| 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: <20171201030627.GS21779@yexl-desktop>
Date: Fri, 1 Dec 2017 11:06:27 +0800
From: kernel test robot <xiaolong.ye@...el.com>
To: Andy Lutomirski <luto@...nel.org>
Cc: Thomas Gleixner <tglx@...utronix.de>,
LKML <linux-kernel@...r.kernel.org>,
Andy Lutomirski <luto@...capital.net>, lkp@...org
Subject: [lkp-robot] [x86/entry/64] 2be1f41c14:
will-it-scale.per_process_ops -18.8% regression
Greeting,
FYI, we noticed a -18.8% regression of will-it-scale.per_process_ops due to commit:
commit: 2be1f41c14fa1921cf2217483f48a0b9b423be37 ("x86/entry/64: Create a percpu SYSCALL entry trampoline")
https://git.kernel.org/cgit/linux/kernel/git/luto/linux.git x86/entry_stack
in testcase: will-it-scale
on test machine: 32 threads Intel(R) Xeon(R) CPU E5-2680 0 @ 2.70GHz with 64G memory
with following parameters:
test: lseek2
cpufreq_governor: performance
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
In addition to that, the commit also has significant impact on the following tests:
+------------------+------------------------------------------------------------------+
| testcase: change | vm-scalability: vm-scalability.median -1.1% improvement |
| test machine | 4 threads Intel(R) Core(TM) i5-2300 CPU @ 2.80GHz with 4G memory |
| test parameters | cpufreq_governor=performance |
| | runtime=300s |
| | test=lru-file-readtwice |
+------------------+------------------------------------------------------------------+
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
testcase/path_params/tbox_group/run: will-it-scale/lseek2-performance/lkp-sb03
9f2327b9ef846266 2be1f41c14fa1921cf2217483f
---------------- --------------------------
470 17% 551 will-it-scale.time.user_time
0.64 8% 0.69 will-it-scale.scalability
2039 -4% 1958 will-it-scale.time.system_time
9781767 -13% 8489674 will-it-scale.per_thread_ops
16296316 -19% 13231829 will-it-scale.per_process_ops
fail:runs %reproduction fail:runs
| | |
:4 25% 1:4 dmesg.RIP:load_balance
:4 50% 2:4 dmesg.RIP:__fget
:4 25% 1:4 dmesg.RIP:native_sched_clock
:4 25% 1:4 dmesg.RIP:__update_idle_core
:4 25% 1:4 dmesg.RIP:rcu_all_qs
:4 25% 1:4 dmesg.RIP:__update_load_avg_se
:4 25% 1:4 dmesg.RIP:delay_tsc
:4 25% 1:4 dmesg.RIP:prepare_to_wait
:4 25% 1:4 dmesg.RIP:trigger_load_balance
4:4 -100% :4 dmesg.RIP:entry_SYSCALL_64
1:4 -25% :4 dmesg.RIP:__switch_to
1:4 -25% :4 dmesg.RIP:ktime_get_update_offsets_now
1:4 -25% :4 dmesg.RIP:update_curr
1:4 -25% :4 dmesg.RIP:_raw_spin_lock
1:4 -25% :4 dmesg.RIP:__tick_nohz_idle_enter
1:4 -25% :4 dmesg.RIP:native_write_msr
0.01 48965% 3.82 perf-stat.branch-miss-rate%
2.349e+08 42829% 1.009e+11 perf-stat.branch-misses
1.128e+08 ± 17% 41% 1.588e+08 ± 16% perf-stat.iTLB-loads
1.20 17% 1.39 perf-stat.cpi
0.00 ± 9% 15% 0.00 perf-stat.dTLB-store-miss-rate%
2.688e+12 ± 3% -12% 2.353e+12 perf-stat.dTLB-stores
3.02e+12 -13% 2.642e+12 perf-stat.branch-instructions
0.84 -14% 0.72 perf-stat.ipc
1.351e+13 -14% 1.156e+13 perf-stat.instructions
4.395e+12 -15% 3.754e+12 perf-stat.dTLB-loads
will-it-scale.per_process_ops
1.7e+07 +-+--------------------------------------------------------------+
|.+.+..+. .+. .+.+..+.+.+.+.+.+..+.+.+.+.+.+..+.+.+.+.+.+.. |
1.65e+07 +-+ + + +.+.+.|
1.6e+07 +-+ |
| |
1.55e+07 +-+ |
1.5e+07 +-+ |
| |
1.45e+07 +-+ |
1.4e+07 +-+ |
| |
1.35e+07 +-+ |
1.3e+07 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 |
1.25e+07 +-+--------------------------------------------------------------+
perf-stat.instructions
1.4e+13 +-+--------------------------------------------------------------+
| .+.. .+. .+. |
1.35e+13 +-+ .+. .+. .+ + +.+.+.+..+ +.+.+.+..+.+.+.|
|.+.+..+.+.+.+ +..+ +.+ |
| |
1.3e+13 +-+ |
| |
1.25e+13 +-+ |
| |
1.2e+13 +-+ |
| O O |
O O O O O O O O O O O O O O O O O O |
1.15e+13 +-+ O O O O O O O |
| |
1.1e+13 +-+--------------------------------------------------------------+
perf-stat.branch-instructions
3.2e+12 +-+---------------------------------------------------------------+
| .+. .+. .+ .+. |
3.1e+12 +-+.+.. .+ +. .+. .+. +..+ + + .+ +.+. |
|.+ + +. + +.+.. + +. +..+. |
3e+12 +-+ .+. .+ +.|
| + + |
2.9e+12 +-+ |
| |
2.8e+12 +-+ |
| |
2.7e+12 +-+ O O O O O O O |
| O O O O O O O O O O O O |
2.6e+12 O-+ O O O O O O O |
| |
2.5e+12 +-+---------------------------------------------------------------+
perf-stat.branch-misses
1.2e+11 +-+---------------------------------------------------------------+
| |
1e+11 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 |
| |
8e+10 +-+ |
| |
6e+10 +-+ |
| |
4e+10 +-+ |
| |
| |
2e+10 +-+ |
| |
0 +-+---------------------------------------------------------------+
perf-stat.dTLB-stores
3e+12 +-+---------------------------------------------------------------+
| .+. .+ |
2.9e+12 +-+ + + : |
| : : |
2.8e+12 +-+ : : |
| : : .+.+ |
2.7e+12 +-+ : : .+ .+. :|
| .+.. .+. .+.. .+.+.+. : : .+.+.+.+..+ + .+ :|
2.6e+12 +-+ +.+ + + + +..+ + |
| |
2.5e+12 +-+ |
| |
2.4e+12 +-+ O O O |
| O O O O O O O O O O O O O O O O O O O |
2.3e+12 O-+---------------------O----O-----O--------O---------------------+
perf-stat.branch-miss-rate_
4 +-+-------------------------------------------------------------------+
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 |
3.5 +-+ |
3 +-+ |
| |
2.5 +-+ |
| |
2 +-+ |
| |
1.5 +-+ |
1 +-+ |
| |
0.5 +-+ |
| |
0 +-+-------------------------------------------------------------------+
perf-stat.ipc
0.86 +-+------------------------------------------------------------------+
| .+.. .+.+ .+.. |
0.84 +-+ .+.+.. .+ + + .+.+. .+.+.+..+.+.+ +.|
0.82 +-+..+. .+..+.+ +.+.+..+ +. +. |
| + |
0.8 +-+ |
| |
0.78 +-+ |
| |
0.76 +-+ |
0.74 +-+ |
| |
0.72 +-O O O O O O O O O O |
O O O O O O O O O O O O O O |
0.7 +-+-------------O------O-O-------------------------------------------+
perf-stat.cpi
1.45 +-+------------------------------------------------------------------+
| O O |
1.4 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 |
| |
1.35 +-+ |
| |
1.3 +-+ |
| |
1.25 +-+ |
| |
|.+.. .+.+..+.+. .+.+.+..+. .+.. .+.. .|
1.2 +-+ + +.+. +. .+. .+ +.+ +.+.+..+.+.+.+..+ |
| +. + |
1.15 +-+------------------------------------------------------------------+
will-it-scale.time.user_time
580 +-+-------------------------------------------------------------------+
O O O |
560 +-+ O O O O O O O O O O O O O O O O O O |
540 +-+ O O O O O O |
| |
520 +-+ |
500 +-+ |
| |
480 +-+ .+.. |
460 +-+ +.+. .+ +.|
|.+..+.+.+..+.+.+..+.+.+..+.+ + +..+.+.+..+.+.+..+ |
440 +-+ : + |
420 +-+ : .+. .+ |
| +. + |
400 +-+-------------------------------------------------------------------+
will-it-scale.time.system_time
2100 +-+------------------------------------------------------------------+
| +.+..+.+ |
2080 +-+ + : |
2060 +-+..+.+. .+. .+..+. .+..+ : .+.+.+..+.+.+..+.+ |
| +. +.+ + +.+. + |
2040 +-+ +.+..+.|
2020 +-+ |
| |
2000 +-+ |
1980 +-+ |
| O O O |
1960 +-+ O O O O O O O O O O O O O O O O O O O O |
1940 O-+ O O |
| O |
1920 +-+------------------------------------------------------------------+
[*] 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,
Xiaolong
View attachment "config-4.14.0-01232-g2be1f41" of type "text/plain" (163502 bytes)
View attachment "job-script" of type "text/plain" (7128 bytes)
View attachment "job.yaml" of type "text/plain" (4801 bytes)
View attachment "reproduce" of type "text/plain" (328 bytes)
Powered by blists - more mailing lists