[PATCH 08/25] perf tools: Add AUX area tracing Snapshot Mode

[Arnaldo Carvalho de Melo <acme@kernel.org>]

From: Adrian Hunter <adrian.hunter@intel.com>

Add support for making snapshots of AUX area tracing data.

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1430404667-10593-9-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/perf.h          |  3 ++
 tools/perf/util/auxtrace.c | 85 ++++++++++++++++++++++++++++++++++++++++------
 tools/perf/util/auxtrace.h | 41 ++++++++++++++++++++++
 3 files changed, 119 insertions(+), 10 deletions(-)

diff --git a/tools/perf/perf.h b/tools/perf/perf.h
index 5042093..aa79fb8 100644
--- a/tools/perf/perf.h
+++ b/tools/perf/perf.h
@@ -55,6 +55,7 @@ struct record_opts {
 	bool	     sample_intr_regs;
 	bool	     running_time;
 	bool	     full_auxtrace;
+	bool	     auxtrace_snapshot_mode;
 	unsigned int freq;
 	unsigned int mmap_pages;
 	unsigned int auxtrace_mmap_pages;
@@ -62,6 +63,8 @@ struct record_opts {
 	u64          branch_stack;
 	u64	     default_interval;
 	u64	     user_interval;
+	size_t	     auxtrace_snapshot_size;
+	const char   *auxtrace_snapshot_opts;
 	bool	     sample_transaction;
 	unsigned     initial_delay;
 	bool         use_clockid;
diff --git a/tools/perf/util/auxtrace.c b/tools/perf/util/auxtrace.c
index 1293710..df66966 100644
--- a/tools/perf/util/auxtrace.c
+++ b/tools/perf/util/auxtrace.c
@@ -504,6 +504,29 @@ void auxtrace_record__free(struct auxtrace_record *itr)
 		itr->free(itr);
 }
 
+int auxtrace_record__snapshot_start(struct auxtrace_record *itr)
+{
+	if (itr && itr->snapshot_start)
+		return itr->snapshot_start(itr);
+	return 0;
+}
+
+int auxtrace_record__snapshot_finish(struct auxtrace_record *itr)
+{
+	if (itr && itr->snapshot_finish)
+		return itr->snapshot_finish(itr);
+	return 0;
+}
+
+int auxtrace_record__find_snapshot(struct auxtrace_record *itr, int idx,
+				   struct auxtrace_mmap *mm,
+				   unsigned char *data, u64 *head, u64 *old)
+{
+	if (itr && itr->find_snapshot)
+		return itr->find_snapshot(itr, idx, mm, data, head, old);
+	return 0;
+}
+
 int auxtrace_record__options(struct auxtrace_record *itr,
 			     struct perf_evlist *evlist,
 			     struct record_opts *opts)
@@ -520,6 +543,19 @@ u64 auxtrace_record__reference(struct auxtrace_record *itr)
 	return 0;
 }
 
+int auxtrace_parse_snapshot_options(struct auxtrace_record *itr,
+				    struct record_opts *opts, const char *str)
+{
+	if (!str)
+		return 0;
+
+	if (itr)
+		return itr->parse_snapshot_options(itr, opts, str);
+
+	pr_err("No AUX area tracing to snapshot\n");
+	return -EINVAL;
+}
+
 struct auxtrace_record *__weak
 auxtrace_record__init(struct perf_evlist *evlist __maybe_unused, int *err)
 {
@@ -1077,16 +1113,26 @@ int perf_event__process_auxtrace_error(struct perf_tool *tool __maybe_unused,
 	return 0;
 }
 
-int auxtrace_mmap__read(struct auxtrace_mmap *mm, struct auxtrace_record *itr,
-			struct perf_tool *tool, process_auxtrace_t fn)
+static int __auxtrace_mmap__read(struct auxtrace_mmap *mm,
+				 struct auxtrace_record *itr,
+				 struct perf_tool *tool, process_auxtrace_t fn,
+				 bool snapshot, size_t snapshot_size)
 {
-	u64 head = auxtrace_mmap__read_head(mm);
-	u64 old = mm->prev, offset, ref;
+	u64 head, old = mm->prev, offset, ref;
 	unsigned char *data = mm->base;
 	size_t size, head_off, old_off, len1, len2, padding;
 	union perf_event ev;
 	void *data1, *data2;
 
+	if (snapshot) {
+		head = auxtrace_mmap__read_snapshot_head(mm);
+		if (auxtrace_record__find_snapshot(itr, mm->idx, mm, data,
+						   &head, &old))
+			return -1;
+	} else {
+		head = auxtrace_mmap__read_head(mm);
+	}
+
 	if (old == head)
 		return 0;
 
@@ -1106,6 +1152,9 @@ int auxtrace_mmap__read(struct auxtrace_mmap *mm, struct auxtrace_record *itr,
 	else
 		size = mm->len - (old_off - head_off);
 
+	if (snapshot && size > snapshot_size)
+		size = snapshot_size;
+
 	ref = auxtrace_record__reference(itr);
 
 	if (head > old || size <= head || mm->mask) {
@@ -1153,18 +1202,34 @@ int auxtrace_mmap__read(struct auxtrace_mmap *mm, struct auxtrace_record *itr,
 
 	mm->prev = head;
 
-	auxtrace_mmap__write_tail(mm, head);
-	if (itr->read_finish) {
-		int err;
+	if (!snapshot) {
+		auxtrace_mmap__write_tail(mm, head);
+		if (itr->read_finish) {
+			int err;
 
-		err = itr->read_finish(itr, mm->idx);
-		if (err < 0)
-			return err;
+			err = itr->read_finish(itr, mm->idx);
+			if (err < 0)
+				return err;
+		}
 	}
 
 	return 1;
 }
 
+int auxtrace_mmap__read(struct auxtrace_mmap *mm, struct auxtrace_record *itr,
+			struct perf_tool *tool, process_auxtrace_t fn)
+{
+	return __auxtrace_mmap__read(mm, itr, tool, fn, false, 0);
+}
+
+int auxtrace_mmap__read_snapshot(struct auxtrace_mmap *mm,
+				 struct auxtrace_record *itr,
+				 struct perf_tool *tool, process_auxtrace_t fn,
+				 size_t snapshot_size)
+{
+	return __auxtrace_mmap__read(mm, itr, tool, fn, true, snapshot_size);
+}
+
 /**
  * struct auxtrace_cache - hash table to implement a cache
  * @hashtable: the hashtable
diff --git a/tools/perf/util/auxtrace.h b/tools/perf/util/auxtrace.h
index 8c6cbb1..c2c677e 100644
--- a/tools/perf/util/auxtrace.h
+++ b/tools/perf/util/auxtrace.h
@@ -276,6 +276,10 @@ struct auxtrace_mmap_params {
  * @info_priv_size: return the size of the private data in auxtrace_info_event
  * @info_fill: fill-in the private data in auxtrace_info_event
  * @free: free this auxtrace record structure
+ * @snapshot_start: starting a snapshot
+ * @snapshot_finish: finishing a snapshot
+ * @find_snapshot: find data to snapshot within auxtrace mmap
+ * @parse_snapshot_options: parse snapshot options
  * @reference: provide a 64-bit reference number for auxtrace_event
  * @read_finish: called after reading from an auxtrace mmap
  */
@@ -289,12 +293,36 @@ struct auxtrace_record {
 			 struct auxtrace_info_event *auxtrace_info,
 			 size_t priv_size);
 	void (*free)(struct auxtrace_record *itr);
+	int (*snapshot_start)(struct auxtrace_record *itr);
+	int (*snapshot_finish)(struct auxtrace_record *itr);
+	int (*find_snapshot)(struct auxtrace_record *itr, int idx,
+			     struct auxtrace_mmap *mm, unsigned char *data,
+			     u64 *head, u64 *old);
+	int (*parse_snapshot_options)(struct auxtrace_record *itr,
+				      struct record_opts *opts,
+				      const char *str);
 	u64 (*reference)(struct auxtrace_record *itr);
 	int (*read_finish)(struct auxtrace_record *itr, int idx);
 };
 
 #ifdef HAVE_AUXTRACE_SUPPORT
 
+/*
+ * In snapshot mode the mmapped page is read-only which makes using
+ * __sync_val_compare_and_swap() problematic.  However, snapshot mode expects
+ * the buffer is not updated while the snapshot is made (e.g. Intel PT disables
+ * the event) so there is not a race anyway.
+ */
+static inline u64 auxtrace_mmap__read_snapshot_head(struct auxtrace_mmap *mm)
+{
+	struct perf_event_mmap_page *pc = mm->userpg;
+	u64 head = ACCESS_ONCE(pc->aux_head);
+
+	/* Ensure all reads are done after we read the head */
+	rmb();
+	return head;
+}
+
 static inline u64 auxtrace_mmap__read_head(struct auxtrace_mmap *mm)
 {
 	struct perf_event_mmap_page *pc = mm->userpg;
@@ -346,6 +374,11 @@ typedef int (*process_auxtrace_t)(struct perf_tool *tool,
 int auxtrace_mmap__read(struct auxtrace_mmap *mm, struct auxtrace_record *itr,
 			struct perf_tool *tool, process_auxtrace_t fn);
 
+int auxtrace_mmap__read_snapshot(struct auxtrace_mmap *mm,
+				 struct auxtrace_record *itr,
+				 struct perf_tool *tool, process_auxtrace_t fn,
+				 size_t snapshot_size);
+
 int auxtrace_queues__init(struct auxtrace_queues *queues);
 int auxtrace_queues__add_event(struct auxtrace_queues *queues,
 			       struct perf_session *session,
@@ -383,6 +416,9 @@ void *auxtrace_cache__lookup(struct auxtrace_cache *c, u32 key);
 struct auxtrace_record *auxtrace_record__init(struct perf_evlist *evlist,
 					      int *err);
 
+int auxtrace_parse_snapshot_options(struct auxtrace_record *itr,
+				    struct record_opts *opts,
+				    const char *str);
 int auxtrace_record__options(struct auxtrace_record *itr,
 			     struct perf_evlist *evlist,
 			     struct record_opts *opts);
@@ -392,6 +428,11 @@ int auxtrace_record__info_fill(struct auxtrace_record *itr,
 			       struct auxtrace_info_event *auxtrace_info,
 			       size_t priv_size);
 void auxtrace_record__free(struct auxtrace_record *itr);
+int auxtrace_record__snapshot_start(struct auxtrace_record *itr);
+int auxtrace_record__snapshot_finish(struct auxtrace_record *itr);
+int auxtrace_record__find_snapshot(struct auxtrace_record *itr, int idx,
+				   struct auxtrace_mmap *mm,
+				   unsigned char *data, u64 *head, u64 *old);
 u64 auxtrace_record__reference(struct auxtrace_record *itr);
 
 int auxtrace_index__auxtrace_event(struct list_head *head, union perf_event *event,
-- 
2.1.0