[PATCH 1/3] tracing: Update event filters for multibuffer

From: Tom Zanussi
Date: Thu Oct 24 2013 - 09:35:08 EST

The trace event filters are still tied to event calls rather than
event files, which means you don't get what you'd expect when using
filters in the multibuffer case:

Before:

# echo 'bytes_alloc > 8192' > /sys/kernel/debug/tracing/events/kmem/kmalloc/filter
# cat /sys/kernel/debug/tracing/events/kmem/kmalloc/filter
bytes_alloc > 8192
# mkdir /sys/kernel/debug/tracing/instances/test1
# echo 'bytes_alloc > 2048' > /sys/kernel/debug/tracing/instances/test1/events/kmem/kmalloc/filter
# cat /sys/kernel/debug/tracing/events/kmem/kmalloc/filter
bytes_alloc > 2048
# cat /sys/kernel/debug/tracing/instances/test1/events/kmem/kmalloc/filter
bytes_alloc > 2048

Setting the filter in tracing/instances/test1/events shouldn't affect
the same event in tracing/events as it does above.

After:

# echo 'bytes_alloc > 8192' > /sys/kernel/debug/tracing/events/kmem/kmalloc/filter
# cat /sys/kernel/debug/tracing/events/kmem/kmalloc/filter
bytes_alloc > 8192
# mkdir /sys/kernel/debug/tracing/instances/test1
# echo 'bytes_alloc > 2048' > /sys/kernel/debug/tracing/instances/test1/events/kmem/kmalloc/filter
# cat /sys/kernel/debug/tracing/events/kmem/kmalloc/filter
bytes_alloc > 8192
# cat /sys/kernel/debug/tracing/instances/test1/events/kmem/kmalloc/filter
bytes_alloc > 2048

We'd like to just move the filter directly from ftrace_event_call to
ftrace_event_file, but there are a couple cases that don't yet have
multibuffer support and therefore have to continue using the current
event_call-based filters. For those cases, a new USE_CALL_FILTER bit
is added to the event_call flags, whose main purpose is to keep the
old behavior for those cases until they can be updated with
multibuffer support; at that point, the USE_CALL_FILTER flag (and the
new associated call_filter_check_discard() function) can go away.

The multibuffer support also made filter_current_check_discard()
redundant, so this change removes that function as well and replaces
it with filter_check_discard() (or call_filter_check_discard() as
appropriate).

Signed-off-by: Tom Zanussi <tom.zanussi@xxxxxxxxxxxxxxx>
---
include/linux/ftrace_event.h | 25 +++-
include/linux/syscalls.h | 4 +-
include/trace/ftrace.h | 7 +-
kernel/trace/trace.c | 40 +++++--
kernel/trace/trace.h | 18 +--
kernel/trace/trace_branch.c | 2 +-
kernel/trace/trace_events.c | 23 ++--
kernel/trace/trace_events_filter.c | 218 ++++++++++++++++++++++++++++-------
kernel/trace/trace_export.c | 2 +-
kernel/trace/trace_functions_graph.c | 4 +-
kernel/trace/trace_kprobe.c | 4 +-
kernel/trace/trace_mmiotrace.c | 4 +-
kernel/trace/trace_sched_switch.c | 4 +-
kernel/trace/trace_syscalls.c | 8 +-
kernel/trace/trace_uprobe.c | 3 +-
15 files changed, 264 insertions(+), 102 deletions(-)

diff --git a/include/linux/ftrace_event.h b/include/linux/ftrace_event.h
index 5eaa746..9abbe63 100644
--- a/include/linux/ftrace_event.h
+++ b/include/linux/ftrace_event.h
@@ -202,6 +202,7 @@ enum {
TRACE_EVENT_FL_NO_SET_FILTER_BIT,
TRACE_EVENT_FL_IGNORE_ENABLE_BIT,
TRACE_EVENT_FL_WAS_ENABLED_BIT,
+ TRACE_EVENT_FL_USE_CALL_FILTER_BIT,
};

/*
@@ -213,6 +214,7 @@ enum {
* WAS_ENABLED - Set and stays set when an event was ever enabled
* (used for module unloading, if a module event is enabled,
* it is best to clear the buffers that used it).
+ * USE_CALL_FILTER - For ftrace internal events, don't use file filter
*/
enum {
TRACE_EVENT_FL_FILTERED = (1 << TRACE_EVENT_FL_FILTERED_BIT),
@@ -220,6 +222,7 @@ enum {
TRACE_EVENT_FL_NO_SET_FILTER = (1 << TRACE_EVENT_FL_NO_SET_FILTER_BIT),
TRACE_EVENT_FL_IGNORE_ENABLE = (1 << TRACE_EVENT_FL_IGNORE_ENABLE_BIT),
TRACE_EVENT_FL_WAS_ENABLED = (1 << TRACE_EVENT_FL_WAS_ENABLED_BIT),
+ TRACE_EVENT_FL_USE_CALL_FILTER = (1 << TRACE_EVENT_FL_USE_CALL_FILTER_BIT),
};

struct ftrace_event_call {
@@ -238,6 +241,7 @@ struct ftrace_event_call {
* bit 2: failed to apply filter
* bit 3: ftrace internal event (do not enable)
* bit 4: Event was enabled by module
+ * bit 5: use call filter rather than file filter
*/
int flags; /* static flags of different events */

@@ -253,6 +257,8 @@ struct ftrace_subsystem_dir;
enum {
FTRACE_EVENT_FL_ENABLED_BIT,
FTRACE_EVENT_FL_RECORDED_CMD_BIT,
+ FTRACE_EVENT_FL_FILTERED_BIT,
+ FTRACE_EVENT_FL_NO_SET_FILTER_BIT,
FTRACE_EVENT_FL_SOFT_MODE_BIT,
FTRACE_EVENT_FL_SOFT_DISABLED_BIT,
};
@@ -261,6 +267,8 @@ enum {
* Ftrace event file flags:
* ENABLED - The event is enabled
* RECORDED_CMD - The comms should be recorded at sched_switch
+ * FILTERED - The event has a filter attached
+ * NO_SET_FILTER - Set when filter has error and is to be ignored
* SOFT_MODE - The event is enabled/disabled by SOFT_DISABLED
* SOFT_DISABLED - When set, do not trace the event (even though its
* tracepoint may be enabled)
@@ -268,6 +276,8 @@ enum {
enum {
FTRACE_EVENT_FL_ENABLED = (1 << FTRACE_EVENT_FL_ENABLED_BIT),
FTRACE_EVENT_FL_RECORDED_CMD = (1 << FTRACE_EVENT_FL_RECORDED_CMD_BIT),
+ FTRACE_EVENT_FL_FILTERED = (1 << FTRACE_EVENT_FL_FILTERED_BIT),
+ FTRACE_EVENT_FL_NO_SET_FILTER = (1 << FTRACE_EVENT_FL_NO_SET_FILTER_BIT),
FTRACE_EVENT_FL_SOFT_MODE = (1 << FTRACE_EVENT_FL_SOFT_MODE_BIT),
FTRACE_EVENT_FL_SOFT_DISABLED = (1 << FTRACE_EVENT_FL_SOFT_DISABLED_BIT),
};
@@ -275,6 +285,7 @@ enum {
struct ftrace_event_file {
struct list_head list;
struct ftrace_event_call *event_call;
+ struct event_filter *filter;
struct dentry *dir;
struct trace_array *tr;
struct ftrace_subsystem_dir *system;
@@ -310,12 +321,16 @@ struct ftrace_event_file {

#define MAX_FILTER_STR_VAL 256 /* Should handle KSYM_SYMBOL_LEN */

-extern void destroy_preds(struct ftrace_event_call *call);
+extern void destroy_preds(struct ftrace_event_file *file);
+extern void destroy_call_preds(struct ftrace_event_call *call);
extern int filter_match_preds(struct event_filter *filter, void *rec);
-extern int filter_current_check_discard(struct ring_buffer *buffer,
- struct ftrace_event_call *call,
- void *rec,
- struct ring_buffer_event *event);
+
+extern int filter_check_discard(struct ftrace_event_file *file, void *rec,
+ struct ring_buffer *buffer,
+ struct ring_buffer_event *event);
+extern int call_filter_check_discard(struct ftrace_event_call *call, void *rec,
+ struct ring_buffer *buffer,
+ struct ring_buffer_event *event);

enum {
FILTER_OTHER = 0,
diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index 7fac04e..10bafa9 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -120,7 +120,7 @@ extern struct trace_event_functions exit_syscall_print_funcs;
.class = &event_class_syscall_enter, \
.event.funcs = &enter_syscall_print_funcs, \
.data = (void *)&__syscall_meta_##sname,\
- .flags = TRACE_EVENT_FL_CAP_ANY, \
+ .flags = TRACE_EVENT_FL_CAP_ANY | TRACE_EVENT_FL_USE_CALL_FILTER,\
}; \
static struct ftrace_event_call __used \
__attribute__((section("_ftrace_events"))) \
@@ -134,7 +134,7 @@ extern struct trace_event_functions exit_syscall_print_funcs;
.class = &event_class_syscall_exit, \
.event.funcs = &exit_syscall_print_funcs, \
.data = (void *)&__syscall_meta_##sname,\
- .flags = TRACE_EVENT_FL_CAP_ANY, \
+ .flags = TRACE_EVENT_FL_CAP_ANY | TRACE_EVENT_FL_USE_CALL_FILTER,\
}; \
static struct ftrace_event_call __used \
__attribute__((section("_ftrace_events"))) \
diff --git a/include/trace/ftrace.h b/include/trace/ftrace.h
index 5c7ab17..52594b2 100644
--- a/include/trace/ftrace.h
+++ b/include/trace/ftrace.h
@@ -437,9 +437,8 @@ static inline notrace int ftrace_get_offsets_##call( \
* { <assign>; } <-- Here we assign the entries by the __field and
* __array macros.
*
- * if (!filter_current_check_discard(buffer, event_call, entry, event))
- * trace_nowake_buffer_unlock_commit(buffer,
- * event, irq_flags, pc);
+ * if (!filter_check_discard(ftrace_file, entry, buffer, event))
+ * trace_buffer_unlock_commit(buffer, event, irq_flags, pc);
* }
*
* static struct trace_event ftrace_event_type_<call> = {
@@ -553,7 +552,7 @@ ftrace_raw_event_##call(void *__data, proto) \
\
{ assign; } \
\
- if (!filter_current_check_discard(buffer, event_call, entry, event)) \
+ if (!filter_check_discard(ftrace_file, entry, buffer, event)) \
trace_buffer_unlock_commit(buffer, event, irq_flags, pc); \
}
/*
diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index 063a92b..489da8b 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -235,13 +235,33 @@ void trace_array_put(struct trace_array *this_tr)
mutex_unlock(&trace_types_lock);
}

-int filter_current_check_discard(struct ring_buffer *buffer,
- struct ftrace_event_call *call, void *rec,
- struct ring_buffer_event *event)
+int filter_check_discard(struct ftrace_event_file *file, void *rec,
+ struct ring_buffer *buffer,
+ struct ring_buffer_event *event)
{
- return filter_check_discard(call, rec, buffer, event);
+ if (unlikely(file->flags & FTRACE_EVENT_FL_FILTERED) &&
+ !filter_match_preds(file->filter, rec)) {
+ ring_buffer_discard_commit(buffer, event);
+ return 1;
+ }
+
+ return 0;
+}
+EXPORT_SYMBOL_GPL(filter_check_discard);
+
+int call_filter_check_discard(struct ftrace_event_call *call, void *rec,
+ struct ring_buffer *buffer,
+ struct ring_buffer_event *event)
+{
+ if (unlikely(call->flags & TRACE_EVENT_FL_FILTERED) &&
+ !filter_match_preds(call->filter, rec)) {
+ ring_buffer_discard_commit(buffer, event);
+ return 1;
+ }
+
+ return 0;
}
-EXPORT_SYMBOL_GPL(filter_current_check_discard);
+EXPORT_SYMBOL_GPL(call_filter_check_discard);

cycle_t buffer_ftrace_now(struct trace_buffer *buf, int cpu)
{
@@ -1633,7 +1653,7 @@ trace_function(struct trace_array *tr,
entry->ip = ip;
entry->parent_ip = parent_ip;

- if (!filter_check_discard(call, entry, buffer, event))
+ if (!call_filter_check_discard(call, entry, buffer, event))
__buffer_unlock_commit(buffer, event);
}

@@ -1717,7 +1737,7 @@ static void __ftrace_trace_stack(struct ring_buffer *buffer,

entry->size = trace.nr_entries;

- if (!filter_check_discard(call, entry, buffer, event))
+ if (!call_filter_check_discard(call, entry, buffer, event))
__buffer_unlock_commit(buffer, event);

out:
@@ -1819,7 +1839,7 @@ ftrace_trace_userstack(struct ring_buffer *buffer, unsigned long flags, int pc)
trace.entries = entry->caller;

save_stack_trace_user(&trace);
- if (!filter_check_discard(call, entry, buffer, event))
+ if (!call_filter_check_discard(call, entry, buffer, event))
__buffer_unlock_commit(buffer, event);

out_drop_count:
@@ -2011,7 +2031,7 @@ int trace_vbprintk(unsigned long ip, const char *fmt, va_list args)
entry->fmt = fmt;

memcpy(entry->buf, tbuffer, sizeof(u32) * len);
- if (!filter_check_discard(call, entry, buffer, event)) {
+ if (!call_filter_check_discard(call, entry, buffer, event)) {
__buffer_unlock_commit(buffer, event);
ftrace_trace_stack(buffer, flags, 6, pc);
}
@@ -2066,7 +2086,7 @@ __trace_array_vprintk(struct ring_buffer *buffer,

memcpy(&entry->buf, tbuffer, len);
entry->buf[len] = '\0';
- if (!filter_check_discard(call, entry, buffer, event)) {
+ if (!call_filter_check_discard(call, entry, buffer, event)) {
__buffer_unlock_commit(buffer, event);
ftrace_trace_stack(buffer, flags, 6, pc);
}
diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h
index d1cf515..12d1a61 100644
--- a/kernel/trace/trace.h
+++ b/kernel/trace/trace.h
@@ -1007,9 +1007,9 @@ struct filter_pred {

extern enum regex_type
filter_parse_regex(char *buff, int len, char **search, int *not);
-extern void print_event_filter(struct ftrace_event_call *call,
+extern void print_event_filter(struct ftrace_event_file *file,
struct trace_seq *s);
-extern int apply_event_filter(struct ftrace_event_call *call,
+extern int apply_event_filter(struct ftrace_event_file *file,
char *filter_string);
extern int apply_subsystem_event_filter(struct ftrace_subsystem_dir *dir,
char *filter_string);
@@ -1020,20 +1020,6 @@ extern int filter_assign_type(const char *type);
struct ftrace_event_field *
trace_find_event_field(struct ftrace_event_call *call, char *name);

-static inline int
-filter_check_discard(struct ftrace_event_call *call, void *rec,
- struct ring_buffer *buffer,
- struct ring_buffer_event *event)
-{
- if (unlikely(call->flags & TRACE_EVENT_FL_FILTERED) &&
- !filter_match_preds(call->filter, rec)) {
- ring_buffer_discard_commit(buffer, event);
- return 1;
- }
-
- return 0;
-}
-
extern void trace_event_enable_cmd_record(bool enable);
extern int event_trace_add_tracer(struct dentry *parent, struct trace_array *tr);
extern int event_trace_del_tracer(struct trace_array *tr);
diff --git a/kernel/trace/trace_branch.c b/kernel/trace/trace_branch.c
index d594da0..697fb9b 100644
--- a/kernel/trace/trace_branch.c
+++ b/kernel/trace/trace_branch.c
@@ -78,7 +78,7 @@ probe_likely_condition(struct ftrace_branch_data *f, int val, int expect)
entry->line = f->line;
entry->correct = val == expect;

- if (!filter_check_discard(call, entry, buffer, event))
+ if (!call_filter_check_discard(call, entry, buffer, event))
__buffer_unlock_commit(buffer, event);

out:
diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c
index 368a4d5..043f833 100644
--- a/kernel/trace/trace_events.c
+++ b/kernel/trace/trace_events.c
@@ -989,7 +989,7 @@ static ssize_t
event_filter_read(struct file *filp, char __user *ubuf, size_t cnt,
loff_t *ppos)
{
- struct ftrace_event_call *call;
+ struct ftrace_event_file *file;
struct trace_seq *s;
int r = -ENODEV;

@@ -1004,12 +1004,12 @@ event_filter_read(struct file *filp, char __user *ubuf, size_t cnt,
trace_seq_init(s);

mutex_lock(&event_mutex);
- call = event_file_data(filp);
- if (call)
- print_event_filter(call, s);
+ file = event_file_data(filp);
+ if (file)
+ print_event_filter(file, s);
mutex_unlock(&event_mutex);

- if (call)
+ if (file)
r = simple_read_from_buffer(ubuf, cnt, ppos, s->buffer, s->len);

kfree(s);
@@ -1021,7 +1021,7 @@ static ssize_t
event_filter_write(struct file *filp, const char __user *ubuf, size_t cnt,
loff_t *ppos)
{
- struct ftrace_event_call *call;
+ struct ftrace_event_file *file;
char *buf;
int err = -ENODEV;

@@ -1039,9 +1039,9 @@ event_filter_write(struct file *filp, const char __user *ubuf, size_t cnt,
buf[cnt] = '\0';

mutex_lock(&event_mutex);
- call = event_file_data(filp);
- if (call)
- err = apply_event_filter(call, buf);
+ file = event_file_data(filp);
+ if (file)
+ err = apply_event_filter(file, buf);
mutex_unlock(&event_mutex);

free_page((unsigned long) buf);
@@ -1539,7 +1539,7 @@ event_create_dir(struct dentry *parent, struct ftrace_event_file *file)
return -1;
}
}
- trace_create_file("filter", 0644, file->dir, call,
+ trace_create_file("filter", 0644, file->dir, file,
&ftrace_event_filter_fops);

trace_create_file("format", 0444, file->dir, call,
@@ -1577,6 +1577,7 @@ static void event_remove(struct ftrace_event_call *call)
if (file->event_call != call)
continue;
ftrace_event_enable_disable(file, 0);
+ destroy_preds(file);
/*
* The do_for_each_event_file() is
* a double loop. After finding the call for this
@@ -1700,7 +1701,7 @@ static void __trace_remove_event_call(struct ftrace_event_call *call)
{
event_remove(call);
trace_destroy_fields(call);
- destroy_preds(call);
+ destroy_call_preds(call);
}

static int probe_remove_event_call(struct ftrace_event_call *call)
diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c
index 97daa8c..2468f56 100644
--- a/kernel/trace/trace_events_filter.c
+++ b/kernel/trace/trace_events_filter.c
@@ -637,10 +637,18 @@ static void append_filter_err(struct filter_parse_state *ps,
free_page((unsigned long) buf);
}

+static inline struct event_filter *event_filter(struct ftrace_event_file *file)
+{
+ if (file->event_call->flags & TRACE_EVENT_FL_USE_CALL_FILTER)
+ return file->event_call->filter;
+ else
+ return file->filter;
+}
+
/* caller must hold event_mutex */
-void print_event_filter(struct ftrace_event_call *call, struct trace_seq *s)
+void print_event_filter(struct ftrace_event_file *file, struct trace_seq *s)
{
- struct event_filter *filter = call->filter;
+ struct event_filter *filter = event_filter(file);

if (filter && filter->filter_string)
trace_seq_printf(s, "%s\n", filter->filter_string);
@@ -766,11 +774,21 @@ static void __free_preds(struct event_filter *filter)
filter->n_preds = 0;
}

-static void filter_disable(struct ftrace_event_call *call)
+static void call_filter_disable(struct ftrace_event_call *call)
{
call->flags &= ~TRACE_EVENT_FL_FILTERED;
}

+static void filter_disable(struct ftrace_event_file *file)
+{
+ struct ftrace_event_call *call = file->event_call;
+
+ if (call->flags & TRACE_EVENT_FL_USE_CALL_FILTER)
+ call_filter_disable(call);
+ else
+ file->flags &= ~FTRACE_EVENT_FL_FILTERED;
+}
+
static void __free_filter(struct event_filter *filter)
{
if (!filter)
@@ -781,16 +799,30 @@ static void __free_filter(struct event_filter *filter)
kfree(filter);
}

+void destroy_call_preds(struct ftrace_event_call *call)
+{
+ __free_filter(call->filter);
+ call->filter = NULL;
+}
+
+static void destroy_file_preds(struct ftrace_event_file *file)
+{
+ __free_filter(file->filter);
+ file->filter = NULL;
+}
+
/*
- * Called when destroying the ftrace_event_call.
- * The call is being freed, so we do not need to worry about
- * the call being currently used. This is for module code removing
+ * Called when destroying the ftrace_event_file.
+ * The file is being freed, so we do not need to worry about
+ * the file being currently used. This is for module code removing
* the tracepoints from within it.
*/
-void destroy_preds(struct ftrace_event_call *call)
+void destroy_preds(struct ftrace_event_file *file)
{
- __free_filter(call->filter);
- call->filter = NULL;
+ if (file->event_call->flags & TRACE_EVENT_FL_USE_CALL_FILTER)
+ destroy_call_preds(file->event_call);
+ else
+ destroy_file_preds(file);
}

static struct event_filter *__alloc_filter(void)
@@ -825,28 +857,56 @@ static int __alloc_preds(struct event_filter *filter, int n_preds)
return 0;
}

-static void filter_free_subsystem_preds(struct event_subsystem *system)
+static inline void __remove_filter(struct ftrace_event_file *file)
{
+ struct ftrace_event_call *call = file->event_call;
+
+ filter_disable(file);
+ if (call->flags & TRACE_EVENT_FL_USE_CALL_FILTER)
+ remove_filter_string(call->filter);
+ else
+ remove_filter_string(file->filter);
+}
+
+static void filter_free_subsystem_preds(struct event_subsystem *system,
+ struct trace_array *tr)
+{
+ struct ftrace_event_file *file;
struct ftrace_event_call *call;

- list_for_each_entry(call, &ftrace_events, list) {
+ list_for_each_entry(file, &tr->events, list) {
+ call = file->event_call;
if (strcmp(call->class->system, system->name) != 0)
continue;

- filter_disable(call);
- remove_filter_string(call->filter);
+ __remove_filter(file);
}
}

-static void filter_free_subsystem_filters(struct event_subsystem *system)
+static inline void __free_subsystem_filter(struct ftrace_event_file *file)
{
+ struct ftrace_event_call *call = file->event_call;
+
+ if (call->flags & TRACE_EVENT_FL_USE_CALL_FILTER) {
+ __free_filter(call->filter);
+ call->filter = NULL;
+ } else {
+ __free_filter(file->filter);
+ file->filter = NULL;
+ }
+}
+
+static void filter_free_subsystem_filters(struct event_subsystem *system,
+ struct trace_array *tr)
+{
+ struct ftrace_event_file *file;
struct ftrace_event_call *call;

- list_for_each_entry(call, &ftrace_events, list) {
+ list_for_each_entry(file, &tr->events, list) {
+ call = file->event_call;
if (strcmp(call->class->system, system->name) != 0)
continue;
- __free_filter(call->filter);
- call->filter = NULL;
+ __free_subsystem_filter(file);
}
}

@@ -1617,15 +1677,85 @@ fail:
return err;
}

+static inline void event_set_filtered_flag(struct ftrace_event_file *file)
+{
+ struct ftrace_event_call *call = file->event_call;
+
+ if (call->flags & TRACE_EVENT_FL_USE_CALL_FILTER)
+ call->flags |= TRACE_EVENT_FL_FILTERED;
+ else
+ file->flags |= FTRACE_EVENT_FL_FILTERED;
+}
+
+static inline void event_set_filter(struct ftrace_event_file *file,
+ struct event_filter *filter)
+{
+ struct ftrace_event_call *call = file->event_call;
+
+ if (call->flags & TRACE_EVENT_FL_USE_CALL_FILTER)
+ rcu_assign_pointer(call->filter, filter);
+ else
+ rcu_assign_pointer(file->filter, filter);
+}
+
+static inline void event_clear_filter(struct ftrace_event_file *file)
+{
+ struct ftrace_event_call *call = file->event_call;
+
+ if (call->flags & TRACE_EVENT_FL_USE_CALL_FILTER)
+ RCU_INIT_POINTER(call->filter, NULL);
+ else
+ RCU_INIT_POINTER(file->filter, NULL);
+}
+
+static inline void
+event_set_no_set_filter_flag(struct ftrace_event_file *file)
+{
+ struct ftrace_event_call *call = file->event_call;
+
+ if (call->flags & TRACE_EVENT_FL_USE_CALL_FILTER)
+ call->flags |= TRACE_EVENT_FL_NO_SET_FILTER;
+ else
+ file->flags |= FTRACE_EVENT_FL_NO_SET_FILTER;
+}
+
+static inline void
+event_clear_no_set_filter_flag(struct ftrace_event_file *file)
+{
+ struct ftrace_event_call *call = file->event_call;
+
+ if (call->flags & TRACE_EVENT_FL_USE_CALL_FILTER)
+ call->flags &= ~TRACE_EVENT_FL_NO_SET_FILTER;
+ else
+ file->flags &= ~FTRACE_EVENT_FL_NO_SET_FILTER;
+}
+
+static inline bool
+event_no_set_filter_flag(struct ftrace_event_file *file)
+{
+ struct ftrace_event_call *call = file->event_call;
+
+ if (file->flags & FTRACE_EVENT_FL_NO_SET_FILTER)
+ return true;
+
+ if ((call->flags & TRACE_EVENT_FL_USE_CALL_FILTER) &&
+ (call->flags & TRACE_EVENT_FL_NO_SET_FILTER))
+ return true;
+
+ return false;
+}
+
struct filter_list {
struct list_head list;
struct event_filter *filter;
};

static int replace_system_preds(struct event_subsystem *system,
+ struct trace_array *tr,
struct filter_parse_state *ps,
char *filter_string)
{
+ struct ftrace_event_file *file;
struct ftrace_event_call *call;
struct filter_list *filter_item;
struct filter_list *tmp;
@@ -1633,8 +1763,8 @@ static int replace_system_preds(struct event_subsystem *system,
bool fail = true;
int err;

- list_for_each_entry(call, &ftrace_events, list) {
-
+ list_for_each_entry(file, &tr->events, list) {
+ call = file->event_call;
if (strcmp(call->class->system, system->name) != 0)
continue;

@@ -1644,18 +1774,20 @@ static int replace_system_preds(struct event_subsystem *system,
*/
err = replace_preds(call, NULL, ps, filter_string, true);
if (err)
- call->flags |= TRACE_EVENT_FL_NO_SET_FILTER;
+ event_set_no_set_filter_flag(file);
else
- call->flags &= ~TRACE_EVENT_FL_NO_SET_FILTER;
+ event_clear_no_set_filter_flag(file);
}

- list_for_each_entry(call, &ftrace_events, list) {
+ list_for_each_entry(file, &tr->events, list) {
struct event_filter *filter;

+ call = file->event_call;
+
if (strcmp(call->class->system, system->name) != 0)
continue;

- if (call->flags & TRACE_EVENT_FL_NO_SET_FILTER)
+ if (event_no_set_filter_flag(file))
continue;

filter_item = kzalloc(sizeof(*filter_item), GFP_KERNEL);
@@ -1676,17 +1808,17 @@ static int replace_system_preds(struct event_subsystem *system,

err = replace_preds(call, filter, ps, filter_string, false);
if (err) {
- filter_disable(call);
+ filter_disable(file);
parse_error(ps, FILT_ERR_BAD_SUBSYS_FILTER, 0);
append_filter_err(ps, filter);
} else
- call->flags |= TRACE_EVENT_FL_FILTERED;
+ event_set_filtered_flag(file);
/*
* Regardless of if this returned an error, we still
* replace the filter for the call.
*/
- filter = call->filter;
- rcu_assign_pointer(call->filter, filter_item->filter);
+ filter = event_filter(file);
+ event_set_filter(file, filter_item->filter);
filter_item->filter = filter;

fail = false;
@@ -1816,6 +1948,7 @@ static int create_filter(struct ftrace_event_call *call,
* and always remembers @filter_str.
*/
static int create_system_filter(struct event_subsystem *system,
+ struct trace_array *tr,
char *filter_str, struct event_filter **filterp)
{
struct event_filter *filter = NULL;
@@ -1824,7 +1957,7 @@ static int create_system_filter(struct event_subsystem *system,

err = create_filter_start(filter_str, true, &ps, &filter);
if (!err) {
- err = replace_system_preds(system, ps, filter_str);
+ err = replace_system_preds(system, tr, ps, filter_str);
if (!err) {
/* System filters just show a default message */
kfree(filter->filter_string);
@@ -1840,20 +1973,25 @@ static int create_system_filter(struct event_subsystem *system,
}

/* caller must hold event_mutex */
-int apply_event_filter(struct ftrace_event_call *call, char *filter_string)
+int apply_event_filter(struct ftrace_event_file *file, char *filter_string)
{
+ struct ftrace_event_call *call = file->event_call;
struct event_filter *filter;
int err;

if (!strcmp(strstrip(filter_string), "0")) {
- filter_disable(call);
- filter = call->filter;
+ filter_disable(file);
+ filter = event_filter(file);
+
if (!filter)
return 0;
- RCU_INIT_POINTER(call->filter, NULL);
+
+ event_clear_filter(file);
+
/* Make sure the filter is not being used */
synchronize_sched();
__free_filter(filter);
+
return 0;
}

@@ -1866,14 +2004,15 @@ int apply_event_filter(struct ftrace_event_call *call, char *filter_string)
* string
*/
if (filter) {
- struct event_filter *tmp = call->filter;
+ struct event_filter *tmp;

+ tmp = event_filter(file);
if (!err)
- call->flags |= TRACE_EVENT_FL_FILTERED;
+ event_set_filtered_flag(file);
else
- filter_disable(call);
+ filter_disable(file);

- rcu_assign_pointer(call->filter, filter);
+ event_set_filter(file, filter);

if (tmp) {
/* Make sure the call is done with the filter */
@@ -1889,6 +2028,7 @@ int apply_subsystem_event_filter(struct ftrace_subsystem_dir *dir,
char *filter_string)
{
struct event_subsystem *system = dir->subsystem;
+ struct trace_array *tr = dir->tr;
struct event_filter *filter;
int err = 0;

@@ -1901,18 +2041,18 @@ int apply_subsystem_event_filter(struct ftrace_subsystem_dir *dir,
}

if (!strcmp(strstrip(filter_string), "0")) {
- filter_free_subsystem_preds(system);
+ filter_free_subsystem_preds(system, tr);
remove_filter_string(system->filter);
filter = system->filter;
system->filter = NULL;
/* Ensure all filters are no longer used */
synchronize_sched();
- filter_free_subsystem_filters(system);
+ filter_free_subsystem_filters(system, tr);
__free_filter(filter);
goto out_unlock;
}

- err = create_system_filter(system, filter_string, &filter);
+ err = create_system_filter(system, tr, filter_string, &filter);
if (filter) {
/*
* No event actually uses the system filter
diff --git a/kernel/trace/trace_export.c b/kernel/trace/trace_export.c
index d21a746..7c3e3e7 100644
--- a/kernel/trace/trace_export.c
+++ b/kernel/trace/trace_export.c
@@ -180,7 +180,7 @@ struct ftrace_event_call __used event_##call = { \
.event.type = etype, \
.class = &event_class_ftrace_##call, \
.print_fmt = print, \
- .flags = TRACE_EVENT_FL_IGNORE_ENABLE, \
+ .flags = TRACE_EVENT_FL_IGNORE_ENABLE | TRACE_EVENT_FL_USE_CALL_FILTER, \
}; \
struct ftrace_event_call __used \
__attribute__((section("_ftrace_events"))) *__event_##call = &event_##call;
diff --git a/kernel/trace/trace_functions_graph.c b/kernel/trace/trace_functions_graph.c
index e08c030..80387d1 100644
--- a/kernel/trace/trace_functions_graph.c
+++ b/kernel/trace/trace_functions_graph.c
@@ -270,7 +270,7 @@ int __trace_graph_entry(struct trace_array *tr,
return 0;
entry = ring_buffer_event_data(event);
entry->graph_ent = *trace;
- if (!filter_current_check_discard(buffer, call, entry, event))
+ if (!call_filter_check_discard(call, entry, buffer, event))
__buffer_unlock_commit(buffer, event);

return 1;
@@ -385,7 +385,7 @@ void __trace_graph_return(struct trace_array *tr,
return;
entry = ring_buffer_event_data(event);
entry->ret = *trace;
- if (!filter_current_check_discard(buffer, call, entry, event))
+ if (!call_filter_check_discard(call, entry, buffer, event))
__buffer_unlock_commit(buffer, event);
}

diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c
index 243f683..dae9541 100644
--- a/kernel/trace/trace_kprobe.c
+++ b/kernel/trace/trace_kprobe.c
@@ -835,7 +835,7 @@ __kprobe_trace_func(struct trace_probe *tp, struct pt_regs *regs,
entry->ip = (unsigned long)tp->rp.kp.addr;
store_trace_args(sizeof(*entry), tp, regs, (u8 *)&entry[1], dsize);

- if (!filter_current_check_discard(buffer, call, entry, event))
+ if (!filter_check_discard(ftrace_file, entry, buffer, event))
trace_buffer_unlock_commit_regs(buffer, event,
irq_flags, pc, regs);
}
@@ -884,7 +884,7 @@ __kretprobe_trace_func(struct trace_probe *tp, struct kretprobe_instance *ri,
entry->ret_ip = (unsigned long)ri->ret_addr;
store_trace_args(sizeof(*entry), tp, regs, (u8 *)&entry[1], dsize);

- if (!filter_current_check_discard(buffer, call, entry, event))
+ if (!filter_check_discard(ftrace_file, entry, buffer, event))
trace_buffer_unlock_commit_regs(buffer, event,
irq_flags, pc, regs);
}
diff --git a/kernel/trace/trace_mmiotrace.c b/kernel/trace/trace_mmiotrace.c
index b3dcfb2..0abd9b8 100644
--- a/kernel/trace/trace_mmiotrace.c
+++ b/kernel/trace/trace_mmiotrace.c
@@ -323,7 +323,7 @@ static void __trace_mmiotrace_rw(struct trace_array *tr,
entry = ring_buffer_event_data(event);
entry->rw = *rw;

- if (!filter_check_discard(call, entry, buffer, event))
+ if (!call_filter_check_discard(call, entry, buffer, event))
trace_buffer_unlock_commit(buffer, event, 0, pc);
}

@@ -353,7 +353,7 @@ static void __trace_mmiotrace_map(struct trace_array *tr,
entry = ring_buffer_event_data(event);
entry->map = *map;

- if (!filter_check_discard(call, entry, buffer, event))
+ if (!call_filter_check_discard(call, entry, buffer, event))
trace_buffer_unlock_commit(buffer, event, 0, pc);
}

diff --git a/kernel/trace/trace_sched_switch.c b/kernel/trace/trace_sched_switch.c
index 4e98e3b..3f34dc9 100644
--- a/kernel/trace/trace_sched_switch.c
+++ b/kernel/trace/trace_sched_switch.c
@@ -45,7 +45,7 @@ tracing_sched_switch_trace(struct trace_array *tr,
entry->next_state = next->state;
entry->next_cpu = task_cpu(next);

- if (!filter_check_discard(call, entry, buffer, event))
+ if (!call_filter_check_discard(call, entry, buffer, event))
trace_buffer_unlock_commit(buffer, event, flags, pc);
}

@@ -101,7 +101,7 @@ tracing_sched_wakeup_trace(struct trace_array *tr,
entry->next_state = wakee->state;
entry->next_cpu = task_cpu(wakee);

- if (!filter_check_discard(call, entry, buffer, event))
+ if (!call_filter_check_discard(call, entry, buffer, event))
trace_buffer_unlock_commit(buffer, event, flags, pc);
}

diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c
index 559329d..32644ee 100644
--- a/kernel/trace/trace_syscalls.c
+++ b/kernel/trace/trace_syscalls.c
@@ -336,8 +336,8 @@ static void ftrace_syscall_enter(void *data, struct pt_regs *regs, long id)
entry->nr = syscall_nr;
syscall_get_arguments(current, regs, 0, sys_data->nb_args, entry->args);

- if (!filter_current_check_discard(buffer, sys_data->enter_event,
- entry, event))
+ if (!call_filter_check_discard(sys_data->enter_event, entry,
+ buffer, event))
trace_current_buffer_unlock_commit(buffer, event,
irq_flags, pc);
}
@@ -377,8 +377,8 @@ static void ftrace_syscall_exit(void *data, struct pt_regs *regs, long ret)
entry->nr = syscall_nr;
entry->ret = syscall_get_return_value(current, regs);

- if (!filter_current_check_discard(buffer, sys_data->exit_event,
- entry, event))
+ if (!call_filter_check_discard(sys_data->exit_event, entry,
+ buffer, event))
trace_current_buffer_unlock_commit(buffer, event,
irq_flags, pc);
}
diff --git a/kernel/trace/trace_uprobe.c b/kernel/trace/trace_uprobe.c
index 272261b..b6dcc42 100644
--- a/kernel/trace/trace_uprobe.c
+++ b/kernel/trace/trace_uprobe.c
@@ -128,6 +128,7 @@ alloc_trace_uprobe(const char *group, const char *event, int nargs, bool is_ret)
if (is_ret)
tu->consumer.ret_handler = uretprobe_dispatcher;
init_trace_uprobe_filter(&tu->filter);
+ tu->call.flags |= TRACE_EVENT_FL_USE_CALL_FILTER;
return tu;

error:
@@ -561,7 +562,7 @@ static void uprobe_trace_print(struct trace_uprobe *tu,
for (i = 0; i < tu->nr_args; i++)
call_fetch(&tu->args[i].fetch, regs, data + tu->args[i].offset);

- if (!filter_current_check_discard(buffer, call, entry, event))
+ if (!call_filter_check_discard(call, entry, buffer, event))
trace_buffer_unlock_commit(buffer, event, 0, 0);
}

--
1.8.3.1

--
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/