Greeting,
FYI, we noticed a 1.3% improvement of pigz.throughput due to commit:
commit: ffb324e6f874121f7dce5bdae5e05d02baae7269 ("tty: vt: always invoke vc->vc_sw->con_resize callback")
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master
in testcase: pigz
on test machine: 272 threads 2 sockets Intel(R) Xeon Phi(TM) CPU 7255 @ 1.10GHz with 112G memory
with following parameters:
nr_threads: 25%
blocksize: 128K
cpufreq_governor: performance
ucode: 0x11
test-description: Pigz, which stands for parallel implementation of gzip, is a fully functional replacement for gzip that exploits multiple processors and multiple cores to the hilt when compressing data.
test-url: https://github.com/madler/pigz
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 split-job --compatible job.yaml # generate the yaml file for lkp run
bin/lkp run generated-yaml-file
=========================================================================================
blocksize/compiler/cpufreq_governor/kconfig/nr_threads/rootfs/tbox_group/testcase/ucode:
128K/gcc-9/performance/x86_64-rhel-8.3/25%/debian-10.4-x86_64-20200603.cgz/lkp-knm02/pigz/0x11
commit:
25a1298726 ("Merge tag 'trace-v5.13-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace")
ffb324e6f8 ("tty: vt: always invoke vc->vc_sw->con_resize callback")
25a1298726e97b9d ffb324e6f874121f7dce5bdae5e
---------------- ---------------------------
%stddev %change %stddev
\ | \
2.46e+08 +1.3% 2.493e+08 pigz.throughput
0.06 ? 17% +96.6% 0.13 ? 37% perf-sched.sch_delay.max.ms.preempt_schedule_common.__cond_resched.stop_one_cpu.sched_exec.bprm_execve
1726001 ? 4% -10.2% 1549532 ? 5% meminfo.DirectMap4k
410.67 ?170% +230.0% 1355 ? 65% meminfo.Mlocked
18201 ? 6% +16.3% 21169 ? 6% softirqs.CPU104.SCHED
20791 ? 9% +13.3% 23550 ? 3% softirqs.CPU165.SCHED
21729 ? 4% -10.1% 19533 ? 7% softirqs.CPU204.SCHED
20348 ? 9% +12.6% 22921 ? 6% softirqs.CPU213.SCHED
20581 ? 9% -11.9% 18124 ? 8% softirqs.CPU9.SCHED
20624 ? 4% +10.4% 22772 ? 4% softirqs.CPU91.SCHED
11.44 ? 3% -1.9 9.56 ? 21% perf-profile.calltrace.cycles-pp.menu_select.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64_no_verify
3.20 ? 5% -0.6 2.56 ? 23% perf-profile.calltrace.cycles-pp.get_next_timer_interrupt.tick_nohz_next_event.tick_nohz_get_sleep_length.menu_select.do_idle
1.28 ? 3% -0.5 0.77 ? 71% perf-profile.calltrace.cycles-pp.idle_cpu.irq_exit_rcu.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state
11.52 ? 3% -1.9 9.65 ? 21% perf-profile.children.cycles-pp.menu_select
8.63 ? 5% -1.5 7.17 ? 16% perf-profile.children.cycles-pp.intel_idle
3.23 ? 5% -0.6 2.59 ? 23% perf-profile.children.cycles-pp.get_next_timer_interrupt
0.43 ? 6% +0.1 0.49 ? 9% perf-profile.children.cycles-pp.tick_program_event
8.62 ? 5% -1.5 7.16 ? 15% perf-profile.self.cycles-pp.intel_idle
1.03 ? 3% -0.2 0.88 ? 15% perf-profile.self.cycles-pp.do_idle
0.19 ? 7% -0.0 0.15 ? 14% perf-profile.self.cycles-pp.sched_clock
0.42 ? 7% +0.1 0.49 ? 10% perf-profile.self.cycles-pp.tick_program_event
2235 ? 20% -21.3% 1759 ? 9% interrupts.CPU0.CAL:Function_call_interrupts
2451 ? 15% -40.3% 1462 ? 11% interrupts.CPU104.TLB:TLB_shootdowns
34.50 ? 98% -69.6% 10.50 ? 31% interrupts.CPU139.RES:Rescheduling_interrupts
3437 ? 5% -16.1% 2885 ? 8% interrupts.CPU140.CAL:Function_call_interrupts
1928 ? 6% -20.1% 1540 ? 15% interrupts.CPU140.TLB:TLB_shootdowns
335.00 ?117% -98.1% 6.50 ? 37% interrupts.CPU153.RES:Rescheduling_interrupts
3160 ? 15% -20.4% 2514 ? 12% interrupts.CPU165.CAL:Function_call_interrupts
2356 ? 8% +47.4% 3473 ? 14% interrupts.CPU172.CAL:Function_call_interrupts
993.17 ? 21% +81.8% 1805 ? 24% interrupts.CPU172.TLB:TLB_shootdowns
3210 ? 13% -22.5% 2487 ? 16% interrupts.CPU209.CAL:Function_call_interrupts
1750 ? 21% -33.6% 1161 ? 31% interrupts.CPU209.TLB:TLB_shootdowns
178.50 ?131% -94.7% 9.50 ? 29% interrupts.CPU221.RES:Rescheduling_interrupts
1942 ? 13% +43.4% 2786 ? 22% interrupts.CPU24.TLB:TLB_shootdowns
2170 ? 77% -83.2% 365.33 ? 82% interrupts.CPU242.NMI:Non-maskable_interrupts
2170 ? 77% -83.2% 365.33 ? 82% interrupts.CPU242.PMI:Performance_monitoring_interrupts
2226 ? 15% +25.0% 2782 ? 12% interrupts.CPU254.CAL:Function_call_interrupts
2389 ? 9% +17.9% 2815 ? 11% interrupts.CPU268.CAL:Function_call_interrupts
1619 ?105% -82.6% 281.33 ? 42% interrupts.CPU39.NMI:Non-maskable_interrupts
1619 ?105% -82.6% 281.33 ? 42% interrupts.CPU39.PMI:Performance_monitoring_interrupts
3060 ? 50% -70.9% 890.67 ?107% interrupts.CPU55.NMI:Non-maskable_interrupts
3060 ? 50% -70.9% 890.67 ?107% interrupts.CPU55.PMI:Performance_monitoring_interrupts
180.17 ?111% -82.9% 30.83 ?140% interrupts.CPU78.RES:Rescheduling_interrupts
323.50 ?123% -96.0% 12.83 ? 45% interrupts.CPU84.RES:Rescheduling_interrupts
430.67 ? 74% +209.9% 1334 ? 67% interrupts.CPU9.NMI:Non-maskable_interrupts
430.67 ? 74% +209.9% 1334 ? 67% interrupts.CPU9.PMI:Performance_monitoring_interrupts
49.50 ? 23% +240.4% 168.50 ? 89% interrupts.CPU9.RES:Rescheduling_interrupts
pigz.throughput
2.6e+08 +----------------------------------------------------------------+
2.58e+08 |-+ |
| O |
2.56e+08 |-+ |
2.54e+08 |-+ O O |
| |
2.52e+08 |-+O |
2.5e+08 |-+ O O O +.. |
2.48e+08 |-+ O O +..+.. + . |
|..+...+.. .+... .. O . O+ + O |
2.46e+08 |-+ +...+. + + +..+ O O |
2.44e+08 |-+ + .. |
| + . |
2.42e+08 |-+ + |
2.4e+08 +----------------------------------------------------------------+
[*] 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.
---
0DAY/LKP+ Test Infrastructure Open Source Technology Center
https://lists.01.org/hyperkitty/list/[email protected] Intel Corporation
Thanks,
Oliver Sang