perf/core: Optimize perf_init_event() for TYPE_SOFTWARE

Andi reported that he was hitting the linear search in
perf_init_event() a lot. Now that all !TYPE_SOFTWARE events should hit
the IDR, make sure the TYPE_SOFTWARE events are at the head of the
list such that we'll quickly find the right PMU (provided a valid
event was given).
Signed-off-by: NLiang, Kan <kan.liang@linux.intel.com>
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Vince Weaver <vincent.weaver@maine.edu>
Signed-off-by: NIngo Molnar <mingo@kernel.org>

kernel/events/core.c

@@ -10180,7 +10180,16 @@ int perf_pmu_register(struct pmu *pmu, const char *name, int type)
 	if (!pmu->event_idx)
 		pmu->event_idx = perf_event_idx_default;
 
-	list_add_rcu(&pmu->entry, &pmus);
+	/*
+	 * Ensure the TYPE_SOFTWARE PMUs are at the head of the list,
+	 * since these cannot be in the IDR. This way the linear search
+	 * is fast, provided a valid software event is provided.
+	 */
+	if (type == PERF_TYPE_SOFTWARE || !name)
+		list_add_rcu(&pmu->entry, &pmus);
+	else
+		list_add_tail_rcu(&pmu->entry, &pmus);
+
 	atomic_set(&pmu->exclusive_cnt, 0);
 	ret = 0;
 unlock: