All of lore.kernel.org
 help / color / mirror / Atom feed
From: kernel test robot <oliver.sang@intel.com>
To: Zhang Yi <yi.zhang@huawei.com>
Cc: <oe-lkp@lists.linux.dev>, <lkp@intel.com>,
	<linux-kernel@vger.kernel.org>, Theodore Ts'o <tytso@mit.edu>,
	Jan Kara <jack@suse.cz>, <linux-ext4@vger.kernel.org>,
	<oliver.sang@intel.com>
Subject: [linus:master] [ext4]  1ad0f42823:  fio.write_iops 23.8% improvement
Date: Thu, 30 Apr 2026 13:38:34 +0800	[thread overview]
Message-ID: <202604301021.faab1f8a-lkp@intel.com> (raw)



Hello,

kernel test robot noticed a 23.8% improvement of fio.write_iops on:


commit: 1ad0f42823291bcac371dafd37533f5e8d92acc3 ("ext4: move pagecache_isize_extended() out of active handle")
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master


testcase: fio-basic
config: x86_64-rhel-9.4
compiler: gcc-14
test machine: 64 threads 2 sockets Intel(R) Xeon(R) Gold 6346 CPU @ 3.10GHz (Ice Lake) with 256G memory
parameters:

	runtime: 300s
	disk: 1HDD
	fs: ext4
	nr_task: 1
	test_size: 128G
	rw: write
	bs: 4k
	ioengine: falloc
	cpufreq_governor: performance



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


The kernel config and materials to reproduce are available at:
https://download.01.org/0day-ci/archive/20260430/202604301021.faab1f8a-lkp@intel.com

=========================================================================================
bs/compiler/cpufreq_governor/disk/fs/ioengine/kconfig/nr_task/rootfs/runtime/rw/tbox_group/test_size/testcase:
  4k/gcc-14/performance/1HDD/ext4/falloc/x86_64-rhel-9.4/1/debian-13-x86_64-20250902.cgz/300s/write/lkp-icl-2sp9/128G/fio-basic

commit: 
  116c0bdac2 ("ext4: remove ctime/mtime update from ext4_alloc_file_blocks()")
  1ad0f42823 ("ext4: move pagecache_isize_extended() out of active handle")

116c0bdac2ec059d 1ad0f42823291bcac371dafd375 
---------------- --------------------------- 
         %stddev     %change         %stddev
             \          |                \  
      0.11 ±  5%      -0.0        0.08 ±  7%  fio.latency_2us%
      0.07 ±  3%      -0.0        0.05 ±  6%  fio.latency_4us%
     34.79           -19.1%      28.14        fio.time.elapsed_time
     34.79           -19.1%      28.14        fio.time.elapsed_time.max
     24.17           -26.0%      17.90        fio.time.system_time
      3736           -18.7%       3037        fio.time.voluntary_context_switches
      3812           +23.8%       4720        fio.write_bw_MBps
    850.67           -22.7%     657.33        fio.write_clat_90%_ns
    857.33           -22.4%     665.33        fio.write_clat_95%_ns
    872.00           -22.5%     676.00        fio.write_clat_99%_ns
    835.45           -22.7%     645.60        fio.write_clat_mean_ns
    192.67 ± 28%    +122.3%     428.28 ± 20%  fio.write_clat_stddev
    975889           +23.8%    1208502        fio.write_iops
      1.13            -7.7%       1.05        iostat.cpu.system
    146867           -17.0%     121930        turbostat.IRQ
 2.317e+09           -17.6%  1.908e+09        cpuidle..time
    121665 ±  9%     -11.5%     107660 ±  5%  cpuidle..usage
      1.13            -0.1        1.04        mpstat.cpu.all.sys%
      0.54 ±  2%      +0.1        0.62 ±  2%  mpstat.cpu.all.usr%
    124.09 ±  2%      -6.1%     116.50        uptime.boot
      7577 ±  2%      -6.3%       7099        uptime.idle
      1507 ±  4%     +13.8%       1714        vmstat.io.bo
      1.58 ±  3%      +9.7%       1.73 ±  2%  vmstat.procs.r
    256506            -3.5%     247467        proc-vmstat.numa_hit
    281823            -4.4%     269466        proc-vmstat.pgalloc_normal
    166349            -7.9%     153190        proc-vmstat.pgfault
    159495            -7.7%     147154        proc-vmstat.pgfree
      7811 ± 20%     -45.1%       4288        proc-vmstat.pgpgout
      7649            -8.2%       7019 ±  2%  proc-vmstat.pgreuse
      4.74 ± 63%      -4.7        0.00        perf-profile.calltrace.cycles-pp.do_mprotect_pkey.__x64_sys_mprotect.do_syscall_64.entry_SYSCALL_64_after_hwframe
      3.64 ±104%      -3.1        0.56 ±223%  perf-profile.calltrace.cycles-pp.acpi_idle_do_entry.acpi_idle_enter.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call
      3.64 ±104%      -3.1        0.56 ±223%  perf-profile.calltrace.cycles-pp.acpi_idle_enter.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call.do_idle
      3.64 ±104%      -3.1        0.56 ±223%  perf-profile.calltrace.cycles-pp.acpi_safe_halt.acpi_idle_do_entry.acpi_idle_enter.cpuidle_enter_state.cpuidle_enter
      3.64 ±104%      -3.1        0.56 ±223%  perf-profile.calltrace.cycles-pp.cpuidle_enter.cpuidle_idle_call.do_idle.cpu_startup_entry.start_secondary
      3.64 ±104%      -3.1        0.56 ±223%  perf-profile.calltrace.cycles-pp.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call.do_idle.cpu_startup_entry
      3.64 ±104%      -3.1        0.56 ±223%  perf-profile.calltrace.cycles-pp.cpuidle_idle_call.do_idle.cpu_startup_entry.start_secondary.common_startup_64
      4.95 ± 81%      -3.4        1.54 ±149%  perf-profile.children.cycles-pp.__split_vma
      3.64 ±104%      -3.1        0.56 ±223%  perf-profile.children.cycles-pp.acpi_idle_do_entry
      3.64 ±104%      -3.1        0.56 ±223%  perf-profile.children.cycles-pp.acpi_idle_enter
      3.64 ±104%      -3.1        0.56 ±223%  perf-profile.children.cycles-pp.acpi_safe_halt
      3.64 ±104%      -3.1        0.56 ±223%  perf-profile.children.cycles-pp.cpuidle_enter
      3.64 ±104%      -3.1        0.56 ±223%  perf-profile.children.cycles-pp.cpuidle_enter_state
      3.64 ±104%      -3.1        0.56 ±223%  perf-profile.children.cycles-pp.cpuidle_idle_call
      3.64 ±104%      -3.1        0.56 ±223%  perf-profile.children.cycles-pp.pv_native_safe_halt
      1.82 ±148%      +5.8        7.59 ± 39%  perf-profile.children.cycles-pp.__mmap_region
      1.82 ±148%      +6.5        8.34 ± 44%  perf-profile.children.cycles-pp.do_mmap
      1.82 ±148%      +6.5        8.34 ± 44%  perf-profile.children.cycles-pp.vm_mmap_pgoff
      0.03 ±  7%     +21.8%       0.04 ±  4%  perf-stat.i.MPKI
      0.26            +0.1        0.32        perf-stat.i.branch-miss-rate%
   4548711           +21.7%    5537175        perf-stat.i.branch-misses
    274432 ±  6%     +20.4%     330549 ±  4%  perf-stat.i.cache-misses
   1860596            +8.1%    2011597        perf-stat.i.cache-references
      0.47            +2.1%       0.48        perf-stat.i.cpi
      2398            +3.0%       2470        perf-stat.i.minor-faults
      2398            +3.0%       2470        perf-stat.i.page-faults
      0.03 ±  6%     +22.3%       0.04 ±  4%  perf-stat.overall.MPKI
      0.26            +0.1        0.32        perf-stat.overall.branch-miss-rate%
     14.77 ±  6%      +1.7       16.46 ±  4%  perf-stat.overall.cache-miss-rate%
      0.47            +1.8%       0.48        perf-stat.overall.cpi
     14177 ±  6%     -16.9%      11783 ±  5%  perf-stat.overall.cycles-between-cache-misses
      2.12            -1.8%       2.08        perf-stat.overall.ipc
      8603           -19.7%       6911        perf-stat.overall.path-length
  1.69e+09            -1.7%  1.662e+09        perf-stat.ps.branch-instructions
   4424559           +21.0%    5353882        perf-stat.ps.branch-misses
    267376 ±  6%     +19.9%     320480 ±  4%  perf-stat.ps.cache-misses
   1810896            +7.5%    1947007        perf-stat.ps.cache-references
 7.998e+09            -2.0%  7.838e+09        perf-stat.ps.instructions
      2334            +2.5%       2393        perf-stat.ps.minor-faults
      2334            +2.5%       2393        perf-stat.ps.page-faults
 2.887e+11           -19.7%  2.319e+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


                 reply	other threads:[~2026-04-30  5:38 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=202604301021.faab1f8a-lkp@intel.com \
    --to=oliver.sang@intel.com \
    --cc=jack@suse.cz \
    --cc=linux-ext4@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=lkp@intel.com \
    --cc=oe-lkp@lists.linux.dev \
    --cc=tytso@mit.edu \
    --cc=yi.zhang@huawei.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.