Greeting, FYI, we noticed a 4.3% improvement of hackbench.throughput due to commit: commit: ba73d98745be1c10dc3cce68e8d7b95012d07d05 ("namei: handle idmapped mounts in may_*() helpers") https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master in testcase: hackbench on test machine: 96 threads 2 sockets Intel(R) Xeon(R) CPU @ 2.30GHz with 128G memory with following parameters: nr_threads: 50% iterations: 8 mode: process ipc: pipe cpufreq_governor: performance ucode: 0x4003006 test-description: Hackbench is both a benchmark and a stress test for the Linux kernel scheduler. test-url: https://github.com/linux-test-project/ltp/blob/master/testcases/kernel/sc... Details are as below: --------------------------------------------------------------------------------------------------> To reproduce: git clone https://github.com/intel/lkp-tests.git cd lkp-tests sudo bin/lkp install job.yaml # job file is attached in this email bin/lkp split-job --compatible job.yaml # generate the yaml file for lkp run sudo bin/lkp run generated-yaml-file # if come across any failure that blocks the test, # please remove ~/.lkp and /lkp dir to run from a clean state. ========================================================================================= compiler/cpufreq_governor/ipc/iterations/kconfig/mode/nr_threads/rootfs/tbox_group/testcase/ucode: gcc-9/performance/pipe/8/x86_64-rhel-8.3/process/50%/debian-10.4-x86_64-20200603.cgz/lkp-csl-2sp4/hackbench/0x4003006 commit: 0d56a4518d ("stat: handle idmapped mounts") ba73d98745 ("namei: handle idmapped mounts in may_*() helpers") 0d56a4518d5eaf59 ba73d98745be1c10dc3cce68e8d ---------------- --------------------------- %stddev %change %stddev \ | \ 351513 +4.3% 366744 hackbench.throughput 9.156e+08 ± 2% -14.7% 7.807e+08 ± 8% hackbench.time.involuntary_context_switches 1679 -5.5% 1586 ± 2% hackbench.time.user_time 2.496e+09 -7.4% 2.311e+09 ± 3% hackbench.time.voluntary_context_switches 3341 -2.0% 3275 boot-time.idle 88548243 +17.6% 1.042e+08 ± 6% cpuidle..usage 0.06 ± 2% +0.0 0.08 ± 5% mpstat.cpu.all.soft% 676.30 ± 9% +23.6% 836.03 ± 18% sched_debug.cpu.nr_uninterruptible.max 12398635 -5.2% 11754370 vmstat.system.cs 1342926 -5.8% 1264455 vmstat.system.in 43240116 ± 2% +22.4% 52937898 ± 8% turbostat.C1 0.77 ± 6% +0.2 0.98 ± 11% turbostat.C1% 3.796e+08 ± 2% -9.3% 3.443e+08 ± 4% turbostat.IRQ 0.00 +0.1 0.07 ± 9% perf-profile.children.cycles-pp.make_kuid 0.00 +0.1 0.09 ± 13% perf-profile.children.cycles-pp.make_kgid 0.00 +0.1 0.14 ± 9% perf-profile.children.cycles-pp.map_id_range_down 0.37 ± 9% +0.2 0.56 ± 9% perf-profile.children.cycles-pp.touch_atime 0.24 ± 13% +0.2 0.43 ± 11% perf-profile.children.cycles-pp.atime_needs_update 0.00 +0.1 0.13 ± 9% perf-profile.self.cycles-pp.map_id_range_down 0.34 ±114% -79.7% 0.07 ± 76% perf-sched.wait_time.avg.ms.preempt_schedule_common._cond_resched.__put_anon_vma.unlink_anon_vmas.free_pgtables 0.01 ±124% +2430.1% 0.35 ±124% perf-sched.wait_time.avg.ms.preempt_schedule_common._cond_resched.dput.step_into.path_openat 0.53 ±115% -100.0% 0.00 perf-sched.wait_time.avg.ms.wait_for_partner.fifo_open.do_dentry_open.path_openat 0.02 ±130% +1574.5% 0.39 ±122% perf-sched.wait_time.max.ms.preempt_schedule_common._cond_resched.dput.step_into.path_openat 27.59 ± 55% +649.8% 206.87 ±147% perf-sched.wait_time.max.ms.preempt_schedule_common._cond_resched.pipe_write.new_sync_write.vfs_write 1.34 ±112% -100.0% 0.00 perf-sched.wait_time.max.ms.wait_for_partner.fifo_open.do_dentry_open.path_openat 14.79 -4.6% 14.11 perf-stat.i.MPKI 3.113e+10 +3.5% 3.222e+10 perf-stat.i.branch-instructions 1.20 -0.0 1.16 perf-stat.i.branch-miss-rate% 2.196e+09 -2.1% 2.151e+09 perf-stat.i.cache-references 12493419 -5.3% 11828083 perf-stat.i.context-switches 1.90 -2.5% 1.85 perf-stat.i.cpi 1567790 -7.1% 1456671 ± 3% perf-stat.i.cpu-migrations 4.432e+10 +2.3% 4.534e+10 perf-stat.i.dTLB-loads 2.594e+10 +1.5% 2.632e+10 perf-stat.i.dTLB-stores 69.41 +1.8 71.25 perf-stat.i.iTLB-load-miss-rate% 2.024e+08 +3.4% 2.092e+08 perf-stat.i.iTLB-load-misses 88106142 -5.4% 83326808 perf-stat.i.iTLB-loads 1.515e+11 +2.7% 1.556e+11 perf-stat.i.instructions 0.54 +2.7% 0.55 perf-stat.i.ipc 1079 +2.4% 1104 perf-stat.i.metric.M/sec 14.51 -4.7% 13.83 perf-stat.overall.MPKI 1.19 -0.0 1.15 perf-stat.overall.branch-miss-rate% 1.87 -2.6% 1.82 perf-stat.overall.cpi 69.64 +1.8 71.48 perf-stat.overall.iTLB-load-miss-rate% 0.53 +2.7% 0.55 perf-stat.overall.ipc 48.98 -1.3 47.65 perf-stat.overall.node-store-miss-rate% 3.101e+10 +3.5% 3.21e+10 perf-stat.ps.branch-instructions 2.189e+09 -2.1% 2.143e+09 perf-stat.ps.cache-references 12461918 -5.3% 11802321 perf-stat.ps.context-switches 1565708 -7.0% 1455380 ± 3% perf-stat.ps.cpu-migrations 4.415e+10 +2.3% 4.518e+10 perf-stat.ps.dTLB-loads 2.584e+10 +1.5% 2.622e+10 perf-stat.ps.dTLB-stores 2.016e+08 +3.4% 2.084e+08 perf-stat.ps.iTLB-load-misses 87887600 -5.4% 83148568 ± 2% perf-stat.ps.iTLB-loads 1.509e+11 +2.7% 1.55e+11 perf-stat.ps.instructions 31228 ± 4% +9.7% 34263 ± 2% softirqs.CPU1.SCHED 29873 ± 2% +10.2% 32919 ± 3% softirqs.CPU15.SCHED 29895 ± 2% +11.0% 33170 ± 2% softirqs.CPU16.SCHED 29884 ± 3% +12.1% 33513 ± 4% softirqs.CPU19.SCHED 29770 ± 2% +9.9% 32729 ± 3% softirqs.CPU20.SCHED 29584 ± 2% +11.0% 32839 ± 2% softirqs.CPU21.SCHED 30153 ± 2% +12.8% 34003 ± 6% softirqs.CPU24.SCHED 30757 ± 3% +10.9% 34123 ± 3% softirqs.CPU25.SCHED 30870 ± 2% +11.1% 34289 ± 3% softirqs.CPU26.SCHED 31234 ± 3% +10.1% 34391 ± 3% softirqs.CPU27.SCHED 30548 ± 2% +9.4% 33405 ± 3% softirqs.CPU30.SCHED 30163 ± 2% +10.7% 33383 ± 4% softirqs.CPU31.SCHED 30903 ± 3% +10.0% 33985 ± 3% softirqs.CPU32.SCHED 30884 ± 3% +11.1% 34297 ± 3% softirqs.CPU33.SCHED 29902 +11.1% 33217 ± 3% softirqs.CPU35.SCHED 30522 ± 3% +9.4% 33393 ± 3% softirqs.CPU36.SCHED 30582 ± 3% +10.3% 33727 ± 2% softirqs.CPU37.SCHED 30650 ± 2% +10.4% 33845 ± 3% softirqs.CPU38.SCHED 30637 ± 3% +11.0% 34002 ± 5% softirqs.CPU39.SCHED 30103 ± 2% +12.0% 33708 ± 2% softirqs.CPU4.SCHED 30160 ± 3% +12.3% 33868 ± 4% softirqs.CPU40.SCHED 30320 ± 3% +10.2% 33407 ± 4% softirqs.CPU41.SCHED 31093 ± 2% +8.1% 33609 ± 4% softirqs.CPU42.SCHED 30899 ± 2% +9.1% 33708 ± 4% softirqs.CPU44.SCHED 30808 ± 2% +10.5% 34051 ± 4% softirqs.CPU45.SCHED 30747 ± 2% +10.9% 34110 ± 3% softirqs.CPU46.SCHED 30485 ± 2% +11.1% 33859 ± 3% softirqs.CPU47.SCHED 30153 ± 3% +9.9% 33139 ± 2% softirqs.CPU48.SCHED 29785 ± 3% +10.8% 33015 ± 2% softirqs.CPU49.SCHED 29772 ± 2% +11.3% 33136 ± 3% softirqs.CPU52.SCHED 30346 ± 2% +11.9% 33958 ± 2% softirqs.CPU57.SCHED 29654 ± 2% +10.8% 32843 softirqs.CPU6.SCHED 30154 ± 2% +9.3% 32950 ± 4% softirqs.CPU63.SCHED 30397 ± 4% +9.2% 33202 ± 2% softirqs.CPU68.SCHED 29773 +11.8% 33283 ± 4% softirqs.CPU72.SCHED 30700 ± 2% +10.5% 33922 ± 4% softirqs.CPU73.SCHED 31490 ± 3% +9.8% 34568 ± 3% softirqs.CPU75.SCHED 30770 ± 2% +9.1% 33564 ± 4% softirqs.CPU77.SCHED 30476 ± 3% +9.9% 33504 ± 4% softirqs.CPU78.SCHED 31012 ± 3% +10.1% 34142 ± 3% softirqs.CPU80.SCHED 31266 ± 4% +10.2% 34464 ± 3% softirqs.CPU81.SCHED 30431 ± 3% +11.9% 34040 ± 3% softirqs.CPU85.SCHED 30765 ± 2% +10.7% 34052 ± 3% softirqs.CPU86.SCHED 30473 ± 3% +11.3% 33932 ± 3% softirqs.CPU88.SCHED 30285 ± 2% +11.7% 33838 ± 3% softirqs.CPU89.SCHED 30872 ± 3% +10.3% 34056 ± 3% softirqs.CPU90.SCHED 31275 ± 3% +9.9% 34361 ± 4% softirqs.CPU91.SCHED 30836 ± 2% +11.3% 34321 ± 4% softirqs.CPU94.SCHED 2929299 +9.9% 3217849 ± 2% softirqs.SCHED 3291660 ± 2% -10.2% 2955467 ± 4% interrupts.CPU0.RES:Rescheduling_interrupts 3209155 -9.9% 2890930 ± 4% interrupts.CPU1.RES:Rescheduling_interrupts 3167361 -10.1% 2848895 ± 5% interrupts.CPU12.RES:Rescheduling_interrupts 3205437 ± 2% -10.2% 2878328 ± 5% interrupts.CPU13.RES:Rescheduling_interrupts 3214503 -10.4% 2878875 ± 5% interrupts.CPU15.RES:Rescheduling_interrupts 3193257 -10.4% 2859582 ± 5% interrupts.CPU23.RES:Rescheduling_interrupts 3197798 ± 4% -10.8% 2851897 ± 4% interrupts.CPU24.RES:Rescheduling_interrupts 3324203 ± 3% -10.9% 2962555 ± 5% interrupts.CPU26.RES:Rescheduling_interrupts 3322402 ± 3% -11.0% 2955560 ± 4% interrupts.CPU27.RES:Rescheduling_interrupts 3331711 ± 2% -13.1% 2893654 ± 5% interrupts.CPU28.RES:Rescheduling_interrupts 3348281 ± 3% -14.5% 2864319 ± 4% interrupts.CPU29.RES:Rescheduling_interrupts 3234270 ± 3% -11.7% 2855014 ± 4% interrupts.CPU30.RES:Rescheduling_interrupts 3277111 ± 2% -12.3% 2873186 ± 4% interrupts.CPU31.RES:Rescheduling_interrupts 3309840 ± 3% -12.5% 2896464 ± 2% interrupts.CPU32.RES:Rescheduling_interrupts 3303859 ± 3% -11.6% 2922260 ± 4% interrupts.CPU33.RES:Rescheduling_interrupts 3287003 ± 4% -12.8% 2866335 ± 4% interrupts.CPU34.RES:Rescheduling_interrupts 3268249 ± 2% -13.1% 2838933 ± 5% interrupts.CPU35.RES:Rescheduling_interrupts 3278415 ± 2% -12.5% 2867775 ± 4% interrupts.CPU36.RES:Rescheduling_interrupts 3252387 ± 3% -12.0% 2861895 ± 4% interrupts.CPU37.RES:Rescheduling_interrupts 3312763 ± 3% -12.3% 2905598 ± 3% interrupts.CPU38.RES:Rescheduling_interrupts 3299906 ± 3% -12.2% 2898481 ± 4% interrupts.CPU39.RES:Rescheduling_interrupts 3227478 ± 3% -10.0% 2904683 ± 3% interrupts.CPU40.RES:Rescheduling_interrupts 3233470 ± 4% -11.8% 2850359 ± 4% interrupts.CPU41.RES:Rescheduling_interrupts 3247904 ± 4% -11.7% 2869199 ± 2% interrupts.CPU42.RES:Rescheduling_interrupts 3313033 ± 4% -11.2% 2941349 ± 4% interrupts.CPU43.RES:Rescheduling_interrupts 3250621 ± 4% -10.9% 2896775 ± 3% interrupts.CPU44.RES:Rescheduling_interrupts 5474 ± 35% +51.3% 8281 interrupts.CPU45.NMI:Non-maskable_interrupts 5474 ± 35% +51.3% 8281 interrupts.CPU45.PMI:Performance_monitoring_interrupts 3278556 ± 3% -11.0% 2918741 ± 5% interrupts.CPU45.RES:Rescheduling_interrupts 3271070 ± 4% -10.3% 2933335 ± 5% interrupts.CPU46.RES:Rescheduling_interrupts 3253683 ± 4% -11.3% 2884737 ± 4% interrupts.CPU47.RES:Rescheduling_interrupts 3186697 ± 2% -9.3% 2891753 ± 5% interrupts.CPU48.RES:Rescheduling_interrupts 3198247 -9.6% 2891256 ± 5% interrupts.CPU49.RES:Rescheduling_interrupts 3252717 -10.1% 2923091 ± 4% interrupts.CPU56.RES:Rescheduling_interrupts 3241606 ± 2% -9.7% 2926074 ± 5% interrupts.CPU57.RES:Rescheduling_interrupts 3200963 ± 2% -9.8% 2886532 ± 4% interrupts.CPU61.RES:Rescheduling_interrupts 3258239 ± 2% -10.3% 2921021 ± 6% interrupts.CPU67.RES:Rescheduling_interrupts 3205346 -9.5% 2901805 ± 4% interrupts.CPU68.RES:Rescheduling_interrupts 3202213 ± 2% -9.4% 2902413 ± 5% interrupts.CPU70.RES:Rescheduling_interrupts 3176968 ± 4% -11.4% 2813497 ± 5% interrupts.CPU72.RES:Rescheduling_interrupts 3259194 ± 4% -10.5% 2917966 ± 4% interrupts.CPU73.RES:Rescheduling_interrupts 3309819 ± 3% -11.3% 2937158 ± 4% interrupts.CPU74.RES:Rescheduling_interrupts 3368701 ± 4% -12.3% 2954496 ± 4% interrupts.CPU75.RES:Rescheduling_interrupts 3302093 ± 3% -12.9% 2874666 ± 3% interrupts.CPU76.RES:Rescheduling_interrupts 3322200 ± 4% -13.3% 2880899 ± 3% interrupts.CPU77.RES:Rescheduling_interrupts 3254923 ± 4% -12.3% 2854791 ± 3% interrupts.CPU78.RES:Rescheduling_interrupts 3298900 ± 2% -12.8% 2877407 ± 5% interrupts.CPU79.RES:Rescheduling_interrupts 3329217 ± 4% -12.0% 2928239 ± 3% interrupts.CPU80.RES:Rescheduling_interrupts 3350850 ± 3% -11.6% 2963550 ± 4% interrupts.CPU81.RES:Rescheduling_interrupts 3261881 ± 4% -11.6% 2883912 ± 3% interrupts.CPU82.RES:Rescheduling_interrupts 3262199 ± 4% -13.6% 2819681 ± 5% interrupts.CPU83.RES:Rescheduling_interrupts 3296607 ± 3% -12.4% 2887541 ± 4% interrupts.CPU84.RES:Rescheduling_interrupts 3247117 ± 4% -11.6% 2871867 ± 4% interrupts.CPU85.RES:Rescheduling_interrupts 3304686 ± 3% -11.4% 2927270 ± 4% interrupts.CPU86.RES:Rescheduling_interrupts 3313258 ± 3% -12.8% 2890225 ± 4% interrupts.CPU87.RES:Rescheduling_interrupts 3276588 ± 4% -11.1% 2911694 ± 4% interrupts.CPU88.RES:Rescheduling_interrupts 3245365 ± 4% -11.8% 2861921 ± 3% interrupts.CPU89.RES:Rescheduling_interrupts 3206511 -9.4% 2905199 ± 4% interrupts.CPU9.RES:Rescheduling_interrupts 3260447 ± 3% -11.2% 2894368 ± 3% interrupts.CPU90.RES:Rescheduling_interrupts 3318146 ± 4% -11.0% 2954676 ± 3% interrupts.CPU91.RES:Rescheduling_interrupts 3253573 ± 4% -10.1% 2924313 ± 4% interrupts.CPU92.RES:Rescheduling_interrupts 3278406 ± 5% -10.5% 2934051 ± 5% interrupts.CPU93.RES:Rescheduling_interrupts 3343345 ± 3% -11.4% 2962741 ± 4% interrupts.CPU94.RES:Rescheduling_interrupts 3292411 ± 4% -12.0% 2897405 ± 4% interrupts.CPU95.RES:Rescheduling_interrupts 3.111e+08 ± 2% -10.5% 2.783e+08 ± 4% interrupts.RES:Rescheduling_interrupts hackbench.throughput 385000 +------------------------------------------------------------------+ 380000 |O+ OO O | | O O O O O | 375000 |-+ O O O O O OO O O| 370000 |-+ O O O O O O OO O | 365000 |-+ O O OO O O | 360000 |-+O O O O + | | + O O + : | 355000 |-+ :: O O + : O O | 350000 |-+ :: + .+O +. + + :+ | 345000 |-++ : : +: +.+ +.++ +.+ + + +. : + | 340000 |-:: + ++ :.+ :.+ + + | |:: :+ + + | 335000 |++ + | 330000 +------------------------------------------------------------------+ [*] 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. --- 0DAY/LKP+ Test Infrastructure Open Source Technology Center https://lists.01.org/hyperkitty/list/lkp@lists.01.org Intel Corporation Thanks, Oliver Sang