bpf: Implement function bpf_perf_event_read() that get the selected hardware PMU...

According to the perf_event_map_fd and index, the function
bpf_perf_event_read() can convert the corresponding map
value to the pointer to struct perf_event and return the
Hardware PMU counter value.

Signed-off-by: Kaixu Xia <xiakaixu@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 4fc1f40..f57d7fe 100644 (file)
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -190,6 +190,7 @@ extern const struct bpf_func_proto bpf_map_lookup_elem_proto;
 extern const struct bpf_func_proto bpf_map_update_elem_proto;
 extern const struct bpf_func_proto bpf_map_delete_elem_proto;
 
+extern const struct bpf_func_proto bpf_perf_event_read_proto;
 extern const struct bpf_func_proto bpf_get_prandom_u32_proto;
 extern const struct bpf_func_proto bpf_get_smp_processor_id_proto;
 extern const struct bpf_func_proto bpf_tail_call_proto;

diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index a1814e8..92a48e2 100644 (file)
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -271,6 +271,7 @@ enum bpf_func_id {
         */
        BPF_FUNC_skb_get_tunnel_key,
        BPF_FUNC_skb_set_tunnel_key,
+       BPF_FUNC_perf_event_read,       /* u64 bpf_perf_event_read(&map, index) */
        __BPF_FUNC_MAX_ID,
 };
 

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index cd307df..48e1c71 100644 (file)
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -238,6 +238,14 @@ static const char * const reg_type_str[] = {
        [CONST_IMM]             = "imm",
 };
 
+static const struct {
+       int map_type;
+       int func_id;
+} func_limit[] = {
+       {BPF_MAP_TYPE_PROG_ARRAY, BPF_FUNC_tail_call},
+       {BPF_MAP_TYPE_PERF_EVENT_ARRAY, BPF_FUNC_perf_event_read},
+};
+
 static void print_verifier_state(struct verifier_env *env)
 {
        enum bpf_reg_type t;
@@ -837,6 +845,28 @@ static int check_func_arg(struct verifier_env *env, u32 regno,
        return err;
 }
 
+static int check_map_func_compatibility(struct bpf_map *map, int func_id)
+{
+       bool bool_map, bool_func;
+       int i;
+
+       if (!map)
+               return 0;
+
+       for (i = 0; i <= ARRAY_SIZE(func_limit); i++) {
+               bool_map = (map->map_type == func_limit[i].map_type);
+               bool_func = (func_id == func_limit[i].func_id);
+               /* only when map & func pair match it can continue.
+                * don't allow any other map type to be passed into
+                * the special func;
+                */
+               if (bool_map != bool_func)
+                       return -EINVAL;
+       }
+
+       return 0;
+}
+
 static int check_call(struct verifier_env *env, int func_id)
 {
        struct verifier_state *state = &env->cur_state;
@@ -912,21 +942,9 @@ static int check_call(struct verifier_env *env, int func_id)
                return -EINVAL;
        }
 
-       if (map && map->map_type == BPF_MAP_TYPE_PROG_ARRAY &&
-           func_id != BPF_FUNC_tail_call)
-               /* prog_array map type needs extra care:
-                * only allow to pass it into bpf_tail_call() for now.
-                * bpf_map_delete_elem() can be allowed in the future,
-                * while bpf_map_update_elem() must only be done via syscall
-                */
-               return -EINVAL;
-
-       if (func_id == BPF_FUNC_tail_call &&
-           map->map_type != BPF_MAP_TYPE_PROG_ARRAY)
-               /* don't allow any other map type to be passed into
-                * bpf_tail_call()
-                */
-               return -EINVAL;
+       err = check_map_func_compatibility(map, func_id);
+       if (err)
+               return err;
 
        return 0;
 }

diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c
index 88a041a..ef9936d 100644 (file)
--- a/kernel/trace/bpf_trace.c
+++ b/kernel/trace/bpf_trace.c
@@ -158,6 +158,35 @@ const struct bpf_func_proto *bpf_get_trace_printk_proto(void)
        return &bpf_trace_printk_proto;
 }
 
+static u64 bpf_perf_event_read(u64 r1, u64 index, u64 r3, u64 r4, u64 r5)
+{
+       struct bpf_map *map = (struct bpf_map *) (unsigned long) r1;
+       struct bpf_array *array = container_of(map, struct bpf_array, map);
+       struct perf_event *event;
+
+       if (unlikely(index >= array->map.max_entries))
+               return -E2BIG;
+
+       event = (struct perf_event *)array->ptrs[index];
+       if (!event)
+               return -ENOENT;
+
+       /*
+        * we don't know if the function is run successfully by the
+        * return value. It can be judged in other places, such as
+        * eBPF programs.
+        */
+       return perf_event_read_local(event);
+}
+
+const struct bpf_func_proto bpf_perf_event_read_proto = {
+       .func           = bpf_perf_event_read,
+       .gpl_only       = false,
+       .ret_type       = RET_INTEGER,
+       .arg1_type      = ARG_CONST_MAP_PTR,
+       .arg2_type      = ARG_ANYTHING,
+};
+
 static const struct bpf_func_proto *kprobe_prog_func_proto(enum bpf_func_id func_id)
 {
        switch (func_id) {
@@ -183,6 +212,8 @@ static const struct bpf_func_proto *kprobe_prog_func_proto(enum bpf_func_id func
                return bpf_get_trace_printk_proto();
        case BPF_FUNC_get_smp_processor_id:
                return &bpf_get_smp_processor_id_proto;
+       case BPF_FUNC_perf_event_read:
+               return &bpf_perf_event_read_proto;
        default:
                return NULL;
        }