Re: [PATCH] cgroup/cpuset: Improve cpuset_css_alloc() description

[Waiman Long <longman@redhat.com>]

On 11/16/22 10:40, Kamalesh Babulal wrote:
>
> On 11/16/22 20:06, Waiman Long wrote:
>> On 11/16/22 00:44, Kamalesh Babulal wrote:
>>> Change the function argument in the description of cpuset_css_alloc()
>>> from 'struct cgroup' -> 'struct cgroup_subsys_state'.  The change to the
>>> argument type was introduced by commit eb95419b023a ("cgroup: pass
>>> around cgroup_subsys_state instead of cgroup in subsystem methods").
>>> Also, add more information to its description.
>>>
>>> Signed-off-by: Kamalesh Babulal <kamalesh.babulal@oracle.com>
>>> ---
>>>    kernel/cgroup/cpuset.c | 11 ++++++++---
>>>    1 file changed, 8 insertions(+), 3 deletions(-)
>>>
>>> diff --git a/kernel/cgroup/cpuset.c b/kernel/cgroup/cpuset.c
>>> index b474289c15b8..aac790462e74 100644
>>> --- a/kernel/cgroup/cpuset.c
>>> +++ b/kernel/cgroup/cpuset.c
>>> @@ -3046,9 +3046,14 @@ static struct cftype dfl_files[] = {
>>>    };
>>>      -/*
>>> - *    cpuset_css_alloc - allocate a cpuset css
>>> - *    cgrp:    control group that the new cpuset will be part of
>>> +/**
>>> + * cpuset_css_alloc - Allocate a cpuset css
>>> + * @parent_css: Parent css of the control group that the new cpuset will be
>>> + *              part of
>>> + * Return: cpuset css on success, -ENOMEM on failure.
>>> + *
>>> + * Allocate and initialize a new cpuset css, for non-root cpuset or return the
>>> + * top cpuset css for root cpuset.
>> Strictly speaking, it returns the css of top cpuset set for NULL input parameter.
> Thank you for reviewing the patch. Does the following, rephrase, sound right:
>
> Allocate and initialize a new cpuset css, for non-NULL @parent_css, return
> top cpuset css otherwise.

That looks fine to me.

Cheers,
Longman

>>>     */
>>>      static struct cgroup_subsys_state *
>> While at it, could you also remove the blank line between the comment and the function body.
> Sure, will remove it.
>