linux-kernel - Re: [syzbot] [mm?] KCSAN: data-race in exec_mmap / vms_clear

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <d45f5c9e-be61-457e-9a38-d8fdd5e942b1@lucifer.local>
Date: Mon, 30 Sep 2024 18:47:13 +0100
From: Lorenzo Stoakes <lorenzo.stoakes@...cle.com>
To: syzbot <syzbot+d207c41e97001109b49d@...kaller.appspotmail.com>
Cc: Liam.Howlett@...cle.com, akpm@...ux-foundation.org,
        linux-kernel@...r.kernel.org, linux-mm@...ck.org,
        syzkaller-bugs@...glegroups.com, vbabka@...e.cz
Subject: Re: [syzbot] [mm?] KCSAN: data-race in exec_mmap / vms_clear_ptes

On Mon, Sep 30, 2024 at 12:39:24AM GMT, syzbot wrote:
> Hello,
>
> syzbot found the following issue on:
>
> HEAD commit:    e7ed34365879 Merge tag 'mailbox-v6.12' of git://git.kernel..
> git tree:       upstream
> console output: https://syzkaller.appspot.com/x/log.txt?x=15238ea9980000
> kernel config:  https://syzkaller.appspot.com/x/.config?x=8bfe37bd3f5983d6
> dashboard link: https://syzkaller.appspot.com/bug?extid=d207c41e97001109b49d
> compiler:       Debian clang version 15.0.6, GNU ld (GNU Binutils for Debian) 2.40
>
> Unfortunately, I don't have any reproducer for this issue yet.
>
> Downloadable assets:
> disk image: https://storage.googleapis.com/syzbot-assets/c86c1297298e/disk-e7ed3436.raw.xz
> vmlinux: https://storage.googleapis.com/syzbot-assets/8313c0846b3b/vmlinux-e7ed3436.xz
> kernel image: https://storage.googleapis.com/syzbot-assets/8af10d85db09/bzImage-e7ed3436.xz
>
> IMPORTANT: if you fix the issue, please add the following tag to the commit:
> Reported-by: syzbot+d207c41e97001109b49d@...kaller.appspotmail.com
>
> ==================================================================
> BUG: KCSAN: data-race in exec_mmap / vms_clear_ptes
>
> write to 0xffff888102fbaae8 of 8 bytes by task 3004 on cpu 1:
>  update_hiwater_rss include/linux/mm.h:2655 [inline]

OK so this sets mm_struct->hiwater_rss on munmap() via update_hiwater_rss() in
vms_clear_ptes()...

>  vms_clear_ptes+0x1a7/0x300 mm/vma.c:1088
>  vms_complete_munmap_vmas+0x170/0x480 mm/vma.c:1140
>  do_vmi_align_munmap+0x349/0x390 mm/vma.c:1349
>  do_vmi_munmap+0x1eb/0x230 mm/vma.c:1397
>  __vm_munmap+0xfd/0x220 mm/mmap.c:1600
>  __do_sys_munmap mm/mmap.c:1617 [inline]
>  __se_sys_munmap mm/mmap.c:1614 [inline]
>  __x64_sys_munmap+0x36/0x40 mm/mmap.c:1614
>  x64_sys_call+0xd32/0x2d60 arch/x86/include/generated/asm/syscalls_64.h:12
>  do_syscall_x64 arch/x86/entry/common.c:52 [inline]
>  do_syscall_64+0xc9/0x1c0 arch/x86/entry/common.c:83
>  entry_SYSCALL_64_after_hwframe+0x77/0x7f
>

> read to 0xffff888102fbaae8 of 8 bytes by task 5494 on cpu 0:
>  get_mm_hiwater_rss include/linux/mm.h:2642 [inline]

...And a racing execve() is trying to read it.

This is a bit of a tricky race, as we downgrade the write lock to a read lock in
vms_complete_munmap_vmas(), which allows exec_mmap() to proceed (as it's under a
read lock), and update_hiwater_rss() is called against vms->vma->vm_mm which is
set to the old_mm and...

I wonder if we should just be referencing current->mm in vms_clear_ptes() which
would avoid this as will be the new mm that has been execve'd in (and presumably
do nothing)?

I don't think in practice this should have too egregious an impact as the
process is being execve()'d anyway, so I think we can wait for Liam to return
from leave to give his input.

On your return - Liam what do you think? :)

>  setmax_mm_hiwater_rss include/linux/mm.h:2672 [inline]
>  exec_mmap+0x2d0/0x440 fs/exec.c:1012
>  begin_new_exec+0xaf9/0x10b0 fs/exec.c:1280
>  load_elf_binary+0x63b/0x19e0 fs/binfmt_elf.c:996
>  search_binary_handler fs/exec.c:1752 [inline]
>  exec_binprm fs/exec.c:1794 [inline]
>  bprm_execve+0x4e6/0xc30 fs/exec.c:1845
>  do_execveat_common+0x793/0x800 fs/exec.c:1952
>  do_execve fs/exec.c:2026 [inline]
>  __do_sys_execve fs/exec.c:2102 [inline]
>  __se_sys_execve fs/exec.c:2097 [inline]
>  __x64_sys_execve+0x5a/0x70 fs/exec.c:2097
>  x64_sys_call+0x1277/0x2d60 arch/x86/include/generated/asm/syscalls_64.h:60
>  do_syscall_x64 arch/x86/entry/common.c:52 [inline]
>  do_syscall_64+0xc9/0x1c0 arch/x86/entry/common.c:83
>  entry_SYSCALL_64_after_hwframe+0x77/0x7f
>
> value changed: 0x000000000000022b -> 0x000000000000024b
>
> Reported by Kernel Concurrency Sanitizer on:
> CPU: 0 UID: 0 PID: 5494 Comm: dhcpcd Not tainted 6.11.0-syzkaller-12113-ge7ed34365879 #0
> Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 09/13/2024
> ==================================================================
>
>
> ---
> 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@...glegroups.com.
>
> syzbot will keep track of this issue. See:
> https://goo.gl/tpsmEJ#status for how to communicate with syzbot.
>
> If the report is already addressed, let syzbot know by replying with:
> #syz fix: exact-commit-title
>
> If you want to overwrite report's subsystems, reply with:
> #syz set subsystems: new-subsystem
> (See the list of subsystem names on the web dashboard)
>
> If the report is a duplicate of another one, reply with:
> #syz dup: exact-subject-of-another-report
>
> If you want to undo deduplication, reply with:
> #syz undup