Reverting to a Previous Kernel (l1tf bad) Solved

You have a problem with Salix? Post here and we'll do what we can to help.
Post Reply
User avatar
Luffy
Posts: 41
Joined: 3. Jul 2017, 22:03

Reverting to a Previous Kernel (l1tf bad) Solved

Post by Luffy » 31. Aug 2018, 18:13

https://docs.salixos.org/wiki/How_to_in ... ent_kernel
http://slackware.oregonstate.edu/slackware/slackware/a/

Only the kernel is found 4.4.14 ...
Where can I download:

/var/log/packages/kernel-firmware-201807XXX (downgrade required?)
/var/log/packages/kernel-headers-4.4.144_smp-x86-1
/var/log/packages/kernel-huge-smp-4.4.144_smp-i686-1
/var/log/packages/kernel-modules-smp-4.4.144_smp-i686-1

Excuse this comparison, in ArchLinux the packages are stored in /var/cache/pacman/pkg ... In Salix, what is that directory?

Finally it is safe, remove the contents of /tmp (700MB)

Code: Select all

> uname -a
Linux hostname 4.4.153-smp #2 SMP Tue Aug 28 14:36:05 CDT 2018 i686 Intel(R) Pentium(R) 4 CPU 2.80GHz GenuineIntel GNU/Linux

> grep . /sys/devices/system/cpu/vulnerabilities/*
/sys/devices/system/cpu/vulnerabilities/l1tf:Mitigation: Page Table Inversion
/sys/devices/system/cpu/vulnerabilities/meltdown:Vulnerable
/sys/devices/system/cpu/vulnerabilities/spec_store_bypass:Vulnerable
/sys/devices/system/cpu/vulnerabilities/spectre_v1:Mitigation: __user pointer sanitization
/sys/devices/system/cpu/vulnerabilities/spectre_v2:Mitigation: Full generic retpoline
> mpv *.avi CPU usage shoots up to 55% now
> mpv *.mp4 CPU usage shoots up to 95% now

There are no changes with l1tf = off (kernel parameters) ... vintage computing. :-)
Last edited by Luffy on 3. Sep 2018, 21:49, edited 1 time in total.

galmei
Posts: 64
Joined: 1. Jun 2018, 21:54

Re: Reverting to a Previous Kernel (l1tf bad)

Post by galmei » 31. Aug 2018, 20:40

Luffy wrote:
31. Aug 2018, 18:13
Where can I download:

/var/log/packages/kernel-firmware-201807XXX (downgrade required?)
/var/log/packages/kernel-headers-4.4.144_smp-x86-1
/var/log/packages/kernel-huge-smp-4.4.144_smp-i686-1
/var/log/packages/kernel-modules-smp-4.4.144_smp-i686-1

Excuse this comparison, in ArchLinux the packages are stored in /var/cache/pacman/pkg ... In Salix, what is that directory?
If these modules were already loaded before, look here: /var/slapt-get/patches/packages
Finally it is safe, remove the contents of /tmp (700MB)
Yes, all data that does not come from the current session.

User avatar
Luffy
Posts: 41
Joined: 3. Jul 2017, 22:03

Re: Reverting to a Previous Kernel (l1tf bad)

Post by Luffy » 31. Aug 2018, 22:23

Thanks colega ... For this PC, the 4.4.144 is the latest update.
> mpv *.avi CPU usage shoots up to 25% now
> mpv *.mp4 CPU usage shoots up to 60% now

Question, I must update to kernel-firmware-20180825_fea76a0-noarch-1 ...

Code: Select all

> slapt-get -l | grep kernel-firmware
kernel-firmware-20160628git-noarch-1 [inst=no]: kernel-firmware (Firmware for the kernel)
kernel-firmware-20180825_fea76a0-noarch-1 [inst=no]: kernel-firmware (Firmware for the kernel)
kernel-firmware-20180727_b01151b-noarch-1 [inst=sí]: kernel-firmware (Firmware for the kernel)
Saludos.

galmei
Posts: 64
Joined: 1. Jun 2018, 21:54

Re: Reverting to a Previous Kernel (l1tf bad)

Post by galmei » 1. Sep 2018, 01:26

Luffy wrote:
31. Aug 2018, 22:23
Question, I must update to kernel-firmware-20180825_fea76a0-noarch-1 ...

Code: Select all

> slapt-get -l | grep kernel-firmware
kernel-firmware-20160628git-noarch-1 [inst=no]: kernel-firmware (Firmware for the kernel)
kernel-firmware-20180825_fea76a0-noarch-1 [inst=no]: kernel-firmware (Firmware for the kernel)
kernel-firmware-20180727_b01151b-noarch-1 [inst=sí]: kernel-firmware (Firmware for the kernel)
I use other kernels and that's why my package names look different. But from the creation date you can still recognize the togetherness.

I still have these kernel packages available because I did not clean the package archive for a long time:

Code: Select all

export LANG=en_US.utf8
cd /var/slapt-get/patches/packages
ls -ld linux-4.4.*
drwxr-xr-x 2 root root    4096 Jan 16  2018 linux-4.4.111/
drwxr-xr-x 2 root root    4096 Feb 23  2018 linux-4.4.115/
drwxr-xr-x 2 root root    4096 Mar  4 12:56 linux-4.4.118/
drwxr-xr-x 2 root root    4096 May 31 19:50 linux-4.4.132/
drwxr-xr-x 2 root root    4096 Aug  7 13:43 linux-4.4.144/
drwxr-xr-x 2 root root    4096 Aug 31 22:29 linux-4.4.153/
drwxr-xr-x 2 root root    4096 Sep 13  2017 linux-4.4.75/
drwxr-xr-x 2 root root    4096 Oct  5  2017 linux-4.4.88/
On my system, the two apparently interesting kernel package versions have the following composition:

linux-4.4.153:

Code: Select all

[shortened ls excerpt]
72280920 Aug 27 20:52 kernel-firmware-20180825_fea76a0-noarch-1.txz
  745512 Aug 28 22:54 kernel-headers-4.4.153-x86-1.txz
 8308272 Aug 28 23:21 kernel-huge-4.4.153-x86_64-1.txz
29301688 Aug 28 23:13 kernel-modules-4.4.153-x86_64-1.txz
linux-4.4.144:

Code: Select all

[shortened ls excerpt]
71615888 Jul 27 20:50 kernel-firmware-20180727_b01151b-noarch-1.txz
  745608 Jul 26 18:58 kernel-headers-4.4.144-x86-1.txz
 8322944 Jul 26 19:46 kernel-huge-4.4.144-x86_64-1.txz
29297752 Jul 26 19:36 kernel-modules-4.4.144-x86_64-1.txz
If you go back to version 4.4.144, then in my opinion, the firmware package kernel-firmware-20180727_b01151b-noarch-1.txz belongs to it.

You can see the connection better with 'ls -l' than with 'slapt-get -l'. The 4.4.144 packages are all from the 26th to the 27th of July. The file kernel-firmware-20180727_b01151b-noarch-1.txz also contains the date in the file name.
20180727 stands for 2018-07-27 (yyyy-mm-dd) This coincides with the package file date and matches the other packages.

Unfortunately, I can not explain it better. Hope it helps anyway.

DidierSpaier
Posts: 312
Joined: 20. Jun 2016, 20:15

Re: Reverting to a Previous Kernel (l1tf bad)

Post by DidierSpaier » 1. Sep 2018, 02:40

I wouldn't go as far as downgrading the kernel, see this post.

PS But if you insist, see here.

User avatar
Luffy
Posts: 41
Joined: 3. Jul 2017, 22:03

Re: Reverting to a Previous Kernel (l1tf bad) Solved

Post by Luffy » 3. Sep 2018, 22:23

Thanks Galmei ...

@ DidierSpaier, I agree with HalseyTaylor ... "Kernel boot parameters did not seem to have much effect"

Post Reply