| 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, 2 Jun 2016 14:45:07 +0800
From: kernel test robot <xiaolong.ye@...el.com>
To: Johannes Weiner <hannes@...xchg.org>
Cc: Linus Torvalds <torvalds@...ux-foundation.org>,
Mel Gorman <mgorman@...e.de>, Rik van Riel <riel@...hat.com>,
David Rientjes <rientjes@...gle.com>,
Joonsoo Kim <iamjoonsoo.kim@....com>,
Andrew Morton <akpm@...ux-foundation.org>,
LKML <linux-kernel@...r.kernel.org>, lkp@...org
Subject: [lkp] [mm] 795ae7a0de: pixz.throughput -9.1% regression
FYI, we noticed pixz.throughput -9.1% regression due to commit:
commit 795ae7a0de6b834a0cc202aa55c190ef81496665 ("mm: scale kswapd watermarks in proportion to memory")
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
in testcase: pixz
on test machine: ivb43: 48 threads Ivytown Ivy Bridge-EP with 64G memory with following parameters: cpufreq_governor=performance/nr_threads=100%
In addition to that, the commit also has significant impact on the following tests:
will-it-scale: will-it-scale.per_process_ops -8.5% regression on test machine - ivb42: 48 threads Ivytown Ivy Bridge-EP with 64G memory
with test parameters: cpufreq_governor=performance/test=page_fault1
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
=========================================================================================
compiler/cpufreq_governor/kconfig/nr_threads/rootfs/tbox_group/testcase:
gcc-4.9/performance/x86_64-rhel/100%/debian-x86_64-2015-02-07.cgz/ivb43/pixz
commit:
3ed3a4f0ddffece942bb2661924d87be4ce63cb7
795ae7a0de6b834a0cc202aa55c190ef81496665
3ed3a4f0ddffece9 795ae7a0de6b834a0cc202aa55
---------------- --------------------------
%stddev %change %stddev
\ | \
78505362 ± 0% -9.1% 71324131 ± 0% pixz.throughput
4530 ± 0% +1.0% 4575 ± 0% pixz.time.percent_of_cpu_this_job_got
14911 ± 0% +2.3% 15251 ± 0% pixz.time.user_time
6586930 ± 0% -7.5% 6093751 ± 1% pixz.time.voluntary_context_switches
49869 ± 1% -9.0% 45401 ± 0% vmstat.system.cs
26406 ± 4% -9.4% 23922 ± 5% numa-meminfo.node0.SReclaimable
4803 ± 85% -87.0% 625.25 ± 16% numa-meminfo.node1.Inactive(anon)
946.75 ± 3% +775.4% 8288 ± 1% proc-vmstat.nr_alloc_batch
2403080 ± 2% -58.4% 999765 ± 0% proc-vmstat.pgalloc_dma32
651.75 ± 5% +12.0% 730.25 ± 6% sched_debug.cfs_rq:/.util_avg.min
81.92 ± 23% -31.2% 56.34 ± 11% sched_debug.cfs_rq:/.util_avg.stddev
-16.00 ±-13% -23.2% -12.29 ± -9% sched_debug.cpu.nr_uninterruptible.min
193807 ± 15% -32.7% 130369 ± 10% cpuidle.C1-IVT.usage
8.866e+08 ± 2% -15.4% 7.498e+08 ± 5% cpuidle.C6-IVT.time
93283 ± 0% -13.2% 80986 ± 2% cpuidle.C6-IVT.usage
771388 ± 9% -38.5% 474559 ± 6% cpuidle.POLL.usage
454.25 ± 2% +772.9% 3965 ± 2% numa-vmstat.node0.nr_alloc_batch
6600 ± 4% -9.4% 5980 ± 5% numa-vmstat.node0.nr_slab_reclaimable
563.00 ± 2% +658.1% 4268 ± 2% numa-vmstat.node1.nr_alloc_batch
1201 ± 85% -87.0% 156.00 ± 16% numa-vmstat.node1.nr_inactive_anon
792.00 ± 11% +25.0% 990.00 ± 11% slabinfo.blkdev_requests.active_objs
792.00 ± 11% +25.0% 990.00 ± 11% slabinfo.blkdev_requests.num_objs
507.00 ± 9% +22.8% 622.75 ± 8% slabinfo.file_lock_cache.active_objs
507.00 ± 9% +22.8% 622.75 ± 8% slabinfo.file_lock_cache.num_objs
94.35 ± 0% +1.0% 95.28 ± 0% turbostat.%Busy
2824 ± 0% +1.0% 2852 ± 0% turbostat.Avg_MHz
3.57 ± 3% -21.1% 2.82 ± 1% turbostat.CPU%c1
22.36 ± 0% -8.2% 20.52 ± 0% turbostat.RAMWatt
0.54 ± 17% +89.8% 1.02 ± 25% perf-profile.cycles-pp.do_huge_pmd_anonymous_page
0.40 ± 65% +140.3% 0.96 ± 31% perf-profile.cycles-pp.mga_imageblit
2.49 ± 40% -63.2% 0.92 ± 94% perf-profile.cycles-pp.poll_idle
1.11 ± 61% -95.7% 0.05 ±149% perf-profile.cycles-pp.rest_init
1.11 ± 61% -95.7% 0.05 ±149% perf-profile.cycles-pp.start_kernel
1.11 ± 61% -95.7% 0.05 ±149% perf-profile.cycles-pp.x86_64_start_kernel
1.11 ± 61% -95.7% 0.05 ±149% perf-profile.cycles-pp.x86_64_start_reservations
pixz.throughput
8e+07 ++----------------------------------------------------------------+
7.9e+07 ++ *. .*. *. .*. *. .*. |
*. + * *.*.*..*.*.*. + * *.*. + * *.*.*..*.*.*.* |
7.8e+07 ++* * * |
7.7e+07 ++ |
| |
7.6e+07 ++ |
7.5e+07 ++ |
7.4e+07 ++ |
| |
7.3e+07 ++ |
7.2e+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 O O O O |
7.1e+07 ++ O O
7e+07 ++----------------------------------------------------------------+
[*] bisect-good sample
[O] bisect-bad sample
***************************************************************************************************
ivb42: 48 threads Ivytown Ivy Bridge-EP with 64G memory
=========================================================================================
compiler/cpufreq_governor/kconfig/rootfs/tbox_group/test/testcase:
gcc-4.9/performance/x86_64-rhel/debian-x86_64-2015-02-07.cgz/ivb42/page_fault1/will-it-scale
commit:
3ed3a4f0ddffece942bb2661924d87be4ce63cb7
795ae7a0de6b834a0cc202aa55c190ef81496665
3ed3a4f0ddffece9 795ae7a0de6b834a0cc202aa55
---------------- --------------------------
%stddev %change %stddev
\ | \
442409 ± 0% -8.5% 404670 ± 0% will-it-scale.per_process_ops
397397 ± 0% -6.2% 372741 ± 0% will-it-scale.per_thread_ops
0.11 ± 1% -15.1% 0.10 ± 0% will-it-scale.scalability
9933 ± 10% +17.8% 11696 ± 4% will-it-scale.time.involuntary_context_switches
5158470 ± 3% +5.4% 5438873 ± 0% will-it-scale.time.maximum_resident_set_size
10701739 ± 0% -11.6% 9456315 ± 0% will-it-scale.time.minor_page_faults
825.00 ± 0% +7.8% 889.75 ± 0% will-it-scale.time.percent_of_cpu_this_job_got
2484 ± 0% +7.8% 2678 ± 0% will-it-scale.time.system_time
81.98 ± 0% +8.7% 89.08 ± 0% will-it-scale.time.user_time
848972 ± 1% -13.3% 735967 ± 0% will-it-scale.time.voluntary_context_switches
19395253 ± 0% -20.0% 15511908 ± 0% numa-numastat.node0.local_node
19400671 ± 0% -20.0% 15518877 ± 0% numa-numastat.node0.numa_hit
7954 ± 2% -7.9% 7326 ± 4% vmstat.system.cs
21796 ± 0% +3.2% 22492 ± 0% vmstat.system.in
15.92 ± 27% +307.9% 64.92 ± 76% sched_debug.cfs_rq:/.util_avg.min
186.67 ± 37% -47.7% 97.58 ± 20% sched_debug.cpu.load.max
39.14 ± 22% -33.7% 25.95 ± 8% sched_debug.cpu.load.stddev
9933 ± 10% +17.8% 11696 ± 4% time.involuntary_context_switches
10701739 ± 0% -11.6% 9456315 ± 0% time.minor_page_faults
848972 ± 1% -13.3% 735967 ± 0% time.voluntary_context_switches
4654910 ± 7% -33.3% 3102519 ± 7% cpuidle.C3-IVT.time
9253 ± 10% -38.9% 5650 ± 7% cpuidle.C3-IVT.usage
1302023 ± 2% -15.8% 1096274 ± 1% cpuidle.C6-IVT.usage
8842237 ± 20% -51.7% 4269878 ± 15% cpuidle.POLL.time
578.75 ± 6% +640.7% 4286 ± 0% numa-vmstat.node0.nr_alloc_batch
9201698 ± 0% -19.2% 7432378 ± 0% numa-vmstat.node0.numa_hit
9127517 ± 0% -19.4% 7356249 ± 0% numa-vmstat.node0.numa_local
678.00 ± 3% +569.4% 4538 ± 5% numa-vmstat.node1.nr_alloc_batch
42.41 ± 0% +3.2% 43.75 ± 0% turbostat.%Busy
1270 ± 0% +3.2% 1310 ± 0% turbostat.Avg_MHz
0.03 ± 0% -33.3% 0.02 ± 0% turbostat.CPU%c3
15.59 ± 0% -10.4% 13.97 ± 0% turbostat.RAMWatt
8.81 ± 5% -41.9% 5.12 ± 54% perf-profile.cycles-pp.call_cpuidle
8.99 ± 5% -41.4% 5.27 ± 52% perf-profile.cycles-pp.cpu_startup_entry
8.81 ± 5% -42.0% 5.12 ± 54% perf-profile.cycles-pp.cpuidle_enter
8.18 ± 2% -41.4% 4.80 ± 53% perf-profile.cycles-pp.cpuidle_enter_state
8.16 ± 2% -41.3% 4.79 ± 53% perf-profile.cycles-pp.intel_idle
1169 ± 3% +654.1% 8820 ± 1% proc-vmstat.nr_alloc_batch
28654816 ± 0% -13.7% 24716699 ± 0% proc-vmstat.numa_hit
28645510 ± 0% -13.7% 24707399 ± 0% proc-vmstat.numa_local
3.309e+08 ± 0% -70.4% 97894895 ± 0% proc-vmstat.pgalloc_dma32
25869020 ± 0% -13.6% 22338006 ± 0% proc-vmstat.pgfault
1.437e+09 ± 0% -14.0% 1.236e+09 ± 0% proc-vmstat.pgfree
2755821 ± 0% -14.0% 2369967 ± 0% proc-vmstat.thp_deferred_split_page
2757377 ± 0% -14.0% 2371310 ± 0% proc-vmstat.thp_fault_alloc
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-4.5.0-02575-g795ae7a" of type "text/plain" (151562 bytes)
View attachment "job.yaml" of type "text/plain" (3554 bytes)
View attachment "reproduce" of type "text/plain" (4550 bytes)
Powered by blists - more mailing lists