You are not logged in.

#1 2016-07-27 20:09:32

hinrikh
Member
Registered: 2016-07-27
Posts: 5

[SOLVED] intel-ucode-20160714-1 causing "EFI stub: ERROR:

Just wanted to leave a note somewhere in case someone experienced similar problems. This morning I was staring at:

EFI stub: ERROR: failed to read file.
Trying to load files to higher address
EFI stub: ERROR: failed to read file.

After some research I loaded a USB flashdrive with a fresh archlinux-2016.07.01-dual.iso to boot from:

  • then oriented myself with lsblk -f and looked at /etc/fstab to refresh my memory on my organization (configured the system 10 months ago).

  • proceeded to mount the necessary devices on /mnt (e.g. /mnt, /mnt/boot, /mnt/var).

  • then arch-chroot /mnt /bin/bash into the system

  • navigated to /var/cache/pacman/pkg and determined that my previous microcode was intel-ucode-20160607-2

  • proceeded to downgrade the microcode with pacman -U intel-ucode-20160607-2-any.pkg.tar.xz

  • exited the chroot (with exit), rebooted and was able to boot again.

So apparently I was using pacman -Syu too wrecklessly.

Lenovo Thinkpad W540 Intel(R) Core(TM) i7-4900MQ CPU @ 2.80GHz - Dualboot (Windows 10) with systemd-boot (UEFI)

$ uname -a
Linux xxxxxxxxxx 4.6.4-1-ARCH #1 SMP PREEMPT Mon Jul 11 19:12:32 CEST 2016 x86_64 GNU/Linux

Offline

#2 2016-07-27 20:23:57

alphaniner
Member
From: Ancapistan
Registered: 2010-07-12
Posts: 2,810

Re: [SOLVED] intel-ucode-20160714-1 causing "EFI stub: ERROR:

If you had to downgrade a package, I don't think you're guilty of reckless behaviour unless you ignored an error or something.

Now that you know how to repair, maybe upgrade again to see if you can duplicate the failure. Or, depending on how easily configurable your bootloader is, you could extract the new ucode to /boot/intel-ucode-new.img or such to test so you wouldn't have to repair again.

Once you're sure the new ucode causes boot failure, a bug report may be in order.


But whether the Constitution really be one thing, or another, this much is certain - that it has either authorized such a government as we have had, or has been powerless to prevent it. In either case, it is unfit to exist.
-Lysander Spooner

Offline

#3 2016-07-27 22:06:00

hinrikh
Member
Registered: 2016-07-27
Posts: 5

Re: [SOLVED] intel-ucode-20160714-1 causing "EFI stub: ERROR:

Thank You for your response.

According to your advice I let the upgrade of the microcode go ahead once again - and big surprise - no problems.

The only warning that came up was this:

[ALPM] warning: directory permissions differ on /boot/ filesystem: 700  package: 755

During my investigation I did notice a "Partition 1 does not start on physical sector boundary." warning for fdisk -l.

# fdisk -l
Disk /dev/sda: 477 GiB, 512110190592 bytes, 1000215216 sectors
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disklabel type: gpt
Disk identifier: 6070EDCE-09B5-47AD-8A25-DB4E6EA4F1B1

Device         Start        End   Sectors   Size Type
/dev/sda1       2048     923647    921600   450M Windows recovery environment
/dev/sda2     931770    1980530   1048761 512.1M EFI System
/dev/sda3    1980531    2024827     44297  21.6M Microsoft reserved
/dev/sda4    2028726  500970959 498942234 237.9G Microsoft basic data
/dev/sda5  500971520  635328511 134356992  64.1G Linux filesystem
/dev/sda6  635328512  702437375  67108864    32G Linux filesystem
/dev/sda7  702437376 1000214527 297777152   142G Linux filesystem


Disk /dev/sdb: 931.5 GiB, 1000204886016 bytes, 1953525168 sectors
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 4096 bytes
I/O size (minimum/optimal): 4096 bytes / 4096 bytes
Disklabel type: gpt
Disk identifier: 0D5942A2-A942-45F1-A816-3E6EA63C05DA

Device          Start        End    Sectors  Size Type
/dev/sdb1          34     262177     262144  128M Microsoft reserved
/dev/sdb2      264192 1359218687 1358954496  648G Microsoft basic data
/dev/sdb3  1886416560 1953520064   67103505   32G Linux swap

Partition 1 does not start on physical sector boundary.

So I'm wondering whether intel-ucode.img got corrupted on the SSD - possibly due to some non-optimal choice on my part when I partitioned the drives last September (admittedly I've been ignoring the fact that /var (/dev/sdb6) doesn't seem to unmount during shutdown for awhile).

So I'd appreciate any advice regarding boot process maintenance and monitoring of partition and file system health.

Again, thank you for your time.

Offline

#4 2016-07-27 22:21:36

alphaniner
Member
From: Ancapistan
Registered: 2010-07-12
Posts: 2,810

Re: [SOLVED] intel-ucode-20160714-1 causing "EFI stub: ERROR:

You could check pacman log and journal for anything around the time of the update. Was going to suggest fscking the partition but I guess /boot is the EFI partition and is VFAT? Maybe it wasn't cleanly unmounted and/or maybe Windows got handsy?

I'm still in the stone age with BIOS/MBR/syslinux, so I can't offer any advice beyond that.


But whether the Constitution really be one thing, or another, this much is certain - that it has either authorized such a government as we have had, or has been powerless to prevent it. In either case, it is unfit to exist.
-Lysander Spooner

Offline

#5 2016-07-27 23:22:55

hinrikh
Member
Registered: 2016-07-27
Posts: 5

Re: [SOLVED] intel-ucode-20160714-1 causing "EFI stub: ERROR:

alphaniner wrote:

maybe Windows got handsy?

I'm always ready to blame Windows - but in this case Windows didn't get a kick at the can between the update and the error.

alphaniner wrote:

I'm still in the stone age with BIOS/MBR/syslinux

Whatever works - the simpler the better!

Thank You!

Offline

Board footer

Powered by FluxBB