Revision 4837fe37adff1d159904f0c013471b1ecbcb455e authored by Michal Hocko on 14 December 2017, 23:33:15 UTC, committed by Linus Torvalds on 15 December 2017, 00:00:49 UTC
David Rientjes has reported the following memory corruption while the
oom reaper tries to unmap the victims address space

  BUG: Bad page map in process oom_reaper  pte:6353826300000000 pmd:00000000
  addr:00007f50cab1d000 vm_flags:08100073 anon_vma:ffff9eea335603f0 mapping:          (null) index:7f50cab1d
  file:          (null) fault:          (null) mmap:          (null) readpage:          (null)
  CPU: 2 PID: 1001 Comm: oom_reaper
  Call Trace:
     unmap_page_range+0x1068/0x1130
     __oom_reap_task_mm+0xd5/0x16b
     oom_reaper+0xff/0x14c
     kthread+0xc1/0xe0

Tetsuo Handa has noticed that the synchronization inside exit_mmap is
insufficient.  We only synchronize with the oom reaper if
tsk_is_oom_victim which is not true if the final __mmput is called from
a different context than the oom victim exit path.  This can trivially
happen from context of any task which has grabbed mm reference (e.g.  to
read /proc/<pid>/ file which requires mm etc.).

The race would look like this

  oom_reaper		oom_victim		task
						mmget_not_zero
			do_exit
			  mmput
  __oom_reap_task_mm				mmput
  						  __mmput
						    exit_mmap
						      remove_vma
    unmap_page_range

Fix this issue by providing a new mm_is_oom_victim() helper which
operates on the mm struct rather than a task.  Any context which
operates on a remote mm struct should use this helper in place of
tsk_is_oom_victim.  The flag is set in mark_oom_victim and never cleared
so it is stable in the exit_mmap path.

Debugged by Tetsuo Handa.

Link: http://lkml.kernel.org/r/20171210095130.17110-1-mhocko@kernel.org
Fixes: 212925802454 ("mm: oom: let oom_reap_task and exit_mmap run concurrently")
Signed-off-by: Michal Hocko <mhocko@suse.com>
Reported-by: David Rientjes <rientjes@google.com>
Acked-by: David Rientjes <rientjes@google.com>
Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Cc: Andrea Argangeli <andrea@kernel.org>
Cc: <stable@vger.kernel.org>	[4.14]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
1 parent bdcf0a4
History
File Mode Size
Makefile -rw-r--r-- 1.5 KB
amdtopology.c -rw-r--r-- 4.0 KB
debug_pagetables.c -rw-r--r-- 997 bytes
dump_pagetables.c -rw-r--r-- 13.7 KB
extable.c -rw-r--r-- 7.5 KB
fault.c -rw-r--r-- 37.9 KB
highmem_32.c -rw-r--r-- 3.2 KB
hugetlbpage.c -rw-r--r-- 5.2 KB
ident_map.c -rw-r--r-- 3.1 KB
init.c -rw-r--r-- 24.7 KB
init_32.c -rw-r--r-- 24.5 KB
init_64.c -rw-r--r-- 36.9 KB
iomap_32.c -rw-r--r-- 3.5 KB
ioremap.c -rw-r--r-- 21.8 KB
kasan_init_64.c -rw-r--r-- 8.8 KB
kaslr.c -rw-r--r-- 7.2 KB
kmmio.c -rw-r--r-- 16.1 KB
mem_encrypt.c -rw-r--r-- 23.8 KB
mem_encrypt_boot.S -rw-r--r-- 4.4 KB
mm_internal.h -rw-r--r-- 558 bytes
mmap.c -rw-r--r-- 6.9 KB
mmio-mod.c -rw-r--r-- 11.8 KB
mpx.c -rw-r--r-- 26.0 KB
numa.c -rw-r--r-- 22.7 KB
numa_32.c -rw-r--r-- 3.3 KB
numa_64.c -rw-r--r-- 251 bytes
numa_emulation.c -rw-r--r-- 13.1 KB
numa_internal.h -rw-r--r-- 765 bytes
pageattr-test.c -rw-r--r-- 5.3 KB
pageattr.c -rw-r--r-- 50.4 KB
pat.c -rw-r--r-- 28.2 KB
pat_internal.h -rw-r--r-- 1.5 KB
pat_rbtree.c -rw-r--r-- 6.8 KB
pf_in.c -rw-r--r-- 10.5 KB
pf_in.h -rw-r--r-- 1.4 KB
pgtable.c -rw-r--r-- 16.4 KB
pgtable_32.c -rw-r--r-- 2.2 KB
physaddr.c -rw-r--r-- 2.2 KB
physaddr.h -rw-r--r-- 231 bytes
pkeys.c -rw-r--r-- 6.4 KB
setup_nx.c -rw-r--r-- 1.4 KB
srat.c -rw-r--r-- 2.9 KB
testmmiotrace.c -rw-r--r-- 3.1 KB
tlb.c -rw-r--r-- 18.9 KB

back to top