[PATCH -tip v14 06/12] tracing: ftrace dynamic ftrace_event_callsupport

From: Masami Hiramatsu
Date: Thu Aug 13 2009 - 16:33:26 EST

Add dynamic ftrace_event_call support to ftrace. Trace engines can adds new
ftrace_event_call to ftrace on the fly. Each operator functions of the call
takes a ftrace_event_call data structure as an argument, because these
functions may be shared among several ftrace_event_calls.

Changes from v13:
- Define remove_subsystem_dir() always (revirt a2ca5e03), because
trace_remove_event_call() uses it.
- Modify syscall tracer because of ftrace_event_call change.

Signed-off-by: Masami Hiramatsu <mhiramat@xxxxxxxxxx>
Acked-by: Frederic Weisbecker <fweisbec@xxxxxxxxx>
Cc: Ananth N Mavinakayanahalli <ananth@xxxxxxxxxx>
Cc: Avi Kivity <avi@xxxxxxxxxx>
Cc: Andi Kleen <ak@xxxxxxxxxxxxxxx>
Cc: Christoph Hellwig <hch@xxxxxxxxxxxxx>
Cc: Frank Ch. Eigler <fche@xxxxxxxxxx>
Cc: H. Peter Anvin <hpa@xxxxxxxxx>
Cc: Ingo Molnar <mingo@xxxxxxx>
Cc: Jason Baron <jbaron@xxxxxxxxxx>
Cc: Jim Keniston <jkenisto@xxxxxxxxxx>
Cc: K.Prasad <prasad@xxxxxxxxxxxxxxxxxx>
Cc: Lai Jiangshan <laijs@xxxxxxxxxxxxxx>
Cc: Li Zefan <lizf@xxxxxxxxxxxxxx>
Cc: PrzemysÅaw PaweÅczyk <przemyslaw@xxxxxxxxxxxx>
Cc: Roland McGrath <roland@xxxxxxxxxx>
Cc: Sam Ravnborg <sam@xxxxxxxxxxxx>
Cc: Srikar Dronamraju <srikar@xxxxxxxxxxxxxxxxxx>
Cc: Steven Rostedt <rostedt@xxxxxxxxxxx>
Cc: Tom Zanussi <tzanussi@xxxxxxxxx>
Cc: Vegard Nossum <vegard.nossum@xxxxxxxxx>
---

include/linux/ftrace_event.h | 14 +++--
include/linux/syscalls.h | 4 +
include/trace/ftrace.h | 19 +++----
include/trace/syscall.h | 8 +--
kernel/trace/trace_events.c | 119 +++++++++++++++++++++++++++++------------
kernel/trace/trace_export.c | 23 ++++----
kernel/trace/trace_syscalls.c | 16 +++---
7 files changed, 125 insertions(+), 78 deletions(-)

diff --git a/include/linux/ftrace_event.h b/include/linux/ftrace_event.h
index 189806b..9af68ce 100644
--- a/include/linux/ftrace_event.h
+++ b/include/linux/ftrace_event.h
@@ -112,13 +112,13 @@ struct ftrace_event_call {
struct dentry *dir;
struct trace_event *event;
int enabled;
- int (*regfunc)(void *);
- void (*unregfunc)(void *);
+ int (*regfunc)(struct ftrace_event_call *);
+ void (*unregfunc)(struct ftrace_event_call *);
int id;
- int (*raw_init)(void);
- int (*show_format)(struct ftrace_event_call *call,
- struct trace_seq *s);
- int (*define_fields)(void);
+ int (*raw_init)(struct ftrace_event_call *);
+ int (*show_format)(struct ftrace_event_call *,
+ struct trace_seq *);
+ int (*define_fields)(struct ftrace_event_call *);
struct list_head fields;
int filter_active;
struct event_filter *filter;
@@ -142,6 +142,8 @@ extern int filter_current_check_discard(struct ftrace_event_call *call,

extern int trace_define_field(struct ftrace_event_call *call, char *type,
char *name, int offset, int size, int is_signed);
+extern int trace_add_event_call(struct ftrace_event_call *call);
+extern void trace_remove_event_call(struct ftrace_event_call *call);

#define is_signed_type(type) (((type)(-1)) < 0)

diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index 87d06c1..be59d22 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -165,7 +165,7 @@ static void prof_sysexit_disable_##sname(struct ftrace_event_call *event_call) \
struct trace_event enter_syscall_print_##sname = { \
.trace = print_syscall_enter, \
}; \
- static int init_enter_##sname(void) \
+ static int init_enter_##sname(struct ftrace_event_call *call) \
{ \
int num, id; \
num = syscall_name_to_nr("sys"#sname); \
@@ -201,7 +201,7 @@ static void prof_sysexit_disable_##sname(struct ftrace_event_call *event_call) \
struct trace_event exit_syscall_print_##sname = { \
.trace = print_syscall_exit, \
}; \
- static int init_exit_##sname(void) \
+ static int init_exit_##sname(struct ftrace_event_call *call) \
{ \
int num, id; \
num = syscall_name_to_nr("sys"#sname); \
diff --git a/include/trace/ftrace.h b/include/trace/ftrace.h
index b250b06..6c1b5b1 100644
--- a/include/trace/ftrace.h
+++ b/include/trace/ftrace.h
@@ -294,10 +294,9 @@ ftrace_raw_output_##call(struct trace_iterator *iter, int flags) \
#undef TRACE_EVENT
#define TRACE_EVENT(call, proto, args, tstruct, func, print) \
int \
-ftrace_define_fields_##call(void) \
+ftrace_define_fields_##call(struct ftrace_event_call *event_call) \
{ \
struct ftrace_raw_##call field; \
- struct ftrace_event_call *event_call = &event_##call; \
int ret; \
\
__common_field(int, type, 1); \
@@ -411,7 +410,7 @@ static void ftrace_profile_disable_##call(struct ftrace_event_call *event_call)\
* event_trace_printk(_RET_IP_, "<call>: " <fmt>);
* }
*
- * static int ftrace_reg_event_<call>(void)
+ * static int ftrace_reg_event_<call>(struct ftrace_event_call *unused)
* {
* int ret;
*
@@ -422,7 +421,7 @@ static void ftrace_profile_disable_##call(struct ftrace_event_call *event_call)\
* return ret;
* }
*
- * static void ftrace_unreg_event_<call>(void)
+ * static void ftrace_unreg_event_<call>(struct ftrace_event_call *unused)
* {
* unregister_trace_<call>(ftrace_event_<call>);
* }
@@ -455,7 +454,7 @@ static void ftrace_profile_disable_##call(struct ftrace_event_call *event_call)\
* trace_current_buffer_unlock_commit(event, irq_flags, pc);
* }
*
- * static int ftrace_raw_reg_event_<call>(void)
+ * static int ftrace_raw_reg_event_<call>(struct ftrace_event_call *unused)
* {
* int ret;
*
@@ -466,7 +465,7 @@ static void ftrace_profile_disable_##call(struct ftrace_event_call *event_call)\
* return ret;
* }
*
- * static void ftrace_unreg_event_<call>(void)
+ * static void ftrace_unreg_event_<call>(struct ftrace_event_call *unused)
* {
* unregister_trace_<call>(ftrace_raw_event_<call>);
* }
@@ -475,7 +474,7 @@ static void ftrace_profile_disable_##call(struct ftrace_event_call *event_call)\
* .trace = ftrace_raw_output_<call>, <-- stage 2
* };
*
- * static int ftrace_raw_init_event_<call>(void)
+ * static int ftrace_raw_init_event_<call>(struct ftrace_event_call *unused)
* {
* int id;
*
@@ -569,7 +568,7 @@ static void ftrace_raw_event_##call(proto) \
trace_nowake_buffer_unlock_commit(event, irq_flags, pc); \
} \
\
-static int ftrace_raw_reg_event_##call(void *ptr) \
+static int ftrace_raw_reg_event_##call(struct ftrace_event_call *unused)\
{ \
int ret; \
\
@@ -580,7 +579,7 @@ static int ftrace_raw_reg_event_##call(void *ptr) \
return ret; \
} \
\
-static void ftrace_raw_unreg_event_##call(void *ptr) \
+static void ftrace_raw_unreg_event_##call(struct ftrace_event_call *unused)\
{ \
unregister_trace_##call(ftrace_raw_event_##call); \
} \
@@ -589,7 +588,7 @@ static struct trace_event ftrace_event_type_##call = { \
.trace = ftrace_raw_output_##call, \
}; \
\
-static int ftrace_raw_init_event_##call(void) \
+static int ftrace_raw_init_event_##call(struct ftrace_event_call *unused)\
{ \
int id; \
\
diff --git a/include/trace/syscall.h b/include/trace/syscall.h
index 0cb0362..848b4ae 100644
--- a/include/trace/syscall.h
+++ b/include/trace/syscall.h
@@ -51,10 +51,10 @@ void set_syscall_enter_id(int num, int id);
void set_syscall_exit_id(int num, int id);
extern struct trace_event event_syscall_enter;
extern struct trace_event event_syscall_exit;
-extern int reg_event_syscall_enter(void *ptr);
-extern void unreg_event_syscall_enter(void *ptr);
-extern int reg_event_syscall_exit(void *ptr);
-extern void unreg_event_syscall_exit(void *ptr);
+extern int reg_event_syscall_enter(struct ftrace_event_call *call);
+extern void unreg_event_syscall_enter(struct ftrace_event_call *call);
+extern int reg_event_syscall_exit(struct ftrace_event_call *call);
+extern void unreg_event_syscall_exit(struct ftrace_event_call *call);
extern int
ftrace_format_syscall(struct ftrace_event_call *call, struct trace_seq *s);
enum print_line_t print_syscall_enter(struct trace_iterator *iter, int flags);
diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c
index b568ade..be701d1 100644
--- a/kernel/trace/trace_events.c
+++ b/kernel/trace/trace_events.c
@@ -62,9 +62,7 @@ err:
}
EXPORT_SYMBOL_GPL(trace_define_field);

-#ifdef CONFIG_MODULES
-
-static void trace_destroy_fields(struct ftrace_event_call *call)
+void trace_destroy_fields(struct ftrace_event_call *call)
{
struct ftrace_event_field *field, *next;

@@ -76,8 +74,6 @@ static void trace_destroy_fields(struct ftrace_event_call *call)
}
}

-#endif /* CONFIG_MODULES */
-
static void ftrace_event_enable_disable(struct ftrace_event_call *call,
int enable)
{
@@ -86,14 +82,14 @@ static void ftrace_event_enable_disable(struct ftrace_event_call *call,
if (call->enabled) {
call->enabled = 0;
tracing_stop_cmdline_record();
- call->unregfunc(call->data);
+ call->unregfunc(call);
}
break;
case 1:
if (!call->enabled) {
call->enabled = 1;
tracing_start_cmdline_record();
- call->regfunc(call->data);
+ call->regfunc(call);
}
break;
}
@@ -941,7 +937,7 @@ event_create_dir(struct ftrace_event_call *call, struct dentry *d_events,
id);

if (call->define_fields) {
- ret = call->define_fields();
+ ret = call->define_fields(call);
if (ret < 0) {
pr_warning("Could not initialize trace point"
" events/%s\n", call->name);
@@ -961,27 +957,43 @@ event_create_dir(struct ftrace_event_call *call, struct dentry *d_events,
return 0;
}

-#define for_each_event(event, start, end) \
- for (event = start; \
- (unsigned long)event < (unsigned long)end; \
- event++)
+static int __trace_add_event_call(struct ftrace_event_call *call)
+{
+ struct dentry *d_events;
+ int ret;

-#ifdef CONFIG_MODULES
+ if (!call->name)
+ return -EINVAL;

-static LIST_HEAD(ftrace_module_file_list);
+ if (call->raw_init) {
+ ret = call->raw_init(call);
+ if (ret < 0) {
+ if (ret != -ENOSYS)
+ pr_warning("Could not initialize trace "
+ "events/%s\n", call->name);
+ return ret;
+ }
+ }

-/*
- * Modules must own their file_operations to keep up with
- * reference counting.
- */
-struct ftrace_module_file_ops {
- struct list_head list;
- struct module *mod;
- struct file_operations id;
- struct file_operations enable;
- struct file_operations format;
- struct file_operations filter;
-};
+ d_events = event_trace_events_dir();
+ if (!d_events)
+ return -ENOENT;
+
+ list_add(&call->list, &ftrace_events);
+ return event_create_dir(call, d_events, &ftrace_event_id_fops,
+ &ftrace_enable_fops, &ftrace_event_filter_fops,
+ &ftrace_event_format_fops);
+}
+
+/* Add an additional event_call dynamically */
+int trace_add_event_call(struct ftrace_event_call *call)
+{
+ int ret;
+ mutex_lock(&event_mutex);
+ ret = __trace_add_event_call(call);
+ mutex_unlock(&event_mutex);
+ return ret;
+}

static void remove_subsystem_dir(const char *name)
{
@@ -1009,6 +1021,48 @@ static void remove_subsystem_dir(const char *name)
}
}

+static void __trace_remove_event_call(struct ftrace_event_call *call)
+{
+ ftrace_event_enable_disable(call, 0);
+ if (call->event)
+ __unregister_ftrace_event(call->event);
+ debugfs_remove_recursive(call->dir);
+ list_del(&call->list);
+ trace_destroy_fields(call);
+ destroy_preds(call);
+ remove_subsystem_dir(call->system);
+}
+
+/* Remove an event_call */
+void trace_remove_event_call(struct ftrace_event_call *call)
+{
+ mutex_lock(&event_mutex);
+ __trace_remove_event_call(call);
+ mutex_unlock(&event_mutex);
+}
+
+#define for_each_event(event, start, end) \
+ for (event = start; \
+ (unsigned long)event < (unsigned long)end; \
+ event++)
+
+#ifdef CONFIG_MODULES
+
+static LIST_HEAD(ftrace_module_file_list);
+
+/*
+ * Modules must own their file_operations to keep up with
+ * reference counting.
+ */
+struct ftrace_module_file_ops {
+ struct list_head list;
+ struct module *mod;
+ struct file_operations id;
+ struct file_operations enable;
+ struct file_operations format;
+ struct file_operations filter;
+};
+
static struct ftrace_module_file_ops *
trace_create_file_ops(struct module *mod)
{
@@ -1066,7 +1120,7 @@ static void trace_module_add_events(struct module *mod)
if (!call->name)
continue;
if (call->raw_init) {
- ret = call->raw_init();
+ ret = call->raw_init(call);
if (ret < 0) {
if (ret != -ENOSYS)
pr_warning("Could not initialize trace "
@@ -1101,14 +1155,7 @@ static void trace_module_remove_events(struct module *mod)
list_for_each_entry_safe(call, p, &ftrace_events, list) {
if (call->mod == mod) {
found = true;
- ftrace_event_enable_disable(call, 0);
- if (call->event)
- __unregister_ftrace_event(call->event);
- debugfs_remove_recursive(call->dir);
- list_del(&call->list);
- trace_destroy_fields(call);
- destroy_preds(call);
- remove_subsystem_dir(call->system);
+ __trace_remove_event_call(call);
}
}

@@ -1226,7 +1273,7 @@ static __init int event_trace_init(void)
if (!call->name)
continue;
if (call->raw_init) {
- ret = call->raw_init();
+ ret = call->raw_init(call);
if (ret < 0) {
if (ret != -ENOSYS)
pr_warning("Could not initialize trace "
diff --git a/kernel/trace/trace_export.c b/kernel/trace/trace_export.c
index 956d4bc..71c8d7f 100644
--- a/kernel/trace/trace_export.c
+++ b/kernel/trace/trace_export.c
@@ -117,10 +117,16 @@ ftrace_format_##call(struct ftrace_event_call *unused, \
#define TRACE_FIELD_SPECIAL(type_item, item, len, cmd) \
cmd;

+static int ftrace_raw_init_event(struct ftrace_event_call *event_call)
+{
+ INIT_LIST_HEAD(&event_call->fields);
+ init_preds(event_call);
+ return 0;
+}
+
#undef TRACE_EVENT_FORMAT
#define TRACE_EVENT_FORMAT(call, proto, args, fmt, tstruct, tpfmt) \
-int ftrace_define_fields_##call(void); \
-static int ftrace_raw_init_event_##call(void); \
+int ftrace_define_fields_##call(struct ftrace_event_call *c); \
\
struct ftrace_event_call __used \
__attribute__((__aligned__(4))) \
@@ -128,16 +134,10 @@ __attribute__((section("_ftrace_events"))) event_##call = { \
.name = #call, \
.id = proto, \
.system = __stringify(TRACE_SYSTEM), \
- .raw_init = ftrace_raw_init_event_##call, \
+ .raw_init = ftrace_raw_init_event, \
.show_format = ftrace_format_##call, \
.define_fields = ftrace_define_fields_##call, \
-}; \
-static int ftrace_raw_init_event_##call(void) \
-{ \
- INIT_LIST_HEAD(&event_##call.fields); \
- init_preds(&event_##call); \
- return 0; \
-} \
+};

#undef TRACE_EVENT_FORMAT_NOFILTER
#define TRACE_EVENT_FORMAT_NOFILTER(call, proto, args, fmt, tstruct, \
@@ -184,9 +184,8 @@ __attribute__((section("_ftrace_events"))) event_##call = { \
#undef TRACE_EVENT_FORMAT
#define TRACE_EVENT_FORMAT(call, proto, args, fmt, tstruct, tpfmt) \
int \
-ftrace_define_fields_##call(void) \
+ftrace_define_fields_##call(struct ftrace_event_call *event_call) \
{ \
- struct ftrace_event_call *event_call = &event_##call; \
struct args field; \
int ret; \
\
diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c
index f837ccc..3451621 100644
--- a/kernel/trace/trace_syscalls.c
+++ b/kernel/trace/trace_syscalls.c
@@ -210,13 +210,13 @@ void ftrace_syscall_exit(struct pt_regs *regs, long ret)
trace_wake_up();
}

-int reg_event_syscall_enter(void *ptr)
+int reg_event_syscall_enter(struct ftrace_event_call *call)
{
int ret = 0;
int num;
char *name;

- name = (char *)ptr;
+ name = (char *)call->data;
num = syscall_name_to_nr(name);
if (num < 0 || num >= FTRACE_SYSCALL_MAX)
return -ENOSYS;
@@ -234,12 +234,12 @@ int reg_event_syscall_enter(void *ptr)
return ret;
}

-void unreg_event_syscall_enter(void *ptr)
+void unreg_event_syscall_enter(struct ftrace_event_call *call)
{
int num;
char *name;

- name = (char *)ptr;
+ name = (char *)call->data;
num = syscall_name_to_nr(name);
if (num < 0 || num >= FTRACE_SYSCALL_MAX)
return;
@@ -251,13 +251,13 @@ void unreg_event_syscall_enter(void *ptr)
mutex_unlock(&syscall_trace_lock);
}

-int reg_event_syscall_exit(void *ptr)
+int reg_event_syscall_exit(struct ftrace_event_call *call)
{
int ret = 0;
int num;
char *name;

- name = (char *)ptr;
+ name = (char *)call->data;
num = syscall_name_to_nr(name);
if (num < 0 || num >= FTRACE_SYSCALL_MAX)
return -ENOSYS;
@@ -275,12 +275,12 @@ int reg_event_syscall_exit(void *ptr)
return ret;
}

-void unreg_event_syscall_exit(void *ptr)
+void unreg_event_syscall_exit(struct ftrace_event_call *call)
{
int num;
char *name;

- name = (char *)ptr;
+ name = (char *)call->data;
num = syscall_name_to_nr(name);
if (num < 0 || num >= FTRACE_SYSCALL_MAX)
return;


--
Masami Hiramatsu

Software Engineer
Hitachi Computer Products (America), Inc.
Software Solutions Division

e-mail: mhiramat@xxxxxxxxxx
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/