perf report: Add machine parallelism

Add calculation of the current parallelism level (number of threads actively
running on CPUs). The parallelism level can be shown in reports on its own,
and to calculate latency overheads.

Signed-off-by: Dmitry Vyukov <dvyukov@google.com>
Reviewed-by: Andi Kleen <ak@linux.intel.com>
Link: https://lore.kernel.org/r/0f8c1b8eb12619029e31b3d5c0346f4616a5aeda.1739437531.git.dvyukov@google.com
Signed-off-by: Namhyung Kim <namhyung@kernel.org>

6 files changed, 20 insertions, 0 deletions

diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c
index f5fbd670d619..0d9bd090eda7 100644
--- a/tools/perf/builtin-report.c
+++ b/tools/perf/builtin-report.c
@@ -1568,6 +1568,7 @@ repeat:
 	report.tool.cgroup		 = perf_event__process_cgroup;
 	report.tool.exit		 = perf_event__process_exit;
 	report.tool.fork		 = perf_event__process_fork;
+	report.tool.context_switch	 = perf_event__process_switch;
 	report.tool.lost		 = perf_event__process_lost;
 	report.tool.read		 = process_read_event;
 	report.tool.attr		 = process_attr;
diff --git a/tools/perf/util/addr_location.c b/tools/perf/util/addr_location.c
index 51825ef8c0ab..007a2f5df9a6 100644
--- a/tools/perf/util/addr_location.c
+++ b/tools/perf/util/addr_location.c
@@ -17,6 +17,7 @@ void addr_location__init(struct addr_location *al)
 	al->cpumode = 0;
 	al->cpu = 0;
 	al->socket = 0;
+	al->parallelism = 1;
 }
 
 /*
diff --git a/tools/perf/util/addr_location.h b/tools/perf/util/addr_location.h
index d8ac0428dff2..36aaa45445f2 100644
--- a/tools/perf/util/addr_location.h
+++ b/tools/perf/util/addr_location.h
@@ -21,6 +21,8 @@ struct addr_location {
 	u8	      cpumode;
 	s32	      cpu;
 	s32	      socket;
+	/* Same as machine.parallelism but within [1, nr_cpus]. */
+	int	      parallelism;
 };
 
 void addr_location__init(struct addr_location *al);
diff --git a/tools/perf/util/event.c b/tools/perf/util/event.c
index aac96d5d1917..2f10e3115757 100644
--- a/tools/perf/util/event.c
+++ b/tools/perf/util/event.c
@@ -767,6 +767,9 @@ int machine__resolve(struct machine *machine, struct addr_location *al,
 			al->socket = env->cpu[al->cpu].socket_id;
 	}
 
+	/* Account for possible out-of-order switch events. */
+	al->parallelism = max(1, min(machine->parallelism, machine__nr_cpus_avail(machine)));
+
 	if (al->map) {
 		if (symbol_conf.dso_list &&
 		    (!dso || !(strlist__has_entry(symbol_conf.dso_list,
diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c
index 55d4977b9913..d96cbfd97ad8 100644
--- a/tools/perf/util/machine.c
+++ b/tools/perf/util/machine.c
@@ -94,6 +94,8 @@ int machine__init(struct machine *machine, const char *root_dir, pid_t pid)
 	machine->comm_exec = false;
 	machine->kernel_start = 0;
 	machine->vmlinux_map = NULL;
+	/* There is no initial context switch in, so we start at 1. */
+	machine->parallelism = 1;
 
 	machine->root_dir = strdup(root_dir);
 	if (machine->root_dir == NULL)
@@ -677,8 +679,11 @@ int machine__process_aux_output_hw_id_event(struct machine *machine __maybe_unus
 int machine__process_switch_event(struct machine *machine __maybe_unused,
 				  union perf_event *event)
 {
+	bool out = event->header.misc & PERF_RECORD_MISC_SWITCH_OUT;
+
 	if (dump_trace)
 		perf_event__fprintf_switch(event, stdout);
+	machine->parallelism += out ? -1 : 1;
 	return 0;
 }
 
@@ -1880,6 +1885,8 @@ int machine__process_exit_event(struct machine *machine, union perf_event *event
 	if (dump_trace)
 		perf_event__fprintf_task(event, stdout);
 
+	/* There is no context switch out before exit, so we decrement here. */
+	machine->parallelism--;
 	if (thread != NULL) {
 		if (symbol_conf.keep_exited_threads)
 			thread__set_exited(thread, /*exited=*/true);
diff --git a/tools/perf/util/machine.h b/tools/perf/util/machine.h
index ae3e5542d57d..b56abec84fed 100644
--- a/tools/perf/util/machine.h
+++ b/tools/perf/util/machine.h
@@ -50,6 +50,12 @@ struct machine {
 		u64	  text_start;
 		u64	  text_end;
 	} sched, lock, traceiter, trace;
+	/*
+	 * The current parallelism level (number of threads that run on CPUs).
+	 * This value can be less than 1, or larger than the total number
+	 * of CPUs, if events are poorly ordered.
+	 */
+	int		  parallelism;
 	pid_t		  *current_tid;
 	size_t		  current_tid_sz;
 	union { /* Tool specific area */