Greeting, FYI, we noticed a 3.9% improvement of reaim.jobs_per_min due to commit: commit: f11826446de4c3a65d56ad7ad250e4e17fb79d7c ("mm: don't include in ") https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git master in testcase: reaim on test machine: 144 threads 4 sockets Intel(R) Xeon(R) Gold 5318H CPU @ 2.50GHz with 128G memory with following parameters: runtime: 300s nr_task: 100% test: new_dbase cpufreq_governor: performance ucode: 0x700001e test-description: REAIM is an updated and improved version of AIM 7 benchmark. test-url: https://sourceforge.net/projects/re-aim-7/ In addition to that, the commit also has significant impact on the following tests: +------------------+-------------------------------------------------------------------------------------+ | testcase: change | reaim: reaim.jobs_per_min 1.0% improvement | | test machine | 192 threads 4 sockets Intel(R) Xeon(R) Platinum 9242 CPU @ 2.30GHz with 192G memory | | test parameters | cpufreq_governor=performance | | | nr_task=100% | | | runtime=300s | | | test=dbase | | | ucode=0x5003006 | +------------------+-------------------------------------------------------------------------------------+ Details are as below: --------------------------------------------------------------------------------------------------> To reproduce: git clone https://github.com/intel/lkp-tests.git cd lkp-tests sudo bin/lkp install job.yaml # job file is attached in this email bin/lkp split-job --compatible job.yaml # generate the yaml file for lkp run sudo bin/lkp run generated-yaml-file # if come across any failure that blocks the test, # please remove ~/.lkp and /lkp dir to run from a clean state. ========================================================================================= compiler/cpufreq_governor/kconfig/nr_task/rootfs/runtime/tbox_group/test/testcase/ucode: gcc-9/performance/x86_64-rhel-8.3/100%/debian-10.4-x86_64-20200603.cgz/300s/lkp-cpl-4sp1/new_dbase/reaim/0x700001e commit: 03acadf5a8 ("mm: don't include in ") f11826446d ("mm: don't include in ") 03acadf5a80307b3 f11826446de4c3a65d56ad7ad25 ---------------- --------------------------- %stddev %change %stddev \ | \ 11.36 ± 2% -21.4% 8.92 reaim.child_systime 900112 +3.9% 935573 reaim.jobs_per_min 6250 +3.9% 6497 reaim.jobs_per_min_child 89.04 +1.2% 90.07 reaim.jti 950116 +2.2% 970847 reaim.max_jobs_per_min 0.99 -3.8% 0.95 reaim.parent_time 10.45 -9.7% 9.44 reaim.std_dev_percent 0.09 -12.9% 0.08 reaim.std_dev_time 2.511e+08 +1.5% 2.549e+08 reaim.time.minor_page_faults 1876 -3.0% 1820 reaim.time.percent_of_cpu_this_job_got 1137 ± 2% -20.2% 907.79 reaim.time.system_time 4514 +1.6% 4587 reaim.time.user_time 1589912 +2.1% 1623661 reaim.time.voluntary_context_switches 1440000 +1.5% 1461600 reaim.workload 5312 ± 5% -20.0% 4250 ± 6% interrupts.CPU63.CAL:Function_call_interrupts 14208 +2.4% 14552 vmstat.system.cs 0.10 -0.0 0.09 mpstat.cpu.all.soft% 6.49 ± 2% -1.1 5.34 mpstat.cpu.all.sys% 229.33 ± 5% +32.6% 304.00 ± 15% slabinfo.biovec-64.active_objs 229.33 ± 5% +32.6% 304.00 ± 15% slabinfo.biovec-64.num_objs 93258 ± 6% +13.0% 105408 ± 6% softirqs.CPU1.RCU 92573 ± 3% +9.1% 100979 ± 5% softirqs.CPU54.RCU 94252 ± 4% +10.0% 103668 ± 6% softirqs.CPU63.RCU 49636 ± 2% +4.1% 51671 proc-vmstat.nr_active_anon 61572 ± 2% +4.0% 64027 proc-vmstat.nr_shmem 49636 ± 2% +4.1% 51671 proc-vmstat.nr_zone_active_anon 2.342e+08 +1.5% 2.378e+08 proc-vmstat.numa_hit 2.345e+08 +1.5% 2.38e+08 proc-vmstat.numa_local 361365 +1.9% 368277 proc-vmstat.pgactivate 2.341e+08 +1.5% 2.376e+08 proc-vmstat.pgalloc_normal 2.528e+08 +1.5% 2.566e+08 proc-vmstat.pgfault 2.34e+08 +1.5% 2.375e+08 proc-vmstat.pgfree 16463378 +1.4% 16693131 proc-vmstat.pgreuse 1.72 ± 5% +17.9% 2.02 ± 7% perf-sched.sch_delay.avg.ms.__sched_text_start.__sched_text_start.io_schedule.__lock_page_killable.filemap_fault 1.37 ± 15% -38.6% 0.84 ± 31% perf-sched.sch_delay.avg.ms.__sched_text_start.__sched_text_start.preempt_schedule_common.__cond_resched.__anon_vma_prepare 150.33 ± 33% -52.8% 70.90 ± 30% perf-sched.sch_delay.max.ms.__sched_text_start.__sched_text_start.do_wait.kernel_wait4.__do_sys_wait4 164.50 ± 32% -54.1% 75.46 ± 16% perf-sched.sch_delay.max.ms.__sched_text_start.__sched_text_start.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_sysvec_apic_timer_interrupt 25.97 ± 28% -44.2% 14.49 ± 44% perf-sched.sch_delay.max.ms.__sched_text_start.__sched_text_start.preempt_schedule_common.__cond_resched.__anon_vma_prepare 129289 ± 7% +13.4% 146667 perf-sched.total_wait_and_delay.count.ms 5.36 ± 23% +33.1% 7.14 ± 12% perf-sched.wait_and_delay.avg.ms.__sched_text_start.__sched_text_start.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_exc_page_fault 25.79 ± 6% -12.7% 22.52 perf-sched.wait_and_delay.avg.ms.__sched_text_start.__sched_text_start.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_sysvec_apic_timer_interrupt 102.35 ± 5% -9.3% 92.88 perf-sched.wait_and_delay.avg.ms.__sched_text_start.__sched_text_start.smpboot_thread_fn.kthread.ret_from_fork 29134 ± 8% +11.3% 32428 ± 2% perf-sched.wait_and_delay.count.__sched_text_start.__sched_text_start.do_task_dead.do_exit.do_group_exit 31263 ± 8% +11.7% 34919 ± 2% perf-sched.wait_and_delay.count.__sched_text_start.__sched_text_start.do_wait.kernel_wait4.__do_sys_wait4 11902 ± 9% +18.2% 14064 perf-sched.wait_and_delay.count.__sched_text_start.__sched_text_start.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_sysvec_apic_timer_interrupt 1755 ± 8% +19.3% 2094 ± 4% perf-sched.wait_and_delay.count.__sched_text_start.__sched_text_start.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_sysvec_call_function_single 8019 ± 9% +19.2% 9560 perf-sched.wait_and_delay.count.__sched_text_start.__sched_text_start.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_sysvec_reschedule_ipi 4385 ± 9% +16.3% 5098 ± 2% perf-sched.wait_and_delay.count.__sched_text_start.__sched_text_start.preempt_schedule_common.__cond_resched.stop_one_cpu 82.67 ± 6% +24.6% 103.00 ± 7% perf-sched.wait_and_delay.count.__sched_text_start.__sched_text_start.schedule_hrtimeout_range_clock.do_sys_poll.__x64_sys_poll 11573 ± 8% +14.7% 13274 perf-sched.wait_and_delay.count.__sched_text_start.__sched_text_start.smpboot_thread_fn.kthread.ret_from_fork 4.73 ± 26% +36.1% 6.43 ± 12% perf-sched.wait_time.avg.ms.__sched_text_start.__sched_text_start.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_exc_page_fault 25.49 ± 6% -12.6% 22.27 perf-sched.wait_time.avg.ms.__sched_text_start.__sched_text_start.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_sysvec_apic_timer_interrupt 0.57 ± 31% +1060.6% 6.58 ±180% perf-sched.wait_time.avg.ms.__sched_text_start.__sched_text_start.preempt_schedule_common.__cond_resched.unmap_page_range 13.71 ± 31% -65.7% 4.71 ± 93% perf-sched.wait_time.avg.ms.__sched_text_start.__sched_text_start.preempt_schedule_common.__cond_resched.wp_page_copy 102.35 ± 5% -9.3% 92.87 perf-sched.wait_time.avg.ms.__sched_text_start.__sched_text_start.smpboot_thread_fn.kthread.ret_from_fork 0.43 ± 85% +1876.3% 8.47 ± 58% perf-sched.wait_time.max.ms.__sched_text_start.__sched_text_start.preempt_schedule_common.__cond_resched.__vmalloc_area_node 10.56 ± 41% +1650.8% 184.79 ±199% perf-sched.wait_time.max.ms.__sched_text_start.__sched_text_start.preempt_schedule_common.__cond_resched.unmap_page_range 1.627e+10 +0.9% 1.642e+10 perf-stat.i.branch-instructions 32973172 -8.9% 30048601 perf-stat.i.cache-misses 14218 +2.0% 14501 perf-stat.i.context-switches 6.409e+10 -3.2% 6.205e+10 perf-stat.i.cpu-cycles 2017 +1.0% 2038 perf-stat.i.cpu-migrations 8.326e+09 +0.9% 8.4e+09 perf-stat.i.dTLB-loads 0.06 ± 2% -0.0 0.05 perf-stat.i.dTLB-store-miss-rate% 49.14 +1.8 50.95 perf-stat.i.iTLB-load-miss-rate% 2912410 +1.8% 2965368 perf-stat.i.iTLB-loads 8.246e+10 +0.9% 8.321e+10 perf-stat.i.instructions 0.90 ± 2% +5.4% 0.95 perf-stat.i.ipc 1.00 -2.9% 0.97 perf-stat.i.metric.GHz 419.48 ± 2% -4.1% 402.22 perf-stat.i.metric.K/sec 481.26 +1.2% 486.91 perf-stat.i.metric.M/sec 829642 +0.9% 837144 perf-stat.i.minor-faults 8315387 -7.7% 7674382 perf-stat.i.node-load-misses 72.73 -1.4 71.30 perf-stat.i.node-store-miss-rate% 3045408 -3.6% 2936099 perf-stat.i.node-store-misses 830068 +0.9% 837567 perf-stat.i.page-faults 4.83 -1.4% 4.76 perf-stat.overall.MPKI 8.30 -0.7 7.61 perf-stat.overall.cache-miss-rate% 0.78 -4.1% 0.75 perf-stat.overall.cpi 1938 +6.2% 2058 perf-stat.overall.cycles-between-cache-misses 1.29 +4.3% 1.34 perf-stat.overall.ipc 1.625e+10 +1.2% 1.645e+10 perf-stat.ps.branch-instructions 1.723e+08 +1.3% 1.745e+08 perf-stat.ps.branch-misses 33029764 -8.6% 30195990 perf-stat.ps.cache-misses 14220 +2.3% 14541 perf-stat.ps.context-switches 6.4e+10 -2.9% 6.214e+10 perf-stat.ps.cpu-cycles 2018 +1.3% 2045 perf-stat.ps.cpu-migrations 8.319e+09 +1.2% 8.418e+09 perf-stat.ps.dTLB-loads 3968649 +1.1% 4010723 perf-stat.ps.dTLB-store-misses 5.894e+09 +1.2% 5.968e+09 perf-stat.ps.dTLB-stores 2909823 +2.0% 2969141 perf-stat.ps.iTLB-loads 8.232e+10 +1.2% 8.332e+10 perf-stat.ps.instructions 829765 +1.2% 840028 perf-stat.ps.minor-faults 8327631 -7.4% 7710627 perf-stat.ps.node-load-misses 3049795 -3.3% 2950459 perf-stat.ps.node-store-misses 830191 +1.2% 840454 perf-stat.ps.page-faults 2.486e+13 +1.4% 2.521e+13 perf-stat.total.instructions 4.42 ± 6% -0.9 3.48 ± 8% perf-profile.calltrace.cycles-pp.page_test 3.40 ± 7% -0.9 2.50 ± 10% perf-profile.calltrace.cycles-pp.handle_mm_fault.do_user_addr_fault.exc_page_fault.asm_exc_page_fault.page_test 3.57 ± 7% -0.9 2.68 ± 10% perf-profile.calltrace.cycles-pp.do_user_addr_fault.exc_page_fault.asm_exc_page_fault.page_test 3.58 ± 7% -0.9 2.70 ± 10% perf-profile.calltrace.cycles-pp.exc_page_fault.asm_exc_page_fault.page_test 3.60 ± 7% -0.9 2.72 ± 10% perf-profile.calltrace.cycles-pp.asm_exc_page_fault.page_test 3.95 ± 3% -0.6 3.39 ± 5% perf-profile.calltrace.cycles-pp.brk 3.13 ± 5% -0.6 2.58 ± 6% perf-profile.calltrace.cycles-pp.__x64_sys_brk.do_syscall_64.entry_SYSCALL_64_after_hwframe.brk 3.19 ± 5% -0.6 2.64 ± 6% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.brk 3.16 ± 5% -0.6 2.61 ± 6% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.brk 1.59 ± 13% -0.5 1.04 ± 13% perf-profile.calltrace.cycles-pp.__mem_cgroup_charge.do_anonymous_page.__handle_mm_fault.handle_mm_fault.do_user_addr_fault 2.43 ± 6% -0.5 1.92 ± 9% perf-profile.calltrace.cycles-pp.__do_munmap.__x64_sys_brk.do_syscall_64.entry_SYSCALL_64_after_hwframe.brk 4.48 ± 6% -0.5 3.97 ± 7% perf-profile.calltrace.cycles-pp.__handle_mm_fault.handle_mm_fault.do_user_addr_fault.exc_page_fault.asm_exc_page_fault 2.34 ± 6% -0.5 1.83 ± 9% perf-profile.calltrace.cycles-pp.unmap_region.__do_munmap.__x64_sys_brk.do_syscall_64.entry_SYSCALL_64_after_hwframe 1.50 ± 7% -0.3 1.15 ± 9% perf-profile.calltrace.cycles-pp.release_pages.tlb_flush_mmu.tlb_finish_mmu.unmap_region.__do_munmap 1.78 ± 6% -0.3 1.44 ± 8% perf-profile.calltrace.cycles-pp.tlb_finish_mmu.unmap_region.__do_munmap.__x64_sys_brk.do_syscall_64 1.77 ± 6% -0.3 1.43 ± 8% perf-profile.calltrace.cycles-pp.tlb_flush_mmu.tlb_finish_mmu.unmap_region.__do_munmap.__x64_sys_brk 0.76 ± 10% -0.3 0.47 ± 45% perf-profile.calltrace.cycles-pp.charge_memcg.__mem_cgroup_charge.do_anonymous_page.__handle_mm_fault.handle_mm_fault 0.99 ± 7% -0.2 0.77 ± 8% perf-profile.calltrace.cycles-pp.__mem_cgroup_uncharge_list.release_pages.tlb_flush_mmu.tlb_finish_mmu.unmap_region 0.76 ± 4% -0.1 0.64 ± 6% perf-profile.calltrace.cycles-pp.uncharge_batch.__mem_cgroup_uncharge_list.release_pages.tlb_flush_mmu.tlb_finish_mmu 0.64 ± 4% -0.1 0.57 ± 5% perf-profile.calltrace.cycles-pp.page_counter_uncharge.uncharge_batch.__mem_cgroup_uncharge_list.release_pages.tlb_flush_mmu 0.56 ± 2% -0.0 0.53 ± 2% perf-profile.calltrace.cycles-pp.do_brk_flags.__x64_sys_brk.do_syscall_64.entry_SYSCALL_64_after_hwframe.brk 0.64 +0.0 0.68 ± 6% perf-profile.calltrace.cycles-pp.do_fault.__handle_mm_fault.handle_mm_fault.do_user_addr_fault.exc_page_fault 0.54 +0.1 0.59 ± 6% perf-profile.calltrace.cycles-pp.filemap_map_pages.do_fault.__handle_mm_fault.handle_mm_fault.do_user_addr_fault 1.70 ± 2% +0.2 1.95 ± 9% perf-profile.calltrace.cycles-pp.handle_mm_fault.do_user_addr_fault.exc_page_fault.asm_exc_page_fault 1.84 ± 2% +0.3 2.13 ± 8% perf-profile.calltrace.cycles-pp.do_user_addr_fault.exc_page_fault.asm_exc_page_fault 1.85 ± 2% +0.3 2.14 ± 8% perf-profile.calltrace.cycles-pp.exc_page_fault.asm_exc_page_fault 1.90 ± 2% +0.3 2.20 ± 8% perf-profile.calltrace.cycles-pp.asm_exc_page_fault 5.01 ± 5% -0.9 4.07 ± 7% perf-profile.children.cycles-pp.page_test 6.14 ± 5% -0.7 5.47 ± 7% perf-profile.children.cycles-pp.handle_mm_fault 6.37 ± 4% -0.6 5.77 ± 6% perf-profile.children.cycles-pp.do_user_addr_fault 6.40 ± 4% -0.6 5.81 ± 6% perf-profile.children.cycles-pp.exc_page_fault 4.00 ± 3% -0.6 3.43 ± 5% perf-profile.children.cycles-pp.brk 3.47 ± 8% -0.6 2.90 ± 9% perf-profile.children.cycles-pp.do_anonymous_page 5.41 ± 5% -0.5 4.90 ± 7% perf-profile.children.cycles-pp.__handle_mm_fault 2.02 ± 12% -0.5 1.51 ± 11% perf-profile.children.cycles-pp.__mem_cgroup_charge 1.48 ± 7% -0.4 1.04 ± 13% perf-profile.children.cycles-pp.__mod_memcg_lruvec_state 1.39 ± 8% -0.4 0.96 ± 14% perf-profile.children.cycles-pp.__mod_memcg_state 0.98 ± 7% -0.3 0.69 ± 14% perf-profile.children.cycles-pp.__mod_lruvec_page_state 0.71 ± 8% -0.2 0.49 ± 14% perf-profile.children.cycles-pp.__count_memcg_events 0.67 ± 13% -0.2 0.48 ± 12% perf-profile.children.cycles-pp.get_mem_cgroup_from_mm 0.97 ± 9% -0.2 0.80 ± 8% perf-profile.children.cycles-pp.charge_memcg 0.47 ± 8% -0.1 0.34 ± 13% perf-profile.children.cycles-pp.page_add_new_anon_rmap 0.33 ± 21% -0.1 0.20 ± 19% perf-profile.children.cycles-pp.uncharge_page 0.60 ± 6% -0.1 0.48 ± 8% perf-profile.children.cycles-pp.mod_objcg_state 0.91 ± 4% -0.1 0.82 ± 6% perf-profile.children.cycles-pp.kmem_cache_alloc 0.72 ± 2% -0.1 0.64 ± 7% perf-profile.children.cycles-pp.page_remove_rmap 0.26 ± 10% -0.1 0.20 ± 13% perf-profile.children.cycles-pp.get_mem_cgroup_from_objcg 0.36 ± 3% -0.0 0.33 ± 5% perf-profile.children.cycles-pp.wp_page_copy 0.07 ± 10% -0.0 0.06 ± 13% perf-profile.children.cycles-pp.free_pgd_range 0.07 ± 12% -0.0 0.05 ± 8% perf-profile.children.cycles-pp.free_p4d_range 0.09 ± 7% +0.0 0.11 ± 6% perf-profile.children.cycles-pp.pick_link 0.08 ± 10% +0.0 0.09 ± 13% perf-profile.children.cycles-pp.d_add 0.11 ± 6% +0.0 0.13 ± 4% perf-profile.children.cycles-pp.free_swap_cache 0.20 ± 6% +0.0 0.22 ± 6% perf-profile.children.cycles-pp.__legitimize_path 0.29 ± 5% +0.0 0.32 ± 3% perf-profile.children.cycles-pp._dl_addr 0.22 ± 8% +0.0 0.26 ± 8% perf-profile.children.cycles-pp.up_write 0.95 ± 2% +0.1 1.03 ± 7% perf-profile.children.cycles-pp.filemap_map_pages 0.76 +0.1 0.86 ± 5% perf-profile.children.cycles-pp._raw_spin_lock 1.38 ± 8% -0.4 0.96 ± 14% perf-profile.self.cycles-pp.__mod_memcg_state 0.71 ± 8% -0.2 0.48 ± 14% perf-profile.self.cycles-pp.__count_memcg_events 0.66 ± 13% -0.2 0.47 ± 12% perf-profile.self.cycles-pp.get_mem_cgroup_from_mm 0.48 ± 15% -0.2 0.31 ± 13% perf-profile.self.cycles-pp.charge_memcg 0.38 ± 15% -0.1 0.24 ± 19% perf-profile.self.cycles-pp.__mem_cgroup_charge 0.29 ± 22% -0.1 0.18 ± 19% perf-profile.self.cycles-pp.uncharge_page 0.26 ± 10% -0.1 0.18 ± 10% perf-profile.self.cycles-pp.__mod_lruvec_page_state 0.26 ± 10% -0.1 0.19 ± 13% perf-profile.self.cycles-pp.get_mem_cgroup_from_objcg 0.15 ± 15% -0.0 0.12 ± 8% perf-profile.self.cycles-pp.obj_cgroup_charge_pages 0.21 ± 5% -0.0 0.18 ± 6% perf-profile.self.cycles-pp.mod_objcg_state 0.09 ± 11% -0.0 0.07 ± 14% perf-profile.self.cycles-pp.uncharge_batch 0.08 ± 8% -0.0 0.06 ± 9% perf-profile.self.cycles-pp.get_obj_cgroup_from_current 0.18 ± 5% +0.0 0.20 ± 6% perf-profile.self.cycles-pp.__handle_mm_fault 0.28 ± 6% +0.0 0.31 ± 5% perf-profile.self.cycles-pp._dl_addr 0.21 ± 9% +0.0 0.25 ± 7% perf-profile.self.cycles-pp.up_write 0.29 ± 2% +0.0 0.33 ± 4% perf-profile.self.cycles-pp.page_remove_rmap 0.42 ± 3% +0.0 0.46 ± 4% perf-profile.self.cycles-pp._raw_spin_lock reaim.jobs_per_min 950000 +------------------------------------------------------------------+ | O O O O O O O O O O O O OO O O O O O O O O O O O O | 900000 |.+.+.+.+.+.+.+.+.+.+.+.+.+.+.+.+ +.+. .+. .+ | | : : + + | | : : | 850000 |-+ : : | | : : | 800000 |-+ : : | | : : | 750000 |-+ : : | | : : | | : .+ | 700000 |-+ ++ | | | 650000 +------------------------------------------------------------------+ [*] bisect-good sample [O] bisect-bad sample *************************************************************************************************** lkp-csl-2ap3: 192 threads 4 sockets Intel(R) Xeon(R) Platinum 9242 CPU @ 2.30GHz with 192G memory ========================================================================================= compiler/cpufreq_governor/kconfig/nr_task/rootfs/runtime/tbox_group/test/testcase/ucode: gcc-9/performance/x86_64-rhel-8.3/100%/debian-10.4-x86_64-20200603.cgz/300s/lkp-csl-2ap3/dbase/reaim/0x5003006 commit: 03acadf5a8 ("mm: don't include in ") f11826446d ("mm: don't include in ") 03acadf5a80307b3 f11826446de4c3a65d56ad7ad25 ---------------- --------------------------- %stddev %change %stddev \ | \ 11.09 -9.4% 10.04 reaim.child_systime 855151 +1.0% 864077 reaim.jobs_per_min 4453 +1.0% 4500 reaim.jobs_per_min_child 1.33 -1.0% 1.32 reaim.parent_time 11.58 -2.9% 11.24 reaim.std_dev_percent 0.13 -3.8% 0.12 reaim.std_dev_time 150999 -0.9% 149574 reaim.time.major_page_faults 999.98 -9.4% 905.73 reaim.time.system_time 29446 ± 31% -53.7% 13623 ± 48% softirqs.NET_RX 6.62 -0.3 6.31 perf-stat.overall.cache-miss-rate% 0.76 -1.4% 0.75 perf-stat.overall.cpi 2378 +4.5% 2485 perf-stat.overall.cycles-between-cache-misses 1.32 +1.4% 1.33 perf-stat.overall.ipc 0.16 ± 67% +172.0% 0.43 ± 17% perf-sched.sch_delay.avg.ms.__sched_text_start.__sched_text_start.exit_to_user_mode_prepare.syscall_exit_to_user_mode.do_syscall_64 0.27 ±131% +1018.6% 2.97 ± 65% perf-sched.sch_delay.avg.ms.__sched_text_start.__sched_text_start.preempt_schedule_common.__cond_resched.copy_page_to_iter 0.72 ± 34% -44.0% 0.40 ± 50% perf-sched.sch_delay.avg.ms.__sched_text_start.__sched_text_start.preempt_schedule_common.__cond_resched.copy_pte_range 0.44 ± 9% +34.0% 0.59 ± 14% perf-sched.sch_delay.avg.ms.__sched_text_start.__sched_text_start.preempt_schedule_common.__cond_resched.remove_vma 6.64 ± 31% +142.2% 16.10 ± 42% perf-sched.sch_delay.max.ms.__sched_text_start.__sched_text_start.pipe_read.new_sync_read.vfs_read 1.68 ±132% +466.5% 9.51 ± 40% perf-sched.sch_delay.max.ms.__sched_text_start.__sched_text_start.preempt_schedule_common.__cond_resched.copy_page_to_iter 2.09 ± 41% +199.6% 6.26 ± 35% perf-sched.wait_and_delay.avg.ms.__sched_text_start.__sched_text_start.exit_to_user_mode_prepare.syscall_exit_to_user_mode.do_syscall_64 1703 ± 2% -12.7% 1487 ± 2% perf-sched.wait_and_delay.count.__sched_text_start.__sched_text_start.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_exc_page_fault 1.93 ± 40% +201.9% 5.83 ± 38% perf-sched.wait_time.avg.ms.__sched_text_start.__sched_text_start.exit_to_user_mode_prepare.syscall_exit_to_user_mode.do_syscall_64 2.37 ± 82% +264.4% 8.62 ± 50% perf-sched.wait_time.avg.ms.__sched_text_start.__sched_text_start.io_schedule.__lock_page_killable.filemap_fault 0.24 ± 30% +99.5% 0.47 ± 22% perf-sched.wait_time.avg.ms.__sched_text_start.__sched_text_start.preempt_schedule_common.__cond_resched.down_read 43.01 ±107% -81.1% 8.14 ± 53% perf-sched.wait_time.avg.ms.__sched_text_start.__sched_text_start.preempt_schedule_common.__cond_resched.shmem_undo_range 12.87 ± 36% +111.7% 27.25 ± 25% perf-sched.wait_time.max.ms.__sched_text_start.__sched_text_start.preempt_schedule_common.__cond_resched.down_read 368.26 ±123% -92.9% 25.96 ± 57% perf-sched.wait_time.max.ms.__sched_text_start.__sched_text_start.preempt_schedule_common.__cond_resched.shmem_undo_range 3.41 ± 5% -0.6 2.78 ± 4% perf-profile.calltrace.cycles-pp.page_test 2.63 ± 6% -0.6 2.02 ± 5% perf-profile.calltrace.cycles-pp.exc_page_fault.asm_exc_page_fault.page_test 2.60 ± 6% -0.6 2.00 ± 5% perf-profile.calltrace.cycles-pp.do_user_addr_fault.exc_page_fault.asm_exc_page_fault.page_test 2.64 ± 6% -0.6 2.04 ± 5% perf-profile.calltrace.cycles-pp.asm_exc_page_fault.page_test 2.42 ± 7% -0.6 1.82 ± 5% perf-profile.calltrace.cycles-pp.handle_mm_fault.do_user_addr_fault.exc_page_fault.asm_exc_page_fault.page_test 2.54 ± 4% -0.4 2.11 ± 5% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.brk 2.52 ± 4% -0.4 2.09 ± 5% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.brk 2.48 ± 4% -0.4 2.05 ± 5% perf-profile.calltrace.cycles-pp.__x64_sys_brk.do_syscall_64.entry_SYSCALL_64_after_hwframe.brk 3.30 ± 3% -0.4 2.88 ± 5% perf-profile.calltrace.cycles-pp.brk 1.78 ± 5% -0.4 1.40 ± 6% perf-profile.calltrace.cycles-pp.__do_munmap.__x64_sys_brk.do_syscall_64.entry_SYSCALL_64_after_hwframe.brk 0.95 ± 11% -0.3 0.61 ± 7% perf-profile.calltrace.cycles-pp.__mem_cgroup_charge.do_anonymous_page.__handle_mm_fault.handle_mm_fault.do_user_addr_fault 1.26 ± 4% -0.3 1.00 ± 6% perf-profile.calltrace.cycles-pp.tlb_flush_mmu.tlb_finish_mmu.unmap_region.__do_munmap.__x64_sys_brk 1.02 ± 5% -0.3 0.76 ± 5% perf-profile.calltrace.cycles-pp.release_pages.tlb_flush_mmu.tlb_finish_mmu.unmap_region.__do_munmap 1.27 ± 4% -0.3 1.00 ± 6% perf-profile.calltrace.cycles-pp.tlb_finish_mmu.unmap_region.__do_munmap.__x64_sys_brk.do_syscall_64 0.57 ± 2% -0.1 0.44 ± 44% perf-profile.calltrace.cycles-pp.do_brk_flags.__x64_sys_brk.do_syscall_64.entry_SYSCALL_64_after_hwframe.brk 1.15 +0.2 1.39 ± 15% perf-profile.calltrace.cycles-pp.cpuidle_enter_state.cpuidle_enter.do_idle.cpu_startup_entry.start_secondary 1.20 +0.3 1.45 ± 15% perf-profile.calltrace.cycles-pp.cpuidle_enter.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64_no_verify 1.60 +0.3 1.88 ± 16% perf-profile.calltrace.cycles-pp.cpu_startup_entry.start_secondary.secondary_startup_64_no_verify 1.60 +0.3 1.88 ± 16% perf-profile.calltrace.cycles-pp.start_secondary.secondary_startup_64_no_verify 1.60 +0.3 1.88 ± 16% perf-profile.calltrace.cycles-pp.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64_no_verify 1.63 +0.3 1.91 ± 15% perf-profile.calltrace.cycles-pp.secondary_startup_64_no_verify 4.00 ± 4% -0.6 3.39 ± 5% perf-profile.children.cycles-pp.page_test 3.34 ± 3% -0.4 2.92 ± 5% perf-profile.children.cycles-pp.brk 0.79 ± 11% -0.3 0.48 ± 14% perf-profile.children.cycles-pp.__mod_memcg_state 0.87 ± 10% -0.3 0.56 ± 14% perf-profile.children.cycles-pp.__mod_memcg_lruvec_state 1.22 ± 10% -0.3 0.92 ± 14% perf-profile.children.cycles-pp.__mem_cgroup_charge 0.62 ± 9% -0.2 0.44 ± 14% perf-profile.children.cycles-pp.__mod_lruvec_page_state 0.39 ± 16% -0.1 0.25 ± 14% perf-profile.children.cycles-pp.lock_page_lruvec_irqsave 0.42 ± 9% -0.1 0.29 ± 20% perf-profile.children.cycles-pp.__count_memcg_events 0.39 ± 11% -0.1 0.27 ± 18% perf-profile.children.cycles-pp.get_mem_cgroup_from_mm 0.42 ± 15% -0.1 0.30 ± 12% perf-profile.children.cycles-pp._raw_spin_lock_irqsave 0.58 ± 11% -0.1 0.47 ± 11% perf-profile.children.cycles-pp.native_queued_spin_lock_slowpath 0.17 ± 20% -0.1 0.09 ± 16% perf-profile.children.cycles-pp.uncharge_page 0.36 ± 7% -0.1 0.28 ± 15% perf-profile.children.cycles-pp.mod_objcg_state 0.29 ± 9% -0.1 0.22 ± 15% perf-profile.children.cycles-pp.page_add_new_anon_rmap 0.26 ± 7% -0.1 0.19 ± 11% perf-profile.children.cycles-pp.lru_add_drain_cpu 0.25 ± 9% -0.1 0.19 ± 10% perf-profile.children.cycles-pp.lru_add_drain 0.16 ± 7% -0.1 0.10 ± 17% perf-profile.children.cycles-pp.get_mem_cgroup_from_objcg 0.07 ± 6% +0.0 0.09 ± 12% perf-profile.children.cycles-pp.simple_lookup 0.10 ± 9% +0.0 0.12 ± 14% perf-profile.children.cycles-pp.free_swap_cache 0.16 ± 4% +0.0 0.19 ± 11% perf-profile.children.cycles-pp.__d_lookup_rcu 0.30 ± 3% +0.0 0.35 ± 13% perf-profile.children.cycles-pp._dl_addr 0.32 ± 4% +0.1 0.38 ± 12% perf-profile.children.cycles-pp.vma_interval_tree_insert 0.30 ± 4% +0.1 0.35 ± 10% perf-profile.children.cycles-pp.lookup_fast 0.20 ± 5% +0.1 0.26 ± 10% perf-profile.children.cycles-pp.clockevents_program_event 0.22 ± 9% +0.1 0.29 ± 10% perf-profile.children.cycles-pp.ktime_get 0.36 ± 4% +0.1 0.42 ± 11% perf-profile.children.cycles-pp.dput 0.48 ± 3% +0.1 0.56 ± 11% perf-profile.children.cycles-pp.next_uptodate_page 0.46 ± 3% +0.1 0.55 ± 18% perf-profile.children.cycles-pp.intel_idle 0.62 ± 2% +0.1 0.73 ± 13% perf-profile.children.cycles-pp.filename_lookup 0.62 ± 5% +0.1 0.73 ± 13% perf-profile.children.cycles-pp._raw_spin_lock 0.62 ± 2% +0.1 0.72 ± 13% perf-profile.children.cycles-pp.path_lookupat 0.65 +0.1 0.76 ± 12% perf-profile.children.cycles-pp.vfs_statx 0.67 ± 2% +0.1 0.78 ± 13% perf-profile.children.cycles-pp.user_path_at_empty 0.66 ± 2% +0.1 0.77 ± 12% perf-profile.children.cycles-pp.__do_sys_newstat 0.84 ± 2% +0.1 0.96 ± 12% perf-profile.children.cycles-pp.filemap_map_pages 1.01 ± 4% +0.1 1.14 ± 10% perf-profile.children.cycles-pp.link_path_walk 1.07 ± 3% +0.2 1.23 ± 12% perf-profile.children.cycles-pp.walk_component 1.22 +0.3 1.48 ± 14% perf-profile.children.cycles-pp.cpuidle_enter 1.22 +0.3 1.48 ± 14% perf-profile.children.cycles-pp.cpuidle_enter_state 1.60 +0.3 1.88 ± 16% perf-profile.children.cycles-pp.start_secondary 1.63 +0.3 1.91 ± 15% perf-profile.children.cycles-pp.secondary_startup_64_no_verify 1.63 +0.3 1.91 ± 15% perf-profile.children.cycles-pp.cpu_startup_entry 1.63 +0.3 1.91 ± 15% perf-profile.children.cycles-pp.do_idle 0.79 ± 11% -0.3 0.48 ± 14% perf-profile.self.cycles-pp.__mod_memcg_state 0.42 ± 9% -0.1 0.28 ± 19% perf-profile.self.cycles-pp.__count_memcg_events 0.38 ± 12% -0.1 0.26 ± 18% perf-profile.self.cycles-pp.get_mem_cgroup_from_mm 0.58 ± 11% -0.1 0.47 ± 11% perf-profile.self.cycles-pp.native_queued_spin_lock_slowpath 0.25 ± 17% -0.1 0.15 ± 11% perf-profile.self.cycles-pp.charge_memcg 0.19 ± 16% -0.1 0.10 ± 10% perf-profile.self.cycles-pp.__mem_cgroup_charge 0.15 ± 21% -0.1 0.08 ± 16% perf-profile.self.cycles-pp.uncharge_page 0.16 ± 9% -0.1 0.10 ± 18% perf-profile.self.cycles-pp.get_mem_cgroup_from_objcg 0.09 ± 12% -0.0 0.06 ± 15% perf-profile.self.cycles-pp.obj_cgroup_charge_pages 0.09 ± 6% +0.0 0.11 ± 8% perf-profile.self.cycles-pp._raw_spin_lock_irqsave 0.07 +0.0 0.09 ± 15% perf-profile.self.cycles-pp.cpuidle_enter_state 0.16 ± 4% +0.0 0.19 ± 11% perf-profile.self.cycles-pp.__d_lookup_rcu 0.02 ±141% +0.0 0.06 ± 13% perf-profile.self.cycles-pp.lockref_put_or_lock 0.30 ± 2% +0.0 0.35 ± 14% perf-profile.self.cycles-pp._dl_addr 0.20 ± 10% +0.1 0.26 ± 11% perf-profile.self.cycles-pp.ktime_get 0.37 ± 4% +0.1 0.45 ± 16% perf-profile.self.cycles-pp._raw_spin_lock 0.46 ± 3% +0.1 0.55 ± 18% 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