| 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: <20170703024825.GC7070@yexl-desktop>
Date: Mon, 3 Jul 2017 10:48:25 +0800
From: kernel test robot <xiaolong.ye@...el.com>
To: Rik van Riel <riel@...hat.com>
Cc: Ingo Molnar <mingo@...nel.org>,
Linus Torvalds <torvalds@...ux-foundation.org>,
Mel Gorman <mgorman@...e.de>, Mike Galbraith <efault@....de>,
Peter Zijlstra <peterz@...radead.org>,
Thomas Gleixner <tglx@...utronix.de>,
LKML <linux-kernel@...r.kernel.org>,
Stephen Rothwell <sfr@...b.auug.org.au>, lkp@...org
Subject: [lkp-robot] [sched/numa] 3fed382b46: hackbench.throughput -9.8%
regression
Greeting,
FYI, we noticed a -9.8% regression of hackbench.throughput due to commit:
commit: 3fed382b46baac83703130fe4cd3d9147f427fb9 ("sched/numa: Implement NUMA node level wake_affine()")
https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git master
in testcase: hackbench
on test machine: 12 threads Intel(R) Atom(R) CPU 3958 @ 2.00GHz with 64G memory
with following parameters:
nr_threads: 50%
mode: threads
ipc: pipe
cpufreq_governor: performance
test-description: Hackbench is both a benchmark and a stress test for the Linux kernel scheduler.
test-url: https://github.com/linux-test-project/ltp/blob/master/testcases/kernel/sched/cfs-scheduler/hackbench.c
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: hackbench/50%-threads-pipe-performance/lkp-denverton3
7d894e6e34a5cdd1 3fed382b46baac83703130fe4c
---------------- --------------------------
42031 -10% 37920 hackbench.throughput
5.792e+08 16% 6.694e+08 hackbench.time.voluntary_context_switches
603 5% 631 hackbench.time.elapsed_time
603 5% 631 hackbench.time.elapsed_time.max
58567750 59709574 hackbench.time.involuntary_context_switches
502 -3% 485 hackbench.time.user_time
1456 -4% 1393 hackbench.time.percent_of_cpu_this_job_got
14747 -6% 13935 hackbench.time.minor_page_faults
104278 5% 109719 interrupts.CAL:Function_call_interrupts
746913 5% 782073 vmstat.system.in
1190120 13% 1349585 vmstat.system.cs
19981084 61% 32149358 perf-stat.cpu-migrations
0.56 0.55 perf-stat.ipc
0.75 4% 0.78 perf-stat.branch-miss-rate%
1.457e+10 5% 1.534e+10 perf-stat.branch-misses
3.12e+10 3.076e+10 perf-stat.iTLB-load-misses
1.937e+12 1.965e+12 perf-stat.branch-instructions
1.028e+13 1.042e+13 perf-stat.instructions
0.30 0.29 perf-stat.iTLB-load-miss-rate%
7.206e+08 19% 8.549e+08 perf-stat.context-switches
329.37 338.63 perf-stat.instructions-per-iTLB-miss
633466 5% 667986 perf-stat.page-faults
633460 5% 667986 perf-stat.minor-faults
4.053e+11 10% 4.465e+11 perf-stat.cache-references
1.027e+13 1.041e+13 perf-stat.iTLB-loads
1.843e+13 1.893e+13 perf-stat.cpu-cycles
6.566e+10 6.43e+10 perf-stat.cache-misses
16.20 -11% 14.40 perf-stat.cache-miss-rate%
perf-stat.cpu-migrations
3.4e+07 ++----------------------------------------------------------------+
O O O O |
3.2e+07 ++ O O O |
3e+07 ++ O O O |
| O O O O O O O O O O O |
2.8e+07 ++ O O |
| |
2.6e+07 ++ |
| |
2.4e+07 ++ |
2.2e+07 ++ .* |
*..*.*..*.*..*.*..*.*..*.*..*.*..*.. .*. + .*..*. |
2e+07 ++ *.*..* *..* *..*.*..*
| |
1.8e+07 ++----------------------------------------------------------------+
hackbench.throughput
43000 ++------------------------------------------------------------------+
| .*. .*.. |
42000 ++ .*. *. *.*..*..*.*..*.*..*
41000 ++ .*..*. .*.. .*.. .* |
*..*.*..*.*. *..*..* * *. |
40000 ++ |
| |
39000 ++ |
| |
38000 O+ O O O O O O |
37000 ++ O O O O O O |
| O O O O O O O |
36000 ++ O |
| O O |
35000 ++------------------------------------------------------------------+
[*] 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-rc6-00260-g3fed382" of type "text/plain" (159327 bytes)
View attachment "job-script" of type "text/plain" (6767 bytes)
View attachment "job.yaml" of type "text/plain" (4366 bytes)
View attachment "reproduce" of type "text/plain" (1110 bytes)
Powered by blists - more mailing lists