Greeting,
FYI, we noticed a -22.7% improvement of phoronix-test-suite.tungsten.Non-Exponential.seconds due to commit:
commit: bdacfaf26da166dd56c62f23f27a4b3e71f2d89e ("perf core: Add a kmem_cache for struct perf_event")
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master
in testcase: phoronix-test-suite
on test machine: 96 threads 2 sockets Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz with 128G memory
with following parameters:
test: tungsten-1.0.0
option_a: Non-Exponential
cpufreq_governor: performance
ucode: 0x5003006
test-description: The Phoronix Test Suite is the most comprehensive testing and benchmarking platform available that provides an extensible framework for which new tests can be easily added.
test-url: http://www.phoronix-test-suite.com/
Details are as below:
-------------------------------------------------------------------------------------------------->
To reproduce:
git clone https://github.com/intel/lkp-tests.git
cd lkp-tests
sudo bin/lkp install job.yaml # job file is attached in this email
bin/lkp split-job --compatible job.yaml # generate the yaml file for lkp run
sudo bin/lkp run generated-yaml-file
# if come across any failure that blocks the test,
# please remove ~/.lkp and /lkp dir to run from a clean state.
=========================================================================================
compiler/cpufreq_governor/kconfig/option_a/rootfs/tbox_group/test/testcase/ucode:
gcc-9/performance/x86_64-rhel-8.3/Non-Exponential/debian-x86_64-phoronix/lkp-csl-2sp8/tungsten-1.0.0/phoronix-test-suite/0x5003006
commit:
9483409ab5 ("perf core: Allocate perf_buffer in the target node memory")
bdacfaf26d ("perf core: Add a kmem_cache for struct perf_event")
9483409ab5067941 bdacfaf26da166dd56c62f23f27
---------------- ---------------------------
%stddev %change %stddev
\ | \
4989 ? 2% -9.5% 4516 phoronix-test-suite.time.percent_of_cpu_this_job_got
7.73 ? 12% -22.7% 5.97 phoronix-test-suite.tungsten.Non-Exponential.seconds
441319 ? 48% -75.1% 109832 ?190% numa-meminfo.node1.Shmem
110322 ? 48% -75.1% 27457 ?190% numa-vmstat.node1.nr_shmem
213.04 -4.4% 203.63 turbostat.PkgWatt
47.09 ? 2% +4.4 51.50 mpstat.cpu.all.idle%
1.12 ? 4% +0.4 1.56 ? 8% mpstat.cpu.all.irq%
40260 ? 55% -76.7% 9375 ? 98% numa-numastat.node0.other_node
27885 ? 80% +110.9% 58815 ? 15% numa-numastat.node1.other_node
47.50 ? 3% +9.1% 51.83 vmstat.cpu.id
50.17 ? 3% -9.3% 45.50 ? 2% vmstat.cpu.us
2417 ? 2% +4.7% 2530 vmstat.system.cs
12.83 ?145% -10.2 2.58 ? 11% perf-profile.calltrace.cycles-pp.cpu_startup_entry.start_secondary.secondary_startup_64_no_verify
12.83 ?145% -10.2 2.58 ? 11% perf-profile.calltrace.cycles-pp.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64_no_verify
12.83 ?145% -10.2 2.58 ? 11% perf-profile.calltrace.cycles-pp.start_secondary.secondary_startup_64_no_verify
12.87 ?145% -10.2 2.63 ? 11% perf-profile.calltrace.cycles-pp.secondary_startup_64_no_verify
10.96 ?141% -8.5 2.44 ? 12% perf-profile.calltrace.cycles-pp.cpuidle_enter.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64_no_verify
10.89 ?142% -8.5 2.40 ? 12% perf-profile.calltrace.cycles-pp.cpuidle_enter_state.cpuidle_enter.do_idle.cpu_startup_entry.start_secondary
12.83 ?145% -10.2 2.58 ? 11% perf-profile.children.cycles-pp.start_secondary
12.87 ?145% -10.2 2.63 ? 11% perf-profile.children.cycles-pp.secondary_startup_64_no_verify
12.87 ?145% -10.2 2.63 ? 11% perf-profile.children.cycles-pp.cpu_startup_entry
12.87 ?145% -10.2 2.63 ? 11% perf-profile.children.cycles-pp.do_idle
11.00 ?140% -8.5 2.48 ? 12% perf-profile.children.cycles-pp.cpuidle_enter
11.00 ?140% -8.5 2.48 ? 12% perf-profile.children.cycles-pp.cpuidle_enter_state
4.5e+09 ? 10% +17.8% 5.3e+09 perf-stat.i.branch-instructions
67695126 ? 7% +16.2% 78641337 ? 3% perf-stat.i.branch-misses
1.901e+08 ? 4% -12.4% 1.666e+08 ? 3% perf-stat.i.cache-misses
6.443e+08 ? 3% -16.9% 5.354e+08 ? 2% perf-stat.i.cache-references
2264 ? 3% +5.6% 2391 perf-stat.i.context-switches
1.43e+11 -9.1% 1.3e+11 perf-stat.i.cpu-cycles
1208 ? 12% +65.1% 1996 ? 13% perf-stat.i.cycles-between-cache-misses
1.367e+10 ? 11% +18.3% 1.617e+10 perf-stat.i.dTLB-loads
6.539e+09 ? 12% +18.5% 7.749e+09 perf-stat.i.dTLB-stores
4.362e+10 ? 11% +18.3% 5.161e+10 perf-stat.i.instructions
18628 ? 9% +24.9% 23262 ? 3% perf-stat.i.instructions-per-iTLB-miss
0.36 ? 4% +16.8% 0.42 ? 4% perf-stat.i.ipc
1489433 -9.2% 1353122 perf-stat.i.metric.GHz
2.645e+08 ? 11% +17.3% 3.103e+08 perf-stat.i.metric.M/sec
3110511 ? 3% -18.6% 2532907 ? 4% perf-stat.i.node-loads
12603127 ? 13% +23.6% 15575486 perf-stat.i.node-store-misses
15.00 ? 13% -54.5% 6.83 ? 70% perf-stat.overall.MPKI
3.33 ? 12% -50.0% 1.66 ? 70% perf-stat.overall.cpi
1.885e+08 ? 4% -42.5% 1.084e+08 ? 70% perf-stat.ps.cache-misses
6.388e+08 ? 3% -45.3% 3.493e+08 ? 70% perf-stat.ps.cache-references
1.418e+11 ? 2% -40.0% 8.504e+10 ? 70% perf-stat.ps.cpu-cycles
3083897 ? 3% -47.0% 1634468 ? 70% perf-stat.ps.node-loads
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.
---
0DAY/LKP+ Test Infrastructure Open Source Technology Center
https://lists.01.org/hyperkitty/list/[email protected] Intel Corporation
Thanks,
Oliver Sang