All of lore.kernel.org
 help / color / mirror / Atom feed
From: Xing Zhengjun <zhengjun.xing@linux.intel.com>
To: Ritesh Harjani <riteshh@linux.ibm.com>,
	kernel test robot <rong.a.chen@intel.com>
Cc: Theodore Ts'o <tytso@mit.edu>, kbuild test robot <lkp@intel.com>,
	Jan Kara <jack@suse.cz>,
	"Darrick J. Wong" <darrick.wong@oracle.com>,
	LKML <linux-kernel@vger.kernel.org>,
	lkp@lists.01.org
Subject: Re: [LKP] [ext4] d3b6f23f71: stress-ng.fiemap.ops_per_sec -60.5% regression
Date: Tue, 16 Jun 2020 15:13:47 +0800	[thread overview]
Message-ID: <7bba55fd-955a-355e-4df1-7b4493b125a6@linux.intel.com> (raw)
In-Reply-To: <bf06a978-828a-9676-0ff8-fa47826d0bdd@linux.intel.com>

Hi Ritesh,

    I test, the regression still existed in v5.8-rc1. Do you have time 
to take a look at it? Thanks.

On 4/14/2020 1:49 PM, Xing Zhengjun wrote:
> Thanks for your quick response, if you need any more test information 
> about the regression, please let me known.
> 
> On 4/13/2020 6:56 PM, Ritesh Harjani wrote:
>>
>>
>> On 4/13/20 2:07 PM, Xing Zhengjun wrote:
>>> Hi Harjani,
>>>
>>>     Do you have time to take a look at this? Thanks.
>>
>> Hello Xing,
>>
>> I do want to look into this. But as of now I am stuck with another
>> mballoc failure issue. I will get back at this once I have some handle
>> over that one.
>>
>> BTW, are you planning to take look at this?
>>
>> -ritesh
>>
>>
>>>
>>> On 4/7/2020 4:00 PM, kernel test robot wrote:
>>>> Greeting,
>>>>
>>>> FYI, we noticed a -60.5% regression of stress-ng.fiemap.ops_per_sec 
>>>> due to commit:
>>>>
>>>>
>>>> commit: d3b6f23f71670007817a5d59f3fbafab2b794e8c ("ext4: move 
>>>> ext4_fiemap to use iomap framework")
>>>> https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master
>>>>
>>>> in testcase: stress-ng
>>>> on test machine: 96 threads Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz 
>>>> with 192G memory
>>>> with following parameters:
>>>>
>>>>     nr_threads: 10%
>>>>     disk: 1HDD
>>>>     testtime: 1s
>>>>     class: os
>>>>     cpufreq_governor: performance
>>>>     ucode: 0x500002c
>>>>     fs: ext4
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> 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
>>>>
>>>> ========================================================================================= 
>>>>
>>>> class/compiler/cpufreq_governor/disk/fs/kconfig/nr_threads/rootfs/tbox_group/testcase/testtime/ucode: 
>>>>
>>>> os/gcc-7/performance/1HDD/ext4/x86_64-rhel-7.6/10%/debian-x86_64-20191114.cgz/lkp-csl-2sp5/stress-ng/1s/0x500002c 
>>>>
>>>>
>>>> commit:
>>>>    b2c5764262 ("ext4: make ext4_ind_map_blocks work with fiemap")
>>>>    d3b6f23f71 ("ext4: move ext4_fiemap to use iomap framework")
>>>>
>>>> b2c5764262edded1 d3b6f23f71670007817a5d59f3f
>>>> ---------------- ---------------------------
>>>>         fail:runs  %reproduction    fail:runs
>>>>             |             |             |
>>>>             :4           25%           1:4 
>>>> dmesg.WARNING:at#for_ip_interrupt_entry/0x
>>>>            2:4            5%           2:4 
>>>> perf-profile.calltrace.cycles-pp.sync_regs.error_entry
>>>>            2:4            6%           3:4 
>>>> perf-profile.calltrace.cycles-pp.error_entry
>>>>            3:4            9%           3:4 
>>>> perf-profile.children.cycles-pp.error_entry
>>>>            0:4            1%           0:4 
>>>> perf-profile.self.cycles-pp.error_entry
>>>>           %stddev     %change         %stddev
>>>>               \          |                \
>>>>       28623           +28.2%      36703 ± 12%  stress-ng.daemon.ops
>>>>       28632           +28.2%      36704 ± 12% 
>>>> stress-ng.daemon.ops_per_sec
>>>>      566.00 ± 22%     -53.2%     265.00 ± 53%  stress-ng.dev.ops
>>>>      278.81 ± 22%     -53.0%     131.00 ± 54%  
>>>> stress-ng.dev.ops_per_sec
>>>>       73160           -60.6%      28849 ±  3%  stress-ng.fiemap.ops
>>>>       72471           -60.5%      28612 ±  3% 
>>>> stress-ng.fiemap.ops_per_sec
>>>>       23421 ± 12%     +21.2%      28388 ±  6%  stress-ng.filename.ops
>>>>       22638 ± 12%     +20.3%      27241 ±  6% 
>>>> stress-ng.filename.ops_per_sec
>>>>       21.25 ±  7%     -10.6%      19.00 ±  3%  stress-ng.iomix.ops
>>>>       38.75 ± 49%     -47.7%      20.25 ± 96%  stress-ng.memhotplug.ops
>>>>       34.45 ± 52%     -51.8%      16.62 ±106% 
>>>> stress-ng.memhotplug.ops_per_sec
>>>>        1734 ± 10%     +31.4%       2278 ± 10%  stress-ng.resources.ops
>>>>      807.56 ±  5%     +35.2%       1091 ±  8% 
>>>> stress-ng.resources.ops_per_sec
>>>>     1007356 ±  3%     -16.5%     840642 ±  9%  stress-ng.revio.ops
>>>>     1007692 ±  3%     -16.6%     840711 ±  9% 
>>>> stress-ng.revio.ops_per_sec
>>>>       21812 ±  3%     +16.0%      25294 ±  5%  stress-ng.sysbadaddr.ops
>>>>       21821 ±  3%     +15.9%      25294 ±  5% 
>>>> stress-ng.sysbadaddr.ops_per_sec
>>>>      440.75 ±  4%     +21.9%     537.25 ±  9%  stress-ng.sysfs.ops
>>>>      440.53 ±  4%     +21.9%     536.86 ±  9% 
>>>> stress-ng.sysfs.ops_per_sec
>>>>    13286582           -11.1%   11805520 ±  6% 
>>>> stress-ng.time.file_system_outputs
>>>>    68253896            +2.4%   69860122 
>>>> stress-ng.time.minor_page_faults
>>>>      197.00 ±  4%     -15.9%     165.75 ± 12%  stress-ng.xattr.ops
>>>>      192.45 ±  5%     -16.1%     161.46 ± 11% 
>>>> stress-ng.xattr.ops_per_sec
>>>>       15310           +62.5%      24875 ± 22%  stress-ng.zombie.ops
>>>>       15310           +62.5%      24874 ± 22% 
>>>> stress-ng.zombie.ops_per_sec
>>>>      203.50 ± 12%     -47.3%     107.25 ± 49%  vmstat.io.bi
>>>>      861318 ± 18%     -29.7%     605884 ±  5%  meminfo.AnonHugePages
>>>>     1062742 ± 14%     -20.2%     847853 ±  3%  meminfo.AnonPages
>>>>       31093 ±  6%      +9.6%      34090 ±  3%  meminfo.KernelStack
>>>>        7151 ± 34%     +55.8%      11145 ±  9%  meminfo.Mlocked
>>>>   1.082e+08 ±  5%     -40.2%   64705429 ± 31% 
>>>> numa-numastat.node0.local_node
>>>>   1.082e+08 ±  5%     -40.2%   64739883 ± 31% 
>>>> numa-numastat.node0.numa_hit
>>>>    46032662 ± 21%    +104.3%   94042918 ± 20% 
>>>> numa-numastat.node1.local_node
>>>>    46074205 ± 21%    +104.2%   94072810 ± 20% 
>>>> numa-numastat.node1.numa_hit
>>>>        3942 ±  3%     +14.2%       4501 ±  4% 
>>>> slabinfo.pool_workqueue.active_objs
>>>>        4098 ±  3%     +14.3%       4683 ±  4% 
>>>> slabinfo.pool_workqueue.num_objs
>>>>        4817 ±  7%     +13.3%       5456 ±  8% 
>>>> slabinfo.proc_dir_entry.active_objs
>>>>        5153 ±  6%     +12.5%       5797 ±  8% 
>>>> slabinfo.proc_dir_entry.num_objs
>>>>       18598 ± 13%     -33.1%      12437 ± 20% 
>>>> sched_debug.cfs_rq:/.load.avg
>>>>      452595 ± 56%     -71.4%     129637 ± 76% 
>>>> sched_debug.cfs_rq:/.load.max
>>>>       67675 ± 35%     -55.1%      30377 ± 42% 
>>>> sched_debug.cfs_rq:/.load.stddev
>>>>       18114 ± 12%     -33.7%      12011 ± 20% 
>>>> sched_debug.cfs_rq:/.runnable_weight.avg
>>>>      448215 ± 58%     -72.8%     121789 ± 82% 
>>>> sched_debug.cfs_rq:/.runnable_weight.max
>>>>       67083 ± 37%     -56.3%      29305 ± 43% 
>>>> sched_debug.cfs_rq:/.runnable_weight.stddev
>>>>      -38032          +434.3%    -203212 
>>>> sched_debug.cfs_rq:/.spread0.avg
>>>>     -204466           +95.8%    -400301 
>>>> sched_debug.cfs_rq:/.spread0.min
>>>>       90.02 ± 25%     -58.1%      37.69 ± 52% 
>>>> sched_debug.cfs_rq:/.util_est_enqueued.avg
>>>>      677.54 ±  6%     -39.3%     411.50 ± 22% 
>>>> sched_debug.cfs_rq:/.util_est_enqueued.max
>>>>      196.57 ±  8%     -47.6%     103.05 ± 36% 
>>>> sched_debug.cfs_rq:/.util_est_enqueued.stddev
>>>>        3.34 ± 23%     +34.1%       4.48 ±  4% 
>>>> sched_debug.cpu.clock.stddev
>>>>        3.34 ± 23%     +34.1%       4.48 ±  4% 
>>>> sched_debug.cpu.clock_task.stddev
>>>>      402872 ±  7%     -11.9%     354819 ±  2% 
>>>> proc-vmstat.nr_active_anon
>>>>     1730331            -9.5%    1566418 ±  5%  proc-vmstat.nr_dirtied
>>>>       31042 ±  6%      +9.3%      33915 ±  3% 
>>>> proc-vmstat.nr_kernel_stack
>>>>      229047            -2.4%     223615        proc-vmstat.nr_mapped
>>>>       74008 ±  7%     +20.5%      89163 ±  8%  proc-vmstat.nr_written
>>>>      402872 ±  7%     -11.9%     354819 ±  2% 
>>>> proc-vmstat.nr_zone_active_anon
>>>>       50587 ± 11%     -25.2%      37829 ± 14% 
>>>> proc-vmstat.numa_pages_migrated
>>>>      457500           -23.1%     351918 ± 31% 
>>>> proc-vmstat.numa_pte_updates
>>>>    81382485            +1.9%   82907822        proc-vmstat.pgfault
>>>>   2.885e+08 ±  5%     -13.3%  2.502e+08 ±  6%  proc-vmstat.pgfree
>>>>       42206 ± 12%     -46.9%      22399 ± 49%  proc-vmstat.pgpgin
>>>>      431233 ± 13%     -64.8%     151736 ±109%  proc-vmstat.pgrotated
>>>>      176754 ±  7%     -40.2%     105637 ± 31% 
>>>> proc-vmstat.thp_fault_alloc
>>>>      314.50 ± 82%    +341.5%       1388 ± 44% 
>>>> proc-vmstat.unevictable_pgs_stranded
>>>>     1075269 ± 14%     -41.3%     631388 ± 17%  
>>>> numa-meminfo.node0.Active
>>>>      976056 ± 12%     -39.7%     588727 ± 19% 
>>>> numa-meminfo.node0.Active(anon)
>>>>      426857 ± 22%     -36.4%     271375 ± 13% 
>>>> numa-meminfo.node0.AnonHugePages
>>>>      558590 ± 19%     -36.4%     355402 ± 14% 
>>>> numa-meminfo.node0.AnonPages
>>>>     1794824 ±  9%     -28.8%    1277157 ± 20% 
>>>> numa-meminfo.node0.FilePages
>>>>        8517 ± 92%     -82.7%       1473 ± 89% 
>>>> numa-meminfo.node0.Inactive(file)
>>>>      633118 ±  2%     -41.7%     368920 ± 36%  
>>>> numa-meminfo.node0.Mapped
>>>>     2958038 ± 12%     -27.7%    2139271 ± 12% 
>>>> numa-meminfo.node0.MemUsed
>>>>      181401 ±  5%     -13.7%     156561 ±  4% 
>>>> numa-meminfo.node0.SUnreclaim
>>>>      258124 ±  6%     -13.0%     224535 ±  5%  numa-meminfo.node0.Slab
>>>>      702083 ± 16%     +31.0%     919406 ± 11%  
>>>> numa-meminfo.node1.Active
>>>>       38663 ±107%    +137.8%      91951 ± 31% 
>>>> numa-meminfo.node1.Active(file)
>>>>     1154975 ±  7%     +41.6%    1635593 ± 12% 
>>>> numa-meminfo.node1.FilePages
>>>>      395813 ± 25%     +62.8%     644533 ± 16% 
>>>> numa-meminfo.node1.Inactive
>>>>      394313 ± 25%     +62.5%     640686 ± 16% 
>>>> numa-meminfo.node1.Inactive(anon)
>>>>      273317           +88.8%     515976 ± 25%  
>>>> numa-meminfo.node1.Mapped
>>>>     2279237 ±  6%     +25.7%    2865582 ±  7% 
>>>> numa-meminfo.node1.MemUsed
>>>>       10830 ± 18%     +29.6%      14033 ±  9% 
>>>> numa-meminfo.node1.PageTables
>>>>      149390 ±  3%     +23.2%     184085 ±  3% 
>>>> numa-meminfo.node1.SUnreclaim
>>>>      569542 ± 16%     +74.8%     995336 ± 21%  numa-meminfo.node1.Shmem
>>>>      220774 ±  5%     +20.3%     265656 ±  3%  numa-meminfo.node1.Slab
>>>>    35623587 ±  5%     -11.7%   31444514 ±  3%  perf-stat.i.cache-misses
>>>>   2.576e+08 ±  5%      -6.8%    2.4e+08 ±  2% 
>>>> perf-stat.i.cache-references
>>>>        3585            -7.3%       3323 ±  5% 
>>>> perf-stat.i.cpu-migrations
>>>>      180139 ±  2%      +4.2%     187668        perf-stat.i.minor-faults
>>>>       69.13            +2.6       71.75 
>>>> perf-stat.i.node-load-miss-rate%
>>>>     4313695 ±  2%      -7.4%    3994957 ±  2% 
>>>> perf-stat.i.node-load-misses
>>>>     5466253 ± 11%     -17.3%    4521173 ±  6%  perf-stat.i.node-loads
>>>>     2818674 ±  6%     -15.8%    2372542 ±  5%  perf-stat.i.node-stores
>>>>      227810            +4.6%     238290        perf-stat.i.page-faults
>>>>       12.67 ±  4%      -7.2%      11.76 ±  2%  perf-stat.overall.MPKI
>>>>        1.01 ±  4%      -0.0        0.97 ±  3% 
>>>> perf-stat.overall.branch-miss-rate%
>>>>        1044           +13.1%       1181 ±  4% 
>>>> perf-stat.overall.cycles-between-cache-misses
>>>>       40.37 ±  4%      +3.6       44.00 ±  2% 
>>>> perf-stat.overall.node-store-miss-rate%
>>>>    36139526 ±  5%     -12.5%   31625519 ±  3%  
>>>> perf-stat.ps.cache-misses
>>>>   2.566e+08 ±  5%      -6.9%  2.389e+08 ±  2% 
>>>> perf-stat.ps.cache-references
>>>>        3562            -7.2%       3306 ±  5% 
>>>> perf-stat.ps.cpu-migrations
>>>>      179088            +4.2%     186579        
>>>> perf-stat.ps.minor-faults
>>>>     4323383 ±  2%      -7.5%    3999214 perf-stat.ps.node-load-misses
>>>>     5607721 ± 10%     -18.5%    4568664 ±  6%  perf-stat.ps.node-loads
>>>>     2855134 ±  7%     -16.4%    2387345 ±  5%  perf-stat.ps.node-stores
>>>>      226270            +4.6%     236709        perf-stat.ps.page-faults
>>>>      242305 ± 10%     -42.4%     139551 ± 18% 
>>>> numa-vmstat.node0.nr_active_anon
>>>>      135983 ± 17%     -37.4%      85189 ± 10% 
>>>> numa-vmstat.node0.nr_anon_pages
>>>>      209.25 ± 16%     -38.1%     129.50 ± 10% 
>>>> numa-vmstat.node0.nr_anon_transparent_hugepages
>>>>      449367 ±  9%     -29.7%     315804 ± 20% 
>>>> numa-vmstat.node0.nr_file_pages
>>>>        2167 ± 90%     -80.6%     419.75 ± 98% 
>>>> numa-vmstat.node0.nr_inactive_file
>>>>      157405 ±  3%     -41.4%      92206 ± 35% 
>>>> numa-vmstat.node0.nr_mapped
>>>>        2022 ± 30%     -73.3%     539.25 ± 91% 
>>>> numa-vmstat.node0.nr_mlock
>>>>        3336 ± 10%     -24.3%       2524 ± 25% 
>>>> numa-vmstat.node0.nr_page_table_pages
>>>>      286158 ± 10%     -41.2%     168337 ± 37% 
>>>> numa-vmstat.node0.nr_shmem
>>>>       45493 ±  5%     -14.1%      39094 ±  4% 
>>>> numa-vmstat.node0.nr_slab_unreclaimable
>>>>      242294 ± 10%     -42.4%     139547 ± 18% 
>>>> numa-vmstat.node0.nr_zone_active_anon
>>>>        2167 ± 90%     -80.6%     419.75 ± 98% 
>>>> numa-vmstat.node0.nr_zone_inactive_file
>>>>    54053924 ±  8%     -39.3%   32786242 ± 34% 
>>>> numa-vmstat.node0.numa_hit
>>>>    53929628 ±  8%     -39.5%   32619715 ± 34% 
>>>> numa-vmstat.node0.numa_local
>>>>        9701 ±107%    +136.9%      22985 ± 31% 
>>>> numa-vmstat.node1.nr_active_file
>>>>      202.50 ± 16%     -25.1%     151.75 ± 23% 
>>>> numa-vmstat.node1.nr_anon_transparent_hugepages
>>>>      284922 ±  7%     +43.3%     408195 ± 13% 
>>>> numa-vmstat.node1.nr_file_pages
>>>>       96002 ± 26%     +67.5%     160850 ± 17% 
>>>> numa-vmstat.node1.nr_inactive_anon
>>>>       68077 ±  2%     +90.3%     129533 ± 25% 
>>>> numa-vmstat.node1.nr_mapped
>>>>      138482 ± 15%     +79.2%     248100 ± 22% 
>>>> numa-vmstat.node1.nr_shmem
>>>>       37396 ±  3%     +23.3%      46094 ±  3% 
>>>> numa-vmstat.node1.nr_slab_unreclaimable
>>>>        9701 ±107%    +136.9%      22985 ± 31% 
>>>> numa-vmstat.node1.nr_zone_active_file
>>>>       96005 ± 26%     +67.5%     160846 ± 17% 
>>>> numa-vmstat.node1.nr_zone_inactive_anon
>>>>    23343661 ± 17%     +99.9%   46664267 ± 23% 
>>>> numa-vmstat.node1.numa_hit
>>>>    23248487 ± 17%    +100.5%   46610447 ± 23% 
>>>> numa-vmstat.node1.numa_local
>>>>      105745 ± 23%    +112.6%     224805 ± 24%  softirqs.CPU0.NET_RX
>>>>      133310 ± 36%     -45.3%      72987 ± 52%  softirqs.CPU1.NET_RX
>>>>      170110 ± 55%     -66.8%      56407 ±147%  softirqs.CPU11.NET_RX
>>>>       91465 ± 36%     -65.2%      31858 ±112%  softirqs.CPU13.NET_RX
>>>>      164491 ± 57%     -77.7%      36641 ±121%  softirqs.CPU15.NET_RX
>>>>      121069 ± 55%     -99.3%     816.75 ± 96%  softirqs.CPU17.NET_RX
>>>>       81019 ±  4%      -8.7%      73967 ±  4%  softirqs.CPU20.RCU
>>>>       72143 ± 63%     -89.8%       7360 ±172%  softirqs.CPU22.NET_RX
>>>>      270663 ± 17%     -57.9%     113915 ± 45%  softirqs.CPU24.NET_RX
>>>>       20149 ± 76%    +474.1%     115680 ± 62%  softirqs.CPU26.NET_RX
>>>>       14033 ± 70%    +977.5%     151211 ± 75%  softirqs.CPU27.NET_RX
>>>>       27834 ± 94%    +476.1%     160357 ± 28%  softirqs.CPU28.NET_RX
>>>>       35346 ± 68%    +212.0%     110290 ± 30%  softirqs.CPU29.NET_RX
>>>>       34347 ±103%    +336.5%     149941 ± 32%  softirqs.CPU32.NET_RX
>>>>       70077 ±  3%     +10.8%      77624 ±  3%  softirqs.CPU34.RCU
>>>>       36453 ± 84%    +339.6%     160253 ± 42%  softirqs.CPU36.NET_RX
>>>>       72367 ±  2%     +10.6%      80043        softirqs.CPU37.RCU
>>>>       25239 ±118%    +267.7%      92799 ± 45%  softirqs.CPU38.NET_RX
>>>>        4995 ±170%   +1155.8%      62734 ± 62%  softirqs.CPU39.NET_RX
>>>>        4641 ±145%   +1611.3%      79432 ± 90%  softirqs.CPU42.NET_RX
>>>>        7192 ± 65%    +918.0%      73225 ± 66%  softirqs.CPU45.NET_RX
>>>>        1772 ±166%   +1837.4%      34344 ± 63%  softirqs.CPU46.NET_RX
>>>>       13149 ± 81%    +874.7%     128170 ± 58%  softirqs.CPU47.NET_RX
>>>>       86484 ± 94%     -92.6%       6357 ±172%  softirqs.CPU48.NET_RX
>>>>      129128 ± 27%     -95.8%       5434 ±172%  softirqs.CPU55.NET_RX
>>>>       82772 ± 59%     -91.7%       6891 ±164%  softirqs.CPU56.NET_RX
>>>>      145313 ± 57%     -87.8%      17796 ± 88%  softirqs.CPU57.NET_RX
>>>>      118160 ± 33%     -86.3%      16226 ±109%  softirqs.CPU58.NET_RX
>>>>       94576 ± 56%     -94.1%       5557 ±173%  softirqs.CPU6.NET_RX
>>>>       82900 ± 77%     -66.8%      27508 ±171%  softirqs.CPU62.NET_RX
>>>>      157291 ± 30%     -81.1%      29656 ±111%  softirqs.CPU64.NET_RX
>>>>      135101 ± 28%     -80.2%      26748 ± 90%  softirqs.CPU67.NET_RX
>>>>      146574 ± 56%    -100.0%      69.75 ± 98%  softirqs.CPU68.NET_RX
>>>>       81347 ±  2%      -9.0%      74024 ±  2%  softirqs.CPU68.RCU
>>>>      201729 ± 37%     -99.6%     887.50 ±107%  softirqs.CPU69.NET_RX
>>>>      108454 ± 78%     -97.9%       2254 ±169%  softirqs.CPU70.NET_RX
>>>>       55289 ±104%     -89.3%       5942 ±172%  softirqs.CPU71.NET_RX
>>>>       10112 ±172%    +964.6%     107651 ± 89%  softirqs.CPU72.NET_RX
>>>>        3136 ±171%   +1522.2%      50879 ± 66%  softirqs.CPU73.NET_RX
>>>>       13353 ± 79%    +809.2%     121407 ±101%  softirqs.CPU74.NET_RX
>>>>       75194 ±  3%     +10.3%      82957 ±  5%  softirqs.CPU75.RCU
>>>>       11002 ±173%   +1040.8%     125512 ± 61%  softirqs.CPU76.NET_RX
>>>>        2463 ±173%   +2567.3%      65708 ± 77%  softirqs.CPU78.NET_RX
>>>>       25956 ±  3%      -7.8%      23932 ±  3%  softirqs.CPU78.SCHED
>>>>       16366 ±150%    +340.7%      72125 ± 91%  softirqs.CPU82.NET_RX
>>>>       14553 ±130%   +1513.4%     234809 ± 27%  softirqs.CPU93.NET_RX
>>>>       26314            -9.2%      23884 ±  3%  softirqs.CPU93.SCHED
>>>>        4582 ± 88%   +4903.4%     229268 ± 23%  softirqs.CPU94.NET_RX
>>>>       11214 ±111%   +1762.5%     208867 ± 18%  softirqs.CPU95.NET_RX
>>>>        1.53 ± 27%      -0.5        0.99 ± 17% 
>>>> perf-profile.calltrace.cycles-pp.exit_to_usermode_loop.do_syscall_64.entry_SYSCALL_64_after_hwframe 
>>>>
>>>>        1.52 ± 27%      -0.5        0.99 ± 17% 
>>>> perf-profile.calltrace.cycles-pp.do_signal.exit_to_usermode_loop.do_syscall_64.entry_SYSCALL_64_after_hwframe 
>>>>
>>>>        1.39 ± 29%      -0.5        0.88 ± 21% 
>>>> perf-profile.calltrace.cycles-pp.do_group_exit.get_signal.do_signal.exit_to_usermode_loop.do_syscall_64 
>>>>
>>>>        1.39 ± 29%      -0.5        0.88 ± 21% 
>>>> perf-profile.calltrace.cycles-pp.get_signal.do_signal.exit_to_usermode_loop.do_syscall_64.entry_SYSCALL_64_after_hwframe 
>>>>
>>>>        0.50 ± 59%      +0.3        0.81 ± 13% 
>>>> perf-profile.calltrace.cycles-pp.filemap_map_pages.handle_pte_fault.__handle_mm_fault.handle_mm_fault.do_page_fault 
>>>>
>>>>        5.70 ±  9%      +0.8        6.47 ±  7% 
>>>> perf-profile.calltrace.cycles-pp.do_exit.do_group_exit.get_signal.do_signal.exit_to_usermode_loop 
>>>>
>>>>        5.48 ±  9%      +0.8        6.27 ±  7% 
>>>> perf-profile.calltrace.cycles-pp.exit_mmap.mmput.do_exit.do_group_exit.get_signal 
>>>>
>>>>        5.49 ±  9%      +0.8        6.28 ±  7% 
>>>> perf-profile.calltrace.cycles-pp.mmput.do_exit.do_group_exit.get_signal.do_signal 
>>>>
>>>>        4.30 ±  4%      +1.3        5.60 ±  7% 
>>>> perf-profile.calltrace.cycles-pp.do_group_exit.get_signal.do_signal.exit_to_usermode_loop.prepare_exit_to_usermode 
>>>>
>>>>        4.40 ±  4%      +1.3        5.69 ±  7% 
>>>> perf-profile.calltrace.cycles-pp.prepare_exit_to_usermode.swapgs_restore_regs_and_return_to_usermode 
>>>>
>>>>        4.37 ±  4%      +1.3        5.66 ±  7% 
>>>> perf-profile.calltrace.cycles-pp.exit_to_usermode_loop.prepare_exit_to_usermode.swapgs_restore_regs_and_return_to_usermode 
>>>>
>>>>        4.36 ±  4%      +1.3        5.66 ±  7% 
>>>> perf-profile.calltrace.cycles-pp.do_signal.exit_to_usermode_loop.prepare_exit_to_usermode.swapgs_restore_regs_and_return_to_usermode 
>>>>
>>>>        4.33 ±  4%      +1.3        5.62 ±  7% 
>>>> perf-profile.calltrace.cycles-pp.get_signal.do_signal.exit_to_usermode_loop.prepare_exit_to_usermode.swapgs_restore_regs_and_return_to_usermode 
>>>>
>>>>        4.44 ±  4%      +1.3        5.74 ±  7% 
>>>> perf-profile.calltrace.cycles-pp.swapgs_restore_regs_and_return_to_usermode 
>>>>
>>>>        3.20 ± 10%      -2.4        0.78 ±156% 
>>>> perf-profile.children.cycles-pp.copy_page
>>>>        0.16 ±  9%      -0.1        0.08 ± 64% 
>>>> perf-profile.children.cycles-pp.irq_work_interrupt
>>>>        0.16 ±  9%      -0.1        0.08 ± 64% 
>>>> perf-profile.children.cycles-pp.smp_irq_work_interrupt
>>>>        0.24 ±  5%      -0.1        0.17 ± 18% 
>>>> perf-profile.children.cycles-pp.irq_work_run_list
>>>>        0.16 ±  9%      -0.1        0.10 ± 24% 
>>>> perf-profile.children.cycles-pp.irq_work_run
>>>>        0.16 ±  9%      -0.1        0.10 ± 24% 
>>>> perf-profile.children.cycles-pp.printk
>>>>        0.23 ±  6%      -0.1        0.17 ±  9% 
>>>> perf-profile.children.cycles-pp.__do_execve_file
>>>>        0.08 ± 14%      -0.1        0.03 ±100% 
>>>> perf-profile.children.cycles-pp.delay_tsc
>>>>        0.16 ±  6%      -0.1        0.11 ±  9% 
>>>> perf-profile.children.cycles-pp.load_elf_binary
>>>>        0.16 ±  7%      -0.0        0.12 ± 13% 
>>>> perf-profile.children.cycles-pp.search_binary_handler
>>>>        0.20 ±  7%      -0.0        0.15 ± 10% 
>>>> perf-profile.children.cycles-pp.call_usermodehelper_exec_async
>>>>        0.19 ±  6%      -0.0        0.15 ± 11% 
>>>> perf-profile.children.cycles-pp.do_execve
>>>>        0.08 ± 10%      -0.0        0.04 ± 59% 
>>>> perf-profile.children.cycles-pp.__vunmap
>>>>        0.15 ±  3%      -0.0        0.11 ±  7% 
>>>> perf-profile.children.cycles-pp.rcu_idle_exit
>>>>        0.12 ± 10%      -0.0        0.09 ± 14% 
>>>> perf-profile.children.cycles-pp.__switch_to_asm
>>>>        0.09 ± 13%      -0.0        0.07 ±  5% 
>>>> perf-profile.children.cycles-pp.des3_ede_encrypt
>>>>        0.06 ± 11%      +0.0        0.09 ± 13% 
>>>> perf-profile.children.cycles-pp.mark_page_accessed
>>>>        0.15 ±  5%      +0.0        0.19 ± 12% 
>>>> perf-profile.children.cycles-pp.apparmor_cred_prepare
>>>>        0.22 ±  8%      +0.0        0.27 ± 11% 
>>>> perf-profile.children.cycles-pp.mem_cgroup_throttle_swaprate
>>>>        0.17 ±  2%      +0.0        0.22 ± 12% 
>>>> perf-profile.children.cycles-pp.security_prepare_creds
>>>>        0.95 ± 17%      +0.3        1.22 ± 14% 
>>>> perf-profile.children.cycles-pp.filemap_map_pages
>>>>        5.92 ±  8%      +0.7        6.65 ±  7% 
>>>> perf-profile.children.cycles-pp.get_signal
>>>>        5.66 ±  9%      +0.8        6.44 ±  7% 
>>>> perf-profile.children.cycles-pp.mmput
>>>>        5.65 ±  9%      +0.8        6.43 ±  7% 
>>>> perf-profile.children.cycles-pp.exit_mmap
>>>>        4.40 ±  4%      +1.3        5.70 ±  7% 
>>>> perf-profile.children.cycles-pp.prepare_exit_to_usermode
>>>>        4.45 ±  4%      +1.3        5.75 ±  7% 
>>>> perf-profile.children.cycles-pp.swapgs_restore_regs_and_return_to_usermode 
>>>>
>>>>        3.16 ± 10%      -2.4        0.77 ±155% 
>>>> perf-profile.self.cycles-pp.copy_page
>>>>        0.08 ± 14%      -0.1        0.03 ±100% 
>>>> perf-profile.self.cycles-pp.delay_tsc
>>>>        0.12 ± 10%      -0.0        0.09 ± 14% 
>>>> perf-profile.self.cycles-pp.__switch_to_asm
>>>>        0.08 ± 12%      -0.0        0.06 ± 17% 
>>>> perf-profile.self.cycles-pp.enqueue_task_fair
>>>>        0.09 ± 13%      -0.0        0.07 ±  5% 
>>>> perf-profile.self.cycles-pp.des3_ede_encrypt
>>>>        0.07 ± 13%      +0.0        0.08 ± 19% 
>>>> perf-profile.self.cycles-pp.__lru_cache_add
>>>>        0.19 ±  9%      +0.0        0.22 ± 10% 
>>>> perf-profile.self.cycles-pp.mem_cgroup_throttle_swaprate
>>>>        0.15 ±  5%      +0.0        0.19 ± 11% 
>>>> perf-profile.self.cycles-pp.apparmor_cred_prepare
>>>>        0.05 ± 58%      +0.0        0.09 ± 13% 
>>>> perf-profile.self.cycles-pp.mark_page_accessed
>>>>        0.58 ± 10%      +0.2        0.80 ± 20% 
>>>> perf-profile.self.cycles-pp.release_pages
>>>>        0.75 ±173%  +1.3e+05%       1005 ±100% 
>>>> interrupts.127:PCI-MSI.31981660-edge.i40e-eth0-TxRx-91
>>>>      820.75 ±111%     -99.9%       0.50 ±173% 
>>>> interrupts.47:PCI-MSI.31981580-edge.i40e-eth0-TxRx-11
>>>>      449.25 ± 86%    -100.0%       0.00 
>>>> interrupts.53:PCI-MSI.31981586-edge.i40e-eth0-TxRx-17
>>>>       33.25 ±157%    -100.0%       0.00 
>>>> interrupts.57:PCI-MSI.31981590-edge.i40e-eth0-TxRx-21
>>>>        0.75 ±110%  +63533.3%     477.25 ±162% 
>>>> interrupts.61:PCI-MSI.31981594-edge.i40e-eth0-TxRx-25
>>>>      561.50 ±160%    -100.0%       0.00 
>>>> interrupts.65:PCI-MSI.31981598-edge.i40e-eth0-TxRx-29
>>>>       82921 ±  8%     -11.1%      73748 ±  6% 
>>>> interrupts.CPU11.CAL:Function_call_interrupts
>>>>       66509 ± 30%     -32.6%      44828 ±  8% 
>>>> interrupts.CPU14.TLB:TLB_shootdowns
>>>>       43105 ± 98%     -90.3%       4183 ± 21% 
>>>> interrupts.CPU17.RES:Rescheduling_interrupts
>>>>      148719 ± 70%     -69.4%      45471 ± 16% 
>>>> interrupts.CPU17.TLB:TLB_shootdowns
>>>>       85589 ± 42%     -52.2%      40884 ±  5% 
>>>> interrupts.CPU20.TLB:TLB_shootdowns
>>>>      222472 ± 41%     -98.0%       4360 ± 45% 
>>>> interrupts.CPU22.RES:Rescheduling_interrupts
>>>>        0.50 ±173%  +95350.0%     477.25 ±162% 
>>>> interrupts.CPU25.61:PCI-MSI.31981594-edge.i40e-eth0-TxRx-25
>>>>       76029 ± 10%     +14.9%      87389 ±  5% 
>>>> interrupts.CPU25.CAL:Function_call_interrupts
>>>>      399042 ±  6%     +13.4%     452479 ±  8% 
>>>> interrupts.CPU27.LOC:Local_timer_interrupts
>>>>      561.00 ±161%    -100.0%       0.00 
>>>> interrupts.CPU29.65:PCI-MSI.31981598-edge.i40e-eth0-TxRx-29
>>>>        7034 ± 46%   +1083.8%      83279 ±138% 
>>>> interrupts.CPU29.RES:Rescheduling_interrupts
>>>>       17829 ± 99%     -71.0%       5172 ± 16% 
>>>> interrupts.CPU30.RES:Rescheduling_interrupts
>>>>        5569 ± 15%   +2414.7%     140059 ± 94% 
>>>> interrupts.CPU31.RES:Rescheduling_interrupts
>>>>       37674 ± 16%     +36.6%      51473 ± 25% 
>>>> interrupts.CPU31.TLB:TLB_shootdowns
>>>>       47905 ± 39%     +76.6%      84583 ± 38% 
>>>> interrupts.CPU34.TLB:TLB_shootdowns
>>>>      568.75 ±140%    +224.8%       1847 ± 90% 
>>>> interrupts.CPU36.NMI:Non-maskable_interrupts
>>>>      568.75 ±140%    +224.8%       1847 ± 90% 
>>>> interrupts.CPU36.PMI:Performance_monitoring_interrupts
>>>>        4236 ± 25%   +2168.5%      96092 ± 90% 
>>>> interrupts.CPU36.RES:Rescheduling_interrupts
>>>>       52717 ± 27%     +43.3%      75565 ± 28% 
>>>> interrupts.CPU37.TLB:TLB_shootdowns
>>>>       41418 ±  9%    +136.6%      98010 ± 50% 
>>>> interrupts.CPU39.TLB:TLB_shootdowns
>>>>        5551 ±  8%    +847.8%      52615 ± 66% 
>>>> interrupts.CPU40.RES:Rescheduling_interrupts
>>>>        4746 ± 25%    +865.9%      45841 ± 91% 
>>>> interrupts.CPU42.RES:Rescheduling_interrupts
>>>>       37556 ± 11%     +24.6%      46808 ±  6% 
>>>> interrupts.CPU42.TLB:TLB_shootdowns
>>>>       21846 ±124%     -84.4%       3415 ± 46% 
>>>> interrupts.CPU48.RES:Rescheduling_interrupts
>>>>      891.50 ± 22%     -35.2%     577.25 ± 40% 
>>>> interrupts.CPU49.NMI:Non-maskable_interrupts
>>>>      891.50 ± 22%     -35.2%     577.25 ± 40% 
>>>> interrupts.CPU49.PMI:Performance_monitoring_interrupts
>>>>       20459 ±120%     -79.2%       4263 ± 14% 
>>>> interrupts.CPU49.RES:Rescheduling_interrupts
>>>>       59840 ± 21%     -23.1%      46042 ± 16% 
>>>> interrupts.CPU5.TLB:TLB_shootdowns
>>>>       65200 ± 19%     -34.5%      42678 ±  9% 
>>>> interrupts.CPU51.TLB:TLB_shootdowns
>>>>       70923 ±153%     -94.0%       4270 ± 29% 
>>>> interrupts.CPU53.RES:Rescheduling_interrupts
>>>>       65312 ± 22%     -28.7%      46578 ± 14% 
>>>> interrupts.CPU56.TLB:TLB_shootdowns
>>>>       65828 ± 24%   �� -33.4%      43846 ±  4% 
>>>> interrupts.CPU59.TLB:TLB_shootdowns
>>>>       72558 ±156%     -93.2%       4906 ±  9% 
>>>> interrupts.CPU6.RES:Rescheduling_interrupts
>>>>       68698 ± 34%     -32.6%      46327 ± 18% 
>>>> interrupts.CPU61.TLB:TLB_shootdowns
>>>>      109745 ± 44%     -57.4%      46711 ± 16% 
>>>> interrupts.CPU62.TLB:TLB_shootdowns
>>>>       89714 ± 44%     -48.5%      46198 ±  7% 
>>>> interrupts.CPU63.TLB:TLB_shootdowns
>>>>       59380 ±136%     -91.5%       5066 ± 13% 
>>>> interrupts.CPU69.RES:Rescheduling_interrupts
>>>>       40094 ± 18%    +133.9%      93798 ± 44% 
>>>> interrupts.CPU78.TLB:TLB_shootdowns
>>>>      129884 ± 72%     -55.3%      58034 ±157% 
>>>> interrupts.CPU8.RES:Rescheduling_interrupts
>>>>       69984 ± 11%     +51.4%     105957 ± 20% 
>>>> interrupts.CPU80.CAL:Function_call_interrupts
>>>>       32857 ± 10%    +128.7%      75131 ± 36% 
>>>> interrupts.CPU80.TLB:TLB_shootdowns
>>>>       35726 ± 16%     +34.6%      48081 ± 12% 
>>>> interrupts.CPU82.TLB:TLB_shootdowns
>>>>       73820 ± 17%     +28.2%      94643 ±  8% 
>>>> interrupts.CPU84.CAL:Function_call_interrupts
>>>>       38829 ± 28%    +190.3%     112736 ± 42% 
>>>> interrupts.CPU84.TLB:TLB_shootdowns
>>>>       36129 ±  4%     +47.6%      53329 ± 13% 
>>>> interrupts.CPU85.TLB:TLB_shootdowns
>>>>        4693 ±  7%   +1323.0%      66793 ±145% 
>>>> interrupts.CPU86.RES:Rescheduling_interrupts
>>>>       38003 ± 11%     +94.8%      74031 ± 43% 
>>>> interrupts.CPU86.TLB:TLB_shootdowns
>>>>       78022 ±  3%      +7.9%      84210 ±  3% 
>>>> interrupts.CPU87.CAL:Function_call_interrupts
>>>>       36359 ±  6%     +54.9%      56304 ± 48% 
>>>> interrupts.CPU88.TLB:TLB_shootdowns
>>>>       89031 ±105%     -95.0%       4475 ± 40% 
>>>> interrupts.CPU9.RES:Rescheduling_interrupts
>>>>       40085 ± 11%     +60.6%      64368 ± 27% 
>>>> interrupts.CPU91.TLB:TLB_shootdowns
>>>>       42244 ± 10%     +44.8%      61162 ± 35% 
>>>> interrupts.CPU94.TLB:TLB_shootdowns
>>>>       40959 ± 15%    +109.4%      85780 ± 41% 
>>>> interrupts.CPU95.TLB:TLB_shootdowns
>>>>
>>>>
>>>>                                  stress-ng.fiemap.ops
>>>>    80000 
>>>> +-------------------------------------------------------------------+
>>>>    75000 |..+.             .+..            .+..+..  .+. 
>>>> .+..                |
>>>>          |    +..+..+..+.+.     .+..+..  .+       +.   +. 
>>>> +.+..+..+..+.+..|
>>>>    70000 |-+                   + 
>>>> +.                                    |
>>>>    65000 
>>>> |-+                                                                 |
>>>>    60000 
>>>> |-+                                                                 |
>>>>    55000 
>>>> |-+                                                                 |
>>>> |                                                                   |
>>>>    50000 
>>>> |-+                                                                 |
>>>>    45000 
>>>> |-+                                                                 |
>>>>    40000 
>>>> |-+                                                                 |
>>>>    35000 |-+ O                                                |
>>>>          |  O       O                       O     O 
>>>> O                     |
>>>>    30000 |-+  O  O     O O     O O     O  O    O     O    O  O O  O 
>>>> O  O O  |
>>>>    25000 
>>>> +-------------------------------------------------------------------+
>>>>                              stress-ng.fiemap.ops_per_sec
>>>>    80000 
>>>> +-------------------------------------------------------------------+
>>>>    75000 |..               .+.. .+..                           |
>>>>          |  +.  .+..+..+.+.     .+..+..  .+.+. 
>>>> +..+.+..+..+.+..+..+..+.+..|
>>>>    70000 |-+  +.               + 
>>>> +.                                    |
>>>>    65000 
>>>> |-+                                                                 |
>>>>    60000 
>>>> |-+                                                                 |
>>>>    55000 
>>>> |-+                                                                 |
>>>> |                                                                   |
>>>>    50000 
>>>> |-+                                                                 |
>>>>    45000 
>>>> |-+                                                                 |
>>>>    40000 
>>>> |-+                                                                 |
>>>>    35000 |-+ O                                                |
>>>>          |  O       O                       O 
>>>> O                          |
>>>>    30000 |-+  O  O     O O     O O     O       O     O O  O  O    O 
>>>> O  O O  |
>>>>    25000 
>>>> +-------------------------------------------------------------------+
>>>> [*] bisect-good sample
>>>> [O] bisect-bad  sample
>>>>
>>>>
>>>>
>>>> Disclaimer:
>>>> Results have been estimated based on internal Intel analysis and are 
>>>> provided
>>>> for informational purposes only. Any difference in system hardware 
>>>> or software
>>>> design or configuration may affect actual performance.
>>>>
>>>>
>>>> Thanks,
>>>> Rong Chen
>>>>
>>>>
>>>> _______________________________________________
>>>> LKP mailing list -- lkp@lists.01.org
>>>> To unsubscribe send an email to lkp-leave@lists.01.org
>>>>
>>>
>>
> 

-- 
Zhengjun Xing

WARNING: multiple messages have this Message-ID (diff)
From: Xing Zhengjun <zhengjun.xing@linux.intel.com>
To: lkp@lists.01.org
Subject: Re: [ext4] d3b6f23f71: stress-ng.fiemap.ops_per_sec -60.5% regression
Date: Tue, 16 Jun 2020 15:13:47 +0800	[thread overview]
Message-ID: <7bba55fd-955a-355e-4df1-7b4493b125a6@linux.intel.com> (raw)
In-Reply-To: <bf06a978-828a-9676-0ff8-fa47826d0bdd@linux.intel.com>

[-- Attachment #1: Type: text/plain, Size: 40763 bytes --]

Hi Ritesh,

    I test, the regression still existed in v5.8-rc1. Do you have time 
to take a look at it? Thanks.

On 4/14/2020 1:49 PM, Xing Zhengjun wrote:
> Thanks for your quick response, if you need any more test information 
> about the regression, please let me known.
> 
> On 4/13/2020 6:56 PM, Ritesh Harjani wrote:
>>
>>
>> On 4/13/20 2:07 PM, Xing Zhengjun wrote:
>>> Hi Harjani,
>>>
>>>     Do you have time to take a look at this? Thanks.
>>
>> Hello Xing,
>>
>> I do want to look into this. But as of now I am stuck with another
>> mballoc failure issue. I will get back at this once I have some handle
>> over that one.
>>
>> BTW, are you planning to take look at this?
>>
>> -ritesh
>>
>>
>>>
>>> On 4/7/2020 4:00 PM, kernel test robot wrote:
>>>> Greeting,
>>>>
>>>> FYI, we noticed a -60.5% regression of stress-ng.fiemap.ops_per_sec 
>>>> due to commit:
>>>>
>>>>
>>>> commit: d3b6f23f71670007817a5d59f3fbafab2b794e8c ("ext4: move 
>>>> ext4_fiemap to use iomap framework")
>>>> https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master
>>>>
>>>> in testcase: stress-ng
>>>> on test machine: 96 threads Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz 
>>>> with 192G memory
>>>> with following parameters:
>>>>
>>>>     nr_threads: 10%
>>>>     disk: 1HDD
>>>>     testtime: 1s
>>>>     class: os
>>>>     cpufreq_governor: performance
>>>>     ucode: 0x500002c
>>>>     fs: ext4
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> 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
>>>>
>>>> ========================================================================================= 
>>>>
>>>> class/compiler/cpufreq_governor/disk/fs/kconfig/nr_threads/rootfs/tbox_group/testcase/testtime/ucode: 
>>>>
>>>> os/gcc-7/performance/1HDD/ext4/x86_64-rhel-7.6/10%/debian-x86_64-20191114.cgz/lkp-csl-2sp5/stress-ng/1s/0x500002c 
>>>>
>>>>
>>>> commit:
>>>>    b2c5764262 ("ext4: make ext4_ind_map_blocks work with fiemap")
>>>>    d3b6f23f71 ("ext4: move ext4_fiemap to use iomap framework")
>>>>
>>>> b2c5764262edded1 d3b6f23f71670007817a5d59f3f
>>>> ---------------- ---------------------------
>>>>         fail:runs  %reproduction    fail:runs
>>>>             |             |             |
>>>>             :4           25%           1:4 
>>>> dmesg.WARNING:at#for_ip_interrupt_entry/0x
>>>>            2:4            5%           2:4 
>>>> perf-profile.calltrace.cycles-pp.sync_regs.error_entry
>>>>            2:4            6%           3:4 
>>>> perf-profile.calltrace.cycles-pp.error_entry
>>>>            3:4            9%           3:4 
>>>> perf-profile.children.cycles-pp.error_entry
>>>>            0:4            1%           0:4 
>>>> perf-profile.self.cycles-pp.error_entry
>>>>           %stddev     %change         %stddev
>>>>               \          |                \
>>>>       28623           +28.2%      36703 ± 12%  stress-ng.daemon.ops
>>>>       28632           +28.2%      36704 ± 12% 
>>>> stress-ng.daemon.ops_per_sec
>>>>      566.00 ± 22%     -53.2%     265.00 ± 53%  stress-ng.dev.ops
>>>>      278.81 ± 22%     -53.0%     131.00 ± 54%  
>>>> stress-ng.dev.ops_per_sec
>>>>       73160           -60.6%      28849 ±  3%  stress-ng.fiemap.ops
>>>>       72471           -60.5%      28612 ±  3% 
>>>> stress-ng.fiemap.ops_per_sec
>>>>       23421 ± 12%     +21.2%      28388 ±  6%  stress-ng.filename.ops
>>>>       22638 ± 12%     +20.3%      27241 ±  6% 
>>>> stress-ng.filename.ops_per_sec
>>>>       21.25 ±  7%     -10.6%      19.00 ±  3%  stress-ng.iomix.ops
>>>>       38.75 ± 49%     -47.7%      20.25 ± 96%  stress-ng.memhotplug.ops
>>>>       34.45 ± 52%     -51.8%      16.62 ±106% 
>>>> stress-ng.memhotplug.ops_per_sec
>>>>        1734 ± 10%     +31.4%       2278 ± 10%  stress-ng.resources.ops
>>>>      807.56 ±  5%     +35.2%       1091 ±  8% 
>>>> stress-ng.resources.ops_per_sec
>>>>     1007356 ±  3%     -16.5%     840642 ±  9%  stress-ng.revio.ops
>>>>     1007692 ±  3%     -16.6%     840711 ±  9% 
>>>> stress-ng.revio.ops_per_sec
>>>>       21812 ±  3%     +16.0%      25294 ±  5%  stress-ng.sysbadaddr.ops
>>>>       21821 ±  3%     +15.9%      25294 ±  5% 
>>>> stress-ng.sysbadaddr.ops_per_sec
>>>>      440.75 ±  4%     +21.9%     537.25 ±  9%  stress-ng.sysfs.ops
>>>>      440.53 ±  4%     +21.9%     536.86 ±  9% 
>>>> stress-ng.sysfs.ops_per_sec
>>>>    13286582           -11.1%   11805520 ±  6% 
>>>> stress-ng.time.file_system_outputs
>>>>    68253896            +2.4%   69860122 
>>>> stress-ng.time.minor_page_faults
>>>>      197.00 ±  4%     -15.9%     165.75 ± 12%  stress-ng.xattr.ops
>>>>      192.45 ±  5%     -16.1%     161.46 ± 11% 
>>>> stress-ng.xattr.ops_per_sec
>>>>       15310           +62.5%      24875 ± 22%  stress-ng.zombie.ops
>>>>       15310           +62.5%      24874 ± 22% 
>>>> stress-ng.zombie.ops_per_sec
>>>>      203.50 ± 12%     -47.3%     107.25 ± 49%  vmstat.io.bi
>>>>      861318 ± 18%     -29.7%     605884 ±  5%  meminfo.AnonHugePages
>>>>     1062742 ± 14%     -20.2%     847853 ±  3%  meminfo.AnonPages
>>>>       31093 ±  6%      +9.6%      34090 ±  3%  meminfo.KernelStack
>>>>        7151 ± 34%     +55.8%      11145 ±  9%  meminfo.Mlocked
>>>>   1.082e+08 ±  5%     -40.2%   64705429 ± 31% 
>>>> numa-numastat.node0.local_node
>>>>   1.082e+08 ±  5%     -40.2%   64739883 ± 31% 
>>>> numa-numastat.node0.numa_hit
>>>>    46032662 ± 21%    +104.3%   94042918 ± 20% 
>>>> numa-numastat.node1.local_node
>>>>    46074205 ± 21%    +104.2%   94072810 ± 20% 
>>>> numa-numastat.node1.numa_hit
>>>>        3942 ±  3%     +14.2%       4501 ±  4% 
>>>> slabinfo.pool_workqueue.active_objs
>>>>        4098 ±  3%     +14.3%       4683 ±  4% 
>>>> slabinfo.pool_workqueue.num_objs
>>>>        4817 ±  7%     +13.3%       5456 ±  8% 
>>>> slabinfo.proc_dir_entry.active_objs
>>>>        5153 ±  6%     +12.5%       5797 ±  8% 
>>>> slabinfo.proc_dir_entry.num_objs
>>>>       18598 ± 13%     -33.1%      12437 ± 20% 
>>>> sched_debug.cfs_rq:/.load.avg
>>>>      452595 ± 56%     -71.4%     129637 ± 76% 
>>>> sched_debug.cfs_rq:/.load.max
>>>>       67675 ± 35%     -55.1%      30377 ± 42% 
>>>> sched_debug.cfs_rq:/.load.stddev
>>>>       18114 ± 12%     -33.7%      12011 ± 20% 
>>>> sched_debug.cfs_rq:/.runnable_weight.avg
>>>>      448215 ± 58%     -72.8%     121789 ± 82% 
>>>> sched_debug.cfs_rq:/.runnable_weight.max
>>>>       67083 ± 37%     -56.3%      29305 ± 43% 
>>>> sched_debug.cfs_rq:/.runnable_weight.stddev
>>>>      -38032          +434.3%    -203212 
>>>> sched_debug.cfs_rq:/.spread0.avg
>>>>     -204466           +95.8%    -400301 
>>>> sched_debug.cfs_rq:/.spread0.min
>>>>       90.02 ± 25%     -58.1%      37.69 ± 52% 
>>>> sched_debug.cfs_rq:/.util_est_enqueued.avg
>>>>      677.54 ±  6%     -39.3%     411.50 ± 22% 
>>>> sched_debug.cfs_rq:/.util_est_enqueued.max
>>>>      196.57 ±  8%     -47.6%     103.05 ± 36% 
>>>> sched_debug.cfs_rq:/.util_est_enqueued.stddev
>>>>        3.34 ± 23%     +34.1%       4.48 ±  4% 
>>>> sched_debug.cpu.clock.stddev
>>>>        3.34 ± 23%     +34.1%       4.48 ±  4% 
>>>> sched_debug.cpu.clock_task.stddev
>>>>      402872 ±  7%     -11.9%     354819 ±  2% 
>>>> proc-vmstat.nr_active_anon
>>>>     1730331            -9.5%    1566418 ±  5%  proc-vmstat.nr_dirtied
>>>>       31042 ±  6%      +9.3%      33915 ±  3% 
>>>> proc-vmstat.nr_kernel_stack
>>>>      229047            -2.4%     223615        proc-vmstat.nr_mapped
>>>>       74008 ±  7%     +20.5%      89163 ±  8%  proc-vmstat.nr_written
>>>>      402872 ±  7%     -11.9%     354819 ±  2% 
>>>> proc-vmstat.nr_zone_active_anon
>>>>       50587 ± 11%     -25.2%      37829 ± 14% 
>>>> proc-vmstat.numa_pages_migrated
>>>>      457500           -23.1%     351918 ± 31% 
>>>> proc-vmstat.numa_pte_updates
>>>>    81382485            +1.9%   82907822        proc-vmstat.pgfault
>>>>   2.885e+08 ±  5%     -13.3%  2.502e+08 ±  6%  proc-vmstat.pgfree
>>>>       42206 ± 12%     -46.9%      22399 ± 49%  proc-vmstat.pgpgin
>>>>      431233 ± 13%     -64.8%     151736 ±109%  proc-vmstat.pgrotated
>>>>      176754 ±  7%     -40.2%     105637 ± 31% 
>>>> proc-vmstat.thp_fault_alloc
>>>>      314.50 ± 82%    +341.5%       1388 ± 44% 
>>>> proc-vmstat.unevictable_pgs_stranded
>>>>     1075269 ± 14%     -41.3%     631388 ± 17%  
>>>> numa-meminfo.node0.Active
>>>>      976056 ± 12%     -39.7%     588727 ± 19% 
>>>> numa-meminfo.node0.Active(anon)
>>>>      426857 ± 22%     -36.4%     271375 ± 13% 
>>>> numa-meminfo.node0.AnonHugePages
>>>>      558590 ± 19%     -36.4%     355402 ± 14% 
>>>> numa-meminfo.node0.AnonPages
>>>>     1794824 ±  9%     -28.8%    1277157 ± 20% 
>>>> numa-meminfo.node0.FilePages
>>>>        8517 ± 92%     -82.7%       1473 ± 89% 
>>>> numa-meminfo.node0.Inactive(file)
>>>>      633118 ±  2%     -41.7%     368920 ± 36%  
>>>> numa-meminfo.node0.Mapped
>>>>     2958038 ± 12%     -27.7%    2139271 ± 12% 
>>>> numa-meminfo.node0.MemUsed
>>>>      181401 ±  5%     -13.7%     156561 ±  4% 
>>>> numa-meminfo.node0.SUnreclaim
>>>>      258124 ±  6%     -13.0%     224535 ±  5%  numa-meminfo.node0.Slab
>>>>      702083 ± 16%     +31.0%     919406 ± 11%  
>>>> numa-meminfo.node1.Active
>>>>       38663 ±107%    +137.8%      91951 ± 31% 
>>>> numa-meminfo.node1.Active(file)
>>>>     1154975 ±  7%     +41.6%    1635593 ± 12% 
>>>> numa-meminfo.node1.FilePages
>>>>      395813 ± 25%     +62.8%     644533 ± 16% 
>>>> numa-meminfo.node1.Inactive
>>>>      394313 ± 25%     +62.5%     640686 ± 16% 
>>>> numa-meminfo.node1.Inactive(anon)
>>>>      273317           +88.8%     515976 ± 25%  
>>>> numa-meminfo.node1.Mapped
>>>>     2279237 ±  6%     +25.7%    2865582 ±  7% 
>>>> numa-meminfo.node1.MemUsed
>>>>       10830 ± 18%     +29.6%      14033 ±  9% 
>>>> numa-meminfo.node1.PageTables
>>>>      149390 ±  3%     +23.2%     184085 ±  3% 
>>>> numa-meminfo.node1.SUnreclaim
>>>>      569542 ± 16%     +74.8%     995336 ± 21%  numa-meminfo.node1.Shmem
>>>>      220774 ±  5%     +20.3%     265656 ±  3%  numa-meminfo.node1.Slab
>>>>    35623587 ±  5%     -11.7%   31444514 ±  3%  perf-stat.i.cache-misses
>>>>   2.576e+08 ±  5%      -6.8%    2.4e+08 ±  2% 
>>>> perf-stat.i.cache-references
>>>>        3585            -7.3%       3323 ±  5% 
>>>> perf-stat.i.cpu-migrations
>>>>      180139 ±  2%      +4.2%     187668        perf-stat.i.minor-faults
>>>>       69.13            +2.6       71.75 
>>>> perf-stat.i.node-load-miss-rate%
>>>>     4313695 ±  2%      -7.4%    3994957 ±  2% 
>>>> perf-stat.i.node-load-misses
>>>>     5466253 ± 11%     -17.3%    4521173 ±  6%  perf-stat.i.node-loads
>>>>     2818674 ±  6%     -15.8%    2372542 ±  5%  perf-stat.i.node-stores
>>>>      227810            +4.6%     238290        perf-stat.i.page-faults
>>>>       12.67 ±  4%      -7.2%      11.76 ±  2%  perf-stat.overall.MPKI
>>>>        1.01 ±  4%      -0.0        0.97 ±  3% 
>>>> perf-stat.overall.branch-miss-rate%
>>>>        1044           +13.1%       1181 ±  4% 
>>>> perf-stat.overall.cycles-between-cache-misses
>>>>       40.37 ±  4%      +3.6       44.00 ±  2% 
>>>> perf-stat.overall.node-store-miss-rate%
>>>>    36139526 ±  5%     -12.5%   31625519 ±  3%  
>>>> perf-stat.ps.cache-misses
>>>>   2.566e+08 ±  5%      -6.9%  2.389e+08 ±  2% 
>>>> perf-stat.ps.cache-references
>>>>        3562            -7.2%       3306 ±  5% 
>>>> perf-stat.ps.cpu-migrations
>>>>      179088            +4.2%     186579        
>>>> perf-stat.ps.minor-faults
>>>>     4323383 ±  2%      -7.5%    3999214 perf-stat.ps.node-load-misses
>>>>     5607721 ± 10%     -18.5%    4568664 ±  6%  perf-stat.ps.node-loads
>>>>     2855134 ±  7%     -16.4%    2387345 ±  5%  perf-stat.ps.node-stores
>>>>      226270            +4.6%     236709        perf-stat.ps.page-faults
>>>>      242305 ± 10%     -42.4%     139551 ± 18% 
>>>> numa-vmstat.node0.nr_active_anon
>>>>      135983 ± 17%     -37.4%      85189 ± 10% 
>>>> numa-vmstat.node0.nr_anon_pages
>>>>      209.25 ± 16%     -38.1%     129.50 ± 10% 
>>>> numa-vmstat.node0.nr_anon_transparent_hugepages
>>>>      449367 ±  9%     -29.7%     315804 ± 20% 
>>>> numa-vmstat.node0.nr_file_pages
>>>>        2167 ± 90%     -80.6%     419.75 ± 98% 
>>>> numa-vmstat.node0.nr_inactive_file
>>>>      157405 ±  3%     -41.4%      92206 ± 35% 
>>>> numa-vmstat.node0.nr_mapped
>>>>        2022 ± 30%     -73.3%     539.25 ± 91% 
>>>> numa-vmstat.node0.nr_mlock
>>>>        3336 ± 10%     -24.3%       2524 ± 25% 
>>>> numa-vmstat.node0.nr_page_table_pages
>>>>      286158 ± 10%     -41.2%     168337 ± 37% 
>>>> numa-vmstat.node0.nr_shmem
>>>>       45493 ±  5%     -14.1%      39094 ±  4% 
>>>> numa-vmstat.node0.nr_slab_unreclaimable
>>>>      242294 ± 10%     -42.4%     139547 ± 18% 
>>>> numa-vmstat.node0.nr_zone_active_anon
>>>>        2167 ± 90%     -80.6%     419.75 ± 98% 
>>>> numa-vmstat.node0.nr_zone_inactive_file
>>>>    54053924 ±  8%     -39.3%   32786242 ± 34% 
>>>> numa-vmstat.node0.numa_hit
>>>>    53929628 ±  8%     -39.5%   32619715 ± 34% 
>>>> numa-vmstat.node0.numa_local
>>>>        9701 ±107%    +136.9%      22985 ± 31% 
>>>> numa-vmstat.node1.nr_active_file
>>>>      202.50 ± 16%     -25.1%     151.75 ± 23% 
>>>> numa-vmstat.node1.nr_anon_transparent_hugepages
>>>>      284922 ±  7%     +43.3%     408195 ± 13% 
>>>> numa-vmstat.node1.nr_file_pages
>>>>       96002 ± 26%     +67.5%     160850 ± 17% 
>>>> numa-vmstat.node1.nr_inactive_anon
>>>>       68077 ±  2%     +90.3%     129533 ± 25% 
>>>> numa-vmstat.node1.nr_mapped
>>>>      138482 ± 15%     +79.2%     248100 ± 22% 
>>>> numa-vmstat.node1.nr_shmem
>>>>       37396 ±  3%     +23.3%      46094 ±  3% 
>>>> numa-vmstat.node1.nr_slab_unreclaimable
>>>>        9701 ±107%    +136.9%      22985 ± 31% 
>>>> numa-vmstat.node1.nr_zone_active_file
>>>>       96005 ± 26%     +67.5%     160846 ± 17% 
>>>> numa-vmstat.node1.nr_zone_inactive_anon
>>>>    23343661 ± 17%     +99.9%   46664267 ± 23% 
>>>> numa-vmstat.node1.numa_hit
>>>>    23248487 ± 17%    +100.5%   46610447 ± 23% 
>>>> numa-vmstat.node1.numa_local
>>>>      105745 ± 23%    +112.6%     224805 ± 24%  softirqs.CPU0.NET_RX
>>>>      133310 ± 36%     -45.3%      72987 ± 52%  softirqs.CPU1.NET_RX
>>>>      170110 ± 55%     -66.8%      56407 ±147%  softirqs.CPU11.NET_RX
>>>>       91465 ± 36%     -65.2%      31858 ±112%  softirqs.CPU13.NET_RX
>>>>      164491 ± 57%     -77.7%      36641 ±121%  softirqs.CPU15.NET_RX
>>>>      121069 ± 55%     -99.3%     816.75 ± 96%  softirqs.CPU17.NET_RX
>>>>       81019 ±  4%      -8.7%      73967 ±  4%  softirqs.CPU20.RCU
>>>>       72143 ± 63%     -89.8%       7360 ±172%  softirqs.CPU22.NET_RX
>>>>      270663 ± 17%     -57.9%     113915 ± 45%  softirqs.CPU24.NET_RX
>>>>       20149 ± 76%    +474.1%     115680 ± 62%  softirqs.CPU26.NET_RX
>>>>       14033 ± 70%    +977.5%     151211 ± 75%  softirqs.CPU27.NET_RX
>>>>       27834 ± 94%    +476.1%     160357 ± 28%  softirqs.CPU28.NET_RX
>>>>       35346 ± 68%    +212.0%     110290 ± 30%  softirqs.CPU29.NET_RX
>>>>       34347 ±103%    +336.5%     149941 ± 32%  softirqs.CPU32.NET_RX
>>>>       70077 ±  3%     +10.8%      77624 ±  3%  softirqs.CPU34.RCU
>>>>       36453 ± 84%    +339.6%     160253 ± 42%  softirqs.CPU36.NET_RX
>>>>       72367 ±  2%     +10.6%      80043        softirqs.CPU37.RCU
>>>>       25239 ±118%    +267.7%      92799 ± 45%  softirqs.CPU38.NET_RX
>>>>        4995 ±170%   +1155.8%      62734 ± 62%  softirqs.CPU39.NET_RX
>>>>        4641 ±145%   +1611.3%      79432 ± 90%  softirqs.CPU42.NET_RX
>>>>        7192 ± 65%    +918.0%      73225 ± 66%  softirqs.CPU45.NET_RX
>>>>        1772 ±166%   +1837.4%      34344 ± 63%  softirqs.CPU46.NET_RX
>>>>       13149 ± 81%    +874.7%     128170 ± 58%  softirqs.CPU47.NET_RX
>>>>       86484 ± 94%     -92.6%       6357 ±172%  softirqs.CPU48.NET_RX
>>>>      129128 ± 27%     -95.8%       5434 ±172%  softirqs.CPU55.NET_RX
>>>>       82772 ± 59%     -91.7%       6891 ±164%  softirqs.CPU56.NET_RX
>>>>      145313 ± 57%     -87.8%      17796 ± 88%  softirqs.CPU57.NET_RX
>>>>      118160 ± 33%     -86.3%      16226 ±109%  softirqs.CPU58.NET_RX
>>>>       94576 ± 56%     -94.1%       5557 ±173%  softirqs.CPU6.NET_RX
>>>>       82900 ± 77%     -66.8%      27508 ±171%  softirqs.CPU62.NET_RX
>>>>      157291 ± 30%     -81.1%      29656 ±111%  softirqs.CPU64.NET_RX
>>>>      135101 ± 28%     -80.2%      26748 ± 90%  softirqs.CPU67.NET_RX
>>>>      146574 ± 56%    -100.0%      69.75 ± 98%  softirqs.CPU68.NET_RX
>>>>       81347 ±  2%      -9.0%      74024 ±  2%  softirqs.CPU68.RCU
>>>>      201729 ± 37%     -99.6%     887.50 ±107%  softirqs.CPU69.NET_RX
>>>>      108454 ± 78%     -97.9%       2254 ±169%  softirqs.CPU70.NET_RX
>>>>       55289 ±104%     -89.3%       5942 ±172%  softirqs.CPU71.NET_RX
>>>>       10112 ±172%    +964.6%     107651 ± 89%  softirqs.CPU72.NET_RX
>>>>        3136 ±171%   +1522.2%      50879 ± 66%  softirqs.CPU73.NET_RX
>>>>       13353 ± 79%    +809.2%     121407 ±101%  softirqs.CPU74.NET_RX
>>>>       75194 ±  3%     +10.3%      82957 ±  5%  softirqs.CPU75.RCU
>>>>       11002 ±173%   +1040.8%     125512 ± 61%  softirqs.CPU76.NET_RX
>>>>        2463 ±173%   +2567.3%      65708 ± 77%  softirqs.CPU78.NET_RX
>>>>       25956 ±  3%      -7.8%      23932 ±  3%  softirqs.CPU78.SCHED
>>>>       16366 ±150%    +340.7%      72125 ± 91%  softirqs.CPU82.NET_RX
>>>>       14553 ±130%   +1513.4%     234809 ± 27%  softirqs.CPU93.NET_RX
>>>>       26314            -9.2%      23884 ±  3%  softirqs.CPU93.SCHED
>>>>        4582 ± 88%   +4903.4%     229268 ± 23%  softirqs.CPU94.NET_RX
>>>>       11214 ±111%   +1762.5%     208867 ± 18%  softirqs.CPU95.NET_RX
>>>>        1.53 ± 27%      -0.5        0.99 ± 17% 
>>>> perf-profile.calltrace.cycles-pp.exit_to_usermode_loop.do_syscall_64.entry_SYSCALL_64_after_hwframe 
>>>>
>>>>        1.52 ± 27%      -0.5        0.99 ± 17% 
>>>> perf-profile.calltrace.cycles-pp.do_signal.exit_to_usermode_loop.do_syscall_64.entry_SYSCALL_64_after_hwframe 
>>>>
>>>>        1.39 ± 29%      -0.5        0.88 ± 21% 
>>>> perf-profile.calltrace.cycles-pp.do_group_exit.get_signal.do_signal.exit_to_usermode_loop.do_syscall_64 
>>>>
>>>>        1.39 ± 29%      -0.5        0.88 ± 21% 
>>>> perf-profile.calltrace.cycles-pp.get_signal.do_signal.exit_to_usermode_loop.do_syscall_64.entry_SYSCALL_64_after_hwframe 
>>>>
>>>>        0.50 ± 59%      +0.3        0.81 ± 13% 
>>>> perf-profile.calltrace.cycles-pp.filemap_map_pages.handle_pte_fault.__handle_mm_fault.handle_mm_fault.do_page_fault 
>>>>
>>>>        5.70 ±  9%      +0.8        6.47 ±  7% 
>>>> perf-profile.calltrace.cycles-pp.do_exit.do_group_exit.get_signal.do_signal.exit_to_usermode_loop 
>>>>
>>>>        5.48 ±  9%      +0.8        6.27 ±  7% 
>>>> perf-profile.calltrace.cycles-pp.exit_mmap.mmput.do_exit.do_group_exit.get_signal 
>>>>
>>>>        5.49 ±  9%      +0.8        6.28 ±  7% 
>>>> perf-profile.calltrace.cycles-pp.mmput.do_exit.do_group_exit.get_signal.do_signal 
>>>>
>>>>        4.30 ±  4%      +1.3        5.60 ±  7% 
>>>> perf-profile.calltrace.cycles-pp.do_group_exit.get_signal.do_signal.exit_to_usermode_loop.prepare_exit_to_usermode 
>>>>
>>>>        4.40 ±  4%      +1.3        5.69 ±  7% 
>>>> perf-profile.calltrace.cycles-pp.prepare_exit_to_usermode.swapgs_restore_regs_and_return_to_usermode 
>>>>
>>>>        4.37 ±  4%      +1.3        5.66 ±  7% 
>>>> perf-profile.calltrace.cycles-pp.exit_to_usermode_loop.prepare_exit_to_usermode.swapgs_restore_regs_and_return_to_usermode 
>>>>
>>>>        4.36 ±  4%      +1.3        5.66 ±  7% 
>>>> perf-profile.calltrace.cycles-pp.do_signal.exit_to_usermode_loop.prepare_exit_to_usermode.swapgs_restore_regs_and_return_to_usermode 
>>>>
>>>>        4.33 ±  4%      +1.3        5.62 ±  7% 
>>>> perf-profile.calltrace.cycles-pp.get_signal.do_signal.exit_to_usermode_loop.prepare_exit_to_usermode.swapgs_restore_regs_and_return_to_usermode 
>>>>
>>>>        4.44 ±  4%      +1.3        5.74 ±  7% 
>>>> perf-profile.calltrace.cycles-pp.swapgs_restore_regs_and_return_to_usermode 
>>>>
>>>>        3.20 ± 10%      -2.4        0.78 ±156% 
>>>> perf-profile.children.cycles-pp.copy_page
>>>>        0.16 ±  9%      -0.1        0.08 ± 64% 
>>>> perf-profile.children.cycles-pp.irq_work_interrupt
>>>>        0.16 ±  9%      -0.1        0.08 ± 64% 
>>>> perf-profile.children.cycles-pp.smp_irq_work_interrupt
>>>>        0.24 ±  5%      -0.1        0.17 ± 18% 
>>>> perf-profile.children.cycles-pp.irq_work_run_list
>>>>        0.16 ±  9%      -0.1        0.10 ± 24% 
>>>> perf-profile.children.cycles-pp.irq_work_run
>>>>        0.16 ±  9%      -0.1        0.10 ± 24% 
>>>> perf-profile.children.cycles-pp.printk
>>>>        0.23 ±  6%      -0.1        0.17 ±  9% 
>>>> perf-profile.children.cycles-pp.__do_execve_file
>>>>        0.08 ± 14%      -0.1        0.03 ±100% 
>>>> perf-profile.children.cycles-pp.delay_tsc
>>>>        0.16 ±  6%      -0.1        0.11 ±  9% 
>>>> perf-profile.children.cycles-pp.load_elf_binary
>>>>        0.16 ±  7%      -0.0        0.12 ± 13% 
>>>> perf-profile.children.cycles-pp.search_binary_handler
>>>>        0.20 ±  7%      -0.0        0.15 ± 10% 
>>>> perf-profile.children.cycles-pp.call_usermodehelper_exec_async
>>>>        0.19 ±  6%      -0.0        0.15 ± 11% 
>>>> perf-profile.children.cycles-pp.do_execve
>>>>        0.08 ± 10%      -0.0        0.04 ± 59% 
>>>> perf-profile.children.cycles-pp.__vunmap
>>>>        0.15 ±  3%      -0.0        0.11 ±  7% 
>>>> perf-profile.children.cycles-pp.rcu_idle_exit
>>>>        0.12 ± 10%      -0.0        0.09 ± 14% 
>>>> perf-profile.children.cycles-pp.__switch_to_asm
>>>>        0.09 ± 13%      -0.0        0.07 ±  5% 
>>>> perf-profile.children.cycles-pp.des3_ede_encrypt
>>>>        0.06 ± 11%      +0.0        0.09 ± 13% 
>>>> perf-profile.children.cycles-pp.mark_page_accessed
>>>>        0.15 ±  5%      +0.0        0.19 ± 12% 
>>>> perf-profile.children.cycles-pp.apparmor_cred_prepare
>>>>        0.22 ±  8%      +0.0        0.27 ± 11% 
>>>> perf-profile.children.cycles-pp.mem_cgroup_throttle_swaprate
>>>>        0.17 ±  2%      +0.0        0.22 ± 12% 
>>>> perf-profile.children.cycles-pp.security_prepare_creds
>>>>        0.95 ± 17%      +0.3        1.22 ± 14% 
>>>> perf-profile.children.cycles-pp.filemap_map_pages
>>>>        5.92 ±  8%      +0.7        6.65 ±  7% 
>>>> perf-profile.children.cycles-pp.get_signal
>>>>        5.66 ±  9%      +0.8        6.44 ±  7% 
>>>> perf-profile.children.cycles-pp.mmput
>>>>        5.65 ±  9%      +0.8        6.43 ±  7% 
>>>> perf-profile.children.cycles-pp.exit_mmap
>>>>        4.40 ±  4%      +1.3        5.70 ±  7% 
>>>> perf-profile.children.cycles-pp.prepare_exit_to_usermode
>>>>        4.45 ±  4%      +1.3        5.75 ±  7% 
>>>> perf-profile.children.cycles-pp.swapgs_restore_regs_and_return_to_usermode 
>>>>
>>>>        3.16 ± 10%      -2.4        0.77 ±155% 
>>>> perf-profile.self.cycles-pp.copy_page
>>>>        0.08 ± 14%      -0.1        0.03 ±100% 
>>>> perf-profile.self.cycles-pp.delay_tsc
>>>>        0.12 ± 10%      -0.0        0.09 ± 14% 
>>>> perf-profile.self.cycles-pp.__switch_to_asm
>>>>        0.08 ± 12%      -0.0        0.06 ± 17% 
>>>> perf-profile.self.cycles-pp.enqueue_task_fair
>>>>        0.09 ± 13%      -0.0        0.07 ±  5% 
>>>> perf-profile.self.cycles-pp.des3_ede_encrypt
>>>>        0.07 ± 13%      +0.0        0.08 ± 19% 
>>>> perf-profile.self.cycles-pp.__lru_cache_add
>>>>        0.19 ±  9%      +0.0        0.22 ± 10% 
>>>> perf-profile.self.cycles-pp.mem_cgroup_throttle_swaprate
>>>>        0.15 ±  5%      +0.0        0.19 ± 11% 
>>>> perf-profile.self.cycles-pp.apparmor_cred_prepare
>>>>        0.05 ± 58%      +0.0        0.09 ± 13% 
>>>> perf-profile.self.cycles-pp.mark_page_accessed
>>>>        0.58 ± 10%      +0.2        0.80 ± 20% 
>>>> perf-profile.self.cycles-pp.release_pages
>>>>        0.75 ±173%  +1.3e+05%       1005 ±100% 
>>>> interrupts.127:PCI-MSI.31981660-edge.i40e-eth0-TxRx-91
>>>>      820.75 ±111%     -99.9%       0.50 ±173% 
>>>> interrupts.47:PCI-MSI.31981580-edge.i40e-eth0-TxRx-11
>>>>      449.25 ± 86%    -100.0%       0.00 
>>>> interrupts.53:PCI-MSI.31981586-edge.i40e-eth0-TxRx-17
>>>>       33.25 ±157%    -100.0%       0.00 
>>>> interrupts.57:PCI-MSI.31981590-edge.i40e-eth0-TxRx-21
>>>>        0.75 ±110%  +63533.3%     477.25 ±162% 
>>>> interrupts.61:PCI-MSI.31981594-edge.i40e-eth0-TxRx-25
>>>>      561.50 ±160%    -100.0%       0.00 
>>>> interrupts.65:PCI-MSI.31981598-edge.i40e-eth0-TxRx-29
>>>>       82921 ±  8%     -11.1%      73748 ±  6% 
>>>> interrupts.CPU11.CAL:Function_call_interrupts
>>>>       66509 ± 30%     -32.6%      44828 ±  8% 
>>>> interrupts.CPU14.TLB:TLB_shootdowns
>>>>       43105 ± 98%     -90.3%       4183 ± 21% 
>>>> interrupts.CPU17.RES:Rescheduling_interrupts
>>>>      148719 ± 70%     -69.4%      45471 ± 16% 
>>>> interrupts.CPU17.TLB:TLB_shootdowns
>>>>       85589 ± 42%     -52.2%      40884 ±  5% 
>>>> interrupts.CPU20.TLB:TLB_shootdowns
>>>>      222472 ± 41%     -98.0%       4360 ± 45% 
>>>> interrupts.CPU22.RES:Rescheduling_interrupts
>>>>        0.50 ±173%  +95350.0%     477.25 ±162% 
>>>> interrupts.CPU25.61:PCI-MSI.31981594-edge.i40e-eth0-TxRx-25
>>>>       76029 ± 10%     +14.9%      87389 ±  5% 
>>>> interrupts.CPU25.CAL:Function_call_interrupts
>>>>      399042 ±  6%     +13.4%     452479 ±  8% 
>>>> interrupts.CPU27.LOC:Local_timer_interrupts
>>>>      561.00 ±161%    -100.0%       0.00 
>>>> interrupts.CPU29.65:PCI-MSI.31981598-edge.i40e-eth0-TxRx-29
>>>>        7034 ± 46%   +1083.8%      83279 ±138% 
>>>> interrupts.CPU29.RES:Rescheduling_interrupts
>>>>       17829 ± 99%     -71.0%       5172 ± 16% 
>>>> interrupts.CPU30.RES:Rescheduling_interrupts
>>>>        5569 ± 15%   +2414.7%     140059 ± 94% 
>>>> interrupts.CPU31.RES:Rescheduling_interrupts
>>>>       37674 ± 16%     +36.6%      51473 ± 25% 
>>>> interrupts.CPU31.TLB:TLB_shootdowns
>>>>       47905 ± 39%     +76.6%      84583 ± 38% 
>>>> interrupts.CPU34.TLB:TLB_shootdowns
>>>>      568.75 ±140%    +224.8%       1847 ± 90% 
>>>> interrupts.CPU36.NMI:Non-maskable_interrupts
>>>>      568.75 ±140%    +224.8%       1847 ± 90% 
>>>> interrupts.CPU36.PMI:Performance_monitoring_interrupts
>>>>        4236 ± 25%   +2168.5%      96092 ± 90% 
>>>> interrupts.CPU36.RES:Rescheduling_interrupts
>>>>       52717 ± 27%     +43.3%      75565 ± 28% 
>>>> interrupts.CPU37.TLB:TLB_shootdowns
>>>>       41418 ±  9%    +136.6%      98010 ± 50% 
>>>> interrupts.CPU39.TLB:TLB_shootdowns
>>>>        5551 ±  8%    +847.8%      52615 ± 66% 
>>>> interrupts.CPU40.RES:Rescheduling_interrupts
>>>>        4746 ± 25%    +865.9%      45841 ± 91% 
>>>> interrupts.CPU42.RES:Rescheduling_interrupts
>>>>       37556 ± 11%     +24.6%      46808 ±  6% 
>>>> interrupts.CPU42.TLB:TLB_shootdowns
>>>>       21846 ±124%     -84.4%       3415 ± 46% 
>>>> interrupts.CPU48.RES:Rescheduling_interrupts
>>>>      891.50 ± 22%     -35.2%     577.25 ± 40% 
>>>> interrupts.CPU49.NMI:Non-maskable_interrupts
>>>>      891.50 ± 22%     -35.2%     577.25 ± 40% 
>>>> interrupts.CPU49.PMI:Performance_monitoring_interrupts
>>>>       20459 ±120%     -79.2%       4263 ± 14% 
>>>> interrupts.CPU49.RES:Rescheduling_interrupts
>>>>       59840 ± 21%     -23.1%      46042 ± 16% 
>>>> interrupts.CPU5.TLB:TLB_shootdowns
>>>>       65200 ± 19%     -34.5%      42678 ±  9% 
>>>> interrupts.CPU51.TLB:TLB_shootdowns
>>>>       70923 ±153%     -94.0%       4270 ± 29% 
>>>> interrupts.CPU53.RES:Rescheduling_interrupts
>>>>       65312 ± 22%     -28.7%      46578 ± 14% 
>>>> interrupts.CPU56.TLB:TLB_shootdowns
>>>>       65828 ± 24%   �� -33.4%      43846 ±  4% 
>>>> interrupts.CPU59.TLB:TLB_shootdowns
>>>>       72558 ±156%     -93.2%       4906 ±  9% 
>>>> interrupts.CPU6.RES:Rescheduling_interrupts
>>>>       68698 ± 34%     -32.6%      46327 ± 18% 
>>>> interrupts.CPU61.TLB:TLB_shootdowns
>>>>      109745 ± 44%     -57.4%      46711 ± 16% 
>>>> interrupts.CPU62.TLB:TLB_shootdowns
>>>>       89714 ± 44%     -48.5%      46198 ±  7% 
>>>> interrupts.CPU63.TLB:TLB_shootdowns
>>>>       59380 ±136%     -91.5%       5066 ± 13% 
>>>> interrupts.CPU69.RES:Rescheduling_interrupts
>>>>       40094 ± 18%    +133.9%      93798 ± 44% 
>>>> interrupts.CPU78.TLB:TLB_shootdowns
>>>>      129884 ± 72%     -55.3%      58034 ±157% 
>>>> interrupts.CPU8.RES:Rescheduling_interrupts
>>>>       69984 ± 11%     +51.4%     105957 ± 20% 
>>>> interrupts.CPU80.CAL:Function_call_interrupts
>>>>       32857 ± 10%    +128.7%      75131 ± 36% 
>>>> interrupts.CPU80.TLB:TLB_shootdowns
>>>>       35726 ± 16%     +34.6%      48081 ± 12% 
>>>> interrupts.CPU82.TLB:TLB_shootdowns
>>>>       73820 ± 17%     +28.2%      94643 ±  8% 
>>>> interrupts.CPU84.CAL:Function_call_interrupts
>>>>       38829 ± 28%    +190.3%     112736 ± 42% 
>>>> interrupts.CPU84.TLB:TLB_shootdowns
>>>>       36129 ±  4%     +47.6%      53329 ± 13% 
>>>> interrupts.CPU85.TLB:TLB_shootdowns
>>>>        4693 ±  7%   +1323.0%      66793 ±145% 
>>>> interrupts.CPU86.RES:Rescheduling_interrupts
>>>>       38003 ± 11%     +94.8%      74031 ± 43% 
>>>> interrupts.CPU86.TLB:TLB_shootdowns
>>>>       78022 ±  3%      +7.9%      84210 ±  3% 
>>>> interrupts.CPU87.CAL:Function_call_interrupts
>>>>       36359 ±  6%     +54.9%      56304 ± 48% 
>>>> interrupts.CPU88.TLB:TLB_shootdowns
>>>>       89031 ±105%     -95.0%       4475 ± 40% 
>>>> interrupts.CPU9.RES:Rescheduling_interrupts
>>>>       40085 ± 11%     +60.6%      64368 ± 27% 
>>>> interrupts.CPU91.TLB:TLB_shootdowns
>>>>       42244 ± 10%     +44.8%      61162 ± 35% 
>>>> interrupts.CPU94.TLB:TLB_shootdowns
>>>>       40959 ± 15%    +109.4%      85780 ± 41% 
>>>> interrupts.CPU95.TLB:TLB_shootdowns
>>>>
>>>>
>>>>                                  stress-ng.fiemap.ops
>>>>    80000 
>>>> +-------------------------------------------------------------------+
>>>>    75000 |..+.             .+..            .+..+..  .+. 
>>>> .+..                |
>>>>          |    +..+..+..+.+.     .+..+..  .+       +.   +. 
>>>> +.+..+..+..+.+..|
>>>>    70000 |-+                   + 
>>>> +.                                    |
>>>>    65000 
>>>> |-+                                                                 |
>>>>    60000 
>>>> |-+                                                                 |
>>>>    55000 
>>>> |-+                                                                 |
>>>> |                                                                   |
>>>>    50000 
>>>> |-+                                                                 |
>>>>    45000 
>>>> |-+                                                                 |
>>>>    40000 
>>>> |-+                                                                 |
>>>>    35000 |-+ O                                                |
>>>>          |  O       O                       O     O 
>>>> O                     |
>>>>    30000 |-+  O  O     O O     O O     O  O    O     O    O  O O  O 
>>>> O  O O  |
>>>>    25000 
>>>> +-------------------------------------------------------------------+
>>>>                              stress-ng.fiemap.ops_per_sec
>>>>    80000 
>>>> +-------------------------------------------------------------------+
>>>>    75000 |..               .+.. .+..                           |
>>>>          |  +.  .+..+..+.+.     .+..+..  .+.+. 
>>>> +..+.+..+..+.+..+..+..+.+..|
>>>>    70000 |-+  +.               + 
>>>> +.                                    |
>>>>    65000 
>>>> |-+                                                                 |
>>>>    60000 
>>>> |-+                                                                 |
>>>>    55000 
>>>> |-+                                                                 |
>>>> |                                                                   |
>>>>    50000 
>>>> |-+                                                                 |
>>>>    45000 
>>>> |-+                                                                 |
>>>>    40000 
>>>> |-+                                                                 |
>>>>    35000 |-+ O                                                |
>>>>          |  O       O                       O 
>>>> O                          |
>>>>    30000 |-+  O  O     O O     O O     O       O     O O  O  O    O 
>>>> O  O O  |
>>>>    25000 
>>>> +-------------------------------------------------------------------+
>>>> [*] bisect-good sample
>>>> [O] bisect-bad  sample
>>>>
>>>>
>>>>
>>>> Disclaimer:
>>>> Results have been estimated based on internal Intel analysis and are 
>>>> provided
>>>> for informational purposes only. Any difference in system hardware 
>>>> or software
>>>> design or configuration may affect actual performance.
>>>>
>>>>
>>>> Thanks,
>>>> Rong Chen
>>>>
>>>>
>>>> _______________________________________________
>>>> LKP mailing list -- lkp(a)lists.01.org
>>>> To unsubscribe send an email to lkp-leave(a)lists.01.org
>>>>
>>>
>>
> 

-- 
Zhengjun Xing

  reply	other threads:[~2020-06-16  7:26 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-04-07  8:00 [ext4] d3b6f23f71: stress-ng.fiemap.ops_per_sec -60.5% regression kernel test robot
2020-04-07  8:00 ` kernel test robot
2020-04-13  8:37 ` [LKP] " Xing Zhengjun
2020-04-13  8:37   ` Xing Zhengjun
2020-04-13 10:56   ` [LKP] " Ritesh Harjani
2020-04-14  5:49     ` Xing Zhengjun
2020-04-14  5:49       ` Xing Zhengjun
2020-06-16  7:13       ` Xing Zhengjun [this message]
2020-06-16  7:13         ` Xing Zhengjun
2020-07-15 11:04 ` Ritesh Harjani
2020-07-22  6:17   ` Xing Zhengjun
2020-08-19  1:52     ` [LKP] " Xing Zhengjun
2020-08-19  1:52       ` Xing Zhengjun
2020-08-19 17:49       ` [LKP] " Theodore Y. Ts'o
2020-08-19 17:49         ` Theodore Y. Ts'o

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=7bba55fd-955a-355e-4df1-7b4493b125a6@linux.intel.com \
    --to=zhengjun.xing@linux.intel.com \
    --cc=darrick.wong@oracle.com \
    --cc=jack@suse.cz \
    --cc=linux-kernel@vger.kernel.org \
    --cc=lkp@intel.com \
    --cc=lkp@lists.01.org \
    --cc=riteshh@linux.ibm.com \
    --cc=rong.a.chen@intel.com \
    --cc=tytso@mit.edu \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.