2022-08-16 19:08:17

by Naresh Kamboju

[permalink] [raw]
Subject: [next] arm64: kernel BUG at fs/inode.c:622 - Internal error: Oops - BUG: 0 - pc : clear_inode

Following kernel BUG found while booting arm64 Qcom dragonboard 410c with
Linux next-20220816 kernel Image.

Reported-by: Linux Kernel Functional Testing <[email protected]>

metadata:
git_ref: master
git_repo: https://gitlab.com/Linaro/lkft/mirrors/next/linux-next
git_sha: e1084bacab44f570691c0fdaa1259acf93ed0098
git_describe: next-20220816
kernel_version: 6.0.0-rc1
kernel-config: https://builds.tuxbuild.com/2DQCqsegDOgC0htScIz07RRDr7J/config
artifact-location: https://builds.tuxbuild.com/2DQCqsegDOgC0htScIz07RRDr7J
vmlinux: https://builds.tuxbuild.com/2DQCqsegDOgC0htScIz07RRDr7J/vmlinux.xz
System.map: https://builds.tuxbuild.com/2DQCqsegDOgC0htScIz07RRDr7J/System.map
toolchain: gcc-11

[ 17.950163] kernel BUG at fs/inode.c:622!
[ 17.954719] Internal error: Oops - BUG: 0 [#6] PREEMPT SMP
[ 17.959290] Modules linked in: rfkill snd_soc_hdmi_codec venus_enc
venus_dec videobuf2_dma_contig pm8916_wdt crct10dif_ce qcom_wcnss_pil
adv7511 cec msm qrtr qcom_camss gpu_sched drm_dp_aux_bus venus_core
display_connector videobuf2_dma_sg drm_display_helper qcom_q6v5_mss
v4l2_fwnode qcom_pil_info snd_soc_lpass_apq8016 qcom_q6v5
snd_soc_lpass_cpu v4l2_async snd_soc_msm8916_digital drm_kms_helper
snd_soc_apq8016_sbc qcom_sysmon v4l2_mem2mem snd_soc_qcom_common
snd_soc_lpass_platform qcom_spmi_temp_alarm snd_soc_msm8916_analog
videobuf2_memops rtc_pm8xxx qcom_common videobuf2_v4l2 qcom_spmi_vadc
qcom_glink_smem qcom_vadc_common qcom_pon qmi_helpers videobuf2_common
mdt_loader i2c_qcom_cci drm qcom_stats qnoc_msm8916 qcom_rng
icc_smd_rpm socinfo rmtfs_mem fuse
[ 18.023315] CPU: 3 PID: 457 Comm: systemd-udevd Tainted: G D
6.0.0-rc1-next-20220816 #1
[ 18.030656] Hardware name: Qualcomm Technologies, Inc. APQ 8016 SBC (DT)
[ 18.039973] pstate: 80000005 (Nzcv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
[ 18.046925] pc : clear_inode+0x88/0x9c
[ 18.053570] lr : clear_inode+0x3c/0x9c
[ 18.058064] sp : ffff80000c9f3c30
[ 18.062518] x29: ffff80000c9f3c30 x28: ffff000004f19080 x27: 0000000000000000
[ 18.067099] x26: ffff000005b2ad90 x25: 0000000000000002 x24: 00000000ffffffec
[ 18.073181] x23: ffff80000aaaea80 x22: ffff000005b2ad58 x21: ffff0000037c3340
[ 18.077364] ax88179_178a 1-1.2:1.0 eth0: ax88179 - Link status is: 1
[ 18.080288] x20: ffff000005b2af10 x19: ffff000005b2ad90 x18: 0000000000000000
[ 18.080306] x17: 0000000000000000 x16: 0000000000000000 x15: 0000000000000000
[ 18.080319] x14: 0000ffffffffffff x13: 0000000000000000
[ 18.094909] IPv6: ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready
[ 18.096330] x12: fefefefefefefeff
[ 18.131891] x11: 0000fffffffffff8 x10: ffff00000a0776c0 x9 : ffff800009443f98
[ 18.136734] x8 : ffff80000c9f3d08 x7 : 0000000000000000 x6 : 0000000000000001
[ 18.142836] x5 : ffff80000aaaf000 x4 : ffff80000aaaf2e8 x3 : 0000000000000000
[ 18.149910] x2 : ffff000005043040 x1 : ffff000005b2af90 x0 : ffff000005b2afb0
[ 18.157031] Call trace:
[ 18.164169] clear_inode+0x88/0x9c
[ 18.169006] shmem_evict_inode+0xa4/0x29c
[ 18.173853] evict+0xac/0x190
[ 18.178674] iput+0x174/0x240
[ 18.183499] do_unlinkat+0x1c4/0x270
[ 18.188333] __arm64_sys_unlinkat+0x4c/0x90
[ 18.193195] invoke_syscall+0x50/0x120
[ 18.198036] el0_svc_common.constprop.0+0x104/0x124
[ 18.202891] do_el0_svc+0x3c/0xd0
[ 18.207775] el0_svc+0x30/0x94
[ 18.212631] el0t_64_sync_handler+0xc0/0x13c
[ 18.217542] el0t_64_sync+0x18c/0x190
[ 18.222401] Code: a8c27bfd d50323bf d65f03c0 d4210000 (d4210000)
[ 18.227277] ---[ end trace 0000000000000000 ]---
[ 18.232281] note: systemd-udevd[457] exited with preempt_count 1
[ 18.287738] ------------[ cut here ]------------
[ 18.292616] WARNING: CPU: 3 PID: 0 at kernel/context_tracking.c:128
ct_kernel_exit.constprop.0+0x108/0x120
[ 18.297548] Modules linked in: rfkill snd_soc_hdmi_codec venus_enc
venus_dec videobuf2_dma_contig pm8916_wdt crct10dif_ce qcom_wcnss_pil
adv7511 cec msm qrtr qcom_camss gpu_sched drm_dp_aux_bus venus_core
display_connector videobuf2_dma_sg drm_display_helper qcom_q6v5_mss
v4l2_fwnode qcom_pil_info snd_soc_lpass_apq8016 qcom_q6v5
snd_soc_lpass_cpu v4l2_async snd_soc_msm8916_digital drm_kms_helper
snd_soc_apq8016_sbc qcom_sysmon v4l2_mem2mem snd_soc_qcom_common
snd_soc_lpass_platform qcom_spmi_temp_alarm snd_soc_msm8916_analog
videobuf2_memops rtc_pm8xxx qcom_common videobuf2_v4l2 qcom_spmi_vadc
qcom_glink_smem qcom_vadc_common qcom_pon qmi_helpers videobuf2_common
mdt_loader i2c_qcom_cci drm qcom_stats qnoc_msm8916 qcom_rng
icc_smd_rpm socinfo rmtfs_mem fuse
[ 18.368071] CPU: 3 PID: 0 Comm: swapper/3 Tainted: G D
6.0.0-rc1-next-20220816 #1
[ 18.373948] Hardware name: Qualcomm Technologies, Inc. APQ 8016 SBC (DT)
[ 18.383033] pstate: 200000c5 (nzCv daIF -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
[ 18.389948] pc : ct_kernel_exit.constprop.0+0x108/0x120
[ 18.397848] lr : ct_idle_enter+0x14/0x20
[ 18.404845] sp : ffff80000b0c3d60
[ 18.410655] x29: ffff80000b0c3d60 x28: 0000000000000000 x27: 0000000000000000
[ 18.417235] x26: 0000000000000000 x25: 0000000000000000 x24: 00000004420862ef
[ 18.424585] x23: 0000000000000000 x22: ffff000004cf2880 x21: ffff00003fca5940
[ 18.431994] x20: ffff000004cf2898 x19: ffff00003fca3b00 x18: 0000000000000000
[ 18.439447] x17: 0000000000000000 x16: ffff00003fc89040 x15: 0000000000000000
[ 18.446941] x14: 0000000000000003 x13: 0000000000000000 x12: 0000000000000000
[ 18.454423] x11: 0000000000000001 x10: 0000000000000c00 x9 : ffff80000901f4f0
[ 18.461904] x8 : ffff000003704e60 x7 : 0000000000000000 x6 : 000000000523b3e8
[ 18.469278] x5 : 4000000000000002 x4 : ffff800035adc000 x3 : ffff80000b0c3d60
[ 18.476594] x2 : ffff80000a1c7b00 x1 : 4000000000000000 x0 : ffff80000a1c7b00
[ 18.483850] Call trace:
[ 18.491111] ct_kernel_exit.constprop.0+0x108/0x120
[ 18.498465] ct_idle_enter+0x14/0x20
[ 18.505793] cpuidle_enter_state+0x2cc/0x4c0
[ 18.513112] cpuidle_enter+0x44/0x60
[ 18.518978] do_idle+0x24c/0x2e0
[ 18.524950] cpu_startup_entry+0x34/0x40
[ 18.531686] secondary_start_kernel+0x13c/0x150
[ 18.538141] __secondary_switched+0xb0/0xb4
[ 18.544315] ---[ end trace 0000000000000000 ]---
[ 19.722064] mmcblk1: p1
[ 19.740791] mmcblk1: p1
[ 19.759017] mmcblk1: p1
[ 19.769040] EXT4-fs (mmcblk0p14): resizing filesystem from 652544
to 1826299 blocks
[ 19.852550] EXT4-fs (mmcblk0p14): resized filesystem to 1826299


ref:
https://lkft.validation.linaro.org/scheduler/job/5391332#L2911
https://qa-reports.linaro.org/lkft/linux-next-master/build/next-20220816/testrun/11378204/suite/log-parser-boot/tests/

--
Linaro LKFT
https://lkft.linaro.org


2022-08-16 19:19:12

by Linus Torvalds

[permalink] [raw]
Subject: Re: [next] arm64: kernel BUG at fs/inode.c:622 - Internal error: Oops - BUG: 0 - pc : clear_inode

On Tue, Aug 16, 2022 at 12:00 PM Naresh Kamboju
<[email protected]> wrote:
>
> Following kernel BUG found while booting arm64 Qcom dragonboard 410c with
> Linux next-20220816 kernel Image.

What kind of environment is this?

Havign that inode list corruption makes it smell a *bit* like the
crazy memory corruption that we saw with the google cloud instances,
but that would only happen wif you actually use VIRTIO for your
environment?

Do you see the same issue with plain v6.0-rc1?

Linus

2022-08-16 19:59:09

by Linus Torvalds

[permalink] [raw]
Subject: Re: [next] arm64: kernel BUG at fs/inode.c:622 - Internal error: Oops - BUG: 0 - pc : clear_inode

On Tue, Aug 16, 2022 at 12:39 PM Naresh Kamboju
<[email protected]> wrote:
>
> This is a physical hardware db410c device.
> Following VIRTIO configs enabled.

Yeah, it's not enough to just enable the VIRTIO support, it actually
needs to run in a virt environment, and with a hypervisor that gets
confused by the virtio changes.

Plus:

> > Do you see the same issue with plain v6.0-rc1?
>
> Nope. I do not notice reported BUG on v6.0-rc1.

Ok, so definitely not related to the google cloud issue we had in rc1.

Linus

2022-08-16 20:17:36

by Naresh Kamboju

[permalink] [raw]
Subject: Re: [next] arm64: kernel BUG at fs/inode.c:622 - Internal error: Oops - BUG: 0 - pc : clear_inode

On Wed, 17 Aug 2022 at 00:40, Linus Torvalds
<[email protected]> wrote:
>
> On Tue, Aug 16, 2022 at 12:00 PM Naresh Kamboju
> <[email protected]> wrote:
> >
> > Following kernel BUG found while booting arm64 Qcom dragonboard 410c with
> > Linux next-20220816 kernel Image.
>
> What kind of environment is this?
>
> Havign that inode list corruption makes it smell a *bit* like the
> crazy memory corruption that we saw with the google cloud instances,
> but that would only happen wif you actually use VIRTIO for your
> environment?

This is a physical hardware db410c device.
Following VIRTIO configs enabled.

CONFIG_BLK_MQ_VIRTIO=y
CONFIG_NET_9P_VIRTIO=y
CONFIG_VIRTIO_BLK=y
CONFIG_SCSI_VIRTIO=y
CONFIG_VIRTIO_NET=y
CONFIG_VIRTIO_CONSOLE=y
CONFIG_VIRTIO_ANCHOR=y
CONFIG_VIRTIO=y
CONFIG_VIRTIO_PCI_LIB=y
CONFIG_VIRTIO_PCI_LIB_LEGACY=y
CONFIG_VIRTIO_MENU=y
CONFIG_VIRTIO_PCI=y
CONFIG_VIRTIO_PCI_LEGACY=y
CONFIG_VIRTIO_BALLOON=y
CONFIG_VIRTIO_MMIO=y
CONFIG_CRYPTO_DEV_VIRTIO=y


>
> Do you see the same issue with plain v6.0-rc1?

Nope. I do not notice reported BUG on v6.0-rc1.

- Naresh

2022-08-17 02:57:21

by Stephen Rothwell

[permalink] [raw]
Subject: Re: [next] arm64: kernel BUG at fs/inode.c:622 - Internal error: Oops - BUG: 0 - pc : clear_inode

Hi Naresh,

On Wed, 17 Aug 2022 01:09:40 +0530 Naresh Kamboju <[email protected]> wrote:
>
> On Wed, 17 Aug 2022 at 00:40, Linus Torvalds
> <[email protected]> wrote:
> >
> > On Tue, Aug 16, 2022 at 12:00 PM Naresh Kamboju
> > <[email protected]> wrote:
> > >
> > > Following kernel BUG found while booting arm64 Qcom dragonboard 410c with
> > > Linux next-20220816 kernel Image.
> >
> > What kind of environment is this?
> >
> > Havign that inode list corruption makes it smell a *bit* like the
> > crazy memory corruption that we saw with the google cloud instances,
> > but that would only happen wif you actually use VIRTIO for your
> > environment?
>
> This is a physical hardware db410c device.
> Following VIRTIO configs enabled.
>
> CONFIG_BLK_MQ_VIRTIO=y
> CONFIG_NET_9P_VIRTIO=y
> CONFIG_VIRTIO_BLK=y
> CONFIG_SCSI_VIRTIO=y
> CONFIG_VIRTIO_NET=y
> CONFIG_VIRTIO_CONSOLE=y
> CONFIG_VIRTIO_ANCHOR=y
> CONFIG_VIRTIO=y
> CONFIG_VIRTIO_PCI_LIB=y
> CONFIG_VIRTIO_PCI_LIB_LEGACY=y
> CONFIG_VIRTIO_MENU=y
> CONFIG_VIRTIO_PCI=y
> CONFIG_VIRTIO_PCI_LEGACY=y
> CONFIG_VIRTIO_BALLOON=y
> CONFIG_VIRTIO_MMIO=y
> CONFIG_CRYPTO_DEV_VIRTIO=y
>
>
> >
> > Do you see the same issue with plain v6.0-rc1?
>
> Nope. I do not notice reported BUG on v6.0-rc1.

Is it reliable enough that you could possibly do a bisection between
v6.0-rc1 and next-20220816?

--
Cheers,
Stephen Rothwell


Attachments:
(No filename) (499.00 B)
OpenPGP digital signature