#1029336 Regression: unable to mount root fs on unknown-block

Package:
linux-image-6.1.0-1-rt-amd64
Source:
linux-image-6.1.0-1-rt-amd64
Description:
Linux 6.1 for 64-bit PCs, PREEMPT_RT (signed)
Submitter:
James Addison
Date:
2023-01-21 19:39:06 UTC
Severity:
normal
Tags:
#1029336#5
Date:
2023-01-21 12:40:07 UTC
From:
To:
Dear Maintainer,

When attempting to boot a ThinkPad X230 from kernel build 6.1.0-1-rt-amd64, the
following error occurs (transcribed from screen; apologies in advance for any
typos):

  [...] No filesystem could mount root, tried :
  [...]
  [...] Kernel panic - not syncing: VFS: Unable to mount root fs on unknown-block(0,0)
  [...] CPU: 0 PID: 1 Comm: swapper/0 Not tainted 6.1.0-1-rt-amd64 #1  Debian 6.1.4-1
  [...] Hardware name: <disk and BIOS details>
  [...] Call Trace:
  [...]  <TASK>
  [...]  dump_stack_lvl+0x44/0x5c
  [...]  panic+0x10e/0x2db
  [...]  mount_block_root+0x1d3/0x1e6
  [...]  prepare_namespace+0x136/0x165
  [...]  kernel_init_freeable+0x25c/0x286
  [...]  ? rest_init+0xd0/0xd0
  [...]  kernel_init+0x16/0x130
  [...]  ret_from_fork+0x22/0x30
  [...]  </TASK>
  [...] Kernel Offset: 0x18600000 from 0xffffffff81000000 (relocation range: 0xffffffff80000000-0xffffffffbfffffff)

This prevents the system from booting.

Adding the 'rootwait' option to the kernel commandline as mentioned in a
similar-sounding upstream bug prevents the panic, but the system pauses,
seemingly forever, without finding the filesystem to mount.

The reason that I believe this is a regression is:

 - The problem does _not_ occur for the non-rt build (6.1.0-1-amd64)
 - The problem does _not_ occur with a recent rt build (6.0.0.6-rt-amd64)

Thank you,
James

#1029336#10
Date:
2023-01-21 13:51:33 UTC
From:
To:
Control:tags -1 + moreinfo

Can you please try with 6.1.7-1 which is present in unstable?

Regards,
Salvatore

#1029336#23
Date:
2023-01-21 18:26:19 UTC
From:
To:
Thanks, Salvatore - I can confirm that the issue is fixed after an install of
version 6.1.7-1 from 'unstable'.

The system _also_ boots fine after installing version 6.1.4-1 from 'testing'.
Perhaps there was a corrupt/misconfigured initrd build on the machine before.

Let me know if more details would be useful; the issue is resolved here.

#1029336#28
Date:
2023-01-21 19:38:19 UTC
From:
To:
Hi James,

Thanks for testing and for reporting back the above. Indeed I would
consider then this bug to be closed, doing so now.

If you still have the logs from the previous packages installation
(check /var/log/apt/term.log* for instance), then you can check if
there indeed was a problem on package installation and initrd update.

Regards,
Salvatore

#1029336#33
Date:
2023-01-21 19:38:19 UTC
From:
To:
Hi James,

Thanks for testing and for reporting back the above. Indeed I would
consider then this bug to be closed, doing so now.

If you still have the logs from the previous packages installation
(check /var/log/apt/term.log* for instance), then you can check if
there indeed was a problem on package installation and initrd update.

Regards,
Salvatore