userfaultfd: convert mfill_atomic_hugetlb() to use a folio

Convert hugetlb_mfill_atomic_pte() to take in a folio pointer instead of
a page pointer.

Convert mfill_atomic_hugetlb() to use a folio.

Link: https://lkml.kernel.org/r/20230410133932.32288-5-zhangpeng362@huawei.com
Signed-off-by: ZhangPeng <zhangpeng362@huawei.com>
Reviewed-by: Sidhartha Kumar <sidhartha.kumar@oracle.com>
Reviewed-by: Mike Kravetz <mike.kravetz@oracle.com>
Cc: Kefeng Wang <wangkefeng.wang@huawei.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Muchun Song <muchun.song@linux.dev>
Cc: Nanyong Sun <sunnanyong@huawei.com>
Cc: Vishal Moola (Oracle) <vishal.moola@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

userfaultfd: convert copy_huge_page_from_user() to copy_folio_from_user()

Replace copy_huge_page_from_user() with copy_folio_from_user().
copy_folio_from_user() does the same as copy_huge_page_from_user(), but
takes in a folio instead of a page.

Convert page_kaddr to kaddr in copy_folio_from_user() to do indenting
cleanup.

Link: https://lkml.kernel.org/r/20230410133932.32288-4-zhangpeng362@huawei.com
Signed-off-by: ZhangPeng <zhangpeng362@huawei.com>
Reviewed-by: Sidhartha Kumar <sidhartha.kumar@oracle.com>
Reviewed-by: Mike Kravetz <mike.kravetz@oracle.com>
Cc: Kefeng Wang <wangkefeng.wang@huawei.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Muchun Song <muchun.song@linux.dev>
Cc: Nanyong Sun <sunnanyong@huawei.com>
Cc: Vishal Moola (Oracle) <vishal.moola@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

userfaultfd: use kmap_local_page() in copy_huge_page_from_user()

kmap() and kmap_atomic() are being deprecated in favor of
kmap_local_page() which is appropriate for any thread local context.[1]

Let's replace the kmap() and kmap_atomic() with kmap_local_page() in
copy_huge_page_from_user().  When allow_pagefault is false, disable page
faults to prevent potential deadlock.[2]

[1] https://lore.kernel.org/all/20220813220034.806698-1-ira.weiny@intel.com/
[2] https://lkml.kernel.org/r/20221025220136.2366143-1-ira.weiny@intel.com

Link: https://lkml.kernel.org/r/20230410133932.32288-3-zhangpeng362@huawei.com
Signed-off-by: ZhangPeng <zhangpeng362@huawei.com>
Reviewed-by: Mike Kravetz <mike.kravetz@oracle.com>
Cc: Kefeng Wang <wangkefeng.wang@huawei.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Muchun Song <muchun.song@linux.dev>
Cc: Nanyong Sun <sunnanyong@huawei.com>
Cc: Sidhartha Kumar <sidhartha.kumar@oracle.com>
Cc: Vishal Moola (Oracle) <vishal.moola@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

userfaultfd: convert mfill_atomic_pte_copy() to use a folio

Patch series "userfaultfd: convert userfaultfd functions to use folios",
v6.

This patch series converts several userfaultfd functions to use folios.

This patch (of 6):

Call vma_alloc_folio() directly instead of alloc_page_vma() and convert
page_kaddr to kaddr in mfill_atomic_pte_copy().  Removes several calls to
compound_head().

Link: https://lkml.kernel.org/r/20230410133932.32288-1-zhangpeng362@huawei.com
Link: https://lkml.kernel.org/r/20230410133932.32288-2-zhangpeng362@huawei.com
Signed-off-by: ZhangPeng <zhangpeng362@huawei.com>
Reviewed-by: Sidhartha Kumar <sidhartha.kumar@oracle.com>
Reviewed-by: Mike Kravetz <mike.kravetz@oracle.com>
Cc: Kefeng Wang <wangkefeng.wang@huawei.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Muchun Song <muchun.song@linux.dev>
Cc: Nanyong Sun <sunnanyong@huawei.com>
Cc: Vishal Moola (Oracle) <vishal.moola@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

smaps: fix defined but not used smaps_shmem_walk_ops

When !CONFIG_SHMEM smaps_shmem_walk_ops is defined but not used,
triggering a compiler warning.  To avoid the warning remove the #ifdef
around the usage.  This has no effect because shmem_mapping() is a stub
returning false when !CONFIG_SHMEM so the code will be compiled out,
however we now need to also provide a stub for shmem_swap_usage().

Link: https://lkml.kernel.org/r/20230405103819.151246-1-steven.price@arm.com
Fixes: 7b86ac3371b7 ("pagewalk: separate function pointers from iterator data")
Signed-off-by: Steven Price <steven.price@arm.com>
Reported-by: kernel test robot <lkp@intel.com>
Link: https://lore.kernel.org/oe-kbuild-all/202304031749.UiyJpxzF-lkp@intel.com/
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm, page_alloc: use check_pages_enabled static key to check tail pages

Commit 700d2e9a36b9 ("mm, page_alloc: reduce page alloc/free sanity
checks") has introduced a new static key check_pages_enabled to control
when struct pages are sanity checked during allocation and freeing.  Mel
Gorman suggested that free_tail_pages_check() could use this static key as
well, instead of relying on CONFIG_DEBUG_VM.  That makes sense, so do
that.  Also rename the function to free_tail_page_prepare() because it
works on a single tail page and has a struct page preparation component as
well as the optional checking component.

Also remove some unnecessary unlikely() within static_branch_unlikely()
statements that Mel pointed out for commit 700d2e9a36b9.

Link: https://lkml.kernel.org/r/20230405142840.11068-1-vbabka@suse.cz
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Suggested-by: Mel Gorman <mgorman@techsingularity.net>
Acked-by: Mel Gorman <mgorman@techsingularity.net>
Cc: Alexander Halbuer <halbuer@sra.uni-hannover.de>
Cc: Kees Cook <keescook@chromium.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm/userfaultfd: don't consider uffd-wp bit of writable migration entries

If we end up with a writable migration entry that has the uffd-wp bit set,
we already messed up: the source PTE/PMD was writable, which means we
could have modified the page without notifying uffd first.  Setting the
uffd-wp bit always implies converting migration entries to !writable
migration entries.

Commit 8f34f1eac382 ("mm/userfaultfd: fix uffd-wp special cases for
fork()") documents that "3.  Forget to carry over uffd-wp bit for a write
migration huge pmd entry", but it doesn't really say why that should be
relevant.

So let's remove that code to avoid hiding an eventual underlying issue (in
the future, we might want to warn when creating writable migration entries
that have the uffd-wp bit set -- or even better when turning a PTE
writable that still has the uffd-wp bit set).

This now matches the handling for hugetlb migration entries in
hugetlb_change_protection().

In copy_huge_pmd()/copy_nonpresent_pte()/copy_hugetlb_page_range(), we
still transfer the uffd-bit also for writable migration entries, but
simply because we have unified handling for "writable" and
"readable-exclusive" migration entries, and we care about transferring the
uffd-wp bit for the latter.

Link: https://lkml.kernel.org/r/20230405160236.587705-3-david@redhat.com
Signed-off-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Cc: Muhammad Usama Anjum <usama.anjum@collabora.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm: mlock: use folios_put() in mlock_folio_batch()

Since we have updated mlock to use folios, it's better to call
folios_put() instead of calling release_pages() directly.

Link: https://lkml.kernel.org/r/20230405161854.6931-2-zhengqi.arch@bytedance.com
Signed-off-by: Qi Zheng <zhengqi.arch@bytedance.com>
Acked-by: Mel Gorman <mgorman@suse.de>
Cc: Lorenzo Stoakes <lstoakes@gmail.com>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

prctl: add PR_GET_AUXV to copy auxv to userspace

If a library wants to get information from auxv (for instance,
AT_HWCAP/AT_HWCAP2), it has a few options, none of them perfectly reliable
or ideal:

- Be main or the pre-main startup code, and grub through the stack above
  main. Doesn't work for a library.
- Call libc getauxval. Not ideal for libraries that are trying to be
  libc-independent and/or don't otherwise require anything from other
  libraries.
- Open and read /proc/self/auxv. Doesn't work for libraries that may run
  in arbitrarily constrained environments that may not have /proc
  mounted (e.g. libraries that might be used by an init program or a
  container setup tool).
- Assume you're on the main thread and still on the original stack, and
  try to walk the stack upwards, hoping to find auxv. Extremely bad
  idea.
- Ask the caller to pass auxv in for you. Not ideal for a user-friendly
  library, and then your caller may have the same problem.

Add a prctl that copies current->mm->saved_auxv to a userspace buffer.

Link: https://lkml.kernel.org/r/d81864a7f7f43bca6afa2a09fc2e850e4050ab42.1680611394.git.josh@joshtriplett.org
Signed-off-by: Josh Triplett <josh@joshtriplett.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

maple_tree: simplify mas_wr_node_walk()

Simplify code of mas_wr_node_walk() without changing functionality, and
improve readability.  Remove some special judgments.  Instead of
dynamically recording the min and max in the loop, get the final min and
max directly at the end.

Link: https://lkml.kernel.org/r/20230314124203.91572-3-zhangpeng.00@bytedance.com
Signed-off-by: Peng Zhang <zhangpeng.00@bytedance.com>
Reviewed-by: Liam R. Howlett <Liam.Howlett@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

m68k/mm: use correct bit number in _PAGE_SWP_EXCLUSIVE comment

As noticed by Geert, commit b5c88f21531c ("microblaze/mm: support
__HAVE_ARCH_PTE_SWP_EXCLUSIVE") modified m68k code by accident.  While
replacing 0x080 by CF_PAGE_NOCACHE is correct, although it should have
been part of commit ed4154067a08 ("m68k/mm: support
__HAVE_ARCH_PTE_SWP_EXCLUSIVE"), replacing "bit 7" by "bit 24" in the
comment was wrong.

Let's revert to the previous, correct, comment.

Link: https://lkml.kernel.org/r/20230404085636.121409-1-david@redhat.com
Signed-off-by: David Hildenbrand <david@redhat.com>
Reported-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm/madvise: use vma_lookup() instead of find_vma()

Using vma_lookup() verifies the address is contained in the found vma.
This results in easier to read the code.

Link: https://lkml.kernel.org/r/20230404094515.1883552-1-zhangpeng362@huawei.com
Signed-off-by: ZhangPeng <zhangpeng362@huawei.com>
Reviewed-by: Liam R. Howlett <Liam.Howlett@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

memcg v1: provide read access to memory.pressure_level

cgroups v1 has a unique way of setting up memory pressure notifications:
the user opens "memory.pressure_level" of the cgroup they want to monitor
for pressure, then open "cgroup.event_control" and write the fd (among
other things) to that file.  memory.pressure_level has no other use,
specifically it does not support any read or write operations.
Consequently, no handlers are provided, and cgroup_file_mode() sets the
permissions to 000.  However, to actually use the mechanism, the
subscribing user must have read access to the file and open the fd for
reading, see memcg_write_event_control().

This is all fine as long as the subscribing process runs as root and is
otherwise unconfined by further restrictions.  However, if you add strict
access controls such as selinux, the permission bits will be enforced, and
opening memory.pressure_level for reading will fail, preventing the
process from subscribing, even as root.

To work around this issue, introduce a dummy read handler.  When
memory.pressure_level is created, cgroup_file_mode() will notice the
existence of a handler, and therefore add read permissions to the file.

Link: https://lkml.kernel.org/r/20230404105900.2005-1-flosch@nutanix.com
Signed-off-by: Florian Schmidt <flosch@nutanix.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Muchun Song <muchun.song@linux.dev>
Cc: Roman Gushchin <roman.gushchin@linux.dev>
Cc: Shakeel Butt <shakeelb@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm/khugepaged: maintain page cache uptodate flag

Make sure that collapse_file doesn't interfere with checking the uptodate
flag in the page cache by only inserting hpage into the page cache after
it has been updated and marked uptodate.  This is achieved by simply not
replacing present pages with hpage when iterating over the target range.

The present pages are already locked, so replacing them with the locked
hpage before the collapse is finalized is unnecessary.  However, it is
necessary to stop freezing the present pages after validating them, since
leaving long-term frozen pages in the page cache can lead to deadlocks.
Simply checking the reference count is sufficient to ensure that there are
no long-term references hanging around that would the collapse would
break.  Similar to hpage, there is no reason that the present pages
actually need to be frozen in addition to being locked.

This fixes a race where folio_seek_hole_data would mistake hpage for an
fallocated but unwritten page.  This race is visible to userspace via data
temporarily disappearing from SEEK_DATA/SEEK_HOLE.  This also fixes a
similar race where pages could temporarily disappear from mincore.

Link: https://lkml.kernel.org/r/20230404120117.2562166-5-stevensd@google.com
Fixes: f3f0e1d2150b ("khugepaged: add support of collapse for tmpfs/shmem pages")
Signed-off-by: David Stevens <stevensd@chromium.org>
Cc: David Hildenbrand <david@redhat.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Jiaqi Yan <jiaqiyan@google.com>
Cc: "Kirill A. Shutemov" <kirill@shutemov.name>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: Peter Xu <peterx@redhat.com>
Cc: Yang Shi <shy828301@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm/khugepaged: skip shmem with userfaultfd

Make sure that collapse_file respects any userfaultfds registered with
MODE_MISSING.  If userspace has any such userfaultfds registered, then for
any page which it knows to be missing, it may expect a
UFFD_EVENT_PAGEFAULT.  This means collapse_file needs to be careful when
collapsing a shmem range would result in replacing an empty page with a
THP, to avoid breaking userfaultfd.

Synchronization when checking for userfaultfds in collapse_file is tricky
because the mmap locks can't be used to prevent races with the
registration of new userfaultfds.  Instead, we provide synchronization by
ensuring that userspace cannot observe the fact that pages are missing
before we check for userfaultfds.  Although this allows registration of a
userfaultfd to race with collapse_file, it ensures that userspace cannot
observe any pages transition from missing to present after such a race
occurs.  This makes such a race indistinguishable to the collapse
occurring immediately before the userfaultfd registration.

The first step to provide this synchronization is to stop filling gaps
during the loop iterating over the target range, since the page cache lock
can be dropped during that loop.  The second step is to fill the gaps with
XA_RETRY_ENTRY after the page cache lock is acquired the final time, to
avoid races with accesses to the page cache that only take the RCU read
lock.

The fact that we don't fill holes during the initial iteration means that
collapse_file now has to handle faults occurring during the collapse.
This is done by re-validating the number of missing pages after acquiring
the page cache lock for the final time.

This fix is targeted at khugepaged, but the change also applies to
MADV_COLLAPSE.  MADV_COLLAPSE on a range with a userfaultfd will now
return EBUSY if there are any missing pages (instead of succeeding on
shmem and returning EINVAL on anonymous memory).  There is also now a
window during MADV_COLLAPSE where a fault on a missing page will cause the
syscall to fail with EAGAIN.

The fact that intermediate page cache state can no longer be observed
before the rollback of a failed collapse is also technically a
userspace-visible change (via at least SEEK_DATA and SEEK_END), but it is
exceedingly unlikely that anything relies on being able to observe that
transient state.

Link: https://lkml.kernel.org/r/20230404120117.2562166-4-stevensd@google.com
Signed-off-by: David Stevens <stevensd@chromium.org>
Acked-by: Peter Xu <peterx@redhat.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Jiaqi Yan <jiaqiyan@google.com>
Cc: "Kirill A. Shutemov" <kirill@shutemov.name>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: Yang Shi <shy828301@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm/khugepaged: refactor collapse_file control flow

Add a rollback label to deal with failure, instead of continuously
checking for RESULT_SUCCESS, to make it easier to add more failure cases.
The refactoring also allows the collapse_file tracepoint to include hpage
on success (instead of NULL).

Link: https://lkml.kernel.org/r/20230404120117.2562166-3-stevensd@google.com
Signed-off-by: David Stevens <stevensd@chromium.org>
Acked-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Yang Shi <shy828301@gmail.com>
Acked-by: Hugh Dickins <hughd@google.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Jiaqi Yan <jiaqiyan@google.com>
Cc: "Kirill A. Shutemov" <kirill@shutemov.name>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm/khugepaged: drain lru after swapping in shmem

Patch series "mm/khugepaged: fixes for khugepaged+shmem", v6.

This series reworks collapse_file so that the intermediate state of the
collapse does not leak out of collapse_file. Although this makes
collapse_file a bit more complicated, it means that the rest of the
kernel doesn't have to deal with the unusual state. This directly fixes
races with both lseek and mincore.

This series also fixes the fact that khugepaged completely breaks
userfaultfd+shmem. The rework of collapse_file provides a convenient
place to check for registered userfaultfds without making the shmem
userfaultfd implementation care about khugepaged.

Finally, this series adds a lru_add_drain after swapping in shmem pages,
which makes the subsequent folio_isolate_lru significantly more likely to
succeed.

This patch (of 4):

Call lru_add_drain after swapping in shmem pages so that isolate_lru_page
is more likely to succeed.

Link: https://lkml.kernel.org/r/20230404120117.2562166-1-stevensd@google.com
Link: https://lkml.kernel.org/r/20230404120117.2562166-2-stevensd@google.com
Signed-off-by: David Stevens <stevensd@chromium.org>
Cc: David Hildenbrand <david@redhat.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Jiaqi Yan <jiaqiyan@google.com>
Cc: "Kirill A. Shutemov" <kirill@shutemov.name>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: Peter Xu <peterx@redhat.com>
Cc: Yang Shi <shy828301@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm/khugepaged: recover from poisoned file-backed memory

Make collapse_file roll back when copying pages failed. More concretely:
- extract copying operations into a separate loop
- postpone the updates for nr_none until both scanning and copying
  succeeded
- postpone joining small xarray entries until both scanning and copying
  succeeded
- postpone the update operations to NR_XXX_THPS until both scanning and
  copying succeeded
- for non-SHMEM file, roll back filemap_nr_thps_inc if scan succeeded but
  copying failed

Tested manually:
0. Enable khugepaged on system under test. Mount tmpfs at /mnt/ramdisk.
1. Start a two-thread application. Each thread allocates a chunk of
   non-huge memory buffer from /mnt/ramdisk.
2. Pick 4 random buffer address (2 in each thread) and inject
   uncorrectable memory errors at physical addresses.
3. Signal both threads to make their memory buffer collapsible, i.e.
   calling madvise(MADV_HUGEPAGE).
4. Wait and then check kernel log: khugepaged is able to recover from
   poisoned pages by skipping them.
5. Signal both threads to inspect their buffer contents and make sure no
   data corruption.

Link: https://lkml.kernel.org/r/20230329151121.949896-4-jiaqiyan@google.com
Signed-off-by: Jiaqi Yan <jiaqiyan@google.com>
Reviewed-by: Yang Shi <shy828301@gmail.com>
Acked-by: Hugh Dickins <hughd@google.com>
Cc: David Stevens <stevensd@chromium.org>
Cc: Kefeng Wang <wangkefeng.wang@huawei.com>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: "Kirill A. Shutemov" <kirill@shutemov.name>
Cc: Miaohe Lin <linmiaohe@huawei.com>
Cc: Naoya Horiguchi <naoya.horiguchi@nec.com>
Cc: Oscar Salvador <osalvador@suse.de>
Cc: Tong Tiangen <tongtiangen@huawei.com>
Cc: Tony Luck <tony.luck@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm/hwpoison: introduce copy_mc_highpage

Similar to how copy_mc_user_highpage is implemented for copy_user_highpage
on #MC supported architecture, introduce the #MC handled version of
copy_highpage.

This helper has immediate usage when khugepaged wants to copy file-backed
memory pages and tolerate #MC.

Link: https://lkml.kernel.org/r/20230329151121.949896-3-jiaqiyan@google.com
Signed-off-by: Jiaqi Yan <jiaqiyan@google.com>
Reviewed-by: Yang Shi <shy828301@gmail.com>
Cc: David Stevens <stevensd@chromium.org>
Cc: Hugh Dickins <hughd@google.com>
Cc: Kefeng Wang <wangkefeng.wang@huawei.com>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: "Kirill A. Shutemov" <kirill@shutemov.name>
Cc: Miaohe Lin <linmiaohe@huawei.com>
Cc: Naoya Horiguchi <naoya.horiguchi@nec.com>
Cc: Oscar Salvador <osalvador@suse.de>
Cc: Tong Tiangen <tongtiangen@huawei.com>
Cc: Tony Luck <tony.luck@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm/khugepaged: recover from poisoned anonymous memory

Problem
=======
Memory DIMMs are subject to multi-bit flips, i.e.  memory errors.  As
memory size and density increase, the chances of and number of memory
errors increase.  The increasing size and density of server RAM in the
data center and cloud have shown increased uncorrectable memory errors.
There are already mechanisms in the kernel to recover from uncorrectable
memory errors.  This series of patches provides the recovery mechanism for
the particular kernel agent khugepaged when it collapses memory pages.

Impact
======
The main reason we chose to make khugepaged collapsing tolerant of memory
failures was its high possibility of accessing poisoned memory while
performing functionally optional compaction actions.  Standard
applications typically don't have strict requirements on the size of its
pages.  So they are given 4K pages by the kernel.  The kernel is able to
improve application performance by either

  1) giving applications 2M pages to begin with, or
  2) collapsing 4K pages into 2M pages when possible.

This collapsing operation is done by khugepaged, a kernel agent that is
constantly scanning memory.  When collapsing 4K pages into a 2M page, it
must copy the data from the 4K pages into a physically contiguous 2M page.
Therefore, as long as there exists one poisoned cache line in collapsible
4K pages, khugepaged will eventually access it.  The current impact to
users is a machine check exception triggered kernel panic.  However,
khugepaged’s compaction operations are not functionally required kernel
actions.  Therefore making khugepaged tolerant to poisoned memory will
greatly improve user experience.

This patch series is for cases where khugepaged is the first guy that
detects the memory errors on the poisoned pages.  IOW, the pages are not
known to have memory errors when khugepaged collapsing gets to them.  In
our observation, this happens frequently when the huge page ratio of the
system is relatively low, which is fairly common in virtual machines
running on cloud.

Solution
========
As stated before, it is less desirable to crash the system only because
khugepaged accesses poisoned pages while it is collapsing 4K pages.  The
high level idea of this patch series is to skip the group of pages
(usually 512 4K-size pages) once khugepaged finds one of them is poisoned,
as these pages have become ineligible to be collapsed.

We are also careful to unwind operations khuagepaged has performed before
it detects memory failures.  For example, before copying and collapsing a
group of anonymous pages into a huge page, the source pages will be
isolated and their page table is unlinked from their PMD.  These
operations need to be undone in order to ensure these pages are not
changed/lost from the perspective of other threads (both user and kernel
space).  As for file backed memory pages, there already exists a rollback
case.  This patch just extends it so that khugepaged also correctly rolls
back when it fails to copy poisoned 4K pages.

This patch (of 3):

Make __collapse_huge_page_copy return whether copying anonymous pages
succeeded, and make collapse_huge_page handle the return status.

Break existing PTE scan loop into two for-loops.  The first loop copies
source pages into target huge page, and can fail gracefully when running
into memory errors in source pages.  If copying all pages succeeds, the
second loop releases and clears up these normal pages.  Otherwise, the
second loop rolls back the page table and page states by:

- re-establishing the original PTEs-to-PMD connection.
- releasing source pages back to their LRU list.

Tested manually:
0. Enable khugepaged on system under test.
1. Start a two-thread application. Each thread allocates a chunk of
   non-huge anonymous memory buffer.
2. Pick 4 random buffer locations (2 in each thread) and inject
   uncorrectable memory errors at corresponding physical addresses.
3. Signal both threads to make their memory buffer collapsible, i.e.
   calling madvise(MADV_HUGEPAGE).
4. Wait and check kernel log: khugepaged is able to recover from poisoned
   pages and skips collapsing them.
5. Signal both threads to inspect their buffer contents and make sure no
   data corruption.

Link: https://lkml.kernel.org/r/20230329151121.949896-1-jiaqiyan@google.com
Link: https://lkml.kernel.org/r/20230329151121.949896-2-jiaqiyan@google.com
Signed-off-by: Jiaqi Yan <jiaqiyan@google.com>
Cc: David Stevens <stevensd@chromium.org>
Cc: Hugh Dickins <hughd@google.com>
Cc: Kefeng Wang <wangkefeng.wang@huawei.com>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: "Kirill A. Shutemov" <kirill@shutemov.name>
Cc: Miaohe Lin <linmiaohe@huawei.com>
Cc: Naoya Horiguchi <naoya.horiguchi@nec.com>
Cc: Oscar Salvador <osalvador@suse.de>
Cc: Tong Tiangen <tongtiangen@huawei.com>
Cc: Tony Luck <tony.luck@intel.com>
Cc: Yang Shi <shy828301@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

memcg: do not modify rstat tree for zero updates

In some situations, we may end up calling memcg_rstat_updated() with a
value of 0, which means the stat was not actually updated.  An example is
if we fail to reclaim any pages in shrink_folio_list().

Do not add the cgroup to the rstat updated tree in this case, to avoid
unnecessarily flushing it.

Link: https://lkml.kernel.org/r/20230330191801.1967435-9-yosryahmed@google.com
Signed-off-by: Yosry Ahmed <yosryahmed@google.com>
Acked-by: Shakeel Butt <shakeelb@google.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Michal Hocko <mhocko@suse.com>
Reviewed-by: Michal Koutný <mkoutny@suse.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Josef Bacik <josef@toxicpanda.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Muchun Song <muchun.song@linux.dev>
Cc: Roman Gushchin <roman.gushchin@linux.dev>
Cc: Tejun Heo <tj@kernel.org>
Cc: Vasily Averin <vasily.averin@linux.dev>
Cc: Zefan Li <lizefan.x@bytedance.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

vmscan: memcg: sleep when flushing stats during reclaim

Memory reclaim is a sleepable context.  Flushing is an expensive operaiton
that scales with the number of cpus and the number of cgroups in the
system, so avoid doing it atomically unnecessarily.  This can slow down
reclaim code if flushing stats is taking too long, but there is already
multiple cond_resched()'s in reclaim code.

Link: https://lkml.kernel.org/r/20230330191801.1967435-8-yosryahmed@google.com
Signed-off-by: Yosry Ahmed <yosryahmed@google.com>
Acked-by: Shakeel Butt <shakeelb@google.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Josef Bacik <josef@toxicpanda.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Michal Koutný <mkoutny@suse.com>
Cc: Muchun Song <muchun.song@linux.dev>
Cc: Roman Gushchin <roman.gushchin@linux.dev>
Cc: Tejun Heo <tj@kernel.org>
Cc: Vasily Averin <vasily.averin@linux.dev>
Cc: Zefan Li <lizefan.x@bytedance.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

workingset: memcg: sleep when flushing stats in workingset_refault()

In workingset_refault(), we call
mem_cgroup_flush_stats_atomic_ratelimited() to read accurate stats within
an RCU read section and with sleeping disallowed.  Move the call above the
RCU read section to make it non-atomic.

Flushing is an expensive operation that scales with the number of cpus and
the number of cgroups in the system, so avoid doing it atomically where
possible.

Since workingset_refault() is the only caller of
mem_cgroup_flush_stats_atomic_ratelimited(), just make it non-atomic, and
rename it to mem_cgroup_flush_stats_ratelimited().

Link: https://lkml.kernel.org/r/20230330191801.1967435-7-yosryahmed@google.com
Signed-off-by: Yosry Ahmed <yosryahmed@google.com>
Acked-by: Shakeel Butt <shakeelb@google.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Josef Bacik <josef@toxicpanda.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Michal Koutný <mkoutny@suse.com>
Cc: Muchun Song <muchun.song@linux.dev>
Cc: Roman Gushchin <roman.gushchin@linux.dev>
Cc: Tejun Heo <tj@kernel.org>
Cc: Vasily Averin <vasily.averin@linux.dev>
Cc: Zefan Li <lizefan.x@bytedance.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

memcg: sleep during flushing stats in safe contexts

Currently, all contexts that flush memcg stats do so with sleeping not
allowed.  Some of these contexts are perfectly safe to sleep in, such as
reading cgroup files from userspace or the background periodic flusher.
Flushing is an expensive operation that scales with the number of cpus and
the number of cgroups in the system, so avoid doing it atomically where
possible.

Refactor the code to make mem_cgroup_flush_stats() non-atomic (aka
sleepable), and provide a separate atomic version.  The atomic version is
used in reclaim, refault, writeback, and in mem_cgroup_usage().  All other
code paths are left to use the non-atomic version.  This includes
callbacks for userspace reads and the periodic flusher.

Since refault is the only caller of mem_cgroup_flush_stats_ratelimited(),
change it to mem_cgroup_flush_stats_atomic_ratelimited().  Reclaim and
refault code paths are modified to do non-atomic flushing in separate
later patches -- so it will eventually be changed back to
mem_cgroup_flush_stats_ratelimited().

Link: https://lkml.kernel.org/r/20230330191801.1967435-6-yosryahmed@google.com
Signed-off-by: Yosry Ahmed <yosryahmed@google.com>
Acked-by: Shakeel Butt <shakeelb@google.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Josef Bacik <josef@toxicpanda.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Michal Koutný <mkoutny@suse.com>
Cc: Muchun Song <muchun.song@linux.dev>
Cc: Roman Gushchin <roman.gushchin@linux.dev>
Cc: Tejun Heo <tj@kernel.org>
Cc: Vasily Averin <vasily.averin@linux.dev>
Cc: Zefan Li <lizefan.x@bytedance.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

memcg: replace stats_flush_lock with an atomic

As Johannes notes in [1], stats_flush_lock is currently used to:
(a) Protect updated to stats_flush_threshold.
(b) Protect updates to flush_next_time.
(c) Serializes calls to cgroup_rstat_flush() based on those ratelimits.

However:

1. stats_flush_threshold is already an atomic

2. flush_next_time is not atomic. The writer is locked, but the reader
   is lockless. If the reader races with a flush, you could see this:

                                        if (time_after(jiffies, flush_next_time))
        spin_trylock()
        flush_next_time = now + delay
        flush()
        spin_unlock()
                                        spin_trylock()
                                        flush_next_time = now + delay
                                        flush()
                                        spin_unlock()

   which means we already can get flushes at a higher frequency than
   FLUSH_TIME during races. But it isn't really a problem.

   The reader could also see garbled partial updates if the compiler
   decides to split the write, so it needs at least READ_ONCE and
   WRITE_ONCE protection.

3. Serializing cgroup_rstat_flush() calls against the ratelimit
   factors is currently broken because of the race in 2. But the race
   is actually harmless, all we might get is the occasional earlier
   flush. If there is no delta, the flush won't do much. And if there
   is, the flush is justified.

So the lock can be removed all together. However, the lock also served
the purpose of preventing a thundering herd problem for concurrent
flushers, see [2]. Use an atomic instead to serve the purpose of
unifying concurrent flushers.

[1]https://lore.kernel.org/lkml/20230323172732.GE739026@cmpxchg.org/
[2]https://lore.kernel.org/lkml/20210716212137.1391164-2-shakeelb@google.com/

Link: https://lkml.kernel.org/r/20230330191801.1967435-5-yosryahmed@google.com
Signed-off-by: Yosry Ahmed <yosryahmed@google.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Shakeel Butt <shakeelb@google.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Josef Bacik <josef@toxicpanda.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Michal Koutný <mkoutny@suse.com>
Cc: Muchun Song <muchun.song@linux.dev>
Cc: Roman Gushchin <roman.gushchin@linux.dev>
Cc: Tejun Heo <tj@kernel.org>
Cc: Vasily Averin <vasily.averin@linux.dev>
Cc: Zefan Li <lizefan.x@bytedance.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

memcg: do not flush stats in irq context

Currently, the only context in which we can invoke an rstat flush from irq
context is through mem_cgroup_usage() on the root memcg when called from
memcg_check_events().  An rstat flush is an expensive operation that
should not be done in irq context, so do not flush stats and use the stale
stats in this case.

Arguably, usage threshold events are not reliable on the root memcg anyway
since its usage is ill-defined.

Link: https://lkml.kernel.org/r/20230330191801.1967435-4-yosryahmed@google.com
Signed-off-by: Yosry Ahmed <yosryahmed@google.com>
Suggested-by: Johannes Weiner <hannes@cmpxchg.org>
Suggested-by: Shakeel Butt <shakeelb@google.com>
Acked-by: Shakeel Butt <shakeelb@google.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Josef Bacik <josef@toxicpanda.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Michal Koutný <mkoutny@suse.com>
Cc: Muchun Song <muchun.song@linux.dev>
Cc: Roman Gushchin <roman.gushchin@linux.dev>
Cc: Tejun Heo <tj@kernel.org>
Cc: Vasily Averin <vasily.averin@linux.dev>
Cc: Zefan Li <lizefan.x@bytedance.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

memcg: rename mem_cgroup_flush_stats_"delayed" to "ratelimited"

mem_cgroup_flush_stats_delayed() suggests his is using a delayed_work, but
this is actually sometimes flushing directly from the callsite.

What it's doing is ratelimited calls.  A better name would be
mem_cgroup_flush_stats_ratelimited().

Link: https://lkml.kernel.org/r/20230330191801.1967435-3-yosryahmed@google.com
Signed-off-by: Yosry Ahmed <yosryahmed@google.com>
Suggested-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Shakeel Butt <shakeelb@google.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Josef Bacik <josef@toxicpanda.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Michal Koutný <mkoutny@suse.com>
Cc: Muchun Song <muchun.song@linux.dev>
Cc: Roman Gushchin <roman.gushchin@linux.dev>
Cc: Tejun Heo <tj@kernel.org>
Cc: Vasily Averin <vasily.averin@linux.dev>
Cc: Zefan Li <lizefan.x@bytedance.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

cgroup: rename cgroup_rstat_flush_"irqsafe" to "atomic"

Patch series "memcg: avoid flushing stats atomically where possible", v3.

rstat flushing is an expensive operation that scales with the number of
cpus and the number of cgroups in the system.  The purpose of this series
is to minimize the contexts where we flush stats atomically.

Patches 1 and 2 are cleanups requested during reviews of prior versions of
this series.

Patch 3 makes sure we never try to flush from within an irq context.

Patches 4 to 7 introduce separate variants of mem_cgroup_flush_stats() for
atomic and non-atomic flushing, and make sure we only flush the stats
atomically when necessary.

Patch 8 is a slightly tangential optimization that limits the work done by
rstat flushing in some scenarios.

This patch (of 8):

cgroup_rstat_flush_irqsafe() can be a confusing name.  It may read as
"irqs are disabled throughout", which is what the current implementation
does (currently under discussion [1]), but is not the intention.  The
intention is that this function is safe to call from atomic contexts.
Name it as such.

Link: https://lkml.kernel.org/r/20230330191801.1967435-1-yosryahmed@google.com
Link: https://lkml.kernel.org/r/20230330191801.1967435-2-yosryahmed@google.com
Signed-off-by: Yosry Ahmed <yosryahmed@google.com>
Suggested-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Shakeel Butt <shakeelb@google.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Josef Bacik <josef@toxicpanda.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Michal Koutný <mkoutny@suse.com>
Cc: Muchun Song <muchun.song@linux.dev>
Cc: Roman Gushchin <roman.gushchin@linux.dev>
Cc: Tejun Heo <tj@kernel.org>
Cc: Vasily Averin <vasily.averin@linux.dev>
Cc: Zefan Li <lizefan.x@bytedance.com>
Cc: Michal Hocko <mhocko@suse.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm: kfence: improve the performance of __kfence_alloc() and __kfence_free()

In __kfence_alloc() and __kfence_free(), we will set and check canary.
Assuming that the size of the object is close to 0, nearly 4k memory
accesses are required because setting and checking canary is executed byte
by byte.

canary is now defined like this:
KFENCE_CANARY_PATTERN(addr) ((u8)0xaa ^ (u8)((unsigned long)(addr) & 0x7))

Observe that canary is only related to the lower three bits of the
address, so every 8 bytes of canary are the same.  We can access 8-byte
canary each time instead of byte-by-byte, thereby optimizing nearly 4k
memory accesses to 4k/8 times.

Use the bcc tool funclatency to measure the latency of __kfence_alloc()
and __kfence_free(), the numbers (deleted the distribution of latency) is
posted below.  Though different object sizes will have an impact on the
measurement, we ignore it for now and assume the average object size is
roughly equal.

Before patching:
__kfence_alloc:
avg = 5055 nsecs, total: 5515252 nsecs, count: 1091
__kfence_free:
avg = 5319 nsecs, total: 9735130 nsecs, count: 1830

After patching:
__kfence_alloc:
avg = 3597 nsecs, total: 6428491 nsecs, count: 1787
__kfence_free:
avg = 3046 nsecs, total: 3415390 nsecs, count: 1121

The numbers indicate that there is ~30% - ~40% performance improvement.

Link: https://lkml.kernel.org/r/20230403122738.6006-1-zhangpeng.00@bytedance.com
Signed-off-by: Peng Zhang <zhangpeng.00@bytedance.com>
Reviewed-by: Marco Elver <elver@google.com>
Cc: Alexander Potapenko <glider@google.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm/zswap: delay the initialization of zswap

Since some users may not use zswap, the zswap_pool is wasted.  Save memory
by delaying the initialization of zswap until enabled.

[liushixin2@huawei.com: fix some pattern problem suggested by Christoph]
Link: https://lkml.kernel.org/r/20230411093632.822290-4-liushixin2@huawei.com
Link: https://lkml.kernel.org/r/20230403121318.1876082-4-liushixin2@huawei.com
Signed-off-by: Liu Shixin <liushixin2@huawei.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Cc: Dan Streetman <ddstreet@ieee.org>
Cc: Nathan Chancellor <nathan@kernel.org>
Cc: Seth Jennings <sjenning@redhat.com>
Cc: Vitaly Wool <vitaly.wool@konsulko.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm/zswap: replace zswap_init_{started/failed} with zswap_init_state

The zswap_init_started variable name has a bit confusing.  Actually, there
are three state: uninitialized, initial failed and initial succeed.  Add a
new variable zswap_init_state to replace zswap_init_{started/failed}.

Link: https://lkml.kernel.org/r/20230403121318.1876082-3-liushixin2@huawei.com
Signed-off-by: Liu Shixin <liushixin2@huawei.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Cc: Dan Streetman <ddstreet@ieee.org>
Cc: Nathan Chancellor <nathan@kernel.org>
Cc: Seth Jennings <sjenning@redhat.com>
Cc: Vitaly Wool <vitaly.wool@konsulko.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm/zswap: remove zswap_entry_cache_{create,destroy} helper function

Patch series "Delay the initialization of zswap", v9.

In the initialization of zswap, about 18MB memory will be allocated for
zswap_pool.  Since some users may not use zswap, the zswap_pool is wasted.
Save memory by delaying the initialization of zswap until enabled.

This patch (of 3):

Remove zswap_entry_cache_create and zswap_entry_cache_destroy and use
kmem_cache_* function directly.

Link: https://lkml.kernel.org/r/20230411093632.822290-1-liushixin2@huawei.com
Link: https://lkml.kernel.org/r/20230403121318.1876082-1-liushixin2@huawei.com
Link: https://lkml.kernel.org/r/20230403121318.1876082-2-liushixin2@huawei.com
Signed-off-by: Liu Shixin <liushixin2@huawei.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Cc: Dan Streetman <ddstreet@ieee.org>
Cc: Nathan Chancellor <nathan@kernel.org>
Cc: Seth Jennings <sjenning@redhat.com>
Cc: Vitaly Wool <vitaly.wool@konsulko.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm: vmalloc: rename addr_to_vb_xarray() function

Short the name of the addr_to_vb_xarray() function to the addr_to_vb_xa().
This aligns with other internal function abbreviations.

Link: https://lkml.kernel.org/r/20230331073727.6968-1-urezki@gmail.com
Signed-off-by: Uladzislau Rezki (Sony) <urezki@gmail.com>
Suggested-by: Baoquan He <bhe@redhat.com>
Reviewed-by: Baoquan He <bhe@redhat.com>
Reviewed-by: Lorenzo Stoakes <lstoakes@gmail.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Dave Chinner <david@fromorbit.com>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: Oleksiy Avramchenko <oleksiy.avramchenko@sony.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

kmemleak-test: fix kmemleak_test.c build logic

kmemleak-test.c was moved to the samples directory in 1abbef4f51724
("mm,kmemleak-test.c: move kmemleak-test.c to samples dir").

If CONFIG_DEBUG_KMEMLEAK_TEST=m and CONFIG_SAMPLES is unset,
kmemleak-test.c will be unnecessarily compiled.

So move the entry for CONFIG_DEBUG_KMEMLEAK_TEST from mm/Kconfig and add a
new CONFIG_SAMPLE_KMEMLEAK in samples/ to control whether kmemleak-test.c
is built or not.

Link: https://lkml.kernel.org/r/20230330060904.292975-1-gehao@kylinos.cn
Fixes: 1abbef4f51724 ("mm,kmemleak-test.c: move kmemleak-test.c to samples dir")
Signed-off-by: Hao Ge <gehao@kylinos.cn>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Alex Gaynor <alex.gaynor@gmail.com>
Cc: Alex Williamson <alex.williamson@redhat.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Finn Behrens <me@kloenk.dev>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Jason Gunthorpe <jgg@ziepe.ca>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Miguel Ojeda <ojeda@kernel.org>
Cc: Tony Krowiak <akrowiak@linux.ibm.com>
Cc: Ye Xingchen <ye.xingchen@zte.com.cn>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

lib/test_vmalloc.c: add vm_map_ram()/vm_unmap_ram() test case

Add vm_map_ram()/vm_unmap_ram() test case to our stress test-suite.

[akpm@linux-foundation.org: fix whitespace, per Lorenzo]
Link: https://lkml.kernel.org/r/20230330190639.431589-2-urezki@gmail.com
Signed-off-by: Uladzislau Rezki (Sony) <urezki@gmail.com>
Reviewed-by: Lorenzo Stoakes <lstoakes@gmail.com>
Reviewed-by: Baoquan He <bhe@redhat.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Dave Chinner <david@fromorbit.com>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: Oleksiy Avramchenko <oleksiy.avramchenko@sony.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm: vmalloc: remove a global vmap_blocks xarray

A global vmap_blocks-xarray array can be contented under heavy usage of
the vm_map_ram()/vm_unmap_ram() APIs.  The lock_stat shows that a
"vmap_blocks.xa_lock" lock is a second in a top-list when it comes to
contentions:

<snip>
----------------------------------------
class name con-bounces contentions ...
----------------------------------------
vmap_area_lock:         2554079 2554276 ...
  --------------
  vmap_area_lock        1297948  [<00000000dd41cbaa>] alloc_vmap_area+0x1c7/0x910
  vmap_area_lock        1256330  [<000000009d927bf3>] free_vmap_block+0x4a/0xe0
  vmap_area_lock              1  [<00000000c95c05a7>] find_vm_area+0x16/0x70
  --------------
  vmap_area_lock        1738590  [<00000000dd41cbaa>] alloc_vmap_area+0x1c7/0x910
  vmap_area_lock         815688  [<000000009d927bf3>] free_vmap_block+0x4a/0xe0
  vmap_area_lock              1  [<00000000c1d619d7>] __get_vm_area_node+0xd2/0x170

vmap_blocks.xa_lock:    862689  862698 ...
  -------------------
  vmap_blocks.xa_lock   378418    [<00000000625a5626>] vm_map_ram+0x359/0x4a0
  vmap_blocks.xa_lock   484280    [<00000000caa2ef03>] xa_erase+0xe/0x30
  -------------------
  vmap_blocks.xa_lock   576226    [<00000000caa2ef03>] xa_erase+0xe/0x30
  vmap_blocks.xa_lock   286472    [<00000000625a5626>] vm_map_ram+0x359/0x4a0
...
<snip>

that is a result of running vm_map_ram()/vm_unmap_ram() in
a loop. The test creates 64(on 64 CPUs system) threads and
each one maps/unmaps 1 page.

After this change the "xa_lock" can be considered as a noise
in the same test condition:

<snip>
...
&xa->xa_lock#1:         10333 10394 ...
  --------------
  &xa->xa_lock#1        5349      [<00000000bbbc9751>] xa_erase+0xe/0x30
  &xa->xa_lock#1        5045      [<0000000018def45d>] vm_map_ram+0x3a4/0x4f0
  --------------
  &xa->xa_lock#1        7326      [<0000000018def45d>] vm_map_ram+0x3a4/0x4f0
  &xa->xa_lock#1        3068      [<00000000bbbc9751>] xa_erase+0xe/0x30
...
<snip>

Running the test_vmalloc.sh run_test_mask=1024 nr_threads=64 nr_pages=5
shows around ~8 percent of throughput improvement of vm_map_ram() and
vm_unmap_ram() APIs.

This patch does not fix vmap_area_lock/free_vmap_area_lock and
purge_vmap_area_lock bottle-necks, it is rather a separate rework.

Link: https://lkml.kernel.org/r/20230330190639.431589-1-urezki@gmail.com
Signed-off-by: Uladzislau Rezki (Sony) <urezki@gmail.com>
Reviewed-by: Lorenzo Stoakes <lstoakes@gmail.com>
Reviewed-by: Baoquan He <bhe@redhat.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Dave Chinner <david@fromorbit.com>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: Oleksiy Avramchenko <oleksiy.avramchenko@sony.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm: move free_area_empty() to mm/internal.h

The free_area_empty() helper is only used inside mm/ so move it there to
reduce noise in include/linux/mmzone.h

Link: https://lkml.kernel.org/r/20230326160215.2674531-1-rppt@kernel.org
Signed-off-by: Mike Rapoport (IBM) <rppt@kernel.org>
Suggested-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

kmsan: fix a stale comment in kmsan_save_stack_with_flags()

After commit 446ec83805dd ("mm/page_alloc: use might_alloc()") and commit
84172f4bb752 ("mm/page_alloc: combine __alloc_pages and
__alloc_pages_nodemask"), the comment is no longer accurate.  Flag
'__GFP_DIRECT_RECLAIM' is clear enough on its own, so remove the comment
rather than update it.

Link: https://lkml.kernel.org/r/20230327034149.942-1-thunder.leizhen@huawei.com
Signed-off-by: Zhen Lei <thunder.leizhen@huawei.com>
Reviewed-by: Alexander Potapenko <glider@google.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Marco Elver <elver@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

hugetlb: remove PageHeadHuge()

Sidhartha Kumar removed the last caller of PageHeadHuge(), so we can now
remove it and make folio_test_hugetlb() the real implementation.  Add
kernel-doc for folio_test_hugetlb().

Link: https://lkml.kernel.org/r/20230327151050.1787744-1-willy@infradead.org
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Reviewed-by: Sidhartha Kumar <sidhartha.kumar@oracle.com>
Reviewed-by: Mike Kravetz <mike.kravetz@oracle.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Muchun Song <songmuchun@bytedance.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

xtensa: reword ARCH_FORCE_MAX_ORDER prompt and help text

The prompt and help text of ARCH_FORCE_MAX_ORDER are not even close to
describe this configuration option.

Update both to actually describe what this option does.

Link: https://lkml.kernel.org/r/20230324052233.2654090-15-rppt@kernel.org
Signed-off-by: Mike Rapoport (IBM) <rppt@kernel.org>
Reviewed-by: Max Filippov <jcmvbkbc@gmail.com>
Acked-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Reviewed-by: Zi Yan <ziy@nvidia.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Christophe Leroy <christophe.leroy@csgroup.eu>
Cc: David Miller <davem@davemloft.net>
Cc: Dinh Nguyen <dinguyen@kernel.org>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Guo Ren <guoren@kernel.org>
Cc: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Rich Felker <dalias@libc.org>
Cc: "Russell King (Oracle)" <linux@armlinux.org.uk>
Cc: Will Deacon <will@kernel.org>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

sparc: reword ARCH_FORCE_MAX_ORDER prompt and help text

The prompt and help text of ARCH_FORCE_MAX_ORDER are not even close to
describe this configuration option.

Update both to actually describe what this option does.

Link: https://lkml.kernel.org/r/20230324052233.2654090-14-rppt@kernel.org
Signed-off-by: Mike Rapoport (IBM) <rppt@kernel.org>
Acked-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Reviewed-by: Zi Yan <ziy@nvidia.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Christophe Leroy <christophe.leroy@csgroup.eu>
Cc: David Miller <davem@davemloft.net>
Cc: Dinh Nguyen <dinguyen@kernel.org>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Guo Ren <guoren@kernel.org>
Cc: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Cc: Max Filippov <jcmvbkbc@gmail.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Rich Felker <dalias@libc.org>
Cc: "Russell King (Oracle)" <linux@armlinux.org.uk>
Cc: Will Deacon <will@kernel.org>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

sh: drop ranges for definition of ARCH_FORCE_MAX_ORDER

sh defines insane ranges for ARCH_FORCE_MAX_ORDER allowing MAX_ORDER up to
63, which implies maximal contiguous allocation size of 2^63 pages.

Drop bogus definitions of ranges for ARCH_FORCE_MAX_ORDER and leave it a
simple integer with sensible defaults.

Users that *really* need to change the value of ARCH_FORCE_MAX_ORDER will
be able to do so but they won't be mislead by the bogus ranges.

[rppt@kernel.org: untweak ARCH_FORCE_MAX_ORDER's `range']
Link: https://lkml.kernel.org/r/20230325060828.2662773-13-rppt@kernel.org
Link: https://lkml.kernel.org/r/20230324052233.2654090-13-rppt@kernel.org
Signed-off-by: Mike Rapoport (IBM) <rppt@kernel.org>
Acked-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Reviewed-by: Zi Yan <ziy@nvidia.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Christophe Leroy <christophe.leroy@csgroup.eu>
Cc: David Miller <davem@davemloft.net>
Cc: Dinh Nguyen <dinguyen@kernel.org>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Guo Ren <guoren@kernel.org>
Cc: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Cc: Max Filippov <jcmvbkbc@gmail.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Rich Felker <dalias@libc.org>
Cc: "Russell King (Oracle)" <linux@armlinux.org.uk>
Cc: Will Deacon <will@kernel.org>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

sh: reword ARCH_FORCE_MAX_ORDER prompt and help text

The prompt and help text of ARCH_FORCE_MAX_ORDER are not even close to
describe this configuration option.

Update both to actually describe what this option does.

[rppt@kernel.org: tweak ARCH_FORCE_MAX_ORDER's `range']
Link: https://lkml.kernel.org/r/20230325060828.2662773-12-rppt@kernel.org
Link: https://lkml.kernel.org/r/20230324052233.2654090-12-rppt@kernel.org
Signed-off-by: Mike Rapoport (IBM) <rppt@kernel.org>
Acked-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Reviewed-by: Zi Yan <ziy@nvidia.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Christophe Leroy <christophe.leroy@csgroup.eu>
Cc: David Miller <davem@davemloft.net>
Cc: Dinh Nguyen <dinguyen@kernel.org>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Guo Ren <guoren@kernel.org>
Cc: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Cc: Max Filippov <jcmvbkbc@gmail.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Rich Felker <dalias@libc.org>
Cc: "Russell King (Oracle)" <linux@armlinux.org.uk>
Cc: Will Deacon <will@kernel.org>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

powerpc: drop ranges for definition of ARCH_FORCE_MAX_ORDER

PowerPC defines ranges for ARCH_FORCE_MAX_ORDER some of which are insanely
allowing MAX_ORDER up to 63, which implies maximal contiguous allocation
size of 2^63 pages.

Drop bogus definitions of ranges for ARCH_FORCE_MAX_ORDER and leave it a
simple integer with sensible defaults.

Users that *really* need to change the value of ARCH_FORCE_MAX_ORDER will
be able to do so but they won't be mislead by the bogus ranges.

Link: https://lkml.kernel.org/r/20230324052233.2654090-11-rppt@kernel.org
Signed-off-by: Mike Rapoport (IBM) <rppt@kernel.org>
Acked-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Reviewed-by: Zi Yan <ziy@nvidia.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Christophe Leroy <christophe.leroy@csgroup.eu>
Cc: David Miller <davem@davemloft.net>
Cc: Dinh Nguyen <dinguyen@kernel.org>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Guo Ren <guoren@kernel.org>
Cc: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Cc: Max Filippov <jcmvbkbc@gmail.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Rich Felker <dalias@libc.org>
Cc: "Russell King (Oracle)" <linux@armlinux.org.uk>
Cc: Will Deacon <will@kernel.org>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

powerpc: reword ARCH_FORCE_MAX_ORDER prompt and help text

The prompt and help text of ARCH_FORCE_MAX_ORDER are not even close to
describe this configuration option.

Update both to actually describe what this option does.

Link: https://lkml.kernel.org/r/20230324052233.2654090-10-rppt@kernel.org
Signed-off-by: Mike Rapoport (IBM) <rppt@kernel.org>
Acked-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Reviewed-by: Zi Yan <ziy@nvidia.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Christophe Leroy <christophe.leroy@csgroup.eu>
Cc: David Miller <davem@davemloft.net>
Cc: Dinh Nguyen <dinguyen@kernel.org>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Guo Ren <guoren@kernel.org>
Cc: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Cc: Max Filippov <jcmvbkbc@gmail.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Rich Felker <dalias@libc.org>
Cc: "Russell King (Oracle)" <linux@armlinux.org.uk>
Cc: Will Deacon <will@kernel.org>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

nios2: drop ranges for definition of ARCH_FORCE_MAX_ORDER

nios2 defines range for ARCH_FORCE_MAX_ORDER allowing MAX_ORDER up to 19,
which implies maximal contiguous allocation size of 2^19 pages or 2GiB.

Drop bogus definition of ranges for ARCH_FORCE_MAX_ORDER and leave it a
simple integer with sensible default.

Users that *really* need to change the value of ARCH_FORCE_MAX_ORDER will
be able to do so but they won't be mislead by the bogus ranges.

Link: https://lkml.kernel.org/r/20230324052233.2654090-9-rppt@kernel.org
Signed-off-by: Mike Rapoport (IBM) <rppt@kernel.org>
Acked-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Reviewed-by: Zi Yan <ziy@nvidia.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Christophe Leroy <christophe.leroy@csgroup.eu>
Cc: David Miller <davem@davemloft.net>
Cc: Dinh Nguyen <dinguyen@kernel.org>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Guo Ren <guoren@kernel.org>
Cc: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Cc: Max Filippov <jcmvbkbc@gmail.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Rich Felker <dalias@libc.org>
Cc: "Russell King (Oracle)" <linux@armlinux.org.uk>
Cc: Will Deacon <will@kernel.org>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

nios2: reword ARCH_FORCE_MAX_ORDER prompt and help text

The prompt and help text of ARCH_FORCE_MAX_ORDER are not even close to
describe this configuration option.

Update both to actually describe what this option does.

Link: https://lkml.kernel.org/r/20230324052233.2654090-8-rppt@kernel.org
Signed-off-by: Mike Rapoport (IBM) <rppt@kernel.org>
Acked-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Reviewed-by: Zi Yan <ziy@nvidia.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Christophe Leroy <christophe.leroy@csgroup.eu>
Cc: David Miller <davem@davemloft.net>
Cc: Dinh Nguyen <dinguyen@kernel.org>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Guo Ren <guoren@kernel.org>
Cc: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Cc: Max Filippov <jcmvbkbc@gmail.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Rich Felker <dalias@libc.org>
Cc: "Russell King (Oracle)" <linux@armlinux.org.uk>
Cc: Will Deacon <will@kernel.org>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

m68k: reword ARCH_FORCE_MAX_ORDER prompt and help text

The prompt and help text of ARCH_FORCE_MAX_ORDER are not even close to
describe this configuration option.

Update both to actually describe what this option does.

Link: https://lkml.kernel.org/r/20230324052233.2654090-7-rppt@kernel.org
Signed-off-by: Mike Rapoport (IBM) <rppt@kernel.org>
Acked-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Acked-by: Geert Uytterhoeven <geert@linux-m68k.org>
Reviewed-by: Zi Yan <ziy@nvidia.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Christophe Leroy <christophe.leroy@csgroup.eu>
Cc: David Miller <davem@davemloft.net>
Cc: Dinh Nguyen <dinguyen@kernel.org>
Cc: Guo Ren <guoren@kernel.org>
Cc: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Cc: Max Filippov <jcmvbkbc@gmail.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Rich Felker <dalias@libc.org>
Cc: "Russell King (Oracle)" <linux@armlinux.org.uk>
Cc: Will Deacon <will@kernel.org>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

ia64: don't allow users to override ARCH_FORCE_MAX_ORDER

It is enough to keep default values for base and huge pages without
letting users to override ARCH_FORCE_MAX_ORDER.

Drop the prompt to make the option unvisible in *config.

Link: https://lkml.kernel.org/r/20230324052233.2654090-6-rppt@kernel.org
Signed-off-by: Mike Rapoport (IBM) <rppt@kernel.org>
Acked-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Reviewed-by: Zi Yan <ziy@nvidia.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Christophe Leroy <christophe.leroy@csgroup.eu>
Cc: David Miller <davem@davemloft.net>
Cc: Dinh Nguyen <dinguyen@kernel.org>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Guo Ren <guoren@kernel.org>
Cc: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Cc: Max Filippov <jcmvbkbc@gmail.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Rich Felker <dalias@libc.org>
Cc: "Russell King (Oracle)" <linux@armlinux.org.uk>
Cc: Will Deacon <will@kernel.org>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

csky: drop ARCH_FORCE_MAX_ORDER

The default value of ARCH_FORCE_MAX_ORDER matches the generic default
defined in the MM code, the architecture does not support huge pages, so
there is no need to keep ARCH_FORCE_MAX_ORDER option available.

Drop it.

Link: https://lkml.kernel.org/r/20230324052233.2654090-5-rppt@kernel.org
Signed-off-by: Mike Rapoport (IBM) <rppt@kernel.org>
Acked-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Reviewed-by: Zi Yan <ziy@nvidia.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Christophe Leroy <christophe.leroy@csgroup.eu>
Cc: David Miller <davem@davemloft.net>
Cc: Dinh Nguyen <dinguyen@kernel.org>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Guo Ren <guoren@kernel.org>
Cc: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Cc: Max Filippov <jcmvbkbc@gmail.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Rich Felker <dalias@libc.org>
Cc: "Russell King (Oracle)" <linux@armlinux.org.uk>
Cc: Will Deacon <will@kernel.org>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

arm64: reword ARCH_FORCE_MAX_ORDER prompt and help text

The prompt and help text of ARCH_FORCE_MAX_ORDER are not even close to
describe this configuration option.

Update both to actually describe what this option does.

[rppt@kernel.org: change ARCH_FORCE_MAX_ORDER dependencies]
Link: https://lkml.kernel.org/r/20230325060828.2662773-4-rppt@kernel.org
Link: https://lkml.kernel.org/r/20230324052233.2654090-4-rppt@kernel.org
Signed-off-by: Mike Rapoport (IBM) <rppt@kernel.org>
Acked-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Reviewed-by: Zi Yan <ziy@nvidia.com>
Reviewed-by: Kefeng Wang <wangkefeng.wang@huawei.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Christophe Leroy <christophe.leroy@csgroup.eu>
Cc: David Miller <davem@davemloft.net>
Cc: Dinh Nguyen <dinguyen@kernel.org>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Guo Ren <guoren@kernel.org>
Cc: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Cc: Max Filippov <jcmvbkbc@gmail.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Rich Felker <dalias@libc.org>
Cc: "Russell King (Oracle)" <linux@armlinux.org.uk>
Cc: Will Deacon <will@kernel.org>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

arm64: drop ranges in definition of ARCH_FORCE_MAX_ORDER

It is not a good idea to change fundamental parameters of core memory
management.  Having predefined ranges suggests that the values within
those ranges are sensible, but one has to *really* understand implications
of changing MAX_ORDER before actually amending it and ranges don't help
here.

Drop ranges in definition of ARCH_FORCE_MAX_ORDER and make its prompt
visible only if EXPERT=y

Link: https://lkml.kernel.org/r/20230324052233.2654090-3-rppt@kernel.org
Signed-off-by: Mike Rapoport (IBM) <rppt@kernel.org>
Reviewed-by: Zi Yan <ziy@nvidia.com>
Reviewed-by: Kefeng Wang <wangkefeng.wang@huawei.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Christophe Leroy <christophe.leroy@csgroup.eu>
Cc: David Miller <davem@davemloft.net>
Cc: Dinh Nguyen <dinguyen@kernel.org>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Guo Ren <guoren@kernel.org>
Cc: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Max Filippov <jcmvbkbc@gmail.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Rich Felker <dalias@libc.org>
Cc: "Russell King (Oracle)" <linux@armlinux.org.uk>
Cc: Will Deacon <will@kernel.org>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

arm: reword ARCH_FORCE_MAX_ORDER prompt and help text

Patch series "arch,mm: cleanup Kconfig entries for ARCH_FORCE_MAX_ORDER",
v3.

Several architectures have ARCH_FORCE_MAX_ORDER in their Kconfig and
they all have wrong and misleading prompt and help text for this option.

Besides, some define insane limits for possible values of
ARCH_FORCE_MAX_ORDER, some carefully define ranges only for a subset of
possible configurations, some make this option configurable by users for no
good reason.

This set updates the prompt and help text everywhere and does its best to
update actual definitions of ranges where applicable.

kbuild generated a bunch of false positives because it assigns -1 to
ARCH_FORCE_MAX_ORDER, hopefully this will be fixed soon.

This patch (of 14):

The prompt and help text of ARCH_FORCE_MAX_ORDER are not even close to
describe this configuration option.

Update both to actually describe what this option does.

Link: https://lkml.kernel.org/r/20230325060828.2662773-1-rppt@kernel.org
Link: https://lkml.kernel.org/r/20230324052233.2654090-1-rppt@kernel.org
Link: https://lkml.kernel.org/r/20230324052233.2654090-2-rppt@kernel.org
Signed-off-by: Mike Rapoport (IBM) <rppt@kernel.org>
Acked-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Reviewed-by: Zi Yan <ziy@nvidia.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Christophe Leroy <christophe.leroy@csgroup.eu>
Cc: David Miller <davem@davemloft.net>
Cc: Dinh Nguyen <dinguyen@kernel.org>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Guo Ren <guoren@kernel.org>
Cc: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Cc: Max Filippov <jcmvbkbc@gmail.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Rich Felker <dalias@libc.org>
Cc: "Russell King (Oracle)" <linux@armlinux.org.uk>
Cc: Will Deacon <will@kernel.org>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

memcg: do not drain charge pcp caches on remote isolated cpus

Leonardo Bras has noticed that pcp charge cache draining might be
disruptive on workloads relying on 'isolated cpus', a feature commonly
used on workloads that are sensitive to interruption and context switching
such as vRAN and Industrial Control Systems.

There are essentially two ways how to approach the issue.  We can either
allow the pcp cache to be drained on a different rather than a local cpu
or avoid remote flushing on isolated cpus.

The current pcp charge cache is really optimized for high performance and
it always relies to stick with its cpu.  That means it only requires
local_lock (preempt_disable on !RT) and draining is handed over to pcp WQ
to drain locally again.

The former solution (remote draining) would require to add an additional
locking to prevent local charges from racing with the draining.  This adds
an atomic operation to otherwise simple arithmetic fast path in the
try_charge path.  Another concern is that the remote draining can cause a
lock contention for the isolated workloads and therefore interfere with it
indirectly via user space interfaces.

Another option is to avoid draining scheduling on isolated cpus
altogether.  That means that those remote cpus would keep their charges
even after drain_all_stock returns.  This is certainly not optimal either
but it shouldn't really cause any major problems.  In the worst case (many
isolated cpus with charges - each of them with MEMCG_CHARGE_BATCH i.e 64
page) the memory consumption of a memcg would be artificially higher than
can be immediately used from other cpus.

Theoretically a memcg OOM killer could be triggered pre-maturely.
Currently it is not really clear whether this is a practical problem
though.  Tight memcg limit would be really counter productive to cpu
isolated workloads pretty much by definition because any memory reclaimed
induced by memcg limit could break user space timing expectations as those
usually expect execution in the userspace most of the time.

Also charges could be left behind on memcg removal.  Any future charge on
those isolated cpus will drain that pcp cache so this won't be a permanent
leak.

Considering cons and pros of both approaches this patch is implementing
the second option and simply do not schedule remote draining if the target
cpu is isolated.  This solution is much more simpler.  It doesn't add any
new locking and it is more more predictable from the user space POV.
Should the pre-mature memcg OOM become a real life problem, we can revisit
this decision.

[akpm@linux-foundation.org: memcontrol.c needs sched/isolation.h]
Link: https://lore.kernel.org/oe-kbuild-all/202303180617.7E3aIlHf-lkp@intel.com/
Link: https://lkml.kernel.org/r/20230317134448.11082-3-mhocko@kernel.org
Signed-off-by: Michal Hocko <mhocko@suse.com>
Suggested-by: Roman Gushchin <roman.gushchin@linux.dev>
Acked-by: Roman Gushchin <roman.gushchin@linux.dev>
Reported-by: Leonardo Bras <leobras@redhat.com>
Acked-by: Shakeel Butt <shakeelb@google.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Muchun Song <muchun.song@linux.dev>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Frederic Weisbecker <frederic@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

sched/isolation: add cpu_is_isolated() API

Patch series "memcg, cpuisol: do not interfere pcp cache charges draining
with cpuisol workloads".

Leonardo has reported [1] that pcp memcg charge draining can interfere
with cpu isolated workloads.  The said draining is done from a WQ context
with a pcp worker scheduled on each CPU which holds any cached charges for
a specific memcg hierarchy.  Operation is not really a common operation
[2].  It can be triggered from the userspace though so some care is
definitely due.

Leonardo has tried to address the issue by allowing remote charge draining
[3].  This approach requires an additional locking to synchronize pcp
caches sync from a remote cpu from local pcp consumers.  Even though the
proposed lock was per-cpu there is still potential for contention and less
predictable behavior.

This patchset addresses the issue from a different angle.  Rather than
dealing with a potential synchronization, cpus which are isolated are
simply never scheduled to be drained.  This means that a small amount of
charges could be laying around and waiting for a later use or they are
flushed when a different memcg is charged from the same cpu.  More details
are in patch 2.  The first patch from Frederic is implementing an
abstraction to tell whether a specific cpu has been isolated and therefore
require a special treatment.

This patch (of 2):

Provide this new API to check if a CPU has been isolated either through
isolcpus= or nohz_full= kernel parameter.

It aims at avoiding kernel load deemed to be safely spared on CPUs running
sensitive workload that can't bear any disturbance, such as pcp cache
draining.

Link: https://lkml.kernel.org/r/20230317134448.11082-1-mhocko@kernel.org
Link: https://lkml.kernel.org/r/20230317134448.11082-2-mhocko@kernel.org
Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
Signed-off-by: Michal Hocko <mhocko@suse.com>
Suggested-by: Michal Hocko <mhocko@suse.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Muchun Song <muchun.song@linux.dev>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Roman Gushchin <roman.gushchin@linux.dev>
Cc: Shakeel Butt <shakeelb@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Leonardo Bras <leobras@redhat.com>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm: khugepaged: fix kernel BUG in hpage_collapse_scan_file()

Syzkaller reported the following issue:

kernel BUG at mm/khugepaged.c:1823!
invalid opcode: 0000 [#1] PREEMPT SMP KASAN
CPU: 1 PID: 5097 Comm: syz-executor220 Not tainted 6.2.0-syzkaller-13154-g857f1268a591 #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 02/16/2023
RIP: 0010:collapse_file mm/khugepaged.c:1823 [inline]
RIP: 0010:hpage_collapse_scan_file+0x67c8/0x7580 mm/khugepaged.c:2233
Code: 00 00 89 de e8 c9 66 a3 ff 31 ff 89 de e8 c0 66 a3 ff 45 84 f6 0f 85 28 0d 00 00 e8 22 64 a3 ff e9 dc f7 ff ff e8 18 64 a3 ff <0f> 0b f3 0f 1e fa e8 0d 64 a3 ff e9 93 f6 ff ff f3 0f 1e fa 4c 89
RSP: 0018:ffffc90003dff4e0 EFLAGS: 00010093
RAX: ffffffff81e95988 RBX: 00000000000001c1 RCX: ffff8880205b3a80
RDX: 0000000000000000 RSI: 00000000000001c0 RDI: 00000000000001c1
RBP: ffffc90003dff830 R08: ffffffff81e90e67 R09: fffffbfff1a433c3
R10: 0000000000000000 R11: dffffc0000000001 R12: 0000000000000000
R13: ffffc90003dff6c0 R14: 00000000000001c0 R15: 0000000000000000
FS:  00007fdbae5ee700(0000) GS:ffff8880b9900000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007fdbae6901e0 CR3: 000000007b2dd000 CR4: 00000000003506e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
 <TASK>
 madvise_collapse+0x721/0xf50 mm/khugepaged.c:2693
 madvise_vma_behavior mm/madvise.c:1086 [inline]
 madvise_walk_vmas mm/madvise.c:1260 [inline]
 do_madvise+0x9e5/0x4680 mm/madvise.c:1439
 __do_sys_madvise mm/madvise.c:1452 [inline]
 __se_sys_madvise mm/madvise.c:1450 [inline]
 __x64_sys_madvise+0xa5/0xb0 mm/madvise.c:1450
 do_syscall_x64 arch/x86/entry/common.c:50 [inline]
 do_syscall_64+0x41/0xc0 arch/x86/entry/common.c:80
 entry_SYSCALL_64_after_hwframe+0x63/0xcd

The xas_store() call during page cache scanning can potentially translate
'xas' into the error state (with the reproducer provided by the syzkaller
the error code is -ENOMEM).  However, there are no further checks after
the 'xas_store', and the next call of 'xas_next' at the start of the
scanning cycle doesn't increase the xa_index, and the issue occurs.

This patch will add the xarray state error checking after the xas_store()
and the corresponding result error code.

Tested via syzbot.

[akpm@linux-foundation.org: update include/trace/events/huge_memory.h's SCAN_STATUS]
Link: https://lkml.kernel.org/r/20230329145330.23191-1-ivan.orlov0322@gmail.com
Link: https://syzkaller.appspot.com/bug?id=7d6bb3760e026ece7524500fe44fb024a0e959fc
Signed-off-by: Ivan Orlov <ivan.orlov0322@gmail.com>
Reported-by: syzbot+9578faa5475acb35fa50@syzkaller.appspotmail.com
Tested-by: Zach O'Keefe <zokeefe@google.com>
Cc: Yang Shi <shy828301@gmail.com>
Cc: Himadri Pandya <himadrispandya@gmail.com>
Cc: Ivan Orlov <ivan.orlov0322@gmail.com>
Cc: Shuah Khan <skhan@linuxfoundation.org>
Cc: Song Liu <songliubraving@fb.com>
Cc: Rik van Riel <riel@surriel.com>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

kasan: remove hwasan-kernel-mem-intrinsic-prefix=1 for clang-14

Some unknown -mllvm options (i.e.  those starting with the letter "h")
don't cause an error to be returned by clang, so the cc-option helper adds
the unknown hwasan-kernel-mem-intrinsic-prefix=1 flag to CFLAGS with
compilers that are new enough for hwasan but too old for this option.

This causes a rather unreadable build failure:

fixdep: error opening file: scripts/mod/.empty.o.d: No such file or directory
make[4]: *** [/home/arnd/arm-soc/scripts/Makefile.build:252: scripts/mod/empty.o] Error 2
fixdep: error opening file: scripts/mod/.devicetable-offsets.s.d: No such file or directory
make[4]: *** [/home/arnd/arm-soc/scripts/Makefile.build:114: scripts/mod/devicetable-offsets.s] Error 2

Add a version check to only allow this option with clang-15, gcc-13
or later versions.

Link: https://lkml.kernel.org/r/20230418122350.1646391-1-arnd@kernel.org
Fixes: 51287dcb00cc ("kasan: emit different calls for instrumentable memintrinsics")
Link: https://lore.kernel.org/all/CANpmjNMwYosrvqh4ogDO8rgn+SeDHM2b-shD21wTypm_6MMe=g@mail.gmail.com/
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Marco Elver <elver@google.com>
Reviewed-by: Nathan Chancellor <nathan@kernel.org>
Cc: Alexander Potapenko <glider@google.com>
Cc: Andrey Konovalov <andreyknvl@gmail.com>
Cc: Andrey Ryabinin <ryabinin.a.a@gmail.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Masahiro Yamada <masahiroy@kernel.org>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Nick Desaulniers <ndesaulniers@google.com>
Cc: Nicolas Schier <nicolas@fjasle.eu>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Tom Rix <trix@redhat.com>
Cc: Vincenzo Frascino <Vincenzo.Frascino@arm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

zsmalloc: reset compaction source zspage pointer after putback_zspage()

The current implementation of the compaction loop fails to set the source
zspage pointer to NULL in all cases, leading to a potential issue where
__zs_compact() could use a stale zspage pointer.  This pointer could even
point to a previously freed zspage, causing unexpected behavior in the
putback_zspage() and migrate_write_unlock() functions after returning from
the compaction loop.

Address the issue by ensuring that the source zspage pointer is always set
to NULL when it should be.

Link: https://lkml.kernel.org/r/20230417130850.1784777-1-senozhatsky@chromium.org
Fixes: 5a845e9f2d66 ("zsmalloc: rework compaction algorithm")
Signed-off-by: Sergey Senozhatsky <senozhatsky@chromium.org>
Reported-by: Yu Zhao <yuzhao@google.com>
Tested-by: Yu Zhao <yuzhao@google.com>
Reviewed-by: Yosry Ahmed <yosryahmed@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm: make arch_has_descending_max_zone_pfns() static

clang produces a build failure on x86 for some randconfig builds after a
change that moves around code to mm/mm_init.c:

Cannot find symbol for section 2: .text.
mm/mm_init.o: failed

I have not been able to figure out why this happens, but the __weak
annotation on arch_has_descending_max_zone_pfns() is the trigger here.

Removing the weak function in favor of an open-coded Kconfig option check
avoids the problem and becomes clearer as well as better to optimize by
the compiler.

[arnd@arndb.de: fix logic bug]
Link: https://lkml.kernel.org/r/20230415081904.969049-1-arnd@kernel.org
Link: https://lkml.kernel.org/r/20230414080418.110236-1-arnd@kernel.org
Fixes: 9420f89db2dd ("mm: move most of core MM initialization to mm/mm_init.c")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Tested-by: SeongJae Park <sj@kernel.org>
Tested-by: Geert Uytterhoeven <geert+renesas@glider.be>
Acked-by: Mike Rapoport (IBM) <rppt@kernel.org>
Cc: kernel test robot <oliver.sang@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm: avoid passing 0 to __ffs()

23baf831a32c ("mm, treewide: redefine MAX_ORDER sanely") results in
various boot failures (hang) on arm targets Debug messages reveal the
reason.

########### MAX_ORDER=10 start=0 __ffs(start)=-1 min()=10 min_t=-1
                         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

If start==0, __ffs(start) returns 0xfffffff or (as int) -1, which min_t()
interprets as such, while min() apparently uses the returned unsigned long
value. Obviously a negative order isn't received well by the rest of the
code.

[akpm@linux-foundation.org: fix comment, per Mike]
Link: https://lkml.kernel.org/r/ZDBa7HWZK69dKKzH@kernel.org
Link: https://lkml.kernel.org/r/20230406072529.vupqyrzqnhyozeyh@box.shutemov.name
Fixes: 23baf831a32c ("mm, treewide: redefine MAX_ORDER sanely")
Signed-off-by: "Kirill A. Shutemov" <kirill@shutemov.name>
Reported-by: Guenter Roeck <linux@roeck-us.net>
Link: https://lkml.kernel.org/r/9460377a-38aa-4f39-ad57-fb73725f92db@roeck-us.net
Reviewed-by: Mike Rapoport (IBM) <rppt@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

sync mm-stable with mm-hotfixes-stable to pick up depended-upon upstream changes

nilfs2: initialize unused bytes in segment summary blocks

Syzbot still reports uninit-value in nilfs_add_checksums_on_logs() for
KMSAN enabled kernels after applying commit 7397031622e0 ("nilfs2:
initialize "struct nilfs_binfo_dat"->bi_pad field").

This is because the unused bytes at the end of each block in segment
summaries are not initialized.  So this fixes the issue by padding the
unused bytes with null bytes.

Link: https://lkml.kernel.org/r/20230417173513.12598-1-konishi.ryusuke@gmail.com
Signed-off-by: Ryusuke Konishi <konishi.ryusuke@gmail.com>
Tested-by: Ryusuke Konishi <konishi.ryusuke@gmail.com>
Reported-by: syzbot+048585f3f4227bb2b49b@syzkaller.appspotmail.com
Link: https://syzkaller.appspot.com/bug?extid=048585f3f4227bb2b49b
Cc: Alexander Potapenko <glider@google.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm: page_alloc: skip regions with hugetlbfs pages when allocating 1G pages

A bug was reported by Yuanxi Liu where allocating 1G pages at runtime is
taking an excessive amount of time for large amounts of memory.  Further
testing allocating huge pages that the cost is linear i.e.  if allocating
1G pages in batches of 10 then the time to allocate nr_hugepages from
10->20->30->etc increases linearly even though 10 pages are allocated at
each step.  Profiles indicated that much of the time is spent checking the
validity within already existing huge pages and then attempting a
migration that fails after isolating the range, draining pages and a whole
lot of other useless work.

Commit eb14d4eefdc4 ("mm,page_alloc: drop unnecessary checks from
pfn_range_valid_contig") removed two checks, one which ignored huge pages
for contiguous allocations as huge pages can sometimes migrate.  While
there may be value on migrating a 2M page to satisfy a 1G allocation, it's
potentially expensive if the 1G allocation fails and it's pointless to try
moving a 1G page for a new 1G allocation or scan the tail pages for valid
PFNs.

Reintroduce the PageHuge check and assume any contiguous region with
hugetlbfs pages is unsuitable for a new 1G allocation.

The hpagealloc test allocates huge pages in batches and reports the
average latency per page over time.  This test happens just after boot
when fragmentation is not an issue.  Units are in milliseconds.

hpagealloc
                               6.3.0-rc6              6.3.0-rc6              6.3.0-rc6
                                 vanilla   hugeallocrevert-v1r1   hugeallocsimple-v1r2
Min       Latency       26.42 (   0.00%)        5.07 (  80.82%)       18.94 (  28.30%)
1st-qrtle Latency      356.61 (   0.00%)        5.34 (  98.50%)       19.85 (  94.43%)
2nd-qrtle Latency      697.26 (   0.00%)        5.47 (  99.22%)       20.44 (  97.07%)
3rd-qrtle Latency      972.94 (   0.00%)        5.50 (  99.43%)       20.81 (  97.86%)
Max-1     Latency       26.42 (   0.00%)        5.07 (  80.82%)       18.94 (  28.30%)
Max-5     Latency       82.14 (   0.00%)        5.11 (  93.78%)       19.31 (  76.49%)
Max-10    Latency      150.54 (   0.00%)        5.20 (  96.55%)       19.43 (  87.09%)
Max-90    Latency     1164.45 (   0.00%)        5.53 (  99.52%)       20.97 (  98.20%)
Max-95    Latency     1223.06 (   0.00%)        5.55 (  99.55%)       21.06 (  98.28%)
Max-99    Latency     1278.67 (   0.00%)        5.57 (  99.56%)       22.56 (  98.24%)
Max       Latency     1310.90 (   0.00%)        8.06 (  99.39%)       26.62 (  97.97%)
Amean     Latency      678.36 (   0.00%)        5.44 *  99.20%*       20.44 *  96.99%*

                   6.3.0-rc6   6.3.0-rc6   6.3.0-rc6
                     vanilla   revert-v1   hugeallocfix-v2
Duration User           0.28        0.27        0.30
Duration System       808.66       17.77       35.99
Duration Elapsed      830.87       18.08       36.33

The vanilla kernel is poor, taking up to 1.3 second to allocate a huge
page and almost 10 minutes in total to run the test.  Reverting the
problematic commit reduces it to 8ms at worst and the patch takes 26ms.
This patch fixes the main issue with skipping huge pages but leaves the
page_count() out because a page with an elevated count potentially can
migrate.

BugLink: https://bugzilla.kernel.org/show_bug.cgi?id=217022
Link: https://lkml.kernel.org/r/20230414141429.pwgieuwluxwez3rj@techsingularity.net
Fixes: eb14d4eefdc4 ("mm,page_alloc: drop unnecessary checks from pfn_range_valid_contig")
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Reported-by: Yuanxi Liu <y.liu@naruida.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Reviewed-by: David Hildenbrand <david@redhat.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Reviewed-by: Oscar Salvador <osalvador@suse.de>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm/mmap: regression fix for unmapped_area{_topdown}

The maple tree limits the gap returned to a window that specifically fits
what was asked.  This may not be optimal in the case of switching search
directions or a gap that does not satisfy the requested space for other
reasons.  Fix the search by retrying the operation and limiting the search
window in the rare occasion that a conflict occurs.

Link: https://lkml.kernel.org/r/20230414185919.4175572-1-Liam.Howlett@oracle.com
Fixes: 3499a13168da ("mm/mmap: use maple tree for unmapped_area{_topdown}")
Signed-off-by: Liam R. Howlett <Liam.Howlett@oracle.com>
Reported-by: Rick Edgecombe <rick.p.edgecombe@intel.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

maple_tree: fix mas_empty_area() search

The internal function of mas_awalk() was incorrectly skipping the last
entry in a node, which could potentially be NULL.  This is only a problem
for the left-most node in the tree - otherwise that NULL would not exist.

Fix mas_awalk() by using the metadata to obtain the end of the node for
the loop and the logical pivot as apposed to the raw pivot value.

Link: https://lkml.kernel.org/r/20230414145728.4067069-2-Liam.Howlett@oracle.com
Fixes: 54a611b60590 ("Maple Tree: add new data structure")
Signed-off-by: Liam R. Howlett <Liam.Howlett@oracle.com>
Reported-by: Rick Edgecombe <rick.p.edgecombe@intel.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

maple_tree: make maple state reusable after mas_empty_area_rev()

Stop using maple state min/max for the range by passing through pointers
for those values.  This will allow the maple state to be reused without
resetting.

Also add some logic to fail out early on searching with invalid
arguments.

Link: https://lkml.kernel.org/r/20230414145728.4067069-1-Liam.Howlett@oracle.com
Fixes: 54a611b60590 ("Maple Tree: add new data structure")
Signed-off-by: Liam R. Howlett <Liam.Howlett@oracle.com>
Reported-by: Rick Edgecombe <rick.p.edgecombe@intel.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm: kmsan: handle alloc failures in kmsan_ioremap_page_range()

Similarly to kmsan_vmap_pages_range_noflush(), kmsan_ioremap_page_range()
must also properly handle allocation/mapping failures.  In the case of
such, it must clean up the already created metadata mappings and return an
error code, so that the error can be propagated to ioremap_page_range().
Without doing so, KMSAN may silently fail to bring the metadata for the
page range into a consistent state, which will result in user-visible
crashes when trying to access them.

Link: https://lkml.kernel.org/r/20230413131223.4135168-2-glider@google.com
Fixes: b073d7f8aee4 ("mm: kmsan: maintain KMSAN metadata for page operations")
Signed-off-by: Alexander Potapenko <glider@google.com>
Reported-by: Dipanjan Das <mail.dipanjan.das@gmail.com>
Link: https://lore.kernel.org/linux-mm/CANX2M5ZRrRA64k0hOif02TjmY9kbbO2aCBPyq79es34RXZ=cAw@mail.gmail.com/
Reviewed-by: Marco Elver <elver@google.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Uladzislau Rezki (Sony) <urezki@gmail.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm: kmsan: handle alloc failures in kmsan_vmap_pages_range_noflush()

As reported by Dipanjan Das, when KMSAN is used together with kernel fault
injection (or, generally, even without the latter), calls to kcalloc() or
__vmap_pages_range_noflush() may fail, leaving the metadata mappings for
the virtual mapping in an inconsistent state.  When these metadata
mappings are accessed later, the kernel crashes.

To address the problem, we return a non-zero error code from
kmsan_vmap_pages_range_noflush() in the case of any allocation/mapping
failure inside it, and make vmap_pages_range_noflush() return an error if
KMSAN fails to allocate the metadata.

This patch also removes KMSAN_WARN_ON() from vmap_pages_range_noflush(),
as these allocation failures are not fatal anymore.

Link: https://lkml.kernel.org/r/20230413131223.4135168-1-glider@google.com
Fixes: b073d7f8aee4 ("mm: kmsan: maintain KMSAN metadata for page operations")
Signed-off-by: Alexander Potapenko <glider@google.com>
Reported-by: Dipanjan Das <mail.dipanjan.das@gmail.com>
Link: https://lore.kernel.org/linux-mm/CANX2M5ZRrRA64k0hOif02TjmY9kbbO2aCBPyq79es34RXZ=cAw@mail.gmail.com/
Reviewed-by: Marco Elver <elver@google.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Uladzislau Rezki (Sony) <urezki@gmail.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

tools/Makefile: do missed s/vm/mm/

Commit 799fb82aa132 ("tools/vm: rename tools/vm to tools/mm") missed
renaming 'vm' in 'tools/Makefile' to 'mm'.  As a result, 'make clean'
under 'tools/' directory fails as below:

    $ make -C tools clean
      DESCEND vm
    make[1]: Entering directory '/linux/tools/vm'
    make[1]: *** No rule to make target 'clean'.  Stop.
    make[1]: Leaving directory '/linux/tools/vm'
    make: *** [Makefile:173: vm_clean] Error 2
    make: Leaving directory '/linux/tools'

Do the missed rename.

Link: https://lkml.kernel.org/r/20230415203110.13858-1-sj@kernel.org
Fixes: 799fb82aa132 ("tools/vm: rename tools/vm to tools/mm")
Signed-off-by: SeongJae Park <sj@kernel.org>
Reported-by: Ricardo Pardini <ricardo@pardini.net>
Link: https://lore.kernel.org/linux-mm/20230415202454.13558-1-sj@kernel.org/
Tested-by: Ricardo Pardini <ricardo@pardini.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm: fix memory leak on mm_init error handling

commit f1a7941243c1 ("mm: convert mm's rss stats into percpu_counter")
introduces a memory leak by missing a call to destroy_context() when a
percpu_counter fails to allocate.

Before introducing the per-cpu counter allocations, init_new_context() was
the last call that could fail in mm_init(), and thus there was no need to
ever invoke destroy_context() in the error paths.  Adding the following
percpu counter allocations adds error paths after init_new_context(),
which means its associated destroy_context() needs to be called when
percpu counters fail to allocate.

Link: https://lkml.kernel.org/r/20230330133822.66271-1-mathieu.desnoyers@efficios.com
Fixes: f1a7941243c1 ("mm: convert mm's rss stats into percpu_counter")
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Acked-by: Shakeel Butt <shakeelb@google.com>
Cc: Marek Szyprowski <m.szyprowski@samsung.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm/page_alloc: fix potential deadlock on zonelist_update_seq seqlock

syzbot is reporting circular locking dependency which involves
zonelist_update_seq seqlock [1], for this lock is checked by memory
allocation requests which do not need to be retried.

One deadlock scenario is kmalloc(GFP_ATOMIC) from an interrupt handler.

  CPU0
  ----
  __build_all_zonelists() {
    write_seqlock(&zonelist_update_seq); // makes zonelist_update_seq.seqcount odd
    // e.g. timer interrupt handler runs at this moment
      some_timer_func() {
        kmalloc(GFP_ATOMIC) {
          __alloc_pages_slowpath() {
            read_seqbegin(&zonelist_update_seq) {
              // spins forever because zonelist_update_seq.seqcount is odd
            }
          }
        }
      }
    // e.g. timer interrupt handler finishes
    write_sequnlock(&zonelist_update_seq); // makes zonelist_update_seq.seqcount even
  }

This deadlock scenario can be easily eliminated by not calling
read_seqbegin(&zonelist_update_seq) from !__GFP_DIRECT_RECLAIM allocation
requests, for retry is applicable to only __GFP_DIRECT_RECLAIM allocation
requests.  But Michal Hocko does not know whether we should go with this
approach.

Another deadlock scenario which syzbot is reporting is a race between
kmalloc(GFP_ATOMIC) from tty_insert_flip_string_and_push_buffer() with
port->lock held and printk() from __build_all_zonelists() with
zonelist_update_seq held.

  CPU0                                   CPU1
  ----                                   ----
  pty_write() {
    tty_insert_flip_string_and_push_buffer() {
                                         __build_all_zonelists() {
                                           write_seqlock(&zonelist_update_seq);
                                           build_zonelists() {
                                             printk() {
                                               vprintk() {
                                                 vprintk_default() {
                                                   vprintk_emit() {
                                                     console_unlock() {
                                                       console_flush_all() {
                                                         console_emit_next_record() {
                                                           con->write() = serial8250_console_write() {
      spin_lock_irqsave(&port->lock, flags);
      tty_insert_flip_string() {
        tty_insert_flip_string_fixed_flag() {
          __tty_buffer_request_room() {
            tty_buffer_alloc() {
              kmalloc(GFP_ATOMIC | __GFP_NOWARN) {
                __alloc_pages_slowpath() {
                  zonelist_iter_begin() {
                    read_seqbegin(&zonelist_update_seq); // spins forever because zonelist_update_seq.seqcount is odd
                                                             spin_lock_irqsave(&port->lock, flags); // spins forever because port->lock is held
                    }
                  }
                }
              }
            }
          }
        }
      }
      spin_unlock_irqrestore(&port->lock, flags);
                                                             // message is printed to console
                                                             spin_unlock_irqrestore(&port->lock, flags);
                                                           }
                                                         }
                                                       }
                                                     }
                                                   }
                                                 }
                                               }
                                             }
                                           }
                                           write_sequnlock(&zonelist_update_seq);
                                         }
    }
  }

This deadlock scenario can be eliminated by

  preventing interrupt context from calling kmalloc(GFP_ATOMIC)

and

  preventing printk() from calling console_flush_all()

while zonelist_update_seq.seqcount is odd.

Since Petr Mladek thinks that __build_all_zonelists() can become a
candidate for deferring printk() [2], let's address this problem by

  disabling local interrupts in order to avoid kmalloc(GFP_ATOMIC)

and

  disabling synchronous printk() in order to avoid console_flush_all()

.

As a side effect of minimizing duration of zonelist_update_seq.seqcount
being odd by disabling synchronous printk(), latency at
read_seqbegin(&zonelist_update_seq) for both !__GFP_DIRECT_RECLAIM and
__GFP_DIRECT_RECLAIM allocation requests will be reduced.  Although, from
lockdep perspective, not calling read_seqbegin(&zonelist_update_seq) (i.e.
do not record unnecessary locking dependency) from interrupt context is
still preferable, even if we don't allow calling kmalloc(GFP_ATOMIC)
inside
write_seqlock(&zonelist_update_seq)/write_sequnlock(&zonelist_update_seq)
section...

Link: https://lkml.kernel.org/r/8796b95c-3da3-5885-fddd-6ef55f30e4d3@I-love.SAKURA.ne.jp
Fixes: 3d36424b3b58 ("mm/page_alloc: fix race condition between build_all_zonelists and page allocation")
Link: https://lkml.kernel.org/r/ZCrs+1cDqPWTDFNM@alley
Reported-by: syzbot <syzbot+223c7461c58c58a4cb10@syzkaller.appspotmail.com>
Link: https://syzkaller.appspot.com/bug?extid=223c7461c58c58a4cb10
Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Acked-by: Michal Hocko <mhocko@suse.com>
Acked-by: Mel Gorman <mgorman@techsingularity.net>
Cc: Petr Mladek <pmladek@suse.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Cc: John Ogness <john.ogness@linutronix.de>
Cc: Patrick Daly <quic_pdaly@quicinc.com>
Cc: Sergey Senozhatsky <senozhatsky@chromium.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

kernel/sys.c: fix and improve control flow in __sys_setres[ug]id()

Linux Security Modules (LSMs) that implement the "capable" hook will
usually emit an access denial message to the audit log whenever they
"block" the current task from using the given capability based on their
security policy.

The occurrence of a denial is used as an indication that the given task
has attempted an operation that requires the given access permission, so
the callers of functions that perform LSM permission checks must take care
to avoid calling them too early (before it is decided if the permission is
actually needed to perform the requested operation).

The __sys_setres[ug]id() functions violate this convention by first
calling ns_capable_setid() and only then checking if the operation
requires the capability or not.  It means that any caller that has the
capability granted by DAC (task's capability set) but not by MAC (LSMs)
will generate a "denied" audit record, even if is doing an operation for
which the capability is not required.

Fix this by reordering the checks such that ns_capable_setid() is checked
last and -EPERM is returned immediately if it returns false.

While there, also do two small optimizations:
* move the capability check before prepare_creds() and
* bail out early in case of a no-op.

Link: https://lkml.kernel.org/r/20230217162154.837549-1-omosnace@redhat.com
Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Signed-off-by: Ondrej Mosnacek <omosnace@redhat.com>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

sync mm-stable with mm-hotfixes-stable to pick up depended-upon upstream changes

Revert "userfaultfd: don't fail on unrecognized features"

This is a proposal to revert commit 914eedcb9ba0ff53c33808.

I found this when writing a simple UFFDIO_API test to be the first unit
test in this set.  Two things breaks with the commit:

  - UFFDIO_API check was lost and missing.  According to man page, the
  kernel should reject ioctl(UFFDIO_API) if uffdio_api.api != 0xaa.  This
  check is needed if the api version will be extended in the future, or
  user app won't be able to identify which is a new kernel.

  - Feature flags checks were removed, which means UFFDIO_API with a
  feature that does not exist will also succeed.  According to the man
  page, we should (and it makes sense) to reject ioctl(UFFDIO_API) if
  unknown features passed in.

Link: https://lore.kernel.org/r/20220722201513.1624158-1-axelrasmussen@google.com
Link: https://lkml.kernel.org/r/20230412163922.327282-2-peterx@redhat.com
Fixes: 914eedcb9ba0 ("userfaultfd: don't fail on unrecognized features")
Signed-off-by: Peter Xu <peterx@redhat.com>
Acked-by: David Hildenbrand <david@redhat.com>
Cc: Axel Rasmussen <axelrasmussen@google.com>
Cc: Dmitry Safonov <0x7f454c46@gmail.com>
Cc: Mike Kravetz <mike.kravetz@oracle.com>
Cc: Mike Rapoport (IBM) <rppt@kernel.org>
Cc: Zach O'Keefe <zokeefe@google.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

writeback, cgroup: fix null-ptr-deref write in bdi_split_work_to_wbs

KASAN report null-ptr-deref:
==================================================================
BUG: KASAN: null-ptr-deref in bdi_split_work_to_wbs+0x5c5/0x7b0
Write of size 8 at addr 0000000000000000 by task sync/943
CPU: 5 PID: 943 Comm: sync Tainted: 6.3.0-rc5-next-20230406-dirty #461
Call Trace:
 <TASK>
 dump_stack_lvl+0x7f/0xc0
 print_report+0x2ba/0x340
 kasan_report+0xc4/0x120
 kasan_check_range+0x1b7/0x2e0
 __kasan_check_write+0x24/0x40
 bdi_split_work_to_wbs+0x5c5/0x7b0
 sync_inodes_sb+0x195/0x630
 sync_inodes_one_sb+0x3a/0x50
 iterate_supers+0x106/0x1b0
 ksys_sync+0x98/0x160
[...]
==================================================================

The race that causes the above issue is as follows:

           cpu1                     cpu2
-------------------------|-------------------------
inode_switch_wbs
 INIT_WORK(&isw->work, inode_switch_wbs_work_fn)
 queue_rcu_work(isw_wq, &isw->work)
 // queue_work async
  inode_switch_wbs_work_fn
   wb_put_many(old_wb, nr_switched)
    percpu_ref_put_many
     ref->data->release(ref)
     cgwb_release
      queue_work(cgwb_release_wq, &wb->release_work)
      // queue_work async
       &wb->release_work
       cgwb_release_workfn
                            ksys_sync
                             iterate_supers
                              sync_inodes_one_sb
                               sync_inodes_sb
                                bdi_split_work_to_wbs
                                 kmalloc(sizeof(*work), GFP_ATOMIC)
                                 // alloc memory failed
        percpu_ref_exit
         ref->data = NULL
         kfree(data)
                                 wb_get(wb)
                                  percpu_ref_get(&wb->refcnt)
                                   percpu_ref_get_many(ref, 1)
                                    atomic_long_add(nr, &ref->data->count)
                                     atomic64_add(i, v)
                                     // trigger null-ptr-deref

bdi_split_work_to_wbs() traverses &bdi->wb_list to split work into all
wbs.  If the allocation of new work fails, the on-stack fallback will be
used and the reference count of the current wb is increased afterwards.
If cgroup writeback membership switches occur before getting the reference
count and the current wb is released as old_wd, then calling wb_get() or
wb_put() will trigger the null pointer dereference above.

This issue was introduced in v4.3-rc7 (see fix tag1).  Both
sync_inodes_sb() and __writeback_inodes_sb_nr() calls to
bdi_split_work_to_wbs() can trigger this issue.  For scenarios called via
sync_inodes_sb(), originally commit 7fc5854f8c6e ("writeback: synchronize
sync(2) against cgroup writeback membership switches") reduced the
possibility of the issue by adding wb_switch_rwsem, but in v5.14-rc1 (see
fix tag2) removed the "inode_io_list_del_locked(inode, old_wb)" from
inode_switch_wbs_work_fn() so that wb->state contains WB_has_dirty_io,
thus old_wb is not skipped when traversing wbs in bdi_split_work_to_wbs(),
and the issue becomes easily reproducible again.

To solve this problem, percpu_ref_exit() is called under RCU protection to
avoid race between cgwb_release_workfn() and bdi_split_work_to_wbs().
Moreover, replace wb_get() with wb_tryget() in bdi_split_work_to_wbs(),
and skip the current wb if wb_tryget() fails because the wb has already
been shutdown.

Link: https://lkml.kernel.org/r/20230410130826.1492525-1-libaokun1@huawei.com
Fixes: b817525a4a80 ("writeback: bdi_writeback iteration must not skip dying ones")
Signed-off-by: Baokun Li <libaokun1@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Acked-by: Tejun Heo <tj@kernel.org>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: Andreas Dilger <adilger.kernel@dilger.ca>
Cc: Christian Brauner <brauner@kernel.org>
Cc: Dennis Zhou <dennis@kernel.org>
Cc: Hou Tao <houtao1@huawei.com>
Cc: yangerkun <yangerkun@huawei.com>
Cc: Zhang Yi <yi.zhang@huawei.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

maple_tree: fix a potential memory leak, OOB access, or other unpredictable bug

In mas_alloc_nodes(), "node->node_count = 0" means to initialize the
node_count field of the new node, but the node may not be a new node.  It
may be a node that existed before and node_count has a value, setting it
to 0 will cause a memory leak.  At this time, mas->alloc->total will be
greater than the actual number of nodes in the linked list, which may
cause many other errors.  For example, out-of-bounds access in
mas_pop_node(), and mas_pop_node() may return addresses that should not be
used.  Fix it by initializing node_count only for new nodes.

Also, by the way, an if-else statement was removed to simplify the code.

Link: https://lkml.kernel.org/r/20230411041005.26205-1-zhangpeng.00@bytedance.com
Fixes: 54a611b60590 ("Maple Tree: add new data structure")
Signed-off-by: Peng Zhang <zhangpeng.00@bytedance.com>
Reviewed-by: Liam R. Howlett <Liam.Howlett@oracle.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

tools/mm/page_owner_sort.c: fix TGID output when cull=tg is used

When using cull option with 'tg' flag, the fprintf is using pid instead
of tgid. It should use tgid instead.

Link: https://lkml.kernel.org/r/20230411034929.2071501-1-steve_chou@pesi.com.tw
Fixes: 9c8a0a8e599f4a ("tools/vm/page_owner_sort.c: support for user-defined culling rules")
Signed-off-by: Steve Chou <steve_chou@pesi.com.tw>
Cc: Jiajian Ye <yejiajian2018@email.szu.edu.cn>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mailmap: update jtoppins' entry to reference correct email

Link: https://lkml.kernel.org/r/d79bc6eaf65e68bd1c2a1e1510ab6291ce5926a6.1681162487.git.jtoppins@redhat.com
Signed-off-by: Jonathan Toppins <jtoppins@redhat.com>
Cc: Colin Ian King <colin.i.king@gmail.com>
Cc: Jakub Kicinski <kuba@kernel.org>
Cc: Kirill Tkhai <tkhai@ya.ru>
Cc: Konrad Dybcio <konrad.dybcio@linaro.org>
Cc: Qais Yousef <qyousef@layalina.io>
Cc: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm/mempolicy: fix use-after-free of VMA iterator

set_mempolicy_home_node() iterates over a list of VMAs and calls
mbind_range() on each VMA, which also iterates over the singular list of
the VMA passed in and potentially splits the VMA.  Since the VMA iterator
is not passed through, set_mempolicy_home_node() may now point to a stale
node in the VMA tree.  This can result in a UAF as reported by syzbot.

Avoid the stale maple tree node by passing the VMA iterator through to the
underlying call to split_vma().

mbind_range() is also overly complicated, since there are two calling
functions and one already handles iterating over the VMAs.  Simplify
mbind_range() to only handle merging and splitting of the VMAs.

Align the new loop in do_mbind() and existing loop in
set_mempolicy_home_node() to use the reduced mbind_range() function.  This
allows for a single location of the range calculation and avoids
constantly looking up the previous VMA (since this is a loop over the
VMAs).

Link: https://lore.kernel.org/linux-mm/000000000000c93feb05f87e24ad@google.com/
Fixes: 66850be55e8e ("mm/mempolicy: use vma iterator & maple state instead of vma linked list")
Signed-off-by: Liam R. Howlett <Liam.Howlett@oracle.com>
Reported-by: syzbot+a7c1ec5b1d71ceaa5186@syzkaller.appspotmail.com
Link: https://lkml.kernel.org/r/20230410152205.2294819-1-Liam.Howlett@oracle.com
Tested-by: syzbot+a7c1ec5b1d71ceaa5186@syzkaller.appspotmail.com
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm/huge_memory.c: warn with pr_warn_ratelimited instead of VM_WARN_ON_ONCE_FOLIO

split_huge_page_to_list() WARNs when called for huge zero pages, which
sounds to me too harsh because it does not imply a kernel bug, but just
notifies the event to admins.  On the other hand, this is considered as
critical by syzkaller and makes its testing less efficient, which seems to
me harmful.

So replace the VM_WARN_ON_ONCE_FOLIO with pr_warn_ratelimited.

Link: https://lkml.kernel.org/r/20230406082004.2185420-1-naoya.horiguchi@linux.dev
Fixes: 478d134e9506 ("mm/huge_memory: do not overkill when splitting huge_zero_page")
Signed-off-by: Naoya Horiguchi <naoya.horiguchi@nec.com>
Reported-by: syzbot+07a218429c8d19b1fb25@syzkaller.appspotmail.com
Link: https://lore.kernel.org/lkml/000000000000a6f34a05e6efcd01@google.com/
Reviewed-by: Yang Shi <shy828301@gmail.com>
Cc: Miaohe Lin <linmiaohe@huawei.com>
Cc: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp>
Cc: Xu Yu <xuyu@linux.alibaba.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm/mprotect: fix do_mprotect_pkey() return on error

When the loop over the VMA is terminated early due to an error, the return
code could be overwritten with ENOMEM.  Fix the return code by only
setting the error on early loop termination when the error is not set.

User-visible effects include: attempts to run mprotect() against a
special mapping or with a poorly-aligned hugetlb address should return
-EINVAL, but they presently return -ENOMEM.  In other cases an -EACCESS
should be returned.

Link: https://lkml.kernel.org/r/20230406193050.1363476-1-Liam.Howlett@oracle.com
Fixes: 2286a6914c77 ("mm: change mprotect_fixup to vma iterator")
Signed-off-by: Liam R. Howlett <Liam.Howlett@oracle.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm/khugepaged: check again on anon uffd-wp during isolation

Khugepaged collapse an anonymous thp in two rounds of scans.  The 2nd
round done in __collapse_huge_page_isolate() after
hpage_collapse_scan_pmd(), during which all the locks will be released
temporarily.  It means the pgtable can change during this phase before 2nd
round starts.

It's logically possible some ptes got wr-protected during this phase, and
we can errornously collapse a thp without noticing some ptes are
wr-protected by userfault.  e1e267c7928f wanted to avoid it but it only
did that for the 1st phase, not the 2nd phase.

Since __collapse_huge_page_isolate() happens after a round of small page
swapins, we don't need to worry on any !present ptes - if it existed
khugepaged will already bail out.  So we only need to check present ptes
with uffd-wp bit set there.

This is something I found only but never had a reproducer, I thought it
was one caused a bug in Muhammad's recent pagemap new ioctl work, but it
turns out it's not the cause of that but an userspace bug.  However this
seems to still be a real bug even with a very small race window, still
worth to have it fixed and copy stable.

Link: https://lkml.kernel.org/r/20230405155120.3608140-1-peterx@redhat.com
Fixes: e1e267c7928f ("khugepaged: skip collapse if uffd-wp detected")
Signed-off-by: Peter Xu <peterx@redhat.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Yang Shi <shy828301@gmail.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Axel Rasmussen <axelrasmussen@google.com>
Cc: Mike Rapoport <rppt@linux.vnet.ibm.com>
Cc: Nadav Amit <nadav.amit@gmail.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm/userfaultfd: fix uffd-wp handling for THP migration entries

Looks like what we fixed for hugetlb in commit 44f86392bdd1 ("mm/hugetlb:
fix uffd-wp handling for migration entries in
hugetlb_change_protection()") similarly applies to THP.

Setting/clearing uffd-wp on THP migration entries is not implemented
properly.  Further, while removing migration PMDs considers the uffd-wp
bit, inserting migration PMDs does not consider the uffd-wp bit.

We have to set/clear independently of the migration entry type in
change_huge_pmd() and properly copy the uffd-wp bit in
set_pmd_migration_entry().

Verified using a simple reproducer that triggers migration of a THP, that
the set_pmd_migration_entry() no longer loses the uffd-wp bit.

Link: https://lkml.kernel.org/r/20230405160236.587705-2-david@redhat.com
Fixes: f45ec5ff16a7 ("userfaultfd: wp: support swap and page migration")
Signed-off-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Cc: <stable@vger.kernel.org>
Cc: Muhammad Usama Anjum <usama.anjum@collabora.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm: swap: fix performance regression on sparsetruncate-tiny

The ->percpu_pvec_drained was originally introduced by commit d9ed0d08b6c6
("mm: only drain per-cpu pagevecs once per pagevec usage") to drain
per-cpu pagevecs only once per pagevec usage.  But after converting the
swap code to be more folio-based, the commit c2bc16817aa0 ("mm/swap: add
folio_batch_move_lru()") breaks this logic, which would cause
->percpu_pvec_drained to be reset to false, that means per-cpu pagevecs
will be drained multiple times per pagevec usage.

In theory, there should be no functional changes when converting code to
be more folio-based.  We should call folio_batch_reinit() in
folio_batch_move_lru() instead of folio_batch_init().  And to verify that
we still need ->percpu_pvec_drained, I ran mmtests/sparsetruncate-tiny and
got the following data:

                             baseline                   with
                            baseline/                 patch/
Min       Time      326.00 (   0.00%)      328.00 (  -0.61%)
1st-qrtle Time      334.00 (   0.00%)      336.00 (  -0.60%)
2nd-qrtle Time      338.00 (   0.00%)      341.00 (  -0.89%)
3rd-qrtle Time      343.00 (   0.00%)      347.00 (  -1.17%)
Max-1     Time      326.00 (   0.00%)      328.00 (  -0.61%)
Max-5     Time      327.00 (   0.00%)      330.00 (  -0.92%)
Max-10    Time      328.00 (   0.00%)      331.00 (  -0.91%)
Max-90    Time      350.00 (   0.00%)      357.00 (  -2.00%)
Max-95    Time      395.00 (   0.00%)      390.00 (   1.27%)
Max-99    Time      508.00 (   0.00%)      434.00 (  14.57%)
Max       Time      547.00 (   0.00%)      476.00 (  12.98%)
Amean     Time      344.61 (   0.00%)      345.56 *  -0.28%*
Stddev    Time       30.34 (   0.00%)       19.51 (  35.69%)
CoeffVar  Time        8.81 (   0.00%)        5.65 (  35.87%)
BAmean-99 Time      342.38 (   0.00%)      344.27 (  -0.55%)
BAmean-95 Time      338.58 (   0.00%)      341.87 (  -0.97%)
BAmean-90 Time      336.89 (   0.00%)      340.26 (  -1.00%)
BAmean-75 Time      335.18 (   0.00%)      338.40 (  -0.96%)
BAmean-50 Time      332.54 (   0.00%)      335.42 (  -0.87%)
BAmean-25 Time      329.30 (   0.00%)      332.00 (  -0.82%)

From the above it can be seen that we get similar data to when
->percpu_pvec_drained was introduced, so we still need it.  Let's call
folio_batch_reinit() in folio_batch_move_lru() to restore the original
logic.

Link: https://lkml.kernel.org/r/20230405161854.6931-1-zhengqi.arch@bytedance.com
Fixes: c2bc16817aa0 ("mm/swap: add folio_batch_move_lru()")
Signed-off-by: Qi Zheng <zhengqi.arch@bytedance.com>
Reviewed-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Acked-by: Mel Gorman <mgorman@suse.de>
Cc: Lorenzo Stoakes <lstoakes@gmail.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

sched/numa: use hash_32 to mix up PIDs accessing VMA

before: last 6 bits of PID is used as index to store information about
tasks accessing VMA's.

after: hash_32 is used to take of cases where tasks are created over a
period of time, and thus improve collision probability.

Result:
The patch series overall improves autonuma cost.

Kernbench around more than 5% improvement and system time in mmtest
autonuma showed more than 80% improvement

Link: https://lkml.kernel.org/r/d5a9f75513300caed74e5c8570bba9317b963c2b.1677672277.git.raghavendra.kt@amd.com
Signed-off-by: Raghavendra K T <raghavendra.kt@amd.com>
Suggested-by: Peter Zijlstra <peterz@infradead.org>
Cc: Bharata B Rao <bharata@amd.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Disha Talreja <dishaa.talreja@amd.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Mike Rapoport <rppt@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

sched/numa: implement access PID reset logic

This helps to ensure that only recently accessed PIDs scan the VMAs.

Current implementation: (idea supported by PeterZ)

 1. Accessing PID information is maintained in two windows.
    access_pids[1] being newest.

 2. Reset old access PID info i.e.  access_pid[0] every (4 *
    sysctl_numa_balancing_scan_delay) interval after initial scan delay
    period expires.

The above interval seemed to be experimentally optimum since it avoids
frequent reset of access info as well as helps clearing the old access
info regularly.  The reset logic is implemented in scan path.

Link: https://lkml.kernel.org/r/f7a675f66d1442d048b4216b2baf94515012c405.1677672277.git.raghavendra.kt@amd.com
Signed-off-by: Raghavendra K T <raghavendra.kt@amd.com>
Suggested-by: Mel Gorman <mgorman@techsingularity.net>
Cc: Bharata B Rao <bharata@amd.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Disha Talreja <dishaa.talreja@amd.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

sched/numa: enhance vma scanning logic

During Numa scanning make sure only relevant vmas of the tasks are
scanned.

Before:
 All the tasks of a process participate in scanning the vma even if they
 do not access vma in it's lifespan.

Now:
 Except cases of first few unconditional scans, if a process do
 not touch vma (exluding false positive cases of PID collisions)
 tasks no longer scan all vma

Logic used:

1) 6 bits of PID used to mark active bit in vma numab status during
   fault to remember PIDs accessing vma.  (Thanks Mel)

2) Subsequently in scan path, vma scanning is skipped if current PID
   had not accessed vma.

3) First two times we do allow unconditional scan to preserve earlier
   behaviour of scanning.

Acknowledgement to Bharata B Rao <bharata@amd.com> for initial patch to
store pid information and Peter Zijlstra <peterz@infradead.org> (Usage of
test and set bit)

Link: https://lkml.kernel.org/r/092f03105c7c1d3450f4636b1ea350407f07640e.1677672277.git.raghavendra.kt@amd.com
Signed-off-by: Raghavendra K T <raghavendra.kt@amd.com>
Suggested-by: Mel Gorman <mgorman@techsingularity.net>
Cc: David Hildenbrand <david@redhat.com>
Cc: Disha Talreja <dishaa.talreja@amd.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Mike Rapoport <rppt@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

sched/numa: apply the scan delay to every new vma

Pach series "sched/numa: Enhance vma scanning", v3.

The patchset proposes one of the enhancements to numa vma scanning
suggested by Mel.  This is continuation of [3].

Reposting the rebased patchset to akpm mm-unstable tree (March 1)

Existing mechanism of scan period involves, scan period derived from
per-thread stats.  Process Adaptive autoNUMA [1] proposed to gather NUMA
fault stats at per-process level to capture aplication behaviour better.

During that course of discussion, Mel proposed several ideas to enhance
current numa balancing.  One of the suggestion was below

Track what threads access a VMA.  The suggestion was to use an unsigned
long pid_mask and use the lower bits to tag approximately what threads
access a VMA.  Skip VMAs that did not trap a fault.  This would be
approximate because of PID collisions but would reduce scanning of areas
the thread is not interested in.  The above suggestion intends not to
penalize threads that has no interest in the vma, thus reduce scanning
overhead.

V3 changes are mostly based on PeterZ comments (details below in changes)

Summary of patchset:

Current patchset implements:

1. Delay the vma scanning logic for newly created VMA's so that
   additional overhead of scanning is not incurred for short lived tasks
   (implementation by Mel)

2. Store the information of tasks accessing VMA in 2 windows.  It is
   regularly cleared in (4*sysctl_numa_balancing_scan_delay) interval.
   The above time is derived from experimenting (Suggested by PeterZ) to
   balance between frequent clearing vs obsolete access data

3. hash_32 used to encode task index accessing VMA information

4. VMA's acess information is used to skip scanning for the tasks
   which had not accessed VMA

Changes since V2:
patch1:
 - Renaming of structure, macro to function,
 - Add explanation to heuristics
 - Adding more details from result (PeterZ)
 Patch2:
 - Usage of test and set bit (PeterZ)
 - Move storing access PID info to numa_migrate_prep()
 - Add a note on fainess among tasks allowed to scan
   (PeterZ)
 Patch3:
 - Maintain two windows of access PID information
  (PeterZ supported implementation and Gave idea to extend
   to N if needed)
 Patch4:
 - Apply hash_32 function to track VMA accessing PIDs (PeterZ)

Changes since RFC V1:
 - Include Mel's vma scan delay patch
 - Change the accessing pid store logic (Thanks Mel)
 - Fencing structure / code to NUMA_BALANCING (David, Mel)
 - Adding clearing access PID logic (Mel)
 - Descriptive change log ( Mike Rapoport)

Things to ponder over:
==========================================

- Improvement to clearing accessing PIDs logic (discussed in-detail in
  patch3 itself (Done in this patchset by implementing 2 window history)

- Current scan period is not changed in the patchset, so we do see
  frequent tries to scan.  Relaxing scan period dynamically could improve
  results further.

[1] sched/numa: Process Adaptive autoNUMA
Link: https://lore.kernel.org/lkml/20220128052851.17162-1-bharata@amd.com/T/
[2] RFC V1 Link:
  https://lore.kernel.org/all/cover.1673610485.git.raghavendra.kt@amd.com/

[3] V2 Link:
  https://lore.kernel.org/lkml/cover.1675159422.git.raghavendra.kt@amd.com/

Results:
Summary: Huge autonuma cost reduction seen in mmtest. Kernbench improvement
is more than 5% and huge system time (80%+) improvement from mmtest autonuma.
(dbench had huge std deviation to post)

kernbench
===========
                      6.2.0-mmunstable-base  6.2.0-mmunstable-patched
Amean     user-256    22002.51 (   0.00%)    22649.95 *  -2.94%*
Amean     syst-256    10162.78 (   0.00%)     8214.13 *  19.17%*
Amean     elsp-256      160.74 (   0.00%)      156.92 *   2.38%*

Duration User       66017.43    67959.84
Duration System     30503.15    24657.03
Duration Elapsed      504.61      493.12

                      6.2.0-mmunstable-base  6.2.0-mmunstable-patched
Ops NUMA alloc hit                1738835089.00  1738780310.00
Ops NUMA alloc local              1738834448.00  1738779711.00
Ops NUMA base-page range updates      477310.00      392566.00
Ops NUMA PTE updates                  477310.00      392566.00
Ops NUMA hint faults                   96817.00       87555.00
Ops NUMA hint local faults %           10150.00        2192.00
Ops NUMA hint local percent               10.48           2.50
Ops NUMA pages migrated                86660.00       85363.00
Ops AutoNUMA cost                        489.07         442.14

autonumabench
===============
                      6.2.0-mmunstable-base  6.2.0-mmunstable-patched
Amean     syst-NUMA01                  399.50 (   0.00%)       52.05 *  86.97%*
Amean     syst-NUMA01_THREADLOCAL        0.21 (   0.00%)        0.22 *  -5.41%*
Amean     syst-NUMA02                    0.80 (   0.00%)        0.78 *   2.68%*
Amean     syst-NUMA02_SMT                0.65 (   0.00%)        0.68 *  -3.95%*
Amean     elsp-NUMA01                  313.26 (   0.00%)      313.11 *   0.05%*
Amean     elsp-NUMA01_THREADLOCAL        1.06 (   0.00%)        1.08 *  -1.76%*
Amean     elsp-NUMA02                    3.19 (   0.00%)        3.24 *  -1.52%*
Amean     elsp-NUMA02_SMT                3.72 (   0.00%)        3.61 *   2.92%*

Duration User      396433.47   324835.96
Duration System      2808.70      376.66
Duration Elapsed     2258.61     2258.12

                      6.2.0-mmunstable-base  6.2.0-mmunstable-patched
Ops NUMA alloc hit                  59921806.00    49623489.00
Ops NUMA alloc miss                        0.00           0.00
Ops NUMA interleave hit                    0.00           0.00
Ops NUMA alloc local                59920880.00    49622594.00
Ops NUMA base-page range updates   152259275.00       50075.00
Ops NUMA PTE updates               152259275.00       50075.00
Ops NUMA PMD updates                       0.00           0.00
Ops NUMA hint faults               154660352.00       39014.00
Ops NUMA hint local faults %       138550501.00       23139.00
Ops NUMA hint local percent               89.58          59.31
Ops NUMA pages migrated              8179067.00       14147.00
Ops AutoNUMA cost                     774522.98         195.69

This patch (of 4):

Currently whenever a new task is created we wait for
sysctl_numa_balancing_scan_delay to avoid unnessary scanning overhead.
Extend the same logic to new or very short-lived VMAs.

[raghavendra.kt@amd.com: add initialization in vm_area_dup())]
Link: https://lkml.kernel.org/r/cover.1677672277.git.raghavendra.kt@amd.com
Link: https://lkml.kernel.org/r/7a6fbba87c8b51e67efd3e74285bb4cb311a16ca.1677672277.git.raghavendra.kt@amd.com
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Raghavendra K T <raghavendra.kt@amd.com>
Cc: Bharata B Rao <bharata@amd.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Disha Talreja <dishaa.talreja@amd.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

s390/mm: try VMA lock-based page fault handling first

Attempt VMA lock-based page fault handling first, and fall back to the
existing mmap_lock-based handling if that fails.

This is the s390 variant of "x86/mm: try VMA lock-based page fault handling
first".

Link: https://lkml.kernel.org/r/20230314132808.1266335-1-hca@linux.ibm.com
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
Cc: Alexander Gordeev <agordeev@linux.ibm.com>
Cc: Christian Borntraeger <borntraeger@linux.ibm.com>
Cc: Gerald Schaefer <gerald.schaefer@linux.ibm.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: Sven Schnelle <svens@linux.ibm.com>
Cc: Vasily Gorbik <gor@linux.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm: separate vma->lock from vm_area_struct

vma->lock being part of the vm_area_struct causes performance regression
during page faults because during contention its count and owner fields
are constantly updated and having other parts of vm_area_struct used
during page fault handling next to them causes constant cache line
bouncing.  Fix that by moving the lock outside of the vm_area_struct.

All attempts to keep vma->lock inside vm_area_struct in a separate cache
line still produce performance regression especially on NUMA machines.
Smallest regression was achieved when lock is placed in the fourth cache
line but that bloats vm_area_struct to 256 bytes.

Considering performance and memory impact, separate lock looks like the
best option.  It increases memory footprint of each VMA but that can be
optimized later if the new size causes issues.  Note that after this
change vma_init() does not allocate or initialize vma->lock anymore.  A
number of drivers allocate a pseudo VMA on the stack but they never use
the VMA's lock, therefore it does not need to be allocated.  The future
drivers which might need the VMA lock should use
vm_area_alloc()/vm_area_free() to allocate the VMA.

Link: https://lkml.kernel.org/r/20230227173632.3292573-34-surenb@google.com
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm/mmap: free vm_area_struct without call_rcu in exit_mmap

call_rcu() can take a long time when callback offloading is enabled.  Its
use in the vm_area_free can cause regressions in the exit path when
multiple VMAs are being freed.

Because exit_mmap() is called only after the last mm user drops its
refcount, the page fault handlers can't be racing with it.  Any other
possible user like oom-reaper or process_mrelease are already synchronized
using mmap_lock.  Therefore exit_mmap() can free VMAs directly, without
the use of call_rcu().

Expose __vm_area_free() and use it from exit_mmap() to avoid possible
call_rcu() floods and performance regressions caused by it.

Link: https://lkml.kernel.org/r/20230227173632.3292573-33-surenb@google.com
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

powerc/mm: try VMA lock-based page fault handling first

Attempt VMA lock-based page fault handling first, and fall back to the
existing mmap_lock-based handling if that fails.  Copied from "x86/mm: try
VMA lock-based page fault handling first"

[ldufour@linux.ibm.com: powerpc/mm: fix mmap_lock bad unlock]
Link: https://lkml.kernel.org/r/20230306154244.17560-1-ldufour@linux.ibm.com
Link: https://lore.kernel.org/linux-mm/842502FB-F99C-417C-9648-A37D0ECDC9CE@linux.ibm.com
Link: https://lkml.kernel.org/r/20230227173632.3292573-32-surenb@google.com
Signed-off-by: Laurent Dufour <ldufour@linux.ibm.com>
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Cc: Sachin Sant <sachinp@linux.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

arm64/mm: try VMA lock-based page fault handling first

Attempt VMA lock-based page fault handling first, and fall back to the
existing mmap_lock-based handling if that fails.

Link: https://lkml.kernel.org/r/20230227173632.3292573-31-surenb@google.com
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

x86/mm: try VMA lock-based page fault handling first

Attempt VMA lock-based page fault handling first, and fall back to the
existing mmap_lock-based handling if that fails.

Link: https://lkml.kernel.org/r/20230227173632.3292573-30-surenb@google.com
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm: introduce per-VMA lock statistics

Add a new CONFIG_PER_VMA_LOCK_STATS config option to dump extra statistics
about handling page fault under VMA lock.

Link: https://lkml.kernel.org/r/20230227173632.3292573-29-surenb@google.com
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm: prevent userfaults to be handled under per-vma lock

Due to the possibility of handle_userfault dropping mmap_lock, avoid fault
handling under VMA lock and retry holding mmap_lock.  This can be handled
more gracefully in the future.

Link: https://lkml.kernel.org/r/20230227173632.3292573-28-surenb@google.com
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Suggested-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm: prevent do_swap_page from handling page faults under VMA lock

Due to the possibility of do_swap_page dropping mmap_lock, abort fault
handling under VMA lock and retry holding mmap_lock.  This can be handled
more gracefully in the future.

Link: https://lkml.kernel.org/r/20230227173632.3292573-27-surenb@google.com
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Reviewed-by: Laurent Dufour <laurent.dufour@fr.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm: add FAULT_FLAG_VMA_LOCK flag

Add a new flag to distinguish page faults handled under protection of
per-vma lock.

[surenb@google.com: document FAULT_FLAG_VMA_LOCK flag]
Link: https://lkml.kernel.org/r/20230301022720.1380780-2-surenb@google.com
Link: https://lore.kernel.org/all/20230301113648.7c279865@canb.auug.org.au/
Link: https://lkml.kernel.org/r/20230227173632.3292573-26-surenb@google.com
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Reviewed-by: Laurent Dufour <laurent.dufour@fr.ibm.com>
Cc: Dan Carpenter <error27@gmail.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Jason Gunthorpe <jgg@ziepe.ca>
Cc: Liam R. Howlett <Liam.Howlett@oracle.com>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Pavel Tatashin <pasha.tatashin@soleen.com>
Cc: Yu Zhao <yuzhao@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm: fall back to mmap_lock if vma->anon_vma is not yet set

When vma->anon_vma is not set, page fault handler will set it by either
reusing anon_vma of an adjacent VMA if VMAs are compatible or by
allocating a new one.  find_mergeable_anon_vma() walks VMA tree to find a
compatible adjacent VMA and that requires not only the faulting VMA to be
stable but also the tree structure and other VMAs inside that tree.
Therefore locking just the faulting VMA is not enough for this search.
Fall back to taking mmap_lock when vma->anon_vma is not set.  This
situation happens only on the first page fault and should not affect
overall performance.

Link: https://lkml.kernel.org/r/20230227173632.3292573-25-surenb@google.com
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Reviewed-by: Hyeonggon Yoo <42.hyeyoo@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm: introduce lock_vma_under_rcu to be used from arch-specific code

Introduce lock_vma_under_rcu function to lookup and lock a VMA during page
fault handling.  When VMA is not found, can't be locked or changes after
being locked, the function returns NULL.  The lookup is performed under
RCU protection to prevent the found VMA from being destroyed before the
VMA lock is acquired.  VMA lock statistics are updated according to the
results.  For now only anonymous VMAs can be searched this way.  In other
cases the function returns NULL.

Link: https://lkml.kernel.org/r/20230227173632.3292573-24-surenb@google.com
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>