From nobody Tue Nov 11 11:28:57 2025 Received: from mail-pl1-f202.google.com (mail-pl1-f202.google.com [209.85.214.202]) (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 6724726F29F for ; Tue, 11 Nov 2025 04:04:28 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.214.202 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1762833870; cv=none; b=e6WKxD9hAKD01/ffPvcC6PewOHpoA7Ix+Y0/fQawMlsRfZgXh612/O56/RrEr5uZNnw4cE9612snoRpYF3vJqVT7hmPLeNHN8550ApKWdPy7j5fwi5wXhXsPUWyDFPYchhu8m1nmGqkNYrcSxHfYKjs3VepfwKu9uhJQtXy59Mw= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1762833870; c=relaxed/simple; bh=5gugg7lkraj9TC0kZ6NwAUfMZ08oCZLBOSh79lmJK3Q=; h=Date:In-Reply-To:Mime-Version:References:Message-ID:Subject:From: To:Content-Type; b=V1250kwHo5HmhCHnpt0gM5aHS3R+/7wb1L/eXzWUpgvuGB5xUAENDMDRuvyXDvY6QvwLJUHmV7PcBGLwDE7CwQ+ZRlkjfTiIFe5cHp4QSez/o95NuINQDOvrqSM1zEgXy16pxGC19CLMb6DOJLe1Jdpj2wXU97rXht600QxIwP4= 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=V1eaSCDK; arc=none smtp.client-ip=209.85.214.202 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="V1eaSCDK" Received: by mail-pl1-f202.google.com with SMTP id d9443c01a7336-297e597f6fbso71687255ad.0 for ; Mon, 10 Nov 2025 20:04:28 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1762833868; x=1763438668; darn=vger.kernel.org; h=to:from:subject:message-id:references:mime-version:in-reply-to:date :from:to:cc:subject:date:message-id:reply-to; bh=C/I124Vpc3AJUGEYy3T6Y6dhWjIbeIJN7sVjw18uSEg=; b=V1eaSCDKp3xcy4BzEuGZoR3xVw8A+oaf5KshSak1mED2lNgNQ4jOUe3gM5cQpkutCL NVYlQM+ILlf8Pa9I1FM88SdgUZ+cfnWC+YKy0sc07KaWjZM0Q6MhHuW6K6YkVP+XJDK5 rCoXw/Oue/2bnfbBTrpzR35NUwjeJN4Ab06y65AnUL4MCwAZd79SReKP7DXkvBaz1ge4 Tnji6QjQIdcUUm35zHXVVNFYYguYpjat4azJDu/NS8tF/mmqMyHvJargSqGdsgoSf+la bsaLRXJyyJD7HvYMUkPHaHvTLTZQ0fds+aNYhCHTUWzA+l7bqXfRateY82Gn+XwzSHBB 48Gw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1762833868; x=1763438668; h=to:from:subject:message-id:references:mime-version:in-reply-to:date :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=C/I124Vpc3AJUGEYy3T6Y6dhWjIbeIJN7sVjw18uSEg=; b=JZ83mX7PNdamkPNJUXNZjNA2uNwtgEvqEcih9bJkV+dqQk9tkPVgqYP6GY8jddQfIC tENR3quOozceGS9SYb293vj+8/AduO6BhKu6RGKOrQKBHU1xaKuYo7KiHnkgCh91xEhU Ss17J4KGMjewNVYyjUH5Ydq0qb9XgTcayA44V4xBPrZohFzW9TPvJuSow3zHq3SP/DeB VcYcNcI7/RbClAo1m7gI+BG0GuYOagETAMWO486uhl5quk3pTAkyrFpW0LoalqawKF7C GJmeHGrxRrfKvkV5+2Id+w797SKtim5YARQVAkcWYqeUEmqthEo4nWQqq5BSkLhGw1Jx Je3g== X-Forwarded-Encrypted: i=1; AJvYcCUqIcXo7832+koReR86nvunYT7Ko5SmdA/0UC1WawBg5+AwpXMAXpcwrIxS7A/2OG5AwJyukuwmZGPWyXU=@vger.kernel.org X-Gm-Message-State: AOJu0YwP/jhZVYI4XJdq82G6+9uPr0lQ2lLeF8NMI2g+rKzU9FT7tZtn idcdCvROlMCTXioBbeRK11rcRqnC5YPHyUhj1f1q/vvDJbdV7MMzNF2bQgsgaYU619aI2dCHS+F DW7MAy6yeCQ== X-Google-Smtp-Source: AGHT+IFomYyqh++zhNNCgyW6eUnQVtaZE4eDcAiCVDnxPld217gHT64tQpyx6OdmePxbHM9DBBM4TULNbDTI X-Received: from dlnn4.prod.google.com ([2002:a05:7022:6184:b0:119:b185:ea76]) (user=irogers job=prod-delivery.src-stubby-dispatcher) by 2002:a17:902:b715:b0:295:8db9:305f with SMTP id d9443c01a7336-297e56be263mr100503715ad.34.1762833867710; Mon, 10 Nov 2025 20:04:27 -0800 (PST) Date: Mon, 10 Nov 2025 20:04:00 -0800 In-Reply-To: <20251111040417.270945-1-irogers@google.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 References: <20251111040417.270945-1-irogers@google.com> X-Mailer: git-send-email 2.51.2.1041.gc1ab5b90ca-goog Message-ID: <20251111040417.270945-2-irogers@google.com> Subject: [PATCH v3 01/18] perf metricgroup: Add care to picking the evsel for displaying a metric From: Ian Rogers To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Namhyung Kim , Alexander Shishkin , Jiri Olsa , Ian Rogers , Adrian Hunter , James Clark , Xu Yang , Chun-Tse Shao , Thomas Richter , Sumanth Korikkar , Collin Funk , Thomas Falcon , Howard Chu , Dapeng Mi , Levi Yun , Yang Li , linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org, Andi Kleen , Weilin Wang Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" Rather than using the first evsel in the matched events, try to find the least shared non-tool evsel. The aim is to pick the first evsel that typifies the metric within the list of metrics. This addresses an issue where Default metric group metrics may lose their counter value due to how the stat displaying hides counters for default event/metric output. For a metricgroup like TopdownL1 on an Intel Alderlake the change is, before there are 4 events with metrics: ``` $ perf stat -M topdownL1 -a sleep 1 Performance counter stats for 'system wide': 7,782,334,296 cpu_core/TOPDOWN.SLOTS/ # 10.4 % tma_= bad_speculation # 19.7 % tma_fronten= d_bound 2,668,927,977 cpu_core/topdown-retiring/ # 35.7 % tma_= backend_bound # 34.1 % tma_retiring 803,623,987 cpu_core/topdown-bad-spec/ 167,514,386 cpu_core/topdown-heavy-ops/ 1,555,265,776 cpu_core/topdown-fe-bound/ 2,792,733,013 cpu_core/topdown-be-bound/ 279,769,310 cpu_atom/TOPDOWN_RETIRING.ALL/ # 12.2 % tma_= retiring # 15.1 % tma_bad_spe= culation 457,917,232 cpu_atom/CPU_CLK_UNHALTED.CORE/ # 38.4 % tma_= backend_bound # 34.2 % tma_fronten= d_bound 783,519,226 cpu_atom/TOPDOWN_FE_BOUND.ALL/ 10,790,192 cpu_core/INT_MISC.UOP_DROPPING/ 879,845,633 cpu_atom/TOPDOWN_BE_BOUND.ALL/ ``` After there are 6 events with metrics: ``` $ perf stat -M topdownL1 -a sleep 1 Performance counter stats for 'system wide': 2,377,551,258 cpu_core/TOPDOWN.SLOTS/ # 7.9 % tma_= bad_speculation # 36.4 % tma_fronten= d_bound 480,791,142 cpu_core/topdown-retiring/ # 35.5 % tma_= backend_bound 186,323,991 cpu_core/topdown-bad-spec/ 65,070,590 cpu_core/topdown-heavy-ops/ # 20.1 % tma_= retiring 871,733,444 cpu_core/topdown-fe-bound/ 848,286,598 cpu_core/topdown-be-bound/ 260,936,456 cpu_atom/TOPDOWN_RETIRING.ALL/ # 12.4 % tma_= retiring # 17.6 % tma_bad_spe= culation 419,576,513 cpu_atom/CPU_CLK_UNHALTED.CORE/ 797,132,597 cpu_atom/TOPDOWN_FE_BOUND.ALL/ # 38.0 % tma_= frontend_bound 3,055,447 cpu_core/INT_MISC.UOP_DROPPING/ 671,014,164 cpu_atom/TOPDOWN_BE_BOUND.ALL/ # 32.0 % tma_= backend_bound ``` Signed-off-by: Ian Rogers --- tools/perf/util/metricgroup.c | 48 ++++++++++++++++++++++++++++++++++- 1 file changed, 47 insertions(+), 1 deletion(-) diff --git a/tools/perf/util/metricgroup.c b/tools/perf/util/metricgroup.c index 48936e517803..76092ee26761 100644 --- a/tools/perf/util/metricgroup.c +++ b/tools/perf/util/metricgroup.c @@ -1323,6 +1323,51 @@ static int parse_ids(bool metric_no_merge, bool fake= _pmu, return ret; } =20 +/* How many times will a given evsel be used in a set of metrics? */ +static int count_uses(struct list_head *metric_list, struct evsel *evsel) +{ + const char *metric_id =3D evsel__metric_id(evsel); + struct metric *m; + int uses =3D 0; + + list_for_each_entry(m, metric_list, nd) { + if (hashmap__find(m->pctx->ids, metric_id, NULL)) + uses++; + } + return uses; +} + +/* + * Select the evsel that stat-display will use to trigger shadow/metric + * printing. Pick the least shared non-tool evsel, encouraging metrics to = be + * with a hardware counter that is specific to them. + */ +static struct evsel *pick_display_evsel(struct list_head *metric_list, + struct evsel **metric_events) +{ + struct evsel *selected =3D metric_events[0]; + size_t selected_uses; + bool selected_is_tool; + + if (!selected) + return NULL; + + selected_uses =3D count_uses(metric_list, selected); + selected_is_tool =3D evsel__is_tool(selected); + for (int i =3D 1; metric_events[i]; i++) { + struct evsel *candidate =3D metric_events[i]; + size_t candidate_uses =3D count_uses(metric_list, candidate); + + if ((selected_is_tool && !evsel__is_tool(candidate)) || + (candidate_uses < selected_uses)) { + selected =3D candidate; + selected_uses =3D candidate_uses; + selected_is_tool =3D evsel__is_tool(selected); + } + } + return selected; +} + static int parse_groups(struct evlist *perf_evlist, const char *pmu, const char *str, bool metric_no_group, @@ -1430,7 +1475,8 @@ static int parse_groups(struct evlist *perf_evlist, goto out; } =20 - me =3D metricgroup__lookup(&perf_evlist->metric_events, metric_events[0], + me =3D metricgroup__lookup(&perf_evlist->metric_events, + pick_display_evsel(&metric_list, metric_events), /*create=3D*/true); =20 expr =3D malloc(sizeof(struct metric_expr)); --=20 2.51.2.1041.gc1ab5b90ca-goog From nobody Tue Nov 11 11:28:57 2025 Received: from mail-pg1-f201.google.com (mail-pg1-f201.google.com [209.85.215.201]) (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 65F3B26E6E5 for ; Tue, 11 Nov 2025 04:04:30 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.215.201 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1762833871; cv=none; b=mvjX0mRMBe1AXr9H8PfCZBVBzBY8Sg6zystOavoiAXE3u13Q/sJ2tLzNEg71/+wz40veSck9nHPrOF3OgWY2paPkzsPWtWb6IYpv5+XfAXj/JntHpQjkqXeexRkatPu0ECRghtIDrXWTT6jBYHeLsIgSDWGiefEbGk/I4fMD5yI= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1762833871; c=relaxed/simple; bh=goAJRIvcXmArqjtsnCnAKiuCo6RWRhzBbuO89gkopnI=; h=Date:In-Reply-To:Mime-Version:References:Message-ID:Subject:From: To:Content-Type; b=r9Jtewfm9cW8zGqeE0Bb7A3QQbRaDvhQKwK6NUglU8skjsdL1e9HCy/4WC1DENv53fDxd3T281leDHJAGmO+ktm2wEtTglLEbAy2t9ho29bLs4PisBqQ7LUpMEFlqhTcUAgaOFd4Wy1CfatIThye4xbR85KXgWhkTSGlZObZRjY= 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=VkVXFu3i; arc=none smtp.client-ip=209.85.215.201 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="VkVXFu3i" Received: by mail-pg1-f201.google.com with SMTP id 41be03b00d2f7-bb1875e1416so1843464a12.0 for ; Mon, 10 Nov 2025 20:04:30 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1762833870; x=1763438670; darn=vger.kernel.org; h=to:from:subject:message-id:references:mime-version:in-reply-to:date :from:to:cc:subject:date:message-id:reply-to; bh=+kfrvj9SxzyHWG3QRMI04WWVn93aXW9NO85HAoMrUfM=; b=VkVXFu3iRg2gCoSRFVtVNXc4AeJgSFyFn956Xf/JnNjr7DimIJZ4eSnI4IQ0VKc+6i PubU818PmAfPg/N15Q3lodrL+qis3NTncI2ymZqhRJxNAs/zM5Et6kn5Trhe8vtQ0zJp 11mmJ/q/qcdQHaGfggMZ6YHgborEEpp+EGEWKPsQ86iKcICiDY5OtFt8yKMxGp0GCjA8 OpJa7aND9DowDSE35ssBuUnr0i9kivdlMSofoHewPXFbpIv3M5Bp416y+2As7wl3KdH8 bT8Xu+BaIYkPYTyBF+FFbPN0RRSd/eFUiLDFuAvPDjyEOsHqthgCeCH6yY90K7ujzcYk eWxw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1762833870; x=1763438670; h=to:from:subject:message-id:references:mime-version:in-reply-to:date :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=+kfrvj9SxzyHWG3QRMI04WWVn93aXW9NO85HAoMrUfM=; b=UOqL6/CpWKOcrZtu2HgvQaHDUkSAFGSOok0/pA1qpd0boq2hctl6jN895LoYPerlZi iujqXHMaGHrM/S3czPi6XOltMmyUxgQIMOwkhGqnkiKJ5Q2+Gv3yleuvx7PffkgNqZ4a qk/XTJuTd8bHF8B3HB/LqQOFouSiX6RFVX4kN9rBqkUlmrarb1H0ZeDzNRWFJmyo8f7S aPTKZh52phbzCLBFbuFWXMaknvzxJYOdGtTyCgCM6UbYqoH+aiPIVZbrHQ35AslC2CGV 090SAluPhG0nHhP515+R3NLORGtw1Uy4HOgH3qF1oyxi1OS8jHjtKeyOMhx1NeZbfSDD kAGA== X-Forwarded-Encrypted: i=1; AJvYcCVYdBtRYQ8d+VNHuIqp1pnNQfinJXnt/dXkfQ57CieC4l2XJKYdPBv/jQz6AfeGXFY5Wik2zo619/xwy8Y=@vger.kernel.org X-Gm-Message-State: AOJu0YxuFKbjeO+50WuvZm+sB41LqWEOuXA2lwWkw1UY38hRe/6c8A9c q9bRC3+4Zkfv20QQ8qHI4GxRoHKLFJ8IozgM/epcp1cKIe7c+qwZj6aenYdHycLXyVEvBjlk/vo G96YWHMUuuw== X-Google-Smtp-Source: AGHT+IHVZJ6rPHs+aut4IPB35AikU94fQpOnKja6gvJpXpnu3R3IcAWz1crnWG63f/Ij/vPLbwFlVJFsfgfA X-Received: from dlnn4.prod.google.com ([2002:a05:7022:6184:b0:119:b185:ea76]) (user=irogers job=prod-delivery.src-stubby-dispatcher) by 2002:a17:902:da4b:b0:295:512f:5060 with SMTP id d9443c01a7336-297e540dc24mr140450485ad.7.1762833869681; Mon, 10 Nov 2025 20:04:29 -0800 (PST) Date: Mon, 10 Nov 2025 20:04:01 -0800 In-Reply-To: <20251111040417.270945-1-irogers@google.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 References: <20251111040417.270945-1-irogers@google.com> X-Mailer: git-send-email 2.51.2.1041.gc1ab5b90ca-goog Message-ID: <20251111040417.270945-3-irogers@google.com> Subject: [PATCH v3 02/18] perf expr: Add #target_cpu literal From: Ian Rogers To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Namhyung Kim , Alexander Shishkin , Jiri Olsa , Ian Rogers , Adrian Hunter , James Clark , Xu Yang , Chun-Tse Shao , Thomas Richter , Sumanth Korikkar , Collin Funk , Thomas Falcon , Howard Chu , Dapeng Mi , Levi Yun , Yang Li , linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org, Andi Kleen , Weilin Wang Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" For CPU nanoseconds a lot of the stat-shadow metrics use either task-clock or cpu-clock, the latter being used when target__has_cpu. Add a #target_cpu literal so that json metrics can perform the same test. Signed-off-by: Ian Rogers --- tools/perf/util/expr.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/tools/perf/util/expr.c b/tools/perf/util/expr.c index 7fda0ff89c16..4df56f2b283d 100644 --- a/tools/perf/util/expr.c +++ b/tools/perf/util/expr.c @@ -409,6 +409,9 @@ double expr__get_literal(const char *literal, const str= uct expr_scanner_ctx *ctx } else if (!strcmp("#core_wide", literal)) { result =3D core_wide(ctx->system_wide, ctx->user_requested_cpu_list) ? 1.0 : 0.0; + } else if (!strcmp("#target_cpu", literal)) { + result =3D (ctx->system_wide || ctx->user_requested_cpu_list) + ? 1.0 : 0.0; } else { pr_err("Unrecognized literal '%s'", literal); } --=20 2.51.2.1041.gc1ab5b90ca-goog From nobody Tue Nov 11 11:28:57 2025 Received: from mail-pl1-f202.google.com (mail-pl1-f202.google.com [209.85.214.202]) (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 4888934A79A for ; Tue, 11 Nov 2025 04:04:32 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.214.202 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1762833874; cv=none; b=KI0ymKuz/yq3O+WXajX/QRry/U+I+7l2DX8gy04yHw/mi/IFyqX18aHj7IqquUyBUiEKU8tKliTmpXOmq4LLymCf3KuHTlCMuSR6Whv4ZBEqedhjQCvZ+oMXK42AHxQ9XxTDvXz/iH8Ph1MDfxcI7ENWNM8nNVl2iI7HKulRElY= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1762833874; c=relaxed/simple; bh=7KZoKBULcESJJJNoo/pxl3bu88timebhUFqQdW+GsaI=; h=Date:In-Reply-To:Mime-Version:References:Message-ID:Subject:From: To:Content-Type; b=W66o305bPI7di38NePjF04yhlvM9d62D+EYX05HoHNOpdzDyMqociyW9SPS5O04qyhQFaIHVAWl4ynCM/1G/zWOL0S0/uuMJYgz/kCVkGm9mWVGLjJyR9l4q6ctR2OXfUJcIb+s6nq7g5gVwhFT43KealAbeyUOShuXeg7s49m8= 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=JH14rpV8; arc=none smtp.client-ip=209.85.214.202 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="JH14rpV8" Received: by mail-pl1-f202.google.com with SMTP id d9443c01a7336-2958c80fcabso102670545ad.0 for ; Mon, 10 Nov 2025 20:04:32 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1762833871; x=1763438671; darn=vger.kernel.org; h=content-transfer-encoding:to:from:subject:message-id:references :mime-version:in-reply-to:date:from:to:cc:subject:date:message-id :reply-to; bh=CSAbFr3qDW+/+4VJQdLwUavDOE/BpHjIQoE1yMzrcnw=; b=JH14rpV8wzCaIKKIl4ZgORIbzJYWIDQnXcR6p24TEhbreWOyK4TqEEE5zbAPC+FdI/ Z0d85IRde07am6o46VM+rAJRbZHyf70pb+F22Huc9KtO4MNMb2U7oM37z83Ybn3nC7CX lhzrNYpnc7xk5ndg1SBJsyZLgfHrTYbfpMZGCJLp0AbaMojtGN8oCIloc6PtFdEB7wps gJ+xarm1tKReo+hWwR6uGqBRvo6KZ2vRDc632zFS4KU/0SIM2VhOVLr7uT0fEC0tdoEp qGvcsRpjx4bmF0mbyVYMeK82AFTLtahozmNnqVa1p05r+oIP5+OQ20wuIHC7mneH4NOU 8kXQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1762833871; x=1763438671; h=content-transfer-encoding:to:from:subject:message-id:references :mime-version:in-reply-to:date:x-gm-message-state:from:to:cc:subject :date:message-id:reply-to; bh=CSAbFr3qDW+/+4VJQdLwUavDOE/BpHjIQoE1yMzrcnw=; b=JcOnMtY8S7ILIyC5/FrWemu3cnHVpcY110PPC4Oicdqr5Vch25GnhYcBbeXbmDatvu nmNCFsrckl8a52CPKDC4vtAFuX+vQsHsgewmdKzif6KHA5dTdliTLJQbXgWWmy8dYyDM 1CDVmE6W11sXbnycKtaAKVT3t+K4k/3APAxeraQ2/3NTcKdf1xWrojfX24FRderovgXB jTRWlRcZzQ7zHKEhxYHz8PWi7ZQHWu3LuJpqgliOiVAyM+ZtonEca+Dxzs99qI4BC0jr 3HPD830KzRyEA5SZAwpvky+w2+Wuc1bR6MC3AOLVgvVJeziTTte7BvjyCDcxw7eGZ+qR x2HQ== X-Forwarded-Encrypted: i=1; AJvYcCUM2KfBYfIQUmXVt+XEKNlYL4oCylxClR1Uila4f4vdoVTHcZ2aDv5fsQqXCSn//UgmafzDf4bWSLEwC3Y=@vger.kernel.org X-Gm-Message-State: AOJu0YwB3YMfBr22saBxYUjtT07FaBn4KzNDPTfi5r2uBSmMCHkh55AS pYDZAqfmkItShB7WXwjLgOxK3VePiM224x/RD4HS6AuHEdl7hVjEY7Qjx9Jv5Awiji23wtlvxIt 37TI9NISFXQ== X-Google-Smtp-Source: AGHT+IGLRRG153nTznyGIoV/50JX6eZMg6cTMJRR5YH9mLxMOQ+cHhM1X+TxU7tm+ihGNukkwb/DMxkJFfLi X-Received: from dyckg20.prod.google.com ([2002:a05:7301:d194:b0:2a4:75f7:2135]) (user=irogers job=prod-delivery.src-stubby-dispatcher) by 2002:a17:902:f908:b0:295:70b1:edc8 with SMTP id d9443c01a7336-297e53e7d36mr105713935ad.2.1762833871449; Mon, 10 Nov 2025 20:04:31 -0800 (PST) Date: Mon, 10 Nov 2025 20:04:02 -0800 In-Reply-To: <20251111040417.270945-1-irogers@google.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 References: <20251111040417.270945-1-irogers@google.com> X-Mailer: git-send-email 2.51.2.1041.gc1ab5b90ca-goog Message-ID: <20251111040417.270945-4-irogers@google.com> Subject: [PATCH v3 03/18] perf jevents: Add set of common metrics based on default ones From: Ian Rogers To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Namhyung Kim , Alexander Shishkin , Jiri Olsa , Ian Rogers , Adrian Hunter , James Clark , Xu Yang , Chun-Tse Shao , Thomas Richter , Sumanth Korikkar , Collin Funk , Thomas Falcon , Howard Chu , Dapeng Mi , Levi Yun , Yang Li , linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org, Andi Kleen , Weilin Wang Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" Add support to getting a common set of metrics from a default table. It simplifies the generation to add json metrics at the same time. The metrics added are CPUs_utilized, cs_per_second, migrations_per_second, page_faults_per_second, insn_per_cycle, stalled_cycles_per_instruction, frontend_cycles_idle, backend_cycles_idle, cycles_frequency, branch_frequency and branch_miss_rate based on the shadow metric definitions. Following this change the default perf stat output on an alderlake looks li= ke: ``` $ perf stat -a -- sleep 1 Performance counter stats for 'system wide': 28,165,735,434 cpu-clock # 27.973 CPUs ut= ilized 23,220 context-switches # 824.406 /sec 833 cpu-migrations # 29.575 /sec 35,293 page-faults # 1.253 K/sec 997,341,554 cpu_atom/instructions/ # 0.84 insn pe= r cycle (35.63%) 11,197,053,736 cpu_core/instructions/ # 1.97 insn pe= r cycle (58.21%) 1,184,871,493 cpu_atom/cycles/ # 0.042 GHz = (35.64%) 5,676,692,769 cpu_core/cycles/ # 0.202 GHz = (58.22%) 150,525,309 cpu_atom/branches/ # 5.344 M/sec = (42.80%) 2,277,232,030 cpu_core/branches/ # 80.851 M/sec = (58.21%) 5,248,575 cpu_atom/branch-misses/ # 3.49% of all = branches (42.82%) 28,829,930 cpu_core/branch-misses/ # 1.27% of all = branches (58.22%) (software) # 824.4 cs/sec cs_per= _second TopdownL1 (cpu_core) # 12.6 % tma_bad_spe= culation # 28.8 % tma_fronten= d_bound (66.57%) TopdownL1 (cpu_core) # 25.8 % tma_backend= _bound # 32.8 % tma_retirin= g (66.57%) (software) # 1253.1 faults/sec pa= ge_faults_per_second # 0.0 GHz cycles_fr= equency (42.80%) # 0.2 GHz cycles_fr= equency (74.92%) TopdownL1 (cpu_atom) # 22.3 % tma_bad_spe= culation # 17.2 % tma_retirin= g (49.95%) TopdownL1 (cpu_atom) # 30.6 % tma_backend= _bound # 29.8 % tma_fronten= d_bound (49.94%) (cpu_atom) # 6.9 K/sec branch_= frequency (42.89%) # 80.5 K/sec branch_= frequency (74.93%) # 29.6 migrations/sec= migrations_per_second # 28.0 CPUs CPUs_uti= lized (cpu_atom) # 0.8 instructions = insn_per_cycle (42.91%) # 2.0 instructions = insn_per_cycle (75.14%) (cpu_atom) # 3.8 % branch_miss= _rate (35.75%) # 1.2 % branch_miss= _rate (66.86%) 1.007063529 seconds time elapsed ``` Signed-off-by: Ian Rogers --- .../arch/common/common/metrics.json | 86 +++++++++++++ tools/perf/pmu-events/empty-pmu-events.c | 115 +++++++++++++----- tools/perf/pmu-events/jevents.py | 21 +++- tools/perf/pmu-events/pmu-events.h | 1 + tools/perf/util/metricgroup.c | 31 +++-- 5 files changed, 212 insertions(+), 42 deletions(-) create mode 100644 tools/perf/pmu-events/arch/common/common/metrics.json diff --git a/tools/perf/pmu-events/arch/common/common/metrics.json b/tools/= perf/pmu-events/arch/common/common/metrics.json new file mode 100644 index 000000000000..d1e37db18dc6 --- /dev/null +++ b/tools/perf/pmu-events/arch/common/common/metrics.json @@ -0,0 +1,86 @@ +[ + { + "BriefDescription": "Average CPU utilization", + "MetricExpr": "(software@cpu\\-clock\\,name\\=3Dcpu\\-clock@ if #t= arget_cpu else software@task\\-clock\\,name\\=3Dtask\\-clock@) / (duration_= time * 1e9)", + "MetricGroup": "Default", + "MetricName": "CPUs_utilized", + "ScaleUnit": "1CPUs", + "MetricConstraint": "NO_GROUP_EVENTS" + }, + { + "BriefDescription": "Context switches per CPU second", + "MetricExpr": "(software@context\\-switches\\,name\\=3Dcontext\\-s= witches@ * 1e9) / (software@cpu\\-clock\\,name\\=3Dcpu\\-clock@ if #target_= cpu else software@task\\-clock\\,name\\=3Dtask\\-clock@)", + "MetricGroup": "Default", + "MetricName": "cs_per_second", + "ScaleUnit": "1cs/sec", + "MetricConstraint": "NO_GROUP_EVENTS" + }, + { + "BriefDescription": "Process migrations to a new CPU per CPU secon= d", + "MetricExpr": "(software@cpu\\-migrations\\,name\\=3Dcpu\\-migrati= ons@ * 1e9) / (software@cpu\\-clock\\,name\\=3Dcpu\\-clock@ if #target_cpu = else software@task\\-clock\\,name\\=3Dtask\\-clock@)", + "MetricGroup": "Default", + "MetricName": "migrations_per_second", + "ScaleUnit": "1migrations/sec", + "MetricConstraint": "NO_GROUP_EVENTS" + }, + { + "BriefDescription": "Page faults per CPU second", + "MetricExpr": "(software@page\\-faults\\,name\\=3Dpage\\-faults@ *= 1e9) / (software@cpu\\-clock\\,name\\=3Dcpu\\-clock@ if #target_cpu else s= oftware@task\\-clock\\,name\\=3Dtask\\-clock@)", + "MetricGroup": "Default", + "MetricName": "page_faults_per_second", + "ScaleUnit": "1faults/sec", + "MetricConstraint": "NO_GROUP_EVENTS" + }, + { + "BriefDescription": "Instructions Per Cycle", + "MetricExpr": "instructions / cpu\\-cycles", + "MetricGroup": "Default", + "MetricName": "insn_per_cycle", + "MetricThreshold": "insn_per_cycle < 1", + "ScaleUnit": "1instructions" + }, + { + "BriefDescription": "Max front or backend stalls per instruction", + "MetricExpr": "max(stalled\\-cycles\\-frontend, stalled\\-cycles\\= -backend) / instructions", + "MetricGroup": "Default", + "MetricName": "stalled_cycles_per_instruction" + }, + { + "BriefDescription": "Frontend stalls per cycle", + "MetricExpr": "stalled\\-cycles\\-frontend / cpu\\-cycles", + "MetricGroup": "Default", + "MetricName": "frontend_cycles_idle", + "MetricThreshold": "frontend_cycles_idle > 0.1" + }, + { + "BriefDescription": "Backend stalls per cycle", + "MetricExpr": "stalled\\-cycles\\-backend / cpu\\-cycles", + "MetricGroup": "Default", + "MetricName": "backend_cycles_idle", + "MetricThreshold": "backend_cycles_idle > 0.2" + }, + { + "BriefDescription": "Cycles per CPU second", + "MetricExpr": "cpu\\-cycles / (software@cpu\\-clock\\,name\\=3Dcpu= \\-clock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtask\\-clock= @)", + "MetricGroup": "Default", + "MetricName": "cycles_frequency", + "ScaleUnit": "1GHz", + "MetricConstraint": "NO_GROUP_EVENTS" + }, + { + "BriefDescription": "Branches per CPU second", + "MetricExpr": "branches / (software@cpu\\-clock\\,name\\=3Dcpu\\-c= lock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtask\\-clock@)", + "MetricGroup": "Default", + "MetricName": "branch_frequency", + "ScaleUnit": "1000K/sec", + "MetricConstraint": "NO_GROUP_EVENTS" + }, + { + "BriefDescription": "Branch miss rate", + "MetricExpr": "branch\\-misses / branches", + "MetricGroup": "Default", + "MetricName": "branch_miss_rate", + "MetricThreshold": "branch_miss_rate > 0.05", + "ScaleUnit": "100%" + } +] diff --git a/tools/perf/pmu-events/empty-pmu-events.c b/tools/perf/pmu-even= ts/empty-pmu-events.c index 2fdf4fbf36e2..d78c1857ac4b 100644 --- a/tools/perf/pmu-events/empty-pmu-events.c +++ b/tools/perf/pmu-events/empty-pmu-events.c @@ -1303,21 +1303,32 @@ static const char *const big_c_string =3D /* offset=3D127519 */ "sys_ccn_pmu.read_cycles\000uncore\000ccn read-cycle= s event\000config=3D0x2c\0000x01\00000\000\000\000\000\000" /* offset=3D127596 */ "uncore_sys_cmn_pmu\000" /* offset=3D127615 */ "sys_cmn_pmu.hnf_cache_miss\000uncore\000Counts tota= l cache misses in first lookup result (high priority)\000eventid=3D1,type= =3D5\000(434|436|43c|43a).*\00000\000\000\000\000\000" -/* offset=3D127758 */ "CPI\000\0001 / IPC\000\000\000\000\000\000\000\0000= 0" -/* offset=3D127780 */ "IPC\000group1\000inst_retired.any / cpu_clk_unhalte= d.thread\000\000\000\000\000\000\000\00000" -/* offset=3D127843 */ "Frontend_Bound_SMT\000\000idq_uops_not_delivered.co= re / (4 * (cpu_clk_unhalted.thread / 2 * (1 + cpu_clk_unhalted.one_thread_a= ctive / cpu_clk_unhalted.ref_xclk)))\000\000\000\000\000\000\000\00000" -/* offset=3D128009 */ "dcache_miss_cpi\000\000l1d\\-loads\\-misses / inst_= retired.any\000\000\000\000\000\000\000\00000" -/* offset=3D128073 */ "icache_miss_cycles\000\000l1i\\-loads\\-misses / in= st_retired.any\000\000\000\000\000\000\000\00000" -/* offset=3D128140 */ "cache_miss_cycles\000group1\000dcache_miss_cpi + ic= ache_miss_cycles\000\000\000\000\000\000\000\00000" -/* offset=3D128211 */ "DCache_L2_All_Hits\000\000l2_rqsts.demand_data_rd_h= it + l2_rqsts.pf_hit + l2_rqsts.rfo_hit\000\000\000\000\000\000\000\00000" -/* offset=3D128305 */ "DCache_L2_All_Miss\000\000max(l2_rqsts.all_demand_d= ata_rd - l2_rqsts.demand_data_rd_hit, 0) + l2_rqsts.pf_miss + l2_rqsts.rfo_= miss\000\000\000\000\000\000\000\00000" -/* offset=3D128439 */ "DCache_L2_All\000\000DCache_L2_All_Hits + DCache_L2= _All_Miss\000\000\000\000\000\000\000\00000" -/* offset=3D128503 */ "DCache_L2_Hits\000\000d_ratio(DCache_L2_All_Hits, D= Cache_L2_All)\000\000\000\000\000\000\000\00000" -/* offset=3D128571 */ "DCache_L2_Misses\000\000d_ratio(DCache_L2_All_Miss,= DCache_L2_All)\000\000\000\000\000\000\000\00000" -/* offset=3D128641 */ "M1\000\000ipc + M2\000\000\000\000\000\000\000\0000= 0" -/* offset=3D128663 */ "M2\000\000ipc + M1\000\000\000\000\000\000\000\0000= 0" -/* offset=3D128685 */ "M3\000\0001 / M3\000\000\000\000\000\000\000\00000" -/* offset=3D128705 */ "L1D_Cache_Fill_BW\000\00064 * l1d.replacement / 1e9= / duration_time\000\000\000\000\000\000\000\00000" +/* offset=3D127758 */ "CPUs_utilized\000Default\000(software@cpu\\-clock\\= ,name\\=3Dcpu\\-clock@ if #target_cpu else software@task\\-clock\\,name\\= =3Dtask\\-clock@) / (duration_time * 1e9)\000\000Average CPU utilization\00= 0\0001CPUs\000\000\000\00001" +/* offset=3D127943 */ "cs_per_second\000Default\000software@context\\-swit= ches\\,name\\=3Dcontext\\-switches@ * 1e9 / (software@cpu\\-clock\\,name\\= =3Dcpu\\-clock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtask\\= -clock@)\000\000Context switches per CPU second\000\0001cs/sec\000\000\000\= 00001" +/* offset=3D128175 */ "migrations_per_second\000Default\000software@cpu\\-= migrations\\,name\\=3Dcpu\\-migrations@ * 1e9 / (software@cpu\\-clock\\,nam= e\\=3Dcpu\\-clock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtas= k\\-clock@)\000\000Process migrations to a new CPU per CPU second\000\0001m= igrations/sec\000\000\000\00001" +/* offset=3D128434 */ "page_faults_per_second\000Default\000software@page\= \-faults\\,name\\=3Dpage\\-faults@ * 1e9 / (software@cpu\\-clock\\,name\\= =3Dcpu\\-clock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtask\\= -clock@)\000\000Page faults per CPU second\000\0001faults/sec\000\000\000\0= 0001" +/* offset=3D128664 */ "insn_per_cycle\000Default\000instructions / cpu\\-c= ycles\000insn_per_cycle < 1\000Instructions Per Cycle\000\0001instructions\= 000\000\000\00000" +/* offset=3D128776 */ "stalled_cycles_per_instruction\000Default\000max(st= alled\\-cycles\\-frontend, stalled\\-cycles\\-backend) / instructions\000\0= 00Max front or backend stalls per instruction\000\000\000\000\000\00000" +/* offset=3D128939 */ "frontend_cycles_idle\000Default\000stalled\\-cycles= \\-frontend / cpu\\-cycles\000frontend_cycles_idle > 0.1\000Frontend stalls= per cycle\000\000\000\000\000\00000" +/* offset=3D129068 */ "backend_cycles_idle\000Default\000stalled\\-cycles\= \-backend / cpu\\-cycles\000backend_cycles_idle > 0.2\000Backend stalls per= cycle\000\000\000\000\000\00000" +/* offset=3D129193 */ "cycles_frequency\000Default\000cpu\\-cycles / (soft= ware@cpu\\-clock\\,name\\=3Dcpu\\-clock@ if #target_cpu else software@task\= \-clock\\,name\\=3Dtask\\-clock@)\000\000Cycles per CPU second\000\0001GHz\= 000\000\000\00001" +/* offset=3D129368 */ "branch_frequency\000Default\000branches / (software= @cpu\\-clock\\,name\\=3Dcpu\\-clock@ if #target_cpu else software@task\\-cl= ock\\,name\\=3Dtask\\-clock@)\000\000Branches per CPU second\000\0001000K/s= ec\000\000\000\00001" +/* offset=3D129547 */ "branch_miss_rate\000Default\000branch\\-misses / br= anches\000branch_miss_rate > 0.05\000Branch miss rate\000\000100%\000\000\0= 00\00000" +/* offset=3D129650 */ "CPI\000\0001 / IPC\000\000\000\000\000\000\000\0000= 0" +/* offset=3D129672 */ "IPC\000group1\000inst_retired.any / cpu_clk_unhalte= d.thread\000\000\000\000\000\000\000\00000" +/* offset=3D129735 */ "Frontend_Bound_SMT\000\000idq_uops_not_delivered.co= re / (4 * (cpu_clk_unhalted.thread / 2 * (1 + cpu_clk_unhalted.one_thread_a= ctive / cpu_clk_unhalted.ref_xclk)))\000\000\000\000\000\000\000\00000" +/* offset=3D129901 */ "dcache_miss_cpi\000\000l1d\\-loads\\-misses / inst_= retired.any\000\000\000\000\000\000\000\00000" +/* offset=3D129965 */ "icache_miss_cycles\000\000l1i\\-loads\\-misses / in= st_retired.any\000\000\000\000\000\000\000\00000" +/* offset=3D130032 */ "cache_miss_cycles\000group1\000dcache_miss_cpi + ic= ache_miss_cycles\000\000\000\000\000\000\000\00000" +/* offset=3D130103 */ "DCache_L2_All_Hits\000\000l2_rqsts.demand_data_rd_h= it + l2_rqsts.pf_hit + l2_rqsts.rfo_hit\000\000\000\000\000\000\000\00000" +/* offset=3D130197 */ "DCache_L2_All_Miss\000\000max(l2_rqsts.all_demand_d= ata_rd - l2_rqsts.demand_data_rd_hit, 0) + l2_rqsts.pf_miss + l2_rqsts.rfo_= miss\000\000\000\000\000\000\000\00000" +/* offset=3D130331 */ "DCache_L2_All\000\000DCache_L2_All_Hits + DCache_L2= _All_Miss\000\000\000\000\000\000\000\00000" +/* offset=3D130395 */ "DCache_L2_Hits\000\000d_ratio(DCache_L2_All_Hits, D= Cache_L2_All)\000\000\000\000\000\000\000\00000" +/* offset=3D130463 */ "DCache_L2_Misses\000\000d_ratio(DCache_L2_All_Miss,= DCache_L2_All)\000\000\000\000\000\000\000\00000" +/* offset=3D130533 */ "M1\000\000ipc + M2\000\000\000\000\000\000\000\0000= 0" +/* offset=3D130555 */ "M2\000\000ipc + M1\000\000\000\000\000\000\000\0000= 0" +/* offset=3D130577 */ "M3\000\0001 / M3\000\000\000\000\000\000\000\00000" +/* offset=3D130597 */ "L1D_Cache_Fill_BW\000\00064 * l1d.replacement / 1e9= / duration_time\000\000\000\000\000\000\000\00000" ; =20 static const struct compact_pmu_event pmu_events__common_default_core[] = =3D { @@ -2603,6 +2614,29 @@ static const struct pmu_table_entry pmu_events__comm= on[] =3D { }, }; =20 +static const struct compact_pmu_event pmu_metrics__common_default_core[] = =3D { +{ 127758 }, /* CPUs_utilized\000Default\000(software@cpu\\-clock\\,name\\= =3Dcpu\\-clock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtask\\= -clock@) / (duration_time * 1e9)\000\000Average CPU utilization\000\0001CPU= s\000\000\000\00001 */ +{ 129068 }, /* backend_cycles_idle\000Default\000stalled\\-cycles\\-backen= d / cpu\\-cycles\000backend_cycles_idle > 0.2\000Backend stalls per cycle\0= 00\000\000\000\000\00000 */ +{ 129368 }, /* branch_frequency\000Default\000branches / (software@cpu\\-c= lock\\,name\\=3Dcpu\\-clock@ if #target_cpu else software@task\\-clock\\,na= me\\=3Dtask\\-clock@)\000\000Branches per CPU second\000\0001000K/sec\000\0= 00\000\00001 */ +{ 129547 }, /* branch_miss_rate\000Default\000branch\\-misses / branches\0= 00branch_miss_rate > 0.05\000Branch miss rate\000\000100%\000\000\000\00000= */ +{ 127943 }, /* cs_per_second\000Default\000software@context\\-switches\\,n= ame\\=3Dcontext\\-switches@ * 1e9 / (software@cpu\\-clock\\,name\\=3Dcpu\\-= clock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtask\\-clock@)\= 000\000Context switches per CPU second\000\0001cs/sec\000\000\000\00001 */ +{ 129193 }, /* cycles_frequency\000Default\000cpu\\-cycles / (software@cpu= \\-clock\\,name\\=3Dcpu\\-clock@ if #target_cpu else software@task\\-clock\= \,name\\=3Dtask\\-clock@)\000\000Cycles per CPU second\000\0001GHz\000\000\= 000\00001 */ +{ 128939 }, /* frontend_cycles_idle\000Default\000stalled\\-cycles\\-front= end / cpu\\-cycles\000frontend_cycles_idle > 0.1\000Frontend stalls per cyc= le\000\000\000\000\000\00000 */ +{ 128664 }, /* insn_per_cycle\000Default\000instructions / cpu\\-cycles\00= 0insn_per_cycle < 1\000Instructions Per Cycle\000\0001instructions\000\000\= 000\00000 */ +{ 128175 }, /* migrations_per_second\000Default\000software@cpu\\-migratio= ns\\,name\\=3Dcpu\\-migrations@ * 1e9 / (software@cpu\\-clock\\,name\\=3Dcp= u\\-clock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtask\\-cloc= k@)\000\000Process migrations to a new CPU per CPU second\000\0001migration= s/sec\000\000\000\00001 */ +{ 128434 }, /* page_faults_per_second\000Default\000software@page\\-faults= \\,name\\=3Dpage\\-faults@ * 1e9 / (software@cpu\\-clock\\,name\\=3Dcpu\\-c= lock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtask\\-clock@)\0= 00\000Page faults per CPU second\000\0001faults/sec\000\000\000\00001 */ +{ 128776 }, /* stalled_cycles_per_instruction\000Default\000max(stalled\\-= cycles\\-frontend, stalled\\-cycles\\-backend) / instructions\000\000Max fr= ont or backend stalls per instruction\000\000\000\000\000\00000 */ + +}; + +static const struct pmu_table_entry pmu_metrics__common[] =3D { +{ + .entries =3D pmu_metrics__common_default_core, + .num_entries =3D ARRAY_SIZE(pmu_metrics__common_default_core), + .pmu_name =3D { 0 /* default_core\000 */ }, +}, +}; + static const struct compact_pmu_event pmu_events__test_soc_cpu_default_cor= e[] =3D { { 126205 }, /* bp_l1_btb_correct\000branch\000L1 BTB Correction\000event= =3D0x8a\000\00000\000\000\000\000\000 */ { 126267 }, /* bp_l2_btb_correct\000branch\000L2 BTB Correction\000event= =3D0x8b\000\00000\000\000\000\000\000 */ @@ -2664,21 +2698,21 @@ static const struct pmu_table_entry pmu_events__tes= t_soc_cpu[] =3D { }; =20 static const struct compact_pmu_event pmu_metrics__test_soc_cpu_default_co= re[] =3D { -{ 127758 }, /* CPI\000\0001 / IPC\000\000\000\000\000\000\000\00000 */ -{ 128439 }, /* DCache_L2_All\000\000DCache_L2_All_Hits + DCache_L2_All_Mis= s\000\000\000\000\000\000\000\00000 */ -{ 128211 }, /* DCache_L2_All_Hits\000\000l2_rqsts.demand_data_rd_hit + l2_= rqsts.pf_hit + l2_rqsts.rfo_hit\000\000\000\000\000\000\000\00000 */ -{ 128305 }, /* DCache_L2_All_Miss\000\000max(l2_rqsts.all_demand_data_rd -= l2_rqsts.demand_data_rd_hit, 0) + l2_rqsts.pf_miss + l2_rqsts.rfo_miss\000= \000\000\000\000\000\000\00000 */ -{ 128503 }, /* DCache_L2_Hits\000\000d_ratio(DCache_L2_All_Hits, DCache_L2= _All)\000\000\000\000\000\000\000\00000 */ -{ 128571 }, /* DCache_L2_Misses\000\000d_ratio(DCache_L2_All_Miss, DCache_= L2_All)\000\000\000\000\000\000\000\00000 */ -{ 127843 }, /* Frontend_Bound_SMT\000\000idq_uops_not_delivered.core / (4 = * (cpu_clk_unhalted.thread / 2 * (1 + cpu_clk_unhalted.one_thread_active / = cpu_clk_unhalted.ref_xclk)))\000\000\000\000\000\000\000\00000 */ -{ 127780 }, /* IPC\000group1\000inst_retired.any / cpu_clk_unhalted.thread= \000\000\000\000\000\000\000\00000 */ -{ 128705 }, /* L1D_Cache_Fill_BW\000\00064 * l1d.replacement / 1e9 / durat= ion_time\000\000\000\000\000\000\000\00000 */ -{ 128641 }, /* M1\000\000ipc + M2\000\000\000\000\000\000\000\00000 */ -{ 128663 }, /* M2\000\000ipc + M1\000\000\000\000\000\000\000\00000 */ -{ 128685 }, /* M3\000\0001 / M3\000\000\000\000\000\000\000\00000 */ -{ 128140 }, /* cache_miss_cycles\000group1\000dcache_miss_cpi + icache_mis= s_cycles\000\000\000\000\000\000\000\00000 */ -{ 128009 }, /* dcache_miss_cpi\000\000l1d\\-loads\\-misses / inst_retired.= any\000\000\000\000\000\000\000\00000 */ -{ 128073 }, /* icache_miss_cycles\000\000l1i\\-loads\\-misses / inst_retir= ed.any\000\000\000\000\000\000\000\00000 */ +{ 129650 }, /* CPI\000\0001 / IPC\000\000\000\000\000\000\000\00000 */ +{ 130331 }, /* DCache_L2_All\000\000DCache_L2_All_Hits + DCache_L2_All_Mis= s\000\000\000\000\000\000\000\00000 */ +{ 130103 }, /* DCache_L2_All_Hits\000\000l2_rqsts.demand_data_rd_hit + l2_= rqsts.pf_hit + l2_rqsts.rfo_hit\000\000\000\000\000\000\000\00000 */ +{ 130197 }, /* DCache_L2_All_Miss\000\000max(l2_rqsts.all_demand_data_rd -= l2_rqsts.demand_data_rd_hit, 0) + l2_rqsts.pf_miss + l2_rqsts.rfo_miss\000= \000\000\000\000\000\000\00000 */ +{ 130395 }, /* DCache_L2_Hits\000\000d_ratio(DCache_L2_All_Hits, DCache_L2= _All)\000\000\000\000\000\000\000\00000 */ +{ 130463 }, /* DCache_L2_Misses\000\000d_ratio(DCache_L2_All_Miss, DCache_= L2_All)\000\000\000\000\000\000\000\00000 */ +{ 129735 }, /* Frontend_Bound_SMT\000\000idq_uops_not_delivered.core / (4 = * (cpu_clk_unhalted.thread / 2 * (1 + cpu_clk_unhalted.one_thread_active / = cpu_clk_unhalted.ref_xclk)))\000\000\000\000\000\000\000\00000 */ +{ 129672 }, /* IPC\000group1\000inst_retired.any / cpu_clk_unhalted.thread= \000\000\000\000\000\000\000\00000 */ +{ 130597 }, /* L1D_Cache_Fill_BW\000\00064 * l1d.replacement / 1e9 / durat= ion_time\000\000\000\000\000\000\000\00000 */ +{ 130533 }, /* M1\000\000ipc + M2\000\000\000\000\000\000\000\00000 */ +{ 130555 }, /* M2\000\000ipc + M1\000\000\000\000\000\000\000\00000 */ +{ 130577 }, /* M3\000\0001 / M3\000\000\000\000\000\000\000\00000 */ +{ 130032 }, /* cache_miss_cycles\000group1\000dcache_miss_cpi + icache_mis= s_cycles\000\000\000\000\000\000\000\00000 */ +{ 129901 }, /* dcache_miss_cpi\000\000l1d\\-loads\\-misses / inst_retired.= any\000\000\000\000\000\000\000\00000 */ +{ 129965 }, /* icache_miss_cycles\000\000l1i\\-loads\\-misses / inst_retir= ed.any\000\000\000\000\000\000\000\00000 */ =20 }; =20 @@ -2759,7 +2793,10 @@ static const struct pmu_events_map pmu_events_map[] = =3D { .pmus =3D pmu_events__common, .num_pmus =3D ARRAY_SIZE(pmu_events__common), }, - .metric_table =3D {}, + .metric_table =3D { + .pmus =3D pmu_metrics__common, + .num_pmus =3D ARRAY_SIZE(pmu_metrics__common), + }, }, { .arch =3D "testarch", @@ -3208,6 +3245,22 @@ const struct pmu_metrics_table *pmu_metrics_table__f= ind(void) return map ? &map->metric_table : NULL; } =20 +const struct pmu_metrics_table *pmu_metrics_table__default(void) +{ + int i =3D 0; + + for (;;) { + const struct pmu_events_map *map =3D &pmu_events_map[i++]; + + if (!map->arch) + break; + + if (!strcmp(map->cpuid, "common")) + return &map->metric_table; + } + return NULL; +} + const struct pmu_events_table *find_core_events_table(const char *arch, co= nst char *cpuid) { for (const struct pmu_events_map *tables =3D &pmu_events_map[0]; diff --git a/tools/perf/pmu-events/jevents.py b/tools/perf/pmu-events/jeven= ts.py index 786a7049363f..5d3f4b44cfb7 100755 --- a/tools/perf/pmu-events/jevents.py +++ b/tools/perf/pmu-events/jevents.py @@ -755,7 +755,10 @@ static const struct pmu_events_map pmu_events_map[] = =3D { \t\t.pmus =3D pmu_events__common, \t\t.num_pmus =3D ARRAY_SIZE(pmu_events__common), \t}, -\t.metric_table =3D {}, +\t.metric_table =3D { +\t\t.pmus =3D pmu_metrics__common, +\t\t.num_pmus =3D ARRAY_SIZE(pmu_metrics__common), +\t}, }, """) else: @@ -1237,6 +1240,22 @@ const struct pmu_metrics_table *pmu_metrics_table__f= ind(void) return map ? &map->metric_table : NULL; } =20 +const struct pmu_metrics_table *pmu_metrics_table__default(void) +{ + int i =3D 0; + + for (;;) { + const struct pmu_events_map *map =3D &pmu_events_map[i++]; + + if (!map->arch) + break; + + if (!strcmp(map->cpuid, "common")) + return &map->metric_table; + } + return NULL; +} + const struct pmu_events_table *find_core_events_table(const char *arch, co= nst char *cpuid) { for (const struct pmu_events_map *tables =3D &pmu_events_map[0]; diff --git a/tools/perf/pmu-events/pmu-events.h b/tools/perf/pmu-events/pmu= -events.h index e0535380c0b2..559265a903c8 100644 --- a/tools/perf/pmu-events/pmu-events.h +++ b/tools/perf/pmu-events/pmu-events.h @@ -127,6 +127,7 @@ int pmu_metrics_table__find_metric(const struct pmu_met= rics_table *table, const struct pmu_events_table *perf_pmu__find_events_table(struct perf_pmu= *pmu); const struct pmu_events_table *perf_pmu__default_core_events_table(void); const struct pmu_metrics_table *pmu_metrics_table__find(void); +const struct pmu_metrics_table *pmu_metrics_table__default(void); const struct pmu_events_table *find_core_events_table(const char *arch, co= nst char *cpuid); const struct pmu_metrics_table *find_core_metrics_table(const char *arch, = const char *cpuid); int pmu_for_each_core_event(pmu_event_iter_fn fn, void *data); diff --git a/tools/perf/util/metricgroup.c b/tools/perf/util/metricgroup.c index 76092ee26761..e67e04ce01c9 100644 --- a/tools/perf/util/metricgroup.c +++ b/tools/perf/util/metricgroup.c @@ -424,10 +424,18 @@ int metricgroup__for_each_metric(const struct pmu_met= rics_table *table, pmu_metr .fn =3D fn, .data =3D data, }; + const struct pmu_metrics_table *tables[2] =3D { + table, + pmu_metrics_table__default(), + }; + + for (size_t i =3D 0; i < ARRAY_SIZE(tables); i++) { + int ret; =20 - if (table) { - int ret =3D pmu_metrics_table__for_each_metric(table, fn, data); + if (!tables[i]) + continue; =20 + ret =3D pmu_metrics_table__for_each_metric(tables[i], fn, data); if (ret) return ret; } @@ -1581,19 +1589,22 @@ static int metricgroup__has_metric_or_groups_callba= ck(const struct pmu_metric *p =20 bool metricgroup__has_metric_or_groups(const char *pmu, const char *metric= _or_groups) { - const struct pmu_metrics_table *table =3D pmu_metrics_table__find(); + const struct pmu_metrics_table *tables[2] =3D { + pmu_metrics_table__find(), + pmu_metrics_table__default(), + }; struct metricgroup__has_metric_data data =3D { .pmu =3D pmu, .metric_or_groups =3D metric_or_groups, }; =20 - if (!table) - return false; - - return pmu_metrics_table__for_each_metric(table, - metricgroup__has_metric_or_groups_callback, - &data) - ? true : false; + for (size_t i =3D 0; i < ARRAY_SIZE(tables); i++) { + if (pmu_metrics_table__for_each_metric(tables[i], + metricgroup__has_metric_or_groups_callback, + &data)) + return true; + } + return false; } =20 static int metricgroup__topdown_max_level_callback(const struct pmu_metric= *pm, --=20 2.51.2.1041.gc1ab5b90ca-goog From nobody Tue Nov 11 11:28:57 2025 Received: from mail-pl1-f201.google.com (mail-pl1-f201.google.com [209.85.214.201]) (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 DE9C834AB16 for ; Tue, 11 Nov 2025 04:04:33 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.214.201 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1762833876; cv=none; b=no1ZEb7eMiBL+p39bKv7jhe4+dnvzQ2VzLDP16uKWnryo/a3+0X/27z6ar1o85FSCE9FPlYq3XzeVuEksSBdJf2q8tufppqsapZDuWTccE7J3iem8hD78Se3yuQcRsEnKTw2QpMXTWKu3n3Q8bwvLc4Osq+JVE8FP62C7gRg/kc= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1762833876; c=relaxed/simple; bh=LyoH6LXIxZIl0HCqMImAAJVwoqBr+2LrDfoLTLGcA7E=; h=Date:In-Reply-To:Mime-Version:References:Message-ID:Subject:From: To:Content-Type; b=YrQlZtu44nMKsLqwAmgyU+V2rXcuCaLbMVPTXYGvFSdhBcbfPrwFA9ahTjV6GgKfIj21yQj3UZOL9xPs/7pzFFvAV+6dVeXX28oy4Fh0N6ORt8mSFyIVQcYtU1Ls97NUVlt1FCQFaYGCaU2lomtefRhERYmQ+AEa+gLHpvH2uWg= 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=Kgiv3LFe; arc=none smtp.client-ip=209.85.214.201 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="Kgiv3LFe" Received: by mail-pl1-f201.google.com with SMTP id d9443c01a7336-297dfae179bso81290745ad.1 for ; Mon, 10 Nov 2025 20:04:33 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1762833873; x=1763438673; darn=vger.kernel.org; h=content-transfer-encoding:to:from:subject:message-id:references :mime-version:in-reply-to:date:from:to:cc:subject:date:message-id :reply-to; bh=EQCAarvbxj6k3e4I9VkNijQc2u+whr0zkBwCpCWRbqE=; b=Kgiv3LFetKEiVbTh5KCbCDHdvSdOtp/zs1A435nPEhmS3W1RU23HiV2v0em8ZA0SIP xluinICRs1zReTgIRoubIYx9a8bf1DxRi+jhkgnTp2T0j+QNJRq+NLhCKmYeWfP7IK1+ 2kdSU7z327sZ5rvWeZy62J6v9FwCUeLKSVhVHUklImV88GscMtWq+MiW0fShm3XwgSxo yQbrL3mwQj7+opXizCOR3ixTl43qqB3Q9vVtBeLIub84X1liD5OFQR7mT8dzXW/LnVI5 02yj2A30WyLRM+oPCx1RdCo2W5K5mxMk6tDu4r9cbfLbFx7P/eSt6F6IHp9wAE6DcvuE stOQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1762833873; x=1763438673; h=content-transfer-encoding:to:from:subject:message-id:references :mime-version:in-reply-to:date:x-gm-message-state:from:to:cc:subject :date:message-id:reply-to; bh=EQCAarvbxj6k3e4I9VkNijQc2u+whr0zkBwCpCWRbqE=; b=Oavia41ArBAYRYAc1/pRC+AON2IWozPT0PYQD22LI4uXxZ+AZyzAr7Bpz9NjBQRxFn UV00AhnqvtMWaTd8aj4eTYFM357I7LvoyJ6nwCo0H+LoSYBtszvKizsvs2VOyCE1Rp4N l3rookLhCOhjWg2eNSHLX9vOUHp+gkzthjfyvafL1OrmSJCDgsm04ywUBsKPIvza5NQt C2Tc/dpalUCJAVpYFghWxdu4vJ+Dkb/O0t6O51nqdUgPGFc635GLwxbaMcowcAvEqLrh Ay54uPJP2hdBACHQqTXOCnJeMbVQnzk8BWRv/ItVX60gRtQ6BeU2g9czp2oX+rPS1+84 YAHg== X-Forwarded-Encrypted: i=1; AJvYcCVLlchq1f8+uQjdFCFnOm4oyDzNsPPlyCVv+LW/v4aAtuwcvRqEstE7CDlm/TzkAku6lSzul8rS1I03AGM=@vger.kernel.org X-Gm-Message-State: AOJu0Ywb4oRkxdiqxT+cogTWV9GZswHfNbsIKcrcHZvtNGYk1IYS923d nOuMzTon31xGsBPIp6oytIg8kzWfzNsRh7cuSGWgjiSBw3Lam1pBVWvpsPwl9u7ILFxM24/rZOt f5vtUkN/ShA== X-Google-Smtp-Source: AGHT+IHksNz9vjS6mUWIxzWSDQWhSVj7x8jcPk5okM+B8flbbVp357GUEEP7kH7yio1Lbm+uDq/r+NgkxFNu X-Received: from dybgj5.prod.google.com ([2002:a05:7301:105:b0:2a2:4eb1:3771]) (user=irogers job=prod-delivery.src-stubby-dispatcher) by 2002:a17:902:ce0c:b0:298:3aa6:c034 with SMTP id d9443c01a7336-2983aa6c13dmr39076005ad.32.1762833873196; Mon, 10 Nov 2025 20:04:33 -0800 (PST) Date: Mon, 10 Nov 2025 20:04:03 -0800 In-Reply-To: <20251111040417.270945-1-irogers@google.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 References: <20251111040417.270945-1-irogers@google.com> X-Mailer: git-send-email 2.51.2.1041.gc1ab5b90ca-goog Message-ID: <20251111040417.270945-5-irogers@google.com> Subject: [PATCH v3 04/18] perf jevents: Add metric DefaultShowEvents From: Ian Rogers To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Namhyung Kim , Alexander Shishkin , Jiri Olsa , Ian Rogers , Adrian Hunter , James Clark , Xu Yang , Chun-Tse Shao , Thomas Richter , Sumanth Korikkar , Collin Funk , Thomas Falcon , Howard Chu , Dapeng Mi , Levi Yun , Yang Li , linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org, Andi Kleen , Weilin Wang Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" Some Default group metrics require their events showing for consistency with perf's previous behavior. Add a flag to indicate when this is the case and use it in stat-display. As events are coming from Default metrics remove that default hardware and software events from perf stat. Following this change the default perf stat output on an alderlake looks li= ke: ``` $ perf stat -a -- sleep 1 Performance counter stats for 'system wide': 20,759 context-switches # 735.7 cs/sec = cs_per_second TopdownL1 (cpu_core) # 7.8 % tma_bad_spe= culation # 34.8 % tma_fronten= d_bound TopdownL1 (cpu_core) # 39.0 % tma_backend= _bound # 18.4 % tma_retiring 769 page-faults # 27.3 faults/= sec page_faults_per_second 531,102,439 cpu_atom/cpu-cycles/ # 0.0 GHz cy= cles_frequency (49.80%) 785,144,850 cpu_core/cpu-cycles/ # 0.0 GHz cy= cles_frequency # 17.6 % tma_bad_spe= culation # 14.4 % tma_retirin= g (50.20%) # 37.0 % tma_backend= _bound # 31.0 % tma_fronten= d_bound (50.37%) 47,631,924 cpu_atom/branches/ # 1.7 K/sec = branch_frequency (60.31%) 138,036,825 cpu_core/branches/ # 4.9 K/sec = branch_frequency 779 cpu-migrations # 27.6 migrati= ons/sec migrations_per_second 28,218,162,085 cpu-clock # 28.0 CPUs C= PUs_utilized 522,230,152 cpu_atom/cpu-cycles/ # 0.5 instruc= tions insn_per_cycle (60.12%) 785,133,103 cpu_core/cpu-cycles/ # 1.0 instruc= tions insn_per_cycle 2,541,997 cpu_atom/branch-misses/ # 5.5 % bran= ch_miss_rate (49.63%) 3,106,064 cpu_core/branch-misses/ # 2.3 % bran= ch_miss_rate 1.007489028 seconds time elapsed ``` Signed-off-by: Ian Rogers --- tools/perf/builtin-stat.c | 42 +------ .../arch/common/common/metrics.json | 33 ++++-- tools/perf/pmu-events/empty-pmu-events.c | 106 +++++++++--------- tools/perf/pmu-events/jevents.py | 7 +- tools/perf/pmu-events/pmu-events.h | 1 + tools/perf/util/evsel.h | 1 + tools/perf/util/metricgroup.c | 13 +++ tools/perf/util/stat-display.c | 4 +- tools/perf/util/stat-shadow.c | 2 +- 9 files changed, 102 insertions(+), 107 deletions(-) diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c index 3c46b92a53ab..31c762695d4b 100644 --- a/tools/perf/builtin-stat.c +++ b/tools/perf/builtin-stat.c @@ -1857,16 +1857,6 @@ static int perf_stat_init_aggr_mode_file(struct perf= _stat *st) return 0; } =20 -/* Add given software event to evlist without wildcarding. */ -static int parse_software_event(struct evlist *evlist, const char *event, - struct parse_events_error *err) -{ - char buf[256]; - - snprintf(buf, sizeof(buf), "software/%s,name=3D%s/", event, event); - return parse_events(evlist, buf, err); -} - /* Add legacy hardware/hardware-cache event to evlist for all core PMUs wi= thout wildcarding. */ static int parse_hardware_event(struct evlist *evlist, const char *event, struct parse_events_error *err) @@ -2011,36 +2001,10 @@ static int add_default_events(void) stat_config.topdown_level =3D 1; =20 if (!evlist->core.nr_entries && !evsel_list->core.nr_entries) { - /* No events so add defaults. */ - const char *sw_events[] =3D { - target__has_cpu(&target) ? "cpu-clock" : "task-clock", - "context-switches", - "cpu-migrations", - "page-faults", - }; - const char *hw_events[] =3D { - "instructions", - "cycles", - "stalled-cycles-frontend", - "stalled-cycles-backend", - "branches", - "branch-misses", - }; - - for (size_t i =3D 0; i < ARRAY_SIZE(sw_events); i++) { - ret =3D parse_software_event(evlist, sw_events[i], &err); - if (ret) - goto out; - } - for (size_t i =3D 0; i < ARRAY_SIZE(hw_events); i++) { - ret =3D parse_hardware_event(evlist, hw_events[i], &err); - if (ret) - goto out; - } - /* - * Add TopdownL1 metrics if they exist. To minimize - * multiplexing, don't request threshold computation. + * Add Default metrics. To minimize multiplexing, don't request + * threshold computation, but it will be computed if the events + * are present. */ if (metricgroup__has_metric_or_groups(pmu, "Default")) { struct evlist *metric_evlist =3D evlist__new(); diff --git a/tools/perf/pmu-events/arch/common/common/metrics.json b/tools/= perf/pmu-events/arch/common/common/metrics.json index d1e37db18dc6..017bbdede3d7 100644 --- a/tools/perf/pmu-events/arch/common/common/metrics.json +++ b/tools/perf/pmu-events/arch/common/common/metrics.json @@ -5,7 +5,8 @@ "MetricGroup": "Default", "MetricName": "CPUs_utilized", "ScaleUnit": "1CPUs", - "MetricConstraint": "NO_GROUP_EVENTS" + "MetricConstraint": "NO_GROUP_EVENTS", + "DefaultShowEvents": "1" }, { "BriefDescription": "Context switches per CPU second", @@ -13,7 +14,8 @@ "MetricGroup": "Default", "MetricName": "cs_per_second", "ScaleUnit": "1cs/sec", - "MetricConstraint": "NO_GROUP_EVENTS" + "MetricConstraint": "NO_GROUP_EVENTS", + "DefaultShowEvents": "1" }, { "BriefDescription": "Process migrations to a new CPU per CPU secon= d", @@ -21,7 +23,8 @@ "MetricGroup": "Default", "MetricName": "migrations_per_second", "ScaleUnit": "1migrations/sec", - "MetricConstraint": "NO_GROUP_EVENTS" + "MetricConstraint": "NO_GROUP_EVENTS", + "DefaultShowEvents": "1" }, { "BriefDescription": "Page faults per CPU second", @@ -29,7 +32,8 @@ "MetricGroup": "Default", "MetricName": "page_faults_per_second", "ScaleUnit": "1faults/sec", - "MetricConstraint": "NO_GROUP_EVENTS" + "MetricConstraint": "NO_GROUP_EVENTS", + "DefaultShowEvents": "1" }, { "BriefDescription": "Instructions Per Cycle", @@ -37,27 +41,31 @@ "MetricGroup": "Default", "MetricName": "insn_per_cycle", "MetricThreshold": "insn_per_cycle < 1", - "ScaleUnit": "1instructions" + "ScaleUnit": "1instructions", + "DefaultShowEvents": "1" }, { "BriefDescription": "Max front or backend stalls per instruction", "MetricExpr": "max(stalled\\-cycles\\-frontend, stalled\\-cycles\\= -backend) / instructions", "MetricGroup": "Default", - "MetricName": "stalled_cycles_per_instruction" + "MetricName": "stalled_cycles_per_instruction", + "DefaultShowEvents": "1" }, { "BriefDescription": "Frontend stalls per cycle", "MetricExpr": "stalled\\-cycles\\-frontend / cpu\\-cycles", "MetricGroup": "Default", "MetricName": "frontend_cycles_idle", - "MetricThreshold": "frontend_cycles_idle > 0.1" + "MetricThreshold": "frontend_cycles_idle > 0.1", + "DefaultShowEvents": "1" }, { "BriefDescription": "Backend stalls per cycle", "MetricExpr": "stalled\\-cycles\\-backend / cpu\\-cycles", "MetricGroup": "Default", "MetricName": "backend_cycles_idle", - "MetricThreshold": "backend_cycles_idle > 0.2" + "MetricThreshold": "backend_cycles_idle > 0.2", + "DefaultShowEvents": "1" }, { "BriefDescription": "Cycles per CPU second", @@ -65,7 +73,8 @@ "MetricGroup": "Default", "MetricName": "cycles_frequency", "ScaleUnit": "1GHz", - "MetricConstraint": "NO_GROUP_EVENTS" + "MetricConstraint": "NO_GROUP_EVENTS", + "DefaultShowEvents": "1" }, { "BriefDescription": "Branches per CPU second", @@ -73,7 +82,8 @@ "MetricGroup": "Default", "MetricName": "branch_frequency", "ScaleUnit": "1000K/sec", - "MetricConstraint": "NO_GROUP_EVENTS" + "MetricConstraint": "NO_GROUP_EVENTS", + "DefaultShowEvents": "1" }, { "BriefDescription": "Branch miss rate", @@ -81,6 +91,7 @@ "MetricGroup": "Default", "MetricName": "branch_miss_rate", "MetricThreshold": "branch_miss_rate > 0.05", - "ScaleUnit": "100%" + "ScaleUnit": "100%", + "DefaultShowEvents": "1" } ] diff --git a/tools/perf/pmu-events/empty-pmu-events.c b/tools/perf/pmu-even= ts/empty-pmu-events.c index d78c1857ac4b..bdf02b667f94 100644 --- a/tools/perf/pmu-events/empty-pmu-events.c +++ b/tools/perf/pmu-events/empty-pmu-events.c @@ -1303,32 +1303,32 @@ static const char *const big_c_string =3D /* offset=3D127519 */ "sys_ccn_pmu.read_cycles\000uncore\000ccn read-cycle= s event\000config=3D0x2c\0000x01\00000\000\000\000\000\000" /* offset=3D127596 */ "uncore_sys_cmn_pmu\000" /* offset=3D127615 */ "sys_cmn_pmu.hnf_cache_miss\000uncore\000Counts tota= l cache misses in first lookup result (high priority)\000eventid=3D1,type= =3D5\000(434|436|43c|43a).*\00000\000\000\000\000\000" -/* offset=3D127758 */ "CPUs_utilized\000Default\000(software@cpu\\-clock\\= ,name\\=3Dcpu\\-clock@ if #target_cpu else software@task\\-clock\\,name\\= =3Dtask\\-clock@) / (duration_time * 1e9)\000\000Average CPU utilization\00= 0\0001CPUs\000\000\000\00001" -/* offset=3D127943 */ "cs_per_second\000Default\000software@context\\-swit= ches\\,name\\=3Dcontext\\-switches@ * 1e9 / (software@cpu\\-clock\\,name\\= =3Dcpu\\-clock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtask\\= -clock@)\000\000Context switches per CPU second\000\0001cs/sec\000\000\000\= 00001" -/* offset=3D128175 */ "migrations_per_second\000Default\000software@cpu\\-= migrations\\,name\\=3Dcpu\\-migrations@ * 1e9 / (software@cpu\\-clock\\,nam= e\\=3Dcpu\\-clock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtas= k\\-clock@)\000\000Process migrations to a new CPU per CPU second\000\0001m= igrations/sec\000\000\000\00001" -/* offset=3D128434 */ "page_faults_per_second\000Default\000software@page\= \-faults\\,name\\=3Dpage\\-faults@ * 1e9 / (software@cpu\\-clock\\,name\\= =3Dcpu\\-clock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtask\\= -clock@)\000\000Page faults per CPU second\000\0001faults/sec\000\000\000\0= 0001" -/* offset=3D128664 */ "insn_per_cycle\000Default\000instructions / cpu\\-c= ycles\000insn_per_cycle < 1\000Instructions Per Cycle\000\0001instructions\= 000\000\000\00000" -/* offset=3D128776 */ "stalled_cycles_per_instruction\000Default\000max(st= alled\\-cycles\\-frontend, stalled\\-cycles\\-backend) / instructions\000\0= 00Max front or backend stalls per instruction\000\000\000\000\000\00000" -/* offset=3D128939 */ "frontend_cycles_idle\000Default\000stalled\\-cycles= \\-frontend / cpu\\-cycles\000frontend_cycles_idle > 0.1\000Frontend stalls= per cycle\000\000\000\000\000\00000" -/* offset=3D129068 */ "backend_cycles_idle\000Default\000stalled\\-cycles\= \-backend / cpu\\-cycles\000backend_cycles_idle > 0.2\000Backend stalls per= cycle\000\000\000\000\000\00000" -/* offset=3D129193 */ "cycles_frequency\000Default\000cpu\\-cycles / (soft= ware@cpu\\-clock\\,name\\=3Dcpu\\-clock@ if #target_cpu else software@task\= \-clock\\,name\\=3Dtask\\-clock@)\000\000Cycles per CPU second\000\0001GHz\= 000\000\000\00001" -/* offset=3D129368 */ "branch_frequency\000Default\000branches / (software= @cpu\\-clock\\,name\\=3Dcpu\\-clock@ if #target_cpu else software@task\\-cl= ock\\,name\\=3Dtask\\-clock@)\000\000Branches per CPU second\000\0001000K/s= ec\000\000\000\00001" -/* offset=3D129547 */ "branch_miss_rate\000Default\000branch\\-misses / br= anches\000branch_miss_rate > 0.05\000Branch miss rate\000\000100%\000\000\0= 00\00000" -/* offset=3D129650 */ "CPI\000\0001 / IPC\000\000\000\000\000\000\000\0000= 0" -/* offset=3D129672 */ "IPC\000group1\000inst_retired.any / cpu_clk_unhalte= d.thread\000\000\000\000\000\000\000\00000" -/* offset=3D129735 */ "Frontend_Bound_SMT\000\000idq_uops_not_delivered.co= re / (4 * (cpu_clk_unhalted.thread / 2 * (1 + cpu_clk_unhalted.one_thread_a= ctive / cpu_clk_unhalted.ref_xclk)))\000\000\000\000\000\000\000\00000" -/* offset=3D129901 */ "dcache_miss_cpi\000\000l1d\\-loads\\-misses / inst_= retired.any\000\000\000\000\000\000\000\00000" -/* offset=3D129965 */ "icache_miss_cycles\000\000l1i\\-loads\\-misses / in= st_retired.any\000\000\000\000\000\000\000\00000" -/* offset=3D130032 */ "cache_miss_cycles\000group1\000dcache_miss_cpi + ic= ache_miss_cycles\000\000\000\000\000\000\000\00000" -/* offset=3D130103 */ "DCache_L2_All_Hits\000\000l2_rqsts.demand_data_rd_h= it + l2_rqsts.pf_hit + l2_rqsts.rfo_hit\000\000\000\000\000\000\000\00000" -/* offset=3D130197 */ "DCache_L2_All_Miss\000\000max(l2_rqsts.all_demand_d= ata_rd - l2_rqsts.demand_data_rd_hit, 0) + l2_rqsts.pf_miss + l2_rqsts.rfo_= miss\000\000\000\000\000\000\000\00000" -/* offset=3D130331 */ "DCache_L2_All\000\000DCache_L2_All_Hits + DCache_L2= _All_Miss\000\000\000\000\000\000\000\00000" -/* offset=3D130395 */ "DCache_L2_Hits\000\000d_ratio(DCache_L2_All_Hits, D= Cache_L2_All)\000\000\000\000\000\000\000\00000" -/* offset=3D130463 */ "DCache_L2_Misses\000\000d_ratio(DCache_L2_All_Miss,= DCache_L2_All)\000\000\000\000\000\000\000\00000" -/* offset=3D130533 */ "M1\000\000ipc + M2\000\000\000\000\000\000\000\0000= 0" -/* offset=3D130555 */ "M2\000\000ipc + M1\000\000\000\000\000\000\000\0000= 0" -/* offset=3D130577 */ "M3\000\0001 / M3\000\000\000\000\000\000\000\00000" -/* offset=3D130597 */ "L1D_Cache_Fill_BW\000\00064 * l1d.replacement / 1e9= / duration_time\000\000\000\000\000\000\000\00000" +/* offset=3D127758 */ "CPUs_utilized\000Default\000(software@cpu\\-clock\\= ,name\\=3Dcpu\\-clock@ if #target_cpu else software@task\\-clock\\,name\\= =3Dtask\\-clock@) / (duration_time * 1e9)\000\000Average CPU utilization\00= 0\0001CPUs\000\000\000\000011" +/* offset=3D127944 */ "cs_per_second\000Default\000software@context\\-swit= ches\\,name\\=3Dcontext\\-switches@ * 1e9 / (software@cpu\\-clock\\,name\\= =3Dcpu\\-clock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtask\\= -clock@)\000\000Context switches per CPU second\000\0001cs/sec\000\000\000\= 000011" +/* offset=3D128177 */ "migrations_per_second\000Default\000software@cpu\\-= migrations\\,name\\=3Dcpu\\-migrations@ * 1e9 / (software@cpu\\-clock\\,nam= e\\=3Dcpu\\-clock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtas= k\\-clock@)\000\000Process migrations to a new CPU per CPU second\000\0001m= igrations/sec\000\000\000\000011" +/* offset=3D128437 */ "page_faults_per_second\000Default\000software@page\= \-faults\\,name\\=3Dpage\\-faults@ * 1e9 / (software@cpu\\-clock\\,name\\= =3Dcpu\\-clock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtask\\= -clock@)\000\000Page faults per CPU second\000\0001faults/sec\000\000\000\0= 00011" +/* offset=3D128668 */ "insn_per_cycle\000Default\000instructions / cpu\\-c= ycles\000insn_per_cycle < 1\000Instructions Per Cycle\000\0001instructions\= 000\000\000\000001" +/* offset=3D128781 */ "stalled_cycles_per_instruction\000Default\000max(st= alled\\-cycles\\-frontend, stalled\\-cycles\\-backend) / instructions\000\0= 00Max front or backend stalls per instruction\000\000\000\000\000\000001" +/* offset=3D128945 */ "frontend_cycles_idle\000Default\000stalled\\-cycles= \\-frontend / cpu\\-cycles\000frontend_cycles_idle > 0.1\000Frontend stalls= per cycle\000\000\000\000\000\000001" +/* offset=3D129075 */ "backend_cycles_idle\000Default\000stalled\\-cycles\= \-backend / cpu\\-cycles\000backend_cycles_idle > 0.2\000Backend stalls per= cycle\000\000\000\000\000\000001" +/* offset=3D129201 */ "cycles_frequency\000Default\000cpu\\-cycles / (soft= ware@cpu\\-clock\\,name\\=3Dcpu\\-clock@ if #target_cpu else software@task\= \-clock\\,name\\=3Dtask\\-clock@)\000\000Cycles per CPU second\000\0001GHz\= 000\000\000\000011" +/* offset=3D129377 */ "branch_frequency\000Default\000branches / (software= @cpu\\-clock\\,name\\=3Dcpu\\-clock@ if #target_cpu else software@task\\-cl= ock\\,name\\=3Dtask\\-clock@)\000\000Branches per CPU second\000\0001000K/s= ec\000\000\000\000011" +/* offset=3D129557 */ "branch_miss_rate\000Default\000branch\\-misses / br= anches\000branch_miss_rate > 0.05\000Branch miss rate\000\000100%\000\000\0= 00\000001" +/* offset=3D129661 */ "CPI\000\0001 / IPC\000\000\000\000\000\000\000\0000= 00" +/* offset=3D129684 */ "IPC\000group1\000inst_retired.any / cpu_clk_unhalte= d.thread\000\000\000\000\000\000\000\000000" +/* offset=3D129748 */ "Frontend_Bound_SMT\000\000idq_uops_not_delivered.co= re / (4 * (cpu_clk_unhalted.thread / 2 * (1 + cpu_clk_unhalted.one_thread_a= ctive / cpu_clk_unhalted.ref_xclk)))\000\000\000\000\000\000\000\000000" +/* offset=3D129915 */ "dcache_miss_cpi\000\000l1d\\-loads\\-misses / inst_= retired.any\000\000\000\000\000\000\000\000000" +/* offset=3D129980 */ "icache_miss_cycles\000\000l1i\\-loads\\-misses / in= st_retired.any\000\000\000\000\000\000\000\000000" +/* offset=3D130048 */ "cache_miss_cycles\000group1\000dcache_miss_cpi + ic= ache_miss_cycles\000\000\000\000\000\000\000\000000" +/* offset=3D130120 */ "DCache_L2_All_Hits\000\000l2_rqsts.demand_data_rd_h= it + l2_rqsts.pf_hit + l2_rqsts.rfo_hit\000\000\000\000\000\000\000\000000" +/* offset=3D130215 */ "DCache_L2_All_Miss\000\000max(l2_rqsts.all_demand_d= ata_rd - l2_rqsts.demand_data_rd_hit, 0) + l2_rqsts.pf_miss + l2_rqsts.rfo_= miss\000\000\000\000\000\000\000\000000" +/* offset=3D130350 */ "DCache_L2_All\000\000DCache_L2_All_Hits + DCache_L2= _All_Miss\000\000\000\000\000\000\000\000000" +/* offset=3D130415 */ "DCache_L2_Hits\000\000d_ratio(DCache_L2_All_Hits, D= Cache_L2_All)\000\000\000\000\000\000\000\000000" +/* offset=3D130484 */ "DCache_L2_Misses\000\000d_ratio(DCache_L2_All_Miss,= DCache_L2_All)\000\000\000\000\000\000\000\000000" +/* offset=3D130555 */ "M1\000\000ipc + M2\000\000\000\000\000\000\000\0000= 00" +/* offset=3D130578 */ "M2\000\000ipc + M1\000\000\000\000\000\000\000\0000= 00" +/* offset=3D130601 */ "M3\000\0001 / M3\000\000\000\000\000\000\000\000000" +/* offset=3D130622 */ "L1D_Cache_Fill_BW\000\00064 * l1d.replacement / 1e9= / duration_time\000\000\000\000\000\000\000\000000" ; =20 static const struct compact_pmu_event pmu_events__common_default_core[] = =3D { @@ -2615,17 +2615,17 @@ static const struct pmu_table_entry pmu_events__com= mon[] =3D { }; =20 static const struct compact_pmu_event pmu_metrics__common_default_core[] = =3D { -{ 127758 }, /* CPUs_utilized\000Default\000(software@cpu\\-clock\\,name\\= =3Dcpu\\-clock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtask\\= -clock@) / (duration_time * 1e9)\000\000Average CPU utilization\000\0001CPU= s\000\000\000\00001 */ -{ 129068 }, /* backend_cycles_idle\000Default\000stalled\\-cycles\\-backen= d / cpu\\-cycles\000backend_cycles_idle > 0.2\000Backend stalls per cycle\0= 00\000\000\000\000\00000 */ -{ 129368 }, /* branch_frequency\000Default\000branches / (software@cpu\\-c= lock\\,name\\=3Dcpu\\-clock@ if #target_cpu else software@task\\-clock\\,na= me\\=3Dtask\\-clock@)\000\000Branches per CPU second\000\0001000K/sec\000\0= 00\000\00001 */ -{ 129547 }, /* branch_miss_rate\000Default\000branch\\-misses / branches\0= 00branch_miss_rate > 0.05\000Branch miss rate\000\000100%\000\000\000\00000= */ -{ 127943 }, /* cs_per_second\000Default\000software@context\\-switches\\,n= ame\\=3Dcontext\\-switches@ * 1e9 / (software@cpu\\-clock\\,name\\=3Dcpu\\-= clock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtask\\-clock@)\= 000\000Context switches per CPU second\000\0001cs/sec\000\000\000\00001 */ -{ 129193 }, /* cycles_frequency\000Default\000cpu\\-cycles / (software@cpu= \\-clock\\,name\\=3Dcpu\\-clock@ if #target_cpu else software@task\\-clock\= \,name\\=3Dtask\\-clock@)\000\000Cycles per CPU second\000\0001GHz\000\000\= 000\00001 */ -{ 128939 }, /* frontend_cycles_idle\000Default\000stalled\\-cycles\\-front= end / cpu\\-cycles\000frontend_cycles_idle > 0.1\000Frontend stalls per cyc= le\000\000\000\000\000\00000 */ -{ 128664 }, /* insn_per_cycle\000Default\000instructions / cpu\\-cycles\00= 0insn_per_cycle < 1\000Instructions Per Cycle\000\0001instructions\000\000\= 000\00000 */ -{ 128175 }, /* migrations_per_second\000Default\000software@cpu\\-migratio= ns\\,name\\=3Dcpu\\-migrations@ * 1e9 / (software@cpu\\-clock\\,name\\=3Dcp= u\\-clock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtask\\-cloc= k@)\000\000Process migrations to a new CPU per CPU second\000\0001migration= s/sec\000\000\000\00001 */ -{ 128434 }, /* page_faults_per_second\000Default\000software@page\\-faults= \\,name\\=3Dpage\\-faults@ * 1e9 / (software@cpu\\-clock\\,name\\=3Dcpu\\-c= lock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtask\\-clock@)\0= 00\000Page faults per CPU second\000\0001faults/sec\000\000\000\00001 */ -{ 128776 }, /* stalled_cycles_per_instruction\000Default\000max(stalled\\-= cycles\\-frontend, stalled\\-cycles\\-backend) / instructions\000\000Max fr= ont or backend stalls per instruction\000\000\000\000\000\00000 */ +{ 127758 }, /* CPUs_utilized\000Default\000(software@cpu\\-clock\\,name\\= =3Dcpu\\-clock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtask\\= -clock@) / (duration_time * 1e9)\000\000Average CPU utilization\000\0001CPU= s\000\000\000\000011 */ +{ 129075 }, /* backend_cycles_idle\000Default\000stalled\\-cycles\\-backen= d / cpu\\-cycles\000backend_cycles_idle > 0.2\000Backend stalls per cycle\0= 00\000\000\000\000\000001 */ +{ 129377 }, /* branch_frequency\000Default\000branches / (software@cpu\\-c= lock\\,name\\=3Dcpu\\-clock@ if #target_cpu else software@task\\-clock\\,na= me\\=3Dtask\\-clock@)\000\000Branches per CPU second\000\0001000K/sec\000\0= 00\000\000011 */ +{ 129557 }, /* branch_miss_rate\000Default\000branch\\-misses / branches\0= 00branch_miss_rate > 0.05\000Branch miss rate\000\000100%\000\000\000\00000= 1 */ +{ 127944 }, /* cs_per_second\000Default\000software@context\\-switches\\,n= ame\\=3Dcontext\\-switches@ * 1e9 / (software@cpu\\-clock\\,name\\=3Dcpu\\-= clock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtask\\-clock@)\= 000\000Context switches per CPU second\000\0001cs/sec\000\000\000\000011 */ +{ 129201 }, /* cycles_frequency\000Default\000cpu\\-cycles / (software@cpu= \\-clock\\,name\\=3Dcpu\\-clock@ if #target_cpu else software@task\\-clock\= \,name\\=3Dtask\\-clock@)\000\000Cycles per CPU second\000\0001GHz\000\000\= 000\000011 */ +{ 128945 }, /* frontend_cycles_idle\000Default\000stalled\\-cycles\\-front= end / cpu\\-cycles\000frontend_cycles_idle > 0.1\000Frontend stalls per cyc= le\000\000\000\000\000\000001 */ +{ 128668 }, /* insn_per_cycle\000Default\000instructions / cpu\\-cycles\00= 0insn_per_cycle < 1\000Instructions Per Cycle\000\0001instructions\000\000\= 000\000001 */ +{ 128177 }, /* migrations_per_second\000Default\000software@cpu\\-migratio= ns\\,name\\=3Dcpu\\-migrations@ * 1e9 / (software@cpu\\-clock\\,name\\=3Dcp= u\\-clock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtask\\-cloc= k@)\000\000Process migrations to a new CPU per CPU second\000\0001migration= s/sec\000\000\000\000011 */ +{ 128437 }, /* page_faults_per_second\000Default\000software@page\\-faults= \\,name\\=3Dpage\\-faults@ * 1e9 / (software@cpu\\-clock\\,name\\=3Dcpu\\-c= lock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtask\\-clock@)\0= 00\000Page faults per CPU second\000\0001faults/sec\000\000\000\000011 */ +{ 128781 }, /* stalled_cycles_per_instruction\000Default\000max(stalled\\-= cycles\\-frontend, stalled\\-cycles\\-backend) / instructions\000\000Max fr= ont or backend stalls per instruction\000\000\000\000\000\000001 */ =20 }; =20 @@ -2698,21 +2698,21 @@ static const struct pmu_table_entry pmu_events__tes= t_soc_cpu[] =3D { }; =20 static const struct compact_pmu_event pmu_metrics__test_soc_cpu_default_co= re[] =3D { -{ 129650 }, /* CPI\000\0001 / IPC\000\000\000\000\000\000\000\00000 */ -{ 130331 }, /* DCache_L2_All\000\000DCache_L2_All_Hits + DCache_L2_All_Mis= s\000\000\000\000\000\000\000\00000 */ -{ 130103 }, /* DCache_L2_All_Hits\000\000l2_rqsts.demand_data_rd_hit + l2_= rqsts.pf_hit + l2_rqsts.rfo_hit\000\000\000\000\000\000\000\00000 */ -{ 130197 }, /* DCache_L2_All_Miss\000\000max(l2_rqsts.all_demand_data_rd -= l2_rqsts.demand_data_rd_hit, 0) + l2_rqsts.pf_miss + l2_rqsts.rfo_miss\000= \000\000\000\000\000\000\00000 */ -{ 130395 }, /* DCache_L2_Hits\000\000d_ratio(DCache_L2_All_Hits, DCache_L2= _All)\000\000\000\000\000\000\000\00000 */ -{ 130463 }, /* DCache_L2_Misses\000\000d_ratio(DCache_L2_All_Miss, DCache_= L2_All)\000\000\000\000\000\000\000\00000 */ -{ 129735 }, /* Frontend_Bound_SMT\000\000idq_uops_not_delivered.core / (4 = * (cpu_clk_unhalted.thread / 2 * (1 + cpu_clk_unhalted.one_thread_active / = cpu_clk_unhalted.ref_xclk)))\000\000\000\000\000\000\000\00000 */ -{ 129672 }, /* IPC\000group1\000inst_retired.any / cpu_clk_unhalted.thread= \000\000\000\000\000\000\000\00000 */ -{ 130597 }, /* L1D_Cache_Fill_BW\000\00064 * l1d.replacement / 1e9 / durat= ion_time\000\000\000\000\000\000\000\00000 */ -{ 130533 }, /* M1\000\000ipc + M2\000\000\000\000\000\000\000\00000 */ -{ 130555 }, /* M2\000\000ipc + M1\000\000\000\000\000\000\000\00000 */ -{ 130577 }, /* M3\000\0001 / M3\000\000\000\000\000\000\000\00000 */ -{ 130032 }, /* cache_miss_cycles\000group1\000dcache_miss_cpi + icache_mis= s_cycles\000\000\000\000\000\000\000\00000 */ -{ 129901 }, /* dcache_miss_cpi\000\000l1d\\-loads\\-misses / inst_retired.= any\000\000\000\000\000\000\000\00000 */ -{ 129965 }, /* icache_miss_cycles\000\000l1i\\-loads\\-misses / inst_retir= ed.any\000\000\000\000\000\000\000\00000 */ +{ 129661 }, /* CPI\000\0001 / IPC\000\000\000\000\000\000\000\000000 */ +{ 130350 }, /* DCache_L2_All\000\000DCache_L2_All_Hits + DCache_L2_All_Mis= s\000\000\000\000\000\000\000\000000 */ +{ 130120 }, /* DCache_L2_All_Hits\000\000l2_rqsts.demand_data_rd_hit + l2_= rqsts.pf_hit + l2_rqsts.rfo_hit\000\000\000\000\000\000\000\000000 */ +{ 130215 }, /* DCache_L2_All_Miss\000\000max(l2_rqsts.all_demand_data_rd -= l2_rqsts.demand_data_rd_hit, 0) + l2_rqsts.pf_miss + l2_rqsts.rfo_miss\000= \000\000\000\000\000\000\000000 */ +{ 130415 }, /* DCache_L2_Hits\000\000d_ratio(DCache_L2_All_Hits, DCache_L2= _All)\000\000\000\000\000\000\000\000000 */ +{ 130484 }, /* DCache_L2_Misses\000\000d_ratio(DCache_L2_All_Miss, DCache_= L2_All)\000\000\000\000\000\000\000\000000 */ +{ 129748 }, /* Frontend_Bound_SMT\000\000idq_uops_not_delivered.core / (4 = * (cpu_clk_unhalted.thread / 2 * (1 + cpu_clk_unhalted.one_thread_active / = cpu_clk_unhalted.ref_xclk)))\000\000\000\000\000\000\000\000000 */ +{ 129684 }, /* IPC\000group1\000inst_retired.any / cpu_clk_unhalted.thread= \000\000\000\000\000\000\000\000000 */ +{ 130622 }, /* L1D_Cache_Fill_BW\000\00064 * l1d.replacement / 1e9 / durat= ion_time\000\000\000\000\000\000\000\000000 */ +{ 130555 }, /* M1\000\000ipc + M2\000\000\000\000\000\000\000\000000 */ +{ 130578 }, /* M2\000\000ipc + M1\000\000\000\000\000\000\000\000000 */ +{ 130601 }, /* M3\000\0001 / M3\000\000\000\000\000\000\000\000000 */ +{ 130048 }, /* cache_miss_cycles\000group1\000dcache_miss_cpi + icache_mis= s_cycles\000\000\000\000\000\000\000\000000 */ +{ 129915 }, /* dcache_miss_cpi\000\000l1d\\-loads\\-misses / inst_retired.= any\000\000\000\000\000\000\000\000000 */ +{ 129980 }, /* icache_miss_cycles\000\000l1i\\-loads\\-misses / inst_retir= ed.any\000\000\000\000\000\000\000\000000 */ =20 }; =20 @@ -2894,6 +2894,8 @@ static void decompress_metric(int offset, struct pmu_= metric *pm) pm->aggr_mode =3D *p - '0'; p++; pm->event_grouping =3D *p - '0'; + p++; + pm->default_show_events =3D *p - '0'; } =20 static int pmu_events_table__for_each_event_pmu(const struct pmu_events_ta= ble *table, diff --git a/tools/perf/pmu-events/jevents.py b/tools/perf/pmu-events/jeven= ts.py index 5d3f4b44cfb7..3413ee5d0227 100755 --- a/tools/perf/pmu-events/jevents.py +++ b/tools/perf/pmu-events/jevents.py @@ -58,10 +58,12 @@ _json_event_attributes =3D [ _json_metric_attributes =3D [ 'metric_name', 'metric_group', 'metric_expr', 'metric_threshold', 'desc', 'long_desc', 'unit', 'compat', 'metricgroup_no_group', - 'default_metricgroup_name', 'aggr_mode', 'event_grouping' + 'default_metricgroup_name', 'aggr_mode', 'event_grouping', + 'default_show_events' ] # Attributes that are bools or enum int values, encoded as '0', '1',... -_json_enum_attributes =3D ['aggr_mode', 'deprecated', 'event_grouping', 'p= erpkg'] +_json_enum_attributes =3D ['aggr_mode', 'deprecated', 'event_grouping', 'p= erpkg', + 'default_show_events'] =20 def removesuffix(s: str, suffix: str) -> str: """Remove the suffix from a string @@ -356,6 +358,7 @@ class JsonEvent: self.metricgroup_no_group =3D jd.get('MetricgroupNoGroup') self.default_metricgroup_name =3D jd.get('DefaultMetricgroupName') self.event_grouping =3D convert_metric_constraint(jd.get('MetricConstr= aint')) + self.default_show_events =3D jd.get('DefaultShowEvents') self.metric_expr =3D None if 'MetricExpr' in jd: self.metric_expr =3D metric.ParsePerfJson(jd['MetricExpr']).Simplify= () diff --git a/tools/perf/pmu-events/pmu-events.h b/tools/perf/pmu-events/pmu= -events.h index 559265a903c8..d3b24014c6ff 100644 --- a/tools/perf/pmu-events/pmu-events.h +++ b/tools/perf/pmu-events/pmu-events.h @@ -74,6 +74,7 @@ struct pmu_metric { const char *default_metricgroup_name; enum aggr_mode_class aggr_mode; enum metric_event_groups event_grouping; + bool default_show_events; }; =20 struct pmu_events_table; diff --git a/tools/perf/util/evsel.h b/tools/perf/util/evsel.h index 71f74c7036ef..3ae4ac8f9a37 100644 --- a/tools/perf/util/evsel.h +++ b/tools/perf/util/evsel.h @@ -122,6 +122,7 @@ struct evsel { bool reset_group; bool needs_auxtrace_mmap; bool default_metricgroup; /* A member of the Default metricgroup */ + bool default_show_events; /* If a default group member, show the event = */ bool needs_uniquify; struct hashmap *per_pkg_mask; int err; diff --git a/tools/perf/util/metricgroup.c b/tools/perf/util/metricgroup.c index e67e04ce01c9..25c75fdbfc52 100644 --- a/tools/perf/util/metricgroup.c +++ b/tools/perf/util/metricgroup.c @@ -152,6 +152,8 @@ struct metric { * Should events of the metric be grouped? */ bool group_events; + /** Show events even if in the Default metric group. */ + bool default_show_events; /** * Parsed events for the metric. Optional as events may be taken from a * different metric whose group contains all the IDs necessary for this @@ -255,6 +257,7 @@ static struct metric *metric__new(const struct pmu_metr= ic *pm, m->pctx->sctx.runtime =3D runtime; m->pctx->sctx.system_wide =3D system_wide; m->group_events =3D !metric_no_group && metric__group_events(pm, metric_n= o_threshold); + m->default_show_events =3D pm->default_show_events; m->metric_refs =3D NULL; m->evlist =3D NULL; =20 @@ -1513,6 +1516,16 @@ static int parse_groups(struct evlist *perf_evlist, free(metric_events); goto out; } + if (m->default_show_events) { + struct evsel *pos; + + for (int i =3D 0; metric_events[i]; i++) + metric_events[i]->default_show_events =3D true; + evlist__for_each_entry(metric_evlist, pos) { + if (pos->metric_leader && pos->metric_leader->default_show_events) + pos->default_show_events =3D true; + } + } expr->metric_threshold =3D m->metric_threshold; expr->metric_unit =3D m->metric_unit; expr->metric_events =3D metric_events; diff --git a/tools/perf/util/stat-display.c b/tools/perf/util/stat-display.c index a67b991f4e81..4d0e353846ea 100644 --- a/tools/perf/util/stat-display.c +++ b/tools/perf/util/stat-display.c @@ -872,7 +872,7 @@ static void printout(struct perf_stat_config *config, s= truct outstate *os, out.ctx =3D os; out.force_header =3D false; =20 - if (!config->metric_only && !counter->default_metricgroup) { + if (!config->metric_only && (!counter->default_metricgroup || counter->de= fault_show_events)) { abs_printout(config, os, os->id, os->aggr_nr, counter, uval, ok); =20 print_noise(config, os, counter, noise, /*before_metric=3D*/true); @@ -880,7 +880,7 @@ static void printout(struct perf_stat_config *config, s= truct outstate *os, } =20 if (ok) { - if (!config->metric_only && counter->default_metricgroup) { + if (!config->metric_only && counter->default_metricgroup && !counter->de= fault_show_events) { void *from =3D NULL; =20 aggr_printout(config, os, os->evsel, os->id, os->aggr_nr); diff --git a/tools/perf/util/stat-shadow.c b/tools/perf/util/stat-shadow.c index abaf6b579bfc..4df614f8e200 100644 --- a/tools/perf/util/stat-shadow.c +++ b/tools/perf/util/stat-shadow.c @@ -665,7 +665,7 @@ void *perf_stat__print_shadow_stats_metricgroup(struct = perf_stat_config *config, if (strcmp(name, mexp->default_metricgroup_name)) return (void *)mexp; /* Only print the name of the metricgroup once */ - if (!header_printed) { + if (!header_printed && !evsel->default_show_events) { header_printed =3D true; perf_stat__print_metricgroup_header(config, evsel, ctxp, name, out); --=20 2.51.2.1041.gc1ab5b90ca-goog From nobody Tue Nov 11 11:28:57 2025 Received: from mail-pl1-f202.google.com (mail-pl1-f202.google.com [209.85.214.202]) (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 1DA5D2E1722 for ; Tue, 11 Nov 2025 04:04:35 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.214.202 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1762833878; cv=none; b=QjmeVXVSkrPYpP0Ee6Hs8PQQ8fDHq3ZVB3QlHu6u4/bRnyvTnp8+bkQMnzZatHKeBW5Ffc+Trc3daiTp/XFn5aXi0P2uJQ7euUZzvjaKbLBqusYR6aC3nMqEqgAuRZiOjlRh7HRheqmcx5Yq1U2jJ2vO+XVVixbsDJ7uo4QPzw4= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1762833878; c=relaxed/simple; bh=R178Ucl3QxqPTa3sc13fEY8U83OM569eeVC6oIADPmM=; h=Date:In-Reply-To:Mime-Version:References:Message-ID:Subject:From: To:Content-Type; b=dKceYvYxuGF77frfSQetsW0zZ7WDMS8Dm43j31u7SSAwueSZgDSnK6v+v3qofGTyP5xSKC05E477YfXGoO61cQsg+pDOOjvvEQnDm/Mi8HKtReVaOEi5eHK7MWKxl1tk6V/s/V4GsmFMlGObewovaYUhpkHYq/b7KMvr3Pjl08k= 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=l6QOZm2s; arc=none smtp.client-ip=209.85.214.202 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="l6QOZm2s" Received: by mail-pl1-f202.google.com with SMTP id d9443c01a7336-2982dec5ccbso32103725ad.3 for ; Mon, 10 Nov 2025 20:04:35 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1762833875; x=1763438675; darn=vger.kernel.org; h=content-transfer-encoding:to:from:subject:message-id:references :mime-version:in-reply-to:date:from:to:cc:subject:date:message-id :reply-to; bh=uCCveltFilMzlnQriVXwfyMsINsqp4xDj/nk6J2vWWw=; b=l6QOZm2sibfwrFAgYS95HaMyjNIL+bbfjce5loxYpgjrZZhp8u+pgE2JFi2AUPf90w GU5ERmmfIdyQgWlAYflIBCpjVRCQZlCNc2m08Sjudcq0799k4FGSXJ/mwhR+KG57kz6S 7EKYCQTp4lognzf3SwXp6cryuaWdGkKAAmn5l1wJ63FWEIuz0kIwzT7lt2i745zPw5sB zM9LK4D5VTejFgC0U8F664pbyvn8AjVPcMSjMMGwJKGbKvubosx936151ytRetAQ6khz +ol01brDLqC/qkMfvVvt0Jk2C0RE2Ls5gu/e5ZQz6+wv03qDMRAiJHJC41dnES/up5fJ QbOg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1762833875; x=1763438675; h=content-transfer-encoding:to:from:subject:message-id:references :mime-version:in-reply-to:date:x-gm-message-state:from:to:cc:subject :date:message-id:reply-to; bh=uCCveltFilMzlnQriVXwfyMsINsqp4xDj/nk6J2vWWw=; b=oSdZJ1gVky3cDh3FZE71eqbsSYuBLik/rll5o1xNE1ferJslvnX6yEHLppzf3/eSGE GojIMcW4I13kMMxxvu91QZjBwy96/Ah1FcfxY0NP9czC4dgV2fgsfeXOerAZ6LKi1Dft l3B4SlDT0/wAdOVJaSXHWWW/01jkW4UoSkJtOGdfo4KGG43RBbRssgK8C5o6Pc9/1wmc HOJfd92/DO73fMz/CxY5oY/w6q5FAH8JlqkKynaTNqyr4xAOLFVfLCEIYdpVsaGnBqXX e1SxaC5+UrBKeJpf5h9eQCvEBLVdK9YXf/V6XV9e899bKrfMCOEr7lEKIwUHj6qkMP6I CSBw== X-Forwarded-Encrypted: i=1; AJvYcCUMYrJVG6Xw4QwVrzMV1SWlDhGzUlSaff+ANCz9R7hOT0c5hCUdLJYfUHt3hHYUfSF7NLkMSYQ27EzCu78=@vger.kernel.org X-Gm-Message-State: AOJu0YxtEBVI2ltCLRdYdd80LKVZjRvpv0wyBNpkgNetb7GyIWBEjW/F NtVrsxdkVzzUJ3LBJ/NN4QuBHDbFLwVhAuB+s4672ba7899Z82iIVyHr5/hiCUUKZpQWozH8POj JPYPehraMKg== X-Google-Smtp-Source: AGHT+IFdTKUAS6WYxt2C4e7TqdsnhKDOsbLsGRxI4yLl1GdrDbKAchrEq0IfPfe/abPEUXfObwO8JuLlYn42 X-Received: from dycue1.prod.google.com ([2002:a05:7300:81:b0:2a4:6577:68f1]) (user=irogers job=prod-delivery.src-stubby-dispatcher) by 2002:a17:903:fad:b0:297:dade:456e with SMTP id d9443c01a7336-297e56dc6ccmr129898955ad.44.1762833875320; Mon, 10 Nov 2025 20:04:35 -0800 (PST) Date: Mon, 10 Nov 2025 20:04:04 -0800 In-Reply-To: <20251111040417.270945-1-irogers@google.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 References: <20251111040417.270945-1-irogers@google.com> X-Mailer: git-send-email 2.51.2.1041.gc1ab5b90ca-goog Message-ID: <20251111040417.270945-6-irogers@google.com> Subject: [PATCH v3 05/18] perf stat: Add detail -d,-dd,-ddd metrics From: Ian Rogers To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Namhyung Kim , Alexander Shishkin , Jiri Olsa , Ian Rogers , Adrian Hunter , James Clark , Xu Yang , Chun-Tse Shao , Thomas Richter , Sumanth Korikkar , Collin Funk , Thomas Falcon , Howard Chu , Dapeng Mi , Levi Yun , Yang Li , linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org, Andi Kleen , Weilin Wang Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" Add metrics for the stat-shadow -d, -dd and -ddd events and hard coded metrics. Remove the events as these now come from the metrics. Following this change a detailed perf stat output looks like: ``` $ perf stat -a -ddd -- sleep 1 Performance counter stats for 'system wide': 18,446 context-switches # 653.0 cs/sec = cs_per_second TopdownL1 (cpu_core) # 6.8 % tma_bad_spe= culation # 37.0 % tma_fronten= d_bound (30.32%) TopdownL1 (cpu_core) # 40.1 % tma_backend= _bound # 16.1 % tma_retirin= g (30.32%) 177 page-faults # 6.3 faults/= sec page_faults_per_second 472,170,922 cpu_atom/cpu-cycles/ # 0.0 GHz cy= cles_frequency (28.57%) 656,868,742 cpu_core/cpu-cycles/ # 0.0 GHz cy= cles_frequency (38.24%) # 22.2 % tma_bad_spe= culation # 12.2 % tma_retirin= g (28.55%) # 32.4 % tma_backend= _bound # 33.1 % tma_fronten= d_bound (35.71%) 43,583,604 cpu_atom/branches/ # 1.5 K/sec = branch_frequency (42.85%) 87,140,541 cpu_core/branches/ # 3.1 K/sec = branch_frequency (54.09%) 493 cpu-migrations # 17.5 migrati= ons/sec migrations_per_second 28,247,893,219 cpu-clock # 28.0 CPUs C= PUs_utilized 445,297,600 cpu_atom/cpu-cycles/ # 0.4 instruc= tions insn_per_cycle (42.87%) 642,323,993 cpu_core/cpu-cycles/ # 0.8 instruc= tions insn_per_cycle (62.01%) 2,126,311 cpu_atom/branch-misses/ # 6.8 % bran= ch_miss_rate (35.73%) 2,172,775 cpu_core/branch-misses/ # 2.5 % bran= ch_miss_rate (62.36%) 1,855,042 cpu_atom/LLC-loads/ # 0.0 % llc_= miss_rate (28.56%) 2,671,549 cpu_core/LLC-loads/ # 32.5 % llc_= miss_rate (46.31%) 8,440,231 cpu_core/L1-dcache-load-misses/ # nan % l1d_= miss_rate (37.99%) 10,823,925 cpu_atom/L1-icache-load-misses/ # 19.0 % l1i_= miss_rate (21.43%) 22,602,344 cpu_atom/dTLB-loads/ # 2.0 % dtlb= _miss_rate (21.44%) 136,524,528 cpu_core/dTLB-loads/ # 0.3 % dtlb= _miss_rate (15.06%) 1.007665494 seconds time elapsed ``` Signed-off-by: Ian Rogers --- tools/perf/builtin-stat.c | 100 +++--------------- .../arch/common/common/metrics.json | 54 ++++++++++ tools/perf/pmu-events/empty-pmu-events.c | 72 +++++++------ 3 files changed, 113 insertions(+), 113 deletions(-) diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c index 31c762695d4b..7862094b93c8 100644 --- a/tools/perf/builtin-stat.c +++ b/tools/perf/builtin-stat.c @@ -1857,28 +1857,6 @@ static int perf_stat_init_aggr_mode_file(struct perf= _stat *st) return 0; } =20 -/* Add legacy hardware/hardware-cache event to evlist for all core PMUs wi= thout wildcarding. */ -static int parse_hardware_event(struct evlist *evlist, const char *event, - struct parse_events_error *err) -{ - char buf[256]; - struct perf_pmu *pmu =3D NULL; - - while ((pmu =3D perf_pmus__scan_core(pmu)) !=3D NULL) { - int ret; - - if (perf_pmus__num_core_pmus() =3D=3D 1) - snprintf(buf, sizeof(buf), "%s/%s,name=3D%s/", pmu->name, event, event); - else - snprintf(buf, sizeof(buf), "%s/%s/", pmu->name, event); - - ret =3D parse_events(evlist, buf, err); - if (ret) - return ret; - } - return 0; -} - /* * Add default events, if there were no attributes specified or * if -d/--detailed, -d -d or -d -d -d is used: @@ -2006,22 +1984,34 @@ static int add_default_events(void) * threshold computation, but it will be computed if the events * are present. */ - if (metricgroup__has_metric_or_groups(pmu, "Default")) { - struct evlist *metric_evlist =3D evlist__new(); + const char *default_metricgroup_names[] =3D { + "Default", "Default2", "Default3", "Default4", + }; + + for (size_t i =3D 0; i < ARRAY_SIZE(default_metricgroup_names); i++) { + struct evlist *metric_evlist; + + if (!metricgroup__has_metric_or_groups(pmu, default_metricgroup_names[i= ])) + continue; + + if ((int)i > detailed_run) + break; =20 + metric_evlist =3D evlist__new(); if (!metric_evlist) { ret =3D -ENOMEM; - goto out; + break; } - if (metricgroup__parse_groups(metric_evlist, pmu, "Default", + if (metricgroup__parse_groups(metric_evlist, pmu, default_metricgroup_n= ames[i], /*metric_no_group=3D*/false, /*metric_no_merge=3D*/false, /*metric_no_threshold=3D*/true, stat_config.user_requested_cpu_list, stat_config.system_wide, stat_config.hardware_aware_grouping) < 0) { + evlist__delete(metric_evlist); ret =3D -1; - goto out; + break; } =20 evlist__for_each_entry(metric_evlist, evsel) @@ -2034,62 +2024,6 @@ static int add_default_events(void) evlist__delete(metric_evlist); } } - - /* Detailed events get appended to the event list: */ - - if (!ret && detailed_run >=3D 1) { - /* - * Detailed stats (-d), covering the L1 and last level data - * caches: - */ - const char *hw_events[] =3D { - "L1-dcache-loads", - "L1-dcache-load-misses", - "LLC-loads", - "LLC-load-misses", - }; - - for (size_t i =3D 0; i < ARRAY_SIZE(hw_events); i++) { - ret =3D parse_hardware_event(evlist, hw_events[i], &err); - if (ret) - goto out; - } - } - if (!ret && detailed_run >=3D 2) { - /* - * Very detailed stats (-d -d), covering the instruction cache - * and the TLB caches: - */ - const char *hw_events[] =3D { - "L1-icache-loads", - "L1-icache-load-misses", - "dTLB-loads", - "dTLB-load-misses", - "iTLB-loads", - "iTLB-load-misses", - }; - - for (size_t i =3D 0; i < ARRAY_SIZE(hw_events); i++) { - ret =3D parse_hardware_event(evlist, hw_events[i], &err); - if (ret) - goto out; - } - } - if (!ret && detailed_run >=3D 3) { - /* - * Very, very detailed stats (-d -d -d), adding prefetch events: - */ - const char *hw_events[] =3D { - "L1-dcache-prefetches", - "L1-dcache-prefetch-misses", - }; - - for (size_t i =3D 0; i < ARRAY_SIZE(hw_events); i++) { - ret =3D parse_hardware_event(evlist, hw_events[i], &err); - if (ret) - goto out; - } - } out: if (!ret) { evlist__for_each_entry(evlist, evsel) { diff --git a/tools/perf/pmu-events/arch/common/common/metrics.json b/tools/= perf/pmu-events/arch/common/common/metrics.json index 017bbdede3d7..89d1d9f61014 100644 --- a/tools/perf/pmu-events/arch/common/common/metrics.json +++ b/tools/perf/pmu-events/arch/common/common/metrics.json @@ -93,5 +93,59 @@ "MetricThreshold": "branch_miss_rate > 0.05", "ScaleUnit": "100%", "DefaultShowEvents": "1" + }, + { + "BriefDescription": "L1D miss rate", + "MetricExpr": "L1\\-dcache\\-load\\-misses / L1\\-dcache\\-loads", + "MetricGroup": "Default2", + "MetricName": "l1d_miss_rate", + "MetricThreshold": "l1d_miss_rate > 0.05", + "ScaleUnit": "100%", + "DefaultShowEvents": "1" + }, + { + "BriefDescription": "LLC miss rate", + "MetricExpr": "LLC\\-load\\-misses / LLC\\-loads", + "MetricGroup": "Default2", + "MetricName": "llc_miss_rate", + "MetricThreshold": "llc_miss_rate > 0.05", + "ScaleUnit": "100%", + "DefaultShowEvents": "1" + }, + { + "BriefDescription": "L1I miss rate", + "MetricExpr": "L1\\-icache\\-load\\-misses / L1\\-icache\\-loads", + "MetricGroup": "Default3", + "MetricName": "l1i_miss_rate", + "MetricThreshold": "l1i_miss_rate > 0.05", + "ScaleUnit": "100%", + "DefaultShowEvents": "1" + }, + { + "BriefDescription": "dTLB miss rate", + "MetricExpr": "dTLB\\-load\\-misses / dTLB\\-loads", + "MetricGroup": "Default3", + "MetricName": "dtlb_miss_rate", + "MetricThreshold": "dtlb_miss_rate > 0.05", + "ScaleUnit": "100%", + "DefaultShowEvents": "1" + }, + { + "BriefDescription": "iTLB miss rate", + "MetricExpr": "iTLB\\-load\\-misses / iTLB\\-loads", + "MetricGroup": "Default3", + "MetricName": "itlb_miss_rate", + "MetricThreshold": "itlb_miss_rate > 0.05", + "ScaleUnit": "100%", + "DefaultShowEvents": "1" + }, + { + "BriefDescription": "L1 prefetch miss rate", + "MetricExpr": "L1\\-dcache\\-prefetch\\-misses / L1\\-dcache\\-pre= fetches", + "MetricGroup": "Default4", + "MetricName": "l1_prefetch_miss_rate", + "MetricThreshold": "l1_prefetch_miss_rate > 0.05", + "ScaleUnit": "100%", + "DefaultShowEvents": "1" } ] diff --git a/tools/perf/pmu-events/empty-pmu-events.c b/tools/perf/pmu-even= ts/empty-pmu-events.c index bdf02b667f94..6fc490b96a02 100644 --- a/tools/perf/pmu-events/empty-pmu-events.c +++ b/tools/perf/pmu-events/empty-pmu-events.c @@ -1314,21 +1314,27 @@ static const char *const big_c_string =3D /* offset=3D129201 */ "cycles_frequency\000Default\000cpu\\-cycles / (soft= ware@cpu\\-clock\\,name\\=3Dcpu\\-clock@ if #target_cpu else software@task\= \-clock\\,name\\=3Dtask\\-clock@)\000\000Cycles per CPU second\000\0001GHz\= 000\000\000\000011" /* offset=3D129377 */ "branch_frequency\000Default\000branches / (software= @cpu\\-clock\\,name\\=3Dcpu\\-clock@ if #target_cpu else software@task\\-cl= ock\\,name\\=3Dtask\\-clock@)\000\000Branches per CPU second\000\0001000K/s= ec\000\000\000\000011" /* offset=3D129557 */ "branch_miss_rate\000Default\000branch\\-misses / br= anches\000branch_miss_rate > 0.05\000Branch miss rate\000\000100%\000\000\0= 00\000001" -/* offset=3D129661 */ "CPI\000\0001 / IPC\000\000\000\000\000\000\000\0000= 00" -/* offset=3D129684 */ "IPC\000group1\000inst_retired.any / cpu_clk_unhalte= d.thread\000\000\000\000\000\000\000\000000" -/* offset=3D129748 */ "Frontend_Bound_SMT\000\000idq_uops_not_delivered.co= re / (4 * (cpu_clk_unhalted.thread / 2 * (1 + cpu_clk_unhalted.one_thread_a= ctive / cpu_clk_unhalted.ref_xclk)))\000\000\000\000\000\000\000\000000" -/* offset=3D129915 */ "dcache_miss_cpi\000\000l1d\\-loads\\-misses / inst_= retired.any\000\000\000\000\000\000\000\000000" -/* offset=3D129980 */ "icache_miss_cycles\000\000l1i\\-loads\\-misses / in= st_retired.any\000\000\000\000\000\000\000\000000" -/* offset=3D130048 */ "cache_miss_cycles\000group1\000dcache_miss_cpi + ic= ache_miss_cycles\000\000\000\000\000\000\000\000000" -/* offset=3D130120 */ "DCache_L2_All_Hits\000\000l2_rqsts.demand_data_rd_h= it + l2_rqsts.pf_hit + l2_rqsts.rfo_hit\000\000\000\000\000\000\000\000000" -/* offset=3D130215 */ "DCache_L2_All_Miss\000\000max(l2_rqsts.all_demand_d= ata_rd - l2_rqsts.demand_data_rd_hit, 0) + l2_rqsts.pf_miss + l2_rqsts.rfo_= miss\000\000\000\000\000\000\000\000000" -/* offset=3D130350 */ "DCache_L2_All\000\000DCache_L2_All_Hits + DCache_L2= _All_Miss\000\000\000\000\000\000\000\000000" -/* offset=3D130415 */ "DCache_L2_Hits\000\000d_ratio(DCache_L2_All_Hits, D= Cache_L2_All)\000\000\000\000\000\000\000\000000" -/* offset=3D130484 */ "DCache_L2_Misses\000\000d_ratio(DCache_L2_All_Miss,= DCache_L2_All)\000\000\000\000\000\000\000\000000" -/* offset=3D130555 */ "M1\000\000ipc + M2\000\000\000\000\000\000\000\0000= 00" -/* offset=3D130578 */ "M2\000\000ipc + M1\000\000\000\000\000\000\000\0000= 00" -/* offset=3D130601 */ "M3\000\0001 / M3\000\000\000\000\000\000\000\000000" -/* offset=3D130622 */ "L1D_Cache_Fill_BW\000\00064 * l1d.replacement / 1e9= / duration_time\000\000\000\000\000\000\000\000000" +/* offset=3D129661 */ "l1d_miss_rate\000Default2\000L1\\-dcache\\-load\\-m= isses / L1\\-dcache\\-loads\000l1d_miss_rate > 0.05\000L1D miss rate\000\0= 00100%\000\000\000\000001" +/* offset=3D129777 */ "llc_miss_rate\000Default2\000LLC\\-load\\-misses / = LLC\\-loads\000llc_miss_rate > 0.05\000LLC miss rate\000\000100%\000\000\00= 0\000001" +/* offset=3D129878 */ "l1i_miss_rate\000Default3\000L1\\-icache\\-load\\-m= isses / L1\\-icache\\-loads\000l1i_miss_rate > 0.05\000L1I miss rate\000\00= 0100%\000\000\000\000001" +/* offset=3D129993 */ "dtlb_miss_rate\000Default3\000dTLB\\-load\\-misses = / dTLB\\-loads\000dtlb_miss_rate > 0.05\000dTLB miss rate\000\000100%\000\0= 00\000\000001" +/* offset=3D130099 */ "itlb_miss_rate\000Default3\000iTLB\\-load\\-misses = / iTLB\\-loads\000itlb_miss_rate > 0.05\000iTLB miss rate\000\000100%\000\0= 00\000\000001" +/* offset=3D130205 */ "l1_prefetch_miss_rate\000Default4\000L1\\-dcache\\-= prefetch\\-misses / L1\\-dcache\\-prefetches\000l1_prefetch_miss_rate > 0.0= 5\000L1 prefetch miss rate\000\000100%\000\000\000\000001" +/* offset=3D130353 */ "CPI\000\0001 / IPC\000\000\000\000\000\000\000\0000= 00" +/* offset=3D130376 */ "IPC\000group1\000inst_retired.any / cpu_clk_unhalte= d.thread\000\000\000\000\000\000\000\000000" +/* offset=3D130440 */ "Frontend_Bound_SMT\000\000idq_uops_not_delivered.co= re / (4 * (cpu_clk_unhalted.thread / 2 * (1 + cpu_clk_unhalted.one_thread_a= ctive / cpu_clk_unhalted.ref_xclk)))\000\000\000\000\000\000\000\000000" +/* offset=3D130607 */ "dcache_miss_cpi\000\000l1d\\-loads\\-misses / inst_= retired.any\000\000\000\000\000\000\000\000000" +/* offset=3D130672 */ "icache_miss_cycles\000\000l1i\\-loads\\-misses / in= st_retired.any\000\000\000\000\000\000\000\000000" +/* offset=3D130740 */ "cache_miss_cycles\000group1\000dcache_miss_cpi + ic= ache_miss_cycles\000\000\000\000\000\000\000\000000" +/* offset=3D130812 */ "DCache_L2_All_Hits\000\000l2_rqsts.demand_data_rd_h= it + l2_rqsts.pf_hit + l2_rqsts.rfo_hit\000\000\000\000\000\000\000\000000" +/* offset=3D130907 */ "DCache_L2_All_Miss\000\000max(l2_rqsts.all_demand_d= ata_rd - l2_rqsts.demand_data_rd_hit, 0) + l2_rqsts.pf_miss + l2_rqsts.rfo_= miss\000\000\000\000\000\000\000\000000" +/* offset=3D131042 */ "DCache_L2_All\000\000DCache_L2_All_Hits + DCache_L2= _All_Miss\000\000\000\000\000\000\000\000000" +/* offset=3D131107 */ "DCache_L2_Hits\000\000d_ratio(DCache_L2_All_Hits, D= Cache_L2_All)\000\000\000\000\000\000\000\000000" +/* offset=3D131176 */ "DCache_L2_Misses\000\000d_ratio(DCache_L2_All_Miss,= DCache_L2_All)\000\000\000\000\000\000\000\000000" +/* offset=3D131247 */ "M1\000\000ipc + M2\000\000\000\000\000\000\000\0000= 00" +/* offset=3D131270 */ "M2\000\000ipc + M1\000\000\000\000\000\000\000\0000= 00" +/* offset=3D131293 */ "M3\000\0001 / M3\000\000\000\000\000\000\000\000000" +/* offset=3D131314 */ "L1D_Cache_Fill_BW\000\00064 * l1d.replacement / 1e9= / duration_time\000\000\000\000\000\000\000\000000" ; =20 static const struct compact_pmu_event pmu_events__common_default_core[] = =3D { @@ -2621,8 +2627,14 @@ static const struct compact_pmu_event pmu_metrics__c= ommon_default_core[] =3D { { 129557 }, /* branch_miss_rate\000Default\000branch\\-misses / branches\0= 00branch_miss_rate > 0.05\000Branch miss rate\000\000100%\000\000\000\00000= 1 */ { 127944 }, /* cs_per_second\000Default\000software@context\\-switches\\,n= ame\\=3Dcontext\\-switches@ * 1e9 / (software@cpu\\-clock\\,name\\=3Dcpu\\-= clock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtask\\-clock@)\= 000\000Context switches per CPU second\000\0001cs/sec\000\000\000\000011 */ { 129201 }, /* cycles_frequency\000Default\000cpu\\-cycles / (software@cpu= \\-clock\\,name\\=3Dcpu\\-clock@ if #target_cpu else software@task\\-clock\= \,name\\=3Dtask\\-clock@)\000\000Cycles per CPU second\000\0001GHz\000\000\= 000\000011 */ +{ 129993 }, /* dtlb_miss_rate\000Default3\000dTLB\\-load\\-misses / dTLB\\= -loads\000dtlb_miss_rate > 0.05\000dTLB miss rate\000\000100%\000\000\000\0= 00001 */ { 128945 }, /* frontend_cycles_idle\000Default\000stalled\\-cycles\\-front= end / cpu\\-cycles\000frontend_cycles_idle > 0.1\000Frontend stalls per cyc= le\000\000\000\000\000\000001 */ { 128668 }, /* insn_per_cycle\000Default\000instructions / cpu\\-cycles\00= 0insn_per_cycle < 1\000Instructions Per Cycle\000\0001instructions\000\000\= 000\000001 */ +{ 130099 }, /* itlb_miss_rate\000Default3\000iTLB\\-load\\-misses / iTLB\\= -loads\000itlb_miss_rate > 0.05\000iTLB miss rate\000\000100%\000\000\000\0= 00001 */ +{ 130205 }, /* l1_prefetch_miss_rate\000Default4\000L1\\-dcache\\-prefetch= \\-misses / L1\\-dcache\\-prefetches\000l1_prefetch_miss_rate > 0.05\000L1 = prefetch miss rate\000\000100%\000\000\000\000001 */ +{ 129661 }, /* l1d_miss_rate\000Default2\000L1\\-dcache\\-load\\-misses / = L1\\-dcache\\-loads\000l1d_miss_rate > 0.05\000L1D miss rate\000\000100%\0= 00\000\000\000001 */ +{ 129878 }, /* l1i_miss_rate\000Default3\000L1\\-icache\\-load\\-misses / = L1\\-icache\\-loads\000l1i_miss_rate > 0.05\000L1I miss rate\000\000100%\00= 0\000\000\000001 */ +{ 129777 }, /* llc_miss_rate\000Default2\000LLC\\-load\\-misses / LLC\\-lo= ads\000llc_miss_rate > 0.05\000LLC miss rate\000\000100%\000\000\000\000001= */ { 128177 }, /* migrations_per_second\000Default\000software@cpu\\-migratio= ns\\,name\\=3Dcpu\\-migrations@ * 1e9 / (software@cpu\\-clock\\,name\\=3Dcp= u\\-clock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtask\\-cloc= k@)\000\000Process migrations to a new CPU per CPU second\000\0001migration= s/sec\000\000\000\000011 */ { 128437 }, /* page_faults_per_second\000Default\000software@page\\-faults= \\,name\\=3Dpage\\-faults@ * 1e9 / (software@cpu\\-clock\\,name\\=3Dcpu\\-c= lock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtask\\-clock@)\0= 00\000Page faults per CPU second\000\0001faults/sec\000\000\000\000011 */ { 128781 }, /* stalled_cycles_per_instruction\000Default\000max(stalled\\-= cycles\\-frontend, stalled\\-cycles\\-backend) / instructions\000\000Max fr= ont or backend stalls per instruction\000\000\000\000\000\000001 */ @@ -2698,21 +2710,21 @@ static const struct pmu_table_entry pmu_events__tes= t_soc_cpu[] =3D { }; =20 static const struct compact_pmu_event pmu_metrics__test_soc_cpu_default_co= re[] =3D { -{ 129661 }, /* CPI\000\0001 / IPC\000\000\000\000\000\000\000\000000 */ -{ 130350 }, /* DCache_L2_All\000\000DCache_L2_All_Hits + DCache_L2_All_Mis= s\000\000\000\000\000\000\000\000000 */ -{ 130120 }, /* DCache_L2_All_Hits\000\000l2_rqsts.demand_data_rd_hit + l2_= rqsts.pf_hit + l2_rqsts.rfo_hit\000\000\000\000\000\000\000\000000 */ -{ 130215 }, /* DCache_L2_All_Miss\000\000max(l2_rqsts.all_demand_data_rd -= l2_rqsts.demand_data_rd_hit, 0) + l2_rqsts.pf_miss + l2_rqsts.rfo_miss\000= \000\000\000\000\000\000\000000 */ -{ 130415 }, /* DCache_L2_Hits\000\000d_ratio(DCache_L2_All_Hits, DCache_L2= _All)\000\000\000\000\000\000\000\000000 */ -{ 130484 }, /* DCache_L2_Misses\000\000d_ratio(DCache_L2_All_Miss, DCache_= L2_All)\000\000\000\000\000\000\000\000000 */ -{ 129748 }, /* Frontend_Bound_SMT\000\000idq_uops_not_delivered.core / (4 = * (cpu_clk_unhalted.thread / 2 * (1 + cpu_clk_unhalted.one_thread_active / = cpu_clk_unhalted.ref_xclk)))\000\000\000\000\000\000\000\000000 */ -{ 129684 }, /* IPC\000group1\000inst_retired.any / cpu_clk_unhalted.thread= \000\000\000\000\000\000\000\000000 */ -{ 130622 }, /* L1D_Cache_Fill_BW\000\00064 * l1d.replacement / 1e9 / durat= ion_time\000\000\000\000\000\000\000\000000 */ -{ 130555 }, /* M1\000\000ipc + M2\000\000\000\000\000\000\000\000000 */ -{ 130578 }, /* M2\000\000ipc + M1\000\000\000\000\000\000\000\000000 */ -{ 130601 }, /* M3\000\0001 / M3\000\000\000\000\000\000\000\000000 */ -{ 130048 }, /* cache_miss_cycles\000group1\000dcache_miss_cpi + icache_mis= s_cycles\000\000\000\000\000\000\000\000000 */ -{ 129915 }, /* dcache_miss_cpi\000\000l1d\\-loads\\-misses / inst_retired.= any\000\000\000\000\000\000\000\000000 */ -{ 129980 }, /* icache_miss_cycles\000\000l1i\\-loads\\-misses / inst_retir= ed.any\000\000\000\000\000\000\000\000000 */ +{ 130353 }, /* CPI\000\0001 / IPC\000\000\000\000\000\000\000\000000 */ +{ 131042 }, /* DCache_L2_All\000\000DCache_L2_All_Hits + DCache_L2_All_Mis= s\000\000\000\000\000\000\000\000000 */ +{ 130812 }, /* DCache_L2_All_Hits\000\000l2_rqsts.demand_data_rd_hit + l2_= rqsts.pf_hit + l2_rqsts.rfo_hit\000\000\000\000\000\000\000\000000 */ +{ 130907 }, /* DCache_L2_All_Miss\000\000max(l2_rqsts.all_demand_data_rd -= l2_rqsts.demand_data_rd_hit, 0) + l2_rqsts.pf_miss + l2_rqsts.rfo_miss\000= \000\000\000\000\000\000\000000 */ +{ 131107 }, /* DCache_L2_Hits\000\000d_ratio(DCache_L2_All_Hits, DCache_L2= _All)\000\000\000\000\000\000\000\000000 */ +{ 131176 }, /* DCache_L2_Misses\000\000d_ratio(DCache_L2_All_Miss, DCache_= L2_All)\000\000\000\000\000\000\000\000000 */ +{ 130440 }, /* Frontend_Bound_SMT\000\000idq_uops_not_delivered.core / (4 = * (cpu_clk_unhalted.thread / 2 * (1 + cpu_clk_unhalted.one_thread_active / = cpu_clk_unhalted.ref_xclk)))\000\000\000\000\000\000\000\000000 */ +{ 130376 }, /* IPC\000group1\000inst_retired.any / cpu_clk_unhalted.thread= \000\000\000\000\000\000\000\000000 */ +{ 131314 }, /* L1D_Cache_Fill_BW\000\00064 * l1d.replacement / 1e9 / durat= ion_time\000\000\000\000\000\000\000\000000 */ +{ 131247 }, /* M1\000\000ipc + M2\000\000\000\000\000\000\000\000000 */ +{ 131270 }, /* M2\000\000ipc + M1\000\000\000\000\000\000\000\000000 */ +{ 131293 }, /* M3\000\0001 / M3\000\000\000\000\000\000\000\000000 */ +{ 130740 }, /* cache_miss_cycles\000group1\000dcache_miss_cpi + icache_mis= s_cycles\000\000\000\000\000\000\000\000000 */ +{ 130607 }, /* dcache_miss_cpi\000\000l1d\\-loads\\-misses / inst_retired.= any\000\000\000\000\000\000\000\000000 */ +{ 130672 }, /* icache_miss_cycles\000\000l1i\\-loads\\-misses / inst_retir= ed.any\000\000\000\000\000\000\000\000000 */ =20 }; =20 --=20 2.51.2.1041.gc1ab5b90ca-goog From nobody Tue Nov 11 11:28:57 2025 Received: from mail-pj1-f73.google.com (mail-pj1-f73.google.com [209.85.216.73]) (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 D6E252EDD53 for ; Tue, 11 Nov 2025 04:04:37 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.216.73 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1762833879; cv=none; b=dBqCKhHv5VlLU3qarkQNbRnhsKJ4X28S47nsqcqkMyKXURoZ2HdI7BLMiduTPq/PE4rVrXNGVR/lokw7n6NBWqM+KXxeQlbHkCC56Be0O56L0iggtNOSTwektjPqjgE9M6OEQDIdSprPLf0P8AEdMfqzHEiouExSk2VvYYUr/48= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1762833879; c=relaxed/simple; bh=c6CS8pyaASg3SGSc1hCSAT5qh+Xngw1FLjg93Oor/hY=; h=Date:In-Reply-To:Mime-Version:References:Message-ID:Subject:From: To:Content-Type; b=FxpvbozoeFoRwuPbQ6HsxcBPbgE62985gAstPHh9dcRR0aV9gJ/1LhA10ueVaYv/DUuBY6SjLO/Y0EDQPtduTuVB/oW1FPzUCkdg7Rizs9t1g5Qioo8EppVeeuFpFH6Fr0Cm3C8pfMNV3LbLczBmMstsFeEkIordS1k1smOumM8= 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=YweIX7LW; arc=none smtp.client-ip=209.85.216.73 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="YweIX7LW" Received: by mail-pj1-f73.google.com with SMTP id 98e67ed59e1d1-3437b43eec4so4238844a91.3 for ; Mon, 10 Nov 2025 20:04:37 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1762833877; x=1763438677; darn=vger.kernel.org; h=to:from:subject:message-id:references:mime-version:in-reply-to:date :from:to:cc:subject:date:message-id:reply-to; bh=Svu0mc053rqCWz40HG7pjwTWIembvieSe2J5PJ8yob4=; b=YweIX7LWgvGkcZJc1R9u2N7f/8W/Xgbes6EzT+kJySTgu1GtDU04QUH1dDYbnjZnRV 0A1FRy8ODsDSAi6hZAnODAwRHDd2KSSlkRiVp/PCKKVzPKV4AglB9I80xSj9wJ1GxuBr gnTKnbZWEBcLWXubZRBrzU8sGxE24263amWHbgD/uxx0yDlWhMzBbhHBPx0Acuhq4dD2 AAh/K5YeB2PMfJKcHhJPMBggd0KW5ppZR3o3T354S/XRCd7mr+lfrNluZtKKggeZZ6z0 aq5WZgnQxD71gkzQsNoGIQiY5RnyLdjW59qFvrLZJwFQUEWMjASdOoURdGkdD69cR328 Htcg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1762833877; x=1763438677; h=to:from:subject:message-id:references:mime-version:in-reply-to:date :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=Svu0mc053rqCWz40HG7pjwTWIembvieSe2J5PJ8yob4=; b=ZmLVy1txrejfxPV+7tdMtoKs1lpZMJQcVt4AUfqmNPza+8dVyZAjKy7JuO23VMAMgu O27OXNCz6/Cq9VwOOplVGhFu4zX6CM03DCxadOtxn6vqEQbDx8wmsk/3k6r6osszH+dq sGDkUt848hujeLbiSYyPj/wPQBYsqoP4uJZhCq58QIJO1BlF6MLhJsmzH3+4LMrfn3v4 Y2sBr5TUkNIytZ6v3Jz1XbJSf4X341z/wwuDn1nGQe5FLElCJeO9vS8FJQksP1TBV6Pt w8/5M3i/IVJhRen8vtKoSXF62RIhRbfV1ErVQMbIEgCukzY/FfMmY1UxsVhTh3JY9/Iw z3HA== X-Forwarded-Encrypted: i=1; AJvYcCVoGXmLFHzoiUbl9yftLe1RrBJlVtR9kJs/2g0Sm7P3YJonh0Gafp1oxFLvhcB5ITgmFu1QzPN6/reTBkA=@vger.kernel.org X-Gm-Message-State: AOJu0YyQVzHMOrqHR/QoT08GHpGYMd3blMLMlZ/XZMx60PWAMlloBNHS XXxEX7jQo45Br/IPe3JGlNpNqfk0VHiXAAX3JSz1gS4TXVEqaNiliCXKfZ1lUCsjy6F8+dqF0HH 74yZrn6fXaQ== X-Google-Smtp-Source: AGHT+IFZ4cmTdJets0uatcWhFJWHsq2cLSkNWliCuH5bJJ3nYjw/nVMxFyjVHKchLBTnJVp1iiYPJYZ3mfW6 X-Received: from dybnj4.prod.google.com ([2002:a05:7300:d084:b0:2a4:5a4a:cc6d]) (user=irogers job=prod-delivery.src-stubby-dispatcher) by 2002:a17:903:2281:b0:295:9cb5:ae2a with SMTP id d9443c01a7336-297e53e79damr111796025ad.9.1762833877060; Mon, 10 Nov 2025 20:04:37 -0800 (PST) Date: Mon, 10 Nov 2025 20:04:05 -0800 In-Reply-To: <20251111040417.270945-1-irogers@google.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 References: <20251111040417.270945-1-irogers@google.com> X-Mailer: git-send-email 2.51.2.1041.gc1ab5b90ca-goog Message-ID: <20251111040417.270945-7-irogers@google.com> Subject: [PATCH v3 06/18] perf script: Change metric format to use json metrics From: Ian Rogers To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Namhyung Kim , Alexander Shishkin , Jiri Olsa , Ian Rogers , Adrian Hunter , James Clark , Xu Yang , Chun-Tse Shao , Thomas Richter , Sumanth Korikkar , Collin Funk , Thomas Falcon , Howard Chu , Dapeng Mi , Levi Yun , Yang Li , linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org, Andi Kleen , Weilin Wang Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" The metric format option isn't properly supported. This change improves that by making the sample events update the counts of an evsel, where the shadow metric code expects to read the values. To support printing metrics, metrics need to be found. This is done on the first attempt to print a metric. Every metric is parsed and then the evsels in the metric's evlist compared to those in perf script using the perf_event_attr type and config. If the metric matches then it is added for printing. As an event in the perf script's evlist may have >1 metric id, or different leader for aggregation, the first metric matched will be displayed in those cases. An example use is: ``` $ perf record -a -e '{instructions,cpu-cycles}:S' -a -- sleep 1 $ perf script -F period,metric ... 867817 metric: 0.30 insn per cycle 125394 metric: 0.04 insn per cycle 313516 metric: 0.11 insn per cycle metric: 1.00 insn per cycle ``` Signed-off-by: Ian Rogers --- tools/perf/builtin-script.c | 239 ++++++++++++++++++++++++++++++++---- 1 file changed, 217 insertions(+), 22 deletions(-) diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c index d813adbf9889..2966d6f1316e 100644 --- a/tools/perf/builtin-script.c +++ b/tools/perf/builtin-script.c @@ -33,6 +33,7 @@ #include "util/path.h" #include "util/event.h" #include "util/mem-info.h" +#include "util/metricgroup.h" #include "ui/ui.h" #include "print_binary.h" #include "print_insn.h" @@ -341,9 +342,6 @@ struct evsel_script { char *filename; FILE *fp; u64 samples; - /* For metric output */ - u64 val; - int gnum; }; =20 static inline struct evsel_script *evsel_script(struct evsel *evsel) @@ -2132,13 +2130,161 @@ static void script_new_line(struct perf_stat_confi= g *config __maybe_unused, fputs("\tmetric: ", mctx->fp); } =20 -static void perf_sample__fprint_metric(struct perf_script *script, - struct thread *thread, +struct script_find_metrics_args { + struct evlist *evlist; + bool system_wide; +}; + +static struct evsel *map_metric_evsel_to_script_evsel(struct evlist *scrip= t_evlist, + struct evsel *metric_evsel) +{ + struct evsel *script_evsel; + + evlist__for_each_entry(script_evlist, script_evsel) { + /* Skip if perf_event_attr differ. */ + if (metric_evsel->core.attr.type !=3D script_evsel->core.attr.type) + continue; + if (metric_evsel->core.attr.config !=3D script_evsel->core.attr.config) + continue; + /* Skip if the script event has a metric_id that doesn't match. */ + if (script_evsel->metric_id && + strcmp(evsel__metric_id(metric_evsel), evsel__metric_id(script_evsel= ))) { + pr_debug("Skipping matching evsel due to differing metric ids '%s' vs '= %s'\n", + evsel__metric_id(metric_evsel), evsel__metric_id(script_evsel)); + continue; + } + return script_evsel; + } + return NULL; +} + +static int script_find_metrics(const struct pmu_metric *pm, + const struct pmu_metrics_table *table __maybe_unused, + void *data) +{ + struct script_find_metrics_args *args =3D data; + struct evlist *script_evlist =3D args->evlist; + struct evlist *metric_evlist =3D evlist__new(); + struct evsel *metric_evsel; + int ret =3D metricgroup__parse_groups(metric_evlist, + /*pmu=3D*/"all", + pm->metric_name, + /*metric_no_group=3D*/false, + /*metric_no_merge=3D*/false, + /*metric_no_threshold=3D*/true, + /*user_requested_cpu_list=3D*/NULL, + args->system_wide, + /*hardware_aware_grouping=3D*/false); + + if (ret) { + /* Metric parsing failed but continue the search. */ + goto out; + } + + /* + * Check the script_evlist has an entry for each metric_evlist entry. If + * the script evsel was already set up avoid changing data that may + * break it. + */ + evlist__for_each_entry(metric_evlist, metric_evsel) { + struct evsel *script_evsel =3D + map_metric_evsel_to_script_evsel(script_evlist, metric_evsel); + struct evsel *new_metric_leader; + + if (!script_evsel) { + pr_debug("Skipping metric '%s' as evsel '%s' / '%s' is missing\n", + pm->metric_name, evsel__name(metric_evsel), + evsel__metric_id(metric_evsel)); + goto out; + } + + if (script_evsel->metric_leader =3D=3D NULL) + continue; + + if (metric_evsel->metric_leader =3D=3D metric_evsel) { + new_metric_leader =3D script_evsel; + } else { + new_metric_leader =3D + map_metric_evsel_to_script_evsel(script_evlist, + metric_evsel->metric_leader); + } + /* Mismatching evsel leaders. */ + if (script_evsel->metric_leader !=3D new_metric_leader) { + pr_debug("Skipping metric '%s' due to mismatching evsel metric leaders = '%s' vs '%s'\n", + pm->metric_name, evsel__metric_id(metric_evsel), + evsel__metric_id(script_evsel)); + goto out; + } + } + /* + * Metric events match those in the script evlist, copy metric evsel + * data into the script evlist. + */ + evlist__for_each_entry(metric_evlist, metric_evsel) { + struct evsel *script_evsel =3D + map_metric_evsel_to_script_evsel(script_evlist, metric_evsel); + struct metric_event *metric_me =3D metricgroup__lookup(&metric_evlist->m= etric_events, + metric_evsel, + /*create=3D*/false); + + if (script_evsel->metric_id =3D=3D NULL) { + script_evsel->metric_id =3D metric_evsel->metric_id; + metric_evsel->metric_id =3D NULL; + } + + if (script_evsel->metric_leader =3D=3D NULL) { + if (metric_evsel->metric_leader =3D=3D metric_evsel) { + script_evsel->metric_leader =3D script_evsel; + } else { + script_evsel->metric_leader =3D + map_metric_evsel_to_script_evsel(script_evlist, + metric_evsel->metric_leader); + } + } + + if (metric_me) { + struct metric_expr *expr; + struct metric_event *script_me =3D + metricgroup__lookup(&script_evlist->metric_events, + script_evsel, + /*create=3D*/true); + + if (!script_me) { + /* + * As the metric_expr is created, the only + * failure is a lack of memory. + */ + goto out; + } + list_splice_init(&metric_me->head, &script_me->head); + list_for_each_entry(expr, &script_me->head, nd) { + for (int i =3D 0; expr->metric_events[i]; i++) { + expr->metric_events[i] =3D + map_metric_evsel_to_script_evsel(script_evlist, + expr->metric_events[i]); + } + } + } + } + pr_debug("Found metric '%s' whose evsels match those of in the perf data\= n", + pm->metric_name); + evlist__delete(metric_evlist); +out: + return 0; +} + +static struct aggr_cpu_id script_aggr_cpu_id_get(struct perf_stat_config *= config __maybe_unused, + struct perf_cpu cpu) +{ + return aggr_cpu_id__global(cpu, /*data=3D*/NULL); +} + +static void perf_sample__fprint_metric(struct thread *thread, struct evsel *evsel, struct perf_sample *sample, FILE *fp) { - struct evsel *leader =3D evsel__leader(evsel); + static bool init_metrics; struct perf_stat_output_ctx ctx =3D { .print_metric =3D script_print_metric, .new_line =3D script_new_line, @@ -2150,23 +2296,72 @@ static void perf_sample__fprint_metric(struct perf_= script *script, }, .force_header =3D false, }; - struct evsel *ev2; - u64 val; + struct perf_counts_values *count, *old_count; + int cpu_map_idx, thread_map_idx, aggr_idx; + struct evsel *pos; + + if (!init_metrics) { + /* One time initialization of stat_config and metric data. */ + struct script_find_metrics_args args =3D { + .evlist =3D evsel->evlist, + /* TODO: Determine system-wide based on evlist.. */ + .system_wide =3D true, + }; + if (!stat_config.output) + stat_config.output =3D stdout; + + if (!stat_config.aggr_map) { + /* TODO: currently only global aggregation is supported. */ + assert(stat_config.aggr_mode =3D=3D AGGR_GLOBAL); + stat_config.aggr_get_id =3D script_aggr_cpu_id_get; + stat_config.aggr_map =3D + cpu_aggr_map__new(evsel->evlist->core.user_requested_cpus, + aggr_cpu_id__global, /*data=3D*/NULL, + /*needs_sort=3D*/false); + } =20 - if (!evsel->stats) - evlist__alloc_stats(&stat_config, script->session->evlist, /*alloc_raw= =3D*/false); - if (evsel_script(leader)->gnum++ =3D=3D 0) - perf_stat__reset_shadow_stats(); - val =3D sample->period * evsel->scale; - evsel_script(evsel)->val =3D val; - if (evsel_script(leader)->gnum =3D=3D leader->core.nr_members) { - for_each_group_member (ev2, leader) { - perf_stat__print_shadow_stats(&stat_config, ev2, - evsel_script(ev2)->val, - sample->cpu, - &ctx); + metricgroup__for_each_metric(pmu_metrics_table__find(), script_find_metr= ics, &args); + init_metrics =3D true; + } + + if (!evsel->stats) { + if (evlist__alloc_stats(&stat_config, evsel->evlist, /*alloc_raw=3D*/tru= e) < 0) + return; + } + if (!evsel->stats->aggr) { + if (evlist__alloc_aggr_stats(evsel->evlist, stat_config.aggr_map->nr) < = 0) + return; + } + + /* Update the evsel's count using the sample's data. */ + cpu_map_idx =3D perf_cpu_map__idx(evsel->core.cpus, (struct perf_cpu){sam= ple->cpu}); + thread_map_idx =3D perf_thread_map__idx(evsel->core.threads, sample->tid); + if (thread_map_idx < 0) { + /* Missing thread, check for any thread. */ + if (perf_thread_map__pid(evsel->core.threads, /*idx=3D*/0) =3D=3D -1) { + thread_map_idx =3D 0; + } else { + pr_info("Missing thread map entry for thread %d\n", sample->tid); + return; + } + } + count =3D perf_counts(evsel->counts, cpu_map_idx, thread_map_idx); + old_count =3D perf_counts(evsel->prev_raw_counts, cpu_map_idx, thread_map= _idx); + count->val =3D old_count->val + sample->period; + count->run =3D old_count->run + 1; + count->ena =3D old_count->ena + 1; + + /* Update the aggregated stats. */ + perf_stat_process_counter(&stat_config, evsel); + + /* Display all metrics. */ + evlist__for_each_entry(evsel->evlist, pos) { + cpu_aggr_map__for_each_idx(aggr_idx, stat_config.aggr_map) { + perf_stat__print_shadow_stats(&stat_config, pos, + count->val, + aggr_idx, + &ctx); } - evsel_script(leader)->gnum =3D 0; } } =20 @@ -2348,7 +2543,7 @@ static void process_event(struct perf_script *script, } =20 if (PRINT_FIELD(METRIC)) - perf_sample__fprint_metric(script, thread, evsel, sample, fp); + perf_sample__fprint_metric(thread, evsel, sample, fp); =20 if (verbose > 0) fflush(fp); --=20 2.51.2.1041.gc1ab5b90ca-goog From nobody Tue Nov 11 11:28:57 2025 Received: from mail-pl1-f202.google.com (mail-pl1-f202.google.com [209.85.214.202]) (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 A29A134B1AF for ; Tue, 11 Nov 2025 04:04:39 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.214.202 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1762833881; cv=none; b=dRUEUw+M35ggbFNfHMvWpegynD07+E0GEXFrrdtT9f2WR4lxhD493Vn6Pz+DgNvQr4ajzVPMcN3GWVdkleddcQJrq6/AxxXgJiVi3kJwmYnuo6dBg21SKN+rP/zwSa6LqhzvCV3gk4HHRJmZIVkKOzhcGRAdOqxm4lyBnkRDOiw= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1762833881; c=relaxed/simple; bh=qUm5bx6/0XmDAeu6sv2CUrTMbOhB3wBeqNrPm7lQTZE=; h=Date:In-Reply-To:Mime-Version:References:Message-ID:Subject:From: To:Content-Type; b=MQNoY8b8ag/RDHRktoHe91SbrasS5yT5jLF4eFa0L+ctg50FsbiiVqV2JDV+dznXiD0D0IT2cCxEj2kqFumMN4LKhvYOK5xTNky6M2OxOyvQ6gD06bewvz454gl1tPd975kTIjmaBNOmuklTGi8dfAAkRtAb1y0TrBm4w1kMmPc= 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=wZ7jMP0M; arc=none smtp.client-ip=209.85.214.202 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="wZ7jMP0M" Received: by mail-pl1-f202.google.com with SMTP id d9443c01a7336-2982b47ce35so16860275ad.2 for ; Mon, 10 Nov 2025 20:04:39 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1762833879; x=1763438679; darn=vger.kernel.org; h=to:from:subject:message-id:references:mime-version:in-reply-to:date :from:to:cc:subject:date:message-id:reply-to; bh=TDoLCBgO0tcXliKlxdM4YCl/0JA9B7dwGS5M2BAbZyM=; b=wZ7jMP0MMZ9RwL9TsAtyKt9YesLjYlbc9U2we951Uw1uF+xAPFggP5dGzNaTfY42IQ z29Uzx1MWTaefyxHBRP/G9YBKrma216sttGjKXMLHuzkscpplZT5+0FtHlxD8xd0hTI4 nHtJ4xs0xywf9xnuRKg9MkA7oX43bzoFzcwFT8RXJG7AXX7VRzGv0SEbjdSQLZDty6wN tl14REo+IRW5qkLTSMtdYi/TbZKXqUaH3FaW2qHr2oORHtNfTi5jruaH33ylbfK7DZsQ HmpOGc/AFXOG8BJW8PZWSITwq6r2Fv0J0MQyJ6QsQZp0bSjOR+RVQjEoGf5mvdUGn+ec 81eg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1762833879; x=1763438679; h=to:from:subject:message-id:references:mime-version:in-reply-to:date :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=TDoLCBgO0tcXliKlxdM4YCl/0JA9B7dwGS5M2BAbZyM=; b=N9E9WDCwjxrhSNdUrRIIb8nNAxWwhsilRwUsS0P9J3QywdCnrq6RjK65LKfdcEADe8 BCwDVTywJyuEN/l1wVsjz+wxlddBjUUmL71IA58i96HCC0SHvrQBJ/G33vJzj0UjI2B5 ZDdT/EQ5qz2UXuhtc7aUSCRkBtZoEWSF5VoKfLjGKoHge+2Yivp1kwVE5jKoUqOnL/5J oLINmrayNwmetrrPn51VKcI01m07NTS44VPbHc+jcNqFzn4AZ+kbG6hGCmhdJ/TwUP+n lxia4W6eh1crQCrYhtH56EVyZX93XN57gOEWKKW76LQGZmGaRY6SeqM+xP5RLZsrYMpq cbIg== X-Forwarded-Encrypted: i=1; AJvYcCUS3WS1JvY0aDZwHTgoUDg/bXZDbrw76cXZ/AO8Xcz28EWfJsKFn67YT18aQqjyLaXPwxWOEFvvTHejjj0=@vger.kernel.org X-Gm-Message-State: AOJu0YzZkC1/FU+dD0QxCntMzaGDkTbw7PrSTRQeJIZpRj/4tIcnZzWW itjgHSo39tQbPFkKbkDSYvp7gBbTsoEGsIvCEgjWkX238W9p9GYwhD1QGvh43G5rHziY068wJRO IgzU7EmmE4A== X-Google-Smtp-Source: AGHT+IG0tdChuTLKBtEvp/BgUyBD047haky1gD/MXzE6Sqj8bB3iyElIkrSbGdCG0yZvQbzlxgMtuvW4Q5wa X-Received: from dybph5.prod.google.com ([2002:a05:7301:4185:b0:2a4:603a:d428]) (user=irogers job=prod-delivery.src-stubby-dispatcher) by 2002:a17:903:1a27:b0:297:ec1a:9db8 with SMTP id d9443c01a7336-297ec1a9e0dmr149289465ad.49.1762833879157; Mon, 10 Nov 2025 20:04:39 -0800 (PST) Date: Mon, 10 Nov 2025 20:04:06 -0800 In-Reply-To: <20251111040417.270945-1-irogers@google.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 References: <20251111040417.270945-1-irogers@google.com> X-Mailer: git-send-email 2.51.2.1041.gc1ab5b90ca-goog Message-ID: <20251111040417.270945-8-irogers@google.com> Subject: [PATCH v3 07/18] perf stat: Remove hard coded shadow metrics From: Ian Rogers To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Namhyung Kim , Alexander Shishkin , Jiri Olsa , Ian Rogers , Adrian Hunter , James Clark , Xu Yang , Chun-Tse Shao , Thomas Richter , Sumanth Korikkar , Collin Funk , Thomas Falcon , Howard Chu , Dapeng Mi , Levi Yun , Yang Li , linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org, Andi Kleen , Weilin Wang Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" Now that the metrics are encoded in common json the hard coded printing means the metrics are shown twice. Remove the hard coded version. This means that when specifying events, and those events correspond to a hard coded metric, the metric will no longer be displayed. The metric will be displayed if the metric is requested. Due to the adhoc printing in the previous approach it was often found frustrating, the new approach avoids this. The default perf stat output on an alderlake now looks like: ``` $ perf stat -a -- sleep 1 Performance counter stats for 'system wide': 7,932 context-switches # 281.7 cs/sec = cs_per_second TopdownL1 (cpu_core) # 10.3 % tma_bad_spe= culation # 17.3 % tma_fronten= d_bound TopdownL1 (cpu_core) # 37.3 % tma_backend= _bound # 35.2 % tma_retiring 5,901 page-faults # 209.5 faults/= sec page_faults_per_second 418,955,116 cpu_atom/cpu-cycles/ # 0.0 GHz cy= cles_frequency (49.77%) 1,113,933,476 cpu_core/cpu-cycles/ # 0.0 GHz cy= cles_frequency # 14.6 % tma_bad_spe= culation # 8.5 % tma_retirin= g (50.17%) # 41.8 % tma_backend= _bound # 35.1 % tma_fronten= d_bound (50.31%) 32,196,918 cpu_atom/branches/ # 1.1 K/sec = branch_frequency (60.24%) 445,404,717 cpu_core/branches/ # 15.8 K/sec = branch_frequency 235 cpu-migrations # 8.3 migrati= ons/sec migrations_per_second 28,160,951,165 cpu-clock # 28.0 CPUs C= PUs_utilized 382,285,763 cpu_atom/cpu-cycles/ # 0.4 instruc= tions insn_per_cycle (60.18%) 1,114,029,255 cpu_core/cpu-cycles/ # 2.3 instruc= tions insn_per_cycle 1,768,727 cpu_atom/branches-misses/ # 6.5 % bran= ch_miss_rate (49.68%) 4,505,904 cpu_core/branches-misses/ # 1.0 % bran= ch_miss_rate 1.007137632 seconds time elapsed ``` Signed-off-by: Ian Rogers --- tools/perf/builtin-script.c | 1 - tools/perf/util/stat-display.c | 4 +- tools/perf/util/stat-shadow.c | 392 +-------------------------------- tools/perf/util/stat.h | 2 +- 4 files changed, 6 insertions(+), 393 deletions(-) diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c index 2966d6f1316e..d43797acb91b 100644 --- a/tools/perf/builtin-script.c +++ b/tools/perf/builtin-script.c @@ -2358,7 +2358,6 @@ static void perf_sample__fprint_metric(struct thread = *thread, evlist__for_each_entry(evsel->evlist, pos) { cpu_aggr_map__for_each_idx(aggr_idx, stat_config.aggr_map) { perf_stat__print_shadow_stats(&stat_config, pos, - count->val, aggr_idx, &ctx); } diff --git a/tools/perf/util/stat-display.c b/tools/perf/util/stat-display.c index 4d0e353846ea..eabeab5e6614 100644 --- a/tools/perf/util/stat-display.c +++ b/tools/perf/util/stat-display.c @@ -902,7 +902,7 @@ static void printout(struct perf_stat_config *config, s= truct outstate *os, &num, from, &out); } while (from !=3D NULL); } else { - perf_stat__print_shadow_stats(config, counter, uval, aggr_idx, &out); + perf_stat__print_shadow_stats(config, counter, aggr_idx, &out); } } else { pm(config, os, METRIC_THRESHOLD_UNKNOWN, /*format=3D*/NULL, /*unit=3D*/N= ULL, /*val=3D*/0); @@ -1274,7 +1274,7 @@ static void print_metric_headers(struct perf_stat_con= fig *config, =20 os.evsel =3D counter; =20 - perf_stat__print_shadow_stats(config, counter, 0, 0, &out); + perf_stat__print_shadow_stats(config, counter, /*aggr_idx=3D*/0, &out); } =20 if (!config->json_output) diff --git a/tools/perf/util/stat-shadow.c b/tools/perf/util/stat-shadow.c index 4df614f8e200..afbc49e8cb31 100644 --- a/tools/perf/util/stat-shadow.c +++ b/tools/perf/util/stat-shadow.c @@ -20,357 +20,12 @@ struct stats walltime_nsecs_stats; struct rusage_stats ru_stats; =20 -enum { - CTX_BIT_USER =3D 1 << 0, - CTX_BIT_KERNEL =3D 1 << 1, - CTX_BIT_HV =3D 1 << 2, - CTX_BIT_HOST =3D 1 << 3, - CTX_BIT_IDLE =3D 1 << 4, - CTX_BIT_MAX =3D 1 << 5, -}; - -enum stat_type { - STAT_NONE =3D 0, - STAT_NSECS, - STAT_CYCLES, - STAT_INSTRUCTIONS, - STAT_STALLED_CYCLES_FRONT, - STAT_STALLED_CYCLES_BACK, - STAT_BRANCHES, - STAT_BRANCH_MISS, - STAT_CACHE_REFS, - STAT_CACHE_MISSES, - STAT_L1_DCACHE, - STAT_L1_ICACHE, - STAT_LL_CACHE, - STAT_ITLB_CACHE, - STAT_DTLB_CACHE, - STAT_L1D_MISS, - STAT_L1I_MISS, - STAT_LL_MISS, - STAT_DTLB_MISS, - STAT_ITLB_MISS, - STAT_MAX -}; - -static int evsel_context(const struct evsel *evsel) -{ - int ctx =3D 0; - - if (evsel->core.attr.exclude_kernel) - ctx |=3D CTX_BIT_KERNEL; - if (evsel->core.attr.exclude_user) - ctx |=3D CTX_BIT_USER; - if (evsel->core.attr.exclude_hv) - ctx |=3D CTX_BIT_HV; - if (evsel->core.attr.exclude_host) - ctx |=3D CTX_BIT_HOST; - if (evsel->core.attr.exclude_idle) - ctx |=3D CTX_BIT_IDLE; - - return ctx; -} - void perf_stat__reset_shadow_stats(void) { memset(&walltime_nsecs_stats, 0, sizeof(walltime_nsecs_stats)); memset(&ru_stats, 0, sizeof(ru_stats)); } =20 -static enum stat_type evsel__stat_type(struct evsel *evsel) -{ - /* Fake perf_hw_cache_op_id values for use with evsel__match. */ - u64 PERF_COUNT_hw_cache_l1d_miss =3D PERF_COUNT_HW_CACHE_L1D | - ((PERF_COUNT_HW_CACHE_OP_READ) << 8) | - ((PERF_COUNT_HW_CACHE_RESULT_MISS) << 16); - u64 PERF_COUNT_hw_cache_l1i_miss =3D PERF_COUNT_HW_CACHE_L1I | - ((PERF_COUNT_HW_CACHE_OP_READ) << 8) | - ((PERF_COUNT_HW_CACHE_RESULT_MISS) << 16); - u64 PERF_COUNT_hw_cache_ll_miss =3D PERF_COUNT_HW_CACHE_LL | - ((PERF_COUNT_HW_CACHE_OP_READ) << 8) | - ((PERF_COUNT_HW_CACHE_RESULT_MISS) << 16); - u64 PERF_COUNT_hw_cache_dtlb_miss =3D PERF_COUNT_HW_CACHE_DTLB | - ((PERF_COUNT_HW_CACHE_OP_READ) << 8) | - ((PERF_COUNT_HW_CACHE_RESULT_MISS) << 16); - u64 PERF_COUNT_hw_cache_itlb_miss =3D PERF_COUNT_HW_CACHE_ITLB | - ((PERF_COUNT_HW_CACHE_OP_READ) << 8) | - ((PERF_COUNT_HW_CACHE_RESULT_MISS) << 16); - - if (evsel__is_clock(evsel)) - return STAT_NSECS; - else if (evsel__match(evsel, HARDWARE, HW_CPU_CYCLES)) - return STAT_CYCLES; - else if (evsel__match(evsel, HARDWARE, HW_INSTRUCTIONS)) - return STAT_INSTRUCTIONS; - else if (evsel__match(evsel, HARDWARE, HW_STALLED_CYCLES_FRONTEND)) - return STAT_STALLED_CYCLES_FRONT; - else if (evsel__match(evsel, HARDWARE, HW_STALLED_CYCLES_BACKEND)) - return STAT_STALLED_CYCLES_BACK; - else if (evsel__match(evsel, HARDWARE, HW_BRANCH_INSTRUCTIONS)) - return STAT_BRANCHES; - else if (evsel__match(evsel, HARDWARE, HW_BRANCH_MISSES)) - return STAT_BRANCH_MISS; - else if (evsel__match(evsel, HARDWARE, HW_CACHE_REFERENCES)) - return STAT_CACHE_REFS; - else if (evsel__match(evsel, HARDWARE, HW_CACHE_MISSES)) - return STAT_CACHE_MISSES; - else if (evsel__match(evsel, HW_CACHE, HW_CACHE_L1D)) - return STAT_L1_DCACHE; - else if (evsel__match(evsel, HW_CACHE, HW_CACHE_L1I)) - return STAT_L1_ICACHE; - else if (evsel__match(evsel, HW_CACHE, HW_CACHE_LL)) - return STAT_LL_CACHE; - else if (evsel__match(evsel, HW_CACHE, HW_CACHE_DTLB)) - return STAT_DTLB_CACHE; - else if (evsel__match(evsel, HW_CACHE, HW_CACHE_ITLB)) - return STAT_ITLB_CACHE; - else if (evsel__match(evsel, HW_CACHE, hw_cache_l1d_miss)) - return STAT_L1D_MISS; - else if (evsel__match(evsel, HW_CACHE, hw_cache_l1i_miss)) - return STAT_L1I_MISS; - else if (evsel__match(evsel, HW_CACHE, hw_cache_ll_miss)) - return STAT_LL_MISS; - else if (evsel__match(evsel, HW_CACHE, hw_cache_dtlb_miss)) - return STAT_DTLB_MISS; - else if (evsel__match(evsel, HW_CACHE, hw_cache_itlb_miss)) - return STAT_ITLB_MISS; - return STAT_NONE; -} - -static enum metric_threshold_classify get_ratio_thresh(const double ratios= [3], double val) -{ - assert(ratios[0] > ratios[1]); - assert(ratios[1] > ratios[2]); - - return val > ratios[1] - ? (val > ratios[0] ? METRIC_THRESHOLD_BAD : METRIC_THRESHOLD_NEARLY_BAD) - : (val > ratios[2] ? METRIC_THRESHOLD_LESS_GOOD : METRIC_THRESHOLD_GOOD); -} - -static double find_stat(const struct evsel *evsel, int aggr_idx, enum stat= _type type) -{ - struct evsel *cur; - int evsel_ctx =3D evsel_context(evsel); - struct perf_pmu *evsel_pmu =3D evsel__find_pmu(evsel); - - evlist__for_each_entry(evsel->evlist, cur) { - struct perf_stat_aggr *aggr; - - /* Ignore the evsel that is being searched from. */ - if (evsel =3D=3D cur) - continue; - - /* Ignore evsels that are part of different groups. */ - if (evsel->core.leader->nr_members > 1 && - evsel->core.leader !=3D cur->core.leader) - continue; - /* Ignore evsels with mismatched modifiers. */ - if (evsel_ctx !=3D evsel_context(cur)) - continue; - /* Ignore if not the cgroup we're looking for. */ - if (evsel->cgrp !=3D cur->cgrp) - continue; - /* Ignore if not the stat we're looking for. */ - if (type !=3D evsel__stat_type(cur)) - continue; - - /* - * Except the SW CLOCK events, - * ignore if not the PMU we're looking for. - */ - if ((type !=3D STAT_NSECS) && (evsel_pmu !=3D evsel__find_pmu(cur))) - continue; - - aggr =3D &cur->stats->aggr[aggr_idx]; - if (type =3D=3D STAT_NSECS) - return aggr->counts.val; - return aggr->counts.val * cur->scale; - } - return 0.0; -} - -static void print_ratio(struct perf_stat_config *config, - const struct evsel *evsel, int aggr_idx, - double numerator, struct perf_stat_output_ctx *out, - enum stat_type denominator_type, - const double thresh_ratios[3], const char *_unit) -{ - double denominator =3D find_stat(evsel, aggr_idx, denominator_type); - double ratio =3D 0; - enum metric_threshold_classify thresh =3D METRIC_THRESHOLD_UNKNOWN; - const char *fmt =3D NULL; - const char *unit =3D NULL; - - if (numerator && denominator) { - ratio =3D numerator / denominator * 100.0; - thresh =3D get_ratio_thresh(thresh_ratios, ratio); - fmt =3D "%7.2f%%"; - unit =3D _unit; - } - out->print_metric(config, out->ctx, thresh, fmt, unit, ratio); -} - -static void print_stalled_cycles_front(struct perf_stat_config *config, - const struct evsel *evsel, - int aggr_idx, double stalled, - struct perf_stat_output_ctx *out) -{ - const double thresh_ratios[3] =3D {50.0, 30.0, 10.0}; - - print_ratio(config, evsel, aggr_idx, stalled, out, STAT_CYCLES, thresh_ra= tios, - "frontend cycles idle"); -} - -static void print_stalled_cycles_back(struct perf_stat_config *config, - const struct evsel *evsel, - int aggr_idx, double stalled, - struct perf_stat_output_ctx *out) -{ - const double thresh_ratios[3] =3D {75.0, 50.0, 20.0}; - - print_ratio(config, evsel, aggr_idx, stalled, out, STAT_CYCLES, thresh_ra= tios, - "backend cycles idle"); -} - -static void print_branch_miss(struct perf_stat_config *config, - const struct evsel *evsel, - int aggr_idx, double misses, - struct perf_stat_output_ctx *out) -{ - const double thresh_ratios[3] =3D {20.0, 10.0, 5.0}; - - print_ratio(config, evsel, aggr_idx, misses, out, STAT_BRANCHES, thresh_r= atios, - "of all branches"); -} - -static void print_l1d_miss(struct perf_stat_config *config, - const struct evsel *evsel, - int aggr_idx, double misses, - struct perf_stat_output_ctx *out) -{ - const double thresh_ratios[3] =3D {20.0, 10.0, 5.0}; - - print_ratio(config, evsel, aggr_idx, misses, out, STAT_L1_DCACHE, thresh_= ratios, - "of all L1-dcache accesses"); -} - -static void print_l1i_miss(struct perf_stat_config *config, - const struct evsel *evsel, - int aggr_idx, double misses, - struct perf_stat_output_ctx *out) -{ - const double thresh_ratios[3] =3D {20.0, 10.0, 5.0}; - - print_ratio(config, evsel, aggr_idx, misses, out, STAT_L1_ICACHE, thresh_= ratios, - "of all L1-icache accesses"); -} - -static void print_ll_miss(struct perf_stat_config *config, - const struct evsel *evsel, - int aggr_idx, double misses, - struct perf_stat_output_ctx *out) -{ - const double thresh_ratios[3] =3D {20.0, 10.0, 5.0}; - - print_ratio(config, evsel, aggr_idx, misses, out, STAT_LL_CACHE, thresh_r= atios, - "of all LL-cache accesses"); -} - -static void print_dtlb_miss(struct perf_stat_config *config, - const struct evsel *evsel, - int aggr_idx, double misses, - struct perf_stat_output_ctx *out) -{ - const double thresh_ratios[3] =3D {20.0, 10.0, 5.0}; - - print_ratio(config, evsel, aggr_idx, misses, out, STAT_DTLB_CACHE, thresh= _ratios, - "of all dTLB cache accesses"); -} - -static void print_itlb_miss(struct perf_stat_config *config, - const struct evsel *evsel, - int aggr_idx, double misses, - struct perf_stat_output_ctx *out) -{ - const double thresh_ratios[3] =3D {20.0, 10.0, 5.0}; - - print_ratio(config, evsel, aggr_idx, misses, out, STAT_ITLB_CACHE, thresh= _ratios, - "of all iTLB cache accesses"); -} - -static void print_cache_miss(struct perf_stat_config *config, - const struct evsel *evsel, - int aggr_idx, double misses, - struct perf_stat_output_ctx *out) -{ - const double thresh_ratios[3] =3D {20.0, 10.0, 5.0}; - - print_ratio(config, evsel, aggr_idx, misses, out, STAT_CACHE_REFS, thresh= _ratios, - "of all cache refs"); -} - -static void print_instructions(struct perf_stat_config *config, - const struct evsel *evsel, - int aggr_idx, double instructions, - struct perf_stat_output_ctx *out) -{ - print_metric_t print_metric =3D out->print_metric; - void *ctxp =3D out->ctx; - double cycles =3D find_stat(evsel, aggr_idx, STAT_CYCLES); - double max_stalled =3D max(find_stat(evsel, aggr_idx, STAT_STALLED_CYCLES= _FRONT), - find_stat(evsel, aggr_idx, STAT_STALLED_CYCLES_BACK)); - - if (cycles) { - print_metric(config, ctxp, METRIC_THRESHOLD_UNKNOWN, "%7.2f ", - "insn per cycle", instructions / cycles); - } else { - print_metric(config, ctxp, METRIC_THRESHOLD_UNKNOWN, /*fmt=3D*/NULL, - "insn per cycle", 0); - } - if (max_stalled && instructions) { - if (out->new_line) - out->new_line(config, ctxp); - print_metric(config, ctxp, METRIC_THRESHOLD_UNKNOWN, "%7.2f ", - "stalled cycles per insn", max_stalled / instructions); - } -} - -static void print_cycles(struct perf_stat_config *config, - const struct evsel *evsel, - int aggr_idx, double cycles, - struct perf_stat_output_ctx *out) -{ - double nsecs =3D find_stat(evsel, aggr_idx, STAT_NSECS); - - if (cycles && nsecs) { - double ratio =3D cycles / nsecs; - - out->print_metric(config, out->ctx, METRIC_THRESHOLD_UNKNOWN, "%8.3f", - "GHz", ratio); - } else { - out->print_metric(config, out->ctx, METRIC_THRESHOLD_UNKNOWN, /*fmt=3D*/= NULL, - "GHz", 0); - } -} - -static void print_nsecs(struct perf_stat_config *config, - const struct evsel *evsel, - int aggr_idx __maybe_unused, double nsecs, - struct perf_stat_output_ctx *out) -{ - print_metric_t print_metric =3D out->print_metric; - void *ctxp =3D out->ctx; - double wall_time =3D avg_stats(&walltime_nsecs_stats); - - if (wall_time) { - print_metric(config, ctxp, METRIC_THRESHOLD_UNKNOWN, "%8.3f", "CPUs util= ized", - nsecs / (wall_time * evsel->scale)); - } else { - print_metric(config, ctxp, METRIC_THRESHOLD_UNKNOWN, /*fmt=3D*/NULL, - "CPUs utilized", 0); - } -} - static int prepare_metric(const struct metric_expr *mexp, const struct evsel *evsel, struct expr_parse_ctx *pctx, @@ -682,56 +337,15 @@ void *perf_stat__print_shadow_stats_metricgroup(struc= t perf_stat_config *config, =20 void perf_stat__print_shadow_stats(struct perf_stat_config *config, struct evsel *evsel, - double avg, int aggr_idx, + int aggr_idx, struct perf_stat_output_ctx *out) { - typedef void (*stat_print_function_t)(struct perf_stat_config *config, - const struct evsel *evsel, - int aggr_idx, double misses, - struct perf_stat_output_ctx *out); - static const stat_print_function_t stat_print_function[STAT_MAX] =3D { - [STAT_INSTRUCTIONS] =3D print_instructions, - [STAT_BRANCH_MISS] =3D print_branch_miss, - [STAT_L1D_MISS] =3D print_l1d_miss, - [STAT_L1I_MISS] =3D print_l1i_miss, - [STAT_DTLB_MISS] =3D print_dtlb_miss, - [STAT_ITLB_MISS] =3D print_itlb_miss, - [STAT_LL_MISS] =3D print_ll_miss, - [STAT_CACHE_MISSES] =3D print_cache_miss, - [STAT_STALLED_CYCLES_FRONT] =3D print_stalled_cycles_front, - [STAT_STALLED_CYCLES_BACK] =3D print_stalled_cycles_back, - [STAT_CYCLES] =3D print_cycles, - [STAT_NSECS] =3D print_nsecs, - }; print_metric_t print_metric =3D out->print_metric; void *ctxp =3D out->ctx; - int num =3D 1; + int num =3D 0; =20 - if (config->iostat_run) { + if (config->iostat_run) iostat_print_metric(config, evsel, out); - } else { - stat_print_function_t fn =3D stat_print_function[evsel__stat_type(evsel)= ]; - - if (fn) - fn(config, evsel, aggr_idx, avg, out); - else { - double nsecs =3D find_stat(evsel, aggr_idx, STAT_NSECS); - - if (nsecs) { - char unit =3D ' '; - char unit_buf[10] =3D "/sec"; - double ratio =3D convert_unit_double(1000000000.0 * avg / nsecs, - &unit); - - if (unit !=3D ' ') - snprintf(unit_buf, sizeof(unit_buf), "%c/sec", unit); - print_metric(config, ctxp, METRIC_THRESHOLD_UNKNOWN, "%8.3f", - unit_buf, ratio); - } else { - num =3D 0; - } - } - } =20 perf_stat__print_shadow_stats_metricgroup(config, evsel, aggr_idx, &num, NULL, out); diff --git a/tools/perf/util/stat.h b/tools/perf/util/stat.h index 230474f49315..b42da4a29c44 100644 --- a/tools/perf/util/stat.h +++ b/tools/perf/util/stat.h @@ -184,7 +184,7 @@ struct perf_stat_output_ctx { =20 void perf_stat__print_shadow_stats(struct perf_stat_config *config, struct evsel *evsel, - double avg, int aggr_idx, + int aggr_idx, struct perf_stat_output_ctx *out); bool perf_stat__skip_metric_event(struct evsel *evsel, u64 ena, u64 run); void *perf_stat__print_shadow_stats_metricgroup(struct perf_stat_config *c= onfig, --=20 2.51.2.1041.gc1ab5b90ca-goog From nobody Tue Nov 11 11:28:57 2025 Received: from mail-pl1-f201.google.com (mail-pl1-f201.google.com [209.85.214.201]) (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 CAD8734B41E for ; Tue, 11 Nov 2025 04:04:41 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.214.201 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1762833883; cv=none; b=pG2/QPGezvwdo970goC9Signhpe9PMna88VotE817z8oKKqbEE2xzSBNtCTYAHi3MlLtlpaLPvVBqCuRi8f1FsA27aUuymr2bnfx+x9wKC5ikP5Lnf64O/CAmdaYSuw/e4uUdt7F/7At/sgmwRcE3jJ7IyAdl2ufhQ3Cp1hh608= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1762833883; c=relaxed/simple; bh=cZF5rXK0ot4f5Ll54hzIMYMkIxY+GXfM/HIllfqaohs=; h=Date:In-Reply-To:Mime-Version:References:Message-ID:Subject:From: To:Content-Type; b=QVo5LVxfFMC1eQA+fVRcFNkCuhoWyGH4LVgFlTkOdaABoNimyJttCfPg5x9yJdC4I4HwQSJtLZt8J/4eUVRHfwlmobr1mmbEv8SQ2HkU+nsSFrZ8ZxHxgdfljn6uAzsNCDgeeK+HuzFjYYsK/U43oobTLrNY54hLbnDS8le4tfc= 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=tMHZ+qj8; arc=none smtp.client-ip=209.85.214.201 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="tMHZ+qj8" Received: by mail-pl1-f201.google.com with SMTP id d9443c01a7336-297b35951b7so62947525ad.3 for ; Mon, 10 Nov 2025 20:04:41 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1762833881; x=1763438681; darn=vger.kernel.org; h=to:from:subject:message-id:references:mime-version:in-reply-to:date :from:to:cc:subject:date:message-id:reply-to; bh=CJrFojSq6nlkAfypEMFi1zUKz5050bK3dx0VztPcSL4=; b=tMHZ+qj8E2QRfuhNUb4UXJNNDTw2JFFhWTcQT612muYlXUcXRAj772/9qMEvXT6Z3Q zL4BGjDwroZe4GLZueozy4yLTGLuTeOyPv2zsOGnSca339MkO4DTiPT0YXEmzwpimVGC +vN/slGRzVT+HZGC57XXAiUZQ1sGvw6i03BJHAUmvfk6HB6thkujsBRXlwtwiEOJ0Fhs aWMCJI6Y6QrmWbwh2kuGcA7J9NCSRIMATVmk7IvqFltowbluJe+zVFfuYzerWt+ds7f8 P5qq5+YXTWF7GniPmE0j/QNhBljK6Yq1Vlx0DjOYPatLdcs2IKhmaPmIu0xxYolDFM7I Rqlg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1762833881; x=1763438681; h=to:from:subject:message-id:references:mime-version:in-reply-to:date :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=CJrFojSq6nlkAfypEMFi1zUKz5050bK3dx0VztPcSL4=; b=byANOIeWNAiY5aLXZt3Fn7zPd0DBgfpXQvY4QTBLzfGH25yga2Tptug3ln09mXvo2I AxX1kESK+orOlKidWFov7oF5M6pT/fse05kvYFda64MpQzglFhjq1GGlTI68hXJ0qXXf XnkwM8JY85C+yXbn9kOPG20QOS4CMpMp6en6j7O8m5swFQPrkRnF85YofSa8YWJQsDeZ edNAgZUgJKgKzd2yWUX13gNOH25zs+bpjX9O6IYkOjvfEPLC9WDL7NmrKsGSjVJ8wdwZ WLpZKmMqElTUV6aopeks9ZJoos0SA3pifvtcngjjJBhTSFHX64uxT3KdtkebS0uoLxx+ Hx8Q== X-Forwarded-Encrypted: i=1; AJvYcCUFLYELFWix3Q8Su1w2anIl/TnfsvnDoJvnMyp5qEjegtdxzwYB8DXbCKMSKLGdNezFzoxsP7cr1n94Oi8=@vger.kernel.org X-Gm-Message-State: AOJu0Yz85k9FLkrGUK9CB4UqQ95rDznsaNALEHmRvgT4wM2kEhV/8pP6 E+DGVbpf8r2gkglmdE7kHIN6JhXBoMja/vaqHwJ7DoSMJ6/MrDgd3V+ElJTVRuF7ONv+1HmPY4d PY55XgHQ2IA== X-Google-Smtp-Source: AGHT+IF4k0uFcaS7/tC4cyXC8wsIhicVjSl/j9UnN2s9TsgyF6DRp0WIZCwddGEHn70Zm1RF4d0DU8ZooCF5 X-Received: from dlbqd3.prod.google.com ([2002:a05:7023:b03:b0:119:49ca:6b8c]) (user=irogers job=prod-delivery.src-stubby-dispatcher) by 2002:a17:902:da8c:b0:298:3e3a:ae6 with SMTP id d9443c01a7336-2983e3a0b4bmr30376125ad.48.1762833881039; Mon, 10 Nov 2025 20:04:41 -0800 (PST) Date: Mon, 10 Nov 2025 20:04:07 -0800 In-Reply-To: <20251111040417.270945-1-irogers@google.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 References: <20251111040417.270945-1-irogers@google.com> X-Mailer: git-send-email 2.51.2.1041.gc1ab5b90ca-goog Message-ID: <20251111040417.270945-9-irogers@google.com> Subject: [PATCH v3 08/18] perf stat: Fix default metricgroup display on hybrid From: Ian Rogers To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Namhyung Kim , Alexander Shishkin , Jiri Olsa , Ian Rogers , Adrian Hunter , James Clark , Xu Yang , Chun-Tse Shao , Thomas Richter , Sumanth Korikkar , Collin Funk , Thomas Falcon , Howard Chu , Dapeng Mi , Levi Yun , Yang Li , linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org, Andi Kleen , Weilin Wang Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" The logic to skip output of a default metric line was firing on Alderlake and not displaying 'TopdownL1 (cpu_atom)'. Remove the need_full_name check as it is equivalent to the different PMU test in the cases we care about, merge the 'if's and flip the evsel of the PMU test. The 'if' is now basically saying, if the output matches the last printed output then skip the output. Before: ``` TopdownL1 (cpu_core) # 11.3 % tma_bad_spe= culation # 24.3 % tma_fronten= d_bound TopdownL1 (cpu_core) # 33.9 % tma_backend= _bound # 30.6 % tma_retiring # 42.2 % tma_backend= _bound # 25.0 % tma_fronten= d_bound (49.81%) # 12.8 % tma_bad_spe= culation # 20.0 % tma_retirin= g (59.46%) ``` After: ``` TopdownL1 (cpu_core) # 8.3 % tma_bad_spe= culation # 43.7 % tma_fronten= d_bound # 30.7 % tma_backend= _bound # 17.2 % tma_retiring TopdownL1 (cpu_atom) # 31.9 % tma_backend= _bound # 37.6 % tma_fronten= d_bound (49.66%) # 18.0 % tma_bad_spe= culation # 12.6 % tma_retirin= g (59.58%) ``` Signed-off-by: Ian Rogers --- tools/perf/util/stat-shadow.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) diff --git a/tools/perf/util/stat-shadow.c b/tools/perf/util/stat-shadow.c index afbc49e8cb31..c1547128c396 100644 --- a/tools/perf/util/stat-shadow.c +++ b/tools/perf/util/stat-shadow.c @@ -256,11 +256,9 @@ static void perf_stat__print_metricgroup_header(struct= perf_stat_config *config, * event. Only align with other metics from * different metric events. */ - if (last_name && !strcmp(last_name, name)) { - if (!need_full_name || last_pmu !=3D evsel->pmu) { - out->print_metricgroup_header(config, ctxp, NULL); - return; - } + if (last_name && !strcmp(last_name, name) && last_pmu =3D=3D evsel->pmu) { + out->print_metricgroup_header(config, ctxp, NULL); + return; } =20 if (need_full_name && evsel->pmu) --=20 2.51.2.1041.gc1ab5b90ca-goog From nobody Tue Nov 11 11:28:57 2025 Received: from mail-pl1-f202.google.com (mail-pl1-f202.google.com [209.85.214.202]) (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 82A5234B68F for ; Tue, 11 Nov 2025 04:04:43 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.214.202 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1762833885; cv=none; b=OIgnd3M4cTT4PWjjdSsajMY31eWOWNj9/u4J7GVueC0j4E9NAcqSRf+IxM3Tioesz+3i0rNLnUh0OKrj6A9fH04kYq6NXQyDrRbgVJSE9hjkQD02lK8howSiBWNIiHVuMkYe0QNC4jduzxrAyPrDOME6kyAOPo5iurFbuo7E4vo= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1762833885; c=relaxed/simple; bh=z5ai+zFpH2QBuK5+a38Mk4V/HsbjWKL3F0Fj9eLHmOU=; h=Date:In-Reply-To:Mime-Version:References:Message-ID:Subject:From: To:Content-Type; b=amspQirVeoJ5lnmQKMrkg/GKza3UswdRkH0O05vyHZBaAdaB5KYTtu6wciwFlCAQPEvPFv/J06+yuCI6qf+qpvtbOkscYi1JDuIFeMZVLJ8L9YXPre1QN4Oq5sFy+nEdpV/aSArCbpO+nukiy8T46qwfeuWRmnPgHe4znkyJRCk= 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=IDrUy1Ag; arc=none smtp.client-ip=209.85.214.202 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="IDrUy1Ag" Received: by mail-pl1-f202.google.com with SMTP id d9443c01a7336-295592eb5dbso34301285ad.0 for ; Mon, 10 Nov 2025 20:04:43 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1762833883; x=1763438683; darn=vger.kernel.org; h=to:from:subject:message-id:references:mime-version:in-reply-to:date :from:to:cc:subject:date:message-id:reply-to; bh=ISmadZFruDrBTfreeqdnoWbjJP8/Vr5LKfdKCjHv7vI=; b=IDrUy1Ag++ETbjpvi4DjUAPyxQmO5YO/mEGDjD61V2YDudaWqmvr3EbNtmVRtDh5Pa UyrZdmF9i690iBwiWDH++QZP8JHcz/1ChbbfYs6wf4eMfNu8+Liicem8EEmzUQB3Pw5h z0241W/d3tA75ZSKa5XN1UvGOJAk8z7D96/DEW3RhXvKLelXfc+CYxpUQIz7LuIauAFy zbGB45kcn6PmRbxDhCRvO9bh/KZhoehF/9Gywv5tJ8OSHiLxZbEC2aGCdJbWBzHKucb1 hieZg0dMpRbTMEn2A+pHhBKvxftKeviNzR858VHCLiv3x9PzgJe4UZfq19CGWhi2nNoK UT/A== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1762833883; x=1763438683; h=to:from:subject:message-id:references:mime-version:in-reply-to:date :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=ISmadZFruDrBTfreeqdnoWbjJP8/Vr5LKfdKCjHv7vI=; b=nDI8D1ajRtLQ0iP+N90c0AgfO6P9Fcq3PXKSdddKBl+qB1lVuMK20L0+jvXvoE29UJ RjUgOfKyPgXp3B8/gbtD+eyxx4Dq6aDlNLcdMfIYnZVtNHpaaWtV7XHCya7/YxqUJeWM LpY2st9GHatWw9PNsSdn7KumKHaZOI6mLgehfKfpwW8bSVnajA7QBodGlJ6wsf797mOO mauL1wIZxiZ5xCvm4BFSjyTU86icHXNQhV46WRNjbZOJIEvbyDVkr6Ld8Ehohy0u0Cx0 8znvYv+uFmcvB5gSED9jQZ5r/1RhQXyd7cdv499jD0RlECyrIY0cdXCgybBniCxqKqEC 3wCQ== X-Forwarded-Encrypted: i=1; AJvYcCXfagqsWle9YMKumWcg1YFxxyuCvcJfw63W6ZDSINHfRRYPG/c554hejBajeAbbaK+3Q87Nhb2fnm0eyAc=@vger.kernel.org X-Gm-Message-State: AOJu0YywEfxTH6LeElk2VmrpTRxe601pmv2qpc2zAeguVb+l9dDXD20+ WF8QuXYtSe6lyEoOJSSRkYOofEmrS1M1n4lPcchsqT8DQrIH5c2D6fHbjwMKy5MDKujJrTEC2s6 DwiKf5mvXSQ== X-Google-Smtp-Source: AGHT+IGiBXsZJvHje+7C1vX1eoueo2kY601eMqrQfibmvzA4SzBeU37Zi/9PcAuxsjuO8ZRjv+M6a392ts8N X-Received: from dlah8.prod.google.com ([2002:a05:701b:2608:b0:11a:51f9:d69]) (user=irogers job=prod-delivery.src-stubby-dispatcher) by 2002:a17:902:da4b:b0:27e:f018:d2fb with SMTP id d9443c01a7336-297e540dbf8mr130637515ad.6.1762833882935; Mon, 10 Nov 2025 20:04:42 -0800 (PST) Date: Mon, 10 Nov 2025 20:04:08 -0800 In-Reply-To: <20251111040417.270945-1-irogers@google.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 References: <20251111040417.270945-1-irogers@google.com> X-Mailer: git-send-email 2.51.2.1041.gc1ab5b90ca-goog Message-ID: <20251111040417.270945-10-irogers@google.com> Subject: [PATCH v3 09/18] perf stat: Sort default events/metrics From: Ian Rogers To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Namhyung Kim , Alexander Shishkin , Jiri Olsa , Ian Rogers , Adrian Hunter , James Clark , Xu Yang , Chun-Tse Shao , Thomas Richter , Sumanth Korikkar , Collin Funk , Thomas Falcon , Howard Chu , Dapeng Mi , Levi Yun , Yang Li , linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org, Andi Kleen , Weilin Wang Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" To improve the readability of default events/metrics, sort the evsels after the Default metric groups have be parsed. Before: ``` $ perf stat -a sleep 1 Performance counter stats for 'system wide': 21,194 context-switches # 752.1 cs/sec = cs_per_second TopdownL1 (cpu_core) # 9.4 % tma_bad_spe= culation # 25.0 % tma_fronten= d_bound # 37.0 % tma_backend= _bound # 28.7 % tma_retiring 6,371 page-faults # 226.1 faults/= sec page_faults_per_second 734,456,525 cpu_atom/cpu-cycles/ # 0.0 GHz cy= cles_frequency (49.77%) 1,679,085,181 cpu_core/cpu-cycles/ # 0.1 GHz cy= cles_frequency TopdownL1 (cpu_atom) # 19.2 % tma_bad_spe= culation # 15.1 % tma_retirin= g (50.15%) # 32.9 % tma_backend= _bound # 32.9 % tma_fronten= d_bound (50.34%) 86,758,824 cpu_atom/branches/ # 3.1 K/sec = branch_frequency (60.26%) 524,281,539 cpu_core/branches/ # 18.6 K/sec = branch_frequency 1,458 cpu-migrations # 51.7 migrati= ons/sec migrations_per_second 28,178,124,975 cpu-clock # 28.0 CPUs C= PUs_utilized 776,037,182 cpu_atom/cpu-cycles/ # 0.6 instruc= tions insn_per_cycle (60.18%) 1,679,168,140 cpu_core/cpu-cycles/ # 1.8 instruc= tions insn_per_cycle 4,045,615 cpu_atom/branches-misses/ # 5.3 % bran= ch_miss_rate (49.65%) 6,656,795 cpu_core/branches-misses/ # 1.3 % bran= ch_miss_rate 1.007340329 seconds time elapsed ``` After: ``` $ perf stat -a sleep 1 Performance counter stats for 'system wide': 25,701 context-switches # 911.8 cs/sec = cs_per_second 28,187,404,943 cpu-clock # 28.0 CPUs C= PUs_utilized 2,053 cpu-migrations # 72.8 migrati= ons/sec migrations_per_second 12,390 page-faults # 439.6 faults/= sec page_faults_per_second 592,082,798 cpu_core/branches/ # 21.0 K/sec = branch_frequency 7,762,204 cpu_core/branches-misses/ # 1.3 % bran= ch_miss_rate 1,925,833,804 cpu_core/cpu-cycles/ # 0.1 GHz cy= cles_frequency 1,925,848,650 cpu_core/cpu-cycles/ # 1.7 instruc= tions insn_per_cycle 95,449,119 cpu_atom/branches/ # 3.4 K/sec = branch_frequency (59.78%) 4,278,932 cpu_atom/branches-misses/ # 4.3 % bran= ch_miss_rate (50.26%) 980,441,753 cpu_atom/cpu-cycles/ # 0.0 GHz cy= cles_frequency (50.34%) 1,091,626,599 cpu_atom/cpu-cycles/ # 0.6 instruc= tions insn_per_cycle (50.37%) TopdownL1 (cpu_core) # 9.1 % tma_bad_spe= culation # 27.3 % tma_fronten= d_bound # 35.7 % tma_backend= _bound # 27.9 % tma_retiring TopdownL1 (cpu_atom) # 31.1 % tma_backend= _bound # 34.3 % tma_fronten= d_bound (49.74%) # 24.1 % tma_bad_spe= culation # 10.5 % tma_retirin= g (59.57%) ``` Signed-off-by: Ian Rogers --- tools/perf/builtin-stat.c | 32 ++++++++++++++++++++++++++++++++ 1 file changed, 32 insertions(+) diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c index 7862094b93c8..095016b2209e 100644 --- a/tools/perf/builtin-stat.c +++ b/tools/perf/builtin-stat.c @@ -74,6 +74,7 @@ #include "util/intel-tpebs.h" #include "asm/bug.h" =20 +#include #include #include #include @@ -1857,6 +1858,35 @@ static int perf_stat_init_aggr_mode_file(struct perf= _stat *st) return 0; } =20 +static int default_evlist_evsel_cmp(void *priv __maybe_unused, + const struct list_head *l, + const struct list_head *r) +{ + const struct perf_evsel *lhs_core =3D container_of(l, struct perf_evsel, = node); + const struct evsel *lhs =3D container_of(lhs_core, struct evsel, core); + const struct perf_evsel *rhs_core =3D container_of(r, struct perf_evsel, = node); + const struct evsel *rhs =3D container_of(rhs_core, struct evsel, core); + + if (evsel__leader(lhs) =3D=3D evsel__leader(rhs)) { + /* Within the same group, respect the original order. */ + return lhs_core->idx - rhs_core->idx; + } + + /* Sort default metrics evsels first, and default show events before thos= e. */ + if (lhs->default_metricgroup !=3D rhs->default_metricgroup) + return lhs->default_metricgroup ? -1 : 1; + + if (lhs->default_show_events !=3D rhs->default_show_events) + return lhs->default_show_events ? -1 : 1; + + /* Sort by PMU type (prefers legacy types first). */ + if (lhs->pmu !=3D rhs->pmu) + return lhs->pmu->type - rhs->pmu->type; + + /* Sort by name. */ + return strcmp(evsel__name((struct evsel *)lhs), evsel__name((struct evsel= *)rhs)); +} + /* * Add default events, if there were no attributes specified or * if -d/--detailed, -d -d or -d -d -d is used: @@ -2023,6 +2053,8 @@ static int add_default_events(void) &metric_evlist->metric_events); evlist__delete(metric_evlist); } + list_sort(/*priv=3D*/NULL, &evlist->core.entries, default_evlist_evsel_c= mp); + } out: if (!ret) { --=20 2.51.2.1041.gc1ab5b90ca-goog From nobody Tue Nov 11 11:28:57 2025 Received: from mail-pl1-f202.google.com (mail-pl1-f202.google.com [209.85.214.202]) (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 7A09333374B for ; Tue, 11 Nov 2025 04:04:45 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.214.202 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1762833887; cv=none; b=mu56RcywgxI/+Uid/pWnrLzU5DXjKTbBhgEtSj1ieE0cY2eNDnY5aXHvKG8qWhSpTkOTF3XLEasMxU/EEK2r7RvQRDCQ4mVNlKrvYY7YTeD6og/qBKbnN1Y7zsKuQAPY8lTHvM0D17xJ6vUlpfkWDLCMBUUvZKK3IhVW2SS9ymY= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1762833887; c=relaxed/simple; bh=3kfszh9GvrFi8Y4HcNCDpFFabOoNANjoDq9+KWDc9ZY=; h=Date:In-Reply-To:Mime-Version:References:Message-ID:Subject:From: To:Content-Type; b=jkZ+o0NDiuaX7jACSFQrwvw4rlbiRKkLBaYz01Uv3LMOkL8iKHMf8FpzwmmsLTTGUCTg5mfO+DOiq/0FiDTtPC0qkR5DiR1EfavdjySuVDI4PZkOqZjY4hVOgWMTovaV/RJZf5nXmuQySnrfDR4gbP5t90T9dxgrU6a833fUdtI= 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=l3Cy+iYw; arc=none smtp.client-ip=209.85.214.202 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="l3Cy+iYw" Received: by mail-pl1-f202.google.com with SMTP id d9443c01a7336-297dde580c8so69843955ad.1 for ; Mon, 10 Nov 2025 20:04:45 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1762833885; x=1763438685; darn=vger.kernel.org; h=to:from:subject:message-id:references:mime-version:in-reply-to:date :from:to:cc:subject:date:message-id:reply-to; bh=blAae1b4UuHw+1SOJAzDhtKVe5EnHpicO6Ta+cSn2vM=; b=l3Cy+iYw7wDFdkgGm4ZOVXDbymTf+r+hi+ZjpREjXkIc21KUTihrrP/mzXUX4hZmiv q9dVaexZmg6WTzF7BLwn9chVMddfflsd5aYuZYkHlZ+Ssm9eOoPue5RgZINkerKqddXa Hvjk8xlgbRrovEx2M3HWkgtBoI5KSyvtrdEOAoW8SHKAVDuwz+xjTL1h9uWfsiveApGS IKWm1sMKnxzisEZUq94TgBlwK6IvsS0TS8YfoZvO80cRLpZNlVoIoLXdCdIy28tMz08M fPIIEo5qSm1SO6cIEstoxq7+t7Ue7uF4HcTnumkzceeMhvjPryi7ZPl1srXLqRWBnWHH qwWA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1762833885; x=1763438685; h=to:from:subject:message-id:references:mime-version:in-reply-to:date :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=blAae1b4UuHw+1SOJAzDhtKVe5EnHpicO6Ta+cSn2vM=; b=G3+S+DZWEijLikoQUumFgx4W4AJd5hAPBjiWtuvdyLPN/nmgKNlnp2+aWv83gf/rD7 dCZp+OePlyJvuNaCdnz9Vup5CpSNfnkO+fvp8LxvVerOp9rxoXkO9e4/Ib4h2DERQhvX 60WM6IdN3uishk2GOxqqdFTj1rXsauH8htBOm9+OTVYIZ4EdTow9k6VcOVd/jpquhSta KijpMvq9NVUqyHSTQ4HPka/jfO5Otei9d/ghv7Z8yNDIZNa6OVAtPwEndglm7E2BuSDi uMYBp9XxxM2JiFb/Ln4brGiIUogy3N8AcZQd70O5HSjWp+DUUZo4CktXuO5VIjSOs7/K 0JyA== X-Forwarded-Encrypted: i=1; AJvYcCXXH/ui5uA1HgreSMwjWLtB6znH0WR6x+fovyqoY4MQkxT8Uou5FfHT8oLiQ291Z88hQIss5JDzd9gi9BQ=@vger.kernel.org X-Gm-Message-State: AOJu0YyvC65z3vdPi+4mlvPpdwjBCLZxfCsHenkA+kLSkUPOgcZE3CBh 7xYP+6jlXFi5Iy/7gLI92/4wmRcf8HSIwLEpw6w5ZLLQmdolwYikaM/wxce9H1K+uLdDUD0lRYY RDHhJweHEow== X-Google-Smtp-Source: AGHT+IFT1U0MlMaA4gQNz5KQzuVH4tab70rQutRTJR46tLMPWDKCUCuZI05XF/rgJDn4c5J837kdXrmVmuyN X-Received: from dycme15.prod.google.com ([2002:a05:693c:240f:b0:2a4:59fe:f2fa]) (user=irogers job=prod-delivery.src-stubby-dispatcher) by 2002:a17:903:2304:b0:294:f30f:ea4b with SMTP id d9443c01a7336-297e540a0efmr111599945ad.8.1762833884788; Mon, 10 Nov 2025 20:04:44 -0800 (PST) Date: Mon, 10 Nov 2025 20:04:09 -0800 In-Reply-To: <20251111040417.270945-1-irogers@google.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 References: <20251111040417.270945-1-irogers@google.com> X-Mailer: git-send-email 2.51.2.1041.gc1ab5b90ca-goog Message-ID: <20251111040417.270945-11-irogers@google.com> Subject: [PATCH v3 10/18] perf stat: Remove "unit" workarounds for metric-only From: Ian Rogers To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Namhyung Kim , Alexander Shishkin , Jiri Olsa , Ian Rogers , Adrian Hunter , James Clark , Xu Yang , Chun-Tse Shao , Thomas Richter , Sumanth Korikkar , Collin Funk , Thomas Falcon , Howard Chu , Dapeng Mi , Levi Yun , Yang Li , linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org, Andi Kleen , Weilin Wang Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" Remove code that tested the "unit" as in KB/sec for certain hard coded metric values and did workarounds. Signed-off-by: Ian Rogers --- tools/perf/util/stat-display.c | 47 ++++++---------------------------- 1 file changed, 8 insertions(+), 39 deletions(-) diff --git a/tools/perf/util/stat-display.c b/tools/perf/util/stat-display.c index eabeab5e6614..b3596f9f5cdd 100644 --- a/tools/perf/util/stat-display.c +++ b/tools/perf/util/stat-display.c @@ -592,42 +592,18 @@ static void print_metricgroup_header_std(struct perf_= stat_config *config, fprintf(config->output, "%*s", MGROUP_LEN - n - 1, ""); } =20 -/* Filter out some columns that don't work well in metrics only mode */ - -static bool valid_only_metric(const char *unit) -{ - if (!unit) - return false; - if (strstr(unit, "/sec") || - strstr(unit, "CPUs utilized")) - return false; - return true; -} - -static const char *fixunit(char *buf, struct evsel *evsel, - const char *unit) -{ - if (!strncmp(unit, "of all", 6)) { - snprintf(buf, 1024, "%s %s", evsel__name(evsel), - unit); - return buf; - } - return unit; -} - static void print_metric_only(struct perf_stat_config *config, void *ctx, enum metric_threshold_classify thresh, const char *fmt, const char *unit, double val) { struct outstate *os =3D ctx; FILE *out =3D os->fh; - char buf[1024], str[1024]; + char str[1024]; unsigned mlen =3D config->metric_only_len; const char *color =3D metric_threshold_classify__color(thresh); =20 - if (!valid_only_metric(unit)) - return; - unit =3D fixunit(buf, os->evsel, unit); + if (!unit) + unit =3D ""; if (mlen < strlen(unit)) mlen =3D strlen(unit) + 1; =20 @@ -643,16 +619,15 @@ static void print_metric_only_csv(struct perf_stat_co= nfig *config __maybe_unused void *ctx, enum metric_threshold_classify thresh __maybe_unused, const char *fmt, - const char *unit, double val) + const char *unit __maybe_unused, double val) { struct outstate *os =3D ctx; FILE *out =3D os->fh; char buf[64], *vals, *ends; - char tbuf[1024]; =20 - if (!valid_only_metric(unit)) + if (!unit) return; - unit =3D fixunit(tbuf, os->evsel, unit); + snprintf(buf, sizeof(buf), fmt ?: "", val); ends =3D vals =3D skip_spaces(buf); while (isdigit(*ends) || *ends =3D=3D '.') @@ -670,13 +645,9 @@ static void print_metric_only_json(struct perf_stat_co= nfig *config __maybe_unuse { struct outstate *os =3D ctx; char buf[64], *ends; - char tbuf[1024]; const char *vals; =20 - if (!valid_only_metric(unit)) - return; - unit =3D fixunit(tbuf, os->evsel, unit); - if (!unit[0]) + if (!unit || !unit[0]) return; snprintf(buf, sizeof(buf), fmt ?: "", val); vals =3D ends =3D skip_spaces(buf); @@ -695,7 +666,6 @@ static void print_metric_header(struct perf_stat_config= *config, const char *unit, double val __maybe_unused) { struct outstate *os =3D ctx; - char tbuf[1024]; =20 /* In case of iostat, print metric header for first root port only */ if (config->iostat_run && @@ -705,9 +675,8 @@ static void print_metric_header(struct perf_stat_config= *config, if (os->evsel->cgrp !=3D os->cgrp) return; =20 - if (!valid_only_metric(unit)) + if (!unit) return; - unit =3D fixunit(tbuf, os->evsel, unit); =20 if (config->json_output) return; --=20 2.51.2.1041.gc1ab5b90ca-goog From nobody Tue Nov 11 11:28:57 2025 Received: from mail-pl1-f202.google.com (mail-pl1-f202.google.com [209.85.214.202]) (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 545DF34C83A for ; Tue, 11 Nov 2025 04:04:47 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.214.202 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1762833888; cv=none; b=Z8HkYzw1vRMOyoY4Qui81DcljfVMb2u55glUR3yVoAPuonf2EnJOi51gEhiu9DnwGWy/Oh3apcBzfDmVE3Q4YkPPb3VKIggxrv06TXyQEwPeX9pN+qspNowGO141Aj43L4Fb2FemhIi/4o8nAFhYAGWIbDArkD27ToG2+V82Rq4= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1762833888; c=relaxed/simple; bh=vsAC+bALI30qElPFV06AjAurdz8WlWhRwpAQxUMz2cg=; h=Date:In-Reply-To:Mime-Version:References:Message-ID:Subject:From: To:Content-Type; b=BJoRtYWCPwSAudcufsg8hCMVxzgOmfRBwdEUVJgerU5GpbfV8tA3b8GSnw+EsCV68lG8s2D2hhpOgY7VvYd1sCbFuS9Y5Flv1TCDhvyuAqAu/PuCRvFP15ArCkXkK3WNexaz7XuNZMtZtCbPxYgphwK5HaFAUah1OH+qromysCA= 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=wU4usRMc; arc=none smtp.client-ip=209.85.214.202 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="wU4usRMc" Received: by mail-pl1-f202.google.com with SMTP id d9443c01a7336-297b35951b7so62948285ad.3 for ; Mon, 10 Nov 2025 20:04:47 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1762833887; x=1763438687; darn=vger.kernel.org; h=to:from:subject:message-id:references:mime-version:in-reply-to:date :from:to:cc:subject:date:message-id:reply-to; bh=ScLENsyYSFxSsSXeHlp0LDfphHjWjgXO0/fUWwSEsxM=; b=wU4usRMclFPGhyRk83oPZ20LdUbM0GM8bbgRkdodS1ww6DFB1mZGwDUNt5mDThKzAI f7NhS4bVhbnR0KjLM7mO1xX7CmQA45/ypUnF/diVHwRuO/R9jK4qxTupq1MFxM5timRN 7qMH3E3xlxHys4MCGB5vXllRtmYFk44GCgyUOliyk7yhYfM9A+s66tGOIf0rsxX27vBR BBbzMfBqxHxoV0ExdK50HIOiRcA+dOVscZeBCCxiM2NHOY1qXrir4PkKYp+u7ObuKh3B kywE9LhiKw3m8DgEykIRzpK3o2RyxGY4dvAGke+Y5IY/2gVM/1yFS0cIFb2ElJLEsOrB NPnA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1762833887; x=1763438687; h=to:from:subject:message-id:references:mime-version:in-reply-to:date :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=ScLENsyYSFxSsSXeHlp0LDfphHjWjgXO0/fUWwSEsxM=; b=fY7PFYGvSvL34JrBC9lrD3ZwOYETCPd4AWv3JPZjIp/WtqTaisl9gstYZSIfPan9UA ijp+U9jV+P764dI8Iz7S9OBoxAkiyjoAhKkX7zr1dtApSz7nSV2wlWjgELdG+9UdcjH4 kzki9utbX6i2yldptxw8P96cwp0nbkaPpcV6bmJ1Asx+uj+L8dCvuwjy32jVI6oLrmRo +P9C2LwY3NzXks+TFOq9bhhvt6rUKZwm07Qsa25BvE4jz+EQEjQJWk+oavbrBB2wLZlZ Vt8gKXrQMGbjFeJX6xX18xKb8t9jCeR1rHaymZ9sH9ysx3GSJ7euGaOWJfIVT57S6Qca +lhQ== X-Forwarded-Encrypted: i=1; AJvYcCWLBCMcXP5uYfoWwbhtRRM2bgPJ3dFkEZKW64jY2cLMR6YljqgbWyrV/exjqvJvWXirJR12Jaw4be6iq/E=@vger.kernel.org X-Gm-Message-State: AOJu0YwOCWoUZALVa4rwth7ox5DiogXJYyuDhY4M/eTwxAcy+V95fUJt 4Uvt4n8gIG4htFjtYmra0ya7M9NqhKIiOOkXob+mdQbLQ+JUsLFd4gHCrsQ8o36eA/MDNef7P95 AjhJkQo5QpA== X-Google-Smtp-Source: AGHT+IFq2mZoiOME+cl0yg/05U1bbEZzMY6aSFKy7XGWVIioRgl1uICIEeUsh828yUm7ED7z/hVL/j+JnmQJ X-Received: from dyu22.prod.google.com ([2002:a05:693c:8116:b0:2a4:7453:cd4e]) (user=irogers job=prod-delivery.src-stubby-dispatcher) by 2002:a17:903:40c9:b0:295:275d:21d8 with SMTP id d9443c01a7336-297e4bff754mr140263145ad.0.1762833886687; Mon, 10 Nov 2025 20:04:46 -0800 (PST) Date: Mon, 10 Nov 2025 20:04:10 -0800 In-Reply-To: <20251111040417.270945-1-irogers@google.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 References: <20251111040417.270945-1-irogers@google.com> X-Mailer: git-send-email 2.51.2.1041.gc1ab5b90ca-goog Message-ID: <20251111040417.270945-12-irogers@google.com> Subject: [PATCH v3 11/18] perf test stat+json: Improve metric-only testing From: Ian Rogers To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Namhyung Kim , Alexander Shishkin , Jiri Olsa , Ian Rogers , Adrian Hunter , James Clark , Xu Yang , Chun-Tse Shao , Thomas Richter , Sumanth Korikkar , Collin Funk , Thomas Falcon , Howard Chu , Dapeng Mi , Levi Yun , Yang Li , linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org, Andi Kleen , Weilin Wang Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" When testing metric-only, pass a metric to perf rather than expecting a hard coded metric value to be generated. Remove keys that were really metric-only units and instead don't expect metric only to have a matching json key as it encodes metrics as {"metric_name", "metric_value"}. Signed-off-by: Ian Rogers --- tools/perf/tests/shell/lib/perf_json_output_lint.py | 4 ++-- tools/perf/tests/shell/stat+json_output.sh | 2 +- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/tools/perf/tests/shell/lib/perf_json_output_lint.py b/tools/pe= rf/tests/shell/lib/perf_json_output_lint.py index c6750ef06c0f..1369baaa0361 100644 --- a/tools/perf/tests/shell/lib/perf_json_output_lint.py +++ b/tools/perf/tests/shell/lib/perf_json_output_lint.py @@ -65,8 +65,6 @@ def check_json_output(expected_items): 'socket': lambda x: True, 'thread': lambda x: True, 'unit': lambda x: True, - 'insn per cycle': lambda x: isfloat(x), - 'GHz': lambda x: True, # FIXME: it seems unintended for --metric-on= ly } input =3D '[\n' + ','.join(Lines) + '\n]' for item in json.loads(input): @@ -88,6 +86,8 @@ def check_json_output(expected_items): f' in \'{item}\'') for key, value in item.items(): if key not in checks: + if args.metric_only: + continue raise RuntimeError(f'Unexpected key: key=3D{key} value=3D{value}') if not checks[key](value): raise RuntimeError(f'Check failed for: key=3D{key} value=3D{value}= ') diff --git a/tools/perf/tests/shell/stat+json_output.sh b/tools/perf/tests/= shell/stat+json_output.sh index 98fb65274ac4..85d1ad7186c6 100755 --- a/tools/perf/tests/shell/stat+json_output.sh +++ b/tools/perf/tests/shell/stat+json_output.sh @@ -181,7 +181,7 @@ check_metric_only() echo "[Skip] CPU-measurement counter facility not installed" return fi - perf stat -j --metric-only -e instructions,cycles -o "${stat_output}" true + perf stat -j --metric-only -M page_faults_per_second -o "${stat_output}" = true $PYTHON $pythonchecker --metric-only --file "${stat_output}" echo "[Success]" } --=20 2.51.2.1041.gc1ab5b90ca-goog From nobody Tue Nov 11 11:28:57 2025 Received: from mail-pl1-f202.google.com (mail-pl1-f202.google.com [209.85.214.202]) (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 3DBE234CFCC for ; Tue, 11 Nov 2025 04:04:48 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.214.202 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1762833890; cv=none; b=ry9oZyROp3PUZtNMutCZg2tHvutdqXo6KEhXNhJ5kDs0Io3uuS8FK4fOpo1tAsgmmuTq5IUKwaWH2kZa+flpNA4kLOm/RP7dUmXi/3Nc3w0+igAEXW9GxdMg1/N9RboE1aEL5pxSp0MPJCFB/eg+Jj7e3wkVu8dZ6jpp/IzLWec= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1762833890; c=relaxed/simple; bh=QjO8gSjusfZ0mPendJr0+QHEFToVhbHiU4PqR4DkANY=; h=Date:In-Reply-To:Mime-Version:References:Message-ID:Subject:From: To:Content-Type; b=jWPDvYYqsY21+PjfO6jTmKMOZhNfTliTNCeoG0k5xcjcpwPVwheWkbm2XkRmAJ0aATmGGwvyshf6Ei4H0hcpejnw1iFsFlgp18jQ7HoRCn7fyPlTUN81+cgSIg++u6DX+be6FXmjwk9l77JLoPT8Vp1UmRd9cXZmkQGz6a4qOgk= 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=pMcx3RBm; arc=none smtp.client-ip=209.85.214.202 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="pMcx3RBm" Received: by mail-pl1-f202.google.com with SMTP id d9443c01a7336-297f3710070so70937725ad.2 for ; Mon, 10 Nov 2025 20:04:48 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1762833888; x=1763438688; darn=vger.kernel.org; h=to:from:subject:message-id:references:mime-version:in-reply-to:date :from:to:cc:subject:date:message-id:reply-to; bh=DlhPdv0WtdedRwVAzlOAlu+LBijAGUItUQ61/Soo80E=; b=pMcx3RBmxGyfJ/3RiYYYLiJT8SzYYWyA7VRce7MTus/vfpdebkBR8w5vTOWeFdkEEL XeZsfEZpTauXhVBuCxNr7pwMo//fDgjHZ7zAJJfmx6gGrkYDb7BeAH3Il8YxwRLwSild 3F4odrOywosmQlNH3IENXyx3wYen/ir8DfnNZE1/bNXAmL6+aCkulBRBRylwyaw57YEB Zc+SJdezoNsDwtEa6s/rvbMhiViDntVLg1arY57jhJkLgNArvImSnhmIMpJf13VhMqlt CHzg9sbx5ggk/s22/XVXMQyjkXL9pgoEeClh8D9hAIzMPWuZCu3T1JyAlo4U2D3eCv7N t3qA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1762833888; x=1763438688; h=to:from:subject:message-id:references:mime-version:in-reply-to:date :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=DlhPdv0WtdedRwVAzlOAlu+LBijAGUItUQ61/Soo80E=; b=B4QFFG23xlLMyMkY3qNqjLMKtKzVPqfTr9TbEHGdXrx5y2et3tW6mhHJxo/sO5/Vas 1aHphBkTI0g3cqe4Py81NTtZT5jzp+4SK+51dH5GtEAhD8W5TmGSrN6yeCnEFUJ1TmsU ad2tbLZ785c3IqNvwV/LgvAfb37LK6z70FMcbjZRTvdCMWKFJfi8tnJsIajA40S+T5aM f2aKOyXLwnYU+EzijIJbmj/N/WxTR7FraPA41CI2cE0WDT+DS9JPvj+u3Tbh9Zl9kHX2 bh52n4VsNOXL4qOTUe6vB6e4zyivhBjP97oajwJ/atYyoxJp51Vz7X0yU1z2RBMOJMZZ Obcg== X-Forwarded-Encrypted: i=1; AJvYcCWaY6PRYMjfo+cGrFdVo9n3aUrcarcHe7WAYytr3GSEq4GefZ69d3ez9T6hSmFoLrNWGoeD5+e2eIxF/Mo=@vger.kernel.org X-Gm-Message-State: AOJu0YzWU94VHHhaMVFAjBVnc+wmmRLZdK1Pg8CoPQMF0VIyD2wYEDhp f+9hwXPgNOiWy3fw2W2ApO1VgIpoHoSAVNjB7zisvDvgpSJuTUNoHYaRvNTySzVnSmAuRqPuTs6 py5aJZBv8Gw== X-Google-Smtp-Source: AGHT+IHnQHwE7ZVnLqbv4sMxN6fuUu9/VHBggmkrVxMIdeFVE3Yq6GszmnnFYMuvX5r+GV9wJ+cLh3U9USGv X-Received: from dycoh8.prod.google.com ([2002:a05:7301:da08:b0:2a4:6ae0:3502]) (user=irogers job=prod-delivery.src-stubby-dispatcher) by 2002:a17:902:e947:b0:295:543a:f7e3 with SMTP id d9443c01a7336-297e564cc94mr135808275ad.27.1762833888290; Mon, 10 Nov 2025 20:04:48 -0800 (PST) Date: Mon, 10 Nov 2025 20:04:11 -0800 In-Reply-To: <20251111040417.270945-1-irogers@google.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 References: <20251111040417.270945-1-irogers@google.com> X-Mailer: git-send-email 2.51.2.1041.gc1ab5b90ca-goog Message-ID: <20251111040417.270945-13-irogers@google.com> Subject: [PATCH v3 12/18] perf test stat: Ignore failures in Default[234] metricgroups From: Ian Rogers To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Namhyung Kim , Alexander Shishkin , Jiri Olsa , Ian Rogers , Adrian Hunter , James Clark , Xu Yang , Chun-Tse Shao , Thomas Richter , Sumanth Korikkar , Collin Funk , Thomas Falcon , Howard Chu , Dapeng Mi , Levi Yun , Yang Li , linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org, Andi Kleen , Weilin Wang Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" The Default[234] metric groups may contain unsupported legacy events. Allow those metric groups to fail. Signed-off-by: Ian Rogers --- tools/perf/tests/shell/stat_all_metricgroups.sh | 3 +++ 1 file changed, 3 insertions(+) diff --git a/tools/perf/tests/shell/stat_all_metricgroups.sh b/tools/perf/t= ests/shell/stat_all_metricgroups.sh index c6d61a4ac3e7..1400880ec01f 100755 --- a/tools/perf/tests/shell/stat_all_metricgroups.sh +++ b/tools/perf/tests/shell/stat_all_metricgroups.sh @@ -37,6 +37,9 @@ do then err=3D2 # Skip fi + elif [[ "$m" =3D=3D @(Default2|Default3|Default4) ]] + then + echo "Ignoring failures in $m that may contain unsupported legacy ev= ents" else echo "Metric group $m failed" echo $result --=20 2.51.2.1041.gc1ab5b90ca-goog From nobody Tue Nov 11 11:28:57 2025 Received: from mail-pl1-f202.google.com (mail-pl1-f202.google.com [209.85.214.202]) (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 ACFDE34D4E2 for ; Tue, 11 Nov 2025 04:04:50 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.214.202 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1762833892; cv=none; b=FfXReJr4vjTgTEKdyIL5EQUlqJrxXpy/mrFBIevYTrrR1XRBayI5ApwLZ6H6/uATpoc3PMCgEfXQDH0qmMjptQWCeRQqRu/Zxnm115r+iqdkhjWd8svqiM9fdg99TyOcsjP8Y+cR+NGeb5RFk5p6auwjPAOOkcwUpZXMgH/j5Sw= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1762833892; c=relaxed/simple; bh=7NP/VCWcDtaQ9ddY1vDzUSspDiDop3c4KIH6lorm7/g=; h=Date:In-Reply-To:Mime-Version:References:Message-ID:Subject:From: To:Content-Type; b=hTKqtoy4l6jBceVfaFc5UEaiqovv/7vB/pQo4M5wDJHgfF0K6XVmv5AuQnRFSPVK6y9UE7MNHU+CPHpwV8LYxa0y7V1CBh6lDdrcm2I1z4kSlWIsMh9Y995Vdx+ALu2PQvpVVuOFM6pBkb3GT7hozoErMpJRF34OIxl3+311XUg= 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=kt9tyjrP; arc=none smtp.client-ip=209.85.214.202 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="kt9tyjrP" Received: by mail-pl1-f202.google.com with SMTP id d9443c01a7336-2980ef53fc5so55436355ad.1 for ; Mon, 10 Nov 2025 20:04:50 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1762833890; x=1763438690; darn=vger.kernel.org; h=to:from:subject:message-id:references:mime-version:in-reply-to:date :from:to:cc:subject:date:message-id:reply-to; bh=bf2tG0VSbGuTREXl21iJqSVPQmFIFuxPe8RAulQFiSU=; b=kt9tyjrP4hZkDeoWC7TBY4nQAe1Pt2atieP+gx/2kh12misHkMrsy7YOm+Iygmlb12 B8t29ODOvvhV9CadZcrQr3SRw67fowUXo2B3sr6WJO7eamy2zhYEwd9BmMCTxXhjG8H8 O2jWyWQ1FZiH5ZqrCPUnKNzGT3MHNDJrYvKesrxdqZ3LrQojyPeB5zYKwCOEbocmeZSS M2FkChLNhnG8J/ewKI7s5cI09unqX+KLIH/TilFTGXAKJR5SPvUohiQpPQJPiEiUEYoZ XFv1WIS4S+FQIwwdp0nXWIY1NZhR1KlrNamGzw59QiMVPfsCInjQgYczT4RC0COOzb2i fT0g== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1762833890; x=1763438690; h=to:from:subject:message-id:references:mime-version:in-reply-to:date :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=bf2tG0VSbGuTREXl21iJqSVPQmFIFuxPe8RAulQFiSU=; b=Sys/9GqhIIoo+6jFGhl1kFk5q59Jtu3Us6lzFrxspjZTJdXl68VmfvzHrhw+y0hxUX BXXXtfbTm65E+EYMF/Q9PsGFzFMhOTKEuTwfc2qBhU0XE5tkZL1MAetmrqHUj0kEYPZd nHN3mRhzRoRodfQuuURxFEMD8XiflBJlmWVMU1vb1tlexYUM+NvDw3Ruo5SKaiObQJW/ OlEkuacFrjympqiprG8MIwpmbhxAmw98dDc5pCpJZBkFCQdPJ02TireugYsIgSi2oDQv 0v54dfNRTQQ9dPIb0yT+50Z77ZwQmJB0EZFMURsXh9hQEFCxyToPe4SZgxrm6cjGoN16 1lBA== X-Forwarded-Encrypted: i=1; AJvYcCWro98SqfDy2HatHve5JO5/Nt4SwaOAa64ab42ETwqsCbhK5Sj0HfBobtxof69sOW+WeCKdBSea9eim5YM=@vger.kernel.org X-Gm-Message-State: AOJu0YzDnczgGKX4NAY6dk8w9QsEC/eaQw1s2/4Z9yLGfHAnWxSEpndF 7WUbz/2i6Q7BjSm67eBpawT/UnXNuBPVhIpo7ZCfZlrEL3rMpr3PdjoUWKdnTyw6vfZ8dAwaztp xamzm1NroZg== X-Google-Smtp-Source: AGHT+IF0XIujBIQ4BHEF6rVrrs4Ei1rf7m9r0QAR9t7GsRl8tXu89zXHQL+Hk/jzWA/cU46Hzzz7xVolj7+P X-Received: from dlbrf23.prod.google.com ([2002:a05:7022:f217:b0:119:49ca:6b87]) (user=irogers job=prod-delivery.src-stubby-dispatcher) by 2002:a17:903:19c8:b0:295:fdc2:1f52 with SMTP id d9443c01a7336-297e5625cf9mr142122905ad.14.1762833890175; Mon, 10 Nov 2025 20:04:50 -0800 (PST) Date: Mon, 10 Nov 2025 20:04:12 -0800 In-Reply-To: <20251111040417.270945-1-irogers@google.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 References: <20251111040417.270945-1-irogers@google.com> X-Mailer: git-send-email 2.51.2.1041.gc1ab5b90ca-goog Message-ID: <20251111040417.270945-14-irogers@google.com> Subject: [PATCH v3 13/18] perf test stat: Update std_output testing metric expectations From: Ian Rogers To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Namhyung Kim , Alexander Shishkin , Jiri Olsa , Ian Rogers , Adrian Hunter , James Clark , Xu Yang , Chun-Tse Shao , Thomas Richter , Sumanth Korikkar , Collin Funk , Thomas Falcon , Howard Chu , Dapeng Mi , Levi Yun , Yang Li , linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org, Andi Kleen , Weilin Wang Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" Make the expectations match json metrics rather than the previous hard coded ones. Signed-off-by: Ian Rogers --- tools/perf/tests/shell/stat+std_output.sh | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tools/perf/tests/shell/stat+std_output.sh b/tools/perf/tests/s= hell/stat+std_output.sh index ec41f24299d9..9c4b92ecf448 100755 --- a/tools/perf/tests/shell/stat+std_output.sh +++ b/tools/perf/tests/shell/stat+std_output.sh @@ -12,8 +12,8 @@ set -e stat_output=3D$(mktemp /tmp/__perf_test.stat_output.std.XXXXX) =20 event_name=3D(cpu-clock task-clock context-switches cpu-migrations page-fa= ults stalled-cycles-frontend stalled-cycles-backend cycles instructions bra= nches branch-misses) -event_metric=3D("CPUs utilized" "CPUs utilized" "/sec" "/sec" "/sec" "fron= tend cycles idle" "backend cycles idle" "GHz" "insn per cycle" "/sec" "of a= ll branches") -skip_metric=3D("stalled cycles per insn" "tma_" "retiring" "frontend_bound= " "bad_speculation" "backend_bound" "TopdownL1" "percent of slots") +event_metric=3D("CPUs_utilized" "CPUs_utilized" "cs/sec" "migrations/sec" = "faults/sec" "frontend_cycles_idle" "backend_cycles_idle" "GHz" "insn_per_c= ycle" "/sec" "branch_miss_rate") +skip_metric=3D("tma_" "TopdownL1") =20 cleanup() { rm -f "${stat_output}" --=20 2.51.2.1041.gc1ab5b90ca-goog From nobody Tue Nov 11 11:28:57 2025 Received: from mail-pl1-f202.google.com (mail-pl1-f202.google.com [209.85.214.202]) (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 9FC7734C83A for ; Tue, 11 Nov 2025 04:04:52 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.214.202 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1762833894; cv=none; b=SVkwDW/RZ0aMLjxEFRl7JzXwbcfoIuXbjQW7js1vWbf6qeo9oPnETwDiJTP6jjwezCGkk18lAFcd8AbgdILUFLbYY2bTq/kiI2Cq5pSLls2CQg37fwED6c6TE7yyq+qkAfPne+zOPg0Juy6LV41lL7yBcJLnzjHQDMDFdJfuWyQ= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1762833894; c=relaxed/simple; bh=bWlWZmoSbjMlQJvhT1Kc+8bbCD51npTRBtHHE+YWJoI=; h=Date:In-Reply-To:Mime-Version:References:Message-ID:Subject:From: To:Content-Type; b=qIMCvRpjuxbYy7z4V/bD8yzYRAJ8TKhBR0BL4OVk5/DRGBcwf3U/5e3IWRvEriJTY/Bfe44KuyLuRrNzEuquEJrm5EuhmcYdaWw+MPH5xHbvf3VcLdQQ88hd3Y2egRZjTlLxWFBhuikx2DvsWO5T2HYSX53GJVVGUhAM1rpuXfM= 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=aelZsAfp; arc=none smtp.client-ip=209.85.214.202 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="aelZsAfp" Received: by mail-pl1-f202.google.com with SMTP id d9443c01a7336-2958c80fcabso102676075ad.0 for ; Mon, 10 Nov 2025 20:04:52 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1762833892; x=1763438692; darn=vger.kernel.org; h=to:from:subject:message-id:references:mime-version:in-reply-to:date :from:to:cc:subject:date:message-id:reply-to; bh=ja604hKaX8qDrZr8dkd1ADzyH8Os+oHlUQ1fcptNWsg=; b=aelZsAfpOM3JxfqCQ/zPPvgLAjX2plnqRxNhPoYwHz2KsxCR3s5f2uu2afV3X7n7yi ee5vaZwb4zBKcLTeaBDHqjU1zTBxa3WFdGISgPuY/VsHyV5TzcXCdENKdpzM/njbtbqF RD5GvkdvF8otfYAB+Dhv+iI0YhnkJoUwkd/BiE/4z0LrUYr1I0pcwArcnALBnOm7Tv78 6G2ezJpd6xo3wYaKSgJun16QASLJna6MV/wcW7NBra0Iw6LEPnRzde0kguiiS2dj0FyF 2suU7KwwvnMG+SDyqBX81dJTMn1/FZzn2fBVoY1FnlsCiazqGXj1yBh7eEn83AqRq/xx OK3A== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1762833892; x=1763438692; h=to:from:subject:message-id:references:mime-version:in-reply-to:date :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=ja604hKaX8qDrZr8dkd1ADzyH8Os+oHlUQ1fcptNWsg=; b=N8Bx6mPMskeoD56XwxThI/1P+mjJ9J4jWXJuJqdBv1RyQPkG2KmxhrsmXFZlz249D0 IZA9qWawWCLmch7XvEZydWFMHwQBBSUmvp0N7mp9GUuZ+gl+uCVV7OkjA0BB565Jo4Bf lLFFmgISollJqNQz7KJq6qvpPUty7IBAMrzhBn6ZO2A2/3X9uW+K4ijH6PcSVvgpUeLQ y2nXOml4iVY0Z6ML78WkjTgp41P5wHGxjSYmM/uwYtBgDw17JDqtBtyytE8MTadT9hcZ 4zWPz5zUreTJqze3ZEVYIpZZ8ReX5zNMlztXO6YJ0okXqs6b40Vlt0y//hqQ9bpYg/j9 +ZjA== X-Forwarded-Encrypted: i=1; AJvYcCUrHLsONYXU6VyodG6KTEXnc5TRYRLnKfsMBDq2yfENaGrdxPCNamHTX59ruHpfIlzpLViAIWIVGVB4LZ8=@vger.kernel.org X-Gm-Message-State: AOJu0YyM1Hy5Pn5rRYRRDti52pY/N6CTItD1ceXMExgKoC8ZCx21MP4i yWQ6EB8gEVOfa6yxqczHStgFEKXPFdwd7gr8rUxY1cvRysWLKvGtRHdYcPBtXYJ1NlnJFnbfZ9R DEtoZbtylgA== X-Google-Smtp-Source: AGHT+IErWkHeqed2GrqiOy88wuRY0tqtNPYJOqc0LJ1kt+lmnAJczl2sta01fh79caMHsqjCpCPCmGcxFpwB X-Received: from dybpf10.prod.google.com ([2002:a05:7301:410a:b0:2a4:6c32:42a2]) (user=irogers job=prod-delivery.src-stubby-dispatcher) by 2002:a17:902:d485:b0:294:f310:5218 with SMTP id d9443c01a7336-297e4be8af6mr152083305ad.0.1762833891972; Mon, 10 Nov 2025 20:04:51 -0800 (PST) Date: Mon, 10 Nov 2025 20:04:13 -0800 In-Reply-To: <20251111040417.270945-1-irogers@google.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 References: <20251111040417.270945-1-irogers@google.com> X-Mailer: git-send-email 2.51.2.1041.gc1ab5b90ca-goog Message-ID: <20251111040417.270945-15-irogers@google.com> Subject: [PATCH v3 14/18] perf test metrics: Update all metrics for possibly failing default metrics From: Ian Rogers To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Namhyung Kim , Alexander Shishkin , Jiri Olsa , Ian Rogers , Adrian Hunter , James Clark , Xu Yang , Chun-Tse Shao , Thomas Richter , Sumanth Korikkar , Collin Funk , Thomas Falcon , Howard Chu , Dapeng Mi , Levi Yun , Yang Li , linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org, Andi Kleen , Weilin Wang Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" Default metrics may use unsupported events and be ignored. These metrics shouldn't cause metric testing to fail. Signed-off-by: Ian Rogers --- tools/perf/tests/shell/stat_all_metrics.sh | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/tools/perf/tests/shell/stat_all_metrics.sh b/tools/perf/tests/= shell/stat_all_metrics.sh index 6fa585a1e34c..a7edf01b3943 100755 --- a/tools/perf/tests/shell/stat_all_metrics.sh +++ b/tools/perf/tests/shell/stat_all_metrics.sh @@ -25,8 +25,13 @@ for m in $(perf list --raw-dump metrics); do # No error result and metric shown. continue fi - if [[ "$result" =3D~ "Cannot resolve IDs for" ]] + if [[ "$result" =3D~ "Cannot resolve IDs for" || "$result" =3D~ "No supp= orted events found" ]] then + if [[ "$m" =3D=3D @(l1_prefetch_miss_rate|stalled_cycles_per_instructi= on) ]] + then + # Default metrics that may use unsupported events. + continue + fi echo "Metric contains missing events" echo $result err=3D1 # Fail --=20 2.51.2.1041.gc1ab5b90ca-goog From nobody Tue Nov 11 11:28:57 2025 Received: from mail-pl1-f202.google.com (mail-pl1-f202.google.com [209.85.214.202]) (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 E1D1034DB61 for ; Tue, 11 Nov 2025 04:04:54 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.214.202 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1762833896; cv=none; b=hG2C+crLFkGBiNZ7FPn2uogQKW6FyN1Rbq0JxGGL9UA700mbnSj685AKyZlNpwriMDX59/2EcAH51hph0E5a/u9X/lPJHpZU1DWpgnqyqLcTv5D1YZfVkTuTZ2Mo5L+Dt68VmLwmPacyP13gxQi2n1gvsDss5u6JoZOXdrKULPU= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1762833896; c=relaxed/simple; bh=NQZLwrJVuAq7AnAGYojD2+FNeBDfMGlmTWdY5sb/Yxo=; h=Date:In-Reply-To:Mime-Version:References:Message-ID:Subject:From: To:Content-Type; b=lw8KNctladV+b3He8SogjsaUpVdT39Hd78PvZVRf2rdNjOkfl5pQPlehZPsQ4/pwH68I03dIvpcxWWv1Oq9tFnyCGbbxA9Ft4k1z0Zcw5wFc21KREXOF9ziTq/SB7sgKD8VDY0Jw27EghN8syyBaJVdIKEl5H09crx6hB1Vv8bk= 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=P63sEqHM; arc=none smtp.client-ip=209.85.214.202 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="P63sEqHM" Received: by mail-pl1-f202.google.com with SMTP id d9443c01a7336-297fbfb4e53so37555405ad.1 for ; Mon, 10 Nov 2025 20:04:54 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1762833894; x=1763438694; darn=vger.kernel.org; h=to:from:subject:message-id:references:mime-version:in-reply-to:date :from:to:cc:subject:date:message-id:reply-to; bh=Snl9m+Gsuclyili9Jia8BWYLwnDqexlNgmsV+/z8jHs=; b=P63sEqHMw9XlacbDKf6mDPgjU7FsXKY9lblmrMyPGFAvEmNTRtXHFPO5SD/7sk0BkU 0cX957J2IXAnu570ernOIDDo9OABskxWXiIe79UJ2rCYvmVIw55eqGONTlLVsLrn407b 1nV2F3cGo6umTrY9QgLdXbqOXf6FOlcNRMezs9SC24bZ8k6TujI3SKpUbxnBT9UrsTKz w7Mqi/zdoUz0LBwUhYFlHgB0WspkzOnxhxSrPK+T5OBPxE+h+Fc0zokYEQ8EKsCagGcz mq7G4nKVTpsiF60csRGLRSXCtochhbgArtEmNBsLflj6cjSzNPex0eZbIFEIImAQof9F N7NA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1762833894; x=1763438694; h=to:from:subject:message-id:references:mime-version:in-reply-to:date :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=Snl9m+Gsuclyili9Jia8BWYLwnDqexlNgmsV+/z8jHs=; b=M+6dg/EAo9OYU7jZcIlqCuuMOBs3uUwjWqpLDs0KKk0kdZYyo8vFoJ+nUXYJkJ+B6j IJkoYELzlwgXeYM4rIBImwu9AkcD3o0SRRFCylnADMH/QSg0325SvNv9D0ykGDMm4Xhz KGx5h7u+lblroAlbiHb5NT7HrB/FuplzQ86m4LUkmtjMA71OndUy56mJ2EPzSCgZkDWt r5IbwNpLXX259j+uPv/omZSgyUh7/KzZn+1Qf5dR4Ng1nj+zi1KiFlHiYmgQtix/D3u+ VkN99z9Qz60yWWSxz36/dmHOiPdPJB+t0ccHfFqlXDaABy4y8DI9sT0iDzp/9MNHcEiY nYdQ== X-Forwarded-Encrypted: i=1; AJvYcCWEg2LEFtJ7AKQmyzRp5JuzD9bQUQmGLKoOVAi4Mhx/R6lH5Hga7DFKmJY7EqHsT8VMXFNg8Xj7nxy4xQo=@vger.kernel.org X-Gm-Message-State: AOJu0Yz48ICi1qEqFV138lJJiPyhDP1bXJWgKKxA2+dudWQXKsb2BQjH ZZ3+8wQA1h/poYw1FmPHyhAtmGGh0gVsl0TZ2N2y+jCAGGh1cm8S8t1r7bPgH52xvJVPiUM8lKJ yqJMgGsu71Q== X-Google-Smtp-Source: AGHT+IEFASiKWqzsDwLv3yb92ygqFV4xil2iBgRJzUJt4a1IPtI16V7BlKTsos1+lpG3Vc0sJ+Qo8umUeilC X-Received: from dybmh40.prod.google.com ([2002:a05:7301:f28:b0:2a4:6c58:332a]) (user=irogers job=prod-delivery.src-stubby-dispatcher) by 2002:a17:903:2c0c:b0:298:42ba:c422 with SMTP id d9443c01a7336-29842bac56cmr12840675ad.31.1762833893842; Mon, 10 Nov 2025 20:04:53 -0800 (PST) Date: Mon, 10 Nov 2025 20:04:14 -0800 In-Reply-To: <20251111040417.270945-1-irogers@google.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 References: <20251111040417.270945-1-irogers@google.com> X-Mailer: git-send-email 2.51.2.1041.gc1ab5b90ca-goog Message-ID: <20251111040417.270945-16-irogers@google.com> Subject: [PATCH v3 15/18] perf test stat: Update shadow test to use metrics From: Ian Rogers To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Namhyung Kim , Alexander Shishkin , Jiri Olsa , Ian Rogers , Adrian Hunter , James Clark , Xu Yang , Chun-Tse Shao , Thomas Richter , Sumanth Korikkar , Collin Funk , Thomas Falcon , Howard Chu , Dapeng Mi , Levi Yun , Yang Li , linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org, Andi Kleen , Weilin Wang Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" Previously '-e cycles,instructions' would implicitly create an IPC metric. This now has to be explicit with '-M insn_per_cycle'. Signed-off-by: Ian Rogers --- tools/perf/tests/shell/stat+shadow_stat.sh | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tools/perf/tests/shell/stat+shadow_stat.sh b/tools/perf/tests/= shell/stat+shadow_stat.sh index 8824f445d343..cabbbf17c662 100755 --- a/tools/perf/tests/shell/stat+shadow_stat.sh +++ b/tools/perf/tests/shell/stat+shadow_stat.sh @@ -14,7 +14,7 @@ perf stat -a -e cycles sleep 1 2>&1 | grep -e cpu_core &&= exit 2 =20 test_global_aggr() { - perf stat -a --no-big-num -e cycles,instructions sleep 1 2>&1 | \ + perf stat -a --no-big-num -M insn_per_cycle sleep 1 2>&1 | \ grep -e cycles -e instructions | \ while read num evt _ ipc rest do @@ -53,7 +53,7 @@ test_global_aggr() =20 test_no_aggr() { - perf stat -a -A --no-big-num -e cycles,instructions sleep 1 2>&1 | \ + perf stat -a -A --no-big-num -M insn_per_cycle sleep 1 2>&1 | \ grep ^CPU | \ while read cpu num evt _ ipc rest do --=20 2.51.2.1041.gc1ab5b90ca-goog From nobody Tue Nov 11 11:28:57 2025 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 AC23434DB72 for ; Tue, 11 Nov 2025 04:04:56 +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=1762833898; cv=none; b=uEMY1jzuUWxZXoxuCz2xvJ/xCBBEtBUKZ76NuZMDXdd9cP+brS01PSjfvQ1GJQa5wGA4SuNEqQLVLGTG9fPolRvajDb3ufzuRcNgf17cq8ySHwhvkJAtcCk/RnWWE2apKZ1LPm9ROsoNwZ2U3dESeZzkgtirrdpghf3w/u/3m1Y= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1762833898; c=relaxed/simple; bh=LLn9FDAlPcwSWZjUFvp9EB4XY1jJAmyC35CaM/eue2A=; h=Date:In-Reply-To:Mime-Version:References:Message-ID:Subject:From: To:Content-Type; b=a7WfztanP8d8naEce28vEoaGsJ8v4HueFyaKa3O5orQcis3C3UVuvYeeJtYuQYM/WEQ/qxsv/fwJPOvPZzvkEnacvT8upOZLqpqGeB/CbRViSgIx4wmKX7sjMshADblJe+IQHIhDvbogxzbjvoHdBzPnp6pZ8sfabs/rqeZlRH0= 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=Xbx1Ui41; 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="Xbx1Ui41" Received: by mail-pg1-f202.google.com with SMTP id 41be03b00d2f7-b55283ff3fcso2820321a12.3 for ; Mon, 10 Nov 2025 20:04:56 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1762833896; x=1763438696; darn=vger.kernel.org; h=to:from:subject:message-id:references:mime-version:in-reply-to:date :from:to:cc:subject:date:message-id:reply-to; bh=5Et4hjqXaxJjcCbzTA1Cjej6Oy43GlvxgQk6XgvQSd0=; b=Xbx1Ui41dRNN7nunAER/3qwKYd9Z8U9S5EE6pnP2LTyMEkYcn4Wx7uOZsHWu3J7Suh GYr/OcWmoId7WFFP9qbbDMT/hAZHZ1IGf5D77dOkzasaLTXRPgb2BCUWHn98xomb3yY5 UI02dJBU6I8ogRxKdosCaZ/D5P5dhNR5og8IwnjJG6fR96KEzPWVECq1C39ifrgVzfOr SX6DcozH1YLuTNnly4wX4FpnDUZpoh6Byqh64jQUllmUpkFdf0OcYkomCUvp8A0ratIR g4EzObhm0BcHI74DDixZ4uZbtanutLch8r3zx848u/SwtR1jcZN9htCoYPUhQ6ajDXDQ 07mQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1762833896; x=1763438696; h=to:from:subject:message-id:references:mime-version:in-reply-to:date :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=5Et4hjqXaxJjcCbzTA1Cjej6Oy43GlvxgQk6XgvQSd0=; b=VA3mogfRUNXO0SOUMpEVRmZizU8PLDaTght/jE3dKZrJaL7s25Hgy+4dJvPJXNi6aN f8jk//v1OjWRWVY1K5tLVNlpjCA8waaPsq57Lgly8IsT4KTqutPbCgeTCcIeBFT6W7xU xYhsF1t52gahQ75+vgQysdkJBlrNe1dLNbB/QMqnNuv2fKwBIEKLXLZyX4d6hr/QygzG w6s7aKrI8CsROKAW6BJbt6TTQv2lxFp3H0fwT3Zcfgb6+8AvZZOz0xZUfvzAboDsz898 jCK8ZqcKYB7HbBi4l/UoyCObcdSe9QYEcr5saLsoeSmI5FQYtv/NFtSr+ZxlyfxBSIE7 LzXw== X-Forwarded-Encrypted: i=1; AJvYcCXBJrFKNkCt1/Cu2a85wHyhpL5oCf0L/9uvjZM1NCnof6JEFSoJ0m6EYU6CVxS3SaboEfmP5VdeDlhGsU4=@vger.kernel.org X-Gm-Message-State: AOJu0YxKUM6NNTFBLxCEatFmSYBLdPPpXPNtbV6M+VKSlXqnK8ZDbDSe F73oVfYQ1u5V/jGAuCs3zXYIdYZIE6divv5LkE+w5+qGlTxzVUYD/bRRw8SodBTUaAuB/e7My+l m98UAgCN+Pw== X-Google-Smtp-Source: AGHT+IGn0tBeyvmCXlP0g5sTiUGrIOYaiPwsUS9X8uiEM3cJe2lhyUnPrjTMn+mtiKy1m8WpFhJu4/X5j5P7 X-Received: from dlbbo21.prod.google.com ([2002:a05:7022:4395:b0:119:78ff:fe16]) (user=irogers job=prod-delivery.src-stubby-dispatcher) by 2002:a17:903:1aad:b0:295:615d:f1d2 with SMTP id d9443c01a7336-297e56ca36emr130639925ad.48.1762833895913; Mon, 10 Nov 2025 20:04:55 -0800 (PST) Date: Mon, 10 Nov 2025 20:04:15 -0800 In-Reply-To: <20251111040417.270945-1-irogers@google.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 References: <20251111040417.270945-1-irogers@google.com> X-Mailer: git-send-email 2.51.2.1041.gc1ab5b90ca-goog Message-ID: <20251111040417.270945-17-irogers@google.com> Subject: [PATCH v3 16/18] perf test stat: Update test expectations and events From: Ian Rogers To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Namhyung Kim , Alexander Shishkin , Jiri Olsa , Ian Rogers , Adrian Hunter , James Clark , Xu Yang , Chun-Tse Shao , Thomas Richter , Sumanth Korikkar , Collin Funk , Thomas Falcon , Howard Chu , Dapeng Mi , Levi Yun , Yang Li , linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org, Andi Kleen , Weilin Wang Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" test_stat_record_report and test_stat_record_script used default output which triggers a bug when sending metrics. As this isn't relevant to the test switch to using named software events. Update the match in test_hybrid as the cycles event is now cpu-cycles to workaround potential ARM issues. Signed-off-by: Ian Rogers --- tools/perf/tests/shell/stat.sh | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/tools/perf/tests/shell/stat.sh b/tools/perf/tests/shell/stat.sh index 8a100a7f2dc1..985adc02749e 100755 --- a/tools/perf/tests/shell/stat.sh +++ b/tools/perf/tests/shell/stat.sh @@ -18,7 +18,7 @@ test_default_stat() { =20 test_stat_record_report() { echo "stat record and report test" - if ! perf stat record -o - true | perf stat report -i - 2>&1 | \ + if ! perf stat record -e task-clock -o - true | perf stat report -i - 2>= &1 | \ grep -E -q "Performance counter stats for 'pipe':" then echo "stat record and report test [Failed]" @@ -30,7 +30,7 @@ test_stat_record_report() { =20 test_stat_record_script() { echo "stat record and script test" - if ! perf stat record -o - true | perf script -i - 2>&1 | \ + if ! perf stat record -e task-clock -o - true | perf script -i - 2>&1 | \ grep -E -q "CPU[[:space:]]+THREAD[[:space:]]+VAL[[:space:]]+ENA[[:spac= e:]]+RUN[[:space:]]+TIME[[:space:]]+EVENT" then echo "stat record and script test [Failed]" @@ -196,7 +196,7 @@ test_hybrid() { fi =20 # Run default Perf stat - cycles_events=3D$(perf stat -- true 2>&1 | grep -E "/cycles/[uH]*| cycl= es[:uH]* " -c) + cycles_events=3D$(perf stat -a -- sleep 0.1 2>&1 | grep -E "/cpu-cycles/= [uH]*| cpu-cycles[:uH]* " -c) =20 # The expectation is that default output will have a cycles events on ea= ch # hybrid PMU. In situations with no cycles PMU events, like virtualized,= this --=20 2.51.2.1041.gc1ab5b90ca-goog From nobody Tue Nov 11 11:28:57 2025 Received: from mail-pl1-f201.google.com (mail-pl1-f201.google.com [209.85.214.201]) (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 C9DD234DCF3 for ; Tue, 11 Nov 2025 04:04:58 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.214.201 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1762833900; cv=none; b=kjW+Ti5GV6mrzEBV3GJiR5rRMv+iC+tq4Crlj8Zp8hz/ydr3JswRWltinz5P7BnV7iiuFAtbbcy8Uhrqe5Ab2rHxvwW3U4Lvx4Lost9BO0pYsc0TV6jqAKru/TSGRXOvndR119eGGFZOELp9zSRGJjqPOxbBr4Z5UpNCxpoNJls= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1762833900; c=relaxed/simple; bh=sWmDIev9bpAFqHQEit38MtJHcI5r4PnZTO5Rq4PESvU=; h=Date:In-Reply-To:Mime-Version:References:Message-ID:Subject:From: To:Content-Type; b=pXOCeHj44tKzrhWfecyBv5dndfTcMbIpRTJtkm4S6pJP2deonnZn41xXceKD5KQO21cKXyVrj4zXrR+gF7TedZIARE8Lb2TPgz8f5WYaiNPFpw09RoA0RwN4+GyIRzfOAGvOHADOWXaQX9Y8+MiV/fAzkF/WRveGyVpwqH5jVQQ= 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=TOJjg4s2; arc=none smtp.client-ip=209.85.214.201 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="TOJjg4s2" Received: by mail-pl1-f201.google.com with SMTP id d9443c01a7336-295b713530cso44560015ad.1 for ; Mon, 10 Nov 2025 20:04:58 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1762833898; x=1763438698; darn=vger.kernel.org; h=to:from:subject:message-id:references:mime-version:in-reply-to:date :from:to:cc:subject:date:message-id:reply-to; bh=9rrubnbGxO/GZTc54MaEZH3FM5tqPTfV8BFx99302Q0=; b=TOJjg4s2uUYgXGYo57ZGDTu+pwW+z8Wb52s2YHvyROvAMg1XM+DUW6IFpMMhqtrEbT b24oVaLLjP8pRlgBxbtO1heHthVgZ89m7+lOm4+ULVjQeesW3vYfx25uOv6cZ8ZfOKbU 2XVBXN0aFFw1ZRWIYNpwU52YbMAhVSsK8p1t/08zvzENk3nBYlJyjBCDaUX3/U1mU2gg Dbq8yJHWkuPvHsF5303wM2ut8fHnJMSwAyM5hc941XpHqyPzVDbOaJk8Pi3xF7Yi4H5H ZdtPEVTNnA9xHBMZX9nu73fN14Bfy7uwO2LJ2fMdvH7g3bCY8bnvBWdK3tKaV7aCLFPF jt/w== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1762833898; x=1763438698; h=to:from:subject:message-id:references:mime-version:in-reply-to:date :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=9rrubnbGxO/GZTc54MaEZH3FM5tqPTfV8BFx99302Q0=; b=XJHYNAEIs/2+Mv+MgNqfHvMUtYB1m6Rcnj/qxTzkUP5rSj6wUNLwBVRvl2sVZS2OmJ AfedktZv1Em0PREv2KVEx4hM8HhxfU21i7C/YqC6Mv5FwZfxJ9cpQZ7MbwOhz7omBcz4 Srwb84m4a21TyN+jtGZijZI82aM2y7rGbBc80ObFBgPKro+AyIj29YOcPkFpdGNjZBKg 2+R1wHgByGtPW00Usmsw9niyDXsmtfDkYBzChPZ4rcYTwgI9UKYZaVZEYynUuGugZjYS 28Bxi5j/NZVfauKimZQcSTAQ+D+YvkETPSrbse6/mzrfz6vHgd62s26zYdg5w4xKMY0L 72kw== X-Forwarded-Encrypted: i=1; AJvYcCWm4Eh5SwXWH68zXZxg0E1saMgbE1PaMlTj7FfN+mZapDnxRzHZVelLrgdhJ/aFNaV8BhGiDmlOzlBmjZY=@vger.kernel.org X-Gm-Message-State: AOJu0YyZAS2/zOpmQAEq2EhrS8fDoip6E1YoevMs/ACw0GWFd5JF3pXq HVEO+aRp57h2CIYBRtsdTI3FSwWSRleMX3r/z3/E1TbOFRHllUxLlXTuoPCujtPdpEd/LTGLj1W UHNi5mPPEsA== X-Google-Smtp-Source: AGHT+IHBUjhm1HAHK4fknr5+U2WBmlW8IIK5MxW46rm68+SMCCxdxu4P7aNsuW9rnpZloIRA3zNCNzCWCDVW X-Received: from dyctx7.prod.google.com ([2002:a05:7301:5f87:b0:2a4:612e:b41f]) (user=irogers job=prod-delivery.src-stubby-dispatcher) by 2002:a17:902:cf42:b0:248:ff5a:b768 with SMTP id d9443c01a7336-297e540d533mr131917455ad.10.1762833898118; Mon, 10 Nov 2025 20:04:58 -0800 (PST) Date: Mon, 10 Nov 2025 20:04:16 -0800 In-Reply-To: <20251111040417.270945-1-irogers@google.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 References: <20251111040417.270945-1-irogers@google.com> X-Mailer: git-send-email 2.51.2.1041.gc1ab5b90ca-goog Message-ID: <20251111040417.270945-18-irogers@google.com> Subject: [PATCH v3 17/18] perf test stat csv: Update test expectations and events From: Ian Rogers To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Namhyung Kim , Alexander Shishkin , Jiri Olsa , Ian Rogers , Adrian Hunter , James Clark , Xu Yang , Chun-Tse Shao , Thomas Richter , Sumanth Korikkar , Collin Funk , Thomas Falcon , Howard Chu , Dapeng Mi , Levi Yun , Yang Li , linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org, Andi Kleen , Weilin Wang Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" Explicitly use a metric rather than implicitly expecting '-e instructions,cycles' to produce a metric. Use a metric with software events to make it more compatible. Signed-off-by: Ian Rogers --- tools/perf/tests/shell/lib/stat_output.sh | 2 +- tools/perf/tests/shell/stat+csv_output.sh | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/tools/perf/tests/shell/lib/stat_output.sh b/tools/perf/tests/s= hell/lib/stat_output.sh index c2ec7881ec1d..3c36e80fe422 100644 --- a/tools/perf/tests/shell/lib/stat_output.sh +++ b/tools/perf/tests/shell/lib/stat_output.sh @@ -156,7 +156,7 @@ check_metric_only() echo "[Skip] CPU-measurement counter facility not installed" return fi - perf stat --metric-only $2 -e instructions,cycles true + perf stat --metric-only $2 -M page_faults_per_second true commachecker --metric-only echo "[Success]" } diff --git a/tools/perf/tests/shell/stat+csv_output.sh b/tools/perf/tests/s= hell/stat+csv_output.sh index 7a6f6e177402..cd6fff597091 100755 --- a/tools/perf/tests/shell/stat+csv_output.sh +++ b/tools/perf/tests/shell/stat+csv_output.sh @@ -44,7 +44,7 @@ function commachecker() ;; "--per-die") exp=3D8 ;; "--per-cluster") exp=3D8 ;; "--per-cache") exp=3D8 - ;; "--metric-only") exp=3D2 + ;; "--metric-only") exp=3D1 esac =20 while read line --=20 2.51.2.1041.gc1ab5b90ca-goog From nobody Tue Nov 11 11:28:57 2025 Received: from mail-pl1-f201.google.com (mail-pl1-f201.google.com [209.85.214.201]) (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 94B9234E759 for ; Tue, 11 Nov 2025 04:05:00 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.214.201 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1762833903; cv=none; b=PL805uyNuH0kpyILSTbA66bSSA1mDhNB6OXXCmU1rm0x8xuQRNcJBDjH16CXg36EyQdtuuptoll6mlSQ+ojNIdgMcXVqAU2W2hf1SIdq1x4boyAdEAdt3Eii5/dEqsC5uVpQ9t7RKCdJie29lhREWBPQwrydv9tx87itcO3Kf4A= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1762833903; c=relaxed/simple; bh=ewmX78hkfP1SRFYSr0MCWFxxlYVl4LAwuJeOP5hTQZU=; h=Date:In-Reply-To:Mime-Version:References:Message-ID:Subject:From: To:Content-Type; b=T9NLzPI2XCUvDzTAO2Cvm4NESyITfY/lIGhHpm5jOJpC7pZa6F7bm4bm1SwUSHS5vv9NYjzwvgWt8C/aJFHfvoKv1dDYkNnIfosJp6Pe7MiLkl/py9dxxfbhTNICTKfXtmL3I+e9zK+/mDEDShqtuX/KRYG7NxKk+UTaR9KeQlU= 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=mVIYDVfb; arc=none smtp.client-ip=209.85.214.201 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="mVIYDVfb" Received: by mail-pl1-f201.google.com with SMTP id d9443c01a7336-2954d676f9dso33243365ad.0 for ; Mon, 10 Nov 2025 20:05:00 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1762833900; x=1763438700; darn=vger.kernel.org; h=content-transfer-encoding:to:from:subject:message-id:references :mime-version:in-reply-to:date:from:to:cc:subject:date:message-id :reply-to; bh=JpGHvbA3zCa4V+xaJipcETglb/v3JOWW9yIsHsvCwgc=; b=mVIYDVfbGWstdvOp5KIyU1Ad0SIVZn7kCLjbfYOKWWjv8FsNPpwErehV80TGuFJYxp 7W8A4vQ99ZRD/TKc2pHm82JmjRZG33JgPCusPIDdn6eJGTIBKN7qedR8RSkx6wsKYwnd OTV/X8ytzlb81FJxfaXnxbQOHbot1+CNqSB9U0kTKP/18KfPUzlJ8GU2WgQ5/4BADL7V eE8CuxRizE8FZWzyZcic6xPHLKJJhYr0zcwNRN3vDG35iPi0xUOG9XIy+MCHlDcGgy4c TXzSJzngbxqHQgaxaE9awgjwTA3UVGFjxvOdOq42/8S2Jl5ig1BZyWpPkryOb0NDFrmd x20w== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1762833900; x=1763438700; h=content-transfer-encoding:to:from:subject:message-id:references :mime-version:in-reply-to:date:x-gm-message-state:from:to:cc:subject :date:message-id:reply-to; bh=JpGHvbA3zCa4V+xaJipcETglb/v3JOWW9yIsHsvCwgc=; b=dli5zmQLM9PvAZ20Qvm3HPXIFhTRxa49FBiO6GwVBSP8y1szvpC9i8vbcPuB7NErJp tIqlwjOjC1gXkLYvVUOy3j5iEEbTu0qzs+GYMAvj5Z9jfJVliuvkaxKS+6vsUWVGH7Oi VwXuDIV2Myd/2eIyDJecUEnGlQxOLjRL2M6pwli2gOYv3v/BMRX/t//9MLgD3WMCHfX8 6aCBhcx/35je1algMIDhKMkmHStYH437RsO+EZTSjamPrW3k8bRbC+HjdIv9jJJLoKb3 +tx8QHSxR2lGK95we/F3gQf5gGbv6rlwJvepH5FXV4MZzsSPLzeVBf35fgP5+NfLUOUn YThg== X-Forwarded-Encrypted: i=1; AJvYcCVacPFf+QatNWxel/CWtf4mNcBWlzu76dpuQjLw4abbqQ1T7OcQtluYD9RlF3npLgyOhc9rxPNiIes6Id4=@vger.kernel.org X-Gm-Message-State: AOJu0YxeLSoLfgz7N2cGzjUWImlxoNPOD0z0CwILhK1HfbH7nu5FbIH0 Q6WQbnFwzwqNM95X+WXbcdDncSKMnn+45dO/eagabVoqdBmYUfIhWS11nevwE1Vq+NWPrMtQaKU Y6lZblnDWuA== X-Google-Smtp-Source: AGHT+IHONuQujXTKcbbXt3nrjmzkvsSWtwr6xMu1FzK900sZqc6PnK73uyQqX7pNJiTfrR6775E5uIgfxzA1 X-Received: from dlbbz27.prod.google.com ([2002:a05:7022:239b:b0:119:49ca:6b8a]) (user=irogers job=prod-delivery.src-stubby-dispatcher) by 2002:a17:902:d4cb:b0:295:460b:5619 with SMTP id d9443c01a7336-297e53f6a2fmr102072945ad.8.1762833899960; Mon, 10 Nov 2025 20:04:59 -0800 (PST) Date: Mon, 10 Nov 2025 20:04:17 -0800 In-Reply-To: <20251111040417.270945-1-irogers@google.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 References: <20251111040417.270945-1-irogers@google.com> X-Mailer: git-send-email 2.51.2.1041.gc1ab5b90ca-goog Message-ID: <20251111040417.270945-19-irogers@google.com> Subject: [PATCH v3 18/18] perf tool_pmu: Make core_wide and target_cpu json events From: Ian Rogers To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Namhyung Kim , Alexander Shishkin , Jiri Olsa , Ian Rogers , Adrian Hunter , James Clark , Xu Yang , Chun-Tse Shao , Thomas Richter , Sumanth Korikkar , Collin Funk , Thomas Falcon , Howard Chu , Dapeng Mi , Levi Yun , Yang Li , linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org, Andi Kleen , Weilin Wang Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" For the sake of better documentation, add core_wide and target_cpu to the tool.json. When the values of system_wide and user_requested_cpu_list are unknown, use the values from the global stat_config. Example output showing how '-a' modifies the values in `perf stat`: ``` $ perf stat -e core_wide,target_cpu true Performance counter stats for 'true': 0 core_wide 0 target_cpu 0.000993787 seconds time elapsed 0.001128000 seconds user 0.000000000 seconds sys $ perf stat -e core_wide,target_cpu -a true Performance counter stats for 'system wide': 1 core_wide 1 target_cpu 0.002271723 seconds time elapsed $ perf list ... tool: core_wide [1 if not SMT,if SMT are events being gathered on all SMT threads 1 = otherwise 0. Unit: tool] ... target_cpu [1 if CPUs being analyzed,0 if threads/processes. Unit: tool] ... ``` Signed-off-by: Ian Rogers --- .../pmu-events/arch/common/common/tool.json | 12 + tools/perf/pmu-events/empty-pmu-events.c | 228 +++++++++--------- tools/perf/util/expr.c | 11 +- tools/perf/util/stat-shadow.c | 2 + tools/perf/util/tool_pmu.c | 24 +- tools/perf/util/tool_pmu.h | 9 +- 6 files changed, 163 insertions(+), 123 deletions(-) diff --git a/tools/perf/pmu-events/arch/common/common/tool.json b/tools/per= f/pmu-events/arch/common/common/tool.json index 12f2ef1813a6..14d0d60a1976 100644 --- a/tools/perf/pmu-events/arch/common/common/tool.json +++ b/tools/perf/pmu-events/arch/common/common/tool.json @@ -70,5 +70,17 @@ "EventName": "system_tsc_freq", "BriefDescription": "The amount a Time Stamp Counter (TSC) increases p= er second", "ConfigCode": "12" + }, + { + "Unit": "tool", + "EventName": "core_wide", + "BriefDescription": "1 if not SMT, if SMT are events being gathered on= all SMT threads 1 otherwise 0", + "ConfigCode": "13" + }, + { + "Unit": "tool", + "EventName": "target_cpu", + "BriefDescription": "1 if CPUs being analyzed, 0 if threads/processes", + "ConfigCode": "14" } ] diff --git a/tools/perf/pmu-events/empty-pmu-events.c b/tools/perf/pmu-even= ts/empty-pmu-events.c index 6fc490b96a02..969d1ce6c7b9 100644 --- a/tools/perf/pmu-events/empty-pmu-events.c +++ b/tools/perf/pmu-events/empty-pmu-events.c @@ -1279,62 +1279,64 @@ static const char *const big_c_string =3D /* offset=3D125889 */ "slots\000tool\000Number of functional units that in= parallel can execute parts of an instruction\000config=3D0xa\000\00000\000= \000\000\000\000" /* offset=3D125999 */ "smt_on\000tool\0001 if simultaneous multithreading = (aka hyperthreading) is enable otherwise 0\000config=3D0xb\000\00000\000\00= 0\000\000\000" /* offset=3D126106 */ "system_tsc_freq\000tool\000The amount a Time Stamp = Counter (TSC) increases per second\000config=3D0xc\000\00000\000\000\000\00= 0\000" -/* offset=3D126205 */ "bp_l1_btb_correct\000branch\000L1 BTB Correction\00= 0event=3D0x8a\000\00000\000\000\000\000\000" -/* offset=3D126267 */ "bp_l2_btb_correct\000branch\000L2 BTB Correction\00= 0event=3D0x8b\000\00000\000\000\000\000\000" -/* offset=3D126329 */ "l3_cache_rd\000cache\000L3 cache access, read\000ev= ent=3D0x40\000\00000\000\000\000\000Attributable Level 3 cache access, read= \000" -/* offset=3D126427 */ "segment_reg_loads.any\000other\000Number of segment= register loads\000event=3D6,period=3D200000,umask=3D0x80\000\00000\000\000= \000\000\000" -/* offset=3D126529 */ "dispatch_blocked.any\000other\000Memory cluster sig= nals to block micro-op dispatch for any reason\000event=3D9,period=3D200000= ,umask=3D0x20\000\00000\000\000\000\000\000" -/* offset=3D126662 */ "eist_trans\000other\000Number of Enhanced Intel Spe= edStep(R) Technology (EIST) transitions\000event=3D0x3a,period=3D200000\000= \00000\000\000\000\000\000" -/* offset=3D126780 */ "hisi_sccl,ddrc\000" -/* offset=3D126795 */ "uncore_hisi_ddrc.flux_wcmd\000uncore\000DDRC write = commands\000event=3D2\000\00000\000\000\000\000\000" -/* offset=3D126865 */ "uncore_cbox\000" -/* offset=3D126877 */ "unc_cbo_xsnp_response.miss_eviction\000uncore\000A = cross-core snoop resulted from L3 Eviction which misses in some processor c= ore\000event=3D0x22,umask=3D0x81\000\00000\000\000\000\000\000" -/* offset=3D127031 */ "event-hyphen\000uncore\000UNC_CBO_HYPHEN\000event= =3D0xe0\000\00000\000\000\000\000\000" -/* offset=3D127085 */ "event-two-hyph\000uncore\000UNC_CBO_TWO_HYPH\000eve= nt=3D0xc0\000\00000\000\000\000\000\000" -/* offset=3D127143 */ "hisi_sccl,l3c\000" -/* offset=3D127157 */ "uncore_hisi_l3c.rd_hit_cpipe\000uncore\000Total rea= d hits\000event=3D7\000\00000\000\000\000\000\000" -/* offset=3D127225 */ "uncore_imc_free_running\000" -/* offset=3D127249 */ "uncore_imc_free_running.cache_miss\000uncore\000Tot= al cache misses\000event=3D0x12\000\00000\000\000\000\000\000" -/* offset=3D127329 */ "uncore_imc\000" -/* offset=3D127340 */ "uncore_imc.cache_hits\000uncore\000Total cache hits= \000event=3D0x34\000\00000\000\000\000\000\000" -/* offset=3D127405 */ "uncore_sys_ddr_pmu\000" -/* offset=3D127424 */ "sys_ddr_pmu.write_cycles\000uncore\000ddr write-cyc= les event\000event=3D0x2b\000v8\00000\000\000\000\000\000" -/* offset=3D127500 */ "uncore_sys_ccn_pmu\000" -/* offset=3D127519 */ "sys_ccn_pmu.read_cycles\000uncore\000ccn read-cycle= s event\000config=3D0x2c\0000x01\00000\000\000\000\000\000" -/* offset=3D127596 */ "uncore_sys_cmn_pmu\000" -/* offset=3D127615 */ "sys_cmn_pmu.hnf_cache_miss\000uncore\000Counts tota= l cache misses in first lookup result (high priority)\000eventid=3D1,type= =3D5\000(434|436|43c|43a).*\00000\000\000\000\000\000" -/* offset=3D127758 */ "CPUs_utilized\000Default\000(software@cpu\\-clock\\= ,name\\=3Dcpu\\-clock@ if #target_cpu else software@task\\-clock\\,name\\= =3Dtask\\-clock@) / (duration_time * 1e9)\000\000Average CPU utilization\00= 0\0001CPUs\000\000\000\000011" -/* offset=3D127944 */ "cs_per_second\000Default\000software@context\\-swit= ches\\,name\\=3Dcontext\\-switches@ * 1e9 / (software@cpu\\-clock\\,name\\= =3Dcpu\\-clock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtask\\= -clock@)\000\000Context switches per CPU second\000\0001cs/sec\000\000\000\= 000011" -/* offset=3D128177 */ "migrations_per_second\000Default\000software@cpu\\-= migrations\\,name\\=3Dcpu\\-migrations@ * 1e9 / (software@cpu\\-clock\\,nam= e\\=3Dcpu\\-clock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtas= k\\-clock@)\000\000Process migrations to a new CPU per CPU second\000\0001m= igrations/sec\000\000\000\000011" -/* offset=3D128437 */ "page_faults_per_second\000Default\000software@page\= \-faults\\,name\\=3Dpage\\-faults@ * 1e9 / (software@cpu\\-clock\\,name\\= =3Dcpu\\-clock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtask\\= -clock@)\000\000Page faults per CPU second\000\0001faults/sec\000\000\000\0= 00011" -/* offset=3D128668 */ "insn_per_cycle\000Default\000instructions / cpu\\-c= ycles\000insn_per_cycle < 1\000Instructions Per Cycle\000\0001instructions\= 000\000\000\000001" -/* offset=3D128781 */ "stalled_cycles_per_instruction\000Default\000max(st= alled\\-cycles\\-frontend, stalled\\-cycles\\-backend) / instructions\000\0= 00Max front or backend stalls per instruction\000\000\000\000\000\000001" -/* offset=3D128945 */ "frontend_cycles_idle\000Default\000stalled\\-cycles= \\-frontend / cpu\\-cycles\000frontend_cycles_idle > 0.1\000Frontend stalls= per cycle\000\000\000\000\000\000001" -/* offset=3D129075 */ "backend_cycles_idle\000Default\000stalled\\-cycles\= \-backend / cpu\\-cycles\000backend_cycles_idle > 0.2\000Backend stalls per= cycle\000\000\000\000\000\000001" -/* offset=3D129201 */ "cycles_frequency\000Default\000cpu\\-cycles / (soft= ware@cpu\\-clock\\,name\\=3Dcpu\\-clock@ if #target_cpu else software@task\= \-clock\\,name\\=3Dtask\\-clock@)\000\000Cycles per CPU second\000\0001GHz\= 000\000\000\000011" -/* offset=3D129377 */ "branch_frequency\000Default\000branches / (software= @cpu\\-clock\\,name\\=3Dcpu\\-clock@ if #target_cpu else software@task\\-cl= ock\\,name\\=3Dtask\\-clock@)\000\000Branches per CPU second\000\0001000K/s= ec\000\000\000\000011" -/* offset=3D129557 */ "branch_miss_rate\000Default\000branch\\-misses / br= anches\000branch_miss_rate > 0.05\000Branch miss rate\000\000100%\000\000\0= 00\000001" -/* offset=3D129661 */ "l1d_miss_rate\000Default2\000L1\\-dcache\\-load\\-m= isses / L1\\-dcache\\-loads\000l1d_miss_rate > 0.05\000L1D miss rate\000\0= 00100%\000\000\000\000001" -/* offset=3D129777 */ "llc_miss_rate\000Default2\000LLC\\-load\\-misses / = LLC\\-loads\000llc_miss_rate > 0.05\000LLC miss rate\000\000100%\000\000\00= 0\000001" -/* offset=3D129878 */ "l1i_miss_rate\000Default3\000L1\\-icache\\-load\\-m= isses / L1\\-icache\\-loads\000l1i_miss_rate > 0.05\000L1I miss rate\000\00= 0100%\000\000\000\000001" -/* offset=3D129993 */ "dtlb_miss_rate\000Default3\000dTLB\\-load\\-misses = / dTLB\\-loads\000dtlb_miss_rate > 0.05\000dTLB miss rate\000\000100%\000\0= 00\000\000001" -/* offset=3D130099 */ "itlb_miss_rate\000Default3\000iTLB\\-load\\-misses = / iTLB\\-loads\000itlb_miss_rate > 0.05\000iTLB miss rate\000\000100%\000\0= 00\000\000001" -/* offset=3D130205 */ "l1_prefetch_miss_rate\000Default4\000L1\\-dcache\\-= prefetch\\-misses / L1\\-dcache\\-prefetches\000l1_prefetch_miss_rate > 0.0= 5\000L1 prefetch miss rate\000\000100%\000\000\000\000001" -/* offset=3D130353 */ "CPI\000\0001 / IPC\000\000\000\000\000\000\000\0000= 00" -/* offset=3D130376 */ "IPC\000group1\000inst_retired.any / cpu_clk_unhalte= d.thread\000\000\000\000\000\000\000\000000" -/* offset=3D130440 */ "Frontend_Bound_SMT\000\000idq_uops_not_delivered.co= re / (4 * (cpu_clk_unhalted.thread / 2 * (1 + cpu_clk_unhalted.one_thread_a= ctive / cpu_clk_unhalted.ref_xclk)))\000\000\000\000\000\000\000\000000" -/* offset=3D130607 */ "dcache_miss_cpi\000\000l1d\\-loads\\-misses / inst_= retired.any\000\000\000\000\000\000\000\000000" -/* offset=3D130672 */ "icache_miss_cycles\000\000l1i\\-loads\\-misses / in= st_retired.any\000\000\000\000\000\000\000\000000" -/* offset=3D130740 */ "cache_miss_cycles\000group1\000dcache_miss_cpi + ic= ache_miss_cycles\000\000\000\000\000\000\000\000000" -/* offset=3D130812 */ "DCache_L2_All_Hits\000\000l2_rqsts.demand_data_rd_h= it + l2_rqsts.pf_hit + l2_rqsts.rfo_hit\000\000\000\000\000\000\000\000000" -/* offset=3D130907 */ "DCache_L2_All_Miss\000\000max(l2_rqsts.all_demand_d= ata_rd - l2_rqsts.demand_data_rd_hit, 0) + l2_rqsts.pf_miss + l2_rqsts.rfo_= miss\000\000\000\000\000\000\000\000000" -/* offset=3D131042 */ "DCache_L2_All\000\000DCache_L2_All_Hits + DCache_L2= _All_Miss\000\000\000\000\000\000\000\000000" -/* offset=3D131107 */ "DCache_L2_Hits\000\000d_ratio(DCache_L2_All_Hits, D= Cache_L2_All)\000\000\000\000\000\000\000\000000" -/* offset=3D131176 */ "DCache_L2_Misses\000\000d_ratio(DCache_L2_All_Miss,= DCache_L2_All)\000\000\000\000\000\000\000\000000" -/* offset=3D131247 */ "M1\000\000ipc + M2\000\000\000\000\000\000\000\0000= 00" -/* offset=3D131270 */ "M2\000\000ipc + M1\000\000\000\000\000\000\000\0000= 00" -/* offset=3D131293 */ "M3\000\0001 / M3\000\000\000\000\000\000\000\000000" -/* offset=3D131314 */ "L1D_Cache_Fill_BW\000\00064 * l1d.replacement / 1e9= / duration_time\000\000\000\000\000\000\000\000000" +/* offset=3D126205 */ "core_wide\000tool\0001 if not SMT, if SMT are event= s being gathered on all SMT threads 1 otherwise 0\000config=3D0xd\000\00000= \000\000\000\000\000" +/* offset=3D126319 */ "target_cpu\000tool\0001 if CPUs being analyzed, 0 i= f threads/processes\000config=3D0xe\000\00000\000\000\000\000\000" +/* offset=3D126403 */ "bp_l1_btb_correct\000branch\000L1 BTB Correction\00= 0event=3D0x8a\000\00000\000\000\000\000\000" +/* offset=3D126465 */ "bp_l2_btb_correct\000branch\000L2 BTB Correction\00= 0event=3D0x8b\000\00000\000\000\000\000\000" +/* offset=3D126527 */ "l3_cache_rd\000cache\000L3 cache access, read\000ev= ent=3D0x40\000\00000\000\000\000\000Attributable Level 3 cache access, read= \000" +/* offset=3D126625 */ "segment_reg_loads.any\000other\000Number of segment= register loads\000event=3D6,period=3D200000,umask=3D0x80\000\00000\000\000= \000\000\000" +/* offset=3D126727 */ "dispatch_blocked.any\000other\000Memory cluster sig= nals to block micro-op dispatch for any reason\000event=3D9,period=3D200000= ,umask=3D0x20\000\00000\000\000\000\000\000" +/* offset=3D126860 */ "eist_trans\000other\000Number of Enhanced Intel Spe= edStep(R) Technology (EIST) transitions\000event=3D0x3a,period=3D200000\000= \00000\000\000\000\000\000" +/* offset=3D126978 */ "hisi_sccl,ddrc\000" +/* offset=3D126993 */ "uncore_hisi_ddrc.flux_wcmd\000uncore\000DDRC write = commands\000event=3D2\000\00000\000\000\000\000\000" +/* offset=3D127063 */ "uncore_cbox\000" +/* offset=3D127075 */ "unc_cbo_xsnp_response.miss_eviction\000uncore\000A = cross-core snoop resulted from L3 Eviction which misses in some processor c= ore\000event=3D0x22,umask=3D0x81\000\00000\000\000\000\000\000" +/* offset=3D127229 */ "event-hyphen\000uncore\000UNC_CBO_HYPHEN\000event= =3D0xe0\000\00000\000\000\000\000\000" +/* offset=3D127283 */ "event-two-hyph\000uncore\000UNC_CBO_TWO_HYPH\000eve= nt=3D0xc0\000\00000\000\000\000\000\000" +/* offset=3D127341 */ "hisi_sccl,l3c\000" +/* offset=3D127355 */ "uncore_hisi_l3c.rd_hit_cpipe\000uncore\000Total rea= d hits\000event=3D7\000\00000\000\000\000\000\000" +/* offset=3D127423 */ "uncore_imc_free_running\000" +/* offset=3D127447 */ "uncore_imc_free_running.cache_miss\000uncore\000Tot= al cache misses\000event=3D0x12\000\00000\000\000\000\000\000" +/* offset=3D127527 */ "uncore_imc\000" +/* offset=3D127538 */ "uncore_imc.cache_hits\000uncore\000Total cache hits= \000event=3D0x34\000\00000\000\000\000\000\000" +/* offset=3D127603 */ "uncore_sys_ddr_pmu\000" +/* offset=3D127622 */ "sys_ddr_pmu.write_cycles\000uncore\000ddr write-cyc= les event\000event=3D0x2b\000v8\00000\000\000\000\000\000" +/* offset=3D127698 */ "uncore_sys_ccn_pmu\000" +/* offset=3D127717 */ "sys_ccn_pmu.read_cycles\000uncore\000ccn read-cycle= s event\000config=3D0x2c\0000x01\00000\000\000\000\000\000" +/* offset=3D127794 */ "uncore_sys_cmn_pmu\000" +/* offset=3D127813 */ "sys_cmn_pmu.hnf_cache_miss\000uncore\000Counts tota= l cache misses in first lookup result (high priority)\000eventid=3D1,type= =3D5\000(434|436|43c|43a).*\00000\000\000\000\000\000" +/* offset=3D127956 */ "CPUs_utilized\000Default\000(software@cpu\\-clock\\= ,name\\=3Dcpu\\-clock@ if #target_cpu else software@task\\-clock\\,name\\= =3Dtask\\-clock@) / (duration_time * 1e9)\000\000Average CPU utilization\00= 0\0001CPUs\000\000\000\000011" +/* offset=3D128142 */ "cs_per_second\000Default\000software@context\\-swit= ches\\,name\\=3Dcontext\\-switches@ * 1e9 / (software@cpu\\-clock\\,name\\= =3Dcpu\\-clock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtask\\= -clock@)\000\000Context switches per CPU second\000\0001cs/sec\000\000\000\= 000011" +/* offset=3D128375 */ "migrations_per_second\000Default\000software@cpu\\-= migrations\\,name\\=3Dcpu\\-migrations@ * 1e9 / (software@cpu\\-clock\\,nam= e\\=3Dcpu\\-clock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtas= k\\-clock@)\000\000Process migrations to a new CPU per CPU second\000\0001m= igrations/sec\000\000\000\000011" +/* offset=3D128635 */ "page_faults_per_second\000Default\000software@page\= \-faults\\,name\\=3Dpage\\-faults@ * 1e9 / (software@cpu\\-clock\\,name\\= =3Dcpu\\-clock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtask\\= -clock@)\000\000Page faults per CPU second\000\0001faults/sec\000\000\000\0= 00011" +/* offset=3D128866 */ "insn_per_cycle\000Default\000instructions / cpu\\-c= ycles\000insn_per_cycle < 1\000Instructions Per Cycle\000\0001instructions\= 000\000\000\000001" +/* offset=3D128979 */ "stalled_cycles_per_instruction\000Default\000max(st= alled\\-cycles\\-frontend, stalled\\-cycles\\-backend) / instructions\000\0= 00Max front or backend stalls per instruction\000\000\000\000\000\000001" +/* offset=3D129143 */ "frontend_cycles_idle\000Default\000stalled\\-cycles= \\-frontend / cpu\\-cycles\000frontend_cycles_idle > 0.1\000Frontend stalls= per cycle\000\000\000\000\000\000001" +/* offset=3D129273 */ "backend_cycles_idle\000Default\000stalled\\-cycles\= \-backend / cpu\\-cycles\000backend_cycles_idle > 0.2\000Backend stalls per= cycle\000\000\000\000\000\000001" +/* offset=3D129399 */ "cycles_frequency\000Default\000cpu\\-cycles / (soft= ware@cpu\\-clock\\,name\\=3Dcpu\\-clock@ if #target_cpu else software@task\= \-clock\\,name\\=3Dtask\\-clock@)\000\000Cycles per CPU second\000\0001GHz\= 000\000\000\000011" +/* offset=3D129575 */ "branch_frequency\000Default\000branches / (software= @cpu\\-clock\\,name\\=3Dcpu\\-clock@ if #target_cpu else software@task\\-cl= ock\\,name\\=3Dtask\\-clock@)\000\000Branches per CPU second\000\0001000K/s= ec\000\000\000\000011" +/* offset=3D129755 */ "branch_miss_rate\000Default\000branch\\-misses / br= anches\000branch_miss_rate > 0.05\000Branch miss rate\000\000100%\000\000\0= 00\000001" +/* offset=3D129859 */ "l1d_miss_rate\000Default2\000L1\\-dcache\\-load\\-m= isses / L1\\-dcache\\-loads\000l1d_miss_rate > 0.05\000L1D miss rate\000\0= 00100%\000\000\000\000001" +/* offset=3D129975 */ "llc_miss_rate\000Default2\000LLC\\-load\\-misses / = LLC\\-loads\000llc_miss_rate > 0.05\000LLC miss rate\000\000100%\000\000\00= 0\000001" +/* offset=3D130076 */ "l1i_miss_rate\000Default3\000L1\\-icache\\-load\\-m= isses / L1\\-icache\\-loads\000l1i_miss_rate > 0.05\000L1I miss rate\000\00= 0100%\000\000\000\000001" +/* offset=3D130191 */ "dtlb_miss_rate\000Default3\000dTLB\\-load\\-misses = / dTLB\\-loads\000dtlb_miss_rate > 0.05\000dTLB miss rate\000\000100%\000\0= 00\000\000001" +/* offset=3D130297 */ "itlb_miss_rate\000Default3\000iTLB\\-load\\-misses = / iTLB\\-loads\000itlb_miss_rate > 0.05\000iTLB miss rate\000\000100%\000\0= 00\000\000001" +/* offset=3D130403 */ "l1_prefetch_miss_rate\000Default4\000L1\\-dcache\\-= prefetch\\-misses / L1\\-dcache\\-prefetches\000l1_prefetch_miss_rate > 0.0= 5\000L1 prefetch miss rate\000\000100%\000\000\000\000001" +/* offset=3D130551 */ "CPI\000\0001 / IPC\000\000\000\000\000\000\000\0000= 00" +/* offset=3D130574 */ "IPC\000group1\000inst_retired.any / cpu_clk_unhalte= d.thread\000\000\000\000\000\000\000\000000" +/* offset=3D130638 */ "Frontend_Bound_SMT\000\000idq_uops_not_delivered.co= re / (4 * (cpu_clk_unhalted.thread / 2 * (1 + cpu_clk_unhalted.one_thread_a= ctive / cpu_clk_unhalted.ref_xclk)))\000\000\000\000\000\000\000\000000" +/* offset=3D130805 */ "dcache_miss_cpi\000\000l1d\\-loads\\-misses / inst_= retired.any\000\000\000\000\000\000\000\000000" +/* offset=3D130870 */ "icache_miss_cycles\000\000l1i\\-loads\\-misses / in= st_retired.any\000\000\000\000\000\000\000\000000" +/* offset=3D130938 */ "cache_miss_cycles\000group1\000dcache_miss_cpi + ic= ache_miss_cycles\000\000\000\000\000\000\000\000000" +/* offset=3D131010 */ "DCache_L2_All_Hits\000\000l2_rqsts.demand_data_rd_h= it + l2_rqsts.pf_hit + l2_rqsts.rfo_hit\000\000\000\000\000\000\000\000000" +/* offset=3D131105 */ "DCache_L2_All_Miss\000\000max(l2_rqsts.all_demand_d= ata_rd - l2_rqsts.demand_data_rd_hit, 0) + l2_rqsts.pf_miss + l2_rqsts.rfo_= miss\000\000\000\000\000\000\000\000000" +/* offset=3D131240 */ "DCache_L2_All\000\000DCache_L2_All_Hits + DCache_L2= _All_Miss\000\000\000\000\000\000\000\000000" +/* offset=3D131305 */ "DCache_L2_Hits\000\000d_ratio(DCache_L2_All_Hits, D= Cache_L2_All)\000\000\000\000\000\000\000\000000" +/* offset=3D131374 */ "DCache_L2_Misses\000\000d_ratio(DCache_L2_All_Miss,= DCache_L2_All)\000\000\000\000\000\000\000\000000" +/* offset=3D131445 */ "M1\000\000ipc + M2\000\000\000\000\000\000\000\0000= 00" +/* offset=3D131468 */ "M2\000\000ipc + M1\000\000\000\000\000\000\000\0000= 00" +/* offset=3D131491 */ "M3\000\0001 / M3\000\000\000\000\000\000\000\000000" +/* offset=3D131512 */ "L1D_Cache_Fill_BW\000\00064 * l1d.replacement / 1e9= / duration_time\000\000\000\000\000\000\000\000000" ; =20 static const struct compact_pmu_event pmu_events__common_default_core[] = =3D { @@ -2587,6 +2589,7 @@ static const struct compact_pmu_event pmu_events__com= mon_software[] =3D { { 123607 }, /* task-clock\000software\000Per-task high-resolution timer ba= sed event\000config=3D1\000\000001e-6msec\000\000\000\000\000 */ }; static const struct compact_pmu_event pmu_events__common_tool[] =3D { +{ 126205 }, /* core_wide\000tool\0001 if not SMT, if SMT are events being = gathered on all SMT threads 1 otherwise 0\000config=3D0xd\000\00000\000\000= \000\000\000 */ { 125072 }, /* duration_time\000tool\000Wall clock interval time in nanose= conds\000config=3D1\000\00000\000\000\000\000\000 */ { 125286 }, /* has_pmem\000tool\0001 if persistent memory installed otherw= ise 0\000config=3D4\000\00000\000\000\000\000\000 */ { 125362 }, /* num_cores\000tool\000Number of cores. A core consists of 1 = or more thread, with each thread being associated with a logical Linux CPU\= 000config=3D5\000\00000\000\000\000\000\000 */ @@ -2598,6 +2601,7 @@ static const struct compact_pmu_event pmu_events__com= mon_tool[] =3D { { 125999 }, /* smt_on\000tool\0001 if simultaneous multithreading (aka hyp= erthreading) is enable otherwise 0\000config=3D0xb\000\00000\000\000\000\00= 0\000 */ { 125218 }, /* system_time\000tool\000System/kernel time in nanoseconds\00= 0config=3D3\000\00000\000\000\000\000\000 */ { 126106 }, /* system_tsc_freq\000tool\000The amount a Time Stamp Counter = (TSC) increases per second\000config=3D0xc\000\00000\000\000\000\000\000 */ +{ 126319 }, /* target_cpu\000tool\0001 if CPUs being analyzed, 0 if thread= s/processes\000config=3D0xe\000\00000\000\000\000\000\000 */ { 125148 }, /* user_time\000tool\000User (non-kernel) time in nanoseconds\= 000config=3D2\000\00000\000\000\000\000\000 */ =20 }; @@ -2621,23 +2625,23 @@ static const struct pmu_table_entry pmu_events__com= mon[] =3D { }; =20 static const struct compact_pmu_event pmu_metrics__common_default_core[] = =3D { -{ 127758 }, /* CPUs_utilized\000Default\000(software@cpu\\-clock\\,name\\= =3Dcpu\\-clock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtask\\= -clock@) / (duration_time * 1e9)\000\000Average CPU utilization\000\0001CPU= s\000\000\000\000011 */ -{ 129075 }, /* backend_cycles_idle\000Default\000stalled\\-cycles\\-backen= d / cpu\\-cycles\000backend_cycles_idle > 0.2\000Backend stalls per cycle\0= 00\000\000\000\000\000001 */ -{ 129377 }, /* branch_frequency\000Default\000branches / (software@cpu\\-c= lock\\,name\\=3Dcpu\\-clock@ if #target_cpu else software@task\\-clock\\,na= me\\=3Dtask\\-clock@)\000\000Branches per CPU second\000\0001000K/sec\000\0= 00\000\000011 */ -{ 129557 }, /* branch_miss_rate\000Default\000branch\\-misses / branches\0= 00branch_miss_rate > 0.05\000Branch miss rate\000\000100%\000\000\000\00000= 1 */ -{ 127944 }, /* cs_per_second\000Default\000software@context\\-switches\\,n= ame\\=3Dcontext\\-switches@ * 1e9 / (software@cpu\\-clock\\,name\\=3Dcpu\\-= clock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtask\\-clock@)\= 000\000Context switches per CPU second\000\0001cs/sec\000\000\000\000011 */ -{ 129201 }, /* cycles_frequency\000Default\000cpu\\-cycles / (software@cpu= \\-clock\\,name\\=3Dcpu\\-clock@ if #target_cpu else software@task\\-clock\= \,name\\=3Dtask\\-clock@)\000\000Cycles per CPU second\000\0001GHz\000\000\= 000\000011 */ -{ 129993 }, /* dtlb_miss_rate\000Default3\000dTLB\\-load\\-misses / dTLB\\= -loads\000dtlb_miss_rate > 0.05\000dTLB miss rate\000\000100%\000\000\000\0= 00001 */ -{ 128945 }, /* frontend_cycles_idle\000Default\000stalled\\-cycles\\-front= end / cpu\\-cycles\000frontend_cycles_idle > 0.1\000Frontend stalls per cyc= le\000\000\000\000\000\000001 */ -{ 128668 }, /* insn_per_cycle\000Default\000instructions / cpu\\-cycles\00= 0insn_per_cycle < 1\000Instructions Per Cycle\000\0001instructions\000\000\= 000\000001 */ -{ 130099 }, /* itlb_miss_rate\000Default3\000iTLB\\-load\\-misses / iTLB\\= -loads\000itlb_miss_rate > 0.05\000iTLB miss rate\000\000100%\000\000\000\0= 00001 */ -{ 130205 }, /* l1_prefetch_miss_rate\000Default4\000L1\\-dcache\\-prefetch= \\-misses / L1\\-dcache\\-prefetches\000l1_prefetch_miss_rate > 0.05\000L1 = prefetch miss rate\000\000100%\000\000\000\000001 */ -{ 129661 }, /* l1d_miss_rate\000Default2\000L1\\-dcache\\-load\\-misses / = L1\\-dcache\\-loads\000l1d_miss_rate > 0.05\000L1D miss rate\000\000100%\0= 00\000\000\000001 */ -{ 129878 }, /* l1i_miss_rate\000Default3\000L1\\-icache\\-load\\-misses / = L1\\-icache\\-loads\000l1i_miss_rate > 0.05\000L1I miss rate\000\000100%\00= 0\000\000\000001 */ -{ 129777 }, /* llc_miss_rate\000Default2\000LLC\\-load\\-misses / LLC\\-lo= ads\000llc_miss_rate > 0.05\000LLC miss rate\000\000100%\000\000\000\000001= */ -{ 128177 }, /* migrations_per_second\000Default\000software@cpu\\-migratio= ns\\,name\\=3Dcpu\\-migrations@ * 1e9 / (software@cpu\\-clock\\,name\\=3Dcp= u\\-clock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtask\\-cloc= k@)\000\000Process migrations to a new CPU per CPU second\000\0001migration= s/sec\000\000\000\000011 */ -{ 128437 }, /* page_faults_per_second\000Default\000software@page\\-faults= \\,name\\=3Dpage\\-faults@ * 1e9 / (software@cpu\\-clock\\,name\\=3Dcpu\\-c= lock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtask\\-clock@)\0= 00\000Page faults per CPU second\000\0001faults/sec\000\000\000\000011 */ -{ 128781 }, /* stalled_cycles_per_instruction\000Default\000max(stalled\\-= cycles\\-frontend, stalled\\-cycles\\-backend) / instructions\000\000Max fr= ont or backend stalls per instruction\000\000\000\000\000\000001 */ +{ 127956 }, /* CPUs_utilized\000Default\000(software@cpu\\-clock\\,name\\= =3Dcpu\\-clock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtask\\= -clock@) / (duration_time * 1e9)\000\000Average CPU utilization\000\0001CPU= s\000\000\000\000011 */ +{ 129273 }, /* backend_cycles_idle\000Default\000stalled\\-cycles\\-backen= d / cpu\\-cycles\000backend_cycles_idle > 0.2\000Backend stalls per cycle\0= 00\000\000\000\000\000001 */ +{ 129575 }, /* branch_frequency\000Default\000branches / (software@cpu\\-c= lock\\,name\\=3Dcpu\\-clock@ if #target_cpu else software@task\\-clock\\,na= me\\=3Dtask\\-clock@)\000\000Branches per CPU second\000\0001000K/sec\000\0= 00\000\000011 */ +{ 129755 }, /* branch_miss_rate\000Default\000branch\\-misses / branches\0= 00branch_miss_rate > 0.05\000Branch miss rate\000\000100%\000\000\000\00000= 1 */ +{ 128142 }, /* cs_per_second\000Default\000software@context\\-switches\\,n= ame\\=3Dcontext\\-switches@ * 1e9 / (software@cpu\\-clock\\,name\\=3Dcpu\\-= clock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtask\\-clock@)\= 000\000Context switches per CPU second\000\0001cs/sec\000\000\000\000011 */ +{ 129399 }, /* cycles_frequency\000Default\000cpu\\-cycles / (software@cpu= \\-clock\\,name\\=3Dcpu\\-clock@ if #target_cpu else software@task\\-clock\= \,name\\=3Dtask\\-clock@)\000\000Cycles per CPU second\000\0001GHz\000\000\= 000\000011 */ +{ 130191 }, /* dtlb_miss_rate\000Default3\000dTLB\\-load\\-misses / dTLB\\= -loads\000dtlb_miss_rate > 0.05\000dTLB miss rate\000\000100%\000\000\000\0= 00001 */ +{ 129143 }, /* frontend_cycles_idle\000Default\000stalled\\-cycles\\-front= end / cpu\\-cycles\000frontend_cycles_idle > 0.1\000Frontend stalls per cyc= le\000\000\000\000\000\000001 */ +{ 128866 }, /* insn_per_cycle\000Default\000instructions / cpu\\-cycles\00= 0insn_per_cycle < 1\000Instructions Per Cycle\000\0001instructions\000\000\= 000\000001 */ +{ 130297 }, /* itlb_miss_rate\000Default3\000iTLB\\-load\\-misses / iTLB\\= -loads\000itlb_miss_rate > 0.05\000iTLB miss rate\000\000100%\000\000\000\0= 00001 */ +{ 130403 }, /* l1_prefetch_miss_rate\000Default4\000L1\\-dcache\\-prefetch= \\-misses / L1\\-dcache\\-prefetches\000l1_prefetch_miss_rate > 0.05\000L1 = prefetch miss rate\000\000100%\000\000\000\000001 */ +{ 129859 }, /* l1d_miss_rate\000Default2\000L1\\-dcache\\-load\\-misses / = L1\\-dcache\\-loads\000l1d_miss_rate > 0.05\000L1D miss rate\000\000100%\0= 00\000\000\000001 */ +{ 130076 }, /* l1i_miss_rate\000Default3\000L1\\-icache\\-load\\-misses / = L1\\-icache\\-loads\000l1i_miss_rate > 0.05\000L1I miss rate\000\000100%\00= 0\000\000\000001 */ +{ 129975 }, /* llc_miss_rate\000Default2\000LLC\\-load\\-misses / LLC\\-lo= ads\000llc_miss_rate > 0.05\000LLC miss rate\000\000100%\000\000\000\000001= */ +{ 128375 }, /* migrations_per_second\000Default\000software@cpu\\-migratio= ns\\,name\\=3Dcpu\\-migrations@ * 1e9 / (software@cpu\\-clock\\,name\\=3Dcp= u\\-clock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtask\\-cloc= k@)\000\000Process migrations to a new CPU per CPU second\000\0001migration= s/sec\000\000\000\000011 */ +{ 128635 }, /* page_faults_per_second\000Default\000software@page\\-faults= \\,name\\=3Dpage\\-faults@ * 1e9 / (software@cpu\\-clock\\,name\\=3Dcpu\\-c= lock@ if #target_cpu else software@task\\-clock\\,name\\=3Dtask\\-clock@)\0= 00\000Page faults per CPU second\000\0001faults/sec\000\000\000\000011 */ +{ 128979 }, /* stalled_cycles_per_instruction\000Default\000max(stalled\\-= cycles\\-frontend, stalled\\-cycles\\-backend) / instructions\000\000Max fr= ont or backend stalls per instruction\000\000\000\000\000\000001 */ =20 }; =20 @@ -2650,29 +2654,29 @@ static const struct pmu_table_entry pmu_metrics__co= mmon[] =3D { }; =20 static const struct compact_pmu_event pmu_events__test_soc_cpu_default_cor= e[] =3D { -{ 126205 }, /* bp_l1_btb_correct\000branch\000L1 BTB Correction\000event= =3D0x8a\000\00000\000\000\000\000\000 */ -{ 126267 }, /* bp_l2_btb_correct\000branch\000L2 BTB Correction\000event= =3D0x8b\000\00000\000\000\000\000\000 */ -{ 126529 }, /* dispatch_blocked.any\000other\000Memory cluster signals to = block micro-op dispatch for any reason\000event=3D9,period=3D200000,umask= =3D0x20\000\00000\000\000\000\000\000 */ -{ 126662 }, /* eist_trans\000other\000Number of Enhanced Intel SpeedStep(R= ) Technology (EIST) transitions\000event=3D0x3a,period=3D200000\000\00000\0= 00\000\000\000\000 */ -{ 126329 }, /* l3_cache_rd\000cache\000L3 cache access, read\000event=3D0x= 40\000\00000\000\000\000\000Attributable Level 3 cache access, read\000 */ -{ 126427 }, /* segment_reg_loads.any\000other\000Number of segment registe= r loads\000event=3D6,period=3D200000,umask=3D0x80\000\00000\000\000\000\000= \000 */ +{ 126403 }, /* bp_l1_btb_correct\000branch\000L1 BTB Correction\000event= =3D0x8a\000\00000\000\000\000\000\000 */ +{ 126465 }, /* bp_l2_btb_correct\000branch\000L2 BTB Correction\000event= =3D0x8b\000\00000\000\000\000\000\000 */ +{ 126727 }, /* dispatch_blocked.any\000other\000Memory cluster signals to = block micro-op dispatch for any reason\000event=3D9,period=3D200000,umask= =3D0x20\000\00000\000\000\000\000\000 */ +{ 126860 }, /* eist_trans\000other\000Number of Enhanced Intel SpeedStep(R= ) Technology (EIST) transitions\000event=3D0x3a,period=3D200000\000\00000\0= 00\000\000\000\000 */ +{ 126527 }, /* l3_cache_rd\000cache\000L3 cache access, read\000event=3D0x= 40\000\00000\000\000\000\000Attributable Level 3 cache access, read\000 */ +{ 126625 }, /* segment_reg_loads.any\000other\000Number of segment registe= r loads\000event=3D6,period=3D200000,umask=3D0x80\000\00000\000\000\000\000= \000 */ }; static const struct compact_pmu_event pmu_events__test_soc_cpu_hisi_sccl_d= drc[] =3D { -{ 126795 }, /* uncore_hisi_ddrc.flux_wcmd\000uncore\000DDRC write commands= \000event=3D2\000\00000\000\000\000\000\000 */ +{ 126993 }, /* uncore_hisi_ddrc.flux_wcmd\000uncore\000DDRC write commands= \000event=3D2\000\00000\000\000\000\000\000 */ }; static const struct compact_pmu_event pmu_events__test_soc_cpu_hisi_sccl_l= 3c[] =3D { -{ 127157 }, /* uncore_hisi_l3c.rd_hit_cpipe\000uncore\000Total read hits\0= 00event=3D7\000\00000\000\000\000\000\000 */ +{ 127355 }, /* uncore_hisi_l3c.rd_hit_cpipe\000uncore\000Total read hits\0= 00event=3D7\000\00000\000\000\000\000\000 */ }; static const struct compact_pmu_event pmu_events__test_soc_cpu_uncore_cbox= [] =3D { -{ 127031 }, /* event-hyphen\000uncore\000UNC_CBO_HYPHEN\000event=3D0xe0\00= 0\00000\000\000\000\000\000 */ -{ 127085 }, /* event-two-hyph\000uncore\000UNC_CBO_TWO_HYPH\000event=3D0xc= 0\000\00000\000\000\000\000\000 */ -{ 126877 }, /* unc_cbo_xsnp_response.miss_eviction\000uncore\000A cross-co= re snoop resulted from L3 Eviction which misses in some processor core\000e= vent=3D0x22,umask=3D0x81\000\00000\000\000\000\000\000 */ +{ 127229 }, /* event-hyphen\000uncore\000UNC_CBO_HYPHEN\000event=3D0xe0\00= 0\00000\000\000\000\000\000 */ +{ 127283 }, /* event-two-hyph\000uncore\000UNC_CBO_TWO_HYPH\000event=3D0xc= 0\000\00000\000\000\000\000\000 */ +{ 127075 }, /* unc_cbo_xsnp_response.miss_eviction\000uncore\000A cross-co= re snoop resulted from L3 Eviction which misses in some processor core\000e= vent=3D0x22,umask=3D0x81\000\00000\000\000\000\000\000 */ }; static const struct compact_pmu_event pmu_events__test_soc_cpu_uncore_imc[= ] =3D { -{ 127340 }, /* uncore_imc.cache_hits\000uncore\000Total cache hits\000even= t=3D0x34\000\00000\000\000\000\000\000 */ +{ 127538 }, /* uncore_imc.cache_hits\000uncore\000Total cache hits\000even= t=3D0x34\000\00000\000\000\000\000\000 */ }; static const struct compact_pmu_event pmu_events__test_soc_cpu_uncore_imc_= free_running[] =3D { -{ 127249 }, /* uncore_imc_free_running.cache_miss\000uncore\000Total cache= misses\000event=3D0x12\000\00000\000\000\000\000\000 */ +{ 127447 }, /* uncore_imc_free_running.cache_miss\000uncore\000Total cache= misses\000event=3D0x12\000\00000\000\000\000\000\000 */ =20 }; =20 @@ -2685,46 +2689,46 @@ static const struct pmu_table_entry pmu_events__tes= t_soc_cpu[] =3D { { .entries =3D pmu_events__test_soc_cpu_hisi_sccl_ddrc, .num_entries =3D ARRAY_SIZE(pmu_events__test_soc_cpu_hisi_sccl_ddrc), - .pmu_name =3D { 126780 /* hisi_sccl,ddrc\000 */ }, + .pmu_name =3D { 126978 /* hisi_sccl,ddrc\000 */ }, }, { .entries =3D pmu_events__test_soc_cpu_hisi_sccl_l3c, .num_entries =3D ARRAY_SIZE(pmu_events__test_soc_cpu_hisi_sccl_l3c), - .pmu_name =3D { 127143 /* hisi_sccl,l3c\000 */ }, + .pmu_name =3D { 127341 /* hisi_sccl,l3c\000 */ }, }, { .entries =3D pmu_events__test_soc_cpu_uncore_cbox, .num_entries =3D ARRAY_SIZE(pmu_events__test_soc_cpu_uncore_cbox), - .pmu_name =3D { 126865 /* uncore_cbox\000 */ }, + .pmu_name =3D { 127063 /* uncore_cbox\000 */ }, }, { .entries =3D pmu_events__test_soc_cpu_uncore_imc, .num_entries =3D ARRAY_SIZE(pmu_events__test_soc_cpu_uncore_imc), - .pmu_name =3D { 127329 /* uncore_imc\000 */ }, + .pmu_name =3D { 127527 /* uncore_imc\000 */ }, }, { .entries =3D pmu_events__test_soc_cpu_uncore_imc_free_running, .num_entries =3D ARRAY_SIZE(pmu_events__test_soc_cpu_uncore_imc_free_= running), - .pmu_name =3D { 127225 /* uncore_imc_free_running\000 */ }, + .pmu_name =3D { 127423 /* uncore_imc_free_running\000 */ }, }, }; =20 static const struct compact_pmu_event pmu_metrics__test_soc_cpu_default_co= re[] =3D { -{ 130353 }, /* CPI\000\0001 / IPC\000\000\000\000\000\000\000\000000 */ -{ 131042 }, /* DCache_L2_All\000\000DCache_L2_All_Hits + DCache_L2_All_Mis= s\000\000\000\000\000\000\000\000000 */ -{ 130812 }, /* DCache_L2_All_Hits\000\000l2_rqsts.demand_data_rd_hit + l2_= rqsts.pf_hit + l2_rqsts.rfo_hit\000\000\000\000\000\000\000\000000 */ -{ 130907 }, /* DCache_L2_All_Miss\000\000max(l2_rqsts.all_demand_data_rd -= l2_rqsts.demand_data_rd_hit, 0) + l2_rqsts.pf_miss + l2_rqsts.rfo_miss\000= \000\000\000\000\000\000\000000 */ -{ 131107 }, /* DCache_L2_Hits\000\000d_ratio(DCache_L2_All_Hits, DCache_L2= _All)\000\000\000\000\000\000\000\000000 */ -{ 131176 }, /* DCache_L2_Misses\000\000d_ratio(DCache_L2_All_Miss, DCache_= L2_All)\000\000\000\000\000\000\000\000000 */ -{ 130440 }, /* Frontend_Bound_SMT\000\000idq_uops_not_delivered.core / (4 = * (cpu_clk_unhalted.thread / 2 * (1 + cpu_clk_unhalted.one_thread_active / = cpu_clk_unhalted.ref_xclk)))\000\000\000\000\000\000\000\000000 */ -{ 130376 }, /* IPC\000group1\000inst_retired.any / cpu_clk_unhalted.thread= \000\000\000\000\000\000\000\000000 */ -{ 131314 }, /* L1D_Cache_Fill_BW\000\00064 * l1d.replacement / 1e9 / durat= ion_time\000\000\000\000\000\000\000\000000 */ -{ 131247 }, /* M1\000\000ipc + M2\000\000\000\000\000\000\000\000000 */ -{ 131270 }, /* M2\000\000ipc + M1\000\000\000\000\000\000\000\000000 */ -{ 131293 }, /* M3\000\0001 / M3\000\000\000\000\000\000\000\000000 */ -{ 130740 }, /* cache_miss_cycles\000group1\000dcache_miss_cpi + icache_mis= s_cycles\000\000\000\000\000\000\000\000000 */ -{ 130607 }, /* dcache_miss_cpi\000\000l1d\\-loads\\-misses / inst_retired.= any\000\000\000\000\000\000\000\000000 */ -{ 130672 }, /* icache_miss_cycles\000\000l1i\\-loads\\-misses / inst_retir= ed.any\000\000\000\000\000\000\000\000000 */ +{ 130551 }, /* CPI\000\0001 / IPC\000\000\000\000\000\000\000\000000 */ +{ 131240 }, /* DCache_L2_All\000\000DCache_L2_All_Hits + DCache_L2_All_Mis= s\000\000\000\000\000\000\000\000000 */ +{ 131010 }, /* DCache_L2_All_Hits\000\000l2_rqsts.demand_data_rd_hit + l2_= rqsts.pf_hit + l2_rqsts.rfo_hit\000\000\000\000\000\000\000\000000 */ +{ 131105 }, /* DCache_L2_All_Miss\000\000max(l2_rqsts.all_demand_data_rd -= l2_rqsts.demand_data_rd_hit, 0) + l2_rqsts.pf_miss + l2_rqsts.rfo_miss\000= \000\000\000\000\000\000\000000 */ +{ 131305 }, /* DCache_L2_Hits\000\000d_ratio(DCache_L2_All_Hits, DCache_L2= _All)\000\000\000\000\000\000\000\000000 */ +{ 131374 }, /* DCache_L2_Misses\000\000d_ratio(DCache_L2_All_Miss, DCache_= L2_All)\000\000\000\000\000\000\000\000000 */ +{ 130638 }, /* Frontend_Bound_SMT\000\000idq_uops_not_delivered.core / (4 = * (cpu_clk_unhalted.thread / 2 * (1 + cpu_clk_unhalted.one_thread_active / = cpu_clk_unhalted.ref_xclk)))\000\000\000\000\000\000\000\000000 */ +{ 130574 }, /* IPC\000group1\000inst_retired.any / cpu_clk_unhalted.thread= \000\000\000\000\000\000\000\000000 */ +{ 131512 }, /* L1D_Cache_Fill_BW\000\00064 * l1d.replacement / 1e9 / durat= ion_time\000\000\000\000\000\000\000\000000 */ +{ 131445 }, /* M1\000\000ipc + M2\000\000\000\000\000\000\000\000000 */ +{ 131468 }, /* M2\000\000ipc + M1\000\000\000\000\000\000\000\000000 */ +{ 131491 }, /* M3\000\0001 / M3\000\000\000\000\000\000\000\000000 */ +{ 130938 }, /* cache_miss_cycles\000group1\000dcache_miss_cpi + icache_mis= s_cycles\000\000\000\000\000\000\000\000000 */ +{ 130805 }, /* dcache_miss_cpi\000\000l1d\\-loads\\-misses / inst_retired.= any\000\000\000\000\000\000\000\000000 */ +{ 130870 }, /* icache_miss_cycles\000\000l1i\\-loads\\-misses / inst_retir= ed.any\000\000\000\000\000\000\000\000000 */ =20 }; =20 @@ -2737,13 +2741,13 @@ static const struct pmu_table_entry pmu_metrics__te= st_soc_cpu[] =3D { }; =20 static const struct compact_pmu_event pmu_events__test_soc_sys_uncore_sys_= ccn_pmu[] =3D { -{ 127519 }, /* sys_ccn_pmu.read_cycles\000uncore\000ccn read-cycles event\= 000config=3D0x2c\0000x01\00000\000\000\000\000\000 */ +{ 127717 }, /* sys_ccn_pmu.read_cycles\000uncore\000ccn read-cycles event\= 000config=3D0x2c\0000x01\00000\000\000\000\000\000 */ }; static const struct compact_pmu_event pmu_events__test_soc_sys_uncore_sys_= cmn_pmu[] =3D { -{ 127615 }, /* sys_cmn_pmu.hnf_cache_miss\000uncore\000Counts total cache = misses in first lookup result (high priority)\000eventid=3D1,type=3D5\000(4= 34|436|43c|43a).*\00000\000\000\000\000\000 */ +{ 127813 }, /* sys_cmn_pmu.hnf_cache_miss\000uncore\000Counts total cache = misses in first lookup result (high priority)\000eventid=3D1,type=3D5\000(4= 34|436|43c|43a).*\00000\000\000\000\000\000 */ }; static const struct compact_pmu_event pmu_events__test_soc_sys_uncore_sys_= ddr_pmu[] =3D { -{ 127424 }, /* sys_ddr_pmu.write_cycles\000uncore\000ddr write-cycles even= t\000event=3D0x2b\000v8\00000\000\000\000\000\000 */ +{ 127622 }, /* sys_ddr_pmu.write_cycles\000uncore\000ddr write-cycles even= t\000event=3D0x2b\000v8\00000\000\000\000\000\000 */ =20 }; =20 @@ -2751,17 +2755,17 @@ static const struct pmu_table_entry pmu_events__tes= t_soc_sys[] =3D { { .entries =3D pmu_events__test_soc_sys_uncore_sys_ccn_pmu, .num_entries =3D ARRAY_SIZE(pmu_events__test_soc_sys_uncore_sys_ccn_p= mu), - .pmu_name =3D { 127500 /* uncore_sys_ccn_pmu\000 */ }, + .pmu_name =3D { 127698 /* uncore_sys_ccn_pmu\000 */ }, }, { .entries =3D pmu_events__test_soc_sys_uncore_sys_cmn_pmu, .num_entries =3D ARRAY_SIZE(pmu_events__test_soc_sys_uncore_sys_cmn_p= mu), - .pmu_name =3D { 127596 /* uncore_sys_cmn_pmu\000 */ }, + .pmu_name =3D { 127794 /* uncore_sys_cmn_pmu\000 */ }, }, { .entries =3D pmu_events__test_soc_sys_uncore_sys_ddr_pmu, .num_entries =3D ARRAY_SIZE(pmu_events__test_soc_sys_uncore_sys_ddr_p= mu), - .pmu_name =3D { 127405 /* uncore_sys_ddr_pmu\000 */ }, + .pmu_name =3D { 127603 /* uncore_sys_ddr_pmu\000 */ }, }, }; =20 diff --git a/tools/perf/util/expr.c b/tools/perf/util/expr.c index 4df56f2b283d..465fe2e9bbbe 100644 --- a/tools/perf/util/expr.c +++ b/tools/perf/util/expr.c @@ -401,17 +401,12 @@ double expr__get_literal(const char *literal, const s= truct expr_scanner_ctx *ctx if (ev !=3D TOOL_PMU__EVENT_NONE) { u64 count; =20 - if (tool_pmu__read_event(ev, /*evsel=3D*/NULL, &count)) + if (tool_pmu__read_event(ev, /*evsel=3D*/NULL, + ctx->system_wide, ctx->user_requested_cpu_list, + &count)) result =3D count; else pr_err("Failure to read '%s'", literal); - - } else if (!strcmp("#core_wide", literal)) { - result =3D core_wide(ctx->system_wide, ctx->user_requested_cpu_list) - ? 1.0 : 0.0; - } else if (!strcmp("#target_cpu", literal)) { - result =3D (ctx->system_wide || ctx->user_requested_cpu_list) - ? 1.0 : 0.0; } else { pr_err("Unrecognized literal '%s'", literal); } diff --git a/tools/perf/util/stat-shadow.c b/tools/perf/util/stat-shadow.c index c1547128c396..b3b482e1808f 100644 --- a/tools/perf/util/stat-shadow.c +++ b/tools/perf/util/stat-shadow.c @@ -72,6 +72,8 @@ static int prepare_metric(const struct metric_expr *mexp, case TOOL_PMU__EVENT_SLOTS: case TOOL_PMU__EVENT_SMT_ON: case TOOL_PMU__EVENT_SYSTEM_TSC_FREQ: + case TOOL_PMU__EVENT_CORE_WIDE: + case TOOL_PMU__EVENT_TARGET_CPU: default: pr_err("Unexpected tool event '%s'", evsel__name(metric_events[i])); abort(); diff --git a/tools/perf/util/tool_pmu.c b/tools/perf/util/tool_pmu.c index f075098488ba..a72c665ee644 100644 --- a/tools/perf/util/tool_pmu.c +++ b/tools/perf/util/tool_pmu.c @@ -6,6 +6,7 @@ #include "pmu.h" #include "print-events.h" #include "smt.h" +#include "stat.h" #include "time-utils.h" #include "tool_pmu.h" #include "tsc.h" @@ -30,6 +31,8 @@ static const char *const tool_pmu__event_names[TOOL_PMU__= EVENT_MAX] =3D { "slots", "smt_on", "system_tsc_freq", + "core_wide", + "target_cpu", }; =20 bool tool_pmu__skip_event(const char *name __maybe_unused) @@ -329,7 +332,11 @@ static bool has_pmem(void) return has_pmem; } =20 -bool tool_pmu__read_event(enum tool_pmu_event ev, struct evsel *evsel, u64= *result) +bool tool_pmu__read_event(enum tool_pmu_event ev, + struct evsel *evsel, + bool system_wide, + const char *user_requested_cpu_list, + u64 *result) { const struct cpu_topology *topology; =20 @@ -421,6 +428,14 @@ bool tool_pmu__read_event(enum tool_pmu_event ev, stru= ct evsel *evsel, u64 *resu *result =3D arch_get_tsc_freq(); return true; =20 + case TOOL_PMU__EVENT_CORE_WIDE: + *result =3D core_wide(system_wide, user_requested_cpu_list) ? 1 : 0; + return true; + + case TOOL_PMU__EVENT_TARGET_CPU: + *result =3D system_wide || (user_requested_cpu_list !=3D NULL) ? 1 : 0; + return true; + case TOOL_PMU__EVENT_NONE: case TOOL_PMU__EVENT_DURATION_TIME: case TOOL_PMU__EVENT_USER_TIME: @@ -452,11 +467,16 @@ int evsel__tool_pmu_read(struct evsel *evsel, int cpu= _map_idx, int thread) case TOOL_PMU__EVENT_SLOTS: case TOOL_PMU__EVENT_SMT_ON: case TOOL_PMU__EVENT_SYSTEM_TSC_FREQ: + case TOOL_PMU__EVENT_CORE_WIDE: + case TOOL_PMU__EVENT_TARGET_CPU: if (evsel->prev_raw_counts) old_count =3D perf_counts(evsel->prev_raw_counts, cpu_map_idx, thread); val =3D 0; if (cpu_map_idx =3D=3D 0 && thread =3D=3D 0) { - if (!tool_pmu__read_event(ev, evsel, &val)) { + if (!tool_pmu__read_event(ev, evsel, + stat_config.system_wide, + stat_config.user_requested_cpu_list, + &val)) { count->lost++; val =3D 0; } diff --git a/tools/perf/util/tool_pmu.h b/tools/perf/util/tool_pmu.h index d642e7d73910..f1714001bc1d 100644 --- a/tools/perf/util/tool_pmu.h +++ b/tools/perf/util/tool_pmu.h @@ -22,6 +22,8 @@ enum tool_pmu_event { TOOL_PMU__EVENT_SLOTS, TOOL_PMU__EVENT_SMT_ON, TOOL_PMU__EVENT_SYSTEM_TSC_FREQ, + TOOL_PMU__EVENT_CORE_WIDE, + TOOL_PMU__EVENT_TARGET_CPU, =20 TOOL_PMU__EVENT_MAX, }; @@ -34,7 +36,12 @@ enum tool_pmu_event tool_pmu__str_to_event(const char *s= tr); bool tool_pmu__skip_event(const char *name); int tool_pmu__num_skip_events(void); =20 -bool tool_pmu__read_event(enum tool_pmu_event ev, struct evsel *evsel, u64= *result); +bool tool_pmu__read_event(enum tool_pmu_event ev, + struct evsel *evsel, + bool system_wide, + const char *user_requested_cpu_list, + u64 *result); + =20 u64 tool_pmu__cpu_slots_per_cycle(void); =20 --=20 2.51.2.1041.gc1ab5b90ca-goog