From: Paul Cercueil <paul@crapouillou.net>
To: Guenter Roeck <linux@roeck-us.net>
Cc: Jean Delvare <jdelvare@suse.com>,
Rob Herring <robh+dt@kernel.org>,
linux-hwmon@vger.kernel.org, devicetree@vger.kernel.org,
linux-kernel@vger.kernel.org,
Cosmin Tanislav <cosmin.tanislav@analog.com>
Subject: Re: [PATCH 2/2] hwmon: Add "label" attribute
Date: Wed, 22 Dec 2021 14:18:46 +0000 [thread overview]
Message-ID: <A3TI4R.0KWPRNVE7E7M3@crapouillou.net> (raw)
In-Reply-To: <78620358-737e-368e-e5f5-82a673adf26a@roeck-us.net>
Hi Guenter,
Comments noted, thanks. I'll send a V2 later today.
Cheers,
-Paul
Le mar., déc. 21 2021 at 10:17:03 -0800, Guenter Roeck
<linux@roeck-us.net> a écrit :
> On 12/21/21 9:50 AM, Paul Cercueil wrote:
>> If a label is defined in the device tree for this device add that
>> to the device specific attributes. This is useful for userspace to
>> be able to identify an individual device when multiple identical
>> chips are present in the system.
>>
>
> This is an ABI change which needs to be documented in
> Documentation/hwmon/sysfs-interface.rst.
>
>> Signed-off-by: Paul Cercueil <paul@crapouillou.net>
>> Tested-by: Cosmin Tanislav <cosmin.tanislav@analog.com>
>> ---
>> drivers/hwmon/hwmon.c | 22 +++++++++++++++++++++-
>> 1 file changed, 21 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/hwmon/hwmon.c b/drivers/hwmon/hwmon.c
>> index 3501a3ead4ba..15826260a463 100644
>> --- a/drivers/hwmon/hwmon.c
>> +++ b/drivers/hwmon/hwmon.c
>> @@ -71,8 +71,23 @@ name_show(struct device *dev, struct
>> device_attribute *attr, char *buf)
>> }
>> static DEVICE_ATTR_RO(name);
>> \x7f+static ssize_t
>> +label_show(struct device *dev, struct device_attribute *attr, char
>> *buf)
>> +{
>> + const char *label;
>> + int ret;
>> +
>> + ret = device_property_read_string(dev, "label", &label);
>
> Requires "#include <linux/property.h>". Also, reading and verifying
> the label
> each time it is read is excessive. More on that see below.
>
>> + if (ret < 0)
>> + return ret;
>> +
>> + return sysfs_emit(buf, "%s\n", label);
>> +}
>> +static DEVICE_ATTR_RO(label);
>> +
>> static struct attribute *hwmon_dev_attrs[] = {
>> &dev_attr_name.attr,
>> + &dev_attr_label.attr,
>> NULL
>> };
>> \x7f@@ -81,7 +96,12 @@ static umode_t
>> hwmon_dev_name_is_visible(struct kobject *kobj,
>
> That function name is no longer appropriate and should be changed to
> something like hwmon_dev_attr_is_visible.
>
>> {
>> struct device *dev = kobj_to_dev(kobj);
>> \x7f- if (to_hwmon_device(dev)->name == NULL)
>> + if (attr == &dev_attr_name.attr &&
>> + to_hwmon_device(dev)->name == NULL)
>
> Unnecessary continuation line.
>
>> + return 0;
>> +
>> + if (attr == &dev_attr_label.attr &&
>> + !device_property_present(dev, "label"))
>
> If the property is present but not a string, each read of "label"
> would
> return a runtime error. I don't like that. I would suggest to store
> a pointer to the label in struct hwmon_device, set it during
> registration
> (eg by using devm_strdup() if it is defined), and use
>
> if (attr == &dev_attr_label.attr && to_hwmon_device(dev)->label ==
> NULL)
> return 0;
>
> to check if it is present.
>
>> return 0;
>> \x7f return attr->mode;
>>
>
prev parent reply other threads:[~2021-12-22 14:18 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-12-21 17:50 [PATCH 0/2] hwmon: Add "label" attribute Paul Cercueil
2021-12-21 17:50 ` [PATCH 1/2] dt-bindings: hwmon: Introduce common properties Paul Cercueil
2021-12-22 19:36 ` Rob Herring
2021-12-23 9:19 ` Paul Cercueil
2021-12-23 10:12 ` Paul Cercueil
2022-01-05 9:41 ` Paul Cercueil
2021-12-21 17:50 ` [PATCH 2/2] hwmon: Add "label" attribute Paul Cercueil
2021-12-21 18:17 ` Guenter Roeck
2021-12-22 14:18 ` Paul Cercueil [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=A3TI4R.0KWPRNVE7E7M3@crapouillou.net \
--to=paul@crapouillou.net \
--cc=cosmin.tanislav@analog.com \
--cc=devicetree@vger.kernel.org \
--cc=jdelvare@suse.com \
--cc=linux-hwmon@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux@roeck-us.net \
--cc=robh+dt@kernel.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