I wonder if intel-ucode is being used

[pebcak]

This is my kernel boot line:

cat /proc/cmdline
BOOT_IMAGE=/@/boot/vmlinuz-linux-zen root=UUID=76692c65-fa62-439e-aece-737c5176ed94 rw rootflags=subvol=@ resume=UUID=6875e2e7-c2b1-4bb3-b5ca-1ca2d133e8d8 apparmor=1 lsm=lockdown,yama,apparmor loglevel=3 nowatchdog

and the entry in grub.cfg:

menuentry 'EndeavourOS' --class endeavouros --class gnu-linux --class gnu --class os $menuentry_id_option 'gnulinux-simple-76692c65-fa62-439e-aece-737c5176ed94' {
	load_video
	set gfxpayload=keep
	insmod gzio
	insmod part_gpt
	insmod btrfs
	if [ x$feature_platform_search_hint = xy ]; then
	  search --no-floppy --fs-uuid --set=root  76692c65-fa62-439e-aece-737c5176ed94
	else
	  search --no-floppy --fs-uuid --set=root 76692c65-fa62-439e-aece-737c5176ed94
	fi
	echo	'Loading kernel linux-zen ...'
	linux	/@/boot/vmlinuz-linux-zen root=UUID=76692c65-fa62-439e-aece-737c5176ed94 rw rootflags=subvol=@  resume=UUID=6875e2e7-c2b1-4bb3-b5ca-1ca2d133e8d8 apparmor=1 lsm=lockdown,yama,apparmor loglevel=3 nowatchdog
	echo	'Loading initial ramdisk ...'
	initrd	/@/boot/initramfs-linux-zen.img

I am new to installing with btrfs, so this is maybe alright? I am used to seeing the intel-ucode in the initrd line.

[manuel]

Doesn’t look like intel-ucode is in use…

[pebcak]

That doesn’t sound good. Do you have any idea what might be wrong? Please tell me if I should run other commands to provide more info.

I am using the os-prober from Manjaro:

First experiment with installing on BTRFS

Hello btrfs users! Testing needed! All btrfs (and grub) users here, please test the following (same commands as above) and report the results here! Maybe we can make this a package later, if this works for all. cd /tmp git clone https://gitlab.manjaro.org/packages/community/os-prober.git cd os-prober makepkg -cf sudo pacman -U os-prober-1.*.pkg.tar.* # this line installs it

and also grub-btrfs.

[BONK]

@pebcak

I am using the os-prober from Manjaro:

That could be potentially problematic. Manjaro packages are usually modified Arch packages. Not exactly K.I.S.S

[pebcak]

Sure, but that was kind of a “workaround” to get the grub to recognize other systems installed with btrfs on a dual/multi -boot system.

I suspect there is something with the generation of the menuentries that fails. That is, something perhaps failing with grub-mkconfig. The os-prober part seems to run okay and discovers the other system. Don’t know.

[BONK]

[pebcak]

In the meantime, I removed Manjaro’s os-prober and installed the one from repo. I also removed grub-btrfs.

sudo os-prober doesn’t capture the other system, as expected.

And sudo grub-mkconfig -o /boot/grub/grub.cfg produces the same menuentry as before:

menuentry 'EndeavourOS' --class endeavouros --class gnu-linux --class gnu --class os $menuentry_id_option 'gnulinux-simple-76692c65-fa62-439e-aece-737c5176ed94' {
	load_video
	set gfxpayload=keep
	insmod gzio
	insmod part_gpt
	insmod btrfs
	if [ x$feature_platform_search_hint = xy ]; then
	  search --no-floppy --fs-uuid --set=root  76692c65-fa62-439e-aece-737c5176ed94
	else
	  search --no-floppy --fs-uuid --set=root 76692c65-fa62-439e-aece-737c5176ed94
	fi
	echo	'Loading kernel linux-zen ...'
	linux	/@/boot/vmlinuz-linux-zen root=UUID=76692c65-fa62-439e-aece-737c5176ed94 rw rootflags=subvol=@  resume=UUID=6875e2e7-c2b1-4bb3-b5ca-1ca2d133e8d8 apparmor=1 lsm=lockdown,yama,apparmor loglevel=3 nowatchdog
	echo	'Loading initial ramdisk ...'
	initrd	/@/boot/initramfs-linux-zen.img

No intel-ucode in sight.

[ringo]

is intel-ucode installed , normall if any ucode is installed with grub it will add to the fish…

[ricklinux]

Do you have Manjaro installed?

[pebcak]

Yes, I just checked it:

 pacman -Qi intel-ucode
Name            : intel-ucode
Version         : 20200616-1
Description     : Microcode update files for Intel CPUs
Architecture    : any
URL             : https://github.com/intel/Intel-Linux-Processor-Microcode-Data-Files
Licenses        : custom
Groups          : None
Provides        : None
Depends On      : None
Optional Deps   : None
Required By     : None
Optional For    : None
Conflicts With  : None
Replaces        : microcode_ctl
Installed Size  : 3.02 MiB
Packager        : Christian Hesse <arch@eworm.de>
Build Date      : Tue 16 Jun 2020 07:50:07 PM CEST
Install Date    : Wed 15 Jul 2020 01:42:30 PM CEST
Install Reason  : Explicitly installed
Install Script  : No
Validated By    : Signature

pacman -Ql intel-ucode

intel-ucode /boot/
intel-ucode /boot/intel-ucode.img
intel-ucode /usr/
intel-ucode /usr/share/
intel-ucode /usr/share/licenses/
intel-ucode /usr/share/licenses/intel-ucode/
intel-ucode /usr/share/licenses/intel-ucode/LICENSE

But the odd thing:

ls /boot 
efi  grub  initramfs-linux-fallback.img  initramfs-linux.img  initramfs-linux-zen-fallback.img  initramfs-linux-zen.img  memtest86+  vmlinuz-linux  vmlinuz-linux-zen

The inel-ucode.img is not there.

[pebcak]

No. The other system is Linux Mint. I use EnOS’ Grub to boot.

[ricklinux]

I can only say one thing. I have gone through this grub thing so many times and I don’t find it that easy to understand how it works. For me there was always a problem with certain distros. When I was using Antergos I also had Manjaro installed with Windows 10 and I let Manjaro control the boot. It always seemed to work. But, I have done a lot of multiboot where grub is always an issue with microcode or something.

Endeavour works good under most but not all scenarios. So I like what I have now because I don’t have that problem! Each grub I have now is on its own installation of Btrfs onLuks or without Encryption and I am using rEFInd and booting with grubx64.efi as opposed to the vmlinux-limux image for UEFi. I like it and grub is a non issue for me unless I do a normal install then the grub picks it up. If it’s installed using @2000 Btrfs setup they are minding their own business! I also have them on separate drives.

[pebcak]

This seems to be one of those odd, inexplicable things. Even though the output from pacman -Qi intel-ucode and pacman -Ql intel-ucode above show that it is installed but it wasn’t present in /boot.

I just reistalled the intel-ucode and voilà:

ls /boot
efi grub initramfs-linux-fallback.img initramfs-linux.img initramfs-linux-zen-fallback.img initramfs-linux-zen.img intel-ucode.img memtest86+ vmlinuz-linux vmlinuz-linux-zen

And:

sudo grub-mkconfig -o /boot/grub/grub.cfg

Generating grub configuration file …
Found theme: /boot/grub/themes/EndeavourOS/theme.txt
Found linux image: /boot/vmlinuz-linux-zen
Found initrd image: /boot/intel-ucode.img /boot/initramfs-linux-zen.img
Found fallback initrd image(s) in /boot: initramfs-linux-zen-fallback.img
Found linux image: /boot/vmlinuz-linux
Found initrd image: /boot/intel-ucode.img /boot/initramfs-linux.img
Found fallback initrd image(s) in /boot: initramfs-linux-fallback.img
Found memtest86+ image: /boot/memtest86+/memtest.bin
/usr/bin/grub-probe: warning: unknown device type nvme0n1.
done

So all is good now.

But how come at the first place …?

@ricklinux, I think I will need to get my act together and have a look at rEFInd some day soon. I don’t seem to be able giving up on multi-booting any time soon

[ricklinux]

Well I’m liking it with this Btrfs setup that @2000 has done. Although I may not totally understand it. It’s like EndeavourOS. I know that it works. A lot of stuff doesn’t. At least from my perspective as a normal user. For most people they aren’t the Linux gurus or Arch Elites. They just want to install something that works with no issues. This is why I like EndeavourOS because it’s mostly KISS! You want something fancy …you need to do it.

[manuel]

If I remember correctly, indeed not having intel-ucode really installed was a problem with a recent ISO. Which ISO did you use for installing?
Anyway, installing intel-ucode manually should fix that issue (as it did for you).

[pebcak]

@manuel

indeed not having intel-ucode really installed was a problem with a recent ISO. Which ISO did you use for installing?

I used the older ISO (the one before the current one) and used the offline install and updated the system after the installation. Now that you mention it, I remember that I have come across some topic regarding this issue in the forum. So that explains why pacman -Qi and pacman -Ql showed that the intel-ucode was iinstalled and the intel-ucode.img present in /boot while it wasn’t in reality.

Thanks @manuel for pointing it out!
And thank you @all for your help and support!

[mandog]

@manuel

If I remember correctly, indeed not having intel-ucode really installed was a problem with a recent ISO. Which ISO did you use for installing?
Anyway, installing intel-ucode manually should fix that issue (as it did for you).

@pebcak

I just reistalled the intel-ucode and voilà:

Thanks for that I was reading this post I did a install a few days before the latest ISO
The results on checking were the same as yours, the solution was the same so not a isolated incidence

[BONK]

There was an announcement

September ISO received a minor bugfix Torrent seeders please seed again

Our ISO that was released on Saturday had a bug in the offline installer. This bug has been fixed and that’s why we released a new ISO image. endeavouros-2020-09-20.iso (The previous was endeavouros-202-09-19) For our Torrent seeders, we kindly request to seed this updated release. You can get it over here: https://endeavouros.com/latest-release/