Greeting, FYI, we noticed a 24.4% improvement of stress-ng.tmpfs.ops_per_sec due to commit: commit: 07ca760673088f262da57ff42c15558688565aa2 ("mm/munlock: maintain page->mlock_count while unevictable") https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git master in testcase: stress-ng on test machine: 128 threads 2 sockets Intel(R) Xeon(R) Platinum 8358 CPU @ 2.60GHz with 128G memory with following parameters: nr_threads: 100% testtime: 60s class: memory test: tmpfs cpufreq_governor: performance ucode: 0xd000331 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. ========================================================================================= class/compiler/cpufreq_governor/kconfig/nr_threads/rootfs/tbox_group/test/testcase/testtime/ucode: memory/gcc-9/performance/x86_64-rhel-8.3/100%/debian-10.4-x86_64-20200603.cgz/lkp-icl-2sp6/tmpfs/stress-ng/60s/0xd000331 commit: b109b87050 ("mm/munlock: replace clear_page_mlock() by final clearance") 07ca760673 ("mm/munlock: maintain page->mlock_count while unevictable") b109b87050df5438 07ca760673088f262da57ff42c1 ---------------- --------------------------- %stddev %change %stddev \ | \ 39.44 ± 5% -18.7% 32.08 ± 3% stress-ng.time.elapsed_time 39.44 ± 5% -18.7% 32.08 ± 3% stress-ng.time.elapsed_time.max 10529 ± 6% -22.7% 8134 ± 5% stress-ng.time.involuntary_context_switches 9988 ± 2% -3.4% 9646 stress-ng.time.percent_of_cpu_this_job_got 2663 ± 8% -33.3% 1776 ± 5% stress-ng.time.system_time 1276 +3.2% 1317 stress-ng.time.user_time 2955 ± 5% +24.4% 3677 ± 2% stress-ng.tmpfs.ops_per_sec 630625 ± 6% -13.9% 543204 ± 11% numa-numastat.node1.numa_hit 12458183 ± 9% -29.8% 8745870 ± 3% turbostat.IRQ 22.83 ± 6% +25.5% 28.67 ± 3% vmstat.cpu.us 3825 ± 2% +15.7% 4424 vmstat.system.cs 291800 ± 4% -15.1% 247782 vmstat.system.in 160426 +11.2% 178462 ± 2% meminfo.Active 160426 +11.2% 178462 ± 2% meminfo.Active(anon) 235963 ± 5% -16.1% 197931 ± 2% meminfo.Mapped 117810 ± 6% -24.1% 89437 ± 3% meminfo.Mlocked 23.66 ± 5% +3.1 26.73 ± 3% mpstat.cpu.all.idle% 0.02 ± 27% +0.0 0.03 ± 18% mpstat.cpu.all.soft% 50.99 ± 4% -9.5 41.53 ± 3% mpstat.cpu.all.sys% 24.59 ± 5% +6.3 30.87 ± 3% mpstat.cpu.all.usr% 78820 ± 3% +14.4% 90181 ± 3% numa-meminfo.node0.Active 78820 ± 3% +14.4% 90181 ± 3% numa-meminfo.node0.Active(anon) 59902 ± 10% -23.5% 45810 ± 4% numa-meminfo.node0.Mlocked 152030 ± 18% -25.1% 113836 ± 12% numa-meminfo.node1.Inactive 152030 ± 18% -25.1% 113836 ± 12% numa-meminfo.node1.Inactive(anon) 57489 ± 10% -26.6% 42196 ± 7% numa-meminfo.node1.Mlocked 233645 ± 5% -14.0% 201029 ± 3% numa-meminfo.node1.Shmem 19568 ± 2% +14.8% 22461 ± 3% numa-vmstat.node0.nr_active_anon 14599 ± 9% -23.2% 11217 ± 7% numa-vmstat.node0.nr_mlock 19562 ± 2% +14.8% 22462 ± 3% numa-vmstat.node0.nr_zone_active_anon 37654 ± 18% -24.1% 28589 ± 10% numa-vmstat.node1.nr_inactive_anon 14616 ± 10% -27.5% 10599 ± 10% numa-vmstat.node1.nr_mlock 58341 ± 5% -14.1% 50124 ± 3% numa-vmstat.node1.nr_shmem 37656 ± 18% -24.1% 28586 ± 10% numa-vmstat.node1.nr_zone_inactive_anon 39985 +10.5% 44174 ± 2% proc-vmstat.nr_active_anon 725611 -1.3% 716100 proc-vmstat.nr_file_pages 121814 ± 2% -6.2% 114239 proc-vmstat.nr_inactive_anon 59502 ± 5% -16.7% 49539 proc-vmstat.nr_mapped 29485 ± 6% -24.6% 22222 ± 3% proc-vmstat.nr_mlock 115710 ± 3% -8.2% 106199 proc-vmstat.nr_shmem 639255 -1.1% 632112 proc-vmstat.nr_unevictable 39985 +10.5% 44174 ± 2% proc-vmstat.nr_zone_active_anon 121814 ± 2% -6.2% 114239 proc-vmstat.nr_zone_inactive_anon 639256 -1.1% 632112 proc-vmstat.nr_zone_unevictable 1233200 -7.1% 1145498 proc-vmstat.numa_hit 1117527 -7.9% 1029792 proc-vmstat.numa_local 1233229 -7.1% 1145564 proc-vmstat.pgalloc_normal 1000156 -4.7% 952725 proc-vmstat.pgfree 17338 -7.4% 16058 ± 3% proc-vmstat.pgreuse 7.718e+10 ± 4% +20.3% 9.289e+10 ± 2% perf-stat.i.branch-instructions 2.104e+08 ± 5% +19.5% 2.515e+08 ± 3% perf-stat.i.branch-misses 1.232e+08 ± 2% +13.2% 1.395e+08 ± 2% perf-stat.i.cache-misses 5.521e+08 ± 6% +13.8% 6.284e+08 ± 3% perf-stat.i.cache-references 3388 ± 2% +15.3% 3905 perf-stat.i.context-switches 1.04 ± 4% -19.6% 0.83 ± 2% perf-stat.i.cpi 3.32e+11 ± 2% -4.3% 3.176e+11 perf-stat.i.cpu-cycles 281.54 ± 4% +11.5% 313.86 ± 3% perf-stat.i.cpu-migrations 2740 -8.8% 2497 ± 6% perf-stat.i.cycles-between-cache-misses 3.883e+10 ± 4% +17.8% 4.575e+10 ± 2% perf-stat.i.dTLB-loads 37861529 ± 5% +22.5% 46399080 ± 2% perf-stat.i.dTLB-store-misses 2.93e+10 ± 5% +22.1% 3.577e+10 ± 2% perf-stat.i.dTLB-stores 3.177e+11 ± 4% +20.4% 3.824e+11 ± 2% perf-stat.i.instructions 1.01 ± 5% +24.4% 1.25 ± 2% perf-stat.i.ipc 1385 ± 6% +28.2% 1776 ± 4% perf-stat.i.major-faults 2.59 ± 2% -4.3% 2.48 perf-stat.i.metric.GHz 670.12 +17.6% 788.18 perf-stat.i.metric.K/sec 1138 ± 4% +16.7% 1329 ± 2% perf-stat.i.metric.M/sec 4303675 ± 5% +22.3% 5264697 ± 2% perf-stat.i.minor-faults 4223176 ± 9% +21.5% 5132933 ± 5% perf-stat.i.node-loads 12619285 ± 9% +26.6% 15970682 ± 5% perf-stat.i.node-stores 4305061 ± 5% +22.3% 5266473 ± 2% perf-stat.i.page-faults 1.74 ± 2% -5.4% 1.64 perf-stat.overall.MPKI 1.05 ± 5% -20.6% 0.83 ± 3% perf-stat.overall.cpi 2694 ± 2% -15.5% 2276 perf-stat.overall.cycles-between-cache-misses 0.96 ± 5% +25.7% 1.20 ± 3% perf-stat.overall.ipc 7.499e+10 ± 4% +20.1% 9.003e+10 ± 2% perf-stat.ps.branch-instructions 2.044e+08 ± 5% +19.2% 2.437e+08 ± 3% perf-stat.ps.branch-misses 1.197e+08 ± 2% +13.0% 1.352e+08 ± 2% perf-stat.ps.cache-misses 5.364e+08 ± 6% +13.5% 6.09e+08 ± 3% perf-stat.ps.cache-references 3296 ± 2% +14.8% 3784 perf-stat.ps.context-switches 3.223e+11 ± 2% -4.5% 3.078e+11 perf-stat.ps.cpu-cycles 273.70 ± 4% +11.1% 304.20 ± 3% perf-stat.ps.cpu-migrations 3.773e+10 ± 4% +17.5% 4.434e+10 ± 2% perf-stat.ps.dTLB-loads 36783176 ± 5% +22.3% 44970404 ± 2% perf-stat.ps.dTLB-store-misses 2.847e+10 ± 5% +21.8% 3.467e+10 ± 2% perf-stat.ps.dTLB-stores 3.087e+11 ± 4% +20.1% 3.706e+11 ± 2% perf-stat.ps.instructions 1347 ± 6% +27.7% 1721 ± 4% perf-stat.ps.major-faults 4181461 ± 5% +22.0% 5102628 ± 2% perf-stat.ps.minor-faults 4097306 ± 9% +21.4% 4973600 ± 5% perf-stat.ps.node-loads 12251815 ± 9% +26.3% 15476164 ± 5% perf-stat.ps.node-stores 4182809 ± 5% +22.0% 5104350 ± 2% perf-stat.ps.page-faults 1.243e+13 -1.7% 1.222e+13 perf-stat.total.instructions 38.79 ± 13% -33.6 5.18 ±101% perf-profile.calltrace.cycles-pp.zap_pte_range.unmap_page_range.unmap_vmas.unmap_region.__do_munmap 39.23 ± 13% -33.6 5.65 ± 96% perf-profile.calltrace.cycles-pp.unmap_vmas.unmap_region.__do_munmap.__vm_munmap.__x64_sys_munmap 39.10 ± 13% -33.6 5.54 ± 97% perf-profile.calltrace.cycles-pp.unmap_page_range.unmap_vmas.unmap_region.__do_munmap.__vm_munmap 40.08 ± 12% -33.5 6.58 ± 88% perf-profile.calltrace.cycles-pp.unmap_region.__do_munmap.__vm_munmap.__x64_sys_munmap.do_syscall_64 42.54 ± 10% -33.2 9.33 ± 75% perf-profile.calltrace.cycles-pp.__do_munmap.__vm_munmap.__x64_sys_munmap.do_syscall_64.entry_SYSCALL_64_after_hwframe 42.84 ± 10% -33.2 9.65 ± 73% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.__munmap.stress_oomable_child 42.75 ± 10% -33.2 9.57 ± 74% perf-profile.calltrace.cycles-pp.__vm_munmap.__x64_sys_munmap.do_syscall_64.entry_SYSCALL_64_after_hwframe.__munmap 42.77 ± 10% -33.2 9.60 ± 74% perf-profile.calltrace.cycles-pp.__x64_sys_munmap.do_syscall_64.entry_SYSCALL_64_after_hwframe.__munmap.stress_oomable_child 42.96 ± 10% -33.1 9.81 ± 73% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.__munmap.stress_oomable_child 43.23 ± 10% -33.1 10.12 ± 72% perf-profile.calltrace.cycles-pp.__munmap.stress_oomable_child 79.55 -30.9 48.66 ± 51% perf-profile.calltrace.cycles-pp.stress_oomable_child 22.84 ± 14% -21.7 1.14 ±162% perf-profile.calltrace.cycles-pp.free_pages_and_swap_cache.tlb_flush_mmu.zap_pte_range.unmap_page_range.unmap_vmas 22.81 ± 14% -21.7 1.12 ±163% perf-profile.calltrace.cycles-pp.lru_add_drain.free_pages_and_swap_cache.tlb_flush_mmu.zap_pte_range.unmap_page_range 23.00 ± 14% -21.7 1.33 ±142% perf-profile.calltrace.cycles-pp.tlb_flush_mmu.zap_pte_range.unmap_page_range.unmap_vmas.unmap_region 13.86 ± 15% -13.9 0.00 perf-profile.calltrace.cycles-pp.lru_add_drain_cpu.lru_add_drain.free_pages_and_swap_cache.tlb_flush_mmu.zap_pte_range 13.83 ± 15% -13.8 0.00 perf-profile.calltrace.cycles-pp.__pagevec_lru_add.lru_add_drain_cpu.lru_add_drain.free_pages_and_swap_cache.tlb_flush_mmu 13.62 ± 15% -13.6 0.00 perf-profile.calltrace.cycles-pp.isolate_lru_page.munlock_page.zap_pte_range.unmap_page_range.unmap_vmas 13.62 ± 15% -13.6 0.00 perf-profile.calltrace.cycles-pp.isolate_lru_page.mlock_page.do_set_pte.filemap_map_pages.do_fault 15.07 ± 14% -13.5 1.59 ±168% perf-profile.calltrace.cycles-pp.handle_mm_fault.__get_user_pages.populate_vma_page_range.__mm_populate.vm_mmap_pgoff 15.07 ± 14% -13.5 1.59 ±168% perf-profile.calltrace.cycles-pp.__handle_mm_fault.handle_mm_fault.__get_user_pages.populate_vma_page_range.__mm_populate 15.06 ± 14% -13.5 1.59 ±167% perf-profile.calltrace.cycles-pp.do_fault.__handle_mm_fault.handle_mm_fault.__get_user_pages.populate_vma_page_range 15.06 ± 14% -13.5 1.59 ±167% perf-profile.calltrace.cycles-pp.filemap_map_pages.do_fault.__handle_mm_fault.handle_mm_fault.__get_user_pages 15.18 ± 14% -13.5 1.71 ±155% perf-profile.calltrace.cycles-pp.__mm_populate.vm_mmap_pgoff.ksys_mmap_pgoff.do_syscall_64.entry_SYSCALL_64_after_hwframe 15.18 ± 14% -13.5 1.71 ±155% perf-profile.calltrace.cycles-pp.populate_vma_page_range.__mm_populate.vm_mmap_pgoff.ksys_mmap_pgoff.do_syscall_64 15.01 ± 14% -13.5 1.54 ±169% perf-profile.calltrace.cycles-pp.do_set_pte.filemap_map_pages.do_fault.__handle_mm_fault.handle_mm_fault 15.18 ± 14% -13.5 1.70 ±155% perf-profile.calltrace.cycles-pp.__get_user_pages.populate_vma_page_range.__mm_populate.vm_mmap_pgoff.ksys_mmap_pgoff 13.43 ± 15% -13.4 0.00 perf-profile.calltrace.cycles-pp.folio_lruvec_lock_irq.isolate_lru_page.munlock_page.zap_pte_range.unmap_page_range 13.40 ± 15% -13.4 0.00 perf-profile.calltrace.cycles-pp.folio_lruvec_lock_irq.isolate_lru_page.mlock_page.do_set_pte.filemap_map_pages 13.38 ± 15% -13.4 0.00 perf-profile.calltrace.cycles-pp._raw_spin_lock_irq.folio_lruvec_lock_irq.isolate_lru_page.munlock_page.zap_pte_range 13.38 ± 15% -13.4 0.00 perf-profile.calltrace.cycles-pp._raw_spin_lock_irq.folio_lruvec_lock_irq.isolate_lru_page.mlock_page.do_set_pte 13.35 ± 15% -13.3 0.00 perf-profile.calltrace.cycles-pp.folio_lruvec_lock_irqsave.__pagevec_lru_add.lru_add_drain_cpu.lru_add_drain.free_pages_and_swap_cache 13.30 ± 15% -13.3 0.00 perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.folio_lruvec_lock_irqsave.__pagevec_lru_add.lru_add_drain_cpu.lru_add_drain 13.28 ± 15% -13.3 0.00 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irq.folio_lruvec_lock_irq.isolate_lru_page.mlock_page 13.26 ± 15% -13.3 0.00 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irq.folio_lruvec_lock_irq.isolate_lru_page.munlock_page 13.17 ± 16% -13.2 0.00 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.folio_lruvec_lock_irqsave.__pagevec_lru_add.lru_add_drain_cpu 17.87 ± 9% -13.0 4.91 ± 76% perf-profile.calltrace.cycles-pp.vm_mmap_pgoff.ksys_mmap_pgoff.do_syscall_64.entry_SYSCALL_64_after_hwframe.__mmap 17.97 ± 9% -12.9 5.02 ± 74% perf-profile.calltrace.cycles-pp.ksys_mmap_pgoff.do_syscall_64.entry_SYSCALL_64_after_hwframe.__mmap 18.02 ± 9% -12.9 5.09 ± 74% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.__mmap 18.01 ± 9% -12.9 5.08 ± 74% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.__mmap 18.15 ± 9% -12.9 5.25 ± 73% perf-profile.calltrace.cycles-pp.__mmap 13.83 ± 15% -12.2 1.60 ±169% perf-profile.calltrace.cycles-pp.munlock_page.zap_pte_range.unmap_page_range.unmap_vmas.unmap_region 13.65 ± 15% -12.1 1.51 ±169% perf-profile.calltrace.cycles-pp.mlock_page.do_set_pte.filemap_map_pages.do_fault.__handle_mm_fault 8.93 ± 14% -7.8 1.09 ±164% perf-profile.calltrace.cycles-pp.pagevec_lru_move_fn.lru_add_drain.free_pages_and_swap_cache.tlb_flush_mmu.zap_pte_range 8.53 ± 15% -7.6 0.93 ±170% perf-profile.calltrace.cycles-pp.folio_lruvec_lock_irqsave.pagevec_lru_move_fn.lru_add_drain.free_pages_and_swap_cache.tlb_flush_mmu 8.47 ± 15% -7.6 0.91 ±171% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.folio_lruvec_lock_irqsave.pagevec_lru_move_fn.lru_add_drain.free_pages_and_swap_cache 8.38 ± 15% -7.5 0.88 ±172% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.folio_lruvec_lock_irqsave.pagevec_lru_move_fn.lru_add_drain 0.00 +2.9 2.95 ± 92% perf-profile.calltrace.cycles-pp.menu_select.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64_no_verify 0.10 ±223% +3.9 3.98 ± 97% perf-profile.calltrace.cycles-pp.__hrtimer_run_queues.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt 0.13 ±223% +5.9 6.02 ± 87% perf-profile.calltrace.cycles-pp.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state 0.14 ±223% +6.0 6.18 ± 88% perf-profile.calltrace.cycles-pp.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state.cpuidle_enter 0.20 ±223% +9.5 9.66 ± 85% perf-profile.calltrace.cycles-pp.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state.cpuidle_enter.do_idle 0.22 ±223% +12.0 12.18 ± 84% perf-profile.calltrace.cycles-pp.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state.cpuidle_enter.do_idle.cpu_startup_entry 0.72 ±114% +21.5 22.26 ± 59% perf-profile.calltrace.cycles-pp.intel_idle.cpuidle_enter_state.cpuidle_enter.do_idle.cpu_startup_entry 1.34 ± 87% +33.8 35.10 ± 68% perf-profile.calltrace.cycles-pp.cpuidle_enter_state.cpuidle_enter.do_idle.cpu_startup_entry.start_secondary 1.39 ± 87% +34.5 35.90 ± 67% perf-profile.calltrace.cycles-pp.cpuidle_enter.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64_no_verify 1.53 ± 89% +38.5 39.98 ± 70% perf-profile.calltrace.cycles-pp.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64_no_verify 1.53 ± 89% +38.5 40.04 ± 70% perf-profile.calltrace.cycles-pp.cpu_startup_entry.start_secondary.secondary_startup_64_no_verify 1.53 ± 89% +38.5 40.04 ± 70% perf-profile.calltrace.cycles-pp.start_secondary.secondary_startup_64_no_verify 1.54 ± 89% +38.8 40.32 ± 70% perf-profile.calltrace.cycles-pp.secondary_startup_64_no_verify 49.06 ± 15% -45.4 3.62 ±171% perf-profile.children.cycles-pp.native_queued_spin_lock_slowpath 63.53 ± 9% -42.6 20.90 ± 55% perf-profile.children.cycles-pp.do_syscall_64 63.70 ± 9% -42.6 21.12 ± 55% perf-profile.children.cycles-pp.entry_SYSCALL_64_after_hwframe 38.83 ± 13% -33.6 5.24 ±100% perf-profile.children.cycles-pp.zap_pte_range 39.24 ± 13% -33.6 5.69 ± 95% perf-profile.children.cycles-pp.unmap_vmas 39.12 ± 13% -33.5 5.58 ± 96% perf-profile.children.cycles-pp.unmap_page_range 40.10 ± 12% -33.5 6.60 ± 88% perf-profile.children.cycles-pp.unmap_region 42.57 ± 10% -33.2 9.39 ± 74% perf-profile.children.cycles-pp.__do_munmap 42.76 ± 10% -33.2 9.60 ± 74% perf-profile.children.cycles-pp.__vm_munmap 42.77 ± 10% -33.2 9.61 ± 74% perf-profile.children.cycles-pp.__x64_sys_munmap 43.32 ± 10% -33.1 10.21 ± 72% perf-profile.children.cycles-pp.__munmap 79.55 -30.9 48.66 ± 51% perf-profile.children.cycles-pp.stress_oomable_child 27.28 ± 15% -27.3 0.00 perf-profile.children.cycles-pp.isolate_lru_page 26.84 ± 15% -24.0 2.82 ±169% perf-profile.children.cycles-pp.folio_lruvec_lock_irq 26.77 ± 15% -23.9 2.87 ±165% perf-profile.children.cycles-pp._raw_spin_lock_irq 22.82 ± 15% -21.9 0.96 ±166% perf-profile.children.cycles-pp.folio_lruvec_lock_irqsave 22.90 ± 14% -21.6 1.30 ±136% perf-profile.children.cycles-pp.lru_add_drain 22.85 ± 14% -21.6 1.30 ±136% perf-profile.children.cycles-pp.free_pages_and_swap_cache 23.04 ± 14% -21.5 1.53 ±118% perf-profile.children.cycles-pp.tlb_flush_mmu 22.76 ± 15% -21.5 1.25 ±118% perf-profile.children.cycles-pp._raw_spin_lock_irqsave 15.03 ± 14% -15.0 0.00 perf-profile.children.cycles-pp.__pagevec_lru_add 13.92 ± 14% -13.8 0.08 ± 55% perf-profile.children.cycles-pp.lru_add_drain_cpu 15.37 ± 14% -13.5 1.90 ±141% perf-profile.children.cycles-pp.do_set_pte 15.18 ± 14% -13.4 1.79 ±145% perf-profile.children.cycles-pp.__mm_populate 15.18 ± 14% -13.4 1.79 ±145% perf-profile.children.cycles-pp.populate_vma_page_range 15.18 ± 14% -13.4 1.80 ±143% perf-profile.children.cycles-pp.__get_user_pages 16.02 ± 13% -13.3 2.68 ±108% perf-profile.children.cycles-pp.do_fault 15.06 ± 14% -13.3 1.76 ±145% perf-profile.children.cycles-pp.filemap_map_pages 16.20 ± 12% -13.2 2.98 ± 97% perf-profile.children.cycles-pp.__handle_mm_fault 16.42 ± 12% -13.2 3.22 ± 92% perf-profile.children.cycles-pp.handle_mm_fault 17.88 ± 9% -12.9 4.96 ± 74% perf-profile.children.cycles-pp.vm_mmap_pgoff 17.97 ± 9% -12.9 5.06 ± 74% perf-profile.children.cycles-pp.ksys_mmap_pgoff 18.20 ± 9% -12.9 5.32 ± 72% perf-profile.children.cycles-pp.__mmap 13.83 ± 15% -12.1 1.75 ±150% perf-profile.children.cycles-pp.munlock_page 13.65 ± 15% -12.1 1.60 ±157% perf-profile.children.cycles-pp.mlock_page 8.94 ± 14% -7.8 1.18 ±148% perf-profile.children.cycles-pp.pagevec_lru_move_fn 0.23 ± 4% -0.1 0.10 ± 75% perf-profile.children.cycles-pp.__list_del_entry_valid 0.06 ± 16% +0.0 0.10 ± 20% perf-profile.children.cycles-pp.fput_many 0.01 ±223% +0.1 0.10 ± 52% perf-profile.children.cycles-pp.__fput 0.01 ±223% +0.1 0.10 ± 49% perf-profile.children.cycles-pp.task_work_run 0.02 ±141% +0.1 0.12 ± 37% perf-profile.children.cycles-pp._find_next_bit 0.10 ± 20% +0.1 0.22 ± 40% perf-profile.children.cycles-pp.exit_to_user_mode_prepare 0.15 ± 18% +0.1 0.28 ± 35% perf-profile.children.cycles-pp.syscall_exit_to_user_mode 0.00 +0.1 0.14 ± 57% perf-profile.children.cycles-pp.rcu_core 0.00 +0.1 0.14 ± 92% perf-profile.children.cycles-pp.exit_mmap 0.00 +0.1 0.14 ± 91% perf-profile.children.cycles-pp.mmput 0.00 +0.2 0.15 ± 82% perf-profile.children.cycles-pp.__x64_sys_exit_group 0.00 +0.2 0.19 ±122% perf-profile.children.cycles-pp.__schedule 0.00 +0.2 0.19 ± 48% perf-profile.children.cycles-pp._raw_spin_trylock 0.00 +0.2 0.20 ±100% perf-profile.children.cycles-pp.do_group_exit 0.00 +0.2 0.20 ±100% perf-profile.children.cycles-pp.do_exit 0.00 +0.2 0.22 ± 62% perf-profile.children.cycles-pp.timekeeping_max_deferment 0.00 +0.2 0.22 ± 53% perf-profile.children.cycles-pp.__intel_pmu_enable_all 0.01 ±223% +0.3 0.28 ± 85% perf-profile.children.cycles-pp.update_rq_clock 0.00 +0.3 0.28 ± 63% perf-profile.children.cycles-pp.start_kernel 0.01 ±223% +0.3 0.29 ± 84% perf-profile.children.cycles-pp.rcu_sched_clock_irq 0.00 +0.3 0.28 ± 66% perf-profile.children.cycles-pp.arch_scale_freq_tick 0.00 +0.3 0.31 ± 83% perf-profile.children.cycles-pp.update_blocked_averages 0.00 +0.3 0.32 ± 81% perf-profile.children.cycles-pp.run_rebalance_domains 0.12 ± 22% +0.3 0.44 ± 57% perf-profile.children.cycles-pp._raw_spin_lock 0.00 +0.3 0.33 ± 77% perf-profile.children.cycles-pp.ktime_get_update_offsets_now 0.01 ±223% +0.3 0.34 ± 94% perf-profile.children.cycles-pp.update_sd_lb_stats 0.00 +0.3 0.34 ± 48% perf-profile.children.cycles-pp.tick_nohz_irq_exit 0.01 ±223% +0.3 0.35 ± 91% perf-profile.children.cycles-pp.find_busiest_group 0.00 +0.3 0.34 ±123% perf-profile.children.cycles-pp.hrtimer_next_event_without 0.00 +0.3 0.35 ±107% perf-profile.children.cycles-pp.get_next_timer_interrupt 0.00 +0.4 0.36 ± 87% perf-profile.children.cycles-pp.rcu_idle_exit 0.01 ±223% +0.4 0.38 ±103% perf-profile.children.cycles-pp.timerqueue_del 0.01 ±223% +0.4 0.43 ± 92% perf-profile.children.cycles-pp.__remove_hrtimer 0.01 ±223% +0.5 0.47 ±104% perf-profile.children.cycles-pp.irqtime_account_irq 0.01 ±223% +0.5 0.50 ± 80% perf-profile.children.cycles-pp.load_balance 0.01 ±223% +0.5 0.53 ± 95% perf-profile.children.cycles-pp.native_sched_clock 0.01 ±223% +0.6 0.62 ± 89% perf-profile.children.cycles-pp.lapic_next_deadline 0.01 ±223% +0.6 0.63 ± 93% perf-profile.children.cycles-pp.sched_clock_cpu 0.01 ±223% +0.6 0.65 ± 71% perf-profile.children.cycles-pp.perf_mux_hrtimer_handler 0.01 ±223% +0.6 0.66 ± 66% perf-profile.children.cycles-pp.rebalance_domains 0.01 ±223% +0.7 0.66 ±100% perf-profile.children.cycles-pp.read_tsc 0.47 ± 18% +0.7 1.19 ± 11% perf-profile.children.cycles-pp.native_irq_return_iret 0.02 ±223% +0.8 0.86 ± 62% perf-profile.children.cycles-pp.tick_nohz_next_event 0.02 ±223% +0.9 0.94 ± 99% perf-profile.children.cycles-pp.tick_irq_enter 0.02 ±223% +1.0 0.99 ±101% perf-profile.children.cycles-pp.irq_enter_rcu 0.13 ± 43% +1.0 1.11 ± 90% perf-profile.children.cycles-pp.scheduler_tick 0.08 ± 36% +1.1 1.13 ± 65% perf-profile.children.cycles-pp.clockevents_program_event 0.02 ±223% +1.3 1.30 ± 82% perf-profile.children.cycles-pp.tick_nohz_get_sleep_length 0.04 ±124% +1.3 1.34 ± 66% perf-profile.children.cycles-pp.__softirqentry_text_start 0.08 ± 32% +1.4 1.50 ± 54% perf-profile.children.cycles-pp.ktime_get 0.24 ± 55% +1.6 1.84 ± 44% perf-profile.children.cycles-pp.kthread 0.08 ± 63% +1.6 1.68 ± 69% perf-profile.children.cycles-pp.irq_exit_rcu 0.24 ± 55% +1.6 1.84 ± 44% perf-profile.children.cycles-pp.ret_from_fork 0.18 ± 61% +1.8 2.02 ± 95% perf-profile.children.cycles-pp.update_process_times 0.20 ± 64% +2.0 2.18 ± 97% perf-profile.children.cycles-pp.tick_sched_handle 0.21 ± 64% +2.2 2.43 ± 93% perf-profile.children.cycles-pp.tick_sched_timer 0.09 ±129% +2.9 2.99 ± 92% perf-profile.children.cycles-pp.menu_select 0.30 ± 66% +3.8 4.12 ± 93% perf-profile.children.cycles-pp.__hrtimer_run_queues 0.43 ± 60% +5.8 6.20 ± 84% perf-profile.children.cycles-pp.hrtimer_interrupt 0.44 ± 62% +5.9 6.36 ± 85% perf-profile.children.cycles-pp.__sysvec_apic_timer_interrupt 0.60 ± 64% +9.3 9.91 ± 83% perf-profile.children.cycles-pp.sysvec_apic_timer_interrupt 0.71 ± 60% +10.7 11.44 ± 81% perf-profile.children.cycles-pp.asm_sysvec_apic_timer_interrupt 0.88 ± 82% +21.6 22.48 ± 59% perf-profile.children.cycles-pp.intel_idle 1.40 ± 87% +34.7 36.14 ± 67% perf-profile.children.cycles-pp.cpuidle_enter_state 1.40 ± 87% +34.8 36.18 ± 67% perf-profile.children.cycles-pp.cpuidle_enter 1.53 ± 89% +38.5 40.04 ± 70% perf-profile.children.cycles-pp.start_secondary 1.54 ± 89% +38.8 40.32 ± 70% perf-profile.children.cycles-pp.secondary_startup_64_no_verify 1.54 ± 89% +38.8 40.32 ± 70% perf-profile.children.cycles-pp.cpu_startup_entry 1.54 ± 89% +38.8 40.32 ± 70% perf-profile.children.cycles-pp.do_idle 49.06 ± 15% -45.4 3.62 ±171% perf-profile.self.cycles-pp.native_queued_spin_lock_slowpath 0.23 ± 4% -0.1 0.10 ± 75% perf-profile.self.cycles-pp.__list_del_entry_valid 0.27 ± 8% -0.1 0.14 ± 37% perf-profile.self.cycles-pp.release_pages 0.01 ±223% +0.1 0.11 ± 28% perf-profile.self.cycles-pp._find_next_bit 0.07 ± 20% +0.1 0.18 ± 39% perf-profile.self.cycles-pp.error_entry 0.00 +0.2 0.18 ± 67% perf-profile.self.cycles-pp.asm_sysvec_apic_timer_interrupt 0.00 +0.2 0.19 ± 48% perf-profile.self.cycles-pp._raw_spin_trylock 0.00 +0.2 0.21 ± 71% perf-profile.self.cycles-pp.ktime_get_update_offsets_now 0.00 +0.2 0.21 ± 76% perf-profile.self.cycles-pp.tick_nohz_next_event 0.00 +0.2 0.22 ± 53% perf-profile.self.cycles-pp.__intel_pmu_enable_all 0.00 +0.2 0.23 ±121% perf-profile.self.cycles-pp.update_process_times 0.00 +0.2 0.24 ±100% perf-profile.self.cycles-pp.update_sd_lb_stats 0.01 ±223% +0.3 0.27 ± 82% perf-profile.self.cycles-pp.rcu_sched_clock_irq 0.00 +0.3 0.28 ± 67% perf-profile.self.cycles-pp.arch_scale_freq_tick 0.11 ± 24% +0.3 0.41 ± 57% perf-profile.self.cycles-pp._raw_spin_lock 0.01 ±223% +0.4 0.42 ±118% perf-profile.self.cycles-pp.do_idle 0.01 ±223% +0.5 0.51 ± 95% perf-profile.self.cycles-pp.native_sched_clock 0.01 ±223% +0.6 0.62 ± 89% perf-profile.self.cycles-pp.lapic_next_deadline 0.01 ±223% +0.6 0.64 ±100% perf-profile.self.cycles-pp.read_tsc 0.47 ± 18% +0.7 1.19 ± 11% perf-profile.self.cycles-pp.native_irq_return_iret 0.07 ± 33% +0.9 0.95 ± 35% perf-profile.self.cycles-pp.ktime_get 0.03 ±223% +1.4 1.39 ± 98% perf-profile.self.cycles-pp.menu_select 0.04 ±171% +2.0 2.02 ± 72% perf-profile.self.cycles-pp.cpuidle_enter_state 0.88 ± 82% +21.6 22.48 ± 59% perf-profile.self.cycles-pp.intel_idle 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