Merge branch 'perf/core' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip into trace/tip/tracing/core-7

Conflicts:
	include/linux/ftrace_event.h
	include/trace/ftrace.h
	kernel/trace/trace_event_perf.c
	kernel/trace/trace_kprobe.c
	kernel/trace/trace_syscalls.c

Signed-off-by: Steven Rostedt <rostedt@goodmis.org>

3 files changed, 111 insertions, 113 deletions

diff --git a/kernel/trace/trace_event_perf.c b/kernel/trace/trace_event_perf.c
index 0a47e8d6b491..26b8607a0abc 100644
--- a/kernel/trace/trace_event_perf.c
+++ b/kernel/trace/trace_event_perf.c
@@ -9,13 +9,9 @@
 #include <linux/kprobes.h>
 #include "trace.h"
 
-DEFINE_PER_CPU(struct pt_regs, perf_trace_regs);
-EXPORT_PER_CPU_SYMBOL_GPL(perf_trace_regs);
-
 EXPORT_SYMBOL_GPL(perf_arch_fetch_caller_regs);
 
-static char *perf_trace_buf;
-static char *perf_trace_buf_nmi;
+static char *perf_trace_buf[4];
 
 /*
  * Force it to be aligned to unsigned long to avoid misaligned accesses
@@ -27,63 +23,82 @@ typedef typeof(unsigned long [PERF_MAX_TRACE_SIZE / sizeof(unsigned long)])
 /* Count the events in use (per event id, not per instance) */
 static int      total_ref_count;
 
-static int perf_trace_event_enable(struct ftrace_event_call *event)
+static int perf_trace_event_init(struct ftrace_event_call *tp_event,
+                                 struct perf_event *p_event)
 {
-        char *buf;
+        struct hlist_head *list;
         int ret = -ENOMEM;
+        int cpu;
 
-        if (event->perf_refcount++ > 0)
+        p_event->tp_event = tp_event;
+        if (tp_event->perf_refcount++ > 0)
                 return 0;
 
-        if (!total_ref_count) {
-                buf = (char *)alloc_percpu(perf_trace_t);
-                if (!buf)
-                        goto fail_buf;
+        list = alloc_percpu(struct hlist_head);
+        if (!list)
+                goto fail;
+
+        for_each_possible_cpu(cpu)
+                INIT_HLIST_HEAD(per_cpu_ptr(list, cpu));
 
-                rcu_assign_pointer(perf_trace_buf, buf);
+        tp_event->perf_events = list;
+
+        if (!total_ref_count) {
+                char *buf;
+                int i;
 
-                buf = (char *)alloc_percpu(perf_trace_t);
-                if (!buf)
-                        goto fail_buf_nmi;
+                for (i = 0; i < 4; i++) {
+                        buf = (char *)alloc_percpu(perf_trace_t);
+                        if (!buf)
+                                goto fail;
 
-                rcu_assign_pointer(perf_trace_buf_nmi, buf);
+                        perf_trace_buf[i] = buf;
+                }
         }
 
-        if (event->class->reg)
-                ret = event->class->reg(event, TRACE_REG_PERF_REGISTER);
+        if (tp_event->class->reg)
+                ret = tp_event->class->reg(tp_event, TRACE_REG_PERF_REGISTER);
         else
-                ret = tracepoint_probe_register(event->name,
-                                                event->class->perf_probe,
-                                                event);
-        if (!ret) {
-                total_ref_count++;
-                return 0;
-        }
+                ret = tracepoint_probe_register(tp_event->name,
+                                                tp_event->class->perf_probe,
+                                                tp_event);
+
+        if (ret)
+                goto fail;
 
-fail_buf_nmi:
+        total_ref_count++;
+        return 0;
+
+fail:
         if (!total_ref_count) {
-                free_percpu(perf_trace_buf_nmi);
-                free_percpu(perf_trace_buf);
-                perf_trace_buf_nmi = NULL;
-                perf_trace_buf = NULL;
+                int i;
+
+                for (i = 0; i < 4; i++) {
+                        free_percpu(perf_trace_buf[i]);
+                        perf_trace_buf[i] = NULL;
+                }
+        }
+
+        if (!--tp_event->perf_refcount) {
+                free_percpu(tp_event->perf_events);
+                tp_event->perf_events = NULL;
         }
-fail_buf:
-        event->perf_refcount--;
 
         return ret;
 }
 
-int perf_trace_enable(int event_id)
+int perf_trace_init(struct perf_event *p_event)
 {
-        struct ftrace_event_call *event;
+        struct ftrace_event_call *tp_event;
+        int event_id = p_event->attr.config;
         int ret = -EINVAL;
 
         mutex_lock(&event_mutex);
-        list_for_each_entry(event, &ftrace_events, list) {
-                if (event->event.type == event_id &&
-                    event->class && event->class->perf_probe &&
-                    try_module_get(event->mod)) {
-                        ret = perf_trace_event_enable(event);
+        list_for_each_entry(tp_event, &ftrace_events, list) {
+                if (tp_event->event.type == event_id &&
+                    tp_event->class && tp_event->class->perf_probe &&
+                    try_module_get(tp_event->mod)) {
+                        ret = perf_trace_event_init(tp_event, p_event);
                         break;
                 }
         }
@@ -92,93 +107,76 @@ int perf_trace_enable(int event_id)
         return ret;
 }
 
-static void perf_trace_event_disable(struct ftrace_event_call *event)
+int perf_trace_enable(struct perf_event *p_event)
 {
-        char *buf, *nmi_buf;
-
-        if (--event->perf_refcount > 0)
-                return;
-
-        if (event->class->reg)
-                event->class->reg(event, TRACE_REG_PERF_UNREGISTER);
-        else
-                tracepoint_probe_unregister(event->name, event->class->perf_probe, event);
+        struct ftrace_event_call *tp_event = p_event->tp_event;
+        struct hlist_head *list;
 
-        if (!--total_ref_count) {
-                buf = perf_trace_buf;
-                rcu_assign_pointer(perf_trace_buf, NULL);
+        list = tp_event->perf_events;
+        if (WARN_ON_ONCE(!list))
+                return -EINVAL;
 
-                nmi_buf = perf_trace_buf_nmi;
-                rcu_assign_pointer(perf_trace_buf_nmi, NULL);
+        list = per_cpu_ptr(list, smp_processor_id());
+        hlist_add_head_rcu(&p_event->hlist_entry, list);
 
-                /*
-                 * Ensure every events in profiling have finished before
-                 * releasing the buffers
-                 */
-                synchronize_sched();
+        return 0;
+}
 
-                free_percpu(buf);
-                free_percpu(nmi_buf);
-        }
+void perf_trace_disable(struct perf_event *p_event)
+{
+        hlist_del_rcu(&p_event->hlist_entry);
 }
 
-void perf_trace_disable(int event_id)
+void perf_trace_destroy(struct perf_event *p_event)
 {
-        struct ftrace_event_call *event;
+        struct ftrace_event_call *tp_event = p_event->tp_event;
+        int i;
 
-        mutex_lock(&event_mutex);
-        list_for_each_entry(event, &ftrace_events, list) {
-                if (event->event.type == event_id) {
-                        perf_trace_event_disable(event);
-                        module_put(event->mod);
-                        break;
+        if (--tp_event->perf_refcount > 0)
+                return;
+
+        if (tp_event->class->reg)
+                tp_event->class->reg(tp_event, TRACE_REG_PERF_UNREGISTER);
+        else
+                tracepoint_probe_unregister(tp_event->name,
+                                            tp_event->class->perf_probe,
+                                            tp_event);
+
+        free_percpu(tp_event->perf_events);
+        tp_event->perf_events = NULL;
+
+        if (!--total_ref_count) {
+                for (i = 0; i < 4; i++) {
+                        free_percpu(perf_trace_buf[i]);
+                        perf_trace_buf[i] = NULL;
                 }
         }
-        mutex_unlock(&event_mutex);
 }
 
 __kprobes void *perf_trace_buf_prepare(int size, unsigned short type,
-                                       int *rctxp, unsigned long *irq_flags)
+                                       struct pt_regs *regs, int *rctxp)
 {
         struct trace_entry *entry;
-        char *trace_buf, *raw_data;
-        int pc, cpu;
+        char *raw_data;
+        int pc;
 
         BUILD_BUG_ON(PERF_MAX_TRACE_SIZE % sizeof(unsigned long));
 
         pc = preempt_count();
 
-        /* Protect the per cpu buffer, begin the rcu read side */
-        local_irq_save(*irq_flags);
-
         *rctxp = perf_swevent_get_recursion_context();
         if (*rctxp < 0)
-                goto err_recursion;
-
-        cpu = smp_processor_id();
-
-        if (in_nmi())
-                trace_buf = rcu_dereference_sched(perf_trace_buf_nmi);
-        else
-                trace_buf = rcu_dereference_sched(perf_trace_buf);
-
-        if (!trace_buf)
-                goto err;
+                return NULL;
 
-        raw_data = per_cpu_ptr(trace_buf, cpu);
+        raw_data = per_cpu_ptr(perf_trace_buf[*rctxp], smp_processor_id());
 
         /* zero the dead bytes from align to not leak stack to user */
         memset(&raw_data[size - sizeof(u64)], 0, sizeof(u64));
 
         entry = (struct trace_entry *)raw_data;
-        tracing_generic_entry_update(entry, *irq_flags, pc);
+        tracing_generic_entry_update(entry, regs->flags, pc);
         entry->type = type;
 
         return raw_data;
-err:
-        perf_swevent_put_recursion_context(*rctxp);
-err_recursion:
-        local_irq_restore(*irq_flags);
-        return NULL;
 }
 EXPORT_SYMBOL_GPL(perf_trace_buf_prepare);
diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c
index 9a082bba9537..faf7cefd15da 100644
--- a/kernel/trace/trace_kprobe.c
+++ b/kernel/trace/trace_kprobe.c
@@ -1338,9 +1338,9 @@ 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_head *entry;
+        struct hlist_head *head;
         u8 *data;
         int size, __size, i;
-        unsigned long irq_flags;
         int rctx;
 
         __size = sizeof(*entry) + tp->size;
@@ -1350,8 +1350,7 @@ static __kprobes void kprobe_perf_func(struct kprobe *kp,
                      "profile buffer not large enough"))
                 return;
 
-        entry = perf_trace_buf_prepare(size, call->event.type,
-                                       &rctx, &irq_flags);
+        entry = perf_trace_buf_prepare(size, call->event.type, regs, &rctx);
         if (!entry)
                 return;
 
@@ -1360,7 +1359,8 @@ static __kprobes void kprobe_perf_func(struct kprobe *kp,
         for (i = 0; i < tp->nr_args; i++)
                 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);
+        head = per_cpu_ptr(call->perf_events, smp_processor_id());
+        perf_trace_buf_submit(entry, size, rctx, entry->ip, 1, regs, head);
 }
 
 /* Kretprobe profile handler */
@@ -1370,9 +1370,9 @@ 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_head *entry;
+        struct hlist_head *head;
         u8 *data;
         int size, __size, i;
-        unsigned long irq_flags;
         int rctx;
 
         __size = sizeof(*entry) + tp->size;
@@ -1382,8 +1382,7 @@ static __kprobes void kretprobe_perf_func(struct kretprobe_instance *ri,
                      "profile buffer not large enough"))
                 return;
 
-        entry = perf_trace_buf_prepare(size, call->event.type,
-                                       &rctx, &irq_flags);
+        entry = perf_trace_buf_prepare(size, call->event.type, regs, &rctx);
         if (!entry)
                 return;
 
@@ -1393,8 +1392,8 @@ static __kprobes void kretprobe_perf_func(struct kretprobe_instance *ri,
         for (i = 0; i < tp->nr_args; i++)
                 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);
+        head = per_cpu_ptr(call->perf_events, smp_processor_id());
+        perf_trace_buf_submit(entry, size, rctx, entry->ret_ip, 1, regs, head);
 }
 
 static int probe_perf_enable(struct ftrace_event_call *call)
diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c
index 9d358301ae3e..d2c859cec9ea 100644
--- a/kernel/trace/trace_syscalls.c
+++ b/kernel/trace/trace_syscalls.c
@@ -488,7 +488,7 @@ static void perf_syscall_enter(void *ignore, struct pt_regs *regs, long id)
 {
         struct syscall_metadata *sys_data;
         struct syscall_trace_enter *rec;
-        unsigned long flags;
+        struct hlist_head *head;
         int syscall_nr;
         int rctx;
         int size;
@@ -511,15 +511,16 @@ static void perf_syscall_enter(void *ignore, struct pt_regs *regs, long id)
                 return;
 
         rec = (struct syscall_trace_enter *)perf_trace_buf_prepare(size,
-                                sys_data->enter_event->event.type,
-                                &rctx, &flags);
+                                sys_data->enter_event->event.type, regs, &rctx);
         if (!rec)
                 return;
 
         rec->nr = syscall_nr;
         syscall_get_arguments(current, regs, 0, sys_data->nb_args,
                                (unsigned long *)&rec->args);
-        perf_trace_buf_submit(rec, size, rctx, 0, 1, flags, regs);
+
+        head = per_cpu_ptr(sys_data->enter_event->perf_events, smp_processor_id());
+        perf_trace_buf_submit(rec, size, rctx, 0, 1, regs, head);
 }
 
 int perf_sysenter_enable(struct ftrace_event_call *call)
@@ -561,7 +562,7 @@ static void perf_syscall_exit(void *ignore, struct pt_regs *regs, long ret)
 {
         struct syscall_metadata *sys_data;
         struct syscall_trace_exit *rec;
-        unsigned long flags;
+        struct hlist_head *head;
         int syscall_nr;
         int rctx;
         int size;
@@ -587,15 +588,15 @@ static void perf_syscall_exit(void *ignore, struct pt_regs *regs, long ret)
                 return;
 
         rec = (struct syscall_trace_exit *)perf_trace_buf_prepare(size,
-                                sys_data->exit_event->event.type,
-                                &rctx, &flags);
+                                sys_data->exit_event->event.type, regs, &rctx);
         if (!rec)
                 return;
 
         rec->nr = syscall_nr;
         rec->ret = syscall_get_return_value(current, regs);
 
-        perf_trace_buf_submit(rec, size, rctx, 0, 1, flags, regs);
+        head = per_cpu_ptr(sys_data->exit_event->perf_events, smp_processor_id());
+        perf_trace_buf_submit(rec, size, rctx, 0, 1, regs, head);
 }
 
 int perf_sysexit_enable(struct ftrace_event_call *call)