What were you trying to do that didn't work?

Kickstart a new UEFI-based CentOS 9 System, booting normally for first time

What is the impact of this issue to you?

New c9s systems are unbootable after kickstart. I have to revert shim package to shim-x64-15-15.el8_2.x86_64.rpm or enter the UEFI shell and launching grubx64.efi by hand

Please provide the package NVR for which the bug is seen:

shim-x64-15.8-2.el9.x86_64

How reproducible is this bug?:

I have two Supermicro A1SRi / Intel Atom systems I can reliably reproduce this on. Secure Boot is not enabled, systems do not have TPMs

Steps to reproduce

1. Install c9s, which brings in the shim-x64-15.8-2.el9.x86_64 package
2. Reboot
3. Let system attempt to boot or go into UEFI shell and run fs0:\EFI\centos\shimx64.efi directly
4. System locks up, often require power reset

Expected results

Expect shim to load grub, and then load the kernel and boot normally.

Actual results

System locks up as soon as shimx64.efi, shim.efi, or shimx64-centos.efi are run

fs0:\EFI\centos> dir
Directory of: fs0:\EFI\centos

  09/22/25  10:47p <DIR>          4,096  .
  09/22/25  10:47p <DIR>          4,096  ..
  08/19/25  07:53a              856,552  mmx64.efi
  08/19/25  07:53a              947,880  shim.efi
  08/19/25  07:53a              949,456  shimx64-centos.efi
  08/19/25  07:53a              947,880  shimx64.efi
  08/19/25  07:53a                  108  BOOTX64.CSV
  09/22/25  10:58p                  144  grub.cfg
  09/09/25  05:31p            2,565,672  grubx64.efi
          7 File(s)   6,267,692 bytes
          2 Dir(s)
fs0:\EFI\centos> shimx64.efi

...no more output...

 If I'm in the UEFI shell I can skip shim and run grubx64.efi directly and this boots the system just fine. mmx64.efi works too, but all shim binaries lock up the system.

If I downgrade to shim-x64-15-15.el8_2.x86_64.rpm the system boots normally without any problems.