Bug 33555 - My AMD processor is still at risk after latest microcode update
Summary: My AMD processor is still at risk after latest microcode update
Status: RESOLVED INVALID
Alias: None
Product: Mageia
Classification: Unclassified
Component: Security (show other bugs)
Version: 9
Hardware: All Linux
Priority: Normal normal
Target Milestone: ---
Assignee: Nicolas Salguero
QA Contact: Sec team
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2024-09-11 15:36 CEST by Jose Manuel López
Modified: 2024-09-12 23:22 CEST (History)
4 users (show)

See Also:
Source RPM: Microcode
CVE:
Status comment:


Attachments

Description Jose Manuel López 2024-09-11 15:36:29 CEST
Description of problem:After the last validated update: microcode-0.20240813-1.mga9, my AMD Ryzen 7 4800H processor is still vulnerable as shown in terminal:


journalctl -xb | grep microcode
sep 10 20:39:29 localhost kernel: Zenbleed: please update your microcode for the most optimal fix
sep 10 20:39:29 localhost kernel: microcode: CPU2: patch_level=0x08600103
sep 10 20:39:29 localhost kernel: microcode: CPU1: patch_level=0x08600103
sep 10 20:39:29 localhost kernel: microcode: CPU3: patch_level=0x08600103
sep 10 20:39:29 localhost kernel: microcode: CPU6: patch_level=0x08600103
sep 10 20:39:29 localhost kernel: microcode: CPU0: patch_level=0x08600103
sep 10 20:39:29 localhost kernel: microcode: CPU7: patch_level=0x08600103
sep 10 20:39:29 localhost kernel: microcode: CPU15: patch_level=0x08600103
sep 10 20:39:29 localhost kernel: microcode: CPU14: patch_level=0x08600103
sep 10 20:39:29 localhost kernel: microcode: CPU4: patch_level=0x08600103
sep 10 20:39:29 localhost kernel: microcode: CPU5: patch_level=0x08600103
sep 10 20:39:29 localhost kernel: microcode: CPU9: patch_level=0x08600103
sep 10 20:39:29 localhost kernel: microcode: CPU8: patch_level=0x08600103
sep 10 20:39:29 localhost kernel: microcode: CPU11: patch_level=0x08600103
sep 10 20:39:29 localhost kernel: microcode: CPU10: patch_level=0x08600103
sep 10 20:39:29 localhost kernel: microcode: CPU13: patch_level=0x08600103
sep 10 20:39:29 localhost kernel: microcode: CPU12: patch_level=0x08600103
sep 10 20:39:29 localhost kernel: microcode: Microcode Update Driver: v2.2.


As far as I have been able to check, in github there is a more current version of microcode: ver0B40401C_2024-07-15. 

Check this link to verify this information and update microcode to the latest version for our AMD users in Mageia.
Comment 1 Len Lawrence 2024-09-11 16:09:20 CEST
Yes, I see the same problem with AMD Ryzen 7 5700U.

CC: (none) => tarazed25

Comment 2 Morgan Leijström 2024-09-11 16:17:55 CEST
Thank you for spotting this and reporting.

I set this to last packager. Previous was tmb.
- Nicholas, if you ca not do this now, please re-assign to all packagers.

I set this to security.
- If wrong please adjust.

QA Contact: (none) => security
Component: RPM Packages => Security
Assignee: bugsquad => nicolas.salguero
CC: (none) => fri

Comment 3 Nicolas Salguero 2024-09-11 17:02:44 CEST
(In reply to Jose Manuel López from comment #0)
> As far as I have been able to check, in github there is a more current
> version of microcode: ver0B40401C_2024-07-15. 
> 
> Check this link to verify this information and update microcode to the
> latest version for our AMD users in Mageia.

Hi,

As far as I can see, AMD microcodes are only available in https://git.kernel.org/pub/scm/linux/kernel/git/firmware/linux-firmware.git, contrary to Intel ones, that are provided in a github repository.

Best regards,

Nico.
Comment 4 Len Lawrence 2024-09-11 20:02:19 CEST
Is this link any use?  It says "Public".
https://github.com/platomav/CPUMicrocodes/tree/master/AMD
Comment 5 Martin Whitaker 2024-09-11 20:23:47 CEST
The message "Zenbleed: please update your microcode for the most optimal fix" does not mean you machine is vulnerable, it means that the bug is being mitigated in software, which is less efficient than mitigating it in the microcode.

In AMD's announcement of CVE-2023-20593:

  https://www.amd.com/en/resources/product-security/bulletin/amd-sb-7008.html

they said it would only be mitigated by a microcode update for the EPYC server-class CPUs. For other classes of CPU it would be mitigated by a firmware (BIOS) update from your machine/motherboard vendor. I can't find anything to say the situation has changed.

CC: (none) => mageia

Comment 6 Nicolas Salguero 2024-09-12 10:15:16 CEST
Hi,

I checked from where some other distributions get their microcode updates.  I found that we work exactly as the other do.

For Intel, the source is: https://github.com/intel/Intel-Linux-Processor-Microcode-Data-Files/

For AMD, the source is: https://git.kernel.org/pub/scm/linux/kernel/git/firmware/linux-firmware.git (mainly the directories "amd" and "amd-ucode").

Nobody is using https://github.com/platomav/CPUMicrocodes/.  Moreover, the README states that:
"""
It is generally advised to request and/or wait for your OEM/OS to release newer fixes. Latest is not always better or tested. Manufacturers and OS mainteners usually have some insider/confidential info from microcode vendors on what got changed/fixed at newer microcode releases so if they ship older microcodes, it could be that newer versions have not been thoroughly tested, have been retracted/downgraded by the microcode vendor or not contain anything important enough to warrant an update. The microcodes here are gathered and provided with the sole purpose of helping people who are out of other viable solutions. Thus, they can be extremely helpful to those who have major problems with their systems for which their manufacturer refuses to assist due to indifference and/or system age.
"""

For all those reasons, I consider that bug as invalid.

Best regards,

Nico.

Resolution: (none) => INVALID
Status: NEW => RESOLVED

Comment 7 Morgan Leijström 2024-09-12 12:06:05 CEST
(In reply to Nicolas Salguero from comment #6)
> For all those reasons, I consider that bug as invalid.

I agree, so maybe instead patch our kernel to agree with us and not say "please update your microcode" ?
Morgan Leijström 2024-09-12 12:06:32 CEST

CC: (none) => kernel

Comment 8 Martin Whitaker 2024-09-12 23:22:44 CEST
(In reply to Morgan Leijström from comment #7)
> I agree, so maybe instead patch our kernel to agree with us and not say
> "please update your microcode" ?

That message does tell you if your BIOS contains the fixed microcode, so it has some value. For example, after updating the BIOS on my desktop machine, that message has gone, but after updating the BIOS on my laptop it is still present.

Note You need to log in before you can comment on or make changes to this bug.