Merge branch 'perf/core' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip into trace/tip/tracing/core-6

Conflicts:
	include/trace/ftrace.h
	kernel/trace/trace_kprobe.c

Acked-by: Masami Hiramatsu <mhiramat@redhat.com>
Acked-by: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>

18 files changed, 849 insertions, 910 deletions

diff --git a/kernel/cgroup_freezer.c b/kernel/cgroup_freezer.c
index da5e139..e5c0244 100644
--- a/kernel/cgroup_freezer.c
+++ b/kernel/cgroup_freezer.c
@@ -205,9 +205,12 @@ static void freezer_fork(struct cgroup_subsys *ss, struct task_struct *task)
 	 * No lock is needed, since the task isn't on tasklist yet,
 	 * so it can't be moved to another cgroup, which means the
 	 * freezer won't be removed and will be valid during this
-	 * function call.
+	 * function call.  Nevertheless, apply RCU read-side critical
+	 * section to suppress RCU lockdep false positives.
 	 */
+	rcu_read_lock();
 	freezer = task_freezer(task);
+	rcu_read_unlock();
 
 	/*
 	 * The root cgroup is non-freezable, so we can skip the
diff --git a/kernel/fork.c b/kernel/fork.c
index 44b0791..5d3592de 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -1111,9 +1111,6 @@ static struct task_struct *copy_process(unsigned long clone_flags,
 	p->memcg_batch.do_batch = 0;
 	p->memcg_batch.memcg = NULL;
 #endif
-
-	p->bts = NULL;
-
 	p->stack_start = stack_start;
 
 	/* Perform scheduler related setup. Assign this task to a CPU. */
diff --git a/kernel/hw_breakpoint.c b/kernel/hw_breakpoint.c
index 03808ed..7a56b22 100644
--- a/kernel/hw_breakpoint.c
+++ b/kernel/hw_breakpoint.c
@@ -40,23 +40,29 @@
 #include <linux/percpu.h>
 #include <linux/sched.h>
 #include <linux/init.h>
+#include <linux/slab.h>
 #include <linux/cpu.h>
 #include <linux/smp.h>
 
 #include <linux/hw_breakpoint.h>
 
+
 /*
  * Constraints data
  */
 
 /* Number of pinned cpu breakpoints in a cpu */
-static DEFINE_PER_CPU(unsigned int, nr_cpu_bp_pinned);
+static DEFINE_PER_CPU(unsigned int, nr_cpu_bp_pinned[TYPE_MAX]);
 
 /* Number of pinned task breakpoints in a cpu */
-static DEFINE_PER_CPU(unsigned int, nr_task_bp_pinned[HBP_NUM]);
+static DEFINE_PER_CPU(unsigned int *, nr_task_bp_pinned[TYPE_MAX]);
 
 /* Number of non-pinned cpu/task breakpoints in a cpu */
-static DEFINE_PER_CPU(unsigned int, nr_bp_flexible);
+static DEFINE_PER_CPU(unsigned int, nr_bp_flexible[TYPE_MAX]);
+
+static int nr_slots[TYPE_MAX];
+
+static int constraints_initialized;
 
 /* Gather the number of total pinned and un-pinned bp in a cpuset */
 struct bp_busy_slots {
@@ -67,16 +73,29 @@ struct bp_busy_slots {
 /* Serialize accesses to the above constraints */
 static DEFINE_MUTEX(nr_bp_mutex);
 
+__weak int hw_breakpoint_weight(struct perf_event *bp)
+{
+	return 1;
+}
+
+static inline enum bp_type_idx find_slot_idx(struct perf_event *bp)
+{
+	if (bp->attr.bp_type & HW_BREAKPOINT_RW)
+		return TYPE_DATA;
+
+	return TYPE_INST;
+}
+
 /*
  * Report the maximum number of pinned breakpoints a task
  * have in this cpu
  */
-static unsigned int max_task_bp_pinned(int cpu)
+static unsigned int max_task_bp_pinned(int cpu, enum bp_type_idx type)
 {
 	int i;
-	unsigned int *tsk_pinned = per_cpu(nr_task_bp_pinned, cpu);
+	unsigned int *tsk_pinned = per_cpu(nr_task_bp_pinned[type], cpu);
 
-	for (i = HBP_NUM -1; i >= 0; i--) {
+	for (i = nr_slots[type] - 1; i >= 0; i--) {
 		if (tsk_pinned[i] > 0)
 			return i + 1;
 	}
@@ -84,7 +103,7 @@ static unsigned int max_task_bp_pinned(int cpu)
 	return 0;
 }
 
-static int task_bp_pinned(struct task_struct *tsk)
+static int task_bp_pinned(struct task_struct *tsk, enum bp_type_idx type)
 {
 	struct perf_event_context *ctx = tsk->perf_event_ctxp;
 	struct list_head *list;
@@ -105,7 +124,8 @@ static int task_bp_pinned(struct task_struct *tsk)
 	 */
 	list_for_each_entry(bp, list, event_entry) {
 		if (bp->attr.type == PERF_TYPE_BREAKPOINT)
-			count++;
+			if (find_slot_idx(bp) == type)
+				count += hw_breakpoint_weight(bp);
 	}
 
 	raw_spin_unlock_irqrestore(&ctx->lock, flags);
@@ -118,18 +138,19 @@ static int task_bp_pinned(struct task_struct *tsk)
  * a given cpu (cpu > -1) or in all of them (cpu = -1).
  */
 static void
-fetch_bp_busy_slots(struct bp_busy_slots *slots, struct perf_event *bp)
+fetch_bp_busy_slots(struct bp_busy_slots *slots, struct perf_event *bp,
+		    enum bp_type_idx type)
 {
 	int cpu = bp->cpu;
 	struct task_struct *tsk = bp->ctx->task;
 
 	if (cpu >= 0) {
-		slots->pinned = per_cpu(nr_cpu_bp_pinned, cpu);
+		slots->pinned = per_cpu(nr_cpu_bp_pinned[type], cpu);
 		if (!tsk)
-			slots->pinned += max_task_bp_pinned(cpu);
+			slots->pinned += max_task_bp_pinned(cpu, type);
 		else
-			slots->pinned += task_bp_pinned(tsk);
-		slots->flexible = per_cpu(nr_bp_flexible, cpu);
+			slots->pinned += task_bp_pinned(tsk, type);
+		slots->flexible = per_cpu(nr_bp_flexible[type], cpu);
 
 		return;
 	}
@@ -137,16 +158,16 @@ fetch_bp_busy_slots(struct bp_busy_slots *slots, struct perf_event *bp)
 	for_each_online_cpu(cpu) {
 		unsigned int nr;
 
-		nr = per_cpu(nr_cpu_bp_pinned, cpu);
+		nr = per_cpu(nr_cpu_bp_pinned[type], cpu);
 		if (!tsk)
-			nr += max_task_bp_pinned(cpu);
+			nr += max_task_bp_pinned(cpu, type);
 		else
-			nr += task_bp_pinned(tsk);
+			nr += task_bp_pinned(tsk, type);
 
 		if (nr > slots->pinned)
 			slots->pinned = nr;
 
-		nr = per_cpu(nr_bp_flexible, cpu);
+		nr = per_cpu(nr_bp_flexible[type], cpu);
 
 		if (nr > slots->flexible)
 			slots->flexible = nr;
@@ -154,31 +175,49 @@ fetch_bp_busy_slots(struct bp_busy_slots *slots, struct perf_event *bp)
 }
 
 /*
+ * For now, continue to consider flexible as pinned, until we can
+ * ensure no flexible event can ever be scheduled before a pinned event
+ * in a same cpu.
+ */
+static void
+fetch_this_slot(struct bp_busy_slots *slots, int weight)
+{
+	slots->pinned += weight;
+}
+
+/*
  * Add a pinned breakpoint for the given task in our constraint table
  */
-static void toggle_bp_task_slot(struct task_struct *tsk, int cpu, bool enable)
+static void toggle_bp_task_slot(struct task_struct *tsk, int cpu, bool enable,
+				enum bp_type_idx type, int weight)
 {
 	unsigned int *tsk_pinned;
-	int count = 0;
+	int old_count = 0;
+	int old_idx = 0;
+	int idx = 0;
 
-	count = task_bp_pinned(tsk);
+	old_count = task_bp_pinned(tsk, type);
+	old_idx = old_count - 1;
+	idx = old_idx + weight;
 
-	tsk_pinned = per_cpu(nr_task_bp_pinned, cpu);
+	tsk_pinned = per_cpu(nr_task_bp_pinned[type], cpu);
 	if (enable) {
-		tsk_pinned[count]++;
-		if (count > 0)
-			tsk_pinned[count-1]--;
+		tsk_pinned[idx]++;
+		if (old_count > 0)
+			tsk_pinned[old_idx]--;
 	} else {
-		tsk_pinned[count]--;
-		if (count > 0)
-			tsk_pinned[count-1]++;
+		tsk_pinned[idx]--;
+		if (old_count > 0)
+			tsk_pinned[old_idx]++;
 	}
 }
 
 /*
  * Add/remove the given breakpoint in our constraint table
  */
-static void toggle_bp_slot(struct perf_event *bp, bool enable)
+static void
+toggle_bp_slot(struct perf_event *bp, bool enable, enum bp_type_idx type,
+	       int weight)
 {
 	int cpu = bp->cpu;
 	struct task_struct *tsk = bp->ctx->task;
@@ -186,20 +225,20 @@ static void toggle_bp_slot(struct perf_event *bp, bool enable)
 	/* Pinned counter task profiling */
 	if (tsk) {
 		if (cpu >= 0) {
-			toggle_bp_task_slot(tsk, cpu, enable);
+			toggle_bp_task_slot(tsk, cpu, enable, type, weight);
 			return;
 		}
 
 		for_each_online_cpu(cpu)
-			toggle_bp_task_slot(tsk, cpu, enable);
+			toggle_bp_task_slot(tsk, cpu, enable, type, weight);
 		return;
 	}
 
 	/* Pinned counter cpu profiling */
 	if (enable)
-		per_cpu(nr_cpu_bp_pinned, bp->cpu)++;
+		per_cpu(nr_cpu_bp_pinned[type], bp->cpu) += weight;
 	else
-		per_cpu(nr_cpu_bp_pinned, bp->cpu)--;
+		per_cpu(nr_cpu_bp_pinned[type], bp->cpu) -= weight;
 }
 
 /*
@@ -246,14 +285,29 @@ static void toggle_bp_slot(struct perf_event *bp, bool enable)
 static int __reserve_bp_slot(struct perf_event *bp)
 {
 	struct bp_busy_slots slots = {0};
+	enum bp_type_idx type;
+	int weight;
 
-	fetch_bp_busy_slots(&slots, bp);
+	/* We couldn't initialize breakpoint constraints on boot */
+	if (!constraints_initialized)
+		return -ENOMEM;
+
+	/* Basic checks */
+	if (bp->attr.bp_type == HW_BREAKPOINT_EMPTY ||
+	    bp->attr.bp_type == HW_BREAKPOINT_INVALID)
+		return -EINVAL;
+
+	type = find_slot_idx(bp);
+	weight = hw_breakpoint_weight(bp);
+
+	fetch_bp_busy_slots(&slots, bp, type);
+	fetch_this_slot(&slots, weight);
 
 	/* Flexible counters need to keep at least one slot */
-	if (slots.pinned + (!!slots.flexible) == HBP_NUM)
+	if (slots.pinned + (!!slots.flexible) > nr_slots[type])
 		return -ENOSPC;
 
-	toggle_bp_slot(bp, true);
+	toggle_bp_slot(bp, true, type, weight);
 
 	return 0;
 }
@@ -273,7 +327,12 @@ int reserve_bp_slot(struct perf_event *bp)
 
 static void __release_bp_slot(struct perf_event *bp)
 {
-	toggle_bp_slot(bp, false);
+	enum bp_type_idx type;
+	int weight;
+
+	type = find_slot_idx(bp);
+	weight = hw_breakpoint_weight(bp);
+	toggle_bp_slot(bp, false, type, weight);
 }
 
 void release_bp_slot(struct perf_event *bp)
@@ -308,6 +367,28 @@ int dbg_release_bp_slot(struct perf_event *bp)
 	return 0;
 }
 
+static int validate_hw_breakpoint(struct perf_event *bp)
+{
+	int ret;
+
+	ret = arch_validate_hwbkpt_settings(bp);
+	if (ret)
+		return ret;
+
+	if (arch_check_bp_in_kernelspace(bp)) {
+		if (bp->attr.exclude_kernel)
+			return -EINVAL;
+		/*
+		 * Don't let unprivileged users set a breakpoint in the trap
+		 * path to avoid trap recursion attacks.
+		 */
+		if (!capable(CAP_SYS_ADMIN))
+			return -EPERM;
+	}
+
+	return 0;
+}
+
 int register_perf_hw_breakpoint(struct perf_event *bp)
 {
 	int ret;
@@ -316,17 +397,7 @@ int register_perf_hw_breakpoint(struct perf_event *bp)
 	if (ret)
 		return ret;
 
-	/*
-	 * Ptrace breakpoints can be temporary perf events only
-	 * meant to reserve a slot. In this case, it is created disabled and
-	 * we don't want to check the params right now (as we put a null addr)
-	 * But perf tools create events as disabled and we want to check
-	 * the params for them.
-	 * This is a quick hack that will be removed soon, once we remove
-	 * the tmp breakpoints from ptrace
-	 */
-	if (!bp->attr.disabled || !bp->overflow_handler)
-		ret = arch_validate_hwbkpt_settings(bp, bp->ctx->task);
+	ret = validate_hw_breakpoint(bp);
 
 	/* if arch_validate_hwbkpt_settings() fails then release bp slot */
 	if (ret)
@@ -373,7 +444,7 @@ int modify_user_hw_breakpoint(struct perf_event *bp, struct perf_event_attr *att
 	if (attr->disabled)
 		goto end;
 
-	err = arch_validate_hwbkpt_settings(bp, bp->ctx->task);
+	err = validate_hw_breakpoint(bp);
 	if (!err)
 		perf_event_enable(bp);
 
@@ -480,7 +551,36 @@ static struct notifier_block hw_breakpoint_exceptions_nb = {
 
 static int __init init_hw_breakpoint(void)
 {
+	unsigned int **task_bp_pinned;
+	int cpu, err_cpu;
+	int i;
+
+	for (i = 0; i < TYPE_MAX; i++)
+		nr_slots[i] = hw_breakpoint_slots(i);
+
+	for_each_possible_cpu(cpu) {
+		for (i = 0; i < TYPE_MAX; i++) {
+			task_bp_pinned = &per_cpu(nr_task_bp_pinned[i], cpu);
+			*task_bp_pinned = kzalloc(sizeof(int) * nr_slots[i],
+						  GFP_KERNEL);
+			if (!*task_bp_pinned)
+				goto err_alloc;
+		}
+	}
+
+	constraints_initialized = 1;
+
 	return register_die_notifier(&hw_breakpoint_exceptions_nb);
+
+ err_alloc:
+	for_each_possible_cpu(err_cpu) {
+		if (err_cpu == cpu)
+			break;
+		for (i = 0; i < TYPE_MAX; i++)
+			kfree(per_cpu(nr_task_bp_pinned[i], cpu));
+	}
+
+	return -ENOMEM;
 }
 core_initcall(init_hw_breakpoint);
 
diff --git a/kernel/kprobes.c b/kernel/kprobes.c
index 0ed46f3..282035f 100644
--- a/kernel/kprobes.c
+++ b/kernel/kprobes.c
@@ -1588,6 +1588,72 @@ static void __kprobes kill_kprobe(struct kprobe *p)
 	arch_remove_kprobe(p);
 }
 
+/* Disable one kprobe */
+int __kprobes disable_kprobe(struct kprobe *kp)
+{
+	int ret = 0;
+	struct kprobe *p;
+
+	mutex_lock(&kprobe_mutex);
+
+	/* Check whether specified probe is valid. */
+	p = __get_valid_kprobe(kp);
+	if (unlikely(p == NULL)) {
+		ret = -EINVAL;
+		goto out;
+	}
+
+	/* If the probe is already disabled (or gone), just return */
+	if (kprobe_disabled(kp))
+		goto out;
+
+	kp->flags |= KPROBE_FLAG_DISABLED;
+	if (p != kp)
+		/* When kp != p, p is always enabled. */
+		try_to_disable_aggr_kprobe(p);
+
+	if (!kprobes_all_disarmed && kprobe_disabled(p))
+		disarm_kprobe(p);
+out:
+	mutex_unlock(&kprobe_mutex);
+	return ret;
+}
+EXPORT_SYMBOL_GPL(disable_kprobe);
+
+/* Enable one kprobe */
+int __kprobes enable_kprobe(struct kprobe *kp)
+{
+	int ret = 0;
+	struct kprobe *p;
+
+	mutex_lock(&kprobe_mutex);
+
+	/* Check whether specified probe is valid. */
+	p = __get_valid_kprobe(kp);
+	if (unlikely(p == NULL)) {
+		ret = -EINVAL;
+		goto out;
+	}
+
+	if (kprobe_gone(kp)) {
+		/* This kprobe has gone, we couldn't enable it. */
+		ret = -EINVAL;
+		goto out;
+	}
+
+	if (p != kp)
+		kp->flags &= ~KPROBE_FLAG_DISABLED;
+
+	if (!kprobes_all_disarmed && kprobe_disabled(p)) {
+		p->flags &= ~KPROBE_FLAG_DISABLED;
+		arm_kprobe(p);
+	}
+out:
+	mutex_unlock(&kprobe_mutex);
+	return ret;
+}
+EXPORT_SYMBOL_GPL(enable_kprobe);
+
 void __kprobes dump_kprobe(struct kprobe *kp)
 {
 	printk(KERN_WARNING "Dumping kprobe:\n");
@@ -1805,72 +1871,6 @@ static const struct file_operations debugfs_kprobes_operations = {
 	.release        = seq_release,
 };
 
-/* Disable one kprobe */
-int __kprobes disable_kprobe(struct kprobe *kp)
-{
-	int ret = 0;
-	struct kprobe *p;
-
-	mutex_lock(&kprobe_mutex);
-
-	/* Check whether specified probe is valid. */
-	p = __get_valid_kprobe(kp);
-	if (unlikely(p == NULL)) {
-		ret = -EINVAL;
-		goto out;
-	}
-
-	/* If the probe is already disabled (or gone), just return */
-	if (kprobe_disabled(kp))
-		goto out;
-
-	kp->flags |= KPROBE_FLAG_DISABLED;
-	if (p != kp)
-		/* When kp != p, p is always enabled. */
-		try_to_disable_aggr_kprobe(p);
-
-	if (!kprobes_all_disarmed && kprobe_disabled(p))
-		disarm_kprobe(p);
-out:
-	mutex_unlock(&kprobe_mutex);
-	return ret;
-}
-EXPORT_SYMBOL_GPL(disable_kprobe);
-
-/* Enable one kprobe */
-int __kprobes enable_kprobe(struct kprobe *kp)
-{
-	int ret = 0;
-	struct kprobe *p;
-
-	mutex_lock(&kprobe_mutex);
-
-	/* Check whether specified probe is valid. */
-	p = __get_valid_kprobe(kp);
-	if (unlikely(p == NULL)) {
-		ret = -EINVAL;
-		goto out;
-	}
-
-	if (kprobe_gone(kp)) {
-		/* This kprobe has gone, we couldn't enable it. */
-		ret = -EINVAL;
-		goto out;
-	}
-
-	if (p != kp)
-		kp->flags &= ~KPROBE_FLAG_DISABLED;
-
-	if (!kprobes_all_disarmed && kprobe_disabled(p)) {
-		p->flags &= ~KPROBE_FLAG_DISABLED;
-		arm_kprobe(p);
-	}
-out:
-	mutex_unlock(&kprobe_mutex);
-	return ret;
-}
-EXPORT_SYMBOL_GPL(enable_kprobe);
-
 static void __kprobes arm_all_kprobes(void)
 {
 	struct hlist_head *head;
diff --git a/kernel/lockdep.c b/kernel/lockdep.c
index 2594e1c..e9c759f 100644
--- a/kernel/lockdep.c
+++ b/kernel/lockdep.c
@@ -3227,7 +3227,7 @@ void lock_release(struct lockdep_map *lock, int nested,
 	raw_local_irq_save(flags);
 	check_flags(flags);
 	current->lockdep_recursion = 1;
-	trace_lock_release(lock, nested, ip);
+	trace_lock_release(lock, ip);
 	__lock_release(lock, nested, ip);
 	current->lockdep_recursion = 0;
 	raw_local_irq_restore(flags);
@@ -3380,7 +3380,7 @@ found_it:
 		hlock->holdtime_stamp = now;
 	}
 
-	trace_lock_acquired(lock, ip, waittime);
+	trace_lock_acquired(lock, ip);
 
 	stats = get_lock_stats(hlock_class(hlock));
 	if (waittime) {
diff --git a/kernel/perf_event.c b/kernel/perf_event.c
index 2f3fbf8..a4fa381 100644
--- a/kernel/perf_event.c
+++ b/kernel/perf_event.c
@@ -16,6 +16,7 @@
 #include <linux/file.h>
 #include <linux/poll.h>
 #include <linux/slab.h>
+#include <linux/hash.h>
 #include <linux/sysfs.h>
 #include <linux/dcache.h>
 #include <linux/percpu.h>
@@ -82,14 +83,6 @@ extern __weak const struct pmu *hw_perf_event_init(struct perf_event *event)
 void __weak hw_perf_disable(void)		{ barrier(); }
 void __weak hw_perf_enable(void)		{ barrier(); }
 
-int __weak
-hw_perf_group_sched_in(struct perf_event *group_leader,
-	       struct perf_cpu_context *cpuctx,
-	       struct perf_event_context *ctx)
-{
-	return 0;
-}
-
 void __weak perf_event_print_debug(void)	{ }
 
 static DEFINE_PER_CPU(int, perf_disable_count);
@@ -262,6 +255,18 @@ static void update_event_times(struct perf_event *event)
 	event->total_time_running = run_end - event->tstamp_running;
 }
 
+/*
+ * Update total_time_enabled and total_time_running for all events in a group.
+ */
+static void update_group_times(struct perf_event *leader)
+{
+	struct perf_event *event;
+
+	update_event_times(leader);
+	list_for_each_entry(event, &leader->sibling_list, group_entry)
+		update_event_times(event);
+}
+
 static struct list_head *
 ctx_group_list(struct perf_event *event, struct perf_event_context *ctx)
 {
@@ -315,8 +320,6 @@ list_add_event(struct perf_event *event, struct perf_event_context *ctx)
 static void
 list_del_event(struct perf_event *event, struct perf_event_context *ctx)
 {
-	struct perf_event *sibling, *tmp;
-
 	if (list_empty(&event->group_entry))
 		return;
 	ctx->nr_events--;
@@ -329,7 +332,7 @@ list_del_event(struct perf_event *event, struct perf_event_context *ctx)
 	if (event->group_leader != event)
 		event->group_leader->nr_siblings--;
 
-	update_event_times(event);
+	update_group_times(event);
 
 	/*
 	 * If event was in error state, then keep it
@@ -340,6 +343,12 @@ list_del_event(struct perf_event *event, struct perf_event_context *ctx)
 	 */
 	if (event->state > PERF_EVENT_STATE_OFF)
 		event->state = PERF_EVENT_STATE_OFF;
+}
+
+static void
+perf_destroy_group(struct perf_event *event, struct perf_event_context *ctx)
+{
+	struct perf_event *sibling, *tmp;
 
 	/*
 	 * If this was a group event with sibling events then
@@ -505,18 +514,6 @@ retry:
 }
 
 /*
- * Update total_time_enabled and total_time_running for all events in a group.
- */
-static void update_group_times(struct perf_event *leader)
-{
-	struct perf_event *event;
-
-	update_event_times(leader);
-	list_for_each_entry(event, &leader->sibling_list, group_entry)
-		update_event_times(event);
-}
-
-/*
  * Cross CPU call to disable a performance event
  */
 static void __perf_event_disable(void *info)
@@ -640,15 +637,20 @@ group_sched_in(struct perf_event *group_event,
 	       struct perf_cpu_context *cpuctx,
 	       struct perf_event_context *ctx)
 {
-	struct perf_event *event, *partial_group;
+	struct perf_event *event, *partial_group = NULL;
+	const struct pmu *pmu = group_event->pmu;
+	bool txn = false;
 	int ret;
 
 	if (group_event->state == PERF_EVENT_STATE_OFF)
 		return 0;
 
-	ret = hw_perf_group_sched_in(group_event, cpuctx, ctx);
-	if (ret)
-		return ret < 0 ? ret : 0;
+	/* Check if group transaction availabe */
+	if (pmu->start_txn)
+		txn = true;
+
+	if (txn)
+		pmu->start_txn(pmu);
 
 	if (event_sched_in(group_event, cpuctx, ctx))
 		return -EAGAIN;
@@ -663,9 +665,19 @@ group_sched_in(struct perf_event *group_event,
 		}
 	}
 
-	return 0;
+	if (!txn)
+		return 0;
+
+	ret = pmu->commit_txn(pmu);
+	if (!ret) {
+		pmu->cancel_txn(pmu);
+		return 0;
+	}
 
 group_error:
+	if (txn)
+		pmu->cancel_txn(pmu);
+
 	/*
 	 * Groups can be scheduled in as one unit only, so undo any
 	 * partial group before returning:
@@ -1367,6 +1379,8 @@ void perf_event_task_sched_in(struct task_struct *task)
 	if (cpuctx->task_ctx == ctx)
 		return;
 
+	perf_disable();
+
 	/*
 	 * We want to keep the following priority order:
 	 * cpu pinned (that don't need to move), task pinned,
@@ -1379,6 +1393,8 @@ void perf_event_task_sched_in(struct task_struct *task)
 	ctx_sched_in(ctx, cpuctx, EVENT_FLEXIBLE);
 
 	cpuctx->task_ctx = ctx;
+
+	perf_enable();
 }
 
 #define MAX_INTERRUPTS (~0ULL)
@@ -1856,9 +1872,30 @@ int perf_event_release_kernel(struct perf_event *event)
 {
 	struct perf_event_context *ctx = event->ctx;
 
+	/*
+	 * Remove from the PMU, can't get re-enabled since we got
+	 * here because the last ref went.
+	 */
+	perf_event_disable(event);
+
 	WARN_ON_ONCE(ctx->parent_ctx);
-	mutex_lock(&ctx->mutex);
-	perf_event_remove_from_context(event);
+	/*
+	 * There are two ways this annotation is useful:
+	 *
+	 *  1) there is a lock recursion from perf_event_exit_task
+	 *     see the comment there.
+	 *
+	 *  2) there is a lock-inversion with mmap_sem through
+	 *     perf_event_read_group(), which takes faults while
+	 *     holding ctx->mutex, however this is called after
+	 *     the last filedesc died, so there is no possibility
+	 *     to trigger the AB-BA case.
+	 */
+	mutex_lock_nested(&ctx->mutex, SINGLE_DEPTH_NESTING);
+	raw_spin_lock_irq(&ctx->lock);
+	list_del_event(event, ctx);
+	perf_destroy_group(event, ctx);
+	raw_spin_unlock_irq(&ctx->lock);
 	mutex_unlock(&ctx->mutex);
 
 	mutex_lock(&event->owner->perf_event_mutex);
@@ -2642,6 +2679,7 @@ static int perf_fasync(int fd, struct file *filp, int on)
 }
 
 static const struct file_operations perf_fops = {
+	.llseek			= no_llseek,
 	.release		= perf_release,
 	.read			= perf_read,
 	.poll			= perf_poll,
@@ -2792,6 +2830,27 @@ void perf_arch_fetch_caller_regs(struct pt_regs *regs, unsigned long ip, int ski
 
 
 /*
+ * We assume there is only KVM supporting the callbacks.
+ * Later on, we might change it to a list if there is
+ * another virtualization implementation supporting the callbacks.
+ */
+struct perf_guest_info_callbacks *perf_guest_cbs;
+
+int perf_register_guest_info_callbacks(struct perf_guest_info_callbacks *cbs)
+{
+	perf_guest_cbs = cbs;
+	return 0;
+}
+EXPORT_SYMBOL_GPL(perf_register_guest_info_callbacks);
+
+int perf_unregister_guest_info_callbacks(struct perf_guest_info_callbacks *cbs)
+{
+	perf_guest_cbs = NULL;
+	return 0;
+}
+EXPORT_SYMBOL_GPL(perf_unregister_guest_info_callbacks);
+
+/*
  * Output
  */
 static bool perf_output_space(struct perf_mmap_data *data, unsigned long tail,
@@ -3743,7 +3802,7 @@ void __perf_event_mmap(struct vm_area_struct *vma)
 		.event_id  = {
 			.header = {
 				.type = PERF_RECORD_MMAP,
-				.misc = 0,
+				.misc = PERF_RECORD_MISC_USER,
 				/* .size */
 			},
 			/* .pid */
@@ -3961,36 +4020,6 @@ static void perf_swevent_add(struct perf_event *event, u64 nr,
 	perf_swevent_overflow(event, 0, nmi, data, regs);
 }
 
-static int perf_swevent_is_counting(struct perf_event *event)
-{
-	/*
-	 * The event is active, we're good!
-	 */
-	if (event->state == PERF_EVENT_STATE_ACTIVE)
-		return 1;
-
-	/*
-	 * The event is off/error, not counting.
-	 */
-	if (event->state != PERF_EVENT_STATE_INACTIVE)
-		return 0;
-
-	/*
-	 * The event is inactive, if the context is active
-	 * we're part of a group that didn't make it on the 'pmu',
-	 * not counting.
-	 */
-	if (event->ctx->is_active)
-		return 0;
-
-	/*
-	 * We're inactive and the context is too, this means the
-	 * task is scheduled out, we're counting events that happen
-	 * to us, like migration events.
-	 */
-	return 1;
-}
-
 static int perf_tp_event_match(struct perf_event *event,
 				struct perf_sample_data *data);
 
@@ -4014,12 +4043,6 @@ static int perf_swevent_match(struct perf_event *event,
 				struct perf_sample_data *data,
 				struct pt_regs *regs)
 {
-	if (event->cpu != -1 && event->cpu != smp_processor_id())
-		return 0;
-
-	if (!perf_swevent_is_counting(event))
-		return 0;
-
 	if (event->attr.type != type)
 		return 0;
 
@@ -4036,18 +4059,53 @@ static int perf_swevent_match(struct perf_event *event,
 	return 1;
 }
 
-static void perf_swevent_ctx_event(struct perf_event_context *ctx,
-				     enum perf_type_id type,
-				     u32 event_id, u64 nr, int nmi,
-				     struct perf_sample_data *data,
-				     struct pt_regs *regs)
+static inline u64 swevent_hash(u64 type, u32 event_id)
+{
+	u64 val = event_id | (type << 32);
+
+	return hash_64(val, SWEVENT_HLIST_BITS);
+}
+
+static struct hlist_head *
+find_swevent_head(struct perf_cpu_context *ctx, u64 type, u32 event_id)
+{
+	u64 hash;
+	struct swevent_hlist *hlist;
+
+	hash = swevent_hash(type, event_id);
+
+	hlist = rcu_dereference(ctx->swevent_hlist);
+	if (!hlist)
+		return NULL;
+
+	return &hlist->heads[hash];
+}
+
+static void do_perf_sw_event(enum perf_type_id type, u32 event_id,
+				    u64 nr, int nmi,
+				    struct perf_sample_data *data,
+				    struct pt_regs *regs)
 {
+	struct perf_cpu_context *cpuctx;
 	struct perf_event *event;
+	struct hlist_node *node;
+	struct hlist_head *head;
 
-	list_for_each_entry_rcu(event, &ctx->event_list, event_entry) {
+	cpuctx = &__get_cpu_var(perf_cpu_context);
+
+	rcu_read_lock();
+
+	head = find_swevent_head(cpuctx, type, event_id);
+
+	if (!head)
+		goto end;
+
+	hlist_for_each_entry_rcu(event, node, head, hlist_entry) {
 		if (perf_swevent_match(event, type, event_id, data, regs))
 			perf_swevent_add(event, nr, nmi, data, regs);
 	}
+end:
+	rcu_read_unlock();
 }
 
 int perf_swevent_get_recursion_context(void)
@@ -4085,27 +4143,6 @@ void perf_swevent_put_recursion_context(int rctx)
 }
 EXPORT_SYMBOL_GPL(perf_swevent_put_recursion_context);
 
-static void do_perf_sw_event(enum perf_type_id type, u32 event_id,
-				    u64 nr, int nmi,
-				    struct perf_sample_data *data,
-				    struct pt_regs *regs)
-{
-	struct perf_cpu_context *cpuctx;
-	struct perf_event_context *ctx;
-
-	cpuctx = &__get_cpu_var(perf_cpu_context);
-	rcu_read_lock();
-	perf_swevent_ctx_event(&cpuctx->ctx, type, event_id,
-				 nr, nmi, data, regs);
-	/*
-	 * doesn't really matter which of the child contexts the
-	 * events ends up in.
-	 */
-	ctx = rcu_dereference(current->perf_event_ctxp);
-	if (ctx)
-		perf_swevent_ctx_event(ctx, type, event_id, nr, nmi, data, regs);
-	rcu_read_unlock();
-}
 
 void __perf_sw_event(u32 event_id, u64 nr, int nmi,
 			    struct pt_regs *regs, u64 addr)
@@ -4131,16 +4168,28 @@ static void perf_swevent_read(struct perf_event *event)
 static int perf_swevent_enable(struct perf_event *event)
 {
 	struct hw_perf_event *hwc = &event->hw;
+	struct perf_cpu_context *cpuctx;
+	struct hlist_head *head;
+
+	cpuctx = &__get_cpu_var(perf_cpu_context);
 
 	if (hwc->sample_period) {
 		hwc->last_period = hwc->sample_period;
 		perf_swevent_set_period(event);
 	}
+
+	head = find_swevent_head(cpuctx, event->attr.type, event->attr.config);
+	if (WARN_ON_ONCE(!head))
+		return -EINVAL;
+
+	hlist_add_head_rcu(&event->hlist_entry, head);
+
 	return 0;
 }
 
 static void perf_swevent_disable(struct perf_event *event)
 {
+	hlist_del_rcu(&event->hlist_entry);
 }
 
 static const struct pmu perf_ops_generic = {
@@ -4168,15 +4217,8 @@ static enum hrtimer_restart perf_swevent_hrtimer(struct hrtimer *hrtimer)
 	perf_sample_data_init(&data, 0);
 	data.period = event->hw.last_period;
 	regs = get_irq_regs();
-	/*
-	 * In case we exclude kernel IPs or are somehow not in interrupt
-	 * context, provide the next best thing, the user IP.
-	 */
-	if ((event->attr.exclude_kernel || !regs) &&
-			!event->attr.exclude_user)
-		regs = task_pt_regs(current);
 
-	if (regs) {
+	if (regs && !perf_exclude_event(event, regs)) {
 		if (!(event->attr.exclude_idle && current->pid == 0))
 			if (perf_event_overflow(event, 0, &data, regs))
 				ret = HRTIMER_NORESTART;
@@ -4324,6 +4366,105 @@ static const struct pmu perf_ops_task_clock = {
 	.read		= task_clock_perf_event_read,
 };
 
+static void swevent_hlist_release_rcu(struct rcu_head *rcu_head)
+{
+	struct swevent_hlist *hlist;
+
+	hlist = container_of(rcu_head, struct swevent_hlist, rcu_head);
+	kfree(hlist);
+}
+
+static void swevent_hlist_release(struct perf_cpu_context *cpuctx)
+{
+	struct swevent_hlist *hlist;
+
+	if (!cpuctx->swevent_hlist)
+		return;
+
+	hlist = cpuctx->swevent_hlist;
+	rcu_assign_pointer(cpuctx->swevent_hlist, NULL);
+	call_rcu(&hlist->rcu_head, swevent_hlist_release_rcu);
+}
+
+static void swevent_hlist_put_cpu(struct perf_event *event, int cpu)
+{
+	struct perf_cpu_context *cpuctx = &per_cpu(perf_cpu_context, cpu);
+
+	mutex_lock(&cpuctx->hlist_mutex);
+
+	if (!--cpuctx->hlist_refcount)
+		swevent_hlist_release(cpuctx);
+
+	mutex_unlock(&cpuctx->hlist_mutex);
+}
+
+static void swevent_hlist_put(struct perf_event *event)
+{
+	int cpu;
+
+	if (event->cpu != -1) {
+		swevent_hlist_put_cpu(event, event->cpu);
+		return;
+	}
+
+	for_each_possible_cpu(cpu)
+		swevent_hlist_put_cpu(event, cpu);
+}
+
+static int swevent_hlist_get_cpu(struct perf_event *event, int cpu)
+{
+	struct perf_cpu_context *cpuctx = &per_cpu(perf_cpu_context, cpu);
+	int err = 0;
+
+	mutex_lock(&cpuctx->hlist_mutex);
+
+	if (!cpuctx->swevent_hlist && cpu_online(cpu)) {
+		struct swevent_hlist *hlist;
+
+		hlist = kzalloc(sizeof(*hlist), GFP_KERNEL);
+		if (!hlist) {
+			err = -ENOMEM;
+			goto exit;
+		}
+		rcu_assign_pointer(cpuctx->swevent_hlist, hlist);
+	}
+	cpuctx->hlist_refcount++;
+ exit:
+	mutex_unlock(&cpuctx->hlist_mutex);
+
+	return err;
+}
+
+static int swevent_hlist_get(struct perf_event *event)
+{
+	int err;
+	int cpu, failed_cpu;
+
+	if (event->cpu != -1)
+		return swevent_hlist_get_cpu(event, event->cpu);
+
+	get_online_cpus();
+	for_each_possible_cpu(cpu) {
+		err = swevent_hlist_get_cpu(event, cpu);
+		if (err) {
+			failed_cpu = cpu;
+			goto fail;
+		}
+	}
+	put_online_cpus();
+
+	return 0;
+ fail:
+	for_each_possible_cpu(cpu) {
+		if (cpu == failed_cpu)
+			break;
+		swevent_hlist_put_cpu(event, cpu);
+	}
+
+	put_online_cpus();
+	return err;
+}
+
 #ifdef CONFIG_EVENT_TRACING
 
 void perf_tp_event(int event_id, u64 addr, u64 count, void *record,
@@ -4357,10 +4498,13 @@ static int perf_tp_event_match(struct perf_event *event,
 static void tp_perf_event_destroy(struct perf_event *event)
 {
 	perf_trace_disable(event->attr.config);
+	swevent_hlist_put(event);
 }
 
 static const struct pmu *tp_perf_event_init(struct perf_event *event)
 {
+	int err;
+
 	/*
 	 * Raw tracepoint data is a severe data leak, only allow root to
 	 * have these.
@@ -4374,6 +4518,11 @@ static const struct pmu *tp_perf_event_init(struct perf_event *event)
 		return NULL;
 
 	event->destroy = tp_perf_event_destroy;
+	err = swevent_hlist_get(event);
+	if (err) {
+		perf_trace_disable(event->attr.config);
+		return ERR_PTR(err);
+	}
 
 	return &perf_ops_generic;
 }
@@ -4474,6 +4623,7 @@ static void sw_perf_event_destroy(struct perf_event *event)
 	WARN_ON(event->parent);
 
 	atomic_dec(&perf_swevent_enabled[event_id]);
+	swevent_hlist_put(event);
 }
 
 static const struct pmu *sw_perf_event_init(struct perf_event *event)
@@ -4512,6 +4662,12 @@ static const struct pmu *sw_perf_event_init(struct perf_event *event)
 	case PERF_COUNT_SW_ALIGNMENT_FAULTS:
 	case PERF_COUNT_SW_EMULATION_FAULTS:
 		if (!event->parent) {
+			int err;
+
+			err = swevent_hlist_get(event);
+			if (err)
+				return ERR_PTR(err);
+
 			atomic_inc(&perf_swevent_enabled[event_id]);
 			event->destroy = sw_perf_event_destroy;
 		}
@@ -4897,7 +5053,7 @@ err_fput_free_put_context:
 
 err_free_put_context:
 	if (err < 0)
-		kfree(event);
+		free_event(event);
 
 err_put_context:
 	if (err < 0)
@@ -5176,7 +5332,7 @@ void perf_event_exit_task(struct task_struct *child)
 	 *
 	 * But since its the parent context it won't be the same instance.
 	 */
-	mutex_lock_nested(&child_ctx->mutex, SINGLE_DEPTH_NESTING);
+	mutex_lock(&child_ctx->mutex);
 
 again:
 	list_for_each_entry_safe(child_event, tmp, &child_ctx->pinned_groups,
@@ -5384,6 +5540,7 @@ static void __init perf_event_init_all_cpus(void)
 
 	for_each_possible_cpu(cpu) {
 		cpuctx = &per_cpu(perf_cpu_context, cpu);
+		mutex_init(&cpuctx->hlist_mutex);
 		__perf_event_init_context(&cpuctx->ctx, NULL);
 	}
 }
@@ -5397,6 +5554,16 @@ static void __cpuinit perf_event_init_cpu(int cpu)
 	spin_lock(&perf_resource_lock);
 	cpuctx->max_pertask = perf_max_events - perf_reserved_percpu;
 	spin_unlock(&perf_resource_lock);
+
+	mutex_lock(&cpuctx->hlist_mutex);
+	if (cpuctx->hlist_refcount > 0) {
+		struct swevent_hlist *hlist;
+
+		hlist = kzalloc(sizeof(*hlist), GFP_KERNEL);
+		WARN_ON_ONCE(!hlist);
+		rcu_assign_pointer(cpuctx->swevent_hlist, hlist);
+	}
+	mutex_unlock(&cpuctx->hlist_mutex);
 }
 
 #ifdef CONFIG_HOTPLUG_CPU
@@ -5416,6 +5583,10 @@ static void perf_event_exit_cpu(int cpu)
 	struct perf_cpu_context *cpuctx = &per_cpu(perf_cpu_context, cpu);
 	struct perf_event_context *ctx = &cpuctx->ctx;
 
+	mutex_lock(&cpuctx->hlist_mutex);
+	swevent_hlist_release(cpuctx);
+	mutex_unlock(&cpuctx->hlist_mutex);
+
 	mutex_lock(&ctx->mutex);
 	smp_call_function_single(cpu, __perf_event_exit_cpu, NULL, 1);
 	mutex_unlock(&ctx->mutex);
diff --git a/kernel/ptrace.c b/kernel/ptrace.c
index 42ad8ae..9fb5123 100644
--- a/kernel/ptrace.c
+++ b/kernel/ptrace.c
@@ -76,7 +76,6 @@ void __ptrace_unlink(struct task_struct *child)
 	child->parent = child->real_parent;
 	list_del_init(&child->ptrace_entry);
 
-	arch_ptrace_untrace(child);
 	if (task_is_traced(child))
 		ptrace_untrace(child);
 }
diff --git a/kernel/sched.c b/kernel/sched.c
index b531d79..78554dd 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -323,6 +323,15 @@ static inline struct task_group *task_group(struct task_struct *p)
 /* Change a task's cfs_rq and parent entity if it moves across CPUs/groups */
 static inline void set_task_rq(struct task_struct *p, unsigned int cpu)
 {
+	/*
+	 * Strictly speaking this rcu_read_lock() is not needed since the
+	 * task_group is tied to the cgroup, which in turn can never go away
+	 * as long as there are tasks attached to it.
+	 *
+	 * However since task_group() uses task_subsys_state() which is an
+	 * rcu_dereference() user, this quiets CONFIG_PROVE_RCU.
+	 */
+	rcu_read_lock();
 #ifdef CONFIG_FAIR_GROUP_SCHED
 	p->se.cfs_rq = task_group(p)->cfs_rq[cpu];
 	p->se.parent = task_group(p)->se[cpu];
@@ -332,6 +341,7 @@ static inline void set_task_rq(struct task_struct *p, unsigned int cpu)
 	p->rt.rt_rq  = task_group(p)->rt_rq[cpu];
 	p->rt.parent = task_group(p)->rt_se[cpu];
 #endif
+	rcu_read_unlock();
 }
 
 #else
@@ -2058,49 +2068,6 @@ static bool migrate_task(struct task_struct *p, int dest_cpu)
 }
 
 /*
- * wait_task_context_switch -	wait for a thread to complete at least one
- *				context switch.
- *
- * @p must not be current.
- */
-void wait_task_context_switch(struct task_struct *p)
-{
-	unsigned long nvcsw, nivcsw, flags;
-	int running;
-	struct rq *rq;
-
-	nvcsw	= p->nvcsw;
-	nivcsw	= p->nivcsw;
-	for (;;) {
-		/*
-		 * The runqueue is assigned before the actual context
-		 * switch. We need to take the runqueue lock.
-		 *
-		 * We could check initially without the lock but it is
-		 * very likely that we need to take the lock in every
-		 * iteration.
-		 */
-		rq = task_rq_lock(p, &flags);
-		running = task_running(rq, p);
-		task_rq_unlock(rq, &flags);
-
-		if (likely(!running))
-			break;
-		/*
-		 * The switch count is incremented before the actual
-		 * context switch. We thus wait for two switches to be
-		 * sure at least one completed.
-		 */
-		if ((p->nvcsw - nvcsw) > 1)
-			break;
-		if ((p->nivcsw - nivcsw) > 1)
-			break;
-
-		cpu_relax();
-	}
-}
-
-/*
  * wait_task_inactive - wait for a thread to unschedule.
  *
  * If @match_state is nonzero, it's the @p->state value just checked and
@@ -3724,7 +3691,7 @@ int mutex_spin_on_owner(struct mutex *lock, struct thread_info *owner)
 	 * the mutex owner just released it and exited.
 	 */
 	if (probe_kernel_address(&owner->cpu, cpu))
-		goto out;
+		return 0;
 #else
 	cpu = owner->cpu;
 #endif
@@ -3734,14 +3701,14 @@ int mutex_spin_on_owner(struct mutex *lock, struct thread_info *owner)
 	 * the cpu field may no longer be valid.
 	 */
 	if (cpu >= nr_cpumask_bits)
-		goto out;
+		return 0;
 
 	/*
 	 * We need to validate that we can do a
 	 * get_cpu() and that we have the percpu area.
 	 */
 	if (!cpu_online(cpu))
-		goto out;
+		return 0;
 
 	rq = cpu_rq(cpu);
 
@@ -3760,7 +3727,7 @@ int mutex_spin_on_owner(struct mutex *lock, struct thread_info *owner)
 
 		cpu_relax();
 	}
-out:
+
 	return 1;
 }
 #endif
diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig
index 13e13d4..8b1797c 100644
--- a/kernel/trace/Kconfig
+++ b/kernel/trace/Kconfig
@@ -44,9 +44,6 @@ config HAVE_FTRACE_MCOUNT_RECORD
 	help
 	  See Documentation/trace/ftrace-design.txt
 
-config HAVE_HW_BRANCH_TRACER
-	bool
-
 config HAVE_SYSCALL_TRACEPOINTS
 	bool
 	help
@@ -374,14 +371,6 @@ config STACK_TRACER
 
 	  Say N if unsure.
 
-config HW_BRANCH_TRACER
-	depends on HAVE_HW_BRANCH_TRACER
-	bool "Trace hw branches"
-	select GENERIC_TRACER
-	help
-	  This tracer records all branches on the system in a circular
-	  buffer, giving access to the last N branches for each cpu.
-
 config KMEMTRACE
 	bool "Trace SLAB allocations"
 	select GENERIC_TRACER
diff --git a/kernel/trace/Makefile b/kernel/trace/Makefile
index 78edc64..ffb1a5b 100644
--- a/kernel/trace/Makefile
+++ b/kernel/trace/Makefile
@@ -41,7 +41,6 @@ obj-$(CONFIG_MMIOTRACE) += trace_mmiotrace.o
 obj-$(CONFIG_BOOT_TRACER) += trace_boot.o
 obj-$(CONFIG_FUNCTION_GRAPH_TRACER) += trace_functions_graph.o
 obj-$(CONFIG_TRACE_BRANCH_PROFILING) += trace_branch.o
-obj-$(CONFIG_HW_BRANCH_TRACER) += trace_hw_branches.o
 obj-$(CONFIG_KMEMTRACE) += kmemtrace.o
 obj-$(CONFIG_WORKQUEUE_TRACER) += trace_workqueue.o
 obj-$(CONFIG_BLK_DEV_IO_TRACE) += blktrace.o
diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h
index 40cd171..2cd9639 100644
--- a/kernel/trace/trace.h
+++ b/kernel/trace/trace.h
@@ -34,7 +34,6 @@ enum trace_type {
 	TRACE_GRAPH_RET,
 	TRACE_GRAPH_ENT,
 	TRACE_USER_STACK,
-	TRACE_HW_BRANCHES,
 	TRACE_KMEM_ALLOC,
 	TRACE_KMEM_FREE,
 	TRACE_BLK,
@@ -103,29 +102,17 @@ struct syscall_trace_exit {
 	long			ret;
 };
 
-struct kprobe_trace_entry {
+struct kprobe_trace_entry_head {
 	struct trace_entry	ent;
 	unsigned long		ip;
-	int			nargs;
-	unsigned long		args[];
 };
 
-#define SIZEOF_KPROBE_TRACE_ENTRY(n)			\
-	(offsetof(struct kprobe_trace_entry, args) +	\
-	(sizeof(unsigned long) * (n)))
-
-struct kretprobe_trace_entry {
+struct kretprobe_trace_entry_head {
 	struct trace_entry	ent;
 	unsigned long		func;
 	unsigned long		ret_ip;
-	int			nargs;
-	unsigned long		args[];
 };
 
-#define SIZEOF_KRETPROBE_TRACE_ENTRY(n)			\
-	(offsetof(struct kretprobe_trace_entry, args) +	\
-	(sizeof(unsigned long) * (n)))
-
 /*
  * trace_flag_type is an enumeration that holds different
  * states when a trace occurs. These are:
@@ -229,7 +216,6 @@ extern void __ftrace_bad_type(void);
 			  TRACE_GRAPH_ENT);		\
 		IF_ASSIGN(var, ent, struct ftrace_graph_ret_entry,	\
 			  TRACE_GRAPH_RET);		\
-		IF_ASSIGN(var, ent, struct hw_branch_entry, TRACE_HW_BRANCHES);\
 		IF_ASSIGN(var, ent, struct kmemtrace_alloc_entry,	\
 			  TRACE_KMEM_ALLOC);	\
 		IF_ASSIGN(var, ent, struct kmemtrace_free_entry,	\
@@ -470,8 +456,6 @@ extern int trace_selftest_startup_sysprof(struct tracer *trace,
 					       struct trace_array *tr);
 extern int trace_selftest_startup_branch(struct tracer *trace,
 					 struct trace_array *tr);
-extern int trace_selftest_startup_hw_branches(struct tracer *trace,
-					      struct trace_array *tr);
 extern int trace_selftest_startup_ksym(struct tracer *trace,
 					 struct trace_array *tr);
 #endif /* CONFIG_FTRACE_STARTUP_TEST */
diff --git a/kernel/trace/trace_entries.h b/kernel/trace/trace_entries.h
index c16a08f..dc008c1 100644
--- a/kernel/trace/trace_entries.h
+++ b/kernel/trace/trace_entries.h
@@ -318,18 +318,6 @@ FTRACE_ENTRY(branch, trace_branch,
 		 __entry->func, __entry->file, __entry->correct)
 );
 
-FTRACE_ENTRY(hw_branch, hw_branch_entry,
-
-	TRACE_HW_BRANCHES,
-
-	F_STRUCT(
-		__field(	u64,	from	)
-		__field(	u64,	to	)
-	),
-
-	F_printk("from: %llx to: %llx", __entry->from, __entry->to)
-);
-
 FTRACE_ENTRY(kmem_alloc, kmemtrace_alloc_entry,
 
 	TRACE_KMEM_ALLOC,
diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c
index 239ea5d..57bb1bb 100644
--- a/kernel/trace/trace_events_filter.c
+++ b/kernel/trace/trace_events_filter.c
@@ -1400,7 +1400,7 @@ int ftrace_profile_set_filter(struct perf_event *event, int event_id,
 	}
 
 	err = -EINVAL;
-	if (!call)
+	if (&call->list == &ftrace_events)
 		goto out_unlock;
 
 	err = -EEXIST;
diff --git a/kernel/trace/trace_hw_branches.c b/kernel/trace/trace_hw_branches.c
deleted file mode 100644
index 7b97000..0000000
--- a/kernel/trace/trace_hw_branches.c
+++ /dev/null
@@ -1,312 +0,0 @@
-/*
- * h/w branch tracer for x86 based on BTS
- *
- * Copyright (C) 2008-2009 Intel Corporation.
- * Markus Metzger <markus.t.metzger@gmail.com>, 2008-2009
- */
-#include <linux/kallsyms.h>
-#include <linux/debugfs.h>
-#include <linux/ftrace.h>
-#include <linux/module.h>
-#include <linux/cpu.h>
-#include <linux/smp.h>
-#include <linux/fs.h>
-
-#include <asm/ds.h>
-
-#include "trace_output.h"
-#include "trace.h"
-
-
-#define BTS_BUFFER_SIZE (1 << 13)
-
-static DEFINE_PER_CPU(struct bts_tracer *, hwb_tracer);
-static DEFINE_PER_CPU(unsigned char[BTS_BUFFER_SIZE], hwb_buffer);
-
-#define this_tracer per_cpu(hwb_tracer, smp_processor_id())
-
-static int trace_hw_branches_enabled __read_mostly;
-static int trace_hw_branches_suspended __read_mostly;
-static struct trace_array *hw_branch_trace __read_mostly;
-
-
-static void bts_trace_init_cpu(int cpu)
-{
-	per_cpu(hwb_tracer, cpu) =
-		ds_request_bts_cpu(cpu, per_cpu(hwb_buffer, cpu),
-				   BTS_BUFFER_SIZE, NULL, (size_t)-1,
-				   BTS_KERNEL);
-
-	if (IS_ERR(per_cpu(hwb_tracer, cpu)))
-		per_cpu(hwb_tracer, cpu) = NULL;
-}
-
-static int bts_trace_init(struct trace_array *tr)
-{
-	int cpu;
-
-	hw_branch_trace = tr;
-	trace_hw_branches_enabled = 0;
-
-	get_online_cpus();
-	for_each_online_cpu(cpu) {
-		bts_trace_init_cpu(cpu);
-
-		if (likely(per_cpu(hwb_tracer, cpu)))
-			trace_hw_branches_enabled = 1;
-	}
-	trace_hw_branches_suspended = 0;
-	put_online_cpus();
-
-	/* If we could not enable tracing on a single cpu, we fail. */
-	return trace_hw_branches_enabled ? 0 : -EOPNOTSUPP;
-}
-
-static void bts_trace_reset(struct trace_array *tr)
-{
-	int cpu;
-
-	get_online_cpus();
-	for_each_online_cpu(cpu) {
-		if (likely(per_cpu(hwb_tracer, cpu))) {
-			ds_release_bts(per_cpu(hwb_tracer, cpu));
-			per_cpu(hwb_tracer, cpu) = NULL;
-		}
-	}
-	trace_hw_branches_enabled = 0;
-	trace_hw_branches_suspended = 0;
-	put_online_cpus();
-}
-
-static void bts_trace_start(struct trace_array *tr)
-{
-	int cpu;
-
-	get_online_cpus();
-	for_each_online_cpu(cpu)
-		if (likely(per_cpu(hwb_tracer, cpu)))
-			ds_resume_bts(per_cpu(hwb_tracer, cpu));
-	trace_hw_branches_suspended = 0;
-	put_online_cpus();
-}
-
-static void bts_trace_stop(struct trace_array *tr)
-{
-	int cpu;
-
-	get_online_cpus();
-	for_each_online_cpu(cpu)
-		if (likely(per_cpu(hwb_tracer, cpu)))
-			ds_suspend_bts(per_cpu(hwb_tracer, cpu));
-	trace_hw_branches_suspended = 1;
-	put_online_cpus();
-}
-
-static int __cpuinit bts_hotcpu_handler(struct notifier_block *nfb,
-				     unsigned long action, void *hcpu)
-{
-	int cpu = (long)hcpu;
-
-	switch (action) {
-	case CPU_ONLINE:
-	case CPU_DOWN_FAILED:
-		/* The notification is sent with interrupts enabled. */
-		if (trace_hw_branches_enabled) {
-			bts_trace_init_cpu(cpu);
-
-			if (trace_hw_branches_suspended &&
-			    likely(per_cpu(hwb_tracer, cpu)))
-				ds_suspend_bts(per_cpu(hwb_tracer, cpu));
-		}
-		break;
-
-	case CPU_DOWN_PREPARE:
-		/* The notification is sent with interrupts enabled. */
-		if (likely(per_cpu(hwb_tracer, cpu))) {
-			ds_release_bts(per_cpu(hwb_tracer, cpu));
-			per_cpu(hwb_tracer, cpu) = NULL;
-		}
-	}
-
-	return NOTIFY_DONE;
-}
-
-static struct notifier_block bts_hotcpu_notifier __cpuinitdata = {
-	.notifier_call = bts_hotcpu_handler
-};
-
-static void bts_trace_print_header(struct seq_file *m)
-{
-	seq_puts(m, "# CPU#        TO  <-  FROM\n");
-}
-
-static enum print_line_t bts_trace_print_line(struct trace_iterator *iter)
-{
-	unsigned long symflags = TRACE_ITER_SYM_OFFSET;
-	struct trace_entry *entry = iter->ent;
-	struct trace_seq *seq = &iter->seq;
-	struct hw_branch_entry *it;
-
-	trace_assign_type(it, entry);
-
-	if (entry->type == TRACE_HW_BRANCHES) {
-		if (trace_seq_printf(seq, "%4d  ", iter->cpu) &&
-		    seq_print_ip_sym(seq, it->to, symflags) &&
-		    trace_seq_printf(seq, "\t  <-  ") &&
-		    seq_print_ip_sym(seq, it->from, symflags) &&
-		    trace_seq_printf(seq, "\n"))
-			return TRACE_TYPE_HANDLED;
-		return TRACE_TYPE_PARTIAL_LINE;
-	}
-	return TRACE_TYPE_UNHANDLED;
-}
-
-void trace_hw_branch(u64 from, u64 to)
-{
-	struct ftrace_event_call *call = &event_hw_branch;
-	struct trace_array *tr = hw_branch_trace;
-	struct ring_buffer_event *event;
-	struct ring_buffer *buf;
-	struct hw_branch_entry *entry;
-	unsigned long irq1;
-	int cpu;
-
-	if (unlikely(!tr))
-		return;
-
-	if (unlikely(!trace_hw_branches_enabled))
-		return;
-
-	local_irq_save(irq1);
-	cpu = raw_smp_processor_id();
-	if (atomic_inc_return(&tr->data[cpu]->disabled) != 1)
-		goto out;
-
-	buf = tr->buffer;
-	event = trace_buffer_lock_reserve(buf, TRACE_HW_BRANCHES,
-					  sizeof(*entry), 0, 0);
-	if (!event)
-		goto out;
-	entry	= ring_buffer_event_data(event);
-	tracing_generic_entry_update(&entry->ent, 0, from);
-	entry->ent.type = TRACE_HW_BRANCHES;
-	entry->from = from;
-	entry->to   = to;
-	if (!filter_check_discard(call, entry, buf, event))
-		trace_buffer_unlock_commit(buf, event, 0, 0);
-
- out:
-	atomic_dec(&tr->data[cpu]->disabled);
-	local_irq_restore(irq1);
-}
-
-static void trace_bts_at(const struct bts_trace *trace, void *at)
-{
-	struct bts_struct bts;
-	int err = 0;
-
-	WARN_ON_ONCE(!trace->read);
-	if (!trace->read)
-		return;
-
-	err = trace->read(this_tracer, at, &bts);
-	if (err < 0)
-		return;
-
-	switch (bts.qualifier) {
-	case BTS_BRANCH:
-		trace_hw_branch(bts.variant.lbr.from, bts.variant.lbr.to);
-		break;
-	}
-}
-
-/*
- * Collect the trace on the current cpu and write it into the ftrace buffer.
- *
- * pre: tracing must be suspended on the current cpu
- */
-static void trace_bts_cpu(void *arg)
-{
-	struct trace_array *tr = (struct trace_array *)arg;
-	const struct bts_trace *trace;
-	unsigned char *at;
-
-	if (unlikely(!tr))
-		return;
-
-	if (unlikely(atomic_read(&tr->data[raw_smp_processor_id()]->disabled)))
-		return;
-
-	if (unlikely(!this_tracer))
-		return;
-
-	trace = ds_read_bts(this_tracer);
-	if (!trace)
-		return;
-
-	for (at = trace->ds.top; (void *)at < trace->ds.end;
-	     at += trace->ds.size)
-		trace_bts_at(trace, at);
-
-	for (at = trace->ds.begin; (void *)at < trace->ds.top;
-	     at += trace->ds.size)
-		trace_bts_at(trace, at);
-}
-
-static void trace_bts_prepare(struct trace_iterator *iter)
-{
-	int cpu;
-
-	get_online_cpus();
-	for_each_online_cpu(cpu)
-		if (likely(per_cpu(hwb_tracer, cpu)))
-			ds_suspend_bts(per_cpu(hwb_tracer, cpu));
-	/*
-	 * We need to collect the trace on the respective cpu since ftrace
-	 * implicitly adds the record for the current cpu.
-	 * Once that is more flexible, we could collect the data from any cpu.
-	 */
-	on_each_cpu(trace_bts_cpu, iter->tr, 1);
-
-	for_each_online_cpu(cpu)
-		if (likely(per_cpu(hwb_tracer, cpu)))
-			ds_resume_bts(per_cpu(hwb_tracer, cpu));
-	put_online_cpus();
-}
-
-static void trace_bts_close(struct trace_iterator *iter)
-{
-	tracing_reset_online_cpus(iter->tr);
-}
-
-void trace_hw_branch_oops(void)
-{
-	if (this_tracer) {
-		ds_suspend_bts_noirq(this_tracer);
-		trace_bts_cpu(hw_branch_trace);
-		ds_resume_bts_noirq(this_tracer);
-	}
-}
-
-struct tracer bts_tracer __read_mostly =
-{
-	.name		= "hw-branch-tracer",
-	.init		= bts_trace_init,
-	.reset		= bts_trace_reset,
-	.print_header	= bts_trace_print_header,
-	.print_line	= bts_trace_print_line,
-	.start		= bts_trace_start,
-	.stop		= bts_trace_stop,
-	.open		= trace_bts_prepare,
-	.close		= trace_bts_close,
-#ifdef CONFIG_FTRACE_SELFTEST
-	.selftest	= trace_selftest_startup_hw_branches,
-#endif /* CONFIG_FTRACE_SELFTEST */
-};
-
-__init static int init_bts_trace(void)
-{
-	register_hotcpu_notifier(&bts_hotcpu_notifier);
-	return register_tracer(&bts_tracer);
-}
-device_initcall(init_bts_trace);
diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c
index 0e3ded6..9a082bb 100644
--- a/kernel/trace/trace_kprobe.c
+++ b/kernel/trace/trace_kprobe.c
@@ -29,6 +29,8 @@
 #include <linux/ctype.h>
 #include <linux/ptrace.h>
 #include <linux/perf_event.h>
+#include <linux/stringify.h>
+#include <asm/bitsperlong.h>
 
 #include "trace.h"
 #include "trace_output.h"
@@ -40,7 +42,6 @@
 
 /* Reserved field names */
 #define FIELD_STRING_IP "__probe_ip"
-#define FIELD_STRING_NARGS "__probe_nargs"
 #define FIELD_STRING_RETIP "__probe_ret_ip"
 #define FIELD_STRING_FUNC "__probe_func"
 
@@ -52,56 +53,102 @@ const char *reserved_field_names[] = {
 	"common_tgid",
 	"common_lock_depth",
 	FIELD_STRING_IP,
-	FIELD_STRING_NARGS,
 	FIELD_STRING_RETIP,
 	FIELD_STRING_FUNC,
 };
 
-struct fetch_func {
-	unsigned long (*func)(struct pt_regs *, void *);
+/* Printing function type */
+typedef int (*print_type_func_t)(struct trace_seq *, const char *, void *);
+#define PRINT_TYPE_FUNC_NAME(type)	print_type_##type
+#define PRINT_TYPE_FMT_NAME(type)	print_type_format_##type
+
+/* Printing  in basic type function template */
+#define DEFINE_BASIC_PRINT_TYPE_FUNC(type, fmt, cast)			\
+static __kprobes int PRINT_TYPE_FUNC_NAME(type)(struct trace_seq *s,	\
+						const char *name, void *data)\
+{									\
+	return trace_seq_printf(s, " %s=" fmt, name, (cast)*(type *)data);\
+}									\
+static const char PRINT_TYPE_FMT_NAME(type)[] = fmt;
+
+DEFINE_BASIC_PRINT_TYPE_FUNC(u8, "%x", unsigned int)
+DEFINE_BASIC_PRINT_TYPE_FUNC(u16, "%x", unsigned int)
+DEFINE_BASIC_PRINT_TYPE_FUNC(u32, "%lx", unsigned long)
+DEFINE_BASIC_PRINT_TYPE_FUNC(u64, "%llx", unsigned long long)
+DEFINE_BASIC_PRINT_TYPE_FUNC(s8, "%d", int)
+DEFINE_BASIC_PRINT_TYPE_FUNC(s16, "%d", int)
+DEFINE_BASIC_PRINT_TYPE_FUNC(s32, "%ld", long)
+DEFINE_BASIC_PRINT_TYPE_FUNC(s64, "%lld", long long)
+
+/* Data fetch function type */
+typedef	void (*fetch_func_t)(struct pt_regs *, void *, void *);
+
+struct fetch_param {
+	fetch_func_t	fn;
 	void *data;
 };
 
-static __kprobes unsigned long call_fetch(struct fetch_func *f,
-					  struct pt_regs *regs)
+static __kprobes void call_fetch(struct fetch_param *fprm,
+				 struct pt_regs *regs, void *dest)
 {
-	return f->func(regs, f->data);
+	return fprm->fn(regs, fprm->data, dest);
 }
 
-/* fetch handlers */
-static __kprobes unsigned long fetch_register(struct pt_regs *regs,
-					      void *offset)
-{
-	return regs_get_register(regs, (unsigned int)((unsigned long)offset));
+#define FETCH_FUNC_NAME(kind, type)	fetch_##kind##_##type
+/*
+ * Define macro for basic types - we don't need to define s* types, because
+ * we have to care only about bitwidth at recording time.
+ */
+#define DEFINE_BASIC_FETCH_FUNCS(kind)  \
+DEFINE_FETCH_##kind(u8)			\
+DEFINE_FETCH_##kind(u16)		\
+DEFINE_FETCH_##kind(u32)		\
+DEFINE_FETCH_##kind(u64)
+
+#define CHECK_BASIC_FETCH_FUNCS(kind, fn)	\
+	((FETCH_FUNC_NAME(kind, u8) == fn) ||	\
+	 (FETCH_FUNC_NAME(kind, u16) == fn) ||	\
+	 (FETCH_FUNC_NAME(kind, u32) == fn) ||	\
+	 (FETCH_FUNC_NAME(kind, u64) == fn))
+
+/* Data fetch function templates */
+#define DEFINE_FETCH_reg(type)						\
+static __kprobes void FETCH_FUNC_NAME(reg, type)(struct pt_regs *regs,	\
+					  void *offset, void *dest)	\
+{									\
+	*(type *)dest = (type)regs_get_register(regs,			\
+				(unsigned int)((unsigned long)offset));	\
 }
-
-static __kprobes unsigned long fetch_stack(struct pt_regs *regs,
-					   void *num)
-{
-	return regs_get_kernel_stack_nth(regs,
-					 (unsigned int)((unsigned long)num));
+DEFINE_BASIC_FETCH_FUNCS(reg)
+
+#define DEFINE_FETCH_stack(type)					\
+static __kprobes void FETCH_FUNC_NAME(stack, type)(struct pt_regs *regs,\
+					  void *offset, void *dest)	\
+{									\
+	*(type *)dest = (type)regs_get_kernel_stack_nth(regs,		\
+				(unsigned int)((unsigned long)offset));	\
 }
+DEFINE_BASIC_FETCH_FUNCS(stack)
 
-static __kprobes unsigned long fetch_memory(struct pt_regs *regs, void *addr)
-{
-	unsigned long retval;
-
-	if (probe_kernel_address(addr, retval))
-		return 0;
-	return retval;
+#define DEFINE_FETCH_retval(type)					\
+static __kprobes void FETCH_FUNC_NAME(retval, type)(struct pt_regs *regs,\
+					  void *dummy, void *dest)	\
+{									\
+	*(type *)dest = (type)regs_return_value(regs);			\
 }
-
-static __kprobes unsigned long fetch_retvalue(struct pt_regs *regs,
-					      void *dummy)
-{
-	return regs_return_value(regs);
-}
-
-static __kprobes unsigned long fetch_stack_address(struct pt_regs *regs,
-						   void *dummy)
-{
-	return kernel_stack_pointer(regs);
+DEFINE_BASIC_FETCH_FUNCS(retval)
+
+#define DEFINE_FETCH_memory(type)					\
+static __kprobes void FETCH_FUNC_NAME(memory, type)(struct pt_regs *regs,\
+					  void *addr, void *dest)	\
+{									\
+	type retval;							\
+	if (probe_kernel_address(addr, retval))				\
+		*(type *)dest = 0;					\
+	else								\
+		*(type *)dest = retval;					\
 }
+DEFINE_BASIC_FETCH_FUNCS(memory)
 
 /* Memory fetching by symbol */
 struct symbol_cache {
@@ -145,51 +192,126 @@ static struct symbol_cache *alloc_symbol_cache(const char *sym, long offset)
 	return sc;
 }
 
-static __kprobes unsigned long fetch_symbol(struct pt_regs *regs, void *data)
-{
-	struct symbol_cache *sc = data;
-
-	if (sc->addr)
-		return fetch_memory(regs, (void *)sc->addr);
-	else
-		return 0;
+#define DEFINE_FETCH_symbol(type)					\
+static __kprobes void FETCH_FUNC_NAME(symbol, type)(struct pt_regs *regs,\
+					  void *data, void *dest)	\
+{									\
+	struct symbol_cache *sc = data;					\
+	if (sc->addr)							\
+		fetch_memory_##type(regs, (void *)sc->addr, dest);	\
+	else								\
+		*(type *)dest = 0;					\
 }
+DEFINE_BASIC_FETCH_FUNCS(symbol)
 
-/* Special indirect memory access interface */
-struct indirect_fetch_data {
-	struct fetch_func orig;
+/* Dereference memory access function */
+struct deref_fetch_param {
+	struct fetch_param orig;
 	long offset;
 };
 
-static __kprobes unsigned long fetch_indirect(struct pt_regs *regs, void *data)
-{
-	struct indirect_fetch_data *ind = data;
-	unsigned long addr;
-
-	addr = call_fetch(&ind->orig, regs);
-	if (addr) {
-		addr += ind->offset;
-		return fetch_memory(regs, (void *)addr);
-	} else
-		return 0;
+#define DEFINE_FETCH_deref(type)					\
+static __kprobes void FETCH_FUNC_NAME(deref, type)(struct pt_regs *regs,\
+					    void *data, void *dest)	\
+{									\
+	struct deref_fetch_param *dprm = data;				\
+	unsigned long addr;						\
+	call_fetch(&dprm->orig, regs, &addr);				\
+	if (addr) {							\
+		addr += dprm->offset;					\
+		fetch_memory_##type(regs, (void *)addr, dest);		\
+	} else								\
+		*(type *)dest = 0;					\
 }
+DEFINE_BASIC_FETCH_FUNCS(deref)
 
-static __kprobes void free_indirect_fetch_data(struct indirect_fetch_data *data)
+static __kprobes void free_deref_fetch_param(struct deref_fetch_param *data)
 {
-	if (data->orig.func == fetch_indirect)
-		free_indirect_fetch_data(data->orig.data);
-	else if (data->orig.func == fetch_symbol)
+	if (CHECK_BASIC_FETCH_FUNCS(deref, data->orig.fn))
+		free_deref_fetch_param(data->orig.data);
+	else if (CHECK_BASIC_FETCH_FUNCS(symbol, data->orig.fn))
 		free_symbol_cache(data->orig.data);
 	kfree(data);
 }
 
+/* Default (unsigned long) fetch type */
+#define __DEFAULT_FETCH_TYPE(t) u##t
+#define _DEFAULT_FETCH_TYPE(t) __DEFAULT_FETCH_TYPE(t)
+#define DEFAULT_FETCH_TYPE _DEFAULT_FETCH_TYPE(BITS_PER_LONG)
+#define DEFAULT_FETCH_TYPE_STR __stringify(DEFAULT_FETCH_TYPE)
+
+#define ASSIGN_FETCH_FUNC(kind, type)	\
+	.kind = FETCH_FUNC_NAME(kind, type)
+
+#define ASSIGN_FETCH_TYPE(ptype, ftype, sign)	\
+	{.name = #ptype,			\
+	 .size = sizeof(ftype),			\
+	 .is_signed = sign,			\
+	 .print = PRINT_TYPE_FUNC_NAME(ptype),	\
+	 .fmt = PRINT_TYPE_FMT_NAME(ptype),	\
+ASSIGN_FETCH_FUNC(reg, ftype),			\
+ASSIGN_FETCH_FUNC(stack, ftype),		\
+ASSIGN_FETCH_FUNC(retval, ftype),		\
+ASSIGN_FETCH_FUNC(memory, ftype),		\
+ASSIGN_FETCH_FUNC(symbol, ftype),		\
+ASSIGN_FETCH_FUNC(deref, ftype),		\
+	}
+
+/* Fetch type information table */
+static const struct fetch_type {
+	const char	*name;		/* Name of type */
+	size_t		size;		/* Byte size of type */
+	int		is_signed;	/* Signed flag */
+	print_type_func_t	print;	/* Print functions */
+	const char	*fmt;		/* Fromat string */
+	/* Fetch functions */
+	fetch_func_t	reg;
+	fetch_func_t	stack;
+	fetch_func_t	retval;
+	fetch_func_t	memory;
+	fetch_func_t	symbol;
+	fetch_func_t	deref;
+} fetch_type_table[] = {
+	ASSIGN_FETCH_TYPE(u8,  u8,  0),
+	ASSIGN_FETCH_TYPE(u16, u16, 0),
+	ASSIGN_FETCH_TYPE(u32, u32, 0),
+	ASSIGN_FETCH_TYPE(u64, u64, 0),
+	ASSIGN_FETCH_TYPE(s8,  u8,  1),
+	ASSIGN_FETCH_TYPE(s16, u16, 1),
+	ASSIGN_FETCH_TYPE(s32, u32, 1),
+	ASSIGN_FETCH_TYPE(s64, u64, 1),
+};
+
+static const struct fetch_type *find_fetch_type(const char *type)
+{
+	int i;
+
+	if (!type)
+		type = DEFAULT_FETCH_TYPE_STR;
+
+	for (i = 0; i < ARRAY_SIZE(fetch_type_table); i++)
+		if (strcmp(type, fetch_type_table[i].name) == 0)
+			return &fetch_type_table[i];
+	return NULL;
+}
+
+/* Special function : only accept unsigned long */
+static __kprobes void fetch_stack_address(struct pt_regs *regs,
+					  void *dummy, void *dest)
+{
+	*(unsigned long *)dest = kernel_stack_pointer(regs);
+}
+
 /**
  * Kprobe event core functions
  */
 
 struct probe_arg {
-	struct fetch_func	fetch;
-	const char		*name;
+	struct fetch_param	fetch;
+	unsigned int		offset;	/* Offset from argument entry */
+	const char		*name;	/* Name of this argument */
+	const char		*comm;	/* Command of this argument */
+	const struct fetch_type	*type;	/* Type of this argument */
 };
 
 /* Flags for trace_probe */
@@ -204,6 +326,7 @@ struct trace_probe {
 	const char		*symbol;	/* symbol name */
 	struct ftrace_event_class	class;
 	struct ftrace_event_call	call;
+	ssize_t			size;		/* trace entry size */
 	unsigned int		nr_args;
 	struct probe_arg	args[];
 };
@@ -212,6 +335,7 @@ struct trace_probe {
 	(offsetof(struct trace_probe, args) +	\
 	(sizeof(struct probe_arg) * (n)))
 
+
 static __kprobes int probe_is_return(struct trace_probe *tp)
 {
 	return tp->rp.handler != NULL;
@@ -222,49 +346,6 @@ static __kprobes const char *probe_symbol(struct trace_probe *tp)
 	return tp->symbol ? tp->symbol : "unknown";
 }
 
-static int probe_arg_string(char *buf, size_t n, struct fetch_func *ff)
-{
-	int ret = -EINVAL;
-
-	if (ff->func == fetch_register) {
-		const char *name;
-		name = regs_query_register_name((unsigned int)((long)ff->data));
-		ret = snprintf(buf, n, "%%%s", name);
-	} else if (ff->func == fetch_stack)
-		ret = snprintf(buf, n, "$stack%lu", (unsigned long)ff->data);
-	else if (ff->func == fetch_memory)
-		ret = snprintf(buf, n, "@0x%p", ff->data);
-	else if (ff->func == fetch_symbol) {
-		struct symbol_cache *sc = ff->data;
-		if (sc->offset)
-			ret = snprintf(buf, n, "@%s%+ld", sc->symbol,
-					sc->offset);
-		else
-			ret = snprintf(buf, n, "@%s", sc->symbol);
-	} else if (ff->func == fetch_retvalue)
-		ret = snprintf(buf, n, "$retval");
-	else if (ff->func == fetch_stack_address)
-		ret = snprintf(buf, n, "$stack");
-	else if (ff->func == fetch_indirect) {
-		struct indirect_fetch_data *id = ff->data;
-		size_t l = 0;
-		ret = snprintf(buf, n, "%+ld(", id->offset);
-		if (ret >= n)
-			goto end;
-		l += ret;
-		ret = probe_arg_string(buf + l, n - l, &id->orig);
-		if (ret < 0)
-			goto end;
-		l += ret;
-		ret = snprintf(buf + l, n - l, ")");
-		ret += l;
-	}
-end:
-	if (ret >= n)
-		return -ENOSPC;
-	return ret;
-}
-
 static int register_probe_event(struct trace_probe *tp);
 static void unregister_probe_event(struct trace_probe *tp);
 
@@ -348,11 +429,12 @@ error:
 
 static void free_probe_arg(struct probe_arg *arg)
 {
-	if (arg->fetch.func == fetch_symbol)
+	if (CHECK_BASIC_FETCH_FUNCS(deref, arg->fetch.fn))
+		free_deref_fetch_param(arg->fetch.data);
+	else if (CHECK_BASIC_FETCH_FUNCS(symbol, arg->fetch.fn))
 		free_symbol_cache(arg->fetch.data);
-	else if (arg->fetch.func == fetch_indirect)
-		free_indirect_fetch_data(arg->fetch.data);
 	kfree(arg->name);
+	kfree(arg->comm);
 }
 
 static void free_trace_probe(struct trace_probe *tp)
@@ -458,28 +540,30 @@ static int split_symbol_offset(char *symbol, unsigned long *offset)
 #define PARAM_MAX_ARGS 16
 #define PARAM_MAX_STACK (THREAD_SIZE / sizeof(unsigned long))
 
-static int parse_probe_vars(char *arg, struct fetch_func *ff, int is_return)
+static int parse_probe_vars(char *arg, const struct fetch_type *t,
+			    struct fetch_param *f, int is_return)
 {
 	int ret = 0;
 	unsigned long param;
 
 	if (strcmp(arg, "retval") == 0) {
-		if (is_return) {
-			ff->func = fetch_retvalue;
-			ff->data = NULL;
-		} else
+		if (is_return)
+			f->fn = t->retval;
+		else
 			ret = -EINVAL;
 	} else if (strncmp(arg, "stack", 5) == 0) {
 		if (arg[5] == '\0') {
-			ff->func = fetch_stack_address;
-			ff->data = NULL;
+			if (strcmp(t->name, DEFAULT_FETCH_TYPE_STR) == 0)
+				f->fn = fetch_stack_address;
+			else
+				ret = -EINVAL;
 		} else if (isdigit(arg[5])) {
 			ret = strict_strtoul(arg + 5, 10, &param);
 			if (ret || param > PARAM_MAX_STACK)
 				ret = -EINVAL;
 			else {
-				ff->func = fetch_stack;
-				ff->data = (void *)param;
+				f->fn = t->stack;
+				f->data = (void *)param;
 			}
 		} else
 			ret = -EINVAL;
@@ -489,7 +573,8 @@ static int parse_probe_vars(char *arg, struct fetch_func *ff, int is_return)
 }
 
 /* Recursive argument parser */
-static int __parse_probe_arg(char *arg, struct fetch_func *ff, int is_return)
+static int __parse_probe_arg(char *arg, const struct fetch_type *t,
+			     struct fetch_param *f, int is_return)
 {
 	int ret = 0;
 	unsigned long param;
@@ -498,13 +583,13 @@ static int __parse_probe_arg(char *arg, struct fetch_func *ff, int is_return)
 
 	switch (arg[0]) {
 	case '$':
-		ret = parse_probe_vars(arg + 1, ff, is_return);
+		ret = parse_probe_vars(arg + 1, t, f, is_return);
 		break;
 	case '%':	/* named register */
 		ret = regs_query_register_offset(arg + 1);
 		if (ret >= 0) {
-			ff->func = fetch_register;
-			ff->data = (void *)(unsigned long)ret;
+			f->fn = t->reg;
+			f->data = (void *)(unsigned long)ret;
 			ret = 0;
 		}
 		break;
@@ -513,26 +598,22 @@ static int __parse_probe_arg(char *arg, struct fetch_func *ff, int is_return)
 			ret = strict_strtoul(arg + 1, 0, &param);
 			if (ret)
 				break;
-			ff->func = fetch_memory;
-			ff->data = (void *)param;
+			f->fn = t->memory;
+			f->data = (void *)param;
 		} else {
 			ret = split_symbol_offset(arg + 1, &offset);
 			if (ret)
 				break;
-			ff->data = alloc_symbol_cache(arg + 1, offset);
-			if (ff->data)
-				ff->func = fetch_symbol;
-			else
-				ret = -EINVAL;
+			f->data = alloc_symbol_cache(arg + 1, offset);
+			if (f->data)
+				f->fn = t->symbol;
 		}
 		break;
-	case '+':	/* indirect memory */
+	case '+':	/* deref memory */
 	case '-':
 		tmp = strchr(arg, '(');
-		if (!tmp) {
-			ret = -EINVAL;
+		if (!tmp)
 			break;
-		}
 		*tmp = '\0';
 		ret = strict_strtol(arg + 1, 0, &offset);
 		if (ret)
@@ -542,38 +623,58 @@ static int __parse_probe_arg(char *arg, struct fetch_func *ff, int is_return)
 		arg = tmp + 1;
 		tmp = strrchr(arg, ')');
 		if (tmp) {
-			struct indirect_fetch_data *id;
+			struct deref_fetch_param *dprm;
+			const struct fetch_type *t2 = find_fetch_type(NULL);
 			*tmp = '\0';
-			id = kzalloc(sizeof(struct indirect_fetch_data),
-				     GFP_KERNEL);
-			if (!id)
+			dprm = kzalloc(sizeof(struct deref_fetch_param),
+				       GFP_KERNEL);
+			if (!dprm)
 				return -ENOMEM;
-			id->offset = offset;
-			ret = __parse_probe_arg(arg, &id->orig, is_return);
+			dprm->offset = offset;
+			ret = __parse_probe_arg(arg, t2, &dprm->orig,
+						is_return);
 			if (ret)
-				kfree(id);
+				kfree(dprm);
 			else {
-				ff->func = fetch_indirect;
-				ff->data = (void *)id;
+				f->fn = t->deref;
+				f->data = (void *)dprm;
 			}
-		} else
-			ret = -EINVAL;
+		}
 		break;
-	default:
-		/* TODO: support custom handler */
-		ret = -EINVAL;
 	}
+	if (!ret && !f->fn)
+		ret = -EINVAL;
 	return ret;
 }
 
 /* String length checking wrapper */
-static int parse_probe_arg(char *arg, struct fetch_func *ff, int is_return)
+static int parse_probe_arg(char *arg, struct trace_probe *tp,
+			   struct probe_arg *parg, int is_return)
 {
+	const char *t;
+
 	if (strlen(arg) > MAX_ARGSTR_LEN) {
 		pr_info("Argument is too long.: %s\n",  arg);
 		return -ENOSPC;
 	}
-	return __parse_probe_arg(arg, ff, is_return);
+	parg->comm = kstrdup(arg, GFP_KERNEL);
+	if (!parg->comm) {
+		pr_info("Failed to allocate memory for command '%s'.\n", arg);
+		return -ENOMEM;
+	}
+	t = strchr(parg->comm, ':');
+	if (t) {
+		arg[t - parg->comm] = '\0';
+		t++;
+	}
+	parg->type = find_fetch_type(t);
+	if (!parg->type) {
+		pr_info("Unsupported type: %s\n", t);
+		return -EINVAL;
+	}
+	parg->offset = tp->size;
+	tp->size += parg->type->size;
+	return __parse_probe_arg(arg, parg->type, &parg->fetch, is_return);
 }
 
 /* Return 1 if name is reserved or already used by another argument */
@@ -603,15 +704,18 @@ static int create_trace_probe(int argc, char **argv)
 	 *  @ADDR	: fetch memory at ADDR (ADDR should be in kernel)
 	 *  @SYM[+|-offs] : fetch memory at SYM +|- offs (SYM is a data symbol)
 	 *  %REG	: fetch register REG
-	 * Indirect memory fetch:
+	 * Dereferencing memory fetch:
 	 *  +|-offs(ARG) : fetch memory at ARG +|- offs address.
 	 * Alias name of args:
 	 *  NAME=FETCHARG : set NAME as alias of FETCHARG.
+	 * Type of args:
+	 *  FETCHARG:TYPE : use TYPE instead of unsigned long.
 	 */
 	struct trace_probe *tp;
 	int i, ret = 0;
 	int is_return = 0, is_delete = 0;
-	char *symbol = NULL, *event = NULL, *arg = NULL, *group = NULL;
+	char *symbol = NULL, *event = NULL, *group = NULL;
+	char *arg, *tmp;
 	unsigned long offset = 0;
 	void *addr = NULL;
 	char buf[MAX_EVENT_NAME_LEN];
@@ -724,13 +828,6 @@ static int create_trace_probe(int argc, char **argv)
 		else
 			arg = argv[i];
 
-		if (conflict_field_name(argv[i], tp->args, i)) {
-			pr_info("Argument%d name '%s' conflicts with "
-				"another field.\n", i, argv[i]);
-			ret = -EINVAL;
-			goto error;
-		}
-
 		tp->args[i].name = kstrdup(argv[i], GFP_KERNEL);
 		if (!tp->args[i].name) {
 			pr_info("Failed to allocate argument%d name '%s'.\n",
@@ -738,9 +835,19 @@ static int create_trace_probe(int argc, char **argv)
 			ret = -ENOMEM;
 			goto error;
 		}
+		tmp = strchr(tp->args[i].name, ':');
+		if (tmp)
+			*tmp = '_';	/* convert : to _ */
+
+		if (conflict_field_name(tp->args[i].name, tp->args, i)) {
+			pr_info("Argument%d name '%s' conflicts with "
+				"another field.\n", i, argv[i]);
+			ret = -EINVAL;
+			goto error;
+		}
 
 		/* Parse fetch argument */
-		ret = parse_probe_arg(arg, &tp->args[i].fetch, is_return);
+		ret = parse_probe_arg(arg, tp, &tp->args[i], is_return);
 		if (ret) {
 			pr_info("Parse error at argument%d. (%d)\n", i, ret);
 			kfree(tp->args[i].name);
@@ -795,8 +902,7 @@ static void probes_seq_stop(struct seq_file *m, void *v)
 static int probes_seq_show(struct seq_file *m, void *v)
 {
 	struct trace_probe *tp = v;
-	int i, ret;
-	char buf[MAX_ARGSTR_LEN + 1];
+	int i;
 
 	seq_printf(m, "%c", probe_is_return(tp) ? 'r' : 'p');
 	seq_printf(m, ":%s/%s", tp->call.class->system, tp->call.name);
@@ -808,15 +914,10 @@ static int probes_seq_show(struct seq_file *m, void *v)
 	else
 		seq_printf(m, " %s", probe_symbol(tp));
 
-	for (i = 0; i < tp->nr_args; i++) {
-		ret = probe_arg_string(buf, MAX_ARGSTR_LEN, &tp->args[i].fetch);
-		if (ret < 0) {
-			pr_warning("Argument%d decoding error(%d).\n", i, ret);
-			return ret;
-		}
-		seq_printf(m, " %s=%s", tp->args[i].name, buf);
-	}
+	for (i = 0; i < tp->nr_args; i++)
+		seq_printf(m, " %s=%s", tp->args[i].name, tp->args[i].comm);
 	seq_printf(m, "\n");
+
 	return 0;
 }
 
@@ -946,9 +1047,10 @@ static const struct file_operations kprobe_profile_ops = {
 static __kprobes void kprobe_trace_func(struct kprobe *kp, struct pt_regs *regs)
 {
 	struct trace_probe *tp = container_of(kp, struct trace_probe, rp.kp);
-	struct kprobe_trace_entry *entry;
+	struct kprobe_trace_entry_head *entry;
 	struct ring_buffer_event *event;
 	struct ring_buffer *buffer;
+	u8 *data;
 	int size, i, pc;
 	unsigned long irq_flags;
 	struct ftrace_event_call *call = &tp->call;
@@ -958,7 +1060,7 @@ static __kprobes void kprobe_trace_func(struct kprobe *kp, struct pt_regs *regs)
 	local_save_flags(irq_flags);
 	pc = preempt_count();
 
-	size = SIZEOF_KPROBE_TRACE_ENTRY(tp->nr_args);
+	size = sizeof(*entry) + tp->size;
 
 	event = trace_current_buffer_lock_reserve(&buffer, call->event.type,
 						  size, irq_flags, pc);
@@ -966,10 +1068,10 @@ static __kprobes void kprobe_trace_func(struct kprobe *kp, struct pt_regs *regs)
 		return;
 
 	entry = ring_buffer_event_data(event);
-	entry->nargs = tp->nr_args;
 	entry->ip = (unsigned long)kp->addr;
+	data = (u8 *)&entry[1];
 	for (i = 0; i < tp->nr_args; i++)
-		entry->args[i] = call_fetch(&tp->args[i].fetch, regs);
+		call_fetch(&tp->args[i].fetch, regs, data + tp->args[i].offset);
 
 	if (!filter_current_check_discard(buffer, call, entry, event))
 		trace_nowake_buffer_unlock_commit(buffer, event, irq_flags, pc);
@@ -980,9 +1082,10 @@ static __kprobes void kretprobe_trace_func(struct kretprobe_instance *ri,
 					  struct pt_regs *regs)
 {
 	struct trace_probe *tp = container_of(ri->rp, struct trace_probe, rp);
-	struct kretprobe_trace_entry *entry;
+	struct kretprobe_trace_entry_head *entry;
 	struct ring_buffer_event *event;
 	struct ring_buffer *buffer;
+	u8 *data;
 	int size, i, pc;
 	unsigned long irq_flags;
 	struct ftrace_event_call *call = &tp->call;
@@ -990,7 +1093,7 @@ static __kprobes void kretprobe_trace_func(struct kretprobe_instance *ri,
 	local_save_flags(irq_flags);
 	pc = preempt_count();
 
-	size = SIZEOF_KRETPROBE_TRACE_ENTRY(tp->nr_args);
+	size = sizeof(*entry) + tp->size;
 
 	event = trace_current_buffer_lock_reserve(&buffer, call->event.type,
 						  size, irq_flags, pc);
@@ -998,11 +1101,11 @@ static __kprobes void kretprobe_trace_func(struct kretprobe_instance *ri,
 		return;
 
 	entry = ring_buffer_event_data(event);
-	entry->nargs = tp->nr_args;
 	entry->func = (unsigned long)tp->rp.kp.addr;
 	entry->ret_ip = (unsigned long)ri->ret_addr;
+	data = (u8 *)&entry[1];
 	for (i = 0; i < tp->nr_args; i++)
-		entry->args[i] = call_fetch(&tp->args[i].fetch, regs);
+		call_fetch(&tp->args[i].fetch, regs, data + tp->args[i].offset);
 
 	if (!filter_current_check_discard(buffer, call, entry, event))
 		trace_nowake_buffer_unlock_commit(buffer, event, irq_flags, pc);
@@ -1013,12 +1116,13 @@ enum print_line_t
 print_kprobe_event(struct trace_iterator *iter, int flags,
 		   struct trace_event *event)
 {
-	struct kprobe_trace_entry *field;
+	struct kprobe_trace_entry_head *field;
 	struct trace_seq *s = &iter->seq;
 	struct trace_probe *tp;
+	u8 *data;
 	int i;
 
-	field = (struct kprobe_trace_entry *)iter->ent;
+	field = (struct kprobe_trace_entry_head *)iter->ent;
 	tp = container_of(event, struct trace_probe, call.event);
 
 	if (!trace_seq_printf(s, "%s: (", tp->call.name))
@@ -1030,9 +1134,10 @@ print_kprobe_event(struct trace_iterator *iter, int flags,
 	if (!trace_seq_puts(s, ")"))
 		goto partial;
 
-	for (i = 0; i < field->nargs; i++)
-		if (!trace_seq_printf(s, " %s=%lx",
-				      tp->args[i].name, field->args[i]))
+	data = (u8 *)&field[1];
+	for (i = 0; i < tp->nr_args; i++)
+		if (!tp->args[i].type->print(s, tp->args[i].name,
+					     data + tp->args[i].offset))
 			goto partial;
 
 	if (!trace_seq_puts(s, "\n"))
@@ -1047,12 +1152,13 @@ enum print_line_t
 print_kretprobe_event(struct trace_iterator *iter, int flags,
 		      struct trace_event *event)
 {
-	struct kretprobe_trace_entry *field;
+	struct kretprobe_trace_entry_head *field;
 	struct trace_seq *s = &iter->seq;
 	struct trace_probe *tp;
+	u8 *data;
 	int i;
 
-	field = (struct kretprobe_trace_entry *)iter->ent;
+	field = (struct kretprobe_trace_entry_head *)iter->ent;
 	tp = container_of(event, struct trace_probe, call.event);
 
 	if (!trace_seq_printf(s, "%s: (", tp->call.name))
@@ -1070,9 +1176,10 @@ print_kretprobe_event(struct trace_iterator *iter, int flags,
 	if (!trace_seq_puts(s, ")"))
 		goto partial;
 
-	for (i = 0; i < field->nargs; i++)
-		if (!trace_seq_printf(s, " %s=%lx",
-				      tp->args[i].name, field->args[i]))
+	data = (u8 *)&field[1];
+	for (i = 0; i < tp->nr_args; i++)
+		if (!tp->args[i].type->print(s, tp->args[i].name,
+					     data + tp->args[i].offset))
 			goto partial;
 
 	if (!trace_seq_puts(s, "\n"))
@@ -1126,29 +1233,43 @@ static int probe_event_raw_init(struct ftrace_event_call *event_call)
 static int kprobe_event_define_fields(struct ftrace_event_call *event_call)
 {
 	int ret, i;
-	struct kprobe_trace_entry field;
+	struct kprobe_trace_entry_head field;
 	struct trace_probe *tp = (struct trace_probe *)event_call->data;
 
 	DEFINE_FIELD(unsigned long, ip, FIELD_STRING_IP, 0);
-	DEFINE_FIELD(int, nargs, FIELD_STRING_NARGS, 1);
 	/* Set argument names as fields */
-	for (i = 0; i < tp->nr_args; i++)
-		DEFINE_FIELD(unsigned long, args[i], tp->args[i].name, 0);
+	for (i = 0; i < tp->nr_args; i++) {
+		ret = trace_define_field(event_call, tp->args[i].type->name,
+					 tp->args[i].name,
+					 sizeof(field) + tp->args[i].offset,
+					 tp->args[i].type->size,
+					 tp->args[i].type->is_signed,
+					 FILTER_OTHER);
+		if (ret)
+			return ret;
+	}
 	return 0;
 }
 
 static int kretprobe_event_define_fields(struct ftrace_event_call *event_call)
 {
 	int ret, i;
-	struct kretprobe_trace_entry field;
+	struct kretprobe_trace_entry_head field;
 	struct trace_probe *tp = (struct trace_probe *)event_call->data;
 
 	DEFINE_FIELD(unsigned long, func, FIELD_STRING_FUNC, 0);
 	DEFINE_FIELD(unsigned long, ret_ip, FIELD_STRING_RETIP, 0);
-	DEFINE_FIELD(int, nargs, FIELD_STRING_NARGS, 1);
 	/* Set argument names as fields */
-	for (i = 0; i < tp->nr_args; i++)
-		DEFINE_FIELD(unsigned long, args[i], tp->args[i].name, 0);
+	for (i = 0; i < tp->nr_args; i++) {
+		ret = trace_define_field(event_call, tp->args[i].type->name,
+					 tp->args[i].name,
+					 sizeof(field) + tp->args[i].offset,
+					 tp->args[i].type->size,
+					 tp->args[i].type->is_signed,
+					 FILTER_OTHER);
+		if (ret)
+			return ret;
+	}
 	return 0;
 }
 
@@ -1173,8 +1294,8 @@ static int __set_print_fmt(struct trace_probe *tp, char *buf, int len)
 	pos += snprintf(buf + pos, LEN_OR_ZERO, "\"%s", fmt);
 
 	for (i = 0; i < tp->nr_args; i++) {
-		pos += snprintf(buf + pos, LEN_OR_ZERO, " %s=%%lx",
-				tp->args[i].name);
+		pos += snprintf(buf + pos, LEN_OR_ZERO, " %s=%s",
+				tp->args[i].name, tp->args[i].type->fmt);
 	}
 
 	pos += snprintf(buf + pos, LEN_OR_ZERO, "\", %s", arg);
@@ -1216,12 +1337,13 @@ static __kprobes void kprobe_perf_func(struct kprobe *kp,
 {
 	struct trace_probe *tp = container_of(kp, struct trace_probe, rp.kp);
 	struct ftrace_event_call *call = &tp->call;
-	struct kprobe_trace_entry *entry;
+	struct kprobe_trace_entry_head *entry;
+	u8 *data;
 	int size, __size, i;
 	unsigned long irq_flags;
 	int rctx;
 
-	__size = SIZEOF_KPROBE_TRACE_ENTRY(tp->nr_args);
+	__size = sizeof(*entry) + tp->size;
 	size = ALIGN(__size + sizeof(u32), sizeof(u64));
 	size -= sizeof(u32);
 	if (WARN_ONCE(size > PERF_MAX_TRACE_SIZE,
@@ -1233,10 +1355,10 @@ static __kprobes void kprobe_perf_func(struct kprobe *kp,
 	if (!entry)
 		return;
 
-	entry->nargs = tp->nr_args;
 	entry->ip = (unsigned long)kp->addr;
+	data = (u8 *)&entry[1];
 	for (i = 0; i < tp->nr_args; i++)
-		entry->args[i] = call_fetch(&tp->args[i].fetch, regs);
+		call_fetch(&tp->args[i].fetch, regs, data + tp->args[i].offset);
 
 	perf_trace_buf_submit(entry, size, rctx, entry->ip, 1, irq_flags, regs);
 }
@@ -1247,12 +1369,13 @@ static __kprobes void kretprobe_perf_func(struct kretprobe_instance *ri,
 {
 	struct trace_probe *tp = container_of(ri->rp, struct trace_probe, rp);
 	struct ftrace_event_call *call = &tp->call;
-	struct kretprobe_trace_entry *entry;
+	struct kretprobe_trace_entry_head *entry;
+	u8 *data;
 	int size, __size, i;
 	unsigned long irq_flags;
 	int rctx;
 
-	__size = SIZEOF_KRETPROBE_TRACE_ENTRY(tp->nr_args);
+	__size = sizeof(*entry) + tp->size;
 	size = ALIGN(__size + sizeof(u32), sizeof(u64));
 	size -= sizeof(u32);
 	if (WARN_ONCE(size > PERF_MAX_TRACE_SIZE,
@@ -1264,11 +1387,11 @@ static __kprobes void kretprobe_perf_func(struct kretprobe_instance *ri,
 	if (!entry)
 		return;
 
-	entry->nargs = tp->nr_args;
 	entry->func = (unsigned long)tp->rp.kp.addr;
 	entry->ret_ip = (unsigned long)ri->ret_addr;
+	data = (u8 *)&entry[1];
 	for (i = 0; i < tp->nr_args; i++)
-		entry->args[i] = call_fetch(&tp->args[i].fetch, regs);
+		call_fetch(&tp->args[i].fetch, regs, data + tp->args[i].offset);
 
 	perf_trace_buf_submit(entry, size, rctx, entry->ret_ip, 1,
 			       irq_flags, regs);
diff --git a/kernel/trace/trace_ksym.c b/kernel/trace/trace_ksym.c
index d59cd68..8eaf007 100644
--- a/kernel/trace/trace_ksym.c
+++ b/kernel/trace/trace_ksym.c
@@ -34,12 +34,6 @@
 
 #include <asm/atomic.h>
 
-/*
- * For now, let us restrict the no. of symbols traced simultaneously to number
- * of available hardware breakpoint registers.
- */
-#define KSYM_TRACER_MAX HBP_NUM
-
 #define KSYM_TRACER_OP_LEN 3 /* rw- */
 
 struct trace_ksym {
@@ -53,7 +47,6 @@ struct trace_ksym {
 
 static struct trace_array *ksym_trace_array;
 
-static unsigned int ksym_filter_entry_count;
 static unsigned int ksym_tracing_enabled;
 
 static HLIST_HEAD(ksym_filter_head);
@@ -181,13 +174,6 @@ int process_new_ksym_entry(char *ksymname, int op, unsigned long addr)
 	struct trace_ksym *entry;
 	int ret = -ENOMEM;
 
-	if (ksym_filter_entry_count >= KSYM_TRACER_MAX) {
-		printk(KERN_ERR "ksym_tracer: Maximum limit:(%d) reached. No"
-		" new requests for tracing can be accepted now.\n",
-			KSYM_TRACER_MAX);
-		return -ENOSPC;
-	}
-
 	entry = kzalloc(sizeof(struct trace_ksym), GFP_KERNEL);
 	if (!entry)
 		return -ENOMEM;
@@ -203,13 +189,17 @@ int process_new_ksym_entry(char *ksymname, int op, unsigned long addr)
 
 	if (IS_ERR(entry->ksym_hbp)) {
 		ret = PTR_ERR(entry->ksym_hbp);
-		printk(KERN_INFO "ksym_tracer request failed. Try again"
-					" later!!\n");
+		if (ret == -ENOSPC) {
+			printk(KERN_ERR "ksym_tracer: Maximum limit reached."
+			" No new requests for tracing can be accepted now.\n");
+		} else {
+			printk(KERN_INFO "ksym_tracer request failed. Try again"
+					 " later!!\n");
+		}
 		goto err;
 	}
 
 	hlist_add_head_rcu(&(entry->ksym_hlist), &ksym_filter_head);
-	ksym_filter_entry_count++;
 
 	return 0;
 
@@ -265,7 +255,6 @@ static void __ksym_trace_reset(void)
 	hlist_for_each_entry_safe(entry, node, node1, &ksym_filter_head,
 								ksym_hlist) {
 		unregister_wide_hw_breakpoint(entry->ksym_hbp);
-		ksym_filter_entry_count--;
 		hlist_del_rcu(&(entry->ksym_hlist));
 		synchronize_rcu();
 		kfree(entry);
@@ -338,7 +327,6 @@ static ssize_t ksym_trace_filter_write(struct file *file,
 				goto out_unlock;
 		}
 		/* Error or "symbol:---" case: drop it */
-		ksym_filter_entry_count--;
 		hlist_del_rcu(&(entry->ksym_hlist));
 		synchronize_rcu();
 		kfree(entry);
diff --git a/kernel/trace/trace_selftest.c b/kernel/trace/trace_selftest.c
index 6a9d36d..250e7f9 100644
--- a/kernel/trace/trace_selftest.c
+++ b/kernel/trace/trace_selftest.c
@@ -17,7 +17,6 @@ static inline int trace_valid_entry(struct trace_entry *entry)
 	case TRACE_BRANCH:
 	case TRACE_GRAPH_ENT:
 	case TRACE_GRAPH_RET:
-	case TRACE_HW_BRANCHES:
 	case TRACE_KSYM:
 		return 1;
 	}
@@ -756,62 +755,6 @@ trace_selftest_startup_branch(struct tracer *trace, struct trace_array *tr)
 }
 #endif /* CONFIG_BRANCH_TRACER */
 
-#ifdef CONFIG_HW_BRANCH_TRACER
-int
-trace_selftest_startup_hw_branches(struct tracer *trace,
-				   struct trace_array *tr)
-{
-	struct trace_iterator *iter;
-	struct tracer tracer;
-	unsigned long count;
-	int ret;
-
-	if (!trace->open) {
-		printk(KERN_CONT "missing open function...");
-		return -1;
-	}
-
-	ret = tracer_init(trace, tr);
-	if (ret) {
-		warn_failed_init_tracer(trace, ret);
-		return ret;
-	}
-
-	/*
-	 * The hw-branch tracer needs to collect the trace from the various
-	 * cpu trace buffers - before tracing is stopped.
-	 */
-	iter = kzalloc(sizeof(*iter), GFP_KERNEL);
-	if (!iter)
-		return -ENOMEM;
-
-	memcpy(&tracer, trace, sizeof(tracer));
-
-	iter->trace = &tracer;
-	iter->tr = tr;
-	iter->pos = -1;
-	mutex_init(&iter->mutex);
-
-	trace->open(iter);
-
-	mutex_destroy(&iter->mutex);
-	kfree(iter);
-
-	tracing_stop();
-
-	ret = trace_test_buffer(tr, &count);
-	trace->reset(tr);
-	tracing_start();
-
-	if (!ret && !count) {
-		printk(KERN_CONT "no entries found..");
-		ret = -1;
-	}
-
-	return ret;
-}
-#endif /* CONFIG_HW_BRANCH_TRACER */
-
 #ifdef CONFIG_KSYM_TRACER
 static int ksym_selftest_dummy;
 
diff --git a/kernel/workqueue.c b/kernel/workqueue.c
index dee4865..5bfb213 100644
--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
@@ -774,7 +774,7 @@ void flush_delayed_work(struct delayed_work *dwork)
 {
 	if (del_timer_sync(&dwork->timer)) {
 		struct cpu_workqueue_struct *cwq;
-		cwq = wq_per_cpu(keventd_wq, get_cpu());
+		cwq = wq_per_cpu(get_wq_data(&dwork->work)->wq, get_cpu());
 		__queue_work(cwq, &dwork->work);
 		put_cpu();
 	}