[<prev] [next>] [day] [month] [year] [list]
Message-ID: <20170810013920.GH24602@yexl-desktop>
Date: Thu, 10 Aug 2017 09:39:20 +0800
From: kernel test robot <xiaolong.ye@...el.com>
To: Brian Foster <bfoster@...hat.com>
Cc: "Darrick J. Wong" <darrick.wong@...cle.com>,
LKML <linux-kernel@...r.kernel.org>,
Linus Torvalds <torvalds@...ux-foundation.org>, lkp@...org
Subject: [lkp-robot] [xfs] e2f2342639: aim7.jobs-per-min 8.2% improvement
Greeting,
FYI, we noticed a 8.2% improvement of aim7.jobs-per-min due to commit:
commit: e2f2342639a414b60de3876a8b437eac2b795dbe ("xfs: refactor xlog_cil_insert_items() to facilitate transaction dump")
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master
in testcase: aim7
on test machine: 40 threads Intel(R) Xeon(R) CPU E5-2690 v2 @ 3.00GHz with 384G memory
with following parameters:
disk: 1BRD_48G
fs: xfs
test: disk_cp
load: 3000
cpufreq_governor: performance
test-description: AIM7 is a traditional UNIX system level benchmark suite which is used to test and measure the performance of multiuser system.
test-url: https://sourceforge.net/projects/aimbench/files/aim-suite7/
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: aim7/1BRD_48G-xfs-disk_cp-3000-performance/lkp-ivb-ep01
7d2d5653460443e0 e2f2342639a414b60de3876a8b
---------------- --------------------------
%stddev change %stddev
\ | \
80127 8% 86664 aim7.jobs-per-min
225 -8% 208 aim7.time.elapsed_time
225 -8% 208 aim7.time.elapsed_time.max
8212 -8% 7536 aim7.time.system_time
430021 ± 4% -12% 379972 ± 4% aim7.time.involuntary_context_switches
0.26 0.27 perf-stat.ipc
3.80 3.72 perf-stat.cpi
1.803e+12 -6% 1.695e+12 perf-stat.dTLB-loads
6.486e+12 -6% 6.081e+12 perf-stat.instructions
1.423e+12 -7% 1.322e+12 perf-stat.branch-instructions
2.463e+13 -8% 2.262e+13 perf-stat.cpu-cycles
66.00 ± 27% -44% 37.19 ± 20% perf-stat.iTLB-load-miss-rate%
aim7.jobs-per-min
89000 ++O----O-------------------------------------------O----------------+
88000 O+ O O O |
| O O O O O |
87000 ++ O O O |
86000 ++ O O O O O O |
85000 ++ O O |
84000 ++ * * |
| : + + : |
83000 ++ *.. : * *.. .*.. + : .*.* * |
82000 ++ : * : : * *. .* *. + + + *.. |
81000 ++ : : : * + .* + .* : |
80000 ++*..*.*.. : : : * * + : *.|
* : * + : *
79000 ++ * * |
78000 ++------------------------------------------------------------------+
[*] 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-rc6-00007-ge2f2342" of type "text/plain" (159256 bytes)
View attachment "job-script" of type "text/plain" (7420 bytes)
View attachment "job.yaml" of type "text/plain" (5052 bytes)
View attachment "reproduce" of type "text/plain" (695 bytes)
Powered by blists - more mailing lists