Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Convert kernel hwmon interface to use PWM (0-255) intsead of RPM #261

Draft
wants to merge 4 commits into
base: main
Choose a base branch
from

Conversation

johnfanv2
Copy link
Owner

This will make it compliant with Kernel standard for hwmon and compatible with different fan curves (rpm, percentage, ...) of different models.

@johnfanv2 johnfanv2 changed the title Convert kernel hwmon interface to use PWM (0-255) intead of RPM Convert kernel hwmon interface to use PWM (0-255) intsead of RPM Sep 22, 2024
@MrDuartePT
Copy link
Collaborator

LGTM, I gonna make the python changes and this can be merge.

#185 need to be update since it have some fixes to new devices

johnfan and others added 2 commits September 23, 2024 12:20
* this is the cherry-pick commit of #185 need testing

Signed-off-by: Gonçalo Negrier Duarte <gonegrier.duarte@gmail.com>
@MrDuartePT
Copy link
Collaborator

MrDuartePT commented Sep 26, 2024

The commit I just push need be tested in a newer device but the code did compile.
The temperature is still hardcoded for the legion go values.

@antheas can you give a look if I miss something.

@antheas
Copy link

antheas commented Sep 26, 2024

kernel lgmt

I would personally keep using rpm for older models and change the driver name though. Perhaps even split the EC fan stuff to a different module that is out-of-tree, and try to merge only the WMI to the kernel.

@MrDuartePT
Copy link
Collaborator

I would personally keep using rpm for older models and change the driver name though. Perhaps even split the EC fan stuff to a different module that is out-of-tree, and try to merge only the WMI to the kernel.

Probably you are right and that would need to be done, but since I don't think the drive would be push to mainline anytime soon, let wait.
But I think is one of the things with should ask when making a proposal to kernel mailing list.

@@ -1712,6 +1712,12 @@ enum OtherMethodFeature {
OtherMethodFeature_C_U1 = 0x05010000,
OtherMethodFeature_TEMP_CPU = 0x05040000,
OtherMethodFeature_TEMP_GPU = 0x05050000,
OtherMethodFeature_FAN_FULLSPEED = 0x04020000,
Copy link
Owner Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

What is the source for that? Is there already a function that uses it that is not in this commit? Maybe you already have info how to use it. It is currently not used in this file.

u32 FST6;
u32 FST7;
u32 FST8;
u32 FST9;
Copy link
Owner Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

What is FTLE? Length?
FSTL: is this field renamed in Legion Go?
FST* = temp;
FSS* = speed; is it 0-100 (percent), pwn (0-255), or in rpm?

} __packed;

static ssize_t wmi_read_fancurve_custom(const struct model_config *model,
struct fancurve *fancurve)
{
u8 buffer[88];
struct WMIFanTableRead fantable;
Copy link
Owner Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Good idea. By adding this new fields we get exactly 88 Bytes. These fields where not defined in the ACPI code of other models.

fancurve->current_point_i = 0;
fancurve->size = 10;
fancurve->size = fantable.FSTL;
Copy link
Owner Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Hmm. This does not work for other models because FSTL is not set to 10 but just not set.

true);
err = wmi_exec_arg(WMI_GUID_LENOVO_FAN_METHOD, 0,
WMI_METHOD_ID_FAN_SET_TABLE, buffer, sizeof(buffer));
WMI_METHOD_ID_FAN_SET_TABLE, &fan, sizeof(fan));
Copy link
Owner Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

In the old code we write a buffer of size 0x20. Now we write an object of size 0x34. Will this break stuff in old models?

@@ -5694,6 +5736,34 @@ static struct attribute *fancurve_hwmon_attributes[] = {
&sensor_dev_attr_pwm1_mode.dev_attr.attr, NULL
};

static struct attribute *fancurve_hwmon_attributes_legion_go[] = {
Copy link
Owner Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Why create new hwmon entries? Do we need new ones with other properties?

&sensor_dev_attr_pwm1_auto_point10_pwm.dev_attr.attr,
// These should be read only if the driver is properly implemented
// and either fail to set or noop
// &sensor_dev_attr_pwm1_auto_point1_temp.dev_attr.attr,
Copy link
Owner Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I guess, we can make set read/write properties depending on model or just error out/ignore it.

@@ -5720,6 +5790,11 @@ static const struct attribute_group legion_hwmon_fancurve_group = {
.is_visible = legion_hwmon_is_visible,
};

static const struct attribute_group legion_go_hwmon_fancurve_group = {
Copy link
Owner Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

See above. I guess, this might be removed.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants