Greeting, FYI, we noticed a -39.8% regression of will-it-scale.per_process_ops due to commit: commit: 059dd9003ad894ad7b2e47126cde89a3ec46e630 ("[patch 101/212] memcg: enable accounting for file lock caches") url: https://github.com/0day-ci/linux/commits/Andrew-Morton/ia64-fix-typo-in-a-comment/20210903-065028 in testcase: will-it-scale on test machine: 96 threads 2 sockets Ice Lake with 256G memory with following parameters: nr_task: 100% mode: process test: lock1 cpufreq_governor: performance ucode: 0xb000280 test-description: Will It Scale takes a testcase and runs it from 1 through to n parallel copies to see if the testcase will scale. It builds both a process and threads based test in order to see any differences between the two. test-url: https://github.com/antonblanchard/will-it-scale In addition to that, the commit also has significant impact on the following tests: +------------------+---------------------------------------------------------------------------------+ | testcase: change | will-it-scale: will-it-scale.per_process_ops -42.9% regression | | test machine | 128 threads 2 sockets Intel(R) Xeon(R) Gold 6338 CPU @ 2.00GHz with 256G memory | | test parameters | cpufreq_governor=performance | | | mode=process | | | nr_task=100% | | | test=lock1 | | | ucode=0xd000280 | +------------------+---------------------------------------------------------------------------------+ | testcase: change | will-it-scale: will-it-scale.per_process_ops -38.0% regression | | test machine | 104 threads 2 sockets Skylake with 192G memory | | test parameters | cpufreq_governor=performance | | | mode=process | | | nr_task=100% | | | test=lock1 | | | ucode=0x2006a0a | +------------------+---------------------------------------------------------------------------------+ | testcase: change | will-it-scale: will-it-scale.per_process_ops -34.5% regression | | test machine | 96 threads 2 sockets Ice Lake with 256G memory | | test parameters | cpufreq_governor=performance | | | mode=process | | | nr_task=16 | | | test=lock1 | | | ucode=0xb000280 | +------------------+---------------------------------------------------------------------------------+ If you fix the issue, kindly add following tag Reported-by: kernel test robot 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 split-job --compatible job.yaml # generate the yaml file for lkp run bin/lkp run generated-yaml-file ========================================================================================= compiler/cpufreq_governor/kconfig/mode/nr_task/rootfs/tbox_group/test/testcase/ucode: gcc-9/performance/x86_64-rhel-8.3/process/100%/debian-10.4-x86_64-20200603.cgz/lkp-icl-2sp1/lock1/will-it-scale/0xb000280 commit: fa4e6b1ad5 ("memcg: enable accounting for pollfd and select bits arrays") 059dd9003a ("memcg: enable accounting for file lock caches") fa4e6b1ad57df096 059dd9003ad894ad7b2e47126cd ---------------- --------------------------- %stddev %change %stddev \ | \ 1.498e+08 -39.8% 90225756 will-it-scale.96.processes 1560769 -39.8% 939851 will-it-scale.per_process_ops 1.498e+08 -39.8% 90225756 will-it-scale.workload 2901 +1.8% 2954 turbostat.Bzy_MHz 11.00 -45.5% 6.00 vmstat.cpu.us 0.01 ± 6% -0.0 0.01 ± 3% mpstat.cpu.all.soft% 11.77 -4.8 6.95 mpstat.cpu.all.usr% 6970 -24.9% 5231 ± 11% slabinfo.Acpi-Parse.active_objs 6970 -24.9% 5231 ± 11% slabinfo.Acpi-Parse.num_objs 74280 +7.7% 80016 proc-vmstat.nr_active_anon 81994 +7.4% 88022 proc-vmstat.nr_shmem 74280 +7.7% 80016 proc-vmstat.nr_zone_active_anon 1250 ± 56% +85.8% 2322 ± 39% interrupts.CPU10.RES:Rescheduling_interrupts 1634 ± 36% -40.5% 972.00 ± 20% interrupts.CPU50.RES:Rescheduling_interrupts 1554 ± 33% -35.5% 1002 ± 25% interrupts.CPU60.RES:Rescheduling_interrupts 591.67 ± 60% +131.4% 1368 ± 36% interrupts.CPU61.CAL:Function_call_interrupts 556.67 ± 18% +134.4% 1305 ± 22% interrupts.CPU71.RES:Rescheduling_interrupts 10513 ± 17% +41.0% 14823 interrupts.CPU8.NMI:Non-maskable_interrupts 10513 ± 17% +41.0% 14823 interrupts.CPU8.PMI:Performance_monitoring_interrupts 11329 ± 20% +30.8% 14814 interrupts.CPU9.NMI:Non-maskable_interrupts 11329 ± 20% +30.8% 14814 interrupts.CPU9.PMI:Performance_monitoring_interrupts 371.67 ± 2% +21.5% 451.67 ± 2% interrupts.IWI:IRQ_work_interrupts 0.01 ± 4% +20.7% 0.01 ± 6% perf-sched.sch_delay.avg.ms.rcu_gp_kthread.kthread.ret_from_fork 0.01 +16.7% 0.01 perf-sched.sch_delay.avg.ms.schedule_timeout.rcu_gp_kthread.kthread.ret_from_fork 1195 ± 19% -43.4% 676.69 ± 18% perf-sched.wait_and_delay.avg.ms.devkmsg_read.vfs_read.ksys_read.do_syscall_64 469.30 ± 6% -19.0% 380.17 ± 17% perf-sched.wait_and_delay.avg.ms.schedule_hrtimeout_range_clock.ep_poll.do_epoll_wait.__x64_sys_epoll_wait 94.52 ± 2% -20.2% 75.46 ± 7% perf-sched.wait_and_delay.avg.ms.schedule_hrtimeout_range_clock.poll_schedule_timeout.constprop.0.do_sys_poll 607.67 ± 13% -66.2% 205.50 ± 9% perf-sched.wait_and_delay.count.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_sysvec_apic_timer_interrupt.[unknown] 875.83 ± 5% -49.9% 439.00 ± 10% perf-sched.wait_and_delay.count.preempt_schedule_common.__cond_resched.kmem_cache_alloc.locks_alloc_lock.fcntl_setlk 1236 ± 10% +84.4% 2281 ± 33% perf-sched.wait_and_delay.count.preempt_schedule_common.__cond_resched.posix_lock_inode.do_lock_file_wait.fcntl_setlk 284.67 ± 2% +28.7% 366.50 ± 2% perf-sched.wait_and_delay.count.schedule_hrtimeout_range_clock.poll_schedule_timeout.constprop.0.do_sys_poll 3752 ± 18% -43.9% 2103 ± 14% perf-sched.wait_and_delay.max.ms.devkmsg_read.vfs_read.ksys_read.do_syscall_64 3755 ± 18% -43.9% 2107 ± 14% perf-sched.wait_and_delay.max.ms.pipe_read.new_sync_read.vfs_read.ksys_read 2022 ± 42% -49.7% 1017 ± 3% perf-sched.wait_and_delay.max.ms.preempt_schedule_common.__cond_resched.wait_for_completion.affine_move_task.__set_cpus_allowed_ptr 3385 ± 20% -48.3% 1749 ± 19% perf-sched.wait_and_delay.max.ms.schedule_hrtimeout_range_clock.ep_poll.do_epoll_wait.__x64_sys_epoll_wait 1195 ± 19% -43.4% 676.33 ± 18% perf-sched.wait_time.avg.ms.devkmsg_read.vfs_read.ksys_read.do_syscall_64 3.00 ± 17% -50.6% 1.48 ± 16% perf-sched.wait_time.avg.ms.rcu_gp_kthread.kthread.ret_from_fork 469.23 ± 6% -19.0% 380.15 ± 17% perf-sched.wait_time.avg.ms.schedule_hrtimeout_range_clock.ep_poll.do_epoll_wait.__x64_sys_epoll_wait 94.52 ± 2% -20.2% 75.45 ± 7% perf-sched.wait_time.avg.ms.schedule_hrtimeout_range_clock.poll_schedule_timeout.constprop.0.do_sys_poll 3752 ± 18% -43.9% 2103 ± 14% perf-sched.wait_time.max.ms.devkmsg_read.vfs_read.ksys_read.do_syscall_64 3755 ± 18% -43.9% 2107 ± 14% perf-sched.wait_time.max.ms.pipe_read.new_sync_read.vfs_read.ksys_read 2021 ± 43% -49.7% 1017 ± 3% perf-sched.wait_time.max.ms.preempt_schedule_common.__cond_resched.wait_for_completion.affine_move_task.__set_cpus_allowed_ptr 3383 ± 20% -48.3% 1749 ± 19% perf-sched.wait_time.max.ms.schedule_hrtimeout_range_clock.ep_poll.do_epoll_wait.__x64_sys_epoll_wait 0.02 ± 12% +1472.5% 0.29 ± 2% perf-stat.i.MPKI 5.912e+10 -4.4% 5.649e+10 perf-stat.i.branch-instructions 85960000 -5.0% 81678832 perf-stat.i.branch-misses 12.55 +14.6 27.15 ± 2% perf-stat.i.cache-miss-rate% 569950 ± 4% +3655.7% 21405416 perf-stat.i.cache-misses 4221711 ± 4% +1785.6% 79604401 ± 2% perf-stat.i.cache-references 0.94 +7.8% 1.01 perf-stat.i.cpi 2.759e+11 +1.9% 2.812e+11 perf-stat.i.cpu-cycles 138.11 -5.2% 130.88 perf-stat.i.cpu-migrations 797859 -98.3% 13247 perf-stat.i.cycles-between-cache-misses 118331 ± 11% -26.3% 87177 ± 13% perf-stat.i.dTLB-load-misses 8.647e+10 -2.0% 8.473e+10 perf-stat.i.dTLB-loads 0.00 ± 2% -0.0 0.00 ± 5% perf-stat.i.dTLB-store-miss-rate% 109661 -31.9% 74676 perf-stat.i.dTLB-store-misses 5.475e+10 -9.5% 4.953e+10 perf-stat.i.dTLB-stores 2.947e+11 -5.3% 2.792e+11 perf-stat.i.instructions 1.07 -7.0% 0.99 perf-stat.i.ipc 2.87 +1.9% 2.93 perf-stat.i.metric.GHz 45.11 +2065.8% 976.94 perf-stat.i.metric.K/sec 2086 -4.8% 1987 perf-stat.i.metric.M/sec 88.85 -4.9 83.91 perf-stat.i.node-load-miss-rate% 67988 ± 4% +10628.1% 7293874 ± 3% perf-stat.i.node-load-misses 29378 ± 10% +4671.2% 1401709 ± 10% perf-stat.i.node-loads 32.56 ± 12% +55.7 88.23 perf-stat.i.node-store-miss-rate% 57616 ± 10% +14021.2% 8136104 ± 2% perf-stat.i.node-store-misses 128311 ± 2% +743.3% 1082066 ± 3% perf-stat.i.node-stores 0.01 ± 4% +1873.4% 0.29 ± 2% perf-stat.overall.MPKI 13.50 +13.4 26.90 ± 2% perf-stat.overall.cache-miss-rate% 0.94 +7.6% 1.01 perf-stat.overall.cpi 480654 ± 4% -97.3% 13131 perf-stat.overall.cycles-between-cache-misses 0.00 ± 10% -0.0 0.00 ± 13% perf-stat.overall.dTLB-load-miss-rate% 0.00 -0.0 0.00 perf-stat.overall.dTLB-store-miss-rate% 1.07 -7.0% 0.99 perf-stat.overall.ipc 68.90 ± 4% +15.0 83.91 perf-stat.overall.node-load-miss-rate% 30.88 ± 7% +57.4 88.26 perf-stat.overall.node-store-miss-rate% 593897 +57.1% 933119 perf-stat.overall.path-length 5.892e+10 -4.4% 5.631e+10 perf-stat.ps.branch-instructions 85676218 -5.0% 81415292 perf-stat.ps.branch-misses 573086 ± 4% +3624.0% 21341565 perf-stat.ps.cache-misses 4246570 ± 4% +1769.4% 79385352 ± 2% perf-stat.ps.cache-references 2.75e+11 +1.9% 2.802e+11 perf-stat.ps.cpu-cycles 137.70 -5.3% 130.46 perf-stat.ps.cpu-migrations 119781 ± 10% -24.8% 90043 ± 13% perf-stat.ps.dTLB-load-misses 8.618e+10 -2.0% 8.445e+10 perf-stat.ps.dTLB-loads 109428 -31.8% 74634 perf-stat.ps.dTLB-store-misses 5.457e+10 -9.5% 4.936e+10 perf-stat.ps.dTLB-stores 2.937e+11 -5.3% 2.783e+11 perf-stat.ps.instructions 68059 ± 4% +10582.2% 7270327 ± 3% perf-stat.ps.node-load-misses 30738 ± 10% +4449.2% 1398381 ± 10% perf-stat.ps.node-loads 57570 ± 10% +13987.0% 8110007 ± 2% perf-stat.ps.node-store-misses 128514 ± 2% +739.5% 1078903 ± 3% perf-stat.ps.node-stores 8.899e+13 -5.4% 8.419e+13 perf-stat.total.instructions 18148 ± 6% -51.4% 8827 ± 13% softirqs.CPU0.RCU 18865 ± 13% -52.5% 8967 ± 6% softirqs.CPU1.RCU 16986 ± 8% -50.3% 8436 ± 6% softirqs.CPU10.RCU 16457 ± 7% -51.8% 7927 ± 11% softirqs.CPU11.RCU 16715 ± 8% -51.9% 8032 ± 4% softirqs.CPU12.RCU 16563 ± 5% -50.4% 8213 ± 4% softirqs.CPU13.RCU 16100 ± 10% -48.7% 8254 ± 5% softirqs.CPU14.RCU 17325 ± 13% -53.6% 8047 ± 4% softirqs.CPU15.RCU 14981 ± 6% -46.7% 7990 ± 4% softirqs.CPU16.RCU 15296 ± 8% -47.4% 8049 ± 6% softirqs.CPU17.RCU 15001 ± 5% -45.8% 8131 ± 6% softirqs.CPU18.RCU 16582 ± 18% -51.4% 8055 ± 3% softirqs.CPU19.RCU 16276 ± 5% -48.9% 8318 ± 2% softirqs.CPU2.RCU 15326 ± 9% -48.0% 7965 ± 2% softirqs.CPU20.RCU 16997 ± 17% -51.1% 8316 ± 3% softirqs.CPU21.RCU 15550 ± 8% -50.3% 7734 ± 10% softirqs.CPU24.RCU 14960 ± 9% -51.3% 7285 ± 2% softirqs.CPU25.RCU 14903 ± 7% -52.5% 7085 ± 3% softirqs.CPU26.RCU 14994 ± 9% -51.6% 7252 ± 5% softirqs.CPU27.RCU 14700 ± 9% -51.7% 7094 ± 3% softirqs.CPU28.RCU 14467 ± 9% -52.6% 6858 ± 4% softirqs.CPU29.RCU 16738 ± 4% -49.1% 8515 ± 2% softirqs.CPU3.RCU 14710 ± 9% -52.6% 6967 ± 3% softirqs.CPU30.RCU 14745 ± 8% -52.7% 6973 ± 3% softirqs.CPU31.RCU 15594 ± 7% -50.8% 7675 ± 5% softirqs.CPU32.RCU 15851 ± 8% -52.1% 7593 ± 9% softirqs.CPU33.RCU 15532 ± 10% -51.0% 7610 ± 3% softirqs.CPU34.RCU 16073 ± 7% -51.8% 7741 ± 4% softirqs.CPU35.RCU 16585 ± 9% -54.0% 7636 ± 8% softirqs.CPU36.RCU 16297 ± 7% -50.8% 8017 ± 6% softirqs.CPU37.RCU 15701 ± 7% -52.6% 7448 ± 2% softirqs.CPU38.RCU 16009 ± 8% -51.7% 7740 ± 2% softirqs.CPU39.RCU 16791 ± 4% -50.0% 8403 softirqs.CPU4.RCU 15760 ± 9% -49.7% 7928 ± 5% softirqs.CPU40.RCU 15574 ± 7% -51.0% 7636 ± 7% softirqs.CPU41.RCU 16007 ± 7% -52.4% 7623 ± 2% softirqs.CPU42.RCU 16174 ± 7% -52.5% 7677 ± 2% softirqs.CPU43.RCU 16230 ± 8% -52.6% 7686 ± 3% softirqs.CPU44.RCU 16293 ± 7% -52.3% 7775 ± 4% softirqs.CPU45.RCU 16168 ± 8% -53.1% 7591 softirqs.CPU46.RCU 16864 ± 7% -51.5% 8174 ± 6% softirqs.CPU47.RCU 15260 ± 3% -53.0% 7178 ± 12% softirqs.CPU48.RCU 15617 ± 6% -50.6% 7711 ± 4% softirqs.CPU49.RCU 16409 ± 5% -50.6% 8110 ± 3% softirqs.CPU5.RCU 14664 ± 14% -47.7% 7669 ± 2% softirqs.CPU50.RCU 15632 ± 5% -51.2% 7624 softirqs.CPU51.RCU 15531 ± 6% -49.6% 7827 ± 3% softirqs.CPU52.RCU 15909 ± 6% -51.1% 7776 ± 5% softirqs.CPU53.RCU 15694 ± 6% -51.1% 7680 ± 3% softirqs.CPU54.RCU 15511 ± 5% -50.8% 7633 ± 3% softirqs.CPU55.RCU 16052 ± 11% -52.4% 7639 ± 5% softirqs.CPU56.RCU 15381 ± 5% -50.7% 7579 ± 4% softirqs.CPU57.RCU 14483 ± 14% -49.0% 7385 softirqs.CPU58.RCU 16075 ± 5% -55.4% 7166 ± 18% softirqs.CPU59.RCU 16475 ± 5% -50.9% 8091 softirqs.CPU6.RCU 16172 ± 5% -53.1% 7584 softirqs.CPU60.RCU 16477 ± 6% -52.7% 7787 ± 4% softirqs.CPU61.RCU 15055 ± 17% -48.4% 7771 ± 3% softirqs.CPU62.RCU 16297 ± 6% -52.5% 7740 ± 3% softirqs.CPU63.RCU 14576 ± 8% -46.8% 7751 ± 6% softirqs.CPU64.RCU 14792 ± 6% -48.6% 7605 ± 4% softirqs.CPU65.RCU 14592 ± 7% -42.4% 8404 ± 14% softirqs.CPU66.RCU 15482 ± 17% -50.3% 7691 ± 5% softirqs.CPU68.RCU 14718 ± 7% -46.2% 7924 ± 7% softirqs.CPU69.RCU 16704 ± 5% -48.9% 8533 ± 3% softirqs.CPU7.RCU 14688 ± 6% -45.5% 8004 ± 4% softirqs.CPU70.RCU 15187 ± 10% -53.6% 7039 ± 2% softirqs.CPU72.RCU 15234 ± 9% -52.8% 7187 ± 5% softirqs.CPU73.RCU 14964 ± 8% -52.4% 7129 ± 6% softirqs.CPU74.RCU 15085 ± 9% -55.5% 6706 ± 6% softirqs.CPU75.RCU 15006 ± 9% -54.0% 6905 ± 3% softirqs.CPU76.RCU 14734 ± 10% -52.1% 7061 ± 2% softirqs.CPU77.RCU 15116 ± 9% -54.0% 6954 ± 3% softirqs.CPU78.RCU 15093 ± 9% -53.8% 6979 ± 5% softirqs.CPU79.RCU 16858 ± 5% -50.3% 8383 ± 3% softirqs.CPU8.RCU 15918 ± 6% -53.0% 7474 ± 3% softirqs.CPU80.RCU 16031 ± 6% -54.2% 7346 ± 11% softirqs.CPU81.RCU 15201 ± 15% -50.5% 7523 ± 3% softirqs.CPU82.RCU 16258 ± 7% -52.5% 7717 ± 2% softirqs.CPU83.RCU 16706 ± 8% -55.6% 7414 ± 8% softirqs.CPU84.RCU 16434 ± 8% -53.1% 7706 ± 3% softirqs.CPU85.RCU 15773 ± 7% -52.5% 7489 ± 4% softirqs.CPU86.RCU 16212 ± 7% -51.8% 7807 ± 4% softirqs.CPU87.RCU 15755 ± 10% -51.5% 7637 ± 3% softirqs.CPU88.RCU 15795 ± 7% -52.4% 7525 ± 7% softirqs.CPU89.RCU 16574 ± 7% -46.4% 8886 ± 15% softirqs.CPU9.RCU 16332 ± 8% -52.1% 7830 ± 5% softirqs.CPU90.RCU 16246 ± 6% -52.9% 7654 ± 2% softirqs.CPU91.RCU 16578 ± 8% -52.5% 7876 ± 4% softirqs.CPU92.RCU 16571 ± 8% -52.5% 7868 ± 5% softirqs.CPU93.RCU 16917 ± 10% -53.0% 7955 ± 4% softirqs.CPU94.RCU 17424 ± 7% -52.1% 8337 ± 4% softirqs.CPU95.RCU 1519697 ± 6% -50.6% 750455 softirqs.RCU 8.07 -3.5 4.61 perf-profile.calltrace.cycles-pp.__entry_text_start.__libc_fcntl64 5.66 -2.6 3.08 perf-profile.calltrace.cycles-pp.memset_erms.kmem_cache_alloc.locks_alloc_lock.posix_lock_inode.do_lock_file_wait 5.49 -2.4 3.08 perf-profile.calltrace.cycles-pp._copy_from_user.do_fcntl.__x64_sys_fcntl.do_syscall_64.entry_SYSCALL_64_after_hwframe 5.12 -2.1 2.98 perf-profile.calltrace.cycles-pp.security_file_lock.do_lock_file_wait.fcntl_setlk.do_fcntl.__x64_sys_fcntl 3.98 -1.7 2.24 perf-profile.calltrace.cycles-pp.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_fcntl64 3.75 -1.6 2.20 perf-profile.calltrace.cycles-pp.common_file_perm.security_file_lock.do_lock_file_wait.fcntl_setlk.do_fcntl 2.85 -1.3 1.56 ± 2% perf-profile.calltrace.cycles-pp.memset_erms.kmem_cache_alloc.locks_alloc_lock.fcntl_setlk.do_fcntl 2.16 -0.9 1.22 perf-profile.calltrace.cycles-pp.copy_user_generic_unrolled._copy_from_user.do_fcntl.__x64_sys_fcntl.do_syscall_64 1.69 -0.7 0.96 perf-profile.calltrace.cycles-pp.___might_sleep.kmem_cache_alloc.locks_alloc_lock.posix_lock_inode.do_lock_file_wait 1.59 -0.7 0.93 perf-profile.calltrace.cycles-pp.__might_sleep.kmem_cache_alloc.locks_alloc_lock.posix_lock_inode.do_lock_file_wait 1.70 -0.6 1.06 ± 2% perf-profile.calltrace.cycles-pp.syscall_return_via_sysret.__libc_fcntl64 1.29 -0.6 0.71 ± 2% perf-profile.calltrace.cycles-pp.__might_fault._copy_from_user.do_fcntl.__x64_sys_fcntl.do_syscall_64 1.20 -0.5 0.66 ± 2% perf-profile.calltrace.cycles-pp.copy_user_enhanced_fast_string._copy_from_user.do_fcntl.__x64_sys_fcntl.do_syscall_64 1.11 -0.5 0.60 ± 2% perf-profile.calltrace.cycles-pp.locks_release_private.posix_lock_inode.do_lock_file_wait.fcntl_setlk.do_fcntl 1.11 -0.5 0.60 ± 2% perf-profile.calltrace.cycles-pp.aa_file_perm.common_file_perm.security_file_lock.do_lock_file_wait.fcntl_setlk 1.06 -0.5 0.57 ± 3% perf-profile.calltrace.cycles-pp.flock64_to_posix_lock.fcntl_setlk.do_fcntl.__x64_sys_fcntl.do_syscall_64 1.32 ± 2% -0.5 0.83 perf-profile.calltrace.cycles-pp.__fget_light.__x64_sys_fcntl.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_fcntl64 1.08 -0.5 0.60 ± 2% perf-profile.calltrace.cycles-pp.exit_to_user_mode_prepare.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_fcntl64 0.61 -0.3 0.34 ± 70% perf-profile.calltrace.cycles-pp.locks_delete_block.do_lock_file_wait.fcntl_setlk.do_fcntl.__x64_sys_fcntl 0.71 +0.2 0.90 ± 7% perf-profile.calltrace.cycles-pp._raw_spin_lock.fcntl_setlk.do_fcntl.__x64_sys_fcntl.do_syscall_64 98.94 +0.3 99.25 perf-profile.calltrace.cycles-pp.__libc_fcntl64 0.72 +0.4 1.09 ± 9% perf-profile.calltrace.cycles-pp._raw_spin_lock.locks_insert_lock_ctx.posix_lock_inode.do_lock_file_wait.fcntl_setlk 0.70 +0.4 1.13 ± 9% perf-profile.calltrace.cycles-pp._raw_spin_lock.locks_unlink_lock_ctx.locks_delete_lock_ctx.posix_lock_inode.do_lock_file_wait 0.00 +0.6 0.59 perf-profile.calltrace.cycles-pp.refill_obj_stock.kmem_cache_free.fcntl_setlk.do_fcntl.__x64_sys_fcntl 1.48 +0.6 2.08 ± 5% perf-profile.calltrace.cycles-pp._raw_spin_lock.posix_lock_inode.do_lock_file_wait.fcntl_setlk.do_fcntl 0.00 +0.8 0.76 ± 24% perf-profile.calltrace.cycles-pp.page_counter_cancel.page_counter_uncharge.obj_cgroup_uncharge_pages.kmem_cache_free.locks_dispose_list 0.00 +0.8 0.80 perf-profile.calltrace.cycles-pp.mod_objcg_state.kmem_cache_free.fcntl_setlk.do_fcntl.__x64_sys_fcntl 3.14 +0.8 3.96 ± 4% perf-profile.calltrace.cycles-pp.locks_dispose_list.posix_lock_inode.do_lock_file_wait.fcntl_setlk.do_fcntl 0.00 +0.8 0.83 perf-profile.calltrace.cycles-pp.mod_objcg_state.kmem_cache_alloc.locks_alloc_lock.fcntl_setlk.do_fcntl 0.00 +0.9 0.91 perf-profile.calltrace.cycles-pp.refill_obj_stock.kmem_cache_free.posix_lock_inode.do_lock_file_wait.fcntl_setlk 0.00 +1.0 0.96 ± 21% perf-profile.calltrace.cycles-pp.page_counter_uncharge.obj_cgroup_uncharge_pages.kmem_cache_free.locks_dispose_list.posix_lock_inode 0.00 +1.0 0.98 ± 21% perf-profile.calltrace.cycles-pp.obj_cgroup_uncharge_pages.kmem_cache_free.locks_dispose_list.posix_lock_inode.do_lock_file_wait 0.00 +1.3 1.29 ± 2% perf-profile.calltrace.cycles-pp.mod_objcg_state.kmem_cache_free.posix_lock_inode.do_lock_file_wait.fcntl_setlk 1.81 +1.4 3.18 ± 6% perf-profile.calltrace.cycles-pp.kmem_cache_free.locks_dispose_list.posix_lock_inode.do_lock_file_wait.fcntl_setlk 0.00 +1.7 1.69 ± 2% perf-profile.calltrace.cycles-pp.mod_objcg_state.kmem_cache_alloc.locks_alloc_lock.posix_lock_inode.do_lock_file_wait 0.00 +1.8 1.84 ± 2% perf-profile.calltrace.cycles-pp.get_obj_cgroup_from_current.kmem_cache_alloc.locks_alloc_lock.fcntl_setlk.do_fcntl 0.00 +2.0 1.96 ± 14% perf-profile.calltrace.cycles-pp.page_counter_cancel.page_counter_uncharge.obj_cgroup_uncharge_pages.kmem_cache_free.fcntl_setlk 0.00 +2.3 2.31 ± 4% perf-profile.calltrace.cycles-pp.page_counter_cancel.page_counter_uncharge.obj_cgroup_uncharge_pages.kmem_cache_free.posix_lock_inode 0.00 +2.5 2.51 ± 11% perf-profile.calltrace.cycles-pp.page_counter_uncharge.obj_cgroup_uncharge_pages.kmem_cache_free.fcntl_setlk.do_fcntl 0.00 +2.5 2.52 ± 3% perf-profile.calltrace.cycles-pp.propagate_protected_usage.page_counter_try_charge.obj_cgroup_charge_pages.obj_cgroup_charge.kmem_cache_alloc 0.00 +2.6 2.56 ± 11% perf-profile.calltrace.cycles-pp.obj_cgroup_uncharge_pages.kmem_cache_free.fcntl_setlk.do_fcntl.__x64_sys_fcntl 0.00 +3.0 2.96 ± 6% perf-profile.calltrace.cycles-pp.page_counter_uncharge.obj_cgroup_uncharge_pages.kmem_cache_free.posix_lock_inode.do_lock_file_wait 0.00 +3.0 3.02 ± 6% perf-profile.calltrace.cycles-pp.obj_cgroup_uncharge_pages.kmem_cache_free.posix_lock_inode.do_lock_file_wait.fcntl_setlk 0.00 +3.5 3.49 ± 3% perf-profile.calltrace.cycles-pp.get_obj_cgroup_from_current.kmem_cache_alloc.locks_alloc_lock.posix_lock_inode.do_lock_file_wait 3.33 +3.5 6.82 ± 3% perf-profile.calltrace.cycles-pp.kmem_cache_free.fcntl_setlk.do_fcntl.__x64_sys_fcntl.do_syscall_64 0.00 +3.8 3.75 ± 11% perf-profile.calltrace.cycles-pp.obj_cgroup_charge_pages.obj_cgroup_charge.kmem_cache_alloc.locks_alloc_lock.fcntl_setlk 0.00 +4.4 4.38 ± 9% perf-profile.calltrace.cycles-pp.obj_cgroup_charge.kmem_cache_alloc.locks_alloc_lock.fcntl_setlk.do_fcntl 5.21 +4.5 9.67 ± 2% perf-profile.calltrace.cycles-pp.kmem_cache_free.posix_lock_inode.do_lock_file_wait.fcntl_setlk.do_fcntl 8.71 +5.3 14.03 ± 2% perf-profile.calltrace.cycles-pp.locks_alloc_lock.fcntl_setlk.do_fcntl.__x64_sys_fcntl.do_syscall_64 86.55 +5.4 91.97 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.__libc_fcntl64 7.45 +5.6 13.04 ± 2% perf-profile.calltrace.cycles-pp.kmem_cache_alloc.locks_alloc_lock.fcntl_setlk.do_fcntl.__x64_sys_fcntl 85.00 +6.1 91.07 perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_fcntl64 50.49 +6.3 56.80 perf-profile.calltrace.cycles-pp.do_lock_file_wait.fcntl_setlk.do_fcntl.__x64_sys_fcntl.do_syscall_64 0.00 +6.5 6.46 ± 2% perf-profile.calltrace.cycles-pp.obj_cgroup_charge_pages.obj_cgroup_charge.kmem_cache_alloc.locks_alloc_lock.posix_lock_inode 0.00 +7.7 7.69 ± 2% perf-profile.calltrace.cycles-pp.obj_cgroup_charge.kmem_cache_alloc.locks_alloc_lock.posix_lock_inode.do_lock_file_wait 19.77 +7.9 27.69 perf-profile.calltrace.cycles-pp.locks_alloc_lock.posix_lock_inode.do_lock_file_wait.fcntl_setlk.do_fcntl 79.14 +8.6 87.73 perf-profile.calltrace.cycles-pp.__x64_sys_fcntl.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_fcntl64 16.98 +8.6 25.60 perf-profile.calltrace.cycles-pp.kmem_cache_alloc.locks_alloc_lock.posix_lock_inode.do_lock_file_wait.fcntl_setlk 76.05 +9.9 85.92 perf-profile.calltrace.cycles-pp.do_fcntl.__x64_sys_fcntl.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_fcntl64 41.83 +9.9 51.72 perf-profile.calltrace.cycles-pp.posix_lock_inode.do_lock_file_wait.fcntl_setlk.do_fcntl.__x64_sys_fcntl 0.00 +10.1 10.08 ± 5% perf-profile.calltrace.cycles-pp.page_counter_try_charge.obj_cgroup_charge_pages.obj_cgroup_charge.kmem_cache_alloc.locks_alloc_lock 68.46 +13.2 81.70 perf-profile.calltrace.cycles-pp.fcntl_setlk.do_fcntl.__x64_sys_fcntl.do_syscall_64.entry_SYSCALL_64_after_hwframe 8.83 -4.0 4.81 perf-profile.children.cycles-pp.memset_erms 5.85 -2.6 3.28 perf-profile.children.cycles-pp._copy_from_user 5.38 -2.3 3.12 perf-profile.children.cycles-pp.security_file_lock 4.93 -2.1 2.84 perf-profile.children.cycles-pp.__entry_text_start 4.84 -2.0 2.84 perf-profile.children.cycles-pp.syscall_return_via_sysret 4.26 -1.9 2.40 perf-profile.children.cycles-pp.syscall_exit_to_user_mode 4.01 -1.8 2.24 perf-profile.children.cycles-pp.__might_sleep 4.00 -1.7 2.33 perf-profile.children.cycles-pp.common_file_perm 3.88 -1.7 2.22 perf-profile.children.cycles-pp.___might_sleep 2.46 -1.1 1.36 ± 2% perf-profile.children.cycles-pp.locks_release_private 2.37 -1.0 1.36 perf-profile.children.cycles-pp.copy_user_generic_unrolled 2.20 -0.9 1.26 perf-profile.children.cycles-pp.__cond_resched 1.48 -0.7 0.82 ± 2% perf-profile.children.cycles-pp.__might_fault 1.38 -0.6 0.76 perf-profile.children.cycles-pp.aa_file_perm 1.37 -0.6 0.78 ± 2% perf-profile.children.cycles-pp.copy_user_enhanced_fast_string 1.16 -0.5 0.62 ± 3% perf-profile.children.cycles-pp.flock64_to_posix_lock 1.17 -0.5 0.65 ± 2% perf-profile.children.cycles-pp.exit_to_user_mode_prepare 1.40 -0.5 0.88 perf-profile.children.cycles-pp.__fget_light 1.10 -0.5 0.62 perf-profile.children.cycles-pp.rcu_all_qs 0.96 -0.5 0.50 ± 2% perf-profile.children.cycles-pp.__init_waitqueue_head 0.94 -0.4 0.54 perf-profile.children.cycles-pp.syscall_enter_from_user_mode 0.71 ± 2% -0.3 0.37 ± 2% perf-profile.children.cycles-pp.syscall_exit_to_user_mode_prepare 0.74 -0.3 0.40 perf-profile.children.cycles-pp.apparmor_file_lock 0.84 ± 2% -0.3 0.52 ± 8% perf-profile.children.cycles-pp.testcase 0.61 -0.3 0.31 ± 2% perf-profile.children.cycles-pp.__list_del_entry_valid 0.62 -0.3 0.34 ± 2% perf-profile.children.cycles-pp.memset 0.66 -0.3 0.40 ± 3% perf-profile.children.cycles-pp.should_failslab 0.61 -0.3 0.35 ± 2% perf-profile.children.cycles-pp.locks_copy_lock 0.56 -0.3 0.31 perf-profile.children.cycles-pp.locks_get_lock_context 0.48 -0.2 0.26 perf-profile.children.cycles-pp.security_file_fcntl 2.06 -0.2 1.86 ± 5% perf-profile.children.cycles-pp.locks_delete_lock_ctx 0.42 -0.2 0.22 ± 3% perf-profile.children.cycles-pp.__list_add_valid 0.38 -0.2 0.21 ± 3% perf-profile.children.cycles-pp.vfs_lock_file 0.31 ± 2% -0.2 0.16 ± 3% perf-profile.children.cycles-pp.locks_move_blocks 0.32 ± 2% -0.1 0.18 ± 3% perf-profile.children.cycles-pp.locks_copy_conflock 0.71 -0.1 0.56 perf-profile.children.cycles-pp.locks_delete_block 0.34 -0.1 0.23 perf-profile.children.cycles-pp.entry_SYSCALL_64_safe_stack 0.25 -0.1 0.14 perf-profile.children.cycles-pp.__fdget_raw 0.22 -0.1 0.12 ± 8% perf-profile.children.cycles-pp.fcntl@plt 0.00 +0.1 0.05 perf-profile.children.cycles-pp.new_sync_write 0.00 +0.1 0.05 perf-profile.children.cycles-pp.try_charge_memcg 0.00 +0.1 0.05 perf-profile.children.cycles-pp.generic_file_write_iter 0.00 +0.1 0.05 perf-profile.children.cycles-pp.__generic_file_write_iter 0.00 +0.1 0.05 perf-profile.children.cycles-pp.generic_perform_write 0.00 +0.1 0.06 ± 9% perf-profile.children.cycles-pp.refill_stock 0.00 +0.1 0.09 ± 10% perf-profile.children.cycles-pp.get_mem_cgroup_from_objcg 99.04 +0.3 99.30 perf-profile.children.cycles-pp.__libc_fcntl64 0.00 +0.3 0.34 perf-profile.children.cycles-pp.obj_cgroup_uncharge 3.45 +0.7 4.12 ± 4% perf-profile.children.cycles-pp.locks_dispose_list 0.00 +0.8 0.84 ± 15% perf-profile.children.cycles-pp.mem_cgroup_from_task 3.84 +1.5 5.34 ± 7% perf-profile.children.cycles-pp._raw_spin_lock 0.28 ± 2% +1.9 2.16 perf-profile.children.cycles-pp.rcu_read_unlock_strict 0.00 +2.0 1.97 perf-profile.children.cycles-pp.refill_obj_stock 0.00 +3.9 3.94 perf-profile.children.cycles-pp.propagate_protected_usage 0.00 +5.0 5.03 ± 8% perf-profile.children.cycles-pp.page_counter_cancel 0.00 +5.3 5.32 perf-profile.children.cycles-pp.mod_objcg_state 86.79 +5.4 92.16 perf-profile.children.cycles-pp.entry_SYSCALL_64_after_hwframe 85.64 +5.8 91.48 perf-profile.children.cycles-pp.do_syscall_64 51.02 +6.1 57.10 perf-profile.children.cycles-pp.do_lock_file_wait 0.00 +6.2 6.16 ± 2% perf-profile.children.cycles-pp.get_obj_cgroup_from_current 0.00 +6.4 6.44 ± 4% perf-profile.children.cycles-pp.page_counter_uncharge 0.00 +6.6 6.56 ± 4% perf-profile.children.cycles-pp.obj_cgroup_uncharge_pages 79.60 +8.4 87.99 perf-profile.children.cycles-pp.__x64_sys_fcntl 43.39 +9.2 52.57 perf-profile.children.cycles-pp.posix_lock_inode 76.44 +9.7 86.14 perf-profile.children.cycles-pp.do_fcntl 10.67 +9.8 20.44 perf-profile.children.cycles-pp.kmem_cache_free 0.00 +10.1 10.10 ± 5% perf-profile.children.cycles-pp.page_counter_try_charge 0.00 +10.2 10.23 ± 5% perf-profile.children.cycles-pp.obj_cgroup_charge_pages 0.00 +12.2 12.23 ± 4% perf-profile.children.cycles-pp.obj_cgroup_charge 29.38 +12.8 42.19 perf-profile.children.cycles-pp.locks_alloc_lock 69.21 +12.9 82.11 perf-profile.children.cycles-pp.fcntl_setlk 26.20 +14.3 40.48 perf-profile.children.cycles-pp.kmem_cache_alloc 8.51 -3.9 4.63 perf-profile.self.cycles-pp.memset_erms 5.30 -2.4 2.90 ± 2% perf-profile.self.cycles-pp.posix_lock_inode 5.26 -2.2 3.06 perf-profile.self.cycles-pp.__libc_fcntl64 4.80 -2.0 2.82 perf-profile.self.cycles-pp.syscall_return_via_sysret 3.60 -1.6 2.02 ± 3% perf-profile.self.cycles-pp.fcntl_setlk 3.41 -1.5 1.95 perf-profile.self.cycles-pp.___might_sleep 3.12 -1.4 1.76 perf-profile.self.cycles-pp.__might_sleep 10.28 -1.3 9.02 perf-profile.self.cycles-pp.kmem_cache_free 2.51 -1.2 1.36 ± 3% perf-profile.self.cycles-pp.locks_alloc_lock 2.66 -1.1 1.60 perf-profile.self.cycles-pp.common_file_perm 2.39 -1.0 1.38 perf-profile.self.cycles-pp.syscall_exit_to_user_mode 2.16 -1.0 1.19 ± 2% perf-profile.self.cycles-pp.locks_release_private 2.20 -0.9 1.26 perf-profile.self.cycles-pp.copy_user_generic_unrolled 1.78 -0.7 1.06 ± 2% perf-profile.self.cycles-pp.__entry_text_start 1.56 -0.7 0.85 perf-profile.self.cycles-pp.do_fcntl 1.46 -0.6 0.83 perf-profile.self.cycles-pp.do_lock_file_wait 1.30 -0.6 0.73 perf-profile.self.cycles-pp.__x64_sys_fcntl 1.18 -0.5 0.67 ± 2% perf-profile.self.cycles-pp.copy_user_enhanced_fast_string 1.26 -0.5 0.74 perf-profile.self.cycles-pp.entry_SYSCALL_64_after_hwframe 1.12 -0.5 0.62 perf-profile.self.cycles-pp.aa_file_perm 1.06 -0.5 0.57 ± 3% perf-profile.self.cycles-pp.flock64_to_posix_lock 1.31 -0.5 0.82 perf-profile.self.cycles-pp.__fget_light 1.07 -0.5 0.59 ± 2% perf-profile.self.cycles-pp.exit_to_user_mode_prepare 1.08 -0.5 0.62 perf-profile.self.cycles-pp.do_syscall_64 1.10 -0.5 0.64 perf-profile.self.cycles-pp.__cond_resched 0.88 -0.4 0.48 perf-profile.self.cycles-pp._copy_from_user 0.80 -0.4 0.40 ± 2% perf-profile.self.cycles-pp.locks_dispose_list 0.81 -0.3 0.47 perf-profile.self.cycles-pp.syscall_enter_from_user_mode 0.75 -0.3 0.42 perf-profile.self.cycles-pp.rcu_all_qs 0.75 -0.3 0.42 ± 5% perf-profile.self.cycles-pp.security_file_lock 0.68 -0.3 0.36 perf-profile.self.cycles-pp.locks_insert_lock_ctx 0.63 -0.3 0.34 perf-profile.self.cycles-pp.__init_waitqueue_head 0.60 ± 2% -0.3 0.32 ± 3% perf-profile.self.cycles-pp.syscall_exit_to_user_mode_prepare 0.61 -0.3 0.32 ± 2% perf-profile.self.cycles-pp.locks_unlink_lock_ctx 0.74 ± 2% -0.3 0.46 ± 8% perf-profile.self.cycles-pp.testcase 0.57 -0.3 0.32 ± 2% perf-profile.self.cycles-pp.apparmor_file_lock 0.49 -0.2 0.25 perf-profile.self.cycles-pp.__list_del_entry_valid 0.38 ± 2% -0.2 0.20 ± 2% perf-profile.self.cycles-pp.locks_get_lock_context 0.38 -0.2 0.21 ± 2% perf-profile.self.cycles-pp.locks_delete_lock_ctx 0.38 -0.2 0.20 ± 2% perf-profile.self.cycles-pp.security_file_fcntl 0.32 -0.2 0.16 ± 3% perf-profile.self.cycles-pp.__list_add_valid 0.30 -0.1 0.16 ± 2% perf-profile.self.cycles-pp.__might_fault 0.26 -0.1 0.12 ± 4% perf-profile.self.cycles-pp.locks_move_blocks 0.28 -0.1 0.16 ± 4% perf-profile.self.cycles-pp.vfs_lock_file 0.34 -0.1 0.22 ± 3% perf-profile.self.cycles-pp.should_failslab 0.28 -0.1 0.16 ± 3% perf-profile.self.cycles-pp.locks_copy_lock 0.28 -0.1 0.16 ± 3% perf-profile.self.cycles-pp.locks_copy_conflock 0.27 ± 2% -0.1 0.16 ± 4% perf-profile.self.cycles-pp.memset 0.33 ± 2% -0.1 0.22 perf-profile.self.cycles-pp.entry_SYSCALL_64_safe_stack 0.51 -0.1 0.45 perf-profile.self.cycles-pp.locks_delete_block 0.12 ± 4% -0.1 0.06 ± 14% perf-profile.self.cycles-pp.fcntl@plt 0.09 ± 5% -0.0 0.04 ± 44% perf-profile.self.cycles-pp.__fdget_raw 0.00 +0.1 0.05 ± 8% perf-profile.self.cycles-pp.refill_stock 0.00 +0.1 0.08 ± 8% perf-profile.self.cycles-pp.get_mem_cgroup_from_objcg 0.00 +0.2 0.15 perf-profile.self.cycles-pp.obj_cgroup_uncharge 0.00 +0.7 0.68 ± 18% perf-profile.self.cycles-pp.mem_cgroup_from_task 0.10 ± 3% +0.7 0.83 ± 2% perf-profile.self.cycles-pp.rcu_read_unlock_strict 3.58 +1.6 5.18 ± 7% perf-profile.self.cycles-pp._raw_spin_lock 0.00 +1.8 1.82 perf-profile.self.cycles-pp.refill_obj_stock 0.00 +1.8 1.82 perf-profile.self.cycles-pp.obj_cgroup_charge 0.00 +3.9 3.90 perf-profile.self.cycles-pp.propagate_protected_usage 0.00 +5.0 4.98 ± 2% perf-profile.self.cycles-pp.get_obj_cgroup_from_current 0.00 +5.0 5.00 ± 8% perf-profile.self.cycles-pp.page_counter_cancel 0.00 +5.0 5.00 perf-profile.self.cycles-pp.mod_objcg_state 0.00 +7.5 7.52 ± 6% perf-profile.self.cycles-pp.page_counter_try_charge 3500 +--------------------------------------------------------------------+ | | | O | 3000 |-+ + | | + + | |.++.+.+.++.+.+.++.+.++.+.+.++.+.+.++.+.++.+.+.++ ++.+.+.++.+.+.++.| 2500 |-+ | | | 2000 |-OO O O OO O O O O O OO O O O OO OO O O O O | | O O O O O O O O | | | 1500 |-+ | | | | O O | 1000 +--------------------------------------------------------------------+ 115 +---------------------------------------------------------------------+ 110 |-+ : + + | | :: : : .+. | 105 |-+ : : : : : : + + | 100 |-+ + : : : : : : + : : | | +. + : + + : : : : : : :: : + + +. +.| 95 |.+.+ +.+.+ + : + : : :+.+ :+ :.+ : .+ : + + :+ +.+ | 90 |-+ + :: + + + +.+ + +.+ + | 85 |-+ + O | | | 80 |-+ O O O O | 75 |-+ O O O O O O O O OO O | | O O O O O O O O O O O O O O O O | 70 |-+ O O O O O | 65 +---------------------------------------------------------------------+ 380 +---------------------------------------------------------------------+ | O O O O O O O O O O O O | 360 |-+ O O OO O O O O O O O O O | 340 |-+ O OO O O O | | | 320 |-+ | | | 300 |-+ + .+ .+. | |.+.+ .+.+ .+ +: .+ ++ +.+ + + + .+ ++.+.+.+ | 280 |-+ :.+ + + .++ : ++ : :: : :+ + : + +.| 260 |-+ + + : : : : : : + : .+ : | | : : : : : : + +: | 240 |-+ :: : : + | | + + + | 220 +---------------------------------------------------------------------+ 3500 +--------------------------------------------------------------------+ | | | O | 3000 |-+ + | | + + | |.++.+.+.++.+.+.++.+.++.+.+.++.+.+.++.+.++.+.+.++ ++.+.+.++.+.+.++.| 2500 |-+ | | | 2000 |-OO O O OO O O O O O OO O O O OO OO O O O O | | O O O O O O O O | | | 1500 |-+ | | | | O O | 1000 +--------------------------------------------------------------------+ 115 +---------------------------------------------------------------------+ 110 |-+ : + + | | :: : : .+. | 105 |-+ : : : : : : + + | 100 |-+ + : : : : : : + : : | | +. + : + + : : : : : : :: : + + +. +.| 95 |.+.+ +.+.+ + : + : : :+.+ :+ :.+ : .+ : + + :+ +.+ | 90 |-+ + :: + + + +.+ + +.+ + | 85 |-+ + O | | | 80 |-+ O O O O | 75 |-+ O O O O O O O O OO O | | O O O O O O O O O O O O O O O O | 70 |-+ O O O O O | 65 +---------------------------------------------------------------------+ 800 +---------------------------------------------------------------------+ | +.| 700 |-+ + : | 600 |-+ :: .+.: | |.+. : : .+ .+ + | 500 |-+ + .+.+ +.+.+. .+ .+. .+ + +. .+. .+.++ | | +. .+.++ ++.+.+ +.+ ++ + + ++.+ | 400 |-+ + | | | 300 |-+ | 200 |-+ O O O O OO O O O O O O O O O O O O | | O O O O O O O O O O OO O O O O | 100 |-+ | | | 0 +---------------------------------------------------------------------+ will-it-scale.96.processes 1.6e+08 +-----------------------------------------------------------------+ |.++.+.++.+.++.+.++.++.+.++.+.++.+.++.+.++. .+.++. | 1.5e+08 |-+ +.++ ++.+.++.+.++.| 1.4e+08 |-+ | | | 1.3e+08 |-+ | | | 1.2e+08 |-+ | | | 1.1e+08 |-+ | 1e+08 |-+ | | | 9e+07 |-OO O OO O OO O OO OO O OO O OO O OO O OO O OO O OO OO O OO O | | | 8e+07 +-----------------------------------------------------------------+ will-it-scale.per_process_ops 1.7e+06 +-----------------------------------------------------------------+ |.++.+.+ .+.++. | 1.6e+06 |-+ +.+.++.+.++.++ +.++.+.++.+.++.+.++.+.++.++.+.++.+.++.| | | 1.5e+06 |-+ | 1.4e+06 |-+ | | | 1.3e+06 |-+ | | | 1.2e+06 |-+ | 1.1e+06 |-+ | | | 1e+06 |-+ | | OO O OO O OO O OO OO O OO O OO O OO O OO O OO O OO OO O OO O | 900000 +-----------------------------------------------------------------+ will-it-scale.workload 1.6e+08 +-----------------------------------------------------------------+ |.++.+.++.+.++.+.++.++.+.++.+.++.+.++.+.++. .+.++. | 1.5e+08 |-+ +.++ ++.+.++.+.++.| 1.4e+08 |-+ | | | 1.3e+08 |-+ | | | 1.2e+08 |-+ | | | 1.1e+08 |-+ | 1e+08 |-+ | | | 9e+07 |-OO O OO O OO O OO OO O OO O OO O OO O OO O OO O OO OO O OO O | | | 8e+07 +-----------------------------------------------------------------+ [*] bisect-good sample [O] bisect-bad sample *************************************************************************************************** lkp-icl-2sp2: 128 threads 2 sockets Intel(R) Xeon(R) Gold 6338 CPU @ 2.00GHz with 256G memory ========================================================================================= compiler/cpufreq_governor/kconfig/mode/nr_task/rootfs/tbox_group/test/testcase/ucode: gcc-9/performance/x86_64-rhel-8.3/process/100%/debian-10.4-x86_64-20200603.cgz/lkp-icl-2sp2/lock1/will-it-scale/0xd000280 commit: fa4e6b1ad5 ("memcg: enable accounting for pollfd and select bits arrays") 059dd9003a ("memcg: enable accounting for file lock caches") fa4e6b1ad57df096 059dd9003ad894ad7b2e47126cd ---------------- --------------------------- %stddev %change %stddev \ | \ 1.85e+08 -42.9% 1.056e+08 will-it-scale.128.processes 1445470 -42.9% 825255 will-it-scale.per_process_ops 1.85e+08 -42.9% 1.056e+08 will-it-scale.workload 45587 ± 30% +202.7% 138001 ± 62% numa-meminfo.node1.AnonPages 11399 ± 30% +202.5% 34478 ± 62% numa-vmstat.node1.nr_anon_pages 0.01 ± 10% -0.0 0.01 ± 3% mpstat.cpu.all.soft% 9.40 -4.3 5.08 mpstat.cpu.all.usr% 46636 ± 97% +109.8% 97839 proc-vmstat.nr_active_anon 690907 ± 7% +8.3% 748337 proc-vmstat.nr_file_pages 54066 ± 89% +99.7% 107950 proc-vmstat.nr_shmem 46636 ± 97% +109.8% 97839 proc-vmstat.nr_zone_active_anon 9277 -91.5% 790.83 ± 12% slabinfo.Acpi-Parse.active_objs 126.83 -91.5% 10.83 ± 12% slabinfo.Acpi-Parse.active_slabs 9277 -91.5% 790.83 ± 12% slabinfo.Acpi-Parse.num_objs 126.83 -91.5% 10.83 ± 12% slabinfo.Acpi-Parse.num_slabs 698.17 ± 27% +207.2% 2144 ± 67% interrupts.CPU107.CAL:Function_call_interrupts 715.50 ± 38% +138.9% 1709 ± 48% interrupts.CPU109.CAL:Function_call_interrupts 583.67 ± 38% +60.3% 935.83 ± 13% interrupts.CPU120.CAL:Function_call_interrupts 581.83 ± 40% +85.1% 1076 ± 42% interrupts.CPU122.CAL:Function_call_interrupts 342.83 ± 31% +237.6% 1157 ± 61% interrupts.CPU124.RES:Rescheduling_interrupts 584.50 ± 33% +163.5% 1540 ± 55% interrupts.CPU17.CAL:Function_call_interrupts 547.50 ± 44% +76.9% 968.33 ± 22% interrupts.CPU23.CAL:Function_call_interrupts 442.83 ± 71% +172.9% 1208 ± 26% interrupts.CPU23.RES:Rescheduling_interrupts 776.17 ± 22% +129.5% 1781 ± 33% interrupts.CPU36.RES:Rescheduling_interrupts 772.17 ± 49% +142.5% 1872 ± 49% interrupts.CPU42.RES:Rescheduling_interrupts 612.67 ± 30% +93.1% 1183 ± 46% interrupts.CPU51.CAL:Function_call_interrupts 1017 ± 59% +117.4% 2212 ± 43% interrupts.CPU77.CAL:Function_call_interrupts 678.00 ± 30% +70.5% 1156 ± 51% interrupts.CPU85.CAL:Function_call_interrupts 0.01 ±100% +2194.4% 0.27 ± 2% perf-stat.i.MPKI 0.03 ±100% +0.1 0.08 ± 5% perf-stat.i.branch-miss-rate% 17154527 ±100% +199.5% 51380873 ± 4% perf-stat.i.branch-misses 4.60 ±100% +19.7 24.32 ± 4% perf-stat.i.cache-miss-rate% 250616 ±100% +8128.4% 20621644 ± 2% perf-stat.i.cache-misses 2658612 ±100% +3108.7% 85307043 ± 2% perf-stat.i.cache-references 0.45 ±100% +122.0% 0.99 perf-stat.i.cpi 21.15 ±100% +3655.9% 794.38 perf-stat.i.metric.K/sec 47456 ±100% +16815.1% 8027297 perf-stat.i.node-load-misses 16971 ±100% +8483.1% 1456700 ± 11% perf-stat.i.node-loads 20.86 ±101% +73.7 94.54 perf-stat.i.node-store-miss-rate% 22206 ±101% +33753.7% 7517713 ± 7% perf-stat.i.node-store-misses 36448 ±100% +1060.5% 422977 ± 4% perf-stat.i.node-stores 0.01 ±100% +3411.1% 0.26 ± 2% perf-stat.overall.MPKI 0.02 ±100% +0.1 0.08 ± 4% perf-stat.overall.branch-miss-rate% 4.71 ±100% +19.5 24.19 ± 4% perf-stat.overall.cache-miss-rate% 0.45 ±100% +121.4% 0.99 perf-stat.overall.cpi 35.88 ±100% +48.7 84.63 ± 2% perf-stat.overall.node-load-miss-rate% 18.89 ±101% +75.8 94.64 perf-stat.overall.node-store-miss-rate% 296973 ±100% +214.5% 934077 perf-stat.overall.path-length 17097545 ±100% +199.5% 51211786 ± 4% perf-stat.ps.branch-misses 253875 ±100% +7998.7% 20560626 ± 2% perf-stat.ps.cache-misses 2695172 ±100% +3056.8% 85081638 ± 2% perf-stat.ps.cache-references 47799 ±100% +16638.3% 8000747 perf-stat.ps.node-load-misses 18828 ±100% +7627.2% 1454946 ± 11% perf-stat.ps.node-loads 22114 ±101% +33783.4% 7493201 ± 7% perf-stat.ps.node-store-misses 36368 ±100% +1059.8% 421810 ± 4% perf-stat.ps.node-stores 0.00 ±100% +125.9% 0.01 ± 3% perf-sched.sch_delay.avg.ms.__x64_sys_pause.do_syscall_64.entry_SYSCALL_64_after_hwframe.[unknown] 0.00 ± 99% +141.7% 0.01 ± 4% perf-sched.sch_delay.avg.ms.do_nanosleep.hrtimer_nanosleep.__x64_sys_nanosleep.do_syscall_64 0.01 ±100% +153.1% 0.01 ± 5% perf-sched.sch_delay.avg.ms.rcu_gp_kthread.kthread.ret_from_fork 0.00 ±100% +138.1% 0.01 ± 5% perf-sched.sch_delay.avg.ms.schedule_timeout.rcu_gp_kthread.kthread.ret_from_fork 0.01 ±100% +154.8% 0.01 ± 18% perf-sched.sch_delay.max.ms.__x64_sys_pause.do_syscall_64.entry_SYSCALL_64_after_hwframe.[unknown] 0.01 ±100% +161.0% 0.03 ± 16% perf-sched.sch_delay.max.ms.do_wait.kernel_wait4.__do_sys_wait4.do_syscall_64 0.01 ±100% +406.9% 0.07 ±121% perf-sched.sch_delay.max.ms.preempt_schedule_common.__cond_resched.wait_for_completion.affine_move_task.__set_cpus_allowed_ptr 0.01 ±101% +148.9% 0.02 ± 14% perf-sched.sch_delay.max.ms.rcu_gp_kthread.kthread.ret_from_fork 0.18 ±216% +287.3% 0.70 ± 72% perf-sched.sch_delay.max.ms.schedule_hrtimeout_range_clock.ep_poll.do_epoll_wait.__x64_sys_epoll_wait 0.31 ±220% +260.5% 1.13 ± 98% perf-sched.sch_delay.max.ms.schedule_hrtimeout_range_clock.poll_schedule_timeout.constprop.0.do_sys_poll 0.00 ±100% +196.6% 0.01 ± 14% perf-sched.sch_delay.max.ms.schedule_timeout.kcompactd.kthread.ret_from_fork 0.01 ±100% +186.5% 0.02 ± 21% perf-sched.sch_delay.max.ms.schedule_timeout.rcu_gp_kthread.kthread.ret_from_fork 0.03 ±100% +187.4% 0.08 ± 23% perf-sched.wait_and_delay.avg.ms.preempt_schedule_common.__cond_resched.posix_lock_inode.do_lock_file_wait.fcntl_setlk 202.00 ±100% +141.4% 487.67 ± 11% perf-sched.wait_and_delay.count.pipe_read.new_sync_read.vfs_read.ksys_read 2034 ±105% +183.2% 5761 ± 13% perf-sched.wait_and_delay.count.preempt_schedule_common.__cond_resched.kmem_cache_alloc.locks_alloc_lock.posix_lock_inode 594.00 ±104% +303.3% 2395 ± 23% perf-sched.wait_and_delay.count.preempt_schedule_common.__cond_resched.posix_lock_inode.do_lock_file_wait.fcntl_setlk 7.50 ±126% +253.3% 26.50 ± 35% perf-sched.wait_and_delay.count.schedule_hrtimeout_range_clock.ep_poll.do_epoll_wait.__x64_sys_epoll_wait 123.00 ±100% +212.3% 384.17 ± 4% perf-sched.wait_and_delay.count.schedule_hrtimeout_range_clock.poll_schedule_timeout.constprop.0.do_sys_poll 944.33 ±100% +129.5% 2167 ± 2% perf-sched.wait_and_delay.count.schedule_timeout.rcu_gp_kthread.kthread.ret_from_fork 8.72 ±104% +323.2% 36.89 ± 41% perf-sched.wait_and_delay.max.ms.preempt_schedule_common.__cond_resched.kmem_cache_alloc.locks_alloc_lock.posix_lock_inode 4.15 ±127% +271.7% 15.44 ± 22% perf-sched.wait_and_delay.max.ms.preempt_schedule_common.__cond_resched.posix_lock_inode.do_lock_file_wait.fcntl_setlk 0.44 ±142% +3436.9% 15.53 ±203% perf-sched.wait_time.avg.ms.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_exc_page_fault.[unknown] 0.07 ±101% +276.4% 0.27 ± 38% perf-sched.wait_time.avg.ms.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_sysvec_apic_timer_interrupt.[unknown] 3.02 ±152% +13152.1% 399.82 ±127% perf-sched.wait_time.avg.ms.preempt_schedule_common.__cond_resched.__alloc_pages.alloc_pages_vma.shmem_alloc_page 0.03 ±100% +187.4% 0.08 ± 23% perf-sched.wait_time.avg.ms.preempt_schedule_common.__cond_resched.posix_lock_inode.do_lock_file_wait.fcntl_setlk 0.00 ±223% +609.1% 0.01 ± 27% perf-sched.wait_time.avg.ms.schedule_timeout.wait_for_completion.affine_move_task.__set_cpus_allowed_ptr 3.92 ±142% +4443.1% 177.91 ±209% perf-sched.wait_time.max.ms.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_exc_page_fault.[unknown] 6.36 ±100% +24150.4% 1543 ±132% perf-sched.wait_time.max.ms.preempt_schedule_common.__cond_resched.__alloc_pages.alloc_pages_vma.shmem_alloc_page 8.72 ±104% +323.2% 36.89 ± 41% perf-sched.wait_time.max.ms.preempt_schedule_common.__cond_resched.kmem_cache_alloc.locks_alloc_lock.posix_lock_inode 4.15 ±127% +271.7% 15.44 ± 22% perf-sched.wait_time.max.ms.preempt_schedule_common.__cond_resched.posix_lock_inode.do_lock_file_wait.fcntl_setlk 0.00 ±223% +618.2% 0.01 ± 26% perf-sched.wait_time.max.ms.schedule_timeout.wait_for_completion.affine_move_task.__set_cpus_allowed_ptr 0.00 +0.6 0.63 ± 8% perf-profile.calltrace.cycles-pp.propagate_protected_usage.page_counter_uncharge.obj_cgroup_uncharge_pages.kmem_cache_free.posix_lock_inode 0.38 ±100% +0.6 1.01 ± 7% perf-profile.calltrace.cycles-pp._raw_spin_lock.fcntl_setlk.do_fcntl.__x64_sys_fcntl.do_syscall_64 0.00 +0.6 0.64 ± 2% perf-profile.calltrace.cycles-pp.refill_obj_stock.kmem_cache_free.fcntl_setlk.do_fcntl.__x64_sys_fcntl 0.00 +0.7 0.73 ± 23% perf-profile.calltrace.cycles-pp.page_counter_cancel.page_counter_uncharge.obj_cgroup_uncharge_pages.kmem_cache_free.locks_dispose_list 0.37 ±100% +0.8 1.22 ± 11% perf-profile.calltrace.cycles-pp._raw_spin_lock.locks_unlink_lock_ctx.locks_delete_lock_ctx.posix_lock_inode.do_lock_file_wait 0.00 +0.9 0.86 ± 3% perf-profile.calltrace.cycles-pp.mod_objcg_state.kmem_cache_free.fcntl_setlk.do_fcntl.__x64_sys_fcntl 0.00 +0.9 0.88 ± 4% perf-profile.calltrace.cycles-pp.mod_objcg_state.kmem_cache_alloc.locks_alloc_lock.fcntl_setlk.do_fcntl 0.38 ±100% +0.9 1.27 ± 11% perf-profile.calltrace.cycles-pp._raw_spin_lock.locks_insert_lock_ctx.posix_lock_inode.do_lock_file_wait.fcntl_setlk 0.77 ±100% +0.9 1.68 ± 7% perf-profile.calltrace.cycles-pp.locks_insert_lock_ctx.posix_lock_inode.do_lock_file_wait.fcntl_setlk.do_fcntl 0.00 +0.9 0.92 ± 22% perf-profile.calltrace.cycles-pp.page_counter_uncharge.obj_cgroup_uncharge_pages.kmem_cache_free.locks_dispose_list.posix_lock_inode 0.00 +0.9 0.94 ± 22% perf-profile.calltrace.cycles-pp.obj_cgroup_uncharge_pages.kmem_cache_free.locks_dispose_list.posix_lock_inode.do_lock_file_wait 0.00 +1.0 0.97 ± 3% perf-profile.calltrace.cycles-pp.refill_obj_stock.kmem_cache_free.posix_lock_inode.do_lock_file_wait.fcntl_setlk 0.00 +1.3 1.35 ± 3% perf-profile.calltrace.cycles-pp.mod_objcg_state.kmem_cache_free.posix_lock_inode.do_lock_file_wait.fcntl_setlk 0.77 ±100% +1.7 2.43 ± 11% perf-profile.calltrace.cycles-pp._raw_spin_lock.posix_lock_inode.do_lock_file_wait.fcntl_setlk.do_fcntl 0.00 +1.8 1.79 ± 4% perf-profile.calltrace.cycles-pp.mod_objcg_state.kmem_cache_alloc.locks_alloc_lock.posix_lock_inode.do_lock_file_wait 0.00 +1.8 1.84 ± 15% perf-profile.calltrace.cycles-pp.page_counter_cancel.page_counter_uncharge.obj_cgroup_uncharge_pages.kmem_cache_free.fcntl_setlk 0.00 +2.0 1.99 ± 6% perf-profile.calltrace.cycles-pp.get_obj_cgroup_from_current.kmem_cache_alloc.locks_alloc_lock.fcntl_setlk.do_fcntl 0.93 ±100% +2.3 3.20 ± 5% perf-profile.calltrace.cycles-pp.kmem_cache_free.locks_dispose_list.posix_lock_inode.do_lock_file_wait.fcntl_setlk 0.00 +2.3 2.32 ± 14% perf-profile.calltrace.cycles-pp.page_counter_uncharge.obj_cgroup_uncharge_pages.kmem_cache_free.fcntl_setlk.do_fcntl 0.00 +2.4 2.36 ± 14% perf-profile.calltrace.cycles-pp.obj_cgroup_uncharge_pages.kmem_cache_free.fcntl_setlk.do_fcntl.__x64_sys_fcntl 1.57 ±100% +2.4 3.94 ± 4% perf-profile.calltrace.cycles-pp.locks_dispose_list.posix_lock_inode.do_lock_file_wait.fcntl_setlk.do_fcntl 0.00 +2.5 2.45 ± 5% perf-profile.calltrace.cycles-pp.propagate_protected_usage.page_counter_try_charge.obj_cgroup_charge_pages.obj_cgroup_charge.kmem_cache_alloc 0.00 +2.5 2.49 ± 11% perf-profile.calltrace.cycles-pp.page_counter_cancel.page_counter_uncharge.obj_cgroup_uncharge_pages.kmem_cache_free.posix_lock_inode 0.00 +3.1 3.12 ± 9% perf-profile.calltrace.cycles-pp.page_counter_uncharge.obj_cgroup_uncharge_pages.kmem_cache_free.posix_lock_inode.do_lock_file_wait 0.00 +3.2 3.18 ± 9% perf-profile.calltrace.cycles-pp.obj_cgroup_uncharge_pages.kmem_cache_free.posix_lock_inode.do_lock_file_wait.fcntl_setlk 0.00 +3.4 3.40 ± 12% perf-profile.calltrace.cycles-pp.obj_cgroup_charge_pages.obj_cgroup_charge.kmem_cache_alloc.locks_alloc_lock.fcntl_setlk 0.00 +3.6 3.59 ± 8% perf-profile.calltrace.cycles-pp.get_obj_cgroup_from_current.kmem_cache_alloc.locks_alloc_lock.posix_lock_inode.do_lock_file_wait 0.00 +4.1 4.06 ± 9% perf-profile.calltrace.cycles-pp.obj_cgroup_charge.kmem_cache_alloc.locks_alloc_lock.fcntl_setlk.do_fcntl 1.67 ±100% +5.1 6.72 ± 4% perf-profile.calltrace.cycles-pp.kmem_cache_free.fcntl_setlk.do_fcntl.__x64_sys_fcntl.do_syscall_64 0.00 +6.7 6.70 ± 5% perf-profile.calltrace.cycles-pp.obj_cgroup_charge_pages.obj_cgroup_charge.kmem_cache_alloc.locks_alloc_lock.posix_lock_inode 2.61 ±100% +7.3 9.89 ± 2% perf-profile.calltrace.cycles-pp.kmem_cache_free.posix_lock_inode.do_lock_file_wait.fcntl_setlk.do_fcntl 0.00 +8.0 8.01 ± 4% perf-profile.calltrace.cycles-pp.obj_cgroup_charge.kmem_cache_alloc.locks_alloc_lock.posix_lock_inode.do_lock_file_wait 4.16 ±100% +9.1 13.22 ± 2% perf-profile.calltrace.cycles-pp.kmem_cache_alloc.locks_alloc_lock.fcntl_setlk.do_fcntl.__x64_sys_fcntl 4.88 ±100% +9.4 14.32 perf-profile.calltrace.cycles-pp.locks_alloc_lock.fcntl_setlk.do_fcntl.__x64_sys_fcntl.do_syscall_64 0.00 +10.0 9.97 ± 7% perf-profile.calltrace.cycles-pp.page_counter_try_charge.obj_cgroup_charge_pages.obj_cgroup_charge.kmem_cache_alloc.locks_alloc_lock 8.66 ±100% +17.8 26.45 perf-profile.calltrace.cycles-pp.kmem_cache_alloc.locks_alloc_lock.posix_lock_inode.do_lock_file_wait.fcntl_setlk 10.13 ±100% +18.6 28.72 perf-profile.calltrace.cycles-pp.locks_alloc_lock.posix_lock_inode.do_lock_file_wait.fcntl_setlk.do_fcntl 21.41 ±100% +32.3 53.75 perf-profile.calltrace.cycles-pp.posix_lock_inode.do_lock_file_wait.fcntl_setlk.do_fcntl.__x64_sys_fcntl 25.56 ±100% +32.8 58.38 perf-profile.calltrace.cycles-pp.do_lock_file_wait.fcntl_setlk.do_fcntl.__x64_sys_fcntl.do_syscall_64 35.19 ±100% +48.4 83.61 perf-profile.calltrace.cycles-pp.fcntl_setlk.do_fcntl.__x64_sys_fcntl.do_syscall_64.entry_SYSCALL_64_after_hwframe 39.34 ±100% +48.8 88.10 perf-profile.calltrace.cycles-pp.do_fcntl.__x64_sys_fcntl.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_fcntl64 41.05 ±100% +49.0 90.05 perf-profile.calltrace.cycles-pp.__x64_sys_fcntl.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_fcntl64 43.20 ±100% +49.0 92.24 perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_fcntl64 44.04 ±100% +49.1 93.13 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.__libc_fcntl64 49.40 ±100% +49.8 99.18 perf-profile.calltrace.cycles-pp.__libc_fcntl64 0.03 ±103% +0.0 0.08 ± 16% perf-profile.children.cycles-pp.record__finish_output 0.03 ±103% +0.0 0.08 ± 16% perf-profile.children.cycles-pp.perf_session__process_events 0.00 +0.1 0.05 perf-profile.children.cycles-pp.try_charge_memcg 0.00 +0.1 0.05 perf-profile.children.cycles-pp.refill_stock 0.00 +0.1 0.08 ± 5% perf-profile.children.cycles-pp.get_mem_cgroup_from_objcg 0.08 ±100% +0.1 0.18 ± 2% perf-profile.children.cycles-pp.perf_callchain_kernel 0.11 ±100% +0.1 0.24 perf-profile.children.cycles-pp.perf_callchain 0.11 ±100% +0.1 0.24 perf-profile.children.cycles-pp.get_perf_callchain 0.12 ±100% +0.1 0.25 perf-profile.children.cycles-pp.perf_prepare_sample 0.13 ±100% +0.1 0.27 ± 3% perf-profile.children.cycles-pp.perf_swevent_overflow 0.13 ±100% +0.1 0.27 ± 3% perf-profile.children.cycles-pp.perf_trace_sched_stat_runtime 0.13 ±100% +0.1 0.27 ± 3% perf-profile.children.cycles-pp.__perf_event_overflow 0.13 ±100% +0.1 0.27 ± 3% perf-profile.children.cycles-pp.perf_event_output_forward 0.13 ±100% +0.1 0.28 ± 3% perf-profile.children.cycles-pp.perf_tp_event 0.14 ±100% +0.1 0.28 ± 2% perf-profile.children.cycles-pp.update_curr 0.16 ±100% +0.2 0.32 ± 2% perf-profile.children.cycles-pp.task_tick_fair 0.00 +0.4 0.38 ± 3% perf-profile.children.cycles-pp.obj_cgroup_uncharge 0.00 +0.9 0.88 ± 17% perf-profile.children.cycles-pp.mem_cgroup_from_task 0.00 +2.1 2.14 ± 2% perf-profile.children.cycles-pp.refill_obj_stock 0.16 ±100% +2.3 2.43 ± 5% perf-profile.children.cycles-pp.rcu_read_unlock_strict 1.72 ±100% +2.4 4.10 ± 3% perf-profile.children.cycles-pp.locks_dispose_list 0.00 +3.8 3.76 ± 4% perf-profile.children.cycles-pp.propagate_protected_usage 2.02 ±100% +4.0 6.06 ± 10% perf-profile.children.cycles-pp._raw_spin_lock 0.00 +5.1 5.07 ± 11% perf-profile.children.cycles-pp.page_counter_cancel 0.00 +5.7 5.66 ± 3% perf-profile.children.cycles-pp.mod_objcg_state 0.00 +6.4 6.38 ± 9% perf-profile.children.cycles-pp.page_counter_uncharge 0.00 +6.5 6.49 ± 9% perf-profile.children.cycles-pp.obj_cgroup_uncharge_pages 0.00 +6.5 6.51 ± 6% perf-profile.children.cycles-pp.get_obj_cgroup_from_current 0.00 +10.0 9.99 ± 7% perf-profile.children.cycles-pp.page_counter_try_charge 0.00 +10.1 10.12 ± 7% perf-profile.children.cycles-pp.obj_cgroup_charge_pages 0.00 +12.2 12.24 ± 5% perf-profile.children.cycles-pp.obj_cgroup_charge 5.37 ±100% +15.3 20.69 perf-profile.children.cycles-pp.kmem_cache_free 13.81 ±100% +27.9 41.71 perf-profile.children.cycles-pp.kmem_cache_alloc 15.49 ±100% +28.1 43.55 perf-profile.children.cycles-pp.locks_alloc_lock 22.23 ±100% +32.4 54.63 perf-profile.children.cycles-pp.posix_lock_inode 25.86 ±100% +32.9 58.71 perf-profile.children.cycles-pp.do_lock_file_wait 35.60 ±100% +48.5 84.06 perf-profile.children.cycles-pp.fcntl_setlk 39.56 ±100% +48.8 88.34 perf-profile.children.cycles-pp.do_fcntl 41.31 ±100% +49.0 90.33 perf-profile.children.cycles-pp.__x64_sys_fcntl 43.55 ±100% +49.1 92.66 perf-profile.children.cycles-pp.do_syscall_64 44.15 ±100% +49.2 93.30 perf-profile.children.cycles-pp.entry_SYSCALL_64_after_hwframe 49.51 ±100% +49.8 99.30 perf-profile.children.cycles-pp.__libc_fcntl64 0.00 +0.1 0.07 ± 8% perf-profile.self.cycles-pp.get_mem_cgroup_from_objcg 0.00 +0.2 0.17 ± 3% perf-profile.self.cycles-pp.obj_cgroup_uncharge 0.00 +0.7 0.70 ± 22% perf-profile.self.cycles-pp.mem_cgroup_from_task 0.06 ±100% +0.9 0.92 ± 2% perf-profile.self.cycles-pp.rcu_read_unlock_strict 0.00 +1.9 1.93 ± 4% perf-profile.self.cycles-pp.obj_cgroup_charge 0.00 +2.0 1.95 ± 2% perf-profile.self.cycles-pp.refill_obj_stock 0.00 +3.7 3.72 ± 4% perf-profile.self.cycles-pp.propagate_protected_usage 1.87 ±100% +4.0 5.88 ± 10% perf-profile.self.cycles-pp._raw_spin_lock 0.00 +5.0 5.03 ± 11% perf-profile.self.cycles-pp.page_counter_cancel 0.00 +5.3 5.27 ± 5% perf-profile.self.cycles-pp.get_obj_cgroup_from_current 0.00 +5.3 5.29 ± 3% perf-profile.self.cycles-pp.mod_objcg_state 0.00 +7.5 7.49 ± 8% perf-profile.self.cycles-pp.page_counter_try_charge 12800 ± 6% -31.2% 8812 ± 3% softirqs.CPU0.RCU 13416 ± 4% -32.4% 9075 ± 11% softirqs.CPU10.RCU 13811 ± 2% -37.9% 8574 softirqs.CPU100.RCU 13975 ± 2% -39.6% 8438 ± 9% softirqs.CPU101.RCU 13859 ± 3% -37.7% 8638 ± 3% softirqs.CPU102.RCU 13851 ± 5% -37.3% 8688 ± 4% softirqs.CPU103.RCU 13956 ± 2% -41.3% 8189 ± 12% softirqs.CPU104.RCU 13191 ± 11% -33.6% 8753 ± 3% softirqs.CPU105.RCU 13684 ± 2% -38.1% 8474 ± 2% softirqs.CPU106.RCU 13755 ± 2% -38.9% 8410 ± 2% softirqs.CPU107.RCU 13945 ± 2% -38.6% 8566 ± 2% softirqs.CPU108.RCU 13983 ± 2% -39.5% 8465 ± 2% softirqs.CPU109.RCU 13072 ± 6% -41.8% 7612 ± 21% softirqs.CPU11.RCU 13711 ± 2% -38.6% 8423 ± 3% softirqs.CPU110.RCU 13809 ± 3% -38.9% 8436 softirqs.CPU111.RCU 12448 -35.2% 8064 ± 3% softirqs.CPU112.RCU 12530 -37.4% 7847 ± 2% softirqs.CPU113.RCU 12226 ± 2% -36.7% 7744 ± 2% softirqs.CPU114.RCU 12396 -37.0% 7815 ± 2% softirqs.CPU115.RCU 12556 ± 4% -34.3% 8245 ± 5% softirqs.CPU116.RCU 12336 -37.5% 7706 ± 2% softirqs.CPU117.RCU 12324 ± 2% -36.4% 7839 ± 3% softirqs.CPU118.RCU 12385 ± 2% -36.7% 7842 ± 2% softirqs.CPU119.RCU 14037 ± 18% -40.1% 8411 ± 4% softirqs.CPU12.RCU 12469 ± 3% -37.5% 7797 ± 4% softirqs.CPU120.RCU 12579 ± 2% -36.8% 7953 ± 2% softirqs.CPU121.RCU 12248 -36.7% 7755 ± 5% softirqs.CPU122.RCU 12406 -37.4% 7765 ± 2% softirqs.CPU123.RCU 12318 ± 4% -36.3% 7852 ± 3% softirqs.CPU124.RCU 12302 -34.1% 8103 ± 7% softirqs.CPU125.RCU 12446 -36.6% 7895 ± 4% softirqs.CPU126.RCU 12710 ± 4% -34.0% 8385 ± 2% softirqs.CPU127.RCU 13037 ± 6% -35.7% 8387 ± 5% softirqs.CPU13.RCU 12279 ± 11% -31.4% 8421 ± 5% softirqs.CPU15.RCU 13662 ± 18% -38.0% 8468 ± 4% softirqs.CPU16.RCU 12807 ± 3% -35.1% 8309 ± 3% softirqs.CPU17.RCU 12922 ± 4% -35.9% 8281 ± 4% softirqs.CPU18.RCU 12688 ± 3% -31.6% 8673 ± 9% softirqs.CPU19.RCU 13269 ± 7% -33.7% 8801 ± 5% softirqs.CPU2.RCU 12736 ± 4% -27.8% 9198 ± 14% softirqs.CPU20.RCU 12599 ± 3% -33.7% 8351 ± 2% softirqs.CPU21.RCU 12971 ± 5% -34.8% 8457 ± 4% softirqs.CPU22.RCU 12929 ± 5% -28.3% 9272 ± 13% softirqs.CPU23.RCU 12656 ± 3% -33.1% 8463 ± 6% softirqs.CPU24.RCU 12381 ± 3% -31.7% 8454 ± 3% softirqs.CPU25.RCU 13212 ± 12% -36.9% 8334 ± 5% softirqs.CPU26.RCU 12859 ± 8% -35.7% 8273 ± 4% softirqs.CPU27.RCU 12672 ± 4% -33.4% 8436 ± 3% softirqs.CPU28.RCU 12821 ± 5% -35.8% 8225 ± 2% softirqs.CPU29.RCU 12638 ± 2% -34.5% 8277 ± 4% softirqs.CPU30.RCU 13077 ± 4% -35.5% 8432 ± 2% softirqs.CPU31.RCU 13497 ± 2% -35.6% 8695 ± 4% softirqs.CPU32.RCU 13436 -36.3% 8552 ± 2% softirqs.CPU33.RCU 13252 -36.9% 8361 softirqs.CPU34.RCU 13247 ± 2% -36.2% 8447 ± 2% softirqs.CPU35.RCU 13313 ± 3% -35.8% 8541 ± 2% softirqs.CPU36.RCU 13274 ± 2% -38.9% 8104 ± 9% softirqs.CPU37.RCU 13294 -36.9% 8392 ± 2% softirqs.CPU38.RCU 13965 ± 11% -38.1% 8645 ± 4% softirqs.CPU39.RCU 13455 ± 7% -33.1% 9005 ± 4% softirqs.CPU4.RCU 13260 ± 2% -37.2% 8330 ± 3% softirqs.CPU40.RCU 13098 ± 7% -35.6% 8437 ± 2% softirqs.CPU41.RCU 13357 ± 2% -37.2% 8394 softirqs.CPU42.RCU 13271 -37.0% 8354 ± 2% softirqs.CPU43.RCU 13378 ± 4% -37.3% 8393 ± 2% softirqs.CPU44.RCU 13385 ± 3% -37.2% 8406 ± 2% softirqs.CPU45.RCU 13321 ± 2% -30.3% 9283 ± 18% softirqs.CPU46.RCU 13325 ± 2% -33.4% 8870 ± 7% softirqs.CPU47.RCU 12571 -34.7% 8207 ± 5% softirqs.CPU48.RCU 12660 -36.8% 7998 ± 2% softirqs.CPU49.RCU 13419 ± 5% -32.1% 9115 ± 13% softirqs.CPU5.RCU 12295 ± 2% -34.4% 8071 ± 4% softirqs.CPU50.RCU 12623 ± 2% -37.0% 7948 ± 2% softirqs.CPU51.RCU 12544 -33.2% 8376 ± 4% softirqs.CPU52.RCU 12436 ± 3% -35.2% 8053 ± 5% softirqs.CPU53.RCU 12518 -35.5% 8073 ± 4% softirqs.CPU54.RCU 12414 ± 2% -34.9% 8085 ± 7% softirqs.CPU55.RCU 12657 ± 4% -36.5% 8036 ± 3% softirqs.CPU56.RCU 12582 -36.3% 8014 softirqs.CPU57.RCU 12316 -34.9% 8023 ± 4% softirqs.CPU58.RCU 12452 -34.3% 8186 ± 3% softirqs.CPU59.RCU 13498 ± 8% -36.3% 8594 ± 5% softirqs.CPU6.RCU 12411 ± 2% -35.0% 8063 ± 2% softirqs.CPU60.RCU 12462 -36.3% 7940 softirqs.CPU61.RCU 12660 ± 3% -36.8% 8005 ± 4% softirqs.CPU62.RCU 12779 ± 2% -35.0% 8307 ± 4% softirqs.CPU63.RCU 12543 ± 7% -26.2% 9253 ± 14% softirqs.CPU64.RCU 12829 ± 5% -30.3% 8937 ± 3% softirqs.CPU66.RCU 13484 ± 3% -31.9% 9182 ± 4% softirqs.CPU68.RCU 13068 ± 2% -33.1% 8746 ± 4% softirqs.CPU69.RCU 13209 ± 4% -38.0% 8192 ± 13% softirqs.CPU7.RCU 13112 ± 2% -33.3% 8741 ± 4% softirqs.CPU70.RCU 12940 -38.1% 8015 ± 11% softirqs.CPU71.RCU 13004 ± 2% -35.3% 8410 ± 4% softirqs.CPU72.RCU 13365 ± 2% -35.5% 8623 ± 16% softirqs.CPU73.RCU 13249 ± 3% -32.4% 8960 ± 9% softirqs.CPU74.RCU 12927 ± 3% -39.8% 7783 ± 13% softirqs.CPU75.RCU 12927 ± 3% -33.4% 8614 ± 4% softirqs.CPU76.RCU 13333 ± 2% -36.9% 8418 ± 3% softirqs.CPU77.RCU 12430 ± 10% -27.9% 8960 ± 7% softirqs.CPU79.RCU 12965 ± 6% -35.3% 8393 ± 5% softirqs.CPU8.RCU 13350 ± 6% -38.4% 8226 ± 2% softirqs.CPU80.RCU 12678 ± 3% -31.3% 8714 ± 4% softirqs.CPU81.RCU 13051 ± 4% -33.9% 8625 ± 3% softirqs.CPU82.RCU 12733 ± 6% -33.5% 8465 ± 6% softirqs.CPU83.RCU 12958 ± 5% -31.0% 8940 ± 11% softirqs.CPU84.RCU 13007 ± 4% -33.1% 8696 ± 6% softirqs.CPU85.RCU 12716 ± 2% -32.9% 8530 ± 7% softirqs.CPU86.RCU 13032 ± 3% -35.7% 8374 ± 5% softirqs.CPU87.RCU 12770 ± 4% -34.5% 8359 ± 4% softirqs.CPU88.RCU 12612 ± 4% -33.0% 8453 ± 3% softirqs.CPU89.RCU 13368 ± 8% -34.1% 8810 ± 20% softirqs.CPU9.RCU 13062 ± 4% -37.5% 8158 ± 4% softirqs.CPU90.RCU 13360 ± 11% -36.8% 8438 ± 2% softirqs.CPU91.RCU 12526 ± 3% -32.4% 8462 ± 4% softirqs.CPU92.RCU 13170 ± 5% -37.4% 8245 ± 3% softirqs.CPU93.RCU 12706 ± 6% -33.5% 8454 ± 3% softirqs.CPU94.RCU 12706 ± 3% -31.6% 8695 ± 4% softirqs.CPU95.RCU 14269 ± 3% -36.9% 8999 ± 2% softirqs.CPU96.RCU 14113 ± 2% -37.1% 8882 ± 2% softirqs.CPU97.RCU 13982 -36.7% 8853 ± 2% softirqs.CPU98.RCU 13875 ± 3% -37.1% 8725 ± 3% softirqs.CPU99.RCU 1662582 -35.0% 1080630 softirqs.RCU *************************************************************************************************** lkp-skl-fpga01: 104 threads 2 sockets Skylake with 192G memory ========================================================================================= compiler/cpufreq_governor/kconfig/mode/nr_task/rootfs/tbox_group/test/testcase/ucode: gcc-9/performance/x86_64-rhel-8.3/process/100%/debian-10.4-x86_64-20200603.cgz/lkp-skl-fpga01/lock1/will-it-scale/0x2006a0a commit: fa4e6b1ad5 ("memcg: enable accounting for pollfd and select bits arrays") 059dd9003a ("memcg: enable accounting for file lock caches") fa4e6b1ad57df096 059dd9003ad894ad7b2e47126cd ---------------- --------------------------- %stddev %change %stddev \ | \ 77280953 -38.0% 47885066 will-it-scale.104.processes 743085 -38.0% 460433 will-it-scale.per_process_ops 77280953 -38.0% 47885066 will-it-scale.workload 0.01 ± 2% -0.0 0.01 ± 2% mpstat.cpu.all.soft% 30.18 -5.9 24.24 mpstat.cpu.all.usr% 7524 -92.9% 535.33 ± 6% slabinfo.Acpi-Parse.active_objs 7524 -92.9% 535.33 ± 6% slabinfo.Acpi-Parse.num_objs 29.00 -17.2% 24.00 vmstat.cpu.us 2014 +6.7% 2149 vmstat.system.cs 2720 +1.4% 2757 turbostat.Bzy_MHz 0.67 -13.4% 0.58 ± 3% turbostat.CPU%c1 196.34 -30.0 166.35 ± 2% turbostat.PKG_% 103.96 +4.3% 108.48 turbostat.RAMWatt 372.50 ± 17% +94.2% 723.50 ± 67% interrupts.CPU0.CAL:Function_call_interrupts 330.17 ± 12% +237.5% 1114 ± 61% interrupts.CPU4.CAL:Function_call_interrupts 987.33 ± 59% +77.2% 1750 ± 24% interrupts.CPU55.RES:Rescheduling_interrupts 360.83 ± 9% +174.1% 989.00 ± 84% interrupts.CPU67.CAL:Function_call_interrupts 383.67 ± 7% +180.8% 1077 ± 68% interrupts.CPU82.CAL:Function_call_interrupts 569.00 ± 63% +194.6% 1676 ± 67% interrupts.CPU89.CAL:Function_call_interrupts 623.33 ± 70% +108.1% 1297 ± 32% interrupts.CPU91.RES:Rescheduling_interrupts 16642 ± 11% -40.5% 9902 ± 2% softirqs.CPU0.RCU 16302 ± 4% -45.3% 8920 ± 7% softirqs.CPU1.RCU 15642 ± 2% -39.3% 9494 ± 13% softirqs.CPU10.RCU 14815 ± 4% -44.3% 8257 ± 4% softirqs.CPU100.RCU 14861 ± 4% -42.3% 8574 ± 6% softirqs.CPU101.RCU 14624 ± 5% -43.2% 8304 ± 5% softirqs.CPU102.RCU 14689 ± 6% -41.9% 8538 ± 8% softirqs.CPU103.RCU 15639 ± 3% -35.7% 10057 ± 24% softirqs.CPU11.RCU 15856 ± 3% -47.4% 8335 ± 3% softirqs.CPU12.RCU 15876 ± 4% -47.1% 8395 ± 4% softirqs.CPU13.RCU 15756 ± 4% -47.3% 8311 ± 3% softirqs.CPU14.RCU 16862 ± 5% -44.0% 9450 ± 7% softirqs.CPU15.RCU 17162 ± 3% -47.5% 9015 ± 3% softirqs.CPU16.RCU 17351 ± 6% -47.1% 9182 ± 3% softirqs.CPU17.RCU 17546 ± 4% -46.6% 9375 ± 4% softirqs.CPU18.RCU 17416 ± 6% -47.4% 9167 ± 2% softirqs.CPU19.RCU 15233 ± 9% -43.7% 8576 ± 3% softirqs.CPU2.RCU 17719 ± 14% -48.8% 9067 ± 6% softirqs.CPU20.RCU 16953 ± 2% -44.1% 9469 ± 8% softirqs.CPU21.RCU 17378 ± 7% -47.9% 9052 ± 2% softirqs.CPU22.RCU 17142 ± 6% -47.4% 9020 ± 3% softirqs.CPU23.RCU 17545 ± 5% -50.7% 8642 ± 12% softirqs.CPU24.RCU 17300 ± 4% -45.7% 9389 ± 7% softirqs.CPU25.RCU 14672 ± 4% -42.0% 8507 ± 9% softirqs.CPU26.RCU 15253 ± 3% -41.3% 8954 ± 6% softirqs.CPU27.RCU 14806 ± 4% -43.3% 8398 ± 7% softirqs.CPU28.RCU 13654 ± 13% -33.2% 9118 ± 21% softirqs.CPU29.RCU 16430 ± 3% -46.1% 8853 ± 8% softirqs.CPU3.RCU 14992 ± 5% -40.5% 8921 ± 7% softirqs.CPU30.RCU 15179 ± 5% -39.6% 9167 ± 9% softirqs.CPU31.RCU 15216 ± 5% -38.6% 9336 ± 7% softirqs.CPU32.RCU 14965 ± 5% -39.4% 9067 ± 8% softirqs.CPU33.RCU 15161 ± 5% -42.2% 8768 ± 5% softirqs.CPU34.RCU 15097 ± 5% -42.8% 8635 ± 4% softirqs.CPU35.RCU 15122 ± 6% -42.9% 8634 ± 4% softirqs.CPU36.RCU 15193 ± 5% -44.5% 8426 ± 11% softirqs.CPU37.RCU 15082 ± 5% -42.7% 8649 ± 5% softirqs.CPU38.RCU 15220 ± 6% -42.7% 8717 ± 7% softirqs.CPU39.RCU 16040 ± 4% -46.5% 8585 ± 5% softirqs.CPU4.RCU 14629 ± 8% -39.7% 8828 ± 6% softirqs.CPU40.RCU 14559 ± 7% -40.7% 8634 ± 5% softirqs.CPU41.RCU 14919 ± 5% -42.8% 8527 ± 5% softirqs.CPU42.RCU 15085 ± 5% -40.1% 9034 ± 9% softirqs.CPU43.RCU 15018 ± 6% -44.4% 8355 ± 8% softirqs.CPU44.RCU 15649 ± 2% -45.6% 8520 ± 3% softirqs.CPU45.RCU 15775 ± 2% -44.1% 8820 ± 7% softirqs.CPU46.RCU 15614 ± 2% -43.5% 8826 ± 5% softirqs.CPU47.RCU 15783 ± 2% -45.0% 8673 ± 8% softirqs.CPU48.RCU 15499 ± 3% -44.6% 8591 ± 8% softirqs.CPU49.RCU 16032 ± 3% -45.5% 8739 ± 6% softirqs.CPU5.RCU 15372 ± 3% -45.5% 8382 ± 3% softirqs.CPU50.RCU 15523 ± 4% -44.7% 8577 ± 3% softirqs.CPU51.RCU 16918 ± 7% -45.1% 9283 ± 4% softirqs.CPU52.RCU 16956 ± 14% -50.4% 8405 ± 17% softirqs.CPU53.RCU 17276 ± 5% -45.7% 9376 ± 3% softirqs.CPU54.RCU 17339 ± 4% -45.1% 9510 ± 6% softirqs.CPU55.RCU 17509 ± 4% -47.3% 9224 ± 2% softirqs.CPU56.RCU 17348 ± 5% -46.5% 9289 ± 4% softirqs.CPU57.RCU 17142 ± 3% -45.8% 9292 ± 6% softirqs.CPU58.RCU 17700 ± 5% -46.1% 9532 ± 5% softirqs.CPU59.RCU 15856 ± 3% -45.7% 8615 ± 4% softirqs.CPU6.RCU 16017 ± 3% -44.6% 8870 ± 6% softirqs.CPU60.RCU 16498 ± 4% -46.8% 8784 ± 4% softirqs.CPU61.RCU 16190 ± 4% -42.4% 9328 ± 16% softirqs.CPU62.RCU 16491 -45.8% 8935 ± 4% softirqs.CPU63.RCU 16504 ± 3% -45.0% 9080 ± 6% softirqs.CPU64.RCU 16232 ± 5% -47.3% 8562 ± 4% softirqs.CPU65.RCU 15822 ± 2% -44.8% 8726 ± 4% softirqs.CPU66.RCU 16801 ± 5% -48.4% 8662 ± 4% softirqs.CPU67.RCU 16812 ± 4% -47.6% 8812 ± 5% softirqs.CPU68.RCU 16720 ± 3% -46.0% 9031 ± 4% softirqs.CPU69.RCU 15795 ± 5% -47.1% 8359 ± 4% softirqs.CPU7.RCU 16875 ± 4% -46.5% 9025 ± 3% softirqs.CPU70.RCU 16876 ± 2% -46.8% 8977 ± 4% softirqs.CPU71.RCU 16864 ± 3% -45.9% 9122 ± 7% softirqs.CPU72.RCU 16398 ± 3% -45.1% 8996 ± 7% softirqs.CPU73.RCU 16418 ± 2% -43.9% 9211 ± 7% softirqs.CPU74.RCU 15707 ± 7% -45.4% 8570 ± 7% softirqs.CPU75.RCU 15498 ± 5% -45.9% 8391 ± 6% softirqs.CPU76.RCU 15363 ± 7% -46.7% 8192 ± 5% softirqs.CPU77.RCU 16108 ± 5% -43.0% 9176 ± 4% softirqs.CPU78.RCU 16078 ± 5% -42.4% 9268 ± 4% softirqs.CPU79.RCU 15729 ± 3% -47.2% 8305 ± 4% softirqs.CPU8.RCU 16445 ± 5% -42.7% 9427 ± 5% softirqs.CPU80.RCU 16028 ± 9% -42.4% 9235 ± 5% softirqs.CPU81.RCU 16194 ± 6% -42.8% 9257 ± 4% softirqs.CPU82.RCU 15968 ± 5% -42.0% 9261 ± 6% softirqs.CPU83.RCU 16105 ± 5% -37.5% 10071 ± 16% softirqs.CPU84.RCU 15917 ± 6% -41.5% 9310 ± 4% softirqs.CPU85.RCU 15961 ± 5% -41.5% 9333 ± 5% softirqs.CPU86.RCU 15897 ± 5% -42.7% 9110 ± 5% softirqs.CPU87.RCU 16110 ± 7% -42.7% 9225 ± 4% softirqs.CPU88.RCU 15825 ± 5% -42.4% 9112 ± 4% softirqs.CPU89.RCU 15663 ± 5% -46.6% 8360 ± 2% softirqs.CPU9.RCU 14789 ± 5% -41.2% 8697 ± 5% softirqs.CPU90.RCU 14887 ± 4% -39.1% 9060 ± 9% softirqs.CPU91.RCU 14299 ± 12% -40.4% 8521 ± 6% softirqs.CPU92.RCU 14215 ± 11% -40.1% 8511 ± 6% softirqs.CPU93.RCU 14813 ± 4% -42.5% 8524 ± 4% softirqs.CPU94.RCU 14963 ± 5% -44.6% 8282 ± 4% softirqs.CPU95.RCU 14744 ± 5% -45.7% 8002 ± 12% softirqs.CPU96.RCU 14968 ± 5% -43.3% 8485 ± 7% softirqs.CPU97.RCU 15030 ± 4% -43.2% 8530 ± 8% softirqs.CPU98.RCU 15961 ± 12% -46.6% 8526 ± 3% softirqs.CPU99.RCU 1654543 ± 3% -44.2% 923546 ± 3% softirqs.RCU 42461 ± 2% +33.1% 56527 ± 3% softirqs.TIMER *************************************************************************************************** lkp-icl-2sp1: 96 threads 2 sockets Ice Lake with 256G memory ========================================================================================= compiler/cpufreq_governor/kconfig/mode/nr_task/rootfs/tbox_group/test/testcase/ucode: gcc-9/performance/x86_64-rhel-8.3/process/16/debian-10.4-x86_64-20200603.cgz/lkp-icl-2sp1/lock1/will-it-scale/0xb000280 commit: fa4e6b1ad5 ("memcg: enable accounting for pollfd and select bits arrays") 059dd9003a ("memcg: enable accounting for file lock caches") fa4e6b1ad57df096 059dd9003ad894ad7b2e47126cd ---------------- --------------------------- %stddev %change %stddev \ | \ 51162267 -34.5% 33497392 ± 5% will-it-scale.16.processes 3197641 -34.5% 2093586 ± 5% will-it-scale.per_process_ops 51162267 -34.5% 33497392 ± 5% will-it-scale.workload 1223202 ± 4% +29.1% 1578754 ± 6% cpuidle..usage 1089773 ± 2% +35.5% 1476283 ± 7% turbostat.C6 19528 +6.4% 20781 vmstat.system.in 5751 ± 8% -24.5% 4341 ± 12% slabinfo.Acpi-Parse.active_objs 5751 ± 8% -24.5% 4341 ± 12% slabinfo.Acpi-Parse.num_objs 0.03 ± 2% +0.0 0.03 ± 3% mpstat.cpu.all.irq% 0.01 ± 5% +0.0 0.01 ± 6% mpstat.cpu.all.soft% 2.49 -0.8 1.65 ± 5% mpstat.cpu.all.usr% 87990 ± 3% +24.3% 109405 ± 18% interrupts.CAL:Function_call_interrupts 51303 ± 18% +29.3% 66321 ± 5% interrupts.CPU0.CAL:Function_call_interrupts 19410 ± 53% +135.2% 45644 ± 40% interrupts.CPU67.LOC:Local_timer_interrupts 5235 ±112% +285.6% 20185 ± 73% interrupts.CPU79.LOC:Local_timer_interrupts 9028 ± 18% -19.9% 7236 ± 2% softirqs.CPU0.RCU 292709 ± 11% -15.8% 246430 ± 3% softirqs.RCU 408900 +14.1% 466365 ± 3% softirqs.SCHED 35525 ± 4% +16.2% 41283 ± 5% softirqs.TIMER 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