| 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: <20160713012444.GA25851@yexl-desktop>
Date: Wed, 13 Jul 2016 09:24:44 +0800
From: kernel test robot <xiaolong.ye@...el.com>
To: Thomas Gleixner <tglx@...utronix.de>
Cc: Ingo Molnar <mingo@...nel.org>,
Arjan van de Ven <arjan@...radead.org>,
Chris Mason <clm@...com>, Eric Dumazet <edumazet@...gle.com>,
Frederic Weisbecker <fweisbec@...il.com>,
George Spelvin <linux@...encehorizons.net>,
Josh Triplett <josh@...htriplett.org>,
Len Brown <lenb@...nel.org>,
Linus Torvalds <torvalds@...ux-foundation.org>,
"Paul E. McKenney" <paulmck@...ux.vnet.ibm.com>,
Peter Zijlstra <peterz@...radead.org>,
Rik van Riel <riel@...hat.com>,
LKML <linux-kernel@...r.kernel.org>,
Stephen Rothwell <sfr@...b.auug.org.au>, lkp@...org
Subject: [lkp] [timers] 500462a9de: netperf.Throughput_Mbps -3.8% regression
FYI, we noticed a -3.8% regression of netperf.Throughput_Mbps due to commit:
commit 500462a9de657f86edaa102f8ab6bff7f7e43fc2 ("timers: Switch to a non-cascading wheel")
https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git master
in testcase: netperf
on test machine: 16 threads Broadwell-DE with 8G memory
with following parameters: cluster=cs-localhost/cpufreq_governor=performance/ip=ipv4/nr_threads=200%/runtime=300s/test=SCTP_STREAM
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.
Details are as below:
-------------------------------------------------------------------------------------------------->
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
=========================================================================================
cluster/compiler/cpufreq_governor/ip/kconfig/nr_threads/rootfs/runtime/tbox_group/test/testcase:
cs-localhost/gcc-4.9/performance/ipv4/x86_64-rhel/200%/debian-x86_64-2015-02-07.cgz/300s/lkp-bdw-de1/SCTP_STREAM/netperf
commit:
b0d6e2dcb2 ("timers: Reduce the CPU index space to 256k")
500462a9de ("timers: Switch to a non-cascading wheel")
b0d6e2dcb284f1f4 500462a9de657f86edaa102f8a
---------------- --------------------------
fail:runs %reproduction fail:runs
| | |
%stddev %change %stddev
\ | \
4.25 ± 0% -3.8% 4.09 ± 0% netperf.Throughput_Mbps
24354 ± 0% -3.9% 23406 ± 0% netperf.time.voluntary_context_switches
442966 ± 15% -18.7% 359983 ± 3% cpuidle.C1-BDW.time
8781285 ± 58% -48.3% 4544291 ± 82% cpuidle.POLL.time
0.06 ± 9% -21.2% 0.04 ± 10% turbostat.CPU%c3
0.20 ± 7% -55.0% 0.09 ± 18% turbostat.Pkg%pc3
1.782e+09 ± 1% -6.4% 1.668e+09 ± 0% perf-stat.L1-dcache-load-misses
7.241e+08 ± 4% +28.2% 9.282e+08 ± 0% perf-stat.LLC-stores
6.285e+08 ± 0% +10.4% 6.937e+08 ± 0% perf-stat.branch-load-misses
5.831e+08 ± 2% +12.2% 6.542e+08 ± 0% perf-stat.branch-misses
4.192e+09 ± 0% +2.5% 4.299e+09 ± 0% perf-stat.cache-references
340227 ± 0% -1.7% 334278 ± 0% perf-stat.context-switches
60151 ± 0% -1.9% 59012 ± 0% perf-stat.cpu-migrations
4.863e+11 ± 1% +3.3% 5.022e+11 ± 1% perf-stat.ref-cycles
15.21 ± 14% +38.9% 21.13 ± 18% sched_debug.cfs_rq:/.runnable_load_avg.avg
-4768 ±-21% +27.5% -6078 ±-22% sched_debug.cfs_rq:/.spread0.avg
104776 ± 3% -12.8% 91330 ± 13% sched_debug.cpu.avg_idle.stddev
195.24 ± 4% -19.7% 156.75 ± 16% sched_debug.cpu.clock.stddev
195.24 ± 4% -19.7% 156.75 ± 16% sched_debug.cpu.clock_task.stddev
12.54 ± 17% +26.0% 15.80 ± 4% sched_debug.cpu.cpu_load[0].avg
3.67 ± 51% +116.7% 7.94 ± 24% sched_debug.cpu.cpu_load[1].min
6.04 ± 16% +56.3% 9.44 ± 5% sched_debug.cpu.cpu_load[2].min
5.33 ± 20% +44.8% 7.72 ± 11% sched_debug.cpu.cpu_load[3].min
1765 ± 8% +14.2% 2016 ± 3% sched_debug.cpu.nr_load_updates.min
2247 ± 8% +20.2% 2701 ± 8% sched_debug.cpu.nr_switches.min
9.71 ± 20% +160.9% 25.33 ± 47% sched_debug.cpu.nr_uninterruptible.max
-13.92 ±-24% +77.6% -24.72 ± -8% sched_debug.cpu.nr_uninterruptible.min
5.98 ± 14% +112.0% 12.68 ± 33% sched_debug.cpu.nr_uninterruptible.stddev
2.97 ± 12% -31.4% 2.04 ± 32% perf-profile.cycles-pp.__const_udelay.wait_for_xmitr.serial8250_console_putchar.uart_console_write.serial8250_console_write
1.63 ± 21% -29.3% 1.15 ± 31% perf-profile.cycles-pp.__netif_receive_skb.process_backlog.net_rx_action.__do_softirq.irq_exit
1.63 ± 21% -29.3% 1.15 ± 31% perf-profile.cycles-pp.__netif_receive_skb_core.__netif_receive_skb.process_backlog.net_rx_action.__do_softirq
7.85 ± 10% -38.9% 4.80 ± 24% perf-profile.cycles-pp.call_console_drivers.constprop.23.console_unlock.vprintk_emit.vprintk_default.printk
7.85 ± 10% -38.9% 4.80 ± 24% perf-profile.cycles-pp.console_unlock.vprintk_emit.vprintk_default.printk.perf_duration_warn
2.97 ± 12% -31.4% 2.04 ± 32% perf-profile.cycles-pp.delay_tsc.__const_udelay.wait_for_xmitr.serial8250_console_putchar.uart_console_write
0.00 ± -1% +Inf% 1.28 ± 32% perf-profile.cycles-pp.get_next_timer_interrupt.tick_nohz_stop_sched_tick.__tick_nohz_idle_enter.tick_nohz_irq_exit.irq_exit
4.46 ± 11% -46.9% 2.37 ± 15% perf-profile.cycles-pp.io_serial_in.wait_for_xmitr.serial8250_console_putchar.uart_console_write.serial8250_console_write
1.61 ± 20% -29.7% 1.13 ± 29% perf-profile.cycles-pp.ip_local_deliver.ip_rcv_finish.ip_rcv.__netif_receive_skb_core.__netif_receive_skb
1.61 ± 20% -29.7% 1.13 ± 29% perf-profile.cycles-pp.ip_local_deliver_finish.ip_local_deliver.ip_rcv_finish.ip_rcv.__netif_receive_skb_core
1.61 ± 20% -28.5% 1.15 ± 31% perf-profile.cycles-pp.ip_rcv.__netif_receive_skb_core.__netif_receive_skb.process_backlog.net_rx_action
1.61 ± 20% -29.7% 1.13 ± 29% perf-profile.cycles-pp.ip_rcv_finish.ip_rcv.__netif_receive_skb_core.__netif_receive_skb.process_backlog
1.63 ± 21% -29.3% 1.15 ± 31% perf-profile.cycles-pp.net_rx_action.__do_softirq.irq_exit.smp_apic_timer_interrupt.apic_timer_interrupt
7.85 ± 10% -38.9% 4.80 ± 24% perf-profile.cycles-pp.perf_duration_warn.irq_work_run_list.irq_work_run.smp_irq_work_interrupt.irq_work_interrupt
1.56 ± 14% -23.2% 1.20 ± 15% perf-profile.cycles-pp.perf_mux_hrtimer_handler.__hrtimer_run_queues.hrtimer_interrupt.local_apic_timer_interrupt.smp_apic_timer_interrupt
7.85 ± 10% -38.9% 4.80 ± 24% perf-profile.cycles-pp.printk.perf_duration_warn.irq_work_run_list.irq_work_run.smp_irq_work_interrupt
1.63 ± 21% -29.3% 1.15 ± 31% perf-profile.cycles-pp.process_backlog.net_rx_action.__do_softirq.irq_exit.smp_apic_timer_interrupt
1.54 ± 19% -26.9% 1.13 ± 29% perf-profile.cycles-pp.sctp_rcv.ip_local_deliver_finish.ip_local_deliver.ip_rcv_finish.ip_rcv
7.57 ± 10% -39.0% 4.62 ± 24% perf-profile.cycles-pp.serial8250_console_putchar.uart_console_write.serial8250_console_write.univ8250_console_write.call_console_drivers.constprop.23
7.85 ± 10% -38.9% 4.80 ± 24% perf-profile.cycles-pp.serial8250_console_write.univ8250_console_write.call_console_drivers.constprop.23.console_unlock.vprintk_emit
7.57 ± 10% -39.0% 4.62 ± 24% perf-profile.cycles-pp.uart_console_write.serial8250_console_write.univ8250_console_write.call_console_drivers.constprop.23.console_unlock
7.85 ± 10% -38.9% 4.80 ± 24% perf-profile.cycles-pp.univ8250_console_write.call_console_drivers.constprop.23.console_unlock.vprintk_emit.vprintk_default
7.85 ± 10% -38.9% 4.80 ± 24% perf-profile.cycles-pp.vprintk_default.printk.perf_duration_warn.irq_work_run_list.irq_work_run
7.85 ± 10% -38.9% 4.80 ± 24% perf-profile.cycles-pp.vprintk_emit.vprintk_default.printk.perf_duration_warn.irq_work_run_list
7.51 ± 9% -39.4% 4.55 ± 23% perf-profile.cycles-pp.wait_for_xmitr.serial8250_console_putchar.uart_console_write.serial8250_console_write.univ8250_console_write
netperf.Throughput_Mbps
4.5 ++--------------------------------------------------------------------+
O.O.O.*.OO *.O.* O O.*.O.O*.O.O.O.O.O.O.O.O.OO.*.O.O.O.*.*.*.**.*.*.*.*
4 ++ : : : : |
3.5 ++ : : : : |
| : : : : |
3 ++ : : : : |
2.5 ++ : : : : |
| : : : : |
2 ++ :: : : |
1.5 ++ :: : : |
| :: : : |
1 ++ : : |
0.5 ++ : : |
| : : |
0 ++----O--*-O---O-*---O----O--------------------O----------------------+
netperf.time.voluntary_context_switches
25000 *+*-*-**-*-*-*-*-**-*-*-*-*-**-*-*-*-*-**-*-*-*-*-**-*-*-*-*-**-*-*-*
O O O OO O O O O OO O O O O OO O O O O OO O O O OO |
| |
20000 ++ |
| |
| |
15000 ++ |
| |
10000 ++ |
| |
| |
5000 ++ |
| |
| |
0 ++--------------------------------------------O---------------------+
[*] bisect-good sample
[O] bisect-bad sample
Thanks,
Xiaolong
View attachment "config-4.7.0-rc6-00014-g500462a" of type "text/plain" (150949 bytes)
View attachment "job.yaml" of type "text/plain" (3703 bytes)
View attachment "reproduce" of type "text/plain" (3896 bytes)
Powered by blists - more mailing lists