| 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: Sun, 8 Dec 2019 23:39:49 +0800
From: kernel test robot <rong.a.chen@...el.com>
To: David Howells <dhowells@...hat.com>
Cc: LKML <linux-kernel@...r.kernel.org>,
Linus Torvalds <torvalds@...ux-foundation.org>,
lkp@...ts.01.org
Subject: [pipe] 3c0edea9b2: lmbench3.PIPE.bandwidth.MB/sec -17.0% regression
Greeting,
FYI, we noticed a -17.0% regression of lmbench3.PIPE.bandwidth.MB/sec due to commit:
commit: 3c0edea9b29f9be6c093f236f762202b30ac9431 ("pipe: Remove sync on wake_ups")
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master
in testcase: lmbench3
on test machine: 16 threads Intel(R) Xeon(R) CPU D-1541 @ 2.10GHz with 48G memory
with following parameters:
test_memory_size: 50%
nr_threads: 100%
mode: development
test: PIPE
cpufreq_governor: performance
ucode: 0x7000019
test-url: http://www.bitmover.com/lmbench/
If you fix the issue, kindly add following tag
Reported-by: kernel test robot <rong.a.chen@...el.com>
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_threads/rootfs/tbox_group/test/test_memory_size/testcase/ucode:
gcc-7/performance/x86_64-rhel-7.6/development/100%/debian-x86_64-2019-11-14.cgz/lkp-bdw-de1/PIPE/50%/lmbench3/0x7000019
commit:
cefa80ced5 ("pipe: Increase the writer-wakeup threshold to reduce context-switch count")
3c0edea9b2 ("pipe: Remove sync on wake_ups")
cefa80ced57a2917 3c0edea9b29f9be6c093f236f76
---------------- ---------------------------
%stddev %change %stddev
\ | \
21204 -17.0% 17604 lmbench3.PIPE.bandwidth.MB/sec
14.06 +91.9% 26.99 lmbench3.PIPE.latency.us
134.47 +25.9% 169.28 ± 7% lmbench3.time.elapsed_time
134.47 +25.9% 169.28 ± 7% lmbench3.time.elapsed_time.max
3334392 ± 9% -78.2% 725930 ± 4% lmbench3.time.involuntary_context_switches
1230 ± 2% +25.4% 1542 lmbench3.time.system_time
66167960 ± 2% -24.0% 50257054 lmbench3.time.voluntary_context_switches
5.83 -1.0 4.80 mpstat.cpu.all.usr%
539839 ± 2% -17.4% 445879 ± 6% vmstat.system.cs
39.29 -2.5% 38.31 boot-time.boot
563.99 -2.7% 548.62 boot-time.idle
6437 -1.7% 6330 proc-vmstat.nr_mapped
683646 +5.9% 724211 ± 2% proc-vmstat.pgfault
1330273 ± 37% +1281.6% 18379318 turbostat.C1
0.74 ± 65% +2.6 3.37 ± 6% turbostat.C1%
4392819 ± 5% +31.9% 5792490 ± 5% turbostat.IRQ
16257279 ± 65% +462.7% 91474668 cpuidle.C1.time
1331447 ± 37% +1280.5% 18380948 cpuidle.C1.usage
5997091 ± 32% +266.2% 21964255 cpuidle.POLL.time
3142700 ± 5% +121.5% 6962552 cpuidle.POLL.usage
0.00 +9e+11% 8988 ±111% sched_debug.cfs_rq:/.MIN_vruntime.avg
0.00 +7.7e+12% 76535 ± 86% sched_debug.cfs_rq:/.MIN_vruntime.max
0.00 ± 15% +1.2e+28% 22897 ± 92% sched_debug.cfs_rq:/.MIN_vruntime.stddev
17.75 ± 16% +66.7% 29.58 ± 20% sched_debug.cfs_rq:/.load_avg.min
155.25 ± 8% -22.5% 120.28 ± 7% sched_debug.cfs_rq:/.load_avg.stddev
0.00 +9e+11% 8988 ±111% sched_debug.cfs_rq:/.max_vruntime.avg
0.00 +7.7e+12% 76535 ± 86% sched_debug.cfs_rq:/.max_vruntime.max
0.00 ± 15% +1.2e+28% 22897 ± 92% sched_debug.cfs_rq:/.max_vruntime.stddev
63.45 ± 22% -41.5% 37.15 ± 32% sched_debug.cfs_rq:/.runnable_load_avg.stddev
247.25 ± 20% +93.6% 478.77 ± 16% sched_debug.cfs_rq:/.util_avg.min
380.01 ± 9% -27.2% 276.67 ± 7% sched_debug.cfs_rq:/.util_avg.stddev
342.89 ± 3% -15.9% 288.48 ± 6% sched_debug.cfs_rq:/.util_est_enqueued.avg
997.25 ± 3% -23.2% 766.29 ± 15% sched_debug.cfs_rq:/.util_est_enqueued.max
294.50 ± 4% -30.6% 204.35 ± 14% sched_debug.cfs_rq:/.util_est_enqueued.stddev
534028 ± 9% -37.7% 332865 ± 17% sched_debug.cpu.avg_idle.avg
896297 ± 5% -29.2% 634560 ± 8% sched_debug.cpu.avg_idle.max
260841 ± 4% -28.5% 186459 ± 8% sched_debug.cpu.avg_idle.stddev
1597 ± 2% +31.5% 2099 ± 5% sched_debug.cpu.curr->pid.avg
0.95 ± 11% +39.9% 1.33 ± 11% sched_debug.cpu.nr_running.avg
0.59 ± 11% +20.8% 0.71 ± 9% sched_debug.cpu.nr_running.stddev
525011 ± 8% +16.5% 611478 ± 2% sched_debug.cpu.nr_switches.stddev
36.67 ± 12% +51.6% 55.58 ± 18% sched_debug.cpu.nr_uninterruptible.max
-37.75 +68.8% -63.73 sched_debug.cpu.nr_uninterruptible.min
20.71 ± 16% +52.8% 31.65 ± 9% sched_debug.cpu.nr_uninterruptible.stddev
524055 ± 8% +16.4% 610231 ± 2% sched_debug.cpu.sched_count.stddev
179082 ± 15% +431.0% 950904 ± 4% sched_debug.cpu.sched_goidle.avg
941875 ± 8% +90.8% 1797084 sched_debug.cpu.sched_goidle.max
57580 ± 47% +1330.6% 823745 ± 6% sched_debug.cpu.sched_goidle.min
261290 ± 8% +16.8% 305219 ± 2% sched_debug.cpu.sched_goidle.stddev
2351907 ± 2% -22.8% 1815202 ± 5% sched_debug.cpu.ttwu_count.avg
3121048 ± 4% -14.7% 2661346 ± 2% sched_debug.cpu.ttwu_count.max
2204420 ± 2% -23.6% 1683662 ± 6% sched_debug.cpu.ttwu_count.min
263284 ± 8% +15.9% 305082 ± 3% sched_debug.cpu.ttwu_count.stddev
2145030 ± 3% -94.7% 113807 ± 2% sched_debug.cpu.ttwu_local.avg
2170062 ± 3% -94.2% 125644 ± 10% sched_debug.cpu.ttwu_local.max
2117317 ± 3% -94.8% 109613 ± 3% sched_debug.cpu.ttwu_local.min
13155 ± 17% -70.2% 3923 ± 89% sched_debug.cpu.ttwu_local.stddev
8071 ± 4% +15.3% 9304 ± 11% softirqs.CPU0.RCU
13782 ± 12% +105.3% 28301 ± 6% softirqs.CPU0.SCHED
57292 ± 5% +20.8% 69210 ± 6% softirqs.CPU0.TIMER
10409 ± 12% +146.9% 25705 ± 5% softirqs.CPU1.SCHED
8100 ± 2% +18.5% 9598 ± 10% softirqs.CPU10.RCU
10535 ± 9% +131.8% 24423 ± 7% softirqs.CPU10.SCHED
63176 ± 7% +12.3% 70916 ± 4% softirqs.CPU10.TIMER
10881 ± 9% +139.9% 26107 ± 8% softirqs.CPU11.SCHED
59833 ± 4% +19.4% 71424 ± 14% softirqs.CPU11.TIMER
10829 ± 2% +129.5% 24855 ± 8% softirqs.CPU12.SCHED
56830 ± 4% +21.9% 69297 ± 10% softirqs.CPU12.TIMER
8338 ± 7% +15.6% 9636 ± 13% softirqs.CPU13.RCU
9965 ± 19% +164.7% 26377 ± 8% softirqs.CPU13.SCHED
7913 ± 5% +18.8% 9400 ± 13% softirqs.CPU14.RCU
10301 ± 13% +137.8% 24499 ± 13% softirqs.CPU14.SCHED
58108 ± 4% +23.3% 71658 ± 3% softirqs.CPU14.TIMER
10402 ± 9% +141.5% 25126 ± 7% softirqs.CPU15.SCHED
60348 ± 2% +17.2% 70757 ± 10% softirqs.CPU15.TIMER
12252 ± 8% +116.1% 26482 ± 5% softirqs.CPU2.SCHED
60066 ± 4% +21.8% 73163 ± 12% softirqs.CPU2.TIMER
8195 ± 4% +14.2% 9358 ± 7% softirqs.CPU3.RCU
11267 ± 17% +123.3% 25161 ± 3% softirqs.CPU3.SCHED
57990 ± 5% +25.0% 72490 ± 13% softirqs.CPU3.TIMER
11242 ± 12% +134.0% 26309 ± 9% softirqs.CPU4.SCHED
62003 ± 7% +18.1% 73208 softirqs.CPU4.TIMER
10623 ± 21% +150.0% 26564 ± 8% softirqs.CPU5.SCHED
64978 ± 9% +16.4% 75636 ± 9% softirqs.CPU5.TIMER
9820 ± 20% +159.9% 25525 ± 7% softirqs.CPU6.SCHED
60525 ± 13% +25.3% 75836 ± 8% softirqs.CPU6.TIMER
8310 ± 9% +15.4% 9587 ± 7% softirqs.CPU7.RCU
10202 ± 6% +155.3% 26044 ± 11% softirqs.CPU7.SCHED
61012 +20.3% 73383 ± 8% softirqs.CPU7.TIMER
7810 ± 6% +28.2% 10012 ± 12% softirqs.CPU8.RCU
11163 ± 3% +135.9% 26339 ± 11% softirqs.CPU8.SCHED
57386 ± 5% +22.0% 70029 ± 7% softirqs.CPU8.TIMER
11643 ± 15% +121.4% 25780 ± 3% softirqs.CPU9.SCHED
138325 ± 2% +13.1% 156403 ± 10% softirqs.RCU
175325 ± 2% +135.9% 413606 ± 6% softirqs.SCHED
967685 ± 3% +18.5% 1146565 ± 7% softirqs.TIMER
21179 ± 4% +15.9% 24541 ± 6% interrupts.CAL:Function_call_interrupts
1346 ± 4% +16.0% 1561 ± 10% interrupts.CPU0.CAL:Function_call_interrupts
255151 ± 7% +33.5% 340712 ± 7% interrupts.CPU0.LOC:Local_timer_interrupts
1597 ± 17% +536.7% 10169 ± 3% interrupts.CPU0.RES:Rescheduling_interrupts
262988 ± 3% +29.4% 340284 ± 7% interrupts.CPU1.LOC:Local_timer_interrupts
1255 ± 15% +656.0% 9492 interrupts.CPU1.RES:Rescheduling_interrupts
259958 ± 5% +31.0% 340654 ± 7% interrupts.CPU10.LOC:Local_timer_interrupts
1448 ± 21% +526.9% 9079 interrupts.CPU10.RES:Rescheduling_interrupts
264562 ± 2% +28.7% 340366 ± 7% interrupts.CPU11.LOC:Local_timer_interrupts
1176 ± 9% +682.6% 9207 ± 2% interrupts.CPU11.RES:Rescheduling_interrupts
1202 ± 17% +31.8% 1584 ± 3% interrupts.CPU12.CAL:Function_call_interrupts
256269 ± 7% +33.0% 340874 ± 7% interrupts.CPU12.LOC:Local_timer_interrupts
1240 ± 11% +644.6% 9232 interrupts.CPU12.RES:Rescheduling_interrupts
258695 ± 6% +31.8% 340891 ± 7% interrupts.CPU13.LOC:Local_timer_interrupts
1329 ± 34% +605.9% 9383 interrupts.CPU13.RES:Rescheduling_interrupts
1161 ± 16% +39.3% 1618 ± 7% interrupts.CPU14.CAL:Function_call_interrupts
259395 ± 5% +31.3% 340636 ± 7% interrupts.CPU14.LOC:Local_timer_interrupts
1313 ± 26% +603.5% 9240 interrupts.CPU14.RES:Rescheduling_interrupts
256597 ± 7% +32.6% 340354 ± 7% interrupts.CPU15.LOC:Local_timer_interrupts
1293 ± 10% +615.1% 9250 ± 2% interrupts.CPU15.RES:Rescheduling_interrupts
259187 ± 5% +31.2% 340034 ± 7% interrupts.CPU2.LOC:Local_timer_interrupts
1482 ± 18% +566.1% 9873 ± 3% interrupts.CPU2.RES:Rescheduling_interrupts
264129 ± 2% +28.8% 340116 ± 7% interrupts.CPU3.LOC:Local_timer_interrupts
1387 ± 19% +584.8% 9503 interrupts.CPU3.RES:Rescheduling_interrupts
1249 ± 19% +19.5% 1493 ± 3% interrupts.CPU4.CAL:Function_call_interrupts
256073 ± 7% +32.8% 340112 ± 7% interrupts.CPU4.LOC:Local_timer_interrupts
1372 ± 18% +594.0% 9521 ± 4% interrupts.CPU4.RES:Rescheduling_interrupts
258411 ± 6% +31.6% 340115 ± 7% interrupts.CPU5.LOC:Local_timer_interrupts
1267 ± 17% +655.9% 9582 ± 2% interrupts.CPU5.RES:Rescheduling_interrupts
258716 ± 5% +31.3% 339719 ± 7% interrupts.CPU6.LOC:Local_timer_interrupts
1220 ± 7% +666.4% 9350 ± 2% interrupts.CPU6.RES:Rescheduling_interrupts
1192 ± 15% +33.0% 1586 ± 8% interrupts.CPU7.CAL:Function_call_interrupts
255847 ± 7% +32.8% 339836 ± 7% interrupts.CPU7.LOC:Local_timer_interrupts
1217 ± 13% +687.4% 9588 ± 2% interrupts.CPU7.RES:Rescheduling_interrupts
1369 ± 5% +14.4% 1566 ± 6% interrupts.CPU8.CAL:Function_call_interrupts
256492 ± 7% +32.9% 340775 ± 7% interrupts.CPU8.LOC:Local_timer_interrupts
1222 ± 11% +688.9% 9640 ± 7% interrupts.CPU8.RES:Rescheduling_interrupts
263589 ± 3% +29.1% 340310 ± 7% interrupts.CPU9.LOC:Local_timer_interrupts
1486 ± 22% +518.1% 9189 ± 2% interrupts.CPU9.RES:Rescheduling_interrupts
4146065 ± 5% +31.3% 5445793 ± 7% interrupts.LOC:Local_timer_interrupts
21310 ± 11% +610.0% 151304 interrupts.RES:Rescheduling_interrupts
41.41 -8.2% 38.01 ± 3% perf-stat.i.MPKI
3.181e+09 -6.2% 2.984e+09 ± 6% perf-stat.i.branch-instructions
1.81 ± 2% -0.1 1.69 perf-stat.i.branch-miss-rate%
41240020 -15.2% 34981946 ± 8% perf-stat.i.branch-misses
5.891e+08 ± 2% -10.6% 5.264e+08 ± 6% perf-stat.i.cache-misses
5.891e+08 ± 2% -10.6% 5.264e+08 ± 6% perf-stat.i.cache-references
549435 ± 2% -17.1% 455328 ± 7% perf-stat.i.context-switches
1.76 +9.8% 1.93 perf-stat.i.cpi
8164 ± 30% +1194.2% 105659 ± 7% perf-stat.i.cpu-migrations
0.10 ± 6% +0.2 0.27 ± 3% perf-stat.i.dTLB-load-miss-rate%
3249270 ± 8% +223.7% 10518062 ± 6% perf-stat.i.dTLB-load-misses
4.53e+09 -11.1% 4.025e+09 ± 5% perf-stat.i.dTLB-loads
0.05 ± 6% +0.0 0.09 ± 3% perf-stat.i.dTLB-store-miss-rate%
1269803 ± 4% +34.4% 1706370 ± 7% perf-stat.i.dTLB-store-misses
2.468e+09 -22.9% 1.904e+09 ± 6% perf-stat.i.dTLB-stores
51.76 ± 6% +6.1 57.84 ± 2% perf-stat.i.iTLB-load-miss-rate%
2394143 ± 12% +28.6% 3079996 ± 5% perf-stat.i.iTLB-load-misses
2553375 ± 2% +46.1% 3730425 ± 6% perf-stat.i.iTLB-loads
1.464e+10 -8.7% 1.337e+10 ± 5% perf-stat.i.instructions
0.62 -9.2% 0.56 perf-stat.i.ipc
4962 -15.3% 4204 ± 5% perf-stat.i.minor-faults
4962 -15.3% 4204 ± 5% perf-stat.i.page-faults
40.24 -2.1% 39.38 perf-stat.overall.MPKI
1.30 ± 2% -0.1 1.17 ± 2% perf-stat.overall.branch-miss-rate%
1.73 +11.8% 1.93 perf-stat.overall.cpi
43.00 +14.2% 49.10 perf-stat.overall.cycles-between-cache-misses
0.07 ± 9% +0.2 0.26 ± 2% perf-stat.overall.dTLB-load-miss-rate%
0.05 ± 4% +0.0 0.09 perf-stat.overall.dTLB-store-miss-rate%
48.21 ± 5% -3.0 45.24 perf-stat.overall.iTLB-load-miss-rate%
6211 ± 12% -30.1% 4340 perf-stat.overall.instructions-per-iTLB-miss
0.58 -10.5% 0.52 perf-stat.overall.ipc
3.157e+09 -6.0% 2.966e+09 ± 6% perf-stat.ps.branch-instructions
40940076 -15.1% 34774814 ± 8% perf-stat.ps.branch-misses
5.846e+08 ± 2% -10.5% 5.232e+08 ± 5% perf-stat.ps.cache-misses
5.846e+08 ± 2% -10.5% 5.232e+08 ± 5% perf-stat.ps.cache-references
545185 ± 2% -17.0% 452472 ± 7% perf-stat.ps.context-switches
8100 ± 30% +1196.1% 104990 ± 7% perf-stat.ps.cpu-migrations
3224891 ± 8% +224.1% 10452308 ± 6% perf-stat.ps.dTLB-load-misses
4.495e+09 -11.0% 4e+09 ± 5% perf-stat.ps.dTLB-loads
1260065 ± 4% +34.6% 1695722 ± 7% perf-stat.ps.dTLB-store-misses
2.449e+09 -22.7% 1.892e+09 ± 6% perf-stat.ps.dTLB-stores
2375668 ± 12% +28.8% 3060737 ± 5% perf-stat.ps.iTLB-load-misses
2533621 ± 2% +46.3% 3706931 ± 6% perf-stat.ps.iTLB-loads
1.453e+10 -8.5% 1.328e+10 ± 5% perf-stat.ps.instructions
4930 -15.2% 4181 ± 5% perf-stat.ps.minor-faults
4931 -15.2% 4181 ± 5% perf-stat.ps.page-faults
1.964e+12 ± 2% +14.3% 2.244e+12 perf-stat.total.instructions
10.67 ± 18% -10.7 0.00 perf-profile.calltrace.cycles-pp.activate_task.ttwu_do_activate.try_to_wake_up.autoremove_wake_function.__wake_up_common
10.60 ± 18% -10.6 0.00 perf-profile.calltrace.cycles-pp.enqueue_task_fair.activate_task.ttwu_do_activate.try_to_wake_up.autoremove_wake_function
21.89 ± 6% -9.8 12.08 ± 17% perf-profile.calltrace.cycles-pp.copy_page_to_iter.pipe_read.new_sync_read.vfs_read.ksys_read
21.03 ± 4% -9.4 11.63 ± 33% perf-profile.calltrace.cycles-pp.copy_page_from_iter.pipe_write.new_sync_write.vfs_write.ksys_write
8.95 ± 18% -9.0 0.00 perf-profile.calltrace.cycles-pp.enqueue_entity.enqueue_task_fair.activate_task.ttwu_do_activate.try_to_wake_up
18.52 ± 5% -8.8 9.70 ± 32% perf-profile.calltrace.cycles-pp.copyin.copy_page_from_iter.pipe_write.new_sync_write.vfs_write
18.21 ± 5% -8.6 9.58 ± 32% perf-profile.calltrace.cycles-pp.copy_user_enhanced_fast_string.copyin.copy_page_from_iter.pipe_write.new_sync_write
19.21 ± 7% -8.4 10.80 ± 16% perf-profile.calltrace.cycles-pp.copyout.copy_page_to_iter.pipe_read.new_sync_read.vfs_read
18.95 ± 7% -8.3 10.68 ± 16% perf-profile.calltrace.cycles-pp.copy_user_enhanced_fast_string.copyout.copy_page_to_iter.pipe_read.new_sync_read
7.84 ± 18% -7.8 0.00 perf-profile.calltrace.cycles-pp.__account_scheduler_latency.enqueue_entity.enqueue_task_fair.activate_task.ttwu_do_activate
7.68 ± 17% -7.7 0.00 perf-profile.calltrace.cycles-pp.__wake_up_common.pipe_read.new_sync_read.vfs_read.ksys_read
7.60 ± 17% -7.6 0.00 perf-profile.calltrace.cycles-pp.autoremove_wake_function.__wake_up_common.pipe_read.new_sync_read.vfs_read
7.51 ± 17% -7.5 0.00 perf-profile.calltrace.cycles-pp.try_to_wake_up.autoremove_wake_function.__wake_up_common.pipe_read.new_sync_read
6.83 ± 17% -6.8 0.00 perf-profile.calltrace.cycles-pp.__wake_up_common.pipe_write.new_sync_write.vfs_write.ksys_write
6.51 ± 17% -6.5 0.00 perf-profile.calltrace.cycles-pp.autoremove_wake_function.__wake_up_common.pipe_write.new_sync_write.vfs_write
6.43 ± 17% -6.4 0.00 perf-profile.calltrace.cycles-pp.try_to_wake_up.autoremove_wake_function.__wake_up_common.pipe_write.new_sync_write
6.27 ± 18% -6.3 0.00 perf-profile.calltrace.cycles-pp.stack_trace_save_tsk.__account_scheduler_latency.enqueue_entity.enqueue_task_fair.activate_task
5.90 ± 18% -5.9 0.00 perf-profile.calltrace.cycles-pp.arch_stack_walk.stack_trace_save_tsk.__account_scheduler_latency.enqueue_entity.enqueue_task_fair
5.77 ± 18% -5.8 0.00 perf-profile.calltrace.cycles-pp.ttwu_do_activate.try_to_wake_up.autoremove_wake_function.__wake_up_common.pipe_read
4.95 ± 18% -4.9 0.00 perf-profile.calltrace.cycles-pp.ttwu_do_activate.try_to_wake_up.autoremove_wake_function.__wake_up_common.pipe_write
0.91 ± 4% -0.5 0.41 ± 58% perf-profile.calltrace.cycles-pp.alloc_pages_current.pipe_write.new_sync_write.vfs_write.ksys_write
0.73 ± 11% +0.4 1.17 ± 5% perf-profile.calltrace.cycles-pp._raw_spin_lock_irq.pipe_read.new_sync_read.vfs_read.ksys_read
0.35 ±173% +3.3 3.62 ± 45% perf-profile.calltrace.cycles-pp.free_pcppages_bulk.free_unref_page.pipe_read.new_sync_read.vfs_read
0.00 +14.4 14.43 ±154% perf-profile.calltrace.cycles-pp.intel_idle.cpuidle_enter_state.cpuidle_enter.do_idle.cpu_startup_entry
0.00 +15.1 15.06 ±154% perf-profile.calltrace.cycles-pp.cpuidle_enter_state.cpuidle_enter.do_idle.cpu_startup_entry.start_secondary
0.00 +15.1 15.13 ±154% perf-profile.calltrace.cycles-pp.cpuidle_enter.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64
2.68 ±111% +15.2 17.90 ± 37% perf-profile.calltrace.cycles-pp.mutex_spin_on_owner.__mutex_lock.pipe_write.new_sync_write.vfs_write
2.72 ±111% +15.5 18.19 ± 37% perf-profile.calltrace.cycles-pp.__mutex_lock.pipe_write.new_sync_write.vfs_write.ksys_write
0.00 +15.7 15.75 ±153% perf-profile.calltrace.cycles-pp.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64
0.00 +15.8 15.75 ±153% perf-profile.calltrace.cycles-pp.cpu_startup_entry.start_secondary.secondary_startup_64
0.00 +15.8 15.75 ±153% perf-profile.calltrace.cycles-pp.start_secondary.secondary_startup_64
0.00 +16.2 16.17 ±153% perf-profile.calltrace.cycles-pp.secondary_startup_64
2.86 ±113% +16.3 19.18 ± 24% perf-profile.calltrace.cycles-pp.mutex_spin_on_owner.__mutex_lock.pipe_read.new_sync_read.vfs_read
2.90 ±113% +16.5 19.44 ± 24% perf-profile.calltrace.cycles-pp.__mutex_lock.pipe_read.new_sync_read.vfs_read.ksys_read
37.47 ± 6% -17.1 20.39 ± 23% perf-profile.children.cycles-pp.copy_user_enhanced_fast_string
14.56 ± 17% -14.1 0.45 ± 51% perf-profile.children.cycles-pp.__wake_up_common
14.43 ± 15% -14.0 0.41 ± 50% perf-profile.children.cycles-pp.try_to_wake_up
14.12 ± 17% -13.8 0.33 ± 57% perf-profile.children.cycles-pp.autoremove_wake_function
11.05 ± 16% -10.8 0.27 ± 57% perf-profile.children.cycles-pp.ttwu_do_activate
11.02 ± 16% -10.7 0.27 ± 57% perf-profile.children.cycles-pp.activate_task
10.95 ± 16% -10.7 0.27 ± 57% perf-profile.children.cycles-pp.enqueue_task_fair
22.01 ± 6% -9.9 12.13 ± 17% perf-profile.children.cycles-pp.copy_page_to_iter
21.16 ± 4% -9.5 11.69 ± 33% perf-profile.children.cycles-pp.copy_page_from_iter
9.27 ± 16% -9.0 0.22 ± 57% perf-profile.children.cycles-pp.enqueue_entity
18.59 ± 5% -8.9 9.73 ± 32% perf-profile.children.cycles-pp.copyin
19.27 ± 7% -8.4 10.84 ± 16% perf-profile.children.cycles-pp.copyout
8.10 ± 16% -7.9 0.19 ± 57% perf-profile.children.cycles-pp.__account_scheduler_latency
7.89 ± 15% -7.3 0.61 ± 47% perf-profile.children.cycles-pp.pipe_wait
7.24 ± 14% -6.9 0.37 ± 48% perf-profile.children.cycles-pp.__schedule
7.12 ± 15% -6.8 0.29 ± 59% perf-profile.children.cycles-pp.schedule
6.48 ± 16% -6.3 0.16 ± 57% perf-profile.children.cycles-pp.stack_trace_save_tsk
6.21 ± 16% -6.0 0.15 ± 57% perf-profile.children.cycles-pp.arch_stack_walk
3.81 ± 16% -3.7 0.09 ± 59% perf-profile.children.cycles-pp.unwind_next_frame
2.79 ± 15% -2.7 0.11 ± 59% perf-profile.children.cycles-pp.dequeue_task_fair
2.13 ± 14% -2.1 0.08 ± 57% perf-profile.children.cycles-pp.select_task_rq_fair
1.69 ± 14% -1.6 0.06 ± 59% perf-profile.children.cycles-pp.select_idle_sibling
1.48 ± 15% -1.4 0.07 ± 58% perf-profile.children.cycles-pp.update_curr
1.19 ± 16% -1.2 0.03 ±100% perf-profile.children.cycles-pp.reweight_entity
1.25 ± 12% -1.2 0.10 ± 59% perf-profile.children.cycles-pp.pick_next_task_fair
1.20 ± 12% -1.1 0.08 ± 58% perf-profile.children.cycles-pp.update_load_avg
1.11 ± 16% -1.1 0.06 ± 63% perf-profile.children.cycles-pp.dequeue_entity
1.09 ± 19% -1.0 0.13 ± 19% perf-profile.children.cycles-pp._raw_spin_lock_irqsave
0.99 ± 12% -0.9 0.05 ± 58% perf-profile.children.cycles-pp.switch_mm_irqs_off
1.05 ± 6% -0.6 0.41 ± 38% perf-profile.children.cycles-pp.security_file_permission
1.67 ± 3% -0.6 1.09 ± 46% perf-profile.children.cycles-pp.__might_fault
1.81 -0.6 1.24 ± 20% perf-profile.children.cycles-pp.___might_sleep
0.74 ± 8% -0.5 0.26 ± 46% perf-profile.children.cycles-pp.free_unref_page_prepare
0.95 ± 4% -0.5 0.49 ± 32% perf-profile.children.cycles-pp.alloc_pages_current
1.38 ± 3% -0.5 0.91 ± 41% perf-profile.children.cycles-pp.__might_sleep
0.79 ± 3% -0.4 0.35 ± 21% perf-profile.children.cycles-pp.syscall_return_via_sysret
0.79 ± 5% -0.4 0.36 ± 21% perf-profile.children.cycles-pp._cond_resched
0.66 ± 4% -0.4 0.24 ± 39% perf-profile.children.cycles-pp.selinux_file_permission
0.51 ± 11% -0.4 0.10 ± 57% perf-profile.children.cycles-pp.generic_pipe_buf_confirm
0.47 ± 45% -0.4 0.07 ± 58% perf-profile.children.cycles-pp.__mutex_unlock_slowpath
0.46 ± 45% -0.4 0.07 ± 59% perf-profile.children.cycles-pp.wake_up_q
0.80 ± 7% -0.4 0.40 ± 32% perf-profile.children.cycles-pp.free_unref_page_commit
0.68 ± 8% -0.4 0.31 ± 20% perf-profile.children.cycles-pp.entry_SYSCALL_64
0.40 ± 8% -0.3 0.10 ± 79% perf-profile.children.cycles-pp.native_write_msr
0.43 ± 8% -0.3 0.16 ± 19% perf-profile.children.cycles-pp.__fdget_pos
0.41 ± 7% -0.3 0.15 ± 20% perf-profile.children.cycles-pp.__fget_light
0.45 ± 9% -0.2 0.20 ± 25% perf-profile.children.cycles-pp.__put_page
0.42 ± 7% -0.2 0.18 ± 13% perf-profile.children.cycles-pp.__x86_indirect_thunk_rax
0.36 ± 3% -0.2 0.17 ± 28% perf-profile.children.cycles-pp.rcu_all_qs
0.28 ± 12% -0.2 0.10 ± 58% perf-profile.children.cycles-pp.file_has_perm
0.40 ± 9% -0.2 0.24 ± 14% perf-profile.children.cycles-pp.mutex_lock
0.23 ± 9% -0.1 0.08 ± 58% perf-profile.children.cycles-pp.fsnotify
0.21 ± 14% -0.1 0.08 ± 57% perf-profile.children.cycles-pp.avc_has_perm
0.26 ± 10% -0.1 0.13 ± 57% perf-profile.children.cycles-pp.__inc_numa_state
0.15 ± 11% -0.1 0.03 ±100% perf-profile.children.cycles-pp.__fsnotify_parent
0.21 ± 8% -0.1 0.09 ± 57% perf-profile.children.cycles-pp.get_task_policy
0.20 ± 6% -0.1 0.09 ± 27% perf-profile.children.cycles-pp.__page_cache_release
0.17 ± 15% -0.1 0.06 ± 58% perf-profile.children.cycles-pp.mem_cgroup_uncharge
0.21 ± 2% -0.1 0.12 ± 28% perf-profile.children.cycles-pp.touch_atime
0.15 ± 9% -0.1 0.06 ± 58% perf-profile.children.cycles-pp.free_pcp_prepare
0.13 ± 20% -0.1 0.05 ± 58% perf-profile.children.cycles-pp.prep_new_page
0.12 ± 8% -0.1 0.06 ± 58% perf-profile.children.cycles-pp.policy_nodemask
0.13 ± 9% -0.1 0.07 ± 58% perf-profile.children.cycles-pp.atime_needs_update
0.11 ± 9% -0.1 0.05 ± 58% perf-profile.children.cycles-pp.file_update_time
0.10 ± 7% -0.1 0.04 ± 58% perf-profile.children.cycles-pp.__inode_security_revalidate
0.10 ± 15% -0.1 0.04 ± 58% perf-profile.children.cycles-pp.should_fail_alloc_page
0.09 ± 8% -0.1 0.04 ± 58% perf-profile.children.cycles-pp.policy_node
0.09 ± 7% -0.0 0.05 ± 58% perf-profile.children.cycles-pp.current_time
0.14 ± 5% -0.0 0.12 ± 9% perf-profile.children.cycles-pp.__wake_up_common_lock
0.07 ± 17% +0.1 0.13 ± 45% perf-profile.children.cycles-pp.scheduler_tick
0.00 +0.1 0.08 ± 26% perf-profile.children.cycles-pp.osq_lock
0.03 ±173% +0.2 0.20 ± 40% perf-profile.children.cycles-pp.__mod_zone_page_state
0.00 +0.3 0.27 ±130% perf-profile.children.cycles-pp.irq_exit
0.00 +0.4 0.42 ±125% perf-profile.children.cycles-pp.start_kernel
1.56 ± 6% +0.5 2.08 ± 12% perf-profile.children.cycles-pp._raw_spin_lock_irq
0.34 ± 8% +0.7 1.09 ±108% perf-profile.children.cycles-pp.smp_apic_timer_interrupt
0.49 ±110% +3.2 3.65 ± 44% perf-profile.children.cycles-pp.free_pcppages_bulk
0.08 ±133% +14.4 14.51 ±153% perf-profile.children.cycles-pp.intel_idle
0.09 ±132% +15.4 15.54 ±153% perf-profile.children.cycles-pp.cpuidle_enter_state
0.09 ±132% +15.5 15.54 ±153% perf-profile.children.cycles-pp.cpuidle_enter
0.10 ±134% +15.7 15.75 ±153% perf-profile.children.cycles-pp.start_secondary
0.11 ±132% +16.1 16.17 ±153% perf-profile.children.cycles-pp.secondary_startup_64
0.11 ±132% +16.1 16.17 ±153% perf-profile.children.cycles-pp.cpu_startup_entry
0.11 ±132% +16.1 16.18 ±153% perf-profile.children.cycles-pp.do_idle
5.57 ±112% +31.8 37.39 ± 31% perf-profile.children.cycles-pp.mutex_spin_on_owner
5.83 ±110% +32.1 37.97 ± 30% perf-profile.children.cycles-pp.__mutex_lock
37.23 ± 6% -17.0 20.18 ± 24% perf-profile.self.cycles-pp.copy_user_enhanced_fast_string
1.59 ± 15% -1.5 0.04 ± 58% perf-profile.self.cycles-pp.unwind_next_frame
2.09 ± 6% -1.1 0.99 ± 18% perf-profile.self.cycles-pp.do_syscall_64
3.12 -0.9 2.21 ± 43% perf-profile.self.cycles-pp.get_page_from_freelist
1.27 ± 5% -0.7 0.53 ± 38% perf-profile.self.cycles-pp.pipe_write
0.74 ± 17% -0.7 0.03 ±100% perf-profile.self.cycles-pp.update_curr
1.21 ± 10% -0.7 0.53 ± 26% perf-profile.self.cycles-pp.pipe_read
0.69 ± 14% -0.7 0.03 ±100% perf-profile.self.cycles-pp.__schedule
1.59 ± 3% -0.6 1.01 ± 40% perf-profile.self.cycles-pp.__alloc_pages_nodemask
1.77 -0.6 1.21 ± 20% perf-profile.self.cycles-pp.___might_sleep
1.09 ± 6% -0.5 0.58 ± 30% perf-profile.self.cycles-pp.copy_page_to_iter
0.71 ± 8% -0.5 0.22 ± 44% perf-profile.self.cycles-pp.free_unref_page_prepare
0.61 ± 12% -0.5 0.12 ± 18% perf-profile.self.cycles-pp._raw_spin_lock_irqsave
0.79 ± 3% -0.4 0.35 ± 21% perf-profile.self.cycles-pp.syscall_return_via_sysret
0.99 ± 4% -0.4 0.59 ± 39% perf-profile.self.cycles-pp.copy_page_from_iter
0.46 ± 11% -0.4 0.08 ± 57% perf-profile.self.cycles-pp.generic_pipe_buf_confirm
1.22 ± 3% -0.4 0.85 ± 41% perf-profile.self.cycles-pp.__might_sleep
0.68 ± 8% -0.4 0.31 ± 20% perf-profile.self.cycles-pp.entry_SYSCALL_64
0.55 ± 6% -0.4 0.20 ± 37% perf-profile.self.cycles-pp.selinux_file_permission
0.69 ± 8% -0.3 0.36 ± 23% perf-profile.self.cycles-pp.free_unref_page
0.43 ± 7% -0.3 0.11 ± 57% perf-profile.self.cycles-pp.__wake_up_common
0.47 ± 5% -0.3 0.17 ± 40% perf-profile.self.cycles-pp.__might_fault
0.39 ± 9% -0.3 0.10 ± 79% perf-profile.self.cycles-pp.native_write_msr
0.55 ± 7% -0.3 0.28 ± 28% perf-profile.self.cycles-pp.free_unref_page_commit
0.53 ± 8% -0.3 0.26 ± 40% perf-profile.self.cycles-pp.alloc_pages_current
0.40 ± 8% -0.3 0.15 ± 19% perf-profile.self.cycles-pp.__fget_light
0.27 ± 12% -0.2 0.04 ± 59% perf-profile.self.cycles-pp.entry_SYSCALL_64_after_hwframe
0.37 ± 8% -0.2 0.17 ± 10% perf-profile.self.cycles-pp.__x86_indirect_thunk_rax
0.32 ± 3% -0.2 0.15 ± 30% perf-profile.self.cycles-pp._cond_resched
0.22 ± 6% -0.1 0.08 ± 62% perf-profile.self.cycles-pp.copyin
0.17 ± 7% -0.1 0.03 ±100% perf-profile.self.cycles-pp.new_sync_write
0.18 ± 9% -0.1 0.04 ± 58% perf-profile.self.cycles-pp.new_sync_read
0.27 ± 5% -0.1 0.13 ± 36% perf-profile.self.cycles-pp.rcu_all_qs
0.21 ± 8% -0.1 0.08 ± 58% perf-profile.self.cycles-pp.fsnotify
0.21 ± 14% -0.1 0.08 ± 57% perf-profile.self.cycles-pp.avc_has_perm
0.14 ± 8% -0.1 0.03 ±100% perf-profile.self.cycles-pp.__fsnotify_parent
0.23 ± 10% -0.1 0.11 ± 57% perf-profile.self.cycles-pp.__inc_numa_state
0.20 ± 9% -0.1 0.09 ± 58% perf-profile.self.cycles-pp.get_task_policy
0.18 ± 8% -0.1 0.09 ± 25% perf-profile.self.cycles-pp.__page_cache_release
0.18 ± 4% -0.1 0.09 ± 26% perf-profile.self.cycles-pp.copyout
0.13 ± 13% -0.1 0.05 ± 58% perf-profile.self.cycles-pp.mem_cgroup_uncharge
0.14 ± 6% -0.1 0.06 ± 58% perf-profile.self.cycles-pp.free_pcp_prepare
0.12 ± 21% -0.1 0.05 ± 58% perf-profile.self.cycles-pp.prep_new_page
0.12 ± 14% -0.1 0.06 ± 14% perf-profile.self.cycles-pp.__put_page
0.09 ± 9% -0.1 0.03 ±100% perf-profile.self.cycles-pp.policy_node
0.11 ± 7% -0.1 0.05 ± 58% perf-profile.self.cycles-pp.policy_nodemask
0.10 ± 9% -0.1 0.04 ± 58% perf-profile.self.cycles-pp.atime_needs_update
0.09 ± 12% -0.0 0.04 ± 58% perf-profile.self.cycles-pp.security_file_permission
0.00 +0.1 0.08 ± 26% perf-profile.self.cycles-pp.osq_lock
0.03 ±173% +0.2 0.19 ± 42% perf-profile.self.cycles-pp.__mod_zone_page_state
0.06 ±128% +0.3 0.38 ± 25% perf-profile.self.cycles-pp.__mutex_lock
1.50 ± 7% +0.6 2.06 ± 12% perf-profile.self.cycles-pp._raw_spin_lock_irq
0.18 ±128% +1.1 1.29 ± 36% perf-profile.self.cycles-pp.free_pcppages_bulk
0.08 ±133% +14.4 14.50 ±153% perf-profile.self.cycles-pp.intel_idle
5.54 ±112% +31.7 37.20 ± 31% perf-profile.self.cycles-pp.mutex_spin_on_owner
lmbench3.time.voluntary_context_switches
8.5e+07 +-+---------------------------------------------------------------+
| + |
8e+07 +-+ : |
7.5e+07 +-+ : : + + |
|+ : :+ : :+ .+ .+ .+ .+ + |
7e+07 +-++. .++ + +.+ +.+ ++. .++ + + +.++ :+ : .++.++. |
6.5e+07 +-+ ++ +.+ ++ + +.+ + +|
| +. : |
6e+07 +-+ + |
5.5e+07 +-+ |
O O O O O O O O O |
5e+07 +-O O O O O O O O O O O OO OO OO OO |
4.5e+07 +-+ O O O O O |
| |
4e+07 +-+---------------------------------------------------------------+
lmbench3.PIPE.latency.us
28 +-+--------------------------------------------------------------------+
O OO OO O OO O OO OO O OO OO O OO O OO OO O OO OO O O |
26 +-+ |
24 +-+ |
| |
22 +-+ |
| |
20 +-+ |
| |
18 +-+ |
16 +-+ |
| |
14 +-++.++.+.++.+.++.++.+.++.++.+.++.+.++.++.+.++.++.+.++.++.+.++.+.++.++.|
| |
12 +-+--------------------------------------------------------------------+
[*] 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-5.4.0-rc2-00012-g3c0edea9b29f9" of type "text/plain" (200609 bytes)
View attachment "job-script" of type "text/plain" (7928 bytes)
View attachment "job.yaml" of type "text/plain" (5658 bytes)
View attachment "reproduce" of type "text/plain" (566 bytes)
Powered by blists - more mailing lists