| 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
| ||
|
Message-ID: <20170531063226.GE1795@yexl-desktop>
Date: Wed, 31 May 2017 14:32:26 +0800
From: kernel test robot <xiaolong.ye@...el.com>
To: Filipe Manana <fdmanana@...e.com>
Cc: LKML <linux-kernel@...r.kernel.org>,
Linus Torvalds <torvalds@...ux-foundation.org>, lkp@...org
Subject: [lkp-robot] [Btrfs] a7e3b975a0: aim7.jobs-per-min -9.8% regression
Greeting,
FYI, we noticed a -9.8% regression of aim7.jobs-per-min due to commit:
commit: a7e3b975a0f9296162b72ac6ab7fad9631a07630 ("Btrfs: fix reported number of inode blocks")
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: btrfs
test: disk_wrt
load: 1500
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/
In addition to that, the commit also has significant impact on the following tests:
+------------------+-----------------------------------------------------------------------+
| testcase: change | aim7: aim7.jobs-per-min -10.6% regression |
| test machine | 40 threads Intel(R) Xeon(R) CPU E5-2690 v2 @ 3.00GHz with 384G memory |
| test parameters | cpufreq_governor=performance |
| | disk=1BRD_48G |
| | fs=btrfs |
| | load=1500 |
| | test=disk_cp |
+------------------+-----------------------------------------------------------------------+
| testcase: change | aim7: aim7.jobs-per-min -2.8% regression |
| test machine | 40 threads Intel(R) Xeon(R) CPU E5-2690 v2 @ 3.00GHz with 384G memory |
| test parameters | cpufreq_governor=performance |
| | disk=4BRD_12G |
| | fs=btrfs |
| | load=1500 |
| | md=RAID0 |
| | test=disk_rr |
+------------------+-----------------------------------------------------------------------+
| testcase: change | aim7: aim7.jobs-per-min -10.3% regression |
| test machine | 40 threads Intel(R) Xeon(R) CPU E5-2690 v2 @ 3.00GHz with 384G memory |
| test parameters | cpufreq_governor=performance |
| | disk=4BRD_12G |
| | fs=btrfs |
| | load=1500 |
| | md=RAID0 |
| | test=disk_cp |
+------------------+-----------------------------------------------------------------------+
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-btrfs-disk_wrt-1500-performance/lkp-ivb-ep01
e1cbfd7bf6dabdac a7e3b975a0f9296162b72ac6ab
---------------- --------------------------
54221 -10% 48881 aim7.jobs-per-min
453341 426% 2385786 ± 4% aim7.time.voluntary_context_switches
166 11% 184 aim7.time.elapsed_time
166 11% 184 aim7.time.elapsed_time.max
6149 4% 6386 aim7.time.system_time
789480 -45% 431017 ± 4% aim7.time.involuntary_context_switches
528100 12% 589553 ± 5% interrupts.CAL:Function_call_interrupts
2852 ± 10% -49% 1468 ± 7% vmstat.io.bo
13002 124% 29105 vmstat.system.cs
165 -7% 154 turbostat.PkgWatt
138 -8% 126 turbostat.CorWatt
92.60 -6% 87.04 turbostat.%Busy
2842 -9% 2586 turbostat.Avg_MHz
2182198 149% 5435023 ± 3% perf-stat.context-switches
54816290 ± 12% 19% 65316524 ± 9% perf-stat.iTLB-load-misses
302089 76% 530355 ± 5% perf-stat.cpu-migrations
0.21 ± 3% 45% 0.30 perf-stat.branch-miss-rate%
2.136e+09 ± 3% 52% 3.237e+09 perf-stat.branch-misses
1.707e+10 26% 2.143e+10 perf-stat.cache-references
2.197e+09 16% 2.553e+09 perf-stat.node-load-misses
1.907e+09 -12% 1.678e+09 perf-stat.node-store-misses
736175 9% 799843 ± 3% perf-stat.page-faults
736322 9% 799871 ± 3% perf-stat.minor-faults
2.669e+09 -18% 2.2e+09 perf-stat.node-stores
2.759e+11 11% 3.074e+11 ± 4% perf-stat.dTLB-stores
5.516e+09 -3% 5.333e+09 perf-stat.cache-misses
45.63 46.21 perf-stat.node-load-miss-rate%
0.24 4% 0.25 perf-stat.ipc
2.618e+09 14% 2.971e+09 perf-stat.node-loads
41.67 4% 43.26 perf-stat.node-store-miss-rate%
4.524e+12 5% 4.757e+12 perf-stat.instructions
1.192e+12 5% 1.249e+12 perf-stat.dTLB-loads
1.038e+12 4% 1.085e+12 perf-stat.branch-instructions
32.32 -23% 24.89 perf-stat.cache-miss-rate%
perf-stat.cache-references
2.5e+10 ++----------------------------------------------------------------+
| |
| OO OOOO OOOO OOOO OOO O |
2e+10 OO OOOO OOOO O |
| * * *.* *.*** ** .* ** * * * ** *
**.* **.*** : : ** *.* * **.* *.** *.* *.****.* **.* *.*|
1.5e+10 ++ : : : |
| : :: : |
1e+10 ++ : :: : |
| : : :: |
| :: : |
5e+09 ++ :: : |
| : : |
| : : |
0 ++---------*--*---------------------------------------------------+
perf-stat.node-store-miss-rate_
45 +O----------------O------------O-O---OO--------------------------------+
O*.OOO.OO.OOO OOO.*OO.OO.OOO.OO*.*O.O**.***.***.**.***.***.***.**.***.**
40 ++ : : : |
35 ++ : : : |
| : : : |
30 ++ : :: : |
25 ++ : :: : |
| : :: : |
20 ++ :: : |
15 ++ :: : |
| :: : |
10 ++ : : |
5 ++ : : |
| : : |
0 ++----------*--*-------------------------------------------------------+
aim7.jobs-per-min
57000 ++------------------------------------------------------------------+
56000 ++.* |
|* : *.* .* *. * * .* * * |
55000 *+ :: : * * * :+ *. :*.***. ** : .** :.* : *.* *.* *.* *. |
54000 ++ * ** * ** ** * ** * *.** * :* * *|
| * *
53000 ++ |
52000 ++ |
51000 ++ |
| O |
50000 O+ O O O |
49000 +O O O O O O OO OO O O O |
| O O O O O O O |
48000 ++ O O O |
47000 ++---------------O--------------------------------------------------+
[*] 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-4.10.0-rc8-00213-ga7e3b97" of type "text/plain" (156159 bytes)
View attachment "job-script" of type "text/plain" (6983 bytes)
View attachment "job.yaml" of type "text/plain" (4566 bytes)
View attachment "reproduce" of type "text/plain" (529 bytes)
Powered by blists - more mailing lists