lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <aBl4wcX889otz_ms@google.com>
Date: Tue, 6 May 2025 10:49:37 +0800
From: Sung-Chi Li <lschyi@...omium.org>
To: Thomas Weißschuh <linux@...ssschuh.net>
Cc: Benson Leung <bleung@...omium.org>, Guenter Roeck <groeck@...omium.org>,
	Jean Delvare <jdelvare@...e.com>,
	Guenter Roeck <linux@...ck-us.net>,
	Jonathan Corbet <corbet@....net>, chrome-platform@...ts.linux.dev,
	linux-kernel@...r.kernel.org, linux-hwmon@...r.kernel.org,
	linux-doc@...r.kernel.org
Subject: Re: [PATCH v2 3/3] hwmon: (cros_ec) register fans into thermal
 framework cooling devices

On Sat, May 03, 2025 at 09:27:18AM +0200, Thomas Weißschuh wrote:
> On 2025-05-02 13:34:47+0800, Sung-Chi Li via B4 Relay wrote:
> > From: Sung-Chi Li <lschyi@...omium.org>
> > 
> > Register fans connected under EC as thermal cooling devices as well, so
> > these fans can then work with the thermal framework.
> > 
> > During the driver probing phase, we will also try to register each fan
> > as a thermal cooling device based on previous probe result (whether the
> > there are fans connected on that channel, and whether EC supports fan
> > control). The basic get max state, get current state, and set current
> > state methods are then implemented as well.
> > 
> > Signed-off-by: Sung-Chi Li <lschyi@...omium.org>
> > ---
> >  Documentation/hwmon/cros_ec_hwmon.rst |  2 ++
> >  drivers/hwmon/cros_ec_hwmon.c         | 66 +++++++++++++++++++++++++++++++++++
> >  2 files changed, 68 insertions(+)
> > 
> > diff --git a/Documentation/hwmon/cros_ec_hwmon.rst b/Documentation/hwmon/cros_ec_hwmon.rst
> > index 5b802be120438732529c3d25b1afa8b4ee353305..82c75bdaf912a116eaafa3149dc1252b3f7007d2 100644
> > --- a/Documentation/hwmon/cros_ec_hwmon.rst
> > +++ b/Documentation/hwmon/cros_ec_hwmon.rst
> > @@ -27,3 +27,5 @@ Fan and temperature readings are supported. PWM fan control is also supported if
> >  the EC also supports setting fan PWM values and fan mode. Note that EC will
> >  switch fan control mode back to auto when suspended. This driver will restore
> >  the fan state before suspended.
> > +If a fan is controllable, this driver will register that fan as a cooling device
> > +in the thermal framework as well.
> > diff --git a/drivers/hwmon/cros_ec_hwmon.c b/drivers/hwmon/cros_ec_hwmon.c
> > index c5e42e2a03a0c8c68d3f8afbb2bb45b93a58b955..abfcf44fb7505189124e78c651b0eb1e0533b4e8 100644
> > --- a/drivers/hwmon/cros_ec_hwmon.c
> > +++ b/drivers/hwmon/cros_ec_hwmon.c
> > @@ -13,6 +13,7 @@
> >  #include <linux/platform_device.h>
> >  #include <linux/platform_data/cros_ec_commands.h>
> >  #include <linux/platform_data/cros_ec_proto.h>
> > +#include <linux/thermal.h>
> 
> Needs a dependency on CONFIG_THERMAL.
> 

I think adding the `if (!IS_ENABLED(CONFIG_THERMAL))` you suggested is
sufficient, and turning on or off CONFIG_THERMAL both can compile, so I'll only
add the guarding statement in the `cros_ec_hwmon_register_fan_cooling_devices`.

> > +
> > +	if (!priv->fan_control_supported)
> > +		return;
> > +
> > +	for (i = 0; i < EC_FAN_SPEED_ENTRIES; i++) {
> > +		if (!(priv->usable_fans & BIT(i)))
> > +			continue;
> > +
> > +		cpriv = devm_kzalloc(dev, sizeof(*cpriv), GFP_KERNEL);
> > +		if (!cpriv)
> > +			return;
> > +
> > +		cpriv->hwmon_priv = priv;
> > +		cpriv->index = i;
> > +		devm_thermal_of_cooling_device_register(
> > +			dev, NULL, devm_kasprintf(dev, GFP_KERNEL, "cros-ec-fan%zu", i), cpriv,
> 
> What happens for multiple/chained ECs? If both provide sensors the
> thermal device names will collide.
> 

How about changing the "cros-ec-fan%zu" to "%s-fan%zu", which prefixes the
`dev_name()`? Here is an example from a device: cros-ec-hwmon.12.auto-fan0.

> Error handling for devm_kasprintf() is missing.
> 

Thank you for catching this, I will skip registering that device if the
devm_kasprintf() fails.

> > +			&cros_ec_thermal_cooling_ops);
> 
> Error handling for devm_thermal_of_cooling_device_register() is missing.
> 

I think we should continue registering other fans, so maybe we add a warning
here if the registration fails?

> > +	}
> > +}
> > +
> >  static int cros_ec_hwmon_probe(struct platform_device *pdev)
> >  {
> >  	struct device *dev = &pdev->dev;
> > @@ -402,6 +467,7 @@ static int cros_ec_hwmon_probe(struct platform_device *pdev)
> >  	cros_ec_hwmon_probe_fans(priv);
> >  	priv->fan_control_supported =
> >  		cros_ec_hwmon_probe_fan_control_supported(priv->cros_ec);
> > +	cros_ec_hwmon_register_fan_cooling_devices(dev, priv);
> >  
> >  	hwmon_dev = devm_hwmon_device_register_with_info(dev, "cros_ec", priv,
> >  							 &cros_ec_hwmon_chip_info, NULL);
> > 
> > -- 
> > 2.49.0.906.g1f30a19c02-goog
> > 
> > 

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ