| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <20200308140241.GP5972@shao2-debian>
Date: Sun, 8 Mar 2020 22:02:41 +0800
From: kernel test robot <rong.a.chen@...el.com>
To: Eric Biggers <ebiggers@...gle.com>
Cc: Theodore Ts'o <tytso@....edu>, LKML <linux-kernel@...r.kernel.org>,
Linus Torvalds <torvalds@...ux-foundation.org>,
lkp@...ts.01.org
Subject: [fscrypt] 22d94f493b: will-it-scale.per_thread_ops 1.7% improvement
Greeting,
FYI, we noticed a 1.7% improvement of will-it-scale.per_thread_ops due to commit:
commit: 22d94f493bfb408fdd764f7b1d0363af2122fba5 ("fscrypt: add FS_IOC_ADD_ENCRYPTION_KEY ioctl")
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master
in testcase: will-it-scale
on test machine: 192 threads Intel(R) Xeon(R) Platinum 9242 CPU @ 2.30GHz with 192G memory
with following parameters:
nr_task: 50%
mode: thread
test: poll2
cpufreq_governor: performance
ucode: 0x500002c
test-description: Will It Scale takes a testcase and runs it from 1 through to n parallel copies to see if the testcase will scale. It builds both a process and threads based test in order to see any differences between the two.
test-url: https://github.com/antonblanchard/will-it-scale
Details are as below:
-------------------------------------------------------------------------------------------------->
To reproduce:
git clone https://github.com/intel/lkp-tests.git
cd lkp-tests
bin/lkp install job.yaml # job file is attached in this email
bin/lkp run job.yaml
=========================================================================================
compiler/cpufreq_governor/kconfig/mode/nr_task/rootfs/tbox_group/test/testcase/ucode:
gcc-7/performance/x86_64-rhel-7.6/thread/50%/debian-x86_64-20191114.cgz/lkp-csl-2ap3/poll2/will-it-scale/0x500002c
commit:
feed825861 ("fscrypt: rename keyinfo.c to keysetup.c")
22d94f493b ("fscrypt: add FS_IOC_ADD_ENCRYPTION_KEY ioctl")
feed825861919ac3 22d94f493bfb408fdd764f7b1d0
---------------- ---------------------------
%stddev %change %stddev
\ | \
218250 +1.7% 221854 will-it-scale.per_thread_ops
20952104 +1.7% 21298053 will-it-scale.workload
4968 ± 4% -5.4% 4700 boot-time.idle
499.00 ± 84% +440.2% 2695 ± 52% numa-meminfo.node1.Inactive
368.25 ± 54% +590.0% 2540 ± 55% numa-meminfo.node1.Inactive(anon)
441.50 ± 50% +664.1% 3373 ± 61% numa-meminfo.node1.Shmem
91.75 ± 55% +591.8% 634.75 ± 56% numa-vmstat.node1.nr_inactive_anon
110.00 ± 50% +666.6% 843.25 ± 61% numa-vmstat.node1.nr_shmem
91.75 ± 55% +591.8% 634.75 ± 56% numa-vmstat.node1.nr_zone_inactive_anon
6706982 ± 84% -92.7% 487432 ±150% turbostat.C6
10.14 ± 88% -9.6 0.55 ±166% turbostat.C6%
314.29 +2.5% 322.01 turbostat.PkgWatt
1.169e+10 ± 86% -94.9% 6.005e+08 ±162% cpuidle.C6.time
13367929 ± 81% -93.2% 909676 ±146% cpuidle.C6.usage
339257 ± 76% -53.8% 156796 ± 2% cpuidle.POLL.time
81726 ± 10% -12.6% 71463 ± 3% cpuidle.POLL.usage
0.14 ± 18% -0.1 0.09 ± 29% perf-profile.children.cycles-pp.ktime_get
0.12 ± 12% -0.0 0.09 ± 19% perf-profile.children.cycles-pp.clockevents_program_event
30.57 ± 8% -3.6 26.96 ± 3% perf-profile.self.cycles-pp.do_sys_poll
0.14 ± 16% -0.1 0.08 ± 30% perf-profile.self.cycles-pp.ktime_get
7.443e+10 +1.6% 7.563e+10 perf-stat.i.branch-instructions
1.298e+08 +2.5% 1.33e+08 perf-stat.i.branch-misses
1082242 ± 3% +10.4% 1194939 ± 2% perf-stat.i.cache-misses
0.92 -1.8% 0.90 perf-stat.i.cpi
298917 ± 4% -10.2% 268385 ± 3% perf-stat.i.cycles-between-cache-misses
0.00 ± 37% -0.0 0.00 ± 16% perf-stat.i.dTLB-load-miss-rate%
8.085e+10 +1.6% 8.213e+10 perf-stat.i.dTLB-loads
4.63e+10 +1.6% 4.705e+10 perf-stat.i.dTLB-stores
85.43 +0.8 86.25 perf-stat.i.iTLB-load-miss-rate%
19168997 +6.2% 20347953 perf-stat.i.iTLB-load-misses
3.278e+11 +1.6% 3.331e+11 perf-stat.i.instructions
17066 -4.3% 16335 perf-stat.i.instructions-per-iTLB-miss
1.10 +1.7% 1.12 perf-stat.i.ipc
224154 +9.7% 245958 ± 2% perf-stat.i.node-load-misses
87483 +4.6% 91483 ± 2% perf-stat.i.node-store-misses
6727 ± 35% -37.8% 4185 ± 10% perf-stat.i.node-stores
0.91 -1.7% 0.90 perf-stat.overall.cpi
275723 ± 3% -9.5% 249423 ± 2% perf-stat.overall.cycles-between-cache-misses
85.51 +0.8 86.33 perf-stat.overall.iTLB-load-miss-rate%
17105 -4.3% 16370 perf-stat.overall.instructions-per-iTLB-miss
1.10 +1.7% 1.12 perf-stat.overall.ipc
7.418e+10 +1.6% 7.538e+10 perf-stat.ps.branch-instructions
1.294e+08 +2.4% 1.326e+08 perf-stat.ps.branch-misses
1080311 ± 3% +10.4% 1192449 ± 2% perf-stat.ps.cache-misses
8.059e+10 +1.6% 8.186e+10 perf-stat.ps.dTLB-loads
4.614e+10 +1.6% 4.689e+10 perf-stat.ps.dTLB-stores
19104494 +6.2% 20281309 perf-stat.ps.iTLB-load-misses
3.267e+11 +1.6% 3.32e+11 perf-stat.ps.instructions
223583 +9.8% 245558 ± 2% perf-stat.ps.node-load-misses
87233 +4.6% 91235 ± 2% perf-stat.ps.node-store-misses
6756 ± 35% -37.8% 4202 ± 10% perf-stat.ps.node-stores
9.876e+13 +1.7% 1.005e+14 perf-stat.total.instructions
11105 ±130% -77.1% 2542 ± 2% softirqs.CPU102.SCHED
98455 ± 19% -16.8% 81922 softirqs.CPU102.TIMER
99133 ± 22% -17.7% 81540 softirqs.CPU107.TIMER
14881 ±166% -97.8% 334.75 ± 13% softirqs.CPU11.NET_RX
109442 ± 22% -22.4% 84913 ± 7% softirqs.CPU113.TIMER
104590 ± 20% -22.1% 81527 softirqs.CPU116.TIMER
2325 ± 4% +363.2% 10770 ±133% softirqs.CPU124.SCHED
2334 ± 5% +345.6% 10402 ±132% softirqs.CPU141.SCHED
101587 ± 19% -17.4% 83945 ± 5% softirqs.CPU149.TIMER
89513 ± 3% -8.3% 82066 softirqs.CPU171.TIMER
89802 ± 4% -9.4% 81366 softirqs.CPU172.TIMER
89801 ± 4% -9.3% 81470 softirqs.CPU179.TIMER
89682 ± 4% -9.5% 81140 softirqs.CPU180.TIMER
100621 ± 17% -19.5% 81038 softirqs.CPU181.TIMER
100607 ± 19% -19.2% 81284 softirqs.CPU182.TIMER
90055 ± 3% -9.7% 81329 softirqs.CPU183.TIMER
89853 ± 4% -9.3% 81474 softirqs.CPU185.TIMER
89557 ± 4% -9.2% 81357 softirqs.CPU186.TIMER
99993 ± 17% -18.9% 81106 softirqs.CPU188.TIMER
92462 -12.3% 81128 softirqs.CPU189.TIMER
18909 ± 86% -87.3% 2405 ± 5% softirqs.CPU190.SCHED
102576 ± 15% -20.8% 81284 softirqs.CPU190.TIMER
91153 ± 5% -9.3% 82633 softirqs.CPU191.TIMER
32090 ± 51% -66.0% 10910 ±132% softirqs.CPU3.SCHED
115259 ± 21% -24.4% 87083 ± 8% softirqs.CPU3.TIMER
127475 ± 13% -27.1% 92878 ± 6% softirqs.CPU38.TIMER
29617 ± 48% -63.2% 10891 ±131% softirqs.CPU4.SCHED
114752 ± 21% -24.3% 86835 ± 7% softirqs.CPU4.TIMER
125958 ± 13% -23.9% 95830 softirqs.CPU61.TIMER
119596 ± 17% -22.6% 92563 ± 5% softirqs.CPU63.TIMER
29270 ± 49% -61.7% 11207 ±123% softirqs.CPU80.SCHED
116200 ± 19% -25.1% 87075 ± 7% softirqs.CPU80.TIMER
116402 ± 27% -28.7% 82942 softirqs.CPU96.TIMER
29576 ±167% -98.2% 537.25 ± 15% interrupts.33:PCI-MSI.524291-edge.eth0-TxRx-2
765.50 ± 26% -48.4% 395.00 ± 59% interrupts.CPU0.RES:Rescheduling_interrupts
324.25 ± 55% +122.4% 721.25 ± 38% interrupts.CPU1.RES:Rescheduling_interrupts
6128 ± 62% -78.0% 1346 ± 73% interrupts.CPU100.CAL:Function_call_interrupts
4980 ± 81% -91.1% 444.25 ±169% interrupts.CPU100.TLB:TLB_shootdowns
29576 ±167% -98.2% 537.25 ± 15% interrupts.CPU11.33:PCI-MSI.524291-edge.eth0-TxRx-2
3244 ± 95% -69.1% 1004 ± 18% interrupts.CPU11.CAL:Function_call_interrupts
5437 ± 40% +47.3% 8009 ± 16% interrupts.CPU149.NMI:Non-maskable_interrupts
5437 ± 40% +47.3% 8009 ± 16% interrupts.CPU149.PMI:Performance_monitoring_interrupts
3583 ± 34% +55.3% 5565 ± 4% interrupts.CPU18.NMI:Non-maskable_interrupts
3583 ± 34% +55.3% 5565 ± 4% interrupts.CPU18.PMI:Performance_monitoring_interrupts
8764 -31.5% 6001 ± 27% interrupts.CPU187.NMI:Non-maskable_interrupts
8764 -31.5% 6001 ± 27% interrupts.CPU187.PMI:Performance_monitoring_interrupts
3583 ± 34% +55.2% 5560 ± 5% interrupts.CPU19.NMI:Non-maskable_interrupts
3583 ± 34% +55.2% 5560 ± 5% interrupts.CPU19.PMI:Performance_monitoring_interrupts
3221 ± 48% +101.5% 6492 ± 20% interrupts.CPU22.NMI:Non-maskable_interrupts
3221 ± 48% +101.5% 6492 ± 20% interrupts.CPU22.PMI:Performance_monitoring_interrupts
7.25 ± 84% +1565.5% 120.75 ± 92% interrupts.CPU27.RES:Rescheduling_interrupts
3.50 ± 71% +55528.6% 1947 ±171% interrupts.CPU27.TLB:TLB_shootdowns
3581 ± 34% +77.3% 6351 ± 22% interrupts.CPU28.NMI:Non-maskable_interrupts
3581 ± 34% +77.3% 6351 ± 22% interrupts.CPU28.PMI:Performance_monitoring_interrupts
7.50 ±120% +1556.7% 124.25 ± 79% interrupts.CPU28.RES:Rescheduling_interrupts
2.50 ±103% +16560.0% 416.50 ±167% interrupts.CPU28.TLB:TLB_shootdowns
1361 ± 43% +340.0% 5987 ± 48% interrupts.CPU3.CAL:Function_call_interrupts
1.50 ±110% +27500.0% 414.00 ±169% interrupts.CPU35.TLB:TLB_shootdowns
1.50 ±110% +92466.7% 1388 ±171% interrupts.CPU38.TLB:TLB_shootdowns
1367 ± 44% +343.2% 6061 ± 48% interrupts.CPU4.CAL:Function_call_interrupts
2.50 ± 60% +59360.0% 1486 ±171% interrupts.CPU45.TLB:TLB_shootdowns
9.75 ±127% +26959.0% 2638 ± 99% interrupts.CPU5.TLB:TLB_shootdowns
229.75 ±166% +1277.9% 3165 ± 76% interrupts.CPU80.TLB:TLB_shootdowns
2865 +121.4% 6345 ± 22% interrupts.CPU82.NMI:Non-maskable_interrupts
2865 +121.4% 6345 ± 22% interrupts.CPU82.PMI:Performance_monitoring_interrupts
3583 ± 34% +55.6% 5576 ± 4% interrupts.CPU87.NMI:Non-maskable_interrupts
3583 ± 34% +55.6% 5576 ± 4% interrupts.CPU87.PMI:Performance_monitoring_interrupts
3247 ± 20% +118.4% 7091 ± 23% interrupts.CPU88.NMI:Non-maskable_interrupts
3247 ± 20% +118.4% 7091 ± 23% interrupts.CPU88.PMI:Performance_monitoring_interrupts
2865 +94.8% 5579 ± 4% interrupts.CPU89.NMI:Non-maskable_interrupts
2865 +94.8% 5579 ± 4% interrupts.CPU89.PMI:Performance_monitoring_interrupts
2865 +94.6% 5578 ± 4% interrupts.CPU90.NMI:Non-maskable_interrupts
2865 +94.6% 5578 ± 4% interrupts.CPU90.PMI:Performance_monitoring_interrupts
3222 ± 48% +124.8% 7245 ± 20% interrupts.CPU91.NMI:Non-maskable_interrupts
3222 ± 48% +124.8% 7245 ± 20% interrupts.CPU91.PMI:Performance_monitoring_interrupts
6.75 ± 97% +40055.6% 2710 ±100% interrupts.CPU91.TLB:TLB_shootdowns
76.75 ±149% -94.1% 4.50 ±122% interrupts.CPU92.RES:Rescheduling_interrupts
3249 ± 20% +71.5% 5572 ± 4% interrupts.CPU93.NMI:Non-maskable_interrupts
3249 ± 20% +71.5% 5572 ± 4% interrupts.CPU93.PMI:Performance_monitoring_interrupts
85.25 ±132% -95.6% 3.75 ± 96% interrupts.CPU93.RES:Rescheduling_interrupts
146.75 ± 88% -95.4% 6.75 ±113% interrupts.CPU94.RES:Rescheduling_interrupts
570.75 ±102% -98.5% 8.50 ± 51% interrupts.CPU94.TLB:TLB_shootdowns
398.50 ± 59% +73.9% 693.00 ± 6% interrupts.CPU95.TLB:TLB_shootdowns
6114 ± 63% -76.4% 1444 ± 63% interrupts.CPU99.CAL:Function_call_interrupts
4980 ± 81% -91.0% 446.25 ±169% interrupts.CPU99.TLB:TLB_shootdowns
will-it-scale.per_thread_ops
224000 +------------------------------------------------------------------+
| O O O O |
223000 |-O O O O O O O O O |
222000 |-+ O O O O O |
| O O O O O |
221000 |-+ O O O |
| |
220000 |-+ |
| +.. + |
219000 |-+ + +.+..+ : + |
218000 |.+.. + : : +..+.+..+.+..+.+..+. .+ |
| +. .. : : +. |
217000 |-+ + + |
| |
216000 +------------------------------------------------------------------+
[*] 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-5.3.0-rc3-00015-g22d94f493bfb4" of type "text/plain" (199535 bytes)
View attachment "job-script" of type "text/plain" (7654 bytes)
View attachment "job.yaml" of type "text/plain" (4999 bytes)
View attachment "reproduce" of type "text/plain" (308 bytes)
Powered by blists - more mailing lists