From nobody Thu Sep 11 12:49:20 2025 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 5B1EEC83F2C for ; Mon, 4 Sep 2023 02:37:03 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1351068AbjIDChD (ORCPT ); Sun, 3 Sep 2023 22:37:03 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:60490 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1350946AbjIDCg4 (ORCPT ); Sun, 3 Sep 2023 22:36:56 -0400 Received: from szxga08-in.huawei.com (szxga08-in.huawei.com [45.249.212.255]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 8F56D110; Sun, 3 Sep 2023 19:36:52 -0700 (PDT) Received: from kwepemm600003.china.huawei.com (unknown [172.30.72.53]) by szxga08-in.huawei.com (SkyGuard) with ESMTP id 4RfCQx5SRpz1M8v2; Mon, 4 Sep 2023 10:35:05 +0800 (CST) Received: from localhost.localdomain (10.67.174.95) by kwepemm600003.china.huawei.com (7.193.23.202) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.2507.31; Mon, 4 Sep 2023 10:36:49 +0800 From: Yang Jihong To: , , , , , , , , , , , , , , , CC: Subject: [PATCH v8 4/6] perf record: Track sideband events for all CPUs when tracing selected CPUs Date: Mon, 4 Sep 2023 02:33:38 +0000 Message-ID: <20230904023340.12707-5-yangjihong1@huawei.com> X-Mailer: git-send-email 2.30.GIT In-Reply-To: <20230904023340.12707-1-yangjihong1@huawei.com> References: <20230904023340.12707-1-yangjihong1@huawei.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Originating-IP: [10.67.174.95] X-ClientProxiedBy: dggems705-chm.china.huawei.com (10.3.19.182) To kwepemm600003.china.huawei.com (7.193.23.202) X-CFilter-Loop: Reflected Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" User space tasks can migrate between CPUs, we need to track side-band events for all CPUs. The specific scenarios are as follows: CPU0 CPU1 perf record -C 0 start taskA starts to be created and executed -> PERF_RECORD_COMM and PERF_RECORD_MMAP events only deliver to CPU1 ...... | migrate to CPU0 | Running on CPU0 <----------/ ... perf record -C 0 stop Now perf samples the PC of taskA. However, perf does not record the PERF_RECORD_COMM and PERF_RECORD_MMAP events of taskA. Therefore, the comm and symbols of taskA cannot be parsed. The solution is to record sideband events for all CPUs when tracing selected CPUs. Because this modifies the default behavior, add related comments to the perf record man page. The sys_perf_event_open invoked is as follows: # perf --debug verbose=3D3 record -e cpu-clock -C 1 true Opening: cpu-clock ------------------------------------------------------------ perf_event_attr: type 1 (PERF_TYPE_SOFTWARE) size 136 config 0 (PERF_COUNT_SW_CPU_CLOCK) { sample_period, sample_freq } 4000 sample_type IP|TID|TIME|CPU|PERIOD|IDENTIFIER read_format ID|LOST disabled 1 inherit 1 freq 1 sample_id_all 1 exclude_guest 1 ------------------------------------------------------------ sys_perf_event_open: pid -1 cpu 1 group_fd -1 flags 0x8 =3D 5 Opening: dummy:u ------------------------------------------------------------ perf_event_attr: type 1 (PERF_TYPE_SOFTWARE) size 136 config 0x9 (PERF_COUNT_SW_DUMMY) { sample_period, sample_freq } 1 sample_type IP|TID|TIME|CPU|IDENTIFIER read_format ID|LOST inherit 1 exclude_kernel 1 exclude_hv 1 mmap 1 comm 1 task 1 sample_id_all 1 exclude_guest 1 mmap2 1 comm_exec 1 ksymbol 1 bpf_event 1 ------------------------------------------------------------ sys_perf_event_open: pid -1 cpu 0 group_fd -1 flags 0x8 =3D 6 sys_perf_event_open: pid -1 cpu 1 group_fd -1 flags 0x8 =3D 7 sys_perf_event_open: pid -1 cpu 2 group_fd -1 flags 0x8 =3D 9 sys_perf_event_open: pid -1 cpu 3 group_fd -1 flags 0x8 =3D 10 sys_perf_event_open: pid -1 cpu 4 group_fd -1 flags 0x8 =3D 11 sys_perf_event_open: pid -1 cpu 5 group_fd -1 flags 0x8 =3D 12 sys_perf_event_open: pid -1 cpu 6 group_fd -1 flags 0x8 =3D 13 sys_perf_event_open: pid -1 cpu 7 group_fd -1 flags 0x8 =3D 14 Signed-off-by: Yang Jihong --- tools/perf/Documentation/perf-record.txt | 3 +++ tools/perf/builtin-record.c | 30 +++++++++++++++++++++++- 2 files changed, 32 insertions(+), 1 deletion(-) diff --git a/tools/perf/Documentation/perf-record.txt b/tools/perf/Document= ation/perf-record.txt index d5217be012d7..1889f66addf2 100644 --- a/tools/perf/Documentation/perf-record.txt +++ b/tools/perf/Documentation/perf-record.txt @@ -374,6 +374,9 @@ comma-separated list with no space: 0,1. Ranges of CPUs= are specified with -: 0- In per-thread mode with inheritance mode on (default), samples are capture= d only when the thread executes on the designated CPUs. Default is to monitor all CPUs. =20 +User space tasks can migrate between CPUs, so when tracing selected CPUs, +a dummy event is created to track sideband for all CPUs. + -B:: --no-buildid:: Do not save the build ids of binaries in the perf.data files. This skips diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c index 83bd1f117191..dcf288a4fb9a 100644 --- a/tools/perf/builtin-record.c +++ b/tools/perf/builtin-record.c @@ -906,10 +906,30 @@ static int record__config_off_cpu(struct record *rec) return off_cpu_prepare(rec->evlist, &rec->opts.target, &rec->opts); } =20 +static bool record__tracking_system_wide(struct record *rec) +{ + struct evlist *evlist =3D rec->evlist; + struct evsel *evsel; + + /* + * If non-dummy evsel exists, system_wide sideband is need to + * help parse sample information. + * For example, PERF_EVENT_MMAP event to help parse symbol, + * and PERF_EVENT_COMM event to help parse task executable name. + */ + evlist__for_each_entry(evlist, evsel) { + if (!evsel__is_dummy_event(evsel)) + return true; + } + + return false; +} + static int record__config_tracking_events(struct record *rec) { struct record_opts *opts =3D &rec->opts; struct evlist *evlist =3D rec->evlist; + bool system_wide =3D false; struct evsel *evsel; =20 /* @@ -919,7 +939,15 @@ static int record__config_tracking_events(struct recor= d *rec) */ if (opts->target.initial_delay || target__has_cpu(&opts->target) || perf_pmus__num_core_pmus() > 1) { - evsel =3D evlist__findnew_tracking_event(evlist, false); + + /* + * User space tasks can migrate between CPUs, so when tracing + * selected CPUs, sideband for all CPUs is still needed. + */ + if (!!opts->target.cpu_list && record__tracking_system_wide(rec)) + system_wide =3D true; + + evsel =3D evlist__findnew_tracking_event(evlist, system_wide); if (!evsel) return -ENOMEM; =20 --=20 2.30.GIT