Skip to content

K
Kernel

openeuler / Kernel
1 年多 前同步成功

通知 8
Star 0
Fork 0
K
Kernel
切换分支/标签
  1. 20 9月, 2020 3 次提交
  3. 18 9月, 2020 2 次提交
    • S
      percpu: fix first chunk size calculation for populated bitmap · b3b33d3c
      由 Sunghyun Jin 提交于 
      Variable populated, which is a member of struct pcpu_chunk, is used as a
unit of size of unsigned long.
However, size of populated is miscounted. So, I fix this minor part.

Fixes: 8ab16c43 ("percpu: change the number of pages marked in the first_chunk pop bitmap")
Cc: <stable@vger.kernel.org> # 4.14+
Signed-off-by: NSunghyun Jin <mcsmonk@gmail.com>
Signed-off-by: NDennis Zhou <dennis@kernel.org>
      b3b33d3c
    • L
      mm: allow a controlled amount of unfairness in the page lock · 5ef64cc8
      由 Linus Torvalds 提交于 
      Commit 2a9127fc ("mm: rewrite wait_on_page_bit_common() logic") made
the page locking entirely fair, in that if a waiter came in while the
lock was held, the lock would be transferred to the lockers strictly in
order.

That was intended to finally get rid of the long-reported watchdog
failures that involved the page lock under extreme load, where a process
could end up waiting essentially forever, as other page lockers stole
the lock from under it.

It also improved some benchmarks, but it ended up causing huge
performance regressions on others, simply because fair lock behavior
doesn't end up giving out the lock as aggressively, causing better
worst-case latency, but potentially much worse average latencies and
throughput.

Instead of reverting that change entirely, this introduces a controlled
amount of unfairness, with a sysctl knob to tune it if somebody needs
to.  But the default value should hopefully be good for any normal load,
allowing a few rounds of lock stealing, but enforcing the strict
ordering before the lock has been stolen too many times.

There is also a hint from Matthieu Baerts that the fair page coloring
may end up exposing an ABBA deadlock that is hidden by the usual
optimistic lock stealing, and while the unfairness doesn't fix the
fundamental issue (and I'm still looking at that), it avoids it in
practice.

The amount of unfairness can be modified by writing a new value to the
'sysctl_page_lock_unfairness' variable (default value of 5, exposed
through /proc/sys/vm/page_lock_unfairness), but that is hopefully
something we'd use mainly for debugging rather than being necessary for
any deep system tuning.

This whole issue has exposed just how critical the page lock can be, and
how contended it gets under certain locks.  And the main contention
doesn't really seem to be anything related to IO (which was the origin
of this lock), but for things like just verifying that the page file
mapping is stable while faulting in the page into a page table.

Link: https://lore.kernel.org/linux-fsdevel/ed8442fd-6f54-dd84-cd4a-941e8b7ee603@MichaelLarabel.com/
Link: https://www.phoronix.com/scan.php?page=article&item=linux-50-59&num=1
Link: https://lore.kernel.org/linux-fsdevel/c560a38d-8313-51fb-b1ec-e904bd8836bc@tessares.net/Reported-and-tested-by: NMichael Larabel <Michael@michaellarabel.com>
Tested-by: NMatthieu Baerts <matthieu.baerts@tessares.net>
Cc: Dave Chinner <david@fromorbit.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Chris Mason <clm@fb.com>
Cc: Jan Kara <jack@suse.cz>
Cc: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      5ef64cc8
  5. 06 9月, 2020 12 次提交
    • D
      mm/khugepaged.c: fix khugepaged's request size in collapse_file · e5a59d30
      由 David Howells 提交于 
      collapse_file() in khugepaged passes PAGE_SIZE as the number of pages to
be read to page_cache_sync_readahead().  The intent was probably to read
a single page.  Fix it to use the number of pages to the end of the
window instead.

Fixes: 99cb0dbd ("mm,thp: add read-only THP support for (non-shmem) FS")
Signed-off-by: NDavid Howells <dhowells@redhat.com>
Signed-off-by: NMatthew Wilcox (Oracle) <willy@infradead.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Reviewed-by: NMatthew Wilcox (Oracle) <willy@infradead.org>
Acked-by: NSong Liu <songliubraving@fb.com>
Acked-by: NYang Shi <shy828301@gmail.com>
Acked-by: NPankaj Gupta <pankaj.gupta.linux@gmail.com>
Cc: Eric Biggers <ebiggers@google.com>
Link: https://lkml.kernel.org/r/20200903140844.14194-2-willy@infradead.orgSigned-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      e5a59d30
    • M
      mm/hugetlb: fix a race between hugetlb sysctl handlers · 17743798
      由 Muchun Song 提交于 
      There is a race between the assignment of `table->data` and write value
to the pointer of `table->data` in the __do_proc_doulongvec_minmax() on
the other thread.

  CPU0:                                 CPU1:
                                        proc_sys_write
  hugetlb_sysctl_handler                  proc_sys_call_handler
  hugetlb_sysctl_handler_common             hugetlb_sysctl_handler
    table->data = &tmp;                       hugetlb_sysctl_handler_common
                                                table->data = &tmp;
      proc_doulongvec_minmax
        do_proc_doulongvec_minmax           sysctl_head_finish
          __do_proc_doulongvec_minmax         unuse_table
            i = table->data;
            *i = val;  // corrupt CPU1's stack

Fix this by duplicating the `table`, and only update the duplicate of
it.  And introduce a helper of proc_hugetlb_doulongvec_minmax() to
simplify the code.

The following oops was seen:

    BUG: kernel NULL pointer dereference, address: 0000000000000000
    #PF: supervisor instruction fetch in kernel mode
    #PF: error_code(0x0010) - not-present page
    Code: Bad RIP value.
    ...
    Call Trace:
     ? set_max_huge_pages+0x3da/0x4f0
     ? alloc_pool_huge_page+0x150/0x150
     ? proc_doulongvec_minmax+0x46/0x60
     ? hugetlb_sysctl_handler_common+0x1c7/0x200
     ? nr_hugepages_store+0x20/0x20
     ? copy_fd_bitmaps+0x170/0x170
     ? hugetlb_sysctl_handler+0x1e/0x20
     ? proc_sys_call_handler+0x2f1/0x300
     ? unregister_sysctl_table+0xb0/0xb0
     ? __fd_install+0x78/0x100
     ? proc_sys_write+0x14/0x20
     ? __vfs_write+0x4d/0x90
     ? vfs_write+0xef/0x240
     ? ksys_write+0xc0/0x160
     ? __ia32_sys_read+0x50/0x50
     ? __close_fd+0x129/0x150
     ? __x64_sys_write+0x43/0x50
     ? do_syscall_64+0x6c/0x200
     ? entry_SYSCALL_64_after_hwframe+0x44/0xa9

Fixes: e5ff2159 ("hugetlb: multiple hstates for multiple page sizes")
Signed-off-by: NMuchun Song <songmuchun@bytedance.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Reviewed-by: NMike Kravetz <mike.kravetz@oracle.com>
Cc: Andi Kleen <ak@linux.intel.com>
Link: http://lkml.kernel.org/r/20200828031146.43035-1-songmuchun@bytedance.comSigned-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      17743798
    • L
      mm/hugetlb: try preferred node first when alloc gigantic page from cma · 953f064a
      由 Li Xinhai 提交于 
      Since commit cf11e85f ("mm: hugetlb: optionally allocate gigantic
hugepages using cma"), the gigantic page would be allocated from node
which is not the preferred node, although there are pages available from
that node.  The reason is that the nid parameter has been ignored in
alloc_gigantic_page().

Besides, the __GFP_THISNODE also need be checked if user required to
alloc only from the preferred node.

After this patch, the preferred node is tried first before other allowed
nodes, and don't try to allocate from other nodes if __GFP_THISNODE is
specified.  If user don't specify the preferred node, the current node
will be used as preferred node, which makes sure consistent behavior of
allocating gigantic and non-gigantic hugetlb page.

Fixes: cf11e85f ("mm: hugetlb: optionally allocate gigantic hugepages using cma")
Signed-off-by: NLi Xinhai <lixinhai.lxh@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Reviewed-by: NMike Kravetz <mike.kravetz@oracle.com>
Acked-by: NMichal Hocko <mhocko@suse.com>
Cc: Roman Gushchin <guro@fb.com>
Link: https://lkml.kernel.org/r/20200902025016.697260-1-lixinhai.lxh@gmail.comSigned-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      953f064a
    • R
      mm/migrate: preserve soft dirty in remove_migration_pte() · 3d321bf8
      由 Ralph Campbell 提交于 
      The code to remove a migration PTE and replace it with a device private
PTE was not copying the soft dirty bit from the migration entry.  This
could lead to page contents not being marked dirty when faulting the page
back from device private memory.
Signed-off-by: NRalph Campbell <rcampbell@nvidia.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Cc: Jerome Glisse <jglisse@redhat.com>
Cc: Alistair Popple <apopple@nvidia.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Jason Gunthorpe <jgg@nvidia.com>
Cc: Bharata B Rao <bharata@linux.ibm.com>
Link: https://lkml.kernel.org/r/20200831212222.22409-3-rcampbell@nvidia.comSigned-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      3d321bf8
    • R
      mm/migrate: remove unnecessary is_zone_device_page() check · 6128763f
      由 Ralph Campbell 提交于 
      Patch series "mm/migrate: preserve soft dirty in remove_migration_pte()".

I happened to notice this from code inspection after seeing Alistair
Popple's patch ("mm/rmap: Fixup copying of soft dirty and uffd ptes").

This patch (of 2):

The check for is_zone_device_page() and is_device_private_page() is
unnecessary since the latter is sufficient to determine if the page is a
device private page.  Simplify the code for easier reading.
Signed-off-by: NRalph Campbell <rcampbell@nvidia.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Cc: Jerome Glisse <jglisse@redhat.com>
Cc: Alistair Popple <apopple@nvidia.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Jason Gunthorpe <jgg@nvidia.com>
Cc: Bharata B Rao <bharata@linux.ibm.com>
Link: https://lkml.kernel.org/r/20200831212222.22409-1-rcampbell@nvidia.com
Link: https://lkml.kernel.org/r/20200831212222.22409-2-rcampbell@nvidia.comSigned-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      6128763f
    • A
      mm/rmap: fixup copying of soft dirty and uffd ptes · ad7df764
      由 Alistair Popple 提交于 
      During memory migration a pte is temporarily replaced with a migration
swap pte.  Some pte bits from the existing mapping such as the soft-dirty
and uffd write-protect bits are preserved by copying these to the
temporary migration swap pte.

However these bits are not stored at the same location for swap and
non-swap ptes.  Therefore testing these bits requires using the
appropriate helper function for the given pte type.

Unfortunately several code locations were found where the wrong helper
function is being used to test soft_dirty and uffd_wp bits which leads to
them getting incorrectly set or cleared during page-migration.

Fix these by using the correct tests based on pte type.

Fixes: a5430dda ("mm/migrate: support un-addressable ZONE_DEVICE page in migration")
Fixes: 8c3328f1 ("mm/migrate: migrate_vma() unmap page from vma while collecting pages")
Fixes: f45ec5ff ("userfaultfd: wp: support swap and page migration")
Signed-off-by: NAlistair Popple <alistair@popple.id.au>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Reviewed-by: NPeter Xu <peterx@redhat.com>
Cc: Jérôme Glisse <jglisse@redhat.com>
Cc: John Hubbard <jhubbard@nvidia.com>
Cc: Ralph Campbell <rcampbell@nvidia.com>
Cc: Alistair Popple <alistair@popple.id.au>
Cc: <stable@vger.kernel.org>
Link: https://lkml.kernel.org/r/20200825064232.10023-2-alistair@popple.id.auSigned-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      ad7df764
    • A
      mm/migrate: fixup setting UFFD_WP flag · ebdf8321
      由 Alistair Popple 提交于 
      Commit f45ec5ff ("userfaultfd: wp: support swap and page migration")
introduced support for tracking the uffd wp bit during page migration.
However the non-swap PTE variant was used to set the flag for zone device
private pages which are a type of swap page.

This leads to corruption of the swap offset if the original PTE has the
uffd_wp flag set.

Fixes: f45ec5ff ("userfaultfd: wp: support swap and page migration")
Signed-off-by: NAlistair Popple <alistair@popple.id.au>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Reviewed-by: NPeter Xu <peterx@redhat.com>
Cc: Jérôme Glisse <jglisse@redhat.com>
Cc: John Hubbard <jhubbard@nvidia.com>
Cc: Ralph Campbell <rcampbell@nvidia.com>
Link: https://lkml.kernel.org/r/20200825064232.10023-1-alistair@popple.id.auSigned-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      ebdf8321
    • Y
      mm: madvise: fix vma user-after-free · 7867fd7c
      由 Yang Shi 提交于 
      The syzbot reported the below use-after-free:

  BUG: KASAN: use-after-free in madvise_willneed mm/madvise.c:293 [inline]
  BUG: KASAN: use-after-free in madvise_vma mm/madvise.c:942 [inline]
  BUG: KASAN: use-after-free in do_madvise.part.0+0x1c8b/0x1cf0 mm/madvise.c:1145
  Read of size 8 at addr ffff8880a6163eb0 by task syz-executor.0/9996

  CPU: 0 PID: 9996 Comm: syz-executor.0 Not tainted 5.9.0-rc1-syzkaller #0
  Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
  Call Trace:
    __dump_stack lib/dump_stack.c:77 [inline]
    dump_stack+0x18f/0x20d lib/dump_stack.c:118
    print_address_description.constprop.0.cold+0xae/0x497 mm/kasan/report.c:383
    __kasan_report mm/kasan/report.c:513 [inline]
    kasan_report.cold+0x1f/0x37 mm/kasan/report.c:530
    madvise_willneed mm/madvise.c:293 [inline]
    madvise_vma mm/madvise.c:942 [inline]
    do_madvise.part.0+0x1c8b/0x1cf0 mm/madvise.c:1145
    do_madvise mm/madvise.c:1169 [inline]
    __do_sys_madvise mm/madvise.c:1171 [inline]
    __se_sys_madvise mm/madvise.c:1169 [inline]
    __x64_sys_madvise+0xd9/0x110 mm/madvise.c:1169
    do_syscall_64+0x2d/0x70 arch/x86/entry/common.c:46
    entry_SYSCALL_64_after_hwframe+0x44/0xa9

  Allocated by task 9992:
    kmem_cache_alloc+0x138/0x3a0 mm/slab.c:3482
    vm_area_alloc+0x1c/0x110 kernel/fork.c:347
    mmap_region+0x8e5/0x1780 mm/mmap.c:1743
    do_mmap+0xcf9/0x11d0 mm/mmap.c:1545
    vm_mmap_pgoff+0x195/0x200 mm/util.c:506
    ksys_mmap_pgoff+0x43a/0x560 mm/mmap.c:1596
    do_syscall_64+0x2d/0x70 arch/x86/entry/common.c:46
    entry_SYSCALL_64_after_hwframe+0x44/0xa9

  Freed by task 9992:
    kmem_cache_free.part.0+0x67/0x1f0 mm/slab.c:3693
    remove_vma+0x132/0x170 mm/mmap.c:184
    remove_vma_list mm/mmap.c:2613 [inline]
    __do_munmap+0x743/0x1170 mm/mmap.c:2869
    do_munmap mm/mmap.c:2877 [inline]
    mmap_region+0x257/0x1780 mm/mmap.c:1716
    do_mmap+0xcf9/0x11d0 mm/mmap.c:1545
    vm_mmap_pgoff+0x195/0x200 mm/util.c:506
    ksys_mmap_pgoff+0x43a/0x560 mm/mmap.c:1596
    do_syscall_64+0x2d/0x70 arch/x86/entry/common.c:46
    entry_SYSCALL_64_after_hwframe+0x44/0xa9

It is because vma is accessed after releasing mmap_lock, but someone
else acquired the mmap_lock and the vma is gone.

Releasing mmap_lock after accessing vma should fix the problem.

Fixes: 692fe624 ("mm: Handle MADV_WILLNEED through vfs_fadvise()")
Reported-by: syzbot+b90df26038d1d5d85c97@syzkaller.appspotmail.com
Signed-off-by: NYang Shi <shy828301@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Reviewed-by: NAndrew Morton <akpm@linux-foundation.org>
Reviewed-by: NJan Kara <jack@suse.cz>
Cc: <stable@vger.kernel.org>	[5.4+]
Link: https://lkml.kernel.org/r/20200816141204.162624-1-shy828301@gmail.comSigned-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      7867fd7c
    • J
      mm: track page table modifications in __apply_to_page_range() · e80d3909
      由 Joerg Roedel 提交于 
      __apply_to_page_range() is also used to change and/or allocate
page-table pages in the vmalloc area of the address space.  Make sure
these changes get synchronized to other page-tables in the system by
calling arch_sync_kernel_mappings() when necessary.

The impact appears limited to x86-32, where apply_to_page_range may miss
updating the PMD.  That leads to explosions in drivers like

  BUG: unable to handle page fault for address: fe036000
  #PF: supervisor write access in kernel mode
  #PF: error_code(0x0002) - not-present page
  *pde = 00000000
  Oops: 0002 [#1] SMP
  CPU: 3 PID: 1300 Comm: gem_concurrent_ Not tainted 5.9.0-rc1+ #16
  Hardware name:  /NUC6i3SYB, BIOS SYSKLi35.86A.0024.2015.1027.2142 10/27/2015
  EIP: __execlists_context_alloc+0x132/0x2d0 [i915]
  Code: 31 d2 89 f0 e8 2f 55 02 00 89 45 e8 3d 00 f0 ff ff 0f 87 11 01 00 00 8b 4d e8 03 4b 30 b8 5a 5a 5a 5a ba 01 00 00 00 8d 79 04 <c7> 01 5a 5a 5a 5a c7 81 fc 0f 00 00 5a 5a 5a 5a 83 e7 fc 29 f9 81
  EAX: 5a5a5a5a EBX: f60ca000 ECX: fe036000 EDX: 00000001
  ESI: f43b7340 EDI: fe036004 EBP: f6389cb8 ESP: f6389c9c
  DS: 007b ES: 007b FS: 00d8 GS: 00e0 SS: 0068 EFLAGS: 00010286
  CR0: 80050033 CR2: fe036000 CR3: 2d361000 CR4: 001506d0
  DR0: 00000000 DR1: 00000000 DR2: 00000000 DR3: 00000000
  DR6: fffe0ff0 DR7: 00000400
  Call Trace:
    execlists_context_alloc+0x10/0x20 [i915]
    intel_context_alloc_state+0x3f/0x70 [i915]
    __intel_context_do_pin+0x117/0x170 [i915]
    i915_gem_do_execbuffer+0xcc7/0x2500 [i915]
    i915_gem_execbuffer2_ioctl+0xcd/0x1f0 [i915]
    drm_ioctl_kernel+0x8f/0xd0
    drm_ioctl+0x223/0x3d0
    __ia32_sys_ioctl+0x1ab/0x760
    __do_fast_syscall_32+0x3f/0x70
    do_fast_syscall_32+0x29/0x60
    do_SYSENTER_32+0x15/0x20
    entry_SYSENTER_32+0x9f/0xf2
  EIP: 0xb7f28559
  Code: 03 74 c0 01 10 05 03 74 b8 01 10 06 03 74 b4 01 10 07 03 74 b0 01 10 08 03 74 d8 01 00 00 00 00 00 51 52 55 89 e5 0f 34 cd 80 <5d> 5a 59 c3 90 90 90 90 8d 76 00 58 b8 77 00 00 00 cd 80 90 8d 76
  EAX: ffffffda EBX: 00000005 ECX: c0406469 EDX: bf95556c
  ESI: b7e68000 EDI: c0406469 EBP: 00000005 ESP: bf9554d8
  DS: 007b ES: 007b FS: 0000 GS: 0033 SS: 007b EFLAGS: 00000296
  Modules linked in: i915 x86_pkg_temp_thermal intel_powerclamp crc32_pclmul crc32c_intel intel_cstate intel_uncore intel_gtt drm_kms_helper intel_pch_thermal video button autofs4 i2c_i801 i2c_smbus fan
  CR2: 00000000fe036000

It looks like kasan, xen and i915 are vulnerable.

Actual impact is "on thinkpad X60 in 5.9-rc1, screen starts blinking
after 30-or-so minutes, and machine is unusable"

[sfr@canb.auug.org.au: ARCH_PAGE_TABLE_SYNC_MASK needs vmalloc.h]
  Link: https://lkml.kernel.org/r/20200825172508.16800a4f@canb.auug.org.au
[chris@chris-wilson.co.uk: changelog addition]
[pavel@ucw.cz: changelog addition]

Fixes: 2ba3e694 ("mm/vmalloc: track which page-table levels were modified")
Fixes: 86cf69f1 ("x86/mm/32: implement arch_sync_kernel_mappings()")
Signed-off-by: NJoerg Roedel <jroedel@suse.de>
Signed-off-by: NStephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Tested-by: Chris Wilson <chris@chris-wilson.co.uk>	[x86-32]
Tested-by: NPavel Machek <pavel@ucw.cz>
Acked-by: NLinus Torvalds <torvalds@linux-foundation.org>
Cc: <stable@vger.kernel.org>	[5.8+]
Link: https://lkml.kernel.org/r/20200821123746.16904-1-joro@8bytes.orgSigned-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      e80d3909
    • E
      mm: slub: fix conversion of freelist_corrupted() · dc07a728
      由 Eugeniu Rosca 提交于 
      Commit 52f23478 ("mm/slub.c: fix corrupted freechain in
deactivate_slab()") suffered an update when picked up from LKML [1].

Specifically, relocating 'freelist = NULL' into 'freelist_corrupted()'
created a no-op statement.  Fix it by sticking to the behavior intended
in the original patch [1].  In addition, make freelist_corrupted()
immune to passing NULL instead of &freelist.

The issue has been spotted via static analysis and code review.

[1] https://lore.kernel.org/linux-mm/20200331031450.12182-1-dongli.zhang@oracle.com/

Fixes: 52f23478 ("mm/slub.c: fix corrupted freechain in deactivate_slab()")
Signed-off-by: NEugeniu Rosca <erosca@de.adit-jv.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Cc: Dongli Zhang <dongli.zhang@oracle.com>
Cc: Joe Jin <joe.jin@oracle.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: <stable@vger.kernel.org>
Link: https://lkml.kernel.org/r/20200824130643.10291-1-erosca@de.adit-jv.comSigned-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      dc07a728
    • X
      mm: memcg: fix memcg reclaim soft lockup · e3336cab
      由 Xunlei Pang 提交于 
      We've met softlockup with "CONFIG_PREEMPT_NONE=y", when the target memcg
doesn't have any reclaimable memory.

It can be easily reproduced as below:

  watchdog: BUG: soft lockup - CPU#0 stuck for 111s![memcg_test:2204]
  CPU: 0 PID: 2204 Comm: memcg_test Not tainted 5.9.0-rc2+ #12
  Call Trace:
    shrink_lruvec+0x49f/0x640
    shrink_node+0x2a6/0x6f0
    do_try_to_free_pages+0xe9/0x3e0
    try_to_free_mem_cgroup_pages+0xef/0x1f0
    try_charge+0x2c1/0x750
    mem_cgroup_charge+0xd7/0x240
    __add_to_page_cache_locked+0x2fd/0x370
    add_to_page_cache_lru+0x4a/0xc0
    pagecache_get_page+0x10b/0x2f0
    filemap_fault+0x661/0xad0
    ext4_filemap_fault+0x2c/0x40
    __do_fault+0x4d/0xf9
    handle_mm_fault+0x1080/0x1790

It only happens on our 1-vcpu instances, because there's no chance for
oom reaper to run to reclaim the to-be-killed process.

Add a cond_resched() at the upper shrink_node_memcgs() to solve this
issue, this will mean that we will get a scheduling point for each memcg
in the reclaimed hierarchy without any dependency on the reclaimable
memory in that memcg thus making it more predictable.
Suggested-by: NMichal Hocko <mhocko@suse.com>
Signed-off-by: NXunlei Pang <xlpang@linux.alibaba.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Acked-by: NChris Down <chris@chrisdown.name>
Acked-by: NMichal Hocko <mhocko@suse.com>
Acked-by: NJohannes Weiner <hannes@cmpxchg.org>
Link: http://lkml.kernel.org/r/1598495549-67324-1-git-send-email-xlpang@linux.alibaba.comSigned-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      e3336cab
    • M
      memcg: fix use-after-free in uncharge_batch · f1796544
      由 Michal Hocko 提交于 
      syzbot has reported an use-after-free in the uncharge_batch path

  BUG: KASAN: use-after-free in instrument_atomic_write include/linux/instrumented.h:71 [inline]
  BUG: KASAN: use-after-free in atomic64_sub_return include/asm-generic/atomic-instrumented.h:970 [inline]
  BUG: KASAN: use-after-free in atomic_long_sub_return include/asm-generic/atomic-long.h:113 [inline]
  BUG: KASAN: use-after-free in page_counter_cancel mm/page_counter.c:54 [inline]
  BUG: KASAN: use-after-free in page_counter_uncharge+0x3d/0xc0 mm/page_counter.c:155
  Write of size 8 at addr ffff8880371c0148 by task syz-executor.0/9304

  CPU: 0 PID: 9304 Comm: syz-executor.0 Not tainted 5.8.0-syzkaller #0
  Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
  Call Trace:
    __dump_stack lib/dump_stack.c:77 [inline]
    dump_stack+0x1f0/0x31e lib/dump_stack.c:118
    print_address_description+0x66/0x620 mm/kasan/report.c:383
    __kasan_report mm/kasan/report.c:513 [inline]
    kasan_report+0x132/0x1d0 mm/kasan/report.c:530
    check_memory_region_inline mm/kasan/generic.c:183 [inline]
    check_memory_region+0x2b5/0x2f0 mm/kasan/generic.c:192
    instrument_atomic_write include/linux/instrumented.h:71 [inline]
    atomic64_sub_return include/asm-generic/atomic-instrumented.h:970 [inline]
    atomic_long_sub_return include/asm-generic/atomic-long.h:113 [inline]
    page_counter_cancel mm/page_counter.c:54 [inline]
    page_counter_uncharge+0x3d/0xc0 mm/page_counter.c:155
    uncharge_batch+0x6c/0x350 mm/memcontrol.c:6764
    uncharge_page+0x115/0x430 mm/memcontrol.c:6796
    uncharge_list mm/memcontrol.c:6835 [inline]
    mem_cgroup_uncharge_list+0x70/0xe0 mm/memcontrol.c:6877
    release_pages+0x13a2/0x1550 mm/swap.c:911
    tlb_batch_pages_flush mm/mmu_gather.c:49 [inline]
    tlb_flush_mmu_free mm/mmu_gather.c:242 [inline]
    tlb_flush_mmu+0x780/0x910 mm/mmu_gather.c:249
    tlb_finish_mmu+0xcb/0x200 mm/mmu_gather.c:328
    exit_mmap+0x296/0x550 mm/mmap.c:3185
    __mmput+0x113/0x370 kernel/fork.c:1076
    exit_mm+0x4cd/0x550 kernel/exit.c:483
    do_exit+0x576/0x1f20 kernel/exit.c:793
    do_group_exit+0x161/0x2d0 kernel/exit.c:903
    get_signal+0x139b/0x1d30 kernel/signal.c:2743
    arch_do_signal+0x33/0x610 arch/x86/kernel/signal.c:811
    exit_to_user_mode_loop kernel/entry/common.c:135 [inline]
    exit_to_user_mode_prepare+0x8d/0x1b0 kernel/entry/common.c:166
    syscall_exit_to_user_mode+0x5e/0x1a0 kernel/entry/common.c:241
    entry_SYSCALL_64_after_hwframe+0x44/0xa9

Commit 1a3e1f40 ("mm: memcontrol: decouple reference counting from
page accounting") reworked the memcg lifetime to be bound the the struct
page rather than charges.  It also removed the css_put_many from
uncharge_batch and that is causing the above splat.

uncharge_batch() is supposed to uncharge accumulated charges for all
pages freed from the same memcg.  The queuing is done by uncharge_page
which however drops the memcg reference after it adds charges to the
batch.  If the current page happens to be the last one holding the
reference for its memcg then the memcg is OK to go and the next page to
be freed will trigger batched uncharge which needs to access the memcg
which is gone already.

Fix the issue by taking a reference for the memcg in the current batch.

Fixes: 1a3e1f40 ("mm: memcontrol: decouple reference counting from page accounting")
Reported-by: syzbot+b305848212deec86eabe@syzkaller.appspotmail.com
Reported-by: syzbot+b5ea6fb6f139c8b9482b@syzkaller.appspotmail.com
Signed-off-by: NMichal Hocko <mhocko@suse.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Reviewed-by: NShakeel Butt <shakeelb@google.com>
Acked-by: NJohannes Weiner <hannes@cmpxchg.org>
Cc: Roman Gushchin <guro@fb.com>
Cc: Hugh Dickins <hughd@google.com>
Link: https://lkml.kernel.org/r/20200820090341.GC5033@dhcp22.suse.czSigned-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      f1796544
  7. 05 9月, 2020 4 次提交
  9. 04 9月, 2020 2 次提交
    • R
      memremap: rename MEMORY_DEVICE_DEVDAX to MEMORY_DEVICE_GENERIC · 4533d3ae
      由 Roger Pau Monne 提交于 
      This is in preparation for the logic behind MEMORY_DEVICE_DEVDAX also
being used by non DAX devices.

No functional change intended.
Signed-off-by: NRoger Pau Monné <roger.pau@citrix.com>
Reviewed-by: NIra Weiny <ira.weiny@intel.com>
Acked-by: NAndrew Morton <akpm@linux-foundation.org>
Reviewed-by: NPankaj Gupta <pankaj.gupta.linux@gmail.com>
Link: https://lore.kernel.org/r/20200901083326.21264-3-roger.pau@citrix.comSigned-off-by: NJuergen Gross <jgross@suse.com>
      4533d3ae
    • D
      mm: fix pin vs. gup mismatch with gate pages · 9fa2dd94
      由 Dave Hansen 提交于 
      Gate pages were missed when converting from get to pin_user_pages().
This can lead to refcount imbalances.  This is reliably and quickly
reproducible running the x86 selftests when vsyscall=emulate is enabled
(the default).  Fix by using try_grab_page() with appropriate flags
passed.

The long story:

Today, pin_user_pages() and get_user_pages() are similar interfaces for
manipulating page reference counts.  However, "pins" use a "bias" value
and manipulate the actual reference count by 1024 instead of 1 used by
plain "gets".

That means that pin_user_pages() must be matched with unpin_user_pages()
and can't be mixed with a plain put_user_pages() or put_page().

Enter gate pages, like the vsyscall page.  They are pages usually in the
kernel image, but which are mapped to userspace.  Userspace is allowed
access to them, including interfaces using get/pin_user_pages().  The
refcount of these kernel pages is manipulated just like a normal user
page on the get/pin side so that the put/unpin side can work the same
for normal user pages or gate pages.

get_gate_page() uses try_get_page() which only bumps the refcount by
1, not 1024, even if called in the pin_user_pages() path.  If someone
pins a gate page, this happens:

	pin_user_pages()
		get_gate_page()
			try_get_page() // bump refcount +1
	... some time later
	unpin_user_pages()
		page_ref_sub_and_test(page, 1024))

... and boom, we get a refcount off by 1023.  This is reliably and
quickly reproducible running the x86 selftests when booted with
vsyscall=emulate (the default).  The selftests use ptrace(), but I
suspect anything using pin_user_pages() on gate pages could hit this.

To fix it, simply use try_grab_page() instead of try_get_page(), and
pass 'gup_flags' in so that FOLL_PIN can be respected.

This bug traces back to the very beginning of the FOLL_PIN support in
commit 3faa52c0 ("mm/gup: track FOLL_PIN pages"), which showed up in
the 5.7 release.
Signed-off-by: NDave Hansen <dave.hansen@linux.intel.com>
Fixes: 3faa52c0 ("mm/gup: track FOLL_PIN pages")
Reported-by: NPeter Zijlstra <peterz@infradead.org>
Reviewed-by: NJohn Hubbard <jhubbard@nvidia.com>
Acked-by: NAndy Lutomirski <luto@kernel.org>
Cc: x86@kernel.org
Cc: Jann Horn <jannh@google.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      9fa2dd94
  11. 24 8月, 2020 1 次提交
  13. 22 8月, 2020 6 次提交
  15. 19 8月, 2020 1 次提交
  17. 15 8月, 2020 9 次提交
    • Q
      mm/swap.c: annotate data races for lru_rotate_pvecs · 7e0cc01e
      由 Qian Cai 提交于 
      Read to lru_add_pvec->nr could be interrupted and then write to the same
variable.  The write has local interrupt disabled, but the plain reads
result in data races.  However, it is unlikely the compilers could do much
damage here given that lru_add_pvec->nr is a "unsigned char" and there is
an existing compiler barrier.  Thus, annotate the reads using the
data_race() macro.  The data races were reported by KCSAN,

 BUG: KCSAN: data-race in lru_add_drain_cpu / rotate_reclaimable_page

 write to 0xffff9291ebcb8a40 of 1 bytes by interrupt on cpu 23:
  rotate_reclaimable_page+0x2df/0x490
  pagevec_add at include/linux/pagevec.h:81
  (inlined by) rotate_reclaimable_page at mm/swap.c:259
  end_page_writeback+0x1b5/0x2b0
  end_swap_bio_write+0x1d0/0x280
  bio_endio+0x297/0x560
  dec_pending+0x218/0x430 [dm_mod]
  clone_endio+0xe4/0x2c0 [dm_mod]
  bio_endio+0x297/0x560
  blk_update_request+0x201/0x920
  scsi_end_request+0x6b/0x4a0
  scsi_io_completion+0xb7/0x7e0
  scsi_finish_command+0x1ed/0x2a0
  scsi_softirq_done+0x1c9/0x1d0
  blk_done_softirq+0x181/0x1d0
  __do_softirq+0xd9/0x57c
  irq_exit+0xa2/0xc0
  do_IRQ+0x8b/0x190
  ret_from_intr+0x0/0x42
  delay_tsc+0x46/0x80
  __const_udelay+0x3c/0x40
  __udelay+0x10/0x20
  kcsan_setup_watchpoint+0x202/0x3a0
  __tsan_read1+0xc2/0x100
  lru_add_drain_cpu+0xb8/0x3f0
  lru_add_drain+0x25/0x40
  shrink_active_list+0xe1/0xc80
  shrink_lruvec+0x766/0xb70
  shrink_node+0x2d6/0xca0
  do_try_to_free_pages+0x1f7/0x9a0
  try_to_free_pages+0x252/0x5b0
  __alloc_pages_slowpath+0x458/0x1290
  __alloc_pages_nodemask+0x3bb/0x450
  alloc_pages_vma+0x8a/0x2c0
  do_anonymous_page+0x16e/0x6f0
  __handle_mm_fault+0xcd5/0xd40
  handle_mm_fault+0xfc/0x2f0
  do_page_fault+0x263/0x6f9
  page_fault+0x34/0x40

 read to 0xffff9291ebcb8a40 of 1 bytes by task 37761 on cpu 23:
  lru_add_drain_cpu+0xb8/0x3f0
  lru_add_drain_cpu at mm/swap.c:602
  lru_add_drain+0x25/0x40
  shrink_active_list+0xe1/0xc80
  shrink_lruvec+0x766/0xb70
  shrink_node+0x2d6/0xca0
  do_try_to_free_pages+0x1f7/0x9a0
  try_to_free_pages+0x252/0x5b0
  __alloc_pages_slowpath+0x458/0x1290
  __alloc_pages_nodemask+0x3bb/0x450
  alloc_pages_vma+0x8a/0x2c0
  do_anonymous_page+0x16e/0x6f0
  __handle_mm_fault+0xcd5/0xd40
  handle_mm_fault+0xfc/0x2f0
  do_page_fault+0x263/0x6f9
  page_fault+0x34/0x40

 2 locks held by oom02/37761:
  #0: ffff9281e5928808 (&mm->mmap_sem#2){++++}, at: do_page_fault
  #1: ffffffffb3ade380 (fs_reclaim){+.+.}, at: fs_reclaim_acquire.part
 irq event stamp: 1949217
 trace_hardirqs_on_thunk+0x1a/0x1c
 __do_softirq+0x2e7/0x57c
 __do_softirq+0x34c/0x57c
 irq_exit+0xa2/0xc0

 Reported by Kernel Concurrency Sanitizer on:
 CPU: 23 PID: 37761 Comm: oom02 Not tainted 5.6.0-rc3-next-20200226+ #6
 Hardware name: HP ProLiant BL660c Gen9, BIOS I38 10/17/2018
Signed-off-by: NQian Cai <cai@lca.pw>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Acked-by: NMarco Elver <elver@google.com>
Link: http://lkml.kernel.org/r/20200228044018.1263-1-cai@lca.pwSigned-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      7e0cc01e
    • Q
      mm/rmap: annotate a data race at tlb_flush_batched · 9c1177b6
      由 Qian Cai 提交于 
      mm->tlb_flush_batched could be accessed concurrently as noticed by
KCSAN,

 BUG: KCSAN: data-race in flush_tlb_batched_pending / try_to_unmap_one

 write to 0xffff93f754880bd0 of 1 bytes by task 822 on cpu 6:
  try_to_unmap_one+0x59a/0x1ab0
  set_tlb_ubc_flush_pending at mm/rmap.c:635
  (inlined by) try_to_unmap_one at mm/rmap.c:1538
  rmap_walk_anon+0x296/0x650
  rmap_walk+0xdf/0x100
  try_to_unmap+0x18a/0x2f0
  shrink_page_list+0xef6/0x2870
  shrink_inactive_list+0x316/0x880
  shrink_lruvec+0x8dc/0x1380
  shrink_node+0x317/0xd80
  balance_pgdat+0x652/0xd90
  kswapd+0x396/0x8d0
  kthread+0x1e0/0x200
  ret_from_fork+0x27/0x50

 read to 0xffff93f754880bd0 of 1 bytes by task 6364 on cpu 4:
  flush_tlb_batched_pending+0x29/0x90
  flush_tlb_batched_pending at mm/rmap.c:682
  change_p4d_range+0x5dd/0x1030
  change_pte_range at mm/mprotect.c:44
  (inlined by) change_pmd_range at mm/mprotect.c:212
  (inlined by) change_pud_range at mm/mprotect.c:240
  (inlined by) change_p4d_range at mm/mprotect.c:260
  change_protection+0x222/0x310
  change_prot_numa+0x3e/0x60
  task_numa_work+0x219/0x350
  task_work_run+0xed/0x140
  prepare_exit_to_usermode+0x2cc/0x2e0
  ret_from_intr+0x32/0x42

 Reported by Kernel Concurrency Sanitizer on:
 CPU: 4 PID: 6364 Comm: mtest01 Tainted: G        W    L 5.5.0-next-20200210+ #5
 Hardware name: HPE ProLiant DL385 Gen10/ProLiant DL385 Gen10, BIOS A40 07/10/2019

flush_tlb_batched_pending() is under PTL but the write is not, but
mm->tlb_flush_batched is only a bool type, so the value is unlikely to be
shattered.  Thus, mark it as an intentional data race by using the data
race macro.
Signed-off-by: NQian Cai <cai@lca.pw>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Cc: Marco Elver <elver@google.com>
Link: http://lkml.kernel.org/r/1581450783-8262-1-git-send-email-cai@lca.pwSigned-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      9c1177b6
    • Q
      mm/mempool: fix a data race in mempool_free() · abe1de42
      由 Qian Cai 提交于 
      mempool_t pool.curr_nr could be accessed concurrently as noticed by
KCSAN,

 BUG: KCSAN: data-race in mempool_free / remove_element

 write to 0xffffffffa937638c of 4 bytes by task 6359 on cpu 113:
  remove_element+0x4a/0x1c0
  remove_element at mm/mempool.c:132
  mempool_alloc+0x102/0x210
  (inlined by) mempool_alloc at mm/mempool.c:399
  bio_alloc_bioset+0x106/0x2c0
  get_swap_bio+0x49/0x230
  __swap_writepage+0x680/0xc30
  swap_writepage+0x9c/0xf0
  pageout+0x33e/0xae0
  shrink_page_list+0x1f57/0x2870
  shrink_inactive_list+0x316/0x880
  shrink_lruvec+0x8dc/0x1380
  shrink_node+0x317/0xd80
  do_try_to_free_pages+0x1f7/0xa10
  try_to_free_pages+0x26c/0x5e0
  __alloc_pages_slowpath+0x458/0x1290
  <snip>

 read to 0xffffffffa937638c of 4 bytes by interrupt on cpu 64:
  mempool_free+0x3e/0x150
  mempool_free at mm/mempool.c:492
  bio_free+0x192/0x280
  bio_put+0x91/0xd0
  end_swap_bio_write+0x1d8/0x280
  bio_endio+0x2c2/0x5b0
  dec_pending+0x22b/0x440 [dm_mod]
  clone_endio+0xe4/0x2c0 [dm_mod]
  bio_endio+0x2c2/0x5b0
  blk_update_request+0x217/0x940
  scsi_end_request+0x6b/0x4d0
  scsi_io_completion+0xb7/0x7e0
  scsi_finish_command+0x223/0x310
  scsi_softirq_done+0x1d5/0x210
  blk_mq_complete_request+0x224/0x250
  scsi_mq_done+0xc2/0x250
  pqi_raid_io_complete+0x5a/0x70 [smartpqi]
  pqi_irq_handler+0x150/0x1410 [smartpqi]
  __handle_irq_event_percpu+0x90/0x540
  handle_irq_event_percpu+0x49/0xd0
  handle_irq_event+0x85/0xca
  handle_edge_irq+0x13f/0x3e0
  do_IRQ+0x86/0x190
  <snip>

Since the write is under pool->lock but the read is done as lockless.
Even though the commit 5b990546 ("mempool: fix and document
synchronization and memory barrier usage") introduced the smp_wmb() and
smp_rmb() pair to improve the situation, it is adequate to protect it
from data races which could lead to a logic bug, so fix it by adding
READ_ONCE() for the read.
Signed-off-by: NQian Cai <cai@lca.pw>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Cc: Marco Elver <elver@google.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Link: http://lkml.kernel.org/r/1581446384-2131-1-git-send-email-cai@lca.pwSigned-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      abe1de42
    • Q
      mm/list_lru: fix a data race in list_lru_count_one · a1f45935
      由 Qian Cai 提交于 
      struct list_lru_one l.nr_items could be accessed concurrently as noticed
by KCSAN,

 BUG: KCSAN: data-race in list_lru_count_one / list_lru_isolate_move

 write to 0xffffa102789c4510 of 8 bytes by task 823 on cpu 39:
  list_lru_isolate_move+0xf9/0x130
  list_lru_isolate_move at mm/list_lru.c:180
  inode_lru_isolate+0x12b/0x2a0
  __list_lru_walk_one+0x122/0x3d0
  list_lru_walk_one+0x75/0xa0
  prune_icache_sb+0x8b/0xc0
  super_cache_scan+0x1b8/0x250
  do_shrink_slab+0x256/0x6d0
  shrink_slab+0x41b/0x4a0
  shrink_node+0x35c/0xd80
  balance_pgdat+0x652/0xd90
  kswapd+0x396/0x8d0
  kthread+0x1e0/0x200
  ret_from_fork+0x27/0x50

 read to 0xffffa102789c4510 of 8 bytes by task 6345 on cpu 56:
  list_lru_count_one+0x116/0x2f0
  list_lru_count_one at mm/list_lru.c:193
  super_cache_count+0xe8/0x170
  do_shrink_slab+0x95/0x6d0
  shrink_slab+0x41b/0x4a0
  shrink_node+0x35c/0xd80
  do_try_to_free_pages+0x1f7/0xa10
  try_to_free_pages+0x26c/0x5e0
  __alloc_pages_slowpath+0x458/0x1290
  __alloc_pages_nodemask+0x3bb/0x450
  alloc_pages_vma+0x8a/0x2c0
  do_anonymous_page+0x170/0x700
  __handle_mm_fault+0xc9f/0xd00
  handle_mm_fault+0xfc/0x2f0
  do_page_fault+0x263/0x6f9
  page_fault+0x34/0x40

 Reported by Kernel Concurrency Sanitizer on:
 CPU: 56 PID: 6345 Comm: oom01 Tainted: G        W    L 5.5.0-next-20200205+ #4
 Hardware name: HPE ProLiant DL385 Gen10/ProLiant DL385 Gen10, BIOS A40 07/10/2019

A shattered l.nr_items could affect the shrinker behaviour due to a data
race. Fix it by adding READ_ONCE() for the read. Since the writes are
aligned and up to word-size, assume those are safe from data races to
avoid readability issues of writing WRITE_ONCE(var, var + val).
Signed-off-by: NQian Cai <cai@lca.pw>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Cc: Marco Elver <elver@google.com>
Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Link: http://lkml.kernel.org/r/1581114679-5488-1-git-send-email-cai@lca.pwSigned-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      a1f45935
    • Q
      mm/page_counter: fix various data races at memsw · 6e4bd50f
      由 Qian Cai 提交于 
      Commit 3e32cb2e ("mm: memcontrol: lockless page counters") could had
memcg->memsw->watermark and memcg->memsw->failcnt been accessed
concurrently as reported by KCSAN,

 BUG: KCSAN: data-race in page_counter_try_charge / page_counter_try_charge

 read to 0xffff8fb18c4cd190 of 8 bytes by task 1081 on cpu 59:
  page_counter_try_charge+0x4d/0x150 mm/page_counter.c:138
  try_charge+0x131/0xd50 mm/memcontrol.c:2405
  __memcg_kmem_charge_memcg+0x58/0x140
  __memcg_kmem_charge+0xcc/0x280
  __alloc_pages_nodemask+0x1e1/0x450
  alloc_pages_current+0xa6/0x120
  pte_alloc_one+0x17/0xd0
  __pte_alloc+0x3a/0x1f0
  copy_p4d_range+0xc36/0x1990
  copy_page_range+0x21d/0x360
  dup_mmap+0x5f5/0x7a0
  dup_mm+0xa2/0x240
  copy_process+0x1b3f/0x3460
  _do_fork+0xaa/0xa20
  __x64_sys_clone+0x13b/0x170
  do_syscall_64+0x91/0xb47
  entry_SYSCALL_64_after_hwframe+0x49/0xbe

 write to 0xffff8fb18c4cd190 of 8 bytes by task 1153 on cpu 120:
  page_counter_try_charge+0x5b/0x150 mm/page_counter.c:139
  try_charge+0x131/0xd50 mm/memcontrol.c:2405
  mem_cgroup_try_charge+0x159/0x460
  mem_cgroup_try_charge_delay+0x3d/0xa0
  wp_page_copy+0x14d/0x930
  do_wp_page+0x107/0x7b0
  __handle_mm_fault+0xce6/0xd40
  handle_mm_fault+0xfc/0x2f0
  do_page_fault+0x263/0x6f9
  page_fault+0x34/0x40

 BUG: KCSAN: data-race in page_counter_try_charge / page_counter_try_charge

 write to 0xffff88809bbf2158 of 8 bytes by task 11782 on cpu 0:
  page_counter_try_charge+0x100/0x170 mm/page_counter.c:129
  try_charge+0x185/0xbf0 mm/memcontrol.c:2405
  __memcg_kmem_charge_memcg+0x4a/0xe0 mm/memcontrol.c:2837
  __memcg_kmem_charge+0xcf/0x1b0 mm/memcontrol.c:2877
  __alloc_pages_nodemask+0x26c/0x310 mm/page_alloc.c:4780

 read to 0xffff88809bbf2158 of 8 bytes by task 11814 on cpu 1:
  page_counter_try_charge+0xef/0x170 mm/page_counter.c:129
  try_charge+0x185/0xbf0 mm/memcontrol.c:2405
  __memcg_kmem_charge_memcg+0x4a/0xe0 mm/memcontrol.c:2837
  __memcg_kmem_charge+0xcf/0x1b0 mm/memcontrol.c:2877
  __alloc_pages_nodemask+0x26c/0x310 mm/page_alloc.c:4780

Since watermark could be compared or set to garbage due to a data race
which would change the code logic, fix it by adding a pair of READ_ONCE()
and WRITE_ONCE() in those places.

The "failcnt" counter is tolerant of some degree of inaccuracy and is only
used to report stats, a data race will not be harmful, thus mark it as an
intentional data race using the data_race() macro.

Fixes: 3e32cb2e ("mm: memcontrol: lockless page counters")
Reported-by: syzbot+f36cfe60b1006a94f9dc@syzkaller.appspotmail.com
Signed-off-by: NQian Cai <cai@lca.pw>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Acked-by: NMichal Hocko <mhocko@suse.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Cc: Marco Elver <elver@google.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Link: http://lkml.kernel.org/r/1581519682-23594-1-git-send-email-cai@lca.pwSigned-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      6e4bd50f
    • Q
      mm/swapfile: fix and annotate various data races · a449bf58
      由 Qian Cai 提交于 
      swap_info_struct si.highest_bit, si.swap_map[offset] and si.flags could
be accessed concurrently separately as noticed by KCSAN,

=== si.highest_bit ===

 write to 0xffff8d5abccdc4d4 of 4 bytes by task 5353 on cpu 24:
  swap_range_alloc+0x81/0x130
  swap_range_alloc at mm/swapfile.c:681
  scan_swap_map_slots+0x371/0xb90
  get_swap_pages+0x39d/0x5c0
  get_swap_page+0xf2/0x524
  add_to_swap+0xe4/0x1c0
  shrink_page_list+0x1795/0x2870
  shrink_inactive_list+0x316/0x880
  shrink_lruvec+0x8dc/0x1380
  shrink_node+0x317/0xd80
  do_try_to_free_pages+0x1f7/0xa10
  try_to_free_pages+0x26c/0x5e0
  __alloc_pages_slowpath+0x458/0x1290

 read to 0xffff8d5abccdc4d4 of 4 bytes by task 6672 on cpu 70:
  scan_swap_map_slots+0x4a6/0xb90
  scan_swap_map_slots at mm/swapfile.c:892
  get_swap_pages+0x39d/0x5c0
  get_swap_page+0xf2/0x524
  add_to_swap+0xe4/0x1c0
  shrink_page_list+0x1795/0x2870
  shrink_inactive_list+0x316/0x880
  shrink_lruvec+0x8dc/0x1380
  shrink_node+0x317/0xd80
  do_try_to_free_pages+0x1f7/0xa10
  try_to_free_pages+0x26c/0x5e0
  __alloc_pages_slowpath+0x458/0x1290

 Reported by Kernel Concurrency Sanitizer on:
 CPU: 70 PID: 6672 Comm: oom01 Tainted: G        W    L 5.5.0-next-20200205+ #3
 Hardware name: HPE ProLiant DL385 Gen10/ProLiant DL385 Gen10, BIOS A40 07/10/2019

=== si.swap_map[offset] ===

 write to 0xffffbc370c29a64c of 1 bytes by task 6856 on cpu 86:
  __swap_entry_free_locked+0x8c/0x100
  __swap_entry_free_locked at mm/swapfile.c:1209 (discriminator 4)
  __swap_entry_free.constprop.20+0x69/0xb0
  free_swap_and_cache+0x53/0xa0
  unmap_page_range+0x7f8/0x1d70
  unmap_single_vma+0xcd/0x170
  unmap_vmas+0x18b/0x220
  exit_mmap+0xee/0x220
  mmput+0x10e/0x270
  do_exit+0x59b/0xf40
  do_group_exit+0x8b/0x180

 read to 0xffffbc370c29a64c of 1 bytes by task 6855 on cpu 20:
  _swap_info_get+0x81/0xa0
  _swap_info_get at mm/swapfile.c:1140
  free_swap_and_cache+0x40/0xa0
  unmap_page_range+0x7f8/0x1d70
  unmap_single_vma+0xcd/0x170
  unmap_vmas+0x18b/0x220
  exit_mmap+0xee/0x220
  mmput+0x10e/0x270
  do_exit+0x59b/0xf40
  do_group_exit+0x8b/0x180

=== si.flags ===

 write to 0xffff956c8fc6c400 of 8 bytes by task 6087 on cpu 23:
  scan_swap_map_slots+0x6fe/0xb50
  scan_swap_map_slots at mm/swapfile.c:887
  get_swap_pages+0x39d/0x5c0
  get_swap_page+0x377/0x524
  add_to_swap+0xe4/0x1c0
  shrink_page_list+0x1795/0x2870
  shrink_inactive_list+0x316/0x880
  shrink_lruvec+0x8dc/0x1380
  shrink_node+0x317/0xd80
  do_try_to_free_pages+0x1f7/0xa10
  try_to_free_pages+0x26c/0x5e0
  __alloc_pages_slowpath+0x458/0x1290

 read to 0xffff956c8fc6c400 of 8 bytes by task 6207 on cpu 63:
  _swap_info_get+0x41/0xa0
  __swap_info_get at mm/swapfile.c:1114
  put_swap_page+0x84/0x490
  __remove_mapping+0x384/0x5f0
  shrink_page_list+0xff1/0x2870
  shrink_inactive_list+0x316/0x880
  shrink_lruvec+0x8dc/0x1380
  shrink_node+0x317/0xd80
  do_try_to_free_pages+0x1f7/0xa10
  try_to_free_pages+0x26c/0x5e0
  __alloc_pages_slowpath+0x458/0x1290

The writes are under si->lock but the reads are not. For si.highest_bit
and si.swap_map[offset], data race could trigger logic bugs, so fix them
by having WRITE_ONCE() for the writes and READ_ONCE() for the reads
except those isolated reads where they compare against zero which a data
race would cause no harm. Thus, annotate them as intentional data races
using the data_race() macro.

For si.flags, the readers are only interested in a single bit where a
data race there would cause no issue there.

[cai@lca.pw: add a missing annotation for si->flags in memory.c]
  Link: http://lkml.kernel.org/r/1581612647-5958-1-git-send-email-cai@lca.pwSigned-off-by: NQian Cai <cai@lca.pw>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Cc: Marco Elver <elver@google.com>
Cc: Hugh Dickins <hughd@google.com>
Link: http://lkml.kernel.org/r/1581095163-12198-1-git-send-email-cai@lca.pwSigned-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      a449bf58
    • K
      mm/filemap.c: fix a data race in filemap_fault() · e630bfac
      由 Kirill A. Shutemov 提交于 
      struct file_ra_state ra.mmap_miss could be accessed concurrently during
page faults as noticed by KCSAN,

 BUG: KCSAN: data-race in filemap_fault / filemap_map_pages

 write to 0xffff9b1700a2c1b4 of 4 bytes by task 3292 on cpu 30:
  filemap_fault+0x920/0xfc0
  do_sync_mmap_readahead at mm/filemap.c:2384
  (inlined by) filemap_fault at mm/filemap.c:2486
  __xfs_filemap_fault+0x112/0x3e0 [xfs]
  xfs_filemap_fault+0x74/0x90 [xfs]
  __do_fault+0x9e/0x220
  do_fault+0x4a0/0x920
  __handle_mm_fault+0xc69/0xd00
  handle_mm_fault+0xfc/0x2f0
  do_page_fault+0x263/0x6f9
  page_fault+0x34/0x40

 read to 0xffff9b1700a2c1b4 of 4 bytes by task 3313 on cpu 32:
  filemap_map_pages+0xc2e/0xd80
  filemap_map_pages at mm/filemap.c:2625
  do_fault+0x3da/0x920
  __handle_mm_fault+0xc69/0xd00
  handle_mm_fault+0xfc/0x2f0
  do_page_fault+0x263/0x6f9
  page_fault+0x34/0x40

 Reported by Kernel Concurrency Sanitizer on:
 CPU: 32 PID: 3313 Comm: systemd-udevd Tainted: G        W    L 5.5.0-next-20200210+ #1
 Hardware name: HPE ProLiant DL385 Gen10/ProLiant DL385 Gen10, BIOS A40 07/10/2019

ra.mmap_miss is used to contribute the readahead decisions, a data race
could be undesirable.  Both the read and write is only under non-exclusive
mmap_sem, two concurrent writers could even underflow the counter.  Fix
the underflow by writing to a local variable before committing a final
store to ra.mmap_miss given a small inaccuracy of the counter should be
acceptable.
Signed-off-by: NKirill A. Shutemov <kirill@shutemov.name>
Signed-off-by: NQian Cai <cai@lca.pw>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Tested-by: NQian Cai <cai@lca.pw>
Reviewed-by: NMatthew Wilcox (Oracle) <willy@infradead.org>
Cc: Marco Elver <elver@google.com>
Link: http://lkml.kernel.org/r/20200211030134.1847-1-cai@lca.pwSigned-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      e630bfac
    • Q
      mm/swap_state: mark various intentional data races · b96a3db2
      由 Qian Cai 提交于 
      swap_cache_info.* could be accessed concurrently as noticed by
KCSAN,

 BUG: KCSAN: data-race in lookup_swap_cache / lookup_swap_cache

 write to 0xffffffff85517318 of 8 bytes by task 94138 on cpu 101:
  lookup_swap_cache+0x12e/0x460
  lookup_swap_cache at mm/swap_state.c:322
  do_swap_page+0x112/0xeb0
  __handle_mm_fault+0xc7a/0xd00
  handle_mm_fault+0xfc/0x2f0
  do_page_fault+0x263/0x6f9
  page_fault+0x34/0x40

 read to 0xffffffff85517318 of 8 bytes by task 91655 on cpu 100:
  lookup_swap_cache+0x117/0x460
  lookup_swap_cache at mm/swap_state.c:322
  shmem_swapin_page+0xc7/0x9e0
  shmem_getpage_gfp+0x2ca/0x16c0
  shmem_fault+0xef/0x3c0
  __do_fault+0x9e/0x220
  do_fault+0x4a0/0x920
  __handle_mm_fault+0xc69/0xd00
  handle_mm_fault+0xfc/0x2f0
  do_page_fault+0x263/0x6f9
  page_fault+0x34/0x40

 Reported by Kernel Concurrency Sanitizer on:
 CPU: 100 PID: 91655 Comm: systemd-journal Tainted: G        W  O L 5.5.0-next-20200204+ #6
 Hardware name: HPE ProLiant DL385 Gen10/ProLiant DL385 Gen10, BIOS A40 07/10/2019

 write to 0xffffffff8d717308 of 8 bytes by task 11365 on cpu 87:
   __delete_from_swap_cache+0x681/0x8b0
   __delete_from_swap_cache at mm/swap_state.c:178

 read to 0xffffffff8d717308 of 8 bytes by task 11275 on cpu 53:
   __delete_from_swap_cache+0x66e/0x8b0
   __delete_from_swap_cache at mm/swap_state.c:178

Both the read and write are done as lockless. Since swap_cache_info.*
are only used to print out counter information, even if any of them
missed a few incremental due to data races, it will be harmless, so just
mark it as an intentional data race using the data_race() macro.

While at it, fix a checkpatch.pl warning,

WARNING: Single statement macros should not use a do {} while (0) loop
Signed-off-by: NQian Cai <cai@lca.pw>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Cc: Marco Elver <elver@google.com>
Link: http://lkml.kernel.org/r/20200207003715.1578-1-cai@lca.pwSigned-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      b96a3db2
    • Q
      mm/page_io: mark various intentional data races · 7b37e226
      由 Qian Cai 提交于 
      struct swap_info_struct si.flags could be accessed concurrently as noticed
by KCSAN,

 BUG: KCSAN: data-race in scan_swap_map_slots / swap_readpage

 write to 0xffff9c77b80ac400 of 8 bytes by task 91325 on cpu 16:
  scan_swap_map_slots+0x6fe/0xb50
  scan_swap_map_slots at mm/swapfile.c:887
  get_swap_pages+0x39d/0x5c0
  get_swap_page+0x377/0x524
  add_to_swap+0xe4/0x1c0
  shrink_page_list+0x1740/0x2820
  shrink_inactive_list+0x316/0x8b0
  shrink_lruvec+0x8dc/0x1380
  shrink_node+0x317/0xd80
  do_try_to_free_pages+0x1f7/0xa10
  try_to_free_pages+0x26c/0x5e0
  __alloc_pages_slowpath+0x458/0x1290
  __alloc_pages_nodemask+0x3bb/0x450
  alloc_pages_vma+0x8a/0x2c0
  do_anonymous_page+0x170/0x700
  __handle_mm_fault+0xc9f/0xd00
  handle_mm_fault+0xfc/0x2f0
  do_page_fault+0x263/0x6f9
  page_fault+0x34/0x40

 read to 0xffff9c77b80ac400 of 8 bytes by task 5422 on cpu 7:
  swap_readpage+0x204/0x6a0
  swap_readpage at mm/page_io.c:380
  read_swap_cache_async+0xa2/0xb0
  swapin_readahead+0x6a0/0x890
  do_swap_page+0x465/0xeb0
  __handle_mm_fault+0xc7a/0xd00
  handle_mm_fault+0xfc/0x2f0
  do_page_fault+0x263/0x6f9
  page_fault+0x34/0x40

 Reported by Kernel Concurrency Sanitizer on:
 CPU: 7 PID: 5422 Comm: gmain Tainted: G        W  O L 5.5.0-next-20200204+ #6
 Hardware name: HPE ProLiant DL385 Gen10/ProLiant DL385 Gen10, BIOS A40 07/10/2019

Other reads,

 read to 0xffff91ea33eac400 of 8 bytes by task 11276 on cpu 120:
  __swap_writepage+0x140/0xc20
  __swap_writepage at mm/page_io.c:289

 read to 0xffff91ea33eac400 of 8 bytes by task 11264 on cpu 16:
  swap_set_page_dirty+0x44/0x1f4
  swap_set_page_dirty at mm/page_io.c:442

The write is under &si->lock, but the reads are done as lockless.  Since
the reads only check for a specific bit in the flag, it is harmless even
if load tearing happens.  Thus, just mark them as intentional data races
using the data_race() macro.

[cai@lca.pw: add a missing annotation]
  Link: http://lkml.kernel.org/r/1581612585-5812-1-git-send-email-cai@lca.pwSigned-off-by: NQian Cai <cai@lca.pw>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Cc: Marco Elver <elver@google.com>
Link: http://lkml.kernel.org/r/20200207003601.1526-1-cai@lca.pwSigned-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      7b37e226