Merge remote-tracking branch 'origin/master' into next

(Merge in order to get the PCIe mps/mrss code fixes)

7 files changed, 111 insertions, 42 deletions

diff --git a/kernel/events/core.c b/kernel/events/core.c
index b8785e26ee1c..0f857782d06f 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -399,14 +399,54 @@ void perf_cgroup_switch(struct task_struct *task, int mode)
         local_irq_restore(flags);
 }
 
-static inline void perf_cgroup_sched_out(struct task_struct *task)
+static inline void perf_cgroup_sched_out(struct task_struct *task,
+                                         struct task_struct *next)
 {
-        perf_cgroup_switch(task, PERF_CGROUP_SWOUT);
+        struct perf_cgroup *cgrp1;
+        struct perf_cgroup *cgrp2 = NULL;
+
+        /*
+         * we come here when we know perf_cgroup_events > 0
+         */
+        cgrp1 = perf_cgroup_from_task(task);
+
+        /*
+         * next is NULL when called from perf_event_enable_on_exec()
+         * that will systematically cause a cgroup_switch()
+         */
+        if (next)
+                cgrp2 = perf_cgroup_from_task(next);
+
+        /*
+         * only schedule out current cgroup events if we know
+         * that we are switching to a different cgroup. Otherwise,
+         * do no touch the cgroup events.
+         */
+        if (cgrp1 != cgrp2)
+                perf_cgroup_switch(task, PERF_CGROUP_SWOUT);
 }
 
-static inline void perf_cgroup_sched_in(struct task_struct *task)
+static inline void perf_cgroup_sched_in(struct task_struct *prev,
+                                        struct task_struct *task)
 {
-        perf_cgroup_switch(task, PERF_CGROUP_SWIN);
+        struct perf_cgroup *cgrp1;
+        struct perf_cgroup *cgrp2 = NULL;
+
+        /*
+         * we come here when we know perf_cgroup_events > 0
+         */
+        cgrp1 = perf_cgroup_from_task(task);
+
+        /* prev can never be NULL */
+        cgrp2 = perf_cgroup_from_task(prev);
+
+        /*
+         * only need to schedule in cgroup events if we are changing
+         * cgroup during ctxsw. Cgroup events were not scheduled
+         * out of ctxsw out if that was not the case.
+         */
+        if (cgrp1 != cgrp2)
+                perf_cgroup_switch(task, PERF_CGROUP_SWIN);
 }
 
 static inline int perf_cgroup_connect(int fd, struct perf_event *event,
@@ -518,11 +558,13 @@ static inline void update_cgrp_time_from_cpuctx(struct perf_cpu_context *cpuctx)
 {
 }
 
-static inline void perf_cgroup_sched_out(struct task_struct *task)
+static inline void perf_cgroup_sched_out(struct task_struct *task,
+                                         struct task_struct *next)
 {
 }
 
-static inline void perf_cgroup_sched_in(struct task_struct *task)
+static inline void perf_cgroup_sched_in(struct task_struct *prev,
+                                        struct task_struct *task)
 {
 }
 
@@ -1988,7 +2030,7 @@ void __perf_event_task_sched_out(struct task_struct *task,
          * cgroup event are system-wide mode only
          */
         if (atomic_read(&__get_cpu_var(perf_cgroup_events)))
-                perf_cgroup_sched_out(task);
+                perf_cgroup_sched_out(task, next);
 }
 
 static void task_ctx_sched_out(struct perf_event_context *ctx)
@@ -2153,7 +2195,8 @@ static void perf_event_context_sched_in(struct perf_event_context *ctx,
  * accessing the event control register. If a NMI hits, then it will
  * keep the event running.
  */
-void __perf_event_task_sched_in(struct task_struct *task)
+void __perf_event_task_sched_in(struct task_struct *prev,
+                                struct task_struct *task)
 {
         struct perf_event_context *ctx;
         int ctxn;
@@ -2171,7 +2214,7 @@ void __perf_event_task_sched_in(struct task_struct *task)
          * cgroup event are system-wide mode only
          */
         if (atomic_read(&__get_cpu_var(perf_cgroup_events)))
-                perf_cgroup_sched_in(task);
+                perf_cgroup_sched_in(prev, task);
 }
 
 static u64 perf_calculate_period(struct perf_event *event, u64 nsec, u64 count)
@@ -2427,7 +2470,7 @@ static void perf_event_enable_on_exec(struct perf_event_context *ctx)
          * ctxswin cgroup events which are already scheduled
          * in.
          */
-        perf_cgroup_sched_out(current);
+        perf_cgroup_sched_out(current, NULL);
 
         raw_spin_lock(&ctx->lock);
         task_ctx_sched_out(ctx);
@@ -3353,8 +3396,8 @@ static int perf_event_index(struct perf_event *event)
 }
 
 static void calc_timer_values(struct perf_event *event,
-                                u64 *running,
-                                u64 *enabled)
+                                u64 *enabled,
+                                u64 *running)
 {
         u64 now, ctx_time;
 
diff --git a/kernel/irq/chip.c b/kernel/irq/chip.c
index d5a3009da71a..dc5114b4c16c 100644
--- a/kernel/irq/chip.c
+++ b/kernel/irq/chip.c
@@ -178,7 +178,7 @@ void irq_shutdown(struct irq_desc *desc)
         desc->depth = 1;
         if (desc->irq_data.chip->irq_shutdown)
                 desc->irq_data.chip->irq_shutdown(&desc->irq_data);
-        if (desc->irq_data.chip->irq_disable)
+        else if (desc->irq_data.chip->irq_disable)
                 desc->irq_data.chip->irq_disable(&desc->irq_data);
         else
                 desc->irq_data.chip->irq_mask(&desc->irq_data);
diff --git a/kernel/sched.c b/kernel/sched.c
index ccacdbdecf45..ec5f472bc5b9 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -3065,7 +3065,7 @@ static void finish_task_switch(struct rq *rq, struct task_struct *prev)
 #ifdef __ARCH_WANT_INTERRUPTS_ON_CTXSW
         local_irq_disable();
 #endif /* __ARCH_WANT_INTERRUPTS_ON_CTXSW */
-        perf_event_task_sched_in(current);
+        perf_event_task_sched_in(prev, current);
 #ifdef __ARCH_WANT_INTERRUPTS_ON_CTXSW
         local_irq_enable();
 #endif /* __ARCH_WANT_INTERRUPTS_ON_CTXSW */
@@ -4279,9 +4279,9 @@ pick_next_task(struct rq *rq)
 }
 
 /*
- * schedule() is the main scheduler function.
+ * __schedule() is the main scheduler function.
  */
-asmlinkage void __sched schedule(void)
+static void __sched __schedule(void)
 {
         struct task_struct *prev, *next;
         unsigned long *switch_count;
@@ -4322,16 +4322,6 @@ need_resched:
                                 if (to_wakeup)
                                         try_to_wake_up_local(to_wakeup);
                         }
-
-                        /*
-                         * If we are going to sleep and we have plugged IO
-                         * queued, make sure to submit it to avoid deadlocks.
-                         */
-                        if (blk_needs_flush_plug(prev)) {
-                                raw_spin_unlock(&rq->lock);
-                                blk_schedule_flush_plug(prev);
-                                raw_spin_lock(&rq->lock);
-                        }
                 }
                 switch_count = &prev->nvcsw;
         }
@@ -4369,6 +4359,26 @@ need_resched:
         if (need_resched())
                 goto need_resched;
 }
+
+static inline void sched_submit_work(struct task_struct *tsk)
+{
+        if (!tsk->state)
+                return;
+        /*
+         * If we are going to sleep and we have plugged IO queued,
+         * make sure to submit it to avoid deadlocks.
+         */
+        if (blk_needs_flush_plug(tsk))
+                blk_schedule_flush_plug(tsk);
+}
+
+asmlinkage void schedule(void)
+{
+        struct task_struct *tsk = current;
+
+        sched_submit_work(tsk);
+        __schedule();
+}
 EXPORT_SYMBOL(schedule);
 
 #ifdef CONFIG_MUTEX_SPIN_ON_OWNER
@@ -4435,7 +4445,7 @@ asmlinkage void __sched notrace preempt_schedule(void)
 
         do {
                 add_preempt_count_notrace(PREEMPT_ACTIVE);
-                schedule();
+                __schedule();
                 sub_preempt_count_notrace(PREEMPT_ACTIVE);
 
                 /*
@@ -4463,7 +4473,7 @@ asmlinkage void __sched preempt_schedule_irq(void)
         do {
                 add_preempt_count(PREEMPT_ACTIVE);
                 local_irq_enable();
-                schedule();
+                __schedule();
                 local_irq_disable();
                 sub_preempt_count(PREEMPT_ACTIVE);
 
@@ -5588,7 +5598,7 @@ static inline int should_resched(void)
 static void __cond_resched(void)
 {
         add_preempt_count(PREEMPT_ACTIVE);
-        schedule();
+        __schedule();
         sub_preempt_count(PREEMPT_ACTIVE);
 }
 
@@ -7443,6 +7453,7 @@ static void __sdt_free(const struct cpumask *cpu_map)
                         struct sched_domain *sd = *per_cpu_ptr(sdd->sd, j);
                         if (sd && (sd->flags & SD_OVERLAP))
                                 free_sched_groups(sd->groups, 0);
+                        kfree(*per_cpu_ptr(sdd->sd, j));
                         kfree(*per_cpu_ptr(sdd->sg, j));
                         kfree(*per_cpu_ptr(sdd->sgp, j));
                 }
diff --git a/kernel/taskstats.c b/kernel/taskstats.c
index e19ce1454ee1..e66046456f4f 100644
--- a/kernel/taskstats.c
+++ b/kernel/taskstats.c
@@ -655,6 +655,7 @@ static struct genl_ops taskstats_ops = {
         .cmd            = TASKSTATS_CMD_GET,
         .doit           = taskstats_user_cmd,
         .policy         = taskstats_cmd_get_policy,
+        .flags          = GENL_ADMIN_PERM,
 };
 
 static struct genl_ops cgroupstats_ops = {
diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c
index 59f369f98a04..ea5e1a928d5b 100644
--- a/kernel/time/alarmtimer.c
+++ b/kernel/time/alarmtimer.c
@@ -441,6 +441,8 @@ static int alarm_timer_create(struct k_itimer *new_timer)
 static void alarm_timer_get(struct k_itimer *timr,
                                 struct itimerspec *cur_setting)
 {
+        memset(cur_setting, 0, sizeof(struct itimerspec));
+
         cur_setting->it_interval =
                         ktime_to_timespec(timr->it.alarmtimer.period);
         cur_setting->it_value =
@@ -479,11 +481,17 @@ static int alarm_timer_set(struct k_itimer *timr, int flags,
         if (!rtcdev)
                 return -ENOTSUPP;
 
-        /* Save old values */
-        old_setting->it_interval =
-                        ktime_to_timespec(timr->it.alarmtimer.period);
-        old_setting->it_value =
-                        ktime_to_timespec(timr->it.alarmtimer.node.expires);
+        /*
+         * XXX HACK! Currently we can DOS a system if the interval
+         * period on alarmtimers is too small. Cap the interval here
+         * to 100us and solve this properly in a future patch! -jstultz
+         */
+        if ((new_setting->it_interval.tv_sec == 0) &&
+                        (new_setting->it_interval.tv_nsec < 100000))
+                new_setting->it_interval.tv_nsec = 100000;
+
+        if (old_setting)
+                alarm_timer_get(timr, old_setting);
 
         /* If the timer was already set, cancel it */
         alarm_cancel(&timr->it.alarmtimer);
diff --git a/kernel/tsacct.c b/kernel/tsacct.c
index 24dc60d9fa1f..5bbfac85866e 100644
--- a/kernel/tsacct.c
+++ b/kernel/tsacct.c
@@ -78,6 +78,7 @@ void bacct_add_tsk(struct taskstats *stats, struct task_struct *tsk)
 
 #define KB 1024
 #define MB (1024*KB)
+#define KB_MASK (~(KB-1))
 /*
  * fill in extended accounting fields
  */
@@ -95,14 +96,14 @@ void xacct_add_tsk(struct taskstats *stats, struct task_struct *p)
                 stats->hiwater_vm    = get_mm_hiwater_vm(mm)  * PAGE_SIZE / KB;
                 mmput(mm);
         }
-        stats->read_char        = p->ioac.rchar;
-        stats->write_char       = p->ioac.wchar;
-        stats->read_syscalls    = p->ioac.syscr;
-        stats->write_syscalls   = p->ioac.syscw;
+        stats->read_char        = p->ioac.rchar & KB_MASK;
+        stats->write_char       = p->ioac.wchar & KB_MASK;
+        stats->read_syscalls    = p->ioac.syscr & KB_MASK;
+        stats->write_syscalls   = p->ioac.syscw & KB_MASK;
 #ifdef CONFIG_TASK_IO_ACCOUNTING
-        stats->read_bytes       = p->ioac.read_bytes;
-        stats->write_bytes      = p->ioac.write_bytes;
-        stats->cancelled_write_bytes = p->ioac.cancelled_write_bytes;
+        stats->read_bytes       = p->ioac.read_bytes & KB_MASK;
+        stats->write_bytes      = p->ioac.write_bytes & KB_MASK;
+        stats->cancelled_write_bytes = p->ioac.cancelled_write_bytes & KB_MASK;
 #else
         stats->read_bytes       = 0;
         stats->write_bytes      = 0;
diff --git a/kernel/workqueue.c b/kernel/workqueue.c
index 25fb1b0e53fa..1783aabc6128 100644
--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
@@ -2412,8 +2412,13 @@ reflush:
 
         for_each_cwq_cpu(cpu, wq) {
                 struct cpu_workqueue_struct *cwq = get_cwq(cpu, wq);
+                bool drained;
 
-                if (!cwq->nr_active && list_empty(&cwq->delayed_works))
+                spin_lock_irq(&cwq->gcwq->lock);
+                drained = !cwq->nr_active && list_empty(&cwq->delayed_works);
+                spin_unlock_irq(&cwq->gcwq->lock);
+
+                if (drained)
                         continue;
 
                 if (++flush_cnt == 10 ||