shared among other cgroups. In that case,
 	mapped_file is accounted only when the memory cgroup is owner of page
 	cache.)
diff --git a/Documentation/admin-guide/cgroup-v2.rst b/Documentation/admin-guide/cgroup-v2.rst
index 175e9435ad5c1..53ada5c2620a7 100644
--- a/Documentation/admin-guide/cgroup-v2.rst
+++ b/Documentation/admin-guide/cgroup-v2.rst
@@ -1448,7 +1448,10 @@ The following nested keys are defined.
 
 	  anon
 		Amount of memory used in anonymous mappings such as
-		brk(), sbrk(), and mmap(MAP_ANONYMOUS)
+		brk(), sbrk(), and mmap(MAP_ANONYMOUS). Note that
+		some kernel configurations might account complete larger
+		allocations (e.g., THP) if only some, but not all the
+		memory of such an allocation is mapped anymore.
 
 	  file
 		Amount of memory used to cache filesystem data,
@@ -1491,7 +1494,10 @@ The following nested keys are defined.
 		Amount of application memory swapped out to zswap.
 
 	  file_mapped
-		Amount of cached filesystem data mapped with mmap()
+		Amount of cached filesystem data mapped with mmap(). Note
+		that some kernel configurations might account complete
+		larger allocations (e.g., THP) if only some, but not
+		not all the memory of such an allocation is mapped.
 
 	  file_dirty
 		Amount of cached filesystem data that was modified but
diff --git a/Documentation/filesystems/proc.rst b/Documentation/filesystems/proc.rst
index 57d55274a1f42..c5052acfa0747 100644
--- a/Documentation/filesystems/proc.rst
+++ b/Documentation/filesystems/proc.rst
@@ -1150,9 +1150,15 @@ Dirty
 Writeback
               Memory which is actively being written back to the disk
 AnonPages
-              Non-file backed pages mapped into userspace page tables
+              Non-file backed pages mapped into userspace page tables. Note that
+              some kernel configurations might consider all pages part of a
+              larger allocation (e.g., THP) as "mapped", as soon as a single
+              page is mapped.
 Mapped
-              files which have been mmapped, such as libraries
+              files which have been mmapped, such as libraries. Note that some
+              kernel configurations might consider all pages part of a larger
+              allocation (e.g., THP) as "mapped", as soon as a single page is
+              mapped.
 Shmem
               Total memory used by shared memory (shmem) and tmpfs
 KReclaimable
diff --git a/Documentation/mm/transhuge.rst b/Documentation/mm/transhuge.rst
index baa17d718a762..0e7f8e4cd2e33 100644
--- a/Documentation/mm/transhuge.rst
+++ b/Documentation/mm/transhuge.rst
@@ -116,23 +116,28 @@ pages:
     succeeds on tail pages.
 
   - map/unmap of a PMD entry for the whole THP increment/decrement
-    folio->_entire_mapcount, increment/decrement folio->_large_mapcount
-    and also increment/decrement folio->_nr_pages_mapped by ENTIRELY_MAPPED
-    when _entire_mapcount goes from -1 to 0 or 0 to -1.
+    folio->_entire_mapcount and folio->_large_mapcount.
 
     We also maintain the two slots for tracking MM owners (MM ID and
     corresponding mapcount), and the current status ("maybe mapped shared" vs.
     "mapped exclusively").
 
+    With CONFIG_PAGE_MAPCOUNT, we also increment/decrement
+    folio->_nr_pages_mapped by ENTIRELY_MAPPED when _entire_mapcount goes
+    from -1 to 0 or 0 to -1.
+
   - map/unmap of individual pages with PTE entry increment/decrement
-    page->_mapcount, increment/decrement folio->_large_mapcount and also
-    increment/decrement folio->_nr_pages_mapped when page->_mapcount goes
-    from -1 to 0 or 0 to -1 as this counts the number of pages mapped by PTE.
+    folio->_large_mapcount.
 
     We also maintain the two slots for tracking MM owners (MM ID and
     corresponding mapcount), and the current status ("maybe mapped shared" vs.
     "mapped exclusively").
 
+    With CONFIG_PAGE_MAPCOUNT, we also increment/decrement
+    page->_mapcount and increment/decrement folio->_nr_pages_mapped when
+    page->_mapcount goes from -1 to 0 or 0 to -1 as this counts the number
+    of pages mapped by PTE.
+
 split_huge_page internally has to distribute the refcounts in the head
 page to the tail pages before clearing all PG_head/tail bits from the page
 structures. It can be done easily for refcounts taken by page table
@@ -159,8 +164,8 @@ clear where references should go after split: it will stay on the head page.
 Note that split_huge_pmd() doesn't have any limitations on refcounting:
 pmd can be split at any point and never fails.
 
-Partial unmap and deferred_split_folio()
-========================================
+Partial unmap and deferred_split_folio() (anon THP only)
+========================================================
 
 Unmapping part of THP (with munmap() or other way) is not going to free
 memory immediately. Instead, we detect that a subpage of THP is not in use
@@ -175,3 +180,13 @@ a THP crosses a VMA boundary.
 The function deferred_split_folio() is used to queue a folio for splitting.
 The splitting itself will happen when we get memory pressure via shrinker
 interface.
+
+With CONFIG_PAGE_MAPCOUNT, we reliably detect partial mappings based on
+folio->_nr_pages_mapped.
+
+With CONFIG_NO_PAGE_MAPCOUNT, we detect partial mappings based on the
+average per-page mapcount in a THP: if the average is < 1, an anon THP is
+certainly partially mapped. As long as only a single process maps a THP,
+this detection is reliable. With long-running child processes, there can
+be scenarios where partial mappings can currently not be detected, and
+might need asynchronous detection during memory reclaim in the future.
diff --git a/include/linux/rmap.h b/include/linux/rmap.h
index c131b0efff0fa..6b82b618846ee 100644
--- a/include/linux/rmap.h
+++ b/include/linux/rmap.h
@@ -240,7 +240,7 @@ static __always_inline void folio_set_large_mapcount(struct folio *folio,
 	folio_set_mm_id(folio, 0, vma->vm_mm->mm_id);
 }
 
-static __always_inline void folio_add_large_mapcount(struct folio *folio,
+static __always_inline int folio_add_return_large_mapcount(struct folio *folio,
 		int diff, struct vm_area_struct *vma)
 {
 	const mm_id_t mm_id = vma->vm_mm->mm_id;
@@ -286,9 +286,11 @@ static __always_inline void folio_add_large_mapcount(struct folio *folio,
 		folio->_mm_ids |= FOLIO_MM_IDS_SHARED_BIT;
 	}
 	folio_unlock_large_mapcount(folio);
+	return new_mapcount_val + 1;
 }
+#define folio_add_large_mapcount folio_add_return_large_mapcount
 
-static __always_inline void folio_sub_large_mapcount(struct folio *folio,
+static __always_inline int folio_sub_return_large_mapcount(struct folio *folio,
 		int diff, struct vm_area_struct *vma)
 {
 	const mm_id_t mm_id = vma->vm_mm->mm_id;
@@ -331,7 +333,9 @@ static __always_inline void folio_sub_large_mapcount(struct folio *folio,
 		folio->_mm_ids &= ~FOLIO_MM_IDS_SHARED_BIT;
 out:
 	folio_unlock_large_mapcount(folio);
+	return new_mapcount_val + 1;
 }
+#define folio_sub_large_mapcount folio_sub_return_large_mapcount
 #else /* !CONFIG_MM_ID */
 /*
  * See __folio_rmap_sanity_checks(), we might map large folios even without
@@ -350,17 +354,33 @@ static inline void folio_add_large_mapcount(struct folio *folio,
 	atomic_add(diff, &folio->_large_mapcount);
 }
 
+static inline int folio_add_return_large_mapcount(struct folio *folio,
+		int diff, struct vm_area_struct *vma)
+{
+	BUILD_BUG();
+}
+
 static inline void folio_sub_large_mapcount(struct folio *folio,
 		int diff, struct vm_area_struct *vma)
 {
 	atomic_sub(diff, &folio->_large_mapcount);
 }
+
+static inline int folio_sub_return_large_mapcount(struct folio *folio,
+		int diff, struct vm_area_struct *vma)
+{
+	BUILD_BUG();
+}
 #endif /* CONFIG_MM_ID */
 
 #define folio_inc_large_mapcount(folio, vma) \
 	folio_add_large_mapcount(folio, 1, vma)
+#define folio_inc_return_large_mapcount(folio, vma) \
+	folio_add_return_large_mapcount(folio, 1, vma)
 #define folio_dec_large_mapcount(folio, vma) \
 	folio_sub_large_mapcount(folio, 1, vma)
+#define folio_dec_return_large_mapcount(folio, vma) \
+	folio_sub_return_large_mapcount(folio, 1, vma)
 
 /* RMAP flags, currently only relevant for some anon rmap operations. */
 typedef int __bitwise rmap_t;
@@ -538,9 +558,11 @@ static __always_inline void __folio_dup_file_rmap(struct folio *folio,
 			break;
 		}
 
-		do {
-			atomic_inc(&page->_mapcount);
-		} while (page++, --nr_pages > 0);
+		if (IS_ENABLED(CONFIG_PAGE_MAPCOUNT)) {
+			do {
+				atomic_inc(&page->_mapcount);
+			} while (page++, --nr_pages > 0);
+		}
 		folio_add_large_mapcount(folio, orig_nr_pages, dst_vma);
 		break;
 	case RMAP_LEVEL_PMD:
@@ -638,7 +660,8 @@ static __always_inline int __folio_try_dup_anon_rmap(struct folio *folio,
 		do {
 			if (PageAnonExclusive(page))
 				ClearPageAnonExclusive(page);
-			atomic_inc(&page->_mapcount);
+			if (IS_ENABLED(CONFIG_PAGE_MAPCOUNT))
+				atomic_inc(&page->_mapcount);
 		} while (page++, --nr_pages > 0);
 		folio_add_large_mapcount(folio, orig_nr_pages, dst_vma);
 		break;
diff --git a/mm/internal.h b/mm/internal.h
index 7303ddd9dac04..7a4f81a6edd66 100644
--- a/mm/internal.h
+++ b/mm/internal.h
@@ -84,6 +84,8 @@ void page_writeback_init(void);
  */
 static inline int folio_nr_pages_mapped(const struct folio *folio)
 {
+	if (IS_ENABLED(CONFIG_NO_PAGE_MAPCOUNT))
+		return -1;
 	return atomic_read(&folio->_nr_pages_mapped) & FOLIO_PAGES_MAPPED;
 }
 
@@ -721,7 +723,8 @@ static inline void prep_compound_head(struct page *page, unsigned int order)
 
 	folio_set_order(folio, order);
 	atomic_set(&folio->_large_mapcount, -1);
-	atomic_set(&folio->_nr_pages_mapped, 0);
+	if (IS_ENABLED(CONFIG_PAGE_MAPCOUNT))
+		atomic_set(&folio->_nr_pages_mapped, 0);
 	if (IS_ENABLED(CONFIG_MM_ID)) {
 		folio->_mm_ids = 0;
 		folio->_mm_id_mapcount[0] = -1;
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 08caa92553998..4402672b5d838 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -951,7 +951,8 @@ static int free_tail_page_prepare(struct page *head_page, struct page *page)
 			bad_page(page, "nonzero large_mapcount");
 			goto out;
 		}
-		if (unlikely(atomic_read(&folio->_nr_pages_mapped))) {
+		if (IS_ENABLED(CONFIG_PAGE_MAPCOUNT) &&
+		    unlikely(atomic_read(&folio->_nr_pages_mapped))) {
 			bad_page(page, "nonzero nr_pages_mapped");
 			goto out;
 		}
diff --git a/mm/rmap.c b/mm/rmap.c
index 8de415157bc8d..67bb273dfb80d 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -1258,6 +1258,16 @@ static __always_inline unsigned int __folio_add_rmap(struct folio *folio,
 			break;
 		}
 
+		if (IS_ENABLED(CONFIG_NO_PAGE_MAPCOUNT)) {
+			nr = folio_add_return_large_mapcount(folio, orig_nr_pages, vma);
+			if (nr == orig_nr_pages)
+				/* Was completely unmapped. */
+				nr = folio_large_nr_pages(folio);
+			else
+				nr = 0;
+			break;
+		}
+
 		do {
 			first += atomic_inc_and_test(&page->_mapcount);
 		} while (page++, --nr_pages > 0);
@@ -1271,6 +1281,18 @@ static __always_inline unsigned int __folio_add_rmap(struct folio *folio,
 	case RMAP_LEVEL_PMD:
 	case RMAP_LEVEL_PUD:
 		first = atomic_inc_and_test(&folio->_entire_mapcount);
+		if (IS_ENABLED(CONFIG_NO_PAGE_MAPCOUNT)) {
+			if (level == RMAP_LEVEL_PMD && first)
+				*nr_pmdmapped = folio_large_nr_pages(folio);
+			nr = folio_inc_return_large_mapcount(folio, vma);
+			if (nr == 1)
+				/* Was completely unmapped. */
+				nr = folio_large_nr_pages(folio);
+			else
+				nr = 0;
+			break;
+		}
+
 		if (first) {
 			nr = atomic_add_return_relaxed(ENTIRELY_MAPPED, mapped);
 			if (likely(nr < ENTIRELY_MAPPED + ENTIRELY_MAPPED)) {
@@ -1436,13 +1458,23 @@ static __always_inline void __folio_add_anon_rmap(struct folio *folio,
 			break;
 		}
 	}
+
+	VM_WARN_ON_FOLIO(!folio_test_large(folio) && PageAnonExclusive(page) &&
+			 atomic_read(&folio->_mapcount) > 0, folio);
 	for (i = 0; i < nr_pages; i++) {
 		struct page *cur_page = page + i;
 
-		/* While PTE-mapping a THP we have a PMD and a PTE mapping. */
-		VM_WARN_ON_FOLIO((atomic_read(&cur_page->_mapcount) > 0 ||
-				  (folio_test_large(folio) &&
-				   folio_entire_mapcount(folio) > 1)) &&
+		VM_WARN_ON_FOLIO(folio_test_large(folio) &&
+				 folio_entire_mapcount(folio) > 1 &&
+				 PageAnonExclusive(cur_page), folio);
+		if (IS_ENABLED(CONFIG_NO_PAGE_MAPCOUNT))
+			continue;
+
+		/*
+		 * While PTE-mapping a THP we have a PMD and a PTE
+		 * mapping.
+		 */
+		VM_WARN_ON_FOLIO(atomic_read(&cur_page->_mapcount) > 0 &&
 				 PageAnonExclusive(cur_page), folio);
 	}
 
@@ -1548,20 +1580,23 @@ void folio_add_new_anon_rmap(struct folio *folio, struct vm_area_struct *vma,
 		for (i = 0; i < nr; i++) {
 			struct page *page = folio_page(folio, i);
 
-			/* increment count (starts at -1) */
-			atomic_set(&page->_mapcount, 0);
+			if (IS_ENABLED(CONFIG_PAGE_MAPCOUNT))
+				/* increment count (starts at -1) */
+				atomic_set(&page->_mapcount, 0);
 			if (exclusive)
 				SetPageAnonExclusive(page);
 		}
 
 		folio_set_large_mapcount(folio, nr, vma);
-		atomic_set(&folio->_nr_pages_mapped, nr);
+		if (IS_ENABLED(CONFIG_PAGE_MAPCOUNT))
+			atomic_set(&folio->_nr_pages_mapped, nr);
 	} else {
 		nr = folio_large_nr_pages(folio);
 		/* increment count (starts at -1) */
 		atomic_set(&folio->_entire_mapcount, 0);
 		folio_set_large_mapcount(folio, 1, vma);
-		atomic_set(&folio->_nr_pages_mapped, ENTIRELY_MAPPED);
+		if (IS_ENABLED(CONFIG_PAGE_MAPCOUNT))
+			atomic_set(&folio->_nr_pages_mapped, ENTIRELY_MAPPED);
 		if (exclusive)
 			SetPageAnonExclusive(&folio->page);
 		nr_pmdmapped = nr;
@@ -1665,6 +1700,19 @@ static __always_inline void __folio_remove_rmap(struct folio *folio,
 			break;
 		}
 
+		if (IS_ENABLED(CONFIG_NO_PAGE_MAPCOUNT)) {
+			nr = folio_sub_return_large_mapcount(folio, nr_pages, vma);
+			if (!nr) {
+				/* Now completely unmapped. */
+				nr = folio_nr_pages(folio);
+			} else {
+				partially_mapped = nr < folio_large_nr_pages(folio) &&
+						   !folio_entire_mapcount(folio);
+				nr = 0;
+			}
+			break;
+		}
+
 		folio_sub_large_mapcount(folio, nr_pages, vma);
 		do {
 			last += atomic_add_negative(-1, &page->_mapcount);
@@ -1678,6 +1726,22 @@ static __always_inline void __folio_remove_rmap(struct folio *folio,
 		break;
 	case RMAP_LEVEL_PMD:
 	case RMAP_LEVEL_PUD:
+		if (IS_ENABLED(CONFIG_NO_PAGE_MAPCOUNT)) {
+			last = atomic_add_negative(-1, &folio->_entire_mapcount);
+			if (level == RMAP_LEVEL_PMD && last)
+				nr_pmdmapped = folio_large_nr_pages(folio);
+			nr = folio_dec_return_large_mapcount(folio, vma);
+			if (!nr) {
+				/* Now completely unmapped. */
+				nr = folio_large_nr_pages(folio);
+			} else {
+				partially_mapped = last &&
+						   nr < folio_large_nr_pages(folio);
+				nr = 0;
+			}
+			break;
+		}
+
 		folio_dec_large_mapcount(folio, vma);
 		last = atomic_add_negative(-1, &folio->_entire_mapcount);
 		if (last) {
-- 
2.48.1

Re: [PATCH v2 16/20] fs/proc/page: remove per-page mapcount dependency for /proc/kpagecount (CONFIG_NO_PAGE_MAPCOUNT)

[Zi Yan]

On Mon Feb 24, 2025 at 11:55 AM EST, David Hildenbrand wrote:
> Let's implement an alternative when per-page mapcounts in large folios
> are no longer maintained -- soon with CONFIG_NO_PAGE_MAPCOUNT.
>
> For large folios, we'll return the per-page average mapcount within the
> folio, except when the average is 0 but the folio is mapped: then we
> return 1.
>
> For hugetlb folios and for large folios that are fully mapped
> into all address spaces, there is no change.
>
> As an alternative, we could simply return 0 for non-hugetlb large folios,
> or disable this legacy interface with CONFIG_NO_PAGE_MAPCOUNT.
>
> But the information exposed by this interface can still be valuable, and
> frequently we deal with fully-mapped large folios where the average
> corresponds to the actual page mapcount. So we'll leave it like this for
> now and document the new behavior.
>
> Note: this interface is likely not very relevant for performance. If
> ever required, we could try doing a rather expensive rmap walk to collect
> precisely how often this folio page is mapped.
>
> Signed-off-by: David Hildenbrand <david@redhat.com>
> ---
>  Documentation/admin-guide/mm/pagemap.rst |  7 +++++-
>  fs/proc/internal.h                       | 31 ++++++++++++++++++++++++
>  fs/proc/page.c                           | 19 ++++++++++++---
>  3 files changed, 53 insertions(+), 4 deletions(-)
>
> diff --git a/Documentation/admin-guide/mm/pagemap.rst b/Documentation/admin-guide/mm/pagemap.rst
> index caba0f52dd36c..49590306c61a0 100644
> --- a/Documentation/admin-guide/mm/pagemap.rst
> +++ b/Documentation/admin-guide/mm/pagemap.rst
> @@ -42,7 +42,12 @@ There are four components to pagemap:
>     skip over unmapped regions.
>  
>   * ``/proc/kpagecount``.  This file contains a 64-bit count of the number of
> -   times each page is mapped, indexed by PFN.
> +   times each page is mapped, indexed by PFN. Some kernel configurations do
> +   not track the precise number of times a page part of a larger allocation
> +   (e.g., THP) is mapped. In these configurations, the average number of
> +   mappings per page in this larger allocation is returned instead. However,
> +   if any page of the large allocation is mapped, the returned value will
> +   be at least 1.
>  
>  The page-types tool in the tools/mm directory can be used to query the
>  number of times a page is mapped.
> diff --git a/fs/proc/internal.h b/fs/proc/internal.h
> index 1695509370b88..16aa1fd260771 100644
> --- a/fs/proc/internal.h
> +++ b/fs/proc/internal.h
> @@ -174,6 +174,37 @@ static inline int folio_precise_page_mapcount(struct folio *folio,
>  	return mapcount;
>  }
>  
> +/**
> + * folio_average_page_mapcount() - Average number of mappings per page in this
> + *				   folio
> + * @folio: The folio.
> + *
> + * The average number of present user page table entries that reference each
> + * page in this folio as tracked via the RMAP: either referenced directly
> + * (PTE) or as part of a larger area that covers this page (e.g., PMD).
> + *
> + * Returns: The average number of mappings per page in this folio. 0 for
> + * folios that are not mapped to user space or are not tracked via the RMAP
> + * (e.g., shared zeropage).
> + */
> +static inline int folio_average_page_mapcount(struct folio *folio)
> +{
> +	int mapcount, entire_mapcount;
> +	unsigned int adjust;
> +
> +	if (!folio_test_large(folio))
> +		return atomic_read(&folio->_mapcount) + 1;
> +
> +	mapcount = folio_large_mapcount(folio);
> +	entire_mapcount = folio_entire_mapcount(folio);
> +	if (mapcount <= entire_mapcount)
> +		return entire_mapcount;
> +	mapcount -= entire_mapcount;
> +
> +	adjust = folio_large_nr_pages(folio) / 2;

Is there any reason for choosing this adjust number? A comment might be
helpful in case people want to change it later, either with some reasoning
or just saying it is chosen empirically.

> +	return ((mapcount + adjust) >> folio_large_order(folio)) +
> +		entire_mapcount;
> +}
>  /*
>   * array.c
>   */
> diff --git a/fs/proc/page.c b/fs/proc/page.c
> index a55f5acefa974..4d3290cc69667 100644
> --- a/fs/proc/page.c
> +++ b/fs/proc/page.c
> @@ -67,9 +67,22 @@ static ssize_t kpagecount_read(struct file *file, char __user *buf,
>  		 * memmaps that were actually initialized.
>  		 */
>  		page = pfn_to_online_page(pfn);
> -		if (page)
> -			mapcount = folio_precise_page_mapcount(page_folio(page),
> -							       page);
> +		if (page) {
> +			struct folio *folio = page_folio(page);
> +
> +			if (IS_ENABLED(CONFIG_PAGE_MAPCOUNT)) {
> +				mapcount = folio_precise_page_mapcount(folio, page);
> +			} else {
> +				/*
> +				 * Indicate the per-page average, but at least "1" for
> +				 * mapped folios.
> +				 */
> +				mapcount = folio_average_page_mapcount(folio);
> +				if (!mapcount && folio_test_large(folio) &&
> +				    folio_mapped(folio))
> +					mapcount = 1;

This should be part of folio_average_page_mapcount() right?
Otherwise, the comment on folio_average_page_mapcount() is not correct,
since it can return 0 when a folio is mapped to user space.

> +			}
> +		}
>  
>  		if (put_user(mapcount, out)) {
>  			ret = -EFAULT;




-- 
Best Regards,
Yan, Zi

Re: [PATCH v2 18/20] fs/proc/task_mmu: remove per-page mapcount dependency for "mapmax" (CONFIG_NO_PAGE_MAPCOUNT)

[Zi Yan]

On Mon Feb 24, 2025 at 11:56 AM EST, David Hildenbrand wrote:
> Let's implement an alternative when per-page mapcounts in large folios are
> no longer maintained -- soon with CONFIG_NO_PAGE_MAPCOUNT.
>
> For calculating "mapmax", we now use the average per-page mapcount in
> a large folio instead of the per-page mapcount.
>
> For hugetlb folios and folios that are not partially mapped into MMs,
> there is no change.
>
> Likely, this change will not matter much in practice, and an alternative
> might be to simple remove this stat with CONFIG_NO_PAGE_MAPCOUNT.
> However, there might be value to it, so let's keep it like that and
> document the behavior.
>
> Signed-off-by: David Hildenbrand <david@redhat.com>
> ---
>  Documentation/filesystems/proc.rst | 5 +++++
>  fs/proc/task_mmu.c                 | 7 ++++++-
>  2 files changed, 11 insertions(+), 1 deletion(-)
>
> diff --git a/Documentation/filesystems/proc.rst b/Documentation/filesystems/proc.rst
> index 09f0aed5a08ba..1aa190017f796 100644
> --- a/Documentation/filesystems/proc.rst
> +++ b/Documentation/filesystems/proc.rst
> @@ -686,6 +686,11 @@ Where:
>  node locality page counters (N0 == node0, N1 == node1, ...) and the kernel page
>  size, in KB, that is backing the mapping up.
>  
> +Note that some kernel configurations do not track the precise number of times
> +a page part of a larger allocation (e.g., THP) is mapped. In these
> +configurations, "mapmax" might corresponds to the average number of mappings
> +per page in such a larger allocation instead.
> +
>  1.2 Kernel data
>  ---------------
>  
> diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
> index 80839bbf9657f..d7ee842367f0f 100644
> --- a/fs/proc/task_mmu.c
> +++ b/fs/proc/task_mmu.c
> @@ -2862,7 +2862,12 @@ static void gather_stats(struct page *page, struct numa_maps *md, int pte_dirty,
>  			unsigned long nr_pages)
>  {
>  	struct folio *folio = page_folio(page);
> -	int count = folio_precise_page_mapcount(folio, page);
> +	int count;
> +
> +	if (IS_ENABLED(CONFIG_PAGE_MAPCOUNT))
> +		count = folio_precise_page_mapcount(folio, page);
> +	else
> +		count = min_t(int, folio_average_page_mapcount(folio), 1);

s/min/max ?

Otherwise, count is at most 1. Anyway, if you change
folio_average_page_mapcount() as I indicated in patch 16, this
will become count = folio_average_page_mapcount(folio).

>  
>  	md->pages += nr_pages;
>  	if (pte_dirty || folio_test_dirty(folio))




-- 
Best Regards,
Yan, Zi

Re: [PATCH v2 19/20] fs/proc/task_mmu: remove per-page mapcount dependency for smaps/smaps_rollup (CONFIG_NO_PAGE_MAPCOUNT)

[Zi Yan]

On Mon Feb 24, 2025 at 11:56 AM EST, David Hildenbrand wrote:
> Let's implement an alternative when per-page mapcounts in large folios are
> no longer maintained -- soon with CONFIG_NO_PAGE_MAPCOUNT.
>
> When computing the output for smaps / smaps_rollups, in particular when
> calculating the USS (Unique Set Size) and the PSS (Proportional Set Size),
> we still rely on per-page mapcounts.
>
> To determine private vs. shared, we'll use folio_likely_mapped_shared(),
> similar to how we handle PM_MMAP_EXCLUSIVE. Similarly, we might now
> under-estimate the USS and count pages towards "shared" that are
> actually "private" ("exclusively mapped").
>
> When calculating the PSS, we'll now also use the average per-page
> mapcount for large folios: this can result in both, an over-estimation
> and an under-estimation of the PSS. The difference is not expected to
> matter much in practice, but we'll have to learn as we go.
>
> We can now provide folio_precise_page_mapcount() only with
> CONFIG_PAGE_MAPCOUNT, and remove one of the last users of per-page
> mapcounts when CONFIG_NO_PAGE_MAPCOUNT is enabled.
>
> Document the new behavior.
>
> Signed-off-by: David Hildenbrand <david@redhat.com>
> ---
>  Documentation/filesystems/proc.rst | 13 +++++++++++++
>  fs/proc/internal.h                 |  8 ++++++++
>  fs/proc/task_mmu.c                 | 17 +++++++++++++++--
>  3 files changed, 36 insertions(+), 2 deletions(-)
>
> diff --git a/Documentation/filesystems/proc.rst b/Documentation/filesystems/proc.rst
> index 1aa190017f796..57d55274a1f42 100644
> --- a/Documentation/filesystems/proc.rst
> +++ b/Documentation/filesystems/proc.rst
> @@ -506,6 +506,19 @@ Note that even a page which is part of a MAP_SHARED mapping, but has only
>  a single pte mapped, i.e.  is currently used by only one process, is accounted
>  as private and not as shared.
>  
> +Note that in some kernel configurations, all pages part of a larger allocation
> +(e.g., THP) might be considered "shared" if the large allocation is
> +considered "shared": if not all pages are exclusive to the same process.
> +Further, some kernel configurations might consider larger allocations "shared",
> +if they were at one point considered "shared", even if they would now be
> +considered "exclusive".
> +
> +Some kernel configurations do not track the precise number of times a page part
> +of a larger allocation is mapped. In this case, when calculating the PSS, the
> +average number of mappings per page in this larger allocation might be used
> +as an approximation for the number of mappings of a page. The PSS calculation
> +will be imprecise in this case.
> +
>  "Referenced" indicates the amount of memory currently marked as referenced or
>  accessed.
>  
> diff --git a/fs/proc/internal.h b/fs/proc/internal.h
> index 16aa1fd260771..70205425a2daa 100644
> --- a/fs/proc/internal.h
> +++ b/fs/proc/internal.h
> @@ -143,6 +143,7 @@ unsigned name_to_int(const struct qstr *qstr);
>  /* Worst case buffer size needed for holding an integer. */
>  #define PROC_NUMBUF 13
>  
> +#ifdef CONFIG_PAGE_MAPCOUNT
>  /**
>   * folio_precise_page_mapcount() - Number of mappings of this folio page.
>   * @folio: The folio.
> @@ -173,6 +174,13 @@ static inline int folio_precise_page_mapcount(struct folio *folio,
>  
>  	return mapcount;
>  }
> +#else /* !CONFIG_PAGE_MAPCOUNT */
> +static inline int folio_precise_page_mapcount(struct folio *folio,
> +		struct page *page)
> +{
> +	BUILD_BUG();
> +}
> +#endif /* CONFIG_PAGE_MAPCOUNT */
>  
>  /**
>   * folio_average_page_mapcount() - Average number of mappings per page in this
> diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
> index d7ee842367f0f..7ca0bc3bf417d 100644
> --- a/fs/proc/task_mmu.c
> +++ b/fs/proc/task_mmu.c
> @@ -707,6 +707,8 @@ static void smaps_account(struct mem_size_stats *mss, struct page *page,
>  	struct folio *folio = page_folio(page);
>  	int i, nr = compound ? compound_nr(page) : 1;
>  	unsigned long size = nr * PAGE_SIZE;
> +	bool exclusive;
> +	int mapcount;
>  
>  	/*
>  	 * First accumulate quantities that depend only on |size| and the type
> @@ -747,18 +749,29 @@ static void smaps_account(struct mem_size_stats *mss, struct page *page,
>  				      dirty, locked, present);
>  		return;
>  	}
> +
> +	if (IS_ENABLED(CONFIG_NO_PAGE_MAPCOUNT)) {
> +		mapcount = folio_average_page_mapcount(folio);

This seems inconsistent with how folio_average_page_mapcount() is used
in patch 16 and 18.

> +		exclusive = !folio_maybe_mapped_shared(folio);
> +	}
> +
>  	/*
>  	 * We obtain a snapshot of the mapcount. Without holding the folio lock
>  	 * this snapshot can be slightly wrong as we cannot always read the
>  	 * mapcount atomically.
>  	 */
>  	for (i = 0; i < nr; i++, page++) {
> -		int mapcount = folio_precise_page_mapcount(folio, page);
>  		unsigned long pss = PAGE_SIZE << PSS_SHIFT;
> +
> +		if (IS_ENABLED(CONFIG_PAGE_MAPCOUNT)) {
> +			mapcount = folio_precise_page_mapcount(folio, page);
> +			exclusive = mapcount < 2;
> +		}
> +
>  		if (mapcount >= 2)
>  			pss /= mapcount;
>  		smaps_page_accumulate(mss, folio, PAGE_SIZE, pss,
> -				dirty, locked, mapcount < 2);
> +				dirty, locked, exclusive);
>  	}
>  }
>  




-- 
Best Regards,
Yan, Zi

Re: [PATCH v2 16/20] fs/proc/page: remove per-page mapcount dependency for /proc/kpagecount (CONFIG_NO_PAGE_MAPCOUNT)

[David Hildenbrand]

On 24.02.25 21:40, Zi Yan wrote:
> On Mon Feb 24, 2025 at 11:55 AM EST, David Hildenbrand wrote:
>> Let's implement an alternative when per-page mapcounts in large folios
>> are no longer maintained -- soon with CONFIG_NO_PAGE_MAPCOUNT.
>>
>> For large folios, we'll return the per-page average mapcount within the
>> folio, except when the average is 0 but the folio is mapped: then we
>> return 1.
>>
>> For hugetlb folios and for large folios that are fully mapped
>> into all address spaces, there is no change.
>>
>> As an alternative, we could simply return 0 for non-hugetlb large folios,
>> or disable this legacy interface with CONFIG_NO_PAGE_MAPCOUNT.
>>
>> But the information exposed by this interface can still be valuable, and
>> frequently we deal with fully-mapped large folios where the average
>> corresponds to the actual page mapcount. So we'll leave it like this for
>> now and document the new behavior.
>>
>> Note: this interface is likely not very relevant for performance. If
>> ever required, we could try doing a rather expensive rmap walk to collect
>> precisely how often this folio page is mapped.
>>
>> Signed-off-by: David Hildenbrand <david@redhat.com>
>> ---
>>   Documentation/admin-guide/mm/pagemap.rst |  7 +++++-
>>   fs/proc/internal.h                       | 31 ++++++++++++++++++++++++
>>   fs/proc/page.c                           | 19 ++++++++++++---
>>   3 files changed, 53 insertions(+), 4 deletions(-)
>>
>> diff --git a/Documentation/admin-guide/mm/pagemap.rst b/Documentation/admin-guide/mm/pagemap.rst
>> index caba0f52dd36c..49590306c61a0 100644
>> --- a/Documentation/admin-guide/mm/pagemap.rst
>> +++ b/Documentation/admin-guide/mm/pagemap.rst
>> @@ -42,7 +42,12 @@ There are four components to pagemap:
>>      skip over unmapped regions.
>>   
>>    * ``/proc/kpagecount``.  This file contains a 64-bit count of the number of
>> -   times each page is mapped, indexed by PFN.
>> +   times each page is mapped, indexed by PFN. Some kernel configurations do
>> +   not track the precise number of times a page part of a larger allocation
>> +   (e.g., THP) is mapped. In these configurations, the average number of
>> +   mappings per page in this larger allocation is returned instead. However,
>> +   if any page of the large allocation is mapped, the returned value will
>> +   be at least 1.
>>   
>>   The page-types tool in the tools/mm directory can be used to query the
>>   number of times a page is mapped.
>> diff --git a/fs/proc/internal.h b/fs/proc/internal.h
>> index 1695509370b88..16aa1fd260771 100644
>> --- a/fs/proc/internal.h
>> +++ b/fs/proc/internal.h
>> @@ -174,6 +174,37 @@ static inline int folio_precise_page_mapcount(struct folio *folio,
>>   	return mapcount;
>>   }
>>   
>> +/**
>> + * folio_average_page_mapcount() - Average number of mappings per page in this
>> + *				   folio
>> + * @folio: The folio.
>> + *
>> + * The average number of present user page table entries that reference each
>> + * page in this folio as tracked via the RMAP: either referenced directly
>> + * (PTE) or as part of a larger area that covers this page (e.g., PMD).
>> + *
>> + * Returns: The average number of mappings per page in this folio. 0 for
>> + * folios that are not mapped to user space or are not tracked via the RMAP
>> + * (e.g., shared zeropage).
>> + */
>> +static inline int folio_average_page_mapcount(struct folio *folio)
>> +{
>> +	int mapcount, entire_mapcount;
>> +	unsigned int adjust;
>> +
>> +	if (!folio_test_large(folio))
>> +		return atomic_read(&folio->_mapcount) + 1;
>> +
>> +	mapcount = folio_large_mapcount(folio);
>> +	entire_mapcount = folio_entire_mapcount(folio);
>> +	if (mapcount <= entire_mapcount)
>> +		return entire_mapcount;
>> +	mapcount -= entire_mapcount;
>> +
>> +	adjust = folio_large_nr_pages(folio) / 2;

Thanks for the review!

> 
> Is there any reason for choosing this adjust number? A comment might be
> helpful in case people want to change it later, either with some reasoning
> or just saying it is chosen empirically.

We're dividing by folio_large_nr_pages(folio) (shifting by 
folio_large_order(folio)), so this is not a magic number at all.

So this should be "ordinary" rounding.

Assume nr_pages = 512.

With 255 we want to round down, with 256 we want to round up.

255 / 512 = 0 :)
256 / 512 = 0 :(

Compared to:

(255 + (512 / 2)) / 512 = (255 + 256) / 512 = 0 :)
(256 + (512 / 2)) / 512 = (256 + 256) / 512 = 1 :)

> 
>> +	return ((mapcount + adjust) >> folio_large_order(folio)) +
>> +		entire_mapcount;
>> +}
>>   /*
>>    * array.c
>>    */
>> diff --git a/fs/proc/page.c b/fs/proc/page.c
>> index a55f5acefa974..4d3290cc69667 100644
>> --- a/fs/proc/page.c
>> +++ b/fs/proc/page.c
>> @@ -67,9 +67,22 @@ static ssize_t kpagecount_read(struct file *file, char __user *buf,
>>   		 * memmaps that were actually initialized.
>>   		 */
>>   		page = pfn_to_online_page(pfn);
>> -		if (page)
>> -			mapcount = folio_precise_page_mapcount(page_folio(page),
>> -							       page);
>> +		if (page) {
>> +			struct folio *folio = page_folio(page);
>> +
>> +			if (IS_ENABLED(CONFIG_PAGE_MAPCOUNT)) {
>> +				mapcount = folio_precise_page_mapcount(folio, page);
>> +			} else {
>> +				/*
>> +				 * Indicate the per-page average, but at least "1" for
>> +				 * mapped folios.
>> +				 */
>> +				mapcount = folio_average_page_mapcount(folio);
>> +				if (!mapcount && folio_test_large(folio) &&
>> +				    folio_mapped(folio))
>> +					mapcount = 1;
> 
> This should be part of folio_average_page_mapcount() right?

No, that's not desired.

> Otherwise, the comment on folio_average_page_mapcount() is not correct,
> since it can return 0 when a folio is mapped to user space.

It's misleading. I'll clarify the comment, probably simply saying:

Returns: The average number of mappings per page in this folio.

Thanks!

-- 
Cheers,

David / dhildenb

Re: [PATCH v2 18/20] fs/proc/task_mmu: remove per-page mapcount dependency for "mapmax" (CONFIG_NO_PAGE_MAPCOUNT)

[David Hildenbrand]

On 24.02.25 21:45, Zi Yan wrote:
> On Mon Feb 24, 2025 at 11:56 AM EST, David Hildenbrand wrote:
>> Let's implement an alternative when per-page mapcounts in large folios are
>> no longer maintained -- soon with CONFIG_NO_PAGE_MAPCOUNT.
>>
>> For calculating "mapmax", we now use the average per-page mapcount in
>> a large folio instead of the per-page mapcount.
>>
>> For hugetlb folios and folios that are not partially mapped into MMs,
>> there is no change.
>>
>> Likely, this change will not matter much in practice, and an alternative
>> might be to simple remove this stat with CONFIG_NO_PAGE_MAPCOUNT.
>> However, there might be value to it, so let's keep it like that and
>> document the behavior.
>>
>> Signed-off-by: David Hildenbrand <david@redhat.com>
>> ---
>>   Documentation/filesystems/proc.rst | 5 +++++
>>   fs/proc/task_mmu.c                 | 7 ++++++-
>>   2 files changed, 11 insertions(+), 1 deletion(-)
>>
>> diff --git a/Documentation/filesystems/proc.rst b/Documentation/filesystems/proc.rst
>> index 09f0aed5a08ba..1aa190017f796 100644
>> --- a/Documentation/filesystems/proc.rst
>> +++ b/Documentation/filesystems/proc.rst
>> @@ -686,6 +686,11 @@ Where:
>>   node locality page counters (N0 == node0, N1 == node1, ...) and the kernel page
>>   size, in KB, that is backing the mapping up.
>>   
>> +Note that some kernel configurations do not track the precise number of times
>> +a page part of a larger allocation (e.g., THP) is mapped. In these
>> +configurations, "mapmax" might corresponds to the average number of mappings
>> +per page in such a larger allocation instead.
>> +
>>   1.2 Kernel data
>>   ---------------
>>   
>> diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
>> index 80839bbf9657f..d7ee842367f0f 100644
>> --- a/fs/proc/task_mmu.c
>> +++ b/fs/proc/task_mmu.c
>> @@ -2862,7 +2862,12 @@ static void gather_stats(struct page *page, struct numa_maps *md, int pte_dirty,
>>   			unsigned long nr_pages)
>>   {
>>   	struct folio *folio = page_folio(page);
>> -	int count = folio_precise_page_mapcount(folio, page);
>> +	int count;
>> +
>> +	if (IS_ENABLED(CONFIG_PAGE_MAPCOUNT))
>> +		count = folio_precise_page_mapcount(folio, page);
>> +	else
>> +		count = min_t(int, folio_average_page_mapcount(folio), 1);
> 
> s/min/max ?

Indeed, thanks!

> 
> Otherwise, count is at most 1. Anyway, if you change
> folio_average_page_mapcount() as I indicated in patch 16, this
> will become count = folio_average_page_mapcount(folio).

No, the average should not be 1 just because a single subpage is mapped.

-- 
Cheers,

David / dhildenb

Re: [PATCH v2 16/20] fs/proc/page: remove per-page mapcount dependency for /proc/kpagecount (CONFIG_NO_PAGE_MAPCOUNT)

[David Hildenbrand]

On 24.02.25 22:02, David Hildenbrand wrote:
> On 24.02.25 21:40, Zi Yan wrote:
>> On Mon Feb 24, 2025 at 11:55 AM EST, David Hildenbrand wrote:
>>> Let's implement an alternative when per-page mapcounts in large folios
>>> are no longer maintained -- soon with CONFIG_NO_PAGE_MAPCOUNT.
>>>
>>> For large folios, we'll return the per-page average mapcount within the
>>> folio, except when the average is 0 but the folio is mapped: then we
>>> return 1.
>>>
>>> For hugetlb folios and for large folios that are fully mapped
>>> into all address spaces, there is no change.
>>>
>>> As an alternative, we could simply return 0 for non-hugetlb large folios,
>>> or disable this legacy interface with CONFIG_NO_PAGE_MAPCOUNT.
>>>
>>> But the information exposed by this interface can still be valuable, and
>>> frequently we deal with fully-mapped large folios where the average
>>> corresponds to the actual page mapcount. So we'll leave it like this for
>>> now and document the new behavior.
>>>
>>> Note: this interface is likely not very relevant for performance. If
>>> ever required, we could try doing a rather expensive rmap walk to collect
>>> precisely how often this folio page is mapped.
>>>
>>> Signed-off-by: David Hildenbrand <david@redhat.com>
>>> ---
>>>    Documentation/admin-guide/mm/pagemap.rst |  7 +++++-
>>>    fs/proc/internal.h                       | 31 ++++++++++++++++++++++++
>>>    fs/proc/page.c                           | 19 ++++++++++++---
>>>    3 files changed, 53 insertions(+), 4 deletions(-)
>>>
>>> diff --git a/Documentation/admin-guide/mm/pagemap.rst b/Documentation/admin-guide/mm/pagemap.rst
>>> index caba0f52dd36c..49590306c61a0 100644
>>> --- a/Documentation/admin-guide/mm/pagemap.rst
>>> +++ b/Documentation/admin-guide/mm/pagemap.rst
>>> @@ -42,7 +42,12 @@ There are four components to pagemap:
>>>       skip over unmapped regions.
>>>    
>>>     * ``/proc/kpagecount``.  This file contains a 64-bit count of the number of
>>> -   times each page is mapped, indexed by PFN.
>>> +   times each page is mapped, indexed by PFN. Some kernel configurations do
>>> +   not track the precise number of times a page part of a larger allocation
>>> +   (e.g., THP) is mapped. In these configurations, the average number of
>>> +   mappings per page in this larger allocation is returned instead. However,
>>> +   if any page of the large allocation is mapped, the returned value will
>>> +   be at least 1.
>>>    
>>>    The page-types tool in the tools/mm directory can be used to query the
>>>    number of times a page is mapped.
>>> diff --git a/fs/proc/internal.h b/fs/proc/internal.h
>>> index 1695509370b88..16aa1fd260771 100644
>>> --- a/fs/proc/internal.h
>>> +++ b/fs/proc/internal.h
>>> @@ -174,6 +174,37 @@ static inline int folio_precise_page_mapcount(struct folio *folio,
>>>    	return mapcount;
>>>    }
>>>    
>>> +/**
>>> + * folio_average_page_mapcount() - Average number of mappings per page in this
>>> + *				   folio
>>> + * @folio: The folio.
>>> + *
>>> + * The average number of present user page table entries that reference each
>>> + * page in this folio as tracked via the RMAP: either referenced directly
>>> + * (PTE) or as part of a larger area that covers this page (e.g., PMD).
>>> + *
>>> + * Returns: The average number of mappings per page in this folio. 0 for
>>> + * folios that are not mapped to user space or are not tracked via the RMAP
>>> + * (e.g., shared zeropage).
>>> + */
>>> +static inline int folio_average_page_mapcount(struct folio *folio)
>>> +{
>>> +	int mapcount, entire_mapcount;
>>> +	unsigned int adjust;
>>> +
>>> +	if (!folio_test_large(folio))
>>> +		return atomic_read(&folio->_mapcount) + 1;
>>> +
>>> +	mapcount = folio_large_mapcount(folio);
>>> +	entire_mapcount = folio_entire_mapcount(folio);
>>> +	if (mapcount <= entire_mapcount)
>>> +		return entire_mapcount;
>>> +	mapcount -= entire_mapcount;
>>> +
>>> +	adjust = folio_large_nr_pages(folio) / 2;
> 
> Thanks for the review!
> 
>>
>> Is there any reason for choosing this adjust number? A comment might be
>> helpful in case people want to change it later, either with some reasoning
>> or just saying it is chosen empirically.
> 
> We're dividing by folio_large_nr_pages(folio) (shifting by
> folio_large_order(folio)), so this is not a magic number at all.
> 
> So this should be "ordinary" rounding.
> 
> Assume nr_pages = 512.
> 
> With 255 we want to round down, with 256 we want to round up.
> 
> 255 / 512 = 0 :)
> 256 / 512 = 0 :(
> 
> Compared to:
> 
> (255 + (512 / 2)) / 512 = (255 + 256) / 512 = 0 :)
> (256 + (512 / 2)) / 512 = (256 + 256) / 512 = 1 :)

I think adding to the function doc:

"The average is calculated by rounding to the nearest integer."

might make it clearer.

-- 
Cheers,

David / dhildenb

Re: [PATCH v2 16/20] fs/proc/page: remove per-page mapcount dependency for /proc/kpagecount (CONFIG_NO_PAGE_MAPCOUNT)

[Zi Yan]

On 24 Feb 2025, at 16:02, David Hildenbrand wrote:

> On 24.02.25 21:40, Zi Yan wrote:
>> On Mon Feb 24, 2025 at 11:55 AM EST, David Hildenbrand wrote:
>>> Let's implement an alternative when per-page mapcounts in large folios
>>> are no longer maintained -- soon with CONFIG_NO_PAGE_MAPCOUNT.
>>>
>>> For large folios, we'll return the per-page average mapcount within the
>>> folio, except when the average is 0 but the folio is mapped: then we
>>> return 1.
>>>
>>> For hugetlb folios and for large folios that are fully mapped
>>> into all address spaces, there is no change.
>>>
>>> As an alternative, we could simply return 0 for non-hugetlb large folios,
>>> or disable this legacy interface with CONFIG_NO_PAGE_MAPCOUNT.
>>>
>>> But the information exposed by this interface can still be valuable, and
>>> frequently we deal with fully-mapped large folios where the average
>>> corresponds to the actual page mapcount. So we'll leave it like this for
>>> now and document the new behavior.
>>>
>>> Note: this interface is likely not very relevant for performance. If
>>> ever required, we could try doing a rather expensive rmap walk to collect
>>> precisely how often this folio page is mapped.
>>>
>>> Signed-off-by: David Hildenbrand <david@redhat.com>
>>> ---
>>>   Documentation/admin-guide/mm/pagemap.rst |  7 +++++-
>>>   fs/proc/internal.h                       | 31 ++++++++++++++++++++++++
>>>   fs/proc/page.c                           | 19 ++++++++++++---
>>>   3 files changed, 53 insertions(+), 4 deletions(-)
>>>
>>> diff --git a/Documentation/admin-guide/mm/pagemap.rst b/Documentation/admin-guide/mm/pagemap.rst
>>> index caba0f52dd36c..49590306c61a0 100644
>>> --- a/Documentation/admin-guide/mm/pagemap.rst
>>> +++ b/Documentation/admin-guide/mm/pagemap.rst
>>> @@ -42,7 +42,12 @@ There are four components to pagemap:
>>>      skip over unmapped regions.
>>>     * ``/proc/kpagecount``.  This file contains a 64-bit count of the number of
>>> -   times each page is mapped, indexed by PFN.
>>> +   times each page is mapped, indexed by PFN. Some kernel configurations do
>>> +   not track the precise number of times a page part of a larger allocation
>>> +   (e.g., THP) is mapped. In these configurations, the average number of
>>> +   mappings per page in this larger allocation is returned instead. However,
>>> +   if any page of the large allocation is mapped, the returned value will
>>> +   be at least 1.
>>>    The page-types tool in the tools/mm directory can be used to query the
>>>   number of times a page is mapped.
>>> diff --git a/fs/proc/internal.h b/fs/proc/internal.h
>>> index 1695509370b88..16aa1fd260771 100644
>>> --- a/fs/proc/internal.h
>>> +++ b/fs/proc/internal.h
>>> @@ -174,6 +174,37 @@ static inline int folio_precise_page_mapcount(struct folio *folio,
>>>   	return mapcount;
>>>   }
>>>  +/**
>>> + * folio_average_page_mapcount() - Average number of mappings per page in this
>>> + *				   folio
>>> + * @folio: The folio.
>>> + *
>>> + * The average number of present user page table entries that reference each
>>> + * page in this folio as tracked via the RMAP: either referenced directly
>>> + * (PTE) or as part of a larger area that covers this page (e.g., PMD).
>>> + *
>>> + * Returns: The average number of mappings per page in this folio. 0 for
>>> + * folios that are not mapped to user space or are not tracked via the RMAP
>>> + * (e.g., shared zeropage).
>>> + */
>>> +static inline int folio_average_page_mapcount(struct folio *folio)
>>> +{
>>> +	int mapcount, entire_mapcount;
>>> +	unsigned int adjust;
>>> +
>>> +	if (!folio_test_large(folio))
>>> +		return atomic_read(&folio->_mapcount) + 1;
>>> +
>>> +	mapcount = folio_large_mapcount(folio);
>>> +	entire_mapcount = folio_entire_mapcount(folio);
>>> +	if (mapcount <= entire_mapcount)
>>> +		return entire_mapcount;
>>> +	mapcount -= entire_mapcount;
>>> +
>>> +	adjust = folio_large_nr_pages(folio) / 2;
>
> Thanks for the review!
>
>>
>> Is there any reason for choosing this adjust number? A comment might be
>> helpful in case people want to change it later, either with some reasoning
>> or just saying it is chosen empirically.
>
> We're dividing by folio_large_nr_pages(folio) (shifting by folio_large_order(folio)), so this is not a magic number at all.
>
> So this should be "ordinary" rounding.

I thought the rounding would be (mapcount + 511) / 512. But
that means if one subpage is mapped, the average will be 1.
Your rounding means if at least half of the subpages is mapped,
the average will be 1. Others might think 1/3 is mapped,
the average will be 1. That is why I think adjust looks like
a magic number.

>
> Assume nr_pages = 512.
>
> With 255 we want to round down, with 256 we want to round up.
>
> 255 / 512 = 0 :)
> 256 / 512 = 0 :(
>
> Compared to:
>
> (255 + (512 / 2)) / 512 = (255 + 256) / 512 = 0 :)
> (256 + (512 / 2)) / 512 = (256 + 256) / 512 = 1 :)
>
>>
>>> +	return ((mapcount + adjust) >> folio_large_order(folio)) +
>>> +		entire_mapcount;
>>> +}
>>>   /*
>>>    * array.c
>>>    */
>>> diff --git a/fs/proc/page.c b/fs/proc/page.c
>>> index a55f5acefa974..4d3290cc69667 100644
>>> --- a/fs/proc/page.c
>>> +++ b/fs/proc/page.c
>>> @@ -67,9 +67,22 @@ static ssize_t kpagecount_read(struct file *file, char __user *buf,
>>>   		 * memmaps that were actually initialized.
>>>   		 */
>>>   		page = pfn_to_online_page(pfn);
>>> -		if (page)
>>> -			mapcount = folio_precise_page_mapcount(page_folio(page),
>>> -							       page);
>>> +		if (page) {
>>> +			struct folio *folio = page_folio(page);
>>> +
>>> +			if (IS_ENABLED(CONFIG_PAGE_MAPCOUNT)) {
>>> +				mapcount = folio_precise_page_mapcount(folio, page);
>>> +			} else {
>>> +				/*
>>> +				 * Indicate the per-page average, but at least "1" for
>>> +				 * mapped folios.
>>> +				 */
>>> +				mapcount = folio_average_page_mapcount(folio);
>>> +				if (!mapcount && folio_test_large(folio) &&
>>> +				    folio_mapped(folio))
>>> +					mapcount = 1;
>>
>> This should be part of folio_average_page_mapcount() right?
>
> No, that's not desired.
>
>> Otherwise, the comment on folio_average_page_mapcount() is not correct,
>> since it can return 0 when a folio is mapped to user space.
>
> It's misleading. I'll clarify the comment, probably simply saying:
>
> Returns: The average number of mappings per page in this folio.

Got it.

Best Regards,
Yan, Zi

Re: [PATCH v2 16/20] fs/proc/page: remove per-page mapcount dependency for /proc/kpagecount (CONFIG_NO_PAGE_MAPCOUNT)

[David Hildenbrand]

On 24.02.25 22:10, Zi Yan wrote:
> On 24 Feb 2025, at 16:02, David Hildenbrand wrote:
> 
>> On 24.02.25 21:40, Zi Yan wrote:
>>> On Mon Feb 24, 2025 at 11:55 AM EST, David Hildenbrand wrote:
>>>> Let's implement an alternative when per-page mapcounts in large folios
>>>> are no longer maintained -- soon with CONFIG_NO_PAGE_MAPCOUNT.
>>>>
>>>> For large folios, we'll return the per-page average mapcount within the
>>>> folio, except when the average is 0 but the folio is mapped: then we
>>>> return 1.
>>>>
>>>> For hugetlb folios and for large folios that are fully mapped
>>>> into all address spaces, there is no change.
>>>>
>>>> As an alternative, we could simply return 0 for non-hugetlb large folios,
>>>> or disable this legacy interface with CONFIG_NO_PAGE_MAPCOUNT.
>>>>
>>>> But the information exposed by this interface can still be valuable, and
>>>> frequently we deal with fully-mapped large folios where the average
>>>> corresponds to the actual page mapcount. So we'll leave it like this for
>>>> now and document the new behavior.
>>>>
>>>> Note: this interface is likely not very relevant for performance. If
>>>> ever required, we could try doing a rather expensive rmap walk to collect
>>>> precisely how often this folio page is mapped.
>>>>
>>>> Signed-off-by: David Hildenbrand <david@redhat.com>
>>>> ---
>>>>    Documentation/admin-guide/mm/pagemap.rst |  7 +++++-
>>>>    fs/proc/internal.h                       | 31 ++++++++++++++++++++++++
>>>>    fs/proc/page.c                           | 19 ++++++++++++---
>>>>    3 files changed, 53 insertions(+), 4 deletions(-)
>>>>
>>>> diff --git a/Documentation/admin-guide/mm/pagemap.rst b/Documentation/admin-guide/mm/pagemap.rst
>>>> index caba0f52dd36c..49590306c61a0 100644
>>>> --- a/Documentation/admin-guide/mm/pagemap.rst
>>>> +++ b/Documentation/admin-guide/mm/pagemap.rst
>>>> @@ -42,7 +42,12 @@ There are four components to pagemap:
>>>>       skip over unmapped regions.
>>>>      * ``/proc/kpagecount``.  This file contains a 64-bit count of the number of
>>>> -   times each page is mapped, indexed by PFN.
>>>> +   times each page is mapped, indexed by PFN. Some kernel configurations do
>>>> +   not track the precise number of times a page part of a larger allocation
>>>> +   (e.g., THP) is mapped. In these configurations, the average number of
>>>> +   mappings per page in this larger allocation is returned instead. However,
>>>> +   if any page of the large allocation is mapped, the returned value will
>>>> +   be at least 1.
>>>>     The page-types tool in the tools/mm directory can be used to query the
>>>>    number of times a page is mapped.
>>>> diff --git a/fs/proc/internal.h b/fs/proc/internal.h
>>>> index 1695509370b88..16aa1fd260771 100644
>>>> --- a/fs/proc/internal.h
>>>> +++ b/fs/proc/internal.h
>>>> @@ -174,6 +174,37 @@ static inline int folio_precise_page_mapcount(struct folio *folio,
>>>>    	return mapcount;
>>>>    }
>>>>   +/**
>>>> + * folio_average_page_mapcount() - Average number of mappings per page in this
>>>> + *				   folio
>>>> + * @folio: The folio.
>>>> + *
>>>> + * The average number of present user page table entries that reference each
>>>> + * page in this folio as tracked via the RMAP: either referenced directly
>>>> + * (PTE) or as part of a larger area that covers this page (e.g., PMD).
>>>> + *
>>>> + * Returns: The average number of mappings per page in this folio. 0 for
>>>> + * folios that are not mapped to user space or are not tracked via the RMAP
>>>> + * (e.g., shared zeropage).
>>>> + */
>>>> +static inline int folio_average_page_mapcount(struct folio *folio)
>>>> +{
>>>> +	int mapcount, entire_mapcount;
>>>> +	unsigned int adjust;
>>>> +
>>>> +	if (!folio_test_large(folio))
>>>> +		return atomic_read(&folio->_mapcount) + 1;
>>>> +
>>>> +	mapcount = folio_large_mapcount(folio);
>>>> +	entire_mapcount = folio_entire_mapcount(folio);
>>>> +	if (mapcount <= entire_mapcount)
>>>> +		return entire_mapcount;
>>>> +	mapcount -= entire_mapcount;
>>>> +
>>>> +	adjust = folio_large_nr_pages(folio) / 2;
>>
>> Thanks for the review!
>>
>>>
>>> Is there any reason for choosing this adjust number? A comment might be
>>> helpful in case people want to change it later, either with some reasoning
>>> or just saying it is chosen empirically.
>>
>> We're dividing by folio_large_nr_pages(folio) (shifting by folio_large_order(folio)), so this is not a magic number at all.
>>
>> So this should be "ordinary" rounding.
> 
> I thought the rounding would be (mapcount + 511) / 512.

Yes, that's "rounding up".

> But
> that means if one subpage is mapped, the average will be 1.
> Your rounding means if at least half of the subpages is mapped,
> the average will be 1. Others might think 1/3 is mapped,
> the average will be 1. That is why I think adjust looks like
> a magic number.

I think all callers could tolerate (or benefit) from 
folio_average_page_mapcount() returning at least 1 in case any page is 
mapped.

There was a reason why I decided to round to the nearest integer instead.

Let me think about this once more, I went back and forth a couple of 
times on this.

-- 
Cheers,

David / dhildenb

Re: [PATCH v2 19/20] fs/proc/task_mmu: remove per-page mapcount dependency for smaps/smaps_rollup (CONFIG_NO_PAGE_MAPCOUNT)

[David Hildenbrand]

On 24.02.25 21:53, Zi Yan wrote:
> On Mon Feb 24, 2025 at 11:56 AM EST, David Hildenbrand wrote:
>> Let's implement an alternative when per-page mapcounts in large folios are
>> no longer maintained -- soon with CONFIG_NO_PAGE_MAPCOUNT.
>>
>> When computing the output for smaps / smaps_rollups, in particular when
>> calculating the USS (Unique Set Size) and the PSS (Proportional Set Size),
>> we still rely on per-page mapcounts.
>>
>> To determine private vs. shared, we'll use folio_likely_mapped_shared(),
>> similar to how we handle PM_MMAP_EXCLUSIVE. Similarly, we might now
>> under-estimate the USS and count pages towards "shared" that are
>> actually "private" ("exclusively mapped").
>>
>> When calculating the PSS, we'll now also use the average per-page
>> mapcount for large folios: this can result in both, an over-estimation
>> and an under-estimation of the PSS. The difference is not expected to
>> matter much in practice, but we'll have to learn as we go.
>>
>> We can now provide folio_precise_page_mapcount() only with
>> CONFIG_PAGE_MAPCOUNT, and remove one of the last users of per-page
>> mapcounts when CONFIG_NO_PAGE_MAPCOUNT is enabled.
>>
>> Document the new behavior.
>>
>> Signed-off-by: David Hildenbrand <david@redhat.com>
>> ---
>>   Documentation/filesystems/proc.rst | 13 +++++++++++++
>>   fs/proc/internal.h                 |  8 ++++++++
>>   fs/proc/task_mmu.c                 | 17 +++++++++++++++--
>>   3 files changed, 36 insertions(+), 2 deletions(-)
>>
>> diff --git a/Documentation/filesystems/proc.rst b/Documentation/filesystems/proc.rst
>> index 1aa190017f796..57d55274a1f42 100644
>> --- a/Documentation/filesystems/proc.rst
>> +++ b/Documentation/filesystems/proc.rst
>> @@ -506,6 +506,19 @@ Note that even a page which is part of a MAP_SHARED mapping, but has only
>>   a single pte mapped, i.e.  is currently used by only one process, is accounted
>>   as private and not as shared.
>>   
>> +Note that in some kernel configurations, all pages part of a larger allocation
>> +(e.g., THP) might be considered "shared" if the large allocation is
>> +considered "shared": if not all pages are exclusive to the same process.
>> +Further, some kernel configurations might consider larger allocations "shared",
>> +if they were at one point considered "shared", even if they would now be
>> +considered "exclusive".
>> +
>> +Some kernel configurations do not track the precise number of times a page part
>> +of a larger allocation is mapped. In this case, when calculating the PSS, the
>> +average number of mappings per page in this larger allocation might be used
>> +as an approximation for the number of mappings of a page. The PSS calculation
>> +will be imprecise in this case.
>> +
>>   "Referenced" indicates the amount of memory currently marked as referenced or
>>   accessed.
>>   
>> diff --git a/fs/proc/internal.h b/fs/proc/internal.h
>> index 16aa1fd260771..70205425a2daa 100644
>> --- a/fs/proc/internal.h
>> +++ b/fs/proc/internal.h
>> @@ -143,6 +143,7 @@ unsigned name_to_int(const struct qstr *qstr);
>>   /* Worst case buffer size needed for holding an integer. */
>>   #define PROC_NUMBUF 13
>>   
>> +#ifdef CONFIG_PAGE_MAPCOUNT
>>   /**
>>    * folio_precise_page_mapcount() - Number of mappings of this folio page.
>>    * @folio: The folio.
>> @@ -173,6 +174,13 @@ static inline int folio_precise_page_mapcount(struct folio *folio,
>>   
>>   	return mapcount;
>>   }
>> +#else /* !CONFIG_PAGE_MAPCOUNT */
>> +static inline int folio_precise_page_mapcount(struct folio *folio,
>> +		struct page *page)
>> +{
>> +	BUILD_BUG();
>> +}
>> +#endif /* CONFIG_PAGE_MAPCOUNT */
>>   
>>   /**
>>    * folio_average_page_mapcount() - Average number of mappings per page in this
>> diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
>> index d7ee842367f0f..7ca0bc3bf417d 100644
>> --- a/fs/proc/task_mmu.c
>> +++ b/fs/proc/task_mmu.c
>> @@ -707,6 +707,8 @@ static void smaps_account(struct mem_size_stats *mss, struct page *page,
>>   	struct folio *folio = page_folio(page);
>>   	int i, nr = compound ? compound_nr(page) : 1;
>>   	unsigned long size = nr * PAGE_SIZE;
>> +	bool exclusive;
>> +	int mapcount;
>>   
>>   	/*
>>   	 * First accumulate quantities that depend only on |size| and the type
>> @@ -747,18 +749,29 @@ static void smaps_account(struct mem_size_stats *mss, struct page *page,
>>   				      dirty, locked, present);
>>   		return;
>>   	}
>> +
>> +	if (IS_ENABLED(CONFIG_NO_PAGE_MAPCOUNT)) {
>> +		mapcount = folio_average_page_mapcount(folio);
> 
> This seems inconsistent with how folio_average_page_mapcount() is used
> in patch 16 and 18.

It only looks that way. Note that the code below only does

>>   		if (mapcount >= 2)
>>   			pss /= mapcount;

Having it as 0 or 1 doesn't matter in that regard.

Thanks!

-- 
Cheers,

David / dhildenb

Re: [PATCH v2 16/20] fs/proc/page: remove per-page mapcount dependency for /proc/kpagecount (CONFIG_NO_PAGE_MAPCOUNT)

[Zi Yan]

On 24 Feb 2025, at 16:15, David Hildenbrand wrote:

> On 24.02.25 22:10, Zi Yan wrote:
>> On 24 Feb 2025, at 16:02, David Hildenbrand wrote:
>>
>>> On 24.02.25 21:40, Zi Yan wrote:
>>>> On Mon Feb 24, 2025 at 11:55 AM EST, David Hildenbrand wrote:
>>>>> Let's implement an alternative when per-page mapcounts in large folios
>>>>> are no longer maintained -- soon with CONFIG_NO_PAGE_MAPCOUNT.
>>>>>
>>>>> For large folios, we'll return the per-page average mapcount within the
>>>>> folio, except when the average is 0 but the folio is mapped: then we
>>>>> return 1.
>>>>>
>>>>> For hugetlb folios and for large folios that are fully mapped
>>>>> into all address spaces, there is no change.
>>>>>
>>>>> As an alternative, we could simply return 0 for non-hugetlb large folios,
>>>>> or disable this legacy interface with CONFIG_NO_PAGE_MAPCOUNT.
>>>>>
>>>>> But the information exposed by this interface can still be valuable, and
>>>>> frequently we deal with fully-mapped large folios where the average
>>>>> corresponds to the actual page mapcount. So we'll leave it like this for
>>>>> now and document the new behavior.
>>>>>
>>>>> Note: this interface is likely not very relevant for performance. If
>>>>> ever required, we could try doing a rather expensive rmap walk to collect
>>>>> precisely how often this folio page is mapped.
>>>>>
>>>>> Signed-off-by: David Hildenbrand <david@redhat.com>
>>>>> ---
>>>>>    Documentation/admin-guide/mm/pagemap.rst |  7 +++++-
>>>>>    fs/proc/internal.h                       | 31 ++++++++++++++++++++++++
>>>>>    fs/proc/page.c                           | 19 ++++++++++++---
>>>>>    3 files changed, 53 insertions(+), 4 deletions(-)
>>>>>
>>>>> diff --git a/Documentation/admin-guide/mm/pagemap.rst b/Documentation/admin-guide/mm/pagemap.rst
>>>>> index caba0f52dd36c..49590306c61a0 100644
>>>>> --- a/Documentation/admin-guide/mm/pagemap.rst
>>>>> +++ b/Documentation/admin-guide/mm/pagemap.rst
>>>>> @@ -42,7 +42,12 @@ There are four components to pagemap:
>>>>>       skip over unmapped regions.
>>>>>      * ``/proc/kpagecount``.  This file contains a 64-bit count of the number of
>>>>> -   times each page is mapped, indexed by PFN.
>>>>> +   times each page is mapped, indexed by PFN. Some kernel configurations do
>>>>> +   not track the precise number of times a page part of a larger allocation
>>>>> +   (e.g., THP) is mapped. In these configurations, the average number of
>>>>> +   mappings per page in this larger allocation is returned instead. However,
>>>>> +   if any page of the large allocation is mapped, the returned value will
>>>>> +   be at least 1.
>>>>>     The page-types tool in the tools/mm directory can be used to query the
>>>>>    number of times a page is mapped.
>>>>> diff --git a/fs/proc/internal.h b/fs/proc/internal.h
>>>>> index 1695509370b88..16aa1fd260771 100644
>>>>> --- a/fs/proc/internal.h
>>>>> +++ b/fs/proc/internal.h
>>>>> @@ -174,6 +174,37 @@ static inline int folio_precise_page_mapcount(struct folio *folio,
>>>>>    	return mapcount;
>>>>>    }
>>>>>   +/**
>>>>> + * folio_average_page_mapcount() - Average number of mappings per page in this
>>>>> + *				   folio
>>>>> + * @folio: The folio.
>>>>> + *
>>>>> + * The average number of present user page table entries that reference each
>>>>> + * page in this folio as tracked via the RMAP: either referenced directly
>>>>> + * (PTE) or as part of a larger area that covers this page (e.g., PMD).
>>>>> + *
>>>>> + * Returns: The average number of mappings per page in this folio. 0 for
>>>>> + * folios that are not mapped to user space or are not tracked via the RMAP
>>>>> + * (e.g., shared zeropage).
>>>>> + */
>>>>> +static inline int folio_average_page_mapcount(struct folio *folio)
>>>>> +{
>>>>> +	int mapcount, entire_mapcount;
>>>>> +	unsigned int adjust;
>>>>> +
>>>>> +	if (!folio_test_large(folio))
>>>>> +		return atomic_read(&folio->_mapcount) + 1;
>>>>> +
>>>>> +	mapcount = folio_large_mapcount(folio);
>>>>> +	entire_mapcount = folio_entire_mapcount(folio);
>>>>> +	if (mapcount <= entire_mapcount)
>>>>> +		return entire_mapcount;
>>>>> +	mapcount -= entire_mapcount;
>>>>> +
>>>>> +	adjust = folio_large_nr_pages(folio) / 2;
>>>
>>> Thanks for the review!
>>>
>>>>
>>>> Is there any reason for choosing this adjust number? A comment might be
>>>> helpful in case people want to change it later, either with some reasoning
>>>> or just saying it is chosen empirically.
>>>
>>> We're dividing by folio_large_nr_pages(folio) (shifting by folio_large_order(folio)), so this is not a magic number at all.
>>>
>>> So this should be "ordinary" rounding.
>>
>> I thought the rounding would be (mapcount + 511) / 512.
>
> Yes, that's "rounding up".
>
>> But
>> that means if one subpage is mapped, the average will be 1.
>> Your rounding means if at least half of the subpages is mapped,
>> the average will be 1. Others might think 1/3 is mapped,
>> the average will be 1. That is why I think adjust looks like
>> a magic number.
>
> I think all callers could tolerate (or benefit) from folio_average_page_mapcount() returning at least 1 in case any page is mapped.
>
> There was a reason why I decided to round to the nearest integer instead.
>
> Let me think about this once more, I went back and forth a couple of times on this.

Sure. Your current choice might be good enough for now. My intend of
adding a comment here is just to let people know the adjust can be
changed in the future. :)


Best Regards,
Yan, Zi

Re: [PATCH v2 16/20] fs/proc/page: remove per-page mapcount dependency for /proc/kpagecount (CONFIG_NO_PAGE_MAPCOUNT)

[David Hildenbrand]

On 24.02.25 22:23, Zi Yan wrote:
> On 24 Feb 2025, at 16:15, David Hildenbrand wrote:
> 
>> On 24.02.25 22:10, Zi Yan wrote:
>>> On 24 Feb 2025, at 16:02, David Hildenbrand wrote:
>>>
>>>> On 24.02.25 21:40, Zi Yan wrote:
>>>>> On Mon Feb 24, 2025 at 11:55 AM EST, David Hildenbrand wrote:
>>>>>> Let's implement an alternative when per-page mapcounts in large folios
>>>>>> are no longer maintained -- soon with CONFIG_NO_PAGE_MAPCOUNT.
>>>>>>
>>>>>> For large folios, we'll return the per-page average mapcount within the
>>>>>> folio, except when the average is 0 but the folio is mapped: then we
>>>>>> return 1.
>>>>>>
>>>>>> For hugetlb folios and for large folios that are fully mapped
>>>>>> into all address spaces, there is no change.
>>>>>>
>>>>>> As an alternative, we could simply return 0 for non-hugetlb large folios,
>>>>>> or disable this legacy interface with CONFIG_NO_PAGE_MAPCOUNT.
>>>>>>
>>>>>> But the information exposed by this interface can still be valuable, and
>>>>>> frequently we deal with fully-mapped large folios where the average
>>>>>> corresponds to the actual page mapcount. So we'll leave it like this for
>>>>>> now and document the new behavior.
>>>>>>
>>>>>> Note: this interface is likely not very relevant for performance. If
>>>>>> ever required, we could try doing a rather expensive rmap walk to collect
>>>>>> precisely how often this folio page is mapped.
>>>>>>
>>>>>> Signed-off-by: David Hildenbrand <david@redhat.com>
>>>>>> ---
>>>>>>     Documentation/admin-guide/mm/pagemap.rst |  7 +++++-
>>>>>>     fs/proc/internal.h                       | 31 ++++++++++++++++++++++++
>>>>>>     fs/proc/page.c                           | 19 ++++++++++++---
>>>>>>     3 files changed, 53 insertions(+), 4 deletions(-)
>>>>>>
>>>>>> diff --git a/Documentation/admin-guide/mm/pagemap.rst b/Documentation/admin-guide/mm/pagemap.rst
>>>>>> index caba0f52dd36c..49590306c61a0 100644
>>>>>> --- a/Documentation/admin-guide/mm/pagemap.rst
>>>>>> +++ b/Documentation/admin-guide/mm/pagemap.rst
>>>>>> @@ -42,7 +42,12 @@ There are four components to pagemap:
>>>>>>        skip over unmapped regions.
>>>>>>       * ``/proc/kpagecount``.  This file contains a 64-bit count of the number of
>>>>>> -   times each page is mapped, indexed by PFN.
>>>>>> +   times each page is mapped, indexed by PFN. Some kernel configurations do
>>>>>> +   not track the precise number of times a page part of a larger allocation
>>>>>> +   (e.g., THP) is mapped. In these configurations, the average number of
>>>>>> +   mappings per page in this larger allocation is returned instead. However,
>>>>>> +   if any page of the large allocation is mapped, the returned value will
>>>>>> +   be at least 1.
>>>>>>      The page-types tool in the tools/mm directory can be used to query the
>>>>>>     number of times a page is mapped.
>>>>>> diff --git a/fs/proc/internal.h b/fs/proc/internal.h
>>>>>> index 1695509370b88..16aa1fd260771 100644
>>>>>> --- a/fs/proc/internal.h
>>>>>> +++ b/fs/proc/internal.h
>>>>>> @@ -174,6 +174,37 @@ static inline int folio_precise_page_mapcount(struct folio *folio,
>>>>>>     	return mapcount;
>>>>>>     }
>>>>>>    +/**
>>>>>> + * folio_average_page_mapcount() - Average number of mappings per page in this
>>>>>> + *				   folio
>>>>>> + * @folio: The folio.
>>>>>> + *
>>>>>> + * The average number of present user page table entries that reference each
>>>>>> + * page in this folio as tracked via the RMAP: either referenced directly
>>>>>> + * (PTE) or as part of a larger area that covers this page (e.g., PMD).
>>>>>> + *
>>>>>> + * Returns: The average number of mappings per page in this folio. 0 for
>>>>>> + * folios that are not mapped to user space or are not tracked via the RMAP
>>>>>> + * (e.g., shared zeropage).
>>>>>> + */
>>>>>> +static inline int folio_average_page_mapcount(struct folio *folio)
>>>>>> +{
>>>>>> +	int mapcount, entire_mapcount;
>>>>>> +	unsigned int adjust;
>>>>>> +
>>>>>> +	if (!folio_test_large(folio))
>>>>>> +		return atomic_read(&folio->_mapcount) + 1;
>>>>>> +
>>>>>> +	mapcount = folio_large_mapcount(folio);
>>>>>> +	entire_mapcount = folio_entire_mapcount(folio);
>>>>>> +	if (mapcount <= entire_mapcount)
>>>>>> +		return entire_mapcount;
>>>>>> +	mapcount -= entire_mapcount;
>>>>>> +
>>>>>> +	adjust = folio_large_nr_pages(folio) / 2;
>>>>
>>>> Thanks for the review!
>>>>
>>>>>
>>>>> Is there any reason for choosing this adjust number? A comment might be
>>>>> helpful in case people want to change it later, either with some reasoning
>>>>> or just saying it is chosen empirically.
>>>>
>>>> We're dividing by folio_large_nr_pages(folio) (shifting by folio_large_order(folio)), so this is not a magic number at all.
>>>>
>>>> So this should be "ordinary" rounding.
>>>
>>> I thought the rounding would be (mapcount + 511) / 512.
>>
>> Yes, that's "rounding up".
>>
>>> But
>>> that means if one subpage is mapped, the average will be 1.
>>> Your rounding means if at least half of the subpages is mapped,
>>> the average will be 1. Others might think 1/3 is mapped,
>>> the average will be 1. That is why I think adjust looks like
>>> a magic number.
>>
>> I think all callers could tolerate (or benefit) from folio_average_page_mapcount() returning at least 1 in case any page is mapped.
>>
>> There was a reason why I decided to round to the nearest integer instead.
>>
>> Let me think about this once more, I went back and forth a couple of times on this.
> 
> Sure. Your current choice might be good enough for now. My intend of
> adding a comment here is just to let people know the adjust can be
> changed in the future. :)

The following will make the callers easier to read, while keeping
the rounding to the next integer for the other cases untouched.

+/**
+ * folio_average_page_mapcount() - Average number of mappings per page in this
+ *                                folio
+ * @folio: The folio.
+ *
+ * The average number of present user page table entries that reference each
+ * page in this folio as tracked via the RMAP: either referenced directly
+ * (PTE) or as part of a larger area that covers this page (e.g., PMD).
+ *
+ * The average is calculated by rounding to the nearest integer; however,
+ * if at least a single page is mapped, the average will be at least 1.
+ *
+ * Returns: The average number of mappings per page in this folio.
+ */
+static inline int folio_average_page_mapcount(struct folio *folio)
+{
+       int mapcount, entire_mapcount, avg;
+
+       if (!folio_test_large(folio))
+               return atomic_read(&folio->_mapcount) + 1;
+
+       mapcount = folio_large_mapcount(folio);
+       if (unlikely(mapcount <= 0))
+               return 0;
+       entire_mapcount = folio_entire_mapcount(folio);
+       if (mapcount <= entire_mapcount)
+               return entire_mapcount;
+       mapcount -= entire_mapcount;
+
+       /* Round to closest integer ... */
+       avg = (mapcount + folio_large_nr_pages(folio) / 2) >> folio_large_order(folio);
+       avg += entire_mapcount;
+       /* ... but return at least 1. */
+       return max_t(int, avg, 1);
+}


-- 
Cheers,

David / dhildenb

Re: [PATCH v2 16/20] fs/proc/page: remove per-page mapcount dependency for /proc/kpagecount (CONFIG_NO_PAGE_MAPCOUNT)

[Zi Yan]

On 24 Feb 2025, at 16:42, David Hildenbrand wrote:

> On 24.02.25 22:23, Zi Yan wrote:
>> On 24 Feb 2025, at 16:15, David Hildenbrand wrote:
>>
>>> On 24.02.25 22:10, Zi Yan wrote:
>>>> On 24 Feb 2025, at 16:02, David Hildenbrand wrote:
>>>>
>>>>> On 24.02.25 21:40, Zi Yan wrote:
>>>>>> On Mon Feb 24, 2025 at 11:55 AM EST, David Hildenbrand wrote:
>>>>>>> Let's implement an alternative when per-page mapcounts in large folios
>>>>>>> are no longer maintained -- soon with CONFIG_NO_PAGE_MAPCOUNT.
>>>>>>>
>>>>>>> For large folios, we'll return the per-page average mapcount within the
>>>>>>> folio, except when the average is 0 but the folio is mapped: then we
>>>>>>> return 1.
>>>>>>>
>>>>>>> For hugetlb folios and for large folios that are fully mapped
>>>>>>> into all address spaces, there is no change.
>>>>>>>
>>>>>>> As an alternative, we could simply return 0 for non-hugetlb large folios,
>>>>>>> or disable this legacy interface with CONFIG_NO_PAGE_MAPCOUNT.
>>>>>>>
>>>>>>> But the information exposed by this interface can still be valuable, and
>>>>>>> frequently we deal with fully-mapped large folios where the average
>>>>>>> corresponds to the actual page mapcount. So we'll leave it like this for
>>>>>>> now and document the new behavior.
>>>>>>>
>>>>>>> Note: this interface is likely not very relevant for performance. If
>>>>>>> ever required, we could try doing a rather expensive rmap walk to collect
>>>>>>> precisely how often this folio page is mapped.
>>>>>>>
>>>>>>> Signed-off-by: David Hildenbrand <david@redhat.com>
>>>>>>> ---
>>>>>>>     Documentation/admin-guide/mm/pagemap.rst |  7 +++++-
>>>>>>>     fs/proc/internal.h                       | 31 ++++++++++++++++++++++++
>>>>>>>     fs/proc/page.c                           | 19 ++++++++++++---
>>>>>>>     3 files changed, 53 insertions(+), 4 deletions(-)
>>>>>>>
>>>>>>> diff --git a/Documentation/admin-guide/mm/pagemap.rst b/Documentation/admin-guide/mm/pagemap.rst
>>>>>>> index caba0f52dd36c..49590306c61a0 100644
>>>>>>> --- a/Documentation/admin-guide/mm/pagemap.rst
>>>>>>> +++ b/Documentation/admin-guide/mm/pagemap.rst
>>>>>>> @@ -42,7 +42,12 @@ There are four components to pagemap:
>>>>>>>        skip over unmapped regions.
>>>>>>>       * ``/proc/kpagecount``.  This file contains a 64-bit count of the number of
>>>>>>> -   times each page is mapped, indexed by PFN.
>>>>>>> +   times each page is mapped, indexed by PFN. Some kernel configurations do
>>>>>>> +   not track the precise number of times a page part of a larger allocation
>>>>>>> +   (e.g., THP) is mapped. In these configurations, the average number of
>>>>>>> +   mappings per page in this larger allocation is returned instead. However,
>>>>>>> +   if any page of the large allocation is mapped, the returned value will
>>>>>>> +   be at least 1.
>>>>>>>      The page-types tool in the tools/mm directory can be used to query the
>>>>>>>     number of times a page is mapped.
>>>>>>> diff --git a/fs/proc/internal.h b/fs/proc/internal.h
>>>>>>> index 1695509370b88..16aa1fd260771 100644
>>>>>>> --- a/fs/proc/internal.h
>>>>>>> +++ b/fs/proc/internal.h
>>>>>>> @@ -174,6 +174,37 @@ static inline int folio_precise_page_mapcount(struct folio *folio,
>>>>>>>     	return mapcount;
>>>>>>>     }
>>>>>>>    +/**
>>>>>>> + * folio_average_page_mapcount() - Average number of mappings per page in this
>>>>>>> + *				   folio
>>>>>>> + * @folio: The folio.
>>>>>>> + *
>>>>>>> + * The average number of present user page table entries that reference each
>>>>>>> + * page in this folio as tracked via the RMAP: either referenced directly
>>>>>>> + * (PTE) or as part of a larger area that covers this page (e.g., PMD).
>>>>>>> + *
>>>>>>> + * Returns: The average number of mappings per page in this folio. 0 for
>>>>>>> + * folios that are not mapped to user space or are not tracked via the RMAP
>>>>>>> + * (e.g., shared zeropage).
>>>>>>> + */
>>>>>>> +static inline int folio_average_page_mapcount(struct folio *folio)
>>>>>>> +{
>>>>>>> +	int mapcount, entire_mapcount;
>>>>>>> +	unsigned int adjust;
>>>>>>> +
>>>>>>> +	if (!folio_test_large(folio))
>>>>>>> +		return atomic_read(&folio->_mapcount) + 1;
>>>>>>> +
>>>>>>> +	mapcount = folio_large_mapcount(folio);
>>>>>>> +	entire_mapcount = folio_entire_mapcount(folio);
>>>>>>> +	if (mapcount <= entire_mapcount)
>>>>>>> +		return entire_mapcount;
>>>>>>> +	mapcount -= entire_mapcount;
>>>>>>> +
>>>>>>> +	adjust = folio_large_nr_pages(folio) / 2;
>>>>>
>>>>> Thanks for the review!
>>>>>
>>>>>>
>>>>>> Is there any reason for choosing this adjust number? A comment might be
>>>>>> helpful in case people want to change it later, either with some reasoning
>>>>>> or just saying it is chosen empirically.
>>>>>
>>>>> We're dividing by folio_large_nr_pages(folio) (shifting by folio_large_order(folio)), so this is not a magic number at all.
>>>>>
>>>>> So this should be "ordinary" rounding.
>>>>
>>>> I thought the rounding would be (mapcount + 511) / 512.
>>>
>>> Yes, that's "rounding up".
>>>
>>>> But
>>>> that means if one subpage is mapped, the average will be 1.
>>>> Your rounding means if at least half of the subpages is mapped,
>>>> the average will be 1. Others might think 1/3 is mapped,
>>>> the average will be 1. That is why I think adjust looks like
>>>> a magic number.
>>>
>>> I think all callers could tolerate (or benefit) from folio_average_page_mapcount() returning at least 1 in case any page is mapped.
>>>
>>> There was a reason why I decided to round to the nearest integer instead.
>>>
>>> Let me think about this once more, I went back and forth a couple of times on this.
>>
>> Sure. Your current choice might be good enough for now. My intend of
>> adding a comment here is just to let people know the adjust can be
>> changed in the future. :)
>
> The following will make the callers easier to read, while keeping
> the rounding to the next integer for the other cases untouched.
>
> +/**
> + * folio_average_page_mapcount() - Average number of mappings per page in this
> + *                                folio
> + * @folio: The folio.
> + *
> + * The average number of present user page table entries that reference each
> + * page in this folio as tracked via the RMAP: either referenced directly
> + * (PTE) or as part of a larger area that covers this page (e.g., PMD).
> + *
> + * The average is calculated by rounding to the nearest integer; however,
> + * if at least a single page is mapped, the average will be at least 1.
> + *
> + * Returns: The average number of mappings per page in this folio.
> + */
> +static inline int folio_average_page_mapcount(struct folio *folio)
> +{
> +       int mapcount, entire_mapcount, avg;
> +
> +       if (!folio_test_large(folio))
> +               return atomic_read(&folio->_mapcount) + 1;
> +
> +       mapcount = folio_large_mapcount(folio);
> +       if (unlikely(mapcount <= 0))
> +               return 0;
> +       entire_mapcount = folio_entire_mapcount(folio);
> +       if (mapcount <= entire_mapcount)
> +               return entire_mapcount;
> +       mapcount -= entire_mapcount;
> +
> +       /* Round to closest integer ... */
> +       avg = (mapcount + folio_large_nr_pages(folio) / 2) >> folio_large_order(folio);
> +       avg += entire_mapcount;
> +       /* ... but return at least 1. */
> +       return max_t(int, avg, 1);
> +}

LGTM. Thanks.


Best Regards,
Yan, Zi

Re: [PATCH v2 16/20] fs/proc/page: remove per-page mapcount dependency for /proc/kpagecount (CONFIG_NO_PAGE_MAPCOUNT)

[David Hildenbrand]

On 24.02.25 22:44, Zi Yan wrote:
> On 24 Feb 2025, at 16:42, David Hildenbrand wrote:
> 
>> On 24.02.25 22:23, Zi Yan wrote:
>>> On 24 Feb 2025, at 16:15, David Hildenbrand wrote:
>>>
>>>> On 24.02.25 22:10, Zi Yan wrote:
>>>>> On 24 Feb 2025, at 16:02, David Hildenbrand wrote:
>>>>>
>>>>>> On 24.02.25 21:40, Zi Yan wrote:
>>>>>>> On Mon Feb 24, 2025 at 11:55 AM EST, David Hildenbrand wrote:
>>>>>>>> Let's implement an alternative when per-page mapcounts in large folios
>>>>>>>> are no longer maintained -- soon with CONFIG_NO_PAGE_MAPCOUNT.
>>>>>>>>
>>>>>>>> For large folios, we'll return the per-page average mapcount within the
>>>>>>>> folio, except when the average is 0 but the folio is mapped: then we
>>>>>>>> return 1.
>>>>>>>>
>>>>>>>> For hugetlb folios and for large folios that are fully mapped
>>>>>>>> into all address spaces, there is no change.
>>>>>>>>
>>>>>>>> As an alternative, we could simply return 0 for non-hugetlb large folios,
>>>>>>>> or disable this legacy interface with CONFIG_NO_PAGE_MAPCOUNT.
>>>>>>>>
>>>>>>>> But the information exposed by this interface can still be valuable, and
>>>>>>>> frequently we deal with fully-mapped large folios where the average
>>>>>>>> corresponds to the actual page mapcount. So we'll leave it like this for
>>>>>>>> now and document the new behavior.
>>>>>>>>
>>>>>>>> Note: this interface is likely not very relevant for performance. If
>>>>>>>> ever required, we could try doing a rather expensive rmap walk to collect
>>>>>>>> precisely how often this folio page is mapped.
>>>>>>>>
>>>>>>>> Signed-off-by: David Hildenbrand <david@redhat.com>
>>>>>>>> ---
>>>>>>>>      Documentation/admin-guide/mm/pagemap.rst |  7 +++++-
>>>>>>>>      fs/proc/internal.h                       | 31 ++++++++++++++++++++++++
>>>>>>>>      fs/proc/page.c                           | 19 ++++++++++++---
>>>>>>>>      3 files changed, 53 insertions(+), 4 deletions(-)
>>>>>>>>
>>>>>>>> diff --git a/Documentation/admin-guide/mm/pagemap.rst b/Documentation/admin-guide/mm/pagemap.rst
>>>>>>>> index caba0f52dd36c..49590306c61a0 100644
>>>>>>>> --- a/Documentation/admin-guide/mm/pagemap.rst
>>>>>>>> +++ b/Documentation/admin-guide/mm/pagemap.rst
>>>>>>>> @@ -42,7 +42,12 @@ There are four components to pagemap:
>>>>>>>>         skip over unmapped regions.
>>>>>>>>        * ``/proc/kpagecount``.  This file contains a 64-bit count of the number of
>>>>>>>> -   times each page is mapped, indexed by PFN.
>>>>>>>> +   times each page is mapped, indexed by PFN. Some kernel configurations do
>>>>>>>> +   not track the precise number of times a page part of a larger allocation
>>>>>>>> +   (e.g., THP) is mapped. In these configurations, the average number of
>>>>>>>> +   mappings per page in this larger allocation is returned instead. However,
>>>>>>>> +   if any page of the large allocation is mapped, the returned value will
>>>>>>>> +   be at least 1.
>>>>>>>>       The page-types tool in the tools/mm directory can be used to query the
>>>>>>>>      number of times a page is mapped.
>>>>>>>> diff --git a/fs/proc/internal.h b/fs/proc/internal.h
>>>>>>>> index 1695509370b88..16aa1fd260771 100644
>>>>>>>> --- a/fs/proc/internal.h
>>>>>>>> +++ b/fs/proc/internal.h
>>>>>>>> @@ -174,6 +174,37 @@ static inline int folio_precise_page_mapcount(struct folio *folio,
>>>>>>>>      	return mapcount;
>>>>>>>>      }
>>>>>>>>     +/**
>>>>>>>> + * folio_average_page_mapcount() - Average number of mappings per page in this
>>>>>>>> + *				   folio
>>>>>>>> + * @folio: The folio.
>>>>>>>> + *
>>>>>>>> + * The average number of present user page table entries that reference each
>>>>>>>> + * page in this folio as tracked via the RMAP: either referenced directly
>>>>>>>> + * (PTE) or as part of a larger area that covers this page (e.g., PMD).
>>>>>>>> + *
>>>>>>>> + * Returns: The average number of mappings per page in this folio. 0 for
>>>>>>>> + * folios that are not mapped to user space or are not tracked via the RMAP
>>>>>>>> + * (e.g., shared zeropage).
>>>>>>>> + */
>>>>>>>> +static inline int folio_average_page_mapcount(struct folio *folio)
>>>>>>>> +{
>>>>>>>> +	int mapcount, entire_mapcount;
>>>>>>>> +	unsigned int adjust;
>>>>>>>> +
>>>>>>>> +	if (!folio_test_large(folio))
>>>>>>>> +		return atomic_read(&folio->_mapcount) + 1;
>>>>>>>> +
>>>>>>>> +	mapcount = folio_large_mapcount(folio);
>>>>>>>> +	entire_mapcount = folio_entire_mapcount(folio);
>>>>>>>> +	if (mapcount <= entire_mapcount)
>>>>>>>> +		return entire_mapcount;
>>>>>>>> +	mapcount -= entire_mapcount;
>>>>>>>> +
>>>>>>>> +	adjust = folio_large_nr_pages(folio) / 2;
>>>>>>
>>>>>> Thanks for the review!
>>>>>>
>>>>>>>
>>>>>>> Is there any reason for choosing this adjust number? A comment might be
>>>>>>> helpful in case people want to change it later, either with some reasoning
>>>>>>> or just saying it is chosen empirically.
>>>>>>
>>>>>> We're dividing by folio_large_nr_pages(folio) (shifting by folio_large_order(folio)), so this is not a magic number at all.
>>>>>>
>>>>>> So this should be "ordinary" rounding.
>>>>>
>>>>> I thought the rounding would be (mapcount + 511) / 512.
>>>>
>>>> Yes, that's "rounding up".
>>>>
>>>>> But
>>>>> that means if one subpage is mapped, the average will be 1.
>>>>> Your rounding means if at least half of the subpages is mapped,
>>>>> the average will be 1. Others might think 1/3 is mapped,
>>>>> the average will be 1. That is why I think adjust looks like
>>>>> a magic number.
>>>>
>>>> I think all callers could tolerate (or benefit) from folio_average_page_mapcount() returning at least 1 in case any page is mapped.
>>>>
>>>> There was a reason why I decided to round to the nearest integer instead.
>>>>
>>>> Let me think about this once more, I went back and forth a couple of times on this.
>>>
>>> Sure. Your current choice might be good enough for now. My intend of
>>> adding a comment here is just to let people know the adjust can be
>>> changed in the future. :)
>>
>> The following will make the callers easier to read, while keeping
>> the rounding to the next integer for the other cases untouched.
>>
>> +/**
>> + * folio_average_page_mapcount() - Average number of mappings per page in this
>> + *                                folio
>> + * @folio: The folio.
>> + *
>> + * The average number of present user page table entries that reference each
>> + * page in this folio as tracked via the RMAP: either referenced directly
>> + * (PTE) or as part of a larger area that covers this page (e.g., PMD).
>> + *
>> + * The average is calculated by rounding to the nearest integer; however,
>> + * if at least a single page is mapped, the average will be at least 1.
>> + *
>> + * Returns: The average number of mappings per page in this folio.
>> + */
>> +static inline int folio_average_page_mapcount(struct folio *folio)
>> +{
>> +       int mapcount, entire_mapcount, avg;
>> +
>> +       if (!folio_test_large(folio))
>> +               return atomic_read(&folio->_mapcount) + 1;
>> +
>> +       mapcount = folio_large_mapcount(folio);
>> +       if (unlikely(mapcount <= 0))
>> +               return 0;
>> +       entire_mapcount = folio_entire_mapcount(folio);
>> +       if (mapcount <= entire_mapcount)
>> +               return entire_mapcount;
>> +       mapcount -= entire_mapcount;
>> +
>> +       /* Round to closest integer ... */
>> +       avg = (mapcount + folio_large_nr_pages(folio) / 2) >> folio_large_order(folio);
>> +       avg += entire_mapcount;
>> +       /* ... but return at least 1. */
>> +       return max_t(int, avg, 1);
>> +}
> 
> LGTM. Thanks.

Thanks! BTW, I think I chose the "round to closest integer" primarily
to make the PSS estimate a bit better. But that is indeed something
that can be adjusted easily later.

BTW, as commented in the cover letter, being able to calculate the
avg without the entire_mapcount could clean this function up quite a bit
(and make it completely atomic), but that will require more work.

-- 
Cheers,

David / dhildenb

Re: [PATCH v2 17/20] fs/proc/task_mmu: remove per-page mapcount dependency for PM_MMAP_EXCLUSIVE (CONFIG_NO_PAGE_MAPCOUNT)

[David Hildenbrand]

On 24.02.25 17:55, David Hildenbrand wrote:
> Let's implement an alternative when per-page mapcounts in large folios are
> no longer maintained -- soon with CONFIG_NO_PAGE_MAPCOUNT.
> 
> PM_MMAP_EXCLUSIVE will now be set if folio_likely_mapped_shared() is
> true -- when the folio is considered "mapped shared", including when
> it once was "mapped shared" but no longer is, as documented.
> 
> This might result in and under-indication of "exclusively mapped", which
> is considered better than over-indicating it: under-estimating the USS
> (Unique Set Size) is better than over-estimating it.
> 
> As an alternative, we could simply remove that flag with
> CONFIG_NO_PAGE_MAPCOUNT completely, but there might be value to it. So,
> let's keep it like that and document the behavior.
> 
> Signed-off-by: David Hildenbrand <david@redhat.com>
> ---
>   Documentation/admin-guide/mm/pagemap.rst |  9 +++++++++
>   fs/proc/task_mmu.c                       | 11 +++++++++--
>   2 files changed, 18 insertions(+), 2 deletions(-)
> 
> diff --git a/Documentation/admin-guide/mm/pagemap.rst b/Documentation/admin-guide/mm/pagemap.rst
> index 49590306c61a0..131c86574c39a 100644
> --- a/Documentation/admin-guide/mm/pagemap.rst
> +++ b/Documentation/admin-guide/mm/pagemap.rst
> @@ -37,6 +37,15 @@ There are four components to pagemap:
>      precisely which pages are mapped (or in swap) and comparing mapped
>      pages between processes.
>   
> +   Note that in some kernel configurations, all pages part of a larger
> +   allocation (e.g., THP) might be considered "mapped shared" if the large
> +   allocation is considered "mapped shared": if not all pages are exclusive to
> +   the same process. Further, some kernel configurations might consider larger
> +   allocations "mapped shared", if they were at one point considered
> +   "mapped shared", even if they would now be considered "exclusively mapped".
> +   Consequently, in these kernel configurations, bit 56 might be set although
> +   the page is actually "exclusively mapped"

I rewrote this yet another time to maybe make it clearer ...

+   Traditionally, bit 56 indicates that a page is mapped exactly once and bit
+   56 is clear when a page is mapped multiple times, even when mapped in the
+   same process multiple times. In some kernel configurations, the semantics
+   for pages part of a larger allocation (e.g., THP) differ: bit 56 is set if
+   all pages part of the corresponding large allocation are *certainly* mapped
+   in the same process, even if the page is mapped multiple times in that
+   process. Bit 56 is clear when any page page of the larger allocation
+   is *maybe* mapped in a different process. In some cases, a large allocation
+   might be treated as "maybe mapped by multiple processes" even though this
+   is no longer the case.

(talking about "process" is not completely correct, it's actually "MMs"; but
that might add more confusion here)

-- 
Cheers,

David / dhildenb