WSL2-Linux-Kernel/mm
Ilya Lipnitskiy e720e7d0e9 mm: fix race by making init_zero_pfn() early_initcall
There are code paths that rely on zero_pfn to be fully initialized
before core_initcall.  For example, wq_sysfs_init() is a core_initcall
function that eventually results in a call to kernel_execve, which
causes a page fault with a subsequent mmput.  If zero_pfn is not
initialized by then it may not get cleaned up properly and result in an
error:

  BUG: Bad rss-counter state mm:(ptrval) type:MM_ANONPAGES val:1

Here is an analysis of the race as seen on a MIPS device. On this
particular MT7621 device (Ubiquiti ER-X), zero_pfn is PFN 0 until
initialized, at which point it becomes PFN 5120:

  1. wq_sysfs_init calls into kobject_uevent_env at core_initcall:
       kobject_uevent_env+0x7e4/0x7ec
       kset_register+0x68/0x88
       bus_register+0xdc/0x34c
       subsys_virtual_register+0x34/0x78
       wq_sysfs_init+0x1c/0x4c
       do_one_initcall+0x50/0x1a8
       kernel_init_freeable+0x230/0x2c8
       kernel_init+0x10/0x100
       ret_from_kernel_thread+0x14/0x1c

  2. kobject_uevent_env() calls call_usermodehelper_exec() which executes
     kernel_execve asynchronously.

  3. Memory allocations in kernel_execve cause a page fault, bumping the
     MM reference counter:
       add_mm_counter_fast+0xb4/0xc0
       handle_mm_fault+0x6e4/0xea0
       __get_user_pages.part.78+0x190/0x37c
       __get_user_pages_remote+0x128/0x360
       get_arg_page+0x34/0xa0
       copy_string_kernel+0x194/0x2a4
       kernel_execve+0x11c/0x298
       call_usermodehelper_exec_async+0x114/0x194

  4. In case zero_pfn has not been initialized yet, zap_pte_range does
     not decrement the MM_ANONPAGES RSS counter and the BUG message is
     triggered shortly afterwards when __mmdrop checks the ref counters:
       __mmdrop+0x98/0x1d0
       free_bprm+0x44/0x118
       kernel_execve+0x160/0x1d8
       call_usermodehelper_exec_async+0x114/0x194
       ret_from_kernel_thread+0x14/0x1c

To avoid races such as described above, initialize init_zero_pfn at
early_initcall level.  Depending on the architecture, ZERO_PAGE is
either constant or gets initialized even earlier, at paging_init, so
there is no issue with initializing zero_pfn earlier.

Link: https://lkml.kernel.org/r/CALCv0x2YqOXEAy2Q=hafjhHCtTHVodChv1qpM=niAXOpqEbt7w@mail.gmail.com
Signed-off-by: Ilya Lipnitskiy <ilya.lipnitskiy@gmail.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: stable@vger.kernel.org
Tested-by: 周琰杰 (Zhou Yanjie) <zhouyanjie@wanyeetech.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2021-03-30 09:46:12 -07:00
..
kasan kasan: clarify that only first bug is reported in HW_TAGS 2021-02-26 09:41:03 -08:00
kfence kfence: make compatible with kmemleak 2021-03-25 09:22:55 -07:00
Kconfig media: videobuf2: Move frame_vector into media subsystem 2021-01-12 14:15:31 +01:00
Kconfig.debug mm, page_poison: remove CONFIG_PAGE_POISONING_ZERO 2020-12-15 12:13:46 -08:00
Makefile mm: add Kernel Electric-Fence infrastructure 2021-02-26 09:41:02 -08:00
backing-dev.c mm/backing-dev.c: use might_alloc() 2021-02-26 09:41:01 -08:00
balloon_compaction.c
cleancache.c
cma.c mm: cma: print region name on failure 2021-02-26 09:41:00 -08:00
cma.h
cma_debug.c
compaction.c mm, compaction: make fast_isolate_freepages() stay within zone 2021-02-24 13:38:34 -08:00
debug.c mm/debug: improve memcg debugging 2021-02-24 13:38:27 -08:00
debug_page_ref.c
debug_vm_pgtable.c mm/debug_vm_pgtable/basic: iterate over entire protection_map[] 2021-02-24 13:38:27 -08:00
dmapool.c mm/dmapool: use might_alloc() 2021-02-26 09:41:01 -08:00
early_ioremap.c mm/early_ioremap.c: use __func__ instead of function name 2021-02-26 09:41:02 -08:00
fadvise.c
failslab.c
filemap.c mm: pass pvec directly to find_get_entries 2021-02-26 09:40:59 -08:00
frontswap.c
gup.c mm/hugetlb: grab head page refcount once for group of subpages 2021-02-24 13:38:32 -08:00
gup_test.c mm/gup_test.c: mark gup_test_init as __init function 2020-12-15 12:13:38 -08:00
gup_test.h selftests/vm: gup_test: introduce the dump_pages() sub-test 2020-12-15 12:13:38 -08:00
highmem.c mm/highmem: fix CONFIG_DEBUG_KMAP_LOCAL_FORCE_MAP 2021-03-25 09:22:55 -07:00
hmm.c
huge_memory.c mm/memcg: rename mem_cgroup_split_huge_fixup to split_page_memcg and add nr_pages argument 2021-03-13 11:27:31 -08:00
hugetlb.c hugetlb_cgroup: fix imbalanced css_get and css_put pair for shared mappings 2021-03-25 09:22:55 -07:00
hugetlb_cgroup.c hugetlb_cgroup: fix imbalanced css_get and css_put pair for shared mappings 2021-03-25 09:22:55 -07:00
hwpoison-inject.c
init-mm.c mm/gup: prevent gup_fast from racing with COW during fork 2020-12-15 12:13:39 -08:00
internal.h mm: introduce page_needs_cow_for_dma() for deciding whether cow 2021-03-13 11:27:30 -08:00
interval_tree.c
ioremap.c
khugepaged.c mm,thp,shmem: make khugepaged obey tmpfs mount flags 2021-02-26 09:40:59 -08:00
kmemleak.c kfence: make compatible with kmemleak 2021-03-25 09:22:55 -07:00
ksm.c mm: cleanup kstrto*() usage 2020-12-15 12:13:47 -08:00
list_lru.c mm/list_lru.c: remove kvfree_rcu_local() 2021-02-24 13:38:30 -08:00
maccess.c
madvise.c mm/madvise: replace ptrace attach requirement for process_madvise 2021-03-13 11:27:30 -08:00
mapping_dirty_helpers.c mm/mapping_dirty_helpers: enhance the kernel-doc markups 2020-12-15 12:13:41 -08:00
memblock.c memblock: remove return value of memblock_free_all() 2021-02-22 13:01:23 -08:00
memcontrol.c mm/memcg: rename mem_cgroup_split_huge_fixup to split_page_memcg and add nr_pages argument 2021-03-13 11:27:31 -08:00
memfd.c
memory-failure.c mm: fix memory_failure() handling of dax-namespace metadata 2021-02-26 09:41:00 -08:00
memory.c mm: fix race by making init_zero_pfn() early_initcall 2021-03-30 09:46:12 -07:00
memory_hotplug.c arm64: mte: Map hotplugged memory as Normal Tagged 2021-03-10 10:56:46 +00:00
mempolicy.c mm/mempolicy: use helper range_in_vma() in queue_pages_test_walk() 2021-02-24 13:38:34 -08:00
mempool.c kasan: move _RET_IP_ to inline wrappers 2021-02-24 13:38:31 -08:00
memremap.c mm/memory_hotplug: prevalidate the address range being added with platform 2021-02-26 09:41:00 -08:00
memtest.c
migrate.c mm: memcg: add swapcache stat for memcg v2 2021-02-24 13:38:29 -08:00
mincore.c inode: make init and permission helpers idmapped mount aware 2021-01-24 14:27:16 +01:00
mlock.c mm/mlock: stop counting mlocked pages when none vma is found 2021-02-26 09:41:01 -08:00
mm_init.c mm: fix fall-through warnings for Clang 2020-12-15 12:13:47 -08:00
mmap.c mm/mmap.c: remove unnecessary local variable 2021-02-24 13:38:30 -08:00
mmap_lock.c mm: mmap_lock: add tracepoints around lock acquisition 2020-12-15 12:13:41 -08:00
mmu_gather.c tlb: mmu_gather: Remove start/end arguments from tlb_gather_mmu() 2021-01-29 20:02:29 +01:00
mmu_notifier.c mm/mmu_notifiers: ensure range_end() is paired with range_start() 2021-03-25 09:22:55 -07:00
mmzone.c mm/lru: replace pgdat lru_lock with lruvec lock 2020-12-15 14:48:04 -08:00
mprotect.c mm/mprotect.c: optimize error detection in do_mprotect_pkey() 2021-02-24 13:38:30 -08:00
mremap.c mm: mremap: unlink anon_vmas when mremap with MREMAP_DONTUNMAP success 2021-02-24 13:38:30 -08:00
msync.c
nommu.c mm/nommu: Fix return type of filemap_map_pages() 2021-01-28 14:10:31 +00:00
oom_kill.c mm, oom: fix a comment in dump_task() 2021-02-24 13:38:34 -08:00
page-writeback.c mm/writeback: Add wait_on_page_writeback_killable 2021-03-23 20:54:29 +00:00
page_alloc.c mm/memcg: set memcg when splitting page 2021-03-13 11:27:31 -08:00
page_counter.c mm/page_counter: use page_counter_read in page_counter_set_max 2020-12-15 12:13:40 -08:00
page_ext.c mm: fix some spelling mistakes in comments 2020-12-15 22:46:19 -08:00
page_idle.c mm: page_idle_get_page() does not need lru_lock 2020-12-15 14:48:03 -08:00
page_io.c swap: fix swapfile read/write offset 2021-03-02 17:25:46 -07:00
page_isolation.c mm/page_isolation: do not isolate the max order page 2020-12-15 12:13:45 -08:00
page_owner.c mm/page_owner: use helper function zone_end_pfn() to get end_pfn 2021-02-24 13:38:27 -08:00
page_poison.c kasan, mm: reset tags when accessing metadata 2020-12-22 12:55:08 -08:00
page_reporting.c mm/page_reporting: use list_entry_is_head() in page_reporting_cycle() 2021-02-24 13:38:30 -08:00
page_reporting.h
page_vma_mapped.c mm/page_vma_mapped.c: add colon to fix kernel-doc markups error for check_pte 2020-12-15 12:13:41 -08:00
pagewalk.c
percpu-internal.h
percpu-km.c
percpu-stats.c
percpu-vm.c
percpu.c percpu: fix clang modpost section mismatch 2021-02-14 18:15:15 +00:00
pgalloc-track.h
pgtable-generic.c mm/pgtable-generic.c: optimize the VM_BUG_ON condition in pmdp_huge_clear_flush() 2021-02-24 13:38:30 -08:00
process_vm_access.c mm/process_vm_access.c: include compat.h 2021-01-12 18:12:54 -08:00
ptdump.c kasan, arm64: expand CONFIG_KASAN checks 2020-12-22 12:55:08 -08:00
readahead.c
rmap.c mm/rmap: correct obsolete comment of page_get_anon_vma() 2021-02-26 09:41:01 -08:00
rodata_test.c
shmem.c mm,shmem,thp: limit shmem THP allocations to requested zones 2021-02-26 09:40:59 -08:00
shuffle.c
shuffle.h
slab.c kfence, slab: fix cache_alloc_debugcheck_after() for bulk allocations 2021-03-13 11:27:30 -08:00
slab.h mm: memcg/slab: pre-allocate obj_cgroups for slab caches with SLAB_ACCOUNT 2021-02-24 13:38:29 -08:00
slab_common.c kasan, mm: optimize krealloc poisoning 2021-02-26 09:41:03 -08:00
slob.c mm, tracing: record slab name for kmem_cache_free() 2021-02-24 13:38:26 -08:00
slub.c Revert "mm, slub: consider rest of partial list if acquire_slab() fails" 2021-03-10 10:18:04 -08:00
sparse-vmemmap.c
sparse.c
swap.c mm: remove pagevec_lookup_entries 2021-02-26 09:40:59 -08:00
swap_cgroup.c
swap_slots.c mm/swap_slots.c: remove redundant NULL check 2021-02-24 13:38:28 -08:00
swap_state.c mm: add FGP_ENTRY 2021-02-26 09:40:59 -08:00
swapfile.c swap: fix swapfile read/write offset 2021-03-02 17:25:46 -07:00
truncate.c mm: remove pagevec_lookup_entries 2021-02-26 09:40:59 -08:00
usercopy.c
userfaultfd.c
util.c mm: Make mem_dump_obj() handle vmalloc() memory 2021-01-22 15:24:04 -08:00
vmacache.c
vmalloc.c Merge branch 'for-mingo-rcu' of git://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-rcu into core/rcu 2021-02-12 12:56:55 +01:00
vmpressure.c
vmscan.c mm/vmscan: restore zone_reclaim_mode ABI 2021-02-24 13:38:34 -08:00
vmstat.c mm/vmstat.c: erase latency in vmstat_shepherd 2021-02-26 09:41:00 -08:00
workingset.c mm: workingset: clarify eviction order and distance calculation 2021-02-24 13:38:34 -08:00
z3fold.c z3fold: prevent reclaim/free race for headless pages 2021-03-25 09:22:55 -07:00
zbud.c mm: set the sleep_mapped to true for zbud and z3fold 2021-02-26 09:41:01 -08:00
zpool.c mm/zswap: add the flag can_sleep_mapped 2021-02-26 09:41:01 -08:00
zsmalloc.c mm/zsmalloc.c: use page_private() to access page->private 2021-02-26 09:41:01 -08:00
zswap.c mm/zswap: add the flag can_sleep_mapped 2021-02-26 09:41:01 -08:00