linux-kernel - [lkp-robot] [sched/fair] 6c362d9657: netperf.Throughput

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: <20170910141730.GC29265@yexl-desktop>
Date:   Sun, 10 Sep 2017 22:17:30 +0800
From:   kernel test robot <xiaolong.ye@...el.com>
To:     Peter Zijlstra <peterz@...radead.org>
Cc:     Chris Wilson <chris@...is-wilson.co.uk>,
        LKML <linux-kernel@...r.kernel.org>,
        Daniel Vetter <daniel.vetter@...ll.ch>,
        intel-gfx@...ts.freedesktop.org, dri-devel@...ts.freedesktop.org,
        lkp@...org
Subject: [lkp-robot] [sched/fair]  6c362d9657:  netperf.Throughput_Mbps
 -10.1% regression


Greeting,

FYI, we noticed a -10.1% regression of netperf.Throughput_Mbps due to commit:


commit: 6c362d9657293d700a8299acbeb2f1e24378f488 ("sched/fair: Fix wake_affine() for !NUMA_BALANCING")
git://anongit.freedesktop.org/drm-intel topic/core-for-CI

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

	ip: ipv4
	runtime: 300s
	nr_threads: 200%
	cluster: cs-localhost
	test: TCP_SENDFILE
	cpufreq_governor: performance

test-description: Netperf is a benchmark that can be use to measure various aspect of networking performance.
test-url: http://www.netperf.org/netperf/


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

testcase/path_params/tbox_group/run: netperf/ipv4-300s-200%-cs-localhost-TCP_SENDFILE-performance/lkp-bdw-de1

0e8e8706dfe2a17a  6c362d9657293d700a8299acbe  
----------------  --------------------------  
     11039             -10%       9929        netperf.Throughput_Mbps
     36660 ± 20%      1257%     497519        netperf.time.involuntary_context_switches
      7069                        6997        netperf.time.minor_page_faults
      2899              -6%       2734        netperf.time.system_time
       983              -6%        925        netperf.time.percent_of_cpu_this_job_got
     62.39             -13%      54.04        netperf.time.user_time
   3137334             -63%    1171532        netperf.time.voluntary_context_switches
     16915              46%      24638        vmstat.system.in
     31608              69%      53301        vmstat.system.cs
     64003 ±  6%       443%     347394        perf-stat.cpu-migrations
     35981 ±  6%       205%     109811 ±  4%  perf-stat.instructions-per-iTLB-miss
      0.16              63%       0.25        perf-stat.branch-miss-rate%
 2.538e+09              57%  3.977e+09        perf-stat.branch-misses
      0.81              -4%       0.77        perf-stat.ipc
      1.24               5%       1.30        perf-stat.cpi
    370027                      364969        perf-stat.page-faults
    370025                      364952        perf-stat.minor-faults
 2.094e+12              -6%   1.97e+12        perf-stat.dTLB-stores
 3.212e+12              -5%   3.04e+12        perf-stat.dTLB-loads
 1.632e+12              -4%  1.566e+12        perf-stat.branch-instructions
 8.997e+12              -4%  8.606e+12        perf-stat.instructions
 1.874e+11              11%  2.079e+11        perf-stat.cache-references
 1.874e+11              11%  2.079e+11        perf-stat.cache-misses
      8.82 ±  6%       -82%       1.59 ±  5%  perf-stat.iTLB-load-miss-rate%
      0.00             157%       0.01        perf-stat.dTLB-load-miss-rate%
 1.104e+08             143%  2.682e+08        perf-stat.dTLB-load-misses
 2.511e+08 ±  6%       -69%   78564886 ±  5%  perf-stat.iTLB-load-misses
 2.594e+09              88%   4.87e+09        perf-stat.iTLB-loads
   9571397              69%   16145347        perf-stat.context-switches




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-4.13.0-00005-g6c362d9" of type "text/plain" (161025 bytes)

View attachment "job-script" of type "text/plain" (7324 bytes)

View attachment "job.yaml" of type "text/plain" (4979 bytes)

View attachment "reproduce" of type "text/plain" (2064 bytes)