| 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: <20200629003838.GC5535@shao2-debian>
Date: Mon, 29 Jun 2020 08:38:38 +0800
From: kernel test robot <rong.a.chen@...el.com>
To: Vamshi K Sthambamkadi <vamshi.k.sthambamkadi@...il.com>
Cc: Steven Rostedt <rostedt@...dmis.org>,
Masami Hiramatsu <mhiramat@...nel.org>,
LKML <linux-kernel@...r.kernel.org>, lkp@...ts.01.org
Subject: [tracing/probe] 3aa8fdc37d: will-it-scale.per_process_ops 3.6%
improvement
Greeting,
FYI, we noticed a 3.6% improvement of will-it-scale.per_process_ops due to commit:
commit: 3aa8fdc37d16735e8891035becf25b3857d3efe0 ("tracing/probe: Fix memleak in fetch_op_data operations")
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master
in testcase: will-it-scale
on test machine: 192 threads Intel(R) Xeon(R) Platinum 9242 CPU @ 2.30GHz with 192G memory
with following parameters:
nr_task: 16
mode: process
test: poll2
cpufreq_governor: performance
ucode: 0x5002f01
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-7.6/process/16/debian-x86_64-20191114.cgz/lkp-csl-2ap2/poll2/will-it-scale/0x5002f01
commit:
48a42f5d13 ("trace: Fix typo in allocate_ftrace_ops()'s comment")
3aa8fdc37d ("tracing/probe: Fix memleak in fetch_op_data operations")
48a42f5d13843524 3aa8fdc37d16735e8891035becf
---------------- ---------------------------
%stddev %change %stddev
\ | \
347710 +3.6% 360154 will-it-scale.per_process_ops
5563370 +3.6% 5762481 will-it-scale.workload
70322 ± 6% -20.6% 55836 ± 13% numa-meminfo.node3.SUnreclaim
21205 ± 10% -28.4% 15186 ± 12% numa-meminfo.node3.Shmem
5295 ± 10% -28.4% 3791 ± 12% numa-vmstat.node3.nr_shmem
17580 ± 6% -20.6% 13959 ± 13% numa-vmstat.node3.nr_slab_unreclaimable
4958 ± 62% -82.5% 867.50 ± 74% proc-vmstat.numa_pages_migrated
4958 ± 62% -82.5% 867.50 ± 74% proc-vmstat.pgmigrate_success
2111735 ± 5% -16.0% 1773749 ± 11% sched_debug.cfs_rq:/.spread0.max
3844 ± 3% -5.8% 3623 ± 4% sched_debug.cpu.nr_switches.avg
1764 ± 10% -15.0% 1500 ± 10% sched_debug.cpu.sched_count.avg
518.38 ± 8% -10.1% 466.17 ± 9% sched_debug.cpu.sched_count.min
839.49 ± 15% -18.0% 688.32 ± 11% sched_debug.cpu.sched_goidle.avg
739.42 ± 8% -12.8% 644.87 ± 10% sched_debug.cpu.ttwu_count.avg
396.78 ± 7% -14.6% 338.80 ± 11% sched_debug.cpu.ttwu_local.avg
6507 ± 6% +39.3% 9062 ± 23% sched_debug.cpu.ttwu_local.max
1569 ± 7% -16.3% 1314 ± 8% slabinfo.Acpi-Parse.active_objs
1569 ± 7% -16.3% 1314 ± 8% slabinfo.Acpi-Parse.num_objs
1112 ± 4% -10.1% 1000 ± 2% slabinfo.kmem_cache_node.active_objs
1280 ± 3% -8.8% 1168 ± 2% slabinfo.kmem_cache_node.num_objs
14608 ± 2% -7.3% 13536 ± 3% slabinfo.skbuff_head_cache.active_objs
14608 ± 2% -7.1% 13568 ± 3% slabinfo.skbuff_head_cache.num_objs
1143 ± 3% +9.3% 1250 ± 5% slabinfo.task_group.active_objs
1143 ± 3% +9.3% 1250 ± 5% slabinfo.task_group.num_objs
4344 ± 62% +101.3% 8745 interrupts.CPU14.NMI:Non-maskable_interrupts
4344 ± 62% +101.3% 8745 interrupts.CPU14.PMI:Performance_monitoring_interrupts
42751 ±169% -98.1% 802.75 interrupts.CPU19.CAL:Function_call_interrupts
115.00 ± 5% -27.8% 83.00 ± 32% interrupts.CPU20.NMI:Non-maskable_interrupts
115.00 ± 5% -27.8% 83.00 ± 32% interrupts.CPU20.PMI:Performance_monitoring_interrupts
114.25 ± 6% -33.9% 75.50 ± 24% interrupts.CPU94.NMI:Non-maskable_interrupts
114.25 ± 6% -33.9% 75.50 ± 24% interrupts.CPU94.PMI:Performance_monitoring_interrupts
131.00 ± 15% -38.9% 80.00 ± 26% interrupts.CPU95.NMI:Non-maskable_interrupts
131.00 ± 15% -38.9% 80.00 ± 26% interrupts.CPU95.PMI:Performance_monitoring_interrupts
7636 ± 24% -42.8% 4368 interrupts.CPU97.NMI:Non-maskable_interrupts
7636 ± 24% -42.8% 4368 interrupts.CPU97.PMI:Performance_monitoring_interrupts
2021 ± 14% -15.7% 1703 ± 11% interrupts.RES:Rescheduling_interrupts
1.295e+10 +3.6% 1.341e+10 perf-stat.i.branch-instructions
0.88 -2.6% 0.86 perf-stat.i.cpi
1.48e+10 +4.0% 1.539e+10 perf-stat.i.dTLB-loads
6.859e+09 +3.7% 7.11e+09 perf-stat.i.dTLB-stores
6.382e+10 +3.6% 6.611e+10 perf-stat.i.instructions
180.49 +3.8% 187.27 perf-stat.i.metric.M/sec
0.88 -2.6% 0.85 perf-stat.overall.cpi
1.29e+10 +3.6% 1.337e+10 perf-stat.ps.branch-instructions
1.475e+10 +4.0% 1.533e+10 perf-stat.ps.dTLB-loads
6.835e+09 +3.7% 7.086e+09 perf-stat.ps.dTLB-stores
6.361e+10 +3.6% 6.588e+10 perf-stat.ps.instructions
1.92e+13 +3.7% 1.991e+13 perf-stat.total.instructions
66.73 ± 2% -8.2 58.51 ± 8% perf-profile.calltrace.cycles-pp.poll
65.83 ± 2% -8.1 57.74 ± 8% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.poll
65.67 ± 2% -8.1 57.59 ± 8% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.poll
65.44 ± 2% -8.1 57.37 ± 8% perf-profile.calltrace.cycles-pp.__x64_sys_poll.do_syscall_64.entry_SYSCALL_64_after_hwframe.poll
64.12 ± 2% -7.5 56.59 ± 8% perf-profile.calltrace.cycles-pp.do_sys_poll.__x64_sys_poll.do_syscall_64.entry_SYSCALL_64_after_hwframe.poll
9.68 ± 2% -2.1 7.58 ± 8% perf-profile.calltrace.cycles-pp.__fget_light.do_sys_poll.__x64_sys_poll.do_syscall_64.entry_SYSCALL_64_after_hwframe
31.02 ± 4% +8.0 39.07 ± 11% perf-profile.calltrace.cycles-pp.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64
31.03 ± 4% +8.1 39.08 ± 11% perf-profile.calltrace.cycles-pp.cpu_startup_entry.start_secondary.secondary_startup_64
31.03 ± 4% +8.1 39.08 ± 11% perf-profile.calltrace.cycles-pp.start_secondary.secondary_startup_64
32.73 ± 5% +8.3 40.99 ± 11% perf-profile.calltrace.cycles-pp.secondary_startup_64
26.44 ± 4% +8.6 35.04 ± 15% perf-profile.calltrace.cycles-pp.intel_idle.cpuidle_enter_state.cpuidle_enter.do_idle.cpu_startup_entry
67.09 ± 2% -8.2 58.85 ± 8% perf-profile.children.cycles-pp.poll
65.91 ± 2% -8.1 57.81 ± 8% perf-profile.children.cycles-pp.entry_SYSCALL_64_after_hwframe
65.77 ± 2% -8.1 57.67 ± 8% perf-profile.children.cycles-pp.do_syscall_64
65.45 ± 2% -8.1 57.38 ± 8% perf-profile.children.cycles-pp.__x64_sys_poll
65.21 ± 2% -8.0 57.18 ± 8% perf-profile.children.cycles-pp.do_sys_poll
9.07 ± 2% -1.8 7.22 ± 8% perf-profile.children.cycles-pp.__fget_light
0.43 ± 37% -0.3 0.17 ± 49% perf-profile.children.cycles-pp.tick_irq_enter
0.44 ± 37% -0.3 0.19 ± 46% perf-profile.children.cycles-pp.irq_enter_rcu
0.29 ± 46% -0.1 0.17 ± 14% perf-profile.children.cycles-pp.rebalance_domains
31.03 ± 4% +8.1 39.08 ± 11% perf-profile.children.cycles-pp.start_secondary
32.73 ± 5% +8.3 40.99 ± 11% perf-profile.children.cycles-pp.secondary_startup_64
32.73 ± 5% +8.3 40.99 ± 11% perf-profile.children.cycles-pp.cpu_startup_entry
32.74 ± 5% +8.3 41.00 ± 11% perf-profile.children.cycles-pp.do_idle
26.44 ± 4% +8.6 35.05 ± 15% perf-profile.children.cycles-pp.intel_idle
51.94 ± 2% -5.8 46.15 ± 8% perf-profile.self.cycles-pp.do_sys_poll
8.02 ± 2% -1.7 6.28 ± 8% perf-profile.self.cycles-pp.__fget_light
0.01 ±173% +0.1 0.07 ± 15% perf-profile.self.cycles-pp.tick_nohz_next_event
26.44 ± 4% +8.6 35.04 ± 15% perf-profile.self.cycles-pp.intel_idle
23202 ± 60% +71.0% 39673 ± 3% softirqs.CPU106.SCHED
14488 ± 71% +172.4% 39472 ± 3% softirqs.CPU110.SCHED
42153 -26.7% 30891 ± 37% softirqs.CPU122.SCHED
14224 -39.3% 8628 ± 34% softirqs.CPU124.RCU
42603 ± 2% -13.6% 36813 ± 8% softirqs.CPU126.SCHED
42916 -14.1% 36878 ± 10% softirqs.CPU127.SCHED
42869 -21.9% 33476 ± 5% softirqs.CPU128.SCHED
42422 -14.1% 36450 ± 10% softirqs.CPU129.SCHED
42822 ± 2% -13.4% 37071 ± 9% softirqs.CPU131.SCHED
42635 ± 3% -8.2% 39120 ± 6% softirqs.CPU137.SCHED
24144 ± 52% -83.3% 4020 ± 8% softirqs.CPU14.SCHED
9280 ± 6% -13.0% 8073 ± 10% softirqs.CPU142.RCU
42676 -11.0% 37972 ± 5% softirqs.CPU158.SCHED
36631 +20.0% 43954 ± 11% softirqs.CPU16.SCHED
98634 ± 3% +46.1% 144110 ± 16% softirqs.CPU16.TIMER
99955 ± 2% +43.8% 143735 ± 15% softirqs.CPU17.TIMER
35680 ± 4% +13.6% 40519 ± 2% softirqs.CPU18.SCHED
100588 ± 2% +42.9% 143744 ± 15% softirqs.CPU18.TIMER
9472 ± 8% -16.6% 7901 ± 8% softirqs.CPU180.RCU
157427 ± 2% -22.3% 122314 ± 18% softirqs.CPU180.TIMER
9348 ± 10% -17.8% 7687 ± 6% softirqs.CPU181.RCU
32413 ± 14% +36.2% 44132 ± 13% softirqs.CPU20.SCHED
99195 ± 2% +45.1% 143898 ± 16% softirqs.CPU20.TIMER
33571 ± 15% +128.3% 76633 ± 46% softirqs.CPU21.SCHED
99376 ± 2% +43.5% 142592 ± 15% softirqs.CPU21.TIMER
34301 ± 7% +13.9% 39054 ± 4% softirqs.CPU22.SCHED
99471 ± 2% +44.0% 143262 ± 16% softirqs.CPU22.TIMER
32370 ± 13% +21.5% 39325 ± 3% softirqs.CPU23.SCHED
98963 ± 2% +45.3% 143820 ± 15% softirqs.CPU23.TIMER
7483 ± 6% +66.1% 12427 ± 18% softirqs.CPU26.RCU
42506 -9.5% 38455 ± 6% softirqs.CPU34.SCHED
9668 ± 4% -13.3% 8380 ± 6% softirqs.CPU37.RCU
42093 -8.5% 38518 ± 5% softirqs.CPU40.SCHED
42498 -7.0% 39502 ± 4% softirqs.CPU41.SCHED
42491 -8.8% 38742 ± 5% softirqs.CPU43.SCHED
42658 -8.6% 38969 ± 4% softirqs.CPU44.SCHED
42316 ± 2% -8.2% 38860 ± 4% softirqs.CPU45.SCHED
42694 -7.2% 39613 ± 3% softirqs.CPU46.SCHED
42703 -7.9% 39322 ± 4% softirqs.CPU47.SCHED
42650 -11.7% 37650 ± 2% softirqs.CPU50.SCHED
42700 -7.4% 39519 ± 3% softirqs.CPU52.SCHED
42857 -8.3% 39314 ± 3% softirqs.CPU54.SCHED
42221 ± 2% -9.7% 38138 ± 6% softirqs.CPU58.SCHED
42871 -13.6% 37028 ± 10% softirqs.CPU59.SCHED
42262 ± 2% -8.3% 38774 ± 6% softirqs.CPU60.SCHED
42151 ± 2% -8.1% 38757 ± 5% softirqs.CPU64.SCHED
42748 -11.0% 38045 ± 7% softirqs.CPU65.SCHED
42927 -7.5% 39696 ± 4% softirqs.CPU68.SCHED
42528 -6.6% 39740 ± 4% softirqs.CPU70.SCHED
41989 -8.1% 38576 ± 4% softirqs.CPU74.SCHED
42191 -7.7% 38927 ± 4% softirqs.CPU75.SCHED
42104 -8.4% 38583 ± 4% softirqs.CPU76.SCHED
42143 -8.2% 38666 ± 4% softirqs.CPU77.SCHED
42075 -7.2% 39062 ± 4% softirqs.CPU78.SCHED
42260 -8.5% 38682 ± 4% softirqs.CPU80.SCHED
42505 -8.3% 38993 ± 4% softirqs.CPU81.SCHED
41778 ± 3% -6.3% 39148 ± 4% softirqs.CPU83.SCHED
42004 -7.3% 38956 ± 3% softirqs.CPU85.SCHED
41965 -7.7% 38737 ± 4% softirqs.CPU86.SCHED
42209 -9.3% 38301 ± 5% softirqs.CPU87.SCHED
41937 -13.5% 36293 ± 13% softirqs.CPU93.SCHED
will-it-scale.per_process_ops
385000 +------------------------------------------------------------------+
| O O O O O O O O O |
380000 |-+ O O |
375000 |-+ |
| |
370000 |-+ |
| |
365000 |-+ |
| |
360000 |-+ O O O O O O |
355000 |-+ O O O O O |
| |
350000 |-.+..+.+..+.. .+.. .+.. .+..+..+.+.. .+..+.+..+.. |
|. +. + +. +. +..+..+.+..+..|
345000 +------------------------------------------------------------------+
[*] 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-rc1-00011-g3aa8fdc37d167" of type "text/plain" (206210 bytes)
View attachment "job-script" of type "text/plain" (7558 bytes)
View attachment "job.yaml" of type "text/plain" (4934 bytes)
View attachment "reproduce" of type "text/plain" (337 bytes)
Powered by blists - more mailing lists