linux-kernel - Re: [RFC PATCH 2/3] platform/x86: Add Uniwill laptop driver

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <c50c011f-6ad5-4122-a0e7-39880a4740ca@gmx.de>
Date: Sun, 22 Jun 2025 20:58:16 +0200
From: Armin Wolf <W_Armin@....de>
To: Werner Sembach <wse@...edocomputers.com>, ilpo.jarvinen@...ux.intel.com,
 hdegoede@...hat.com, chumuzero@...il.com, corbet@....net, cs@...edo.de,
 ggo@...edocomputers.com
Cc: linux-doc@...r.kernel.org, linux-kernel@...r.kernel.org,
 platform-driver-x86@...r.kernel.org
Subject: Re: [RFC PATCH 2/3] platform/x86: Add Uniwill laptop driver

Am 18.06.25 um 13:15 schrieb Werner Sembach:

> Hi,
>
> Am 17.06.25 um 23:50 schrieb Armin Wolf:
>> Am 16.06.25 um 14:32 schrieb Werner Sembach:
>>
>>> Hi,
>>>
>>> Am 15.06.25 um 19:59 schrieb Armin Wolf:
> [...]
>>>> +The EC behind the ``GetSetULong`` method is used by the OEM 
>>>> software supplied by the manufacturer.
>>>> +Reverse-engineering of this software is difficult since it uses an 
>>>> obfuscator, however some parts
>>>> +are not obfuscated.
>>> Iirc Christoffer said there is also another non wmi, but acpi, 
>>> get/set cmd that can write only specific regions but is faster. Just 
>>> wanted to leave this here for potential future optimizations if you 
>>> are interested.
>>
>> What interface are you referring to exactly?
>
> I think he was referring to these 2 methods in the dsdt:
>
>     Scope (_SB)
>     {
>         Device (INOU)
>         {
>             [...]
>
>             Method (ECRR, 1, NotSerialized)
>             {
>                 Local0 = (0xFE410000 + Arg0)
>                 Local1 = MMRW (Local0, Zero, Zero, Zero)
>                 Return (Local1)
>             }
>
>             Method (ECRW, 2, NotSerialized)
>             {
>                 Local0 = (0xFE410000 + Arg0)
>                 MMRW (Local0, One, Zero, Arg1)
>             }
>
>             [...]
>         }
>     }
>
> They can read/write the 0x04xx and 0x07xx addresses but not the 0x18xx 
> ones iirc, don't know about the 0x0fxx ones.
>
> I think 0x18xx was never intended to be written probably, but we use 
> it to implement a custom fan curve option in tuxedo-drivers on older 
> devices that officially don't support it.
>
> Best regards,
>
> Werner
>
I see, i too noticed the INOU0000 device during reverse engineering. I will update the documentation to inform future developers about this potentially
feaster interface. However for now i suggest that we stick to the WMI interface.

Thanks,
Armin Wolf