lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [day] [month] [year] [list]
Date:	Mon, 15 Dec 2014 11:26:55 +0800
From:	Huang Ying <ying.huang@...el.com>
To:	Jaegeuk Kim <jaegeuk@...nel.org>
Cc:	LKML <linux-kernel@...r.kernel.org>, LKP ML <lkp@...org>
Subject: [LKP] [f2fs] 8b26ef98da3: + 27.4% iostat.sda.wrqm/s

FYI, we noticed the below changes on

commit 8b26ef98da3387eb57a8a5c1747c6e628948ee0c ("f2fs: use rw_semaphore for nat entry lock")

testbox/testcase/testparams: lkp-ne04/fsmark/performance-1x-32t-1HDD-f2fs-8K-400M-fsyncBeforeClose-16d-256fpd

4634d71ed190c99e  8b26ef98da3387eb57a8a5c174  
----------------  --------------------------  
         %stddev     %change         %stddev
             \          |                \  
       420 ±  0%     +12.0%        470 ±  0%  fsmark.files_per_sec
      7.37 ± 22%     -84.0%       1.18 ± 26%  turbostat.%pc6
      2122 ±  2%    +929.0%      21838 ±  1%  proc-vmstat.pgactivate
     41341 ± 34%    +226.9%     135151 ± 40%  sched_debug.cpu#4.sched_count
      4093 ± 29%    +266.1%      14988 ± 21%  sched_debug.cpu#12.ttwu_count
  20670219 ± 24%    +243.7%   71049994 ± 11%  cpuidle.C1-NHM.time
      4279 ± 25%    +237.2%      14431 ± 19%  sched_debug.cpu#14.ttwu_count
      3995 ± 19%    +237.7%      13492 ± 22%  sched_debug.cpu#11.ttwu_count
      4092 ± 25%    +230.0%      13503 ± 19%  sched_debug.cpu#15.ttwu_count
      7241 ± 14%    +218.7%      23080 ± 18%  sched_debug.cpu#3.ttwu_count
      4065 ± 28%    +251.5%      14291 ± 24%  sched_debug.cpu#13.ttwu_count
        23 ± 48%    +201.1%         69 ± 12%  cpuidle.POLL.usage
     12604 ± 11%    +161.1%      32904 ± 28%  sched_debug.cpu#11.nr_switches
      5441 ± 15%    +164.0%      14365 ± 27%  sched_debug.cpu#11.sched_goidle
     12902 ±  9%    +163.0%      33936 ± 33%  sched_debug.cpu#13.nr_switches
      8230 ± 13%    +182.2%      23230 ± 20%  sched_debug.cpu#1.ttwu_count
     13010 ±  9%    +153.2%      32947 ± 28%  sched_debug.cpu#15.nr_switches
      5571 ± 11%    +160.5%      14511 ± 30%  sched_debug.cpu#13.sched_goidle
     13596 ± 13%    +172.7%      37082 ± 38%  sched_debug.cpu#15.sched_count
      7563 ± 16%    +200.9%      22762 ± 22%  sched_debug.cpu#7.ttwu_count
      5598 ± 12%    +156.2%      14342 ± 26%  sched_debug.cpu#15.sched_goidle
     16069 ± 23%    +117.8%      34992 ± 25%  sched_debug.cpu#14.nr_switches
     14194 ±  8%    +152.8%      35879 ± 26%  sched_debug.cpu#12.nr_switches
     13397 ± 11%    +158.2%      34598 ± 22%  sched_debug.cpu#11.sched_count
     14596 ±  9%    +148.3%      36240 ± 25%  sched_debug.cpu#12.sched_count
     13647 ± 10%    +150.2%      34139 ± 32%  sched_debug.cpu#13.sched_count
      6705 ± 20%    +127.1%      15225 ± 23%  sched_debug.cpu#14.sched_goidle
      6177 ± 10%    +151.7%      15546 ± 24%  sched_debug.cpu#12.sched_goidle
     16275 ± 23%    +139.7%      39015 ± 17%  sched_debug.cpu#14.sched_count
      6218 ± 15%    +209.6%      19252 ± 45%  sched_debug.cpu#10.sched_goidle
     21820 ±  6%    +123.4%      48742 ± 25%  sched_debug.cpu#7.nr_switches
     22931 ± 10%    +159.5%      59497 ± 44%  sched_debug.cpu#5.nr_switches
      9865 ±  8%    +120.0%      21709 ± 24%  sched_debug.cpu#7.sched_goidle
     10505 ± 12%    +141.8%      25405 ± 37%  sched_debug.cpu#5.sched_goidle
     12980 ±  6%    +107.7%      26956 ± 16%  sched_debug.cpu#4.ttwu_count
     24231 ± 18%    +103.6%      49334 ± 24%  sched_debug.cpu#3.nr_switches
     11147 ± 14%     +99.2%      22210 ± 22%  sched_debug.cpu#1.sched_goidle
     11092 ± 21%     +99.0%      22076 ± 23%  sched_debug.cpu#3.sched_goidle
     29443 ±  8%     +89.3%      55744 ± 20%  sched_debug.cpu#4.nr_switches
     32087 ±  7%     +81.3%      58169 ± 18%  sched_debug.cpu#2.nr_switches
     12984 ± 17%    +111.4%      27446 ± 12%  sched_debug.cpu#2.ttwu_count
     26458 ± 18%     +89.7%      50191 ± 24%  sched_debug.cpu#1.nr_switches
     14505 ±  8%     +98.6%      28807 ± 29%  sched_debug.cpu#0.sched_goidle
     13628 ±  8%     +81.1%      24686 ± 17%  sched_debug.cpu#2.sched_goidle
     13700 ±  9%     +82.6%      25012 ± 18%  sched_debug.cpu#4.sched_goidle
     33822 ±  9%    +102.3%      68417 ± 35%  sched_debug.cpu#0.nr_switches
     18438 ± 28%    +160.1%      47957 ± 23%  cpuidle.C1-NHM.usage
      6.50 ± 10%     +73.2%      11.25 ±  7%  turbostat.%c1
        14 ± 13%     +52.5%         22 ± 12%  sched_debug.cfs_rq[13]:/.tg_runnable_contrib
    135553 ±  6%     +73.5%     235188 ±  6%  cpuidle.C3-NHM.usage
       723 ± 13%     +48.3%       1072 ± 10%  sched_debug.cfs_rq[13]:/.avg->runnable_avg_sum
     28.84 ±  9%     +52.2%      43.89 ±  5%  turbostat.%c3
     63.48 ±  3%     -31.8%      43.29 ±  5%  turbostat.%c6
     30737 ±  0%     -31.0%      21223 ±  1%  softirqs.BLOCK
      2329 ±  5%     +31.1%       3052 ± 11%  sched_debug.cfs_rq[14]:/.min_vruntime
 3.494e+08 ± 12%     +48.6%  5.192e+08 ±  5%  cpuidle.C3-NHM.time
 1.545e+09 ±  2%     -27.1%  1.126e+09 ±  2%  cpuidle.C6-NHM.time
  26451473 ±  5%     -28.7%   18850454 ± 17%  cpuidle.C1E-NHM.time
    304184 ±  6%     +36.3%     414743 ±  6%  cpuidle.C6-NHM.usage
       362 ±  2%     +28.7%        466 ±  6%  sched_debug.cfs_rq[0]:/.tg->runnable_avg
       363 ±  2%     +28.6%        467 ±  6%  sched_debug.cfs_rq[1]:/.tg->runnable_avg
       364 ±  1%     +28.4%        467 ±  6%  sched_debug.cfs_rq[2]:/.tg->runnable_avg
       367 ±  1%     +28.0%        470 ±  6%  sched_debug.cfs_rq[3]:/.tg->runnable_avg
       369 ±  1%     +27.9%        472 ±  5%  sched_debug.cfs_rq[4]:/.tg->runnable_avg
    977486 ±  1%     -21.6%     766721 ± 11%  sched_debug.cpu#13.avg_idle
       372 ±  1%     +27.2%        473 ±  6%  sched_debug.cfs_rq[5]:/.tg->runnable_avg
       373 ±  1%     +27.6%        476 ±  6%  sched_debug.cfs_rq[6]:/.tg->runnable_avg
       379 ±  1%     +27.2%        482 ±  6%  sched_debug.cfs_rq[8]:/.tg->runnable_avg
       376 ±  1%     +27.5%        479 ±  5%  sched_debug.cfs_rq[7]:/.tg->runnable_avg
       381 ±  1%     +26.8%        484 ±  6%  sched_debug.cfs_rq[9]:/.tg->runnable_avg
     41363 ±  5%     +59.4%      65923 ± 48%  sched_debug.cpu#0.ttwu_count
       384 ±  1%     +23.1%        473 ±  8%  sched_debug.cfs_rq[10]:/.tg->runnable_avg
    986988 ±  0%     -19.5%     794664 ±  4%  sched_debug.cpu#11.avg_idle
       386 ±  1%     +22.8%        474 ±  8%  sched_debug.cfs_rq[11]:/.tg->runnable_avg
       389 ±  2%     +22.0%        475 ±  8%  sched_debug.cfs_rq[13]:/.tg->runnable_avg
       392 ±  2%     +21.2%        476 ±  8%  sched_debug.cfs_rq[14]:/.tg->runnable_avg
       388 ±  1%     +22.1%        474 ±  8%  sched_debug.cfs_rq[12]:/.tg->runnable_avg
       396 ±  2%     +20.8%        478 ±  7%  sched_debug.cfs_rq[15]:/.tg->runnable_avg
    940409 ±  1%     -12.3%     824690 ±  4%  sched_debug.cpu#0.avg_idle
    927692 ±  3%     -12.9%     807567 ±  5%  sched_debug.cpu#2.avg_idle
      3216 ±  5%     -10.8%       2870 ±  3%  proc-vmstat.nr_alloc_batch
    979736 ±  0%     -13.5%     847782 ±  4%  sched_debug.cpu#12.avg_idle
    245057 ±  6%     -12.1%     215473 ± 11%  numa-vmstat.node1.numa_local
      1620 ±  4%     -11.5%       1435 ±  8%  numa-vmstat.node0.nr_alloc_batch
    894470 ±  3%     -12.4%     783635 ±  7%  sched_debug.cpu#7.avg_idle
    965398 ±  2%     -11.1%     858414 ±  6%  sched_debug.cpu#14.avg_idle
    167233 ±  0%    +239.7%     568014 ±  0%  time.voluntary_context_switches
      5760 ±  0%    +115.2%      12394 ±  1%  vmstat.system.cs
      7938 ±  2%     +86.4%      14800 ±  2%  time.involuntary_context_switches
         9 ±  7%     +72.2%         15 ±  5%  time.percent_of_cpu_this_job_got
     10.79 ±  4%     +52.9%      16.50 ±  4%  time.system_time
      1.18 ±  2%     +33.8%       1.57 ±  3%  turbostat.%c0
       394 ±  1%     +27.4%        502 ±  1%  iostat.sda.wrqm/s
     17.69 ±  0%     -13.3%      15.33 ±  0%  iostat.sda.avgqu-sz
      5140 ±  1%     +14.8%       5900 ±  0%  vmstat.io.bo
      5183 ±  1%     +14.5%       5935 ±  0%  iostat.sda.wkB/s
       833 ±  0%     -10.5%        746 ±  0%  iostat.sda.w/s
       122 ±  0%     -10.4%        109 ±  0%  time.elapsed_time
      1174 ±  1%      +5.4%       1238 ±  1%  vmstat.system.in
      2.17 ±  1%      -4.6%       2.06 ±  1%  turbostat.GHz
   1280314 ±  0%      +2.9%    1317252 ±  0%  time.file_system_outputs

lkp-ne04: Nehalem-EP
Memory: 12G




                                 iostat.sda.wrqm/s

  600 ++--------------------------------------------------------------------+
      |                                                                     |
  500 O+O O  O O O O O  O O O O O O  O O O O O  O O O O O  O O O            |
      |                                                                     |
      |                                                                     |
  400 *+*.*..*.*.*.*.*..*.*.*.*.*.*..*.*.*.*.*..*.*.*   *    *.*.*.*.*..*.*.*
      |                                             :   :    :              |
  300 ++                                            :   ::   :              |
      |                                              : : :  :               |
  200 ++                                             : : :  :               |
      |                                              : :  : :               |
      |                                              : :  : :               |
  100 ++                                              :   ::                |
      |                                               :    :                |
    0 ++----------------------------------------------*----*----------------+


	[*] bisect-good sample
	[O] bisect-bad  sample

To reproduce:

	apt-get install ruby ruby-oj
	git clone git://git.kernel.org/pub/scm/linux/kernel/git/wfg/lkp-tests.git
	cd lkp-tests
	bin/setup-local job.yaml # the job file attached in this email
	bin/run-local   job.yaml


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,
Fengguang


View attachment "job.yaml" of type "text/plain" (1879 bytes)

View attachment "reproduce" of type "text/plain" (1750 bytes)

_______________________________________________
LKP mailing list
LKP@...ux.intel.com

Powered by blists - more mailing lists