This defines and exports a platform specific custom vm_get_page_prot() via
subscribing ARCH_HAS_VM_GET_PAGE_PROT. It localizes arch_vm_get_page_prot()
and moves it near vm_get_page_prot().

Link: https://lkml.kernel.org/r/20220414062125.609297-4-anshuman.khandual@arm.comSigned-off-by: NAnshuman Khandual <anshuman.khandual@arm.com>
Reviewed-by: NCatalin Marinas <catalin.marinas@arm.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Will Deacon <will@kernel.org>
Cc: Christophe Leroy <christophe.leroy@csgroup.eu>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: David S. Miller <davem@davemloft.net>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Khalid Aziz <khalid.aziz@oracle.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

This defines and exports a platform specific custom vm_get_page_prot() via
subscribing ARCH_HAS_VM_GET_PAGE_PROT.  While here, this also localizes
arch_vm_get_page_prot() as __vm_get_page_prot() and moves it near
vm_get_page_prot().

Link: https://lkml.kernel.org/r/20220414062125.609297-3-anshuman.khandual@arm.comSigned-off-by: NAnshuman Khandual <anshuman.khandual@arm.com>
Reviewed-by: NChristophe Leroy <christophe.leroy@csgroup.eu>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: David S. Miller <davem@davemloft.net>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Khalid Aziz <khalid.aziz@oracle.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Will Deacon <will@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

Patch series "mm/mmap: Drop arch_vm_get_page_prot() and arch_filter_pgprot()", v7.

protection_map[] is an array based construct that translates given
vm_flags combination.  This array contains page protection map, which is
populated by the platform via [__S000 ..  __S111] and [__P000 ..  __P111]
exported macros.  Primary usage for protection_map[] is for
vm_get_page_prot(), which is used to determine page protection value for a
given vm_flags.  vm_get_page_prot() implementation, could again call
platform overrides arch_vm_get_page_prot() and arch_filter_pgprot().  Some
platforms override protection_map[] that was originally built with
__SXXX/__PXXX with different runtime values.

Currently there are multiple layers of abstraction i.e __SXXX/__PXXX
macros , protection_map[], arch_vm_get_page_prot() and
arch_filter_pgprot() built between the platform and generic MM, finally
defining vm_get_page_prot().

Hence this series proposes to drop later two abstraction levels and
instead just move the responsibility of defining vm_get_page_prot() to the
platform (still utilizing generic protection_map[] array) itself making it
clean and simple.

This first introduces ARCH_HAS_VM_GET_PAGE_PROT which enables the
platforms to define custom vm_get_page_prot().  This starts converting
platforms that define the overrides arch_filter_pgprot() or
arch_vm_get_page_prot() which enables for those constructs to be dropped
off completely.

The series has been inspired from an earlier discuss with Christoph Hellwig

https://lore.kernel.org/all/1632712920-8171-1-git-send-email-anshuman.khandual@arm.com/


This patch (of 7):

Add a new config ARCH_HAS_VM_GET_PAGE_PROT, which when subscribed enables
a given platform to define its own vm_get_page_prot() but still utilizing
the generic protection_map[] array.

Link: https://lkml.kernel.org/r/20220414062125.609297-1-anshuman.khandual@arm.com
Link: https://lkml.kernel.org/r/20220414062125.609297-2-anshuman.khandual@arm.comSigned-off-by: NAnshuman Khandual <anshuman.khandual@arm.com>
Reviewed-by: NChristophe Leroy <christophe.leroy@csgroup.eu>
Reviewed-by: NCatalin Marinas <catalin.marinas@arm.com>
Suggested-by: NChristoph Hellwig <hch@infradead.org>
Cc: David S. Miller <davem@davemloft.net>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Khalid Aziz <khalid.aziz@oracle.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Will Deacon <will@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

Use helper mlock_future_check() to check whether it's safe to enlarge the
locked_vm to simplify the code.  Minor readability improvement.

Link: https://lkml.kernel.org/r/20220402032231.64974-1-linmiaohe@huawei.comSigned-off-by: NMiaohe Lin <linmiaohe@huawei.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

protection_map[] maps vm_flags access combinations into page protection
value as defined by the platform via __PXXX and __SXXX macros.  The array
indices in protection_map[], represents vm_flags access combinations but
it's not very intuitive to derive.  This makes it clear and explicit.

Link: https://lkml.kernel.org/r/20220404031840.588321-3-anshuman.khandual@arm.comSigned-off-by: NAnshuman Khandual <anshuman.khandual@arm.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

Patch series "mm: protection_map[] cleanups".


This patch (of 2):

Although protection_map[] contains the platform defined page protection
map for a given vm_flags combination, vm_get_page_prot() is the right
interface to use.  This will also reduce dependency on protection_map[]
which is going to be dropped off completely later on.

Link: https://lkml.kernel.org/r/20220404031840.588321-1-anshuman.khandual@arm.com
Link: https://lkml.kernel.org/r/20220404031840.588321-2-anshuman.khandual@arm.comSigned-off-by: NAnshuman Khandual <anshuman.khandual@arm.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

free a large list of pages maybe cause rcu_sched starved on
non-preemptible kernels.  howerver free_unref_page_list maybe can't
cond_resched as it maybe called in interrupt or atomic context, especially
can't detect atomic context in CONFIG_PREEMPTION=n.

The issue is detected in guest with kvm cpu 200% overcommit, however I
didn't see the warning in the host with the same application.  I'm sure
that the patch is needed for guest kernel, but no sure for host.

To reproduce, set up two virtual machines in one host machine, per vm has
the same number cpu and half memory of host.  the run ltpstress.sh in per
vm, then will see rcu stall warning.kernel is preempt disabled, append
kernel command 'preempt=none' if enable dynamic preempt .  It could
detected in loongson machine(32 core, 128G mem) and ProLiant DL380
Gen9(x86 E5-2680, 28 core, 64G mem)

tlb flush batch count depends on PAGE_SIZE, it's too large if PAGE_SIZE >
4K, here limit free batch count with 512.  And add schedule point in
tlb_batch_pages_flush.

rcu: rcu_sched kthread starved for 5359 jiffies! g454793 f0x0
RCU_GP_WAIT_FQS(5) ->state=0x0 ->cpu=19
[...]
Call Trace:
   free_unref_page_list+0x19c/0x270
   release_pages+0x3cc/0x498
   tlb_flush_mmu_free+0x44/0x70
   zap_pte_range+0x450/0x738
   unmap_page_range+0x108/0x240
   unmap_vmas+0x74/0xf0
   unmap_region+0xb0/0x120
   do_munmap+0x264/0x438
   vm_munmap+0x58/0xa0
   sys_munmap+0x10/0x20
   syscall_common+0x24/0x38

Link: https://lkml.kernel.org/r/20220317072857.2635262-1-wangjianxing@loongson.cnSigned-off-by: NJianxing Wang <wangjianxing@loongson.cn>
Signed-off-by: NPeter Zijlstra <peterz@infradead.org>
Cc: Will Deacon <will@kernel.org>
Cc: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

In case the lock is actually not held at this point.

Link: https://lkml.kernel.org/r/5827758.TJ1SttVevJ@mobilepool36.emlix.comSigned-off-by: NRolf Eike Beer <eb@emlix.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

These might not be issues yet, but they make the script more fragile. 
Also by fixing them we give a better example to future readers, who might
copy/paste or otherwise re-use snippets from our script.

- Use "read -r", since we don't ever want read to be interpreting '\'
  characters as escape sequences...
- Quote variables, to deal with spaces properly.
- Use $() instead of the older and harder-to-nest ``.
- Get rid of superfluous "$" prefixes inside arithmetic $(()).

Link: https://lkml.kernel.org/r/20220421224928.1848230-2-axelrasmussen@google.comSigned-off-by: NAxel Rasmussen <axelrasmussen@google.com>
Cc: Shuah Khan <shuah@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

Previously, each test printed out its own header, dealt with its own
return code, etc.  By just putting this standard stuff in a function, we
can delete > 300 lines from the script.

This also makes adding future tests easier. And, it gets rid of various
inconsistencies that already exist:

- Some tests correctly deal with ksft_skip, but others don't.
- Some tests just print the executable name, others print arguments, and
  yet others print some comment in the header.
- Most tests print out a header with two separator lines, but not the
  HMM smoke test or the memfd_secret test, which only print one.
- We had a redundant "exit" at the end, with all the boilerplate it's an
  easy oversight.

Link: https://lkml.kernel.org/r/20220421224928.1848230-1-axelrasmussen@google.comSigned-off-by: NAxel Rasmussen <axelrasmussen@google.com>
Cc: Shuah Khan <shuah@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

This introduces three tests:

1) Sanity check soft dirty basic semantics: allocate area, clean,
   dirty, check if the SD bit is flipped.

2) Check VMA reuse: validate the VM_SOFTDIRTY usage

3) Check soft-dirty on huge pages

This was motivated by Will Deacon's fix commit 912efa17 ("mm: proc:
Invalidate TLB after clearing soft-dirty page state").  I was tracking the
same issue that he fixed, and this test would have caught it.

Link: https://lkml.kernel.org/r/20220420084036.4101604-2-usama.anjum@collabora.comSigned-off-by: NGabriel Krisman Bertazi <krisman@collabora.com>
Signed-off-by: NMuhammad Usama Anjum <usama.anjum@collabora.com>
Co-developed-by: NMuhammad Usama Anjum <usama.anjum@collabora.com>
Cc: Will Deacon <will@kernel.org>
Cc: Shuah Khan <shuah@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

Bring common functions to a new file while keeping code as much same as
possible.  These functions can be used in the new tests.  This helps in
avoiding code duplication.

Link: https://lkml.kernel.org/r/20220420084036.4101604-1-usama.anjum@collabora.comSigned-off-by: NMuhammad Usama Anjum <usama.anjum@collabora.com>
Acked-by: NDavid Hildenbrand <david@redhat.com>
Cc: Gabriel Krisman Bertazi <krisman@collabora.com>
Cc: Shuah Khan <shuah@kernel.org>
Cc: Will Deacon <will@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

Print three possible reasons /sys/kernel/debug/gup_test cannot be opened
to help users of this test diagnose failures.

Link: https://lkml.kernel.org/r/20220405214809.3351223-1-sidhartha.kumar@oracle.comSigned-off-by: NSidhartha Kumar <sidhartha.kumar@oracle.com>
Reviewed-by: NShuah Khan <skhan@linuxfoundation.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

The only user (DAX) of range and pmdpp parameters of
follow_invalidate_pte() is gone, it is safe to remove them and make it
static to simlify the code.  This is revertant of the following commits:

  09796395 ("mm: add follow_pte_pmd()")
  a4d1a885 ("dax: update to new mmu_notifier semantic")

There is only one caller of the follow_invalidate_pte().  So just fold it
into follow_pte() and remove it.

Link: https://lkml.kernel.org/r/20220403053957.10770-7-songmuchun@bytedance.comSigned-off-by: NMuchun Song <songmuchun@bytedance.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Cc: Alistair Popple <apopple@nvidia.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Jan Kara <jack@suse.cz>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Ralph Campbell <rcampbell@nvidia.com>
Cc: Ross Zwisler <zwisler@kernel.org>
Cc: Xiongchun Duan <duanxiongchun@bytedance.com>
Cc: Xiyu Yang <xiyuyang19@fudan.edu.cn>
Cc: Yang Shi <shy828301@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

Currently dax_mapping_entry_mkclean() fails to clean and write protect the
pte entry within a DAX PMD entry during an *sync operation.  This can
result in data loss in the following sequence:

  1) process A mmap write to DAX PMD, dirtying PMD radix tree entry and
     making the pmd entry dirty and writeable.
  2) process B mmap with the @offset (e.g. 4K) and @length (e.g. 4K)
     write to the same file, dirtying PMD radix tree entry (already
     done in 1)) and making the pte entry dirty and writeable.
  3) fsync, flushing out PMD data and cleaning the radix tree entry. We
     currently fail to mark the pte entry as clean and write protected
     since the vma of process B is not covered in dax_entry_mkclean().
  4) process B writes to the pte. These don't cause any page faults since
     the pte entry is dirty and writeable. The radix tree entry remains
     clean.
  5) fsync, which fails to flush the dirty PMD data because the radix tree
     entry was clean.
  6) crash - dirty data that should have been fsync'd as part of 5) could
     still have been in the processor cache, and is lost.

Just to use pfn_mkclean_range() to clean the pfns to fix this issue.

Link: https://lkml.kernel.org/r/20220403053957.10770-6-songmuchun@bytedance.com
Fixes: 4b4bb46d ("dax: clear dirty entry tags on cache flush")
Signed-off-by: NMuchun Song <songmuchun@bytedance.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Cc: Alistair Popple <apopple@nvidia.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Jan Kara <jack@suse.cz>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Ralph Campbell <rcampbell@nvidia.com>
Cc: Ross Zwisler <zwisler@kernel.org>
Cc: Xiongchun Duan <duanxiongchun@bytedance.com>
Cc: Xiyu Yang <xiyuyang19@fudan.edu.cn>
Cc: Yang Shi <shy828301@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

The devmap pages can not use page_vma_mapped_walk() to check if a huge
devmap page is mapped into a vma.  Add support for walking huge devmap
pages so that DAX can use it in the next patch.

Link: https://lkml.kernel.org/r/20220403053957.10770-5-songmuchun@bytedance.comSigned-off-by: NMuchun Song <songmuchun@bytedance.com>
Cc: Alistair Popple <apopple@nvidia.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Jan Kara <jack@suse.cz>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Ralph Campbell <rcampbell@nvidia.com>
Cc: Ross Zwisler <zwisler@kernel.org>
Cc: Xiongchun Duan <duanxiongchun@bytedance.com>
Cc: Xiyu Yang <xiyuyang19@fudan.edu.cn>
Cc: Yang Shi <shy828301@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

The page_mkclean_one() is supposed to be used with the pfn that has a
associated struct page, but not all the pfns (e.g.  DAX) have a struct
page.  Introduce a new function pfn_mkclean_range() to cleans the PTEs
(including PMDs) mapped with range of pfns which has no struct page
associated with them.  This helper will be used by DAX device in the next
patch to make pfns clean.

Link: https://lkml.kernel.org/r/20220403053957.10770-4-songmuchun@bytedance.comSigned-off-by: NMuchun Song <songmuchun@bytedance.com>
Cc: Alistair Popple <apopple@nvidia.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Jan Kara <jack@suse.cz>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Ralph Campbell <rcampbell@nvidia.com>
Cc: Ross Zwisler <zwisler@kernel.org>
Cc: Xiongchun Duan <duanxiongchun@bytedance.com>
Cc: Xiyu Yang <xiyuyang19@fudan.edu.cn>
Cc: Yang Shi <shy828301@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

The flush_cache_page() only remove a PAGE_SIZE sized range from the cache.
However, it does not cover the full pages in a THP except a head page. 
Replace it with flush_cache_range() to fix this issue.  This is just a
documentation issue with the respect to properly documenting the expected
usage of cache flushing before modifying the pmd.  However, in practice
this is not a problem due to the fact that DAX is not available on
architectures with virtually indexed caches per:

  commit d92576f1 ("dax: does not work correctly with virtual aliasing caches")

Link: https://lkml.kernel.org/r/20220403053957.10770-3-songmuchun@bytedance.com
Fixes: f729c8c9 ("dax: wrprotect pmd_t in dax_mapping_entry_mkclean")
Signed-off-by: NMuchun Song <songmuchun@bytedance.com>
Reviewed-by: NDan Williams <dan.j.williams@intel.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Cc: Alistair Popple <apopple@nvidia.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Hugh Dickins <hughd@google.com>
Cc: Jan Kara <jack@suse.cz>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Ralph Campbell <rcampbell@nvidia.com>
Cc: Ross Zwisler <zwisler@kernel.org>
Cc: Xiongchun Duan <duanxiongchun@bytedance.com>
Cc: Xiyu Yang <xiyuyang19@fudan.edu.cn>
Cc: Yang Shi <shy828301@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

Patch series "Fix some bugs related to ramp and dax", v7.

Patch 1-2 fix a cache flush bug, because subsequent patches depend on
those on those changes, there are placed in this series.  Patch 3-4 are
preparation for fixing a dax bug in patch 5.  Patch 6 is code cleanup
since the previous patch removes the usage of follow_invalidate_pte().


This patch (of 6):

The flush_cache_page() only remove a PAGE_SIZE sized range from the cache.
However, it does not cover the full pages in a THP except a head page. 
Replace it with flush_cache_range() to fix this issue.  At least, no
problems were found due to this.  Maybe because the architectures that
have virtual indexed caches is less.

Link: https://lkml.kernel.org/r/20220403053957.10770-1-songmuchun@bytedance.com
Link: https://lkml.kernel.org/r/20220403053957.10770-2-songmuchun@bytedance.com
Fixes: f27176cf ("mm: convert page_mkclean_one() to use page_vma_mapped_walk()")
Signed-off-by: NMuchun Song <songmuchun@bytedance.com>
Reviewed-by: NYang Shi <shy828301@gmail.com>
Reviewed-by: NDan Williams <dan.j.williams@intel.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Jan Kara <jack@suse.cz>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Alistair Popple <apopple@nvidia.com>
Cc: Yang Shi <shy828301@gmail.com>
Cc: Ralph Campbell <rcampbell@nvidia.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Xiyu Yang <xiyuyang19@fudan.edu.cn>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Ross Zwisler <zwisler@kernel.org>
Cc: Xiongchun Duan <duanxiongchun@bytedance.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

We can't assume pte_offset_map_lock will return same orig_pte value. So
it's necessary to reacquire the orig_pte or pte_unmap_unlock will unmap
the stale pte.

Link: https://lkml.kernel.org/r/20220416081416.23304-1-linmiaohe@huawei.com
Fixes: 9c276cc6 ("mm: introduce MADV_COLD")
Fixes: 854e9ed0 ("mm: support madvise(MADV_FREE)")
Signed-off-by: NMiaohe Lin <linmiaohe@huawei.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Hugh Dickins <hughd@google.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

At the time demote-on-reclaim was introduced, it was tied to
CONFIG_HOTPLUG_CPU + CONFIG_MIGRATE, but that is not really accurate.

The only two things we need to depend on are CONFIG_NUMA + CONFIG_MIGRATE,
so clean this up.  Furthermore, we only register the hotplug memory
notifier when the system has CONFIG_MEMORY_HOTPLUG.

Link: https://lkml.kernel.org/r/20220322224016.4574-1-osalvador@suse.deSigned-off-by: NOscar Salvador <osalvador@suse.de>
Suggested-by: N"Huang, Ying" <ying.huang@intel.com>
Reviewed-by: N"Huang, Ying" <ying.huang@intel.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Abhishek Goel <huntbag@linux.vnet.ibm.com>
Cc: Baolin Wang <baolin.wang@linux.alibaba.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

There is no need to validate the hugetlb page's refcount before trying to
freeze the hugetlb page's expected refcount, instead we can just rely on
the page_ref_freeze() to simplify the validation.

Moreover we are always under the page lock when migrating the hugetlb page
mapping, which means nowhere else can remove it from the page cache, so we
can remove the xas_load() validation under the i_pages lock.

Link: https://lkml.kernel.org/r/eb2fbbeaef2b1714097b9dec457426d682ee0635.1649676424.git.baolin.wang@linux.alibaba.comSigned-off-by: NBaolin Wang <baolin.wang@linux.alibaba.com>
Acked-by: NMike Kravetz <mike.kravetz@oracle.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: "Huang, Ying" <ying.huang@intel.com>
Cc: Muchun Song <songmuchun@bytedance.com>
Cc: Alistair Popple <apopple@nvidia.com>
Cc: Zi Yan <ziy@nvidia.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

When follow_page peeks a page, the page could be migrated and then be
offlined while it's still being used by the do_pages_stat_array().  Use
FOLL_GET to hold the page refcnt to fix this potential race.

Link: https://lkml.kernel.org/r/20220318111709.60311-12-linmiaohe@huawei.comSigned-off-by: NMiaohe Lin <linmiaohe@huawei.com>
Acked-by: N"Huang, Ying" <ying.huang@intel.com>
Reviewed-by: NMuchun Song <songmuchun@bytedance.com>
Cc: Alistair Popple <apopple@nvidia.com>
Cc: Baolin Wang <baolin.wang@linux.alibaba.com>
Cc: Zi Yan <ziy@nvidia.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

If we failed to setup hotplug state callbacks for mm/demotion:online in
some corner cases, node_demotion will be left uninitialized.  Invalid node
might be returned from the next_demotion_node() when doing reclaim-based
migration.  Use kcalloc to allocate node_demotion to fix the issue.

Link: https://lkml.kernel.org/r/20220318111709.60311-11-linmiaohe@huawei.com
Fixes: ac16ec83 ("mm: migrate: support multiple target nodes demotion")
Signed-off-by: NMiaohe Lin <linmiaohe@huawei.com>
Reviewed-by: N"Huang, Ying" <ying.huang@intel.com>
Cc: Alistair Popple <apopple@nvidia.com>
Cc: Baolin Wang <baolin.wang@linux.alibaba.com>
Cc: Muchun Song <songmuchun@bytedance.com>
Cc: Zi Yan <ziy@nvidia.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

In -ENOMEM case, there might be some subpages of fail-to-migrate THPs left
in thp_split_pages list.  We should move them back to migration list so
that they could be put back to the right list by the caller otherwise the
page refcnt will be leaked here.  Also adjust nr_failed and nr_thp_failed
accordingly to make vm events account more accurate.

Link: https://lkml.kernel.org/r/20220318111709.60311-10-linmiaohe@huawei.com
Fixes: b5bade97 ("mm: migrate: fix the return value of migrate_pages()")
Signed-off-by: NMiaohe Lin <linmiaohe@huawei.com>
Reviewed-by: NZi Yan <ziy@nvidia.com>
Reviewed-by: N"Huang, Ying" <ying.huang@intel.com>
Reviewed-by: NBaolin Wang <baolin.wang@linux.alibaba.com>
Reviewed-by: NMuchun Song <songmuchun@bytedance.com>
Cc: Alistair Popple <apopple@nvidia.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

Remove the duplicated codes in migrate_pages to simplify the code.  Minor
readability improvement.  No functional change intended.

Link: https://lkml.kernel.org/r/20220318111709.60311-9-linmiaohe@huawei.comSigned-off-by: NMiaohe Lin <linmiaohe@huawei.com>
Reviewed-by: NZi Yan <ziy@nvidia.com>
Reviewed-by: NBaolin Wang <baolin.wang@linux.alibaba.com>
Cc: Alistair Popple <apopple@nvidia.com>
Cc: "Huang, Ying" <ying.huang@intel.com>
Cc: Muchun Song <songmuchun@bytedance.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

Avoid unneeded next_pass and this_pass initialization as they're always
set before using to save possible cpu cycles when there are plenty of
nodes in the system.

Link: https://lkml.kernel.org/r/20220318111709.60311-8-linmiaohe@huawei.comSigned-off-by: NMiaohe Lin <linmiaohe@huawei.com>
Reviewed-by: NMuchun Song <songmuchun@bytedance.com>
Reviewed-by: NBaolin Wang <baolin.wang@linux.alibaba.com>
Cc: Alistair Popple <apopple@nvidia.com>
Cc: "Huang, Ying" <ying.huang@intel.com>
Cc: Zi Yan <ziy@nvidia.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

We could use helper macro min to help set the chunk_nr to simplify the
code.

Link: https://lkml.kernel.org/r/20220318111709.60311-7-linmiaohe@huawei.comSigned-off-by: NMiaohe Lin <linmiaohe@huawei.com>
Reviewed-by: NMuchun Song <songmuchun@bytedance.com>
Cc: Alistair Popple <apopple@nvidia.com>
Cc: Baolin Wang <baolin.wang@linux.alibaba.com>
Cc: "Huang, Ying" <ying.huang@intel.com>
Cc: Zi Yan <ziy@nvidia.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

We could use helper function vma_lookup() to lookup the needed vma to
simplify the code.

Link: https://lkml.kernel.org/r/20220318111709.60311-6-linmiaohe@huawei.comSigned-off-by: NMiaohe Lin <linmiaohe@huawei.com>
Cc: Alistair Popple <apopple@nvidia.com>
Cc: Baolin Wang <baolin.wang@linux.alibaba.com>
Cc: "Huang, Ying" <ying.huang@intel.com>
Cc: Muchun Song <songmuchun@bytedance.com>
Cc: Zi Yan <ziy@nvidia.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

We can use page_is_file_lru() directly to help account the isolated pages
to simplify the code a bit.

Link: https://lkml.kernel.org/r/20220318111709.60311-4-linmiaohe@huawei.comSigned-off-by: NMiaohe Lin <linmiaohe@huawei.com>
Cc: Alistair Popple <apopple@nvidia.com>
Cc: Baolin Wang <baolin.wang@linux.alibaba.com>
Cc: "Huang, Ying" <ying.huang@intel.com>
Cc: Muchun Song <songmuchun@bytedance.com>
Cc: Zi Yan <ziy@nvidia.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

Patch series "A few cleanup and fixup patches for migration", v2.

This series contains a few patches to remove unneeded variables, jump
label and use helper to simplify the code.  Also we fix some bugs such as
page refcounts leak , invalid node access and so on.  More details can be
found in the respective changelogs.


This patch (of 11):

When mapping_locked is true, TTU_RMAP_LOCKED is always set to ttu.  We can
check ttu instead so mapping_locked can be removed.  And ttu is either 0
or TTU_RMAP_LOCKED now.  Change '|=' to '=' to reflect this.

Link: https://lkml.kernel.org/r/20220318111709.60311-1-linmiaohe@huawei.com
Link: https://lkml.kernel.org/r/20220318111709.60311-2-linmiaohe@huawei.comSigned-off-by: NMiaohe Lin <linmiaohe@huawei.com>
Reviewed-by: NMuchun Song <songmuchun@bytedance.com>
Cc: Zi Yan <ziy@nvidia.com>
Cc: Baolin Wang <baolin.wang@linux.alibaba.com>
Cc: "Huang, Ying" <ying.huang@intel.com>
Cc: Muchun Song <songmuchun@bytedance.com>
Cc: Alistair Popple <apopple@nvidia.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

Add some basic migration tests and in particular tests that will
stress both the pte and pmd migration entry wait paths.

Link: https://lkml.kernel.org/r/20220324014349.229253-1-apopple@nvidia.comSigned-off-by: NAlistair Popple <apopple@nvidia.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Jan Kara <jack@suse.cz>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: Ralph Campbell <rcampbell@nvidia.com>
Cc: Muchun Song <songmuchun@bytedance.com>
Cc: John Hubbard <jhubbard@nvidia.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Cc: Shuah Khan <shuah@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

Since commit e5947d23 ("mm: mempolicy: don't have to split pmd for
huge zero page"), THP is never splited in queue_pages_pmd.  Thus 2 is
never returned now.  We can remove such unnecessary ret != 2 check and
clean up the relevant comment.  Minor improvements in readability.

Link: https://lkml.kernel.org/r/20220419122234.45083-1-linmiaohe@huawei.comSigned-off-by: NMiaohe Lin <linmiaohe@huawei.com>
Reviewed-by: NYang Shi <shy828301@gmail.com>
Acked-by: NDavid Rientjes <rientjes@google.com>
Cc: Zi Yan <ziy@nvidia.com>
Cc: Michal Hocko <mhocko@suse.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

Compaction sysfs file is created via compaction_register_node in
register_node.  But we forgot to remove it in unregister_node.  Thus
compaction sysfs file is leaked.  Using compaction_unregister_node to fix
this issue.

Link: https://lkml.kernel.org/r/20220401070905.43679-1-linmiaohe@huawei.com
Fixes: ed4a6d7f ("mm: compaction: add /sys trigger for per-node memory compaction")
Signed-off-by: NMiaohe Lin <linmiaohe@huawei.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Rafael J. Wysocki <rafael@kernel.org>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Minchan Kim <minchan.kim@gmail.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

Use helper isolation_suitable() to check whether page is suitable to
isolate to simplify the code.  Minor readability improvement.

Link: https://lkml.kernel.org/r/20220322110750.60311-1-linmiaohe@huawei.comSigned-off-by: NMiaohe Lin <linmiaohe@huawei.com>
Reviewed-by: NAndrew Morton <akpm@linux-foundation.org>
Reviewed-by: NWei Yang <richard.weiyang@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

The only caller z3fold_free() never calls free_handle() in PAGE_HEADLESS
case.  Remove this unneeded check.

Link: https://lkml.kernel.org/r/20220308134311.59086-9-linmiaohe@huawei.comSigned-off-by: NMiaohe Lin <linmiaohe@huawei.com>
Reviewed-by: NVitaly Wool <vitaly.wool@konsulko.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

do_compact_page() will do list_del_init(&zhdr->buddy) for us.  Remove this
extra one to save some possible cpu cycles.

Link: https://lkml.kernel.org/r/20220308134311.59086-8-linmiaohe@huawei.comSigned-off-by: NMiaohe Lin <linmiaohe@huawei.com>
Reviewed-by: NVitaly Wool <vitaly.wool@konsulko.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

The z3fold will always do atomic64_dec(&pool->pages_nr) when the
__release_z3fold_page() is called.  Thus we can move decrement of
pool->pages_nr into __release_z3fold_page() to simplify the code.

Also we can reduce the size of z3fold.o ~1k.

Without this patch:
   text	   data	    bss	    dec	    hex	filename
  15444	   1376	      8	  16828	   41bc	mm/z3fold.o
With this patch:
   text	   data	    bss	    dec	    hex	filename
  15044	   1248	      8	  16300	   3fac	mm/z3fold.o

Link: https://lkml.kernel.org/r/20220308134311.59086-7-linmiaohe@huawei.comSigned-off-by: NMiaohe Lin <linmiaohe@huawei.com>
Cc: Vitaly Wool <vitaly.wool@konsulko.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

The local variable l holds the address of unbuddied[i] which won't change
after we take the pool lock.  Remove it to avoid confusion.

Link: https://lkml.kernel.org/r/20220308134311.59086-6-linmiaohe@huawei.comSigned-off-by: NMiaohe Lin <linmiaohe@huawei.com>
Reviewed-by: NVitaly Wool <vitaly.wool@konsulko.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

Page->page_type and PagePrivate are not used in z3fold.  We should remove
these confusing unneeded operations.  The z3fold do these here is due to
referring to zsmalloc's migration code which does need these operations.

Link: https://lkml.kernel.org/r/20220308134311.59086-5-linmiaohe@huawei.comSigned-off-by: NMiaohe Lin <linmiaohe@huawei.com>
Reviewed-by: NVitaly Wool <vitaly.wool@konsulko.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>