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] [day] [month] [year] [list]
Message-ID: <20201217094911.GA32379@xsang-OptiPlex-9020>
Date:   Thu, 17 Dec 2020 17:49:11 +0800
From:   kernel test robot <oliver.sang@...el.com>
To:     Sven Schnelle <svens@...ux.ibm.com>
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@...el.com, Thomas Gleixner <tglx@...utronix.de>,
        hca@...ux.ibm.com, Sven Schnelle <svens@...ux.ibm.com>
Subject: [entry]  a358d5636c:  unixbench.score -1.9% regression


Greeting,

FYI, we noticed a -1.9% regression of unixbench.score due to commit:


commit: a358d5636cfcb8b278c766ac261ad99143215041 ("[PATCH] entry: split lockdep and syscall work functions")
url: https://github.com/0day-ci/linux/commits/Sven-Schnelle/entry-split-lockdep-and-syscall-work-functions/20201201-163931
base: https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git 09162bc32c880a791c6c0668ce0745cf7958f576

in testcase: will-it-scale
on test machine: 192 threads Intel(R) Xeon(R) Platinum 9242 CPU @ 2.30GHz with 192G memory
with following parameters:

	nr_task: 16
	mode: process
	test: futex3
	cpufreq_governor: performance
	ucode: 0x5003003

test-description: Will It Scale takes a testcase and runs it from 1 through to n parallel copies to see if the testcase will scale. It builds both a process and threads based test in order to see any differences between the two.
test-url: https://github.com/antonblanchard/will-it-scale

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


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/mode/nr_task/rootfs/tbox_group/test/testcase/ucode:
  gcc-9/performance/x86_64-rhel-8.3/process/16/debian-10.4-x86_64-20200603.cgz/lkp-csl-2ap2/futex3/will-it-scale/0x5003003

commit: 
  v5.10-rc4
  a358d5636c ("entry: split lockdep and syscall work functions")

       v5.10-rc4 a358d5636cfcb8b278c766ac261 
---------------- --------------------------- 
       fail:runs  %reproduction    fail:runs
           |             |             |    
           :7          129%           9:9     stderr.Events_disabled
           :7          129%           9:9     stderr.Events_enabled
           :7          129%           9:9     stderr.[perf_record:Captured_and_wrote#MB/tmp/lkp/perf-sched.data(#samples)]
           :7          129%           9:9     stderr.[perf_record:Woken_up#times_to_write_data]
           :7          129%           9:9     stderr.has_stderr
         %stddev     %change         %stddev
             \          |                \  
      5516 ±  4%     -10.3%       4947 ±  4%  boot-time.idle
    291515 ± 68%     -41.2%     171377 ±  4%  cpuidle.POLL.time
     22225 ± 62%    +280.3%      84528 ±  6%  numa-numastat.node1.other_node
      4076 ±101%    -100.0%       0.00        softirqs.CPU14.NET_RX
     10586           +14.2%      12085        slabinfo.kmalloc-2k.active_objs
     10704 ±  2%     +14.5%      12252        slabinfo.kmalloc-2k.num_objs
      4098 ± 15%    +203.3%      12431 ±  5%  numa-vmstat.node3.nr_active_anon
     76843 ±  6%     +15.0%      88374 ±  5%  numa-vmstat.node3.nr_file_pages
      4679 ± 23%    +183.0%      13240 ±  6%  numa-vmstat.node3.nr_shmem
      4098 ± 15%    +203.3%      12431 ±  5%  numa-vmstat.node3.nr_zone_active_anon
     16418 ± 15%    +202.6%      49674 ±  5%  numa-meminfo.node3.Active
     16418 ± 15%    +202.6%      49674 ±  5%  numa-meminfo.node3.Active(anon)
    307349 ±  6%     +15.0%     353508 ±  5%  numa-meminfo.node3.FilePages
    697704 ± 20%     +32.9%     927015 ± 13%  numa-meminfo.node3.MemUsed
     18690 ± 23%    +183.4%      52973 ±  6%  numa-meminfo.node3.Shmem
     19935 ±  8%    +169.7%      53758 ±  3%  meminfo.Active
     19935 ±  8%    +169.7%      53758 ±  3%  meminfo.Active(anon)
    470080            +9.8%     516343        meminfo.Committed_AS
   2798824           +15.1%    3220140        meminfo.Memused
     31425 ±  5%    +109.3%      65771 ±  3%  meminfo.Shmem
     11733           +27.0%      14898        meminfo.max_used_kB
      4977 ±  8%    +169.7%      13426 ±  3%  proc-vmstat.nr_active_anon
     67216            +4.0%      69896        proc-vmstat.nr_anon_pages
    293542            +5.6%     310056        proc-vmstat.nr_file_pages
     69999            +3.9%      72758        proc-vmstat.nr_inactive_anon
      7851 ±  5%    +109.3%      16436 ±  3%  proc-vmstat.nr_shmem
     69914            +2.5%      71683        proc-vmstat.nr_slab_unreclaimable
    285726            +2.8%     293640        proc-vmstat.nr_unevictable
      4977 ±  8%    +169.7%      13426 ±  3%  proc-vmstat.nr_zone_active_anon
     69999            +3.9%      72758        proc-vmstat.nr_zone_inactive_anon
    285726            +2.8%     293640        proc-vmstat.nr_zone_unevictable
    989357           +25.2%    1238643        proc-vmstat.numa_hit
    896081            +9.2%     978823        proc-vmstat.numa_local
     93275          +178.6%     259820        proc-vmstat.numa_other
      5379 ± 15%    +229.0%      17697 ±  4%  proc-vmstat.pgactivate
   1060503           +24.6%    1321886        proc-vmstat.pgalloc_normal
   1160918            +2.7%    1192221        proc-vmstat.pgfault
   1177873           +20.8%    1422376        proc-vmstat.pgfree
 9.075e+09 ± 36%     +25.9%  1.142e+10        perf-stat.i.branch-instructions
   5471478 ± 11%     -20.5%    4349395 ±  6%  perf-stat.i.cache-misses
      1.22 ± 55%     -36.0%       0.78        perf-stat.i.cpi
    198.19            +1.8%     201.74        perf-stat.i.cpu-migrations
     10702 ± 12%     +24.4%      13314 ±  5%  perf-stat.i.cycles-between-cache-misses
 1.488e+10 ± 36%     +24.5%  1.853e+10        perf-stat.i.dTLB-loads
 1.124e+10 ± 36%     +25.2%  1.408e+10        perf-stat.i.dTLB-stores
      1.03 ± 36%     +25.3%       1.29        perf-stat.i.ipc
      0.73 ± 12%     +95.6%       1.44 ± 12%  perf-stat.i.major-faults
    183.61 ± 36%     +25.0%     229.60        perf-stat.i.metric.M/sec
      3733            +2.4%       3822        perf-stat.i.minor-faults
      3734            +2.4%       3823        perf-stat.i.page-faults
      1.22 ± 55%     -36.2%       0.78        perf-stat.overall.cpi
     10582 ± 11%     +23.5%      13066 ±  6%  perf-stat.overall.cycles-between-cache-misses
      1.03 ± 36%     +25.3%       1.29        perf-stat.overall.ipc
 9.044e+09 ± 36%     +25.9%  1.138e+10        perf-stat.ps.branch-instructions
   5452944 ± 11%     -20.5%    4334998 ±  6%  perf-stat.ps.cache-misses
    197.71            +1.7%     201.11        perf-stat.ps.cpu-migrations
 1.483e+10 ± 36%     +24.5%  1.847e+10        perf-stat.ps.dTLB-loads
  1.12e+10 ± 36%     +25.2%  1.403e+10        perf-stat.ps.dTLB-stores
      0.73 ± 12%     +95.9%       1.43 ± 12%  perf-stat.ps.major-faults
      3720            +2.4%       3809        perf-stat.ps.minor-faults
      3721            +2.4%       3810        perf-stat.ps.page-faults
     21770 ± 23%    -100.0%       0.00        sched_debug.cfs_rq:/.exec_clock.avg
    110841 ± 11%    -100.0%       0.00        sched_debug.cfs_rq:/.exec_clock.max
     18.36 ± 18%    -100.0%       0.00        sched_debug.cfs_rq:/.exec_clock.min
     33023 ± 14%    -100.0%       0.00        sched_debug.cfs_rq:/.exec_clock.stddev
      0.03 ± 78%    -100.0%       0.00        sched_debug.cfs_rq:/.nr_spread_over.avg
      1.34 ± 28%    -100.0%       0.00        sched_debug.cfs_rq:/.nr_spread_over.max
      0.17 ± 49%    -100.0%       0.00        sched_debug.cfs_rq:/.nr_spread_over.stddev
      1670 ± 11%    -100.0%       0.00        sched_debug.cpu.sched_count.avg
     67230 ± 12%    -100.0%       0.00        sched_debug.cpu.sched_count.max
    512.85 ±  9%    -100.0%       0.00        sched_debug.cpu.sched_count.min
      5175 ± 10%    -100.0%       0.00        sched_debug.cpu.sched_count.stddev
    755.90 ± 11%    -100.0%       0.00        sched_debug.cpu.sched_goidle.avg
     33544 ± 12%    -100.0%       0.00        sched_debug.cpu.sched_goidle.max
     70.86 ± 29%    -100.0%       0.00        sched_debug.cpu.sched_goidle.min
      2591 ± 10%    -100.0%       0.00        sched_debug.cpu.sched_goidle.stddev
    734.02 ± 12%    -100.0%       0.00        sched_debug.cpu.ttwu_count.avg
     33982 ± 13%    -100.0%       0.00        sched_debug.cpu.ttwu_count.max
    181.24 ±  9%    -100.0%       0.00        sched_debug.cpu.ttwu_count.min
      2616 ± 10%    -100.0%       0.00        sched_debug.cpu.ttwu_count.stddev
    308.36 ± 15%    -100.0%       0.00        sched_debug.cpu.ttwu_local.avg
      3731 ±105%    -100.0%       0.00        sched_debug.cpu.ttwu_local.max
    176.38 ± 10%    -100.0%       0.00        sched_debug.cpu.ttwu_local.min
    371.08 ± 77%    -100.0%       0.00        sched_debug.cpu.ttwu_local.stddev
      0.00            +0.6        0.62 ±  7%  perf-profile.calltrace.cycles-pp.syscall_exit_to_user_mode_prepare.syscall_exit_to_user_mode.entry_SYSCALL_64_after_hwframe.syscall
      0.00            +0.1        0.06 ± 18%  perf-profile.children.cycles-pp.unwind_next_frame
      0.00            +0.1        0.09 ± 13%  perf-profile.children.cycles-pp.perf_callchain_kernel
      0.00            +0.1        0.10 ± 13%  perf-profile.children.cycles-pp.get_perf_callchain
      0.00            +0.1        0.10 ± 13%  perf-profile.children.cycles-pp.perf_callchain
      0.00            +0.1        0.11 ± 12%  perf-profile.children.cycles-pp.perf_prepare_sample
      0.00            +0.1        0.11 ± 23%  perf-profile.children.cycles-pp.process_simple
      0.00            +0.1        0.11 ± 25%  perf-profile.children.cycles-pp.perf_session__process_events
      0.00            +0.1        0.11 ± 25%  perf-profile.children.cycles-pp.record__finish_output
      0.21 ± 39%      +0.1        0.33 ±  8%  perf-profile.children.cycles-pp.syscall@plt
      0.00            +0.1        0.12 ± 13%  perf-profile.children.cycles-pp.perf_trace_sched_stat_runtime
      0.00            +0.1        0.12 ± 12%  perf-profile.children.cycles-pp.perf_event_output_forward
      0.00            +0.1        0.13 ± 12%  perf-profile.children.cycles-pp.perf_swevent_overflow
      0.00            +0.1        0.13 ± 12%  perf-profile.children.cycles-pp.__perf_event_overflow
      0.00            +0.1        0.13 ± 14%  perf-profile.children.cycles-pp.update_curr
      0.00            +0.1        0.13 ± 12%  perf-profile.children.cycles-pp.perf_tp_event
      0.00            +0.1        0.14 ± 24%  perf-profile.children.cycles-pp.cmd_sched
      0.00            +0.1        0.14 ± 24%  perf-profile.children.cycles-pp.cmd_record
      0.21 ± 27%      +0.1        0.36 ± 17%  perf-profile.children.cycles-pp.scheduler_tick
      0.00            +0.1        0.15 ± 12%  perf-profile.children.cycles-pp.task_tick_fair
      0.00            +0.2        0.16 ± 21%  perf-profile.children.cycles-pp.__libc_start_main
      0.00            +0.2        0.16 ± 21%  perf-profile.children.cycles-pp.main
      0.00            +0.2        0.16 ± 21%  perf-profile.children.cycles-pp.run_builtin
      0.00            +0.6        0.62 ±  7%  perf-profile.children.cycles-pp.syscall_exit_to_user_mode_prepare
     12.66 ±140%     -11.4        1.22 ±  8%  perf-profile.self.cycles-pp.syscall_exit_to_user_mode
      0.00            +0.6        0.62 ±  7%  perf-profile.self.cycles-pp.syscall_exit_to_user_mode_prepare
      8798 ±204%    -100.0%       0.00        interrupts.161:PCI-MSI.524289-edge.eth0-TxRx-0
      2564 ±100%    -100.0%       0.00        interrupts.162:PCI-MSI.524290-edge.eth0-TxRx-1
     10781 ±172%    -100.0%       0.00        interrupts.163:PCI-MSI.524291-edge.eth0-TxRx-2
      7838 ±101%    -100.0%       0.00        interrupts.164:PCI-MSI.524292-edge.eth0-TxRx-3
     34.71 ± 31%    +448.9%     190.56 ± 68%  interrupts.CPU10.RES:Rescheduling_interrupts
      8798 ±204%    -100.0%       0.00        interrupts.CPU11.161:PCI-MSI.524289-edge.eth0-TxRx-0
      2156 ± 21%     -28.4%       1544 ± 20%  interrupts.CPU115.CAL:Function_call_interrupts
      2090 ± 17%     -25.8%       1551 ± 20%  interrupts.CPU116.CAL:Function_call_interrupts
      2564 ±100%    -100.0%       0.00        interrupts.CPU12.162:PCI-MSI.524290-edge.eth0-TxRx-1
      2330 ± 40%     -31.8%       1588 ± 18%  interrupts.CPU123.CAL:Function_call_interrupts
      2014 ±  4%     -21.2%       1587 ± 19%  interrupts.CPU124.CAL:Function_call_interrupts
      2689 ± 42%     -40.8%       1593 ± 16%  interrupts.CPU125.CAL:Function_call_interrupts
      2214 ± 27%     -29.9%       1552 ± 19%  interrupts.CPU126.CAL:Function_call_interrupts
     10781 ±172%    -100.0%       0.00        interrupts.CPU13.163:PCI-MSI.524291-edge.eth0-TxRx-2
      7838 ±101%    -100.0%       0.00        interrupts.CPU14.164:PCI-MSI.524292-edge.eth0-TxRx-3
      1962 ±  9%     -16.7%       1634 ± 13%  interrupts.CPU144.CAL:Function_call_interrupts
      1913 ±  5%     -17.8%       1572 ± 19%  interrupts.CPU146.CAL:Function_call_interrupts
      1877 ±  3%     -17.7%       1545 ± 20%  interrupts.CPU151.CAL:Function_call_interrupts
      2622 ± 69%     -40.6%       1558 ± 20%  interrupts.CPU165.CAL:Function_call_interrupts
    118.43 ± 29%    +332.2%     511.89 ± 24%  interrupts.CPU191.NMI:Non-maskable_interrupts
    118.43 ± 29%    +332.2%     511.89 ± 24%  interrupts.CPU191.PMI:Performance_monitoring_interrupts
      2625 ± 70%     -40.7%       1556 ± 19%  interrupts.CPU41.CAL:Function_call_interrupts
      1903 ±  4%     -18.8%       1544 ± 20%  interrupts.CPU45.CAL:Function_call_interrupts
      1955 ±  6%     -17.9%       1606 ± 20%  interrupts.CPU49.CAL:Function_call_interrupts
      2521 ± 40%     -38.1%       1560 ± 20%  interrupts.CPU75.CAL:Function_call_interrupts
    147.43 ±  9%    +121.8%     327.00 ± 22%  interrupts.CPU95.NMI:Non-maskable_interrupts
    147.43 ±  9%    +121.8%     327.00 ± 22%  interrupts.CPU95.PMI:Performance_monitoring_interrupts
      6669 ± 35%     -39.2%       4053 ± 17%  interrupts.RES:Rescheduling_interrupts


                                                                                
                              will-it-scale.per_process_ops                     
                                                                                
   1.04e+07 +---------------------------------------------------------------+   
            |             +..+    +..                                       |   
  1.035e+07 |..+..+.+.. ..    + ..                                          |   
            |          +       +     +..+..                                 |   
            |                              +    +..           +..        +..|   
   1.03e+07 |-+                             + ..     .+..   ..   +..   ..   |   
            |                                +     +.    +.+        +.+     |   
  1.025e+07 |-+                                                             |   
            |                O       O                   O       O  O       |   
   1.02e+07 |-+                O  O     O  O O     O  O    O  O       O  O  |   
            |     O O  O  O                     O                           |   
            |                                                               |   
  1.015e+07 |-+                                                             |   
            |  O                                                            |   
   1.01e+07 +---------------------------------------------------------------+   
                                                                                
                                                                                
[*] bisect-good sample
[O] bisect-bad  sample

***************************************************************************************************
lkp-csl-2sp4: 96 threads Intel(R) Xeon(R) CPU @ 2.30GHz with 128G memory
=========================================================================================
compiler/cpufreq_governor/kconfig/nr_task/rootfs/runtime/tbox_group/test/testcase/ucode:
  gcc-9/performance/x86_64-rhel-8.3/1/debian-10.4-x86_64-20200603.cgz/300s/lkp-csl-2sp4/syscall/unixbench/0x4003003

commit: 
  v5.10-rc4
  a358d5636c ("entry: split lockdep and syscall work functions")

       v5.10-rc4 a358d5636cfcb8b278c766ac261 
---------------- --------------------------- 
       fail:runs  %reproduction    fail:runs
           |             |             |    
           :9           44%           4:4     stderr.Events_disabled
           :9           44%           4:4     stderr.Events_enabled
           :9           44%           4:4     stderr.[perf_record:Captured_and_wrote#MB/tmp/lkp/perf-sched.data(#samples)]
           :9           44%           4:4     stderr.[perf_record:Woken_up#times_to_write_data]
           :9           44%           4:4     stderr.has_stderr
          1:9          -10%           0:4     perf-profile.children.cycles-pp.error_entry
          1:9           -9%           0:4     perf-profile.self.cycles-pp.error_entry
         %stddev     %change         %stddev
             \          |                \  
      1506            -1.9%       1478        unixbench.score
     54357            -7.7%      50172        unixbench.time.minor_page_faults
     92.23            +1.7%      93.76        unixbench.time.user_time
 8.815e+08            -1.8%  8.654e+08        unixbench.workload
     21.47            +3.1%      22.13        boot-time.boot
     32298 ± 41%     -37.1%      20310 ±  7%  softirqs.CPU62.RCU
      6703 ±  3%     +11.1%       7448        slabinfo.kmalloc-2k.active_objs
      6757 ±  3%     +12.0%       7567        slabinfo.kmalloc-2k.num_objs
    837.78 ± 11%     +27.9%       1071 ±  5%  slabinfo.mnt_cache.active_objs
    837.78 ± 11%     +27.9%       1071 ±  5%  slabinfo.mnt_cache.num_objs
      3003 ± 52%    +196.2%       8895 ±  7%  numa-meminfo.node1.Active
      2964 ± 53%    +198.2%       8839 ±  7%  numa-meminfo.node1.Active(anon)
    450792 ±  6%     +11.8%     503916 ±  9%  numa-meminfo.node1.FilePages
      8176 ± 40%     +87.8%      15358 ± 20%  numa-meminfo.node1.Shmem
    442607 ±  6%     +10.4%     488571 ±  9%  numa-meminfo.node1.Unevictable
      6811 ±  3%     +75.3%      11943 ±  5%  meminfo.Active
      6545 ±  4%     +71.0%      11193 ±  5%  meminfo.Active(anon)
    265.89 ±107%    +181.6%     748.75 ± 74%  meminfo.Active(file)
   2086104           +12.7%    2350606        meminfo.Memused
     16185           +28.7%      20830 ±  3%  meminfo.Shmem
      5936 ±  2%     +21.3%       7199 ±  2%  meminfo.max_used_kB
    740.67 ± 53%    +198.6%       2211 ±  8%  numa-vmstat.node1.nr_active_anon
    112698 ±  6%     +11.8%     125979 ±  9%  numa-vmstat.node1.nr_file_pages
      2044 ± 40%     +87.9%       3840 ± 20%  numa-vmstat.node1.nr_shmem
    110651 ±  6%     +10.4%     122142 ±  9%  numa-vmstat.node1.nr_unevictable
    740.67 ± 53%    +198.6%       2211 ±  8%  numa-vmstat.node1.nr_zone_active_anon
    110651 ±  6%     +10.4%     122142 ±  9%  numa-vmstat.node1.nr_zone_unevictable
   1770566 ±  9%     -26.4%    1303772 ± 23%  perf-stat.i.cache-misses
      4073 ±  6%     +32.6%       5402 ± 13%  perf-stat.i.cycles-between-cache-misses
 6.456e+08            +1.3%  6.541e+08        perf-stat.i.dTLB-stores
      0.48 ±  9%     +11.1%       0.53 ±  8%  perf-stat.i.ipc
      2903            +1.1%       2936        perf-stat.i.minor-faults
      2903            +1.1%       2936        perf-stat.i.page-faults
      3677           +25.2%       4604 ±  9%  perf-stat.overall.cycles-between-cache-misses
   1766768 ±  9%     -26.4%    1301105 ± 23%  perf-stat.ps.cache-misses
 6.438e+08            +1.3%  6.524e+08        perf-stat.ps.dTLB-stores
      2896            +1.1%       2929        perf-stat.ps.minor-faults
      2897            +1.1%       2929        perf-stat.ps.page-faults
      0.55 ± 38%      +0.4        0.96 ± 13%  perf-profile.calltrace.cycles-pp.exit_to_user_mode_prepare.syscall_exit_to_user_mode.entry_SYSCALL_64_after_hwframe
      1.13 ± 12%      +0.6        1.74 ± 14%  perf-profile.calltrace.cycles-pp.syscall_exit_to_user_mode.entry_SYSCALL_64_after_hwframe
      0.14 ± 13%      +0.0        0.18 ±  8%  perf-profile.children.cycles-pp.call_cpuidle
      0.12 ± 22%      +0.1        0.18 ± 24%  perf-profile.children.cycles-pp.do_user_addr_fault
      0.14 ± 19%      +0.1        0.20 ± 18%  perf-profile.children.cycles-pp.asm_exc_page_fault
      0.13 ± 22%      +0.1        0.18 ± 20%  perf-profile.children.cycles-pp.exc_page_fault
      0.11 ± 26%      +0.1        0.16 ± 22%  perf-profile.children.cycles-pp.__handle_mm_fault
      0.11 ± 23%      +0.1        0.17 ± 22%  perf-profile.children.cycles-pp.handle_mm_fault
      0.77 ± 11%      +0.5        1.24 ± 13%  perf-profile.children.cycles-pp.exit_to_user_mode_prepare
      0.00            +0.6        0.60 ± 20%  perf-profile.children.cycles-pp.syscall_exit_to_user_mode_prepare
      1.70 ±  8%      +0.8        2.51 ± 15%  perf-profile.children.cycles-pp.syscall_exit_to_user_mode
      0.14 ± 15%      +0.0        0.18 ±  8%  perf-profile.self.cycles-pp.call_cpuidle
      0.06 ± 59%      +0.0        0.10 ± 27%  perf-profile.self.cycles-pp.trigger_load_balance
      0.63 ± 12%      +0.5        1.12 ± 15%  perf-profile.self.cycles-pp.exit_to_user_mode_prepare
      0.00            +0.5        0.53 ± 17%  perf-profile.self.cycles-pp.syscall_exit_to_user_mode_prepare
      1635 ±  4%     +71.0%       2795 ±  5%  proc-vmstat.nr_active_anon
     65.89 ±108%    +183.8%     187.00 ± 74%  proc-vmstat.nr_active_file
     58918            +5.3%      62014        proc-vmstat.nr_anon_pages
    239429            +3.9%     248685        proc-vmstat.nr_file_pages
     61168            +5.0%      64234        proc-vmstat.nr_inactive_anon
      8304            +2.2%       8483        proc-vmstat.nr_mapped
      4045           +28.7%       5208 ±  3%  proc-vmstat.nr_shmem
    235310            +3.4%     243282        proc-vmstat.nr_unevictable
      1635 ±  4%     +71.0%       2795 ±  5%  proc-vmstat.nr_zone_active_anon
     65.89 ±108%    +183.8%     187.00 ± 74%  proc-vmstat.nr_zone_active_file
     61168            +5.0%      64234        proc-vmstat.nr_zone_inactive_anon
    235310            +3.4%     243282        proc-vmstat.nr_zone_unevictable
    937318           +14.1%    1069639        proc-vmstat.numa_hit
    906134            +8.1%     979942        proc-vmstat.numa_local
     31184          +187.6%      89696 ±  5%  proc-vmstat.numa_other
    999.44 ± 15%    +282.9%       3827 ±  3%  proc-vmstat.pgactivate
    996499           +13.8%    1134186        proc-vmstat.pgalloc_normal
   1174226            +1.3%    1189681        proc-vmstat.pgfault
   1031237           +12.2%    1157491        proc-vmstat.pgfree
    625.89 ±222%     -87.5%      78.25 ± 30%  interrupts.CPU11.NMI:Non-maskable_interrupts
    625.89 ±222%     -87.5%      78.25 ± 30%  interrupts.CPU11.PMI:Performance_monitoring_interrupts
     24.89 ±102%    +152.1%      62.75 ± 61%  interrupts.CPU12.RES:Rescheduling_interrupts
    337.44 ± 44%    +195.5%     997.00 ± 82%  interrupts.CPU36.CAL:Function_call_interrupts
    363.89 ± 33%     +67.4%     609.25 ± 19%  interrupts.CPU38.CAL:Function_call_interrupts
    408.78 ± 56%   +5575.0%      23198 ±166%  interrupts.CPU4.CAL:Function_call_interrupts
    347.89 ± 47%    +106.7%     719.25 ± 44%  interrupts.CPU42.CAL:Function_call_interrupts
    347.89 ± 47%     +56.9%     545.75 ±  3%  interrupts.CPU45.CAL:Function_call_interrupts
    652.78 ±216%     -87.8%      79.75 ± 42%  interrupts.CPU53.NMI:Non-maskable_interrupts
    652.78 ±216%     -87.8%      79.75 ± 42%  interrupts.CPU53.PMI:Performance_monitoring_interrupts
    288.11 ±145%     -67.2%      94.50 ± 27%  interrupts.CPU54.NMI:Non-maskable_interrupts
    288.11 ±145%     -67.2%      94.50 ± 27%  interrupts.CPU54.PMI:Performance_monitoring_interrupts
    128.44 ± 18%     -49.4%      65.00 ± 33%  interrupts.CPU55.NMI:Non-maskable_interrupts
    128.44 ± 18%     -49.4%      65.00 ± 33%  interrupts.CPU55.PMI:Performance_monitoring_interrupts
    142.33 ± 14%     -36.2%      90.75 ± 27%  interrupts.CPU56.NMI:Non-maskable_interrupts
    142.33 ± 14%     -36.2%      90.75 ± 27%  interrupts.CPU56.PMI:Performance_monitoring_interrupts
    141.78 ± 15%     -44.8%      78.25 ± 23%  interrupts.CPU58.NMI:Non-maskable_interrupts
    141.78 ± 15%     -44.8%      78.25 ± 23%  interrupts.CPU58.PMI:Performance_monitoring_interrupts
    351.22 ± 47%    +221.2%       1128 ± 85%  interrupts.CPU81.CAL:Function_call_interrupts
    363.22 ± 47%    +500.0%       2179 ± 76%  interrupts.CPU90.CAL:Function_call_interrupts
      3295 ±  9%     -35.4%       2128 ± 10%  interrupts.RES:Rescheduling_interrupts
      1796 ± 23%    -100.0%       0.00        sched_debug.cfs_rq:/.exec_clock.avg
     35813 ± 23%    -100.0%       0.00        sched_debug.cfs_rq:/.exec_clock.max
     11.90 ± 29%    -100.0%       0.00        sched_debug.cfs_rq:/.exec_clock.min
      6114 ± 21%    -100.0%       0.00        sched_debug.cfs_rq:/.exec_clock.stddev
      0.04 ±  7%     +11.2%       0.05 ±  9%  sched_debug.cpu.nr_running.avg
      2585 ± 16%    -100.0%       0.00        sched_debug.cpu.sched_count.avg
     56814 ± 16%    -100.0%       0.00        sched_debug.cpu.sched_count.max
    602.46 ±  8%    -100.0%       0.00        sched_debug.cpu.sched_count.min
      6511 ± 23%    -100.0%       0.00        sched_debug.cpu.sched_count.stddev
      1213 ± 16%    -100.0%       0.00        sched_debug.cpu.sched_goidle.avg
     28314 ± 16%    -100.0%       0.00        sched_debug.cpu.sched_goidle.max
    216.95 ±  8%    -100.0%       0.00        sched_debug.cpu.sched_goidle.min
      3257 ± 23%    -100.0%       0.00        sched_debug.cpu.sched_goidle.stddev
      1155 ± 17%    -100.0%       0.00        sched_debug.cpu.ttwu_count.avg
     26529 ± 21%    -100.0%       0.00        sched_debug.cpu.ttwu_count.max
    212.36 ±  8%    -100.0%       0.00        sched_debug.cpu.ttwu_count.min
      3187 ± 26%    -100.0%       0.00        sched_debug.cpu.ttwu_count.stddev
    535.27 ± 28%    -100.0%       0.00        sched_debug.cpu.ttwu_local.avg
     11898 ± 98%    -100.0%       0.00        sched_debug.cpu.ttwu_local.max
    210.09 ±  8%    -100.0%       0.00        sched_debug.cpu.ttwu_local.min
      1252 ± 91%    -100.0%       0.00        sched_debug.cpu.ttwu_local.stddev





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-5.10.0-rc4-00001-ga358d5636cfc" of type "text/plain" (170102 bytes)

View attachment "job-script" of type "text/plain" (7933 bytes)

View attachment "job.yaml" of type "text/plain" (5178 bytes)

View attachment "reproduce" of type "text/plain" (338 bytes)

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ