Skip to contentSkip to conversation

tracing: add per-subsystem filtering

PATCH In review Tom Zanussi 4 participants 9 messages Mar 22, 2009 – Mar 25, 2009 Tracing

lore
Share & tools
lore
#0 lore raw

PATCH 4/4 tracing: add per-subsystem filtering

This patch adds per-subsystem filtering to the event tracing subsystem.

It adds a 'filter' debugfs file to each subsystem directory. This file can be written to to set filters; reading from it will display the current set of filters set for that subsystem.

Basically what it does is propagate the filter down to each event contained in the subsystem. If a particular event doesn't have a field with the name specified in the filter, it simply doesn't get set for that event. You can verify whether or not the filter was set for a particular event by looking at the filter file for that event.

As with per-event filters, compound expressions are supported, echoing '0' to the subsystem's filter file clears all filters in the subsystem, etc.

---
 kernel/trace/trace.h               |   15 ++++++
 kernel/trace/trace_events.c        |   86 +++++++++++++++++++++++++++++++++---
 kernel/trace/trace_events_filter.c |   80 +++++++++++++++++++++++++++++++++
 3 files changed, 175 insertions(+), 6 deletions(-)
kernel/trace/trace.h +15 -0
index d9eb39e..f267723 100644
@@ -804,6 +804,18 @@ struct ftrace_event_call {
 #endif
 };
 
+struct event_subsystem {
+	struct list_head	list;
+	const char		*name;
+	struct dentry		*entry;
+	struct filter_pred	**preds;
+};
+
+#define events_for_each(event)						\
+	for (event = __start_ftrace_events;				\
+	     (unsigned long)event < (unsigned long)__stop_ftrace_events; \
+	     event++)
+
 #define MAX_FILTER_PRED 8
 
 struct filter_pred;
@@ -832,6 +844,9 @@ extern int filter_add_pred(struct ftrace_event_call *call,
 			   struct filter_pred *pred);
 extern void filter_free_preds(struct ftrace_event_call *call);
 extern int filter_match_preds(struct ftrace_event_call *call, void *rec);
+extern void filter_free_subsystem_preds(struct event_subsystem *system);
+extern int filter_add_subsystem_pred(struct event_subsystem *system,
+				     struct filter_pred *pred);
 
 void event_trace_printk(unsigned long ip, const char *fmt, ...);
 extern struct ftrace_event_call __start_ftrace_events[];
kernel/trace/trace_events.c +80 -6

116 lines not on this page · raw message

kernel/trace/trace_events_filter.c +80 -0

92 lines not on this page · raw message

Signature · 1 line
1.5.6.3
Acked-by on v1 4/4
#1 lore raw
Quoted 167 lines · Tom Zanussi

This patch adds per-subsystem filtering to the event tracing subsystem.

It adds a 'filter' debugfs file to each subsystem directory. This file can be written to to set filters; reading from it will display the current set of filters set for that subsystem.

Basically what it does is propagate the filter down to each event contained in the subsystem. If a particular event doesn't have a field with the name specified in the filter, it simply doesn't get set for that event. You can verify whether or not the filter was set for a particular event by looking at the filter file for that event.

As with per-event filters, compound expressions are supported, echoing '0' to the subsystem's filter file clears all filters in the subsystem, etc.

---
 kernel/trace/trace.h               |   15 ++++++
 kernel/trace/trace_events.c        |   86 +++++++++++++++++++++++++++++++++---
 kernel/trace/trace_events_filter.c |   80 +++++++++++++++++++++++++++++++++
 3 files changed, 175 insertions(+), 6 deletions(-)
kernel/trace/trace.h +15 -0
index d9eb39e..f267723 100644
@@ -804,6 +804,18 @@ struct ftrace_event_call {
 #endif
 };
 
+struct event_subsystem {
+	struct list_head	list;
+	const char		*name;
+	struct dentry		*entry;
+	struct filter_pred	**preds;
+};
+
+#define events_for_each(event)						\
+	for (event = __start_ftrace_events;				\
+	     (unsigned long)event < (unsigned long)__stop_ftrace_events; \
+	     event++)
+
 #define MAX_FILTER_PRED 8
 
 struct filter_pred;
@@ -832,6 +844,9 @@ extern int filter_add_pred(struct ftrace_event_call *call,
 			   struct filter_pred *pred);
 extern void filter_free_preds(struct ftrace_event_call *call);
 extern int filter_match_preds(struct ftrace_event_call *call, void *rec);
+extern void filter_free_subsystem_preds(struct event_subsystem *system);
+extern int filter_add_subsystem_pred(struct event_subsystem *system,
+				     struct filter_pred *pred);
 
 void event_trace_printk(unsigned long ip, const char *fmt, ...);
 extern struct ftrace_event_call __start_ftrace_events[];
kernel/trace/trace_events.c +72 -6
index 97470c4..97d4daa 100644
@@ -524,6 +524,71 @@ event_filter_write(struct file *filp, const char __user *ubuf, size_t cnt,
 	return cnt;
 }
 
+static ssize_t
+subsystem_filter_read(struct file *filp, char __user *ubuf, size_t cnt,
+		      loff_t *ppos)
+{
+	struct event_subsystem *system = filp->private_data;
+	struct trace_seq *s;
+	int r;
+
+	if (*ppos)
+		return 0;
+
+	s = kmalloc(sizeof(*s), GFP_KERNEL);
+	if (!s)
+		return -ENOMEM;
+
+	trace_seq_init(s);
+
+	r = filter_print_preds(system->preds, s->buffer);
+	r = simple_read_from_buffer(ubuf, cnt, ppos, s->buffer, r);
+
+	kfree(s);
+
+	return r;
+}
+
+static ssize_t
+subsystem_filter_write(struct file *filp, const char __user *ubuf, size_t cnt,
+		       loff_t *ppos)
+{
+	struct event_subsystem *system = filp->private_data;
+	char buf[64], *pbuf = buf;
+	struct filter_pred *pred;
+	int err;
+
+	if (cnt >= sizeof(buf))
+		return -EINVAL;
+
+	if (copy_from_user(&buf, ubuf, cnt))
+		return -EFAULT;
+
+	pred = kzalloc(sizeof(*pred), GFP_KERNEL);
+	if (!pred)
+		return -ENOMEM;
+
+	err = filter_parse(&pbuf, pred);
+	if (err < 0) {
+		filter_free_pred(pred);
+		return err;
+	}
+
+	if (pred->clear) {
+		filter_free_subsystem_preds(system);
+		return cnt;
+	}
+
+	if (filter_add_subsystem_pred(system, pred)) {
+		filter_free_pred(pred);
+		return -EINVAL;
+	}
+
+	*ppos += cnt;
+
+	return cnt;
+}
+
 static const struct seq_operations show_event_seq_ops = {
 	.start = t_start,
 	.next = t_next,
@@ -575,6 +640,12 @@ static const struct file_operations ftrace_event_filter_fops = {
 	.write = event_filter_write,
 };
 
+static const struct file_operations ftrace_subsystem_filter_fops = {
+	.open = tracing_open_generic,
+	.read = subsystem_filter_read,
+	.write = subsystem_filter_write,
+};
+
 static struct dentry *event_trace_events_dir(void)
 {
 	static struct dentry *d_tracer;
@@ -595,18 +666,13 @@ static struct dentry *event_trace_events_dir(void)
 	return d_events;
 }
 
-struct event_subsystem {
-	struct list_head	list;
-	const char		*name;
-	struct dentry		*entry;
-};
-
 static LIST_HEAD(event_subsystems);
 
 static struct dentry *
 event_subsystem_dir(const char *name, struct dentry *d_events)
 {
 	struct event_subsystem *system;
+	struct dentry *entry;
 
 	/* First see if we did not already create this dir */
 	list_for_each_entry(system, &event_subsystems, list) {

kernel/trace/trace_events.c event_subsystem_dir(const char *name, struct dentry *d_events) · line 704

Quoted 6 lines
kernel/trace/trace_events.c +3 -0
@@ -633,6 +699,14 @@ event_subsystem_dir(const char *name, struct dentry *d_events)
 	system->name = name;
 	list_add(&system->list, &event_subsystems);
 
+	system->preds = NULL;
+
+	entry = debugfs_create_file("filter", 0444, system->entry, system,

Should be 0644.

Quoted 77 lines
kernel/trace/trace_events.c +5 -0
+				    &ftrace_subsystem_filter_fops);
+	if (!entry)
+		pr_warning("Could not create debugfs "
+			   "'%s/filter' entry\n", name);
+
 	return system->entry;
 }
 
kernel/trace/trace_events_filter.c +54 -0
index 8e8c5fa..1ab20ce 100644
@@ -181,6 +181,27 @@ void filter_free_preds(struct ftrace_event_call *call)
 	}
 }
 
+void filter_free_subsystem_preds(struct event_subsystem *system)
+{
+	struct ftrace_event_call *call = __start_ftrace_events;
+	int i;
+
+	if (system->preds) {
+		for (i = 0; i < MAX_FILTER_PRED; i++)
+			filter_free_pred(system->preds[i]);
+		kfree(system->preds);
+		system->preds = NULL;
+	}
+
+	events_for_each(call) {
+		if (!call->name || !call->regfunc)
+			continue;
+
+		if (!strcmp(call->system, system->name))
+			filter_free_preds(call);
+	}
+}
+
 static int __filter_add_pred(struct ftrace_event_call *call,
 			     struct filter_pred *pred)
 {
@@ -250,6 +271,65 @@ int filter_add_pred(struct ftrace_event_call *call, struct filter_pred *pred)
 	return __filter_add_pred(call, pred);
 }
 
+static struct filter_pred *copy_pred(struct filter_pred *pred)
+{
+	struct filter_pred *new_pred = kmalloc(sizeof(*pred), GFP_KERNEL);
+	if (!new_pred)
+		return NULL;
+
+	memcpy(new_pred, pred, sizeof(*pred));
+	if (pred->str_val) {
+		new_pred->str_val = kstrdup(pred->str_val, GFP_KERNEL);
+		new_pred->field_name = kstrdup(pred->field_name, GFP_KERNEL);
+		if (!new_pred->str_val) {
+			kfree(new_pred);
+			return NULL;
+		}
+	}
+
+	return new_pred;
+}
+
+int filter_add_subsystem_pred(struct event_subsystem *system,
+			      struct filter_pred *pred)
+{
+	struct ftrace_event_call *call = __start_ftrace_events;
+	struct filter_pred *event_pred;
+	int i;
+
+	if (system->preds && !pred->compound)
+		filter_free_subsystem_preds(system);
+
+	if (!system->preds) {
+		system->preds = kzalloc(MAX_FILTER_PRED * sizeof(pred),
+					GFP_KERNEL);
+		if (!system->preds)

kernel/trace/trace_events_filter.c int filter_add_pred(struct ftrace_event_call *call, struct filter_pred *pred) · line 316

Quoted 10 lines
kernel/trace/trace_events_filter.c +10 -0
+			return -ENOMEM;
+	}
+
+	for (i = 0; i < MAX_FILTER_PRED; i++) {
+		if (!system->preds[i]) {
+			system->preds[i] = pred;
+			break;
+		}
+		if (i == MAX_FILTER_PRED - 1)
+			return -EINVAL;

Shouldn't it be i == MAX_FILTER_PRED ?

Quoted 24 lines
kernel/trace/trace_events_filter.c +16 -0
+	}
+
+	events_for_each(call) {
+		if (!call->name || !call->regfunc)
+			continue;
+
+		if (!strcmp(call->system, system->name)) {
+			event_pred = copy_pred(pred);
+			if (event_pred)
+				filter_add_pred(call, event_pred);
+		}
+	}
+
+	return 0;
+}
+
 int filter_parse(char **pbuf, struct filter_pred *pred)
 {
 	char *tmp, *tok, *val_str = NULL;
Signature · 1 line
1.5.6.3

Looks good too.

#4 lore raw
In reply to Tom Zanussi
Commit-ID:  cfb180f3e71b2a280a254c8646a9ab1beab63f84
Gitweb:     http://git.kernel.org/tip/cfb180f3e71b2a280a254c8646a9ab1beab63f84
Author:     Tom Zanussi <[email protected]>
AuthorDate: Sun, 22 Mar 2009 03:31:17 -0500
Committer:  Ingo Molnar <[email protected]>
CommitDate: Sun, 22 Mar 2009 18:38:47 +0100

tracing: add per-subsystem filtering

This patch adds per-subsystem filtering to the event tracing subsystem.

It adds a 'filter' debugfs file to each subsystem directory. This file can be written to to set filters; reading from it will display the current set of filters set for that subsystem.

Basically what it does is propagate the filter down to each event contained in the subsystem. If a particular event doesn't have a field with the name specified in the filter, it simply doesn't get set for that event. You can verify whether or not the filter was set for a particular event by looking at the filter file for that event.

As with per-event filters, compound expressions are supported, echoing '0' to the subsystem's filter file clears all filters in the subsystem, etc.

LKML-Reference: <1237710677.7703.49.camel@charm-linux>

---
 kernel/trace/trace.h               |   15 ++++++
 kernel/trace/trace_events.c        |   86 +++++++++++++++++++++++++++++++++---
 kernel/trace/trace_events_filter.c |   80 +++++++++++++++++++++++++++++++++
 3 files changed, 175 insertions(+), 6 deletions(-)
kernel/trace/trace.h +15 -0
index d9eb39e..f267723 100644
@@ -804,6 +804,18 @@ struct ftrace_event_call {
 #endif
 };
 
+struct event_subsystem {
+	struct list_head	list;
+	const char		*name;
+	struct dentry		*entry;
+	struct filter_pred	**preds;
+};
+
+#define events_for_each(event)						\
+	for (event = __start_ftrace_events;				\
+	     (unsigned long)event < (unsigned long)__stop_ftrace_events; \
+	     event++)
+
 #define MAX_FILTER_PRED 8
 
 struct filter_pred;
@@ -832,6 +844,9 @@ extern int filter_add_pred(struct ftrace_event_call *call,
 			   struct filter_pred *pred);
 extern void filter_free_preds(struct ftrace_event_call *call);
 extern int filter_match_preds(struct ftrace_event_call *call, void *rec);
+extern void filter_free_subsystem_preds(struct event_subsystem *system);
+extern int filter_add_subsystem_pred(struct event_subsystem *system,
+				     struct filter_pred *pred);
 
 void event_trace_printk(unsigned long ip, const char *fmt, ...);
 extern struct ftrace_event_call __start_ftrace_events[];
kernel/trace/trace_events.c +80 -6

116 lines not on this page · raw message

kernel/trace/trace_events_filter.c +80 -0

92 lines not on this page · raw message

#2 lore raw
In reply to Frederic Weisbecker
Quoted 19 lines · Frederic Weisbecker
Quoted 19 lines
kernel/trace/trace_events_filter.c +19 -0
+int filter_add_subsystem_pred(struct event_subsystem *system,
+			      struct filter_pred *pred)
+{
+	struct ftrace_event_call *call = __start_ftrace_events;
+	struct filter_pred *event_pred;
+	int i;
+
+	if (system->preds && !pred->compound)
+		filter_free_subsystem_preds(system);
+
+	if (!system->preds) {
+		system->preds = kzalloc(MAX_FILTER_PRED * sizeof(pred),
+					GFP_KERNEL);
+		if (!system->preds)
+			return -ENOMEM;
+	}
+
+	for (i = 0; i < MAX_FILTER_PRED; i++) {
+		if (!system->preds[i]) {
Quoted 8 lines

kernel/trace/trace_events_filter.c int filter_add_pred(struct ftrace_event_call *call, struct filter_pred *pred) · line 316

Quoted 5 lines
kernel/trace/trace_events_filter.c +5 -0
+			system->preds[i] = pred;
+			break;
+		}
+		if (i == MAX_FILTER_PRED - 1)
+			return -EINVAL;

Shouldn't it be i == MAX_FILTER_PRED ?

Here we search for a free slot in the array of sub-expressions of the subsystem level filters. That condition cannot even be true inside a 'i < MAX_FILTER_PRED' loop.

Checking for i==MAX would be fine if done outside of the loop - and should probably be done that way. But the code is correct this way too i think.

Ingo

#3 lore raw
Quoted 31 lines · Ingo Molnar
Quoted 19 lines · Frederic Weisbecker
Quoted 19 lines
kernel/trace/trace_events_filter.c +19 -0
+int filter_add_subsystem_pred(struct event_subsystem *system,
+			      struct filter_pred *pred)
+{
+	struct ftrace_event_call *call = __start_ftrace_events;
+	struct filter_pred *event_pred;
+	int i;
+
+	if (system->preds && !pred->compound)
+		filter_free_subsystem_preds(system);
+
+	if (!system->preds) {
+		system->preds = kzalloc(MAX_FILTER_PRED * sizeof(pred),
+					GFP_KERNEL);
+		if (!system->preds)
+			return -ENOMEM;
+	}
+
+	for (i = 0; i < MAX_FILTER_PRED; i++) {
+		if (!system->preds[i]) {
Quoted 8 lines

kernel/trace/trace_events_filter.c int filter_add_pred(struct ftrace_event_call *call, struct filter_pred *pred) · line 316

Quoted 5 lines
kernel/trace/trace_events_filter.c +5 -0
+			system->preds[i] = pred;
+			break;
+		}
+		if (i == MAX_FILTER_PRED - 1)
+			return -EINVAL;

Shouldn't it be i == MAX_FILTER_PRED ?

Quoted 3 lines

Here we search for a free slot in the array of sub-expressions of the subsystem level filters. That condition cannot even be true inside a 'i < MAX_FILTER_PRED' loop.

Darn, I should sleep more!

Quoted 4 lines

Checking for i==MAX would be fine if done outside of the loop - and should probably be done that way. But the code is correct this way too i think.

Yes, at least it's harmless.

maintainer
#5 lore raw
In reply to Tom Zanussi
Quoted 31 lines · Tom Zanussi
kernel/trace/trace.h +15 -0
+struct event_subsystem {
+	struct list_head	list;
+	const char		*name;
+	struct dentry		*entry;
+	struct filter_pred	**preds;
+};
+
+#define events_for_each(event)						\
+	for (event = __start_ftrace_events;				\
+	     (unsigned long)event < (unsigned long)__stop_ftrace_events; \
+	     event++)
+
 #define MAX_FILTER_PRED 8
 
 struct filter_pred;
@@ -832,6 +844,9 @@ extern int filter_add_pred(struct ftrace_event_call *call,
 			   struct filter_pred *pred);
 extern void filter_free_preds(struct ftrace_event_call *call);
 extern int filter_match_preds(struct ftrace_event_call *call, void *rec);
+extern void filter_free_subsystem_preds(struct event_subsystem *system);
+extern int filter_add_subsystem_pred(struct event_subsystem *system,
+				     struct filter_pred *pred);
 
 void event_trace_printk(unsigned long ip, const char *fmt, ...);
 extern struct ftrace_event_call __start_ftrace_events[];
kernel/trace/trace_events.c +0 -0
index 97470c4..97d4daa 100644

kernel/trace/trace_events.c event_filter_write(struct file *filp, const char __user *ubuf, size_t cnt, · line 532

Quoted 9 lines
kernel/trace/trace_events.c +6 -0
@@ -524,6 +524,71 @@ event_filter_write(struct file *filp, const char __user *ubuf, size_t cnt,
 	return cnt;
 }
 
+static ssize_t
+subsystem_filter_read(struct file *filp, char __user *ubuf, size_t cnt,
+		      loff_t *ppos)
+{
+	struct event_subsystem *system = filp->private_data;
+	struct trace_seq *s;

Again, trace_seq is not used, might as well use your own buffer.

Quoted 36 lines
kernel/trace/trace_events.c +36 -0
+	int r;
+
+	if (*ppos)
+		return 0;
+
+	s = kmalloc(sizeof(*s), GFP_KERNEL);
+	if (!s)
+		return -ENOMEM;
+
+	trace_seq_init(s);
+
+	r = filter_print_preds(system->preds, s->buffer);
+	r = simple_read_from_buffer(ubuf, cnt, ppos, s->buffer, r);
+
+	kfree(s);
+
+	return r;
+}
+
+static ssize_t
+subsystem_filter_write(struct file *filp, const char __user *ubuf, size_t cnt,
+		       loff_t *ppos)
+{
+	struct event_subsystem *system = filp->private_data;
+	char buf[64], *pbuf = buf;
+	struct filter_pred *pred;
+	int err;
+
+	if (cnt >= sizeof(buf))
+		return -EINVAL;
+
+	if (copy_from_user(&buf, ubuf, cnt))
+		return -EFAULT;
+
+	pred = kzalloc(sizeof(*pred), GFP_KERNEL);
+	if (!pred)

kernel/trace/trace_events.c event_filter_write(struct file *filp, const char __user *ubuf, size_t cnt, · line 578

Quoted 10 lines
kernel/trace/trace_events.c +10 -0
+		return -ENOMEM;
+
+	err = filter_parse(&pbuf, pred);
+	if (err < 0) {
+		filter_free_pred(pred);
+		return err;
+	}
+
+	if (pred->clear) {
+		filter_free_subsystem_preds(system);

is "system" correct here?

Quoted 102 lines
kernel/trace/trace_events.c +28 -6
+		return cnt;
+	}
+
+	if (filter_add_subsystem_pred(system, pred)) {
+		filter_free_pred(pred);
+		return -EINVAL;
+	}
+
+	*ppos += cnt;
+
+	return cnt;
+}
+
 static const struct seq_operations show_event_seq_ops = {
 	.start = t_start,
 	.next = t_next,
@@ -575,6 +640,12 @@ static const struct file_operations ftrace_event_filter_fops = {
 	.write = event_filter_write,
 };
 
+static const struct file_operations ftrace_subsystem_filter_fops = {
+	.open = tracing_open_generic,
+	.read = subsystem_filter_read,
+	.write = subsystem_filter_write,
+};
+
 static struct dentry *event_trace_events_dir(void)
 {
 	static struct dentry *d_tracer;
@@ -595,18 +666,13 @@ static struct dentry *event_trace_events_dir(void)
 	return d_events;
 }
 
-struct event_subsystem {
-	struct list_head	list;
-	const char		*name;
-	struct dentry		*entry;
-};
-
 static LIST_HEAD(event_subsystems);
 
 static struct dentry *
 event_subsystem_dir(const char *name, struct dentry *d_events)
 {
 	struct event_subsystem *system;
+	struct dentry *entry;
 
 	/* First see if we did not already create this dir */
 	list_for_each_entry(system, &event_subsystems, list) {
@@ -633,6 +699,14 @@ event_subsystem_dir(const char *name, struct dentry *d_events)
 	system->name = name;
 	list_add(&system->list, &event_subsystems);
 
+	system->preds = NULL;
+
+	entry = debugfs_create_file("filter", 0444, system->entry, system,
+				    &ftrace_subsystem_filter_fops);
+	if (!entry)
+		pr_warning("Could not create debugfs "
+			   "'%s/filter' entry\n", name);
+
 	return system->entry;
 }
 
kernel/trace/trace_events_filter.c +23 -0
index 8e8c5fa..1ab20ce 100644
@@ -181,6 +181,27 @@ void filter_free_preds(struct ftrace_event_call *call)
 	}
 }
 
+void filter_free_subsystem_preds(struct event_subsystem *system)
+{
+	struct ftrace_event_call *call = __start_ftrace_events;
+	int i;
+
+	if (system->preds) {
+		for (i = 0; i < MAX_FILTER_PRED; i++)
+			filter_free_pred(system->preds[i]);
+		kfree(system->preds);
+		system->preds = NULL;
+	}
+
+	events_for_each(call) {
+		if (!call->name || !call->regfunc)
+			continue;
+
+		if (!strcmp(call->system, system->name))
+			filter_free_preds(call);
+	}
+}
+
 static int __filter_add_pred(struct ftrace_event_call *call,
 			     struct filter_pred *pred)
 {
@@ -250,6 +271,65 @@ int filter_add_pred(struct ftrace_event_call *call, struct filter_pred *pred)
 	return __filter_add_pred(call, pred);
 }
 
+static struct filter_pred *copy_pred(struct filter_pred *pred)
+{

kernel/trace/trace_events_filter.c int filter_add_pred(struct ftrace_event_call *call, struct filter_pred *pred) · line 285

Quoted 10 lines
kernel/trace/trace_events_filter.c +10 -0
+	struct filter_pred *new_pred = kmalloc(sizeof(*pred), GFP_KERNEL);
+	if (!new_pred)
+		return NULL;
+
+	memcpy(new_pred, pred, sizeof(*pred));
+	if (pred->str_val) {
+		new_pred->str_val = kstrdup(pred->str_val, GFP_KERNEL);
+		new_pred->field_name = kstrdup(pred->field_name, GFP_KERNEL);
+		if (!new_pred->str_val) {
+			kfree(new_pred);

Shouldn't there be a check for field_name too?

-- Steve

Quoted 7 lines
kernel/trace/trace_events_filter.c +7 -0
+			return NULL;
+		}
+	}
+
+	return new_pred;
+}
+
#6 lore raw

Hi,

Quoted 84 lines · Steven Rostedt
Quoted 31 lines · Tom Zanussi
kernel/trace/trace.h +15 -0
+struct event_subsystem {
+	struct list_head	list;
+	const char		*name;
+	struct dentry		*entry;
+	struct filter_pred	**preds;
+};
+
+#define events_for_each(event)						\
+	for (event = __start_ftrace_events;				\
+	     (unsigned long)event < (unsigned long)__stop_ftrace_events; \
+	     event++)
+
 #define MAX_FILTER_PRED 8
 
 struct filter_pred;
@@ -832,6 +844,9 @@ extern int filter_add_pred(struct ftrace_event_call *call,
 			   struct filter_pred *pred);
 extern void filter_free_preds(struct ftrace_event_call *call);
 extern int filter_match_preds(struct ftrace_event_call *call, void *rec);
+extern void filter_free_subsystem_preds(struct event_subsystem *system);
+extern int filter_add_subsystem_pred(struct event_subsystem *system,
+				     struct filter_pred *pred);
 
 void event_trace_printk(unsigned long ip, const char *fmt, ...);
 extern struct ftrace_event_call __start_ftrace_events[];
kernel/trace/trace_events.c +0 -0
index 97470c4..97d4daa 100644

kernel/trace/trace_events.c event_filter_write(struct file *filp, const char __user *ubuf, size_t cnt, · line 532

Quoted 9 lines
kernel/trace/trace_events.c +6 -0
@@ -524,6 +524,71 @@ event_filter_write(struct file *filp, const char __user *ubuf, size_t cnt,
 	return cnt;
 }
 
+static ssize_t
+subsystem_filter_read(struct file *filp, char __user *ubuf, size_t cnt,
+		      loff_t *ppos)
+{
+	struct event_subsystem *system = filp->private_data;
+	struct trace_seq *s;

Again, trace_seq is not used, might as well use your own buffer.

Quoted 40 lines
kernel/trace/trace_events.c +40 -0
+	int r;
+
+	if (*ppos)
+		return 0;
+
+	s = kmalloc(sizeof(*s), GFP_KERNEL);
+	if (!s)
+		return -ENOMEM;
+
+	trace_seq_init(s);
+
+	r = filter_print_preds(system->preds, s->buffer);
+	r = simple_read_from_buffer(ubuf, cnt, ppos, s->buffer, r);
+
+	kfree(s);
+
+	return r;
+}
+
+static ssize_t
+subsystem_filter_write(struct file *filp, const char __user *ubuf, size_t cnt,
+		       loff_t *ppos)
+{
+	struct event_subsystem *system = filp->private_data;
+	char buf[64], *pbuf = buf;
+	struct filter_pred *pred;
+	int err;
+
+	if (cnt >= sizeof(buf))
+		return -EINVAL;
+
+	if (copy_from_user(&buf, ubuf, cnt))
+		return -EFAULT;
+
+	pred = kzalloc(sizeof(*pred), GFP_KERNEL);
+	if (!pred)
+		return -ENOMEM;
+
+	err = filter_parse(&pbuf, pred);
+	if (err < 0) {
Quoted 8 lines

kernel/trace/trace_events.c event_filter_write(struct file *filp, const char __user *ubuf, size_t cnt, · line 578

Quoted 6 lines
kernel/trace/trace_events.c +6 -0
+		filter_free_pred(pred);
+		return err;
+	}
+
+	if (pred->clear) {
+		filter_free_subsystem_preds(system);

is "system" correct here?

Do you mean the naming i.e. would be better as "subsystem"? Otherwise, I think it's correct.

Quoted 101 lines
Quoted 101 lines
kernel/trace/trace_events.c +28 -6
+		return cnt;
+	}
+
+	if (filter_add_subsystem_pred(system, pred)) {
+		filter_free_pred(pred);
+		return -EINVAL;
+	}
+
+	*ppos += cnt;
+
+	return cnt;
+}
+
 static const struct seq_operations show_event_seq_ops = {
 	.start = t_start,
 	.next = t_next,
@@ -575,6 +640,12 @@ static const struct file_operations ftrace_event_filter_fops = {
 	.write = event_filter_write,
 };
 
+static const struct file_operations ftrace_subsystem_filter_fops = {
+	.open = tracing_open_generic,
+	.read = subsystem_filter_read,
+	.write = subsystem_filter_write,
+};
+
 static struct dentry *event_trace_events_dir(void)
 {
 	static struct dentry *d_tracer;
@@ -595,18 +666,13 @@ static struct dentry *event_trace_events_dir(void)
 	return d_events;
 }
 
-struct event_subsystem {
-	struct list_head	list;
-	const char		*name;
-	struct dentry		*entry;
-};
-
 static LIST_HEAD(event_subsystems);
 
 static struct dentry *
 event_subsystem_dir(const char *name, struct dentry *d_events)
 {
 	struct event_subsystem *system;
+	struct dentry *entry;
 
 	/* First see if we did not already create this dir */
 	list_for_each_entry(system, &event_subsystems, list) {
@@ -633,6 +699,14 @@ event_subsystem_dir(const char *name, struct dentry *d_events)
 	system->name = name;
 	list_add(&system->list, &event_subsystems);
 
+	system->preds = NULL;
+
+	entry = debugfs_create_file("filter", 0444, system->entry, system,
+				    &ftrace_subsystem_filter_fops);
+	if (!entry)
+		pr_warning("Could not create debugfs "
+			   "'%s/filter' entry\n", name);
+
 	return system->entry;
 }
 
kernel/trace/trace_events_filter.c +21 -0
index 8e8c5fa..1ab20ce 100644
@@ -181,6 +181,27 @@ void filter_free_preds(struct ftrace_event_call *call)
 	}
 }
 
+void filter_free_subsystem_preds(struct event_subsystem *system)
+{
+	struct ftrace_event_call *call = __start_ftrace_events;
+	int i;
+
+	if (system->preds) {
+		for (i = 0; i < MAX_FILTER_PRED; i++)
+			filter_free_pred(system->preds[i]);
+		kfree(system->preds);
+		system->preds = NULL;
+	}
+
+	events_for_each(call) {
+		if (!call->name || !call->regfunc)
+			continue;
+
+		if (!strcmp(call->system, system->name))
+			filter_free_preds(call);
+	}
+}
+
 static int __filter_add_pred(struct ftrace_event_call *call,
 			     struct filter_pred *pred)
 {
@@ -250,6 +271,65 @@ int filter_add_pred(struct ftrace_event_call *call, struct filter_pred *pred)
 	return __filter_add_pred(call, pred);
 }
 
Quoted 15 lines

kernel/trace/trace_events_filter.c int filter_add_pred(struct ftrace_event_call *call, struct filter_pred *pred) · line 285

Quoted 12 lines
kernel/trace/trace_events_filter.c +12 -0
+static struct filter_pred *copy_pred(struct filter_pred *pred)
+{
+	struct filter_pred *new_pred = kmalloc(sizeof(*pred), GFP_KERNEL);
+	if (!new_pred)
+		return NULL;
+
+	memcpy(new_pred, pred, sizeof(*pred));
+	if (pred->str_val) {
+		new_pred->str_val = kstrdup(pred->str_val, GFP_KERNEL);
+		new_pred->field_name = kstrdup(pred->field_name, GFP_KERNEL);
+		if (!new_pred->str_val) {
+			kfree(new_pred);

Shouldn't there be a check for field_name too?

Yes - I posted a patch to copy_pred() yesterday to fix that.

Tom

Quoted 9 lines

-- Steve

Quoted 7 lines
kernel/trace/trace_events_filter.c +7 -0
+			return NULL;
+		}
+	}
+
+	return new_pred;
+}
+
maintainer
#7 lore raw
Quoted 31 lines · Tom Zanussi
Quoted 22 lines
Quoted 22 lines
kernel/trace/trace_events.c +22 -0
+
+static ssize_t
+subsystem_filter_write(struct file *filp, const char __user *ubuf, size_t cnt,
+		       loff_t *ppos)
+{
+	struct event_subsystem *system = filp->private_data;
+	char buf[64], *pbuf = buf;
+	struct filter_pred *pred;
+	int err;
+
+	if (cnt >= sizeof(buf))
+		return -EINVAL;
+
+	if (copy_from_user(&buf, ubuf, cnt))
+		return -EFAULT;
+
+	pred = kzalloc(sizeof(*pred), GFP_KERNEL);
+	if (!pred)
+		return -ENOMEM;
+
+	err = filter_parse(&pbuf, pred);
+	if (err < 0) {
Quoted 8 lines

kernel/trace/trace_events.c event_filter_write(struct file *filp, const char __user *ubuf, size_t cnt, · line 578

Quoted 6 lines
kernel/trace/trace_events.c +6 -0
+		filter_free_pred(pred);
+		return err;
+	}
+
+	if (pred->clear) {
+		filter_free_subsystem_preds(system);

is "system" correct here?

Quoted 2 lines

Do you mean the naming i.e. would be better as "subsystem"? Otherwise, I think it's correct.

No, I'm just confused how the system could see the pred before it was added below.

-- Steve

Quoted 15 lines
Quoted 15 lines
Quoted 15 lines
kernel/trace/trace_events.c +13 -0
+		return cnt;
+	}
+
+	if (filter_add_subsystem_pred(system, pred)) {
+		filter_free_pred(pred);
+		return -EINVAL;
+	}
+
+	*ppos += cnt;
+
+	return cnt;
+}
+
#8 lore raw
Quoted 32 lines · Steven Rostedt
Quoted 31 lines · Tom Zanussi
Quoted 30 lines
Quoted 18 lines
kernel/trace/trace_events.c +18 -0
+
+static ssize_t
+subsystem_filter_write(struct file *filp, const char __user *ubuf, size_t cnt,
+		       loff_t *ppos)
+{
+	struct event_subsystem *system = filp->private_data;
+	char buf[64], *pbuf = buf;
+	struct filter_pred *pred;
+	int err;
+
+	if (cnt >= sizeof(buf))
+		return -EINVAL;
+
+	if (copy_from_user(&buf, ubuf, cnt))
+		return -EFAULT;
+
+	pred = kzalloc(sizeof(*pred), GFP_KERNEL);
+	if (!pred)

kernel/trace/trace_events.c event_filter_write(struct file *filp, const char __user *ubuf, size_t cnt, · line 578

Quoted 10 lines
kernel/trace/trace_events.c +10 -0
+		return -ENOMEM;
+
+	err = filter_parse(&pbuf, pred);
+	if (err < 0) {
+		filter_free_pred(pred);
+		return err;
+	}
+
+	if (pred->clear) {
+		filter_free_subsystem_preds(system);

is "system" correct here?

Quoted 5 lines
Quoted 2 lines

Do you mean the naming i.e. would be better as "subsystem"? Otherwise, I think it's correct.

No, I'm just confused how the system could see the pred before it was added below.

Again, in this case the pred is only used to flag clearing and never gets added.

Tom

Quoted 18 lines

-- Steve

Quoted 15 lines
Quoted 15 lines
Quoted 15 lines
kernel/trace/trace_events.c +13 -0
+		return cnt;
+	}
+
+	if (filter_add_subsystem_pred(system, pred)) {
+		filter_free_pred(pred);
+		return -EINVAL;
+	}
+
+	*ppos += cnt;
+
+	return cnt;
+}
+