Greeting, We noticed a +19.4% improvement of unixbench.score due to commit: commit: d59ad49131bbfd57293d00a007383cc98aff2710 ("sched/fair: Remove check in idle_balance against migration_cost") url: https://github.com/0day-ci/linux/commits/Rohit-Jain/sched-fair-Remove-che... in testcase: unixbench on test machine: 8 threads Intel(R) Core(TM) i7 CPU 870 @ 2.93GHz with 6G memory with following parameters: runtime: 300s nr_task: 100% test: spawn test-description: UnixBench is the original BYTE UNIX benchmark suite aims to test performance of Unix-like system. test-url: https://github.com/kdlucas/byte-unixbench 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 ========================================================================================= tbox_group/testcase/rootfs/kconfig/compiler/runtime/nr_task/test: nhm-white/unixbench/debian-x86_64-2016-08-31.cgz/x86_64-rhel-7.2/gcc-7/300s/100%/spawn commit: v4.16-rc4 d59ad49131bbfd57293d00a007383cc98aff2710 v4.16-rc4 d59ad49131bbfd57293d00a007 ---------------- -------------------------- %stddev %change %stddev \ | \ 3202 +19.4% 3824 unixbench.score 4062446 ± 4% -12.2% 3565055 ± 3% unixbench.time.involuntary_context_switches 2.809e+08 +18.9% 3.34e+08 unixbench.time.minor_page_faults 611.56 +12.3% 686.50 unixbench.time.percent_of_cpu_this_job_got 862.38 +14.4% 986.68 unixbench.time.system_time 1170 +10.3% 1290 unixbench.time.user_time 23973299 +15.5% 27696636 unixbench.time.voluntary_context_switches 13386322 +19.2% 15962986 unixbench.workload 24295 +46.7% 35646 ± 10% interrupts.CAL:Function_call_interrupts 93923 ± 7% -25.3% 70144 ± 10% meminfo.DirectMap4k 42426 ± 3% +13.0% 47960 meminfo.Shmem 21.95 ± 5% -9.7 12.28 mpstat.cpu.idle% 71.10 +9.3 80.37 mpstat.cpu.sys% 2430631 +13.5% 2758915 softirqs.RCU 921699 +26.9% 1169816 softirqs.SCHED 355.45 -100.0% 0.00 uptime.boot 780.22 ± 6% -100.0% 0.00 uptime.idle 4.00 +9800.0% 396.00 vmstat.memory.buff 138810 -11.7% 122590 vmstat.system.cs 2.163e+08 -76.8% 50241754 cpuidle.C1.time 8231502 -68.3% 2607969 ± 2% cpuidle.C1.usage 67306881 -88.4% 7799355 ± 2% cpuidle.C1E.time 2343272 -86.1% 325846 ± 2% cpuidle.C1E.usage 277490 ± 9% -37.1% 174635 ± 2% cpuidle.C3.usage 4062446 ± 4% -12.2% 3565055 ± 3% time.involuntary_context_switches 2.809e+08 +18.9% 3.34e+08 time.minor_page_faults 611.56 +12.3% 686.50 time.percent_of_cpu_this_job_got 862.38 +14.4% 986.68 time.system_time 1170 +10.3% 1290 time.user_time 23973299 +15.5% 27696636 time.voluntary_context_switches 124.00 ± 19% -100.0% 0.00 slabinfo.btrfs_path.active_objs 124.00 ± 19% -100.0% 0.00 slabinfo.btrfs_path.num_objs 249.33 ± 10% -33.5% 165.75 ± 10% slabinfo.buffer_head.active_objs 249.33 ± 10% -33.5% 165.75 ± 10% slabinfo.buffer_head.num_objs 10443 ± 5% +19.4% 12471 ± 6% slabinfo.kmalloc-32.active_objs 10825 ± 5% +18.0% 12773 ± 7% slabinfo.kmalloc-32.num_objs 3987 +35.8% 5413 slabinfo.kmalloc-64.active_objs 3987 +35.8% 5413 slabinfo.kmalloc-64.num_objs 36.89 ± 7% +171.1% 100.00 proc-vmstat.nr_active_file 10606 ± 3% +13.1% 11991 proc-vmstat.nr_shmem unixbench.score 4000 O-+-O-O--------------------------------------------------------------+ | O O O O O O O O O O O O O O O O O O O O | 3500 +-+ | 3000 +-+.+.+..+.+.+.+.+.+..+.+.+ +.+..+.+.+.+.+.+.+ +.+.+.+.+..+.+.+.| | : : : : | 2500 +-+ : : : : | | : : : : | 2000 +-+ : : : : | | : : : : | 1500 +-+ : : : : | 1000 +-+ : : : : | | : : : : | 500 +-+ : :: | | : : | 0 +-+------------------------------------------------------------------+ unixbench.workload 1.8e+07 +-+---------------------------------------------------------------+ O O O O | 1.6e+07 +-O O O O O O O O O O O O O O O O O O O | 1.4e+07 +-+ .+.+. .+.+.| |.+.+.+.+.+.+.+..+.+.+.+.+ +.+.+.+.+.+ + +.+.+.+.+.+ | 1.2e+07 +-+ : : : : | 1e+07 +-+ : : : : | | : : : : | 8e+06 +-+ : : : : | 6e+06 +-+ : : : : | | : : : : | 4e+06 +-+ : : : : | 2e+06 +-+ : :: | | : : | 0 +-+---------------------------------------------------------------+ [*] 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