| 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: <20160324015135.GA23869@yexl-desktop>
Date: Thu, 24 Mar 2016 09:51:36 +0800
From: kernel test robot <xiaolong.ye@...el.com>
To: Christoph Lameter <cl@...ux.com>
Cc: LKML <linux-kernel@...r.kernel.org>, lkp@...org
Subject: [lkp] [vmstat] 0eb77e9880: +2.7% pigz.throughput
FYI, we noticed that pigz.throughput +2.7% improvement on
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
commit 0eb77e9880321915322d42913c3b53241739c8aa ("vmstat: make vmstat_updater deferrable again and shut down on idle")
=========================================================================================
blocksize/compiler/cpufreq_governor/kconfig/nr_threads/rootfs/tbox_group/testcase:
128K/gcc-4.9/performance/x86_64-rhel/100%/debian-x86_64-2015-02-07.cgz/brickland1/pigz
commit:
686739f6af5e8d5687ffebbf1193ff066aada6d9
0eb77e9880321915322d42913c3b53241739c8aa
686739f6af5e8d56 0eb77e9880321915322d42913c
---------------- --------------------------
%stddev %change %stddev
\ | \
1.989e+08 ± 0% +2.7% 2.043e+08 ± 0% pigz.throughput
19017 ± 2% -17.3% 15725 ± 4% pigz.time.involuntary_context_switches
1929 ± 0% +6.2% 2049 ± 0% pigz.time.percent_of_cpu_this_job_got
262.60 ± 0% +6.6% 279.90 ± 0% pigz.time.system_time
5532 ± 0% +6.2% 5873 ± 0% pigz.time.user_time
15031795 ± 0% +2.7% 15439431 ± 0% pigz.time.voluntary_context_switches
3978 ± 96% -99.8% 8.25 ± 33% numa-numastat.node3.other_node
19017 ± 2% -17.3% 15725 ± 4% time.involuntary_context_switches
1537 ± 1% -10.2% 1380 ± 1% proc-vmstat.nr_alloc_batch
768.25 ± 33% -100.0% 0.33 ±141% proc-vmstat.pgmigrate_fail
101204 ± 0% +2.4% 103648 ± 0% vmstat.system.cs
33988 ± 1% +4.6% 35542 ± 1% vmstat.system.in
16.57 ± 0% +6.3% 17.61 ± 0% turbostat.%Busy
346.00 ± 0% +3.0% 356.50 ± 0% turbostat.Avg_MHz
0.92 ± 1% -62.6% 0.35 ± 4% turbostat.CPU%c3
665.14 ± 0% +6.5% 708.22 ± 0% turbostat.CorWatt
677.67 ± 0% +6.4% 720.77 ± 0% turbostat.PkgWatt
401974 ± 10% +28.1% 514828 ± 10% numa-meminfo.node0.MemUsed
15203 ± 6% -11.9% 13388 ± 3% numa-meminfo.node2.SReclaimable
38174 ± 18% +25.5% 47898 ± 6% numa-meminfo.node3.Active(anon)
332312 ± 3% +14.4% 380028 ± 1% numa-meminfo.node3.MemUsed
15391 ± 19% +29.6% 19949 ± 6% numa-meminfo.node3.SReclaimable
51955 ± 7% +13.4% 58906 ± 2% numa-meminfo.node3.Slab
21485 ± 19% +23.2% 26468 ± 0% numa-vmstat.node1.numa_other
411.50 ± 4% -15.6% 347.50 ± 2% numa-vmstat.node2.nr_alloc_batch
3800 ± 6% -11.9% 3346 ± 3% numa-vmstat.node2.nr_slab_reclaimable
9542 ± 18% +25.4% 11969 ± 6% numa-vmstat.node3.nr_active_anon
403.25 ± 2% -12.6% 352.50 ± 3% numa-vmstat.node3.nr_alloc_batch
3847 ± 19% +29.6% 4986 ± 6% numa-vmstat.node3.nr_slab_reclaimable
428.00 ± 8% +38.0% 590.75 ± 21% slabinfo.blkdev_requests.active_objs
428.00 ± 8% +38.0% 590.75 ± 21% slabinfo.blkdev_requests.num_objs
26723 ± 2% -7.0% 24850 ± 3% slabinfo.kmalloc-256.active_objs
27745 ± 2% -7.0% 25801 ± 3% slabinfo.kmalloc-256.num_objs
1165 ± 5% +16.2% 1354 ± 3% slabinfo.mnt_cache.active_objs
1165 ± 5% +16.2% 1354 ± 3% slabinfo.mnt_cache.num_objs
5.87e+08 ± 2% +19.4% 7.012e+08 ± 7% cpuidle.C1-IVT-4S.time
651990 ± 0% -10.3% 584857 ± 0% cpuidle.C1-IVT-4S.usage
64334504 ± 2% -60.4% 25444312 ± 8% cpuidle.C1E-IVT-4S.time
121454 ± 1% -59.8% 48868 ± 7% cpuidle.C1E-IVT-4S.usage
2.634e+08 ± 1% -61.0% 1.028e+08 ± 5% cpuidle.C3-IVT-4S.time
451711 ± 1% -59.6% 182479 ± 7% cpuidle.C3-IVT-4S.usage
30487 ± 6% +41.4% 43112 ± 5% cpuidle.POLL.usage
1.15 ± 7% -83.2% 0.19 ± 10% perf-profile.cycles-pp.__remove_hrtimer.hrtimer_try_to_cancel.hrtimer_cancel.tick_nohz_restart.tick_nohz_idle_exit
2.65 ± 5% +11.7% 2.96 ± 2% perf-profile.cycles-pp.__schedule.schedule.futex_wait_queue_me.futex_wait.do_futex
3.67 ± 3% -25.4% 2.73 ± 8% perf-profile.cycles-pp.__tick_nohz_idle_enter.tick_nohz_idle_enter.cpu_startup_entry.start_secondary
3.70 ± 5% +8.9% 4.02 ± 2% perf-profile.cycles-pp.futex_wait.do_futex.sys_futex.entry_SYSCALL_64_fastpath
1.39 ± 6% -69.4% 0.42 ± 5% perf-profile.cycles-pp.hrtimer_cancel.tick_nohz_restart.tick_nohz_idle_exit.cpu_startup_entry.start_secondary
2.16 ± 4% -39.1% 1.31 ± 9% perf-profile.cycles-pp.hrtimer_start_range_ns.tick_nohz_stop_sched_tick.__tick_nohz_idle_enter.tick_nohz_idle_enter.cpu_startup_entry
1.36 ± 7% -72.1% 0.38 ± 8% perf-profile.cycles-pp.hrtimer_try_to_cancel.hrtimer_cancel.tick_nohz_restart.tick_nohz_idle_exit.cpu_startup_entry
0.88 ± 8% -15.3% 0.74 ± 9% perf-profile.cycles-pp.ktime_get.clockevents_program_event.tick_program_event.__remove_hrtimer.hrtimer_start_range_ns
0.97 ± 6% -10.1% 0.87 ± 6% perf-profile.cycles-pp.ktime_get.sched_clock_tick.sched_clock_idle_wakeup_event.tick_nohz_stop_idle.tick_nohz_idle_exit
0.65 ± 6% +44.1% 0.94 ± 8% perf-profile.cycles-pp.pick_next_task_fair.__schedule.schedule.futex_wait_queue_me.futex_wait
0.00 ± -1% +Inf% 2.02 ± 5% perf-profile.cycles-pp.quiet_vmstat.cpu_startup_entry.start_secondary
1.62 ± 6% -51.6% 0.78 ± 9% perf-profile.cycles-pp.read_hpet.ktime_get.clockevents_program_event.tick_program_event.__remove_hrtimer
1.67 ± 4% -49.9% 0.84 ± 6% perf-profile.cycles-pp.read_hpet.ktime_get.clockevents_program_event.tick_program_event.hrtimer_start_range_ns
0.00 ± -1% +Inf% 1.71 ± 5% perf-profile.cycles-pp.refresh_cpu_vm_stats.quiet_vmstat.cpu_startup_entry.start_secondary
2.79 ± 5% +10.1% 3.07 ± 2% perf-profile.cycles-pp.schedule.futex_wait_queue_me.futex_wait.do_futex.sys_futex
3.78 ± 3% -24.6% 2.85 ± 8% perf-profile.cycles-pp.tick_nohz_idle_enter.cpu_startup_entry.start_secondary
5.26 ± 4% -20.9% 4.16 ± 4% perf-profile.cycles-pp.tick_nohz_idle_exit.cpu_startup_entry.start_secondary
2.86 ± 6% -35.2% 1.85 ± 4% perf-profile.cycles-pp.tick_nohz_restart.tick_nohz_idle_exit.cpu_startup_entry.start_secondary
2.56 ± 3% -33.2% 1.71 ± 9% perf-profile.cycles-pp.tick_nohz_stop_sched_tick.__tick_nohz_idle_enter.tick_nohz_idle_enter.cpu_startup_entry.start_secondary
0.94 ± 4% -94.7% 0.05 ± 58% perf-profile.cycles-pp.tick_program_event.__remove_hrtimer.hrtimer_try_to_cancel.hrtimer_cancel.tick_nohz_restart
brickland1: Brickland Ivy Bridge-EX
Memory: 128G
pigz.time.voluntary_context_switches
1.6e+07 ++----------------------------------------------------------------+
| O O O O O O O O O O O O O O O |
1.5e+07 O+*.*.O.O.*.*.*..*.O.*.O.O.O.O.*.*.*.*.*.*.*.*.*..*.*.*.*.*.*.*.*.*
| : |
| : |
1.4e+07 ++ |
|: |
1.3e+07 ++ |
|: |
1.2e+07 ++ |
|: |
| |
1.1e+07 ++ |
* |
1e+07 ++----------------------------------------------------------------+
cpuidle.C1E-IVT-4S.time
9e+07 ++------------------------------------------------------------------+
| * |
8e+07 ++ :: |
| : : |
7e+07 ++ : *. .*. |
| .*.. .*.*.*.*. .*.. .*.*. .*. .*. : *.*.*. *.*. .*
6e+07 ++*.* * *.* *.* *.* *. *.* * |
| : |
5e+07 ++ |
|: |
4e+07 ++ |
|: O |
3e+07 O+ O O O O O O O O O |
* O O O O O O O O O O O |
2e+07 ++---------------------------------------------O--------------------+
cpuidle.C1E-IVT-4S.usage
130000 ++--------*-------------------------------------------*------------+
120000 ++ .*. + *.*.*.*.*.*. .*.*. .*.*.*.*.*. .*.*.* *.*.*.*.*.*
| *.* * * *..* * |
110000 ++: |
100000 ++: |
|: |
90000 ++ |
80000 ++ |
70000 ++ |
|: |
60000 O+O O O O |
50000 ++ O O O O O O O O O O O O O |
| O O O O O |
40000 *+ |
30000 ++-----------------------------------------------------------------+
cpuidle.C3-IVT-4S.time
2.8e+08 ++------------*---------------------------------------------------+
2.6e+08 ++ .*.*. .*.* *.*.*.*. .*.*. .*.*.*.*.*. .*..*.*.*.*.*.*.*.*.*
| * * * *.* * |
2.4e+08 ++: |
2.2e+08 ++: |
|: |
2e+08 ++ |
1.8e+08 ++ |
1.6e+08 ++ |
|: |
1.4e+08 ++ |
1.2e+08 ++O O O |
O O O O O O O O O O O O O O |
1e+08 ++ O O O O O O |
8e+07 ++----------------------------------------------------------------+
cpuidle.C3-IVT-4S.usage
500000 ++-----------------------------------------------------------------+
| *.. .*.*.. |
450000 ++*.*.*. + *.*.*.*.*.*. .*.*. .*.*.*.*.*.*.*.* *.*.*.*.*.*
| : * * *..* |
400000 ++: |
| : |
350000 ++ |
|: |
300000 ++ |
|: |
250000 ++ |
|: O |
200000 O+O O O O O O O O O O O O O |
| O O O O O O O O |
150000 *+-----------------------------------------------------------------+
turbostat.CPU_c3
1 ++--------------------------------------------------------------------+
| .*. .*.*.*..*.*. .*. .*.. .*. .*..*.*.*.*.*. .*. |
0.9 ++*.*. * * *.*..*.*.*.* * *.* *. *.*
| : |
0.8 ++: |
|: |
0.7 ++ |
|: |
0.6 ++ |
|: |
0.5 ++ |
| |
0.4 *+O O O |
O O O O O O O O O O O O O O O O O O O |
0.3 ++----------------------------------------------O---------------------+
[*] 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
View attachment "job.yaml" of type "text/plain" (3430 bytes)
View attachment "reproduce" of type "text/plain" (11291 bytes)
Powered by blists - more mailing lists