FYI, we noticed a -51.9% regression of vm-scalability.throughput due to commit: commit 84b1c66da9468dcbcf33f656a1bd6cfab2d6c80d ("sched/fair: Remove scale_load_down() for load_avg") git://bee.sh.intel.com/git/ydu19/tip sched_avg_opt_and_cleanup_v2.7 in testcase: vm-scalability on test machine: lkp-hsw01: 56 threads Grantley Haswell-EP with 64G memory with following parameters: cpufreq_governor=performance/runtime=300s/size=2T/test=shm-xread-seq-mt Details are as below: --------------------------------------------------------------------------------------------------> To reproduce: git clone git://git.kernel.org/pub/scm/linux/kernel/git/wfg/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/rootfs/runtime/size/tbox_group/test/testcase: gcc-4.9/performance/x86_64-rhel/debian-x86_64-2015-02-07.cgz/300s/2T/lkp-hsw01/shm-xread-seq-mt/vm-scalability commit: 1b3afd693c00319f4303d470d62807e245a79f65 84b1c66da9468dcbcf33f656a1bd6cfab2d6c80d 1b3afd693c00319f 84b1c66da9468dcbcf33f656a1 ---------------- -------------------------- fail:runs %reproduction fail:runs | | | :4 25% 1:4 kmsg.Spurious_LAPIC_timer_interrupt_on_cpu %stddev %change %stddev \ | \ 37668391 ± 0% -51.9% 18116025 ± 0% vm-scalability.throughput 115.57 ± 0% +103.6% 235.25 ± 0% vm-scalability.time.elapsed_time 115.57 ± 0% +103.6% 235.25 ± 0% vm-scalability.time.elapsed_time.max 93867376 ± 0% +114.0% 2.008e+08 ± 0% vm-scalability.time.minor_page_faults 912.20 ± 0% +261.3% 3296 ± 0% vm-scalability.time.system_time 1255 ± 0% -12.4% 1099 ± 0% vm-scalability.time.user_time 48785949 ± 0% +155.8% 1.248e+08 ± 0% vm-scalability.time.voluntary_context_switches 132413 ± 1% +9.3% 144712 ± 2% meminfo.Active 30501 ± 4% +38.0% 42097 ± 6% meminfo.Active(anon) 9932 ± 15% -80.9% 1893 ±164% numa-numastat.node0.other_node 917.50 ±170% +890.5% 9087 ± 34% numa-numastat.node1.other_node 62614 ± 1% +17.6% 73665 ± 8% numa-meminfo.node0.Active 11382 ± 12% +103.4% 23149 ± 22% numa-meminfo.node0.Active(anon) 6983 ± 14% -41.9% 4057 ± 47% numa-meminfo.node1.AnonHugePages 25.75 ± 1% -30.1% 18.00 ± 0% vmstat.procs.r 714664 ± 0% +41.6% 1011775 ± 0% vmstat.system.cs 60803 ± 0% +1.5% 61729 ± 0% vmstat.system.in 7772 ± 3% +27.0% 9868 ± 5% softirqs.NET_RX 174881 ± 9% +291.3% 684355 ± 1% softirqs.RCU 850755 ± 1% +73.2% 1473464 ± 0% softirqs.SCHED 1252182 ± 3% +86.2% 2331071 ± 1% softirqs.TIMER 2845 ± 12% +103.4% 5786 ± 22% numa-vmstat.node0.nr_active_anon 6889706 ± 7% +13.2% 7800953 ± 0% numa-vmstat.node0.numa_hit 6814259 ± 7% +13.7% 7749981 ± 0% numa-vmstat.node0.numa_local 75446 ± 2% -32.4% 50971 ± 57% numa-vmstat.node0.numa_other 16485 ± 9% +148.8% 41020 ± 70% numa-vmstat.node1.numa_other 1074 ± 0% +1.9% 1094 ± 0% turbostat.Avg_MHz 24.50 ± 0% +154.7% 62.39 ± 0% turbostat.CPU%c1 40.00 ± 0% -94.6% 2.18 ± 2% turbostat.CPU%c6 55.75 ± 1% +15.2% 64.25 ± 1% turbostat.CoreTmp 32.69 ± 0% -97.2% 0.91 ± 19% turbostat.Pkg%pc2 60.25 ± 2% +14.5% 69.00 ± 1% turbostat.PkgTmp 174.85 ± 0% +20.1% 209.98 ± 0% turbostat.PkgWatt 21.27 ± 2% +36.3% 28.99 ± 0% turbostat.RAMWatt 9.077e+08 ± 0% +464.7% 5.126e+09 ± 0% cpuidle.C1-HSW.time 34488483 ± 0% +204.4% 1.05e+08 ± 0% cpuidle.C1-HSW.usage 32091141 ± 0% +2179.6% 7.316e+08 ± 1% cpuidle.C1E-HSW.time 425737 ± 1% +2620.6% 11582628 ± 1% cpuidle.C1E-HSW.usage 2503376 ± 4% +1103.9% 30136985 ± 1% cpuidle.C3-HSW.time 9350 ± 4% +1849.4% 182271 ± 1% cpuidle.C3-HSW.usage 3.183e+09 ± 0% -22.4% 2.469e+09 ± 0% cpuidle.C6-HSW.time 3397123 ± 0% -23.8% 2587962 ± 0% cpuidle.C6-HSW.usage 48923914 ± 3% +260.5% 1.764e+08 ± 2% cpuidle.POLL.time 725127 ± 0% +232.9% 2413961 ± 0% cpuidle.POLL.usage 7624 ± 4% +38.0% 10524 ± 6% proc-vmstat.nr_active_anon 7966 ± 5% +90.8% 15197 ± 2% proc-vmstat.numa_hint_faults 4343 ± 6% +86.1% 8082 ± 2% proc-vmstat.numa_hint_faults_local 26506353 ± 0% +11.5% 29544930 ± 0% proc-vmstat.numa_hit 26495504 ± 0% +11.5% 29533949 ± 0% proc-vmstat.numa_local 1239 ± 4% +609.5% 8793 ± 97% proc-vmstat.numa_pages_migrated 8062 ± 5% +703.3% 64769 ± 87% proc-vmstat.numa_pte_updates 6587 ± 7% +29.0% 8498 ± 8% proc-vmstat.pgactivate 25724719 ± 0% +11.5% 28680319 ± 0% proc-vmstat.pgalloc_normal 94133198 ± 0% +113.9% 2.014e+08 ± 0% proc-vmstat.pgfault 24677874 ± 2% +13.9% 28118027 ± 2% proc-vmstat.pgfree 1239 ± 4% +609.5% 8793 ± 97% proc-vmstat.pgmigrate_success 9.148e+10 ± 0% +45.6% 1.332e+11 ± 0% perf-stat.L1-dcache-load-misses 2.408e+12 ± 0% +35.9% 3.273e+12 ± 1% perf-stat.L1-dcache-loads 7.486e+11 ± 0% +50.4% 1.126e+12 ± 1% perf-stat.L1-dcache-stores 1.858e+10 ± 0% +54.6% 2.874e+10 ± 0% perf-stat.L1-icache-load-misses 90333622 ± 2% +6383.1% 5.856e+09 ± 1% perf-stat.LLC-load-misses 1.831e+10 ± 0% +78.9% 3.275e+10 ± 1% perf-stat.LLC-loads 65489933 ± 1% +4490.6% 3.006e+09 ± 2% perf-stat.LLC-store-misses 5.403e+09 ± 0% +109.8% 1.134e+10 ± 0% perf-stat.LLC-stores 3.322e+12 ± 2% +25.8% 4.179e+12 ± 3% perf-stat.branch-instructions 4.477e+09 ± 2% +125.8% 1.011e+10 ± 0% perf-stat.branch-load-misses 3.251e+12 ± 1% +27.7% 4.152e+12 ± 1% perf-stat.branch-loads 4.533e+09 ± 3% +122.6% 1.009e+10 ± 1% perf-stat.branch-misses 2.258e+11 ± 0% +102.4% 4.571e+11 ± 0% perf-stat.bus-cycles 1.664e+08 ± 4% +5294.7% 8.976e+09 ± 2% perf-stat.cache-misses 3.473e+10 ± 2% +59.0% 5.523e+10 ± 1% perf-stat.cache-references 84147940 ± 0% +185.6% 2.403e+08 ± 0% perf-stat.context-switches 6.927e+12 ± 2% +108.4% 1.444e+13 ± 0% perf-stat.cpu-cycles 19621508 ± 0% +115.8% 42339374 ± 0% perf-stat.cpu-migrations 1.247e+09 ± 5% +102.5% 2.525e+09 ± 4% perf-stat.dTLB-load-misses 2.413e+12 ± 0% +33.7% 3.225e+12 ± 2% perf-stat.dTLB-loads 1.103e+08 ± 6% +26.7% 1.398e+08 ± 4% perf-stat.dTLB-store-misses 7.431e+11 ± 1% +50.5% 1.118e+12 ± 1% perf-stat.dTLB-stores 6.621e+08 ± 1% +174.7% 1.819e+09 ± 0% perf-stat.iTLB-loads 1.014e+13 ± 2% +33.2% 1.351e+13 ± 1% perf-stat.instructions 94122563 ± 0% +113.9% 2.013e+08 ± 0% perf-stat.minor-faults 77552115 ± 8% +7339.8% 5.77e+09 ± 2% perf-stat.node-load-misses 17020549 ± 9% +584.8% 1.165e+08 ± 2% perf-stat.node-loads 10450654 ± 5% +17487.2% 1.838e+09 ± 2% perf-stat.node-store-misses 52951139 ± 2% +2079.9% 1.154e+09 ± 2% perf-stat.node-stores 94122616 ± 0% +113.9% 2.013e+08 ± 0% perf-stat.page-faults 5.857e+12 ± 0% +103.7% 1.193e+13 ± 2% perf-stat.ref-cycles 8.88 ± 2% -26.3% 6.54 ± 1% perf-profile.cycles-pp.__account_scheduler_latency.enqueue_entity.enqueue_task_fair.activate_task.ttwu_do_activate 7.95 ± 1% -34.1% 5.24 ± 1% perf-profile.cycles-pp.__do_fault.handle_pte_fault.handle_mm_fault.__do_page_fault.do_page_fault 4.66 ± 2% -32.3% 3.15 ± 1% perf-profile.cycles-pp.__kernel_text_address.print_context_stack.dump_trace.save_stack_trace_tsk.__account_scheduler_latency 4.83 ± 1% -26.7% 3.54 ± 1% perf-profile.cycles-pp.__lock_page.find_lock_entry.shmem_getpage_gfp.shmem_fault.__do_fault 2.82 ± 4% -53.9% 1.30 ± 1% perf-profile.cycles-pp.__schedule.schedule.schedule_preempt_disabled.cpu_startup_entry.start_secondary 2.67 ± 1% -24.9% 2.01 ± 2% perf-profile.cycles-pp.__schedule.schedule.schedule_timeout.io_schedule_timeout.bit_wait_io 4.72 ± 1% -26.4% 3.48 ± 1% perf-profile.cycles-pp.__wait_on_bit_lock.__lock_page.find_lock_entry.shmem_getpage_gfp.shmem_fault 15.90 ± 2% -47.1% 8.41 ± 1% perf-profile.cycles-pp.__wake_up.__wake_up_bit.unlock_page.handle_pte_fault.handle_mm_fault 16.06 ± 2% -46.6% 8.57 ± 1% perf-profile.cycles-pp.__wake_up_bit.unlock_page.handle_pte_fault.handle_mm_fault.__do_page_fault 15.43 ± 2% -46.5% 8.25 ± 1% perf-profile.cycles-pp.__wake_up_common.__wake_up.__wake_up_bit.unlock_page.handle_pte_fault 1.69 ± 6% -100.0% 0.00 ± -1% perf-profile.cycles-pp._raw_spin_lock.__schedule.schedule.schedule_preempt_disabled.cpu_startup_entry 16.88 ± 2% +83.5% 30.96 ± 1% perf-profile.cycles-pp._raw_spin_lock.handle_pte_fault.handle_mm_fault.__do_page_fault.do_page_fault 2.10 ± 2% -100.0% 0.00 ± -1% perf-profile.cycles-pp._raw_spin_lock.try_to_wake_up.default_wake_function.autoremove_wake_function.wake_bit_function 0.00 ± -1% +Inf% 2.46 ± 2% perf-profile.cycles-pp.activate_task.ttwu_do_activate.sched_ttwu_pending.cpu_startup_entry.start_secondary 10.39 ± 2% -47.5% 5.45 ± 1% perf-profile.cycles-pp.activate_task.ttwu_do_activate.try_to_wake_up.default_wake_function.autoremove_wake_function 15.20 ± 2% -47.1% 8.04 ± 1% perf-profile.cycles-pp.autoremove_wake_function.wake_bit_function.__wake_up_common.__wake_up.__wake_up_bit 3.36 ± 1% -27.4% 2.44 ± 2% perf-profile.cycles-pp.bit_wait_io.__wait_on_bit_lock.__lock_page.find_lock_entry.shmem_getpage_gfp 45.79 ± 2% -7.5% 42.36 ± 1% perf-profile.cycles-pp.call_cpuidle.cpu_startup_entry.start_secondary 45.76 ± 2% -7.5% 42.34 ± 1% perf-profile.cycles-pp.cpuidle_enter.call_cpuidle.cpu_startup_entry.start_secondary 1.44 ± 3% -30.3% 1.01 ± 1% perf-profile.cycles-pp.deactivate_task.__schedule.schedule.schedule_timeout.io_schedule_timeout 15.11 ± 2% -47.7% 7.90 ± 1% perf-profile.cycles-pp.default_wake_function.autoremove_wake_function.wake_bit_function.__wake_up_common.__wake_up 1.17 ± 3% -26.3% 0.86 ± 0% perf-profile.cycles-pp.dequeue_entity.dequeue_task_fair.deactivate_task.__schedule.schedule 1.31 ± 3% -29.5% 0.93 ± 1% perf-profile.cycles-pp.dequeue_task_fair.deactivate_task.__schedule.schedule.schedule_timeout 8.06 ± 2% -30.3% 5.62 ± 0% perf-profile.cycles-pp.dump_trace.save_stack_trace_tsk.__account_scheduler_latency.enqueue_entity.enqueue_task_fair 0.00 ± -1% +Inf% 2.28 ± 2% perf-profile.cycles-pp.enqueue_entity.enqueue_task_fair.activate_task.ttwu_do_activate.sched_ttwu_pending 9.95 ± 2% -47.3% 5.24 ± 1% perf-profile.cycles-pp.enqueue_entity.enqueue_task_fair.activate_task.ttwu_do_activate.try_to_wake_up 0.00 ± -1% +Inf% 2.39 ± 2% perf-profile.cycles-pp.enqueue_task_fair.activate_task.ttwu_do_activate.sched_ttwu_pending.cpu_startup_entry 10.19 ± 2% -47.7% 5.33 ± 1% perf-profile.cycles-pp.enqueue_task_fair.activate_task.ttwu_do_activate.try_to_wake_up.default_wake_function 1.42 ± 3% -22.1% 1.11 ± 1% perf-profile.cycles-pp.filemap_map_pages.handle_pte_fault.handle_mm_fault.__do_page_fault.do_page_fault 5.42 ± 1% -24.6% 4.09 ± 1% perf-profile.cycles-pp.find_lock_entry.shmem_getpage_gfp.shmem_fault.__do_fault.handle_pte_fault 3.34 ± 1% -27.3% 2.43 ± 1% perf-profile.cycles-pp.io_schedule_timeout.bit_wait_io.__wait_on_bit_lock.__lock_page.find_lock_entry 1.04 ± 2% -49.4% 0.53 ± 2% perf-profile.cycles-pp.is_ftrace_trampoline.__kernel_text_address.print_context_stack.dump_trace.save_stack_trace_tsk 1.57 ± 1% -52.9% 0.74 ± 2% perf-profile.cycles-pp.is_module_text_address.__kernel_text_address.print_context_stack.dump_trace.save_stack_trace_tsk 1.69 ± 6% -100.0% 0.00 ± -1% perf-profile.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock.__schedule.schedule.schedule_preempt_disabled 16.79 ± 2% +83.4% 30.80 ± 1% perf-profile.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock.handle_pte_fault.handle_mm_fault.__do_page_fault 2.10 ± 2% -100.0% 0.00 ± -1% perf-profile.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock.try_to_wake_up.default_wake_function.autoremove_wake_function 7.35 ± 2% -30.5% 5.11 ± 0% perf-profile.cycles-pp.print_context_stack.dump_trace.save_stack_trace_tsk.__account_scheduler_latency.enqueue_entity 0.98 ± 3% -27.9% 0.70 ± 1% perf-profile.cycles-pp.radix_tree_next_chunk.filemap_map_pages.handle_pte_fault.handle_mm_fault.__do_page_fault 8.08 ± 2% -30.3% 5.63 ± 0% perf-profile.cycles-pp.save_stack_trace_tsk.__account_scheduler_latency.enqueue_entity.enqueue_task_fair.activate_task 0.00 ± -1% +Inf% 2.74 ± 1% perf-profile.cycles-pp.sched_ttwu_pending.cpu_startup_entry.start_secondary 2.90 ± 4% -52.2% 1.39 ± 1% perf-profile.cycles-pp.schedule.schedule_preempt_disabled.cpu_startup_entry.start_secondary 2.78 ± 1% -25.1% 2.08 ± 2% perf-profile.cycles-pp.schedule.schedule_timeout.io_schedule_timeout.bit_wait_io.__wait_on_bit_lock 2.92 ± 4% -52.0% 1.40 ± 2% perf-profile.cycles-pp.schedule_preempt_disabled.cpu_startup_entry.start_secondary 2.81 ± 1% -25.7% 2.09 ± 2% perf-profile.cycles-pp.schedule_timeout.io_schedule_timeout.bit_wait_io.__wait_on_bit_lock.__lock_page 1.33 ± 2% -27.8% 0.96 ± 1% perf-profile.cycles-pp.select_task_rq_fair.try_to_wake_up.default_wake_function.autoremove_wake_function.wake_bit_function 7.89 ± 1% -34.0% 5.21 ± 1% perf-profile.cycles-pp.shmem_fault.__do_fault.handle_pte_fault.handle_mm_fault.__do_page_fault 7.26 ± 1% -31.4% 4.98 ± 1% perf-profile.cycles-pp.shmem_getpage_gfp.shmem_fault.__do_fault.handle_pte_fault.handle_mm_fault 14.94 ± 2% -48.1% 7.75 ± 1% perf-profile.cycles-pp.try_to_wake_up.default_wake_function.autoremove_wake_function.wake_bit_function.__wake_up_common 0.00 ± -1% +Inf% 2.57 ± 1% perf-profile.cycles-pp.ttwu_do_activate.sched_ttwu_pending.cpu_startup_entry.start_secondary 10.79 ± 3% -47.2% 5.70 ± 1% perf-profile.cycles-pp.ttwu_do_activate.try_to_wake_up.default_wake_function.autoremove_wake_function.wake_bit_function 16.31 ± 2% -44.8% 9.01 ± 1% perf-profile.cycles-pp.unlock_page.handle_pte_fault.handle_mm_fault.__do_page_fault.do_page_fault 15.22 ± 2% -47.1% 8.05 ± 1% perf-profile.cycles-pp.wake_bit_function.__wake_up_common.__wake_up.__wake_up_bit.unlock_page 354.29 ± 56% +25157.9% 89486 ± 38% sched_debug.cfs_rq:/.MIN_vruntime.avg 11512 ± 14% +17893.8% 2071559 ± 24% sched_debug.cfs_rq:/.MIN_vruntime.max 1888 ± 32% +20508.3% 389104 ± 27% sched_debug.cfs_rq:/.MIN_vruntime.stddev 9944 ± 0% +203.0% 30129 ± 0% sched_debug.cfs_rq:/.exec_clock.avg 12551 ± 5% +220.5% 40231 ± 0% sched_debug.cfs_rq:/.exec_clock.max 7832 ± 3% +143.3% 19059 ± 1% sched_debug.cfs_rq:/.exec_clock.min 1731 ± 7% +436.0% 9283 ± 0% sched_debug.cfs_rq:/.exec_clock.stddev 306374 ± 9% -97.8% 6654 ± 13% sched_debug.cfs_rq:/.load.avg 1025203 ± 0% -96.2% 39390 ± 6% sched_debug.cfs_rq:/.load.max 448149 ± 3% -97.7% 10314 ± 3% sched_debug.cfs_rq:/.load.stddev 116.25 ± 4% +14275.9% 16711 ± 13% sched_debug.cfs_rq:/.load_avg.avg 653.62 ± 6% +38475.7% 252140 ± 3% sched_debug.cfs_rq:/.load_avg.max 148.90 ± 17% +38464.5% 57422 ± 9% sched_debug.cfs_rq:/.load_avg.stddev 354.29 ± 56% +25157.9% 89486 ± 38% sched_debug.cfs_rq:/.max_vruntime.avg 11512 ± 14% +17893.8% 2071559 ± 24% sched_debug.cfs_rq:/.max_vruntime.max 1888 ± 32% +20508.3% 389104 ± 27% sched_debug.cfs_rq:/.max_vruntime.stddev 11289 ± 0% +17228.9% 1956363 ± 0% sched_debug.cfs_rq:/.min_vruntime.avg 24343 ± 6% +10296.9% 2530993 ± 0% sched_debug.cfs_rq:/.min_vruntime.max 8522 ± 3% +15221.3% 1305727 ± 1% sched_debug.cfs_rq:/.min_vruntime.min 3036 ± 5% +17356.2% 530134 ± 0% sched_debug.cfs_rq:/.min_vruntime.stddev 37.28 ± 46% +2740.5% 1058 ± 40% sched_debug.cfs_rq:/.runnable_load_avg.avg 209.75 ± 22% +10452.6% 22134 ±117% sched_debug.cfs_rq:/.runnable_load_avg.max 55.64 ± 37% +6209.2% 3510 ± 92% sched_debug.cfs_rq:/.runnable_load_avg.stddev -11180 ±-13% +4743.3% -541529 ± 0% sched_debug.cfs_rq:/.spread0.avg 1874 ± 99% +1671.2% 33198 ± 17% sched_debug.cfs_rq:/.spread0.max -13949 ±-12% +8446.9% -1192219 ± -1% sched_debug.cfs_rq:/.spread0.min 3037 ± 5% +17354.8% 530186 ± 0% sched_debug.cfs_rq:/.spread0.stddev 248.46 ± 6% +14.5% 284.43 ± 0% sched_debug.cfs_rq:/.util_avg.avg 641.38 ± 7% -17.7% 527.69 ± 2% sched_debug.cfs_rq:/.util_avg.max 62.62 ± 88% -95.7% 2.69 ± 98% sched_debug.cfs_rq:/.util_avg.min 391160 ± 3% -45.1% 214776 ± 10% sched_debug.cpu.avg_idle.avg 71428 ± 6% +81.9% 129931 ± 3% sched_debug.cpu.clock.avg 71441 ± 6% +81.9% 129935 ± 3% sched_debug.cpu.clock.max 71398 ± 7% +82.0% 129924 ± 3% sched_debug.cpu.clock.min 71428 ± 6% +81.9% 129931 ± 3% sched_debug.cpu.clock_task.avg 71441 ± 6% +81.9% 129935 ± 3% sched_debug.cpu.clock_task.max 71398 ± 7% +82.0% 129924 ± 3% sched_debug.cpu.clock_task.min 17.79 ± 9% +4314.3% 785.21 ± 62% sched_debug.cpu.cpu_load[0].avg 207.50 ± 22% +10351.3% 21686 ±120% sched_debug.cpu.cpu_load[0].max 42.02 ± 16% +7431.5% 3165 ±104% sched_debug.cpu.cpu_load[0].stddev 87.72 ± 6% +3905.5% 3513 ± 45% sched_debug.cpu.cpu_load[1].avg 261.00 ± 4% +48754.4% 127510 ± 73% sched_debug.cpu.cpu_load[1].max 78.13 ± 7% +22146.8% 17380 ± 69% sched_debug.cpu.cpu_load[1].stddev 83.88 ± 5% +3295.3% 2848 ± 34% sched_debug.cpu.cpu_load[2].avg 255.75 ± 10% +36120.3% 92633 ± 64% sched_debug.cpu.cpu_load[2].max 75.91 ± 9% +16771.2% 12807 ± 60% sched_debug.cpu.cpu_load[2].stddev 79.24 ± 6% +2565.6% 2112 ± 26% sched_debug.cpu.cpu_load[3].avg 246.88 ± 15% +24717.2% 61267 ± 52% sched_debug.cpu.cpu_load[3].max 72.85 ± 11% +11627.4% 8543 ± 49% sched_debug.cpu.cpu_load[3].stddev 75.02 ± 9% +1959.3% 1544 ± 22% sched_debug.cpu.cpu_load[4].avg 229.00 ± 11% +16544.1% 38114 ± 45% sched_debug.cpu.cpu_load[4].max 68.73 ± 12% +7759.6% 5401 ± 42% sched_debug.cpu.cpu_load[4].stddev 708.89 ± 8% +49.1% 1056 ± 0% sched_debug.cpu.curr->pid.avg 2307 ± 0% +57.8% 3642 ± 4% sched_debug.cpu.curr->pid.max 1009 ± 3% +41.6% 1428 ± 1% sched_debug.cpu.curr->pid.stddev 297811 ± 11% -97.8% 6577 ± 6% sched_debug.cpu.load.avg 1025387 ± 0% -96.3% 37468 ± 16% sched_debug.cpu.load.max 445802 ± 4% -97.8% 9997 ± 3% sched_debug.cpu.load.stddev 0.00 ± 30% -46.8% 0.00 ± 18% sched_debug.cpu.next_balance.stddev 19083 ± 3% +307.4% 77750 ± 0% sched_debug.cpu.nr_load_updates.avg 24124 ± 2% +300.4% 96586 ± 1% sched_debug.cpu.nr_load_updates.max 15854 ± 6% +265.1% 57892 ± 1% sched_debug.cpu.nr_load_updates.min 2205 ± 26% +586.6% 15141 ± 0% sched_debug.cpu.nr_load_updates.stddev 2.75 ± 15% -34.1% 1.81 ± 11% sched_debug.cpu.nr_running.max 0.67 ± 3% -20.6% 0.53 ± 6% sched_debug.cpu.nr_running.stddev 383955 ± 0% +328.5% 1645106 ± 0% sched_debug.cpu.nr_switches.avg 468859 ± 3% +368.6% 2197053 ± 0% sched_debug.cpu.nr_switches.max 310050 ± 4% +236.1% 1041992 ± 1% sched_debug.cpu.nr_switches.min 57708 ± 8% +782.2% 509119 ± 0% sched_debug.cpu.nr_switches.stddev 0.24 ± 8% +101.4% 0.49 ± 12% sched_debug.cpu.nr_uninterruptible.avg 1057 ± 8% -22.5% 819.00 ± 13% sched_debug.cpu.nr_uninterruptible.max -1123 ± -3% -29.1% -796.88 ± -2% sched_debug.cpu.nr_uninterruptible.min 735.97 ± 5% -10.7% 657.49 ± 2% sched_debug.cpu.nr_uninterruptible.stddev 383743 ± 0% +328.8% 1645352 ± 0% sched_debug.cpu.sched_count.avg 489071 ± 3% +354.4% 2222429 ± 0% sched_debug.cpu.sched_count.max 309779 ± 4% +236.3% 1041674 ± 1% sched_debug.cpu.sched_count.min 57799 ± 8% +781.7% 509642 ± 0% sched_debug.cpu.sched_count.stddev 160215 ± 0% +392.3% 788761 ± 0% sched_debug.cpu.sched_goidle.avg 185615 ± 3% +453.2% 1026840 ± 0% sched_debug.cpu.sched_goidle.max 137784 ± 4% +277.4% 519987 ± 1% sched_debug.cpu.sched_goidle.min 16111 ± 19% +1276.8% 221816 ± 0% sched_debug.cpu.sched_goidle.stddev 222670 ± 0% +284.0% 855043 ± 0% sched_debug.cpu.ttwu_count.avg 280679 ± 3% +317.7% 1172374 ± 0% sched_debug.cpu.ttwu_count.max 170029 ± 4% +204.5% 517707 ± 1% sched_debug.cpu.ttwu_count.min 43065 ± 4% +569.8% 288430 ± 0% sched_debug.cpu.ttwu_count.stddev 14304 ± 0% -96.3% 532.37 ± 5% sched_debug.cpu.ttwu_local.avg 21467 ± 4% -95.6% 950.25 ± 16% sched_debug.cpu.ttwu_local.max 7884 ± 3% -95.6% 350.69 ± 8% sched_debug.cpu.ttwu_local.min 5851 ± 1% -97.8% 128.39 ± 16% sched_debug.cpu.ttwu_local.stddev 71435 ± 6% +81.9% 129928 ± 3% sched_debug.cpu_clk 68712 ± 7% +85.1% 127177 ± 3% sched_debug.ktime 71435 ± 6% +81.9% 129928 ± 3% sched_debug.sched_clk vm-scalability.throughput 4e+07 ++----------------------------------------------------------------+ | *.****.** ***.****. *.* *.*** | ***.****.* * * .* *** ** | 3.5e+07 ***.****.****.****.* * * | | | | | 3e+07 ++ | | | 2.5e+07 ++ | | | | | 2e+07 ++ | | OOOO OOOO OOOO | OOO OOOO OOOO | 1.5e+07 ++----------------------------------------------------------------+ [*] 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, Xiaolong