RE: [x86/split_lock] b041b525da: stress-ng.lockbus.ops_per_sec -48.3% regression

From: Luck, Tony
Date: Fri May 06 2022 - 11:36:50 EST


Only -43.8%? That just sounds like I made life "pretty unpleasant", not "miserable" :-)

This regression is expected. No plans to fix it.

You might look into a benchmark that measures performance of other applications while this
stress-ng.lockbus.ops_per_sec is running. The intent of this patch is to reduce the impact
of bus locks on the rest of the system.

-Tony

-----Original Message-----
From: Sang, Oliver <oliver.sang@xxxxxxxxx>
Sent: Friday, May 06, 2022 1:43 AM
To: Luck, Tony <tony.luck@xxxxxxxxx>
Cc: Thomas Gleixner <tglx@xxxxxxxxxxxxx>; LKML <linux-kernel@xxxxxxxxxxxxxxx>; x86@xxxxxxxxxx; lkp@xxxxxxxxxxxx; lkp <lkp@xxxxxxxxx>; Huang, Ying <ying.huang@xxxxxxxxx>; Tang, Feng <feng.tang@xxxxxxxxx>; zhengjun.xing@xxxxxxxxxxxxxxx; Yin, Fengwei <fengwei.yin@xxxxxxxxx>
Subject: [x86/split_lock] b041b525da: stress-ng.lockbus.ops_per_sec -48.3% regression



Greeting,

FYI, we noticed a -48.3% regression of stress-ng.lockbus.ops_per_sec due to commit:


commit: b041b525dab95352fbd666b14dc73ab898df465f ("x86/split_lock: Make life miserable for split lockers")
https://git.kernel.org/cgit/linux/kernel/git/tip/tip.git x86/splitlock

in testcase: stress-ng
on test machine: 128 threads 2 sockets Intel(R) Xeon(R) Platinum 8358 CPU @ 2.60GHz with 128G memory
with following parameters:

nr_threads: 100%
testtime: 60s
class: memory
test: lockbus
cpufreq_governor: performance
ucode: 0xd000331




If you fix the issue, kindly add following tag
Reported-by: kernel test robot <oliver.sang@xxxxxxxxx>


Details are as below:
-------------------------------------------------------------------------------------------------->


To reproduce:

git clone https://github.com/intel/lkp-tests.git
cd lkp-tests
sudo 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
sudo bin/lkp run generated-yaml-file

# if come across any failure that blocks the test,
# please remove ~/.lkp and /lkp dir to run from a clean state.

=========================================================================================
class/compiler/cpufreq_governor/kconfig/nr_threads/rootfs/tbox_group/test/testcase/testtime/ucode:
memory/gcc-11/performance/x86_64-rhel-8.3/100%/debian-10.4-x86_64-20200603.cgz/lkp-icl-2sp6/lockbus/stress-ng/60s/0xd000331

commit:
v5.18-rc4
b041b525da ("x86/split_lock: Make life miserable for split lockers")

v5.18-rc4 b041b525dab95352fbd666b14dc
---------------- ---------------------------
%stddev %change %stddev
\ | \
84375 -48.3% 43587 ± 3% stress-ng.lockbus.ops
1400 -48.3% 723.64 ± 3% stress-ng.lockbus.ops_per_sec
46977 ± 2% -61.4% 18141 ± 11% stress-ng.time.involuntary_context_switches
7779 -98.7% 97.67 stress-ng.time.percent_of_cpu_this_job_got
251.13 ± 21% -99.3% 1.80 ± 3% stress-ng.time.system_time
4660 -98.7% 59.55 stress-ng.time.user_time
431.40 ± 10% +8470.7% 36974 ± 11% stress-ng.time.voluntary_context_switches
293203 +12.9% 331056 meminfo.AnonHugePages
4712 ± 3% +156.4% 12081 uptime.idle
5.616e+08 ± 2% +1311.6% 7.927e+09 cpuidle..time
1017813 ± 20% +1502.7% 16312100 cpuidle..usage
6.44 ± 8% +90.9 97.30 mpstat.cpu.all.idle%
34.59 ± 7% -32.8 1.76 ± 31% mpstat.cpu.all.irq%
0.89 ± 13% -0.8 0.09 ± 44% mpstat.cpu.all.soft%
5.10 ± 8% -5.0 0.07 ± 3% mpstat.cpu.all.sys%
52.99 ± 6% -52.2 0.78 mpstat.cpu.all.usr%
878964 ± 5% +10.8% 974067 numa-meminfo.node0.Shmem
1018902 ± 7% +16.1% 1183099 ± 5% numa-meminfo.node1.Inactive
1018902 ± 7% +16.1% 1183099 ± 5% numa-meminfo.node1.Inactive(anon)
842061 ± 6% +17.5% 989596 ± 2% numa-meminfo.node1.Mapped
848272 ± 5% +15.2% 976855 numa-meminfo.node1.Shmem
9.80 ± 4% +886.4% 96.67 vmstat.cpu.id
50.00 ± 6% -100.0% 0.00 vmstat.cpu.us
121.40 -99.2% 1.00 vmstat.procs.r
2548 +34.1% 3418 ± 10% vmstat.system.cs
144004 ± 9% +74.3% 250999 vmstat.system.in
282203 ± 4% +14.8% 323935 ± 5% numa-vmstat.node0.nr_inactive_anon
211496 ± 6% +16.6% 246585 ± 2% numa-vmstat.node0.nr_mapped
206680 ± 5% +18.9% 245754 ± 2% numa-vmstat.node0.nr_shmem
282195 ± 4% +14.8% 323935 ± 5% numa-vmstat.node0.nr_zone_inactive_anon
240558 ± 8% +24.0% 298317 ± 5% numa-vmstat.node1.nr_inactive_anon
197879 ± 7% +26.2% 249742 ± 2% numa-vmstat.node1.nr_mapped
1642 ± 40% +27.4% 2093 ± 40% numa-vmstat.node1.nr_page_table_pages
199997 ± 6% +23.2% 246443 numa-vmstat.node1.nr_shmem
240505 ± 8% +24.0% 298317 ± 5% numa-vmstat.node1.nr_zone_inactive_anon
3066 -97.4% 78.67 ± 19% turbostat.Avg_MHz
93.28 -89.2 4.07 ± 22% turbostat.Busy%
3296 -40.8% 1952 ± 5% turbostat.Bzy_MHz
781444 ± 54% +1628.9% 13510398 ± 28% turbostat.C1E
4.23 ± 61% +64.0 68.28 ± 57% turbostat.C1E%
4.44 ± 55% +1453.3% 69.03 ± 55% turbostat.CPU%c1
63.20 -16.4% 52.83 ± 4% turbostat.CoreTmp
8352767 ± 8% +97.6% 16501874 turbostat.IRQ
63.00 ± 2% -16.1% 52.83 ± 2% turbostat.PkgTmp
375.56 -39.9% 225.74 ± 6% turbostat.PkgWatt
121375 +7.5% 130425 proc-vmstat.nr_anon_pages
987453 +7.0% 1056187 proc-vmstat.nr_file_pages
537372 +14.8% 616827 proc-vmstat.nr_inactive_anon
422895 +16.2% 491550 proc-vmstat.nr_mapped
4107 ± 2% +6.8% 4385 ± 2% proc-vmstat.nr_page_table_pages
418750 +16.4% 487474 proc-vmstat.nr_shmem
537370 +14.8% 616827 proc-vmstat.nr_zone_inactive_anon
3330 ± 61% -99.7% 11.50 ±186% proc-vmstat.numa_hint_faults
2035 ± 59% -99.9% 1.67 ±197% proc-vmstat.numa_hint_faults_local
1002524 +4.6% 1048399 proc-vmstat.numa_hit
886626 +5.0% 931095 proc-vmstat.numa_local
5946 ±116% -98.4% 95.00 ±223% proc-vmstat.numa_pages_migrated
493749 ± 3% -99.8% 853.33 ±223% proc-vmstat.numa_pte_updates
1001222 +4.6% 1046837 proc-vmstat.pgalloc_normal
841720 +8.6% 914206 proc-vmstat.pgfault
775048 +18.1% 915055 proc-vmstat.pgfree
5946 ±116% -98.4% 95.00 ±223% proc-vmstat.pgmigrate_success
17361 +54.0% 26742 proc-vmstat.pgreuse
5.129e+08 ± 7% -41.1% 3.022e+08 ± 2% perf-stat.i.branch-instructions
0.95 ± 11% +3.6 4.58 ± 99% perf-stat.i.branch-miss-rate%
2382071 ± 7% -44.4% 1324699 ± 8% perf-stat.i.cache-misses
2539 ± 2% +20.4% 3057 ± 12% perf-stat.i.context-switches
426.49 ± 4% -97.6% 10.23 ± 13% perf-stat.i.cpi
139782 -8.4% 128033 perf-stat.i.cpu-clock
2.639e+11 -95.8% 1.118e+10 ± 11% perf-stat.i.cpu-cycles
752036 ± 5% -98.0% 14941 ± 26% perf-stat.i.cycles-between-cache-misses
5.608e+08 ± 9% -32.0% 3.812e+08 perf-stat.i.dTLB-loads
0.41 ± 6% -0.3 0.13 ±118% perf-stat.i.dTLB-store-miss-rate%
2.391e+08 ± 9% -20.5% 1.9e+08 perf-stat.i.dTLB-stores
2.44e+09 ± 9% -38.9% 1.49e+09 ± 2% perf-stat.i.instructions
54.70 ± 5% -77.1% 12.51 perf-stat.i.major-faults
1.88 ± 2% -95.4% 0.09 ± 11% perf-stat.i.metric.GHz
944.84 ± 19% -64.4% 336.04 ±112% perf-stat.i.metric.K/sec
8.61 ± 8% -20.7% 6.83 perf-stat.i.metric.M/sec
10898 ± 6% -58.0% 4579 perf-stat.i.minor-faults
65.86 ± 4% +18.6 84.51 perf-stat.i.node-load-miss-rate%
141261 ± 8% +56.6% 221158 ± 15% perf-stat.i.node-load-misses
91471 ± 5% -46.1% 49309 ± 16% perf-stat.i.node-loads
89300 ± 11% -54.1% 40980 ± 21% perf-stat.i.node-store-misses
1212554 ± 7% -57.6% 514245 ± 4% perf-stat.i.node-stores
10953 ± 6% -58.1% 4591 perf-stat.i.page-faults
139787 -8.4% 128033 perf-stat.i.task-clock
254.01 ± 8% -97.0% 7.52 ± 12% perf-stat.overall.cpi
286326 ± 9% -97.0% 8577 ± 19% perf-stat.overall.cycles-between-cache-misses
0.00 ± 9% +3300.4% 0.13 ± 11% perf-stat.overall.ipc
61.44 ± 9% +20.3 81.74 perf-stat.overall.node-load-miss-rate%
1.77e+08 ± 9% +67.8% 2.97e+08 ± 2% perf-stat.ps.branch-instructions
2702583 ± 18% +329.2% 11598999 ± 82% perf-stat.ps.branch-misses
716805 ± 10% +81.6% 1301900 ± 8% perf-stat.ps.cache-misses
1459 ± 5% +106.1% 3008 ± 12% perf-stat.ps.context-switches
95205 ± 8% +32.3% 125997 perf-stat.ps.cpu-clock
2.037e+11 ± 7% -94.6% 1.1e+10 ± 11% perf-stat.ps.cpu-cycles
120.76 ± 7% +42.3% 171.90 ± 9% perf-stat.ps.cpu-migrations
2.002e+08 ± 10% +87.2% 3.748e+08 perf-stat.ps.dTLB-loads
89201266 ± 11% +109.5% 1.868e+08 perf-stat.ps.dTLB-stores
8.074e+08 ± 10% +81.4% 1.465e+09 ± 2% perf-stat.ps.instructions
3184 ± 8% +41.1% 4492 perf-stat.ps.minor-faults
61589 ± 10% +253.4% 217652 ± 15% perf-stat.ps.node-load-misses
29168 ± 6% +38.2% 40305 ± 21% perf-stat.ps.node-store-misses
303641 ± 13% +66.2% 504661 ± 4% perf-stat.ps.node-stores
3198 ± 8% +40.8% 4504 perf-stat.ps.page-faults
95207 ± 8% +32.3% 125997 perf-stat.ps.task-clock
4.66e+10 ± 12% +99.4% 9.294e+10 ± 2% perf-stat.total.instructions
32354 ± 57% -99.8% 61.76 ± 70% sched_debug.cfs_rq:/.MIN_vruntime.avg
2247989 ± 26% -99.8% 5064 ± 56% sched_debug.cfs_rq:/.MIN_vruntime.max
247145 ± 39% -99.8% 537.60 ± 62% sched_debug.cfs_rq:/.MIN_vruntime.stddev
0.57 ± 4% -83.4% 0.10 ± 28% sched_debug.cfs_rq:/.h_nr_running.avg
32354 ± 57% -99.8% 61.76 ± 70% sched_debug.cfs_rq:/.max_vruntime.avg
2247991 ± 26% -99.8% 5064 ± 56% sched_debug.cfs_rq:/.max_vruntime.max
247145 ± 39% -99.8% 537.60 ± 62% sched_debug.cfs_rq:/.max_vruntime.stddev
2763104 -99.7% 8913 ± 36% sched_debug.cfs_rq:/.min_vruntime.avg
3358182 ± 4% -99.3% 24400 ± 13% sched_debug.cfs_rq:/.min_vruntime.max
1538537 ± 12% -99.8% 2844 ± 28% sched_debug.cfs_rq:/.min_vruntime.min
156471 ± 5% -97.6% 3686 ± 24% sched_debug.cfs_rq:/.min_vruntime.stddev
0.55 ± 3% -82.7% 0.10 ± 28% sched_debug.cfs_rq:/.nr_running.avg
119.88 ± 20% +55.8% 186.73 ± 29% sched_debug.cfs_rq:/.removed.load_avg.stddev
59.29 ± 28% +51.4% 89.77 ± 30% sched_debug.cfs_rq:/.removed.util_avg.stddev
665.61 -64.8% 234.43 ± 18% sched_debug.cfs_rq:/.runnable_avg.avg
1886 ± 8% -37.4% 1181 ± 16% sched_debug.cfs_rq:/.runnable_avg.max
528967 ± 9% -101.6% -8642 sched_debug.cfs_rq:/.spread0.avg
1119786 ± 10% -99.4% 6842 ± 55% sched_debug.cfs_rq:/.spread0.max
-715437 -97.9% -14712 sched_debug.cfs_rq:/.spread0.min
152488 ± 7% -97.6% 3686 ± 24% sched_debug.cfs_rq:/.spread0.stddev
628.76 -62.8% 233.60 ± 18% sched_debug.cfs_rq:/.util_avg.avg
1585 ± 13% -25.6% 1179 ± 16% sched_debug.cfs_rq:/.util_avg.max
303.61 ± 3% -93.0% 21.11 ± 34% sched_debug.cfs_rq:/.util_est_enqueued.avg
1127 -32.9% 756.50 ± 18% sched_debug.cfs_rq:/.util_est_enqueued.max
206.96 ± 4% -51.3% 100.87 ± 22% sched_debug.cfs_rq:/.util_est_enqueued.stddev
2525689 ± 11% -64.2% 903412 ± 2% sched_debug.cpu.avg_idle.avg
6234029 ± 9% -79.7% 1266582 ± 16% sched_debug.cpu.avg_idle.max
299393 ± 18% -95.8% 12702 ±196% sched_debug.cpu.avg_idle.min
1331474 ± 13% -86.1% 185160 ± 11% sched_debug.cpu.avg_idle.stddev
68351 -39.0% 41714 ± 24% sched_debug.cpu.clock.avg
68613 -39.2% 41719 ± 24% sched_debug.cpu.clock.max
67731 -38.4% 41708 ± 24% sched_debug.cpu.clock.min
275.33 ± 28% -98.9% 3.00 ± 33% sched_debug.cpu.clock.stddev
1145 ± 11% -58.0% 481.20 ± 36% sched_debug.cpu.clock_task.stddev
3293 ± 2% -89.3% 352.66 ± 26% sched_debug.cpu.curr->pid.avg
1704382 ± 8% -70.6% 501841 sched_debug.cpu.max_idle_balance_cost.avg
3528217 ± 7% -80.3% 695415 ± 14% sched_debug.cpu.max_idle_balance_cost.max
638236 ± 16% -97.1% 18481 ± 54% sched_debug.cpu.max_idle_balance_cost.stddev
0.00 ± 37% -97.0% 0.00 ± 32% sched_debug.cpu.next_balance.stddev
0.57 ± 3% -83.3% 0.10 ± 26% sched_debug.cpu.nr_running.avg
2.40 ± 8% -51.4% 1.17 ± 31% sched_debug.cpu.nr_running.max
0.34 ± 4% -13.2% 0.29 ± 15% sched_debug.cpu.nr_running.stddev
67624 -38.3% 41712 ± 24% sched_debug.cpu_clk
66904 -38.7% 40993 ± 24% sched_debug.ktime
0.00 -100.0% 0.00 sched_debug.rt_rq:/.rt_nr_migratory.avg
0.50 -100.0% 0.00 sched_debug.rt_rq:/.rt_nr_migratory.max
0.04 -100.0% 0.00 sched_debug.rt_rq:/.rt_nr_migratory.stddev
0.00 ± 33% -100.0% 0.00 sched_debug.rt_rq:/.rt_nr_running.avg
0.50 -100.0% 0.00 sched_debug.rt_rq:/.rt_nr_running.max
0.05 ± 15% -100.0% 0.00 sched_debug.rt_rq:/.rt_nr_running.stddev
2.18 ± 19% -100.0% 0.00 sched_debug.rt_rq:/.rt_time.avg
273.15 ± 24% -100.0% 0.00 sched_debug.rt_rq:/.rt_time.max
24.05 ± 24% -100.0% 0.00 sched_debug.rt_rq:/.rt_time.stddev
68997 -37.0% 43458 ± 25% sched_debug.sched_clk
128.87 ± 15% -122.8 6.06 ± 45% perf-profile.calltrace.cycles-pp.asm_sysvec_apic_timer_interrupt.stress_lockbus
43.54 ± 21% -18.7 24.80 ± 11% perf-profile.calltrace.cycles-pp.stress_lockbus
15.57 ± 3% -15.6 0.00 perf-profile.calltrace.cycles-pp.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.stress_lockbus
13.77 ± 3% -13.8 0.00 perf-profile.calltrace.cycles-pp.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.stress_lockbus
13.65 ± 3% -13.7 0.00 perf-profile.calltrace.cycles-pp.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.stress_lockbus
10.18 ± 7% -5.4 4.76 ± 55% perf-profile.calltrace.cycles-pp.__hrtimer_run_queues.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt
6.71 ± 12% -3.7 2.97 ± 54% perf-profile.calltrace.cycles-pp.tick_sched_timer.__hrtimer_run_queues.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt
5.91 ± 14% -3.6 2.32 ± 55% perf-profile.calltrace.cycles-pp.update_process_times.tick_sched_handle.tick_sched_timer.__hrtimer_run_queues.hrtimer_interrupt
6.06 ± 14% -3.5 2.54 ± 60% perf-profile.calltrace.cycles-pp.tick_sched_handle.tick_sched_timer.__hrtimer_run_queues.hrtimer_interrupt.__sysvec_apic_timer_interrupt
4.80 ± 15% -3.5 1.31 ± 44% perf-profile.calltrace.cycles-pp.scheduler_tick.update_process_times.tick_sched_handle.tick_sched_timer.__hrtimer_run_queues
2.12 ± 10% -1.3 0.84 ± 46% perf-profile.calltrace.cycles-pp.perf_mux_hrtimer_handler.__hrtimer_run_queues.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt
2.01 ± 8% -0.5 1.49 ± 19% perf-profile.calltrace.cycles-pp.clockevents_program_event.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt
0.58 ± 6% +0.4 1.02 ± 20% perf-profile.calltrace.cycles-pp.rebalance_domains.__softirqentry_text_start.__irq_exit_rcu.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt
0.00 +0.6 0.61 ± 7% perf-profile.calltrace.cycles-pp.lapic_next_deadline.clockevents_program_event.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt
0.00 +0.7 0.70 ± 22% perf-profile.calltrace.cycles-pp.load_balance.rebalance_domains.__softirqentry_text_start.__irq_exit_rcu.sysvec_apic_timer_interrupt
0.00 +0.8 0.78 ± 37% perf-profile.calltrace.cycles-pp.ktime_get.tick_irq_enter.irq_enter_rcu.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt
0.00 +1.2 1.17 ± 27% perf-profile.calltrace.cycles-pp.tick_irq_enter.irq_enter_rcu.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state
0.00 +1.2 1.21 ± 26% perf-profile.calltrace.cycles-pp.irq_enter_rcu.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state.cpuidle_enter
0.00 +1.4 1.38 ± 12% perf-profile.calltrace.cycles-pp.tick_nohz_next_event.tick_nohz_get_sleep_length.menu_select.cpuidle_idle_call.do_idle
0.00 +1.9 1.87 ± 28% perf-profile.calltrace.cycles-pp.__softirqentry_text_start.__irq_exit_rcu.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state
0.00 +2.1 2.10 ± 6% perf-profile.calltrace.cycles-pp.tick_nohz_get_sleep_length.menu_select.cpuidle_idle_call.do_idle.cpu_startup_entry
0.00 +2.8 2.84 ± 14% perf-profile.calltrace.cycles-pp.__irq_exit_rcu.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state.cpuidle_enter
0.00 +4.4 4.42 ± 18% perf-profile.calltrace.cycles-pp.menu_select.cpuidle_idle_call.do_idle.cpu_startup_entry.secondary_startup_64_no_verify
0.00 +7.3 7.34 ± 36% perf-profile.calltrace.cycles-pp.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state
0.00 +7.5 7.53 ± 38% perf-profile.calltrace.cycles-pp.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state.cpuidle_enter
0.00 +12.6 12.59 ± 31% perf-profile.calltrace.cycles-pp.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call
0.00 +15.9 15.90 ± 26% perf-profile.calltrace.cycles-pp.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call.do_idle
0.00 +47.0 47.02 ± 8% perf-profile.calltrace.cycles-pp.mwait_idle_with_hints.intel_idle.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call
0.00 +47.2 47.21 ± 8% perf-profile.calltrace.cycles-pp.intel_idle.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call.do_idle
0.00 +65.0 64.98 ± 3% perf-profile.calltrace.cycles-pp.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call.do_idle.cpu_startup_entry
0.00 +65.3 65.26 ± 3% perf-profile.calltrace.cycles-pp.cpuidle_enter.cpuidle_idle_call.do_idle.cpu_startup_entry.secondary_startup_64_no_verify
0.00 +70.6 70.57 ± 4% perf-profile.calltrace.cycles-pp.cpuidle_idle_call.do_idle.cpu_startup_entry.secondary_startup_64_no_verify
0.00 +71.3 71.30 ± 4% perf-profile.calltrace.cycles-pp.do_idle.cpu_startup_entry.secondary_startup_64_no_verify
0.00 +71.4 71.39 ± 4% perf-profile.calltrace.cycles-pp.cpu_startup_entry.secondary_startup_64_no_verify
0.00 +72.0 71.98 ± 4% perf-profile.calltrace.cycles-pp.secondary_startup_64_no_verify
95.06 -68.0 27.09 ± 12% perf-profile.children.cycles-pp.stress_lockbus
80.12 ± 13% -61.7 18.47 ± 23% perf-profile.children.cycles-pp.asm_sysvec_apic_timer_interrupt
14.64 ± 3% -7.2 7.48 ± 36% perf-profile.children.cycles-pp.hrtimer_interrupt
14.76 ± 3% -7.1 7.66 ± 38% perf-profile.children.cycles-pp.__sysvec_apic_timer_interrupt
10.91 ± 7% -6.1 4.86 ± 54% perf-profile.children.cycles-pp.__hrtimer_run_queues
7.20 ± 12% -4.2 3.02 ± 54% perf-profile.children.cycles-pp.tick_sched_timer
6.33 ± 14% -4.0 2.38 ± 55% perf-profile.children.cycles-pp.update_process_times
6.48 ± 14% -3.9 2.58 ± 60% perf-profile.children.cycles-pp.tick_sched_handle
5.18 ± 15% -3.8 1.35 ± 43% perf-profile.children.cycles-pp.scheduler_tick
2.95 ± 9% -2.5 0.43 ± 17% perf-profile.children.cycles-pp.entry_SYSCALL_64_after_hwframe
2.95 ± 9% -2.5 0.43 ± 17% perf-profile.children.cycles-pp.do_syscall_64
2.32 ± 10% -1.4 0.87 ± 43% perf-profile.children.cycles-pp.perf_mux_hrtimer_handler
1.61 ± 5% -1.3 0.27 ± 10% perf-profile.children.cycles-pp.__intel_pmu_enable_all
1.37 ± 12% -1.3 0.07 ± 55% perf-profile.children.cycles-pp.update_load_avg
1.41 ± 2% -1.3 0.15 ± 28% perf-profile.children.cycles-pp.error_entry
1.46 ± 4% -0.9 0.56 ± 21% perf-profile.children.cycles-pp.native_sched_clock
0.82 ± 10% -0.7 0.10 ± 27% perf-profile.children.cycles-pp.asm_exc_page_fault
2.21 ± 8% -0.7 1.52 ± 18% perf-profile.children.cycles-pp.clockevents_program_event
0.71 ± 9% -0.6 0.09 ± 28% perf-profile.children.cycles-pp.exc_page_fault
0.70 ± 9% -0.6 0.09 ± 27% perf-profile.children.cycles-pp.do_user_addr_fault
0.58 ± 7% -0.5 0.08 ± 29% perf-profile.children.cycles-pp.handle_mm_fault
0.53 ± 7% -0.5 0.08 ± 35% perf-profile.children.cycles-pp.__handle_mm_fault
0.53 ± 10% -0.5 0.07 ± 39% perf-profile.children.cycles-pp.__update_load_avg_cfs_rq
0.51 ± 4% -0.4 0.09 ± 11% perf-profile.children.cycles-pp.__x64_sys_execve
0.51 ± 3% -0.4 0.09 ± 11% perf-profile.children.cycles-pp.do_execveat_common
0.50 ± 2% -0.4 0.09 ± 11% perf-profile.children.cycles-pp.execve
0.67 ± 17% -0.4 0.28 ± 30% perf-profile.children.cycles-pp.ret_from_fork
0.65 ± 17% -0.4 0.28 ± 31% perf-profile.children.cycles-pp.kthread
0.76 ± 10% -0.3 0.41 ± 41% perf-profile.children.cycles-pp.update_rq_clock
0.40 ± 4% -0.3 0.07 ± 18% perf-profile.children.cycles-pp.bprm_execve
0.36 ± 4% -0.3 0.06 ± 46% perf-profile.children.cycles-pp.exec_binprm
0.35 ± 7% -0.3 0.04 ± 73% perf-profile.children.cycles-pp.exit_mmap
0.36 ± 4% -0.3 0.06 ± 46% perf-profile.children.cycles-pp.search_binary_handler
0.36 ± 3% -0.3 0.06 ± 46% perf-profile.children.cycles-pp.load_elf_binary
0.35 ± 6% -0.3 0.05 ± 48% perf-profile.children.cycles-pp.mmput
0.40 ± 7% -0.3 0.12 ± 20% perf-profile.children.cycles-pp.__schedule
0.36 ± 8% -0.3 0.08 ± 19% perf-profile.children.cycles-pp.schedule
0.72 ± 5% -0.3 0.45 ± 27% perf-profile.children.cycles-pp._raw_spin_lock
0.30 ± 3% -0.3 0.03 ±100% perf-profile.children.cycles-pp.__x64_sys_openat
0.30 ± 4% -0.3 0.03 ±100% perf-profile.children.cycles-pp.do_sys_openat2
0.34 ± 8% -0.3 0.08 ± 36% perf-profile.children.cycles-pp.ksys_read
0.33 ± 7% -0.3 0.08 ± 37% perf-profile.children.cycles-pp.vfs_read
0.38 ± 2% -0.2 0.13 ± 27% perf-profile.children.cycles-pp._raw_spin_lock_irq
0.27 ± 2% -0.2 0.03 ±100% perf-profile.children.cycles-pp.do_filp_open
0.27 ± 2% -0.2 0.03 ±100% perf-profile.children.cycles-pp.path_openat
0.65 ± 13% -0.2 0.45 ± 9% perf-profile.children.cycles-pp.ktime_get_update_offsets_now
0.33 ± 17% -0.2 0.16 ± 55% perf-profile.children.cycles-pp.process_one_work
0.35 ± 17% -0.2 0.18 ± 42% perf-profile.children.cycles-pp.worker_thread
0.12 ± 10% -0.1 0.02 ± 99% perf-profile.children.cycles-pp.smpboot_thread_fn
0.20 ± 7% -0.1 0.11 ± 33% perf-profile.children.cycles-pp._raw_spin_unlock_irqrestore
0.07 ± 9% +0.0 0.10 ± 20% perf-profile.children.cycles-pp.cpumask_next_and
0.02 ±122% +0.1 0.07 ± 28% perf-profile.children.cycles-pp.update_group_capacity
0.00 +0.1 0.06 ± 19% perf-profile.children.cycles-pp.cpuidle_reflect
0.13 ± 7% +0.1 0.20 ± 27% perf-profile.children.cycles-pp._raw_spin_trylock
0.00 +0.1 0.07 ± 23% perf-profile.children.cycles-pp.sched_clock
0.00 +0.1 0.10 ± 26% perf-profile.children.cycles-pp.arch_cpu_idle_exit
0.00 +0.1 0.11 ± 41% perf-profile.children.cycles-pp.rcu_eqs_enter
0.01 ±200% +0.1 0.12 ± 28% perf-profile.children.cycles-pp.__libc_start_main
0.01 ±200% +0.1 0.12 ± 28% perf-profile.children.cycles-pp.main
0.01 ±200% +0.1 0.12 ± 28% perf-profile.children.cycles-pp.run_builtin
0.24 ± 4% +0.1 0.36 ± 11% perf-profile.children.cycles-pp.update_sg_lb_stats
0.00 +0.1 0.12 ± 47% perf-profile.children.cycles-pp.calc_global_load_tick
0.00 +0.1 0.12 ± 13% perf-profile.children.cycles-pp.menu_reflect
0.00 +0.1 0.13 ± 67% perf-profile.children.cycles-pp.tick_nohz_tick_stopped
0.00 +0.1 0.14 ± 26% perf-profile.children.cycles-pp.call_cpuidle
0.00 +0.2 0.15 ± 25% perf-profile.children.cycles-pp.rcu_dynticks_inc
0.00 +0.2 0.18 ± 58% perf-profile.children.cycles-pp.timerqueue_iterate_next
0.00 +0.2 0.19 ± 42% perf-profile.children.cycles-pp.io_serial_in
0.23 ± 13% +0.2 0.44 ± 17% perf-profile.children.cycles-pp.__hrtimer_next_event_base
0.03 ± 82% +0.2 0.26 ± 45% perf-profile.children.cycles-pp.__update_blocked_fair
0.33 ± 6% +0.2 0.56 ± 10% perf-profile.children.cycles-pp.update_sd_lb_stats
0.00 +0.2 0.24 ± 65% perf-profile.children.cycles-pp.nr_iowait_cpu
0.34 ± 6% +0.3 0.59 ± 11% perf-profile.children.cycles-pp.find_busiest_group
0.00 +0.3 0.25 ± 46% perf-profile.children.cycles-pp.serial8250_console_putchar
0.52 ± 7% +0.3 0.78 ± 17% perf-profile.children.cycles-pp.load_balance
0.39 ± 15% +0.3 0.64 ± 7% perf-profile.children.cycles-pp.lapic_next_deadline
0.00 +0.3 0.26 ± 43% perf-profile.children.cycles-pp.wait_for_xmitr
0.00 +0.3 0.26 ± 44% perf-profile.children.cycles-pp.uart_console_write
0.22 ± 10% +0.3 0.48 ± 48% perf-profile.children.cycles-pp.update_blocked_averages
0.00 +0.3 0.27 ± 44% perf-profile.children.cycles-pp.serial8250_console_write
0.00 +0.3 0.27 ± 18% perf-profile.children.cycles-pp.hrtimer_get_next_event
0.23 ± 11% +0.3 0.50 ± 50% perf-profile.children.cycles-pp.run_rebalance_domains
0.00 +0.3 0.27 ± 42% perf-profile.children.cycles-pp.console_unlock
0.00 +0.3 0.27 ± 42% perf-profile.children.cycles-pp.call_console_drivers
0.03 ± 82% +0.3 0.31 ± 42% perf-profile.children.cycles-pp.irq_work_run_list
0.00 +0.3 0.28 ± 42% perf-profile.children.cycles-pp._printk
0.00 +0.3 0.28 ± 42% perf-profile.children.cycles-pp.vprintk_emit
0.00 +0.3 0.28 ± 42% perf-profile.children.cycles-pp.asm_sysvec_irq_work
0.00 +0.3 0.28 ± 42% perf-profile.children.cycles-pp.sysvec_irq_work
0.00 +0.3 0.28 ± 42% perf-profile.children.cycles-pp.__sysvec_irq_work
0.00 +0.3 0.28 ± 42% perf-profile.children.cycles-pp.irq_work_run
0.00 +0.3 0.28 ± 42% perf-profile.children.cycles-pp.irq_work_single
0.00 +0.3 0.29 ± 55% perf-profile.children.cycles-pp.rcu_eqs_exit
0.00 +0.3 0.31 ± 16% perf-profile.children.cycles-pp.tsc_verify_tsc_adjust
0.00 +0.3 0.32 ± 18% perf-profile.children.cycles-pp.arch_cpu_idle_enter
0.00 +0.3 0.33 ± 63% perf-profile.children.cycles-pp.cpuidle_governor_latency_req
0.20 ± 13% +0.4 0.56 ± 6% perf-profile.children.cycles-pp.read_tsc
0.67 ± 6% +0.4 1.06 ± 18% perf-profile.children.cycles-pp.rebalance_domains
0.00 +0.4 0.41 ± 16% perf-profile.children.cycles-pp.tick_nohz_irq_exit
0.00 +0.4 0.43 ± 23% perf-profile.children.cycles-pp.timekeeping_max_deferment
0.00 +0.5 0.48 ± 19% perf-profile.children.cycles-pp.get_next_timer_interrupt
0.00 +0.5 0.51 ± 45% perf-profile.children.cycles-pp.rcu_idle_exit
1.40 ± 15% +0.5 1.94 ± 27% perf-profile.children.cycles-pp.__softirqentry_text_start
0.10 ± 11% +0.5 0.65 ± 45% perf-profile.children.cycles-pp.idle_cpu
0.00 +0.6 0.59 ± 9% perf-profile.children.cycles-pp.hrtimer_next_event_without
0.00 +0.6 0.60 ± 14% perf-profile.children.cycles-pp.start_kernel
1.43 ± 9% +0.8 2.20 ± 29% perf-profile.children.cycles-pp.ktime_get
0.00 +0.9 0.94 ± 13% perf-profile.children.cycles-pp.native_irq_return_iret
1.92 ± 10% +1.0 2.92 ± 13% perf-profile.children.cycles-pp.__irq_exit_rcu
0.00 +1.2 1.20 ± 27% perf-profile.children.cycles-pp.tick_irq_enter
0.00 +1.2 1.23 ± 27% perf-profile.children.cycles-pp.irq_enter_rcu
0.00 +1.4 1.41 ± 12% perf-profile.children.cycles-pp.tick_nohz_next_event
0.00 +2.1 2.14 ± 6% perf-profile.children.cycles-pp.tick_nohz_get_sleep_length
0.00 +4.5 4.50 ± 18% perf-profile.children.cycles-pp.menu_select
0.00 +47.4 47.42 ± 8% perf-profile.children.cycles-pp.mwait_idle_with_hints
0.00 +47.6 47.62 ± 8% perf-profile.children.cycles-pp.intel_idle
0.00 +65.6 65.64 ± 3% perf-profile.children.cycles-pp.cpuidle_enter_state
0.00 +65.8 65.83 ± 3% perf-profile.children.cycles-pp.cpuidle_enter
0.00 +71.2 71.22 ± 4% perf-profile.children.cycles-pp.cpuidle_idle_call
0.01 ±200% +72.0 71.98 ± 4% perf-profile.children.cycles-pp.secondary_startup_64_no_verify
0.01 ±200% +72.0 71.98 ± 4% perf-profile.children.cycles-pp.cpu_startup_entry
0.01 ±200% +72.0 71.98 ± 4% perf-profile.children.cycles-pp.do_idle
66.95 -41.4 25.58 ± 13% perf-profile.self.cycles-pp.stress_lockbus
5.20 ± 24% -3.5 1.73 ± 55% perf-profile.self.cycles-pp.asm_sysvec_apic_timer_interrupt
1.56 ± 5% -1.3 0.27 ± 10% perf-profile.self.cycles-pp.__intel_pmu_enable_all
1.31 ± 13% -1.2 0.09 ± 36% perf-profile.self.cycles-pp.clockevents_program_event
1.44 ± 4% -0.9 0.52 ± 19% perf-profile.self.cycles-pp.native_sched_clock
0.45 ± 10% -0.4 0.07 ± 39% perf-profile.self.cycles-pp.__update_load_avg_cfs_rq
0.42 ± 10% -0.3 0.14 ± 30% perf-profile.self.cycles-pp.error_entry
0.61 ± 13% -0.3 0.34 ± 13% perf-profile.self.cycles-pp.ktime_get_update_offsets_now
0.36 ± 3% -0.2 0.12 ± 31% perf-profile.self.cycles-pp._raw_spin_lock_irq
0.12 ± 10% -0.1 0.04 ±105% perf-profile.self.cycles-pp.tick_sched_timer
0.02 ±125% +0.1 0.09 ± 35% perf-profile.self.cycles-pp.__softirqentry_text_start
0.00 +0.1 0.07 ± 27% perf-profile.self.cycles-pp.update_group_capacity
0.00 +0.1 0.08 ± 27% perf-profile.self.cycles-pp.menu_reflect
0.29 ± 8% +0.1 0.38 ± 11% perf-profile.self.cycles-pp._raw_spin_lock_irqsave
0.11 ± 9% +0.1 0.20 ± 27% perf-profile.self.cycles-pp._raw_spin_trylock
0.00 +0.1 0.09 ± 17% perf-profile.self.cycles-pp.sched_clock_cpu
0.18 ± 6% +0.1 0.27 ± 15% perf-profile.self.cycles-pp.update_sg_lb_stats
0.00 +0.1 0.10 ± 36% perf-profile.self.cycles-pp.rcu_eqs_enter
0.00 +0.1 0.10 ± 22% perf-profile.self.cycles-pp.update_sd_lb_stats
0.02 ±122% +0.1 0.13 ± 37% perf-profile.self.cycles-pp.load_balance
0.05 +0.1 0.16 ± 61% perf-profile.self.cycles-pp.rebalance_domains
0.00 +0.1 0.12 ± 26% perf-profile.self.cycles-pp.tick_nohz_get_sleep_length
0.00 +0.1 0.12 ± 51% perf-profile.self.cycles-pp.cpuidle_governor_latency_req
0.00 +0.1 0.12 ± 48% perf-profile.self.cycles-pp.calc_global_load_tick
0.00 +0.1 0.13 ± 27% perf-profile.self.cycles-pp.call_cpuidle
0.00 +0.1 0.13 ± 36% perf-profile.self.cycles-pp.cpuidle_enter
0.02 ±122% +0.1 0.16 ± 87% perf-profile.self.cycles-pp.sysvec_apic_timer_interrupt
0.00 +0.1 0.14 ± 24% perf-profile.self.cycles-pp.rcu_dynticks_inc
0.00 +0.1 0.14 ± 64% perf-profile.self.cycles-pp.get_next_timer_interrupt
0.00 +0.1 0.15 ± 25% perf-profile.self.cycles-pp.hrtimer_get_next_event
0.00 +0.2 0.15 ± 58% perf-profile.self.cycles-pp.__update_blocked_fair
0.00 +0.2 0.17 ± 4% perf-profile.self.cycles-pp.hrtimer_next_event_without
0.00 +0.2 0.18 ± 61% perf-profile.self.cycles-pp.timerqueue_iterate_next
0.00 +0.2 0.19 ± 60% perf-profile.self.cycles-pp.rcu_eqs_exit
0.00 +0.2 0.19 ± 42% perf-profile.self.cycles-pp.io_serial_in
0.00 +0.2 0.20 ± 36% perf-profile.self.cycles-pp.do_idle
0.00 +0.2 0.20 ± 19% perf-profile.self.cycles-pp.rcu_idle_exit
0.00 +0.2 0.21 ± 27% perf-profile.self.cycles-pp.intel_idle
0.00 +0.2 0.23 ± 69% perf-profile.self.cycles-pp.nr_iowait_cpu
0.00 +0.3 0.26 ± 15% perf-profile.self.cycles-pp.tsc_verify_tsc_adjust
0.37 ± 15% +0.3 0.64 ± 7% perf-profile.self.cycles-pp.lapic_next_deadline
0.00 +0.4 0.38 ± 18% perf-profile.self.cycles-pp.cpuidle_idle_call
0.16 ± 16% +0.4 0.54 ± 6% perf-profile.self.cycles-pp.read_tsc
0.00 +0.4 0.42 ± 37% perf-profile.self.cycles-pp.tick_nohz_next_event
0.00 +0.4 0.42 ± 26% perf-profile.self.cycles-pp.timekeeping_max_deferment
0.09 ± 12% +0.6 0.64 ± 45% perf-profile.self.cycles-pp.idle_cpu
0.00 +0.9 0.94 ± 13% perf-profile.self.cycles-pp.native_irq_return_iret
0.00 +1.9 1.93 ± 28% perf-profile.self.cycles-pp.menu_select
0.00 +2.5 2.48 ± 16% perf-profile.self.cycles-pp.cpuidle_enter_state
0.00 +47.4 47.40 ± 8% perf-profile.self.cycles-pp.mwait_idle_with_hints




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.


--
0-DAY CI Kernel Test Service
https://01.org/lkp