* [linus:master] [xfs] c91d38b57f: stress-ng.chmod.ops_per_sec 3807.5% improvement
@ 2025-11-07 7:21 kernel test robot
0 siblings, 0 replies; only message in thread
From: kernel test robot @ 2025-11-07 7:21 UTC (permalink / raw)
To: Dave Chinner
Cc: oe-lkp, lkp, linux-kernel, Carlos Maiolino, Jan Kara,
Christoph Hellwig, linux-xfs, oliver.sang
Hello,
kernel test robot noticed a 3807.5% improvement of stress-ng.chmod.ops_per_sec on:
commit: c91d38b57f2c4784d885c874b2a1234a01361afd ("xfs: rework datasync tracking and execution")
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: 64 threads 2 sockets Intel(R) Xeon(R) Gold 6346 CPU @ 3.10GHz (Ice Lake) with 256G memory
parameters:
nr_threads: 100%
disk: 1HDD
testtime: 60s
fs: xfs
test: chmod
cpufreq_governor: performance
Details are as below:
-------------------------------------------------------------------------------------------------->
The kernel config and materials to reproduce are available at:
https://download.01.org/0day-ci/archive/20251107/202511071447.b5bd3ddf-lkp@intel.com
=========================================================================================
compiler/cpufreq_governor/disk/fs/kconfig/nr_threads/rootfs/tbox_group/test/testcase/testtime:
gcc-14/performance/1HDD/xfs/x86_64-rhel-9.4/100%/debian-13-x86_64-20250902.cgz/lkp-icl-2sp8/chmod/stress-ng/60s
commit:
bc7d684fea ("xfs: rearrange code in xfs_inode_item_precommit")
c91d38b57f ("xfs: rework datasync tracking and execution")
bc7d684fea18cc48 c91d38b57f2c4784d885c874b2a
---------------- ---------------------------
%stddev %change %stddev
\ | \
6943 ± 2% -31.3% 4767 uptime.idle
3.968e+09 -19.6% 3.191e+09 cpuidle..time
267038 ± 6% +228.0% 875840 ± 6% cpuidle..usage
202471 ± 87% +110.4% 425943 ± 53% numa-meminfo.node1.Active
202464 ± 87% +110.4% 425935 ± 53% numa-meminfo.node1.Active(anon)
11019 ± 11% +614.5% 78736 ± 19% numa-meminfo.node1.Shmem
50689 ± 87% +110.2% 106536 ± 53% numa-vmstat.node1.nr_active_anon
2736 ± 11% +620.7% 19718 ± 19% numa-vmstat.node1.nr_shmem
50689 ± 87% +110.2% 106536 ± 53% numa-vmstat.node1.nr_zone_active_anon
98.29 -47.8% 51.30 ± 2% iostat.cpu.idle
0.01 ± 34% +4.4e+05% 29.80 iostat.cpu.iowait
0.86 ± 8% +1983.4% 17.82 ± 6% iostat.cpu.system
0.84 ± 13% +27.9% 1.08 ± 2% iostat.cpu.user
678151 +10.8% 751607 meminfo.Active
678134 +10.8% 751591 meminfo.Active(anon)
74229 ± 4% -10.1% 66725 ± 3% meminfo.AnonHugePages
16590 ± 5% +432.3% 88307 ± 14% meminfo.Shmem
15.83 ± 23% +1103.2% 190.50 ± 8% perf-c2c.DRAM.local
358.00 ± 30% +3312.1% 12215 ± 7% perf-c2c.DRAM.remote
321.17 ± 32% +2900.4% 9636 ± 7% perf-c2c.HITM.local
241.00 ± 30% +3726.9% 9222 ± 7% perf-c2c.HITM.remote
562.17 ± 31% +3254.7% 18859 ± 7% perf-c2c.HITM.total
1225 ± 14% +3762.4% 47314 ± 4% stress-ng.chmod.ops
20.17 ± 15% +3807.5% 788.15 ± 4% stress-ng.chmod.ops_per_sec
146.50 ± 11% +752.3% 1248 ± 4% stress-ng.time.involuntary_context_switches
11984 +20.8% 14481 stress-ng.time.minor_page_faults
34.17 ± 16% +3293.2% 1159 ± 6% stress-ng.time.percent_of_cpu_this_job_got
21.27 ± 15% +3173.1% 696.14 ± 6% stress-ng.time.system_time
62731 ± 16% +1072.8% 735741 ± 9% stress-ng.time.voluntary_context_switches
98.30 -48.5 49.84 ± 2% mpstat.cpu.all.idle%
0.00 ±105% +30.7 30.72 mpstat.cpu.all.iowait%
0.02 ± 8% +0.1 0.09 ± 6% mpstat.cpu.all.irq%
0.01 ± 18% +0.0 0.02 ± 5% mpstat.cpu.all.soft%
0.83 ± 9% +17.4 18.24 ± 6% mpstat.cpu.all.sys%
0.85 ± 14% +0.2 1.09 ± 2% mpstat.cpu.all.usr%
1.00 +4033.3% 41.33 ± 33% mpstat.max_utilization.seconds
6.13 ± 6% +333.0% 26.54 ± 4% mpstat.max_utilization_pct
65.17 ± 11% +969.1% 696.67 ± 6% turbostat.Avg_MHz
1.81 ± 11% +17.6 19.39 ± 6% turbostat.Busy%
98.22 -17.5 80.69 turbostat.C1%
96.82 -25.6% 72.07 turbostat.CPU%c1
0.72 ± 4% -55.4% 0.32 ± 2% turbostat.IPC
372011 ± 8% +706.5% 3000169 ± 5% turbostat.IRQ
23024 ± 30% +2629.0% 628341 ± 6% turbostat.NMI
243.27 +9.4% 266.04 turbostat.PkgWatt
15.14 +2.9% 15.57 turbostat.RAMWatt
225.26 ± 17% -92.7% 16.48 ± 9% perf-sched.total_wait_and_delay.average.ms
4552 ± 19% +1482.3% 72026 ± 10% perf-sched.total_wait_and_delay.count.ms
4996 -22.0% 3898 ± 10% perf-sched.total_wait_and_delay.max.ms
225.25 ± 17% -92.7% 16.46 ± 9% perf-sched.total_wait_time.average.ms
4996 -22.0% 3898 ± 10% perf-sched.total_wait_time.max.ms
225.26 ± 17% -92.7% 16.48 ± 9% perf-sched.wait_and_delay.avg.ms.[unknown].[unknown].[unknown].[unknown].[unknown]
4552 ± 19% +1482.3% 72026 ± 10% perf-sched.wait_and_delay.count.[unknown].[unknown].[unknown].[unknown].[unknown]
4996 -22.0% 3898 ± 10% perf-sched.wait_and_delay.max.ms.[unknown].[unknown].[unknown].[unknown].[unknown]
225.25 ± 17% -92.7% 16.46 ± 9% perf-sched.wait_time.avg.ms.[unknown].[unknown].[unknown].[unknown].[unknown]
4996 -22.0% 3898 ± 10% perf-sched.wait_time.max.ms.[unknown].[unknown].[unknown].[unknown].[unknown]
169578 +10.8% 187927 proc-vmstat.nr_active_anon
935816 +1.9% 953408 proc-vmstat.nr_file_pages
20878 +4.0% 21711 proc-vmstat.nr_mapped
4142 ± 5% +433.8% 22109 ± 14% proc-vmstat.nr_shmem
41118 +1.0% 41531 proc-vmstat.nr_slab_unreclaimable
169578 +10.8% 187927 proc-vmstat.nr_zone_active_anon
615.17 ±218% +464.4% 3472 ± 14% proc-vmstat.numa_hint_faults_local
345828 ± 4% +10.1% 380782 ± 2% proc-vmstat.numa_hit
279652 ± 5% +12.5% 314600 ± 3% proc-vmstat.numa_local
60782 ± 60% +78.2% 108287 ± 14% proc-vmstat.numa_pte_updates
382421 ± 4% +9.6% 419096 ± 2% proc-vmstat.pgalloc_normal
5338 ± 20% +193.0% 15639 ± 6% proc-vmstat.pgpgout
7731 ± 7% +1078.0% 91080 ± 10% sched_debug.cfs_rq:/.avg_vruntime.avg
61046 ± 13% +183.6% 173149 ± 14% sched_debug.cfs_rq:/.avg_vruntime.max
747.60 ± 23% +3826.2% 29352 ± 13% sched_debug.cfs_rq:/.avg_vruntime.min
10401 ± 7% +72.6% 17954 ± 9% sched_debug.cfs_rq:/.avg_vruntime.stddev
564880 ± 44% -71.2% 162663 ±113% sched_debug.cfs_rq:/.load.max
79058 ± 37% -64.4% 28177 ± 90% sched_debug.cfs_rq:/.load.stddev
7731 ± 7% +1078.0% 91080 ± 10% sched_debug.cfs_rq:/.min_vruntime.avg
61046 ± 13% +183.6% 173149 ± 14% sched_debug.cfs_rq:/.min_vruntime.max
747.60 ± 23% +3826.2% 29352 ± 13% sched_debug.cfs_rq:/.min_vruntime.min
10401 ± 7% +72.6% 17954 ± 9% sched_debug.cfs_rq:/.min_vruntime.stddev
254.61 ± 6% +23.3% 313.88 ± 6% sched_debug.cfs_rq:/.runnable_avg.avg
254.13 ± 6% +23.4% 313.61 ± 6% sched_debug.cfs_rq:/.util_avg.avg
1388025 ± 3% -8.9% 1264183 ± 4% sched_debug.cpu.avg_idle.avg
4207 ± 7% +257.6% 15044 ± 6% sched_debug.cpu.nr_switches.avg
13511 ± 20% +276.6% 50888 ± 62% sched_debug.cpu.nr_switches.max
1440 ± 9% +592.5% 9975 ± 4% sched_debug.cpu.nr_switches.min
2591 ± 14% +129.9% 5958 ± 61% sched_debug.cpu.nr_switches.stddev
19.42 ± 31% +137.8% 46.17 ± 17% sched_debug.cpu.nr_uninterruptible.max
-14.67 +891.5% -145.42 sched_debug.cpu.nr_uninterruptible.min
4.79 ± 10% +353.6% 21.71 ± 15% sched_debug.cpu.nr_uninterruptible.stddev
0.94 ± 5% -30.7% 0.65 ± 2% perf-stat.i.MPKI
6.179e+08 ± 12% +373.6% 2.927e+09 ± 4% perf-stat.i.branch-instructions
3.29 ± 11% -1.8 1.47 ± 8% perf-stat.i.branch-miss-rate%
43227294 ± 15% +20.8% 52203685 ± 3% perf-stat.i.branch-misses
17.76 ± 6% +19.1 36.85 perf-stat.i.cache-miss-rate%
1306832 ± 11% +601.7% 9169533 ± 4% perf-stat.i.cache-misses
8673239 ± 10% +206.9% 26621069 ± 3% perf-stat.i.cache-references
3772 ± 10% +618.3% 27098 ± 8% perf-stat.i.context-switches
2.43 ± 4% +32.5% 3.23 perf-stat.i.cpi
4.243e+09 ± 11% +964.2% 4.515e+10 ± 6% perf-stat.i.cpu-cycles
116.32 ± 2% +203.4% 352.94 ± 3% perf-stat.i.cpu-migrations
3133 ± 3% +59.5% 4996 ± 2% perf-stat.i.cycles-between-cache-misses
3.037e+09 ± 12% +376.7% 1.448e+10 ± 4% perf-stat.i.instructions
0.55 ± 4% -40.3% 0.33 ± 2% perf-stat.i.ipc
0.43 ± 4% +46.9% 0.63 ± 2% perf-stat.overall.MPKI
6.97 ± 3% -5.2 1.79 ± 7% perf-stat.overall.branch-miss-rate%
15.06 ± 4% +19.3 34.39 perf-stat.overall.cache-miss-rate%
1.40 ± 3% +122.3% 3.12 perf-stat.overall.cpi
3252 ± 2% +51.3% 4921 ± 3% perf-stat.overall.cycles-between-cache-misses
0.71 ± 3% -55.1% 0.32 perf-stat.overall.ipc
6.093e+08 ± 12% +372.7% 2.88e+09 ± 4% perf-stat.ps.branch-instructions
42634564 ± 15% +20.7% 51439404 ± 3% perf-stat.ps.branch-misses
1287683 ± 11% +600.7% 9022871 ± 4% perf-stat.ps.cache-misses
8556164 ± 10% +206.5% 26227629 ± 3% perf-stat.ps.cache-references
3710 ± 10% +618.7% 26664 ± 8% perf-stat.ps.context-switches
4.185e+09 ± 11% +961.5% 4.442e+10 ± 6% perf-stat.ps.cpu-cycles
114.49 ± 2% +203.8% 347.78 ± 3% perf-stat.ps.cpu-migrations
2.995e+09 ± 12% +375.7% 1.425e+10 ± 4% perf-stat.ps.instructions
1.865e+11 ± 12% +369.7% 8.759e+11 ± 4% 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
^ permalink raw reply [flat|nested] only message in thread
only message in thread, other threads:[~2025-11-07 7:22 UTC | newest]
Thread overview: (only message) (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-11-07 7:21 [linus:master] [xfs] c91d38b57f: stress-ng.chmod.ops_per_sec 3807.5% improvement kernel test robot
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.