Message-ID: <20230504084019.wdtrjvn7s4vs6koz@quack3>
Date:   Thu, 4 May 2023 10:40:19 +0200
From:   Jan Kara <jack@...e.cz>
To:     syzbot <syzbot+6898da502aef574c5f8a@...kaller.appspotmail.com>
Cc:     linux-ext4@...r.kernel.org, akpm@...ux-foundation.org,
        linux-fsdevel@...r.kernel.org, linux-kernel@...r.kernel.org,
        linux-mm@...ck.org, syzkaller-bugs@...glegroups.com,
        willy@...radead.org
Subject: Re: [syzbot] [fs?] [mm?] possible deadlock in do_writepages

On Mon 24-04-23 01:11:37, syzbot wrote:
> Hello,
> 
> syzbot found the following issue on:
> 
> HEAD commit:    e8d018dd0257 Linux 6.3-rc3
> git tree:       upstream
> console output: https://syzkaller.appspot.com/x/log.txt?x=13cd8a86c80000
> kernel config:  https://syzkaller.appspot.com/x/.config?x=cde06fe2cf5765b7
> dashboard link: https://syzkaller.appspot.com/bug?extid=6898da502aef574c5f8a
> compiler:       gcc (Debian 10.2.1-6) 10.2.1 20210110, GNU ld (GNU Binutils for Debian) 2.35.2
> 
> Unfortunately, I don't have any reproducer for this issue yet.
> 
> IMPORTANT: if you fix the issue, please add the following tag to the commit:
> Reported-by: syzbot+6898da502aef574c5f8a@...kaller.appspotmail.com
> 
> ======================================================
> WARNING: possible circular locking dependency detected
> 6.3.0-rc3-syzkaller #0 Not tainted
> ------------------------------------------------------
> kswapd0/100 is trying to acquire lock:
> ffff88801e636b98 (&sbi->s_writepages_rwsem){++++}-{0:0}, at: do_writepages+0x1a8/0x640 mm/page-writeback.c:2551
> 
> but task is already holding lock:
> ffffffff8c8e1280 (fs_reclaim){+.+.}-{0:0}, at: set_task_reclaim_state mm/vmscan.c:200 [inline]
> ffffffff8c8e1280 (fs_reclaim){+.+.}-{0:0}, at: balance_pgdat+0x170/0x1ac0 mm/vmscan.c:7338
> 
> which lock already depends on the new lock.

This looks like a real deadlock opportunity in ext4 due to
EXT4_SB(sb)->s_writepages_rwsem. The problem is like:

CPU0				CPU1

ext4_writepages()
  percpu_down_read(sbi->s_writepages_rwsem);
				ext4_change_inode_journal_flag()
				  percpu_down_write(sbi->s_writepages_rwsem);
				    - blocks, all readers block from now on
  ext4_do_writepages()
    ext4_init_io_end()
      kmem_cache_zalloc(io_end_cachep, GFP_KERNEL)
        fs_reclaim frees dentry...
          dentry_unlink_inode()
            iput() - last ref =>
              iput_final() - inode dirty =>
                write_inode_now()...
                  ext4_writepages() tries to acquire sbi->s_writepages_rwsem

So we need to disable fs reclaim whenever we acquire s_writepages_rwsem.
I'll send a patch.

								Honza

> 
> the existing dependency chain (in reverse order) is:
> 
> -> #1 (fs_reclaim){+.+.}-{0:0}:
>        __fs_reclaim_acquire mm/page_alloc.c:4716 [inline]
>        fs_reclaim_acquire+0x11d/0x160 mm/page_alloc.c:4730
>        might_alloc include/linux/sched/mm.h:271 [inline]
>        slab_pre_alloc_hook mm/slab.h:728 [inline]
>        slab_alloc_node mm/slab.c:3241 [inline]
>        slab_alloc mm/slab.c:3266 [inline]
>        __kmem_cache_alloc_lru mm/slab.c:3443 [inline]
>        kmem_cache_alloc+0x3d/0x3f0 mm/slab.c:3452
>        kmem_cache_zalloc include/linux/slab.h:710 [inline]
>        ext4_init_io_end+0x27/0x180 fs/ext4/page-io.c:278
>        ext4_do_writepages+0xbea/0x31e0 fs/ext4/inode.c:2824
>        ext4_writepages+0x27c/0x5e0 fs/ext4/inode.c:2964
>        do_writepages+0x1a8/0x640 mm/page-writeback.c:2551
>        __writeback_single_inode+0x121/0xdc0 fs/fs-writeback.c:1600
>        writeback_sb_inodes+0x54d/0xe70 fs/fs-writeback.c:1891
>        __writeback_inodes_wb+0xc6/0x280 fs/fs-writeback.c:1962
>        wb_writeback+0x7e9/0xa50 fs/fs-writeback.c:2067
>        wb_check_background_flush fs/fs-writeback.c:2133 [inline]
>        wb_do_writeback fs/fs-writeback.c:2221 [inline]
>        wb_workfn+0x8b0/0xfc0 fs/fs-writeback.c:2248
>        process_one_work+0x991/0x15c0 kernel/workqueue.c:2390
>        worker_thread+0x669/0x1090 kernel/workqueue.c:2537
>        kthread+0x2e8/0x3a0 kernel/kthread.c:376
>        ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:308
> 
> -> #0 (&sbi->s_writepages_rwsem){++++}-{0:0}:
>        check_prev_add kernel/locking/lockdep.c:3098 [inline]
>        check_prevs_add kernel/locking/lockdep.c:3217 [inline]
>        validate_chain kernel/locking/lockdep.c:3832 [inline]
>        __lock_acquire+0x2ec7/0x5d40 kernel/locking/lockdep.c:5056
>        lock_acquire kernel/locking/lockdep.c:5669 [inline]
>        lock_acquire+0x1af/0x520 kernel/locking/lockdep.c:5634
>        percpu_down_read include/linux/percpu-rwsem.h:51 [inline]
>        ext4_writepages+0x1a9/0x5e0 fs/ext4/inode.c:2963
>        do_writepages+0x1a8/0x640 mm/page-writeback.c:2551
>        __writeback_single_inode+0x121/0xdc0 fs/fs-writeback.c:1600
>        writeback_single_inode+0x2ad/0x590 fs/fs-writeback.c:1721
>        write_inode_now+0x16e/0x1e0 fs/fs-writeback.c:2757
>        iput_final fs/inode.c:1735 [inline]
>        iput.part.0+0x408/0x740 fs/inode.c:1774
>        iput+0x5c/0x80 fs/inode.c:1764
>        dentry_unlink_inode+0x2b1/0x460 fs/dcache.c:401
>        __dentry_kill+0x3c0/0x640 fs/dcache.c:607
>        shrink_dentry_list+0x12c/0x4f0 fs/dcache.c:1201
>        prune_dcache_sb+0xeb/0x150 fs/dcache.c:1282
>        super_cache_scan+0x33a/0x590 fs/super.c:104
>        do_shrink_slab+0x428/0xaa0 mm/vmscan.c:853
>        shrink_slab_memcg mm/vmscan.c:922 [inline]
>        shrink_slab+0x388/0x660 mm/vmscan.c:1001
>        shrink_one+0x502/0x810 mm/vmscan.c:5343
>        shrink_many mm/vmscan.c:5394 [inline]
>        lru_gen_shrink_node mm/vmscan.c:5511 [inline]
>        shrink_node+0x2064/0x35f0 mm/vmscan.c:6459
>        kswapd_shrink_node mm/vmscan.c:7262 [inline]
>        balance_pgdat+0xa02/0x1ac0 mm/vmscan.c:7452
>        kswapd+0x677/0xd60 mm/vmscan.c:7712
>        kthread+0x2e8/0x3a0 kernel/kthread.c:376
>        ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:308
> 
> other info that might help us debug this:
> 
>  Possible unsafe locking scenario:
> 
>        CPU0                    CPU1
>        ----                    ----
>   lock(fs_reclaim);
>                                lock(&sbi->s_writepages_rwsem);
>                                lock(fs_reclaim);
>   lock(&sbi->s_writepages_rwsem);
> 
>  *** DEADLOCK ***
> 
> 3 locks held by kswapd0/100:
>  #0: ffffffff8c8e1280 (fs_reclaim){+.+.}-{0:0}, at: set_task_reclaim_state mm/vmscan.c:200 [inline]
>  #0: ffffffff8c8e1280 (fs_reclaim){+.+.}-{0:0}, at: balance_pgdat+0x170/0x1ac0 mm/vmscan.c:7338
>  #1: ffffffff8c8980b0 (shrinker_rwsem){++++}-{3:3}, at: shrink_slab_memcg mm/vmscan.c:895 [inline]
>  #1: ffffffff8c8980b0 (shrinker_rwsem){++++}-{3:3}, at: shrink_slab+0x2a0/0x660 mm/vmscan.c:1001
>  #2: ffff88801e6300e0 (&type->s_umount_key#50){++++}-{3:3}, at: trylock_super fs/super.c:414 [inline]
>  #2: ffff88801e6300e0 (&type->s_umount_key#50){++++}-{3:3}, at: super_cache_scan+0x70/0x590 fs/super.c:79
> 
> stack backtrace:
> CPU: 1 PID: 100 Comm: kswapd0 Not tainted 6.3.0-rc3-syzkaller #0
> Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.14.0-2 04/01/2014
> Call Trace:
>  <TASK>
>  __dump_stack lib/dump_stack.c:88 [inline]
>  dump_stack_lvl+0xd9/0x150 lib/dump_stack.c:106
>  check_noncircular+0x25f/0x2e0 kernel/locking/lockdep.c:2178
>  check_prev_add kernel/locking/lockdep.c:3098 [inline]
>  check_prevs_add kernel/locking/lockdep.c:3217 [inline]
>  validate_chain kernel/locking/lockdep.c:3832 [inline]
>  __lock_acquire+0x2ec7/0x5d40 kernel/locking/lockdep.c:5056
>  lock_acquire kernel/locking/lockdep.c:5669 [inline]
>  lock_acquire+0x1af/0x520 kernel/locking/lockdep.c:5634
>  percpu_down_read include/linux/percpu-rwsem.h:51 [inline]
>  ext4_writepages+0x1a9/0x5e0 fs/ext4/inode.c:2963
>  do_writepages+0x1a8/0x640 mm/page-writeback.c:2551
>  __writeback_single_inode+0x121/0xdc0 fs/fs-writeback.c:1600
>  writeback_single_inode+0x2ad/0x590 fs/fs-writeback.c:1721
>  write_inode_now+0x16e/0x1e0 fs/fs-writeback.c:2757
>  iput_final fs/inode.c:1735 [inline]
>  iput.part.0+0x408/0x740 fs/inode.c:1774
>  iput+0x5c/0x80 fs/inode.c:1764
>  dentry_unlink_inode+0x2b1/0x460 fs/dcache.c:401
>  __dentry_kill+0x3c0/0x640 fs/dcache.c:607
>  shrink_dentry_list+0x12c/0x4f0 fs/dcache.c:1201
>  prune_dcache_sb+0xeb/0x150 fs/dcache.c:1282
>  super_cache_scan+0x33a/0x590 fs/super.c:104
>  do_shrink_slab+0x428/0xaa0 mm/vmscan.c:853
>  shrink_slab_memcg mm/vmscan.c:922 [inline]
>  shrink_slab+0x388/0x660 mm/vmscan.c:1001
>  shrink_one+0x502/0x810 mm/vmscan.c:5343
>  shrink_many mm/vmscan.c:5394 [inline]
>  lru_gen_shrink_node mm/vmscan.c:5511 [inline]
>  shrink_node+0x2064/0x35f0 mm/vmscan.c:6459
>  kswapd_shrink_node mm/vmscan.c:7262 [inline]
>  balance_pgdat+0xa02/0x1ac0 mm/vmscan.c:7452
>  kswapd+0x677/0xd60 mm/vmscan.c:7712
>  kthread+0x2e8/0x3a0 kernel/kthread.c:376
>  ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:308
>  </TASK>
> ------------[ cut here ]------------
> WARNING: CPU: 1 PID: 100 at fs/ext4/inode.c:5325 ext4_write_inode+0x337/0x590 fs/ext4/inode.c:5325
> Modules linked in:
> CPU: 1 PID: 100 Comm: kswapd0 Not tainted 6.3.0-rc3-syzkaller #0
> Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.14.0-2 04/01/2014
> RIP: 0010:ext4_write_inode+0x337/0x590 fs/ext4/inode.c:5325
> Code: b6 04 02 84 c0 74 08 3c 03 0f 8e 3d 02 00 00 8b b5 b0 06 00 00 4c 89 f7 e8 c6 94 12 00 41 89 c4 e9 ed fd ff ff e8 e9 4e 5a ff <0f> 0b 45 31 e4 e9 de fd ff ff e8 da 4e 5a ff 48 89 ef 48 8d 74 24
> RSP: 0018:ffffc90000c87360 EFLAGS: 00010293
> RAX: 0000000000000000 RBX: 1ffff92000190e6c RCX: 0000000000000000
> RDX: ffff8880169700c0 RSI: ffffffff8227b6c7 RDI: 0000000000000005
> RBP: ffff8880283852f0 R08: 0000000000000005 R09: 0000000000000000
> R10: 0000000000000800 R11: 3e4b5341542f3c20 R12: 0000000000000800
> R13: ffffc90000c874e0 R14: dffffc0000000000 R15: ffff888028385318
> FS:  0000000000000000(0000) GS:ffff88802ca80000(0000) knlGS:0000000000000000
> CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> CR2: 00007fe7eabfe718 CR3: 00000000738f6000 CR4: 0000000000150ee0
> DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
> Call Trace:
>  <TASK>
>  write_inode fs/fs-writeback.c:1453 [inline]
>  __writeback_single_inode+0x9f8/0xdc0 fs/fs-writeback.c:1665
>  writeback_single_inode+0x2ad/0x590 fs/fs-writeback.c:1721
>  write_inode_now+0x16e/0x1e0 fs/fs-writeback.c:2757
>  iput_final fs/inode.c:1735 [inline]
>  iput.part.0+0x408/0x740 fs/inode.c:1774
>  iput+0x5c/0x80 fs/inode.c:1764
>  dentry_unlink_inode+0x2b1/0x460 fs/dcache.c:401
>  __dentry_kill+0x3c0/0x640 fs/dcache.c:607
>  shrink_dentry_list+0x12c/0x4f0 fs/dcache.c:1201
>  prune_dcache_sb+0xeb/0x150 fs/dcache.c:1282
>  super_cache_scan+0x33a/0x590 fs/super.c:104
>  do_shrink_slab+0x428/0xaa0 mm/vmscan.c:853
>  shrink_slab_memcg mm/vmscan.c:922 [inline]
>  shrink_slab+0x388/0x660 mm/vmscan.c:1001
>  shrink_one+0x502/0x810 mm/vmscan.c:5343
>  shrink_many mm/vmscan.c:5394 [inline]
>  lru_gen_shrink_node mm/vmscan.c:5511 [inline]
>  shrink_node+0x2064/0x35f0 mm/vmscan.c:6459
>  kswapd_shrink_node mm/vmscan.c:7262 [inline]
>  balance_pgdat+0xa02/0x1ac0 mm/vmscan.c:7452
>  kswapd+0x677/0xd60 mm/vmscan.c:7712
>  kthread+0x2e8/0x3a0 kernel/kthread.c:376
>  ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:308
>  </TASK>
> 
> 
> ---
> This report is generated by a bot. It may contain errors.
> See https://goo.gl/tpsmEJ for more information about syzbot.
> syzbot engineers can be reached at syzkaller@...glegroups.com.
> 
> syzbot will keep track of this issue. See:
> https://goo.gl/tpsmEJ#status for how to communicate with syzbot.
-- 
Jan Kara <jack@...e.com>
SUSE Labs, CR

Powered by blists - more mailing lists