Greeting, FYI, we noticed a -3.4% regression of reaim.jobs_per_min due to commit: commit: 46ad0840b1584b92b5ff2cc3ed0b011dd6b8e0f1 ("locking/rwsem: Remove arch specific rwsem files") https://git.kernel.org/cgit/linux/kernel/git/tip/tip.git WIP.locking/core in testcase: reaim on test machine: 48 threads Intel(R) Xeon(R) CPU E5-2697 v2 @ 2.70GHz with 64G memory with following parameters: runtime: 300s nr_task: 100t test: fork_test cpufreq_governor: performance ucode: 0x42d test-description: REAIM is an updated and improved version of AIM 7 benchmark. test-url: https://sourceforge.net/projects/re-aim-7/ Details are as below: --------------------------------------------------------------------------------------------------> To reproduce: git clone https://github.com/intel/lkp-tests.git cd lkp-tests bin/lkp install job.yaml # job file is attached in this email bin/lkp run job.yaml ========================================================================================= compiler/cpufreq_governor/kconfig/nr_task/rootfs/runtime/tbox_group/test/testcase/ucode: gcc-7/performance/x86_64-rhel-7.6/100t/debian-x86_64-2018-04-03.cgz/300s/ivb44/fork_test/reaim/0x42d commit: a1247d06d0 ("locking/static_key: Fix false positive warnings on concurrent dec/inc") 46ad0840b1 ("locking/rwsem: Remove arch specific rwsem files") a1247d06d01045d7 46ad0840b1584b92b5ff2cc3ed0 ---------------- --------------------------- fail:runs %reproduction fail:runs | | | :5 40% 2:6 dmesg.WARNING:at#for_ip_swapgs_restore_regs_and_return_to_usermode/0x 1:5 8% 2:6 perf-profile.calltrace.cycles-pp.error_entry %stddev %change %stddev \ | \ 319.20 +4.5% 333.70 reaim.child_systime 228.14 +1.2% 230.79 reaim.child_utime 50164 -3.4% 48470 reaim.jobs_per_min 501.65 -3.4% 484.70 reaim.jobs_per_min_child 51181 -2.4% 49944 reaim.max_jobs_per_min 11.96 +3.5% 12.38 reaim.parent_time 307.80 -1.7% 302.61 reaim.time.elapsed_time 307.80 -1.7% 302.61 reaim.time.elapsed_time.max 21833132 -4.8% 20791331 reaim.time.involuntary_context_switches 7.594e+08 -4.4% 7.263e+08 reaim.time.minor_page_faults 5038 -3.4% 4865 reaim.time.user_time 220000 -4.5% 210000 reaim.workload 14617 ± 2% +10.3% 16128 ± 3% meminfo.PageTables 0.03 ± 11% -0.0 0.02 ± 14% mpstat.cpu.all.soft% 320607 +2.6% 328953 vmstat.system.cs 15491 -86.2% 2133 ±134% numa-numastat.node0.other_node 228.33 ± 5% +5855.5% 13598 ± 21% numa-numastat.node1.other_node 108917 ± 4% -7.8% 100423 ± 3% softirqs.CPU11.TIMER 113369 -8.1% 104175 ± 5% softirqs.CPU40.TIMER 8966761 ± 5% +10.7% 9926924 ± 4% cpuidle.C1.usage 964538 ± 7% +88.9% 1821601 ± 11% cpuidle.POLL.time 803275 ± 10% +118.2% 1752897 ± 13% cpuidle.POLL.usage 8957861 ± 5% +10.8% 9923511 ± 4% turbostat.C1 0.43 ± 69% -0.4 0.00 turbostat.PKG_% 0.02 ±141% +660.0% 0.13 ± 65% turbostat.Pkg%pc3 0.34 ± 62% -0.3 0.00 turbostat.RAM_% 789.33 ± 3% -21.6% 618.67 ± 9% slabinfo.kmalloc-rcl-128.active_objs 789.33 ± 3% -21.6% 618.67 ± 9% slabinfo.kmalloc-rcl-128.num_objs 1366 -10.2% 1227 ± 5% slabinfo.pool_workqueue.active_objs 1367 -10.0% 1230 ± 5% slabinfo.pool_workqueue.num_objs 341.33 ± 8% +25.0% 426.67 ± 7% slabinfo.skbuff_fclone_cache.active_objs 341.33 ± 8% +25.0% 426.67 ± 7% slabinfo.skbuff_fclone_cache.num_objs 153786 +9.3% 168087 numa-vmstat.node0.nr_file_pages 2967 ± 30% +61.5% 4793 ± 5% numa-vmstat.node0.nr_inactive_anon 3513 ± 15% +30.3% 4578 ± 2% numa-vmstat.node0.nr_mapped 3349 ± 26% +382.3% 16155 ± 2% numa-vmstat.node0.nr_shmem 6987 ± 10% +65.6% 11569 ± 3% numa-vmstat.node0.nr_slab_reclaimable 2967 ± 30% +61.5% 4793 ± 5% numa-vmstat.node0.nr_zone_inactive_anon 25044 ± 54% -90.7% 2340 ±121% numa-vmstat.node0.numa_other 655.67 ± 71% +62.0% 1062 numa-vmstat.node1.nr_active_file 169454 -8.4% 155176 numa-vmstat.node1.nr_file_pages 2152 ± 41% -82.8% 369.33 ± 59% numa-vmstat.node1.nr_inactive_anon 3903 ± 14% -29.0% 2770 ± 2% numa-vmstat.node1.nr_mapped 13327 ± 5% -95.1% 652.33 ± 35% numa-vmstat.node1.nr_shmem 11241 ± 6% -46.1% 6064 ± 6% numa-vmstat.node1.nr_slab_reclaimable 655.67 ± 71% +62.0% 1062 numa-vmstat.node1.nr_zone_active_file 2152 ± 41% -82.8% 369.33 ± 59% numa-vmstat.node1.nr_zone_inactive_anon 79276 -2.1% 77572 proc-vmstat.nr_active_anon 66149 -2.9% 64243 proc-vmstat.nr_anon_pages 7467 -1.6% 7346 proc-vmstat.nr_mapped 3756 ± 2% +5.3% 3956 proc-vmstat.nr_page_table_pages 18126 -3.1% 17561 proc-vmstat.nr_slab_reclaimable 79276 -2.1% 77572 proc-vmstat.nr_zone_active_anon 8473 +8.1% 9162 ± 7% proc-vmstat.numa_hint_faults 7.766e+08 -3.9% 7.459e+08 proc-vmstat.numa_hit 7.765e+08 -3.9% 7.459e+08 proc-vmstat.numa_local 984.00 ± 74% +885.9% 9701 ± 97% proc-vmstat.numa_pages_migrated 38689 ± 39% +47.3% 56992 ± 12% proc-vmstat.numa_pte_updates 8.078e+08 -4.0% 7.753e+08 proc-vmstat.pgalloc_normal 7.608e+08 -4.4% 7.276e+08 proc-vmstat.pgfault 8.077e+08 -4.0% 7.752e+08 proc-vmstat.pgfree 984.00 ± 74% +885.9% 9701 ± 97% proc-vmstat.pgmigrate_success 615148 +9.3% 672325 numa-meminfo.node0.FilePages 12293 ± 25% +55.2% 19080 ± 5% numa-meminfo.node0.Inactive 11872 ± 30% +60.7% 19080 ± 5% numa-meminfo.node0.Inactive(anon) 27949 ± 10% +64.6% 46006 ± 3% numa-meminfo.node0.KReclaimable 14085 ± 15% +28.0% 18031 numa-meminfo.node0.Mapped 27949 ± 10% +64.6% 46006 ± 3% numa-meminfo.node0.SReclaimable 13401 ± 26% +382.0% 64597 ± 2% numa-meminfo.node0.Shmem 104587 ± 4% +14.0% 119233 ± 2% numa-meminfo.node0.Slab 2624 ± 71% +61.9% 4249 numa-meminfo.node1.Active(file) 677803 -8.4% 620707 numa-meminfo.node1.FilePages 9431 ± 32% -72.2% 2619 ± 33% numa-meminfo.node1.Inactive 8710 ± 41% -83.0% 1479 ± 59% numa-meminfo.node1.Inactive(anon) 44608 ± 4% -45.6% 24259 ± 6% numa-meminfo.node1.KReclaimable 15357 ± 14% -28.9% 10924 numa-meminfo.node1.Mapped 44608 ± 4% -45.6% 24259 ± 6% numa-meminfo.node1.SReclaimable 53293 ± 5% -95.1% 2611 ± 35% numa-meminfo.node1.Shmem 109835 ± 4% -16.5% 91708 ± 2% numa-meminfo.node1.Slab 198438 ± 19% -48.3% 102613 ± 29% sched_debug.cfs_rq:/.MIN_vruntime.avg 872601 ± 15% -44.0% 488425 ± 20% sched_debug.cfs_rq:/.MIN_vruntime.stddev 58877 ± 4% -32.5% 39716 ± 12% sched_debug.cfs_rq:/.load.avg 884652 ± 13% -36.3% 563479 sched_debug.cfs_rq:/.load.max 170539 ± 10% -44.5% 94726 ± 11% sched_debug.cfs_rq:/.load.stddev 198438 ± 19% -48.3% 102613 ± 29% sched_debug.cfs_rq:/.max_vruntime.avg 872601 ± 15% -44.0% 488425 ± 20% sched_debug.cfs_rq:/.max_vruntime.stddev 98.43 ± 20% +44.2% 141.96 ± 17% sched_debug.cfs_rq:/.removed.load_avg.avg 4591 ± 20% +42.7% 6553 ± 17% sched_debug.cfs_rq:/.removed.runnable_sum.avg 20.87 ± 11% +50.0% 31.30 ± 14% sched_debug.cfs_rq:/.removed.util_avg.avg 67.81 ± 5% +19.5% 81.04 ± 11% sched_debug.cfs_rq:/.removed.util_avg.stddev 40.35 ± 15% -23.4% 30.90 ± 10% sched_debug.cfs_rq:/.runnable_load_avg.avg 629.17 ± 11% -23.5% 481.17 ± 16% sched_debug.cfs_rq:/.runnable_load_avg.max 118.51 ± 16% -31.1% 81.69 ± 4% sched_debug.cfs_rq:/.runnable_load_avg.stddev 56059 ± 8% -37.8% 34866 ± 13% sched_debug.cfs_rq:/.runnable_weight.avg 876174 ± 14% -37.1% 551504 sched_debug.cfs_rq:/.runnable_weight.max 171272 ± 11% -45.1% 94102 ± 11% sched_debug.cfs_rq:/.runnable_weight.stddev 31227 ± 19% -32.2% 21174 ± 32% sched_debug.cfs_rq:/.spread0.avg 1223 ± 4% +16.5% 1425 ± 3% sched_debug.cfs_rq:/.util_avg.max 213.82 ± 2% +11.2% 237.75 sched_debug.cfs_rq:/.util_avg.stddev 124.17 ± 4% +12.4% 139.59 ± 4% sched_debug.cfs_rq:/.util_est_enqueued.stddev 219522 ± 24% -23.3% 168384 ± 18% sched_debug.cpu.avg_idle.avg 11161 ± 7% -26.2% 8235 ± 16% sched_debug.cpu.avg_idle.min 451.78 ± 7% -15.1% 383.78 ± 7% sched_debug.cpu.cpu_load[1].max 85.00 ± 10% -21.7% 66.56 ± 10% sched_debug.cpu.cpu_load[1].stddev 277.61 ± 10% -17.9% 227.94 ± 9% sched_debug.cpu.cpu_load[2].max 56.14 ± 11% -18.5% 45.73 ± 9% sched_debug.cpu.cpu_load[2].stddev 52735 ± 8% +32.7% 69975 ± 8% sched_debug.cpu.curr->pid.avg 77254 ± 6% +23.7% 95588 ± 6% sched_debug.cpu.curr->pid.max 102587 ± 23% +28.5% 131794 ± 7% sched_debug.cpu.load.stddev 2153 ± 8% -20.9% 1703 ± 23% sched_debug.cpu.nr_load_updates.stddev 12689 ± 7% -11.4% 11237 ± 3% sched_debug.cpu.nr_switches.stddev 0.05 ± 48% +126.8% 0.11 ± 24% sched_debug.cpu.nr_uninterruptible.avg -333.33 +32.1% -440.17 sched_debug.cpu.nr_uninterruptible.min 111.56 ± 2% +18.4% 132.09 sched_debug.cpu.nr_uninterruptible.stddev 8.47e+09 +2.8% 8.707e+09 perf-stat.i.branch-instructions 1.006e+08 -2.5% 98077383 perf-stat.i.branch-misses 1.205e+08 -2.5% 1.174e+08 perf-stat.i.cache-misses 1.034e+09 -2.5% 1.008e+09 perf-stat.i.cache-references 323541 +2.8% 332474 perf-stat.i.context-switches 25454 +5.5% 26847 ± 2% perf-stat.i.cpu-migrations 1.217e+10 +3.9% 1.264e+10 perf-stat.i.dTLB-loads 5.319e+09 -2.7% 5.174e+09 perf-stat.i.dTLB-stores 29193114 -2.4% 28497000 perf-stat.i.iTLB-load-misses 4.178e+10 +3.4% 4.319e+10 perf-stat.i.instructions 1424 +4.5% 1487 perf-stat.i.instructions-per-iTLB-miss 2472806 -2.9% 2402150 perf-stat.i.minor-faults 78070119 -2.8% 75920079 perf-stat.i.node-load-misses 82100776 -2.7% 79853820 perf-stat.i.node-loads 2472804 -2.9% 2402149 perf-stat.i.page-faults 24.75 -5.7% 23.35 perf-stat.overall.MPKI 1.19 -0.1 1.13 perf-stat.overall.branch-miss-rate% 2.85 -3.1% 2.77 perf-stat.overall.cpi 989.57 +2.8% 1017 perf-stat.overall.cycles-between-cache-misses 1431 +5.9% 1515 perf-stat.overall.instructions-per-iTLB-miss 0.35 +3.2% 0.36 perf-stat.overall.ipc 58325624 +6.7% 62258667 perf-stat.overall.path-length 8.442e+09 +2.8% 8.678e+09 perf-stat.ps.branch-instructions 1.003e+08 -2.5% 97753949 perf-stat.ps.branch-misses 1.201e+08 -2.5% 1.171e+08 perf-stat.ps.cache-misses 1.031e+09 -2.5% 1.005e+09 perf-stat.ps.cache-references 322473 +2.8% 331365 perf-stat.ps.context-switches 25370 +5.5% 26758 ± 2% perf-stat.ps.cpu-migrations 1.213e+10 +3.9% 1.26e+10 perf-stat.ps.dTLB-loads 5.302e+09 -2.7% 5.157e+09 perf-stat.ps.dTLB-stores 29096816 -2.4% 28402005 perf-stat.ps.iTLB-load-misses 4.165e+10 +3.4% 4.305e+10 perf-stat.ps.instructions 2464656 -2.9% 2394142 perf-stat.ps.minor-faults 77812636 -2.8% 75666866 perf-stat.ps.node-load-misses 81830010 -2.7% 79587500 perf-stat.ps.node-loads 2464651 -2.9% 2394141 perf-stat.ps.page-faults 244.00 ± 2% +125.7% 550.67 ± 36% interrupts.35:IR-PCI-MSI.2621441-edge.eth0-TxRx-0 184.67 ± 6% -15.2% 156.67 interrupts.36:IR-PCI-MSI.2621442-edge.eth0-TxRx-1 358.33 ± 74% -56.3% 156.67 interrupts.37:IR-PCI-MSI.2621443-edge.eth0-TxRx-2 201599 -1.2% 199147 interrupts.CAL:Function_call_interrupts 283.00 ± 9% -18.4% 231.00 ± 7% interrupts.CPU1.TLB:TLB_shootdowns 1163 ±131% +473.3% 6671 interrupts.CPU10.NMI:Non-maskable_interrupts 1163 ±131% +473.3% 6671 interrupts.CPU10.PMI:Performance_monitoring_interrupts 345.33 ± 8% -34.8% 225.00 ± 16% interrupts.CPU11.TLB:TLB_shootdowns 4375 ± 2% -7.2% 4060 ± 5% interrupts.CPU12.CAL:Function_call_interrupts 4277 ± 3% -9.1% 3888 ± 3% interrupts.CPU13.CAL:Function_call_interrupts 331.00 ± 5% -23.0% 255.00 ± 5% interrupts.CPU14.TLB:TLB_shootdowns 1115 ±141% +303.1% 4496 ± 35% interrupts.CPU16.NMI:Non-maskable_interrupts 1115 ±141% +303.1% 4496 ± 35% interrupts.CPU16.PMI:Performance_monitoring_interrupts 1136 ±137% +197.0% 3375 interrupts.CPU18.NMI:Non-maskable_interrupts 1136 ±137% +197.0% 3375 interrupts.CPU18.PMI:Performance_monitoring_interrupts 247.67 ± 15% +21.0% 299.67 ± 7% interrupts.CPU19.TLB:TLB_shootdowns 245.00 ± 18% +26.3% 309.33 ± 7% interrupts.CPU20.TLB:TLB_shootdowns 270.67 ± 27% +39.5% 377.67 ± 9% interrupts.CPU21.TLB:TLB_shootdowns 99.00 ± 53% +3296.0% 3362 interrupts.CPU22.NMI:Non-maskable_interrupts 99.00 ± 53% +3296.0% 3362 interrupts.CPU22.PMI:Performance_monitoring_interrupts 244.00 ± 2% +125.7% 550.67 ± 36% interrupts.CPU24.35:IR-PCI-MSI.2621441-edge.eth0-TxRx-0 184.67 ± 6% -15.2% 156.67 interrupts.CPU25.36:IR-PCI-MSI.2621442-edge.eth0-TxRx-1 358.33 ± 74% -56.3% 156.67 interrupts.CPU26.37:IR-PCI-MSI.2621443-edge.eth0-TxRx-2 2.00 ± 81% +57150.0% 1145 ±138% interrupts.CPU26.NMI:Non-maskable_interrupts 2.00 ± 81% +57150.0% 1145 ±138% interrupts.CPU26.PMI:Performance_monitoring_interrupts 310.00 ± 5% -28.0% 223.33 ± 5% interrupts.CPU27.TLB:TLB_shootdowns 3360 -66.9% 1113 ±140% interrupts.CPU29.NMI:Non-maskable_interrupts 3360 -66.9% 1113 ±140% interrupts.CPU29.PMI:Performance_monitoring_interrupts 1112 ±141% +402.7% 5592 ± 28% interrupts.CPU3.NMI:Non-maskable_interrupts 1112 ±141% +402.7% 5592 ± 28% interrupts.CPU3.PMI:Performance_monitoring_interrupts 213.67 ± 14% +21.8% 260.33 ± 17% interrupts.CPU31.TLB:TLB_shootdowns 3356 ± 80% -99.5% 17.00 ± 84% interrupts.CPU34.NMI:Non-maskable_interrupts 3356 ± 80% -99.5% 17.00 ± 84% interrupts.CPU34.PMI:Performance_monitoring_interrupts 1123 ±138% +401.6% 5633 ± 28% interrupts.CPU37.NMI:Non-maskable_interrupts 1123 ±138% +401.6% 5633 ± 28% interrupts.CPU37.PMI:Performance_monitoring_interrupts 1143 ±136% +292.3% 4484 ± 34% interrupts.CPU41.NMI:Non-maskable_interrupts 1143 ±136% +292.3% 4484 ± 34% interrupts.CPU41.PMI:Performance_monitoring_interrupts 230.33 ± 9% +25.5% 289.00 ± 12% interrupts.CPU42.TLB:TLB_shootdowns 1134 ±137% +209.6% 3511 ± 71% interrupts.CPU43.NMI:Non-maskable_interrupts 1134 ±137% +209.6% 3511 ± 71% interrupts.CPU43.PMI:Performance_monitoring_interrupts 319.00 ± 7% -33.4% 212.33 ± 5% interrupts.CPU43.TLB:TLB_shootdowns 1146 ±133% +289.2% 4462 ± 35% interrupts.CPU44.NMI:Non-maskable_interrupts 1146 ±133% +289.2% 4462 ± 35% interrupts.CPU44.PMI:Performance_monitoring_interrupts 633.00 ±140% +604.5% 4459 ± 34% interrupts.CPU5.NMI:Non-maskable_interrupts 633.00 ±140% +604.5% 4459 ± 34% interrupts.CPU5.PMI:Performance_monitoring_interrupts 285.33 ± 4% -31.2% 196.33 ± 26% interrupts.CPU6.TLB:TLB_shootdowns 308.00 ± 6% -18.4% 251.33 ± 20% interrupts.CPU9.TLB:TLB_shootdowns 108747 ± 5% +32.4% 144006 ± 13% interrupts.NMI:Non-maskable_interrupts 108747 ± 5% +32.4% 144006 ± 13% interrupts.PMI:Performance_monitoring_interrupts 25.95 ± 4% -19.2 6.78 ± 11% perf-profile.calltrace.cycles-pp.__libc_fork 23.30 ± 5% -17.2 6.15 ± 11% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.__libc_fork 23.30 ± 5% -17.2 6.15 ± 11% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_fork 23.29 ± 5% -17.1 6.14 ± 11% perf-profile.calltrace.cycles-pp._do_fork.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_fork 22.32 ± 4% -16.4 5.91 ± 11% perf-profile.calltrace.cycles-pp.copy_process._do_fork.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_fork 5.57 ± 3% -5.6 0.00 perf-profile.calltrace.cycles-pp.call_rwsem_down_write_failed.down_write.anon_vma_fork.copy_process._do_fork 5.55 ± 3% -5.5 0.00 perf-profile.calltrace.cycles-pp.rwsem_down_write_failed.call_rwsem_down_write_failed.down_write.anon_vma_fork.copy_process 5.34 ± 2% -5.3 0.00 perf-profile.calltrace.cycles-pp.call_rwsem_down_write_failed.down_write.unlink_anon_vmas.free_pgtables.exit_mmap 5.33 ± 2% -5.3 0.00 perf-profile.calltrace.cycles-pp.rwsem_down_write_failed.call_rwsem_down_write_failed.down_write.unlink_anon_vmas.free_pgtables 4.40 ± 10% -3.4 0.99 ± 9% perf-profile.calltrace.cycles-pp.wait 3.97 ± 10% -3.1 0.88 ± 9% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.wait 3.95 ± 10% -3.1 0.88 ± 9% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.wait 3.93 ± 10% -3.0 0.88 ± 9% perf-profile.calltrace.cycles-pp.__do_sys_wait4.do_syscall_64.entry_SYSCALL_64_after_hwframe.wait 3.93 ± 10% -3.0 0.88 ± 9% perf-profile.calltrace.cycles-pp.kernel_wait4.__do_sys_wait4.do_syscall_64.entry_SYSCALL_64_after_hwframe.wait 2.52 ± 9% -0.7 1.84 perf-profile.calltrace.cycles-pp.queued_read_lock_slowpath.do_wait.kernel_wait4.__do_sys_wait4.do_syscall_64 2.83 -0.6 2.26 ± 13% perf-profile.calltrace.cycles-pp.anon_vma_interval_tree_insert.anon_vma_clone.anon_vma_fork.copy_process._do_fork 7.70 ± 2% -0.5 7.15 ± 3% perf-profile.calltrace.cycles-pp.tlb_finish_mmu.exit_mmap.mmput.do_exit.do_group_exit 7.69 ± 2% -0.5 7.14 ± 3% perf-profile.calltrace.cycles-pp.arch_tlb_finish_mmu.tlb_finish_mmu.exit_mmap.mmput.do_exit 7.62 ± 2% -0.5 7.09 ± 3% perf-profile.calltrace.cycles-pp.tlb_flush_mmu_free.arch_tlb_finish_mmu.tlb_finish_mmu.exit_mmap.mmput 5.01 ± 4% -0.5 4.48 ± 2% perf-profile.calltrace.cycles-pp.do_wait.kernel_wait4.__do_sys_wait4.do_syscall_64.entry_SYSCALL_64_after_hwframe 8.09 ± 3% -0.5 7.62 ± 3% perf-profile.calltrace.cycles-pp.unmap_vmas.exit_mmap.mmput.do_exit.do_group_exit 6.17 ± 2% -0.4 5.74 ± 3% perf-profile.calltrace.cycles-pp.release_pages.tlb_flush_mmu_free.arch_tlb_finish_mmu.tlb_finish_mmu.exit_mmap 1.83 ± 2% -0.4 1.43 ± 2% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath.queued_read_lock_slowpath.do_wait.kernel_wait4.__do_sys_wait4 0.61 ± 3% -0.2 0.36 ± 70% perf-profile.calltrace.cycles-pp.lru_add_drain_cpu.lru_add_drain.exit_mmap.mmput.do_exit 0.61 ± 3% -0.2 0.37 ± 70% perf-profile.calltrace.cycles-pp.lru_add_drain.exit_mmap.mmput.do_exit.do_group_exit 0.59 ± 3% -0.2 0.36 ± 70% perf-profile.calltrace.cycles-pp.pagevec_lru_move_fn.lru_add_drain_cpu.lru_add_drain.exit_mmap.mmput 4.11 ± 2% -0.2 3.92 perf-profile.calltrace.cycles-pp.copy_page_range.copy_process._do_fork.do_syscall_64.entry_SYSCALL_64_after_hwframe 3.94 ± 2% -0.2 3.75 perf-profile.calltrace.cycles-pp.copy_p4d_range.copy_page_range.copy_process._do_fork.do_syscall_64 1.01 ± 8% -0.1 0.86 ± 3% perf-profile.calltrace.cycles-pp.queued_write_lock_slowpath.do_exit.do_group_exit.__x64_sys_exit_group.do_syscall_64 0.79 ± 9% -0.1 0.66 ± 5% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath.queued_write_lock_slowpath.do_exit.do_group_exit.__x64_sys_exit_group 1.44 ± 4% -0.1 1.33 perf-profile.calltrace.cycles-pp.free_pages_and_swap_cache.tlb_flush_mmu_free.arch_tlb_finish_mmu.tlb_finish_mmu.exit_mmap 0.76 ± 2% -0.1 0.71 perf-profile.calltrace.cycles-pp.remove_vma.exit_mmap.mmput.do_exit.do_group_exit 1.01 -0.0 0.97 ± 2% perf-profile.calltrace.cycles-pp.__slab_free.kmem_cache_free.unlink_anon_vmas.free_pgtables.exit_mmap 1.10 -0.0 1.07 perf-profile.calltrace.cycles-pp.kmem_cache_free.unlink_anon_vmas.free_pgtables.exit_mmap.mmput 0.60 +0.0 0.65 ± 3% perf-profile.calltrace.cycles-pp.alloc_pages_vma.wp_page_copy.do_wp_page.__handle_mm_fault.handle_mm_fault 0.65 ± 2% +0.1 0.71 ± 6% perf-profile.calltrace.cycles-pp.swapgs_restore_regs_and_return_to_usermode 3.57 ± 5% +0.4 3.93 ± 5% perf-profile.calltrace.cycles-pp.intel_idle.cpuidle_enter_state.do_idle.cpu_startup_entry.start_secondary 4.02 ± 6% +0.4 4.47 ± 6% perf-profile.calltrace.cycles-pp.cpuidle_enter_state.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64 4.65 ± 3% +0.5 5.10 ± 4% perf-profile.calltrace.cycles-pp.secondary_startup_64 4.50 ± 5% +0.5 5.01 ± 5% perf-profile.calltrace.cycles-pp.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64 4.51 ± 6% +0.5 5.02 ± 5% perf-profile.calltrace.cycles-pp.start_secondary.secondary_startup_64 4.50 ± 5% +0.5 5.02 ± 5% perf-profile.calltrace.cycles-pp.cpu_startup_entry.start_secondary.secondary_startup_64 0.00 +0.5 0.52 perf-profile.calltrace.cycles-pp.__alloc_pages_nodemask.alloc_pages_vma.wp_page_copy.do_wp_page.__handle_mm_fault 0.00 +0.6 0.56 ± 2% perf-profile.calltrace.cycles-pp.page_fault.__put_user_4.schedule_tail.ret_from_fork 0.00 +0.6 0.56 ± 2% perf-profile.calltrace.cycles-pp.__put_user_4.schedule_tail.ret_from_fork 1.13 ± 8% +0.6 1.69 perf-profile.calltrace.cycles-pp.ret_from_fork 0.00 +0.7 0.72 ± 3% perf-profile.calltrace.cycles-pp.rwsem_spin_on_owner.rwsem_down_write_failed.down_write.anon_vma_clone.anon_vma_fork 0.00 +0.8 0.83 perf-profile.calltrace.cycles-pp.rwsem_spin_on_owner.rwsem_down_write_failed.down_write.anon_vma_fork.copy_process 0.00 +0.8 0.84 ± 2% perf-profile.calltrace.cycles-pp.rwsem_spin_on_owner.rwsem_down_write_failed.down_write.unlink_anon_vmas.free_pgtables 0.00 +0.9 0.86 perf-profile.calltrace.cycles-pp.schedule_tail.ret_from_fork 9.74 ± 2% +0.9 10.68 ± 5% perf-profile.calltrace.cycles-pp.anon_vma_clone.anon_vma_fork.copy_process._do_fork.do_syscall_64 0.00 +0.9 0.95 perf-profile.calltrace.cycles-pp.wake_up_new_task._do_fork.do_syscall_64.entry_SYSCALL_64_after_hwframe 5.78 ± 3% +1.0 6.81 ± 9% perf-profile.calltrace.cycles-pp.down_write.anon_vma_fork.copy_process._do_fork.do_syscall_64 5.08 ± 3% +1.2 6.33 ± 10% perf-profile.calltrace.cycles-pp.down_write.anon_vma_clone.anon_vma_fork.copy_process._do_fork 5.69 ± 2% +1.3 7.00 ± 10% perf-profile.calltrace.cycles-pp.down_write.unlink_anon_vmas.free_pgtables.exit_mmap.mmput 35.70 +1.5 37.20 perf-profile.calltrace.cycles-pp.do_exit.do_group_exit.__x64_sys_exit_group.do_syscall_64.entry_SYSCALL_64_after_hwframe 35.71 +1.5 37.22 perf-profile.calltrace.cycles-pp.do_group_exit.__x64_sys_exit_group.do_syscall_64.entry_SYSCALL_64_after_hwframe 35.71 +1.5 37.22 perf-profile.calltrace.cycles-pp.__x64_sys_exit_group.do_syscall_64.entry_SYSCALL_64_after_hwframe 32.20 +1.7 33.86 ± 2% perf-profile.calltrace.cycles-pp.mmput.do_exit.do_group_exit.__x64_sys_exit_group.do_syscall_64 32.13 +1.7 33.79 ± 2% perf-profile.calltrace.cycles-pp.exit_mmap.mmput.do_exit.do_group_exit.__x64_sys_exit_group 2.96 ± 8% +1.7 4.65 ± 13% perf-profile.calltrace.cycles-pp.down_write.__put_anon_vma.unlink_anon_vmas.free_pgtables.exit_mmap 3.74 ± 8% +1.7 5.45 ± 11% perf-profile.calltrace.cycles-pp.__put_anon_vma.unlink_anon_vmas.free_pgtables.exit_mmap.mmput 16.96 ± 2% +1.9 18.89 ± 6% perf-profile.calltrace.cycles-pp.anon_vma_fork.copy_process._do_fork.do_syscall_64.entry_SYSCALL_64_after_hwframe 1.13 ± 18% +2.5 3.64 ± 2% perf-profile.calltrace.cycles-pp.kernel_wait4.__do_sys_wait4.do_syscall_64.entry_SYSCALL_64_after_hwframe 1.13 ± 18% +2.5 3.65 perf-profile.calltrace.cycles-pp.__do_sys_wait4.do_syscall_64.entry_SYSCALL_64_after_hwframe 14.70 ± 3% +2.8 17.53 ± 6% perf-profile.calltrace.cycles-pp.free_pgtables.exit_mmap.mmput.do_exit.do_group_exit 12.95 ± 3% +3.0 15.93 ± 8% perf-profile.calltrace.cycles-pp.unlink_anon_vmas.free_pgtables.exit_mmap.mmput.do_exit 0.00 +3.8 3.83 ± 15% perf-profile.calltrace.cycles-pp.osq_lock.rwsem_down_write_failed.down_write.__put_anon_vma.unlink_anon_vmas 0.00 +4.6 4.59 ± 13% perf-profile.calltrace.cycles-pp.rwsem_down_write_failed.down_write.__put_anon_vma.unlink_anon_vmas.free_pgtables 0.00 +4.6 4.65 ± 13% perf-profile.calltrace.cycles-pp.osq_lock.rwsem_down_write_failed.down_write.anon_vma_clone.anon_vma_fork 0.00 +5.0 5.02 ± 11% perf-profile.calltrace.cycles-pp.osq_lock.rwsem_down_write_failed.down_write.anon_vma_fork.copy_process 0.00 +5.2 5.21 ± 13% perf-profile.calltrace.cycles-pp.osq_lock.rwsem_down_write_failed.down_write.unlink_anon_vmas.free_pgtables 0.00 +6.1 6.07 ± 10% perf-profile.calltrace.cycles-pp.rwsem_down_write_failed.down_write.anon_vma_clone.anon_vma_fork.copy_process 0.00 +6.6 6.60 ± 9% perf-profile.calltrace.cycles-pp.rwsem_down_write_failed.down_write.anon_vma_fork.copy_process._do_fork 0.00 +6.6 6.62 ± 11% perf-profile.calltrace.cycles-pp.rwsem_down_write_failed.down_write.unlink_anon_vmas.free_pgtables.exit_mmap 6.72 ± 19% +17.5 24.22 ± 4% perf-profile.calltrace.cycles-pp.copy_process._do_fork.do_syscall_64.entry_SYSCALL_64_after_hwframe 7.01 ± 19% +18.2 25.22 ± 4% perf-profile.calltrace.cycles-pp._do_fork.do_syscall_64.entry_SYSCALL_64_after_hwframe 44.08 ± 3% +22.8 66.84 ± 2% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe 44.08 ± 3% +22.8 66.86 ± 2% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe reaim.time.minor_page_faults 8e+08 +-+-----------------------------------------------------------------+ | OO + O O.O+ O O O +O O OO + O +.+ + + + ++ +.| 7e+08 +-+: : : : : : : :: : : : : : : : : :: : | 6e+08 +-+: : : : : : : :: : : : : : : : : :: : | | : : : : : : : :: : : : : : : : : :: : | 5e+08 +-+:: :: :: : : : : :: : : :: :: : : : : :: : : :: : : : | | :: :: :: : : : : :: : : :: :: : : : : :: : : :: : : : | 4e+08 +-+:: :: :: : : : : :: : : :: :: : : : : :: : : :: : : : | | : :: : : :: : : :: : : : : :: :: :: : : :: :: : : :: | 3e+08 +-+ :: : : :: : : :: : : : : :: :: :: : : :: :: : : :: | 2e+08 +-+ :: : : :: : : :: : : : : :: :: :: : : :: :: : : :: | | : : : : :: : : : : : : : :: : : : : | 1e+08 +-+ : : : :: : : : : : : : :: : : : : | | : : : : :: : : : : : : : :: : : : : | 0 O-+--OO-O--O----O--O---O--O-O---O-O----OO---------------------------+ reaim.time.involuntary_context_switches 2.5e+07 +-+---------------------------------------------------------------+ | | | OO + O OO.+O O O +O O OO + O +.+ + + + ++ +.| 2e+07 +-+: : : : : : : :: : : : : : : : : :: : | | : : : : : : : :: : : : : : : : : :: : | | :: :: :: : : :: : : : : : : : : :: :: : : : : | 1.5e+07 +-+:: :: :: : : :: :: : : :: :: :: : : :: :: :: : : : | | :: :: :: : : :: :: : : :: :: :: : : :: :: :: : : : | 1e+07 +-+:: :: :: : : :: : :: : : :: :: :: : :: :: : :: : : | | : :: :: : : : : : : :: : : :: :: :: : : :: : : :: :: | | : :: :: : : : : : : :: : : :: :: :: : : :: : : :: :: | 5e+06 +-+ : : :: : : :: :: : : :: :: :: : : : :: :: : | | : : : : : : : : : : : : : : : : : : : | | : : : : : : : : : : : : : : : : : : : | 0 O-+--OO-O--O----O--O--O--O-O---O-O----OO--------------------------+ reaim.parent_time 14 +-+--------------------------------------------------------------------+ | OO O O | 12 +-OO + O ++.+ O O O ++ O + O + O +.+ + + + +.+ +.| | : : : : : : : :: : : : : : : : : : : : | 10 +-+: : : : : : : :: : : : : : : : : : : : | | :: : :: : : :: :: : : :: : :: : : :: :: :: : : : | 8 +-+:: :: : : : : : : :: : : :: :: : : : : : : :: : : : : : | | :: :: : : : : : : :: : : :: :: : : : : : : :: : : : : : | 6 +-+:: : :: : : : : : :: : : :: : :: : : : : : :: : : : : : | | : : : :: : : : : :: : : : : : : :: :: : : :: : : :: :: | 4 +-+ : : :: : : : : :: : : : : : : :: :: : : :: : : :: :: | | : :: :: :: : : : :: : : :: :: : : : : :: : : | 2 +-+ : : : :: : : : : : : : :: : : : : | | : : : : :: : : : : : : : :: : : : : | 0 O-+--O-OO---O----O--O---O--O-O----O-O----OO----------------------------+ reaim.child_systime 350 +-+----------O-O---------O----O-O----O--------------------------------+ | OO + O +.+.+O O + +.+ +O + O ++ + + + +.+ +.| 300 +-+: : : : : : : : : : : : :: : : : : : : | | : : : : : : : : : : : : :: : : : : : : | 250 +-+: : : : : : : : : : : : :: : : : : : : | | :: :: : : : : :: : : : : :: : : :: : : :: :: : : : : : | 200 +-+:: :: : : : : :: : : : : :: : : :: : : :: :: : : : : : | | :: :: : : : : :: : : : : :: : : :: : : :: :: : : : : : | 150 +-+ : : :: :: : : :: :: : : :: :: : : : : : : :: :: :: | | : : : :: :: : : :: :: : : :: :: : : : : : : :: :: :: | 100 +-+ : : :: :: : : :: :: : : :: :: : : : : : : :: :: :: | | : : : : : : : : : : : : : : : : : : : | 50 +-+ : : : : : : : : : : : : : : : : : : | | : : : : : : : : : : : : : : : : : : : | 0 O-+--O-OO---O----O--O--O---OO----O-O----O-O---------------------------+ reaim.jobs_per_min 60000 +-+-----------------------------------------------------------------+ | | 50000 +-++ + + +.++ + + ++ + + + +.+ + + + ++ +.| | OO : O O O: O O O :O O OO : O : : : : : :: : | | : : : : : : : :: : : : : : : : : :: : | 40000 +-+:: :: : : : :: :: : : : : :: : : : :: :: : : : | | :: :: :: : : : : :: : : :: :: : : : : :: : : :: : : : | 30000 +-+:: :: :: : : : : :: : : :: :: : : : : :: : : :: : : : | | :: :: : :: : : : :: : : : :: :: : : : : :: : :: : : : | 20000 +-+ :: : : :: : : :: : : : : :: :: :: : : :: :: : : :: | | : :: : : :: : : :: : : : : :: :: :: : : :: :: : : :: | | : : :: :: : : : :: : : :: :: : : : :: : :: : | 10000 +-+ : : : :: : : : : : : : :: : : : : | | : : : : :: : : : : : : : :: : : : : | 0 O-+--OO-O--O----O--O---O--O-O---O-O----OO---------------------------+ reaim.jobs_per_min_child 600 +-+-------------------------------------------------------------------+ | | 500 +-++ + + +.+.+ + + +.+ + + + ++ + + + +.+ +.| | OO : O O O :O O O : O O OO : O :: : : : : : : | | : : : : : : : : : : : : :: : : : : : : | 400 +-+:: : :: : : : :: : : : :: :: : : :: : :: : : : | | :: :: : : : : :: : : : : :: : : :: : : :: :: : : : : : | 300 +-+:: :: : : : : :: : : : : :: : : :: : : :: :: : : : : : | | :: : :: : : : : :: : : : : :: : :: : : :: : :: : : : : | 200 +-+ : : :: :: : : :: :: : : :: :: : : : : : : :: :: :: | | : : : :: :: : : :: :: : : :: :: : : : : : : :: :: :: | | : :: :: : : : :: : : : :: : :: : : :: :: : : | 100 +-+ : : : : : : : : : : : : : : : : : : | | : : : : : : : : : : : : : : : : : : : | 0 O-+--O-OO---O----O--O--O---OO----O-O----O-O---------------------------+ reaim.workload 250000 +-+----------------------------------------------------------------+ | | | OO + O OO.+ O O O +.OO O O+ O ++ + + + ++ +.| 200000 +-+: : : : : : : : : : : : :: : : : :: : | | : : : : : : : : : : : : :: : : : :: : | | :: :: : : : : :: : : :: :: :: : : :: :: :: : : : | 150000 +-+:: :: :: : : :: : : : : :: :: :: : : : : :: :: : : : | | :: :: :: : : :: : : : : :: :: :: : : : : :: :: : : : | 100000 +-+:: :: : :: : : :: : : : :: :: :: : : : : :: :: : : : | | : :: : : :: : : :: :: : : :: :: : : : : :: :: : : :: | | : :: : : :: : : :: :: : : :: :: : : : : :: :: : : :: | 50000 +-+ : :: :: : : :: : : : : : :: : : : : :: : | | : : : : :: : : : : : : : :: : : : : | | : : : : :: : : : : : : : :: : : : : | 0 O-+--OO-O--O----O--O--O---OO----O-O---O-O--------------------------+ [*] bisect-good sample [O] bisect-bad sample 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. Thanks, Rong Chen