Hello,
kernel test robot noticed a -1.7% regression of sysbench-fileio.write_operations/s on:
commit: 060406c61c7cb4bbd82a02d179decca9c9bb3443 ("block: add plug while submitting IO")
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master
testcase: sysbench-fileio
test machine: 128 threads 2 sockets Intel(R) Xeon(R) Platinum 8358 CPU @ 2.60GHz (Ice Lake) with 128G memory
parameters:
period: 600s
nr_threads: 100%
disk: 1HDD
fs: xfs
size: 64G
filenum: 1024f
rwmode: rndwr
iomode: sync
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@xxxxxxxxx>
| Closes: https://lore.kernel.org/oe-lkp/202406252157.348c160f-oliver.sang@xxxxxxxxx
Details are as below:
-------------------------------------------------------------------------------------------------->
The kernel config and materials to reproduce are available at:
https://download.01.org/0day-ci/archive/20240625/202406252157.348c160f-oliver.sang@xxxxxxxxx
=========================================================================================
compiler/cpufreq_governor/disk/filenum/fs/iomode/kconfig/nr_threads/period/rootfs/rwmode/size/tbox_group/testcase:
gcc-13/performance/1HDD/1024f/xfs/sync/x86_64-rhel-8.3/100%/600s/debian-12-x86_64-20240206.cgz/rndwr/64G/lkp-icl-2sp6/sysbench-fileio
commit:
3a861560cc ("bcache: fix variable length array abuse in btree_iter")
060406c61c ("block: add plug while submitting IO")
3a861560ccb35f2a 060406c61c7cb4bbd82a02d179d
---------------- ---------------------------
%stddev %change %stddev
\ | \
24.53 ± 5% -3.0 21.49 ± 3% mpstat.cpu.all.idle%
52604 ± 70% +99.5% 104921 ± 10% numa-numastat.node1.other_node
24267 ± 4% -10.3% 21772 ± 3% uptime.idle
8218 -1.6% 8084 vmstat.system.cs
24.75 ± 5% -12.3% 21.72 ± 3% iostat.cpu.idle
75.08 +4.0% 78.12 iostat.cpu.iowait
132522 ± 11% +23.4% 163559 ± 6% numa-meminfo.node0.Inactive(file)
237987 ± 6% -15.5% 201059 ± 5% numa-meminfo.node1.Inactive(file)
22.53 ± 11% -6.3 16.22 ± 25% perf-profile.children.cycles-pp.native_flush_tlb_one_user
0.58 ± 29% +0.5 1.05 ± 28% perf-profile.children.cycles-pp.set_pte_vaddr_p4d
22.53 ± 11% -6.3 16.22 ± 25% perf-profile.self.cycles-pp.native_flush_tlb_one_user
0.58 ± 29% +0.5 1.05 ± 28% perf-profile.self.cycles-pp.set_pte_vaddr_p4d
0.03 ± 6% +13.9% 0.03 ± 4% sched_debug.cfs_rq:/.h_nr_running.avg
0.03 ± 6% +13.9% 0.03 ± 4% sched_debug.cfs_rq:/.nr_running.avg
-226.99 +41.3% -320.83 sched_debug.cpu.nr_uninterruptible.min
25.90 ± 20% +40.7% 36.44 ± 10% sched_debug.cpu.nr_uninterruptible.stddev
194997 -1.7% 191663 proc-vmstat.nr_dirtied
92615 -1.6% 91119 proc-vmstat.nr_inactive_file
194997 -1.7% 191663 proc-vmstat.nr_written
92615 -1.6% 91119 proc-vmstat.nr_zone_inactive_file
952833 -1.7% 936954 proc-vmstat.pgpgout
1035 -1.4% 1020 sysbench-fileio.fsync_operations/s
140.12 +1.7% 142.56 sysbench-fileio.latency_avg_ms
1559978 -1.7% 1533306 sysbench-fileio.time.file_system_outputs
1497 ± 3% -51.2% 730.50 ± 10% sysbench-fileio.time.involuntary_context_switches
1871907 -1.5% 1843314 sysbench-fileio.time.voluntary_context_switches
1.31 -1.7% 1.29 sysbench-fileio.write_bytes_MB/s
1.25 -1.7% 1.23 sysbench-fileio.write_bytes_MiB/s
80.13 -1.7% 78.74 sysbench-fileio.write_operations/s
71858 ± 7% +18.7% 85272 ± 3% numa-vmstat.node0.nr_dirtied
33121 ± 11% +23.4% 40882 ± 6% numa-vmstat.node0.nr_inactive_file
71858 ± 7% +18.7% 85272 ± 3% numa-vmstat.node0.nr_written
33121 ± 11% +23.4% 40882 ± 6% numa-vmstat.node0.nr_zone_inactive_file
123139 ± 4% -13.6% 106391 ± 2% numa-vmstat.node1.nr_dirtied
59483 ± 6% -15.5% 50254 ± 5% numa-vmstat.node1.nr_inactive_file
123139 ± 4% -13.6% 106391 ± 2% numa-vmstat.node1.nr_written
59483 ± 6% -15.5% 50254 ± 5% numa-vmstat.node1.nr_zone_inactive_file
52604 ± 70% +99.5% 104921 ± 10% numa-vmstat.node1.numa_other
2.55 +4.8% 2.68 perf-stat.i.MPKI
92991408 ± 2% -3.7% 89545806 ± 2% perf-stat.i.branch-instructions
6.14 +0.3 6.41 perf-stat.i.branch-miss-rate%
18893060 -3.6% 18215592 perf-stat.i.cache-references
1.83 +4.9% 1.92 perf-stat.i.cpi
6.848e+08 -2.5% 6.675e+08 ± 2% perf-stat.i.cpu-cycles
214.11 ± 2% -7.2% 198.60 perf-stat.i.cpu-migrations
4.678e+08 ± 2% -3.9% 4.495e+08 ± 2% perf-stat.i.instructions
0.58 -4.3% 0.56 perf-stat.i.ipc
92993821 ± 2% -3.7% 89564915 ± 2% perf-stat.ps.branch-instructions
18876617 -3.6% 18201664 perf-stat.ps.cache-references
6.863e+08 -2.5% 6.692e+08 ± 2% perf-stat.ps.cpu-cycles
213.74 ± 2% -7.2% 198.25 perf-stat.ps.cpu-migrations
4.678e+08 ± 2% -3.9% 4.497e+08 ± 2% perf-stat.ps.instructions
3.003e+11 ± 2% -3.9% 2.887e+11 ± 2% 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
