[linus:master] [slab] ed30c4adfc: stress-ng.timerfd.ops_per_sec 13.5% regression

[kernel test robot <oliver.sang@intel.com>]

Hello,

kernel test robot noticed a 13.5% regression of stress-ng.timerfd.ops_per_sec on:


commit: ed30c4adfc2b56909ca43fb5e4750a646928cbf4 ("slab: add optimized sheaf refill from partial list")
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master

[still regression on linus/master      a95f71ad3e2e224277508e006580c333d0a5fe36]
[still regression on linux-next/master d4906ae14a5f136ceb671bb14cedbf13fa560da6]

testcase: stress-ng
config: x86_64-rhel-9.4
compiler: gcc-14
test machine: 224 threads 2 sockets Intel(R) Xeon(R) Platinum 8480CTDX (Sapphire Rapids) with 512G memory
parameters:

	nr_threads: 100%
	testtime: 60s
	test: timerfd
	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/202602241519.7f9aea41-lkp@intel.com


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


The kernel config and materials to reproduce are available at:
https://download.01.org/0day-ci/archive/20260224/202602241519.7f9aea41-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-spr-2sp4/timerfd/stress-ng/60s

commit: 
  913ffd3a1b ("slab: handle kmalloc sheaves bootstrap")
  ed30c4adfc ("slab: add optimized sheaf refill from partial list")

913ffd3a1bf5d154 ed30c4adfc2b56909ca43fb5e47 
---------------- --------------------------- 
         %stddev     %change         %stddev
             \          |                \  
     17379            +2.9%      17891        stress-ng.time.percent_of_cpu_this_job_got
      7100           +28.7%       9137        stress-ng.time.system_time
      3397           -51.2%       1657 ±  3%  stress-ng.time.user_time
 2.162e+10           -13.5%   1.87e+10        stress-ng.timerfd.ops
 3.596e+08           -13.5%  3.111e+08        stress-ng.timerfd.ops_per_sec
     24.13           -50.5%      11.96 ±  2%  vmstat.cpu.us
     20.84            -2.2       18.66        mpstat.cpu.all.irq%
     51.25           +14.5       65.79        mpstat.cpu.all.sys%
     24.90           -12.6       12.34 ±  2%  mpstat.cpu.all.usr%
      0.92           -12.4%       0.81        turbostat.IPC
    114.75           -13.2      101.54        turbostat.PKG_%
     21.44            +5.0%      22.52 ±  3%  turbostat.RAMWatt
      0.02 ± 20%    +187.5%       0.06 ± 35%  perf-stat.i.MPKI
 1.146e+11           -12.7%  1.001e+11        perf-stat.i.branch-instructions
      0.17 ±  7%      +0.0        0.21 ±  5%  perf-stat.i.branch-miss-rate%
 1.397e+08 ±  4%     +16.8%  1.632e+08 ±  2%  perf-stat.i.branch-misses
     12.23 ± 47%     -10.3        1.92 ± 40%  perf-stat.i.cache-miss-rate%
   2332655 ± 45%    +742.7%   19656432 ± 53%  perf-stat.i.cache-misses
  16606819 ±  9%   +8113.0%  1.364e+09 ±  8%  perf-stat.i.cache-references
      1.11           +14.1%       1.27        perf-stat.i.cpi
    730605 ± 33%     -91.1%      65146 ± 48%  perf-stat.i.cycles-between-cache-misses
 5.892e+11           -12.4%  5.162e+11        perf-stat.i.instructions
      0.91           -12.3%       0.80        perf-stat.i.ipc
      0.00 ± 46%    +868.2%       0.04 ± 54%  perf-stat.overall.MPKI
      0.12 ±  4%      +0.0        0.16 ±  3%  perf-stat.overall.branch-miss-rate%
     13.80 ± 38%     -12.3        1.45 ± 53%  perf-stat.overall.cache-miss-rate%
      1.09           +14.0%       1.24        perf-stat.overall.cpi
    310396 ± 24%     -87.4%      39114 ± 34%  perf-stat.overall.cycles-between-cache-misses
      0.92           -12.2%       0.81        perf-stat.overall.ipc
 1.128e+11           -12.7%  9.852e+10        perf-stat.ps.branch-instructions
 1.371e+08 ±  4%     +16.8%  1.602e+08 ±  2%  perf-stat.ps.branch-misses
   2278016 ± 45%    +747.6%   19307920 ± 54%  perf-stat.ps.cache-misses
  16326694 ±  9%   +8118.6%  1.342e+09 ±  8%  perf-stat.ps.cache-references
 5.796e+11           -12.4%  5.078e+11        perf-stat.ps.instructions
 3.566e+13           -12.2%  3.129e+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