From: Song Liu <songliubraving@fb.com>
To: Namhyung Kim <namhyung@kernel.org>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>,
Jiri Olsa <jolsa@redhat.com>, Ingo Molnar <mingo@kernel.org>,
Peter Zijlstra <peterz@infradead.org>,
LKML <linux-kernel@vger.kernel.org>,
Andi Kleen <ak@linux.intel.com>, Ian Rogers <irogers@google.com>,
Stephane Eranian <eranian@google.com>
Subject: Re: [PATCH 4/4] perf stat: Enable BPF counter with --for-each-cgroup
Date: Wed, 30 Jun 2021 18:47:16 +0000 [thread overview]
Message-ID: <43811928-C46C-45CE-AB5A-4DE84DCDB1AF@fb.com> (raw)
In-Reply-To: <20210625071826.608504-5-namhyung@kernel.org>
> On Jun 25, 2021, at 12:18 AM, Namhyung Kim <namhyung@kernel.org> wrote:
>
> Recently bperf was added to use BPF to count perf events for various
> purposes. This is an extension for the approach and targetting to
> cgroup usages.
>
> Unlike the other bperf, it doesn't share the events with other
> processes but it'd reduce unnecessary events (and the overhead of
> multiplexing) for each monitored cgroup within the perf session.
>
> When --for-each-cgroup is used with --bpf-counters, it will open
> cgroup-switches event per cpu internally and attach the new BPF
> program to read given perf_events and to aggregate the results for
> cgroups. It's only called when task is switched to a task in a
> different cgroup.
>
> Cc: Song Liu <songliubraving@fb.com>
> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
> ---
> tools/perf/Makefile.perf | 17 +-
> tools/perf/util/Build | 1 +
> tools/perf/util/bpf_counter.c | 5 +
> tools/perf/util/bpf_counter_cgroup.c | 299 ++++++++++++++++++++
> tools/perf/util/bpf_skel/bperf_cgroup.bpf.c | 191 +++++++++++++
> tools/perf/util/cgroup.c | 2 +
> tools/perf/util/cgroup.h | 1 +
> 7 files changed, 515 insertions(+), 1 deletion(-)
> create mode 100644 tools/perf/util/bpf_counter_cgroup.c
> create mode 100644 tools/perf/util/bpf_skel/bperf_cgroup.bpf.c
[...]
> diff --git a/tools/perf/util/bpf_counter_cgroup.c b/tools/perf/util/bpf_counter_cgroup.c
> new file mode 100644
> index 000000000000..327f97a23a84
> --- /dev/null
> +++ b/tools/perf/util/bpf_counter_cgroup.c
> @@ -0,0 +1,299 @@
> +// SPDX-License-Identifier: GPL-2.0
> +
> +/* Copyright (c) 2019 Facebook */
I am not sure whether this ^^^ is accurate.
> +/* Copyright (c) 2021 Google */
> +
> +#include <assert.h>
> +#include <limits.h>
> +#include <unistd.h>
> +#include <sys/file.h>
> +#include <sys/time.h>
> +#include <sys/resource.h>
> +#include <linux/err.h>
> +#include <linux/zalloc.h>
> +#include <linux/perf_event.h>
> +#include <api/fs/fs.h>
> +#include <perf/bpf_perf.h>
> +
> +#include "affinity.h"
> +#include "bpf_counter.h"
> +#include "cgroup.h"
> +#include "counts.h"
> +#include "debug.h"
> +#include "evsel.h"
> +#include "evlist.h"
> +#include "target.h"
> +#include "cpumap.h"
> +#include "thread_map.h"
> +
> +#include "bpf_skel/bperf_cgroup.skel.h"
> +
> +static struct perf_event_attr cgrp_switch_attr = {
> + .type = PERF_TYPE_SOFTWARE,
> + .config = PERF_COUNT_SW_CGROUP_SWITCHES,
> + .size = sizeof(cgrp_switch_attr),
> + .sample_period = 1,
> + .disabled = 1,
> +};
> +
> +static struct evsel *cgrp_switch;
> +static struct bperf_cgroup_bpf *skel;
> +
> +#define FD(evt, cpu) (*(int *)xyarray__entry(evt->core.fd, cpu, 0))
> +
> +static int bperf_load_program(struct evlist *evlist)
> +{
> + struct bpf_link *link;
> + struct evsel *evsel;
> + struct cgroup *cgrp, *leader_cgrp;
> + __u32 i, cpu;
> + int nr_cpus = evlist->core.all_cpus->nr;
> + int total_cpus = cpu__max_cpu();
> + int map_size, map_fd;
> + int prog_fd, err;
> +
> + skel = bperf_cgroup_bpf__open();
> + if (!skel) {
> + pr_err("Failed to open cgroup skeleton\n");
> + return -1;
> + }
> +
> + skel->rodata->num_cpus = total_cpus;
> + skel->rodata->num_events = evlist->core.nr_entries / nr_cgroups;
> +
> + BUG_ON(evlist->core.nr_entries % nr_cgroups != 0);
> +
> + /* we need one copy of events per cpu for reading */
> + map_size = total_cpus * evlist->core.nr_entries / nr_cgroups;
> + bpf_map__resize(skel->maps.events, map_size);
> + bpf_map__resize(skel->maps.cgrp_idx, nr_cgroups);
> + /* previous result is saved in a per-cpu array */
> + map_size = evlist->core.nr_entries / nr_cgroups;
> + bpf_map__resize(skel->maps.prev_readings, map_size);
> + /* cgroup result needs all events (per-cpu) */
> + map_size = evlist->core.nr_entries;
> + bpf_map__resize(skel->maps.cgrp_readings, map_size);
> +
> + set_max_rlimit();
> +
> + err = bperf_cgroup_bpf__load(skel);
> + if (err) {
> + pr_err("Failed to load cgroup skeleton\n");
> + goto out;
> + }
> +
> + if (cgroup_is_v2("perf_event") > 0)
> + skel->bss->use_cgroup_v2 = 1;
> +
> + err = -1;
> +
> + cgrp_switch = evsel__new(&cgrp_switch_attr);
> + if (evsel__open_per_cpu(cgrp_switch, evlist->core.all_cpus, -1) < 0) {
> + pr_err("Failed to open cgroup switches event\n");
> + goto out;
> + }
> +
> + for (i = 0; i < nr_cpus; i++) {
> + link = bpf_program__attach_perf_event(skel->progs.on_cgrp_switch,
> + FD(cgrp_switch, i));
> + if (IS_ERR(link)) {
> + pr_err("Failed to attach cgroup program\n");
> + err = PTR_ERR(link);
> + goto out;
> + }
> + }
> +
> + /*
> + * Update cgrp_idx map from cgroup-id to event index.
> + */
> + cgrp = NULL;
> + i = 0;
> +
> + evlist__for_each_entry(evlist, evsel) {
> + if (cgrp == NULL || evsel->cgrp == leader_cgrp) {
> + leader_cgrp = evsel->cgrp;
> + evsel->cgrp = NULL;
> +
> + /* open single copy of the events w/o cgroup */
> + err = evsel__open_per_cpu(evsel, evlist->core.all_cpus, -1);
> + if (err) {
> + pr_err("Failed to open first cgroup events\n");
> + goto out;
> + }
> +
> + map_fd = bpf_map__fd(skel->maps.events);
> + for (cpu = 0; cpu < nr_cpus; cpu++) {
> + int fd = FD(evsel, cpu);
> + __u32 idx = evsel->idx * total_cpus +
> + evlist->core.all_cpus->map[cpu];
> +
> + err = bpf_map_update_elem(map_fd, &idx, &fd,
> + BPF_ANY);
> + if (err < 0) {
> + pr_err("Failed to update perf_event fd\n");
> + goto out;
> + }
> + }
> +
> + evsel->cgrp = leader_cgrp;
> + }
> + evsel->supported = true;
> +
> + if (evsel->cgrp == cgrp)
> + continue;
> +
> + cgrp = evsel->cgrp;
> +
> + if (read_cgroup_id(cgrp) < 0) {
> + pr_err("Failed to get cgroup id\n");
> + err = -1;
> + goto out;
> + }
> +
> + map_fd = bpf_map__fd(skel->maps.cgrp_idx);
> + err = bpf_map_update_elem(map_fd, &cgrp->id, &i, BPF_ANY);
> + if (err < 0) {
> + pr_err("Failed to update cgroup index map\n");
> + goto out;
> + }
> +
> + i++;
> + }
> +
> + /*
> + * bperf uses BPF_PROG_TEST_RUN to get accurate reading. Check
> + * whether the kernel support it
> + */
> + prog_fd = bpf_program__fd(skel->progs.trigger_read);
> + err = bperf_trigger_reading(prog_fd, 0);
> + if (err) {
> + pr_debug("The kernel does not support test_run for raw_tp BPF programs.\n"
> + "Therefore, --for-each-cgroup might show inaccurate readings\n");
I think this should be a warning, and we should set err = 0 to continue?
> + }
> +
> +out:
> + return err;
> +}
> +
[...]
> +
> +/*
> + * trigger the leader prog on each cpu, so the cgrp_reading map could get
> + * the latest results.
> + */
> +static int bperf_cgrp__sync_counters(struct evlist *evlist)
> +{
> + int i, cpu;
> + int nr_cpus = evlist->core.all_cpus->nr;
> + int prog_fd = bpf_program__fd(skel->progs.trigger_read);
> +
> + for (i = 0; i < nr_cpus; i++) {
> + cpu = evlist->core.all_cpus->map[i];
> + bperf_trigger_reading(prog_fd, cpu);
> + }
> +
> + return 0;
> +}
> +
> +static int bperf_cgrp__enable(struct evsel *evsel)
> +{
Do we need to call bperf_cgrp__sync_counters() before setting enabled to 1?
If we don't, we may count some numbers before setting enabled to 1, no?
> + skel->bss->enabled = 1;
> + return 0;
> +}
[...]
next prev parent reply other threads:[~2021-06-30 18:47 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-06-25 7:18 [PATCHSET v4 0/4] perf stat: Enable BPF counters with --for-each-cgroup Namhyung Kim
2021-06-25 7:18 ` [PATCH 1/4] perf tools: Add read_cgroup_id() function Namhyung Kim
2021-07-01 17:59 ` Arnaldo Carvalho de Melo
2021-06-25 7:18 ` [PATCH 2/4] perf tools: Add cgroup_is_v2() helper Namhyung Kim
2021-06-29 15:51 ` Ian Rogers
2021-06-30 6:35 ` Namhyung Kim
2021-06-30 18:43 ` Arnaldo Carvalho de Melo
2021-06-25 7:18 ` [PATCH 3/4] perf tools: Move common bpf functions to bpf_counter.h Namhyung Kim
2021-06-30 18:28 ` Song Liu
2021-07-01 19:09 ` Arnaldo Carvalho de Melo
2021-07-01 20:11 ` Namhyung Kim
2021-06-25 7:18 ` [PATCH 4/4] perf stat: Enable BPF counter with --for-each-cgroup Namhyung Kim
2021-06-30 18:47 ` Song Liu [this message]
2021-06-30 20:09 ` Namhyung Kim
2021-07-01 20:16 ` Namhyung Kim
2021-06-30 18:50 ` Arnaldo Carvalho de Melo
2021-06-30 20:12 ` Namhyung Kim
2021-07-01 13:43 ` Arnaldo Carvalho de Melo
2021-07-01 17:10 ` Namhyung Kim
2021-06-27 15:29 ` [PATCHSET v4 0/4] perf stat: Enable BPF counters " Namhyung Kim
2021-06-30 6:19 ` Namhyung Kim
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=43811928-C46C-45CE-AB5A-4DE84DCDB1AF@fb.com \
--to=songliubraving@fb.com \
--cc=acme@kernel.org \
--cc=ak@linux.intel.com \
--cc=eranian@google.com \
--cc=irogers@google.com \
--cc=jolsa@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=namhyung@kernel.org \
--cc=peterz@infradead.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).