linux-kernel - [linus:master] [futex] b04b8f3032: perf-bench-futex.ops/s 4.1% regression

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [day] [month] [year] [list]

Message-ID: <202506131635.f3240923-lkp@intel.com>
Date: Fri, 13 Jun 2025 17:19:10 +0800
From: kernel test robot <oliver.sang@...el.com>
To: Sebastian Andrzej Siewior <bigeasy@...utronix.de>
CC: <oe-lkp@...ts.linux.dev>, <lkp@...el.com>, <linux-kernel@...r.kernel.org>,
	Peter Zijlstra <peterz@...radead.org>, <oliver.sang@...el.com>
Subject: [linus:master] [futex]  b04b8f3032:  perf-bench-futex.ops/s 4.1%
 regression


Hello,

we reported
"[tip:locking/futex] [futex]  b04b8f3032:  perf-bench-futex.ops/s 3.9% regression"
in
https://lore.kernel.org/all/202505191048.c1220460-lkp@intel.com/

now the commit is in mainline. we rebuild the kernels and rerun the tests, still
see the similar regression. report again FYI.


kernel test robot noticed a 4.1% regression of perf-bench-futex.ops/s on:


commit: b04b8f3032aae6121303bfa324c768faba032242 ("futex: Introduce futex_q_lockptr_lock()")
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master

[still regression on linus/master      f09079bd04a924c72d555cd97942d5f8d7eca98c]
[still regression on linux-next/master b27cc623e01be9de1580eaa913508b237a7a9673]

testcase: perf-bench-futex
config: x86_64-rhel-9.4
compiler: gcc-12
test machine: 192 threads 2 sockets Intel(R) Xeon(R) 6740E  CPU @ 2.4GHz (Sierra Forest) with 256G memory
parameters:

	runtime: 300s
	nr_task: 100%
	test: hash
	shared: shared
	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@...el.com>
| Closes: https://lore.kernel.org/oe-lkp/202506131635.f3240923-lkp@intel.com


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


The kernel config and materials to reproduce are available at:
https://download.01.org/0day-ci/archive/20250613/202506131635.f3240923-lkp@intel.com

=========================================================================================
compiler/cpufreq_governor/kconfig/nr_task/rootfs/runtime/shared/tbox_group/test/testcase:
  gcc-12/performance/x86_64-rhel-9.4/100%/debian-12-x86_64-20240206.cgz/300s/shared/lkp-srf-2sp2/hash/perf-bench-futex

commit: 
  fe00e88d21 ("futex: Decrease the waiter count before the unlock operation")
  b04b8f3032 ("futex: Introduce futex_q_lockptr_lock()")

fe00e88d217a7bf7 b04b8f3032aae6121303bfa324c 
---------------- --------------------------- 
         %stddev     %change         %stddev
             \          |                \  
   2390532            -4.1%    2293415        perf-bench-futex.ops/s
      6999            -4.7%       6669        perf-bench-futex.time.user_time
     18413 ±  4%     -19.3%      14855 ±  2%  perf-c2c.DRAM.remote
    102464 ±  3%     -15.0%      87089 ±  4%  perf-c2c.HITM.local
     18212 ±  4%     -19.4%      14683 ±  2%  perf-c2c.HITM.remote
    120676 ±  3%     -15.7%     101772 ±  4%  perf-c2c.HITM.total
      0.39            +1.9%       0.39        perf-stat.i.MPKI
  9.27e+10            -4.0%  8.894e+10        perf-stat.i.branch-instructions
 2.175e+08            -4.1%  2.086e+08        perf-stat.i.cache-misses
  5.28e+08            -4.2%   5.06e+08        perf-stat.i.cache-references
      1.09            +6.3%       1.16        perf-stat.i.cpi
      2810            +4.3%       2931        perf-stat.i.cycles-between-cache-misses
 5.633e+11            -5.9%  5.299e+11        perf-stat.i.instructions
      0.92            -5.9%       0.87        perf-stat.i.ipc
      0.39            +1.9%       0.39        perf-stat.overall.MPKI
      1.09            +6.3%       1.15        perf-stat.overall.cpi
      2811            +4.3%       2932        perf-stat.overall.cycles-between-cache-misses
      0.92            -5.9%       0.87        perf-stat.overall.ipc
 9.239e+10            -4.0%  8.865e+10        perf-stat.ps.branch-instructions
 2.168e+08            -4.1%  2.079e+08        perf-stat.ps.cache-misses
 5.263e+08            -4.2%  5.043e+08        perf-stat.ps.cache-references
 5.614e+11            -5.9%  5.282e+11        perf-stat.ps.instructions
  1.69e+14            -6.0%  1.589e+14        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