Re: [RFC/PATCH 1/8]: CGroup Files: Add locking mode to cgroups control files

[Li Zefan <lizf@cn.fujitsu.com>]

>>  > @@ -1518,16 +1580,21 @@ static ssize_t cgroup_file_read(struct f
>>  >       struct cftype *cft = __d_cft(file->f_dentry);
>>  >       struct cgroup *cgrp = __d_cgrp(file->f_dentry->d_parent);
>>  >
>>  > -     if (!cft || cgroup_is_removed(cgrp))
>>  > +     if (cgroup_is_removed(cgrp))
>>  >               return -ENODEV;
>>  >
>>
>>  This check seems redundant now.
>>
> 
> It's not needed for safety, but it doesn't seem to hurt to check
> cgroup_is_removed() prior to doing any copying, since we'll fail after
> copying anyway if cgroup_is_removed() returns true (once we've taken
> any relevant locks).
> 

It's a bit odd to me that cgroup_is_removed() is removed in cgroup_file_write(),
but it's ressered here.