Message-ID: <f1787927-b655-4321-b9d9-bc12353c72db@gmx.de>
Date: Sun, 26 Oct 2025 23:59:38 +0100
From: Armin Wolf <W_Armin@....de>
To: Rong Zhang <i@...g.moe>, Derek John Clark <derekjohn.clark@...il.com>
Cc: Mark Pearson <mpearson-lenovo@...ebb.ca>, Hans de Goede
 <hansg@...nel.org>, Ilpo Järvinen
 <ilpo.jarvinen@...ux.intel.com>, Guenter Roeck <linux@...ck-us.net>,
 platform-driver-x86@...r.kernel.org, linux-kernel@...r.kernel.org,
 linux-hwmon@...r.kernel.org
Subject: Re: [PATCH 0/6] platform/x86: lenovo-wmi-{capdata,other}: Add HWMON
 for fan speed

Am 26.10.25 um 18:11 schrieb Rong Zhang:

> Hi Derek,
>
> On Sat, 2025-10-25 at 21:39 -0700, Derek John Clark wrote:
>> On Sun, Oct 19, 2025 at 2:05 PM Rong Zhang <i@...g.moe> wrote:
>>> Lenovo WMI Other Mode interface also supports querying or setting fan
>>> speed RPM. This capability is decribed by LENOVO_CAPABILITY_DATA_00.
>>> Besides, LENOVO_FAN_TEST_DATA provides reference data for self-test of
>>> cooling fans, including minimum and maximum fan speed RPM.
>>>
>>> This patchset turns lenovo-wmi-capdata01 into a unified driver (now
>>> named lenovo-wmi-capdata) for LENOVO_CAPABILITY_DATA_{00,01} and
>>> LENOVO_FAN_TEST_DATA; then adds HWMON support for lenovo-wmi-other:
>>>
>>>   - fanX_enable: enable/disable the fan (tunable)
>>>   - fanX_input: current RPM
>>>   - fanX_max: maximum RPM
>>>   - fanX_min: minimum RPM
>>>   - fanX_target: target RPM (tunable)
>>>
>>> This implementation doesn't require all capability data to be available,
>>> and is capable to expose interfaces accordingly:
>>>
>>>   - Having LENOVO_CAPABILITY_DATA_00: exposes fanX_{enable,input,target}
>>>   - Having LENOVO_CAPABILITY_DATA_01: exposes firmware_attributes
>>>   - Having LENOVO_FAN_TEST_DATA: exposes fanX_{max,min}
>>>
>>> Rong Zhang (6):
>>>    platform/x86: Rename lenovo-wmi-capdata01 to lenovo-wmi-capdata
>>>    platform/x86: lenovo-wmi-{capdata,other}: Support multiple Capability
>>>      Data
>>>    platform/x86: lenovo-wmi-capdata: Add support for Capability Data 00
>>>    platform/x86: lenovo-wmi-other: Add HWMON for fan speed RPM
>>>    platform/x86: lenovo-wmi-capdata: Add support for Fan Test Data
>>>    platform/x86: lenovo-wmi-other: Report min/max RPM and hide dummy fans
>>>
>>>   .../wmi/devices/lenovo-wmi-other.rst          |  32 +
>>>   drivers/platform/x86/lenovo/Kconfig           |   5 +-
>>>   drivers/platform/x86/lenovo/Makefile          |   2 +-
>>>   drivers/platform/x86/lenovo/wmi-capdata.c     | 545 ++++++++++++++++++
>>>   drivers/platform/x86/lenovo/wmi-capdata.h     |  46 ++
>>>   drivers/platform/x86/lenovo/wmi-capdata01.c   | 302 ----------
>>>   drivers/platform/x86/lenovo/wmi-capdata01.h   |  25 -
>>>   drivers/platform/x86/lenovo/wmi-other.c       | 422 +++++++++++++-
>>>   8 files changed, 1028 insertions(+), 351 deletions(-)
>>>   create mode 100644 drivers/platform/x86/lenovo/wmi-capdata.c
>>>   create mode 100644 drivers/platform/x86/lenovo/wmi-capdata.h
>>>   delete mode 100644 drivers/platform/x86/lenovo/wmi-capdata01.c
>>>   delete mode 100644 drivers/platform/x86/lenovo/wmi-capdata01.h
>>>
>>>
>>> base-commit: 3a8660878839faadb4f1a6dd72c3179c1df56787
>>> --
>>> 2.51.0
>>>
>> The series' intention looks good overall. The composable methods for
>> additional capdata interfaces is a welcome change. I have a few
>> comments I'll add for a couple of the patches. My apologies for the
>> slow review timeline, I've been on travel and wanted to test the
>> changes before submitting a review.
> Thanks for you review and testing! Hope you have/had a nice trip ;)
>
>> For testing I'm using my Legion Go 2. It apparently doesn't have the
>> FAN_TEST_DATA GUID, and the hwmon interface errors on all inputs
>> despite being visible. I know for the Legion Go series they use a fan
>> table with 10 auto_set points in the Other Method interface tied to
>> the platform profile, but the documentation I have says the methods
>> you're adding here should be available on all models, so that is a bit
>> strange.
> Yeah, that sounds weird.
>
> As for the fan table on your device, did you mean
> LENOVO_FAN_TABLE_DATA/LENOVO_FAN_METHOD? My device doesn't use a fan
> table, the corresponding ACPI methods are dummy (see below).
>
> My device is ThinkBook 14 G7+ ASP (forgot to mention when submitting,
> sorry). I don't have any documentation and I finished the patchset
> according to the MOF as well as the decompiled ASL code of its ACPI
> tables. The information from the documentation (including those in your
> following replies) is very useful, thanks for that!
>
> As it's branded as ThinkBook, most GAMEZONE/WMI_OTHER interfaces on my
> device may differ from Legion devices. To summerize:
>
> - LENOVO_GAMEZONE_DATA: dummy ACPI method.
> - LENOVO_GAMEZONE_CPU_OC_DATA: presents in MOF; missing ACPI method.
> - LENOVO_GAMEZONE_GPU_OC_DATA: dummy ACPI method.
> - LENOVO_CAPABILITY_DATA_00: works fine.
> - LENOVO_CAPABILITY_DATA_01: dummy ACPI method, data still presents
>    (\_SB.GZFD.CD01).
> - LENOVO_FAN_TEST_DATA: works fine.
> - LENOVO_FAN_TABLE_DATA: dummy ACPI method.
> - LENOVO_FAN_METHOD: dummy ACPI method.
> - LENOVO_OTHER_METHOD:
>    * Despite missing LENOVO_CAPABILITY_DATA_01, SPPT/SPL/FPPT can still
>      be get/set. There is also CHTC (FEATURE_ID=4, get/set) which I am
>      not sure what it means.
>    * FAN1/2: get method reads data from the EC; set method for FAN1
>      updates the EC, for FAN2 is dummy (no-op, returns 0).
>
>> dmesg output:
>> [    3.995549] lenovo_wmi_cd 362A3AFE-3D96-4665-8530-96DAD5BB300E-13:
>> registered LENOVO_CAPABILITY_DATA_00 with 33 items
>> [    4.000266] lenovo_wmi_cd 7A8F5407-CB67-4D6E-B547-39B3BE018154-9:
>> registered LENOVO_CAPABILITY_DATA_01 with 80 items
>> [    4.005603] lenovo_wmi_other
>> DC2A8805-3A8C-41BA-A6F7-092E0089CD3B-3: bound
>> 362A3AFE-3D96-4665-8530-96DAD5BB300E-13 (ops lwmi_cd_component_ops
>> [lenovo_wmi_capdata])
>> [    4.005611] lenovo_wmi_other
>> DC2A8805-3A8C-41BA-A6F7-092E0089CD3B-3: bound
>> 7A8F5407-CB67-4D6E-B547-39B3BE018154-9 (ops lwmi_cd_component_ops
>> [lenovo_wmi_capdata])
>> [    4.005614] lenovo_wmi_other
>> DC2A8805-3A8C-41BA-A6F7-092E0089CD3B-3: fan capdata unavailable
>>
>> Testing results:
>> (deck@...o2 hwmon5)$ ls
>> device  fan1_enable  fan1_input  fan1_target  name  power  subsystem  uevent
>> (deck@...o2 hwmon5)$ cat fan1_enable
>> cat: fan1_enable: No data available
>> (1)(deck@...o2 hwmon5)$ echo 1 | sudo tee fan1_enable
>> [sudo] password for deck:
>> 1
>> tee: fan1_enable: Input/output error
>> (1)(deck@...o2 hwmon5)$ echo 0 | sudo tee fan1_enable
>> 0
>> tee: fan1_enable: Input/output error
>> (1)(deck@...o2 hwmon5)$ echo 3000 | sudo tee fan1_target
>> 3000
>> tee: fan1_target: Input/output error
> -EIO was returned when the set method didn't return 1 (as long as
> lwmi_dev_evaluate_int() didn't return this due to ACPI_FAILURE).
> Despite the return value, did the fan speed change after writing?
> Otherwise the method might be dummy and LENOVO_CAPABILITY_DATA_00
> simply returned mistaken data :(
>
>> (1)(deck@...o2 hwmon5)$ cat fan1_input
>> cat: fan1_input: No such device or address
> -ENXIO was returned by lwmi_dev_evaluate_int() as the return value was
> not an integer. It's really weird. Could you check the type of the
> return value? Some clues may also lie in the ASL code of the ACPI
> method.

The Windows WMI-ACPI driver converts all ACPI objects into a common buffer
format, so returning a buffer with four bytes will look like an integer
for WMI consumers under Windows.

I already have patches for that, but for now i suggest that you handle
this inside lwmi_dev_evaluate_int() yourself.

Thanks,
Armin Wolf

>> Thanks,
>> Derek
> Thanks,
> Rong
>

Powered by blists - more mailing lists