linux-kernel - [lkp-robot] [xfs] 19957a1816: blogbench.write

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, 2 Apr 2018 11:16:41 +0800
From:   kernel test robot <xiaolong.ye@...el.com>
To:     Vratislav Bendel <vbendel@...hat.com>
Cc:     "Darrick J. Wong" <darrick.wong@...cle.com>,
        Brian Foster <bfoster@...hat.com>,
        Christoph Hellwig <hch@....de>,
        LKML <linux-kernel@...r.kernel.org>, linux-xfs@...r.kernel.org,
        lkp@...org
Subject: [lkp-robot] [xfs]  19957a1816:  blogbench.write_score 5.2%
 improvement


Greeting,

FYI, we noticed a 5.2% improvement of blogbench.write_score due to commit:


commit: 19957a181608d25c8f4136652d0ea00b3738972d ("xfs: Correctly invert xfs_buftarg LRU isolation logic")
https://git.kernel.org/cgit/fs/xfs/xfs-linux.git xfs-4.17-merge

in testcase: blogbench
on test machine: 16 threads Intel(R) Xeon(R) CPU D-1541 @ 2.10GHz with 8G memory
with following parameters:

	disk: 1SSD
	fs: xfs
	cpufreq_governor: performance

test-description: Blogbench is a portable filesystem benchmark that tries to reproduce the load of a real-world busy file server.
test-url: https://www.pureftpd.org/project/blogbench



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/disk/fs/kconfig/rootfs/tbox_group/testcase:
  gcc-7/performance/1SSD/xfs/x86_64-rhel-7.2/debian-x86_64-2016-08-31.cgz/lkp-bdw-de1/blogbench

commit: 
  4df0f7f145 ("xfs: fix transaction allocation deadlock in IO path")
  19957a1816 ("xfs: Correctly invert xfs_buftarg LRU isolation logic")

4df0f7f145f2ac14 19957a181608d25c8f4136652d 
---------------- -------------------------- 
         %stddev     %change         %stddev
             \          |                \  
     10702            +5.2%      11259        blogbench.write_score
  76694732           -12.2%   67350219        blogbench.time.file_system_inputs
  76001327            +5.4%   80102749        blogbench.time.file_system_outputs
    661724            -3.2%     640345        blogbench.time.involuntary_context_switches
    578.00            -1.2%     571.20        blogbench.time.percent_of_cpu_this_job_got
      1603            -1.1%       1585        blogbench.time.system_time
   3413089            -8.1%    3138095        blogbench.time.voluntary_context_switches
      1.42 ±  5%     -45.0%       0.78 ±  6%  turbostat.CPU%c6
    128458           -12.8%     112013        vmstat.io.bi
    144255            +7.8%     155524        vmstat.io.bo
     36242            -4.6%      34586        vmstat.system.cs
     25403            -2.2%      24852        vmstat.system.in
      2.13            +0.0        2.15        perf-stat.branch-miss-rate%
 1.078e+11            -1.5%  1.062e+11        perf-stat.cache-misses
 1.078e+11            -1.5%  1.062e+11        perf-stat.cache-references
  10993347            -4.7%   10482153        perf-stat.context-switches
      2.09            -1.1%       2.07        perf-stat.cpi
    329741            -9.2%     299471        perf-stat.cpu-migrations
      0.48            +1.1%       0.48        perf-stat.ipc
   2468477 ± 31%     -50.9%    1212905 ± 32%  proc-vmstat.compact_daemon_free_scanned
   2468726 ± 31%     -50.7%    1216044 ± 31%  proc-vmstat.compact_free_scanned
   2923481           -13.6%    2527242        proc-vmstat.kswapd_inodesteal
      1837 ±  5%     -16.7%       1530 ±  4%  proc-vmstat.kswapd_low_wmark_hit_quickly
      1707 ±  7%     +16.6%       1989 ±  6%  proc-vmstat.nr_writeback
      2536 ±  3%     -10.8%       2262 ±  2%  proc-vmstat.pageoutrun
  38900337           -12.7%   33941396        proc-vmstat.pgpgin
    813055 ±  3%     -12.7%     709846 ±  2%  proc-vmstat.workingset_activate
   1041709 ±  2%     -10.4%     932898        proc-vmstat.workingset_refault
      2331           -12.0%       2052        slabinfo.kmalloc-2048.active_objs
      2568 ±  2%     -12.6%       2244 ±  2%  slabinfo.kmalloc-2048.num_objs
    749.80 ±  2%     -15.0%     637.20 ±  2%  slabinfo.kmalloc-4096.active_objs
    860.50 ±  2%     -15.1%     730.60 ±  3%  slabinfo.kmalloc-4096.num_objs
     25369 ±  2%     +12.1%      28440 ±  3%  slabinfo.kmalloc-512.num_objs
      1051 ±  2%     -15.2%     891.80 ±  2%  slabinfo.kmalloc-8192.active_objs
      1131 ±  2%     -16.2%     947.90 ±  2%  slabinfo.kmalloc-8192.num_objs
     18441           +16.3%      21452        slabinfo.mnt_cache.active_objs
     19600           +14.6%      22462        slabinfo.mnt_cache.num_objs
      5358 ±  3%     -27.5%       3886 ±  4%  slabinfo.pool_workqueue.active_objs
      6708 ±  4%     -25.8%       4974 ±  6%  slabinfo.pool_workqueue.num_objs



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-4.16.0-rc5-00008-g19957a1" of type "text/plain" (165943 bytes)

View attachment "job-script" of type "text/plain" (7108 bytes)

View attachment "job.yaml" of type "text/plain" (4777 bytes)

View attachment "reproduce" of type "text/plain" (419 bytes)