On Tue, Oct 10, 2023 at 4:20 PM Johannes Weiner <hannes@xxxxxxxxxxx> wrote: > > This is the earlier version of the hugetlb cgroup accounting patches > that trigger on an uncharged hugetlbfs: > > 7547 /* > 7548 * Note that it is normal to see !memcg for a hugetlb folio. > 7549 * It could have been allocated when memory_hugetlb_accounting was not > 7550 * selected, for e.g. > 7551 */ > 7552 VM_WARN_ON_ONCE_FOLIO(!memcg, old); > > It's been fixed in the revision that's in the latest next release: > > 7539 /* > 7540 * Note that it is normal to see !memcg for a hugetlb folio. > 7541 * For e.g, itt could have been allocated when memory_hugetlb_accounting > 7542 * was not selected. > 7543 */ > 7544 VM_WARN_ON_ONCE_FOLIO(!folio_test_hugetlb(old) && !memcg, old); > 7545 if (!memcg) > 7546 return; > > > Modules linked in: > > CPU: 1 PID: 5208 Comm: syz-executor.1 Not tainted 6.6.0-rc4-next-20231005-syzkaller #0 > > Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 09/06/2023 > > RIP: 0010:mem_cgroup_migrate+0x2fa/0x390 mm/memcontrol.c:7552 > > Code: f7 ff e9 36 ff ff ff 80 3d 84 b2 d1 0c 00 0f 85 54 ff ff ff 48 c7 c6 a0 9e 9b 8a 48 89 ef e8 0d 5c df ff c6 05 68 b2 d1 0c 01 <0f> 0b e9 37 ff ff ff 48 c7 c6 e0 9a 9b 8a 48 89 df e8 f0 5b df ff > > RSP: 0018:ffffc90004b2fa38 EFLAGS: 00010246 > > RAX: 0000000000040000 RBX: ffffea0005338000 RCX: ffffc90005439000 > > RDX: 0000000000040000 RSI: ffffffff81e76463 RDI: ffffffff8ae96da0 > > RBP: ffffea0001d98000 R08: 0000000000000000 R09: fffffbfff1d9db9a > > R10: ffffffff8ecedcd7 R11: 0000000000000000 R12: 0000000000000000 > > R13: 0000000000000200 R14: 0000000000000000 R15: ffffea0001d98018 > > FS: 00007fc15e89d6c0(0000) GS:ffff8880b9900000(0000) knlGS:0000000000000000 > > CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 > > CR2: 0000001b31820000 CR3: 000000007f5e1000 CR4: 00000000003506f0 > > DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 > > DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 > > Call Trace: > > <TASK> > > hugetlbfs_migrate_folio fs/hugetlbfs/inode.c:1066 [inline] > > hugetlbfs_migrate_folio+0xd0/0x120 fs/hugetlbfs/inode.c:1049 > > move_to_new_folio+0x183/0x690 mm/migrate.c:966 > > unmap_and_move_huge_page mm/migrate.c:1428 [inline] > > migrate_hugetlbs mm/migrate.c:1546 [inline] > > migrate_pages+0x16ac/0x27c0 mm/migrate.c:1900 > > migrate_to_node mm/mempolicy.c:1072 [inline] > > do_migrate_pages+0x43e/0x690 mm/mempolicy.c:1171 > > kernel_migrate_pages+0x59b/0x780 mm/mempolicy.c:1682 > > __do_sys_migrate_pages mm/mempolicy.c:1700 [inline] > > __se_sys_migrate_pages mm/mempolicy.c:1696 [inline] > > __x64_sys_migrate_pages+0x96/0x100 mm/mempolicy.c:1696 > > do_syscall_x64 arch/x86/entry/common.c:51 [inline] > > do_syscall_64+0x38/0xb0 arch/x86/entry/common.c:81 > > entry_SYSCALL_64_after_hwframe+0x63/0xcd > > RIP: 0033:0x7fc15da7cae9 > > Code: 28 00 00 00 75 05 48 83 c4 28 c3 e8 e1 20 00 00 90 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 b0 ff ff ff f7 d8 64 89 01 48 > > RSP: 002b:00007fc15e89d0c8 EFLAGS: 00000246 ORIG_RAX: 0000000000000100 > > RAX: ffffffffffffffda RBX: 00007fc15db9bf80 RCX: 00007fc15da7cae9 > > RDX: 0000000020000340 RSI: 0000000000000080 RDI: 0000000000000000 > > RBP: 00007fc15dac847a R08: 0000000000000000 R09: 0000000000000000 > > R10: 00000000200003c0 R11: 0000000000000246 R12: 0000000000000000 > > R13: 000000000000000b R14: 00007fc15db9bf80 R15: 00007ffd87d7c058 > > </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. > > > > If the bug is already fixed, let syzbot know by replying with: > > #syz fix: exact-commit-title > > #syz fix: next-20231010 > Thanks for sharing the info and updating the issue! If there's no fixing commit (the faulty series is dropped or replaced), it's better to just invalidate the report: #syz invalid Otherwise, as in this case, syzbot would start looking for the "next-20231010" commit (and won't find it because it's a tag) and, after some time, start complaining that no such commit is reachable from any of the master branches of the tested trees. -- Aleksandr
