| 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
| ||
|
Date: Wed, 9 Aug 2023 10:34:15 +0800
From: kernel test robot <oliver.sang@...el.com>
To: Peter Zijlstra <peterz@...radead.org>
CC: <oe-lkp@...ts.linux.dev>, <lkp@...el.com>,
<linux-kernel@...r.kernel.org>, <x86@...nel.org>,
Ingo Molnar <mingo@...nel.org>, <ying.huang@...el.com>,
<feng.tang@...el.com>, <fengwei.yin@...el.com>,
<aubrey.li@...ux.intel.com>, <yu.c.chen@...el.com>,
<oliver.sang@...el.com>
Subject: [tip:sched/eevdf] [sched/smp] e8f331bcc2: ebizzy.throughput -4.2%
regression
Hello,
kernel test robot noticed a -4.2% regression of ebizzy.throughput on:
commit: e8f331bcc270354a803c2127c486190d33eac441 ("sched/smp: Use lag to simplify cross-runqueue placement")
https://git.kernel.org/cgit/linux/kernel/git/tip/tip.git sched/eevdf
testcase: ebizzy
test machine: 8 threads 1 sockets Intel(R) Core(TM) i7-4770K CPU @ 3.50GHz (Haswell) with 8G memory
parameters:
memory.high: 90%
memory.low: 50%
memory.max: max
pids.max: 10000
nr_threads: 200%
iterations: 100x
duration: 10s
cpufreq_governor: performance
If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <oliver.sang@...el.com>
| Closes: https://lore.kernel.org/oe-lkp/202308090904.e05ff18a-oliver.sang@intel.com
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.
=========================================================================================
compiler/cpufreq_governor/duration/iterations/kconfig/memory.high/memory.low/memory.max/nr_threads/pids.max/rootfs/tbox_group/testcase:
gcc-12/performance/10s/100x/x86_64-rhel-8.3/90%/50%/max/200%/10000/debian-11.1-x86_64-20220510.cgz/lkp-hsw-d04/ebizzy
commit:
76cae9dbe1 ("sched/fair: Commit to lag based placement")
e8f331bcc2 ("sched/smp: Use lag to simplify cross-runqueue placement")
76cae9dbe185b82a e8f331bcc270354a803c2127c48
---------------- ---------------------------
%stddev %change %stddev
\ | \
11.80 ± 34% +164.1% 31.17 ± 22% perf-c2c.DRAM.local
7697 ± 2% -11.7% 6796 ± 3% turbostat.C1E
2926 ± 2% +53.2% 4481 vmstat.system.cs
0.90 +0.2 1.09 ± 3% mpstat.cpu.all.irq%
0.02 ± 2% +0.0 0.03 ± 2% mpstat.cpu.all.soft%
90268 ± 6% +14.3% 103195 meminfo.Active
90188 ± 6% +14.3% 103115 meminfo.Active(anon)
92123 ± 6% +14.3% 105307 meminfo.Shmem
178465 -4.2% 170966 ebizzy.throughput
14333 -8.5% 13116 ebizzy.throughput.per_thread.max
9163 +5.4% 9655 ebizzy.throughput.per_thread.min
1337358 ± 3% +109.5% 2801631 ebizzy.time.involuntary_context_switches
0.01 ± 2% -10.7% 0.01 ± 3% ebizzy.time.sys
1787961 -4.3% 1710187 ebizzy.workload
22562 ± 6% +14.3% 25782 proc-vmstat.nr_active_anon
23048 ± 6% +14.2% 26325 proc-vmstat.nr_shmem
22562 ± 6% +14.3% 25782 proc-vmstat.nr_zone_active_anon
1162372 +7.0% 1243659 proc-vmstat.numa_hit
1162379 +7.0% 1243741 proc-vmstat.numa_local
21357 ± 7% +15.3% 24634 proc-vmstat.pgactivate
1243135 +4.9% 1303428 proc-vmstat.pgalloc_normal
1462811 +8.6% 1588942 proc-vmstat.pgfault
1211314 +4.8% 1269526 proc-vmstat.pgfree
50286 +14.8% 57742 proc-vmstat.pgreuse
422591 ± 33% -32.5% 285138 ± 13% sched_debug.cfs_rq:/.load.max
112707 ± 43% -43.6% 63567 ± 18% sched_debug.cfs_rq:/.load.stddev
85.31 ± 23% +28.5% 109.63 ± 4% sched_debug.cfs_rq:/.load_avg.min
578.22 ± 3% -18.6% 470.88 ± 8% sched_debug.cfs_rq:/.runnable_avg.stddev
1108 ± 7% +11.9% 1240 sched_debug.cfs_rq:/.util_avg.max
699.87 ± 22% +20.9% 846.47 sched_debug.cfs_rq:/.util_avg.min
525.64 ± 13% +20.5% 633.34 sched_debug.cfs_rq:/.util_est_enqueued.avg
154.93 ± 26% +53.1% 237.21 ± 13% sched_debug.cfs_rq:/.util_est_enqueued.min
593.14 ± 3% -17.1% 491.61 ± 7% sched_debug.cfs_rq:/ebizzy.1.runnable_avg.stddev
593.41 ± 3% -17.1% 491.68 ± 7% sched_debug.cfs_rq:/ebizzy.1.se->avg.runnable_avg.stddev
1104 ± 8% +11.7% 1233 sched_debug.cfs_rq:/ebizzy.1.se->avg.util_avg.max
25.83 ± 13% -98.4% 0.41 ± 12% sched_debug.cfs_rq:/ebizzy.1.spread.avg
129.52 ± 15% -98.4% 2.08 ± 6% sched_debug.cfs_rq:/ebizzy.1.spread.max
45.16 ± 15% -98.4% 0.74 ± 7% sched_debug.cfs_rq:/ebizzy.1.spread.stddev
1104 ± 8% +11.7% 1233 sched_debug.cfs_rq:/ebizzy.1.util_avg.max
0.02 ±111% +48.3% 0.03 ± 80% sched_debug.cfs_rq:/init.scope.avg_vruntime.stddev
0.02 ±111% +48.3% 0.03 ± 80% sched_debug.cfs_rq:/init.scope.min_vruntime.stddev
4802 ±115% +253.9% 16995 ± 36% sched_debug.cfs_rq:/init.scope.se->load.weight.stddev
0.06 ± 62% -88.6% 0.01 sched_debug.cfs_rq:/system.slice.h_nr_running.avg
0.25 ± 48% -76.2% 0.06 sched_debug.cfs_rq:/system.slice.h_nr_running.max
0.08 ± 60% -76.7% 0.02 sched_debug.cfs_rq:/system.slice.h_nr_running.stddev
67467 ± 62% -88.6% 7710 sched_debug.cfs_rq:/system.slice.load.avg
259059 ± 48% -76.2% 61680 sched_debug.cfs_rq:/system.slice.load.max
87483 ± 60% -76.7% 20399 sched_debug.cfs_rq:/system.slice.load.stddev
0.06 ± 62% -88.6% 0.01 sched_debug.cfs_rq:/system.slice.nr_running.avg
0.25 ± 48% -76.2% 0.06 sched_debug.cfs_rq:/system.slice.nr_running.max
0.08 ± 60% -76.7% 0.02 sched_debug.cfs_rq:/system.slice.nr_running.stddev
347341 ± 4% -19.0% 281444 ± 3% sched_debug.cpu.avg_idle.avg
717321 ± 7% -11.9% 631662 ± 6% sched_debug.cpu.avg_idle.max
7948 ± 12% +23.1% 9784 sched_debug.cpu.curr->pid.avg
6533 ± 22% +32.5% 8653 ± 4% sched_debug.cpu.curr->pid.min
262685 ± 2% +43.5% 376980 sched_debug.cpu.nr_switches.avg
454358 ± 14% +23.9% 563129 ± 10% sched_debug.cpu.nr_switches.max
115263 ± 13% +90.5% 219574 ± 10% sched_debug.cpu.nr_switches.min
0.28 ± 2% -8.6% 0.25 ± 5% sched_debug.cpu.nr_uninterruptible.avg
993.50 +8.4% 1076 perf-stat.i.MPKI
79679114 -12.0% 70154708 perf-stat.i.branch-instructions
0.90 ± 3% +0.1 0.97 perf-stat.i.branch-miss-rate%
1494675 ± 7% -17.1% 1239019 perf-stat.i.branch-misses
0.54 ± 10% +0.8 1.34 ± 12% perf-stat.i.cache-miss-rate%
1668656 ± 13% +161.5% 4364316 ± 13% perf-stat.i.cache-misses
3152 ± 4% +42.1% 4479 perf-stat.i.context-switches
86.40 +5.8% 91.44 perf-stat.i.cpi
8183 -2.2% 8000 perf-stat.i.cpu-clock
2.831e+10 -2.1% 2.772e+10 perf-stat.i.cpu-cycles
63.49 +10.2% 69.98 perf-stat.i.cpu-migrations
19545 ± 11% -63.6% 7110 ± 12% perf-stat.i.cycles-between-cache-misses
87729 ± 3% +12.9% 99006 ± 3% perf-stat.i.dTLB-load-misses
2.751e+09 -8.2% 2.526e+09 perf-stat.i.dTLB-loads
49937 ± 3% +49.5% 74663 ± 3% perf-stat.i.dTLB-store-misses
3.02e+09 -6.4% 2.827e+09 perf-stat.i.dTLB-stores
71.96 +4.8 76.78 perf-stat.i.iTLB-load-miss-rate%
4.283e+08 -11.1% 3.806e+08 perf-stat.i.instructions
9029 -22.6% 6986 ± 2% perf-stat.i.instructions-per-iTLB-miss
0.02 ± 9% -13.9% 0.01 ± 2% perf-stat.i.ipc
195.49 ± 13% +172.5% 532.79 ± 14% perf-stat.i.metric.K/sec
754.47 -4.7% 718.68 perf-stat.i.metric.M/sec
1486 +3.0% 1531 perf-stat.i.minor-faults
241878 ± 4% +102.3% 489275 ± 8% perf-stat.i.node-loads
1272828 ± 18% +189.5% 3684478 ± 15% perf-stat.i.node-stores
1486 +3.0% 1532 perf-stat.i.page-faults
8183 -2.2% 8000 perf-stat.i.task-clock
807.36 +6.2% 857.22 perf-stat.overall.MPKI
1.65 ± 2% +0.1 1.77 perf-stat.overall.branch-miss-rate%
0.51 ± 14% +0.8 1.34 ± 13% perf-stat.overall.cache-miss-rate%
70.20 +3.6% 72.72 perf-stat.overall.cpi
17545 ± 13% -63.1% 6465 ± 13% perf-stat.overall.cycles-between-cache-misses
0.00 ± 2% +0.0 0.00 ± 3% perf-stat.overall.dTLB-load-miss-rate%
0.00 ± 3% +0.0 0.00 ± 3% perf-stat.overall.dTLB-store-miss-rate%
7875 ± 6% -17.9% 6463 ± 2% perf-stat.overall.instructions-per-iTLB-miss
0.01 -3.5% 0.01 perf-stat.overall.ipc
72763224 -3.5% 70199360 perf-stat.ps.branch-instructions
1607278 ± 14% +171.3% 4360199 ± 13% perf-stat.ps.cache-misses
3.182e+08 +2.6% 3.264e+08 perf-stat.ps.cache-references
2890 ± 2% +55.0% 4479 perf-stat.ps.context-switches
61.68 +13.3% 69.91 perf-stat.ps.cpu-migrations
83192 ± 2% +19.0% 98961 ± 3% perf-stat.ps.dTLB-load-misses
2.689e+09 -6.2% 2.523e+09 perf-stat.ps.dTLB-loads
46972 ± 3% +58.9% 74634 ± 3% perf-stat.ps.dTLB-store-misses
2.956e+09 -4.4% 2.824e+09 perf-stat.ps.dTLB-stores
50294 ± 7% +17.2% 58948 ± 2% perf-stat.ps.iTLB-load-misses
20231 +7.1% 21671 ± 2% perf-stat.ps.iTLB-loads
3.941e+08 -3.4% 3.808e+08 perf-stat.ps.instructions
1404 +9.0% 1530 perf-stat.ps.minor-faults
227772 ± 5% +114.7% 489108 ± 8% perf-stat.ps.node-loads
1231787 ± 18% +198.8% 3680693 ± 15% perf-stat.ps.node-stores
1404 +9.0% 1530 perf-stat.ps.page-faults
3.963e+11 -3.5% 3.823e+11 perf-stat.total.instructions
49.68 ± 41% -95.0% 2.47 ± 36% perf-sched.sch_delay.avg.ms.__cond_resched.stop_one_cpu.sched_exec.bprm_execve.part
1.55 ± 78% -78.6% 0.33 ± 91% perf-sched.sch_delay.avg.ms.devkmsg_read.vfs_read.ksys_read.do_syscall_64
0.04 ± 31% -62.2% 0.01 ± 2% perf-sched.sch_delay.avg.ms.do_nanosleep.hrtimer_nanosleep.common_nsleep.__x64_sys_clock_nanosleep
12.48 ± 83% -92.7% 0.92 ± 35% perf-sched.sch_delay.avg.ms.do_task_dead.do_exit.do_group_exit.__x64_sys_exit_group.do_syscall_64
0.00 ± 57% -100.0% 0.00 perf-sched.sch_delay.avg.ms.exit_to_user_mode_loop.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_sysvec_apic_timer_interrupt
9.85 ±124% -99.9% 0.01 ±173% perf-sched.sch_delay.avg.ms.schedule_hrtimeout_range_clock.do_poll.constprop.0.do_sys_poll
0.03 ± 83% +255.5% 0.11 ± 50% perf-sched.sch_delay.avg.ms.schedule_preempt_disabled.rwsem_down_write_slowpath.down_write_killable.vm_mmap_pgoff
7.43 ±122% -99.8% 0.01 ± 5% perf-sched.sch_delay.avg.ms.schedule_timeout.io_schedule_timeout.__wait_for_common.blk_execute_rq
270.83 ± 28% -97.9% 5.58 ± 22% perf-sched.sch_delay.max.ms.__cond_resched.stop_one_cpu.sched_exec.bprm_execve.part
44.68 ±107% -93.6% 2.85 ± 57% perf-sched.sch_delay.max.ms.devkmsg_read.vfs_read.ksys_read.do_syscall_64
256.51 ± 35% -97.0% 7.74 ± 23% perf-sched.sch_delay.max.ms.do_nanosleep.hrtimer_nanosleep.common_nsleep.__x64_sys_clock_nanosleep
406.43 ± 84% -98.4% 6.67 ± 21% perf-sched.sch_delay.max.ms.do_task_dead.do_exit.do_group_exit.__x64_sys_exit_group.do_syscall_64
94.12 ± 86% -95.5% 4.24 ± 10% perf-sched.sch_delay.max.ms.exit_to_user_mode_loop.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_sysvec_apic_timer_interrupt
257.42 ± 59% -98.1% 4.83 ± 22% perf-sched.sch_delay.max.ms.exit_to_user_mode_loop.exit_to_user_mode_prepare.syscall_exit_to_user_mode.do_syscall_64
67.90 ±100% -92.8% 4.91 ± 29% perf-sched.sch_delay.max.ms.pipe_read.vfs_read.ksys_read.do_syscall_64
339.82 ±138% -99.8% 0.51 ±217% perf-sched.sch_delay.max.ms.schedule_hrtimeout_range_clock.do_poll.constprop.0.do_sys_poll
0.18 ± 71% +102.3% 0.37 ± 3% perf-sched.sch_delay.max.ms.schedule_preempt_disabled.rwsem_down_write_slowpath.down_write_killable.vm_mmap_pgoff
0.01 ± 10% +23.9% 0.02 ± 9% perf-sched.sch_delay.max.ms.schedule_timeout.__wait_for_common.__wait_rcu_gp.synchronize_rcu
136.20 ±122% -100.0% 0.02 ± 39% perf-sched.sch_delay.max.ms.schedule_timeout.io_schedule_timeout.__wait_for_common.blk_execute_rq
3.83 ±125% -99.5% 0.02 ± 27% perf-sched.sch_delay.max.ms.smpboot_thread_fn.kthread.ret_from_fork.ret_from_fork_asm
0.06 ± 41% -85.1% 0.01 ± 5% perf-sched.total_sch_delay.average.ms
579.34 ± 59% -98.1% 11.03 ± 23% perf-sched.total_sch_delay.max.ms
2.82 ± 10% -25.6% 2.10 perf-sched.total_wait_and_delay.average.ms
107770 ± 9% +22.7% 132242 perf-sched.total_wait_and_delay.count.ms
3725 ± 22% -34.4% 2443 ± 23% perf-sched.total_wait_and_delay.max.ms
2.76 ± 9% -24.3% 2.09 perf-sched.total_wait_time.average.ms
3725 ± 22% -34.4% 2443 ± 23% perf-sched.total_wait_time.max.ms
132.33 ± 9% -36.7% 83.74 perf-sched.wait_and_delay.avg.ms.__cond_resched.__wait_for_common.affine_move_task.__set_cpus_allowed_ptr.__sched_setaffinity
50.05 ± 41% -96.8% 1.59 ±105% perf-sched.wait_and_delay.avg.ms.__cond_resched.stop_one_cpu.sched_exec.bprm_execve.part
0.14 ± 20% -30.3% 0.10 ± 11% perf-sched.wait_and_delay.avg.ms.do_nanosleep.hrtimer_nanosleep.common_nsleep.__x64_sys_clock_nanosleep
27.91 ± 41% -89.9% 2.82 ± 49% perf-sched.wait_and_delay.avg.ms.do_wait.kernel_wait4.__do_sys_wait4.do_syscall_64
10.18 ± 50% -86.7% 1.36 ± 66% perf-sched.wait_and_delay.avg.ms.exit_to_user_mode_loop.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_sysvec_reschedule_ipi
334.11 ± 15% -24.8% 251.12 ± 4% perf-sched.wait_and_delay.avg.ms.schedule_hrtimeout_range_clock.do_poll.constprop.0.do_sys_poll
16.36 ±114% +206.1% 50.09 ± 7% perf-sched.wait_and_delay.avg.ms.schedule_hrtimeout_range_clock.do_select.core_sys_select.kern_select
529.57 ± 5% -17.8% 435.23 ± 8% perf-sched.wait_and_delay.avg.ms.smpboot_thread_fn.kthread.ret_from_fork.ret_from_fork_asm
33.00 ± 12% +47.0% 48.50 perf-sched.wait_and_delay.count.__cond_resched.__wait_for_common.affine_move_task.__set_cpus_allowed_ptr.__sched_setaffinity
50141 ± 9% +20.2% 60292 perf-sched.wait_and_delay.count.do_nanosleep.hrtimer_nanosleep.common_nsleep.__x64_sys_clock_nanosleep
56.60 ± 9% -29.3% 40.00 ± 45% perf-sched.wait_and_delay.count.do_wait.kernel_wait4.__do_sys_wait4.do_syscall_64
52857 ± 9% +27.7% 67519 perf-sched.wait_and_delay.count.exit_to_user_mode_loop.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_sysvec_apic_timer_interrupt
921.80 ± 29% -40.6% 548.00 ± 24% perf-sched.wait_and_delay.count.exit_to_user_mode_loop.exit_to_user_mode_prepare.syscall_exit_to_user_mode.do_syscall_64
47.20 ± 8% +21.5% 57.33 perf-sched.wait_and_delay.count.schedule_hrtimeout_range_clock.do_poll.constprop.0.do_sys_poll
104.00 ± 11% +16.2% 120.83 ± 6% perf-sched.wait_and_delay.count.smpboot_thread_fn.kthread.ret_from_fork.ret_from_fork_asm
270.83 ± 28% -98.7% 3.39 ±100% perf-sched.wait_and_delay.max.ms.__cond_resched.stop_one_cpu.sched_exec.bprm_execve.part
473.76 ± 29% -90.3% 46.10 ± 51% perf-sched.wait_and_delay.max.ms.do_wait.kernel_wait4.__do_sys_wait4.do_syscall_64
476.59 ± 19% -86.0% 66.65 ±181% perf-sched.wait_and_delay.max.ms.exit_to_user_mode_loop.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_sysvec_reschedule_ipi
2013 ± 18% -49.4% 1018 perf-sched.wait_and_delay.max.ms.pipe_read.vfs_read.ksys_read.do_syscall_64
1225 ± 15% -18.3% 1001 perf-sched.wait_and_delay.max.ms.schedule_hrtimeout_range_clock.do_poll.constprop.0.do_sys_poll
192.72 ±116% +207.3% 592.15 ± 8% perf-sched.wait_and_delay.max.ms.schedule_hrtimeout_range_clock.do_select.core_sys_select.kern_select
132.20 ± 9% -36.7% 83.63 perf-sched.wait_time.avg.ms.__cond_resched.__wait_for_common.affine_move_task.__set_cpus_allowed_ptr.__sched_setaffinity
27.28 ± 44% -88.8% 3.05 ± 25% perf-sched.wait_time.avg.ms.do_wait.kernel_wait4.__do_sys_wait4.do_syscall_64
10.10 ± 50% -83.3% 1.69 ± 40% perf-sched.wait_time.avg.ms.exit_to_user_mode_loop.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_sysvec_reschedule_ipi
324.26 ± 12% -22.6% 251.11 ± 4% perf-sched.wait_time.avg.ms.schedule_hrtimeout_range_clock.do_poll.constprop.0.do_sys_poll
16.48 ±113% +202.0% 49.76 ± 8% perf-sched.wait_time.avg.ms.schedule_hrtimeout_range_clock.do_select.core_sys_select.kern_select
0.01 ± 50% +39.9% 0.01 ± 4% perf-sched.wait_time.avg.ms.schedule_timeout.__wait_for_common.ata_exec_internal_sg.ata_exec_internal
529.52 ± 5% -17.8% 435.22 ± 8% perf-sched.wait_time.avg.ms.smpboot_thread_fn.kthread.ret_from_fork.ret_from_fork_asm
446.14 ± 37% -89.4% 47.27 ± 45% perf-sched.wait_time.max.ms.do_wait.kernel_wait4.__do_sys_wait4.do_syscall_64
476.59 ± 19% -85.8% 67.84 ±176% perf-sched.wait_time.max.ms.exit_to_user_mode_loop.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_sysvec_reschedule_ipi
2013 ± 18% -49.5% 1016 perf-sched.wait_time.max.ms.pipe_read.vfs_read.ksys_read.do_syscall_64
193.14 ±116% +206.6% 592.14 ± 8% perf-sched.wait_time.max.ms.schedule_hrtimeout_range_clock.do_select.core_sys_select.kern_select
34.48 ± 13% -11.7 22.77 ± 31% perf-profile.calltrace.cycles-pp.ata_scsi_error.scsi_error_handler.kthread.ret_from_fork.ret_from_fork_asm
34.48 ± 13% -11.7 22.77 ± 31% perf-profile.calltrace.cycles-pp.ata_scsi_port_error_handler.ata_scsi_error.scsi_error_handler.kthread.ret_from_fork
34.45 ± 13% -11.7 22.74 ± 31% perf-profile.calltrace.cycles-pp.ahci_error_handler.ata_scsi_port_error_handler.ata_scsi_error.scsi_error_handler.kthread
34.45 ± 13% -11.7 22.74 ± 31% perf-profile.calltrace.cycles-pp.sata_pmp_error_handler.ahci_error_handler.ata_scsi_port_error_handler.ata_scsi_error.scsi_error_handler
34.45 ± 13% -11.7 22.74 ± 31% perf-profile.calltrace.cycles-pp.sata_pmp_eh_recover.sata_pmp_error_handler.ahci_error_handler.ata_scsi_port_error_handler.ata_scsi_error
34.45 ± 13% -11.7 22.74 ± 31% perf-profile.calltrace.cycles-pp.ata_eh_recover.sata_pmp_eh_recover.sata_pmp_error_handler.ahci_error_handler.ata_scsi_port_error_handler
35.13 ± 13% -11.7 23.43 ± 30% perf-profile.calltrace.cycles-pp.ret_from_fork_asm
35.13 ± 13% -11.7 23.43 ± 30% perf-profile.calltrace.cycles-pp.ret_from_fork.ret_from_fork_asm
35.13 ± 13% -11.7 23.43 ± 30% perf-profile.calltrace.cycles-pp.kthread.ret_from_fork.ret_from_fork_asm
34.48 ± 13% -11.7 22.79 ± 31% perf-profile.calltrace.cycles-pp.scsi_error_handler.kthread.ret_from_fork.ret_from_fork_asm
30.68 ± 15% -10.7 20.00 ± 35% perf-profile.calltrace.cycles-pp.serial8250_console_write.console_flush_all.console_unlock.vprintk_emit._printk
25.72 ± 14% -8.1 17.64 ± 22% perf-profile.calltrace.cycles-pp.wait_for_lsr.serial8250_console_write.console_flush_all.console_unlock.vprintk_emit
16.01 ± 19% -7.3 8.67 ± 23% perf-profile.calltrace.cycles-pp.vprintk_emit._printk.ata_dev_set_mode.ata_do_set_mode.ata_set_mode
16.01 ± 19% -7.3 8.67 ± 23% perf-profile.calltrace.cycles-pp._printk.ata_dev_set_mode.ata_do_set_mode.ata_set_mode.ata_eh_recover
16.02 ± 19% -7.3 8.68 ± 23% perf-profile.calltrace.cycles-pp.ata_set_mode.ata_eh_recover.sata_pmp_eh_recover.sata_pmp_error_handler.ahci_error_handler
16.02 ± 19% -7.3 8.68 ± 23% perf-profile.calltrace.cycles-pp.ata_do_set_mode.ata_set_mode.ata_eh_recover.sata_pmp_eh_recover.sata_pmp_error_handler
16.02 ± 19% -7.3 8.68 ± 23% perf-profile.calltrace.cycles-pp.ata_dev_set_mode.ata_do_set_mode.ata_set_mode.ata_eh_recover.sata_pmp_eh_recover
15.99 ± 19% -7.3 8.65 ± 23% perf-profile.calltrace.cycles-pp.console_unlock.vprintk_emit._printk.ata_dev_set_mode.ata_do_set_mode
15.99 ± 19% -7.3 8.65 ± 23% perf-profile.calltrace.cycles-pp.console_flush_all.console_unlock.vprintk_emit._printk.ata_dev_set_mode
16.27 ± 10% -5.7 10.59 ± 23% perf-profile.calltrace.cycles-pp.io_serial_in.wait_for_lsr.serial8250_console_write.console_flush_all.console_unlock
0.67 ± 18% +0.5 1.14 ± 32% perf-profile.calltrace.cycles-pp.update_curr.task_tick_fair.scheduler_tick.update_process_times.tick_sched_handle
0.87 ± 26% +0.6 1.50 ± 17% perf-profile.calltrace.cycles-pp.asm_exc_page_fault
0.66 ± 57% +0.7 1.33 ± 18% perf-profile.calltrace.cycles-pp.exc_page_fault.asm_exc_page_fault
0.63 ± 55% +0.7 1.33 ± 18% perf-profile.calltrace.cycles-pp.do_user_addr_fault.exc_page_fault.asm_exc_page_fault
1.61 ± 12% +0.8 2.44 ± 26% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe
1.61 ± 12% +0.8 2.44 ± 26% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe
1.05 ± 38% +1.1 2.12 ± 12% perf-profile.calltrace.cycles-pp.update_load_avg.task_tick_fair.scheduler_tick.update_process_times.tick_sched_handle
0.23 ±123% +1.1 1.36 ± 33% perf-profile.calltrace.cycles-pp.__schedule.schedule.exit_to_user_mode_loop.exit_to_user_mode_prepare.irqentry_exit_to_user_mode
0.36 ± 83% +1.1 1.50 ± 34% perf-profile.calltrace.cycles-pp.exit_to_user_mode_loop.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_sysvec_apic_timer_interrupt.thread_run
0.26 ±123% +1.2 1.43 ± 32% perf-profile.calltrace.cycles-pp.schedule.exit_to_user_mode_loop.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_sysvec_apic_timer_interrupt
1.63 ± 16% +1.3 2.90 ± 23% perf-profile.calltrace.cycles-pp.irqentry_exit_to_user_mode.asm_sysvec_apic_timer_interrupt.thread_run.start_thread
0.43 ± 81% +1.5 1.98 ± 23% perf-profile.calltrace.cycles-pp.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_sysvec_apic_timer_interrupt.thread_run.start_thread
2.53 ± 16% +1.7 4.25 ± 13% perf-profile.calltrace.cycles-pp.task_tick_fair.scheduler_tick.update_process_times.tick_sched_handle.tick_sched_timer
3.65 ± 10% +2.4 6.00 ± 12% perf-profile.calltrace.cycles-pp.scheduler_tick.update_process_times.tick_sched_handle.tick_sched_timer.__hrtimer_run_queues
4.67 ± 9% +3.4 8.04 ± 14% perf-profile.calltrace.cycles-pp.update_process_times.tick_sched_handle.tick_sched_timer.__hrtimer_run_queues.hrtimer_interrupt
4.77 ± 9% +3.4 8.22 ± 15% perf-profile.calltrace.cycles-pp.tick_sched_handle.tick_sched_timer.__hrtimer_run_queues.hrtimer_interrupt.__sysvec_apic_timer_interrupt
5.32 ± 11% +3.7 9.06 ± 14% perf-profile.calltrace.cycles-pp.tick_sched_timer.__hrtimer_run_queues.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt
7.62 ± 9% +4.0 11.66 ± 12% perf-profile.calltrace.cycles-pp.__hrtimer_run_queues.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt
9.14 ± 7% +4.3 13.47 ± 11% perf-profile.calltrace.cycles-pp.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.thread_run
9.24 ± 7% +4.5 13.74 ± 11% perf-profile.calltrace.cycles-pp.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.thread_run.start_thread
10.73 ± 6% +4.9 15.66 ± 9% perf-profile.calltrace.cycles-pp.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.thread_run.start_thread
12.78 ± 7% +6.3 19.08 ± 10% perf-profile.calltrace.cycles-pp.asm_sysvec_apic_timer_interrupt.thread_run.start_thread
34.01 ± 9% +6.6 40.58 ± 9% perf-profile.calltrace.cycles-pp.start_thread
34.33 ± 8% +6.9 41.28 ± 9% perf-profile.calltrace.cycles-pp.thread_run.start_thread
34.48 ± 13% -11.7 22.77 ± 31% perf-profile.children.cycles-pp.ata_scsi_error
34.48 ± 13% -11.7 22.77 ± 31% perf-profile.children.cycles-pp.ata_scsi_port_error_handler
34.45 ± 13% -11.7 22.74 ± 31% perf-profile.children.cycles-pp.ahci_error_handler
34.45 ± 13% -11.7 22.74 ± 31% perf-profile.children.cycles-pp.sata_pmp_error_handler
34.45 ± 13% -11.7 22.74 ± 31% perf-profile.children.cycles-pp.sata_pmp_eh_recover
34.45 ± 13% -11.7 22.74 ± 31% perf-profile.children.cycles-pp.ata_eh_recover
35.13 ± 13% -11.7 23.43 ± 30% perf-profile.children.cycles-pp.kthread
35.14 ± 13% -11.7 23.45 ± 30% perf-profile.children.cycles-pp.ret_from_fork_asm
35.14 ± 13% -11.7 23.45 ± 30% perf-profile.children.cycles-pp.ret_from_fork
34.48 ± 13% -11.7 22.79 ± 31% perf-profile.children.cycles-pp.scsi_error_handler
34.11 ± 13% -11.6 22.52 ± 31% perf-profile.children.cycles-pp._printk
36.49 ± 12% -10.9 25.57 ± 21% perf-profile.children.cycles-pp.console_unlock
36.49 ± 12% -10.9 25.57 ± 21% perf-profile.children.cycles-pp.console_flush_all
36.51 ± 12% -10.4 26.08 ± 20% perf-profile.children.cycles-pp.vprintk_emit
32.50 ± 14% -9.8 22.69 ± 24% perf-profile.children.cycles-pp.serial8250_console_write
32.10 ± 14% -9.8 22.32 ± 23% perf-profile.children.cycles-pp.wait_for_lsr
16.02 ± 19% -7.3 8.68 ± 23% perf-profile.children.cycles-pp.ata_set_mode
16.02 ± 19% -7.3 8.68 ± 23% perf-profile.children.cycles-pp.ata_do_set_mode
16.02 ± 19% -7.3 8.68 ± 23% perf-profile.children.cycles-pp.ata_dev_set_mode
20.25 ± 12% -6.7 13.58 ± 23% perf-profile.children.cycles-pp.io_serial_in
6.29 ± 15% -1.8 4.52 ± 30% perf-profile.children.cycles-pp.wait_for_xmitr
2.88 ± 7% -1.2 1.66 ± 42% perf-profile.children.cycles-pp.intel_fbdev_defio_imageblit
2.80 ± 5% -1.2 1.62 ± 44% perf-profile.children.cycles-pp.cfb_imageblit
2.74 ± 6% -1.1 1.60 ± 43% perf-profile.children.cycles-pp.fast_imageblit
0.05 ± 84% +0.1 0.16 ± 36% perf-profile.children.cycles-pp.release_task
0.19 ± 44% +0.1 0.31 ± 44% perf-profile.children.cycles-pp.printk_get_next_message
0.15 ± 56% +0.1 0.28 ± 27% perf-profile.children.cycles-pp.wait4
0.06 ± 87% +0.1 0.19 ± 54% perf-profile.children.cycles-pp.wp_page_copy
0.02 ±200% +0.1 0.15 ± 49% perf-profile.children.cycles-pp.os_xsave
0.06 ± 87% +0.2 0.21 ± 35% perf-profile.children.cycles-pp.release_pages
0.12 ± 69% +0.2 0.30 ± 29% perf-profile.children.cycles-pp.exec_mmap
0.26 ± 26% +0.2 0.44 ± 40% perf-profile.children.cycles-pp.__open64_nocancel
0.05 ±135% +0.2 0.24 ± 34% perf-profile.children.cycles-pp.error_return
0.09 ± 33% +0.2 0.28 ± 35% perf-profile.children.cycles-pp.__libc_read
0.16 ± 58% +0.2 0.36 ± 51% perf-profile.children.cycles-pp.load_balance
0.08 ±111% +0.2 0.31 ± 59% perf-profile.children.cycles-pp.alloc_mem
0.21 ± 60% +0.2 0.44 ± 29% perf-profile.children.cycles-pp.seq_read_iter
0.15 ± 52% +0.2 0.38 ± 22% perf-profile.children.cycles-pp.begin_new_exec
0.26 ± 59% +0.2 0.51 ± 29% perf-profile.children.cycles-pp.account_user_time
0.03 ±200% +0.3 0.28 ± 62% perf-profile.children.cycles-pp.__switch_to_asm
0.19 ± 35% +0.3 0.48 ± 37% perf-profile.children.cycles-pp.prepare_task_switch
0.05 ± 81% +0.3 0.34 ± 32% perf-profile.children.cycles-pp.restore_fpregs_from_fpstate
0.36 ± 30% +0.3 0.65 ± 39% perf-profile.children.cycles-pp.__update_load_avg_se
0.57 ± 26% +0.3 0.87 ± 36% perf-profile.children.cycles-pp.path_openat
0.29 ± 43% +0.3 0.61 ± 20% perf-profile.children.cycles-pp.___perf_sw_event
0.57 ± 26% +0.3 0.90 ± 36% perf-profile.children.cycles-pp.do_filp_open
0.56 ± 20% +0.4 0.92 ± 31% perf-profile.children.cycles-pp.__mmput
0.56 ± 20% +0.4 0.92 ± 31% perf-profile.children.cycles-pp.exit_mmap
0.08 ± 12% +0.4 0.46 ± 22% perf-profile.children.cycles-pp.switch_fpu_return
0.19 ± 61% +0.4 0.57 ± 57% perf-profile.children.cycles-pp.rcu_sched_clock_irq
0.98 ± 18% +0.5 1.44 ± 23% perf-profile.children.cycles-pp.update_curr
1.38 ± 20% +0.8 2.20 ± 26% perf-profile.children.cycles-pp.schedule
1.29 ± 14% +0.9 2.18 ± 26% perf-profile.children.cycles-pp.__schedule
0.94 ± 20% +1.0 1.95 ± 31% perf-profile.children.cycles-pp.exit_to_user_mode_loop
1.21 ± 35% +1.2 2.41 ± 11% perf-profile.children.cycles-pp.update_load_avg
2.57 ± 20% +1.3 3.92 ± 22% perf-profile.children.cycles-pp.irqentry_exit_to_user_mode
1.11 ± 18% +1.4 2.51 ± 22% perf-profile.children.cycles-pp.exit_to_user_mode_prepare
2.56 ± 15% +1.8 4.32 ± 13% perf-profile.children.cycles-pp.task_tick_fair
3.70 ± 11% +2.4 6.07 ± 12% perf-profile.children.cycles-pp.scheduler_tick
4.76 ± 9% +3.4 8.16 ± 14% perf-profile.children.cycles-pp.update_process_times
4.84 ± 9% +3.5 8.34 ± 15% perf-profile.children.cycles-pp.tick_sched_handle
5.41 ± 10% +3.8 9.20 ± 14% perf-profile.children.cycles-pp.tick_sched_timer
7.78 ± 9% +4.1 11.87 ± 12% perf-profile.children.cycles-pp.__hrtimer_run_queues
9.52 ± 16% +4.1 13.63 ± 30% perf-profile.children.cycles-pp.entry_SYSCALL_64_after_hwframe
9.48 ± 16% +4.1 13.61 ± 30% perf-profile.children.cycles-pp.do_syscall_64
9.30 ± 8% +4.4 13.66 ± 11% perf-profile.children.cycles-pp.hrtimer_interrupt
9.38 ± 7% +4.6 13.93 ± 11% perf-profile.children.cycles-pp.__sysvec_apic_timer_interrupt
10.93 ± 6% +5.0 15.90 ± 9% perf-profile.children.cycles-pp.sysvec_apic_timer_interrupt
12.97 ± 7% +6.4 19.40 ± 10% perf-profile.children.cycles-pp.asm_sysvec_apic_timer_interrupt
34.01 ± 9% +6.6 40.58 ± 9% perf-profile.children.cycles-pp.start_thread
33.98 ± 9% +6.6 40.56 ± 9% perf-profile.children.cycles-pp.thread_run
20.23 ± 12% -6.6 13.58 ± 23% perf-profile.self.cycles-pp.io_serial_in
2.74 ± 6% -1.1 1.60 ± 43% perf-profile.self.cycles-pp.fast_imageblit
0.03 ±124% +0.1 0.12 ± 51% perf-profile.self.cycles-pp.switch_fpu_return
0.06 ± 98% +0.1 0.16 ± 48% perf-profile.self.cycles-pp.__sysvec_apic_timer_interrupt
0.02 ±200% +0.1 0.15 ± 49% perf-profile.self.cycles-pp.os_xsave
0.07 ± 95% +0.2 0.22 ± 49% perf-profile.self.cycles-pp.rep_movs_alternative
0.10 ± 72% +0.2 0.31 ± 23% perf-profile.self.cycles-pp.task_tick_fair
0.08 ±111% +0.2 0.31 ± 59% perf-profile.self.cycles-pp.alloc_mem
0.03 ±200% +0.3 0.28 ± 62% perf-profile.self.cycles-pp.__switch_to_asm
0.22 ± 17% +0.3 0.48 ± 45% perf-profile.self.cycles-pp.ktime_get
0.26 ± 53% +0.3 0.53 ± 25% perf-profile.self.cycles-pp.___perf_sw_event
0.05 ± 81% +0.3 0.34 ± 32% perf-profile.self.cycles-pp.restore_fpregs_from_fpstate
0.14 ± 93% +0.4 0.49 ± 38% perf-profile.self.cycles-pp.__schedule
0.54 ± 32% +0.4 0.92 ± 32% perf-profile.self.cycles-pp.update_curr
0.23 ± 43% +0.4 0.62 ± 35% perf-profile.self.cycles-pp.hrtimer_interrupt
0.45 ± 57% +0.7 1.14 ± 30% perf-profile.self.cycles-pp.update_load_avg
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://github.com/intel/lkp-tests/wiki
View attachment "config-6.5.0-rc2-00022-ge8f331bcc270" of type "text/plain" (159815 bytes)
View attachment "job-script" of type "text/plain" (9141 bytes)
View attachment "job.yaml" of type "text/plain" (6524 bytes)
View attachment "reproduce" of type "text/plain" (3005 bytes)
Powered by blists - more mailing lists