[Bug 219085] New: kvm_spurious_fault in L1 when running a nested kvm instance on AMD Opteron_G5_qemu L0

[bugzilla-daemon@kernel.org]

https://bugzilla.kernel.org/show_bug.cgi?id=219085

            Bug ID: 219085
           Summary: kvm_spurious_fault in L1 when running a nested kvm
                    instance on AMD Opteron_G5_qemu L0
           Product: Virtualization
           Version: unspecified
          Hardware: All
                OS: Linux
            Status: NEW
          Severity: normal
          Priority: P3
         Component: kvm
          Assignee: virtualization_kvm@kernel-bugs.osdl.org
          Reporter: ununpta@mailto.plus
        Regression: No

Hello,

when I try to run `qemu-system-x86_64 -accel kvm` in L1 bash in linux kvm guest
with kernel 6.10.0 x86_64, I get this message in L1 guest:

[  104.446685] kvm_amd: Nested Virtualization enabled
[  104.446688] kvm_amd: Nested Paging disabled
[  104.446690] kvm_amd: PMU virtualization is disabled
[  112.940705] clocksource: timekeeping watchdog on CPU0: hpet wd-wd read-back
delay of 50500ns
[  112.940746] clocksource: wd-tsc-wd read-back delay of 1385000ns, clock-skew
test skipped!
[  355.714362] unchecked MSR access error: WRMSR to 0xc0000080 (tried to write
0x0000000000001d01) at rIP: 0xffffffff9228a274 (native_write_msr+0x4/0x20)
[  355.714373] Call Trace:
[  355.714376]  <TASK>
[  355.714379]  ? ex_handler_msr+0xd3/0x150
[  355.714381]  ? fixup_exception+0x276/0x2e0
[  355.714383]  ? exc_general_protection+0x14f/0x440
[  355.714388]  ? asm_exc_general_protection+0x22/0x30
[  355.714391]  ? native_write_msr+0x4/0x20
[  355.714397]  svm_hardware_enable+0xd5/0x2f0 [kvm_amd]
[  355.714405]  kvm_arch_hardware_enable+0xc7/0x280 [kvm]
[  355.714469]  hardware_enable_nolock+0x1d/0x50 [kvm]
[  355.714489]  smp_call_function_many_cond+0xcf/0x4d0
[  355.714494]  ? kmalloc_trace_noprof+0x2c8/0x2f0
[  355.714497]  ? __pfx_hardware_enable_nolock+0x10/0x10 [kvm]
[  355.714516]  on_each_cpu_cond_mask+0x20/0x40
[  355.714517]  kvm_dev_ioctl+0x815/0xb40 [kvm]
[  355.714538]  __x64_sys_ioctl+0x93/0xd0
[  355.714542]  do_syscall_64+0x7e/0x190
[  355.714545]  ? kvm_dev_ioctl+0x2fb/0xb40 [kvm]
[  355.714564]  ? __schedule+0x3f3/0xb40
[  355.714566]  ? syscall_exit_to_user_mode+0x73/0x200
[  355.714567]  ? do_syscall_64+0x8a/0x190
[  355.714568]  ? do_syscall_64+0x8a/0x190
[  355.714569]  ? tomoyo_init_request_info+0x95/0xc0
[  355.714573]  ? tomoyo_path_number_perm+0x88/0x200
[  355.714576]  ? kvm_dev_ioctl+0x2fb/0xb40 [kvm]
[  355.714595]  ? syscall_exit_to_user_mode+0x73/0x200
[  355.714597]  ? syscall_exit_to_user_mode+0x73/0x200
[  355.714598]  ? do_syscall_64+0x8a/0x190
[  355.714599]  ? __count_memcg_events+0x54/0xf0
[  355.714601]  ? __rseq_handle_notify_resume+0xa4/0x4f0
[  355.714604]  ? handle_mm_fault+0xaa/0x320
[  355.714608]  ? restore_fpregs_from_fpstate+0x38/0x90
[  355.714611]  ? switch_fpu_return+0x4b/0xc0
[  355.714612]  entry_SYSCALL_64_after_hwframe+0x76/0x7e
[  355.714614] RIP: 0033:0x7fb24aab7c5b
[  355.714616] Code: 00 48 89 44 24 18 31 c0 48 8d 44 24 60 c7 04 24 10 00 00
00 48 89 44 24 08 48 8d 44 24 20 48 89 44 24 10 b8 10 00 00 00 0f 05 <89> c2 3d
00 f0 ff ff 77 1c 48 8b 44 24 18 64 48 2b 04 25 28 00 00
[  355.714617] RSP: 002b:00007ffee1205880 EFLAGS: 00000246 ORIG_RAX:
0000000000000010
[  355.714619] RAX: ffffffffffffffda RBX: 000000000000ae01 RCX:
00007fb24aab7c5b
[  355.714620] RDX: 0000000000000000 RSI: 000000000000ae01 RDI:
000000000000000a
[  355.714620] RBP: 000055b5ba0d2160 R08: 00007fb24ab8cc68 R09:
0000000000000006
[  355.714621] R10: 0000000000000000 R11: 0000000000000246 R12:
0000000000000000
[  355.714621] R13: 00007ffee1205b80 R14: 0000000000000000 R15:
00007ffee1205ac0
[  355.714622]  </TASK>
[  355.880539] ------------[ cut here ]------------
[  355.880542] kernel BUG at arch/x86/kvm/x86.c:510!
[  355.880548] Oops: invalid opcode: 0000 [#1] PREEMPT SMP NOPTI
[  355.880551] CPU: 0 PID: 1550 Comm: qemu-system-x86 Not tainted 6.10.0 #8
[  355.880553] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS
edk2-stable202402-prebuilt.qemu.org 02/14/2024
[  355.880554] RIP: 0010:kvm_spurious_fault+0xe/0x10 [kvm]
[  355.880584] Code: 00 00 85 c0 0f 95 c0 e9 90 79 e7 d1 90 90 90 90 90 90 90
90 90 90 90 90 90 90 90 90 80 3d f9 1c 02 00 00 74 05 e9 72 79 e7 d1 <0f> 0b 90
90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 66 90 e9 59
[  355.880586] RSP: 0018:ffffb618806fbc38 EFLAGS: 00010246
[  355.880587] RAX: 00000001025d0000 RBX: ffff94884d6c99b0 RCX:
0000000000000027
[  355.880588] RDX: 0000000000000003 RSI: 000000000188d000 RDI:
ffff94884d6c99b0
[  355.880589] RBP: 0000000000038060 R08: 0000000000000001 R09:
0000000000000027
[  355.880590] R10: 0000000000000001 R11: 0000000000400dc0 R12:
ffff9488bbc38060
[  355.880590] R13: 0000000000000000 R14: ffff9488411da000 R15:
0000000000000000
[  355.880591] FS:  00007fb2390006c0(0000) GS:ffff9488bbc00000(0000)
knlGS:0000000000000000
[  355.880592] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  355.880593] CR2: 00007fbe78a5e030 CR3: 000000010d630000 CR4:
0000000000350ef0
[  355.880595] Call Trace:
[  355.880598]  <TASK>
[  355.880599]  ? die+0x32/0x80
[  355.880603]  ? do_trap+0xd9/0x100
[  355.880605]  ? kvm_spurious_fault+0xe/0x10 [kvm]
[  355.880627]  ? do_error_trap+0x6a/0x90
[  355.880628]  ? kvm_spurious_fault+0xe/0x10 [kvm]
[  355.880648]  ? exc_invalid_op+0x4c/0x60
[  355.880652]  ? kvm_spurious_fault+0xe/0x10 [kvm]
[  355.880672]  ? asm_exc_invalid_op+0x16/0x20
[  355.880675]  ? kvm_spurious_fault+0xe/0x10 [kvm]
[  355.880695]  svm_prepare_switch_to_guest+0xe4/0x160 [kvm_amd]
[  355.880701]  kvm_arch_vcpu_ioctl_run+0x441/0x15b0 [kvm]
[  355.880729]  kvm_vcpu_ioctl+0x23d/0x6f0 [kvm]
[  355.880749]  ? check_preempt_wakeup_fair+0x136/0x1d0
[  355.880753]  __x64_sys_ioctl+0x93/0xd0
[  355.880757]  do_syscall_64+0x7e/0x190
[  355.880760]  ? wake_up_q+0x4a/0x90
[  355.880762]  ? futex_wake+0x155/0x190
[  355.880765]  ? do_futex+0xeb/0x1c0
[  355.880766]  ? __x64_sys_futex+0x8e/0x1d0
[  355.880767]  ? syscall_exit_to_user_mode+0x73/0x200
[  355.880769]  ? syscall_exit_to_user_mode+0x73/0x200
[  355.880770]  ? do_syscall_64+0x8a/0x190
[  355.880771]  ? do_syscall_64+0x8a/0x190
[  355.880772]  ? exc_page_fault+0x72/0x170
[  355.880773]  entry_SYSCALL_64_after_hwframe+0x76/0x7e
[  355.880775] RIP: 0033:0x7fb24aab7c5b
[  355.880776] Code: 00 48 89 44 24 18 31 c0 48 8d 44 24 60 c7 04 24 10 00 00
00 48 89 44 24 08 48 8d 44 24 20 48 89 44 24 10 b8 10 00 00 00 0f 05 <89> c2 3d
00 f0 ff ff 77 1c 48 8b 44 24 18 64 48 2b 04 25 28 00 00
[  355.880777] RSP: 002b:00007fb238fff530 EFLAGS: 00000246 ORIG_RAX:
0000000000000010
[  355.880778] RAX: ffffffffffffffda RBX: 000000000000ae80 RCX:
00007fb24aab7c5b
[  355.880779] RDX: 0000000000000000 RSI: 000000000000ae80 RDI:
000000000000000c
[  355.880780] RBP: 000055b5ba0d7e60 R08: 000055b5b32412d0 R09:
0000000000000000
[  355.880780] R10: 00007fb24ab2bf70 R11: 0000000000000246 R12:
0000000000000000
[  355.880781] R13: 0000000000000007 R14: 00007ffee1205360 R15:
00007fb238800000
[  355.880782]  </TASK>
[  355.880783] Modules linked in: kvm_amd ccp kvm qrtr rfkill binfmt_misc
nls_ascii nls_cp437 vfat fat crc32_pclmul ghash_clmulni_intel sha512_ssse3
sha256_ssse3 sha1_ssse3 aesni_intel crypto_simd iTCO_wdt cryptd intel_pmc_bxt
joydev iTCO_vendor_support pcspkr watchdog button sg evdev serio_raw parport_pc
ppdev lp parport fuse loop efi_pstore dm_mod configfs qemu_fw_cfg ip_tables
x_tables autofs4 ext4 crc16 mbcache jbd2 btrfs blake2b_generic efivarfs raid10
raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx libcrc32c
crc32c_generic xor raid6_pq raid1 raid0 md_mod hid_generic usbhid bochs
drm_vram_helper hid sd_mod t10_pi drm_kms_helper crc64_rocksoft crc64
crc_t10dif crct10dif_generic drm_ttm_helper ttm ahci libahci ehci_pci uhci_hcd
virtio_scsi libata ehci_hcd scsi_mod e1000e psmouse usbcore virtio_pci virtio
virtio_pci_legacy_dev virtio_pci_modern_dev crct10dif_pclmul crct10dif_common
crc32c_intel drm virtio_ring i2c_i801 lpc_ich usb_common scsi_common i2c_smbus
[last unloaded: ccp]
[  355.880835] ---[ end trace 0000000000000000 ]---
[  355.884034] RIP: 0010:kvm_spurious_fault+0xe/0x10 [kvm]
[  355.884060] Code: 00 00 85 c0 0f 95 c0 e9 90 79 e7 d1 90 90 90 90 90 90 90
90 90 90 90 90 90 90 90 90 80 3d f9 1c 02 00 00 74 05 e9 72 79 e7 d1 <0f> 0b 90
90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 66 90 e9 59
[  355.884062] RSP: 0018:ffffb618806fbc38 EFLAGS: 00010246
[  355.884063] RAX: 00000001025d0000 RBX: ffff94884d6c99b0 RCX:
0000000000000027
[  355.884064] RDX: 0000000000000003 RSI: 000000000188d000 RDI:
ffff94884d6c99b0
[  355.884064] RBP: 0000000000038060 R08: 0000000000000001 R09:
0000000000000027
[  355.884065] R10: 0000000000000001 R11: 0000000000400dc0 R12:
ffff9488bbc38060
[  355.884066] R13: 0000000000000000 R14: ffff9488411da000 R15:
0000000000000000
[  355.884066] FS:  00007fb2390006c0(0000) GS:ffff9488bbc00000(0000)
knlGS:0000000000000000
[  355.884067] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  355.884068] CR2: 00007fbe78a5e030 CR3: 000000010d630000 CR4:
0000000000350ef0
[  355.884069] note: qemu-system-x86[1550] exited with preempt_count 1

If I run `qemu-system-x86_64 -accel tcg` in L1 bash, it correctly boots into
qemu BIOS.

Any ideas about what could have caused it?

-- 
You may reply to this email to add a comment.

You are receiving this mail because:
You are watching the assignee of the bug.