Re: [syzbot] [kvm?] WARNING in kvm_tdp_mmu_invalidate_all_roots

[Sean Christopherson]

On Fri, Apr 21, 2023, syzbot wrote:
> Hello,
> 
> syzbot found the following issue on:
> 
> HEAD commit:    d3e1ee0e67e7 Add linux-next specific files for 20230421
> git tree:       linux-next
> console output: https://syzkaller.appspot.com/x/log.txt?x=16ac3de0280000
> kernel config:  https://syzkaller.appspot.com/x/.config?x=53c789efbcc06cf6
> dashboard link: https://syzkaller.appspot.com/bug?extid=094a74444165dbcd3a54
> 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.
> 
> Downloadable assets:
> disk image: https://storage.googleapis.com/syzbot-assets/c558a9e1fe6a/disk-d3e1ee0e.raw.xz
> vmlinux: https://storage.googleapis.com/syzbot-assets/2ec100a34c4c/vmlinux-d3e1ee0e.xz
> kernel image: https://storage.googleapis.com/syzbot-assets/1afcd9936dc1/bzImage-d3e1ee0e.xz
> 
> IMPORTANT: if you fix the issue, please add the following tag to the commit:
> Reported-by: syzbot+094a74444165dbcd3a54@xxxxxxxxxxxxxxxxxxxxxxxxx
> 
> ------------[ cut here ]------------
> WARNING: CPU: 0 PID: 12623 at arch/x86/kvm/mmu/tdp_mmu.c:943 kvm_tdp_mmu_invalidate_all_roots+0x2bd/0x370 arch/x86/kvm/mmu/tdp_mmu.c:943
> Modules linked in:
> CPU: 0 PID: 12623 Comm: syz-executor.3 Not tainted 6.3.0-rc7-next-20230421-syzkaller #0
> Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 03/30/2023
> RIP: 0010:kvm_tdp_mmu_invalidate_all_roots+0x2bd/0x370 arch/x86/kvm/mmu/tdp_mmu.c:943
> Call Trace:
>  <TASK>
>  kvm_mmu_uninit_tdp_mmu+0x16/0x100 arch/x86/kvm/mmu/tdp_mmu.c:48
>  kvm_mmu_uninit_vm+0x6a/0x70 arch/x86/kvm/mmu/mmu.c:6239
>  kvm_arch_destroy_vm+0x369/0x490 arch/x86/kvm/x86.c:12465
>  kvm_create_vm arch/x86/kvm/../../../virt/kvm/kvm_main.c:1245 [inline]
>  kvm_dev_ioctl_create_vm arch/x86/kvm/../../../virt/kvm/kvm_main.c:5017 [inline]
>  kvm_dev_ioctl+0x11be/0x1bb0 arch/x86/kvm/../../../virt/kvm/kvm_main.c:5059
>  vfs_ioctl fs/ioctl.c:51 [inline]
>  __do_sys_ioctl fs/ioctl.c:870 [inline]
>  __se_sys_ioctl fs/ioctl.c:856 [inline]
>  __x64_sys_ioctl+0x197/0x210 fs/ioctl.c:856
>  do_syscall_x64 arch/x86/entry/common.c:50 [inline]
>  do_syscall_64+0x39/0xb0 arch/x86/entry/common.c:80
>  entry_SYSCALL_64_after_hwframe+0x63/0xcd

Gah, kvm->users_count is elevated when the VM is destroyed if creation fails,
but mmu_lock isn't held in that case, and lockdep yells.  This particular bug is
effectively a false positive, nothing else holds a reference to the VM.

However, syzbot found another bug that _is_ a real problem (hasn't been reported
upstream yet for whatever reason).  Not holding mmu_lock during "real" VM destruction
is problematic because walking the list can race with the asynchronous worker
deleting and freeing from the list.

Posted a fixed version[*] that should resolve both issues by protecting the walk
with RCU.

[*] https://lore.kernel.org/all/20230421214946.2571580-1-seanjc@xxxxxxxxxx

==================================================================
BUG: KASAN: slab-use-after-free in kvm_tdp_mmu_invalidate_all_roots+0x2e3/0x370 arch/x86/kvm/mmu/tdp_mmu.c:945
Read of size 8 at addr ffff88801c8cf948 by task syz-executor.1/17012

CPU: 0 PID: 17012 Comm: syz-executor.1 Not tainted 6.3.0-rc7-next-20230421-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 03/30/2023
Call Trace:
<TASK>
__dump_stack lib/dump_stack.c:88 [inline]
dump_stack_lvl+0xd9/0x150 lib/dump_stack.c:106
print_address_description.constprop.0+0x2c/0x3c0 mm/kasan/report.c:351
print_report mm/kasan/report.c:462 [inline]
kasan_report+0x11c/0x130 mm/kasan/report.c:572
kvm_tdp_mmu_invalidate_all_roots+0x2e3/0x370 arch/x86/kvm/mmu/tdp_mmu.c:945
kvm_mmu_uninit_tdp_mmu+0x16/0x100 arch/x86/kvm/mmu/tdp_mmu.c:48
kvm_mmu_uninit_vm+0x6a/0x70 arch/x86/kvm/mmu/mmu.c:6239
kvm_arch_destroy_vm+0x369/0x490 arch/x86/kvm/x86.c:12465
kvm_destroy_vm arch/x86/kvm/../../../virt/kvm/kvm_main.c:1313 [inline]
kvm_put_kvm+0x4da/0xae0 arch/x86/kvm/../../../virt/kvm/kvm_main.c:1347
kvm_vcpu_release+0x51/0x70 arch/x86/kvm/../../../virt/kvm/kvm_main.c:3846
__fput+0x27c/0xa90 fs/file_table.c:321
task_work_run+0x16f/0x270 kernel/task_work.c:179
resume_user_mode_work include/linux/resume_user_mode.h:49 [inline]
exit_to_user_mode_loop kernel/entry/common.c:171 [inline]
exit_to_user_mode_prepare+0x210/0x240 kernel/entry/common.c:204
__syscall_exit_to_user_mode_work kernel/entry/common.c:286 [inline]
syscall_exit_to_user_mode+0x1d/0x50 kernel/entry/common.c:297
do_syscall_64+0x46/0xb0 arch/x86/entry/common.c:86
entry_SYSCALL_64_after_hwframe+0x63/0xcd
RIP: 0033:0x7fd7a583e01b
Code: 0f 05 48 3d 00 f0 ff ff 77 45 c3 0f 1f 40 00 48 83 ec 18 89 7c 24 0c e8 63 fc ff ff 8b 7c 24 0c 41 89 c0 b8 03 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 35 44 89 c7 89 44 24 0c e8 a1 fc ff ff 8b 44
RSP: 002b:00007ffc821a2fb0 EFLAGS: 00000293 ORIG_RAX: 0000000000000003
RAX: 0000000000000000 RBX: 000000000000000a RCX: 00007fd7a583e01b
RDX: 0000000000000000 RSI: 00007fd7a5400000 RDI: 0000000000000009
RBP: 00007fd7a59ad980 R08: 0000000000000000 R09: 0000000071612a4e
R10: 0000000000000000 R11: 0000000000000293 R12: 000000000014ffb4
R13: 00007ffc821a30b0 R14: 00007fd7a59ac050 R15: 0000000000000032
</TASK>