From mboxrd@z Thu Jan 1 00:00:00 1970 Content-Type: multipart/mixed; boundary="===============3283624816074374174==" MIME-Version: 1.0 From: Rong Chen To: lkp@lists.01.org Subject: Re: [sched/numa] f6183ef98b: phoronix-test-suite.aom-av1.0.frames_per_second -25.0% regression Date: Thu, 27 Feb 2020 10:57:38 +0800 Message-ID: <173e9ed6-c93c-e6da-3cdb-782da0f9f3f4@intel.com> In-Reply-To: List-Id: --===============3283624816074374174== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable On 2/26/20 4:03 PM, Vincent Guittot wrote: > Hi Rong, > > On Wed, 26 Feb 2020 at 02:33, kernel test robot = wrote: >> Greeting, >> >> FYI, we noticed a -25.0% regression of phoronix-test-suite.aom-av1.0.fra= mes_per_second due to commit: >> >> >> commit: f6183ef98bba39a5de563d4ab4bc889ff2b600d7 ("sched/numa: replace r= unnable_load_avg by load_avg") >> https://git.kernel.org/cgit/linux/kernel/git/mel/linux.git sched-lbnuma-= rewrite-v2r6 > The following patches in the branch recover the perf regression. Do > you see regression on the whole branch ? Hi Guittot, I confirmed that the regression can be found on the head commit = (b92430cf87eb "sched/numa: Stop an exhastive search if a reasonable swap = candidate or idle CPU is found"). Best Regards, Rong Chen > >> in testcase: phoronix-test-suite >> on test machine: 16 threads Intel(R) Xeon(R) CPU X5570 @ 2.93GHz with 48= G memory >> with following parameters: >> >> test: aom-av1-1.2.0 >> cpufreq_governor: performance >> ucode: 0x1d >> >> test-description: The Phoronix Test Suite is the most comprehensive test= ing and benchmarking platform available that provides an extensible framewo= rk for which new tests can be easily added. >> test-url: http://www.phoronix-test-suite.com/ >> >> >> >> 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 >> >> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D >> compiler/cpufreq_governor/kconfig/rootfs/tbox_group/test/testcase/ucode: >> gcc-7/performance/x86_64-rhel-7.6/debian-x86_64-phoronix/lkp-nhm-2ep1= /aom-av1-1.2.0/phoronix-test-suite/0x1d >> >> commit: >> 91d7637912 ("sched/fair: reorder enqueue/dequeue_task_fair path") >> f6183ef98b ("sched/numa: replace runnable_load_avg by load_avg") >> >> 91d76379120e4640 f6183ef98bba39a5de563d4ab4b >> ---------------- --------------------------- >> %stddev %change %stddev >> \ | \ >> 0.04 -25.0% 0.03 phoronix-test-suite.aom-a= v1.0.frames_per_second >> 1648 +24.6% 2055 phoronix-test-suite.time.= elapsed_time >> 1648 +24.6% 2055 phoronix-test-suite.time.= elapsed_time.max >> 1935 -19.6% 1556 meminfo.max_used_kB >> 1.518e+10 =C2=B1 29% +39.9% 2.124e+10 cpuidle.C1E.time >> 33131701 =C2=B1 34% +35.9% 45011103 =C2=B1 2% cpuidle.C1E.usa= ge >> 672.00 -100.0% 0.00 slabinfo.dmaengine-unmap-= 16.active_objs >> 672.00 -100.0% 0.00 slabinfo.dmaengine-unmap-= 16.num_objs >> 67.50 -2.2% 66.00 vmstat.cpu.id >> 31.00 +6.5% 33.00 vmstat.cpu.us >> 467.75 +5.0% 491.00 proc-vmstat.nr_mlock >> 484.75 +4.8% 508.00 proc-vmstat.nr_unevictable >> 484.75 +4.8% 508.00 proc-vmstat.nr_zone_unevi= ctable >> 3492646 =C2=B1 7% +67.4% 5846384 =C2=B1 3% proc-vmstat.num= a_hint_faults >> 2478506 =C2=B1 6% +127.8% 5646434 =C2=B1 4% proc-vmstat.num= a_hint_faults_local >> 3650509 +13.5% 4141674 proc-vmstat.numa_hit >> 76734 +98.1% 151980 =C2=B1 2% proc-vmstat.numa_hug= e_pte_updates >> 3647139 +13.5% 4138249 proc-vmstat.numa_local >> 3408776 =C2=B1 7% -80.4% 667824 =C2=B1 8% proc-vmstat.num= a_pages_migrated >> 42828088 +95.3% 83628988 =C2=B1 2% proc-vmstat.numa_pte= _updates >> 7766914 =C2=B1 3% -25.6% 5776215 proc-vmstat.pgalloc_= normal >> 7771023 =C2=B1 3% +43.5% 11147904 =C2=B1 2% proc-vmstat.pgf= ault >> 7716195 =C2=B1 3% -25.8% 5726088 proc-vmstat.pgfree >> 1152 =C2=B1 19% -88.9% 128.00 =C2=B1173% proc-vmstat.pgm= igrate_fail >> 3408776 =C2=B1 7% -80.4% 667824 =C2=B1 8% proc-vmstat.pgm= igrate_success >> 0.13 =C2=B1173% +0.5 0.63 =C2=B1 7% perf-profile.ca= lltrace.cycles-pp.smp_apic_timer_interrupt.apic_timer_interrupt.cpuidle_ent= er_state.cpuidle_enter.do_idle >> 0.14 =C2=B1173% +0.5 0.67 =C2=B1 8% perf-profile.ca= lltrace.cycles-pp.apic_timer_interrupt.cpuidle_enter_state.cpuidle_enter.do= _idle.cpu_startup_entry >> 4.40 =C2=B1 47% +3.5 7.87 =C2=B1 16% perf-profile.ca= lltrace.cycles-pp.cpuidle_enter_state.cpuidle_enter.do_idle.cpu_startup_ent= ry.start_kernel >> 4.40 =C2=B1 47% +3.5 7.87 =C2=B1 16% perf-profile.ca= lltrace.cycles-pp.cpuidle_enter.do_idle.cpu_startup_entry.start_kernel.seco= ndary_startup_64 >> 4.41 =C2=B1 47% +3.5 7.94 =C2=B1 16% perf-profile.ca= lltrace.cycles-pp.cpu_startup_entry.start_kernel.secondary_startup_64 >> 4.41 =C2=B1 47% +3.5 7.94 =C2=B1 16% perf-profile.ca= lltrace.cycles-pp.do_idle.cpu_startup_entry.start_kernel.secondary_startup_= 64 >> 4.41 =C2=B1 47% +3.5 7.94 =C2=B1 16% perf-profile.ca= lltrace.cycles-pp.start_kernel.secondary_startup_64 >> 0.05 =C2=B1 67% +0.1 0.12 =C2=B1 17% perf-profile.ch= ildren.cycles-pp.ktime_get >> 0.01 =C2=B1173% +0.1 0.10 =C2=B1 27% perf-profile.ch= ildren.cycles-pp._raw_spin_lock_irqsave >> 0.00 +0.1 0.10 =C2=B1 36% perf-profile.childre= n.cycles-pp.rcu_core >> 0.32 =C2=B1 86% +0.5 0.85 =C2=B1 54% perf-profile.ch= ildren.cycles-pp.__libc_start_main >> 0.60 =C2=B1 46% +0.6 1.19 =C2=B1 41% perf-profile.ch= ildren.cycles-pp.do_syscall_64 >> 0.60 =C2=B1 46% +0.6 1.19 =C2=B1 41% perf-profile.ch= ildren.cycles-pp.entry_SYSCALL_64_after_hwframe >> 0.34 =C2=B1 82% +0.6 0.94 =C2=B1 51% perf-profile.ch= ildren.cycles-pp.vfs_read >> 0.34 =C2=B1 82% +0.6 0.94 =C2=B1 51% perf-profile.ch= ildren.cycles-pp.ksys_read >> 0.29 =C2=B1 96% +0.6 0.90 =C2=B1 53% perf-profile.ch= ildren.cycles-pp.perf_event_read >> 0.29 =C2=B1 96% +0.6 0.90 =C2=B1 53% perf-profile.ch= ildren.cycles-pp.smp_call_function_single >> 0.30 =C2=B1 96% +0.6 0.91 =C2=B1 53% perf-profile.ch= ildren.cycles-pp.__libc_read >> 0.29 =C2=B1 96% +0.6 0.91 =C2=B1 52% perf-profile.ch= ildren.cycles-pp.perf_read >> 4.41 =C2=B1 47% +3.5 7.94 =C2=B1 16% perf-profile.ch= ildren.cycles-pp.start_kernel >> 0.01 =C2=B1173% +0.1 0.07 =C2=B1 19% perf-profile.se= lf.cycles-pp.cpuidle_enter_state >> 0.29 =C2=B1 98% +0.6 0.90 =C2=B1 53% perf-profile.se= lf.cycles-pp.smp_call_function_single >> 161465 =C2=B1 7% +66.5% 268796 =C2=B1 16% softirqs.CPU0.S= CHED >> 154149 =C2=B1 4% +57.3% 242412 =C2=B1 5% softirqs.CPU1.S= CHED >> 217783 =C2=B1 10% +17.4% 255655 =C2=B1 3% softirqs.CPU10.= RCU >> 685538 =C2=B1 6% +16.3% 797133 =C2=B1 3% softirqs.CPU10.= TIMER >> 647482 +29.6% 839256 =C2=B1 14% softirqs.CPU11.TIMER >> 215207 =C2=B1 9% +19.8% 257893 =C2=B1 5% softirqs.CPU12.= RCU >> 676241 =C2=B1 4% +20.5% 814890 softirqs.CPU12.TIMER >> 218513 =C2=B1 8% +17.8% 257505 =C2=B1 4% softirqs.CPU14.= RCU >> 665815 =C2=B1 5% +20.2% 800244 =C2=B1 3% softirqs.CPU14.= TIMER >> 217764 =C2=B1 10% +16.0% 252579 =C2=B1 2% softirqs.CPU15.= RCU >> 671996 =C2=B1 2% +18.0% 792743 =C2=B1 10% softirqs.CPU15.= TIMER >> 216810 =C2=B1 10% +16.2% 251864 =C2=B1 3% softirqs.CPU2.R= CU >> 168050 =C2=B1 8% +53.2% 257377 =C2=B1 4% softirqs.CPU2.S= CHED >> 687608 =C2=B1 7% +13.9% 783359 =C2=B1 5% softirqs.CPU2.T= IMER >> 214659 =C2=B1 10% +17.4% 252030 =C2=B1 2% softirqs.CPU3.R= CU >> 201381 =C2=B1 8% +26.8% 255401 =C2=B1 10% softirqs.CPU3.S= CHED >> 218567 =C2=B1 9% +15.1% 251588 =C2=B1 4% softirqs.CPU4.R= CU >> 162228 =C2=B1 14% +39.7% 226690 =C2=B1 21% softirqs.CPU4.S= CHED >> 686728 =C2=B1 7% +15.4% 792699 =C2=B1 4% softirqs.CPU4.T= IMER >> 598763 =C2=B1 4% +33.1% 796983 =C2=B1 15% softirqs.CPU5.T= IMER >> 219495 =C2=B1 10% +15.8% 254260 =C2=B1 4% softirqs.CPU6.R= CU >> 685621 =C2=B1 6% +15.4% 790913 =C2=B1 4% softirqs.CPU6.T= IMER >> 213675 =C2=B1 10% +18.1% 252314 =C2=B1 2% softirqs.CPU7.R= CU >> 607580 =C2=B1 4% +27.7% 776035 =C2=B1 7% softirqs.CPU7.T= IMER >> 217328 =C2=B1 10% +17.9% 256261 =C2=B1 4% softirqs.CPU8.R= CU >> 672277 =C2=B1 8% +18.4% 795798 =C2=B1 3% softirqs.CPU8.T= IMER >> 662696 =C2=B1 15% +22.2% 809936 =C2=B1 6% softirqs.CPU9.T= IMER >> 3471815 =C2=B1 10% +16.7% 4051741 =C2=B1 2% softirqs.RCU >> 2788089 +21.0% 3372673 =C2=B1 2% softirqs.SCHED >> 10668573 =C2=B1 2% +17.5% 12539415 =C2=B1 5% softirqs.TIMER >> 250541 +31.8% 330137 =C2=B1 3% sched_debug.cfs_rq:/= .exec_clock.avg >> 446877 =C2=B1 7% +68.2% 751493 =C2=B1 18% sched_debug.cfs= _rq:/.exec_clock.max >> 112331 =C2=B1 17% +122.7% 250151 =C2=B1 35% sched_debug.cfs= _rq:/.exec_clock.stddev >> 1517645 +35.0% 2049553 =C2=B1 3% sched_debug.cfs_rq:/= .min_vruntime.avg >> 2633991 =C2=B1 6% +65.2% 4352247 =C2=B1 18% sched_debug.cfs= _rq:/.min_vruntime.max >> 684251 =C2=B1 18% +118.2% 1492888 =C2=B1 37% sched_debug.cfs= _rq:/.min_vruntime.stddev >> 0.46 =C2=B1 2% +8.1% 0.50 =C2=B1 4% sched_debug.cfs= _rq:/.nr_running.avg >> -292907 -321.0% 647293 =C2=B1 49% sched_debug.cfs_rq:/= .spread0.avg >> 823435 =C2=B1 35% +258.3% 2949987 =C2=B1 32% sched_debug.cfs= _rq:/.spread0.max >> 684251 =C2=B1 18% +118.2% 1492888 =C2=B1 37% sched_debug.cfs= _rq:/.spread0.stddev >> 389.08 +10.3% 429.08 =C2=B1 3% sched_debug.cfs_rq:/= .util_avg.avg >> 282.99 =C2=B1 2% +13.1% 320.08 =C2=B1 6% sched_debug.cfs= _rq:/.util_est_enqueued.avg >> 383.98 =C2=B1 2% +8.6% 416.84 =C2=B1 3% sched_debug.cfs= _rq:/.util_est_enqueued.stddev >> 842328 +25.8% 1059503 sched_debug.cpu.clock.avg >> 842329 +25.8% 1059503 sched_debug.cpu.clock.max >> 842327 +25.8% 1059502 sched_debug.cpu.clock.min >> 1.02 =C2=B1 69% -49.5% 0.51 sched_debug.cpu.cloc= k.stddev >> 842328 +25.8% 1059503 sched_debug.cpu.clock_tas= k.avg >> 842329 +25.8% 1059503 sched_debug.cpu.clock_tas= k.max >> 842327 +25.8% 1059502 sched_debug.cpu.clock_tas= k.min >> 1.02 =C2=B1 69% -49.5% 0.51 sched_debug.cpu.cloc= k_task.stddev >> 6159 =C2=B1 3% +29.3% 7966 =C2=B1 6% sched_debug.cpu= .curr->pid.avg >> 21434 +25.0% 26795 sched_debug.cpu.curr->pid= .max >> 7691 =C2=B1 2% +26.8% 9755 sched_debug.cpu.curr= ->pid.stddev >> 57507 =C2=B1 2% +27.2% 73171 sched_debug.cpu.nr_s= witches.avg >> 98634 =C2=B1 8% +51.9% 149817 =C2=B1 22% sched_debug.cpu= .nr_switches.max >> 21105 =C2=B1 18% +85.3% 39113 =C2=B1 41% sched_debug.cpu= .nr_switches.stddev >> 52692 =C2=B1 2% +29.9% 68458 sched_debug.cpu.sche= d_count.avg >> 94966 =C2=B1 8% +50.3% 142769 =C2=B1 25% sched_debug.cpu= .sched_count.max >> 25787 =C2=B1 2% +27.6% 32917 sched_debug.cpu.sche= d_goidle.avg >> 47100 =C2=B1 8% +48.8% 70064 =C2=B1 26% sched_debug.cpu= .sched_goidle.max >> 25109 =C2=B1 2% +28.2% 32200 sched_debug.cpu.ttwu= _count.avg >> 47736 =C2=B1 4% +55.9% 74430 =C2=B1 9% sched_debug.cpu= .ttwu_count.max >> 10778 =C2=B1 13% +80.8% 19490 =C2=B1 29% sched_debug.cpu= .ttwu_count.stddev >> 11888 =C2=B1 4% +30.1% 15462 sched_debug.cpu.ttwu= _local.avg >> 842327 +25.8% 1059502 sched_debug.cpu_clk >> 839741 +25.9% 1056916 sched_debug.ktime >> 843284 +25.8% 1060461 sched_debug.sched_clk >> 1307 +27.3% 1663 =C2=B1 10% interrupts.35:PCI-MS= I.524289-edge.eth0-rx-0 >> 884.50 =C2=B1 2% +45.9% 1290 =C2=B1 26% interrupts.42:P= CI-MSI.524296-edge.eth0-tx-3 >> 1810554 =C2=B1 9% -24.9% 1359040 interrupts.CAL:Funct= ion_call_interrupts >> 128994 =C2=B1 13% -54.5% 58688 =C2=B1 11% interrupts.CPU0= .CAL:Function_call_interrupts >> 3136833 =C2=B1 9% +30.6% 4097106 interrupts.CPU0.LOC:= Local_timer_interrupts >> 147624 =C2=B1 12% -63.5% 53903 =C2=B1 7% interrupts.CPU0= .TLB:TLB_shootdowns >> 131316 =C2=B1 12% -58.5% 54545 =C2=B1 3% interrupts.CPU1= .CAL:Function_call_interrupts >> 3196967 =C2=B1 6% +28.8% 4117877 interrupts.CPU1.LOC:= Local_timer_interrupts >> 1316 =C2=B1 44% +134.5% 3086 =C2=B1 54% interrupts.CPU1= .RES:Rescheduling_interrupts >> 150414 =C2=B1 17% -65.7% 51585 =C2=B1 14% interrupts.CPU1= .TLB:TLB_shootdowns >> 3214233 =C2=B1 5% +28.6% 4132526 interrupts.CPU10.LOC= :Local_timer_interrupts >> 9879 =C2=B1 80% -85.4% 1445 =C2=B1 45% interrupts.CPU1= 0.RES:Rescheduling_interrupts >> 3177700 =C2=B1 6% +29.8% 4126192 interrupts.CPU11.LOC= :Local_timer_interrupts >> 559.50 =C2=B1 46% +253.3% 1976 =C2=B1 51% interrupts.CPU1= 1.RES:Rescheduling_interrupts >> 1307 +27.3% 1663 =C2=B1 10% interrupts.CPU12.35:= PCI-MSI.524289-edge.eth0-rx-0 >> 3234073 =C2=B1 4% +27.7% 4130544 interrupts.CPU12.LOC= :Local_timer_interrupts >> 9543 =C2=B1 47% -86.8% 1263 =C2=B1 55% interrupts.CPU1= 2.RES:Rescheduling_interrupts >> 3178037 =C2=B1 7% +29.8% 4124112 interrupts.CPU13.LOC= :Local_timer_interrupts >> 3234950 =C2=B1 4% +27.7% 4130372 interrupts.CPU14.LOC= :Local_timer_interrupts >> 7845 =C2=B1 61% -86.3% 1074 =C2=B1 77% interrupts.CPU1= 4.RES:Rescheduling_interrupts >> 3177518 =C2=B1 7% +29.8% 4123846 interrupts.CPU15.LOC= :Local_timer_interrupts >> 115381 =C2=B1 14% -46.8% 61375 =C2=B1 38% interrupts.CPU2= .CAL:Function_call_interrupts >> 3211923 =C2=B1 4% +28.3% 4122113 interrupts.CPU2.LOC:= Local_timer_interrupts >> 128739 =C2=B1 17% -57.7% 54470 =C2=B1 48% interrupts.CPU2= .TLB:TLB_shootdowns >> 884.50 =C2=B1 2% +45.9% 1290 =C2=B1 26% interrupts.CPU3= .42:PCI-MSI.524296-edge.eth0-tx-3 >> 3189990 =C2=B1 6% +29.1% 4119429 interrupts.CPU3.LOC:= Local_timer_interrupts >> 1680 =C2=B1 89% +222.4% 5418 =C2=B1 50% interrupts.CPU3= .RES:Rescheduling_interrupts >> 3207702 =C2=B1 5% +28.7% 4126732 interrupts.CPU4.LOC:= Local_timer_interrupts >> 132646 =C2=B1 21% -46.4% 71034 =C2=B1 60% interrupts.CPU4= .TLB:TLB_shootdowns >> 3185259 =C2=B1 7% +29.4% 4121092 interrupts.CPU5.LOC:= Local_timer_interrupts >> 1237 =C2=B1 56% +203.8% 3757 =C2=B1 37% interrupts.CPU5= .RES:Rescheduling_interrupts >> 3200005 =C2=B1 5% +29.0% 4128870 interrupts.CPU6.LOC:= Local_timer_interrupts >> 3187267 =C2=B1 7% +29.4% 4125551 interrupts.CPU7.LOC:= Local_timer_interrupts >> 802.00 =C2=B1 53% +280.5% 3051 =C2=B1 26% interrupts.CPU7= .RES:Rescheduling_interrupts >> 3266308 =C2=B1 2% +26.5% 4132734 interrupts.CPU8.LOC:= Local_timer_interrupts >> 23706 =C2=B1 90% -82.0% 4262 =C2=B1 66% interrupts.CPU8= .RES:Rescheduling_interrupts >> 3163000 =C2=B1 8% +30.5% 4128371 interrupts.CPU9.LOC:= Local_timer_interrupts >> 51161770 =C2=B1 6% +29.0% 65987472 interrupts.LOC:Local= _timer_interrupts >> 1944653 =C2=B1 10% -31.9% 1325214 interrupts.TLB:TLB_s= hootdowns >> 4.79 +19.6% 5.73 perf-stat.i.MPKI >> 1.596e+09 -19.9% 1.278e+09 perf-stat.i.branch-instru= ctions >> 1.66 +0.1 1.76 perf-stat.i.branch-miss-r= ate% >> 26901389 -14.4% 23036625 perf-stat.i.branch-misses >> 1.39e+08 -2.8% 1.351e+08 perf-stat.i.cache-referen= ces >> 0.61 =C2=B1 6% +17.9% 0.72 perf-stat.i.cpi >> 16.50 =C2=B1 5% +145.5% 40.52 =C2=B1 2% perf-stat.i.cpu= -migrations >> 5919 -3.6% 5706 =C2=B1 2% perf-stat.i.cycles-b= etween-cache-misses >> 0.05 =C2=B1 2% +0.0 0.07 perf-stat.i.dTLB-loa= d-miss-rate% >> 3285384 =C2=B1 3% +15.2% 3784496 perf-stat.i.dTLB-loa= d-misses >> 7.201e+09 -19.5% 5.793e+09 perf-stat.i.dTLB-loads >> 0.20 =C2=B1 2% +0.0 0.22 =C2=B1 4% perf-stat.i.dTL= B-store-miss-rate% >> 3641778 =C2=B1 2% -7.6% 3366197 =C2=B1 5% perf-stat.i.dTL= B-store-misses >> 1.925e+09 -19.5% 1.549e+09 perf-stat.i.dTLB-stores >> 0.01 =C2=B1 5% +0.0 0.02 =C2=B1 2% perf-stat.i.iTL= B-load-miss-rate% >> 2872660 =C2=B1 2% +25.9% 3617061 =C2=B1 2% perf-stat.i.iTL= B-load-misses >> 2.816e+10 -19.7% 2.261e+10 perf-stat.i.iTLB-loads >> 2.819e+10 -19.8% 2.261e+10 perf-stat.i.instructions >> 10654 =C2=B1 3% -27.1% 7764 =C2=B1 2% perf-stat.i.ins= tructions-per-iTLB-miss >> 1.70 =C2=B1 4% -14.9% 1.45 perf-stat.i.ipc >> 0.08 =C2=B1 2% -20.0% 0.07 =C2=B1 2% perf-stat.i.maj= or-faults >> 4638 =C2=B1 3% +15.4% 5354 =C2=B1 2% perf-stat.i.min= or-faults >> 4638 =C2=B1 3% +15.4% 5354 =C2=B1 2% perf-stat.i.pag= e-faults >> 4.93 +21.2% 5.97 perf-stat.overall.MPKI >> 1.69 +0.1 1.80 perf-stat.overall.branch-= miss-rate% >> 2.48 =C2=B1 4% +0.2 2.66 =C2=B1 2% perf-stat.overa= ll.cache-miss-rate% >> 0.58 =C2=B1 2% +25.2% 0.73 perf-stat.overall.cpi >> 0.05 =C2=B1 3% +0.0 0.07 perf-stat.overall.dT= LB-load-miss-rate% >> 0.19 =C2=B1 2% +0.0 0.22 =C2=B1 5% perf-stat.overa= ll.dTLB-store-miss-rate% >> 0.01 +0.0 0.02 =C2=B1 2% perf-stat.overall.iT= LB-load-miss-rate% >> 9815 =C2=B1 2% -36.3% 6253 =C2=B1 2% perf-stat.overa= ll.instructions-per-iTLB-miss >> 1.72 =C2=B1 2% -20.2% 1.37 perf-stat.overall.ipc >> 1.595e+09 -19.9% 1.278e+09 perf-stat.ps.branch-instr= uctions >> 26891393 -14.3% 23033727 perf-stat.ps.branch-misses >> 1.389e+08 -2.8% 1.35e+08 perf-stat.ps.cache-refere= nces >> 16.50 =C2=B1 5% +145.4% 40.50 =C2=B1 2% perf-stat.ps.cp= u-migrations >> 3283187 =C2=B1 3% +15.2% 3782278 perf-stat.ps.dTLB-lo= ad-misses >> 7.196e+09 -19.5% 5.79e+09 perf-stat.ps.dTLB-loads >> 3639422 =C2=B1 2% -7.6% 3364334 =C2=B1 5% perf-stat.ps.dT= LB-store-misses >> 1.924e+09 -19.5% 1.548e+09 perf-stat.ps.dTLB-stores >> 2870820 =C2=B1 2% +25.9% 3614797 =C2=B1 2% perf-stat.ps.iT= LB-load-misses >> 2.814e+10 -19.7% 2.26e+10 perf-stat.ps.iTLB-loads >> 2.817e+10 -19.8% 2.259e+10 perf-stat.ps.instructions >> 0.08 =C2=B1 2% -20.0% 0.07 =C2=B1 2% perf-stat.ps.ma= jor-faults >> 4636 =C2=B1 3% +15.5% 5353 =C2=B1 2% perf-stat.ps.mi= nor-faults >> 4636 =C2=B1 3% +15.5% 5353 =C2=B1 2% perf-stat.ps.pa= ge-faults >> >> >> >> phoronix-test-suite.aom-av1.0.frames_per_second >> >> 0.04 +--------------------------------------------------------------= -----+ >> | = | >> 0.035 |-+ = | >> 0.03 |-O O O O O O O O O O O O O O O O O O O O O O O O O = O O | >> | = | >> 0.025 |-+ = | >> | = | >> 0.02 |-+ = | >> | = | >> 0.015 |-+ = | >> 0.01 |-+ = | >> | = | >> 0.005 |-+ = | >> | = | >> 0 +--------------------------------------------------------------= -----+ >> >> >> [*] bisect-good sample >> [O] bisect-bad sample >> >> >> >> Disclaimer: >> Results have been estimated based on internal Intel analysis and are pro= vided >> for informational purposes only. Any difference in system hardware or so= ftware >> design or configuration may affect actual performance. >> >> >> Thanks, >> Rong Chen >> --===============3283624816074374174==--