kexec issues with backported kernels

[Mike Purvis]

Hi Core developers,

I'm having an issue trying to kexec to a HWE kernel in xenial (from a
running xenial system). Basically I do the kexec -l / kexec -e thing, but
instead of getting log output from the new kernel, my system reboots back
to the BIOS.

The kernel in question is the one from this package:

https://packages.ubuntu.com/xenial/linux-image-4.15.0-45-generic
(4.15.0-45.48~16.04.1)

Curiously, though, the exact same kernel from Bionic works just fine:

https://packages.ubuntu.com/bionic-updates/amd64/linux-image-4.15.0-45-generic
(4.15.0-45.48)

The two files are almost identical, though the Bionic one is about 2%
larger:

-rw------- 8128088 Jan 30 10:08 vmlinuz-4.15.0-45-generic-xenial
-rw------- 8281848 Jan 29 16:11 vmlinuz-4.15.0-45-generic-bionic

This looks similar to what is discussed on the Arch wiki here
<https://wiki.archlinux.org/index.php/kexec#System_hangs_or_reboots_after_%22kexec_core:_Starting_new_kernel%22>,
but the suggested fix is not applicable in my case as I'm using legacy boot
rather than EFI.

I had a look at the DSCs for the two packages, but unfortunately there's a
lot of magic going on and I'm not seeing how to quickly tell if/what is any
different about the two. Like, are there additional patches applied on the
16.04 version of the kernel, or is literally the only difference the
version of GCC used?

Thanks for any insight or pointers on next steps!

Mike
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.ubuntu.com/archives/ubuntu-devel-discuss/attachments/20190207/a6e3a691/attachment.html>