| 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: Mon, 1 Mar 2021 15:52:49 +0800
From: kernel test robot <oliver.sang@...el.com>
To: Thomas Gleixner <tglx@...utronix.de>
Cc: Alexandre Belloni <alexandre.belloni@...tlin.com>,
LKML <linux-kernel@...r.kernel.org>, lkp@...ts.01.org,
lkp@...el.com, ying.huang@...el.com, feng.tang@...el.com,
zhengjun.xing@...el.com
Subject: [rtc] 05a0302c35: stress-ng.rtc.ops_per_sec -23.2% regression
Greeting,
FYI, we noticed a -23.2% regression of stress-ng.rtc.ops_per_sec due to commit:
commit: 05a0302c35481e9b47fb90ba40922b0a4cae40d8 ("rtc: mc146818: Prevent reading garbage")
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master
in testcase: stress-ng
on test machine: 96 threads Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz with 192G memory
with following parameters:
nr_threads: 10%
disk: 1HDD
testtime: 60s
fs: ext4
class: os
test: rtc
cpufreq_governor: performance
ucode: 0x5003006
If you fix the issue, kindly add following tag
Reported-by: kernel test robot <oliver.sang@...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 split-job --compatible job.yaml
bin/lkp run compatible-job.yaml
=========================================================================================
class/compiler/cpufreq_governor/disk/fs/kconfig/nr_threads/rootfs/tbox_group/test/testcase/testtime/ucode:
os/gcc-9/performance/1HDD/ext4/x86_64-rhel-8.3/10%/debian-10.4-x86_64-20200603.cgz/lkp-csl-2sp5/rtc/stress-ng/60s/0x5003006
commit:
fef92cd2bc ("Merge tag 'timers-v5.11' of https://git.linaro.org/people/daniel.lezcano/linux into timers/core")
05a0302c35 ("rtc: mc146818: Prevent reading garbage")
fef92cd2bc04c64b 05a0302c35481e9b47fb90ba409
---------------- ---------------------------
%stddev %change %stddev
\ | \
266718 -23.2% 204802 stress-ng.rtc.ops
4445 -23.2% 3413 stress-ng.rtc.ops_per_sec
17181 -15.7% 14487 stress-ng.time.involuntary_context_switches
23753 ± 4% -16.8% 19771 ± 5% stress-ng.time.voluntary_context_switches
3074 -6.6% 2873 vmstat.system.cs
168341 ± 23% +30.3% 219394 ± 9% numa-numastat.node0.numa_hit
197610 ± 19% -30.4% 137589 ± 16% numa-numastat.node1.numa_hit
1046 -14.7% 892.83 ± 8% slabinfo.file_lock_cache.active_objs
1046 -14.7% 892.83 ± 8% slabinfo.file_lock_cache.num_objs
4038 ± 52% +81.4% 7327 ± 40% softirqs.CPU1.RCU
4018 ± 89% -55.4% 1793 ± 4% softirqs.CPU33.RCU
399840 -2.3% 390797 proc-vmstat.numa_hit
313382 -2.9% 304319 proc-vmstat.numa_local
442269 -4.2% 423682 proc-vmstat.pgalloc_normal
315987 -5.8% 297770 proc-vmstat.pgfree
262.33 ±106% +1329.7% 3750 ± 75% interrupts.CPU24.NMI:Non-maskable_interrupts
262.33 ±106% +1329.7% 3750 ± 75% interrupts.CPU24.PMI:Performance_monitoring_interrupts
113.83 ± 19% +1230.5% 1514 ± 65% interrupts.CPU38.NMI:Non-maskable_interrupts
113.83 ± 19% +1230.5% 1514 ± 65% interrupts.CPU38.PMI:Performance_monitoring_interrupts
185.67 ± 47% +1779.9% 3490 ± 76% interrupts.CPU72.NMI:Non-maskable_interrupts
185.67 ± 47% +1779.9% 3490 ± 76% interrupts.CPU72.PMI:Performance_monitoring_interrupts
114.33 ± 19% +1334.8% 1640 ±108% interrupts.CPU86.NMI:Non-maskable_interrupts
114.33 ± 19% +1334.8% 1640 ±108% interrupts.CPU86.PMI:Performance_monitoring_interrupts
0.63 ± 11% +0.8 1.43 ± 9% perf-profile.calltrace.cycles-pp.cmos_procfs.rtc_proc_show.seq_read.proc_reg_read.vfs_read
7.10 ± 12% +0.9 8.02 ± 9% perf-profile.calltrace.cycles-pp.osq_lock.__mutex_lock.rtc_read_time.time_show.dev_attr_show
6.93 ± 12% +0.9 7.85 ± 10% perf-profile.calltrace.cycles-pp.osq_lock.__mutex_lock.rtc_read_time.since_epoch_show.dev_attr_show
8.07 ± 12% +1.0 9.11 ± 10% perf-profile.calltrace.cycles-pp.__mutex_lock.rtc_read_time.since_epoch_show.dev_attr_show.sysfs_kf_seq_show
9.65 ± 12% +1.1 10.73 ± 10% perf-profile.calltrace.cycles-pp.since_epoch_show.dev_attr_show.sysfs_kf_seq_show.seq_read.vfs_read
9.64 ± 12% +1.1 10.72 ± 10% perf-profile.calltrace.cycles-pp.rtc_read_time.since_epoch_show.dev_attr_show.sysfs_kf_seq_show.seq_read
0.00 +1.2 1.15 ± 9% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irq.cmos_procfs.rtc_proc_show.seq_read
0.00 +1.2 1.15 ± 9% perf-profile.calltrace.cycles-pp._raw_spin_lock_irq.cmos_procfs.rtc_proc_show.seq_read.proc_reg_read
0.44 ± 10% -0.1 0.33 ± 7% perf-profile.children.cycles-pp._raw_spin_lock_irqsave
0.07 ± 6% -0.0 0.05 ± 45% perf-profile.children.cycles-pp.vsnprintf
0.39 ± 5% +0.1 0.46 ± 6% perf-profile.children.cycles-pp.worker_thread
0.39 ± 5% +0.1 0.46 ± 6% perf-profile.children.cycles-pp.process_one_work
0.38 ± 5% +0.1 0.46 ± 6% perf-profile.children.cycles-pp.rtc_timer_do_work
0.60 ± 10% +0.8 1.38 ± 8% perf-profile.children.cycles-pp.native_queued_spin_lock_slowpath
0.63 ± 11% +0.8 1.44 ± 9% perf-profile.children.cycles-pp.cmos_procfs
0.29 ± 11% +0.9 1.18 ± 9% perf-profile.children.cycles-pp._raw_spin_lock_irq
9.65 ± 12% +1.1 10.73 ± 10% perf-profile.children.cycles-pp.since_epoch_show
0.60 ± 10% +0.8 1.38 ± 8% perf-profile.self.cycles-pp.native_queued_spin_lock_slowpath
2.26 ± 12% -14.1% 1.94 ± 3% perf-stat.i.MPKI
1.573e+09 -1.9% 1.544e+09 perf-stat.i.branch-instructions
5662672 ± 4% -8.8% 5163661 ± 2% perf-stat.i.branch-misses
4334555 ± 5% -13.3% 3758431 ± 5% perf-stat.i.cache-misses
15484636 ± 4% -13.1% 13450378 ± 3% perf-stat.i.cache-references
2967 -7.4% 2747 perf-stat.i.context-switches
3.63 +2.1% 3.71 perf-stat.i.cpi
6257 ± 5% +16.8% 7305 ± 5% perf-stat.i.cycles-between-cache-misses
2.678e+08 -11.8% 2.361e+08 perf-stat.i.dTLB-stores
59.28 -2.1 57.18 perf-stat.i.iTLB-load-miss-rate%
3045003 -9.1% 2769045 perf-stat.i.iTLB-load-misses
7.407e+09 -1.9% 7.267e+09 perf-stat.i.instructions
2422 +7.9% 2615 perf-stat.i.instructions-per-iTLB-miss
0.75 ± 33% +42.8% 1.08 ± 4% perf-stat.i.metric.K/sec
39.57 -2.5% 38.59 perf-stat.i.metric.M/sec
1079593 ± 2% -15.7% 910198 ± 2% perf-stat.i.node-load-misses
109822 ± 4% -10.3% 98550 ± 4% perf-stat.i.node-loads
655123 -15.7% 552328 perf-stat.i.node-store-misses
2.09 ± 4% -11.5% 1.85 ± 3% perf-stat.overall.MPKI
0.36 ± 4% -0.0 0.33 ± 2% perf-stat.overall.branch-miss-rate%
3.66 +2.2% 3.74 perf-stat.overall.cpi
6264 ± 5% +15.6% 7244 ± 5% perf-stat.overall.cycles-between-cache-misses
59.36 -2.1 57.22 perf-stat.overall.iTLB-load-miss-rate%
2433 +7.9% 2624 perf-stat.overall.instructions-per-iTLB-miss
0.27 -2.1% 0.27 perf-stat.overall.ipc
1.548e+09 -1.9% 1.519e+09 perf-stat.ps.branch-instructions
5566347 ± 4% -8.8% 5075900 ± 2% perf-stat.ps.branch-misses
4265723 ± 5% -13.3% 3697689 ± 5% perf-stat.ps.cache-misses
15234536 ± 4% -13.1% 13232157 ± 3% perf-stat.ps.cache-references
2919 -7.4% 2703 perf-stat.ps.context-switches
2.634e+08 -11.9% 2.322e+08 perf-stat.ps.dTLB-stores
2996642 -9.1% 2724338 perf-stat.ps.iTLB-load-misses
7.29e+09 -1.9% 7.15e+09 perf-stat.ps.instructions
1062628 ± 2% -15.7% 895597 ± 2% perf-stat.ps.node-load-misses
108035 ± 4% -10.3% 96952 ± 4% perf-stat.ps.node-loads
644869 -15.7% 543523 perf-stat.ps.node-store-misses
4.605e+11 -1.8% 4.521e+11 perf-stat.total.instructions
stress-ng.time.involuntary_context_switches
18500 +-------------------------------------------------------------------+
18000 |-+ .+ + + |
|: + .+ + + + + + .+ +. + .+. |
17500 |:+ + + + +.+.+ + + +.+ + : + +. +. .+.+.|
17000 |-+ + + .+.+. .+.+ : + +. + + |
| + + + + |
16500 |-+ |
16000 |-+ |
15500 |-+ |
| |
15000 |-+ O O O |
14500 |-+ O O O O O O O O O O O O O O O O |
| O O O O O O O O O O O |
14000 |-O O |
13500 +-------------------------------------------------------------------+
stress-ng.rtc.ops
270000 +------------------------------------------------------------------+
|.+.+.+.+.+.+.+.+.+.+.+.+.+.+.+.+ +.+ +.+.+ +.+ + +.+.|
260000 |-+ |
250000 |-+ |
| |
240000 |-+ |
230000 |-+ |
| |
220000 |-+ |
210000 |-+ |
| O O O O O O O O O O O O OO O O O O O O O O O O O O O O O |
200000 |-+ |
190000 |-+ O |
| O O O |
180000 +------------------------------------------------------------------+
stress-ng.rtc.ops_per_sec
4600 +--------------------------------------------------------------------+
|.+.+.+. .+. .+.+.+.+.+.+.+. .+.+..+.+.+.+.+.+.+.+.+.+.+.+.+.+.+.+.+.|
4400 |-+ + + + |
4200 |-+ |
| |
4000 |-+ |
| |
3800 |-+ |
| |
3600 |-+ |
3400 |-+ O O O O O O O O O O O O O O O O O O O O O O O O O O O O O |
| |
3200 |-+ O |
| O O O |
3000 +--------------------------------------------------------------------+
[*] 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,
Oliver Sang
View attachment "config-5.10.0-rc1-00034-g05a0302c3548" of type "text/plain" (171317 bytes)
View attachment "job-script" of type "text/plain" (8088 bytes)
View attachment "job.yaml" of type "text/plain" (5601 bytes)
View attachment "reproduce" of type "text/plain" (531 bytes)
Powered by blists - more mailing lists