Greeting, FYI, we noticed a -3% regression of vm-scalability.throughput due to commit: commit: 27ff9bf7cdf588eda950f5ddfe8a6f0045258c58 ("mm, THP, swap: make reuse_swap_page() works for THP swapped out") https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git master in testcase: vm-scalability on test machine: 88 threads Intel(R) Xeon(R) CPU E5-2699 v4 @ 2.20GHz with 128G memory with following parameters: runtime: 300s size: 8T test: anon-cow-seq cpufreq_governor: performance test-description: The motivation behind this suite is to exercise functions and regions of the mm/ of the Linux kernel which are of interest to us. test-url: https://git.kernel.org/cgit/linux/kernel/git/wfg/vm-scalability.git/ Details are as below: --------------------------------------------------------------------------------------------------> To reproduce: git clone https://github.com/01org/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: vm-scalability/300s-8T-anon-cow-seq-performance/lkp-bdw-ep2 91c807e922afbb73 27ff9bf7cdf588eda950f5ddfe ---------------- -------------------------- %stddev change %stddev \ | \ 350099 346898 vm-scalability.median 31125270 -3% 30202484 vm-scalability.throughput 7.846e+09 -4% 7.568e+09 vm-scalability.workload 0.04 ± 4% -26% 0.03 ± 9% vm-scalability.median_stddev 212213 ± 4% 84% 391164 vm-scalability.time.voluntary_context_switches 171709 8% 185688 vm-scalability.time.involuntary_context_switches 9936 10124 vm-scalability.time.system_time 7302 7237 vm-scalability.time.percent_of_cpu_this_job_got 11544055 11272539 vm-scalability.time.minor_page_faults 12152 11810 vm-scalability.time.user_time 2558212 ± 3% -32% 1734828 ± 3% interrupts.CAL:Function_call_interrupts 62707 -3% 60767 pmeter.performance_per_watt 8843 ± 13% 32% 11632 ± 11% vmstat.system.cs 100508 97669 vmstat.system.in 0 1e+04 9948 ± 17% latency_stats.max.io_schedule.__lock_page.do_huge_pmd_wp_page.__handle_mm_fault.handle_mm_fault.__do_page_fault.do_page_fault.page_fault 20099 ± 88% -1e+04 6788 ± 73% latency_stats.max.call_rwsem_down_read_failed.page_lock_anon_vma_read.do_huge_pmd_numa_page.__handle_mm_fault.handle_mm_fault.__do_page_fault.do_page_fault.page_fault 0 8e+06 7773763 latency_stats.sum.io_schedule.__lock_page.do_huge_pmd_wp_page.__handle_mm_fault.handle_mm_fault.__do_page_fault.do_page_fault.page_fault 39557 ± 21% -3e+04 6076 ± 24% latency_stats.sum.call_rwsem_down_read_failed.page_lock_anon_vma_read.rmap_walk_anon.rmap_walk.try_to_unmap.migrate_pages.migrate_misplaced_page.__handle_mm_fault.handle_mm_fault.__do_page_fault.do_page_fault.page_fault 0.36 66% 0.60 perf-stat.node-store-miss-rate% 96588646 55% 1.494e+08 perf-stat.node-store-misses 2.624e+09 47% 3.868e+09 perf-stat.node-loads 3.328e+09 ± 3% 44% 4.787e+09 perf-stat.node-load-misses 139122 ± 3% 40% 194820 perf-stat.cpu-migrations 2682774 ± 13% 32% 3537536 ± 11% perf-stat.context-switches 1.106e+08 ± 3% 9% 1.211e+08 ± 4% perf-stat.iTLB-load-misses 0.00 8% 0.00 perf-stat.dTLB-load-miss-rate% 0.01 5% 0.01 perf-stat.branch-miss-rate% 55.13 4% 57.26 perf-stat.iTLB-load-miss-rate% 0.00 3% 0.00 perf-stat.dTLB-store-miss-rate% 2.38 3% 2.46 perf-stat.cpi 27.21 27.70 perf-stat.cache-miss-rate% 1.087e+09 1.106e+09 perf-stat.branch-misses 4.403e+11 4.357e+11 perf-stat.cache-references 12092547 11842335 perf-stat.minor-faults 12092547 11842319 perf-stat.page-faults 0.42 0.41 perf-stat.ipc 2.129e+12 -3% 2.059e+12 perf-stat.dTLB-stores 8.182e+12 -3% 7.906e+12 perf-stat.branch-instructions 2.571e+13 -3% 2.484e+13 perf-stat.instructions 5.747e+12 -3% 5.549e+12 perf-stat.dTLB-loads 2.659e+10 -7% 2.475e+10 perf-stat.node-stores 232836 ± 4% -12% 205592 ± 4% perf-stat.instructions-per-iTLB-miss vm-scalability.time.user_time 12300 ++------------------------------------------------------------------+ | * * * | 12200 ++: .*. .*.*. : .* : .*.* * *.| |: : *. * * *. .* : : * : : : * : :: : * 12100 ++ : : : : *. + .*.*. : : + : : : : : : : : | * :: : : * *.* :+ * :: : : * | 12000 ++ * * * * * | | | 11900 ++ | | O O O | 11800 ++ O O O O O O O O O O O O O | | | 11700 ++ O O O | O O O O | 11600 ++------------------------------------------------------------------+ vm-scalability.time.voluntary_context_switches 450000 ++-----------------------------------------------------------------+ O O O | 400000 ++ O O O O O O O O O O O O | | O O O O O O O O | | | 350000 ++ | | | 300000 ++ | | | 250000 ++ | | *.. | |.*. + .*. .*. .*.*.* 200000 *+ *.*.* *.*.*.*. .*.*. .*.* *.*.*.* *.*.*..*.*.* | | *.*.* *. | 150000 ++-----------------------------------------------------------------+ [*] 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