Greeting, FYI, we noticed a 18.3% improvement of will-it-scale.per_process_ops due to commit: commit: e83a3e282ebeab6bcf3eec024b93fcab0d6e4dc5 ("readfile: implement readfile syscall") https://git.kernel.org/cgit/linux/kernel/git/gregkh/driver-core.git readfile in testcase: will-it-scale on test machine: 192 threads Intel(R) Xeon(R) Platinum 9242 CPU @ 2.30GHz with 192G memory with following parameters: nr_task: 50% mode: process test: signal1 cpufreq_governor: performance ucode: 0x5002f01 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 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/mode/nr_task/rootfs/tbox_group/test/testcase/ucode: gcc-9/performance/x86_64-rhel-8.3/process/50%/debian-10.4-x86_64-20200603.cgz/lkp-csl-2ap3/signal1/will-it-scale/0x5002f01 commit: v5.8-rc7 e83a3e282e ("readfile: implement readfile syscall") v5.8-rc7 e83a3e282ebeab6bcf3eec024b9 ---------------- --------------------------- %stddev %change %stddev \ | \ 58045 +18.3% 68672 will-it-scale.per_process_ops 5572453 +18.3% 6592569 will-it-scale.workload 1.04 +0.3 1.33 mpstat.cpu.all.usr% 1432 ± 12% -20.8% 1133 ± 5% slabinfo.dmaengine-unmap-16.active_objs 1432 ± 12% -20.8% 1133 ± 5% slabinfo.dmaengine-unmap-16.num_objs 70.90 ± 25% -54.1% 32.56 ± 21% sched_debug.cfs_rq:/.exec_clock.min 29948 ± 10% -24.9% 22490 ± 11% sched_debug.cfs_rq:/.min_vruntime.min 1174 ± 13% -16.9% 976.50 ± 10% sched_debug.cpu.nr_switches.min 10734 ± 11% -22.6% 8304 ± 10% numa-vmstat.node0.nr_slab_reclaimable 27316 ± 8% -17.3% 22580 ± 3% numa-vmstat.node0.nr_slab_unreclaimable 69281 ± 4% +8.0% 74797 ± 5% numa-vmstat.node1.nr_unevictable 69281 ± 4% +8.0% 74797 ± 5% numa-vmstat.node1.nr_zone_unevictable 42938 ± 11% -22.6% 33220 ± 10% numa-meminfo.node0.KReclaimable 939002 ± 5% -11.0% 836158 ± 6% numa-meminfo.node0.MemUsed 42938 ± 11% -22.6% 33220 ± 10% numa-meminfo.node0.SReclaimable 109266 ± 8% -17.3% 90326 ± 3% numa-meminfo.node0.SUnreclaim 152205 ± 5% -18.8% 123547 ± 4% numa-meminfo.node0.Slab 277125 ± 4% +8.0% 299188 ± 5% numa-meminfo.node1.Unevictable 989829 +1.2% 1002099 proc-vmstat.numa_hit 896406 +1.4% 908669 proc-vmstat.numa_local 3928 ± 77% +272.9% 14648 ± 44% proc-vmstat.numa_pages_migrated 14026 ± 77% +183.9% 39817 ± 36% proc-vmstat.numa_pte_updates 1065214 +2.2% 1088172 proc-vmstat.pgalloc_normal 1164976 +1.9% 1186815 proc-vmstat.pgfault 3928 ± 77% +272.9% 14648 ± 44% proc-vmstat.pgmigrate_success 25786 ± 62% -86.3% 3536 ± 9% softirqs.CPU102.SCHED 31809 ± 43% -60.6% 12534 ±121% softirqs.CPU112.SCHED 3727 ± 3% +371.9% 17592 ± 83% softirqs.CPU177.SCHED 4832 ± 48% +337.9% 21159 ± 82% softirqs.CPU18.SCHED 31756 ± 43% -61.1% 12360 ±120% softirqs.CPU27.SCHED 10268 ±128% +104.4% 20989 ± 82% softirqs.CPU50.SCHED 26586 ± 55% -53.5% 12360 ±120% softirqs.CPU93.SCHED 5.569e+09 +17.2% 6.53e+09 perf-stat.i.branch-instructions 0.91 -0.0 0.89 perf-stat.i.branch-miss-rate% 50387334 +15.0% 57968632 perf-stat.i.branch-misses 10.87 -14.9% 9.25 perf-stat.i.cpi 8.449e+09 +17.4% 9.917e+09 perf-stat.i.dTLB-loads 20126 +14.4% 23016 ± 2% perf-stat.i.dTLB-store-misses 5.366e+09 +17.6% 6.307e+09 perf-stat.i.dTLB-stores 92.97 +0.8 93.78 perf-stat.i.iTLB-load-miss-rate% 54730723 ± 6% +14.9% 62868063 perf-stat.i.iTLB-load-misses 4065855 +0.9% 4101208 perf-stat.i.iTLB-loads 2.767e+10 +17.3% 3.245e+10 perf-stat.i.instructions 0.09 +17.6% 0.11 perf-stat.i.ipc 0.16 +7.2% 0.18 ± 2% perf-stat.i.metric.K/sec 101.76 +17.3% 119.36 perf-stat.i.metric.M/sec 3764 +1.7% 3829 perf-stat.i.minor-faults 288173 ± 7% +9.9% 316612 ± 4% perf-stat.i.node-loads 8940354 +18.0% 10550802 perf-stat.i.node-store-misses 3764 +1.7% 3829 perf-stat.i.page-faults 0.90 -0.0 0.89 perf-stat.overall.branch-miss-rate% 10.89 -14.9% 9.26 perf-stat.overall.cpi 93.06 +0.8 93.88 perf-stat.overall.iTLB-load-miss-rate% 0.09 +17.5% 0.11 perf-stat.overall.ipc 5.55e+09 +17.2% 6.507e+09 perf-stat.ps.branch-instructions 50135115 +15.0% 57663630 perf-stat.ps.branch-misses 8.421e+09 +17.4% 9.883e+09 perf-stat.ps.dTLB-loads 19887 +14.2% 22708 ± 2% perf-stat.ps.dTLB-store-misses 5.348e+09 +17.5% 6.287e+09 perf-stat.ps.dTLB-stores 54503912 ± 6% +14.9% 62633375 perf-stat.ps.iTLB-load-misses 4046767 +0.9% 4082601 perf-stat.ps.iTLB-loads 2.758e+10 +17.3% 3.234e+10 perf-stat.ps.instructions 3723 +1.5% 3780 perf-stat.ps.minor-faults 8911288 +18.0% 10514645 perf-stat.ps.node-store-misses 3723 +1.5% 3780 perf-stat.ps.page-faults 8.331e+12 +17.4% 9.777e+12 perf-stat.total.instructions 55.39 ± 9% -20.2 35.22 ± 9% perf-profile.calltrace.cycles-pp.security_task_kill.do_send_specific.do_tkill.__x64_sys_tgkill.do_syscall_64 55.34 ± 9% -20.1 35.20 ± 9% perf-profile.calltrace.cycles-pp.apparmor_task_kill.security_task_kill.do_send_specific.do_tkill.__x64_sys_tgkill 32.20 ± 8% -16.4 15.76 ± 10% perf-profile.calltrace.cycles-pp.aa_get_task_label.apparmor_task_kill.security_task_kill.do_send_specific.do_tkill 60.76 ± 9% -13.9 46.88 ± 9% perf-profile.calltrace.cycles-pp.do_tkill.__x64_sys_tgkill.do_syscall_64.entry_SYSCALL_64_after_hwframe.raise 60.77 ± 9% -13.9 46.89 ± 9% perf-profile.calltrace.cycles-pp.__x64_sys_tgkill.do_syscall_64.entry_SYSCALL_64_after_hwframe.raise 60.50 ± 9% -13.8 46.70 ± 9% perf-profile.calltrace.cycles-pp.do_send_specific.do_tkill.__x64_sys_tgkill.do_syscall_64.entry_SYSCALL_64_after_hwframe 1.74 ± 22% +0.5 2.28 ± 8% perf-profile.calltrace.cycles-pp.handler 4.76 ± 11% +5.9 10.67 ± 10% perf-profile.calltrace.cycles-pp.dequeue_signal.get_signal.do_signal.__prepare_exit_to_usermode.do_syscall_64 4.70 ± 11% +5.9 10.61 ± 10% perf-profile.calltrace.cycles-pp.__dequeue_signal.dequeue_signal.get_signal.do_signal.__prepare_exit_to_usermode 4.62 ± 12% +5.9 10.55 ± 10% perf-profile.calltrace.cycles-pp.__sigqueue_free.__dequeue_signal.dequeue_signal.get_signal.do_signal 4.88 ± 11% +6.0 10.85 ± 10% perf-profile.calltrace.cycles-pp.get_signal.do_signal.__prepare_exit_to_usermode.do_syscall_64.entry_SYSCALL_64_after_hwframe 5.86 ± 11% +6.0 11.88 ± 10% perf-profile.calltrace.cycles-pp.__prepare_exit_to_usermode.do_syscall_64.entry_SYSCALL_64_after_hwframe.raise 5.78 ± 11% +6.0 11.81 ± 10% perf-profile.calltrace.cycles-pp.do_signal.__prepare_exit_to_usermode.do_syscall_64.entry_SYSCALL_64_after_hwframe.raise 4.79 ± 11% +6.3 11.07 ± 10% perf-profile.calltrace.cycles-pp.__sigqueue_alloc.__send_signal.do_send_sig_info.do_send_specific.do_tkill 4.85 ± 11% +6.3 11.14 ± 10% perf-profile.calltrace.cycles-pp.__send_signal.do_send_sig_info.do_send_specific.do_tkill.__x64_sys_tgkill 5.01 ± 11% +6.3 11.34 ± 10% perf-profile.calltrace.cycles-pp.do_send_sig_info.do_send_specific.do_tkill.__x64_sys_tgkill.do_syscall_64 55.39 ± 9% -20.2 35.22 ± 9% perf-profile.children.cycles-pp.security_task_kill 55.38 ± 9% -20.2 35.21 ± 9% perf-profile.children.cycles-pp.apparmor_task_kill 32.21 ± 8% -16.4 15.77 ± 10% perf-profile.children.cycles-pp.aa_get_task_label 60.77 ± 9% -13.9 46.89 ± 9% perf-profile.children.cycles-pp.__x64_sys_tgkill 60.76 ± 9% -13.9 46.89 ± 9% perf-profile.children.cycles-pp.do_tkill 60.50 ± 9% -13.8 46.70 ± 9% perf-profile.children.cycles-pp.do_send_specific 0.06 ± 6% +0.0 0.10 ± 10% perf-profile.children.cycles-pp.__lock_text_start 0.00 +0.1 0.06 ± 14% perf-profile.children.cycles-pp.clockevents_program_event 0.64 ± 11% +0.2 0.84 ± 7% perf-profile.children.cycles-pp.native_irq_return_iret 1.26 ± 10% +0.3 1.51 ± 8% perf-profile.children.cycles-pp.handler 4.76 ± 11% +5.9 10.67 ± 10% perf-profile.children.cycles-pp.dequeue_signal 4.70 ± 11% +5.9 10.62 ± 10% perf-profile.children.cycles-pp.__dequeue_signal 4.62 ± 12% +5.9 10.55 ± 10% perf-profile.children.cycles-pp.__sigqueue_free 4.89 ± 11% +6.0 10.86 ± 10% perf-profile.children.cycles-pp.get_signal 6.41 ± 11% +6.1 12.48 ± 10% perf-profile.children.cycles-pp.__prepare_exit_to_usermode 6.31 ± 11% +6.1 12.38 ± 10% perf-profile.children.cycles-pp.do_signal 4.79 ± 11% +6.3 11.07 ± 10% perf-profile.children.cycles-pp.__sigqueue_alloc 4.86 ± 11% +6.3 11.14 ± 10% perf-profile.children.cycles-pp.__send_signal 5.02 ± 11% +6.3 11.35 ± 10% perf-profile.children.cycles-pp.do_send_sig_info 32.07 ± 8% -16.4 15.68 ± 10% perf-profile.self.cycles-pp.aa_get_task_label 22.56 ± 9% -3.7 18.91 ± 9% perf-profile.self.cycles-pp.apparmor_task_kill 0.64 ± 11% +0.2 0.84 ± 8% perf-profile.self.cycles-pp.native_irq_return_iret 4.61 ± 12% +5.9 10.55 ± 10% perf-profile.self.cycles-pp.__sigqueue_free 4.75 ± 11% +6.3 11.04 ± 10% perf-profile.self.cycles-pp.__sigqueue_alloc 169158 +15.1% 194640 ± 8% interrupts.CAL:Function_call_interrupts 3163 ± 44% +58.4% 5011 ± 24% interrupts.CPU0.NMI:Non-maskable_interrupts 3163 ± 44% +58.4% 5011 ± 24% interrupts.CPU0.PMI:Performance_monitoring_interrupts 4045 ± 56% +78.2% 7209 ± 20% interrupts.CPU1.NMI:Non-maskable_interrupts 4045 ± 56% +78.2% 7209 ± 20% interrupts.CPU1.PMI:Performance_monitoring_interrupts 5731 ± 44% +25.6% 7200 ± 20% interrupts.CPU100.NMI:Non-maskable_interrupts 5731 ± 44% +25.6% 7200 ± 20% interrupts.CPU100.PMI:Performance_monitoring_interrupts 5159 ± 54% +68.0% 8666 interrupts.CPU102.NMI:Non-maskable_interrupts 5159 ± 54% +68.0% 8666 interrupts.CPU102.PMI:Performance_monitoring_interrupts 4587 ± 49% +56.7% 7188 ± 20% interrupts.CPU104.NMI:Non-maskable_interrupts 4587 ± 49% +56.7% 7188 ± 20% interrupts.CPU104.PMI:Performance_monitoring_interrupts 4589 ± 49% +72.7% 7926 ± 16% interrupts.CPU11.NMI:Non-maskable_interrupts 4589 ± 49% +72.7% 7926 ± 16% interrupts.CPU11.PMI:Performance_monitoring_interrupts 4038 ± 56% +69.4% 6843 ± 27% interrupts.CPU112.NMI:Non-maskable_interrupts 4038 ± 56% +69.4% 6843 ± 27% interrupts.CPU112.PMI:Performance_monitoring_interrupts 4626 ± 49% +48.0% 6845 ± 27% interrupts.CPU116.NMI:Non-maskable_interrupts 4626 ± 49% +48.0% 6845 ± 27% interrupts.CPU116.PMI:Performance_monitoring_interrupts 4892 ± 43% +62.0% 7927 ± 16% interrupts.CPU118.NMI:Non-maskable_interrupts 4892 ± 43% +62.0% 7927 ± 16% interrupts.CPU118.PMI:Performance_monitoring_interrupts 4587 ± 50% +57.1% 7205 ± 20% interrupts.CPU120.NMI:Non-maskable_interrupts 4587 ± 50% +57.1% 7205 ± 20% interrupts.CPU120.PMI:Performance_monitoring_interrupts 3446 ± 77% +88.6% 6498 ± 37% interrupts.CPU136.NMI:Non-maskable_interrupts 3446 ± 77% +88.6% 6498 ± 37% interrupts.CPU136.PMI:Performance_monitoring_interrupts 3489 ± 31% +106.8% 7216 ± 20% interrupts.CPU140.NMI:Non-maskable_interrupts 3489 ± 31% +106.8% 7216 ± 20% interrupts.CPU140.PMI:Performance_monitoring_interrupts 4615 ± 49% +56.3% 7212 ± 20% interrupts.CPU142.NMI:Non-maskable_interrupts 4615 ± 49% +56.3% 7212 ± 20% interrupts.CPU142.PMI:Performance_monitoring_interrupts 4020 ± 57% +79.3% 7206 ± 20% interrupts.CPU144.NMI:Non-maskable_interrupts 4020 ± 57% +79.3% 7206 ± 20% interrupts.CPU144.PMI:Performance_monitoring_interrupts 4598 ± 49% +56.3% 7188 ± 20% interrupts.CPU15.NMI:Non-maskable_interrupts 4598 ± 49% +56.3% 7188 ± 20% interrupts.CPU15.PMI:Performance_monitoring_interrupts 4589 ± 50% +57.1% 7211 ± 20% interrupts.CPU156.NMI:Non-maskable_interrupts 4589 ± 50% +57.1% 7211 ± 20% interrupts.CPU156.PMI:Performance_monitoring_interrupts 4590 ± 50% +57.1% 7210 ± 20% interrupts.CPU160.NMI:Non-maskable_interrupts 4590 ± 50% +57.1% 7210 ± 20% interrupts.CPU160.PMI:Performance_monitoring_interrupts 5180 ± 54% +53.6% 7956 ± 16% interrupts.CPU161.NMI:Non-maskable_interrupts 5180 ± 54% +53.6% 7956 ± 16% interrupts.CPU161.PMI:Performance_monitoring_interrupts 4595 ± 50% +56.9% 7210 ± 20% interrupts.CPU162.NMI:Non-maskable_interrupts 4595 ± 50% +56.9% 7210 ± 20% interrupts.CPU162.PMI:Performance_monitoring_interrupts 4587 ± 50% +57.2% 7209 ± 20% interrupts.CPU164.NMI:Non-maskable_interrupts 4587 ± 50% +57.2% 7209 ± 20% interrupts.CPU164.PMI:Performance_monitoring_interrupts 4510 ± 50% +59.8% 7209 ± 20% interrupts.CPU166.NMI:Non-maskable_interrupts 4510 ± 50% +59.8% 7209 ± 20% interrupts.CPU166.PMI:Performance_monitoring_interrupts 4.40 ± 23% +809.1% 40.00 ±123% interrupts.CPU167.RES:Rescheduling_interrupts 4344 ± 59% +65.5% 7188 ± 20% interrupts.CPU17.NMI:Non-maskable_interrupts 4344 ± 59% +65.5% 7188 ± 20% interrupts.CPU17.PMI:Performance_monitoring_interrupts 4609 ± 49% +88.3% 8682 interrupts.CPU170.NMI:Non-maskable_interrupts 4609 ± 49% +88.3% 8682 interrupts.CPU170.PMI:Performance_monitoring_interrupts 5755 ± 44% +25.1% 7199 ± 20% interrupts.CPU172.NMI:Non-maskable_interrupts 5755 ± 44% +25.1% 7199 ± 20% interrupts.CPU172.PMI:Performance_monitoring_interrupts 4307 ± 59% +67.3% 7205 ± 20% interrupts.CPU182.NMI:Non-maskable_interrupts 4307 ± 59% +67.3% 7205 ± 20% interrupts.CPU182.PMI:Performance_monitoring_interrupts 5.80 ±143% +576.7% 39.25 ± 98% interrupts.CPU182.RES:Rescheduling_interrupts 3439 ± 33% +109.5% 7204 ± 20% interrupts.CPU186.NMI:Non-maskable_interrupts 3439 ± 33% +109.5% 7204 ± 20% interrupts.CPU186.PMI:Performance_monitoring_interrupts 4590 ± 50% +56.9% 7199 ± 20% interrupts.CPU188.NMI:Non-maskable_interrupts 4590 ± 50% +56.9% 7199 ± 20% interrupts.CPU188.PMI:Performance_monitoring_interrupts 4597 ± 49% +72.5% 7928 ± 16% interrupts.CPU19.NMI:Non-maskable_interrupts 4597 ± 49% +72.5% 7928 ± 16% interrupts.CPU19.PMI:Performance_monitoring_interrupts 4305 ± 59% +84.0% 7924 ± 16% interrupts.CPU21.NMI:Non-maskable_interrupts 4305 ± 59% +84.0% 7924 ± 16% interrupts.CPU21.PMI:Performance_monitoring_interrupts 4320 ± 59% +84.5% 7969 ± 16% interrupts.CPU27.NMI:Non-maskable_interrupts 4320 ± 59% +84.5% 7969 ± 16% interrupts.CPU27.PMI:Performance_monitoring_interrupts 5192 ± 54% +39.0% 7214 ± 20% interrupts.CPU33.NMI:Non-maskable_interrupts 5192 ± 54% +39.0% 7214 ± 20% interrupts.CPU33.PMI:Performance_monitoring_interrupts 4360 ± 58% +65.4% 7211 ± 20% interrupts.CPU35.NMI:Non-maskable_interrupts 4360 ± 58% +65.4% 7211 ± 20% interrupts.CPU35.PMI:Performance_monitoring_interrupts 224.60 ±112% -86.6% 30.00 ±156% interrupts.CPU38.RES:Rescheduling_interrupts 4311 ± 59% +67.3% 7212 ± 20% interrupts.CPU39.NMI:Non-maskable_interrupts 4311 ± 59% +67.3% 7212 ± 20% interrupts.CPU39.PMI:Performance_monitoring_interrupts 3726 ± 30% +93.5% 7210 ± 20% interrupts.CPU41.NMI:Non-maskable_interrupts 3726 ± 30% +93.5% 7210 ± 20% interrupts.CPU41.PMI:Performance_monitoring_interrupts 4026 ± 57% +79.1% 7211 ± 20% interrupts.CPU43.NMI:Non-maskable_interrupts 4026 ± 57% +79.1% 7211 ± 20% interrupts.CPU43.PMI:Performance_monitoring_interrupts 4361 ± 50% +65.4% 7215 ± 20% interrupts.CPU45.NMI:Non-maskable_interrupts 4361 ± 50% +65.4% 7215 ± 20% interrupts.CPU45.PMI:Performance_monitoring_interrupts 17.80 ±143% +329.8% 76.50 ± 54% interrupts.CPU48.RES:Rescheduling_interrupts 29.20 ±108% +145.7% 71.75 ± 54% interrupts.CPU54.RES:Rescheduling_interrupts 4.80 ± 24% +1301.0% 67.25 ± 43% interrupts.CPU60.RES:Rescheduling_interrupts 4593 ± 50% +73.2% 7957 ± 16% interrupts.CPU67.NMI:Non-maskable_interrupts 4593 ± 50% +73.2% 7957 ± 16% interrupts.CPU67.PMI:Performance_monitoring_interrupts 4592 ± 50% +57.0% 7210 ± 20% interrupts.CPU69.NMI:Non-maskable_interrupts 4592 ± 50% +57.0% 7210 ± 20% interrupts.CPU69.PMI:Performance_monitoring_interrupts 4017 ± 57% +78.9% 7186 ± 20% interrupts.CPU7.NMI:Non-maskable_interrupts 4017 ± 57% +78.9% 7186 ± 20% interrupts.CPU7.PMI:Performance_monitoring_interrupts 60.20 ± 89% -87.1% 7.75 ± 27% interrupts.CPU7.RES:Rescheduling_interrupts 12.60 ±117% +511.1% 77.00 ± 33% interrupts.CPU70.RES:Rescheduling_interrupts 13.40 ± 83% +394.4% 66.25 ± 45% interrupts.CPU72.RES:Rescheduling_interrupts 4592 ± 50% +72.9% 7940 ± 16% interrupts.CPU79.NMI:Non-maskable_interrupts 4592 ± 50% +72.9% 7940 ± 16% interrupts.CPU79.PMI:Performance_monitoring_interrupts 4588 ± 49% +56.6% 7187 ± 20% interrupts.CPU9.NMI:Non-maskable_interrupts 4588 ± 49% +56.6% 7187 ± 20% interrupts.CPU9.PMI:Performance_monitoring_interrupts 4609 ± 49% +57.6% 7263 ± 19% interrupts.CPU98.NMI:Non-maskable_interrupts 4609 ± 49% +57.6% 7263 ± 19% interrupts.CPU98.PMI:Performance_monitoring_interrupts will-it-scale.per_process_ops 74000 +-------------------------------------------------------------------+ | O O O O O O | 72000 |-+ O O O | 70000 |-+ | | O O O O O O O O O O O O O O O O O | 68000 |-+ | 66000 |-+ | | | 64000 |-+ | 62000 |-+ | | | 60000 |-+ | 58000 |.+.. .+.+. .+. .+.+.. .+. .+.+ | | + +..+.+.+.+. + + +. | 56000 +-------------------------------------------------------------------+ will-it-scale.workload 7.2e+06 +-----------------------------------------------------------------+ | O O O O O | 7e+06 |-+ O O O O O O O | 6.8e+06 |-+ | | | 6.6e+06 |-+ O O O O O O O O O O O O O O O O O | 6.4e+06 |-+ | | | 6.2e+06 |-+ | 6e+06 |-+ | | | 5.8e+06 |-+ | 5.6e+06 |.+. .+.+. .+. .+..+. .+. .+..+ | | +. +.+.+..+.+ + + + | 5.4e+06 +-----------------------------------------------------------------+ [*] 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, Rong Chen