Drop minor version (NEEDS TESTING)

[Xabi08YT]

Hi,

I just finished the rework on version check during module load.
However, I cannot test that since I only have access to WSL atm (Intel please publish the freacking patch).

So if someone could test it would be great. Some regression may occur since I did some deduplicating.

I also tried adding a blacklist but it should also be tested. I know that it is not needed now but it may be used one day or another.

Like always, no AI. So if you want to check.

[EricLin0509]

@Xabi08YT

Hi, I just tested your commits, but it cause 2 compilation errors

msi-ec.c:60:2: error: expected ';' after top level declarator
   60 | }
      |  ^
      |  ;


msi-ec.c:2831:53: error: expected ';' after expression
 2831 |                 pr_err("Your firmware version is blacklisted !\n")
      |                                                                   ^
      | 

I'm using my own forks with your commits

[Xabi08YT]

@Xabi08YT

Hi, I just tested your commits, but it cause 2 compilation errors

msi-ec.c:60:2: error: expected ';' after top level declarator
   60 | }
      |  ^
      |  ;


msi-ec.c:2831:53: error: expected ';' after expression
 2831 |                 pr_err("Your firmware version is blacklisted !\n")
      |                                                                   ^
      | 

I'm using my own forks with your commits

Thank you very much. I will correct that right away

[Xabi08YT]

@EricLin0509 Thank you it should be fixed

[Xabi08YT]

Hello @glpnk,
I just wanted to check if you had time to test the PR ?
Also, small question, do we keep the version check like this or do we edit it to use the sku number as you suggested ?

[glpnk]

Sorry for the slow reply, but I think it is worth dropping only the last 2 digits of the version, not all 3 with + dot.

The SKU is the BIOS version, which also, probably printed on the board as MS-xxxxy, but no proofs. At the current stage of config list shrink, we are only dropping the version (or release number) part.

In the case of EC versions and features, we currently have 6 viable bytes: xxxx_MSy.z--. In place of the _ is the EC chip brand ITE/ENE. In place of MS, there can be different values, like WS for workstation, and few pre-release codes, which are questionable to support. Basically, we have no samples of those devices.

[Xabi08YT]

Ok I see. I will correct that later today

[Xabi08YT]

But you know, since the .1 duplicate data and almost never changes I thought dropping it would not change much

[Xabi08YT]

Yeah ok i just saw a .2 so yes we should keep it then. I am resolving the merge conflict so I will redo most of the deduplicating

[Xabi08YT]

@glpnk I corrected like you said. However it needs to be tested still. Some regression may occur but I think it should be fine

[glpnk]

Sorted and squashed commits

Backup branch

📩

At least it compiles

[Xabi08YT]

OK wow good job @glpnk
Much cleaner than what I did lol. But i kinda hate making regex lol

[glpnk]

Well, it was joke. Anyway I've deduped them manually

[Xabi08YT]

Well, it was joke. Anyway I've deduped them manually

Wow thx. It is really a painful job

[Xabi08YT]

OK I see what you meant now. Yeah right it is also good.

[Xabi08YT]

I think everything looks good. You can merge if you want

[Xabi08YT]

Hey guys, just wanted to know if there is any conflicts that need to be resolved ATM before merging this.

[glpnk]

I'll fix them and merge soon

[glpnk]

I think we can drop bigger part of the name, because recently we got request from Workstation/Creator devices and devices made for Turkey, but maybe address this problem in newer PR after this

[Xabi08YT]

OK. But for dropping bigger part of the name idk. I don't feel like it is a good idea. I think that would make that too broad. IDK.

[Xabi08YT]

So maybe we should start like this, and drop more later if we feel like it is something needed

[Xabi08YT]

Hey @glpnk I just updated this branch to match main.