linux-kernel - Re: [WARN] LOCKDEP: MAX_LOCKDEP_CHAIN

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <c0fb67e3-ebe0-45ec-9f61-a8cd5949cc42@redhat.com>
Date: Tue, 27 May 2025 00:48:21 -0400
From: Waiman Long <llong@...hat.com>
To: Lance Yang <ioworker0@...il.com>, peterz@...radead.org
Cc: mingo@...hat.com, will@...nel.org, boqun.feng@...il.com,
 linux-kernel@...r.kernel.org
Subject: Re: [WARN] LOCKDEP: MAX_LOCKDEP_CHAIN_HLOCKS too low

On 5/26/25 10:02 PM, Lance Yang wrote:
> From: Lance Yang <lance.yang@...ux.dev>
>
> Hi all,
>
> With CONFIG_LOCKDEP on, I got this warning during kernel builds:
>
> [Tue May 27 00:22:59 2025] BUG: MAX_LOCKDEP_CHAIN_HLOCKS too low!
> [Tue May 27 00:22:59 2025] turning off the locking correctness validator.
> [Tue May 27 00:22:59 2025] CPU: 56 UID: 0 PID: 3362352 Comm: cc1 Kdump: loaded Tainted: G S                  6.15.0-rc6 #5 PREEMPT(voluntary)
> [Tue May 27 00:22:59 2025] Tainted: [S]=CPU_OUT_OF_SPEC
> [Tue May 27 00:22:59 2025] Hardware name: New H3C Technologies Co., Ltd. H3C UniServer R4900 G5/RS35M2C16SB, BIOS 5.69 10/18/2023
> [Tue May 27 00:22:59 2025] Call Trace:
> [Tue May 27 00:22:59 2025]  <TASK>
> [Tue May 27 00:22:59 2025]  show_stack+0x4d/0x60
> [Tue May 27 00:22:59 2025]  dump_stack_lvl+0x72/0xa0
> [Tue May 27 00:22:59 2025]  dump_stack+0x14/0x1a
> [Tue May 27 00:22:59 2025]  add_chain_cache+0x304/0x330
> [Tue May 27 00:22:59 2025]  __lock_acquire+0x7d3/0xfd0
> [Tue May 27 00:22:59 2025]  lock_acquire.part.0+0xb4/0x210
> [Tue May 27 00:22:59 2025]  ? bad_range+0xa6/0x320
> [Tue May 27 00:22:59 2025]  ? mark_usage+0x68/0x130
> [Tue May 27 00:22:59 2025]  lock_acquire+0x62/0x120
> [Tue May 27 00:22:59 2025]  ? bad_range+0xa6/0x320
> [Tue May 27 00:22:59 2025]  seqcount_lockdep_reader_access+0x3d/0xa0
> [Tue May 27 00:22:59 2025]  ? bad_range+0xa6/0x320
> [Tue May 27 00:22:59 2025]  bad_range+0xa6/0x320
> [Tue May 27 00:22:59 2025]  ? __kasan_check_write+0x18/0x20
> [Tue May 27 00:22:59 2025]  expand+0x91/0x3c0
> [Tue May 27 00:22:59 2025]  ? __del_page_from_free_list+0x82/0x4b0
> [Tue May 27 00:22:59 2025]  rmqueue_bulk+0x13a/0xc00
> [Tue May 27 00:22:59 2025]  ? lock_acquire.part.0+0xb4/0x210
> [Tue May 27 00:22:59 2025]  __rmqueue_pcplist+0x4a7/0x8f0
> [Tue May 27 00:22:59 2025]  rmqueue_pcplist+0xcc/0x2a0
> [Tue May 27 00:22:59 2025]  rmqueue.isra.0+0xd26/0x1470
> [Tue May 27 00:22:59 2025]  ? stack_trace_save+0x96/0xd0
> [Tue May 27 00:22:59 2025]  ? __pfx_stack_trace_save+0x10/0x10
> [Tue May 27 00:22:59 2025]  ? stack_depot_save_flags+0x41/0x6a0
> [Tue May 27 00:22:59 2025]  ? lock_acquire.part.0+0xb4/0x210
> [Tue May 27 00:22:59 2025]  get_page_from_freelist+0x262/0x11a0
> [Tue May 27 00:22:59 2025]  ? kasan_save_stack+0x3e/0x50
> [Tue May 27 00:22:59 2025]  ? kasan_save_stack+0x2f/0x50
> [Tue May 27 00:22:59 2025]  ? __call_rcu_common.constprop.0+0xc4/0x950
> [Tue May 27 00:22:59 2025]  ? commit_merge+0x634/0x1100
> [Tue May 27 00:22:59 2025]  __alloc_frozen_pages_noprof+0x30e/0x6c0
> [Tue May 27 00:22:59 2025]  ? __pfx___alloc_frozen_pages_noprof+0x10/0x10
> [Tue May 27 00:22:59 2025]  ? __lock_acquire+0x3dc/0xfd0
> [Tue May 27 00:22:59 2025]  ? mark_usage+0x68/0x130
> [Tue May 27 00:22:59 2025]  ? policy_nodemask+0x21d/0x350
> [Tue May 27 00:22:59 2025]  alloc_pages_mpol+0x163/0x460
> [Tue May 27 00:22:59 2025]  ? __pfx_alloc_pages_mpol+0x10/0x10
> [Tue May 27 00:22:59 2025]  ? ___slab_alloc+0xe3/0x10f0
> [Tue May 27 00:22:59 2025]  ? find_held_lock+0x31/0x90
> [Tue May 27 00:22:59 2025]  alloc_frozen_pages_noprof+0x4b/0x130
> [Tue May 27 00:22:59 2025]  allocate_slab+0x23a/0x380
> [Tue May 27 00:22:59 2025]  ___slab_alloc+0x985/0x10f0
> [Tue May 27 00:22:59 2025]  ? find_held_lock+0x31/0x90
> [Tue May 27 00:22:59 2025]  ? xfs_buf_item_init+0x7b/0x660 [xfs]
> [Tue May 27 00:22:59 2025]  ? xfs_buf_item_init+0x7b/0x660 [xfs]
> [Tue May 27 00:22:59 2025]  ? lock_release.part.0+0x20/0x60
> [Tue May 27 00:22:59 2025]  ? fs_reclaim_acquire+0x83/0x120
> [Tue May 27 00:22:59 2025]  ? xfs_buf_item_init+0x7b/0x660 [xfs]
> [Tue May 27 00:22:59 2025]  kmem_cache_alloc_noprof+0x1ed/0x430
> [Tue May 27 00:22:59 2025]  ? kmem_cache_alloc_noprof+0x1ed/0x430
> [Tue May 27 00:22:59 2025]  xfs_buf_item_init+0x7b/0x660 [xfs]
> [Tue May 27 00:22:59 2025]  ? xfs_imap_to_bp+0x10b/0x2b0 [xfs]
> [Tue May 27 00:22:59 2025]  ? __pfx_xfs_buf_read_map+0x10/0x10 [xfs]
> [Tue May 27 00:22:59 2025]  ? xfs_file_buffered_write+0x14c/0xa50 [xfs]
> [Tue May 27 00:22:59 2025]  _xfs_trans_bjoin+0x45/0x130 [xfs]
> [Tue May 27 00:22:59 2025]  xfs_trans_read_buf_map+0x38c/0x840 [xfs]
> [Tue May 27 00:22:59 2025]  ? __pfx_xfs_trans_read_buf_map+0x10/0x10 [xfs]
> [Tue May 27 00:22:59 2025]  ? lock_acquire.part.0+0xb4/0x210
> [Tue May 27 00:22:59 2025]  xfs_imap_to_bp+0x10b/0x2b0 [xfs]
> [Tue May 27 00:22:59 2025]  ? __pfx_xfs_imap_to_bp+0x10/0x10 [xfs]
> [Tue May 27 00:22:59 2025]  ? __kasan_check_read+0x15/0x20
> [Tue May 27 00:22:59 2025]  ? do_raw_spin_unlock+0x5d/0x1f0
> [Tue May 27 00:22:59 2025]  xfs_inode_item_precommit+0x538/0xc10 [xfs]
> [Tue May 27 00:22:59 2025]  ? __pfx_xfs_inode_item_precommit+0x10/0x10 [xfs]
> [Tue May 27 00:22:59 2025]  __xfs_trans_commit+0x2a3/0xba0 [xfs]
> [Tue May 27 00:22:59 2025]  ? __pfx___xfs_trans_commit+0x10/0x10 [xfs]
> [Tue May 27 00:22:59 2025]  ? ktime_get_coarse_real_ts64_mg+0x61/0x1d0
> [Tue May 27 00:22:59 2025]  ? __kasan_check_read+0x15/0x20
> [Tue May 27 00:22:59 2025]  xfs_trans_commit+0xce/0x150 [xfs]
> [Tue May 27 00:22:59 2025]  ? xfs_trans_ijoin+0xcf/0x170 [xfs]
> [Tue May 27 00:22:59 2025]  ? __pfx_xfs_trans_commit+0x10/0x10 [xfs]
> [Tue May 27 00:22:59 2025]  xfs_vn_update_time+0x1fc/0x440 [xfs]
> [Tue May 27 00:22:59 2025]  ? __pfx_xfs_vn_update_time+0x10/0x10 [xfs]
> [Tue May 27 00:22:59 2025]  ? __kasan_check_read+0x15/0x20
> [Tue May 27 00:22:59 2025]  kiocb_modified+0x1a6/0x240
> [Tue May 27 00:22:59 2025]  xfs_file_write_checks.constprop.0+0x451/0x860 [xfs]
> [Tue May 27 00:22:59 2025]  xfs_file_buffered_write+0x14c/0xa50 [xfs]
> [Tue May 27 00:22:59 2025]  ? __pfx_xfs_file_buffered_write+0x10/0x10 [xfs]
> [Tue May 27 00:22:59 2025]  ? ovl_other_xattr_get+0xee/0x160 [overlay]
> [Tue May 27 00:22:59 2025]  ? find_held_lock+0x31/0x90
> [Tue May 27 00:22:59 2025]  ? __pfx_ovl_other_xattr_get+0x10/0x10 [overlay]
> [Tue May 27 00:22:59 2025]  ? mark_usage+0x68/0x130
> [Tue May 27 00:22:59 2025]  xfs_file_write_iter+0x553/0x830 [xfs]
> [Tue May 27 00:22:59 2025]  do_iter_readv_writev+0x422/0x910
> [Tue May 27 00:22:59 2025]  ? lock_acquire.part.0+0xb4/0x210
> [Tue May 27 00:22:59 2025]  ? backing_file_write_iter.part.0+0x4ee/0x7e0
> [Tue May 27 00:22:59 2025]  ? __pfx_do_iter_readv_writev+0x10/0x10
> [Tue May 27 00:22:59 2025]  ? selinux_file_permission+0x389/0x470
> [Tue May 27 00:22:59 2025]  ? lock_is_held_type+0xa8/0x120
> [Tue May 27 00:22:59 2025]  vfs_iter_write+0x17b/0x7a0
> [Tue May 27 00:22:59 2025]  backing_file_write_iter.part.0+0x4ee/0x7e0
> [Tue May 27 00:22:59 2025]  ? ovl_real_file+0x16a/0x1b0 [overlay]
> [Tue May 27 00:22:59 2025]  backing_file_write_iter+0xc8/0x110
> [Tue May 27 00:22:59 2025]  ovl_write_iter+0x2cc/0x450 [overlay]
> [Tue May 27 00:22:59 2025]  ? __pfx_ovl_write_iter+0x10/0x10 [overlay]
> [Tue May 27 00:22:59 2025]  ? __pfx_ovl_file_end_write+0x10/0x10 [overlay]
> [Tue May 27 00:22:59 2025]  ? lock_is_held_type+0xa8/0x120
> [Tue May 27 00:22:59 2025]  vfs_write+0x5c1/0x1050
> [Tue May 27 00:22:59 2025]  ? __pfx_vfs_write+0x10/0x10
> [Tue May 27 00:22:59 2025]  ? ktime_get_coarse_real_ts64+0x44/0xd0
> [Tue May 27 00:22:59 2025]  ? lockdep_hardirqs_on_prepare.part.0+0xa3/0x140
> [Tue May 27 00:22:59 2025]  ksys_write+0x109/0x200
> [Tue May 27 00:22:59 2025]  ? __lock_release.isra.0+0x60/0x160
> [Tue May 27 00:22:59 2025]  ? __pfx_ksys_write+0x10/0x10
> [Tue May 27 00:22:59 2025]  ? __audit_syscall_entry+0x2ef/0x540
> [Tue May 27 00:22:59 2025]  ? irqentry_exit_to_user_mode+0x7d/0x290
> [Tue May 27 00:22:59 2025]  ? irqentry_exit+0x6f/0xa0
> [Tue May 27 00:22:59 2025]  __x64_sys_write+0x76/0xb0
> [Tue May 27 00:22:59 2025]  x64_sys_call+0x28a/0x1d70
> [Tue May 27 00:22:59 2025]  do_syscall_64+0x77/0x180
> [Tue May 27 00:22:59 2025]  entry_SYSCALL_64_after_hwframe+0x76/0x7e
> [Tue May 27 00:22:59 2025] RIP: 0033:0x7f56cbac9687
> [Tue May 27 00:22:59 2025] Code: 48 89 fa 4c 89 df e8 58 b3 00 00 8b 93 08 03 00 00 59 5e 48 83 f8 fc 74 1a 5b c3 0f 1f 84 00 00 00 00 00 48 8b 44 24 10 0f 05 <5b> c3 0f 1f 80 00 00 00 00 83 e2 39 83 fa 08 75 de e8 23 ff ff ff
> [Tue May 27 00:22:59 2025] RSP: 002b:00007ffe65c16880 EFLAGS: 00000202 ORIG_RAX: 0000000000000001
> [Tue May 27 00:22:59 2025] RAX: ffffffffffffffda RBX: 00007f56cba39440 RCX: 00007f56cbac9687
> [Tue May 27 00:22:59 2025] RDX: 0000000000001000 RSI: 0000000004735f60 RDI: 0000000000000003
> [Tue May 27 00:22:59 2025] RBP: 0000000004735f60 R08: 0000000000000000 R09: 0000000000000000
> [Tue May 27 00:22:59 2025] R10: 0000000000000000 R11: 0000000000000202 R12: 0000000000001000
> [Tue May 27 00:22:59 2025] R13: 00000000046ecad0 R14: 00007f56cbc1fe80 R15: 0000000000000028
> [Tue May 27 00:22:59 2025]  </TASK>
>
> $ cat .config|grep CONFIG_LOCKDEP
> CONFIG_LOCKDEP_SUPPORT=y
> CONFIG_LOCKDEP=y
> CONFIG_LOCKDEP_BITS=15
> CONFIG_LOCKDEP_CHAINS_BITS=16
> CONFIG_LOCKDEP_STACK_TRACE_BITS=19
> CONFIG_LOCKDEP_STACK_TRACE_HASH_BITS=14
> CONFIG_LOCKDEP_CIRCULAR_QUEUE_BITS=12
>
> Is it safe? Or could this be a real locking issue?

The lock chains store the locking order of nested locks. The default 
value of 16 may be too low now as the kernel is becoming more complex in 
term of possible nested locking orders. Anyway, I would suggest upping 
the CONFIG_LOCKDEP_CHAIN_BITS to 17 or even 18 to prevent this kind of 
problem. In fact, the latest RHEL debug kernel sets 
CONFIG_LOCKDEP_CHAINS_BITS to 18.

Cheers,
Longman