10 files changed, 109 insertions, 57 deletions

diff --git a/kernel/audit.c b/kernel/audit.c
index e8692a5748c2..e092f1c0ce30 100644
--- a/kernel/audit.c
+++ b/kernel/audit.c
@@ -738,7 +738,7 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
                 if (!audit_enabled && msg_type != AUDIT_USER_AVC)
                         return 0;
 
-                err = audit_filter_user(&NETLINK_CB(skb), msg_type);
+                err = audit_filter_user(&NETLINK_CB(skb));
                 if (err == 1) {
                         err = 0;
                         if (msg_type == AUDIT_USER_TTY) {
@@ -779,7 +779,7 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
                 }
                 /* fallthrough */
         case AUDIT_LIST:
-                err = audit_receive_filter(nlh->nlmsg_type, NETLINK_CB(skb).pid,
+                err = audit_receive_filter(msg_type, NETLINK_CB(skb).pid,
                                            uid, seq, data, nlmsg_len(nlh),
                                            loginuid, sessionid, sid);
                 break;
@@ -798,7 +798,7 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
                 }
                 /* fallthrough */
         case AUDIT_LIST_RULES:
-                err = audit_receive_filter(nlh->nlmsg_type, NETLINK_CB(skb).pid,
+                err = audit_receive_filter(msg_type, NETLINK_CB(skb).pid,
                                            uid, seq, data, nlmsg_len(nlh),
                                            loginuid, sessionid, sid);
                 break;
diff --git a/kernel/auditfilter.c b/kernel/auditfilter.c
index 0e0bd27e6512..98c50cc671bb 100644
--- a/kernel/auditfilter.c
+++ b/kernel/auditfilter.c
@@ -1544,6 +1544,7 @@ static void audit_log_rule_change(uid_t loginuid, u32 sessionid, u32 sid,
  * @data: payload data
  * @datasz: size of payload data
  * @loginuid: loginuid of sender
+ * @sessionid: sessionid for netlink audit message
  * @sid: SE Linux Security ID of sender
  */
 int audit_receive_filter(int type, int pid, int uid, int seq, void *data,
@@ -1720,7 +1721,7 @@ static int audit_filter_user_rules(struct netlink_skb_parms *cb,
         return 1;
 }
 
-int audit_filter_user(struct netlink_skb_parms *cb, int type)
+int audit_filter_user(struct netlink_skb_parms *cb)
 {
         enum audit_state state = AUDIT_DISABLED;
         struct audit_entry *e;
diff --git a/kernel/futex.c b/kernel/futex.c
index 449def8074fe..7d1136e97c14 100644
--- a/kernel/futex.c
+++ b/kernel/futex.c
@@ -1096,21 +1096,64 @@ static void unqueue_me_pi(struct futex_q *q)
  * private futexes.
  */
 static int fixup_pi_state_owner(u32 __user *uaddr, struct futex_q *q,
-                                struct task_struct *newowner)
+                                struct task_struct *newowner,
+                                struct rw_semaphore *fshared)
 {
         u32 newtid = task_pid_vnr(newowner) | FUTEX_WAITERS;
         struct futex_pi_state *pi_state = q->pi_state;
+        struct task_struct *oldowner = pi_state->owner;
         u32 uval, curval, newval;
-        int ret;
+        int ret, attempt = 0;
 
         /* Owner died? */
+        if (!pi_state->owner)
+                newtid |= FUTEX_OWNER_DIED;
+
+        /*
+         * We are here either because we stole the rtmutex from the
+         * pending owner or we are the pending owner which failed to
+         * get the rtmutex. We have to replace the pending owner TID
+         * in the user space variable. This must be atomic as we have
+         * to preserve the owner died bit here.
+         *
+         * Note: We write the user space value _before_ changing the
+         * pi_state because we can fault here. Imagine swapped out
+         * pages or a fork, which was running right before we acquired
+         * mmap_sem, that marked all the anonymous memory readonly for
+         * cow.
+         *
+         * Modifying pi_state _before_ the user space value would
+         * leave the pi_state in an inconsistent state when we fault
+         * here, because we need to drop the hash bucket lock to
+         * handle the fault. This might be observed in the PID check
+         * in lookup_pi_state.
+         */
+retry:
+        if (get_futex_value_locked(&uval, uaddr))
+                goto handle_fault;
+
+        while (1) {
+                newval = (uval & FUTEX_OWNER_DIED) | newtid;
+
+                curval = cmpxchg_futex_value_locked(uaddr, uval, newval);
+
+                if (curval == -EFAULT)
+                        goto handle_fault;
+                if (curval == uval)
+                        break;
+                uval = curval;
+        }
+
+        /*
+         * We fixed up user space. Now we need to fix the pi_state
+         * itself.
+         */
         if (pi_state->owner != NULL) {
                 spin_lock_irq(&pi_state->owner->pi_lock);
                 WARN_ON(list_empty(&pi_state->list));
                 list_del_init(&pi_state->list);
                 spin_unlock_irq(&pi_state->owner->pi_lock);
-        } else
-                newtid |= FUTEX_OWNER_DIED;
+        }
 
         pi_state->owner = newowner;
 
@@ -1118,26 +1161,35 @@ static int fixup_pi_state_owner(u32 __user *uaddr, struct futex_q *q,
         WARN_ON(!list_empty(&pi_state->list));
         list_add(&pi_state->list, &newowner->pi_state_list);
         spin_unlock_irq(&newowner->pi_lock);
+        return 0;
 
         /*
-         * We own it, so we have to replace the pending owner
-         * TID. This must be atomic as we have preserve the
-         * owner died bit here.
+         * To handle the page fault we need to drop the hash bucket
+         * lock here. That gives the other task (either the pending
+         * owner itself or the task which stole the rtmutex) the
+         * chance to try the fixup of the pi_state. So once we are
+         * back from handling the fault we need to check the pi_state
+         * after reacquiring the hash bucket lock and before trying to
+         * do another fixup. When the fixup has been done already we
+         * simply return.
          */
-        ret = get_futex_value_locked(&uval, uaddr);
+handle_fault:
+        spin_unlock(q->lock_ptr);
 
-        while (!ret) {
-                newval = (uval & FUTEX_OWNER_DIED) | newtid;
+        ret = futex_handle_fault((unsigned long)uaddr, fshared, attempt++);
 
-                curval = cmpxchg_futex_value_locked(uaddr, uval, newval);
+        spin_lock(q->lock_ptr);
 
-                if (curval == -EFAULT)
-                        ret = -EFAULT;
-                if (curval == uval)
-                        break;
-                uval = curval;
-        }
-        return ret;
+        /*
+         * Check if someone else fixed it for us:
+         */
+        if (pi_state->owner != oldowner)
+                return 0;
+
+        if (ret)
+                return ret;
+
+        goto retry;
 }
 
 /*
@@ -1507,7 +1559,7 @@ static int futex_lock_pi(u32 __user *uaddr, struct rw_semaphore *fshared,
                  * that case:
                  */
                 if (q.pi_state->owner != curr)
-                        ret = fixup_pi_state_owner(uaddr, &q, curr);
+                        ret = fixup_pi_state_owner(uaddr, &q, curr, fshared);
         } else {
                 /*
                  * Catch the rare case, where the lock was released
@@ -1539,7 +1591,8 @@ static int futex_lock_pi(u32 __user *uaddr, struct rw_semaphore *fshared,
                                 int res;
 
                                 owner = rt_mutex_owner(&q.pi_state->pi_mutex);
-                                res = fixup_pi_state_owner(uaddr, &q, owner);
+                                res = fixup_pi_state_owner(uaddr, &q, owner,
+                                                           fshared);
 
                                 /* propagate -EFAULT, if the fixup failed */
                                 if (res)
diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c
index 421be5fe5cc7..543d9ca9b4f4 100644
--- a/kernel/hrtimer.c
+++ b/kernel/hrtimer.c
@@ -1078,7 +1078,7 @@ ktime_t hrtimer_get_remaining(const struct hrtimer *timer)
 }
 EXPORT_SYMBOL_GPL(hrtimer_get_remaining);
 
-#if defined(CONFIG_NO_IDLE_HZ) || defined(CONFIG_NO_HZ)
+#ifdef CONFIG_NO_HZ
 /**
  * hrtimer_get_next_event - get the time until next expiry event
  *
diff --git a/kernel/kgdb.c b/kernel/kgdb.c
index 79e3c90113c2..3ec23c3ec97f 100644
--- a/kernel/kgdb.c
+++ b/kernel/kgdb.c
@@ -1499,7 +1499,8 @@ int kgdb_nmicallback(int cpu, void *regs)
         return 1;
 }
 
-void kgdb_console_write(struct console *co, const char *s, unsigned count)
+static void kgdb_console_write(struct console *co, const char *s,
+   unsigned count)
 {
         unsigned long flags;
 
diff --git a/kernel/rcuclassic.c b/kernel/rcuclassic.c
index f4ffbd0f306f..a38895a5b8e2 100644
--- a/kernel/rcuclassic.c
+++ b/kernel/rcuclassic.c
@@ -89,8 +89,22 @@ static void force_quiescent_state(struct rcu_data *rdp,
                 /*
                  * Don't send IPI to itself. With irqs disabled,
                  * rdp->cpu is the current cpu.
+                 *
+                 * cpu_online_map is updated by the _cpu_down()
+                 * using stop_machine_run(). Since we're in irqs disabled
+                 * section, stop_machine_run() is not exectuting, hence
+                 * the cpu_online_map is stable.
+                 *
+                 * However,  a cpu might have been offlined _just_ before
+                 * we disabled irqs while entering here.
+                 * And rcu subsystem might not yet have handled the CPU_DEAD
+                 * notification, leading to the offlined cpu's bit
+                 * being set in the rcp->cpumask.
+                 *
+                 * Hence cpumask = (rcp->cpumask & cpu_online_map) to prevent
+                 * sending smp_reschedule() to an offlined CPU.
                  */
-                cpumask = rcp->cpumask;
+                cpus_and(cpumask, rcp->cpumask, cpu_online_map);
                 cpu_clear(rdp->cpu, cpumask);
                 for_each_cpu_mask(cpu, cpumask)
                         smp_send_reschedule(cpu);
diff --git a/kernel/sched.c b/kernel/sched.c
index b048ad8a11af..94ead43eda62 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -4398,22 +4398,20 @@ do_wait_for_common(struct completion *x, long timeout, int state)
                              signal_pending(current)) ||
                             (state == TASK_KILLABLE &&
                              fatal_signal_pending(current))) {
-                                __remove_wait_queue(&x->wait, &wait);
-                                return -ERESTARTSYS;
+                                timeout = -ERESTARTSYS;
+                                break;
                         }
                         __set_current_state(state);
                         spin_unlock_irq(&x->wait.lock);
                         timeout = schedule_timeout(timeout);
                         spin_lock_irq(&x->wait.lock);
-                        if (!timeout) {
-                                __remove_wait_queue(&x->wait, &wait);
-                                return timeout;
-                        }
-                } while (!x->done);
+                } while (!x->done && timeout);
                 __remove_wait_queue(&x->wait, &wait);
+                if (!x->done)
+                        return timeout;
         }
         x->done--;
-        return timeout;
+        return timeout ?: 1;
 }
 
 static long __sched
@@ -5889,6 +5887,7 @@ static void migrate_dead_tasks(unsigned int dead_cpu)
                 next = pick_next_task(rq, rq->curr);
                 if (!next)
                         break;
+                next->sched_class->put_prev_task(rq, next);
                 migrate_dead(dead_cpu, next);
 
         }
@@ -8503,6 +8502,9 @@ int sched_group_set_rt_period(struct task_group *tg, long rt_period_us)
         rt_period = (u64)rt_period_us * NSEC_PER_USEC;
         rt_runtime = tg->rt_bandwidth.rt_runtime;
 
+        if (rt_period == 0)
+                return -EINVAL;
+
         return tg_set_bandwidth(tg, rt_period, rt_runtime);
 }
 
diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c
index 1dad5bbb59b6..0f3c19197fa4 100644
--- a/kernel/sched_rt.c
+++ b/kernel/sched_rt.c
@@ -250,7 +250,8 @@ static int do_sched_rt_period_timer(struct rt_bandwidth *rt_b, int overrun)
                         if (rt_rq->rt_time || rt_rq->rt_nr_running)
                                 idle = 0;
                         spin_unlock(&rt_rq->rt_runtime_lock);
-                }
+                } else if (rt_rq->rt_nr_running)
+                        idle = 0;
 
                 if (enqueue)
                         sched_rt_rq_enqueue(rt_rq);
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 29116652dca8..f6d2e57b99a0 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -132,8 +132,6 @@ extern int sysctl_userprocess_debug;
 extern int spin_retry;
 #endif
 
-extern int sysctl_hz_timer;
-
 #ifdef CONFIG_BSD_PROCESS_ACCT
 extern int acct_parm[];
 #endif
@@ -563,16 +561,6 @@ static struct ctl_table kern_table[] = {
                 .proc_handler   = &proc_dointvec,
         },
 #endif
-#ifdef CONFIG_NO_IDLE_HZ
-        {
-                .ctl_name       = KERN_HZ_TIMER,
-                .procname       = "hz_timer",
-                .data           = &sysctl_hz_timer,
-                .maxlen         = sizeof(int),
-                .mode           = 0644,
-                .proc_handler   = &proc_dointvec,
-        },
-#endif
         {
                 .ctl_name       = KERN_S390_USER_DEBUG_LOGGING,
                 .procname       = "userprocess_debug",
diff --git a/kernel/timer.c b/kernel/timer.c
index ceacc6626572..ef3fa6906e8f 100644
--- a/kernel/timer.c
+++ b/kernel/timer.c
@@ -812,7 +812,7 @@ static inline void __run_timers(struct tvec_base *base)
         spin_unlock_irq(&base->lock);
 }
 
-#if defined(CONFIG_NO_IDLE_HZ) || defined(CONFIG_NO_HZ)
+#ifdef CONFIG_NO_HZ
 /*
  * Find out when the next timer event is due to happen. This
  * is used on S/390 to stop all activity when a cpus is idle.
@@ -947,14 +947,6 @@ unsigned long get_next_timer_interrupt(unsigned long now)
 
         return cmp_next_hrtimer_event(now, expires);
 }
-
-#ifdef CONFIG_NO_IDLE_HZ
-unsigned long next_timer_interrupt(void)
-{
-        return get_next_timer_interrupt(jiffies);
-}
-#endif
-
 #endif
 
 #ifndef CONFIG_VIRT_CPU_ACCOUNTING