提交 223dbc38 编写于 作者: T Tejun Heo

cgroup: bounce cgroup_subsys_state ref kill confirmation to a work item

css (cgroup_subsys_state) offlining, which requires process context,
will be moved to ref kill confirmation.  In preparation, bounce
css_killed handling through css->destroy_work.

css_ref_killed_fn() is renamed to css_killed_ref_fn() so that it's
consistent with the new css_killed_work_fn().

This patch adds an additional work item bouncing but doesn't change
the actual logic.
Signed-off-by: NTejun Heo <tj@kernel.org>
Acked-by: NLi Zefan <lizefan@huawei.com>
上级 ae7f164a
......@@ -4555,12 +4555,27 @@ static void cgroup_css_killed(struct cgroup *cgrp)
schedule_work(&cgrp->destroy_work);
}
static void css_ref_killed_fn(struct percpu_ref *ref)
/*
* This is called when the refcnt of a css is confirmed to be killed.
* css_tryget() is now guaranteed to fail.
*/
static void css_killed_work_fn(struct work_struct *work)
{
struct cgroup_subsys_state *css =
container_of(work, struct cgroup_subsys_state, destroy_work);
struct cgroup *cgrp = css->cgroup;
cgroup_css_killed(cgrp);
}
/* css kill confirmation processing requires process context, bounce */
static void css_killed_ref_fn(struct percpu_ref *ref)
{
struct cgroup_subsys_state *css =
container_of(ref, struct cgroup_subsys_state, refcnt);
cgroup_css_killed(css->cgroup);
INIT_WORK(&css->destroy_work, css_killed_work_fn);
schedule_work(&css->destroy_work);
}
/**
......@@ -4634,7 +4649,7 @@ static int cgroup_destroy_locked(struct cgroup *cgrp)
percpu_ref_get(&css->refcnt);
atomic_inc(&cgrp->css_kill_cnt);
percpu_ref_kill_and_confirm(&css->refcnt, css_ref_killed_fn);
percpu_ref_kill_and_confirm(&css->refcnt, css_killed_ref_fn);
}
cgroup_css_killed(cgrp);
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册