From: kernel test robot <oliver.sang@intel.com>
To: "Liam R. Howlett" <Liam.Howlett@oracle.com>
Cc: <oe-lkp@lists.linux.dev>, <lkp@intel.com>,
<linux-kernel@vger.kernel.org>, Vlastimil Babka <vbabka@suse.cz>,
Suren Baghdasaryan <surenb@google.com>,
<maple-tree@lists.infradead.org>, <linux-mm@kvack.org>,
<oliver.sang@intel.com>
Subject: [linus:master] [maple_tree] 9b05890a25: stress-ng.mmap.ops_per_sec 3.0% improvement
Date: Thu, 6 Nov 2025 15:26:17 +0800 [thread overview]
Message-ID: <202511061420.a984bb7e-lkp@intel.com> (raw)
Hello,
kernel test robot noticed a 3.0% improvement of stress-ng.mmap.ops_per_sec on:
commit: 9b05890a25d9197e39fcf5b2298f0b911c323306 ("maple_tree: Prefilled sheaf conversion and testing")
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%
testtime: 60s
test: mmap
cpufreq_governor: performance
In addition to that, the commit also has significant impact on the following tests:
+------------------+-------------------------------------------------------------------------------------------+
| testcase: change | stress-ng: stress-ng.tmpfs.ops_per_sec 3.2% improvement |
| test machine | 64 threads 2 sockets Intel(R) Xeon(R) Gold 6346 CPU @ 3.10GHz (Ice Lake) with 256G memory |
| test parameters | cpufreq_governor=performance |
| | nr_threads=100% |
| | test=tmpfs |
| | testtime=60s |
+------------------+-------------------------------------------------------------------------------------------+
Details are as below:
-------------------------------------------------------------------------------------------------->
The kernel config and materials to reproduce are available at:
https://download.01.org/0day-ci/archive/20251106/202511061420.a984bb7e-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-icl-2sp7/mmap/stress-ng/60s
commit:
fdbebab19f ("tools/testing: Add support for prefilled slab sheafs")
9b05890a25 ("maple_tree: Prefilled sheaf conversion and testing")
fdbebab19f147af6 9b05890a25d9197e39fcf5b2298
---------------- ---------------------------
%stddev %change %stddev
\ | \
14.17 -1.3 12.88 mpstat.cpu.all.sys%
38.05 ± 2% -22.8% 29.38 ± 2% mpstat.max_utilization_pct
13835 ± 11% -29.9% 9698 ± 19% perf-c2c.DRAM.local
4110 ± 4% -17.7% 3381 ± 7% perf-c2c.HITM.local
3.03 ±100% -3.0 0.00 perf-profile.calltrace.cycles-pp.mas_alloc_nodes.mas_preallocate.__mmap_new_vma.__mmap_region.do_mmap
2.72 ±100% -2.7 0.00 perf-profile.calltrace.cycles-pp.mas_alloc_nodes.mas_preallocate.commit_merge.vma_expand.vma_merge_new_range
5.78 ±100% -5.8 0.00 perf-profile.children.cycles-pp.mas_alloc_nodes
7045634 -10.8% 6283116 meminfo.Memused
807198 ± 11% -67.1% 265386 meminfo.SUnreclaim
429663 ± 2% -9.9% 387130 ± 2% meminfo.Shmem
899957 ± 10% -60.2% 358148 meminfo.Slab
417173 ± 41% -67.2% 136751 ± 4% numa-meminfo.node0.SUnreclaim
465472 ± 37% -60.1% 185577 ± 14% numa-meminfo.node0.Slab
390418 ± 30% -66.8% 129555 ± 5% numa-meminfo.node1.SUnreclaim
434877 ± 28% -60.1% 173494 ± 15% numa-meminfo.node1.Slab
5573487 ± 5% +52.3% 8486524 ± 2% numa-numastat.node0.local_node
5603800 ± 4% +51.8% 8506515 ± 2% numa-numastat.node0.numa_hit
5715530 ± 4% +55.4% 8881057 ± 2% numa-numastat.node1.local_node
5752421 ± 4% +55.2% 8927246 ± 2% numa-numastat.node1.numa_hit
0.98 ± 4% -0.1 0.86 turbostat.C1E%
802545 -10.7% 716562 turbostat.NMI
179.86 -1.0% 178.07 turbostat.PkgWatt
16.46 -1.0% 16.29 turbostat.RAMWatt
104549 ± 41% -67.0% 34522 ± 4% numa-vmstat.node0.nr_slab_unreclaimable
5603946 ± 4% +51.8% 8506697 ± 2% numa-vmstat.node0.numa_hit
5573633 ± 5% +52.3% 8486707 ± 2% numa-vmstat.node0.numa_local
97906 ± 29% -66.4% 32924 ± 5% numa-vmstat.node1.nr_slab_unreclaimable
5751995 ± 4% +55.2% 8926887 ± 2% numa-vmstat.node1.numa_hit
5715104 ± 4% +55.4% 8880697 ± 2% numa-vmstat.node1.numa_local
273239 +3.0% 281447 stress-ng.mmap.ops
4554 +3.0% 4690 stress-ng.mmap.ops_per_sec
9280 ± 3% -15.9% 7808 ± 3% stress-ng.time.involuntary_context_switches
7742463 +2.0% 7893647 stress-ng.time.minor_page_faults
933.17 -8.8% 850.83 stress-ng.time.percent_of_cpu_this_job_got
534.72 -9.3% 484.88 stress-ng.time.system_time
337713 -3.1% 327210 proc-vmstat.nr_active_anon
1026876 -1.0% 1016123 proc-vmstat.nr_file_pages
107891 ± 2% -10.0% 97114 ± 2% proc-vmstat.nr_shmem
201980 ± 11% -67.2% 66273 ± 2% proc-vmstat.nr_slab_unreclaimable
337713 -3.1% 327209 proc-vmstat.nr_zone_active_anon
11358395 +53.5% 17435898 proc-vmstat.numa_hit
11291191 +53.8% 17369717 proc-vmstat.numa_local
35250051 +35.6% 47789964 proc-vmstat.pgalloc_normal
8103365 +1.9% 8259384 proc-vmstat.pgfault
34497114 +37.2% 47343827 proc-vmstat.pgfree
13461 +2.4% 13780 proc-vmstat.thp_deferred_split_page
41345 +2.2% 42260 proc-vmstat.thp_fault_alloc
13525 +2.4% 13844 proc-vmstat.thp_split_pmd
28491365 +2.1% 29102247 proc-vmstat.unevictable_pgs_culled
28491780 +2.1% 29102937 proc-vmstat.unevictable_pgs_mlocked
28491774 +2.1% 29102933 proc-vmstat.unevictable_pgs_munlocked
28490836 +2.1% 29101721 proc-vmstat.unevictable_pgs_rescued
1.87 -6.3% 1.75 perf-stat.i.MPKI
7.634e+09 -1.2% 7.543e+09 perf-stat.i.branch-instructions
63.21 -1.7 61.53 perf-stat.i.cache-miss-rate%
67800530 -6.9% 63093634 perf-stat.i.cache-misses
1.067e+08 -4.6% 1.018e+08 perf-stat.i.cache-references
1.08 -6.0% 1.02 perf-stat.i.cpi
3.949e+10 -7.1% 3.668e+10 perf-stat.i.cpu-cycles
229.21 -4.5% 218.93 perf-stat.i.cpu-migrations
0.93 +6.5% 0.99 perf-stat.i.ipc
1.86 -6.3% 1.75 perf-stat.overall.MPKI
63.49 -1.5 61.97 perf-stat.overall.cache-miss-rate%
1.09 -6.5% 1.02 perf-stat.overall.cpi
0.92 +7.0% 0.99 perf-stat.overall.ipc
7.516e+09 -1.3% 7.421e+09 perf-stat.ps.branch-instructions
66736273 -7.0% 62060020 perf-stat.ps.cache-misses
1.051e+08 -4.7% 1.001e+08 perf-stat.ps.cache-references
3.89e+10 -7.2% 3.61e+10 perf-stat.ps.cpu-cycles
226.17 -4.7% 215.61 perf-stat.ps.cpu-migrations
***************************************************************************************************
lkp-icl-2sp8: 64 threads 2 sockets Intel(R) Xeon(R) Gold 6346 CPU @ 3.10GHz (Ice Lake) with 256G memory
=========================================================================================
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-icl-2sp8/tmpfs/stress-ng/60s
commit:
fdbebab19f ("tools/testing: Add support for prefilled slab sheafs")
9b05890a25 ("maple_tree: Prefilled sheaf conversion and testing")
fdbebab19f147af6 9b05890a25d9197e39fcf5b2298
---------------- ---------------------------
%stddev %change %stddev
\ | \
1.47 ± 9% +0.4 1.85 ± 9% mpstat.cpu.all.soft%
598993 ± 14% -36.3% 381625 ± 12% meminfo.SUnreclaim
689901 ± 12% -31.4% 473344 ± 10% meminfo.Slab
403.27 -5.2% 382.31 stress-ng.time.system_time
1800 +3.2% 1859 stress-ng.tmpfs.ops_per_sec
315677 ± 33% -37.9% 196080 ± 12% numa-meminfo.node0.SUnreclaim
1446782 ±106% +99.7% 2889569 ± 35% numa-meminfo.node0.Unevictable
283500 ± 25% -34.4% 186029 ± 13% numa-meminfo.node1.SUnreclaim
334463 ± 25% -34.0% 220868 ± 15% numa-meminfo.node1.Slab
701526 ± 8% +216.2% 2218043 ± 3% numa-numastat.node0.local_node
731546 ± 7% +207.8% 2251790 ± 2% numa-numastat.node0.numa_hit
686642 ± 5% +212.7% 2147291 numa-numastat.node1.local_node
723003 ± 5% +201.5% 2179707 numa-numastat.node1.numa_hit
79111 ± 33% -37.6% 49392 ± 12% numa-vmstat.node0.nr_slab_unreclaimable
731773 ± 7% +207.7% 2251794 ± 2% numa-vmstat.node0.numa_hit
701753 ± 8% +216.1% 2218046 ± 3% numa-vmstat.node0.numa_local
71014 ± 25% -33.7% 47056 ± 13% numa-vmstat.node1.nr_slab_unreclaimable
722610 ± 5% +201.4% 2178261 numa-vmstat.node1.numa_hit
686249 ± 5% +212.7% 2145846 numa-vmstat.node1.numa_local
149703 ± 14% -36.3% 95370 ± 12% proc-vmstat.nr_slab_unreclaimable
1457881 ± 6% +204.1% 4433426 proc-vmstat.numa_hit
1391484 ± 6% +213.9% 4367252 proc-vmstat.numa_local
2289318 ± 8% +260.5% 8252931 proc-vmstat.pgalloc_normal
2045446 ± 8% +297.0% 8120146 proc-vmstat.pgfree
6961 ± 2% -5.1% 6607 ± 2% proc-vmstat.pgreuse
9.26 ± 71% -9.3 0.00 perf-profile.calltrace.cycles-pp.__x64_sys_exit_group.x64_sys_call.do_syscall_64.entry_SYSCALL_64_after_hwframe
9.26 ± 71% -9.3 0.00 perf-profile.calltrace.cycles-pp.do_exit.do_group_exit.__x64_sys_exit_group.x64_sys_call.do_syscall_64
9.26 ± 71% -9.3 0.00 perf-profile.calltrace.cycles-pp.do_group_exit.__x64_sys_exit_group.x64_sys_call.do_syscall_64.entry_SYSCALL_64_after_hwframe
9.26 ± 71% -9.3 0.00 perf-profile.calltrace.cycles-pp.x64_sys_call.do_syscall_64.entry_SYSCALL_64_after_hwframe
9.26 ± 71% -9.3 0.00 perf-profile.children.cycles-pp.__x64_sys_exit_group
9.26 ± 71% -9.3 0.00 perf-profile.children.cycles-pp.x64_sys_call
8.13 ± 92% -4.1 4.05 ±144% perf-profile.children.cycles-pp.free_pcppages_bulk
6.31 ± 63% -0.9 5.42 ±146% perf-profile.children.cycles-pp.__schedule
1.99 -5.1% 1.89 perf-stat.i.MPKI
2.678e+10 +2.2% 2.736e+10 perf-stat.i.branch-instructions
38.13 -1.1 36.98 perf-stat.i.cache-miss-rate%
2.698e+08 -2.8% 2.624e+08 perf-stat.i.cache-misses
1.20 -3.0% 1.16 perf-stat.i.cpi
730.94 ± 7% +15.1% 841.39 ± 3% perf-stat.i.cpu-migrations
606.32 +2.2% 619.60 perf-stat.i.cycles-between-cache-misses
1.319e+11 +2.5% 1.352e+11 perf-stat.i.instructions
0.84 +3.0% 0.87 perf-stat.i.ipc
149.32 +3.6% 154.64 perf-stat.i.metric.K/sec
4777975 +3.7% 4952420 perf-stat.i.minor-faults
4777975 +3.7% 4952420 perf-stat.i.page-faults
2.05 -5.2% 1.94 perf-stat.overall.MPKI
1.83 -0.0 1.81 perf-stat.overall.branch-miss-rate%
38.31 -1.2 37.08 perf-stat.overall.cache-miss-rate%
1.22 -3.1% 1.19 perf-stat.overall.cpi
598.57 +2.2% 611.83 perf-stat.overall.cycles-between-cache-misses
0.82 +3.2% 0.84 perf-stat.overall.ipc
2.451e+10 +1.6% 2.489e+10 perf-stat.ps.branch-instructions
2.47e+08 -3.3% 2.388e+08 perf-stat.ps.cache-misses
671.65 ± 8% +14.3% 767.68 ± 3% perf-stat.ps.cpu-migrations
1.207e+11 +1.9% 1.23e+11 perf-stat.ps.instructions
4374027 +3.1% 4507555 perf-stat.ps.minor-faults
4374027 +3.1% 4507555 perf-stat.ps.page-faults
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:[~2025-11-06 7:26 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=202511061420.a984bb7e-lkp@intel.com \
--to=oliver.sang@intel.com \
--cc=Liam.Howlett@oracle.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=lkp@intel.com \
--cc=maple-tree@lists.infradead.org \
--cc=oe-lkp@lists.linux.dev \
--cc=surenb@google.com \
--cc=vbabka@suse.cz \
/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.