[PATCH v3 6/6] perf report: Display columns Predicted/Abort/Cycles in --branch-history

[Jin Yao <yao.jin@linux.intel.com>]

Use current sort mechanism but the real .se_cmp() just returns 0 so
that new columns "Predicted", "Abort" and "Cycles" are created in display
but actually these keys are not the sort keys.

For example:

Overhead  Source:Line   Symbol    Shared Object  Predicted  Abort  Cycles
........  ............  ........  .............  .........  .....  ......

  38.25%  div.c:45      [.] main  div            97.6%      0      3

Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
---
 tools/perf/Documentation/perf-report.txt |  12 ++++
 tools/perf/builtin-report.c              |   6 +-
 tools/perf/util/hist.c                   |   3 +
 tools/perf/util/hist.h                   |   3 +
 tools/perf/util/sort.c                   | 113 ++++++++++++++++++++++++++++++-
 tools/perf/util/sort.h                   |   3 +
 6 files changed, 138 insertions(+), 2 deletions(-)

diff --git a/tools/perf/Documentation/perf-report.txt b/tools/perf/Documentation/perf-report.txt
index 2d17462..703b564 100644
--- a/tools/perf/Documentation/perf-report.txt
+++ b/tools/perf/Documentation/perf-report.txt
@@ -335,6 +335,18 @@ OPTIONS
 --branch-history::
 	Add the addresses of sampled taken branches to the callstack.
 	This allows to examine the path the program took to each sample.
+
+	Also show with some branch flags that can be:
+	- Predicted: display the average percentage of predicated branches.
+		     (predicated number / total number)
+	- Abort: display the number of tsx aborted branches.
+	- Cycles: cycles in basic block.
+
+	- removed loops: the number of removed loops.
+	It also shows the number of removed loops in callchain list.
+	The branch history code has a loop detection that removes small loops.
+	It would be nice for knowing how many loops were removed.
+
 	The data collection must have used -b (or -j) and -g.
 
 --objdump=<path>::
diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c
index 3dfbfff..fbcd035 100644
--- a/tools/perf/builtin-report.c
+++ b/tools/perf/builtin-report.c
@@ -670,6 +670,10 @@ const char report_callchain_help[] = "Display call graph (stack chain/backtrace)
 				     CALLCHAIN_REPORT_HELP
 				     "\n\t\t\t\tDefault: " CALLCHAIN_DEFAULT_OPT;
 
+#define CALLCHAIN_BRANCH_SORT_ORDER	\
+	"srcline,symbol,dso,callchain_branch_predicted," \
+	"callchain_branch_abort,callchain_branch_cycles"
+
 int cmd_report(int argc, const char **argv, const char *prefix __maybe_unused)
 {
 	struct perf_session *session;
@@ -930,7 +934,7 @@ int cmd_report(int argc, const char **argv, const char *prefix __maybe_unused)
 		symbol_conf.use_callchain = true;
 		callchain_register_param(&callchain_param);
 		if (sort_order == NULL)
-			sort_order = "srcline,symbol,dso";
+			sort_order = CALLCHAIN_BRANCH_SORT_ORDER;
 	}
 
 	if (report.mem_mode) {
diff --git a/tools/perf/util/hist.c b/tools/perf/util/hist.c
index b02992e..89ab66b 100644
--- a/tools/perf/util/hist.c
+++ b/tools/perf/util/hist.c
@@ -176,6 +176,9 @@ void hists__calc_col_len(struct hists *hists, struct hist_entry *h)
 	hists__new_col_len(hists, HISTC_MEM_LVL, 21 + 3);
 	hists__new_col_len(hists, HISTC_LOCAL_WEIGHT, 12);
 	hists__new_col_len(hists, HISTC_GLOBAL_WEIGHT, 12);
+	hists__new_col_len(hists, HISTC_CALLCHAIN_BRANCH_PREDICTED, 9);
+	hists__new_col_len(hists, HISTC_CALLCHAIN_BRANCH_ABORT, 5);
+	hists__new_col_len(hists, HISTC_CALLCHAIN_BRANCH_CYCLES, 6);
 
 	if (h->srcline) {
 		len = MAX(strlen(h->srcline), strlen(sort_srcline.se_header));
diff --git a/tools/perf/util/hist.h b/tools/perf/util/hist.h
index 9928fed..505561d 100644
--- a/tools/perf/util/hist.h
+++ b/tools/perf/util/hist.h
@@ -56,6 +56,9 @@ enum hist_column {
 	HISTC_SRCLINE_FROM,
 	HISTC_SRCLINE_TO,
 	HISTC_TRACE,
+	HISTC_CALLCHAIN_BRANCH_PREDICTED,
+	HISTC_CALLCHAIN_BRANCH_ABORT,
+	HISTC_CALLCHAIN_BRANCH_CYCLES,
 	HISTC_NR_COLS, /* Last entry */
 };
 
diff --git a/tools/perf/util/sort.c b/tools/perf/util/sort.c
index 452e15a..c720751 100644
--- a/tools/perf/util/sort.c
+++ b/tools/perf/util/sort.c
@@ -435,6 +435,102 @@ struct sort_entry sort_srcline_to = {
 	.se_width_idx	= HISTC_SRCLINE_TO,
 };
 
+/* --sort callchain_branch_predicted */
+
+static int64_t
+sort__callchain_branch_predicted_cmp(struct hist_entry *left __maybe_unused,
+				     struct hist_entry *right __maybe_unused)
+{
+	return 0;
+}
+
+static int hist_entry__callchain_branch_predicted_snprintf(
+	struct hist_entry *he, char *bf, size_t size, unsigned int width)
+{
+	u64 branch_count, predicted_count;
+	double percent = 0.0;
+	char str[32];
+
+	callchain_branch_counts(he->callchain, &branch_count,
+				&predicted_count, NULL, NULL);
+
+	if (branch_count)
+		percent = predicted_count * 100.0 / branch_count;
+
+	snprintf(str, sizeof(str), "%.1f%%", percent);
+	return repsep_snprintf(bf, size, "%-*.*s", width, width, str);
+}
+
+struct sort_entry sort_callchain_branch_predicted = {
+	.se_header	= "Predicted",
+	.se_cmp		= sort__callchain_branch_predicted_cmp,
+	.se_snprintf	= hist_entry__callchain_branch_predicted_snprintf,
+	.se_width_idx	= HISTC_CALLCHAIN_BRANCH_PREDICTED,
+};
+
+/* --sort callchain_branch_abort */
+
+static int64_t
+sort__callchain_branch_abort_cmp(struct hist_entry *left __maybe_unused,
+				 struct hist_entry *right __maybe_unused)
+{
+	return 0;
+}
+
+static int hist_entry__callchain_branch_abort_snprintf(struct hist_entry *he,
+						       char *bf, size_t size,
+						       unsigned int width)
+{
+	u64 branch_count, abort_count;
+	char str[32];
+
+	callchain_branch_counts(he->callchain, &branch_count,
+				NULL, &abort_count, NULL);
+
+	snprintf(str, sizeof(str), "%" PRId64, abort_count);
+	return repsep_snprintf(bf, size, "%-*.*s", width, width, str);
+}
+
+struct sort_entry sort_callchain_branch_abort = {
+	.se_header	= "Abort",
+	.se_cmp		= sort__callchain_branch_abort_cmp,
+	.se_snprintf	= hist_entry__callchain_branch_abort_snprintf,
+	.se_width_idx	= HISTC_CALLCHAIN_BRANCH_ABORT,
+};
+
+/* --sort callchain_branch_cycles */
+
+static int64_t
+sort__callchain_branch_cycles_cmp(struct hist_entry *left __maybe_unused,
+				  struct hist_entry *right __maybe_unused)
+{
+	return 0;
+}
+
+static int hist_entry__callchain_branch_cycles_snprintf(struct hist_entry *he,
+							char *bf, size_t size,
+							unsigned int width)
+{
+	u64 branch_count, cycles_count, cycles = 0;
+	char str[32];
+
+	callchain_branch_counts(he->callchain, &branch_count,
+				NULL, NULL, &cycles_count);
+
+	if (branch_count)
+		cycles = cycles_count / branch_count;
+
+	snprintf(str, sizeof(str), "%" PRId64 "", cycles);
+	return repsep_snprintf(bf, size, "%-*.*s", width, width, str);
+}
+
+struct sort_entry sort_callchain_branch_cycles = {
+	.se_header	= "Cycles",
+	.se_cmp		= sort__callchain_branch_cycles_cmp,
+	.se_snprintf	= hist_entry__callchain_branch_cycles_snprintf,
+	.se_width_idx	= HISTC_CALLCHAIN_BRANCH_CYCLES,
+};
+
 /* --sort srcfile */
 
 static char no_srcfile[1];
@@ -1435,6 +1531,15 @@ static struct sort_dimension bstack_sort_dimensions[] = {
 	DIM(SORT_CYCLES, "cycles", sort_cycles),
 	DIM(SORT_SRCLINE_FROM, "srcline_from", sort_srcline_from),
 	DIM(SORT_SRCLINE_TO, "srcline_to", sort_srcline_to),
+	DIM(SORT_CALLCHAIN_BRANCH_PREDICTED,
+		"callchain_branch_predicted",
+		sort_callchain_branch_predicted),
+	DIM(SORT_CALLCHAIN_BRANCH_ABORT,
+		"callchain_branch_abort",
+		sort_callchain_branch_abort),
+	DIM(SORT_CALLCHAIN_BRANCH_CYCLES,
+		"callchain_branch_cycles",
+		sort_callchain_branch_cycles),
 };
 
 #undef DIM
@@ -2369,7 +2474,13 @@ int sort_dimension__add(struct perf_hpp_list *list, const char *tok,
 		if (strncasecmp(tok, sd->name, strlen(tok)))
 			continue;
 
-		if (sort__mode != SORT_MODE__BRANCH)
+		if ((sort__mode != SORT_MODE__BRANCH) &&
+			strncasecmp(tok, "callchain_branch_predicted",
+				    strlen(tok)) &&
+			strncasecmp(tok, "callchain_branch_abort",
+				    strlen(tok)) &&
+			strncasecmp(tok, "callchain_branch_cycles",
+				    strlen(tok)))
 			return -EINVAL;
 
 		if (sd->entry == &sort_sym_from || sd->entry == &sort_sym_to)
diff --git a/tools/perf/util/sort.h b/tools/perf/util/sort.h
index 099c975..d55b4b6 100644
--- a/tools/perf/util/sort.h
+++ b/tools/perf/util/sort.h
@@ -224,6 +224,9 @@ enum sort_type {
 	SORT_CYCLES,
 	SORT_SRCLINE_FROM,
 	SORT_SRCLINE_TO,
+	SORT_CALLCHAIN_BRANCH_PREDICTED,
+	SORT_CALLCHAIN_BRANCH_ABORT,
+	SORT_CALLCHAIN_BRANCH_CYCLES,
 
 	/* memory mode specific sort keys */
 	__SORT_MEMORY_MODE,
-- 
2.7.4