Greeting, FYI, we noticed a -19.1% regression of stress-ng.sync-file.ops_per_sec due to commit: commit: 0279bbbbc03f2ce574c5754f90dfb36e29811bbd ("xfs: CIL checkpoint flushes caches unconditionally") https://git.kernel.org/cgit/linux/kernel/git/dgc/linux-xfs.git xfs-cil-scale-2 in testcase: stress-ng on test machine: 96 threads 2 sockets Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz with 512G memory with following parameters: nr_threads: 10% disk: 1HDD testtime: 60s fs: xfs class: filesystem test: sync-file cpufreq_governor: performance ucode: 0x5003006 If you fix the issue, kindly add following tag Reported-by: kernel test robot <oliver.sang(a)intel.com&gt; Details are as below: --------------------------------------------------------------------------------------------------> To reproduce: git clone https://github.com/intel/lkp-tests.git cd lkp-tests 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 bin/lkp run generated-yaml-file ========================================================================================= class/compiler/cpufreq_governor/disk/fs/kconfig/nr_threads/rootfs/tbox_group/test/testcase/testtime/ucode: filesystem/gcc-9/performance/1HDD/xfs/x86_64-rhel-8.3/10%/debian-10.4-x86_64-20200603.cgz/lkp-csl-2sp7/sync-file/stress-ng/60s/0x5003006 commit: db7e30204e ("xfs: async blkdev cache flush") 0279bbbbc0 ("xfs: CIL checkpoint flushes caches unconditionally") db7e30204e4c2054 0279bbbbc03f2ce574c5754f90d ---------------- --------------------------- %stddev %change %stddev \ | \ 7785 -19.1% 6299 ± 3% stress-ng.sync-file.ops 129.73 -19.1% 104.96 ± 3% stress-ng.sync-file.ops_per_sec 171.83 ± 66% +477.4% 992.17 ± 39% stress-ng.time.involuntary_context_switches 94054 -15.4% 79613 stress-ng.time.voluntary_context_switches 3797269 ± 38% -34.4% 2492527 ± 8% cpuidle.C1.time 21288 +12.2% 23894 ± 2% interrupts.315:PCI-MSI.376832-edge.ahci[0000:00:17.0] 34683 -21.9% 27074 proc-vmstat.pgpgout 12144 +10.8% 13451 ± 2% softirqs.BLOCK 0.06 ± 37% -0.0 0.04 ± 12% turbostat.C1% 95.66 +1.5% 97.12 iostat.cpu.idle 2.04 -69.6% 0.62 ± 18% iostat.cpu.iowait 1.91 ± 2% -1.3 0.58 ± 18% mpstat.cpu.all.iowait% 0.20 -0.0 0.17 ± 3% mpstat.cpu.all.sys% 886.83 ± 9% +71.3% 1519 ± 38% slabinfo.ip6-frags.active_objs 886.83 ± 9% +71.3% 1519 ± 38% slabinfo.ip6-frags.num_objs 534.83 -21.9% 417.83 vmstat.io.bo 1.67 ± 28% -100.0% 0.00 vmstat.procs.b 0.01 ±223% +0.1 0.09 ± 33% perf-profile.children.cycles-pp.wait_for_completion_io 0.01 ±223% +0.1 0.09 ± 33% perf-profile.children.cycles-pp.io_schedule_timeout 0.04 ±106% +0.1 0.18 ± 25% perf-profile.children.cycles-pp.__flush_work 0.03 ±102% +0.1 0.18 ± 27% perf-profile.children.cycles-pp.wait_for_completion 0.10 ± 17% +0.2 0.25 ± 26% perf-profile.children.cycles-pp.xlog_cil_force_lsn 0.08 ± 60% +0.2 0.26 ± 24% perf-profile.children.cycles-pp.schedule_timeout 0.19 ± 17% +0.2 0.38 ± 17% perf-profile.children.cycles-pp.xfs_log_force_lsn 0.19 ± 21% +0.3 0.48 ± 28% perf-profile.children.cycles-pp.pick_next_task_fair 0.50 ± 12% +0.3 0.81 ± 19% perf-profile.children.cycles-pp.__schedule 0.34 ± 10% +0.3 0.66 ± 21% perf-profile.children.cycles-pp.schedule 0.17 ± 24% +0.3 0.49 ± 31% perf-profile.children.cycles-pp.newidle_balance 0.12 ± 12% -0.0 0.08 ± 27% perf-profile.self.cycles-pp.tick_sched_timer 0.40 ± 16% +0.1 0.50 ± 16% perf-profile.self.cycles-pp.__hrtimer_next_event_base stress-ng.time.voluntary_context_switches 100000 +------------------------------------------------------------------+ | + + + | | :: : :: | 95000 |-+ : :: : : : | |+++. ++ ++.+++++++.++++++ ++ ++.+++++++.++++++.+++++++.+++++ +++| | + + | 90000 |-+ | | | 85000 |-+ | | | | O | 80000 |-+ O O | | OOOOO OOOOO O O OOOO OOO OO O | |OOO O O OOO OOO | 75000 +------------------------------------------------------------------+ stress-ng.sync-file.ops 8000 +--------------------------------------------------------------------+ |++. + .+ ++ +.+ ++ .+ + : ++ ++. + +.+++ +. ++ .++++++.++++ .++| | ++ ++ + + + + + + +++ + + ++ + + | 7500 |-+ | | | | | 7000 |-+ | | | 6500 |-+ OO | | O | |OO OO O O O | 6000 |-+ OOO OOOOOO O O OO O O OOO O O | | O O O O OOO | | | 5500 +--------------------------------------------------------------------+ stress-ng.sync-file.ops_per_sec 135 +---------------------------------------------------------------------+ | + +. + | 130 |++.+++++.+++++.++++.+++++.+ ++ +++++.+++++.+++++.++++.+++++.+++ +.++| 125 |-+ | | | 120 |-+ | | | 115 |-+ | | | 110 |-+ OO | 105 |-+ O | |OO OO O O O | 100 |-+ OOO OOOOO O O OO OO O OO O O | | O O O O O OOO | 95 +---------------------------------------------------------------------+ [*] 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. --- 0DAY/LKP+ Test Infrastructure Open Source Technology Center https://lists.01.org/hyperkitty/list/lkp@lists.01.org Intel Corporation Thanks, Oliver Sang