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]
Date:   Wed, 5 Feb 2020 20:28:05 +0800
From:   kernel test robot <rong.a.chen@...el.com>
To:     Ming Lei <ming.lei@...hat.com>
Cc:     0day robot <lkp@...el.com>,
        Sathya Prakash <sathya.prakash@...adcom.com>,
        Chaitra P B <chaitra.basappa@...adcom.com>,
        Suganath Prabu Subramani 
        <suganath-prabu.subramani@...adcom.com>,
        Kashyap Desai <kashyap.desai@...adcom.com>,
        Sumit Saxena <sumit.saxena@...adcom.com>,
        Shivasharan S <shivasharan.srikanteshwara@...adcom.com>,
        "Ewan D. Milne" <emilne@...hat.com>,
        Hannes Reinecke <hare@...e.de>,
        Bart Van Assche <bart.vanassche@....com>,
        LKML <linux-kernel@...r.kernel.org>, lkp@...ts.01.org
Subject: [scsi] d4ea3041f3: fsmark.files_per_sec -11.8% regression

Greeting,

FYI, we noticed a -11.8% regression of fsmark.files_per_sec due to commit:


commit: d4ea3041f3fd005c1be676f66031c24d06c9dc9b ("scsi: sd: register request queue after sd_revalidate_disk is done")
https://github.com/0day-ci/linux/commits/Ming-Lei/scsi-support-bypass-device-busy-check-for-some-high-end-HBA-with-SSD/20200119-151701

in testcase: fsmark
on test machine: 96 threads Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz with 256G memory
with following parameters:

	iterations: 1x
	nr_threads: 32t
	disk: 1SSD
	fs: xfs
	filesize: 9B
	test_size: 400M
	sync_method: fsyncBeforeClose
	nr_directories: 16d
	nr_files_per_directory: 256fpd
	cpufreq_governor: performance
	ucode: 0x500002c

test-description: The fsmark is a file system benchmark to test synchronous write workloads, for example, mail servers workload.
test-url: https://sourceforge.net/projects/fsmark/

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

+------------------+----------------------------------------------------------------------+
| testcase: change | filebench: filebench.sum_bytes_mb/s -4.1% regression                 |
| test machine     | 8 threads Intel(R) Core(TM) i7-4770 CPU @ 3.40GHz with 8G memory     |
| test parameters  | cpufreq_governor=performance                                         |
|                  | disk=1HDD                                                            |
|                  | fs=xfs                                                               |
|                  | test=singlestreamreaddirect.f                                        |
|                  | ucode=0x27                                                           |
+------------------+----------------------------------------------------------------------+
| testcase: change | fsmark: fsmark.files_per_sec -11.2% regression                       |
| test machine     | 96 threads Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz with 256G memory |
| test parameters  | cpufreq_governor=performance                                         |
|                  | disk=1SSD                                                            |
|                  | filesize=8K                                                          |
|                  | fs=xfs                                                               |
|                  | iterations=1x                                                        |
|                  | nr_directories=16d                                                   |
|                  | nr_files_per_directory=256fpd                                        |
|                  | nr_threads=32t                                                       |
|                  | sync_method=fsyncBeforeClose                                         |
|                  | test_size=400M                                                       |
|                  | ucode=0x500002c                                                      |
+------------------+----------------------------------------------------------------------+


If you fix the issue, kindly add following tag
Reported-by: kernel test robot <rong.a.chen@...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/disk/filesize/fs/iterations/kconfig/nr_directories/nr_files_per_directory/nr_threads/rootfs/sync_method/tbox_group/test_size/testcase/ucode:
  gcc-7/performance/1SSD/9B/xfs/1x/x86_64-rhel-7.6/16d/256fpd/32t/debian-x86_64-20191114.cgz/fsyncBeforeClose/lkp-csl-2sp7/400M/fsmark/0x500002c

commit: 
  8186413d49 ("scsi: remove .for_blk_mq")
  d4ea3041f3 ("scsi: sd: register request queue after sd_revalidate_disk is done")

8186413d49989aa6 d4ea3041f3fd005c1be676f6603 
---------------- --------------------------- 
         %stddev     %change         %stddev
             \          |                \  
     19966           -11.8%      17603        fsmark.files_per_sec
     43027           +13.3%      48749 ±  2%  fsmark.time.involuntary_context_switches
    237.67 ±  2%      -9.5%     215.00        fsmark.time.percent_of_cpu_this_job_got
    575974            +0.9%     581111        fsmark.time.voluntary_context_switches
     50117 ± 59%     +61.6%      81010 ± 34%  numa-numastat.node1.numa_hit
      1.28 ±  2%      +3.2%       1.32 ±  2%  perf-stat.overall.cpi
    250801 ±  5%     +70.1%     426653 ± 50%  turbostat.C1
    268.67 ±104%   +2589.0%       7224 ±120%  numa-vmstat.node1.nr_inactive_file
    268.67 ±104%   +2589.0%       7224 ±120%  numa-vmstat.node1.nr_zone_inactive_file
    251408 ±  5%     +70.1%     427525 ± 50%  cpuidle.C1.usage
    226827 ±  4%    +103.3%     461250 ± 65%  cpuidle.POLL.time
     42260 ±  6%    +102.4%      85537 ± 59%  cpuidle.POLL.usage
     13.39 ± 71%     -87.7%       1.65 ±125%  iostat.sda.await.max
    127.48 ± 71%     -89.6%      13.30 ±136%  iostat.sda.r_await.max
      2.51 ±141%     -70.5%       0.74 ±100%  iostat.sdb.await.max
     23.30 ±141%     -80.7%       4.48 ±100%  iostat.sdb.r_await.max
      6252 ± 60%     -84.3%     982.25 ±125%  proc-vmstat.numa_hint_faults
      4011 ± 67%    +183.0%      11354 ± 22%  proc-vmstat.numa_pages_migrated
      4011 ± 67%    +183.0%      11354 ± 22%  proc-vmstat.pgmigrate_success
   1102146            -1.7%    1083816        proc-vmstat.pgpgout
    123676 ±  6%     -13.5%     106962 ±  6%  numa-meminfo.node0.Inactive
     12985 ± 24%    +209.6%      40200 ± 72%  numa-meminfo.node1.Inactive
      1071 ±104%   +2576.6%      28684 ±122%  numa-meminfo.node1.Inactive(file)
   1219741 ±  8%      +9.5%    1335221 ±  3%  numa-meminfo.node1.MemUsed
    121749 ±  3%      +8.8%     132487 ±  6%  numa-meminfo.node1.Slab
      9101 ±  3%     -11.5%       8058 ±  5%  slabinfo.kmalloc-1k.active_objs
      9173 ±  3%     -11.6%       8107 ±  5%  slabinfo.kmalloc-1k.num_objs
      2287 ± 17%     +16.1%       2655 ±  3%  slabinfo.pool_workqueue.active_objs
      2287 ± 17%     +16.2%       2656 ±  3%  slabinfo.pool_workqueue.num_objs
    640.00           -20.0%     512.00        slabinfo.scsi_sense_cache.active_objs
    640.00           -20.0%     512.00        slabinfo.scsi_sense_cache.num_objs
      2187 ± 24%     +24.0%       2713 ±  5%  slabinfo.xfs_btree_cur.active_objs
      2187 ± 24%     +24.0%       2713 ±  5%  slabinfo.xfs_btree_cur.num_objs
      3225 ± 15%     +16.8%       3766 ±  4%  slabinfo.xfs_buf.active_objs
      3271 ± 16%     +16.8%       3822 ±  4%  slabinfo.xfs_buf.num_objs
      0.12 ± 20%     +29.0%       0.15 ± 10%  sched_debug.cfs_rq:/.nr_running.avg
      0.33 ± 11%     +13.2%       0.37 ±  3%  sched_debug.cfs_rq:/.nr_running.stddev
     -1551          +293.8%      -6110        sched_debug.cfs_rq:/.spread0.min
    257.42           +18.6%     305.18 ± 10%  sched_debug.cfs_rq:/.util_avg.avg
      1814 ± 14%     -27.6%       1314 ± 17%  sched_debug.cfs_rq:/.util_avg.max
     13631 ±123%    +419.4%      70797 ± 75%  sched_debug.cpu.avg_idle.min
    -20.67           +25.8%     -26.00        sched_debug.cpu.nr_uninterruptible.min
      6.41 ± 92%     -93.8%       0.40 ±144%  sched_debug.cpu.sched_count.avg
    136.67 ± 87%     -92.3%      10.50 ±151%  sched_debug.cpu.sched_count.max
     26.08 ± 89%     -92.7%       1.91 ±148%  sched_debug.cpu.sched_count.stddev
     72.00 ± 85%     -92.7%       5.25 ±151%  sched_debug.cpu.sched_goidle.max
     13.64 ± 87%     -92.9%       0.96 ±146%  sched_debug.cpu.sched_goidle.stddev
     12.85 ± 56%     -12.8        0.00        perf-profile.calltrace.cycles-pp.__x64_sys_exit_group.do_syscall_64.entry_SYSCALL_64_after_hwframe
     12.85 ± 56%     -12.8        0.00        perf-profile.calltrace.cycles-pp.do_group_exit.__x64_sys_exit_group.do_syscall_64.entry_SYSCALL_64_after_hwframe
     12.85 ± 56%     -12.8        0.00        perf-profile.calltrace.cycles-pp.do_exit.do_group_exit.__x64_sys_exit_group.do_syscall_64.entry_SYSCALL_64_after_hwframe
      7.90 ± 79%      -7.9        0.00        perf-profile.calltrace.cycles-pp.exit_mmap.mmput.do_exit.do_group_exit.__x64_sys_exit_group
      7.90 ± 79%      -7.9        0.00        perf-profile.calltrace.cycles-pp.mmput.do_exit.do_group_exit.__x64_sys_exit_group.do_syscall_64
      7.72 ±  6%      -5.8        1.92 ±173%  perf-profile.calltrace.cycles-pp.page_fault
      7.72 ±  6%      -5.8        1.92 ±173%  perf-profile.calltrace.cycles-pp.do_page_fault.page_fault
      7.72 ±  6%      -5.8        1.92 ±173%  perf-profile.calltrace.cycles-pp.__do_page_fault.do_page_fault.page_fault
      7.72 ±  6%      -5.8        1.92 ±173%  perf-profile.calltrace.cycles-pp.handle_mm_fault.__do_page_fault.do_page_fault.page_fault
      4.94 ± 70%      -4.9        0.00        perf-profile.calltrace.cycles-pp.do_wp_page.handle_pte_fault.__handle_mm_fault.handle_mm_fault.__do_page_fault
     12.85 ± 56%     -12.8        0.00        perf-profile.children.cycles-pp.__x64_sys_exit_group
      7.90 ± 79%      -6.0        1.92 ±173%  perf-profile.children.cycles-pp.mmput
      7.90 ± 79%      -6.0        1.92 ±173%  perf-profile.children.cycles-pp.exit_mmap
      4.94 ± 70%      -4.9        0.00        perf-profile.children.cycles-pp.do_wp_page
     68230           +19.1%      81258        softirqs.BLOCK
     11147 ±  3%     -15.5%       9418 ±  4%  softirqs.CPU12.TIMER
      9184 ±  5%     +20.9%      11108 ± 11%  softirqs.CPU25.TIMER
      9146 ±  5%     +22.3%      11186 ± 10%  softirqs.CPU26.TIMER
      9110 ±  3%     +23.3%      11233 ± 12%  softirqs.CPU28.TIMER
      9832 ±  5%     +18.8%      11678 ±  6%  softirqs.CPU3.TIMER
      9172 ±  4%     +15.3%      10572 ±  8%  softirqs.CPU30.TIMER
      8881 ±  6%     +23.5%      10967 ± 12%  softirqs.CPU35.TIMER
      9032 ±  5%     +18.8%      10726 ±  6%  softirqs.CPU36.TIMER
      8979 ±  8%     +15.1%      10339 ± 10%  softirqs.CPU37.TIMER
      9099 ±  8%     +19.2%      10843 ± 12%  softirqs.CPU38.TIMER
      9012 ±  7%     +14.6%      10325 ±  9%  softirqs.CPU40.TIMER
      9027 ±  5%     +23.0%      11106 ± 15%  softirqs.CPU43.TIMER
      8985 ±  8%     +27.3%      11433 ± 10%  softirqs.CPU44.TIMER
      9463 ±  5%     +26.4%      11965 ± 19%  softirqs.CPU47.TIMER
     10177 ±  7%     -13.0%       8857 ±  2%  softirqs.CPU67.TIMER
      9335 ±  6%     +16.8%      10904 ±  6%  softirqs.CPU78.TIMER
      8584 ±  6%     +16.0%       9961 ±  9%  softirqs.CPU84.TIMER
      8580 ±  7%     +23.4%      10587 ± 14%  softirqs.CPU89.TIMER
      8550 ±  8%     +18.2%      10104 ± 10%  softirqs.CPU90.TIMER
      8451 ±  5%     +24.6%      10531 ± 11%  softirqs.CPU93.TIMER
      8436 ±  4%     +32.6%      11186 ± 18%  softirqs.CPU94.TIMER
      8780 ±  8%     +31.5%      11546 ±  7%  softirqs.CPU95.TIMER
    131591           +19.7%     157488        interrupts.316:PCI-MSI.376832-edge.ahci[0000:00:17.0]
     15638 ± 11%    +224.4%      50724 ± 56%  interrupts.CAL:Function_call_interrupts
     94.33 ± 14%     -43.0%      53.75 ± 43%  interrupts.CPU20.RES:Rescheduling_interrupts
    262.67 ± 68%     -71.2%      75.75 ± 20%  interrupts.CPU23.RES:Rescheduling_interrupts
    171.33 ± 15%    +386.6%     833.75 ± 71%  interrupts.CPU24.CAL:Function_call_interrupts
    167.67 ± 14%    +436.9%     900.25 ± 68%  interrupts.CPU25.CAL:Function_call_interrupts
    170.33 ± 15%    +318.2%     712.25 ± 58%  interrupts.CPU26.CAL:Function_call_interrupts
    167.67 ± 12%    +421.3%     874.00 ± 69%  interrupts.CPU27.CAL:Function_call_interrupts
    179.00 ± 20%    +388.7%     874.75 ± 71%  interrupts.CPU29.CAL:Function_call_interrupts
      3.67 ±105%   +5006.8%     187.25 ±139%  interrupts.CPU29.RES:Rescheduling_interrupts
    152.00 ± 10%     +20.6%     183.25 ±  8%  interrupts.CPU3.CAL:Function_call_interrupts
    130.33 ± 49%    +531.5%     823.00 ± 66%  interrupts.CPU30.CAL:Function_call_interrupts
      4.00 ±108%   +4112.5%     168.50 ±137%  interrupts.CPU31.RES:Rescheduling_interrupts
    166.67 ± 11%    +417.8%     863.00 ± 66%  interrupts.CPU32.CAL:Function_call_interrupts
    162.00 ± 22%    +414.4%     833.25 ± 68%  interrupts.CPU33.CAL:Function_call_interrupts
    120.33 ± 38%    +617.6%     863.50 ± 69%  interrupts.CPU34.CAL:Function_call_interrupts
    165.67 ± 11%    +408.7%     842.75 ± 69%  interrupts.CPU35.CAL:Function_call_interrupts
    163.33 ±  9%    +420.7%     850.50 ± 68%  interrupts.CPU36.CAL:Function_call_interrupts
    169.33 ± 12%    +451.3%     933.50 ± 75%  interrupts.CPU37.CAL:Function_call_interrupts
    169.00 ± 13%    +401.8%     848.00 ± 70%  interrupts.CPU38.CAL:Function_call_interrupts
    164.67 ±  9%    +433.5%     878.50 ± 76%  interrupts.CPU39.CAL:Function_call_interrupts
    147.67 ± 12%     +18.7%     175.25 ± 11%  interrupts.CPU4.CAL:Function_call_interrupts
    174.67 ± 16%    +393.4%     861.75 ± 67%  interrupts.CPU40.CAL:Function_call_interrupts
      2.67 ±141%  +26928.1%     720.75 ±164%  interrupts.CPU40.RES:Rescheduling_interrupts
    166.00 ±  9%    +427.6%     875.75 ± 69%  interrupts.CPU41.CAL:Function_call_interrupts
    166.67 ± 10%    +456.1%     926.75 ± 70%  interrupts.CPU42.CAL:Function_call_interrupts
    175.00 ± 15%    +511.4%       1070 ± 70%  interrupts.CPU43.CAL:Function_call_interrupts
    173.33 ± 13%    +437.1%     931.00 ± 71%  interrupts.CPU44.CAL:Function_call_interrupts
    178.00 ± 14%    +479.1%       1030 ± 77%  interrupts.CPU45.CAL:Function_call_interrupts
    174.00 ± 13%    +455.2%     966.00 ± 77%  interrupts.CPU46.CAL:Function_call_interrupts
    162.67 ±  8%    +335.9%     709.00 ± 64%  interrupts.CPU47.CAL:Function_call_interrupts
    154.33 ± 12%     +13.4%     175.00 ± 11%  interrupts.CPU57.CAL:Function_call_interrupts
    101.33 ± 14%     -34.6%      66.25 ± 36%  interrupts.CPU57.RES:Rescheduling_interrupts
    101.00 ± 25%     -51.0%      49.50 ± 39%  interrupts.CPU6.RES:Rescheduling_interrupts
    109.00 ±  5%     -28.2%      78.25 ± 24%  interrupts.CPU60.RES:Rescheduling_interrupts
    179.33 ± 18%    +428.6%     948.00 ± 69%  interrupts.CPU72.CAL:Function_call_interrupts
    176.33 ± 18%    +382.9%     851.50 ± 66%  interrupts.CPU73.CAL:Function_call_interrupts
    173.67 ± 15%    +317.9%     725.75 ± 58%  interrupts.CPU74.CAL:Function_call_interrupts
    170.33 ± 13%    +422.7%     890.25 ± 72%  interrupts.CPU75.CAL:Function_call_interrupts
    171.67 ± 14%    +493.9%       1019 ± 70%  interrupts.CPU76.CAL:Function_call_interrupts
    179.00 ± 21%    +471.4%       1022 ± 70%  interrupts.CPU77.CAL:Function_call_interrupts
    170.67 ± 16%    +442.1%     925.25 ± 72%  interrupts.CPU78.CAL:Function_call_interrupts
    173.33 ± 17%    +429.2%     917.25 ± 76%  interrupts.CPU79.CAL:Function_call_interrupts
    171.00 ± 72%     -73.5%      45.25 ± 34%  interrupts.CPU8.RES:Rescheduling_interrupts
    135.00 ± 57%    +533.0%     854.50 ± 67%  interrupts.CPU80.CAL:Function_call_interrupts
    161.67 ± 22%    +421.3%     842.75 ± 67%  interrupts.CPU81.CAL:Function_call_interrupts
    171.00 ± 17%    +426.9%     901.00 ± 71%  interrupts.CPU82.CAL:Function_call_interrupts
    172.33 ± 17%    +376.3%     820.75 ± 75%  interrupts.CPU83.CAL:Function_call_interrupts
    176.00 ± 18%    +448.6%     965.50 ± 70%  interrupts.CPU84.CAL:Function_call_interrupts
    175.67 ± 18%    +428.3%     928.00 ± 72%  interrupts.CPU85.CAL:Function_call_interrupts
    176.67 ± 17%    +382.0%     851.50 ± 68%  interrupts.CPU86.CAL:Function_call_interrupts
    174.00 ± 16%    +459.2%     973.00 ± 76%  interrupts.CPU87.CAL:Function_call_interrupts
    178.00 ± 16%    +517.3%       1098 ± 89%  interrupts.CPU88.CAL:Function_call_interrupts
    181.33 ± 21%    +422.7%     947.75 ± 77%  interrupts.CPU89.CAL:Function_call_interrupts
    173.67 ± 16%    +421.7%     906.00 ± 73%  interrupts.CPU90.CAL:Function_call_interrupts
    173.67 ± 20%    +394.2%     858.25 ± 70%  interrupts.CPU91.CAL:Function_call_interrupts
    173.67 ± 20%    +425.7%     913.00 ± 69%  interrupts.CPU92.CAL:Function_call_interrupts
    170.67 ± 18%    +402.7%     858.00 ± 67%  interrupts.CPU93.CAL:Function_call_interrupts
    173.67 ± 21%    +404.8%     876.75 ± 71%  interrupts.CPU94.CAL:Function_call_interrupts
    158.00 ±  8%    +316.0%     657.25 ± 57%  interrupts.CPU95.CAL:Function_call_interrupts


                                                                                
                      fsmark.time.percent_of_cpu_this_job_got                   
                                                                                
  250 +-+-------------------------------------------------------------------+   
      | +   +   +.+   :  +.+   +.+.+.+     +.+.+.+.+.+   +.++   :   +   +.+ |   
      O O O O O O O   O  O O O   O O O   O O O O O O O      :   :   :   :   |   
  200 +-+             :  :                                  :   :   :   :   |   
      |                : :                                  :  :    :   :   |   
      |                : :                                   : :     : :    |   
  150 +-+              : :                                   : :     : :    |   
      |                : :                                   : :     : :    |   
  100 +-+              ::                                    : :     : :    |   
      |                ::                                    : :     : :    |   
      |                ::                                    : :     : :    |   
   50 +-+               :                                     :       :     |   
      |                 :                                     :       :     |   
      |                 :                                     :       :     |   
    0 +-+-----------O---O------O-------O------------------------------------+   
                                                                                
                                                                                                                                                                
                                fsmark.files_per_sec                            
                                                                                
  25000 +-+-----------------------------------------------------------------+   
        |                                                                   |   
        |                                                                   |   
  20000 +-+.+.+.++.+.+.+   +.+.+.++.+.+.+.+.+.+.+.++.+.+.+.+.+   +.++   +.+.|   
        O O O O OO O   O   O O O  O O O   O O O O OO O       :   :  :   :   |   
        |              :   :                                 :   :  :   :   |   
  15000 +-+             : :                                   : :    : :    |   
        |               : :                                   : :    : :    |   
  10000 +-+             : :                                   : :    : :    |   
        |               : :                                   : :    : :    |   
        |               : :                                   : :    : :    |   
   5000 +-+             : :                                   : :    : :    |   
        |                :                                     :      :     |   
        |                :                                     :      :     |   
      0 +-+----------O---O-------O------O-----------------------------------+   
                                                                                
                                                                                
[*] bisect-good sample
[O] bisect-bad  sample

***************************************************************************************************
lkp-hsw-d01: 8 threads Intel(R) Core(TM) i7-4770 CPU @ 3.40GHz with 8G memory
=========================================================================================
compiler/cpufreq_governor/disk/fs/kconfig/rootfs/tbox_group/test/testcase/ucode:
  gcc-7/performance/1HDD/xfs/x86_64-rhel-7.6/debian-x86_64-20191114.cgz/lkp-hsw-d01/singlestreamreaddirect.f/filebench/0x27

commit: 
  8186413d49 ("scsi: remove .for_blk_mq")
  d4ea3041f3 ("scsi: sd: register request queue after sd_revalidate_disk is done")

8186413d49989aa6 d4ea3041f3fd005c1be676f6603 
---------------- --------------------------- 
       fail:runs  %reproduction    fail:runs
           |             |             |    
          1:4          -25%            :4     dmesg.WARNING:at#for_ip_interrupt_entry/0x
         %stddev     %change         %stddev
             \          |                \  
     75.95            -4.1%      72.83        filebench.sum_bytes_mb/s
      4557            -4.1%       4371        filebench.sum_operations
     75.94            -4.1%      72.85        filebench.sum_operations/s
     76.00            -3.9%      73.00        filebench.sum_reads/s
     13.17            +4.2%      13.72        filebench.sum_time_ms/op
   9332769            -4.1%    8952523        filebench.time.file_system_inputs
    162436 ±  6%     -11.8%     143321        meminfo.DirectMap4k
     44022            -4.4%      42098        vmstat.io.bi
      3125 ±  5%      -6.3%       2928 ±  5%  slabinfo.filp.active_objs
      7419 ±  3%      -8.0%       6825 ±  3%  slabinfo.kmalloc-8.active_objs
      7419 ±  3%      -8.0%       6825 ±  3%  slabinfo.kmalloc-8.num_objs
     18.34            +2.1%      18.73        iostat.cpu.iowait
      1.46           -37.9%       0.90        iostat.sda.avgqu-sz
     44025            -4.4%      42098        iostat.sda.rkB/s
      1.05            -3.2%       1.01        iostat.sda.svctm
     59194 ±  7%     -14.2%      50811 ±  8%  softirqs.CPU1.TIMER
     32902 ±  5%      -7.0%      30592 ±  6%  softirqs.CPU2.RCU
     12304 ± 11%     +32.4%      16287 ±  2%  softirqs.CPU2.SCHED
     54266 ±  5%      -8.7%      49545 ±  5%  softirqs.CPU2.TIMER
     30586 ±  5%      -6.0%      28753 ±  2%  softirqs.CPU6.RCU
 1.057e+08 ±  5%     -27.1%   77008220 ±  5%  cpuidle.C1E.time
   1127971           -11.1%    1002432 ±  2%  cpuidle.C1E.usage
 2.882e+08 ±  5%    +122.2%  6.403e+08 ± 15%  cpuidle.C3.time
    794565           +73.5%    1378623 ±  9%  cpuidle.C3.usage
   6567827 ± 68%   +1171.0%   83478482 ±114%  cpuidle.C6.time
     10816 ± 64%    +938.8%     112360 ±107%  cpuidle.C6.usage
 4.024e+08 ±  7%     -98.2%    7380622 ± 74%  cpuidle.C7s.time
    577131           -98.2%      10454 ± 66%  cpuidle.C7s.usage
    132590            +3.5%     137195        proc-vmstat.nr_dirty
   1080688            +1.3%    1095121        proc-vmstat.nr_file_pages
     18355            -5.9%      17263 ±  3%  proc-vmstat.nr_free_cma
    831702            -1.8%     817002        proc-vmstat.nr_free_pages
    799686            +1.8%     814042        proc-vmstat.nr_inactive_file
      3957            +2.4%       4050        proc-vmstat.nr_kernel_stack
    799690            +1.8%     814044        proc-vmstat.nr_zone_inactive_file
    139862            +3.2%     144400        proc-vmstat.nr_zone_write_pending
   1397097 ±  5%      -4.1%    1340270 ±  7%  proc-vmstat.pgfree
   4666401            -4.1%    4476278        proc-vmstat.pgpgin
    104.50            -8.8%      95.33 ±  6%  turbostat.Avg_MHz
      3.14 ±  2%      -0.3        2.86 ±  3%  turbostat.Busy%
   1127727           -11.1%    1002183 ±  2%  turbostat.C1E
     12.67 ±  5%      -3.5        9.19 ±  5%  turbostat.C1E%
    794359           +73.5%    1378587 ±  9%  turbostat.C3
     34.56 ±  6%     +41.9       76.42 ± 15%  turbostat.C3%
     10756 ± 65%    +944.5%     112347 ±107%  turbostat.C6
      0.78 ± 69%      +9.2       10.01 ±115%  turbostat.C6%
    577213           -98.2%      10268 ± 67%  turbostat.C7s
     48.25 ±  7%     -47.4        0.87 ± 76%  turbostat.C7s%
     28.55 ±  5%     -19.7%      22.92 ±  3%  turbostat.CPU%c1
     35.85           +96.3%      70.39 ±  6%  turbostat.CPU%c3
     31.56 ±  8%     -99.5%       0.17 ± 92%  turbostat.CPU%c7
     12.76            -4.1%      12.23        turbostat.PkgWatt
    434484 ± 25%     -32.5%     293281 ±  9%  interrupts.CPU0.LOC:Local_timer_interrupts
    620.50 ± 24%     -64.3%     221.33 ± 15%  interrupts.CPU0.RES:Rescheduling_interrupts
    423916 ± 18%     -24.0%     322268 ± 23%  interrupts.CPU2.LOC:Local_timer_interrupts
    575.00 ± 67%     -71.1%     166.33 ±106%  interrupts.CPU2.RES:Rescheduling_interrupts
    276350 ± 23%     +83.1%     506099 ± 34%  interrupts.CPU3.LOC:Local_timer_interrupts
    247.50 ±  3%     +55.6%     385.00 ±  7%  interrupts.CPU3.NMI:Non-maskable_interrupts
    247.50 ±  3%     +55.6%     385.00 ±  7%  interrupts.CPU3.PMI:Performance_monitoring_interrupts
    430.00           -41.9%     250.00 ± 29%  interrupts.CPU3.RES:Rescheduling_interrupts
    288708 ± 18%      -9.5%     261253 ± 17%  interrupts.CPU5.LOC:Local_timer_interrupts
    506.00 ± 52%     -57.6%     214.67 ± 60%  interrupts.CPU5.RES:Rescheduling_interrupts
    212556           +26.4%     268764 ± 16%  interrupts.CPU6.LOC:Local_timer_interrupts
    121.00 ± 24%    +223.1%     391.00 ± 45%  interrupts.CPU6.RES:Rescheduling_interrupts
      1051 ±  7%     -28.4%     752.67 ± 24%  interrupts.CPU7.CAL:Function_call_interrupts
    166.50 ±  8%     +58.6%     264.00 ±  2%  interrupts.CPU7.NMI:Non-maskable_interrupts
    166.50 ±  8%     +58.6%     264.00 ±  2%  interrupts.CPU7.PMI:Performance_monitoring_interrupts
      1621           +20.1%       1947 ± 10%  interrupts.NMI:Non-maskable_interrupts
      1621           +20.1%       1947 ± 10%  interrupts.PMI:Performance_monitoring_interrupts
      3323           -39.7%       2004 ± 26%  interrupts.RES:Rescheduling_interrupts
     98.93 ± 99%    +139.4%     236.87 ± 11%  sched_debug.cfs_rq:/.exec_clock.min
      1157 ± 25%     -32.8%     777.88 ± 15%  sched_debug.cfs_rq:/.load_avg.avg
      3709 ± 21%     -39.3%       2252 ± 12%  sched_debug.cfs_rq:/.load_avg.max
      1164 ± 11%     -29.0%     826.83 ± 18%  sched_debug.cfs_rq:/.load_avg.stddev
      3315 ±  3%     +19.4%       3959 ±  4%  sched_debug.cfs_rq:/.min_vruntime.min
    609.00 ± 32%     -41.8%     354.50 ± 10%  sched_debug.cfs_rq:/.util_avg.min
    149.59 ±  9%     +54.4%     230.96 ±  9%  sched_debug.cfs_rq:/.util_avg.stddev
     70559 ± 28%     -33.1%      47195 ± 49%  sched_debug.cpu.avg_idle.min
    311669 ±  3%      -7.6%     288062 ±  6%  sched_debug.cpu.avg_idle.stddev
    778.66 ± 17%     -20.9%     616.04 ±  8%  sched_debug.cpu.curr->pid.avg
      1.59 ± 17%     -47.7%       0.83 ± 21%  sched_debug.cpu.nr_running.avg
      3.50 ± 14%     -42.9%       2.00 ± 20%  sched_debug.cpu.nr_running.max
      1.06 ±  9%     -37.7%       0.66 ± 21%  sched_debug.cpu.nr_running.stddev
      8410 ± 78%    +382.4%      40568 ± 38%  sched_debug.cpu.nr_switches.min
      5158 ±100%    +590.5%      35618 ± 42%  sched_debug.cpu.sched_count.min
      2495 ±100%    +612.5%      17778 ± 42%  sched_debug.cpu.sched_goidle.min
      3940 ±100%    +349.8%      17726 ± 35%  sched_debug.cpu.ttwu_count.min
      2288 ±100%    +593.7%      15877 ± 43%  sched_debug.cpu.ttwu_local.min
     28.82 ±  2%     -12.1%      25.33 ±  4%  perf-stat.i.MPKI
      3.44 ±  2%      -1.4        2.02 ± 16%  perf-stat.i.branch-miss-rate%
   3203155           -33.2%    2140684 ± 35%  perf-stat.i.branch-misses
     10.17            -0.9        9.28 ±  2%  perf-stat.i.cache-miss-rate%
    698479           -12.8%     609001 ±  5%  perf-stat.i.cache-misses
   6854696            -6.9%    6382981 ±  4%  perf-stat.i.cache-references
      2.81 ±  2%     -10.4%       2.52 ±  2%  perf-stat.i.cpi
 8.278e+08            -7.9%  7.621e+08 ±  6%  perf-stat.i.cpu-cycles
      6.09 ±  4%     +62.2%       9.88 ± 23%  perf-stat.i.cpu-migrations
      1215           +11.8%       1359 ±  2%  perf-stat.i.cycles-between-cache-misses
    159444 ±  2%      -3.8%     153313 ±  2%  perf-stat.i.iTLB-load-misses
    194692            -2.9%     189082 ±  2%  perf-stat.i.iTLB-loads
      0.42            +6.9%       0.45 ±  2%  perf-stat.i.ipc
    196601 ±  3%      -6.9%     183066 ±  6%  perf-stat.i.node-loads
      3.66 ±  2%      -1.2        2.49 ± 26%  perf-stat.overall.branch-miss-rate%
     10.19            -0.7        9.54 ±  2%  perf-stat.overall.cache-miss-rate%
      1185            +5.5%       1251        perf-stat.overall.cycles-between-cache-misses
   3177943           -33.3%    2120771 ± 35%  perf-stat.ps.branch-misses
    691895           -12.8%     603163 ±  5%  perf-stat.ps.cache-misses
   6789018            -6.9%    6321639 ±  4%  perf-stat.ps.cache-references
 8.201e+08            -8.0%  7.549e+08 ±  6%  perf-stat.ps.cpu-cycles
      6.03 ±  4%     +62.2%       9.78 ± 23%  perf-stat.ps.cpu-migrations
    157910 ±  2%      -3.8%     151843 ±  2%  perf-stat.ps.iTLB-load-misses
    192827            -2.9%     187263 ±  2%  perf-stat.ps.iTLB-loads
    194815 ±  3%      -6.9%     181321 ±  6%  perf-stat.ps.node-loads
     23.48 ±  6%      -5.8       17.70 ±  3%  perf-profile.calltrace.cycles-pp.apic_timer_interrupt.cpuidle_enter_state.cpuidle_enter.do_idle.cpu_startup_entry
     13.07 ±  6%      -4.4        8.68 ± 35%  perf-profile.calltrace.cycles-pp.cpu_startup_entry.start_kernel.secondary_startup_64
     13.07 ±  6%      -4.4        8.68 ± 35%  perf-profile.calltrace.cycles-pp.do_idle.cpu_startup_entry.start_kernel.secondary_startup_64
     13.07 ±  6%      -4.4        8.68 ± 35%  perf-profile.calltrace.cycles-pp.start_kernel.secondary_startup_64
     11.43            -4.4        7.04 ±  7%  perf-profile.calltrace.cycles-pp.hrtimer_interrupt.smp_apic_timer_interrupt.apic_timer_interrupt.cpuidle_enter_state.cpuidle_enter
     11.85 ±  5%      -3.9        7.94 ± 40%  perf-profile.calltrace.cycles-pp.cpuidle_enter.do_idle.cpu_startup_entry.start_kernel.secondary_startup_64
      8.30 ± 11%      -3.8        4.49 ± 12%  perf-profile.calltrace.cycles-pp.__hrtimer_run_queues.hrtimer_interrupt.smp_apic_timer_interrupt.apic_timer_interrupt.cpuidle_enter_state
     19.09 ±  5%      -3.8       15.32 ±  5%  perf-profile.calltrace.cycles-pp.smp_apic_timer_interrupt.apic_timer_interrupt.cpuidle_enter_state.cpuidle_enter.do_idle
     11.80 ± 15%      -2.8        9.05 ±  5%  perf-profile.calltrace.cycles-pp.irq_work_run_list.irq_work_run.smp_irq_work_interrupt.irq_work_interrupt.cpuidle_enter_state
     11.80 ± 15%      -2.8        9.05 ±  5%  perf-profile.calltrace.cycles-pp.irq_work_interrupt.cpuidle_enter_state.cpuidle_enter.do_idle.cpu_startup_entry
     11.80 ± 15%      -2.8        9.05 ±  5%  perf-profile.calltrace.cycles-pp.smp_irq_work_interrupt.irq_work_interrupt.cpuidle_enter_state.cpuidle_enter.do_idle
     11.80 ± 15%      -2.8        9.05 ±  5%  perf-profile.calltrace.cycles-pp.irq_work_run.smp_irq_work_interrupt.irq_work_interrupt.cpuidle_enter_state.cpuidle_enter
     12.91 ±  5%      -2.5       10.41 ± 11%  perf-profile.calltrace.cycles-pp.console_unlock.vprintk_emit.printk.irq_work_run_list.irq_work_run
     12.91 ±  5%      -2.5       10.41 ± 11%  perf-profile.calltrace.cycles-pp.vprintk_emit.printk.irq_work_run_list.irq_work_run.smp_irq_work_interrupt
     12.91 ±  5%      -2.5       10.41 ± 11%  perf-profile.calltrace.cycles-pp.printk.irq_work_run_list.irq_work_run.smp_irq_work_interrupt.irq_work_interrupt
     12.41 ±  5%      -2.5        9.92 ± 11%  perf-profile.calltrace.cycles-pp.serial8250_console_write.console_unlock.vprintk_emit.printk.irq_work_run_list
     12.41 ±  5%      -2.5        9.92 ± 11%  perf-profile.calltrace.cycles-pp.uart_console_write.serial8250_console_write.console_unlock.vprintk_emit.printk
     13.82 ±  7%      -2.4       11.38 ± 19%  perf-profile.calltrace.cycles-pp.wait_for_xmitr.serial8250_console_putchar.uart_console_write.serial8250_console_write.console_unlock
     12.93            -2.0       10.98 ± 16%  perf-profile.calltrace.cycles-pp.serial8250_console_putchar.uart_console_write.serial8250_console_write.console_unlock.vprintk_emit
      9.78 ±  7%      -1.6        8.14 ± 19%  perf-profile.calltrace.cycles-pp.io_serial_in.wait_for_xmitr.serial8250_console_putchar.uart_console_write.serial8250_console_write
      3.68 ± 28%      -1.2        2.46 ± 10%  perf-profile.calltrace.cycles-pp.tick_sched_timer.__hrtimer_run_queues.hrtimer_interrupt.smp_apic_timer_interrupt.apic_timer_interrupt
      2.80 ± 20%      -0.7        2.06 ±  9%  perf-profile.calltrace.cycles-pp.tick_sched_handle.tick_sched_timer.__hrtimer_run_queues.hrtimer_interrupt.smp_apic_timer_interrupt
      3.58 ± 13%      -0.7        2.85 ± 17%  perf-profile.calltrace.cycles-pp.delay_tsc.wait_for_xmitr.serial8250_console_putchar.uart_console_write.serial8250_console_write
      2.62 ± 21%      -0.7        1.90 ± 10%  perf-profile.calltrace.cycles-pp.update_process_times.tick_sched_handle.tick_sched_timer.__hrtimer_run_queues.hrtimer_interrupt
      1.29 ±  5%      -0.4        0.92 ± 18%  perf-profile.calltrace.cycles-pp.irq_enter.smp_apic_timer_interrupt.apic_timer_interrupt.cpuidle_enter_state.cpuidle_enter
      1.03 ± 13%      -0.3        0.76 ± 19%  perf-profile.calltrace.cycles-pp.tick_irq_enter.irq_enter.smp_apic_timer_interrupt.apic_timer_interrupt.cpuidle_enter_state
      2.63            -0.2        2.43        perf-profile.calltrace.cycles-pp.tick_nohz_get_sleep_length.menu_select.do_idle.cpu_startup_entry.start_secondary
      0.66 ±  6%      +0.4        1.04 ± 11%  perf-profile.calltrace.cycles-pp.load_balance.rebalance_domains.__softirqentry_text_start.irq_exit.smp_apic_timer_interrupt
      0.28 ±100%      +0.5        0.79 ± 19%  perf-profile.calltrace.cycles-pp.rcu_core.__softirqentry_text_start.irq_exit.smp_apic_timer_interrupt.apic_timer_interrupt
      0.76 ±  9%      +0.6        1.31 ± 13%  perf-profile.calltrace.cycles-pp.rebalance_domains.__softirqentry_text_start.irq_exit.smp_apic_timer_interrupt.apic_timer_interrupt
      0.95            +0.9        1.85 ± 19%  perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.unlinkat
      0.95            +0.9        1.85 ± 19%  perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.unlinkat
      0.95            +0.9        1.85 ± 19%  perf-profile.calltrace.cycles-pp.unlinkat
      0.95            +0.9        1.85 ± 19%  perf-profile.calltrace.cycles-pp.do_unlinkat.do_syscall_64.entry_SYSCALL_64_after_hwframe.unlinkat
      0.95            +0.9        1.85 ± 19%  perf-profile.calltrace.cycles-pp.evict.do_unlinkat.do_syscall_64.entry_SYSCALL_64_after_hwframe.unlinkat
      0.91 ±  2%      +0.9        1.85 ± 19%  perf-profile.calltrace.cycles-pp.truncate_inode_pages_range.evict.do_unlinkat.do_syscall_64.entry_SYSCALL_64_after_hwframe
     49.55            +8.9       58.46 ±  3%  perf-profile.calltrace.cycles-pp.intel_idle.cpuidle_enter_state.cpuidle_enter.do_idle.cpu_startup_entry
     21.98 ±  5%      -5.3       16.66 ±  4%  perf-profile.children.cycles-pp.apic_timer_interrupt
     13.07 ±  6%      -4.4        8.68 ± 35%  perf-profile.children.cycles-pp.start_kernel
     11.46            -4.2        7.28 ±  8%  perf-profile.children.cycles-pp.hrtimer_interrupt
     19.12 ±  4%      -3.8       15.34 ±  5%  perf-profile.children.cycles-pp.smp_apic_timer_interrupt
      8.37 ± 11%      -3.6        4.80 ± 10%  perf-profile.children.cycles-pp.__hrtimer_run_queues
     13.87 ± 12%      -3.0       10.89 ± 16%  perf-profile.children.cycles-pp.irq_work_run_list
     13.95 ±  7%      -2.6       11.39 ± 19%  perf-profile.children.cycles-pp.serial8250_console_putchar
     14.04 ±  7%      -2.5       11.50 ± 20%  perf-profile.children.cycles-pp.wait_for_xmitr
     14.11 ±  8%      -2.5       11.60 ± 19%  perf-profile.children.cycles-pp.uart_console_write
     12.91 ±  5%      -2.5       10.41 ± 11%  perf-profile.children.cycles-pp.irq_work_interrupt
     12.91 ±  5%      -2.5       10.41 ± 11%  perf-profile.children.cycles-pp.smp_irq_work_interrupt
     12.91 ±  5%      -2.5       10.41 ± 11%  perf-profile.children.cycles-pp.irq_work_run
     12.91 ±  5%      -2.5       10.41 ± 11%  perf-profile.children.cycles-pp.printk
     14.20 ±  8%      -2.5       11.71 ± 20%  perf-profile.children.cycles-pp.serial8250_console_write
     14.79 ±  7%      -2.4       12.42 ± 20%  perf-profile.children.cycles-pp.console_unlock
     13.87            -1.9       11.95 ± 17%  perf-profile.children.cycles-pp.vprintk_emit
      9.93 ±  7%      -1.7        8.18 ± 19%  perf-profile.children.cycles-pp.io_serial_in
      3.97 ± 22%      -1.4        2.60 ± 10%  perf-profile.children.cycles-pp.tick_sched_timer
      1.51 ±  5%      -1.0        0.53 ± 33%  perf-profile.children.cycles-pp.__remove_hrtimer
      1.38 ± 10%      -1.0        0.43 ± 28%  perf-profile.children.cycles-pp.timerqueue_del
      3.12 ± 22%      -0.9        2.18 ±  9%  perf-profile.children.cycles-pp.tick_sched_handle
      2.87 ± 26%      -0.9        2.01 ±  9%  perf-profile.children.cycles-pp.update_process_times
      1.16            -0.8        0.32 ± 42%  perf-profile.children.cycles-pp.interrupt_entry
      6.37 ±  3%      -0.8        5.55 ±  7%  perf-profile.children.cycles-pp.menu_select
      0.66 ± 12%      -0.6        0.09 ± 72%  perf-profile.children.cycles-pp.rb_next
      2.52 ±  4%      -0.5        2.01 ± 13%  perf-profile.children.cycles-pp.tick_nohz_next_event
      3.10 ±  3%      -0.5        2.61 ±  5%  perf-profile.children.cycles-pp.tick_nohz_get_sleep_length
      1.50            -0.5        1.04 ± 18%  perf-profile.children.cycles-pp.irq_enter
      0.79            -0.4        0.37 ± 22%  perf-profile.children.cycles-pp.enqueue_hrtimer
      1.23 ±  6%      -0.4        0.84 ± 24%  perf-profile.children.cycles-pp.tick_irq_enter
      0.73 ±  6%      -0.4        0.35 ± 16%  perf-profile.children.cycles-pp.timerqueue_add
      0.47 ± 19%      -0.4        0.10 ± 37%  perf-profile.children.cycles-pp.tick_check_oneshot_broadcast_this_cpu
      0.91 ± 12%      -0.4        0.55 ± 41%  perf-profile.children.cycles-pp.native_write_msr
      0.53 ± 28%      -0.3        0.20 ± 32%  perf-profile.children.cycles-pp.rcu_sched_clock_irq
      0.50 ± 13%      -0.3        0.22 ± 14%  perf-profile.children.cycles-pp.tick_nohz_irq_exit
      1.25 ± 11%      -0.3        0.99 ±  9%  perf-profile.children.cycles-pp.perf_mux_hrtimer_handler
      1.60 ±  4%      -0.3        1.34 ± 20%  perf-profile.children.cycles-pp.get_next_timer_interrupt
      0.38 ± 18%      -0.2        0.15 ± 46%  perf-profile.children.cycles-pp.__intel_pmu_enable_all
      0.81 ±  5%      -0.2        0.59 ± 10%  perf-profile.children.cycles-pp.sched_clock_cpu
      0.34 ± 10%      -0.2        0.14 ± 26%  perf-profile.children.cycles-pp.irq_work_needs_cpu
      0.44 ± 56%      -0.2        0.24 ± 71%  perf-profile.children.cycles-pp._raw_spin_unlock_irqrestore
      0.42 ± 54%      -0.2        0.22 ± 27%  perf-profile.children.cycles-pp.ret_from_fork
      0.25 ± 52%      -0.2        0.07 ± 25%  perf-profile.children.cycles-pp.rcu_needs_cpu
      0.44 ± 15%      -0.2        0.27 ± 34%  perf-profile.children.cycles-pp.tick_sched_do_timer
      0.34 ± 10%      -0.2        0.18 ± 48%  perf-profile.children.cycles-pp.__x86_indirect_thunk_rax
      0.56 ±  9%      -0.1        0.42 ± 19%  perf-profile.children.cycles-pp.native_sched_clock
      0.22 ± 45%      -0.1        0.08 ± 20%  perf-profile.children.cycles-pp.account_process_tick
      0.26            -0.1        0.12 ± 32%  perf-profile.children.cycles-pp._raw_spin_lock_irq
      1.20 ± 12%      -0.1        1.06 ± 12%  perf-profile.children.cycles-pp.ktime_get
      0.56 ±  9%      -0.1        0.44 ± 14%  perf-profile.children.cycles-pp.sched_clock
      0.25 ± 52%      -0.1        0.13 ± 91%  perf-profile.children.cycles-pp.idle_cpu
      0.34 ± 10%      -0.1        0.23 ± 29%  perf-profile.children.cycles-pp.timekeeping_advance
      0.22 ± 40%      -0.1        0.11 ±110%  perf-profile.children.cycles-pp.do_IRQ
      0.44 ± 15%      -0.1        0.34 ±  9%  perf-profile.children.cycles-pp.try_to_wake_up
      0.25 ± 52%      -0.1        0.16 ± 91%  perf-profile.children.cycles-pp.__x64_sys_exit_group
      0.25 ± 52%      -0.1        0.16 ± 91%  perf-profile.children.cycles-pp.do_group_exit
      0.25 ± 52%      -0.1        0.16 ± 91%  perf-profile.children.cycles-pp.do_exit
      0.22 ± 45%      -0.1        0.13 ± 55%  perf-profile.children.cycles-pp.rcu_irq_exit
      0.12 ±  4%      -0.1        0.04 ± 71%  perf-profile.children.cycles-pp.hrtimer_forward
      0.19            -0.1        0.10 ± 40%  perf-profile.children.cycles-pp.__update_load_avg_cfs_rq
      0.12 ± 52%      -0.1        0.05 ± 81%  perf-profile.children.cycles-pp.rcu_qs
      0.12 ± 52%      -0.1        0.05 ±  8%  perf-profile.children.cycles-pp.__hrtimer_get_next_event
      0.12 ± 52%      -0.1        0.06 ± 86%  perf-profile.children.cycles-pp.swapgs_restore_regs_and_return_to_usermode
      0.12 ±  4%      -0.1        0.07 ± 25%  perf-profile.children.cycles-pp.unwind_next_frame
      0.09 ± 33%      -0.1        0.03 ± 70%  perf-profile.children.cycles-pp.pipe_write
      0.09 ± 33%      -0.1        0.04 ± 71%  perf-profile.children.cycles-pp.prepare_exit_to_usermode
      0.15 ± 22%      -0.1        0.10 ± 40%  perf-profile.children.cycles-pp.swake_up_one
      0.15 ± 22%      -0.1        0.10 ± 40%  perf-profile.children.cycles-pp.swake_up_locked
      0.19 ± 36%      -0.0        0.14 ± 47%  perf-profile.children.cycles-pp.pm_qos_read_value
      0.10 ± 36%      -0.0        0.05 ±  8%  perf-profile.children.cycles-pp.truncate_cleanup_page
      0.10 ± 36%      -0.0        0.06 ± 86%  perf-profile.children.cycles-pp.do_sys_open
      0.06            -0.0        0.03 ± 70%  perf-profile.children.cycles-pp.handle_fasteoi_irq
      0.06            -0.0        0.03 ± 70%  perf-profile.children.cycles-pp.handle_irq_event
      0.06            -0.0        0.03 ± 70%  perf-profile.children.cycles-pp.handle_irq_event_percpu
      0.12 ±  4%      +0.1        0.21 ± 11%  perf-profile.children.cycles-pp.call_timer_fn
      0.00            +0.1        0.09 ± 32%  perf-profile.children.cycles-pp.__x64_sys_clone
      0.00            +0.1        0.09 ± 32%  perf-profile.children.cycles-pp._do_fork
      0.00            +0.1        0.09 ± 32%  perf-profile.children.cycles-pp.copy_process
      0.00            +0.1        0.09 ± 65%  perf-profile.children.cycles-pp.workingset_update_node
      0.12 ±  4%      +0.1        0.22 ± 14%  perf-profile.children.cycles-pp.leave_mm
      0.10 ± 36%      +0.1        0.20 ± 21%  perf-profile.children.cycles-pp.xas_init_marks
      0.03 ±100%      +0.1        0.14 ± 11%  perf-profile.children.cycles-pp.timekeeping_max_deferment
      0.10 ± 36%      +0.1        0.21 ± 11%  perf-profile.children.cycles-pp.pagevec_lookup_entries
      0.06            +0.1        0.20 ± 39%  perf-profile.children.cycles-pp.rcu_gp_kthread
      0.12 ±  4%      +0.2        0.32 ± 18%  perf-profile.children.cycles-pp.native_queued_spin_lock_slowpath
      0.10 ± 36%      +0.2        0.30 ± 42%  perf-profile.children.cycles-pp.__delete_from_page_cache
      0.47 ±  8%      +0.3        0.76 ± 28%  perf-profile.children.cycles-pp.find_busiest_group
      0.23 ± 15%      +0.3        0.52 ± 34%  perf-profile.children.cycles-pp.xas_store
      0.15 ± 22%      +0.3        0.46 ± 30%  perf-profile.children.cycles-pp.pm_qos_request
      0.34 ± 10%      +0.3        0.67 ± 22%  perf-profile.children.cycles-pp.delete_from_page_cache_batch
      0.84 ±  5%      +0.4        1.20 ± 16%  perf-profile.children.cycles-pp.load_balance
      0.00            +0.4        0.43 ± 43%  perf-profile.children.cycles-pp.blk_mq_try_issue_list_directly
      0.00            +0.4        0.43 ± 43%  perf-profile.children.cycles-pp.blk_mq_request_issue_directly
      0.00            +0.4        0.43 ± 43%  perf-profile.children.cycles-pp.__blk_mq_try_issue_directly
      0.41 ±  9%      +0.6        1.01 ± 11%  perf-profile.children.cycles-pp._raw_spin_lock_irqsave
      0.94 ±  8%      +0.6        1.56 ± 14%  perf-profile.children.cycles-pp.rebalance_domains
      1.07 ± 13%      +0.9        1.95 ± 19%  perf-profile.children.cycles-pp.evict
      0.95            +0.9        1.85 ± 19%  perf-profile.children.cycles-pp.truncate_inode_pages_range
      0.95            +0.9        1.85 ± 19%  perf-profile.children.cycles-pp.unlinkat
      0.95            +0.9        1.85 ± 19%  perf-profile.children.cycles-pp.do_unlinkat
     49.55            +8.9       58.49 ±  3%  perf-profile.children.cycles-pp.intel_idle
      9.93 ±  7%      -1.7        8.18 ± 19%  perf-profile.self.cycles-pp.io_serial_in
      2.56 ± 13%      -0.8        1.73 ±  4%  perf-profile.self.cycles-pp.cpuidle_enter_state
      1.06            -0.7        0.32 ± 42%  perf-profile.self.cycles-pp.interrupt_entry
      0.75 ± 24%      -0.7        0.06 ± 86%  perf-profile.self.cycles-pp.apic_timer_interrupt
      2.61 ±  7%      -0.6        2.03 ±  3%  perf-profile.self.cycles-pp.menu_select
      0.66 ± 12%      -0.6        0.09 ± 72%  perf-profile.self.cycles-pp.rb_next
      0.84 ± 24%      -0.4        0.43 ± 44%  perf-profile.self.cycles-pp.hrtimer_interrupt
      0.67 ± 15%      -0.4        0.28 ± 12%  perf-profile.self.cycles-pp.timerqueue_add
      0.91 ± 12%      -0.4        0.54 ± 38%  perf-profile.self.cycles-pp.native_write_msr
      0.47 ± 19%      -0.4        0.10 ± 37%  perf-profile.self.cycles-pp.tick_check_oneshot_broadcast_this_cpu
      0.57 ± 20%      -0.3        0.22 ± 32%  perf-profile.self.cycles-pp.__hrtimer_run_queues
      0.62 ± 18%      -0.2        0.38 ± 39%  perf-profile.self.cycles-pp.do_idle
      0.32 ± 17%      -0.2        0.09 ± 32%  perf-profile.self.cycles-pp.get_next_timer_interrupt
      0.38 ± 49%      -0.2        0.18 ± 48%  perf-profile.self.cycles-pp.perf_mux_hrtimer_handler
      0.32 ± 20%      -0.2        0.13 ± 43%  perf-profile.self.cycles-pp.irq_work_needs_cpu
      0.25 ± 52%      -0.2        0.07 ± 25%  perf-profile.self.cycles-pp.rcu_needs_cpu
      0.56 ±  9%      -0.2        0.38 ± 16%  perf-profile.self.cycles-pp.native_sched_clock
      0.32            -0.2        0.14 ± 54%  perf-profile.self.cycles-pp.__x86_indirect_thunk_rax
      0.41 ± 53%      -0.2        0.24 ± 71%  perf-profile.self.cycles-pp._raw_spin_unlock_irqrestore
      0.43 ± 12%      -0.2        0.28 ± 12%  perf-profile.self.cycles-pp.run_timer_softirq
      0.19 ± 31%      -0.1        0.05 ± 78%  perf-profile.self.cycles-pp.clockevents_program_event
      0.22 ± 45%      -0.1        0.08 ± 20%  perf-profile.self.cycles-pp.account_process_tick
      0.25 ± 52%      -0.1        0.13 ± 91%  perf-profile.self.cycles-pp.idle_cpu
      0.25 ± 24%      -0.1        0.13 ± 55%  perf-profile.self.cycles-pp.rcu_sched_clock_irq
      0.19 ± 36%      -0.1        0.08 ± 20%  perf-profile.self.cycles-pp.tick_sched_timer
      0.23 ± 15%      -0.1        0.12 ± 32%  perf-profile.self.cycles-pp._raw_spin_lock_irq
      0.26            -0.1        0.15 ± 16%  perf-profile.self.cycles-pp.sched_clock_cpu
      0.15 ± 22%      -0.1        0.06 ± 71%  perf-profile.self.cycles-pp.irq_exit
      0.22 ± 45%      -0.1        0.13 ± 55%  perf-profile.self.cycles-pp.rcu_irq_exit
      0.23 ± 15%      -0.1        0.14 ± 44%  perf-profile.self.cycles-pp.release_pages
      0.12 ±  4%      -0.1        0.04 ± 71%  perf-profile.self.cycles-pp.hrtimer_forward
      0.19            -0.1        0.10 ± 40%  perf-profile.self.cycles-pp.__update_load_avg_cfs_rq
      0.12 ± 52%      -0.1        0.05 ± 81%  perf-profile.self.cycles-pp.rcu_qs
      0.09 ± 33%      -0.1        0.04 ± 71%  perf-profile.self.cycles-pp.prepare_exit_to_usermode
      0.19 ± 36%      -0.0        0.14 ± 47%  perf-profile.self.cycles-pp.pm_qos_read_value
      0.10 ± 36%      -0.0        0.05 ±  8%  perf-profile.self.cycles-pp.run_local_timers
      0.00            +0.1        0.05 ±  8%  perf-profile.self.cycles-pp.sched_clock
      0.12 ±  4%      +0.1        0.19 ± 15%  perf-profile.self.cycles-pp.leave_mm
      0.00            +0.1        0.09 ± 65%  perf-profile.self.cycles-pp.workingset_update_node
      0.00            +0.1        0.14 ± 11%  perf-profile.self.cycles-pp.timekeeping_max_deferment
      0.12 ±  4%      +0.2        0.32 ± 18%  perf-profile.self.cycles-pp.native_queued_spin_lock_slowpath
      0.34 ± 10%      +0.2        0.57 ± 16%  perf-profile.self.cycles-pp.find_busiest_group
      0.06            +0.3        0.32 ± 75%  perf-profile.self.cycles-pp.rebalance_domains
      0.15 ± 22%      +0.3        0.45 ± 34%  perf-profile.self.cycles-pp.pm_qos_request
      0.29 ± 12%      +0.4        0.69 ± 22%  perf-profile.self.cycles-pp._raw_spin_lock_irqsave
     49.43            +8.9       58.30 ±  3%  perf-profile.self.cycles-pp.intel_idle



***************************************************************************************************
lkp-csl-2sp7: 96 threads Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz with 256G memory
=========================================================================================
compiler/cpufreq_governor/disk/filesize/fs/iterations/kconfig/nr_directories/nr_files_per_directory/nr_threads/rootfs/sync_method/tbox_group/test_size/testcase/ucode:
  gcc-7/performance/1SSD/8K/xfs/1x/x86_64-rhel-7.6/16d/256fpd/32t/debian-x86_64-20191114.cgz/fsyncBeforeClose/lkp-csl-2sp7/400M/fsmark/0x500002c

commit: 
  8186413d49 ("scsi: remove .for_blk_mq")
  d4ea3041f3 ("scsi: sd: register request queue after sd_revalidate_disk is done")

8186413d49989aa6 d4ea3041f3fd005c1be676f6603 
---------------- --------------------------- 
         %stddev     %change         %stddev
             \          |                \  
     17532           -11.2%      15568        fsmark.files_per_sec
     21452 ±  4%     +34.3%      28806        fsmark.time.involuntary_context_switches
    220.50            -9.6%     199.33        fsmark.time.percent_of_cpu_this_job_got
    287664            +1.1%     290828        fsmark.time.voluntary_context_switches
      1231            +5.4%       1298        proc-vmstat.nr_page_table_pages
    142907 ±  6%     +21.4%     173537 ±  5%  turbostat.C1
    143727 ±  6%     +21.3%     174368 ±  5%  cpuidle.C1.usage
    125251 ±  9%     +25.4%     157022 ± 15%  cpuidle.POLL.time
     24561 ± 10%     +27.3%      31275 ± 17%  cpuidle.POLL.usage
    176229 ±  8%     -24.8%     132584 ± 25%  numa-meminfo.node1.Active
    175236 ±  7%     -25.1%     131325 ± 25%  numa-meminfo.node1.Active(anon)
    174944 ±  7%     -25.1%     130998 ± 25%  numa-meminfo.node1.AnonPages
     43836 ±  7%     -25.1%      32847 ± 25%  numa-vmstat.node1.nr_active_anon
     43761 ±  7%     -25.1%      32756 ± 25%  numa-vmstat.node1.nr_anon_pages
     43836 ±  7%     -25.1%      32847 ± 25%  numa-vmstat.node1.nr_zone_active_anon
     21491 ± 14%     -29.9%      15061 ± 26%  sched_debug.cfs_rq:/.min_vruntime.avg
    485.50 ±  9%     -52.9%     228.67 ± 77%  sched_debug.cfs_rq:/.removed.util_avg.max
     22518 ± 88%     -91.6%       1896 ±  2%  sched_debug.cpu.avg_idle.min
     34973           +19.9%      41935        softirqs.BLOCK
      8249 ±  2%     +29.3%      10668 ±  4%  softirqs.CPU10.TIMER
     10384 ± 21%     -17.7%       8545 ±  4%  softirqs.CPU28.TIMER
      9712 ± 10%     -13.5%       8400 ±  5%  softirqs.CPU8.TIMER
     96.09            -0.7%      95.40        iostat.cpu.idle
      6.06 ± 16%     -60.6%       2.38 ± 71%  iostat.sdb.avgqu-sz
     10.43 ± 63%     -85.6%       1.50 ± 80%  iostat.sdb.await.max
     96.57 ± 67%     -88.9%      10.71 ± 88%  iostat.sdb.r_await.max
      1436 ± 16%     -93.4%      95.39 ± 71%  iostat.sdb.wrqm/s
      4368 ±  7%     -18.5%       3560 ± 19%  slabinfo.eventpoll_pwq.active_objs
      4368 ±  7%     -18.5%       3560 ± 19%  slabinfo.eventpoll_pwq.num_objs
    752.00 ± 10%     -12.0%     661.67 ± 11%  slabinfo.file_lock_cache.active_objs
    752.00 ± 10%     -12.0%     661.67 ± 11%  slabinfo.file_lock_cache.num_objs
      7200 ±  3%      -9.9%       6487        slabinfo.kmalloc-1k.active_objs
      7251 ±  3%      -9.3%       6574        slabinfo.kmalloc-1k.num_objs
      9913 ±  4%      -4.8%       9436 ±  4%  slabinfo.kmalloc-512.active_objs
      9983 ±  4%      -4.8%       9505 ±  5%  slabinfo.kmalloc-512.num_objs
    924.25 ±  8%     -13.9%     796.00 ±  2%  slabinfo.mnt_cache.active_objs
    924.25 ±  8%     -13.9%     796.00 ±  2%  slabinfo.mnt_cache.num_objs
    640.00           -20.0%     512.00        slabinfo.scsi_sense_cache.active_objs
    640.00           -20.0%     512.00        slabinfo.scsi_sense_cache.num_objs
    687.00 ± 19%     -24.6%     518.33 ±  5%  slabinfo.skbuff_fclone_cache.active_objs
    687.00 ± 19%     -24.6%     518.33 ±  5%  slabinfo.skbuff_fclone_cache.num_objs
      8.56 ±  5%     +17.9%      10.10 ±  3%  perf-stat.i.MPKI
     10.90 ± 26%      -5.1        5.75 ± 12%  perf-stat.i.cache-miss-rate%
  74744370 ±  7%     +19.1%   89057238 ±  4%  perf-stat.i.cache-references
    193622 ±  6%     +14.4%     221455 ±  5%  perf-stat.i.context-switches
      1731 ±109%    +227.2%       5664 ± 46%  perf-stat.i.cpu-migrations
      0.04 ± 20%      +0.0        0.05 ± 17%  perf-stat.i.dTLB-load-miss-rate%
    690430 ± 34%     +53.8%    1061755 ± 23%  perf-stat.i.dTLB-load-misses
      0.02 ±  6%      +0.0        0.02 ± 13%  perf-stat.i.dTLB-store-miss-rate%
     97896 ± 11%     +29.2%     126471 ± 10%  perf-stat.i.dTLB-store-misses
     43.10            +0.7       43.79        perf-stat.i.iTLB-load-miss-rate%
   3798429 ±  4%      +7.9%    4100118 ±  3%  perf-stat.i.iTLB-load-misses
      8.97 ±  7%     +18.2%      10.61 ±  5%  perf-stat.overall.MPKI
      5.65 ± 15%      -1.6        4.03 ±  7%  perf-stat.overall.cache-miss-rate%
      1.30 ±  2%      +4.1%       1.35 ±  3%  perf-stat.overall.cpi
      0.03 ± 32%      +0.0        0.05 ± 23%  perf-stat.overall.dTLB-load-miss-rate%
      0.01 ± 11%      +0.0        0.01 ± 16%  perf-stat.overall.dTLB-store-miss-rate%
     41.58            +0.6       42.14        perf-stat.overall.iTLB-load-miss-rate%
      0.77 ±  2%      -3.9%       0.74 ±  3%  perf-stat.overall.ipc
     75.22 ±  2%      +5.7       80.88 ±  2%  perf-stat.overall.node-load-miss-rate%
  60564747 ±  7%     +15.0%   69654648 ±  6%  perf-stat.ps.cache-references
      1431 ±110%    +208.5%       4415 ± 47%  perf-stat.ps.cpu-migrations
     79706 ± 12%     +24.1%      98906 ±  8%  perf-stat.ps.dTLB-store-misses
     10.62 ± 63%     -10.6        0.00        perf-profile.calltrace.cycles-pp.__libc_start_main
      7.27 ±113%      -7.3        0.00        perf-profile.calltrace.cycles-pp.page_fault
      6.07 ±103%      -6.1        0.00        perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.ioctl.__libc_start_main
      6.07 ±103%      -6.1        0.00        perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.ioctl.__libc_start_main
      6.07 ±103%      -6.1        0.00        perf-profile.calltrace.cycles-pp.smp_call_function_single.event_function_call.perf_event_for_each_child._perf_ioctl.perf_ioctl
      6.07 ±103%      -6.1        0.00        perf-profile.calltrace.cycles-pp.event_function_call.perf_event_for_each_child._perf_ioctl.perf_ioctl.ksys_ioctl
      6.07 ±103%      -6.1        0.00        perf-profile.calltrace.cycles-pp.__x64_sys_ioctl.do_syscall_64.entry_SYSCALL_64_after_hwframe.ioctl.__libc_start_main
      6.07 ±103%      -6.1        0.00        perf-profile.calltrace.cycles-pp._perf_ioctl.perf_ioctl.ksys_ioctl.__x64_sys_ioctl.do_syscall_64
      6.07 ±103%      -6.1        0.00        perf-profile.calltrace.cycles-pp.ksys_ioctl.__x64_sys_ioctl.do_syscall_64.entry_SYSCALL_64_after_hwframe.ioctl
      6.07 ±103%      -6.1        0.00        perf-profile.calltrace.cycles-pp.ioctl.__libc_start_main
      6.07 ±103%      -6.1        0.00        perf-profile.calltrace.cycles-pp.perf_ioctl.ksys_ioctl.__x64_sys_ioctl.do_syscall_64.entry_SYSCALL_64_after_hwframe
      6.07 ±103%      -6.1        0.00        perf-profile.calltrace.cycles-pp.perf_event_for_each_child._perf_ioctl.perf_ioctl.ksys_ioctl.__x64_sys_ioctl
      5.28 ±100%      -1.6        3.70 ±141%  perf-profile.calltrace.cycles-pp.setlocale
      8.34 ± 62%      -8.3        0.00        perf-profile.children.cycles-pp.__libc_start_main
      6.07 ±103%      -6.1        0.00        perf-profile.children.cycles-pp.__x64_sys_ioctl
      6.07 ±103%      -6.1        0.00        perf-profile.children.cycles-pp._perf_ioctl
      6.07 ±103%      -6.1        0.00        perf-profile.children.cycles-pp.ksys_ioctl
      6.07 ±103%      -6.1        0.00        perf-profile.children.cycles-pp.ioctl
      6.07 ±103%      -6.1        0.00        perf-profile.children.cycles-pp.perf_ioctl
      6.07 ±103%      -6.1        0.00        perf-profile.children.cycles-pp.perf_event_for_each_child
      8.57 ±102%      -4.9        3.70 ±141%  perf-profile.children.cycles-pp.event_function_call
      8.57 ±102%      -4.9        3.70 ±141%  perf-profile.children.cycles-pp.smp_call_function_single
      4.77 ±100%      -4.8        0.00        perf-profile.children.cycles-pp.page_fault
      5.28 ±100%      -1.6        3.70 ±141%  perf-profile.children.cycles-pp.setlocale
      4.77 ±100%      -1.1        3.70 ±141%  perf-profile.children.cycles-pp.link_path_walk
      5.05 ±100%      -0.9        4.17 ±141%  perf-profile.children.cycles-pp.mutex_lock
      8.57 ±102%      -4.9        3.70 ±141%  perf-profile.self.cycles-pp.smp_call_function_single
     48515 ± 57%     +62.1%      78654        interrupts.316:PCI-MSI.376832-edge.ahci[0000:00:17.0]
     19956 ± 54%     +95.5%      39006 ± 31%  interrupts.CAL:Function_call_interrupts
    535.00 ± 72%     -90.2%      52.67 ± 79%  interrupts.CPU2.RES:Rescheduling_interrupts
    290.25 ± 78%    +139.7%     695.67 ± 32%  interrupts.CPU24.CAL:Function_call_interrupts
    304.00 ± 76%    +189.3%     879.33 ± 42%  interrupts.CPU25.CAL:Function_call_interrupts
    315.50 ± 77%    +124.8%     709.33 ± 36%  interrupts.CPU26.CAL:Function_call_interrupts
    291.50 ± 72%    +150.5%     730.33 ± 44%  interrupts.CPU29.CAL:Function_call_interrupts
    407.75 ±134%     -95.0%      20.33 ± 85%  interrupts.CPU3.RES:Rescheduling_interrupts
    299.50 ± 68%    +147.4%     741.00 ± 44%  interrupts.CPU32.CAL:Function_call_interrupts
    337.00 ± 78%    +131.0%     778.33 ± 39%  interrupts.CPU33.CAL:Function_call_interrupts
    285.25 ± 71%    +249.6%     997.33 ± 42%  interrupts.CPU34.CAL:Function_call_interrupts
    268.25 ± 88%    +179.8%     750.67 ± 38%  interrupts.CPU37.CAL:Function_call_interrupts
    291.00 ± 75%    +133.1%     678.33 ± 36%  interrupts.CPU39.CAL:Function_call_interrupts
      7.50 ±119%    +740.0%      63.00 ± 86%  interrupts.CPU40.RES:Rescheduling_interrupts
    291.50 ± 74%    +156.6%     748.00 ± 33%  interrupts.CPU42.CAL:Function_call_interrupts
    279.50 ± 69%    +139.7%     670.00 ± 39%  interrupts.CPU47.CAL:Function_call_interrupts
    127.75 ±  9%     -13.6%     110.33 ±  2%  interrupts.CPU52.CAL:Function_call_interrupts
    127.50 ±  9%     -12.4%     111.67 ±  2%  interrupts.CPU53.CAL:Function_call_interrupts
    127.25 ±  9%     -13.0%     110.67 ±  2%  interrupts.CPU54.CAL:Function_call_interrupts
    127.75 ±  9%     -14.9%     108.67 ±  2%  interrupts.CPU56.CAL:Function_call_interrupts
    126.25 ±  5%      -8.1%     116.00 ±  5%  interrupts.CPU71.CAL:Function_call_interrupts
    264.00 ± 68%    +174.7%     725.33 ± 41%  interrupts.CPU72.CAL:Function_call_interrupts
    280.75 ± 70%    +157.5%     723.00 ± 39%  interrupts.CPU73.CAL:Function_call_interrupts
    253.50 ± 89%    +186.0%     725.00 ± 37%  interrupts.CPU76.CAL:Function_call_interrupts
     30.50 ± 59%   +3123.0%     983.00 ± 65%  interrupts.CPU8.RES:Rescheduling_interrupts
    306.25 ± 78%    +140.1%     735.33 ± 34%  interrupts.CPU82.CAL:Function_call_interrupts
    314.00 ± 78%    +123.6%     702.00 ± 35%  interrupts.CPU83.CAL:Function_call_interrupts
    311.00 ± 72%    +131.7%     720.67 ± 33%  interrupts.CPU84.CAL:Function_call_interrupts
    304.75 ± 78%    +133.1%     710.33 ± 33%  interrupts.CPU89.CAL:Function_call_interrupts





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,
Rong Chen


View attachment "config-5.5.0-rc4-00142-gd4ea3041f3fd0" of type "text/plain" (202322 bytes)

View attachment "job-script" of type "text/plain" (8219 bytes)

View attachment "job.yaml" of type "text/plain" (5800 bytes)

View attachment "reproduce" of type "text/plain" (1057 bytes)

Powered by blists - more mailing lists