Greeting, FYI, we noticed a 164.9% improvement of fxmark.ssd_xfs_MRPH_72_bufferedio.works/sec due to commit: commit: f5fe24ef17b5fbe6db49534163e77499fb10ae8c ("lockref: stop doing cpu_relax in the cmpxchg loop") https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master in testcase: fxmark on test machine: 128 threads 2 sockets Intel(R) Xeon(R) Platinum 8358 CPU @ 2.60GHz (Ice Lake) with 128G memory with following parameters: disk: 1SSD media: ssd test: MRPH fstype: xfs directio: bufferedio cpufreq_governor: performance test-description: FxMark is a filesystem benchmark that test multicore scalability. test-url: https://github.com/sslab-gatech/fxmark In addition to that, the commit also has significant impact on the following tests: +------------------+----------------------------------------------------------------+ | testcase: change | will-it-scale: will-it-scale.per_process_ops 73.1% improvement | | test machine | 104 threads 2 sockets (Skylake) with 192G memory | | test parameters | cpufreq_governor=performance | | | mode=process | | | nr_task=16 | | | test=open3 | +------------------+----------------------------------------------------------------+ Details are as below: --------------------------------------------------------------------------------------------------> To reproduce: git clone https://github.com/intel/lkp-tests.git cd lkp-tests sudo 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 sudo bin/lkp run generated-yaml-file # if come across any failure that blocks the test, # please remove ~/.lkp and /lkp dir to run from a clean state. ========================================================================================= compiler/cpufreq_governor/directio/disk/fstype/kconfig/media/rootfs/tbox_group/test/testcase: gcc-11/performance/bufferedio/1SSD/xfs/x86_64-rhel-8.3/ssd/debian-11.1-x86_64-20220510.cgz/lkp-icl-2sp5/MRPH/fxmark commit: 0bf913e07b ("Merge tag 'efi-fixes-for-v6.2-1' of git://git.kernel.org/pub/scm/linux/kernel/git/efi/efi") f5fe24ef17 ("lockref: stop doing cpu_relax in the cmpxchg loop") 0bf913e07b377cfc f5fe24ef17b5fbe6db49534163e ---------------- --------------------------- %stddev %change %stddev \ | \ 113.20 ± 9% +24.2% 140.63 ± 3% fxmark.ssd_xfs_MRPH_18_bufferedio.user_sec 12.57 ± 9% +24.2% 15.61 ± 3% fxmark.ssd_xfs_MRPH_18_bufferedio.user_util 2.898e+08 ± 9% +23.4% 3.577e+08 ± 2% fxmark.ssd_xfs_MRPH_18_bufferedio.works 5796755 ± 9% +23.4% 7154469 ± 2% fxmark.ssd_xfs_MRPH_18_bufferedio.works/sec 69.00 ± 19% +47.6% 101.84 ± 10% fxmark.ssd_xfs_MRPH_36_bufferedio.user_sec 3.83 ± 19% +47.6% 5.65 ± 10% fxmark.ssd_xfs_MRPH_36_bufferedio.user_util 1.759e+08 ± 18% +42.2% 2.502e+08 ± 11% fxmark.ssd_xfs_MRPH_36_bufferedio.works 3518906 ± 18% +42.2% 5004235 ± 11% fxmark.ssd_xfs_MRPH_36_bufferedio.works/sec 67.29 ± 33% +74.4% 117.34 ± 12% fxmark.ssd_xfs_MRPH_54_bufferedio.user_sec 2.49 ± 33% +74.4% 4.34 ± 12% fxmark.ssd_xfs_MRPH_54_bufferedio.user_util 1.677e+08 ± 34% +74.8% 2.932e+08 ± 12% fxmark.ssd_xfs_MRPH_54_bufferedio.works 3354868 ± 34% +74.8% 5864422 ± 12% fxmark.ssd_xfs_MRPH_54_bufferedio.works/sec 19.26 ± 2% -12.0% 16.95 ± 6% fxmark.ssd_xfs_MRPH_72_bufferedio.irq_sec 0.53 ± 2% -12.0% 0.47 ± 6% fxmark.ssd_xfs_MRPH_72_bufferedio.irq_util 41.56 ± 2% +159.7% 107.93 ± 33% fxmark.ssd_xfs_MRPH_72_bufferedio.user_sec 1.15 ± 2% +159.7% 3.00 ± 33% fxmark.ssd_xfs_MRPH_72_bufferedio.user_util 92456310 ± 2% +164.9% 2.449e+08 ± 34% fxmark.ssd_xfs_MRPH_72_bufferedio.works 1849128 ± 2% +164.9% 4897580 ± 34% fxmark.ssd_xfs_MRPH_72_bufferedio.works/sec 0.72 ± 2% +6.5% 0.77 ± 2% perf-stat.ps.major-faults 151.91 ± 10% +23.4% 187.41 ± 16% sched_debug.cfs_rq:/.util_est_enqueued.avg 13144 ± 28% -41.2% 7729 ± 56% proc-vmstat.numa_pages_migrated 13144 ± 28% -41.2% 7729 ± 56% proc-vmstat.pgmigrate_success *************************************************************************************************** lkp-skl-fpga01: 104 threads 2 sockets (Skylake) with 192G memory ========================================================================================= compiler/cpufreq_governor/kconfig/mode/nr_task/rootfs/tbox_group/test/testcase: gcc-11/performance/x86_64-rhel-8.3/process/16/debian-11.1-x86_64-20220510.cgz/lkp-skl-fpga01/open3/will-it-scale commit: 0bf913e07b ("Merge tag 'efi-fixes-for-v6.2-1' of git://git.kernel.org/pub/scm/linux/kernel/git/efi/efi") f5fe24ef17 ("lockref: stop doing cpu_relax in the cmpxchg loop") 0bf913e07b377cfc f5fe24ef17b5fbe6db49534163e ---------------- --------------------------- %stddev %change %stddev \ | \ 1073256 ± 3% +73.1% 1857962 ± 6% will-it-scale.16.processes 67078 ± 3% +73.1% 116122 ± 6% will-it-scale.per_process_ops 1073256 ± 3% +73.1% 1857962 ± 6% will-it-scale.workload 11873 ±127% -92.2% 929.17 ± 37% numa-meminfo.node1.Mapped 9042529 ± 3% +66.5% 15055720 ± 6% numa-numastat.node0.local_node 9082850 ± 3% +66.1% 15090608 ± 6% numa-numastat.node0.numa_hit 0.05 ± 10% +60.7% 0.08 ± 6% turbostat.IPC 220.56 +3.1% 227.36 turbostat.PkgWatt 1.68 ± 10% -0.3 1.40 ± 2% mpstat.cpu.all.irq% 0.18 ± 4% +0.1 0.26 ± 8% mpstat.cpu.all.soft% 1.19 ± 15% +0.7 1.87 ± 2% mpstat.cpu.all.usr% 9082921 ± 3% +66.1% 15090591 ± 6% numa-vmstat.node0.numa_hit 9042600 ± 3% +66.5% 15055704 ± 6% numa-vmstat.node0.numa_local 2967 ±127% -91.9% 240.17 ± 34% numa-vmstat.node1.nr_mapped 9679445 ± 2% +62.2% 15697699 ± 5% proc-vmstat.numa_hit 9584229 ± 2% +62.8% 15603315 ± 5% proc-vmstat.numa_local 18656711 ± 3% +64.6% 30700650 ± 5% proc-vmstat.pgalloc_normal 18625336 ± 3% +64.7% 30666686 ± 5% proc-vmstat.pgfree 1.465e+09 ± 2% +60.9% 2.356e+09 ± 5% perf-stat.i.branch-instructions 1.80 ± 11% -0.4 1.36 perf-stat.i.branch-miss-rate% 26733272 ± 10% +21.2% 32400604 ± 4% perf-stat.i.branch-misses 1.713e+08 ± 12% +46.9% 2.516e+08 ± 9% perf-stat.i.cache-references 6.38 ± 3% -39.4% 3.87 ± 5% perf-stat.i.cpi 4.787e+10 -2.0% 4.693e+10 perf-stat.i.cpu-cycles 3361695 ± 13% +19.0% 4002066 ± 5% perf-stat.i.dTLB-load-misses 2.136e+09 ± 2% +63.8% 3.5e+09 ± 5% perf-stat.i.dTLB-loads 0.01 ± 39% -0.0 0.00 ± 14% perf-stat.i.dTLB-store-miss-rate% 1.196e+09 ± 2% +64.2% 1.964e+09 ± 5% perf-stat.i.dTLB-stores 2959344 ± 2% +65.7% 4904453 ± 4% perf-stat.i.iTLB-load-misses 4632906 ± 13% +65.6% 7672899 ± 13% perf-stat.i.iTLB-loads 7.574e+09 ± 2% +61.3% 1.222e+10 ± 5% perf-stat.i.instructions 0.16 ± 3% +64.6% 0.26 ± 5% perf-stat.i.ipc 0.46 -2.0% 0.45 perf-stat.i.metric.GHz 53.96 ± 13% +58.4% 85.45 ± 12% perf-stat.i.metric.K/sec 47.77 ± 2% +62.5% 77.61 ± 5% perf-stat.i.metric.M/sec 24.18 ± 24% -9.3 14.93 ± 20% perf-stat.i.node-load-miss-rate% 83431 ± 8% +21.0% 100988 ± 3% perf-stat.i.node-load-misses 291228 ± 27% +110.6% 613461 ± 23% perf-stat.i.node-loads 7.86 ± 28% -3.4 4.50 ± 20% perf-stat.i.node-store-miss-rate% 242828 ± 23% +93.7% 470348 ± 22% perf-stat.i.node-stores 1.83 ± 10% -0.4 1.38 perf-stat.overall.branch-miss-rate% 6.32 ± 3% -39.1% 3.85 ± 5% perf-stat.overall.cpi 0.01 ± 38% -0.0 0.00 ± 14% perf-stat.overall.dTLB-store-miss-rate% 0.16 ± 3% +64.6% 0.26 ± 5% perf-stat.overall.ipc 23.33 ± 23% -8.6 14.70 ± 19% perf-stat.overall.node-load-miss-rate% 7.90 ± 27% -3.4 4.51 ± 18% perf-stat.overall.node-store-miss-rate% 2131429 -6.7% 1987757 perf-stat.overall.path-length 1.46e+09 ± 2% +60.9% 2.35e+09 ± 5% perf-stat.ps.branch-instructions 26658961 ± 10% +21.4% 32361331 ± 4% perf-stat.ps.branch-misses 1.707e+08 ± 12% +46.9% 2.507e+08 ± 9% perf-stat.ps.cache-references 4.771e+10 -2.0% 4.678e+10 perf-stat.ps.cpu-cycles 110.34 +0.9% 111.38 perf-stat.ps.cpu-migrations 3350393 ± 13% +19.1% 3988846 ± 5% perf-stat.ps.dTLB-load-misses 2.129e+09 ± 2% +63.9% 3.489e+09 ± 5% perf-stat.ps.dTLB-loads 1.192e+09 ± 2% +64.2% 1.958e+09 ± 5% perf-stat.ps.dTLB-stores 2949621 ± 2% +65.7% 4888216 ± 4% perf-stat.ps.iTLB-load-misses 4617332 ± 13% +65.6% 7647709 ± 13% perf-stat.ps.iTLB-loads 7.55e+09 ± 2% +61.4% 1.219e+10 ± 5% perf-stat.ps.instructions 83148 ± 8% +21.2% 100761 ± 3% perf-stat.ps.node-load-misses 290435 ± 27% +110.8% 612364 ± 23% perf-stat.ps.node-loads 19456 ± 3% +9.4% 21281 ± 4% perf-stat.ps.node-store-misses 242169 ± 23% +94.0% 469843 ± 22% perf-stat.ps.node-stores 2.287e+12 ± 2% +61.5% 3.692e+12 ± 5% perf-stat.total.instructions 18.81 ± 59% -18.8 0.00 perf-profile.calltrace.cycles-pp.intel_idle_ibrs.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call.do_idle 18.70 ± 59% -18.7 0.00 perf-profile.calltrace.cycles-pp.mwait_idle_with_hints.intel_idle_ibrs.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call 9.85 ± 8% -4.6 5.23 ± 10% perf-profile.calltrace.cycles-pp.lockref_get_not_dead.__legitimize_path.try_to_unlazy.complete_walk.do_open 10.79 ± 10% -4.6 6.22 ± 9% perf-profile.calltrace.cycles-pp.lockref_get.do_dentry_open.do_open.path_openat.do_filp_open 10.70 ± 11% -4.5 6.22 ± 12% perf-profile.calltrace.cycles-pp.lockref_put_or_lock.dput.terminate_walk.path_openat.do_filp_open 10.96 ± 10% -4.3 6.65 ± 10% perf-profile.calltrace.cycles-pp.dput.terminate_walk.path_openat.do_filp_open.do_sys_openat2 11.02 ± 10% -4.3 6.73 ± 10% perf-profile.calltrace.cycles-pp.terminate_walk.path_openat.do_filp_open.do_sys_openat2.__x64_sys_openat 11.86 ± 7% -4.2 7.67 ± 11% perf-profile.calltrace.cycles-pp.__legitimize_path.try_to_unlazy.complete_walk.do_open.path_openat 11.89 ± 7% -4.2 7.72 ± 11% perf-profile.calltrace.cycles-pp.try_to_unlazy.complete_walk.do_open.path_openat.do_filp_open 11.91 ± 7% -4.2 7.75 ± 11% perf-profile.calltrace.cycles-pp.complete_walk.do_open.path_openat.do_filp_open.do_sys_openat2 10.92 ± 10% -4.2 6.76 ± 12% perf-profile.calltrace.cycles-pp.lockref_put_or_lock.dput.__fput.task_work_run.exit_to_user_mode_loop 11.20 ± 8% -3.9 7.26 ± 11% perf-profile.calltrace.cycles-pp.dput.__fput.task_work_run.exit_to_user_mode_loop.exit_to_user_mode_prepare 12.49 ± 6% -2.2 10.27 ± 11% perf-profile.calltrace.cycles-pp.__fput.task_work_run.exit_to_user_mode_loop.exit_to_user_mode_prepare.syscall_exit_to_user_mode 0.61 ± 5% +0.4 1.04 ± 10% perf-profile.calltrace.cycles-pp.kmem_cache_alloc.__alloc_file.alloc_empty_file.path_openat.do_filp_open 0.66 ± 6% +0.5 1.13 ± 11% perf-profile.calltrace.cycles-pp.getname_flags.do_sys_openat2.__x64_sys_openat.do_syscall_64.entry_SYSCALL_64_after_hwframe 2.55 ± 11% +0.6 3.16 ± 12% perf-profile.calltrace.cycles-pp.__d_lookup_rcu.lookup_fast.open_last_lookups.path_openat.do_filp_open 1.02 ± 6% +0.7 1.70 ± 11% perf-profile.calltrace.cycles-pp.syscall_return_via_sysret.open64 1.02 ± 5% +0.7 1.70 ± 11% perf-profile.calltrace.cycles-pp.syscall_return_via_sysret.__close 0.00 +0.7 0.72 ± 11% perf-profile.calltrace.cycles-pp.__entry_text_start.__close 0.48 ± 44% +0.7 1.20 ± 11% perf-profile.calltrace.cycles-pp.__x64_sys_close.do_syscall_64.entry_SYSCALL_64_after_hwframe.__close 0.00 +0.7 0.73 ± 11% perf-profile.calltrace.cycles-pp.__entry_text_start.open64 0.00 +0.8 0.75 ± 11% perf-profile.calltrace.cycles-pp.strncpy_from_user.getname_flags.do_sys_openat2.__x64_sys_openat.do_syscall_64 0.00 +0.8 0.82 ± 11% perf-profile.calltrace.cycles-pp.link_path_walk.path_openat.do_filp_open.do_sys_openat2.__x64_sys_openat 0.00 +0.9 0.88 ± 13% perf-profile.calltrace.cycles-pp.apparmor_file_free_security.security_file_free.__fput.task_work_run.exit_to_user_mode_loop 0.74 ± 6% +0.9 1.63 ± 13% perf-profile.calltrace.cycles-pp.security_file_alloc.__alloc_file.alloc_empty_file.path_openat.do_filp_open 0.00 +0.9 0.90 ± 12% perf-profile.calltrace.cycles-pp.security_file_free.__fput.task_work_run.exit_to_user_mode_loop.exit_to_user_mode_prepare 0.00 +1.0 0.96 ± 11% perf-profile.calltrace.cycles-pp.filp_close.__x64_sys_close.do_syscall_64.entry_SYSCALL_64_after_hwframe.__close 0.26 ±100% +1.0 1.24 ± 13% perf-profile.calltrace.cycles-pp.apparmor_file_alloc_security.security_file_alloc.__alloc_file.alloc_empty_file.path_openat 1.50 ± 5% +1.4 2.93 ± 12% perf-profile.calltrace.cycles-pp.__alloc_file.alloc_empty_file.path_openat.do_filp_open.do_sys_openat2 1.54 ± 5% +1.5 2.99 ± 12% perf-profile.calltrace.cycles-pp.alloc_empty_file.path_openat.do_filp_open.do_sys_openat2.__x64_sys_openat 2.55 ± 6% +1.5 4.07 ± 11% perf-profile.calltrace.cycles-pp.apparmor_file_open.security_file_open.do_dentry_open.do_open.path_openat 2.56 ± 6% +1.5 4.09 ± 11% perf-profile.calltrace.cycles-pp.security_file_open.do_dentry_open.do_open.path_openat.do_filp_open 21.63 ± 10% -8.6 12.99 ± 12% perf-profile.children.cycles-pp.lockref_put_or_lock 22.16 ± 9% -8.2 13.92 ± 10% perf-profile.children.cycles-pp.dput 9.86 ± 8% -4.6 5.23 ± 10% perf-profile.children.cycles-pp.lockref_get_not_dead 10.82 ± 10% -4.6 6.25 ± 10% perf-profile.children.cycles-pp.lockref_get 11.02 ± 10% -4.3 6.74 ± 10% perf-profile.children.cycles-pp.terminate_walk 11.86 ± 7% -4.2 7.67 ± 11% perf-profile.children.cycles-pp.__legitimize_path 11.89 ± 7% -4.2 7.73 ± 11% perf-profile.children.cycles-pp.try_to_unlazy 11.91 ± 7% -4.2 7.75 ± 11% perf-profile.children.cycles-pp.complete_walk 12.51 ± 6% -2.2 10.30 ± 11% perf-profile.children.cycles-pp.__fput 0.06 ± 15% +0.0 0.09 ± 15% perf-profile.children.cycles-pp._raw_spin_lock_irq 0.05 ± 8% +0.0 0.09 ± 16% perf-profile.children.cycles-pp.build_open_flags 0.06 ± 13% +0.0 0.09 ± 10% perf-profile.children.cycles-pp.should_failslab 0.05 ± 8% +0.0 0.09 ± 15% perf-profile.children.cycles-pp.path_init 0.06 ± 14% +0.0 0.11 ± 14% perf-profile.children.cycles-pp.__lookup_mnt 0.04 ± 71% +0.0 0.08 ± 19% perf-profile.children.cycles-pp.__might_fault 0.02 ±141% +0.0 0.06 ± 14% perf-profile.children.cycles-pp.rcu_segcblist_enqueue 0.06 ± 11% +0.0 0.11 ± 12% perf-profile.children.cycles-pp.generic_permission 0.09 ± 7% +0.1 0.14 ± 10% perf-profile.children.cycles-pp.walk_component 0.08 ± 13% +0.1 0.13 ± 13% perf-profile.children.cycles-pp.__get_obj_cgroup_from_memcg 0.07 ± 7% +0.1 0.12 ± 12% perf-profile.children.cycles-pp.__check_heap_object 0.08 ± 25% +0.1 0.13 ± 20% perf-profile.children.cycles-pp.rcu_cblist_dequeue 0.02 ±141% +0.1 0.07 ± 10% perf-profile.children.cycles-pp.testcase 0.13 ± 5% +0.1 0.18 ± 9% perf-profile.children.cycles-pp._raw_spin_lock 0.02 ±141% +0.1 0.07 ± 18% perf-profile.children.cycles-pp.obj_cgroup_charge 0.06 ± 7% +0.1 0.12 ± 8% perf-profile.children.cycles-pp.__might_sleep 0.06 ± 13% +0.1 0.12 ± 13% perf-profile.children.cycles-pp.__mnt_want_write 0.10 ± 6% +0.1 0.16 ± 10% perf-profile.children.cycles-pp.close_fd 0.10 ± 7% +0.1 0.17 ± 11% perf-profile.children.cycles-pp.__cond_resched 0.09 ± 9% +0.1 0.16 ± 10% perf-profile.children.cycles-pp.task_work_add 0.09 ± 10% +0.1 0.15 ± 14% perf-profile.children.cycles-pp.memset_erms 0.10 ± 14% +0.1 0.17 ± 10% perf-profile.children.cycles-pp.___slab_alloc 0.10 ± 7% +0.1 0.18 ± 13% perf-profile.children.cycles-pp.mod_objcg_state 0.10 ± 10% +0.1 0.17 ± 12% perf-profile.children.cycles-pp.__virt_addr_valid 0.12 ± 11% +0.1 0.19 ± 12% perf-profile.children.cycles-pp.__slab_free 0.13 ± 11% +0.1 0.21 ± 9% perf-profile.children.cycles-pp.__call_rcu_common 0.14 ± 7% +0.1 0.23 ± 10% perf-profile.children.cycles-pp.fput 0.13 ± 7% +0.1 0.22 ± 16% perf-profile.children.cycles-pp.alloc_fd 0.15 ± 8% +0.1 0.24 ± 12% perf-profile.children.cycles-pp.step_into 0.03 ±100% +0.1 0.12 ± 37% perf-profile.children.cycles-pp.dnotify_flush 0.13 ± 10% +0.1 0.22 ± 14% perf-profile.children.cycles-pp.file_free_rcu 0.14 ± 8% +0.1 0.24 ± 9% perf-profile.children.cycles-pp.get_obj_cgroup_from_current 0.11 ± 14% +0.1 0.21 ± 13% perf-profile.children.cycles-pp.syscall_enter_from_user_mode 0.15 ± 11% +0.1 0.26 ± 11% perf-profile.children.cycles-pp.check_heap_object 0.15 ± 7% +0.1 0.25 ± 11% perf-profile.children.cycles-pp.entry_SYSRETQ_unsafe_stack 0.16 ± 9% +0.1 0.29 ± 9% perf-profile.children.cycles-pp.__might_resched 0.18 ± 7% +0.1 0.31 ± 8% perf-profile.children.cycles-pp.entry_SYSCALL_64_safe_stack 0.16 ± 8% +0.1 0.29 ± 14% perf-profile.children.cycles-pp.inode_permission 0.19 ± 7% +0.1 0.33 ± 14% perf-profile.children.cycles-pp.memcg_slab_post_alloc_hook 0.27 ± 3% +0.2 0.45 ± 11% perf-profile.children.cycles-pp.kmem_cache_free 0.26 ± 8% +0.2 0.45 ± 9% perf-profile.children.cycles-pp.__check_object_size 0.20 ± 25% +0.2 0.43 ± 14% perf-profile.children.cycles-pp.__fsnotify_parent 0.08 ± 19% +0.3 0.40 ± 14% perf-profile.children.cycles-pp.locks_remove_file 0.44 ± 6% +0.3 0.76 ± 11% perf-profile.children.cycles-pp.strncpy_from_user 0.72 ± 5% +0.4 1.07 ± 12% perf-profile.children.cycles-pp.__irq_exit_rcu 0.50 ± 5% +0.4 0.85 ± 11% perf-profile.children.cycles-pp.link_path_walk 0.68 ± 3% +0.4 1.03 ± 12% perf-profile.children.cycles-pp.__do_softirq 0.23 ± 11% +0.4 0.60 ± 13% perf-profile.children.cycles-pp.locks_remove_posix 0.54 ± 7% +0.4 0.92 ± 12% perf-profile.children.cycles-pp.rcu_do_batch 0.55 ± 6% +0.4 0.93 ± 12% perf-profile.children.cycles-pp.rcu_core 0.66 ± 6% +0.5 1.13 ± 11% perf-profile.children.cycles-pp.getname_flags 0.42 ± 6% +0.5 0.90 ± 12% perf-profile.children.cycles-pp.security_file_free 0.39 ± 7% +0.5 0.88 ± 13% perf-profile.children.cycles-pp.apparmor_file_free_security 0.42 ± 7% +0.5 0.97 ± 11% perf-profile.children.cycles-pp.filp_close 0.99 ± 6% +0.6 1.63 ± 10% perf-profile.children.cycles-pp.__entry_text_start 0.56 ± 6% +0.6 1.20 ± 11% perf-profile.children.cycles-pp.__x64_sys_close 2.61 ± 10% +0.7 3.27 ± 12% perf-profile.children.cycles-pp.__d_lookup_rcu 1.00 ± 7% +0.7 1.68 ± 11% perf-profile.children.cycles-pp.kmem_cache_alloc 0.50 ± 5% +0.7 1.25 ± 13% perf-profile.children.cycles-pp.apparmor_file_alloc_security 0.74 ± 6% +0.9 1.63 ± 13% perf-profile.children.cycles-pp.security_file_alloc 2.07 ± 5% +1.4 3.45 ± 11% perf-profile.children.cycles-pp.syscall_return_via_sysret 1.51 ± 5% +1.4 2.94 ± 12% perf-profile.children.cycles-pp.__alloc_file 1.54 ± 5% +1.5 3.00 ± 12% perf-profile.children.cycles-pp.alloc_empty_file 2.55 ± 6% +1.5 4.08 ± 11% perf-profile.children.cycles-pp.apparmor_file_open 2.56 ± 6% +1.5 4.09 ± 11% perf-profile.children.cycles-pp.security_file_open 21.33 ± 10% -8.6 12.71 ± 12% perf-profile.self.cycles-pp.lockref_put_or_lock 9.72 ± 8% -4.6 5.13 ± 10% perf-profile.self.cycles-pp.lockref_get_not_dead 10.67 ± 10% -4.6 6.12 ± 10% perf-profile.self.cycles-pp.lockref_get 0.07 ± 14% +0.0 0.10 ± 4% perf-profile.self.cycles-pp.mntput_no_expire 0.06 ± 16% +0.0 0.09 ± 14% perf-profile.self.cycles-pp._raw_spin_lock_irq 0.05 +0.0 0.08 ± 14% perf-profile.self.cycles-pp.build_open_flags 0.05 ± 8% +0.0 0.09 ± 9% perf-profile.self.cycles-pp.exit_to_user_mode_loop 0.06 ± 11% +0.0 0.10 ± 13% perf-profile.self.cycles-pp.__cond_resched 0.06 ± 14% +0.0 0.10 ± 12% perf-profile.self.cycles-pp.__lookup_mnt 0.03 ± 70% +0.0 0.07 ± 12% perf-profile.self.cycles-pp.fput 0.07 ± 10% +0.0 0.11 ± 14% perf-profile.self.cycles-pp.task_work_run 0.06 ± 6% +0.0 0.10 ± 22% perf-profile.self.cycles-pp.alloc_fd 0.07 ± 14% +0.0 0.12 ± 17% perf-profile.self.cycles-pp.do_syscall_64 0.02 ± 99% +0.0 0.07 ± 8% perf-profile.self.cycles-pp.path_openat 0.05 ± 8% +0.0 0.10 ± 10% perf-profile.self.cycles-pp.generic_permission 0.08 ± 10% +0.0 0.13 ± 13% perf-profile.self.cycles-pp.__get_obj_cgroup_from_memcg 0.07 ± 12% +0.0 0.12 ± 10% perf-profile.self.cycles-pp.syscall_enter_from_user_mode 0.07 ± 7% +0.0 0.12 ± 13% perf-profile.self.cycles-pp.__check_heap_object 0.06 ± 11% +0.1 0.11 ± 6% perf-profile.self.cycles-pp.get_obj_cgroup_from_current 0.06 ± 6% +0.1 0.11 ± 8% perf-profile.self.cycles-pp.__might_sleep 0.04 ± 45% +0.1 0.10 ± 19% perf-profile.self.cycles-pp.__x64_sys_openat 0.12 ± 7% +0.1 0.18 ± 9% perf-profile.self.cycles-pp._raw_spin_lock 0.08 ± 7% +0.1 0.14 ± 10% perf-profile.self.cycles-pp.step_into 0.01 ±223% +0.1 0.06 ± 11% perf-profile.self.cycles-pp.rcu_segcblist_enqueue 0.06 ± 11% +0.1 0.11 ± 14% perf-profile.self.cycles-pp.__mnt_want_write 0.07 ± 26% +0.1 0.13 ± 20% perf-profile.self.cycles-pp.rcu_cblist_dequeue 0.07 ± 11% +0.1 0.13 ± 14% perf-profile.self.cycles-pp.entry_SYSCALL_64_safe_stack 0.02 ±141% +0.1 0.07 ± 10% perf-profile.self.cycles-pp.__x64_sys_close 0.08 ± 8% +0.1 0.14 ± 10% perf-profile.self.cycles-pp.task_work_add 0.08 ± 11% +0.1 0.14 ± 15% perf-profile.self.cycles-pp.memset_erms 0.00 +0.1 0.06 ± 11% perf-profile.self.cycles-pp.__check_object_size 0.10 ± 9% +0.1 0.16 ± 12% perf-profile.self.cycles-pp.mod_objcg_state 0.10 ± 13% +0.1 0.16 ± 12% perf-profile.self.cycles-pp.__virt_addr_valid 0.00 +0.1 0.07 ± 11% perf-profile.self.cycles-pp.getname_flags 0.12 ± 9% +0.1 0.19 ± 12% perf-profile.self.cycles-pp.__slab_free 0.09 ± 10% +0.1 0.18 ± 15% perf-profile.self.cycles-pp.inode_permission 0.03 ±100% +0.1 0.12 ± 34% perf-profile.self.cycles-pp.dnotify_flush 0.13 ± 10% +0.1 0.22 ± 16% perf-profile.self.cycles-pp.__alloc_file 0.13 ± 10% +0.1 0.22 ± 14% perf-profile.self.cycles-pp.file_free_rcu 0.13 ± 8% +0.1 0.22 ± 9% perf-profile.self.cycles-pp.entry_SYSRETQ_unsafe_stack 0.13 ± 10% +0.1 0.24 ± 13% perf-profile.self.cycles-pp.strncpy_from_user 0.14 ± 7% +0.1 0.25 ± 12% perf-profile.self.cycles-pp.memcg_slab_post_alloc_hook 0.16 ± 4% +0.1 0.28 ± 12% perf-profile.self.cycles-pp.kmem_cache_free 0.11 ± 13% +0.1 0.23 ± 11% perf-profile.self.cycles-pp.__close 0.16 ± 9% +0.1 0.28 ± 9% perf-profile.self.cycles-pp.__might_resched 0.15 ± 7% +0.1 0.27 ± 15% perf-profile.self.cycles-pp.link_path_walk 0.14 ± 7% +0.1 0.27 ± 11% perf-profile.self.cycles-pp.open64 0.30 ± 9% +0.2 0.49 ± 12% perf-profile.self.cycles-pp.kmem_cache_alloc 0.20 ± 26% +0.2 0.42 ± 14% perf-profile.self.cycles-pp.__fsnotify_parent 0.08 ± 19% +0.3 0.39 ± 14% perf-profile.self.cycles-pp.locks_remove_file 0.22 ± 12% +0.4 0.58 ± 13% perf-profile.self.cycles-pp.locks_remove_posix 0.38 ± 8% +0.5 0.86 ± 13% perf-profile.self.cycles-pp.apparmor_file_free_security 0.86 ± 5% +0.6 1.42 ± 10% perf-profile.self.cycles-pp.__entry_text_start 2.57 ± 10% +0.6 3.20 ± 12% perf-profile.self.cycles-pp.__d_lookup_rcu 0.46 ± 5% +0.7 1.16 ± 13% perf-profile.self.cycles-pp.apparmor_file_alloc_security 0.58 ± 49% +0.7 1.33 ± 30% perf-profile.self.cycles-pp.__fput 1.16 ± 6% +0.8 1.98 ± 10% perf-profile.self.cycles-pp.entry_SYSCALL_64_after_hwframe 2.29 ± 25% +1.3 3.61 ± 12% perf-profile.self.cycles-pp.syscall_exit_to_user_mode 2.07 ± 5% +1.4 3.44 ± 11% perf-profile.self.cycles-pp.syscall_return_via_sysret 2.51 ± 6% +1.5 4.00 ± 11% perf-profile.self.cycles-pp.apparmor_file_open 2.29 ± 7% +1.7 4.01 ± 12% perf-profile.self.cycles-pp.do_dentry_open 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. -- 0-DAY CI Kernel Test Service https://github.com/intel/lkp-tests