[PATCH] cpuset: use rcu_read_lock() to protect task_cs()

[PATCH] cpuset: use rcu_read_lock() to protect task_cs()

[Li Zefan]

We no longer use task_lock() to protect tsk->cgroups.

Reported-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: Li Zefan <lizefan@huawei.com>
---
 kernel/cpuset.c | 24 +++++++++++++-----------
 1 file changed, 13 insertions(+), 11 deletions(-)

diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index c63a0d9..4b3560d 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -2235,10 +2235,10 @@ void cpuset_cpus_allowed(struct task_struct *tsk, struct cpumask *pmask)
 	struct cpuset *cpus_cs;
 
 	mutex_lock(&callback_mutex);
-	task_lock(tsk);
+	rcu_read_lock();
 	cpus_cs = effective_cpumask_cpuset(task_cs(tsk));
 	guarantee_online_cpus(cpus_cs, pmask);
-	task_unlock(tsk);
+	rcu_read_unlock();
 	mutex_unlock(&callback_mutex);
 }
 
@@ -2291,10 +2291,10 @@ nodemask_t cpuset_mems_allowed(struct task_struct *tsk)
 	nodemask_t mask;
 
 	mutex_lock(&callback_mutex);
-	task_lock(tsk);
+	rcu_read_lock();
 	mems_cs = effective_nodemask_cpuset(task_cs(tsk));
 	guarantee_online_mems(mems_cs, &mask);
-	task_unlock(tsk);
+	rcu_read_unlock();
 	mutex_unlock(&callback_mutex);
 
 	return mask;
@@ -2410,10 +2410,10 @@ int __cpuset_node_allowed_softwall(int node, gfp_t gfp_mask)
 	/* Not hardwall and node outside mems_allowed: scan up cpusets */
 	mutex_lock(&callback_mutex);
 
-	task_lock(current);
+	rcu_read_lock();
 	cs = nearest_hardwall_ancestor(task_cs(current));
 	allowed = node_isset(node, cs->mems_allowed);
-	task_unlock(current);
+	rcu_read_unlock();
 
 	mutex_unlock(&callback_mutex);
 	return allowed;
@@ -2539,24 +2539,26 @@ int cpuset_mems_allowed_intersects(const struct task_struct *tsk1,
  * @task: pointer to task_struct of some task.
  *
  * Description: Prints @task's name, cpuset name, and cached copy of its
- * mems_allowed to the kernel log.  Must hold task_lock(task) to allow
- * dereferencing task_cs(task).
+ * mems_allowed to the kernel log.
  */
 void cpuset_print_task_mems_allowed(struct task_struct *tsk)
 {
 	 /* Statically allocated to prevent using excess stack. */
 	static char cpuset_nodelist[CPUSET_NODELIST_LEN];
 	static DEFINE_SPINLOCK(cpuset_buffer_lock);
-	struct cgroup *cgrp = task_cs(tsk)->css.cgroup;
+	struct cgroup *cgrp;
 
 	spin_lock(&cpuset_buffer_lock);
+	rcu_read_lock();
 
+	cgrp = task_cs(tsk)->css.cgroup;
 	nodelist_scnprintf(cpuset_nodelist, CPUSET_NODELIST_LEN,
 			   tsk->mems_allowed);
 	printk(KERN_INFO "%s cpuset=", tsk->comm);
 	pr_cont_cgroup_name(cgrp);
 	pr_cont(" mems_allowed=%s\n", cpuset_nodelist);
 
+	rcu_read_unlock();
 	spin_unlock(&cpuset_buffer_lock);
 }
 
@@ -2588,9 +2590,9 @@ int cpuset_memory_pressure_enabled __read_mostly;
 
 void __cpuset_memory_pressure_bump(void)
 {
-	task_lock(current);
+	rcu_read_lock();
 	fmeter_markevent(&task_cs(current)->fmeter);
-	task_unlock(current);
+	rcu_read_unlock();
 }
 
 #ifdef CONFIG_PROC_PID_CPUSET
-- 
1.8.0.2

[PATCH] cpuset: use rcu_read_lock() to protect task_cs()

[Li Zefan]

We no longer use task_lock() to protect tsk->cgroups.

Reported-by: Fengguang Wu <fengguang.wu-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
Signed-off-by: Li Zefan <lizefan-hv44wF8Li93QT0dZR+AlfA@public.gmane.org>
---
 kernel/cpuset.c | 24 +++++++++++++-----------
 1 file changed, 13 insertions(+), 11 deletions(-)

diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index c63a0d9..4b3560d 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -2235,10 +2235,10 @@ void cpuset_cpus_allowed(struct task_struct *tsk, struct cpumask *pmask)
 	struct cpuset *cpus_cs;
 
 	mutex_lock(&callback_mutex);
-	task_lock(tsk);
+	rcu_read_lock();
 	cpus_cs = effective_cpumask_cpuset(task_cs(tsk));
 	guarantee_online_cpus(cpus_cs, pmask);
-	task_unlock(tsk);
+	rcu_read_unlock();
 	mutex_unlock(&callback_mutex);
 }
 
@@ -2291,10 +2291,10 @@ nodemask_t cpuset_mems_allowed(struct task_struct *tsk)
 	nodemask_t mask;
 
 	mutex_lock(&callback_mutex);
-	task_lock(tsk);
+	rcu_read_lock();
 	mems_cs = effective_nodemask_cpuset(task_cs(tsk));
 	guarantee_online_mems(mems_cs, &mask);
-	task_unlock(tsk);
+	rcu_read_unlock();
 	mutex_unlock(&callback_mutex);
 
 	return mask;
@@ -2410,10 +2410,10 @@ int __cpuset_node_allowed_softwall(int node, gfp_t gfp_mask)
 	/* Not hardwall and node outside mems_allowed: scan up cpusets */
 	mutex_lock(&callback_mutex);
 
-	task_lock(current);
+	rcu_read_lock();
 	cs = nearest_hardwall_ancestor(task_cs(current));
 	allowed = node_isset(node, cs->mems_allowed);
-	task_unlock(current);
+	rcu_read_unlock();
 
 	mutex_unlock(&callback_mutex);
 	return allowed;
@@ -2539,24 +2539,26 @@ int cpuset_mems_allowed_intersects(const struct task_struct *tsk1,
  * @task: pointer to task_struct of some task.
  *
  * Description: Prints @task's name, cpuset name, and cached copy of its
- * mems_allowed to the kernel log.  Must hold task_lock(task) to allow
- * dereferencing task_cs(task).
+ * mems_allowed to the kernel log.
  */
 void cpuset_print_task_mems_allowed(struct task_struct *tsk)
 {
 	 /* Statically allocated to prevent using excess stack. */
 	static char cpuset_nodelist[CPUSET_NODELIST_LEN];
 	static DEFINE_SPINLOCK(cpuset_buffer_lock);
-	struct cgroup *cgrp = task_cs(tsk)->css.cgroup;
+	struct cgroup *cgrp;
 
 	spin_lock(&cpuset_buffer_lock);
+	rcu_read_lock();
 
+	cgrp = task_cs(tsk)->css.cgroup;
 	nodelist_scnprintf(cpuset_nodelist, CPUSET_NODELIST_LEN,
 			   tsk->mems_allowed);
 	printk(KERN_INFO "%s cpuset=", tsk->comm);
 	pr_cont_cgroup_name(cgrp);
 	pr_cont(" mems_allowed=%s\n", cpuset_nodelist);
 
+	rcu_read_unlock();
 	spin_unlock(&cpuset_buffer_lock);
 }
 
@@ -2588,9 +2590,9 @@ int cpuset_memory_pressure_enabled __read_mostly;
 
 void __cpuset_memory_pressure_bump(void)
 {
-	task_lock(current);
+	rcu_read_lock();
 	fmeter_markevent(&task_cs(current)->fmeter);
-	task_unlock(current);
+	rcu_read_unlock();
 }
 
 #ifdef CONFIG_PROC_PID_CPUSET
-- 
1.8.0.2

Re: [PATCH] cpuset: use rcu_read_lock() to protect task_cs()

[Tejun Heo]

On Mon, Mar 03, 2014 at 04:49:39PM +0800, Li Zefan wrote:
> We no longer use task_lock() to protect tsk->cgroups.
> 
> Reported-by: Fengguang Wu <fengguang.wu-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
> Signed-off-by: Li Zefan <lizefan-hv44wF8Li93QT0dZR+AlfA@public.gmane.org>

Applied to for-3.15 (do we need this for 3.14-fixes?).

Thanks.

-- 
tejun