android_kernel_motorola_sm6375/mm
Repository files (latest commit first)
Filename Latest commit message Latest commit date
Aneesh Kumar K.V 482efd771f UPSTREAM: mm/mremap: hold the rmap lock in write mode when moving page table entries.
commit 97113eb39fa7972722ff490b947d8af023e1f6a2 upstream.

To avoid a race between rmap walk and mremap, mremap does
take_rmap_locks().  The lock was taken to ensure that rmap walk don't miss
a page table entry due to PTE moves via move_pagetables().  The kernel
does further optimization of this lock such that if we are going to find
the newly added vma after the old vma, the rmap lock is not taken.  This
is because rmap walk would find the vmas in the same order and if we don't
find the page table attached to older vma we would find it with the new
vma which we would iterate later.

As explained in commit eb66ae0308 ("mremap: properly flush TLB before
releasing the page") mremap is special in that it doesn't take ownership
of the page.  The optimized version for PUD/PMD aligned mremap also
doesn't hold the ptl lock.  This can result in stale TLB entries as show
below.

This patch updates the rmap locking requirement in mremap to handle the race condition
explained below with optimized mremap::

Optmized PMD move

    CPU 1                           CPU 2                                   CPU 3

    mremap(old_addr, new_addr)      page_shrinker/try_to_unmap_one

    mmap_write_lock_killable()

                                    addr = old_addr
                                    lock(pte_ptl)
    lock(pmd_ptl)
    pmd = *old_pmd
    pmd_clear(old_pmd)
    flush_tlb_range(old_addr)

    *new_pmd = pmd
                                                                            *new_addr = 10; and fills
                                                                            TLB with new addr
                                                                            and old pfn

    unlock(pmd_ptl)
                                    ptep_clear_flush()
                                    old pfn is free.
                                                                            Stale TLB entry

Optimized PUD move also suffers from a similar race.  Both the above race
condition can be fixed if we force mremap path to take rmap lock.

Bug: 248354871
Link: https://lkml.kernel.org/r/20210616045239.370802-7-aneesh.kumar@linux.ibm.com
Fixes: 2c91bd4a4e ("mm: speed up mremap by 20x on large regions")
Fixes: c49dd3401802 ("mm: speedup mremap on 1GB or larger regions")
Link: https://lore.kernel.org/linux-mm/CAHk-=wgXVR04eBNtxQfevontWnP6FDm+oj5vauQXP3S-huwbPw@mail.gmail.com
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Acked-by: Hugh Dickins <hughd@google.com>
Acked-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Christophe Leroy <christophe.leroy@csgroup.eu>
Cc: Joel Fernandes <joel@joelfernandes.org>
Cc: Kalesh Singh <kaleshsingh@google.com>
Cc: Kirill A. Shutemov <kirill@shutemov.name>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Nicholas Piggin <npiggin@gmail.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
[patch rewritten for backport since the code was refactored since]
Signed-off-by: Jann Horn <jannh@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Lee Jones <joneslee@google.com>
Change-Id: I046ea082f78b4c35c364dca305953a3f9ed28f1d
2022-11-07 11:17:25 +00:00
..
kasan kasan: fix incorrect arguments passing in kasan_add_zero_shadow 2021-01-27 11:47:53 +01:00
backing-dev.c mm: bdi: initialize bdi_min_ratio when bdi is unregistered 2021-12-14 14:49:00 +01:00
balloon_compaction.c
cleancache.c
cma.c UPSTREAM: mm/cma.c: remove redundant cma_mutex lock 2021-05-06 09:34:26 +00:00
cma.h
cma_debug.c
compaction.c mm, compaction: fast_find_migrateblock() should return pfn in the target zone 2022-06-14 18:11:46 +02:00
debug.c mm/debug.c: always print flags in dump_page() 2020-03-05 16:43:51 +01:00
debug_page_ref.c
dmapool.c
early_ioremap.c
fadvise.c
failslab.c
filemap.c mm/filemap: fix storing to a THP shadow entry 2021-06-10 13:37:15 +02:00
frame_vector.c
frontswap.c
gup.c mm/gup: fix gup_fast with dynamic page table folding 2020-10-01 13:18:24 +02:00
gup_benchmark.c mm/gup: fix memory leak in __gup_benchmark_ioctl 2020-01-09 10:20:00 +01:00
highmem.c
hmm.c
huge_memory.c mm/huge_memory.c: don't discard hugepage if other processes are mapping it 2021-07-14 16:53:47 +02:00
hugetlb.c hugetlb: fix huge_pmd_unshare address update 2022-06-14 18:11:48 +02:00
hugetlb_cgroup.c
hwpoison-inject.c
init-mm.c
internal.h mm/thp: fix vma_address() if virtual address below file offset 2021-06-30 08:47:52 -04:00
interval_tree.c
Kconfig mm/zsmalloc.c: drop ZSMALLOC_PGTABLE_MAPPING 2020-12-16 10:56:59 +01:00
Kconfig.debug
khugepaged.c khugepaged: fix wrong result value for trace_mm_collapse_huge_page_isolate() 2021-05-19 10:08:27 +02:00
kmemleak-test.c
kmemleak.c mm: kmemleak: take a full lowmem check in kmemleak_*_phys() 2022-04-20 09:19:38 +02:00
ksm.c UPSTREAM: mm/ksm: Remove reuse_ksm_page() 2022-07-06 18:36:53 -07:00
list_lru.c mm: list_lru: set shrinker map bit when child nr_items is not zero 2020-12-11 13:23:31 +01:00
maccess.c uaccess: Add non-pagefault user-space write function 2020-01-17 19:48:40 +01:00
madvise.c This is the 5.4.69 stable release 2020-10-01 18:00:35 +02:00
Makefile
memblock.c This is the 5.4.182 stable release 2022-03-02 15:29:51 +01:00
memcontrol.c mm/memcontrol: return 1 from cgroup.memory __setup() handler 2022-04-15 14:18:29 +02:00
memfd.c memfd: fix F_SEAL_WRITE after shmem huge page allocated 2022-03-08 19:07:49 +01:00
memory-failure.c mm/memory-failure: make sure wait for page writeback in memory_failure 2021-06-23 14:41:23 +02:00
memory.c Merge tag 'android11-5.4.197_r00' into 'android11-5.4' 2022-07-27 11:19:48 +02:00
memory_hotplug.c mm/memory_hotplug: use "unsigned long" for PFN in zone_for_pfn_range() 2021-09-22 12:26:43 +02:00
mempolicy.c This is the 5.4.208 stable release 2022-07-30 14:33:25 +02:00
mempool.c
memremap.c mm/memory_hotplug: shrink zones when offlining memory 2020-01-09 10:19:56 +01:00
memtest.c
migrate.c mm: fix missing cache flush for all tail pages of compound page 2022-05-15 19:54:47 +02:00
mincore.c
mlock.c
mm_init.c
mmap.c UPSTREAM: mm: Fix TLB flush for not-first PFNMAP mappings in unmap_region() 2022-11-03 08:06:02 +00:00
mmu_context.c mm: fix kthread_use_mm() vs TLB invalidate 2020-09-03 11:26:51 +02:00
mmu_gather.c mm/mmu_gather: invalidate TLB correctly on batch allocation failure and flush 2020-02-11 04:35:42 -08:00
mmu_notifier.c
mmzone.c arm: remove CONFIG_ARCH_HAS_HOLES_MEMORYMODEL 2022-05-15 19:54:46 +02:00
mprotect.c Revert "BACKPORT: FROMGIT: mm: improve mprotect(R|W) efficiency on pages referenced once" 2021-01-18 16:25:18 +00:00
mremap.c UPSTREAM: mm/mremap: hold the rmap lock in write mode when moving page table entries. 2022-11-07 11:17:25 +00:00
msync.c
nommu.c x86/mm: split vmalloc_sync_all() 2020-03-25 08:25:58 +01:00
oom_kill.c Revert "oom_kill.c: futex: delay the OOM reaper to allow time for proper futex cleanup" 2022-04-27 15:56:55 +02:00
OWNERS ANDROID: Add OWNERS files referring to the respective android-mainline OWNERS 2021-04-01 13:45:14 +00:00
page-writeback.c mm/page-writeback.c: avoid potential division by zero in wb_min_max_ratio() 2020-01-23 08:22:41 +01:00
page_alloc.c This is the 5.4.191 stable release 2022-04-27 14:24:26 +02:00
page_counter.c Revert "ANDROID: Revert: Merge 5.4.60 into android11-5.4" 2020-08-23 13:12:51 +02:00
page_ext.c
page_idle.c
page_io.c This is the 5.4.193 stable release 2022-05-16 08:51:00 +02:00
page_isolation.c mm/memory_hotplug: drain per-cpu pages again during memory offline 2020-09-23 12:40:47 +02:00
page_owner.c mm/page_owner: change split_page_owner to take a count 2020-10-29 09:57:52 +01:00
page_poison.c
page_vma_mapped.c mm/thp: another PVMW_SYNC fix in page_vma_mapped_walk() 2021-06-30 08:47:55 -04:00
pagewalk.c mm: pagewalk: fix termination condition in walk_pte_range() 2020-10-01 13:17:30 +02:00
percpu-internal.h
percpu-km.c
percpu-stats.c
percpu-vm.c
percpu.c percpu: fix first chunk size calculation for populated bitmap 2020-09-23 12:40:45 +02:00
pgtable-generic.c mm/thp: fix __split_huge_pmd_locked() on shmem migration entry 2021-06-30 08:47:52 -04:00
process_vm_access.c
readahead.c
rmap.c mm: fix race between MADV_FREE reclaim and blkdev direct IO read 2022-04-15 14:18:36 +02:00
rodata_test.c
shmem.c This is the 5.4.174 stable release 2022-01-31 15:14:13 +01:00
shuffle.c mm/shuffle: don't move pages between zones and don't read garbage memmaps 2020-09-03 11:26:51 +02:00
shuffle.h
slab.c mm, debug_pagealloc: don't rely on static keys too early 2020-01-23 08:22:40 +01:00
slab.h mm: kmemleak: slob: respect SLAB_NOLEAKTRACE flag 2021-11-26 10:47:21 +01:00
slab_common.c mm: slab: fix kmem_cache_create failed when sysfs node not destroyed 2021-07-25 14:35:14 +02:00
slob.c UPSTREAM: mm/sl[uo]b: export __kmalloc_track(_node)_caller 2020-11-02 16:12:00 +00:00
slub.c This is the 5.4.205 stable release 2022-07-12 21:08:00 +02:00
sparse-vmemmap.c
sparse.c mm/sparse: add the missing sparse_buffer_fini() in error branch 2021-05-14 09:44:32 +02:00
swap.c
swap_cgroup.c
swap_slots.c
swap_state.c mm/swap_state: fix a data race in swapin_nr_pages 2020-10-01 13:18:08 +02:00
swapfile.c This is the 5.4.103 stable release 2021-03-07 12:51:29 +01:00
truncate.c mm/thp: unmap_mapping_page() to fix THP truncate_cleanup_page() 2021-06-30 08:47:53 -04:00
usercopy.c mm/usercopy: return 1 from hardened_usercopy __setup() handler 2022-04-15 14:18:30 +02:00
userfaultfd.c mm: userfaultfd: fix missing cache flush in mcopy_atomic_pte() and __mcopy_atomic() 2022-05-15 19:54:47 +02:00
util.c random: move randomize_page() into mm where it belongs 2022-06-22 14:11:17 +02:00
vmacache.c
vmalloc.c mm/vunmap: add cond_resched() in vunmap_pmd_range 2020-09-03 11:26:52 +02:00
vmpressure.c
vmscan.c mm,vmscan: fix divide by zero in get_scan_count 2021-09-22 12:26:37 +02:00
vmstat.c Merge 5.4.194 into android11-5.4-lts 2022-05-18 16:49:42 +02:00
workingset.c
z3fold.c mm/z3fold: fix potential memory leak in z3fold_destroy_pool() 2021-07-14 16:53:47 +02:00
zbud.c
zpool.c
zsmalloc.c zsmalloc: fix races between asynchronous zspage free and page migration 2022-06-06 08:33:50 +02:00
zswap.c