[jlayton:delstid] [nfsd] 1436c81cbe: filebench.sum_operations/s 15.7% regression

[jlayton:delstid] [nfsd] 1436c81cbe: filebench.sum_operations/s 15.7% regression

[kernel test robot]

Hello,

kernel test robot noticed a 15.7% regression of filebench.sum_operations/s on:


commit: 1436c81cbe9bef840cccc8a921948238e281442b ("nfsd: handle delegated timestamps in SETATTR")
https://git.kernel.org/cgit/linux/kernel/git/jlayton/linux.git delstid

testcase: filebench
config: x86_64-rhel-9.4
compiler: gcc-12
test machine: 128 threads 2 sockets Intel(R) Xeon(R) Platinum 8358 CPU @ 2.60GHz (Ice Lake) with 128G memory
parameters:

	disk: 1HDD
	fs: ext4
	fs2: nfsv4
	test: fileserver.f
	cpufreq_governor: performance




If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <oliver.sang@intel.com>
| Closes: https://lore.kernel.org/oe-lkp/202412252119.8934ca82-lkp@intel.com


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


The kernel config and materials to reproduce are available at:
https://download.01.org/0day-ci/archive/20241225/202412252119.8934ca82-lkp@intel.com

=========================================================================================
compiler/cpufreq_governor/disk/fs2/fs/kconfig/rootfs/tbox_group/test/testcase:
  gcc-12/performance/1HDD/nfsv4/ext4/x86_64-rhel-9.4/debian-12-x86_64-20240206.cgz/lkp-icl-2sp6/fileserver.f/filebench

commit: 
  e42e5990ce ("nfsd: add support for delegated timestamps")
  1436c81cbe ("nfsd: handle delegated timestamps in SETATTR")

e42e5990ceb8ddbb 1436c81cbe9bef840cccc8a9219 
---------------- --------------------------- 
         %stddev     %change         %stddev
             \          |                \  
   5176774           +10.1%    5700789        cpuidle..usage
     81.80 ± 16%     +37.1%     112.11 ± 21%  perf-sched.wait_time.max.ms.io_schedule.folio_wait_bit_common.folio_wait_writeback.__filemap_fdatawait_range
      1467            -5.2%       1390        vmstat.io.bo
      2162            -4.1%       2073        vmstat.system.cs
      2962            +6.0%       3138        vmstat.system.in
    613132            -1.5%     604009        proc-vmstat.nr_dirtied
     20689            +3.2%      21341        proc-vmstat.nr_shmem
    613062            -1.5%     603944        proc-vmstat.nr_written
   3388546            +3.8%    3516202        proc-vmstat.numa_hit
   3255944            +3.9%    3383613        proc-vmstat.numa_local
   4280567            +3.1%    4415074        proc-vmstat.pgalloc_normal
   3620218            +5.9%    3833758        proc-vmstat.pgfault
   4169646            +3.0%    4296404        proc-vmstat.pgfree
    158758            +5.9%     168050        proc-vmstat.pgreuse
      4.87           -16.1%       4.08        filebench.sum_bytes_mb/s
     12799           -15.7%      10791        filebench.sum_operations
    213.31           -15.7%     179.83        filebench.sum_operations/s
     19.00           -15.8%      16.00        filebench.sum_reads/s
    232.79           +18.3%     275.40        filebench.sum_time_ms/op
     39.00           -15.4%      33.00        filebench.sum_writes/s
      1427            +5.6%       1507        filebench.time.elapsed_time
      1427            +5.6%       1507        filebench.time.elapsed_time.max
   2377240            -2.1%    2328234        filebench.time.file_system_outputs
     58841            -4.9%      55954        filebench.time.voluntary_context_switches
      2.70 ± 10%      -0.4        2.30 ±  7%  perf-profile.calltrace.cycles-pp.asm_exc_page_fault
      0.36 ± 22%      -0.2        0.16 ± 34%  perf-profile.children.cycles-pp.irqentry_enter
      0.29 ± 25%      -0.1        0.14 ± 37%  perf-profile.children.cycles-pp.kfree
      0.39 ± 10%      -0.1        0.31 ± 15%  perf-profile.children.cycles-pp.wait4
      0.02 ±223%      +0.1        0.14 ± 27%  perf-profile.children.cycles-pp.__kmalloc_noprof
      0.43 ± 13%      -0.2        0.25 ± 15%  perf-profile.self.cycles-pp.kmem_cache_free
      0.18 ± 23%      -0.1        0.06 ± 80%  perf-profile.self.cycles-pp.irqentry_enter
      0.23 ± 29%      -0.1        0.11 ± 54%  perf-profile.self.cycles-pp.kfree
      0.10 ± 57%      +0.1        0.19 ± 17%  perf-profile.self.cycles-pp.__block_commit_write
      0.01 ±223%      +0.1        0.11 ± 32%  perf-profile.self.cycles-pp.__kmalloc_noprof
      0.09 ± 53%      +0.1        0.19 ± 24%  perf-profile.self.cycles-pp.rcu_sched_clock_irq
  32545397            -1.4%   32080624        perf-stat.i.branch-instructions
      5.56            +0.1        5.69        perf-stat.i.branch-miss-rate%
  11214449            +4.0%   11660477        perf-stat.i.cache-references
      2150            -4.2%       2059        perf-stat.i.context-switches
      2.25            +1.3%       2.28        perf-stat.i.cpi
 1.585e+08            -1.4%  1.563e+08        perf-stat.i.instructions
      0.46            -1.2%       0.45        perf-stat.i.ipc
      5.32            +0.1        5.44        perf-stat.overall.branch-miss-rate%
      1.87            +1.9%       1.90        perf-stat.overall.cpi
      0.54            -1.9%       0.53        perf-stat.overall.ipc
  32487857            -1.4%   32026463        perf-stat.ps.branch-instructions
  11205362            +4.0%   11651731        perf-stat.ps.cache-references
      2148            -4.2%       2058        perf-stat.ps.context-switches
 1.583e+08            -1.4%   1.56e+08        perf-stat.ps.instructions
 2.261e+11            +4.9%  2.371e+11        perf-stat.total.instructions




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.


-- 
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki