Greeting,

FYI, we noticed a 1.9% improvement of fio.write_bw_MBps due to commit:


commit: 753598e253c8f50eb5c71fdc9a85df4df9cffbf2 ("mm: test code to write THP to swap device as a whole")
https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git master

in testcase: fio-basic
on test machine: 48 threads Intel(R) Xeon(R) CPU E5-2697 v2 @ 2.70GHz with 64G memory
with following parameters:

	runtime: 300s
	disk: 1HDD
	fs: xfs
	nr_task: 1
	test_size: 128G
	rw: randwrite
	bs: 4M
	ioengine: sync
	cpufreq_governor: performance

test-description: Fio is a tool that will spawn a number of threads or processes doing a particular type of I/O action as specified by the user.
test-url: https://github.com/axboe/fio



Details are as below:
-------------------------------------------------------------------------------------------------->


To reproduce:

        git clone https://github.com/01org/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: fio-basic/300s-1HDD-xfs-1-128G-randwrite-4M-sync-performance/ivb44

bd3c0e3c0819eb55  753598e253c8f50eb5c71fdc9a  
----------------  --------------------------  
         %stddev      change         %stddev
             \          |                \  
     35.16                       35.84        fio.write_iops
       140                         143        fio.write_bw_MBps
     28386                       27848        fio.write_clat_mean_us
     15416              -4%      14809        fio.write_clat_stddev
     40192              -4%      38528        fio.write_clat_90%_us
     48640              -9%      44416        fio.write_clat_99%_us
  86415360                    88078336        fio.time.file_system_outputs
    106174                      108997        iostat.sda.wkB/s
    107017                      109838        vmstat.io.bo
      0.73 ±  7%        15%       0.84        perf-stat.dTLB-load-miss-rate%
     39.41              -5%      37.50        perf-stat.node-load-miss-rate%
  42764176              -7%   39787608        perf-stat.iTLB-loads
 2.343e+08              -8%  2.154e+08        perf-stat.iTLB-load-misses



                                fio.write_clat_90__us

  40500 ++------------------------------------------------------------------+
        *.*.*.*.**.*.*.*.*.*.*.*.**.*.*.*.*.*.*.*.**.*.*.*.*.*.*.*.**.*.*.*.*
  40000 ++                                                                  |
        |                    O                                              |
  39500 ++                                                                  |
        |   O          O   O      O O O O                                   |
  39000 ++                                                                  |
        O O                    O            O                               |
  38500 ++                                                                  |
        |        O O O   O       O        O   O                             |
  38000 ++                                                                  |
        |                                                                   |
  37500 ++                                                                  |
        |                                                                   |
  37000 ++----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