Greeting, We noticed a -14% regression of aim7.jobs_per_min due to commit: commit: 0db1ff222d40f1601c961f0edb86d10426992595 ("ext4: add shutdown bit and check for it") https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git master in testcase: aim7 on test machine: 40 threads Intel(R) Xeon(R) CPU E5-2690 v2 @ 3.00GHz with 384G memory with following parameters: disk: 4BRD_12G md: RAID0 fs: ext4 test: creat-clo load: 1000 cpufreq_governor: performance test-description: AIM7 is a traditional UNIX system level benchmark suite which is used to test and measure the performance of multiuser system. test-url: https://sourceforge.net/projects/aimbench/files/aim-suite7/ 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: aim7/4BRD_12G-RAID0-ext4-creat-clo-1000-performance/lkp-ivb-ep01 9549a168bd500db1 0db1ff222d40f1601c961f0edb ---------------- -------------------------- %stddev change %stddev \ | \ 18884 -14% 16185 aim7.jobs-per-min 249058 ± 3% 20% 297672 ± 4% aim7.time.involuntary_context_switches 10788 18% 12722 ± 3% aim7.time.system_time 317 17% 371 aim7.time.elapsed_time 317 17% 371 aim7.time.elapsed_time.max 693826 ± 4% 15% 798510 aim7.time.voluntary_context_switches 176592 ± 4% 7% 189018 ± 4% aim7.time.minor_page_faults 241675 ± 4% 13% 272360 interrupts.CAL:Function_call_interrupts 6612 -15% 5636 iostat.md0.wkB/s 4.753e+09 ± 6% 23% 5.84e+09 ± 8% perf-stat.node-loads 4.401e+09 ± 6% 22% 5.391e+09 ± 8% perf-stat.node-load-misses 47335355 ± 6% 21% 57398968 ± 11% perf-stat.iTLB-loads 3.22e+13 18% 3.801e+13 ± 4% perf-stat.cpu-cycles 7.966e+09 ± 4% 18% 9.364e+09 ± 5% perf-stat.cache-misses 2.526e+10 ± 3% 17% 2.961e+10 ± 5% perf-stat.cache-references 449434 ± 3% 16% 521919 ± 3% perf-stat.cpu-migrations 1998247 ± 3% 16% 2316905 perf-stat.context-switches 3.565e+12 16% 4.127e+12 ± 3% perf-stat.branch-instructions 1.799e+13 16% 2.08e+13 ± 3% perf-stat.instructions 4.834e+12 15% 5.56e+12 ± 3% perf-stat.dTLB-loads 780935 14% 890503 perf-stat.page-faults 780942 14% 890503 perf-stat.minor-faults 3.101e+09 9% 3.392e+09 perf-stat.node-stores 2.238e+09 9% 2.437e+09 perf-stat.branch-misses 2.229e+09 7% 2.383e+09 perf-stat.node-store-misses 41.81 41.26 perf-stat.node-store-miss-rate% 0.56 0.55 perf-stat.ipc 0.06 -6% 0.06 perf-stat.branch-miss-rate% perf-stat.node-stores 3.45e+09 ++---------------------------------------------------------------+ O O O O | 3.4e+09 ++O O O O O O O O O O | 3.35e+09 ++ O O O | | | 3.3e+09 ++ | 3.25e+09 ++ | | | 3.2e+09 ++ .* | 3.15e+09 ++ * + *. .*. | | *. : * .*. + *.* *.*.* *.*.* | 3.1e+09 *+ + *.* : + .* * .* : .*. + :| 3.05e+09 ++*.* + : * + .*. : .* *.* :| | * * * * 3e+09 ++---------------------------------------------------------------+ perf-stat.node-store-misses 2.45e+09 ++---------------------------------------------------------------+ O | 2.4e+09 ++ O O | | O O O O O O O O O O O | | O O O | 2.35e+09 ++ | | | 2.3e+09 ++ * | | + + *. *. | 2.25e+09 ++ *. * * * : *. + *. .* *.*.* | * + *.* : + + + : * * : .*. + :| |+ .* + : *.* *. .*..* : .* *.* :| 2.2e+09 ++* * * * | | * 2.15e+09 ++---------------------------------------------------------------+ perf-stat.page-faults 960000 ++-----------------------------------------------------------------+ 940000 ++ O | | O | 920000 ++O O O O | 900000 O+ O O O O O O | 880000 ++ O O O O | 860000 ++ | | | 840000 ++ | 820000 ++ .* | 800000 *+ * *.* + .* .* .*.* .* * | 780000 ++ .*. + + + *.* *. + .* + .* + .*. + + + .* | | * *.*.*..* * + + * * *.* * * +| 760000 ++ * * 740000 ++-----------------------------------------------------------------+ perf-stat.minor-faults 960000 ++-----------------------------------------------------------------+ 940000 ++ O | | O | 920000 ++O O O O | 900000 O+ O O O O O O | 880000 ++ O O O O | 860000 ++ | | | 840000 ++ | 820000 ++ .* | 800000 *+ * *.* + .* .* .*.* .* * | 780000 ++ .*. + + + *.* *. + .* + .* + .*. + + + .* | | * *.*.*..* * + + * * *.* * * +| 760000 ++ * * 740000 ++-----------------------------------------------------------------+ perf-stat.node-store-miss-rate_ 42 ++-------------------------------------------------------------------+ *. .*.*.. * *. .*. * * * *.. | 41.9 ++* *. + + + *..* *.* *..*. + + + + .*. + + + *. .* | 41.8 ++ * * + + * *.* *..* * * * +| | * * 41.7 ++ | 41.6 ++ | | | 41.5 ++ | 41.4 O+ | | O O O O O O | 41.3 ++O O O O O | 41.2 ++ O O O O | | O | 41.1 ++-------------------------------------------------------------------+ aim7.jobs-per-min 20000 ++------------------------------------------------------------------+ | * | 19500 ++ .* * + + * * * 19000 ++ .*.*..*.* + + : *.* * + + + : +| |.* * : .. + .*.*.*..*.*.*.* * : .*.*.*.* | 18500 *+ *.*.* * *. | 18000 ++ | | | 17500 ++ | 17000 ++ | | | 16500 ++ O O O O O O | 16000 O+O O O O O | | O O O O O | 15500 ++------------------------------------------------------------------+ aim7.time.elapsed_time 390 ++--------------------------------------------------------------------+ | O O | 380 ++ O O O O O O | 370 O+O O O O | | O O O O | 360 ++ | 350 ++ | | | 340 ++ | 330 ++ | *. *.. *.*.* .*. .*.. *. | 320 ++*. + + + .*.. * *.*..*.*.*.* .*. + *.*..*.*.| 310 ++ *..*.*.*.* * * + * * * | * | 300 ++--------------------------------------------------------------------+ aim7.time.elapsed_time.max 390 ++--------------------------------------------------------------------+ | O O | 380 ++ O O O O O O | 370 O+O O O O | | O O O O | 360 ++ | 350 ++ | | | 340 ++ | 330 ++ | *. *.. *.*.* .*. .*.. *. | 320 ++*. + + + .*.. * *.*..*.*.*.* .*. + *.*..*.*.| 310 ++ *..*.*.*.* * * + * * * | * | 300 ++--------------------------------------------------------------------+ iostat.md0.wkB_s 8000 ++-------------------------------------------------------------------+ | | 7000 ++*.*.*..*.*.*.*.*. .*.*.*.*.*.*..*. .*. .*. .* * *. *.*.*.*.*.*..* * *.*..*.*.* | 6000 ++ O O O O O O O O O | | O O O O O O | 5000 ++ | | | 4000 ++ | | | 3000 ++ | | | 2000 O+O | | | 1000 ++-------------------------------------------------------------------+ [*] 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