perf: Remove the 'match' callback for auxiliary events processing

It gives the following benefits:

- only one function pointer is passed along the way

- the 'match' function is called within output function
and could be inlined by the compiler

Suggested-by: Peter Zijlstra <[e-mail]>
Signed-off-by: Jiri Olsa <[e-mail]>
Cc: Corey Ashford <[e-mail]>
Cc: Frederic Weisbecker <[e-mail]>
Cc: Ingo Molnar <[e-mail]>
Cc: Namhyung Kim <[e-mail]>
Cc: Paul Mackerras <[e-mail]>
Cc: Arnaldo Carvalho de Melo <[e-mail]>
Signed-off-by: Peter Zijlstra <[e-mail]>
Link: http://lkml.kernel.org/r/[e-mail]
Signed-off-by: Ingo Molnar <[e-mail]>

kernel/events/core.c

···
 	perf_output_end(&handle);
 }
 
-typedef int  (perf_event_aux_match_cb)(struct perf_event *event, void *data);
 typedef void (perf_event_aux_output_cb)(struct perf_event *event, void *data);
 
 static void
 perf_event_aux_ctx(struct perf_event_context *ctx,
-		   perf_event_aux_match_cb match,
 		   perf_event_aux_output_cb output,
 		   void *data)
 {
···
 			continue;
 		if (!event_filter_match(event))
 			continue;
-		if (match(event, data))
-			output(event, data);
+		output(event, data);
 	}
 }
 
 static void
-perf_event_aux(perf_event_aux_match_cb match,
-	       perf_event_aux_output_cb output,
-	       void *data,
+perf_event_aux(perf_event_aux_output_cb output, void *data,
 	       struct perf_event_context *task_ctx)
 {
 	struct perf_cpu_context *cpuctx;
···
 		cpuctx = get_cpu_ptr(pmu->pmu_cpu_context);
 		if (cpuctx->unique_pmu != pmu)
 			goto next;
-		perf_event_aux_ctx(&cpuctx->ctx, match, output, data);
+		perf_event_aux_ctx(&cpuctx->ctx, output, data);
 		if (task_ctx)
 			goto next;
 		ctxn = pmu->task_ctx_nr;
···
 			goto next;
 		ctx = rcu_dereference(current->perf_event_ctxp[ctxn]);
 		if (ctx)
-			perf_event_aux_ctx(ctx, match, output, data);
+			perf_event_aux_ctx(ctx, output, data);
 next:
 		put_cpu_ptr(pmu->pmu_cpu_context);
 	}
 
 	if (task_ctx) {
 		preempt_disable();
-		perf_event_aux_ctx(task_ctx, match, output, data);
+		perf_event_aux_ctx(task_ctx, output, data);
 		preempt_enable();
 	}
 	rcu_read_unlock();
···
 	} event_id;
 };
 
+static int perf_event_task_match(struct perf_event *event)
+{
+	return event->attr.comm || event->attr.mmap ||
+	       event->attr.mmap_data || event->attr.task;
+}
+
 static void perf_event_task_output(struct perf_event *event,
 				   void *data)
 {
···
 	struct perf_sample_data	sample;
 	struct task_struct *task = task_event->task;
 	int ret, size = task_event->event_id.header.size;
+
+	if (!perf_event_task_match(event))
+		return;
 
 	perf_event_header__init_id(&task_event->event_id.header, &sample, event);
 
···
 	perf_output_end(&handle);
 out:
 	task_event->event_id.header.size = size;
-}
-
-static int perf_event_task_match(struct perf_event *event,
-				 void *data __maybe_unused)
-{
-	return event->attr.comm || event->attr.mmap ||
-	       event->attr.mmap_data || event->attr.task;
 }
 
 static void perf_event_task(struct task_struct *task,
···
 		},
 	};
 
-	perf_event_aux(perf_event_task_match,
-		       perf_event_task_output,
+	perf_event_aux(perf_event_task_output,
 		       &task_event,
 		       task_ctx);
 }
···
 	} event_id;
 };
 
+static int perf_event_comm_match(struct perf_event *event)
+{
+	return event->attr.comm;
+}
+
 static void perf_event_comm_output(struct perf_event *event,
 				   void *data)
 {
···
 	struct perf_sample_data sample;
 	int size = comm_event->event_id.header.size;
 	int ret;
+
+	if (!perf_event_comm_match(event))
+		return;
 
 	perf_event_header__init_id(&comm_event->event_id.header, &sample, event);
 	ret = perf_output_begin(&handle, event,
···
 	comm_event->event_id.header.size = size;
 }
 
-static int perf_event_comm_match(struct perf_event *event,
-				 void *data __maybe_unused)
-{
-	return event->attr.comm;
-}
-
 static void perf_event_comm_event(struct perf_comm_event *comm_event)
 {
 	char comm[TASK_COMM_LEN];
···
 
 	comm_event->event_id.header.size = sizeof(comm_event->event_id) + size;
 
-	perf_event_aux(perf_event_comm_match,
-		       perf_event_comm_output,
+	perf_event_aux(perf_event_comm_output,
 		       comm_event,
 		       NULL);
 }
···
 	} event_id;
 };
 
+static int perf_event_mmap_match(struct perf_event *event,
+				 void *data)
+{
+	struct perf_mmap_event *mmap_event = data;
+	struct vm_area_struct *vma = mmap_event->vma;
+	int executable = vma->vm_flags & VM_EXEC;
+
+	return (!executable && event->attr.mmap_data) ||
+	       (executable && event->attr.mmap);
+}
+
 static void perf_event_mmap_output(struct perf_event *event,
 				   void *data)
 {
···
 	struct perf_sample_data sample;
 	int size = mmap_event->event_id.header.size;
 	int ret;
+
+	if (!perf_event_mmap_match(event, data))
+		return;
 
 	perf_event_header__init_id(&mmap_event->event_id.header, &sample, event);
 	ret = perf_output_begin(&handle, event,
···
 	perf_output_end(&handle);
 out:
 	mmap_event->event_id.header.size = size;
-}
-
-static int perf_event_mmap_match(struct perf_event *event,
-				 void *data)
-{
-	struct perf_mmap_event *mmap_event = data;
-	struct vm_area_struct *vma = mmap_event->vma;
-	int executable = vma->vm_flags & VM_EXEC;
-
-	return (!executable && event->attr.mmap_data) ||
-	       (executable && event->attr.mmap);
 }
 
 static void perf_event_mmap_event(struct perf_mmap_event *mmap_event)
···
 
 	mmap_event->event_id.header.size = sizeof(mmap_event->event_id) + size;
 
-	perf_event_aux(perf_event_mmap_match,
-		       perf_event_mmap_output,
+	perf_event_aux(perf_event_mmap_output,
 		       mmap_event,
 		       NULL);