lowmemorykiller: Killing 'syz-executor.5' (8480) (tgid 8480), adj 1000,
   to free 41400kB on behalf of 'syz-executor.5' (9987) because
   cache 64348kB is below limit 65536kB for oom_score_adj 12
   Free memory is -12840kB above reserved
=================================
[ INFO: inconsistent lock state ]
4.9.194+ #0 Not tainted
---------------------------------
inconsistent {RECLAIM_FS-ON-W} -> {IN-RECLAIM_FS-R} usage.
syz-executor.5/9987 [HC0[0]:SC0[0]:HE1:SE1] takes:
 (&mm->mmap_sem){+++++?}, at: [<0000000087737afc>] get_cmdline+0xa3/0x2d0 mm/util.c:641
  mark_held_locks+0xb1/0x100 kernel/locking/lockdep.c:2660
  __lockdep_trace_alloc kernel/locking/lockdep.c:2882 [inline]
  lockdep_trace_alloc+0x18c/0x2b0 kernel/locking/lockdep.c:2897
  __alloc_pages_nodemask+0x143/0x1a80 mm/page_alloc.c:3803
  __alloc_pages include/linux/gfp.h:433 [inline]
  __alloc_pages_node include/linux/gfp.h:446 [inline]
  alloc_pages_node include/linux/gfp.h:460 [inline]
  pmd_alloc_one arch/x86/include/asm/pgalloc.h:88 [inline]
  __pmd_alloc+0x4a/0x330 mm/memory.c:3742
  pmd_alloc include/linux/mm.h:1625 [inline]
  alloc_new_pmd mm/mremap.c:64 [inline]
  move_page_tables+0xadb/0xd60 mm/mremap.c:212
  shift_arg_pages+0x1ae/0x470 fs/exec.c:642
  setup_arg_pages+0x60d/0x7c0 fs/exec.c:754
  load_elf_binary+0xa84/0x4a90 fs/binfmt_elf.c:860
  search_binary_handler fs/exec.c:1621 [inline]
  search_binary_handler+0x14f/0x700 fs/exec.c:1599
  exec_binprm fs/exec.c:1663 [inline]
  do_execveat_common.isra.0+0xf81/0x1db0 fs/exec.c:1785
  do_execve+0x3a/0x50 fs/exec.c:1829
  run_init_process+0x33/0x37 init/main.c:904
  try_to_run_init_process+0x18/0x48 init/main.c:913
  kernel_init+0xf2/0x163 init/main.c:984
  ret_from_fork+0x5c/0x70 arch/x86/entry/entry_64.S:375
irq event stamp: 1103507
hardirqs last  enabled at (1103507): [<00000000bad5286b>] vprintk_emit+0x25c/0x6f0 kernel/printk/printk.c:1897
hardirqs last disabled at (1103506): [<000000009f2ac57c>] vprintk_emit+0x6d/0x6f0 kernel/printk/printk.c:1801
softirqs last  enabled at (1103492): [<00000000ef4e0a37>] spin_unlock_bh include/linux/spinlock.h:352 [inline]
softirqs last  enabled at (1103492): [<00000000ef4e0a37>] wb_queue_work+0x1ed/0x450 fs/fs-writeback.c:203
softirqs last disabled at (1103488): [<00000000a7e2a1b6>] spin_lock_bh include/linux/spinlock.h:307 [inline]
softirqs last disabled at (1103488): [<00000000a7e2a1b6>] wb_queue_work+0x90/0x450 fs/fs-writeback.c:195

other info that might help us debug this:
 Possible unsafe locking scenario:

       CPU0
       ----
  lock(&mm->mmap_sem);
  <Interrupt>
    lock(&mm->mmap_sem);

 *** DEADLOCK ***

1 lock held by syz-executor.5/9987:
 #0:  (shrinker_rwsem){++++..}, at: [<000000003ecc3654>] shrink_slab.part.0+0xb2/0xa20 mm/vmscan.c:472

stack backtrace:
CPU: 1 PID: 9987 Comm: syz-executor.5 Not tainted 4.9.194+ #0
 ffff88018fd4f070 ffffffff81b67001 00000000000000f0 ffff8801ab1c2f80
 ffffffff83cb0630 ffff8801ab1c3878 ffffffff84252000 ffff88018fd4f0e8
 ffffffff81408710 0000000000000000 ffffffff00000001 0000000000000001
Call Trace:
 [<00000000d9d18b13>] __dump_stack lib/dump_stack.c:15 [inline]
 [<00000000d9d18b13>] dump_stack+0xc1/0x120 lib/dump_stack.c:51
 [<000000008e0aa140>] print_usage_bug kernel/locking/lockdep.c:2387 [inline]
 [<000000008e0aa140>] print_usage_bug.cold+0x452/0x5a2 kernel/locking/lockdep.c:2354
 [<00000000e13179e8>] valid_state kernel/locking/lockdep.c:2400 [inline]
 [<00000000e13179e8>] mark_lock_irq kernel/locking/lockdep.c:2602 [inline]
 [<00000000e13179e8>] mark_lock+0x6c7/0x12e0 kernel/locking/lockdep.c:3065
 [<00000000ba79a463>] mark_irqflags kernel/locking/lockdep.c:2958 [inline]
 [<00000000ba79a463>] __lock_acquire+0x5be/0x4390 kernel/locking/lockdep.c:3302
 [<000000002df3bd06>] lock_acquire+0x133/0x3d0 kernel/locking/lockdep.c:3756
 [<00000000f3e02c33>] down_read+0x44/0xb0 kernel/locking/rwsem.c:22
 [<0000000087737afc>] get_cmdline+0xa3/0x2d0 mm/util.c:641
 [<000000007eed0dad>] handle_lmk_event+0x169/0x920 drivers/staging/android/lowmemorykiller.c:116
 [<0000000037f82294>] lowmem_scan+0x6f3/0xb70 drivers/staging/android/lowmemorykiller.c:354
 [<00000000299601f9>] do_shrink_slab mm/vmscan.c:399 [inline]
 [<00000000299601f9>] shrink_slab.part.0+0x3cf/0xa20 mm/vmscan.c:502
 [<000000008d299679>] shrink_slab mm/vmscan.c:466 [inline]
 [<000000008d299679>] shrink_node+0x1ed/0x750 mm/vmscan.c:2604
 [<000000007f03a3ba>] shrink_zones mm/vmscan.c:2751 [inline]
 [<000000007f03a3ba>] do_try_to_free_pages mm/vmscan.c:2793 [inline]
 [<000000007f03a3ba>] try_to_free_pages+0x397/0xbd0 mm/vmscan.c:3004
 [<00000000afe2d137>] __perform_reclaim mm/page_alloc.c:3332 [inline]
 [<00000000afe2d137>] __alloc_pages_direct_reclaim mm/page_alloc.c:3354 [inline]
 [<00000000afe2d137>] __alloc_pages_slowpath mm/page_alloc.c:3704 [inline]
 [<00000000afe2d137>] __alloc_pages_nodemask+0x930/0x1a80 mm/page_alloc.c:3861
 [<000000002c29679c>] __alloc_pages include/linux/gfp.h:433 [inline]
 [<000000002c29679c>] __alloc_pages_node include/linux/gfp.h:446 [inline]
 [<000000002c29679c>] alloc_pages_node include/linux/gfp.h:460 [inline]
 [<000000002c29679c>] __vmalloc_area_node mm/vmalloc.c:1648 [inline]
 [<000000002c29679c>] __vmalloc_node_range+0x25b/0x610 mm/vmalloc.c:1706
 [<000000006830cfc1>] __vmalloc_node mm/vmalloc.c:1755 [inline]
 [<000000006830cfc1>] __vmalloc_node_flags mm/vmalloc.c:1769 [inline]
 [<000000006830cfc1>] vmalloc+0x5c/0x70 mm/vmalloc.c:1784