Merge branch 'for-4.8-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj...

Pull cgroup fixes from Tejun Heo:
 "Two fixes for cgroup.

   - There still was a hole in enforcing cpuset rules, fixed by Li.

   - The recent switch to global percpu_rwseom for threadgroup locking
     revealed a couple issues in how percpu_rwsem is implemented and
     used by cgroup.  Balbir found that the read locking section was too
     wide unnecessarily including operations which can often depend on
     IOs.  With percpu_rwsem updates (coming through a different tree)
     and reduction of read locking section, all the reported locking
     latency issues, including the android one, are resolved.

  It looks like we can keep global percpu_rwsem locking for now.  If
  there actually are cases which can't be resolved, we can go back to
  more complex per-signal_struct locking"

* 'for-4.8-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup:
  cgroup: reduce read locked section of cgroup_threadgroup_rwsem during fork
  cpuset: make sure new tasks conform to the current config of the cpuset

diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index c7fd277..c27e533 100644 (file)
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -2069,6 +2069,20 @@ static void cpuset_bind(struct cgroup_subsys_state *root_css)
        mutex_unlock(&cpuset_mutex);
 }
 
+/*
+ * Make sure the new task conform to the current state of its parent,
+ * which could have been changed by cpuset just after it inherits the
+ * state from the parent and before it sits on the cgroup's task list.
+ */
+void cpuset_fork(struct task_struct *task)
+{
+       if (task_css_is_root(task, cpuset_cgrp_id))
+               return;
+
+       set_cpus_allowed_ptr(task, &current->cpus_allowed);
+       task->mems_allowed = current->mems_allowed;
+}
+
 struct cgroup_subsys cpuset_cgrp_subsys = {
        .css_alloc      = cpuset_css_alloc,
        .css_online     = cpuset_css_online,
@@ -2079,6 +2093,7 @@ struct cgroup_subsys cpuset_cgrp_subsys = {
        .attach         = cpuset_attach,
        .post_attach    = cpuset_post_attach,
        .bind           = cpuset_bind,
+       .fork           = cpuset_fork,
        .legacy_cftypes = files,
        .early_init     = true,
 };

diff --git a/kernel/fork.c b/kernel/fork.c
index 52e725d..aaf7823 100644 (file)
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -1404,7 +1404,6 @@ static struct task_struct *copy_process(unsigned long clone_flags,
        p->real_start_time = ktime_get_boot_ns();
        p->io_context = NULL;
        p->audit_context = NULL;
-       threadgroup_change_begin(current);
        cgroup_fork(p);
 #ifdef CONFIG_NUMA
        p->mempolicy = mpol_dup(p->mempolicy);
@@ -1556,6 +1555,7 @@ static struct task_struct *copy_process(unsigned long clone_flags,
        INIT_LIST_HEAD(&p->thread_group);
        p->task_works = NULL;
 
+       threadgroup_change_begin(current);
        /*
         * Ensure that the cgroup subsystem policies allow the new process to be
         * forked. It should be noted the the new process's css_set can be changed
@@ -1656,6 +1656,7 @@ static struct task_struct *copy_process(unsigned long clone_flags,
 bad_fork_cancel_cgroup:
        cgroup_cancel_fork(p);
 bad_fork_free_pid:
+       threadgroup_change_end(current);
        if (pid != &init_struct_pid)
                free_pid(pid);
 bad_fork_cleanup_thread:
@@ -1688,7 +1689,6 @@ bad_fork_cleanup_policy:
        mpol_put(p->mempolicy);
 bad_fork_cleanup_threadgroup_lock:
 #endif
-       threadgroup_change_end(current);
        delayacct_tsk_free(p);
 bad_fork_cleanup_count:
        atomic_dec(&p->cred->user->processes);