lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [day] [month] [year] [list]
Message-ID: <20191002140503.GB1741@shao2-debian>
Date:   Wed, 2 Oct 2019 22:05:03 +0800
From:   kernel test robot <rong.a.chen@...el.com>
To:     Kairui Song <kasong@...hat.com>
Cc:     Thomas Gleixner <tglx@...utronix.de>, Baoquan He <bhe@...hat.com>,
        Jiri Bohac <jbohac@...e.cz>, Borislav Petkov <bp@...en8.de>,
        "H. Peter Anvin" <hpa@...or.com>,
        Alexey Dobriyan <adobriyan@...il.com>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Omar Sandoval <osandov@...com>, Dave Young <dyoung@...hat.com>,
        LKML <linux-kernel@...r.kernel.org>,
        Linus Torvalds <torvalds@...ux-foundation.org>, lkp@...org
Subject: [x86/gart]  ffc8599aa9:  will-it-scale.per_thread_ops -1.3%
 regression

Greeting,

FYI, we noticed a -1.3% regression of will-it-scale.per_thread_ops due to commit:


commit: ffc8599aa9763f39f6736a79da4d1575e7006f9a ("x86/gart: Exclude GART aperture from kcore")
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master

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

	nr_task: 100%
	mode: thread
	test: sched_yield
	cpufreq_governor: performance

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

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

+------------------+--------------------------------------------------------------+
| testcase: change | will-it-scale: will-it-scale.per_thread_ops -2.0% regression |
| test machine     | 192 threads Skylake-SP with 256G memory                      |
| test parameters  | cpufreq_governor=performance                                 |
|                  | mode=thread                                                  |
|                  | nr_task=100%                                                 |
|                  | test=sched_yield                                             |
+------------------+--------------------------------------------------------------+


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/kconfig/mode/nr_task/rootfs/tbox_group/test/testcase:
  gcc-7/performance/x86_64-rhel-7.6/thread/100%/debian-x86_64-2019-05-14.cgz/lkp-csl-2ap4/sched_yield/will-it-scale

commit: 
  f7798711ad ("Merge branch 'x86/cpu' into x86/urgent")
  ffc8599aa9 ("x86/gart: Exclude GART aperture from kcore")

f7798711adeebde3 ffc8599aa9763f39f6736a79da4 
---------------- --------------------------- 
       fail:runs  %reproduction    fail:runs
           |             |             |    
          1:4          -25%            :4     dmesg.WARNING:at#for_ip_interrupt_entry/0x
         %stddev     %change         %stddev
             \          |                \  
   3241706            -1.3%    3199007        will-it-scale.per_thread_ops
 6.224e+08            -1.3%  6.142e+08        will-it-scale.workload
     76990 ± 16%    +186.3%     220413 ± 94%  cpuidle.C6.usage
    137.00            -2.9%     133.00 ±  2%  proc-vmstat.nr_anon_transparent_hugepages
     70.00 ±  3%      -7.9%      64.50 ±  4%  turbostat.PkgTmp
     34.66 ±  3%      +5.2%      36.48        boot-time.boot
      5579 ±  4%      +6.8%       5957 ±  2%  boot-time.idle
    105.50 ± 54%    +589.8%     727.75 ± 99%  interrupts.CPU14.RES:Rescheduling_interrupts
    726.25 ±150%     -93.1%      49.75 ± 22%  interrupts.CPU2.RES:Rescheduling_interrupts
    834.50 ±106%     -92.4%      63.75 ± 54%  interrupts.CPU6.RES:Rescheduling_interrupts
    105.50 ± 77%   +1122.7%       1290 ± 76%  interrupts.CPU8.RES:Rescheduling_interrupts
     81431 ± 11%     -14.6%      69520 ± 12%  numa-vmstat.node1.nr_file_pages
      2295 ±  9%     -29.4%       1620 ±  9%  numa-vmstat.node1.nr_mapped
     38612 ± 22%     -57.9%      16263 ± 43%  numa-vmstat.node2.nr_active_anon
    369.00 ± 74%     -67.5%     120.00 ± 14%  numa-vmstat.node2.nr_page_table_pages
     38612 ± 22%     -57.9%      16263 ± 43%  numa-vmstat.node2.nr_zone_active_anon
     72.21           +15.6%      83.50 ±  8%  sched_debug.cpu.ttwu_count.min
     46.46 ±  2%     +21.1%      56.25 ± 11%  sched_debug.cpu.ttwu_local.min
      0.22 ± 51%    +196.5%       0.65 ± 26%  sched_debug.rt_rq:/.rt_time.avg
     41.84 ± 51%    +196.5%     124.05 ± 26%  sched_debug.rt_rq:/.rt_time.max
      3.01 ± 51%    +196.5%       8.93 ± 26%  sched_debug.rt_rq:/.rt_time.stddev
    325759 ± 11%     -14.6%     278092 ± 12%  numa-meminfo.node1.FilePages
      8962 ±  8%     -27.0%       6541 ± 10%  numa-meminfo.node1.Mapped
    154772 ± 22%     -58.0%      65021 ± 43%  numa-meminfo.node2.Active
    154386 ± 22%     -57.9%      65021 ± 43%  numa-meminfo.node2.Active(anon)
     83648 ± 44%     -57.2%      35824 ± 53%  numa-meminfo.node2.AnonHugePages
    622984 ±  9%     -17.2%     515783 ±  6%  numa-meminfo.node2.MemUsed
      1484 ± 74%     -66.9%     491.75 ± 17%  numa-meminfo.node2.PageTables
      1825 ± 11%     +15.6%       2110 ±  3%  slabinfo.UNIX.active_objs
      1825 ± 11%     +15.6%       2110 ±  3%  slabinfo.UNIX.num_objs
      3514 ± 16%     +25.4%       4408 ±  6%  slabinfo.eventpoll_pwq.active_objs
      3514 ± 16%     +25.4%       4408 ±  6%  slabinfo.eventpoll_pwq.num_objs
      3996 ± 11%     +14.8%       4590 ±  3%  slabinfo.sock_inode_cache.active_objs
      3996 ± 11%     +14.8%       4590 ±  3%  slabinfo.sock_inode_cache.num_objs
      1307 ±  7%     +18.6%       1549 ±  3%  slabinfo.task_group.active_objs
      1307 ±  7%     +18.6%       1549 ±  3%  slabinfo.task_group.num_objs
     92621 ±  2%      +5.3%      97488 ±  4%  softirqs.CPU0.TIMER
     29278 ±  5%      +8.9%      31873 ±  5%  softirqs.CPU104.RCU
     29770 ±  6%      +8.2%      32217 ±  4%  softirqs.CPU106.RCU
     29452 ±  5%      +8.1%      31837 ±  3%  softirqs.CPU111.RCU
     27058 ±  7%     +13.5%      30700 ±  8%  softirqs.CPU119.RCU
     27840 ±  4%     +17.4%      32687 ±  7%  softirqs.CPU150.RCU
     27467 ±  4%     +13.2%      31093 ±  4%  softirqs.CPU51.RCU
     27155 ±  4%     +14.0%      30949 ±  6%  softirqs.CPU52.RCU
     93817 ±  4%     +20.9%     113413 ± 23%  softirqs.CPU72.TIMER
     27250 ±  8%     +15.1%      31373 ±  7%  softirqs.CPU73.RCU
  1.16e+11            -1.4%  1.144e+11        perf-stat.i.branch-instructions
 7.639e+08            -1.4%   7.53e+08        perf-stat.i.branch-misses
      0.99            +2.1%       1.02        perf-stat.i.cpi
 1.751e+11            -1.4%  1.726e+11        perf-stat.i.dTLB-loads
 1.066e+11            -1.4%  1.051e+11        perf-stat.i.dTLB-stores
 5.721e+11            -1.4%  5.639e+11        perf-stat.i.instructions
    829.37            -1.1%     820.58        perf-stat.i.instructions-per-iTLB-miss
      1.01            -1.6%       0.99        perf-stat.i.ipc
    156668 ±  3%     +26.3%     197814 ±  7%  perf-stat.i.node-load-misses
      0.99            +1.5%       1.01        perf-stat.overall.cpi
    829.33            -1.1%     820.50        perf-stat.overall.instructions-per-iTLB-miss
      1.01            -1.5%       0.99        perf-stat.overall.ipc
 1.156e+11            -1.4%   1.14e+11        perf-stat.ps.branch-instructions
 7.613e+08            -1.4%  7.504e+08        perf-stat.ps.branch-misses
 1.745e+11            -1.4%   1.72e+11        perf-stat.ps.dTLB-loads
 1.062e+11            -1.4%  1.047e+11        perf-stat.ps.dTLB-stores
 5.701e+11            -1.4%  5.619e+11        perf-stat.ps.instructions
    156146 ±  3%     +26.3%     197163 ±  7%  perf-stat.ps.node-load-misses
 1.703e+14            -1.1%  1.685e+14        perf-stat.total.instructions
     16.68            -0.2       16.44        perf-profile.calltrace.cycles-pp.entry_SYSCALL_64.__sched_yield
      2.57            -0.2        2.35        perf-profile.calltrace.cycles-pp.update_min_vruntime.update_curr.pick_next_task_fair.__sched_text_start.schedule
      1.72            -0.1        1.67        perf-profile.calltrace.cycles-pp.yield_task_fair.do_sched_yield.__x64_sys_sched_yield.do_syscall_64.entry_SYSCALL_64_after_hwframe
      1.24            -0.0        1.21        perf-profile.calltrace.cycles-pp.__list_del_entry_valid.pick_next_task_fair.__sched_text_start.schedule.__x64_sys_sched_yield
      1.15            +0.1        1.21        perf-profile.calltrace.cycles-pp.__list_add_valid.pick_next_task_fair.__sched_text_start.schedule.__x64_sys_sched_yield
      7.43            +0.3        7.73        perf-profile.calltrace.cycles-pp.__calc_delta.update_curr.pick_next_task_fair.__sched_text_start.schedule
     77.71            +0.3       78.05        perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.__sched_yield
     74.90            +0.4       75.27        perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.__sched_yield
     71.74            +0.4       72.14        perf-profile.calltrace.cycles-pp.__x64_sys_sched_yield.do_syscall_64.entry_SYSCALL_64_after_hwframe.__sched_yield
     62.26            +0.5       62.78        perf-profile.calltrace.cycles-pp.schedule.__x64_sys_sched_yield.do_syscall_64.entry_SYSCALL_64_after_hwframe.__sched_yield
     60.38            +0.6       60.94        perf-profile.calltrace.cycles-pp.__sched_text_start.schedule.__x64_sys_sched_yield.do_syscall_64.entry_SYSCALL_64_after_hwframe
      2.34            +0.6        2.91        perf-profile.calltrace.cycles-pp.clear_buddies.pick_next_entity.pick_next_task_fair.__sched_text_start.schedule
     38.47            +0.7       39.15        perf-profile.calltrace.cycles-pp.pick_next_task_fair.__sched_text_start.schedule.__x64_sys_sched_yield.do_syscall_64
     21.48            +0.8       22.26        perf-profile.calltrace.cycles-pp.update_curr.pick_next_task_fair.__sched_text_start.schedule.__x64_sys_sched_yield
      2.60            -0.2        2.39        perf-profile.children.cycles-pp.update_min_vruntime
     11.18            -0.2       11.01        perf-profile.children.cycles-pp.entry_SYSCALL_64
      8.66            -0.2        8.50        perf-profile.children.cycles-pp.syscall_return_via_sysret
      0.92 ±  2%      -0.1        0.84        perf-profile.children.cycles-pp.check_cfs_rq_runtime
      1.75            -0.1        1.70        perf-profile.children.cycles-pp.yield_task_fair
      1.29            -0.0        1.27        perf-profile.children.cycles-pp.__list_del_entry_valid
      0.08 ± 10%      +0.0        0.12 ± 16%  perf-profile.children.cycles-pp.ktime_get_update_offsets_now
      1.20            +0.1        1.25        perf-profile.children.cycles-pp.__list_add_valid
      5.48            +0.3        5.75        perf-profile.children.cycles-pp.pick_next_entity
      1.73            +0.3        2.02        perf-profile.children.cycles-pp.clear_buddies
      8.05            +0.3        8.35        perf-profile.children.cycles-pp.__calc_delta
     77.85            +0.3       78.19        perf-profile.children.cycles-pp.entry_SYSCALL_64_after_hwframe
     75.11            +0.4       75.49        perf-profile.children.cycles-pp.do_syscall_64
     71.97            +0.4       72.37        perf-profile.children.cycles-pp.__x64_sys_sched_yield
     62.29            +0.5       62.80        perf-profile.children.cycles-pp.schedule
     60.91            +0.6       61.47        perf-profile.children.cycles-pp.__sched_text_start
     39.84            +0.7       40.53        perf-profile.children.cycles-pp.pick_next_task_fair
     21.79            +0.7       22.54        perf-profile.children.cycles-pp.update_curr
     10.29            -0.2       10.04        perf-profile.self.cycles-pp.pick_next_task_fair
      2.36            -0.2        2.17        perf-profile.self.cycles-pp.update_min_vruntime
      8.59            -0.2        8.42        perf-profile.self.cycles-pp.syscall_return_via_sysret
      7.21            -0.1        7.10        perf-profile.self.cycles-pp.__sched_yield
      5.61            -0.1        5.53        perf-profile.self.cycles-pp.entry_SYSCALL_64
      1.63            -0.1        1.57        perf-profile.self.cycles-pp.yield_task_fair
      0.50 ±  2%      -0.0        0.46        perf-profile.self.cycles-pp.check_cfs_rq_runtime
      2.72            -0.0        2.68        perf-profile.self.cycles-pp.entry_SYSCALL_64_after_hwframe
      1.38            -0.0        1.34        perf-profile.self.cycles-pp.schedule
      1.43            -0.0        1.40        perf-profile.self.cycles-pp.__x64_sys_sched_yield
      0.08 ± 10%      +0.0        0.12 ± 15%  perf-profile.self.cycles-pp.ktime_get_update_offsets_now
      3.88            +0.2        4.09        perf-profile.self.cycles-pp.pick_next_entity
      7.77            +0.3        8.09        perf-profile.self.cycles-pp.__calc_delta
      8.41            +0.7        9.06 ±  3%  perf-profile.self.cycles-pp.update_curr


                                                                                
                             will-it-scale.per_thread_ops                       
                                                                                
  3.5e+06 +-+---------------------------------------------------------------+   
          |.+.+.+.+.+.+.O.OO.O.O.O.O.O.O.O.O.O.O.O.+.+.+.+.++.+.+.+.+.+.+.+.|   
    3e+06 O-O O O   O O                                                     |   
          |                                                                 |   
  2.5e+06 +-+                                                               |   
          |                                                                 |   
    2e+06 +-+                                                               |   
          |                                                                 |   
  1.5e+06 +-+                                                               |   
          |                                                                 |   
    1e+06 +-+                                                               |   
          |                                                                 |   
   500000 +-+                                                               |   
          |                                                                 |   
        0 +-+-----O---------------------------------------------------------+   
                                                                                
                                                                                
[*] bisect-good sample
[O] bisect-bad  sample

***************************************************************************************************
lkp-skl-4sp1: 192 threads Skylake-SP with 256G memory
=========================================================================================
compiler/cpufreq_governor/kconfig/mode/nr_task/rootfs/tbox_group/test/testcase:
  gcc-7/performance/x86_64-rhel-7.6/thread/100%/debian-x86_64-2018-04-03.cgz/lkp-skl-4sp1/sched_yield/will-it-scale

commit: 
  f7798711ad ("Merge branch 'x86/cpu' into x86/urgent")
  ffc8599aa9 ("x86/gart: Exclude GART aperture from kcore")

f7798711adeebde3 ffc8599aa9763f39f6736a79da4 
---------------- --------------------------- 
       fail:runs  %reproduction    fail:runs
           |             |             |    
          1:4          -25%            :4     dmesg.WARNING:at#for_ip_interrupt_entry/0x
         %stddev     %change         %stddev
             \          |                \  
   1562038            -2.0%    1530890        will-it-scale.per_thread_ops
      9111            +1.9%       9281        will-it-scale.time.minor_page_faults
 2.999e+08            -2.0%  2.939e+08        will-it-scale.workload
    510.25 ± 18%     -28.1%     366.75 ± 16%  kthread_noise.total_time
    261316 ±  5%      +3.5%     270361 ±  5%  numa-meminfo.node0.Unevictable
     27269 ± 49%     +61.1%      43927 ± 16%  numa-numastat.node3.other_node
      2187 ±  6%     +92.1%       4202 ± 67%  proc-vmstat.numa_hint_faults
      1909           +15.5%       2205 ± 15%  proc-vmstat.numa_hint_faults_local
    714.50 ±  7%     -16.4%     597.50 ±  8%  slabinfo.skbuff_fclone_cache.active_objs
    714.50 ±  7%     -16.4%     597.50 ±  8%  slabinfo.skbuff_fclone_cache.num_objs
     65329 ±  5%      +3.5%      67590 ±  5%  numa-vmstat.node0.nr_unevictable
     65329 ±  5%      +3.5%      67590 ±  5%  numa-vmstat.node0.nr_zone_unevictable
    111616 ± 12%     +14.9%     128279 ±  5%  numa-vmstat.node3.numa_other
      0.73           +23.8%       0.90 ± 17%  irq_exception_noise.__do_page_fault.50th
      0.74           +54.2%       1.14 ± 31%  irq_exception_noise.__do_page_fault.60th
      0.75          +106.8%       1.56 ± 34%  irq_exception_noise.__do_page_fault.70th
      0.77 ±  2%    +149.6%       1.92 ± 34%  irq_exception_noise.__do_page_fault.80th
    297.75 ± 12%     -43.6%     168.00 ±  3%  interrupts.33:PCI-MSI.26738690-edge.eth0-TxRx-1
     19.75 ± 57%    +738.0%     165.50 ±130%  interrupts.CPU100.RES:Rescheduling_interrupts
    162.00 ± 87%    +451.2%     893.00 ± 65%  interrupts.CPU12.RES:Rescheduling_interrupts
    112.50 ± 46%     -80.4%      22.00 ± 33%  interrupts.CPU120.RES:Rescheduling_interrupts
     55.00 ± 65%     -55.0%      24.75 ± 21%  interrupts.CPU122.RES:Rescheduling_interrupts
    233.00 ±116%     -90.6%      22.00 ± 58%  interrupts.CPU130.RES:Rescheduling_interrupts
    325.00 ±150%     -94.7%      17.25 ± 31%  interrupts.CPU142.RES:Rescheduling_interrupts
     11.75 ± 36%    +463.8%      66.25 ±123%  interrupts.CPU146.RES:Rescheduling_interrupts
     11.25 ± 52%    +364.4%      52.25 ± 73%  interrupts.CPU153.RES:Rescheduling_interrupts
    507.50 ±112%     -93.6%      32.50 ± 68%  interrupts.CPU163.RES:Rescheduling_interrupts
     82.25 ± 78%   +2841.6%       2419 ± 77%  interrupts.CPU17.RES:Rescheduling_interrupts
     76.00 ± 66%   +1346.1%       1099 ±120%  interrupts.CPU18.RES:Rescheduling_interrupts
    168.50 ±132%    +534.1%       1068 ±113%  interrupts.CPU20.RES:Rescheduling_interrupts
      2015 ±100%     -93.1%     139.75 ± 87%  interrupts.CPU24.RES:Rescheduling_interrupts
    242.50 ±159%    +443.4%       1317 ± 69%  interrupts.CPU3.RES:Rescheduling_interrupts
      4543 ± 56%     -85.4%     661.25 ±157%  interrupts.CPU34.RES:Rescheduling_interrupts
      3070 ± 52%     -93.0%     215.75 ±102%  interrupts.CPU40.RES:Rescheduling_interrupts
      1129 ± 85%     -92.5%      84.75 ± 67%  interrupts.CPU44.RES:Rescheduling_interrupts
      2609 ± 59%     -91.6%     220.25 ±122%  interrupts.CPU45.RES:Rescheduling_interrupts
    297.75 ± 12%     -43.6%     168.00 ±  3%  interrupts.CPU9.33:PCI-MSI.26738690-edge.eth0-TxRx-1
     26.33            -0.8       25.52        perf-profile.calltrace.cycles-pp.entry_SYSCALL_64.__sched_yield
     19.40            -0.3       19.06        perf-profile.calltrace.cycles-pp.syscall_return_via_sysret.__sched_yield
      6.55            -0.3        6.29        perf-profile.calltrace.cycles-pp.do_sched_yield.__x64_sys_sched_yield.do_syscall_64.entry_SYSCALL_64_after_hwframe.__sched_yield
      2.21            -0.1        2.10        perf-profile.calltrace.cycles-pp.yield_task_fair.do_sched_yield.__x64_sys_sched_yield.do_syscall_64.entry_SYSCALL_64_after_hwframe
      0.77            -0.0        0.74        perf-profile.calltrace.cycles-pp.__list_add_valid.pick_next_task_fair.__sched_text_start.schedule.__x64_sys_sched_yield
      0.58            +0.0        0.62        perf-profile.calltrace.cycles-pp.__x86_indirect_thunk_rax.__sched_yield
      3.55            +0.1        3.61        perf-profile.calltrace.cycles-pp.sched_clock_cpu.update_rq_clock.__sched_text_start.schedule.__x64_sys_sched_yield
      1.86            +0.1        1.93        perf-profile.calltrace.cycles-pp.testcase
      1.46            +0.1        1.52        perf-profile.calltrace.cycles-pp.update_min_vruntime.update_curr.pick_next_task_fair.__sched_text_start.schedule
      3.20            +0.1        3.27        perf-profile.calltrace.cycles-pp.native_sched_clock.sched_clock.sched_clock_cpu.update_rq_clock.__sched_text_start
      3.15            +0.1        3.23        perf-profile.calltrace.cycles-pp.sched_clock.sched_clock_cpu.update_rq_clock.__sched_text_start.schedule
      4.35            +0.1        4.45        perf-profile.calltrace.cycles-pp.update_rq_clock.__sched_text_start.schedule.__x64_sys_sched_yield.do_syscall_64
     51.64            +1.1       52.77        perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.__sched_yield
     47.58            +1.1       48.72        perf-profile.calltrace.cycles-pp.__x64_sys_sched_yield.do_syscall_64.entry_SYSCALL_64_after_hwframe.__sched_yield
     49.88            +1.1       51.03        perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.__sched_yield
      4.25            +1.2        5.50        perf-profile.calltrace.cycles-pp.__calc_delta.update_curr.pick_next_task_fair.__sched_text_start.schedule
     12.46            +1.4       13.89        perf-profile.calltrace.cycles-pp.update_curr.pick_next_task_fair.__sched_text_start.schedule.__x64_sys_sched_yield
     24.97            +1.5       26.45        perf-profile.calltrace.cycles-pp.pick_next_task_fair.__sched_text_start.schedule.__x64_sys_sched_yield.do_syscall_64
     38.00            +1.5       39.49        perf-profile.calltrace.cycles-pp.__sched_text_start.schedule.__x64_sys_sched_yield.do_syscall_64.entry_SYSCALL_64_after_hwframe
     39.20            +1.5       40.70        perf-profile.calltrace.cycles-pp.schedule.__x64_sys_sched_yield.do_syscall_64.entry_SYSCALL_64_after_hwframe.__sched_yield
 5.318e+10            -1.9%  5.217e+10        perf-stat.i.branch-instructions
 9.172e+08            -1.9%  8.999e+08        perf-stat.i.branch-misses
      1.51            +2.0%       1.54        perf-stat.i.cpi
      0.00 ± 10%      -0.0        0.00 ±  5%  perf-stat.i.dTLB-load-miss-rate%
    389507 ± 10%     -19.4%     314009 ±  5%  perf-stat.i.dTLB-load-misses
 8.384e+10            -1.9%  8.221e+10        perf-stat.i.dTLB-loads
 5.157e+10            -1.9%  5.058e+10        perf-stat.i.dTLB-stores
 3.031e+08           +65.0%  5.001e+08        perf-stat.i.iTLB-load-misses
   8332137           +36.7%   11386301        perf-stat.i.iTLB-loads
 2.773e+11            -1.9%   2.72e+11        perf-stat.i.instructions
    918.98           -39.8%     553.30        perf-stat.i.instructions-per-iTLB-miss
      0.66            -1.9%       0.65        perf-stat.i.ipc
      0.03 ±  5%      +8.9%       0.04 ±  4%  perf-stat.overall.MPKI
      1.51            +2.0%       1.54        perf-stat.overall.cpi
      0.00 ± 10%      -0.0        0.00 ±  5%  perf-stat.overall.dTLB-load-miss-rate%
    914.60           -40.6%     543.59        perf-stat.overall.instructions-per-iTLB-miss
      0.66            -1.9%       0.65        perf-stat.overall.ipc
 5.298e+10            -2.0%  5.195e+10        perf-stat.ps.branch-instructions
  9.14e+08            -1.9%  8.963e+08        perf-stat.ps.branch-misses
    388156 ± 10%     -19.4%     312919 ±  5%  perf-stat.ps.dTLB-load-misses
 8.354e+10            -1.9%  8.192e+10        perf-stat.ps.dTLB-loads
 5.139e+10            -1.9%   5.04e+10        perf-stat.ps.dTLB-stores
  3.02e+08           +65.0%  4.983e+08        perf-stat.ps.iTLB-load-misses
   8300078           +36.7%   11347114        perf-stat.ps.iTLB-loads
 2.762e+11            -2.0%  2.708e+11        perf-stat.ps.instructions
 8.308e+13            -2.0%  8.144e+13        perf-stat.total.instructions





Disclaimer:
Results have been estimated based on internal Intel analysis and are provided
for informational purposes only. Any difference in system hardware or software
design or configuration may affect actual performance.


Thanks,
Rong Chen


View attachment "config-5.1.0-rc1-00013-gffc8599aa9763" of type "text/plain" (188904 bytes)

View attachment "job-script" of type "text/plain" (7390 bytes)

View attachment "job.yaml" of type "text/plain" (5029 bytes)

View attachment "reproduce" of type "text/plain" (315 bytes)

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ