Re: [Syzkaller & bisect] There is BUG: soft lockup after sendmsg syscall in v6.8-rc4

[Jason Xing]

On Wed, Feb 21, 2024 at 10:35 AM Pengfei Xu <pengfei.xu@xxxxxxxxx> wrote:
>
> Hi Kuniyuki Iwashima and kernel experts,
>
> Greeting!
> There is BUG: soft lockup after sendmsg syscall in v6.8-rc4 in guest.
>
> All detailed info: https://github.com/xupengfe/syzkaller_logs/tree/main/240220_161242_softlockup
> Syzkaller reproduced code: https://github.com/xupengfe/syzkaller_logs/blob/main/240220_161242_softlockup/repro.c
> Syzkaller syscall reproduced steps: https://github.com/xupengfe/syzkaller_logs/blob/main/240220_161242_softlockup/repro.prog
> Kconfig(need make olddefconfig): https://github.com/xupengfe/syzkaller_logs/blob/main/240220_161242_softlockup/kconfig_origin
> Bisect info: https://github.com/xupengfe/syzkaller_logs/blob/main/240220_161242_softlockup/bisect_info.log
> v6.8-rc4 issue dmesg: https://github.com/xupengfe/syzkaller_logs/blob/main/240220_161242_softlockup/841c35169323cd833294798e58b9bf63fa4fa1de_dmesg.log
> bzImage_v6.8-rc4: https://github.com/xupengfe/syzkaller_logs/raw/main/240220_161242_softlockup/bzImage_v6.8-rc4.tar.gz
>
> Bisected and found first bad commit:
> "
> 1279f9d9dec2 af_unix: Call kfree_skb() for dead unix_(sk)->oob_skb in GC.
> "
>
> After reverted above commit on top of v6.8-rc4 kernel, this issue was gone.
>
> Syzkaller repro.report: https://github.com/xupengfe/syzkaller_logs/blob/main/240220_161242_softlockup/repro.report
> "
> watchdog: BUG: soft lockup - CPU#0 stuck for 26s! [gdbus:349]
> Modules linked in:
> irq event stamp: 25162
> hardirqs last  enabled at (25161): [<ffffffff855dff2e>] irqentry_exit+0x3e/0xa0 kernel/entry/common.c:351
> hardirqs last disabled at (25162): [<ffffffff855dded4>] sysvec_apic_timer_interrupt+0x14/0xc0 arch/x86/kernel/apic/apic.c:1076
> softirqs last  enabled at (25140): [<ffffffff8127fcc8>] invoke_softirq kernel/softirq.c:427 [inline]
> softirqs last  enabled at (25140): [<ffffffff8127fcc8>] __irq_exit_rcu+0xa8/0x110 kernel/softirq.c:632
> softirqs last disabled at (25135): [<ffffffff8127fcc8>] invoke_softirq kernel/softirq.c:427 [inline]
> softirqs last disabled at (25135): [<ffffffff8127fcc8>] __irq_exit_rcu+0xa8/0x110 kernel/softirq.c:632
> CPU: 0 PID: 349 Comm: gdbus Not tainted 6.8.0-rc4-2024-02-12-intel-next-e92619743fcb+ #1
> Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.16.0-0-gd239552ce722-prebuilt.qemu.org 04/01/2014
> RIP: 0010:__sanitizer_cov_trace_pc+0x38/0x70 kernel/kcov.c:207
> Code: 65 8b 05 73 89 a0 7e 48 89 e5 48 8b 75 08 a9 00 01 ff 00 74 0f f6 c4 01 74 35 8b 82 e4 1d 00 00 85 c0 74 2b 8b 82 c0 1d 00 00 <83> f8 02 75 20 48 8b 8a c8 1d 00 00 8b 92 c4 1d 00 00 48 8b 01 48
> RSP: 0018:ffff88800b48f7b0 EFLAGS: 00000246
> RAX: 0000000000000000 RBX: ffff888013018000 RCX: 1ffffffff12150bb
> RDX: ffff888011ab8000 RSI: ffffffff8515e235 RDI: ffff888013018770
> RBP: ffff88800b48f7b0 R08: 0000000000000001 R09: fffffbfff120f86e
> R10: ffffffff8907c377 R11: 0000000000000001 R12: ffff888013018000
> R13: dffffc0000000000 R14: ffff888013018630 R15: ffff88800b48f840
> FS:  0000000000000000(0000) GS:ffff88806ca00000(0000) knlGS:0000000000000000
> CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> CR2: 00007fffcc5fd648 CR3: 0000000006c82000 CR4: 0000000000750ef0
> PKRU: 55555554
> Call Trace:
>  <IRQ>
>  </IRQ>
>  <TASK>
>  unix_gc+0x465/0xfd0 net/unix/garbage.c:319
>  unix_release_sock+0xb8c/0xe80 net/unix/af_unix.c:683
>  unix_release+0x9c/0x100 net/unix/af_unix.c:1064
>  __sock_release+0xb6/0x280 net/socket.c:659
>  sock_close+0x27/0x40 net/socket.c:1421
>  __fput+0x284/0xb70 fs/file_table.c:376
>  ____fput+0x1f/0x30 fs/file_table.c:404
>  task_work_run+0x19d/0x2b0 kernel/task_work.c:180
>  exit_task_work include/linux/task_work.h:38 [inline]
>  do_exit+0xb51/0x28c0 kernel/exit.c:871
>  do_group_exit+0xe5/0x2c0 kernel/exit.c:1020
>  get_signal+0x2715/0x27d0 kernel/signal.c:2893
>  arch_do_signal_or_restart+0x8e/0x7e0 arch/x86/kernel/signal.c:311
>  exit_to_user_mode_loop kernel/entry/common.c:105 [inline]
>  exit_to_user_mode_prepare include/linux/entry-common.h:328 [inline]
>  __syscall_exit_to_user_mode_work kernel/entry/common.c:201 [inline]
>  syscall_exit_to_user_mode+0x129/0x190 kernel/entry/common.c:212
>  do_syscall_64+0x82/0x150 arch/x86/entry/common.c:89
>  entry_SYSCALL_64_after_hwframe+0x6e/0x76
> RIP: 0033:0x7fe8f8b4296f
> Code: Unable to access opcode bytes at 0x7fe8f8b42945.
> RSP: 002b:00007fe8d7dff9c0 EFLAGS: 00000293 ORIG_RAX: 0000000000000007
> RAX: fffffffffffffdfc RBX: 00007fe8f8f73071 RCX: 00007fe8f8b4296f
> RDX: 00000000ffffffff RSI: 0000000000000002 RDI: 00007fe8a801f3f0
> RBP: 00007fe8a801f3f0 R08: 0000000000000000 R09: 00007fe8d7dff850
> R10: 00007ffec9196080 R11: 0000000000000293 R12: 0000000000000002
> R13: 0000000000000002 R14: 00007fe8d7dffa30 R15: 00007fe8a801c4c0
>  </TASK>
> "

Hello, did you try this one:

commit 25236c91b5ab4a26a56ba2e79b8060cf4e047839
Author: Kuniyuki Iwashima <kuniyu@xxxxxxxxxx>
Date:   Fri Feb 9 14:04:53 2024 -0800

    af_unix: Fix task hung while purging oob_skb in GC.

    syzbot reported a task hung; at the same time, GC was looping infinitely
    in list_for_each_entry_safe() for OOB skb.  [0]

    syzbot demonstrated that the list_for_each_entry_safe() was not actually
    safe in this case.

    A single skb could have references for multiple sockets.  If we free such
    a skb in the list_for_each_entry_safe(), the current and next sockets could
    be unlinked in a single iteration.

It should work, I think :)

Thanks,
Jason

>
> Thanks!
>
> ---
>
> If you don't need the following environment to reproduce the problem or if you
> already have one reproduced environment, please ignore the following information.
>
> How to reproduce:
> git clone https://gitlab.com/xupengfe/repro_vm_env.git
> cd repro_vm_env
> tar -xvf repro_vm_env.tar.gz
> cd repro_vm_env; ./start3.sh  // it needs qemu-system-x86_64 and I used v7.1.0
>   // start3.sh will load bzImage_2241ab53cbb5cdb08a6b2d4688feb13971058f65 v6.2-rc5 kernel
>   // You could change the bzImage_xxx as you want
>   // Maybe you need to remove line "-drive if=pflash,format=raw,readonly=on,file=./OVMF_CODE.fd \" for different qemu version
> You could use below command to log in, there is no password for root.
> ssh -p 10023 root@localhost
>
> After login vm(virtual machine) successfully, you could transfer reproduced
> binary to the vm by below way, and reproduce the problem in vm:
> gcc -pthread -o repro repro.c
> scp -P 10023 repro root@localhost:/root/
>
> Get the bzImage for target kernel:
> Please use target kconfig and copy it to kernel_src/.config
> make olddefconfig
> make -jx bzImage           //x should equal or less than cpu num your pc has
>
> Fill the bzImage file into above start3.sh to load the target kernel in vm.
>
>
> Tips:
> If you already have qemu-system-x86_64, please ignore below info.
> If you want to install qemu v7.1.0 version:
> git clone https://github.com/qemu/qemu.git
> cd qemu
> git checkout -f v7.1.0
> mkdir build
> cd build
> yum install -y ninja-build.x86_64
> yum -y install libslirp-devel.x86_64
> ../configure --target-list=x86_64-softmmu --enable-kvm --enable-vnc --enable-gtk --enable-sdl --enable-usb-redir --enable-slirp
> make
> make install
>
> Best Regards,
> Thanks!
>