Re: [syzbot ci] Re: Eliminate Dying Memory Cgroup

[Qi Zheng <qi.zheng@linux.dev>]

On 1/15/26 1:07 AM, syzbot ci wrote:
> syzbot ci has tested the following series
> 
> [v3] Eliminate Dying Memory Cgroup
> https://lore.kernel.org/all/cover.1768389889.git.zhengqi.arch@bytedance.com
> * [PATCH v3 01/30] mm: memcontrol: remove dead code of checking parent memory cgroup
> * [PATCH v3 02/30] mm: workingset: use folio_lruvec() in workingset_refault()
> * [PATCH v3 03/30] mm: rename unlock_page_lruvec_irq and its variants
> * [PATCH v3 04/30] mm: vmscan: prepare for the refactoring the move_folios_to_lru()
> * [PATCH v3 05/30] mm: vmscan: refactor move_folios_to_lru()
> * [PATCH v3 06/30] mm: memcontrol: allocate object cgroup for non-kmem case
> * [PATCH v3 07/30] mm: memcontrol: return root object cgroup for root memory cgroup
> * [PATCH v3 08/30] mm: memcontrol: prevent memory cgroup release in get_mem_cgroup_from_folio()
> * [PATCH v3 09/30] buffer: prevent memory cgroup release in folio_alloc_buffers()
> * [PATCH v3 10/30] writeback: prevent memory cgroup release in writeback module
> * [PATCH v3 11/30] mm: memcontrol: prevent memory cgroup release in count_memcg_folio_events()
> * [PATCH v3 12/30] mm: page_io: prevent memory cgroup release in page_io module
> * [PATCH v3 13/30] mm: migrate: prevent memory cgroup release in folio_migrate_mapping()
> * [PATCH v3 14/30] mm: mglru: prevent memory cgroup release in mglru
> * [PATCH v3 15/30] mm: memcontrol: prevent memory cgroup release in mem_cgroup_swap_full()
> * [PATCH v3 16/30] mm: workingset: prevent memory cgroup release in lru_gen_eviction()
> * [PATCH v3 17/30] mm: thp: prevent memory cgroup release in folio_split_queue_lock{_irqsave}()
> * [PATCH v3 18/30] mm: zswap: prevent memory cgroup release in zswap_compress()
> * [PATCH v3 19/30] mm: workingset: prevent lruvec release in workingset_refault()
> * [PATCH v3 20/30] mm: zswap: prevent lruvec release in zswap_folio_swapin()
> * [PATCH v3 21/30] mm: swap: prevent lruvec release in lru_gen_clear_refs()
> * [PATCH v3 22/30] mm: workingset: prevent lruvec release in workingset_activation()
> * [PATCH v3 23/30] mm: do not open-code lruvec lock
> * [PATCH v3 24/30] mm: memcontrol: prepare for reparenting LRU pages for lruvec lock
> * [PATCH v3 25/30] mm: vmscan: prepare for reparenting traditional LRU folios
> * [PATCH v3 26/30] mm: vmscan: prepare for reparenting MGLRU folios
> * [PATCH v3 27/30] mm: memcontrol: refactor memcg_reparent_objcgs()
> * [PATCH v3 28/30] mm: memcontrol: prepare for reparenting state_local
> * [PATCH v3 29/30] mm: memcontrol: eliminate the problem of dying memory cgroup for LRU folios
> * [PATCH v3 30/30] mm: lru: add VM_WARN_ON_ONCE_FOLIO to lru maintenance helpers
> 
> and found the following issue:
> UBSAN: array-index-out-of-bounds in reparent_memcg_lruvec_state_local
> 
> Full report is available here:
> https://ci.syzbot.org/series/45c0b58d-255a-4579-9880-497bdbd4fb99
> 
> ***
> 
> UBSAN: array-index-out-of-bounds in reparent_memcg_lruvec_state_local
> 
> tree:      linux-next
> URL:       https://kernel.googlesource.com/pub/scm/linux/kernel/git/next/linux-next
> base:      b775e489bec70895b7ef6b66927886bbac79598f
> arch:      amd64
> compiler:  Debian clang version 21.1.8 (++20251221033036+2078da43e25a-1~exp1~20251221153213.50), Debian LLD 21.1.8
> config:    https://ci.syzbot.org/builds/4d8819ab-0f94-42e8-bd70-87c7e83c37d2/config
> syz repro: https://ci.syzbot.org/findings/7850f5dd-4ac7-4b74-85ff-a75ddddebbee/syz_repro
> 
> ------------[ cut here ]------------
> UBSAN: array-index-out-of-bounds in mm/memcontrol.c:530:3
> index 33 is out of range for type 'long[33]'

Oh, the size of lruvec_stats->state_local is NR_MEMCG_NODE_STAT_ITEMS,
but memcg1_stats contains MEMCG_SWAP, which is outside the array range.

It seems that only the following items need to be reparented:

1). NR_LRU_LISTS
2). NR_SLAB_RECLAIMABLE_B + NR_SLAB_UNRECLAIMABLE_B

But for 2), since we reparented the slab page a long time ago, it seems
there has always been a problem. So this patchset will only handle 1).


> CPU: 1 UID: 0 PID: 31 Comm: kworker/1:1 Not tainted syzkaller #0 PREEMPT(full)
> Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.2-debian-1.16.2-1 04/01/2014
> Workqueue: cgroup_offline css_killed_work_fn
> Call Trace:
>   <TASK>
>   dump_stack_lvl+0xe8/0x150 lib/dump_stack.c:120
>   ubsan_epilogue+0xa/0x30 lib/ubsan.c:233
>   __ubsan_handle_out_of_bounds+0xe8/0xf0 lib/ubsan.c:455
>   reparent_memcg_lruvec_state_local+0x34f/0x460 mm/memcontrol.c:530
>   reparent_memcg1_lruvec_state_local+0xa7/0xc0 mm/memcontrol-v1.c:1917
>   reparent_state_local mm/memcontrol.c:242 [inline]
>   memcg_reparent_objcgs mm/memcontrol.c:299 [inline]
>   mem_cgroup_css_offline+0xc7c/0xc90 mm/memcontrol.c:4054
>   offline_css kernel/cgroup/cgroup.c:5760 [inline]
>   css_killed_work_fn+0x12f/0x570 kernel/cgroup/cgroup.c:6055
>   process_one_work+0x949/0x15a0 kernel/workqueue.c:3279
>   process_scheduled_works kernel/workqueue.c:3362 [inline]
>   worker_thread+0x9af/0xee0 kernel/workqueue.c:3443
>   kthread+0x388/0x470 kernel/kthread.c:467
>   ret_from_fork+0x51b/0xa40 arch/x86/kernel/process.c:158
>   ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:246
>   </TASK>
> ---[ end trace ]---
> Kernel panic - not syncing: UBSAN: panic_on_warn set ...
> CPU: 1 UID: 0 PID: 31 Comm: kworker/1:1 Not tainted syzkaller #0 PREEMPT(full)
> Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.2-debian-1.16.2-1 04/01/2014
> Workqueue: cgroup_offline css_killed_work_fn
> Call Trace:
>   <TASK>
>   vpanic+0x1e0/0x670 kernel/panic.c:490
>   panic+0xc5/0xd0 kernel/panic.c:627
>   check_panic_on_warn+0x89/0xb0 kernel/panic.c:377
>   __ubsan_handle_out_of_bounds+0xe8/0xf0 lib/ubsan.c:455
>   reparent_memcg_lruvec_state_local+0x34f/0x460 mm/memcontrol.c:530
>   reparent_memcg1_lruvec_state_local+0xa7/0xc0 mm/memcontrol-v1.c:1917
>   reparent_state_local mm/memcontrol.c:242 [inline]
>   memcg_reparent_objcgs mm/memcontrol.c:299 [inline]
>   mem_cgroup_css_offline+0xc7c/0xc90 mm/memcontrol.c:4054
>   offline_css kernel/cgroup/cgroup.c:5760 [inline]
>   css_killed_work_fn+0x12f/0x570 kernel/cgroup/cgroup.c:6055
>   process_one_work+0x949/0x15a0 kernel/workqueue.c:3279
>   process_scheduled_works kernel/workqueue.c:3362 [inline]
>   worker_thread+0x9af/0xee0 kernel/workqueue.c:3443
>   kthread+0x388/0x470 kernel/kthread.c:467
>   ret_from_fork+0x51b/0xa40 arch/x86/kernel/process.c:158
>   ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:246
>   </TASK>
> Kernel Offset: disabled
> Rebooting in 86400 seconds..
> 
> 
> ***
> 
> If these findings have caused you to resend the series or submit a
> separate fix, please add the following tag to your commit message:
>    Tested-by: syzbot@syzkaller.appspotmail.com
> 
> ---
> This report is generated by a bot. It may contain errors.
> syzbot ci engineers can be reached at syzkaller@googlegroups.com.