From patchwork Wed Nov 29 06:01:58 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Ian Rogers X-Patchwork-Id: 13472215 X-Patchwork-Delegate: bpf@iogearbox.net Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=google.com header.i=@google.com header.b="kX7c6gCv" Received: from mail-yw1-x114a.google.com (mail-yw1-x114a.google.com [IPv6:2607:f8b0:4864:20::114a]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 234DF19B1 for ; Tue, 28 Nov 2023 22:02:18 -0800 (PST) Received: by mail-yw1-x114a.google.com with SMTP id 00721157ae682-5cdde93973aso83917367b3.1 for ; Tue, 28 Nov 2023 22:02:18 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1701237737; x=1701842537; darn=vger.kernel.org; h=to:from:subject:references:mime-version:message-id:in-reply-to:date :from:to:cc:subject:date:message-id:reply-to; bh=6thajjQRkz+HkHW7D3sSCmpzEeB477d5Mr+y0iQJVzs=; b=kX7c6gCvFHw5an55jNpK+MCu0n06UnFZxpirz/t1R5a+0/qU6FHchFYK0P3WAbTCNe 7jCpy0TUz3XvRTvuxNSDzEe6NHfQvoroF07Bf2C49yGQuXLSvgmYoLKq0qeXUYC3cEgD niQMAV1yT7yxf9enxT5SYiFJwaJx8O81E1dJfadtGep4iH6fMzCC5zfduJBLOH0te26/ eqiqV62q0rHdJj3IQCVbYLlhxcO3Sl5b2kChwLyJx9Cyd79ycy6KwxoMbH27JBFNvY9T 3M4kxZX/x2AHpOHW3yYE1ehG4/zjlra1zTcMRZlrBaTbzPUPC/lOJ6BXBcpdvkZkmCR5 GYBg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1701237737; x=1701842537; h=to:from:subject:references:mime-version:message-id:in-reply-to:date :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=6thajjQRkz+HkHW7D3sSCmpzEeB477d5Mr+y0iQJVzs=; b=W2ZIecAC0uPxQgW0b19q7FwDs8GdP21c+5QRaDFII09anUFhqY8k/38Ga4/XZVbQxb sLumlVN+srXakES1wG9/KXobkZs6McL3apA6hjRQdmWP98IPeKFc9AK060y07IXJKuzg XjYrRN2He0yPOytJYMU6tgl4o9tROAfPWwXSCMwVSE4BxY3W543Fr/z4vkji4DDfOr/o uEAwRsE75/EfCHdM1JdgNl4mxStE7HgnUt/bcIm1hQ08I3Gsx43f70Gq4hab4Fnd0Dd/ 8HcRMKPD7alATV7j0lNeC745kfbdNSqQNWrwDVEC6iNMFjkDgM/7NyjabVYLHtZmfta5 Jy+Q== X-Gm-Message-State: AOJu0Yxq81WPTyU/9bh1T8EUDdDB5Ubms0WftNeZHjgGRyjcdZ7T/EUo fJd1Wu95WYnVl1123ocaCakIqKTDdqQZ X-Google-Smtp-Source: AGHT+IHStCf02eR5Z7G2oMb+qjaqd+bye5wyKkPORIP+SdC4EPzelKmD6TjLiarOMrn6W9b9mIBoWBllGFGX X-Received: from irogers.svl.corp.google.com ([2620:15c:2a3:200:763b:80fa:23ca:96f8]) (user=irogers job=sendgmr) by 2002:a05:690c:2e8a:b0:5ce:dff:f7a1 with SMTP id eu10-20020a05690c2e8a00b005ce0dfff7a1mr482531ywb.9.1701237737183; Tue, 28 Nov 2023 22:02:17 -0800 (PST) Date: Tue, 28 Nov 2023 22:01:58 -0800 In-Reply-To: <20231129060211.1890454-1-irogers@google.com> Message-Id: <20231129060211.1890454-2-irogers@google.com> Precedence: bulk X-Mailing-List: bpf@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 References: <20231129060211.1890454-1-irogers@google.com> X-Mailer: git-send-email 2.43.0.rc1.413.gea7ed67945-goog Subject: [PATCH v1 01/14] libperf cpumap: Rename perf_cpu_map__dummy_new From: Ian Rogers To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Mark Rutland , Alexander Shishkin , Jiri Olsa , Namhyung Kim , Ian Rogers , Adrian Hunter , Suzuki K Poulose , Mike Leach , James Clark , Leo Yan , John Garry , Will Deacon , Thomas Gleixner , Darren Hart , Davidlohr Bueso , " =?utf-8?q?Andr=C3=A9_Almeida?= " , Kan Liang , K Prateek Nayak , Sean Christopherson , Paolo Bonzini , Kajol Jain , Athira Rajeev , Andrew Jones , Alexandre Ghiti , Atish Patra , "Steinar H. Gunderson" , Yang Jihong , Yang Li , Changbin Du , Sandipan Das , Ravi Bangoria , Paran Lee , Nick Desaulniers , Huacai Chen , Yanteng Si , linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org, coresight@lists.linaro.org, linux-arm-kernel@lists.infradead.org, bpf@vger.kernel.org Rename perf_cpu_map__dummy_new to perf_cpu_map__new_any_cpu to better indicate this is creating a CPU map for the perf_event_open "any" CPU case. Signed-off-by: Ian Rogers Reviewed-by: James Clark --- tools/lib/perf/Documentation/libperf.txt | 2 +- tools/lib/perf/cpumap.c | 4 ++-- tools/lib/perf/evsel.c | 2 +- tools/lib/perf/include/perf/cpumap.h | 4 ++-- tools/lib/perf/libperf.map | 2 +- tools/lib/perf/tests/test-cpumap.c | 2 +- tools/lib/perf/tests/test-evlist.c | 2 +- tools/perf/tests/cpumap.c | 2 +- tools/perf/tests/sw-clock.c | 2 +- tools/perf/tests/task-exit.c | 2 +- tools/perf/util/evlist.c | 2 +- tools/perf/util/evsel.c | 2 +- 12 files changed, 14 insertions(+), 14 deletions(-) diff --git a/tools/lib/perf/Documentation/libperf.txt b/tools/lib/perf/Documentation/libperf.txt index a8f1a237931b..a256a26598b0 100644 --- a/tools/lib/perf/Documentation/libperf.txt +++ b/tools/lib/perf/Documentation/libperf.txt @@ -37,7 +37,7 @@ SYNOPSIS struct perf_cpu_map; - struct perf_cpu_map *perf_cpu_map__dummy_new(void); + struct perf_cpu_map *perf_cpu_map__new_any_cpu(void); struct perf_cpu_map *perf_cpu_map__new(const char *cpu_list); struct perf_cpu_map *perf_cpu_map__read(FILE *file); struct perf_cpu_map *perf_cpu_map__get(struct perf_cpu_map *map); diff --git a/tools/lib/perf/cpumap.c b/tools/lib/perf/cpumap.c index 2a5a29217374..2bd6aba3d8c9 100644 --- a/tools/lib/perf/cpumap.c +++ b/tools/lib/perf/cpumap.c @@ -27,7 +27,7 @@ struct perf_cpu_map *perf_cpu_map__alloc(int nr_cpus) return result; } -struct perf_cpu_map *perf_cpu_map__dummy_new(void) +struct perf_cpu_map *perf_cpu_map__new_any_cpu(void) { struct perf_cpu_map *cpus = perf_cpu_map__alloc(1); @@ -271,7 +271,7 @@ struct perf_cpu_map *perf_cpu_map__new(const char *cpu_list) else if (*cpu_list != '\0') cpus = cpu_map__default_new(); else - cpus = perf_cpu_map__dummy_new(); + cpus = perf_cpu_map__new_any_cpu(); invalid: free(tmp_cpus); out: diff --git a/tools/lib/perf/evsel.c b/tools/lib/perf/evsel.c index 8b51b008a81f..c07160953224 100644 --- a/tools/lib/perf/evsel.c +++ b/tools/lib/perf/evsel.c @@ -120,7 +120,7 @@ int perf_evsel__open(struct perf_evsel *evsel, struct perf_cpu_map *cpus, static struct perf_cpu_map *empty_cpu_map; if (empty_cpu_map == NULL) { - empty_cpu_map = perf_cpu_map__dummy_new(); + empty_cpu_map = perf_cpu_map__new_any_cpu(); if (empty_cpu_map == NULL) return -ENOMEM; } diff --git a/tools/lib/perf/include/perf/cpumap.h b/tools/lib/perf/include/perf/cpumap.h index e38d859a384d..d0bf218ada11 100644 --- a/tools/lib/perf/include/perf/cpumap.h +++ b/tools/lib/perf/include/perf/cpumap.h @@ -19,9 +19,9 @@ struct perf_cache { struct perf_cpu_map; /** - * perf_cpu_map__dummy_new - a map with a singular "any CPU"/dummy -1 value. + * perf_cpu_map__new_any_cpu - a map with a singular "any CPU"/dummy -1 value. */ -LIBPERF_API struct perf_cpu_map *perf_cpu_map__dummy_new(void); +LIBPERF_API struct perf_cpu_map *perf_cpu_map__new_any_cpu(void); LIBPERF_API struct perf_cpu_map *perf_cpu_map__default_new(void); LIBPERF_API struct perf_cpu_map *perf_cpu_map__new(const char *cpu_list); LIBPERF_API struct perf_cpu_map *perf_cpu_map__read(FILE *file); diff --git a/tools/lib/perf/libperf.map b/tools/lib/perf/libperf.map index 190b56ae923a..a8ff64baea3e 100644 --- a/tools/lib/perf/libperf.map +++ b/tools/lib/perf/libperf.map @@ -1,7 +1,7 @@ LIBPERF_0.0.1 { global: libperf_init; - perf_cpu_map__dummy_new; + perf_cpu_map__new_any_cpu; perf_cpu_map__default_new; perf_cpu_map__get; perf_cpu_map__put; diff --git a/tools/lib/perf/tests/test-cpumap.c b/tools/lib/perf/tests/test-cpumap.c index 87b0510a556f..2c359bdb951e 100644 --- a/tools/lib/perf/tests/test-cpumap.c +++ b/tools/lib/perf/tests/test-cpumap.c @@ -21,7 +21,7 @@ int test_cpumap(int argc, char **argv) libperf_init(libperf_print); - cpus = perf_cpu_map__dummy_new(); + cpus = perf_cpu_map__new_any_cpu(); if (!cpus) return -1; diff --git a/tools/lib/perf/tests/test-evlist.c b/tools/lib/perf/tests/test-evlist.c index ed616fc19b4f..ab63878bacb9 100644 --- a/tools/lib/perf/tests/test-evlist.c +++ b/tools/lib/perf/tests/test-evlist.c @@ -261,7 +261,7 @@ static int test_mmap_thread(void) threads = perf_thread_map__new_dummy(); __T("failed to create threads", threads); - cpus = perf_cpu_map__dummy_new(); + cpus = perf_cpu_map__new_any_cpu(); __T("failed to create cpus", cpus); perf_thread_map__set_pid(threads, 0, pid); diff --git a/tools/perf/tests/cpumap.c b/tools/perf/tests/cpumap.c index 7730fc2ab40b..bd8e396f3e57 100644 --- a/tools/perf/tests/cpumap.c +++ b/tools/perf/tests/cpumap.c @@ -213,7 +213,7 @@ static int test__cpu_map_intersect(struct test_suite *test __maybe_unused, static int test__cpu_map_equal(struct test_suite *test __maybe_unused, int subtest __maybe_unused) { - struct perf_cpu_map *any = perf_cpu_map__dummy_new(); + struct perf_cpu_map *any = perf_cpu_map__new_any_cpu(); struct perf_cpu_map *one = perf_cpu_map__new("1"); struct perf_cpu_map *two = perf_cpu_map__new("2"); struct perf_cpu_map *empty = perf_cpu_map__intersect(one, two); diff --git a/tools/perf/tests/sw-clock.c b/tools/perf/tests/sw-clock.c index 4d7493fa0105..290716783ac6 100644 --- a/tools/perf/tests/sw-clock.c +++ b/tools/perf/tests/sw-clock.c @@ -62,7 +62,7 @@ static int __test__sw_clock_freq(enum perf_sw_ids clock_id) } evlist__add(evlist, evsel); - cpus = perf_cpu_map__dummy_new(); + cpus = perf_cpu_map__new_any_cpu(); threads = thread_map__new_by_tid(getpid()); if (!cpus || !threads) { err = -ENOMEM; diff --git a/tools/perf/tests/task-exit.c b/tools/perf/tests/task-exit.c index 968dddde6dda..d33d0952025c 100644 --- a/tools/perf/tests/task-exit.c +++ b/tools/perf/tests/task-exit.c @@ -70,7 +70,7 @@ static int test__task_exit(struct test_suite *test __maybe_unused, int subtest _ * evlist__prepare_workload we'll fill in the only thread * we're monitoring, the one forked there. */ - cpus = perf_cpu_map__dummy_new(); + cpus = perf_cpu_map__new_any_cpu(); threads = thread_map__new_by_tid(-1); if (!cpus || !threads) { err = -ENOMEM; diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c index e36da58522ef..ff7f85ded89d 100644 --- a/tools/perf/util/evlist.c +++ b/tools/perf/util/evlist.c @@ -1056,7 +1056,7 @@ int evlist__create_maps(struct evlist *evlist, struct target *target) return -1; if (target__uses_dummy_map(target)) - cpus = perf_cpu_map__dummy_new(); + cpus = perf_cpu_map__new_any_cpu(); else cpus = perf_cpu_map__new(target->cpu_list); diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c index a5da74e3a517..76ef3ab488a2 100644 --- a/tools/perf/util/evsel.c +++ b/tools/perf/util/evsel.c @@ -1801,7 +1801,7 @@ static int __evsel__prepare_open(struct evsel *evsel, struct perf_cpu_map *cpus, if (cpus == NULL) { if (empty_cpu_map == NULL) { - empty_cpu_map = perf_cpu_map__dummy_new(); + empty_cpu_map = perf_cpu_map__new_any_cpu(); if (empty_cpu_map == NULL) return -ENOMEM; } From patchwork Wed Nov 29 06:01:59 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Ian Rogers X-Patchwork-Id: 13472216 X-Patchwork-Delegate: bpf@iogearbox.net Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=google.com header.i=@google.com header.b="uspoxPX8" Received: from mail-yb1-xb49.google.com (mail-yb1-xb49.google.com [IPv6:2607:f8b0:4864:20::b49]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 72EC81BC0 for ; Tue, 28 Nov 2023 22:02:20 -0800 (PST) Received: by mail-yb1-xb49.google.com with SMTP id 3f1490d57ef6-db4004a8aa9so7330775276.1 for ; Tue, 28 Nov 2023 22:02:20 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1701237739; x=1701842539; darn=vger.kernel.org; h=to:from:subject:references:mime-version:message-id:in-reply-to:date :from:to:cc:subject:date:message-id:reply-to; bh=T8ykSaAbLFyuzcPYepQRuQ5xp7FIWrmaxmvAYiTpoEM=; b=uspoxPX8E+ukWpbpDaIPltcSuLuYXkZjkyQAFnD+ZHlccd93ygNbhE26g8N6A4gGZz trPWszcg5+fakGMKJ+JFfhMMI4CN3LnxyRb/fYgJNWnTFEmTjgUcg0G+GprK510Yy3Xd 08AhJLv98drog7NjeiFW1WzE4hZeOG8BW1Y5C8ski6VCPzKnnBdjclcklHDBFqMTh43s zsorQtJWI5V+yphwovihsaN6UUnQOypJmvwifGk7w90+kKsc7xaK8pDO8RDweob5FO3X jVH8kgHR8Zzeb0YWDJpcaVTopuDEraS6pKdliysP9MenV4fiimta7aabEoSVfldTzCcq Kneg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1701237739; x=1701842539; h=to:from:subject:references:mime-version:message-id:in-reply-to:date :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=T8ykSaAbLFyuzcPYepQRuQ5xp7FIWrmaxmvAYiTpoEM=; b=KVS77OrUKNKZoRds1csTjUm0o2STk2q5XETdRws1IVWcJv5HajOdgLhz5qXb5hrc7G C3sAstVPleXUcRs8tGBtke8EgqqS/G7NlIYD1mlXKZBygsa4fBW9qjJWnpy+QjUW88WT IxWZ8Vb1eOtUKkP2PfAp7xNJ3pFJC610nmK/pAN55dwlK556tYlVIHCOjK3TAv5+X+3C 0sKWevu3QtVddQbOUBg6/swaEZ/zeTn6dWQpgmcp91JMobZuNm4/nglCmAIbOYSAy4wC hgYleSktGUc7SwVkHUY113fu0yLWM8JFD/AZz6PKtRFK+CPCoZepmpUScxs5PetmWPVB Q5OA== X-Gm-Message-State: AOJu0Yz+XvN5VYqqOoAYeGQO3T1J8lw+dxyha8O2zdMBIzRyTFFezdjh dTFeGjAMXW6a/ySSDGPWExmNYk5V+Tqt X-Google-Smtp-Source: AGHT+IGpkfJM/vGDzeJGawlB8gonUbNHatIumaAJ9ml4Cx9YmaCI6/1HJPOtb7yX5CQXz6dt/8yBM1CTLG9d X-Received: from irogers.svl.corp.google.com ([2620:15c:2a3:200:763b:80fa:23ca:96f8]) (user=irogers job=sendgmr) by 2002:a25:ca50:0:b0:daf:5d6b:ef39 with SMTP id a77-20020a25ca50000000b00daf5d6bef39mr547103ybg.12.1701237739607; Tue, 28 Nov 2023 22:02:19 -0800 (PST) Date: Tue, 28 Nov 2023 22:01:59 -0800 In-Reply-To: <20231129060211.1890454-1-irogers@google.com> Message-Id: <20231129060211.1890454-3-irogers@google.com> Precedence: bulk X-Mailing-List: bpf@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 References: <20231129060211.1890454-1-irogers@google.com> X-Mailer: git-send-email 2.43.0.rc1.413.gea7ed67945-goog Subject: [PATCH v1 02/14] libperf cpumap: Rename and prefer sysfs for perf_cpu_map__default_new From: Ian Rogers To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Mark Rutland , Alexander Shishkin , Jiri Olsa , Namhyung Kim , Ian Rogers , Adrian Hunter , Suzuki K Poulose , Mike Leach , James Clark , Leo Yan , John Garry , Will Deacon , Thomas Gleixner , Darren Hart , Davidlohr Bueso , " =?utf-8?q?Andr=C3=A9_Almeida?= " , Kan Liang , K Prateek Nayak , Sean Christopherson , Paolo Bonzini , Kajol Jain , Athira Rajeev , Andrew Jones , Alexandre Ghiti , Atish Patra , "Steinar H. Gunderson" , Yang Jihong , Yang Li , Changbin Du , Sandipan Das , Ravi Bangoria , Paran Lee , Nick Desaulniers , Huacai Chen , Yanteng Si , linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org, coresight@lists.linaro.org, linux-arm-kernel@lists.infradead.org, bpf@vger.kernel.org Rename perf_cpu_map__default_new to perf_cpu_map__new_online_cpus to better indicate what the implementation does. Read the online CPUs from /sys/devices/system/cpu/online first before using sysconf as sysconf can't accurately configure holes in the CPU map. If sysconf is used, warn when the configured and online processors disagree. When reading from a file, if the read doesn't yield a CPU map then return an empty map rather than the default online. This avoids recursion but also better yields being able to detect failures. Add more comments. Signed-off-by: Ian Rogers Reviewed-by: James Clark --- tools/lib/perf/cpumap.c | 59 +++++++++++++++++----------- tools/lib/perf/include/perf/cpumap.h | 15 ++++++- tools/lib/perf/libperf.map | 2 +- tools/lib/perf/tests/test-cpumap.c | 2 +- 4 files changed, 51 insertions(+), 27 deletions(-) diff --git a/tools/lib/perf/cpumap.c b/tools/lib/perf/cpumap.c index 2bd6aba3d8c9..463ca8b37d45 100644 --- a/tools/lib/perf/cpumap.c +++ b/tools/lib/perf/cpumap.c @@ -9,6 +9,7 @@ #include #include #include +#include "internal.h" void perf_cpu_map__set_nr(struct perf_cpu_map *map, int nr_cpus) { @@ -66,15 +67,21 @@ void perf_cpu_map__put(struct perf_cpu_map *map) } } -static struct perf_cpu_map *cpu_map__default_new(void) +static struct perf_cpu_map *cpu_map__new_sysconf(void) { struct perf_cpu_map *cpus; - int nr_cpus; + int nr_cpus, nr_cpus_conf; nr_cpus = sysconf(_SC_NPROCESSORS_ONLN); if (nr_cpus < 0) return NULL; + nr_cpus_conf = sysconf(_SC_NPROCESSORS_CONF); + if (nr_cpus != nr_cpus_conf) { + pr_warning("Number of online CPUs (%d) differs from the number configured (%d) the CPU map will only cover the first %d CPUs.", + nr_cpus, nr_cpus_conf, nr_cpus); + } + cpus = perf_cpu_map__alloc(nr_cpus); if (cpus != NULL) { int i; @@ -86,9 +93,27 @@ static struct perf_cpu_map *cpu_map__default_new(void) return cpus; } -struct perf_cpu_map *perf_cpu_map__default_new(void) +static struct perf_cpu_map *cpu_map__new_syfs_online(void) { - return cpu_map__default_new(); + struct perf_cpu_map *cpus = NULL; + FILE *onlnf; + + onlnf = fopen("/sys/devices/system/cpu/online", "r"); + if (onlnf) { + cpus = perf_cpu_map__read(onlnf); + fclose(onlnf); + } + return cpus; +} + +struct perf_cpu_map *perf_cpu_map__new_online_cpus(void) +{ + struct perf_cpu_map *cpus = cpu_map__new_syfs_online(); + + if (cpus) + return cpus; + + return cpu_map__new_sysconf(); } @@ -180,27 +205,11 @@ struct perf_cpu_map *perf_cpu_map__read(FILE *file) if (nr_cpus > 0) cpus = cpu_map__trim_new(nr_cpus, tmp_cpus); - else - cpus = cpu_map__default_new(); out_free_tmp: free(tmp_cpus); return cpus; } -static struct perf_cpu_map *cpu_map__read_all_cpu_map(void) -{ - struct perf_cpu_map *cpus = NULL; - FILE *onlnf; - - onlnf = fopen("/sys/devices/system/cpu/online", "r"); - if (!onlnf) - return cpu_map__default_new(); - - cpus = perf_cpu_map__read(onlnf); - fclose(onlnf); - return cpus; -} - struct perf_cpu_map *perf_cpu_map__new(const char *cpu_list) { struct perf_cpu_map *cpus = NULL; @@ -211,7 +220,7 @@ struct perf_cpu_map *perf_cpu_map__new(const char *cpu_list) int max_entries = 0; if (!cpu_list) - return cpu_map__read_all_cpu_map(); + return perf_cpu_map__new_online_cpus(); /* * must handle the case of empty cpumap to cover @@ -268,9 +277,11 @@ struct perf_cpu_map *perf_cpu_map__new(const char *cpu_list) if (nr_cpus > 0) cpus = cpu_map__trim_new(nr_cpus, tmp_cpus); - else if (*cpu_list != '\0') - cpus = cpu_map__default_new(); - else + else if (*cpu_list != '\0') { + pr_warning("Unexpected characters at end of cpu list ('%s'), using online CPUs.", + cpu_list); + cpus = perf_cpu_map__new_online_cpus(); + } else cpus = perf_cpu_map__new_any_cpu(); invalid: free(tmp_cpus); diff --git a/tools/lib/perf/include/perf/cpumap.h b/tools/lib/perf/include/perf/cpumap.h index d0bf218ada11..b24bd8b8f34e 100644 --- a/tools/lib/perf/include/perf/cpumap.h +++ b/tools/lib/perf/include/perf/cpumap.h @@ -22,7 +22,20 @@ struct perf_cpu_map; * perf_cpu_map__new_any_cpu - a map with a singular "any CPU"/dummy -1 value. */ LIBPERF_API struct perf_cpu_map *perf_cpu_map__new_any_cpu(void); -LIBPERF_API struct perf_cpu_map *perf_cpu_map__default_new(void); +/** + * perf_cpu_map__new_online_cpus - a map read from + * /sys/devices/system/cpu/online if + * available. If reading wasn't possible a map + * is created using the online processors + * assuming the first 'n' processors are all + * online. + */ +LIBPERF_API struct perf_cpu_map *perf_cpu_map__new_online_cpus(void); +/** + * perf_cpu_map__new - create a map from the given cpu_list such as "0-7". If no + * cpu_list argument is provided then + * perf_cpu_map__new_online_cpus is returned. + */ LIBPERF_API struct perf_cpu_map *perf_cpu_map__new(const char *cpu_list); LIBPERF_API struct perf_cpu_map *perf_cpu_map__read(FILE *file); LIBPERF_API struct perf_cpu_map *perf_cpu_map__get(struct perf_cpu_map *map); diff --git a/tools/lib/perf/libperf.map b/tools/lib/perf/libperf.map index a8ff64baea3e..8a71f841498e 100644 --- a/tools/lib/perf/libperf.map +++ b/tools/lib/perf/libperf.map @@ -2,7 +2,7 @@ LIBPERF_0.0.1 { global: libperf_init; perf_cpu_map__new_any_cpu; - perf_cpu_map__default_new; + perf_cpu_map__new_online_cpus; perf_cpu_map__get; perf_cpu_map__put; perf_cpu_map__new; diff --git a/tools/lib/perf/tests/test-cpumap.c b/tools/lib/perf/tests/test-cpumap.c index 2c359bdb951e..c998b1dae863 100644 --- a/tools/lib/perf/tests/test-cpumap.c +++ b/tools/lib/perf/tests/test-cpumap.c @@ -29,7 +29,7 @@ int test_cpumap(int argc, char **argv) perf_cpu_map__put(cpus); perf_cpu_map__put(cpus); - cpus = perf_cpu_map__default_new(); + cpus = perf_cpu_map__new_online_cpus(); if (!cpus) return -1; From patchwork Wed Nov 29 06:02:00 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Ian Rogers X-Patchwork-Id: 13472223 X-Patchwork-Delegate: bpf@iogearbox.net Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=google.com header.i=@google.com header.b="cBS8cH2k" Received: from mail-yw1-x114a.google.com (mail-yw1-x114a.google.com [IPv6:2607:f8b0:4864:20::114a]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id E3CFD1BCD for ; Tue, 28 Nov 2023 22:02:22 -0800 (PST) Received: by mail-yw1-x114a.google.com with SMTP id 00721157ae682-5cc3e334d16so101391367b3.3 for ; Tue, 28 Nov 2023 22:02:22 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1701237742; x=1701842542; darn=vger.kernel.org; h=to:from:subject:references:mime-version:message-id:in-reply-to:date :from:to:cc:subject:date:message-id:reply-to; bh=rZsrGWSv6swOR+t3Iv27SAPHcTaNTC2gPF1xFShNf5I=; b=cBS8cH2kO4WOshBtnY+Y4tVy0M9agxclMBmnsQ1xQwTKu3SfDOmmqvcSux1qJYzKlj sGmV7ggj4CDxnlw3pWmPoNAlTjeSoB8xnLHVPYS4QHFec1VjBZ089OygvsBBAEHLtIQ7 SimJ6vwDsAgQ8OlgpCw1fQuqg8xIyF1s17EzsCvmahSsavelpatFqGTXrlkC4c+x1eG/ qdx0wq3UP4L8Kn/0Gz5cDRZocu7U++weEdBTh1r4HDbVLd1mvfTxvy2AO2y5QMXDEyiv Zv/HzbFOgaWMf+Qbmx538DcUFmZgitSbwDyOGpMhHkf3f6E6mlV3LRnEiUapSbB4QG10 xccQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1701237742; x=1701842542; h=to:from:subject:references:mime-version:message-id:in-reply-to:date :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=rZsrGWSv6swOR+t3Iv27SAPHcTaNTC2gPF1xFShNf5I=; b=RunNPSZODM4xDbqNVONXOqbC3V5ssCEyY+lGNS5tqVGTuqKFzJu0UVWRmuaVI2fC37 /FMaUhRhYWytC5kj/41xKoRnelnLctORnfenRvSl5+YCMyTMGVYfoH1L156UrXNX1hsR mdNbMHgdhdILdrk2pnGwIDUvpNeJUtGUU8FAwZwGoEwKavTxAjP/rWtqoWowODHujH6P Xjc73SI6Gs5oY9sIYi2WsK5asAK5gxniQcR83T1PRSjH2mLp4s41eaWwG5gY2fthX67A FIa05aJ+6VRAz3bDLJo7Eh5XJ5rBVGNUD7y0s/cn5j7MsyZDVUURg+7/cYJwZtnV0EdX oYXw== X-Gm-Message-State: AOJu0Yx8/edgDv6Xtxk3ipPe/BM42ly3P9lh7l4BrmypvIMirQZeFiZg YTNi9cbY+62rsIbjkfLDoceA+iQlh1ns X-Google-Smtp-Source: AGHT+IF/EFntS8k3BTSYeBHJpiZIVPzHJV/7srchCsdGMVKGukm38qeKP6ku7fSxxYCx3fcKdX/lv2DcQRM1 X-Received: from irogers.svl.corp.google.com ([2620:15c:2a3:200:763b:80fa:23ca:96f8]) (user=irogers job=sendgmr) by 2002:a05:690c:98d:b0:5cd:c47d:d8a0 with SMTP id ce13-20020a05690c098d00b005cdc47dd8a0mr560233ywb.7.1701237742052; Tue, 28 Nov 2023 22:02:22 -0800 (PST) Date: Tue, 28 Nov 2023 22:02:00 -0800 In-Reply-To: <20231129060211.1890454-1-irogers@google.com> Message-Id: <20231129060211.1890454-4-irogers@google.com> Precedence: bulk X-Mailing-List: bpf@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 References: <20231129060211.1890454-1-irogers@google.com> X-Mailer: git-send-email 2.43.0.rc1.413.gea7ed67945-goog Subject: [PATCH v1 03/14] libperf cpumap: Rename perf_cpu_map__empty From: Ian Rogers To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Mark Rutland , Alexander Shishkin , Jiri Olsa , Namhyung Kim , Ian Rogers , Adrian Hunter , Suzuki K Poulose , Mike Leach , James Clark , Leo Yan , John Garry , Will Deacon , Thomas Gleixner , Darren Hart , Davidlohr Bueso , " =?utf-8?q?Andr=C3=A9_Almeida?= " , Kan Liang , K Prateek Nayak , Sean Christopherson , Paolo Bonzini , Kajol Jain , Athira Rajeev , Andrew Jones , Alexandre Ghiti , Atish Patra , "Steinar H. Gunderson" , Yang Jihong , Yang Li , Changbin Du , Sandipan Das , Ravi Bangoria , Paran Lee , Nick Desaulniers , Huacai Chen , Yanteng Si , linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org, coresight@lists.linaro.org, linux-arm-kernel@lists.infradead.org, bpf@vger.kernel.org The name perf_cpu_map_empty is misleading as true is also returned when the map contains an "any" CPU (aka dummy) map. Rename to perf_cpu_map__has_any_cpu_or_is_empty, later changes will (re)introduce perf_cpu_map__empty and perf_cpu_map__has_any_cpu. Signed-off-by: Ian Rogers Reviewed-by: James Clark --- tools/lib/perf/Documentation/libperf.txt | 2 +- tools/lib/perf/cpumap.c | 2 +- tools/lib/perf/evlist.c | 4 ++-- tools/lib/perf/include/perf/cpumap.h | 4 ++-- tools/lib/perf/libperf.map | 2 +- tools/perf/arch/arm/util/cs-etm.c | 10 +++++----- tools/perf/arch/arm64/util/arm-spe.c | 4 ++-- tools/perf/arch/x86/util/intel-bts.c | 4 ++-- tools/perf/arch/x86/util/intel-pt.c | 10 +++++----- tools/perf/builtin-c2c.c | 2 +- tools/perf/builtin-stat.c | 6 +++--- tools/perf/util/auxtrace.c | 4 ++-- tools/perf/util/record.c | 2 +- tools/perf/util/stat.c | 2 +- 14 files changed, 29 insertions(+), 29 deletions(-) diff --git a/tools/lib/perf/Documentation/libperf.txt b/tools/lib/perf/Documentation/libperf.txt index a256a26598b0..fcfb9499ef9c 100644 --- a/tools/lib/perf/Documentation/libperf.txt +++ b/tools/lib/perf/Documentation/libperf.txt @@ -46,7 +46,7 @@ SYNOPSIS void perf_cpu_map__put(struct perf_cpu_map *map); int perf_cpu_map__cpu(const struct perf_cpu_map *cpus, int idx); int perf_cpu_map__nr(const struct perf_cpu_map *cpus); - bool perf_cpu_map__empty(const struct perf_cpu_map *map); + bool perf_cpu_map__has_any_cpu_or_is_empty(const struct perf_cpu_map *map); int perf_cpu_map__max(struct perf_cpu_map *map); bool perf_cpu_map__has(const struct perf_cpu_map *map, int cpu); diff --git a/tools/lib/perf/cpumap.c b/tools/lib/perf/cpumap.c index 463ca8b37d45..49fc98e16514 100644 --- a/tools/lib/perf/cpumap.c +++ b/tools/lib/perf/cpumap.c @@ -311,7 +311,7 @@ int perf_cpu_map__nr(const struct perf_cpu_map *cpus) return cpus ? __perf_cpu_map__nr(cpus) : 1; } -bool perf_cpu_map__empty(const struct perf_cpu_map *map) +bool perf_cpu_map__has_any_cpu_or_is_empty(const struct perf_cpu_map *map) { return map ? __perf_cpu_map__cpu(map, 0).cpu == -1 : true; } diff --git a/tools/lib/perf/evlist.c b/tools/lib/perf/evlist.c index 3acbbccc1901..75f36218fdd9 100644 --- a/tools/lib/perf/evlist.c +++ b/tools/lib/perf/evlist.c @@ -619,7 +619,7 @@ static int perf_evlist__nr_mmaps(struct perf_evlist *evlist) /* One for each CPU */ nr_mmaps = perf_cpu_map__nr(evlist->all_cpus); - if (perf_cpu_map__empty(evlist->all_cpus)) { + if (perf_cpu_map__has_any_cpu_or_is_empty(evlist->all_cpus)) { /* Plus one for each thread */ nr_mmaps += perf_thread_map__nr(evlist->threads); /* Minus the per-thread CPU (-1) */ @@ -653,7 +653,7 @@ int perf_evlist__mmap_ops(struct perf_evlist *evlist, if (evlist->pollfd.entries == NULL && perf_evlist__alloc_pollfd(evlist) < 0) return -ENOMEM; - if (perf_cpu_map__empty(cpus)) + if (perf_cpu_map__has_any_cpu_or_is_empty(cpus)) return mmap_per_thread(evlist, ops, mp); return mmap_per_cpu(evlist, ops, mp); diff --git a/tools/lib/perf/include/perf/cpumap.h b/tools/lib/perf/include/perf/cpumap.h index b24bd8b8f34e..9cf361fc5edc 100644 --- a/tools/lib/perf/include/perf/cpumap.h +++ b/tools/lib/perf/include/perf/cpumap.h @@ -47,9 +47,9 @@ LIBPERF_API void perf_cpu_map__put(struct perf_cpu_map *map); LIBPERF_API struct perf_cpu perf_cpu_map__cpu(const struct perf_cpu_map *cpus, int idx); LIBPERF_API int perf_cpu_map__nr(const struct perf_cpu_map *cpus); /** - * perf_cpu_map__empty - is map either empty or the "any CPU"/dummy value. + * perf_cpu_map__has_any_cpu_or_is_empty - is map either empty or has the "any CPU"/dummy value. */ -LIBPERF_API bool perf_cpu_map__empty(const struct perf_cpu_map *map); +LIBPERF_API bool perf_cpu_map__has_any_cpu_or_is_empty(const struct perf_cpu_map *map); LIBPERF_API struct perf_cpu perf_cpu_map__max(const struct perf_cpu_map *map); LIBPERF_API bool perf_cpu_map__has(const struct perf_cpu_map *map, struct perf_cpu cpu); LIBPERF_API bool perf_cpu_map__equal(const struct perf_cpu_map *lhs, diff --git a/tools/lib/perf/libperf.map b/tools/lib/perf/libperf.map index 8a71f841498e..10b3f3722642 100644 --- a/tools/lib/perf/libperf.map +++ b/tools/lib/perf/libperf.map @@ -9,7 +9,7 @@ LIBPERF_0.0.1 { perf_cpu_map__read; perf_cpu_map__nr; perf_cpu_map__cpu; - perf_cpu_map__empty; + perf_cpu_map__has_any_cpu_or_is_empty; perf_cpu_map__max; perf_cpu_map__has; perf_thread_map__new_array; diff --git a/tools/perf/arch/arm/util/cs-etm.c b/tools/perf/arch/arm/util/cs-etm.c index 2cf873d71dff..c6b7b3066324 100644 --- a/tools/perf/arch/arm/util/cs-etm.c +++ b/tools/perf/arch/arm/util/cs-etm.c @@ -211,7 +211,7 @@ static int cs_etm_validate_config(struct auxtrace_record *itr, * program can run on any CPUs in this case, thus don't skip * validation. */ - if (!perf_cpu_map__empty(event_cpus) && + if (!perf_cpu_map__has_any_cpu_or_is_empty(event_cpus) && !perf_cpu_map__has(event_cpus, cpu)) continue; @@ -435,7 +435,7 @@ static int cs_etm_recording_options(struct auxtrace_record *itr, * Also the case of per-cpu mmaps, need the contextID in order to be notified * when a context switch happened. */ - if (!perf_cpu_map__empty(cpus)) { + if (!perf_cpu_map__has_any_cpu_or_is_empty(cpus)) { evsel__set_config_if_unset(cs_etm_pmu, cs_etm_evsel, "timestamp", 1); evsel__set_config_if_unset(cs_etm_pmu, cs_etm_evsel, @@ -461,7 +461,7 @@ static int cs_etm_recording_options(struct auxtrace_record *itr, evsel->core.attr.sample_period = 1; /* In per-cpu case, always need the time of mmap events etc */ - if (!perf_cpu_map__empty(cpus)) + if (!perf_cpu_map__has_any_cpu_or_is_empty(cpus)) evsel__set_sample_bit(evsel, TIME); err = cs_etm_validate_config(itr, cs_etm_evsel); @@ -539,7 +539,7 @@ cs_etm_info_priv_size(struct auxtrace_record *itr __maybe_unused, struct perf_cpu_map *online_cpus = perf_cpu_map__new(NULL); /* cpu map is not empty, we have specific CPUs to work with */ - if (!perf_cpu_map__empty(event_cpus)) { + if (!perf_cpu_map__has_any_cpu_or_is_empty(event_cpus)) { for (i = 0; i < cpu__max_cpu().cpu; i++) { struct perf_cpu cpu = { .cpu = i, }; @@ -814,7 +814,7 @@ static int cs_etm_info_fill(struct auxtrace_record *itr, return -EINVAL; /* If the cpu_map is empty all online CPUs are involved */ - if (perf_cpu_map__empty(event_cpus)) { + if (perf_cpu_map__has_any_cpu_or_is_empty(event_cpus)) { cpu_map = online_cpus; } else { /* Make sure all specified CPUs are online */ diff --git a/tools/perf/arch/arm64/util/arm-spe.c b/tools/perf/arch/arm64/util/arm-spe.c index e3acc739bd00..51ccbfd3d246 100644 --- a/tools/perf/arch/arm64/util/arm-spe.c +++ b/tools/perf/arch/arm64/util/arm-spe.c @@ -232,7 +232,7 @@ static int arm_spe_recording_options(struct auxtrace_record *itr, * In the case of per-cpu mmaps, sample CPU for AUX event; * also enable the timestamp tracing for samples correlation. */ - if (!perf_cpu_map__empty(cpus)) { + if (!perf_cpu_map__has_any_cpu_or_is_empty(cpus)) { evsel__set_sample_bit(arm_spe_evsel, CPU); evsel__set_config_if_unset(arm_spe_pmu, arm_spe_evsel, "ts_enable", 1); @@ -265,7 +265,7 @@ static int arm_spe_recording_options(struct auxtrace_record *itr, tracking_evsel->core.attr.sample_period = 1; /* In per-cpu case, always need the time of mmap events etc */ - if (!perf_cpu_map__empty(cpus)) { + if (!perf_cpu_map__has_any_cpu_or_is_empty(cpus)) { evsel__set_sample_bit(tracking_evsel, TIME); evsel__set_sample_bit(tracking_evsel, CPU); diff --git a/tools/perf/arch/x86/util/intel-bts.c b/tools/perf/arch/x86/util/intel-bts.c index d2c8cac11470..af8ae4647585 100644 --- a/tools/perf/arch/x86/util/intel-bts.c +++ b/tools/perf/arch/x86/util/intel-bts.c @@ -143,7 +143,7 @@ static int intel_bts_recording_options(struct auxtrace_record *itr, if (!opts->full_auxtrace) return 0; - if (opts->full_auxtrace && !perf_cpu_map__empty(cpus)) { + if (opts->full_auxtrace && !perf_cpu_map__has_any_cpu_or_is_empty(cpus)) { pr_err(INTEL_BTS_PMU_NAME " does not support per-cpu recording\n"); return -EINVAL; } @@ -224,7 +224,7 @@ static int intel_bts_recording_options(struct auxtrace_record *itr, * In the case of per-cpu mmaps, we need the CPU on the * AUX event. */ - if (!perf_cpu_map__empty(cpus)) + if (!perf_cpu_map__has_any_cpu_or_is_empty(cpus)) evsel__set_sample_bit(intel_bts_evsel, CPU); } diff --git a/tools/perf/arch/x86/util/intel-pt.c b/tools/perf/arch/x86/util/intel-pt.c index fa0c718b9e72..d199619df3ab 100644 --- a/tools/perf/arch/x86/util/intel-pt.c +++ b/tools/perf/arch/x86/util/intel-pt.c @@ -369,7 +369,7 @@ static int intel_pt_info_fill(struct auxtrace_record *itr, ui__warning("Intel Processor Trace: TSC not available\n"); } - per_cpu_mmaps = !perf_cpu_map__empty(session->evlist->core.user_requested_cpus); + per_cpu_mmaps = !perf_cpu_map__has_any_cpu_or_is_empty(session->evlist->core.user_requested_cpus); auxtrace_info->type = PERF_AUXTRACE_INTEL_PT; auxtrace_info->priv[INTEL_PT_PMU_TYPE] = intel_pt_pmu->type; @@ -774,7 +774,7 @@ static int intel_pt_recording_options(struct auxtrace_record *itr, * Per-cpu recording needs sched_switch events to distinguish different * threads. */ - if (have_timing_info && !perf_cpu_map__empty(cpus) && + if (have_timing_info && !perf_cpu_map__has_any_cpu_or_is_empty(cpus) && !record_opts__no_switch_events(opts)) { if (perf_can_record_switch_events()) { bool cpu_wide = !target__none(&opts->target) && @@ -832,7 +832,7 @@ static int intel_pt_recording_options(struct auxtrace_record *itr, * In the case of per-cpu mmaps, we need the CPU on the * AUX event. */ - if (!perf_cpu_map__empty(cpus)) + if (!perf_cpu_map__has_any_cpu_or_is_empty(cpus)) evsel__set_sample_bit(intel_pt_evsel, CPU); } @@ -858,7 +858,7 @@ static int intel_pt_recording_options(struct auxtrace_record *itr, tracking_evsel->immediate = true; /* In per-cpu case, always need the time of mmap events etc */ - if (!perf_cpu_map__empty(cpus)) { + if (!perf_cpu_map__has_any_cpu_or_is_empty(cpus)) { evsel__set_sample_bit(tracking_evsel, TIME); /* And the CPU for switch events */ evsel__set_sample_bit(tracking_evsel, CPU); @@ -870,7 +870,7 @@ static int intel_pt_recording_options(struct auxtrace_record *itr, * Warn the user when we do not have enough information to decode i.e. * per-cpu with no sched_switch (except workload-only). */ - if (!ptr->have_sched_switch && !perf_cpu_map__empty(cpus) && + if (!ptr->have_sched_switch && !perf_cpu_map__has_any_cpu_or_is_empty(cpus) && !target__none(&opts->target) && !intel_pt_evsel->core.attr.exclude_user) ui__warning("Intel Processor Trace decoding will not be possible except for kernel tracing!\n"); diff --git a/tools/perf/builtin-c2c.c b/tools/perf/builtin-c2c.c index a4cf9de7a7b5..f78eea9e2153 100644 --- a/tools/perf/builtin-c2c.c +++ b/tools/perf/builtin-c2c.c @@ -2320,7 +2320,7 @@ static int setup_nodes(struct perf_session *session) nodes[node] = set; /* empty node, skip */ - if (perf_cpu_map__empty(map)) + if (perf_cpu_map__has_any_cpu_or_is_empty(map)) continue; perf_cpu_map__for_each_cpu(cpu, idx, map) { diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c index a3af805a1d57..3303aa20f326 100644 --- a/tools/perf/builtin-stat.c +++ b/tools/perf/builtin-stat.c @@ -1316,7 +1316,7 @@ static int cpu__get_cache_id_from_map(struct perf_cpu cpu, char *map) * be the first online CPU in the cache domain else use the * first online CPU of the cache domain as the ID. */ - if (perf_cpu_map__empty(cpu_map)) + if (perf_cpu_map__has_any_cpu_or_is_empty(cpu_map)) id = cpu.cpu; else id = perf_cpu_map__cpu(cpu_map, 0).cpu; @@ -1622,7 +1622,7 @@ static int perf_stat_init_aggr_mode(void) * taking the highest cpu number to be the size of * the aggregation translate cpumap. */ - if (!perf_cpu_map__empty(evsel_list->core.user_requested_cpus)) + if (!perf_cpu_map__has_any_cpu_or_is_empty(evsel_list->core.user_requested_cpus)) nr = perf_cpu_map__max(evsel_list->core.user_requested_cpus).cpu; else nr = 0; @@ -2289,7 +2289,7 @@ int process_stat_config_event(struct perf_session *session, perf_event__read_stat_config(&stat_config, &event->stat_config); - if (perf_cpu_map__empty(st->cpus)) { + if (perf_cpu_map__has_any_cpu_or_is_empty(st->cpus)) { if (st->aggr_mode != AGGR_UNSET) pr_warning("warning: processing task data, aggregation mode not set\n"); } else if (st->aggr_mode != AGGR_UNSET) { diff --git a/tools/perf/util/auxtrace.c b/tools/perf/util/auxtrace.c index f528c4364d23..3684e6009b63 100644 --- a/tools/perf/util/auxtrace.c +++ b/tools/perf/util/auxtrace.c @@ -174,7 +174,7 @@ void auxtrace_mmap_params__set_idx(struct auxtrace_mmap_params *mp, struct evlist *evlist, struct evsel *evsel, int idx) { - bool per_cpu = !perf_cpu_map__empty(evlist->core.user_requested_cpus); + bool per_cpu = !perf_cpu_map__has_any_cpu_or_is_empty(evlist->core.user_requested_cpus); mp->mmap_needed = evsel->needs_auxtrace_mmap; @@ -648,7 +648,7 @@ int auxtrace_parse_snapshot_options(struct auxtrace_record *itr, static int evlist__enable_event_idx(struct evlist *evlist, struct evsel *evsel, int idx) { - bool per_cpu_mmaps = !perf_cpu_map__empty(evlist->core.user_requested_cpus); + bool per_cpu_mmaps = !perf_cpu_map__has_any_cpu_or_is_empty(evlist->core.user_requested_cpus); if (per_cpu_mmaps) { struct perf_cpu evlist_cpu = perf_cpu_map__cpu(evlist->core.all_cpus, idx); diff --git a/tools/perf/util/record.c b/tools/perf/util/record.c index 9eb5c6a08999..40290382b2d7 100644 --- a/tools/perf/util/record.c +++ b/tools/perf/util/record.c @@ -237,7 +237,7 @@ bool evlist__can_select_event(struct evlist *evlist, const char *str) evsel = evlist__last(temp_evlist); - if (!evlist || perf_cpu_map__empty(evlist->core.user_requested_cpus)) { + if (!evlist || perf_cpu_map__has_any_cpu_or_is_empty(evlist->core.user_requested_cpus)) { struct perf_cpu_map *cpus = perf_cpu_map__new(NULL); if (cpus) diff --git a/tools/perf/util/stat.c b/tools/perf/util/stat.c index ec3506042217..012c4946b9c4 100644 --- a/tools/perf/util/stat.c +++ b/tools/perf/util/stat.c @@ -315,7 +315,7 @@ static int check_per_pkg(struct evsel *counter, struct perf_counts_values *vals, if (!counter->per_pkg) return 0; - if (perf_cpu_map__empty(cpus)) + if (perf_cpu_map__has_any_cpu_or_is_empty(cpus)) return 0; if (!mask) { From patchwork Wed Nov 29 06:02:01 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Ian Rogers X-Patchwork-Id: 13472224 X-Patchwork-Delegate: bpf@iogearbox.net Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=google.com header.i=@google.com header.b="xyNO/x5Q" Received: from mail-yw1-x114a.google.com (mail-yw1-x114a.google.com [IPv6:2607:f8b0:4864:20::114a]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 8434619B7 for ; Tue, 28 Nov 2023 22:02:25 -0800 (PST) Received: by mail-yw1-x114a.google.com with SMTP id 00721157ae682-5d064f9e2a1so46932407b3.1 for ; Tue, 28 Nov 2023 22:02:25 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1701237744; x=1701842544; darn=vger.kernel.org; h=to:from:subject:references:mime-version:message-id:in-reply-to:date :from:to:cc:subject:date:message-id:reply-to; bh=iENnNsoCTzlwBGsMv7Mnpwt6IdDhoBtVXLX3WJvXL6w=; b=xyNO/x5Q5B55tPncfi/0HYpAAcWLCWs9QK9Z4JtmdEWi/FcdYnRLrSPAJILzgBmWRR GpK9IrUqAQF9N2TBxBaSJ1d676cJXuBcWfHNlihM/nXnC5YTD2NXitQ/h9t7+7Ocs4EJ LmUKgXB2z8DKxeRG/iFf1C8Cm89qDvCJd4OAzAzUjtJ2oTp/fKL/pQoa6+gmpS753Bma XYEj9iGkdoqygGX1HZLD3rXRG1wU/LeM95DsfHtnu6Ur/XnbiaVD8ZQisXt/WIcVDPdD qAbn4n6UM0kry3lhRiJROtmx/DpWmhrIr/44uMRTIdC3hrMSIxqcGp5T8aNmHOMQDU69 WiXQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1701237745; x=1701842545; h=to:from:subject:references:mime-version:message-id:in-reply-to:date :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=iENnNsoCTzlwBGsMv7Mnpwt6IdDhoBtVXLX3WJvXL6w=; b=w9j0OCyG2wzEmcwGP8qZydifBeN0vzroTSSPomTcQC6qYruVx38kOPYkJb+guJpXMC V0qLg7MSi/cwSoU7Wa19xXGjg9ureNBcGW+n1sikEUn1ABbZW1ETFR1+jd0+WOPKreBv ZD7Hy0/Mnu7G7SUtD+Vca9pawJKU23JlgrAK8MWkrntnD2DBRlIYyLNoWPcQ2J/DNozy wdcYPjyQO66EvC/434owC4BEMyL6OdiQfNpxEFwfXmp2eLLvUvcMVkKLyyTzYsjKzTsK b+wS7/2USlppcW4OzfSzc8IZx73CfhDvUo8Y+HvwEyugBXv4aysZj6L9qjcsO36khPb/ NiwA== X-Gm-Message-State: AOJu0YxiDDlFupjFN6JUbm2j+gqtYCFvz/DCWPxRbhK8q4qNe2Z9fXwv u0tC280By80V51CHqJXEwCs6gugL6rQ6 X-Google-Smtp-Source: AGHT+IE/tqLBOTQD/q2r4cLJRocgvvytnTzBBe323GFWCNFmGci9wQpHD+kdIp3dXQs4Qwhedz415vJ0gHkd X-Received: from irogers.svl.corp.google.com ([2620:15c:2a3:200:763b:80fa:23ca:96f8]) (user=irogers job=sendgmr) by 2002:a25:d4c6:0:b0:dae:292e:68de with SMTP id m189-20020a25d4c6000000b00dae292e68demr539714ybf.6.1701237744640; Tue, 28 Nov 2023 22:02:24 -0800 (PST) Date: Tue, 28 Nov 2023 22:02:01 -0800 In-Reply-To: <20231129060211.1890454-1-irogers@google.com> Message-Id: <20231129060211.1890454-5-irogers@google.com> Precedence: bulk X-Mailing-List: bpf@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 References: <20231129060211.1890454-1-irogers@google.com> X-Mailer: git-send-email 2.43.0.rc1.413.gea7ed67945-goog Subject: [PATCH v1 04/14] libperf cpumap: Replace usage of perf_cpu_map__new(NULL) From: Ian Rogers To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Mark Rutland , Alexander Shishkin , Jiri Olsa , Namhyung Kim , Ian Rogers , Adrian Hunter , Suzuki K Poulose , Mike Leach , James Clark , Leo Yan , John Garry , Will Deacon , Thomas Gleixner , Darren Hart , Davidlohr Bueso , " =?utf-8?q?Andr=C3=A9_Almeida?= " , Kan Liang , K Prateek Nayak , Sean Christopherson , Paolo Bonzini , Kajol Jain , Athira Rajeev , Andrew Jones , Alexandre Ghiti , Atish Patra , "Steinar H. Gunderson" , Yang Jihong , Yang Li , Changbin Du , Sandipan Das , Ravi Bangoria , Paran Lee , Nick Desaulniers , Huacai Chen , Yanteng Si , linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org, coresight@lists.linaro.org, linux-arm-kernel@lists.infradead.org, bpf@vger.kernel.org Passing NULL to perf_cpu_map__new performs perf_cpu_map__new_online_cpus, just directly call perf_cpu_map__new_online_cpus to be more intention revealing. Signed-off-by: Ian Rogers Reviewed-by: James Clark --- tools/lib/perf/Documentation/examples/sampling.c | 2 +- tools/lib/perf/Documentation/libperf-sampling.txt | 2 +- tools/lib/perf/evlist.c | 2 +- tools/lib/perf/tests/test-evlist.c | 4 ++-- tools/lib/perf/tests/test-evsel.c | 2 +- tools/perf/arch/arm/util/cs-etm.c | 6 +++--- tools/perf/arch/arm64/util/header.c | 2 +- tools/perf/bench/epoll-ctl.c | 2 +- tools/perf/bench/epoll-wait.c | 2 +- tools/perf/bench/futex-hash.c | 2 +- tools/perf/bench/futex-lock-pi.c | 2 +- tools/perf/bench/futex-requeue.c | 2 +- tools/perf/bench/futex-wake-parallel.c | 2 +- tools/perf/bench/futex-wake.c | 2 +- tools/perf/builtin-ftrace.c | 2 +- tools/perf/tests/code-reading.c | 2 +- tools/perf/tests/keep-tracking.c | 2 +- tools/perf/tests/mmap-basic.c | 2 +- tools/perf/tests/openat-syscall-all-cpus.c | 2 +- tools/perf/tests/perf-time-to-tsc.c | 2 +- tools/perf/tests/switch-tracking.c | 2 +- tools/perf/tests/topology.c | 2 +- tools/perf/util/bpf_counter.c | 2 +- tools/perf/util/cpumap.c | 2 +- tools/perf/util/cputopo.c | 2 +- tools/perf/util/evlist.c | 2 +- tools/perf/util/perf_api_probe.c | 4 ++-- tools/perf/util/record.c | 2 +- 28 files changed, 32 insertions(+), 32 deletions(-) diff --git a/tools/lib/perf/Documentation/examples/sampling.c b/tools/lib/perf/Documentation/examples/sampling.c index 8e1a926a9cfe..bc142f0664b5 100644 --- a/tools/lib/perf/Documentation/examples/sampling.c +++ b/tools/lib/perf/Documentation/examples/sampling.c @@ -39,7 +39,7 @@ int main(int argc, char **argv) libperf_init(libperf_print); - cpus = perf_cpu_map__new(NULL); + cpus = perf_cpu_map__new_online_cpus(); if (!cpus) { fprintf(stderr, "failed to create cpus\n"); return -1; diff --git a/tools/lib/perf/Documentation/libperf-sampling.txt b/tools/lib/perf/Documentation/libperf-sampling.txt index d6ca24f6ef78..2378980fab8a 100644 --- a/tools/lib/perf/Documentation/libperf-sampling.txt +++ b/tools/lib/perf/Documentation/libperf-sampling.txt @@ -97,7 +97,7 @@ In this case we will monitor all the available CPUs: [source,c] -- - 42 cpus = perf_cpu_map__new(NULL); + 42 cpus = perf_cpu_map__new_online_cpus(); 43 if (!cpus) { 44 fprintf(stderr, "failed to create cpus\n"); 45 return -1; diff --git a/tools/lib/perf/evlist.c b/tools/lib/perf/evlist.c index 75f36218fdd9..058e3ff10f9b 100644 --- a/tools/lib/perf/evlist.c +++ b/tools/lib/perf/evlist.c @@ -39,7 +39,7 @@ static void __perf_evlist__propagate_maps(struct perf_evlist *evlist, if (evsel->system_wide) { /* System wide: set the cpu map of the evsel to all online CPUs. */ perf_cpu_map__put(evsel->cpus); - evsel->cpus = perf_cpu_map__new(NULL); + evsel->cpus = perf_cpu_map__new_online_cpus(); } else if (evlist->has_user_cpus && evsel->is_pmu_core) { /* * User requested CPUs on a core PMU, ensure the requested CPUs diff --git a/tools/lib/perf/tests/test-evlist.c b/tools/lib/perf/tests/test-evlist.c index ab63878bacb9..10f70cb41ff1 100644 --- a/tools/lib/perf/tests/test-evlist.c +++ b/tools/lib/perf/tests/test-evlist.c @@ -46,7 +46,7 @@ static int test_stat_cpu(void) }; int err, idx; - cpus = perf_cpu_map__new(NULL); + cpus = perf_cpu_map__new_online_cpus(); __T("failed to create cpus", cpus); evlist = perf_evlist__new(); @@ -350,7 +350,7 @@ static int test_mmap_cpus(void) attr.config = id; - cpus = perf_cpu_map__new(NULL); + cpus = perf_cpu_map__new_online_cpus(); __T("failed to create cpus", cpus); evlist = perf_evlist__new(); diff --git a/tools/lib/perf/tests/test-evsel.c b/tools/lib/perf/tests/test-evsel.c index a11fc51bfb68..545ec3150546 100644 --- a/tools/lib/perf/tests/test-evsel.c +++ b/tools/lib/perf/tests/test-evsel.c @@ -27,7 +27,7 @@ static int test_stat_cpu(void) }; int err, idx; - cpus = perf_cpu_map__new(NULL); + cpus = perf_cpu_map__new_online_cpus(); __T("failed to create cpus", cpus); evsel = perf_evsel__new(&attr); diff --git a/tools/perf/arch/arm/util/cs-etm.c b/tools/perf/arch/arm/util/cs-etm.c index c6b7b3066324..77e6663c1703 100644 --- a/tools/perf/arch/arm/util/cs-etm.c +++ b/tools/perf/arch/arm/util/cs-etm.c @@ -199,7 +199,7 @@ static int cs_etm_validate_config(struct auxtrace_record *itr, { int i, err = -EINVAL; struct perf_cpu_map *event_cpus = evsel->evlist->core.user_requested_cpus; - struct perf_cpu_map *online_cpus = perf_cpu_map__new(NULL); + struct perf_cpu_map *online_cpus = perf_cpu_map__new_online_cpus(); /* Set option of each CPU we have */ for (i = 0; i < cpu__max_cpu().cpu; i++) { @@ -536,7 +536,7 @@ cs_etm_info_priv_size(struct auxtrace_record *itr __maybe_unused, int i; int etmv3 = 0, etmv4 = 0, ete = 0; struct perf_cpu_map *event_cpus = evlist->core.user_requested_cpus; - struct perf_cpu_map *online_cpus = perf_cpu_map__new(NULL); + struct perf_cpu_map *online_cpus = perf_cpu_map__new_online_cpus(); /* cpu map is not empty, we have specific CPUs to work with */ if (!perf_cpu_map__has_any_cpu_or_is_empty(event_cpus)) { @@ -802,7 +802,7 @@ static int cs_etm_info_fill(struct auxtrace_record *itr, u64 nr_cpu, type; struct perf_cpu_map *cpu_map; struct perf_cpu_map *event_cpus = session->evlist->core.user_requested_cpus; - struct perf_cpu_map *online_cpus = perf_cpu_map__new(NULL); + struct perf_cpu_map *online_cpus = perf_cpu_map__new_online_cpus(); struct cs_etm_recording *ptr = container_of(itr, struct cs_etm_recording, itr); struct perf_pmu *cs_etm_pmu = ptr->cs_etm_pmu; diff --git a/tools/perf/arch/arm64/util/header.c b/tools/perf/arch/arm64/util/header.c index a2eef9ec5491..97037499152e 100644 --- a/tools/perf/arch/arm64/util/header.c +++ b/tools/perf/arch/arm64/util/header.c @@ -57,7 +57,7 @@ static int _get_cpuid(char *buf, size_t sz, struct perf_cpu_map *cpus) int get_cpuid(char *buf, size_t sz) { - struct perf_cpu_map *cpus = perf_cpu_map__new(NULL); + struct perf_cpu_map *cpus = perf_cpu_map__new_online_cpus(); int ret; if (!cpus) diff --git a/tools/perf/bench/epoll-ctl.c b/tools/perf/bench/epoll-ctl.c index 6bfffe83dde9..d3db73dac66a 100644 --- a/tools/perf/bench/epoll-ctl.c +++ b/tools/perf/bench/epoll-ctl.c @@ -330,7 +330,7 @@ int bench_epoll_ctl(int argc, const char **argv) act.sa_sigaction = toggle_done; sigaction(SIGINT, &act, NULL); - cpu = perf_cpu_map__new(NULL); + cpu = perf_cpu_map__new_online_cpus(); if (!cpu) goto errmem; diff --git a/tools/perf/bench/epoll-wait.c b/tools/perf/bench/epoll-wait.c index cb5174b53940..06bb3187660a 100644 --- a/tools/perf/bench/epoll-wait.c +++ b/tools/perf/bench/epoll-wait.c @@ -444,7 +444,7 @@ int bench_epoll_wait(int argc, const char **argv) act.sa_sigaction = toggle_done; sigaction(SIGINT, &act, NULL); - cpu = perf_cpu_map__new(NULL); + cpu = perf_cpu_map__new_online_cpus(); if (!cpu) goto errmem; diff --git a/tools/perf/bench/futex-hash.c b/tools/perf/bench/futex-hash.c index 2005a3fa3026..0c69d20efa32 100644 --- a/tools/perf/bench/futex-hash.c +++ b/tools/perf/bench/futex-hash.c @@ -138,7 +138,7 @@ int bench_futex_hash(int argc, const char **argv) exit(EXIT_FAILURE); } - cpu = perf_cpu_map__new(NULL); + cpu = perf_cpu_map__new_online_cpus(); if (!cpu) goto errmem; diff --git a/tools/perf/bench/futex-lock-pi.c b/tools/perf/bench/futex-lock-pi.c index 092cbd52db82..7a4973346180 100644 --- a/tools/perf/bench/futex-lock-pi.c +++ b/tools/perf/bench/futex-lock-pi.c @@ -172,7 +172,7 @@ int bench_futex_lock_pi(int argc, const char **argv) if (argc) goto err; - cpu = perf_cpu_map__new(NULL); + cpu = perf_cpu_map__new_online_cpus(); if (!cpu) err(EXIT_FAILURE, "calloc"); diff --git a/tools/perf/bench/futex-requeue.c b/tools/perf/bench/futex-requeue.c index c0035990a33c..d9ad736c1a3e 100644 --- a/tools/perf/bench/futex-requeue.c +++ b/tools/perf/bench/futex-requeue.c @@ -174,7 +174,7 @@ int bench_futex_requeue(int argc, const char **argv) if (argc) goto err; - cpu = perf_cpu_map__new(NULL); + cpu = perf_cpu_map__new_online_cpus(); if (!cpu) err(EXIT_FAILURE, "cpu_map__new"); diff --git a/tools/perf/bench/futex-wake-parallel.c b/tools/perf/bench/futex-wake-parallel.c index 5ab0234d74e6..b66df553e561 100644 --- a/tools/perf/bench/futex-wake-parallel.c +++ b/tools/perf/bench/futex-wake-parallel.c @@ -264,7 +264,7 @@ int bench_futex_wake_parallel(int argc, const char **argv) err(EXIT_FAILURE, "mlockall"); } - cpu = perf_cpu_map__new(NULL); + cpu = perf_cpu_map__new_online_cpus(); if (!cpu) err(EXIT_FAILURE, "calloc"); diff --git a/tools/perf/bench/futex-wake.c b/tools/perf/bench/futex-wake.c index 18a5894af8bb..690fd6d3da13 100644 --- a/tools/perf/bench/futex-wake.c +++ b/tools/perf/bench/futex-wake.c @@ -149,7 +149,7 @@ int bench_futex_wake(int argc, const char **argv) exit(EXIT_FAILURE); } - cpu = perf_cpu_map__new(NULL); + cpu = perf_cpu_map__new_online_cpus(); if (!cpu) err(EXIT_FAILURE, "calloc"); diff --git a/tools/perf/builtin-ftrace.c b/tools/perf/builtin-ftrace.c index ac2e6c75f912..eb30c8eca488 100644 --- a/tools/perf/builtin-ftrace.c +++ b/tools/perf/builtin-ftrace.c @@ -333,7 +333,7 @@ static int set_tracing_func_irqinfo(struct perf_ftrace *ftrace) static int reset_tracing_cpu(void) { - struct perf_cpu_map *cpumap = perf_cpu_map__new(NULL); + struct perf_cpu_map *cpumap = perf_cpu_map__new_online_cpus(); int ret; ret = set_tracing_cpumask(cpumap); diff --git a/tools/perf/tests/code-reading.c b/tools/perf/tests/code-reading.c index 8620146d0378..7a3a7bbbec71 100644 --- a/tools/perf/tests/code-reading.c +++ b/tools/perf/tests/code-reading.c @@ -610,7 +610,7 @@ static int do_test_code_reading(bool try_kcore) goto out_put; } - cpus = perf_cpu_map__new(NULL); + cpus = perf_cpu_map__new_online_cpus(); if (!cpus) { pr_debug("perf_cpu_map__new failed\n"); goto out_put; diff --git a/tools/perf/tests/keep-tracking.c b/tools/perf/tests/keep-tracking.c index 8f4f9b632e1e..5a3b2bed07f3 100644 --- a/tools/perf/tests/keep-tracking.c +++ b/tools/perf/tests/keep-tracking.c @@ -81,7 +81,7 @@ static int test__keep_tracking(struct test_suite *test __maybe_unused, int subte threads = thread_map__new(-1, getpid(), UINT_MAX); CHECK_NOT_NULL__(threads); - cpus = perf_cpu_map__new(NULL); + cpus = perf_cpu_map__new_online_cpus(); CHECK_NOT_NULL__(cpus); evlist = evlist__new(); diff --git a/tools/perf/tests/mmap-basic.c b/tools/perf/tests/mmap-basic.c index 886a13a77a16..012c8ae439fd 100644 --- a/tools/perf/tests/mmap-basic.c +++ b/tools/perf/tests/mmap-basic.c @@ -52,7 +52,7 @@ static int test__basic_mmap(struct test_suite *test __maybe_unused, int subtest return -1; } - cpus = perf_cpu_map__new(NULL); + cpus = perf_cpu_map__new_online_cpus(); if (cpus == NULL) { pr_debug("perf_cpu_map__new\n"); goto out_free_threads; diff --git a/tools/perf/tests/openat-syscall-all-cpus.c b/tools/perf/tests/openat-syscall-all-cpus.c index f3275be83a33..fb114118c876 100644 --- a/tools/perf/tests/openat-syscall-all-cpus.c +++ b/tools/perf/tests/openat-syscall-all-cpus.c @@ -37,7 +37,7 @@ static int test__openat_syscall_event_on_all_cpus(struct test_suite *test __mayb return -1; } - cpus = perf_cpu_map__new(NULL); + cpus = perf_cpu_map__new_online_cpus(); if (cpus == NULL) { pr_debug("perf_cpu_map__new\n"); goto out_thread_map_delete; diff --git a/tools/perf/tests/perf-time-to-tsc.c b/tools/perf/tests/perf-time-to-tsc.c index efcd71c2738a..bbe2ddeb9b74 100644 --- a/tools/perf/tests/perf-time-to-tsc.c +++ b/tools/perf/tests/perf-time-to-tsc.c @@ -93,7 +93,7 @@ static int test__perf_time_to_tsc(struct test_suite *test __maybe_unused, int su threads = thread_map__new(-1, getpid(), UINT_MAX); CHECK_NOT_NULL__(threads); - cpus = perf_cpu_map__new(NULL); + cpus = perf_cpu_map__new_online_cpus(); CHECK_NOT_NULL__(cpus); evlist = evlist__new(); diff --git a/tools/perf/tests/switch-tracking.c b/tools/perf/tests/switch-tracking.c index e52b031bedc5..5cab17a1942e 100644 --- a/tools/perf/tests/switch-tracking.c +++ b/tools/perf/tests/switch-tracking.c @@ -351,7 +351,7 @@ static int test__switch_tracking(struct test_suite *test __maybe_unused, int sub goto out_err; } - cpus = perf_cpu_map__new(NULL); + cpus = perf_cpu_map__new_online_cpus(); if (!cpus) { pr_debug("perf_cpu_map__new failed!\n"); goto out_err; diff --git a/tools/perf/tests/topology.c b/tools/perf/tests/topology.c index 9dee63734e66..2a842f53fbb5 100644 --- a/tools/perf/tests/topology.c +++ b/tools/perf/tests/topology.c @@ -215,7 +215,7 @@ static int test__session_topology(struct test_suite *test __maybe_unused, int su if (session_write_header(path)) goto free_path; - map = perf_cpu_map__new(NULL); + map = perf_cpu_map__new_online_cpus(); if (map == NULL) { pr_debug("failed to get system cpumap\n"); goto free_path; diff --git a/tools/perf/util/bpf_counter.c b/tools/perf/util/bpf_counter.c index 7f9b0e46e008..7a8af60e0f51 100644 --- a/tools/perf/util/bpf_counter.c +++ b/tools/perf/util/bpf_counter.c @@ -455,7 +455,7 @@ static int bperf__load(struct evsel *evsel, struct target *target) return -1; if (!all_cpu_map) { - all_cpu_map = perf_cpu_map__new(NULL); + all_cpu_map = perf_cpu_map__new_online_cpus(); if (!all_cpu_map) return -1; } diff --git a/tools/perf/util/cpumap.c b/tools/perf/util/cpumap.c index 0e090e8bc334..0581ee0fa5f2 100644 --- a/tools/perf/util/cpumap.c +++ b/tools/perf/util/cpumap.c @@ -672,7 +672,7 @@ struct perf_cpu_map *cpu_map__online(void) /* thread unsafe */ static struct perf_cpu_map *online; if (!online) - online = perf_cpu_map__new(NULL); /* from /sys/devices/system/cpu/online */ + online = perf_cpu_map__new_online_cpus(); /* from /sys/devices/system/cpu/online */ return online; } diff --git a/tools/perf/util/cputopo.c b/tools/perf/util/cputopo.c index 81cfc85f4668..8bbeb2dc76fd 100644 --- a/tools/perf/util/cputopo.c +++ b/tools/perf/util/cputopo.c @@ -267,7 +267,7 @@ struct cpu_topology *cpu_topology__new(void) ncpus = cpu__max_present_cpu().cpu; /* build online CPU map */ - map = perf_cpu_map__new(NULL); + map = perf_cpu_map__new_online_cpus(); if (map == NULL) { pr_debug("failed to get system cpumap\n"); return NULL; diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c index ff7f85ded89d..0ed3ce2aa8eb 100644 --- a/tools/perf/util/evlist.c +++ b/tools/perf/util/evlist.c @@ -1352,7 +1352,7 @@ static int evlist__create_syswide_maps(struct evlist *evlist) * error, and we may not want to do that fallback to a * default cpu identity map :-\ */ - cpus = perf_cpu_map__new(NULL); + cpus = perf_cpu_map__new_online_cpus(); if (!cpus) goto out; diff --git a/tools/perf/util/perf_api_probe.c b/tools/perf/util/perf_api_probe.c index e1e2d701599c..1de3b69cdf4a 100644 --- a/tools/perf/util/perf_api_probe.c +++ b/tools/perf/util/perf_api_probe.c @@ -64,7 +64,7 @@ static bool perf_probe_api(setup_probe_fn_t fn) struct perf_cpu cpu; int ret, i = 0; - cpus = perf_cpu_map__new(NULL); + cpus = perf_cpu_map__new_online_cpus(); if (!cpus) return false; cpu = perf_cpu_map__cpu(cpus, 0); @@ -140,7 +140,7 @@ bool perf_can_record_cpu_wide(void) struct perf_cpu cpu; int fd; - cpus = perf_cpu_map__new(NULL); + cpus = perf_cpu_map__new_online_cpus(); if (!cpus) return false; diff --git a/tools/perf/util/record.c b/tools/perf/util/record.c index 40290382b2d7..87e817b3cf7e 100644 --- a/tools/perf/util/record.c +++ b/tools/perf/util/record.c @@ -238,7 +238,7 @@ bool evlist__can_select_event(struct evlist *evlist, const char *str) evsel = evlist__last(temp_evlist); if (!evlist || perf_cpu_map__has_any_cpu_or_is_empty(evlist->core.user_requested_cpus)) { - struct perf_cpu_map *cpus = perf_cpu_map__new(NULL); + struct perf_cpu_map *cpus = perf_cpu_map__new_online_cpus(); if (cpus) cpu = perf_cpu_map__cpu(cpus, 0); From patchwork Wed Nov 29 06:02:02 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Ian Rogers X-Patchwork-Id: 13472225 X-Patchwork-Delegate: bpf@iogearbox.net Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=google.com header.i=@google.com header.b="ybnh1lIH" Received: from mail-yb1-xb4a.google.com (mail-yb1-xb4a.google.com [IPv6:2607:f8b0:4864:20::b4a]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id EAB0C1BC0 for ; Tue, 28 Nov 2023 22:02:27 -0800 (PST) Received: by mail-yb1-xb4a.google.com with SMTP id 3f1490d57ef6-daf702bde7eso6645480276.3 for ; Tue, 28 Nov 2023 22:02:27 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1701237747; x=1701842547; darn=vger.kernel.org; h=to:from:subject:references:mime-version:message-id:in-reply-to:date :from:to:cc:subject:date:message-id:reply-to; bh=u4ZVQm6BXfm09/tS57asaxt2XeV8jv3rjSJhFgaw+iA=; b=ybnh1lIHrqCD75MWPzDGjjlZEGSE6JvL1bGsP63SmVP5TDDczIG0mK42B2wyYyKJAe GhcmUFT2PqIqEqK3kr+XjXEhpcv+O8K28eAikeVJ73WIPedf8e0+uUveJ233kGX3JH8g dP85102lX13NMXZenBa2nDjZ6b4QhCi/Jlo9QPKv3Xf7cOeP/toykEweiA1zY1MPrtdq CW9hwAk4MaBwoYaLIDmsEUrNDky2rJvMNtFNjHNhtHFh5VoN4UrZwol6AgZ1sym/pjMK OVpuBRsaiz4pudo/ld/F+Tc2tvFl/TUNukOV7/y90p1vHrNs4VYAbgEPGFxNkX/7840g iZAQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1701237747; x=1701842547; h=to:from:subject:references:mime-version:message-id:in-reply-to:date :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=u4ZVQm6BXfm09/tS57asaxt2XeV8jv3rjSJhFgaw+iA=; b=NBJU3VSWRr9pgA0cTIsYz7wnmQn/xworo6j+yLSCA3s5+vssl/O+uAyJnHj+NRpN2Q /xNV4ayBaginSCo6lxQ4/I/k9VjUZ4bUmRckNbey84Qe4N5ysVF/OUohH6aVfW4uzKHR G86ZXTwx0UVyVhhPSqnY7wlz3884p3GUgPjakd0vBkS3Z2a+chfHd2qD3P78wtEj2ZfQ MmHORlRos51u1mqY2lriz+nviS5J1EDrzZ3nlgtE+ij/I25fqeWd53hinTdpmkyH94VG hmdfZPL+VqdxsYbEjxVexQAuOUDORyMHWoXEd6RMC3UFdfxjFYjfj2TV+kDdnVkpJ+QT QEdA== X-Gm-Message-State: AOJu0YyHwJ4Vxw7GiCHyBZVLxQjBtiRzx6wL4pz8GwU1hEbEIXUDiAyD L1mj5YBb94MycI+KEZ0DJYB+8vSbfq1S X-Google-Smtp-Source: AGHT+IGrbmAcgGFKWGO4CN4DZuwGzNNirWTnRPFt7368NYxc+4IVjA+ETsxRxMz4iHCrslIZ6bnFdP+hkjGF X-Received: from irogers.svl.corp.google.com ([2620:15c:2a3:200:763b:80fa:23ca:96f8]) (user=irogers job=sendgmr) by 2002:a25:7141:0:b0:da0:c9a5:b529 with SMTP id m62-20020a257141000000b00da0c9a5b529mr472146ybc.12.1701237746973; Tue, 28 Nov 2023 22:02:26 -0800 (PST) Date: Tue, 28 Nov 2023 22:02:02 -0800 In-Reply-To: <20231129060211.1890454-1-irogers@google.com> Message-Id: <20231129060211.1890454-6-irogers@google.com> Precedence: bulk X-Mailing-List: bpf@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 References: <20231129060211.1890454-1-irogers@google.com> X-Mailer: git-send-email 2.43.0.rc1.413.gea7ed67945-goog Subject: [PATCH v1 05/14] libperf cpumap: Add for_each_cpu that skips the "any CPU" case From: Ian Rogers To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Mark Rutland , Alexander Shishkin , Jiri Olsa , Namhyung Kim , Ian Rogers , Adrian Hunter , Suzuki K Poulose , Mike Leach , James Clark , Leo Yan , John Garry , Will Deacon , Thomas Gleixner , Darren Hart , Davidlohr Bueso , " =?utf-8?q?Andr=C3=A9_Almeida?= " , Kan Liang , K Prateek Nayak , Sean Christopherson , Paolo Bonzini , Kajol Jain , Athira Rajeev , Andrew Jones , Alexandre Ghiti , Atish Patra , "Steinar H. Gunderson" , Yang Jihong , Yang Li , Changbin Du , Sandipan Das , Ravi Bangoria , Paran Lee , Nick Desaulniers , Huacai Chen , Yanteng Si , linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org, coresight@lists.linaro.org, linux-arm-kernel@lists.infradead.org, bpf@vger.kernel.org When iterating CPUs in a CPU map it is often desirable to skip the "any CPU" (aka dummy) case. Add a helper for this and use in builtin-record. Signed-off-by: Ian Rogers Reviewed-by: James Clark --- tools/lib/perf/include/perf/cpumap.h | 6 ++++++ tools/perf/builtin-record.c | 4 +--- 2 files changed, 7 insertions(+), 3 deletions(-) diff --git a/tools/lib/perf/include/perf/cpumap.h b/tools/lib/perf/include/perf/cpumap.h index 9cf361fc5edc..dbe0a7352b64 100644 --- a/tools/lib/perf/include/perf/cpumap.h +++ b/tools/lib/perf/include/perf/cpumap.h @@ -64,6 +64,12 @@ LIBPERF_API bool perf_cpu_map__has_any_cpu(const struct perf_cpu_map *map); (idx) < perf_cpu_map__nr(cpus); \ (idx)++, (cpu) = perf_cpu_map__cpu(cpus, idx)) +#define perf_cpu_map__for_each_cpu_skip_any(_cpu, idx, cpus) \ + for ((idx) = 0, (_cpu) = perf_cpu_map__cpu(cpus, idx); \ + (idx) < perf_cpu_map__nr(cpus); \ + (idx)++, (_cpu) = perf_cpu_map__cpu(cpus, idx)) \ + if ((_cpu).cpu != -1) + #define perf_cpu_map__for_each_idx(idx, cpus) \ for ((idx) = 0; (idx) < perf_cpu_map__nr(cpus); (idx)++) diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c index 8ec818568662..066f9232e947 100644 --- a/tools/perf/builtin-record.c +++ b/tools/perf/builtin-record.c @@ -3580,9 +3580,7 @@ static int record__mmap_cpu_mask_init(struct mmap_cpu_mask *mask, struct perf_cp if (cpu_map__is_dummy(cpus)) return 0; - perf_cpu_map__for_each_cpu(cpu, idx, cpus) { - if (cpu.cpu == -1) - continue; + perf_cpu_map__for_each_cpu_skip_any(cpu, idx, cpus) { /* Return ENODEV is input cpu is greater than max cpu */ if ((unsigned long)cpu.cpu > mask->nbits) return -ENODEV; From patchwork Wed Nov 29 06:02:03 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Ian Rogers X-Patchwork-Id: 13472226 X-Patchwork-Delegate: bpf@iogearbox.net Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=google.com header.i=@google.com header.b="r8F3/jBS" Received: from mail-yw1-x114a.google.com (mail-yw1-x114a.google.com [IPv6:2607:f8b0:4864:20::114a]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 35C901BCF for ; Tue, 28 Nov 2023 22:02:30 -0800 (PST) Received: by mail-yw1-x114a.google.com with SMTP id 00721157ae682-5d032ab478fso47665247b3.0 for ; Tue, 28 Nov 2023 22:02:30 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1701237749; x=1701842549; darn=vger.kernel.org; h=to:from:subject:references:mime-version:message-id:in-reply-to:date :from:to:cc:subject:date:message-id:reply-to; bh=oKpplEK/d17CHcbqAj9G9D9QvSt6Rl/kdbFVrBTbVJE=; b=r8F3/jBSyyKMVrpC9Hj8PeyFnViEY4aP9iji/6hoW8nwhIQ2Y5C+IygZ3PS3AsmkXp UJ+YGJ+HYj3zd4bF9BNnW4PVCqdQfZgbm2iP3HngZR5Fk5AagGgCvOmmhR/uU4/g8uRR Pp647zAa4QbJ0uB7CPO6cn0EsqAykWQn+H98jGOQp6qzU5zo+aXYA1k31BkGPvPviYBb Gr5LvP46omEIi1m97hCtcpjXoRnwRKsX3R8dfYU+fEHcei1rFx+jQdwBOpJwIogD2sI4 sJ561UJus7eNwWlSwCpjPEezoLYxhBkreEWTHVVjpLK8jODMfQ94TJ+zz3n2XMfRGgpg ZpGA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1701237749; x=1701842549; h=to:from:subject:references:mime-version:message-id:in-reply-to:date :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=oKpplEK/d17CHcbqAj9G9D9QvSt6Rl/kdbFVrBTbVJE=; b=LoCzHN7wjvQDtNnhBQolNpdlEU7PUahu/DwfXXAo1NY0mcsUhR7JObgnxyrAzsKxu8 h3SEdMFT4JdF2dbj3Hp2f+2YhYw9KaXuuEY8SBihAvvrvw63ba6SF9+Duz5X8KiXrPkX dw117oNBP2TcHcQGdhPCEGIIBvoAevvgrsG5kq8aGDR4ALYEq0N0uBPiD7kt8+xEC7RT iOU3aAyi26IA/4Hw7AuEx4W/t0wnYa5Uljo5RsmOn/Je43oFNEjYa/tEE8fzYjA35T9Y Vje2bMYXfh+Uxn41TqqydHHijjnuWCIsUGrxeA3z4Notqj/ITJuuFhALMJxSm0VkMeyq Ac3A== X-Gm-Message-State: AOJu0YxVn2vZap8w6UwrjmAj/O7i0EDPM8o7RGz9nXoGkLk+3Q4lyAQz EfcQOSb5P9WaBrItjnGVaetVM9XM0rkU X-Google-Smtp-Source: AGHT+IEN+gAwFsf4MSRmBYmw9rtmwPFX3A/a9Ww3oOkFZ/AmBgDB+5vxiQ7AVxAvJsCiACy2MQaf8zl4pEkT X-Received: from irogers.svl.corp.google.com ([2620:15c:2a3:200:763b:80fa:23ca:96f8]) (user=irogers job=sendgmr) by 2002:a05:690c:fd3:b0:5cc:41c1:606 with SMTP id dg19-20020a05690c0fd300b005cc41c10606mr691036ywb.6.1701237749390; Tue, 28 Nov 2023 22:02:29 -0800 (PST) Date: Tue, 28 Nov 2023 22:02:03 -0800 In-Reply-To: <20231129060211.1890454-1-irogers@google.com> Message-Id: <20231129060211.1890454-7-irogers@google.com> Precedence: bulk X-Mailing-List: bpf@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 References: <20231129060211.1890454-1-irogers@google.com> X-Mailer: git-send-email 2.43.0.rc1.413.gea7ed67945-goog Subject: [PATCH v1 06/14] libperf cpumap: Add any, empty and min helpers From: Ian Rogers To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Mark Rutland , Alexander Shishkin , Jiri Olsa , Namhyung Kim , Ian Rogers , Adrian Hunter , Suzuki K Poulose , Mike Leach , James Clark , Leo Yan , John Garry , Will Deacon , Thomas Gleixner , Darren Hart , Davidlohr Bueso , " =?utf-8?q?Andr=C3=A9_Almeida?= " , Kan Liang , K Prateek Nayak , Sean Christopherson , Paolo Bonzini , Kajol Jain , Athira Rajeev , Andrew Jones , Alexandre Ghiti , Atish Patra , "Steinar H. Gunderson" , Yang Jihong , Yang Li , Changbin Du , Sandipan Das , Ravi Bangoria , Paran Lee , Nick Desaulniers , Huacai Chen , Yanteng Si , linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org, coresight@lists.linaro.org, linux-arm-kernel@lists.infradead.org, bpf@vger.kernel.org Additional helpers to better replace perf_cpu_map__has_any_cpu_or_is_empty. Signed-off-by: Ian Rogers --- tools/lib/perf/cpumap.c | 27 +++++++++++++++++++++++++++ tools/lib/perf/include/perf/cpumap.h | 16 ++++++++++++++++ tools/lib/perf/libperf.map | 4 ++++ 3 files changed, 47 insertions(+) diff --git a/tools/lib/perf/cpumap.c b/tools/lib/perf/cpumap.c index 49fc98e16514..7403819da8fd 100644 --- a/tools/lib/perf/cpumap.c +++ b/tools/lib/perf/cpumap.c @@ -316,6 +316,19 @@ bool perf_cpu_map__has_any_cpu_or_is_empty(const struct perf_cpu_map *map) return map ? __perf_cpu_map__cpu(map, 0).cpu == -1 : true; } +bool perf_cpu_map__is_any_cpu_or_is_empty(const struct perf_cpu_map *map) +{ + if (!map) + return true; + + return __perf_cpu_map__nr(map) == 1 && __perf_cpu_map__cpu(map, 0).cpu == -1; +} + +bool perf_cpu_map__is_empty(const struct perf_cpu_map *map) +{ + return map == NULL; +} + int perf_cpu_map__idx(const struct perf_cpu_map *cpus, struct perf_cpu cpu) { int low, high; @@ -372,6 +385,20 @@ bool perf_cpu_map__has_any_cpu(const struct perf_cpu_map *map) return map && __perf_cpu_map__cpu(map, 0).cpu == -1; } +struct perf_cpu perf_cpu_map__min(const struct perf_cpu_map *map) +{ + struct perf_cpu cpu, result = { + .cpu = -1 + }; + int idx; + + perf_cpu_map__for_each_cpu_skip_any(cpu, idx, map) { + result = cpu; + break; + } + return result; +} + struct perf_cpu perf_cpu_map__max(const struct perf_cpu_map *map) { struct perf_cpu result = { diff --git a/tools/lib/perf/include/perf/cpumap.h b/tools/lib/perf/include/perf/cpumap.h index dbe0a7352b64..523e4348fc96 100644 --- a/tools/lib/perf/include/perf/cpumap.h +++ b/tools/lib/perf/include/perf/cpumap.h @@ -50,6 +50,22 @@ LIBPERF_API int perf_cpu_map__nr(const struct perf_cpu_map *cpus); * perf_cpu_map__has_any_cpu_or_is_empty - is map either empty or has the "any CPU"/dummy value. */ LIBPERF_API bool perf_cpu_map__has_any_cpu_or_is_empty(const struct perf_cpu_map *map); +/** + * perf_cpu_map__is_any_cpu_or_is_empty - is map either empty or the "any CPU"/dummy value. + */ +LIBPERF_API bool perf_cpu_map__is_any_cpu_or_is_empty(const struct perf_cpu_map *map); +/** + * perf_cpu_map__is_empty - does the map contain no values and it doesn't + * contain the special "any CPU"/dummy value. + */ +LIBPERF_API bool perf_cpu_map__is_empty(const struct perf_cpu_map *map); +/** + * perf_cpu_map__min - the minimum CPU value or -1 if empty or just the "any CPU"/dummy value. + */ +LIBPERF_API struct perf_cpu perf_cpu_map__min(const struct perf_cpu_map *map); +/** + * perf_cpu_map__max - the maximum CPU value or -1 if empty or just the "any CPU"/dummy value. + */ LIBPERF_API struct perf_cpu perf_cpu_map__max(const struct perf_cpu_map *map); LIBPERF_API bool perf_cpu_map__has(const struct perf_cpu_map *map, struct perf_cpu cpu); LIBPERF_API bool perf_cpu_map__equal(const struct perf_cpu_map *lhs, diff --git a/tools/lib/perf/libperf.map b/tools/lib/perf/libperf.map index 10b3f3722642..2aa79b696032 100644 --- a/tools/lib/perf/libperf.map +++ b/tools/lib/perf/libperf.map @@ -10,6 +10,10 @@ LIBPERF_0.0.1 { perf_cpu_map__nr; perf_cpu_map__cpu; perf_cpu_map__has_any_cpu_or_is_empty; + perf_cpu_map__is_any_cpu_or_is_empty; + perf_cpu_map__is_empty; + perf_cpu_map__has_any_cpu; + perf_cpu_map__min; perf_cpu_map__max; perf_cpu_map__has; perf_thread_map__new_array; From patchwork Wed Nov 29 06:02:04 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Ian Rogers X-Patchwork-Id: 13472227 X-Patchwork-Delegate: bpf@iogearbox.net Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=google.com header.i=@google.com header.b="3wDaubIm" Received: from mail-yw1-x114a.google.com (mail-yw1-x114a.google.com [IPv6:2607:f8b0:4864:20::114a]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 825321BD8 for ; Tue, 28 Nov 2023 22:02:32 -0800 (PST) Received: by mail-yw1-x114a.google.com with SMTP id 00721157ae682-5cccfca81b2so90619637b3.2 for ; Tue, 28 Nov 2023 22:02:32 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1701237751; x=1701842551; darn=vger.kernel.org; h=to:from:subject:references:mime-version:message-id:in-reply-to:date :from:to:cc:subject:date:message-id:reply-to; bh=fls1hKYyOHnf3/bDsf3iPKJd0wwxSBN0A5pCMLik0gc=; b=3wDaubImVNAsdlj/q51pC/fauMjUFw9sRFujTynC/0lClwFFa7FPFXiBIiOR7bo0WY sm+0sHLTCRrT9lS/iejuJ/xVbu5p7iMsXcrTxX7vp4OfhbNP/UOJLRfaFvP6N6qgi7ax wWh2ERc+EDgS9gsmgtQYkuzRoDnKqdI+NPUM2dTQD3c8Oxb1JOy6g7ga3/6BWkWQieYJ eh7HHarWCXEgS6B4hlaBdCMPNF70x5m8t2IDEUB0mgn/VLMqspXQ2wTRiEeZqQE9dnOz Cxv6gxydrZR1Amo/xgnmaFSxPHoZUHKqZhIw3aUznfaRn+CAowGp3i4jGUIZtaIyPguv Vs4Q== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1701237751; x=1701842551; h=to:from:subject:references:mime-version:message-id:in-reply-to:date :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=fls1hKYyOHnf3/bDsf3iPKJd0wwxSBN0A5pCMLik0gc=; b=B8q2cIeHqzdcVqRRrmyrBSFCX1GTlmmXQqrlAxUXr7+ul/UL4hAPIqiK9ez3Tb7jGp 7XrcHMfZG0B98sIHtClHWTP4uJXdKX8jnx6O6D+dWniuVffasRqXRJJNrwvhgWA4c63/ jkRSkMtYQe8xlJc8RuM10ZdKEIVaGfV5e5jxPo+NYPtVn6MtAPo67KMkLXJ+U0iVTGvS eWMa7DxveZQDTByVSyTx+zNmYbzqyMAalkZs1nVCFf11QoGwx/QrMMgUjUdRZh4agDwt OTT9sINCBHj+jN2sbS9QDdPpIXIIJsLutX9pMCDm57yi3JYnFth9k7Q+K2D6DgT/838l bl9Q== X-Gm-Message-State: AOJu0YyaIgAlIc1Uf3bgMqW5L/3XI93X57Z5DMnPdSxGmFiiLoGgmPIV 4Z+BxevXE+iYghztsKBI5cvxLGSvFvF8 X-Google-Smtp-Source: AGHT+IE+lyuX8dfYIgqcySLvvqClEvuXw1syxnF2VGdznLlQKEF0kdQlIAQFG3Eq5mP6kXDREMLbGDtcMcML X-Received: from irogers.svl.corp.google.com ([2620:15c:2a3:200:763b:80fa:23ca:96f8]) (user=irogers job=sendgmr) by 2002:a05:690c:470e:b0:5cb:d80c:3d34 with SMTP id gz14-20020a05690c470e00b005cbd80c3d34mr370105ywb.8.1701237751669; Tue, 28 Nov 2023 22:02:31 -0800 (PST) Date: Tue, 28 Nov 2023 22:02:04 -0800 In-Reply-To: <20231129060211.1890454-1-irogers@google.com> Message-Id: <20231129060211.1890454-8-irogers@google.com> Precedence: bulk X-Mailing-List: bpf@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 References: <20231129060211.1890454-1-irogers@google.com> X-Mailer: git-send-email 2.43.0.rc1.413.gea7ed67945-goog Subject: [PATCH v1 07/14] perf arm-spe/cs-etm: Directly iterate CPU maps From: Ian Rogers To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Mark Rutland , Alexander Shishkin , Jiri Olsa , Namhyung Kim , Ian Rogers , Adrian Hunter , Suzuki K Poulose , Mike Leach , James Clark , Leo Yan , John Garry , Will Deacon , Thomas Gleixner , Darren Hart , Davidlohr Bueso , " =?utf-8?q?Andr=C3=A9_Almeida?= " , Kan Liang , K Prateek Nayak , Sean Christopherson , Paolo Bonzini , Kajol Jain , Athira Rajeev , Andrew Jones , Alexandre Ghiti , Atish Patra , "Steinar H. Gunderson" , Yang Jihong , Yang Li , Changbin Du , Sandipan Das , Ravi Bangoria , Paran Lee , Nick Desaulniers , Huacai Chen , Yanteng Si , linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org, coresight@lists.linaro.org, linux-arm-kernel@lists.infradead.org, bpf@vger.kernel.org Rather than iterate all CPUs and see if they are in CPU maps, directly iterate the CPU map. Similarly make use of the intersect function. Switch perf_cpu_map__has_any_cpu_or_is_empty to more appropriate alternatives. Signed-off-by: Ian Rogers --- tools/perf/arch/arm/util/cs-etm.c | 77 ++++++++++++---------------- tools/perf/arch/arm64/util/arm-spe.c | 4 +- 2 files changed, 34 insertions(+), 47 deletions(-) diff --git a/tools/perf/arch/arm/util/cs-etm.c b/tools/perf/arch/arm/util/cs-etm.c index 77e6663c1703..a68a72f2f668 100644 --- a/tools/perf/arch/arm/util/cs-etm.c +++ b/tools/perf/arch/arm/util/cs-etm.c @@ -197,38 +197,32 @@ static int cs_etm_validate_timestamp(struct auxtrace_record *itr, static int cs_etm_validate_config(struct auxtrace_record *itr, struct evsel *evsel) { - int i, err = -EINVAL; + int idx, err = -EINVAL; struct perf_cpu_map *event_cpus = evsel->evlist->core.user_requested_cpus; struct perf_cpu_map *online_cpus = perf_cpu_map__new_online_cpus(); + struct perf_cpu_map *intersect_cpus = perf_cpu_map__intersect(event_cpus, online_cpus); + struct perf_cpu cpu; - /* Set option of each CPU we have */ - for (i = 0; i < cpu__max_cpu().cpu; i++) { - struct perf_cpu cpu = { .cpu = i, }; - - /* - * In per-cpu case, do the validation for CPUs to work with. - * In per-thread case, the CPU map is empty. Since the traced - * program can run on any CPUs in this case, thus don't skip - * validation. - */ - if (!perf_cpu_map__has_any_cpu_or_is_empty(event_cpus) && - !perf_cpu_map__has(event_cpus, cpu)) - continue; - - if (!perf_cpu_map__has(online_cpus, cpu)) - continue; + perf_cpu_map__put(online_cpus); - err = cs_etm_validate_context_id(itr, evsel, i); + /* + * Set option of each CPU we have. In per-cpu case, do the validation + * for CPUs to work with. In per-thread case, the CPU map is empty. + * Since the traced program can run on any CPUs in this case, thus don't + * skip validation. + */ + perf_cpu_map__for_each_cpu_skip_any(cpu, idx, intersect_cpus) { + err = cs_etm_validate_context_id(itr, evsel, cpu.cpu); if (err) goto out; - err = cs_etm_validate_timestamp(itr, evsel, i); + err = cs_etm_validate_timestamp(itr, evsel, idx); if (err) goto out; } err = 0; out: - perf_cpu_map__put(online_cpus); + perf_cpu_map__put(intersect_cpus); return err; } @@ -435,7 +429,7 @@ static int cs_etm_recording_options(struct auxtrace_record *itr, * Also the case of per-cpu mmaps, need the contextID in order to be notified * when a context switch happened. */ - if (!perf_cpu_map__has_any_cpu_or_is_empty(cpus)) { + if (!perf_cpu_map__is_any_cpu_or_is_empty(cpus)) { evsel__set_config_if_unset(cs_etm_pmu, cs_etm_evsel, "timestamp", 1); evsel__set_config_if_unset(cs_etm_pmu, cs_etm_evsel, @@ -461,7 +455,7 @@ static int cs_etm_recording_options(struct auxtrace_record *itr, evsel->core.attr.sample_period = 1; /* In per-cpu case, always need the time of mmap events etc */ - if (!perf_cpu_map__has_any_cpu_or_is_empty(cpus)) + if (!perf_cpu_map__is_any_cpu_or_is_empty(cpus)) evsel__set_sample_bit(evsel, TIME); err = cs_etm_validate_config(itr, cs_etm_evsel); @@ -533,38 +527,32 @@ static size_t cs_etm_info_priv_size(struct auxtrace_record *itr __maybe_unused, struct evlist *evlist __maybe_unused) { - int i; + int idx; int etmv3 = 0, etmv4 = 0, ete = 0; struct perf_cpu_map *event_cpus = evlist->core.user_requested_cpus; struct perf_cpu_map *online_cpus = perf_cpu_map__new_online_cpus(); + struct perf_cpu cpu; /* cpu map is not empty, we have specific CPUs to work with */ - if (!perf_cpu_map__has_any_cpu_or_is_empty(event_cpus)) { - for (i = 0; i < cpu__max_cpu().cpu; i++) { - struct perf_cpu cpu = { .cpu = i, }; - - if (!perf_cpu_map__has(event_cpus, cpu) || - !perf_cpu_map__has(online_cpus, cpu)) - continue; + if (!perf_cpu_map__is_empty(event_cpus)) { + struct perf_cpu_map *intersect_cpus = + perf_cpu_map__intersect(event_cpus, online_cpus); - if (cs_etm_is_ete(itr, i)) + perf_cpu_map__for_each_cpu_skip_any(cpu, idx, intersect_cpus) { + if (cs_etm_is_ete(itr, cpu.cpu)) ete++; - else if (cs_etm_is_etmv4(itr, i)) + else if (cs_etm_is_etmv4(itr, cpu.cpu)) etmv4++; else etmv3++; } + perf_cpu_map__put(intersect_cpus); } else { /* get configuration for all CPUs in the system */ - for (i = 0; i < cpu__max_cpu().cpu; i++) { - struct perf_cpu cpu = { .cpu = i, }; - - if (!perf_cpu_map__has(online_cpus, cpu)) - continue; - - if (cs_etm_is_ete(itr, i)) + perf_cpu_map__for_each_cpu(cpu, idx, online_cpus) { + if (cs_etm_is_ete(itr, cpu.cpu)) ete++; - else if (cs_etm_is_etmv4(itr, i)) + else if (cs_etm_is_etmv4(itr, cpu.cpu)) etmv4++; else etmv3++; @@ -814,15 +802,14 @@ static int cs_etm_info_fill(struct auxtrace_record *itr, return -EINVAL; /* If the cpu_map is empty all online CPUs are involved */ - if (perf_cpu_map__has_any_cpu_or_is_empty(event_cpus)) { + if (perf_cpu_map__is_empty(event_cpus)) { cpu_map = online_cpus; } else { /* Make sure all specified CPUs are online */ - for (i = 0; i < perf_cpu_map__nr(event_cpus); i++) { - struct perf_cpu cpu = { .cpu = i, }; + struct perf_cpu cpu; - if (perf_cpu_map__has(event_cpus, cpu) && - !perf_cpu_map__has(online_cpus, cpu)) + perf_cpu_map__for_each_cpu(cpu, i, event_cpus) { + if (!perf_cpu_map__has(online_cpus, cpu)) return -EINVAL; } diff --git a/tools/perf/arch/arm64/util/arm-spe.c b/tools/perf/arch/arm64/util/arm-spe.c index 51ccbfd3d246..0b52e67edb3b 100644 --- a/tools/perf/arch/arm64/util/arm-spe.c +++ b/tools/perf/arch/arm64/util/arm-spe.c @@ -232,7 +232,7 @@ static int arm_spe_recording_options(struct auxtrace_record *itr, * In the case of per-cpu mmaps, sample CPU for AUX event; * also enable the timestamp tracing for samples correlation. */ - if (!perf_cpu_map__has_any_cpu_or_is_empty(cpus)) { + if (!perf_cpu_map__is_any_cpu_or_is_empty(cpus)) { evsel__set_sample_bit(arm_spe_evsel, CPU); evsel__set_config_if_unset(arm_spe_pmu, arm_spe_evsel, "ts_enable", 1); @@ -265,7 +265,7 @@ static int arm_spe_recording_options(struct auxtrace_record *itr, tracking_evsel->core.attr.sample_period = 1; /* In per-cpu case, always need the time of mmap events etc */ - if (!perf_cpu_map__has_any_cpu_or_is_empty(cpus)) { + if (!perf_cpu_map__is_any_cpu_or_is_empty(cpus)) { evsel__set_sample_bit(tracking_evsel, TIME); evsel__set_sample_bit(tracking_evsel, CPU); From patchwork Wed Nov 29 06:02:05 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Ian Rogers X-Patchwork-Id: 13472228 X-Patchwork-Delegate: bpf@iogearbox.net Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=google.com header.i=@google.com header.b="plg/BvyH" Received: from mail-yw1-x114a.google.com (mail-yw1-x114a.google.com [IPv6:2607:f8b0:4864:20::114a]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id C2D711BCF for ; Tue, 28 Nov 2023 22:02:34 -0800 (PST) Received: by mail-yw1-x114a.google.com with SMTP id 00721157ae682-5d0c4ba7081so38044797b3.0 for ; Tue, 28 Nov 2023 22:02:34 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1701237754; x=1701842554; darn=vger.kernel.org; h=to:from:subject:references:mime-version:message-id:in-reply-to:date :from:to:cc:subject:date:message-id:reply-to; bh=VgvVWlesLuugbgRkTEvAvehfG8qq5p36CQo20xvLh64=; b=plg/BvyHeIdYw5/sTw/8O42pXvNpBZZ4xtGoLVCUiJ8muIkzDqGuBwIo95BlhMoKFP o/i3w6YTXdOnDTxRV/0RqWsZm1doL2DbJnJXxvjhVN/rkqy9j/FB0qds3WgYZ0wiJc9+ 36Q3FUJIMVZZuyaEvBd9T7YM8nNaFu2wip6TbRFCM5lMKpmKvaXmqL3sC9eyM31/ZNXw +PcvoJJDV/r6HB4n4V+Ubndu0hujzHDUnucFIp5H5nrcPARH7K1Jtk0lj4a21xEvx1Ta RcaIkHQqNrN3/RXdXd9fGyBYU3vaYlv6IF7qSr455Iq6b52U/Kbdhr2FPs0Dn/BXru3E JrMw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1701237754; x=1701842554; h=to:from:subject:references:mime-version:message-id:in-reply-to:date :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=VgvVWlesLuugbgRkTEvAvehfG8qq5p36CQo20xvLh64=; b=J5OPqE/fLN0ZwS4y5N1ZsC2sGCGHRl4W7rtpnnf/joUQJaO5xpz2mGrOIz+ltS1Lco BNQEYgtRiFlp6MkcBbAFbGVyAKfrXHpiHuVeIvg3VW1fcTJwPodcI3cK+KPljO6EnQEk /+i1vogqLAi/xFr4xOCAM27wn5IAOZg6L6aGwgUvBIfOSQ75bYB10NXDNv0F92iyv16/ RJxVsxM5KDPtOpHEwfHI0GQJzRPWC+9AUWHqDR0sv8xYCXl9Ktk9ywUozb/aA4ltBzgm 2TBtaAmLSzQiay0/085PoSFAIDy2MsmR4s/N6wvIkysZFeWuAIg4cukz2efRcdeYR27p cMCA== X-Gm-Message-State: AOJu0YxcUk5S/b0RYOPrxlJs6YuLX6+jY7SYV1cNsM9Mdo26WYPpfHXu XTlAAJwxLLgjzh6fg9TVqhk+ieooJBAG X-Google-Smtp-Source: AGHT+IHNr4ejdtETuIXhfF0YTzQe2IWwh7lZHWr7MqF4UN0ZlhfgRrJfjgfRl1a5b9HrVLvNyeXWn4aGPi+3 X-Received: from irogers.svl.corp.google.com ([2620:15c:2a3:200:763b:80fa:23ca:96f8]) (user=irogers job=sendgmr) by 2002:a05:690c:2e10:b0:5cc:20a1:90e4 with SMTP id et16-20020a05690c2e1000b005cc20a190e4mr530191ywb.6.1701237753930; Tue, 28 Nov 2023 22:02:33 -0800 (PST) Date: Tue, 28 Nov 2023 22:02:05 -0800 In-Reply-To: <20231129060211.1890454-1-irogers@google.com> Message-Id: <20231129060211.1890454-9-irogers@google.com> Precedence: bulk X-Mailing-List: bpf@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 References: <20231129060211.1890454-1-irogers@google.com> X-Mailer: git-send-email 2.43.0.rc1.413.gea7ed67945-goog Subject: [PATCH v1 08/14] perf intel-pt/intel-bts: Switch perf_cpu_map__has_any_cpu_or_is_empty use From: Ian Rogers To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Mark Rutland , Alexander Shishkin , Jiri Olsa , Namhyung Kim , Ian Rogers , Adrian Hunter , Suzuki K Poulose , Mike Leach , James Clark , Leo Yan , John Garry , Will Deacon , Thomas Gleixner , Darren Hart , Davidlohr Bueso , " =?utf-8?q?Andr=C3=A9_Almeida?= " , Kan Liang , K Prateek Nayak , Sean Christopherson , Paolo Bonzini , Kajol Jain , Athira Rajeev , Andrew Jones , Alexandre Ghiti , Atish Patra , "Steinar H. Gunderson" , Yang Jihong , Yang Li , Changbin Du , Sandipan Das , Ravi Bangoria , Paran Lee , Nick Desaulniers , Huacai Chen , Yanteng Si , linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org, coresight@lists.linaro.org, linux-arm-kernel@lists.infradead.org, bpf@vger.kernel.org Switch perf_cpu_map__has_any_cpu_or_is_empty to perf_cpu_map__is_any_cpu_or_is_empty as a CPU map may contain CPUs as well as the dummy event and perf_cpu_map__is_any_cpu_or_is_empty is a more correct alternative. Signed-off-by: Ian Rogers --- tools/perf/arch/x86/util/intel-bts.c | 4 ++-- tools/perf/arch/x86/util/intel-pt.c | 10 +++++----- 2 files changed, 7 insertions(+), 7 deletions(-) diff --git a/tools/perf/arch/x86/util/intel-bts.c b/tools/perf/arch/x86/util/intel-bts.c index af8ae4647585..34696f3d3d5d 100644 --- a/tools/perf/arch/x86/util/intel-bts.c +++ b/tools/perf/arch/x86/util/intel-bts.c @@ -143,7 +143,7 @@ static int intel_bts_recording_options(struct auxtrace_record *itr, if (!opts->full_auxtrace) return 0; - if (opts->full_auxtrace && !perf_cpu_map__has_any_cpu_or_is_empty(cpus)) { + if (opts->full_auxtrace && !perf_cpu_map__is_any_cpu_or_is_empty(cpus)) { pr_err(INTEL_BTS_PMU_NAME " does not support per-cpu recording\n"); return -EINVAL; } @@ -224,7 +224,7 @@ static int intel_bts_recording_options(struct auxtrace_record *itr, * In the case of per-cpu mmaps, we need the CPU on the * AUX event. */ - if (!perf_cpu_map__has_any_cpu_or_is_empty(cpus)) + if (!perf_cpu_map__is_any_cpu_or_is_empty(cpus)) evsel__set_sample_bit(intel_bts_evsel, CPU); } diff --git a/tools/perf/arch/x86/util/intel-pt.c b/tools/perf/arch/x86/util/intel-pt.c index d199619df3ab..6de7e2d21075 100644 --- a/tools/perf/arch/x86/util/intel-pt.c +++ b/tools/perf/arch/x86/util/intel-pt.c @@ -369,7 +369,7 @@ static int intel_pt_info_fill(struct auxtrace_record *itr, ui__warning("Intel Processor Trace: TSC not available\n"); } - per_cpu_mmaps = !perf_cpu_map__has_any_cpu_or_is_empty(session->evlist->core.user_requested_cpus); + per_cpu_mmaps = !perf_cpu_map__is_any_cpu_or_is_empty(session->evlist->core.user_requested_cpus); auxtrace_info->type = PERF_AUXTRACE_INTEL_PT; auxtrace_info->priv[INTEL_PT_PMU_TYPE] = intel_pt_pmu->type; @@ -774,7 +774,7 @@ static int intel_pt_recording_options(struct auxtrace_record *itr, * Per-cpu recording needs sched_switch events to distinguish different * threads. */ - if (have_timing_info && !perf_cpu_map__has_any_cpu_or_is_empty(cpus) && + if (have_timing_info && !perf_cpu_map__is_any_cpu_or_is_empty(cpus) && !record_opts__no_switch_events(opts)) { if (perf_can_record_switch_events()) { bool cpu_wide = !target__none(&opts->target) && @@ -832,7 +832,7 @@ static int intel_pt_recording_options(struct auxtrace_record *itr, * In the case of per-cpu mmaps, we need the CPU on the * AUX event. */ - if (!perf_cpu_map__has_any_cpu_or_is_empty(cpus)) + if (!perf_cpu_map__is_any_cpu_or_is_empty(cpus)) evsel__set_sample_bit(intel_pt_evsel, CPU); } @@ -858,7 +858,7 @@ static int intel_pt_recording_options(struct auxtrace_record *itr, tracking_evsel->immediate = true; /* In per-cpu case, always need the time of mmap events etc */ - if (!perf_cpu_map__has_any_cpu_or_is_empty(cpus)) { + if (!perf_cpu_map__is_any_cpu_or_is_empty(cpus)) { evsel__set_sample_bit(tracking_evsel, TIME); /* And the CPU for switch events */ evsel__set_sample_bit(tracking_evsel, CPU); @@ -870,7 +870,7 @@ static int intel_pt_recording_options(struct auxtrace_record *itr, * Warn the user when we do not have enough information to decode i.e. * per-cpu with no sched_switch (except workload-only). */ - if (!ptr->have_sched_switch && !perf_cpu_map__has_any_cpu_or_is_empty(cpus) && + if (!ptr->have_sched_switch && !perf_cpu_map__is_any_cpu_or_is_empty(cpus) && !target__none(&opts->target) && !intel_pt_evsel->core.attr.exclude_user) ui__warning("Intel Processor Trace decoding will not be possible except for kernel tracing!\n"); From patchwork Wed Nov 29 06:02:06 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Ian Rogers X-Patchwork-Id: 13472229 X-Patchwork-Delegate: bpf@iogearbox.net Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=google.com header.i=@google.com header.b="MBY45QlV" Received: from mail-yw1-x1149.google.com (mail-yw1-x1149.google.com [IPv6:2607:f8b0:4864:20::1149]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 023051BE3 for ; Tue, 28 Nov 2023 22:02:37 -0800 (PST) Received: by mail-yw1-x1149.google.com with SMTP id 00721157ae682-5cbfcf3ae48so4461337b3.1 for ; Tue, 28 Nov 2023 22:02:37 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1701237756; x=1701842556; darn=vger.kernel.org; h=to:from:subject:references:mime-version:message-id:in-reply-to:date :from:to:cc:subject:date:message-id:reply-to; bh=uG9dHqe+W7QKF/tqT6hMSrmaRYWeZm2ANVmXQYyVcMw=; b=MBY45QlVN+VLPP18MF80I+TOuNvbEVxLQXwYNqvHMktrELL/abBHP+Q5kA5CnVRIQR mJLH0YtVeHKd8ruEswkwl2IJx4tBEzzjqhUCFnAQ8qHCZajJGwVvdCSVebNySoPeCf6V INWDBPUK0Phkoe6v3gfYnHubHHoc7jaR6U9EIcLGyLAqUErf9r2/zaoE947hrC2IWLjf dcroVHPNUhVXDg6rpt1fYKTk5Ee+5rpRY5AZJPHQDKI+L7XLqSayZ3v+S8ttqVPnhl9m yN3kLHsHebcO8oEQfXV/JTfk7qAHLx732RrijpmdXnFLCUTGuyYtUaPxsbF7x68sDaDq 6H/Q== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1701237756; x=1701842556; h=to:from:subject:references:mime-version:message-id:in-reply-to:date :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=uG9dHqe+W7QKF/tqT6hMSrmaRYWeZm2ANVmXQYyVcMw=; b=lJSaG92oU4KOfpZnTejYgLuT0UlEx/NfAZiLKMfuDnIblHx79z6eROrkXR6+oy2tUJ ZXnff+T/F/4QBxgOiRS+S8csvrGEBRxX4kxaE83lEnWcf0hgX5WJQQFuGUiZhQSWiVjA 1RErtb3X2pQylV1rA6H9t3eruSi60C1QDxKDs3Mgb3sTEVnVyT5vZnGrO2EvzytnSsPp ZQal61rtQXfzFMOei+lnGgj8KKPGT0h9Du61to80kmcnSOhoELpwZr+BmO5OzU47Sk9e HwIrA8nmLvtlXZrA0PGiuZK+Raj+VgYSl1bikLt3r4+W/sdjePSWB9Um/RGygBcfB0qd aM9Q== X-Gm-Message-State: AOJu0Yw3EkOAkaQAEECsqOrQkFAV/O0jvMueaaU77QtZHmdErgO5bGQY 3BX2+2WSXMP4tTkqhag6TLGW5Z6e2+yN X-Google-Smtp-Source: AGHT+IHZ7VKuLrd0RTaphxrTR9JPVsSgxwODprHGGe8tro5qukldPn6DgcQliDLHW2yuWD/XgfiuGYaXrgpT X-Received: from irogers.svl.corp.google.com ([2620:15c:2a3:200:763b:80fa:23ca:96f8]) (user=irogers job=sendgmr) by 2002:a25:aa49:0:b0:d89:42d7:e72d with SMTP id s67-20020a25aa49000000b00d8942d7e72dmr669750ybi.3.1701237756399; Tue, 28 Nov 2023 22:02:36 -0800 (PST) Date: Tue, 28 Nov 2023 22:02:06 -0800 In-Reply-To: <20231129060211.1890454-1-irogers@google.com> Message-Id: <20231129060211.1890454-10-irogers@google.com> Precedence: bulk X-Mailing-List: bpf@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 References: <20231129060211.1890454-1-irogers@google.com> X-Mailer: git-send-email 2.43.0.rc1.413.gea7ed67945-goog Subject: [PATCH v1 09/14] perf cpumap: Clean up use of perf_cpu_map__has_any_cpu_or_is_empty From: Ian Rogers To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Mark Rutland , Alexander Shishkin , Jiri Olsa , Namhyung Kim , Ian Rogers , Adrian Hunter , Suzuki K Poulose , Mike Leach , James Clark , Leo Yan , John Garry , Will Deacon , Thomas Gleixner , Darren Hart , Davidlohr Bueso , " =?utf-8?q?Andr=C3=A9_Almeida?= " , Kan Liang , K Prateek Nayak , Sean Christopherson , Paolo Bonzini , Kajol Jain , Athira Rajeev , Andrew Jones , Alexandre Ghiti , Atish Patra , "Steinar H. Gunderson" , Yang Jihong , Yang Li , Changbin Du , Sandipan Das , Ravi Bangoria , Paran Lee , Nick Desaulniers , Huacai Chen , Yanteng Si , linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org, coresight@lists.linaro.org, linux-arm-kernel@lists.infradead.org, bpf@vger.kernel.org Most uses of what was perf_cpu_map__empty but is now perf_cpu_map__has_any_cpu_or_is_empty want to do something with the CPU map if it contains CPUs. Replace uses of perf_cpu_map__has_any_cpu_or_is_empty with other helpers so that CPUs within the map can be handled. Signed-off-by: Ian Rogers Reviewed-by: James Clark --- tools/perf/builtin-c2c.c | 6 +----- tools/perf/builtin-stat.c | 9 ++++----- tools/perf/util/auxtrace.c | 4 ++-- tools/perf/util/record.c | 2 +- tools/perf/util/stat.c | 2 +- 5 files changed, 9 insertions(+), 14 deletions(-) diff --git a/tools/perf/builtin-c2c.c b/tools/perf/builtin-c2c.c index f78eea9e2153..ef7ed53a4b4e 100644 --- a/tools/perf/builtin-c2c.c +++ b/tools/perf/builtin-c2c.c @@ -2319,11 +2319,7 @@ static int setup_nodes(struct perf_session *session) nodes[node] = set; - /* empty node, skip */ - if (perf_cpu_map__has_any_cpu_or_is_empty(map)) - continue; - - perf_cpu_map__for_each_cpu(cpu, idx, map) { + perf_cpu_map__for_each_cpu_skip_any(cpu, idx, map) { __set_bit(cpu.cpu, set); if (WARN_ONCE(cpu2node[cpu.cpu] != -1, "node/cpu topology bug")) diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c index 3303aa20f326..f583027a0639 100644 --- a/tools/perf/builtin-stat.c +++ b/tools/perf/builtin-stat.c @@ -1316,10 +1316,9 @@ static int cpu__get_cache_id_from_map(struct perf_cpu cpu, char *map) * be the first online CPU in the cache domain else use the * first online CPU of the cache domain as the ID. */ - if (perf_cpu_map__has_any_cpu_or_is_empty(cpu_map)) + id = perf_cpu_map__min(cpu_map).cpu; + if (id == -1) id = cpu.cpu; - else - id = perf_cpu_map__cpu(cpu_map, 0).cpu; /* Free the perf_cpu_map used to find the cache ID */ perf_cpu_map__put(cpu_map); @@ -1622,7 +1621,7 @@ static int perf_stat_init_aggr_mode(void) * taking the highest cpu number to be the size of * the aggregation translate cpumap. */ - if (!perf_cpu_map__has_any_cpu_or_is_empty(evsel_list->core.user_requested_cpus)) + if (!perf_cpu_map__is_any_cpu_or_is_empty(evsel_list->core.user_requested_cpus)) nr = perf_cpu_map__max(evsel_list->core.user_requested_cpus).cpu; else nr = 0; @@ -2289,7 +2288,7 @@ int process_stat_config_event(struct perf_session *session, perf_event__read_stat_config(&stat_config, &event->stat_config); - if (perf_cpu_map__has_any_cpu_or_is_empty(st->cpus)) { + if (perf_cpu_map__is_empty(st->cpus)) { if (st->aggr_mode != AGGR_UNSET) pr_warning("warning: processing task data, aggregation mode not set\n"); } else if (st->aggr_mode != AGGR_UNSET) { diff --git a/tools/perf/util/auxtrace.c b/tools/perf/util/auxtrace.c index 3684e6009b63..6b1d4bafad59 100644 --- a/tools/perf/util/auxtrace.c +++ b/tools/perf/util/auxtrace.c @@ -174,7 +174,7 @@ void auxtrace_mmap_params__set_idx(struct auxtrace_mmap_params *mp, struct evlist *evlist, struct evsel *evsel, int idx) { - bool per_cpu = !perf_cpu_map__has_any_cpu_or_is_empty(evlist->core.user_requested_cpus); + bool per_cpu = !perf_cpu_map__has_any_cpu(evlist->core.user_requested_cpus); mp->mmap_needed = evsel->needs_auxtrace_mmap; @@ -648,7 +648,7 @@ int auxtrace_parse_snapshot_options(struct auxtrace_record *itr, static int evlist__enable_event_idx(struct evlist *evlist, struct evsel *evsel, int idx) { - bool per_cpu_mmaps = !perf_cpu_map__has_any_cpu_or_is_empty(evlist->core.user_requested_cpus); + bool per_cpu_mmaps = !perf_cpu_map__has_any_cpu(evlist->core.user_requested_cpus); if (per_cpu_mmaps) { struct perf_cpu evlist_cpu = perf_cpu_map__cpu(evlist->core.all_cpus, idx); diff --git a/tools/perf/util/record.c b/tools/perf/util/record.c index 87e817b3cf7e..e867de8ddaaa 100644 --- a/tools/perf/util/record.c +++ b/tools/perf/util/record.c @@ -237,7 +237,7 @@ bool evlist__can_select_event(struct evlist *evlist, const char *str) evsel = evlist__last(temp_evlist); - if (!evlist || perf_cpu_map__has_any_cpu_or_is_empty(evlist->core.user_requested_cpus)) { + if (!evlist || perf_cpu_map__is_any_cpu_or_is_empty(evlist->core.user_requested_cpus)) { struct perf_cpu_map *cpus = perf_cpu_map__new_online_cpus(); if (cpus) diff --git a/tools/perf/util/stat.c b/tools/perf/util/stat.c index 012c4946b9c4..915808a6211a 100644 --- a/tools/perf/util/stat.c +++ b/tools/perf/util/stat.c @@ -315,7 +315,7 @@ static int check_per_pkg(struct evsel *counter, struct perf_counts_values *vals, if (!counter->per_pkg) return 0; - if (perf_cpu_map__has_any_cpu_or_is_empty(cpus)) + if (perf_cpu_map__is_any_cpu_or_is_empty(cpus)) return 0; if (!mask) { From patchwork Wed Nov 29 06:02:07 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Ian Rogers X-Patchwork-Id: 13472230 X-Patchwork-Delegate: bpf@iogearbox.net Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=google.com header.i=@google.com header.b="3sI5pIGv" Received: from mail-yw1-x114a.google.com (mail-yw1-x114a.google.com [IPv6:2607:f8b0:4864:20::114a]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id B3FD11BC8 for ; Tue, 28 Nov 2023 22:02:39 -0800 (PST) Received: by mail-yw1-x114a.google.com with SMTP id 00721157ae682-5ce16bc121aso68651007b3.1 for ; Tue, 28 Nov 2023 22:02:39 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1701237759; x=1701842559; darn=vger.kernel.org; h=to:from:subject:references:mime-version:message-id:in-reply-to:date :from:to:cc:subject:date:message-id:reply-to; bh=LT6pUkq6nOZyulqCEeWY4llreF+FWbGDdI1GlLo0K+Y=; b=3sI5pIGvs/fAI6Opb6EAFgnk3xrzuTRPcS9vxNKmuSZNIyPTDZC0jOo3c14GBriIXC Q2RnZlK7OaUe1ia7sa3p1SNKl3gSeuMuWPsaDE9xjqcGmD3OwrHAdkM+yhzhWNCMvmkR 9gpQWR6Z+ESxHMpgRoxvi+XFHtqQsCZXpPKGdxoBbDMZGf6BBCaJHa5bL57ZVbO8tcY0 D7OfQgQbecbOXAbOwEmskBN9GUiVn04u50jx44iSqT+yt8+p1/Uvqh+dlwlc20AgS3SE /Kr6+ObDr7si897Ss5wEV6AE1jvv6y7YPJLoUHq7vfkGvjB9nhRPcT2ljWrRsjl05tRN SjKA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1701237759; x=1701842559; h=to:from:subject:references:mime-version:message-id:in-reply-to:date :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=LT6pUkq6nOZyulqCEeWY4llreF+FWbGDdI1GlLo0K+Y=; b=b8d5ZT5+wQH4MJx/s4T0BbYKeJqRvttzMoBo0Zor5bRsvxa0I6+2le7XmZ+LOuvb8a 6d8TtbBZcnbxc+GrB4BwWfUbP1bdgyPkpNLqBOOg1ZGaNJNX1PuQPpO71s7bx7EiOLmW nwv7kAYpMSJSbeaLJDonaYhEOuECn52OkiK9yszk8nxzypKgSgLstqAF62HIFFEmeK5X CgDbN8bx2Sv7LVxAe3kTP7zqswuzVh0QaWYRfuL166bOQS0r5iw9KccLeUgIz2ii6fQc lB5pK041UnCUUAfTlpERzX7ZAzQq9C8Gi5d6RS3gVyci2ITQC7WAVvZg80u74ngrx+Gm E58A== X-Gm-Message-State: AOJu0YyHjYfI8zkmkxDjIq73rmkUi+RDV6XnvZPE5PkluReSTtxXvVk0 3z48W8hhTrCIM61DGhFDUG9mKQivjfu3 X-Google-Smtp-Source: AGHT+IHcLhZcuRAnfvMJq6pSrYjIxfVYzDocHNgvrzM6qynXE4Nqh8rL8wha0ZAZQAWY4sv3ZhFZ2JZcU4oz X-Received: from irogers.svl.corp.google.com ([2620:15c:2a3:200:763b:80fa:23ca:96f8]) (user=irogers job=sendgmr) by 2002:a81:9b10:0:b0:5ca:5fcd:7063 with SMTP id s16-20020a819b10000000b005ca5fcd7063mr598732ywg.3.1701237758929; Tue, 28 Nov 2023 22:02:38 -0800 (PST) Date: Tue, 28 Nov 2023 22:02:07 -0800 In-Reply-To: <20231129060211.1890454-1-irogers@google.com> Message-Id: <20231129060211.1890454-11-irogers@google.com> Precedence: bulk X-Mailing-List: bpf@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 References: <20231129060211.1890454-1-irogers@google.com> X-Mailer: git-send-email 2.43.0.rc1.413.gea7ed67945-goog Subject: [PATCH v1 10/14] perf top: Avoid repeated function calls From: Ian Rogers To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Mark Rutland , Alexander Shishkin , Jiri Olsa , Namhyung Kim , Ian Rogers , Adrian Hunter , Suzuki K Poulose , Mike Leach , James Clark , Leo Yan , John Garry , Will Deacon , Thomas Gleixner , Darren Hart , Davidlohr Bueso , " =?utf-8?q?Andr=C3=A9_Almeida?= " , Kan Liang , K Prateek Nayak , Sean Christopherson , Paolo Bonzini , Kajol Jain , Athira Rajeev , Andrew Jones , Alexandre Ghiti , Atish Patra , "Steinar H. Gunderson" , Yang Jihong , Yang Li , Changbin Du , Sandipan Das , Ravi Bangoria , Paran Lee , Nick Desaulniers , Huacai Chen , Yanteng Si , linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org, coresight@lists.linaro.org, linux-arm-kernel@lists.infradead.org, bpf@vger.kernel.org Add a local variable to avoid repeated calls to perf_cpu_map__nr. Signed-off-by: Ian Rogers Reviewed-by: James Clark --- tools/perf/util/top.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) diff --git a/tools/perf/util/top.c b/tools/perf/util/top.c index be7157de0451..4db3d1bd686c 100644 --- a/tools/perf/util/top.c +++ b/tools/perf/util/top.c @@ -28,6 +28,7 @@ size_t perf_top__header_snprintf(struct perf_top *top, char *bf, size_t size) struct record_opts *opts = &top->record_opts; struct target *target = &opts->target; size_t ret = 0; + int nr_cpus; if (top->samples) { samples_per_sec = top->samples / top->delay_secs; @@ -93,19 +94,17 @@ size_t perf_top__header_snprintf(struct perf_top *top, char *bf, size_t size) else ret += SNPRINTF(bf + ret, size - ret, " (all"); + nr_cpus = perf_cpu_map__nr(top->evlist->core.user_requested_cpus); if (target->cpu_list) ret += SNPRINTF(bf + ret, size - ret, ", CPU%s: %s)", - perf_cpu_map__nr(top->evlist->core.user_requested_cpus) > 1 - ? "s" : "", + nr_cpus > 1 ? "s" : "", target->cpu_list); else { if (target->tid) ret += SNPRINTF(bf + ret, size - ret, ")"); else ret += SNPRINTF(bf + ret, size - ret, ", %d CPU%s)", - perf_cpu_map__nr(top->evlist->core.user_requested_cpus), - perf_cpu_map__nr(top->evlist->core.user_requested_cpus) > 1 - ? "s" : ""); + nr_cpus, nr_cpus > 1 ? "s" : ""); } perf_top__reset_sample_counters(top); From patchwork Wed Nov 29 06:02:08 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Ian Rogers X-Patchwork-Id: 13472231 X-Patchwork-Delegate: bpf@iogearbox.net Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=google.com header.i=@google.com header.b="a4zY/nC+" Received: from mail-yb1-xb49.google.com (mail-yb1-xb49.google.com [IPv6:2607:f8b0:4864:20::b49]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 01D061BC3 for ; Tue, 28 Nov 2023 22:02:42 -0800 (PST) Received: by mail-yb1-xb49.google.com with SMTP id 3f1490d57ef6-db49a6b9be5so4171606276.1 for ; Tue, 28 Nov 2023 22:02:41 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1701237761; x=1701842561; darn=vger.kernel.org; h=to:from:subject:references:mime-version:message-id:in-reply-to:date :from:to:cc:subject:date:message-id:reply-to; bh=0DVMbIRV+NQmnATStJOD8jb70rfi854C2wLg/boNolQ=; b=a4zY/nC+0QtkxYXQUOj8uwuD+fonU+1WoTol7K1a0yha0+XUCYnbvn4CPLXb5tK3aH vJt/cKfnVcxmP24DjxQnvxwPtO54xENEs+k/qZd9ROc2TF5IAIZsRGPE6rqvypeJEiG1 /HC/F5uEqO9iUVYIaSciOoD6+UV7MLN+sOphw68wcXtTHGYqLcijMEeWKxuAvNeaXwGN QBiCewTJkOcL76fD7sQPmI4XVSgN/ON246iDDGrjvUROdQvoI9W7COlfzGYRnikJSRDU iriSFPigvOQ04ojlxspTj8iOOcmdZHREqd64sZtPw1D/xJXdMW8dhTEOr8jdHR2n5+Nl YqOA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1701237761; x=1701842561; h=to:from:subject:references:mime-version:message-id:in-reply-to:date :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=0DVMbIRV+NQmnATStJOD8jb70rfi854C2wLg/boNolQ=; b=fnbkCk9lTEPscOMAdnVFqaMyo5f1/9GaGwHREMPWr12jeIPY+bShQd5sg0pEwuEyu9 DKPnxnIq+Y4BgCFEPQypflq+/XAWF75AzE6Zso1zayIMCopu8pZO3NIZ80/F0lIjykfR jG/Wpgmttt/NBpsAgHHpLZgNFQqoYmkM++xkx5oRItYZZLpFSHgdJ4tHN0PJhS/NJpsx B/NTpDEgroJQdeXdgebqe1o/OGKIddWkD1q8sKTuapX/sE/CFIi9XJPJ3htLbJ54kJB6 lffkgmbkEZbqgqVehbKs6m462z/t+rp4j1ZH8ZVZLOEdkAJxVgMLLdZpSjlRhrrN5aYc zz2Q== X-Gm-Message-State: AOJu0YzBdps4GltYWogHCPESJxANUw6fWF3Fvew/5nqk5yq2RKTYcj+3 3zngt2sHgCBQ+YD5rMvcCpScwAR5GIzU X-Google-Smtp-Source: AGHT+IHJMRqYtdqiq8wS3BvtGG3WjhuhilD2hoypqMhKE4FVBpXobXp8Db6UyUfeaAWJBMxnRYUQrmZznlQj X-Received: from irogers.svl.corp.google.com ([2620:15c:2a3:200:763b:80fa:23ca:96f8]) (user=irogers job=sendgmr) by 2002:a25:e4c5:0:b0:d9c:801c:4230 with SMTP id b188-20020a25e4c5000000b00d9c801c4230mr494820ybh.5.1701237761085; Tue, 28 Nov 2023 22:02:41 -0800 (PST) Date: Tue, 28 Nov 2023 22:02:08 -0800 In-Reply-To: <20231129060211.1890454-1-irogers@google.com> Message-Id: <20231129060211.1890454-12-irogers@google.com> Precedence: bulk X-Mailing-List: bpf@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 References: <20231129060211.1890454-1-irogers@google.com> X-Mailer: git-send-email 2.43.0.rc1.413.gea7ed67945-goog Subject: [PATCH v1 11/14] perf arm64 header: Remove unnecessary CPU map get and put From: Ian Rogers To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Mark Rutland , Alexander Shishkin , Jiri Olsa , Namhyung Kim , Ian Rogers , Adrian Hunter , Suzuki K Poulose , Mike Leach , James Clark , Leo Yan , John Garry , Will Deacon , Thomas Gleixner , Darren Hart , Davidlohr Bueso , " =?utf-8?q?Andr=C3=A9_Almeida?= " , Kan Liang , K Prateek Nayak , Sean Christopherson , Paolo Bonzini , Kajol Jain , Athira Rajeev , Andrew Jones , Alexandre Ghiti , Atish Patra , "Steinar H. Gunderson" , Yang Jihong , Yang Li , Changbin Du , Sandipan Das , Ravi Bangoria , Paran Lee , Nick Desaulniers , Huacai Chen , Yanteng Si , linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org, coresight@lists.linaro.org, linux-arm-kernel@lists.infradead.org, bpf@vger.kernel.org In both cases the CPU map is known owned by either the caller or a PMU. Signed-off-by: Ian Rogers Reviewed-by: James Clark --- tools/perf/arch/arm64/util/header.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/tools/perf/arch/arm64/util/header.c b/tools/perf/arch/arm64/util/header.c index 97037499152e..a9de0b5187dd 100644 --- a/tools/perf/arch/arm64/util/header.c +++ b/tools/perf/arch/arm64/util/header.c @@ -25,8 +25,6 @@ static int _get_cpuid(char *buf, size_t sz, struct perf_cpu_map *cpus) if (!sysfs || sz < MIDR_SIZE) return EINVAL; - cpus = perf_cpu_map__get(cpus); - for (cpu = 0; cpu < perf_cpu_map__nr(cpus); cpu++) { char path[PATH_MAX]; FILE *file; @@ -51,7 +49,6 @@ static int _get_cpuid(char *buf, size_t sz, struct perf_cpu_map *cpus) break; } - perf_cpu_map__put(cpus); return ret; } From patchwork Wed Nov 29 06:02:09 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Ian Rogers X-Patchwork-Id: 13472232 X-Patchwork-Delegate: bpf@iogearbox.net Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=google.com header.i=@google.com header.b="cZfJJGiR" Received: from mail-yw1-x1149.google.com (mail-yw1-x1149.google.com [IPv6:2607:f8b0:4864:20::1149]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 4A2F01BC0 for ; Tue, 28 Nov 2023 22:02:44 -0800 (PST) Received: by mail-yw1-x1149.google.com with SMTP id 00721157ae682-5d04540d5aaso47810337b3.1 for ; Tue, 28 Nov 2023 22:02:44 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1701237763; x=1701842563; darn=vger.kernel.org; h=to:from:subject:references:mime-version:message-id:in-reply-to:date :from:to:cc:subject:date:message-id:reply-to; bh=hx1FXofSm6/tFgZiy1TvVwbEYJz6ssRcV+dGOMGcvW0=; b=cZfJJGiRnIlA8GdwUq/oU0RlrfPGoiItm5nZ2HcpyvEr4tVzQOY/2hl5ZythbdT+xr v3AT9s/llHVuRpVJRf7Qc/GqfPYyTA/X0nQKUlk/EK/NhDAagPZrdlyhQMNNwe9g2O/9 c8Npi68QpzjfSqfMeQRSSnGhfireqB4MBtCNovKGuweTbegohaaKkD1ojALt55jSmJGJ GwQweBtlnxWbsutlbZYItnHsP5w+mfkrq9M2F28Pv9trvcBGUNsbxJaPDjxRT3TtBDM0 wnYjHsPC/wcRG6vtupUxyTAVL94FpTEL4H5JYlC1kVBOwjuC7ey/JUUOFU4vpztW5QA8 iRbA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1701237763; x=1701842563; h=to:from:subject:references:mime-version:message-id:in-reply-to:date :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=hx1FXofSm6/tFgZiy1TvVwbEYJz6ssRcV+dGOMGcvW0=; b=OygDGwo27F57tT7dClrkzNgr7qeBVeWH+br0jxpUL1aB8tWpD7YSop2K2IhNVkEKcz SF9+/vrdZxUKhXbyU2R85v/gIbFq7oHQfVy/+cD062PXeX0Q3eQRc6eqajtmL1Eqx64O O7OY6vysPm2FjHKkRHyq3RBUZSiSUchLJOouit5xDMLB0NSG75TA8KwCRR2fsMrSKyaH ekwGo+VYDijjEOIEg36Sb9sQvzQIEVFe8LaJEpTsZw2xQ3QbaDUN+ghqu4133fB2n3d3 0UEdhKKiNQvlijLDqbU/Jzq3cu+ffuxSS+71ZQLu3GrwHdgjydCdZTmUQPq020PXoBGm ou+w== X-Gm-Message-State: AOJu0YyUlKq34ROR7GcugJ8InjSGzMFCVWnyadNF4hm9kwtyeArYxuew mZeI5bip+FV6jsqWRNn7QxayevB62vmY X-Google-Smtp-Source: AGHT+IFkHheFXSho6uLG11gC+NuzV7oPEGtX6gzG3KyX8XN+0sHsL9fCMoG1bThASPDwSTf3YxUPIjvReV9+ X-Received: from irogers.svl.corp.google.com ([2620:15c:2a3:200:763b:80fa:23ca:96f8]) (user=irogers job=sendgmr) by 2002:a05:690c:2f0d:b0:5cb:e3c9:ca22 with SMTP id ev13-20020a05690c2f0d00b005cbe3c9ca22mr465449ywb.7.1701237763521; Tue, 28 Nov 2023 22:02:43 -0800 (PST) Date: Tue, 28 Nov 2023 22:02:09 -0800 In-Reply-To: <20231129060211.1890454-1-irogers@google.com> Message-Id: <20231129060211.1890454-13-irogers@google.com> Precedence: bulk X-Mailing-List: bpf@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 References: <20231129060211.1890454-1-irogers@google.com> X-Mailer: git-send-email 2.43.0.rc1.413.gea7ed67945-goog Subject: [PATCH v1 12/14] perf stat: Remove duplicate cpus_map_matched function From: Ian Rogers To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Mark Rutland , Alexander Shishkin , Jiri Olsa , Namhyung Kim , Ian Rogers , Adrian Hunter , Suzuki K Poulose , Mike Leach , James Clark , Leo Yan , John Garry , Will Deacon , Thomas Gleixner , Darren Hart , Davidlohr Bueso , " =?utf-8?q?Andr=C3=A9_Almeida?= " , Kan Liang , K Prateek Nayak , Sean Christopherson , Paolo Bonzini , Kajol Jain , Athira Rajeev , Andrew Jones , Alexandre Ghiti , Atish Patra , "Steinar H. Gunderson" , Yang Jihong , Yang Li , Changbin Du , Sandipan Das , Ravi Bangoria , Paran Lee , Nick Desaulniers , Huacai Chen , Yanteng Si , linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org, coresight@lists.linaro.org, linux-arm-kernel@lists.infradead.org, bpf@vger.kernel.org Use libperf's perf_cpu_map__equal that performs the same function. Signed-off-by: Ian Rogers Reviewed-by: James Clark --- tools/perf/builtin-stat.c | 22 +--------------------- 1 file changed, 1 insertion(+), 21 deletions(-) diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c index f583027a0639..8e2f90b5c276 100644 --- a/tools/perf/builtin-stat.c +++ b/tools/perf/builtin-stat.c @@ -164,26 +164,6 @@ static struct perf_stat_config stat_config = { .iostat_run = false, }; -static bool cpus_map_matched(struct evsel *a, struct evsel *b) -{ - if (!a->core.cpus && !b->core.cpus) - return true; - - if (!a->core.cpus || !b->core.cpus) - return false; - - if (perf_cpu_map__nr(a->core.cpus) != perf_cpu_map__nr(b->core.cpus)) - return false; - - for (int i = 0; i < perf_cpu_map__nr(a->core.cpus); i++) { - if (perf_cpu_map__cpu(a->core.cpus, i).cpu != - perf_cpu_map__cpu(b->core.cpus, i).cpu) - return false; - } - - return true; -} - static void evlist__check_cpu_maps(struct evlist *evlist) { struct evsel *evsel, *warned_leader = NULL; @@ -194,7 +174,7 @@ static void evlist__check_cpu_maps(struct evlist *evlist) /* Check that leader matches cpus with each member. */ if (leader == evsel) continue; - if (cpus_map_matched(leader, evsel)) + if (perf_cpu_map__equal(leader->core.cpus, evsel->core.cpus)) continue; /* If there's mismatch disable the group and warn user. */ From patchwork Wed Nov 29 06:02:10 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Ian Rogers X-Patchwork-Id: 13472233 X-Patchwork-Delegate: bpf@iogearbox.net Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=google.com header.i=@google.com header.b="gT2zFlD5" Received: from mail-yb1-xb49.google.com (mail-yb1-xb49.google.com [IPv6:2607:f8b0:4864:20::b49]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 7AEB21BC1 for ; Tue, 28 Nov 2023 22:02:46 -0800 (PST) Received: by mail-yb1-xb49.google.com with SMTP id 3f1490d57ef6-da033914f7cso7731117276.0 for ; Tue, 28 Nov 2023 22:02:46 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1701237765; x=1701842565; darn=vger.kernel.org; h=to:from:subject:references:mime-version:message-id:in-reply-to:date :from:to:cc:subject:date:message-id:reply-to; bh=KjG+yuGvRcuoj12S9jtBpCqgse8AcR4nWFWWBDY4R7c=; b=gT2zFlD54uN8C+jM6E48ajrVnH6LAq6ijqSdtSx9I7gFFAId7v+sKVWWvoiESA1Hqa +mV5e6OKfcdAdtLgrnIbbX2aNiiQJ2SmFkZvvRW0X0wD27RzOGvUwahulxKPMJbGrG+o SUDLooU2dd7+ZNJspDs3bH9C1cSGaifltDyz423QBM4IbXT1XjPIs4qAEvEpau14lsE+ dIAYy6oxaEofM6reEJEdMeuDpEdpK2/RjjJQTlAsA6mlP8hAMM/KbrItUlqjx6j60asi KNpAaAuN7577eLLlUEQksi5yc6hfLx93yzmF0oRP1xFkDnOTLbZnYZtwIdJMgNVT6f9S F/kA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1701237765; x=1701842565; h=to:from:subject:references:mime-version:message-id:in-reply-to:date :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=KjG+yuGvRcuoj12S9jtBpCqgse8AcR4nWFWWBDY4R7c=; b=raCcMbAMRufQB98civRNpF3ATNEf5dI/3mvoXKK1B/Qhmbi7THAucTnAu4Ei+eVTlV S0p3+bgthdfC8W3bbI0MVH4mSpru2sYuPMpzbHK/enNng67iqYJJxCfuOAymAIgs9C+X QVldpl+Yzhjpao4Bz2e+vWk5kXMj1FFeJVWI5kD0qiVeQeYSuoShiNGUkUFQ1d/p4zZC vwKSABjqhq/pm/7rszXUrrHRtqlpfymGRuW3M5eSOOBPCUjI4bYdKXC0nRs4lvjBwd2v qTkg82K69ulvxmmlkMJ/zeZwPT9xBQHJnniO/AzDzHvScDWK6aktRvOedcrPEZ1Qh/VI 2H9w== X-Gm-Message-State: AOJu0YyK2M9TPnKgHZ3UchLTUMHxlLkOl3M5S2Cb+jVGcNCC0wwNSW/p 44rBSKAby+zWo48z/s22roZpVnpQu2Yp X-Google-Smtp-Source: AGHT+IEIIHTmJROlBWSzZtCcoUGOKooMLGfswN+i7t4wFFSQ1/MHr4QAJhupvuajGvXbqTHRTtAT1qvSVoeK X-Received: from irogers.svl.corp.google.com ([2620:15c:2a3:200:763b:80fa:23ca:96f8]) (user=irogers job=sendgmr) by 2002:a25:6810:0:b0:db5:2a4:45d5 with SMTP id d16-20020a256810000000b00db502a445d5mr113790ybc.13.1701237765746; Tue, 28 Nov 2023 22:02:45 -0800 (PST) Date: Tue, 28 Nov 2023 22:02:10 -0800 In-Reply-To: <20231129060211.1890454-1-irogers@google.com> Message-Id: <20231129060211.1890454-14-irogers@google.com> Precedence: bulk X-Mailing-List: bpf@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 References: <20231129060211.1890454-1-irogers@google.com> X-Mailer: git-send-email 2.43.0.rc1.413.gea7ed67945-goog Subject: [PATCH v1 13/14] perf cpumap: Use perf_cpu_map__for_each_cpu when possible From: Ian Rogers To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Mark Rutland , Alexander Shishkin , Jiri Olsa , Namhyung Kim , Ian Rogers , Adrian Hunter , Suzuki K Poulose , Mike Leach , James Clark , Leo Yan , John Garry , Will Deacon , Thomas Gleixner , Darren Hart , Davidlohr Bueso , " =?utf-8?q?Andr=C3=A9_Almeida?= " , Kan Liang , K Prateek Nayak , Sean Christopherson , Paolo Bonzini , Kajol Jain , Athira Rajeev , Andrew Jones , Alexandre Ghiti , Atish Patra , "Steinar H. Gunderson" , Yang Jihong , Yang Li , Changbin Du , Sandipan Das , Ravi Bangoria , Paran Lee , Nick Desaulniers , Huacai Chen , Yanteng Si , linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org, coresight@lists.linaro.org, linux-arm-kernel@lists.infradead.org, bpf@vger.kernel.org Rather than manually iterating the CPU map, use perf_cpu_map__for_each_cpu. When possible tidy local variables. Signed-off-by: Ian Rogers Reviewed-by: James Clark --- tools/perf/arch/arm64/util/header.c | 10 ++-- tools/perf/tests/bitmap.c | 13 +++--- tools/perf/tests/topology.c | 46 +++++++++---------- tools/perf/util/bpf_kwork.c | 16 ++++--- tools/perf/util/bpf_kwork_top.c | 12 ++--- tools/perf/util/cpumap.c | 12 ++--- .../scripting-engines/trace-event-python.c | 12 +++-- tools/perf/util/session.c | 5 +- tools/perf/util/svghelper.c | 20 ++++---- 9 files changed, 72 insertions(+), 74 deletions(-) diff --git a/tools/perf/arch/arm64/util/header.c b/tools/perf/arch/arm64/util/header.c index a9de0b5187dd..741df3614a09 100644 --- a/tools/perf/arch/arm64/util/header.c +++ b/tools/perf/arch/arm64/util/header.c @@ -4,8 +4,6 @@ #include #include #include -#include -#include #include #include #include "debug.h" @@ -19,18 +17,18 @@ static int _get_cpuid(char *buf, size_t sz, struct perf_cpu_map *cpus) { const char *sysfs = sysfs__mountpoint(); - int cpu; - int ret = EINVAL; + struct perf_cpu cpu; + int idx, ret = EINVAL; if (!sysfs || sz < MIDR_SIZE) return EINVAL; - for (cpu = 0; cpu < perf_cpu_map__nr(cpus); cpu++) { + perf_cpu_map__for_each_cpu(cpu, idx, cpus) { char path[PATH_MAX]; FILE *file; scnprintf(path, PATH_MAX, "%s/devices/system/cpu/cpu%d" MIDR, - sysfs, RC_CHK_ACCESS(cpus)->map[cpu].cpu); + sysfs, cpu.cpu); file = fopen(path, "r"); if (!file) { diff --git a/tools/perf/tests/bitmap.c b/tools/perf/tests/bitmap.c index 0173f5402a35..98956e0e0765 100644 --- a/tools/perf/tests/bitmap.c +++ b/tools/perf/tests/bitmap.c @@ -11,18 +11,19 @@ static unsigned long *get_bitmap(const char *str, int nbits) { struct perf_cpu_map *map = perf_cpu_map__new(str); - unsigned long *bm = NULL; - int i; + unsigned long *bm; bm = bitmap_zalloc(nbits); if (map && bm) { - for (i = 0; i < perf_cpu_map__nr(map); i++) - __set_bit(perf_cpu_map__cpu(map, i).cpu, bm); + int i; + struct perf_cpu cpu; + + perf_cpu_map__for_each_cpu(cpu, i, map) + __set_bit(cpu.cpu, bm); } - if (map) - perf_cpu_map__put(map); + perf_cpu_map__put(map); return bm; } diff --git a/tools/perf/tests/topology.c b/tools/perf/tests/topology.c index 2a842f53fbb5..a8cb5ba898ab 100644 --- a/tools/perf/tests/topology.c +++ b/tools/perf/tests/topology.c @@ -68,6 +68,7 @@ static int check_cpu_topology(char *path, struct perf_cpu_map *map) }; int i; struct aggr_cpu_id id; + struct perf_cpu cpu; session = perf_session__new(&data, NULL); TEST_ASSERT_VAL("can't get session", !IS_ERR(session)); @@ -113,8 +114,7 @@ static int check_cpu_topology(char *path, struct perf_cpu_map *map) TEST_ASSERT_VAL("Session header CPU map not set", session->header.env.cpu); for (i = 0; i < session->header.env.nr_cpus_avail; i++) { - struct perf_cpu cpu = { .cpu = i }; - + cpu.cpu = i; if (!perf_cpu_map__has(map, cpu)) continue; pr_debug("CPU %d, core %d, socket %d\n", i, @@ -123,48 +123,48 @@ static int check_cpu_topology(char *path, struct perf_cpu_map *map) } // Test that CPU ID contains socket, die, core and CPU - for (i = 0; i < perf_cpu_map__nr(map); i++) { - id = aggr_cpu_id__cpu(perf_cpu_map__cpu(map, i), NULL); + perf_cpu_map__for_each_cpu(cpu, i, map) { + id = aggr_cpu_id__cpu(cpu, NULL); TEST_ASSERT_VAL("Cpu map - CPU ID doesn't match", - perf_cpu_map__cpu(map, i).cpu == id.cpu.cpu); + cpu.cpu == id.cpu.cpu); TEST_ASSERT_VAL("Cpu map - Core ID doesn't match", - session->header.env.cpu[perf_cpu_map__cpu(map, i).cpu].core_id == id.core); + session->header.env.cpu[cpu.cpu].core_id == id.core); TEST_ASSERT_VAL("Cpu map - Socket ID doesn't match", - session->header.env.cpu[perf_cpu_map__cpu(map, i).cpu].socket_id == + session->header.env.cpu[cpu.cpu].socket_id == id.socket); TEST_ASSERT_VAL("Cpu map - Die ID doesn't match", - session->header.env.cpu[perf_cpu_map__cpu(map, i).cpu].die_id == id.die); + session->header.env.cpu[cpu.cpu].die_id == id.die); TEST_ASSERT_VAL("Cpu map - Node ID is set", id.node == -1); TEST_ASSERT_VAL("Cpu map - Thread IDX is set", id.thread_idx == -1); } // Test that core ID contains socket, die and core - for (i = 0; i < perf_cpu_map__nr(map); i++) { - id = aggr_cpu_id__core(perf_cpu_map__cpu(map, i), NULL); + perf_cpu_map__for_each_cpu(cpu, i, map) { + id = aggr_cpu_id__core(cpu, NULL); TEST_ASSERT_VAL("Core map - Core ID doesn't match", - session->header.env.cpu[perf_cpu_map__cpu(map, i).cpu].core_id == id.core); + session->header.env.cpu[cpu.cpu].core_id == id.core); TEST_ASSERT_VAL("Core map - Socket ID doesn't match", - session->header.env.cpu[perf_cpu_map__cpu(map, i).cpu].socket_id == + session->header.env.cpu[cpu.cpu].socket_id == id.socket); TEST_ASSERT_VAL("Core map - Die ID doesn't match", - session->header.env.cpu[perf_cpu_map__cpu(map, i).cpu].die_id == id.die); + session->header.env.cpu[cpu.cpu].die_id == id.die); TEST_ASSERT_VAL("Core map - Node ID is set", id.node == -1); TEST_ASSERT_VAL("Core map - Thread IDX is set", id.thread_idx == -1); } // Test that die ID contains socket and die - for (i = 0; i < perf_cpu_map__nr(map); i++) { - id = aggr_cpu_id__die(perf_cpu_map__cpu(map, i), NULL); + perf_cpu_map__for_each_cpu(cpu, i, map) { + id = aggr_cpu_id__die(cpu, NULL); TEST_ASSERT_VAL("Die map - Socket ID doesn't match", - session->header.env.cpu[perf_cpu_map__cpu(map, i).cpu].socket_id == + session->header.env.cpu[cpu.cpu].socket_id == id.socket); TEST_ASSERT_VAL("Die map - Die ID doesn't match", - session->header.env.cpu[perf_cpu_map__cpu(map, i).cpu].die_id == id.die); + session->header.env.cpu[cpu.cpu].die_id == id.die); TEST_ASSERT_VAL("Die map - Node ID is set", id.node == -1); TEST_ASSERT_VAL("Die map - Core is set", id.core == -1); @@ -173,10 +173,10 @@ static int check_cpu_topology(char *path, struct perf_cpu_map *map) } // Test that socket ID contains only socket - for (i = 0; i < perf_cpu_map__nr(map); i++) { - id = aggr_cpu_id__socket(perf_cpu_map__cpu(map, i), NULL); + perf_cpu_map__for_each_cpu(cpu, i, map) { + id = aggr_cpu_id__socket(cpu, NULL); TEST_ASSERT_VAL("Socket map - Socket ID doesn't match", - session->header.env.cpu[perf_cpu_map__cpu(map, i).cpu].socket_id == + session->header.env.cpu[cpu.cpu].socket_id == id.socket); TEST_ASSERT_VAL("Socket map - Node ID is set", id.node == -1); @@ -187,10 +187,10 @@ static int check_cpu_topology(char *path, struct perf_cpu_map *map) } // Test that node ID contains only node - for (i = 0; i < perf_cpu_map__nr(map); i++) { - id = aggr_cpu_id__node(perf_cpu_map__cpu(map, i), NULL); + perf_cpu_map__for_each_cpu(cpu, i, map) { + id = aggr_cpu_id__node(cpu, NULL); TEST_ASSERT_VAL("Node map - Node ID doesn't match", - cpu__get_node(perf_cpu_map__cpu(map, i)) == id.node); + cpu__get_node(cpu) == id.node); TEST_ASSERT_VAL("Node map - Socket is set", id.socket == -1); TEST_ASSERT_VAL("Node map - Die ID is set", id.die == -1); TEST_ASSERT_VAL("Node map - Core is set", id.core == -1); diff --git a/tools/perf/util/bpf_kwork.c b/tools/perf/util/bpf_kwork.c index 6eb2c78fd7f4..44f0f708a15d 100644 --- a/tools/perf/util/bpf_kwork.c +++ b/tools/perf/util/bpf_kwork.c @@ -147,12 +147,12 @@ static bool valid_kwork_class_type(enum kwork_class_type type) static int setup_filters(struct perf_kwork *kwork) { - u8 val = 1; - int i, nr_cpus, key, fd; - struct perf_cpu_map *map; - if (kwork->cpu_list != NULL) { - fd = bpf_map__fd(skel->maps.perf_kwork_cpu_filter); + int idx, nr_cpus; + struct perf_cpu_map *map; + struct perf_cpu cpu; + int fd = bpf_map__fd(skel->maps.perf_kwork_cpu_filter); + if (fd < 0) { pr_debug("Invalid cpu filter fd\n"); return -1; @@ -165,8 +165,8 @@ static int setup_filters(struct perf_kwork *kwork) } nr_cpus = libbpf_num_possible_cpus(); - for (i = 0; i < perf_cpu_map__nr(map); i++) { - struct perf_cpu cpu = perf_cpu_map__cpu(map, i); + perf_cpu_map__for_each_cpu(cpu, idx, map) { + u8 val = 1; if (cpu.cpu >= nr_cpus) { perf_cpu_map__put(map); @@ -181,6 +181,8 @@ static int setup_filters(struct perf_kwork *kwork) } if (kwork->profile_name != NULL) { + int key, fd; + if (strlen(kwork->profile_name) >= MAX_KWORKNAME) { pr_err("Requested name filter %s too large, limit to %d\n", kwork->profile_name, MAX_KWORKNAME - 1); diff --git a/tools/perf/util/bpf_kwork_top.c b/tools/perf/util/bpf_kwork_top.c index 035e02272790..22a3b00a1e23 100644 --- a/tools/perf/util/bpf_kwork_top.c +++ b/tools/perf/util/bpf_kwork_top.c @@ -122,11 +122,11 @@ static bool valid_kwork_class_type(enum kwork_class_type type) static int setup_filters(struct perf_kwork *kwork) { - u8 val = 1; - int i, nr_cpus, fd; - struct perf_cpu_map *map; - if (kwork->cpu_list) { + int idx, nr_cpus, fd; + struct perf_cpu_map *map; + struct perf_cpu cpu; + fd = bpf_map__fd(skel->maps.kwork_top_cpu_filter); if (fd < 0) { pr_debug("Invalid cpu filter fd\n"); @@ -140,8 +140,8 @@ static int setup_filters(struct perf_kwork *kwork) } nr_cpus = libbpf_num_possible_cpus(); - for (i = 0; i < perf_cpu_map__nr(map); i++) { - struct perf_cpu cpu = perf_cpu_map__cpu(map, i); + perf_cpu_map__for_each_cpu(cpu, idx, map) { + u8 val = 1; if (cpu.cpu >= nr_cpus) { perf_cpu_map__put(map); diff --git a/tools/perf/util/cpumap.c b/tools/perf/util/cpumap.c index 0581ee0fa5f2..e2287187babd 100644 --- a/tools/perf/util/cpumap.c +++ b/tools/perf/util/cpumap.c @@ -629,10 +629,10 @@ static char hex_char(unsigned char val) size_t cpu_map__snprint_mask(struct perf_cpu_map *map, char *buf, size_t size) { - int i, cpu; + int idx; char *ptr = buf; unsigned char *bitmap; - struct perf_cpu last_cpu = perf_cpu_map__cpu(map, perf_cpu_map__nr(map) - 1); + struct perf_cpu c, last_cpu = perf_cpu_map__max(map); if (buf == NULL) return 0; @@ -643,12 +643,10 @@ size_t cpu_map__snprint_mask(struct perf_cpu_map *map, char *buf, size_t size) return 0; } - for (i = 0; i < perf_cpu_map__nr(map); i++) { - cpu = perf_cpu_map__cpu(map, i).cpu; - bitmap[cpu / 8] |= 1 << (cpu % 8); - } + perf_cpu_map__for_each_cpu(c, idx, map) + bitmap[c.cpu / 8] |= 1 << (c.cpu % 8); - for (cpu = last_cpu.cpu / 4 * 4; cpu >= 0; cpu -= 4) { + for (int cpu = last_cpu.cpu / 4 * 4; cpu >= 0; cpu -= 4) { unsigned char bits = bitmap[cpu / 8]; if (cpu % 8) diff --git a/tools/perf/util/scripting-engines/trace-event-python.c b/tools/perf/util/scripting-engines/trace-event-python.c index 860e1837ba96..8ef0e5ac03c2 100644 --- a/tools/perf/util/scripting-engines/trace-event-python.c +++ b/tools/perf/util/scripting-engines/trace-event-python.c @@ -1693,13 +1693,15 @@ static void python_process_stat(struct perf_stat_config *config, { struct perf_thread_map *threads = counter->core.threads; struct perf_cpu_map *cpus = counter->core.cpus; - int cpu, thread; - for (thread = 0; thread < perf_thread_map__nr(threads); thread++) { - for (cpu = 0; cpu < perf_cpu_map__nr(cpus); cpu++) { - process_stat(counter, perf_cpu_map__cpu(cpus, cpu), + for (int thread = 0; thread < perf_thread_map__nr(threads); thread++) { + int idx; + struct perf_cpu cpu; + + perf_cpu_map__for_each_cpu(cpu, idx, cpus) { + process_stat(counter, cpu, perf_thread_map__pid(threads, thread), tstamp, - perf_counts(counter->counts, cpu, thread)); + perf_counts(counter->counts, idx, thread)); } } } diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c index 199d3e8df315..d52b58344dbc 100644 --- a/tools/perf/util/session.c +++ b/tools/perf/util/session.c @@ -2738,6 +2738,7 @@ int perf_session__cpu_bitmap(struct perf_session *session, int i, err = -1; struct perf_cpu_map *map; int nr_cpus = min(session->header.env.nr_cpus_avail, MAX_NR_CPUS); + struct perf_cpu cpu; for (i = 0; i < PERF_TYPE_MAX; ++i) { struct evsel *evsel; @@ -2759,9 +2760,7 @@ int perf_session__cpu_bitmap(struct perf_session *session, return -1; } - for (i = 0; i < perf_cpu_map__nr(map); i++) { - struct perf_cpu cpu = perf_cpu_map__cpu(map, i); - + perf_cpu_map__for_each_cpu(cpu, i, map) { if (cpu.cpu >= nr_cpus) { pr_err("Requested CPU %d too large. " "Consider raising MAX_NR_CPUS\n", cpu.cpu); diff --git a/tools/perf/util/svghelper.c b/tools/perf/util/svghelper.c index 1892e9b6aa7f..2b04f47f4db0 100644 --- a/tools/perf/util/svghelper.c +++ b/tools/perf/util/svghelper.c @@ -725,26 +725,24 @@ static void scan_core_topology(int *map, struct topology *t, int nr_cpus) static int str_to_bitmap(char *s, cpumask_t *b, int nr_cpus) { - int i; - int ret = 0; - struct perf_cpu_map *m; - struct perf_cpu c; + int idx, ret = 0; + struct perf_cpu_map *map; + struct perf_cpu cpu; - m = perf_cpu_map__new(s); - if (!m) + map = perf_cpu_map__new(s); + if (!map) return -1; - for (i = 0; i < perf_cpu_map__nr(m); i++) { - c = perf_cpu_map__cpu(m, i); - if (c.cpu >= nr_cpus) { + perf_cpu_map__for_each_cpu(cpu, idx, map) { + if (cpu.cpu >= nr_cpus) { ret = -1; break; } - __set_bit(c.cpu, cpumask_bits(b)); + __set_bit(cpu.cpu, cpumask_bits(b)); } - perf_cpu_map__put(m); + perf_cpu_map__put(map); return ret; } From patchwork Wed Nov 29 06:02:11 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Ian Rogers X-Patchwork-Id: 13472234 X-Patchwork-Delegate: bpf@iogearbox.net Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=google.com header.i=@google.com header.b="gotxKEv8" Received: from mail-yw1-x1149.google.com (mail-yw1-x1149.google.com [IPv6:2607:f8b0:4864:20::1149]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 095001BC6 for ; Tue, 28 Nov 2023 22:02:49 -0800 (PST) Received: by mail-yw1-x1149.google.com with SMTP id 00721157ae682-5cd2100f467so84273727b3.1 for ; Tue, 28 Nov 2023 22:02:49 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1701237768; x=1701842568; darn=vger.kernel.org; h=to:from:subject:references:mime-version:message-id:in-reply-to:date :from:to:cc:subject:date:message-id:reply-to; bh=gdtDmuZlpO8R+sCZuGq6XIuCqB6n5MWFW+Mek7m4zb4=; b=gotxKEv80wkHKfwOnrYUFZ1P/QiffG8FS3MtPA9e9ms22h6ijUbW+/gk/3Zc3rYqX0 eZz+enL5hNKX61jBBi0F1WtSYodUp0KpUX0MiQqCsSrT/qXGNmklX1bKK2dXW1i/3Guh MM7V904BCR9zGFc5AfXVYQ5S73yHjMZy5c+XG61hU70H/MmXc3WWwh8KYVR4SjJ0Syni Rn5Hs4cyevaUfF4b13San60VPFuoeKeaYqYnr/RLHSErNHnx9jmQPu6EDahzQGId4VAh azirywDrQQrMois+BFSzJlyIygMrwJDTghbRccfFE9s6SSQCoVEgWC2fXFTexinuJ6R2 1OiA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1701237768; x=1701842568; h=to:from:subject:references:mime-version:message-id:in-reply-to:date :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=gdtDmuZlpO8R+sCZuGq6XIuCqB6n5MWFW+Mek7m4zb4=; b=eolEnkESgSZ2UNRCt2SmEhklmlXuLfOQZYqfOaPpThj3dWJlKE9tm7gZUYukjRs0Jf 3TElvQXBLDWFccr6vLq9q0Xbsc3otp0VNRhUqPy8Pdsq51bHb/WiTLshSjuo3JAf+mJd I+3rkoOqERYR2F3eT9GyimLSuyRpRd3zi91m6IivJBGeELSiTatVB6TyfmxypZO95LFh eC0rIUaOkaDAVGPf1z4V6AOEzIVdFJ5T6Z2VBBW9lSGnIer3RldmsEeDARhr7BeS3OBX xQl8t5VZIDIMsAHYymiEufEfO2H5QmCL83E/FBy4UdyrnOh3q7AJfJ8AlWfR0hM0Zv7o H6AQ== X-Gm-Message-State: AOJu0YyIk4krxFEvxaHJjtQGLaCanD2blceGq2WhB/8I2BVa3TAOc8a/ ZHLS7DrLAMzQurzhIHDkqb/MR9LmznCU X-Google-Smtp-Source: AGHT+IFWysPah8NVdJGuOS/cub7nCqzkPRr9kIA04P1LJGn4EJzUePlje+77t8xYTmBpNHAmkcV+MNHmEHhD X-Received: from irogers.svl.corp.google.com ([2620:15c:2a3:200:763b:80fa:23ca:96f8]) (user=irogers job=sendgmr) by 2002:a05:690c:fcd:b0:5ca:fef:82a4 with SMTP id dg13-20020a05690c0fcd00b005ca0fef82a4mr623872ywb.4.1701237768039; Tue, 28 Nov 2023 22:02:48 -0800 (PST) Date: Tue, 28 Nov 2023 22:02:11 -0800 In-Reply-To: <20231129060211.1890454-1-irogers@google.com> Message-Id: <20231129060211.1890454-15-irogers@google.com> Precedence: bulk X-Mailing-List: bpf@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 References: <20231129060211.1890454-1-irogers@google.com> X-Mailer: git-send-email 2.43.0.rc1.413.gea7ed67945-goog Subject: [PATCH v1 14/14] libperf cpumap: Document perf_cpu_map__nr's behavior From: Ian Rogers To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Mark Rutland , Alexander Shishkin , Jiri Olsa , Namhyung Kim , Ian Rogers , Adrian Hunter , Suzuki K Poulose , Mike Leach , James Clark , Leo Yan , John Garry , Will Deacon , Thomas Gleixner , Darren Hart , Davidlohr Bueso , " =?utf-8?q?Andr=C3=A9_Almeida?= " , Kan Liang , K Prateek Nayak , Sean Christopherson , Paolo Bonzini , Kajol Jain , Athira Rajeev , Andrew Jones , Alexandre Ghiti , Atish Patra , "Steinar H. Gunderson" , Yang Jihong , Yang Li , Changbin Du , Sandipan Das , Ravi Bangoria , Paran Lee , Nick Desaulniers , Huacai Chen , Yanteng Si , linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org, coresight@lists.linaro.org, linux-arm-kernel@lists.infradead.org, bpf@vger.kernel.org perf_cpu_map__nr's behavior around an empty CPU map is strange as it returns that there is 1 CPU. Changing code that may rely on this behavior is hard, we can at least document the behavior. Signed-off-by: Ian Rogers Reviewed-by: James Clark --- tools/lib/perf/include/perf/cpumap.h | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/tools/lib/perf/include/perf/cpumap.h b/tools/lib/perf/include/perf/cpumap.h index 523e4348fc96..90457d17fb2f 100644 --- a/tools/lib/perf/include/perf/cpumap.h +++ b/tools/lib/perf/include/perf/cpumap.h @@ -44,7 +44,18 @@ LIBPERF_API struct perf_cpu_map *perf_cpu_map__merge(struct perf_cpu_map *orig, LIBPERF_API struct perf_cpu_map *perf_cpu_map__intersect(struct perf_cpu_map *orig, struct perf_cpu_map *other); LIBPERF_API void perf_cpu_map__put(struct perf_cpu_map *map); +/** + * perf_cpu_map__cpu - get the CPU value at the given index. Returns -1 if index + * is invalid. + */ LIBPERF_API struct perf_cpu perf_cpu_map__cpu(const struct perf_cpu_map *cpus, int idx); +/** + * perf_cpu_map__nr - for an empty map returns 1, as perf_cpu_map__cpu returns a + * cpu of -1 for an invalid index, this makes an empty map + * look like it contains the "any CPU"/dummy value. Otherwise + * the result is the number CPUs in the map plus one if the + * "any CPU"/dummy value is present. + */ LIBPERF_API int perf_cpu_map__nr(const struct perf_cpu_map *cpus); /** * perf_cpu_map__has_any_cpu_or_is_empty - is map either empty or has the "any CPU"/dummy value.