[<prev] [next>] [day] [month] [year] [list]
Message-ID: <20200730140659.GP23458@shao2-debian>
Date: Thu, 30 Jul 2020 22:07:00 +0800
From: kernel test robot <rong.a.chen@...el.com>
To: Like Xu <like.xu@...ux.intel.com>
Cc: Peter Zijlstra <peterz@...radead.org>,
LKML <linux-kernel@...r.kernel.org>, lkp@...ts.01.org
Subject: [perf/x86] 097e4311cd: will-it-scale.per_process_ops 7.3% improvement
Greeting,
FYI, we noticed a 7.3% improvement of will-it-scale.per_process_ops due to commit:
commit: 097e4311cda952dfb047f2a49d35aa5de500d474 ("perf/x86: Add constraint to create guest LBR event without hw counter")
https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git master
in testcase: will-it-scale
on test machine: 104 threads Skylake with 192G memory
with following parameters:
nr_task: 100%
mode: process
test: signal1
cpufreq_governor: performance
ucode: 0x2006906
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
=========================================================================================
compiler/cpufreq_governor/kconfig/mode/nr_task/rootfs/tbox_group/test/testcase/ucode:
gcc-9/performance/x86_64-rhel-8.3/process/100%/debian-10.4-x86_64-20200603.cgz/lkp-skl-fpga01/signal1/will-it-scale/0x2006906
commit:
b2d6504761 ("perf/x86/lbr: Add interface to get LBR information")
097e4311cd ("perf/x86: Add constraint to create guest LBR event without hw counter")
b2d6504761a50b94 097e4311cda952dfb047f2a49d3
---------------- ---------------------------
%stddev %change %stddev
\ | \
86644 +7.3% 92967 will-it-scale.per_process_ops
9011113 +7.3% 9668655 will-it-scale.workload
449.40 ± 36% +68.3% 756.40 ± 42% interrupts.CPU58.CAL:Function_call_interrupts
469.28 ± 5% -14.3% 402.14 ± 12% sched_debug.cpu.curr->pid.stddev
1102 ± 9% -22.1% 859.00 ± 13% slabinfo.file_lock_cache.active_objs
1102 ± 9% -22.1% 859.00 ± 13% slabinfo.file_lock_cache.num_objs
73.00 -2.2% 71.40 vmstat.cpu.sy
25.00 +5.6% 26.40 vmstat.cpu.us
18725 ± 2% +12.6% 21082 ± 5% softirqs.CPU10.RCU
18903 ± 3% +9.3% 20662 ± 3% softirqs.CPU103.RCU
20904 ± 2% +9.5% 22893 ± 3% softirqs.CPU24.RCU
20333 ± 8% +12.1% 22802 ± 2% softirqs.CPU25.RCU
20544 ± 13% +15.9% 23820 ± 3% softirqs.CPU55.RCU
17201 ± 9% +18.4% 20373 ± 5% softirqs.CPU77.RCU
18801 ± 4% +13.8% 21400 ± 8% softirqs.CPU8.RCU
21434 ± 2% +15.0% 24640 ± 7% softirqs.CPU81.RCU
18689 ± 3% +11.3% 20794 ± 4% softirqs.CPU9.RCU
18841 ± 2% +10.8% 20882 ± 3% softirqs.CPU92.RCU
18786 ± 2% +10.7% 20792 ± 3% softirqs.CPU97.RCU
18295 ± 11% +15.0% 21030 ± 4% softirqs.CPU98.RCU
18809 ± 2% +10.0% 20695 ± 2% softirqs.CPU99.RCU
will-it-scale.per_process_ops
98000 +-------------------------------------------------------------------+
| O O O |
96000 |-+ |
| O O |
94000 |-+ O O O O O |
| O O O O O O O O O O O O |
92000 |-+ |
| |
90000 |-+ .+ |
| +. : |
88000 |-+ : : |
| : : .+..+.. |
86000 |.. .+..+..+..+. .+.. : +.+. +..+ |
| .+..+.+..+. +. + |
84000 +-------------------------------------------------------------------+
[*] 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,
Rong Chen
View attachment "config-5.8.0-rc3-00004-g097e4311cda95" of type "text/plain" (158415 bytes)
View attachment "job-script" of type "text/plain" (7392 bytes)
View attachment "job.yaml" of type "text/plain" (5083 bytes)
View attachment "reproduce" of type "text/plain" (340 bytes)
Powered by blists - more mailing lists