From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mail-pg1-f202.google.com (mail-pg1-f202.google.com [209.85.215.202]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id E22A4348891 for ; Tue, 4 Nov 2025 23:41:59 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.215.202 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1762299721; cv=none; b=u0zYp2FSlI1NU1hHoc+wsYNkfjaNJAGGIlOOP4lKCsBRsd6xJbIXA/yUj250gV8r53W/s0AyZ/WFNHv8uXOf7sx+vhA8SgXE4woQxmOZzXEi9LCtECi5hf3iasRKkLv7IkZHWv8UgkrAmGKdQcsb29b2tcwlsdDgNCKz18mUMbQ= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1762299721; c=relaxed/simple; bh=bJbxS9bBatxeYgFbkEtaODM1/57yTM4qiiWIKLBpyXA=; h=Date:Mime-Version:Message-ID:Subject:From:To:Content-Type; b=hhC9NzbfmneS8xPYDXDEdMipe/Uo4x4YhCq0RKtGs8f6TN9sgokaizZiKOKe9UNPG0mNW2ZQlBuJRFIwB5gABgB4ijlLQcfdSJgeIyGg0kM0m4wxVZhRBZlCG6gVJnD3DScdAKEXqtQBc+XfqJiJsvK0t3lfrs4OILkcACJTDkg= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dmarc=pass (p=reject dis=none) header.from=google.com; spf=pass smtp.mailfrom=flex--irogers.bounces.google.com; dkim=pass (2048-bit key) header.d=google.com header.i=@google.com header.b=xzZGNcdk; arc=none smtp.client-ip=209.85.215.202 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=reject dis=none) header.from=google.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=flex--irogers.bounces.google.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=google.com header.i=@google.com header.b="xzZGNcdk" Received: by mail-pg1-f202.google.com with SMTP id 41be03b00d2f7-b969f3f5bb1so7625393a12.0 for ; Tue, 04 Nov 2025 15:41:59 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1762299719; x=1762904519; darn=vger.kernel.org; h=to:from:subject:message-id:mime-version:date:from:to:cc:subject :date:message-id:reply-to; bh=GDR4fNhLPPj2XAMnNCUJwi9KHHqZBJ1VgCgYtFssSHU=; b=xzZGNcdkpwuDyh2MQ1ZLAv10zyHlu9+5C5CZAu+iSGYU7b0tyLiYRH7gNeDhDMVckE ml2gIHiY6PZ7mKu3PWrRVir7SxGyGGG1xChUuKwiCzxHQMrw6wJGngODYJZI0qIDr9EL LtDHmWRCqnHxrHTcjQiEPF91uduJeu68kOs9dqfgz12GrNKgU+2lOfD9jw6Xoo/NRn0U ECjsO5YMwCCzSzZFpveyEeba5FDG/C9/GfexUjk6AdPadQfHEnhKeAKyNQuEk6v5jUzu VgI9/LcsWqnZbYG1oyApjwVRSUasONph3qMU24js0Z0lBEdfrYTnU24WY+hiQ3dpQumQ ckBQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1762299719; x=1762904519; h=to:from:subject:message-id:mime-version:date:x-gm-message-state :from:to:cc:subject:date:message-id:reply-to; bh=GDR4fNhLPPj2XAMnNCUJwi9KHHqZBJ1VgCgYtFssSHU=; b=RC+cxTkJCkFnjenR+i2aoNOANOkfCF0BSNEGKznz57+pkPKXBi0kZtpWkqB1d3eJ17 lmYXa3bAkp63nSkA4mo74xM1goDHGEkv2KfVP0N/vvmMammXJ9ZH5kg2cndGvGMgni/z UW6HczrJ/oPiO1VFqLjE1iJF4Po9u1px8ziDjBILpXVqvErIscZumJiWIyIfJL9garRV KA8cevWMCRl//iFc35ZAEra4ygIP/mOGiG0VUon8gkP7ZzbqVekc6y3odgm3VTU/Lf0L xAtgteXzhMXCOAh+liMigUS4+ZRezCyLAZynkGu32QPiJ+GkgSmhAMMkCAMAj0hsUkg6 ro8w== X-Forwarded-Encrypted: i=1; AJvYcCUF5B0aIaufrvSEAHY66yuvKI4sD0y3fFIKzN8aCzI9fLgS26D2dji2UJVsyJ43v1YD2EkBcHEsEo8j4EhGaXtc@vger.kernel.org X-Gm-Message-State: AOJu0YyYZZ/xdguD7VeV/Cp1dbcfH+FYz7qwLdSBuVcQMGTkOgO0xmyS 1RTdqGUd4NoC6idgWCaXIznA92k7iDgjn2PPFR+2+7Y198gtRWlRBeLvZE/PkOchO+HNBikkFBY HJWZfWAok/A== X-Google-Smtp-Source: AGHT+IGQKFWkbw/0w02HmgEtGqNY5xF/QMNYbw7JTx6i8rZxEONlur5j1kykfmfyFQbKqO6q0uvP85j7pdho X-Received: from dlag5.prod.google.com ([2002:a05:701b:2505:b0:119:49ca:6ba2]) (user=irogers job=prod-delivery.src-stubby-dispatcher) by 2002:a05:6a20:6a03:b0:341:3b91:69c7 with SMTP id adf61e73a8af0-34f8450a80bmr1483585637.11.1762299719222; Tue, 04 Nov 2025 15:41:59 -0800 (PST) Date: Tue, 4 Nov 2025 15:41:46 -0800 Precedence: bulk X-Mailing-List: linux-perf-users@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 X-Mailer: git-send-email 2.51.2.1006.ga50a493c49-goog Message-ID: <20251104234148.3103176-1-irogers@google.com> Subject: [PATCH v2 1/3] perf tool_pmu: Use old_count when computing count values for time events From: Ian Rogers To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Namhyung Kim , Alexander Shishkin , Jiri Olsa , Ian Rogers , Adrian Hunter , "Dr. David Alan Gilbert" , Yang Li , James Clark , Thomas Falcon , Thomas Richter , linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org Content-Type: text/plain; charset="UTF-8" When running in interval mode every third count of a time event isn't showing properly: ``` $ perf stat -e duration_time -a -I 1000 1.001082862 1,002,290,425 duration_time 2.004264262 1,003,183,516 duration_time 3.007381401 duration_time 4.011160141 1,003,705,631 duration_time 5.014515385 1,003,290,110 duration_time 6.018539680 duration_time 7.022065321 1,003,591,720 duration_time ``` The regression came in with a different fix, found through bisection, commit 68cb1567439f ("perf tool_pmu: Fix aggregation on duration_time"). The issue is caused by the enabled and running time of the event matching the old_count's and creating a delta of 0, which is indicative of an error. Fixes: 68cb1567439f ("perf tool_pmu: Fix aggregation on duration_time") Signed-off-by: Ian Rogers --- tools/perf/builtin-stat.c | 16 +++++++++-- tools/perf/util/tool_pmu.c | 59 +++++++++++++++++++++----------------- 2 files changed, 46 insertions(+), 29 deletions(-) diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c index 886727ae8529..d89fa9468f89 100644 --- a/tools/perf/builtin-stat.c +++ b/tools/perf/builtin-stat.c @@ -282,17 +282,27 @@ static int read_single_counter(struct evsel *counter, int cpu_map_idx, int threa if (err && cpu_map_idx == 0 && (evsel__tool_event(counter) == TOOL_PMU__EVENT_USER_TIME || evsel__tool_event(counter) == TOOL_PMU__EVENT_SYSTEM_TIME)) { - u64 val, *start_time; struct perf_counts_values *count = perf_counts(counter->counts, cpu_map_idx, thread); + struct perf_counts_values *old_count = NULL; + u64 val; + + if (counter->prev_raw_counts) + old_count = perf_counts(counter->prev_raw_counts, cpu_map_idx, thread); - start_time = xyarray__entry(counter->start_times, cpu_map_idx, thread); if (evsel__tool_event(counter) == TOOL_PMU__EVENT_USER_TIME) val = ru_stats.ru_utime_usec_stat.mean; else val = ru_stats.ru_stime_usec_stat.mean; - count->ena = count->run = *start_time + val; + count->val = val; + if (old_count) { + count->run = old_count->run + 1; + count->ena = old_count->ena + 1; + } else { + count->run++; + count->ena++; + } return 0; } return err; diff --git a/tools/perf/util/tool_pmu.c b/tools/perf/util/tool_pmu.c index f075098488ba..b895e88ff740 100644 --- a/tools/perf/util/tool_pmu.c +++ b/tools/perf/util/tool_pmu.c @@ -431,16 +431,39 @@ bool tool_pmu__read_event(enum tool_pmu_event ev, struct evsel *evsel, u64 *resu } } +static void perf_counts__update(struct perf_counts_values *count, + const struct perf_counts_values *old_count, + bool raw, u64 val) +{ + /* + * The values of enabled and running must make a ratio of 100%. The + * exact values don't matter as long as they are non-zero to avoid + * issues with evsel__count_has_error. + */ + if (old_count) { + count->val = raw ? val : old_count->val + val; + count->run = old_count->run + 1; + count->ena = old_count->ena + 1; + count->lost = old_count->lost; + } else { + count->val = val; + count->run++; + count->ena++; + count->lost = 0; + } +} + int evsel__tool_pmu_read(struct evsel *evsel, int cpu_map_idx, int thread) { __u64 *start_time, cur_time, delta_start; - u64 val; - int fd, err = 0; + int err = 0; struct perf_counts_values *count, *old_count = NULL; bool adjust = false; enum tool_pmu_event ev = evsel__tool_event(evsel); count = perf_counts(evsel->counts, cpu_map_idx, thread); + if (evsel->prev_raw_counts) + old_count = perf_counts(evsel->prev_raw_counts, cpu_map_idx, thread); switch (ev) { case TOOL_PMU__EVENT_HAS_PMEM: @@ -451,26 +474,18 @@ int evsel__tool_pmu_read(struct evsel *evsel, int cpu_map_idx, int thread) case TOOL_PMU__EVENT_NUM_PACKAGES: case TOOL_PMU__EVENT_SLOTS: case TOOL_PMU__EVENT_SMT_ON: - case TOOL_PMU__EVENT_SYSTEM_TSC_FREQ: - if (evsel->prev_raw_counts) - old_count = perf_counts(evsel->prev_raw_counts, cpu_map_idx, thread); - val = 0; + case TOOL_PMU__EVENT_SYSTEM_TSC_FREQ: { + u64 val = 0; + if (cpu_map_idx == 0 && thread == 0) { if (!tool_pmu__read_event(ev, evsel, &val)) { count->lost++; val = 0; } } - if (old_count) { - count->val = old_count->val + val; - count->run = old_count->run + 1; - count->ena = old_count->ena + 1; - } else { - count->val = val; - count->run++; - count->ena++; - } + perf_counts__update(count, old_count, /*raw=*/false, val); return 0; + } case TOOL_PMU__EVENT_DURATION_TIME: /* * Pretend duration_time is only on the first CPU and thread, or @@ -486,9 +501,9 @@ int evsel__tool_pmu_read(struct evsel *evsel, int cpu_map_idx, int thread) case TOOL_PMU__EVENT_USER_TIME: case TOOL_PMU__EVENT_SYSTEM_TIME: { bool system = evsel__tool_event(evsel) == TOOL_PMU__EVENT_SYSTEM_TIME; + int fd = FD(evsel, cpu_map_idx, thread); start_time = xyarray__entry(evsel->start_times, cpu_map_idx, thread); - fd = FD(evsel, cpu_map_idx, thread); lseek(fd, SEEK_SET, 0); if (evsel->pid_stat) { /* The event exists solely on 1 CPU. */ @@ -522,17 +537,9 @@ int evsel__tool_pmu_read(struct evsel *evsel, int cpu_map_idx, int thread) if (adjust) { __u64 ticks_per_sec = sysconf(_SC_CLK_TCK); - delta_start *= 1000000000 / ticks_per_sec; + delta_start *= 1e9 / ticks_per_sec; } - count->val = delta_start; - count->lost = 0; - /* - * The values of enabled and running must make a ratio of 100%. The - * exact values don't matter as long as they are non-zero to avoid - * issues with evsel__count_has_error. - */ - count->ena++; - count->run++; + perf_counts__update(count, old_count, /*raw=*/true, delta_start); return 0; } -- 2.51.2.1006.ga50a493c49-goog