From patchwork Fri Mar 8 21:20:45 2019 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Daniel Lezcano X-Patchwork-Id: 160002 Delivered-To: patch@linaro.org Received: by 2002:a02:5cc1:0:0:0:0:0 with SMTP id w62csp9204478jad; Fri, 8 Mar 2019 13:21:11 -0800 (PST) X-Google-Smtp-Source: APXvYqyfiN4rCprANp0oUEtsLE6kH1I9fYuU4j05s+Dv2n+X8htEa1SXKuHXvGlIDShw/vJbID4M X-Received: by 2002:a63:1a25:: with SMTP id a37mr19210396pga.428.1552080070932; Fri, 08 Mar 2019 13:21:10 -0800 (PST) ARC-Seal: i=1; a=rsa-sha256; t=1552080070; cv=none; d=google.com; s=arc-20160816; b=g8DLEr+vm9YMpA75RefZhPxpYgNPBhFQl3M4ZPZiClDFfoACjscsl09a/TCZGZ3OYC iyZ4QzQs4mXoUAjOTZwl52kRqHKE1qUVskcyiZZGiibBHrgRcDk24kSpDpcb+e4Ug7Aq Bb8UfFPDO2ET7QgDxI28xmYWqncsnlRI0ecViLYToxlr7sBXolhoZyYFda23fzw+jnmq neMBFAr83SJjNPhzSAKmVskrtFdleSj0+QExoRLs1/tcuncSPBDwXkf02DOSb6s/tpM+ az8WumD9JtVDF2aQKShagYi7qrBW/0a75Eehu54uzM1W95V7SiPAsSWrp8DvwsjC9G8t TvcA== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=list-id:precedence:sender:references:in-reply-to:message-id:date :subject:cc:to:from:dkim-signature; bh=aycm97Plm6qVrne/2MmodSPEzVygGxEgmJGgQsC6tWo=; b=HEASHSOnDwdG3smmpbIX1h6G4MAa3goQ08ffbh6dF5o8JxB3GOoWXzMtPBDA0fh6tf rZ11kQGlzl9rObcZr/hcsZ+NEBzg4jtj8c88RbHE2Dn4WYjIWl7sSoGnSLrlBJQp7Q9h DfeER5Px8XwrMR+usZY66++fmcm8BTQmPgqLt5gPylbGlqOtcVlQhwKQtAtTdpWBd3po DGR0IbdsZLJcWZCA07u2d6ECB8pUy4zx0LEAVxcAfWy34h8tCJBuJqg83VwnrX1WJd2P e9HLd4ssv4eQ9Z6CzR5ey5LAO+m456HlkejPeCoGlaZHPuTkZvhuS/vUYE043lYyZ/B1 OM7Q== ARC-Authentication-Results: i=1; mx.google.com; dkim=pass header.i=@linaro.org header.s=google header.b="QN/JjxLA"; spf=pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=linaro.org Return-Path: Received: from vger.kernel.org (vger.kernel.org. [209.132.180.67]) by mx.google.com with ESMTP id z20si7598214pgf.324.2019.03.08.13.21.10; Fri, 08 Mar 2019 13:21:10 -0800 (PST) Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67; Authentication-Results: mx.google.com; dkim=pass header.i=@linaro.org header.s=google header.b="QN/JjxLA"; spf=pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=linaro.org Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1726654AbfCHVVJ (ORCPT + 31 others); Fri, 8 Mar 2019 16:21:09 -0500 Received: from mail-wm1-f47.google.com ([209.85.128.47]:38805 "EHLO mail-wm1-f47.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1726551AbfCHVVI (ORCPT ); Fri, 8 Mar 2019 16:21:08 -0500 Received: by mail-wm1-f47.google.com with SMTP id a188so13458124wmf.3 for ; Fri, 08 Mar 2019 13:21:05 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linaro.org; s=google; h=from:to:cc:subject:date:message-id:in-reply-to:references; bh=aycm97Plm6qVrne/2MmodSPEzVygGxEgmJGgQsC6tWo=; b=QN/JjxLAA/59J87r1Yc9f6qVcjvgxNhIFISHPDN4YHNLet0YQl44MRnM4RruJC/1Cx PKhZ1sxd0b5FCQzTsDzq1zGGoyr3DO4GJPRrr2TpoTp5P0uNuAmH2XpZNj/l0W8mDrRq 5ODOSRluiplQgeS3xkaoToLKZ6Xc1aKdN3qZMlGj66LU+chAH/yUICLGy8aCqRVZm9J6 W1dThQlM3oy5lr2lbXwq/diqXn5OfzbpESp+BC32bMGo2GyE9w3uQzEmZm22aSS/ej7g YNMlePYE5TJu3kYgolmI/buJ0OnbG3xPUnc/itO4N1RMim9YasdOQpwsMSWLwKpialKs JyGQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:to:cc:subject:date:message-id:in-reply-to :references; bh=aycm97Plm6qVrne/2MmodSPEzVygGxEgmJGgQsC6tWo=; b=FnLUnx5SoF6uGVNx99TDfEsQ7d+OUoHFWYTkhiFZNeTxpzkNeQlp/4A5dOquzofgJy 0MrZUen3/yyDDo9r+J+yEPTI0bGGb7NZWtAf9uTpaq3CDNY6Om+fUuJ0ZMGxidb8f2if +lE+j8yr5ZsD5+BKhRBGU/XKr1U04gUmqMdDPur5tsMjiAzg0W46QwCK9PeWXkO8RxiW 0Qnby0FR/X75Sd5ar9/HWp2YOo+EBY4BjCRYbJG9yYA7VETBt05cJCaUqAxpfdRKTQIp DrnOUs+j37Jmq8z8Tf1K7xQYENZ7tpXD+Ou0c1Q6zvQsPJrpv1oJ+UCInN8geeM4Av1G 7B3Q== X-Gm-Message-State: APjAAAX+FuMG8nIlIRt8PbjC1zu4GUmlKEFcLQH8O7bG/xBGo/JG+bIv jM/p0SGwyDVbNH9nMpRcN1YXkw== X-Received: by 2002:a7b:c0d5:: with SMTP id s21mr11194049wmh.153.1552080064666; Fri, 08 Mar 2019 13:21:04 -0800 (PST) Received: from clegane.local (246.127.130.77.rev.sfr.net. [77.130.127.246]) by smtp.gmail.com with ESMTPSA id d2sm13568507wrq.94.2019.03.08.13.21.03 (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Fri, 08 Mar 2019 13:21:03 -0800 (PST) From: Daniel Lezcano To: rjw@rjwysocki.net, tglx@linutronix.de Cc: ulf.hansson@linaro.org, linux-pm@vger.kernel.org, linux-kernel@vger.kernel.org Subject: [PATCH 1/3] genirq/timings: Remove variance computation code Date: Fri, 8 Mar 2019 22:20:45 +0100 Message-Id: <20190308212047.28767-2-daniel.lezcano@linaro.org> X-Mailer: git-send-email 2.17.1 In-Reply-To: <20190308212047.28767-1-daniel.lezcano@linaro.org> References: <20190308212047.28767-1-daniel.lezcano@linaro.org> Sender: linux-kernel-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org The next patch will introduce another approach to compute the next interrupt based on the array suffixes derived algorithm. This one will replace the variance computation code. The patch review will be too complex if we change the code little by little, it is much simpler to remove the variance code and add in the next patch the suffix array code. Remove the variance code in timings.c Signed-off-by: Daniel Lezcano --- kernel/irq/timings.c | 252 +------------------------------------------ 1 file changed, 2 insertions(+), 250 deletions(-) -- 2.17.1 diff --git a/kernel/irq/timings.c b/kernel/irq/timings.c index 1e4cb63a5c82..3cde046a2bc8 100644 --- a/kernel/irq/timings.c +++ b/kernel/irq/timings.c @@ -8,7 +8,6 @@ #include #include #include -#include #include @@ -19,13 +18,7 @@ DEFINE_STATIC_KEY_FALSE(irq_timing_enabled); DEFINE_PER_CPU(struct irq_timings, irq_timings); struct irqt_stat { - u64 next_evt; - u64 last_ts; - u64 variance; - u32 avg; - u32 nr_samples; - int anomalies; - int valid; + u64 next_evt; }; static DEFINE_IDR(irqt_stats); @@ -40,184 +33,6 @@ void irq_timings_disable(void) static_branch_disable(&irq_timing_enabled); } -/** - * irqs_update - update the irq timing statistics with a new timestamp - * - * @irqs: an irqt_stat struct pointer - * @ts: the new timestamp - * - * The statistics are computed online, in other words, the code is - * designed to compute the statistics on a stream of values rather - * than doing multiple passes on the values to compute the average, - * then the variance. The integer division introduces a loss of - * precision but with an acceptable error margin regarding the results - * we would have with the double floating precision: we are dealing - * with nanosec, so big numbers, consequently the mantisse is - * negligeable, especially when converting the time in usec - * afterwards. - * - * The computation happens at idle time. When the CPU is not idle, the - * interrupts' timestamps are stored in the circular buffer, when the - * CPU goes idle and this routine is called, all the buffer's values - * are injected in the statistical model continuying to extend the - * statistics from the previous busy-idle cycle. - * - * The observations showed a device will trigger a burst of periodic - * interrupts followed by one or two peaks of longer time, for - * instance when a SD card device flushes its cache, then the periodic - * intervals occur again. A one second inactivity period resets the - * stats, that gives us the certitude the statistical values won't - * exceed 1x10^9, thus the computation won't overflow. - * - * Basically, the purpose of the algorithm is to watch the periodic - * interrupts and eliminate the peaks. - * - * An interrupt is considered periodically stable if the interval of - * its occurences follow the normal distribution, thus the values - * comply with: - * - * avg - 3 x stddev < value < avg + 3 x stddev - * - * Which can be simplified to: - * - * -3 x stddev < value - avg < 3 x stddev - * - * abs(value - avg) < 3 x stddev - * - * In order to save a costly square root computation, we use the - * variance. For the record, stddev = sqrt(variance). The equation - * above becomes: - * - * abs(value - avg) < 3 x sqrt(variance) - * - * And finally we square it: - * - * (value - avg) ^ 2 < (3 x sqrt(variance)) ^ 2 - * - * (value - avg) x (value - avg) < 9 x variance - * - * Statistically speaking, any values out of this interval is - * considered as an anomaly and is discarded. However, a normal - * distribution appears when the number of samples is 30 (it is the - * rule of thumb in statistics, cf. "30 samples" on Internet). When - * there are three consecutive anomalies, the statistics are resetted. - * - */ -static void irqs_update(struct irqt_stat *irqs, u64 ts) -{ - u64 old_ts = irqs->last_ts; - u64 variance = 0; - u64 interval; - s64 diff; - - /* - * The timestamps are absolute time values, we need to compute - * the timing interval between two interrupts. - */ - irqs->last_ts = ts; - - /* - * The interval type is u64 in order to deal with the same - * type in our computation, that prevent mindfuck issues with - * overflow, sign and division. - */ - interval = ts - old_ts; - - /* - * The interrupt triggered more than one second apart, that - * ends the sequence as predictible for our purpose. In this - * case, assume we have the beginning of a sequence and the - * timestamp is the first value. As it is impossible to - * predict anything at this point, return. - * - * Note the first timestamp of the sequence will always fall - * in this test because the old_ts is zero. That is what we - * want as we need another timestamp to compute an interval. - */ - if (interval >= NSEC_PER_SEC) { - memset(irqs, 0, sizeof(*irqs)); - irqs->last_ts = ts; - return; - } - - /* - * Pre-compute the delta with the average as the result is - * used several times in this function. - */ - diff = interval - irqs->avg; - - /* - * Increment the number of samples. - */ - irqs->nr_samples++; - - /* - * Online variance divided by the number of elements if there - * is more than one sample. Normally the formula is division - * by nr_samples - 1 but we assume the number of element will be - * more than 32 and dividing by 32 instead of 31 is enough - * precise. - */ - if (likely(irqs->nr_samples > 1)) - variance = irqs->variance >> IRQ_TIMINGS_SHIFT; - - /* - * The rule of thumb in statistics for the normal distribution - * is having at least 30 samples in order to have the model to - * apply. Values outside the interval are considered as an - * anomaly. - */ - if ((irqs->nr_samples >= 30) && ((diff * diff) > (9 * variance))) { - /* - * After three consecutive anomalies, we reset the - * stats as it is no longer stable enough. - */ - if (irqs->anomalies++ >= 3) { - memset(irqs, 0, sizeof(*irqs)); - irqs->last_ts = ts; - return; - } - } else { - /* - * The anomalies must be consecutives, so at this - * point, we reset the anomalies counter. - */ - irqs->anomalies = 0; - } - - /* - * The interrupt is considered stable enough to try to predict - * the next event on it. - */ - irqs->valid = 1; - - /* - * Online average algorithm: - * - * new_average = average + ((value - average) / count) - * - * The variance computation depends on the new average - * to be computed here first. - * - */ - irqs->avg = irqs->avg + (diff >> IRQ_TIMINGS_SHIFT); - - /* - * Online variance algorithm: - * - * new_variance = variance + (value - average) x (value - new_average) - * - * Warning: irqs->avg is updated with the line above, hence - * 'interval - irqs->avg' is no longer equal to 'diff' - */ - irqs->variance = irqs->variance + (diff * (interval - irqs->avg)); - - /* - * Update the next event - */ - irqs->next_evt = ts + irqs->avg; -} - /** * irq_timings_next_event - Return when the next event is supposed to arrive * @@ -246,12 +61,6 @@ static void irqs_update(struct irqt_stat *irqs, u64 ts) */ u64 irq_timings_next_event(u64 now) { - struct irq_timings *irqts = this_cpu_ptr(&irq_timings); - struct irqt_stat *irqs; - struct irqt_stat __percpu *s; - u64 ts, next_evt = U64_MAX; - int i, irq = 0; - /* * This function must be called with the local irq disabled in * order to prevent the timings circular buffer to be updated @@ -259,64 +68,7 @@ u64 irq_timings_next_event(u64 now) */ lockdep_assert_irqs_disabled(); - /* - * Number of elements in the circular buffer: If it happens it - * was flushed before, then the number of elements could be - * smaller than IRQ_TIMINGS_SIZE, so the count is used, - * otherwise the array size is used as we wrapped. The index - * begins from zero when we did not wrap. That could be done - * in a nicer way with the proper circular array structure - * type but with the cost of extra computation in the - * interrupt handler hot path. We choose efficiency. - * - * Inject measured irq/timestamp to the statistical model - * while decrementing the counter because we consume the data - * from our circular buffer. - */ - for (i = irqts->count & IRQ_TIMINGS_MASK, - irqts->count = min(IRQ_TIMINGS_SIZE, irqts->count); - irqts->count > 0; irqts->count--, i = (i + 1) & IRQ_TIMINGS_MASK) { - - irq = irq_timing_decode(irqts->values[i], &ts); - - s = idr_find(&irqt_stats, irq); - if (s) { - irqs = this_cpu_ptr(s); - irqs_update(irqs, ts); - } - } - - /* - * Look in the list of interrupts' statistics, the earliest - * next event. - */ - idr_for_each_entry(&irqt_stats, s, i) { - - irqs = this_cpu_ptr(s); - - if (!irqs->valid) - continue; - - if (irqs->next_evt <= now) { - irq = i; - next_evt = now; - - /* - * This interrupt mustn't use in the future - * until new events occur and update the - * statistics. - */ - irqs->valid = 0; - break; - } - - if (irqs->next_evt < next_evt) { - irq = i; - next_evt = irqs->next_evt; - } - } - - return next_evt; + return 0; } void irq_timings_free(int irq) From patchwork Fri Mar 8 21:20:47 2019 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Daniel Lezcano X-Patchwork-Id: 160003 Delivered-To: patch@linaro.org Received: by 2002:a02:5cc1:0:0:0:0:0 with SMTP id w62csp9204548jad; Fri, 8 Mar 2019 13:21:15 -0800 (PST) X-Google-Smtp-Source: APXvYqxfKYKsPBrXaqdglyAjMHVSUzkmb0vwjEcx0ebUaDnK1/5REDmUqgrIJMpUv5PITVGM6FOY X-Received: by 2002:a63:e106:: with SMTP id z6mr18631485pgh.418.1552080075101; Fri, 08 Mar 2019 13:21:15 -0800 (PST) ARC-Seal: i=1; a=rsa-sha256; t=1552080075; cv=none; d=google.com; s=arc-20160816; b=JvW6aia69jweXY23ihFMDHFF4bN41Or4nqz6isK75pX9y8QhA2a7j29fMEX+d/DOsm EUj46W2F9ftmPKF+iz9b0hfCUk3wSGBRZuyqjEo/dvhzP8ZpzUsHqAycsDKEvKgyEN2q 2A90dXTGfCZfdH7Mpd+gvPAKU5DzcprgdQdz6Q4cVix8nEUXRyMtE+6l54RyKSTx+bs/ NCt07viMYgoNzu2B3WK31rLJCCWOp7Ls5aXwE5FGMp8wRNZrUlg/+nizM9g83pO7r878 gnVWBefk5Z/CYJBsmLpm+ZVe/NSyVoj2wMfq6v2SaeED7It+jLZ36yX+3bOsr5/P1MzB ZDRQ== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=list-id:precedence:sender:references:in-reply-to:message-id:date :subject:cc:to:from:dkim-signature; bh=EEyzRstIdCADM4UwDA7Tr71E6kOlhMYB3Q5LdWH90L0=; b=0kl7kq4vaQ4/lN0PxfOS1R/fwVoE1SvPymG8Wgh96MvCWyRnE1DWNBJdkgjJvC+sQL nwBXQjW07KipnBIIXQnDkgVliAusDkFVwQENxvhtMjmDOOOf2AyN/+A5G78znRGz+FkY LmD6KoLSI6TvF60F/GFXc0k4JOZX+wnZfdmsU0mbP8dFLav8grVp2MOiILWxyuf1X04u VQKVw0kA3xqdZuP7CbUOabVsjgpWfrgi1CeJox7Pvpguj0HeCJW2+iu2+QgViPp05krl 2ktS2vekcxDod3IvBB/tjLFgAbSlPKYVEaqyPcDeQ5G7OZ2TQ9UfX0sLw3rar9ggsI+3 8XXw== ARC-Authentication-Results: i=1; mx.google.com; dkim=pass header.i=@linaro.org header.s=google header.b="ccZam//5"; spf=pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=linaro.org Return-Path: Received: from vger.kernel.org (vger.kernel.org. [209.132.180.67]) by mx.google.com with ESMTP id 33si7999051pls.161.2019.03.08.13.21.14; Fri, 08 Mar 2019 13:21:15 -0800 (PST) Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67; Authentication-Results: mx.google.com; dkim=pass header.i=@linaro.org header.s=google header.b="ccZam//5"; spf=pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=linaro.org Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1726696AbfCHVVM (ORCPT + 31 others); Fri, 8 Mar 2019 16:21:12 -0500 Received: from mail-wm1-f66.google.com ([209.85.128.66]:36710 "EHLO mail-wm1-f66.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1726568AbfCHVVK (ORCPT ); Fri, 8 Mar 2019 16:21:10 -0500 Received: by mail-wm1-f66.google.com with SMTP id j125so13624908wmj.1 for ; Fri, 08 Mar 2019 13:21:08 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linaro.org; s=google; h=from:to:cc:subject:date:message-id:in-reply-to:references; bh=EEyzRstIdCADM4UwDA7Tr71E6kOlhMYB3Q5LdWH90L0=; b=ccZam//5Ujjc4iHjRbtfWX7hOfxSWxx08p7z1vDm6LTtsRMbo7ylkfLdayL/c698eG R5NjMOui3XPJlGyGyPq0H4IbpmK6hB/l8UhTy8bw0/ole05Vpi4jiBZm8ARCmTpKXsU1 v8G16xCNPwbN9YlhN1JuMhegUUWm/aIbl47diEBnGXcqQwtv3C1DbL7Jx3m/x/+BzkSN 7uwTyZiF5DlLhwyaf0o2KPKhgSsvrO8OddLd8kut6+OYLigRlitPaR+I/Zla3ETVk/Nz H8uMG3VRNseIrls5UTwRPd/V+luyH5eF9PmvKZXNj8l6nxgim+8nrFuliEEm1NH8V8Mn Leyw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:to:cc:subject:date:message-id:in-reply-to :references; bh=EEyzRstIdCADM4UwDA7Tr71E6kOlhMYB3Q5LdWH90L0=; b=i6M4sSyeXN+DhAV8P00ibTp1MenE9oF0AusCiMDe0PlKkIvZMnSwLVBbOA0tVZTAF7 BJBXPGbRPPmB/xM08kwlwVNvDLeFSphBP+J/rIANHfxCezHyLdTG+GuKPLENSPZvAqFE fnFRbg10N0JEuTjfH2aV31V9ube/znuE5x9KCJHzzTCnrNn2yvoRURX1WCNQgiXFciMa Tso2p7t/chqyoJdjm/R/cUeJDOwiHAb8hFJrYukMZtYJ6p5HamXJ+/b64zWkdpu6Z8b7 O3S3hucm4WdQIxrKfMQsxXYeTZxYyjQCOz3sBgulil70oXGxNFvYESngbFYC9r38dWJ0 1PsA== X-Gm-Message-State: APjAAAWaAHjOYH4L0ELAyoFpWEwCCzHrVL9LPkTLvp8vXJ+T1SALO4Bs gD6vjOxt9Ga6ve9x8aCytzfeiF8z+o0= X-Received: by 2002:a1c:5546:: with SMTP id j67mr9915695wmb.95.1552080067580; Fri, 08 Mar 2019 13:21:07 -0800 (PST) Received: from clegane.local (246.127.130.77.rev.sfr.net. [77.130.127.246]) by smtp.gmail.com with ESMTPSA id d2sm13568507wrq.94.2019.03.08.13.21.06 (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Fri, 08 Mar 2019 13:21:06 -0800 (PST) From: Daniel Lezcano To: rjw@rjwysocki.net, tglx@linutronix.de Cc: ulf.hansson@linaro.org, linux-pm@vger.kernel.org, linux-kernel@vger.kernel.org Subject: [PATCH 3/3] cpuidle/drivers/mbed: Add new governor for embedded systems Date: Fri, 8 Mar 2019 22:20:47 +0100 Message-Id: <20190308212047.28767-4-daniel.lezcano@linaro.org> X-Mailer: git-send-email 2.17.1 In-Reply-To: <20190308212047.28767-1-daniel.lezcano@linaro.org> References: <20190308212047.28767-1-daniel.lezcano@linaro.org> Sender: linux-kernel-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org The objective is the same for all the governors: save energy, but at the end the governors menu, ladder and teo aim to improve the performances with an acceptable energy drop for some workloads which are identified for servers and desktops (with the help of a firmware). It is very difficult to do changes in these governors for embedded systems without impacting performances on servers/desktops or ruin the optimizations for the workloads on these platforms. The mbed governor is a new governor targeting embedded systems running on battery where the energy saving has a higher priority than servers or desktops. This governor aims to save energy as much as possible but accepting a performance degradation. In this way, we can optimize the governor for specific mobile workload and more generally embedded systems without impacting other platforms. The governor is based on the irq timings where we predict the next interrupt occurrences on the current CPU and the next timer. It is well suited for mobile and more generally embedded systems where the interrupts are usually pinned on one CPU and where the power is more important than the performances. The multimedia applications on the embedded system spawns several threads which are migrated across the different CPUs and waking between them up. In order to catch this situation we have also to track the idle task rescheduling duration with a relative degree of confidence as the scheduler is involved in the task migrations. The resched information is in the scope of the governor via the reflect callback. The governor begins with a clean foundation basing the prediction on the irq behavior returned by the irq timings, the timers and the idle task rescheduling. The advantage of the approach is we have a full view of the wakeup sources as we identify them separately and then we can control the situation without relying on biased heuristics. Despite the naive idle state selection for this first iteration, the governor provides a performance improvement of 30% for Jankbench throughout for the same amount of energy than the menu governor and a similar energy saving for other workloads. There are areas of improvement for the different workloads which will be optimized iteratively with non-regression testing for previous identified workloads on an Android reference platform. Signed-off-by: Daniel Lezcano --- drivers/cpuidle/Kconfig | 11 ++- drivers/cpuidle/governors/Makefile | 1 + drivers/cpuidle/governors/mbed.c | 143 +++++++++++++++++++++++++++++ 3 files changed, 154 insertions(+), 1 deletion(-) create mode 100644 drivers/cpuidle/governors/mbed.c -- 2.17.1 diff --git a/drivers/cpuidle/Kconfig b/drivers/cpuidle/Kconfig index 8caccbbd7353..6fef94c9bde0 100644 --- a/drivers/cpuidle/Kconfig +++ b/drivers/cpuidle/Kconfig @@ -4,7 +4,7 @@ config CPU_IDLE bool "CPU idle PM support" default y if ACPI || PPC_PSERIES select CPU_IDLE_GOV_LADDER if (!NO_HZ && !NO_HZ_IDLE) - select CPU_IDLE_GOV_MENU if (NO_HZ || NO_HZ_IDLE) && !CPU_IDLE_GOV_TEO + select CPU_IDLE_GOV_MENU if (NO_HZ || NO_HZ_IDLE) && !CPU_IDLE_GOV_TEO && !CPU_IDLE_GOV_MBED help CPU idle is a generic framework for supporting software-controlled idle processor power management. It includes modular cross-platform @@ -32,6 +32,15 @@ config CPU_IDLE_GOV_TEO Some workloads benefit from using it and it generally should be safe to use. Say Y here if you are not happy with the alternatives. +config CPU_IDLE_GOV_MBED + bool "Embedded governor" + select IRQ_TIMINGS + help + The embedded governor is based on irq timings measurements and + pattern research combined with the next timer. This governor + suits very well on embedded systems where the interrupts are + grouped on a single core and the power is the priority. + config DT_IDLE_STATES bool diff --git a/drivers/cpuidle/governors/Makefile b/drivers/cpuidle/governors/Makefile index 4d8aff5248a8..d4571935575c 100644 --- a/drivers/cpuidle/governors/Makefile +++ b/drivers/cpuidle/governors/Makefile @@ -5,3 +5,4 @@ obj-$(CONFIG_CPU_IDLE_GOV_LADDER) += ladder.o obj-$(CONFIG_CPU_IDLE_GOV_MENU) += menu.o obj-$(CONFIG_CPU_IDLE_GOV_TEO) += teo.o +obj-$(CONFIG_CPU_IDLE_GOV_MBED) += mbed.o diff --git a/drivers/cpuidle/governors/mbed.c b/drivers/cpuidle/governors/mbed.c new file mode 100644 index 000000000000..8d8c8ba92533 --- /dev/null +++ b/drivers/cpuidle/governors/mbed.c @@ -0,0 +1,143 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (C) 2019, Linaro Ltd + * Author: Daniel Lezcano + */ +#include +#include +#include +#include +#include +#include +#include + +struct mbed_device { + u64 idle_ema_avg; + u64 idle_total; + unsigned long last_jiffies; +}; + +#define EMA_ALPHA_VAL 64 +#define EMA_ALPHA_SHIFT 7 +#define MAX_RESCHED_INTERVAL_MS 100 + +static DEFINE_PER_CPU(struct mbed_device, mbed_devices); + +static int mbed_ema_new(s64 value, s64 ema_old) +{ + if (likely(ema_old)) + return ema_old + (((value - ema_old) * EMA_ALPHA_VAL) >> + EMA_ALPHA_SHIFT); + return value; +} + +static void mbed_reflect(struct cpuidle_device *dev, int index) +{ + struct mbed_device *mbed_dev = this_cpu_ptr(&mbed_devices); + + /* + * The idle task was not rescheduled since + * MAX_RESCHED_INTERVAL_MS, let's consider the duration is + * long enough to clear our stats. + */ + if (time_after(jiffies, mbed_dev->last_jiffies + + msecs_to_jiffies(MAX_RESCHED_INTERVAL_MS))) + mbed_dev->idle_ema_avg = 0; + + /* + * Sum all the residencies in order to compute the total + * duration of the idle task. + */ + mbed_dev->idle_total += dev->last_residency; + + /* + * We exited the idle state with the need_resched() flag, the + * idle task will be rescheduled, so store the duration the + * idle task was scheduled in an exponential moving average and + * reset the total of the idle duration. + */ + if (need_resched()) { + mbed_dev->idle_ema_avg = mbed_ema_new(mbed_dev->idle_total, + mbed_dev->idle_ema_avg); + mbed_dev->idle_total = 0; + mbed_dev->last_jiffies = jiffies; + } +} + +static int mbed_select(struct cpuidle_driver *drv, struct cpuidle_device *dev, + bool *stop_tick) +{ + struct mbed_device *mbed_dev = this_cpu_ptr(&mbed_devices); + int latency_req = cpuidle_governor_latency_req(dev->cpu); + int i, index = 0; + ktime_t delta_next; + u64 now, irq_length, timer_length; + u64 idle_duration_us; + + /* + * Get the present time as reference for the next steps + */ + now = local_clock(); + + /* + * Get the next interrupt event giving the 'now' as a + * reference, if the next event appears to have already + * expired then we get the 'now' returned which ends up with a + * zero duration. + */ + irq_length = irq_timings_next_event(now) - now; + + /* + * Get the timer duration before expiration. + */ + timer_length = ktime_to_ns(tick_nohz_get_sleep_length(&delta_next)); + + /* + * Get the smallest duration between the timer and the irq next event. + */ + idle_duration_us = min_t(u64, irq_length, timer_length) / NSEC_PER_USEC; + + /* + * Get the idle task duration average if the information is + * available. + */ + if (mbed_dev->idle_ema_avg) + idle_duration_us = min_t(u64, idle_duration_us, + mbed_dev->idle_ema_avg); + + for (i = 0; i < drv->state_count; i++) { + struct cpuidle_state *s = &drv->states[i]; + struct cpuidle_state_usage *su = &dev->states_usage[i]; + + if (s->disabled || su->disable) + continue; + + if (s->exit_latency > latency_req) + break; + + if (s->target_residency > idle_duration_us) + break; + + index = i; + } + + if (!index) + *stop_tick = false; + + return index; +} + +static struct cpuidle_governor mbed_governor = { + .name = "mbed", + .rating = 20, + .select = mbed_select, + .reflect = mbed_reflect, +}; + +static int __init init_governor(void) +{ + irq_timings_enable(); + return cpuidle_register_governor(&mbed_governor); +} + +postcore_initcall(init_governor);