[syzbot] [kvmarm?] [kvm?] BUG: unable to handle kernel paging request in kvm_vgic_destroy

[syzbot]

Hello,

syzbot found the following issue on:

HEAD commit: 6316366129d2 Merge branch kvm-arm64/misc-6.20 into kvmarm-..
git tree: git://git.kernel.org/pub/scm/linux/kernel/git/kvmarm/kvmarm.git next
console output: https://syzkaller.appspot.com/x/log.txt?x=15e59c4a580000
kernel config: https://syzkaller.appspot.com/x/.config?x=148fc9aa8e041d0a
dashboard link: https://syzkaller.appspot.com/bug?extid=f6a46b038fc243ac0175
compiler: Debian clang version 21.1.8 (++20251221033036+2078da43e25a-1~exp1~20251221153213.50), Debian LLD 21.1.8
userspace arch: arm64
syz repro: https://syzkaller.appspot.com/x/repro.syz?x=13182006580000
C reproducer: https://syzkaller.appspot.com/x/repro.c?x=173900ba580000

Downloadable assets:
disk image (non-bootable): https://storage.googleapis.com/syzbot-assets/fa3fbcfdac58/non_bootable_disk-63163661.raw.xz
vmlinux: https://storage.googleapis.com/syzbot-assets/1018400deda3/vmlinux-63163661.xz
kernel image: https://storage.googleapis.com/syzbot-assets/fb8a8bb5d8a4/Image-63163661.gz.xz

IMPORTANT: if you fix the issue, please add the following tag to the commit:
Reported-by: syzbot+f6a46b...@syzkaller.appspotmail.com

__invoke_syscall arch/arm64/kernel/syscall.c:35 [inline]
invoke_syscall+0x90/0x230 arch/arm64/kernel/syscall.c:49
el0_svc_common+0x120/0x2f4 arch/arm64/kernel/syscall.c:132
do_el0_svc+0x58/0x74 arch/arm64/kernel/syscall.c:151
el0_svc+0x5c/0x238 arch/arm64/kernel/entry-common.c:724
el0t_64_sync_handler+0x84/0x12c arch/arm64/kernel/entry-common.c:743
el0t_64_sync+0x198/0x19c arch/arm64/kernel/entry.S:596
Unable to handle kernel paging request at virtual address ffef800000000000
KASAN: maybe wild-memory-access in range [0xff00000000000000-0xff0000000000000f]
Mem abort info:
ESR = 0x0000000096000004
EC = 0x25: DABT (current EL), IL = 32 bits
SET = 0, FnV = 0
EA = 0, S1PTW = 0
FSC = 0x04: level 0 translation fault
Data abort info:
ISV = 0, ISS = 0x00000004, ISS2 = 0x00000000
CM = 0, WnR = 0, TnD = 0, TagAccess = 0
GCS = 0, Overlay = 0, DirtyBit = 0, Xs = 0
[ffef800000000000] address between user and kernel address ranges
Internal error: Oops: 0000000096000004 [#1] SMP
Modules linked in:
CPU: 0 UID: 0 PID: 3651 Comm: syz.2.17 Not tainted syzkaller #0 PREEMPT
Hardware name: linux,dummy-virt (DT)
pstate: 01402009 (nzcv daif +PAN -UAO -TCO +DIT -SSBS BTYPE=--)
pc : kvm_vgic_dist_destroy arch/arm64/kvm/vgic/vgic-init.c:445 [inline]
pc : kvm_vgic_destroy+0x2d4/0x624 arch/arm64/kvm/vgic/vgic-init.c:518
lr : kvm_vgic_dist_destroy arch/arm64/kvm/vgic/vgic-init.c:444 [inline]
lr : kvm_vgic_destroy+0x290/0x624 arch/arm64/kvm/vgic/vgic-init.c:518
sp : ffff80008e647b90
x29: ffff80008e647ba0 x28: 0000000000000005 x27: cdf00000200a52d8
x26: cdf00000200a4db0 x25: 00000000000000cd x24: cdf00000200a4d8c
x23: 00000000000000cd x22: 00000000000000cd x21: cdf00000200a4ad0
x20: efff800000000000 x19: cdf00000200a4000 x18: 00000000030f4b63
x17: 0000000000000031 x16: 0000000000000000 x15: ffff800088209a68
x14: ffffffffffffffff x13: 0000000000000028 x12: 5df000001795c1f0
x11: ffff800088209a68 x10: 0000000000ff0100 x9 : 0ff0000000000000
x8 : 0000000000000000 x7 : ffff80008672f958 x6 : 0000000000000000
x5 : 0000000000000000 x4 : 0000000000000000 x3 : 0000000000000002
x2 : 0000000000000008 x1 : 0000000000000000 x0 : 0000000000000007
Call trace:
kvm_vgic_dist_destroy arch/arm64/kvm/vgic/vgic-init.c:445 [inline] (P)
kvm_vgic_destroy+0x2d4/0x624 arch/arm64/kvm/vgic/vgic-init.c:518 (P)
kvm_arch_destroy_vm+0x88/0x138 arch/arm64/kvm/arm.c:299
kvm_destroy_vm virt/kvm/kvm_main.c:1317 [inline]
kvm_put_kvm+0x778/0xbe0 virt/kvm/kvm_main.c:1354
kvm_vm_release+0x58/0x78 virt/kvm/kvm_main.c:1377
__fput+0x4ac/0x978 fs/file_table.c:468
____fput+0x20/0x58 fs/file_table.c:496
task_work_run+0x1b8/0x250 kernel/task_work.c:233
resume_user_mode_work include/linux/resume_user_mode.h:50 [inline]
__exit_to_user_mode_loop kernel/entry/common.c:44 [inline]
exit_to_user_mode_loop+0x110/0x188 kernel/entry/common.c:75
__exit_to_user_mode_prepare include/linux/irq-entry-common.h:226 [inline]
exit_to_user_mode_prepare_legacy include/linux/irq-entry-common.h:242 [inline]
arm64_exit_to_user_mode arch/arm64/kernel/entry-common.c:81 [inline]
el0_svc+0x17c/0x238 arch/arm64/kernel/entry-common.c:725
el0t_64_sync_handler+0x84/0x12c arch/arm64/kernel/entry-common.c:743
el0t_64_sync+0x198/0x19c arch/arm64/kernel/entry.S:596
Code: 54000420 b2481c28 d344fd09 d378fc28 (38696a89)
---[ end trace 0000000000000000 ]---
----------------
Code disassembly (best guess):
0: 54000420 b.eq 0x84 // b.none
4: b2481c28 orr x8, x1, #0xff00000000000000
8: d344fd09 lsr x9, x8, #4
c: d378fc28 lsr x8, x1, #56
* 10: 38696a89 ldrb w9, [x20, x9] <-- trapping instruction


---
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 syzk...@googlegroups.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 syzbot to run the reproducer, reply with:
#syz test: git://repo/address.git branch-or-commit-hash
If you attach or paste a git patch, syzbot will apply it before testing.

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

[Marc Zyngier]

Oh gawd, fault injection. Because we didn't have enough bona fide,
directly triggerable bugs, we're tricking the kernel into generating
more. Oh well.

Thankfully, that's an easy one: vgic_allocate_private_irqs_locked()
fails, we exit kvm_vgic_create() early, leaving dist->rd_regions
uninitialised. kvm_vgic_dist_destroy() comes along and walks into the
weeds.

Note to the syzcaller folks: being a lazy bastard, I run the test case
(both kernel and C reproducer) as a nested guest using kvmtool.
kvmtool has a very simple init that doesn't mount debugfs by default.
It'd be great if the reproducer could check that the debugfs files are
accessible and stop if it can't configure them. I initially couldn't
reproduce the issue because of this.

Anyway, that being said:

#syz test: git://git.kernel.org/pub/scm/linux/kernel/git/maz/arm-platforms.git kvm-arm64/vgic-fixes-7.0

Thanks,

M.

--
Jazz isn't dead. It just smells funny.

[syzbot]

Hello,

syzbot has tested the proposed patch and the reproducer did not trigger any issue:

Reported-by: syzbot+f6a46b...@syzkaller.appspotmail.com
Tested-by: syzbot+f6a46b...@syzkaller.appspotmail.com

Tested on:

commit: 949862c2 KVM: arm64: Eagerly init vgic dist/redist on ..
git tree: git://git.kernel.org/pub/scm/linux/kernel/git/maz/arm-platforms.git kvm-arm64/vgic-fixes-7.0
console output: https://syzkaller.appspot.com/x/log.txt?x=15f88952580000
kernel config: https://syzkaller.appspot.com/x/.config?x=31d0db0315ae5d07

dashboard link: https://syzkaller.appspot.com/bug?extid=f6a46b038fc243ac0175
compiler: Debian clang version 21.1.8 (++20251221033036+2078da43e25a-1~exp1~20251221153213.50), Debian LLD 21.1.8
userspace arch: arm64

Note: no patches were applied.
Note: testing is done by a robot and is best-effort only.