* [PATCH] hwmon: (asus_rog_ryujin) Update cached duty cycle after cooler PWM write
@ 2026-11-05 4:48 Guenter Roeck
2026-11-05 4:54 ` sashiko-bot
0 siblings, 1 reply; 2+ messages in thread
From: Guenter Roeck @ 2026-11-05 4:48 UTC (permalink / raw)
To: linux-hwmon; +Cc: Guenter Roeck, Aleksa Savic
In rog_ryujin_write_fixed_duty(), writing a PWM duty cycle to channel 0
(pump) or channel 1 (internal fan) returns directly from
rog_ryujin_write_cooler_duty(), skipping the priv->duty_input[channel]
update at the end of the function. Because the function first fetches
the pre-write cooler duty via get_cooler_duty_cmd and SET confirmation
reports do not update priv->duty_input, subsequent sysfs reads within
the STATUS_VALIDITY (1500 ms) cache window return stale duty values (or,
on Ryujin III when switching pump source at the reported duty, the
temporary neighbouring duty written by rog_ryujin_select_pump_source()).
Furthermore, rog_ryujin_write_fixed_duty() overwrote val in place with
rog_ryujin_pwm_to_percent(val), so even before the early return was
introduced, priv->duty_input[channel] stored a 0-100 percentage rather
than the 0-255 PWM value expected by rog_ryujin_read().
Keep the 0-100 percentage in a separate local variable and fall through
on success so priv->duty_input[channel] caches the written 0-255 PWM
value until the next status refresh.
Fixes: ed3e03790c5c ("hwmon: Add driver for ASUS ROG RYUJIN II 360 AIO cooler")
Fixes: b4306c0c8e24 ("hwmon: (asus_rog_ryujin) Rely on subsystem locking")
Cc: Aleksa Savic <savicaleksa83@gmail.com>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
---
drivers/hwmon/asus_rog_ryujin.c | 11 +++++++----
1 file changed, 7 insertions(+), 4 deletions(-)
diff --git a/drivers/hwmon/asus_rog_ryujin.c b/drivers/hwmon/asus_rog_ryujin.c
index 695e74cf71e4..7b80c46a0930 100644
--- a/drivers/hwmon/asus_rog_ryujin.c
+++ b/drivers/hwmon/asus_rog_ryujin.c
@@ -420,6 +420,7 @@ static int rog_ryujin_select_pump_source(struct rog_ryujin_data *priv, u8 *set_c
static int rog_ryujin_write_fixed_duty(struct rog_ryujin_data *priv, int channel, int val)
{
u8 set_cmd[SET_CMD_LENGTH];
+ int duty;
int ret;
if (channel < 2) {
@@ -437,18 +438,18 @@ static int rog_ryujin_write_fixed_duty(struct rog_ryujin_data *priv, int channel
set_cmd[RYUJIN_SET_COOLER_CHANNEL_OFFSET] = priv->info->duty_channel;
/* Cooler duties are set as 0-100% */
- val = rog_ryujin_pwm_to_percent(val);
+ duty = rog_ryujin_pwm_to_percent(val);
if (channel == 0) {
/* Cooler pump duty */
- set_cmd[RYUJIN_SET_COOLER_PUMP_DUTY_OFFSET] = val;
+ set_cmd[RYUJIN_SET_COOLER_PUMP_DUTY_OFFSET] = duty;
set_cmd[RYUJIN_SET_COOLER_FAN_DUTY_OFFSET] =
rog_ryujin_pwm_to_percent(priv->duty_input[1]);
} else if (channel == 1) {
/* Cooler internal fan duty */
set_cmd[RYUJIN_SET_COOLER_PUMP_DUTY_OFFSET] =
rog_ryujin_pwm_to_percent(priv->duty_input[0]);
- set_cmd[RYUJIN_SET_COOLER_FAN_DUTY_OFFSET] = val;
+ set_cmd[RYUJIN_SET_COOLER_FAN_DUTY_OFFSET] = duty;
}
if (priv->info->duty_channel) {
@@ -457,7 +458,9 @@ static int rog_ryujin_write_fixed_duty(struct rog_ryujin_data *priv, int channel
return ret;
}
- return rog_ryujin_write_cooler_duty(priv, set_cmd);
+ ret = rog_ryujin_write_cooler_duty(priv, set_cmd);
+ if (ret < 0)
+ return ret;
} else {
/*
* Controller fan duty (channel == 2). No need to retrieve current
--
2.45.2
^ permalink raw reply related [flat|nested] 2+ messages in thread
* Re: [PATCH] hwmon: (asus_rog_ryujin) Update cached duty cycle after cooler PWM write
2026-11-05 4:48 [PATCH] hwmon: (asus_rog_ryujin) Update cached duty cycle after cooler PWM write Guenter Roeck
@ 2026-11-05 4:54 ` sashiko-bot
0 siblings, 0 replies; 2+ messages in thread
From: sashiko-bot @ 2026-11-05 4:54 UTC (permalink / raw)
To: Guenter Roeck, linux-hwmon
> In rog_ryujin_write_fixed_duty(), writing a PWM duty cycle to channel 0
> (pump) or channel 1 (internal fan) returns directly from
> rog_ryujin_write_cooler_duty(), skipping the priv->duty_input[channel]
> update at the end of the function. Because the function first fetches
> the pre-write cooler duty via get_cooler_duty_cmd and SET confirmation
> reports do not update priv->duty_input, subsequent sysfs reads within
> the STATUS_VALIDITY (1500 ms) cache window return stale duty values (or,
> on Ryujin III when switching pump source at the reported duty, the
> temporary neighbouring duty written by rog_ryujin_select_pump_source()).
>
> Furthermore, rog_ryujin_write_fixed_duty() overwrote val in place with
> rog_ryujin_pwm_to_percent(val), so even before the early return was
> introduced, priv->duty_input[channel] stored a 0-100 percentage rather
> than the 0-255 PWM value expected by rog_ryujin_read().
>
> Keep the 0-100 percentage in a separate local variable and fall through
> on success so priv->duty_input[channel] caches the written 0-255 PWM
> value until the next status refresh.
>
> Fixes: ed3e03790c5c ("hwmon: Add driver for ASUS ROG RYUJIN II 360 AIO cooler")
> Fixes: b4306c0c8e24 ("hwmon: (asus_rog_ryujin) Rely on subsystem locking")
> Cc: Aleksa Savic <savicaleksa83@gmail.com>
> Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Sashiko has reviewed this patch and found no issues.
--
Sashiko AI review · https://sashiko.dev/#/patchset/20261105044843.3262037-1-linux@roeck-us.net?part=1
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2026-11-05 4:54 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-11-05 4:48 [PATCH] hwmon: (asus_rog_ryujin) Update cached duty cycle after cooler PWM write Guenter Roeck
2026-11-05 4:54 ` sashiko-bot
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.