Greeting,
There is no primary kpi change in this test, below is the data collected through multiple
monitors running background just for your information.
commit: e4ca24b7a8a47de06ea539b6559d011a7f47ae2b ("mm: fix handling PTE-mapped THPs
in page_referenced()")
https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git master
in testcase: chromeswap
on test machine: 8 threads Intel(R) Core(TM) i7-4770 CPU @ 3.40GHz with 8G memory
with following parameters:
nr_pmem: 1
selection: sequential
nr_procs: 50
swap_targets: 0.5
cpufreq_governor: performance
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:
chromeswap/1-sequential-50-0.5-performance/lkp-hsw-d01
9493a7b4766de76d e4ca24b7a8a47de06ea539b655
---------------- --------------------------
%stddev change %stddev
\ | \
433.33 ± 1% +14.2% 494.75 ± 2%
chromeswap.faults_swap_0.50_sequential_faults_95th_percentile
446.67 ± 1% +15.0% 513.50 ± 1%
chromeswap.faults_swap_0.50_sequential_faults_99th_percentile
12117099 ± 10% -26.6% 8892877 ± 3% chromeswap.time.minor_page_faults
2.64 ± 1% +11.9% 2.96 ± 5%
chromeswap.us_fault_swap_0.50_sequential_fault_time_95th_percentile
2.43 ± 1% -8.7% 2.21 ± 5%
chromeswap.us_fault_swap_0.50_sequential_fault_time_median
721065 ± 13% -19.8% 578011 ± 4% interrupts.CAL:Function_call_interrupts
12117099 ± 10% -26.6% 8892877 ± 3% time.minor_page_faults
12447239 ± 9% -25.9% 9223106 ± 3% perf-stat.minor-faults
8.433e+09 ± 0% -2.9% 8.187e+09 ± 2% perf-stat.node-loads
1.112e+08 ± 8% +33.4% 1.483e+08 ± 8% proc-vmstat.pgdeactivate
1371 ± 2% -25.5% 1022 ± 0% proc-vmstat.thp_split_page
23.91 ± 0% +4.2% 24.90 ± 2% turbostat.CorWatt
32.91 ± 0% +3.2% 33.97 ± 2% turbostat.PkgWatt
221.56 ± 4% -22.6% 171.46 ± 10% sched_debug.cfs_rq:/.runnable_load_avg.avg
10721 ±119% -193.9% -10072 ±-144% sched_debug.cfs_rq:/.spread0.avg
52123 ± 12% +30.5% 68030 ± 7% sched_debug.cpu.nr_load_updates.min
16878 ± 21% -37.8% 10494 ± 17% sched_debug.cpu.nr_load_updates.stddev
8290 ± 10% -12.3% 7272 ± 5% slabinfo.anon_vma_chain.active_objs
2352 ± 7% -19.8% 1886 ± 8% slabinfo.cred_jar.active_objs
2479 ± 5% -13.9% 2136 ± 8% slabinfo.cred_jar.num_objs
9471 ± 3% -26.4% 6973 ± 18% slabinfo.kmalloc-32.active_objs
9471 ± 3% -26.4% 6973 ± 18% slabinfo.kmalloc-32.num_objs
1491 ± 11% -24.6% 1124 ± 3% slabinfo.kmalloc-512.active_objs
1492 ± 11% -23.1% 1147 ± 4% slabinfo.kmalloc-512.num_objs
619071 ± 31% -7% 578011 ± 4% interrupts.CAL:Function_call_interrupts
33.22 33.97 turbostat.PkgWatt
24.10 3% 24.90 turbostat.CorWatt
7.053e+09 ± 33% 16% 8.187e+09 perf-stat.node-loads
10577095 ± 32% -13% 9223106 ± 3% perf-stat.minor-faults
chromeswap.faults_swap_0.50_sequential_faults_95th_percentile
600 ++--------------------------------------------------------------------+
| |
500 ++O O O O O O O O O |
| O O O O O O O O O O |
| *.*.* * *.*..*.*.*.*.*.* * *.*.* *.*.* *.*.*.*.* *.*
400 ++: : : : : : : : : : : : : |
| : : : : : : : : : : : : : |
300 ++ : : : : : : : : : : : : : : |
|: : : : : : : : : : : : : : : |
200 ++ : : : : : : : : : : : : : : |
|: : : : : : : : : : : : : : : |
|: : : : : : : : : : : : : : : |
100 ++ : : : : : : :: : |
| : : : : : : : : |
0 O+--O---*---O----------------O-*---*-------*-*--------*-----------*---+
chromeswap.faults_swap_0.50_sequential_faults_99th_percentile
600 ++--------------------------------------------------------------------+
| |
500 ++O O O O O O O O O O O O O O O O O O O |
| * .*
| *.*.* : *.*..*.*.*.*.*.* * *.*.* *.*.* *.*.*.*.* * |
400 ++: : : : : : : : : : : : : |
| : : : : : : : : : : : : : : |
300 ++ : : : : : : : : : : : : : : |
|: : : : : : : : : : : : : : : |
200 ++ : : : : : : : : : : : : : : |
|: : : : : : : : : : : : : : : |
|: : : : : : : : : : : : : : : |
100 ++ : : : : : : :: : |
| : : : : : : : : |
0 O+--O---*---O----------------O-*---*-------*-*--------*-----------*---+
[*] 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