mm/rmap: basic MM owner tracking for large folios (!hugetlb)

For small folios, we traditionally use the mapcount to decide whether it
was "certainly mapped exclusively" by a single MM (mapcount == 1

mm/rmap: basic MM owner tracking for large folios (!hugetlb)

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.

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: David Hildenbrand <[email protected]>
Cc: Andy Lutomirks^H^Hski <[email protected]>
Cc: Borislav Betkov <[email protected]>
Cc: Dave Hansen <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jann Horn <[email protected]>
Cc: Johannes Weiner <[email protected]>
Cc: Jonathan Corbet <[email protected]>
Cc: Kirill A. Shutemov <[email protected]>
Cc: Lance Yang <[email protected]>
Cc: Liam Howlett <[email protected]>
Cc: Lorenzo Stoakes <[email protected]>
Cc: Matthew Wilcow (Oracle) <[email protected]>
Cc: Michal Koutn <[email protected]>
Cc: Muchun Song <[email protected]>
Cc: tejun heo <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Cc: Vlastimil Babka <[email protected]>
Cc: Zefan Li <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>

show more ...

mm: move _entire_mapcount in folio to page[2] on 32bit

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 mapping

mm: move _entire_mapcount in folio to page[2] on 32bit

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.

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: David Hildenbrand <[email protected]>
Cc: Andy Lutomirks^H^Hski <[email protected]>
Cc: Borislav Betkov <[email protected]>
Cc: Dave Hansen <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jann Horn <[email protected]>
Cc: Johannes Weiner <[email protected]>
Cc: Jonathan Corbet <[email protected]>
Cc: Kirill A. Shutemov <[email protected]>
Cc: Lance Yang <[email protected]>
Cc: Liam Howlett <[email protected]>
Cc: Lorenzo Stoakes <[email protected]>
Cc: Matthew Wilcow (Oracle) <[email protected]>
Cc: Michal Koutn <[email protected]>
Cc: Muchun Song <[email protected]>
Cc: tejun heo <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Cc: Vlastimil Babka <[email protected]>
Cc: Zefan Li <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>

show more ...

mm: move _pincount in folio to page[2] on 32bit

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 u

mm: move _pincount in folio to page[2] on 32bit

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".

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: David Hildenbrand <[email protected]>
Cc: Andy Lutomirks^H^Hski <[email protected]>
Cc: Borislav Betkov <[email protected]>
Cc: Dave Hansen <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jann Horn <[email protected]>
Cc: Johannes Weiner <[email protected]>
Cc: Jonathan Corbet <[email protected]>
Cc: Kirill A. Shutemov <[email protected]>
Cc: Lance Yang <[email protected]>
Cc: Liam Howlett <[email protected]>
Cc: Lorenzo Stoakes <[email protected]>
Cc: Matthew Wilcow (Oracle) <[email protected]>
Cc: Michal Koutn <[email protected]>
Cc: Muchun Song <[email protected]>
Cc: tejun heo <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Cc: Vlastimil Babka <[email protected]>
Cc: Zefan Li <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>

show more ...

mm: move hugetlb specific things in folio to page[3]

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 pag

mm: move hugetlb specific things in folio to page[3]

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.

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: David Hildenbrand <[email protected]>
Cc: Andy Lutomirks^H^Hski <[email protected]>
Cc: Borislav Betkov <[email protected]>
Cc: Dave Hansen <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jann Horn <[email protected]>
Cc: Johannes Weiner <[email protected]>
Cc: Jonathan Corbet <[email protected]>
Cc: Kirill A. Shutemov <[email protected]>
Cc: Lance Yang <[email protected]>
Cc: Liam Howlett <[email protected]>
Cc: Lorenzo Stoakes <[email protected]>
Cc: Matthew Wilcow (Oracle) <[email protected]>
Cc: Michal Koutn <[email protected]>
Cc: Muchun Song <[email protected]>
Cc: tejun heo <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Cc: Vlastimil Babka <[email protected]>
Cc: Zefan Li <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>

show more ...

mm: let _folio_nr_pages overlay memcg_data in first tail page

Let's free up some more of the "unconditionally available on 64BIT" space
in order-1 folios by letting _folio_nr_pages overlay memcg_dat

mm: let _folio_nr_pages overlay memcg_data in first tail page

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.

[[email protected]: make "make htmldoc" happy]
Link: https://lkml.kernel.org/r/[email protected]
Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: David Hildenbrand <[email protected]>
Reviewed-by: Kirill A. Shutemov <[email protected]>
Cc: Andy Lutomirks^H^Hski <[email protected]>
Cc: Borislav Betkov <[email protected]>
Cc: Dave Hansen <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jann Horn <[email protected]>
Cc: Johannes Weiner <[email protected]>
Cc: Jonathan Corbet <[email protected]>
Cc: Lance Yang <[email protected]>
Cc: Liam Howlett <[email protected]>
Cc: Lorenzo Stoakes <[email protected]>
Cc: Matthew Wilcow (Oracle) <[email protected]>
Cc: Michal Koutn <[email protected]>
Cc: Muchun Song <[email protected]>
Cc: tejun heo <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Cc: Vlastimil Babka <[email protected]>
Cc: Zefan Li <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>

show more ...

fs/dax: properly refcount fs dax pages

Currently fs dax pages are considered free when the refcount drops to one
and their refcounts are not increased when mapped via PTEs or decreased
when unmapped

fs/dax: properly refcount fs dax pages

Currently fs dax pages are considered free when the refcount drops to one
and their refcounts are not increased when mapped via PTEs or decreased
when unmapped. This requires special logic in mm paths to detect that
these pages should not be properly refcounted, and to detect when the
refcount drops to one instead of zero.

On the other hand get_user_pages(), etc. will properly refcount fs dax
pages by taking a reference and dropping it when the page is unpinned.

Tracking this special behaviour requires extra PTE bits (eg. pte_devmap)
and introduces rules that are potentially confusing and specific to FS DAX
pages. To fix this, and to possibly allow removal of the special PTE bits
in future, convert the fs dax page refcounts to be zero based and instead
take a reference on the page each time it is mapped as is currently the
case for normal pages.

This may also allow a future clean-up to remove the pgmap refcounting that
is currently done in mm/gup.c.

Link: https://lkml.kernel.org/r/c7d886ad7468a20452ef6e0ddab6cfe220874e7c.1740713401.git-series.apopple@nvidia.com
Signed-off-by: Alistair Popple <[email protected]>
Reviewed-by: Dan Williams <[email protected]>
Tested-by: Alison Schofield <[email protected]>
Acked-by: David Hildenbrand <[email protected]>
Cc: Alexander Gordeev <[email protected]>
Cc: Asahi Lina <[email protected]>
Cc: Balbir Singh <[email protected]>
Cc: Bjorn Helgaas <[email protected]>
Cc: Catalin Marinas <[email protected]>
Cc: Christian Borntraeger <[email protected]>
Cc: Christoph Hellwig <[email protected]>
Cc: Chunyan Zhang <[email protected]>
Cc: "Darrick J. Wong" <[email protected]>
Cc: Dave Chinner <[email protected]>
Cc: Dave Hansen <[email protected]>
Cc: Dave Jiang <[email protected]>
Cc: Gerald Schaefer <[email protected]>
Cc: Heiko Carstens <[email protected]>
Cc: Huacai Chen <[email protected]>
Cc: Ira Weiny <[email protected]>
Cc: Jan Kara <[email protected]>
Cc: Jason Gunthorpe <[email protected]>
Cc: Jason Gunthorpe <[email protected]>
Cc: John Hubbard <[email protected]>
Cc: linmiaohe <[email protected]>
Cc: Logan Gunthorpe <[email protected]>
Cc: Matthew Wilcow (Oracle) <[email protected]>
Cc: Michael "Camp Drill Sergeant" Ellerman <[email protected]>
Cc: Nicholas Piggin <[email protected]>
Cc: Peter Xu <[email protected]>
Cc: Sven Schnelle <[email protected]>
Cc: Ted Ts'o <[email protected]>
Cc: Vasily Gorbik <[email protected]>
Cc: Vishal Verma <[email protected]>
Cc: Vivek Goyal <[email protected]>
Cc: WANG Xuerui <[email protected]>
Cc: Will Deacon <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>

show more ...

mm: allow compound zone device pages

Zone device pages are used to represent various type of device memory
managed by device drivers. Currently compound zone device pages are not
supported. This i

mm: allow compound zone device pages

Zone device pages are used to represent various type of device memory
managed by device drivers. Currently compound zone device pages are not
supported. This is because MEMORY_DEVICE_FS_DAX pages are the only user
of higher order zone device pages and have their own page reference
counting.

A future change will unify FS DAX reference counting with normal page
reference counting rules and remove the special FS DAX reference counting.
Supporting that requires compound zone device pages.

Supporting compound zone device pages requires compound_head() to
distinguish between head and tail pages whilst still preserving the
special struct page fields that are specific to zone device pages.

A tail page is distinguished by having bit zero being set in
page->compound_head, with the remaining bits pointing to the head page.
For zone device pages page->compound_head is shared with page->pgmap.

The page->pgmap field must be common to all pages within a folio, even if
the folio spans memory sections. Therefore pgmap is the same for both
head and tail pages and can be moved into the folio and we can use the
standard scheme to find compound_head from a tail page.

Link: https://lkml.kernel.org/r/67055d772e6102accf85161d0b57b0b3944292bf.1740713401.git-series.apopple@nvidia.com
Signed-off-by: Alistair Popple <[email protected]>
Signed-off-by: Balbir Singh <[email protected]>
Reviewed-by: Jason Gunthorpe <[email protected]>
Reviewed-by: Dan Williams <[email protected]>
Acked-by: David Hildenbrand <[email protected]>
Tested-by: Alison Schofield <[email protected]>
Cc: Alexander Gordeev <[email protected]>
Cc: Asahi Lina <[email protected]>
Cc: Bjorn Helgaas <[email protected]>
Cc: Catalin Marinas <[email protected]>
Cc: Christian Borntraeger <[email protected]>
Cc: Christoph Hellwig <[email protected]>
Cc: Chunyan Zhang <[email protected]>
Cc: "Darrick J. Wong" <[email protected]>
Cc: Dave Chinner <[email protected]>
Cc: Dave Hansen <[email protected]>
Cc: Dave Jiang <[email protected]>
Cc: Gerald Schaefer <[email protected]>
Cc: Heiko Carstens <[email protected]>
Cc: Huacai Chen <[email protected]>
Cc: Ira Weiny <[email protected]>
Cc: Jan Kara <[email protected]>
Cc: Jason Gunthorpe <[email protected]>
Cc: John Hubbard <[email protected]>
Cc: linmiaohe <[email protected]>
Cc: Logan Gunthorpe <[email protected]>
Cc: Matthew Wilcow (Oracle) <[email protected]>
Cc: Michael "Camp Drill Sergeant" Ellerman <[email protected]>
Cc: Nicholas Piggin <[email protected]>
Cc: Peter Xu <[email protected]>
Cc: Sven Schnelle <[email protected]>
Cc: Ted Ts'o <[email protected]>
Cc: Vasily Gorbik <[email protected]>
Cc: Vishal Verma <[email protected]>
Cc: Vivek Goyal <[email protected]>
Cc: WANG Xuerui <[email protected]>
Cc: Will Deacon <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>

show more ...

mm: make vma cache SLAB_TYPESAFE_BY_RCU

To enable SLAB_TYPESAFE_BY_RCU for vma cache we need to ensure that
object reuse before RCU grace period is over will be detected by
lock_vma_under_rcu().

Cu

mm: make vma cache SLAB_TYPESAFE_BY_RCU

To enable SLAB_TYPESAFE_BY_RCU for vma cache we need to ensure that
object reuse before RCU grace period is over will be detected by
lock_vma_under_rcu().

Current checks are sufficient as long as vma is detached before it is
freed. The only place this is not currently happening is in exit_mmap().
Add the missing vma_mark_detached() in exit_mmap().

Another issue which might trick lock_vma_under_rcu() during vma reuse is
vm_area_dup(), which copies the entire content of the vma into a new one,
overriding new vma's vm_refcnt and temporarily making it appear as
attached. This might trick a racing lock_vma_under_rcu() to operate on a
reused vma if it found the vma before it got reused. To prevent this
situation, we should ensure that vm_refcnt stays at detached state (0)
when it is copied and advances to attached state only after it is added
into the vma tree. Introduce vm_area_init_from() which preserves new
vma's vm_refcnt and use it in vm_area_dup(). Since all vmas are in
detached state with no current readers when they are freed,

lock_vma_under_rcu() will not be able to take vm_refcnt after vma got
detached even if vma is reused. vma_mark_attached() in modified to
include a release fence to ensure all stores to the vma happen before
vm_refcnt gets initialized.

Finally, make vm_area_cachep SLAB_TYPESAFE_BY_RCU. This will facilitate
vm_area_struct reuse and will minimize the number of call_rcu() calls.

[[email protected]: remove atomic_set_release() usage in tools/]
Link: https://lkml.kernel.org/r/[email protected]
Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Suren Baghdasaryan <[email protected]>
Reviewed-by: Vlastimil Babka <[email protected]>
Tested-by: Shivank Garg <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
Cc: Christian Brauner <[email protected]>
Cc: David Hildenbrand <[email protected]>
Cc: David Howells <[email protected]>
Cc: Davidlohr Bueso <[email protected]>
Cc: Hugh Dickins <[email protected]>
Cc: Jann Horn <[email protected]>
Cc: Johannes Weiner <[email protected]>
Cc: Jonathan Corbet <[email protected]>
Cc: Klara Modin <[email protected]>
Cc: Liam R. Howlett <[email protected]>
Cc: Lokesh Gidra <[email protected]>
Cc: Lorenzo Stoakes <[email protected]>
Cc: Mateusz Guzik <[email protected]>
Cc: Matthew Wilcox <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Minchan Kim <[email protected]>
Cc: Oleg Nesterov <[email protected]>
Cc: Pasha Tatashin <[email protected]>
Cc: "Paul E . McKenney" <[email protected]>
Cc: Peter Xu <[email protected]>
Cc: Peter Zijlstra (Intel) <[email protected]>
Cc: Shakeel Butt <[email protected]>
Cc: Sourav Panda <[email protected]>
Cc: Wei Yang <[email protected]>
Cc: Will Deacon <[email protected]>
Cc: Heiko Carstens <[email protected]>
Cc: Stephen Rothwell <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>

show more ...

mm: move lesser used vma_area_struct members into the last cacheline

Move several vma_area_struct members which are rarely or never used during
page fault handling into the last cacheline to better

mm: move lesser used vma_area_struct members into the last cacheline

Move several vma_area_struct members which are rarely or never used during
page fault handling into the last cacheline to better pack vm_area_struct.
As a result vm_area_struct will fit into 3 as opposed to 4 cachelines.
New typical vm_area_struct layout:

struct vm_area_struct {
union {
struct {
long unsigned int vm_start; /* 0 8 */
long unsigned int vm_end; /* 8 8 */
}; /* 0 16 */
freeptr_t vm_freeptr; /* 0 8 */
}; /* 0 16 */
struct mm_struct * vm_mm; /* 16 8 */
pgprot_t vm_page_prot; /* 24 8 */
union {
const vm_flags_t vm_flags; /* 32 8 */
vm_flags_t __vm_flags; /* 32 8 */
}; /* 32 8 */
unsigned int vm_lock_seq; /* 40 4 */

/* XXX 4 bytes hole, try to pack */

struct list_head anon_vma_chain; /* 48 16 */
/* --- cacheline 1 boundary (64 bytes) --- */
struct anon_vma * anon_vma; /* 64 8 */
const struct vm_operations_struct * vm_ops; /* 72 8 */
long unsigned int vm_pgoff; /* 80 8 */
struct file * vm_file; /* 88 8 */
void * vm_private_data; /* 96 8 */
atomic_long_t swap_readahead_info; /* 104 8 */
struct mempolicy * vm_policy; /* 112 8 */
struct vma_numab_state * numab_state; /* 120 8 */
/* --- cacheline 2 boundary (128 bytes) --- */
refcount_t vm_refcnt (__aligned__(64)); /* 128 4 */

/* XXX 4 bytes hole, try to pack */

struct {
struct rb_node rb (__aligned__(8)); /* 136 24 */
long unsigned int rb_subtree_last; /* 160 8 */
} __attribute__((__aligned__(8))) shared; /* 136 32 */
struct anon_vma_name * anon_name; /* 168 8 */
struct vm_userfaultfd_ctx vm_userfaultfd_ctx; /* 176 8 */

/* size: 192, cachelines: 3, members: 18 */
/* sum members: 176, holes: 2, sum holes: 8 */
/* padding: 8 */
/* forced alignments: 2, forced holes: 1, sum forced holes: 4 */
} __attribute__((__aligned__(64)));

Memory consumption per 1000 VMAs becomes 48 pages:

slabinfo after vm_area_struct changes:
<name> ... <objsize> <objperslab> <pagesperslab> : ...
vm_area_struct ... 192 42 2 : ...

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Suren Baghdasaryan <[email protected]>
Reviewed-by: Lorenzo Stoakes <[email protected]>
Tested-by: Shivank Garg <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
Reviewed-by: Vlastimil Babka <[email protected]>
Cc: Christian Brauner <[email protected]>
Cc: David Hildenbrand <[email protected]>
Cc: David Howells <[email protected]>
Cc: Davidlohr Bueso <[email protected]>
Cc: Hugh Dickins <[email protected]>
Cc: Jann Horn <[email protected]>
Cc: Johannes Weiner <[email protected]>
Cc: Jonathan Corbet <[email protected]>
Cc: Klara Modin <[email protected]>
Cc: Liam R. Howlett <[email protected]>
Cc: Lokesh Gidra <[email protected]>
Cc: Mateusz Guzik <[email protected]>
Cc: Matthew Wilcox <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Minchan Kim <[email protected]>
Cc: Oleg Nesterov <[email protected]>
Cc: Pasha Tatashin <[email protected]>
Cc: "Paul E . McKenney" <[email protected]>
Cc: Peter Xu <[email protected]>
Cc: Peter Zijlstra (Intel) <[email protected]>
Cc: Shakeel Butt <[email protected]>
Cc: Sourav Panda <[email protected]>
Cc: Wei Yang <[email protected]>
Cc: Will Deacon <[email protected]>
Cc: Heiko Carstens <[email protected]>
Cc: Stephen Rothwell <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>

show more ...

mm: replace vm_lock and detached flag with a reference count

rw_semaphore is a sizable structure of 40 bytes and consumes considerable
space for each vm_area_struct. However vma_lock has two import

mm: replace vm_lock and detached flag with a reference count

rw_semaphore is a sizable structure of 40 bytes and consumes considerable
space for each vm_area_struct. However vma_lock has two important
specifics which can be used to replace rw_semaphore with a simpler
structure:

1. Readers never wait. They try to take the vma_lock and fall back to
mmap_lock if that fails.

2. Only one writer at a time will ever try to write-lock a vma_lock
because writers first take mmap_lock in write mode. Because of these
requirements, full rw_semaphore functionality is not needed and we can
replace rw_semaphore and the vma->detached flag with a refcount
(vm_refcnt).

When vma is in detached state, vm_refcnt is 0 and only a call to
vma_mark_attached() can take it out of this state. Note that unlike
before, now we enforce both vma_mark_attached() and vma_mark_detached() to
be done only after vma has been write-locked. vma_mark_attached() changes
vm_refcnt to 1 to indicate that it has been attached to the vma tree.
When a reader takes read lock, it increments vm_refcnt, unless the top
usable bit of vm_refcnt (0x40000000) is set, indicating presence of a
writer. When writer takes write lock, it sets the top usable bit to
indicate its presence. If there are readers, writer will wait using newly
introduced mm->vma_writer_wait. Since all writers take mmap_lock in write
mode first, there can be only one writer at a time. The last reader to
release the lock will signal the writer to wake up. refcount might
overflow if there are many competing readers, in which case read-locking
will fail. Readers are expected to handle such failures.

In summary:
1. all readers increment the vm_refcnt;
2. writer sets top usable (writer) bit of vm_refcnt;
3. readers cannot increment the vm_refcnt if the writer bit is set;
4. in the presence of readers, writer must wait for the vm_refcnt to drop
to 1 (plus the VMA_LOCK_OFFSET writer bit), indicating an attached vma
with no readers;
5. vm_refcnt overflow is handled by the readers.

While this vm_lock replacement does not yet result in a smaller
vm_area_struct (it stays at 256 bytes due to cacheline alignment), it
allows for further size optimization by structure member regrouping to
bring the size of vm_area_struct below 192 bytes.

[[email protected]: fix a crash due to vma_end_read() that should have been removed]
Link: https://lkml.kernel.org/r/[email protected]
Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Suren Baghdasaryan <[email protected]>
Suggested-by: Peter Zijlstra <[email protected]>
Suggested-by: Matthew Wilcox <[email protected]>
Tested-by: Shivank Garg <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
Reviewed-by: Vlastimil Babka <[email protected]>
Cc: Christian Brauner <[email protected]>
Cc: David Hildenbrand <[email protected]>
Cc: David Howells <[email protected]>
Cc: Davidlohr Bueso <[email protected]>
Cc: Hugh Dickins <[email protected]>
Cc: Jann Horn <[email protected]>
Cc: Johannes Weiner <[email protected]>
Cc: Jonathan Corbet <[email protected]>
Cc: Klara Modin <[email protected]>
Cc: Liam R. Howlett <[email protected]>
Cc: Lokesh Gidra <[email protected]>
Cc: Lorenzo Stoakes <[email protected]>
Cc: Mateusz Guzik <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Minchan Kim <[email protected]>
Cc: Oleg Nesterov <[email protected]>
Cc: Pasha Tatashin <[email protected]>
Cc: "Paul E . McKenney" <[email protected]>
Cc: Peter Xu <[email protected]>
Cc: Shakeel Butt <[email protected]>
Cc: Sourav Panda <[email protected]>
Cc: Wei Yang <[email protected]>
Cc: Will Deacon <[email protected]>
Cc: Heiko Carstens <[email protected]>
Cc: Stephen Rothwell <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>

show more ...

mm: move per-vma lock into vm_area_struct

Back when per-vma locks were introduces, vm_lock was moved out of
vm_area_struct in [1] because of the performance regression caused by
false cacheline shar

mm: move per-vma lock into vm_area_struct

Back when per-vma locks were introduces, vm_lock was moved out of
vm_area_struct in [1] because of the performance regression caused by
false cacheline sharing. Recent investigation [2] revealed that the
regressions is limited to a rather old Broadwell microarchitecture and
even there it can be mitigated by disabling adjacent cacheline
prefetching, see [3].

Splitting single logical structure into multiple ones leads to more
complicated management, extra pointer dereferences and overall less
maintainable code. When that split-away part is a lock, it complicates
things even further. With no performance benefits, there are no reasons
for this split. Merging the vm_lock back into vm_area_struct also allows
vm_area_struct to use SLAB_TYPESAFE_BY_RCU later in this patchset. Move
vm_lock back into vm_area_struct, aligning it at the cacheline boundary
and changing the cache to be cacheline-aligned as well. With kernel
compiled using defconfig, this causes VMA memory consumption to grow from
160 (vm_area_struct) + 40 (vm_lock) bytes to 256 bytes:

slabinfo before:
<name> ... <objsize> <objperslab> <pagesperslab> : ...
vma_lock ... 40 102 1 : ...
vm_area_struct ... 160 51 2 : ...

slabinfo after moving vm_lock:
<name> ... <objsize> <objperslab> <pagesperslab> : ...
vm_area_struct ... 256 32 2 : ...

Aggregate VMA memory consumption per 1000 VMAs grows from 50 to 64 pages,
which is 5.5MB per 100000 VMAs. Note that the size of this structure is
dependent on the kernel configuration and typically the original size is
higher than 160 bytes. Therefore these calculations are close to the
worst case scenario. A more realistic vm_area_struct usage before this
change is:

<name> ... <objsize> <objperslab> <pagesperslab> : ...
vma_lock ... 40 102 1 : ...
vm_area_struct ... 176 46 2 : ...

Aggregate VMA memory consumption per 1000 VMAs grows from 54 to 64 pages,
which is 3.9MB per 100000 VMAs. This memory consumption growth can be
addressed later by optimizing the vm_lock.

[1] https://lore.kernel.org/all/[email protected]/
[2] https://lore.kernel.org/all/ZsQyI%2F087V34JoIt@xsang-OptiPlex-9020/
[3] https://lore.kernel.org/all/CAJuCfpEisU8Lfe96AYJDZ+OM4NoPmnw9bP53cT_kbfP_pR+-2g@mail.gmail.com/

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Suren Baghdasaryan <[email protected]>
Reviewed-by: Lorenzo Stoakes <[email protected]>
Reviewed-by: Shakeel Butt <[email protected]>
Reviewed-by: Vlastimil Babka <[email protected]>
Reviewed-by: Liam R. Howlett <[email protected]>
Tested-by: Shivank Garg <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
Cc: Christian Brauner <[email protected]>
Cc: David Hildenbrand <[email protected]>
Cc: David Howells <[email protected]>
Cc: Davidlohr Bueso <[email protected]>
Cc: Hugh Dickins <[email protected]>
Cc: Jann Horn <[email protected]>
Cc: Johannes Weiner <[email protected]>
Cc: Jonathan Corbet <[email protected]>
Cc: Klara Modin <[email protected]>
Cc: Lokesh Gidra <[email protected]>
Cc: Mateusz Guzik <[email protected]>
Cc: Matthew Wilcox <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Minchan Kim <[email protected]>
Cc: Oleg Nesterov <[email protected]>
Cc: Pasha Tatashin <[email protected]>
Cc: "Paul E . McKenney" <[email protected]>
Cc: Peter Xu <[email protected]>
Cc: Peter Zijlstra (Intel) <[email protected]>
Cc: Sourav Panda <[email protected]>
Cc: Wei Yang <[email protected]>
Cc: Will Deacon <[email protected]>
Cc: Heiko Carstens <[email protected]>
Cc: Stephen Rothwell <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>

show more ...

mm, bpf: Introduce free_pages_nolock()

Introduce free_pages_nolock() that can free pages without taking locks.
It relies on trylock and can be called from any context.
Since spin_trylock() cannot be

mm, bpf: Introduce free_pages_nolock()

Introduce free_pages_nolock() that can free pages without taking locks.
It relies on trylock and can be called from any context.
Since spin_trylock() cannot be used in PREEMPT_RT from hard IRQ or NMI
it uses lockless link list to stash the pages which will be freed
by subsequent free_pages() from good context.

Do not use llist unconditionally. BPF maps continuously
allocate/free, so we cannot unconditionally delay the freeing to
llist. When the memory becomes free make it available to the
kernel and BPF users right away if possible, and fallback to
llist as the last resort.

Acked-by: Vlastimil Babka <[email protected]>
Acked-by: Sebastian Andrzej Siewior <[email protected]>
Reviewed-by: Shakeel Butt <[email protected]>
Signed-off-by: Alexei Starovoitov <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Alexei Starovoitov <[email protected]>

show more ...

sched: Compact RSEQ concurrency IDs with reduced threads and affinity

When a process reduces its number of threads or clears bits in its CPU
affinity mask, the mm_cid allocation should eventually co

sched: Compact RSEQ concurrency IDs with reduced threads and affinity

When a process reduces its number of threads or clears bits in its CPU
affinity mask, the mm_cid allocation should eventually converge towards
smaller values.

However, the change introduced by:

commit 7e019dcc470f ("sched: Improve cache locality of RSEQ concurrency
IDs for intermittent workloads")

adds a per-mm/CPU recent_cid which is never unset unless a thread
migrates.

This is a tradeoff between:

A) Preserving cache locality after a transition from many threads to few
threads, or after reducing the hamming weight of the allowed CPU mask.

B) Making the mm_cid upper bounds wrt nr threads and allowed CPU mask
easy to document and understand.

C) Allowing applications to eventually react to mm_cid compaction after
reduction of the nr threads or allowed CPU mask, making the tracking
of mm_cid compaction easier by shrinking it back towards 0 or not.

D) Making sure applications that periodically reduce and then increase
again the nr threads or allowed CPU mask still benefit from good
cache locality with mm_cid.

Introduce the following changes:

* After shrinking the number of threads or reducing the number of
allowed CPUs, reduce the value of max_nr_cid so expansion of CID
allocation will preserve cache locality if the number of threads or
allowed CPUs increase again.

* Only re-use a recent_cid if it is within the max_nr_cid upper bound,
else find the first available CID.

Fixes: 7e019dcc470f ("sched: Improve cache locality of RSEQ concurrency IDs for intermittent workloads")
Signed-off-by: Mathieu Desnoyers <[email protected]>
Signed-off-by: Gabriele Monaco <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Tested-by: Gabriele Monaco <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]

show more ...

mm: convert mm_lock_seq to a proper seqcount

Convert mm_lock_seq to be seqcount_t and change all mmap_write_lock
variants to increment it, in-line with the usual seqcount usage pattern.
This lets us

mm: convert mm_lock_seq to a proper seqcount

Convert mm_lock_seq to be seqcount_t and change all mmap_write_lock
variants to increment it, in-line with the usual seqcount usage pattern.
This lets us check whether the mmap_lock is write-locked by checking
mm_lock_seq.sequence counter (odd=locked, even=unlocked). This will be
used when implementing mmap_lock speculation functions.
As a result vm_lock_seq is also change to be unsigned to match the type
of mm_lock_seq.sequence.

Link: https://lkml.kernel.org/r/[email protected]
Suggested-by: Peter Zijlstra <[email protected]>
Signed-off-by: Suren Baghdasaryan <[email protected]>
Reviewed-by: Liam R. Howlett <[email protected]>
Cc: Christian Brauner <[email protected]>
Cc: David Hildenbrand <[email protected]>
Cc: David Howells <[email protected]>
Cc: Davidlohr Bueso <[email protected]>
Cc: Hillf Danton <[email protected]>
Cc: Hugh Dickins <[email protected]>
Cc: Jann Horn <[email protected]>
Cc: Johannes Weiner <[email protected]>
Cc: Jonathan Corbet <[email protected]>
Cc: Lorenzo Stoakes <[email protected]>
Cc: Mateusz Guzik <[email protected]>
Cc: Matthew Wilcox <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Minchan Kim <[email protected]>
Cc: Oleg Nesterov <[email protected]>
Cc: Pasha Tatashin <[email protected]>
Cc: Paul E. McKenney <[email protected]>
Cc: Peter Xu <[email protected]>
Cc: Shakeel Butt <[email protected]>
Cc: Sourav Panda <[email protected]>
Cc: Vlastimil Babka <[email protected]>
Cc: Wei Yang <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>

show more ...

x86: mm: free page table pages by RCU instead of semi RCU

Now, if CONFIG_MMU_GATHER_RCU_TABLE_FREE is selected, the page table pages
will be freed by semi RCU, that is:

- batch table freeing: asyn

x86: mm: free page table pages by RCU instead of semi RCU

Now, if CONFIG_MMU_GATHER_RCU_TABLE_FREE is selected, the page table pages
will be freed by semi RCU, that is:

- batch table freeing: asynchronous free by RCU
- single table freeing: IPI + synchronous free

In this way, the page table can be lockless traversed by disabling IRQ in
paths such as fast GUP. But this is not enough to free the empty PTE page
table pages in paths other that munmap and exit_mmap path, because IPI
cannot be synchronized with rcu_read_lock() in pte_offset_map{_lock}().

In preparation for supporting empty PTE page table pages reclaimation, let
single table also be freed by RCU like batch table freeing. Then we can
also use pte_offset_map() etc to prevent PTE page from being freed.

Like pte_free_defer(), we can also safely use ptdesc->pt_rcu_head to free
the page table pages:

- The pt_rcu_head is unioned with pt_list and pmd_huge_pte.

- For pt_list, it is used to manage the PGD page in x86. Fortunately
tlb_remove_table() will not be used for free PGD pages, so it is safe
to use pt_rcu_head.

- For pmd_huge_pte, it is used for THPs, so it is safe.

After applying this patch, if CONFIG_PT_RECLAIM is enabled, the function
call of free_pte() is as follows:

free_pte
pte_free_tlb
__pte_free_tlb
___pte_free_tlb
paravirt_tlb_remove_table
tlb_remove_table [!CONFIG_PARAVIRT, Xen PV, Hyper-V, KVM]
[no-free-memory slowpath:]
tlb_table_invalidate
tlb_remove_table_one
__tlb_remove_table_one [frees via RCU]
[fastpath:]
tlb_table_flush
tlb_remove_table_free [frees via RCU]
native_tlb_remove_table [CONFIG_PARAVIRT on native]
tlb_remove_table [see above]

Link: https://lkml.kernel.org/r/0287d442a973150b0e1019cc406e6322d148277a.1733305182.git.zhengqi.arch@bytedance.com
Signed-off-by: Qi Zheng <[email protected]>
Cc: Dave Hansen <[email protected]>
Cc: Andy Lutomirski <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Catalin Marinas <[email protected]>
Cc: David Hildenbrand <[email protected]>
Cc: David Rientjes <[email protected]>
Cc: Hugh Dickins <[email protected]>
Cc: Jann Horn <[email protected]>
Cc: Lorenzo Stoakes <[email protected]>
Cc: Matthew Wilcox <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: Muchun Song <[email protected]>
Cc: Peter Xu <[email protected]>
Cc: Will Deacon <[email protected]>
Cc: Zach O'Keefe <[email protected]>
Cc: Dan Carpenter <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>

show more ...

mm: hugetlb: independent PMD page table shared count

The folio refcount may be increased unexpectly through try_get_folio() by
caller such as split_huge_pages. In huge_pmd_unshare(), we use refcoun

mm: hugetlb: independent PMD page table shared count

The folio refcount may be increased unexpectly through try_get_folio() by
caller such as split_huge_pages. In huge_pmd_unshare(), we use refcount
to check whether a pmd page table is shared. The check is incorrect if
the refcount is increased by the above caller, and this can cause the page
table leaked:

BUG: Bad page state in process sh pfn:109324
page: refcount:0 mapcount:0 mapping:0000000000000000 index:0x66 pfn:0x109324
flags: 0x17ffff800000000(node=0|zone=2|lastcpupid=0xfffff)
page_type: f2(table)
raw: 017ffff800000000 0000000000000000 0000000000000000 0000000000000000
raw: 0000000000000066 0000000000000000 00000000f2000000 0000000000000000
page dumped because: nonzero mapcount
...
CPU: 31 UID: 0 PID: 7515 Comm: sh Kdump: loaded Tainted: G B 6.13.0-rc2master+ #7
Tainted: [B]=BAD_PAGE
Hardware name: QEMU KVM Virtual Machine, BIOS 0.0.0 02/06/2015
Call trace:
show_stack+0x20/0x38 (C)
dump_stack_lvl+0x80/0xf8
dump_stack+0x18/0x28
bad_page+0x8c/0x130
free_page_is_bad_report+0xa4/0xb0
free_unref_page+0x3cc/0x620
__folio_put+0xf4/0x158
split_huge_pages_all+0x1e0/0x3e8
split_huge_pages_write+0x25c/0x2d8
full_proxy_write+0x64/0xd8
vfs_write+0xcc/0x280
ksys_write+0x70/0x110
__arm64_sys_write+0x24/0x38
invoke_syscall+0x50/0x120
el0_svc_common.constprop.0+0xc8/0xf0
do_el0_svc+0x24/0x38
el0_svc+0x34/0x128
el0t_64_sync_handler+0xc8/0xd0
el0t_64_sync+0x190/0x198

The issue may be triggered by damon, offline_page, page_idle, etc, which
will increase the refcount of page table.

1. The page table itself will be discarded after reporting the
"nonzero mapcount".

2. The HugeTLB page mapped by the page table miss freeing since we
treat the page table as shared and a shared page table will not be
unmapped.

Fix it by introducing independent PMD page table shared count. As
described by comment, pt_index/pt_mm/pt_frag_refcount are used for s390
gmap, x86 pgds and powerpc, pt_share_count is used for x86/arm64/riscv
pmds, so we can reuse the field as pt_share_count.

Link: https://lkml.kernel.org/r/[email protected]
Fixes: 39dde65c9940 ("[PATCH] shared page table for hugetlb page")
Signed-off-by: Liu Shixin <[email protected]>
Cc: Kefeng Wang <[email protected]>
Cc: Ken Chen <[email protected]>
Cc: Muchun Song <[email protected]>
Cc: Nanyong Sun <[email protected]>
Cc: Jane Chu <[email protected]>
Cc: <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>

show more ...

mm: convert mm_lock_seq to a proper seqcount

Convert mm_lock_seq to be seqcount_t and change all mmap_write_lock
variants to increment it, in-line with the usual seqcount usage pattern.
This lets us

mm: convert mm_lock_seq to a proper seqcount

Convert mm_lock_seq to be seqcount_t and change all mmap_write_lock
variants to increment it, in-line with the usual seqcount usage pattern.
This lets us check whether the mmap_lock is write-locked by checking
mm_lock_seq.sequence counter (odd=locked, even=unlocked). This will be
used when implementing mmap_lock speculation functions.
As a result vm_lock_seq is also change to be unsigned to match the type
of mm_lock_seq.sequence.

Suggested-by: Peter Zijlstra <[email protected]>
Signed-off-by: Suren Baghdasaryan <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Reviewed-by: Liam R. Howlett <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]

show more ...

x86/mm/tlb: Add tracepoint for TLB flush IPI to stale CPU

Add a tracepoint when we send a TLB flush IPI to a CPU that used
to be in the mm_cpumask, but isn't any more.

Suggested-by: Dave Hansen <da

x86/mm/tlb: Add tracepoint for TLB flush IPI to stale CPU

Add a tracepoint when we send a TLB flush IPI to a CPU that used
to be in the mm_cpumask, but isn't any more.

Suggested-by: Dave Hansen <[email protected]>
Signed-off-by: Rik van Riel <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
Link: https://lore.kernel.org/r/[email protected]

show more ...

mm: move the page fragment allocator from page_alloc into its own file

Inspired by [1], move the page fragment allocator from page_alloc
into its own c file and header file, as we are about to make

mm: move the page fragment allocator from page_alloc into its own file

Inspired by [1], move the page fragment allocator from page_alloc
into its own c file and header file, as we are about to make more
change for it to replace another page_frag implementation in
sock.c

As this patchset is going to replace 'struct page_frag' with
'struct page_frag_cache' in sched.h, including page_frag_cache.h
in sched.h has a compiler error caused by interdependence between
mm_types.h and mm.h for asm-offsets.c, see [2]. So avoid the compiler
error by moving 'struct page_frag_cache' to mm_types_task.h as
suggested by Alexander, see [3].

1. https://lore.kernel.org/all/[email protected]/
2. https://lore.kernel.org/all/[email protected]/
3. https://lore.kernel.org/all/CAKgT0UdH1yD=LSCXFJ=YM_aiA4OomD-2wXykO42bizaWMt_HOA@mail.gmail.com/
CC: David Howells <[email protected]>
CC: Linux-MM <[email protected]>
Signed-off-by: Yunsheng Lin <[email protected]>
Acked-by: Andrew Morton <[email protected]>
Reviewed-by: Alexander Duyck <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Jakub Kicinski <[email protected]>

show more ...

mm: move mm flags to mm_types.h

The types of mm flags are now far beyond the core dump related features.
This patch moves mm flags from linux/sched/coredump.h to linux/mm_types.h.
The linux/sched/c

mm: move mm flags to mm_types.h

The types of mm flags are now far beyond the core dump related features.
This patch moves mm flags from linux/sched/coredump.h to linux/mm_types.h.
The linux/sched/coredump.h has include the mm_types.h, so the C files
related to coredump does not need to change head file inclusion. In
addition, the inclusion of sched/coredump.h now can be deleted from the C
files that irrelevant to core dump.

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Nanyong Sun <[email protected]>
Cc: Kefeng Wang <[email protected]>
Cc: Masami Hiramatsu <[email protected]>
Cc: Matthew Wilcox <[email protected]>
Cc: Oleg Nesterov <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>

show more ...

sched: Improve cache locality of RSEQ concurrency IDs for intermittent workloads

commit 223baf9d17f25 ("sched: Fix performance regression introduced by mm_cid")
introduced a per-mm/cpu current concu

sched: Improve cache locality of RSEQ concurrency IDs for intermittent workloads

commit 223baf9d17f25 ("sched: Fix performance regression introduced by mm_cid")
introduced a per-mm/cpu current concurrency id (mm_cid), which keeps
a reference to the concurrency id allocated for each CPU. This reference
expires shortly after a 100ms delay.

These per-CPU references keep the per-mm-cid data cache-local in
situations where threads are running at least once on each CPU within
each 100ms window, thus keeping the per-cpu reference alive.

However, intermittent workloads behaving in bursts spaced by more than
100ms on each CPU exhibit bad cache locality and degraded performance
compared to purely per-cpu data indexing, because concurrency IDs are
allocated over various CPUs and cores, therefore losing cache locality
of the associated data.

Introduce the following changes to improve per-mm-cid cache locality:

- Add a "recent_cid" field to the per-mm/cpu mm_cid structure to keep
track of which mm_cid value was last used, and use it as a hint to
attempt re-allocating the same concurrency ID the next time this
mm/cpu needs to allocate a concurrency ID,

- Add a per-mm CPUs allowed mask, which keeps track of the union of
CPUs allowed for all threads belonging to this mm. This cpumask is
only set during the lifetime of the mm, never cleared, so it
represents the union of all the CPUs allowed since the beginning of
the mm lifetime (note that the mm_cpumask() is really arch-specific
and tailored to the TLB flush needs, and is thus _not_ a viable
approach for this),

- Add a per-mm nr_cpus_allowed to keep track of the weight of the
per-mm CPUs allowed mask (for fast access),

- Add a per-mm max_nr_cid to keep track of the highest number of
concurrency IDs allocated for the mm. This is used for expanding the
concurrency ID allocation within the upper bound defined by:

min(mm->nr_cpus_allowed, mm->mm_users)

When the next unused CID value reaches this threshold, stop trying
to expand the cid allocation and use the first available cid value
instead.

Spreading allocation to use all the cid values within the range

[ 0, min(mm->nr_cpus_allowed, mm->mm_users) - 1 ]

improves cache locality while preserving mm_cid compactness within the
expected user limits,

- In __mm_cid_try_get, only return cid values within the range
[ 0, mm->nr_cpus_allowed ] rather than [ 0, nr_cpu_ids ]. This
prevents allocating cids above the number of allowed cpus in
rare scenarios where cid allocation races with a concurrent
remote-clear of the per-mm/cpu cid. This improvement is made
possible by the addition of the per-mm CPUs allowed mask,

- In sched_mm_cid_migrate_to, use mm->nr_cpus_allowed rather than
t->nr_cpus_allowed. This criterion was really meant to compare
the number of mm->mm_users to the number of CPUs allowed for the
entire mm. Therefore, the prior comparison worked fine when all
threads shared the same CPUs allowed mask, but not so much in
scenarios where those threads have different masks (e.g. each
thread pinned to a single CPU). This improvement is made
possible by the addition of the per-mm CPUs allowed mask.

* Benchmarks

Each thread increments 16kB worth of 8-bit integers in bursts, with
a configurable delay between each thread's execution. Each thread run
one after the other (no threads run concurrently). The order of
thread execution in the sequence is random. The thread execution
sequence begins again after all threads have executed. The 16kB areas
are allocated with rseq_mempool and indexed by either cpu_id, mm_cid
(not cache-local), or cache-local mm_cid. Each thread is pinned to its
own core.

Testing configurations:

8-core/1-L3: Use 8 cores within a single L3
24-core/24-L3: Use 24 cores, 1 core per L3
192-core/24-L3: Use 192 cores (all cores in the system)
384-thread/24-L3: Use 384 HW threads (all HW threads in the system)

Intermittent workload delays between threads: 200ms, 10ms.

Hardware:

CPU(s): 384
On-line CPU(s) list: 0-383
Vendor ID: AuthenticAMD
Model name: AMD EPYC 9654 96-Core Processor
Thread(s) per core: 2
Core(s) per socket: 96
Socket(s): 2
Caches (sum of all):
L1d: 6 MiB (192 instances)
L1i: 6 MiB (192 instances)
L2: 192 MiB (192 instances)
L3: 768 MiB (24 instances)

Each result is an average of 5 test runs. The cache-local speedup
is calculated as: (cache-local mm_cid) / (mm_cid).

Intermittent workload delay: 200ms

per-cpu mm_cid cache-local mm_cid cache-local speedup
(ns) (ns) (ns)
8-core/1-L3 1374 19289 1336 14.4x
24-core/24-L3 2423 26721 1594 16.7x
192-core/24-L3 2291 15826 2153 7.3x
384-thread/24-L3 1874 13234 1907 6.9x

Intermittent workload delay: 10ms

per-cpu mm_cid cache-local mm_cid cache-local speedup
(ns) (ns) (ns)
8-core/1-L3 662 756 686 1.1x
24-core/24-L3 1378 3648 1035 3.5x
192-core/24-L3 1439 10833 1482 7.3x
384-thread/24-L3 1503 10570 1556 6.8x

[ This deprecates the prior "sched: NUMA-aware per-memory-map concurrency IDs"
patch series with a simpler and more general approach. ]

[ This patch applies on top of v6.12-rc1. ]

Signed-off-by: Mathieu Desnoyers <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Acked-by: Marco Elver <[email protected]>
Link: https://lore.kernel.org/lkml/[email protected]/

show more ...

mm: remove PageSwapCache

This flag is now only used on folios, so we can remove all the page
accessors and reword the comments that refer to them.

Link: https://lkml.kernel.org/r/20240821193445.229

mm: remove PageSwapCache

This flag is now only used on folios, so we can remove all the page
accessors and reword the comments that refer to them.

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Matthew Wilcox (Oracle) <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>

show more ...

mm: add optional close() to struct vm_special_mapping

Add an optional close() callback to struct vm_special_mapping. It will be
used, by powerpc at least, to handle unmapping of the VDSO.

Although

mm: add optional close() to struct vm_special_mapping

Add an optional close() callback to struct vm_special_mapping. It will be
used, by powerpc at least, to handle unmapping of the VDSO.

Although support for unmapping the VDSO was initially added for CRIU[1],
it is not desirable to guard that support behind
CONFIG_CHECKPOINT_RESTORE.

There are other known users of unmapping the VDSO which are not related to
CRIU, eg. Valgrind [2] and void-ship [3].

The powerpc arch_unmap() hook has been in place for ~9 years, with no
ifdef, so there may be other unknown users that have come to rely on
unmapping the VDSO. Even if the code was behind an ifdef, major distros
enable CHECKPOINT_RESTORE so users may not realise unmapping the VDSO
depends on that configuration option.

It's also undesirable to have such core mm behaviour behind a relatively
obscure CONFIG option.

Longer term the unmap behaviour should be standardised across
architectures, however that is complicated by the fact the VDSO pointer is
stored differently across architectures. There was a previous attempt to
unify that handling [4], which could be revived.

See [5] for further discussion.

[1]: commit 83d3f0e90c6c ("powerpc/mm: tracking vDSO remap")
[2]: https://sourceware.org/git/?p=valgrind.git;a=commit;h=3a004915a2cbdcdebafc1612427576bf3321eef5
[3]: https://github.com/insanitybit/void-ship
[4]: https://lore.kernel.org/lkml/[email protected]/
[5]: https://lore.kernel.org/linuxppc-dev/shiq5v3jrmyi6ncwke7wgl76ojysgbhrchsk32q4lbx2hadqqc@kzyy2igem256

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Michael Ellerman <[email protected]>
Suggested-by: Linus Torvalds <[email protected]>
Reviewed-by: David Hildenbrand <[email protected]>
Reviewed-by: Liam R. Howlett <[email protected]>
Cc: Christophe Leroy <[email protected]>
Cc: Jeff Xu <[email protected]>
Cc: Nicholas Piggin <[email protected]>
Cc: Pedro Falcato <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>

show more ...

mm: fix (harmless) type confusion in lock_vma_under_rcu()

There is a (harmless) type confusion in lock_vma_under_rcu(): After
vma_start_read(), we have taken the VMA lock but don't know yet whether

mm: fix (harmless) type confusion in lock_vma_under_rcu()

There is a (harmless) type confusion in lock_vma_under_rcu(): After
vma_start_read(), we have taken the VMA lock but don't know yet whether
the VMA has already been detached and scheduled for RCU freeing. At this
point, ->vm_start and ->vm_end are accessed.

vm_area_struct contains a union such that ->vm_rcu uses the same memory as
->vm_start and ->vm_end; so accessing ->vm_start and ->vm_end of a
detached VMA is illegal and leads to type confusion between union members.

Fix it by reordering the vma->detached check above the address checks, and
document the rules for RCU readers accessing VMAs.

This will probably change the number of observed VMA_LOCK_MISS events
(since previously, trying to access a detached VMA whose ->vm_rcu has been
scheduled would bail out when checking the fault address against the
rcu_head members reinterpreted as VMA bounds).

Link: https://lkml.kernel.org/r/20240805-fix-vma-lock-type-confusion-v1-1-9f25443a9a71@google.com
Fixes: 50ee32537206 ("mm: introduce lock_vma_under_rcu to be used from arch-specific code")
Signed-off-by: Jann Horn <[email protected]>
Acked-by: Suren Baghdasaryan <[email protected]>
Cc: Matthew Wilcox <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>

show more ...

mm: turn USE_SPLIT_PTE_PTLOCKS / USE_SPLIT_PTE_PTLOCKS into Kconfig options

Patch series "mm: split PTE/PMD PT table Kconfig cleanups+clarifications".

This series is a follow up to the fixes:
"[PA

mm: turn USE_SPLIT_PTE_PTLOCKS / USE_SPLIT_PTE_PTLOCKS into Kconfig options

Patch series "mm: split PTE/PMD PT table Kconfig cleanups+clarifications".

This series is a follow up to the fixes:
"[PATCH v1 0/2] mm/hugetlb: fix hugetlb vs. core-mm PT locking"

When working on the fixes, I wondered why 8xx is fine (-> never uses split
PT locks) and how PT locking even works properly with PMD page table
sharing (-> always requires split PMD PT locks).

Let's improve the split PT lock detection, make hugetlb properly depend on
it and make 8xx bail out if it would ever get enabled by accident.

As an alternative to patch #3 we could extend the Kconfig
SPLIT_PTE_PTLOCKS option from patch #2 -- but enforcing it closer to the
code that actually implements it feels a bit nicer for documentation
purposes, and there is no need to actually disable it because it should
always be disabled (!SMP).

Did a bunch of cross-compilations to make sure that split PTE/PMD PT locks
are still getting used where we would expect them.

[1] https://lkml.kernel.org/r/[email protected]


This patch (of 3):

Let's clean that up a bit and prepare for depending on
CONFIG_SPLIT_PMD_PTLOCKS in other Kconfig options.

More cleanups would be reasonable (like the arch-specific "depends on" for
CONFIG_SPLIT_PTE_PTLOCKS), but we'll leave that for another day.

Link: https://lkml.kernel.org/r/[email protected]
Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: David Hildenbrand <[email protected]>
Acked-by: Mike Rapoport (Microsoft) <[email protected]>
Reviewed-by: Russell King (Oracle) <[email protected]>
Reviewed-by: Qi Zheng <[email protected]>
Cc: Alexander Viro <[email protected]>
Cc: Borislav Petkov <[email protected]>
Cc: Boris Ostrovsky <[email protected]>
Cc: Christian Brauner <[email protected]>
Cc: Christophe Leroy <[email protected]>
Cc: Dave Hansen <[email protected]>
Cc: "H. Peter Anvin" <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Juergen Gross <[email protected]>
Cc: Michael Ellerman <[email protected]>
Cc: Muchun Song <[email protected]>
Cc: "Naveen N. Rao" <[email protected]>
Cc: Nicholas Piggin <[email protected]>
Cc: Oscar Salvador <[email protected]>
Cc: Peter Xu <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>

show more ...