From nobody Thu Dec 11 19:22:03 2025 Received: from mail-pf1-f201.google.com (mail-pf1-f201.google.com [209.85.210.201]) (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 E45FB2E6CC8 for ; Tue, 2 Dec 2025 17:54:31 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.210.201 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1764698074; cv=none; b=Shq90xX6NPHYoiX2pG7EswjptcyWVWsSzJi0FArb3FddB7CaY41ejWjKxbQKopVbepdRT2b95daCT/jzOM3EVm2ExloKyy/zxCeMJqk33FFx9iIEOKFtxwEHrHwCuWpH5zAqEIFfHc+YtpDtO7ypPs/A1qb+aiC2o9awHSMq1aE= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1764698074; c=relaxed/simple; bh=Hv4UhU6NXENZa+Z1+6L4rRfnL/6waxqGPRgqBzUWSE4=; h=Date:In-Reply-To:Mime-Version:References:Message-ID:Subject:From: To:Content-Type; b=RSt23QqU1no2F8v284lkATPysRArDn++PSIXKExRMDxo4veDBUCJtSn2ObQc7aWrDwrokFZde7cVxzjK9Nnn3R/rsCe2oUPokJ6O/X+N69kX36FxOxBL1cu8kjKvqX3k5knoY+CmISNfvDbJtjmtj/orbZSStFDZlD7KHVIAM7g= 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=LaobBi28; arc=none smtp.client-ip=209.85.210.201 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="LaobBi28" Received: by mail-pf1-f201.google.com with SMTP id d2e1a72fcca58-7b8a12f0cb4so5404186b3a.3 for ; Tue, 02 Dec 2025 09:54:31 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1764698071; x=1765302871; 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=j6uZvpU1CmHPVZiDBD5+c8ap8dhyWb5RPEdJMn1xJn8=; b=LaobBi288U0kEFwCWHvSgASAed02S19dJy0wL+kIvsoNwejy+7Nb7rrJYRvSlESBhT lA7HL7eCHPnBI/SEWXfVtxKsnvNNKNuAidvoBYBgnxcOkGaE0UnVYlNqGseIi6dCbD/D 4iduR2VRTxNr7l4rdjmKvojZQWD4qaFUsS7h/bCQAkXIxTrTsfzBfqFFBAbBMC9/546o RTYrea4L/iC9arz5n04VxhvGLDhJhIMVI7Sc1QvSyD7gjDpg0WITwI7RJr26OYgOzrhU 1KQqY8mepEtg0fC7bCrJOLexTNpirlpDu6lgqb30JuUteNso22GwTEh0M42Klu8s8e4k wGKQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1764698071; x=1765302871; 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=j6uZvpU1CmHPVZiDBD5+c8ap8dhyWb5RPEdJMn1xJn8=; b=F043IGZq6gmMHICuzix9rd2vvhIUo++rDOuKyUt3Yl4KvKPukDV0MPaGgCx4yJurdW scDKlAyvS2Ao+6aB1ArZhZTQoaNUwfG2oYfiNEl1az321WOyd0urMwbUnRWDZrR7KaUb jzT3v9BABocFDp0R96soqBLIEHSkGOi3ab8hKLLvejSN3IhwAZpvSG/a+sL2exD4s755 tjvB5uLAfULHpkHnIqrdG/YzW3y3f09U5J5TnYpJ3Fx3k4X1MMhq/Dg1s+C8WVRacyM6 nd7jMsjMuaiihkJOCSTUfXJI9NQNnEf0RtZ6HIQJAF893mzFqJPBPD+8YviW7KeaVVrt 29/w== X-Forwarded-Encrypted: i=1; AJvYcCXAaoBRW+c4fii4NHKt7bXmTGD4bbvpqyeWa/4NWhSZPxTm/btGzknt0TZnUWHi7kPhMDzyQw/shmew/oA=@vger.kernel.org X-Gm-Message-State: AOJu0YyVGUssC9386E/alipCpt9+vEh5/4WNuQcKz+lpLzgmx9kA3x/J 2PZ1EyXmE6T6hc2zFHckRxK+Z+ZPu5Mp3wvf7jvvNc5JEyOdVAlM3WOJuAmy5zFSkNNC4lals9H 7/wUgNjcHgQ== X-Google-Smtp-Source: AGHT+IEQS4yH/MnzCpTnSvSxrkfRl/QtAgjK7RdA6AiWk79tkC6mhXBKHY4vP+R4KKcoA6cLh4RkPsikxUlB X-Received: from dycqv11.prod.google.com ([2002:a05:7300:df4b:b0:2a4:63df:76a5]) (user=irogers job=prod-delivery.src-stubby-dispatcher) by 2002:a05:7022:660a:b0:11b:a73b:2327 with SMTP id a92af1059eb24-11cbba52d48mr20931064c88.30.1764698071166; Tue, 02 Dec 2025 09:54:31 -0800 (PST) Date: Tue, 2 Dec 2025 09:50:38 -0800 In-Reply-To: <20251202175043.623597-1-irogers@google.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 References: <20251202175043.623597-1-irogers@google.com> X-Mailer: git-send-email 2.52.0.158.g65b55ccf14-goog Message-ID: <20251202175043.623597-44-irogers@google.com> Subject: [PATCH v9 43/48] perf jevents: Add local/remote miss latency metrics for Intel From: Ian Rogers To: Adrian Hunter , Alexander Shishkin , Arnaldo Carvalho de Melo , Benjamin Gray , Caleb Biggers , Edward Baker , Ian Rogers , Ingo Molnar , James Clark , Jing Zhang , Jiri Olsa , John Garry , Leo Yan , Namhyung Kim , Perry Taylor , Peter Zijlstra , Samantha Alt , Sandipan Das , Thomas Falcon , Weilin Wang , Xu Yang , linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" Derive from CBOX/CHA occupancy and inserts the average latency as is provided in Intel's uncore performance monitoring reference. Signed-off-by: Ian Rogers Tested-by: Thomas Falcon --- tools/perf/pmu-events/intel_metrics.py | 70 ++++++++++++++++++++++++-- 1 file changed, 67 insertions(+), 3 deletions(-) diff --git a/tools/perf/pmu-events/intel_metrics.py b/tools/perf/pmu-events= /intel_metrics.py index 118fe0fc05a3..037f9b2ea1b6 100755 --- a/tools/perf/pmu-events/intel_metrics.py +++ b/tools/perf/pmu-events/intel_metrics.py @@ -6,9 +6,10 @@ import math import os import re from typing import Optional -from metric import (d_ratio, has_event, max, CheckPmu, Event, JsonEncodeMe= tric, - JsonEncodeMetricGroupDescriptions, Literal, LoadEvents, - Metric, MetricConstraint, MetricGroup, MetricRef, Sele= ct) +from metric import (d_ratio, has_event, max, source_count, CheckPmu, Event, + JsonEncodeMetric, JsonEncodeMetricGroupDescriptions, + Literal, LoadEvents, Metric, MetricConstraint, MetricG= roup, + MetricRef, Select) =20 # Global command line arguments. _args =3D None @@ -624,6 +625,68 @@ def IntelL2() -> Optional[MetricGroup]: ], description=3D"L2 data cache analysis") =20 =20 +def IntelMissLat() -> Optional[MetricGroup]: + try: + ticks =3D Event("UNC_CHA_CLOCKTICKS", "UNC_C_CLOCKTICKS") + data_rd_loc_occ =3D Event("UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_LOCAL= ", + "UNC_CHA_TOR_OCCUPANCY.IA_MISS", + "UNC_C_TOR_OCCUPANCY.MISS_LOCAL_OPCODE", + "UNC_C_TOR_OCCUPANCY.MISS_OPCODE") + data_rd_loc_ins =3D Event("UNC_CHA_TOR_INSERTS.IA_MISS_DRD_LOCAL", + "UNC_CHA_TOR_INSERTS.IA_MISS", + "UNC_C_TOR_INSERTS.MISS_LOCAL_OPCODE", + "UNC_C_TOR_INSERTS.MISS_OPCODE") + data_rd_rem_occ =3D Event("UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOT= E", + "UNC_CHA_TOR_OCCUPANCY.IA_MISS", + "UNC_C_TOR_OCCUPANCY.MISS_REMOTE_OPCODE", + "UNC_C_TOR_OCCUPANCY.NID_MISS_OPCODE") + data_rd_rem_ins =3D Event("UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE", + "UNC_CHA_TOR_INSERTS.IA_MISS", + "UNC_C_TOR_INSERTS.MISS_REMOTE_OPCODE", + "UNC_C_TOR_INSERTS.NID_MISS_OPCODE") + except: + return None + + if (data_rd_loc_occ.name =3D=3D "UNC_C_TOR_OCCUPANCY.MISS_LOCAL_OPCODE= " or + data_rd_loc_occ.name =3D=3D "UNC_C_TOR_OCCUPANCY.MISS_OPCODE"): + data_rd =3D 0x182 + for e in [data_rd_loc_occ, data_rd_loc_ins, data_rd_rem_occ, data_= rd_rem_ins]: + e.name +=3D f"/filter_opc=3D{hex(data_rd)}/" + elif data_rd_loc_occ.name =3D=3D "UNC_CHA_TOR_OCCUPANCY.IA_MISS": + # Demand Data Read - Full cache-line read requests from core for + # lines to be cached in S or E, typically for data + demand_data_rd =3D 0x202 + # LLC Prefetch Data - Uncore will first look up the line in the + # LLC; for a cache hit, the LRU will be updated, on a miss, the + # DRd will be initiated + llc_prefetch_data =3D 0x25a + local_filter =3D (f"/filter_opc0=3D{hex(demand_data_rd)}," + f"filter_opc1=3D{hex(llc_prefetch_data)}," + "filter_loc,filter_nm,filter_not_nm/") + remote_filter =3D (f"/filter_opc0=3D{hex(demand_data_rd)}," + f"filter_opc1=3D{hex(llc_prefetch_data)}," + "filter_rem,filter_nm,filter_not_nm/") + for e in [data_rd_loc_occ, data_rd_loc_ins]: + e.name +=3D local_filter + for e in [data_rd_rem_occ, data_rd_rem_ins]: + e.name +=3D remote_filter + else: + assert data_rd_loc_occ.name =3D=3D "UNC_CHA_TOR_OCCUPANCY.IA_MISS_= DRD_LOCAL", data_rd_loc_occ + + ticks_per_cha =3D ticks / source_count(data_rd_loc_ins) + loc_lat =3D interval_sec * 1e9 * data_rd_loc_occ / \ + (ticks_per_cha * data_rd_loc_ins) + ticks_per_cha =3D ticks / source_count(data_rd_rem_ins) + rem_lat =3D interval_sec * 1e9 * data_rd_rem_occ / \ + (ticks_per_cha * data_rd_rem_ins) + return MetricGroup("lpm_miss_lat", [ + Metric("lpm_miss_lat_loc", "Local to a socket miss latency in nano= seconds", + loc_lat, "ns"), + Metric("lpm_miss_lat_rem", "Remote to a socket miss latency in nan= oseconds", + rem_lat, "ns"), + ]) + + def IntelMlp() -> Optional[Metric]: try: l1d =3D Event("L1D_PEND_MISS.PENDING") @@ -1005,6 +1068,7 @@ def main() -> None: IntelIlp(), IntelL2(), IntelLdSt(), + IntelMissLat(), IntelMlp(), IntelPorts(), IntelSwpf(), --=20 2.52.0.158.g65b55ccf14-goog