From: "Ilpo Järvinen" <ilpo.jarvinen@linux.intel.com>
To: Radhey Kalra <radheykalra901@gmail.com>
Cc: "Kürşat Abaylı" <hello@kursatabayli.dev>,
krishna.chomal108@gmail.com, "Hans de Goede" <hansg@kernel.org>,
emreleno@gmail.com, edip@medip.dev,
platform-driver-x86@vger.kernel.org,
LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH v3] platform/x86: hp-wmi: Add dual-channel PWM fan control for Victus S
Date: Wed, 1 Jul 2026 13:39:54 +0300 (EEST) [thread overview]
Message-ID: <537bf20d-15e1-b3bf-f02a-09b96405c398@linux.intel.com> (raw)
In-Reply-To: <CAJ5VyXK_3wGmJtEwLLAtNFhKb0Ymf8TTSaO6cfozfN1GnComZw@mail.gmail.com>
[-- Attachment #1: Type: text/plain, Size: 10939 bytes --]
On Thu, 11 Jun 2026, Radhey Kalra wrote:
> Hi Ilpo,
>
> I tested Kürşat's v3 patch on my 8A3D machine:
>
> board_name: 8A3D
> product_name: Victus by HP Gaming Laptop 15-fb0xxx
> kernel: 7.0.11-1-cachyos-bore-lto
>
> For the test I used current platform-drivers-x86/for-next plus the v3 fan
> patch, with a temporary 8A3D DMI match added locally because 8A3D is not in
> for-next yet.
>
> The patched module exposed the expected interface:
>
> fan1_input=3800
> fan2_input=3000
> pwm1_enable=2
> pwm1=158
> pwm2=123
>
> Manual mode also worked:
>
> write pwm1_enable=1
> fan1_input=3300
> fan2_input=2500
> pwm1_enable=1
> pwm1=145
> pwm2=109
>
> Writing pwm1 and pwm2 separately affected the two fans independently:
>
> write pwm1=140
> fan1_input=3000
> fan2_input=2500
> pwm1=131
> pwm2=109
>
> write pwm2=160
> fan1_input=3000
> fan2_input=3500
> pwm1=131
> pwm2=153
>
> Brief max mode and return to auto also worked:
>
> write pwm1_enable=0
> fan1_input=3900
> fan2_input=4500
> pwm1_enable=0
>
> write pwm1_enable=2
> fan1_input=3300
> fan2_input=3700
> pwm1_enable=2
>
> I did not see a fan-control failure during the test.
Thanks for testing. Do you want to give your Tested-by tag to be credited
for you testing efforts?
(Up to you, not mandatory, of course. The tag will become permanent public
record in the commit.)
--
i.
> Thanks,
> Radhey
>
> On Wed, Jun 10, 2026 at 3:00 PM Ilpo Järvinen
> <ilpo.jarvinen@linux.intel.com> wrote:
> >
> > + Radhey
> >
> > On Sun, 31 May 2026, Kürşat Abaylı wrote:
> >
> > > Currently, manual fan control on supported Victus S models uses a single
> > > PWM value for both CPU and GPU fans, linking their speeds via a hardcoded
> > > gpu_delta offset. This prevents userspace tools from managing the thermal
> > > profiles of the CPU and GPU independently.
> > >
> > > Refactor the hwmon implementation to support independent dual-channel
> > > PWM control:
> > > - Split the single 'pwm' state into 'cpu_pwm' and 'gpu_pwm'.
> > > - Expose a second PWM channel ('pwm2') to userspace via hwmon_channel_info.
> > > - Remove the gpu_delta mechanism entirely.
> > >
> > > The 'pwm1_enable' mode remains shared, as the underlying hardware does
> > > not support per-fan modes. When switching to manual mode, both fans are
> > > smoothly initialized to their current RPMs. Additionally, ensure that
> > > the HP_FAN_SPEED_AUTOMATIC flag is isolated from rpm_to_pwm mathematical
> > > interpolations during mode resets to prevent unintended fan states.
> > >
> > > Tested on: HP Victus 16-s0xxx
> > >
> > > Signed-off-by: Kürşat Abaylı <hello@kursatabayli.dev>
> >
> > Has Radhey perhaps had an opportunity to test this patch?
> >
> > Radhey's series initially had the gpu delta init patch which was dropped
> > because this change was anticipated. It would be nice to confirm if this
> > patch successfully avoids the gpu delta init problem without introducing
> > some other problems?
> >
> > --
> > i.
> >
> > > ---
> > > Changes in v3:
> > > - Addressed review feedback from Krishna Chomal:
> > > - Expanded ternary operators in hp_wmi_fan_speed_set to standard if-else blocks.
> > > - Replaced hardcoded channel numbers with CPU_FAN and GPU_FAN macros.
> > >
> > > Changes in v2:
> > > - Made RPM readings atomic during the transition to manual mode.
> > > - Fixed a variable scoping issue in the pwm_input block.
> > > ---
> > > ---
> > > drivers/platform/x86/hp/hp-wmi.c | 60 +++++++++++++++++---------------
> > > 1 file changed, 32 insertions(+), 28 deletions(-)
> > >
> > > diff --git a/drivers/platform/x86/hp/hp-wmi.c b/drivers/platform/x86/hp/hp-wmi.c
> > > index f63bc00d9a9b..cef0c4436bd0 100644
> > > --- a/drivers/platform/x86/hp/hp-wmi.c
> > > +++ b/drivers/platform/x86/hp/hp-wmi.c
> > > @@ -488,9 +488,9 @@ struct hp_wmi_hwmon_priv {
> > > struct mutex lock; /* protects mode, pwm */
> > > u8 min_rpm;
> > > u8 max_rpm;
> > > - int gpu_delta;
> > > u8 mode;
> > > - u8 pwm;
> > > + u8 cpu_pwm;
> > > + u8 gpu_pwm;
> > > struct delayed_work keep_alive_dwork;
> > > };
> > >
> > > @@ -817,24 +817,20 @@ static int hp_wmi_fan_speed_max_set(int enabled)
> > > return enabled;
> > > }
> > >
> > > -static int hp_wmi_fan_speed_set(struct hp_wmi_hwmon_priv *priv, u8 speed)
> > > +static int hp_wmi_fan_speed_set(struct hp_wmi_hwmon_priv *priv)
> > > {
> > > u8 fan_speed[2];
> > > - int gpu_speed, ret;
> > > + int ret;
> > >
> > > - fan_speed[CPU_FAN] = speed;
> > > - fan_speed[GPU_FAN] = speed;
> > > + if (priv->cpu_pwm == HP_FAN_SPEED_AUTOMATIC)
> > > + fan_speed[CPU_FAN] = HP_FAN_SPEED_AUTOMATIC;
> > > + else
> > > + fan_speed[CPU_FAN] = pwm_to_rpm(priv->cpu_pwm, priv);
> > >
> > > - /*
> > > - * GPU fan speed is always a little higher than CPU fan speed, we fetch
> > > - * this delta value from the fan table during hwmon init.
> > > - * Exception: Speed is set to HP_FAN_SPEED_AUTOMATIC, to revert to
> > > - * automatic mode.
> > > - */
> > > - if (speed != HP_FAN_SPEED_AUTOMATIC) {
> > > - gpu_speed = speed + priv->gpu_delta;
> > > - fan_speed[GPU_FAN] = clamp_val(gpu_speed, 0, U8_MAX);
> > > - }
> > > + if (priv->gpu_pwm == HP_FAN_SPEED_AUTOMATIC)
> > > + fan_speed[GPU_FAN] = HP_FAN_SPEED_AUTOMATIC;
> > > + else
> > > + fan_speed[GPU_FAN] = pwm_to_rpm(priv->gpu_pwm, priv);
> > >
> > > ret = hp_wmi_get_fan_count_userdefine_trigger();
> > > if (ret < 0)
> > > @@ -851,7 +847,9 @@ static int hp_wmi_fan_speed_set(struct hp_wmi_hwmon_priv *priv, u8 speed)
> > >
> > > static int hp_wmi_fan_speed_reset(struct hp_wmi_hwmon_priv *priv)
> > > {
> > > - return hp_wmi_fan_speed_set(priv, HP_FAN_SPEED_AUTOMATIC);
> > > + priv->cpu_pwm = HP_FAN_SPEED_AUTOMATIC;
> > > + priv->gpu_pwm = HP_FAN_SPEED_AUTOMATIC;
> > > + return hp_wmi_fan_speed_set(priv);
> > > }
> > >
> > > static int hp_wmi_fan_speed_max_reset(struct hp_wmi_hwmon_priv *priv)
> > > @@ -2402,7 +2400,7 @@ static int hp_wmi_apply_fan_settings(struct hp_wmi_hwmon_priv *priv)
> > > case PWM_MODE_MANUAL:
> > > if (!is_victus_s_thermal_profile())
> > > return -EOPNOTSUPP;
> > > - ret = hp_wmi_fan_speed_set(priv, pwm_to_rpm(priv->pwm, priv));
> > > + ret = hp_wmi_fan_speed_set(priv);
> > > if (ret < 0)
> > > return ret;
> > > mod_delayed_work(system_wq, &priv->keep_alive_dwork,
> > > @@ -2502,13 +2500,14 @@ static int hp_wmi_hwmon_write(struct device *dev, enum hwmon_sensor_types type,
> > > u32 attr, int channel, long val)
> > > {
> > > struct hp_wmi_hwmon_priv *priv;
> > > - int rpm;
> > > + int cpu_rpm, gpu_rpm;
> > >
> > > priv = dev_get_drvdata(dev);
> > > guard(mutex)(&priv->lock);
> > > switch (type) {
> > > case hwmon_pwm:
> > > if (attr == hwmon_pwm_input) {
> > > + int rpm;
> > > if (!is_victus_s_thermal_profile())
> > > return -EOPNOTSUPP;
> > > /* PWM input is invalid when not in manual mode */
> > > @@ -2518,7 +2517,10 @@ static int hp_wmi_hwmon_write(struct device *dev, enum hwmon_sensor_types type,
> > > /* ensure PWM input is within valid fan speeds */
> > > rpm = pwm_to_rpm(val, priv);
> > > rpm = clamp_val(rpm, priv->min_rpm, priv->max_rpm);
> > > - priv->pwm = rpm_to_pwm(rpm, priv);
> > > + if (channel == CPU_FAN)
> > > + priv->cpu_pwm = rpm_to_pwm(rpm, priv);
> > > + else if (channel == GPU_FAN)
> > > + priv->gpu_pwm = rpm_to_pwm(rpm, priv);
> > > return hp_wmi_apply_fan_settings(priv);
> > > }
> > > switch (val) {
> > > @@ -2532,10 +2534,14 @@ static int hp_wmi_hwmon_write(struct device *dev, enum hwmon_sensor_types type,
> > > * When switching to manual mode, set fan speed to
> > > * current RPM values to ensure a smooth transition.
> > > */
> > > - rpm = hp_wmi_get_fan_speed_victus_s(channel);
> > > - if (rpm < 0)
> > > - return rpm;
> > > - priv->pwm = rpm_to_pwm(rpm / 100, priv);
> > > + cpu_rpm = hp_wmi_get_fan_speed_victus_s(CPU_FAN);
> > > + if (cpu_rpm < 0)
> > > + return cpu_rpm;
> > > + gpu_rpm = hp_wmi_get_fan_speed_victus_s(GPU_FAN);
> > > + if (gpu_rpm < 0)
> > > + return gpu_rpm;
> > > + priv->cpu_pwm = rpm_to_pwm(cpu_rpm / 100, priv);
> > > + priv->gpu_pwm = rpm_to_pwm(gpu_rpm / 100, priv);
> > > priv->mode = PWM_MODE_MANUAL;
> > > return hp_wmi_apply_fan_settings(priv);
> > > case PWM_MODE_AUTO:
> > > @@ -2551,7 +2557,7 @@ static int hp_wmi_hwmon_write(struct device *dev, enum hwmon_sensor_types type,
> > >
> > > static const struct hwmon_channel_info * const info[] = {
> > > HWMON_CHANNEL_INFO(fan, HWMON_F_INPUT, HWMON_F_INPUT),
> > > - HWMON_CHANNEL_INFO(pwm, HWMON_PWM_ENABLE | HWMON_PWM_INPUT),
> > > + HWMON_CHANNEL_INFO(pwm, HWMON_PWM_ENABLE | HWMON_PWM_INPUT, HWMON_PWM_INPUT),
> > > NULL
> > > };
> > >
> > > @@ -2592,7 +2598,7 @@ static int hp_wmi_setup_fan_settings(struct hp_wmi_hwmon_priv *priv)
> > > struct victus_s_fan_table *fan_table;
> > > u8 min_rpm, max_rpm;
> > > u8 cpu_rpm, gpu_rpm, noise_db;
> > > - int gpu_delta, i, num_entries, ret;
> > > + int i, num_entries, ret;
> > > size_t header_size, entry_size;
> > >
> > > /* Default behaviour on hwmon init is automatic mode */
> > > @@ -2638,10 +2644,8 @@ static int hp_wmi_setup_fan_settings(struct hp_wmi_hwmon_priv *priv)
> > > if (min_rpm == U8_MAX || max_rpm == 0)
> > > return -EINVAL;
> > >
> > > - gpu_delta = fan_table->entries[0].gpu_rpm - fan_table->entries[0].cpu_rpm;
> > > priv->min_rpm = min_rpm;
> > > priv->max_rpm = max_rpm;
> > > - priv->gpu_delta = gpu_delta;
> > >
> > > return 0;
> > > }
> > >
>
next prev parent reply other threads:[~2026-07-01 10:40 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-05-31 18:28 [PATCH v3] platform/x86: hp-wmi: Add dual-channel PWM fan control for Victus S Kürşat Abaylı
2026-06-10 9:30 ` Ilpo Järvinen
2026-06-11 15:19 ` Radhey Kalra
2026-07-01 10:39 ` Ilpo Järvinen [this message]
2026-07-01 12:37 ` Radhey Kalra
2026-07-07 12:57 ` Ilpo Järvinen
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=537bf20d-15e1-b3bf-f02a-09b96405c398@linux.intel.com \
--to=ilpo.jarvinen@linux.intel.com \
--cc=edip@medip.dev \
--cc=emreleno@gmail.com \
--cc=hansg@kernel.org \
--cc=hello@kursatabayli.dev \
--cc=krishna.chomal108@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=platform-driver-x86@vger.kernel.org \
--cc=radheykalra901@gmail.com \
/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