Greeting, FYI, we noticed a -3.7% regression of unixbench.score due to commit: commit: a23a7dc57615679ee21ac350c36bde4f863ab249 ("[RFC][PATCH] fanotify: introduce filesystem view mark") url: https://github.com/0day-ci/linux/commits/Amir-Goldstein/fanotify-introduce-filesystem-view-mark/20201110-020535 base: https://git.kernel.org/cgit/linux/kernel/git/jack/linux-fs.git fsnotify in testcase: unixbench on test machine: 96 threads Intel(R) Xeon(R) CPU @ 2.30GHz with 128G memory with following parameters: runtime: 300s nr_task: 30% test: pipe cpufreq_governor: performance ucode: 0x4002f01 test-description: UnixBench is the original BYTE UNIX benchmark suite aims to test performance of Unix-like system. test-url: https://github.com/kdlucas/byte-unixbench 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 run job.yaml ========================================================================================= compiler/cpufreq_governor/kconfig/nr_task/rootfs/runtime/tbox_group/test/testcase/ucode: gcc-9/performance/x86_64-rhel-8.3/30%/debian-10.4-x86_64-20200603.cgz/300s/lkp-csl-2sp4/pipe/unixbench/0x4002f01 commit: 9289012374 ("inotify: Increase default inotify.max_user_watches limit to 1048576") a23a7dc576 ("fanotify: introduce filesystem view mark") 92890123749bafc3 a23a7dc57615679ee21ac350c36 ---------------- --------------------------- %stddev %change %stddev \ | \ 41413 -3.7% 39864 unixbench.score 943.86 -2.2% 922.97 unixbench.time.user_time 2.01e+10 -3.7% 1.937e+10 unixbench.workload 24527 ± 14% +19.8% 29395 ± 2% numa-meminfo.node1.Shmem 6126 ± 14% +19.9% 7344 ± 2% numa-vmstat.node1.nr_shmem 1621 ± 2% -3.0% 1572 vmstat.system.cs 924.00 ± 5% -12.2% 811.50 ± 9% slabinfo.buffer_head.active_objs 924.00 ± 5% -12.2% 811.50 ± 9% slabinfo.buffer_head.num_objs 0.00 ± 4% +53.7% 0.00 ± 24% sched_debug.cpu.next_balance.stddev 34803 ± 29% -47.8% 18172 ± 18% sched_debug.cpu.nr_switches.max 4463 ± 18% -30.2% 3116 ± 8% sched_debug.cpu.nr_switches.stddev 31716 ± 33% -52.0% 15219 ± 26% sched_debug.cpu.sched_count.max 3928 ± 25% -37.1% 2471 ± 14% sched_debug.cpu.sched_count.stddev 15722 ± 33% -52.5% 7466 ± 27% sched_debug.cpu.sched_goidle.max 1968 ± 25% -37.3% 1235 ± 14% sched_debug.cpu.sched_goidle.stddev 8399 ± 76% -57.9% 3532 ± 43% sched_debug.cpu.ttwu_local.max 16361 ± 16% -19.6% 13155 ± 6% softirqs.CPU14.RCU 16928 ± 13% -14.9% 14408 ± 4% softirqs.CPU16.RCU 17013 ± 13% -16.1% 14273 ± 5% softirqs.CPU17.RCU 16952 ± 12% -15.8% 14278 ± 5% softirqs.CPU18.RCU 16941 ± 12% -17.1% 14038 ± 5% softirqs.CPU19.RCU 17005 ± 12% -16.3% 14239 ± 4% softirqs.CPU20.RCU 17035 ± 12% -21.7% 13337 ± 9% softirqs.CPU21.RCU 17190 ± 11% -17.3% 14219 ± 4% softirqs.CPU23.RCU 16941 ± 11% -24.8% 12739 ± 5% softirqs.CPU48.RCU 39459 ± 4% -12.2% 34647 ± 4% softirqs.CPU55.SCHED 38878 ± 10% -11.7% 34325 ± 3% softirqs.CPU56.SCHED 38813 ± 5% -10.7% 34663 ± 2% softirqs.CPU57.SCHED 43268 ± 6% -13.9% 37250 ± 9% softirqs.CPU67.SCHED 42397 ± 8% -14.7% 36159 ± 2% softirqs.CPU68.SCHED 17926 ± 12% -14.9% 15255 ± 3% softirqs.CPU69.RCU 0.50 ±173% +9750.0% 49.25 ±140% interrupts.75:PCI-MSI.31981608-edge.i40e-eth0-TxRx-39 9.50 ±107% +1397.4% 142.25 ± 83% interrupts.CPU1.TLB:TLB_shootdowns 2302 ± 71% +83.9% 4233 ± 37% interrupts.CPU2.NMI:Non-maskable_interrupts 2302 ± 71% +83.9% 4233 ± 37% interrupts.CPU2.PMI:Performance_monitoring_interrupts 2775 ± 48% +53.6% 4263 ± 37% interrupts.CPU3.NMI:Non-maskable_interrupts 2775 ± 48% +53.6% 4263 ± 37% interrupts.CPU3.PMI:Performance_monitoring_interrupts 27.75 ± 79% +847.7% 263.00 ± 30% interrupts.CPU30.TLB:TLB_shootdowns 0.50 ±173% +9550.0% 48.25 ±143% interrupts.CPU39.75:PCI-MSI.31981608-edge.i40e-eth0-TxRx-39 106.25 ± 96% -96.0% 4.25 ± 10% interrupts.CPU42.TLB:TLB_shootdowns 1979 ± 26% +72.2% 3408 ± 24% interrupts.CPU48.NMI:Non-maskable_interrupts 1979 ± 26% +72.2% 3408 ± 24% interrupts.CPU48.PMI:Performance_monitoring_interrupts 1730 ± 50% +106.7% 3577 ± 25% interrupts.CPU50.NMI:Non-maskable_interrupts 1730 ± 50% +106.7% 3577 ± 25% interrupts.CPU50.PMI:Performance_monitoring_interrupts 42.25 ± 49% +221.3% 135.75 ± 53% interrupts.CPU56.RES:Rescheduling_interrupts 2.00 ± 50% +2750.0% 57.00 ±142% interrupts.CPU57.TLB:TLB_shootdowns 451.00 ± 5% +78.8% 806.50 ± 22% interrupts.CPU58.CAL:Function_call_interrupts 59.75 ± 33% +92.5% 115.00 ± 55% interrupts.CPU58.RES:Rescheduling_interrupts 2.00 ± 50% +3287.5% 67.75 ±162% interrupts.CPU58.TLB:TLB_shootdowns 3245 ± 53% +73.9% 5642 ± 7% interrupts.CPU6.NMI:Non-maskable_interrupts 3245 ± 53% +73.9% 5642 ± 7% interrupts.CPU6.PMI:Performance_monitoring_interrupts 130.00 ± 31% -56.9% 56.00 ± 23% interrupts.CPU61.RES:Rescheduling_interrupts 3.25 ± 33% +3930.8% 131.00 ±166% interrupts.CPU74.TLB:TLB_shootdowns 2751 ± 25% -61.9% 1049 ±116% interrupts.CPU78.NMI:Non-maskable_interrupts 2751 ± 25% -61.9% 1049 ±116% interrupts.CPU78.PMI:Performance_monitoring_interrupts 595.75 ± 28% +59.8% 951.75 ± 45% interrupts.CPU88.CAL:Function_call_interrupts 3193 ± 31% +75.1% 5590 ± 10% interrupts.CPU9.NMI:Non-maskable_interrupts 3193 ± 31% +75.1% 5590 ± 10% interrupts.CPU9.PMI:Performance_monitoring_interrupts 2.03 ± 9% -21.7% 1.59 ± 10% perf-stat.i.MPKI 2.045e+10 -3.8% 1.968e+10 perf-stat.i.branch-instructions 1.765e+08 -4.1% 1.693e+08 perf-stat.i.branch-misses 15.34 ± 2% -2.3 13.00 ± 5% perf-stat.i.cache-miss-rate% 1361616 ± 5% -30.6% 945163 ± 11% perf-stat.i.cache-misses 8752341 ± 2% -20.0% 7004871 ± 6% perf-stat.i.cache-references 1592 ± 2% -3.0% 1545 perf-stat.i.context-switches 61359 ± 4% +51.4% 92911 ± 9% perf-stat.i.cycles-between-cache-misses 83930 ± 7% -18.4% 68483 ± 9% perf-stat.i.dTLB-load-misses 3.066e+10 -3.4% 2.962e+10 perf-stat.i.dTLB-loads 0.00 ± 2% +0.0 0.00 ± 6% perf-stat.i.dTLB-store-miss-rate% 1.924e+10 -6.7% 1.795e+10 perf-stat.i.dTLB-stores 1.756e+08 -4.4% 1.679e+08 perf-stat.i.iTLB-load-misses 1.019e+11 -4.0% 9.778e+10 perf-stat.i.instructions 1.28 -3.4% 1.24 perf-stat.i.ipc 1.21 +11.7% 1.36 ± 6% perf-stat.i.metric.K/sec 732.97 -4.4% 700.60 perf-stat.i.metric.M/sec 86.90 -1.0 85.92 perf-stat.i.node-load-miss-rate% 166355 -26.0% 123022 ± 4% perf-stat.i.node-load-misses 28716 ± 5% -13.8% 24766 ± 6% perf-stat.i.node-loads 0.09 ± 2% -16.6% 0.07 ± 6% perf-stat.overall.MPKI 15.54 ± 2% -2.1 13.44 ± 5% perf-stat.overall.cache-miss-rate% 0.67 +3.7% 0.70 perf-stat.overall.cpi 50659 ± 5% +44.8% 73365 ± 10% perf-stat.overall.cycles-between-cache-misses 0.00 +0.0 0.00 perf-stat.overall.dTLB-store-miss-rate% 1.48 -3.6% 1.43 perf-stat.overall.ipc 85.27 -2.0 83.24 perf-stat.overall.node-load-miss-rate% 2.044e+10 -3.8% 1.967e+10 perf-stat.ps.branch-instructions 1.763e+08 -4.0% 1.692e+08 perf-stat.ps.branch-misses 1357072 ± 5% -30.6% 941556 ± 11% perf-stat.ps.cache-misses 8725659 ± 2% -20.0% 6983651 ± 6% perf-stat.ps.cache-references 1588 ± 2% -3.0% 1541 perf-stat.ps.context-switches 83826 ± 7% -18.4% 68430 ± 9% perf-stat.ps.dTLB-load-misses 3.063e+10 -3.3% 2.961e+10 perf-stat.ps.dTLB-loads 1.923e+10 -6.7% 1.794e+10 perf-stat.ps.dTLB-stores 1.755e+08 -4.4% 1.678e+08 perf-stat.ps.iTLB-load-misses 1.018e+11 -4.0% 9.774e+10 perf-stat.ps.instructions 165887 -26.1% 122672 ± 4% perf-stat.ps.node-load-misses 28657 ± 5% -13.8% 24710 ± 6% perf-stat.ps.node-loads 3.99e+13 -3.9% 3.834e+13 perf-stat.total.instructions 46.88 ± 8% -6.7 40.22 ± 6% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe 43.90 ± 8% -6.0 37.91 ± 6% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe 21.23 ± 8% -3.9 17.29 ± 4% perf-profile.calltrace.cycles-pp.new_sync_write.vfs_write.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe 19.96 ± 9% -3.7 16.22 ± 4% perf-profile.calltrace.cycles-pp.pipe_write.new_sync_write.vfs_write.ksys_write.do_syscall_64 22.16 ± 8% -3.2 18.94 ± 6% perf-profile.calltrace.cycles-pp.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe 17.24 ± 8% -3.2 14.06 ± 5% perf-profile.calltrace.cycles-pp.new_sync_read.vfs_read.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe 20.52 ± 8% -2.9 17.57 ± 7% perf-profile.calltrace.cycles-pp.vfs_write.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe 15.90 ± 8% -2.9 12.97 ± 5% perf-profile.calltrace.cycles-pp.pipe_read.new_sync_read.vfs_read.ksys_read.do_syscall_64 19.78 ± 8% -2.5 17.32 ± 6% perf-profile.calltrace.cycles-pp.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe 18.21 ± 8% -2.2 16.05 ± 6% perf-profile.calltrace.cycles-pp.vfs_read.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe 7.55 ± 10% -1.4 6.11 ± 4% perf-profile.calltrace.cycles-pp.write 6.60 ± 9% -1.3 5.35 ± 5% perf-profile.calltrace.cycles-pp.copy_page_from_iter.pipe_write.new_sync_write.vfs_write.ksys_write 7.02 ± 9% -1.2 5.80 ± 4% perf-profile.calltrace.cycles-pp.read 6.54 ± 8% -1.2 5.33 ± 4% perf-profile.calltrace.cycles-pp.copy_page_to_iter.pipe_read.new_sync_read.vfs_read.ksys_read 6.21 ± 10% -1.2 5.04 ± 4% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.write 5.84 ± 10% -1.1 4.76 ± 3% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.write 5.61 ± 10% -1.0 4.58 ± 4% perf-profile.calltrace.cycles-pp.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe.write 5.69 ± 9% -0.9 4.75 ± 4% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.read 5.34 ± 9% -0.9 4.47 ± 4% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.read 5.10 ± 9% -0.8 4.29 ± 4% perf-profile.calltrace.cycles-pp.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe.read 4.14 ± 8% -0.7 3.40 ± 4% perf-profile.calltrace.cycles-pp.copyout.copy_page_to_iter.pipe_read.new_sync_read.vfs_read 3.98 ± 9% -0.7 3.26 ± 6% perf-profile.calltrace.cycles-pp.copyin.copy_page_from_iter.pipe_write.new_sync_write.vfs_write 3.60 ± 8% -0.6 2.95 ± 4% perf-profile.calltrace.cycles-pp.copy_user_enhanced_fast_string.copyout.copy_page_to_iter.pipe_read.new_sync_read 3.48 ± 9% -0.6 2.87 ± 6% perf-profile.calltrace.cycles-pp.copy_user_enhanced_fast_string.copyin.copy_page_from_iter.pipe_write.new_sync_write 3.51 ± 8% -0.6 2.90 ± 5% perf-profile.calltrace.cycles-pp.__wake_up_common_lock.pipe_write.new_sync_write.vfs_write.ksys_write 2.33 ± 9% -0.4 1.93 ± 5% perf-profile.calltrace.cycles-pp.__entry_text_start 1.45 ± 6% -0.4 1.07 ± 3% perf-profile.calltrace.cycles-pp.syscall_exit_to_user_mode.entry_SYSCALL_64_after_hwframe 1.71 ± 7% -0.3 1.38 ± 7% perf-profile.calltrace.cycles-pp.mutex_lock.pipe_read.new_sync_read.vfs_read.ksys_read 1.64 ± 8% -0.3 1.34 ± 6% perf-profile.calltrace.cycles-pp.mutex_lock.pipe_write.new_sync_write.vfs_write.ksys_write 1.11 ± 9% -0.2 0.89 ± 6% perf-profile.calltrace.cycles-pp.__fdget_pos.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe 0.98 ± 10% -0.2 0.78 ± 5% perf-profile.calltrace.cycles-pp.__fget_light.__fdget_pos.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe 1.13 ± 7% -0.2 0.94 ± 5% perf-profile.calltrace.cycles-pp.__fdget_pos.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe 0.89 ± 7% -0.2 0.71 ± 8% perf-profile.calltrace.cycles-pp._raw_spin_unlock_irqrestore.__wake_up_common_lock.pipe_write.new_sync_write.vfs_write 1.22 ± 7% -0.2 1.05 ± 5% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.__wake_up_common_lock.pipe_write.new_sync_write.vfs_write 1.20 ± 8% -0.2 1.02 ± 8% perf-profile.calltrace.cycles-pp.touch_atime.pipe_read.new_sync_read.vfs_read.ksys_read 1.42 ± 7% -0.2 1.24 ± 6% perf-profile.calltrace.cycles-pp.syscall_return_via_sysret 0.93 ± 7% -0.2 0.76 ± 11% perf-profile.calltrace.cycles-pp.anon_pipe_buf_release.pipe_read.new_sync_read.vfs_read.ksys_read 0.75 ± 8% -0.2 0.58 ± 7% perf-profile.calltrace.cycles-pp.mutex_unlock.pipe_read.new_sync_read.vfs_read.ksys_read 0.99 ± 8% -0.2 0.82 ± 7% perf-profile.calltrace.cycles-pp.__fget_light.__fdget_pos.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe 1.35 ± 7% -0.2 1.18 ± 8% perf-profile.calltrace.cycles-pp.syscall_enter_from_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe 0.90 ± 9% -0.2 0.74 ± 5% perf-profile.calltrace.cycles-pp._raw_spin_lock_irq.pipe_read.new_sync_read.vfs_read.ksys_read 0.73 ± 9% -0.1 0.58 ± 7% perf-profile.calltrace.cycles-pp.mutex_unlock.pipe_write.new_sync_write.vfs_write.ksys_write 0.82 ± 6% -0.1 0.68 ± 6% perf-profile.calltrace.cycles-pp._raw_spin_lock_irq.pipe_write.new_sync_write.vfs_write.ksys_write 0.97 ± 8% -0.1 0.83 ± 7% perf-profile.calltrace.cycles-pp.atime_needs_update.touch_atime.pipe_read.new_sync_read.vfs_read 0.71 ± 8% -0.1 0.57 ± 7% perf-profile.calltrace.cycles-pp.__sb_end_write.pipe_write.new_sync_write.vfs_write.ksys_write 0.64 ± 10% -0.1 0.54 ± 8% perf-profile.calltrace.cycles-pp.current_time.atime_needs_update.touch_atime.pipe_read.new_sync_read 0.68 ± 7% -0.1 0.59 ± 5% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_safe_stack 0.85 ± 7% +0.3 1.12 ± 5% perf-profile.calltrace.cycles-pp.fsnotify.vfs_read.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe 0.76 ± 6% +0.4 1.17 ± 11% perf-profile.calltrace.cycles-pp.fsnotify.vfs_write.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe 0.60 ± 7% +0.4 1.04 ± 7% perf-profile.calltrace.cycles-pp.fsnotify.security_file_permission.vfs_read.ksys_read.do_syscall_64 30.90 ± 18% +10.5 41.43 ± 6% perf-profile.calltrace.cycles-pp.secondary_startup_64_no_verify 30.70 ± 18% +10.6 41.30 ± 7% perf-profile.calltrace.cycles-pp.cpu_startup_entry.start_secondary.secondary_startup_64_no_verify 30.70 ± 18% +10.6 41.30 ± 7% perf-profile.calltrace.cycles-pp.start_secondary.secondary_startup_64_no_verify 30.70 ± 18% +10.6 41.30 ± 7% perf-profile.calltrace.cycles-pp.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64_no_verify 30.16 ± 19% +10.8 40.99 ± 7% perf-profile.calltrace.cycles-pp.cpuidle_enter.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64_no_verify 30.02 ± 19% +10.9 40.91 ± 7% perf-profile.calltrace.cycles-pp.cpuidle_enter_state.cpuidle_enter.do_idle.cpu_startup_entry.start_secondary 29.11 ± 20% +11.0 40.09 ± 7% perf-profile.calltrace.cycles-pp.intel_idle.cpuidle_enter_state.cpuidle_enter.do_idle.cpu_startup_entry 58.91 ± 8% -8.8 50.13 ± 5% perf-profile.children.cycles-pp.entry_SYSCALL_64_after_hwframe 55.24 ± 8% -8.0 47.28 ± 5% perf-profile.children.cycles-pp.do_syscall_64 27.80 ± 8% -4.3 23.52 ± 5% perf-profile.children.cycles-pp.ksys_write 21.30 ± 8% -4.0 17.35 ± 4% perf-profile.children.cycles-pp.new_sync_write 25.84 ± 8% -3.9 21.94 ± 5% perf-profile.children.cycles-pp.vfs_write 20.19 ± 9% -3.8 16.39 ± 4% perf-profile.children.cycles-pp.pipe_write 24.93 ± 8% -3.3 21.64 ± 4% perf-profile.children.cycles-pp.ksys_read 17.32 ± 8% -3.2 14.13 ± 5% perf-profile.children.cycles-pp.new_sync_read 16.12 ± 8% -3.0 13.15 ± 5% perf-profile.children.cycles-pp.pipe_read 23.04 ± 8% -2.9 20.12 ± 4% perf-profile.children.cycles-pp.vfs_read 7.90 ± 10% -1.5 6.38 ± 4% perf-profile.children.cycles-pp.write 7.71 ± 8% -1.4 6.33 ± 5% perf-profile.children.cycles-pp.copy_user_enhanced_fast_string 6.73 ± 9% -1.3 5.42 ± 5% perf-profile.children.cycles-pp.copy_page_from_iter 7.38 ± 9% -1.3 6.08 ± 4% perf-profile.children.cycles-pp.read 6.71 ± 8% -1.2 5.47 ± 5% perf-profile.children.cycles-pp.copy_page_to_iter 5.39 ± 8% -0.9 4.53 ± 5% perf-profile.children.cycles-pp.syscall_return_via_sysret 4.36 ± 8% -0.9 3.50 ± 4% perf-profile.children.cycles-pp.mutex_lock 4.18 ± 8% -0.7 3.43 ± 4% perf-profile.children.cycles-pp.copyout 4.03 ± 9% -0.7 3.30 ± 6% perf-profile.children.cycles-pp.copyin 3.58 ± 9% -0.7 2.88 ± 2% perf-profile.children.cycles-pp.__entry_text_start 3.58 ± 9% -0.6 2.95 ± 5% perf-profile.children.cycles-pp.__wake_up_common_lock 2.96 ± 8% -0.6 2.41 ± 4% perf-profile.children.cycles-pp.__fdget_pos 2.00 ± 7% -0.5 1.49 ± 2% perf-profile.children.cycles-pp.syscall_exit_to_user_mode 3.59 ± 7% -0.5 3.10 ± 7% perf-profile.children.cycles-pp.common_file_perm 2.44 ± 8% -0.5 1.96 ± 5% perf-profile.children.cycles-pp.__fget_light 1.92 ± 8% -0.4 1.50 ± 5% perf-profile.children.cycles-pp.mutex_unlock 1.95 ± 9% -0.4 1.56 ± 5% perf-profile.children.cycles-pp.___might_sleep 2.21 ± 8% -0.4 1.82 ± 4% perf-profile.children.cycles-pp._raw_spin_lock_irq 1.76 ± 9% -0.4 1.39 ± 3% perf-profile.children.cycles-pp.__might_sleep 1.22 ± 9% -0.3 0.96 ± 3% perf-profile.children.cycles-pp.__might_fault 0.51 ± 21% -0.2 0.27 ± 17% perf-profile.children.cycles-pp.menu_select 1.75 ± 7% -0.2 1.51 ± 6% perf-profile.children.cycles-pp.syscall_enter_from_user_mode 1.53 ± 8% -0.2 1.31 ± 5% perf-profile.children.cycles-pp._raw_spin_lock_irqsave 1.11 ± 8% -0.2 0.89 ± 6% perf-profile.children.cycles-pp._raw_spin_unlock_irqrestore 1.56 ± 8% -0.2 1.34 ± 6% perf-profile.children.cycles-pp.touch_atime 1.19 ± 7% -0.2 0.97 ± 8% perf-profile.children.cycles-pp.anon_pipe_buf_release 1.08 ± 8% -0.2 0.88 ± 4% perf-profile.children.cycles-pp._cond_resched 0.93 ± 8% -0.2 0.73 ± 5% perf-profile.children.cycles-pp.__sb_end_write 1.27 ± 9% -0.2 1.09 ± 6% perf-profile.children.cycles-pp.atime_needs_update 0.96 ± 8% -0.2 0.78 ± 5% perf-profile.children.cycles-pp.aa_file_perm 0.78 ± 6% -0.2 0.61 ± 3% perf-profile.children.cycles-pp.exit_to_user_mode_prepare 0.93 ± 8% -0.1 0.78 ± 4% perf-profile.children.cycles-pp.entry_SYSCALL_64_safe_stack 0.67 ± 7% -0.1 0.53 ± 4% perf-profile.children.cycles-pp.__sb_start_write 0.54 ± 9% -0.1 0.44 ± 3% perf-profile.children.cycles-pp.rcu_all_qs 0.45 ± 10% -0.1 0.36 ± 5% perf-profile.children.cycles-pp.__wake_up_common 0.43 ± 8% -0.1 0.35 ± 6% perf-profile.children.cycles-pp.__x86_indirect_thunk_rax 0.39 ± 5% -0.1 0.32 ± 3% perf-profile.children.cycles-pp.rw_verify_area 0.26 ± 7% -0.0 0.21 ± 4% perf-profile.children.cycles-pp.kill_fasync 0.28 ± 5% -0.0 0.23 ± 7% perf-profile.children.cycles-pp.iov_iter_init 0.17 ± 5% -0.0 0.14 ± 8% perf-profile.children.cycles-pp.rcu_read_unlock_strict 0.10 ± 7% -0.0 0.08 ± 5% perf-profile.children.cycles-pp.__x64_sys_read 0.10 ± 8% -0.0 0.08 perf-profile.children.cycles-pp.__x64_sys_write 2.84 ± 7% +1.4 4.25 ± 5% perf-profile.children.cycles-pp.fsnotify 30.90 ± 18% +10.5 41.43 ± 6% perf-profile.children.cycles-pp.secondary_startup_64_no_verify 30.90 ± 18% +10.5 41.43 ± 6% perf-profile.children.cycles-pp.cpu_startup_entry 30.90 ± 18% +10.5 41.43 ± 6% perf-profile.children.cycles-pp.do_idle 30.70 ± 18% +10.6 41.30 ± 7% perf-profile.children.cycles-pp.start_secondary 30.36 ± 19% +10.8 41.12 ± 7% perf-profile.children.cycles-pp.cpuidle_enter 30.35 ± 19% +10.8 41.12 ± 7% perf-profile.children.cycles-pp.cpuidle_enter_state 29.17 ± 20% +11.0 40.22 ± 7% perf-profile.children.cycles-pp.intel_idle 7.59 ± 8% -1.4 6.23 ± 5% perf-profile.self.cycles-pp.copy_user_enhanced_fast_string 5.39 ± 8% -0.9 4.53 ± 5% perf-profile.self.cycles-pp.syscall_return_via_sysret 2.89 ± 9% -0.5 2.35 ± 4% perf-profile.self.cycles-pp.__entry_text_start 2.62 ± 8% -0.5 2.13 ± 3% perf-profile.self.cycles-pp.pipe_write 2.45 ± 8% -0.5 1.98 ± 5% perf-profile.self.cycles-pp.pipe_read 2.33 ± 8% -0.5 1.87 ± 5% perf-profile.self.cycles-pp.__fget_light 2.33 ± 8% -0.5 1.87 ± 4% perf-profile.self.cycles-pp.mutex_lock 1.85 ± 8% -0.4 1.44 ± 4% perf-profile.self.cycles-pp.mutex_unlock 1.93 ± 8% -0.4 1.54 ± 5% perf-profile.self.cycles-pp.___might_sleep 2.12 ± 7% -0.4 1.75 ± 4% perf-profile.self.cycles-pp._raw_spin_lock_irq 1.27 ± 7% -0.3 0.93 ± 2% perf-profile.self.cycles-pp.syscall_exit_to_user_mode 1.50 ± 9% -0.3 1.19 ± 4% perf-profile.self.cycles-pp.__might_sleep 2.63 ± 7% -0.3 2.32 ± 8% perf-profile.self.cycles-pp.common_file_perm 1.69 ± 8% -0.3 1.40 ± 6% perf-profile.self.cycles-pp.entry_SYSCALL_64_after_hwframe 1.31 ± 11% -0.3 1.01 ± 4% perf-profile.self.cycles-pp.copy_page_from_iter 0.36 ± 19% -0.2 0.12 ± 38% perf-profile.self.cycles-pp.menu_select 1.26 ± 10% -0.2 1.02 ± 6% perf-profile.self.cycles-pp.copy_page_to_iter 1.71 ± 7% -0.2 1.48 ± 7% perf-profile.self.cycles-pp.syscall_enter_from_user_mode 1.09 ± 8% -0.2 0.88 ± 6% perf-profile.self.cycles-pp._raw_spin_unlock_irqrestore 1.48 ± 8% -0.2 1.27 ± 5% perf-profile.self.cycles-pp._raw_spin_lock_irqsave 1.07 ± 8% -0.2 0.86 ± 3% perf-profile.self.cycles-pp.new_sync_read 1.27 ± 8% -0.2 1.06 perf-profile.self.cycles-pp.vfs_read 1.14 ± 7% -0.2 0.94 ± 8% perf-profile.self.cycles-pp.anon_pipe_buf_release 0.35 ± 10% -0.2 0.15 ± 12% perf-profile.self.cycles-pp.cpuidle_enter_state 0.92 ± 9% -0.2 0.73 ± 5% perf-profile.self.cycles-pp.__sb_end_write 0.82 ± 8% -0.2 0.66 ± 5% perf-profile.self.cycles-pp.aa_file_perm 0.69 ± 6% -0.1 0.54 perf-profile.self.cycles-pp.exit_to_user_mode_prepare 0.93 ± 8% -0.1 0.78 ± 4% perf-profile.self.cycles-pp.entry_SYSCALL_64_safe_stack 0.75 ± 8% -0.1 0.61 ± 6% perf-profile.self.cycles-pp.security_file_permission 0.67 ± 7% -0.1 0.53 ± 4% perf-profile.self.cycles-pp.__sb_start_write 0.93 ± 5% -0.1 0.80 perf-profile.self.cycles-pp.new_sync_write 0.60 ± 11% -0.1 0.47 ± 4% perf-profile.self.cycles-pp.__wake_up_common_lock 0.62 ± 9% -0.1 0.50 ± 5% perf-profile.self.cycles-pp.current_time 0.54 ± 8% -0.1 0.43 ± 5% perf-profile.self.cycles-pp._cond_resched 0.42 ± 7% -0.1 0.31 ± 7% perf-profile.self.cycles-pp.do_syscall_64 0.45 ± 11% -0.1 0.34 ± 5% perf-profile.self.cycles-pp.write 0.45 ± 7% -0.1 0.35 ± 4% perf-profile.self.cycles-pp.ksys_write 0.45 ± 9% -0.1 0.35 ± 6% perf-profile.self.cycles-pp.read 0.43 ± 6% -0.1 0.34 ± 4% perf-profile.self.cycles-pp.ksys_read 0.61 ± 7% -0.1 0.53 ± 4% perf-profile.self.cycles-pp.__fdget_pos 0.49 ± 9% -0.1 0.41 ± 4% perf-profile.self.cycles-pp.file_update_time 0.43 ± 10% -0.1 0.35 ± 5% perf-profile.self.cycles-pp.__wake_up_common 0.43 ± 5% -0.1 0.36 ± 7% perf-profile.self.cycles-pp.atime_needs_update 0.31 ± 9% -0.1 0.25 ± 6% perf-profile.self.cycles-pp.__x86_indirect_thunk_rax 0.28 ± 8% -0.1 0.23 ± 2% perf-profile.self.cycles-pp.__might_fault 0.31 ± 5% -0.1 0.26 ± 4% perf-profile.self.cycles-pp.rw_verify_area 0.27 ± 6% -0.0 0.22 ± 3% perf-profile.self.cycles-pp.copyin 0.25 ± 5% -0.0 0.21 ± 5% perf-profile.self.cycles-pp.iov_iter_init 0.21 ± 6% -0.0 0.18 ± 3% perf-profile.self.cycles-pp.kill_fasync 2.75 ± 7% +1.4 4.18 ± 6% perf-profile.self.cycles-pp.fsnotify 29.17 ± 20% +11.0 40.22 ± 7% perf-profile.self.cycles-pp.intel_idle unixbench.score 42000 +-------------------------------------------------------------------+ | +.+ O : : +. | 40000 |-O O O O O O O O O O O O O O O O O O : O : O O O | 38000 |-+ : : | | : : | 36000 |-+ : : | 34000 |-+ : : | | : : | 32000 |-+ : : | 30000 |-+ :: | | :: | 28000 |-+ : | 26000 |-+ + | | | 24000 +-------------------------------------------------------------------+ [*] bisect-good sample [O] bisect-bad sample Disclaimer: Results have been estimated based on internal Intel analysis and are provided for informational purposes only. Any difference in system hardware or software design or configuration may affect actual performance. Thanks, Oliver Sang