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; }