[LKP] [NFS] a08a8cd375d: +1.5% will-it-scale.per_process_ops
From: Huang Ying
Date: Mon Mar 30 2015 - 02:29:56 EST
FYI, we noticed the below changes on
git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
commit a08a8cd375db9769588257e7782f6b6b68561b88 ("NFS: Add attribute update barriers to NFS writebacks")
testbox/testcase/testparams: ivb42/will-it-scale/performance-pthread_mutex2
f5062003465c20cf a08a8cd375db9769588257e778
---------------- --------------------------
%stddev %change %stddev
\ | \
2.012e+08 Â 0% +1.5% 2.042e+08 Â 0% will-it-scale.per_process_ops
3.81 Â 0% +1.6% 3.88 Â 0% turbostat.RAMWatt
387 Â 7% +13.7% 441 Â 6% numa-vmstat.node0.nr_page_table_pages
5639 Â 15% -15.1% 4788 Â 0% meminfo.AnonHugePages
1552 Â 7% +13.7% 1765 Â 6% numa-meminfo.node0.PageTables
29.40 Â 21% +57.6% 46.32 Â 3% perf-profile.cpu-cycles.start_secondary
10.72 Â 30% -100.0% 0.00 Â 0% perf-profile.cpu-cycles.rest_init.start_kernel.x86_64_start_reservations.x86_64_start_kernel
10.72 Â 30% -100.0% 0.00 Â 0% perf-profile.cpu-cycles.cpu_startup_entry.rest_init.start_kernel.x86_64_start_reservations.x86_64_start_kernel
1.23 Â 17% -29.1% 0.87 Â 19% perf-profile.cpu-cycles.cmd_record._start.main.__libc_start_main
1.27 Â 17% -29.4% 0.89 Â 16% perf-profile.cpu-cycles.main.__libc_start_main
1.27 Â 17% -31.2% 0.87 Â 19% perf-profile.cpu-cycles._start.main.__libc_start_main
25.76 Â 19% +66.1% 42.79 Â 5% perf-profile.cpu-cycles.cpuidle_enter.cpu_startup_entry.start_secondary
1.39 Â 16% -30.3% 0.97 Â 11% perf-profile.cpu-cycles.__libc_start_main
1.62 Â 20% +52.2% 2.47 Â 18% perf-profile.cpu-cycles.ktime_get_update_offsets_now.hrtimer_interrupt.local_apic_timer_interrupt.smp_apic_timer_interrupt.apic_timer_interrupt
28.79 Â 21% +60.5% 46.22 Â 3% perf-profile.cpu-cycles.cpu_startup_entry.start_secondary
0.00 Â 0% +Inf% 0.81 Â 36% perf-profile.cpu-cycles._raw_spin_unlock_irqrestore.update_blocked_averages.rebalance_domains.run_rebalance_domains.__do_softirq
10.72 Â 30% -100.0% 0.00 Â 0% perf-profile.cpu-cycles.x86_64_start_kernel
18.38 Â 16% +43.4% 26.35 Â 3% perf-profile.cpu-cycles.intel_idle.cpuidle_enter_state.cpuidle_enter.cpu_startup_entry.start_secondary
1.64 Â 30% -59.1% 0.67 Â 46% perf-profile.cpu-cycles.filemap_map_pages.handle_pte_fault.handle_mm_fault.__do_page_fault.do_page_fault
18.60 Â 17% +42.4% 26.49 Â 4% perf-profile.cpu-cycles.cpuidle_enter_state.cpuidle_enter.cpu_startup_entry.start_secondary
10.42 Â 30% -100.0% 0.00 Â 0% perf-profile.cpu-cycles.cpuidle_enter.cpu_startup_entry.rest_init.start_kernel.x86_64_start_reservations
9.45 Â 34% -100.0% 0.00 Â 0% perf-profile.cpu-cycles.smp_reschedule_interrupt.reschedule_interrupt.cpuidle_enter.cpu_startup_entry.rest_init
10.72 Â 30% -100.0% 0.00 Â 0% perf-profile.cpu-cycles.x86_64_start_reservations.x86_64_start_kernel
13.90 Â 7% -12.0% 12.22 Â 4% perf-profile.cpu-cycles.update_process_times.tick_sched_handle.tick_sched_timer.__run_hrtimer.hrtimer_interrupt
13.92 Â 2% -22.7% 10.77 Â 18% perf-profile.cpu-cycles.__do_softirq.irq_exit.scheduler_ipi.smp_reschedule_interrupt.reschedule_interrupt
10.72 Â 30% -100.0% 0.00 Â 0% perf-profile.cpu-cycles.start_kernel.x86_64_start_reservations.x86_64_start_kernel
9.45 Â 34% -100.0% 0.00 Â 0% perf-profile.cpu-cycles.reschedule_interrupt.cpuidle_enter.cpu_startup_entry.rest_init.start_kernel
1438772 Â 8% +21.2% 1743708 Â 2% sched_debug.cfs_rq[0]:/.min_vruntime
25049 Â 3% +26.4% 31672 Â 1% sched_debug.cfs_rq[0]:/.avg->runnable_avg_sum
17 Â 12% +121.4% 38 Â 19% sched_debug.cfs_rq[0]:/.runnable_load_avg
545 Â 3% +26.4% 689 Â 2% sched_debug.cfs_rq[0]:/.tg_runnable_contrib
48 Â 47% +147.7% 120 Â 30% sched_debug.cfs_rq[0]:/.tg_load_contrib
53474 Â 7% +63.0% 87186 Â 2% sched_debug.cfs_rq[0]:/.exec_clock
61 Â 4% -43.0% 34 Â 9% sched_debug.cfs_rq[24]:/.runnable_load_avg
2241502 Â 4% -17.8% 1843317 Â 1% sched_debug.cfs_rq[24]:/.min_vruntime
802666 Â 24% -87.6% 99533 Â 40% sched_debug.cfs_rq[24]:/.spread0
99 Â 21% +51.9% 151 Â 31% sched_debug.cfs_rq[24]:/.tg_load_contrib
62 Â 5% -44.6% 34 Â 10% sched_debug.cfs_rq[24]:/.load
855 Â 2% -17.4% 706 Â 2% sched_debug.cfs_rq[24]:/.tg_runnable_contrib
39223 Â 2% -17.4% 32411 Â 2% sched_debug.cfs_rq[24]:/.avg->runnable_avg_sum
129807 Â 3% -26.0% 96066 Â 2% sched_debug.cfs_rq[24]:/.exec_clock
1 Â 0% +300.0% 4 Â 50% sched_debug.cfs_rq[39]:/.nr_spread_over
3 Â 33% -66.7% 1 Â 0% sched_debug.cfs_rq[3]:/.nr_spread_over
17 Â 12% +122.9% 39 Â 18% sched_debug.cpu#0.cpu_load[0]
56862 Â 7% +58.5% 90118 Â 2% sched_debug.cpu#0.nr_load_updates
3783 Â 6% +13.9% 4310 Â 0% sched_debug.cpu#0.curr->pid
16708 Â 18% -18.4% 13637 Â 3% sched_debug.cpu#0.sched_count
3907 Â 10% -24.6% 2945 Â 11% sched_debug.cpu#0.ttwu_local
5540 Â 8% -15.4% 4688 Â 6% sched_debug.cpu#0.ttwu_count
19 Â 17% +182.1% 55 Â 33% sched_debug.cpu#0.cpu_load[2]
18 Â 14% +159.5% 48 Â 29% sched_debug.cpu#0.cpu_load[1]
22 Â 27% +181.8% 62 Â 35% sched_debug.cpu#0.cpu_load[3]
24 Â 35% +175.8% 68 Â 35% sched_debug.cpu#0.cpu_load[4]
2253 Â 11% -23.0% 1734 Â 9% sched_debug.cpu#0.sched_goidle
3720 Â 8% -16.5% 3108 Â 13% sched_debug.cpu#13.curr->pid
6862 Â 10% -32.1% 4659 Â 15% sched_debug.cpu#15.nr_switches
3135 Â 11% -36.6% 1987 Â 15% sched_debug.cpu#15.sched_goidle
10 Â 4% +14.6% 11 Â 7% sched_debug.cpu#22.cpu_load[4]
84 Â 16% -55.2% 38 Â 21% sched_debug.cpu#24.cpu_load[3]
94 Â 18% -55.9% 41 Â 22% sched_debug.cpu#24.cpu_load[4]
136876 Â 2% -20.1% 109402 Â 2% sched_debug.cpu#24.nr_load_updates
61 Â 4% -43.0% 34 Â 9% sched_debug.cpu#24.cpu_load[0]
62 Â 5% -44.6% 34 Â 10% sched_debug.cpu#24.load
4589 Â 3% -10.6% 4101 Â 0% sched_debug.cpu#24.curr->pid
65 Â 6% -45.8% 35 Â 15% sched_debug.cpu#24.cpu_load[1]
75 Â 13% -51.2% 37 Â 20% sched_debug.cpu#24.cpu_load[2]
411 Â 29% +82.3% 749 Â 46% sched_debug.cpu#26.ttwu_count
207 Â 26% +110.2% 436 Â 23% sched_debug.cpu#32.ttwu_count
1027 Â 37% -52.7% 486 Â 2% sched_debug.cpu#37.ttwu_local
1550 Â 23% -41.5% 906 Â 39% sched_debug.cpu#42.ttwu_count
773 Â 21% +152.1% 1950 Â 46% sched_debug.cpu#45.ttwu_count
1664 Â 10% +109.2% 3482 Â 27% sched_debug.cpu#45.nr_switches
386 Â 24% +199.9% 1159 Â 40% sched_debug.cpu#45.ttwu_local
10 Â 8% +59.5% 16 Â 36% sched_debug.cpu#45.cpu_load[1]
699 Â 10% +59.6% 1116 Â 30% sched_debug.cpu#45.sched_goidle
1678 Â 10% +158.7% 4341 Â 48% sched_debug.cpu#45.sched_count
2954 Â 4% +15.7% 3417 Â 6% sched_debug.cpu#47.curr->pid
ivb42: Ivytown Ivy Bridge-EP
Memory: 64G
will-it-scale.per_process_ops
2.045e+08 ++------------------------O-------------------------------------+
2.04e+08 ++ O O O O O
| O O O |
2.035e+08 O+ O O O O O O O |
2.03e+08 ++ O O |
| O O O O |
2.025e+08 ++ |
2.02e+08 ++ |
2.015e+08 ++ |
| *..*..*..* |
2.01e+08 ++ *.. .*.. .. |
2.005e+08 ++ *..*..*..*.. : .*. * |
|.. : *. |
2e+08 *+ * |
1.995e+08 ++--------------------------------------------------------------+
[*] bisect-good sample
[O] bisect-bad sample
To reproduce:
apt-get install ruby
git clone git://git.kernel.org/pub/scm/linux/kernel/git/wfg/lkp-tests.git
cd lkp-tests
bin/setup-local job.yaml # the job file attached in this email
bin/run-local 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,
Ying Huang
---
testcase: will-it-scale
default-monitors:
wait: pre-test
uptime:
iostat:
vmstat:
numa-numastat:
numa-vmstat:
numa-meminfo:
proc-vmstat:
proc-stat:
meminfo:
slabinfo:
interrupts:
lock_stat:
latency_stats:
softirqs:
bdi_dev_mapping:
diskstats:
nfsstat:
cpuidle:
cpufreq-stats:
turbostat:
pmeter:
sched_debug:
interval: 10
default_watchdogs:
watch-oom:
watchdog:
cpufreq_governor: performance
commit: 09dd29592b9de9921d5451d18735ddf7b7c41c89
model: Ivytown Ivy Bridge-EP
nr_cpu: 48
memory: 64G
rootfs: debian-x86_64-2015-02-07.cgz
perf-profile:
freq: 800
will-it-scale:
test: pthread_mutex2
testbox: ivb42
tbox_group: ivb42
kconfig: x86_64-rhel
enqueue_time: 2015-03-08 22:23:51.602528045 +08:00
head_commit: 09dd29592b9de9921d5451d18735ddf7b7c41c89
base_commit: 9eccca0843205f87c00404b663188b88eb248051
branch: linux-devel/devel-hourly-2015030909
kernel: "/kernel/x86_64-rhel/09dd29592b9de9921d5451d18735ddf7b7c41c89/vmlinuz-4.0.0-rc3-01030-g09dd295"
user: lkp
queue: cyclic
result_root: "/result/ivb42/will-it-scale/performance-pthread_mutex2/debian-x86_64-2015-02-07.cgz/x86_64-rhel/09dd29592b9de9921d5451d18735ddf7b7c41c89/0"
job_file: "/lkp/scheduled/ivb42/cyclic_will-it-scale-performance-pthread_mutex2-debian-x86_64.cgz-x86_64-rhel-HEAD-09dd29592b9de9921d5451d18735ddf7b7c41c89-0-20150308-40925-iwbdjc.yaml"
dequeue_time: 2015-03-09 13:24:54.233793554 +08:00
job_state: finished
loadavg: 25.26 17.11 7.20 1/422 10086
start_time: '1425878742'
end_time: '1425879052'
version: "/lkp/lkp/.src-20150309-125440"
echo performance > /sys/devices/system/cpu/cpu0/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu1/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu10/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu11/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu12/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu13/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu14/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu15/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu16/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu17/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu18/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu19/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu2/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu20/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu21/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu22/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu23/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu24/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu25/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu26/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu27/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu28/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu29/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu3/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu30/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu31/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu32/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu33/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu34/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu35/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu36/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu37/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu38/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu39/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu4/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu40/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu41/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu42/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu43/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu44/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu45/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu46/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu47/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu5/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu6/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu7/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu8/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu9/cpufreq/scaling_governor
./runtest.py pthread_mutex2 25 both 1 12 24 36 48
_______________________________________________
LKP mailing list
LKP@xxxxxxxxxxxxxxx