Re: [syzbot] WARNING in folio_lruvec_lock_irqsave

From: Muchun Song
Date: Thu Jun 23 2022 - 00:42:54 EST

Next message: Andrii Nakryiko: "Re: [PATCH v2] libbpf: Cleanup the kprobe_event on failed add_kprobe_event_legacy()"
Previous message: Stanley Chu: "[PATCH v2 4/8] scsi: ufs-mediatek: Add stage information for ref-clk control"
In reply to: Roman Gushchin: "Re: [syzbot] WARNING in folio_lruvec_lock_irqsave"
Next in thread: Roman Gushchin: "Re: [syzbot] WARNING in folio_lruvec_lock_irqsave"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

On Thu, Jun 23, 2022 at 10:32:32AM +0800, Muchun Song wrote:
> On Wed, Jun 22, 2022 at 11:33:48PM +0800, Muchun Song wrote:
> > On Wed, Jun 22, 2022 at 06:49:31AM -0700, syzbot wrote:
> > > Hello,
> > >
> > > syzbot found the following issue on:
> > >
> > > HEAD commit: ac0ba5454ca8 Add linux-next specific files for 20220622
> > > git tree: linux-next
> > > console output: https://syzkaller.appspot.com/x/log.txt?x=14354c18080000
> > > kernel config: https://syzkaller.appspot.com/x/.config?x=12809dacb9e7c5e0
> > > dashboard link: https://syzkaller.appspot.com/bug?extid=ec972d37869318fc3ffb
> > > 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+ec972d37869318fc3ffb@xxxxxxxxxxxxxxxxxxxxxxxxx
> > >
> > > folio_put include/linux/mm.h:1227 [inline]
> > > put_page+0x217/0x280 include/linux/mm.h:1279
> > > unmap_and_move_huge_page mm/migrate.c:1343 [inline]
> > > migrate_pages+0x3dc3/0x5a10 mm/migrate.c:1440
> > > do_mbind mm/mempolicy.c:1332 [inline]
> > > kernel_mbind+0x4d7/0x7d0 mm/mempolicy.c:1479
> > > do_syscall_x64 arch/x86/entry/common.c:50 [inline]
> > > do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80
> > > entry_SYSCALL_64_after_hwframe+0x46/0xb0
> > > page has been migrated, last migrate reason: mempolicy_mbind
> > > ------------[ cut here ]------------
> > > WARNING: CPU: 1 PID: 18925 at include/linux/memcontrol.h:800 folio_lruvec include/linux/memcontrol.h:800 [inline]
> >
> > The warning here is "VM_WARN_ON_ONCE_FOLIO(!memcg && !mem_cgroup_disabled(), folio)",
> > the memcg returned by folio_memcg() seems to be NULL which has 2 possibility, one is
> > that objcg returned by folio_objcg() is NULL, another is that obj_cgroup_memcg(objcg)
> > returns NULL. However, obj_cgroup_memcg() always returns a valid memcg. So Most likely
> > objcg is NULL meaning this page is not charged to memcg. Is this possible for LRU pages?
> >
> > I am not sure if this issue is caused by my commit cca700a8e695 ("mm: lru: use lruvec
>
> I have asked Andrew to drop this individual commit (to reduce potential impact) since
> this commit can be treated as a separate optimization patch compared to LRU page
> reparenting work. I will resend this patch again after LRU page reparenting work
> stabilizes.
>
> Thanks.
>
> > lock to serialize memcg changes") since I have removed folio_test_clear_lru() check
> > from folio_batch_move_lru(). We know that a non-lru page may be not charged to memcg.
> > But is it possible for a non-lru page to be passed to folio_batch_move_lru()? Seems

Seems my guess is right. The console log said:

[ 2295.057051][T21698] page:ffffea00013f0000 refcount:3 mapcount:0 mapping:ffff88804394b830 index:0x0 pfn:0x4fc00
[ 2295.057088][T21698] head:ffffea00013f0000 order:9 compound_mapcount:0 compound_pincount:0
[ 2295.057112][T21698] aops:hugetlbfs_aops ino:4 dentry name:"SYSV00000000"
[ 2295.057200][T21698] flags: 0xfff0000001000c(uptodate|dirty|head|node=0|zone=1|lastcpupid=0x7ff)
[ 2295.057237][T21698] raw: 00fff0000001000c ffffea0000458008 ffffffff90f93ad8 ffff88804394b830
[ 2295.057261][T21698] raw: 0000000000000000 0000000000000002 00000003ffffffff 0000000000000000
[ 2295.057275][T21698] page dumped because: VM_WARN_ON_ONCE_FOLIO(!memcg && !mem_cgroup_disabled())
[ 2295.057288][T21698] page_owner tracks the page as allocated
[ 2295.057295][T21698] page last allocated via order 9, migratetype Movable, gfp_mask 0x146cca(GFP_HIGHUSER_MOVABLE|__GFP_NOWARN|__GFP_RETRY_MAYFAIL|__GFP_COMP), pid 15372, tgid 15365 (syz-executor.2), ts 992774327150, free_ts 992742358974
[ 2295.057342][T21698] get_page_from_freelist+0x1f18/0x3a40
[ 2295.057378][T21698] __alloc_pages+0x1c7/0x510
[ 2295.057408][T21698] alloc_fresh_huge_page+0x49a/0x700
[ 2295.057443][T21698] alloc_surplus_huge_page+0x171/0x460
[ 2295.057478][T21698] gather_surplus_pages+0x1e1/0x6c0
[ 2295.057513][T21698] hugetlb_acct_memory.part.0+0x74/0xd0
[ 2295.057550][T21698] hugetlb_reserve_pages+0x4d8/0xde0
[ 2295.057586][T21698] hugetlbfs_file_mmap+0x40c/0x5c0
[ 2295.057610][T21698] shm_mmap+0xf1/0x230
[ 2295.057666][T21698] mmap_region+0x6bf/0x1bf0
[ 2295.057693][T21698] do_mmap+0x825/0xf60
[ 2295.057725][T21698] do_shmat+0xe42/0x10d0
[ 2295.057757][T21698] __x64_sys_shmat+0xcc/0x160
[ 2295.057786][T21698] do_syscall_64+0x35/0xb0
[ 2295.057844][T21698] entry_SYSCALL_64_after_hwframe+0x46/0xb0

This page is a HugeTLB page which should not be put to LRU list. So I think the following diff can
fix this issue.

diff --git a/mm/swap.c b/mm/swap.c
index ae256e333713..a877a2324be8 100644
--- a/mm/swap.c
+++ b/mm/swap.c
@@ -671,7 +671,7 @@ void deactivate_file_folio(struct folio *folio)
struct folio_batch *fbatch;

/* Deactivating an unevictable folio will not accelerate reclaim */
- if (folio_test_unevictable(folio))
+ if (!folio_test_lru(folio) || folio_test_unevictable(folio))
return;

folio_get(folio);

> > impossible. Right? I am not very confident about this commit, hopefully, someone can
> > review it.
> >
> > Thanks.
> >
> > > WARNING: CPU: 1 PID: 18925 at include/linux/memcontrol.h:800 folio_lruvec_lock_irqsave+0x2fd/0x4f0 mm/memcontrol.c:1424
> > > Modules linked in:
> > > CPU: 1 PID: 18925 Comm: syz-executor.3 Not tainted 5.19.0-rc3-next-20220622-syzkaller #0
> > > Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
> > > RIP: 0010:folio_lruvec include/linux/memcontrol.h:800 [inline]
> > > RIP: 0010:folio_lruvec_lock_irqsave+0x2fd/0x4f0 mm/memcontrol.c:1424
> > > Code: 1f 44 00 00 45 31 e4 80 3d 06 3e da 0b 00 0f 85 01 fe ff ff 48 c7 c6 40 6f da 89 4c 89 f7 e8 0a 44 e2 ff c6 05 ea 3d da 0b 01 <0f> 0b e9 e4 fd ff ff e8 67 be ad 07 85 c0 0f 84 37 fd ff ff 80 3d
> > > RSP: 0018:ffffc9000b84f2c8 EFLAGS: 00010246
> > > RAX: 0000000000040000 RBX: fffff9400027e007 RCX: ffffc900135af000
> > > RDX: 0000000000040000 RSI: ffffffff81ce36a6 RDI: fffff52001709e28
> > > RBP: dffffc0000000000 R08: 000000000000003c R09: 0000000000000000
> > > R10: 0000000080000001 R11: 0000000000000001 R12: 0000000000000000
> > > R13: fffff9400027e000 R14: ffffea00013f0000 R15: 0000000000000000
> > > FS: 00007f5cfbb96700(0000) GS:ffff8880b9b00000(0000) knlGS:0000000000000000
> > > CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> > > CR2: 000000002073f000 CR3: 0000000074b9f000 CR4: 00000000003506e0
> > > DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> > > DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
> > > Call Trace:
> > > <TASK>
> > > folio_lruvec_relock_irqsave include/linux/memcontrol.h:1666 [inline]
> > > folio_batch_move_lru+0xf9/0x500 mm/swap.c:242
> > > folio_batch_add_and_move+0xd4/0x130 mm/swap.c:258
> > > deactivate_file_folio+0x222/0x580 mm/swap.c:678
> > > invalidate_mapping_pagevec+0x38d/0x5c0 mm/truncate.c:535
> > > drop_pagecache_sb+0xcf/0x2a0 fs/drop_caches.c:39
> > > iterate_supers+0x13c/0x290 fs/super.c:694
> > > drop_caches_sysctl_handler+0xdb/0x110 fs/drop_caches.c:62
> > > proc_sys_call_handler+0x4a1/0x6e0 fs/proc/proc_sysctl.c:611
> > > call_write_iter include/linux/fs.h:2057 [inline]
> > > do_iter_readv_writev+0x3d1/0x640 fs/read_write.c:742
> > > do_iter_write+0x182/0x700 fs/read_write.c:868
> > > vfs_iter_write+0x70/0xa0 fs/read_write.c:909
> > > iter_file_splice_write+0x723/0xc70 fs/splice.c:689
> > > do_splice_from fs/splice.c:767 [inline]
> > > direct_splice_actor+0x110/0x180 fs/splice.c:936
> > > splice_direct_to_actor+0x34b/0x8c0 fs/splice.c:891
> > > do_splice_direct+0x1a7/0x270 fs/splice.c:979
> > > do_sendfile+0xae0/0x1240 fs/read_write.c:1262
> > > __do_sys_sendfile64 fs/read_write.c:1321 [inline]
> > > __se_sys_sendfile64 fs/read_write.c:1313 [inline]
> > > __x64_sys_sendfile64+0x149/0x210 fs/read_write.c:1313
> > > do_syscall_x64 arch/x86/entry/common.c:50 [inline]
> > > do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80
> > > entry_SYSCALL_64_after_hwframe+0x46/0xb0
> > > RIP: 0033:0x7f5cfaa89109
> > > Code: ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 40 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 b8 ff ff ff f7 d8 64 89 01 48
> > > RSP: 002b:00007f5cfbb96168 EFLAGS: 00000246 ORIG_RAX: 0000000000000028
> > > RAX: ffffffffffffffda RBX: 00007f5cfab9c030 RCX: 00007f5cfaa89109
> > > RDX: 0000000020002080 RSI: 0000000000000005 RDI: 0000000000000006
> > > RBP: 00007f5cfaae305d R08: 0000000000000000 R09: 0000000000000000
> > > R10: 0000000000000262 R11: 0000000000000246 R12: 0000000000000000
> > > R13: 00007fff1ef394df R14: 00007f5cfbb96300 R15: 0000000000022000
> > > </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@xxxxxxxxxxxxxxxx.
> > >
> > > syzbot will keep track of this issue. See:
> > > https://goo.gl/tpsmEJ#status for how to communicate with syzbot.
> > >
> >
>

Next message: Andrii Nakryiko: "Re: [PATCH v2] libbpf: Cleanup the kprobe_event on failed add_kprobe_event_legacy()"
Previous message: Stanley Chu: "[PATCH v2 4/8] scsi: ufs-mediatek: Add stage information for ref-clk control"
In reply to: Roman Gushchin: "Re: [syzbot] WARNING in folio_lruvec_lock_irqsave"
Next in thread: Roman Gushchin: "Re: [syzbot] WARNING in folio_lruvec_lock_irqsave"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]