Skip to contentSkip to conversation

xfs : WARNING: possible circular locking dependency detected

PATCH Applied Xiubo Li 4 participants 8 messages Apr 18, 2024 – Apr 19, 2024 Library code

lore
Share & tools
lore

Xiubo Li

author
#0 lore raw

Hi all

BTW, is this a known issue and has it been fixed already ? I can reproduce this always with my VMs:

<4>[ 9009.171195]
<4>[ 9009.171205] ======================================================
<4>[ 9009.171208] WARNING: possible circular locking dependency detected
<4>[ 9009.171211] 6.9.0-rc3+ #49 Not tainted
<4>[ 9009.171214] ------------------------------------------------------
<4>[ 9009.171216] kswapd0/149 is trying to acquire lock:
<4>[ 9009.171219] ffff88811346a920 (&xfs_nondir_ilock_class){++++}-{4:4}, at: xfs_reclaim_inode+0x3ac/0x590 [xfs]
<4>[ 9009.171580]
<4>[ 9009.171580] but task is already holding lock:
<4>[ 9009.171583] ffffffff8bb33100 (fs_reclaim){+.+.}-{0:0}, at: balance_pgdat+0x5d9/0xad0
<4>[ 9009.171593]
<4>[ 9009.171593] which lock already depends on the new lock.
<4>[ 9009.171593]
<4>[ 9009.171595]
<4>[ 9009.171595] the existing dependency chain (in reverse order) is:
<4>[ 9009.171597]
<4>[ 9009.171597] -> #1 (fs_reclaim){+.+.}-{0:0}:
<4>[ 9009.171603]        __lock_acquire+0x7da/0x1030
<4>[ 9009.171610]        lock_acquire+0x15d/0x400
<4>[ 9009.171614]        fs_reclaim_acquire+0xb5/0x100
<4>[ 9009.171618] prepare_alloc_pages.constprop.0+0xc5/0x230
<4>[ 9009.171622]        __alloc_pages+0x12a/0x3f0
<4>[ 9009.171625]        alloc_pages_mpol+0x175/0x340
<4>[ 9009.171630]        stack_depot_save_flags+0x4c5/0x510
<4>[ 9009.171635]        kasan_save_stack+0x30/0x40
<4>[ 9009.171640]        kasan_save_track+0x10/0x30
<4>[ 9009.171643]        __kasan_slab_alloc+0x83/0x90
<4>[ 9009.171646]        kmem_cache_alloc+0x15e/0x4a0
<4>[ 9009.171652]        __alloc_object+0x35/0x370
<4>[ 9009.171659]        __create_object+0x22/0x90
<4>[ 9009.171665] __kmalloc_node_track_caller+0x477/0x5b0
<4>[ 9009.171672]        krealloc+0x5f/0x110
<4>[ 9009.171679]        xfs_iext_insert_raw+0x4b2/0x6e0 [xfs]
<4>[ 9009.172172]        xfs_iext_insert+0x2e/0x130 [xfs]
<4>[ 9009.172468]        xfs_iread_bmbt_block+0x1a9/0x4d0 [xfs]
<4>[ 9009.172852]        xfs_btree_visit_block+0xfb/0x290 [xfs]
<4>[ 9009.173168]        xfs_btree_visit_blocks+0x215/0x2c0 [xfs]
<4>[ 9009.173477]        xfs_iread_extents+0x1a2/0x2e0 [xfs]
<4>[ 9009.173790] xfs_buffered_write_iomap_begin+0x376/0x10a0 [xfs]
<4>[ 9009.174096]        iomap_iter+0x1d1/0x2d0
<4>[ 9009.174102] iomap_file_buffered_write+0x120/0x1a0
<4>[ 9009.174106]        xfs_file_buffered_write+0x128/0x4b0 [xfs]
<4>[ 9009.174415]        vfs_write+0x675/0x890
<4>[ 9009.174420]        ksys_write+0xc3/0x160
<4>[ 9009.174423]        do_syscall_64+0x94/0x170
<4>[ 9009.174428] entry_SYSCALL_64_after_hwframe+0x71/0x79
<4>[ 9009.174433]
<4>[ 9009.174433] -> #0 (&xfs_nondir_ilock_class){++++}-{4:4}:
<4>[ 9009.174441]        check_prev_add+0x1b2/0x1320
<4>[ 9009.174445]        validate_chain+0x8a2/0x9f0
<4>[ 9009.174448]        __lock_acquire+0x7da/0x1030
<4>[ 9009.174452]        lock_acquire+0x15d/0x400
<4>[ 9009.174455]        down_write_nested+0x9b/0x1d0
<4>[ 9009.174459]        xfs_reclaim_inode+0x3ac/0x590 [xfs]
<4>[ 9009.174770]        xfs_icwalk_ag+0x3c1/0x850 [xfs]
<4>[ 9009.175076]        xfs_icwalk+0x38/0xb0 [xfs]
<4>[ 9009.175389]        xfs_reclaim_inodes_nr+0x107/0x170 [xfs]
<4>[ 9009.175699]        super_cache_scan+0x248/0x290
<4>[ 9009.175704]        do_shrink_slab+0x239/0x8d0
<4>[ 9009.175709]        shrink_slab_memcg+0x39f/0x780
<4>[ 9009.175713]        shrink_slab+0x3a7/0x460
<4>[ 9009.175716]        shrink_node_memcgs+0x26c/0x390
<4>[ 9009.175719]        shrink_node+0x128/0x720
<4>[ 9009.175722]        balance_pgdat+0x53c/0xad0
<4>[ 9009.175725]        kswapd+0x235/0x370
<4>[ 9009.175728]        kthread+0x1a4/0x1e0
<4>[ 9009.175732]        ret_from_fork+0x30/0x50
<4>[ 9009.175736]        ret_from_fork_asm+0x1a/0x30
<4>[ 9009.175740]
<4>[ 9009.175740] other info that might help us debug this:
<4>[ 9009.175740]
<4>[ 9009.175742]  Possible unsafe locking scenario:
<4>[ 9009.175742]
<4>[ 9009.175744]        CPU0                    CPU1
<4>[ 9009.175746]        ----                    ----
<4>[ 9009.175748]   lock(fs_reclaim);
<4>[ 9009.175752] lock(&xfs_nondir_ilock_class);
<4>[ 9009.175756] lock(fs_reclaim);
<4>[ 9009.175760]   lock(&xfs_nondir_ilock_class);
<4>[ 9009.175764]
<4>[ 9009.175764]  *** DEADLOCK ***
<4>[ 9009.175764]
<4>[ 9009.175766] 2 locks held by kswapd0/149:
<4>[ 9009.175769]  #0: ffffffff8bb33100 (fs_reclaim){+.+.}-{0:0}, at: balance_pgdat+0x5d9/0xad0
<4>[ 9009.175777]  #1: ffff88814c5be0e8 (&type->s_umount_key#58){++++}-{4:4}, at: super_cache_scan+0x4b/0x290
<4>[ 9009.175788]
<4>[ 9009.175788] stack backtrace:
<4>[ 9009.175791] CPU: 0 PID: 149 Comm: kswapd0 Kdump: loaded Not tainted 6.9.0-rc3+ #49
<4>[ 9009.175795] Hardware name: innotek GmbH VirtualBox/VirtualBox, BIOS VirtualBox 12/01/2006
<4>[ 9009.175798] Call Trace:
<4>[ 9009.175801]  <TASK>
<4>[ 9009.175805]  dump_stack_lvl+0x7e/0xc0
<4>[ 9009.175811]  check_noncircular+0x267/0x2b0
<4>[ 9009.175816]  ? __pfx_check_noncircular+0x10/0x10
<4>[ 9009.175821]  ? check_path.constprop.0+0x24/0x50
<4>[ 9009.175825]  ? check_noncircular+0x132/0x2b0
<4>[ 9009.175829]  ? add_chain_block+0x1e0/0x290
<4>[ 9009.175835]  check_prev_add+0x1b2/0x1320
<4>[ 9009.175841]  validate_chain+0x8a2/0x9f0
<4>[ 9009.175846]  ? __pfx_validate_chain+0x10/0x10
<4>[ 9009.175850]  ? mark_lock.part.0+0x85/0x510
<4>[ 9009.175855]  __lock_acquire+0x7da/0x1030
<4>[ 9009.175861]  lock_acquire+0x15d/0x400
<4>[ 9009.175865]  ? xfs_reclaim_inode+0x3ac/0x590 [xfs]
<4>[ 9009.176199]  ? __pfx_lock_acquire+0x10/0x10
<4>[ 9009.176204]  ? find_held_lock+0x8c/0xa0
<4>[ 9009.176208]  ? kvm_sched_clock_read+0xd/0x20
<4>[ 9009.176212]  ? local_clock_noinstr+0x9/0xb0
<4>[ 9009.176218]  down_write_nested+0x9b/0x1d0
<4>[ 9009.176222]  ? xfs_reclaim_inode+0x3ac/0x590 [xfs]
<4>[ 9009.176537]  ? __pfx_down_write_nested+0x10/0x10
<4>[ 9009.176542]  ? xfs_reclaim_inode+0x39f/0x590 [xfs]
<4>[ 9009.176869]  ? rcu_is_watching+0x1c/0x50
<4>[ 9009.176874]  xfs_reclaim_inode+0x3ac/0x590 [xfs]
<4>[ 9009.177192]  xfs_icwalk_ag+0x3c1/0x850 [xfs]
<4>[ 9009.177499]  ? xfs_icwalk_ag+0xe4/0x850 [xfs]
<4>[ 9009.177940]  ? __pfx_xfs_icwalk_ag+0x10/0x10 [xfs]
<4>[ 9009.178339]  ? find_held_lock+0x8c/0xa0
<4>[ 9009.178343]  ? kvm_sched_clock_read+0xd/0x20
<4>[ 9009.178347]  ? local_clock_noinstr+0x9/0xb0
<4>[ 9009.178355]  ? xfs_icwalk+0x69/0xb0 [xfs]
<4>[ 9009.178659]  ? xfs_perag_grab_tag+0x154/0x360 [xfs]
<4>[ 9009.178956]  ? __pfx_xfs_perag_grab_tag+0x10/0x10 [xfs]
<4>[ 9009.179256]  ? __pfx_try_to_wake_up+0x10/0x10
<4>[ 9009.179263]  xfs_icwalk+0x38/0xb0 [xfs]
<4>[ 9009.179572]  xfs_reclaim_inodes_nr+0x107/0x170 [xfs]
<4>[ 9009.179891]  ? __pfx_xfs_reclaim_inodes_nr+0x10/0x10 [xfs]
<4>[ 9009.180203]  super_cache_scan+0x248/0x290
<4>[ 9009.180212]  do_shrink_slab+0x239/0x8d0
<4>[ 9009.180219]  shrink_slab_memcg+0x39f/0x780
<4>[ 9009.180223]  ? shrink_slab_memcg+0x20b/0x780
<4>[ 9009.180229]  ? __pfx_shrink_slab_memcg+0x10/0x10
<4>[ 9009.180234]  ? __pfx_shrink_lruvec+0x10/0x10
<4>[ 9009.180238]  ? __lock_release+0x14c/0x530
<4>[ 9009.180243]  shrink_slab+0x3a7/0x460
<4>[ 9009.180247]  ? __pfx_shrink_slab+0x10/0x10
<4>[ 9009.180253]  ? mem_cgroup_calculate_protection+0x15c/0x2a0
<4>[ 9009.180259]  shrink_node_memcgs+0x26c/0x390
<4>[ 9009.180265]  shrink_node+0x128/0x720
<4>[ 9009.180268]  ? __zone_watermark_ok+0xc0/0x2a0
<4>[ 9009.180275]  balance_pgdat+0x53c/0xad0
<4>[ 9009.180280]  ? __pfx_balance_pgdat+0x10/0x10
<4>[ 9009.180283]  ? finish_wait+0x20/0x120
<4>[ 9009.180289]  ? mark_held_locks+0x23/0x90
<4>[ 9009.180294]  ? __pfx_kswapd_try_to_sleep+0x10/0x10
<4>[ 9009.180300]  ? __pfx___lock_acquired+0x10/0x10
<4>[ 9009.180305]  ? __might_resched+0x210/0x2f0
<4>[ 9009.180311]  kswapd+0x235/0x370
<4>[ 9009.180315]  ? __pfx_kswapd+0x10/0x10
<4>[ 9009.180318]  ? lockdep_hardirqs_on_prepare.part.0+0xe9/0x1b0
<4>[ 9009.180323]  ? __kthread_parkme+0xd3/0xf0
<4>[ 9009.180328]  ? __pfx_kswapd+0x10/0x10
<4>[ 9009.180331]  kthread+0x1a4/0x1e0
<4>[ 9009.180335]  ? kthread+0xf7/0x1e0
<4>[ 9009.180338]  ? __pfx_kthread+0x10/0x10
<4>[ 9009.180341]  ret_from_fork+0x30/0x50
<4>[ 9009.180346]  ? __pfx_kthread+0x10/0x10
<4>[ 9009.180349]  ret_from_fork_asm+0x1a/0x30
<4>[ 9009.180356]  </TASK>
<4>[76541.325574] cc1plus invoked oom-killer: gfp_mask=0x140dca(GFP_HIGHUSER_MOVABLE|__GFP_COMP|__GFP_ZERO), order=0, oom_score_adj=0
<4>[76541.325590] CPU: 0 PID: 33609 Comm: cc1plus Kdump: loaded Not tainted 6.9.0-rc3+ #49
<4>[76541.325596] Hardware name: innotek GmbH VirtualBox/VirtualBox, BIOS VirtualBox 12/01/2006
<4>[76541.325599] Call Trace:
<4>[76541.325602]  <TASK>
<4>[76541.325605]  dump_stack_lvl+0xac/0xc0
<4>[76541.325613]  dump_header+0x86/0x460
<4>[76541.325621]  oom_kill_process+0x15f/0x370
<4>[76541.325626]  out_of_memory+0x18e/0x590
<4>[76541.325631]  ? __pfx_out_of_memory+0x10/0x10
<4>[76541.325638] __alloc_pages_slowpath.constprop.0+0xadc/0xea0
<4>[76541.325647]  ? __pfx___alloc_pages_slowpath.constprop.0+0x10/0x10
<4>[76541.325651]  ? __zone_watermark_ok+0xc0/0x2a0
<4>[76541.325657]  ? get_page_from_freelist+0x383/0xa60
<4>[76541.325665]  __alloc_pages+0x3ad/0x3f0
<4>[76541.325669]  ? __pfx___alloc_pages+0x10/0x10
<4>[76541.325674]  ? rcu_is_watching+0x1c/0x50
<4>[76541.325681]  ? lock_acquire+0xc7/0x400
<4>[76541.325686]  ? _find_first_bit+0x28/0x50
<4>[76541.325692]  alloc_pages_mpol+0x175/0x340
<4>[76541.325698]  ? __pfx_alloc_pages_mpol+0x10/0x10
<4>[76541.325703]  ? __pfx_lock_acquire+0x10/0x10
<4>[76541.325706]  ? do_raw_spin_trylock+0xcf/0x120
<4>[76541.325712]  ? rcu_is_watching+0x1c/0x50
<4>[76541.325717]  vma_alloc_folio+0x95/0x100
<4>[76541.325721]  ? __pfx_vma_alloc_folio+0x10/0x10
<4>[76541.325726]  ? __pte_offset_map+0x43/0x290
<4>[76541.325730]  ? __pte_offset_map+0x154/0x290
<4>[76541.325735]  alloc_anon_folio+0x260/0x460
<4>[76541.325742]  do_anonymous_page+0x96/0x780
<4>[76541.325746]  ? lock_release+0x4d/0x1c0
<4>[76541.325751]  ? do_pte_missing+0x2c/0x570
<4>[76541.325757]  ? handle_pte_fault+0x23c/0x2e0
<4>[76541.325762]  __handle_mm_fault+0x4ce/0xa00
<4>[76541.325768]  ? __pfx___handle_mm_fault+0x10/0x10
<4>[76541.325776]  ? __pfx_lock_vma_under_rcu+0x10/0x10
<4>[76541.325782]  handle_mm_fault+0x11b/0x3e0
<4>[76541.325788]  do_user_addr_fault+0x273/0x960
<4>[76541.325795]  exc_page_fault+0x59/0xd0
<4>[76541.325800]  asm_exc_page_fault+0x22/0x30
<4>[76541.325805] RIP: 0033:0x563beed22d16
<4>[76541.325810] Code: 00 00 00 e8 cc 6b ee ff b9 01 00 00 00 31 f6 31 d2 bf 28 00 00 00 49 89 c6 e8 66 87 ed ff 66 0f ef c0 ba 02 00 00 00 48 89 ef <0f> 11 00 48 89 c3 66 89 10 4c 89 78 10 48 c7 40 18 00 00 00 00 4c
<4>[76541.325814] RSP: 002b:00007ffddd3aa8a0 EFLAGS: 00010202
<4>[76541.325819] RAX: 00007f7541549000 RBX: 00007f7541546888 RCX: 0000563c21ca6e50
<4>[76541.325822] RDX: 0000000000000002 RSI: 0000000000000006 RDI: 00007f7541546930
<4>[76541.325825] RBP: 00007f7541546930 R08: 00007f7541549000 R09: 0000563c21ca6e50
<4>[76541.325828] R10: 0000000000000001 R11: 0000000000000001 R12: 00007f7546a2e690
<4>[76541.325831] R13: 00007f75414bf000 R14: 00007f7541524c78 R15: 00007f7541544fa0
<4>[76541.325838]  </TASK>
<4>[76541.325844] Mem-Info:
<4>[76541.325848] active_anon:504 inactive_anon:4920752 isolated_anon:0
<4>[76541.325848]  active_file:1300 inactive_file:1167 isolated_file:0
<4>[76541.325848]  unevictable:0 dirty:82 writeback:0
<4>[76541.325848]  slab_reclaimable:59582 slab_unreclaimable:373508
<4>[76541.325848]  mapped:4582 shmem:5358 pagetables:13948
<4>[76541.325848]  sec_pagetables:0 bounce:0
<4>[76541.325848]  kernel_misc_reclaimable:0
<4>[76541.325848]  free:68497 free_pcp:152 free_cma:0
<4>[76541.325855] Node 0 active_anon:2016kB inactive_anon:19683008kB active_file:5200kB inactive_file:4668kB unevictable:0kB isolated(anon):0kB isolated(file):0kB mapped:18328kB dirty:328kB writeback:0kB shmem:21432kB shmem_thp:0kB shmem_pmdmapped:0kB anon_thp:7714816kB writeback_tmp:0kB kernel_stack:20352kB pagetables:55792kB sec_pagetables:0kB all_unreclaimable? no
<4>[76541.325863] Node 0 DMA free:11264kB boost:0kB min:44kB low:56kB high:68kB reserved_highatomic:0KB active_anon:0kB inactive_anon:0kB active_file:0kB inactive_file:0kB unevictable:0kB writepending:0kB present:15992kB managed:15360kB mlocked:0kB bounce:0kB free_pcp:0kB local_pcp:0kB free_cma:0kB
<4>[76541.325872] lowmem_reserve[]: 0 3113 21217 21217 21217
<4>[76541.325881] Node 0 DMA32 free:82792kB boost:0kB min:9908kB low:13096kB high:16284kB reserved_highatomic:0KB active_anon:0kB inactive_anon:3182860kB active_file:484kB inactive_file:308kB unevictable:0kB writepending:136kB present:3653568kB managed:3325888kB mlocked:0kB bounce:0kB free_pcp:512kB local_pcp:0kB free_cma:0kB
<4>[76541.325889] lowmem_reserve[]: 0 0 18103 18103 18103
<4>[76541.325902] Node 0 Normal free:179932kB boost:142044kB min:199668kB low:218204kB high:236740kB reserved_highatomic:0KB active_anon:2016kB inactive_anon:16499404kB active_file:4952kB inactive_file:4408kB unevictable:0kB writepending:272kB present:22405120kB managed:18550444kB mlocked:0kB bounce:0kB free_pcp:164kB local_pcp:0kB free_cma:0kB
<4>[76541.325910] lowmem_reserve[]: 0 0 0 0 0
<4>[76541.325918] Node 0 DMA: 0*4kB 0*8kB 0*16kB 0*32kB 0*64kB 0*128kB 0*256kB 0*512kB 1*1024kB (U) 1*2048kB (M) 2*4096kB (M) = 11264kB
<4>[76541.325942] Node 0 DMA32: 147*4kB (UM) 866*8kB (UM) 724*16kB (UME) 547*32kB (UME) 318*64kB (UME) 73*128kB (UE) 4*256kB (UM) 5*512kB (M) 12*1024kB (M) 0*2048kB 0*4096kB = 82172kB
<4>[76541.325972] Node 0 Normal: 286*4kB (UME) 3098*8kB (UME) 7684*16kB (UME) 964*32kB (UME) 5*64kB (M) 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 180040kB
<4>[76541.325998] Node 0 hugepages_total=0 hugepages_free=0 hugepages_surp=0 hugepages_size=2048kB
<4>[76541.326001] 7807 total pagecache pages
<4>[76541.326004] 0 pages in swap cache
<4>[76541.326006] Free swap  = 0kB
<4>[76541.326008] Total swap = 0kB
<4>[76541.326011] 6518670 pages RAM
<4>[76541.326013] 0 pages HighMem/MovableOnly
<4>[76541.326015] 1045747 pages reserved
<4>[76541.326017] 0 pages cma reserved
<4>[76541.326020] 0 pages hwpoisoned
<6>[76541.326022] Tasks state (memory values in pages):
<6>[76541.326025] [  pid  ]   uid  tgid total_vm      rss rss_anon rss_file rss_shmem pgtables_bytes swapents oom_score_adj name
<6>[76541.326043] [    751]     0   751     6809 1042      256      178       608    90112        0          -250 systemd-journal
<6>[76541.326050] [    765]     0   765     9020 1150      975      175         0    94208        0         -1000 systemd-udevd
<6>[76541.326058] [    899]     0   899    22958 331      160      171         0    77824        0         -1000 auditd
<6>[76541.326063] [    901]     0   901     1947 270       64      206         0    61440        0             0 sedispatch
<6>[76541.326069] [    922]    81   922     2738 339      210      129         0    73728        0          -900 dbus-broker-lau
<6>[76541.326074] [    923]    81   923     2427 689      565      124         0    57344        0          -900 dbus-broker
<6>[76541.326079] [    924]     0   924   118878 932      736      196         0   172032        0             0 NetworkManager
<6>[76541.326084] [    925]    70   925     3984 396      224      172         0    69632        0             0 avahi-daemon
<6>[76541.326094] [    928]     0   928    19781 241       64      177         0    65536        0             0 irqbalance
<6>[76541.326099] [    929]   993   929      676 99       32       67         0    53248        0             0 lsmd
<6>[76541.326105] [    930]     0   930      833 213       32      181         0    53248        0             0 mcelog
<6>[76541.326109] [    933]   998   933   696894 1606     1426      180         0   253952        0             0 polkitd
<6>[76541.326115] [    934]     0   934   112996 321      128      193         0   131072        0             0 power-profiles-
<6>[76541.326120] [    937]   172   937    38527 243       64      179         0    69632        0             0 rtkit-daemon
<6>[76541.326126] [    939]   982   939    21215 289      140      149         0    73728        0             0 chronyd
<6>[76541.326131] [    940]     0   940   131390 284      160      124         0   139264        0             0 accounts-daemon
<6>[76541.326136] [    942]     0   942   112043 287       96      191         0   122880        0             0 switcheroo-cont
<6>[76541.326141] [    949]     0   949     7297 767      565      202         0    98304        0             0 systemd-logind
<6>[76541.326146] [    952]     0   952    98804 708      480      228         0   147456        0             0 udisksd
<6>[76541.326151] [    953]     0   953   112202 393      192      201         0   126976        0             0 upowerd
<6>[76541.326156] [    957]    70   957     3916 367      205      162         0    65536        0             0 avahi-daemon
<6>[76541.326162] [    972]     0   972     1181 276       64      212         0    49152        0             0 alsactl
<6>[76541.326167] [    982]     0   982    97866 535      416      119         0   151552        0             0 ModemManager
<6>[76541.326172] [   1011]     0  1011    61956 1198      992      206         0   122880        0             0 cupsd
<6>[76541.326177] [   1015]     0  1015     3950 521      288      233         0    69632        0         -1000 sshd
<6>[76541.326182] [   1033]     0  1033     1168 234       32      202         0    45056        0             0 atd
<6>[76541.326188] [   1036]     0  1036    55980 335      192      143         0    69632        0             0 crond
<6>[76541.326193] [   1040]     0  1040   113278 437      224      213         0   135168        0             0 gdm
<6>[76541.326198] [   1093]     0  1093    41209 446      160      121       165   106496        0             0 rsyslogd
<6>[76541.326203] [   1579]     0  1579     3822 370      192      178         0    69632        0             0 wpa_supplicant
<6>[76541.326208] [   1805]   989  1805   132604 664      512      152         0   151552        0             0 colord
<6>[76541.326213] [   2010]     0  2010    96963 673      416      257         0   131072        0             0 gdm-session-wor
<6>[76541.326219] [   2028]  1000  2028     5877 1052      892      160         0    86016        0           100 systemd
<6>[76541.326227] [   2032]  1000  2032     6799 1674     1536      138         0    90112        0           100 (sd-pam)
<6>[76541.326233] [   2057]  1000  2057   113231 414      199      215         0   135168        0             0 gnome-keyring-d
<6>[76541.326242] [   2063]  1000  2063    93539 337       96      241         0    98304        0             0 gdm-wayland-ses
<6>[76541.326246] [   2067]  1000  2067     2665 262       96      166         0    73728        0           200 dbus-broker-lau
<6>[76541.326253] [   2069]  1000  2069     1667 462      320      142         0    49152        0           200 dbus-broker
<6>[76541.326258] [   2072]  1000  2072   128284 649      480      169         0   196608        0             0 gnome-session-b
<6>[76541.326263] [   2117]  1000  2117    75949 289       96      193         0   110592        0           200 gnome-session-c
<6>[76541.326267] [   2119]  1000  2119   165282 760      576      184         0   229376        0           200 gnome-session-b
<6>[76541.326273] [   2135]  1000  2135  1732662    41332 40744      172       416  1613824        0           200 gnome-shell
<6>[76541.326278] [   2178]  1000  2178   113144 359      160      199         0   131072        0           200 gvfsd
<6>[76541.326283] [   2185]  1000  2185    94983 357      160      197         0   110592        0           200 gvfsd-fuse
<6>[76541.326288] [   2196]  1000  2196    77144 339      160      179         0   118784        0           200 at-spi-bus-laun
<6>[76541.326293] [   2202]  1000  2202     2633 277       64      213         0    65536        0           200 dbus-broker-lau
<6>[76541.326298] [   2204]  1000  2204     1247 236       64      172         0    53248        0           200 dbus-broker
<6>[76541.326303] [   2215]  1000  2215   131709 1393     1169      224         0   147456        0           200 ibus-daemon
<6>[76541.326308] [   2219]  1000  2219   112273 318      160      158         0   118784        0           200 ibus-dconf
<6>[76541.326314] [   2220]  1000  2220   148953 2758     2505      157        96   237568        0           200 ibus-extension-
<6>[76541.326319] [   2224]  1000  2224   146388 883      704      179         0   221184        0           200 gnome-shell-cal
<6>[76541.326324] [   2226]  1000  2226   112025 329       96      233         0   126976        0           200 xdg-permission-
<6>[76541.326329] [   2230]  1000  2230   112261 323      128      195         0   122880        0           200 ibus-portal
<6>[76541.326334] [   2235]  1000  2235    81642 1710     1454      256         0   118784        0           200 pipewire
<6>[76541.326339] [   2240]  1000  2240   136741 1446     1248      198         0   172032        0           200 wireplumber
<6>[76541.326345] [   2246]  1000  2246    79192 601      448      153         0   114688        0           200 pipewire-pulse
<6>[76541.326350] [   2252]  1000  2252   321629 1223     1056      167         0   311296        0           200 evolution-sourc
<6>[76541.326355] [   2267]  1000  2267   204877 864      672      192         0   266240        0           200 goa-daemon
<6>[76541.326363] [   2272]  1000  2272   262624 1350     1120      230         0   323584        0           200 evolution-calen
<6>[76541.326368] [   2274]  1000  2274   131890 574      352      222         0   147456        0           200 gvfs-udisks2-vo
<6>[76541.326374] [   2288]  1000  2288   132884 458      320      138         0   155648        0           200 goa-identity-se
<6>[76541.326379] [   2291]  1000  2291   112063 270      128      142         0   118784        0           200 gvfs-mtp-volume
<6>[76541.326384] [   2305]  1000  2305   112520 363      160      203         0   126976        0           200 gvfs-gphoto2-vo
<6>[76541.326388] [   2310]  1000  2310    39001 216       96      120         0    73728        0           200 dconf-service
<6>[76541.326393] [   2314]  1000  2314   238450 1084      864      220         0   299008        0           200 evolution-addre
<6>[76541.326399] [   2318]  1000  2318   112074 317      128      189         0   122880        0           200 gvfs-goa-volume
<6>[76541.326404] [   2325]     0  2325    60133 486      320      166         0   114688        0             0 sssd_kcm
<6>[76541.326410] [   2344]  1000  2344   784123 1523     1300      223         0   294912        0           200 gjs
<6>[76541.326415] [   2345]  1000  2345    40430 358      160      198         0    90112        0           200 at-spi2-registr
<6>[76541.326420] [   2355]  1000  2355   130534 324      128      196         0   139264        0           200 gsd-a11y-settin
<6>[76541.326425] [   2369]  1000  2369   184989 1639     1344      169       126   249856        0           200 gsd-color
<6>[76541.326430] [   2375]  1000  2375   148593 712      576      136         0   233472        0           200 gsd-datetime
<6>[76541.326435] [   2378]  1000  2378    93814 353      128      225         0   110592        0           200 ibus-engine-sim
<6>[76541.326440] [   2387]  1000  2387   131126 399      192      207         0   135168        0           200 gsd-housekeepin
<6>[76541.326446] [   2401]  1000  2401   147567 1611     1280      237        94   217088        0           200 gsd-keyboard
<6>[76541.326451] [   2418]  1000  2418   237038 1706     1344      271        91   286720        0           200 gsd-media-keys
<6>[76541.326455] [   2421]  1000  2421   184825 1497     1280      129        88   245760        0           200 gsd-power
<6>[76541.326460] [   2424]  1000  2424   116824 622      416      206         0   159744        0           200 gsd-print-notif
<6>[76541.326465] [   2430]  1000  2430   167414 339      128      211         0   159744        0           200 gsd-rfkill
<6>[76541.326470] [   2431]  1000  2431   112014 320      128      192         0   118784        0           200 gsd-screensaver
<6>[76541.326485] [   2434]  1000  2434   169561 552      352      200         0   180224        0           200 gsd-sharing
<6>[76541.326490] [   2441]  1000  2441   150846 572      352      220         0   147456        0           200 gsd-smartcard
<6>[76541.326495] [   2443]  1000  2443   132668 441      224      217         0   159744        0           200 gsd-sound
<6>[76541.326499] [   2450]  1000  2450   149435 300      160      140         0   155648        0           200 gsd-usb-protect
<6>[76541.326504] [   2464]  1000  2464   147723 1639     1312      230        97   212992        0           200 gsd-wacom
<6>[76541.326509] [   2472]  1000  2472    58012 407      224      183         0    98304        0           200 gsd-disk-utilit
<6>[76541.326528] [   2479]  1000  2479   259040    11962 11791      171         0   446464        0           200 gnome-software
<6>[76541.326533] [   2487]  1000  2487   228900 2389     2144      148        97   335872        0           200 evolution-alarm
<6>[76541.326538] [   2528]  1000  2528   784122 1547     1344      203         0   294912        0           200 gjs
<6>[76541.326542] [   2536]  1000  2536   172243 615      416      199         0   176128        0           200 xdg-desktop-por
<6>[76541.326547] [   2549]  1000  2549   149605 870      640      230         0   233472        0           200 gsd-printer
<6>[76541.326551] [   2568]  1000  2568   167492 262       96      166         0   172032        0           200 xdg-document-po
<6>[76541.326560] [   2574]  1000  2574      663 172        0      172         0    49152        0           200 fusermount
<6>[76541.326565] [   2597]  1000  2597   223864 1479     1280      199         0   290816        0           200 xdg-desktop-por
<6>[76541.326570] [   2690]  1000  2690   147896 1725     1408      220        97   225280        0           200 xdg-desktop-por
<6>[76541.326575] [   2712]     0  2712   141140 1275     1056      219         0   212992        0             0 fwupd
<6>[76541.326579] [   2724]  1000  2724    93694 206       64      142         0   110592        0           200 gvfsd-metadata
<6>[76541.326584] [   2778]  1000  2778   190541 3240     2560      231       449   294912        0           200 gnome-terminal-
<6>[76541.326592] [   2796]  1000  2796    56028 663      448      215         0    73728        0           200 bash
<6>[76541.326597] [   2835]     0  2835     4859 600      416      184         0    81920        0             0 sshd
<6>[76541.326601] [   2840]  1000  2840     4903 612      487      125         0    81920        0             0 sshd
<6>[76541.326606] [   2843]  1000  2843    56056 618      480      138         0    77824        0             0 bash
<6>[76541.326611] [   3868]     0  3868     4859 634      448      186         0    81920        0             0 sshd
<6>[76541.326615] [   3873]  1000  3873     4903 681      461      220         0    81920        0             0 sshd
<6>[76541.326620] [   3876]  1000  3876    56056 615      480      135         0    77824        0             0 bash
<6>[76541.326630] [  25652]  1000 25652    57500 287      128      159         0    90112        0             0 git
<6>[76541.326636] [  26274]  1000 26274    55461 335      192      143         0    69632        0             0 make
<6>[76541.326640] [  26463]  1000 26463    56106 879      768      111         0    81920        0             0 make
<6>[76541.326646] [  32166]  1000 32166    67643    12357 12215      142         0   176128        0             0 make
<6>[76541.326650] [  32585]  1000 32585    61726 6459     6317      142         0   126976        0             0 make
<6>[76541.326659] [  33364]  1000 33364    55748 240       96      144         0    73728        0             0 g++
<6>[76541.326663] [  33365]  1000 33365  2781960  2651315 2650090     1225         0 21700608        0             0 cc1plus
<6>[76541.326668] [  33587]  1000 33587    55748 252       96      156         0    73728        0             0 g++
<6>[76541.326673] [  33588]  1000 33588   597044   510453 509837      616         0  4243456        0             0 cc1plus
<6>[76541.326678] [  33592]  1000 33592    55748 252       96      156         0    69632        0             0 g++
<6>[76541.326682] [  33593]  1000 33593   532253   445795 444214     1581         0  3723264        0             0 cc1plus
<6>[76541.326687] [  33596]  1000 33596    55748 247       96      151         0    69632        0             0 g++
<6>[76541.326694] [  33597]  1000 33597   505784   435412 435159      253         0  3641344        0             0 cc1plus
<6>[76541.326699] [  33600]  1000 33600    55748 273       96      177         0    65536        0             0 g++
<6>[76541.326704] [  33601]  1000 33601   456709   387653 386025     1628         0  3252224        0             0 cc1plus
<6>[76541.326708] [  33604]  1000 33604    55748 269       96      173         0    73728        0             0 g++
<6>[76541.326713] [  33605]  1000 33605   286890   218179 216781     1398         0  1904640        0             0 cc1plus
<6>[76541.326717] [  33608]  1000 33608    55748 220       96      124         0    77824        0             0 g++
<6>[76541.326725] [  33609]  1000 33609   217521   141052 139828     1224         0  1290240        0             0 cc1plus
<6>[76541.326729] oom-kill:constraint=CONSTRAINT_NONE,nodemask=(null),cpuset=/,mems_allowed=0,global_oom,task_memcg=/user.slice/user-1000.slice/session-4.scope,task=cc1plus,pid=33365,uid=1000
<3>[76541.326764] Out of memory: Killed process 33365 (cc1plus) total-vm:11127840kB, anon-rss:10600360kB, file-rss:4900kB, shmem-rss:0kB, UID:1000 pgtables:21192kB oom_score_adj:0

Thanks

- Xiubo

#1 lore raw
Quoted 34 lines · Xiubo Li

Hi all

BTW, is this a known issue and has it been fixed already ? I can reproduce this always with my VMs:

<4>[ 9009.171195]
<4>[ 9009.171205] ======================================================
<4>[ 9009.171208] WARNING: possible circular locking dependency detected
<4>[ 9009.171211] 6.9.0-rc3+ #49 Not tainted
<4>[ 9009.171214] ------------------------------------------------------
<4>[ 9009.171216] kswapd0/149 is trying to acquire lock:
<4>[ 9009.171219] ffff88811346a920 (&xfs_nondir_ilock_class){++++}-{4:4},
at: xfs_reclaim_inode+0x3ac/0x590 [xfs]
<4>[ 9009.171580]
<4>[ 9009.171580] but task is already holding lock:
<4>[ 9009.171583] ffffffff8bb33100 (fs_reclaim){+.+.}-{0:0}, at:
balance_pgdat+0x5d9/0xad0
<4>[ 9009.171593]
<4>[ 9009.171593] which lock already depends on the new lock.
<4>[ 9009.171593]
<4>[ 9009.171595]
<4>[ 9009.171595] the existing dependency chain (in reverse order) is:
<4>[ 9009.171597]
<4>[ 9009.171597] -> #1 (fs_reclaim){+.+.}-{0:0}:
<4>[ 9009.171603]        __lock_acquire+0x7da/0x1030
<4>[ 9009.171610]        lock_acquire+0x15d/0x400
<4>[ 9009.171614]        fs_reclaim_acquire+0xb5/0x100
<4>[ 9009.171618] prepare_alloc_pages.constprop.0+0xc5/0x230
<4>[ 9009.171622]        __alloc_pages+0x12a/0x3f0
<4>[ 9009.171625]        alloc_pages_mpol+0x175/0x340
<4>[ 9009.171630]        stack_depot_save_flags+0x4c5/0x510
<4>[ 9009.171635]        kasan_save_stack+0x30/0x40
<4>[ 9009.171640]        kasan_save_track+0x10/0x30
Quoted 8 lines
<4>[ 9009.171643]        __kasan_slab_alloc+0x83/0x90
<4>[ 9009.171646]        kmem_cache_alloc+0x15e/0x4a0
<4>[ 9009.171652]        __alloc_object+0x35/0x370
<4>[ 9009.171659]        __create_object+0x22/0x90
<4>[ 9009.171665] __kmalloc_node_track_caller+0x477/0x5b0
<4>[ 9009.171672]        krealloc+0x5f/0x110
<4>[ 9009.171679]        xfs_iext_insert_raw+0x4b2/0x6e0 [xfs]
<4>[ 9009.172172]        xfs_iext_insert+0x2e/0x130 [xfs]

The only krealloc() in this path is:

	new = krealloc(ifp->if_data, new_size,
                        GFP_KERNEL | __GFP_NOLOCKDEP | __GFP_NOFAIL);

And it explicitly uses __GFP_NOLOCKDEP to tell lockdep not to warn about this allocation because of this false positive situation.

Oh. I've seen this before. This is a KASAN bug, and I'm pretty sure I've posted a patch to fix it a fair while back that nobody seemed to care about enough to review or merge it.

That is: kasan_save_stack() is doing a fixed GFP_KERNEL allocation in an context where GFP_KERNEL allocations are known to generate lockdep false positives. This occurs depsite the XFS and general memory allocation code doing exactly the right thing to avoid the lockdep false positives (i.e. using and obeying __GFP_NOLOCKDEP).

The kasan code ends up in stack_depot_save_flags(), which does a GFP_KERNEL allocation but filters out __GFP_NOLOCKDEP and does not add it back. Hence kasan generates the false positive lockdep warnings, not the code doing the original allocation.

kasan and/or stack_depot_save_flags() needs fixing here.

-Dave.

Signature · 2 lines
Dave Chinner
[email protected]
author
#2 lore raw
Quoted 61 lines · Dave Chinner
Quoted 34 lines · Xiubo Li

Hi all

BTW, is this a known issue and has it been fixed already ? I can reproduce this always with my VMs:

<4>[ 9009.171195]
<4>[ 9009.171205] ======================================================
<4>[ 9009.171208] WARNING: possible circular locking dependency detected
<4>[ 9009.171211] 6.9.0-rc3+ #49 Not tainted
<4>[ 9009.171214] ------------------------------------------------------
<4>[ 9009.171216] kswapd0/149 is trying to acquire lock:
<4>[ 9009.171219] ffff88811346a920 (&xfs_nondir_ilock_class){++++}-{4:4},
at: xfs_reclaim_inode+0x3ac/0x590 [xfs]
<4>[ 9009.171580]
<4>[ 9009.171580] but task is already holding lock:
<4>[ 9009.171583] ffffffff8bb33100 (fs_reclaim){+.+.}-{0:0}, at:
balance_pgdat+0x5d9/0xad0
<4>[ 9009.171593]
<4>[ 9009.171593] which lock already depends on the new lock.
<4>[ 9009.171593]
<4>[ 9009.171595]
<4>[ 9009.171595] the existing dependency chain (in reverse order) is:
<4>[ 9009.171597]
<4>[ 9009.171597] -> #1 (fs_reclaim){+.+.}-{0:0}:
<4>[ 9009.171603]        __lock_acquire+0x7da/0x1030
<4>[ 9009.171610]        lock_acquire+0x15d/0x400
<4>[ 9009.171614]        fs_reclaim_acquire+0xb5/0x100
<4>[ 9009.171618] prepare_alloc_pages.constprop.0+0xc5/0x230
<4>[ 9009.171622]        __alloc_pages+0x12a/0x3f0
<4>[ 9009.171625]        alloc_pages_mpol+0x175/0x340
<4>[ 9009.171630]        stack_depot_save_flags+0x4c5/0x510
<4>[ 9009.171635]        kasan_save_stack+0x30/0x40
<4>[ 9009.171640]        kasan_save_track+0x10/0x30
Quoted 8 lines
<4>[ 9009.171643]        __kasan_slab_alloc+0x83/0x90
<4>[ 9009.171646]        kmem_cache_alloc+0x15e/0x4a0
<4>[ 9009.171652]        __alloc_object+0x35/0x370
<4>[ 9009.171659]        __create_object+0x22/0x90
<4>[ 9009.171665] __kmalloc_node_track_caller+0x477/0x5b0
<4>[ 9009.171672]        krealloc+0x5f/0x110
<4>[ 9009.171679]        xfs_iext_insert_raw+0x4b2/0x6e0 [xfs]
<4>[ 9009.172172]        xfs_iext_insert+0x2e/0x130 [xfs]

The only krealloc() in this path is:

	new = krealloc(ifp->if_data, new_size,
                         GFP_KERNEL | __GFP_NOLOCKDEP | __GFP_NOFAIL);

And it explicitly uses __GFP_NOLOCKDEP to tell lockdep not to warn about this allocation because of this false positive situation.

Oh. I've seen this before. This is a KASAN bug, and I'm pretty sure I've posted a patch to fix it a fair while back that nobody seemed to care about enough to review or merge it.

That is: kasan_save_stack() is doing a fixed GFP_KERNEL allocation in an context where GFP_KERNEL allocations are known to generate lockdep false positives. This occurs depsite the XFS and general memory allocation code doing exactly the right thing to avoid the lockdep false positives (i.e. using and obeying __GFP_NOLOCKDEP).

Quoted 6 lines

The kasan code ends up in stack_depot_save_flags(), which does a GFP_KERNEL allocation but filters out __GFP_NOLOCKDEP and does not add it back. Hence kasan generates the false positive lockdep warnings, not the code doing the original allocation.

kasan and/or stack_depot_save_flags() needs fixing here.

Hi Dave,

Thanks very much for your feedback.

BTW, do you have the link of your patch ? I can help test it.

Thanks

- Xiubo

Quoted 1 line

-Dave.

#3 lore raw
In reply to Dave Chinner

Adding the KASAN maintainer so that we actuall have a chane of fixing this instead of a rant that just gets lost on the xfs list..

Quoted 18 lines · Dave Chinner

The only krealloc() in this path is:

	new = krealloc(ifp->if_data, new_size,
                        GFP_KERNEL | __GFP_NOLOCKDEP | __GFP_NOFAIL);

And it explicitly uses __GFP_NOLOCKDEP to tell lockdep not to warn about this allocation because of this false positive situation.

Oh. I've seen this before. This is a KASAN bug, and I'm pretty sure I've posted a patch to fix it a fair while back that nobody seemed to care about enough to review or merge it.

That is: kasan_save_stack() is doing a fixed GFP_KERNEL allocation in an context where GFP_KERNEL allocations are known to generate lockdep false positives. This occurs depsite the XFS and general memory allocation code doing exactly the right thing to avoid the lockdep false positives (i.e. using and obeying __GFP_NOLOCKDEP).

Quoted 12 lines

The kasan code ends up in stack_depot_save_flags(), which does a GFP_KERNEL allocation but filters out __GFP_NOLOCKDEP and does not add it back. Hence kasan generates the false positive lockdep warnings, not the code doing the original allocation.

kasan and/or stack_depot_save_flags() needs fixing here.

-Dave.

Signature · 2 lines
Dave Chinner
[email protected]

---end quoted text---

#4 lore raw
Quoted 4 lines · Christoph Hellwig

Adding the KASAN maintainer so that we actuall have a chane of fixing this instead of a rant that just gets lost on the xfs list..

Thanks.

Quoted 6 lines · Dave Chinner
Quoted 5 lines

The only krealloc() in this path is:

      new = krealloc(ifp->if_data, new_size,
                        GFP_KERNEL | __GFP_NOLOCKDEP | __GFP_NOFAIL);
Quoted 7 lines
Quoted 7 lines

And it explicitly uses __GFP_NOLOCKDEP to tell lockdep not to warn about this allocation because of this false positive situation.

Oh. I've seen this before. This is a KASAN bug, and I'm pretty sure I've posted a patch to fix it a fair while back that nobody seemed to care about enough to review or merge it.

Sorry, must have been my bad. I didn't find the actual patch though, only proposed way to fix this bug:
https://lkml.kernel.org/r/%[email protected]%3E
So I'll cook patch and will send it shortly.

#5 lore raw

PATCH 1/1 stackdepot: respect __GFP_NOLOCKDEP allocation flag

In reply to Xiubo Li

If stack_depot_save_flags() allocates memory it always drops __GFP_NOLOCKDEP flag. So when KASAN tries to track __GFP_NOLOCKDEP allocation we may end up with lockdep splat like bellow:

======================================================
WARNING: possible circular locking dependency detected 6.9.0-rc3+ #49 Not tainted
------------------------------------------------------
kswapd0/149 is trying to acquire lock:
ffff88811346a920
(&xfs_nondir_ilock_class){++++}-{4:4}, at: xfs_reclaim_inode+0x3ac/0x590 [xfs]

but task is already holding lock:
ffffffff8bb33100 (fs_reclaim){+.+.}-{0:0}, at:
balance_pgdat+0x5d9/0xad0

which lock already depends on the new lock.

the existing dependency chain (in reverse order) is:
-> #1 (fs_reclaim){+.+.}-{0:0}:

        __lock_acquire+0x7da/0x1030
        lock_acquire+0x15d/0x400
        fs_reclaim_acquire+0xb5/0x100
 prepare_alloc_pages.constprop.0+0xc5/0x230
        __alloc_pages+0x12a/0x3f0
        alloc_pages_mpol+0x175/0x340
        stack_depot_save_flags+0x4c5/0x510
        kasan_save_stack+0x30/0x40
        kasan_save_track+0x10/0x30
        __kasan_slab_alloc+0x83/0x90
        kmem_cache_alloc+0x15e/0x4a0
        __alloc_object+0x35/0x370
        __create_object+0x22/0x90
 __kmalloc_node_track_caller+0x477/0x5b0
        krealloc+0x5f/0x110
        xfs_iext_insert_raw+0x4b2/0x6e0 [xfs]
        xfs_iext_insert+0x2e/0x130 [xfs]
        xfs_iread_bmbt_block+0x1a9/0x4d0 [xfs]
        xfs_btree_visit_block+0xfb/0x290 [xfs]
        xfs_btree_visit_blocks+0x215/0x2c0 [xfs]
        xfs_iread_extents+0x1a2/0x2e0 [xfs]
 xfs_buffered_write_iomap_begin+0x376/0x10a0 [xfs]
        iomap_iter+0x1d1/0x2d0
 iomap_file_buffered_write+0x120/0x1a0
        xfs_file_buffered_write+0x128/0x4b0 [xfs]
        vfs_write+0x675/0x890
        ksys_write+0xc3/0x160
        do_syscall_64+0x94/0x170
 entry_SYSCALL_64_after_hwframe+0x71/0x79

Always preserve __GFP_NOLOCKDEP to fix this.

Fixes cd11016e5f52 ("mm, kasan: stackdepot implementation. Enable stackdepot for SLAB")
Cc Christoph Hellwig <[email protected]>
---
 lib/stackdepot.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
lib/stackdepot.c +2 -2
index 68c97387aa54..cd8f23455285 100644
@@ -627,10 +627,10 @@ depot_stack_handle_t stack_depot_save_flags(unsigned long *entries,
 		/*
 		 * Zero out zone modifiers, as we don't have specific zone
 		 * requirements. Keep the flags related to allocation in atomic
-		 * contexts and I/O.
+		 * contexts, I/O, nolockdep.
 		 */
 		alloc_flags &= ~GFP_ZONEMASK;
-		alloc_flags &= (GFP_ATOMIC | GFP_KERNEL);
+		alloc_flags &= (GFP_ATOMIC | GFP_KERNEL | __GFP_NOLOCKDEP);
 		alloc_flags |= __GFP_NOWARN;
 		page = alloc_pages(alloc_flags, DEPOT_POOL_ORDER);
 		if (page)
Signature · 1 line
2.43.2
author
#6 lore raw

Re: [PATCH] stackdepot: respect __GFP_NOLOCKDEP allocation flag

Thanks Andrey,

I will test it soon.

- Xiubo

Quoted 82 lines · Andrey Ryabinin

If stack_depot_save_flags() allocates memory it always drops __GFP_NOLOCKDEP flag. So when KASAN tries to track __GFP_NOLOCKDEP allocation we may end up with lockdep splat like bellow:

======================================================
WARNING: possible circular locking dependency detected 6.9.0-rc3+ #49 Not tainted
------------------------------------------------------
kswapd0/149 is trying to acquire lock:
ffff88811346a920
(&xfs_nondir_ilock_class){++++}-{4:4}, at: xfs_reclaim_inode+0x3ac/0x590 [xfs]

but task is already holding lock:
ffffffff8bb33100 (fs_reclaim){+.+.}-{0:0}, at:
balance_pgdat+0x5d9/0xad0

which lock already depends on the new lock.

the existing dependency chain (in reverse order) is:
-> #1 (fs_reclaim){+.+.}-{0:0}:

         __lock_acquire+0x7da/0x1030
         lock_acquire+0x15d/0x400
         fs_reclaim_acquire+0xb5/0x100
  prepare_alloc_pages.constprop.0+0xc5/0x230
         __alloc_pages+0x12a/0x3f0
         alloc_pages_mpol+0x175/0x340
         stack_depot_save_flags+0x4c5/0x510
         kasan_save_stack+0x30/0x40
         kasan_save_track+0x10/0x30
         __kasan_slab_alloc+0x83/0x90
         kmem_cache_alloc+0x15e/0x4a0
         __alloc_object+0x35/0x370
         __create_object+0x22/0x90
  __kmalloc_node_track_caller+0x477/0x5b0
         krealloc+0x5f/0x110
         xfs_iext_insert_raw+0x4b2/0x6e0 [xfs]
         xfs_iext_insert+0x2e/0x130 [xfs]
         xfs_iread_bmbt_block+0x1a9/0x4d0 [xfs]
         xfs_btree_visit_block+0xfb/0x290 [xfs]
         xfs_btree_visit_blocks+0x215/0x2c0 [xfs]
         xfs_iread_extents+0x1a2/0x2e0 [xfs]
  xfs_buffered_write_iomap_begin+0x376/0x10a0 [xfs]
         iomap_iter+0x1d1/0x2d0
  iomap_file_buffered_write+0x120/0x1a0
         xfs_file_buffered_write+0x128/0x4b0 [xfs]
         vfs_write+0x675/0x890
         ksys_write+0xc3/0x160
         do_syscall_64+0x94/0x170
  entry_SYSCALL_64_after_hwframe+0x71/0x79

Always preserve __GFP_NOLOCKDEP to fix this.

Fixes cd11016e5f52 ("mm, kasan: stackdepot implementation. Enable stackdepot for SLAB")
Cc Christoph Hellwig <[email protected]>
---
  lib/stackdepot.c | 4 ++--
  1 file changed, 2 insertions(+), 2 deletions(-)
lib/stackdepot.c +2 -2
index 68c97387aa54..cd8f23455285 100644
@@ -627,10 +627,10 @@ depot_stack_handle_t stack_depot_save_flags(unsigned long *entries,
  		/*
  		 * Zero out zone modifiers, as we don't have specific zone
  		 * requirements. Keep the flags related to allocation in atomic
-		 * contexts and I/O.
+		 * contexts, I/O, nolockdep.
  		 */
  		alloc_flags &= ~GFP_ZONEMASK;
-		alloc_flags &= (GFP_ATOMIC | GFP_KERNEL);
+		alloc_flags &= (GFP_ATOMIC | GFP_KERNEL | __GFP_NOLOCKDEP);
  		alloc_flags |= __GFP_NOWARN;
  		page = alloc_pages(alloc_flags, DEPOT_POOL_ORDER);
  		if (page)
author
#7 lore raw

Re: [PATCH] stackdepot: respect __GFP_NOLOCKDEP allocation flag

In reply to Andrey Ryabinin
Quoted 72 lines · Andrey Ryabinin

If stack_depot_save_flags() allocates memory it always drops __GFP_NOLOCKDEP flag. So when KASAN tries to track __GFP_NOLOCKDEP allocation we may end up with lockdep splat like bellow:

======================================================
WARNING: possible circular locking dependency detected 6.9.0-rc3+ #49 Not tainted
------------------------------------------------------
kswapd0/149 is trying to acquire lock:
ffff88811346a920
(&xfs_nondir_ilock_class){++++}-{4:4}, at: xfs_reclaim_inode+0x3ac/0x590 [xfs]

but task is already holding lock:
ffffffff8bb33100 (fs_reclaim){+.+.}-{0:0}, at:
balance_pgdat+0x5d9/0xad0

which lock already depends on the new lock.

the existing dependency chain (in reverse order) is:
-> #1 (fs_reclaim){+.+.}-{0:0}:

         __lock_acquire+0x7da/0x1030
         lock_acquire+0x15d/0x400
         fs_reclaim_acquire+0xb5/0x100
  prepare_alloc_pages.constprop.0+0xc5/0x230
         __alloc_pages+0x12a/0x3f0
         alloc_pages_mpol+0x175/0x340
         stack_depot_save_flags+0x4c5/0x510
         kasan_save_stack+0x30/0x40
         kasan_save_track+0x10/0x30
         __kasan_slab_alloc+0x83/0x90
         kmem_cache_alloc+0x15e/0x4a0
         __alloc_object+0x35/0x370
         __create_object+0x22/0x90
  __kmalloc_node_track_caller+0x477/0x5b0
         krealloc+0x5f/0x110
         xfs_iext_insert_raw+0x4b2/0x6e0 [xfs]
         xfs_iext_insert+0x2e/0x130 [xfs]
         xfs_iread_bmbt_block+0x1a9/0x4d0 [xfs]
         xfs_btree_visit_block+0xfb/0x290 [xfs]
         xfs_btree_visit_blocks+0x215/0x2c0 [xfs]
         xfs_iread_extents+0x1a2/0x2e0 [xfs]
  xfs_buffered_write_iomap_begin+0x376/0x10a0 [xfs]
         iomap_iter+0x1d1/0x2d0
  iomap_file_buffered_write+0x120/0x1a0
         xfs_file_buffered_write+0x128/0x4b0 [xfs]
         vfs_write+0x675/0x890
         ksys_write+0xc3/0x160
         do_syscall_64+0x94/0x170
  entry_SYSCALL_64_after_hwframe+0x71/0x79

Always preserve __GFP_NOLOCKDEP to fix this.

Fixes cd11016e5f52 ("mm, kasan: stackdepot implementation. Enable stackdepot for SLAB")
Cc Christoph Hellwig <[email protected]>
---
  lib/stackdepot.c | 4 ++--
  1 file changed, 2 insertions(+), 2 deletions(-)
lib/stackdepot.c +0 -0
index 68c97387aa54..cd8f23455285 100644
@@ -627,10 +627,10 @@ depot_stack_handle_t stack_depot_save_flags(unsigned long *entries,
  		/*
  		 * Zero out zone modifiers, as we don't have specific zone

lib/stackdepot.c depot_stack_handle_t stack_depot_save_flags(unsigned long *entries, · line 636

Quoted 10 lines
lib/stackdepot.c +2 -2
  		 * requirements. Keep the flags related to allocation in atomic
-		 * contexts and I/O.
+		 * contexts, I/O, nolockdep.
  		 */
  		alloc_flags &= ~GFP_ZONEMASK;
-		alloc_flags &= (GFP_ATOMIC | GFP_KERNEL);
+		alloc_flags &= (GFP_ATOMIC | GFP_KERNEL | __GFP_NOLOCKDEP);
  		alloc_flags |= __GFP_NOWARN;
  		page = alloc_pages(alloc_flags, DEPOT_POOL_ORDER);
  		if (page)

Tested it and this patch worked well for me.

Thanks

- Xiubo