lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [day] [month] [year] [list]
Message-ID: <20191022060143.GB5781@xsang-OptiPlex-9020>
Date:   Tue, 22 Oct 2019 14:01:43 +0800
From:   kernel test robot <oliver.sang@...el.com>
To:     Song Liu <songliubraving@...com>
Cc:     Ingo Molnar <mingo@...nel.org>,
        Peter Zijlstra <peterz@...radead.org>,
        Arnaldo Carvalho de Melo <acme@...nel.org>,
        Linus Torvalds <torvalds@...ux-foundation.org>,
        Sasha Levin <sashal@...nel.org>,
        Thomas Gleixner <tglx@...utronix.de>,
        LKML <linux-kernel@...r.kernel.org>, lkp@...ts.01.org
Subject: [perf/core]  7fa343b7fd:  vm-scalability.median -1.5% regression

Greeting,

FYI, we noticed a -1.5% regression of vm-scalability.median due to commit:


commit: 7fa343b7fdc4f351de4e3f28d5c285937dd1f42f ("perf/core: Fix corner case in perf_rotate_context()")
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master

in testcase: vm-scalability
on test machine: 144 threads Intel(R) Xeon(R) CPU E7-8890 v3 @ 2.50GHz with 512G memory
with following parameters:

	runtime: 300s
	size: 512G
	test: anon-w-rand
	cpufreq_governor: performance
	ucode: 0x16

test-description: The motivation behind this suite is to exercise functions and regions of the mm/ of the Linux kernel which are of interest to us.
test-url: https://git.kernel.org/cgit/linux/kernel/git/wfg/vm-scalability.git/



If you fix the issue, kindly add following tag
Reported-by: kernel test robot <oliver.sang@...el.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/rootfs/runtime/size/tbox_group/test/testcase/ucode:
  gcc-7/performance/x86_64-rhel-7.6/debian-x86_64-2019-09-23.cgz/300s/512G/lkp-hsw-4ex1/anon-w-rand/vm-scalability/0x16

commit: 
  d44248a413 ("perf/core: Rework memory accounting in perf_mmap()")
  7fa343b7fd ("perf/core: Fix corner case in perf_rotate_context()")

d44248a41337731a 7fa343b7fdc4f351de4e3f28d5c 
---------------- --------------------------- 
       fail:runs  %reproduction    fail:runs
           |             |             |    
          0:4           -5%           0:4     perf-profile.children.cycles-pp.error_entry
          0:4           -4%           0:4     perf-profile.self.cycles-pp.error_entry
         %stddev     %change         %stddev
             \          |                \  
     35614            -1.5%      35070        vm-scalability.median
   4909819            -1.6%    4832961        vm-scalability.throughput
    133541            +5.7%     141118 ±  2%  vm-scalability.time.involuntary_context_switches
      2731            +3.3%       2821        vm-scalability.time.system_time
     35579            +1.4%      36086        vm-scalability.time.user_time
   3708205 ±100%     -86.9%     487323 ±157%  cpuidle.C1E.usage
   3698401 ±101%     -87.1%     478037 ±161%  turbostat.C1E
      4973 ± 22%     +35.4%       6734 ± 15%  numa-vmstat.node2.nr_slab_reclaimable
     68267 ±  4%      -5.5%      64522 ±  5%  numa-vmstat.node3.nr_unevictable
     68267 ±  4%      -5.5%      64522 ±  5%  numa-vmstat.node3.nr_zone_unevictable
     19895 ± 22%     +35.4%      26937 ± 15%  numa-meminfo.node2.KReclaimable
     19895 ± 22%     +35.4%      26937 ± 15%  numa-meminfo.node2.SReclaimable
     69426 ± 11%     -13.6%      59976 ± 10%  numa-meminfo.node3.Slab
    273071 ±  4%      -5.5%     258090 ±  5%  numa-meminfo.node3.Unevictable
    196.00           -11.2%     174.00 ±  9%  proc-vmstat.nr_isolated_anon
      7144            +2.6%       7327        proc-vmstat.nr_mapped
     64412            +1.8%      65599        proc-vmstat.nr_shmem
     51657            +1.0%      52191        proc-vmstat.nr_slab_unreclaimable
      5764 ±  7%     +39.9%       8061 ±  8%  proc-vmstat.pgactivate
      2685 ± 19%     +32.9%       3569        sched_debug.cfs_rq:/.runnable_weight.min
  -1316030           +24.8%   -1642376        sched_debug.cfs_rq:/.spread0.min
    631.04 ±  6%     -10.1%     567.17 ±  3%  sched_debug.cfs_rq:/.util_est_enqueued.max
     17178 ± 10%     +19.8%      20576 ± 15%  sched_debug.cpu.nr_switches.max
      6029 ±  3%     +22.4%       7379 ± 15%  sched_debug.cpu.ttwu_count.max
      5111 ±  5%     +31.8%       6739 ± 19%  sched_debug.cpu.ttwu_local.max
     90333 ± 10%     +34.4%     121364 ± 11%  syscalls.sys_close.max
     14504 ±  2%     +10.9%      16083 ±  3%  syscalls.sys_mmap.med
 1.535e+08 ± 11%  +5.1e+08   6.604e+08 ± 79%  syscalls.sys_mmap.noise.100%
  3.79e+08 ±  4%  +5.1e+08   8.894e+08 ± 57%  syscalls.sys_mmap.noise.2%
 3.469e+08 ±  4%  +5.1e+08   8.548e+08 ± 60%  syscalls.sys_mmap.noise.25%
 3.776e+08 ±  4%  +5.1e+08   8.875e+08 ± 58%  syscalls.sys_mmap.noise.5%
 2.896e+08 ±  5%  +5.1e+08   7.978e+08 ± 65%  syscalls.sys_mmap.noise.50%
    198724 ± 36%     +42.5%     283239 ± 20%  syscalls.sys_write.max
 5.145e+08            +0.9%  5.193e+08        perf-stat.i.cache-references
     78.93            +5.2       84.10        perf-stat.i.iTLB-load-miss-rate%
    394771 ±  9%     +29.5%     511163 ±  4%  perf-stat.i.iTLB-load-misses
    327477 ±  8%     -26.8%     239622 ±  5%  perf-stat.i.iTLB-loads
   5408504 ±  9%     +13.6%    6141622 ±  6%  perf-stat.i.node-store-misses
     54.74 ±  7%     +13.5       68.20 ±  2%  perf-stat.overall.iTLB-load-miss-rate%
    109266 ± 10%     -23.4%      83711 ±  4%  perf-stat.overall.instructions-per-iTLB-miss
      1.17 ± 10%      +0.2        1.33 ±  6%  perf-stat.overall.node-store-miss-rate%
 5.126e+08            +0.9%  5.172e+08        perf-stat.ps.cache-references
    392782 ±  9%     +29.4%     508415 ±  3%  perf-stat.ps.iTLB-load-misses
    323887 ±  8%     -26.8%     237034 ±  5%  perf-stat.ps.iTLB-loads
   5327487 ±  9%     +13.6%    6051553 ±  6%  perf-stat.ps.node-store-misses
    134501 ±  5%     +11.1%     149465 ±  2%  softirqs.CPU109.TIMER
    135152 ±  5%     +10.7%     149653 ±  2%  softirqs.CPU113.TIMER
    131528 ±  4%     +13.0%     148634 ±  3%  softirqs.CPU114.TIMER
     26149 ±  8%     +17.2%      30651        softirqs.CPU16.RCU
    137403 ±  5%      +8.8%     149539 ±  2%  softirqs.CPU16.TIMER
    124932 ±  6%      +8.1%     135007 ±  5%  softirqs.CPU20.TIMER
    133590 ±  5%     +13.3%     151311 ±  2%  softirqs.CPU24.TIMER
    135781 ±  5%     +12.3%     152519        softirqs.CPU37.TIMER
    125112 ±  3%     +18.4%     148121 ±  2%  softirqs.CPU40.TIMER
    135323 ±  4%     +11.8%     151302        softirqs.CPU41.TIMER
    134035 ±  4%     +13.7%     152358        softirqs.CPU42.TIMER
    131180 ±  3%     +10.4%     144805 ±  3%  softirqs.CPU43.TIMER
    142460 ±  7%     -10.8%     127131 ±  4%  softirqs.CPU62.TIMER
    141232 ±  8%     -11.3%     125314 ±  4%  softirqs.CPU68.TIMER
     31741 ±  4%     +16.2%      36894 ±  6%  softirqs.CPU72.RCU
    137740 ±  7%     +16.7%     160714 ±  5%  softirqs.CPU90.TIMER
     32909 ±  5%     -12.7%      28718 ±  6%  softirqs.CPU94.RCU
     10850 ±  3%     +11.3%      12080 ±  3%  softirqs.CPU94.SCHED
    131437 ±  5%     +13.3%     148956 ±  3%  softirqs.CPU96.TIMER
    133760 ±  4%      +9.1%     145885 ±  4%  softirqs.CPU97.TIMER
    262.00 ± 48%     -33.7%     173.75        interrupts.50:PCI-MSI.1572868-edge.eth0-TxRx-4
    465045            +4.6%     486327        interrupts.CAL:Function_call_interrupts
      2969 ± 23%     -42.4%       1709        interrupts.CPU107.NMI:Non-maskable_interrupts
      2969 ± 23%     -42.4%       1709        interrupts.CPU107.PMI:Performance_monitoring_interrupts
    424.00 ±173%    +414.2%       2180 ± 29%  interrupts.CPU11.NMI:Non-maskable_interrupts
    424.00 ±173%    +414.2%       2180 ± 29%  interrupts.CPU11.PMI:Performance_monitoring_interrupts
    260.75 ± 88%     -80.1%      52.00 ± 57%  interrupts.CPU115.RES:Rescheduling_interrupts
    476.00 ±145%    +514.4%       2924 ± 23%  interrupts.CPU120.NMI:Non-maskable_interrupts
    476.00 ±145%    +514.4%       2924 ± 23%  interrupts.CPU120.PMI:Performance_monitoring_interrupts
    179.00 ± 37%    +143.6%     436.00 ± 46%  interrupts.CPU13.RES:Rescheduling_interrupts
      3366           -39.5%       2037 ± 33%  interrupts.CPU131.NMI:Non-maskable_interrupts
      3366           -39.5%       2037 ± 33%  interrupts.CPU131.PMI:Performance_monitoring_interrupts
     80.00 ±104%   +2714.4%       2251 ± 30%  interrupts.CPU137.NMI:Non-maskable_interrupts
     80.00 ±104%   +2714.4%       2251 ± 30%  interrupts.CPU137.PMI:Performance_monitoring_interrupts
    654.00 ± 97%    +288.8%       2542 ± 32%  interrupts.CPU138.NMI:Non-maskable_interrupts
    654.00 ± 97%    +288.8%       2542 ± 32%  interrupts.CPU138.PMI:Performance_monitoring_interrupts
    234.75 ± 60%    +616.6%       1682 ± 49%  interrupts.CPU140.RES:Rescheduling_interrupts
     70.75 ± 47%    +301.8%     284.25 ± 49%  interrupts.CPU142.RES:Rescheduling_interrupts
      2989 ± 24%     -36.2%       1908 ± 21%  interrupts.CPU16.NMI:Non-maskable_interrupts
      2989 ± 24%     -36.2%       1908 ± 21%  interrupts.CPU16.PMI:Performance_monitoring_interrupts
    836.25 ± 99%    +214.6%       2631 ± 27%  interrupts.CPU19.NMI:Non-maskable_interrupts
    836.25 ± 99%    +214.6%       2631 ± 27%  interrupts.CPU19.PMI:Performance_monitoring_interrupts
      3038 ±  4%     +12.2%       3409        interrupts.CPU23.CAL:Function_call_interrupts
      1250 ±110%    +132.8%       2910 ± 25%  interrupts.CPU24.NMI:Non-maskable_interrupts
      1250 ±110%    +132.8%       2910 ± 25%  interrupts.CPU24.PMI:Performance_monitoring_interrupts
      1696 ± 69%     +95.6%       3316 ±  2%  interrupts.CPU3.NMI:Non-maskable_interrupts
      1696 ± 69%     +95.6%       3316 ±  2%  interrupts.CPU3.PMI:Performance_monitoring_interrupts
    341.75 ± 79%    +399.2%       1706 ± 73%  interrupts.CPU37.RES:Rescheduling_interrupts
    667.00 ± 82%     -76.0%     159.75 ± 61%  interrupts.CPU39.RES:Rescheduling_interrupts
    262.00 ± 48%     -33.7%     173.75        interrupts.CPU4.50:PCI-MSI.1572868-edge.eth0-TxRx-4
    407.50 ±172%    +414.8%       2098 ± 29%  interrupts.CPU42.NMI:Non-maskable_interrupts
    407.50 ±172%    +414.8%       2098 ± 29%  interrupts.CPU42.PMI:Performance_monitoring_interrupts
    639.75 ±102%    +267.6%       2351 ± 29%  interrupts.CPU43.NMI:Non-maskable_interrupts
    639.75 ±102%    +267.6%       2351 ± 29%  interrupts.CPU43.PMI:Performance_monitoring_interrupts
    171.25 ± 12%    +162.8%     450.00 ± 64%  interrupts.CPU44.RES:Rescheduling_interrupts
    846.25 ± 99%    +202.2%       2557 ± 29%  interrupts.CPU45.NMI:Non-maskable_interrupts
    846.25 ± 99%    +202.2%       2557 ± 29%  interrupts.CPU45.PMI:Performance_monitoring_interrupts
      1087 ± 64%    +166.0%       2891 ± 24%  interrupts.CPU53.NMI:Non-maskable_interrupts
      1087 ± 64%    +166.0%       2891 ± 24%  interrupts.CPU53.PMI:Performance_monitoring_interrupts
     94.75 ± 53%    +402.9%     476.50 ± 54%  interrupts.CPU68.RES:Rescheduling_interrupts
    946.50 ± 81%    +161.1%       2471 ± 31%  interrupts.CPU7.NMI:Non-maskable_interrupts
    946.50 ± 81%    +161.1%       2471 ± 31%  interrupts.CPU7.PMI:Performance_monitoring_interrupts
    590.75 ±162%    +395.4%       2926 ± 25%  interrupts.CPU88.NMI:Non-maskable_interrupts
    590.75 ±162%    +395.4%       2926 ± 25%  interrupts.CPU88.PMI:Performance_monitoring_interrupts
    858.75 ± 96%    +233.1%       2860 ± 23%  interrupts.CPU91.NMI:Non-maskable_interrupts
    858.75 ± 96%    +233.1%       2860 ± 23%  interrupts.CPU91.PMI:Performance_monitoring_interrupts
     99.00 ± 99%    +344.7%     440.25 ± 76%  interrupts.CPU94.RES:Rescheduling_interrupts
    252147 ±  9%     +39.3%     351310        interrupts.NMI:Non-maskable_interrupts
    252147 ±  9%     +39.3%     351310        interrupts.PMI:Performance_monitoring_interrupts
    741.75 ± 10%     -16.3%     621.00 ± 14%  interrupts.TLB:TLB_shootdowns
     50.17 ±  4%      -4.3       45.90 ±  2%  perf-profile.calltrace.cycles-pp.hrtimer_interrupt.smp_apic_timer_interrupt.apic_timer_interrupt
     28.40 ± 13%      -3.7       24.72 ±  4%  perf-profile.calltrace.cycles-pp.update_process_times.tick_sched_handle.tick_sched_timer.__hrtimer_run_queues.hrtimer_interrupt
     28.57 ± 13%      -3.6       24.93 ±  4%  perf-profile.calltrace.cycles-pp.tick_sched_handle.tick_sched_timer.__hrtimer_run_queues.hrtimer_interrupt.smp_apic_timer_interrupt
     38.27 ±  8%      -3.6       34.66 ±  2%  perf-profile.calltrace.cycles-pp.__hrtimer_run_queues.hrtimer_interrupt.smp_apic_timer_interrupt.apic_timer_interrupt
     61.59 ±  4%      -3.5       58.09        perf-profile.calltrace.cycles-pp.smp_apic_timer_interrupt.apic_timer_interrupt
     64.08 ±  4%      -3.3       60.74        perf-profile.calltrace.cycles-pp.apic_timer_interrupt
      3.09 ± 35%      -1.2        1.86 ±  5%  perf-profile.calltrace.cycles-pp.run_local_timers.update_process_times.tick_sched_handle.tick_sched_timer.__hrtimer_run_queues
      3.43 ± 14%      -0.7        2.75 ±  5%  perf-profile.calltrace.cycles-pp.account_user_time.update_process_times.tick_sched_handle.tick_sched_timer.__hrtimer_run_queues
      2.42 ± 16%      -0.6        1.79 ± 16%  perf-profile.calltrace.cycles-pp.hrtimer_active.task_tick_fair.scheduler_tick.update_process_times.tick_sched_handle
      1.08 ± 22%      -0.6        0.53 ± 58%  perf-profile.calltrace.cycles-pp.memcpy_erms.drm_fb_helper_dirty_work.process_one_work.worker_thread.kthread
      1.24 ± 21%      -0.5        0.76 ± 16%  perf-profile.calltrace.cycles-pp.worker_thread.kthread.ret_from_fork
      1.23 ± 20%      -0.5        0.75 ± 15%  perf-profile.calltrace.cycles-pp.process_one_work.worker_thread.kthread.ret_from_fork
      1.17 ± 20%      -0.5        0.71 ± 16%  perf-profile.calltrace.cycles-pp.drm_fb_helper_dirty_work.process_one_work.worker_thread.kthread.ret_from_fork
      0.83 ±  8%      -0.3        0.49 ± 58%  perf-profile.calltrace.cycles-pp.read_tsc.ktime_get_update_offsets_now.hrtimer_interrupt.smp_apic_timer_interrupt.apic_timer_interrupt
      1.48 ± 12%      +0.3        1.79 ±  7%  perf-profile.calltrace.cycles-pp.fpregs_assert_state_consistent.prepare_exit_to_usermode.swapgs_restore_regs_and_return_to_usermode
      2.09 ±  5%      +0.4        2.47 ± 12%  perf-profile.calltrace.cycles-pp.native_write_msr.lapic_next_deadline.clockevents_program_event.hrtimer_interrupt.smp_apic_timer_interrupt
      2.68 ±  5%      +0.5        3.13 ± 10%  perf-profile.calltrace.cycles-pp.lapic_next_deadline.clockevents_program_event.hrtimer_interrupt.smp_apic_timer_interrupt.apic_timer_interrupt
     19.59 ±  3%      +1.8       21.36 ±  6%  perf-profile.calltrace.cycles-pp.swapgs_restore_regs_and_return_to_usermode
     15.36 ±  4%      +2.0       17.40 ±  6%  perf-profile.calltrace.cycles-pp.prepare_exit_to_usermode.swapgs_restore_regs_and_return_to_usermode
     50.31 ±  4%      -4.3       46.04 ±  2%  perf-profile.children.cycles-pp.hrtimer_interrupt
     28.46 ± 13%      -3.7       24.78 ±  4%  perf-profile.children.cycles-pp.update_process_times
     28.60 ± 13%      -3.6       24.96 ±  4%  perf-profile.children.cycles-pp.tick_sched_handle
     38.36 ±  8%      -3.6       34.78 ±  2%  perf-profile.children.cycles-pp.__hrtimer_run_queues
     61.73 ±  4%      -3.5       58.27        perf-profile.children.cycles-pp.smp_apic_timer_interrupt
     64.31 ±  4%      -3.4       60.93        perf-profile.children.cycles-pp.apic_timer_interrupt
      3.09 ± 34%      -1.2        1.88 ±  5%  perf-profile.children.cycles-pp.run_local_timers
      3.43 ± 14%      -0.7        2.75 ±  5%  perf-profile.children.cycles-pp.account_user_time
      2.42 ± 16%      -0.6        1.81 ± 17%  perf-profile.children.cycles-pp.hrtimer_active
      1.99 ± 14%      -0.6        1.38 ±  9%  perf-profile.children.cycles-pp.irq_work_run
      1.99 ± 14%      -0.6        1.38 ±  9%  perf-profile.children.cycles-pp.printk
      1.99 ± 14%      -0.6        1.38 ±  9%  perf-profile.children.cycles-pp.vprintk_emit
      2.04 ± 14%      -0.6        1.46 ±  7%  perf-profile.children.cycles-pp.irq_work_run_list
      1.95 ± 13%      -0.6        1.38 ±  9%  perf-profile.children.cycles-pp.irq_work_interrupt
      1.95 ± 13%      -0.6        1.38 ±  9%  perf-profile.children.cycles-pp.smp_irq_work_interrupt
      1.24 ± 21%      -0.5        0.76 ± 16%  perf-profile.children.cycles-pp.worker_thread
      1.23 ± 20%      -0.5        0.75 ± 15%  perf-profile.children.cycles-pp.process_one_work
      1.17 ± 20%      -0.5        0.71 ± 16%  perf-profile.children.cycles-pp.drm_fb_helper_dirty_work
      1.15 ± 17%      -0.4        0.70 ± 16%  perf-profile.children.cycles-pp.memcpy_erms
      0.28 ±  9%      -0.1        0.18 ± 23%  perf-profile.children.cycles-pp.native_queued_spin_lock_slowpath
      0.23 ± 27%      -0.1        0.15 ± 27%  perf-profile.children.cycles-pp.run_rebalance_domains
      0.04 ± 59%      +0.0        0.09 ± 17%  perf-profile.children.cycles-pp.new_sync_read
      0.04 ± 57%      +0.1        0.09 ± 30%  perf-profile.children.cycles-pp.force_qs_rnp
      0.03 ±100%      +0.1        0.10 ± 48%  perf-profile.children.cycles-pp.__do_munmap
      0.01 ±173%      +0.1        0.09 ± 26%  perf-profile.children.cycles-pp.select_idle_sibling
      0.10 ± 31%      +0.1        0.21 ± 19%  perf-profile.children.cycles-pp.__alloc_pages_nodemask
      0.45 ± 24%      +0.1        0.60 ± 14%  perf-profile.children.cycles-pp.tick_program_event
      0.24 ± 26%      +0.1        0.39 ± 27%  perf-profile.children.cycles-pp.load_elf_binary
      0.24 ± 26%      +0.2        0.40 ± 26%  perf-profile.children.cycles-pp.search_binary_handler
      0.41 ± 26%      +0.2        0.61 ± 16%  perf-profile.children.cycles-pp.__do_execve_file
      0.07 ± 58%      +0.2        0.26 ± 74%  perf-profile.children.cycles-pp.rcu_gp_kthread
      0.42 ± 28%      +0.2        0.61 ± 15%  perf-profile.children.cycles-pp.__x64_sys_execve
      0.41 ± 29%      +0.2        0.61 ± 18%  perf-profile.children.cycles-pp.execve
      0.23 ± 40%      +0.2        0.45 ± 21%  perf-profile.children.cycles-pp.read
      1.51 ± 11%      +0.3        1.81 ±  6%  perf-profile.children.cycles-pp.fpregs_assert_state_consistent
      0.14 ± 37%      +0.3        0.45 ± 45%  perf-profile.children.cycles-pp.schedule
      0.14 ± 37%      +0.3        0.46 ± 43%  perf-profile.children.cycles-pp.__schedule
      2.71 ±  5%      +0.4        3.15 ± 10%  perf-profile.children.cycles-pp.lapic_next_deadline
     19.62 ±  3%      +1.8       21.43 ±  6%  perf-profile.children.cycles-pp.swapgs_restore_regs_and_return_to_usermode
     15.54 ±  4%      +2.0       17.59 ±  7%  perf-profile.children.cycles-pp.prepare_exit_to_usermode
      3.08 ± 35%      -1.2        1.85 ±  6%  perf-profile.self.cycles-pp.run_local_timers
      2.40 ± 17%      -0.6        1.78 ± 18%  perf-profile.self.cycles-pp.hrtimer_active
      2.48 ± 10%      -0.6        1.87 ±  4%  perf-profile.self.cycles-pp.account_user_time
      0.33 ± 16%      -0.1        0.21 ± 30%  perf-profile.self.cycles-pp.delay_tsc
      0.28 ±  9%      -0.1        0.18 ± 23%  perf-profile.self.cycles-pp.native_queued_spin_lock_slowpath
      0.03 ±100%      +0.1        0.08 ± 36%  perf-profile.self.cycles-pp.clear_page_erms
      1.19 ±  9%      +0.1        1.32 ±  6%  perf-profile.self.cycles-pp.run_timer_softirq


                                                                                
                               vm-scalability.throughput                        
                                                                                
   5.2e+06 +-+--------------------------------------------------------------+   
           |                          +             +                       |   
  5.15e+06 +-+                        :             :                       |   
           |:                        : :           : :                      |   
   5.1e+06 +-+                       : :           : :                      |   
  5.05e+06 +-+                       : :           : :                      |   
           | :                      :   :         :   :                     |   
     5e+06 +-+                      :   :         :   :                     |   
           | :                      :   :         :   :                     |   
  4.95e+06 +-+:                    :     :       :     :                    |   
   4.9e+06 +-++..+..+..+..+..+..+..+     +..+.+..+     +..+..+..+..+..+..+..|   
           |                                                                |   
  4.85e+06 +-+O        O                                                    O   
           O     O  O     O  O  O  O  O  O  O O  O  O  O  O     O  O  O  O  |   
   4.8e+06 +-+-----------------------------------------------O--------------+   
                                                                                
                                                                                                                                                                
                                vm-scalability.median                           
                                                                                
  36000 +-+-----------------------------------------------------------------+   
        |                           +              +                        |   
  35800 +-+                        + :           .. :                       |   
        | :   +..  .+..+...       +   :   +..  .+    :   +...  .+..         |   
        | : ..   +.        +..  .+    : ..   +.      : ..    +.    +..+..   |   
  35600 +-++                  +.       +              +                  +..|   
        |                                                                   |   
  35400 +-+                                                                 |   
        |                                                                   |   
  35200 +-+         O                                                       O   
        |  O                        O        O     O                     O  |   
        |        O     O         O        O     O     O  O                  |   
  35000 +-+                O           O                        O  O  O     |   
        O     O               O                                             |   
  34800 +-+--------------------------------------------------O--------------+   
                                                                                
                                                                                
[*] 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


View attachment "config-5.4.0-rc2-00028-g7fa343b7fdc4f" of type "text/plain" (200620 bytes)

View attachment "job-script" of type "text/plain" (7922 bytes)

View attachment "job.yaml" of type "text/plain" (5402 bytes)

View attachment "reproduce" of type "text/plain" (1063 bytes)

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ