Commit 452477fa authored by Tejun Heo's avatar Tejun Heo
Browse files

cpuset: pin down cpus and mems while a task is being attached 



cpuset is scheduled to be decoupled from cgroup_lock which will make
configuration updates race with task migration.  Any config update
will be allowed to happen between ->can_attach() and ->attach().  If
such config update removes either all cpus or mems, by the time
->attach() is called, the condition verified by ->can_attach(), that
the cpuset is capable of hosting the tasks, is no longer true.

This patch adds cpuset->attach_in_progress which is incremented from
->can_attach() and decremented when the attach operation finishes
either successfully or not.  validate_change() treats cpusets w/
non-zero ->attach_in_progress like cpusets w/ tasks and refuses to
remove all cpus or mems from it.

This currently doesn't make any functional difference as everything is
protected by cgroup_mutex but enables decoupling the locking.

Signed-off-by: default avatarTejun Heo <tj@kernel.org>
Acked-by: default avatarLi Zefan <lizefan@huawei.com>
parent 8d033948

kernel/cpuset.c

+26 −2
Original line number Diff line number Diff line
@@ -91,6 +91,12 @@ struct cpuset { 


	struct fmeter fmeter;		/* memory_pressure filter */



	/*

	 * Tasks are being attached to this cpuset.  Used to prevent

	 * zeroing cpus/mems_allowed between ->can_attach() and ->attach().

	 */

	int attach_in_progress;



	/* partition number for rebuild_sched_domains() */

	int pn;
@@ -468,9 +474,12 @@ static int validate_change(const struct cpuset *cur, const struct cpuset *trial) 
			goto out;

	}



	/* Cpusets with tasks can't have empty cpus_allowed or mems_allowed */

	/*

	 * Cpusets with tasks - existing or newly being attached - can't

	 * have empty cpus_allowed or mems_allowed.

	 */

	ret = -ENOSPC;

	if (cgroup_task_count(cur->css.cgroup) &&

	if ((cgroup_task_count(cur->css.cgroup) || cur->attach_in_progress) &&

	    (cpumask_empty(trial->cpus_allowed) ||

	     nodes_empty(trial->mems_allowed)))

		goto out;
@@ -1386,9 +1395,21 @@ static int cpuset_can_attach(struct cgroup *cgrp, struct cgroup_taskset *tset) 
			return ret;

	}



	/*

	 * Mark attach is in progress.  This makes validate_change() fail

	 * changes which zero cpus/mems_allowed.

	 */

	cs->attach_in_progress++;



	return 0;

}



static void cpuset_cancel_attach(struct cgroup *cgrp,

				 struct cgroup_taskset *tset)

{

	cgroup_cs(cgrp)->attach_in_progress--;

}



/*

 * Protected by cgroup_mutex.  cpus_attach is used only by cpuset_attach()

 * but we can't allocate it dynamically there.  Define it global and
@@ -1441,6 +1462,8 @@ static void cpuset_attach(struct cgroup *cgrp, struct cgroup_taskset *tset) 
					  &cpuset_attach_nodemask_to);

		mmput(mm);

	}



	cs->attach_in_progress--;

}



/* The various types of files and directories in a cpuset file system */
@@ -1908,6 +1931,7 @@ struct cgroup_subsys cpuset_subsys = { 
	.css_offline = cpuset_css_offline,

	.css_free = cpuset_css_free,

	.can_attach = cpuset_can_attach,

	.cancel_attach = cpuset_cancel_attach,

	.attach = cpuset_attach,

	.subsys_id = cpuset_subsys_id,

	.base_cftypes = files,

CRA Git | Maintained and supported by SUSTech CRA and CCSE