mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Jie Zhan <zhanjie9@hisilicon.com>
To: "zhangpengjie (A)" <zhangpengjie2@huawei.com>,
	<myungjoo.ham@samsung.com>, <kyungmin.park@samsung.com>,
	<cw00.choi@samsung.com>
Cc: <linux-pm@vger.kernel.org>, <linux-kernel@vger.kernel.org>,
	<zhenglifeng1@huawei.com>, <lihuisong@huawei.com>,
	<yubowen8@huawei.com>, <linhongye@h-partners.com>,
	<linuxarm@huawei.com>, <jonathan.cameron@huawei.com>
Subject: Re: [PATCH v3] PM / devfreq: use _visible attribute to replace create/remove_sysfs_files()
Date: Thu, 4 Dec 2025 16:41:11 +0800	[thread overview]
Message-ID: <07c098d7-979b-44b5-aae0-7999581d8d90@hisilicon.com> (raw)
In-Reply-To: <1402fe0a-13ff-4e6c-bcea-660acb383177@huawei.com>



...

>>> +static umode_t gov_attr_visible(struct kobject *kobj,
>>> +                struct attribute *attr, int n)
>>>   {
>>> -    if (IS_SUPPORTED_ATTR(gov->attrs, POLLING_INTERVAL))
>>> -        CREATE_SYSFS_FILE(devfreq, polling_interval);
>>> -    if (IS_SUPPORTED_ATTR(gov->attrs, TIMER))
>>> -        CREATE_SYSFS_FILE(devfreq, timer);
>>> +    struct device *dev = kobj_to_dev(kobj);
>>> +    struct devfreq *df = to_devfreq(dev);
>>> +
>>> +    if (!df->governor || !df->governor->attrs)
>>> +        return 0;
>>> +
>>> +    if (IS_SUPPORTED_ATTR(df->governor->attrs, POLLING_INTERVAL))
>>> +        return attr->mode;
>>> +    if (IS_SUPPORTED_ATTR(df->governor->attrs, TIMER))
>>> +        return attr->mode;
>> This would expose both 'timer' and 'polling_interval' if either of them is
>> supported, which is wrong.
> 
> ok,i see,may be
> 
>     if (attr == &dev_attr_polling_interval.attr) {
>         if (IS_SUPPORTED_ATTR(df->governor->attrs, POLLING_INTERVAL))
>             return attr->mode;
>         return 0;
>     }
> 
>     if (attr == &dev_attr_timer.attr) {
>         if (IS_SUPPORTED_ATTR(df->governor->attrs, TIMER))
>             return attr->mode;
>         return 0;
>     }
> 
Yeah.

Could be cleaner like:
	if (attr == &dev_attr_polling_interval.attr &&
	    IS_SUPPORTED_ATTR(df->governor->attrs, POLLING_INTERVAL))
		return attr->mode;

	if (attr == &dev_attr_timer.attr &&
	    IS_SUPPORTED_ATTR(df->governor->attrs, TIMER))
		return attr->mode;

	return 0;
}
>>> +
>>> +    return 0;
>>>   }
>>>   -/* Remove the specific sysfs files which depend on each governor. */
>>> -static void remove_sysfs_files(struct devfreq *devfreq,
>>> -                const struct devfreq_governor *gov)
>>> +static bool gov_group_visible(struct kobject *kobj)
>>>   {
>>> -    if (IS_SUPPORTED_ATTR(gov->attrs, POLLING_INTERVAL))
>>> -        sysfs_remove_file(&devfreq->dev.kobj,
>>> -                &dev_attr_polling_interval.attr);
>>> -    if (IS_SUPPORTED_ATTR(gov->attrs, TIMER))
>>> -        sysfs_remove_file(&devfreq->dev.kobj, &dev_attr_timer.attr);
>>> +    struct device *dev = kobj_to_dev(kobj);
>>> +
>>> +    if (!dev)
>>> +        return false;
>> kobj_to_dev(kobj) can't fail.  This check is unnecessary.
>>> +
>>> +    return true;
>>>   }
>>> +DEFINE_SYSFS_GROUP_VISIBLE(gov);
>> The implementation of gov_group_visible() and gov_attr_visible() doesn't
>> seem to comply with the design of DEFINE_SYSFS_GROUP_VISIBLE().
>>
>> DEFINE_SYSFS_GROUP_VISIBLE is defined as:
>>
>> #define DEFINE_SYSFS_GROUP_VISIBLE(name)                             \
>>          static inline umode_t sysfs_group_visible_##name(            \
>>                  struct kobject *kobj, struct attribute *attr, int n) \
>>          {                                                            \
>>                  if (n == 0 && !name##_group_visible(kobj))           \
>>                          return SYSFS_GROUP_INVISIBLE;                \
>>                  return name##_attr_visible(kobj, attr, n);           \
>>          }
>>
>> That means:
>> _group_visible controls whether to hide all the attrs in this group;
>> _attr_visible further decides whether to show each attr.
>>
>> Hence,
>> 1. gov_group_visible() should check if any attr in 'governor_attrs' should
>> be present, i.e. checking df->governor->attrs.
>> 2. gov_attr_visible identifies which attr it is and checks whether the
>> governor supports it.
>>
>> Neither is done in the above code, so this should be updated.
> 
> we can move the check|if (!df->governor || !df->governor->attrs)|
> 
> from|gov_attr_visible|to|gov_group_visible|.
Exactly.
> 
> thanks !
> 

      reply	other threads:[~2025-12-04  8:41 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-11-07  3:17 Pengjie Zhang
2025-11-21 12:33 ` zhangpengjie (A)
2025-12-04  7:36 ` Jie Zhan
2025-12-04  8:29   ` zhangpengjie (A)
2025-12-04  8:41     ` Jie Zhan [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=07c098d7-979b-44b5-aae0-7999581d8d90@hisilicon.com \
    --to=zhanjie9@hisilicon.com \
    --cc=cw00.choi@samsung.com \
    --cc=jonathan.cameron@huawei.com \
    --cc=kyungmin.park@samsung.com \
    --cc=lihuisong@huawei.com \
    --cc=linhongye@h-partners.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pm@vger.kernel.org \
    --cc=linuxarm@huawei.com \
    --cc=myungjoo.ham@samsung.com \
    --cc=yubowen8@huawei.com \
    --cc=zhangpengjie2@huawei.com \
    --cc=zhenglifeng1@huawei.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

all inboxes | Powered by JetHome®