bpf: enable program stats

JITed BPF programs are indistinguishable from kernel functions, but unlike
kernel code BPF code can be changed often.
Typical approach of "perf record" + "perf report" profiling and tuning of
kernel code works just as well for BPF programs, but kernel code doesn't
need to be monitored whereas BPF programs do.
Users load and run large amount of BPF programs.
These BPF stats allow tools monitor the usage of BPF on the server.
The monitoring tools will turn sysctl kernel.bpf_stats_enabled
on and off for few seconds to sample average cost of the programs.
Aggregated data over hours and days will provide an insight into cost of BPF
and alarms can trigger in case given program suddenly gets more expensive.

The cost of two sched_clock() per program invocation adds ~20 nsec.
Fast BPF progs (like selftests/bpf/progs/test_pkt_access.c) will slow down
from ~10 nsec to ~30 nsec.
static_key minimizes the cost of the stats collection.
There is no measurable difference before/after this patch
with kernel.bpf_stats_enabled=0

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>

2 files changed, 28 insertions, 1 deletions

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index de18227b3d95..a2132e09dc1c 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -16,6 +16,7 @@
 #include <linux/rbtree_latch.h>
 #include <linux/numa.h>
 #include <linux/wait.h>
+#include <linux/u64_stats_sync.h>
 
 struct bpf_verifier_env;
 struct perf_event;
@@ -340,6 +341,12 @@ enum bpf_cgroup_storage_type {
 
 #define MAX_BPF_CGROUP_STORAGE_TYPE __BPF_CGROUP_STORAGE_MAX
 
+struct bpf_prog_stats {
+	u64 cnt;
+	u64 nsecs;
+	struct u64_stats_sync syncp;
+};
+
 struct bpf_prog_aux {
 	atomic_t refcnt;
 	u32 used_map_cnt;
@@ -389,6 +396,7 @@ struct bpf_prog_aux {
 	 * main prog always has linfo_idx == 0
 	 */
 	u32 linfo_idx;
+	struct bpf_prog_stats __percpu *stats;
 	union {
 		struct work_struct work;
 		struct rcu_head	rcu;
@@ -559,6 +567,7 @@ void bpf_map_area_free(void *base);
 void bpf_map_init_from_attr(struct bpf_map *map, union bpf_attr *attr);
 
 extern int sysctl_unprivileged_bpf_disabled;
+extern int sysctl_bpf_stats_enabled;
 
 int bpf_map_new_fd(struct bpf_map *map, int flags);
 int bpf_prog_new_fd(struct bpf_prog *prog);
diff --git a/include/linux/filter.h b/include/linux/filter.h
index f32b3eca5a04..7e5e3db11106 100644
--- a/include/linux/filter.h
+++ b/include/linux/filter.h
@@ -533,7 +533,24 @@ struct sk_filter {
 	struct bpf_prog	*prog;
 };
 
-#define BPF_PROG_RUN(filter, ctx)  ({ cant_sleep(); (*(filter)->bpf_func)(ctx, (filter)->insnsi); })
+DECLARE_STATIC_KEY_FALSE(bpf_stats_enabled_key);
+
+#define BPF_PROG_RUN(prog, ctx)	({				\
+	u32 ret;						\
+	cant_sleep();						\
+	if (static_branch_unlikely(&bpf_stats_enabled_key)) {	\
+		struct bpf_prog_stats *stats;			\
+		u64 start = sched_clock();			\
+		ret = (*(prog)->bpf_func)(ctx, (prog)->insnsi);	\
+		stats = this_cpu_ptr(prog->aux->stats);		\
+		u64_stats_update_begin(&stats->syncp);		\
+		stats->cnt++;					\
+		stats->nsecs += sched_clock() - start;		\
+		u64_stats_update_end(&stats->syncp);		\
+	} else {						\
+		ret = (*(prog)->bpf_func)(ctx, (prog)->insnsi);	\
+	}							\
+	ret; })
 
 #define BPF_SKB_CB_LEN QDISC_CB_PRIV_LEN
 
@@ -764,6 +781,7 @@ void bpf_prog_free_jited_linfo(struct bpf_prog *prog);
 void bpf_prog_free_unused_jited_linfo(struct bpf_prog *prog);
 
 struct bpf_prog *bpf_prog_alloc(unsigned int size, gfp_t gfp_extra_flags);
+struct bpf_prog *bpf_prog_alloc_no_stats(unsigned int size, gfp_t gfp_extra_flags);
 struct bpf_prog *bpf_prog_realloc(struct bpf_prog *fp_old, unsigned int size,
 				  gfp_t gfp_extra_flags);
 void __bpf_prog_free(struct bpf_prog *fp);