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>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20220527062116.GA11731@xsang-OptiPlex-9020>
Date:   Fri, 27 May 2022 14:21:16 +0800
From:   kernel test robot <oliver.sang@...el.com>
To:     Mel Gorman <mgorman@...hsingularity.net>
Cc:     0day robot <lkp@...el.com>, LKML <linux-kernel@...r.kernel.org>,
        lkp@...ts.01.org, ying.huang@...el.com, feng.tang@...el.com,
        zhengjun.xing@...ux.intel.com, fengwei.yin@...el.com,
        Peter Zijlstra <peterz@...radead.org>,
        Ingo Molnar <mingo@...nel.org>,
        Vincent Guittot <vincent.guittot@...aro.org>,
        Valentin Schneider <valentin.schneider@....com>,
        K Prateek Nayak <kprateek.nayak@....com>,
        Aubrey Li <aubrey.li@...ux.intel.com>,
        Mel Gorman <mgorman@...hsingularity.net>, yu.c.chen@...el.com
Subject: [sched/numa]  5278ba412f:  unixbench.score -2.9% regression



Greeting,

FYI, we noticed a -2.9% regression of unixbench.score due to commit:


commit: 5278ba412faff8402e318ad20ab762cc9ba7a801 ("[PATCH 3/4] sched/numa: Apply imbalance limitations consistently")
url: https://github.com/intel-lab-lkp/linux/commits/Mel-Gorman/Mitigate-inconsistent-NUMA-imbalance-behaviour/20220520-183837
base: https://git.kernel.org/cgit/linux/kernel/git/tip/tip.git 991d8d8142cad94f9c5c05db25e67fa83d6f772a
patch link: https://lore.kernel.org/lkml/20220520103519.1863-4-mgorman@techsingularity.net

in testcase: unixbench
on test machine: 128 threads 2 sockets Intel(R) Xeon(R) Gold 6338 CPU @ 2.00GHz with 256G memory
with following parameters:

	runtime: 300s
	nr_task: 1
	test: shell1
	cpufreq_governor: performance
	ucode: 0xd000331

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

In addition to that, the commit also has significant impact on the following tests:

+------------------+---------------------------------------------------------------------------------+
| testcase: change | unixbench: unixbench.score -11.1% regression                                    |
| test machine     | 128 threads 2 sockets Intel(R) Xeon(R) Gold 6338 CPU @ 2.00GHz with 256G memory |
| test parameters  | cpufreq_governor=performance                                                    |
|                  | nr_task=1                                                                       |
|                  | runtime=300s                                                                    |
|                  | test=shell8                                                                     |
|                  | ucode=0xd000331                                                                 |
+------------------+---------------------------------------------------------------------------------+


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
        sudo bin/lkp install job.yaml           # job file is attached in this email
        bin/lkp split-job --compatible job.yaml # generate the yaml file for lkp run
        sudo bin/lkp run generated-yaml-file

        # if come across any failure that blocks the test,
        # please remove ~/.lkp and /lkp dir to run from a clean state.

=========================================================================================
compiler/cpufreq_governor/kconfig/nr_task/rootfs/runtime/tbox_group/test/testcase/ucode:
  gcc-11/performance/x86_64-rhel-8.3/1/debian-10.4-x86_64-20200603.cgz/300s/lkp-icl-2sp2/shell1/unixbench/0xd000331

commit: 
  626db23ac9 ("sched/numa: Do not swap tasks between nodes when spare capacity is available")
  5278ba412f ("sched/numa: Apply imbalance limitations consistently")

626db23ac968c13b 5278ba412faff8402e318ad20ab 
---------------- --------------------------- 
         %stddev     %change         %stddev
             \          |                \  
      2705            -2.9%       2625        unixbench.score
     71105 ±  5%     -13.4%      61547 ±  4%  unixbench.time.involuntary_context_switches
    287.33 ± 10%    +312.2%       1184 ±  6%  unixbench.time.major_page_faults
 1.232e+08            -2.9%  1.196e+08        unixbench.time.minor_page_faults
    143.00            +1.4%     145.00        unixbench.time.percent_of_cpu_this_job_got
    383.13            +7.7%     412.62        unixbench.time.system_time
    521.90            -3.4%     504.36        unixbench.time.user_time
   3400459            -3.6%    3278744        unixbench.time.voluntary_context_switches
   7226950            -2.9%    7015844        unixbench.workload
     30.49            +1.5%      30.96        turbostat.RAMWatt
     16050            -6.6%      14986        vmstat.system.cs
   1293860 ±  6%   +4158.8%   55103032 ± 60%  numa-numastat.node0.local_node
   1344815 ±  5%   +3999.6%   55131871 ± 60%  numa-numastat.node0.numa_hit
  89313120           -63.9%   32257305 ±103%  numa-numastat.node1.local_node
  89365596           -63.8%   32337261 ±102%  numa-numastat.node1.numa_hit
     66387           +30.1%      86365 ±  6%  meminfo.Active
     66201           +30.2%      86181 ±  7%  meminfo.Active(anon)
  12721152 ± 10%     -27.2%    9261056 ±  7%  meminfo.DirectMap2M
     48790           +10.7%      53987 ±  2%  meminfo.Mapped
     92524           +25.1%     115764 ±  5%  meminfo.Shmem
     18726 ±  8%     -15.7%      15785 ±  8%  sched_debug.cfs_rq:/.min_vruntime.stddev
     87.63 ±  5%     -11.0%      77.99 ±  4%  sched_debug.cfs_rq:/.runnable_avg.avg
    618.08 ±  2%     +21.9%     753.36 ± 11%  sched_debug.cfs_rq:/.runnable_avg.max
     10444 ± 28%    -376.2%     -28845        sched_debug.cfs_rq:/.spread0.avg
     60208 ±  4%     -77.6%      13496 ± 74%  sched_debug.cfs_rq:/.spread0.max
     -8539          +479.7%     -49504        sched_debug.cfs_rq:/.spread0.min
     18727 ±  8%     -15.7%      15785 ±  8%  sched_debug.cfs_rq:/.spread0.stddev
     87.61 ±  5%     -11.0%      77.96 ±  4%  sched_debug.cfs_rq:/.util_avg.avg
    617.98 ±  2%     +21.9%     753.33 ± 11%  sched_debug.cfs_rq:/.util_avg.max
      0.00 ± 12%     -24.5%       0.00 ± 23%  sched_debug.cpu.next_balance.stddev
    161071 ±  4%     -22.7%     124508 ± 10%  sched_debug.cpu.nr_switches.max
     44072 ±  6%     -25.6%      32790 ±  3%  sched_debug.cpu.nr_switches.stddev
    184614 ±  6%     -58.9%      75886 ± 28%  numa-meminfo.node0.AnonHugePages
    227436 ±  5%     -50.8%     111893 ± 18%  numa-meminfo.node0.AnonPages
    250375 ±  6%     -39.2%     152234 ± 18%  numa-meminfo.node0.AnonPages.max
    239538 ±  4%     -50.1%     119562 ± 18%  numa-meminfo.node0.Inactive
    239385 ±  4%     -50.1%     119488 ± 18%  numa-meminfo.node0.Inactive(anon)
     17347 ± 20%     -33.1%      11605 ± 35%  numa-meminfo.node0.Shmem
     61415 ±  2%     +35.5%      83195 ±  9%  numa-meminfo.node1.Active
     61415 ±  2%     +35.3%      83107 ±  9%  numa-meminfo.node1.Active(anon)
     15111 ± 77%    +724.0%     124518 ± 17%  numa-meminfo.node1.AnonHugePages
     48170 ± 26%    +252.5%     169822 ± 12%  numa-meminfo.node1.AnonPages
    977352           +19.6%    1168644 ±  5%  numa-meminfo.node1.AnonPages.max
     62321 ± 14%    +205.7%     190528 ± 11%  numa-meminfo.node1.Inactive
     62321 ± 14%    +205.6%     190446 ± 11%  numa-meminfo.node1.Inactive(anon)
     75390 ±  4%     +38.5%     104414 ±  8%  numa-meminfo.node1.Shmem
     56856 ±  5%     -50.8%      27966 ± 18%  numa-vmstat.node0.nr_anon_pages
     59844 ±  4%     -50.1%      29863 ± 18%  numa-vmstat.node0.nr_inactive_anon
      4336 ± 20%     -33.3%       2894 ± 35%  numa-vmstat.node0.nr_shmem
     59844 ±  4%     -50.1%      29862 ± 18%  numa-vmstat.node0.nr_zone_inactive_anon
   1344749 ±  5%   +3999.7%   55131256 ± 60%  numa-vmstat.node0.numa_hit
   1293794 ±  6%   +4159.0%   55102416 ± 60%  numa-vmstat.node0.numa_local
     15374 ±  2%     +35.3%      20797 ±  9%  numa-vmstat.node1.nr_active_anon
     11720 ± 27%    +258.8%      42056 ± 12%  numa-vmstat.node1.nr_anon_pages
     15192 ± 14%    +210.3%      47137 ± 11%  numa-vmstat.node1.nr_inactive_anon
     18800 ±  4%     +38.6%      26056 ±  8%  numa-vmstat.node1.nr_shmem
     15374 ±  2%     +35.3%      20797 ±  9%  numa-vmstat.node1.nr_zone_active_anon
     15192 ± 14%    +210.3%      47137 ± 11%  numa-vmstat.node1.nr_zone_inactive_anon
  89364368           -63.8%   32336320 ±102%  numa-vmstat.node1.numa_hit
  89311892           -63.9%   32256364 ±103%  numa-vmstat.node1.numa_local
     16551           +30.1%      21527 ±  7%  proc-vmstat.nr_active_anon
     68924            +2.2%      70416        proc-vmstat.nr_anon_pages
     75477            +2.7%      77526        proc-vmstat.nr_inactive_anon
     12470           +10.3%      13749 ±  2%  proc-vmstat.nr_mapped
     23215           +25.1%      29034 ±  5%  proc-vmstat.nr_shmem
     16551           +30.1%      21527 ±  7%  proc-vmstat.nr_zone_active_anon
     75477            +2.7%      77526        proc-vmstat.nr_zone_inactive_anon
    242786           +41.2%     342858 ± 16%  proc-vmstat.numa_hint_faults
    241277           +41.0%     340196 ± 16%  proc-vmstat.numa_hint_faults_local
  90713416            -3.6%   87455312        proc-vmstat.numa_hit
  90609984            -3.6%   87346517        proc-vmstat.numa_local
      3339 ± 58%    +412.8%      17123 ± 58%  proc-vmstat.numa_pages_migrated
    343034 ±  3%     +49.8%     513708 ± 14%  proc-vmstat.numa_pte_updates
     86449           +86.6%     161358 ±  2%  proc-vmstat.pgactivate
  90707968            -3.6%   87448830        proc-vmstat.pgalloc_normal
 1.257e+08            -2.7%  1.223e+08        proc-vmstat.pgfault
  90721230            -3.6%   87454560        proc-vmstat.pgfree
      3339 ± 58%    +412.8%      17123 ± 58%  proc-vmstat.pgmigrate_success
   6816403            -2.2%    6668991        proc-vmstat.pgreuse
      3800            -3.2%       3677        proc-vmstat.thp_fault_alloc
   1597074            -2.9%    1550242        proc-vmstat.unevictable_pgs_culled
      0.84 ±  7%      -0.2        0.69 ± 10%  perf-profile.calltrace.cycles-pp.ret_from_fork
      0.81 ±  7%      -0.1        0.66 ± 10%  perf-profile.calltrace.cycles-pp.kthread.ret_from_fork
      0.54 ±  7%      +0.1        0.66 ± 12%  perf-profile.calltrace.cycles-pp.next_uptodate_page.filemap_map_pages.do_read_fault.do_fault.__handle_mm_fault
      0.79 ±  7%      +0.1        0.92 ±  7%  perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.__open64_nocancel.setlocale
      0.78 ±  7%      +0.1        0.91 ±  7%  perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.__open64_nocancel.setlocale
      0.76 ±  8%      +0.1        0.88 ±  7%  perf-profile.calltrace.cycles-pp.do_sys_openat2.__x64_sys_openat.do_syscall_64.entry_SYSCALL_64_after_hwframe.__open64_nocancel
      0.76 ±  8%      +0.1        0.90 ±  8%  perf-profile.calltrace.cycles-pp.__x64_sys_openat.do_syscall_64.entry_SYSCALL_64_after_hwframe.__open64_nocancel.setlocale
      0.83 ±  7%      +0.1        0.96 ±  7%  perf-profile.calltrace.cycles-pp.__open64_nocancel.setlocale
      0.81 ±  7%      -0.1        0.66 ± 10%  perf-profile.children.cycles-pp.kthread
      1.10 ±  6%      -0.1        0.97 ±  8%  perf-profile.children.cycles-pp.ret_from_fork
      0.31 ± 10%      -0.1        0.25 ±  7%  perf-profile.children.cycles-pp.newidle_balance
      0.04 ± 45%      +0.0        0.08 ± 20%  perf-profile.children.cycles-pp.folio_add_lru
      0.08 ± 20%      +0.0        0.12 ± 12%  perf-profile.children.cycles-pp.touch_atime
      0.05 ± 47%      +0.0        0.08 ± 23%  perf-profile.children.cycles-pp.apparmor_file_free_security
      0.04 ± 71%      +0.0        0.07 ± 11%  perf-profile.children.cycles-pp.perf_output_copy
      0.29 ±  3%      +0.0        0.33 ±  8%  perf-profile.children.cycles-pp.page_counter_charge
      0.17 ±  9%      +0.0        0.21 ±  9%  perf-profile.children.cycles-pp.__anon_vma_prepare
      0.14 ±  8%      +0.0        0.18 ± 16%  perf-profile.children.cycles-pp.apparmor_file_alloc_security
      0.46 ±  4%      +0.0        0.51 ±  7%  perf-profile.children.cycles-pp.vfs_read
      0.03 ±100%      +0.1        0.08 ± 19%  perf-profile.children.cycles-pp.propagate_protected_usage
      0.20 ± 15%      +0.1        0.25 ±  4%  perf-profile.children.cycles-pp.copy_string_kernel
      0.00            +0.1        0.06 ± 13%  perf-profile.children.cycles-pp.__mark_inode_dirty
      0.27 ±  5%      +0.1        0.33 ±  6%  perf-profile.children.cycles-pp.vma_interval_tree_remove
      0.20 ±  6%      +0.1        0.26 ±  5%  perf-profile.children.cycles-pp.up_write
      0.00            +0.1        0.06 ±  6%  perf-profile.children.cycles-pp.rmqueue_bulk
      0.36 ±  7%      +0.1        0.43 ±  8%  perf-profile.children.cycles-pp.get_page_from_freelist
      0.34 ±  7%      +0.1        0.41 ±  3%  perf-profile.children.cycles-pp.unlink_file_vma
      0.31 ±  8%      +0.1        0.40 ± 10%  perf-profile.children.cycles-pp.__slab_free
      0.34 ± 10%      +0.1        0.44 ±  7%  perf-profile.children.cycles-pp.ksys_write
      0.72 ±  3%      +0.1        0.82 ±  5%  perf-profile.children.cycles-pp.__split_vma
      0.33 ± 10%      +0.1        0.44 ±  8%  perf-profile.children.cycles-pp.vfs_write
      0.32 ± 10%      +0.1        0.42 ±  9%  perf-profile.children.cycles-pp.new_sync_write
      1.05 ±  4%      +0.2        1.29 ±  8%  perf-profile.children.cycles-pp.next_uptodate_page
      1.41 ±  4%      -0.3        1.10 ± 23%  perf-profile.self.cycles-pp.menu_select
      0.09 ± 10%      -0.0        0.06 ± 11%  perf-profile.self.cycles-pp.vm_normal_page
      0.02 ±141%      +0.1        0.07 ± 21%  perf-profile.self.cycles-pp.propagate_protected_usage
      0.26 ±  5%      +0.1        0.32 ±  6%  perf-profile.self.cycles-pp.vma_interval_tree_remove
      0.20 ±  8%      +0.1        0.26 ±  3%  perf-profile.self.cycles-pp.up_write
      0.52 ±  6%      +0.1        0.59 ±  8%  perf-profile.self.cycles-pp.vma_interval_tree_insert
      0.30 ± 10%      +0.1        0.38 ± 10%  perf-profile.self.cycles-pp.__slab_free
      1.00 ±  4%      +0.2        1.24 ±  8%  perf-profile.self.cycles-pp.next_uptodate_page
      9.50            +3.3%       9.82        perf-stat.i.MPKI
 1.467e+09            -3.1%  1.421e+09        perf-stat.i.branch-instructions
  25259322            -2.8%   24564546        perf-stat.i.branch-misses
      4.64            +3.8        8.40        perf-stat.i.cache-miss-rate%
   3128761           +84.4%    5768597        perf-stat.i.cache-misses
     16064            -6.6%      15000        perf-stat.i.context-switches
      1.28 ±  2%      +4.9%       1.35 ±  2%  perf-stat.i.cpi
    147.24          +456.0%     818.73        perf-stat.i.cpu-migrations
      3222 ±  3%     -37.4%       2017 ±  2%  perf-stat.i.cycles-between-cache-misses
      0.03            +0.0        0.04        perf-stat.i.dTLB-load-miss-rate%
    646123 ±  2%     +10.8%     716013        perf-stat.i.dTLB-load-misses
 1.866e+09            -2.7%  1.816e+09        perf-stat.i.dTLB-loads
    988430            -1.7%     971311        perf-stat.i.dTLB-store-misses
 1.078e+09            -2.4%  1.052e+09        perf-stat.i.dTLB-stores
 7.095e+09            -3.1%  6.873e+09        perf-stat.i.instructions
      0.89 ±  6%    +162.7%       2.35 ±  5%  perf-stat.i.major-faults
     34.46            -2.8%      33.52        perf-stat.i.metric.M/sec
    195370            -2.7%     190058        perf-stat.i.minor-faults
     85.14            +3.7       88.89        perf-stat.i.node-load-miss-rate%
    504104 ±  2%    +121.9%    1118698        perf-stat.i.node-load-misses
    104059 ±  3%     +44.8%     150652 ±  4%  perf-stat.i.node-loads
      5.36 ± 18%     +19.3       24.63 ±  4%  perf-stat.i.node-store-miss-rate%
     45098 ± 16%    +733.8%     376029 ±  4%  perf-stat.i.node-store-misses
    937285           +26.6%    1186448 ±  2%  perf-stat.i.node-stores
    195371            -2.7%     190061        perf-stat.i.page-faults
      9.64            +2.9%       9.91        perf-stat.overall.MPKI
      4.59            +3.9        8.47        perf-stat.overall.cache-miss-rate%
      1.20 ±  2%      +4.6%       1.25 ±  2%  perf-stat.overall.cpi
      2709 ±  2%     -44.9%       1493 ±  2%  perf-stat.overall.cycles-between-cache-misses
      0.03 ±  2%      +0.0        0.04        perf-stat.overall.dTLB-load-miss-rate%
     82.64            +5.4       88.05        perf-stat.overall.node-load-miss-rate%
      4.60 ± 14%     +19.4       24.05 ±  4%  perf-stat.overall.node-store-miss-rate%
 1.465e+09            -3.1%   1.42e+09        perf-stat.ps.branch-instructions
  25223900            -2.7%   24532257        perf-stat.ps.branch-misses
   3133964           +84.0%    5765238        perf-stat.ps.cache-misses
     16037            -6.6%      14976        perf-stat.ps.context-switches
    146.96          +455.6%     816.47        perf-stat.ps.cpu-migrations
    645888 ±  2%     +10.8%     715648        perf-stat.ps.dTLB-load-misses
 1.864e+09            -2.7%  1.814e+09        perf-stat.ps.dTLB-loads
    987054            -1.7%     970089        perf-stat.ps.dTLB-store-misses
 1.077e+09            -2.4%  1.051e+09        perf-stat.ps.dTLB-stores
 7.086e+09            -3.1%  6.866e+09        perf-stat.ps.instructions
      0.89 ±  6%    +162.4%       2.34 ±  5%  perf-stat.ps.major-faults
    195074            -2.7%     189793        perf-stat.ps.minor-faults
    503993          +121.6%    1117027        perf-stat.ps.node-load-misses
    105826 ±  3%     +43.3%     151624 ±  4%  perf-stat.ps.node-loads
     45328 ± 16%    +728.3%     375435 ±  4%  perf-stat.ps.node-store-misses
    937188           +26.5%    1185959 ±  2%  perf-stat.ps.node-stores
    195075            -2.7%     189796        perf-stat.ps.page-faults
 4.472e+12            -3.1%  4.335e+12        perf-stat.total.instructions


***************************************************************************************************
lkp-icl-2sp2: 128 threads 2 sockets Intel(R) Xeon(R) Gold 6338 CPU @ 2.00GHz with 256G memory
=========================================================================================
compiler/cpufreq_governor/kconfig/nr_task/rootfs/runtime/tbox_group/test/testcase/ucode:
  gcc-11/performance/x86_64-rhel-8.3/1/debian-10.4-x86_64-20200603.cgz/300s/lkp-icl-2sp2/shell8/unixbench/0xd000331

commit: 
  626db23ac9 ("sched/numa: Do not swap tasks between nodes when spare capacity is available")
  5278ba412f ("sched/numa: Apply imbalance limitations consistently")

626db23ac968c13b 5278ba412faff8402e318ad20ab 
---------------- --------------------------- 
         %stddev     %change         %stddev
             \          |                \  
      9459           -11.1%       8410        unixbench.score
     49295 ±  2%     -13.0%      42902        unixbench.time.involuntary_context_switches
      1616 ±  3%    +147.0%       3993 ±  4%  unixbench.time.major_page_faults
  45966048           -11.0%   40896241        unixbench.time.minor_page_faults
    173.46           +18.9%     206.17        unixbench.time.system_time
    190.29           -16.7%     158.42        unixbench.time.user_time
   1307674           -12.3%    1146302        unixbench.time.voluntary_context_switches
    358509           -11.3%     317922        unixbench.workload
     31.04            +6.0%      32.92        turbostat.RAMWatt
     53649           -10.2%      48196        vmstat.system.cs
      0.14 ±  3%      +0.0        0.16        mpstat.cpu.all.soft%
      1.63            -0.2        1.45        mpstat.cpu.all.usr%
      0.34 ±223%      +2.3        2.67 ± 83%  perf-profile.calltrace.cycles-pp.do_execveat_common.__x64_sys_execve.do_syscall_64.entry_SYSCALL_64_after_hwframe
      0.34 ±223%      +2.4        2.71 ± 81%  perf-profile.calltrace.cycles-pp.__x64_sys_execve.do_syscall_64.entry_SYSCALL_64_after_hwframe
  12953258 ±  7%     -29.1%    9177429 ±  7%  meminfo.DirectMap2M
     22474           +13.6%      25533 ±  2%  meminfo.KernelStack
      7838          +156.0%      20065 ± 20%  meminfo.PageTables
     62596 ± 16%     -39.1%      38109 ± 36%  numa-vmstat.node0.nr_inactive_anon
     12227 ±  6%     +11.5%      13631 ±  4%  numa-vmstat.node0.nr_kernel_stack
      1163 ± 18%    +134.4%       2726 ± 18%  numa-vmstat.node0.nr_page_table_pages
     62596 ± 16%     -39.1%      38109 ± 36%  numa-vmstat.node0.nr_zone_inactive_anon
     14241 ± 67%    +184.6%      40525 ± 36%  numa-vmstat.node1.nr_anon_pages
     14899 ± 69%    +182.7%      42126 ± 33%  numa-vmstat.node1.nr_inactive_anon
     10252 ±  7%     +16.7%      11959 ±  6%  numa-vmstat.node1.nr_kernel_stack
    809.17 ± 25%    +202.7%       2449 ± 21%  numa-vmstat.node1.nr_page_table_pages
     14899 ± 69%    +182.7%      42126 ± 33%  numa-vmstat.node1.nr_zone_inactive_anon
    244098 ± 15%     -39.8%     146934 ± 40%  numa-meminfo.node0.AnonPages.max
    250449 ± 16%     -39.3%     152036 ± 37%  numa-meminfo.node0.Inactive
    250301 ± 16%     -39.3%     151965 ± 37%  numa-meminfo.node0.Inactive(anon)
      4630 ± 18%    +127.3%      10522 ± 23%  numa-meminfo.node0.PageTables
     56826 ± 67%    +184.7%     161757 ± 35%  numa-meminfo.node1.AnonPages
     64784 ± 57%    +166.9%     172884 ± 33%  numa-meminfo.node1.AnonPages.max
     59459 ± 70%    +183.1%     168330 ± 33%  numa-meminfo.node1.Inactive
     59459 ± 70%    +182.8%     168177 ± 33%  numa-meminfo.node1.Inactive(anon)
     10234 ±  7%     +16.8%      11949 ±  6%  numa-meminfo.node1.KernelStack
      3189 ± 24%    +203.1%       9665 ± 22%  numa-meminfo.node1.PageTables
     73176            +3.2%      75515        proc-vmstat.nr_anon_pages
     77439            +3.4%      80074        proc-vmstat.nr_inactive_anon
     22484           +13.6%      25550 ±  2%  proc-vmstat.nr_kernel_stack
      1964          +159.7%       5101 ± 19%  proc-vmstat.nr_page_table_pages
     77439            +3.4%      80074        proc-vmstat.nr_zone_inactive_anon
  33321568           -11.1%   29636150        proc-vmstat.numa_hit
  33204884           -11.1%   29522839        proc-vmstat.numa_local
      2826 ±  7%    +450.7%      15565        proc-vmstat.pgactivate
  33314916           -11.1%   29632022        proc-vmstat.pgalloc_normal
  46272301           -10.9%   41208227        proc-vmstat.pgfault
  33127507           -11.1%   29444660        proc-vmstat.pgfree
   2594250           -11.2%    2304883        proc-vmstat.pgreuse
      1499           -10.6%       1340        proc-vmstat.thp_fault_alloc
    635894           -11.1%     565409        proc-vmstat.unevictable_pgs_culled
     10.69            +4.3%      11.15        perf-stat.i.MPKI
 4.972e+09           -10.5%  4.448e+09        perf-stat.i.branch-instructions
      1.78            +0.0        1.81        perf-stat.i.branch-miss-rate%
  87366205            -9.0%   79542506        perf-stat.i.branch-misses
      4.39           +10.2       14.62        perf-stat.i.cache-miss-rate%
  10832039          +238.1%   36622848        perf-stat.i.cache-misses
 2.643e+08            -6.7%  2.466e+08        perf-stat.i.cache-references
     55032           -10.2%      49392        perf-stat.i.context-switches
      0.84           +11.6%       0.94        perf-stat.i.cpi
      1083 ±  4%     +90.1%       2058        perf-stat.i.cpu-migrations
      1976           -61.1%     769.37 ±  3%  perf-stat.i.cycles-between-cache-misses
      0.04 ±  4%      +0.0        0.04        perf-stat.i.dTLB-load-miss-rate%
 6.279e+09           -10.2%  5.638e+09        perf-stat.i.dTLB-loads
   3629603           -10.3%    3257478        perf-stat.i.dTLB-store-misses
 3.662e+09           -10.2%  3.289e+09        perf-stat.i.dTLB-stores
 2.408e+10           -10.5%  2.154e+10        perf-stat.i.instructions
      1.19           -10.6%       1.06        perf-stat.i.ipc
     25.60 ±  3%    +145.3%      62.80 ±  4%  perf-stat.i.major-faults
    102.69           +76.3%     181.04        perf-stat.i.metric.K/sec
    118.55           -10.3%     106.39        perf-stat.i.metric.M/sec
    710240           -11.0%     632294        perf-stat.i.minor-faults
     67.54           +24.9       92.47        perf-stat.i.node-load-miss-rate%
    823765 ±  2%    +788.1%    7315469        perf-stat.i.node-load-misses
    399301 ±  3%     +31.5%     525162        perf-stat.i.node-loads
      5.32 ±  6%     +34.6       39.92        perf-stat.i.node-store-miss-rate%
    193239 ±  6%   +1687.7%    3454580        perf-stat.i.node-store-misses
   4318690           +18.6%    5123428        perf-stat.i.node-stores
    710266           -11.0%     632357        perf-stat.i.page-faults
     10.98            +4.2%      11.44        perf-stat.overall.MPKI
      1.76            +0.0        1.79        perf-stat.overall.branch-miss-rate%
      4.10           +10.8       14.85        perf-stat.overall.cache-miss-rate%
      0.83           +12.2%       0.94        perf-stat.overall.cpi
      1853           -70.3%     550.23        perf-stat.overall.cycles-between-cache-misses
      0.04 ±  4%      +0.0        0.05        perf-stat.overall.dTLB-load-miss-rate%
      1.20           -10.9%       1.07        perf-stat.overall.ipc
     67.35           +25.9       93.30        perf-stat.overall.node-load-miss-rate%
      4.28 ±  6%     +36.0       40.27        perf-stat.overall.node-store-miss-rate%
 4.894e+09           -10.5%  4.379e+09        perf-stat.ps.branch-instructions
  86000604            -9.0%   78300554        perf-stat.ps.branch-misses
  10662516          +238.1%   36050907        perf-stat.ps.cache-misses
 2.602e+08            -6.7%  2.427e+08        perf-stat.ps.cache-references
     54171           -10.2%      48620        perf-stat.ps.context-switches
      1066 ±  4%     +90.1%       2026        perf-stat.ps.cpu-migrations
 6.181e+09           -10.2%   5.55e+09        perf-stat.ps.dTLB-loads
   3572776           -10.2%    3206610        perf-stat.ps.dTLB-store-misses
 3.604e+09           -10.2%  3.237e+09        perf-stat.ps.dTLB-stores
  2.37e+10           -10.5%  2.121e+10        perf-stat.ps.instructions
     25.20 ±  3%    +145.3%      61.82 ±  4%  perf-stat.ps.major-faults
    699120           -11.0%     622421        perf-stat.ps.minor-faults
    810873 ±  2%    +788.1%    7201233        perf-stat.ps.node-load-misses
    393054 ±  3%     +31.5%     516951        perf-stat.ps.node-loads
    190213 ±  6%   +1687.8%    3400631        perf-stat.ps.node-store-misses
   4251087           +18.6%    5043399        perf-stat.ps.node-stores
    699145           -11.0%     622482        perf-stat.ps.page-faults
 1.523e+12           -10.5%  1.364e+12        perf-stat.total.instructions





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.


-- 
0-DAY CI Kernel Test Service
https://01.org/lkp



View attachment "config-5.18.0-rc5-00025-g5278ba412faf" of type "text/plain" (162678 bytes)

View attachment "job-script" of type "text/plain" (8036 bytes)

View attachment "job.yaml" of type "text/plain" (5379 bytes)

View attachment "reproduce" of type "text/plain" (278 bytes)

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ