* [PATCHv9 0/7] ftrace, perf: Adding support to use function trace
@ 2012-02-15 14:51 Jiri Olsa
2012-02-15 14:51 ` [PATCH 1/7] ftrace: Add enable/disable ftrace_ops control interface Jiri Olsa
` (6 more replies)
0 siblings, 7 replies; 17+ messages in thread
From: Jiri Olsa @ 2012-02-15 14:51 UTC (permalink / raw)
To: rostedt, fweisbec, mingo, paulus, acme, a.p.zijlstra
Cc: linux-kernel, aarapov
hi,
here's another version of perf support for function trace
with filter.
attached patches:
- 1/7 ftrace: Add enable/disable ftrace_ops control interface
- 2/7 ftrace, perf: Add open/close tracepoint perf registration actions
- 3/7 ftrace, perf: Add add/del tracepoint perf registration actions
- 4/7 ftrace: Add FTRACE_ENTRY_REG macro to allow event registration
- 5/7 ftrace, perf: Add support to use function tracepoint in perf
- 6/7 ftrace: Allow to specify filter field type for ftrace events
- 7/7 ftrace, perf: Add filter support for function trace event
v9 changes:
fixed patches 5/7 and 7/7 to work with disabled options:
CONFIG_FUNCTION_TRACER, CONFIG_DYNAMIC_FTRACE
thanks,
jirka
---
include/linux/ftrace.h | 77 +++++++++++++-
include/linux/ftrace_event.h | 9 ++-
include/linux/perf_event.h | 3 +
kernel/trace/ftrace.c | 132 ++++++++++++++++++++---
kernel/trace/trace.h | 38 +++++--
kernel/trace/trace_entries.h | 54 +++++++---
kernel/trace/trace_event_perf.c | 208 ++++++++++++++++++++++++++++-------
kernel/trace/trace_events.c | 12 ++-
kernel/trace/trace_events_filter.c | 168 ++++++++++++++++++++++++++++-
kernel/trace/trace_export.c | 64 ++++++++----
kernel/trace/trace_kprobe.c | 8 ++-
kernel/trace/trace_syscalls.c | 18 +++-
12 files changed, 671 insertions(+), 120 deletions(-)
^ permalink raw reply [flat|nested] 17+ messages in thread
* [PATCH 1/7] ftrace: Add enable/disable ftrace_ops control interface
2012-02-15 14:51 [PATCHv9 0/7] ftrace, perf: Adding support to use function trace Jiri Olsa
@ 2012-02-15 14:51 ` Jiri Olsa
2012-02-27 9:33 ` [tip:perf/core] ftrace: Add enable/ disable " tip-bot for Jiri Olsa
2012-02-15 14:51 ` [PATCH 2/7] ftrace, perf: Add open/close tracepoint perf registration actions Jiri Olsa
` (5 subsequent siblings)
6 siblings, 1 reply; 17+ messages in thread
From: Jiri Olsa @ 2012-02-15 14:51 UTC (permalink / raw)
To: rostedt, fweisbec, mingo, paulus, acme, a.p.zijlstra
Cc: linux-kernel, aarapov, Jiri Olsa
Adding a way to temporarily enable/disable ftrace_ops. The change
follows the same way as 'global' ftrace_ops are done.
Introducing 2 global ftrace_ops - control_ops and ftrace_control_list
which take over all ftrace_ops registered with FTRACE_OPS_FL_CONTROL
flag. In addition new per cpu flag called 'disabled' is also added to
ftrace_ops to provide the control information for each cpu.
When ftrace_ops with FTRACE_OPS_FL_CONTROL is registered, it is
set as disabled for all cpus.
The ftrace_control_list contains all the registered 'control' ftrace_ops.
The control_ops provides function which iterates ftrace_control_list
and does the check for 'disabled' flag on current cpu.
Adding 3 inline functions:
ftrace_function_local_disable/ftrace_function_local_enable
- enable/disable the ftrace_ops on current cpu
ftrace_function_local_disabled
- get disabled ftrace_ops::disabled value for current cpu
Signed-off-by: Jiri Olsa <jolsa@redhat.com>
Acked-by: Frederic Weisbecker <fweisbec@gmail.com>
---
include/linux/ftrace.h | 66 ++++++++++++++++++++++++++++
kernel/trace/ftrace.c | 111 +++++++++++++++++++++++++++++++++++++++++++++---
kernel/trace/trace.h | 2 +
3 files changed, 172 insertions(+), 7 deletions(-)
diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h
index f33fb3b..64a309d 100644
--- a/include/linux/ftrace.h
+++ b/include/linux/ftrace.h
@@ -31,16 +31,33 @@ ftrace_enable_sysctl(struct ctl_table *table, int write,
typedef void (*ftrace_func_t)(unsigned long ip, unsigned long parent_ip);
+/*
+ * FTRACE_OPS_FL_* bits denote the state of ftrace_ops struct and are
+ * set in the flags member.
+ *
+ * ENABLED - set/unset when ftrace_ops is registered/unregistered
+ * GLOBAL - set manualy by ftrace_ops user to denote the ftrace_ops
+ * is part of the global tracers sharing the same filter
+ * via set_ftrace_* debugfs files.
+ * DYNAMIC - set when ftrace_ops is registered to denote dynamically
+ * allocated ftrace_ops which need special care
+ * CONTROL - set manualy by ftrace_ops user to denote the ftrace_ops
+ * could be controled by following calls:
+ * ftrace_function_local_enable
+ * ftrace_function_local_disable
+ */
enum {
FTRACE_OPS_FL_ENABLED = 1 << 0,
FTRACE_OPS_FL_GLOBAL = 1 << 1,
FTRACE_OPS_FL_DYNAMIC = 1 << 2,
+ FTRACE_OPS_FL_CONTROL = 1 << 3,
};
struct ftrace_ops {
ftrace_func_t func;
struct ftrace_ops *next;
unsigned long flags;
+ int __percpu *disabled;
#ifdef CONFIG_DYNAMIC_FTRACE
struct ftrace_hash *notrace_hash;
struct ftrace_hash *filter_hash;
@@ -97,6 +114,55 @@ int register_ftrace_function(struct ftrace_ops *ops);
int unregister_ftrace_function(struct ftrace_ops *ops);
void clear_ftrace_function(void);
+/**
+ * ftrace_function_local_enable - enable controlled ftrace_ops on current cpu
+ *
+ * This function enables tracing on current cpu by decreasing
+ * the per cpu control variable.
+ * It must be called with preemption disabled and only on ftrace_ops
+ * registered with FTRACE_OPS_FL_CONTROL. If called without preemption
+ * disabled, this_cpu_ptr will complain when CONFIG_DEBUG_PREEMPT is enabled.
+ */
+static inline void ftrace_function_local_enable(struct ftrace_ops *ops)
+{
+ if (WARN_ON_ONCE(!(ops->flags & FTRACE_OPS_FL_CONTROL)))
+ return;
+
+ (*this_cpu_ptr(ops->disabled))--;
+}
+
+/**
+ * ftrace_function_local_disable - enable controlled ftrace_ops on current cpu
+ *
+ * This function enables tracing on current cpu by decreasing
+ * the per cpu control variable.
+ * It must be called with preemption disabled and only on ftrace_ops
+ * registered with FTRACE_OPS_FL_CONTROL. If called without preemption
+ * disabled, this_cpu_ptr will complain when CONFIG_DEBUG_PREEMPT is enabled.
+ */
+static inline void ftrace_function_local_disable(struct ftrace_ops *ops)
+{
+ if (WARN_ON_ONCE(!(ops->flags & FTRACE_OPS_FL_CONTROL)))
+ return;
+
+ (*this_cpu_ptr(ops->disabled))++;
+}
+
+/**
+ * ftrace_function_local_disabled - returns ftrace_ops disabled value
+ * on current cpu
+ *
+ * This function returns value of ftrace_ops::disabled on current cpu.
+ * It must be called with preemption disabled and only on ftrace_ops
+ * registered with FTRACE_OPS_FL_CONTROL. If called without preemption
+ * disabled, this_cpu_ptr will complain when CONFIG_DEBUG_PREEMPT is enabled.
+ */
+static inline int ftrace_function_local_disabled(struct ftrace_ops *ops)
+{
+ WARN_ON_ONCE(!(ops->flags & FTRACE_OPS_FL_CONTROL));
+ return *this_cpu_ptr(ops->disabled);
+}
+
extern void ftrace_stub(unsigned long a0, unsigned long a1);
#else /* !CONFIG_FUNCTION_TRACER */
diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
index e2e0597..c8d2af2 100644
--- a/kernel/trace/ftrace.c
+++ b/kernel/trace/ftrace.c
@@ -62,6 +62,8 @@
#define FTRACE_HASH_DEFAULT_BITS 10
#define FTRACE_HASH_MAX_BITS 12
+#define FL_GLOBAL_CONTROL_MASK (FTRACE_OPS_FL_GLOBAL | FTRACE_OPS_FL_CONTROL)
+
/* ftrace_enabled is a method to turn ftrace on or off */
int ftrace_enabled __read_mostly;
static int last_ftrace_enabled;
@@ -89,12 +91,14 @@ static struct ftrace_ops ftrace_list_end __read_mostly = {
};
static struct ftrace_ops *ftrace_global_list __read_mostly = &ftrace_list_end;
+static struct ftrace_ops *ftrace_control_list __read_mostly = &ftrace_list_end;
static struct ftrace_ops *ftrace_ops_list __read_mostly = &ftrace_list_end;
ftrace_func_t ftrace_trace_function __read_mostly = ftrace_stub;
static ftrace_func_t __ftrace_trace_function_delay __read_mostly = ftrace_stub;
ftrace_func_t __ftrace_trace_function __read_mostly = ftrace_stub;
ftrace_func_t ftrace_pid_function __read_mostly = ftrace_stub;
static struct ftrace_ops global_ops;
+static struct ftrace_ops control_ops;
static void
ftrace_ops_list_func(unsigned long ip, unsigned long parent_ip);
@@ -168,6 +172,32 @@ static void ftrace_test_stop_func(unsigned long ip, unsigned long parent_ip)
}
#endif
+static void control_ops_disable_all(struct ftrace_ops *ops)
+{
+ int cpu;
+
+ for_each_possible_cpu(cpu)
+ *per_cpu_ptr(ops->disabled, cpu) = 1;
+}
+
+static int control_ops_alloc(struct ftrace_ops *ops)
+{
+ int __percpu *disabled;
+
+ disabled = alloc_percpu(int);
+ if (!disabled)
+ return -ENOMEM;
+
+ ops->disabled = disabled;
+ control_ops_disable_all(ops);
+ return 0;
+}
+
+static void control_ops_free(struct ftrace_ops *ops)
+{
+ free_percpu(ops->disabled);
+}
+
static void update_global_ops(void)
{
ftrace_func_t func;
@@ -259,6 +289,26 @@ static int remove_ftrace_ops(struct ftrace_ops **list, struct ftrace_ops *ops)
return 0;
}
+static void add_ftrace_list_ops(struct ftrace_ops **list,
+ struct ftrace_ops *main_ops,
+ struct ftrace_ops *ops)
+{
+ int first = *list == &ftrace_list_end;
+ add_ftrace_ops(list, ops);
+ if (first)
+ add_ftrace_ops(&ftrace_ops_list, main_ops);
+}
+
+static int remove_ftrace_list_ops(struct ftrace_ops **list,
+ struct ftrace_ops *main_ops,
+ struct ftrace_ops *ops)
+{
+ int ret = remove_ftrace_ops(list, ops);
+ if (!ret && *list == &ftrace_list_end)
+ ret = remove_ftrace_ops(&ftrace_ops_list, main_ops);
+ return ret;
+}
+
static int __register_ftrace_function(struct ftrace_ops *ops)
{
if (ftrace_disabled)
@@ -270,15 +320,20 @@ static int __register_ftrace_function(struct ftrace_ops *ops)
if (WARN_ON(ops->flags & FTRACE_OPS_FL_ENABLED))
return -EBUSY;
+ /* We don't support both control and global flags set. */
+ if ((ops->flags & FL_GLOBAL_CONTROL_MASK) == FL_GLOBAL_CONTROL_MASK)
+ return -EINVAL;
+
if (!core_kernel_data((unsigned long)ops))
ops->flags |= FTRACE_OPS_FL_DYNAMIC;
if (ops->flags & FTRACE_OPS_FL_GLOBAL) {
- int first = ftrace_global_list == &ftrace_list_end;
- add_ftrace_ops(&ftrace_global_list, ops);
+ add_ftrace_list_ops(&ftrace_global_list, &global_ops, ops);
ops->flags |= FTRACE_OPS_FL_ENABLED;
- if (first)
- add_ftrace_ops(&ftrace_ops_list, &global_ops);
+ } else if (ops->flags & FTRACE_OPS_FL_CONTROL) {
+ if (control_ops_alloc(ops))
+ return -ENOMEM;
+ add_ftrace_list_ops(&ftrace_control_list, &control_ops, ops);
} else
add_ftrace_ops(&ftrace_ops_list, ops);
@@ -302,11 +357,23 @@ static int __unregister_ftrace_function(struct ftrace_ops *ops)
return -EINVAL;
if (ops->flags & FTRACE_OPS_FL_GLOBAL) {
- ret = remove_ftrace_ops(&ftrace_global_list, ops);
- if (!ret && ftrace_global_list == &ftrace_list_end)
- ret = remove_ftrace_ops(&ftrace_ops_list, &global_ops);
+ ret = remove_ftrace_list_ops(&ftrace_global_list,
+ &global_ops, ops);
if (!ret)
ops->flags &= ~FTRACE_OPS_FL_ENABLED;
+ } else if (ops->flags & FTRACE_OPS_FL_CONTROL) {
+ ret = remove_ftrace_list_ops(&ftrace_control_list,
+ &control_ops, ops);
+ if (!ret) {
+ /*
+ * The ftrace_ops is now removed from the list,
+ * so there'll be no new users. We must ensure
+ * all current users are done before we free
+ * the control data.
+ */
+ synchronize_sched();
+ control_ops_free(ops);
+ }
} else
ret = remove_ftrace_ops(&ftrace_ops_list, ops);
@@ -3874,6 +3941,36 @@ ftrace_ops_test(struct ftrace_ops *ops, unsigned long ip)
#endif /* CONFIG_DYNAMIC_FTRACE */
static void
+ftrace_ops_control_func(unsigned long ip, unsigned long parent_ip)
+{
+ struct ftrace_ops *op;
+
+ if (unlikely(trace_recursion_test(TRACE_CONTROL_BIT)))
+ return;
+
+ /*
+ * Some of the ops may be dynamically allocated,
+ * they must be freed after a synchronize_sched().
+ */
+ preempt_disable_notrace();
+ trace_recursion_set(TRACE_CONTROL_BIT);
+ op = rcu_dereference_raw(ftrace_control_list);
+ while (op != &ftrace_list_end) {
+ if (!ftrace_function_local_disabled(op) &&
+ ftrace_ops_test(op, ip))
+ op->func(ip, parent_ip);
+
+ op = rcu_dereference_raw(op->next);
+ };
+ trace_recursion_clear(TRACE_CONTROL_BIT);
+ preempt_enable_notrace();
+}
+
+static struct ftrace_ops control_ops = {
+ .func = ftrace_ops_control_func,
+};
+
+static void
ftrace_ops_list_func(unsigned long ip, unsigned long parent_ip)
{
struct ftrace_ops *op;
diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h
index b93ecba..55c6ea0 100644
--- a/kernel/trace/trace.h
+++ b/kernel/trace/trace.h
@@ -288,6 +288,8 @@ struct tracer {
/* for function tracing recursion */
#define TRACE_INTERNAL_BIT (1<<11)
#define TRACE_GLOBAL_BIT (1<<12)
+#define TRACE_CONTROL_BIT (1<<13)
+
/*
* Abuse of the trace_recursion.
* As we need a way to maintain state if we are tracing the function
--
1.7.1
^ permalink raw reply related [flat|nested] 17+ messages in thread
* [PATCH 2/7] ftrace, perf: Add open/close tracepoint perf registration actions
2012-02-15 14:51 [PATCHv9 0/7] ftrace, perf: Adding support to use function trace Jiri Olsa
2012-02-15 14:51 ` [PATCH 1/7] ftrace: Add enable/disable ftrace_ops control interface Jiri Olsa
@ 2012-02-15 14:51 ` Jiri Olsa
2012-02-27 9:34 ` [tip:perf/core] ftrace, perf: Add open/ close " tip-bot for Jiri Olsa
2012-02-15 14:51 ` [PATCH 3/7] ftrace, perf: Add add/del " Jiri Olsa
` (4 subsequent siblings)
6 siblings, 1 reply; 17+ messages in thread
From: Jiri Olsa @ 2012-02-15 14:51 UTC (permalink / raw)
To: rostedt, fweisbec, mingo, paulus, acme, a.p.zijlstra
Cc: linux-kernel, aarapov, Jiri Olsa
Adding TRACE_REG_PERF_OPEN and TRACE_REG_PERF_CLOSE to differentiate
register/unregister from open/close actions.
The register/unregister actions are invoked for the first/last
tracepoint user when opening/closing the event.
The open/close actions are invoked for each tracepoint user when
opening/closing the event.
Signed-off-by: Jiri Olsa <jolsa@redhat.com>
Acked-by: Frederic Weisbecker <fweisbec@gmail.com>
---
include/linux/ftrace_event.h | 6 +-
kernel/trace/trace_event_perf.c | 116 +++++++++++++++++++++++++--------------
kernel/trace/trace_events.c | 10 ++-
kernel/trace/trace_kprobe.c | 6 ++-
kernel/trace/trace_syscalls.c | 14 +++-
5 files changed, 101 insertions(+), 51 deletions(-)
diff --git a/include/linux/ftrace_event.h b/include/linux/ftrace_event.h
index c3da42d..195e360 100644
--- a/include/linux/ftrace_event.h
+++ b/include/linux/ftrace_event.h
@@ -146,6 +146,8 @@ enum trace_reg {
TRACE_REG_UNREGISTER,
TRACE_REG_PERF_REGISTER,
TRACE_REG_PERF_UNREGISTER,
+ TRACE_REG_PERF_OPEN,
+ TRACE_REG_PERF_CLOSE,
};
struct ftrace_event_call;
@@ -157,7 +159,7 @@ struct ftrace_event_class {
void *perf_probe;
#endif
int (*reg)(struct ftrace_event_call *event,
- enum trace_reg type);
+ enum trace_reg type, void *data);
int (*define_fields)(struct ftrace_event_call *);
struct list_head *(*get_fields)(struct ftrace_event_call *);
struct list_head fields;
@@ -165,7 +167,7 @@ struct ftrace_event_class {
};
extern int ftrace_event_reg(struct ftrace_event_call *event,
- enum trace_reg type);
+ enum trace_reg type, void *data);
enum {
TRACE_EVENT_FL_ENABLED_BIT,
diff --git a/kernel/trace/trace_event_perf.c b/kernel/trace/trace_event_perf.c
index 19a359d..0cfcc37 100644
--- a/kernel/trace/trace_event_perf.c
+++ b/kernel/trace/trace_event_perf.c
@@ -44,23 +44,17 @@ static int perf_trace_event_perm(struct ftrace_event_call *tp_event,
return 0;
}
-static int perf_trace_event_init(struct ftrace_event_call *tp_event,
- struct perf_event *p_event)
+static int perf_trace_event_reg(struct ftrace_event_call *tp_event,
+ struct perf_event *p_event)
{
struct hlist_head __percpu *list;
- int ret;
+ int ret = -ENOMEM;
int cpu;
- ret = perf_trace_event_perm(tp_event, p_event);
- if (ret)
- return ret;
-
p_event->tp_event = tp_event;
if (tp_event->perf_refcount++ > 0)
return 0;
- ret = -ENOMEM;
-
list = alloc_percpu(struct hlist_head);
if (!list)
goto fail;
@@ -83,7 +77,7 @@ static int perf_trace_event_init(struct ftrace_event_call *tp_event,
}
}
- ret = tp_event->class->reg(tp_event, TRACE_REG_PERF_REGISTER);
+ ret = tp_event->class->reg(tp_event, TRACE_REG_PERF_REGISTER, NULL);
if (ret)
goto fail;
@@ -108,6 +102,69 @@ fail:
return ret;
}
+static void perf_trace_event_unreg(struct perf_event *p_event)
+{
+ struct ftrace_event_call *tp_event = p_event->tp_event;
+ int i;
+
+ if (--tp_event->perf_refcount > 0)
+ goto out;
+
+ tp_event->class->reg(tp_event, TRACE_REG_PERF_UNREGISTER, NULL);
+
+ /*
+ * Ensure our callback won't be called anymore. The buffers
+ * will be freed after that.
+ */
+ tracepoint_synchronize_unregister();
+
+ free_percpu(tp_event->perf_events);
+ tp_event->perf_events = NULL;
+
+ if (!--total_ref_count) {
+ for (i = 0; i < PERF_NR_CONTEXTS; i++) {
+ free_percpu(perf_trace_buf[i]);
+ perf_trace_buf[i] = NULL;
+ }
+ }
+out:
+ module_put(tp_event->mod);
+}
+
+static int perf_trace_event_open(struct perf_event *p_event)
+{
+ struct ftrace_event_call *tp_event = p_event->tp_event;
+ return tp_event->class->reg(tp_event, TRACE_REG_PERF_OPEN, p_event);
+}
+
+static void perf_trace_event_close(struct perf_event *p_event)
+{
+ struct ftrace_event_call *tp_event = p_event->tp_event;
+ tp_event->class->reg(tp_event, TRACE_REG_PERF_CLOSE, p_event);
+}
+
+static int perf_trace_event_init(struct ftrace_event_call *tp_event,
+ struct perf_event *p_event)
+{
+ int ret;
+
+ ret = perf_trace_event_perm(tp_event, p_event);
+ if (ret)
+ return ret;
+
+ ret = perf_trace_event_reg(tp_event, p_event);
+ if (ret)
+ return ret;
+
+ ret = perf_trace_event_open(p_event);
+ if (ret) {
+ perf_trace_event_unreg(p_event);
+ return ret;
+ }
+
+ return 0;
+}
+
int perf_trace_init(struct perf_event *p_event)
{
struct ftrace_event_call *tp_event;
@@ -130,6 +187,14 @@ int perf_trace_init(struct perf_event *p_event)
return ret;
}
+void perf_trace_destroy(struct perf_event *p_event)
+{
+ mutex_lock(&event_mutex);
+ perf_trace_event_close(p_event);
+ perf_trace_event_unreg(p_event);
+ mutex_unlock(&event_mutex);
+}
+
int perf_trace_add(struct perf_event *p_event, int flags)
{
struct ftrace_event_call *tp_event = p_event->tp_event;
@@ -154,37 +219,6 @@ void perf_trace_del(struct perf_event *p_event, int flags)
hlist_del_rcu(&p_event->hlist_entry);
}
-void perf_trace_destroy(struct perf_event *p_event)
-{
- struct ftrace_event_call *tp_event = p_event->tp_event;
- int i;
-
- mutex_lock(&event_mutex);
- if (--tp_event->perf_refcount > 0)
- goto out;
-
- tp_event->class->reg(tp_event, TRACE_REG_PERF_UNREGISTER);
-
- /*
- * Ensure our callback won't be called anymore. The buffers
- * will be freed after that.
- */
- tracepoint_synchronize_unregister();
-
- free_percpu(tp_event->perf_events);
- tp_event->perf_events = NULL;
-
- if (!--total_ref_count) {
- for (i = 0; i < PERF_NR_CONTEXTS; i++) {
- free_percpu(perf_trace_buf[i]);
- perf_trace_buf[i] = NULL;
- }
- }
-out:
- module_put(tp_event->mod);
- mutex_unlock(&event_mutex);
-}
-
__kprobes void *perf_trace_buf_prepare(int size, unsigned short type,
struct pt_regs *regs, int *rctxp)
{
diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c
index c212a7f..5138fea 100644
--- a/kernel/trace/trace_events.c
+++ b/kernel/trace/trace_events.c
@@ -147,7 +147,8 @@ int trace_event_raw_init(struct ftrace_event_call *call)
}
EXPORT_SYMBOL_GPL(trace_event_raw_init);
-int ftrace_event_reg(struct ftrace_event_call *call, enum trace_reg type)
+int ftrace_event_reg(struct ftrace_event_call *call,
+ enum trace_reg type, void *data)
{
switch (type) {
case TRACE_REG_REGISTER:
@@ -170,6 +171,9 @@ int ftrace_event_reg(struct ftrace_event_call *call, enum trace_reg type)
call->class->perf_probe,
call);
return 0;
+ case TRACE_REG_PERF_OPEN:
+ case TRACE_REG_PERF_CLOSE:
+ return 0;
#endif
}
return 0;
@@ -209,7 +213,7 @@ static int ftrace_event_enable_disable(struct ftrace_event_call *call,
tracing_stop_cmdline_record();
call->flags &= ~TRACE_EVENT_FL_RECORDED_CMD;
}
- call->class->reg(call, TRACE_REG_UNREGISTER);
+ call->class->reg(call, TRACE_REG_UNREGISTER, NULL);
}
break;
case 1:
@@ -218,7 +222,7 @@ static int ftrace_event_enable_disable(struct ftrace_event_call *call,
tracing_start_cmdline_record();
call->flags |= TRACE_EVENT_FL_RECORDED_CMD;
}
- ret = call->class->reg(call, TRACE_REG_REGISTER);
+ ret = call->class->reg(call, TRACE_REG_REGISTER, NULL);
if (ret) {
tracing_stop_cmdline_record();
pr_info("event trace: Could not enable event "
diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c
index 00d527c..5667f89 100644
--- a/kernel/trace/trace_kprobe.c
+++ b/kernel/trace/trace_kprobe.c
@@ -1892,7 +1892,8 @@ static __kprobes void kretprobe_perf_func(struct kretprobe_instance *ri,
#endif /* CONFIG_PERF_EVENTS */
static __kprobes
-int kprobe_register(struct ftrace_event_call *event, enum trace_reg type)
+int kprobe_register(struct ftrace_event_call *event,
+ enum trace_reg type, void *data)
{
struct trace_probe *tp = (struct trace_probe *)event->data;
@@ -1909,6 +1910,9 @@ int kprobe_register(struct ftrace_event_call *event, enum trace_reg type)
case TRACE_REG_PERF_UNREGISTER:
disable_trace_probe(tp, TP_FLAG_PROFILE);
return 0;
+ case TRACE_REG_PERF_OPEN:
+ case TRACE_REG_PERF_CLOSE:
+ return 0;
#endif
}
return 0;
diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c
index cb65454..6916b0d 100644
--- a/kernel/trace/trace_syscalls.c
+++ b/kernel/trace/trace_syscalls.c
@@ -17,9 +17,9 @@ static DECLARE_BITMAP(enabled_enter_syscalls, NR_syscalls);
static DECLARE_BITMAP(enabled_exit_syscalls, NR_syscalls);
static int syscall_enter_register(struct ftrace_event_call *event,
- enum trace_reg type);
+ enum trace_reg type, void *data);
static int syscall_exit_register(struct ftrace_event_call *event,
- enum trace_reg type);
+ enum trace_reg type, void *data);
static int syscall_enter_define_fields(struct ftrace_event_call *call);
static int syscall_exit_define_fields(struct ftrace_event_call *call);
@@ -649,7 +649,7 @@ void perf_sysexit_disable(struct ftrace_event_call *call)
#endif /* CONFIG_PERF_EVENTS */
static int syscall_enter_register(struct ftrace_event_call *event,
- enum trace_reg type)
+ enum trace_reg type, void *data)
{
switch (type) {
case TRACE_REG_REGISTER:
@@ -664,13 +664,16 @@ static int syscall_enter_register(struct ftrace_event_call *event,
case TRACE_REG_PERF_UNREGISTER:
perf_sysenter_disable(event);
return 0;
+ case TRACE_REG_PERF_OPEN:
+ case TRACE_REG_PERF_CLOSE:
+ return 0;
#endif
}
return 0;
}
static int syscall_exit_register(struct ftrace_event_call *event,
- enum trace_reg type)
+ enum trace_reg type, void *data)
{
switch (type) {
case TRACE_REG_REGISTER:
@@ -685,6 +688,9 @@ static int syscall_exit_register(struct ftrace_event_call *event,
case TRACE_REG_PERF_UNREGISTER:
perf_sysexit_disable(event);
return 0;
+ case TRACE_REG_PERF_OPEN:
+ case TRACE_REG_PERF_CLOSE:
+ return 0;
#endif
}
return 0;
--
1.7.1
^ permalink raw reply related [flat|nested] 17+ messages in thread
* [PATCH 3/7] ftrace, perf: Add add/del tracepoint perf registration actions
2012-02-15 14:51 [PATCHv9 0/7] ftrace, perf: Adding support to use function trace Jiri Olsa
2012-02-15 14:51 ` [PATCH 1/7] ftrace: Add enable/disable ftrace_ops control interface Jiri Olsa
2012-02-15 14:51 ` [PATCH 2/7] ftrace, perf: Add open/close tracepoint perf registration actions Jiri Olsa
@ 2012-02-15 14:51 ` Jiri Olsa
2012-02-27 9:34 ` [tip:perf/core] ftrace, perf: Add add/ del " tip-bot for Jiri Olsa
2012-02-15 14:51 ` [PATCH 4/7] ftrace: Add FTRACE_ENTRY_REG macro to allow event registration Jiri Olsa
` (3 subsequent siblings)
6 siblings, 1 reply; 17+ messages in thread
From: Jiri Olsa @ 2012-02-15 14:51 UTC (permalink / raw)
To: rostedt, fweisbec, mingo, paulus, acme, a.p.zijlstra
Cc: linux-kernel, aarapov, Jiri Olsa
Adding TRACE_REG_PERF_ADD and TRACE_REG_PERF_DEL to handle
perf event schedule in/out actions.
The add action is invoked for when the perf event is scheduled in,
while the del action is invoked when the event is scheduled out.
Signed-off-by: Jiri Olsa <jolsa@redhat.com>
Acked-by: Frederic Weisbecker <fweisbec@gmail.com>
---
include/linux/ftrace_event.h | 2 ++
kernel/trace/trace_event_perf.c | 4 +++-
kernel/trace/trace_events.c | 2 ++
kernel/trace/trace_kprobe.c | 2 ++
kernel/trace/trace_syscalls.c | 4 ++++
5 files changed, 13 insertions(+), 1 deletions(-)
diff --git a/include/linux/ftrace_event.h b/include/linux/ftrace_event.h
index 195e360..2bf677c 100644
--- a/include/linux/ftrace_event.h
+++ b/include/linux/ftrace_event.h
@@ -148,6 +148,8 @@ enum trace_reg {
TRACE_REG_PERF_UNREGISTER,
TRACE_REG_PERF_OPEN,
TRACE_REG_PERF_CLOSE,
+ TRACE_REG_PERF_ADD,
+ TRACE_REG_PERF_DEL,
};
struct ftrace_event_call;
diff --git a/kernel/trace/trace_event_perf.c b/kernel/trace/trace_event_perf.c
index 0cfcc37..d72af0b 100644
--- a/kernel/trace/trace_event_perf.c
+++ b/kernel/trace/trace_event_perf.c
@@ -211,12 +211,14 @@ int perf_trace_add(struct perf_event *p_event, int flags)
list = this_cpu_ptr(pcpu_list);
hlist_add_head_rcu(&p_event->hlist_entry, list);
- return 0;
+ return tp_event->class->reg(tp_event, TRACE_REG_PERF_ADD, p_event);
}
void perf_trace_del(struct perf_event *p_event, int flags)
{
+ struct ftrace_event_call *tp_event = p_event->tp_event;
hlist_del_rcu(&p_event->hlist_entry);
+ tp_event->class->reg(tp_event, TRACE_REG_PERF_DEL, p_event);
}
__kprobes void *perf_trace_buf_prepare(int size, unsigned short type,
diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c
index 5138fea..079a93a 100644
--- a/kernel/trace/trace_events.c
+++ b/kernel/trace/trace_events.c
@@ -173,6 +173,8 @@ int ftrace_event_reg(struct ftrace_event_call *call,
return 0;
case TRACE_REG_PERF_OPEN:
case TRACE_REG_PERF_CLOSE:
+ case TRACE_REG_PERF_ADD:
+ case TRACE_REG_PERF_DEL:
return 0;
#endif
}
diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c
index 5667f89..580a05e 100644
--- a/kernel/trace/trace_kprobe.c
+++ b/kernel/trace/trace_kprobe.c
@@ -1912,6 +1912,8 @@ int kprobe_register(struct ftrace_event_call *event,
return 0;
case TRACE_REG_PERF_OPEN:
case TRACE_REG_PERF_CLOSE:
+ case TRACE_REG_PERF_ADD:
+ case TRACE_REG_PERF_DEL:
return 0;
#endif
}
diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c
index 6916b0d..dbdd804 100644
--- a/kernel/trace/trace_syscalls.c
+++ b/kernel/trace/trace_syscalls.c
@@ -666,6 +666,8 @@ static int syscall_enter_register(struct ftrace_event_call *event,
return 0;
case TRACE_REG_PERF_OPEN:
case TRACE_REG_PERF_CLOSE:
+ case TRACE_REG_PERF_ADD:
+ case TRACE_REG_PERF_DEL:
return 0;
#endif
}
@@ -690,6 +692,8 @@ static int syscall_exit_register(struct ftrace_event_call *event,
return 0;
case TRACE_REG_PERF_OPEN:
case TRACE_REG_PERF_CLOSE:
+ case TRACE_REG_PERF_ADD:
+ case TRACE_REG_PERF_DEL:
return 0;
#endif
}
--
1.7.1
^ permalink raw reply related [flat|nested] 17+ messages in thread
* [PATCH 4/7] ftrace: Add FTRACE_ENTRY_REG macro to allow event registration
2012-02-15 14:51 [PATCHv9 0/7] ftrace, perf: Adding support to use function trace Jiri Olsa
` (2 preceding siblings ...)
2012-02-15 14:51 ` [PATCH 3/7] ftrace, perf: Add add/del " Jiri Olsa
@ 2012-02-15 14:51 ` Jiri Olsa
2012-02-15 15:13 ` Frederic Weisbecker
2012-02-27 9:35 ` [tip:perf/core] " tip-bot for Jiri Olsa
2012-02-15 14:51 ` [PATCH 5/7] ftrace, perf: Add support to use function tracepoint in perf Jiri Olsa
` (2 subsequent siblings)
6 siblings, 2 replies; 17+ messages in thread
From: Jiri Olsa @ 2012-02-15 14:51 UTC (permalink / raw)
To: rostedt, fweisbec, mingo, paulus, acme, a.p.zijlstra
Cc: linux-kernel, aarapov, Jiri Olsa
Adding FTRACE_ENTRY_REG macro so particular ftrace entries
could specify registration function and thus become accesible
via perf.
This will be used in upcomming patch for function trace.
Signed-off-by: Jiri Olsa <jolsa@redhat.com>
---
kernel/trace/trace.h | 4 ++++
kernel/trace/trace_export.c | 18 ++++++++++++++++--
2 files changed, 20 insertions(+), 2 deletions(-)
diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h
index 55c6ea0..638476a 100644
--- a/kernel/trace/trace.h
+++ b/kernel/trace/trace.h
@@ -68,6 +68,10 @@ enum trace_type {
#undef FTRACE_ENTRY_DUP
#define FTRACE_ENTRY_DUP(name, name_struct, id, tstruct, printk)
+#undef FTRACE_ENTRY_REG
+#define FTRACE_ENTRY_REG(name, struct_name, id, tstruct, print, regfn) \
+ FTRACE_ENTRY(name, struct_name, id, PARAMS(tstruct), PARAMS(print))
+
#include "trace_entries.h"
/*
diff --git a/kernel/trace/trace_export.c b/kernel/trace/trace_export.c
index bbeec31..f74de86 100644
--- a/kernel/trace/trace_export.c
+++ b/kernel/trace/trace_export.c
@@ -18,6 +18,14 @@
#undef TRACE_SYSTEM
#define TRACE_SYSTEM ftrace
+/*
+ * The FTRACE_ENTRY_REG macro allows ftrace entry to define register
+ * function and thus become accesible via perf.
+ */
+#undef FTRACE_ENTRY_REG
+#define FTRACE_ENTRY_REG(name, struct_name, id, tstruct, print, regfn) \
+ FTRACE_ENTRY(name, struct_name, id, PARAMS(tstruct), PARAMS(print))
+
/* not needed for this file */
#undef __field_struct
#define __field_struct(type, item)
@@ -152,13 +160,14 @@ ftrace_define_fields_##name(struct ftrace_event_call *event_call) \
#undef F_printk
#define F_printk(fmt, args...) #fmt ", " __stringify(args)
-#undef FTRACE_ENTRY
-#define FTRACE_ENTRY(call, struct_name, etype, tstruct, print) \
+#undef FTRACE_ENTRY_REG
+#define FTRACE_ENTRY_REG(call, struct_name, etype, tstruct, print, regfn)\
\
struct ftrace_event_class event_class_ftrace_##call = { \
.system = __stringify(TRACE_SYSTEM), \
.define_fields = ftrace_define_fields_##call, \
.fields = LIST_HEAD_INIT(event_class_ftrace_##call.fields),\
+ .reg = regfn, \
}; \
\
struct ftrace_event_call __used event_##call = { \
@@ -170,4 +179,9 @@ struct ftrace_event_call __used event_##call = { \
struct ftrace_event_call __used \
__attribute__((section("_ftrace_events"))) *__event_##call = &event_##call;
+#undef FTRACE_ENTRY
+#define FTRACE_ENTRY(call, struct_name, etype, tstruct, print) \
+ FTRACE_ENTRY_REG(call, struct_name, etype, \
+ PARAMS(tstruct), PARAMS(print), NULL)
+
#include "trace_entries.h"
--
1.7.1
^ permalink raw reply related [flat|nested] 17+ messages in thread
* [PATCH 5/7] ftrace, perf: Add support to use function tracepoint in perf
2012-02-15 14:51 [PATCHv9 0/7] ftrace, perf: Adding support to use function trace Jiri Olsa
` (3 preceding siblings ...)
2012-02-15 14:51 ` [PATCH 4/7] ftrace: Add FTRACE_ENTRY_REG macro to allow event registration Jiri Olsa
@ 2012-02-15 14:51 ` Jiri Olsa
2012-02-15 16:03 ` Frederic Weisbecker
2012-02-27 9:36 ` [tip:perf/core] " tip-bot for Jiri Olsa
2012-02-15 14:51 ` [PATCH 6/7] ftrace: Allow to specify filter field type for ftrace events Jiri Olsa
2012-02-15 14:51 ` [PATCH 7/7] ftrace, perf: Add filter support for function trace event Jiri Olsa
6 siblings, 2 replies; 17+ messages in thread
From: Jiri Olsa @ 2012-02-15 14:51 UTC (permalink / raw)
To: rostedt, fweisbec, mingo, paulus, acme, a.p.zijlstra
Cc: linux-kernel, aarapov, Jiri Olsa
Adding perf registration support for the ftrace function event,
so it is now possible to register it via perf interface.
The perf_event struct statically contains ftrace_ops as a handle
for function tracer. The function tracer is registered/unregistered
in open/close actions.
To be efficient, we enable/disable ftrace_ops each time the traced
process is scheduled in/out (via TRACE_REG_PERF_(ADD|DELL) handlers).
This way tracing is enabled only when the process is running.
Intentionally using this way instead of the event's hw state
PERF_HES_STOPPED, which would not disable the ftrace_ops.
It is now possible to use function trace within perf commands
like:
perf record -e ftrace:function ls
perf stat -e ftrace:function ls
Allowed only for root.
Signed-off-by: Jiri Olsa <jolsa@redhat.com>
---
include/linux/perf_event.h | 3 +
kernel/trace/trace.h | 11 +++++
kernel/trace/trace_entries.h | 6 ++-
kernel/trace/trace_event_perf.c | 86 +++++++++++++++++++++++++++++++++++++++
kernel/trace/trace_export.c | 5 ++
5 files changed, 109 insertions(+), 2 deletions(-)
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 412b790..92a056f 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -859,6 +859,9 @@ struct perf_event {
#ifdef CONFIG_EVENT_TRACING
struct ftrace_event_call *tp_event;
struct event_filter *filter;
+#ifdef CONFIG_FUNCTION_TRACER
+ struct ftrace_ops ftrace_ops;
+#endif
#endif
#ifdef CONFIG_CGROUP_PERF
diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h
index 638476a..76a1c50 100644
--- a/kernel/trace/trace.h
+++ b/kernel/trace/trace.h
@@ -595,6 +595,8 @@ static inline int ftrace_trace_task(struct task_struct *task)
static inline int ftrace_is_dead(void) { return 0; }
#endif
+int ftrace_event_is_function(struct ftrace_event_call *call);
+
/*
* struct trace_parser - servers for reading the user input separated by spaces
* @cont: set if the input is not complete - no final space char was found
@@ -832,4 +834,13 @@ extern const char *__stop___trace_bprintk_fmt[];
FTRACE_ENTRY(call, struct_name, id, PARAMS(tstruct), PARAMS(print))
#include "trace_entries.h"
+#ifdef CONFIG_PERF_EVENTS
+#ifdef CONFIG_FUNCTION_TRACER
+int perf_ftrace_event_register(struct ftrace_event_call *call,
+ enum trace_reg type, void *data);
+#else
+#define perf_ftrace_event_register NULL
+#endif /* CONFIG_FUNCTION_TRACER */
+#endif /* CONFIG_PERF_EVENTS */
+
#endif /* _LINUX_KERNEL_TRACE_H */
diff --git a/kernel/trace/trace_entries.h b/kernel/trace/trace_entries.h
index 9336590..47db7ed 100644
--- a/kernel/trace/trace_entries.h
+++ b/kernel/trace/trace_entries.h
@@ -55,7 +55,7 @@
/*
* Function trace entry - function address and parent function address:
*/
-FTRACE_ENTRY(function, ftrace_entry,
+FTRACE_ENTRY_REG(function, ftrace_entry,
TRACE_FN,
@@ -64,7 +64,9 @@ FTRACE_ENTRY(function, ftrace_entry,
__field( unsigned long, parent_ip )
),
- F_printk(" %lx <-- %lx", __entry->ip, __entry->parent_ip)
+ F_printk(" %lx <-- %lx", __entry->ip, __entry->parent_ip),
+
+ perf_ftrace_event_register
);
/* Function call entry */
diff --git a/kernel/trace/trace_event_perf.c b/kernel/trace/trace_event_perf.c
index d72af0b..fdeeb5c 100644
--- a/kernel/trace/trace_event_perf.c
+++ b/kernel/trace/trace_event_perf.c
@@ -24,6 +24,11 @@ static int total_ref_count;
static int perf_trace_event_perm(struct ftrace_event_call *tp_event,
struct perf_event *p_event)
{
+ /* The ftrace function trace is allowed only for root. */
+ if (ftrace_event_is_function(tp_event) &&
+ perf_paranoid_kernel() && !capable(CAP_SYS_ADMIN))
+ return -EPERM;
+
/* No tracing, just counting, so no obvious leak */
if (!(p_event->attr.sample_type & PERF_SAMPLE_RAW))
return 0;
@@ -250,3 +255,84 @@ __kprobes void *perf_trace_buf_prepare(int size, unsigned short type,
return raw_data;
}
EXPORT_SYMBOL_GPL(perf_trace_buf_prepare);
+
+#ifdef CONFIG_FUNCTION_TRACER
+static void
+perf_ftrace_function_call(unsigned long ip, unsigned long parent_ip)
+{
+ struct ftrace_entry *entry;
+ struct hlist_head *head;
+ struct pt_regs regs;
+ int rctx;
+
+#define ENTRY_SIZE (ALIGN(sizeof(struct ftrace_entry) + sizeof(u32), \
+ sizeof(u64)) - sizeof(u32))
+
+ BUILD_BUG_ON(ENTRY_SIZE > PERF_MAX_TRACE_SIZE);
+
+ perf_fetch_caller_regs(®s);
+
+ entry = perf_trace_buf_prepare(ENTRY_SIZE, TRACE_FN, NULL, &rctx);
+ if (!entry)
+ return;
+
+ entry->ip = ip;
+ entry->parent_ip = parent_ip;
+
+ head = this_cpu_ptr(event_function.perf_events);
+ perf_trace_buf_submit(entry, ENTRY_SIZE, rctx, 0,
+ 1, ®s, head);
+
+#undef ENTRY_SIZE
+}
+
+static int perf_ftrace_function_register(struct perf_event *event)
+{
+ struct ftrace_ops *ops = &event->ftrace_ops;
+
+ ops->flags |= FTRACE_OPS_FL_CONTROL;
+ ops->func = perf_ftrace_function_call;
+ return register_ftrace_function(ops);
+}
+
+static int perf_ftrace_function_unregister(struct perf_event *event)
+{
+ struct ftrace_ops *ops = &event->ftrace_ops;
+ return unregister_ftrace_function(ops);
+}
+
+static void perf_ftrace_function_enable(struct perf_event *event)
+{
+ ftrace_function_local_enable(&event->ftrace_ops);
+}
+
+static void perf_ftrace_function_disable(struct perf_event *event)
+{
+ ftrace_function_local_disable(&event->ftrace_ops);
+}
+
+int perf_ftrace_event_register(struct ftrace_event_call *call,
+ enum trace_reg type, void *data)
+{
+ switch (type) {
+ case TRACE_REG_REGISTER:
+ case TRACE_REG_UNREGISTER:
+ break;
+ case TRACE_REG_PERF_REGISTER:
+ case TRACE_REG_PERF_UNREGISTER:
+ return 0;
+ case TRACE_REG_PERF_OPEN:
+ return perf_ftrace_function_register(data);
+ case TRACE_REG_PERF_CLOSE:
+ return perf_ftrace_function_unregister(data);
+ case TRACE_REG_PERF_ADD:
+ perf_ftrace_function_enable(data);
+ return 0;
+ case TRACE_REG_PERF_DEL:
+ perf_ftrace_function_disable(data);
+ return 0;
+ }
+
+ return -EINVAL;
+}
+#endif /* CONFIG_FUNCTION_TRACER */
diff --git a/kernel/trace/trace_export.c b/kernel/trace/trace_export.c
index f74de86..a3dbee6 100644
--- a/kernel/trace/trace_export.c
+++ b/kernel/trace/trace_export.c
@@ -184,4 +184,9 @@ __attribute__((section("_ftrace_events"))) *__event_##call = &event_##call;
FTRACE_ENTRY_REG(call, struct_name, etype, \
PARAMS(tstruct), PARAMS(print), NULL)
+int ftrace_event_is_function(struct ftrace_event_call *call)
+{
+ return call == &event_function;
+}
+
#include "trace_entries.h"
--
1.7.1
^ permalink raw reply related [flat|nested] 17+ messages in thread
* [PATCH 6/7] ftrace: Allow to specify filter field type for ftrace events
2012-02-15 14:51 [PATCHv9 0/7] ftrace, perf: Adding support to use function trace Jiri Olsa
` (4 preceding siblings ...)
2012-02-15 14:51 ` [PATCH 5/7] ftrace, perf: Add support to use function tracepoint in perf Jiri Olsa
@ 2012-02-15 14:51 ` Jiri Olsa
2012-02-27 9:37 ` [tip:perf/core] " tip-bot for Jiri Olsa
2012-02-15 14:51 ` [PATCH 7/7] ftrace, perf: Add filter support for function trace event Jiri Olsa
6 siblings, 1 reply; 17+ messages in thread
From: Jiri Olsa @ 2012-02-15 14:51 UTC (permalink / raw)
To: rostedt, fweisbec, mingo, paulus, acme, a.p.zijlstra
Cc: linux-kernel, aarapov, Jiri Olsa
Adding FILTER_TRACE_FN event field type for function tracepoint
event, so it can be properly recognized within filtering code.
Currently all fields of ftrace subsystem events share the common
field type FILTER_OTHER. Since the function trace fields need
special care within the filtering code we need to recognize it
properly, hence adding the FILTER_TRACE_FN event type.
Adding filter parameter to the FTRACE_ENTRY macro, to specify the
filter field type for the event.
Signed-off-by: Jiri Olsa <jolsa@redhat.com>
---
include/linux/ftrace_event.h | 1 +
kernel/trace/trace.h | 23 +++++++++-------
kernel/trace/trace_entries.h | 48 ++++++++++++++++++++++++---------
kernel/trace/trace_events_filter.c | 7 ++++-
kernel/trace/trace_export.c | 51 +++++++++++++++++++----------------
5 files changed, 83 insertions(+), 47 deletions(-)
diff --git a/include/linux/ftrace_event.h b/include/linux/ftrace_event.h
index 2bf677c..dd478fc 100644
--- a/include/linux/ftrace_event.h
+++ b/include/linux/ftrace_event.h
@@ -245,6 +245,7 @@ enum {
FILTER_STATIC_STRING,
FILTER_DYN_STRING,
FILTER_PTR_STRING,
+ FILTER_TRACE_FN,
};
#define EVENT_STORAGE_SIZE 128
diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h
index 76a1c50..29f93cd 100644
--- a/kernel/trace/trace.h
+++ b/kernel/trace/trace.h
@@ -56,21 +56,23 @@ enum trace_type {
#define F_STRUCT(args...) args
#undef FTRACE_ENTRY
-#define FTRACE_ENTRY(name, struct_name, id, tstruct, print) \
- struct struct_name { \
- struct trace_entry ent; \
- tstruct \
+#define FTRACE_ENTRY(name, struct_name, id, tstruct, print, filter) \
+ struct struct_name { \
+ struct trace_entry ent; \
+ tstruct \
}
#undef TP_ARGS
#define TP_ARGS(args...) args
#undef FTRACE_ENTRY_DUP
-#define FTRACE_ENTRY_DUP(name, name_struct, id, tstruct, printk)
+#define FTRACE_ENTRY_DUP(name, name_struct, id, tstruct, printk, filter)
#undef FTRACE_ENTRY_REG
-#define FTRACE_ENTRY_REG(name, struct_name, id, tstruct, print, regfn) \
- FTRACE_ENTRY(name, struct_name, id, PARAMS(tstruct), PARAMS(print))
+#define FTRACE_ENTRY_REG(name, struct_name, id, tstruct, print, \
+ filter, regfn) \
+ FTRACE_ENTRY(name, struct_name, id, PARAMS(tstruct), PARAMS(print), \
+ filter)
#include "trace_entries.h"
@@ -826,12 +828,13 @@ extern const char *__start___trace_bprintk_fmt[];
extern const char *__stop___trace_bprintk_fmt[];
#undef FTRACE_ENTRY
-#define FTRACE_ENTRY(call, struct_name, id, tstruct, print) \
+#define FTRACE_ENTRY(call, struct_name, id, tstruct, print, filter) \
extern struct ftrace_event_call \
__attribute__((__aligned__(4))) event_##call;
#undef FTRACE_ENTRY_DUP
-#define FTRACE_ENTRY_DUP(call, struct_name, id, tstruct, print) \
- FTRACE_ENTRY(call, struct_name, id, PARAMS(tstruct), PARAMS(print))
+#define FTRACE_ENTRY_DUP(call, struct_name, id, tstruct, print, filter) \
+ FTRACE_ENTRY(call, struct_name, id, PARAMS(tstruct), PARAMS(print), \
+ filter)
#include "trace_entries.h"
#ifdef CONFIG_PERF_EVENTS
diff --git a/kernel/trace/trace_entries.h b/kernel/trace/trace_entries.h
index 47db7ed..d91eb05 100644
--- a/kernel/trace/trace_entries.h
+++ b/kernel/trace/trace_entries.h
@@ -66,6 +66,8 @@ FTRACE_ENTRY_REG(function, ftrace_entry,
F_printk(" %lx <-- %lx", __entry->ip, __entry->parent_ip),
+ FILTER_TRACE_FN,
+
perf_ftrace_event_register
);
@@ -80,7 +82,9 @@ FTRACE_ENTRY(funcgraph_entry, ftrace_graph_ent_entry,
__field_desc( int, graph_ent, depth )
),
- F_printk("--> %lx (%d)", __entry->func, __entry->depth)
+ F_printk("--> %lx (%d)", __entry->func, __entry->depth),
+
+ FILTER_OTHER
);
/* Function return entry */
@@ -100,7 +104,9 @@ FTRACE_ENTRY(funcgraph_exit, ftrace_graph_ret_entry,
F_printk("<-- %lx (%d) (start: %llx end: %llx) over: %d",
__entry->func, __entry->depth,
__entry->calltime, __entry->rettime,
- __entry->depth)
+ __entry->depth),
+
+ FILTER_OTHER
);
/*
@@ -129,8 +135,9 @@ FTRACE_ENTRY(context_switch, ctx_switch_entry,
F_printk("%u:%u:%u ==> %u:%u:%u [%03u]",
__entry->prev_pid, __entry->prev_prio, __entry->prev_state,
__entry->next_pid, __entry->next_prio, __entry->next_state,
- __entry->next_cpu
- )
+ __entry->next_cpu),
+
+ FILTER_OTHER
);
/*
@@ -148,8 +155,9 @@ FTRACE_ENTRY_DUP(wakeup, ctx_switch_entry,
F_printk("%u:%u:%u ==+ %u:%u:%u [%03u]",
__entry->prev_pid, __entry->prev_prio, __entry->prev_state,
__entry->next_pid, __entry->next_prio, __entry->next_state,
- __entry->next_cpu
- )
+ __entry->next_cpu),
+
+ FILTER_OTHER
);
/*
@@ -171,7 +179,9 @@ FTRACE_ENTRY(kernel_stack, stack_entry,
"\t=> (%08lx)\n\t=> (%08lx)\n\t=> (%08lx)\n\t=> (%08lx)\n",
__entry->caller[0], __entry->caller[1], __entry->caller[2],
__entry->caller[3], __entry->caller[4], __entry->caller[5],
- __entry->caller[6], __entry->caller[7])
+ __entry->caller[6], __entry->caller[7]),
+
+ FILTER_OTHER
);
FTRACE_ENTRY(user_stack, userstack_entry,
@@ -187,7 +197,9 @@ FTRACE_ENTRY(user_stack, userstack_entry,
"\t=> (%08lx)\n\t=> (%08lx)\n\t=> (%08lx)\n\t=> (%08lx)\n",
__entry->caller[0], __entry->caller[1], __entry->caller[2],
__entry->caller[3], __entry->caller[4], __entry->caller[5],
- __entry->caller[6], __entry->caller[7])
+ __entry->caller[6], __entry->caller[7]),
+
+ FILTER_OTHER
);
/*
@@ -204,7 +216,9 @@ FTRACE_ENTRY(bprint, bprint_entry,
),
F_printk("%08lx fmt:%p",
- __entry->ip, __entry->fmt)
+ __entry->ip, __entry->fmt),
+
+ FILTER_OTHER
);
FTRACE_ENTRY(print, print_entry,
@@ -217,7 +231,9 @@ FTRACE_ENTRY(print, print_entry,
),
F_printk("%08lx %s",
- __entry->ip, __entry->buf)
+ __entry->ip, __entry->buf),
+
+ FILTER_OTHER
);
FTRACE_ENTRY(mmiotrace_rw, trace_mmiotrace_rw,
@@ -236,7 +252,9 @@ FTRACE_ENTRY(mmiotrace_rw, trace_mmiotrace_rw,
F_printk("%lx %lx %lx %d %x %x",
(unsigned long)__entry->phys, __entry->value, __entry->pc,
- __entry->map_id, __entry->opcode, __entry->width)
+ __entry->map_id, __entry->opcode, __entry->width),
+
+ FILTER_OTHER
);
FTRACE_ENTRY(mmiotrace_map, trace_mmiotrace_map,
@@ -254,7 +272,9 @@ FTRACE_ENTRY(mmiotrace_map, trace_mmiotrace_map,
F_printk("%lx %lx %lx %d %x",
(unsigned long)__entry->phys, __entry->virt, __entry->len,
- __entry->map_id, __entry->opcode)
+ __entry->map_id, __entry->opcode),
+
+ FILTER_OTHER
);
@@ -274,6 +294,8 @@ FTRACE_ENTRY(branch, trace_branch,
F_printk("%u:%s:%s (%u)",
__entry->line,
- __entry->func, __entry->file, __entry->correct)
+ __entry->func, __entry->file, __entry->correct),
+
+ FILTER_OTHER
);
diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c
index 24aee71..eb04a2a 100644
--- a/kernel/trace/trace_events_filter.c
+++ b/kernel/trace/trace_events_filter.c
@@ -900,6 +900,11 @@ int filter_assign_type(const char *type)
return FILTER_OTHER;
}
+static bool is_function_field(struct ftrace_event_field *field)
+{
+ return field->filter_type == FILTER_TRACE_FN;
+}
+
static bool is_string_field(struct ftrace_event_field *field)
{
return field->filter_type == FILTER_DYN_STRING ||
@@ -987,7 +992,7 @@ static int init_pred(struct filter_parse_state *ps,
fn = filter_pred_strloc;
else
fn = filter_pred_pchar;
- } else {
+ } else if (!is_function_field(field)) {
if (field->is_signed)
ret = strict_strtoll(pred->regex.pattern, 0, &val);
else
diff --git a/kernel/trace/trace_export.c b/kernel/trace/trace_export.c
index a3dbee6..7b46c9b 100644
--- a/kernel/trace/trace_export.c
+++ b/kernel/trace/trace_export.c
@@ -23,8 +23,10 @@
* function and thus become accesible via perf.
*/
#undef FTRACE_ENTRY_REG
-#define FTRACE_ENTRY_REG(name, struct_name, id, tstruct, print, regfn) \
- FTRACE_ENTRY(name, struct_name, id, PARAMS(tstruct), PARAMS(print))
+#define FTRACE_ENTRY_REG(name, struct_name, id, tstruct, print, \
+ filter, regfn) \
+ FTRACE_ENTRY(name, struct_name, id, PARAMS(tstruct), PARAMS(print), \
+ filter)
/* not needed for this file */
#undef __field_struct
@@ -52,21 +54,22 @@
#define F_printk(fmt, args...) fmt, args
#undef FTRACE_ENTRY
-#define FTRACE_ENTRY(name, struct_name, id, tstruct, print) \
-struct ____ftrace_##name { \
- tstruct \
-}; \
-static void __always_unused ____ftrace_check_##name(void) \
-{ \
- struct ____ftrace_##name *__entry = NULL; \
- \
- /* force compile-time check on F_printk() */ \
- printk(print); \
+#define FTRACE_ENTRY(name, struct_name, id, tstruct, print, filter) \
+struct ____ftrace_##name { \
+ tstruct \
+}; \
+static void __always_unused ____ftrace_check_##name(void) \
+{ \
+ struct ____ftrace_##name *__entry = NULL; \
+ \
+ /* force compile-time check on F_printk() */ \
+ printk(print); \
}
#undef FTRACE_ENTRY_DUP
-#define FTRACE_ENTRY_DUP(name, struct_name, id, tstruct, print) \
- FTRACE_ENTRY(name, struct_name, id, PARAMS(tstruct), PARAMS(print))
+#define FTRACE_ENTRY_DUP(name, struct_name, id, tstruct, print, filter) \
+ FTRACE_ENTRY(name, struct_name, id, PARAMS(tstruct), PARAMS(print), \
+ filter)
#include "trace_entries.h"
@@ -75,7 +78,7 @@ static void __always_unused ____ftrace_check_##name(void) \
ret = trace_define_field(event_call, #type, #item, \
offsetof(typeof(field), item), \
sizeof(field.item), \
- is_signed_type(type), FILTER_OTHER); \
+ is_signed_type(type), filter_type); \
if (ret) \
return ret;
@@ -85,7 +88,7 @@ static void __always_unused ____ftrace_check_##name(void) \
offsetof(typeof(field), \
container.item), \
sizeof(field.container.item), \
- is_signed_type(type), FILTER_OTHER); \
+ is_signed_type(type), filter_type); \
if (ret) \
return ret;
@@ -99,7 +102,7 @@ static void __always_unused ____ftrace_check_##name(void) \
ret = trace_define_field(event_call, event_storage, #item, \
offsetof(typeof(field), item), \
sizeof(field.item), \
- is_signed_type(type), FILTER_OTHER); \
+ is_signed_type(type), filter_type); \
mutex_unlock(&event_storage_mutex); \
if (ret) \
return ret; \
@@ -112,7 +115,7 @@ static void __always_unused ____ftrace_check_##name(void) \
offsetof(typeof(field), \
container.item), \
sizeof(field.container.item), \
- is_signed_type(type), FILTER_OTHER); \
+ is_signed_type(type), filter_type); \
if (ret) \
return ret;
@@ -120,17 +123,18 @@ static void __always_unused ____ftrace_check_##name(void) \
#define __dynamic_array(type, item) \
ret = trace_define_field(event_call, #type, #item, \
offsetof(typeof(field), item), \
- 0, is_signed_type(type), FILTER_OTHER);\
+ 0, is_signed_type(type), filter_type);\
if (ret) \
return ret;
#undef FTRACE_ENTRY
-#define FTRACE_ENTRY(name, struct_name, id, tstruct, print) \
+#define FTRACE_ENTRY(name, struct_name, id, tstruct, print, filter) \
int \
ftrace_define_fields_##name(struct ftrace_event_call *event_call) \
{ \
struct struct_name field; \
int ret; \
+ int filter_type = filter; \
\
tstruct; \
\
@@ -161,7 +165,8 @@ ftrace_define_fields_##name(struct ftrace_event_call *event_call) \
#define F_printk(fmt, args...) #fmt ", " __stringify(args)
#undef FTRACE_ENTRY_REG
-#define FTRACE_ENTRY_REG(call, struct_name, etype, tstruct, print, regfn)\
+#define FTRACE_ENTRY_REG(call, struct_name, etype, tstruct, print, filter,\
+ regfn) \
\
struct ftrace_event_class event_class_ftrace_##call = { \
.system = __stringify(TRACE_SYSTEM), \
@@ -180,9 +185,9 @@ struct ftrace_event_call __used \
__attribute__((section("_ftrace_events"))) *__event_##call = &event_##call;
#undef FTRACE_ENTRY
-#define FTRACE_ENTRY(call, struct_name, etype, tstruct, print) \
+#define FTRACE_ENTRY(call, struct_name, etype, tstruct, print, filter) \
FTRACE_ENTRY_REG(call, struct_name, etype, \
- PARAMS(tstruct), PARAMS(print), NULL)
+ PARAMS(tstruct), PARAMS(print), filter, NULL)
int ftrace_event_is_function(struct ftrace_event_call *call)
{
--
1.7.1
^ permalink raw reply related [flat|nested] 17+ messages in thread
* [PATCH 7/7] ftrace, perf: Add filter support for function trace event
2012-02-15 14:51 [PATCHv9 0/7] ftrace, perf: Adding support to use function trace Jiri Olsa
` (5 preceding siblings ...)
2012-02-15 14:51 ` [PATCH 6/7] ftrace: Allow to specify filter field type for ftrace events Jiri Olsa
@ 2012-02-15 14:51 ` Jiri Olsa
2012-02-27 9:38 ` [tip:perf/core] " tip-bot for Jiri Olsa
6 siblings, 1 reply; 17+ messages in thread
From: Jiri Olsa @ 2012-02-15 14:51 UTC (permalink / raw)
To: rostedt, fweisbec, mingo, paulus, acme, a.p.zijlstra
Cc: linux-kernel, aarapov, Jiri Olsa
Adding support to filter function trace event via perf
interface. It is now possible to use filter interface
in the perf tool like:
perf record -e ftrace:function --filter="(ip == mm_*)" ls
The filter syntax is restricted to the the 'ip' field only,
and following operators are accepted '==' '!=' '||', ending
up with the filter strings like:
ip == f1[, ]f2 ... || ip != f3[, ]f4 ...
with comma ',' or space ' ' as a function separator. If the
space ' ' is used as a separator, the right side of the
assignment needs to be enclosed in double quotes '"', e.g.:
perf record -e ftrace:function --filter '(ip == do_execve,sys_*,ext*)' ls
perf record -e ftrace:function --filter '(ip == "do_execve,sys_*,ext*")' ls
perf record -e ftrace:function --filter '(ip == "do_execve sys_* ext*")' ls
The '==' operator adds trace filter with same effect as would
be added via set_ftrace_filter file.
The '!=' operator adds trace filter with same effect as would
be added via set_ftrace_notrace file.
The right side of the '!=', '==' operators is list of functions
or regexp. to be added to filter separated by space.
The '||' operator is used for connecting multiple filter definitions
together. It is possible to have more than one '==' and '!='
operators within one filter string.
Signed-off-by: Jiri Olsa <jolsa@redhat.com>
---
include/linux/ftrace.h | 7 +-
kernel/trace/ftrace.c | 6 ++
kernel/trace/trace.h | 2 -
kernel/trace/trace_event_perf.c | 4 +-
kernel/trace/trace_events_filter.c | 165 ++++++++++++++++++++++++++++++++++--
5 files changed, 172 insertions(+), 12 deletions(-)
diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h
index 64a309d..72a6cab 100644
--- a/include/linux/ftrace.h
+++ b/include/linux/ftrace.h
@@ -250,6 +250,7 @@ int ftrace_set_notrace(struct ftrace_ops *ops, unsigned char *buf,
int len, int reset);
void ftrace_set_global_filter(unsigned char *buf, int len, int reset);
void ftrace_set_global_notrace(unsigned char *buf, int len, int reset);
+void ftrace_free_filter(struct ftrace_ops *ops);
int register_ftrace_command(struct ftrace_func_command *cmd);
int unregister_ftrace_command(struct ftrace_func_command *cmd);
@@ -380,9 +381,6 @@ extern void ftrace_enable_daemon(void);
#else
static inline int skip_trace(unsigned long ip) { return 0; }
static inline int ftrace_force_update(void) { return 0; }
-static inline void ftrace_set_filter(unsigned char *buf, int len, int reset)
-{
-}
static inline void ftrace_disable_daemon(void) { }
static inline void ftrace_enable_daemon(void) { }
static inline void ftrace_release_mod(struct module *mod) {}
@@ -406,6 +404,9 @@ static inline int ftrace_text_reserved(void *start, void *end)
*/
#define ftrace_regex_open(ops, flag, inod, file) ({ -ENODEV; })
#define ftrace_set_early_filter(ops, buf, enable) do { } while (0)
+#define ftrace_set_filter(ops, buf, len, reset) ({ -ENODEV; })
+#define ftrace_set_notrace(ops, buf, len, reset) ({ -ENODEV; })
+#define ftrace_free_filter(ops) do { } while (0)
static inline ssize_t ftrace_filter_write(struct file *file, const char __user *ubuf,
size_t cnt, loff_t *ppos) { return -ENODEV; }
diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
index c8d2af2..239b94a 100644
--- a/kernel/trace/ftrace.c
+++ b/kernel/trace/ftrace.c
@@ -1186,6 +1186,12 @@ static void free_ftrace_hash_rcu(struct ftrace_hash *hash)
call_rcu_sched(&hash->rcu, __free_ftrace_hash_rcu);
}
+void ftrace_free_filter(struct ftrace_ops *ops)
+{
+ free_ftrace_hash(ops->filter_hash);
+ free_ftrace_hash(ops->notrace_hash);
+}
+
static struct ftrace_hash *alloc_ftrace_hash(int size_bits)
{
struct ftrace_hash *hash;
diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h
index 29f93cd..54faec7 100644
--- a/kernel/trace/trace.h
+++ b/kernel/trace/trace.h
@@ -776,9 +776,7 @@ struct filter_pred {
u64 val;
struct regex regex;
unsigned short *ops;
-#ifdef CONFIG_FTRACE_STARTUP_TEST
struct ftrace_event_field *field;
-#endif
int offset;
int not;
int op;
diff --git a/kernel/trace/trace_event_perf.c b/kernel/trace/trace_event_perf.c
index fdeeb5c..fee3752 100644
--- a/kernel/trace/trace_event_perf.c
+++ b/kernel/trace/trace_event_perf.c
@@ -298,7 +298,9 @@ static int perf_ftrace_function_register(struct perf_event *event)
static int perf_ftrace_function_unregister(struct perf_event *event)
{
struct ftrace_ops *ops = &event->ftrace_ops;
- return unregister_ftrace_function(ops);
+ int ret = unregister_ftrace_function(ops);
+ ftrace_free_filter(ops);
+ return ret;
}
static void perf_ftrace_function_enable(struct perf_event *event)
diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c
index eb04a2a..4d7ac74 100644
--- a/kernel/trace/trace_events_filter.c
+++ b/kernel/trace/trace_events_filter.c
@@ -81,6 +81,7 @@ enum {
FILT_ERR_TOO_MANY_PREDS,
FILT_ERR_MISSING_FIELD,
FILT_ERR_INVALID_FILTER,
+ FILT_ERR_IP_FIELD_ONLY,
};
static char *err_text[] = {
@@ -96,6 +97,7 @@ static char *err_text[] = {
"Too many terms in predicate expression",
"Missing field name and/or value",
"Meaningless filter expression",
+ "Only 'ip' field is supported for function trace",
};
struct opstack_op {
@@ -992,7 +994,12 @@ static int init_pred(struct filter_parse_state *ps,
fn = filter_pred_strloc;
else
fn = filter_pred_pchar;
- } else if (!is_function_field(field)) {
+ } else if (is_function_field(field)) {
+ if (strcmp(field->name, "ip")) {
+ parse_error(ps, FILT_ERR_IP_FIELD_ONLY, 0);
+ return -EINVAL;
+ }
+ } else {
if (field->is_signed)
ret = strict_strtoll(pred->regex.pattern, 0, &val);
else
@@ -1339,10 +1346,7 @@ static struct filter_pred *create_pred(struct filter_parse_state *ps,
strcpy(pred.regex.pattern, operand2);
pred.regex.len = strlen(pred.regex.pattern);
-
-#ifdef CONFIG_FTRACE_STARTUP_TEST
pred.field = field;
-#endif
return init_pred(ps, field, &pred) ? NULL : &pred;
}
@@ -1955,6 +1959,148 @@ void ftrace_profile_free_filter(struct perf_event *event)
__free_filter(filter);
}
+struct function_filter_data {
+ struct ftrace_ops *ops;
+ int first_filter;
+ int first_notrace;
+};
+
+#ifdef CONFIG_FUNCTION_TRACER
+static char **
+ftrace_function_filter_re(char *buf, int len, int *count)
+{
+ char *str, *sep, **re;
+
+ str = kstrndup(buf, len, GFP_KERNEL);
+ if (!str)
+ return NULL;
+
+ /*
+ * The argv_split function takes white space
+ * as a separator, so convert ',' into spaces.
+ */
+ while ((sep = strchr(str, ',')))
+ *sep = ' ';
+
+ re = argv_split(GFP_KERNEL, str, count);
+ kfree(str);
+ return re;
+}
+
+static int ftrace_function_set_regexp(struct ftrace_ops *ops, int filter,
+ int reset, char *re, int len)
+{
+ int ret;
+
+ if (filter)
+ ret = ftrace_set_filter(ops, re, len, reset);
+ else
+ ret = ftrace_set_notrace(ops, re, len, reset);
+
+ return ret;
+}
+
+static int __ftrace_function_set_filter(int filter, char *buf, int len,
+ struct function_filter_data *data)
+{
+ int i, re_cnt, ret;
+ int *reset;
+ char **re;
+
+ reset = filter ? &data->first_filter : &data->first_notrace;
+
+ /*
+ * The 'ip' field could have multiple filters set, separated
+ * either by space or comma. We first cut the filter and apply
+ * all pieces separatelly.
+ */
+ re = ftrace_function_filter_re(buf, len, &re_cnt);
+ if (!re)
+ return -EINVAL;
+
+ for (i = 0; i < re_cnt; i++) {
+ ret = ftrace_function_set_regexp(data->ops, filter, *reset,
+ re[i], strlen(re[i]));
+ if (ret)
+ break;
+
+ if (*reset)
+ *reset = 0;
+ }
+
+ argv_free(re);
+ return ret;
+}
+
+static int ftrace_function_check_pred(struct filter_pred *pred, int leaf)
+{
+ struct ftrace_event_field *field = pred->field;
+
+ if (leaf) {
+ /*
+ * Check the leaf predicate for function trace, verify:
+ * - only '==' and '!=' is used
+ * - the 'ip' field is used
+ */
+ if ((pred->op != OP_EQ) && (pred->op != OP_NE))
+ return -EINVAL;
+
+ if (strcmp(field->name, "ip"))
+ return -EINVAL;
+ } else {
+ /*
+ * Check the non leaf predicate for function trace, verify:
+ * - only '||' is used
+ */
+ if (pred->op != OP_OR)
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+static int ftrace_function_set_filter_cb(enum move_type move,
+ struct filter_pred *pred,
+ int *err, void *data)
+{
+ /* Checking the node is valid for function trace. */
+ if ((move != MOVE_DOWN) ||
+ (pred->left != FILTER_PRED_INVALID)) {
+ *err = ftrace_function_check_pred(pred, 0);
+ } else {
+ *err = ftrace_function_check_pred(pred, 1);
+ if (*err)
+ return WALK_PRED_ABORT;
+
+ *err = __ftrace_function_set_filter(pred->op == OP_EQ,
+ pred->regex.pattern,
+ pred->regex.len,
+ data);
+ }
+
+ return (*err) ? WALK_PRED_ABORT : WALK_PRED_DEFAULT;
+}
+
+static int ftrace_function_set_filter(struct perf_event *event,
+ struct event_filter *filter)
+{
+ struct function_filter_data data = {
+ .first_filter = 1,
+ .first_notrace = 1,
+ .ops = &event->ftrace_ops,
+ };
+
+ return walk_pred_tree(filter->preds, filter->root,
+ ftrace_function_set_filter_cb, &data);
+}
+#else
+static int ftrace_function_set_filter(struct perf_event *event,
+ struct event_filter *filter)
+{
+ return -ENODEV;
+}
+#endif /* CONFIG_FUNCTION_TRACER */
+
int ftrace_profile_set_filter(struct perf_event *event, int event_id,
char *filter_str)
{
@@ -1975,9 +2121,16 @@ int ftrace_profile_set_filter(struct perf_event *event, int event_id,
goto out_unlock;
err = create_filter(call, filter_str, false, &filter);
- if (!err)
- event->filter = filter;
+ if (err)
+ goto free_filter;
+
+ if (ftrace_event_is_function(call))
+ err = ftrace_function_set_filter(event, filter);
else
+ event->filter = filter;
+
+free_filter:
+ if (err || ftrace_event_is_function(call))
__free_filter(filter);
out_unlock:
--
1.7.1
^ permalink raw reply related [flat|nested] 17+ messages in thread
* Re: [PATCH 4/7] ftrace: Add FTRACE_ENTRY_REG macro to allow event registration
2012-02-15 14:51 ` [PATCH 4/7] ftrace: Add FTRACE_ENTRY_REG macro to allow event registration Jiri Olsa
@ 2012-02-15 15:13 ` Frederic Weisbecker
2012-02-27 9:35 ` [tip:perf/core] " tip-bot for Jiri Olsa
1 sibling, 0 replies; 17+ messages in thread
From: Frederic Weisbecker @ 2012-02-15 15:13 UTC (permalink / raw)
To: Jiri Olsa
Cc: rostedt, mingo, paulus, acme, a.p.zijlstra, linux-kernel, aarapov
On Wed, Feb 15, 2012 at 03:51:51PM +0100, Jiri Olsa wrote:
> Adding FTRACE_ENTRY_REG macro so particular ftrace entries
> could specify registration function and thus become accesible
> via perf.
>
> This will be used in upcomming patch for function trace.
>
> Signed-off-by: Jiri Olsa <jolsa@redhat.com>
Acked-by: Frederic Weisbecker <fweisbec@gmail.com>
> ---
> kernel/trace/trace.h | 4 ++++
> kernel/trace/trace_export.c | 18 ++++++++++++++++--
> 2 files changed, 20 insertions(+), 2 deletions(-)
>
> diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h
> index 55c6ea0..638476a 100644
> --- a/kernel/trace/trace.h
> +++ b/kernel/trace/trace.h
> @@ -68,6 +68,10 @@ enum trace_type {
> #undef FTRACE_ENTRY_DUP
> #define FTRACE_ENTRY_DUP(name, name_struct, id, tstruct, printk)
>
> +#undef FTRACE_ENTRY_REG
> +#define FTRACE_ENTRY_REG(name, struct_name, id, tstruct, print, regfn) \
> + FTRACE_ENTRY(name, struct_name, id, PARAMS(tstruct), PARAMS(print))
> +
> #include "trace_entries.h"
>
> /*
> diff --git a/kernel/trace/trace_export.c b/kernel/trace/trace_export.c
> index bbeec31..f74de86 100644
> --- a/kernel/trace/trace_export.c
> +++ b/kernel/trace/trace_export.c
> @@ -18,6 +18,14 @@
> #undef TRACE_SYSTEM
> #define TRACE_SYSTEM ftrace
>
> +/*
> + * The FTRACE_ENTRY_REG macro allows ftrace entry to define register
> + * function and thus become accesible via perf.
> + */
> +#undef FTRACE_ENTRY_REG
> +#define FTRACE_ENTRY_REG(name, struct_name, id, tstruct, print, regfn) \
> + FTRACE_ENTRY(name, struct_name, id, PARAMS(tstruct), PARAMS(print))
> +
> /* not needed for this file */
> #undef __field_struct
> #define __field_struct(type, item)
> @@ -152,13 +160,14 @@ ftrace_define_fields_##name(struct ftrace_event_call *event_call) \
> #undef F_printk
> #define F_printk(fmt, args...) #fmt ", " __stringify(args)
>
> -#undef FTRACE_ENTRY
> -#define FTRACE_ENTRY(call, struct_name, etype, tstruct, print) \
> +#undef FTRACE_ENTRY_REG
> +#define FTRACE_ENTRY_REG(call, struct_name, etype, tstruct, print, regfn)\
> \
> struct ftrace_event_class event_class_ftrace_##call = { \
> .system = __stringify(TRACE_SYSTEM), \
> .define_fields = ftrace_define_fields_##call, \
> .fields = LIST_HEAD_INIT(event_class_ftrace_##call.fields),\
> + .reg = regfn, \
> }; \
> \
> struct ftrace_event_call __used event_##call = { \
> @@ -170,4 +179,9 @@ struct ftrace_event_call __used event_##call = { \
> struct ftrace_event_call __used \
> __attribute__((section("_ftrace_events"))) *__event_##call = &event_##call;
>
> +#undef FTRACE_ENTRY
> +#define FTRACE_ENTRY(call, struct_name, etype, tstruct, print) \
> + FTRACE_ENTRY_REG(call, struct_name, etype, \
> + PARAMS(tstruct), PARAMS(print), NULL)
> +
> #include "trace_entries.h"
> --
> 1.7.1
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PATCH 5/7] ftrace, perf: Add support to use function tracepoint in perf
2012-02-15 14:51 ` [PATCH 5/7] ftrace, perf: Add support to use function tracepoint in perf Jiri Olsa
@ 2012-02-15 16:03 ` Frederic Weisbecker
2012-02-27 9:36 ` [tip:perf/core] " tip-bot for Jiri Olsa
1 sibling, 0 replies; 17+ messages in thread
From: Frederic Weisbecker @ 2012-02-15 16:03 UTC (permalink / raw)
To: Jiri Olsa
Cc: rostedt, mingo, paulus, acme, a.p.zijlstra, linux-kernel, aarapov
On Wed, Feb 15, 2012 at 03:51:52PM +0100, Jiri Olsa wrote:
> Adding perf registration support for the ftrace function event,
> so it is now possible to register it via perf interface.
>
> The perf_event struct statically contains ftrace_ops as a handle
> for function tracer. The function tracer is registered/unregistered
> in open/close actions.
>
> To be efficient, we enable/disable ftrace_ops each time the traced
> process is scheduled in/out (via TRACE_REG_PERF_(ADD|DELL) handlers).
> This way tracing is enabled only when the process is running.
> Intentionally using this way instead of the event's hw state
> PERF_HES_STOPPED, which would not disable the ftrace_ops.
>
> It is now possible to use function trace within perf commands
> like:
>
> perf record -e ftrace:function ls
> perf stat -e ftrace:function ls
>
> Allowed only for root.
>
> Signed-off-by: Jiri Olsa <jolsa@redhat.com>
Acked-by: Frederic Weisbecker <fweisbec@gmail.com>
> ---
> include/linux/perf_event.h | 3 +
> kernel/trace/trace.h | 11 +++++
> kernel/trace/trace_entries.h | 6 ++-
> kernel/trace/trace_event_perf.c | 86 +++++++++++++++++++++++++++++++++++++++
> kernel/trace/trace_export.c | 5 ++
> 5 files changed, 109 insertions(+), 2 deletions(-)
>
> diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
> index 412b790..92a056f 100644
> --- a/include/linux/perf_event.h
> +++ b/include/linux/perf_event.h
> @@ -859,6 +859,9 @@ struct perf_event {
> #ifdef CONFIG_EVENT_TRACING
> struct ftrace_event_call *tp_event;
> struct event_filter *filter;
> +#ifdef CONFIG_FUNCTION_TRACER
> + struct ftrace_ops ftrace_ops;
> +#endif
> #endif
>
> #ifdef CONFIG_CGROUP_PERF
> diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h
> index 638476a..76a1c50 100644
> --- a/kernel/trace/trace.h
> +++ b/kernel/trace/trace.h
> @@ -595,6 +595,8 @@ static inline int ftrace_trace_task(struct task_struct *task)
> static inline int ftrace_is_dead(void) { return 0; }
> #endif
>
> +int ftrace_event_is_function(struct ftrace_event_call *call);
> +
> /*
> * struct trace_parser - servers for reading the user input separated by spaces
> * @cont: set if the input is not complete - no final space char was found
> @@ -832,4 +834,13 @@ extern const char *__stop___trace_bprintk_fmt[];
> FTRACE_ENTRY(call, struct_name, id, PARAMS(tstruct), PARAMS(print))
> #include "trace_entries.h"
>
> +#ifdef CONFIG_PERF_EVENTS
> +#ifdef CONFIG_FUNCTION_TRACER
> +int perf_ftrace_event_register(struct ftrace_event_call *call,
> + enum trace_reg type, void *data);
> +#else
> +#define perf_ftrace_event_register NULL
> +#endif /* CONFIG_FUNCTION_TRACER */
> +#endif /* CONFIG_PERF_EVENTS */
> +
> #endif /* _LINUX_KERNEL_TRACE_H */
> diff --git a/kernel/trace/trace_entries.h b/kernel/trace/trace_entries.h
> index 9336590..47db7ed 100644
> --- a/kernel/trace/trace_entries.h
> +++ b/kernel/trace/trace_entries.h
> @@ -55,7 +55,7 @@
> /*
> * Function trace entry - function address and parent function address:
> */
> -FTRACE_ENTRY(function, ftrace_entry,
> +FTRACE_ENTRY_REG(function, ftrace_entry,
>
> TRACE_FN,
>
> @@ -64,7 +64,9 @@ FTRACE_ENTRY(function, ftrace_entry,
> __field( unsigned long, parent_ip )
> ),
>
> - F_printk(" %lx <-- %lx", __entry->ip, __entry->parent_ip)
> + F_printk(" %lx <-- %lx", __entry->ip, __entry->parent_ip),
> +
> + perf_ftrace_event_register
> );
>
> /* Function call entry */
> diff --git a/kernel/trace/trace_event_perf.c b/kernel/trace/trace_event_perf.c
> index d72af0b..fdeeb5c 100644
> --- a/kernel/trace/trace_event_perf.c
> +++ b/kernel/trace/trace_event_perf.c
> @@ -24,6 +24,11 @@ static int total_ref_count;
> static int perf_trace_event_perm(struct ftrace_event_call *tp_event,
> struct perf_event *p_event)
> {
> + /* The ftrace function trace is allowed only for root. */
> + if (ftrace_event_is_function(tp_event) &&
> + perf_paranoid_kernel() && !capable(CAP_SYS_ADMIN))
> + return -EPERM;
> +
> /* No tracing, just counting, so no obvious leak */
> if (!(p_event->attr.sample_type & PERF_SAMPLE_RAW))
> return 0;
> @@ -250,3 +255,84 @@ __kprobes void *perf_trace_buf_prepare(int size, unsigned short type,
> return raw_data;
> }
> EXPORT_SYMBOL_GPL(perf_trace_buf_prepare);
> +
> +#ifdef CONFIG_FUNCTION_TRACER
> +static void
> +perf_ftrace_function_call(unsigned long ip, unsigned long parent_ip)
> +{
> + struct ftrace_entry *entry;
> + struct hlist_head *head;
> + struct pt_regs regs;
> + int rctx;
> +
> +#define ENTRY_SIZE (ALIGN(sizeof(struct ftrace_entry) + sizeof(u32), \
> + sizeof(u64)) - sizeof(u32))
> +
> + BUILD_BUG_ON(ENTRY_SIZE > PERF_MAX_TRACE_SIZE);
> +
> + perf_fetch_caller_regs(®s);
> +
> + entry = perf_trace_buf_prepare(ENTRY_SIZE, TRACE_FN, NULL, &rctx);
> + if (!entry)
> + return;
> +
> + entry->ip = ip;
> + entry->parent_ip = parent_ip;
> +
> + head = this_cpu_ptr(event_function.perf_events);
> + perf_trace_buf_submit(entry, ENTRY_SIZE, rctx, 0,
> + 1, ®s, head);
> +
> +#undef ENTRY_SIZE
> +}
> +
> +static int perf_ftrace_function_register(struct perf_event *event)
> +{
> + struct ftrace_ops *ops = &event->ftrace_ops;
> +
> + ops->flags |= FTRACE_OPS_FL_CONTROL;
> + ops->func = perf_ftrace_function_call;
> + return register_ftrace_function(ops);
> +}
> +
> +static int perf_ftrace_function_unregister(struct perf_event *event)
> +{
> + struct ftrace_ops *ops = &event->ftrace_ops;
> + return unregister_ftrace_function(ops);
> +}
> +
> +static void perf_ftrace_function_enable(struct perf_event *event)
> +{
> + ftrace_function_local_enable(&event->ftrace_ops);
> +}
> +
> +static void perf_ftrace_function_disable(struct perf_event *event)
> +{
> + ftrace_function_local_disable(&event->ftrace_ops);
> +}
> +
> +int perf_ftrace_event_register(struct ftrace_event_call *call,
> + enum trace_reg type, void *data)
> +{
> + switch (type) {
> + case TRACE_REG_REGISTER:
> + case TRACE_REG_UNREGISTER:
> + break;
> + case TRACE_REG_PERF_REGISTER:
> + case TRACE_REG_PERF_UNREGISTER:
> + return 0;
> + case TRACE_REG_PERF_OPEN:
> + return perf_ftrace_function_register(data);
> + case TRACE_REG_PERF_CLOSE:
> + return perf_ftrace_function_unregister(data);
> + case TRACE_REG_PERF_ADD:
> + perf_ftrace_function_enable(data);
> + return 0;
> + case TRACE_REG_PERF_DEL:
> + perf_ftrace_function_disable(data);
> + return 0;
> + }
> +
> + return -EINVAL;
> +}
> +#endif /* CONFIG_FUNCTION_TRACER */
> diff --git a/kernel/trace/trace_export.c b/kernel/trace/trace_export.c
> index f74de86..a3dbee6 100644
> --- a/kernel/trace/trace_export.c
> +++ b/kernel/trace/trace_export.c
> @@ -184,4 +184,9 @@ __attribute__((section("_ftrace_events"))) *__event_##call = &event_##call;
> FTRACE_ENTRY_REG(call, struct_name, etype, \
> PARAMS(tstruct), PARAMS(print), NULL)
>
> +int ftrace_event_is_function(struct ftrace_event_call *call)
> +{
> + return call == &event_function;
> +}
> +
> #include "trace_entries.h"
> --
> 1.7.1
>
^ permalink raw reply [flat|nested] 17+ messages in thread
* [tip:perf/core] ftrace: Add enable/ disable ftrace_ops control interface
2012-02-15 14:51 ` [PATCH 1/7] ftrace: Add enable/disable ftrace_ops control interface Jiri Olsa
@ 2012-02-27 9:33 ` tip-bot for Jiri Olsa
0 siblings, 0 replies; 17+ messages in thread
From: tip-bot for Jiri Olsa @ 2012-02-27 9:33 UTC (permalink / raw)
To: linux-tip-commits
Cc: linux-kernel, hpa, mingo, fweisbec, rostedt, tglx, jolsa
Commit-ID: e248491ac283b516958ca9ab62c8e74b6718bca8
Gitweb: http://git.kernel.org/tip/e248491ac283b516958ca9ab62c8e74b6718bca8
Author: Jiri Olsa <jolsa@redhat.com>
AuthorDate: Wed, 15 Feb 2012 15:51:48 +0100
Committer: Steven Rostedt <rostedt@goodmis.org>
CommitDate: Tue, 21 Feb 2012 11:08:23 -0500
ftrace: Add enable/disable ftrace_ops control interface
Adding a way to temporarily enable/disable ftrace_ops. The change
follows the same way as 'global' ftrace_ops are done.
Introducing 2 global ftrace_ops - control_ops and ftrace_control_list
which take over all ftrace_ops registered with FTRACE_OPS_FL_CONTROL
flag. In addition new per cpu flag called 'disabled' is also added to
ftrace_ops to provide the control information for each cpu.
When ftrace_ops with FTRACE_OPS_FL_CONTROL is registered, it is
set as disabled for all cpus.
The ftrace_control_list contains all the registered 'control' ftrace_ops.
The control_ops provides function which iterates ftrace_control_list
and does the check for 'disabled' flag on current cpu.
Adding 3 inline functions:
ftrace_function_local_disable/ftrace_function_local_enable
- enable/disable the ftrace_ops on current cpu
ftrace_function_local_disabled
- get disabled ftrace_ops::disabled value for current cpu
Link: http://lkml.kernel.org/r/1329317514-8131-2-git-send-email-jolsa@redhat.com
Acked-by: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: Jiri Olsa <jolsa@redhat.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
include/linux/ftrace.h | 66 ++++++++++++++++++++++++++++
kernel/trace/ftrace.c | 111 +++++++++++++++++++++++++++++++++++++++++++++---
kernel/trace/trace.h | 2 +
3 files changed, 172 insertions(+), 7 deletions(-)
diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h
index f33fb3b..64a309d 100644
--- a/include/linux/ftrace.h
+++ b/include/linux/ftrace.h
@@ -31,16 +31,33 @@ ftrace_enable_sysctl(struct ctl_table *table, int write,
typedef void (*ftrace_func_t)(unsigned long ip, unsigned long parent_ip);
+/*
+ * FTRACE_OPS_FL_* bits denote the state of ftrace_ops struct and are
+ * set in the flags member.
+ *
+ * ENABLED - set/unset when ftrace_ops is registered/unregistered
+ * GLOBAL - set manualy by ftrace_ops user to denote the ftrace_ops
+ * is part of the global tracers sharing the same filter
+ * via set_ftrace_* debugfs files.
+ * DYNAMIC - set when ftrace_ops is registered to denote dynamically
+ * allocated ftrace_ops which need special care
+ * CONTROL - set manualy by ftrace_ops user to denote the ftrace_ops
+ * could be controled by following calls:
+ * ftrace_function_local_enable
+ * ftrace_function_local_disable
+ */
enum {
FTRACE_OPS_FL_ENABLED = 1 << 0,
FTRACE_OPS_FL_GLOBAL = 1 << 1,
FTRACE_OPS_FL_DYNAMIC = 1 << 2,
+ FTRACE_OPS_FL_CONTROL = 1 << 3,
};
struct ftrace_ops {
ftrace_func_t func;
struct ftrace_ops *next;
unsigned long flags;
+ int __percpu *disabled;
#ifdef CONFIG_DYNAMIC_FTRACE
struct ftrace_hash *notrace_hash;
struct ftrace_hash *filter_hash;
@@ -97,6 +114,55 @@ int register_ftrace_function(struct ftrace_ops *ops);
int unregister_ftrace_function(struct ftrace_ops *ops);
void clear_ftrace_function(void);
+/**
+ * ftrace_function_local_enable - enable controlled ftrace_ops on current cpu
+ *
+ * This function enables tracing on current cpu by decreasing
+ * the per cpu control variable.
+ * It must be called with preemption disabled and only on ftrace_ops
+ * registered with FTRACE_OPS_FL_CONTROL. If called without preemption
+ * disabled, this_cpu_ptr will complain when CONFIG_DEBUG_PREEMPT is enabled.
+ */
+static inline void ftrace_function_local_enable(struct ftrace_ops *ops)
+{
+ if (WARN_ON_ONCE(!(ops->flags & FTRACE_OPS_FL_CONTROL)))
+ return;
+
+ (*this_cpu_ptr(ops->disabled))--;
+}
+
+/**
+ * ftrace_function_local_disable - enable controlled ftrace_ops on current cpu
+ *
+ * This function enables tracing on current cpu by decreasing
+ * the per cpu control variable.
+ * It must be called with preemption disabled and only on ftrace_ops
+ * registered with FTRACE_OPS_FL_CONTROL. If called without preemption
+ * disabled, this_cpu_ptr will complain when CONFIG_DEBUG_PREEMPT is enabled.
+ */
+static inline void ftrace_function_local_disable(struct ftrace_ops *ops)
+{
+ if (WARN_ON_ONCE(!(ops->flags & FTRACE_OPS_FL_CONTROL)))
+ return;
+
+ (*this_cpu_ptr(ops->disabled))++;
+}
+
+/**
+ * ftrace_function_local_disabled - returns ftrace_ops disabled value
+ * on current cpu
+ *
+ * This function returns value of ftrace_ops::disabled on current cpu.
+ * It must be called with preemption disabled and only on ftrace_ops
+ * registered with FTRACE_OPS_FL_CONTROL. If called without preemption
+ * disabled, this_cpu_ptr will complain when CONFIG_DEBUG_PREEMPT is enabled.
+ */
+static inline int ftrace_function_local_disabled(struct ftrace_ops *ops)
+{
+ WARN_ON_ONCE(!(ops->flags & FTRACE_OPS_FL_CONTROL));
+ return *this_cpu_ptr(ops->disabled);
+}
+
extern void ftrace_stub(unsigned long a0, unsigned long a1);
#else /* !CONFIG_FUNCTION_TRACER */
diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
index d1499e9..f615f97 100644
--- a/kernel/trace/ftrace.c
+++ b/kernel/trace/ftrace.c
@@ -62,6 +62,8 @@
#define FTRACE_HASH_DEFAULT_BITS 10
#define FTRACE_HASH_MAX_BITS 12
+#define FL_GLOBAL_CONTROL_MASK (FTRACE_OPS_FL_GLOBAL | FTRACE_OPS_FL_CONTROL)
+
/* ftrace_enabled is a method to turn ftrace on or off */
int ftrace_enabled __read_mostly;
static int last_ftrace_enabled;
@@ -89,12 +91,14 @@ static struct ftrace_ops ftrace_list_end __read_mostly = {
};
static struct ftrace_ops *ftrace_global_list __read_mostly = &ftrace_list_end;
+static struct ftrace_ops *ftrace_control_list __read_mostly = &ftrace_list_end;
static struct ftrace_ops *ftrace_ops_list __read_mostly = &ftrace_list_end;
ftrace_func_t ftrace_trace_function __read_mostly = ftrace_stub;
static ftrace_func_t __ftrace_trace_function_delay __read_mostly = ftrace_stub;
ftrace_func_t __ftrace_trace_function __read_mostly = ftrace_stub;
ftrace_func_t ftrace_pid_function __read_mostly = ftrace_stub;
static struct ftrace_ops global_ops;
+static struct ftrace_ops control_ops;
static void
ftrace_ops_list_func(unsigned long ip, unsigned long parent_ip);
@@ -168,6 +172,32 @@ static void ftrace_test_stop_func(unsigned long ip, unsigned long parent_ip)
}
#endif
+static void control_ops_disable_all(struct ftrace_ops *ops)
+{
+ int cpu;
+
+ for_each_possible_cpu(cpu)
+ *per_cpu_ptr(ops->disabled, cpu) = 1;
+}
+
+static int control_ops_alloc(struct ftrace_ops *ops)
+{
+ int __percpu *disabled;
+
+ disabled = alloc_percpu(int);
+ if (!disabled)
+ return -ENOMEM;
+
+ ops->disabled = disabled;
+ control_ops_disable_all(ops);
+ return 0;
+}
+
+static void control_ops_free(struct ftrace_ops *ops)
+{
+ free_percpu(ops->disabled);
+}
+
static void update_global_ops(void)
{
ftrace_func_t func;
@@ -259,6 +289,26 @@ static int remove_ftrace_ops(struct ftrace_ops **list, struct ftrace_ops *ops)
return 0;
}
+static void add_ftrace_list_ops(struct ftrace_ops **list,
+ struct ftrace_ops *main_ops,
+ struct ftrace_ops *ops)
+{
+ int first = *list == &ftrace_list_end;
+ add_ftrace_ops(list, ops);
+ if (first)
+ add_ftrace_ops(&ftrace_ops_list, main_ops);
+}
+
+static int remove_ftrace_list_ops(struct ftrace_ops **list,
+ struct ftrace_ops *main_ops,
+ struct ftrace_ops *ops)
+{
+ int ret = remove_ftrace_ops(list, ops);
+ if (!ret && *list == &ftrace_list_end)
+ ret = remove_ftrace_ops(&ftrace_ops_list, main_ops);
+ return ret;
+}
+
static int __register_ftrace_function(struct ftrace_ops *ops)
{
if (ftrace_disabled)
@@ -270,15 +320,20 @@ static int __register_ftrace_function(struct ftrace_ops *ops)
if (WARN_ON(ops->flags & FTRACE_OPS_FL_ENABLED))
return -EBUSY;
+ /* We don't support both control and global flags set. */
+ if ((ops->flags & FL_GLOBAL_CONTROL_MASK) == FL_GLOBAL_CONTROL_MASK)
+ return -EINVAL;
+
if (!core_kernel_data((unsigned long)ops))
ops->flags |= FTRACE_OPS_FL_DYNAMIC;
if (ops->flags & FTRACE_OPS_FL_GLOBAL) {
- int first = ftrace_global_list == &ftrace_list_end;
- add_ftrace_ops(&ftrace_global_list, ops);
+ add_ftrace_list_ops(&ftrace_global_list, &global_ops, ops);
ops->flags |= FTRACE_OPS_FL_ENABLED;
- if (first)
- add_ftrace_ops(&ftrace_ops_list, &global_ops);
+ } else if (ops->flags & FTRACE_OPS_FL_CONTROL) {
+ if (control_ops_alloc(ops))
+ return -ENOMEM;
+ add_ftrace_list_ops(&ftrace_control_list, &control_ops, ops);
} else
add_ftrace_ops(&ftrace_ops_list, ops);
@@ -302,11 +357,23 @@ static int __unregister_ftrace_function(struct ftrace_ops *ops)
return -EINVAL;
if (ops->flags & FTRACE_OPS_FL_GLOBAL) {
- ret = remove_ftrace_ops(&ftrace_global_list, ops);
- if (!ret && ftrace_global_list == &ftrace_list_end)
- ret = remove_ftrace_ops(&ftrace_ops_list, &global_ops);
+ ret = remove_ftrace_list_ops(&ftrace_global_list,
+ &global_ops, ops);
if (!ret)
ops->flags &= ~FTRACE_OPS_FL_ENABLED;
+ } else if (ops->flags & FTRACE_OPS_FL_CONTROL) {
+ ret = remove_ftrace_list_ops(&ftrace_control_list,
+ &control_ops, ops);
+ if (!ret) {
+ /*
+ * The ftrace_ops is now removed from the list,
+ * so there'll be no new users. We must ensure
+ * all current users are done before we free
+ * the control data.
+ */
+ synchronize_sched();
+ control_ops_free(ops);
+ }
} else
ret = remove_ftrace_ops(&ftrace_ops_list, ops);
@@ -3874,6 +3941,36 @@ ftrace_ops_test(struct ftrace_ops *ops, unsigned long ip)
#endif /* CONFIG_DYNAMIC_FTRACE */
static void
+ftrace_ops_control_func(unsigned long ip, unsigned long parent_ip)
+{
+ struct ftrace_ops *op;
+
+ if (unlikely(trace_recursion_test(TRACE_CONTROL_BIT)))
+ return;
+
+ /*
+ * Some of the ops may be dynamically allocated,
+ * they must be freed after a synchronize_sched().
+ */
+ preempt_disable_notrace();
+ trace_recursion_set(TRACE_CONTROL_BIT);
+ op = rcu_dereference_raw(ftrace_control_list);
+ while (op != &ftrace_list_end) {
+ if (!ftrace_function_local_disabled(op) &&
+ ftrace_ops_test(op, ip))
+ op->func(ip, parent_ip);
+
+ op = rcu_dereference_raw(op->next);
+ };
+ trace_recursion_clear(TRACE_CONTROL_BIT);
+ preempt_enable_notrace();
+}
+
+static struct ftrace_ops control_ops = {
+ .func = ftrace_ops_control_func,
+};
+
+static void
ftrace_ops_list_func(unsigned long ip, unsigned long parent_ip)
{
struct ftrace_ops *op;
diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h
index b93ecba..55c6ea0 100644
--- a/kernel/trace/trace.h
+++ b/kernel/trace/trace.h
@@ -288,6 +288,8 @@ struct tracer {
/* for function tracing recursion */
#define TRACE_INTERNAL_BIT (1<<11)
#define TRACE_GLOBAL_BIT (1<<12)
+#define TRACE_CONTROL_BIT (1<<13)
+
/*
* Abuse of the trace_recursion.
* As we need a way to maintain state if we are tracing the function
^ permalink raw reply related [flat|nested] 17+ messages in thread
* [tip:perf/core] ftrace, perf: Add open/ close tracepoint perf registration actions
2012-02-15 14:51 ` [PATCH 2/7] ftrace, perf: Add open/close tracepoint perf registration actions Jiri Olsa
@ 2012-02-27 9:34 ` tip-bot for Jiri Olsa
0 siblings, 0 replies; 17+ messages in thread
From: tip-bot for Jiri Olsa @ 2012-02-27 9:34 UTC (permalink / raw)
To: linux-tip-commits
Cc: linux-kernel, hpa, mingo, fweisbec, rostedt, tglx, jolsa
Commit-ID: ceec0b6fc7cd43b38a40c2d40223f9cd0616f0cd
Gitweb: http://git.kernel.org/tip/ceec0b6fc7cd43b38a40c2d40223f9cd0616f0cd
Author: Jiri Olsa <jolsa@redhat.com>
AuthorDate: Wed, 15 Feb 2012 15:51:49 +0100
Committer: Steven Rostedt <rostedt@goodmis.org>
CommitDate: Tue, 21 Feb 2012 11:08:24 -0500
ftrace, perf: Add open/close tracepoint perf registration actions
Adding TRACE_REG_PERF_OPEN and TRACE_REG_PERF_CLOSE to differentiate
register/unregister from open/close actions.
The register/unregister actions are invoked for the first/last
tracepoint user when opening/closing the event.
The open/close actions are invoked for each tracepoint user when
opening/closing the event.
Link: http://lkml.kernel.org/r/1329317514-8131-3-git-send-email-jolsa@redhat.com
Acked-by: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: Jiri Olsa <jolsa@redhat.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
include/linux/ftrace_event.h | 6 +-
kernel/trace/trace_event_perf.c | 116 +++++++++++++++++++++++++--------------
kernel/trace/trace_events.c | 10 ++-
kernel/trace/trace_kprobe.c | 6 ++-
kernel/trace/trace_syscalls.c | 14 +++-
5 files changed, 101 insertions(+), 51 deletions(-)
diff --git a/include/linux/ftrace_event.h b/include/linux/ftrace_event.h
index c3da42d..195e360 100644
--- a/include/linux/ftrace_event.h
+++ b/include/linux/ftrace_event.h
@@ -146,6 +146,8 @@ enum trace_reg {
TRACE_REG_UNREGISTER,
TRACE_REG_PERF_REGISTER,
TRACE_REG_PERF_UNREGISTER,
+ TRACE_REG_PERF_OPEN,
+ TRACE_REG_PERF_CLOSE,
};
struct ftrace_event_call;
@@ -157,7 +159,7 @@ struct ftrace_event_class {
void *perf_probe;
#endif
int (*reg)(struct ftrace_event_call *event,
- enum trace_reg type);
+ enum trace_reg type, void *data);
int (*define_fields)(struct ftrace_event_call *);
struct list_head *(*get_fields)(struct ftrace_event_call *);
struct list_head fields;
@@ -165,7 +167,7 @@ struct ftrace_event_class {
};
extern int ftrace_event_reg(struct ftrace_event_call *event,
- enum trace_reg type);
+ enum trace_reg type, void *data);
enum {
TRACE_EVENT_FL_ENABLED_BIT,
diff --git a/kernel/trace/trace_event_perf.c b/kernel/trace/trace_event_perf.c
index 19a359d..0cfcc37 100644
--- a/kernel/trace/trace_event_perf.c
+++ b/kernel/trace/trace_event_perf.c
@@ -44,23 +44,17 @@ static int perf_trace_event_perm(struct ftrace_event_call *tp_event,
return 0;
}
-static int perf_trace_event_init(struct ftrace_event_call *tp_event,
- struct perf_event *p_event)
+static int perf_trace_event_reg(struct ftrace_event_call *tp_event,
+ struct perf_event *p_event)
{
struct hlist_head __percpu *list;
- int ret;
+ int ret = -ENOMEM;
int cpu;
- ret = perf_trace_event_perm(tp_event, p_event);
- if (ret)
- return ret;
-
p_event->tp_event = tp_event;
if (tp_event->perf_refcount++ > 0)
return 0;
- ret = -ENOMEM;
-
list = alloc_percpu(struct hlist_head);
if (!list)
goto fail;
@@ -83,7 +77,7 @@ static int perf_trace_event_init(struct ftrace_event_call *tp_event,
}
}
- ret = tp_event->class->reg(tp_event, TRACE_REG_PERF_REGISTER);
+ ret = tp_event->class->reg(tp_event, TRACE_REG_PERF_REGISTER, NULL);
if (ret)
goto fail;
@@ -108,6 +102,69 @@ fail:
return ret;
}
+static void perf_trace_event_unreg(struct perf_event *p_event)
+{
+ struct ftrace_event_call *tp_event = p_event->tp_event;
+ int i;
+
+ if (--tp_event->perf_refcount > 0)
+ goto out;
+
+ tp_event->class->reg(tp_event, TRACE_REG_PERF_UNREGISTER, NULL);
+
+ /*
+ * Ensure our callback won't be called anymore. The buffers
+ * will be freed after that.
+ */
+ tracepoint_synchronize_unregister();
+
+ free_percpu(tp_event->perf_events);
+ tp_event->perf_events = NULL;
+
+ if (!--total_ref_count) {
+ for (i = 0; i < PERF_NR_CONTEXTS; i++) {
+ free_percpu(perf_trace_buf[i]);
+ perf_trace_buf[i] = NULL;
+ }
+ }
+out:
+ module_put(tp_event->mod);
+}
+
+static int perf_trace_event_open(struct perf_event *p_event)
+{
+ struct ftrace_event_call *tp_event = p_event->tp_event;
+ return tp_event->class->reg(tp_event, TRACE_REG_PERF_OPEN, p_event);
+}
+
+static void perf_trace_event_close(struct perf_event *p_event)
+{
+ struct ftrace_event_call *tp_event = p_event->tp_event;
+ tp_event->class->reg(tp_event, TRACE_REG_PERF_CLOSE, p_event);
+}
+
+static int perf_trace_event_init(struct ftrace_event_call *tp_event,
+ struct perf_event *p_event)
+{
+ int ret;
+
+ ret = perf_trace_event_perm(tp_event, p_event);
+ if (ret)
+ return ret;
+
+ ret = perf_trace_event_reg(tp_event, p_event);
+ if (ret)
+ return ret;
+
+ ret = perf_trace_event_open(p_event);
+ if (ret) {
+ perf_trace_event_unreg(p_event);
+ return ret;
+ }
+
+ return 0;
+}
+
int perf_trace_init(struct perf_event *p_event)
{
struct ftrace_event_call *tp_event;
@@ -130,6 +187,14 @@ int perf_trace_init(struct perf_event *p_event)
return ret;
}
+void perf_trace_destroy(struct perf_event *p_event)
+{
+ mutex_lock(&event_mutex);
+ perf_trace_event_close(p_event);
+ perf_trace_event_unreg(p_event);
+ mutex_unlock(&event_mutex);
+}
+
int perf_trace_add(struct perf_event *p_event, int flags)
{
struct ftrace_event_call *tp_event = p_event->tp_event;
@@ -154,37 +219,6 @@ void perf_trace_del(struct perf_event *p_event, int flags)
hlist_del_rcu(&p_event->hlist_entry);
}
-void perf_trace_destroy(struct perf_event *p_event)
-{
- struct ftrace_event_call *tp_event = p_event->tp_event;
- int i;
-
- mutex_lock(&event_mutex);
- if (--tp_event->perf_refcount > 0)
- goto out;
-
- tp_event->class->reg(tp_event, TRACE_REG_PERF_UNREGISTER);
-
- /*
- * Ensure our callback won't be called anymore. The buffers
- * will be freed after that.
- */
- tracepoint_synchronize_unregister();
-
- free_percpu(tp_event->perf_events);
- tp_event->perf_events = NULL;
-
- if (!--total_ref_count) {
- for (i = 0; i < PERF_NR_CONTEXTS; i++) {
- free_percpu(perf_trace_buf[i]);
- perf_trace_buf[i] = NULL;
- }
- }
-out:
- module_put(tp_event->mod);
- mutex_unlock(&event_mutex);
-}
-
__kprobes void *perf_trace_buf_prepare(int size, unsigned short type,
struct pt_regs *regs, int *rctxp)
{
diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c
index c212a7f..5138fea 100644
--- a/kernel/trace/trace_events.c
+++ b/kernel/trace/trace_events.c
@@ -147,7 +147,8 @@ int trace_event_raw_init(struct ftrace_event_call *call)
}
EXPORT_SYMBOL_GPL(trace_event_raw_init);
-int ftrace_event_reg(struct ftrace_event_call *call, enum trace_reg type)
+int ftrace_event_reg(struct ftrace_event_call *call,
+ enum trace_reg type, void *data)
{
switch (type) {
case TRACE_REG_REGISTER:
@@ -170,6 +171,9 @@ int ftrace_event_reg(struct ftrace_event_call *call, enum trace_reg type)
call->class->perf_probe,
call);
return 0;
+ case TRACE_REG_PERF_OPEN:
+ case TRACE_REG_PERF_CLOSE:
+ return 0;
#endif
}
return 0;
@@ -209,7 +213,7 @@ static int ftrace_event_enable_disable(struct ftrace_event_call *call,
tracing_stop_cmdline_record();
call->flags &= ~TRACE_EVENT_FL_RECORDED_CMD;
}
- call->class->reg(call, TRACE_REG_UNREGISTER);
+ call->class->reg(call, TRACE_REG_UNREGISTER, NULL);
}
break;
case 1:
@@ -218,7 +222,7 @@ static int ftrace_event_enable_disable(struct ftrace_event_call *call,
tracing_start_cmdline_record();
call->flags |= TRACE_EVENT_FL_RECORDED_CMD;
}
- ret = call->class->reg(call, TRACE_REG_REGISTER);
+ ret = call->class->reg(call, TRACE_REG_REGISTER, NULL);
if (ret) {
tracing_stop_cmdline_record();
pr_info("event trace: Could not enable event "
diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c
index 00d527c..5667f89 100644
--- a/kernel/trace/trace_kprobe.c
+++ b/kernel/trace/trace_kprobe.c
@@ -1892,7 +1892,8 @@ static __kprobes void kretprobe_perf_func(struct kretprobe_instance *ri,
#endif /* CONFIG_PERF_EVENTS */
static __kprobes
-int kprobe_register(struct ftrace_event_call *event, enum trace_reg type)
+int kprobe_register(struct ftrace_event_call *event,
+ enum trace_reg type, void *data)
{
struct trace_probe *tp = (struct trace_probe *)event->data;
@@ -1909,6 +1910,9 @@ int kprobe_register(struct ftrace_event_call *event, enum trace_reg type)
case TRACE_REG_PERF_UNREGISTER:
disable_trace_probe(tp, TP_FLAG_PROFILE);
return 0;
+ case TRACE_REG_PERF_OPEN:
+ case TRACE_REG_PERF_CLOSE:
+ return 0;
#endif
}
return 0;
diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c
index 4350015..e23515f 100644
--- a/kernel/trace/trace_syscalls.c
+++ b/kernel/trace/trace_syscalls.c
@@ -17,9 +17,9 @@ static DECLARE_BITMAP(enabled_enter_syscalls, NR_syscalls);
static DECLARE_BITMAP(enabled_exit_syscalls, NR_syscalls);
static int syscall_enter_register(struct ftrace_event_call *event,
- enum trace_reg type);
+ enum trace_reg type, void *data);
static int syscall_exit_register(struct ftrace_event_call *event,
- enum trace_reg type);
+ enum trace_reg type, void *data);
static int syscall_enter_define_fields(struct ftrace_event_call *call);
static int syscall_exit_define_fields(struct ftrace_event_call *call);
@@ -649,7 +649,7 @@ void perf_sysexit_disable(struct ftrace_event_call *call)
#endif /* CONFIG_PERF_EVENTS */
static int syscall_enter_register(struct ftrace_event_call *event,
- enum trace_reg type)
+ enum trace_reg type, void *data)
{
switch (type) {
case TRACE_REG_REGISTER:
@@ -664,13 +664,16 @@ static int syscall_enter_register(struct ftrace_event_call *event,
case TRACE_REG_PERF_UNREGISTER:
perf_sysenter_disable(event);
return 0;
+ case TRACE_REG_PERF_OPEN:
+ case TRACE_REG_PERF_CLOSE:
+ return 0;
#endif
}
return 0;
}
static int syscall_exit_register(struct ftrace_event_call *event,
- enum trace_reg type)
+ enum trace_reg type, void *data)
{
switch (type) {
case TRACE_REG_REGISTER:
@@ -685,6 +688,9 @@ static int syscall_exit_register(struct ftrace_event_call *event,
case TRACE_REG_PERF_UNREGISTER:
perf_sysexit_disable(event);
return 0;
+ case TRACE_REG_PERF_OPEN:
+ case TRACE_REG_PERF_CLOSE:
+ return 0;
#endif
}
return 0;
^ permalink raw reply related [flat|nested] 17+ messages in thread
* [tip:perf/core] ftrace, perf: Add add/ del tracepoint perf registration actions
2012-02-15 14:51 ` [PATCH 3/7] ftrace, perf: Add add/del " Jiri Olsa
@ 2012-02-27 9:34 ` tip-bot for Jiri Olsa
0 siblings, 0 replies; 17+ messages in thread
From: tip-bot for Jiri Olsa @ 2012-02-27 9:34 UTC (permalink / raw)
To: linux-tip-commits
Cc: linux-kernel, hpa, mingo, fweisbec, rostedt, tglx, jolsa
Commit-ID: 489c75c3b333dfda4c8d2b7ad1b00e5da024bfa7
Gitweb: http://git.kernel.org/tip/489c75c3b333dfda4c8d2b7ad1b00e5da024bfa7
Author: Jiri Olsa <jolsa@redhat.com>
AuthorDate: Wed, 15 Feb 2012 15:51:50 +0100
Committer: Steven Rostedt <rostedt@goodmis.org>
CommitDate: Tue, 21 Feb 2012 11:08:25 -0500
ftrace, perf: Add add/del tracepoint perf registration actions
Adding TRACE_REG_PERF_ADD and TRACE_REG_PERF_DEL to handle
perf event schedule in/out actions.
The add action is invoked for when the perf event is scheduled in,
while the del action is invoked when the event is scheduled out.
Link: http://lkml.kernel.org/r/1329317514-8131-4-git-send-email-jolsa@redhat.com
Acked-by: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: Jiri Olsa <jolsa@redhat.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
include/linux/ftrace_event.h | 2 ++
kernel/trace/trace_event_perf.c | 4 +++-
kernel/trace/trace_events.c | 2 ++
kernel/trace/trace_kprobe.c | 2 ++
kernel/trace/trace_syscalls.c | 4 ++++
5 files changed, 13 insertions(+), 1 deletions(-)
diff --git a/include/linux/ftrace_event.h b/include/linux/ftrace_event.h
index 195e360..2bf677c 100644
--- a/include/linux/ftrace_event.h
+++ b/include/linux/ftrace_event.h
@@ -148,6 +148,8 @@ enum trace_reg {
TRACE_REG_PERF_UNREGISTER,
TRACE_REG_PERF_OPEN,
TRACE_REG_PERF_CLOSE,
+ TRACE_REG_PERF_ADD,
+ TRACE_REG_PERF_DEL,
};
struct ftrace_event_call;
diff --git a/kernel/trace/trace_event_perf.c b/kernel/trace/trace_event_perf.c
index 0cfcc37..d72af0b 100644
--- a/kernel/trace/trace_event_perf.c
+++ b/kernel/trace/trace_event_perf.c
@@ -211,12 +211,14 @@ int perf_trace_add(struct perf_event *p_event, int flags)
list = this_cpu_ptr(pcpu_list);
hlist_add_head_rcu(&p_event->hlist_entry, list);
- return 0;
+ return tp_event->class->reg(tp_event, TRACE_REG_PERF_ADD, p_event);
}
void perf_trace_del(struct perf_event *p_event, int flags)
{
+ struct ftrace_event_call *tp_event = p_event->tp_event;
hlist_del_rcu(&p_event->hlist_entry);
+ tp_event->class->reg(tp_event, TRACE_REG_PERF_DEL, p_event);
}
__kprobes void *perf_trace_buf_prepare(int size, unsigned short type,
diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c
index 5138fea..079a93a 100644
--- a/kernel/trace/trace_events.c
+++ b/kernel/trace/trace_events.c
@@ -173,6 +173,8 @@ int ftrace_event_reg(struct ftrace_event_call *call,
return 0;
case TRACE_REG_PERF_OPEN:
case TRACE_REG_PERF_CLOSE:
+ case TRACE_REG_PERF_ADD:
+ case TRACE_REG_PERF_DEL:
return 0;
#endif
}
diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c
index 5667f89..580a05e 100644
--- a/kernel/trace/trace_kprobe.c
+++ b/kernel/trace/trace_kprobe.c
@@ -1912,6 +1912,8 @@ int kprobe_register(struct ftrace_event_call *event,
return 0;
case TRACE_REG_PERF_OPEN:
case TRACE_REG_PERF_CLOSE:
+ case TRACE_REG_PERF_ADD:
+ case TRACE_REG_PERF_DEL:
return 0;
#endif
}
diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c
index e23515f..96fc733 100644
--- a/kernel/trace/trace_syscalls.c
+++ b/kernel/trace/trace_syscalls.c
@@ -666,6 +666,8 @@ static int syscall_enter_register(struct ftrace_event_call *event,
return 0;
case TRACE_REG_PERF_OPEN:
case TRACE_REG_PERF_CLOSE:
+ case TRACE_REG_PERF_ADD:
+ case TRACE_REG_PERF_DEL:
return 0;
#endif
}
@@ -690,6 +692,8 @@ static int syscall_exit_register(struct ftrace_event_call *event,
return 0;
case TRACE_REG_PERF_OPEN:
case TRACE_REG_PERF_CLOSE:
+ case TRACE_REG_PERF_ADD:
+ case TRACE_REG_PERF_DEL:
return 0;
#endif
}
^ permalink raw reply related [flat|nested] 17+ messages in thread
* [tip:perf/core] ftrace: Add FTRACE_ENTRY_REG macro to allow event registration
2012-02-15 14:51 ` [PATCH 4/7] ftrace: Add FTRACE_ENTRY_REG macro to allow event registration Jiri Olsa
2012-02-15 15:13 ` Frederic Weisbecker
@ 2012-02-27 9:35 ` tip-bot for Jiri Olsa
1 sibling, 0 replies; 17+ messages in thread
From: tip-bot for Jiri Olsa @ 2012-02-27 9:35 UTC (permalink / raw)
To: linux-tip-commits
Cc: linux-kernel, hpa, mingo, fweisbec, rostedt, tglx, jolsa
Commit-ID: e59a0bff3ecf389951e3c9378ddfd00f6448bfaa
Gitweb: http://git.kernel.org/tip/e59a0bff3ecf389951e3c9378ddfd00f6448bfaa
Author: Jiri Olsa <jolsa@redhat.com>
AuthorDate: Wed, 15 Feb 2012 15:51:51 +0100
Committer: Steven Rostedt <rostedt@goodmis.org>
CommitDate: Tue, 21 Feb 2012 11:08:26 -0500
ftrace: Add FTRACE_ENTRY_REG macro to allow event registration
Adding FTRACE_ENTRY_REG macro so particular ftrace entries
could specify registration function and thus become accesible
via perf.
This will be used in upcomming patch for function trace.
Link: http://lkml.kernel.org/r/1329317514-8131-5-git-send-email-jolsa@redhat.com
Acked-by: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: Jiri Olsa <jolsa@redhat.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
kernel/trace/trace.h | 4 ++++
kernel/trace/trace_export.c | 18 ++++++++++++++++--
2 files changed, 20 insertions(+), 2 deletions(-)
diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h
index 55c6ea0..638476a 100644
--- a/kernel/trace/trace.h
+++ b/kernel/trace/trace.h
@@ -68,6 +68,10 @@ enum trace_type {
#undef FTRACE_ENTRY_DUP
#define FTRACE_ENTRY_DUP(name, name_struct, id, tstruct, printk)
+#undef FTRACE_ENTRY_REG
+#define FTRACE_ENTRY_REG(name, struct_name, id, tstruct, print, regfn) \
+ FTRACE_ENTRY(name, struct_name, id, PARAMS(tstruct), PARAMS(print))
+
#include "trace_entries.h"
/*
diff --git a/kernel/trace/trace_export.c b/kernel/trace/trace_export.c
index bbeec31..f74de86 100644
--- a/kernel/trace/trace_export.c
+++ b/kernel/trace/trace_export.c
@@ -18,6 +18,14 @@
#undef TRACE_SYSTEM
#define TRACE_SYSTEM ftrace
+/*
+ * The FTRACE_ENTRY_REG macro allows ftrace entry to define register
+ * function and thus become accesible via perf.
+ */
+#undef FTRACE_ENTRY_REG
+#define FTRACE_ENTRY_REG(name, struct_name, id, tstruct, print, regfn) \
+ FTRACE_ENTRY(name, struct_name, id, PARAMS(tstruct), PARAMS(print))
+
/* not needed for this file */
#undef __field_struct
#define __field_struct(type, item)
@@ -152,13 +160,14 @@ ftrace_define_fields_##name(struct ftrace_event_call *event_call) \
#undef F_printk
#define F_printk(fmt, args...) #fmt ", " __stringify(args)
-#undef FTRACE_ENTRY
-#define FTRACE_ENTRY(call, struct_name, etype, tstruct, print) \
+#undef FTRACE_ENTRY_REG
+#define FTRACE_ENTRY_REG(call, struct_name, etype, tstruct, print, regfn)\
\
struct ftrace_event_class event_class_ftrace_##call = { \
.system = __stringify(TRACE_SYSTEM), \
.define_fields = ftrace_define_fields_##call, \
.fields = LIST_HEAD_INIT(event_class_ftrace_##call.fields),\
+ .reg = regfn, \
}; \
\
struct ftrace_event_call __used event_##call = { \
@@ -170,4 +179,9 @@ struct ftrace_event_call __used event_##call = { \
struct ftrace_event_call __used \
__attribute__((section("_ftrace_events"))) *__event_##call = &event_##call;
+#undef FTRACE_ENTRY
+#define FTRACE_ENTRY(call, struct_name, etype, tstruct, print) \
+ FTRACE_ENTRY_REG(call, struct_name, etype, \
+ PARAMS(tstruct), PARAMS(print), NULL)
+
#include "trace_entries.h"
^ permalink raw reply related [flat|nested] 17+ messages in thread
* [tip:perf/core] ftrace, perf: Add support to use function tracepoint in perf
2012-02-15 14:51 ` [PATCH 5/7] ftrace, perf: Add support to use function tracepoint in perf Jiri Olsa
2012-02-15 16:03 ` Frederic Weisbecker
@ 2012-02-27 9:36 ` tip-bot for Jiri Olsa
1 sibling, 0 replies; 17+ messages in thread
From: tip-bot for Jiri Olsa @ 2012-02-27 9:36 UTC (permalink / raw)
To: linux-tip-commits
Cc: linux-kernel, hpa, mingo, fweisbec, rostedt, tglx, jolsa
Commit-ID: ced39002f5ea736b716ae233fb68b26d59783912
Gitweb: http://git.kernel.org/tip/ced39002f5ea736b716ae233fb68b26d59783912
Author: Jiri Olsa <jolsa@redhat.com>
AuthorDate: Wed, 15 Feb 2012 15:51:52 +0100
Committer: Steven Rostedt <rostedt@goodmis.org>
CommitDate: Tue, 21 Feb 2012 11:08:27 -0500
ftrace, perf: Add support to use function tracepoint in perf
Adding perf registration support for the ftrace function event,
so it is now possible to register it via perf interface.
The perf_event struct statically contains ftrace_ops as a handle
for function tracer. The function tracer is registered/unregistered
in open/close actions.
To be efficient, we enable/disable ftrace_ops each time the traced
process is scheduled in/out (via TRACE_REG_PERF_(ADD|DELL) handlers).
This way tracing is enabled only when the process is running.
Intentionally using this way instead of the event's hw state
PERF_HES_STOPPED, which would not disable the ftrace_ops.
It is now possible to use function trace within perf commands
like:
perf record -e ftrace:function ls
perf stat -e ftrace:function ls
Allowed only for root.
Link: http://lkml.kernel.org/r/1329317514-8131-6-git-send-email-jolsa@redhat.com
Acked-by: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: Jiri Olsa <jolsa@redhat.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
include/linux/perf_event.h | 3 +
kernel/trace/trace.h | 11 +++++
kernel/trace/trace_entries.h | 6 ++-
kernel/trace/trace_event_perf.c | 86 +++++++++++++++++++++++++++++++++++++++
kernel/trace/trace_export.c | 5 ++
5 files changed, 109 insertions(+), 2 deletions(-)
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 412b790..92a056f 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -859,6 +859,9 @@ struct perf_event {
#ifdef CONFIG_EVENT_TRACING
struct ftrace_event_call *tp_event;
struct event_filter *filter;
+#ifdef CONFIG_FUNCTION_TRACER
+ struct ftrace_ops ftrace_ops;
+#endif
#endif
#ifdef CONFIG_CGROUP_PERF
diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h
index 638476a..76a1c50 100644
--- a/kernel/trace/trace.h
+++ b/kernel/trace/trace.h
@@ -595,6 +595,8 @@ static inline int ftrace_trace_task(struct task_struct *task)
static inline int ftrace_is_dead(void) { return 0; }
#endif
+int ftrace_event_is_function(struct ftrace_event_call *call);
+
/*
* struct trace_parser - servers for reading the user input separated by spaces
* @cont: set if the input is not complete - no final space char was found
@@ -832,4 +834,13 @@ extern const char *__stop___trace_bprintk_fmt[];
FTRACE_ENTRY(call, struct_name, id, PARAMS(tstruct), PARAMS(print))
#include "trace_entries.h"
+#ifdef CONFIG_PERF_EVENTS
+#ifdef CONFIG_FUNCTION_TRACER
+int perf_ftrace_event_register(struct ftrace_event_call *call,
+ enum trace_reg type, void *data);
+#else
+#define perf_ftrace_event_register NULL
+#endif /* CONFIG_FUNCTION_TRACER */
+#endif /* CONFIG_PERF_EVENTS */
+
#endif /* _LINUX_KERNEL_TRACE_H */
diff --git a/kernel/trace/trace_entries.h b/kernel/trace/trace_entries.h
index 9336590..47db7ed 100644
--- a/kernel/trace/trace_entries.h
+++ b/kernel/trace/trace_entries.h
@@ -55,7 +55,7 @@
/*
* Function trace entry - function address and parent function address:
*/
-FTRACE_ENTRY(function, ftrace_entry,
+FTRACE_ENTRY_REG(function, ftrace_entry,
TRACE_FN,
@@ -64,7 +64,9 @@ FTRACE_ENTRY(function, ftrace_entry,
__field( unsigned long, parent_ip )
),
- F_printk(" %lx <-- %lx", __entry->ip, __entry->parent_ip)
+ F_printk(" %lx <-- %lx", __entry->ip, __entry->parent_ip),
+
+ perf_ftrace_event_register
);
/* Function call entry */
diff --git a/kernel/trace/trace_event_perf.c b/kernel/trace/trace_event_perf.c
index d72af0b..fdeeb5c 100644
--- a/kernel/trace/trace_event_perf.c
+++ b/kernel/trace/trace_event_perf.c
@@ -24,6 +24,11 @@ static int total_ref_count;
static int perf_trace_event_perm(struct ftrace_event_call *tp_event,
struct perf_event *p_event)
{
+ /* The ftrace function trace is allowed only for root. */
+ if (ftrace_event_is_function(tp_event) &&
+ perf_paranoid_kernel() && !capable(CAP_SYS_ADMIN))
+ return -EPERM;
+
/* No tracing, just counting, so no obvious leak */
if (!(p_event->attr.sample_type & PERF_SAMPLE_RAW))
return 0;
@@ -250,3 +255,84 @@ __kprobes void *perf_trace_buf_prepare(int size, unsigned short type,
return raw_data;
}
EXPORT_SYMBOL_GPL(perf_trace_buf_prepare);
+
+#ifdef CONFIG_FUNCTION_TRACER
+static void
+perf_ftrace_function_call(unsigned long ip, unsigned long parent_ip)
+{
+ struct ftrace_entry *entry;
+ struct hlist_head *head;
+ struct pt_regs regs;
+ int rctx;
+
+#define ENTRY_SIZE (ALIGN(sizeof(struct ftrace_entry) + sizeof(u32), \
+ sizeof(u64)) - sizeof(u32))
+
+ BUILD_BUG_ON(ENTRY_SIZE > PERF_MAX_TRACE_SIZE);
+
+ perf_fetch_caller_regs(®s);
+
+ entry = perf_trace_buf_prepare(ENTRY_SIZE, TRACE_FN, NULL, &rctx);
+ if (!entry)
+ return;
+
+ entry->ip = ip;
+ entry->parent_ip = parent_ip;
+
+ head = this_cpu_ptr(event_function.perf_events);
+ perf_trace_buf_submit(entry, ENTRY_SIZE, rctx, 0,
+ 1, ®s, head);
+
+#undef ENTRY_SIZE
+}
+
+static int perf_ftrace_function_register(struct perf_event *event)
+{
+ struct ftrace_ops *ops = &event->ftrace_ops;
+
+ ops->flags |= FTRACE_OPS_FL_CONTROL;
+ ops->func = perf_ftrace_function_call;
+ return register_ftrace_function(ops);
+}
+
+static int perf_ftrace_function_unregister(struct perf_event *event)
+{
+ struct ftrace_ops *ops = &event->ftrace_ops;
+ return unregister_ftrace_function(ops);
+}
+
+static void perf_ftrace_function_enable(struct perf_event *event)
+{
+ ftrace_function_local_enable(&event->ftrace_ops);
+}
+
+static void perf_ftrace_function_disable(struct perf_event *event)
+{
+ ftrace_function_local_disable(&event->ftrace_ops);
+}
+
+int perf_ftrace_event_register(struct ftrace_event_call *call,
+ enum trace_reg type, void *data)
+{
+ switch (type) {
+ case TRACE_REG_REGISTER:
+ case TRACE_REG_UNREGISTER:
+ break;
+ case TRACE_REG_PERF_REGISTER:
+ case TRACE_REG_PERF_UNREGISTER:
+ return 0;
+ case TRACE_REG_PERF_OPEN:
+ return perf_ftrace_function_register(data);
+ case TRACE_REG_PERF_CLOSE:
+ return perf_ftrace_function_unregister(data);
+ case TRACE_REG_PERF_ADD:
+ perf_ftrace_function_enable(data);
+ return 0;
+ case TRACE_REG_PERF_DEL:
+ perf_ftrace_function_disable(data);
+ return 0;
+ }
+
+ return -EINVAL;
+}
+#endif /* CONFIG_FUNCTION_TRACER */
diff --git a/kernel/trace/trace_export.c b/kernel/trace/trace_export.c
index f74de86..a3dbee6 100644
--- a/kernel/trace/trace_export.c
+++ b/kernel/trace/trace_export.c
@@ -184,4 +184,9 @@ __attribute__((section("_ftrace_events"))) *__event_##call = &event_##call;
FTRACE_ENTRY_REG(call, struct_name, etype, \
PARAMS(tstruct), PARAMS(print), NULL)
+int ftrace_event_is_function(struct ftrace_event_call *call)
+{
+ return call == &event_function;
+}
+
#include "trace_entries.h"
^ permalink raw reply related [flat|nested] 17+ messages in thread
* [tip:perf/core] ftrace: Allow to specify filter field type for ftrace events
2012-02-15 14:51 ` [PATCH 6/7] ftrace: Allow to specify filter field type for ftrace events Jiri Olsa
@ 2012-02-27 9:37 ` tip-bot for Jiri Olsa
0 siblings, 0 replies; 17+ messages in thread
From: tip-bot for Jiri Olsa @ 2012-02-27 9:37 UTC (permalink / raw)
To: linux-tip-commits; +Cc: linux-kernel, hpa, mingo, rostedt, tglx, jolsa
Commit-ID: 02aa3162edaa166a01d193f80ccde890be8b55da
Gitweb: http://git.kernel.org/tip/02aa3162edaa166a01d193f80ccde890be8b55da
Author: Jiri Olsa <jolsa@redhat.com>
AuthorDate: Wed, 15 Feb 2012 15:51:53 +0100
Committer: Steven Rostedt <rostedt@goodmis.org>
CommitDate: Tue, 21 Feb 2012 11:08:29 -0500
ftrace: Allow to specify filter field type for ftrace events
Adding FILTER_TRACE_FN event field type for function tracepoint
event, so it can be properly recognized within filtering code.
Currently all fields of ftrace subsystem events share the common
field type FILTER_OTHER. Since the function trace fields need
special care within the filtering code we need to recognize it
properly, hence adding the FILTER_TRACE_FN event type.
Adding filter parameter to the FTRACE_ENTRY macro, to specify the
filter field type for the event.
Link: http://lkml.kernel.org/r/1329317514-8131-7-git-send-email-jolsa@redhat.com
Signed-off-by: Jiri Olsa <jolsa@redhat.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
include/linux/ftrace_event.h | 1 +
kernel/trace/trace.h | 23 +++++++++-------
kernel/trace/trace_entries.h | 48 ++++++++++++++++++++++++---------
kernel/trace/trace_events_filter.c | 7 ++++-
kernel/trace/trace_export.c | 51 +++++++++++++++++++----------------
5 files changed, 83 insertions(+), 47 deletions(-)
diff --git a/include/linux/ftrace_event.h b/include/linux/ftrace_event.h
index 2bf677c..dd478fc 100644
--- a/include/linux/ftrace_event.h
+++ b/include/linux/ftrace_event.h
@@ -245,6 +245,7 @@ enum {
FILTER_STATIC_STRING,
FILTER_DYN_STRING,
FILTER_PTR_STRING,
+ FILTER_TRACE_FN,
};
#define EVENT_STORAGE_SIZE 128
diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h
index 76a1c50..29f93cd 100644
--- a/kernel/trace/trace.h
+++ b/kernel/trace/trace.h
@@ -56,21 +56,23 @@ enum trace_type {
#define F_STRUCT(args...) args
#undef FTRACE_ENTRY
-#define FTRACE_ENTRY(name, struct_name, id, tstruct, print) \
- struct struct_name { \
- struct trace_entry ent; \
- tstruct \
+#define FTRACE_ENTRY(name, struct_name, id, tstruct, print, filter) \
+ struct struct_name { \
+ struct trace_entry ent; \
+ tstruct \
}
#undef TP_ARGS
#define TP_ARGS(args...) args
#undef FTRACE_ENTRY_DUP
-#define FTRACE_ENTRY_DUP(name, name_struct, id, tstruct, printk)
+#define FTRACE_ENTRY_DUP(name, name_struct, id, tstruct, printk, filter)
#undef FTRACE_ENTRY_REG
-#define FTRACE_ENTRY_REG(name, struct_name, id, tstruct, print, regfn) \
- FTRACE_ENTRY(name, struct_name, id, PARAMS(tstruct), PARAMS(print))
+#define FTRACE_ENTRY_REG(name, struct_name, id, tstruct, print, \
+ filter, regfn) \
+ FTRACE_ENTRY(name, struct_name, id, PARAMS(tstruct), PARAMS(print), \
+ filter)
#include "trace_entries.h"
@@ -826,12 +828,13 @@ extern const char *__start___trace_bprintk_fmt[];
extern const char *__stop___trace_bprintk_fmt[];
#undef FTRACE_ENTRY
-#define FTRACE_ENTRY(call, struct_name, id, tstruct, print) \
+#define FTRACE_ENTRY(call, struct_name, id, tstruct, print, filter) \
extern struct ftrace_event_call \
__attribute__((__aligned__(4))) event_##call;
#undef FTRACE_ENTRY_DUP
-#define FTRACE_ENTRY_DUP(call, struct_name, id, tstruct, print) \
- FTRACE_ENTRY(call, struct_name, id, PARAMS(tstruct), PARAMS(print))
+#define FTRACE_ENTRY_DUP(call, struct_name, id, tstruct, print, filter) \
+ FTRACE_ENTRY(call, struct_name, id, PARAMS(tstruct), PARAMS(print), \
+ filter)
#include "trace_entries.h"
#ifdef CONFIG_PERF_EVENTS
diff --git a/kernel/trace/trace_entries.h b/kernel/trace/trace_entries.h
index 47db7ed..d91eb05 100644
--- a/kernel/trace/trace_entries.h
+++ b/kernel/trace/trace_entries.h
@@ -66,6 +66,8 @@ FTRACE_ENTRY_REG(function, ftrace_entry,
F_printk(" %lx <-- %lx", __entry->ip, __entry->parent_ip),
+ FILTER_TRACE_FN,
+
perf_ftrace_event_register
);
@@ -80,7 +82,9 @@ FTRACE_ENTRY(funcgraph_entry, ftrace_graph_ent_entry,
__field_desc( int, graph_ent, depth )
),
- F_printk("--> %lx (%d)", __entry->func, __entry->depth)
+ F_printk("--> %lx (%d)", __entry->func, __entry->depth),
+
+ FILTER_OTHER
);
/* Function return entry */
@@ -100,7 +104,9 @@ FTRACE_ENTRY(funcgraph_exit, ftrace_graph_ret_entry,
F_printk("<-- %lx (%d) (start: %llx end: %llx) over: %d",
__entry->func, __entry->depth,
__entry->calltime, __entry->rettime,
- __entry->depth)
+ __entry->depth),
+
+ FILTER_OTHER
);
/*
@@ -129,8 +135,9 @@ FTRACE_ENTRY(context_switch, ctx_switch_entry,
F_printk("%u:%u:%u ==> %u:%u:%u [%03u]",
__entry->prev_pid, __entry->prev_prio, __entry->prev_state,
__entry->next_pid, __entry->next_prio, __entry->next_state,
- __entry->next_cpu
- )
+ __entry->next_cpu),
+
+ FILTER_OTHER
);
/*
@@ -148,8 +155,9 @@ FTRACE_ENTRY_DUP(wakeup, ctx_switch_entry,
F_printk("%u:%u:%u ==+ %u:%u:%u [%03u]",
__entry->prev_pid, __entry->prev_prio, __entry->prev_state,
__entry->next_pid, __entry->next_prio, __entry->next_state,
- __entry->next_cpu
- )
+ __entry->next_cpu),
+
+ FILTER_OTHER
);
/*
@@ -171,7 +179,9 @@ FTRACE_ENTRY(kernel_stack, stack_entry,
"\t=> (%08lx)\n\t=> (%08lx)\n\t=> (%08lx)\n\t=> (%08lx)\n",
__entry->caller[0], __entry->caller[1], __entry->caller[2],
__entry->caller[3], __entry->caller[4], __entry->caller[5],
- __entry->caller[6], __entry->caller[7])
+ __entry->caller[6], __entry->caller[7]),
+
+ FILTER_OTHER
);
FTRACE_ENTRY(user_stack, userstack_entry,
@@ -187,7 +197,9 @@ FTRACE_ENTRY(user_stack, userstack_entry,
"\t=> (%08lx)\n\t=> (%08lx)\n\t=> (%08lx)\n\t=> (%08lx)\n",
__entry->caller[0], __entry->caller[1], __entry->caller[2],
__entry->caller[3], __entry->caller[4], __entry->caller[5],
- __entry->caller[6], __entry->caller[7])
+ __entry->caller[6], __entry->caller[7]),
+
+ FILTER_OTHER
);
/*
@@ -204,7 +216,9 @@ FTRACE_ENTRY(bprint, bprint_entry,
),
F_printk("%08lx fmt:%p",
- __entry->ip, __entry->fmt)
+ __entry->ip, __entry->fmt),
+
+ FILTER_OTHER
);
FTRACE_ENTRY(print, print_entry,
@@ -217,7 +231,9 @@ FTRACE_ENTRY(print, print_entry,
),
F_printk("%08lx %s",
- __entry->ip, __entry->buf)
+ __entry->ip, __entry->buf),
+
+ FILTER_OTHER
);
FTRACE_ENTRY(mmiotrace_rw, trace_mmiotrace_rw,
@@ -236,7 +252,9 @@ FTRACE_ENTRY(mmiotrace_rw, trace_mmiotrace_rw,
F_printk("%lx %lx %lx %d %x %x",
(unsigned long)__entry->phys, __entry->value, __entry->pc,
- __entry->map_id, __entry->opcode, __entry->width)
+ __entry->map_id, __entry->opcode, __entry->width),
+
+ FILTER_OTHER
);
FTRACE_ENTRY(mmiotrace_map, trace_mmiotrace_map,
@@ -254,7 +272,9 @@ FTRACE_ENTRY(mmiotrace_map, trace_mmiotrace_map,
F_printk("%lx %lx %lx %d %x",
(unsigned long)__entry->phys, __entry->virt, __entry->len,
- __entry->map_id, __entry->opcode)
+ __entry->map_id, __entry->opcode),
+
+ FILTER_OTHER
);
@@ -274,6 +294,8 @@ FTRACE_ENTRY(branch, trace_branch,
F_printk("%u:%s:%s (%u)",
__entry->line,
- __entry->func, __entry->file, __entry->correct)
+ __entry->func, __entry->file, __entry->correct),
+
+ FILTER_OTHER
);
diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c
index 76afaee..3da3d0e 100644
--- a/kernel/trace/trace_events_filter.c
+++ b/kernel/trace/trace_events_filter.c
@@ -899,6 +899,11 @@ int filter_assign_type(const char *type)
return FILTER_OTHER;
}
+static bool is_function_field(struct ftrace_event_field *field)
+{
+ return field->filter_type == FILTER_TRACE_FN;
+}
+
static bool is_string_field(struct ftrace_event_field *field)
{
return field->filter_type == FILTER_DYN_STRING ||
@@ -986,7 +991,7 @@ static int init_pred(struct filter_parse_state *ps,
fn = filter_pred_strloc;
else
fn = filter_pred_pchar;
- } else {
+ } else if (!is_function_field(field)) {
if (field->is_signed)
ret = strict_strtoll(pred->regex.pattern, 0, &val);
else
diff --git a/kernel/trace/trace_export.c b/kernel/trace/trace_export.c
index a3dbee6..7b46c9b 100644
--- a/kernel/trace/trace_export.c
+++ b/kernel/trace/trace_export.c
@@ -23,8 +23,10 @@
* function and thus become accesible via perf.
*/
#undef FTRACE_ENTRY_REG
-#define FTRACE_ENTRY_REG(name, struct_name, id, tstruct, print, regfn) \
- FTRACE_ENTRY(name, struct_name, id, PARAMS(tstruct), PARAMS(print))
+#define FTRACE_ENTRY_REG(name, struct_name, id, tstruct, print, \
+ filter, regfn) \
+ FTRACE_ENTRY(name, struct_name, id, PARAMS(tstruct), PARAMS(print), \
+ filter)
/* not needed for this file */
#undef __field_struct
@@ -52,21 +54,22 @@
#define F_printk(fmt, args...) fmt, args
#undef FTRACE_ENTRY
-#define FTRACE_ENTRY(name, struct_name, id, tstruct, print) \
-struct ____ftrace_##name { \
- tstruct \
-}; \
-static void __always_unused ____ftrace_check_##name(void) \
-{ \
- struct ____ftrace_##name *__entry = NULL; \
- \
- /* force compile-time check on F_printk() */ \
- printk(print); \
+#define FTRACE_ENTRY(name, struct_name, id, tstruct, print, filter) \
+struct ____ftrace_##name { \
+ tstruct \
+}; \
+static void __always_unused ____ftrace_check_##name(void) \
+{ \
+ struct ____ftrace_##name *__entry = NULL; \
+ \
+ /* force compile-time check on F_printk() */ \
+ printk(print); \
}
#undef FTRACE_ENTRY_DUP
-#define FTRACE_ENTRY_DUP(name, struct_name, id, tstruct, print) \
- FTRACE_ENTRY(name, struct_name, id, PARAMS(tstruct), PARAMS(print))
+#define FTRACE_ENTRY_DUP(name, struct_name, id, tstruct, print, filter) \
+ FTRACE_ENTRY(name, struct_name, id, PARAMS(tstruct), PARAMS(print), \
+ filter)
#include "trace_entries.h"
@@ -75,7 +78,7 @@ static void __always_unused ____ftrace_check_##name(void) \
ret = trace_define_field(event_call, #type, #item, \
offsetof(typeof(field), item), \
sizeof(field.item), \
- is_signed_type(type), FILTER_OTHER); \
+ is_signed_type(type), filter_type); \
if (ret) \
return ret;
@@ -85,7 +88,7 @@ static void __always_unused ____ftrace_check_##name(void) \
offsetof(typeof(field), \
container.item), \
sizeof(field.container.item), \
- is_signed_type(type), FILTER_OTHER); \
+ is_signed_type(type), filter_type); \
if (ret) \
return ret;
@@ -99,7 +102,7 @@ static void __always_unused ____ftrace_check_##name(void) \
ret = trace_define_field(event_call, event_storage, #item, \
offsetof(typeof(field), item), \
sizeof(field.item), \
- is_signed_type(type), FILTER_OTHER); \
+ is_signed_type(type), filter_type); \
mutex_unlock(&event_storage_mutex); \
if (ret) \
return ret; \
@@ -112,7 +115,7 @@ static void __always_unused ____ftrace_check_##name(void) \
offsetof(typeof(field), \
container.item), \
sizeof(field.container.item), \
- is_signed_type(type), FILTER_OTHER); \
+ is_signed_type(type), filter_type); \
if (ret) \
return ret;
@@ -120,17 +123,18 @@ static void __always_unused ____ftrace_check_##name(void) \
#define __dynamic_array(type, item) \
ret = trace_define_field(event_call, #type, #item, \
offsetof(typeof(field), item), \
- 0, is_signed_type(type), FILTER_OTHER);\
+ 0, is_signed_type(type), filter_type);\
if (ret) \
return ret;
#undef FTRACE_ENTRY
-#define FTRACE_ENTRY(name, struct_name, id, tstruct, print) \
+#define FTRACE_ENTRY(name, struct_name, id, tstruct, print, filter) \
int \
ftrace_define_fields_##name(struct ftrace_event_call *event_call) \
{ \
struct struct_name field; \
int ret; \
+ int filter_type = filter; \
\
tstruct; \
\
@@ -161,7 +165,8 @@ ftrace_define_fields_##name(struct ftrace_event_call *event_call) \
#define F_printk(fmt, args...) #fmt ", " __stringify(args)
#undef FTRACE_ENTRY_REG
-#define FTRACE_ENTRY_REG(call, struct_name, etype, tstruct, print, regfn)\
+#define FTRACE_ENTRY_REG(call, struct_name, etype, tstruct, print, filter,\
+ regfn) \
\
struct ftrace_event_class event_class_ftrace_##call = { \
.system = __stringify(TRACE_SYSTEM), \
@@ -180,9 +185,9 @@ struct ftrace_event_call __used \
__attribute__((section("_ftrace_events"))) *__event_##call = &event_##call;
#undef FTRACE_ENTRY
-#define FTRACE_ENTRY(call, struct_name, etype, tstruct, print) \
+#define FTRACE_ENTRY(call, struct_name, etype, tstruct, print, filter) \
FTRACE_ENTRY_REG(call, struct_name, etype, \
- PARAMS(tstruct), PARAMS(print), NULL)
+ PARAMS(tstruct), PARAMS(print), filter, NULL)
int ftrace_event_is_function(struct ftrace_event_call *call)
{
^ permalink raw reply related [flat|nested] 17+ messages in thread
* [tip:perf/core] ftrace, perf: Add filter support for function trace event
2012-02-15 14:51 ` [PATCH 7/7] ftrace, perf: Add filter support for function trace event Jiri Olsa
@ 2012-02-27 9:38 ` tip-bot for Jiri Olsa
0 siblings, 0 replies; 17+ messages in thread
From: tip-bot for Jiri Olsa @ 2012-02-27 9:38 UTC (permalink / raw)
To: linux-tip-commits; +Cc: linux-kernel, hpa, mingo, rostedt, tglx, jolsa
Commit-ID: 5500fa51199aee770ce53718853732600543619e
Gitweb: http://git.kernel.org/tip/5500fa51199aee770ce53718853732600543619e
Author: Jiri Olsa <jolsa@redhat.com>
AuthorDate: Wed, 15 Feb 2012 15:51:54 +0100
Committer: Steven Rostedt <rostedt@goodmis.org>
CommitDate: Tue, 21 Feb 2012 11:08:30 -0500
ftrace, perf: Add filter support for function trace event
Adding support to filter function trace event via perf
interface. It is now possible to use filter interface
in the perf tool like:
perf record -e ftrace:function --filter="(ip == mm_*)" ls
The filter syntax is restricted to the the 'ip' field only,
and following operators are accepted '==' '!=' '||', ending
up with the filter strings like:
ip == f1[, ]f2 ... || ip != f3[, ]f4 ...
with comma ',' or space ' ' as a function separator. If the
space ' ' is used as a separator, the right side of the
assignment needs to be enclosed in double quotes '"', e.g.:
perf record -e ftrace:function --filter '(ip == do_execve,sys_*,ext*)' ls
perf record -e ftrace:function --filter '(ip == "do_execve,sys_*,ext*")' ls
perf record -e ftrace:function --filter '(ip == "do_execve sys_* ext*")' ls
The '==' operator adds trace filter with same effect as would
be added via set_ftrace_filter file.
The '!=' operator adds trace filter with same effect as would
be added via set_ftrace_notrace file.
The right side of the '!=', '==' operators is list of functions
or regexp. to be added to filter separated by space.
The '||' operator is used for connecting multiple filter definitions
together. It is possible to have more than one '==' and '!='
operators within one filter string.
Link: http://lkml.kernel.org/r/1329317514-8131-8-git-send-email-jolsa@redhat.com
Signed-off-by: Jiri Olsa <jolsa@redhat.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
include/linux/ftrace.h | 7 +-
kernel/trace/ftrace.c | 6 ++
kernel/trace/trace.h | 2 -
kernel/trace/trace_event_perf.c | 4 +-
kernel/trace/trace_events_filter.c | 165 ++++++++++++++++++++++++++++++++++--
5 files changed, 172 insertions(+), 12 deletions(-)
diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h
index 64a309d..72a6cab 100644
--- a/include/linux/ftrace.h
+++ b/include/linux/ftrace.h
@@ -250,6 +250,7 @@ int ftrace_set_notrace(struct ftrace_ops *ops, unsigned char *buf,
int len, int reset);
void ftrace_set_global_filter(unsigned char *buf, int len, int reset);
void ftrace_set_global_notrace(unsigned char *buf, int len, int reset);
+void ftrace_free_filter(struct ftrace_ops *ops);
int register_ftrace_command(struct ftrace_func_command *cmd);
int unregister_ftrace_command(struct ftrace_func_command *cmd);
@@ -380,9 +381,6 @@ extern void ftrace_enable_daemon(void);
#else
static inline int skip_trace(unsigned long ip) { return 0; }
static inline int ftrace_force_update(void) { return 0; }
-static inline void ftrace_set_filter(unsigned char *buf, int len, int reset)
-{
-}
static inline void ftrace_disable_daemon(void) { }
static inline void ftrace_enable_daemon(void) { }
static inline void ftrace_release_mod(struct module *mod) {}
@@ -406,6 +404,9 @@ static inline int ftrace_text_reserved(void *start, void *end)
*/
#define ftrace_regex_open(ops, flag, inod, file) ({ -ENODEV; })
#define ftrace_set_early_filter(ops, buf, enable) do { } while (0)
+#define ftrace_set_filter(ops, buf, len, reset) ({ -ENODEV; })
+#define ftrace_set_notrace(ops, buf, len, reset) ({ -ENODEV; })
+#define ftrace_free_filter(ops) do { } while (0)
static inline ssize_t ftrace_filter_write(struct file *file, const char __user *ubuf,
size_t cnt, loff_t *ppos) { return -ENODEV; }
diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
index f615f97..867bd1d 100644
--- a/kernel/trace/ftrace.c
+++ b/kernel/trace/ftrace.c
@@ -1186,6 +1186,12 @@ static void free_ftrace_hash_rcu(struct ftrace_hash *hash)
call_rcu_sched(&hash->rcu, __free_ftrace_hash_rcu);
}
+void ftrace_free_filter(struct ftrace_ops *ops)
+{
+ free_ftrace_hash(ops->filter_hash);
+ free_ftrace_hash(ops->notrace_hash);
+}
+
static struct ftrace_hash *alloc_ftrace_hash(int size_bits)
{
struct ftrace_hash *hash;
diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h
index 29f93cd..54faec7 100644
--- a/kernel/trace/trace.h
+++ b/kernel/trace/trace.h
@@ -776,9 +776,7 @@ struct filter_pred {
u64 val;
struct regex regex;
unsigned short *ops;
-#ifdef CONFIG_FTRACE_STARTUP_TEST
struct ftrace_event_field *field;
-#endif
int offset;
int not;
int op;
diff --git a/kernel/trace/trace_event_perf.c b/kernel/trace/trace_event_perf.c
index fdeeb5c..fee3752 100644
--- a/kernel/trace/trace_event_perf.c
+++ b/kernel/trace/trace_event_perf.c
@@ -298,7 +298,9 @@ static int perf_ftrace_function_register(struct perf_event *event)
static int perf_ftrace_function_unregister(struct perf_event *event)
{
struct ftrace_ops *ops = &event->ftrace_ops;
- return unregister_ftrace_function(ops);
+ int ret = unregister_ftrace_function(ops);
+ ftrace_free_filter(ops);
+ return ret;
}
static void perf_ftrace_function_enable(struct perf_event *event)
diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c
index 3da3d0e..431dba8 100644
--- a/kernel/trace/trace_events_filter.c
+++ b/kernel/trace/trace_events_filter.c
@@ -81,6 +81,7 @@ enum {
FILT_ERR_TOO_MANY_PREDS,
FILT_ERR_MISSING_FIELD,
FILT_ERR_INVALID_FILTER,
+ FILT_ERR_IP_FIELD_ONLY,
};
static char *err_text[] = {
@@ -96,6 +97,7 @@ static char *err_text[] = {
"Too many terms in predicate expression",
"Missing field name and/or value",
"Meaningless filter expression",
+ "Only 'ip' field is supported for function trace",
};
struct opstack_op {
@@ -991,7 +993,12 @@ static int init_pred(struct filter_parse_state *ps,
fn = filter_pred_strloc;
else
fn = filter_pred_pchar;
- } else if (!is_function_field(field)) {
+ } else if (is_function_field(field)) {
+ if (strcmp(field->name, "ip")) {
+ parse_error(ps, FILT_ERR_IP_FIELD_ONLY, 0);
+ return -EINVAL;
+ }
+ } else {
if (field->is_signed)
ret = strict_strtoll(pred->regex.pattern, 0, &val);
else
@@ -1338,10 +1345,7 @@ static struct filter_pred *create_pred(struct filter_parse_state *ps,
strcpy(pred.regex.pattern, operand2);
pred.regex.len = strlen(pred.regex.pattern);
-
-#ifdef CONFIG_FTRACE_STARTUP_TEST
pred.field = field;
-#endif
return init_pred(ps, field, &pred) ? NULL : &pred;
}
@@ -1954,6 +1958,148 @@ void ftrace_profile_free_filter(struct perf_event *event)
__free_filter(filter);
}
+struct function_filter_data {
+ struct ftrace_ops *ops;
+ int first_filter;
+ int first_notrace;
+};
+
+#ifdef CONFIG_FUNCTION_TRACER
+static char **
+ftrace_function_filter_re(char *buf, int len, int *count)
+{
+ char *str, *sep, **re;
+
+ str = kstrndup(buf, len, GFP_KERNEL);
+ if (!str)
+ return NULL;
+
+ /*
+ * The argv_split function takes white space
+ * as a separator, so convert ',' into spaces.
+ */
+ while ((sep = strchr(str, ',')))
+ *sep = ' ';
+
+ re = argv_split(GFP_KERNEL, str, count);
+ kfree(str);
+ return re;
+}
+
+static int ftrace_function_set_regexp(struct ftrace_ops *ops, int filter,
+ int reset, char *re, int len)
+{
+ int ret;
+
+ if (filter)
+ ret = ftrace_set_filter(ops, re, len, reset);
+ else
+ ret = ftrace_set_notrace(ops, re, len, reset);
+
+ return ret;
+}
+
+static int __ftrace_function_set_filter(int filter, char *buf, int len,
+ struct function_filter_data *data)
+{
+ int i, re_cnt, ret;
+ int *reset;
+ char **re;
+
+ reset = filter ? &data->first_filter : &data->first_notrace;
+
+ /*
+ * The 'ip' field could have multiple filters set, separated
+ * either by space or comma. We first cut the filter and apply
+ * all pieces separatelly.
+ */
+ re = ftrace_function_filter_re(buf, len, &re_cnt);
+ if (!re)
+ return -EINVAL;
+
+ for (i = 0; i < re_cnt; i++) {
+ ret = ftrace_function_set_regexp(data->ops, filter, *reset,
+ re[i], strlen(re[i]));
+ if (ret)
+ break;
+
+ if (*reset)
+ *reset = 0;
+ }
+
+ argv_free(re);
+ return ret;
+}
+
+static int ftrace_function_check_pred(struct filter_pred *pred, int leaf)
+{
+ struct ftrace_event_field *field = pred->field;
+
+ if (leaf) {
+ /*
+ * Check the leaf predicate for function trace, verify:
+ * - only '==' and '!=' is used
+ * - the 'ip' field is used
+ */
+ if ((pred->op != OP_EQ) && (pred->op != OP_NE))
+ return -EINVAL;
+
+ if (strcmp(field->name, "ip"))
+ return -EINVAL;
+ } else {
+ /*
+ * Check the non leaf predicate for function trace, verify:
+ * - only '||' is used
+ */
+ if (pred->op != OP_OR)
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+static int ftrace_function_set_filter_cb(enum move_type move,
+ struct filter_pred *pred,
+ int *err, void *data)
+{
+ /* Checking the node is valid for function trace. */
+ if ((move != MOVE_DOWN) ||
+ (pred->left != FILTER_PRED_INVALID)) {
+ *err = ftrace_function_check_pred(pred, 0);
+ } else {
+ *err = ftrace_function_check_pred(pred, 1);
+ if (*err)
+ return WALK_PRED_ABORT;
+
+ *err = __ftrace_function_set_filter(pred->op == OP_EQ,
+ pred->regex.pattern,
+ pred->regex.len,
+ data);
+ }
+
+ return (*err) ? WALK_PRED_ABORT : WALK_PRED_DEFAULT;
+}
+
+static int ftrace_function_set_filter(struct perf_event *event,
+ struct event_filter *filter)
+{
+ struct function_filter_data data = {
+ .first_filter = 1,
+ .first_notrace = 1,
+ .ops = &event->ftrace_ops,
+ };
+
+ return walk_pred_tree(filter->preds, filter->root,
+ ftrace_function_set_filter_cb, &data);
+}
+#else
+static int ftrace_function_set_filter(struct perf_event *event,
+ struct event_filter *filter)
+{
+ return -ENODEV;
+}
+#endif /* CONFIG_FUNCTION_TRACER */
+
int ftrace_profile_set_filter(struct perf_event *event, int event_id,
char *filter_str)
{
@@ -1974,9 +2120,16 @@ int ftrace_profile_set_filter(struct perf_event *event, int event_id,
goto out_unlock;
err = create_filter(call, filter_str, false, &filter);
- if (!err)
- event->filter = filter;
+ if (err)
+ goto free_filter;
+
+ if (ftrace_event_is_function(call))
+ err = ftrace_function_set_filter(event, filter);
else
+ event->filter = filter;
+
+free_filter:
+ if (err || ftrace_event_is_function(call))
__free_filter(filter);
out_unlock:
^ permalink raw reply related [flat|nested] 17+ messages in thread
end of thread, other threads:[~2012-02-27 9:38 UTC | newest]
Thread overview: 17+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2012-02-15 14:51 [PATCHv9 0/7] ftrace, perf: Adding support to use function trace Jiri Olsa
2012-02-15 14:51 ` [PATCH 1/7] ftrace: Add enable/disable ftrace_ops control interface Jiri Olsa
2012-02-27 9:33 ` [tip:perf/core] ftrace: Add enable/ disable " tip-bot for Jiri Olsa
2012-02-15 14:51 ` [PATCH 2/7] ftrace, perf: Add open/close tracepoint perf registration actions Jiri Olsa
2012-02-27 9:34 ` [tip:perf/core] ftrace, perf: Add open/ close " tip-bot for Jiri Olsa
2012-02-15 14:51 ` [PATCH 3/7] ftrace, perf: Add add/del " Jiri Olsa
2012-02-27 9:34 ` [tip:perf/core] ftrace, perf: Add add/ del " tip-bot for Jiri Olsa
2012-02-15 14:51 ` [PATCH 4/7] ftrace: Add FTRACE_ENTRY_REG macro to allow event registration Jiri Olsa
2012-02-15 15:13 ` Frederic Weisbecker
2012-02-27 9:35 ` [tip:perf/core] " tip-bot for Jiri Olsa
2012-02-15 14:51 ` [PATCH 5/7] ftrace, perf: Add support to use function tracepoint in perf Jiri Olsa
2012-02-15 16:03 ` Frederic Weisbecker
2012-02-27 9:36 ` [tip:perf/core] " tip-bot for Jiri Olsa
2012-02-15 14:51 ` [PATCH 6/7] ftrace: Allow to specify filter field type for ftrace events Jiri Olsa
2012-02-27 9:37 ` [tip:perf/core] " tip-bot for Jiri Olsa
2012-02-15 14:51 ` [PATCH 7/7] ftrace, perf: Add filter support for function trace event Jiri Olsa
2012-02-27 9:38 ` [tip:perf/core] " tip-bot for Jiri Olsa
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.