[PATCH v16 12/23] tracing: Add hist trigger support for stacktraces as keys

[Tom Zanussi <tom.zanussi@linux.intel.com>]

It's often useful to be able to use a stacktrace as a hash key, for
keeping a count of the number of times a particular call path resulted
in a trace event, for instance.  Add a special key named 'stacktrace'
which can be used as key in a 'keys=' param for this purpose:

    # echo hist:keys=stacktrace ... \
               [ if filter] > event/trigger

Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com>
Tested-by: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Reviewed-by: Namhyung Kim <namhyung@kernel.org>
---
 kernel/trace/trace.c             |  16 ++---
 kernel/trace/trace_events_hist.c | 135 +++++++++++++++++++++++++++++----------
 2 files changed, 109 insertions(+), 42 deletions(-)

diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index d1890cc..1a652d8 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -3834,14 +3834,14 @@ static const char readme_msg[] =
 	"\t    table using the key(s) and value(s) named, and the value of a\n"
 	"\t    sum called 'hitcount' is incremented.  Keys and values\n"
 	"\t    correspond to fields in the event's format description.  Keys\n"
-	"\t    can be any field.  Compound keys consisting of up to two\n"
-	"\t    fields can be specified by the 'keys' keyword.  Values must\n"
-	"\t    correspond to numeric fields.  Sort keys consisting of up to\n"
-	"\t    two fields can be specified using the 'sort' keyword.  The\n"
-	"\t    sort direction can be modified by appending '.descending' or\n"
-	"\t    '.ascending' to a sort field.  The 'size' parameter can be\n"
-	"\t    used to specify more or fewer than the default 2048 entries\n"
-	"\t    for the hashtable size.\n\n"
+	"\t    can be any field, or the special string 'stacktrace'.\n"
+	"\t    Compound keys consisting of up to two fields can be specified\n"
+	"\t    by the 'keys' keyword.  Values must correspond to numeric\n"
+	"\t    fields.  Sort keys consisting of up to two fields can be\n"
+	"\t    specified using the 'sort' keyword.  The sort direction can\n"
+	"\t    be modified by appending '.descending' or '.ascending' to a\n"
+	"\t    sort field.  The 'size' parameter can be used to specify more\n"
+	"\t    or fewer than the default 2048 entries for the hashtable size.\n\n"
 	"\t    Reading the 'hist' file for the event will dump the hash\n"
 	"\t    table in its entirety to stdout.  The default format used to\n"
 	"\t    display a given field can be modified by appending any of the\n"
diff --git a/kernel/trace/trace_events_hist.c b/kernel/trace/trace_events_hist.c
index f9e7ecb..21cae42 100644
--- a/kernel/trace/trace_events_hist.c
+++ b/kernel/trace/trace_events_hist.c
@@ -35,6 +35,11 @@ struct hist_field {
 	unsigned int			offset;
 };
 
+static u64 hist_field_none(struct hist_field *field, void *event)
+{
+	return 0;
+}
+
 static u64 hist_field_counter(struct hist_field *field, void *event)
 {
 	return 1;
@@ -73,8 +78,12 @@ DEFINE_HIST_FIELD_FN(u8);
 #define for_each_hist_key_field(i, hist_data)	\
 	for ((i) = (hist_data)->n_vals; (i) < (hist_data)->n_fields; (i)++)
 
+#define HIST_STACKTRACE_DEPTH	16
+#define HIST_STACKTRACE_SIZE	(HIST_STACKTRACE_DEPTH * sizeof(unsigned long))
+#define HIST_STACKTRACE_SKIP	5
+
 #define HITCOUNT_IDX		0
-#define HIST_KEY_SIZE_MAX	(MAX_FILTER_STR_VAL + sizeof(u64))
+#define HIST_KEY_SIZE_MAX	(MAX_FILTER_STR_VAL + HIST_STACKTRACE_SIZE)
 
 enum hist_field_flags {
 	HIST_FIELD_FL_HITCOUNT		= 1,
@@ -85,6 +94,7 @@ enum hist_field_flags {
 	HIST_FIELD_FL_SYM_OFFSET	= 32,
 	HIST_FIELD_FL_EXECNAME		= 64,
 	HIST_FIELD_FL_SYSCALL		= 128,
+	HIST_FIELD_FL_STACKTRACE	= 256,
 };
 
 struct hist_trigger_attrs {
@@ -323,6 +333,11 @@ static struct hist_field *create_hist_field(struct ftrace_event_field *field,
 		goto out;
 	}
 
+	if (flags & HIST_FIELD_FL_STACKTRACE) {
+		hist_field->fn = hist_field_none;
+		goto out;
+	}
+
 	if (is_string_field(field)) {
 		flags |= HIST_FIELD_FL_STRING;
 		hist_field->fn = hist_field_string;
@@ -456,7 +471,6 @@ static int create_key_field(struct hist_trigger_data *hist_data,
 	struct ftrace_event_field *field = NULL;
 	unsigned long flags = 0;
 	unsigned int key_size;
-	char *field_name;
 	int ret = 0;
 
 	if (WARN_ON(key_idx >= TRACING_MAP_FIELDS_MAX))
@@ -464,33 +478,39 @@ static int create_key_field(struct hist_trigger_data *hist_data,
 
 	flags |= HIST_FIELD_FL_KEY;
 
-	field_name = strsep(&field_str, ".");
-	if (field_str) {
-		if (strcmp(field_str, "hex") == 0)
-			flags |= HIST_FIELD_FL_HEX;
-		else if (strcmp(field_str, "sym") == 0)
-			flags |= HIST_FIELD_FL_SYM;
-		else if (strcmp(field_str, "sym-offset") == 0)
-			flags |= HIST_FIELD_FL_SYM_OFFSET;
-		else if ((strcmp(field_str, "execname") == 0) &&
-			 (strcmp(field_name, "common_pid") == 0))
-			flags |= HIST_FIELD_FL_EXECNAME;
-		else if (strcmp(field_str, "syscall") == 0)
-			flags |= HIST_FIELD_FL_SYSCALL;
-		else {
+	if (strcmp(field_str, "stacktrace") == 0) {
+		flags |= HIST_FIELD_FL_STACKTRACE;
+		key_size = sizeof(unsigned long) * HIST_STACKTRACE_DEPTH;
+	} else {
+		char *field_name = strsep(&field_str, ".");
+
+		if (field_str) {
+			if (strcmp(field_str, "hex") == 0)
+				flags |= HIST_FIELD_FL_HEX;
+			else if (strcmp(field_str, "sym") == 0)
+				flags |= HIST_FIELD_FL_SYM;
+			else if (strcmp(field_str, "sym-offset") == 0)
+				flags |= HIST_FIELD_FL_SYM_OFFSET;
+			else if ((strcmp(field_str, "execname") == 0) &&
+				 (strcmp(field_name, "common_pid") == 0))
+				flags |= HIST_FIELD_FL_EXECNAME;
+			else if (strcmp(field_str, "syscall") == 0)
+				flags |= HIST_FIELD_FL_SYSCALL;
+			else {
+				ret = -EINVAL;
+				goto out;
+			}
+		}
+
+		field = trace_find_event_field(file->event_call, field_name);
+		if (!field) {
 			ret = -EINVAL;
 			goto out;
 		}
-	}
 
-	field = trace_find_event_field(file->event_call, field_name);
-	if (!field) {
-		ret = -EINVAL;
-		goto out;
+		key_size = field->size;
 	}
 
-	key_size = field->size;
-
 	hist_data->fields[key_idx] = create_hist_field(field, flags);
 	if (!hist_data->fields[key_idx]) {
 		ret = -ENOMEM;
@@ -679,7 +699,9 @@ static int create_tracing_map_fields(struct hist_trigger_data *hist_data)
 
 			field = hist_field->field;
 
-			if (is_string_field(field))
+			if (hist_field->flags & HIST_FIELD_FL_STACKTRACE)
+				cmp_fn = tracing_map_cmp_none;
+			else if (is_string_field(field))
 				cmp_fn = tracing_map_cmp_string;
 			else
 				cmp_fn = tracing_map_cmp_num(field->size,
@@ -786,7 +808,9 @@ static void hist_trigger_elt_update(struct hist_trigger_data *hist_data,
 static void event_hist_trigger(struct event_trigger_data *data, void *rec)
 {
 	struct hist_trigger_data *hist_data = data->private_data;
+	unsigned long entries[HIST_STACKTRACE_DEPTH];
 	char compound_key[HIST_KEY_SIZE_MAX];
+	struct stack_trace stacktrace;
 	struct hist_field *key_field;
 	struct tracing_map_elt *elt;
 	u64 field_contents;
@@ -799,15 +823,27 @@ static void event_hist_trigger(struct event_trigger_data *data, void *rec)
 	for_each_hist_key_field(i, hist_data) {
 		key_field = hist_data->fields[i];
 
-		field_contents = key_field->fn(key_field, rec);
-		if (key_field->flags & HIST_FIELD_FL_STRING)
-			key = (void *)(unsigned long)field_contents;
-		else
-			key = (void *)&field_contents;
+		if (key_field->flags & HIST_FIELD_FL_STACKTRACE) {
+			stacktrace.max_entries = HIST_STACKTRACE_DEPTH;
+			stacktrace.entries = entries;
+			stacktrace.nr_entries = 0;
+			stacktrace.skip = HIST_STACKTRACE_SKIP;
 
-		if (hist_data->n_keys > 1) {
-			memcpy(compound_key + key_field->offset, key,
-			       key_field->size);
+			memset(stacktrace.entries, 0, HIST_STACKTRACE_SIZE);
+			save_stack_trace(&stacktrace);
+
+			key = entries;
+		} else {
+			field_contents = key_field->fn(key_field, rec);
+			if (key_field->flags & HIST_FIELD_FL_STRING)
+				key = (void *)(unsigned long)field_contents;
+			else
+				key = (void *)&field_contents;
+
+			if (hist_data->n_keys > 1) {
+				memcpy(compound_key + key_field->offset, key,
+				       key_field->size);
+			}
 		}
 	}
 
@@ -819,6 +855,24 @@ static void event_hist_trigger(struct event_trigger_data *data, void *rec)
 		hist_trigger_elt_update(hist_data, elt, rec);
 }
 
+static void hist_trigger_stacktrace_print(struct seq_file *m,
+					  unsigned long *stacktrace_entries,
+					  unsigned int max_entries)
+{
+	char str[KSYM_SYMBOL_LEN];
+	unsigned int spaces = 8;
+	unsigned int i;
+
+	for (i = 0; i < max_entries; i++) {
+		if (stacktrace_entries[i] == ULONG_MAX)
+			return;
+
+		seq_printf(m, "%*c", 1 + spaces, ' ');
+		sprint_symbol(str, stacktrace_entries[i]);
+		seq_printf(m, "%s\n", str);
+	}
+}
+
 static void
 hist_trigger_entry_print(struct seq_file *m,
 			 struct hist_trigger_data *hist_data, void *key,
@@ -826,6 +880,7 @@ hist_trigger_entry_print(struct seq_file *m,
 {
 	struct hist_field *key_field;
 	char str[KSYM_SYMBOL_LEN];
+	bool multiline = false;
 	unsigned int i;
 	u64 uval;
 
@@ -867,6 +922,12 @@ hist_trigger_entry_print(struct seq_file *m,
 
 			seq_printf(m, "%s: %-30s[%3llu]",
 				   key_field->field->name, syscall_name, uval);
+		} else if (key_field->flags & HIST_FIELD_FL_STACKTRACE) {
+			seq_puts(m, "stacktrace:\n");
+			hist_trigger_stacktrace_print(m,
+						      key + key_field->offset,
+						      HIST_STACKTRACE_DEPTH);
+			multiline = true;
 		} else if (key_field->flags & HIST_FIELD_FL_STRING) {
 			seq_printf(m, "%s: %-50s", key_field->field->name,
 				   (char *)(key + key_field->offset));
@@ -877,7 +938,10 @@ hist_trigger_entry_print(struct seq_file *m,
 		}
 	}
 
-	seq_puts(m, " }");
+	if (!multiline)
+		seq_puts(m, " ");
+
+	seq_puts(m, "}");
 
 	seq_printf(m, " hitcount: %10llu",
 		   tracing_map_read_sum(elt, HITCOUNT_IDX));
@@ -1021,7 +1085,10 @@ static int event_hist_trigger_print(struct seq_file *m,
 		if (i > hist_data->n_vals)
 			seq_puts(m, ",");
 
-		hist_field_print(m, key_field);
+		if (key_field->flags & HIST_FIELD_FL_STACKTRACE)
+			seq_puts(m, "stacktrace");
+		else
+			hist_field_print(m, key_field);
 	}
 
 	seq_puts(m, ":vals=");
-- 
1.9.3