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 8EF1219C556 for ; Wed, 30 Jul 2025 22:08:26 +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=1753913308; cv=none; b=jF5i+v+4x+76E9lsvmLS2dRcNec78nSg5EjcOrPD784xI0EosgQz/rsii8ql5fOCdBa9/5LOvqXBiZnN7VGg5KVvYAmavXUFbTNmHu9huWQUQS+GBKQN/jmrSjG6WNNrdy5//NlUb7xlK0B+JEvc5g9ll/4IBdFwJggMTvV8oTo= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1753913308; c=relaxed/simple; bh=PNDe9DmLpP8vQLE52QVnejDJ8eVQpGsQIZ1TWehA3sg=; h=Date:Mime-Version:Message-ID:Subject:From:To:Cc:Content-Type; b=oue5E+nptC8tu1owrxfn+TdldU8qgrUV0YObnMe0VuPNfvO83v932Lw5wun5+53SyqBIvi0EtDqrxgL3lYHDx6Y8BJ+kFoT2P4jOguxRAySr9StlCQRLfQSGPR27DTPc4txMtKj0UhE+hSN58me80AC2U7miJlj2wUTfT2E/LdQ= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dmarc=pass (p=reject dis=none) header.from=google.com; spf=pass smtp.mailfrom=flex--pmalani.bounces.google.com; dkim=pass (2048-bit key) header.d=google.com header.i=@google.com header.b=IvPvF4ep; 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--pmalani.bounces.google.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=google.com header.i=@google.com header.b="IvPvF4ep" Received: by mail-pg1-f202.google.com with SMTP id 41be03b00d2f7-b3f38d3cabeso174714a12.3 for ; Wed, 30 Jul 2025 15:08:26 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1753913306; x=1754518106; darn=vger.kernel.org; h=cc:to:from:subject:message-id:mime-version:date:from:to:cc:subject :date:message-id:reply-to; bh=HYPSzgAZClucp9advPGtHTtoMvoPpzZ3Dtwk5mdPmjs=; b=IvPvF4epxARvaMU2dcOHI7Sq+MA37Grm6Ef7xZt6iPi8BBQ7fpm9+Q16UMw4fV8qOt Share2RzPf9VvT51LUiA/sPra+vgldUvw12bpm17SIuUPokp9kvFoWbHXpT0vl6FPs4c iyMS6m3++ARyE5gOAQ1ayCkSEmH6kyV9ZQxpx5xIAe5KclaoUDlFRcwW7Us8RuLy8rSx Fq6kh+33E7s/2mhLT4DpJmvCq+eoUJX9A4cZIfcSKV0kbBxzJSmmrFoov8/1llp9FN5/ RLTM9SNdECGKost5cm3+I2w07iH6eoTNdrbHKrNxwpPCgttOSv9dxemSZ6mdm1HgKEnw 0vNA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1753913306; x=1754518106; h=cc:to:from:subject:message-id:mime-version:date:x-gm-message-state :from:to:cc:subject:date:message-id:reply-to; bh=HYPSzgAZClucp9advPGtHTtoMvoPpzZ3Dtwk5mdPmjs=; b=JYrSz4bdzMuR5uO/sfKlemGJAN2lAgP5dtU9oGfPwmEUzSWC4GM3KJbNk+xzbojfJl 44wsbLJ6ktym7SC1B++xLz8TRasKsY+wuYBiLp7teilD6dSplIZtfqz7/r0xbg+R0qGk Mx0UY1K33pglhrpK5LszQIRt10naWOEjheLxtMaEun7mnPwV1tuauOsR+7di5DN4KcjB wGHKg+Qw6hfhrk+72LqqZjStPp7VXlrKYikE6xH/XEs+L6yEE/nu0xG88MZtU8FQDlto VjKe2dbYiRzTNdMXrBUArF7a+2Dce+0rkXV/DhteGa+/DhVS+GBY+ONP6sBrpbFSRtmq KRgQ== X-Forwarded-Encrypted: i=1; AJvYcCX+Q5oyBab+JU8lEOZFVCNmISBRmscbMw7DyZp2h0ngJaPdXu2/q2zvBF65WuSmRcJ1IU+uNkUeXA==@vger.kernel.org X-Gm-Message-State: AOJu0YwCvhsLZZykSlbvTmFP1IZhcxH3gCfBiF7uWsZK5vaffyzyp5um GD25iT2vl7u0Kw938uo8hiOaaGkCj2zfNG9EPokSQQiGZWQJuKfVsHZbxSMDJkL9FakyRB87b9J tkGldnwGdYw== X-Google-Smtp-Source: AGHT+IGZ3NDO6jUkDRlidOJBUV0Ja1bR3uiB0LRR/EOuhSKjhT4aH0bvvbHvHtK+qePEip623lThmruu6pu6 X-Received: from pjbnt5.prod.google.com ([2002:a17:90b:2485:b0:2fb:fa85:1678]) (user=pmalani job=prod-delivery.src-stubby-dispatcher) by 2002:a17:90b:3907:b0:311:fde5:c4be with SMTP id 98e67ed59e1d1-31f5ea6b6bbmr6169503a91.35.1753913305877; Wed, 30 Jul 2025 15:08:25 -0700 (PDT) Date: Wed, 30 Jul 2025 22:07:58 +0000 Precedence: bulk X-Mailing-List: linux-pm@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 X-Mailer: git-send-email 2.50.1.552.g942d659e1b-goog Message-ID: <20250730220812.53098-1-pmalani@google.com> Subject: [PATCH] cpufreq: CPPC: Increase delay between perf counter reads From: Prashant Malani To: open list , "open list:CPU FREQUENCY SCALING FRAMEWORK" , "Rafael J. Wysocki" , Viresh Kumar Cc: Prashant Malani , Yang Shi , Ionela Voinescu Content-Type: text/plain; charset="UTF-8" On a heavily loaded CPU, performance counter reads can be erratic. This is due to two factors: - The method used to calculate CPPC delivered performance. - Linux scheduler vagaries. As an example, on a CPU which has a max frequency of 3.4 GHz, if we run stress-ng on the CPU in the background and then read the frequency, we get invalid readings: ./stress_ng --cpu 108 --taskset 3 -t 30s & cat /sys/devices/system/cpu/cpu3/cpufreq/cpuinfo_cur_freq 3600000 3500000 3600000 Per [1] CPPC performance is measured by reading the delivered and reference counters at timestamp t0, then waiting 2us, and then repeating the measurement at t1. So, in theory, one should end up with: Timestamp t0: ref0, del0 Timestamp t1: ref1, del1 However, since the reference and delivered registers are individual register reads (in the case of FFH, it even results in an IPI to the CPU in question), what happens in practice is: Timestamp t0: del0 Timestamp t0 + m: ref0 Timestamp t1: del1 Timestamp t1 + n: ref1 There has been prior discussion[2] about the cause of these differences; it was broadly pegged as due to IRQs and "interconnect congestion". Since the gap between t0 and t1 is very small (2us), differing values of m and n mean that the measurements don't correspond to 2 discrete timestamps, since the delivered performance delta is being measured across a significantly different time period than the reference performance delta. This has an influence on the perf measurement which is: ((del1 - del0) * reference perf) / (ref1 - ref0) Previously collected data[4] shows that cppc_get_perf_ctrs() itself takes anywhere between 4.9us and 3.6us, which further suggests that a 2us delta is too less. If we increase the time delta to a high enough value (i.e if delay >> m,n) then the effects of m and n get mitigated, leading to both the register measurements (ref and del) corresponding to the same timestamp. When this approach was previously proposed[3], there was concern about this function being called with interrupts off but that was later found to be not true [2]. So, waiting for a slightly longer time in between counter samples should be acceptable. Increase the time delay between counter reads to 200 us to reduce the effect of timing discrepancies in reading individual performance registers. [1] https://docs.kernel.org/admin-guide/acpi/cppc_sysfs.html#computing-average-delivered-performance [2] https://lore.kernel.org/all/7b57e680-0ba3-0b8b-851e-7cc369050386@os.amperecomputing.com/ [3] https://lore.kernel.org/all/20230328193846.8757-1-yang@os.amperecomputing.com/ [4] https://lore.kernel.org/all/1ce09fd7-0c1d-fc46-ce12-01b25fbd4afd@os.amperecomputing.com/ Cc: Yang Shi Cc: Ionela Voinescu Signed-off-by: Prashant Malani --- drivers/cpufreq/cppc_cpufreq.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/cpufreq/cppc_cpufreq.c b/drivers/cpufreq/cppc_cpufreq.c index 4a17162a392d..086c3b87bd4e 100644 --- a/drivers/cpufreq/cppc_cpufreq.c +++ b/drivers/cpufreq/cppc_cpufreq.c @@ -718,7 +718,7 @@ static int cppc_get_perf_ctrs_sample(int cpu, if (ret) return ret; - udelay(2); /* 2usec delay between sampling */ + udelay(200); /* 200usec delay between sampling */ return cppc_get_perf_ctrs(cpu, fb_ctrs_t1); } -- 2.50.1.552.g942d659e1b-goog