linux-kernel - [lkp-robot] [sched/core] 8d18fda5c8: unixbench.score 4.4% improvement

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 for Android: free password hash cracker in your pocket
[<prev] [next>] [day] [month] [year] [list]
Message-ID: <20170807024637.GA25554@yexl-desktop>
Date:   Mon, 7 Aug 2017 10:46:37 +0800
From:   kernel test robot <xiaolong.ye@...el.com>
To:     Will Deacon <will.deacon@....com>
Cc:     LKML <linux-kernel@...r.kernel.org>,
        Will Deacon <will.deacon@....com>, lkp@...org
Subject: [lkp-robot] [sched/core]  8d18fda5c8:  unixbench.score 4.4%
 improvement


Greeting,

FYI, we noticed a 4.4% improvement of unixbench.score due to commit:


commit: 8d18fda5c8a2d22d60cdb8cc5ffdea3f9aa465c4 ("sched/core: Avoid rebalancing on exec() in shadow of fork()")
https://git.kernel.org/cgit/linux/kernel/git/will/linux.git aarch64/devel

in testcase: unixbench
on test machine: 4 threads Intel(R) Core(TM) i3-3220 CPU @ 3.30GHz with 4G memory
with following parameters:

	runtime: 300s
	nr_task: 100%
	test: shell1
	cpufreq_governor: performance

test-description: UnixBench is the original BYTE UNIX benchmark suite aims to test performance of Unix-like system.
test-url: https://github.com/kdlucas/byte-unixbench



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: unixbench/300s-100%-shell1-performance/lkp-ivb-d04

1c7cf53658f0fa16  8d18fda5c8a2d22d60cdb8cc5f  
----------------  --------------------------  
         %stddev      change         %stddev
             \          |                \  
      5948               4%       6209        unixbench.score
       307 ±  5%         9%        334        unixbench.time.percent_of_cpu_this_job_got
       694               7%        744        unixbench.time.system_time
      1305               6%       1387        unixbench.time.user_time
   6925375                     7103453        unixbench.time.voluntary_context_switches
 2.149e+08               5%  2.249e+08        unixbench.time.minor_page_faults
   6639557             -15%    5644060        unixbench.time.involuntary_context_switches
    149028             -83%      25130        interrupts.CAL:Function_call_interrupts
     38816 ±  5%       -11%      34499        vmstat.system.cs
  6.881e+08 ±  9%        27%  8.729e+08 ± 10%  perf-stat.dTLB-store-misses
 2.152e+09               8%  2.329e+09        perf-stat.iTLB-load-misses
 6.761e+12               6%  7.151e+12        perf-stat.cpu-cycles
     72.02                       73.98        perf-stat.iTLB-load-miss-rate%
  1.36e+12               4%  1.413e+12        perf-stat.dTLB-loads
 2.729e+10               4%  2.832e+10        perf-stat.branch-misses
 8.979e+11               4%  9.319e+11        perf-stat.dTLB-stores
 2.107e+08               5%  2.205e+08        perf-stat.minor-faults
 2.107e+08               5%  2.205e+08        perf-stat.page-faults
 8.363e+08                    8.19e+08        perf-stat.iTLB-loads
  1.05e+12               4%  1.091e+12        perf-stat.branch-instructions
 5.287e+12               4%  5.492e+12        perf-stat.instructions
      0.78                        0.77        perf-stat.ipc
      1.28                        1.30        perf-stat.cpi
 7.254e+10                   7.468e+10        perf-stat.cache-references
      2456              -4%       2358        perf-stat.instructions-per-iTLB-miss
  25275768             -13%   22037547        perf-stat.context-switches
   4463516             -24%    3396779        perf-stat.cpu-migrations



                                  unixbench.score

  6250 ++-------------------------------------------------------------------+
       |                       O                       O O  O       O  O    |
  6200 O+      O  O         O     O       O    O    O          O O       O  O
       |  O O          O  O         O       O    O                          |
  6150 ++           O                  O                                    |
       |                                                                    |
  6100 ++                                                                   |
       |                                                                    |
  6050 ++                                                                   |
       |                                                                    |
  6000 ++   *                                                               |
       |   : :                                                              |
  5950 ++  :  :   *.*..*..*.  .*..*.*..*..          *..*.*..                |
       *..*   : ..          *.            *.  .*. ..        *               |
  5900 ++------*----------------------------*----*--------------------------+



  [*] 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-4.13.0-rc2-00007-g8d18fda" of type "text/plain" (160961 bytes)

View attachment "job-script" of type "text/plain" (6753 bytes)

View attachment "job.yaml" of type "text/plain" (4388 bytes)

View attachment "reproduce" of type "text/plain" (278 bytes)