| 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: <20171016023917.GG26196@yexl-desktop>
Date: Mon, 16 Oct 2017 10:39:17 +0800
From: kernel test robot <xiaolong.ye@...el.com>
To: Borislav Petkov <bp@...en8.de>
Cc: Andy Lutomirski <luto@...nel.org>, x86@...nel.org,
linux-kernel@...r.kernel.org,
Markus Trippelsdorf <markus@...ppelsdorf.de>,
Adam Borowski <kilobyte@...band.pl>,
Brian Gerst <brgerst@...il.com>,
Johannes Hirte <johannes.hirte@...enkhaos.de>, lkp@...org
Subject: [lkp-robot] [x86/mm] c4c3c3c2d0: will-it-scale.per_process_ops
-61.0% regression
Greeting,
FYI, we noticed a -61.0% regression of will-it-scale.per_process_ops due to commit:
commit: c4c3c3c2d00826c88b5c02c20e80704664424b9b ("x86/mm: Flush more aggressively in lazy TLB mode")
url: https://github.com/0day-ci/linux/commits/Borislav-Petkov/x86-mm-Flush-more-aggressively-in-lazy-TLB-mode/20171011-115901
in testcase: will-it-scale
on test machine: 88 threads Intel(R) Xeon(R) CPU E5-2699 v4 @ 2.20GHz with 64G memory
with following parameters:
nr_task: 50%
mode: process
test: context_switch1
cpufreq_governor: performance
test-description: Will It Scale takes a testcase and runs it from 1 through to n parallel copies to see if the testcase will scale. It builds both a process and threads based test in order to see any differences between the two.
test-url: https://github.com/antonblanchard/will-it-scale
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: will-it-scale/50%-process-context_switch1-performance/lkp-bdw-ep3d
5b8809deb4b0a77f c4c3c3c2d00826c88b5c02c20e
---------------- --------------------------
%stddev change %stddev
\ | \
448554 -61% 174892 will-it-scale.per_process_ops
8.08 17% 9.44 turbostat.RAMWatt
287 -18% 236 turbostat.PkgWatt
858233 -55% 389530 ± 5% vmstat.system.in
39084502 -62% 14806021 ± 5% vmstat.system.cs
1.171e+08 1546% 1.927e+09 perf-stat.node-store-misses
13683987 580% 93114744 perf-stat.node-loads
4.456e+09 317% 1.858e+10 perf-stat.cache-misses
7.64 ± 6% 222% 24.62 perf-stat.cache-miss-rate%
1.11 159% 2.87 perf-stat.cpi
4.858e+08 56% 7.577e+08 perf-stat.node-load-misses
5.857e+10 ± 6% 29% 7.545e+10 perf-stat.cache-references
77.79 27% 98.55 perf-stat.node-store-miss-rate%
0.00 20% 0.00 ± 14% perf-stat.dTLB-load-miss-rate%
0.25 16% 0.29 perf-stat.branch-miss-rate%
4927 5% 5154 perf-stat.instructions-per-iTLB-miss
7.148e+13 7.359e+13 perf-stat.cpu-cycles
97.26 -8% 89.06 perf-stat.node-load-miss-rate%
33424758 ± 3% -15% 28449849 perf-stat.node-stores
159781 -23% 122595 ± 4% perf-stat.cpu-migrations
23.33 -33% 15.67 perf-stat.iTLB-load-miss-rate%
4.31e+10 -38% 2.68e+10 perf-stat.iTLB-loads
4.626e+08 -53% 2.194e+08 ± 15% perf-stat.dTLB-load-misses
3.123e+10 -54% 1.426e+10 perf-stat.branch-misses
6.462e+13 -60% 2.567e+13 perf-stat.instructions
1.209e+13 -60% 4.798e+12 perf-stat.dTLB-stores
1.846e+13 -60% 7.308e+12 perf-stat.dTLB-loads
1.247e+13 -61% 4.922e+12 perf-stat.branch-instructions
1.188e+10 -61% 4.629e+09 perf-stat.context-switches
0.90 -61% 0.35 perf-stat.ipc
1.311e+10 -62% 4.979e+09 perf-stat.iTLB-load-misses
will-it-scale.per_process_ops
500000 +-+----------------------------------------------------------------+
|.+..+.+..+.+.+..+.+..+.+.+..+.+..+.+.+..+.+. |
450000 +-+ +..+.+..+.+.+..+.+..+.|
| |
400000 +-+ |
350000 +-+ |
| |
300000 +-+ |
| |
250000 +-+ |
200000 +-+ |
| O O O O O O O O O O |
150000 O-O O O O O O O O O O O O |
| O O |
100000 +-+----------------------------------------------------------------+
[*] 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-4.14.0-rc3-00696-gc4c3c3c" of type "text/plain" (163429 bytes)
View attachment "job-script" of type "text/plain" (7038 bytes)
View attachment "job.yaml" of type "text/plain" (4657 bytes)
View attachment "reproduce" of type "text/plain" (309 bytes)
Powered by blists - more mailing lists