From: Mingyou Chen <qby140326@gmail.com>
To: "Bozhen Peng" <bozhenpeng93@gmail.com>,
"Hans de Goede" <hansg@kernel.org>,
"Ilpo Järvinen" <ilpo.jarvinen@linux.intel.com>
Cc: Armin Wolf <W_Armin@gmx.de>, Jonathan Corbet <corbet@lwn.net>,
Shuah Khan <skhan@linuxfoundation.org>,
Randy Dunlap <rdunlap@infradead.org>,
platform-driver-x86@vger.kernel.org, linux-doc@vger.kernel.org,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH v2] platform/x86: bitland-mifs-wmi: Support WUJIE15 PRO HPT
Date: Fri, 20 Nov 2026 22:48:06 +0800 [thread overview]
Message-ID: <8b31251f-cb67-4aef-bd93-d2bd43885be1@gmail.com> (raw)
In-Reply-To: <fb4e0678-3028-463e-8a8c-1b680e40c4ee@gmail.com>
On 9/23/26 15:42, Bozhen Peng wrote:
> I found that my WUJIE15 PRO with the HPT board uses three profiles:
> performance (0), balanced (1), and quiet (2). The driver's default
> mapping gets these wrong and exposes a fourth mode that the firmware
> does not implement. My laptop only has USB-C power input, so the
> circular-hole adapter requirement also rules out its supported external-
> power configuration.
>
> I added a check matching the vendor, product and board names to select
> the correct profiles and power check. During testing, I saw
> power_supply_is_system_supplied() return true even after I unplugged the
> charger during sleep. I switched the check to the ACPI adapter's _PSR,
> which reads the current power-present bit on this machine. I also made
> resume select balanced if power was removed, rather than restore the
> saved performance profile. If the power query fails, I return the error
> instead of assuming the charger is connected.
>
> I checked the firmware methods and found only two fan readings, in F2/F1
> order, and no MIFS CPU temperature reading. Some unimplemented GET
> commands return success with zero data, so I could not use a successful
> query to establish sensor support. I register the two fans and leave out
> the temperature input. I also hide the unsupported GPU, RGB mode and fan
> boost controls. I rely on the shared WMI helper to check the remaining SET
> commands. I left normal keyboard backlight control unchanged.
>
> I found that the event GUID also matches redmi-wmi, so I made that
> driver decline this machine and leave its notifications to bitland-mifs-
> wmi. I kept the existing behavior for other machines and documented the
> HPT differences.
>
> Signed-off-by: Bozhen Peng <bozhenpeng93@gmail.com>
> ---
> Changes in v2:
> - Fix the RST heading underline length, as Randy pointed out.
> - Rebase onto current platform-driver-x86 for-next, retaining its shared
> WMI return-code handling.
> - Drop the unreachable redmi-wmi Cc address.
>
> I tested this on my MECHREVO WUJIE15 PRO (WUJIE15 Series-HPT), comparing
> the firmware methods and profile values with the Windows OEM controls.
>
> I ran the hardware tests on Ubuntu 26.04.1, kernel 7.0.0-30-generic. I
> used a small adaptation for that kernel's older WMI API in the test
> modules. I have left that adaptation out of this patch.
>
> I ran 22 automated profile, backlight, binding and sensor checks, all of
> which passed, and tested the actual hotkeys. I recorded about 2280 and
> 2850-2970 RPM from the two fans while running, and zero when stopped. I
> did not independently calibrate the RPM readings.
>
> For the final _PSR change, I ran two s2idle tests. With the charger
> connected, the driver read supplied=1 and restored performance; it
> stayed in performance afterwards. With the charger unplugged during
> sleep, it read supplied=0 and restored balanced; it stayed balanced
> afterwards. Requesting performance on battery then returned EOPNOTSUPP.
> The logs include the resume decision and firmware readbacks.
>
> I built a full x86_64 kernel and both modules against the for-next base
> below. I used x86_64_defconfig with BITLAND_MIFS_WMI and REDMI_WMI as
> modules, GCC 15.2.0 and W=1. The build passed without compiler warnings
> from either changed driver.
> Documentation/wmi/devices/bitland-mifs-wmi.rst | 31 +++++
> drivers/platform/x86/bitland-mifs-wmi.c | 157 ++++++++++++++++++++++++-
> drivers/platform/x86/redmi-wmi.c | 7 ++
> 3 files changed, 189 insertions(+), 6 deletions(-)
>
> diff --git a/Documentation/wmi/devices/bitland-mifs-wmi.rst b/Documentation/wmi/devices/bitland-mifs-wmi.rst
> index 27c3fd248..e73f9666f 100644
> --- a/Documentation/wmi/devices/bitland-mifs-wmi.rst
> +++ b/Documentation/wmi/devices/bitland-mifs-wmi.rst
> @@ -7,6 +7,37 @@ Bitland MIFS driver (bitland-mifs-wmi)
> Introduction
> ============
>
> +MECHREVO WUJIE15 PRO (HPT) firmware variant
> +-------------------------------------------
> +
> +The MECHREVO WUJIE15 PRO with board WUJIE15 Series-HPT uses a reduced
> +interface. The driver selects this variant using exact system vendor,
> +product name and board name matches.
> +
> +The supported platform profiles are low-power (raw value 2), balanced
> +(1), and performance (0). There is no separate full-speed mode.
> +USB-C is the only external power input on this model. Performance is
> +available on external power, but not on battery. The driver queries the
> +ACPI adapter's _PSR for current power presence because USB-C connector
> +ONLINE state may still be stale during resume. A failed query is not
> +treated as external power being present.
> +On resume, a saved performance profile becomes balanced if external
> +power was removed, matching the firmware's AC-loss transition.
> +
> +Command 0x0d returns two little-endian tachometers, EC Fan 2 followed by
> +Fan 1. No third fan or CPU-temperature input is provided. GPU switching,
> +RGB modes and fan boost are not exposed. Keyboard backlight brightness
> +uses the normal 0 through 3 values.
> +
> +Unsupported GET commands can return success (0x8000) with an empty
> +payload. Successful queries alone therefore do not establish the
> +presence of a sensor or control; the driver uses the known capabilities.
> +The shared WMI call helper checks firmware acknowledgments for supported
> +profile and backlight SET calls.
> +
> +The event GUID is shared with redmi-wmi. On this exact DMI combination,
> +redmi-wmi declines binding so bitland-mifs-wmi handles the notifications.
> +
>
> EC WMI interface description
> ============================
> diff --git a/drivers/platform/x86/bitland-mifs-wmi.c b/drivers/platform/x86/bitland-mifs-wmi.c
> index dd1648b7d..0d2b6c80c 100644
> --- a/drivers/platform/x86/bitland-mifs-wmi.c
> +++ b/drivers/platform/x86/bitland-mifs-wmi.c
> @@ -14,6 +14,7 @@
> #include <linux/dev_printk.h>
> #include <linux/device.h>
> #include <linux/device/devres.h>
> +#include <linux/dmi.h>
> #include <linux/err.h>
> #include <linux/hwmon.h>
> #include <linux/init.h>
> @@ -161,6 +162,19 @@ struct bitland_mifs_wmi_data {
> struct device *hwmon_dev;
> struct device *pp_dev;
> enum platform_profile_option saved_profile;
> + bool is_wujie15_hpt;
> +};
> +
> +/* Profile mapping and capabilities for the WUJIE15 PRO HPT board. */
> +static const struct dmi_system_id bitland_wujie15_hpt[] = {
> + {
> + .matches = {
> + DMI_EXACT_MATCH(DMI_SYS_VENDOR, "MECHREVO"),
> + DMI_EXACT_MATCH(DMI_PRODUCT_NAME, "WUJIE15 PRO"),
> + DMI_EXACT_MATCH(DMI_BOARD_NAME, "WUJIE15 Series-HPT"),
> + },
> + },
> + { }
> };
>
> static int bitland_mifs_wmi_call(struct bitland_mifs_wmi_data *data,
> @@ -195,6 +209,12 @@ static int bitland_mifs_wmi_call(struct bitland_mifs_wmi_data *data,
> if (output)
> memcpy(output, result, sizeof(*output));
>
> + if (data->is_wujie15_hpt)
> + dev_dbg(&data->wdev->dev,
> + "HPT op=%#x fn=%#x arg=%u value=%u status=%#x\n",
> + input->operation, input->function, input->payload[0],
> + result->data[0], le16_to_cpu(result->return_code));
> +
> return 0;
> }
>
> @@ -213,6 +233,22 @@ static int laptop_profile_get(struct device *dev,
> if (ret)
> return ret;
>
> + if (data->is_wujie15_hpt) {
> + switch (result.data[0]) {
> + case 0:
> + *profile = PLATFORM_PROFILE_PERFORMANCE;
> + return 0;
> + case 1:
> + *profile = PLATFORM_PROFILE_BALANCED;
> + return 0;
> + case 2:
> + *profile = PLATFORM_PROFILE_LOW_POWER;
> + return 0;
> + default:
> + return -EINVAL;
> + }
> + }
> +
> switch (result.data[0]) {
> case WMI_PP_BALANCED:
> *profile = PLATFORM_PROFILE_BALANCED;
> @@ -232,6 +268,26 @@ static int laptop_profile_get(struct device *dev,
> return 0;
> }
>
> +static int bitland_wujie15_is_supplied(void)
> +{
> + struct acpi_device *adapter;
> + unsigned long long online;
> + acpi_status status;
> +
> + /* USB-C ONLINE can still contain the pre-suspend connector state. */
> + adapter = acpi_dev_get_first_match_dev("ACPI0003", NULL, -1);
> + if (!adapter)
> + return -ENODEV;
> +
> + /* This machine's adapter _PSR reads the current EC power-present bit. */
> + status = acpi_evaluate_integer(adapter->handle, "_PSR", NULL, &online);
> + acpi_dev_put(adapter);
> + if (ACPI_FAILURE(status) || online > 1)
> + return -EIO;
> +
> + return online;
> +}
> +
> static int bitland_check_performance_capability(struct bitland_mifs_wmi_data *data)
> {
> struct bitland_mifs_input input = {
> @@ -241,7 +297,15 @@ static int bitland_check_performance_capability(struct bitland_mifs_wmi_data *da
> struct bitland_mifs_output output;
> int ret;
>
> - /* Full-speed/performance mode requires DC power (not USB-C) */
> + /* WUJIE15 PRO is powered through USB-C only. */
> + if (data->is_wujie15_hpt) {
> + ret = bitland_wujie15_is_supplied();
> + if (ret < 0)
> + return ret;
> + return ret ? 0 : -EOPNOTSUPP;
> + }
> +
> + /* Other models require the firmware's circular-hole adapter type. */
> if (!power_supply_is_system_supplied())
> return -EOPNOTSUPP;
>
> @@ -270,6 +334,27 @@ static int laptop_profile_set(struct device *dev,
> int ret;
> u8 val;
>
> + if (data->is_wujie15_hpt) {
> + switch (profile) {
> + case PLATFORM_PROFILE_LOW_POWER:
> + val = 2;
> + break;
> + case PLATFORM_PROFILE_BALANCED:
> + val = 1;
> + break;
> + case PLATFORM_PROFILE_PERFORMANCE:
> + ret = bitland_check_performance_capability(data);
> + if (ret)
> + return ret;
> + val = 0;
> + break;
> + default:
> + return -EOPNOTSUPP;
> + }
> + input.payload[0] = val;
> + return bitland_mifs_wmi_call(data, &input, NULL);
> + }
> +
> switch (profile) {
> case PLATFORM_PROFILE_LOW_POWER:
> val = WMI_PP_QUIET;
> @@ -300,9 +385,12 @@ static int laptop_profile_set(struct device *dev,
>
> static int platform_profile_probe(void *drvdata, unsigned long *choices)
> {
> + struct bitland_mifs_wmi_data *data = drvdata;
> +
> set_bit(PLATFORM_PROFILE_LOW_POWER, choices);
> set_bit(PLATFORM_PROFILE_BALANCED, choices);
> - set_bit(PLATFORM_PROFILE_BALANCED_PERFORMANCE, choices);
> + if (!data->is_wujie15_hpt)
> + set_bit(PLATFORM_PROFILE_BALANCED_PERFORMANCE, choices);
> set_bit(PLATFORM_PROFILE_PERFORMANCE, choices);
>
> return 0;
> @@ -328,13 +416,25 @@ static int bitland_mifs_wmi_suspend(struct device *dev)
> static int bitland_mifs_wmi_resume(struct device *dev)
> {
> struct bitland_mifs_wmi_data *data = dev_get_drvdata(dev);
> + enum platform_profile_option profile = data->saved_profile;
> + int supplied;
>
> /* Skip event device */
> if (!data->pp_dev)
> return 0;
>
> - dev_dbg(dev, "Resuming, restoring profile %d\n", data->saved_profile);
> - return laptop_profile_set(dev, data->saved_profile);
> + /* Match the firmware's performance-to-balanced transition on AC loss. */
> + if (data->is_wujie15_hpt &&
> + profile == PLATFORM_PROFILE_PERFORMANCE) {
> + supplied = bitland_wujie15_is_supplied();
> + if (supplied < 0)
> + return supplied;
> + if (!supplied)
> + profile = PLATFORM_PROFILE_BALANCED;
> + }
> +
> + dev_dbg(dev, "Resuming, restoring profile %d\n", profile);
> + return laptop_profile_set(dev, profile);
> }
>
> static DEFINE_SIMPLE_DEV_PM_OPS(bitland_mifs_wmi_pm_ops,
> @@ -398,7 +498,16 @@ static int laptop_hwmon_read_string(struct device *dev,
> enum hwmon_sensor_types type, u32 attr,
> int channel, const char **str)
> {
> + struct bitland_mifs_wmi_data *data = dev_get_drvdata(dev);
> + static const char *const hpt_fan_labels[] = { "Fan 2", "Fan 1" };
> +
> if (type == hwmon_fan && attr == hwmon_fan_label) {
> + if (data->is_wujie15_hpt) {
> + if (channel < 0 || channel >= ARRAY_SIZE(hpt_fan_labels))
> + return -EINVAL;
> + *str = hpt_fan_labels[channel];
> + return 0;
> + }
> if (channel >= 0 && channel < ARRAY_SIZE(fan_labels)) {
> *str = fan_labels[channel];
> return 0;
> @@ -426,6 +535,18 @@ static const struct hwmon_chip_info laptop_chip_info = {
> .info = laptop_hwmon_info,
> };
>
> +/* HPT fills only the F2/F1 tachometers; the temperature GET is a no-op. */
> +static const struct hwmon_channel_info *hpt_hwmon_info[] = {
> + HWMON_CHANNEL_INFO(fan, HWMON_F_INPUT | HWMON_F_LABEL,
> + HWMON_F_INPUT | HWMON_F_LABEL),
> + NULL
> +};
> +
> +static const struct hwmon_chip_info hpt_chip_info = {
> + .ops = &laptop_hwmon_ops,
> + .info = hpt_hwmon_info,
> +};
> +
> static int laptop_kbd_led_set(struct led_classdev *led_cdev,
> enum led_brightness value)
> {
> @@ -604,7 +725,28 @@ static const struct attribute *const laptop_attrs[] = {
> &dev_attr_fan_boost.attr,
> NULL,
> };
> -ATTRIBUTE_GROUPS(laptop);
> +
> +static umode_t laptop_attr_is_visible(struct kobject *kobj,
> + const struct attribute *attr, int index)
> +{
> + struct bitland_mifs_wmi_data *data = dev_get_drvdata(kobj_to_dev(kobj));
> +
> + /* All three custom attributes require methods absent on HPT. */
> + if (data && data->is_wujie15_hpt)
> + return 0;
> +
> + return attr->mode;
> +}
> +
> +static const struct attribute_group laptop_group = {
> + .attrs_const = laptop_attrs,
> + .is_visible_const = laptop_attr_is_visible,
> +};
> +
> +static const struct attribute_group *laptop_groups[] = {
> + &laptop_group,
> + NULL
> +};
>
> static const struct key_entry bitland_mifs_wmi_keymap[] = {
> { KE_KEY, WMI_EVENT_OPEN_APP, { KEY_PROG1 } },
> @@ -659,6 +801,7 @@ static int bitland_notifier_callback(struct notifier_block *nb,
> static int bitland_mifs_wmi_probe(struct wmi_device *wdev, const void *context)
> {
> struct bitland_mifs_wmi_data *drv_data;
> + const struct hwmon_chip_info *chip_info;
> enum bitland_wmi_device_type dev_type =
> (enum bitland_wmi_device_type)(unsigned long)context;
> struct led_init_data init_data = {
> @@ -673,6 +816,7 @@ static int bitland_mifs_wmi_probe(struct wmi_device *wdev, const void *context)
> return -ENOMEM;
>
> drv_data->wdev = wdev;
> + drv_data->is_wujie15_hpt = dmi_check_system(bitland_wujie15_hpt);
>
> ret = devm_mutex_init(&wdev->dev, &drv_data->lock);
> if (ret)
> @@ -706,10 +850,11 @@ static int bitland_mifs_wmi_probe(struct wmi_device *wdev, const void *context)
> return PTR_ERR(drv_data->pp_dev);
>
> /* Register hwmon */
> + chip_info = drv_data->is_wujie15_hpt ? &hpt_chip_info : &laptop_chip_info;
> drv_data->hwmon_dev = devm_hwmon_device_register_with_info(&wdev->dev,
> "bitland_mifs",
> drv_data,
> - &laptop_chip_info,
> + chip_info,
> NULL);
> if (IS_ERR(drv_data->hwmon_dev))
> return PTR_ERR(drv_data->hwmon_dev);
> diff --git a/drivers/platform/x86/redmi-wmi.c b/drivers/platform/x86/redmi-wmi.c
> index cc82ef502..de41642b7 100644
> --- a/drivers/platform/x86/redmi-wmi.c
> +++ b/drivers/platform/x86/redmi-wmi.c
> @@ -4,6 +4,7 @@
> #include <linux/acpi.h>
> #include <linux/bits.h>
> #include <linux/device.h>
> +#include <linux/dmi.h>
> #include <linux/input.h>
> #include <linux/input/sparse-keymap.h>
> #include <linux/module.h>
> @@ -69,6 +70,12 @@ static int redmi_wmi_probe(struct wmi_device *wdev, const void *context)
> struct redmi_wmi *data;
> int err;
>
> + /* This shared GUID belongs to bitland-mifs-wmi on the WUJIE15 PRO HPT. */
> + if (dmi_match(DMI_SYS_VENDOR, "MECHREVO") &&
> + dmi_match(DMI_PRODUCT_NAME, "WUJIE15 PRO") &&
> + dmi_match(DMI_BOARD_NAME, "WUJIE15 Series-HPT"))
> + return -ENODEV;
> +
> /* Init dev */
> data = devm_kzalloc(&wdev->dev, sizeof(*data), GFP_KERNEL);
> if (!data)
>
> base-commit: a3371fac5dd51518de4edcdcc3c7284a5c5a6607
Hi, sorry for the very late reply.
We have introduced the quirk table system in the recent commits (in the
pdx86/for-next branch).
You could refine the patch within the quirks system by adding new quirk
items. We have no WUJIE 15 PRO HPT support yet.
prev parent reply other threads:[~2026-11-20 14:48 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-09-14 1:12 [PATCH] " Bozhen Peng
2026-09-14 2:01 ` Randy Dunlap
2026-09-14 2:20 ` Bozhen Peng
2026-09-23 7:42 ` [PATCH v2] " Bozhen Peng
2026-11-20 14:48 ` Mingyou Chen [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=8b31251f-cb67-4aef-bd93-d2bd43885be1@gmail.com \
--to=qby140326@gmail.com \
--cc=W_Armin@gmx.de \
--cc=bozhenpeng93@gmail.com \
--cc=corbet@lwn.net \
--cc=hansg@kernel.org \
--cc=ilpo.jarvinen@linux.intel.com \
--cc=linux-doc@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=platform-driver-x86@vger.kernel.org \
--cc=rdunlap@infradead.org \
--cc=skhan@linuxfoundation.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®