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: <20170525025540.GU1795@yexl-desktop>
Date:   Thu, 25 May 2017 10:55:40 +0800
From:   kernel test robot <xiaolong.ye@...el.com>
To:     Peter Zijlstra <peterz@...radead.org>
Cc:     Ingo Molnar <mingo@...nel.org>,
        Lauro Ramos Venancio <lvenanci@...hat.com>,
        Linus Torvalds <torvalds@...ux-foundation.org>,
        Mike Galbraith <efault@....de>, Rik van Riel <riel@...hat.com>,
        Thomas Gleixner <tglx@...utronix.de>,
        LKML <linux-kernel@...r.kernel.org>,
        "H. Peter Anvin" <hpa@...or.com>, tipbuild@...or.com, lkp@...org
Subject: [lkp-robot] [sched/fair, cpumask]  c743f0a5c5: hackbench.throughput
 -24.9% regression


Greeting,

FYI, we noticed a -24.9% regression of hackbench.throughput due to commit:


commit: c743f0a5c50f2fcbc628526279cfa24f3dabe182 ("sched/fair, cpumask: Export for_each_cpu_wrap()")
https://git.kernel.org/cgit/linux/kernel/git/tip/tip.git sched/core

in testcase: hackbench
on test machine: 88 threads Intel(R) Xeon(R) CPU E5-2699 v4 @ 2.20GHz with 64G memory
with following parameters:

	nr_threads: 50%
	mode: process
	ipc: socket
	cpufreq_governor: performance

test-description: Hackbench is both a benchmark and a stress test for the Linux kernel scheduler.
test-url: https://github.com/linux-test-project/ltp/blob/master/testcases/kernel/sched/cfs-scheduler/hackbench.c



Details are as below:
-------------------------------------------------------------------------------------------------->


To reproduce:

        git clone https://github.com/01org/lkp-tests.git
        cd lkp-tests
        bin/lkp install job.yaml  # job file is attached in this email
        bin/lkp run     job.yaml

testcase/path_params/tbox_group/run: hackbench/50%-process-socket-performance/lkp-bdw-ep3

8c0334697dc37eb3  c743f0a5c50f2fcbc628526279  
----------------  --------------------------  
         %stddev      change         %stddev
             \          |                \  
    179259             -25%     134695 ±  4%  hackbench.throughput
    825171             -23%     632917 ±  4%  hackbench.time.minor_page_faults
      1783 ±  3%        15%       2057        hackbench.time.user_time
 1.596e+09 ±  5%        43%  2.286e+09 ±  6%  hackbench.time.involuntary_context_switches
 4.157e+09 ±  3%        16%  4.841e+09        hackbench.time.voluntary_context_switches
   9066379              24%   11264639 ±  3%  vmstat.system.cs
      1.12 ±  4%        59%       1.78 ± 27%  perf-stat.cache-miss-rate%
  1.91e+10 ±  4%        23%  2.354e+10 ±  4%  perf-stat.iTLB-load-misses
     96.25                       97.43        perf-stat.iTLB-load-miss-rate%
     98.27                       99.31        perf-stat.node-load-miss-rate%
     72.92             -20%      58.24 ±  6%  perf-stat.node-store-miss-rate%
   2037986             -12%    1797391        perf-stat.minor-faults
   2038044             -12%    1797391        perf-stat.page-faults
      0.44              30%       0.57        perf-stat.branch-miss-rate%
 1.707e+09 ± 11%       116%  3.693e+09 ± 28%  perf-stat.node-stores
 7.149e+10              27%  9.098e+10 ±  3%  perf-stat.branch-misses
      0.17               9%       0.18        perf-stat.dTLB-load-miss-rate%
 5.778e+09 ±  4%        24%  7.148e+09        perf-stat.context-switches
 4.118e+10 ±  3%         6%  4.366e+10        perf-stat.dTLB-load-misses
 7.424e+08             -16%  6.203e+08 ±  5%  perf-stat.iTLB-loads
 2.933e+08 ± 22%       188%  8.443e+08 ± 18%  perf-stat.cpu-migrations
      4480 ±  3%       -19%       3630 ±  4%  perf-stat.instructions-per-iTLB-miss
 1.843e+10 ±  6%        52%  2.797e+10 ± 25%  perf-stat.cache-misses
  8.35e+09 ±  8%        78%   1.49e+10 ± 27%  perf-stat.node-load-misses
 1.467e+08 ±  4%       -34%   97529419 ±  9%  perf-stat.node-loads



                          hackbench.time.minor_page_faults

  950000 ++-----------------------------------------------------------------+
         |    .*.      *.           .*.     *.       .*.                    |
  900000 *+*.*   *     : *     *   *   *.*.*  *.*.*.*   *                   |
         |        +   :   :   + : :                      :                  |
  850000 ++        *. :   : .*  : :                      :                  |
         |           *     *     *                        *. .*.*.*.*.*.*.*.*
  800000 ++                                                 *               |
         |           O       O                                              |
  750000 ++            O O O   O     O O O OO                               |
         |         O                                                        |
  700000 ++    O O               O                                          |
         O O O                     O                O                       |
  650000 ++                                                                 |
         |                                            O O                   |
  600000 ++-----------------------------------O-O-O-------------------------+


                                perf-stat.branch-misses

    1e+11 ++-------------------O-----O--------------------------------------+
          |           O O    O           O   O                              |
  9.5e+10 ++    O         OO     O O                 O                      |
          O O     O O                  O   O   O                            |
    9e+10 ++                                     O O     O                  |
          |   O                                        O                    |
  8.5e+10 ++                                                                |
          |                                                                 |
    8e+10 ++                                                                |
          |                                                                 |
  7.5e+10 *+    *       *.*        *.*.*     *   *     *                    |
          |+   + +     +   :  .*. +     +   + + + +   + +    .*.     .*. .*.*
    7e+10 ++*.*   *.*.*    *.*   *       *.*   *   *.*   *.**   *.*.*   *   |
          |                                                                 |
  6.5e+10 ++----------------------------------------------------------------+


                            perf-stat.branch-miss-rate_

  0.62 ++-------------------------------------------------------------------+
   0.6 ++                        O O                                        |
       |                 O O O O                                            |
  0.58 ++            O O             O O  O O       O                       |
  0.56 ++O O O O O O                          O O O   O O                   |
       O                                                                    |
  0.54 ++                                                                   |
  0.52 ++                                                                   |
   0.5 ++                                                                   |
       |                                                                    |
  0.48 ++                                                                   |
  0.46 ++                                                                   |
       *.   .*.             .*.*.   .*.    .*. .*.   .*.                    |
  0.44 ++*.*   *.*.*.*.*.*.*     *.*   *..*   *   *.*   *.*.*.*.*.*.*.*.*.*.*
  0.42 ++-------------------------------------------------------------------+

  [*] 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,
Xiaolong

View attachment "config-4.12.0-rc1-00015-gc743f0a" of type "text/plain" (159405 bytes)

View attachment "job-script" of type "text/plain" (6724 bytes)

View attachment "job.yaml" of type "text/plain" (4327 bytes)

View attachment "reproduce" of type "text/plain" (455 bytes)

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ