[Bug 2084664] [NEW] Unable to boot J-hwe-6.8 kernel on openstack ppc64el VM

Public bug reported:

This issue was found on Oct. 08 and affecting the s2024.09.02 and
2024.09.30 cycle.

The Jammy PowerPC VM on openstack refused to boot the J-hwe-6.8 kernel
installed by running `sudo apt install kernel-testing--linux-hwe-6.8--
full--generic`

This issue can be found on both scalingstack_bos01 and
scalingstack_bos02 region. For instances on scalingstack_bos02 you can
see the system failed to find the root device and panics:

[ 0.288925] clk: Disabling unused clocks
[ 0.289052] md: Waiting for all devices to be available before autodetect
[ 0.289096] md: If you don't use raid, use raid=noautodetect
[ 0.289137] md: Autodetecting RAID arrays.
[ 0.289163] md: autorun ...
[ 0.289181] md: ... autorun DONE.
[ 0.289351] /dev/root: Can't open blockdev
[ 0.289383] VFS: Cannot open root device "UUID=b0bdf9d3-cb0f-4096-8d38-9071b53e0b8c" or unknown-block(0,0): error -6
[ 0.289449] Please append a correct "root=" boot option; here are the available partitions:
[ 0.289501] 0800 20971520 sda
[ 0.289504] driver: sd
[ 0.289547] 0801 20962287 sda1 0004e4f7-ac99-42ec-a415-0cc181049bb1
[ 0.289549]
[ 0.289615] 0802 8192 sda2 8e56d7a3-2727-4f86-b2ba-8ed729683119
[ 0.289617]
[ 0.289684] List of all bdev filesystems:
[ 0.289709] ext3
[ 0.289710] ext2
[ 0.289728] ext4
[ 0.289745] squashfs
[ 0.289763] vfat
[ 0.289781] fuseblk
[ 0.289798]
[ 0.289833] Kernel panic - not syncing: VFS: Unable to mount root fs on unknown-block(0,0)
[ 0.289884] CPU: 1 PID: 1 Comm: swapper/0 Not tainted 6.8.0-45-generic #45~22.04.1-Ubuntu
[ 0.289935] Hardware name: IBM pSeries (emulated by qemu) POWER9 (raw) 0x4e1203 0xf000005 of:SLOF,HEAD hv:linux,kvm pSeries
[ 0.290001] Call Trace:
[ 0.290019] [c00000000738fb40] [c0000000016c20f8] dump_stack_lvl+0xe0/0x10c (unreliable)
[ 0.290075] [c00000000738fb70] [c0000000001933ec] panic+0x300/0x524
[ 0.290120] [c00000000738fc10] [c00000000300ce6c] mount_root_generic+0x208/0x444
[ 0.290173] [c00000000738fce0] [c00000000300d2c8] prepare_namespace+0x98/0x430
[ 0.290225] [c00000000738fd70] [c00000000300c220] kernel_init_freeable+0x32c/0x37c
[ 0.290278] [c00000000738fde0] [c00000000001145c] kernel_init+0x34/0x2a8
[ 0.290322] [c00000000738fe50] [c00000000000debc] ret_from_kernel_user_thread+0x14/0x1c
[ 0.290374] --- interrupt: 0 at 0x0
[ 0.292503] pstore: backend (nvram) writing error (-1)

This can even be reproduced with an older kernel that used to work, just
deploy and Jammy PowerPC vm on openstack and install the 6.8 kernel from
-updates pocket.

From Stefan's investigation, ppc64el in openstack boots with 6.8 if we
only install the virtual subset (linux-image-virtual-hwe-22.04)

** Affects: ubuntu-kernel-tests
Importance: Undecided
Status: New

** Affects: linux-meta-hwe-6.8 (Ubuntu)
Importance: Undecided
Status: New


** Tags: 6.8 jammy ppc64el sru-20240930 sru-s20240902

** Also affects: linux (Ubuntu)
Importance: Undecided
Status: New

** Package changed: linux (Ubuntu) => linux-meta-hwe-6.8 (Ubuntu)

--
You received this bug notification because you are subscribed to linux
in Ubuntu.
Matching subscriptions: Bgg, Bmail, Nb
https://bugs.launchpad.net/bugs/2084664

Title:
Unable to boot J-hwe-6.8 kernel on openstack ppc64el VM

Status in ubuntu-kernel-tests:
New
Status in linux-meta-hwe-6.8 package in Ubuntu:
New

Bug description:
This issue was found on Oct. 08 and affecting the s2024.09.02 and
2024.09.30 cycle.

The Jammy PowerPC VM on openstack refused to boot the J-hwe-6.8 kernel
installed by running `sudo apt install kernel-testing--linux-hwe-6.8--
full--generic`

This issue can be found on both scalingstack_bos01 and
scalingstack_bos02 region. For instances on scalingstack_bos02 you can
see the system failed to find the root device and panics:

[ 0.288925] clk: Disabling unused clocks
[ 0.289052] md: Waiting for all devices to be available before autodetect
[ 0.289096] md: If you don't use raid, use raid=noautodetect
[ 0.289137] md: Autodetecting RAID arrays.
[ 0.289163] md: autorun ...
[ 0.289181] md: ... autorun DONE.
[ 0.289351] /dev/root: Can't open blockdev
[ 0.289383] VFS: Cannot open root device "UUID=b0bdf9d3-cb0f-4096-8d38-9071b53e0b8c" or unknown-block(0,0): error -6
[ 0.289449] Please append a correct "root=" boot option; here are the available partitions:
[ 0.289501] 0800 20971520 sda
[ 0.289504] driver: sd
[ 0.289547] 0801 20962287 sda1 0004e4f7-ac99-42ec-a415-0cc181049bb1
[ 0.289549]
[ 0.289615] 0802 8192 sda2 8e56d7a3-2727-4f86-b2ba-8ed729683119
[ 0.289617]
[ 0.289684] List of all bdev filesystems:
[ 0.289709] ext3
[ 0.289710] ext2
[ 0.289728] ext4
[ 0.289745] squashfs
[ 0.289763] vfat
[ 0.289781] fuseblk
[ 0.289798]
[ 0.289833] Kernel panic - not syncing: VFS: Unable to mount root fs on unknown-block(0,0)
[ 0.289884] CPU: 1 PID: 1 Comm: swapper/0 Not tainted 6.8.0-45-generic #45~22.04.1-Ubuntu
[ 0.289935] Hardware name: IBM pSeries (emulated by qemu) POWER9 (raw) 0x4e1203 0xf000005 of:SLOF,HEAD hv:linux,kvm pSeries
[ 0.290001] Call Trace:
[ 0.290019] [c00000000738fb40] [c0000000016c20f8] dump_stack_lvl+0xe0/0x10c (unreliable)
[ 0.290075] [c00000000738fb70] [c0000000001933ec] panic+0x300/0x524
[ 0.290120] [c00000000738fc10] [c00000000300ce6c] mount_root_generic+0x208/0x444
[ 0.290173] [c00000000738fce0] [c00000000300d2c8] prepare_namespace+0x98/0x430
[ 0.290225] [c00000000738fd70] [c00000000300c220] kernel_init_freeable+0x32c/0x37c
[ 0.290278] [c00000000738fde0] [c00000000001145c] kernel_init+0x34/0x2a8
[ 0.290322] [c00000000738fe50] [c00000000000debc] ret_from_kernel_user_thread+0x14/0x1c
[ 0.290374] --- interrupt: 0 at 0x0
[ 0.292503] pstore: backend (nvram) writing error (-1)

This can even be reproduced with an older kernel that used to work,
just deploy and Jammy PowerPC vm on openstack and install the 6.8
kernel from -updates pocket.

From Stefan's investigation, ppc64el in openstack boots with 6.8 if we
only install the virtual subset (linux-image-virtual-hwe-22.04)

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu-kernel-tests/+bug/2084664/+subscriptions