From: kernel test robot <xiaolong.ye@intel.com>
To: lkp@lists.01.org
Subject: [sched/fair] ec419c38ec: unixbench.score +2.7% improvement
Date: Wed, 13 Apr 2016 09:59:09 +0800 [thread overview]
Message-ID: <20160413015909.GD659@yexl-desktop> (raw)
[-- Attachment #1: Type: text/plain, Size: 14297 bytes --]
FYI, we noticed that unixbench.score +2.7% improvement on
git://bee.sh.intel.com/git/ydu19/tip flat_hierarchy_v2
commit ec419c38ecd766e8e41e3232e9f3d473c7d56b3a ("sched/fair: Drop out incomplete current period when sched averages accrue")
=========================================================================================
compiler/kconfig/nr_task/rootfs/tbox_group/test/testcase:
gcc-4.9/x86_64-rhel/100%/debian-x86_64-2015-02-07.cgz/nhm-white2/shell8/unixbench
commit:
19aa29a06823363298646327ecf461e70758b483
ec419c38ecd766e8e41e3232e9f3d473c7d56b3a
19aa29a068233632 ec419c38ecd766e8e41e3232e9
---------------- --------------------------
fail:runs %reproduction fail:runs
| | |
10397 ± 0% +2.7% 10673 ± 0% unixbench.score
1839526 ± 0% -57.9% 774575 ± 0% unixbench.time.involuntary_context_switches
1.245e+08 ± 0% +2.7% 1.278e+08 ± 0% unixbench.time.minor_page_faults
737.00 ± 0% +1.8% 750.50 ± 0% unixbench.time.percent_of_cpu_this_job_got
556.78 ± 0% +3.7% 577.20 ± 0% unixbench.time.user_time
12060 ± 31% +247.1% 41859 ± 36% latency_stats.sum.pipe_read.__vfs_read.vfs_read.SyS_read.entry_SYSCALL_64_fastpath
5391 ± 1% +16.0% 6252 ± 0% meminfo.KernelStack
137262 ± 0% -38.4% 84501 ± 0% softirqs.SCHED
1839526 ± 0% -57.9% 774575 ± 0% time.involuntary_context_switches
37242 ± 0% -30.6% 25830 ± 0% vmstat.system.cs
10778 ± 0% -8.5% 9857 ± 0% vmstat.system.in
93.38 ± 0% +1.4% 94.71 ± 0% turbostat.%Busy
2731 ± 0% +1.4% 2770 ± 0% turbostat.Avg_MHz
1.48 ± 1% -88.9% 0.17 ± 5% turbostat.CPU%c1
1676 ± 1% +13.2% 1897 ± 1% slabinfo.cred_jar.active_objs
10224 ± 3% +12.1% 11462 ± 4% slabinfo.kmalloc-32.active_objs
10444 ± 3% +11.6% 11658 ± 4% slabinfo.kmalloc-32.num_objs
2548 ± 4% +5.7% 2692 ± 5% slabinfo.pid.num_objs
5442033 ± 1% -92.1% 432360 ± 8% cpuidle.C1-NHM.time
90309 ± 1% -86.4% 12297 ± 27% cpuidle.C1-NHM.usage
9093083 ± 2% -93.8% 567479 ± 10% cpuidle.C1E-NHM.time
138870 ± 1% -94.2% 8031 ± 13% cpuidle.C1E-NHM.usage
7175826 ± 3% -84.9% 1083522 ± 7% cpuidle.C3-NHM.time
91126 ± 3% -95.1% 4434 ± 8% cpuidle.C3-NHM.usage
21635 ± 4% -55.7% 9588 ± 2% cpuidle.C6-NHM.usage
1547 ± 17% -91.6% 130.00 ± 18% cpuidle.POLL.time
81.00 ± 12% -88.9% 9.00 ± 13% cpuidle.POLL.usage
63.96 ± 21% -58.2% 26.70 ± 68% sched_debug.cfs_rq:/.exec_clock.stddev
78.16 ± 39% -87.7% 9.62 ±143% sched_debug.cfs_rq:/.load.avg
13.81 ± 36% -89.1% 1.50 ± 0% sched_debug.cfs_rq:/.load.min
83.85 ± 4% +3.9e+07% 33031208 ± 0% sched_debug.cfs_rq:/.load_avg.avg
147.62 ± 11% +2.8e+07% 41895680 ± 6% sched_debug.cfs_rq:/.load_avg.max
58.88 ± 1% +3.9e+07% 22927040 ± 16% sched_debug.cfs_rq:/.load_avg.min
30.81 ± 15% +2e+07% 6076286 ± 33% sched_debug.cfs_rq:/.load_avg.stddev
1967526 ± 0% +2084.6% 42982191 ± 0% sched_debug.cfs_rq:/.min_vruntime.avg
1986032 ± 0% +2073.0% 43156774 ± 0% sched_debug.cfs_rq:/.min_vruntime.max
1948430 ± 0% +2097.8% 42822398 ± 0% sched_debug.cfs_rq:/.min_vruntime.min
12142 ± 27% +795.8% 108769 ± 21% sched_debug.cfs_rq:/.min_vruntime.stddev
0.09 ±-1066% +18625.0% 17.55 ± 4% sched_debug.cfs_rq:/.nr_spread_over.avg
0.75 ±-133% +5133.3% 39.25 ± 24% sched_debug.cfs_rq:/.nr_spread_over.max
0.25 ±-403% +5061.5% 12.80 ± 22% sched_debug.cfs_rq:/.nr_spread_over.stddev
58.11 ± 3% +1345.5% 840.00 ± 1% sched_debug.cfs_rq:/.runnable_load_avg.avg
68.00 ± 2% +1500.0% 1088 ± 10% sched_debug.cfs_rq:/.runnable_load_avg.max
43.75 ± 33% +1655.4% 768.00 ± 0% sched_debug.cfs_rq:/.runnable_load_avg.min
7.60 ± 56% +1637.8% 132.02 ± 27% sched_debug.cfs_rq:/.runnable_load_avg.stddev
22539 ± 60% +808.0% 204668 ± 55% sched_debug.cfs_rq:/.spread0.max
-15041 ±-70% +761.6% -129597 ±-39% sched_debug.cfs_rq:/.spread0.min
12137 ± 27% +796.4% 108802 ± 20% sched_debug.cfs_rq:/.spread0.stddev
811.48 ± 0% +43541.4% 354139 ± 5% sched_debug.cfs_rq:/.util_avg.avg
869.12 ± 1% +1.9e+05% 1638566 ± 5% sched_debug.cfs_rq:/.util_avg.max
743.88 ± 1% +727.2% 6153 ± 44% sched_debug.cfs_rq:/.util_avg.min
39.33 ± 6% +1.4e+06% 555702 ± 8% sched_debug.cfs_rq:/.util_avg.stddev
480108 ± 5% +13.3% 544065 ± 2% sched_debug.cpu.avg_idle.avg
212617 ± 21% +54.0% 327335 ± 11% sched_debug.cpu.avg_idle.min
186637 ± 6% -31.0% 128767 ± 12% sched_debug.cpu.avg_idle.stddev
57.17 ± 1% +1383.2% 848.00 ± 3% sched_debug.cpu.cpu_load[0].avg
67.75 ± 0% +1505.9% 1088 ± 10% sched_debug.cpu.cpu_load[0].max
38.50 ± 18% +1894.8% 768.00 ± 0% sched_debug.cpu.cpu_load[0].min
9.14 ± 24% +1344.5% 132.01 ± 30% sched_debug.cpu.cpu_load[0].stddev
57.66 ± 0% +1352.1% 837.35 ± 1% sched_debug.cpu.cpu_load[1].avg
66.75 ± 0% +1380.6% 988.31 ± 3% sched_debug.cpu.cpu_load[1].max
45.50 ± 5% +1579.3% 764.06 ± 0% sched_debug.cpu.cpu_load[1].min
6.45 ± 12% +1149.7% 80.65 ± 18% sched_debug.cpu.cpu_load[1].stddev
57.53 ± 0% +1351.2% 834.91 ± 0% sched_debug.cpu.cpu_load[2].avg
66.12 ± 2% +1316.4% 936.62 ± 2% sched_debug.cpu.cpu_load[2].max
48.50 ± 2% +1476.8% 764.75 ± 1% sched_debug.cpu.cpu_load[2].min
5.22 ± 8% +985.5% 56.65 ± 12% sched_debug.cpu.cpu_load[2].stddev
57.43 ± 0% +1350.6% 833.06 ± 0% sched_debug.cpu.cpu_load[3].avg
65.12 ± 2% +1290.8% 905.75 ± 2% sched_debug.cpu.cpu_load[3].max
50.56 ± 1% +1433.4% 775.31 ± 1% sched_debug.cpu.cpu_load[3].min
4.47 ± 6% +834.0% 41.77 ± 17% sched_debug.cpu.cpu_load[3].stddev
57.62 ± 0% +1352.0% 836.62 ± 0% sched_debug.cpu.cpu_load[4].avg
64.62 ± 2% +1281.2% 892.62 ± 3% sched_debug.cpu.cpu_load[4].max
51.56 ± 1% +1437.8% 792.94 ± 1% sched_debug.cpu.cpu_load[4].min
3.96 ± 11% +696.6% 31.52 ± 30% sched_debug.cpu.cpu_load[4].stddev
7178 ± 18% -91.6% 599.44 ± 0% sched_debug.cpu.curr->pid.min
62.48 ± 6% -84.6% 9.62 ±143% sched_debug.cpu.load.avg
14.50 ± 27% -89.7% 1.50 ± 0% sched_debug.cpu.load.min
1280 ± 20% +35.3% 1733 ± 5% sched_debug.cpu.nr_load_updates.stddev
424311 ± 0% -30.2% 296261 ± 0% sched_debug.cpu.nr_switches.avg
442171 ± 0% -29.6% 311189 ± 1% sched_debug.cpu.nr_switches.max
409575 ± 0% -31.9% 278989 ± 1% sched_debug.cpu.nr_switches.min
65.50 ± 13% +167.3% 175.06 ± 21% sched_debug.cpu.nr_uninterruptible.max
-68.31 ±-28% +323.2% -289.12 ±-39% sched_debug.cpu.nr_uninterruptible.min
43.89 ± 17% +210.5% 136.27 ± 29% sched_debug.cpu.nr_uninterruptible.stddev
420593 ± 0% -30.5% 292299 ± 0% sched_debug.cpu.sched_count.avg
435963 ± 0% -30.0% 305050 ± 0% sched_debug.cpu.sched_count.max
406393 ± 0% -32.0% 276478 ± 1% sched_debug.cpu.sched_count.min
20000 ± 2% -90.7% 1865 ± 8% sched_debug.cpu.sched_goidle.avg
21505 ± 2% -78.9% 4535 ± 22% sched_debug.cpu.sched_goidle.max
18583 ± 3% -94.4% 1049 ± 8% sched_debug.cpu.sched_goidle.min
209474 ± 0% -22.9% 161408 ± 0% sched_debug.cpu.ttwu_count.avg
222079 ± 0% -22.9% 171331 ± 1% sched_debug.cpu.ttwu_count.max
198750 ± 0% -24.5% 149958 ± 1% sched_debug.cpu.ttwu_count.min
157706 ± 0% -20.1% 125986 ± 0% sched_debug.cpu.ttwu_local.avg
163360 ± 0% -20.5% 129947 ± 0% sched_debug.cpu.ttwu_local.max
152244 ± 0% -19.5% 122494 ± 0% sched_debug.cpu.ttwu_local.min
3666 ± 6% -35.3% 2373 ± 17% sched_debug.cpu.ttwu_local.stddev
nhm-white2: Nehalem
Memory: 4G
unixbench.time.user_time
580 ++-----------------------O-----O--------------------------------------+
| OO O OO O OO OO |
| O O O OO OO |
575 +O OO O O O O |
O |
| |
570 ++ |
| |
565 ++ |
| |
| * |
560 **.**. *.**. *. .**.**. *.**. *.**.**. *.**. *.* .* : |
| * * **.**. .* * * * * * * : *
| ** *.* **.*|
555 ++--------------------------------------------------------------------+
unixbench.time.percent_of_cpu_this_job_got
752 ++--------------------------------------------------------------------+
| O OO OO OO O O O |
750 OO OO OO OO OO OO OO O O O |
748 ++ |
| |
746 ++ |
| |
744 ++ |
| |
742 ++ |
740 ++ |
| |
738 ++ .* *. .* |
**.**.**.**.**.** *.* ** *.**.**.**.**.**.**.**.**.**.**.**. *.**.**
736 ++-------------------------------------------------------------*------+
unixbench.time.involuntary_context_switches
2e+06 ++----------------------------------------------------------------+
|*. *. .* .* *. *. * .* .* *. * .**. * |
1.8e+06 *+ * *** * **.**.***.**.** * **.* * * * **.* * * *.**.**
| |
1.6e+06 ++ |
| |
1.4e+06 ++ |
| |
1.2e+06 ++ |
| |
1e+06 ++ |
| |
800000 ++ OO O OO OOO OO OOO OO OOO OO OO O |
OO O O |
600000 ++----------------------------------------------------------------+
time.involuntary_context_switches
2e+06 ++----------------------------------------------------------------+
|*. *. .* .* *. *. * .* .* *. * .**. * |
1.8e+06 *+ * *** * **.**.***.**.** * **.* * * * **.* * * *.**.**
| |
1.6e+06 ++ |
| |
1.4e+06 ++ |
| |
1.2e+06 ++ |
| |
1e+06 ++ |
| |
800000 ++ OO O OO OOO OO OOO OO OOO OO OO O |
OO O O |
600000 ++----------------------------------------------------------------+
[*] bisect-good sample
[O] bisect-bad sample
To reproduce:
git clone git://git.kernel.org/pub/scm/linux/kernel/git/wfg/lkp-tests.git
cd lkp-tests
bin/lkp install job.yaml # job file is attached in this email
bin/lkp run job.yaml
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 Ye
[-- Attachment #2: job.yaml --]
[-- Type: text/plain, Size: 3319 bytes --]
---
LKP_SERVER: inn
LKP_CGI_PORT: 80
LKP_CIFS_PORT: 139
testcase: unixbench
default-monitors:
wait: activate-monitor
kmsg:
uptime:
iostat:
heartbeat:
vmstat:
numa-numastat:
numa-vmstat:
numa-meminfo:
proc-vmstat:
proc-stat:
interval: 10
meminfo:
slabinfo:
interrupts:
lock_stat:
latency_stats:
softirqs:
bdi_dev_mapping:
diskstats:
nfsstat:
cpuidle:
cpufreq-stats:
turbostat:
pmeter:
sched_debug:
interval: 60
cpufreq_governor:
default-watchdogs:
oom-killer:
watchdog:
commit: ec419c38ecd766e8e41e3232e9f3d473c7d56b3a
model: Nehalem
memory: 4G
nr_cpu: 8
hdd_partitions:
swap_partitions:
rootfs_partition:
netconsole_port: 6671
category: benchmark
nr_task: 100%
unixbench:
test: shell8
queue: bisect
testbox: nhm-white2
tbox_group: nhm-white2
kconfig: x86_64-rhel
enqueue_time: 2016-04-11 13:42:49.026345495 +08:00
compiler: gcc-4.9
rootfs: debian-x86_64-2015-02-07.cgz
id: aff286cfa94db04bc77c6d15807909fa0c84a838
user: lkp
head_commit: 68ac08a99a391b4288edbbabcaff9e754a4f8bbc
base_commit: 9735a22799b9214d17d3c231fe377fc852f042e9
branch: linux-devel/devel-hourly-2016041020
result_root: "/result/unixbench/100%-shell8/nhm-white2/debian-x86_64-2015-02-07.cgz/x86_64-rhel/gcc-4.9/ec419c38ecd766e8e41e3232e9f3d473c7d56b3a/0"
job_file: "/lkp/scheduled/nhm-white2/bisect_unixbench-100%-shell8-debian-x86_64-2015-02-07.cgz-x86_64-rhel-ec419c38ecd766e8e41e3232e9f3d473c7d56b3a-20160411-80304-15v80b6-0.yaml"
max_uptime: 1017.42
initrd: "/osimage/debian/debian-x86_64-2015-02-07.cgz"
bootloader_append:
- root=/dev/ram0
- user=lkp
- job=/lkp/scheduled/nhm-white2/bisect_unixbench-100%-shell8-debian-x86_64-2015-02-07.cgz-x86_64-rhel-ec419c38ecd766e8e41e3232e9f3d473c7d56b3a-20160411-80304-15v80b6-0.yaml
- ARCH=x86_64
- kconfig=x86_64-rhel
- branch=linux-devel/devel-hourly-2016041020
- commit=ec419c38ecd766e8e41e3232e9f3d473c7d56b3a
- BOOT_IMAGE=/pkg/linux/x86_64-rhel/gcc-4.9/ec419c38ecd766e8e41e3232e9f3d473c7d56b3a/vmlinuz-4.6.0-rc2-00086-gec419c3
- max_uptime=1017
- RESULT_ROOT=/result/unixbench/100%-shell8/nhm-white2/debian-x86_64-2015-02-07.cgz/x86_64-rhel/gcc-4.9/ec419c38ecd766e8e41e3232e9f3d473c7d56b3a/0
- LKP_SERVER=inn
- |2-
earlyprintk=ttyS0,115200 systemd.log_level=err
debug apic=debug sysrq_always_enabled rcupdate.rcu_cpu_stall_timeout=100
panic=-1 softlockup_panic=1 nmi_watchdog=panic oops=panic load_ramdisk=2 prompt_ramdisk=0
console=ttyS0,115200 console=tty0 vga=normal
rw
lkp_initrd: "/lkp/lkp/lkp-x86_64.cgz"
modules_initrd: "/pkg/linux/x86_64-rhel/gcc-4.9/ec419c38ecd766e8e41e3232e9f3d473c7d56b3a/modules.cgz"
bm_initrd: "/osimage/deps/debian-x86_64-2015-02-07.cgz/lkp.cgz,/osimage/deps/debian-x86_64-2015-02-07.cgz/run-ipconfig.cgz,/osimage/deps/debian-x86_64-2015-02-07.cgz/turbostat.cgz,/lkp/benchmarks/turbostat.cgz,/lkp/benchmarks/unixbench.cgz"
linux_headers_initrd: "/pkg/linux/x86_64-rhel/gcc-4.9/ec419c38ecd766e8e41e3232e9f3d473c7d56b3a/linux-headers.cgz"
repeat_to: 2
kernel: "/pkg/linux/x86_64-rhel/gcc-4.9/ec419c38ecd766e8e41e3232e9f3d473c7d56b3a/vmlinuz-4.6.0-rc2-00086-gec419c3"
dequeue_time: 2016-04-11 13:52:37.184945077 +08:00
job_state: finished
loadavg: 39.41 19.32 7.60 1/181 19041
start_time: '1460353982'
end_time: '1460354171'
version: "/lkp/lkp/.src-20160411-122710"
[-- Attachment #3: reproduce.ksh --]
[-- Type: text/plain, Size: 38 bytes --]
2016-04-11 13:53:02 ./Run shell8 -c 8
reply other threads:[~2016-04-13 1:59 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20160413015909.GD659@yexl-desktop \
--to=xiaolong.ye@intel.com \
--cc=lkp@lists.01.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.