linux_dsm_epyc7002/samples/bpf/test_overhead_tp_kern.c
Alexei Starovoitov e3edfdec04 samples/bpf: add tracepoint vs kprobe performance tests
the first microbenchmark does
fd=open("/proc/self/comm");
for() {
  write(fd, "test");
}
and on 4 cpus in parallel:
                                      writes per sec
base (no tracepoints, no kprobes)         930k
with kprobe at __set_task_comm()          420k
with tracepoint at task:task_rename       730k

For kprobe + full bpf program manully fetches oldcomm, newcomm via bpf_probe_read.
For tracepint bpf program does nothing, since arguments are copied by tracepoint.

2nd microbenchmark does:
fd=open("/dev/urandom");
for() {
  read(fd, buf);
}
and on 4 cpus in parallel:
                                       reads per sec
base (no tracepoints, no kprobes)         300k
with kprobe at urandom_read()             279k
with tracepoint at random:urandom_read    290k

bpf progs attached to kprobe and tracepoint are noop.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-07 21:04:27 -04:00

37 lines
828 B
C

/* Copyright (c) 2016 Facebook
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of version 2 of the GNU General Public
* License as published by the Free Software Foundation.
*/
#include <uapi/linux/bpf.h>
#include "bpf_helpers.h"
/* from /sys/kernel/debug/tracing/events/task/task_rename/format */
struct task_rename {
__u64 pad;
__u32 pid;
char oldcomm[16];
char newcomm[16];
__u16 oom_score_adj;
};
SEC("tracepoint/task/task_rename")
int prog(struct task_rename *ctx)
{
return 0;
}
/* from /sys/kernel/debug/tracing/events/random/urandom_read/format */
struct urandom_read {
__u64 pad;
int got_bits;
int pool_left;
int input_left;
};
SEC("tracepoint/random/urandom_read")
int prog2(struct urandom_read *ctx)
{
return 0;
}
char _license[] SEC("license") = "GPL";