| 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: <20180103024709.GB32584@yexl-desktop>
Date: Wed, 3 Jan 2018 10:47:09 +0800
From: kernel test robot <xiaolong.ye@...el.com>
To: Jeff Layton <jlayton@...hat.com>
Cc: LKML <linux-kernel@...r.kernel.org>,
Stephen Rothwell <sfr@...b.auug.org.au>, lkp@...org
Subject: [lkp-robot] [fs] aad9c3c4a9: vm-scalability.throughput +50.0%
improvement
Greeting,
FYI, we noticed a 50.0% improvement of vm-scalability.throughput due to commit:
commit: aad9c3c4a9737891cd470b99565c1ecfbf2d78c6 ("fs: handle inode->i_version more efficiently")
https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git master
in testcase: vm-scalability
on test machine: 4 threads Intel(R) Core(TM) i3-3220 CPU @ 3.30GHz with 8G memory
with following parameters:
runtime: 300s
size: 1T
test: msync-mt
cpufreq_governor: performance
test-description: The motivation behind this suite is to exercise functions and regions of the mm/ of the Linux kernel which are of interest to us.
test-url: https://git.kernel.org/cgit/linux/kernel/git/wfg/vm-scalability.git/
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/rootfs/runtime/size/tbox_group/test/testcase:
gcc-7/performance/x86_64-rhel-7.2/debian-x86_64-2016-08-31.cgz/300s/1T/lkp-ivb-d02/msync-mt/vm-scalability
commit:
9402b098d9 ("btrfs: only dirty the inode in btrfs_update_time if something was changed")
aad9c3c4a9 ("fs: handle inode->i_version more efficiently")
9402b098d9ffb4ce aad9c3c4a9737891cd470b9956
---------------- --------------------------
%stddev %change %stddev
\ | \
1533629 +50.0% 2299900 ± 2% vm-scalability.throughput
383407 +50.0% 574985 ± 2% vm-scalability.median
0.07 ± 26% +84.5% 0.12 ± 19% vm-scalability.median_stddev
1.899e+08 ± 2% +40.6% 2.67e+08 ± 2% vm-scalability.time.file_system_outputs
263.75 -5.2% 250.00 ± 2% vm-scalability.time.percent_of_cpu_this_job_got
481.09 -35.2% 311.56 vm-scalability.time.system_time
335.72 ± 2% +34.8% 452.69 ± 2% vm-scalability.time.user_time
34751652 -36.1% 22204080 vm-scalability.time.voluntary_context_switches
4.257e+08 ± 2% +39.4% 5.932e+08 ± 2% vm-scalability.workload
42910848 ± 3% +62.2% 69606949 ± 3% interrupts.CAL:Function_call_interrupts
342405 -14.9% 291457 softirqs.SCHED
304749 +42.3% 433808 ± 2% vmstat.io.bo
249395 -30.0% 174551 vmstat.system.cs
156123 ± 2% +56.9% 245018 ± 3% vmstat.system.in
23.84 -5.8 18.08 mpstat.cpu.idle%
0.01 ± 18% +0.0 0.01 ± 8% mpstat.cpu.soft%
47.75 -6.2 41.58 mpstat.cpu.sys%
25.86 +9.9 35.74 ± 3% mpstat.cpu.usr%
210816 ± 9% -19.2% 170317 ± 14% slabinfo.buffer_head.active_objs
5406 ± 9% -19.2% 4368 ± 14% slabinfo.buffer_head.active_slabs
210866 ± 9% -19.2% 170383 ± 14% slabinfo.buffer_head.num_objs
5406 ± 9% -19.2% 4368 ± 14% slabinfo.buffer_head.num_slabs
1.45e+08 ± 2% -50.4% 71881872 ± 4% cpuidle.C1.time
33814583 -33.5% 22474611 cpuidle.C1.usage
12171968 ± 26% +123.3% 27183370 ± 25% cpuidle.C1E.time
2847989 ± 20% +55.6% 4430881 ± 21% cpuidle.C3.time
8170 ± 6% +22.9% 10044 ± 10% cpuidle.C3.usage
84854134 ± 6% +39.3% 1.182e+08 ± 6% cpuidle.C6.time
92679 ± 5% +40.0% 129751 ± 6% cpuidle.C6.usage
4451553 ± 4% +12.7% 5016307 ± 7% meminfo.Committed_AS
396145 ± 49% +83.4% 726650 ± 36% meminfo.Inactive(anon)
823798 ± 9% -22.2% 641057 ± 7% meminfo.Inactive(file)
2083271 ± 9% -26.8% 1525429 ± 23% meminfo.Mapped
2219723 ± 8% -25.7% 1649289 ± 24% meminfo.MemAvailable
4270015 ± 4% +13.2% 4834851 ± 8% meminfo.Shmem
17965 ± 6% +34.1% 24092 ± 8% meminfo.Writeback
2640 +2.1% 2695 turbostat.Avg_MHz
33814132 -33.5% 22474013 turbostat.C1
11.69 -5.8 5.87 ± 3% turbostat.C1%
0.98 ± 27% +1.2 2.22 ± 25% turbostat.C1E%
8159 ± 6% +22.8% 10020 ± 10% turbostat.C3
0.23 ± 21% +0.1 0.36 ± 21% turbostat.C3%
92498 ± 5% +40.1% 129594 ± 6% turbostat.C6
6.83 ± 7% +2.8 9.63 ± 6% turbostat.C6%
17.09 ± 3% -13.1% 14.85 ± 9% turbostat.CPU%c1
0.05 ± 47% +142.1% 0.11 ± 22% turbostat.CPU%c3
91582377 ± 3% +58.4% 1.451e+08 ± 3% turbostat.IRQ
34.36 +0.8% 34.65 turbostat.PkgWatt
2382 ± 63% +100.6% 4781 ± 25% sched_debug.cfs_rq:/.load_avg.avg
8167 ± 73% +113.2% 17411 ± 26% sched_debug.cfs_rq:/.load_avg.max
3374 ± 75% +116.9% 7321 ± 27% sched_debug.cfs_rq:/.load_avg.stddev
9711903 -29.6% 6838151 sched_debug.cpu.nr_switches.avg
12121196 ± 5% -24.0% 9217555 ± 5% sched_debug.cpu.nr_switches.max
8342443 -36.8% 5270671 ± 2% sched_debug.cpu.nr_switches.min
9699577 -29.6% 6825116 sched_debug.cpu.sched_count.avg
12108558 ± 5% -24.0% 9205305 ± 5% sched_debug.cpu.sched_count.max
8333087 -36.9% 5259926 ± 2% sched_debug.cpu.sched_count.min
4191762 -32.6% 2823269 sched_debug.cpu.sched_goidle.avg
4475335 ± 3% -25.4% 3337952 ± 6% sched_debug.cpu.sched_goidle.max
4014538 -38.0% 2490817 ± 2% sched_debug.cpu.sched_goidle.min
187011 ± 35% +92.5% 360088 ± 25% sched_debug.cpu.sched_goidle.stddev
4987427 -30.1% 3486044 sched_debug.cpu.ttwu_count.avg
6234460 ± 5% -24.5% 4704313 ± 4% sched_debug.cpu.ttwu_count.max
4307230 -37.6% 2685788 ± 2% sched_debug.cpu.ttwu_count.min
115442 ± 4% -20.6% 91715 ± 11% sched_debug.cpu.ttwu_local.min
7.493e+11 ± 2% +19.4% 8.948e+11 perf-stat.branch-instructions
0.30 ± 2% -0.1 0.18 ± 2% perf-stat.branch-miss-rate%
2.227e+09 -29.3% 1.574e+09 perf-stat.branch-misses
23.15 ± 3% +15.6 38.73 ± 2% perf-stat.cache-miss-rate%
5.133e+09 ± 2% +43.2% 7.349e+09 ± 2% perf-stat.cache-misses
2.219e+10 ± 2% -14.5% 1.897e+10 perf-stat.cache-references
77672159 -30.9% 53710082 perf-stat.context-switches
1.15 -12.6% 1.00 perf-stat.cpi
684224 -53.5% 318492 ± 4% perf-stat.cpu-migrations
0.50 ± 18% -0.1 0.38 ± 11% perf-stat.dTLB-load-miss-rate%
6.891e+11 +13.7% 7.836e+11 perf-stat.dTLB-loads
3.327e+08 ± 5% +14.4% 3.806e+08 ± 5% perf-stat.dTLB-store-misses
3.302e+11 +9.0% 3.6e+11 perf-stat.dTLB-stores
19.94 ± 3% +10.4 30.32 perf-stat.iTLB-load-miss-rate%
5.582e+08 ± 4% -20.4% 4.445e+08 perf-stat.iTLB-load-misses
2.241e+09 ± 2% -54.4% 1.022e+09 perf-stat.iTLB-loads
2.871e+12 +15.4% 3.312e+12 perf-stat.instructions
5152 ± 3% +44.6% 7451 perf-stat.instructions-per-iTLB-miss
0.87 +14.4% 1.00 perf-stat.ipc
6745 -17.2% 5584 perf-stat.path-length
565.50 ± 15% +357.6% 2587 ± 38% proc-vmstat.allocstall_movable
7845 ± 13% +89.4% 14862 ± 28% proc-vmstat.allocstall_normal
2692 ± 17% +21.2% 3264 ± 8% proc-vmstat.kswapd_low_wmark_hit_quickly
23734403 ± 2% +40.6% 33370937 ± 2% proc-vmstat.nr_dirtied
37168 ± 6% -10.9% 33099 ± 10% proc-vmstat.nr_dirty
56817 ± 8% -25.2% 42500 ± 23% proc-vmstat.nr_dirty_background_threshold
113774 ± 8% -25.2% 85106 ± 23% proc-vmstat.nr_dirty_threshold
98919 ± 48% +83.7% 181713 ± 36% proc-vmstat.nr_inactive_anon
206165 ± 9% -22.5% 159735 ± 8% proc-vmstat.nr_inactive_file
521790 ± 9% -26.7% 382658 ± 23% proc-vmstat.nr_mapped
1066314 ± 4% +13.3% 1208373 ± 8% proc-vmstat.nr_shmem
4755 ± 4% +24.4% 5917 ± 10% proc-vmstat.nr_writeback
23729118 ± 2% +40.6% 33367754 ± 2% proc-vmstat.nr_written
98919 ± 48% +83.7% 181714 ± 36% proc-vmstat.nr_zone_inactive_anon
206107 ± 9% -22.5% 159675 ± 8% proc-vmstat.nr_zone_inactive_file
41158789 +47.0% 60498506 ± 3% proc-vmstat.numa_hit
41158789 +47.0% 60498506 ± 3% proc-vmstat.numa_local
26729060 ± 2% +33.7% 35731315 ± 3% proc-vmstat.pgactivate
25084109 ± 11% +72.2% 43199387 ± 12% proc-vmstat.pgalloc_normal
16873808 +53.6% 25924823 ± 9% proc-vmstat.pgdeactivate
40639141 ± 2% +47.0% 59748176 ± 3% proc-vmstat.pgfree
2644 ± 3% +8.2% 2861 ± 3% proc-vmstat.pgpgin
94918758 ± 2% +40.6% 1.335e+08 ± 2% proc-vmstat.pgpgout
16886430 +53.6% 25935650 ± 9% proc-vmstat.pgrefill
536748 ± 46% +56.7% 841292 ± 32% proc-vmstat.pgrotated
52835149 ± 5% +67.5% 88491650 ± 10% proc-vmstat.pgscan_kswapd
451899 ± 11% +100.2% 904615 ± 20% proc-vmstat.pgsteal_direct
26328123 ± 6% +69.9% 44732568 ± 12% proc-vmstat.pgsteal_kswapd
7126269 ± 4% +61.3% 11497297 ± 9% proc-vmstat.workingset_refault
vm-scalability.throughput
2.4e+06 +-+-----------------------------------------------O--------O------+
O O O O O |
2.3e+06 +-+ O O O O O |
2.2e+06 +-+ O O O O O O O O |
| O O O O O |
2.1e+06 +-+ O O |
2e+06 +-+ |
| |
1.9e+06 +-+ |
1.8e+06 +-+ |
| |
1.7e+06 +-+ .+.+.. .+.+. |
1.6e+06 +-+..+.+.+.. .+ +.+.+. +..+.+. |
| +.+. +..+.+.. .+. .+. .+..+.|
1.5e+06 +-+---------------------------------------------------------------+
vm-scalability.median
600000 +-+-----------------------------------------------O---------O------+
O O O O O |
| O O O O O O O O |
550000 +-+ O O O O O O O |
| O O O O O |
| |
500000 +-+ |
| |
450000 +-+ |
| |
| +.+..+. .+.+..+. .+. |
400000 +-+..+.+..+. .. +.+. +.+. +. |
| +.+ +..+.+..+.+.+..+.+..+.|
| |
350000 +-+----------------------------------------------------------------+
perf-stat.branch-misses
2.3e+09 +-+---------------------------------------------------------------+
| .+. +.+.. .|
2.2e+09 +-+.. +.+..+. .+..+ .+.+.. +.+..+ + +.. + + |
2.1e+09 +-+ + +..+.+ + .+..+ + + .. + |
| + + + + |
2e+09 +-+ |
1.9e+09 +-+ |
| |
1.8e+09 +-+ |
1.7e+09 +-+ |
| O O |
1.6e+09 +-+ O O O O O O O O O O |
1.5e+09 +-+ O O O O O O O O O O |
O O O O O |
1.4e+09 +-+---------------------------------------------------------------+
perf-stat.context-switches
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 |
5.5e+07 +-+ O O O O O O O O O |
| O O |
5e+07 +-+----------------------------O----------------------------------+
perf-stat.cpu-migrations
750000 +-+----------------------------------------------------------------+
700000 +-+ .+. .+. +.. .+. .+.. |
| .+. .+.+..+ +..+.+..+ +.. + +.+. +. .+ +.|
650000 +-+..+.+..+ +. +.+ +. |
600000 +-+ |
| |
550000 +-+ |
500000 +-+ |
450000 +-+ |
| |
400000 +-+ |
350000 O-+ O O O |
| O O O O O O O O O O O O O O O O O O O |
300000 +-+ O O O |
250000 +-+-------------------------------O--------------------------------+
perf-stat.cache-miss-rate_
42 +-+--------------------------------------------------------------------+
40 +-+ O O |
| O O O O O |
38 +-+ O O O O |
36 O-+O O O O O O O O O O O O O O O |
34 +-+ |
32 +-+ |
| |
30 +-+ |
28 +-+ |
26 +-+ .+.. |
24 +-+ .+.+..+. .+.+..+..+. .+ +. .|
|..+.+..+.+.. .+. +. +. +..+.+.. .+..+.+.. .+. |
22 +-+ + + + |
20 +-+--------------------------------------------------------------------+
perf-stat.branch-miss-rate_
0.32 +-+------------------------------------------------------------------+
| +. |
0.3 +-+ .+.+..+. .. +.. .|
0.28 +-+ .+.+..+. .+..+. .+.+. .+.+..+.+. + + |
|.+..+.+. +..+ +..+.+. +. |
0.26 +-+ |
| |
0.24 +-+ |
| |
0.22 +-+ |
0.2 +-+ |
| |
0.18 +-+ 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 |
0.16 O-O------------------------------------------------------------------+
perf-stat.ipc
1.02 +-+------------------------------------------------------------------+
| O O |
1 +-+ O O O O O O |
0.98 O-O O O O O O O O O O O O O O |
| O O O O |
0.96 +-+ |
| |
0.94 +-+ |
| |
0.92 +-+ |
0.9 +-+ |
| .+..+.+.. .+.+.+..+. |
0.88 +-+.. .+..+ +.+. +.. |
| +.+..+. .+ +.+..+. .+.+.. .+..+.|
0.86 +-+------------------------------------------------------------------+
perf-stat.cpi
1.16 +-+------------------------------------------------------------------+
| .+.+..+ : .+ +.+. +..+.|
1.14 +-+. : .+. |
1.12 +-+ +..+.+..+.+..+.+..+.+.+..+ |
| |
1.1 +-+ |
1.08 +-+ |
| |
1.06 +-+ |
1.04 +-+ |
| |
1.02 +-+ O O O O O O O O O O |
1 O-O O O O O O O O O O O O O O |
| O O |
0.98 +-+------------------------------------------------------------------+
vm-scalability.time.system_time
500 +-+-------------------------------------------------------------------+
| .+. +. + + +.|
| .+. .+.+. +..+.+..+. +..+.+.. .. +..+ +.+..+ |
450 +-+ +. +.. + +. .+.+ |
| + +. |
| |
400 +-+ |
| |
350 +-+ |
| O O O O |
O O O O O O O O O O O O O |
300 +-+ O O O O O O O O O |
| O |
| |
250 +-+-------------------------------------------------------------------+
vm-scalability.time.voluntary_context_switches
3.6e+07 +-+---------------------------------------------------------------+
| .+..+. +. + + + +.|
3.4e+07 +-+.. .+ +..+.+.+..+. .+..+.+.+.. .+. .. +..+ +..+.+ |
3.2e+07 +-+ + + + + |
| |
3e+07 +-+ |
| |
2.8e+07 +-+ |
| |
2.6e+07 +-+ |
2.4e+07 +-+ O O O O |
O O O O O O O O O O O O O |
2.2e+07 +-+ O O O O O O O O O |
| O |
2e+07 +-+---------------------------------------------------------------+
vm-scalability.time.file_system_outputs
2.8e+08 +-+-----------------------------------------------O--------O------+
2.7e+08 O-+ O O O |
| O O O O O O |
2.6e+08 +-+ O O O O O O O O |
2.5e+08 +-+ O O O O |
| O O |
2.4e+08 +-+ O |
2.3e+08 +-+ |
2.2e+08 +-+ |
| |
2.1e+08 +-+ |
2e+08 +-+.. .+.+.. +.+.+.. .+..+.+. .+. |
| + +. .. +.+ +..+ +..+. .+. .+. .+..+.|
1.9e+08 +-+ + +..+ +. + |
1.8e+08 +-+---------------------------------------------------------------+
vmstat.system.cs
260000 +-+----------------------------------------------------------------+
250000 +-+ .+. +.. +..+. .+..+.|
|.+.. .+..+ +..+.+..+.+. .+.+..+.+. + + +.+..+ |
240000 +-+ + +. +..+.+ + |
230000 +-+ |
| |
220000 +-+ |
210000 +-+ |
200000 +-+ |
| |
190000 +-O O O O |
180000 O-+ O O O O O O O O O |
| O O O O O O O O O O O O |
170000 +-+ O |
160000 +-+----------------------------------------------------------------+
perf-stat.path-length
7000 +-+------------------------------------------------------------------+
| |
6800 +-+ .+..+.+..+.+.. .+..+.|
6600 +-+ .+..+ .+. .+. .+ + |
|.+..+. .+ + .+. +.. .+ +..+.+. |
6400 +-+ +. +..+ +.+. |
| |
6200 +-+ |
| |
6000 +-+ |
5800 +-+ |
| O O O O O O O O O O O |
5600 +-+ O O O O O O O O |
O O O O O O O O |
5400 +-+------------------------------------------------------------------+
[*] 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-4.15.0-rc3-00022-gaad9c3c" of type "text/plain" (163796 bytes)
View attachment "job-script" of type "text/plain" (7150 bytes)
View attachment "job.yaml" of type "text/plain" (4774 bytes)
View attachment "reproduce" of type "text/plain" (7999 bytes)
Powered by blists - more mailing lists