perf, bpf: Introduce PERF_RECORD_BPF_EVENT

For better performance analysis of BPF programs, this patch introduces
PERF_RECORD_BPF_EVENT, a new perf_event_type that exposes BPF program
load/unload information to user space.

Each BPF program may contain up to BPF_MAX_SUBPROGS (256) sub programs.
The following example shows kernel symbols for a BPF program with 7 sub
programs:

    ffffffffa0257cf9 t bpf_prog_b07ccb89267cf242_F
    ffffffffa02592e1 t bpf_prog_2dcecc18072623fc_F
    ffffffffa025b0e9 t bpf_prog_bb7a405ebaec5d5c_F
    ffffffffa025dd2c t bpf_prog_a7540d4a39ec1fc7_F
    ffffffffa025fcca t bpf_prog_05762d4ade0e3737_F
    ffffffffa026108f t bpf_prog_db4bd11e35df90d4_F
    ffffffffa0263f00 t bpf_prog_89d64e4abf0f0126_F
    ffffffffa0257cf9 t bpf_prog_ae31629322c4b018__dummy_tracepoi

When a bpf program is loaded, PERF_RECORD_KSYMBOL is generated for each
of these sub programs. Therefore, PERF_RECORD_BPF_EVENT is not needed
for simple profiling.

For annotation, user space need to listen to PERF_RECORD_BPF_EVENT and
gather more information about these (sub) programs via sys_bpf.

Signed-off-by: Song Liu <songliubraving@fb.com>
Reviewed-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradeaed.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: kernel-team@fb.com
Cc: netdev@vger.kernel.org
Link: http://lkml.kernel.org/r/20190117161521.1341602-4-songliubraving@fb.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

1 files changed, 115 insertions, 0 deletions

diff --git a/kernel/events/core.c b/kernel/events/core.c
index e04ab5f325cf..236bb8ddb7bc 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -386,6 +386,7 @@ static atomic_t nr_task_events __read_mostly;
 static atomic_t nr_freq_events __read_mostly;
 static atomic_t nr_switch_events __read_mostly;
 static atomic_t nr_ksymbol_events __read_mostly;
+static atomic_t nr_bpf_events __read_mostly;
 
 static LIST_HEAD(pmus);
 static DEFINE_MUTEX(pmus_lock);
@@ -4308,6 +4309,8 @@ static void unaccount_event(struct perf_event *event)
                 dec = true;
         if (event->attr.ksymbol)
                 atomic_dec(&nr_ksymbol_events);
+        if (event->attr.bpf_event)
+                atomic_dec(&nr_bpf_events);
 
         if (dec) {
                 if (!atomic_add_unless(&perf_sched_count, -1, 1))
@@ -7747,6 +7750,116 @@ err:
         WARN_ONCE(1, "%s: Invalid KSYMBOL type 0x%x\n", __func__, ksym_type);
 }
 
+/*
+ * bpf program load/unload tracking
+ */
+
+struct perf_bpf_event {
+        struct bpf_prog *prog;
+        struct {
+                struct perf_event_header        header;
+                u16                             type;
+                u16                             flags;
+                u32                             id;
+                u8                              tag[BPF_TAG_SIZE];
+        } event_id;
+};
+
+static int perf_event_bpf_match(struct perf_event *event)
+{
+        return event->attr.bpf_event;
+}
+
+static void perf_event_bpf_output(struct perf_event *event, void *data)
+{
+        struct perf_bpf_event *bpf_event = data;
+        struct perf_output_handle handle;
+        struct perf_sample_data sample;
+        int ret;
+
+        if (!perf_event_bpf_match(event))
+                return;
+
+        perf_event_header__init_id(&bpf_event->event_id.header,
+                                   &sample, event);
+        ret = perf_output_begin(&handle, event,
+                                bpf_event->event_id.header.size);
+        if (ret)
+                return;
+
+        perf_output_put(&handle, bpf_event->event_id);
+        perf_event__output_id_sample(event, &handle, &sample);
+
+        perf_output_end(&handle);
+}
+
+static void perf_event_bpf_emit_ksymbols(struct bpf_prog *prog,
+                                         enum perf_bpf_event_type type)
+{
+        bool unregister = type == PERF_BPF_EVENT_PROG_UNLOAD;
+        char sym[KSYM_NAME_LEN];
+        int i;
+
+        if (prog->aux->func_cnt == 0) {
+                bpf_get_prog_name(prog, sym);
+                perf_event_ksymbol(PERF_RECORD_KSYMBOL_TYPE_BPF,
+                                   (u64)(unsigned long)prog->bpf_func,
+                                   prog->jited_len, unregister, sym);
+        } else {
+                for (i = 0; i < prog->aux->func_cnt; i++) {
+                        struct bpf_prog *subprog = prog->aux->func[i];
+
+                        bpf_get_prog_name(subprog, sym);
+                        perf_event_ksymbol(
+                                PERF_RECORD_KSYMBOL_TYPE_BPF,
+                                (u64)(unsigned long)subprog->bpf_func,
+                                subprog->jited_len, unregister, sym);
+                }
+        }
+}
+
+void perf_event_bpf_event(struct bpf_prog *prog,
+                          enum perf_bpf_event_type type,
+                          u16 flags)
+{
+        struct perf_bpf_event bpf_event;
+
+        if (type <= PERF_BPF_EVENT_UNKNOWN ||
+            type >= PERF_BPF_EVENT_MAX)
+                return;
+
+        switch (type) {
+        case PERF_BPF_EVENT_PROG_LOAD:
+        case PERF_BPF_EVENT_PROG_UNLOAD:
+                if (atomic_read(&nr_ksymbol_events))
+                        perf_event_bpf_emit_ksymbols(prog, type);
+                break;
+        default:
+                break;
+        }
+
+        if (!atomic_read(&nr_bpf_events))
+                return;
+
+        bpf_event = (struct perf_bpf_event){
+                .prog = prog,
+                .event_id = {
+                        .header = {
+                                .type = PERF_RECORD_BPF_EVENT,
+                                .size = sizeof(bpf_event.event_id),
+                        },
+                        .type = type,
+                        .flags = flags,
+                        .id = prog->aux->id,
+                },
+        };
+
+        BUILD_BUG_ON(BPF_TAG_SIZE % sizeof(u64));
+
+        memcpy(bpf_event.event_id.tag, prog->tag, BPF_TAG_SIZE);
+        perf_iterate_sb(perf_event_bpf_output, &bpf_event, NULL);
+}
+
 void perf_event_itrace_started(struct perf_event *event)
 {
         event->attach_state |= PERF_ATTACH_ITRACE;
@@ -10008,6 +10121,8 @@ static void account_event(struct perf_event *event)
                 inc = true;
         if (event->attr.ksymbol)
                 atomic_inc(&nr_ksymbol_events);
+        if (event->attr.bpf_event)
+                atomic_inc(&nr_bpf_events);
 
         if (inc) {
                 /*