Hi everyone,
I’m experiencing several errors during boot on Fedora 41. These issues seem to be related to ACPI, GNOME Keyring, and kernel memory management. Below are the logs from journalctl -b --priority=3:
Dec 07 17:03:04 fedora kernel: ACPI BIOS Error (bug): Failure creating named object [_SB.PCI0.GPP6.WLAN._S0W], AE_ALREADY_EXISTS
Dec 07 17:03:04 fedora kernel: ACPI Error: AE_ALREADY_EXISTS, During name lookup/catalog (20240322/psobject-220)
Dec 07 17:03:11 fedora kernel: ACPI BIOS Error (bug): Could not resolve symbol [_SB.PCI0.GPP0.PEGP.GPS.NVD1], AE_NOT_FOUND
Dec 07 17:03:11 fedora kernel: ACPI Error: Aborting method _SB.PCI0.GPP0.PEGP.GPS due to previous error (AE_NOT_FOUND)
Dec 07 17:03:11 fedora kernel: ACPI Error: Aborting method _SB.PCI0.GPP0.PEGP._DSM due to previous error (AE_NOT_FOUND)
Dec 07 17:03:28 fedora gdm-password][2294]: gkr-pam: unable to locate daemon control file
Dec 07 17:03:41 fedora abrt-notification[3475]: System encountered a non-fatal error in follow_pte()
Dec 07 17:03:42 fedora abrt-notification[3489]: System encountered a non-fatal error in follow_pte()
Dec 07 17:03:43 fedora abrt-notification[3506]: System encountered a non-fatal error in follow_pte()
The ACPI errors you can ignore, normal buggy BIOS stuff.
You should be able to use coredumpctl to find out what program is causing the follow_pte() problem. You might also find clues in sudo dmesg about the program that crashed.
ACPI Errors: There are often kernel command-line options that will eliminate some of these messages. They are often harmless, but clutter up the journalctl output so may cause you to miss important entries. On my Dell systems, I can use: acpi_osi=linux. Kernel Parameters acpi_osi=linux and acpi_backlight=vendor.
Gnome keyring: check that the keyring daemon starts properly. Here:
% journalctl --no-hostname -b -g gkr-pam|cat
Dec 06 15:03:50 gdm-password][11314]: gkr-pam: unable to locate daemon control file
Dec 06 15:03:50 gdm-password][11314]: gkr-pam: sta% journalctl --no-hostname -b -g gkr |cat
Dec 06 15:03:50 gdm-password][11314]: gkr-pam: unable to locate daemon control file
Dec 06 15:03:50 gdm-password][11314]: gkr-pam: stashed password to try later in open session
Dec 06 15:03:50 gdm-password][11314]: gkr-pam: gnome-keyring-daemon started properly and unlocked keyring
Dec 07 14:24:12 gdm-password][33972]: gkr-pam: unlocked login keyring
shed password to try later in open session
Dec 06 15:03:50 gdm-password][11314]: gkr-pam: gnome-keyring-daemon started properly and unlocked keyring
Dec 07 14:24:12 gdm-password][33972]: gkr-pam: unlocked login keyring
non-fatal error in follow_pte(): there should be more details in the “Problem Reporting” app. follow_pte is relatively new, so some software may not yet use it correctly. It is marked `GPL only", see Linux World - On the value of EXPORT_SYMBOL_GPL and comment by Linus Torvalds – I wonder if the non-fatal error is non-GPL code using a GPL symbol.
Thank you for your suggestions! I understand that ACPI errors are harmless and typically caused by a buggy BIOS, but they’re still quite frustrating as they clutter my logs. I tried adding the acpi_osi=linux parameter to the kernel options, but unfortunately, it didn’t resolve the issue – the errors keep appearing, and the logs remain cluttered.
Regarding the follow_pte() issue, it seems to be related to NVIDIA drivers. I’ve tried reinstalling the drivers, but it didn’t help, and I’m running out of ideas on how to fix this.