[PATCH v3 0/2] perf: Fix missing branch counter information in script and annotate

[PATCH v3 0/2] perf: Fix missing branch counter information in script and annotate

[Thomas Falcon]

This series fixes two issues related to branch counter display in perf.

The first patch fixes a missing '+' indicator in 'perf script' output.
The '+' suffix is used to signal that event occurrences may have been
lost due to branch counter overflow.

The second patch fixes a regression in 'perf annotate' TUI mode where
the Branch Count column was never displayed. The check for whether to
enable branch counter display was performed before events were processed
and the counter data were populated, so the condition was never satisfied.
Determine whether to display the branch counter column after processing
events and populating counter data.

v3: For patch 1/2, guarded against printing a spurious '+' if the branch
    counter width is zero, reported by sashiko
v2: Added a cover letter and expanded commit message for patch 1/2.
    Patch 2/2 was missing some conditional checks to enable branch counter
    display, which have been added back.

Dapeng Mi (1):
  perf script: Fix missing '+' indicator when branch counter reaches
    upper limit

Thomas Falcon (1):
  perf/annotate: Fix missing branch counter column in TUI mode

 tools/perf/builtin-annotate.c | 9 +++++----
 tools/perf/builtin-script.c   | 8 ++++++--
 2 files changed, 11 insertions(+), 6 deletions(-)

-- 
2.43.0

[PATCH v3 1/2] perf script: Fix missing '+' indicator when branch counter reaches upper limit

[Thomas Falcon]

From: Dapeng Mi <dapeng1.mi@linux.intel.com>

When displaying branch counter (br_cntr) information, a "+"
suffix represents that event occurrences may have been lost
due to branch counter saturation. However, this indicator was
missing in perf script. Add it back.

Before:

 # Branch counter abbr list:
 # cpu_core/event=0xc4,umask=0x20/ppp = A
 # cpu_core/instructions/ = B
 # cpu_core/MEM_INST_RETIRED.ALL_LOADS/ = C
 # cpu_core/MEM_LOAD_RETIRED.L2_MISS/ = D
 # '-' No event occurs
 # '+' Event occurrences may be lost due to branch counter saturated
...
        datasym+190:
        00005567f9951676        jz 0x5567f995162dr_cntr: BBBC   # PRED 1 cycles [1]
...
After:
...
        datasym+190:
        00005567f9951676        jz 0x5567f995162dr_cntr: BBB+C  # PRED 1 cycles [1]
...

Acked-by: Ian Rogers <irogers@google.com>
Reviewed-by: Dapeng Mi <dapeng1.mi@linux.intel.com>
Signed-off-by: Dapeng Mi <dapeng1.mi@linux.intel.com>
Signed-off-by: Thomas Falcon <thomas.falcon@intel.com>
---
 tools/perf/builtin-script.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c
index c0918006e0ab..7d78adb7995a 100644
--- a/tools/perf/builtin-script.c
+++ b/tools/perf/builtin-script.c
@@ -1288,8 +1288,12 @@ static int ip__fprintf_jump(uint64_t ip, struct branch_entry *en,
 			if (!verbose) {
 				for (j = 0; j < num; j++)
 					printed += fprintf(fp, "%s", pos->abbr_name);
-			} else
-				printed += fprintf(fp, "%s %d ", pos->name, num);
+				if (mask && (num == mask))
+					printed += fprintf(fp, "+");
+			} else {
+				printed += fprintf(fp, "%s %d%s", pos->name,
+						   num, mask && (num == mask) ? "+ " : " ");
+			}
 		}
 		if (numprinted == 0 && !verbose)
 			printed += fprintf(fp, "-");
-- 
2.43.0

[PATCH v3 2/2] perf/annotate: Fix missing branch counter column in TUI mode

[Thomas Falcon]

perf annotate checks that evlist->nr_br_cntr has been incremented
to determine whether to show branch counter information. However,
this data is not populated until after the check when events are
processed. Therefore, this counter will always be less than zero
and the Branch Count column is never shown. Do this check after
events have been processed and branch counter data is updated.

Acked-by: Ian Rogers <irogers@google.com>
Reviewed-by: Dapeng Mi <dapeng1.mi@linux.intel.com>
Signed-off-by: Thomas Falcon <thomas.falcon@intel.com>
---
 tools/perf/builtin-annotate.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/tools/perf/builtin-annotate.c b/tools/perf/builtin-annotate.c
index 719b36d4eed5..5f450c8093c0 100644
--- a/tools/perf/builtin-annotate.c
+++ b/tools/perf/builtin-annotate.c
@@ -559,6 +559,10 @@ static int __cmd_annotate(struct perf_annotate *ann)
 	if (ret)
 		goto out;
 
+	if ((use_browser == 1 || ann->use_stdio2) && ann->has_br_stack)
+		if (session->evlist->nr_br_cntr > 0)
+			annotate_opts.show_br_cntr = true;
+
 	if (dump_trace) {
 		perf_session__fprintf_nr_events(session, stdout);
 		evlist__fprintf_nr_events(session->evlist, stdout);
@@ -922,11 +926,8 @@ int cmd_annotate(int argc, const char **argv)
 	 * branch counters, if the corresponding branch info is available
 	 * in the perf data in the TUI mode.
 	 */
-	if ((use_browser == 1 || annotate.use_stdio2) && annotate.has_br_stack) {
+	if ((use_browser == 1 || annotate.use_stdio2) && annotate.has_br_stack)
 		sort__mode = SORT_MODE__BRANCH;
-		if (annotate.session->evlist->nr_br_cntr > 0)
-			annotate_opts.show_br_cntr = true;
-	}
 
 	if (setup_sorting(/*evlist=*/NULL, perf_session__env(annotate.session)) < 0)
 		usage_with_options(annotate_usage, options);
-- 
2.43.0