lkml.org 
[lkml]   [2020]   [Dec]   [23]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
SubjectRe: [LKP] [locking/rwsem] 617f3ef951: unixbench.score -21.2% regression
From
Date
Hi Waiman,

Do you have time to look at this? Thanks.
As you describe in commit: 617f3ef95177840c77f59c2aec1029d27d5547d6
("locking/rwsem: Remove reader optimistic spinning"), The patch that
disables reader optimistic spinning shows reduced performance at lightly
loaded cases, so for this regression, Is it as expected?

On 12/17/2020 9:33 AM, kernel test robot wrote:
>
> Greeting,
>
> FYI, we noticed a -21.2% regression of unixbench.score due to commit:
>
>
> commit: 617f3ef95177840c77f59c2aec1029d27d5547d6 ("locking/rwsem: Remove reader optimistic spinning")
> https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master
>
>
> in testcase: unixbench
> on test machine: 16 threads Intel(R) Xeon(R) E-2278G CPU @ 3.40GHz with 32G memory
> with following parameters:
>
> runtime: 300s
> nr_task: 30%
> test: shell8
> cpufreq_governor: performance
> ucode: 0xde
>
> 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 <oliver.sang@intel.com>
>
>
> 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-cfl-e1/shell8/unixbench/0xde
>
> commit:
> 1a728dff85 ("locking/rwsem: Enable reader optimistic lock stealing")
> 617f3ef951 ("locking/rwsem: Remove reader optimistic spinning")
>
> 1a728dff855a318b 617f3ef95177840c77f59c2aec1
> ---------------- ---------------------------
> fail:runs %reproduction fail:runs
> | | |
> 39:4 -992% :4 perf-profile.calltrace.cycles-pp.error_entry
> 25:4 -635% :4 perf-profile.children.cycles-pp.error_entry
> %stddev %change %stddev
> \ | \
> 21807 ± 3% -21.2% 17186 unixbench.score
> 1287072 ± 3% -38.7% 788414 unixbench.time.involuntary_context_switches
> 37161 ± 4% +31.3% 48798 unixbench.time.major_page_faults
> 1.047e+08 ± 3% -21.1% 82610985 unixbench.time.minor_page_faults
> 1341 -27.1% 978.00 unixbench.time.percent_of_cpu_this_job_got
> 370.87 -33.3% 247.55 unixbench.time.system_time
> 490.05 -23.3% 376.03 unixbench.time.user_time
> 3083520 ± 3% +59.7% 4924900 unixbench.time.voluntary_context_switches
> 824314 ± 3% -21.2% 649654 unixbench.workload
> 0.03 ± 27% -51.9% 0.02 ± 59% perf-sched.sch_delay.avg.ms.schedule_timeout.rcu_gp_kthread.kthread.ret_from_fork
> 385.15 ± 2% +62.5% 625.72 uptime.idle
> 17.03 -1.8% 16.73 boot-time.boot
> 11.01 -1.6% 10.83 boot-time.dhcp
> 214.12 ± 3% -3.1% 207.49 boot-time.idle
> 13.72 ± 4% +23.5 37.24 mpstat.cpu.all.idle%
> 1.06 -0.1 0.94 mpstat.cpu.all.irq%
> 49.32 ± 2% -11.8 37.53 mpstat.cpu.all.sys%
> 35.24 ± 2% -11.6 23.68 mpstat.cpu.all.usr%
> 15.50 ± 3% +145.2% 38.00 vmstat.cpu.id
> 49.00 ± 2% -22.4% 38.00 vmstat.cpu.sy
> 33.75 ± 2% -33.3% 22.50 ± 2% vmstat.cpu.us
> 21.75 ± 3% -33.3% 14.50 ± 3% vmstat.procs.r
> 97370 ± 3% +56.4% 152258 vmstat.system.cs
> 37589 -2.1% 36804 vmstat.system.in
> 11861 ± 9% -18.0% 9730 slabinfo.filp.active_objs
> 13242 ± 8% -15.5% 11184 slabinfo.filp.num_objs
> 14731 ± 7% -9.5% 13325 ± 5% slabinfo.kmalloc-8.active_objs
> 14731 ± 7% -9.5% 13325 ± 5% slabinfo.kmalloc-8.num_objs
> 5545 ± 2% -13.8% 4780 ± 4% slabinfo.pid.active_objs
> 5563 ± 2% -13.8% 4793 ± 4% slabinfo.pid.num_objs
> 5822 ± 14% -40.4% 3468 ± 5% slabinfo.task_delay_info.active_objs
> 5825 ± 14% -40.5% 3468 ± 5% slabinfo.task_delay_info.num_objs
> 32104492 ± 3% +303.3% 1.295e+08 ± 11% cpuidle.C1.time
> 882330 ± 5% +131.5% 2042656 ± 10% cpuidle.C1.usage
> 21965263 ± 3% +340.5% 96762398 ± 14% cpuidle.C1E.time
> 442911 ± 2% +211.3% 1378866 ± 14% cpuidle.C1E.usage
> 6511399 ± 4% +606.6% 46010023 ± 13% cpuidle.C3.time
> 66498 ± 5% +559.3% 438404 ± 9% cpuidle.C3.usage
> 101453 ± 18% +139.9% 243422 ± 7% cpuidle.C6.usage
> 11612925 ± 66% +228.1% 38105483 ± 21% cpuidle.C8.time
> 12182 ± 65% +220.7% 39070 ± 21% cpuidle.C8.usage
> 173670 ± 30% +166.5% 462831 ± 7% cpuidle.POLL.time
> 17441 ± 18% +195.6% 51559 ± 6% cpuidle.POLL.usage
> 59652 +0.6% 60011 proc-vmstat.nr_inactive_anon
> 5840 +1.4% 5922 proc-vmstat.nr_kernel_stack
> 1404 +3.1% 1447 proc-vmstat.nr_page_table_pages
> 13255 -1.7% 13026 proc-vmstat.nr_slab_reclaimable
> 14861 -6.9% 13835 ± 4% proc-vmstat.nr_slab_unreclaimable
> 59652 +0.6% 60011 proc-vmstat.nr_zone_inactive_anon
> 70705198 ± 3% -21.1% 55762717 proc-vmstat.numa_hit
> 70705198 ± 3% -21.1% 55762717 proc-vmstat.numa_local
> 116793 ± 3% -32.1% 79354 proc-vmstat.pgactivate
> 74803470 ± 3% -20.7% 59317396 proc-vmstat.pgalloc_normal
> 1.049e+08 ± 3% -21.1% 82849276 proc-vmstat.pgfault
> 74776100 ± 3% -20.7% 59288922 proc-vmstat.pgfree
> 5338118 ± 3% -24.5% 4030373 proc-vmstat.pgreuse
> 3413 ± 2% -19.6% 2745 proc-vmstat.thp_fault_alloc
> 1466090 ± 3% -21.2% 1155570 proc-vmstat.unevictable_pgs_culled
> 382612 ± 60% -66.4% 128559 ± 7% sched_debug.cfs_rq:/.load.max
> 274.29 ± 53% -48.2% 142.09 ± 15% sched_debug.cfs_rq:/.load_avg.avg
> 452211 -25.5% 336912 sched_debug.cfs_rq:/.min_vruntime.avg
> 489002 -22.6% 378258 ± 2% sched_debug.cfs_rq:/.min_vruntime.max
> 440439 -26.4% 324253 sched_debug.cfs_rq:/.min_vruntime.min
> 210358 ± 8% -21.6% 164917 ± 9% sched_debug.cpu.avg_idle.avg
> 666992 ± 9% -14.4% 571102 ± 3% sched_debug.cpu.avg_idle.max
> 183734 ± 2% -10.3% 164859 sched_debug.cpu.avg_idle.stddev
> 0.54 ± 19% -29.1% 0.38 ± 15% sched_debug.cpu.clock.stddev
> 199240 ± 3% +53.0% 304861 sched_debug.cpu.nr_switches.avg
> 215766 ± 3% +48.7% 320802 sched_debug.cpu.nr_switches.max
> 186135 ± 3% +55.3% 289040 sched_debug.cpu.nr_switches.min
> 82.12 ± 12% +247.2% 285.12 ± 11% sched_debug.cpu.nr_uninterruptible.max
> -122.88 +312.6% -507.00 sched_debug.cpu.nr_uninterruptible.min
> 51.87 ± 13% +263.6% 188.58 ± 8% sched_debug.cpu.nr_uninterruptible.stddev
> 242600 ± 5% -33.6% 161055 softirqs.CPU0.RCU
> 15144 ± 2% +14.4% 17328 ± 3% softirqs.CPU0.SCHED
> 14371 +13.8% 16354 ± 3% softirqs.CPU1.SCHED
> 14064 ± 2% +10.1% 15480 softirqs.CPU10.SCHED
> 241990 ± 4% -33.1% 161882 softirqs.CPU11.RCU
> 13889 +12.7% 15650 softirqs.CPU11.SCHED
> 239184 ± 7% -33.4% 159310 softirqs.CPU12.RCU
> 13988 ± 3% +12.3% 15711 softirqs.CPU13.SCHED
> 13744 ± 2% +14.1% 15688 ± 2% softirqs.CPU14.SCHED
> 14104 +12.8% 15912 ± 2% softirqs.CPU2.SCHED
> 240729 ± 4% -32.5% 162396 softirqs.CPU3.RCU
> 14143 ± 2% +13.6% 16064 ± 3% softirqs.CPU3.SCHED
> 240581 ± 5% -31.9% 163890 softirqs.CPU4.RCU
> 13936 +10.5% 15402 softirqs.CPU5.SCHED
> 13761 ± 3% +12.9% 15532 softirqs.CPU6.SCHED
> 14059 +11.8% 15724 ± 2% softirqs.CPU7.SCHED
> 241270 ± 5% -34.5% 157931 ± 2% softirqs.CPU8.RCU
> 13695 +11.8% 15314 ± 2% softirqs.CPU8.SCHED
> 225021 +12.1% 252248 softirqs.SCHED
> 14.09 ± 73% -14.1 0.00 perf-profile.calltrace.cycles-pp.do_filp_open.do_sys_openat2.do_sys_open.do_syscall_64.entry_SYSCALL_64_after_hwframe
> 14.09 ± 73% -14.1 0.00 perf-profile.calltrace.cycles-pp.path_openat.do_filp_open.do_sys_openat2.do_sys_open.do_syscall_64
> 10.52 ± 60% -10.5 0.00 perf-profile.calltrace.cycles-pp.setlocale
> 9.13 ±104% -9.1 0.00 perf-profile.calltrace.cycles-pp.free_pgtables.exit_mmap.mmput.do_exit.do_group_exit
> 8.43 ± 62% -8.4 0.00 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.__open64_nocancel.setlocale
> 8.43 ± 62% -8.4 0.00 perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.__open64_nocancel.setlocale
> 8.43 ± 62% -8.4 0.00 perf-profile.calltrace.cycles-pp.do_sys_open.do_syscall_64.entry_SYSCALL_64_after_hwframe.__open64_nocancel.setlocale
> 8.43 ± 62% -8.4 0.00 perf-profile.calltrace.cycles-pp.do_sys_openat2.do_sys_open.do_syscall_64.entry_SYSCALL_64_after_hwframe.__open64_nocancel
> 8.43 ± 62% -8.4 0.00 perf-profile.calltrace.cycles-pp.__open64_nocancel.setlocale
> 6.35 ±101% -6.3 0.00 perf-profile.calltrace.cycles-pp.unlink_file_vma.free_pgtables.exit_mmap.mmput.do_exit
> 4.86 ±102% -4.9 0.00 perf-profile.calltrace.cycles-pp.terminate_walk.path_openat.do_filp_open.do_sys_openat2.do_sys_open
> 6.35 ±101% -3.6 2.78 ±173% perf-profile.calltrace.cycles-pp.exit_mmap.mmput.do_exit.do_group_exit.__x64_sys_exit_group
> 6.35 ±101% -3.6 2.78 ±173% perf-profile.calltrace.cycles-pp.mmput.do_exit.do_group_exit.__x64_sys_exit_group.do_syscall_64
> 7.98 ± 60% -3.0 5.00 ±173% perf-profile.calltrace.cycles-pp.exit_to_user_mode_prepare.syscall_exit_to_user_mode.entry_SYSCALL_64_after_hwframe
> 7.98 ± 60% -3.0 5.00 ±173% perf-profile.calltrace.cycles-pp.arch_do_signal.exit_to_user_mode_prepare.syscall_exit_to_user_mode.entry_SYSCALL_64_after_hwframe
> 7.98 ± 60% -3.0 5.00 ±173% perf-profile.calltrace.cycles-pp.do_group_exit.get_signal.arch_do_signal.exit_to_user_mode_prepare.syscall_exit_to_user_mode
> 7.98 ± 60% -3.0 5.00 ±173% perf-profile.calltrace.cycles-pp.do_exit.do_group_exit.get_signal.arch_do_signal.exit_to_user_mode_prepare
> 7.98 ± 60% -3.0 5.00 ±173% perf-profile.calltrace.cycles-pp.exit_mmap.mmput.do_exit.do_group_exit.get_signal
> 7.98 ± 60% -3.0 5.00 ±173% perf-profile.calltrace.cycles-pp.get_signal.arch_do_signal.exit_to_user_mode_prepare.syscall_exit_to_user_mode.entry_SYSCALL_64_after_hwframe
> 7.98 ± 60% -3.0 5.00 ±173% perf-profile.calltrace.cycles-pp.mmput.do_exit.do_group_exit.get_signal.arch_do_signal
> 7.98 ± 60% -3.0 5.00 ±173% perf-profile.calltrace.cycles-pp.syscall_exit_to_user_mode.entry_SYSCALL_64_after_hwframe
> 6.35 ±101% -0.8 5.55 ±173% perf-profile.calltrace.cycles-pp.__x64_sys_exit_group.do_syscall_64.entry_SYSCALL_64_after_hwframe
> 6.35 ±101% -0.8 5.55 ±173% perf-profile.calltrace.cycles-pp.do_group_exit.__x64_sys_exit_group.do_syscall_64.entry_SYSCALL_64_after_hwframe
> 6.35 ±101% -0.8 5.55 ±173% perf-profile.calltrace.cycles-pp.do_exit.do_group_exit.__x64_sys_exit_group.do_syscall_64.entry_SYSCALL_64_after_hwframe
> 14.09 ± 73% -14.1 0.00 perf-profile.children.cycles-pp.do_filp_open
> 14.09 ± 73% -14.1 0.00 perf-profile.children.cycles-pp.do_sys_open
> 14.09 ± 73% -14.1 0.00 perf-profile.children.cycles-pp.do_sys_openat2
> 14.09 ± 73% -14.1 0.00 perf-profile.children.cycles-pp.path_openat
> 10.52 ± 60% -10.5 0.00 perf-profile.children.cycles-pp.setlocale
> 9.13 ±104% -9.1 0.00 perf-profile.children.cycles-pp.free_pgtables
> 8.43 ± 62% -8.4 0.00 perf-profile.children.cycles-pp.__open64_nocancel
> 6.35 ±101% -6.3 0.00 perf-profile.children.cycles-pp.sync_regs
> 6.35 ±101% -6.3 0.00 perf-profile.children.cycles-pp.unlink_file_vma
> 4.86 ±102% -4.9 0.00 perf-profile.children.cycles-pp.terminate_walk
> 7.98 ± 60% -3.0 5.00 ±173% perf-profile.children.cycles-pp.exit_to_user_mode_prepare
> 7.98 ± 60% -3.0 5.00 ±173% perf-profile.children.cycles-pp.arch_do_signal
> 7.98 ± 60% -3.0 5.00 ±173% perf-profile.children.cycles-pp.get_signal
> 7.98 ± 60% -3.0 5.00 ±173% perf-profile.children.cycles-pp.syscall_exit_to_user_mode
> 6.35 ±101% -0.8 5.55 ±173% perf-profile.children.cycles-pp.__x64_sys_exit_group
> 5.65 ±106% -0.7 5.00 ±173% perf-profile.children.cycles-pp.__mod_memcg_lruvec_state
> 5.65 ±106% -0.1 5.55 ±173% perf-profile.children.cycles-pp.handle_mm_fault
> 6.35 ±101% -6.3 0.00 perf-profile.self.cycles-pp.sync_regs
> 31246 ± 5% +200.7% 93965 interrupts.CAL:Function_call_interrupts
> 2091 ± 11% +179.5% 5844 interrupts.CPU0.CAL:Function_call_interrupts
> 26624 ± 2% -20.9% 21057 interrupts.CPU0.RES:Rescheduling_interrupts
> 500.25 ± 7% +378.2% 2392 interrupts.CPU0.TLB:TLB_shootdowns
> 1931 ± 4% +201.7% 5827 interrupts.CPU1.CAL:Function_call_interrupts
> 26939 ± 4% -22.2% 20957 interrupts.CPU1.RES:Rescheduling_interrupts
> 499.25 ± 5% +374.8% 2370 ± 4% interrupts.CPU1.TLB:TLB_shootdowns
> 1967 ± 6% +195.8% 5820 interrupts.CPU10.CAL:Function_call_interrupts
> 26951 ± 2% -19.9% 21594 interrupts.CPU10.RES:Rescheduling_interrupts
> 546.25 +336.4% 2384 ± 4% interrupts.CPU10.TLB:TLB_shootdowns
> 1927 ± 3% +211.0% 5994 interrupts.CPU11.CAL:Function_call_interrupts
> 26619 ± 2% -18.7% 21630 interrupts.CPU11.RES:Rescheduling_interrupts
> 514.25 ± 6% +361.8% 2374 ± 2% interrupts.CPU11.TLB:TLB_shootdowns
> 1887 ± 9% +206.9% 5791 ± 4% interrupts.CPU12.CAL:Function_call_interrupts
> 26377 ± 4% -19.5% 21237 ± 2% interrupts.CPU12.RES:Rescheduling_interrupts
> 496.00 ± 3% +372.8% 2345 ± 2% interrupts.CPU12.TLB:TLB_shootdowns
> 1912 ± 7% +208.6% 5899 ± 2% interrupts.CPU13.CAL:Function_call_interrupts
> 27187 ± 4% -20.1% 21734 ± 2% interrupts.CPU13.RES:Rescheduling_interrupts
> 500.50 ± 5% +380.5% 2405 ± 3% interrupts.CPU13.TLB:TLB_shootdowns
> 1891 ± 7% +215.1% 5960 ± 2% interrupts.CPU14.CAL:Function_call_interrupts
> 26046 ± 4% -17.0% 21627 interrupts.CPU14.RES:Rescheduling_interrupts
> 506.25 ± 9% +367.9% 2368 ± 2% interrupts.CPU14.TLB:TLB_shootdowns
> 1960 ± 4% +195.8% 5798 interrupts.CPU15.CAL:Function_call_interrupts
> 26567 ± 4% -19.7% 21342 ± 2% interrupts.CPU15.RES:Rescheduling_interrupts
> 536.50 ± 3% +350.6% 2417 interrupts.CPU15.TLB:TLB_shootdowns
> 2001 ± 4% +190.3% 5811 ± 2% interrupts.CPU2.CAL:Function_call_interrupts
> 27172 ± 3% -21.5% 21338 interrupts.CPU2.RES:Rescheduling_interrupts
> 539.00 ± 4% +332.7% 2332 ± 3% interrupts.CPU2.TLB:TLB_shootdowns
> 1950 ± 6% +198.1% 5814 interrupts.CPU3.CAL:Function_call_interrupts
> 26579 ± 2% -19.2% 21479 interrupts.CPU3.RES:Rescheduling_interrupts
> 504.75 ± 8% +369.0% 2367 ± 4% interrupts.CPU3.TLB:TLB_shootdowns
> 1965 ± 10% +199.8% 5891 interrupts.CPU4.CAL:Function_call_interrupts
> 26811 ± 3% -19.8% 21512 interrupts.CPU4.RES:Rescheduling_interrupts
> 515.75 ± 5% +358.1% 2362 ± 2% interrupts.CPU4.TLB:TLB_shootdowns
> 1935 ± 7% +206.3% 5926 interrupts.CPU5.CAL:Function_call_interrupts
> 26997 ± 3% -20.4% 21486 interrupts.CPU5.RES:Rescheduling_interrupts
> 531.25 ± 3% +348.6% 2383 interrupts.CPU5.TLB:TLB_shootdowns
> 1904 ± 5% +214.0% 5979 interrupts.CPU6.CAL:Function_call_interrupts
> 26702 ± 4% -20.1% 21341 interrupts.CPU6.RES:Rescheduling_interrupts
> 505.25 ± 3% +354.6% 2296 ± 2% interrupts.CPU6.TLB:TLB_shootdowns
> 1933 ± 4% +205.3% 5901 interrupts.CPU7.CAL:Function_call_interrupts
> 26505 ± 3% -19.6% 21302 interrupts.CPU7.RES:Rescheduling_interrupts
> 494.00 ± 4% +371.1% 2327 ± 3% interrupts.CPU7.TLB:TLB_shootdowns
> 2003 ± 5% +190.5% 5818 interrupts.CPU8.CAL:Function_call_interrupts
> 26799 ± 3% -20.2% 21376 ± 2% interrupts.CPU8.RES:Rescheduling_interrupts
> 519.00 ± 4% +364.7% 2411 ± 4% interrupts.CPU8.TLB:TLB_shootdowns
> 1986 ± 10% +196.5% 5888 ± 2% interrupts.CPU9.CAL:Function_call_interrupts
> 26994 ± 3% -20.6% 21439 interrupts.CPU9.RES:Rescheduling_interrupts
> 526.00 ± 6% +351.8% 2376 interrupts.CPU9.TLB:TLB_shootdowns
> 427874 ± 3% -20.0% 342458 interrupts.RES:Rescheduling_interrupts
> 8234 ± 2% +360.4% 37914 ± 2% interrupts.TLB:TLB_shootdowns
> 47.20 +6.4% 50.22 perf-stat.i.MPKI
> 9.902e+09 ± 3% -19.6% 7.965e+09 perf-stat.i.branch-instructions
> 2.24 +0.1 2.32 perf-stat.i.branch-miss-rate%
> 2.143e+08 ± 3% -16.4% 1.791e+08 perf-stat.i.branch-misses
> 99172014 ± 2% -15.6% 83714922 perf-stat.i.cache-misses
> 2.363e+09 ± 3% -14.4% 2.022e+09 perf-stat.i.cache-references
> 101957 ± 3% +56.1% 159180 perf-stat.i.context-switches
> 1.20 ± 3% -5.8% 1.13 perf-stat.i.cpi
> 5.571e+10 -23.7% 4.249e+10 perf-stat.i.cpu-cycles
> 15111 ± 3% +116.2% 32667 perf-stat.i.cpu-migrations
> 610.84 ± 2% -10.2% 548.56 perf-stat.i.cycles-between-cache-misses
> 1.234e+10 ± 3% -19.7% 9.908e+09 perf-stat.i.dTLB-loads
> 0.05 +0.0 0.05 perf-stat.i.dTLB-store-miss-rate%
> 3568011 ± 3% -14.8% 3038208 perf-stat.i.dTLB-store-misses
> 7.127e+09 ± 3% -19.9% 5.706e+09 perf-stat.i.dTLB-stores
> 57.81 -2.0 55.78 perf-stat.i.iTLB-load-miss-rate%
> 10339942 ± 3% -16.8% 8603048 perf-stat.i.iTLB-load-misses
> 7643641 ± 3% -7.7% 7057126 perf-stat.i.iTLB-loads
> 4.833e+10 ± 3% -19.6% 3.887e+10 perf-stat.i.instructions
> 5183 -1.9% 5085 perf-stat.i.instructions-per-iTLB-miss
> 0.85 ± 4% +5.6% 0.90 perf-stat.i.ipc
> 577.76 ± 4% +32.0% 762.53 perf-stat.i.major-faults
> 3.48 -23.7% 2.66 perf-stat.i.metric.GHz
> 0.55 ± 44% -77.5% 0.12 ± 30% perf-stat.i.metric.K/sec
> 1986 ± 3% -19.3% 1602 perf-stat.i.metric.M/sec
> 1599403 ± 3% -20.7% 1268462 perf-stat.i.minor-faults
> 5321084 ± 3% -15.9% 4474662 perf-stat.i.node-loads
> 32016465 ± 3% -17.1% 26551337 perf-stat.i.node-stores
> 1599980 ± 3% -20.7% 1269225 perf-stat.i.page-faults
> 48.88 +6.4% 52.03 perf-stat.overall.MPKI
> 2.16 +0.1 2.25 perf-stat.overall.branch-miss-rate%
> 4.20 -0.1 4.14 perf-stat.overall.cache-miss-rate%
> 1.15 ± 4% -5.3% 1.09 perf-stat.overall.cpi
> 562.37 ± 3% -9.7% 507.65 perf-stat.overall.cycles-between-cache-misses
> 0.05 +0.0 0.05 perf-stat.overall.dTLB-store-miss-rate%
> 57.50 -2.6 54.94 perf-stat.overall.iTLB-load-miss-rate%
> 4674 -3.3% 4518 perf-stat.overall.instructions-per-iTLB-miss
> 0.87 ± 4% +5.4% 0.91 perf-stat.overall.ipc
> 9.742e+09 ± 3% -19.6% 7.836e+09 perf-stat.ps.branch-instructions
> 2.108e+08 ± 3% -16.5% 1.762e+08 perf-stat.ps.branch-misses
> 97569500 ± 2% -15.6% 82348192 perf-stat.ps.cache-misses
> 2.324e+09 ± 3% -14.4% 1.989e+09 perf-stat.ps.cache-references
> 100305 ± 3% +56.1% 156566 perf-stat.ps.context-switches
> 5.481e+10 -23.7% 4.18e+10 perf-stat.ps.cpu-cycles
> 14866 ± 3% +116.1% 32130 perf-stat.ps.cpu-migrations
> 1.214e+10 ± 3% -19.7% 9.746e+09 perf-stat.ps.dTLB-loads
> 3510146 ± 3% -14.9% 2988346 perf-stat.ps.dTLB-store-misses
> 7.012e+09 ± 3% -20.0% 5.612e+09 perf-stat.ps.dTLB-stores
> 10172369 ± 3% -16.8% 8461956 perf-stat.ps.iTLB-load-misses
> 7519764 ± 3% -7.7% 6941264 perf-stat.ps.iTLB-loads
> 4.755e+10 ± 3% -19.6% 3.823e+10 perf-stat.ps.instructions
> 568.39 ± 4% +32.0% 750.01 perf-stat.ps.major-faults
> 1573444 ± 3% -20.7% 1247622 perf-stat.ps.minor-faults
> 5235200 ± 3% -15.9% 4401366 perf-stat.ps.node-loads
> 31497428 ± 3% -17.1% 26115335 perf-stat.ps.node-stores
> 1574012 ± 3% -20.7% 1248372 perf-stat.ps.page-faults
> 3.111e+12 ± 3% -20.5% 2.473e+12 perf-stat.total.instructions
>
>
>
> unixbench.time.user_time
>
> 500 +---------------------------------------------------------------------+
> |. +..+..+ +..+..+.+..+..+.+..+..+.+..+..+.+..+..+..+.+..+..+.+..|
> 480 |-+ |
> | |
> 460 |-+ |
> | |
> 440 |-+ |
> | |
> 420 |-+ |
> | |
> 400 |-+ |
> | |
> 380 |-+ O O O O |
> | O O O O O O O O O O O O O O |
> 360 +---------------------------------------------------------------------+
>
>
> unixbench.time.system_time
>
> 380 +---------------------------------------------------------------------+
> |..+.+..+..+.+..+..+..+.+..+. +. +.+..+..+.+..+..+..+.+. +.+..|
> 360 |-+ |
> | |
> 340 |-+ |
> | |
> 320 |-+ |
> | |
> 300 |-+ |
> | |
> 280 |-+ |
> | |
> 260 |-+ O |
> | O O O O O O O O O O O O O O O O O |
> 240 +---------------------------------------------------------------------+
>
>
> unixbench.time.percent_of_cpu_this_job_got
>
> 1400 +--------------------------------------------------------------------+
> |..+.+..+..+.+.. .+.. .+.+..+..+.+..+.+..+..+.+..+..+. |
> 1350 |-+ +..+ +. +..+..+.+..|
> 1300 |-+ |
> | |
> 1250 |-+ |
> 1200 |-+ |
> | |
> 1150 |-+ |
> 1100 |-+ |
> | |
> 1050 |-+ |
> 1000 |-+ |
> | O O O O O O O O O O O O O O O O O O |
> 950 +--------------------------------------------------------------------+
>
>
> unixbench.time.minor_page_faults
>
> 1.1e+08 +----------------------------------------------------------------+
> |..+.+..+.+..+.+..+.+..+.+ + +..+.+..+.+..+.+ + +..+.|
> 1.05e+08 |-+ : :: : : :: : |
> | : : : : : : : : |
> 1e+08 |-+ : : : : : : : : |
> | + + + + |
> 9.5e+07 |-+ |
> | |
> 9e+07 |-+ |
> | |
> 8.5e+07 |-+ |
> | O O O O O O O O O O O O O O O |
> 8e+07 |-+ |
> | |
> 7.5e+07 +----------------------------------------------------------------+
>
>
> unixbench.time.voluntary_context_switches
>
> 5e+06 +-----------------------------------------------------------------+
> | O O O O O O O O O O O |
> | O O |
> 4.5e+06 |-+ O |
> | |
> | |
> 4e+06 |-+ |
> | |
> 3.5e+06 |-+ |
> | |
> | .+. .+.+..+.+..+. +.. +.+..+.+..+.+.. + +.+..|
> 3e+06 |.+ +..+.+. +.. + .. +. .. + .. |
> | + + + + |
> | |
> 2.5e+06 +-----------------------------------------------------------------+
>
>
> unixbench.time.involuntary_context_switches
>
> 1.4e+06 +-----------------------------------------------------------------+
> |..+.+..+.+..+.+.. +.+.. +.. +.+..+.+..+.+..+ + + |
> 1.3e+06 |-+ .. + .. + .. + .. + ..|
> | +.+ + + + + + |
> 1.2e+06 |-+ |
> | |
> 1.1e+06 |-+ |
> | |
> 1e+06 |-+ |
> | |
> 900000 |-+ |
> | |
> 800000 |-+O O O O O O O O O O O O O O |
> | O O O O |
> 700000 +-----------------------------------------------------------------+
>
>
> unixbench.score
>
> 23000 +-------------------------------------------------------------------+
> |..+.+..+.+..+..+.+..+..+.+ + +.+..+.+..+..+.+ + +.+..|
> 22000 |-+ + : : : : : : : |
> 21000 |-+ + : : : : : : : |
> | + : : : : :: |
> 20000 |-+ + + + |
> | |
> 19000 |-+ |
> | |
> 18000 |-+ |
> 17000 |-+O O O O O O O O O O O O |
> | O O O |
> 16000 |-+ |
> | O O O |
> 15000 +-------------------------------------------------------------------+
>
>
> unixbench.workload
>
> 900000 +------------------------------------------------------------------+
> | |
> 850000 |..+.+..+.+..+.+..+..+.+..+ + +..+.+..+.+..+..+ + +.+..|
> | : + + : : + : + |
> 800000 |-+ : + + : : + : + |
> | + + + + |
> 750000 |-+ |
> | |
> 700000 |-+ |
> | |
> 650000 |-+O O O O O O O O O O O O O O O |
> | |
> 600000 |-+ O O O |
> | |
> 550000 +------------------------------------------------------------------+
>
>
> [*] 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
>
>
> _______________________________________________
> LKP mailing list -- lkp@lists.01.org
> To unsubscribe send an email to lkp-leave@lists.01.org
>

--
Zhengjun Xing

\
 
 \ /
  Last update: 2020-12-23 08:40    [W:0.059 / U:1.460 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site