All of lore.kernel.org
 help / color / mirror / Atom feed
From: kernel test robot <oliver.sang@intel.com>
To: Charlie Jenkins <charlie@rivosinc.com>
Cc: <oe-lkp@lists.linux.dev>, <lkp@intel.com>,
	<linux-kernel@vger.kernel.org>, <x86@kernel.org>,
	Thomas Gleixner <tglx@linutronix.de>,
	Alexandre Ghiti <alexghiti@rivosinc.com>, <oliver.sang@intel.com>
Subject: [tip:core/entry] [entry]  e43b8bb56e: stress-ng.personality.ops_per_sec 3.2% improvement
Date: Fri, 16 May 2025 10:39:32 +0800	[thread overview]
Message-ID: <202505160902.c72854e8-lkp@intel.com> (raw)



Hello,

kernel test robot noticed a 3.2% improvement of stress-ng.personality.ops_per_sec on:


commit: e43b8bb56e537bfc8d9076793091e7679020fc9c ("entry: Inline syscall_exit_to_user_mode()")
https://git.kernel.org/cgit/linux/kernel/git/tip/tip.git core/entry


testcase: stress-ng
config: x86_64-rhel-9.4
compiler: gcc-12
test machine: 64 threads 2 sockets Intel(R) Xeon(R) Gold 6346 CPU @ 3.10GHz (Ice Lake) with 256G memory
parameters:

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


In addition to that, the commit also has significant impact on the following tests:

+------------------+--------------------------------------------------------------------------------+
| testcase: change | stress-ng: stress-ng.context.swapcontext_calls_per_sec 2.1% improvement        |
| test machine     | 384 threads 2 sockets Intel(R) Xeon(R) 6972P (Granite Rapids) with 128G memory |
| test parameters  | cpufreq_governor=performance                                                   |
|                  | nr_threads=100%                                                                |
|                  | test=context                                                                   |
|                  | testtime=60s                                                                   |
+------------------+--------------------------------------------------------------------------------+




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


The kernel config and materials to reproduce are available at:
https://download.01.org/0day-ci/archive/20250516/202505160902.c72854e8-lkp@intel.com

=========================================================================================
compiler/cpufreq_governor/kconfig/nr_threads/rootfs/tbox_group/test/testcase/testtime:
  gcc-12/performance/x86_64-rhel-9.4/100%/debian-12-x86_64-20240206.cgz/lkp-icl-2sp7/personality/stress-ng/60s

commit: 
  7ace1602ab ("LoongArch: entry: Migrate ret_from_fork() to C")
  e43b8bb56e ("entry: Inline syscall_exit_to_user_mode()")

7ace1602abf21da5 e43b8bb56e537bfc8d907679309 
---------------- --------------------------- 
         %stddev     %change         %stddev
             \          |                \  
      6.44 ±100%      -6.4        0.00        perf-profile.calltrace.cycles-pp.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe.personality
      6.84 ±100%      -6.8        0.00        perf-profile.children.cycles-pp.syscall_exit_to_user_mode
      5.76 ±100%      -5.8        0.00        perf-profile.self.cycles-pp.syscall_exit_to_user_mode
 3.346e+08            +3.2%  3.455e+08        stress-ng.personality.ops
   5577357            +3.2%    5758223        stress-ng.personality.ops_per_sec
      1278            +1.5%       1297        stress-ng.time.user_time
      1.31            +3.5%       1.36        perf-stat.i.cpi
  1.48e+11            -3.1%  1.435e+11        perf-stat.i.instructions
      0.76            -3.4%       0.74        perf-stat.i.ipc
      1.31            +3.5%       1.36        perf-stat.overall.cpi
      0.76            -3.4%       0.74        perf-stat.overall.ipc
 1.456e+11            -3.1%  1.411e+11        perf-stat.ps.instructions
 8.985e+12            -3.6%  8.659e+12        perf-stat.total.instructions
      0.97 ±111%     -99.9%       0.00 ±141%  perf-sched.sch_delay.avg.ms.irqentry_exit_to_user_mode.asm_exc_page_fault.[unknown]
      1.55 ±  4%    -100.0%       0.00        perf-sched.sch_delay.avg.ms.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe.[unknown]
      1.30 ±103%     -99.9%       0.00 ±141%  perf-sched.sch_delay.max.ms.irqentry_exit_to_user_mode.asm_exc_page_fault.[unknown]
      5.64 ± 25%    -100.0%       0.00        perf-sched.sch_delay.max.ms.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe.[unknown]
      2.42 ± 37%     +49.9%       3.62 ±  4%  perf-sched.sch_delay.max.ms.wait_for_partner.fifo_open.do_dentry_open.vfs_open
    698.88 ± 15%     -40.0%     419.17 ± 20%  perf-sched.wait_and_delay.avg.ms.__cond_resched.smpboot_thread_fn.kthread.ret_from_fork.ret_from_fork_asm
      3.61 ± 12%    -100.0%       0.00        perf-sched.wait_and_delay.avg.ms.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe.[unknown]
      3781 ±  7%    -100.0%       0.00        perf-sched.wait_and_delay.count.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe.[unknown]
    504.70 ± 98%    -100.0%       0.00        perf-sched.wait_and_delay.max.ms.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe.[unknown]
    698.87 ± 15%     -40.0%     419.16 ± 20%  perf-sched.wait_time.avg.ms.__cond_resched.smpboot_thread_fn.kthread.ret_from_fork.ret_from_fork_asm
      2.06 ± 23%    -100.0%       0.00        perf-sched.wait_time.avg.ms.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe.[unknown]
    502.38 ± 98%    -100.0%       0.00        perf-sched.wait_time.max.ms.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe.[unknown]


***************************************************************************************************
lkp-gnr-2ap2: 384 threads 2 sockets Intel(R) Xeon(R) 6972P (Granite Rapids) with 128G memory
=========================================================================================
compiler/cpufreq_governor/kconfig/nr_threads/rootfs/tbox_group/test/testcase/testtime:
  gcc-12/performance/x86_64-rhel-9.4/100%/debian-12-x86_64-20240206.cgz/lkp-gnr-2ap2/context/stress-ng/60s

commit: 
  7ace1602ab ("LoongArch: entry: Migrate ret_from_fork() to C")
  e43b8bb56e ("entry: Inline syscall_exit_to_user_mode()")

7ace1602abf21da5 e43b8bb56e537bfc8d907679309 
---------------- --------------------------- 
         %stddev     %change         %stddev
             \          |                \  
     49055 ±  2%      -4.1%      47032        proc-vmstat.pgreuse
     96.00 ± 19%     -42.5%      55.17 ± 37%  sched_debug.cpu.nr_uninterruptible.max
      4.04 ± 70%      -4.0        0.00        perf-profile.calltrace.cycles-pp.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe.swapcontext
      4.25 ± 70%      -4.2        0.00        perf-profile.children.cycles-pp.syscall_exit_to_user_mode
      3.55 ± 70%      -3.6        0.00        perf-profile.self.cycles-pp.syscall_exit_to_user_mode
 1.047e+08            +2.1%  1.069e+08        stress-ng.context.ops
   1745421            +2.1%    1781401        stress-ng.context.ops_per_sec
   4541476            +2.1%    4636071        stress-ng.context.swapcontext_calls_per_sec
      9344            -1.6%       9196        stress-ng.time.system_time
     13524            +1.1%      13679        stress-ng.time.user_time
      0.06 ± 11%    -100.0%       0.00        perf-sched.sch_delay.avg.ms.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe.[unknown]
     10.75 ± 79%    -100.0%       0.00        perf-sched.sch_delay.max.ms.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe.[unknown]
      0.91 ± 97%    -100.0%       0.00        perf-sched.wait_and_delay.avg.ms.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe.[unknown]
      3274 ±  2%    -100.0%       0.00        perf-sched.wait_and_delay.count.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe.[unknown]
    512.24 ± 95%    -100.0%       0.00        perf-sched.wait_and_delay.max.ms.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe.[unknown]
      0.86 ±104%    -100.0%       0.00        perf-sched.wait_time.avg.ms.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe.[unknown]
    505.95 ± 97%    -100.0%       0.00        perf-sched.wait_time.max.ms.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe.[unknown]
 1.173e+11            -3.6%   1.13e+11        perf-stat.i.branch-instructions
      1.49            +0.1        1.57        perf-stat.i.branch-miss-rate%
 1.727e+09            +2.3%  1.767e+09        perf-stat.i.branch-misses
      2.10            +4.2%       2.19        perf-stat.i.cpi
 6.241e+11            -3.6%  6.019e+11        perf-stat.i.instructions
      0.48            -4.2%       0.46        perf-stat.i.ipc
      1.47            +0.1        1.56        perf-stat.overall.branch-miss-rate%
      2.11            +4.0%       2.19        perf-stat.overall.cpi
      0.47            -3.9%       0.46        perf-stat.overall.ipc
 1.154e+11            -3.6%  1.112e+11        perf-stat.ps.branch-instructions
 1.698e+09            +2.3%  1.738e+09        perf-stat.ps.branch-misses
 6.136e+11            -3.5%  5.919e+11        perf-stat.ps.instructions
 3.804e+13            -4.6%  3.628e+13        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:[~2025-05-16  2:39 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=202505160902.c72854e8-lkp@intel.com \
    --to=oliver.sang@intel.com \
    --cc=alexghiti@rivosinc.com \
    --cc=charlie@rivosinc.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=lkp@intel.com \
    --cc=oe-lkp@lists.linux.dev \
    --cc=tglx@linutronix.de \
    --cc=x86@kernel.org \
    /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.