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: <3a5fa538-2f08-a7f0-f85c-97fdffd93a8b@roeck-us.net>
Date:   Sun, 12 Mar 2017 15:53:56 -0700
From:   Guenter Roeck <linux@...ck-us.net>
To:     Shilpasri G Bhat <shilpa.bhat@...ux.vnet.ibm.com>
Cc:     linux-kernel@...r.kernel.org, linuxppc-dev@...ts.ozlabs.org,
        benh@...nel.crashing.org, mpe@...erman.id.au, paulus@...ba.org,
        svaidy@...ux.vnet.ibm.com, ego@...ux.vnet.ibm.com,
        akshay.adiga@...ux.vnet.ibm.com, andrew@...id.au, clg@...d.org,
        maddy@...ux.vnet.ibm.com, Rob Herring <robh+dt@...nel.org>,
        Mark Rutland <mark.rutland@....com>,
        Jean Delvare <jdelvare@...e.com>,
        Jonathan Corbet <corbet@....net>, devicetree@...r.kernel.org,
        linux-hwmon@...r.kernel.org, linux-doc@...r.kernel.org,
        Eddie James <eajames@...ux.vnet.ibm.com>
Subject: Re: [RFC 2/2] hwmon: powernv: Hwmon driver for OCC inband power and
 temperature sensors

On 03/09/2017 05:30 AM, Shilpasri G Bhat wrote:
> Hi Guenter,
>
> On 03/09/2017 05:40 PM, Guenter Roeck wrote:
>> On Thu, Mar 09, 2017 at 05:19:15PM +0530, Shilpasri G Bhat wrote:
>>> Add support to read power and temperature sensors from OCC inband
>>> sensors which are copied to main memory by OCC.
>>>
>>
>> Is this supposed to be an alternative to the submission from
>> Eddie James ? If so, is there a reason to consider such an
>> alternative ?
>>
>
> This is not an alternative submission. Eddie James' hwmon OCC driver is to
> export the sensors in BMC which is a service processor for POWER{8,9} servers.
> This patch adds a hwmon driver in the POWER9 server.
>

Both are names ...occ... kind of difficult to understand what is what.

Either case, it appears that there are some dependencies to other code
which I was not copied on. that is not very helpful.

Please use the new hwmon API (devm_hwmon_device_register_with_info); this driver
seems to be a perfect candidate. Also, please do not use function macros;
those tend to obfuscate the code and make it hard to understand.
With the new API, such macros should not be necessary.
Couple of additional comments inline.

Thanks,
Guenter

> Thanks and Regards,
> Shilpa
>
>> Thanks,
>> Guenter
>>
>>> Signed-off-by: Shilpasri G Bhat <shilpa.bhat@...ux.vnet.ibm.com>
>>> CC: Rob Herring <robh+dt@...nel.org>
>>> CC: Mark Rutland <mark.rutland@....com>
>>> CC: Jean Delvare <jdelvare@...e.com>
>>> CC: Guenter Roeck <linux@...ck-us.net>
>>> CC: Jonathan Corbet <corbet@....net>
>>> CC: devicetree@...r.kernel.org
>>> CC: linux-hwmon@...r.kernel.orgThat
>>> CC: linux-doc@...r.kernel.org
>>> ---
>>>  .../devicetree/bindings/hwmon/ibmpowernv-occ.txt   |   4 +
>>>  Documentation/hwmon/ibmpowernv-occ                 |  24 ++
>>>  drivers/hwmon/Kconfig                              |  11 +
>>>  drivers/hwmon/Makefile                             |   1 +
>>>  drivers/hwmon/ibmpowernv-occ.c                     | 302 +++++++++++++++++++++
>>>  5 files changed, 342 insertions(+)
>>>  create mode 100644 Documentation/devicetree/bindings/hwmon/ibmpowernv-occ.txt
>>>  create mode 100644 Documentation/hwmon/ibmpowernv-occ
>>>  create mode 100644 drivers/hwmon/ibmpowernv-occ.c
>>>
>>> diff --git a/Documentation/devicetree/bindings/hwmon/ibmpowernv-occ.txt b/Documentation/devicetree/bindings/hwmon/ibmpowernv-occ.txt
>>> new file mode 100644
>>> index 0000000..d03f744
>>> --- /dev/null
>>> +++ b/Documentation/devicetree/bindings/hwmon/ibmpowernv-occ.txt
>>> @@ -0,0 +1,4 @@
>>> +IBM POWERNV OCC inband platform sensors
>>> +
>>> +Required device-tree property:
>>> +- compatible: "ibm,p9-occ-inband-sensor"
>>> diff --git a/Documentation/hwmon/ibmpowernv-occ b/Documentation/hwmon/ibmpowernv-occ
>>> new file mode 100644
>>> index 0000000..151028b
>>> --- /dev/null
>>> +++ b/Documentation/hwmon/ibmpowernv-occ
>>> @@ -0,0 +1,24 @@
>>> +Kernel driver ibmpowernv-occ
>>> +=============================
>>> +
>>> +Supported systems:
>>> + * P9 server based on POWERNV platform
>>> +
>>> +Description
>>> +------------
>>> +
>>> +This driver exports the power and temperature sensors from OCC inband
>>> +sensors on P9 POWERNV platforms.
>>> +
>>> +Sysfs attributes
>>> +----------------
>>> +
>>> +powerX_input		Latest power reading
>>> +powerX_input_highest	Minimum power
>>> +powerX_input_lowest	Maximum power
>>> +powerX_label		Sensor name
>>> +
>>> +tempX_input		Latest temperature reading
>>> +tempX_max		Minimum temperature
>>> +tempX_min		Maximum temperature
>>> +tempX_label		Sensor name
>>> diff --git a/drivers/hwmon/Kconfig b/drivers/hwmon/Kconfig
>>> index 0649d53f3..3b1dbb9 100644
>>> --- a/drivers/hwmon/Kconfig
>>> +++ b/drivers/hwmon/Kconfig
>>> @@ -598,6 +598,17 @@ config SENSORS_IBMPOWERNV
>>>  	  This driver can also be built as a module. If so, the module
>>>  	  will be called ibmpowernv.
>>>
>>> +config SENSORS_IBMPOWERNV_OCC
>>> +	tristate "IBM POWERNV OCC Inband platform sensors"
>>> +	depends on PPC_POWERNV
>>> +	default y
>>> +	help
>>> +	  If you say yes here you get support for the temperature/power
>>> +	  OCC inband sensors on your PowerNV platform.
>>> +
>>> +	  This driver can also be built as a module. If so, the module
>>> +	  will be called ibmpowernv-occ.
>>> +
>>>  config SENSORS_IIO_HWMON
>>>  	tristate "Hwmon driver that uses channels specified via iio maps"
>>>  	depends on IIO
>>> diff --git a/drivers/hwmon/Makefile b/drivers/hwmon/Makefile
>>> index 5509edf..0da2207 100644
>>> --- a/drivers/hwmon/Makefile
>>> +++ b/drivers/hwmon/Makefile
>>> @@ -75,6 +75,7 @@ obj-$(CONFIG_SENSORS_I5K_AMB)	+= i5k_amb.o
>>>  obj-$(CONFIG_SENSORS_IBMAEM)	+= ibmaem.o
>>>  obj-$(CONFIG_SENSORS_IBMPEX)	+= ibmpex.o
>>>  obj-$(CONFIG_SENSORS_IBMPOWERNV)+= ibmpowernv.o
>>> +obj-$(CONFIG_SENSORS_IBMPOWERNV_OCC)+= ibmpowernv-occ.o
>>>  obj-$(CONFIG_SENSORS_IIO_HWMON) += iio_hwmon.o
>>>  obj-$(CONFIG_SENSORS_INA209)	+= ina209.o
>>>  obj-$(CONFIG_SENSORS_INA2XX)	+= ina2xx.o
>>> diff --git a/drivers/hwmon/ibmpowernv-occ.c b/drivers/hwmon/ibmpowernv-occ.c
>>> new file mode 100644
>>> index 0000000..97b1bbe
>>> --- /dev/null
>>> +++ b/drivers/hwmon/ibmpowernv-occ.c
>>> @@ -0,0 +1,302 @@
>>> +/*
>>> + * IBM PowerNV platform OCC inband sensors for temperature/power
>>> + * Copyright (C) 2017 IBM
>>> + *
>>> + * This program is free software; you can redistribute it and/or modify
>>> + * it under the terms of the GNU General Public License as published by
>>> + * the Free Software Foundation; either version 2 of the License, or
>>> + * (at your option) any later version.
>>> + *
>>> + * This program is distributed in the hope that it will be useful,
>>> + * but WITHOUT ANY WARRANTY; without even the implied warranty of
>>> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
>>> + * GNU General Public License for more details.
>>> + *
>>> + * You should have received a copy of the GNU General Public License
>>> + * along with this program.

Unnecessary.

>>> + */
>>> +
>>> +#define DRVNAME		"ibmpowernv_occ"
>>> +#define pr_fmt(fmt)	DRVNAME ": " fmt
>>> +
>>> +#include <linux/module.h>
>>> +#include <linux/hwmon.h>
>>> +#include <linux/slab.h>
>>> +#include <linux/platform_device.h>

alphabetic order please

>>> +
>>> +#include <asm/opal.h>
>>> +
>>> +#define MAX_HWMON_ATTR_LEN	32
>>> +#define MAX_HWMON_LABEL_LEN	(MAX_OCC_SENSOR_NAME_LEN * 2)
>>> +#define HWMON_ATTRS_PER_SENSOR	16
>>> +#define TO_MILLI_UNITS(x)	((x) * 1000)
>>> +#define TO_MICRO_UNITS(x)	((x) * 1000000)
>>> +
>>> +enum sensors {
>>> +	TEMP,
>>> +	POWER,
>>> +	MAX_SENSOR_TYPE,
>>> +};
>>> +
>>> +static struct sensor_type {
>>> +	const char *name;
>>> +	int hwmon_id;
>>> +} sensor_types[] = {
>>> +	{ "temp"},
>>> +	{ "power"},
>>> +};
>>> +
>>> +static struct sensor_data {
>>> +	u32 occ_id;
>>> +	u64 offset;  /* Offset to ping/pong reading buffer */
>>> +	enum sensors type;
>>> +	char label[MAX_HWMON_LABEL_LEN];
>>> +	char name[MAX_HWMON_ATTR_LEN];
>>> +	struct device_attribute attr;
>>> +} *sdata;
>>> +
>>> +static struct attribute_group sensor_attrs_group;
>>> +__ATTRIBUTE_GROUPS(sensor_attrs);
>>> +
>>> +#define show(file_name)							\
>>> +static ssize_t ibmpowernv_occ_show_##file_name				\
>>> +(struct device *dev, struct device_attribute *dattr, char *buf)		\
>>> +{									\
>>> +	struct sensor_data *sdata = container_of(dattr,			\
>>> +						 struct sensor_data,	\
>>> +						 attr);			\
>>> +	u64 val;							\
>>> +	int ret;							\

this does or should generate a checkpatch warning.

>>> +	ret = opal_occ_sensor_get_##file_name(sdata->occ_id,		\
>>> +					      sdata->offset,		\
>>> +					      &val);			\
>>> +	if (ret)							\
>>> +		return ret;						\
>>> +	if (sdata->type == TEMP)					\
>>> +		val = TO_MILLI_UNITS(val);				\
>>> +	else if (sdata->type == POWER)					\
>>> +		val = TO_MICRO_UNITS(val);				\
>>> +	return sprintf(buf, "%llu\n", val);				\
>>> +}
>>> +
>>> +show(sample);
>>> +show(max);
>>> +show(min);
>>> +show(js_min);
>>> +show(js_max);
>>> +show(csm_min);
>>> +show(csm_max);
>>> +show(prof_min);
>>> +show(prof_max);
>>> +
>>> +static struct sensor_view_groups {
>>> +	const char *name;
>>> +	ssize_t (*show_sample)(struct device *dev,
>>> +			       struct device_attribute *attr,
>>> +			       char *buf);
>>> +	ssize_t (*show_min)(struct device *dev,
>>> +			    struct device_attribute *attr,
>>> +			    char *buf);
>>> +	ssize_t (*show_max)(struct device *dev,
>>> +			    struct device_attribute *attr,
>>> +			    char *buf);
>>> +} sensor_views[] = {
>>> +	{
>>> +		.name		= "",
>>> +		.show_sample	= ibmpowernv_occ_show_sample,
>>> +		.show_min	= ibmpowernv_occ_show_min,
>>> +		.show_max	= ibmpowernv_occ_show_max
>>> +	},
>>> +	{
>>> +		.name		= "_JS",
>>> +		.show_sample	= ibmpowernv_occ_show_sample,
>>> +		.show_min	= ibmpowernv_occ_show_js_min,
>>> +		.show_max	= ibmpowernv_occ_show_js_max
>>> +	},
>>> +	{	.name		= "_CSM",
>>> +		.show_sample	= ibmpowernv_occ_show_sample,
>>> +		.show_min	= ibmpowernv_occ_show_csm_min,
>>> +		.show_max	= ibmpowernv_occ_show_csm_max
>>> +	},
>>> +	{	.name		= "_Prof",
>>> +		.show_sample	= ibmpowernv_occ_show_sample,
>>> +		.show_min	= ibmpowernv_occ_show_prof_min,
>>> +		.show_max	= ibmpowernv_occ_show_prof_max
>>> +	},
>>> +};
>>> +
>>> +static ssize_t ibmpowernv_occ_show_label(struct device *dev,
>>> +					 struct device_attribute *dattr,
>>> +					 char *buf)
>>> +{
>>> +	struct sensor_data *sdata = container_of(dattr, struct sensor_data,
>>> +						 attr);
>>> +
>>> +	return sprintf(buf, "%s\n", sdata->label);
>>> +}
>>> +
>>> +static int ibmpowernv_occ_get_sensor_type(enum occ_sensor_type type)
>>> +{
>>> +	switch (type) {
>>> +	case OCC_SENSOR_TYPE_POWER:
>>> +		return POWER;
>>> +	case OCC_SENSOR_TYPE_TEMPERATURE:
>>> +		return TEMP;
>>> +	default:
>>> +		return MAX_SENSOR_TYPE;
>>> +	}
>>> +
>>> +	return MAX_SENSOR_TYPE;

Never reached.

>>> +}
>>> +
>>> +static void ibmpowernv_occ_add_sdata(struct occ_hwmon_sensor sensor,

Passing a data structure (instead of a pointer to it) as parameter is quite
unusual. Is this really needed ? I don't really see the point.

>>> +				     struct sensor_data *sdata, char *name,
>>> +				     int hwmon_id, enum sensors type,
>>> +				     ssize_t (*show)(struct device *dev,
>>> +						struct device_attribute *attr,
>>> +						char *buf))
>>> +{
>>> +	sdata->type = type;
>>> +	sdata->occ_id = sensor.occ_id;
>>> +	sdata->offset = sensor.offset;
>>> +	snprintf(sdata->name, MAX_HWMON_ATTR_LEN, "%s%d_%s",
>>> +		 sensor_types[type].name, hwmon_id, name);
>>> +	sysfs_attr_init(&sdata->attr.attr);
>>> +	sdata->attr.attr.name = sdata->name;
>>> +	sdata->attr.attr.mode = 0444;
>>> +	sdata->attr.show = show;
>>> +}
>>> +
>>> +static void ibmpowernv_occ_add_sensor_attrs(struct occ_hwmon_sensor sensor,
>>> +					    int index)
>>> +{
>>> +	struct attribute **attrs = sensor_attrs_group.attrs;
>>> +	char attr_str[MAX_HWMON_ATTR_LEN];
>>> +	enum sensors type = ibmpowernv_occ_get_sensor_type(sensor.type);
>>> +	int i;
>>> +
>>> +	index *= HWMON_ATTRS_PER_SENSOR;
>>> +	for (i = 0; i < ARRAY_SIZE(sensor_views); i++) {
>>> +		int hid = ++sensor_types[type].hwmon_id;
>>> +
>>> +		/* input */
>>> +		ibmpowernv_occ_add_sdata(sensor, &sdata[index], "input", hid,
>>> +					 type, sensor_views[i].show_sample);
>>> +		attrs[index] = &sdata[index].attr.attr;
>>> +		index++;
>>> +
>>> +		/* min */
>>> +		if (type == POWER)
>>> +			snprintf(attr_str, MAX_HWMON_ATTR_LEN, "%s",
>>> +				 "input_lowest");
>>> +		else
>>> +			snprintf(attr_str, MAX_HWMON_ATTR_LEN, "%s", "min");
>>> +
>>> +		ibmpowernv_occ_add_sdata(sensor, &sdata[index], attr_str, hid,
>>> +					 type, sensor_views[i].show_min);
>>> +		attrs[index] = &sdata[index].attr.attr;
>>> +		index++;
>>> +
>>> +		/* max */
>>> +		if (type == POWER)
>>> +			snprintf(attr_str, MAX_HWMON_ATTR_LEN, "%s",
>>> +				 "input_highest");
>>> +		else
>>> +			snprintf(attr_str, MAX_HWMON_ATTR_LEN, "%s", "max");

I don't really see the point of those snprintf()s and for having attr_str[]
(instead of char *attr_str).

>>> +
>>> +		ibmpowernv_occ_add_sdata(sensor, &sdata[index], attr_str, hid,
>>> +					 type, sensor_views[i].show_max);
>>> +		attrs[index] = &sdata[index].attr.attr;
>>> +		index++;
>>> +
>>> +		/* label */
>>> +		snprintf(sdata[index].label, MAX_HWMON_LABEL_LEN, "%s%s",
>>> +			 sensor.name, sensor_views[i].name);
>>> +		ibmpowernv_occ_add_sdata(sensor, &sdata[index], "label", hid,
>>> +					 type, ibmpowernv_occ_show_label);
>>> +		attrs[index] = &sdata[index].attr.attr;
>>> +		index++;
>>> +	}
>>> +}
>>> +
>>> +static int ibmpowernv_occ_add_device_attrs(struct platform_device *pdev)
>>> +{
>>> +	struct attribute **attrs;
>>> +	struct occ_hwmon_sensor *slist = NULL;

Unnecessary initialization.

>>> +	int nr_sensors = 0, i;
>>> +	int ret = -ENOMEM;
>>> +
>>> +	slist = opal_occ_sensor_get_hwmon_list(&nr_sensors);
>>> +	if (!nr_sensors)
>>> +		return -ENODEV;
>>> +
>>> +	if (!slist)
>>> +		return ret;
>>> +

is nr_sensors guaranteed to be valid if slist is NULL ?

>>> +	sdata = devm_kzalloc(&pdev->dev, nr_sensors * sizeof(*sdata) *
>>> +			     HWMON_ATTRS_PER_SENSOR, GFP_KERNEL);
>>> +	if (!sdata)
>>> +		goto out;
>>> +
>>> +	attrs = devm_kzalloc(&pdev->dev, nr_sensors * sizeof(*attrs) *
>>> +			     HWMON_ATTRS_PER_SENSOR, GFP_KERNEL);
>>> +	if (!attrs)
>>> +		goto out;
>>> +
>>> +	sensor_attrs_group.attrs = attrs;
>>> +	for (i = 0; i < nr_sensors; i++)
>>> +		ibmpowernv_occ_add_sensor_attrs(slist[i], i);
>>> +
>>> +	ret = 0;
>>> +out:
>>> +	kfree(slist);
>>> +	return ret;
>>> +}
>>> +
>>> +static int ibmpowernv_occ_probe(struct platform_device *pdev)
>>> +{
>>> +	struct device *hwmon_dev;
>>> +	int err;
>>> +
>>> +	err = ibmpowernv_occ_add_device_attrs(pdev);
>>> +	if (err)
>>> +		goto out;
>>> +

For -ENODEV and -ENOMEM you don't really want/need an error message.

>>> +	hwmon_dev = devm_hwmon_device_register_with_groups(&pdev->dev, DRVNAME,
>>> +							   NULL,
>>> +							   sensor_attrs_groups);
>>> +
>>> +	err = PTR_ERR_OR_ZERO(hwmon_dev);
>>> +out:
>>> +	if (err)
>>> +		pr_warn("Failed to initialize Hwmon OCC inband sensors\n");

dev_warn() if this is really neceessary. I would suggest a more specific
error message, though. This doesn't tell if ibmpowernv_occ_add_device_attrs()
or devm_hwmon_device_register_with_groups() failed.

>>> +
>>> +	return err;
>>> +}
>>> +
>>> +static const struct platform_device_id occ_sensor_ids[] = {
>>> +	{ .name = "occ-inband-sensor" },
>>> +	{ }
>>> +};
>>> +MODULE_DEVICE_TABLE(platform, occ_sensor_ids);
>>> +
>>> +static const struct of_device_id occ_sensor_of_ids[] = {
>>> +	{ .compatible	= "ibm,p9-occ-inband-sensor" },
>>> +	{ }
>>> +};
>>> +MODULE_DEVICE_TABLE(of, occ_sensor_of_ids);
>>> +
>>> +static struct platform_driver ibmpowernv_occ_driver = {
>>> +	.probe		= ibmpowernv_occ_probe,
>>> +	.id_table	= occ_sensor_ids,
>>> +	.driver		= {
>>> +		.name	= DRVNAME,
>>> +		.of_match_table	= occ_sensor_of_ids,
>>> +	},
>>> +};
>>> +
>>> +module_platform_driver(ibmpowernv_occ_driver);
>>> +
>>> +MODULE_AUTHOR("Shilpasri G Bhat <shilpa.bhat@...ux.vnet.ibm.com>");
>>> +MODULE_DESCRIPTION("IBM POWERNV platform OCC inband sensors");
>>> +MODULE_LICENSE("GPL");
>>> --
>>> 1.8.3.1
>>>
>>
>
>

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ