[PATCH bpf-next 0/3] selftests/bpf: Fixes for perf_buffer test

hi,
sending fixes for perf_buffer test on systems
with offline cpus.

v2:
  - resend due to delivery issues, no changes

thanks,
jirka


---
Jiri Olsa (3):
      selftests/bpf: Fix perf_buffer test on system with offline cpus
      selftests/bpf: Fix possible/online index mismatch in perf_buffer test
      selftests/bpf: Use nanosleep tracepoint in perf buffer test

 tools/testing/selftests/bpf/prog_tests/perf_buffer.c | 17 +++++++++--------
 tools/testing/selftests/bpf/progs/test_perf_buffer.c |  2 +-
 2 files changed, 10 insertions(+), 9 deletions(-)

The perf_buffer fails on system with offline cpus:

  # test_progs -t perf_buffer
  test_perf_buffer:PASS:nr_cpus 0 nsec
  test_perf_buffer:PASS:nr_on_cpus 0 nsec
  test_perf_buffer:PASS:skel_load 0 nsec
  test_perf_buffer:PASS:attach_kprobe 0 nsec
  test_perf_buffer:PASS:perf_buf__new 0 nsec
  test_perf_buffer:PASS:epoll_fd 0 nsec
  skipping offline CPU #24
  skipping offline CPU #25
  skipping offline CPU #26
  skipping offline CPU #27
  skipping offline CPU #28
  skipping offline CPU #29
  skipping offline CPU #30
  skipping offline CPU #31
  test_perf_buffer:PASS:perf_buffer__poll 0 nsec
  test_perf_buffer:PASS:seen_cpu_cnt 0 nsec
  test_perf_buffer:FAIL:buf_cnt got 24, expected 32
  Summary: 0/0 PASSED, 0 SKIPPED, 1 FAILED

Changing the test to check online cpus instead of possible.

Signed-off-by: Jiri Olsa <jolsa@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/perf_buffer.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/perf_buffer.c b/tools/testing/selftests/bpf/prog_tests/perf_buffer.c
index 6979aff4aab2..877600392851 100644
--- a/tools/testing/selftests/bpf/prog_tests/perf_buffer.c
+++ b/tools/testing/selftests/bpf/prog_tests/perf_buffer.c

@@ -107,8 +107,8 @@ void serial_test_perf_buffer(void)
 		  "expect %d, seen %d\n", nr_on_cpus, CPU_COUNT(&cpu_seen)))
 		goto out_free_pb;
 
-	if (CHECK(perf_buffer__buffer_cnt(pb) != nr_cpus, "buf_cnt",
-		  "got %zu, expected %d\n", perf_buffer__buffer_cnt(pb), nr_cpus))
+	if (CHECK(perf_buffer__buffer_cnt(pb) != nr_on_cpus, "buf_cnt",
+		  "got %zu, expected %d\n", perf_buffer__buffer_cnt(pb), nr_on_cpus))
 		goto out_close;
 
 	for (i = 0; i < nr_cpus; i++) {

-- 
2.31.1

The perf_buffer fails on system with offline cpus:

  # test_progs -t perf_buffer
  serial_test_perf_buffer:PASS:nr_cpus 0 nsec
  serial_test_perf_buffer:PASS:nr_on_cpus 0 nsec
  serial_test_perf_buffer:PASS:skel_load 0 nsec
  serial_test_perf_buffer:PASS:attach_kprobe 0 nsec
  serial_test_perf_buffer:PASS:perf_buf__new 0 nsec
  serial_test_perf_buffer:PASS:epoll_fd 0 nsec
  skipping offline CPU #4
  serial_test_perf_buffer:PASS:perf_buffer__poll 0 nsec
  serial_test_perf_buffer:PASS:seen_cpu_cnt 0 nsec
  serial_test_perf_buffer:PASS:buf_cnt 0 nsec
  ...
  serial_test_perf_buffer:PASS:fd_check 0 nsec
  serial_test_perf_buffer:PASS:drain_buf 0 nsec
  serial_test_perf_buffer:PASS:consume_buf 0 nsec
  serial_test_perf_buffer:FAIL:cpu_seen cpu 5 not seen
  #88 perf_buffer:FAIL
  Summary: 0/0 PASSED, 0 SKIPPED, 1 FAILED

If the offline cpu is from the middle of the possible set,
we get mismatch with possible and online cpu buffers.

The perf buffer test calls perf_buffer__consume_buffer for
all 'possible' cpus, but the library holds only 'online'
cpu buffers and perf_buffer__consume_buffer returns them
based on index.

Adding extra (online) index to keep track of online buffers,
we need the original (possible) index to trigger trace on
proper cpu.

Signed-off-by: Jiri Olsa <jolsa@kernel.org>
---
 .../testing/selftests/bpf/prog_tests/perf_buffer.c  | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/perf_buffer.c b/tools/testing/selftests/bpf/prog_tests/perf_buffer.c
index 877600392851..0b0cd045979b 100644
--- a/tools/testing/selftests/bpf/prog_tests/perf_buffer.c
+++ b/tools/testing/selftests/bpf/prog_tests/perf_buffer.c

@@ -45,7 +45,7 @@ int trigger_on_cpu(int cpu)
 
 void serial_test_perf_buffer(void)
 {
-	int err, on_len, nr_on_cpus = 0, nr_cpus, i;
+	int err, on_len, nr_on_cpus = 0, nr_cpus, i, j;
 	struct perf_buffer_opts pb_opts = {};
 	struct test_perf_buffer *skel;
 	cpu_set_t cpu_seen;

@@ -111,15 +111,15 @@ void serial_test_perf_buffer(void)
 		  "got %zu, expected %d\n", perf_buffer__buffer_cnt(pb), nr_on_cpus))
 		goto out_close;
 
-	for (i = 0; i < nr_cpus; i++) {
+	for (i = 0, j = 0; i < nr_cpus; i++) {
 		if (i >= on_len || !online[i])
 			continue;
 
-		fd = perf_buffer__buffer_fd(pb, i);
+		fd = perf_buffer__buffer_fd(pb, j);
 		CHECK(fd < 0 || last_fd == fd, "fd_check", "last fd %d == fd %d\n", last_fd, fd);
 		last_fd = fd;
 
-		err = perf_buffer__consume_buffer(pb, i);
+		err = perf_buffer__consume_buffer(pb, j);
 		if (CHECK(err, "drain_buf", "cpu %d, err %d\n", i, err))
 			goto out_close;
 

@@ -127,12 +127,13 @@ void serial_test_perf_buffer(void)
 		if (trigger_on_cpu(i))
 			goto out_close;
 
-		err = perf_buffer__consume_buffer(pb, i);
-		if (CHECK(err, "consume_buf", "cpu %d, err %d\n", i, err))
+		err = perf_buffer__consume_buffer(pb, j);
+		if (CHECK(err, "consume_buf", "cpu %d, err %d\n", j, err))
 			goto out_close;
 
 		if (CHECK(!CPU_ISSET(i, &cpu_seen), "cpu_seen", "cpu %d not seen\n", i))
 			goto out_close;
+		j++;
 	}
 
 out_free_pb:

-- 
2.31.1

The perf buffer tests triggers trace with nanosleep syscall,
but monitors all syscalls, which results in lot of data in the
buffer and makes it harder to debug. Let's lower the trace
traffic and monitor just nanosleep syscall.

Signed-off-by: Jiri Olsa <jolsa@kernel.org>
---
 tools/testing/selftests/bpf/progs/test_perf_buffer.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/progs/test_perf_buffer.c b/tools/testing/selftests/bpf/progs/test_perf_buffer.c
index d37ce29fd393..a08874c5bdf2 100644
--- a/tools/testing/selftests/bpf/progs/test_perf_buffer.c
+++ b/tools/testing/selftests/bpf/progs/test_perf_buffer.c

@@ -12,7 +12,7 @@ struct {
 	__type(value, int);
 } perf_buf_map SEC(".maps");
 
-SEC("tp/raw_syscalls/sys_enter")
+SEC("tp/syscalls/sys_enter_nanosleep")
 int handle_sys_enter(void *ctx)
 {
 	int cpu = bpf_get_smp_processor_id();

-- 
2.31.1

Jiri Olsa wrote:

hi,
sending fixes for perf_buffer test on systems
with offline cpus.

v2:
  - resend due to delivery issues, no changes

thanks,
jirka


---
Jiri Olsa (3):
      selftests/bpf: Fix perf_buffer test on system with offline cpus
      selftests/bpf: Fix possible/online index mismatch in perf_buffer test
      selftests/bpf: Use nanosleep tracepoint in perf buffer test

 tools/testing/selftests/bpf/prog_tests/perf_buffer.c | 17 +++++++++--------
 tools/testing/selftests/bpf/progs/test_perf_buffer.c |  2 +-
 2 files changed, 10 insertions(+), 9 deletions(-)

For the series.

Acked-by: John Fastabend <john.fastabend@gmail.com>

Hello:

This series was applied to bpf/bpf-next.git (master)
by Andrii Nakryiko [off-list ref]:

On Thu, 21 Oct 2021 13:41:29 +0200 you wrote:

hi,
sending fixes for perf_buffer test on systems
with offline cpus.

v2:
  - resend due to delivery issues, no changes

[...]

Here is the summary with links:
  - [bpf-next,1/3] selftests/bpf: Fix perf_buffer test on system with offline cpus
    https://git.kernel.org/bpf/bpf-next/c/d4121376ac7a
  - [bpf-next,2/3] selftests/bpf: Fix possible/online index mismatch in perf_buffer test
    https://git.kernel.org/bpf/bpf-next/c/aa274f98b269
  - [bpf-next,3/3] selftests/bpf: Use nanosleep tracepoint in perf buffer test
    https://git.kernel.org/bpf/bpf-next/c/99d099757ab4

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html