Greeting, FYI, we noticed a 21.9% improvement of stress-ng.eventfd.ops due to commit: commit: fd7732e033e30b3a586923b57e338c859e17858a ("fs/locks: create a tree of dependent requests.") https://git.kernel.org/cgit/linux/kernel/git/jlayton/linux.git locks-next in testcase: stress-ng on test machine: 88 threads Intel(R) Xeon(R) CPU E5-2699 v4 @ 2.20GHz with 64G memory with following parameters: nr_threads: 100% disk: 1HDD testtime: 1s class: filesystem ucode: 0xb00002e cpufreq_governor: performance 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 run job.yaml ========================================================================================= class/compiler/cpufreq_governor/disk/kconfig/nr_threads/rootfs/tbox_group/testcase/testtime/ucode: filesystem/gcc-7/performance/1HDD/x86_64-rhel-7.2/100%/debian-x86_64-2018-04-03.cgz/lkp-bdw-ep3/stress-ng/1s/0xb00002e commit: c0e1590897 ("fs/locks: change all *_conflict() functions to return bool.") fd7732e033 ("fs/locks: create a tree of dependent requests.") c0e15908979d269a fd7732e033e30b3a586923b57e ---------------- -------------------------- %stddev %change %stddev \ | \ 1184559 ± 14% +21.9% 1443641 ± 12% stress-ng.eventfd.ops 1184698 ± 14% +21.9% 1443665 ± 12% stress-ng.eventfd.ops_per_sec 32423 ± 3% +241.2% 110636 stress-ng.fcntl.ops 32422 ± 3% +241.2% 110628 stress-ng.fcntl.ops_per_sec 1.906e+08 -6.5% 1.781e+08 perf-stat.node-loads 1053 ±142% -95.4% 48.25 meminfo.Active(file) 293886 ± 10% -23.5% 224694 ± 13% meminfo.DirectMap4k 13992028 ± 4% -8.7% 12777987 ± 4% numa-numastat.node0.local_node 13996312 ± 4% -8.7% 12782281 ± 4% numa-numastat.node0.numa_hit 602968 ± 10% -51.1% 294959 ± 31% turbostat.C1E 0.83 ± 4% -0.2 0.60 ± 14% turbostat.C1E% 31784409 ± 4% -27.5% 23043591 ± 14% cpuidle.C1E.time 606812 ± 10% -50.9% 298135 ± 30% cpuidle.C1E.usage 293255 ± 17% -31.2% 201788 ± 10% cpuidle.POLL.time 263.00 ±142% -95.4% 12.00 proc-vmstat.nr_active_file 263.00 ±142% -95.4% 12.00 proc-vmstat.nr_zone_active_file 27429189 ± 3% -8.9% 24977810 ± 6% proc-vmstat.numa_hit 27411960 ± 3% -8.9% 24960610 ± 6% proc-vmstat.numa_local 24864411 ± 4% -6.4% 23283878 ± 5% proc-vmstat.unevictable_pgs_culled 6.16 ± 84% -6.2 0.00 perf-profile.calltrace.cycles-pp.__d_lookup.lookup_fast.path_openat.do_filp_open.do_sys_open 6.16 ± 84% -6.2 0.00 perf-profile.calltrace.cycles-pp.lookup_fast.path_openat.do_filp_open.do_sys_open.do_syscall_64 6.16 ± 84% -6.2 0.00 perf-profile.children.cycles-pp.__d_lookup 6.16 ± 84% -6.2 0.00 perf-profile.children.cycles-pp.lookup_fast 6.57 ±100% -5.5 1.04 ±173% perf-profile.children.cycles-pp.vsnprintf 6.16 ± 84% -6.2 0.00 perf-profile.self.cycles-pp.__d_lookup 4899 ± 24% -27.5% 3550 ± 6% sched_debug.cfs_rq:/.min_vruntime.avg 13466 ± 18% -19.7% 10814 ± 9% sched_debug.cfs_rq:/.min_vruntime.max 604.50 ± 19% -17.3% 500.00 ± 2% sched_debug.cfs_rq:/.util_est_enqueued.max 11536 ±111% -79.6% 2357 ± 22% sched_debug.cpu.avg_idle.min 164.50 ± 4% +15.2% 189.50 ± 5% sched_debug.cpu.nr_switches.min 12.25 ± 23% +67.3% 20.50 ± 14% sched_debug.cpu.nr_uninterruptible.max 3.16 ± 3% +13.7% 3.59 ± 5% sched_debug.cpu.nr_uninterruptible.stddev 226.75 ± 55% -74.4% 58.00 slabinfo.btrfs_extent_buffer.active_objs 226.75 ± 55% -74.4% 58.00 slabinfo.btrfs_extent_buffer.num_objs 213.00 ± 26% +41.9% 302.25 ± 16% slabinfo.nfs_commit_data.active_objs 213.00 ± 26% +41.9% 302.25 ± 16% slabinfo.nfs_commit_data.num_objs 279.75 ± 34% +57.7% 441.25 ± 21% slabinfo.nfs_read_data.active_objs 279.75 ± 34% +57.7% 441.25 ± 21% slabinfo.nfs_read_data.num_objs 654.25 ± 17% +22.2% 799.25 ± 9% slabinfo.skbuff_fclone_cache.active_objs 654.25 ± 17% +22.2% 799.25 ± 9% slabinfo.skbuff_fclone_cache.num_objs stress-ng.fcntl.ops 120000 +-+----------------------------------------------------------------+ O O O O O O O O O O O O O O O O O O | 110000 +-O O O O O O O | 100000 +-+ | | | 90000 +-+ | 80000 +-+ | | | 70000 +-+ | 60000 +-+ | | | 50000 +-+ | 40000 +-+ | |. .+.+..+.+. .+.+..+.+. .+.+.+..+. .+. .+.. .+. .+. .+.| 30000 +-+----------------------------------------------------------------+ stress-ng.fcntl.ops_per_sec 120000 +-+----------------------------------------------------------------+ O O O O O O O O O O O O O O O O O O | 110000 +-O O O O O O O | 100000 +-+ | | | 90000 +-+ | 80000 +-+ | | | 70000 +-+ | 60000 +-+ | | | 50000 +-+ | 40000 +-+ | |. .+.+..+.+. .+.+..+.+. .+.+.+..+. .+. .+.. .+. .+. .+.| 30000 +-+----------------------------------------------------------------+ [*] 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, Rong Chen