cgroup: add __rcu modifier to cgroup->subsys[]
For the planned unified hierarchy, each css (cgroup_subsys_state) will
be RCU protected so that it can be created and destroyed individually
while allowing RCU accesses. Previous changes ensured that all
cgroup->subsys[] accesses use the cgroup_css() accessor. This patch
adds __rcu modifier to cgroup->subsys[], add matching RCU dereference
in cgroup_css() and convert all assignments to either
rcu_assign_pointer() or RCU_INIT_POINTER().
This change prepares for the actual RCUfication of css's and doesn't
introduce any visible behavior change. The conversion is verified
with sparse and all accesses are properly RCU annotated.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Li Zefan <lizefan@huawei.com>
diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h
index 8a5dc91..eb200b5 100644
--- a/include/linux/cgroup.h
+++ b/include/linux/cgroup.h
@@ -204,7 +204,7 @@
struct cgroup_name __rcu *name;
/* Private pointers for each registered subsystem */
- struct cgroup_subsys_state *subsys[CGROUP_SUBSYS_COUNT];
+ struct cgroup_subsys_state __rcu *subsys[CGROUP_SUBSYS_COUNT];
struct cgroupfs_root *root;
diff --git a/kernel/cgroup.c b/kernel/cgroup.c
index d63beffd..c271016 100644
--- a/kernel/cgroup.c
+++ b/kernel/cgroup.c
@@ -229,11 +229,16 @@
* @subsys_id: the subsystem of interest
*
* Return @cgrp's css (cgroup_subsys_state) associated with @subsys_id.
+ * This function must be called either under cgroup_mutex or
+ * rcu_read_lock() and the caller is responsible for pinning the returned
+ * css if it wants to keep accessing it outside the said locks. This
+ * function may return %NULL if @cgrp doesn't have @subsys_id enabled.
*/
static struct cgroup_subsys_state *cgroup_css(struct cgroup *cgrp,
int subsys_id)
{
- return cgrp->subsys[subsys_id];
+ return rcu_dereference_check(cgrp->subsys[subsys_id],
+ lockdep_is_held(&cgroup_mutex));
}
/* convenient tests for these bits */
@@ -1072,8 +1077,10 @@
BUG_ON(!cgroup_css(cgroup_dummy_top, i));
BUG_ON(cgroup_css(cgroup_dummy_top, i)->cgroup != cgroup_dummy_top);
- cgrp->subsys[i] = cgroup_dummy_top->subsys[i];
+ rcu_assign_pointer(cgrp->subsys[i],
+ cgroup_css(cgroup_dummy_top, i));
cgroup_css(cgrp, i)->cgroup = cgrp;
+
list_move(&ss->sibling, &root->subsys_list);
ss->root = root;
if (ss->bind)
@@ -1088,8 +1095,10 @@
if (ss->bind)
ss->bind(cgroup_css(cgroup_dummy_top, i));
+
cgroup_css(cgroup_dummy_top, i)->cgroup = cgroup_dummy_top;
- cgrp->subsys[i] = NULL;
+ RCU_INIT_POINTER(cgrp->subsys[i], NULL);
+
cgroup_subsys[i]->root = &cgroup_dummy_root;
list_move(&ss->sibling, &cgroup_dummy_root.subsys_list);
@@ -4314,7 +4323,7 @@
css->flags |= CSS_ROOT;
BUG_ON(cgroup_css(cgrp, ss->subsys_id));
- cgrp->subsys[ss->subsys_id] = css;
+ rcu_assign_pointer(cgrp->subsys[ss->subsys_id], css);
}
/* invoke ->css_online() on a new CSS and mark it online if successful */
@@ -4962,7 +4971,7 @@
* also takes care of freeing the css_id.
*/
ss->css_free(cgroup_css(cgroup_dummy_top, ss->subsys_id));
- cgroup_dummy_top->subsys[ss->subsys_id] = NULL;
+ RCU_INIT_POINTER(cgroup_dummy_top->subsys[ss->subsys_id], NULL);
mutex_unlock(&cgroup_mutex);
}
