From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from kanga.kvack.org (kanga.kvack.org [205.233.56.17]) by smtp.lore.kernel.org (Postfix) with ESMTP id 97272C3DA6E for ; Wed, 20 Dec 2023 05:27:25 +0000 (UTC) Received: by kanga.kvack.org (Postfix) id F25366B0080; Wed, 20 Dec 2023 00:27:24 -0500 (EST) Received: by kanga.kvack.org (Postfix, from userid 40) id EAE976B0081; Wed, 20 Dec 2023 00:27:24 -0500 (EST) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id C89726B0082; Wed, 20 Dec 2023 00:27:24 -0500 (EST) X-Delivered-To: linux-mm@kvack.org Received: from relay.hostedemail.com (smtprelay0014.hostedemail.com [216.40.44.14]) by kanga.kvack.org (Postfix) with ESMTP id A98C26B0080 for ; Wed, 20 Dec 2023 00:27:24 -0500 (EST) Received: from smtpin26.hostedemail.com (a10.router.float.18 [10.200.18.1]) by unirelay09.hostedemail.com (Postfix) with ESMTP id 6F679804D0 for ; Wed, 20 Dec 2023 05:27:24 +0000 (UTC) X-FDA: 81586063608.26.AD72361 Received: from mail-pl1-f170.google.com (mail-pl1-f170.google.com [209.85.214.170]) by imf03.hostedemail.com (Postfix) with ESMTP id 2687F2000D for ; Wed, 20 Dec 2023 05:27:21 +0000 (UTC) Authentication-Results: imf03.hostedemail.com; dkim=pass header.d=gmail.com header.s=20230601 header.b=d6fXSgp4; dmarc=pass (policy=none) header.from=gmail.com; spf=pass (imf03.hostedemail.com: domain of shy828301@gmail.com designates 209.85.214.170 as permitted sender) smtp.mailfrom=shy828301@gmail.com ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=hostedemail.com; s=arc-20220608; t=1703050042; h=from:from:sender:reply-to:subject:subject:date:date: message-id:message-id:to:to:cc:cc:mime-version:mime-version: content-type:content-type: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references:dkim-signature; bh=HyCa8aNvFFMzS/2cpbfLyfUNa1SzwIXdOb7F8Mv2w/s=; b=tSwZKVSgTWUCDiE4Hq4KqsDt1mNyv6M0TqElfr0/XR0EN9qpH/0H7CvzIE15HyRavoNmzt Mhi2btlAI1Aq6D3auoU2gXg1X8tjsrOsCfWL85CrrO8wEoDv+iBfDFi87tqE+BJymwlYLf dvmkvsaiD0ShKGXtmSJ3/ES3kc+y/Bg= ARC-Authentication-Results: i=1; imf03.hostedemail.com; dkim=pass header.d=gmail.com header.s=20230601 header.b=d6fXSgp4; dmarc=pass (policy=none) header.from=gmail.com; spf=pass (imf03.hostedemail.com: domain of shy828301@gmail.com designates 209.85.214.170 as permitted sender) smtp.mailfrom=shy828301@gmail.com ARC-Seal: i=1; s=arc-20220608; d=hostedemail.com; t=1703050042; a=rsa-sha256; cv=none; b=CTGGnMlWa9li63/otr7TKanFwrSlXTmKC97sSjwXZSQPYrWTj8C2EQWUBFmDFPXJlqyW+Y iO920nRQ1QZPSwcGkuS95MVGDIofZgkgnbMvgL7sIj2ySKQTmyalFwA7Jh02xHcCQATlCu 3w8Zmb5k1qL/kE/eeS1Y1iif2YPe9Kw= Received: by mail-pl1-f170.google.com with SMTP id d9443c01a7336-1d337dc9697so42942735ad.3 for ; Tue, 19 Dec 2023 21:27:21 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20230601; t=1703050041; x=1703654841; darn=kvack.org; h=content-transfer-encoding:cc:to:subject:message-id:date:from :in-reply-to:references:mime-version:from:to:cc:subject:date :message-id:reply-to; bh=HyCa8aNvFFMzS/2cpbfLyfUNa1SzwIXdOb7F8Mv2w/s=; b=d6fXSgp4bherC7fKd3jZ+JN94GjUsq+YdPqk+xDb0kYqDLtuM75xQwWyqad6yE3R5n HEYubLpde95Pq0GknXdOkJ97Oq//l6mVgQaB+Cpk9f3FCYkP9YBFE/1qEzK6edb/V+9i hAy9+CqCr5tf3+cXjeRV2/eeqllAwLlH8TaM9QbeSF6lYBDmtLnDBkFH3Rm5aAe5MuVX o5PBoZlBxTirF+9gYdWGksbrC7VNN8ab4nQG1SZBuhriuu+Ykgwi42hmIjatZZv1B4sU r69EfUhSzgae8huFpBGLjLRapritSM78PEcKC/LOqURW+O7ujineua5vsk8jikEM2zXX yqng== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1703050041; x=1703654841; h=content-transfer-encoding:cc:to:subject:message-id:date:from :in-reply-to:references:mime-version:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=HyCa8aNvFFMzS/2cpbfLyfUNa1SzwIXdOb7F8Mv2w/s=; b=pMf5S3d8T5+jm24lwon7ck9iGRiwrppKTlhw/hJuN3AOa11IbYZdOca1XD8YUFxRMH KIekm55KNgJMNdLIUjnl0dBaIlRIuoxx6Wd9RzpgnFnGomyLNRpaKFpUNqesV9A/B3PB GoVHEWSr/yDGO+0OsPM7jr8tHVrJIBxz5Sqs/wVnZ1WKoQARZoc9FxCvrO3SUBHww4BL 47knTN5+yryZnBZb0sJGxwkWrFqVq5ty8cKf40fmNZEPlCP7+lHIQG+njirDniLwsqZx mLMOxIXsm14ZJJRFuqoFBM2m1tDntsrSDxCm7m5oOmotyybEKP0urr3YTAYEQMY8zYHN lFPA== X-Gm-Message-State: AOJu0YzvxhgOksBFCcs4HXANO94MdB2rB596ZO39RqiiSPtDCCqVaL4n 8oXrvjC4sket+TmjZvwkp7lFLcnBL9alVj3ErHY= X-Google-Smtp-Source: AGHT+IFpaXaKguxX+QoX4tXMfcPd6AMOhSNFJ+rcfbC6yZOr5pKzM1gxfggtNjklpbHXbBt0oT6kyXKTszwYHx7Isdo= X-Received: by 2002:a17:902:dacf:b0:1d3:d72e:ce33 with SMTP id q15-20020a170902dacf00b001d3d72ece33mr4349157plx.107.1703050040181; Tue, 19 Dec 2023 21:27:20 -0800 (PST) MIME-Version: 1.0 References: <202312192310.56367035-oliver.sang@intel.com> In-Reply-To: <202312192310.56367035-oliver.sang@intel.com> From: Yang Shi Date: Tue, 19 Dec 2023 21:27:07 -0800 Message-ID: Subject: Re: [linux-next:master] [mm] 1111d46b5c: stress-ng.pthread.ops_per_sec -84.3% regression To: kernel test robot Cc: Rik van Riel , oe-lkp@lists.linux.dev, lkp@intel.com, Linux Memory Management List , Andrew Morton , Matthew Wilcox , Christopher Lameter , ying.huang@intel.com, feng.tang@intel.com, fengwei.yin@intel.com Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Rspamd-Queue-Id: 2687F2000D X-Rspam-User: X-Rspamd-Server: rspam04 X-Stat-Signature: 19pfnoaicyzfrf3iz7ud5kx95ktqfu9m X-HE-Tag: 1703050041-695461 X-HE-Meta: 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 FF93yQ83 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 X-Bogosity: Ham, tests=bogofilter, spamicity=0.000000, version=1.2.4 Sender: owner-linux-mm@kvack.org Precedence: bulk X-Loop: owner-majordomo@kvack.org List-ID: List-Subscribe: List-Unsubscribe: On Tue, Dec 19, 2023 at 7:41=E2=80=AFAM kernel test robot wrote: > > > > Hello, > > for this commit, we reported > "[mm] 96db82a66d: will-it-scale.per_process_ops -95.3% regression" > in Aug, 2022 when it's in linux-next/master > https://lore.kernel.org/all/YwIoiIYo4qsYBcgd@xsang-OptiPlex-9020/ > > later, we reported > "[mm] f35b5d7d67: will-it-scale.per_process_ops -95.5% regression" > in Oct, 2022 when it's in linus/master > https://lore.kernel.org/all/202210181535.7144dd15-yujie.liu@intel.com/ > > and the commit was reverted finally by > commit 0ba09b1733878afe838fe35c310715fda3d46428 > Author: Linus Torvalds > Date: Sun Dec 4 12:51:59 2022 -0800 > > now we noticed it goes into linux-next/master again. > > we are not sure if there is an agreement that the benefit of this commit > has already overweight performance drop in some mirco benchmark. > > we also noticed from https://lore.kernel.org/all/20231214223423.1133074-1= -yang@os.amperecomputing.com/ > that > "This patch was applied to v6.1, but was reverted due to a regression > report. However it turned out the regression was not due to this patch. > I ping'ed Andrew to reapply this patch, Andrew may forget it. This > patch helps promote THP, so I rebased it onto the latest mm-unstable." IIRC, Huang Ying's analysis showed the regression for will-it-scale micro benchmark is fine, it was actually reverted due to kernel build regression with LLVM reported by Nathan Chancellor. Then the regression was resolved by commit 81e506bec9be1eceaf5a2c654e28ba5176ef48d8 ("mm/thp: check and bail out if page in deferred queue already"). And this patch did improve kernel build with GCC by ~3% if I remember correctly. > > however, unfortunately, in our latest tests, we still observed below regr= ession > upon this commit. just FYI. > > > > kernel test robot noticed a -84.3% regression of stress-ng.pthread.ops_pe= r_sec on: Interesting, wasn't the same regression seen last time? And I'm a little bit confused about how pthread got regressed. I didn't see the pthread benchmark do any intensive memory alloc/free operations. Do the pthread APIs do any intensive memory operations? I saw the benchmark does allocate memory for thread stack, but it should be just 8K per thread, so it should not trigger what this patch does. With 1024 threads, the thread stacks may get merged into one single VMA (8M total), but it may do so even though the patch is not applied. > > > commit: 1111d46b5cbad57486e7a3fab75888accac2f072 ("mm: align larger anony= mous mappings on THP boundaries") > https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git master > > testcase: stress-ng > test machine: 36 threads 1 sockets Intel(R) Core(TM) i9-10980XE CPU @ 3.0= 0GHz (Cascade Lake) with 128G memory > parameters: > > nr_threads: 1 > disk: 1HDD > testtime: 60s > fs: ext4 > class: os > test: pthread > cpufreq_governor: performance > > > In addition to that, the commit also has significant impact on the follow= ing tests: > > +------------------+-----------------------------------------------------= ------------------------------------------+ > | testcase: change | stream: stream.triad_bandwidth_MBps -12.1% regressio= n | > | test machine | 224 threads 2 sockets Intel(R) Xeon(R) Platinum 8480= CTDX (Sapphire Rapids) with 512G memory | > | test parameters | array_size=3D50000000 = | > | | cpufreq_governor=3Dperformance = | > | | iterations=3D10x = | > | | loop=3D100 = | > | | nr_threads=3D25% = | > | | omp=3Dtrue = | > +------------------+-----------------------------------------------------= ------------------------------------------+ > | testcase: change | phoronix-test-suite: phoronix-test-suite.ramspeed.Av= erage.Integer.mb_s -3.5% regression | > | test machine | 12 threads 1 sockets Intel(R) Core(TM) i7-8700 CPU @= 3.20GHz (Coffee Lake) with 16G memory | > | test parameters | cpufreq_governor=3Dperformance = | > | | option_a=3DAverage = | > | | option_b=3DInteger = | > | | test=3Dramspeed-1.4.3 = | > +------------------+-----------------------------------------------------= ------------------------------------------+ > | testcase: change | phoronix-test-suite: phoronix-test-suite.ramspeed.Av= erage.FloatingPoint.mb_s -3.0% regression | > | test machine | 12 threads 1 sockets Intel(R) Core(TM) i7-8700 CPU @= 3.20GHz (Coffee Lake) with 16G memory | > | test parameters | cpufreq_governor=3Dperformance = | > | | option_a=3DAverage = | > | | option_b=3DFloating Point = | > | | test=3Dramspeed-1.4.3 = | > +------------------+-----------------------------------------------------= ------------------------------------------+ > > > If you fix the issue in a separate patch/commit (i.e. not just a new vers= ion of > the same patch/commit), kindly add following tags > | Reported-by: kernel test robot > | Closes: https://lore.kernel.org/oe-lkp/202312192310.56367035-oliver.san= g@intel.com > > > Details are as below: > -------------------------------------------------------------------------= -------------------------> > > > The kernel config and materials to reproduce are available at: > https://download.01.org/0day-ci/archive/20231219/202312192310.56367035-ol= iver.sang@intel.com > > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D > class/compiler/cpufreq_governor/disk/fs/kconfig/nr_threads/rootfs/tbox_gr= oup/test/testcase/testtime: > os/gcc-12/performance/1HDD/ext4/x86_64-rhel-8.3/1/debian-11.1-x86_64-20= 220510.cgz/lkp-csl-d02/pthread/stress-ng/60s > > commit: > 30749e6fbb ("mm/memory: replace kmap() with kmap_local_page()") > 1111d46b5c ("mm: align larger anonymous mappings on THP boundaries") > > 30749e6fbb3d391a 1111d46b5cbad57486e7a3fab75 > ---------------- --------------------------- > %stddev %change %stddev > \ | \ > 13405796 -65.5% 4620124 cpuidle..usage > 8.00 +8.2% 8.66 =C4=85 2% iostat.cpu.system > 1.61 -60.6% 0.63 iostat.cpu.user > 597.50 =C4=85 14% -64.3% 213.50 =C4=85 14% perf-c2c.DRAM.loc= al > 1882 =C4=85 14% -74.7% 476.83 =C4=85 7% perf-c2c.HITM.loc= al > 3768436 -12.9% 3283395 vmstat.memory.cache > 355105 -75.7% 86344 =C4=85 3% vmstat.system.cs > 385435 -20.7% 305714 =C4=85 3% vmstat.system.in > 1.13 -0.2 0.88 mpstat.cpu.all.irq% > 0.29 -0.2 0.10 =C4=85 2% mpstat.cpu.all.soft% > 6.76 =C4=85 2% +1.1 7.88 =C4=85 2% mpstat.cpu.all.sy= s% > 1.62 -1.0 0.62 =C4=85 2% mpstat.cpu.all.usr% > 2234397 -84.3% 350161 =C4=85 5% stress-ng.pthread.ops > 37237 -84.3% 5834 =C4=85 5% stress-ng.pthread.ops_= per_sec > 294706 =C4=85 2% -68.0% 94191 =C4=85 6% stress-ng.time.in= voluntary_context_switches > 41442 =C4=85 2% +5023.4% 2123284 stress-ng.time.maximum= _resident_set_size > 4466457 -83.9% 717053 =C4=85 5% stress-ng.time.minor_p= age_faults The larger RSS and fewer page faults are expected. > 243.33 +13.5% 276.17 =C4=85 3% stress-ng.time.percent= _of_cpu_this_job_got > 131.64 +27.7% 168.11 =C4=85 3% stress-ng.time.system_= time > 19.73 -82.1% 3.53 =C4=85 4% stress-ng.time.user_ti= me Much less user time. And it seems to match the drop of the pthread metric. > 7715609 -80.2% 1530125 =C4=85 4% stress-ng.time.volunta= ry_context_switches > 494566 -59.5% 200338 =C4=85 3% meminfo.Active > 478287 -61.5% 184050 =C4=85 3% meminfo.Active(anon) > 58549 =C4=85 17% +1532.8% 956006 =C4=85 14% meminfo.AnonHugeP= ages > 424631 +194.9% 1252445 =C4=85 10% meminfo.AnonPages > 3677263 -13.0% 3197755 meminfo.Cached > 5829485 =C4=85 4% -19.0% 4724784 =C4=85 10% meminfo.Committed= _AS > 692486 +108.6% 1444669 =C4=85 8% meminfo.Inactive > 662179 +113.6% 1414338 =C4=85 9% meminfo.Inactive(anon) > 182416 -50.2% 90759 meminfo.Mapped > 4614466 +10.0% 5076604 =C4=85 2% meminfo.Memused > 6985 +47.6% 10307 =C4=85 4% meminfo.PageTables > 718445 -66.7% 238913 =C4=85 3% meminfo.Shmem > 35906 -20.7% 28471 =C4=85 3% meminfo.VmallocUsed > 4838522 +25.6% 6075302 meminfo.max_used_kB > 488.83 -20.9% 386.67 =C4=85 2% turbostat.Avg_MHz > 12.95 -2.7 10.26 =C4=85 2% turbostat.Busy% > 7156734 -87.2% 919149 =C4=85 4% turbostat.C1 > 10.59 -8.9 1.65 =C4=85 5% turbostat.C1% > 3702647 -55.1% 1663518 =C4=85 2% turbostat.C1E > 32.99 -20.6 12.36 =C4=85 3% turbostat.C1E% > 1161078 +64.5% 1909611 turbostat.C6 > 44.25 +31.8 76.10 turbostat.C6% > 0.18 -33.3% 0.12 turbostat.IPC > 74338573 =C4=85 2% -33.9% 49159610 =C4=85 4% turbostat.IRQ > 1381661 -91.0% 124075 =C4=85 6% turbostat.POLL > 0.26 -0.2 0.04 =C4=85 12% turbostat.POLL% > 96.15 -5.4% 90.95 turbostat.PkgWatt > 12.12 +19.3% 14.46 turbostat.RAMWatt > 119573 -61.5% 46012 =C4=85 3% proc-vmstat.nr_active_= anon > 106168 +195.8% 314047 =C4=85 10% proc-vmstat.nr_anon_pa= ges > 28.60 =C4=85 17% +1538.5% 468.68 =C4=85 14% proc-vmstat.nr_an= on_transparent_hugepages > 923365 -13.0% 803489 proc-vmstat.nr_file_pages > 165571 +113.5% 353493 =C4=85 9% proc-vmstat.nr_inactiv= e_anon > 45605 -50.2% 22690 proc-vmstat.nr_mapped > 1752 +47.1% 2578 =C4=85 4% proc-vmstat.nr_page_ta= ble_pages > 179613 -66.7% 59728 =C4=85 3% proc-vmstat.nr_shmem > 21490 -2.4% 20981 proc-vmstat.nr_slab_reclaim= able > 28260 -7.3% 26208 proc-vmstat.nr_slab_unrecla= imable > 119573 -61.5% 46012 =C4=85 3% proc-vmstat.nr_zone_ac= tive_anon > 165570 +113.5% 353492 =C4=85 9% proc-vmstat.nr_zone_in= active_anon > 17343640 -76.3% 4116748 =C4=85 4% proc-vmstat.numa_hit > 17364975 -76.3% 4118098 =C4=85 4% proc-vmstat.numa_local > 249252 -66.2% 84187 =C4=85 2% proc-vmstat.pgactivate > 27528916 +567.1% 1.836e+08 =C4=85 5% proc-vmstat.pgalloc_no= rmal > 4912427 -79.2% 1019949 =C4=85 3% proc-vmstat.pgfault > 27227124 +574.1% 1.835e+08 =C4=85 5% proc-vmstat.pgfree > 8728 +3896.4% 348802 =C4=85 5% proc-vmstat.thp_deferr= ed_split_page > 8730 +3895.3% 348814 =C4=85 5% proc-vmstat.thp_fault_= alloc > 8728 +3896.4% 348802 =C4=85 5% proc-vmstat.thp_split_= pmd > 316745 -21.5% 248756 =C4=85 4% sched_debug.cfs_rq:/.a= vg_vruntime.avg > 112735 =C4=85 4% -34.3% 74061 =C4=85 6% sched_debug.cfs_r= q:/.avg_vruntime.min > 0.49 =C4=85 6% -17.2% 0.41 =C4=85 8% sched_debug.cfs_r= q:/.h_nr_running.stddev > 12143 =C4=85120% -99.9% 15.70 =C4=85116% sched_debug.cfs_r= q:/.left_vruntime.avg > 414017 =C4=85126% -99.9% 428.50 =C4=85102% sched_debug.cfs_r= q:/.left_vruntime.max > 68492 =C4=85125% -99.9% 78.15 =C4=85106% sched_debug.cfs_r= q:/.left_vruntime.stddev > 41917 =C4=85 24% -48.3% 21690 =C4=85 57% sched_debug.cfs_r= q:/.load.avg > 176151 =C4=85 30% -56.9% 75963 =C4=85 57% sched_debug.cfs_r= q:/.load.stddev > 6489 =C4=85 17% -29.0% 4608 =C4=85 12% sched_debug.cfs_r= q:/.load_avg.max > 4.42 =C4=85 45% -81.1% 0.83 =C4=85 74% sched_debug.cfs_r= q:/.load_avg.min > 1112 =C4=85 17% -31.0% 767.62 =C4=85 11% sched_debug.cfs_r= q:/.load_avg.stddev > 316745 -21.5% 248756 =C4=85 4% sched_debug.cfs_rq:/.m= in_vruntime.avg > 112735 =C4=85 4% -34.3% 74061 =C4=85 6% sched_debug.cfs_r= q:/.min_vruntime.min > 0.49 =C4=85 6% -17.2% 0.41 =C4=85 8% sched_debug.cfs_r= q:/.nr_running.stddev > 12144 =C4=85120% -99.9% 15.70 =C4=85116% sched_debug.cfs_r= q:/.right_vruntime.avg > 414017 =C4=85126% -99.9% 428.50 =C4=85102% sched_debug.cfs_r= q:/.right_vruntime.max > 68492 =C4=85125% -99.9% 78.15 =C4=85106% sched_debug.cfs_r= q:/.right_vruntime.stddev > 14.25 =C4=85 44% -76.6% 3.33 =C4=85 58% sched_debug.cfs_r= q:/.runnable_avg.min > 11.58 =C4=85 49% -77.7% 2.58 =C4=85 58% sched_debug.cfs_r= q:/.util_avg.min > 423972 =C4=85 23% +59.3% 675379 =C4=85 3% sched_debug.cpu.a= vg_idle.avg > 5720 =C4=85 43% +439.5% 30864 sched_debug.cpu.avg_id= le.min > 99.79 =C4=85 2% -23.7% 76.11 =C4=85 2% sched_debug.cpu.c= lock_task.stddev > 162475 =C4=85 49% -95.8% 6813 =C4=85 26% sched_debug.cpu.c= urr->pid.avg > 1061268 -84.0% 170212 =C4=85 4% sched_debug.cpu.curr->= pid.max > 365404 =C4=85 20% -91.3% 31839 =C4=85 10% sched_debug.cpu.c= urr->pid.stddev > 0.51 =C4=85 3% -20.1% 0.41 =C4=85 9% sched_debug.cpu.n= r_running.stddev > 311923 -74.2% 80615 =C4=85 2% sched_debug.cpu.nr_swi= tches.avg > 565973 =C4=85 4% -77.8% 125597 =C4=85 10% sched_debug.cpu.n= r_switches.max > 192666 =C4=85 4% -70.6% 56695 =C4=85 6% sched_debug.cpu.n= r_switches.min > 67485 =C4=85 8% -79.9% 13558 =C4=85 10% sched_debug.cpu.n= r_switches.stddev > 2.62 +102.1% 5.30 perf-stat.i.MPKI > 2.09e+09 -47.6% 1.095e+09 =C4=85 4% perf-stat.i.branch-ins= tructions > 1.56 -0.5 1.01 perf-stat.i.branch-miss-rat= e% > 31951200 -60.9% 12481432 =C4=85 2% perf-stat.i.branch-mis= ses > 19.38 +23.7 43.08 perf-stat.i.cache-miss-rate= % > 26413597 -5.7% 24899132 =C4=85 4% perf-stat.i.cache-miss= es > 1.363e+08 -58.3% 56906133 =C4=85 4% perf-stat.i.cache-refe= rences > 370628 -75.8% 89743 =C4=85 3% perf-stat.i.context-sw= itches > 1.77 +65.1% 2.92 =C4=85 2% perf-stat.i.cpi > 1.748e+10 -21.8% 1.367e+10 =C4=85 2% perf-stat.i.cpu-cycles > 61611 -79.1% 12901 =C4=85 6% perf-stat.i.cpu-migrat= ions > 716.97 =C4=85 2% -17.2% 593.35 =C4=85 2% perf-stat.i.cycle= s-between-cache-misses > 0.12 =C4=85 4% -0.1 0.05 perf-stat.i.dTLB-load-= miss-rate% > 3066100 =C4=85 3% -81.3% 573066 =C4=85 5% perf-stat.i.dTLB-= load-misses > 2.652e+09 -50.1% 1.324e+09 =C4=85 4% perf-stat.i.dTLB-loads > 0.08 =C4=85 2% -0.0 0.03 perf-stat.i.dTLB-store= -miss-rate% > 1168195 =C4=85 2% -82.9% 199438 =C4=85 5% perf-stat.i.dTLB-= store-misses > 1.478e+09 -56.8% 6.384e+08 =C4=85 3% perf-stat.i.dTLB-store= s > 8080423 -73.2% 2169371 =C4=85 3% perf-stat.i.iTLB-load-= misses > 5601321 -74.3% 1440571 =C4=85 2% perf-stat.i.iTLB-loads > 1.028e+10 -49.7% 5.173e+09 =C4=85 4% perf-stat.i.instructio= ns > 1450 +73.1% 2511 =C4=85 2% perf-stat.i.instructio= ns-per-iTLB-miss > 0.61 -35.9% 0.39 perf-stat.i.ipc > 0.48 -21.4% 0.38 =C4=85 2% perf-stat.i.metric.GHz > 616.28 -17.6% 507.69 =C4=85 4% perf-stat.i.metric.K/s= ec > 175.16 -50.8% 86.18 =C4=85 4% perf-stat.i.metric.M/s= ec > 76728 -80.8% 14724 =C4=85 4% perf-stat.i.minor-faul= ts > 5600408 -61.4% 2160997 =C4=85 5% perf-stat.i.node-loads > 8873996 +52.1% 13499744 =C4=85 5% perf-stat.i.node-store= s > 112409 -81.9% 20305 =C4=85 4% perf-stat.i.page-fault= s > 2.55 +89.6% 4.83 perf-stat.overall.MPKI Much more TLB misses. > 1.51 -0.4 1.13 perf-stat.overall.branch-mi= ss-rate% > 19.26 +24.5 43.71 perf-stat.overall.cache-mis= s-rate% > 1.70 +56.4% 2.65 perf-stat.overall.cpi > 665.84 -17.5% 549.51 =C4=85 2% perf-stat.overall.cycl= es-between-cache-misses > 0.12 =C4=85 4% -0.1 0.04 perf-stat.overall.dTLB= -load-miss-rate% > 0.08 =C4=85 2% -0.0 0.03 perf-stat.overall.dTLB= -store-miss-rate% > 59.16 +0.9 60.04 perf-stat.overall.iTLB-load= -miss-rate% > 1278 +86.1% 2379 =C4=85 2% perf-stat.overall.inst= ructions-per-iTLB-miss > 0.59 -36.1% 0.38 perf-stat.overall.ipc Worse IPC and CPI. > 2.078e+09 -48.3% 1.074e+09 =C4=85 4% perf-stat.ps.branch-in= structions > 31292687 -61.2% 12133349 =C4=85 2% perf-stat.ps.branch-mi= sses > 26057291 -5.9% 24512034 =C4=85 4% perf-stat.ps.cache-mis= ses > 1.353e+08 -58.6% 56072195 =C4=85 4% perf-stat.ps.cache-ref= erences > 365254 -75.8% 88464 =C4=85 3% perf-stat.ps.context-s= witches > 1.735e+10 -22.4% 1.346e+10 =C4=85 2% perf-stat.ps.cpu-cycle= s > 60838 -79.1% 12727 =C4=85 6% perf-stat.ps.cpu-migra= tions > 3056601 =C4=85 4% -81.5% 565354 =C4=85 4% perf-stat.ps.dTLB= -load-misses > 2.636e+09 -50.7% 1.3e+09 =C4=85 4% perf-stat.ps.dTLB-load= s > 1155253 =C4=85 2% -83.0% 196581 =C4=85 5% perf-stat.ps.dTLB= -store-misses > 1.473e+09 -57.4% 6.268e+08 =C4=85 3% perf-stat.ps.dTLB-stor= es > 7997726 -73.3% 2131477 =C4=85 3% perf-stat.ps.iTLB-load= -misses > 5521346 -74.3% 1418623 =C4=85 2% perf-stat.ps.iTLB-load= s > 1.023e+10 -50.4% 5.073e+09 =C4=85 4% perf-stat.ps.instructi= ons > 75671 -80.9% 14479 =C4=85 4% perf-stat.ps.minor-fau= lts > 5549722 -61.4% 2141750 =C4=85 4% perf-stat.ps.node-load= s > 8769156 +51.6% 13296579 =C4=85 5% perf-stat.ps.node-stor= es > 110795 -82.0% 19977 =C4=85 4% perf-stat.ps.page-faul= ts > 6.482e+11 -50.7% 3.197e+11 =C4=85 4% perf-stat.total.instru= ctions > 0.00 =C4=85 37% -100.0% 0.00 perf-sched.sch_delay.a= vg.ms.__cond_resched.__kmem_cache_alloc_node.__kmalloc_node.memcg_alloc_sla= b_cgroups.allocate_slab > 0.01 =C4=85 18% +8373.1% 0.73 =C4=85 49% perf-sched.sch_de= lay.avg.ms.__cond_resched.down_read.do_madvise.__x64_sys_madvise.do_syscall= _64 > 0.01 =C4=85 16% +4600.0% 0.38 =C4=85 24% perf-sched.sch_de= lay.avg.ms.__cond_resched.down_read.exit_mm.do_exit.__x64_sys_exit More time spent in madvise and munmap. but I'm not sure whether this is caused by tearing down the address space when exiting the test. If so it should not count in the regression. > 0.01 =C4=85204% -100.0% 0.00 perf-sched.sch_delay.a= vg.ms.__cond_resched.down_write.do_vmi_align_munmap.do_vmi_munmap.__vm_munm= ap > 0.01 =C4=85 8% +3678.9% 0.36 =C4=85 79% perf-sched.sch_de= lay.avg.ms.__cond_resched.exit_signals.do_exit.__x64_sys_exit.do_syscall_64 > 0.01 =C4=85 14% -38.5% 0.00 perf-sched.sch_delay.a= vg.ms.__cond_resched.kmem_cache_alloc.security_file_alloc.init_file.alloc_e= mpty_file > 0.01 =C4=85 5% +2946.2% 0.26 =C4=85 43% perf-sched.sch_de= lay.avg.ms.__cond_resched.mutex_lock.futex_exit_release.exit_mm_release.exi= t_mm > 0.00 =C4=85 14% +125.0% 0.01 =C4=85 12% perf-sched.sch_de= lay.avg.ms.__cond_resched.smpboot_thread_fn.kthread.ret_from_fork.ret_from_= fork_asm > 0.02 =C4=85170% -83.0% 0.00 perf-sched.sch_delay.a= vg.ms.__cond_resched.switch_task_namespaces.__do_sys_setns.do_syscall_64.en= try_SYSCALL_64_after_hwframe > 0.00 =C4=85 69% +6578.6% 0.31 =C4=85 4% perf-sched.sch_de= lay.avg.ms.__cond_resched.tlb_batch_pages_flush.tlb_finish_mmu.zap_page_ran= ge_single.madvise_vma_behavior > 0.00 +100.0% 0.00 perf-sched.sch_delay.avg.ms= .__cond_resched.unmap_vmas.unmap_region.do_vmi_align_munmap.do_vmi_munmap > 0.02 =C4=85 86% +4234.4% 0.65 =C4=85 4% perf-sched.sch_de= lay.avg.ms.__cond_resched.zap_page_range_single.madvise_vma_behavior.do_mad= vise.__x64_sys_madvise > 0.01 =C4=85 6% +6054.3% 0.47 perf-sched.sch_delay.a= vg.ms.__cond_resched.zap_pmd_range.isra.0.unmap_page_range > 0.00 =C4=85 14% +195.2% 0.01 =C4=85 89% perf-sched.sch_de= lay.avg.ms.__x64_sys_pause.do_syscall_64.entry_SYSCALL_64_after_hwframe.[un= known] > 0.00 =C4=85102% +340.0% 0.01 =C4=85 85% perf-sched.sch_de= lay.avg.ms.devkmsg_read.vfs_read.ksys_read.do_syscall_64 > 0.00 +100.0% 0.00 perf-sched.sch_delay.avg.ms= .do_nanosleep.hrtimer_nanosleep.common_nsleep.__x64_sys_clock_nanosleep > 0.00 =C4=85 11% +66.7% 0.01 =C4=85 21% perf-sched.sch_de= lay.avg.ms.do_wait.kernel_wait4.__do_sys_wait4.do_syscall_64 > 0.01 =C4=85 89% +1096.1% 0.15 =C4=85 30% perf-sched.sch_de= lay.avg.ms.exit_to_user_mode_loop.exit_to_user_mode_prepare.irqentry_exit_t= o_user_mode.asm_sysvec_reschedule_ipi > 0.00 +141.7% 0.01 =C4=85 61% perf-sched.sch_delay.a= vg.ms.pipe_read.vfs_read.ksys_read.do_syscall_64 > 0.00 =C4=85223% +9975.0% 0.07 =C4=85203% perf-sched.sch_de= lay.avg.ms.schedule_hrtimeout_range_clock.do_select.core_sys_select.kern_se= lect > 0.00 =C4=85 10% +789.3% 0.04 =C4=85 69% perf-sched.sch_de= lay.avg.ms.schedule_hrtimeout_range_clock.ep_poll.do_epoll_wait.__x64_sys_e= poll_wait > 0.00 =C4=85 31% +6691.3% 0.26 =C4=85 5% perf-sched.sch_de= lay.avg.ms.schedule_preempt_disabled.rwsem_down_read_slowpath.down_read.do_= madvise > 0.00 =C4=85 28% +14612.5% 0.59 =C4=85 4% perf-sched.sch_de= lay.avg.ms.schedule_preempt_disabled.rwsem_down_read_slowpath.down_read.exi= t_mm > 0.00 =C4=85 24% +4904.2% 0.20 =C4=85 4% perf-sched.sch_de= lay.avg.ms.schedule_preempt_disabled.rwsem_down_read_slowpath.down_read_kil= lable.lock_mm_and_find_vma > 0.00 =C4=85 28% +450.0% 0.01 =C4=85 74% perf-sched.sch_de= lay.avg.ms.schedule_preempt_disabled.rwsem_down_write_slowpath.down_write_k= illable.__vm_munmap > 0.00 =C4=85 17% +984.6% 0.02 =C4=85 79% perf-sched.sch_de= lay.avg.ms.schedule_timeout.__wait_for_common.wait_for_completion_state.ker= nel_clone > 0.00 =C4=85 20% +231.8% 0.01 =C4=85 89% perf-sched.sch_de= lay.avg.ms.schedule_timeout.io_schedule_timeout.__wait_for_common.submit_bi= o_wait > 0.00 +350.0% 0.01 =C4=85 16% perf-sched.sch_delay.a= vg.ms.schedule_timeout.rcu_gp_fqs_loop.rcu_gp_kthread.kthread > 0.02 =C4=85 16% +320.2% 0.07 =C4=85 2% perf-sched.sch_de= lay.avg.ms.smpboot_thread_fn.kthread.ret_from_fork.ret_from_fork_asm > 0.02 =C4=85 2% +282.1% 0.09 =C4=85 5% perf-sched.sch_de= lay.avg.ms.worker_thread.kthread.ret_from_fork.ret_from_fork_asm > 0.00 =C4=85 14% -100.0% 0.00 perf-sched.sch_delay.m= ax.ms.__cond_resched.__kmem_cache_alloc_node.__kmalloc_node.memcg_alloc_sla= b_cgroups.allocate_slab > 0.05 =C4=85 35% +3784.5% 1.92 =C4=85 16% perf-sched.sch_de= lay.max.ms.__cond_resched.down_read.do_madvise.__x64_sys_madvise.do_syscall= _64 > 0.29 =C4=85128% +563.3% 1.92 =C4=85 7% perf-sched.sch_de= lay.max.ms.__cond_resched.down_read.exit_mm.do_exit.__x64_sys_exit > 0.14 =C4=85217% -99.7% 0.00 =C4=85223% perf-sched.sch_de= lay.max.ms.__cond_resched.down_write.do_vmi_align_munmap.do_vmi_munmap.__vm= _munmap > 0.03 =C4=85 49% -74.0% 0.01 =C4=85 51% perf-sched.sch_de= lay.max.ms.__cond_resched.dput.__fput.__x64_sys_close.do_syscall_64 > 0.01 =C4=85 54% -57.4% 0.00 =C4=85 75% perf-sched.sch_de= lay.max.ms.__cond_resched.dput.__ns_get_path.ns_get_path.proc_ns_get_link > 0.12 =C4=85 21% +873.0% 1.19 =C4=85 60% perf-sched.sch_de= lay.max.ms.__cond_resched.exit_signals.do_exit.__x64_sys_exit.do_syscall_64 > 2.27 =C4=85220% -99.7% 0.01 =C4=85 19% perf-sched.sch_de= lay.max.ms.__cond_resched.kmem_cache_alloc.create_new_namespaces.__do_sys_s= etns.do_syscall_64 > 0.02 =C4=85 36% -54.4% 0.01 =C4=85 55% perf-sched.sch_de= lay.max.ms.__cond_resched.kmem_cache_alloc.getname_flags.part.0 > 0.04 =C4=85 36% -77.1% 0.01 =C4=85 31% perf-sched.sch_de= lay.max.ms.__cond_resched.kmem_cache_alloc.security_file_alloc.init_file.al= loc_empty_file > 0.12 =C4=85 32% +1235.8% 1.58 =C4=85 31% perf-sched.sch_de= lay.max.ms.__cond_resched.mutex_lock.futex_exit_release.exit_mm_release.exi= t_mm > 2.25 =C4=85218% -99.3% 0.02 =C4=85 52% perf-sched.sch_de= lay.max.ms.__cond_resched.switch_task_namespaces.__do_sys_setns.do_syscall_= 64.entry_SYSCALL_64_after_hwframe > 0.01 =C4=85 85% +19836.4% 2.56 =C4=85 7% perf-sched.sch_de= lay.max.ms.__cond_resched.tlb_batch_pages_flush.tlb_finish_mmu.zap_page_ran= ge_single.madvise_vma_behavior > 0.03 =C4=85 70% -93.6% 0.00 =C4=85223% perf-sched.sch_de= lay.max.ms.__cond_resched.unmap_page_range.zap_page_range_single.madvise_vm= a_behavior.do_madvise > 0.10 =C4=85 16% +2984.2% 3.21 =C4=85 6% perf-sched.sch_de= lay.max.ms.__cond_resched.zap_pmd_range.isra.0.unmap_page_range > 0.01 =C4=85 20% +883.9% 0.05 =C4=85177% perf-sched.sch_de= lay.max.ms.__x64_sys_pause.do_syscall_64.entry_SYSCALL_64_after_hwframe.[un= known] > 0.01 =C4=85 15% +694.7% 0.08 =C4=85123% perf-sched.sch_de= lay.max.ms.do_wait.kernel_wait4.__do_sys_wait4.do_syscall_64 > 0.00 =C4=85223% +6966.7% 0.07 =C4=85199% perf-sched.sch_de= lay.max.ms.schedule_hrtimeout_range_clock.do_select.core_sys_select.kern_se= lect > 0.01 =C4=85 38% +8384.6% 0.55 =C4=85 72% perf-sched.sch_de= lay.max.ms.schedule_timeout.__wait_for_common.wait_for_completion_state.ker= nel_clone > 0.01 =C4=85 13% +12995.7% 1.51 =C4=85103% perf-sched.sch_de= lay.max.ms.schedule_timeout.rcu_gp_fqs_loop.rcu_gp_kthread.kthread > 117.80 =C4=85 56% -96.4% 4.26 =C4=85 36% perf-sched.sch_de= lay.max.ms.smpboot_thread_fn.kthread.ret_from_fork.ret_from_fork_asm > 0.01 =C4=85 68% +331.9% 0.03 perf-sched.total_sch_d= elay.average.ms > 4.14 +242.6% 14.20 =C4=85 4% perf-sched.total_wait_= and_delay.average.ms > 700841 -69.6% 212977 =C4=85 3% perf-sched.total_wait_= and_delay.count.ms > 4.14 +242.4% 14.16 =C4=85 4% perf-sched.total_wait_= time.average.ms > 11.68 =C4=85 8% +213.3% 36.59 =C4=85 28% perf-sched.wait_a= nd_delay.avg.ms.__cond_resched.apparmor_file_alloc_security.security_file_a= lloc.init_file.alloc_empty_file > 10.00 =C4=85 2% +226.1% 32.62 =C4=85 20% perf-sched.wait_a= nd_delay.avg.ms.__cond_resched.dentry_kill.dput.__fput.__x64_sys_close > 10.55 =C4=85 3% +259.8% 37.96 =C4=85 7% perf-sched.wait_a= nd_delay.avg.ms.__cond_resched.dput.nd_jump_link.proc_ns_get_link.pick_link > 9.80 =C4=85 12% +196.5% 29.07 =C4=85 32% perf-sched.wait_a= nd_delay.avg.ms.__cond_resched.dput.pick_link.step_into.open_last_lookups > 9.80 =C4=85 4% +234.9% 32.83 =C4=85 14% perf-sched.wait_a= nd_delay.avg.ms.__cond_resched.dput.terminate_walk.path_openat.do_filp_open > 10.32 =C4=85 2% +223.8% 33.42 =C4=85 6% perf-sched.wait_a= nd_delay.avg.ms.__cond_resched.kmem_cache_alloc.alloc_empty_file.path_opena= t.do_filp_open > 8.15 =C4=85 14% +271.3% 30.25 =C4=85 35% perf-sched.wait_a= nd_delay.avg.ms.__cond_resched.kmem_cache_alloc.create_new_namespaces.__do_= sys_setns.do_syscall_64 > 9.60 =C4=85 4% +240.8% 32.73 =C4=85 16% perf-sched.wait_a= nd_delay.avg.ms.__cond_resched.kmem_cache_alloc.getname_flags.part.0 > 10.37 =C4=85 4% +232.0% 34.41 =C4=85 10% perf-sched.wait_a= nd_delay.avg.ms.__cond_resched.kmem_cache_alloc.security_file_alloc.init_fi= le.alloc_empty_file > 7.32 =C4=85 46% +269.7% 27.07 =C4=85 49% perf-sched.wait_a= nd_delay.avg.ms.__cond_resched.shmem_inode_acct_blocks.shmem_alloc_and_add_= folio.shmem_get_folio_gfp.shmem_write_begin > 9.88 +236.2% 33.23 =C4=85 4% perf-sched.wait_and_de= lay.avg.ms.__cond_resched.slab_pre_alloc_hook.constprop.0.kmem_cache_alloc_= lru > 4.44 =C4=85 4% +379.0% 21.27 =C4=85 18% perf-sched.wait_a= nd_delay.avg.ms.__cond_resched.smpboot_thread_fn.kthread.ret_from_fork.ret_= from_fork_asm > 10.05 =C4=85 2% +235.6% 33.73 =C4=85 11% perf-sched.wait_a= nd_delay.avg.ms.__cond_resched.switch_task_namespaces.__do_sys_setns.do_sys= call_64.entry_SYSCALL_64_after_hwframe > 0.03 +462.6% 0.15 =C4=85 6% perf-sched.wait_and_de= lay.avg.ms.do_task_dead.do_exit.__x64_sys_exit.do_syscall_64.entry_SYSCALL_= 64_after_hwframe > 6.78 =C4=85 4% +482.1% 39.46 =C4=85 3% perf-sched.wait_a= nd_delay.avg.ms.exit_to_user_mode_loop.exit_to_user_mode_prepare.syscall_ex= it_to_user_mode.do_syscall_64 > 3.17 +683.3% 24.85 =C4=85 8% perf-sched.wait_and_de= lay.avg.ms.futex_wait_queue.__futex_wait.futex_wait.do_futex > 36.64 =C4=85 13% +244.7% 126.32 =C4=85 6% perf-sched.wait_a= nd_delay.avg.ms.schedule_hrtimeout_range_clock.do_poll.constprop.0.do_sys_p= oll > 9.81 +302.4% 39.47 =C4=85 4% perf-sched.wait_and_de= lay.avg.ms.schedule_hrtimeout_range_clock.do_sigtimedwait.__x64_sys_rt_sigt= imedwait.do_syscall_64 > 1.05 +48.2% 1.56 perf-sched.wait_and_delay.a= vg.ms.schedule_preempt_disabled.rwsem_down_read_slowpath.down_read.do_madvi= se > 0.93 +14.2% 1.06 =C4=85 2% perf-sched.wait_and_de= lay.avg.ms.schedule_preempt_disabled.rwsem_down_read_slowpath.down_read_kil= lable.lock_mm_and_find_vma > 9.93 -100.0% 0.00 perf-sched.wait_and_delay.a= vg.ms.schedule_timeout.ext4_lazyinit_thread.part.0.kthread > 12.02 =C4=85 3% +139.8% 28.83 =C4=85 6% perf-sched.wait_a= nd_delay.avg.ms.smpboot_thread_fn.kthread.ret_from_fork.ret_from_fork_asm > 6.09 =C4=85 2% +403.0% 30.64 =C4=85 5% perf-sched.wait_a= nd_delay.avg.ms.worker_thread.kthread.ret_from_fork.ret_from_fork_asm > 23.17 =C4=85 19% -83.5% 3.83 =C4=85143% perf-sched.wait_a= nd_delay.count.__cond_resched.__alloc_pages.alloc_pages_mpol.shmem_alloc_fo= lio.shmem_alloc_and_add_folio > 79.83 =C4=85 9% -55.1% 35.83 =C4=85 16% perf-sched.wait_a= nd_delay.count.__cond_resched.dentry_kill.dput.__fput.__x64_sys_close > 14.83 =C4=85 14% -59.6% 6.00 =C4=85 56% perf-sched.wait_a= nd_delay.count.__cond_resched.dput.__fput.__x64_sys_close.do_syscall_64 > 8.50 =C4=85 17% -80.4% 1.67 =C4=85 89% perf-sched.wait_a= nd_delay.count.__cond_resched.dput.__ns_get_path.ns_get_path.proc_ns_get_li= nk > 114.00 =C4=85 14% -62.4% 42.83 =C4=85 11% perf-sched.wait_a= nd_delay.count.__cond_resched.dput.nd_jump_link.proc_ns_get_link.pick_link > 94.67 =C4=85 7% -48.1% 49.17 =C4=85 13% perf-sched.wait_a= nd_delay.count.__cond_resched.dput.terminate_walk.path_openat.do_filp_open > 59.83 =C4=85 13% -76.0% 14.33 =C4=85 48% perf-sched.wait_a= nd_delay.count.__cond_resched.generic_perform_write.shmem_file_write_iter.v= fs_write.ksys_write > 103.00 =C4=85 12% -48.1% 53.50 =C4=85 20% perf-sched.wait_a= nd_delay.count.__cond_resched.kmem_cache_alloc.alloc_empty_file.path_openat= .do_filp_open > 19.33 =C4=85 16% -56.0% 8.50 =C4=85 29% perf-sched.wait_a= nd_delay.count.__cond_resched.kmem_cache_alloc.create_new_namespaces.__do_s= ys_setns.do_syscall_64 > 68.17 =C4=85 11% -39.1% 41.50 =C4=85 19% perf-sched.wait_a= nd_delay.count.__cond_resched.kmem_cache_alloc.security_file_alloc.init_fil= e.alloc_empty_file > 36.67 =C4=85 22% -79.1% 7.67 =C4=85 46% perf-sched.wait_a= nd_delay.count.__cond_resched.mutex_lock.perf_poll.do_poll.constprop > 465.50 =C4=85 9% -47.4% 244.83 =C4=85 11% perf-sched.wait_a= nd_delay.count.__cond_resched.slab_pre_alloc_hook.constprop.0.kmem_cache_al= loc_lru > 14492 =C4=85 3% -96.3% 533.67 =C4=85 10% perf-sched.wait_a= nd_delay.count.__cond_resched.smpboot_thread_fn.kthread.ret_from_fork.ret_f= rom_fork_asm > 128.67 =C4=85 7% -53.5% 59.83 =C4=85 10% perf-sched.wait_a= nd_delay.count.__cond_resched.switch_task_namespaces.__do_sys_setns.do_sysc= all_64.entry_SYSCALL_64_after_hwframe > 7.67 =C4=85 34% -80.4% 1.50 =C4=85107% perf-sched.wait_a= nd_delay.count.__cond_resched.vunmap_p4d_range.__vunmap_range_noflush.remov= e_vm_area.vfree > 147533 -81.0% 28023 =C4=85 5% perf-sched.wait_and_de= lay.count.do_task_dead.do_exit.__x64_sys_exit.do_syscall_64.entry_SYSCALL_6= 4_after_hwframe > 4394 =C4=85 4% -78.5% 942.83 =C4=85 7% perf-sched.wait_a= nd_delay.count.exit_to_user_mode_loop.exit_to_user_mode_prepare.syscall_exi= t_to_user_mode.do_syscall_64 > 228791 -79.3% 47383 =C4=85 4% perf-sched.wait_and_de= lay.count.futex_wait_queue.__futex_wait.futex_wait.do_futex > 368.50 =C4=85 2% -67.1% 121.33 =C4=85 3% perf-sched.wait_a= nd_delay.count.schedule_hrtimeout_range_clock.do_poll.constprop.0.do_sys_po= ll > 147506 -81.0% 28010 =C4=85 5% perf-sched.wait_and_de= lay.count.schedule_hrtimeout_range_clock.do_sigtimedwait.__x64_sys_rt_sigti= medwait.do_syscall_64 > 5387 =C4=85 6% -16.7% 4488 =C4=85 5% perf-sched.wait_a= nd_delay.count.schedule_preempt_disabled.rwsem_down_read_slowpath.down_read= .do_madvise > 8303 =C4=85 2% -56.9% 3579 =C4=85 5% perf-sched.wait_a= nd_delay.count.schedule_preempt_disabled.rwsem_down_read_slowpath.down_read= _killable.lock_mm_and_find_vma > 14.67 =C4=85 7% -100.0% 0.00 perf-sched.wait_and_de= lay.count.schedule_timeout.ext4_lazyinit_thread.part.0.kthread > 370.50 =C4=85141% +221.9% 1192 =C4=85 5% perf-sched.wait_a= nd_delay.count.schedule_timeout.rcu_gp_fqs_loop.rcu_gp_kthread.kthread > 24395 =C4=85 2% -51.2% 11914 =C4=85 6% perf-sched.wait_a= nd_delay.count.smpboot_thread_fn.kthread.ret_from_fork.ret_from_fork_asm > 31053 =C4=85 2% -80.5% 6047 =C4=85 5% perf-sched.wait_a= nd_delay.count.worker_thread.kthread.ret_from_fork.ret_from_fork_asm > 16.41 =C4=85 2% +342.7% 72.65 =C4=85 29% perf-sched.wait_a= nd_delay.max.ms.__cond_resched.apparmor_file_alloc_security.security_file_a= lloc.init_file.alloc_empty_file > 16.49 =C4=85 3% +463.3% 92.90 =C4=85 27% perf-sched.wait_a= nd_delay.max.ms.__cond_resched.dentry_kill.dput.__fput.__x64_sys_close > 17.32 =C4=85 5% +520.9% 107.52 =C4=85 14% perf-sched.wait_a= nd_delay.max.ms.__cond_resched.dput.nd_jump_link.proc_ns_get_link.pick_link > 15.38 =C4=85 6% +325.2% 65.41 =C4=85 22% perf-sched.wait_a= nd_delay.max.ms.__cond_resched.dput.pick_link.step_into.open_last_lookups > 16.73 =C4=85 4% +456.2% 93.04 =C4=85 11% perf-sched.wait_a= nd_delay.max.ms.__cond_resched.dput.terminate_walk.path_openat.do_filp_open > 17.14 =C4=85 3% +510.6% 104.68 =C4=85 14% perf-sched.wait_a= nd_delay.max.ms.__cond_resched.kmem_cache_alloc.alloc_empty_file.path_opena= t.do_filp_open > 15.70 =C4=85 4% +379.4% 75.25 =C4=85 28% perf-sched.wait_a= nd_delay.max.ms.__cond_resched.kmem_cache_alloc.create_new_namespaces.__do_= sys_setns.do_syscall_64 > 15.70 =C4=85 3% +422.1% 81.97 =C4=85 19% perf-sched.wait_a= nd_delay.max.ms.__cond_resched.kmem_cache_alloc.getname_flags.part.0 > 16.38 +528.4% 102.91 =C4=85 21% perf-sched.wait_and_de= lay.max.ms.__cond_resched.kmem_cache_alloc.security_file_alloc.init_file.al= loc_empty_file > 45.20 =C4=85 48% +166.0% 120.23 =C4=85 27% perf-sched.wait_a= nd_delay.max.ms.__cond_resched.shmem_inode_acct_blocks.shmem_alloc_and_add_= folio.shmem_get_folio_gfp.shmem_write_begin > 17.25 +495.5% 102.71 =C4=85 2% perf-sched.wait_and_de= lay.max.ms.__cond_resched.slab_pre_alloc_hook.constprop.0.kmem_cache_alloc_= lru > 402.57 =C4=85 15% -52.8% 189.90 =C4=85 14% perf-sched.wait_a= nd_delay.max.ms.__cond_resched.smpboot_thread_fn.kthread.ret_from_fork.ret_= from_fork_asm > 16.96 =C4=85 4% +521.3% 105.40 =C4=85 15% perf-sched.wait_a= nd_delay.max.ms.__cond_resched.switch_task_namespaces.__do_sys_setns.do_sys= call_64.entry_SYSCALL_64_after_hwframe > 28.45 +517.3% 175.65 =C4=85 14% perf-sched.wait_and_de= lay.max.ms.do_task_dead.do_exit.__x64_sys_exit.do_syscall_64.entry_SYSCALL_= 64_after_hwframe > 22.49 +628.5% 163.83 =C4=85 16% perf-sched.wait_and_de= lay.max.ms.futex_wait_queue.__futex_wait.futex_wait.do_futex > 26.53 =C4=85 30% +326.9% 113.25 =C4=85 16% perf-sched.wait_a= nd_delay.max.ms.schedule_hrtimeout_range_clock.do_sigtimedwait.__x64_sys_rt= _sigtimedwait.do_syscall_64 > 15.54 -100.0% 0.00 perf-sched.wait_and_delay.m= ax.ms.schedule_timeout.ext4_lazyinit_thread.part.0.kthread > 1.67 =C4=85141% +284.6% 6.44 =C4=85 4% perf-sched.wait_a= nd_delay.max.ms.schedule_timeout.rcu_gp_fqs_loop.rcu_gp_kthread.kthread > 0.07 =C4=85 34% -93.6% 0.00 =C4=85105% perf-sched.wait_t= ime.avg.ms.__cond_resched.__alloc_pages.alloc_pages_mpol.pte_alloc_one.__pt= e_alloc > 10.21 =C4=85 15% +295.8% 40.43 =C4=85 50% perf-sched.wait_t= ime.avg.ms.__cond_resched.__fput.__x64_sys_close.do_syscall_64.entry_SYSCAL= L_64_after_hwframe > 3.89 =C4=85 40% -99.8% 0.01 =C4=85113% perf-sched.wait_t= ime.avg.ms.__cond_resched.__kmem_cache_alloc_node.__kmalloc_node.memcg_allo= c_slab_cgroups.allocate_slab > 11.67 =C4=85 8% +213.5% 36.58 =C4=85 28% perf-sched.wait_t= ime.avg.ms.__cond_resched.apparmor_file_alloc_security.security_file_alloc.= init_file.alloc_empty_file > 9.98 =C4=85 2% +226.8% 32.61 =C4=85 20% perf-sched.wait_t= ime.avg.ms.__cond_resched.dentry_kill.dput.__fput.__x64_sys_close > 1.03 +71.2% 1.77 =C4=85 20% perf-sched.wait_time.a= vg.ms.__cond_resched.down_read.do_madvise.__x64_sys_madvise.do_syscall_64 > 0.06 =C4=85 79% -100.0% 0.00 perf-sched.wait_time.a= vg.ms.__cond_resched.down_write.__split_vma.vma_modify.mprotect_fixup > 0.05 =C4=85 22% -100.0% 0.00 perf-sched.wait_time.a= vg.ms.__cond_resched.down_write.vma_expand.mmap_region.do_mmap > 0.08 =C4=85 82% -98.2% 0.00 =C4=85223% perf-sched.wait_t= ime.avg.ms.__cond_resched.down_write_killable.vm_mmap_pgoff.do_syscall_64.e= ntry_SYSCALL_64_after_hwframe > 10.72 =C4=85 10% +166.9% 28.61 =C4=85 29% perf-sched.wait_t= ime.avg.ms.__cond_resched.dput.__fput.__x64_sys_close.do_syscall_64 > 10.53 =C4=85 3% +260.5% 37.95 =C4=85 7% perf-sched.wait_t= ime.avg.ms.__cond_resched.dput.nd_jump_link.proc_ns_get_link.pick_link > 9.80 =C4=85 12% +196.6% 29.06 =C4=85 32% perf-sched.wait_t= ime.avg.ms.__cond_resched.dput.pick_link.step_into.open_last_lookups > 9.80 =C4=85 4% +235.1% 32.82 =C4=85 14% perf-sched.wait_t= ime.avg.ms.__cond_resched.dput.terminate_walk.path_openat.do_filp_open > 9.50 =C4=85 12% +281.9% 36.27 =C4=85 70% perf-sched.wait_t= ime.avg.ms.__cond_resched.generic_perform_write.shmem_file_write_iter.vfs_w= rite.ksys_write > 10.31 =C4=85 2% +223.9% 33.40 =C4=85 6% perf-sched.wait_t= ime.avg.ms.__cond_resched.kmem_cache_alloc.alloc_empty_file.path_openat.do_= filp_open > 8.04 =C4=85 15% +276.1% 30.25 =C4=85 35% perf-sched.wait_t= ime.avg.ms.__cond_resched.kmem_cache_alloc.create_new_namespaces.__do_sys_s= etns.do_syscall_64 > 9.60 =C4=85 4% +240.9% 32.72 =C4=85 16% perf-sched.wait_t= ime.avg.ms.__cond_resched.kmem_cache_alloc.getname_flags.part.0 > 0.06 =C4=85 66% -98.3% 0.00 =C4=85223% perf-sched.wait_t= ime.avg.ms.__cond_resched.kmem_cache_alloc.mas_alloc_nodes.mas_preallocate.= __split_vma > 10.36 =C4=85 4% +232.1% 34.41 =C4=85 10% perf-sched.wait_t= ime.avg.ms.__cond_resched.kmem_cache_alloc.security_file_alloc.init_file.al= loc_empty_file > 0.08 =C4=85 50% -95.7% 0.00 =C4=85100% perf-sched.wait_t= ime.avg.ms.__cond_resched.kmem_cache_alloc.vm_area_dup.__split_vma.vma_modi= fy > 0.01 =C4=85 49% -100.0% 0.00 perf-sched.wait_time.a= vg.ms.__cond_resched.kmem_cache_alloc_node.alloc_vmap_area.__get_vm_area_no= de.__vmalloc_node_range > 0.03 =C4=85 73% -87.4% 0.00 =C4=85145% perf-sched.wait_t= ime.avg.ms.__cond_resched.kmem_cache_alloc_node.dup_task_struct.copy_proces= s.kernel_clone > 8.01 =C4=85 25% +238.0% 27.07 =C4=85 49% perf-sched.wait_t= ime.avg.ms.__cond_resched.shmem_inode_acct_blocks.shmem_alloc_and_add_folio= .shmem_get_folio_gfp.shmem_write_begin > 9.86 +237.0% 33.23 =C4=85 4% perf-sched.wait_time.a= vg.ms.__cond_resched.slab_pre_alloc_hook.constprop.0.kmem_cache_alloc_lru > 4.44 =C4=85 4% +379.2% 21.26 =C4=85 18% perf-sched.wait_t= ime.avg.ms.__cond_resched.smpboot_thread_fn.kthread.ret_from_fork.ret_from_= fork_asm > 10.03 +236.3% 33.73 =C4=85 11% perf-sched.wait_time.a= vg.ms.__cond_resched.switch_task_namespaces.__do_sys_setns.do_syscall_64.en= try_SYSCALL_64_after_hwframe > 0.97 =C4=85 8% -87.8% 0.12 =C4=85221% perf-sched.wait_t= ime.avg.ms.__cond_resched.unmap_page_range.zap_page_range_single.madvise_vm= a_behavior.do_madvise > 0.02 =C4=85 13% +1846.8% 0.45 =C4=85 11% perf-sched.wait_t= ime.avg.ms.__cond_resched.unmap_vmas.unmap_region.do_vmi_align_munmap.do_vm= i_munmap > 1.01 +64.7% 1.66 perf-sched.wait_time.avg.ms= .__cond_resched.zap_page_range_single.madvise_vma_behavior.do_madvise.__x64= _sys_madvise > 0.75 =C4=85 4% +852.1% 7.10 =C4=85 5% perf-sched.wait_t= ime.avg.ms.__x64_sys_sched_yield.do_syscall_64.entry_SYSCALL_64_after_hwfra= me.[unknown] > 0.03 +462.6% 0.15 =C4=85 6% perf-sched.wait_time.a= vg.ms.do_task_dead.do_exit.__x64_sys_exit.do_syscall_64.entry_SYSCALL_64_af= ter_hwframe > 0.24 =C4=85 4% +25.3% 0.30 =C4=85 8% perf-sched.wait_t= ime.avg.ms.do_wait.kernel_wait4.__do_sys_wait4.do_syscall_64 > 1.98 =C4=85 15% +595.7% 13.80 =C4=85 90% perf-sched.wait_t= ime.avg.ms.exit_to_user_mode_loop.exit_to_user_mode_prepare.irqentry_exit_t= o_user_mode.asm_sysvec_apic_timer_interrupt > 2.78 =C4=85 14% +444.7% 15.12 =C4=85 16% perf-sched.wait_t= ime.avg.ms.exit_to_user_mode_loop.exit_to_user_mode_prepare.irqentry_exit_t= o_user_mode.asm_sysvec_call_function > 6.77 =C4=85 4% +483.0% 39.44 =C4=85 3% perf-sched.wait_t= ime.avg.ms.exit_to_user_mode_loop.exit_to_user_mode_prepare.syscall_exit_to= _user_mode.do_syscall_64 > 3.17 +684.7% 24.85 =C4=85 8% perf-sched.wait_time.a= vg.ms.futex_wait_queue.__futex_wait.futex_wait.do_futex > 36.64 =C4=85 13% +244.7% 126.32 =C4=85 6% perf-sched.wait_t= ime.avg.ms.schedule_hrtimeout_range_clock.do_poll.constprop.0.do_sys_poll > 9.79 +303.0% 39.45 =C4=85 4% perf-sched.wait_time.a= vg.ms.schedule_hrtimeout_range_clock.do_sigtimedwait.__x64_sys_rt_sigtimedw= ait.do_syscall_64 > 1.05 +23.8% 1.30 perf-sched.wait_time.avg.ms= .schedule_preempt_disabled.rwsem_down_read_slowpath.down_read.do_madvise > 0.86 +101.2% 1.73 =C4=85 3% perf-sched.wait_time.a= vg.ms.schedule_preempt_disabled.rwsem_down_read_slowpath.down_read.exit_mm > 0.11 =C4=85 21% +438.9% 0.61 =C4=85 15% perf-sched.wait_t= ime.avg.ms.schedule_preempt_disabled.rwsem_down_write_slowpath.down_write_k= illable.__vm_munmap > 0.32 =C4=85 4% +28.5% 0.41 =C4=85 13% perf-sched.wait_t= ime.avg.ms.schedule_timeout.__wait_for_common.wait_for_completion_state.ker= nel_clone > 12.00 =C4=85 3% +139.6% 28.76 =C4=85 6% perf-sched.wait_t= ime.avg.ms.smpboot_thread_fn.kthread.ret_from_fork.ret_from_fork_asm > 6.07 =C4=85 2% +403.5% 30.56 =C4=85 5% perf-sched.wait_t= ime.avg.ms.worker_thread.kthread.ret_from_fork.ret_from_fork_asm > 0.38 =C4=85 41% -98.8% 0.00 =C4=85105% perf-sched.wait_t= ime.max.ms.__cond_resched.__alloc_pages.alloc_pages_mpol.pte_alloc_one.__pt= e_alloc > 0.36 =C4=85 34% -84.3% 0.06 =C4=85200% perf-sched.wait_t= ime.max.ms.__cond_resched.__alloc_pages.alloc_pages_mpol.vma_alloc_folio.do= _anonymous_page > 0.36 =C4=85 51% -92.9% 0.03 =C4=85114% perf-sched.wait_t= ime.max.ms.__cond_resched.__anon_vma_prepare.do_anonymous_page.__handle_mm_= fault.handle_mm_fault > 15.98 =C4=85 5% +361.7% 73.80 =C4=85 23% perf-sched.wait_t= ime.max.ms.__cond_resched.__fput.__x64_sys_close.do_syscall_64.entry_SYSCAL= L_64_after_hwframe > 0.51 =C4=85 14% -92.8% 0.04 =C4=85196% perf-sched.wait_t= ime.max.ms.__cond_resched.__kmem_cache_alloc_node.__kmalloc_node.__vmalloc_= area_node.__vmalloc_node_range > 8.56 =C4=85 11% -99.9% 0.01 =C4=85126% perf-sched.wait_t= ime.max.ms.__cond_resched.__kmem_cache_alloc_node.__kmalloc_node.memcg_allo= c_slab_cgroups.allocate_slab > 0.43 =C4=85 32% -68.2% 0.14 =C4=85119% perf-sched.wait_t= ime.max.ms.__cond_resched.__kmem_cache_alloc_node.kmalloc_node_trace.__get_= vm_area_node.__vmalloc_node_range > 0.46 =C4=85 20% -89.3% 0.05 =C4=85184% perf-sched.wait_t= ime.max.ms.__cond_resched.__vmalloc_area_node.__vmalloc_node_range.alloc_th= read_stack_node.dup_task_struct > 16.40 =C4=85 2% +342.9% 72.65 =C4=85 29% perf-sched.wait_t= ime.max.ms.__cond_resched.apparmor_file_alloc_security.security_file_alloc.= init_file.alloc_empty_file > 0.31 =C4=85 63% -76.2% 0.07 =C4=85169% perf-sched.wait_t= ime.max.ms.__cond_resched.cgroup_css_set_fork.cgroup_can_fork.copy_process.= kernel_clone > 0.14 =C4=85 93% +258.7% 0.49 =C4=85 14% perf-sched.wait_t= ime.max.ms.__cond_resched.clear_huge_page.__do_huge_pmd_anonymous_page.__ha= ndle_mm_fault.handle_mm_fault > 16.49 =C4=85 3% +463.5% 92.89 =C4=85 27% perf-sched.wait_t= ime.max.ms.__cond_resched.dentry_kill.dput.__fput.__x64_sys_close > 1.09 +171.0% 2.96 =C4=85 10% perf-sched.wait_time.m= ax.ms.__cond_resched.down_read.do_madvise.__x64_sys_madvise.do_syscall_64 > 1.16 =C4=85 7% +155.1% 2.97 =C4=85 4% perf-sched.wait_t= ime.max.ms.__cond_resched.down_read.exit_mm.do_exit.__x64_sys_exit > 0.19 =C4=85 78% -100.0% 0.00 perf-sched.wait_time.m= ax.ms.__cond_resched.down_write.__split_vma.vma_modify.mprotect_fixup > 0.33 =C4=85 35% -100.0% 0.00 perf-sched.wait_time.m= ax.ms.__cond_resched.down_write.vma_expand.mmap_region.do_mmap > 0.20 =C4=85101% -99.3% 0.00 =C4=85223% perf-sched.wait_t= ime.max.ms.__cond_resched.down_write_killable.vm_mmap_pgoff.do_syscall_64.e= ntry_SYSCALL_64_after_hwframe > 17.31 =C4=85 5% +521.0% 107.51 =C4=85 14% perf-sched.wait_t= ime.max.ms.__cond_resched.dput.nd_jump_link.proc_ns_get_link.pick_link > 15.38 =C4=85 6% +325.3% 65.40 =C4=85 22% perf-sched.wait_t= ime.max.ms.__cond_resched.dput.pick_link.step_into.open_last_lookups > 16.72 =C4=85 4% +456.6% 93.04 =C4=85 11% perf-sched.wait_t= ime.max.ms.__cond_resched.dput.terminate_walk.path_openat.do_filp_open > 1.16 =C4=85 2% +88.7% 2.20 =C4=85 33% perf-sched.wait_t= ime.max.ms.__cond_resched.exit_signals.do_exit.__x64_sys_exit.do_syscall_64 > 53.96 =C4=85 32% +444.0% 293.53 =C4=85109% perf-sched.wait_t= ime.max.ms.__cond_resched.generic_perform_write.shmem_file_write_iter.vfs_w= rite.ksys_write > 17.13 =C4=85 2% +511.2% 104.68 =C4=85 14% perf-sched.wait_t= ime.max.ms.__cond_resched.kmem_cache_alloc.alloc_empty_file.path_openat.do_= filp_open > 15.69 =C4=85 4% +379.5% 75.25 =C4=85 28% perf-sched.wait_t= ime.max.ms.__cond_resched.kmem_cache_alloc.create_new_namespaces.__do_sys_s= etns.do_syscall_64 > 15.70 =C4=85 3% +422.2% 81.97 =C4=85 19% perf-sched.wait_t= ime.max.ms.__cond_resched.kmem_cache_alloc.getname_flags.part.0 > 0.27 =C4=85 80% -99.6% 0.00 =C4=85223% perf-sched.wait_t= ime.max.ms.__cond_resched.kmem_cache_alloc.mas_alloc_nodes.mas_preallocate.= __split_vma > 16.37 +528.6% 102.90 =C4=85 21% perf-sched.wait_time.m= ax.ms.__cond_resched.kmem_cache_alloc.security_file_alloc.init_file.alloc_e= mpty_file > 0.44 =C4=85 33% -99.1% 0.00 =C4=85104% perf-sched.wait_t= ime.max.ms.__cond_resched.kmem_cache_alloc.vm_area_dup.__split_vma.vma_modi= fy > 0.02 =C4=85 83% -100.0% 0.00 perf-sched.wait_time.m= ax.ms.__cond_resched.kmem_cache_alloc_node.alloc_vmap_area.__get_vm_area_no= de.__vmalloc_node_range > 0.08 =C4=85 83% -95.4% 0.00 =C4=85147% perf-sched.wait_t= ime.max.ms.__cond_resched.kmem_cache_alloc_node.dup_task_struct.copy_proces= s.kernel_clone > 1.16 =C4=85 2% +134.7% 2.72 =C4=85 19% perf-sched.wait_t= ime.max.ms.__cond_resched.mutex_lock.futex_exit_release.exit_mm_release.exi= t_mm > 49.88 =C4=85 25% +141.0% 120.23 =C4=85 27% perf-sched.wait_t= ime.max.ms.__cond_resched.shmem_inode_acct_blocks.shmem_alloc_and_add_folio= .shmem_get_folio_gfp.shmem_write_begin > 17.24 +495.7% 102.70 =C4=85 2% perf-sched.wait_time.m= ax.ms.__cond_resched.slab_pre_alloc_hook.constprop.0.kmem_cache_alloc_lru > 402.56 =C4=85 15% -52.8% 189.89 =C4=85 14% perf-sched.wait_t= ime.max.ms.__cond_resched.smpboot_thread_fn.kthread.ret_from_fork.ret_from_= fork_asm > 16.96 =C4=85 4% +521.4% 105.39 =C4=85 15% perf-sched.wait_t= ime.max.ms.__cond_resched.switch_task_namespaces.__do_sys_setns.do_syscall_= 64.entry_SYSCALL_64_after_hwframe > 1.06 +241.7% 3.61 =C4=85 4% perf-sched.wait_time.m= ax.ms.__cond_resched.tlb_batch_pages_flush.tlb_finish_mmu.zap_page_range_si= ngle.madvise_vma_behavior > 1.07 -88.9% 0.12 =C4=85221% perf-sched.wait_time.m= ax.ms.__cond_resched.unmap_page_range.zap_page_range_single.madvise_vma_beh= avior.do_madvise > 0.28 =C4=85 27% +499.0% 1.67 =C4=85 18% perf-sched.wait_t= ime.max.ms.__cond_resched.unmap_vmas.unmap_region.do_vmi_align_munmap.do_vm= i_munmap > 1.21 =C4=85 2% +207.2% 3.71 =C4=85 3% perf-sched.wait_t= ime.max.ms.__cond_resched.zap_page_range_single.madvise_vma_behavior.do_mad= vise.__x64_sys_madvise > 13.43 =C4=85 26% +38.8% 18.64 perf-sched.wait_time.m= ax.ms.__x64_sys_sched_yield.do_syscall_64.entry_SYSCALL_64_after_hwframe.[u= nknown] > 28.45 +517.3% 175.65 =C4=85 14% perf-sched.wait_time.m= ax.ms.do_task_dead.do_exit.__x64_sys_exit.do_syscall_64.entry_SYSCALL_64_af= ter_hwframe > 0.79 =C4=85 10% +62.2% 1.28 =C4=85 25% perf-sched.wait_t= ime.max.ms.do_wait.kernel_wait4.__do_sys_wait4.do_syscall_64 > 13.22 =C4=85 2% +317.2% 55.16 =C4=85 35% perf-sched.wait_t= ime.max.ms.exit_to_user_mode_loop.exit_to_user_mode_prepare.irqentry_exit_t= o_user_mode.asm_sysvec_call_function > 834.29 =C4=85 28% -48.5% 429.53 =C4=85 94% perf-sched.wait_t= ime.max.ms.exit_to_user_mode_loop.exit_to_user_mode_prepare.irqentry_exit_t= o_user_mode.asm_sysvec_reschedule_ipi > 22.48 +628.6% 163.83 =C4=85 16% perf-sched.wait_time.m= ax.ms.futex_wait_queue.__futex_wait.futex_wait.do_futex > 22.74 =C4=85 18% +398.0% 113.25 =C4=85 16% perf-sched.wait_t= ime.max.ms.schedule_hrtimeout_range_clock.do_sigtimedwait.__x64_sys_rt_sigt= imedwait.do_syscall_64 > 7.72 =C4=85 7% +80.6% 13.95 =C4=85 2% perf-sched.wait_t= ime.max.ms.schedule_preempt_disabled.rwsem_down_write_slowpath.down_write_k= illable.__vm_munmap > 0.74 =C4=85 4% +77.2% 1.31 =C4=85 32% perf-sched.wait_t= ime.max.ms.schedule_timeout.__wait_for_common.wait_for_completion_state.ker= nel_clone > 5.01 +14.1% 5.72 =C4=85 2% perf-sched.wait_time.m= ax.ms.schedule_timeout.rcu_gp_fqs_loop.rcu_gp_kthread.kthread > 44.98 -19.7 25.32 =C4=85 2% perf-profile.calltrace= .cycles-pp.secondary_startup_64_no_verify > 43.21 -19.6 23.65 =C4=85 3% perf-profile.calltrace= .cycles-pp.start_secondary.secondary_startup_64_no_verify > 43.21 -19.6 23.65 =C4=85 3% perf-profile.calltrace= .cycles-pp.cpu_startup_entry.start_secondary.secondary_startup_64_no_verify > 43.18 -19.5 23.63 =C4=85 3% perf-profile.calltrace= .cycles-pp.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64_n= o_verify > 40.30 -17.5 22.75 =C4=85 3% perf-profile.calltrace= .cycles-pp.cpuidle_idle_call.do_idle.cpu_startup_entry.start_secondary.seco= ndary_startup_64_no_verify > 41.10 -17.4 23.66 =C4=85 2% perf-profile.calltrace= .cycles-pp.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call.do_idle.cpu_= startup_entry > 39.55 -17.3 22.24 =C4=85 3% perf-profile.calltrace= .cycles-pp.cpuidle_enter.cpuidle_idle_call.do_idle.cpu_startup_entry.start_= secondary > 24.76 =C4=85 2% -8.5 16.23 =C4=85 3% perf-profile.call= trace.cycles-pp.intel_idle.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_c= all.do_idle > 8.68 =C4=85 4% -6.5 2.22 =C4=85 6% perf-profile.call= trace.cycles-pp.asm_sysvec_apic_timer_interrupt.intel_idle_irq.cpuidle_ente= r_state.cpuidle_enter.cpuidle_idle_call > 7.23 =C4=85 4% -5.8 1.46 =C4=85 8% perf-profile.call= trace.cycles-pp.entry_SYSCALL_64_after_hwframe > 7.23 =C4=85 4% -5.8 1.46 =C4=85 8% perf-profile.call= trace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe > 7.11 =C4=85 4% -5.7 1.39 =C4=85 7% perf-profile.call= trace.cycles-pp.__x64_sys_exit.do_syscall_64.entry_SYSCALL_64_after_hwframe > 7.09 =C4=85 4% -5.7 1.39 =C4=85 7% perf-profile.call= trace.cycles-pp.do_exit.__x64_sys_exit.do_syscall_64.entry_SYSCALL_64_after= _hwframe > 6.59 =C4=85 3% -5.1 1.47 =C4=85 7% perf-profile.call= trace.cycles-pp.ret_from_fork_asm > 6.59 =C4=85 3% -5.1 1.47 =C4=85 7% perf-profile.call= trace.cycles-pp.ret_from_fork.ret_from_fork_asm > 6.59 =C4=85 3% -5.1 1.47 =C4=85 7% perf-profile.call= trace.cycles-pp.kthread.ret_from_fork.ret_from_fork_asm > 5.76 =C4=85 2% -5.0 0.80 =C4=85 9% perf-profile.call= trace.cycles-pp.start_thread > 7.43 =C4=85 2% -4.9 2.52 =C4=85 7% perf-profile.call= trace.cycles-pp.intel_idle_irq.cpuidle_enter_state.cpuidle_enter.cpuidle_id= le_call.do_idle > 5.51 =C4=85 3% -4.8 0.70 =C4=85 7% perf-profile.call= trace.cycles-pp.entry_SYSCALL_64_after_hwframe.start_thread > 5.50 =C4=85 3% -4.8 0.70 =C4=85 7% perf-profile.call= trace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.start_thread > 5.48 =C4=85 3% -4.8 0.69 =C4=85 7% perf-profile.call= trace.cycles-pp.__x64_sys_exit.do_syscall_64.entry_SYSCALL_64_after_hwframe= .start_thread > 5.42 =C4=85 3% -4.7 0.69 =C4=85 7% perf-profile.call= trace.cycles-pp.do_exit.__x64_sys_exit.do_syscall_64.entry_SYSCALL_64_after= _hwframe.start_thread > 5.90 =C4=85 5% -3.9 2.01 =C4=85 4% perf-profile.call= trace.cycles-pp.flush_tlb_mm_range.tlb_finish_mmu.zap_page_range_single.mad= vise_vma_behavior.do_madvise > 4.18 =C4=85 5% -3.8 0.37 =C4=85 71% perf-profile.call= trace.cycles-pp.exit_notify.do_exit.__x64_sys_exit.do_syscall_64.entry_SYSC= ALL_64_after_hwframe > 5.76 =C4=85 5% -3.8 1.98 =C4=85 4% perf-profile.call= trace.cycles-pp.on_each_cpu_cond_mask.flush_tlb_mm_range.tlb_finish_mmu.zap= _page_range_single.madvise_vma_behavior > 5.04 =C4=85 7% -3.7 1.32 =C4=85 9% perf-profile.call= trace.cycles-pp.entry_SYSCALL_64_after_hwframe.__clone > 5.03 =C4=85 7% -3.7 1.32 =C4=85 9% perf-profile.call= trace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.__clone > 5.02 =C4=85 7% -3.7 1.32 =C4=85 9% perf-profile.call= trace.cycles-pp.__do_sys_clone.do_syscall_64.entry_SYSCALL_64_after_hwframe= .__clone > 5.02 =C4=85 7% -3.7 1.32 =C4=85 9% perf-profile.call= trace.cycles-pp.kernel_clone.__do_sys_clone.do_syscall_64.entry_SYSCALL_64_= after_hwframe.__clone > 5.62 =C4=85 5% -3.7 1.96 =C4=85 3% perf-profile.call= trace.cycles-pp.smp_call_function_many_cond.on_each_cpu_cond_mask.flush_tlb= _mm_range.tlb_finish_mmu.zap_page_range_single > 4.03 =C4=85 4% -3.1 0.92 =C4=85 7% perf-profile.call= trace.cycles-pp.smpboot_thread_fn.kthread.ret_from_fork.ret_from_fork_asm > 6.03 =C4=85 5% -3.1 2.94 =C4=85 3% perf-profile.call= trace.cycles-pp.tlb_finish_mmu.zap_page_range_single.madvise_vma_behavior.d= o_madvise.__x64_sys_madvise > 3.43 =C4=85 5% -2.8 0.67 =C4=85 13% perf-profile.call= trace.cycles-pp.run_ksoftirqd.smpboot_thread_fn.kthread.ret_from_fork.ret_f= rom_fork_asm > 3.43 =C4=85 5% -2.8 0.67 =C4=85 13% perf-profile.call= trace.cycles-pp.__do_softirq.run_ksoftirqd.smpboot_thread_fn.kthread.ret_fr= om_fork > 3.41 =C4=85 5% -2.7 0.66 =C4=85 13% perf-profile.call= trace.cycles-pp.rcu_core.__do_softirq.run_ksoftirqd.smpboot_thread_fn.kthre= ad > 3.40 =C4=85 5% -2.7 0.66 =C4=85 13% perf-profile.call= trace.cycles-pp.rcu_do_batch.rcu_core.__do_softirq.run_ksoftirqd.smpboot_th= read_fn > 3.67 =C4=85 7% -2.7 0.94 =C4=85 10% perf-profile.call= trace.cycles-pp.copy_process.kernel_clone.__do_sys_clone.do_syscall_64.entr= y_SYSCALL_64_after_hwframe > 2.92 =C4=85 7% -2.4 0.50 =C4=85 46% perf-profile.call= trace.cycles-pp.stress_pthread > 2.54 =C4=85 6% -2.2 0.38 =C4=85 70% perf-profile.call= trace.cycles-pp.worker_thread.kthread.ret_from_fork.ret_from_fork_asm > 2.46 =C4=85 6% -1.8 0.63 =C4=85 10% perf-profile.call= trace.cycles-pp.dup_task_struct.copy_process.kernel_clone.__do_sys_clone.do= _syscall_64 > 3.00 =C4=85 6% -1.6 1.43 =C4=85 7% perf-profile.call= trace.cycles-pp.__munmap > 2.96 =C4=85 6% -1.5 1.42 =C4=85 7% perf-profile.call= trace.cycles-pp.entry_SYSCALL_64_after_hwframe.__munmap > 2.96 =C4=85 6% -1.5 1.42 =C4=85 7% perf-profile.call= trace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.__munmap > 2.95 =C4=85 6% -1.5 1.41 =C4=85 7% perf-profile.call= trace.cycles-pp.__x64_sys_munmap.do_syscall_64.entry_SYSCALL_64_after_hwfra= me.__munmap > 2.95 =C4=85 6% -1.5 1.41 =C4=85 7% perf-profile.call= trace.cycles-pp.__vm_munmap.__x64_sys_munmap.do_syscall_64.entry_SYSCALL_64= _after_hwframe.__munmap > 2.02 =C4=85 4% -1.5 0.52 =C4=85 46% perf-profile.call= trace.cycles-pp.__lll_lock_wait > 1.78 =C4=85 3% -1.5 0.30 =C4=85100% perf-profile.call= trace.cycles-pp.entry_SYSCALL_64_after_hwframe.__lll_lock_wait > 1.77 =C4=85 3% -1.5 0.30 =C4=85100% perf-profile.call= trace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.__lll_lock_wai= t > 1.54 =C4=85 6% -1.3 0.26 =C4=85100% perf-profile.call= trace.cycles-pp.schedule_idle.do_idle.cpu_startup_entry.start_secondary.sec= ondary_startup_64_no_verify > 2.54 =C4=85 6% -1.2 1.38 =C4=85 6% perf-profile.call= trace.cycles-pp.do_vmi_munmap.__vm_munmap.__x64_sys_munmap.do_syscall_64.en= try_SYSCALL_64_after_hwframe > 2.51 =C4=85 6% -1.1 1.37 =C4=85 7% perf-profile.call= trace.cycles-pp.do_vmi_align_munmap.do_vmi_munmap.__vm_munmap.__x64_sys_mun= map.do_syscall_64 > 1.13 -0.7 0.40 =C4=85 70% perf-profile.calltrace= .cycles-pp.exit_mm.do_exit.__x64_sys_exit.do_syscall_64.entry_SYSCALL_64_af= ter_hwframe > 1.15 =C4=85 5% -0.7 0.46 =C4=85 45% perf-profile.call= trace.cycles-pp.llist_add_batch.smp_call_function_many_cond.on_each_cpu_con= d_mask.flush_tlb_mm_range.tlb_finish_mmu > 1.58 =C4=85 5% -0.6 0.94 =C4=85 7% perf-profile.call= trace.cycles-pp.unmap_region.do_vmi_align_munmap.do_vmi_munmap.__vm_munmap.= __x64_sys_munmap > 0.99 =C4=85 5% -0.5 0.51 =C4=85 45% perf-profile.call= trace.cycles-pp.__do_softirq.irq_exit_rcu.sysvec_apic_timer_interrupt.asm_s= ysvec_apic_timer_interrupt.cpuidle_enter_state > 1.01 =C4=85 5% -0.5 0.54 =C4=85 45% perf-profile.call= trace.cycles-pp.irq_exit_rcu.sysvec_apic_timer_interrupt.asm_sysvec_apic_ti= mer_interrupt.cpuidle_enter_state.cpuidle_enter > 0.82 =C4=85 4% -0.2 0.59 =C4=85 5% perf-profile.call= trace.cycles-pp.default_send_IPI_mask_sequence_phys.smp_call_function_many_= cond.on_each_cpu_cond_mask.flush_tlb_mm_range.tlb_finish_mmu > 0.00 +0.5 0.54 =C4=85 5% perf-profile.calltrace= .cycles-pp.flush_tlb_func.__flush_smp_call_function_queue.__sysvec_call_fun= ction.sysvec_call_function.asm_sysvec_call_function > 0.00 +0.6 0.60 =C4=85 5% perf-profile.calltrace= .cycles-pp.release_pages.tlb_batch_pages_flush.tlb_finish_mmu.zap_page_rang= e_single.madvise_vma_behavior > 0.00 +0.6 0.61 =C4=85 6% perf-profile.calltrace= .cycles-pp.release_pages.tlb_batch_pages_flush.tlb_finish_mmu.unmap_region.= do_vmi_align_munmap > 0.00 +0.6 0.62 =C4=85 6% perf-profile.calltrace= .cycles-pp.tlb_batch_pages_flush.tlb_finish_mmu.unmap_region.do_vmi_align_m= unmap.do_vmi_munmap > 0.53 =C4=85 5% +0.6 1.17 =C4=85 13% perf-profile.call= trace.cycles-pp.tick_nohz_highres_handler.__hrtimer_run_queues.hrtimer_inte= rrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt > 1.94 =C4=85 2% +0.7 2.64 =C4=85 9% perf-profile.call= trace.cycles-pp.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt= .cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call > 0.00 +0.7 0.73 =C4=85 5% perf-profile.calltrace= .cycles-pp.__mod_memcg_lruvec_state.__mod_lruvec_page_state.page_remove_rma= p.zap_pte_range.zap_pmd_range > 0.00 +0.8 0.75 =C4=85 20% perf-profile.calltrace= .cycles-pp.__cond_resched.clear_huge_page.__do_huge_pmd_anonymous_page.__ha= ndle_mm_fault.handle_mm_fault > 2.02 =C4=85 2% +0.8 2.85 =C4=85 9% perf-profile.call= trace.cycles-pp.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state.cpuidle= _enter.cpuidle_idle_call.do_idle > 0.74 =C4=85 5% +0.8 1.57 =C4=85 11% perf-profile.call= trace.cycles-pp.__hrtimer_run_queues.hrtimer_interrupt.__sysvec_apic_timer_= interrupt.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt > 0.00 +0.9 0.90 =C4=85 4% perf-profile.calltrace= .cycles-pp.tlb_batch_pages_flush.tlb_finish_mmu.zap_page_range_single.madvi= se_vma_behavior.do_madvise > 0.00 +0.9 0.92 =C4=85 13% perf-profile.calltrace= .cycles-pp.scheduler_tick.update_process_times.tick_sched_handle.tick_nohz_= highres_handler.__hrtimer_run_queues > 0.86 =C4=85 4% +1.0 1.82 =C4=85 10% perf-profile.call= trace.cycles-pp.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic= _timer_interrupt.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state > 0.86 =C4=85 4% +1.0 1.83 =C4=85 10% perf-profile.call= trace.cycles-pp.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt.a= sm_sysvec_apic_timer_interrupt.cpuidle_enter_state.cpuidle_enter > 0.00 +1.0 0.98 =C4=85 7% perf-profile.calltrace= .cycles-pp.smp_call_function_many_cond.on_each_cpu_cond_mask.flush_tlb_mm_r= ange.pmdp_invalidate.__split_huge_pmd_locked > 0.09 =C4=85223% +1.0 1.07 =C4=85 11% perf-profile.call= trace.cycles-pp.update_process_times.tick_sched_handle.tick_nohz_highres_ha= ndler.__hrtimer_run_queues.hrtimer_interrupt > 0.00 +1.0 0.99 =C4=85 6% perf-profile.calltrace= .cycles-pp.on_each_cpu_cond_mask.flush_tlb_mm_range.pmdp_invalidate.__split= _huge_pmd_locked.__split_huge_pmd > 0.00 +1.0 1.00 =C4=85 7% perf-profile.calltrace= .cycles-pp.flush_tlb_mm_range.pmdp_invalidate.__split_huge_pmd_locked.__spl= it_huge_pmd.zap_pmd_range > 0.09 =C4=85223% +1.0 1.10 =C4=85 12% perf-profile.call= trace.cycles-pp.tick_sched_handle.tick_nohz_highres_handler.__hrtimer_run_q= ueues.hrtimer_interrupt.__sysvec_apic_timer_interrupt > 0.00 +1.0 1.01 =C4=85 6% perf-profile.calltrace= .cycles-pp.pmdp_invalidate.__split_huge_pmd_locked.__split_huge_pmd.zap_pmd= _range.unmap_page_range > 0.00 +1.1 1.10 =C4=85 5% perf-profile.calltrace= .cycles-pp.__flush_smp_call_function_queue.__sysvec_call_function.sysvec_ca= ll_function.asm_sysvec_call_function.native_queued_spin_lock_slowpath > 0.00 +1.1 1.12 =C4=85 5% perf-profile.calltrace= .cycles-pp.__sysvec_call_function.sysvec_call_function.asm_sysvec_call_func= tion.native_queued_spin_lock_slowpath._raw_spin_lock > 0.00 +1.2 1.23 =C4=85 4% perf-profile.calltrace= .cycles-pp.page_add_anon_rmap.__split_huge_pmd_locked.__split_huge_pmd.zap_= pmd_range.unmap_page_range > 0.00 +1.3 1.32 =C4=85 4% perf-profile.calltrace= .cycles-pp.sysvec_call_function.asm_sysvec_call_function.native_queued_spin= _lock_slowpath._raw_spin_lock.__split_huge_pmd > 0.00 +1.4 1.38 =C4=85 5% perf-profile.calltrace= .cycles-pp.__mod_lruvec_page_state.page_remove_rmap.zap_pte_range.zap_pmd_r= ange.unmap_page_range > 0.00 +2.4 2.44 =C4=85 10% perf-profile.calltrace= .cycles-pp.asm_sysvec_call_function.native_queued_spin_lock_slowpath._raw_s= pin_lock.__split_huge_pmd.zap_pmd_range > 0.00 +3.1 3.10 =C4=85 5% perf-profile.calltrace= .cycles-pp.page_remove_rmap.zap_pte_range.zap_pmd_range.unmap_page_range.za= p_page_range_single > 0.00 +3.5 3.52 =C4=85 5% perf-profile.calltrace= .cycles-pp.__split_huge_pmd_locked.__split_huge_pmd.zap_pmd_range.unmap_pag= e_range.zap_page_range_single > 0.88 =C4=85 4% +3.8 4.69 =C4=85 4% perf-profile.call= trace.cycles-pp.zap_pte_range.zap_pmd_range.unmap_page_range.zap_page_range= _single.madvise_vma_behavior > 6.30 =C4=85 6% +13.5 19.85 =C4=85 7% perf-profile.call= trace.cycles-pp.__clone > 0.00 +16.7 16.69 =C4=85 7% perf-profile.calltrace= .cycles-pp.clear_page_erms.clear_huge_page.__do_huge_pmd_anonymous_page.__h= andle_mm_fault.handle_mm_fault > 1.19 =C4=85 29% +17.1 18.32 =C4=85 7% perf-profile.call= trace.cycles-pp.__handle_mm_fault.handle_mm_fault.do_user_addr_fault.exc_pa= ge_fault.asm_exc_page_fault > 0.00 +17.6 17.56 =C4=85 7% perf-profile.calltrace= .cycles-pp.clear_huge_page.__do_huge_pmd_anonymous_page.__handle_mm_fault.h= andle_mm_fault.do_user_addr_fault > 0.63 =C4=85 7% +17.7 18.35 =C4=85 7% perf-profile.call= trace.cycles-pp.asm_exc_page_fault.__clone > 0.59 =C4=85 5% +17.8 18.34 =C4=85 7% perf-profile.call= trace.cycles-pp.exc_page_fault.asm_exc_page_fault.__clone > 0.59 =C4=85 5% +17.8 18.34 =C4=85 7% perf-profile.call= trace.cycles-pp.do_user_addr_fault.exc_page_fault.asm_exc_page_fault.__clon= e > 0.00 +17.9 17.90 =C4=85 7% perf-profile.calltrace= .cycles-pp.__do_huge_pmd_anonymous_page.__handle_mm_fault.handle_mm_fault.d= o_user_addr_fault.exc_page_fault > 0.36 =C4=85 71% +18.0 18.33 =C4=85 7% perf-profile.call= trace.cycles-pp.handle_mm_fault.do_user_addr_fault.exc_page_fault.asm_exc_p= age_fault.__clone > 0.00 +32.0 32.03 =C4=85 2% perf-profile.calltrace= .cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock.__split_huge_pmd= .zap_pmd_range.unmap_page_range > 0.00 +32.6 32.62 =C4=85 2% perf-profile.calltrace= .cycles-pp._raw_spin_lock.__split_huge_pmd.zap_pmd_range.unmap_page_range.z= ap_page_range_single > 0.00 +36.2 36.19 =C4=85 2% perf-profile.calltrace= .cycles-pp.__split_huge_pmd.zap_pmd_range.unmap_page_range.zap_page_range_s= ingle.madvise_vma_behavior > 7.97 =C4=85 4% +36.6 44.52 =C4=85 2% perf-profile.call= trace.cycles-pp.__madvise > 7.91 =C4=85 4% +36.6 44.46 =C4=85 2% perf-profile.call= trace.cycles-pp.entry_SYSCALL_64_after_hwframe.__madvise > 7.90 =C4=85 4% +36.6 44.46 =C4=85 2% perf-profile.call= trace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.__madvise > 7.87 =C4=85 4% +36.6 44.44 =C4=85 2% perf-profile.call= trace.cycles-pp.__x64_sys_madvise.do_syscall_64.entry_SYSCALL_64_after_hwfr= ame.__madvise > 7.86 =C4=85 4% +36.6 44.44 =C4=85 2% perf-profile.call= trace.cycles-pp.do_madvise.__x64_sys_madvise.do_syscall_64.entry_SYSCALL_64= _after_hwframe.__madvise > 7.32 =C4=85 4% +36.8 44.07 =C4=85 2% perf-profile.call= trace.cycles-pp.madvise_vma_behavior.do_madvise.__x64_sys_madvise.do_syscal= l_64.entry_SYSCALL_64_after_hwframe > 7.25 =C4=85 4% +36.8 44.06 =C4=85 2% perf-profile.call= trace.cycles-pp.zap_page_range_single.madvise_vma_behavior.do_madvise.__x64= _sys_madvise.do_syscall_64 > 1.04 =C4=85 4% +40.0 41.08 =C4=85 2% perf-profile.call= trace.cycles-pp.unmap_page_range.zap_page_range_single.madvise_vma_behavior= .do_madvise.__x64_sys_madvise > 1.00 =C4=85 3% +40.1 41.06 =C4=85 2% perf-profile.call= trace.cycles-pp.zap_pmd_range.unmap_page_range.zap_page_range_single.madvis= e_vma_behavior.do_madvise > 44.98 -19.7 25.32 =C4=85 2% perf-profile.children.= cycles-pp.secondary_startup_64_no_verify > 44.98 -19.7 25.32 =C4=85 2% perf-profile.children.= cycles-pp.cpu_startup_entry > 44.96 -19.6 25.31 =C4=85 2% perf-profile.children.= cycles-pp.do_idle > 43.21 -19.6 23.65 =C4=85 3% perf-profile.children.= cycles-pp.start_secondary > 41.98 -17.6 24.40 =C4=85 2% perf-profile.children.= cycles-pp.cpuidle_idle_call > 41.21 -17.3 23.86 =C4=85 2% perf-profile.children.= cycles-pp.cpuidle_enter > 41.20 -17.3 23.86 =C4=85 2% perf-profile.children.= cycles-pp.cpuidle_enter_state > 12.69 =C4=85 3% -10.6 2.12 =C4=85 6% perf-profile.chil= dren.cycles-pp.do_exit > 12.60 =C4=85 3% -10.5 2.08 =C4=85 7% perf-profile.chil= dren.cycles-pp.__x64_sys_exit > 24.76 =C4=85 2% -8.5 16.31 =C4=85 2% perf-profile.chil= dren.cycles-pp.intel_idle > 12.34 =C4=85 2% -8.4 3.90 =C4=85 5% perf-profile.chil= dren.cycles-pp.intel_idle_irq > 6.96 =C4=85 4% -5.4 1.58 =C4=85 7% perf-profile.chil= dren.cycles-pp.ret_from_fork_asm > 6.69 =C4=85 4% -5.2 1.51 =C4=85 7% perf-profile.chil= dren.cycles-pp.ret_from_fork > 6.59 =C4=85 3% -5.1 1.47 =C4=85 7% perf-profile.chil= dren.cycles-pp.kthread > 5.78 =C4=85 2% -5.0 0.80 =C4=85 8% perf-profile.chil= dren.cycles-pp.start_thread > 4.68 =C4=85 4% -4.5 0.22 =C4=85 10% perf-profile.chil= dren.cycles-pp._raw_spin_lock_irq > 5.03 =C4=85 7% -3.7 1.32 =C4=85 9% perf-profile.chil= dren.cycles-pp.__do_sys_clone > 5.02 =C4=85 7% -3.7 1.32 =C4=85 9% perf-profile.chil= dren.cycles-pp.kernel_clone > 4.20 =C4=85 5% -3.7 0.53 =C4=85 9% perf-profile.chil= dren.cycles-pp.exit_notify > 4.67 =C4=85 5% -3.6 1.10 =C4=85 9% perf-profile.chil= dren.cycles-pp.rcu_core > 4.60 =C4=85 4% -3.5 1.06 =C4=85 10% perf-profile.chil= dren.cycles-pp.rcu_do_batch > 4.89 =C4=85 5% -3.4 1.44 =C4=85 11% perf-profile.chil= dren.cycles-pp.__do_softirq > 5.64 =C4=85 3% -3.2 2.39 =C4=85 6% perf-profile.chil= dren.cycles-pp.__schedule > 6.27 =C4=85 5% -3.2 3.03 =C4=85 4% perf-profile.chil= dren.cycles-pp.flush_tlb_mm_range > 4.03 =C4=85 4% -3.1 0.92 =C4=85 7% perf-profile.chil= dren.cycles-pp.smpboot_thread_fn > 6.68 =C4=85 4% -3.1 3.61 =C4=85 3% perf-profile.chil= dren.cycles-pp.tlb_finish_mmu > 6.04 =C4=85 5% -3.1 2.99 =C4=85 4% perf-profile.chil= dren.cycles-pp.on_each_cpu_cond_mask > 6.04 =C4=85 5% -3.0 2.99 =C4=85 4% perf-profile.chil= dren.cycles-pp.smp_call_function_many_cond > 3.77 =C4=85 2% -3.0 0.73 =C4=85 16% perf-profile.chil= dren.cycles-pp._raw_spin_lock_irqsave > 7.78 -3.0 4.77 =C4=85 5% perf-profile.children.= cycles-pp.asm_sysvec_apic_timer_interrupt > 3.43 =C4=85 5% -2.8 0.67 =C4=85 13% perf-profile.chil= dren.cycles-pp.run_ksoftirqd > 3.67 =C4=85 7% -2.7 0.94 =C4=85 10% perf-profile.chil= dren.cycles-pp.copy_process > 2.80 =C4=85 6% -2.5 0.34 =C4=85 15% perf-profile.chil= dren.cycles-pp.queued_write_lock_slowpath > 3.41 =C4=85 2% -2.5 0.96 =C4=85 16% perf-profile.chil= dren.cycles-pp.do_futex > 3.06 =C4=85 5% -2.4 0.68 =C4=85 16% perf-profile.chil= dren.cycles-pp.free_unref_page_commit > 3.02 =C4=85 5% -2.4 0.67 =C4=85 16% perf-profile.chil= dren.cycles-pp.free_pcppages_bulk > 2.92 =C4=85 7% -2.3 0.58 =C4=85 14% perf-profile.chil= dren.cycles-pp.stress_pthread > 3.22 =C4=85 3% -2.3 0.90 =C4=85 18% perf-profile.chil= dren.cycles-pp.__x64_sys_futex > 2.52 =C4=85 5% -2.2 0.35 =C4=85 7% perf-profile.chil= dren.cycles-pp.release_task > 2.54 =C4=85 6% -2.0 0.53 =C4=85 10% perf-profile.chil= dren.cycles-pp.worker_thread > 3.12 =C4=85 5% -1.9 1.17 =C4=85 11% perf-profile.chil= dren.cycles-pp.free_unref_page > 2.31 =C4=85 6% -1.9 0.45 =C4=85 11% perf-profile.chil= dren.cycles-pp.process_one_work > 2.47 =C4=85 6% -1.8 0.63 =C4=85 10% perf-profile.chil= dren.cycles-pp.dup_task_struct > 2.19 =C4=85 5% -1.8 0.41 =C4=85 12% perf-profile.chil= dren.cycles-pp.delayed_vfree_work > 2.14 =C4=85 5% -1.7 0.40 =C4=85 11% perf-profile.chil= dren.cycles-pp.vfree > 3.19 =C4=85 2% -1.6 1.58 =C4=85 8% perf-profile.chil= dren.cycles-pp.schedule > 2.06 =C4=85 3% -1.6 0.46 =C4=85 7% perf-profile.chil= dren.cycles-pp.__sigtimedwait > 3.02 =C4=85 6% -1.6 1.44 =C4=85 7% perf-profile.chil= dren.cycles-pp.__munmap > 1.94 =C4=85 4% -1.6 0.39 =C4=85 14% perf-profile.chil= dren.cycles-pp.__unfreeze_partials > 2.95 =C4=85 6% -1.5 1.41 =C4=85 7% perf-profile.chil= dren.cycles-pp.__x64_sys_munmap > 2.95 =C4=85 6% -1.5 1.41 =C4=85 7% perf-profile.chil= dren.cycles-pp.__vm_munmap > 2.14 =C4=85 3% -1.5 0.60 =C4=85 21% perf-profile.chil= dren.cycles-pp.futex_wait > 2.08 =C4=85 4% -1.5 0.60 =C4=85 19% perf-profile.chil= dren.cycles-pp.__lll_lock_wait > 2.04 =C4=85 3% -1.5 0.56 =C4=85 20% perf-profile.chil= dren.cycles-pp.__futex_wait > 1.77 =C4=85 5% -1.5 0.32 =C4=85 10% perf-profile.chil= dren.cycles-pp.remove_vm_area > 1.86 =C4=85 5% -1.4 0.46 =C4=85 10% perf-profile.chil= dren.cycles-pp.open64 > 1.74 =C4=85 4% -1.4 0.37 =C4=85 7% perf-profile.chil= dren.cycles-pp.__x64_sys_rt_sigtimedwait > 1.71 =C4=85 4% -1.4 0.36 =C4=85 8% perf-profile.chil= dren.cycles-pp.do_sigtimedwait > 1.79 =C4=85 5% -1.3 0.46 =C4=85 9% perf-profile.chil= dren.cycles-pp.__x64_sys_openat > 1.78 =C4=85 5% -1.3 0.46 =C4=85 8% perf-profile.chil= dren.cycles-pp.do_sys_openat2 > 1.61 =C4=85 4% -1.3 0.32 =C4=85 12% perf-profile.chil= dren.cycles-pp.poll_idle > 1.65 =C4=85 9% -1.3 0.37 =C4=85 14% perf-profile.chil= dren.cycles-pp.pthread_create@@GLIBC_2.2.5 > 1.56 =C4=85 8% -1.2 0.35 =C4=85 7% perf-profile.chil= dren.cycles-pp.alloc_thread_stack_node > 2.32 =C4=85 3% -1.2 1.13 =C4=85 8% perf-profile.chil= dren.cycles-pp.pick_next_task_fair > 2.59 =C4=85 6% -1.2 1.40 =C4=85 7% perf-profile.chil= dren.cycles-pp.do_vmi_munmap > 1.55 =C4=85 4% -1.2 0.40 =C4=85 19% perf-profile.chil= dren.cycles-pp.futex_wait_queue > 1.37 =C4=85 5% -1.1 0.22 =C4=85 12% perf-profile.chil= dren.cycles-pp.find_unlink_vmap_area > 2.52 =C4=85 6% -1.1 1.38 =C4=85 6% perf-profile.chil= dren.cycles-pp.do_vmi_align_munmap > 1.53 =C4=85 5% -1.1 0.39 =C4=85 8% perf-profile.chil= dren.cycles-pp.do_filp_open > 1.52 =C4=85 5% -1.1 0.39 =C4=85 7% perf-profile.chil= dren.cycles-pp.path_openat > 1.25 =C4=85 3% -1.1 0.14 =C4=85 12% perf-profile.chil= dren.cycles-pp.sigpending > 1.58 =C4=85 5% -1.1 0.50 =C4=85 6% perf-profile.chil= dren.cycles-pp.schedule_idle > 1.29 =C4=85 5% -1.1 0.21 =C4=85 21% perf-profile.chil= dren.cycles-pp.__mprotect > 1.40 =C4=85 8% -1.1 0.32 =C4=85 4% perf-profile.chil= dren.cycles-pp.__vmalloc_node_range > 2.06 =C4=85 3% -1.0 1.02 =C4=85 9% perf-profile.chil= dren.cycles-pp.newidle_balance > 1.04 =C4=85 3% -1.0 0.08 =C4=85 23% perf-profile.chil= dren.cycles-pp.__x64_sys_rt_sigpending > 1.14 =C4=85 6% -1.0 0.18 =C4=85 18% perf-profile.chil= dren.cycles-pp.__x64_sys_mprotect > 1.13 =C4=85 6% -1.0 0.18 =C4=85 17% perf-profile.chil= dren.cycles-pp.do_mprotect_pkey > 1.30 =C4=85 7% -0.9 0.36 =C4=85 10% perf-profile.chil= dren.cycles-pp.wake_up_new_task > 1.14 =C4=85 9% -0.9 0.22 =C4=85 16% perf-profile.chil= dren.cycles-pp.do_anonymous_page > 0.95 =C4=85 3% -0.9 0.04 =C4=85 71% perf-profile.chil= dren.cycles-pp.do_sigpending > 1.24 =C4=85 3% -0.9 0.34 =C4=85 9% perf-profile.chil= dren.cycles-pp.futex_wake > 1.02 =C4=85 6% -0.9 0.14 =C4=85 15% perf-profile.chil= dren.cycles-pp.mprotect_fixup > 1.91 =C4=85 2% -0.9 1.06 =C4=85 9% perf-profile.chil= dren.cycles-pp.load_balance > 1.38 =C4=85 5% -0.8 0.53 =C4=85 6% perf-profile.chil= dren.cycles-pp.select_task_rq_fair > 1.14 =C4=85 4% -0.8 0.31 =C4=85 12% perf-profile.chil= dren.cycles-pp.__pthread_mutex_unlock_usercnt > 2.68 =C4=85 3% -0.8 1.91 =C4=85 6% perf-profile.chil= dren.cycles-pp.__flush_smp_call_function_queue > 1.00 =C4=85 4% -0.7 0.26 =C4=85 10% perf-profile.chil= dren.cycles-pp.flush_smp_call_function_queue > 1.44 =C4=85 3% -0.7 0.73 =C4=85 10% perf-profile.chil= dren.cycles-pp.find_busiest_group > 0.81 =C4=85 6% -0.7 0.10 =C4=85 18% perf-profile.chil= dren.cycles-pp.vma_modify > 1.29 =C4=85 3% -0.7 0.60 =C4=85 8% perf-profile.chil= dren.cycles-pp.exit_mm > 1.40 =C4=85 3% -0.7 0.71 =C4=85 10% perf-profile.chil= dren.cycles-pp.update_sd_lb_stats > 0.78 =C4=85 7% -0.7 0.10 =C4=85 19% perf-profile.chil= dren.cycles-pp.__split_vma > 0.90 =C4=85 8% -0.7 0.22 =C4=85 10% perf-profile.chil= dren.cycles-pp.__vmalloc_area_node > 0.75 =C4=85 4% -0.7 0.10 =C4=85 5% perf-profile.chil= dren.cycles-pp.__exit_signal > 1.49 =C4=85 2% -0.7 0.84 =C4=85 7% perf-profile.chil= dren.cycles-pp.try_to_wake_up > 0.89 =C4=85 7% -0.6 0.24 =C4=85 10% perf-profile.chil= dren.cycles-pp.find_idlest_cpu > 1.59 =C4=85 5% -0.6 0.95 =C4=85 7% perf-profile.chil= dren.cycles-pp.unmap_region > 0.86 =C4=85 3% -0.6 0.22 =C4=85 26% perf-profile.chil= dren.cycles-pp.pthread_cond_timedwait@@GLIBC_2.3.2 > 1.59 =C4=85 3% -0.6 0.95 =C4=85 9% perf-profile.chil= dren.cycles-pp.irq_exit_rcu > 1.24 =C4=85 3% -0.6 0.61 =C4=85 10% perf-profile.chil= dren.cycles-pp.update_sg_lb_stats > 0.94 =C4=85 5% -0.6 0.32 =C4=85 11% perf-profile.chil= dren.cycles-pp.do_task_dead > 0.87 =C4=85 3% -0.6 0.25 =C4=85 19% perf-profile.chil= dren.cycles-pp.perf_iterate_sb > 0.82 =C4=85 4% -0.6 0.22 =C4=85 10% perf-profile.chil= dren.cycles-pp.sched_ttwu_pending > 1.14 =C4=85 3% -0.6 0.54 =C4=85 10% perf-profile.chil= dren.cycles-pp.activate_task > 0.84 -0.6 0.25 =C4=85 10% perf-profile.children.= cycles-pp.syscall_exit_to_user_mode > 0.81 =C4=85 6% -0.6 0.22 =C4=85 11% perf-profile.chil= dren.cycles-pp.find_idlest_group > 0.75 =C4=85 5% -0.6 0.18 =C4=85 14% perf-profile.chil= dren.cycles-pp.step_into > 0.74 =C4=85 8% -0.6 0.18 =C4=85 14% perf-profile.chil= dren.cycles-pp.__alloc_pages_bulk > 0.74 =C4=85 6% -0.5 0.19 =C4=85 11% perf-profile.chil= dren.cycles-pp.update_sg_wakeup_stats > 0.72 =C4=85 5% -0.5 0.18 =C4=85 15% perf-profile.chil= dren.cycles-pp.pick_link > 1.06 =C4=85 2% -0.5 0.52 =C4=85 9% perf-profile.chil= dren.cycles-pp.enqueue_task_fair > 0.77 =C4=85 6% -0.5 0.23 =C4=85 12% perf-profile.chil= dren.cycles-pp.unmap_vmas > 0.76 =C4=85 2% -0.5 0.22 =C4=85 8% perf-profile.chil= dren.cycles-pp.exit_to_user_mode_prepare > 0.94 =C4=85 2% -0.5 0.42 =C4=85 10% perf-profile.chil= dren.cycles-pp.dequeue_task_fair > 0.65 =C4=85 5% -0.5 0.15 =C4=85 18% perf-profile.chil= dren.cycles-pp.open_last_lookups > 1.37 =C4=85 3% -0.5 0.87 =C4=85 4% perf-profile.chil= dren.cycles-pp.llist_add_batch > 0.70 =C4=85 4% -0.5 0.22 =C4=85 19% perf-profile.chil= dren.cycles-pp.memcpy_orig > 0.91 =C4=85 4% -0.5 0.44 =C4=85 7% perf-profile.chil= dren.cycles-pp.update_load_avg > 0.67 -0.5 0.20 =C4=85 8% perf-profile.children.= cycles-pp.switch_fpu_return > 0.88 =C4=85 3% -0.5 0.42 =C4=85 8% perf-profile.chil= dren.cycles-pp.enqueue_entity > 0.91 =C4=85 4% -0.5 0.45 =C4=85 12% perf-profile.chil= dren.cycles-pp.ttwu_do_activate > 0.77 =C4=85 4% -0.5 0.32 =C4=85 10% perf-profile.chil= dren.cycles-pp.schedule_hrtimeout_range_clock > 0.63 =C4=85 5% -0.4 0.20 =C4=85 21% perf-profile.chil= dren.cycles-pp.arch_dup_task_struct > 0.74 =C4=85 3% -0.4 0.32 =C4=85 15% perf-profile.chil= dren.cycles-pp.dequeue_entity > 0.62 =C4=85 5% -0.4 0.21 =C4=85 5% perf-profile.chil= dren.cycles-pp.finish_task_switch > 0.56 -0.4 0.16 =C4=85 7% perf-profile.children.= cycles-pp.restore_fpregs_from_fpstate > 0.53 =C4=85 4% -0.4 0.13 =C4=85 9% perf-profile.chil= dren.cycles-pp.syscall > 0.50 =C4=85 9% -0.4 0.11 =C4=85 18% perf-profile.chil= dren.cycles-pp.__get_vm_area_node > 0.51 =C4=85 3% -0.4 0.12 =C4=85 12% perf-profile.chil= dren.cycles-pp.__slab_free > 0.52 =C4=85 2% -0.4 0.14 =C4=85 10% perf-profile.chil= dren.cycles-pp.kmem_cache_free > 0.75 =C4=85 3% -0.4 0.37 =C4=85 9% perf-profile.chil= dren.cycles-pp.exit_mm_release > 0.50 =C4=85 6% -0.4 0.12 =C4=85 21% perf-profile.chil= dren.cycles-pp.do_send_specific > 0.74 =C4=85 3% -0.4 0.37 =C4=85 8% perf-profile.chil= dren.cycles-pp.futex_exit_release > 0.45 =C4=85 10% -0.4 0.09 =C4=85 17% perf-profile.chil= dren.cycles-pp.alloc_vmap_area > 0.47 =C4=85 3% -0.4 0.11 =C4=85 20% perf-profile.chil= dren.cycles-pp.tgkill > 0.68 =C4=85 11% -0.4 0.32 =C4=85 12% perf-profile.chil= dren.cycles-pp.__mmap > 0.48 =C4=85 3% -0.4 0.13 =C4=85 6% perf-profile.chil= dren.cycles-pp.entry_SYSCALL_64 > 0.76 =C4=85 5% -0.3 0.41 =C4=85 10% perf-profile.chil= dren.cycles-pp.wake_up_q > 0.42 =C4=85 7% -0.3 0.08 =C4=85 22% perf-profile.chil= dren.cycles-pp.__close > 0.49 =C4=85 7% -0.3 0.14 =C4=85 25% perf-profile.chil= dren.cycles-pp.kmem_cache_alloc > 0.49 =C4=85 9% -0.3 0.15 =C4=85 14% perf-profile.chil= dren.cycles-pp.mas_store_gfp > 0.46 =C4=85 4% -0.3 0.12 =C4=85 23% perf-profile.chil= dren.cycles-pp.perf_event_task_output > 0.44 =C4=85 10% -0.3 0.10 =C4=85 28% perf-profile.chil= dren.cycles-pp.pthread_sigqueue > 0.46 =C4=85 4% -0.3 0.12 =C4=85 15% perf-profile.chil= dren.cycles-pp.link_path_walk > 0.42 =C4=85 8% -0.3 0.10 =C4=85 20% perf-profile.chil= dren.cycles-pp.proc_ns_get_link > 0.63 =C4=85 10% -0.3 0.32 =C4=85 12% perf-profile.chil= dren.cycles-pp.vm_mmap_pgoff > 0.45 =C4=85 4% -0.3 0.14 =C4=85 13% perf-profile.chil= dren.cycles-pp.sched_move_task > 0.36 =C4=85 8% -0.3 0.06 =C4=85 49% perf-profile.chil= dren.cycles-pp.__x64_sys_close > 0.46 =C4=85 8% -0.3 0.17 =C4=85 14% perf-profile.chil= dren.cycles-pp.prctl > 0.65 =C4=85 3% -0.3 0.35 =C4=85 7% perf-profile.chil= dren.cycles-pp.futex_cleanup > 0.42 =C4=85 7% -0.3 0.12 =C4=85 15% perf-profile.chil= dren.cycles-pp.mas_store_prealloc > 0.49 =C4=85 5% -0.3 0.20 =C4=85 13% perf-profile.chil= dren.cycles-pp.__rmqueue_pcplist > 0.37 =C4=85 7% -0.3 0.08 =C4=85 16% perf-profile.chil= dren.cycles-pp.do_tkill > 0.36 =C4=85 10% -0.3 0.08 =C4=85 20% perf-profile.chil= dren.cycles-pp.ns_get_path > 0.37 =C4=85 4% -0.3 0.09 =C4=85 18% perf-profile.chil= dren.cycles-pp.setns > 0.67 =C4=85 3% -0.3 0.41 =C4=85 8% perf-profile.chil= dren.cycles-pp.hrtimer_wakeup > 0.35 =C4=85 5% -0.3 0.10 =C4=85 16% perf-profile.chil= dren.cycles-pp.__task_pid_nr_ns > 0.41 =C4=85 5% -0.3 0.16 =C4=85 12% perf-profile.chil= dren.cycles-pp.mas_wr_bnode > 0.35 =C4=85 4% -0.3 0.10 =C4=85 20% perf-profile.chil= dren.cycles-pp.rcu_cblist_dequeue > 0.37 =C4=85 5% -0.2 0.12 =C4=85 17% perf-profile.chil= dren.cycles-pp.exit_task_stack_account > 0.56 =C4=85 4% -0.2 0.31 =C4=85 12% perf-profile.chil= dren.cycles-pp.select_task_rq > 0.29 =C4=85 6% -0.2 0.05 =C4=85 46% perf-profile.chil= dren.cycles-pp.mas_wr_store_entry > 0.34 =C4=85 4% -0.2 0.10 =C4=85 27% perf-profile.chil= dren.cycles-pp.perf_event_task > 0.39 =C4=85 9% -0.2 0.15 =C4=85 12% perf-profile.chil= dren.cycles-pp.__switch_to_asm > 0.35 =C4=85 5% -0.2 0.11 =C4=85 11% perf-profile.chil= dren.cycles-pp.account_kernel_stack > 0.30 =C4=85 7% -0.2 0.06 =C4=85 48% perf-profile.chil= dren.cycles-pp.__ns_get_path > 0.31 =C4=85 9% -0.2 0.07 =C4=85 17% perf-profile.chil= dren.cycles-pp.free_vmap_area_noflush > 0.31 =C4=85 5% -0.2 0.08 =C4=85 19% perf-profile.chil= dren.cycles-pp.__do_sys_setns > 0.33 =C4=85 7% -0.2 0.10 =C4=85 7% perf-profile.chil= dren.cycles-pp.__free_one_page > 0.31 =C4=85 11% -0.2 0.08 =C4=85 13% perf-profile.chil= dren.cycles-pp.__pte_alloc > 0.36 =C4=85 6% -0.2 0.13 =C4=85 12% perf-profile.chil= dren.cycles-pp.switch_mm_irqs_off > 0.27 =C4=85 12% -0.2 0.05 =C4=85 71% perf-profile.chil= dren.cycles-pp.__fput > 0.53 =C4=85 9% -0.2 0.31 =C4=85 12% perf-profile.chil= dren.cycles-pp.do_mmap > 0.27 =C4=85 12% -0.2 0.05 =C4=85 77% perf-profile.chil= dren.cycles-pp.__x64_sys_rt_tgsigqueueinfo > 0.28 =C4=85 5% -0.2 0.06 =C4=85 50% perf-profile.chil= dren.cycles-pp.entry_SYSRETQ_unsafe_stack > 0.34 =C4=85 10% -0.2 0.12 =C4=85 29% perf-profile.chil= dren.cycles-pp.futex_wait_setup > 0.27 =C4=85 6% -0.2 0.06 =C4=85 45% perf-profile.chil= dren.cycles-pp.__x64_sys_tgkill > 0.31 =C4=85 7% -0.2 0.11 =C4=85 18% perf-profile.chil= dren.cycles-pp.__switch_to > 0.26 =C4=85 8% -0.2 0.06 =C4=85 21% perf-profile.chil= dren.cycles-pp.__call_rcu_common > 0.33 =C4=85 9% -0.2 0.13 =C4=85 18% perf-profile.chil= dren.cycles-pp.__do_sys_prctl > 0.28 =C4=85 5% -0.2 0.08 =C4=85 17% perf-profile.chil= dren.cycles-pp.mm_release > 0.52 =C4=85 2% -0.2 0.32 =C4=85 9% perf-profile.chil= dren.cycles-pp.__get_user_8 > 0.24 =C4=85 10% -0.2 0.04 =C4=85 72% perf-profile.chil= dren.cycles-pp.dput > 0.25 =C4=85 14% -0.2 0.05 =C4=85 46% perf-profile.chil= dren.cycles-pp.perf_event_mmap > 0.24 =C4=85 7% -0.2 0.06 =C4=85 50% perf-profile.chil= dren.cycles-pp.mas_walk > 0.28 =C4=85 6% -0.2 0.10 =C4=85 24% perf-profile.chil= dren.cycles-pp.rmqueue_bulk > 0.23 =C4=85 15% -0.2 0.05 =C4=85 46% perf-profile.chil= dren.cycles-pp.perf_event_mmap_event > 0.25 =C4=85 15% -0.2 0.08 =C4=85 45% perf-profile.chil= dren.cycles-pp.___slab_alloc > 0.20 =C4=85 14% -0.2 0.03 =C4=85100% perf-profile.chil= dren.cycles-pp.lookup_fast > 0.20 =C4=85 10% -0.2 0.04 =C4=85 75% perf-profile.chil= dren.cycles-pp.memcg_slab_post_alloc_hook > 0.28 =C4=85 7% -0.2 0.12 =C4=85 24% perf-profile.chil= dren.cycles-pp.prepare_task_switch > 0.22 =C4=85 11% -0.2 0.05 =C4=85 8% perf-profile.chil= dren.cycles-pp.ttwu_queue_wakelist > 0.63 =C4=85 5% -0.2 0.47 =C4=85 12% perf-profile.chil= dren.cycles-pp.llist_reverse_order > 0.25 =C4=85 11% -0.2 0.09 =C4=85 34% perf-profile.chil= dren.cycles-pp.futex_q_lock > 0.21 =C4=85 6% -0.2 0.06 =C4=85 47% perf-profile.chil= dren.cycles-pp.kmem_cache_alloc_node > 0.18 =C4=85 11% -0.2 0.03 =C4=85100% perf-profile.chil= dren.cycles-pp.alloc_empty_file > 0.19 =C4=85 5% -0.2 0.04 =C4=85 71% perf-profile.chil= dren.cycles-pp.__put_task_struct > 0.19 =C4=85 15% -0.2 0.03 =C4=85 70% perf-profile.chil= dren.cycles-pp.asm_sysvec_call_function_single > 0.24 =C4=85 6% -0.2 0.09 =C4=85 20% perf-profile.chil= dren.cycles-pp.___perf_sw_event > 0.18 =C4=85 7% -0.2 0.03 =C4=85100% perf-profile.chil= dren.cycles-pp.perf_event_fork > 0.19 =C4=85 11% -0.1 0.04 =C4=85 71% perf-profile.chil= dren.cycles-pp.select_idle_core > 0.30 =C4=85 11% -0.1 0.15 =C4=85 7% perf-profile.chil= dren.cycles-pp.pte_alloc_one > 0.25 =C4=85 6% -0.1 0.11 =C4=85 10% perf-profile.chil= dren.cycles-pp.set_next_entity > 0.20 =C4=85 10% -0.1 0.06 =C4=85 49% perf-profile.chil= dren.cycles-pp.__perf_event_header__init_id > 0.18 =C4=85 15% -0.1 0.03 =C4=85101% perf-profile.chil= dren.cycles-pp.__radix_tree_lookup > 0.22 =C4=85 11% -0.1 0.08 =C4=85 21% perf-profile.chil= dren.cycles-pp.mas_spanning_rebalance > 0.20 =C4=85 9% -0.1 0.06 =C4=85 9% perf-profile.chil= dren.cycles-pp.stress_pthread_func > 0.18 =C4=85 12% -0.1 0.04 =C4=85 73% perf-profile.chil= dren.cycles-pp.__getpid > 0.16 =C4=85 13% -0.1 0.02 =C4=85 99% perf-profile.chil= dren.cycles-pp.walk_component > 0.28 =C4=85 5% -0.1 0.15 =C4=85 13% perf-profile.chil= dren.cycles-pp.update_curr > 0.25 =C4=85 5% -0.1 0.11 =C4=85 22% perf-profile.chil= dren.cycles-pp.balance_fair > 0.16 =C4=85 9% -0.1 0.03 =C4=85100% perf-profile.chil= dren.cycles-pp.futex_wake_mark > 0.16 =C4=85 12% -0.1 0.04 =C4=85 71% perf-profile.chil= dren.cycles-pp.get_futex_key > 0.17 =C4=85 6% -0.1 0.05 =C4=85 47% perf-profile.chil= dren.cycles-pp.memcg_account_kmem > 0.25 =C4=85 11% -0.1 0.12 =C4=85 11% perf-profile.chil= dren.cycles-pp._find_next_bit > 0.15 =C4=85 13% -0.1 0.02 =C4=85 99% perf-profile.chil= dren.cycles-pp.do_open > 0.20 =C4=85 8% -0.1 0.08 =C4=85 16% perf-profile.chil= dren.cycles-pp.mas_rebalance > 0.17 =C4=85 13% -0.1 0.05 =C4=85 45% perf-profile.chil= dren.cycles-pp.__memcg_kmem_charge_page > 0.33 =C4=85 6% -0.1 0.21 =C4=85 10% perf-profile.chil= dren.cycles-pp.select_idle_sibling > 0.14 =C4=85 11% -0.1 0.03 =C4=85100% perf-profile.chil= dren.cycles-pp.get_user_pages_fast > 0.18 =C4=85 7% -0.1 0.07 =C4=85 14% perf-profile.chil= dren.cycles-pp.mas_alloc_nodes > 0.14 =C4=85 11% -0.1 0.03 =C4=85101% perf-profile.chil= dren.cycles-pp.set_task_cpu > 0.14 =C4=85 12% -0.1 0.03 =C4=85101% perf-profile.chil= dren.cycles-pp.vm_unmapped_area > 0.38 =C4=85 6% -0.1 0.27 =C4=85 7% perf-profile.chil= dren.cycles-pp.native_sched_clock > 0.16 =C4=85 10% -0.1 0.05 =C4=85 47% perf-profile.chil= dren.cycles-pp.arch_get_unmapped_area_topdown > 0.36 =C4=85 9% -0.1 0.25 =C4=85 12% perf-profile.chil= dren.cycles-pp.mmap_region > 0.23 =C4=85 7% -0.1 0.12 =C4=85 9% perf-profile.chil= dren.cycles-pp.available_idle_cpu > 0.13 =C4=85 11% -0.1 0.02 =C4=85 99% perf-profile.chil= dren.cycles-pp.internal_get_user_pages_fast > 0.16 =C4=85 10% -0.1 0.06 =C4=85 18% perf-profile.chil= dren.cycles-pp.get_unmapped_area > 0.50 =C4=85 7% -0.1 0.40 =C4=85 6% perf-profile.chil= dren.cycles-pp.menu_select > 0.24 =C4=85 9% -0.1 0.14 =C4=85 13% perf-profile.chil= dren.cycles-pp.rmqueue > 0.17 =C4=85 14% -0.1 0.07 =C4=85 26% perf-profile.chil= dren.cycles-pp.perf_event_comm > 0.17 =C4=85 15% -0.1 0.07 =C4=85 23% perf-profile.chil= dren.cycles-pp.perf_event_comm_event > 0.17 =C4=85 11% -0.1 0.07 =C4=85 14% perf-profile.chil= dren.cycles-pp.pick_next_entity > 0.13 =C4=85 14% -0.1 0.03 =C4=85102% perf-profile.chil= dren.cycles-pp.perf_output_begin > 0.23 =C4=85 6% -0.1 0.13 =C4=85 21% perf-profile.chil= dren.cycles-pp.__update_load_avg_cfs_rq > 0.14 =C4=85 18% -0.1 0.04 =C4=85 72% perf-profile.chil= dren.cycles-pp.perf_event_comm_output > 0.21 =C4=85 9% -0.1 0.12 =C4=85 9% perf-profile.chil= dren.cycles-pp.update_rq_clock > 0.16 =C4=85 8% -0.1 0.06 =C4=85 19% perf-profile.chil= dren.cycles-pp.mas_split > 0.13 =C4=85 14% -0.1 0.04 =C4=85 71% perf-profile.chil= dren.cycles-pp.raw_spin_rq_lock_nested > 0.13 =C4=85 6% -0.1 0.04 =C4=85 71% perf-profile.chil= dren.cycles-pp.syscall_return_via_sysret > 0.13 =C4=85 7% -0.1 0.04 =C4=85 72% perf-profile.chil= dren.cycles-pp.mas_topiary_replace > 0.14 =C4=85 8% -0.1 0.06 =C4=85 9% perf-profile.chil= dren.cycles-pp.mas_preallocate > 0.16 =C4=85 11% -0.1 0.07 =C4=85 18% perf-profile.chil= dren.cycles-pp.__pick_eevdf > 0.11 =C4=85 14% -0.1 0.02 =C4=85 99% perf-profile.chil= dren.cycles-pp.mas_empty_area_rev > 0.25 =C4=85 7% -0.1 0.17 =C4=85 10% perf-profile.chil= dren.cycles-pp.select_idle_cpu > 0.14 =C4=85 12% -0.1 0.06 =C4=85 14% perf-profile.chil= dren.cycles-pp.cpu_stopper_thread > 0.14 =C4=85 10% -0.1 0.06 =C4=85 13% perf-profile.chil= dren.cycles-pp.active_load_balance_cpu_stop > 0.14 =C4=85 14% -0.1 0.06 =C4=85 11% perf-profile.chil= dren.cycles-pp.os_xsave > 0.18 =C4=85 6% -0.1 0.11 =C4=85 14% perf-profile.chil= dren.cycles-pp.idle_cpu > 0.17 =C4=85 4% -0.1 0.10 =C4=85 15% perf-profile.chil= dren.cycles-pp.hrtimer_start_range_ns > 0.11 =C4=85 14% -0.1 0.03 =C4=85100% perf-profile.chil= dren.cycles-pp.__pthread_mutex_lock > 0.32 =C4=85 5% -0.1 0.25 =C4=85 5% perf-profile.chil= dren.cycles-pp.sched_clock > 0.11 =C4=85 6% -0.1 0.03 =C4=85 70% perf-profile.chil= dren.cycles-pp.wakeup_preempt > 0.23 =C4=85 7% -0.1 0.16 =C4=85 13% perf-profile.chil= dren.cycles-pp.update_rq_clock_task > 0.13 =C4=85 8% -0.1 0.06 =C4=85 16% perf-profile.chil= dren.cycles-pp.local_clock_noinstr > 0.11 =C4=85 10% -0.1 0.04 =C4=85 71% perf-profile.chil= dren.cycles-pp.kmem_cache_alloc_bulk > 0.34 =C4=85 4% -0.1 0.27 =C4=85 6% perf-profile.chil= dren.cycles-pp.sched_clock_cpu > 0.11 =C4=85 9% -0.1 0.04 =C4=85 76% perf-profile.chil= dren.cycles-pp.avg_vruntime > 0.15 =C4=85 8% -0.1 0.08 =C4=85 14% perf-profile.chil= dren.cycles-pp.update_cfs_group > 0.10 =C4=85 8% -0.1 0.04 =C4=85 71% perf-profile.chil= dren.cycles-pp.__kmem_cache_alloc_bulk > 0.13 =C4=85 8% -0.1 0.06 =C4=85 11% perf-profile.chil= dren.cycles-pp.sched_use_asym_prio > 0.09 =C4=85 12% -0.1 0.02 =C4=85 99% perf-profile.chil= dren.cycles-pp.getname_flags > 0.18 =C4=85 9% -0.1 0.12 =C4=85 12% perf-profile.chil= dren.cycles-pp.__update_load_avg_se > 0.11 =C4=85 8% -0.1 0.05 =C4=85 46% perf-profile.chil= dren.cycles-pp.place_entity > 0.08 =C4=85 12% -0.0 0.02 =C4=85 99% perf-profile.chil= dren.cycles-pp.folio_add_lru_vma > 0.10 =C4=85 7% -0.0 0.05 =C4=85 46% perf-profile.chil= dren.cycles-pp._find_next_and_bit > 0.10 =C4=85 6% -0.0 0.06 =C4=85 24% perf-profile.chil= dren.cycles-pp.reweight_entity > 0.03 =C4=85 70% +0.0 0.08 =C4=85 14% perf-profile.chil= dren.cycles-pp.perf_rotate_context > 0.19 =C4=85 10% +0.1 0.25 =C4=85 7% perf-profile.chil= dren.cycles-pp.irqtime_account_irq > 0.08 =C4=85 11% +0.1 0.14 =C4=85 21% perf-profile.chil= dren.cycles-pp.perf_mux_hrtimer_handler > 0.00 +0.1 0.06 =C4=85 14% perf-profile.children.= cycles-pp.rcu_pending > 0.10 =C4=85 17% +0.1 0.16 =C4=85 13% perf-profile.chil= dren.cycles-pp.rebalance_domains > 0.14 =C4=85 16% +0.1 0.21 =C4=85 12% perf-profile.chil= dren.cycles-pp.downgrade_write > 0.14 =C4=85 14% +0.1 0.21 =C4=85 10% perf-profile.chil= dren.cycles-pp.down_read_killable > 0.00 +0.1 0.07 =C4=85 11% perf-profile.children.= cycles-pp.free_tail_page_prepare > 0.02 =C4=85141% +0.1 0.09 =C4=85 20% perf-profile.chil= dren.cycles-pp.rcu_sched_clock_irq > 0.01 =C4=85223% +0.1 0.08 =C4=85 25% perf-profile.chil= dren.cycles-pp.arch_scale_freq_tick > 0.55 =C4=85 9% +0.1 0.62 =C4=85 9% perf-profile.chil= dren.cycles-pp.__alloc_pages > 0.34 =C4=85 5% +0.1 0.41 =C4=85 9% perf-profile.chil= dren.cycles-pp.clock_nanosleep > 0.00 +0.1 0.08 =C4=85 23% perf-profile.children.= cycles-pp.tick_nohz_next_event > 0.70 =C4=85 2% +0.1 0.78 =C4=85 5% perf-profile.chil= dren.cycles-pp.flush_tlb_func > 0.14 =C4=85 10% +0.1 0.23 =C4=85 13% perf-profile.chil= dren.cycles-pp.__intel_pmu_enable_all > 0.07 =C4=85 19% +0.1 0.17 =C4=85 17% perf-profile.chil= dren.cycles-pp.cgroup_rstat_updated > 0.04 =C4=85 71% +0.1 0.14 =C4=85 11% perf-profile.chil= dren.cycles-pp.tick_nohz_get_sleep_length > 0.25 =C4=85 9% +0.1 0.38 =C4=85 11% perf-profile.chil= dren.cycles-pp.down_read > 0.43 =C4=85 9% +0.1 0.56 =C4=85 10% perf-profile.chil= dren.cycles-pp.get_page_from_freelist > 0.00 +0.1 0.15 =C4=85 6% perf-profile.children.= cycles-pp.vm_normal_page > 0.31 =C4=85 7% +0.2 0.46 =C4=85 9% perf-profile.chil= dren.cycles-pp.native_flush_tlb_local > 0.00 +0.2 0.16 =C4=85 8% perf-profile.children.= cycles-pp.__tlb_remove_page_size > 0.28 =C4=85 11% +0.2 0.46 =C4=85 13% perf-profile.chil= dren.cycles-pp.vma_alloc_folio > 0.00 +0.2 0.24 =C4=85 5% perf-profile.children.= cycles-pp._compound_head > 0.07 =C4=85 16% +0.2 0.31 =C4=85 6% perf-profile.chil= dren.cycles-pp.__mod_node_page_state > 0.38 =C4=85 5% +0.2 0.62 =C4=85 7% perf-profile.chil= dren.cycles-pp.perf_adjust_freq_unthr_context > 0.22 =C4=85 12% +0.2 0.47 =C4=85 10% perf-profile.chil= dren.cycles-pp.schedule_preempt_disabled > 0.38 =C4=85 5% +0.3 0.64 =C4=85 7% perf-profile.chil= dren.cycles-pp.perf_event_task_tick > 0.00 +0.3 0.27 =C4=85 5% perf-profile.children.= cycles-pp.free_swap_cache > 0.30 =C4=85 10% +0.3 0.58 =C4=85 10% perf-profile.chil= dren.cycles-pp.rwsem_down_read_slowpath > 0.00 +0.3 0.30 =C4=85 4% perf-profile.children.= cycles-pp.free_pages_and_swap_cache > 0.09 =C4=85 10% +0.3 0.42 =C4=85 7% perf-profile.chil= dren.cycles-pp.__mod_lruvec_state > 0.00 +0.3 0.34 =C4=85 9% perf-profile.children.= cycles-pp.deferred_split_folio > 0.00 +0.4 0.36 =C4=85 13% perf-profile.children.= cycles-pp.prep_compound_page > 0.09 =C4=85 10% +0.4 0.50 =C4=85 9% perf-profile.chil= dren.cycles-pp.free_unref_page_prepare > 0.00 +0.4 0.42 =C4=85 11% perf-profile.children.= cycles-pp.do_huge_pmd_anonymous_page > 1.67 =C4=85 3% +0.4 2.12 =C4=85 8% perf-profile.chil= dren.cycles-pp.__hrtimer_run_queues > 0.63 =C4=85 3% +0.5 1.11 =C4=85 12% perf-profile.chil= dren.cycles-pp.scheduler_tick > 1.93 =C4=85 3% +0.5 2.46 =C4=85 8% perf-profile.chil= dren.cycles-pp.__sysvec_apic_timer_interrupt > 1.92 =C4=85 3% +0.5 2.45 =C4=85 8% perf-profile.chil= dren.cycles-pp.hrtimer_interrupt > 0.73 =C4=85 3% +0.6 1.31 =C4=85 11% perf-profile.chil= dren.cycles-pp.update_process_times > 0.74 =C4=85 3% +0.6 1.34 =C4=85 11% perf-profile.chil= dren.cycles-pp.tick_sched_handle > 0.20 =C4=85 8% +0.6 0.83 =C4=85 18% perf-profile.chil= dren.cycles-pp.__cond_resched > 0.78 =C4=85 4% +0.6 1.43 =C4=85 12% perf-profile.chil= dren.cycles-pp.tick_nohz_highres_handler > 0.12 =C4=85 7% +0.7 0.81 =C4=85 5% perf-profile.chil= dren.cycles-pp.__mod_memcg_lruvec_state > 0.28 =C4=85 7% +0.9 1.23 =C4=85 4% perf-profile.chil= dren.cycles-pp.release_pages > 0.00 +1.0 1.01 =C4=85 6% perf-profile.children.= cycles-pp.pmdp_invalidate > 0.35 =C4=85 6% +1.2 1.56 =C4=85 5% perf-profile.chil= dren.cycles-pp.__mod_lruvec_page_state > 0.30 =C4=85 8% +1.2 1.53 =C4=85 4% perf-profile.chil= dren.cycles-pp.tlb_batch_pages_flush > 0.00 +1.3 1.26 =C4=85 4% perf-profile.children.= cycles-pp.page_add_anon_rmap > 0.09 =C4=85 11% +3.1 3.20 =C4=85 5% perf-profile.chil= dren.cycles-pp.page_remove_rmap > 1.60 =C4=85 2% +3.4 5.04 =C4=85 4% perf-profile.chil= dren.cycles-pp.zap_pte_range > 0.03 =C4=85100% +3.5 3.55 =C4=85 5% perf-profile.chil= dren.cycles-pp.__split_huge_pmd_locked > 41.36 +11.6 52.92 =C4=85 2% perf-profile.children.= cycles-pp.entry_SYSCALL_64_after_hwframe > 41.22 +11.7 52.88 =C4=85 2% perf-profile.children.= cycles-pp.do_syscall_64 > 6.42 =C4=85 6% +13.5 19.88 =C4=85 7% perf-profile.chil= dren.cycles-pp.__clone > 0.82 =C4=85 6% +16.2 16.98 =C4=85 7% perf-profile.chil= dren.cycles-pp.clear_page_erms > 2.62 =C4=85 5% +16.4 19.04 =C4=85 7% perf-profile.chil= dren.cycles-pp.asm_exc_page_fault > 2.18 =C4=85 5% +16.8 18.94 =C4=85 7% perf-profile.chil= dren.cycles-pp.exc_page_fault > 2.06 =C4=85 6% +16.8 18.90 =C4=85 7% perf-profile.chil= dren.cycles-pp.do_user_addr_fault > 1.60 =C4=85 8% +17.0 18.60 =C4=85 7% perf-profile.chil= dren.cycles-pp.handle_mm_fault > 1.52 =C4=85 7% +17.1 18.58 =C4=85 7% perf-profile.chil= dren.cycles-pp.__handle_mm_fault > 0.30 =C4=85 7% +17.4 17.72 =C4=85 7% perf-profile.chil= dren.cycles-pp.clear_huge_page > 0.31 =C4=85 8% +17.6 17.90 =C4=85 7% perf-profile.chil= dren.cycles-pp.__do_huge_pmd_anonymous_page > 11.66 =C4=85 3% +22.2 33.89 =C4=85 2% perf-profile.chil= dren.cycles-pp.native_queued_spin_lock_slowpath > 3.29 =C4=85 3% +30.2 33.46 perf-profile.children.= cycles-pp._raw_spin_lock > 0.04 =C4=85 71% +36.2 36.21 =C4=85 2% perf-profile.chil= dren.cycles-pp.__split_huge_pmd > 8.00 =C4=85 4% +36.5 44.54 =C4=85 2% perf-profile.chil= dren.cycles-pp.__madvise > 7.87 =C4=85 4% +36.6 44.44 =C4=85 2% perf-profile.chil= dren.cycles-pp.__x64_sys_madvise > 7.86 =C4=85 4% +36.6 44.44 =C4=85 2% perf-profile.chil= dren.cycles-pp.do_madvise > 7.32 =C4=85 4% +36.8 44.07 =C4=85 2% perf-profile.chil= dren.cycles-pp.madvise_vma_behavior > 7.26 =C4=85 4% +36.8 44.06 =C4=85 2% perf-profile.chil= dren.cycles-pp.zap_page_range_single > 1.78 +39.5 41.30 =C4=85 2% perf-profile.children.= cycles-pp.unmap_page_range > 1.72 +39.6 41.28 =C4=85 2% perf-profile.children.= cycles-pp.zap_pmd_range > 24.76 =C4=85 2% -8.5 16.31 =C4=85 2% perf-profile.self= .cycles-pp.intel_idle > 11.46 =C4=85 2% -7.8 3.65 =C4=85 5% perf-profile.self= .cycles-pp.intel_idle_irq > 3.16 =C4=85 7% -2.1 1.04 =C4=85 6% perf-profile.self= .cycles-pp.smp_call_function_many_cond > 1.49 =C4=85 4% -1.2 0.30 =C4=85 12% perf-profile.self= .cycles-pp.poll_idle > 1.15 =C4=85 3% -0.6 0.50 =C4=85 9% perf-profile.self= .cycles-pp._raw_spin_lock > 0.60 =C4=85 6% -0.6 0.03 =C4=85100% perf-profile.self= .cycles-pp.queued_write_lock_slowpath > 0.69 =C4=85 4% -0.5 0.22 =C4=85 20% perf-profile.self= .cycles-pp.memcpy_orig > 0.66 =C4=85 7% -0.5 0.18 =C4=85 11% perf-profile.self= .cycles-pp.update_sg_wakeup_stats > 0.59 =C4=85 4% -0.5 0.13 =C4=85 8% perf-profile.self= .cycles-pp._raw_spin_lock_irq > 0.86 =C4=85 3% -0.4 0.43 =C4=85 12% perf-profile.self= .cycles-pp.update_sg_lb_stats > 0.56 -0.4 0.16 =C4=85 7% perf-profile.self.cycl= es-pp.restore_fpregs_from_fpstate > 0.48 =C4=85 3% -0.4 0.12 =C4=85 10% perf-profile.self= .cycles-pp.__slab_free > 1.18 =C4=85 2% -0.4 0.82 =C4=85 3% perf-profile.self= .cycles-pp.llist_add_batch > 0.54 =C4=85 5% -0.3 0.19 =C4=85 6% perf-profile.self= .cycles-pp.__schedule > 0.47 =C4=85 7% -0.3 0.18 =C4=85 13% perf-profile.self= .cycles-pp._raw_spin_lock_irqsave > 0.34 =C4=85 5% -0.2 0.09 =C4=85 18% perf-profile.self= .cycles-pp.kmem_cache_free > 0.43 =C4=85 4% -0.2 0.18 =C4=85 11% perf-profile.self= .cycles-pp.update_load_avg > 0.35 =C4=85 4% -0.2 0.10 =C4=85 23% perf-profile.self= .cycles-pp.rcu_cblist_dequeue > 0.38 =C4=85 9% -0.2 0.15 =C4=85 10% perf-profile.self= .cycles-pp.__switch_to_asm > 0.33 =C4=85 5% -0.2 0.10 =C4=85 16% perf-profile.self= .cycles-pp.__task_pid_nr_ns > 0.36 =C4=85 6% -0.2 0.13 =C4=85 14% perf-profile.self= .cycles-pp.switch_mm_irqs_off > 0.31 =C4=85 6% -0.2 0.09 =C4=85 6% perf-profile.self= .cycles-pp.__free_one_page > 0.28 =C4=85 5% -0.2 0.06 =C4=85 50% perf-profile.self= .cycles-pp.entry_SYSRETQ_unsafe_stack > 0.27 =C4=85 13% -0.2 0.06 =C4=85 23% perf-profile.self= .cycles-pp.pthread_create@@GLIBC_2.2.5 > 0.30 =C4=85 7% -0.2 0.10 =C4=85 19% perf-profile.self= .cycles-pp.__switch_to > 0.27 =C4=85 4% -0.2 0.10 =C4=85 17% perf-profile.self= .cycles-pp.finish_task_switch > 0.23 =C4=85 7% -0.2 0.06 =C4=85 50% perf-profile.self= .cycles-pp.mas_walk > 0.22 =C4=85 9% -0.2 0.05 =C4=85 48% perf-profile.self= .cycles-pp.__clone > 0.63 =C4=85 5% -0.2 0.46 =C4=85 12% perf-profile.self= .cycles-pp.llist_reverse_order > 0.20 =C4=85 4% -0.2 0.04 =C4=85 72% perf-profile.self= .cycles-pp.entry_SYSCALL_64 > 0.24 =C4=85 10% -0.1 0.09 =C4=85 19% perf-profile.self= .cycles-pp.rmqueue_bulk > 0.18 =C4=85 13% -0.1 0.03 =C4=85101% perf-profile.self= .cycles-pp.__radix_tree_lookup > 0.18 =C4=85 11% -0.1 0.04 =C4=85 71% perf-profile.self= .cycles-pp.stress_pthread_func > 0.36 =C4=85 8% -0.1 0.22 =C4=85 11% perf-profile.self= .cycles-pp.menu_select > 0.22 =C4=85 4% -0.1 0.08 =C4=85 19% perf-profile.self= .cycles-pp.___perf_sw_event > 0.20 =C4=85 13% -0.1 0.07 =C4=85 20% perf-profile.self= .cycles-pp.start_thread > 0.16 =C4=85 13% -0.1 0.03 =C4=85101% perf-profile.self= .cycles-pp.alloc_vmap_area > 0.17 =C4=85 10% -0.1 0.04 =C4=85 73% perf-profile.self= .cycles-pp.kmem_cache_alloc > 0.14 =C4=85 9% -0.1 0.03 =C4=85100% perf-profile.self= .cycles-pp.futex_wake > 0.17 =C4=85 4% -0.1 0.06 =C4=85 11% perf-profile.self= .cycles-pp.dequeue_task_fair > 0.23 =C4=85 6% -0.1 0.12 =C4=85 11% perf-profile.self= .cycles-pp.available_idle_cpu > 0.22 =C4=85 13% -0.1 0.11 =C4=85 12% perf-profile.self= .cycles-pp._find_next_bit > 0.21 =C4=85 7% -0.1 0.10 =C4=85 6% perf-profile.self= .cycles-pp.__rmqueue_pcplist > 0.37 =C4=85 7% -0.1 0.26 =C4=85 8% perf-profile.self= .cycles-pp.native_sched_clock > 0.22 =C4=85 7% -0.1 0.12 =C4=85 21% perf-profile.self= .cycles-pp.__update_load_avg_cfs_rq > 0.19 =C4=85 7% -0.1 0.10 =C4=85 11% perf-profile.self= .cycles-pp.enqueue_entity > 0.15 =C4=85 5% -0.1 0.06 =C4=85 45% perf-profile.self= .cycles-pp.enqueue_task_fair > 0.15 =C4=85 11% -0.1 0.06 =C4=85 17% perf-profile.self= .cycles-pp.__pick_eevdf > 0.13 =C4=85 13% -0.1 0.05 =C4=85 72% perf-profile.self= .cycles-pp.prepare_task_switch > 0.17 =C4=85 10% -0.1 0.08 =C4=85 8% perf-profile.self= .cycles-pp.update_rq_clock_task > 0.54 =C4=85 4% -0.1 0.46 =C4=85 6% perf-profile.self= .cycles-pp.__flush_smp_call_function_queue > 0.14 =C4=85 14% -0.1 0.06 =C4=85 11% perf-profile.self= .cycles-pp.os_xsave > 0.11 =C4=85 10% -0.1 0.03 =C4=85 70% perf-profile.self= .cycles-pp.try_to_wake_up > 0.10 =C4=85 8% -0.1 0.03 =C4=85100% perf-profile.self= .cycles-pp.futex_wait > 0.14 =C4=85 9% -0.1 0.07 =C4=85 10% perf-profile.self= .cycles-pp.update_curr > 0.18 =C4=85 9% -0.1 0.11 =C4=85 14% perf-profile.self= .cycles-pp.idle_cpu > 0.11 =C4=85 11% -0.1 0.04 =C4=85 76% perf-profile.self= .cycles-pp.avg_vruntime > 0.15 =C4=85 10% -0.1 0.08 =C4=85 14% perf-profile.self= .cycles-pp.update_cfs_group > 0.09 =C4=85 9% -0.1 0.03 =C4=85100% perf-profile.self= .cycles-pp.reweight_entity > 0.12 =C4=85 13% -0.1 0.06 =C4=85 8% perf-profile.self= .cycles-pp.do_idle > 0.18 =C4=85 10% -0.1 0.12 =C4=85 13% perf-profile.self= .cycles-pp.__update_load_avg_se > 0.09 =C4=85 17% -0.1 0.04 =C4=85 71% perf-profile.self= .cycles-pp.cpuidle_idle_call > 0.10 =C4=85 11% -0.0 0.06 =C4=85 45% perf-profile.self= .cycles-pp.update_rq_clock > 0.12 =C4=85 15% -0.0 0.07 =C4=85 16% perf-profile.self= .cycles-pp.update_sd_lb_stats > 0.09 =C4=85 5% -0.0 0.05 =C4=85 46% perf-profile.self= .cycles-pp._find_next_and_bit > 0.01 =C4=85223% +0.1 0.08 =C4=85 25% perf-profile.self= .cycles-pp.arch_scale_freq_tick > 0.78 =C4=85 4% +0.1 0.87 =C4=85 4% perf-profile.self= .cycles-pp.default_send_IPI_mask_sequence_phys > 0.14 =C4=85 10% +0.1 0.23 =C4=85 13% perf-profile.self= .cycles-pp.__intel_pmu_enable_all > 0.06 =C4=85 46% +0.1 0.15 =C4=85 19% perf-profile.self= .cycles-pp.cgroup_rstat_updated > 0.19 =C4=85 3% +0.1 0.29 =C4=85 4% perf-profile.self= .cycles-pp.cpuidle_enter_state > 0.00 +0.1 0.10 =C4=85 11% perf-profile.self.cycl= es-pp.__mod_lruvec_state > 0.00 +0.1 0.11 =C4=85 18% perf-profile.self.cycl= es-pp.__tlb_remove_page_size > 0.00 +0.1 0.12 =C4=85 9% perf-profile.self.cycl= es-pp.vm_normal_page > 0.23 =C4=85 7% +0.1 0.36 =C4=85 8% perf-profile.self= .cycles-pp.perf_adjust_freq_unthr_context > 0.20 =C4=85 8% +0.2 0.35 =C4=85 7% perf-profile.self= .cycles-pp.__mod_lruvec_page_state > 1.12 =C4=85 2% +0.2 1.28 =C4=85 4% perf-profile.self= .cycles-pp.zap_pte_range > 0.31 =C4=85 8% +0.2 0.46 =C4=85 9% perf-profile.self= .cycles-pp.native_flush_tlb_local > 0.00 +0.2 0.16 =C4=85 5% perf-profile.self.cycl= es-pp._compound_head > 0.06 =C4=85 17% +0.2 0.26 =C4=85 4% perf-profile.self= .cycles-pp.__mod_node_page_state > 0.00 +0.2 0.24 =C4=85 6% perf-profile.self.cycl= es-pp.free_swap_cache > 0.00 +0.3 0.27 =C4=85 15% perf-profile.self.cycl= es-pp.clear_huge_page > 0.00 +0.3 0.27 =C4=85 11% perf-profile.self.cycl= es-pp.deferred_split_folio > 0.00 +0.4 0.36 =C4=85 13% perf-profile.self.cycl= es-pp.prep_compound_page > 0.05 =C4=85 47% +0.4 0.43 =C4=85 9% perf-profile.self= .cycles-pp.free_unref_page_prepare > 0.08 =C4=85 7% +0.5 0.57 =C4=85 23% perf-profile.self= .cycles-pp.__cond_resched > 0.08 =C4=85 12% +0.5 0.58 =C4=85 5% perf-profile.self= .cycles-pp.release_pages > 0.10 =C4=85 10% +0.5 0.63 =C4=85 6% perf-profile.self= .cycles-pp.__mod_memcg_lruvec_state > 0.00 +1.1 1.11 =C4=85 7% perf-profile.self.cycl= es-pp.__split_huge_pmd_locked > 0.00 +1.2 1.18 =C4=85 4% perf-profile.self.cycl= es-pp.page_add_anon_rmap > 0.03 =C4=85101% +1.3 1.35 =C4=85 7% perf-profile.self= .cycles-pp.page_remove_rmap > 0.82 =C4=85 5% +16.1 16.88 =C4=85 7% perf-profile.self= .cycles-pp.clear_page_erms > 11.65 =C4=85 3% +20.2 31.88 =C4=85 2% perf-profile.self= .cycles-pp.native_queued_spin_lock_slowpath > > > *************************************************************************= ************************** > lkp-spr-2sp4: 224 threads 2 sockets Intel(R) Xeon(R) Platinum 8480CTDX (S= apphire Rapids) with 512G memory > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D > array_size/compiler/cpufreq_governor/iterations/kconfig/loop/nr_threads/o= mp/rootfs/tbox_group/testcase: > 50000000/gcc-12/performance/10x/x86_64-rhel-8.3/100/25%/true/debian-11.= 1-x86_64-20220510.cgz/lkp-spr-2sp4/stream > > commit: > 30749e6fbb ("mm/memory: replace kmap() with kmap_local_page()") > 1111d46b5c ("mm: align larger anonymous mappings on THP boundaries") > > 30749e6fbb3d391a 1111d46b5cbad57486e7a3fab75 > ---------------- --------------------------- > %stddev %change %stddev > \ | \ > 10.50 =C4=85 14% +55.6% 16.33 =C4=85 16% perf-c2c.DRAM.loc= al > 6724 -11.4% 5954 =C4=85 2% vmstat.system.cs > 2.746e+09 +16.7% 3.205e+09 =C4=85 2% cpuidle..time > 2771516 +16.0% 3213723 =C4=85 2% cpuidle..usage > 0.06 =C4=85 4% -0.0 0.05 =C4=85 5% mpstat.cpu.all.so= ft% > 0.47 =C4=85 2% -0.1 0.39 =C4=85 2% mpstat.cpu.all.sy= s% > 0.01 =C4=85 85% +1700.0% 0.20 =C4=85188% perf-sched.sch_de= lay.avg.ms.syslog_print.do_syslog.kmsg_read.vfs_read > 15.11 =C4=85 13% -28.8% 10.76 =C4=85 34% perf-sched.wait_a= nd_delay.avg.ms.worker_thread.kthread.ret_from_fork.ret_from_fork_asm > 15.09 =C4=85 13% -30.3% 10.51 =C4=85 38% perf-sched.wait_t= ime.avg.ms.worker_thread.kthread.ret_from_fork.ret_from_fork_asm > 1023952 +13.4% 1161219 meminfo.AnonHugePages > 1319741 +10.8% 1461995 meminfo.AnonPages > 1331039 +11.2% 1480149 meminfo.Inactive > 1330865 +11.2% 1479975 meminfo.Inactive(anon) > 1266202 +16.0% 1469399 =C4=85 2% turbostat.C1E > 1509871 +16.6% 1760853 =C4=85 2% turbostat.C6 > 3521203 +17.4% 4134075 =C4=85 3% turbostat.IRQ > 580.32 -3.8% 558.30 turbostat.PkgWatt > 77.42 -14.0% 66.60 =C4=85 2% turbostat.RAMWatt > 330416 +10.8% 366020 proc-vmstat.nr_anon_pages > 500.90 +13.4% 567.99 proc-vmstat.nr_anon_transpa= rent_hugepages > 333197 +11.2% 370536 proc-vmstat.nr_inactive_ano= n > 333197 +11.2% 370536 proc-vmstat.nr_zone_inactiv= e_anon > 129879 =C4=85 11% -46.7% 69207 =C4=85 12% proc-vmstat.numa_= pages_migrated > 3879028 +5.9% 4109180 proc-vmstat.pgalloc_normal > 3403414 +6.6% 3628929 proc-vmstat.pgfree > 129879 =C4=85 11% -46.7% 69207 =C4=85 12% proc-vmstat.pgmig= rate_success > 5763 +9.8% 6327 proc-vmstat.thp_fault_alloc > 350993 -15.6% 296081 =C4=85 2% stream.add_bandwidth_M= Bps > 349830 -16.1% 293492 =C4=85 2% stream.add_bandwidth_M= Bps_harmonicMean > 333973 -20.5% 265439 =C4=85 3% stream.copy_bandwidth_= MBps > 332930 -21.7% 260548 =C4=85 3% stream.copy_bandwidth_= MBps_harmonicMean > 302788 -16.2% 253817 =C4=85 2% stream.scale_bandwidth= _MBps > 302157 -17.1% 250577 =C4=85 2% stream.scale_bandwidth= _MBps_harmonicMean > 1177276 +9.3% 1286614 stream.time.maximum_residen= t_set_size > 5038 +1.1% 5095 stream.time.percent_of_cpu_= this_job_got > 694.19 =C4=85 2% +19.5% 829.85 =C4=85 2% stream.time.user_= time > 339047 -12.1% 298061 stream.triad_bandwidth_MBps > 338186 -12.4% 296218 stream.triad_bandwidth_MBps= _harmonicMean > 8.42 =C4=85100% -8.4 0.00 perf-profile.calltrace= .cycles-pp.asm_sysvec_reschedule_ipi > 8.42 =C4=85100% -8.4 0.00 perf-profile.calltrace= .cycles-pp.irqentry_exit_to_user_mode.asm_sysvec_reschedule_ipi > 8.42 =C4=85100% -8.4 0.00 perf-profile.calltrace= .cycles-pp.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_sysvec_= reschedule_ipi > 8.42 =C4=85100% -8.4 0.00 perf-profile.calltrace= .cycles-pp.exit_to_user_mode_loop.exit_to_user_mode_prepare.irqentry_exit_t= o_user_mode.asm_sysvec_reschedule_ipi > 8.42 =C4=85100% -8.4 0.00 perf-profile.calltrace= .cycles-pp.arch_do_signal_or_restart.exit_to_user_mode_loop.exit_to_user_mo= de_prepare.irqentry_exit_to_user_mode.asm_sysvec_reschedule_ipi > 8.42 =C4=85100% -8.4 0.00 perf-profile.calltrace= .cycles-pp.get_signal.arch_do_signal_or_restart.exit_to_user_mode_loop.exit= _to_user_mode_prepare.irqentry_exit_to_user_mode > 0.84 =C4=85103% +1.7 2.57 =C4=85 59% perf-profile.call= trace.cycles-pp.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state.cpuidle= _enter.cpuidle_idle_call.do_idle > 0.84 =C4=85103% +1.7 2.57 =C4=85 59% perf-profile.call= trace.cycles-pp.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt= .cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call > 0.31 =C4=85223% +2.0 2.33 =C4=85 44% perf-profile.call= trace.cycles-pp.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt.a= sm_sysvec_apic_timer_interrupt.cpuidle_enter_state.cpuidle_enter > 0.31 =C4=85223% +2.0 2.33 =C4=85 44% perf-profile.call= trace.cycles-pp.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic= _timer_interrupt.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state > 3.07 =C4=85 56% +2.8 5.88 =C4=85 28% perf-profile.call= trace.cycles-pp.do_exit.do_group_exit.__x64_sys_exit_group.do_syscall_64.en= try_SYSCALL_64_after_hwframe > 8.42 =C4=85100% -8.4 0.00 perf-profile.children.= cycles-pp.asm_sysvec_reschedule_ipi > 8.42 =C4=85100% -8.1 0.36 =C4=85223% perf-profile.chil= dren.cycles-pp.irqentry_exit_to_user_mode > 12.32 =C4=85 25% -6.6 5.69 =C4=85 69% perf-profile.chil= dren.cycles-pp.vsnprintf > 12.76 =C4=85 27% -6.6 6.19 =C4=85 67% perf-profile.chil= dren.cycles-pp.seq_printf > 3.07 =C4=85 56% +2.8 5.88 =C4=85 28% perf-profile.chil= dren.cycles-pp.__x64_sys_exit_group > 40.11 -11.0% 35.71 =C4=85 2% perf-stat.i.MPKI > 1.563e+10 -12.3% 1.371e+10 =C4=85 2% perf-stat.i.branch-ins= tructions > 3.721e+09 =C4=85 2% -23.2% 2.858e+09 =C4=85 4% perf-stat.i.cache= -misses > 4.471e+09 =C4=85 3% -22.7% 3.458e+09 =C4=85 4% perf-stat.i.cache= -references > 5970 =C4=85 5% -15.9% 5021 =C4=85 4% perf-stat.i.conte= xt-switches > 1.66 =C4=85 2% +15.8% 1.92 =C4=85 2% perf-stat.i.cpi > 41.83 =C4=85 4% +30.6% 54.63 =C4=85 4% perf-stat.i.cycle= s-between-cache-misses > 2.282e+10 =C4=85 2% -14.5% 1.952e+10 =C4=85 2% perf-stat.i.dTLB-= loads > 572602 =C4=85 3% -9.2% 519922 =C4=85 5% perf-stat.i.dTLB-= store-misses > 1.483e+10 =C4=85 2% -15.7% 1.25e+10 =C4=85 2% perf-stat.i.dTLB-= stores > 9.179e+10 -13.7% 7.924e+10 =C4=85 2% perf-stat.i.instructio= ns > 0.61 -13.4% 0.52 =C4=85 2% perf-stat.i.ipc > 373.79 =C4=85 4% -37.8% 232.60 =C4=85 9% perf-stat.i.metri= c.K/sec > 251.45 -13.4% 217.72 =C4=85 2% perf-stat.i.metric.M/s= ec > 21446 =C4=85 3% -24.1% 16278 =C4=85 8% perf-stat.i.minor= -faults > 15.07 =C4=85 5% -6.0 9.10 =C4=85 10% perf-stat.i.node-= load-miss-rate% > 68275790 =C4=85 5% -44.9% 37626128 =C4=85 12% perf-stat.i.node-= load-misses > 21448 =C4=85 3% -24.1% 16281 =C4=85 8% perf-stat.i.page-= faults > 40.71 -11.3% 36.10 =C4=85 2% perf-stat.overall.MPKI > 1.67 +15.3% 1.93 =C4=85 2% perf-stat.overall.cpi > 41.07 =C4=85 3% +30.1% 53.42 =C4=85 4% perf-stat.overall= .cycles-between-cache-misses > 0.00 =C4=85 2% +0.0 0.00 =C4=85 2% perf-stat.overall= .dTLB-store-miss-rate% > 0.60 -13.2% 0.52 =C4=85 2% perf-stat.overall.ipc > 15.19 =C4=85 5% -6.2 9.03 =C4=85 11% perf-stat.overall= .node-load-miss-rate% > 1.4e+10 -9.3% 1.269e+10 perf-stat.ps.branch-instruc= tions > 3.352e+09 =C4=85 3% -20.9% 2.652e+09 =C4=85 4% perf-stat.ps.cach= e-misses > 4.026e+09 =C4=85 3% -20.3% 3.208e+09 =C4=85 4% perf-stat.ps.cach= e-references > 4888 =C4=85 4% -10.8% 4362 =C4=85 3% perf-stat.ps.cont= ext-switches > 206092 +2.1% 210375 perf-stat.ps.cpu-clock > 1.375e+11 +2.8% 1.414e+11 perf-stat.ps.cpu-cycles > 258.23 =C4=85 5% +8.8% 280.85 =C4=85 4% perf-stat.ps.cpu-= migrations > 2.048e+10 -11.7% 1.809e+10 =C4=85 2% perf-stat.ps.dTLB-load= s > 1.333e+10 =C4=85 2% -13.0% 1.16e+10 =C4=85 2% perf-stat.ps.dTLB= -stores > 8.231e+10 -10.8% 7.342e+10 perf-stat.ps.instructions > 15755 =C4=85 3% -16.3% 13187 =C4=85 6% perf-stat.ps.mino= r-faults > 61706790 =C4=85 6% -43.8% 34699716 =C4=85 11% perf-stat.ps.node= -load-misses > 15757 =C4=85 3% -16.3% 13189 =C4=85 6% perf-stat.ps.page= -faults > 206092 +2.1% 210375 perf-stat.ps.task-clock > 1.217e+12 +4.1% 1.267e+12 =C4=85 2% perf-stat.total.instru= ctions > > > > *************************************************************************= ************************** > lkp-cfl-d1: 12 threads 1 sockets Intel(R) Core(TM) i7-8700 CPU @ 3.20GHz = (Coffee Lake) with 16G memory > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D > compiler/cpufreq_governor/kconfig/option_a/option_b/rootfs/tbox_group/tes= t/testcase: > gcc-12/performance/x86_64-rhel-8.3/Average/Integer/debian-x86_64-phoron= ix/lkp-cfl-d1/ramspeed-1.4.3/phoronix-test-suite > > commit: > 30749e6fbb ("mm/memory: replace kmap() with kmap_local_page()") > 1111d46b5c ("mm: align larger anonymous mappings on THP boundaries") > > 30749e6fbb3d391a 1111d46b5cbad57486e7a3fab75 > ---------------- --------------------------- > %stddev %change %stddev > \ | \ > 232.12 =C4=85 7% -12.0% 204.18 =C4=85 8% sched_debug.cfs_r= q:/.load_avg.stddev > 6797 -3.3% 6576 vmstat.system.cs > 15161 -0.9% 15029 vmstat.system.in > 349927 +44.3% 504820 meminfo.AnonHugePages > 507807 +27.1% 645169 meminfo.AnonPages > 1499332 +10.2% 1652612 meminfo.Inactive(anon) > 8.67 =C4=85 62% +184.6% 24.67 =C4=85 25% turbostat.C10 > 1.50 -0.1 1.45 turbostat.C1E% > 3.30 -3.2% 3.20 turbostat.RAMWatt > 1.40 =C4=85 14% -0.3 1.09 =C4=85 13% perf-profile.call= trace.cycles-pp.asm_exc_page_fault > 1.44 =C4=85 12% -0.3 1.12 =C4=85 13% perf-profile.chil= dren.cycles-pp.asm_exc_page_fault > 0.03 =C4=85141% +0.1 0.10 =C4=85 30% perf-profile.chil= dren.cycles-pp.next_uptodate_folio > 0.02 =C4=85141% +0.1 0.10 =C4=85 22% perf-profile.chil= dren.cycles-pp.rcu_nocb_flush_deferred_wakeup > 0.02 =C4=85143% +0.1 0.10 =C4=85 25% perf-profile.self= .cycles-pp.next_uptodate_folio > 0.01 =C4=85223% +0.1 0.09 =C4=85 19% perf-profile.self= .cycles-pp.rcu_nocb_flush_deferred_wakeup > 19806 -3.5% 19109 phoronix-test-suite.ramspee= d.Average.Integer.mb_s > 283.70 +3.8% 294.50 phoronix-test-suite.time.el= apsed_time > 283.70 +3.8% 294.50 phoronix-test-suite.time.el= apsed_time.max > 120454 +1.6% 122334 phoronix-test-suite.time.ma= ximum_resident_set_size > 281337 -54.8% 127194 phoronix-test-suite.time.mi= nor_page_faults > 259.13 +4.1% 269.81 phoronix-test-suite.time.us= er_time > 126951 +27.0% 161291 proc-vmstat.nr_anon_pages > 170.86 +44.3% 246.49 proc-vmstat.nr_anon_transpa= rent_hugepages > 355917 -1.0% 352250 proc-vmstat.nr_dirty_backgr= ound_threshold > 712705 -1.0% 705362 proc-vmstat.nr_dirty_thresh= old > 3265201 -1.1% 3228465 proc-vmstat.nr_free_pages > 374833 +10.2% 413153 proc-vmstat.nr_inactive_ano= n > 1767 +4.8% 1853 proc-vmstat.nr_page_table_p= ages > 374833 +10.2% 413153 proc-vmstat.nr_zone_inactiv= e_anon > 854665 -34.3% 561406 proc-vmstat.numa_hit > 854632 -34.3% 561397 proc-vmstat.numa_local > 5548755 +1.1% 5610598 proc-vmstat.pgalloc_normal > 1083315 -26.2% 799129 proc-vmstat.pgfault > 113425 +3.7% 117656 proc-vmstat.pgreuse > 9025 +7.6% 9714 proc-vmstat.thp_fault_alloc > 3.38 +0.1 3.45 perf-stat.i.branch-miss-rat= e% > 4.135e+08 -3.2% 4.003e+08 perf-stat.i.cache-misses > 5.341e+08 -2.7% 5.197e+08 perf-stat.i.cache-reference= s > 6832 -3.4% 6600 perf-stat.i.context-switche= s > 4.06 +3.1% 4.19 perf-stat.i.cpi > 438639 =C4=85 5% -18.7% 356730 =C4=85 6% perf-stat.i.dTLB-= load-misses > 1.119e+09 -3.8% 1.077e+09 perf-stat.i.dTLB-loads > 0.02 =C4=85 15% -0.0 0.01 =C4=85 26% perf-stat.i.dTLB-= store-miss-rate% > 80407 =C4=85 10% -63.5% 29387 =C4=85 23% perf-stat.i.dTLB-= store-misses > 7.319e+08 -3.8% 7.043e+08 perf-stat.i.dTLB-stores > 57.72 +0.8 58.52 perf-stat.i.iTLB-load-miss-= rate% > 129846 -3.8% 124973 perf-stat.i.iTLB-load-misse= s > 144448 -5.3% 136837 perf-stat.i.iTLB-loads > 2.389e+09 -3.5% 2.305e+09 perf-stat.i.instructions > 0.28 -2.9% 0.27 perf-stat.i.ipc > 220.59 -3.4% 213.11 perf-stat.i.metric.M/sec > 3610 -31.2% 2483 perf-stat.i.minor-faults > 49238342 +1.1% 49776834 perf-stat.i.node-loads > 98106028 -3.1% 95018390 perf-stat.i.node-stores > 3615 -31.2% 2487 perf-stat.i.page-faults > 3.65 +3.7% 3.78 perf-stat.overall.cpi > 21.08 +3.3% 21.79 perf-stat.overall.cycles-be= tween-cache-misses > 0.04 =C4=85 5% -0.0 0.03 =C4=85 6% perf-stat.overall= .dTLB-load-miss-rate% > 0.01 =C4=85 10% -0.0 0.00 =C4=85 23% perf-stat.overall= .dTLB-store-miss-rate% > 0.27 -3.6% 0.26 perf-stat.overall.ipc > 4.122e+08 -3.2% 3.99e+08 perf-stat.ps.cache-misses > 5.324e+08 -2.7% 5.181e+08 perf-stat.ps.cache-referenc= es > 6809 -3.4% 6580 perf-stat.ps.context-switch= es > 437062 =C4=85 5% -18.7% 355481 =C4=85 6% perf-stat.ps.dTLB= -load-misses > 1.115e+09 -3.8% 1.073e+09 perf-stat.ps.dTLB-loads > 80134 =C4=85 10% -63.5% 29283 =C4=85 23% perf-stat.ps.dTLB= -store-misses > 7.295e+08 -3.8% 7.021e+08 perf-stat.ps.dTLB-stores > 129362 -3.7% 124535 perf-stat.ps.iTLB-load-miss= es > 143865 -5.2% 136338 perf-stat.ps.iTLB-loads > 2.381e+09 -3.5% 2.297e+09 perf-stat.ps.instructions > 3596 -31.2% 2473 perf-stat.ps.minor-faults > 49081949 +1.1% 49621463 perf-stat.ps.node-loads > 97795918 -3.1% 94724831 perf-stat.ps.node-stores > 3600 -31.2% 2477 perf-stat.ps.page-faults > > > > *************************************************************************= ************************** > lkp-cfl-d1: 12 threads 1 sockets Intel(R) Core(TM) i7-8700 CPU @ 3.20GHz = (Coffee Lake) with 16G memory > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D > compiler/cpufreq_governor/kconfig/option_a/option_b/rootfs/tbox_group/tes= t/testcase: > gcc-12/performance/x86_64-rhel-8.3/Average/Floating Point/debian-x86_64= -phoronix/lkp-cfl-d1/ramspeed-1.4.3/phoronix-test-suite > > commit: > 30749e6fbb ("mm/memory: replace kmap() with kmap_local_page()") > 1111d46b5c ("mm: align larger anonymous mappings on THP boundaries") > > 30749e6fbb3d391a 1111d46b5cbad57486e7a3fab75 > ---------------- --------------------------- > %stddev %change %stddev > \ | \ > 167.28 =C4=85 5% -13.1% 145.32 =C4=85 6% sched_debug.cfs_r= q:/.util_est_enqueued.avg > 6845 -2.5% 6674 vmstat.system.cs > 351910 =C4=85 2% +40.2% 493341 meminfo.AnonHugePages > 505908 +27.2% 643328 meminfo.AnonPages > 1497656 +10.2% 1650453 meminfo.Inactive(anon) > 18957 =C4=85 13% +26.3% 23947 =C4=85 17% turbostat.C1 > 1.52 -0.0 1.48 turbostat.C1E% > 3.32 -2.9% 3.23 turbostat.RAMWatt > 19978 -3.0% 19379 phoronix-test-suite.ramspee= d.Average.FloatingPoint.mb_s > 280.71 +3.3% 289.93 phoronix-test-suite.time.el= apsed_time > 280.71 +3.3% 289.93 phoronix-test-suite.time.el= apsed_time.max > 120465 +1.5% 122257 phoronix-test-suite.time.ma= ximum_resident_set_size > 281047 -54.7% 127190 phoronix-test-suite.time.mi= nor_page_faults > 257.03 +3.5% 265.95 phoronix-test-suite.time.us= er_time > 126473 +27.2% 160831 proc-vmstat.nr_anon_pages > 171.83 =C4=85 2% +40.2% 240.89 proc-vmstat.nr_anon_tr= ansparent_hugepages > 355973 -1.0% 352304 proc-vmstat.nr_dirty_backgr= ound_threshold > 712818 -1.0% 705471 proc-vmstat.nr_dirty_thresh= old > 3265800 -1.1% 3228879 proc-vmstat.nr_free_pages > 374410 +10.2% 412613 proc-vmstat.nr_inactive_ano= n > 1770 +4.4% 1848 proc-vmstat.nr_page_table_p= ages > 374410 +10.2% 412613 proc-vmstat.nr_zone_inactiv= e_anon > 852082 -34.9% 555093 proc-vmstat.numa_hit > 852125 -34.9% 555018 proc-vmstat.numa_local > 1078293 -26.6% 791038 proc-vmstat.pgfault > 112693 +2.9% 116004 proc-vmstat.pgreuse > 9025 +7.6% 9713 proc-vmstat.thp_fault_alloc > 3.63 =C4=85 6% +0.6 4.25 =C4=85 9% perf-profile.call= trace.cycles-pp.__hrtimer_run_queues.hrtimer_interrupt.__sysvec_apic_timer_= interrupt.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt > 0.25 =C4=85 55% -0.2 0.08 =C4=85 68% perf-profile.chil= dren.cycles-pp.ret_from_fork_asm > 0.25 =C4=85 55% -0.2 0.08 =C4=85 68% perf-profile.chil= dren.cycles-pp.ret_from_fork > 0.23 =C4=85 56% -0.2 0.07 =C4=85 69% perf-profile.chil= dren.cycles-pp.kthread > 0.14 =C4=85 36% -0.1 0.05 =C4=85120% perf-profile.chil= dren.cycles-pp.do_anonymous_page > 0.14 =C4=85 35% -0.1 0.05 =C4=85 76% perf-profile.chil= dren.cycles-pp.copy_mc_enhanced_fast_string > 0.04 =C4=85 72% +0.0 0.08 =C4=85 19% perf-profile.chil= dren.cycles-pp.try_to_wake_up > 0.04 =C4=85118% +0.1 0.10 =C4=85 36% perf-profile.chil= dren.cycles-pp.update_rq_clock > 0.07 =C4=85 79% +0.1 0.17 =C4=85 21% perf-profile.chil= dren.cycles-pp._raw_spin_lock_irqsave > 7.99 =C4=85 11% +1.0 9.02 =C4=85 5% perf-profile.chil= dren.cycles-pp.asm_sysvec_apic_timer_interrupt > 0.23 =C4=85 28% -0.1 0.14 =C4=85 49% perf-profile.self= .cycles-pp.irqentry_exit_to_user_mode > 0.14 =C4=85 35% -0.1 0.05 =C4=85 76% perf-profile.self= .cycles-pp.copy_mc_enhanced_fast_string > 0.06 =C4=85 79% +0.1 0.16 =C4=85 21% perf-profile.self= .cycles-pp._raw_spin_lock_irqsave > 0.21 =C4=85 34% +0.2 0.36 =C4=85 18% perf-profile.self= .cycles-pp.ktime_get > 1.187e+08 -4.6% 1.133e+08 perf-stat.i.branch-instruct= ions > 3.36 +0.1 3.42 perf-stat.i.branch-miss-rat= e% > 5492420 -3.9% 5275592 perf-stat.i.branch-misses > 4.148e+08 -2.8% 4.034e+08 perf-stat.i.cache-misses > 5.251e+08 -2.6% 5.114e+08 perf-stat.i.cache-reference= s > 6880 -2.5% 6711 perf-stat.i.context-switche= s > 4.30 +2.9% 4.43 perf-stat.i.cpi > 0.10 =C4=85 7% -0.0 0.09 =C4=85 2% perf-stat.i.dTLB-= load-miss-rate% > 472268 =C4=85 6% -19.9% 378489 perf-stat.i.dTLB-load-= misses > 8.107e+08 -3.4% 7.831e+08 perf-stat.i.dTLB-loads > 0.02 =C4=85 16% -0.0 0.01 =C4=85 2% perf-stat.i.dTLB-= store-miss-rate% > 90535 =C4=85 11% -59.8% 36371 =C4=85 2% perf-stat.i.dTLB-= store-misses > 5.323e+08 -3.3% 5.145e+08 perf-stat.i.dTLB-stores > 129981 -3.0% 126061 perf-stat.i.iTLB-load-misse= s > 143662 -3.1% 139223 perf-stat.i.iTLB-loads > 2.253e+09 -3.6% 2.172e+09 perf-stat.i.instructions > 0.26 -3.2% 0.25 perf-stat.i.ipc > 4.71 =C4=85 2% -6.4% 4.41 =C4=85 2% perf-stat.i.major= -faults > 180.03 -3.0% 174.57 perf-stat.i.metric.M/sec > 3627 -30.8% 2510 =C4=85 2% perf-stat.i.minor-faul= ts > 3632 -30.8% 2514 =C4=85 2% perf-stat.i.page-fault= s > 3.88 +3.6% 4.02 perf-stat.overall.cpi > 21.08 +2.7% 21.65 perf-stat.overall.cycles-be= tween-cache-misses > 0.06 =C4=85 6% -0.0 0.05 perf-stat.overall.dTLB= -load-miss-rate% > 0.02 =C4=85 11% -0.0 0.01 =C4=85 2% perf-stat.overall= .dTLB-store-miss-rate% > 0.26 -3.5% 0.25 perf-stat.overall.ipc > 1.182e+08 -4.6% 1.128e+08 perf-stat.ps.branch-instruc= tions > 5468166 -4.0% 5251939 perf-stat.ps.branch-misses > 4.135e+08 -2.7% 4.021e+08 perf-stat.ps.cache-misses > 5.234e+08 -2.6% 5.098e+08 perf-stat.ps.cache-referenc= es > 6859 -2.5% 6685 perf-stat.ps.context-switch= es > 470567 =C4=85 6% -19.9% 377127 perf-stat.ps.dTLB-load= -misses > 8.079e+08 -3.4% 7.805e+08 perf-stat.ps.dTLB-loads > 90221 =C4=85 11% -59.8% 36239 =C4=85 2% perf-stat.ps.dTLB= -store-misses > 5.305e+08 -3.3% 5.128e+08 perf-stat.ps.dTLB-stores > 129499 -3.0% 125601 perf-stat.ps.iTLB-load-miss= es > 143121 -3.1% 138638 perf-stat.ps.iTLB-loads > 2.246e+09 -3.6% 2.165e+09 perf-stat.ps.instructions > 4.69 =C4=85 2% -6.3% 4.39 =C4=85 2% perf-stat.ps.majo= r-faults > 3613 -30.8% 2500 =C4=85 2% perf-stat.ps.minor-fau= lts > 3617 -30.8% 2504 =C4=85 2% perf-stat.ps.page-faul= ts > > > > > > Disclaimer: > Results have been estimated based on internal Intel analysis and are prov= ided > for informational purposes only. Any difference in system hardware or sof= tware > design or configuration may affect actual performance. > > > -- > 0-DAY CI Kernel Test Service > https://github.com/intel/lkp-tests/wiki >