Greeting, FYI, we noticed a -11.9% regression of aim9.exec_test.ops_per_sec due to commit: commit: a97056a6fab541e1661fed9ced0f793bda34b717 ("cpuidle: poll_state: Add time limit to poll_idle()") https://git.kernel.org/cgit/linux/kernel/git/rafael/linux-pm.git poll-idle in testcase: aim9 on test machine: 72 threads Intel(R) Xeon(R) CPU E5-2699 v3 @ 2.30GHz with 128G memory with following parameters: testtime: 5s test: all cpufreq_governor: performance test-description: Suite IX is the "AIM Independent Resource Benchmark:" the famous synthetic benchmark. test-url: https://sourceforge.net/projects/aimbench/files/aim-suite9/ 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/tbox_group/test/testcase/testtime: gcc-7/performance/x86_64-rhel-7.2/debian-x86_64-2016-08-31.cgz/lkp-hsw-ep4/all/aim9/5s commit: v4.16-rc5 a97056a6fa ("cpuidle: poll_state: Add time limit to poll_idle()") v4.16-rc5 a97056a6fab541e1661fed9ced ---------------- -------------------------- %stddev %change %stddev \ | \ 2010 -11.9% 1770 aim9.exec_test.ops_per_sec 619320 ± 2% +2.6% 635500 aim9.creat-clo.ops_per_sec 1335376 +1.3% 1353216 aim9.disk_wrt.ops_per_sec 4976 -6.9% 4633 aim9.fork_test.ops_per_sec 4.555e+08 +1.4% 4.62e+08 aim9.fun_cal1.ops_per_sec 2.498e+08 ± 5% +6.8% 2.668e+08 aim9.fun_cal15.ops_per_sec 48323640 ± 2% +2.0% 49295400 aim9.jmp_test.ops_per_sec 413494 +2.7% 424689 aim9.link_test.ops_per_sec 509182 +1.2% 515440 aim9.page_test.ops_per_sec 396.68 -12.2% 348.15 ± 9% aim9.shell_rtns_1.ops_per_sec 701520 ± 2% +2.6% 720100 aim9.signal_test.ops_per_sec 619952 +1.6% 630083 aim9.sync_disk_rw.ops_per_sec 9496 ± 9% -29.1% 6730 ± 13% aim9.time.involuntary_context_switches 6666695 -5.1% 6328538 ± 3% aim9.time.minor_page_faults 117287 -8.4% 107404 ± 3% aim9.time.voluntary_context_switches 5573 ± 62% +108.4% 11614 ± 34% numa-numastat.node0.other_node 2172 -5.4% 2054 vmstat.system.cs 49345778 ± 4% +307.8% 2.012e+08 ± 74% cpuidle.C3.time 4.561e+08 ± 9% -100.0% 123156 ± 62% cpuidle.POLL.time 106130 ±183% -97.4% 2761 ±124% numa-meminfo.node1.Inactive 106067 ±183% -97.5% 2612 ±131% numa-meminfo.node1.Inactive(anon) 132349 ±175% -95.0% 6681 ± 44% numa-meminfo.node1.Shmem 2573 ± 17% +25.6% 3232 ± 13% numa-vmstat.node0.nr_mapped 5961 ± 52% +98.4% 11828 ± 33% numa-vmstat.node0.numa_other 26515 ±183% -97.5% 651.75 ±132% numa-vmstat.node1.nr_inactive_anon 33087 ±175% -95.0% 1665 ± 44% numa-vmstat.node1.nr_shmem 26515 ±183% -97.5% 651.75 ±132% numa-vmstat.node1.nr_zone_inactive_anon 1349 ± 62% +119.0% 2955 ± 11% slabinfo.dmaengine-unmap-16.active_objs 1368 ± 62% +116.1% 2955 ± 11% slabinfo.dmaengine-unmap-16.num_objs 731.60 ± 8% +21.1% 886.25 ± 4% slabinfo.ip6_dst_cache.active_objs 731.60 ± 8% +21.1% 886.25 ± 4% slabinfo.ip6_dst_cache.num_objs 3184 ± 6% -13.4% 2758 ± 3% slabinfo.mm_struct.active_objs 3184 ± 6% -13.0% 2771 ± 3% slabinfo.mm_struct.num_objs 166.80 ± 3% -36.2% 106.50 turbostat.Avg_MHz 5.94 ± 2% -1.1 4.84 ± 5% turbostat.Busy% 2812 -21.2% 2216 ± 4% turbostat.Bzy_MHz 27.11 -10.2% 24.34 turbostat.CPU%c1 19.80 ± 9% +81.7% 35.98 turbostat.Pkg%pc2 0.04 ± 90% +250.0% 0.14 ± 29% turbostat.Pkg%pc6 117.44 ± 3% -15.9% 98.79 turbostat.PkgWatt 8.64 ± 5% -11.1% 7.68 ± 3% turbostat.RAMWatt 2.794e+11 ± 10% -23.1% 2.15e+11 ± 4% perf-stat.branch-instructions 1.46 ± 5% +0.4 1.87 ± 6% perf-stat.branch-miss-rate% 0.93 ± 5% -0.1 0.84 ± 2% perf-stat.cache-miss-rate% 90185384 ± 2% -6.9% 83996045 ± 2% perf-stat.cache-misses 653470 -5.7% 616530 perf-stat.context-switches 2.28 ± 10% -18.9% 1.85 ± 4% perf-stat.cpi 3.505e+12 ± 4% -35.2% 2.273e+12 ± 4% perf-stat.cpu-cycles 31896 -6.3% 29892 ± 2% perf-stat.cpu-migrations 0.12 ± 18% +0.0 0.17 ± 17% perf-stat.dTLB-load-miss-rate% 4.953e+11 ± 12% -29.4% 3.497e+11 ± 10% perf-stat.dTLB-loads 0.44 ± 9% +22.3% 0.54 ± 4% perf-stat.ipc 7355724 -4.6% 7019933 ± 3% perf-stat.minor-faults 7355724 -4.6% 7019932 ± 3% perf-stat.page-faults 2713 ± 6% +20.4% 3268 ± 5% sched_debug.cfs_rq:/.exec_clock.avg 73594 ± 6% +26.1% 92766 ± 10% sched_debug.cfs_rq:/.exec_clock.max 12795 ± 5% +21.7% 15577 ± 6% sched_debug.cfs_rq:/.exec_clock.stddev 18587 ± 10% +16.2% 21602 ± 7% sched_debug.cfs_rq:/.min_vruntime.avg 177486 ± 7% +21.7% 215979 ± 2% sched_debug.cfs_rq:/.min_vruntime.max 28721 ± 5% +18.6% 34065 ± 4% sched_debug.cfs_rq:/.min_vruntime.stddev 1.97 ± 33% -57.6% 0.83 sched_debug.cfs_rq:/.nr_spread_over.max 0.25 ± 62% -100.0% 0.00 sched_debug.cfs_rq:/.nr_spread_over.stddev 4.66 ± 35% -65.4% 1.61 ± 87% sched_debug.cfs_rq:/.removed.util_avg.avg 25.10 ± 25% -57.0% 10.79 ± 83% sched_debug.cfs_rq:/.removed.util_avg.stddev 166666 ± 8% +22.4% 203936 ± 4% sched_debug.cfs_rq:/.spread0.max 28721 ± 5% +18.6% 34066 ± 4% sched_debug.cfs_rq:/.spread0.stddev 649.62 ± 6% -9.4% 588.34 ± 4% sched_debug.cpu.curr->pid.avg 4637 ± 3% -9.0% 4220 ± 2% sched_debug.cpu.curr->pid.stddev 1954 ± 8% +13.5% 2218 ± 10% sched_debug.cpu.sched_goidle.stddev 518.80 ± 24% -35.4% 335.12 ± 36% sched_debug.cpu.ttwu_count.min aim9.exec_test.ops_per_sec 2500 +-+------------------------------------------------------------------+ | | | | 2000 +-++..+..+...+..+..+..+..+..+..+...+..+..+..+..+..+..+..+...+..+..+..| O O O O O O O O O O O O O O O O O O O O O O | | 1500 +-+ | | | 1000 +-+ | | | | | 500 +-+ | | | | | 0 +-+----------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