Greeting, FYI, we noticed a 46.6% improvement of hackbench.throughput due to commit: commit: b08e888d9715f01546da7a40ca3ff72b9d1a78cc ("sched/fair: Force progress on min_vruntime") https://git.kernel.org/cgit/linux/kernel/git/peterz/queue.git sched/wip.migrate in testcase: hackbench on test machine: 128 threads 2 sockets Intel(R) Xeon(R) Gold 6338 CPU @ 2.00GHz with 256G memory with following parameters: nr_threads: 100% iterations: 4 mode: threads ipc: pipe cpufreq_governor: performance ucode: 0xd000280 test-description: Hackbench is both a benchmark and a stress test for the Linux kernel scheduler. test-url: https://github.com/linux-test-project/ltp/blob/master/testcases/kernel/sc... In addition to that, the commit also has significant impact on the following tests: +------------------+------------------------------------------------------------------------------------------------------------+ | testcase: change | phoronix-test-suite: phoronix-test-suite.schbench.32.32.usec,_99.9th_latency_percentile -60.6% improvement | | test machine | 96 threads 2 sockets Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz with 128G memory | | test parameters | cpufreq_governor=performance | | | option_a=32 | | | option_b=32 | | | test=schbench-1.0.0 | | | ucode=0x5003006 | +------------------+------------------------------------------------------------------------------------------------------------+ | testcase: change | hackbench: hackbench.throughput 55.5% improvement | | test machine | 104 threads 2 sockets Skylake with 192G memory | | test parameters | cpufreq_governor=performance | | | ipc=pipe | | | iterations=4 | | | mode=threads | | | nr_threads=100% | | | ucode=0x2006a0a | +------------------+------------------------------------------------------------------------------------------------------------+ Details are as below: --------------------------------------------------------------------------------------------------> To reproduce: git clone https://github.com/intel/lkp-tests.git cd lkp-tests sudo bin/lkp install job.yaml # job file is attached in this email bin/lkp split-job --compatible job.yaml # generate the yaml file for lkp run sudo bin/lkp run generated-yaml-file # if come across any failure that blocks the test, # please remove ~/.lkp and /lkp dir to run from a clean state. ========================================================================================= compiler/cpufreq_governor/ipc/iterations/kconfig/mode/nr_threads/rootfs/tbox_group/testcase/ucode: gcc-9/performance/pipe/4/x86_64-rhel-8.3/threads/100%/debian-10.4-x86_64-20200603.cgz/lkp-icl-2sp2/hackbench/0xd000280 commit: v5.16-rc6 b08e888d97 ("sched/fair: Force progress on min_vruntime") v5.16-rc6 b08e888d9715f01546da7a40ca3 ---------------- --------------------------- %stddev %change %stddev \ | \ 653575 +46.6% 958089 hackbench.throughput 17908198 ± 5% -36.4% 11391663 ± 5% cpuidle..usage 155.26 -24.8% 116.73 ± 3% uptime.boot 0.03 ± 3% -0.0 0.03 ± 6% mpstat.cpu.all.soft% 9.86 +3.3 13.20 mpstat.cpu.all.usr% 2052 ± 20% -35.2% 1329 ± 23% numa-meminfo.node0.Active 2052 ± 20% -35.2% 1329 ± 23% numa-meminfo.node0.Active(anon) 3312 -16.9% 2753 ± 2% meminfo.Active 3312 -16.9% 2753 ± 2% meminfo.Active(anon) 128712 ± 2% -24.9% 96605 meminfo.AnonHugePages 2554 +1.2% 2584 turbostat.Bzy_MHz 17637175 ± 5% -35.5% 11382497 ± 5% turbostat.C1 1.321e+08 ± 4% -77.7% 29492706 ± 3% turbostat.IRQ 1213 +108.1% 2525 ± 3% vmstat.procs.r 23544038 -94.2% 1373085 ± 4% vmstat.system.cs 1267010 ± 3% -65.1% 441932 ± 2% vmstat.system.in 28404703 ± 7% +154.1% 72168987 ± 4% numa-numastat.node0.local_node 28445356 ± 7% +154.0% 72237824 ± 4% numa-numastat.node0.numa_hit 25297912 ± 6% +186.2% 72404938 ± 4% numa-numastat.node1.local_node 25371744 ± 6% +185.5% 72437776 ± 4% numa-numastat.node1.numa_hit 513.50 ± 21% -35.8% 329.67 ± 24% numa-vmstat.node0.nr_active_anon 513.50 ± 21% -35.8% 329.67 ± 24% numa-vmstat.node0.nr_zone_active_anon 16630487 ± 8% +129.4% 38157859 ± 4% numa-vmstat.node0.numa_hit 16587877 ± 8% +129.6% 38086060 ± 4% numa-vmstat.node0.numa_local 14598373 ± 8% +159.4% 37867111 ± 6% numa-vmstat.node1.numa_hit 14483676 ± 8% +160.9% 37790942 ± 6% numa-vmstat.node1.numa_local 6.63 ± 64% -6.0 0.62 ±223% perf-profile.calltrace.cycles-pp.do_filp_open.do_sys_openat2.do_sys_open.do_syscall_64.entry_SYSCALL_64_after_hwframe 6.63 ± 64% -6.0 0.62 ±223% perf-profile.calltrace.cycles-pp.path_openat.do_filp_open.do_sys_openat2.do_sys_open.do_syscall_64 3.75 ± 72% -3.7 0.00 perf-profile.calltrace.cycles-pp.open64 5.44 ± 51% -2.9 2.52 ±163% perf-profile.calltrace.cycles-pp.copy_page_from_iter_atomic.generic_perform_write.__generic_file_write_iter.generic_file_write_iter.new_sync_write 4.48 ± 57% -2.7 1.76 ±141% perf-profile.calltrace.cycles-pp.number.vsnprintf.seq_vprintf.seq_printf.show_interrupts 3.65 ± 77% -1.0 2.67 ±223% perf-profile.calltrace.cycles-pp.proc_pid_status.proc_single_show.seq_read_iter.seq_read.vfs_read 6.63 ± 64% -6.0 0.62 ±223% perf-profile.children.cycles-pp.do_sys_open 6.63 ± 64% -6.0 0.62 ±223% perf-profile.children.cycles-pp.do_sys_openat2 6.63 ± 64% -6.0 0.62 ±223% perf-profile.children.cycles-pp.do_filp_open 6.63 ± 64% -6.0 0.62 ±223% perf-profile.children.cycles-pp.path_openat 5.04 ± 64% -3.3 1.76 ±141% perf-profile.children.cycles-pp.number 5.44 ± 51% -2.9 2.52 ±163% perf-profile.children.cycles-pp.copy_page_from_iter_atomic 5.04 ± 64% -3.3 1.76 ±141% perf-profile.self.cycles-pp.number 826.83 -16.8% 688.33 ± 2% proc-vmstat.nr_active_anon 79216 -1.6% 77968 proc-vmstat.nr_anon_pages 84555 ± 2% -2.6% 82387 proc-vmstat.nr_inactive_anon 92536 -4.6% 88260 proc-vmstat.nr_kernel_stack 6216 ± 31% -16.0% 5220 proc-vmstat.nr_shmem 826.83 -16.8% 688.33 ± 2% proc-vmstat.nr_zone_active_anon 84555 ± 2% -2.6% 82387 proc-vmstat.nr_zone_inactive_anon 40595 ± 24% -85.3% 5957 ± 89% proc-vmstat.numa_hint_faults 20890 ± 41% -87.4% 2629 ±105% proc-vmstat.numa_hint_faults_local 53819891 +168.8% 1.447e+08 proc-vmstat.numa_hit 53705406 +169.2% 1.446e+08 proc-vmstat.numa_local 20887 ± 27% -84.1% 3328 ±100% proc-vmstat.numa_pages_migrated 72657 ± 30% -55.8% 32092 ± 61% proc-vmstat.numa_pte_updates 53822302 +168.7% 1.446e+08 proc-vmstat.pgalloc_normal 449333 ± 13% -33.5% 298589 ± 3% proc-vmstat.pgfault 53610460 +169.4% 1.444e+08 proc-vmstat.pgfree 20887 ± 27% -84.1% 3328 ±100% proc-vmstat.pgmigrate_success 20678 ± 2% -32.9% 13880 ± 4% proc-vmstat.pgreuse 7.20 ± 2% -56.8% 3.11 ± 2% perf-stat.i.MPKI 5.303e+10 -4.0% 5.092e+10 perf-stat.i.branch-instructions 0.63 -0.5 0.16 perf-stat.i.branch-miss-rate% 3.195e+08 -77.6% 71713145 perf-stat.i.branch-misses 7.04 ± 4% +18.3 25.32 ± 3% perf-stat.i.cache-miss-rate% 1.199e+08 ± 3% +41.4% 1.695e+08 ± 2% perf-stat.i.cache-misses 1.824e+09 -59.2% 7.436e+08 ± 2% perf-stat.i.cache-references 24276211 -94.4% 1358591 ± 4% perf-stat.i.context-switches 1521377 ± 3% -87.5% 190633 ± 7% perf-stat.i.cpu-migrations 3127 ± 5% -39.4% 1893 ± 2% perf-stat.i.cycles-between-cache-misses 0.00 ± 6% -0.0 0.00 ± 9% perf-stat.i.dTLB-load-miss-rate% 1804532 ± 5% -23.3% 1383626 ± 10% perf-stat.i.dTLB-load-misses 0.00 ± 8% -0.0 0.00 ± 4% perf-stat.i.dTLB-store-miss-rate% 343753 ± 3% -23.8% 261974 ± 4% perf-stat.i.dTLB-store-misses 4.648e+10 -3.9% 4.464e+10 perf-stat.i.dTLB-stores 2.617e+11 -3.4% 2.528e+11 perf-stat.i.instructions 0.82 -3.5% 0.79 perf-stat.i.ipc 517.04 ± 2% -9.8% 466.28 perf-stat.i.metric.K/sec 1407 -3.3% 1361 perf-stat.i.metric.M/sec 87.20 -17.6 69.59 perf-stat.i.node-load-miss-rate% 2795035 ± 6% +214.8% 8799183 ± 5% perf-stat.i.node-loads 59.48 +3.0 62.46 perf-stat.i.node-store-miss-rate% 11371827 ± 4% +62.7% 18504651 ± 2% perf-stat.i.node-store-misses 7495979 ± 3% +48.1% 11099696 perf-stat.i.node-stores 6.92 ± 2% -57.8% 2.92 ± 2% perf-stat.overall.MPKI 0.60 -0.5 0.14 perf-stat.overall.branch-miss-rate% 6.62 ± 4% +16.3 22.96 ± 4% perf-stat.overall.cache-miss-rate% 1.20 +4.7% 1.26 perf-stat.overall.cpi 2621 ± 3% -28.5% 1873 ± 2% perf-stat.overall.cycles-between-cache-misses 0.00 ± 5% -0.0 0.00 ± 11% perf-stat.overall.dTLB-load-miss-rate% 0.00 ± 2% -0.0 0.00 ± 4% perf-stat.overall.dTLB-store-miss-rate% 0.83 -4.5% 0.80 perf-stat.overall.ipc 86.77 -17.9 68.83 perf-stat.overall.node-load-miss-rate% 60.35 +2.2 62.51 perf-stat.overall.node-store-miss-rate% 5.269e+10 -4.6% 5.025e+10 perf-stat.ps.branch-instructions 3.154e+08 -77.7% 70229688 perf-stat.ps.branch-misses 1.191e+08 ± 3% +40.4% 1.673e+08 ± 2% perf-stat.ps.cache-misses 1.8e+09 ± 2% -59.5% 7.291e+08 ± 2% perf-stat.ps.cache-references 24036022 -94.5% 1320301 ± 4% perf-stat.ps.context-switches 1470959 ± 3% -87.6% 182972 ± 7% perf-stat.ps.cpu-migrations 1759980 ± 5% -23.8% 1341985 ± 11% perf-stat.ps.dTLB-load-misses 335955 ± 2% -24.1% 255091 ± 4% perf-stat.ps.dTLB-store-misses 4.618e+10 -4.6% 4.406e+10 perf-stat.ps.dTLB-stores 2.6e+11 -4.1% 2.495e+11 perf-stat.ps.instructions 2784328 ± 5% +212.5% 8701815 ± 5% perf-stat.ps.node-loads 11244792 ± 4% +62.0% 18220133 ± 2% perf-stat.ps.node-store-misses 7386068 ± 3% +47.9% 10923448 perf-stat.ps.node-stores 2.568e+13 -37.0% 1.619e+13 perf-stat.total.instructions 18746 ± 12% -39.3% 11371 ± 9% softirqs.CPU0.RCU 19003 ± 13% -37.5% 11885 ± 16% softirqs.CPU1.RCU 19070 ± 12% -42.2% 11015 ± 9% softirqs.CPU10.RCU 19263 ± 6% -38.0% 11951 ± 6% softirqs.CPU100.RCU 19660 ± 7% -39.6% 11878 ± 5% softirqs.CPU101.RCU 19396 ± 6% -38.4% 11942 ± 5% softirqs.CPU102.RCU 19208 ± 6% -38.0% 11912 ± 5% softirqs.CPU103.RCU 19759 ± 7% -40.1% 11826 ± 5% softirqs.CPU104.RCU 19676 ± 6% -40.1% 11787 ± 5% softirqs.CPU105.RCU 19141 ± 6% -38.1% 11849 ± 6% softirqs.CPU106.RCU 19236 ± 6% -38.2% 11883 ± 6% softirqs.CPU107.RCU 19948 ± 5% -39.7% 12033 ± 5% softirqs.CPU108.RCU 19426 ± 6% -39.1% 11832 ± 6% softirqs.CPU109.RCU 18843 ± 11% -36.0% 12060 ± 21% softirqs.CPU11.RCU 19548 ± 6% -39.8% 11765 ± 6% softirqs.CPU110.RCU 19460 ± 6% -38.7% 11920 ± 5% softirqs.CPU111.RCU 17625 ± 6% -39.9% 10596 ± 4% softirqs.CPU112.RCU 17477 ± 6% -40.1% 10472 ± 5% softirqs.CPU113.RCU 17702 ± 8% -40.6% 10506 ± 5% softirqs.CPU114.RCU 17433 ± 7% -39.6% 10530 ± 4% softirqs.CPU115.RCU 17184 ± 9% -38.5% 10572 ± 5% softirqs.CPU116.RCU 17123 ± 6% -38.5% 10534 ± 8% softirqs.CPU117.RCU 17453 ± 6% -39.9% 10497 ± 5% softirqs.CPU118.RCU 17449 ± 7% -40.4% 10395 ± 5% softirqs.CPU119.RCU 18501 ± 12% -39.6% 11180 ± 10% softirqs.CPU12.RCU 17417 ± 6% -39.1% 10600 ± 5% softirqs.CPU120.RCU 17554 ± 6% -40.2% 10501 ± 6% softirqs.CPU121.RCU 18068 ± 12% -39.4% 10945 ± 7% softirqs.CPU122.RCU 17356 ± 5% -39.1% 10574 ± 5% softirqs.CPU123.RCU 17411 ± 6% -38.5% 10715 ± 5% softirqs.CPU124.RCU 17615 ± 6% -39.2% 10702 ± 6% softirqs.CPU125.RCU 17569 ± 6% -39.5% 10631 ± 5% softirqs.CPU126.RCU 17797 ± 6% -40.2% 10635 ± 5% softirqs.CPU127.RCU 19570 ± 13% -38.9% 11967 ± 21% softirqs.CPU13.RCU 18846 ± 11% -38.5% 11586 ± 13% softirqs.CPU14.RCU 18642 ± 13% -39.6% 11255 ± 10% softirqs.CPU15.RCU 18684 ± 12% -41.8% 10869 ± 10% softirqs.CPU16.RCU 18890 ± 14% -42.9% 10790 ± 11% softirqs.CPU17.RCU 18794 ± 13% -40.0% 11278 ± 6% softirqs.CPU18.RCU 18930 ± 13% -41.4% 11096 ± 11% softirqs.CPU19.RCU 18567 ± 13% -40.6% 11028 ± 9% softirqs.CPU2.RCU 18769 ± 12% -41.4% 11006 ± 9% softirqs.CPU20.RCU 18500 ± 12% -40.3% 11040 ± 12% softirqs.CPU21.RCU 18477 ± 14% -42.3% 10655 ± 10% softirqs.CPU22.RCU 18551 ± 12% -43.4% 10507 ± 18% softirqs.CPU23.RCU 18638 ± 12% -42.0% 10805 ± 21% softirqs.CPU24.RCU 18602 ± 12% -42.8% 10642 ± 8% softirqs.CPU25.RCU 18739 ± 12% -41.3% 11006 ± 9% softirqs.CPU26.RCU 18763 ± 13% -41.9% 10900 ± 10% softirqs.CPU27.RCU 18268 ± 19% -40.3% 10898 ± 12% softirqs.CPU28.RCU 18584 ± 12% -42.4% 10706 ± 9% softirqs.CPU29.RCU 19256 ± 12% -43.5% 10889 ± 11% softirqs.CPU3.RCU 18713 ± 12% -41.8% 10887 ± 8% softirqs.CPU30.RCU 18710 ± 13% -41.4% 10960 ± 9% softirqs.CPU31.RCU 18442 ± 5% -40.2% 11023 ± 4% softirqs.CPU32.RCU 17972 ± 7% -38.1% 11126 ± 5% softirqs.CPU33.RCU 18139 ± 6% -39.4% 10986 ± 4% softirqs.CPU34.RCU 17619 ± 6% -38.7% 10795 ± 5% softirqs.CPU35.RCU 17452 ± 6% -38.1% 10795 ± 3% softirqs.CPU36.RCU 17860 ± 7% -38.0% 11079 ± 4% softirqs.CPU37.RCU 17703 ± 6% -39.5% 10710 ± 10% softirqs.CPU38.RCU 17733 ± 4% -39.9% 10655 ± 5% softirqs.CPU39.RCU 19506 ± 15% -43.6% 11001 ± 8% softirqs.CPU4.RCU 18044 ± 6% -39.7% 10882 ± 5% softirqs.CPU40.RCU 18256 ± 6% -40.4% 10881 ± 5% softirqs.CPU41.RCU 17318 ± 7% -37.5% 10823 ± 5% softirqs.CPU42.RCU 17660 ± 7% -40.3% 10539 ± 5% softirqs.CPU43.RCU 18660 ± 5% -41.9% 10844 ± 4% softirqs.CPU44.RCU 17768 ± 6% -38.9% 10852 ± 6% softirqs.CPU45.RCU 17655 ± 7% -39.3% 10713 ± 5% softirqs.CPU46.RCU 17747 ± 6% -39.1% 10816 ± 5% softirqs.CPU47.RCU 17656 ± 6% -39.8% 10636 ± 5% softirqs.CPU48.RCU 17592 ± 6% -39.5% 10646 ± 4% softirqs.CPU49.RCU 19129 ± 13% -39.4% 11599 ± 19% softirqs.CPU5.RCU 17586 ± 6% -40.1% 10540 ± 4% softirqs.CPU50.RCU 17758 ± 5% -40.5% 10570 ± 4% softirqs.CPU51.RCU 17960 ± 7% -40.1% 10765 ± 4% softirqs.CPU52.RCU 16892 ± 8% -39.3% 10256 ± 7% softirqs.CPU53.RCU 17597 ± 5% -40.2% 10524 ± 4% softirqs.CPU54.RCU 17614 ± 6% -40.9% 10412 ± 5% softirqs.CPU55.RCU 17682 ± 7% -40.0% 10600 ± 4% softirqs.CPU56.RCU 18064 ± 7% -40.1% 10816 ± 5% softirqs.CPU57.RCU 17623 ± 6% -40.4% 10500 ± 4% softirqs.CPU58.RCU 17585 ± 6% -40.6% 10450 ± 4% softirqs.CPU59.RCU 18840 ± 12% -40.9% 11139 ± 8% softirqs.CPU6.RCU 17667 ± 6% -39.1% 10761 ± 7% softirqs.CPU60.RCU 17780 ± 6% -37.0% 11206 ± 5% softirqs.CPU61.RCU 17766 ± 6% -37.9% 11029 ± 9% softirqs.CPU62.RCU 17961 ± 7% -42.4% 10348 ± 8% softirqs.CPU63.RCU 18806 ± 15% -43.0% 10724 ± 8% softirqs.CPU64.RCU 18615 ± 13% -39.7% 11220 ± 10% softirqs.CPU65.RCU 18447 ± 13% -37.8% 11482 ± 16% softirqs.CPU66.RCU 18479 ± 13% -40.7% 10960 ± 11% softirqs.CPU67.RCU 18815 ± 13% -42.0% 10916 ± 10% softirqs.CPU68.RCU 18691 ± 13% -41.6% 10912 ± 9% softirqs.CPU69.RCU 19163 ± 13% -40.6% 11392 ± 13% softirqs.CPU7.RCU 18358 ± 13% -40.2% 10982 ± 9% softirqs.CPU70.RCU 18096 ± 19% -40.1% 10848 ± 9% softirqs.CPU71.RCU 18737 ± 13% -42.4% 10789 ± 8% softirqs.CPU72.RCU 19085 ± 11% -43.1% 10864 ± 7% softirqs.CPU73.RCU 18868 ± 13% -37.1% 11868 ± 9% softirqs.CPU74.RCU 18523 ± 13% -41.3% 10876 ± 11% softirqs.CPU75.RCU 18366 ± 13% -40.9% 10850 ± 10% softirqs.CPU76.RCU 19087 ± 12% -43.1% 10869 ± 9% softirqs.CPU77.RCU 18624 ± 11% -41.4% 10910 ± 9% softirqs.CPU78.RCU 18660 ± 12% -41.6% 10889 ± 10% softirqs.CPU79.RCU 18591 ± 12% -41.0% 10971 ± 9% softirqs.CPU8.RCU 18670 ± 13% -42.9% 10654 ± 8% softirqs.CPU80.RCU 19309 ± 16% -44.8% 10652 ± 11% softirqs.CPU81.RCU 18675 ± 12% -42.4% 10753 ± 9% softirqs.CPU82.RCU 18285 ± 13% -41.3% 10735 ± 10% softirqs.CPU83.RCU 18831 ± 13% -41.6% 11003 ± 8% softirqs.CPU84.RCU 18235 ± 13% -41.1% 10734 ± 11% softirqs.CPU85.RCU 18344 ± 13% -41.6% 10721 ± 7% softirqs.CPU86.RCU 18706 ± 13% -41.9% 10860 ± 14% softirqs.CPU87.RCU 18387 ± 13% -41.2% 10809 ± 13% softirqs.CPU88.RCU 18812 ± 12% -42.0% 10904 ± 8% softirqs.CPU89.RCU 18664 ± 11% -40.3% 11136 ± 10% softirqs.CPU9.RCU 18902 ± 13% -41.9% 10976 ± 10% softirqs.CPU90.RCU 18687 ± 12% -39.1% 11374 ± 14% softirqs.CPU91.RCU 18283 ± 14% -40.3% 10915 ± 10% softirqs.CPU92.RCU 18476 ± 12% -41.7% 10766 ± 10% softirqs.CPU93.RCU 18580 ± 13% -40.2% 11119 ± 7% softirqs.CPU94.RCU 18675 ± 13% -40.3% 11149 ± 10% softirqs.CPU95.RCU 19611 ± 6% -38.8% 12003 ± 5% softirqs.CPU96.RCU 19352 ± 6% -37.9% 12016 ± 5% softirqs.CPU97.RCU 20041 ± 5% -40.3% 11974 ± 4% softirqs.CPU98.RCU 19370 ± 6% -39.1% 11792 ± 6% softirqs.CPU99.RCU 2358890 ± 9% -40.3% 1409034 ± 6% softirqs.RCU 732676 ± 8% -36.8% 463351 ± 9% softirqs.SCHED 33955 ± 5% -19.4% 27373 ± 9% softirqs.TIMER *************************************************************************************************** lkp-csl-2sp8: 96 threads 2 sockets Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz with 128G memory ========================================================================================= compiler/cpufreq_governor/kconfig/option_a/option_b/rootfs/tbox_group/test/testcase/ucode: gcc-9/performance/x86_64-rhel-8.3/32/32/debian-x86_64-phoronix/lkp-csl-2sp8/schbench-1.0.0/phoronix-test-suite/0x5003006 commit: v5.16-rc6 b08e888d97 ("sched/fair: Force progress on min_vruntime") v5.16-rc6 b08e888d9715f01546da7a40ca3 ---------------- --------------------------- %stddev %change %stddev \ | \ 135133 -60.6% 53294 phoronix-test-suite.schbench.32.32.usec,_99.9th_latency_percentile 2038805 -15.1% 1730336 phoronix-test-suite.time.involuntary_context_switches 0.14 ± 2% -0.0 0.11 ± 4% mpstat.cpu.all.sys% 1096 ± 31% +45.4% 1594 ± 6% proc-vmstat.numa_huge_pte_updates 669991 ± 26% +38.5% 928203 ± 5% proc-vmstat.numa_pte_updates 11.39 ± 5% +1.7 13.12 ± 5% perf-stat.i.cache-miss-rate% 30800 -10.5% 27573 perf-stat.i.context-switches 3842 +22.1% 4690 perf-stat.i.cpu-migrations 609.10 -9.8% 549.40 vmstat.procs.r 29868 -9.4% 27049 vmstat.system.cs 201252 ± 9% +4.6% 210516 vmstat.system.in 0.84 ± 7% +0.2 1.07 ± 5% perf-profile.calltrace.cycles-pp.rebalance_domains.__softirqentry_text_start.irq_exit_rcu.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt 0.78 ± 7% +0.2 1.02 ± 5% perf-profile.calltrace.cycles-pp.load_balance.rebalance_domains.__softirqentry_text_start.irq_exit_rcu.sysvec_apic_timer_interrupt 0.72 ± 11% +0.3 1.03 ± 5% perf-profile.calltrace.cycles-pp.newidle_balance.pick_next_task_fair.__schedule.schedule.futex_wait_queue 0.69 ± 12% +0.3 1.00 ± 5% perf-profile.calltrace.cycles-pp.load_balance.newidle_balance.pick_next_task_fair.__schedule.schedule 0.90 ± 11% +0.3 1.22 ± 4% perf-profile.calltrace.cycles-pp.pick_next_task_fair.__schedule.schedule.futex_wait_queue.futex_wait 5.69 ± 3% +0.5 6.23 ± 4% perf-profile.calltrace.cycles-pp.__schedule.schedule.futex_wait_queue.futex_wait.do_futex 5.73 ± 3% +0.5 6.27 ± 4% perf-profile.calltrace.cycles-pp.schedule.futex_wait_queue.futex_wait.do_futex.__x64_sys_futex 5.82 ± 3% +0.6 6.38 ± 4% perf-profile.calltrace.cycles-pp.futex_wait_queue.futex_wait.do_futex.__x64_sys_futex.do_syscall_64 6.02 ± 3% +0.6 6.58 ± 3% perf-profile.calltrace.cycles-pp.futex_wait.do_futex.__x64_sys_futex.do_syscall_64.entry_SYSCALL_64_after_hwframe 6.50 ± 4% -1.5 5.04 ± 5% perf-profile.children.cycles-pp.exit_to_user_mode_prepare 6.19 ± 5% -1.3 4.89 ± 5% perf-profile.children.cycles-pp.irqentry_exit_to_user_mode 12.46 ± 2% -1.1 11.32 ± 3% perf-profile.children.cycles-pp.__schedule 12.52 ± 2% -1.1 11.39 ± 3% perf-profile.children.cycles-pp.schedule 1.00 ± 15% -0.8 0.20 ± 22% perf-profile.children.cycles-pp.asm_sysvec_reschedule_ipi 6.14 ± 6% -0.8 5.37 ± 4% perf-profile.children.cycles-pp.perf_trace_sched_switch 1.39 ± 10% -0.4 0.96 ± 10% perf-profile.children.cycles-pp._raw_spin_lock 0.89 ± 14% -0.4 0.49 ± 24% perf-profile.children.cycles-pp.native_queued_spin_lock_slowpath 0.69 ± 14% -0.2 0.48 ± 20% perf-profile.children.cycles-pp.__x64_sys_nanosleep 0.68 ± 14% -0.2 0.47 ± 20% perf-profile.children.cycles-pp.hrtimer_nanosleep 0.68 ± 14% -0.2 0.47 ± 20% perf-profile.children.cycles-pp.do_nanosleep 0.38 ± 17% -0.2 0.22 ± 24% perf-profile.children.cycles-pp.syscall_exit_to_user_mode 0.15 ± 19% -0.1 0.08 ± 48% perf-profile.children.cycles-pp.write 0.21 ± 19% +0.1 0.30 ± 12% perf-profile.children.cycles-pp.kvm_is_in_guest 0.84 ± 7% +0.2 1.07 ± 5% perf-profile.children.cycles-pp.rebalance_domains 0.45 ± 9% +0.3 0.70 ± 15% perf-profile.children.cycles-pp.__calc_delta 0.74 ± 11% +0.3 1.08 ± 6% perf-profile.children.cycles-pp.newidle_balance 1.49 ± 9% +0.6 2.07 ± 4% perf-profile.children.cycles-pp.load_balance 0.89 ± 15% -0.4 0.49 ± 23% perf-profile.self.cycles-pp.native_queued_spin_lock_slowpath 0.38 ± 11% -0.1 0.30 ± 9% perf-profile.self.cycles-pp.__schedule 0.19 ± 19% +0.1 0.28 ± 12% perf-profile.self.cycles-pp.kvm_is_in_guest 0.44 ± 8% +0.2 0.69 ± 15% perf-profile.self.cycles-pp.__calc_delta *************************************************************************************************** lkp-skl-fpga01: 104 threads 2 sockets Skylake with 192G memory ========================================================================================= compiler/cpufreq_governor/ipc/iterations/kconfig/mode/nr_threads/rootfs/tbox_group/testcase/ucode: gcc-9/performance/pipe/4/x86_64-rhel-8.3/threads/100%/debian-10.4-x86_64-20200603.cgz/lkp-skl-fpga01/hackbench/0x2006a0a commit: v5.16-rc6 b08e888d97 ("sched/fair: Force progress on min_vruntime") v5.16-rc6 b08e888d9715f01546da7a40ca3 ---------------- --------------------------- %stddev %change %stddev \ | \ 243924 +55.5% 379330 hackbench.throughput 15540713 ± 8% -60.4% 6160950 ± 3% cpuidle..usage 261.57 -29.5% 184.44 uptime.boot 1.18 -0.1 1.05 mpstat.cpu.all.irq% 0.04 ± 5% -0.0 0.04 ± 2% mpstat.cpu.all.soft% 73.85 -11.8 62.02 mpstat.cpu.all.sys% 22.89 +11.7 34.62 mpstat.cpu.all.usr% 20206732 +171.9% 54945573 numa-numastat.node0.local_node 20253752 +171.5% 54982565 numa-numastat.node0.numa_hit 21045850 ± 2% +183.3% 59621383 numa-numastat.node1.local_node 21058134 +183.2% 59642825 numa-numastat.node1.numa_hit 7043843 ± 12% -37.3% 4419834 ± 4% turbostat.C1 0.24 ± 12% +0.1 0.30 ± 4% turbostat.C1% 2.04e+08 ± 2% -84.8% 30989257 turbostat.IRQ 115.28 +14.7% 132.18 turbostat.RAMWatt 22.17 +53.4% 34.00 vmstat.cpu.us 5220877 ± 14% -41.6% 3050905 ± 2% vmstat.memory.cache 1204 +77.7% 2139 vmstat.procs.r 14664745 -95.2% 704900 ± 2% vmstat.system.cs 911202 -74.5% 232488 vmstat.system.in 2250 ± 15% -27.9% 1621 ± 10% numa-meminfo.node0.Active 2095 ± 16% -31.0% 1445 ± 11% numa-meminfo.node0.Active(anon) 188542 ± 50% -58.5% 78333 ± 13% numa-meminfo.node0.Mapped 67992 ± 7% -72.8% 18495 ± 50% numa-meminfo.node1.Active 67992 ± 7% -72.8% 18495 ± 50% numa-meminfo.node1.Active(anon) 72982 ± 46% -58.8% 30071 ± 88% numa-meminfo.node1.AnonHugePages 130528 ± 30% -47.3% 68777 ± 54% numa-meminfo.node1.AnonPages 2539845 ± 29% -81.1% 479406 ± 14% numa-meminfo.node1.Inactive 2539845 ± 29% -81.1% 479406 ± 14% numa-meminfo.node1.Inactive(anon) 502608 ± 21% -79.8% 101561 ± 18% numa-meminfo.node1.Mapped 2492 ± 10% -42.0% 1446 ± 12% numa-meminfo.node1.PageTables 2477433 ± 30% -82.7% 429019 ± 23% numa-meminfo.node1.Shmem 70018 ± 6% -72.8% 19048 ± 42% meminfo.Active 69863 ± 6% -73.0% 18872 ± 42% meminfo.Active(anon) 192446 -26.1% 142292 ± 2% meminfo.AnonHugePages 341625 -10.9% 304256 meminfo.AnonPages 5121644 ± 15% -42.7% 2935312 ± 3% meminfo.Cached 3116939 ± 25% -71.1% 899660 ± 10% meminfo.Committed_AS 2901201 ± 27% -74.9% 728894 ± 12% meminfo.Inactive 2901053 ± 27% -74.9% 728746 ± 12% meminfo.Inactive(anon) 687779 ± 14% -73.8% 180190 ± 11% meminfo.Mapped 7146592 ± 11% -30.2% 4990358 meminfo.Memused 5611 ± 3% -16.2% 4703 meminfo.PageTables 2629586 ± 29% -83.2% 442853 ± 20% meminfo.Shmem 7270248 ± 10% -30.2% 5074455 meminfo.max_used_kB 522.33 ± 16% -30.7% 362.17 ± 11% numa-vmstat.node0.nr_active_anon 46914 ± 50% -56.3% 20487 ± 15% numa-vmstat.node0.nr_mapped 522.33 ± 16% -30.7% 362.17 ± 11% numa-vmstat.node0.nr_zone_active_anon 12264878 +142.5% 29744174 ± 2% numa-vmstat.node0.numa_hit 12214256 +143.1% 29692277 ± 2% numa-vmstat.node0.numa_local 16894 ± 6% -72.1% 4714 ± 49% numa-vmstat.node1.nr_active_anon 32556 ± 30% -47.0% 17259 ± 55% numa-vmstat.node1.nr_anon_pages 631923 ± 30% -80.1% 125630 ± 15% numa-vmstat.node1.nr_inactive_anon 124390 ± 22% -78.5% 26774 ± 17% numa-vmstat.node1.nr_mapped 620.00 ± 10% -41.0% 366.00 ± 11% numa-vmstat.node1.nr_page_table_pages 616293 ± 31% -81.7% 113060 ± 21% numa-vmstat.node1.nr_shmem 16894 ± 6% -72.1% 4714 ± 49% numa-vmstat.node1.nr_zone_active_anon 631922 ± 30% -80.1% 125629 ± 15% numa-vmstat.node1.nr_zone_inactive_anon 11597593 ± 2% +170.4% 31363549 ± 2% numa-vmstat.node1.numa_hit 11533651 ± 2% +171.4% 31297047 ± 2% numa-vmstat.node1.numa_local 17516 ± 6% -73.4% 4662 ± 40% proc-vmstat.nr_active_anon 85408 -11.0% 76023 proc-vmstat.nr_anon_pages 4710721 +1.1% 4764703 proc-vmstat.nr_dirty_background_threshold 9432961 +1.1% 9541057 proc-vmstat.nr_dirty_threshold 1280589 ± 15% -42.8% 732711 ± 2% proc-vmstat.nr_file_pages 47394242 +1.1% 47934851 proc-vmstat.nr_free_pages 725395 ± 26% -75.0% 181084 ± 11% proc-vmstat.nr_inactive_anon 80621 -4.0% 77417 proc-vmstat.nr_kernel_stack 171686 ± 13% -73.8% 44975 ± 11% proc-vmstat.nr_mapped 1402 ± 3% -16.3% 1173 proc-vmstat.nr_page_table_pages 657573 ± 29% -83.3% 109596 ± 19% proc-vmstat.nr_shmem 28926 -5.7% 27281 proc-vmstat.nr_slab_reclaimable 67226 -1.6% 66140 proc-vmstat.nr_slab_unreclaimable 17516 ± 6% -73.4% 4662 ± 40% proc-vmstat.nr_zone_active_anon 725395 ± 26% -75.0% 181084 ± 11% proc-vmstat.nr_zone_inactive_anon 390080 ± 4% -62.6% 145788 ± 20% proc-vmstat.numa_hint_faults 284764 ± 6% -71.6% 80809 ± 21% proc-vmstat.numa_hint_faults_local 41317354 +177.4% 1.146e+08 proc-vmstat.numa_hit 41258052 +177.7% 1.146e+08 proc-vmstat.numa_local 36260 ± 42% -80.8% 6978 ± 47% proc-vmstat.numa_pages_migrated 627669 -50.8% 308691 ± 10% proc-vmstat.numa_pte_updates 41321465 +177.4% 1.146e+08 proc-vmstat.pgalloc_normal 1221322 ± 2% -38.8% 747607 ± 4% proc-vmstat.pgfault 39928898 +185.4% 1.14e+08 proc-vmstat.pgfree 36260 ± 42% -80.8% 6978 ± 47% proc-vmstat.pgmigrate_success 37167 ± 2% -36.7% 23533 ± 2% proc-vmstat.pgreuse 10.86 ± 2% -60.4% 4.30 perf-stat.i.MPKI 2.773e+10 -24.3% 2.098e+10 perf-stat.i.branch-instructions 1.46 -0.2 1.25 perf-stat.i.branch-miss-rate% 3.945e+08 -33.9% 2.609e+08 perf-stat.i.branch-misses 4.19 ± 5% +28.0 32.20 perf-stat.i.cache-miss-rate% 53454898 ± 5% +150.0% 1.336e+08 perf-stat.i.cache-misses 1.412e+09 -68.8% 4.407e+08 perf-stat.i.cache-references 14863174 -95.3% 698605 ± 2% perf-stat.i.context-switches 2.08 +26.7% 2.64 perf-stat.i.cpi 918790 -93.6% 58630 ± 3% perf-stat.i.cpu-migrations 5445 ± 5% -61.5% 2098 perf-stat.i.cycles-between-cache-misses 0.12 +0.1 0.24 perf-stat.i.dTLB-load-miss-rate% 48798801 +56.3% 76261572 perf-stat.i.dTLB-load-misses 3.971e+10 -19.4% 3.201e+10 perf-stat.i.dTLB-loads 0.00 ± 10% -0.0 0.00 ± 3% perf-stat.i.dTLB-store-miss-rate% 303926 ± 3% -46.0% 164007 ± 2% perf-stat.i.dTLB-store-misses 2.368e+10 -21.7% 1.855e+10 perf-stat.i.dTLB-stores 64515637 ± 5% +35.0% 87084500 ± 7% perf-stat.i.iTLB-load-misses 1.352e+11 -22.9% 1.043e+11 perf-stat.i.instructions 2402 ± 12% -48.6% 1235 ± 5% perf-stat.i.instructions-per-iTLB-miss 0.49 -22.8% 0.38 perf-stat.i.ipc 3.42 ± 32% -53.5% 1.59 perf-stat.i.major-faults 361.63 ± 3% +14.5% 413.95 ± 4% perf-stat.i.metric.K/sec 889.15 -22.1% 692.26 perf-stat.i.metric.M/sec 5580 ± 2% -9.5% 5049 ± 5% perf-stat.i.minor-faults 80.85 +2.0 82.88 perf-stat.i.node-load-miss-rate% 9339361 ± 4% +92.4% 17965649 ± 4% perf-stat.i.node-load-misses 2176060 ± 3% +71.5% 3731319 ± 2% perf-stat.i.node-loads 57.54 ± 2% +17.6 75.11 perf-stat.i.node-store-miss-rate% 3921170 ± 4% +231.4% 12995856 ± 3% perf-stat.i.node-store-misses 2780324 +55.0% 4310621 ± 2% perf-stat.i.node-stores 5583 ± 2% -9.5% 5051 ± 5% perf-stat.i.page-faults 10.40 -59.6% 4.20 perf-stat.overall.MPKI 1.42 -0.2 1.24 perf-stat.overall.branch-miss-rate% 3.80 ± 5% +26.8 30.57 perf-stat.overall.cache-miss-rate% 2.03 +30.1% 2.64 perf-stat.overall.cpi 5143 ± 5% -60.0% 2055 perf-stat.overall.cycles-between-cache-misses 0.12 +0.1 0.24 perf-stat.overall.dTLB-load-miss-rate% 0.00 ± 3% -0.0 0.00 ± 2% perf-stat.overall.dTLB-store-miss-rate% 95.23 +1.6 96.82 perf-stat.overall.iTLB-load-miss-rate% 2096 ± 5% -42.6% 1202 ± 7% perf-stat.overall.instructions-per-iTLB-miss 0.49 -23.2% 0.38 perf-stat.overall.ipc 81.04 +1.7 82.71 perf-stat.overall.node-load-miss-rate% 58.43 ± 2% +16.7 75.14 perf-stat.overall.node-store-miss-rate% 2.757e+10 -24.4% 2.085e+10 perf-stat.ps.branch-instructions 3.923e+08 -34.0% 2.59e+08 perf-stat.ps.branch-misses 53174568 ± 5% +150.3% 1.331e+08 perf-stat.ps.cache-misses 1.399e+09 -68.9% 4.354e+08 perf-stat.ps.cache-references 14787641 -95.4% 683155 ± 2% perf-stat.ps.context-switches 901548 -93.8% 55778 ± 3% perf-stat.ps.cpu-migrations 48660610 +55.8% 75815767 perf-stat.ps.dTLB-load-misses 3.951e+10 -19.5% 3.181e+10 perf-stat.ps.dTLB-loads 299532 ± 3% -46.0% 161819 ± 2% perf-stat.ps.dTLB-store-misses 2.356e+10 -21.8% 1.843e+10 perf-stat.ps.dTLB-stores 64330425 ± 5% +34.7% 86669158 ± 7% perf-stat.ps.iTLB-load-misses 1.345e+11 -22.9% 1.036e+11 perf-stat.ps.instructions 3.34 ± 32% -52.4% 1.59 perf-stat.ps.major-faults 5385 -8.5% 4929 ± 5% perf-stat.ps.minor-faults 9247666 ± 4% +93.1% 17857896 ± 4% perf-stat.ps.node-load-misses 2160605 ± 3% +72.6% 3728183 ± 2% perf-stat.ps.node-loads 3886969 ± 4% +232.3% 12914952 ± 3% perf-stat.ps.node-store-misses 2761577 +54.7% 4270821 ± 2% perf-stat.ps.node-stores 5388 -8.5% 4931 ± 5% perf-stat.ps.page-faults 2.83e+13 -51.6% 1.371e+13 perf-stat.total.instructions 35304 ± 9% -35.1% 22902 ± 4% softirqs.CPU0.RCU 34522 ± 10% -34.0% 22784 ± 9% softirqs.CPU1.RCU 7675 ± 19% -34.3% 5038 ± 19% softirqs.CPU1.SCHED 34716 ± 9% -36.7% 21979 ± 6% softirqs.CPU10.RCU 33713 ± 9% -37.0% 21228 ± 4% softirqs.CPU100.RCU 33959 ± 9% -37.6% 21176 ± 5% softirqs.CPU101.RCU 34338 ± 9% -38.6% 21077 ± 4% softirqs.CPU102.RCU 34548 ± 9% -38.1% 21387 ± 5% softirqs.CPU103.RCU 8452 ± 17% -44.1% 4728 ± 18% softirqs.CPU103.SCHED 34720 ± 8% -37.5% 21684 ± 4% softirqs.CPU11.RCU 35190 ± 9% -38.2% 21733 ± 3% softirqs.CPU12.RCU 35034 ± 7% -36.4% 22294 ± 4% softirqs.CPU13.RCU 34598 ± 9% -36.6% 21921 ± 3% softirqs.CPU14.RCU 37841 ± 8% -38.5% 23284 ± 5% softirqs.CPU15.RCU 37752 ± 8% -38.4% 23246 ± 4% softirqs.CPU16.RCU 37597 ± 8% -37.7% 23428 ± 2% softirqs.CPU17.RCU 37341 ± 8% -36.5% 23696 ± 4% softirqs.CPU18.RCU 36883 ± 8% -37.1% 23210 ± 4% softirqs.CPU19.RCU 34873 ± 8% -38.8% 21355 ± 6% softirqs.CPU2.RCU 37958 ± 9% -38.7% 23259 ± 4% softirqs.CPU20.RCU 38116 ± 8% -39.2% 23170 ± 4% softirqs.CPU21.RCU 37496 ± 8% -38.2% 23187 ± 4% softirqs.CPU22.RCU 37683 ± 7% -38.0% 23378 ± 4% softirqs.CPU23.RCU 37514 ± 7% -37.1% 23592 ± 4% softirqs.CPU24.RCU 37848 ± 8% -37.3% 23748 ± 6% softirqs.CPU25.RCU 33410 ± 9% -40.1% 20017 ± 3% softirqs.CPU26.RCU 33936 ± 9% -39.0% 20699 ± 6% softirqs.CPU27.RCU 10205 ± 11% -36.7% 6462 ± 28% softirqs.CPU27.SCHED 33054 ± 9% -40.4% 19714 ± 4% softirqs.CPU28.RCU 33007 ± 10% -39.9% 19836 ± 4% softirqs.CPU29.RCU 35600 ± 11% -40.0% 21365 ± 4% softirqs.CPU3.RCU 35544 ± 10% -36.8% 22467 ± 5% softirqs.CPU30.RCU 34897 ± 8% -36.9% 22011 ± 6% softirqs.CPU31.RCU 34213 ± 6% -36.5% 21719 ± 4% softirqs.CPU32.RCU 33298 ± 5% -36.0% 21323 ± 5% softirqs.CPU33.RCU 7987 ± 17% -42.5% 4595 ± 8% softirqs.CPU33.SCHED 34005 ± 9% -37.8% 21150 ± 4% softirqs.CPU34.RCU 34250 ± 9% -38.2% 21163 ± 4% softirqs.CPU35.RCU 34218 ± 9% -37.9% 21256 ± 4% softirqs.CPU36.RCU 34474 ± 9% -36.2% 22010 ± 5% softirqs.CPU37.RCU 34626 ± 9% -38.0% 21485 ± 5% softirqs.CPU38.RCU 34276 ± 9% -38.1% 21203 ± 5% softirqs.CPU39.RCU 34472 ± 9% -36.9% 21742 ± 4% softirqs.CPU4.RCU 33896 ± 9% -38.1% 20986 ± 5% softirqs.CPU40.RCU 34236 ± 9% -38.2% 21159 ± 4% softirqs.CPU41.RCU 34549 ± 9% -37.9% 21445 ± 5% softirqs.CPU42.RCU 34676 ± 9% -37.7% 21617 ± 6% softirqs.CPU43.RCU 34667 ± 9% -38.1% 21469 ± 4% softirqs.CPU44.RCU 34607 ± 9% -40.1% 20718 ± 4% softirqs.CPU45.RCU 34814 ± 9% -38.9% 21266 ± 6% softirqs.CPU46.RCU 34743 ± 10% -39.6% 20991 ± 4% softirqs.CPU47.RCU 34411 ± 10% -39.1% 20945 ± 5% softirqs.CPU48.RCU 34696 ± 10% -40.0% 20819 ± 4% softirqs.CPU49.RCU 35847 ± 8% -38.3% 22125 ± 4% softirqs.CPU5.RCU 34536 ± 9% -38.7% 21160 ± 6% softirqs.CPU50.RCU 34457 ± 9% -38.6% 21150 ± 4% softirqs.CPU51.RCU 36863 ± 8% -40.1% 22070 ± 3% softirqs.CPU52.RCU 35183 ± 14% -34.2% 23157 ± 5% softirqs.CPU53.RCU 37456 ± 8% -35.6% 24126 ± 8% softirqs.CPU54.RCU 36481 ± 12% -37.5% 22814 ± 5% softirqs.CPU55.RCU 37552 ± 8% -39.1% 22880 ± 5% softirqs.CPU56.RCU 38204 ± 12% -39.5% 23108 ± 4% softirqs.CPU57.RCU 37570 ± 8% -40.6% 22333 ± 5% softirqs.CPU58.RCU 37972 ± 7% -39.2% 23087 ± 5% softirqs.CPU59.RCU 35010 ± 9% -38.0% 21717 ± 5% softirqs.CPU6.RCU 34777 ± 9% -37.3% 21821 ± 4% softirqs.CPU60.RCU 34756 ± 8% -35.2% 22536 ± 6% softirqs.CPU61.RCU 35005 ± 7% -37.6% 21829 ± 5% softirqs.CPU62.RCU 35414 ± 9% -37.9% 21979 ± 4% softirqs.CPU63.RCU 35050 ± 8% -37.2% 22015 ± 4% softirqs.CPU64.RCU 35234 ± 9% -37.6% 21997 ± 4% softirqs.CPU65.RCU 34721 ± 9% -37.5% 21699 ± 3% softirqs.CPU66.RCU 35608 ± 7% -38.9% 21772 ± 5% softirqs.CPU67.RCU 34949 ± 8% -39.0% 21317 ± 4% softirqs.CPU68.RCU 35032 ± 9% -37.8% 21778 ± 4% softirqs.CPU69.RCU 34514 ± 9% -36.3% 21983 ± 9% softirqs.CPU7.RCU 35351 ± 9% -38.1% 21883 ± 2% softirqs.CPU70.RCU 33661 ± 9% -35.2% 21816 ± 4% softirqs.CPU71.RCU 34803 ± 8% -39.5% 21054 ± 7% softirqs.CPU72.RCU 34299 ± 8% -37.4% 21463 ± 2% softirqs.CPU73.RCU 34667 ± 9% -38.3% 21385 ± 3% softirqs.CPU74.RCU 32951 ± 8% -38.2% 20359 ± 7% softirqs.CPU75.RCU 33407 ± 8% -39.0% 20370 ± 4% softirqs.CPU76.RCU 33587 ± 7% -40.7% 19914 ± 4% softirqs.CPU77.RCU 34722 ± 8% -39.0% 21194 ± 7% softirqs.CPU78.RCU 33933 ± 10% -37.8% 21112 ± 4% softirqs.CPU79.RCU 34812 ± 7% -38.0% 21592 ± 5% softirqs.CPU8.RCU 35145 ± 9% -39.1% 21396 ± 4% softirqs.CPU80.RCU 34879 ± 8% -38.8% 21332 ± 5% softirqs.CPU81.RCU 34627 ± 8% -39.3% 21007 ± 4% softirqs.CPU82.RCU 34760 ± 8% -39.0% 21215 ± 4% softirqs.CPU83.RCU 34952 ± 8% -39.0% 21312 ± 4% softirqs.CPU84.RCU 34522 ± 8% -39.1% 21024 ± 5% softirqs.CPU85.RCU 34406 ± 9% -38.6% 21137 ± 5% softirqs.CPU86.RCU 34419 ± 8% -39.0% 20987 ± 4% softirqs.CPU87.RCU 34403 ± 8% -37.9% 21358 ± 3% softirqs.CPU88.RCU 35015 ± 9% -38.0% 21701 ± 4% softirqs.CPU89.RCU 35199 ± 9% -38.3% 21704 ± 6% softirqs.CPU9.RCU 34512 ± 9% -37.9% 21441 ± 4% softirqs.CPU90.RCU 34052 ± 9% -38.0% 21098 ± 4% softirqs.CPU91.RCU 34048 ± 9% -37.6% 21252 ± 4% softirqs.CPU92.RCU 34155 ± 8% -37.3% 21398 ± 5% softirqs.CPU93.RCU 34243 ± 8% -37.7% 21347 ± 4% softirqs.CPU94.RCU 34387 ± 9% -37.4% 21536 ± 5% softirqs.CPU95.RCU 34242 ± 8% -37.3% 21472 ± 4% softirqs.CPU96.RCU 34569 ± 8% -38.6% 21224 ± 4% softirqs.CPU97.RCU 34086 ± 9% -37.5% 21317 ± 4% softirqs.CPU98.RCU 34399 ± 9% -37.8% 21380 ± 5% softirqs.CPU99.RCU 3645591 ± 8% -38.0% 2259598 ± 4% softirqs.RCU 713024 ± 7% -37.9% 442950 ± 11% softirqs.SCHED 45382 ± 4% -22.4% 35207 ± 3% softirqs.TIMER 21.02 ± 10% -20.0 1.07 ± 50% perf-profile.calltrace.cycles-pp.__wake_up_common_lock.pipe_write.new_sync_write.vfs_write.ksys_write 20.54 ± 10% -19.5 1.05 ± 50% perf-profile.calltrace.cycles-pp.__wake_up_common.__wake_up_common_lock.pipe_write.new_sync_write.vfs_write 20.34 ± 10% -19.3 1.04 ± 51% perf-profile.calltrace.cycles-pp.autoremove_wake_function.__wake_up_common.__wake_up_common_lock.pipe_write.new_sync_write 20.02 ± 11% -19.0 1.02 ± 52% perf-profile.calltrace.cycles-pp.try_to_wake_up.autoremove_wake_function.__wake_up_common.__wake_up_common_lock.pipe_write 29.44 ± 5% -14.3 15.18 ± 3% perf-profile.calltrace.cycles-pp.pipe_write.new_sync_write.vfs_write.ksys_write.do_syscall_64 14.88 ± 15% -14.1 0.78 ± 88% perf-profile.calltrace.cycles-pp.schedule.pipe_read.new_sync_read.vfs_read.ksys_read 14.70 ± 15% -14.0 0.70 ±107% perf-profile.calltrace.cycles-pp.__schedule.schedule.pipe_read.new_sync_read.vfs_read 30.12 ± 5% -13.9 16.20 ± 3% perf-profile.calltrace.cycles-pp.new_sync_write.vfs_write.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe 31.81 ± 4% -12.3 19.49 ± 2% perf-profile.calltrace.cycles-pp.vfs_write.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_write 11.62 ± 55% -11.2 0.40 ±223% perf-profile.calltrace.cycles-pp.perf_prepare_sample.perf_event_output_forward.__perf_event_overflow.perf_swevent_overflow.perf_tp_event 33.67 ± 4% -10.9 22.79 perf-profile.calltrace.cycles-pp.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_write 10.54 ± 54% -10.3 0.29 ±223% perf-profile.calltrace.cycles-pp.perf_callchain.perf_prepare_sample.perf_event_output_forward.__perf_event_overflow.perf_swevent_overflow 44.47 ± 2% -10.1 34.37 perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_write 10.32 ± 54% -10.0 0.29 ±223% perf-profile.calltrace.cycles-pp.get_perf_callchain.perf_callchain.perf_prepare_sample.perf_event_output_forward.__perf_event_overflow 44.84 ± 2% -10.0 34.89 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.__libc_write 9.40 ± 53% -9.2 0.19 ±223% perf-profile.calltrace.cycles-pp.perf_callchain_kernel.get_perf_callchain.perf_callchain.perf_prepare_sample.perf_event_output_forward 8.80 ± 14% -8.5 0.35 ±107% perf-profile.calltrace.cycles-pp.ttwu_do_activate.try_to_wake_up.autoremove_wake_function.__wake_up_common.__wake_up_common_lock 8.71 ± 15% -8.4 0.27 ±146% perf-profile.calltrace.cycles-pp.enqueue_task_fair.ttwu_do_activate.try_to_wake_up.autoremove_wake_function.__wake_up_common 24.75 ± 6% -7.1 17.70 perf-profile.calltrace.cycles-pp.pipe_read.new_sync_read.vfs_read.ksys_read.do_syscall_64 7.28 ± 14% -7.0 0.26 ±146% perf-profile.calltrace.cycles-pp.dequeue_task_fair.__schedule.schedule.pipe_read.new_sync_read 25.37 ± 6% -6.9 18.47 perf-profile.calltrace.cycles-pp.new_sync_read.vfs_read.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe 5.92 ± 54% -5.7 0.24 ±223% perf-profile.calltrace.cycles-pp.__perf_event_overflow.perf_swevent_overflow.perf_tp_event.perf_trace_sched_stat_runtime.update_curr 5.88 ± 54% -5.7 0.23 ±223% perf-profile.calltrace.cycles-pp.perf_event_output_forward.__perf_event_overflow.perf_swevent_overflow.perf_tp_event.perf_trace_sched_stat_runtime 5.85 ± 28% -5.6 0.23 ±148% perf-profile.calltrace.cycles-pp.enqueue_entity.enqueue_task_fair.ttwu_do_activate.try_to_wake_up.autoremove_wake_function 27.30 ± 5% -5.3 22.01 perf-profile.calltrace.cycles-pp.vfs_read.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_read 52.39 -4.7 47.68 perf-profile.calltrace.cycles-pp.__libc_write 28.82 ± 4% -4.2 24.66 perf-profile.calltrace.cycles-pp.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_read 0.71 ± 9% +0.3 0.96 perf-profile.calltrace.cycles-pp.__pthread_disable_asynccancel 0.68 ± 9% +0.3 0.96 ± 2% perf-profile.calltrace.cycles-pp.file_update_time.pipe_write.new_sync_write.vfs_write.ksys_write 10.15 ± 2% +0.4 10.53 perf-profile.calltrace.cycles-pp.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_write 0.66 ± 11% +0.4 1.10 ± 2% perf-profile.calltrace.cycles-pp.mutex_unlock.pipe_write.new_sync_write.vfs_write.ksys_write 0.96 ± 9% +0.4 1.41 perf-profile.calltrace.cycles-pp.mutex_lock.pipe_read.new_sync_read.vfs_read.ksys_read 0.00 +0.5 0.52 ± 2% perf-profile.calltrace.cycles-pp.current_time.atime_needs_update.touch_atime.pipe_read.new_sync_read 0.00 +0.5 0.53 ± 2% perf-profile.calltrace.cycles-pp.__x64_sys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_write 0.00 +0.6 0.55 ± 2% perf-profile.calltrace.cycles-pp.fput_many.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_write 0.00 +0.6 0.59 ± 2% perf-profile.calltrace.cycles-pp.fput_many.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_read 0.82 ± 8% +0.6 1.42 ± 11% perf-profile.calltrace.cycles-pp.__fget_light.__fdget_pos.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe 1.03 ± 9% +0.6 1.65 perf-profile.calltrace.cycles-pp.copy_user_enhanced_fast_string.copyout.copy_page_to_iter.pipe_read.new_sync_read 0.82 ± 7% +0.6 1.47 ± 11% perf-profile.calltrace.cycles-pp.__fdget_pos.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_read 0.39 ± 70% +0.7 1.05 ± 2% perf-profile.calltrace.cycles-pp.__pthread_enable_asynccancel 0.86 ± 9% +0.7 1.56 perf-profile.calltrace.cycles-pp.copy_user_enhanced_fast_string.copyin.copy_page_from_iter.pipe_write.new_sync_write 1.10 ± 7% +0.7 1.80 perf-profile.calltrace.cycles-pp.atime_needs_update.touch_atime.pipe_read.new_sync_read.vfs_read 1.39 ± 6% +0.7 2.10 ± 2% perf-profile.calltrace.cycles-pp.touch_atime.pipe_read.new_sync_read.vfs_read.ksys_read 0.00 +0.7 0.71 ± 4% perf-profile.calltrace.cycles-pp.mutex_unlock.pipe_read.new_sync_read.vfs_read.ksys_read 1.29 ± 8% +0.8 2.04 ± 7% perf-profile.calltrace.cycles-pp.__fget_light.__fdget_pos.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe 0.00 +0.8 0.76 ± 6% perf-profile.calltrace.cycles-pp.propagate_protected_usage.page_counter_charge.obj_cgroup_charge_pages.__memcg_kmem_charge_page.__alloc_pages 1.16 ± 9% +0.8 1.93 perf-profile.calltrace.cycles-pp.copyout.copy_page_to_iter.pipe_read.new_sync_read.vfs_read 0.83 ± 6% +0.8 1.61 ± 7% perf-profile.calltrace.cycles-pp.common_file_perm.security_file_permission.vfs_write.ksys_write.do_syscall_64 1.30 ± 8% +0.8 2.09 ± 7% perf-profile.calltrace.cycles-pp.__fdget_pos.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_write 0.86 ± 7% +0.8 1.70 ± 7% perf-profile.calltrace.cycles-pp.common_file_perm.security_file_permission.vfs_read.ksys_read.do_syscall_64 1.02 ± 9% +0.9 1.90 ± 2% perf-profile.calltrace.cycles-pp.copyin.copy_page_from_iter.pipe_write.new_sync_write.vfs_write 1.66 ± 9% +1.0 2.67 ± 4% perf-profile.calltrace.cycles-pp.mutex_lock.pipe_write.new_sync_write.vfs_write.ksys_write 1.11 ± 8% +1.2 2.36 ± 6% perf-profile.calltrace.cycles-pp.security_file_permission.vfs_write.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe 1.35 ± 8% +1.5 2.82 ± 5% perf-profile.calltrace.cycles-pp.security_file_permission.vfs_read.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe 2.22 ± 9% +1.5 3.72 ± 2% perf-profile.calltrace.cycles-pp.copy_page_from_iter.pipe_write.new_sync_write.vfs_write.ksys_write 2.26 ± 9% +1.7 3.97 perf-profile.calltrace.cycles-pp.copy_page_to_iter.pipe_read.new_sync_read.vfs_read.ksys_read 0.00 +1.9 1.94 ± 4% perf-profile.calltrace.cycles-pp.page_counter_charge.obj_cgroup_charge_pages.__memcg_kmem_charge_page.__alloc_pages.pipe_write 2.78 ± 9% +2.1 4.86 perf-profile.calltrace.cycles-pp.syscall_return_via_sysret.__libc_write 2.86 ± 9% +2.1 4.95 perf-profile.calltrace.cycles-pp.syscall_return_via_sysret.__libc_read 0.00 +2.1 2.12 ± 4% perf-profile.calltrace.cycles-pp.obj_cgroup_charge_pages.__memcg_kmem_charge_page.__alloc_pages.pipe_write.new_sync_write 0.00 +2.2 2.16 ± 4% perf-profile.calltrace.cycles-pp.__memcg_kmem_charge_page.__alloc_pages.pipe_write.new_sync_write.vfs_write 1.26 ± 31% +2.3 3.54 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_safe_stack.__libc_write 0.00 +2.3 2.30 ± 4% perf-profile.calltrace.cycles-pp.__alloc_pages.pipe_write.new_sync_write.vfs_write.ksys_write 2.50 ± 9% +2.4 4.91 perf-profile.calltrace.cycles-pp.__entry_text_start.__libc_read 2.51 ± 8% +2.5 5.03 perf-profile.calltrace.cycles-pp.__entry_text_start.__libc_write 0.00 +2.7 2.70 ± 2% perf-profile.calltrace.cycles-pp.page_counter_cancel.page_counter_uncharge.uncharge_batch.__mem_cgroup_uncharge.__put_page 1.61 ± 11% +2.8 4.41 ± 3% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_safe_stack.__libc_read 0.00 +3.0 3.04 ± 4% perf-profile.calltrace.cycles-pp.propagate_protected_usage.page_counter_uncharge.uncharge_batch.__mem_cgroup_uncharge.__put_page 7.16 ± 7% +3.4 10.51 perf-profile.calltrace.cycles-pp.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_read 45.21 +4.7 49.92 perf-profile.calltrace.cycles-pp.__libc_read 0.00 +6.0 6.05 ± 3% perf-profile.calltrace.cycles-pp.page_counter_uncharge.uncharge_batch.__mem_cgroup_uncharge.__put_page.pipe_read 0.00 +6.2 6.20 ± 3% perf-profile.calltrace.cycles-pp.uncharge_batch.__mem_cgroup_uncharge.__put_page.pipe_read.new_sync_read 0.00 +6.3 6.25 ± 3% perf-profile.calltrace.cycles-pp.__mem_cgroup_uncharge.__put_page.pipe_read.new_sync_read.vfs_read 0.00 +6.3 6.26 ± 3% perf-profile.calltrace.cycles-pp.__put_page.pipe_read.new_sync_read.vfs_read.ksys_read 22.50 ± 10% -21.3 1.16 ± 49% perf-profile.children.cycles-pp.__wake_up_common_lock 21.96 ± 11% -20.8 1.11 ± 50% perf-profile.children.cycles-pp.__wake_up_common 21.76 ± 11% -20.7 1.10 ± 51% perf-profile.children.cycles-pp.autoremove_wake_function 21.51 ± 11% -20.2 1.28 ± 47% perf-profile.children.cycles-pp.try_to_wake_up 20.25 ± 14% -19.1 1.12 ± 54% perf-profile.children.cycles-pp.schedule 20.08 ± 15% -19.0 1.13 ± 55% perf-profile.children.cycles-pp.__schedule 29.50 ± 5% -14.2 15.25 ± 3% perf-profile.children.cycles-pp.pipe_write 30.15 ± 5% -13.9 16.24 ± 3% perf-profile.children.cycles-pp.new_sync_write 14.55 ± 49% -13.5 1.08 ±117% perf-profile.children.cycles-pp.perf_tp_event 13.74 ± 51% -12.7 1.03 ±117% perf-profile.children.cycles-pp.perf_swevent_overflow 13.71 ± 51% -12.7 1.03 ±117% perf-profile.children.cycles-pp.__perf_event_overflow 13.62 ± 51% -12.6 1.02 ±117% perf-profile.children.cycles-pp.perf_event_output_forward 31.84 ± 4% -12.3 19.52 ± 2% perf-profile.children.cycles-pp.vfs_write 13.19 ± 51% -12.3 0.94 ±118% perf-profile.children.cycles-pp.perf_prepare_sample 12.04 ± 51% -11.2 0.86 ±118% perf-profile.children.cycles-pp.perf_callchain 11.96 ± 51% -11.1 0.86 ±118% perf-profile.children.cycles-pp.get_perf_callchain 33.70 ± 4% -10.9 22.84 perf-profile.children.cycles-pp.ksys_write 81.19 ± 2% -10.5 70.66 perf-profile.children.cycles-pp.do_syscall_64 11.08 ± 51% -10.3 0.80 ±118% perf-profile.children.cycles-pp.perf_callchain_kernel 81.93 ± 2% -10.3 71.67 perf-profile.children.cycles-pp.entry_SYSCALL_64_after_hwframe 9.70 ± 33% -8.9 0.81 ± 86% perf-profile.children.cycles-pp.update_curr 8.94 ± 13% -8.4 0.56 ± 53% perf-profile.children.cycles-pp.ttwu_do_activate 8.85 ± 14% -8.3 0.55 ± 54% perf-profile.children.cycles-pp.enqueue_task_fair 8.32 ± 51% -7.7 0.59 ±118% perf-profile.children.cycles-pp.unwind_next_frame 7.88 ± 14% -7.4 0.52 ± 60% perf-profile.children.cycles-pp.dequeue_task_fair 24.85 ± 6% -7.1 17.77 perf-profile.children.cycles-pp.pipe_read 25.41 ± 6% -6.9 18.52 perf-profile.children.cycles-pp.new_sync_read 7.05 ± 49% -6.4 0.60 ±117% perf-profile.children.cycles-pp.perf_trace_sched_stat_runtime 6.18 ± 24% -5.8 0.43 ± 70% perf-profile.children.cycles-pp.enqueue_entity 27.33 ± 5% -5.3 22.05 perf-profile.children.cycles-pp.vfs_read 5.83 ± 5% -5.2 0.62 ± 2% perf-profile.children.cycles-pp.exit_to_user_mode_prepare 52.87 -5.0 47.86 perf-profile.children.cycles-pp.__libc_write 5.14 ± 28% -4.8 0.38 ± 82% perf-profile.children.cycles-pp.dequeue_entity 4.87 ± 50% -4.5 0.34 ±117% perf-profile.children.cycles-pp.__unwind_start 28.87 ± 4% -4.2 24.71 perf-profile.children.cycles-pp.ksys_read 4.38 ± 50% -4.1 0.24 ±118% perf-profile.children.cycles-pp.perf_trace_sched_switch 4.28 ± 13% -4.1 0.14 ± 33% perf-profile.children.cycles-pp._raw_spin_lock 3.81 ± 38% -3.5 0.30 ± 91% perf-profile.children.cycles-pp.perf_trace_sched_wakeup_template 3.63 ± 11% -3.5 0.16 ± 3% perf-profile.children.cycles-pp.select_task_rq_fair 3.45 ± 19% -3.4 0.05 ±111% perf-profile.children.cycles-pp.native_queued_spin_lock_slowpath 3.07 ± 11% -2.9 0.19 ± 3% perf-profile.children.cycles-pp.update_load_avg 2.80 ± 11% -2.6 0.16 ± 5% perf-profile.children.cycles-pp.pick_next_task_fair 2.67 ± 11% -2.5 0.12 ± 3% perf-profile.children.cycles-pp.select_idle_sibling 2.08 ± 51% -1.9 0.15 ±116% perf-profile.children.cycles-pp.__orc_find 1.99 ± 12% -1.8 0.16 ± 13% perf-profile.children.cycles-pp.update_cfs_group 1.80 ± 11% -1.7 0.08 ± 6% perf-profile.children.cycles-pp.select_idle_cpu 1.50 ± 51% -1.4 0.11 ±118% perf-profile.children.cycles-pp.unwind_get_return_address 1.30 ± 11% -1.2 0.10 ± 3% perf-profile.children.cycles-pp.reweight_entity 1.25 ± 51% -1.2 0.09 ±118% perf-profile.children.cycles-pp.orc_find 1.21 ± 51% -1.1 0.09 ±118% perf-profile.children.cycles-pp.__kernel_text_address 1.17 ± 11% -1.1 0.10 ± 6% perf-profile.children.cycles-pp.switch_fpu_return 1.12 ± 12% -1.0 0.08 ± 6% perf-profile.children.cycles-pp.set_next_entity 1.01 ± 52% -0.9 0.07 ±117% perf-profile.children.cycles-pp.kernel_text_address 0.98 ± 12% -0.9 0.09 ± 5% perf-profile.children.cycles-pp.prepare_to_wait_event 0.95 ± 51% -0.9 0.06 ±118% perf-profile.children.cycles-pp.stack_access_ok 0.94 ± 11% -0.9 0.07 ± 8% perf-profile.children.cycles-pp.__switch_to_asm 1.05 ± 11% -0.9 0.18 ± 2% perf-profile.children.cycles-pp._raw_spin_lock_irqsave 0.85 ± 9% -0.8 0.07 perf-profile.children.cycles-pp.__update_load_avg_se 0.84 ± 12% -0.8 0.08 ± 4% perf-profile.children.cycles-pp.restore_fpregs_from_fpstate 0.80 ± 11% -0.7 0.06 ± 7% perf-profile.children.cycles-pp.__switch_to 0.78 ± 11% -0.7 0.04 ± 44% perf-profile.children.cycles-pp.__update_load_avg_cfs_rq 0.76 ± 12% -0.7 0.05 perf-profile.children.cycles-pp.update_rq_clock 0.56 ± 5% -0.5 0.03 ±105% perf-profile.children.cycles-pp.sched_clock_cpu 0.42 ± 12% -0.3 0.09 ± 4% perf-profile.children.cycles-pp.__calc_delta 0.39 ± 10% -0.3 0.07 ± 7% perf-profile.children.cycles-pp.__list_del_entry_valid 0.28 ± 11% -0.2 0.09 perf-profile.children.cycles-pp._raw_spin_lock_irq 0.27 ± 12% -0.2 0.09 ± 4% perf-profile.children.cycles-pp.anon_pipe_buf_release 0.11 ± 4% -0.1 0.06 ± 9% perf-profile.children.cycles-pp.generic_update_time 0.09 ± 4% -0.0 0.04 ± 45% perf-profile.children.cycles-pp.__mark_inode_dirty 0.10 ± 7% +0.0 0.13 ± 5% perf-profile.children.cycles-pp.kill_fasync 0.00 +0.1 0.05 ± 7% perf-profile.children.cycles-pp.uncharge_folio 0.00 +0.1 0.05 ± 8% perf-profile.children.cycles-pp.get_mem_cgroup_from_objcg 0.00 +0.1 0.06 ± 8% perf-profile.children.cycles-pp.__count_memcg_events 0.05 ± 46% +0.1 0.11 ± 5% perf-profile.children.cycles-pp.wake_q_add 0.16 ± 14% +0.1 0.22 ± 6% perf-profile.children.cycles-pp.ktime_get_coarse_real_ts64 0.00 +0.1 0.07 ± 5% perf-profile.children.cycles-pp.rmqueue 0.00 +0.1 0.07 ± 5% perf-profile.children.cycles-pp.free_unref_page 0.17 ± 5% +0.1 0.25 ± 2% perf-profile.children.cycles-pp.syscall_exit_to_user_mode_prepare 0.16 ± 25% +0.1 0.24 ± 3% perf-profile.children.cycles-pp.start_thread 0.10 ± 12% +0.1 0.18 ± 2% perf-profile.children.cycles-pp.iov_iter_init 0.00 +0.1 0.09 ± 4% perf-profile.children.cycles-pp.get_page_from_freelist 0.07 ± 10% +0.1 0.16 ± 4% perf-profile.children.cycles-pp.rw_verify_area 0.00 +0.1 0.10 ± 9% perf-profile.children.cycles-pp.page_counter_try_charge 0.22 ± 10% +0.1 0.33 ± 3% perf-profile.children.cycles-pp.rcu_all_qs 0.54 ± 11% +0.1 0.67 ± 2% perf-profile.children.cycles-pp.__cond_resched 0.33 ± 10% +0.1 0.48 perf-profile.children.cycles-pp.aa_file_perm 0.00 +0.1 0.15 ± 8% perf-profile.children.cycles-pp.try_charge_memcg 0.10 ± 16% +0.2 0.25 ± 17% perf-profile.children.cycles-pp.wake_up_q 0.26 ± 7% +0.2 0.42 perf-profile.children.cycles-pp.timestamp_truncate 0.23 ± 9% +0.2 0.40 ± 5% perf-profile.children.cycles-pp.make_kgid 0.24 ± 13% +0.2 0.42 ± 3% perf-profile.children.cycles-pp.syscall_enter_from_user_mode 0.23 ± 9% +0.2 0.42 ± 2% perf-profile.children.cycles-pp.make_kuid 0.28 ± 8% +0.2 0.49 ± 6% perf-profile.children.cycles-pp.__x64_sys_read 0.30 ± 6% +0.3 0.55 ± 3% perf-profile.children.cycles-pp.__x64_sys_write 0.71 ± 10% +0.3 0.97 perf-profile.children.cycles-pp.__pthread_disable_asynccancel 0.00 +0.3 0.27 ± 9% perf-profile.children.cycles-pp.mutex_spin_on_owner 0.70 ± 9% +0.3 0.98 ± 2% perf-profile.children.cycles-pp.file_update_time 0.78 ± 10% +0.3 1.06 perf-profile.children.cycles-pp.__might_sleep 0.26 ± 20% +0.3 0.55 ± 6% perf-profile.children.cycles-pp.__mutex_unlock_slowpath 0.40 ± 9% +0.3 0.73 ± 3% perf-profile.children.cycles-pp.map_id_range_down 0.50 ± 9% +0.4 0.89 ± 2% perf-profile.children.cycles-pp.__might_fault 0.68 ± 9% +0.4 1.09 ± 2% perf-profile.children.cycles-pp.current_time 0.16 ± 18% +0.5 0.65 ± 8% perf-profile.children.cycles-pp.__mutex_lock 0.55 ± 9% +0.5 1.05 ± 2% perf-profile.children.cycles-pp.__pthread_enable_asynccancel 0.45 ± 11% +0.6 1.00 ± 3% perf-profile.children.cycles-pp.apparmor_file_permission 0.56 ± 9% +0.6 1.14 ± 2% perf-profile.children.cycles-pp.fput_many 1.26 ± 8% +0.6 1.91 perf-profile.children.cycles-pp.__might_resched 1.13 ± 7% +0.7 1.84 perf-profile.children.cycles-pp.atime_needs_update 1.40 ± 6% +0.7 2.12 ± 2% perf-profile.children.cycles-pp.touch_atime 1.17 ± 9% +0.8 1.95 perf-profile.children.cycles-pp.copyout 1.08 ± 10% +0.8 1.85 perf-profile.children.cycles-pp.mutex_unlock 1.05 ± 9% +0.9 1.94 perf-profile.children.cycles-pp.copyin 2.12 ± 8% +1.4 3.49 ± 8% perf-profile.children.cycles-pp.__fget_light 2.20 ± 8% +1.4 3.64 ± 8% perf-profile.children.cycles-pp.__fdget_pos 2.69 ± 9% +1.5 4.18 ± 3% perf-profile.children.cycles-pp.mutex_lock 2.09 ± 9% +1.5 3.60 perf-profile.children.cycles-pp.copy_user_enhanced_fast_string 2.27 ± 9% +1.5 3.79 ± 2% perf-profile.children.cycles-pp.copy_page_from_iter 1.72 ± 6% +1.6 3.35 ± 7% perf-profile.children.cycles-pp.common_file_perm 2.30 ± 9% +1.7 4.03 perf-profile.children.cycles-pp.copy_page_to_iter 0.04 ± 71% +1.9 1.94 ± 4% perf-profile.children.cycles-pp.page_counter_charge 0.07 ± 18% +2.1 2.13 ± 4% perf-profile.children.cycles-pp.obj_cgroup_charge_pages 0.07 ± 15% +2.1 2.16 ± 4% perf-profile.children.cycles-pp.__memcg_kmem_charge_page 0.11 ± 16% +2.2 2.30 ± 4% perf-profile.children.cycles-pp.__alloc_pages 0.15 ± 18% +2.6 2.71 ± 2% perf-profile.children.cycles-pp.page_counter_cancel 2.48 ± 8% +2.7 5.21 ± 5% perf-profile.children.cycles-pp.security_file_permission 1.82 ± 11% +2.9 4.75 perf-profile.children.cycles-pp.entry_SYSCALL_64_safe_stack 17.36 ± 4% +3.8 21.12 perf-profile.children.cycles-pp.syscall_exit_to_user_mode 0.08 ± 16% +3.8 3.88 ± 4% perf-profile.children.cycles-pp.propagate_protected_usage 45.67 +4.4 50.06 perf-profile.children.cycles-pp.__libc_read 6.37 ± 9% +4.7 11.04 perf-profile.children.cycles-pp.syscall_return_via_sysret 6.22 ± 9% +5.8 11.98 perf-profile.children.cycles-pp.__entry_text_start 0.22 ± 16% +5.8 6.05 ± 3% perf-profile.children.cycles-pp.page_counter_uncharge 0.23 ± 16% +6.0 6.20 ± 3% perf-profile.children.cycles-pp.uncharge_batch 0.24 ± 15% +6.0 6.25 ± 3% perf-profile.children.cycles-pp.__mem_cgroup_uncharge 0.24 ± 16% +6.0 6.26 ± 3% perf-profile.children.cycles-pp.__put_page 3.97 ± 50% -3.7 0.28 ±117% perf-profile.self.cycles-pp.unwind_next_frame 3.44 ± 20% -3.4 0.05 ±111% perf-profile.self.cycles-pp.native_queued_spin_lock_slowpath 2.08 ± 51% -1.9 0.15 ±118% perf-profile.self.cycles-pp.__orc_find 1.97 ± 12% -1.8 0.16 ± 14% perf-profile.self.cycles-pp.update_cfs_group 1.51 ± 12% -1.4 0.09 ± 4% perf-profile.self.cycles-pp.__schedule 1.40 ± 13% -1.3 0.07 ± 5% perf-profile.self.cycles-pp.update_load_avg 1.19 ± 8% -1.1 0.06 ± 8% perf-profile.self.cycles-pp.update_curr 1.18 ± 51% -1.1 0.08 ±120% perf-profile.self.cycles-pp.orc_find 0.94 ± 11% -0.9 0.07 ± 8% perf-profile.self.cycles-pp.__switch_to_asm 1.01 ± 10% -0.8 0.16 ± 2% perf-profile.self.cycles-pp._raw_spin_lock_irqsave 0.86 ± 12% -0.8 0.09 ± 5% perf-profile.self.cycles-pp._raw_spin_lock 0.83 ± 9% -0.8 0.07 ± 7% perf-profile.self.cycles-pp.__update_load_avg_se 0.84 ± 12% -0.8 0.08 ± 4% perf-profile.self.cycles-pp.restore_fpregs_from_fpstate 0.80 ± 50% -0.8 0.06 ±119% perf-profile.self.cycles-pp.perf_callchain_kernel 0.80 ± 50% -0.7 0.06 ±119% perf-profile.self.cycles-pp.stack_access_ok 0.74 ± 10% -0.7 0.03 ± 70% perf-profile.self.cycles-pp.__update_load_avg_cfs_rq 0.76 ± 11% -0.7 0.06 perf-profile.self.cycles-pp.__switch_to 0.62 ± 9% -0.6 0.06 ± 6% perf-profile.self.cycles-pp.try_to_wake_up 0.42 ± 11% -0.3 0.08 ± 5% perf-profile.self.cycles-pp.__calc_delta 0.34 ± 10% -0.3 0.06 ± 7% perf-profile.self.cycles-pp.__list_del_entry_valid 0.27 ± 12% -0.2 0.09 ± 4% perf-profile.self.cycles-pp._raw_spin_lock_irq 0.26 ± 12% -0.2 0.08 ± 5% perf-profile.self.cycles-pp.anon_pipe_buf_release 0.09 ± 4% -0.0 0.04 ± 45% perf-profile.self.cycles-pp.__mark_inode_dirty 0.18 ± 5% +0.1 0.24 ± 5% perf-profile.self.cycles-pp.touch_atime 0.00 +0.1 0.05 ± 8% perf-profile.self.cycles-pp.wake_up_q 0.00 +0.1 0.05 ± 8% perf-profile.self.cycles-pp.get_mem_cgroup_from_objcg 0.00 +0.1 0.05 ± 8% perf-profile.self.cycles-pp.__count_memcg_events 0.05 ± 46% +0.1 0.11 ± 3% perf-profile.self.cycles-pp.wake_q_add 0.13 ± 14% +0.1 0.19 ± 7% perf-profile.self.cycles-pp.ktime_get_coarse_real_ts64 0.13 ± 11% +0.1 0.19 ± 5% perf-profile.self.cycles-pp.__fdget_pos 0.06 ± 11% +0.1 0.12 ± 3% perf-profile.self.cycles-pp.copyout 0.06 ± 49% +0.1 0.13 ± 8% perf-profile.self.cycles-pp.__mutex_unlock_slowpath 0.16 ± 7% +0.1 0.23 perf-profile.self.cycles-pp.syscall_exit_to_user_mode_prepare 0.09 ± 9% +0.1 0.17 ± 2% perf-profile.self.cycles-pp.__might_fault 0.25 ± 7% +0.1 0.33 ± 2% perf-profile.self.cycles-pp.__cond_resched 0.09 ± 11% +0.1 0.17 ± 2% perf-profile.self.cycles-pp.iov_iter_init 0.00 +0.1 0.08 ± 4% perf-profile.self.cycles-pp.uncharge_batch 0.06 ± 9% +0.1 0.14 ± 5% perf-profile.self.cycles-pp.rw_verify_area 0.17 ± 11% +0.1 0.26 ± 2% perf-profile.self.cycles-pp.rcu_all_qs 0.06 ± 11% +0.1 0.16 ± 4% perf-profile.self.cycles-pp.copyin 0.28 ± 10% +0.1 0.39 ± 2% perf-profile.self.cycles-pp.file_update_time 0.41 ± 8% +0.1 0.52 ± 2% perf-profile.self.cycles-pp.do_syscall_64 0.50 ± 13% +0.1 0.62 ± 2% perf-profile.self.cycles-pp.new_sync_read 0.55 ± 9% +0.1 0.67 ± 3% perf-profile.self.cycles-pp.vfs_read 0.08 ± 23% +0.1 0.21 ± 8% perf-profile.self.cycles-pp.__mutex_lock 0.31 ± 10% +0.1 0.45 perf-profile.self.cycles-pp.aa_file_perm 0.18 ± 14% +0.1 0.32 ± 3% perf-profile.self.cycles-pp.syscall_enter_from_user_mode 0.24 ± 7% +0.1 0.38 perf-profile.self.cycles-pp.timestamp_truncate 0.33 ± 8% +0.2 0.48 ± 3% perf-profile.self.cycles-pp.atime_needs_update 0.25 ± 7% +0.2 0.43 ± 7% perf-profile.self.cycles-pp.__x64_sys_read 1.29 ± 9% +0.2 1.48 perf-profile.self.cycles-pp.pipe_read 0.29 ± 9% +0.2 0.48 ± 3% perf-profile.self.cycles-pp.current_time 0.36 ± 11% +0.2 0.56 ± 8% perf-profile.self.cycles-pp.ksys_read 0.27 ± 5% +0.2 0.49 ± 4% perf-profile.self.cycles-pp.__x64_sys_write 0.65 ± 10% +0.2 0.89 perf-profile.self.cycles-pp.__pthread_disable_asynccancel 0.69 ± 9% +0.2 0.93 perf-profile.self.cycles-pp.__might_sleep 0.76 ± 10% +0.2 1.01 ± 3% perf-profile.self.cycles-pp.entry_SYSCALL_64_after_hwframe 0.00 +0.3 0.27 ± 9% perf-profile.self.cycles-pp.mutex_spin_on_owner 0.00 +0.3 0.29 ± 5% perf-profile.self.cycles-pp.page_counter_uncharge 0.53 ± 7% +0.3 0.84 ± 4% perf-profile.self.cycles-pp.new_sync_write 0.38 ± 8% +0.3 0.70 ± 3% perf-profile.self.cycles-pp.map_id_range_down 0.56 ± 6% +0.3 0.88 ± 2% perf-profile.self.cycles-pp.vfs_write 0.29 ± 10% +0.3 0.62 ± 5% perf-profile.self.cycles-pp.ksys_write 0.42 ± 8% +0.4 0.79 ± 2% perf-profile.self.cycles-pp.copy_page_from_iter 0.65 ± 10% +0.4 1.07 perf-profile.self.cycles-pp.__libc_read 0.54 ± 9% +0.4 0.98 perf-profile.self.cycles-pp.copy_page_to_iter 0.63 ± 10% +0.5 1.12 ± 2% perf-profile.self.cycles-pp.__libc_write 0.53 ± 9% +0.5 1.03 ± 2% perf-profile.self.cycles-pp.__pthread_enable_asynccancel 0.40 ± 11% +0.5 0.93 ± 3% perf-profile.self.cycles-pp.apparmor_file_permission 0.33 ± 11% +0.6 0.88 ± 13% perf-profile.self.cycles-pp.security_file_permission 0.54 ± 9% +0.6 1.11 ± 2% perf-profile.self.cycles-pp.fput_many 1.24 ± 8% +0.6 1.88 perf-profile.self.cycles-pp.__might_resched 1.46 ± 9% +0.7 2.20 ± 2% perf-profile.self.cycles-pp.pipe_write 1.05 ± 10% +0.8 1.82 perf-profile.self.cycles-pp.mutex_unlock 0.60 ± 11% +0.9 1.51 ± 4% perf-profile.self.cycles-pp.entry_SYSCALL_64_safe_stack 1.66 ± 10% +1.1 2.79 ± 4% perf-profile.self.cycles-pp.mutex_lock 0.03 ± 70% +1.1 1.17 ± 3% perf-profile.self.cycles-pp.page_counter_charge 2.04 ± 8% +1.4 3.41 ± 8% perf-profile.self.cycles-pp.__fget_light 1.37 ± 6% +1.5 2.83 ± 9% perf-profile.self.cycles-pp.common_file_perm 2.07 ± 9% +1.5 3.56 perf-profile.self.cycles-pp.copy_user_enhanced_fast_string 0.15 ± 18% +2.5 2.69 ± 2% perf-profile.self.cycles-pp.page_counter_cancel 0.08 ± 19% +3.8 3.85 ± 4% perf-profile.self.cycles-pp.propagate_protected_usage 6.36 ± 9% +4.7 11.02 perf-profile.self.cycles-pp.syscall_return_via_sysret 5.50 ± 9% +5.3 10.78 perf-profile.self.cycles-pp.__entry_text_start 11.40 ± 9% +8.9 20.30 perf-profile.self.cycles-pp.syscall_exit_to_user_mode Disclaimer: Results have been estimated based on internal Intel analysis and are provided for informational purposes only. Any difference in system hardware or software design or configuration may affect actual performance. --- 0DAY/LKP+ Test Infrastructure Open Source Technology Center https://lists.01.org/hyperkitty/list/lkp@lists.01.org Intel Corporation Thanks, Oliver Sang