Greeting, FYI, we noticed a -15.5% regression of vm-scalability.throughput due to commit: commit: 85766f621c492a9aa7904050ad2f80893ab2a8fd ("shmem: Convert shmem_add_to_page_cache to XArray") https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git master in testcase: vm-scalability on test machine: 4 threads Intel(R) Core(TM) i3-3220 CPU @ 3.30GHz with 8G memory with following parameters: runtime: 300s size: 16G test: shm-xread-rand cpufreq_governor: performance test-description: The motivation behind this suite is to exercise functions and regions of the mm/ of the Linux kernel which are of interest to us. test-url: https://git.kernel.org/cgit/linux/kernel/git/wfg/vm-scalability.git/ 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/rootfs/runtime/size/tbox_group/test/testcase: gcc-7/performance/x86_64-rhel-7.2/debian-x86_64-2018-04-03.cgz/300s/16G/lkp-ivb-d02/shm-xread-rand/vm-scalability commit: 85b67aaaa5 ("shmem: Convert find_swap_entry to XArray") 85766f621c ("shmem: Convert shmem_add_to_page_cache to XArray") 85b67aaaa59cd76a 85766f621c492a9aa7904050ad ---------------- -------------------------- %stddev %change %stddev \ | \ 190175 -15.5% 160733 ± 21% vm-scalability.throughput 47700 +99.1% 94958 ± 3% vm-scalability.median 0.03 ± 15% +38.6% 0.05 ± 81% vm-scalability.stddev 96.91 -47.6% 50.75 ± 5% vm-scalability.time.elapsed_time 96.91 -47.6% 50.75 ± 5% vm-scalability.time.elapsed_time.max 878016 ± 2% -98.3% 14720 ± 53% vm-scalability.time.involuntary_context_switches 4024745 -0.0% 4024725 vm-scalability.time.maximum_resident_set_size 1497244 -14.3% 1283635 ± 5% vm-scalability.time.minor_page_faults 4096 +0.0% 4096 vm-scalability.time.page_size 375.00 -54.1% 172.25 ± 22% vm-scalability.time.percent_of_cpu_this_job_got 11.18 ± 3% -53.6% 5.19 ± 5% vm-scalability.time.system_time 352.62 -76.3% 83.60 ± 27% vm-scalability.time.user_time 18105273 -56.2% 7921056 ± 24% vm-scalability.workload 2677 ± 4% -42.1% 1549 ± 9% interrupts.CAL:Function_call_interrupts 74956 ± 11% -49.1% 38140 ± 19% softirqs.RCU 10283 ± 4% +166.8% 27439 ± 11% softirqs.SCHED 196048 -43.5% 110676 ± 6% softirqs.TIMER 79.95 -0.2% 79.79 ± 2% boot-time.boot 70.20 ± 2% -0.0% 70.17 ± 2% boot-time.dhcp 299.76 ± 2% -0.2% 299.06 ± 2% boot-time.idle 72.80 ± 2% -0.2% 72.62 ± 2% boot-time.kernel_boot 2.80 ± 6% +46.8 49.56 ± 20% mpstat.cpu.idle% 0.03 ± 34% -0.0 0.02 ± 28% mpstat.cpu.soft% 4.65 ± 4% +0.2 4.83 ± 34% mpstat.cpu.sys% 92.52 -46.9 45.59 ± 20% mpstat.cpu.usr% 1029 -3.3% 995.00 vmstat.memory.buff 4825193 -3.4% 4663456 vmstat.memory.cache 2875885 +6.4% 3059946 vmstat.memory.free 0.00 -100.0% 0.00 vmstat.procs.b 4.00 -50.0% 2.00 ± 35% vmstat.procs.r 20841 +49.8% 31217 vmstat.system.cs 13121 +37.1% 17983 vmstat.system.in 43219 ± 10% +755.6% 369768 ± 61% cpuidle.C1.time 1375 ± 15% +1197.9% 17855 ± 68% cpuidle.C1.usage 1256903 ± 68% +3335.5% 43180596 ± 3% cpuidle.C1E.time 21308 ± 69% +3424.9% 751082 ± 4% cpuidle.C1E.usage 39665 ± 25% +351.9% 179230 ± 40% cpuidle.C3.time 160.75 ± 18% +311.8% 662.00 ± 32% cpuidle.C3.usage 7135782 ± 18% +666.0% 54661249 ± 27% cpuidle.C6.time 7706 ± 17% +668.0% 59186 ± 23% cpuidle.C6.usage 301.25 ± 39% +1868.8% 5931 ± 37% cpuidle.POLL.time 29.00 ± 44% +3090.5% 925.25 ± 49% cpuidle.POLL.usage 2510 ± 20% +2.6% 2576 ± 16% slabinfo.anon_vma.active_objs 2607 ± 13% +0.7% 2627 ± 15% slabinfo.anon_vma.num_objs 1193 ± 10% +2.9% 1228 ± 9% slabinfo.cred_jar.active_objs 1193 ± 10% +2.9% 1228 ± 9% slabinfo.cred_jar.num_objs 107.75 ± 13% +107.2% 223.25 ± 54% slabinfo.dmaengine-unmap-16.active_objs 115.25 ± 19% +111.9% 244.25 ± 61% slabinfo.dmaengine-unmap-16.num_objs 7200 ± 12% +8.9% 7840 ± 6% slabinfo.kmalloc-32.active_objs 7200 ± 12% +8.9% 7840 ± 6% slabinfo.kmalloc-32.num_objs 5248 ± 8% -0.2% 5238 ± 4% slabinfo.kmalloc-8.active_objs 5248 ± 8% -0.2% 5238 ± 4% slabinfo.kmalloc-8.num_objs 3455 ± 5% +19.2% 4119 ± 4% slabinfo.pid.active_objs 4570 ± 7% +0.9% 4613 ± 5% slabinfo.pid.num_objs 24874 -1.4% 24521 slabinfo.radix_tree_node.active_objs 24874 -1.3% 24563 slabinfo.radix_tree_node.num_objs 3221 -46.7% 1717 ± 18% turbostat.Avg_MHz 97.84 -45.7 52.18 ± 18% turbostat.Busy% 3300 -0.0% 3299 turbostat.Bzy_MHz 1375 ± 15% +1198.2% 17856 ± 68% turbostat.C1 0.01 +0.2 0.18 ± 69% turbostat.C1% 21308 ± 69% +3425.0% 751108 ± 4% turbostat.C1E 0.32 ± 69% +20.5 20.78 ± 2% turbostat.C1E% 161.25 ± 17% +310.7% 662.25 ± 32% turbostat.C3 0.01 +0.1 0.09 ± 36% turbostat.C3% 7713 ± 17% +667.3% 59186 ± 23% turbostat.C6 1.82 ± 18% +24.9 26.73 ± 33% turbostat.C6% 1.38 ± 22% +3185.0% 45.33 ± 14% turbostat.CPU%c1 0.01 ± 34% -60.0% 0.01 ±100% turbostat.CPU%c3 0.77 ± 44% +220.4% 2.48 ±126% turbostat.CPU%c6 13.59 -3.8% 13.08 ± 6% turbostat.CorWatt 49.50 -3.5% 47.75 ± 2% turbostat.CoreTmp 350.00 +0.0% 350.00 turbostat.GFXMHz 1305924 -27.1% 952039 ± 4% turbostat.IRQ 0.04 ± 30% -100.0% 0.00 turbostat.Pkg%pc2 0.12 ± 10% -100.0% 0.00 turbostat.Pkg%pc6 47.00 ± 4% -1.6% 46.25 ± 2% turbostat.PkgTmp 30.99 -1.9% 30.41 ± 3% turbostat.PkgWatt 3292 -0.0% 3292 turbostat.TSC_MHz 5.878e+10 ± 3% -51.3% 2.861e+10 ± 53% perf-stat.branch-instructions 1.64 ± 14% +0.8 2.45 ± 86% perf-stat.branch-miss-rate% 9.698e+08 ± 18% -49.4% 4.904e+08 ±105% perf-stat.branch-misses 91.01 -18.6 72.41 ± 26% perf-stat.cache-miss-rate% 4.227e+09 -55.8% 1.868e+09 ± 87% perf-stat.cache-misses 4.645e+09 -53.7% 2.148e+09 ± 78% perf-stat.cache-references 2026235 -19.2% 1636525 ± 4% perf-stat.context-switches 4.83 ± 4% -50.7% 2.39 ± 32% perf-stat.cpi 1.255e+12 -72.6% 3.435e+11 ± 70% perf-stat.cpu-cycles 13403 ± 17% -85.7% 1914 ± 5% perf-stat.cpu-migrations 5.39 ± 3% -1.8 3.57 ± 58% perf-stat.dTLB-load-miss-rate% 3.877e+09 -57.0% 1.668e+09 ± 88% perf-stat.dTLB-load-misses 6.807e+10 ± 2% -50.5% 3.37e+10 ± 57% perf-stat.dTLB-loads 0.02 ± 20% +0.0 0.07 ± 69% perf-stat.dTLB-store-miss-rate% 6310413 ± 22% -6.8% 5881564 ± 31% perf-stat.dTLB-store-misses 2.53e+10 ± 2% -47.4% 1.331e+10 ± 49% perf-stat.dTLB-stores 89.37 -18.3 71.07 ± 14% perf-stat.iTLB-load-miss-rate% 10852282 ± 31% +98.0% 21484353 ± 50% perf-stat.iTLB-load-misses 1333714 ± 48% +446.8% 7292385 ± 6% perf-stat.iTLB-loads 2.601e+11 ± 3% -50.8% 1.278e+11 ± 51% perf-stat.instructions 25902 ± 23% -66.0% 8814 ± 95% perf-stat.instructions-per-iTLB-miss 0.21 ± 4% +126.5% 0.47 ± 33% perf-stat.ipc 1626010 -16.2% 1362182 ± 5% perf-stat.minor-faults 1626011 -16.2% 1362184 ± 5% perf-stat.page-faults 14363 ± 3% +4.0% 14938 ± 40% perf-stat.path-length 306033 ± 7% +501.9% 1842050 ± 51% meminfo.Active 305962 ± 7% +502.0% 1841980 ± 51% meminfo.Active(anon) 74019 -48.2% 38339 ± 5% meminfo.AnonHugePages 235389 -1.5% 231743 meminfo.AnonPages 1041 -2.2% 1017 meminfo.Buffers 4827393 -2.3% 4714606 meminfo.Cached 18164 ± 8% +32.3% 24031 ± 7% meminfo.CmaFree 204800 +0.0% 204800 meminfo.CmaTotal 4023392 +0.0% 4023392 meminfo.CommitLimit 4265286 -2.5% 4158052 meminfo.Committed_AS 8191029 +0.1% 8195721 meminfo.DirectMap2M 79997 ± 5% -5.9% 75305 ± 2% meminfo.DirectMap4k 2048 +0.0% 2048 meminfo.Hugepagesize 3877131 -42.6% 2224654 ± 41% meminfo.Inactive 3875926 -42.6% 2223471 ± 41% meminfo.Inactive(anon) 1205 -1.8% 1183 meminfo.Inactive(file) 3603 -1.7% 3541 meminfo.KernelStack 3932052 -2.8% 3822786 meminfo.Mapped 2715637 +5.0% 2850785 meminfo.MemAvailable 2824985 +4.8% 2960247 meminfo.MemFree 8046788 +0.0% 8046788 meminfo.MemTotal 34324 -50.5% 16985 ± 18% meminfo.PageTables 48267 -0.4% 48065 meminfo.SReclaimable 20732 -0.7% 20579 meminfo.SUnreclaim 3946053 -2.9% 3833269 meminfo.Shmem 69000 -0.5% 68645 meminfo.Slab 881118 -0.0% 881102 meminfo.Unevictable 3.436e+10 +0.0% 3.436e+10 meminfo.VmallocTotal 76751 ± 4% +500.6% 460938 ± 51% proc-vmstat.nr_active_anon 58905 -1.5% 58008 proc-vmstat.nr_anon_pages 67403 +4.7% 70593 proc-vmstat.nr_dirty_background_threshold 134972 +4.7% 141360 proc-vmstat.nr_dirty_threshold 1205133 -2.2% 1178739 proc-vmstat.nr_file_pages 4613 ± 4% +29.8% 5989 ± 7% proc-vmstat.nr_free_cma 708248 +4.5% 740205 proc-vmstat.nr_free_pages 966805 -42.6% 555343 ± 41% proc-vmstat.nr_inactive_anon 300.50 -2.0% 294.50 proc-vmstat.nr_inactive_file 3601 -1.7% 3542 proc-vmstat.nr_kernel_stack 978954 -2.9% 950733 proc-vmstat.nr_mapped 8544 -50.4% 4237 ± 18% proc-vmstat.nr_page_table_pages 984538 -2.7% 958151 proc-vmstat.nr_shmem 12063 -0.5% 12004 proc-vmstat.nr_slab_reclaimable 5182 -0.7% 5144 proc-vmstat.nr_slab_unreclaimable 220278 -0.0% 220275 proc-vmstat.nr_unevictable 76754 ± 4% +500.5% 460940 ± 51% proc-vmstat.nr_zone_active_anon 966805 -42.6% 555343 ± 41% proc-vmstat.nr_zone_inactive_anon 300.50 -2.0% 294.50 proc-vmstat.nr_zone_inactive_file 220278 -0.0% 220275 proc-vmstat.nr_zone_unevictable 1162127 -3.6% 1120485 proc-vmstat.numa_hit 1162127 -3.6% 1120485 proc-vmstat.numa_local 1006996 -5.2% 954325 ± 5% proc-vmstat.pgactivate 247442 -10.8% 220800 proc-vmstat.pgalloc_dma32 936515 -3.6% 903231 proc-vmstat.pgalloc_normal 1630736 -16.2% 1366671 ± 5% proc-vmstat.pgfault 595443 ± 62% +1.1% 601843 ± 83% proc-vmstat.pgfree 2173 +0.0% 2173 proc-vmstat.pgpgin 2048 +0.0% 2049 proc-vmstat.pgpgout 168.36 ±173% +248.0% 585.83 ±173% sched_debug.cfs_rq:/.MIN_vruntime.avg 673.46 ±173% +248.0% 2343 ±173% sched_debug.cfs_rq:/.MIN_vruntime.max 0.00 +0.0% 0.00 sched_debug.cfs_rq:/.MIN_vruntime.min 291.62 ±173% +248.0% 1014 ±173% sched_debug.cfs_rq:/.MIN_vruntime.stddev 29757 -99.9% 39.23 ± 25% sched_debug.cfs_rq:/.exec_clock.avg 30069 -99.9% 40.37 ± 24% sched_debug.cfs_rq:/.exec_clock.max 29419 -99.9% 37.80 ± 26% sched_debug.cfs_rq:/.exec_clock.min 248.80 ± 68% -99.6% 1.02 ± 13% sched_debug.cfs_rq:/.exec_clock.stddev 292413 ± 19% +13.2% 331037 ± 31% sched_debug.cfs_rq:/.load.avg 512518 ± 46% +11.9% 573270 ± 69% sched_debug.cfs_rq:/.load.max 187469 ± 7% +9.9% 205951 ± 11% sched_debug.cfs_rq:/.load.min 133934 ± 76% +10.8% 148390 ±112% sched_debug.cfs_rq:/.load.stddev 406.56 ± 15% +46.8% 596.94 ± 7% sched_debug.cfs_rq:/.load_avg.avg 634.38 ± 21% +85.7% 1178 ± 28% sched_debug.cfs_rq:/.load_avg.max 240.25 ± 11% +10.7% 266.00 ± 18% sched_debug.cfs_rq:/.load_avg.min 167.36 ± 42% +130.5% 385.79 ± 37% sched_debug.cfs_rq:/.load_avg.stddev 168.36 ±173% +248.0% 585.83 ±173% sched_debug.cfs_rq:/.max_vruntime.avg 673.46 ±173% +248.0% 2343 ±173% sched_debug.cfs_rq:/.max_vruntime.max 0.00 +0.0% 0.00 sched_debug.cfs_rq:/.max_vruntime.min 291.62 ±173% +248.0% 1014 ±173% sched_debug.cfs_rq:/.max_vruntime.stddev 131544 -93.1% 9126 sched_debug.cfs_rq:/.min_vruntime.avg 140526 ± 2% -89.2% 15201 ± 15% sched_debug.cfs_rq:/.min_vruntime.max 121759 ± 3% -97.1% 3486 ± 22% sched_debug.cfs_rq:/.min_vruntime.min 7074 ± 31% -36.0% 4528 ± 20% sched_debug.cfs_rq:/.min_vruntime.stddev 1.03 ± 5% +3.0% 1.06 ± 10% sched_debug.cfs_rq:/.nr_running.avg 1.12 ± 19% +11.1% 1.25 ± 34% sched_debug.cfs_rq:/.nr_running.max 1.00 +0.0% 1.00 sched_debug.cfs_rq:/.nr_running.min 0.05 ±173% +100.0% 0.11 ±173% sched_debug.cfs_rq:/.nr_running.stddev 3.41 ± 7% -100.0% 0.00 sched_debug.cfs_rq:/.nr_spread_over.avg 6.88 ± 13% -100.0% 0.00 sched_debug.cfs_rq:/.nr_spread_over.max 1.12 ± 72% -100.0% 0.00 sched_debug.cfs_rq:/.nr_spread_over.min 2.16 ± 27% -100.0% 0.00 sched_debug.cfs_rq:/.nr_spread_over.stddev 178.72 ± 3% -53.9% 82.38 ± 26% sched_debug.cfs_rq:/.runnable_load_avg.avg 274.75 ± 4% -50.9% 135.00 ± 38% sched_debug.cfs_rq:/.runnable_load_avg.max 141.12 ± 4% -70.8% 41.25 ± 44% sched_debug.cfs_rq:/.runnable_load_avg.min 56.87 ± 7% -35.1% 36.91 ± 66% sched_debug.cfs_rq:/.runnable_load_avg.stddev 212710 ± 24% -63.1% 78556 ± 28% sched_debug.cfs_rq:/.runnable_weight.avg 400157 ± 53% -69.9% 120568 ± 40% sched_debug.cfs_rq:/.runnable_weight.max 141898 ± 2% -70.5% 41804 ± 44% sched_debug.cfs_rq:/.runnable_weight.min 109826 ± 83% -72.0% 30704 ± 76% sched_debug.cfs_rq:/.runnable_weight.stddev -956.92 -348.7% 2379 ±119% sched_debug.cfs_rq:/.spread0.avg 8024 ± 76% +5.3% 8451 ± 28% sched_debug.cfs_rq:/.spread0.max -10742 -69.7% -3260 sched_debug.cfs_rq:/.spread0.min 7074 ± 31% -36.0% 4527 ± 20% sched_debug.cfs_rq:/.spread0.stddev 1028 ± 2% +0.5% 1033 ± 3% sched_debug.cfs_rq:/.util_avg.avg 1105 ± 5% +14.1% 1261 ± 6% sched_debug.cfs_rq:/.util_avg.max 929.62 ± 5% -8.2% 853.00 ± 14% sched_debug.cfs_rq:/.util_avg.min 67.44 ± 63% +130.4% 155.36 ± 41% sched_debug.cfs_rq:/.util_avg.stddev 222.56 ± 34% -71.7% 62.88 ± 58% sched_debug.cfs_rq:/.util_est_enqueued.avg 503.88 ± 33% -51.8% 242.75 ± 59% sched_debug.cfs_rq:/.util_est_enqueued.max 26.00 ±162% -95.2% 1.25 ± 34% sched_debug.cfs_rq:/.util_est_enqueued.min 195.39 ± 35% -46.8% 103.88 ± 60% sched_debug.cfs_rq:/.util_est_enqueued.stddev 473454 ± 14% +38.0% 653593 ± 9% sched_debug.cpu.avg_idle.avg 832956 ± 6% -3.1% 806762 ± 9% sched_debug.cpu.avg_idle.max 193874 ± 57% +124.3% 434865 ± 36% sched_debug.cpu.avg_idle.min 260441 ± 13% -45.2% 142845 ± 55% sched_debug.cpu.avg_idle.stddev 110437 -27.4% 80213 ± 2% sched_debug.cpu.clock.avg 110438 -27.4% 80214 ± 2% sched_debug.cpu.clock.max 110436 -27.4% 80211 ± 2% sched_debug.cpu.clock.min 0.83 ± 13% +25.6% 1.04 ± 12% sched_debug.cpu.clock.stddev 110437 -27.4% 80213 ± 2% sched_debug.cpu.clock_task.avg 110438 -27.4% 80214 ± 2% sched_debug.cpu.clock_task.max 110436 -27.4% 80211 ± 2% sched_debug.cpu.clock_task.min 0.83 ± 13% +25.6% 1.04 ± 12% sched_debug.cpu.clock_task.stddev 189.84 ± 4% -54.7% 85.94 ± 17% sched_debug.cpu.cpu_load[0].avg 289.12 ± 10% -44.1% 161.50 ± 23% sched_debug.cpu.cpu_load[0].max 141.12 ± 4% -71.1% 40.75 ± 17% sched_debug.cpu.cpu_load[0].min 63.08 ± 20% -21.9% 49.24 ± 30% sched_debug.cpu.cpu_load[0].stddev 185.59 ± 4% -52.2% 88.69 ± 12% sched_debug.cpu.cpu_load[1].avg 282.50 ± 6% -47.1% 149.50 ± 29% sched_debug.cpu.cpu_load[1].max 137.25 ± 4% -67.8% 44.25 ± 18% sched_debug.cpu.cpu_load[1].min 60.14 ± 14% -28.4% 43.04 ± 46% sched_debug.cpu.cpu_load[1].stddev 184.72 ± 3% -50.5% 91.38 ± 12% sched_debug.cpu.cpu_load[2].avg 275.38 ± 5% -47.7% 144.00 ± 30% sched_debug.cpu.cpu_load[2].max 138.75 ± 6% -62.3% 52.25 ± 24% sched_debug.cpu.cpu_load[2].min 55.48 ± 11% -30.6% 38.47 ± 50% sched_debug.cpu.cpu_load[2].stddev 186.38 ± 4% -47.5% 97.94 ± 14% sched_debug.cpu.cpu_load[3].avg 258.75 ± 4% -44.8% 142.75 ± 27% sched_debug.cpu.cpu_load[3].max 139.88 ± 6% -57.5% 59.50 ± 25% sched_debug.cpu.cpu_load[3].min 45.63 ± 13% -26.4% 33.58 ± 46% sched_debug.cpu.cpu_load[3].stddev 188.03 ± 5% -42.5% 108.06 ± 13% sched_debug.cpu.cpu_load[4].avg 240.62 ± 5% -40.2% 144.00 ± 23% sched_debug.cpu.cpu_load[4].max 142.50 ± 9% -51.9% 68.50 ± 18% sched_debug.cpu.cpu_load[4].min 36.69 ± 22% -21.2% 28.92 ± 43% sched_debug.cpu.cpu_load[4].stddev 966.06 ± 6% -25.7% 718.06 ± 9% sched_debug.cpu.curr->pid.avg 1299 -33.2% 868.25 sched_debug.cpu.curr->pid.max 705.75 ± 34% -40.1% 422.75 ± 37% sched_debug.cpu.curr->pid.min 237.21 ± 31% -24.8% 178.41 ± 35% sched_debug.cpu.curr->pid.stddev 325755 ± 34% -18.5% 265488 ± 5% sched_debug.cpu.load.avg 617677 ± 67% -44.1% 345304 ± 6% sched_debug.cpu.load.max 187469 ± 7% +9.9% 205951 ± 11% sched_debug.cpu.load.min 176372 ±100% -69.5% 53831 ± 27% sched_debug.cpu.load.stddev 500000 +0.0% 500000 sched_debug.cpu.max_idle_balance_cost.avg 500000 +0.0% 500000 sched_debug.cpu.max_idle_balance_cost.max 500000 +0.0% 500000 sched_debug.cpu.max_idle_balance_cost.min 4294 -0.0% 4294 sched_debug.cpu.next_balance.avg 4294 -0.0% 4294 sched_debug.cpu.next_balance.max 4294 -0.0% 4294 sched_debug.cpu.next_balance.min 0.00 ± 36% -10.2% 0.00 ± 65% sched_debug.cpu.next_balance.stddev 37966 -78.9% 8004 ± 4% sched_debug.cpu.nr_load_updates.avg 40554 ± 2% -70.4% 11997 ± 21% sched_debug.cpu.nr_load_updates.max 34281 ± 2% -87.1% 4427 ± 15% sched_debug.cpu.nr_load_updates.min 2438 ± 22% +19.2% 2905 ± 41% sched_debug.cpu.nr_load_updates.stddev 2.25 ± 17% +8.3% 2.44 ± 29% sched_debug.cpu.nr_running.avg 3.12 ± 20% +20.0% 3.75 ± 54% sched_debug.cpu.nr_running.max 1.62 ± 13% -23.1% 1.25 ± 34% sched_debug.cpu.nr_running.min 0.63 ± 22% +59.9% 1.00 ± 85% sched_debug.cpu.nr_running.stddev 254977 -96.8% 8201 ± 4% sched_debug.cpu.nr_switches.avg 674868 ± 21% -98.2% 11993 ± 22% sched_debug.cpu.nr_switches.max 37283 ± 51% -85.3% 5491 ± 17% sched_debug.cpu.nr_switches.min 258652 ± 28% -99.0% 2585 ± 58% sched_debug.cpu.nr_switches.stddev 0.03 ±173% +700.0% 0.25 ± 70% sched_debug.cpu.nr_uninterruptible.avg 22.25 ± 22% -34.8% 14.50 ± 18% sched_debug.cpu.nr_uninterruptible.max -20.75 -45.8% -11.25 sched_debug.cpu.nr_uninterruptible.min 16.52 ± 24% -38.8% 10.12 ± 14% sched_debug.cpu.nr_uninterruptible.stddev 247705 -99.6% 886.94 ± 51% sched_debug.cpu.sched_count.avg 667148 ± 21% -99.6% 2595 ± 62% sched_debug.cpu.sched_count.max 30369 ± 59% -99.5% 143.75 ± 7% sched_debug.cpu.sched_count.min 258206 ± 27% -99.6% 1005 ± 66% sched_debug.cpu.sched_count.stddev 2815 ± 66% -100.0% 0.00 sched_debug.cpu.sched_goidle.avg 8181 ± 73% -100.0% 0.00 sched_debug.cpu.sched_goidle.max 29.38 ±173% -100.0% 0.00 sched_debug.cpu.sched_goidle.min 3344 ± 71% -100.0% 0.00 sched_debug.cpu.sched_goidle.stddev 123801 -99.6% 440.69 ± 52% sched_debug.cpu.ttwu_count.avg 334777 ± 21% -99.6% 1307 ± 61% sched_debug.cpu.ttwu_count.max 14411 ± 67% -99.4% 81.25 ± 9% sched_debug.cpu.ttwu_count.min 129727 ± 27% -99.6% 507.89 ± 64% sched_debug.cpu.ttwu_count.stddev 122356 -99.7% 411.38 ± 55% sched_debug.cpu.ttwu_local.avg 332928 ± 21% -99.6% 1279 ± 62% sched_debug.cpu.ttwu_local.max 12947 ± 71% -99.6% 50.50 ± 8% sched_debug.cpu.ttwu_local.min 129617 ± 27% -99.6% 508.69 ± 64% sched_debug.cpu.ttwu_local.stddev 110435 -27.4% 80211 ± 2% sched_debug.cpu_clk 996147 +0.0% 996147 sched_debug.dl_rq:.dl_bw->bw.avg 996147 +0.0% 996147 sched_debug.dl_rq:.dl_bw->bw.max 996147 +0.0% 996147 sched_debug.dl_rq:.dl_bw->bw.min 4.295e+09 -0.0% 4.295e+09 sched_debug.jiffies 110435 -27.4% 80211 ± 2% sched_debug.ktime 950.00 +0.0% 950.00 sched_debug.rt_rq:/.rt_runtime.avg 950.00 +0.0% 950.00 sched_debug.rt_rq:/.rt_runtime.max 950.00 +0.0% 950.00 sched_debug.rt_rq:/.rt_runtime.min 110441 -27.4% 80216 ± 2% sched_debug.sched_clk 1.00 +0.0% 1.00 sched_debug.sched_clock_stable() 4118331 +0.0% 4118331 sched_debug.sysctl_sched.sysctl_sched_features 18.00 +0.0% 18.00 sched_debug.sysctl_sched.sysctl_sched_latency 2.25 +0.0% 2.25 sched_debug.sysctl_sched.sysctl_sched_min_granularity 1.00 +0.0% 1.00 sched_debug.sysctl_sched.sysctl_sched_tunable_scaling 3.00 +0.0% 3.00 sched_debug.sysctl_sched.sysctl_sched_wakeup_granularity vm-scalability.workload 2e+07 +-+---------------------------------------------------------------+ | | 1.8e+07 +-++.+.++.+.++.+.++.++.+.++.+.++.+.++.+.++.+.++.+.++.++.+.++.+.++.| 1.6e+07 +-+ | | | 1.4e+07 +-+ | O O O O O | 1.2e+07 +-+ | | | 1e+07 +-+ | 8e+06 +-OO O O O OO | | | 6e+06 +-+ | | | 4e+06 +-+----OO----O---O---O--------------------------------------------+ vm-scalability.time.user_time 400 +-+-------------------------------------------------------------------+ | | 350 +-+.++.+.+.++.+.+.++.+.+.++.+.+.++.+.+.++.+.+.++.+.+.++.+.+.++.+.+.++.| 300 +-+ | | | 250 +-+ | | | 200 O-+ O O O O | | | 150 +-+ | 100 +-O | | O O O OO O | 50 +-+ O O O O O | | | 0 +-+-------------------------------------------------------------------+ vm-scalability.time.system_time 13 +-+--------------------------------------------------------------------+ |:: :: : .+ | 12 +-+: + + :: + : : + + + + .| 11 +-+: :: .+ .++. .+.+ : + .+ :: : + : : : : +. .+ + + .+ | | : : .+ +. .+.+ + :: + :: : + :: :: + +. + + | 10 +-+ + + + + + + + + + | 9 +-+ | | | 8 +-+ | 7 +-+ | O O | 6 +-O OO OO O O | 5 +-+ O O O O O O | | O O | 4 +-+--------------------------------------------------------------------+ vm-scalability.time.percent_of_cpu_this_job_got 400 +-+-------------------------------------------------------------------+ |.+.++.+.+.++.+.+.++.+.+.++.+.+.++.+.+.++.+.+.++.+.+.++.+.+.++.+.+.++.| 350 +-+ | | | | | 300 +-+ | O O O O O | 250 +-+ | | | 200 +-O O O O | | O O O | | | 150 +-+ | | | 100 +-+----O-O----O---O----O----------------------------------------------+ vm-scalability.time.involuntary_context_switches 1e+06 +-+----------------------------------------------------------------+ 900000 +-++.+.++.+. +.+.+ .+.+.++.+.++.+.++.+.+ + .++.| | ++.+.++.+.+ + +. + +.+.+ .+ | 800000 +-+ + + | 700000 +-+ | | | 600000 +-+ | 500000 +-+ | 400000 +-+ | | | 300000 +-+ | 200000 +-+ | O O | 100000 +-+ O | 0 +-OO---OO-O-OO-O-O--O-OO-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, Xiaolong