From patchwork Fri Mar 30 16:24:43 2012 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: Brad Figg X-Patchwork-Id: 149681 Return-Path: X-Original-To: incoming@patchwork.ozlabs.org Delivered-To: patchwork-incoming@bilbo.ozlabs.org Received: from chlorine.canonical.com (chlorine.canonical.com [91.189.94.204]) by ozlabs.org (Postfix) with ESMTP id B7F7EB6EF3 for ; Sat, 31 Mar 2012 03:24:55 +1100 (EST) Received: from localhost ([127.0.0.1] helo=chlorine.canonical.com) by chlorine.canonical.com with esmtp (Exim 4.71) (envelope-from ) id 1SDed7-0006qs-6f; Fri, 30 Mar 2012 16:24:49 +0000 Received: from youngberry.canonical.com ([91.189.89.112]) by chlorine.canonical.com with esmtp (Exim 4.71) (envelope-from ) id 1SDed5-0006q1-09 for kernel-team@lists.ubuntu.com; Fri, 30 Mar 2012 16:24:47 +0000 Received: from static-50-53-107-235.bvtn.or.frontiernet.net ([50.53.107.235] helo=localhost) by youngberry.canonical.com with esmtpsa (TLS1.0:DHE_RSA_AES_128_CBC_SHA1:16) (Exim 4.71) (envelope-from ) id 1SDed4-0003cX-Fy for kernel-team@lists.ubuntu.com; Fri, 30 Mar 2012 16:24:46 +0000 From: Brad Figg To: kernel-team@lists.ubuntu.com Subject: [PATCH 1/1] [PRECISE] sched: Fix nohz load accounting Date: Fri, 30 Mar 2012 09:24:43 -0700 Message-Id: <1333124683-30868-2-git-send-email-brad.figg@canonical.com> X-Mailer: git-send-email 1.7.9.1 In-Reply-To: <1333124683-30868-1-git-send-email-brad.figg@canonical.com> References: <1333124683-30868-1-git-send-email-brad.figg@canonical.com> MIME-Version: 1.0 X-BeenThere: kernel-team@lists.ubuntu.com X-Mailman-Version: 2.1.13 Precedence: list List-Id: Kernel team discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: kernel-team-bounces@lists.ubuntu.com Errors-To: kernel-team-bounces@lists.ubuntu.com BugLink: http://bugs.launchpad.net/bugs/838811 Various people reported nohz load tracking still being wrecked, but Doug spotted the actual problem. We fold the nohz remainder in too soon, causing us to loose samples and under-account. So instead of playing catch-up up-front, always do a single load-fold with whatever state we encounter and only then fold the nohz remainder and play catch-up. Reported-by: Doug Smythies Reported-by: LesÅ=82aw Kope=C4=87 Reported-by: Aman Gupta Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/n/tip-4v31etnhgg9kwd6ocgx3rxl8@git.kernel.org Signed-off-by: Ingo Molnar (backport of upstream: c308b56b5398779cd3da0f62ab26b0453494c3d4) Signed-off-by: Brad Figg --- kernel/sched.c | 53 ++++++++++++++++++++++++++--------------------------- 1 files changed, 26 insertions(+), 27 deletions(-) diff --git a/kernel/sched.c b/kernel/sched.c index 3a9f25d..654cd33 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -3538,13 +3538,10 @@ calc_load_n(unsigned long load, unsigned long exp, * Once we've updated the global active value, we need to apply the exponential * weights adjusted to the number of cycles missed. */ -static void calc_global_nohz(unsigned long ticks) +static void calc_global_nohz(void) { long delta, active, n; - if (time_before(jiffies, calc_load_update)) - return; - /* * If we crossed a calc_load_update boundary, make sure to fold * any pending idle changes, the respective CPUs might have @@ -3556,31 +3553,25 @@ static void calc_global_nohz(unsigned long ticks) atomic_long_add(delta, &calc_load_tasks); /* - * If we were idle for multiple load cycles, apply them. + * It could be the one fold was all it took, we done! */ - if (ticks >= LOAD_FREQ) { - n = ticks / LOAD_FREQ; + if (time_before(jiffies, calc_load_update + 10)) + return; - active = atomic_long_read(&calc_load_tasks); - active = active > 0 ? active * FIXED_1 : 0; + /* + * Catch-up, fold however many we are behind still + */ + delta = jiffies - calc_load_update - 10; + n = 1 + (delta / LOAD_FREQ); - avenrun[0] = calc_load_n(avenrun[0], EXP_1, active, n); - avenrun[1] = calc_load_n(avenrun[1], EXP_5, active, n); - avenrun[2] = calc_load_n(avenrun[2], EXP_15, active, n); + active = atomic_long_read(&calc_load_tasks); + active = active > 0 ? active * FIXED_1 : 0; - calc_load_update += n * LOAD_FREQ; - } + avenrun[0] = calc_load_n(avenrun[0], EXP_1, active, n); + avenrun[1] = calc_load_n(avenrun[1], EXP_5, active, n); + avenrun[2] = calc_load_n(avenrun[2], EXP_15, active, n); - /* - * Its possible the remainder of the above division also crosses - * a LOAD_FREQ period, the regular check in calc_global_load() - * which comes after this will take care of that. - * - * Consider us being 11 ticks before a cycle completion, and us - * sleeping for 4*LOAD_FREQ + 22 ticks, then the above code will - * age us 4 cycles, and the test in calc_global_load() will - * pick up the final one. - */ + calc_load_update += n * LOAD_FREQ; } #else static void calc_load_account_idle(struct rq *this_rq) @@ -3592,7 +3583,7 @@ static inline long calc_load_fold_idle(void) return 0; } -static void calc_global_nohz(unsigned long ticks) +static void calc_global_nohz(void) { } #endif @@ -3620,8 +3611,6 @@ void calc_global_load(unsigned long ticks) { long active; - calc_global_nohz(ticks); - if (time_before(jiffies, calc_load_update + 10)) return; @@ -3633,6 +3622,16 @@ void calc_global_load(unsigned long ticks) avenrun[2] = calc_load(avenrun[2], EXP_15, active); calc_load_update += LOAD_FREQ; + + /* + * Account one period with whatever state we found before + * folding in the nohz state and ageing the entire idle period. + * + * This avoids loosing a sample when we go idle between + * calc_load_account_active() (10 ticks ago) and now and thus + * under-accounting. + */ + calc_global_nohz(); } /*