public inbox for linux-fsdevel@vger.kernel.org
 help / color / mirror / Atom feed
From: kernel test robot <oliver.sang@intel.com>
To: Christian Brauner <brauner@kernel.org>
Cc: <oe-lkp@lists.linux.dev>, <lkp@intel.com>,
	<linux-kernel@vger.kernel.org>, Jan Kara <jack@suse.cz>,
	<linux-fsdevel@vger.kernel.org>, <oliver.sang@intel.com>
Subject: [linus:master] [pidfs]  8021824904: stress-ng.pthread.ops_per_sec 115.4% improvement
Date: Mon, 9 Mar 2026 13:50:58 +0800	[thread overview]
Message-ID: <202603091347.982f38c4-lkp@intel.com> (raw)



Hello,

kernel test robot noticed a 115.4% improvement of stress-ng.pthread.ops_per_sec on:


commit: 802182490445f6bcf5de0e0518fb967c2afb6da1 ("pidfs: convert rb-tree to rhashtable")
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master


testcase: stress-ng
config: x86_64-rhel-9.4
compiler: gcc-14
test machine: 192 threads 2 sockets Intel(R) Xeon(R) 6740E  CPU @ 2.4GHz (Sierra Forest) with 256G memory
parameters:

	nr_threads: 100%
	testtime: 60s
	test: pthread
	cpufreq_governor: performance



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


The kernel config and materials to reproduce are available at:
https://download.01.org/0day-ci/archive/20260309/202603091347.982f38c4-lkp@intel.com

=========================================================================================
compiler/cpufreq_governor/kconfig/nr_threads/rootfs/tbox_group/test/testcase/testtime:
  gcc-14/performance/x86_64-rhel-9.4/100%/debian-13-x86_64-20250902.cgz/lkp-srf-2sp3/pthread/stress-ng/60s

commit: 
  a344860211 ("ipc: Add SPDX license id to mqueue.c")
  8021824904 ("pidfs: convert rb-tree to rhashtable")

a344860211f5c07d 802182490445f6bcf5de0e0518f 
---------------- --------------------------- 
         %stddev     %change         %stddev
             \          |                \  
    238962 ±  2%     -56.6%     103737 ±  4%  stress-ng.pthread.nanosecs_to_start_a_pthread
   8502955          +115.5%   18320200        stress-ng.pthread.ops
    141479          +115.4%     304813        stress-ng.pthread.ops_per_sec
  11806823          +128.1%   26934722        stress-ng.time.involuntary_context_switches
  34893919          +112.5%   74144888        stress-ng.time.minor_page_faults
      8732           +13.1%       9872 ±  2%  stress-ng.time.percent_of_cpu_this_job_got
      5152           +11.6%       5748 ±  2%  stress-ng.time.system_time
    119.70           +79.2%     214.54        stress-ng.time.user_time
  17274026          +115.2%   37181651        stress-ng.time.voluntary_context_switches
 2.536e+09 ±  2%     +27.7%  3.237e+09        cpuidle..time
  11624156          +112.2%   24669418 ±  2%  cpuidle..usage
     21.06 ±  2%      +6.3       27.39        mpstat.cpu.all.idle%
      0.75 ±  2%      +0.1        0.87        mpstat.cpu.all.irq%
      0.45            +0.7        1.11 ±  4%  mpstat.cpu.all.soft%
     76.23            -8.2       68.00        mpstat.cpu.all.sys%
      1.52            +1.1        2.63        mpstat.cpu.all.usr%
     15273 ± 39%    +115.0%      32841 ±  6%  perf-c2c.DRAM.local
     11312 ± 42%    +462.1%      63587 ± 11%  perf-c2c.DRAM.remote
     17041 ± 39%    +521.2%     105868 ±  9%  perf-c2c.HITM.local
      5959 ± 41%    +610.3%      42332 ± 11%  perf-c2c.HITM.remote
     23001 ± 40%    +544.3%     148201 ±  9%  perf-c2c.HITM.total
      0.10 ±  9%     +44.6%       0.15        turbostat.IPC
  41291979 ±  9%     +46.8%   60611834        turbostat.IRQ
    262.50 ±  9%     +16.0%     304.39        turbostat.PkgWatt
     19.84 ±  5%     +23.0%      24.39        turbostat.RAMWatt
      0.02           +50.0%       0.03        turbostat.SysWatt
     23.63 ±  2%     +26.3%      29.83        vmstat.cpu.id
  10289045 ±  2%     +44.9%   14908391 ±  5%  vmstat.memory.cache
    781.86 ±  4%     +13.9%     890.45 ±  6%  vmstat.procs.r
    760169          +123.8%    1700926        vmstat.system.cs
    624448 ±  7%     +46.3%     913594        vmstat.system.in
     94.93           -94.9        0.00        perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe
     94.92           -94.9        0.00        perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe
     48.28           -48.3        0.00        perf-profile.calltrace.cycles-pp.__do_sys_clone3.do_syscall_64.entry_SYSCALL_64_after_hwframe
     48.27           -48.3        0.00        perf-profile.calltrace.cycles-pp.kernel_clone.__do_sys_clone3.do_syscall_64.entry_SYSCALL_64_after_hwframe
     46.51           -46.5        0.00        perf-profile.calltrace.cycles-pp.copy_process.kernel_clone.__do_sys_clone3.do_syscall_64.entry_SYSCALL_64_after_hwframe
     44.15           -44.2        0.00        perf-profile.calltrace.cycles-pp.alloc_pid.copy_process.kernel_clone.__do_sys_clone3.do_syscall_64
     44.00           -44.0        0.00        perf-profile.calltrace.cycles-pp.x64_sys_call.do_syscall_64.entry_SYSCALL_64_after_hwframe
     44.00           -44.0        0.00        perf-profile.calltrace.cycles-pp.__x64_sys_exit.x64_sys_call.do_syscall_64.entry_SYSCALL_64_after_hwframe
     44.00           -44.0        0.00        perf-profile.calltrace.cycles-pp.do_exit.__x64_sys_exit.x64_sys_call.do_syscall_64.entry_SYSCALL_64_after_hwframe
     43.77           -43.8        0.00        perf-profile.calltrace.cycles-pp._raw_spin_lock.alloc_pid.copy_process.kernel_clone.__do_sys_clone3
     43.74           -43.7        0.00        perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock.alloc_pid.copy_process.kernel_clone
     42.86           -42.9        0.00        perf-profile.calltrace.cycles-pp.exit_notify.do_exit.__x64_sys_exit.x64_sys_call.do_syscall_64
     42.71           -42.7        0.00        perf-profile.calltrace.cycles-pp.release_task.exit_notify.do_exit.__x64_sys_exit.x64_sys_call
     42.36           -42.4        0.00        perf-profile.calltrace.cycles-pp.free_pids.release_task.exit_notify.do_exit.__x64_sys_exit
     42.32           -42.3        0.00        perf-profile.calltrace.cycles-pp.free_pid.free_pids.release_task.exit_notify.do_exit
     42.15           -42.2        0.00        perf-profile.calltrace.cycles-pp._raw_spin_lock.free_pid.free_pids.release_task.exit_notify
     42.06           -42.1        0.00        perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock.free_pid.free_pids.release_task
     96.27           -96.3        0.00        perf-profile.children.cycles-pp.entry_SYSCALL_64_after_hwframe
     96.25           -96.2        0.00        perf-profile.children.cycles-pp.do_syscall_64
     87.32           -87.3        0.00        perf-profile.children.cycles-pp._raw_spin_lock
     87.27           -87.3        0.00        perf-profile.children.cycles-pp.native_queued_spin_lock_slowpath
     48.28           -48.3        0.00        perf-profile.children.cycles-pp.__do_sys_clone3
     48.27           -48.3        0.00        perf-profile.children.cycles-pp.kernel_clone
     46.51           -46.5        0.00        perf-profile.children.cycles-pp.copy_process
     44.15           -44.2        0.00        perf-profile.children.cycles-pp.alloc_pid
     44.01           -44.0        0.00        perf-profile.children.cycles-pp.x64_sys_call
     44.00           -44.0        0.00        perf-profile.children.cycles-pp.__x64_sys_exit
     44.00           -44.0        0.00        perf-profile.children.cycles-pp.do_exit
     42.86           -42.9        0.00        perf-profile.children.cycles-pp.exit_notify
     42.71           -42.7        0.00        perf-profile.children.cycles-pp.release_task
     42.36           -42.4        0.00        perf-profile.children.cycles-pp.free_pids
     42.32           -42.3        0.00        perf-profile.children.cycles-pp.free_pid
     86.50           -86.5        0.00        perf-profile.self.cycles-pp.native_queued_spin_lock_slowpath
      3.91           +39.0%       5.44        perf-stat.i.MPKI
 1.071e+10           +39.3%  1.491e+10        perf-stat.i.branch-instructions
      0.55            +0.3        0.89        perf-stat.i.branch-miss-rate%
  58562752          +125.4%   1.32e+08 ±  2%  perf-stat.i.branch-misses
     33.04            -1.1       31.91        perf-stat.i.cache-miss-rate%
 1.846e+08          +103.4%  3.755e+08        perf-stat.i.cache-misses
 5.592e+08          +110.8%  1.179e+09        perf-stat.i.cache-references
    791816          +124.8%    1780366        perf-stat.i.context-switches
     10.42           -34.7%       6.80        perf-stat.i.cpi
 4.923e+11            -6.9%  4.584e+11        perf-stat.i.cpu-cycles
    124149           +94.6%     241625        perf-stat.i.cpu-migrations
      2676           -52.5%       1270 ±  2%  perf-stat.i.cycles-between-cache-misses
 4.769e+10           +45.8%  6.954e+10        perf-stat.i.instructions
      0.10           +55.4%       0.15        perf-stat.i.ipc
     10.82          +128.3%      24.70        perf-stat.i.metric.K/sec
    584711          +110.0%    1228050        perf-stat.i.minor-faults
    722714          +111.3%    1527264        perf-stat.i.page-faults
      3.89           +39.3%       5.42        perf-stat.overall.MPKI
      0.55            +0.3        0.89        perf-stat.overall.branch-miss-rate%
     33.04            -1.2       31.87        perf-stat.overall.cache-miss-rate%
     10.34           -36.2%       6.60        perf-stat.overall.cpi
      2659           -54.2%       1217 ±  2%  perf-stat.overall.cycles-between-cache-misses
      0.10           +56.8%       0.15        perf-stat.overall.ipc
 1.041e+10           +40.0%  1.458e+10        perf-stat.ps.branch-instructions
  57180163          +126.3%  1.294e+08 ±  2%  perf-stat.ps.branch-misses
 1.804e+08          +104.2%  3.683e+08        perf-stat.ps.cache-misses
  5.46e+08          +111.7%  1.156e+09        perf-stat.ps.cache-references
    774516          +125.5%    1746770        perf-stat.ps.context-switches
 4.797e+11            -6.5%  4.483e+11        perf-stat.ps.cpu-cycles
    121274           +95.3%     236908        perf-stat.ps.cpu-migrations
 4.639e+10           +46.5%  6.798e+10        perf-stat.ps.instructions
    568315          +111.7%    1203024        perf-stat.ps.minor-faults
    703458          +112.8%    1496627        perf-stat.ps.page-faults
 2.724e+12 ±  9%     +53.6%  4.184e+12        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:[~2026-03-09  5:51 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=202603091347.982f38c4-lkp@intel.com \
    --to=oliver.sang@intel.com \
    --cc=brauner@kernel.org \
    --cc=jack@suse.cz \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=lkp@intel.com \
    --cc=oe-lkp@lists.linux.dev \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox