From: Boris Burkov Subject: [PATCH cgroup/for-5.8] cgroup: add cpu.stat file to root cgroup Date: Tue, 26 May 2020 13:08:00 -0700 Message-ID: <20200526200800.3969430-1-boris@bur.io> Mime-Version: 1.0 Content-Transfer-Encoding: 8bit Return-path: DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=bur.io; h=from :to:cc:subject:date:message-id:mime-version :content-transfer-encoding; s=fm1; bh=LHN0JqCZjAhlY7hB0/Y3fto1Fs QcamBNWhl7bqCCYFQ=; b=Tf85u/C9df53lBvUWR+YDqK8pTkDrlYwLiHV44qbkS mnHeugttyrDec7o60NcUtnABRLdccSQKc7t59jSSHpVbVIAJrtb9dFc/tI/VCFik ZmenzZbrOUgnS/E5bPrTAnCfY2QUl2V5PbiPLZ6nLYZR3eKjZUKpp9FCkaJb30yO xHPJ/DJA2342C4+6ldrjdpka0rNkcWmswGMP/nbsL8819nj+hYgfqCBTdhVTldJq ZE556X5zBI1eSwczYvHj0naUQeMjaUmhbdB4D/ZL+psVh7Pz15JWuFIRB//rpR0r 2AIZr2SY0irzoSvvg65uX40vdxoYuEbvGN3spSfXJEug== DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d= messagingengine.com; h=cc:content-transfer-encoding:date:from :message-id:mime-version:subject:to:x-me-proxy:x-me-proxy :x-me-sender:x-me-sender:x-sasl-enc; s=fm2; bh=LHN0JqCZjAhlY7hB0 /Y3fto1FsQcamBNWhl7bqCCYFQ=; b=wAS94y4Q5SKaY0h7NxjaHI5iBN210i57D RMFJcWHfxHh4bN4m9zPx+8tecS3ekBMEiCwemkO5TjYMCd7bknJVCfTtp2A6Wr6o TIAbO0h/HNHYhSlcdKB/c5NuXReBNeMfDL5B+uAz/G0R/84JOwTj4keSLoe7nT4n 0QWEJS7F/+XmhW8SAZjxGkxgHr8TUORJZsw7QuJscdG8dSLaJAv1JoutVw9numCH WnY8Q6xN3xtlFgw3Dia2CU7YIiCCRAcZVfh5WoLmyyg0d3nExliI4pMUPGipDn5T eFk9/hflNNQBnrXsfAA1IWt+S/KtY7yCCB0VMBe1DR+aLxSumqazA== Sender: cgroups-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-ID: Content-Type: text/plain; charset="us-ascii" To: Tejun Heo , Li Zefan , Johannes Weiner Cc: Boris Burkov , Ingo Molnar , Peter Zijlstra , Juri Lelli , Vincent Guittot , cgroups-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, kernel-team-b10kYP2dOMg@public.gmane.org Currently, the root cgroup does not have a cpu.stat file. Add one which is consistent with /proc/stat to capture global cpu statistics that might not fall under cgroup accounting. We haven't done this in the past because the data are already presented in /proc/stat and we didn't want to add overhead from collecting root cgroup stats when cgroups are configured, but no cgroups have been created. By keeping the data consistent with /proc/stat, I think we avoid the first problem, while improving the usability of cgroups stats. We avoid the second problem by computing the contents of cpu.stat from existing data collected for /proc/stat anyway. Signed-off-by: Boris Burkov Suggested-by: Tejun Heo --- kernel/cgroup/cgroup.c | 1 - kernel/cgroup/rstat.c | 60 ++++++++++++++++++++++++++++++++++++------ 2 files changed, 52 insertions(+), 9 deletions(-) diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c index 557a9b9d2244..b8a75169c3e4 100644 --- a/kernel/cgroup/cgroup.c +++ b/kernel/cgroup/cgroup.c @@ -4881,7 +4881,6 @@ static struct cftype cgroup_base_files[] = { }, { .name = "cpu.stat", - .flags = CFTYPE_NOT_ON_ROOT, .seq_show = cpu_stat_show, }, #ifdef CONFIG_PSI diff --git a/kernel/cgroup/rstat.c b/kernel/cgroup/rstat.c index 41ca996568df..b6397a186ce9 100644 --- a/kernel/cgroup/rstat.c +++ b/kernel/cgroup/rstat.c @@ -389,18 +389,62 @@ void __cgroup_account_cputime_field(struct cgroup *cgrp, cgroup_base_stat_cputime_account_end(cgrp, rstatc); } +/* + * compute the cputime for the root cgroup by getting the per cpu data + * at a global level, then categorizing the fields in a manner consistent + * with how it is done by __cgroup_account_cputime_field for each bit of + * cpu time attributed to a cgroup. + */ +static void root_cgroup_cputime(struct task_cputime *cputime) +{ + int i; + + cputime->stime = 0; + cputime->utime = 0; + cputime->sum_exec_runtime = 0; + for_each_possible_cpu(i) { + struct kernel_cpustat kcpustat; + u64 *cpustat = kcpustat.cpustat; + u64 user = 0; + u64 sys = 0; + + kcpustat_cpu_fetch(&kcpustat, i); + + user += cpustat[CPUTIME_USER]; + user += cpustat[CPUTIME_NICE]; + cputime->utime += user; + + sys += cpustat[CPUTIME_SYSTEM]; + sys += cpustat[CPUTIME_IRQ]; + sys += cpustat[CPUTIME_SOFTIRQ]; + cputime->stime += sys; + + cputime->sum_exec_runtime += user; + cputime->sum_exec_runtime += sys; + cputime->sum_exec_runtime += cpustat[CPUTIME_STEAL]; + cputime->sum_exec_runtime += cpustat[CPUTIME_GUEST]; + cputime->sum_exec_runtime += cpustat[CPUTIME_GUEST_NICE]; + } +} + void cgroup_base_stat_cputime_show(struct seq_file *seq) { struct cgroup *cgrp = seq_css(seq)->cgroup; u64 usage, utime, stime; - - if (!cgroup_parent(cgrp)) - return; - - cgroup_rstat_flush_hold(cgrp); - usage = cgrp->bstat.cputime.sum_exec_runtime; - cputime_adjust(&cgrp->bstat.cputime, &cgrp->prev_cputime, &utime, &stime); - cgroup_rstat_flush_release(); + struct task_cputime cputime; + + if (cgroup_parent(cgrp)) { + cgroup_rstat_flush_hold(cgrp); + usage = cgrp->bstat.cputime.sum_exec_runtime; + cputime_adjust(&cgrp->bstat.cputime, &cgrp->prev_cputime, + &utime, &stime); + cgroup_rstat_flush_release(); + } else { + root_cgroup_cputime(&cputime); + usage = cputime.sum_exec_runtime; + utime = cputime.utime; + stime = cputime.stime; + } do_div(usage, NSEC_PER_USEC); do_div(utime, NSEC_PER_USEC); -- 2.24.1