perf/core: Implement the 'perf_kprobe' PMU

A new PMU type, perf_kprobe is added. Based on attr from perf_event_open(),
perf_kprobe creates a kprobe (or kretprobe) for the perf_event. This
kprobe is private to this perf_event, and thus not added to global
lists, and not available in tracefs.

Two functions, create_local_trace_kprobe() and
destroy_local_trace_kprobe()  are added to created and destroy these
local trace_kprobe.

Signed-off-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Yonghong Song <yhs@fb.com>
Reviewed-by: Josef Bacik <jbacik@fb.com>
Cc: <daniel@iogearbox.net>
Cc: <davem@davemloft.net>
Cc: <kernel-team@fb.com>
Cc: <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/20171206224518.3598254-6-songliubraving@fb.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>

1 files changed, 107 insertions, 35 deletions

diff --git a/kernel/events/core.c b/kernel/events/core.c
index d99fe3fdec8a..333735531968 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -7992,9 +7992,77 @@ static struct pmu perf_tracepoint = {
 	.read		= perf_swevent_read,
 };
 
+#ifdef CONFIG_KPROBE_EVENTS
+/*
+ * Flags in config, used by dynamic PMU kprobe and uprobe
+ * The flags should match following PMU_FORMAT_ATTR().
+ *
+ * PERF_PROBE_CONFIG_IS_RETPROBE if set, create kretprobe/uretprobe
+ *                               if not set, create kprobe/uprobe
+ */
+enum perf_probe_config {
+	PERF_PROBE_CONFIG_IS_RETPROBE = 1U << 0,  /* [k,u]retprobe */
+};
+
+PMU_FORMAT_ATTR(retprobe, "config:0");
+
+static struct attribute *probe_attrs[] = {
+	&format_attr_retprobe.attr,
+	NULL,
+};
+
+static struct attribute_group probe_format_group = {
+	.name = "format",
+	.attrs = probe_attrs,
+};
+
+static const struct attribute_group *probe_attr_groups[] = {
+	&probe_format_group,
+	NULL,
+};
+
+static int perf_kprobe_event_init(struct perf_event *event);
+static struct pmu perf_kprobe = {
+	.task_ctx_nr	= perf_sw_context,
+	.event_init	= perf_kprobe_event_init,
+	.add		= perf_trace_add,
+	.del		= perf_trace_del,
+	.start		= perf_swevent_start,
+	.stop		= perf_swevent_stop,
+	.read		= perf_swevent_read,
+	.attr_groups	= probe_attr_groups,
+};
+
+static int perf_kprobe_event_init(struct perf_event *event)
+{
+	int err;
+	bool is_retprobe;
+
+	if (event->attr.type != perf_kprobe.type)
+		return -ENOENT;
+	/*
+	 * no branch sampling for probe events
+	 */
+	if (has_branch_stack(event))
+		return -EOPNOTSUPP;
+
+	is_retprobe = event->attr.config & PERF_PROBE_CONFIG_IS_RETPROBE;
+	err = perf_kprobe_init(event, is_retprobe);
+	if (err)
+		return err;
+
+	event->destroy = perf_kprobe_destroy;
+
+	return 0;
+}
+#endif /* CONFIG_KPROBE_EVENTS */
+
 static inline void perf_tp_register(void)
 {
 	perf_pmu_register(&perf_tracepoint, "tracepoint", PERF_TYPE_TRACEPOINT);
+#ifdef CONFIG_KPROBE_EVENTS
+	perf_pmu_register(&perf_kprobe, "kprobe", -1);
+#endif
 }
 
 static void perf_event_free_filter(struct perf_event *event)
@@ -8071,13 +8139,28 @@ static void perf_event_free_bpf_handler(struct perf_event *event)
 }
 #endif
 
+/*
+ * returns true if the event is a tracepoint, or a kprobe/upprobe created
+ * with perf_event_open()
+ */
+static inline bool perf_event_is_tracing(struct perf_event *event)
+{
+	if (event->pmu == &perf_tracepoint)
+		return true;
+#ifdef CONFIG_KPROBE_EVENTS
+	if (event->pmu == &perf_kprobe)
+		return true;
+#endif
+	return false;
+}
+
 static int perf_event_set_bpf_prog(struct perf_event *event, u32 prog_fd)
 {
 	bool is_kprobe, is_tracepoint, is_syscall_tp;
 	struct bpf_prog *prog;
 	int ret;
 
-	if (event->attr.type != PERF_TYPE_TRACEPOINT)
+	if (!perf_event_is_tracing(event))
 		return perf_event_set_bpf_handler(event, prog_fd);
 
 	is_kprobe = event->tp_event->flags & TRACE_EVENT_FL_UKPROBE;
@@ -8116,7 +8199,7 @@ static int perf_event_set_bpf_prog(struct perf_event *event, u32 prog_fd)
 
 static void perf_event_free_bpf_prog(struct perf_event *event)
 {
-	if (event->attr.type != PERF_TYPE_TRACEPOINT) {
+	if (!perf_event_is_tracing(event)) {
 		perf_event_free_bpf_handler(event);
 		return;
 	}
@@ -8535,47 +8618,36 @@ fail_clear_files:
 	return ret;
 }
 
-static int
-perf_tracepoint_set_filter(struct perf_event *event, char *filter_str)
-{
-	struct perf_event_context *ctx = event->ctx;
-	int ret;
-
-	/*
-	 * Beware, here be dragons!!
-	 *
-	 * the tracepoint muck will deadlock against ctx->mutex, but the tracepoint
-	 * stuff does not actually need it. So temporarily drop ctx->mutex. As per
-	 * perf_event_ctx_lock() we already have a reference on ctx.
-	 *
-	 * This can result in event getting moved to a different ctx, but that
-	 * does not affect the tracepoint state.
-	 */
-	mutex_unlock(&ctx->mutex);
-	ret = ftrace_profile_set_filter(event, event->attr.config, filter_str);
-	mutex_lock(&ctx->mutex);
-
-	return ret;
-}
-
 static int perf_event_set_filter(struct perf_event *event, void __user *arg)
 {
-	char *filter_str;
 	int ret = -EINVAL;
-
-	if ((event->attr.type != PERF_TYPE_TRACEPOINT ||
-	    !IS_ENABLED(CONFIG_EVENT_TRACING)) &&
-	    !has_addr_filter(event))
-		return -EINVAL;
+	char *filter_str;
 
 	filter_str = strndup_user(arg, PAGE_SIZE);
 	if (IS_ERR(filter_str))
 		return PTR_ERR(filter_str);
 
-	if (IS_ENABLED(CONFIG_EVENT_TRACING) &&
-	    event->attr.type == PERF_TYPE_TRACEPOINT)
-		ret = perf_tracepoint_set_filter(event, filter_str);
-	else if (has_addr_filter(event))
+#ifdef CONFIG_EVENT_TRACING
+	if (perf_event_is_tracing(event)) {
+		struct perf_event_context *ctx = event->ctx;
+
+		/*
+		 * Beware, here be dragons!!
+		 *
+		 * the tracepoint muck will deadlock against ctx->mutex, but
+		 * the tracepoint stuff does not actually need it. So
+		 * temporarily drop ctx->mutex. As per perf_event_ctx_lock() we
+		 * already have a reference on ctx.
+		 *
+		 * This can result in event getting moved to a different ctx,
+		 * but that does not affect the tracepoint state.
+		 */
+		mutex_unlock(&ctx->mutex);
+		ret = ftrace_profile_set_filter(event, event->attr.config, filter_str);
+		mutex_lock(&ctx->mutex);
+	} else
+#endif
+	if (has_addr_filter(event))
 		ret = perf_event_set_addr_filter(event, filter_str);
 
 	kfree(filter_str);