Re: [syzbot] WARNING in __split_huge_page_tail

[David Hildenbrand]

On 26.10.22 15:25, Dmitry Vyukov wrote:
> On Wed, 26 Oct 2022 at 02:54, David Hildenbrand <david@xxxxxxxxxx> wrote:
> > On 26.10.22 08:59, syzbot wrote:
> > > Hello,
> > >
> > > syzbot found the following issue on:
> > >
> > > HEAD commit:    4da34b7d175d Merge tag 'thermal-6.1-rc2' of git://git.kern..
> > > git tree:       upstream
> > > console+strace: https://syzkaller.appspot.com/x/log.txt?x=113bd8bc880000
> > > kernel config:  https://syzkaller.appspot.com/x/.config?x=4789759e8a6d5f57
> > > dashboard link: https://syzkaller.appspot.com/bug?extid=273b547b15eb58ea35e8
> > > compiler:       Debian clang version 13.0.1-++20220126092033+75e33f71c2da-1~exp1~20220126212112.63, GNU ld (GNU Binutils for Debian) 2.35.2
> > > syz repro:      https://syzkaller.appspot.com/x/repro.syz?x=161e1f62880000
> > > C reproducer:   https://syzkaller.appspot.com/x/repro.c?x=16dd4fe6880000
> > >
> > > Downloadable assets:
> > > disk image: https://storage.googleapis.com/syzbot-assets/a61ddb36c296/disk-4da34b7d.raw.xz
> > > vmlinux: https://storage.googleapis.com/syzbot-assets/ceee41246252/vmlinux-4da34b7d.xz
> > >
> > > IMPORTANT: if you fix the issue, please add the following tag to the commit:
> > > Reported-by: syzbot+273b547b15eb58ea35e8@xxxxxxxxxxxxxxxxxxxxxxxxx
> > >
> > >    tlb_finish_mmu+0xcb/0x200 mm/mmu_gather.c:363
> > >    exit_mmap+0x2b1/0x670 mm/mmap.c:3098
> > >    __mmput+0x114/0x3b0 kernel/fork.c:1185
> > >    exit_mm+0x217/0x2f0 kernel/exit.c:516
> > >    do_exit+0x5e7/0x2070 kernel/exit.c:807
> > >    do_group_exit+0x1fd/0x2b0 kernel/exit.c:950
> > >    __do_sys_exit_group kernel/exit.c:961 [inline]
> > >    __se_sys_exit_group kernel/exit.c:959 [inline]
> > >    __x64_sys_exit_group+0x3b/0x40 kernel/exit.c:959
> > >    do_syscall_x64 arch/x86/entry/common.c:50 [inline]
> > >    do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80
> > >    entry_SYSCALL_64_after_hwframe+0x63/0xcd
> > > ------------[ cut here ]------------
> > > WARNING: CPU: 0 PID: 3908 at mm/huge_memory.c:2465 __split_huge_page_tail+0x81c/0x1080 mm/huge_memory.c:2465
> >
> > Is this the
> >
> > VM_BUG_ON_PAGE(atomic_read(&page_tail->_mapcount) != -1, page_tail);
> >
> > assertion?
>
> Hi David,
>
> You can check the sources for that revision, but on the dashboard
> there are clickable links for all source references:
> https://syzkaller.appspot.com/bug?extid=273b547b15eb58ea35e8
>
> In this case it points to:
> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/mm/huge_memory.c?id=4da34b7d175dc99b8befebd69e96546c960d526c#n2465

Ah, thanks!

... so

	if (!folio_test_swapcache(page_folio(head))) {
		VM_WARN_ON_ONCE_PAGE(page_tail->private != 0, head);
		page_tail->private = 0;
	}

I recall that there was a patch either from Hugh or Mel floating around 
that might be related.

--
Thanks,

David / dhildenb