| 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: Thu, 29 Dec 2016 01:57:22 +0800
From: kernel test robot <xiaolong.ye@...el.com>
To: Thomas Gleixner <tglx@...utronix.de>
Cc: Linus Torvalds <torvalds@...ux-foundation.org>,
Markus Trippelsdorf <markus@...ppelsdorf.de>,
Boris Ostrovsky <boris.ostrovsky@...cle.com>,
LKML <linux-kernel@...r.kernel.org>, lkp@...org
Subject: [lkp-developer] [smp/hotplug] b9d9d6911b: unixbench.score -25.5%
regression
Greeting,
FYI, we noticed a -25.5% regression of unixbench.score due to commit:
commit: b9d9d6911bd5c370ad4b3aa57d758c093d17aed5 ("smp/hotplug: Undo tglxs brainfart")
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
in testcase: unixbench
on test machine: 8 threads Intel(R) Core(TM) i7 CPU 870 @ 2.93GHz with 6G memory
with following parameters:
runtime: 300s
nr_task: 1
test: shell1
test-description: UnixBench is the original BYTE UNIX benchmark suite aims to test performance of Unix-like system.
test-url: https://github.com/kdlucas/byte-unixbench
In addition to that, the commit also has significant impact on the following tests:
+------------------+---------------------------------------------------------------------+
| testcase: change | will-it-scale: will-it-scale.per_process_ops 2.0% improvement |
| test machine | 32 threads Intel(R) Xeon(R) CPU E5-2680 0 @ 2.70GHz with 64G memory |
| test parameters | cpufreq_governor=performance |
| | test=pthread_mutex1 |
+------------------+---------------------------------------------------------------------+
Details are as below:
-------------------------------------------------------------------------------------------------->
To reproduce:
git clone git://git.kernel.org/pub/scm/linux/kernel/git/wfg/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: unixbench/300s-1-shell1/nhm-white
b4b8664d291ac199 b9d9d6911bd5c370ad4b3aa57d
---------------- --------------------------
\ | \
2132 ± 0% -25.5% 1589 ± 0% unixbench.score
531506 ± 1% -17.0% 441203 ± 3% unixbench.time.involuntary_context_switches
76750760 ± 0% -25.6% 57116561 ± 0% unixbench.time.minor_page_faults
134.00 ± 0% -3.0% 130.00 ± 0% unixbench.time.percent_of_cpu_this_job_got
350.25 ± 0% +12.1% 392.47 ± 0% unixbench.time.system_time
500.55 ± 0% -13.6% 432.52 ± 0% unixbench.time.user_time
2832130 ± 0% -26.0% 2096467 ± 0% unixbench.time.voluntary_context_switches
307226 ± 0% -81.7% 56119 ± 0% interrupts.CAL:Function_call_interrupts
19838 ± 0% -17.4% 16392 ± 0% meminfo.Shmem
1192 ± 8% +16.1% 1383 ± 6% slabinfo.kmalloc-512.active_objs
376530 ± 0% -12.2% 330611 ± 0% softirqs.RCU
27117 ± 1% -19.7% 21787 ± 3% vmstat.system.cs
27824 ± 0% -98.6% 396.00 ± 3% latency_stats.hits.sigsuspend.SyS_rt_sigsuspend.entry_SYSCALL_64_fastpath
22958 ± 20% +74.8% 40122 ±126% latency_stats.max.max
10429 ± 70% -97.4% 272.75 ± 68% latency_stats.max.submit_bio_wait.blkdev_issue_flush.ext4_sync_fs.sync_fs_one_sb.iterate_supers.sys_sync.entry_SYSCALL_64_fastpath
1.296e+08 ± 0% -98.7% 1709911 ± 3% latency_stats.sum.sigsuspend.SyS_rt_sigsuspend.entry_SYSCALL_64_fastpath
12641 ± 61% -80.1% 2515 ± 12% latency_stats.sum.submit_bio_wait.blkdev_issue_flush.ext4_sync_fs.sync_fs_one_sb.iterate_supers.sys_sync.entry_SYSCALL_64_fastpath
531506 ± 1% -17.0% 441203 ± 3% time.involuntary_context_switches
76750760 ± 0% -25.6% 57116561 ± 0% time.minor_page_faults
4958 ± 0% -17.4% 4097 ± 0% proc-vmstat.nr_shmem
68574363 ± 0% -25.4% 51144859 ± 0% proc-vmstat.numa_hit
68574329 ± 0% -25.4% 51144828 ± 0% proc-vmstat.numa_local
274537 ± 0% -25.6% 204343 ± 0% proc-vmstat.pgactivate
70178190 ± 0% -25.5% 52306133 ± 0% proc-vmstat.pgalloc_normal
77376625 ± 0% -25.4% 57723437 ± 0% proc-vmstat.pgfault
70169945 ± 0% -25.5% 52297119 ± 0% proc-vmstat.pgfree
3020 ± 1% -27.2% 2197 ± 1% proc-vmstat.thp_deferred_split_page
3024 ± 1% -27.2% 2201 ± 1% proc-vmstat.thp_fault_alloc
3.813e+11 ± 0% -22.9% 2.94e+11 ± 0% perf-stat.branch-instructions
3.17 ± 0% +1.7% 3.22 ± 0% perf-stat.branch-miss-rate%
1.207e+10 ± 0% -21.5% 9.468e+09 ± 0% perf-stat.branch-misses
2.83 ± 1% -22.5% 2.19 ± 1% perf-stat.cache-miss-rate%
7.017e+08 ± 1% -24.8% 5.276e+08 ± 1% perf-stat.cache-misses
17149215 ± 1% -19.7% 13776667 ± 3% perf-stat.context-switches
1.874e+12 ± 0% -12.8% 1.634e+12 ± 3% perf-stat.cpu-cycles
489367 ± 0% -24.1% 371545 ± 0% perf-stat.cpu-migrations
1.657e+09 ± 3% -19.5% 1.334e+09 ± 1% perf-stat.dTLB-load-misses
6.654e+11 ± 0% -23.4% 5.097e+11 ± 0% perf-stat.dTLB-loads
1.19e+09 ± 0% -22.1% 9.273e+08 ± 4% perf-stat.dTLB-store-misses
3.524e+11 ± 0% -20.5% 2.8e+11 ± 0% perf-stat.dTLB-stores
0.03 ± 0% +6.9% 0.03 ± 0% perf-stat.iTLB-load-miss-rate%
5.335e+08 ± 0% -18.1% 4.368e+08 ± 0% perf-stat.iTLB-load-misses
1.946e+12 ± 0% -23.4% 1.491e+12 ± 0% perf-stat.iTLB-loads
1.901e+12 ± 0% -27.1% 1.386e+12 ± 5% perf-stat.instructions
3563 ± 0% -10.9% 3173 ± 4% perf-stat.instructions-per-iTLB-miss
1.01 ± 0% -16.4% 0.85 ± 1% perf-stat.ipc
75575650 ± 0% -25.4% 56399669 ± 0% perf-stat.minor-faults
1.051e+08 ± 2% -22.8% 81186461 ± 2% perf-stat.node-loads
5.711e+08 ± 1% -28.1% 4.104e+08 ± 5% perf-stat.node-stores
75575650 ± 0% -25.4% 56399664 ± 0% perf-stat.page-faults
perf-stat.cpu-cycles
1.95e+12 ++---------------------------------------------------------------+
| .*.. |
1.9e+12 ++ .*.*..* *.*.*..*.*..*. |
1.85e+12 *+*..*.*..* *..*.*.*..*.*..*.*.*..*.*..*.*
| |
1.8e+12 ++ |
| |
1.75e+12 ++ |
| O O O O |
1.7e+12 ++ O O O |
1.65e+12 O+O O O O O O |
| O O O O |
1.6e+12 ++ O O O O |
| O O O |
1.55e+12 ++---------------------------------------------------------------+
perf-stat.instructions
2e+12 ++----------------------------------------------------------------+
| .*. |
1.9e+12 *+*..*.*..*.*..*.*. *.*..*.*..*.*..*.*..*.*.*..*.*..*.*..*.*..*.*
| |
1.8e+12 ++ |
| |
1.7e+12 ++ |
| |
1.6e+12 ++ |
| |
1.5e+12 ++ O O O O |
| O O O O |
1.4e+12 O+O O O O O O O O |
| O O O O O |
1.3e+12 ++-----------------------------------O----O-----------O-----------+
perf-stat.cache-misses
7.5e+08 ++----------------------------------------------------------------+
| |
| *.*.. .*..*. .*. *.*. *.*.. .*
7e+08 *+ .. *.*..* *.*.. .*. *..*. .. *..*. .. *.*..* |
| * * * * |
| |
6.5e+08 ++ |
| |
6e+08 ++ |
| |
| |
5.5e+08 ++ |
| O O O O O O O O O O O O O O |
O O O O O O O O O |
5e+08 ++---O----------------O-------------------------------------------+
perf-stat.branch-instructions
3.9e+11 ++----------------------------------------------------------------+
3.8e+11 *+*..*.*..*.*..*.*..*.*.*..*.*..*.*..*.*..*.*.*..*.*..*.*..*.*..*.*
| |
3.7e+11 ++ |
3.6e+11 ++ |
| |
3.5e+11 ++ |
3.4e+11 ++ |
3.3e+11 ++ |
| |
3.2e+11 ++ |
3.1e+11 ++ |
| |
3e+11 ++ O O O O O O O O O O O O O O O O O O |
2.9e+11 O+O----O-------O-O------O--------------------------O--------------+
perf-stat.branch-misses
1.25e+10 ++---------------------------------------------------------------+
| .*.*..*.*..*.*.*..*.*..*. |
1.2e+10 *+*..*.*..* *..*.*.*..*.*..*.*.*..*.*..*.*
| |
1.15e+10 ++ |
| |
1.1e+10 ++ |
| |
1.05e+10 ++ |
| |
1e+10 ++ |
O O O O O O O O O O O |
9.5e+09 ++ O O O O O O O O O O O O O O |
| |
9e+09 ++---------------------------------------------------------------+
perf-stat.dTLB-loads
6.8e+11 ++----------------------------------------------------------------+
*.*..*.*..*.*..*.*..*.*.*..*.*..*.*..*.*..*.*.*..*.*..*.*..*.*..*.*
6.6e+11 ++ |
6.4e+11 ++ |
| |
6.2e+11 ++ |
6e+11 ++ |
| |
5.8e+11 ++ |
5.6e+11 ++ |
| |
5.4e+11 ++ |
5.2e+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 |
5e+11 ++----------------------------------------------------------------+
perf-stat.dTLB-stores
3.6e+11 ++----------------------------------------------------------------+
*.*..*.*..*.*..*.*..*.*.*..*.*..*.*..*.*..*.*.*..*.*..*.*..*.*..*.*
3.5e+11 ++ |
3.4e+11 ++ |
| |
3.3e+11 ++ |
3.2e+11 ++ |
| |
3.1e+11 ++ |
3e+11 ++ |
| |
2.9e+11 ++ |
2.8e+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 |
2.7e+11 ++----------------------------------------------------------------+
perf-stat.dTLB-store-misses
1.3e+09 ++---------------------------------------------------------------+
| * |
1.25e+09 ++ *. + + |
1.2e+09 ++ .*.*.. + *.. .*.. .*.*..* + .*.*.. .*.*.. .*.. .*
|.*. * * * *. .* *. .* * * |
1.15e+09 *+ *. *. |
1.1e+09 ++ |
| |
1.05e+09 ++ |
1e+09 ++ O |
O |
9.5e+08 ++ O O O O |
9e+08 ++O O O O O O O O O O O O O O O |
| O O O O |
8.5e+08 ++---------------------------------------------------------------+
perf-stat.iTLB-loads
2e+12 ++---------------------------------------------------------------+
1.95e+12 ++*..*.*..*.*.*..*.*..*.*.*..*.*..*. .*.*.*..*. .*.*.*..*.*..*.*
* *. *. |
1.9e+12 ++ |
1.85e+12 ++ |
1.8e+12 ++ |
1.75e+12 ++ |
| |
1.7e+12 ++ |
1.65e+12 ++ |
1.6e+12 ++ |
1.55e+12 ++ |
| |
1.5e+12 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 |
1.45e+12 ++---------------------------------------------------------------+
perf-stat.iTLB-load-misses
5.6e+08 ++----------------------------------------------------------------+
| |
5.4e+08 ++*.. *.. .*..*. *.. *.. |
|+ *.*.. + *.* *..*.*.. + *.*. + *.*.. .*..*.*
5.2e+08 *+ *.*..* * *..* * |
| |
5e+08 ++ |
| |
4.8e+08 ++ |
| |
4.6e+08 ++ |
| |
4.4e+08 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 |
| |
4.2e+08 ++----------------------------------------------------------------+
perf-stat.node-loads
1.15e+08 ++---------------------------------------------------------------+
| *.. |
1.1e+08 ++ .*.. + .*.. .*.. *
1.05e+08 ++*..* .* *.*.*..*.*.. .*.* .*.* +|
* *.*.*. *.*. *.*. *.*..* |
1e+08 ++ |
| |
9.5e+07 ++ |
| |
9e+07 ++ |
8.5e+07 ++ |
O O O O |
8e+07 ++ O O O O O O O O O O O O O O |
| O O O O O O O |
7.5e+07 ++---------------------------------------------------------------+
perf-stat.node-stores
6e+08 ++----------------------------------------------------------------+
| *.*.. .*..*. .*. *.*. *.*.. .*
*. .. *.*..* *.*.. .*. *..*. .. *..*. .. *.*..* |
5.5e+08 ++* * * * |
| |
| |
5e+08 ++ |
| |
4.5e+08 ++ |
| O O O O O |
| O O O O O O O |
4e+08 O+O O O O |
| O O O O O O |
| O O |
3.5e+08 ++----------------------------------------------------------------+
perf-stat.page-faults
7.6e+07 *+*--*-*--*-*--*----*------*-*----*--*-*--*-*-*--*-*--*-*--*-*--*-*
7.4e+07 ++ *. *.*. * |
| |
7.2e+07 ++ |
7e+07 ++ |
6.8e+07 ++ |
6.6e+07 ++ |
| |
6.4e+07 ++ |
6.2e+07 ++ |
6e+07 ++ |
5.8e+07 ++ |
| O O O O O O O O O O |
5.6e+07 O+O O O O O O O O O O O O O O |
5.4e+07 ++----------------------------------------------------------------+
perf-stat.cpu-migrations
520000 ++-----------------------*---------*-------------------------------+
| *..*.*..*.*. *..*.*. : |
500000 ++ + : .*.. |
480000 *+*..*.*..* *..*.*..*.*..*.*..*.*..* *.*
| |
460000 ++ |
| |
440000 ++ |
| |
420000 ++ |
400000 ++ |
| O O O O O |
380000 O+O O O O O O O O O |
| O O O O O O O O O O |
360000 ++-----------------------------------------------------------------+
perf-stat.minor-faults
7.6e+07 *+*--*-*--*-*--*----*------*-*----*--*-*--*-*-*--*-*--*-*--*-*--*-*
7.4e+07 ++ *. *.*. * |
| |
7.2e+07 ++ |
7e+07 ++ |
6.8e+07 ++ |
6.6e+07 ++ |
| |
6.4e+07 ++ |
6.2e+07 ++ |
6e+07 ++ |
5.8e+07 ++ |
| O O O O O O O O O O |
5.6e+07 O+O O O O O O O O O O O O O O |
5.4e+07 ++----------------------------------------------------------------+
perf-stat.cache-miss-rate_
2.9 ++---*--------------------------------------*------------*------------+
*.. + *.*.. .*. .. *.*..*. .. *.*..*. .*
2.8 ++ * *..*.*..*.*..*.. .*..*.*. * * *. |
2.7 ++ * |
| |
2.6 ++ |
2.5 ++ |
| |
2.4 ++ |
2.3 ++ |
| O O |
2.2 ++ O O O O O O O |
2.1 ++ O O O O O O O O O O O O O |
O O O |
2 ++--------------------------------------------------------------------+
perf-stat.iTLB-load-miss-rate_
0.03 ++-----------------------------------------------------------------+
| |
0.0295 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 |
0.029 ++ |
| |
0.0285 ++ |
| |
0.028 ++ |
| *.. .*.. *. *.. |
0.0275 ++ *.* *.. .*..* + *..* : *.* *.. .*
|: + : * *.*..*. + : : + : * |
0.027 *+ + : * : .* + : |
| *.*..* *. * |
0.0265 ++-----------------------------------------------------------------+
perf-stat.ipc
1.02 *+-*----------------------------------*--*---------*--*---------*-*--+
1 ++ *..*.*..*.*..*.*..*..*.*..*.*..* *..*..* *..*.*. *
| |
0.98 ++ |
0.96 ++ |
| |
0.94 ++ |
0.92 ++ |
0.9 ++ |
| |
0.88 ++ |
0.86 ++ O O O O O O |
O O O O O O O O O O O |
0.84 ++ O O O O O O O O |
0.82 ++-------------------------------------------------------------------+
unixbench.score
2200 ++-------------------------------------------------------------------+
*..*.*..*.*.. .*..*.*..*..*.*..*.*..*.*..*.*..*
2100 ++ *.*..*.*..*..*.*..*.*..* |
| |
2000 ++ |
| |
1900 ++ |
| |
1800 ++ |
| |
1700 ++ |
| |
1600 ++ 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 |
1500 ++-------------------------------------------------------------------+
interrupts.CAL:Function_call_interrupts
350000 ++-----------------------------------------------------------------+
| |
300000 *+*..*.*..*.*..*.*..*.*..*.*..*.*..*.*..*.*..*.*..*.*..*.*..*.*..*.*
| |
| |
250000 ++ |
| |
200000 ++ |
| |
150000 ++ |
| |
| |
100000 ++ |
| |
50000 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------+
unixbench.time.user_time
510 ++--------------------------------------------------------------------+
| .*. .*. .*. .*..*
500 *+.*.*. *..*..*.*..*.*..*..*.*..*.*..*.*..*. *..*.*..*. *..* |
490 ++ |
| |
480 ++ |
470 ++ |
| |
460 ++ |
450 ++ |
| |
440 ++ |
430 ++ 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 |
420 ++--------------------------------------------------------------------+
unixbench.time.minor_page_faults
8e+07 ++----------------------------------------------------------------+
| |
*.*..*.*..*.*..*.*..*.*.*..*.*..*.*..*.*..*.*.*..*.*..*.*..*.*..*.*
7.5e+07 ++ |
| |
| |
7e+07 ++ |
| |
6.5e+07 ++ |
| |
| |
6e+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 |
5.5e+07 ++----------------------------------------------------------------+
[*] 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-4.10.0-rc1-00002-gb9d9d691" of type "text/plain" (155604 bytes)
View attachment "job-script" of type "text/plain" (6451 bytes)
View attachment "job.yaml" of type "text/plain" (4204 bytes)
View attachment "reproduce" of type "text/plain" (25 bytes)
Powered by blists - more mailing lists