[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [PATCH 01/11] kset: fix documentation for kset_register()
From: |
Luben Tuikov |
Subject: |
Re: [PATCH 01/11] kset: fix documentation for kset_register() |
Date: |
Fri, 21 Oct 2022 04:18:28 -0400 |
User-agent: |
Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101 Thunderbird/102.3.3 |
On 2022-10-21 04:05, Yang Yingliang wrote:
>
> On 2022/10/21 13:34, Luben Tuikov wrote:
>> On 2022-10-20 22:20, Yang Yingliang wrote:
>>> kset_register() is currently used in some places without calling
>>> kset_put() in error path, because the callers think it should be
>>> kset internal thing to do, but the driver core can not know what
>>> caller doing with that memory at times. The memory could be freed
>>> both in kset_put() and error path of caller, if it is called in
>>> kset_register().
>>>
>>> So make the function documentation more explicit about calling
>>> kset_put() in the error path of caller.
>>>
>>> Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
>>> ---
>>> lib/kobject.c | 3 +++
>>> 1 file changed, 3 insertions(+)
>>>
>>> diff --git a/lib/kobject.c b/lib/kobject.c
>>> index a0b2dbfcfa23..6da04353d974 100644
>>> --- a/lib/kobject.c
>>> +++ b/lib/kobject.c
>>> @@ -834,6 +834,9 @@ EXPORT_SYMBOL_GPL(kobj_sysfs_ops);
>>> /**
>>> * kset_register() - Initialize and add a kset.
>>> * @k: kset.
>>> + *
>>> + * If this function returns an error, kset_put() must be called to
>>> + * properly clean up the memory associated with the object.
>>> */
>> And I'd continue the sentence, with " ... with the object,
>> for instance the memory for the kset.kobj.name when
>> kobj_set_name(&kset.kobj, format, ...)
>> was called before calling kset_register()."
> kobject_cleanup() not only frees name, but aslo calls ->release() to
> free another resources.
Yes, it does. For this reason I said "for instance..." I didn't want to include
this in case in the future if the code changes, the comment would be wrong. IOW,
I wanted to add the minimalist comment possible.
>>
>> This makes it clear what we want to make sure is freed, in case of an early
>> error
>> from kset_register().
>
> How about like this:
>
> If this function returns an error, kset_put() must be called to clean up the
> name of
> kset object and other memory associated with the object.
It's bit too wordy and redundant with what else it does--this can be gleaned
from the code. I'd say:
On error, kset_put() should be called to clean up at least
kset.kobj.name allocated
by kobj_set_name(&kset.kobj, format, ...).
This tells the reader the symmetry of the calls: kobj_set_name() -->
kset_register() --> kset_put();
Because if the code evolves to use other means of allocation, or if the the
user allocates a name
by different means, then they'll understand what to watch out for.
Regards,
Luben
>
>>
>> Regards,
>> Luben
>>
>>> int kset_register(struct kset *k)
>>> {
>> .
- [PATCH 00/11] fix memory leak while kset_register() fails, Yang Yingliang, 2022/10/20
- [PATCH 02/11] kset: add null pointer check in kset_put(), Yang Yingliang, 2022/10/20
- [PATCH 10/11] drm/amdgpu/discovery: fix possible memory leak, Yang Yingliang, 2022/10/20
- [PATCH 07/11] f2fs: fix possible memory leak in f2fs_init_sysfs(), Yang Yingliang, 2022/10/20
- [PATCH 11/11] ubifs: Fix memory leak in ubifs_sysfs_init(), Yang Yingliang, 2022/10/20
- [PATCH 06/11] firmware: qemu_fw_cfg: fix possible memory leak in fw_cfg_build_symlink(), Yang Yingliang, 2022/10/20
- [PATCH 01/11] kset: fix documentation for kset_register(), Yang Yingliang, 2022/10/20
- [PATCH 04/11] kobject: fix possible memory leak in kset_create_and_add(), Yang Yingliang, 2022/10/20
- [PATCH 08/11] erofs: fix possible memory leak in erofs_init_sysfs(), Yang Yingliang, 2022/10/20
- [PATCH 03/11] bus: fix possible memory leak in bus_register(), Yang Yingliang, 2022/10/20
- [PATCH 05/11] class: fix possible memory leak in __class_register(), Yang Yingliang, 2022/10/20
- [PATCH 09/11] ocfs2: possible memory leak in mlog_sys_init(), Yang Yingliang, 2022/10/20
- Re: [PATCH 00/11] fix memory leak while kset_register() fails, Luben Tuikov, 2022/10/21