Greeting, FYI, we noticed a 8.9% improvement of netperf.Throughput_Mbps due to commit: commit: dc86d23b330b040a0d64a9e9c0f2e6fea2dac89a ("mm: page_alloc: High-order per-cpu page allocator") https://git.kernel.org/pub/scm/linux/kernel/git/mel/linux.git mm-pagealloc-highorder-percpu-v2r4 in testcase: netperf on test machine: 16 threads Intel(R) Xeon(R) CPU D-1541 @ 2.10GHz with 8G memory with following parameters: ip: ipv4 runtime: 300s nr_threads: 25% cluster: cs-localhost send_size: 10K test: SCTP_STREAM_MANY cpufreq_governor: performance test-description: Netperf is a benchmark that can be use to measure various aspect of networking performance. test-url: http://www.netperf.org/netperf/ Details are as below: --------------------------------------------------------------------------------------------------> To reproduce: git clone git://git.kernel.org/pub/scm/linux/kernel/git/wfg/lkp-tests.git cd lkp-tests bin/lkp install job.yaml # job file is attached in this email bin/lkp run job.yaml testcase/path_params/tbox_group/run: netperf/ipv4-300s-25%-cs-localhost-10K-SCTP_STREAM_MANY-performance/lkp-bdw-de1 v4.9-rc5 dc86d23b330b040a0d64a9e9c0 ---------------- -------------------------- %stddev change %stddev \ | \ 14333 9% 15603 netperf.Throughput_Mbps 1574 ± 5% 131% 3630 netperf.time.voluntary_context_switches 386 -5% 365 netperf.time.percent_of_cpu_this_job_got 1152 -5% 1089 netperf.time.system_time 395677 9% 429320 vmstat.system.cs 8.25 17% 9.67 turbostat.RAMWatt 35.55 35.95 turbostat.PkgWatt 2.231e+08 16% 2.583e+08 ± 5% perf-stat.dTLB-load-misses 14144 12% 15850 perf-stat.cpu-migrations 0.02 9% 0.02 ± 5% perf-stat.dTLB-load-miss-rate% 1.203e+08 9% 1.305e+08 perf-stat.context-switches 8.175e+11 7% 8.774e+11 perf-stat.dTLB-stores 0.68 6% 0.72 perf-stat.ipc 1.126e+12 6% 1.197e+12 perf-stat.dTLB-loads 5.804e+11 5% 6.123e+11 perf-stat.branch-instructions 3.105e+12 5% 3.27e+12 perf-stat.instructions 2.129e+09 5% 2.226e+09 perf-stat.iTLB-loads 2.13e+11 -6% 2.003e+11 perf-stat.cache-references 2.13e+11 -6% 2.003e+11 perf-stat.cache-misses 1.624e+09 -26% 1.203e+09 ± 3% perf-stat.branch-misses 0.28 -30% 0.20 ± 3% perf-stat.branch-miss-rate% perf-stat.branch-misses 1.8e+09 *+----------------------------------------------------------------+ |+ .*. .*. *.*.*.*.*.*.*.*.*. .*. .*.*. .**.*. .*. .*.*.| 1.6e+09 ++* * *.*.*.* *.* * * *.* * * 1.4e+09 ++ | | O O OO O O O O O O | 1.2e+09 O+O O O O O O O O O O O | 1e+09 ++ | | | 8e+08 ++ | 6e+08 ++ | | | 4e+08 ++ | 2e+08 ++ | | | 0 ++--------O-------------------------------------------------------+ perf-stat.context-switches 1.4e+08 ++----------------------------------------------------------------+ O O O O O O O OO O O O O O O O O O O O O O | 1.2e+08 *+*.*.*.*.*.*.*.**.*.*.*.*.*.*.*.*.*.*.*.*.*.*.*.**.*.*.*.*.*.*.*.* | | 1e+08 ++ | | | 8e+07 ++ | | | 6e+07 ++ | | | 4e+07 ++ | | | 2e+07 ++ | | | 0 ++--------O-------------------------------------------------------+ turbostat.RAMWatt 10 ++O-O-O-O----O-O-O-----O-O---O-O-O--O-O-O-O-O-O------------------------+ 9 O+ O O O | *.*.*.*.*. .*. .*.*.*.*.*.*.*. .*.*.*.*.*.*.*.*.*. .*.*..*.*.*.*.*.* 8 ++ *. *.* *. * | 7 ++ | | | 6 ++ | 5 ++ | 4 ++ | | | 3 ++ | 2 ++ | | | 1 ++ | 0 ++--------O------------------------------------------------------------+ netperf.Throughput_Mbps 16000 O+----O-O---O-O-O-O-O-O-O---O-O-O-O-O-O-O-O-O-----------------------+ *.O.O. .*.*.*.*.*.*.*. .O.*.*.*.*.*. .*.*.*.*.*.*.*.*.*. .*.*.*.*.* 14000 ++ *.* * * * | 12000 ++ | | | 10000 ++ | | | 8000 ++ | | | 6000 ++ | 4000 ++ | | | 2000 ++ | | | 0 ++--------O---------------------------------------------------------+ netperf.time.system_time 1200 ++--------*-------*-----------*--------------------*-----------------+ O.*.O.O.O O.O.O O O.O.*.*.O O O..O.O.O.O.O.O.*.* *.*.*.*.*.*.*.*.* 1000 ++O O O | | | | | 800 ++ | | | 600 ++ | | | 400 ++ | | | | | 200 ++ | | | 0 ++--------O----------------------------------------------------------+ netperf.time.percent_of_cpu_this_job_got 400 ++--*-*---*-*-*-*--*-*-*---*-*-*-*---*-*-*-*-*---*-*--*---*-*-*-*-*---* O.O O O O O O O O O O O O O O O O O O O O O * * * | 350 ++ | 300 ++ | | | 250 ++ | | | 200 ++ | | | 150 ++ | 100 ++ | | | 50 ++ | | | 0 ++--------O-----------------------------------------------------------+ netperf.time.voluntary_context_switches 4500 ++-------------------------------------------------------------------+ | O O O O O | 4000 O+ O O O O O O O O O O O | 3500 ++ O O O O O | | | 3000 ++ | 2500 ++ | | | 2000 *+ | 1500 ++*.*.*.*.*.*.*.*.*.*.*.*.*.*.*.*..*.*.*.*.*.*.*.*.*.*.*.*.*.*.*.*.*.* | | 1000 ++ | 500 ++ | | | 0 ++--------O----------------------------------------------------------+ vmstat.system.cs 450000 ++-----------------------------------------------------------------+ O O O O O O O O O O O O O O O O OO O O O O | 400000 *+*.*.*.*.*.*.*.*.*.*.*.*.*.*.*.*.**.*.*.*.*.*.*.*.*.*.*.*.*.*.*.*.* 350000 ++ | | | 300000 ++ | 250000 ++ | | | 200000 ++ | 150000 ++ | | | 100000 ++ | 50000 ++ | | | 0 ++--------O--------------------------------------------------------+ [*] bisect-good sample [O] bisect-bad sample 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. Thanks, Xiaolong