Hello, kernel test robot noticed a 13.0% improvement of hackbench.throughput on: commit: c8d312e039030edab25836a326bcaeb2a3d4db14 ("slub: Delay freezing of partial slabs") https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git master testcase: hackbench test machine: 224 threads 2 sockets Intel(R) Xeon(R) Platinum 8480CTDX (Sapphire Rapids) with 256G memory parameters: nr_threads: 50% iterations: 4 mode: threads ipc: socket cpufreq_governor: performance In addition to that, the commit also has significant impact on the following tests: +------------------+-----------------------------------------------------------------------------------------------+ | testcase: change | hackbench: hackbench.throughput 3.8% improvement | | test machine | 128 threads 2 sockets Intel(R) Xeon(R) Gold 6338 CPU @ 2.00GHz (Ice Lake) with 256G memory | | test parameters | cpufreq_governor=performance | | | ipc=socket | | | iterations=4 | | | mode=process | | | nr_threads=50% | +------------------+-----------------------------------------------------------------------------------------------+ Details are as below: --------------------------------------------------------------------------------------------------> The kernel config and materials to reproduce are available at: https://download.01.org/0day-ci/archive/20231121/202311211510.d06bb4ec-oliver.sang@xxxxxxxxx ========================================================================================= compiler/cpufreq_governor/ipc/iterations/kconfig/mode/nr_threads/rootfs/tbox_group/testcase: gcc-12/performance/socket/4/x86_64-rhel-8.3/threads/50%/debian-11.1-x86_64-20220510.cgz/lkp-spr-r02/hackbench commit: 00b15a19ee ("slub: Introduce freeze_slab()") c8d312e039 ("slub: Delay freezing of partial slabs") 00b15a19ee543f01 c8d312e039030edab25836a326b ---------------- --------------------------- %stddev %change %stddev \ | \ 523889 ± 21% -36.5% 332913 ± 34% numa-meminfo.node1.AnonPages.max 2087 ± 8% +12.8% 2354 ± 7% perf-c2c.DRAM.remote 4331250 ± 7% -16.1% 3631875 ± 7% proc-vmstat.unevictable_pgs_scanned 11778 ± 9% +17.8% 13873 sched_debug.cpu.curr->pid.avg 67435 ± 17% +39.2% 93869 ± 17% sched_debug.cpu.nr_switches.stddev 4695177 ± 3% +19.5% 5609624 vmstat.system.cs 869277 ± 2% +19.4% 1037802 vmstat.system.in 1.04 +0.2 1.20 mpstat.cpu.all.irq% 0.03 ± 4% +0.0 0.04 ± 2% mpstat.cpu.all.soft% 5.13 ± 3% +1.0 6.15 mpstat.cpu.all.usr% 0.17 ± 13% +0.1 0.22 ± 6% turbostat.C1% 0.32 +13.4% 0.37 turbostat.IPC 90.44 ± 2% +14.7 105.11 turbostat.PKG_% 599288 ± 4% +23.3% 739042 ± 6% turbostat.POLL 21.01 +3.7% 21.78 turbostat.RAMWatt 617318 +13.0% 697416 hackbench.throughput 598514 +14.4% 684462 hackbench.throughput_avg 617318 +13.0% 697416 hackbench.throughput_best 572244 +15.1% 658588 hackbench.throughput_worst 176.95 -12.4% 154.95 hackbench.time.elapsed_time 176.95 -12.4% 154.95 hackbench.time.elapsed_time.max 1.656e+08 +2.8% 1.702e+08 hackbench.time.involuntary_context_switches 36017 -14.0% 30972 hackbench.time.system_time 1987 +3.3% 2052 hackbench.time.user_time 6.275e+10 ± 4% +13.2% 7.102e+10 perf-stat.i.branch-instructions 3.082e+08 ± 4% +12.1% 3.453e+08 perf-stat.i.branch-misses 3.22 ± 5% +0.2 3.46 ± 2% perf-stat.i.cache-miss-rate% 76571285 ± 4% +20.7% 92432946 ± 2% perf-stat.i.cache-misses 2.504e+09 ± 4% +14.6% 2.869e+09 perf-stat.i.cache-references 4755923 ± 4% +19.4% 5677973 perf-stat.i.context-switches 1.87 ± 3% -12.9% 1.63 perf-stat.i.cpi 568753 ± 4% +13.6% 646372 perf-stat.i.cpu-migrations 8666 ± 4% -18.8% 7037 ± 2% perf-stat.i.cycles-between-cache-misses 9208258 ± 8% +30.2% 11991181 ± 11% perf-stat.i.dTLB-load-misses 9.064e+10 ± 4% +14.0% 1.033e+11 perf-stat.i.dTLB-loads 5.5e+10 ± 4% +16.3% 6.398e+10 perf-stat.i.dTLB-stores 3.356e+11 ± 4% +13.9% 3.824e+11 perf-stat.i.instructions 0.54 +14.1% 0.62 perf-stat.i.ipc 940.99 ± 4% +14.3% 1075 perf-stat.i.metric.M/sec 8058 ± 5% +16.9% 9418 ± 3% perf-stat.i.minor-faults 13116654 ± 3% +17.3% 15384778 ± 4% perf-stat.i.node-load-misses 8058 ± 5% +16.9% 9418 ± 3% perf-stat.i.page-faults 0.23 ± 2% +5.6% 0.24 ± 2% perf-stat.overall.MPKI 1.83 -11.2% 1.62 perf-stat.overall.cpi 7976 ± 2% -16.0% 6703 ± 2% perf-stat.overall.cycles-between-cache-misses 0.55 +12.6% 0.62 perf-stat.overall.ipc 6.256e+10 ± 3% +12.9% 7.061e+10 perf-stat.ps.branch-instructions 3.071e+08 ± 3% +11.7% 3.431e+08 perf-stat.ps.branch-misses 76645185 ± 3% +20.0% 91985443 ± 2% perf-stat.ps.cache-misses 2.501e+09 ± 3% +14.1% 2.854e+09 perf-stat.ps.cache-references 4734725 ± 3% +19.3% 5647562 perf-stat.ps.context-switches 568804 ± 3% +13.1% 643041 perf-stat.ps.cpu-migrations 9839896 ± 8% +30.4% 12827789 ± 10% perf-stat.ps.dTLB-load-misses 9.038e+10 ± 3% +13.7% 1.027e+11 perf-stat.ps.dTLB-loads 5.484e+10 ± 3% +16.0% 6.362e+10 perf-stat.ps.dTLB-stores 3.347e+11 ± 3% +13.6% 3.802e+11 perf-stat.ps.instructions 7688 ± 5% +14.4% 8792 ± 3% perf-stat.ps.minor-faults 13160460 ± 3% +16.3% 15304482 ± 4% perf-stat.ps.node-load-misses 7688 ± 5% +14.4% 8792 ± 3% perf-stat.ps.page-faults 6.072e+13 -2.0% 5.948e+13 perf-stat.total.instructions 11.63 ± 13% -5.3 6.38 ± 6% perf-profile.calltrace.cycles-pp.__unfreeze_partials.skb_release_data.consume_skb.unix_stream_read_generic.unix_stream_recvmsg 11.28 ± 14% -5.2 6.10 ± 6% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.__unfreeze_partials.skb_release_data.consume_skb 11.39 ± 14% -5.2 6.24 ± 6% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.__unfreeze_partials.skb_release_data.consume_skb.unix_stream_read_generic 16.09 ± 9% -4.4 11.70 ± 2% perf-profile.calltrace.cycles-pp.skb_release_data.consume_skb.unix_stream_read_generic.unix_stream_recvmsg.sock_recvmsg 21.92 -3.8 18.15 ± 2% perf-profile.calltrace.cycles-pp.__alloc_skb.alloc_skb_with_frags.sock_alloc_send_pskb.unix_stream_sendmsg.sock_write_iter 22.11 -3.7 18.37 ± 2% perf-profile.calltrace.cycles-pp.alloc_skb_with_frags.sock_alloc_send_pskb.unix_stream_sendmsg.sock_write_iter.vfs_write 19.14 ± 7% -3.7 15.40 perf-profile.calltrace.cycles-pp.consume_skb.unix_stream_read_generic.unix_stream_recvmsg.sock_recvmsg.sock_read_iter 7.82 ± 13% -3.5 4.32 ± 6% perf-profile.calltrace.cycles-pp.get_partial_node.___slab_alloc.__kmem_cache_alloc_node.__kmalloc_node_track_caller.kmalloc_reserve 7.47 ± 14% -3.4 4.04 ± 6% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.get_partial_node.___slab_alloc.__kmem_cache_alloc_node.__kmalloc_node_track_caller 7.44 ± 14% -3.4 4.02 ± 6% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.get_partial_node.___slab_alloc.__kmem_cache_alloc_node 8.31 ± 13% -3.4 4.96 ± 5% perf-profile.calltrace.cycles-pp.___slab_alloc.__kmem_cache_alloc_node.__kmalloc_node_track_caller.kmalloc_reserve.__alloc_skb 23.82 -3.3 20.51 ± 2% perf-profile.calltrace.cycles-pp.sock_alloc_send_pskb.unix_stream_sendmsg.sock_write_iter.vfs_write.ksys_write 45.82 -3.2 42.59 perf-profile.calltrace.cycles-pp.unix_stream_read_generic.unix_stream_recvmsg.sock_recvmsg.sock_read_iter.vfs_read 46.08 -3.2 42.90 perf-profile.calltrace.cycles-pp.unix_stream_recvmsg.sock_recvmsg.sock_read_iter.vfs_read.ksys_read 46.71 -3.1 43.66 perf-profile.calltrace.cycles-pp.sock_recvmsg.sock_read_iter.vfs_read.ksys_read.do_syscall_64 47.22 -3.0 44.24 perf-profile.calltrace.cycles-pp.sock_read_iter.vfs_read.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe 9.54 ± 12% -2.9 6.60 ± 8% perf-profile.calltrace.cycles-pp.__unfreeze_partials.unix_stream_read_generic.unix_stream_recvmsg.sock_recvmsg.sock_read_iter 9.20 ± 12% -2.9 6.33 ± 8% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.__unfreeze_partials.unix_stream_read_generic.unix_stream_recvmsg 10.74 ± 9% -2.9 7.88 ± 3% perf-profile.calltrace.cycles-pp.__kmem_cache_alloc_node.__kmalloc_node_track_caller.kmalloc_reserve.__alloc_skb.alloc_skb_with_frags 9.31 ± 12% -2.9 6.46 ± 8% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.__unfreeze_partials.unix_stream_read_generic.unix_stream_recvmsg.sock_recvmsg 11.03 ± 9% -2.8 8.21 ± 3% perf-profile.calltrace.cycles-pp.__kmalloc_node_track_caller.kmalloc_reserve.__alloc_skb.alloc_skb_with_frags.sock_alloc_send_pskb 11.30 ± 9% -2.8 8.54 ± 3% perf-profile.calltrace.cycles-pp.kmalloc_reserve.__alloc_skb.alloc_skb_with_frags.sock_alloc_send_pskb.unix_stream_sendmsg 48.67 -2.7 45.95 perf-profile.calltrace.cycles-pp.vfs_read.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_read 49.78 -2.5 47.27 perf-profile.calltrace.cycles-pp.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_read.start_thread 50.67 -2.3 48.33 perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_read.start_thread 50.86 -2.3 48.56 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.__libc_read.start_thread 52.23 -2.1 50.15 perf-profile.calltrace.cycles-pp.__libc_read.start_thread 6.40 ± 11% -2.0 4.45 ± 8% perf-profile.calltrace.cycles-pp.get_partial_node.___slab_alloc.kmem_cache_alloc_node.__alloc_skb.alloc_skb_with_frags 6.07 ± 12% -1.9 4.17 ± 8% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.get_partial_node.___slab_alloc.kmem_cache_alloc_node.__alloc_skb 6.04 ± 12% -1.9 4.15 ± 8% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.get_partial_node.___slab_alloc.kmem_cache_alloc_node 6.86 ± 11% -1.8 5.04 ± 7% perf-profile.calltrace.cycles-pp.___slab_alloc.kmem_cache_alloc_node.__alloc_skb.alloc_skb_with_frags.sock_alloc_send_pskb 35.95 -0.3 35.65 perf-profile.calltrace.cycles-pp.unix_stream_sendmsg.sock_write_iter.vfs_write.ksys_write.do_syscall_64 0.53 +0.1 0.61 perf-profile.calltrace.cycles-pp.mod_objcg_state.kmem_cache_free.unix_stream_read_generic.unix_stream_recvmsg.sock_recvmsg 0.52 +0.1 0.60 perf-profile.calltrace.cycles-pp.mod_objcg_state.__kmem_cache_free.skb_release_data.consume_skb.unix_stream_read_generic 0.54 +0.1 0.63 perf-profile.calltrace.cycles-pp.unix_write_space.sock_wfree.unix_destruct_scm.skb_release_head_state.consume_skb 0.55 +0.1 0.65 perf-profile.calltrace.cycles-pp._copy_from_iter.skb_copy_datagram_from_iter.unix_stream_sendmsg.sock_write_iter.vfs_write 0.65 +0.1 0.75 perf-profile.calltrace.cycles-pp.__check_object_size.skb_copy_datagram_from_iter.unix_stream_sendmsg.sock_write_iter.vfs_write 0.59 ± 6% +0.1 0.70 ± 7% perf-profile.calltrace.cycles-pp.__fdget_pos.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_read 0.58 +0.1 0.70 perf-profile.calltrace.cycles-pp.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_read.start_thread 0.60 +0.1 0.73 perf-profile.calltrace.cycles-pp.__pthread_enable_asynccancel.start_thread 0.52 +0.1 0.66 ± 2% perf-profile.calltrace.cycles-pp.dequeue_task_fair.__schedule.schedule.schedule_timeout.unix_stream_data_wait 0.60 ± 3% +0.2 0.75 ± 3% perf-profile.calltrace.cycles-pp.security_socket_sendmsg.sock_write_iter.vfs_write.ksys_write.do_syscall_64 1.01 +0.2 1.18 perf-profile.calltrace.cycles-pp.__slab_free.unix_stream_read_generic.unix_stream_recvmsg.sock_recvmsg.sock_read_iter 0.45 ± 44% +0.2 0.62 ± 7% perf-profile.calltrace.cycles-pp.__fget_light.__fdget_pos.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe 0.84 +0.2 1.01 perf-profile.calltrace.cycles-pp.memcg_slab_post_alloc_hook.kmem_cache_alloc_node.__alloc_skb.alloc_skb_with_frags.sock_alloc_send_pskb 0.73 +0.2 0.92 perf-profile.calltrace.cycles-pp.enqueue_task_fair.activate_task.ttwu_do_activate.try_to_wake_up.autoremove_wake_function 0.76 +0.2 0.95 perf-profile.calltrace.cycles-pp.activate_task.ttwu_do_activate.try_to_wake_up.autoremove_wake_function.__wake_up_common 1.38 +0.2 1.61 perf-profile.calltrace.cycles-pp.skb_copy_datagram_from_iter.unix_stream_sendmsg.sock_write_iter.vfs_write.ksys_write 0.89 +0.2 1.12 perf-profile.calltrace.cycles-pp.ttwu_do_activate.try_to_wake_up.autoremove_wake_function.__wake_up_common.__wake_up_sync_key 1.72 +0.3 1.98 perf-profile.calltrace.cycles-pp.__entry_text_start.__libc_write.start_thread 1.69 +0.3 1.95 perf-profile.calltrace.cycles-pp.__entry_text_start.__libc_read.start_thread 1.24 +0.3 1.50 perf-profile.calltrace.cycles-pp.memcg_slab_post_alloc_hook.__kmem_cache_alloc_node.__kmalloc_node_track_caller.kmalloc_reserve.__alloc_skb 1.12 +0.3 1.43 ± 2% perf-profile.calltrace.cycles-pp.skb_set_owner_w.sock_alloc_send_pskb.unix_stream_sendmsg.sock_write_iter.vfs_write 1.77 ± 2% +0.4 2.13 ± 3% perf-profile.calltrace.cycles-pp.__slab_free.skb_release_data.consume_skb.unix_stream_read_generic.unix_stream_recvmsg 1.98 +0.4 2.37 perf-profile.calltrace.cycles-pp.__kmem_cache_free.skb_release_data.consume_skb.unix_stream_read_generic.unix_stream_recvmsg 1.64 ± 3% +0.4 2.04 ± 5% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.skb_queue_tail.unix_stream_sendmsg.sock_write_iter.vfs_write 1.60 +0.4 2.00 perf-profile.calltrace.cycles-pp.__schedule.schedule.schedule_timeout.unix_stream_data_wait.unix_stream_read_generic 1.80 +0.4 2.21 ± 2% perf-profile.calltrace.cycles-pp.__pthread_disable_asynccancel.start_thread 1.73 ± 3% +0.4 2.14 ± 5% perf-profile.calltrace.cycles-pp.skb_queue_tail.unix_stream_sendmsg.sock_write_iter.vfs_write.ksys_write 1.64 +0.4 2.05 perf-profile.calltrace.cycles-pp.schedule.schedule_timeout.unix_stream_data_wait.unix_stream_read_generic.unix_stream_recvmsg 2.00 +0.4 2.43 ± 2% perf-profile.calltrace.cycles-pp.check_heap_object.__check_object_size.simple_copy_to_iter.__skb_datagram_iter.skb_copy_datagram_iter 1.74 +0.4 2.18 perf-profile.calltrace.cycles-pp.schedule_timeout.unix_stream_data_wait.unix_stream_read_generic.unix_stream_recvmsg.sock_recvmsg 1.76 +0.4 2.21 ± 2% perf-profile.calltrace.cycles-pp.try_to_wake_up.autoremove_wake_function.__wake_up_common.__wake_up_sync_key.sock_def_readable 1.80 +0.5 2.25 ± 2% perf-profile.calltrace.cycles-pp.autoremove_wake_function.__wake_up_common.__wake_up_sync_key.sock_def_readable.unix_stream_sendmsg 1.88 +0.5 2.35 ± 2% perf-profile.calltrace.cycles-pp.__wake_up_common.__wake_up_sync_key.sock_def_readable.unix_stream_sendmsg.sock_write_iter 2.24 +0.5 2.71 ± 3% perf-profile.calltrace.cycles-pp._copy_to_iter.__skb_datagram_iter.skb_copy_datagram_iter.unix_stream_read_actor.unix_stream_read_generic 2.50 +0.5 3.02 perf-profile.calltrace.cycles-pp.__check_object_size.simple_copy_to_iter.__skb_datagram_iter.skb_copy_datagram_iter.unix_stream_read_actor 0.08 ±223% +0.5 0.60 perf-profile.calltrace.cycles-pp.security_socket_recvmsg.sock_recvmsg.sock_read_iter.vfs_read.ksys_read 2.13 +0.5 2.65 perf-profile.calltrace.cycles-pp.unix_stream_data_wait.unix_stream_read_generic.unix_stream_recvmsg.sock_recvmsg.sock_read_iter 0.00 +0.5 0.52 perf-profile.calltrace.cycles-pp.mutex_lock.unix_stream_read_generic.unix_stream_recvmsg.sock_recvmsg.sock_read_iter 2.62 +0.5 3.15 perf-profile.calltrace.cycles-pp.simple_copy_to_iter.__skb_datagram_iter.skb_copy_datagram_iter.unix_stream_read_actor.unix_stream_read_generic 0.00 +0.5 0.54 perf-profile.calltrace.cycles-pp.pick_next_task_fair.__schedule.schedule.schedule_timeout.unix_stream_data_wait 2.50 +0.5 3.04 ± 2% perf-profile.calltrace.cycles-pp.sock_wfree.unix_destruct_scm.skb_release_head_state.consume_skb.unix_stream_read_generic 0.00 +0.5 0.54 perf-profile.calltrace.cycles-pp.exit_to_user_mode_prepare.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_read 0.00 +0.5 0.55 ± 2% perf-profile.calltrace.cycles-pp.skb_unlink.unix_stream_read_generic.unix_stream_recvmsg.sock_recvmsg.sock_read_iter 0.00 +0.6 0.56 ± 2% perf-profile.calltrace.cycles-pp.enqueue_entity.enqueue_task_fair.activate_task.ttwu_do_activate.try_to_wake_up 2.01 ± 2% +0.6 2.56 ± 4% perf-profile.calltrace.cycles-pp.__get_file_rcu.__fget_light.__fdget_pos.ksys_write.do_syscall_64 2.63 +0.6 3.20 ± 2% perf-profile.calltrace.cycles-pp.unix_destruct_scm.skb_release_head_state.consume_skb.unix_stream_read_generic.unix_stream_recvmsg 0.00 +0.6 0.56 perf-profile.calltrace.cycles-pp.obj_cgroup_charge.__kmem_cache_alloc_node.__kmalloc_node_track_caller.kmalloc_reserve.__alloc_skb 2.70 +0.6 3.28 ± 2% perf-profile.calltrace.cycles-pp.skb_release_head_state.consume_skb.unix_stream_read_generic.unix_stream_recvmsg.sock_recvmsg 2.20 ± 3% +0.6 2.78 ± 5% perf-profile.calltrace.cycles-pp.__fget_light.__fdget_pos.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe 2.27 ± 3% +0.6 2.86 ± 5% perf-profile.calltrace.cycles-pp.__fdget_pos.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_write 0.00 +0.6 0.59 ± 5% perf-profile.calltrace.cycles-pp.aa_sk_perm.security_socket_sendmsg.sock_write_iter.vfs_write.ksys_write 2.17 +0.6 2.77 ± 2% perf-profile.calltrace.cycles-pp.__wake_up_sync_key.sock_def_readable.unix_stream_sendmsg.sock_write_iter.vfs_write 2.63 ± 2% +0.8 3.43 ± 3% perf-profile.calltrace.cycles-pp._raw_spin_lock.unix_stream_sendmsg.sock_write_iter.vfs_write.ksys_write 4.44 +0.8 5.26 perf-profile.calltrace.cycles-pp.kmem_cache_free.unix_stream_read_generic.unix_stream_recvmsg.sock_recvmsg.sock_read_iter 41.38 +1.0 42.34 perf-profile.calltrace.cycles-pp.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_write.start_thread 3.78 ± 2% +1.0 4.79 ± 3% perf-profile.calltrace.cycles-pp.sock_def_readable.unix_stream_sendmsg.sock_write_iter.vfs_write.ksys_write 5.02 +1.0 6.05 ± 2% perf-profile.calltrace.cycles-pp.__skb_datagram_iter.skb_copy_datagram_iter.unix_stream_read_actor.unix_stream_read_generic.unix_stream_recvmsg 5.12 +1.0 6.16 ± 2% perf-profile.calltrace.cycles-pp.skb_copy_datagram_iter.unix_stream_read_actor.unix_stream_read_generic.unix_stream_recvmsg.sock_recvmsg 5.18 +1.1 6.23 ± 2% perf-profile.calltrace.cycles-pp.unix_stream_read_actor.unix_stream_read_generic.unix_stream_recvmsg.sock_recvmsg.sock_read_iter 42.11 +1.1 43.19 perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_write.start_thread 42.31 +1.1 43.42 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.__libc_write.start_thread 43.43 +1.3 44.75 perf-profile.calltrace.cycles-pp.__libc_write.start_thread 34.93 ± 2% -13.0 21.96 ± 6% perf-profile.children.cycles-pp.native_queued_spin_lock_slowpath 36.94 ± 2% -12.7 24.29 ± 5% perf-profile.children.cycles-pp._raw_spin_lock_irqsave 21.20 ± 2% -8.2 12.99 ± 6% perf-profile.children.cycles-pp.__unfreeze_partials 14.30 ± 2% -5.4 8.85 ± 6% perf-profile.children.cycles-pp.get_partial_node 15.17 ± 2% -5.2 10.00 ± 5% perf-profile.children.cycles-pp.___slab_alloc 16.12 ± 9% -4.4 11.74 ± 2% perf-profile.children.cycles-pp.skb_release_data 22.00 -3.8 18.23 ± 2% perf-profile.children.cycles-pp.__alloc_skb 22.14 -3.7 18.40 ± 2% perf-profile.children.cycles-pp.alloc_skb_with_frags 19.18 ± 7% -3.7 15.45 perf-profile.children.cycles-pp.consume_skb 23.86 -3.3 20.56 ± 2% perf-profile.children.cycles-pp.sock_alloc_send_pskb 45.98 -3.2 42.76 perf-profile.children.cycles-pp.unix_stream_read_generic 46.11 -3.2 42.92 perf-profile.children.cycles-pp.unix_stream_recvmsg 46.76 -3.0 43.71 perf-profile.children.cycles-pp.sock_recvmsg 47.25 -3.0 44.26 perf-profile.children.cycles-pp.sock_read_iter 10.86 ± 9% -2.8 8.01 ± 3% perf-profile.children.cycles-pp.__kmem_cache_alloc_node 11.08 ± 9% -2.8 8.26 ± 3% perf-profile.children.cycles-pp.__kmalloc_node_track_caller 11.35 ± 9% -2.8 8.60 ± 3% perf-profile.children.cycles-pp.kmalloc_reserve 48.74 -2.7 46.03 perf-profile.children.cycles-pp.vfs_read 49.84 -2.5 47.33 perf-profile.children.cycles-pp.ksys_read 53.04 -2.0 51.06 perf-profile.children.cycles-pp.__libc_read 93.00 -1.3 91.74 perf-profile.children.cycles-pp.do_syscall_64 93.28 -1.2 92.07 perf-profile.children.cycles-pp.entry_SYSCALL_64_after_hwframe 36.13 -0.3 35.85 perf-profile.children.cycles-pp.unix_stream_sendmsg 0.06 -0.0 0.05 perf-profile.children.cycles-pp.native_irq_return_iret 0.05 +0.0 0.06 perf-profile.children.cycles-pp.__x64_sys_read 0.05 +0.0 0.06 perf-profile.children.cycles-pp.native_sched_clock 0.05 +0.0 0.06 perf-profile.children.cycles-pp.__x64_sys_write 0.06 +0.0 0.07 ± 5% perf-profile.children.cycles-pp.update_min_vruntime 0.13 +0.0 0.14 ± 2% perf-profile.children.cycles-pp.unix_scm_to_skb 0.07 ± 6% +0.0 0.09 ± 8% perf-profile.children.cycles-pp.cgroup_rstat_updated 0.06 +0.0 0.07 ± 6% perf-profile.children.cycles-pp.check_preempt_wakeup_fair 0.11 ± 4% +0.0 0.12 ± 3% perf-profile.children.cycles-pp.maybe_add_creds 0.07 ± 5% +0.0 0.09 ± 8% perf-profile.children.cycles-pp.os_xsave 0.10 ± 4% +0.0 0.12 perf-profile.children.cycles-pp.is_vmalloc_addr 0.07 ± 7% +0.0 0.08 ± 5% perf-profile.children.cycles-pp.cpuacct_charge 0.05 ± 8% +0.0 0.07 perf-profile.children.cycles-pp.put_prev_entity 0.08 ± 4% +0.0 0.10 ± 3% perf-profile.children.cycles-pp.fsnotify_perm 0.32 +0.0 0.34 perf-profile.children.cycles-pp.__list_del_entry_valid_or_report 0.13 ± 3% +0.0 0.15 ± 2% perf-profile.children.cycles-pp.rw_verify_area 0.05 +0.0 0.07 ± 10% perf-profile.children.cycles-pp.available_idle_cpu 0.13 ± 3% +0.0 0.15 ± 3% perf-profile.children.cycles-pp.__mod_memcg_state 0.10 ± 3% +0.0 0.12 ± 3% perf-profile.children.cycles-pp.refill_stock 0.16 ± 3% +0.0 0.18 ± 2% perf-profile.children.cycles-pp.put_cpu_partial 0.15 +0.0 0.17 ± 3% perf-profile.children.cycles-pp.__scm_recv_common 0.12 +0.0 0.14 perf-profile.children.cycles-pp.security_socket_getpeersec_dgram 0.11 ± 3% +0.0 0.13 ± 3% perf-profile.children.cycles-pp.__update_load_avg_cfs_rq 0.07 ± 5% +0.0 0.09 ± 5% perf-profile.children.cycles-pp.update_rq_clock 0.14 ± 2% +0.0 0.16 ± 4% perf-profile.children.cycles-pp.kfree 0.12 ± 4% +0.0 0.15 ± 2% perf-profile.children.cycles-pp.check_stack_object 0.12 +0.0 0.14 ± 3% perf-profile.children.cycles-pp.current_obj_cgroup 0.13 ± 4% +0.0 0.16 ± 3% perf-profile.children.cycles-pp.try_charge_memcg 0.09 ± 4% +0.0 0.11 ± 4% perf-profile.children.cycles-pp.asm_sysvec_reschedule_ipi 0.06 +0.0 0.08 ± 8% perf-profile.children.cycles-pp.cpuidle_enter 0.15 ± 2% +0.0 0.18 ± 2% perf-profile.children.cycles-pp.syscall_exit_to_user_mode_prepare 0.16 ± 2% +0.0 0.18 ± 2% perf-profile.children.cycles-pp.__build_skb_around 0.16 ± 3% +0.0 0.19 ± 6% perf-profile.children.cycles-pp.load_balance 0.12 ± 4% +0.0 0.15 ± 3% perf-profile.children.cycles-pp.__mod_memcg_lruvec_state 0.12 +0.0 0.15 ± 3% perf-profile.children.cycles-pp.__dequeue_entity 0.07 +0.0 0.10 ± 4% perf-profile.children.cycles-pp.select_idle_sibling 0.07 ± 7% +0.0 0.09 ± 10% perf-profile.children.cycles-pp.cpuidle_idle_call 0.06 ± 6% +0.0 0.08 ± 8% perf-profile.children.cycles-pp.cpuidle_enter_state 0.28 +0.0 0.31 ± 3% perf-profile.children.cycles-pp.syscall_return_via_sysret 0.15 ± 3% +0.0 0.18 ± 2% perf-profile.children.cycles-pp.entry_SYSCALL_64_safe_stack 0.12 ± 3% +0.0 0.15 ± 3% perf-profile.children.cycles-pp.prepare_task_switch 0.16 ± 2% +0.0 0.19 perf-profile.children.cycles-pp.__update_load_avg_se 0.21 +0.0 0.24 perf-profile.children.cycles-pp._raw_spin_unlock_irqrestore 0.13 ± 2% +0.0 0.16 ± 2% perf-profile.children.cycles-pp.__switch_to 0.18 ± 2% +0.0 0.21 perf-profile.children.cycles-pp.wait_for_unix_gc 0.16 ± 3% +0.0 0.20 perf-profile.children.cycles-pp.reweight_entity 0.12 ± 5% +0.0 0.16 ± 3% perf-profile.children.cycles-pp.update_rq_clock_task 0.22 +0.0 0.25 perf-profile.children.cycles-pp.memcg_account_kmem 0.14 ± 2% +0.0 0.17 ± 2% perf-profile.children.cycles-pp.finish_task_switch 0.12 ± 4% +0.0 0.16 perf-profile.children.cycles-pp.__switch_to_asm 0.15 ± 3% +0.0 0.18 ± 4% perf-profile.children.cycles-pp.wake_affine 0.24 +0.0 0.27 perf-profile.children.cycles-pp.syscall_enter_from_user_mode 0.13 +0.0 0.17 ± 2% perf-profile.children.cycles-pp.wakeup_preempt 0.20 ± 2% +0.0 0.25 perf-profile.children.cycles-pp.restore_fpregs_from_fpstate 0.20 +0.0 0.24 perf-profile.children.cycles-pp.kmalloc_slab 0.32 +0.0 0.37 perf-profile.children.cycles-pp.__virt_addr_valid 0.25 +0.0 0.29 ± 2% perf-profile.children.cycles-pp.obj_cgroup_uncharge_pages 0.22 ± 2% +0.0 0.27 ± 3% perf-profile.children.cycles-pp.set_next_entity 0.18 ± 2% +0.0 0.23 perf-profile.children.cycles-pp.pick_next_entity 0.10 ± 5% +0.0 0.15 ± 10% perf-profile.children.cycles-pp.secondary_startup_64_no_verify 0.10 ± 5% +0.0 0.15 ± 10% perf-profile.children.cycles-pp.cpu_startup_entry 0.10 ± 5% +0.0 0.15 ± 10% perf-profile.children.cycles-pp.do_idle 0.10 ± 5% +0.0 0.15 ± 10% perf-profile.children.cycles-pp.start_secondary 0.16 ± 3% +0.0 0.20 ± 2% perf-profile.children.cycles-pp.kmalloc_size_roundup 0.28 ± 2% +0.0 0.33 perf-profile.children.cycles-pp.aa_file_perm 0.20 +0.1 0.25 perf-profile.children.cycles-pp.__pick_eevdf 0.00 +0.1 0.05 ± 7% perf-profile.children.cycles-pp.finish_wait 0.29 +0.1 0.34 perf-profile.children.cycles-pp.mutex_unlock 0.29 +0.1 0.34 ± 2% perf-profile.children.cycles-pp.__get_task_ioprio 0.20 ± 2% +0.1 0.25 ± 2% perf-profile.children.cycles-pp.__enqueue_entity 0.00 +0.1 0.06 ± 8% perf-profile.children.cycles-pp.intel_idle 0.32 +0.1 0.38 perf-profile.children.cycles-pp.exit_to_user_mode_loop 0.25 +0.1 0.32 ± 2% perf-profile.children.cycles-pp.select_task_rq_fair 0.50 +0.1 0.57 perf-profile.children.cycles-pp.refill_obj_stock 0.32 +0.1 0.38 perf-profile.children.cycles-pp.switch_fpu_return 0.27 ± 2% +0.1 0.34 ± 2% perf-profile.children.cycles-pp.prepare_to_wait 0.45 ± 3% +0.1 0.52 perf-profile.children.cycles-pp.__check_heap_object 0.38 +0.1 0.45 perf-profile.children.cycles-pp.rcu_all_qs 0.28 +0.1 0.36 ± 2% perf-profile.children.cycles-pp.select_task_rq 0.31 +0.1 0.38 ± 2% perf-profile.children.cycles-pp.dequeue_entity 0.47 +0.1 0.56 perf-profile.children.cycles-pp.mutex_lock 0.36 +0.1 0.45 ± 3% perf-profile.children.cycles-pp.update_curr 0.22 ± 5% +0.1 0.31 ± 6% perf-profile.children.cycles-pp.task_tick_fair 0.56 +0.1 0.66 perf-profile.children.cycles-pp._copy_from_iter 0.48 +0.1 0.57 perf-profile.children.cycles-pp.skb_unlink 0.55 +0.1 0.65 perf-profile.children.cycles-pp.unix_write_space 0.25 ± 5% +0.1 0.35 ± 6% perf-profile.children.cycles-pp.scheduler_tick 0.08 ± 5% +0.1 0.18 ± 2% perf-profile.children.cycles-pp.__list_add_valid_or_report 0.26 ± 4% +0.1 0.36 ± 4% perf-profile.children.cycles-pp.update_cfs_group 0.28 ± 4% +0.1 0.38 ± 5% perf-profile.children.cycles-pp.update_process_times 0.28 ± 4% +0.1 0.39 ± 6% perf-profile.children.cycles-pp.tick_sched_handle 0.53 +0.1 0.64 perf-profile.children.cycles-pp.update_load_avg 0.51 ± 2% +0.1 0.62 perf-profile.children.cycles-pp.security_socket_recvmsg 0.32 ± 4% +0.1 0.44 ± 5% perf-profile.children.cycles-pp.__hrtimer_run_queues 0.29 ± 4% +0.1 0.40 ± 6% perf-profile.children.cycles-pp.tick_nohz_highres_handler 0.35 ± 3% +0.1 0.47 ± 5% perf-profile.children.cycles-pp.hrtimer_interrupt 0.36 ± 3% +0.1 0.48 ± 5% perf-profile.children.cycles-pp.__sysvec_apic_timer_interrupt 0.43 ± 3% +0.1 0.55 ± 5% perf-profile.children.cycles-pp.asm_sysvec_apic_timer_interrupt 0.40 ± 3% +0.1 0.52 ± 5% perf-profile.children.cycles-pp.sysvec_apic_timer_interrupt 0.70 ± 4% +0.1 0.82 ± 3% perf-profile.children.cycles-pp.apparmor_file_permission 0.72 +0.1 0.85 perf-profile.children.cycles-pp.__cond_resched 0.52 +0.1 0.65 ± 2% perf-profile.children.cycles-pp.enqueue_entity 0.62 +0.1 0.76 perf-profile.children.cycles-pp.__pthread_enable_asynccancel 0.83 +0.1 0.98 perf-profile.children.cycles-pp.entry_SYSRETQ_unsafe_stack 0.83 +0.2 0.98 perf-profile.children.cycles-pp.obj_cgroup_charge 0.81 +0.2 0.96 perf-profile.children.cycles-pp.exit_to_user_mode_prepare 0.62 ± 3% +0.2 0.78 ± 3% perf-profile.children.cycles-pp.security_socket_sendmsg 0.62 +0.2 0.77 perf-profile.children.cycles-pp.dequeue_task_fair 0.91 ± 3% +0.2 1.06 ± 2% perf-profile.children.cycles-pp.security_file_permission 0.73 +0.2 0.89 perf-profile.children.cycles-pp.pick_next_task_fair 0.60 +0.2 0.77 perf-profile.children.cycles-pp.fput 1.12 +0.2 1.32 perf-profile.children.cycles-pp.syscall_exit_to_user_mode 0.85 +0.2 1.06 perf-profile.children.cycles-pp.enqueue_task_fair 0.84 ± 3% +0.2 1.05 ± 2% perf-profile.children.cycles-pp.aa_sk_perm 0.88 +0.2 1.10 perf-profile.children.cycles-pp.activate_task 1.43 +0.2 1.66 perf-profile.children.cycles-pp.skb_copy_datagram_from_iter 1.00 +0.3 1.25 perf-profile.children.cycles-pp.ttwu_do_activate 1.77 +0.3 2.06 perf-profile.children.cycles-pp.mod_objcg_state 1.93 +0.3 2.23 perf-profile.children.cycles-pp.__entry_text_start 1.14 +0.3 1.46 ± 2% perf-profile.children.cycles-pp.skb_set_owner_w 2.01 +0.4 2.41 perf-profile.children.cycles-pp.__kmem_cache_free 1.82 +0.4 2.23 ± 2% perf-profile.children.cycles-pp.__pthread_disable_asynccancel 1.76 ± 3% +0.4 2.17 ± 5% perf-profile.children.cycles-pp.skb_queue_tail 2.13 +0.4 2.57 perf-profile.children.cycles-pp.memcg_slab_post_alloc_hook 2.26 +0.5 2.73 ± 3% perf-profile.children.cycles-pp._copy_to_iter 1.98 +0.5 2.46 perf-profile.children.cycles-pp.schedule_timeout 1.94 +0.5 2.43 ± 2% perf-profile.children.cycles-pp.try_to_wake_up 1.98 +0.5 2.47 ± 2% perf-profile.children.cycles-pp.autoremove_wake_function 2.44 +0.5 2.94 perf-profile.children.cycles-pp.check_heap_object 2.06 +0.5 2.58 ± 2% perf-profile.children.cycles-pp.__wake_up_common 2.15 +0.5 2.68 perf-profile.children.cycles-pp.unix_stream_data_wait 2.16 +0.5 2.69 perf-profile.children.cycles-pp.schedule 2.65 +0.5 3.18 perf-profile.children.cycles-pp.simple_copy_to_iter 2.80 +0.5 3.34 ± 2% perf-profile.children.cycles-pp.__slab_free 2.17 +0.5 2.71 perf-profile.children.cycles-pp.__schedule 2.54 +0.6 3.09 ± 2% perf-profile.children.cycles-pp.sock_wfree 2.66 +0.6 3.23 ± 2% perf-profile.children.cycles-pp.unix_destruct_scm 2.73 +0.6 3.30 ± 2% perf-profile.children.cycles-pp.skb_release_head_state 3.34 +0.6 3.98 perf-profile.children.cycles-pp.__check_object_size 2.36 +0.6 3.00 ± 2% perf-profile.children.cycles-pp.__wake_up_sync_key 2.40 ± 2% +0.6 3.04 ± 4% perf-profile.children.cycles-pp.__get_file_rcu 2.78 ± 3% +0.7 3.46 ± 4% perf-profile.children.cycles-pp.__fget_light 2.91 ± 3% +0.7 3.62 ± 4% perf-profile.children.cycles-pp.__fdget_pos 4.48 +0.8 5.30 perf-profile.children.cycles-pp.kmem_cache_free 41.46 +1.0 42.42 perf-profile.children.cycles-pp.ksys_write 3.46 +1.0 4.44 ± 3% perf-profile.children.cycles-pp._raw_spin_lock 3.80 ± 2% +1.0 4.82 ± 3% perf-profile.children.cycles-pp.sock_def_readable 5.05 +1.0 6.08 ± 2% perf-profile.children.cycles-pp.__skb_datagram_iter 5.14 +1.0 6.19 ± 2% perf-profile.children.cycles-pp.skb_copy_datagram_iter 5.20 +1.1 6.26 ± 2% perf-profile.children.cycles-pp.unix_stream_read_actor 44.28 +1.4 45.71 perf-profile.children.cycles-pp.__libc_write 34.92 ± 2% -13.0 21.95 ± 6% perf-profile.self.cycles-pp.native_queued_spin_lock_slowpath 0.37 -0.3 0.09 perf-profile.self.cycles-pp.__unfreeze_partials 0.38 -0.1 0.28 ± 2% perf-profile.self.cycles-pp.get_partial_node 0.06 -0.0 0.05 perf-profile.self.cycles-pp.native_irq_return_iret 0.05 +0.0 0.06 perf-profile.self.cycles-pp.unix_stream_read_actor 0.05 +0.0 0.06 perf-profile.self.cycles-pp.skb_free_head 0.07 +0.0 0.08 perf-profile.self.cycles-pp.put_pid 0.08 +0.0 0.09 perf-profile.self.cycles-pp.is_vmalloc_addr 0.11 ± 3% +0.0 0.12 perf-profile.self.cycles-pp.unix_scm_to_skb 0.05 +0.0 0.06 ± 6% perf-profile.self.cycles-pp.___perf_sw_event 0.10 +0.0 0.11 ± 4% perf-profile.self.cycles-pp.skb_queue_tail 0.07 +0.0 0.08 ± 5% perf-profile.self.cycles-pp.fsnotify_perm 0.05 +0.0 0.06 ± 7% perf-profile.self.cycles-pp.enqueue_entity 0.08 ± 6% +0.0 0.09 perf-profile.self.cycles-pp.refill_stock 0.10 ± 3% +0.0 0.11 ± 4% perf-profile.self.cycles-pp.check_stack_object 0.10 ± 4% +0.0 0.12 perf-profile.self.cycles-pp.rw_verify_area 0.10 ± 4% +0.0 0.11 ± 3% perf-profile.self.cycles-pp.skb_unlink 0.07 +0.0 0.08 ± 5% perf-profile.self.cycles-pp.prepare_task_switch 0.07 +0.0 0.08 ± 5% perf-profile.self.cycles-pp.os_xsave 0.06 ± 9% +0.0 0.07 perf-profile.self.cycles-pp.update_min_vruntime 0.12 +0.0 0.14 ± 3% perf-profile.self.cycles-pp.__scm_recv_common 0.32 +0.0 0.33 perf-profile.self.cycles-pp.__list_del_entry_valid_or_report 0.13 +0.0 0.15 ± 3% perf-profile.self.cycles-pp.__kmalloc_node_track_caller 0.09 +0.0 0.11 ± 4% perf-profile.self.cycles-pp.memcg_account_kmem 0.09 +0.0 0.11 ± 4% perf-profile.self.cycles-pp.security_socket_getpeersec_dgram 0.06 ± 7% +0.0 0.08 ± 4% perf-profile.self.cycles-pp.kmalloc_size_roundup 0.06 ± 7% +0.0 0.08 ± 4% perf-profile.self.cycles-pp.cpuacct_charge 0.08 ± 4% +0.0 0.10 ± 3% perf-profile.self.cycles-pp.maybe_add_creds 0.08 ± 4% +0.0 0.10 ± 5% perf-profile.self.cycles-pp.prepare_to_wait 0.11 ± 3% +0.0 0.13 ± 3% perf-profile.self.cycles-pp.__update_load_avg_cfs_rq 0.09 ± 4% +0.0 0.11 ± 4% perf-profile.self.cycles-pp.__mod_memcg_state 0.14 +0.0 0.16 perf-profile.self.cycles-pp.kmalloc_reserve 0.09 +0.0 0.11 perf-profile.self.cycles-pp.current_obj_cgroup 0.08 ± 5% +0.0 0.10 ± 4% perf-profile.self.cycles-pp.reweight_entity 0.11 ± 4% +0.0 0.13 ± 3% perf-profile.self.cycles-pp.syscall_exit_to_user_mode_prepare 0.08 +0.0 0.10 perf-profile.self.cycles-pp.pick_next_task_fair 0.16 +0.0 0.18 ± 2% perf-profile.self.cycles-pp.skb_copy_datagram_from_iter 0.14 ± 3% +0.0 0.16 ± 3% perf-profile.self.cycles-pp.__fdget_pos 0.10 ± 5% +0.0 0.12 ± 4% perf-profile.self.cycles-pp.try_charge_memcg 0.15 +0.0 0.17 ± 2% perf-profile.self.cycles-pp.put_cpu_partial 0.10 ± 3% +0.0 0.12 ± 3% perf-profile.self.cycles-pp.unix_destruct_scm 0.10 ± 4% +0.0 0.12 ± 4% perf-profile.self.cycles-pp.security_socket_sendmsg 0.09 ± 4% +0.0 0.11 perf-profile.self.cycles-pp.__dequeue_entity 0.06 +0.0 0.08 ± 4% perf-profile.self.cycles-pp.dequeue_task_fair 0.18 ± 2% +0.0 0.20 perf-profile.self.cycles-pp.syscall_exit_to_user_mode 0.13 +0.0 0.15 ± 3% perf-profile.self.cycles-pp.__build_skb_around 0.10 ± 4% +0.0 0.12 perf-profile.self.cycles-pp.finish_task_switch 0.09 ± 5% +0.0 0.12 ± 4% perf-profile.self.cycles-pp.__mod_memcg_lruvec_state 0.12 ± 3% +0.0 0.15 ± 3% perf-profile.self.cycles-pp.security_socket_recvmsg 0.12 ± 3% +0.0 0.14 ± 3% perf-profile.self.cycles-pp.kfree 0.08 ± 5% +0.0 0.11 ± 4% perf-profile.self.cycles-pp.__wake_up_common 0.16 ± 2% +0.0 0.19 perf-profile.self.cycles-pp.__skb_datagram_iter 0.14 ± 3% +0.0 0.17 ± 2% perf-profile.self.cycles-pp.alloc_skb_with_frags 0.14 ± 3% +0.0 0.16 ± 5% perf-profile.self.cycles-pp.unix_stream_recvmsg 0.09 ± 5% +0.0 0.12 ± 3% perf-profile.self.cycles-pp.enqueue_task_fair 0.11 ± 4% +0.0 0.13 ± 2% perf-profile.self.cycles-pp.schedule_timeout 0.15 +0.0 0.18 ± 2% perf-profile.self.cycles-pp.exit_to_user_mode_prepare 0.27 +0.0 0.30 ± 3% perf-profile.self.cycles-pp.syscall_return_via_sysret 0.13 +0.0 0.16 ± 4% perf-profile.self.cycles-pp.__switch_to 0.10 ± 3% +0.0 0.13 ± 2% perf-profile.self.cycles-pp.switch_fpu_return 0.16 ± 2% +0.0 0.19 ± 3% perf-profile.self.cycles-pp.sock_recvmsg 0.15 ± 3% +0.0 0.18 ± 2% perf-profile.self.cycles-pp.entry_SYSCALL_64_safe_stack 0.14 ± 2% +0.0 0.17 perf-profile.self.cycles-pp.__update_load_avg_se 0.20 +0.0 0.23 ± 2% perf-profile.self.cycles-pp.syscall_enter_from_user_mode 0.12 ± 4% +0.0 0.14 ± 3% perf-profile.self.cycles-pp.try_to_wake_up 0.03 ± 70% +0.0 0.06 ± 7% perf-profile.self.cycles-pp.available_idle_cpu 0.18 ± 2% +0.0 0.21 perf-profile.self.cycles-pp._raw_spin_unlock_irqrestore 0.20 +0.0 0.23 perf-profile.self.cycles-pp.ksys_read 0.11 ± 6% +0.0 0.14 ± 3% perf-profile.self.cycles-pp.update_rq_clock_task 0.20 +0.0 0.23 ± 2% perf-profile.self.cycles-pp.ksys_write 0.12 ± 3% +0.0 0.16 ± 2% perf-profile.self.cycles-pp.__switch_to_asm 0.23 +0.0 0.27 ± 2% perf-profile.self.cycles-pp.security_file_permission 0.20 ± 2% +0.0 0.24 ± 2% perf-profile.self.cycles-pp.restore_fpregs_from_fpstate 0.30 +0.0 0.34 ± 2% perf-profile.self.cycles-pp.__virt_addr_valid 0.17 +0.0 0.21 perf-profile.self.cycles-pp.kmalloc_slab 0.18 ± 2% +0.0 0.22 perf-profile.self.cycles-pp.__pick_eevdf 0.23 ± 2% +0.0 0.28 perf-profile.self.cycles-pp.rcu_all_qs 0.25 ± 2% +0.0 0.29 perf-profile.self.cycles-pp.aa_file_perm 0.30 +0.0 0.34 perf-profile.self.cycles-pp.entry_SYSCALL_64_after_hwframe 0.25 +0.0 0.30 ± 2% perf-profile.self.cycles-pp.__get_task_ioprio 0.17 ± 2% +0.0 0.22 ± 3% perf-profile.self.cycles-pp.update_curr 0.28 +0.0 0.33 perf-profile.self.cycles-pp.mutex_unlock 0.36 +0.1 0.41 perf-profile.self.cycles-pp.do_syscall_64 0.19 ± 2% +0.1 0.25 perf-profile.self.cycles-pp.__enqueue_entity 0.33 ± 2% +0.1 0.39 ± 2% perf-profile.self.cycles-pp.consume_skb 0.00 +0.1 0.06 ± 8% perf-profile.self.cycles-pp.intel_idle 0.28 ± 3% +0.1 0.34 ± 2% perf-profile.self.cycles-pp.sock_alloc_send_pskb 0.34 +0.1 0.40 perf-profile.self.cycles-pp.unix_write_space 0.37 +0.1 0.44 perf-profile.self.cycles-pp.__check_object_size 0.31 +0.1 0.38 perf-profile.self.cycles-pp.mutex_lock 0.26 +0.1 0.32 perf-profile.self.cycles-pp.update_load_avg 0.47 +0.1 0.54 perf-profile.self.cycles-pp.refill_obj_stock 0.48 ± 2% +0.1 0.55 ± 2% perf-profile.self.cycles-pp.sock_read_iter 0.40 ± 6% +0.1 0.46 ± 5% perf-profile.self.cycles-pp.apparmor_file_permission 0.42 ± 3% +0.1 0.48 perf-profile.self.cycles-pp.__check_heap_object 0.37 +0.1 0.44 perf-profile.self.cycles-pp.__cond_resched 0.45 +0.1 0.53 perf-profile.self.cycles-pp.__entry_text_start 0.44 +0.1 0.52 ± 3% perf-profile.self.cycles-pp.skb_release_data 0.49 +0.1 0.58 perf-profile.self.cycles-pp.obj_cgroup_charge 0.31 ± 3% +0.1 0.41 ± 5% perf-profile.self.cycles-pp.__schedule 0.54 +0.1 0.63 perf-profile.self.cycles-pp._copy_from_iter 0.08 +0.1 0.18 ± 2% perf-profile.self.cycles-pp.__list_add_valid_or_report 0.26 ± 3% +0.1 0.36 ± 5% perf-profile.self.cycles-pp.update_cfs_group 0.54 +0.1 0.64 perf-profile.self.cycles-pp.sock_write_iter 0.66 +0.1 0.77 perf-profile.self.cycles-pp.__kmem_cache_alloc_node 0.60 +0.1 0.72 perf-profile.self.cycles-pp.__pthread_enable_asynccancel 0.62 +0.1 0.75 perf-profile.self.cycles-pp.kmem_cache_alloc_node 0.97 +0.1 1.11 perf-profile.self.cycles-pp.__libc_read 0.81 +0.1 0.95 perf-profile.self.cycles-pp.entry_SYSRETQ_unsafe_stack 0.80 +0.1 0.94 perf-profile.self.cycles-pp.vfs_read 0.99 +0.2 1.15 perf-profile.self.cycles-pp.__libc_write 0.57 +0.2 0.73 perf-profile.self.cycles-pp.fput 0.76 +0.2 0.92 perf-profile.self.cycles-pp.vfs_write 0.62 ± 4% +0.2 0.80 ± 3% perf-profile.self.cycles-pp.aa_sk_perm 1.10 +0.2 1.34 perf-profile.self.cycles-pp.__kmem_cache_free 1.60 +0.3 1.85 perf-profile.self.cycles-pp.mod_objcg_state 0.86 +0.3 1.14 perf-profile.self.cycles-pp.___slab_alloc 1.12 +0.3 1.43 perf-profile.self.cycles-pp.skb_set_owner_w 1.38 +0.3 1.70 perf-profile.self.cycles-pp.memcg_slab_post_alloc_hook 1.67 +0.4 2.04 ± 2% perf-profile.self.cycles-pp.__alloc_skb 1.59 ± 3% +0.4 1.99 ± 6% perf-profile.self.cycles-pp.sock_def_readable 1.73 +0.4 2.14 ± 2% perf-profile.self.cycles-pp.__pthread_disable_asynccancel 2.00 +0.4 2.43 ± 2% perf-profile.self.cycles-pp.check_heap_object 1.98 ± 2% +0.5 2.43 ± 2% perf-profile.self.cycles-pp.sock_wfree 2.02 ± 2% +0.5 2.49 ± 5% perf-profile.self.cycles-pp.unix_stream_sendmsg 2.22 +0.5 2.70 ± 3% perf-profile.self.cycles-pp._copy_to_iter 2.76 +0.5 3.29 ± 2% perf-profile.self.cycles-pp.__slab_free 2.58 ± 2% +0.6 3.13 ± 4% perf-profile.self.cycles-pp._raw_spin_lock_irqsave 2.60 +0.6 3.18 ± 3% perf-profile.self.cycles-pp.unix_stream_read_generic 2.36 ± 2% +0.6 2.99 ± 4% perf-profile.self.cycles-pp.__get_file_rcu 3.61 +0.7 4.30 ± 2% perf-profile.self.cycles-pp.kmem_cache_free 2.82 +0.7 3.55 ± 3% perf-profile.self.cycles-pp._raw_spin_lock *************************************************************************************************** lkp-icl-2sp2: 128 threads 2 sockets Intel(R) Xeon(R) Gold 6338 CPU @ 2.00GHz (Ice Lake) with 256G memory ========================================================================================= compiler/cpufreq_governor/ipc/iterations/kconfig/mode/nr_threads/rootfs/tbox_group/testcase: gcc-12/performance/socket/4/x86_64-rhel-8.3/process/50%/debian-11.1-x86_64-20220510.cgz/lkp-icl-2sp2/hackbench commit: 00b15a19ee ("slub: Introduce freeze_slab()") c8d312e039 ("slub: Delay freezing of partial slabs") 00b15a19ee543f01 c8d312e039030edab25836a326b ---------------- --------------------------- %stddev %change %stddev \ | \ 1492 ± 9% +71.4% 2558 ± 29% turbostat.POLL 5914740 ± 7% +17.5% 6949608 ± 8% vmstat.memory.cache 448744 +3.8% 465846 hackbench.throughput 441391 +3.8% 457980 hackbench.throughput_avg 448744 +3.8% 465846 hackbench.throughput_best 427227 +4.7% 447449 hackbench.throughput_worst 83156624 -1.8% 81700614 hackbench.time.involuntary_context_switches 16199 -4.5% 15469 hackbench.time.system_time 0.54 ± 4% -0.0 0.51 perf-stat.i.branch-miss-rate% 2280 ± 44% -22.9% 1758 ± 2% perf-stat.i.cycles-between-cache-misses 0.53 -0.0 0.51 perf-stat.overall.branch-miss-rate% 1.36 -3.9% 1.30 perf-stat.overall.cpi 1822 -3.5% 1758 perf-stat.overall.cycles-between-cache-misses 0.74 +4.1% 0.77 perf-stat.overall.ipc 5777253 ± 7% +17.8% 6802747 ± 8% meminfo.Cached 4716622 ± 7% +22.9% 5794899 ± 10% meminfo.Committed_AS 3306409 ± 14% +38.3% 4572192 ± 13% meminfo.Inactive 3306021 ± 14% +38.3% 4571804 ± 13% meminfo.Inactive(anon) 8940810 ± 4% +12.4% 10051649 ± 5% meminfo.Memused 3088924 ± 13% +33.2% 4114417 ± 14% meminfo.Shmem 9221505 ± 4% +11.3% 10259268 ± 5% meminfo.max_used_kB 1444106 ± 7% +17.8% 1701674 ± 8% proc-vmstat.nr_file_pages 825202 ± 14% +38.6% 1143844 ± 13% proc-vmstat.nr_inactive_anon 772022 ± 13% +33.4% 1029591 ± 14% proc-vmstat.nr_shmem 33082 +1.6% 33620 proc-vmstat.nr_slab_reclaimable 825202 ± 14% +38.6% 1143844 ± 13% proc-vmstat.nr_zone_inactive_anon 2333093 ± 5% +15.4% 2693344 ± 7% proc-vmstat.numa_hit 2199304 ± 5% +16.2% 2554985 ± 7% proc-vmstat.numa_local 3312352 ± 3% +9.3% 3621949 ± 5% proc-vmstat.pgalloc_normal 2086440 ± 5% -8.4% 1911146 proc-vmstat.pgfree 10.82 ±216% -99.9% 0.01 ±223% perf-sched.sch_delay.avg.ms.__cond_resched.__kmem_cache_alloc_node.__kmalloc_node.memcg_alloc_slab_cgroups.memcg_slab_post_alloc_hook 1.29 ± 8% -89.8% 0.13 ±223% perf-sched.sch_delay.avg.ms.__cond_resched.__kmem_cache_alloc_node.__kmalloc_node_track_caller.kmalloc_reserve.__alloc_skb 1.46 ± 9% -83.0% 0.25 ±223% perf-sched.sch_delay.avg.ms.__cond_resched.aa_sk_perm.security_socket_recvmsg.sock_recvmsg.sock_read_iter 1.88 ±113% -98.6% 0.03 ±223% perf-sched.sch_delay.avg.ms.__cond_resched.exit_mmap.__mmput.exit_mm.do_exit 0.82 ± 16% -87.3% 0.10 ±223% perf-sched.sch_delay.avg.ms.__cond_resched.kmem_cache_alloc_node.__alloc_skb.alloc_skb_with_frags.sock_alloc_send_pskb 0.97 ± 35% -85.2% 0.14 ±223% perf-sched.sch_delay.avg.ms.__cond_resched.wait_for_unix_gc.unix_stream_sendmsg.sock_write_iter.vfs_write 1.29 ±130% -99.7% 0.00 ±223% perf-sched.sch_delay.avg.ms.__cond_resched.zap_pmd_range.isra.0.unmap_page_range 13.65 ± 64% -97.0% 0.41 ±223% perf-sched.sch_delay.avg.ms.do_wait.kernel_wait4.__do_sys_wait4.do_syscall_64 4.64 ± 28% -83.9% 0.75 ±223% perf-sched.sch_delay.avg.ms.exit_to_user_mode_loop.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_sysvec_apic_timer_interrupt 6.59 ± 12% -87.2% 0.84 ±223% perf-sched.sch_delay.avg.ms.exit_to_user_mode_loop.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_sysvec_reschedule_ipi 0.94 ± 18% -86.3% 0.13 ±223% perf-sched.sch_delay.avg.ms.exit_to_user_mode_loop.exit_to_user_mode_prepare.syscall_exit_to_user_mode.do_syscall_64 32.34 ± 99% -93.9% 1.99 ±223% perf-sched.sch_delay.avg.ms.pipe_read.vfs_read.ksys_read.do_syscall_64 14.51 ± 84% -96.9% 0.45 ±223% perf-sched.sch_delay.avg.ms.rcu_gp_kthread.kthread.ret_from_fork.ret_from_fork_asm 3.08 ± 37% -96.5% 0.11 ±223% perf-sched.sch_delay.avg.ms.schedule_hrtimeout_range_clock.do_select.core_sys_select.kern_select 0.48 ±122% -89.9% 0.05 ±223% perf-sched.sch_delay.avg.ms.schedule_hrtimeout_range_clock.ep_poll.do_epoll_wait.__x64_sys_epoll_wait 5.98 ±135% -99.4% 0.04 ±223% perf-sched.sch_delay.avg.ms.schedule_preempt_disabled.rwsem_down_write_slowpath.down_write.unlink_anon_vmas 24.69 ±149% -99.3% 0.16 ±223% perf-sched.sch_delay.avg.ms.schedule_timeout.__wait_for_common.wait_for_completion_state.kernel_clone 6.64 ± 29% -86.6% 0.89 ±223% perf-sched.sch_delay.avg.ms.schedule_timeout.rcu_gp_fqs_loop.rcu_gp_kthread.kthread 7.86 ± 3% -84.7% 1.20 ±223% perf-sched.sch_delay.avg.ms.schedule_timeout.sock_alloc_send_pskb.unix_stream_sendmsg.sock_write_iter 1.04 ± 11% -86.4% 0.14 ±223% perf-sched.sch_delay.avg.ms.schedule_timeout.unix_stream_data_wait.unix_stream_read_generic.unix_stream_recvmsg 2.99 ± 91% -99.9% 0.00 ±223% perf-sched.sch_delay.avg.ms.syslog_print.do_syslog.kmsg_read.vfs_read 0.45 ± 78% -91.8% 0.04 ±223% perf-sched.sch_delay.avg.ms.wait_for_partner.fifo_open.do_dentry_open.do_open 14.68 ± 78% -99.6% 0.07 ±223% perf-sched.sch_delay.avg.ms.worker_thread.kthread.ret_from_fork.ret_from_fork_asm 87.77 ±213% -99.7% 0.23 ±223% perf-sched.sch_delay.max.ms.__cond_resched.__kmem_cache_alloc_node.__kmalloc_node.memcg_alloc_slab_cgroups.memcg_slab_post_alloc_hook 0.34 ±101% -100.0% 0.00 perf-sched.sch_delay.max.ms.__cond_resched.__wait_for_common.affine_move_task.__set_cpus_allowed_ptr.__sched_setaffinity 39.05 ±129% -98.8% 0.48 ±223% perf-sched.sch_delay.max.ms.__cond_resched.exit_mmap.__mmput.exit_mm.do_exit 434.35 ± 30% -90.7% 40.29 ±223% perf-sched.sch_delay.max.ms.__cond_resched.kmem_cache_alloc_node.__alloc_skb.alloc_skb_with_frags.sock_alloc_send_pskb 302.77 ± 53% -85.7% 43.20 ±223% perf-sched.sch_delay.max.ms.__cond_resched.mutex_lock.unix_stream_read_generic.unix_stream_recvmsg.sock_recvmsg 47.44 ±153% -99.6% 0.19 ±223% perf-sched.sch_delay.max.ms.__cond_resched.zap_pmd_range.isra.0.unmap_page_range 844.60 ± 68% -94.1% 49.49 ±223% perf-sched.sch_delay.max.ms.do_wait.kernel_wait4.__do_sys_wait4.do_syscall_64 542.84 ± 41% -87.6% 67.21 ±223% perf-sched.sch_delay.max.ms.exit_to_user_mode_loop.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_sysvec_reschedule_ipi 2386 ± 52% -92.8% 171.54 ±223% perf-sched.sch_delay.max.ms.pipe_read.vfs_read.ksys_read.do_syscall_64 234.85 ±122% -94.3% 13.50 ±223% perf-sched.sch_delay.max.ms.rcu_gp_kthread.kthread.ret_from_fork.ret_from_fork_asm 5.27 ± 40% -89.0% 0.58 ±223% perf-sched.sch_delay.max.ms.schedule_hrtimeout_range_clock.do_select.core_sys_select.kern_select 167.84 ±134% -99.4% 1.01 ±223% perf-sched.sch_delay.max.ms.schedule_preempt_disabled.rwsem_down_write_slowpath.down_write.unlink_anon_vmas 718.40 ± 10% -86.9% 94.02 ±223% perf-sched.sch_delay.max.ms.schedule_timeout.unix_stream_data_wait.unix_stream_read_generic.unix_stream_recvmsg 0.75 ± 78% -94.1% 0.04 ±223% perf-sched.sch_delay.max.ms.smpboot_thread_fn.kthread.ret_from_fork.ret_from_fork_asm 3.20 ± 81% -99.9% 0.00 ±223% perf-sched.sch_delay.max.ms.syslog_print.do_syslog.kmsg_read.vfs_read 4.72 ± 84% -91.2% 0.41 ±223% perf-sched.sch_delay.max.ms.wait_for_partner.fifo_open.do_dentry_open.do_open 2577 ± 60% -97.9% 54.50 ±223% perf-sched.sch_delay.max.ms.worker_thread.kthread.ret_from_fork.ret_from_fork_asm 1.46 ± 12% -86.0% 0.20 ±223% perf-sched.total_sch_delay.average.ms 3445 ± 15% -90.3% 335.24 ±223% perf-sched.total_sch_delay.max.ms 4.59 ± 11% -85.6% 0.66 ±223% perf-sched.total_wait_and_delay.average.ms 3.13 ± 11% -85.4% 0.46 ±223% perf-sched.total_wait_time.average.ms 7.32 ± 13% -82.1% 1.31 ±223% perf-sched.wait_and_delay.avg.ms.__cond_resched.aa_sk_perm.security_socket_sendmsg.sock_write_iter.vfs_write 33.50 ± 19% -86.8% 4.42 ±223% perf-sched.wait_and_delay.avg.ms.do_task_dead.do_exit.do_group_exit.__x64_sys_exit_group.do_syscall_64 41.53 ± 58% -94.3% 2.35 ±223% perf-sched.wait_and_delay.avg.ms.do_wait.kernel_wait4.__do_sys_wait4.do_syscall_64 16.23 ± 16% -86.5% 2.20 ±223% perf-sched.wait_and_delay.avg.ms.exit_to_user_mode_loop.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_sysvec_apic_timer_interrupt 21.47 ± 7% -86.7% 2.86 ±223% perf-sched.wait_and_delay.avg.ms.exit_to_user_mode_loop.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_sysvec_reschedule_ipi 3.02 ± 16% -86.2% 0.42 ±223% perf-sched.wait_and_delay.avg.ms.exit_to_user_mode_loop.exit_to_user_mode_prepare.syscall_exit_to_user_mode.do_syscall_64 111.86 ± 47% -85.5% 16.22 ±223% perf-sched.wait_and_delay.avg.ms.pipe_read.vfs_read.ksys_read.do_syscall_64 41.81 ± 48% -94.7% 2.20 ±223% perf-sched.wait_and_delay.avg.ms.rcu_gp_kthread.kthread.ret_from_fork.ret_from_fork_asm 466.43 ± 41% -98.7% 5.96 ±223% perf-sched.wait_and_delay.avg.ms.schedule_hrtimeout_range_clock.do_poll.constprop.0.do_sys_poll 517.70 ± 26% -94.5% 28.29 ±223% perf-sched.wait_and_delay.avg.ms.schedule_hrtimeout_range_clock.ep_poll.do_epoll_wait.__x64_sys_epoll_wait 19.49 ±107% -100.0% 0.00 perf-sched.wait_and_delay.avg.ms.schedule_preempt_disabled.rwsem_down_write_slowpath.down_write.unlink_anon_vmas 130.50 ± 66% -96.7% 4.25 ±223% perf-sched.wait_and_delay.avg.ms.schedule_timeout.__wait_for_common.wait_for_completion_state.kernel_clone 827.96 ± 22% -84.2% 130.69 ±223% perf-sched.wait_and_delay.avg.ms.schedule_timeout.kcompactd.kthread.ret_from_fork 22.22 ± 19% -86.4% 3.03 ±223% perf-sched.wait_and_delay.avg.ms.schedule_timeout.rcu_gp_fqs_loop.rcu_gp_kthread.kthread 23.97 ± 3% -84.5% 3.70 ±223% perf-sched.wait_and_delay.avg.ms.schedule_timeout.sock_alloc_send_pskb.unix_stream_sendmsg.sock_write_iter 3.05 ± 11% -86.2% 0.42 ±223% perf-sched.wait_and_delay.avg.ms.schedule_timeout.unix_stream_data_wait.unix_stream_read_generic.unix_stream_recvmsg 699.70 ± 8% -89.9% 70.43 ±223% perf-sched.wait_and_delay.avg.ms.worker_thread.kthread.ret_from_fork.ret_from_fork_asm 57.50 ±105% -100.0% 0.00 perf-sched.wait_and_delay.count.schedule_preempt_disabled.rwsem_down_write_slowpath.down_write.unlink_anon_vmas 506.78 ± 90% -90.8% 46.75 ±223% perf-sched.wait_and_delay.max.ms.__cond_resched.__put_anon_vma.unlink_anon_vmas.free_pgtables.exit_mmap 1145 ± 2% -83.6% 187.73 ±223% perf-sched.wait_and_delay.max.ms.__cond_resched.__wait_for_common.affine_move_task.__set_cpus_allowed_ptr.__sched_setaffinity 1425 ± 55% -88.1% 170.08 ±223% perf-sched.wait_and_delay.max.ms.do_task_dead.do_exit.do_group_exit.__x64_sys_exit_group.do_syscall_64 1687 ± 69% -89.5% 176.52 ±223% perf-sched.wait_and_delay.max.ms.do_wait.kernel_wait4.__do_sys_wait4.do_syscall_64 1142 ± 37% -88.1% 135.48 ±223% perf-sched.wait_and_delay.max.ms.exit_to_user_mode_loop.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_sysvec_reschedule_ipi 4606 ± 21% -85.4% 674.23 ±223% perf-sched.wait_and_delay.max.ms.pipe_read.vfs_read.ksys_read.do_syscall_64 528.00 ±104% -94.0% 31.50 ±223% perf-sched.wait_and_delay.max.ms.rcu_gp_kthread.kthread.ret_from_fork.ret_from_fork_asm 1506 ± 33% -88.9% 166.94 ±223% perf-sched.wait_and_delay.max.ms.schedule_hrtimeout_range_clock.do_poll.constprop.0.do_sys_poll 359.82 ±121% -100.0% 0.00 perf-sched.wait_and_delay.max.ms.schedule_preempt_disabled.rwsem_down_write_slowpath.down_write.unlink_anon_vmas 1932 ± 23% -82.6% 336.00 ±223% perf-sched.wait_and_delay.max.ms.schedule_timeout.kcompactd.kthread.ret_from_fork 1457 ± 10% -87.1% 188.18 ±223% perf-sched.wait_and_delay.max.ms.schedule_timeout.unix_stream_data_wait.unix_stream_read_generic.unix_stream_recvmsg 4952 ± 31% -88.1% 587.66 ±223% perf-sched.wait_and_delay.max.ms.worker_thread.kthread.ret_from_fork.ret_from_fork_asm 2.12 ± 53% -94.2% 0.12 ±223% perf-sched.wait_time.avg.ms.__cond_resched.__alloc_pages.alloc_pages_mpol.shmem_alloc_folio.shmem_alloc_and_add_folio 12.37 ±201% -99.9% 0.02 ±223% perf-sched.wait_time.avg.ms.__cond_resched.__alloc_pages.alloc_pages_mpol.vma_alloc_folio.wp_page_copy 20.52 ± 86% -97.2% 0.57 ±223% perf-sched.wait_time.avg.ms.__cond_resched.__kmem_cache_alloc_node.__kmalloc_node.memcg_alloc_slab_cgroups.allocate_slab 3.06 ± 6% -88.2% 0.36 ±223% perf-sched.wait_time.avg.ms.__cond_resched.__kmem_cache_alloc_node.__kmalloc_node_track_caller.kmalloc_reserve.__alloc_skb 2.82 ± 9% -85.4% 0.41 ±223% perf-sched.wait_time.avg.ms.__cond_resched.aa_sk_perm.security_socket_recvmsg.sock_recvmsg.sock_read_iter 5.25 ± 12% -82.8% 0.90 ±223% perf-sched.wait_time.avg.ms.__cond_resched.aa_sk_perm.security_socket_sendmsg.sock_write_iter.vfs_write 0.50 ±108% -99.4% 0.00 ±223% perf-sched.wait_time.avg.ms.__cond_resched.down_read.acct_collect.do_exit.do_group_exit 2.67 ±145% -99.8% 0.00 ±223% perf-sched.wait_time.avg.ms.__cond_resched.down_write.unlink_anon_vmas.free_pgtables.exit_mmap 0.97 ±156% -99.8% 0.00 ±223% perf-sched.wait_time.avg.ms.__cond_resched.dput.__fput.__x64_sys_close.do_syscall_64 1.53 ± 44% -87.3% 0.19 ±223% perf-sched.wait_time.avg.ms.__cond_resched.generic_perform_write.shmem_file_write_iter.vfs_write.ksys_write 0.22 ±117% -97.4% 0.01 ±223% perf-sched.wait_time.avg.ms.__cond_resched.kmem_cache_alloc.alloc_empty_file.path_openat.do_filp_open 2.74 ± 10% -84.9% 0.41 ±223% perf-sched.wait_time.avg.ms.__cond_resched.kmem_cache_alloc_node.__alloc_skb.alloc_skb_with_frags.sock_alloc_send_pskb 0.74 ±140% -94.7% 0.04 ±223% perf-sched.wait_time.avg.ms.__cond_resched.migrate_pages_batch.migrate_pages.migrate_misplaced_folio.do_numa_page 1.75 ± 12% -84.8% 0.27 ±223% perf-sched.wait_time.avg.ms.__cond_resched.mutex_lock.unix_stream_read_generic.unix_stream_recvmsg.sock_recvmsg 4.91 ± 86% -94.0% 0.30 ±223% perf-sched.wait_time.avg.ms.__cond_resched.rmap_walk_anon.try_to_migrate.migrate_folio_unmap.migrate_pages_batch 1.59 ± 41% -87.0% 0.21 ±223% perf-sched.wait_time.avg.ms.__cond_resched.shmem_inode_acct_blocks.shmem_alloc_and_add_folio.shmem_get_folio_gfp.shmem_write_begin 0.06 ± 91% -100.0% 0.00 perf-sched.wait_time.avg.ms.__cond_resched.tlb_batch_pages_flush.tlb_finish_mmu.shift_arg_pages.setup_arg_pages 0.08 ± 93% -87.7% 0.01 ±223% perf-sched.wait_time.avg.ms.__cond_resched.unmap_vmas.exit_mmap.__mmput.exit_mm 0.07 ± 78% -100.0% 0.00 perf-sched.wait_time.avg.ms.__cond_resched.unmap_vmas.unmap_region.constprop.0 2.38 ± 23% -85.0% 0.36 ±223% perf-sched.wait_time.avg.ms.__cond_resched.wait_for_unix_gc.unix_stream_sendmsg.sock_write_iter.vfs_write 2.47 ±154% -99.2% 0.02 ±223% perf-sched.wait_time.avg.ms.__cond_resched.wp_page_copy.__handle_mm_fault.handle_mm_fault.do_user_addr_fault 2.61 ±109% -99.0% 0.03 ±223% perf-sched.wait_time.avg.ms.__cond_resched.zap_pmd_range.isra.0.unmap_page_range 27.00 ± 17% -88.0% 3.25 ±223% perf-sched.wait_time.avg.ms.do_task_dead.do_exit.do_group_exit.__x64_sys_exit_group.do_syscall_64 27.88 ± 56% -93.0% 1.94 ±223% perf-sched.wait_time.avg.ms.do_wait.kernel_wait4.__do_sys_wait4.do_syscall_64 11.59 ± 19% -87.5% 1.45 ±223% perf-sched.wait_time.avg.ms.exit_to_user_mode_loop.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_sysvec_apic_timer_interrupt 14.88 ± 6% -86.5% 2.01 ±223% perf-sched.wait_time.avg.ms.exit_to_user_mode_loop.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_sysvec_reschedule_ipi 2.09 ± 15% -86.1% 0.29 ±223% perf-sched.wait_time.avg.ms.exit_to_user_mode_loop.exit_to_user_mode_prepare.syscall_exit_to_user_mode.do_syscall_64 27.30 ± 37% -93.6% 1.75 ±223% perf-sched.wait_time.avg.ms.rcu_gp_kthread.kthread.ret_from_fork.ret_from_fork_asm 465.41 ± 41% -98.8% 5.36 ±223% perf-sched.wait_time.avg.ms.schedule_hrtimeout_range_clock.do_poll.constprop.0.do_sys_poll 517.22 ± 26% -94.5% 28.24 ±223% perf-sched.wait_time.avg.ms.schedule_hrtimeout_range_clock.ep_poll.do_epoll_wait.__x64_sys_epoll_wait 13.51 ± 96% -95.9% 0.55 ±223% perf-sched.wait_time.avg.ms.schedule_preempt_disabled.rwsem_down_write_slowpath.down_write.unlink_anon_vmas 105.81 ± 47% -96.1% 4.09 ±223% perf-sched.wait_time.avg.ms.schedule_timeout.__wait_for_common.wait_for_completion_state.kernel_clone 827.95 ± 22% -84.2% 130.67 ±223% perf-sched.wait_time.avg.ms.schedule_timeout.kcompactd.kthread.ret_from_fork 15.58 ± 17% -86.3% 2.14 ±223% perf-sched.wait_time.avg.ms.schedule_timeout.rcu_gp_fqs_loop.rcu_gp_kthread.kthread 16.10 ± 3% -84.5% 2.50 ±223% perf-sched.wait_time.avg.ms.schedule_timeout.sock_alloc_send_pskb.unix_stream_sendmsg.sock_write_iter 2.01 ± 11% -86.2% 0.28 ±223% perf-sched.wait_time.avg.ms.schedule_timeout.unix_stream_data_wait.unix_stream_read_generic.unix_stream_recvmsg 0.16 ± 76% -99.1% 0.00 ±223% perf-sched.wait_time.avg.ms.wait_for_partner.fifo_open.do_dentry_open.do_open 685.02 ± 8% -89.7% 70.37 ±223% perf-sched.wait_time.avg.ms.worker_thread.kthread.ret_from_fork.ret_from_fork_asm 247.60 ± 69% -85.0% 37.19 ±223% perf-sched.wait_time.max.ms.__cond_resched.__alloc_pages.alloc_pages_mpol.shmem_alloc_folio.shmem_alloc_and_add_folio 193.46 ±189% -100.0% 0.08 ±223% perf-sched.wait_time.max.ms.__cond_resched.__alloc_pages.alloc_pages_mpol.vma_alloc_folio.wp_page_copy 170.06 ± 94% -96.5% 5.94 ±223% perf-sched.wait_time.max.ms.__cond_resched.__kmem_cache_alloc_node.__kmalloc_node.memcg_alloc_slab_cgroups.allocate_slab 353.19 ± 71% -86.8% 46.75 ±223% perf-sched.wait_time.max.ms.__cond_resched.__put_anon_vma.unlink_anon_vmas.free_pgtables.exit_mmap 1145 ± 2% -83.6% 187.73 ±223% perf-sched.wait_time.max.ms.__cond_resched.__wait_for_common.affine_move_task.__set_cpus_allowed_ptr.__sched_setaffinity 2.91 ±140% -99.7% 0.01 ±223% perf-sched.wait_time.max.ms.__cond_resched.down_read.acct_collect.do_exit.do_group_exit 16.59 ±159% -100.0% 0.01 ±223% perf-sched.wait_time.max.ms.__cond_resched.down_write.unlink_anon_vmas.free_pgtables.exit_mmap 1.59 ± 98% -99.9% 0.00 ±223% perf-sched.wait_time.max.ms.__cond_resched.dput.__fput.__x64_sys_close.do_syscall_64 55.59 ± 80% -86.7% 7.40 ±223% perf-sched.wait_time.max.ms.__cond_resched.exit_mmap.__mmput.exit_mm.do_exit 0.37 ± 93% -98.5% 0.01 ±223% perf-sched.wait_time.max.ms.__cond_resched.kmem_cache_alloc.alloc_empty_file.path_openat.do_filp_open 620.61 ± 26% -88.4% 71.76 ±223% perf-sched.wait_time.max.ms.__cond_resched.kmem_cache_alloc_node.__alloc_skb.alloc_skb_with_frags.sock_alloc_send_pskb 3.36 ±152% -96.8% 0.11 ±223% perf-sched.wait_time.max.ms.__cond_resched.migrate_pages_batch.migrate_pages.migrate_misplaced_folio.do_numa_page 134.38 ±117% -95.1% 6.60 ±223% perf-sched.wait_time.max.ms.__cond_resched.rmap_walk_anon.try_to_migrate.migrate_folio_unmap.migrate_pages_batch 116.95 ±144% -97.1% 3.42 ±223% perf-sched.wait_time.max.ms.__cond_resched.tlb_batch_pages_flush.tlb_finish_mmu.exit_mmap.__mmput 0.07 ± 84% -100.0% 0.00 perf-sched.wait_time.max.ms.__cond_resched.tlb_batch_pages_flush.tlb_finish_mmu.shift_arg_pages.setup_arg_pages 0.13 ± 84% -100.0% 0.00 perf-sched.wait_time.max.ms.__cond_resched.unmap_vmas.unmap_region.constprop.0 366.28 ± 24% -81.8% 66.58 ±223% perf-sched.wait_time.max.ms.__cond_resched.wait_for_unix_gc.unix_stream_sendmsg.sock_write_iter.vfs_write 29.17 ±158% -99.7% 0.08 ±223% perf-sched.wait_time.max.ms.__cond_resched.wp_page_copy.__handle_mm_fault.handle_mm_fault.do_user_addr_fault 100.14 ±128% -99.3% 0.73 ±223% perf-sched.wait_time.max.ms.__cond_resched.zap_pmd_range.isra.0.unmap_page_range 1420 ± 55% -88.2% 168.37 ±223% perf-sched.wait_time.max.ms.do_task_dead.do_exit.do_group_exit.__x64_sys_exit_group.do_syscall_64 417.03 ± 26% -86.6% 55.67 ±223% perf-sched.wait_time.max.ms.exit_to_user_mode_loop.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_sysvec_apic_timer_interrupt 828.97 ± 30% -91.5% 70.61 ±223% perf-sched.wait_time.max.ms.exit_to_user_mode_loop.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_sysvec_reschedule_ipi 303.65 ± 86% -94.1% 18.00 ±223% perf-sched.wait_time.max.ms.rcu_gp_kthread.kthread.ret_from_fork.ret_from_fork_asm 1503 ± 33% -88.9% 166.86 ±223% perf-sched.wait_time.max.ms.schedule_hrtimeout_range_clock.do_poll.constprop.0.do_sys_poll 199.45 ±107% -93.2% 13.49 ±223% perf-sched.wait_time.max.ms.schedule_preempt_disabled.rwsem_down_write_slowpath.down_write.unlink_anon_vmas 1932 ± 23% -82.6% 336.00 ±223% perf-sched.wait_time.max.ms.schedule_timeout.kcompactd.kthread.ret_from_fork 781.42 ± 12% -84.4% 121.90 ±223% perf-sched.wait_time.max.ms.schedule_timeout.sock_alloc_send_pskb.unix_stream_sendmsg.sock_write_iter 822.69 ± 9% -87.9% 99.56 ±223% perf-sched.wait_time.max.ms.schedule_timeout.unix_stream_data_wait.unix_stream_read_generic.unix_stream_recvmsg 0.82 ± 78% -98.4% 0.01 ±223% perf-sched.wait_time.max.ms.wait_for_partner.fifo_open.do_dentry_open.do_open 4444 ± 13% -86.8% 587.66 ±223% perf-sched.wait_time.max.ms.worker_thread.kthread.ret_from_fork.ret_from_fork_asm 3.68 ± 7% -1.5 2.22 ± 7% perf-profile.calltrace.cycles-pp.__unfreeze_partials.skb_release_data.consume_skb.unix_stream_read_generic.unix_stream_recvmsg 3.38 ± 8% -1.4 2.02 ± 8% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.__unfreeze_partials.skb_release_data.consume_skb 3.46 ± 8% -1.3 2.13 ± 7% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.__unfreeze_partials.skb_release_data.consume_skb.unix_stream_read_generic 10.21 ± 2% -1.2 9.04 perf-profile.calltrace.cycles-pp.skb_release_data.consume_skb.unix_stream_read_generic.unix_stream_recvmsg.sock_recvmsg 3.06 ± 8% -1.1 1.96 ± 5% perf-profile.calltrace.cycles-pp.__unfreeze_partials.unix_stream_read_generic.unix_stream_recvmsg.sock_recvmsg.sock_read_iter 14.35 -1.0 13.34 perf-profile.calltrace.cycles-pp.consume_skb.unix_stream_read_generic.unix_stream_recvmsg.sock_recvmsg.sock_read_iter 2.76 ± 9% -1.0 1.76 ± 5% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.__unfreeze_partials.unix_stream_read_generic.unix_stream_recvmsg 2.84 ± 9% -1.0 1.86 ± 5% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.__unfreeze_partials.unix_stream_read_generic.unix_stream_recvmsg.sock_recvmsg 2.50 ± 7% -1.0 1.55 ± 7% perf-profile.calltrace.cycles-pp.get_partial_node.___slab_alloc.__kmem_cache_alloc_node.__kmalloc_node_track_caller.kmalloc_reserve 16.14 -0.9 15.20 perf-profile.calltrace.cycles-pp.__alloc_skb.alloc_skb_with_frags.sock_alloc_send_pskb.unix_stream_sendmsg.sock_write_iter 16.46 -0.9 15.54 perf-profile.calltrace.cycles-pp.alloc_skb_with_frags.sock_alloc_send_pskb.unix_stream_sendmsg.sock_write_iter.vfs_write 41.42 -0.9 40.52 perf-profile.calltrace.cycles-pp.unix_stream_read_generic.unix_stream_recvmsg.sock_recvmsg.sock_read_iter.vfs_read 2.24 ± 7% -0.9 1.35 ± 8% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.get_partial_node.___slab_alloc.__kmem_cache_alloc_node.__kmalloc_node_track_caller 2.22 ± 7% -0.9 1.34 ± 8% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.get_partial_node.___slab_alloc.__kmem_cache_alloc_node 41.86 -0.9 40.98 perf-profile.calltrace.cycles-pp.unix_stream_recvmsg.sock_recvmsg.sock_read_iter.vfs_read.ksys_read 2.98 ± 6% -0.9 2.11 ± 6% perf-profile.calltrace.cycles-pp.___slab_alloc.__kmem_cache_alloc_node.__kmalloc_node_track_caller.kmalloc_reserve.__alloc_skb 19.50 -0.8 18.71 perf-profile.calltrace.cycles-pp.sock_alloc_send_pskb.unix_stream_sendmsg.sock_write_iter.vfs_write.ksys_write 43.08 -0.8 42.30 perf-profile.calltrace.cycles-pp.sock_recvmsg.sock_read_iter.vfs_read.ksys_read.do_syscall_64 44.05 -0.7 43.31 perf-profile.calltrace.cycles-pp.sock_read_iter.vfs_read.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe 2.06 ± 8% -0.7 1.36 ± 4% perf-profile.calltrace.cycles-pp.get_partial_node.___slab_alloc.kmem_cache_alloc_node.__alloc_skb.alloc_skb_with_frags 7.16 ± 2% -0.7 6.50 ± 2% perf-profile.calltrace.cycles-pp.__kmem_cache_alloc_node.__kmalloc_node_track_caller.kmalloc_reserve.__alloc_skb.alloc_skb_with_frags 1.82 ± 9% -0.6 1.17 ± 5% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.get_partial_node.___slab_alloc.kmem_cache_alloc_node.__alloc_skb 7.65 ± 2% -0.6 7.01 perf-profile.calltrace.cycles-pp.__kmalloc_node_track_caller.kmalloc_reserve.__alloc_skb.alloc_skb_with_frags.sock_alloc_send_pskb 1.80 ± 9% -0.6 1.16 ± 5% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.get_partial_node.___slab_alloc.kmem_cache_alloc_node 2.54 ± 6% -0.6 1.90 ± 3% perf-profile.calltrace.cycles-pp.___slab_alloc.kmem_cache_alloc_node.__alloc_skb.alloc_skb_with_frags.sock_alloc_send_pskb 46.80 -0.6 46.18 perf-profile.calltrace.cycles-pp.vfs_read.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_read 8.22 ± 2% -0.6 7.61 perf-profile.calltrace.cycles-pp.kmalloc_reserve.__alloc_skb.alloc_skb_with_frags.sock_alloc_send_pskb.unix_stream_sendmsg 47.75 -0.6 47.20 perf-profile.calltrace.cycles-pp.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_read 49.22 -0.5 48.74 perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_read 49.54 -0.5 49.08 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.__libc_read 6.01 ± 2% -0.4 5.57 perf-profile.calltrace.cycles-pp.kmem_cache_alloc_node.__alloc_skb.alloc_skb_with_frags.sock_alloc_send_pskb.unix_stream_sendmsg 52.34 -0.3 52.03 perf-profile.calltrace.cycles-pp.__libc_read 0.58 +0.0 0.60 perf-profile.calltrace.cycles-pp.exit_to_user_mode_prepare.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_write 0.60 +0.0 0.62 perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.skb_unlink.unix_stream_read_generic.unix_stream_recvmsg.sock_recvmsg 0.57 +0.0 0.60 perf-profile.calltrace.cycles-pp.mod_objcg_state.memcg_slab_post_alloc_hook.kmem_cache_alloc_node.__alloc_skb.alloc_skb_with_frags 0.83 +0.0 0.86 perf-profile.calltrace.cycles-pp.mod_objcg_state.__kmem_cache_free.skb_release_data.consume_skb.unix_stream_read_generic 0.74 +0.0 0.77 perf-profile.calltrace.cycles-pp.skb_unlink.unix_stream_read_generic.unix_stream_recvmsg.sock_recvmsg.sock_read_iter 0.76 +0.0 0.79 perf-profile.calltrace.cycles-pp.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_write 0.90 +0.0 0.94 ± 2% perf-profile.calltrace.cycles-pp.security_file_permission.vfs_read.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe 0.61 +0.0 0.64 perf-profile.calltrace.cycles-pp.__build_skb_around.__alloc_skb.alloc_skb_with_frags.sock_alloc_send_pskb.unix_stream_sendmsg 0.54 +0.0 0.58 perf-profile.calltrace.cycles-pp.obj_cgroup_charge.kmem_cache_alloc_node.__alloc_skb.alloc_skb_with_frags.sock_alloc_send_pskb 0.65 +0.0 0.68 perf-profile.calltrace.cycles-pp.mutex_lock.unix_stream_read_generic.unix_stream_recvmsg.sock_recvmsg.sock_read_iter 0.89 +0.0 0.92 perf-profile.calltrace.cycles-pp.unix_write_space.sock_wfree.unix_destruct_scm.skb_release_head_state.consume_skb 0.71 +0.0 0.75 perf-profile.calltrace.cycles-pp.exit_to_user_mode_prepare.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_read 0.78 +0.0 0.81 perf-profile.calltrace.cycles-pp.mod_objcg_state.kmem_cache_free.unix_stream_read_generic.unix_stream_recvmsg.sock_recvmsg 0.75 +0.0 0.79 perf-profile.calltrace.cycles-pp.obj_cgroup_charge.__kmem_cache_alloc_node.__kmalloc_node_track_caller.kmalloc_reserve.__alloc_skb 0.57 +0.0 0.60 ± 2% perf-profile.calltrace.cycles-pp.enqueue_entity.enqueue_task_fair.activate_task.ttwu_do_activate.try_to_wake_up 0.64 +0.0 0.68 perf-profile.calltrace.cycles-pp.check_heap_object.__check_object_size.skb_copy_datagram_from_iter.unix_stream_sendmsg.sock_write_iter 0.68 +0.0 0.72 ± 3% perf-profile.calltrace.cycles-pp.dequeue_task_fair.__schedule.schedule.schedule_timeout.unix_stream_data_wait 0.92 +0.1 0.98 perf-profile.calltrace.cycles-pp.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe.__libc_read 0.63 +0.1 0.69 ± 3% perf-profile.calltrace.cycles-pp.aa_sk_perm.security_socket_recvmsg.sock_recvmsg.sock_read_iter.vfs_read 0.96 +0.1 1.02 ± 2% perf-profile.calltrace.cycles-pp.activate_task.ttwu_do_activate.try_to_wake_up.autoremove_wake_function.__wake_up_common 0.91 +0.1 0.97 ± 2% perf-profile.calltrace.cycles-pp.enqueue_task_fair.activate_task.ttwu_do_activate.try_to_wake_up.autoremove_wake_function 1.14 +0.1 1.20 perf-profile.calltrace.cycles-pp.ttwu_do_activate.try_to_wake_up.autoremove_wake_function.__wake_up_common.__wake_up_sync_key 1.68 +0.1 1.75 perf-profile.calltrace.cycles-pp.memcg_slab_post_alloc_hook.__kmem_cache_alloc_node.__kmalloc_node_track_caller.kmalloc_reserve.__alloc_skb 1.84 +0.1 1.92 perf-profile.calltrace.cycles-pp.__entry_text_start.__libc_write 1.70 +0.1 1.77 perf-profile.calltrace.cycles-pp.__slab_free.unix_stream_read_generic.unix_stream_recvmsg.sock_recvmsg.sock_read_iter 1.59 +0.1 1.67 perf-profile.calltrace.cycles-pp.memcg_slab_post_alloc_hook.kmem_cache_alloc_node.__alloc_skb.alloc_skb_with_frags.sock_alloc_send_pskb 1.91 +0.1 1.98 perf-profile.calltrace.cycles-pp.__slab_free.skb_release_data.consume_skb.unix_stream_read_generic.unix_stream_recvmsg 0.93 +0.1 1.01 perf-profile.calltrace.cycles-pp.security_socket_recvmsg.sock_recvmsg.sock_read_iter.vfs_read.ksys_read 1.24 +0.1 1.32 perf-profile.calltrace.cycles-pp.__check_object_size.skb_copy_datagram_from_iter.unix_stream_sendmsg.sock_write_iter.vfs_write 1.75 +0.1 1.86 perf-profile.calltrace.cycles-pp.__entry_text_start.__libc_read 2.40 +0.1 2.52 perf-profile.calltrace.cycles-pp.check_heap_object.__check_object_size.simple_copy_to_iter.__skb_datagram_iter.skb_copy_datagram_iter 2.41 +0.1 2.54 ± 2% perf-profile.calltrace.cycles-pp.__schedule.schedule.schedule_timeout.unix_stream_data_wait.unix_stream_read_generic 3.05 +0.1 3.18 perf-profile.calltrace.cycles-pp.skb_copy_datagram_from_iter.unix_stream_sendmsg.sock_write_iter.vfs_write.ksys_write 2.45 +0.1 2.59 ± 2% perf-profile.calltrace.cycles-pp.schedule.schedule_timeout.unix_stream_data_wait.unix_stream_read_generic.unix_stream_recvmsg 3.48 +0.1 3.62 ± 2% perf-profile.calltrace.cycles-pp.unix_destruct_scm.skb_release_head_state.consume_skb.unix_stream_read_generic.unix_stream_recvmsg 2.55 +0.1 2.70 ± 2% perf-profile.calltrace.cycles-pp.schedule_timeout.unix_stream_data_wait.unix_stream_read_generic.unix_stream_recvmsg.sock_recvmsg 3.58 +0.2 3.73 perf-profile.calltrace.cycles-pp.__kmem_cache_free.skb_release_data.consume_skb.unix_stream_read_generic.unix_stream_recvmsg 3.07 +0.2 3.23 ± 2% perf-profile.calltrace.cycles-pp.unix_stream_data_wait.unix_stream_read_generic.unix_stream_recvmsg.sock_recvmsg.sock_read_iter 3.19 +0.2 3.36 perf-profile.calltrace.cycles-pp.__check_object_size.simple_copy_to_iter.__skb_datagram_iter.skb_copy_datagram_iter.unix_stream_read_actor 3.43 +0.2 3.62 perf-profile.calltrace.cycles-pp.simple_copy_to_iter.__skb_datagram_iter.skb_copy_datagram_iter.unix_stream_read_actor.unix_stream_read_generic 3.44 +0.2 3.66 ± 2% perf-profile.calltrace.cycles-pp._copy_to_iter.__skb_datagram_iter.skb_copy_datagram_iter.unix_stream_read_actor.unix_stream_read_generic 5.51 +0.3 5.76 perf-profile.calltrace.cycles-pp.kmem_cache_free.unix_stream_read_generic.unix_stream_recvmsg.sock_recvmsg.sock_read_iter 7.13 +0.4 7.56 perf-profile.calltrace.cycles-pp.__skb_datagram_iter.skb_copy_datagram_iter.unix_stream_read_actor.unix_stream_read_generic.unix_stream_recvmsg 7.33 +0.4 7.77 perf-profile.calltrace.cycles-pp.skb_copy_datagram_iter.unix_stream_read_actor.unix_stream_read_generic.unix_stream_recvmsg.sock_recvmsg 7.43 +0.4 7.87 perf-profile.calltrace.cycles-pp.unix_stream_read_actor.unix_stream_read_generic.unix_stream_recvmsg.sock_recvmsg.sock_read_iter 11.75 ± 3% -3.9 7.87 ± 4% perf-profile.children.cycles-pp.native_queued_spin_lock_slowpath 14.09 ± 2% -3.8 10.26 ± 3% perf-profile.children.cycles-pp._raw_spin_lock_irqsave 6.76 ± 3% -2.6 4.19 ± 6% perf-profile.children.cycles-pp.__unfreeze_partials 4.66 ± 3% -1.7 2.98 ± 5% perf-profile.children.cycles-pp.get_partial_node 5.53 ± 2% -1.5 4.02 ± 4% perf-profile.children.cycles-pp.___slab_alloc 10.25 ± 2% -1.2 9.08 perf-profile.children.cycles-pp.skb_release_data 14.45 -1.0 13.44 perf-profile.children.cycles-pp.consume_skb 16.26 -0.9 15.32 perf-profile.children.cycles-pp.__alloc_skb 16.53 -0.9 15.61 perf-profile.children.cycles-pp.alloc_skb_with_frags 41.65 -0.9 40.76 perf-profile.children.cycles-pp.unix_stream_read_generic 41.90 -0.9 41.03 perf-profile.children.cycles-pp.unix_stream_recvmsg 19.57 -0.8 18.78 perf-profile.children.cycles-pp.sock_alloc_send_pskb 43.15 -0.8 42.37 perf-profile.children.cycles-pp.sock_recvmsg 44.10 -0.7 43.36 perf-profile.children.cycles-pp.sock_read_iter 7.32 ± 2% -0.7 6.66 perf-profile.children.cycles-pp.__kmem_cache_alloc_node 7.75 ± 2% -0.6 7.11 perf-profile.children.cycles-pp.__kmalloc_node_track_caller 46.92 -0.6 46.30 perf-profile.children.cycles-pp.vfs_read 8.32 ± 2% -0.6 7.72 perf-profile.children.cycles-pp.kmalloc_reserve 47.82 -0.6 47.26 perf-profile.children.cycles-pp.ksys_read 6.16 ± 2% -0.4 5.72 perf-profile.children.cycles-pp.kmem_cache_alloc_node 93.67 -0.3 93.33 perf-profile.children.cycles-pp.do_syscall_64 94.20 -0.3 93.88 perf-profile.children.cycles-pp.entry_SYSCALL_64_after_hwframe 52.43 -0.3 52.12 perf-profile.children.cycles-pp.__libc_read 0.10 ± 7% -0.0 0.07 ± 5% perf-profile.children.cycles-pp.get_any_partial 0.18 ± 2% +0.0 0.20 ± 2% perf-profile.children.cycles-pp.wake_affine 0.30 +0.0 0.31 perf-profile.children.cycles-pp.__scm_recv_common 0.22 +0.0 0.23 ± 2% perf-profile.children.cycles-pp.kfree 0.17 ± 2% +0.0 0.19 ± 3% perf-profile.children.cycles-pp.__switch_to 0.29 +0.0 0.31 ± 2% perf-profile.children.cycles-pp.restore_fpregs_from_fpstate 0.34 +0.0 0.35 perf-profile.children.cycles-pp.wait_for_unix_gc 0.22 ± 2% +0.0 0.24 ± 2% perf-profile.children.cycles-pp.check_stack_object 0.46 ± 2% +0.0 0.48 perf-profile.children.cycles-pp.rcu_all_qs 0.26 +0.0 0.28 ± 3% perf-profile.children.cycles-pp.current_obj_cgroup 0.48 +0.0 0.51 perf-profile.children.cycles-pp.mutex_unlock 0.33 ± 2% +0.0 0.36 perf-profile.children.cycles-pp._raw_spin_unlock_irqrestore 0.68 +0.0 0.71 perf-profile.children.cycles-pp.update_load_avg 0.35 +0.0 0.38 ± 2% perf-profile.children.cycles-pp.kmalloc_size_roundup 0.42 +0.0 0.44 perf-profile.children.cycles-pp.syscall_enter_from_user_mode 0.40 +0.0 0.43 perf-profile.children.cycles-pp.kmalloc_slab 0.40 +0.0 0.42 ± 2% perf-profile.children.cycles-pp.aa_file_perm 0.40 +0.0 0.43 perf-profile.children.cycles-pp.switch_fpu_return 0.67 +0.0 0.70 perf-profile.children.cycles-pp.exit_to_user_mode_loop 0.79 +0.0 0.81 perf-profile.children.cycles-pp.skb_unlink 0.48 +0.0 0.51 ± 2% perf-profile.children.cycles-pp.__get_task_ioprio 0.54 ± 2% +0.0 0.56 perf-profile.children.cycles-pp.update_curr 0.29 +0.0 0.32 ± 2% perf-profile.children.cycles-pp.__list_del_entry_valid_or_report 0.48 ± 2% +0.0 0.50 perf-profile.children.cycles-pp.dequeue_entity 0.50 +0.0 0.53 ± 2% perf-profile.children.cycles-pp.switch_mm_irqs_off 0.64 +0.0 0.67 perf-profile.children.cycles-pp.__build_skb_around 0.91 +0.0 0.95 perf-profile.children.cycles-pp.unix_write_space 0.70 +0.0 0.74 perf-profile.children.cycles-pp.enqueue_entity 0.73 +0.0 0.77 perf-profile.children.cycles-pp.refill_obj_stock 0.69 +0.0 0.74 perf-profile.children.cycles-pp.mutex_lock 0.86 +0.0 0.91 perf-profile.children.cycles-pp.dequeue_task_fair 0.78 +0.1 0.83 perf-profile.children.cycles-pp.__check_heap_object 0.14 ± 3% +0.1 0.19 ± 3% perf-profile.children.cycles-pp.__list_add_valid_or_report 1.44 +0.1 1.49 perf-profile.children.cycles-pp.apparmor_file_permission 1.18 ± 2% +0.1 1.24 perf-profile.children.cycles-pp.security_socket_sendmsg 1.20 +0.1 1.26 perf-profile.children.cycles-pp.activate_task 1.11 +0.1 1.17 perf-profile.children.cycles-pp.enqueue_task_fair 1.22 +0.1 1.28 perf-profile.children.cycles-pp.__cond_resched 1.36 +0.1 1.42 perf-profile.children.cycles-pp.exit_to_user_mode_prepare 1.83 +0.1 1.90 perf-profile.children.cycles-pp.security_file_permission 1.32 +0.1 1.40 perf-profile.children.cycles-pp.ttwu_do_activate 1.36 +0.1 1.44 perf-profile.children.cycles-pp.obj_cgroup_charge 0.98 +0.1 1.06 perf-profile.children.cycles-pp.security_socket_recvmsg 1.85 +0.1 1.94 perf-profile.children.cycles-pp.syscall_exit_to_user_mode 1.54 +0.1 1.63 perf-profile.children.cycles-pp.entry_SYSRETQ_unsafe_stack 2.14 +0.1 2.24 perf-profile.children.cycles-pp.__entry_text_start 1.58 ± 2% +0.1 1.68 ± 2% perf-profile.children.cycles-pp.aa_sk_perm 2.81 +0.1 2.94 perf-profile.children.cycles-pp.mod_objcg_state 3.26 +0.1 3.40 ± 2% perf-profile.children.cycles-pp.sock_wfree 3.68 +0.1 3.82 perf-profile.children.cycles-pp.skb_release_head_state 3.13 +0.1 3.27 perf-profile.children.cycles-pp.skb_copy_datagram_from_iter 3.65 +0.2 3.80 perf-profile.children.cycles-pp.__slab_free 3.06 +0.2 3.22 perf-profile.children.cycles-pp.schedule_timeout 3.36 +0.2 3.52 perf-profile.children.cycles-pp.memcg_slab_post_alloc_hook 3.64 +0.2 3.80 perf-profile.children.cycles-pp.__kmem_cache_free 3.09 +0.2 3.25 ± 2% perf-profile.children.cycles-pp.unix_stream_data_wait 3.18 +0.2 3.34 perf-profile.children.cycles-pp.check_heap_object 3.42 +0.2 3.60 perf-profile.children.cycles-pp.schedule 3.44 +0.2 3.61 perf-profile.children.cycles-pp.__schedule 4.34 +0.2 4.52 ± 2% perf-profile.children.cycles-pp._raw_spin_lock 3.47 +0.2 3.66 perf-profile.children.cycles-pp.simple_copy_to_iter 3.46 +0.2 3.69 ± 2% perf-profile.children.cycles-pp._copy_to_iter 5.58 +0.3 5.84 perf-profile.children.cycles-pp.kmem_cache_free 4.77 +0.3 5.05 perf-profile.children.cycles-pp.__check_object_size 7.19 +0.4 7.62 perf-profile.children.cycles-pp.__skb_datagram_iter 7.37 +0.4 7.81 perf-profile.children.cycles-pp.skb_copy_datagram_iter 7.47 +0.4 7.92 perf-profile.children.cycles-pp.unix_stream_read_actor 11.73 ± 3% -3.9 7.85 ± 5% perf-profile.self.cycles-pp.native_queued_spin_lock_slowpath 0.35 ± 2% -0.3 0.06 ± 7% perf-profile.self.cycles-pp.__unfreeze_partials 0.30 -0.1 0.16 ± 3% perf-profile.self.cycles-pp.get_partial_node 0.23 +0.0 0.24 perf-profile.self.cycles-pp.__scm_recv_common 0.12 +0.0 0.13 perf-profile.self.cycles-pp.pick_next_task_fair 0.20 +0.0 0.21 perf-profile.self.cycles-pp.kfree 0.17 +0.0 0.18 ± 2% perf-profile.self.cycles-pp.check_stack_object 0.16 ± 2% +0.0 0.18 ± 2% perf-profile.self.cycles-pp.kmalloc_size_roundup 0.30 +0.0 0.31 perf-profile.self.cycles-pp.sock_recvmsg 0.24 +0.0 0.25 perf-profile.self.cycles-pp.__kmalloc_node_track_caller 0.17 ± 2% +0.0 0.18 ± 2% perf-profile.self.cycles-pp.rw_verify_area 0.11 ± 3% +0.0 0.12 ± 3% perf-profile.self.cycles-pp.try_charge_memcg 0.25 ± 2% +0.0 0.27 ± 2% perf-profile.self.cycles-pp.exit_to_user_mode_prepare 0.23 ± 2% +0.0 0.25 ± 3% perf-profile.self.cycles-pp.__enqueue_entity 0.21 +0.0 0.23 ± 3% perf-profile.self.cycles-pp.current_obj_cgroup 0.26 +0.0 0.28 ± 2% perf-profile.self.cycles-pp.ksys_read 0.28 ± 2% +0.0 0.30 perf-profile.self.cycles-pp.kmalloc_reserve 0.30 +0.0 0.31 perf-profile.self.cycles-pp.alloc_skb_with_frags 0.28 +0.0 0.30 perf-profile.self.cycles-pp._raw_spin_unlock_irqrestore 0.30 +0.0 0.32 perf-profile.self.cycles-pp.__skb_datagram_iter 0.42 +0.0 0.44 ± 2% perf-profile.self.cycles-pp.__get_task_ioprio 0.30 +0.0 0.32 perf-profile.self.cycles-pp.syscall_exit_to_user_mode 0.24 +0.0 0.26 perf-profile.self.cycles-pp.update_curr 0.46 +0.0 0.48 perf-profile.self.cycles-pp.mutex_unlock 0.36 +0.0 0.38 perf-profile.self.cycles-pp.syscall_enter_from_user_mode 0.44 ± 2% +0.0 0.47 perf-profile.self.cycles-pp.mutex_lock 0.34 ± 2% +0.0 0.36 perf-profile.self.cycles-pp.skb_copy_datagram_from_iter 0.56 +0.0 0.58 ± 2% perf-profile.self.cycles-pp.__virt_addr_valid 0.50 +0.0 0.52 ± 2% perf-profile.self.cycles-pp.switch_mm_irqs_off 0.61 +0.0 0.64 perf-profile.self.cycles-pp.do_syscall_64 0.29 +0.0 0.32 perf-profile.self.cycles-pp.__list_del_entry_valid_or_report 0.56 +0.0 0.59 perf-profile.self.cycles-pp.entry_SYSCALL_64_after_hwframe 0.48 +0.0 0.51 perf-profile.self.cycles-pp.sock_alloc_send_pskb 0.92 +0.0 0.95 perf-profile.self.cycles-pp.__libc_write 0.53 +0.0 0.57 perf-profile.self.cycles-pp.unix_write_space 0.42 +0.0 0.45 ± 2% perf-profile.self.cycles-pp.__schedule 0.59 +0.0 0.62 perf-profile.self.cycles-pp.__build_skb_around 0.68 +0.0 0.72 perf-profile.self.cycles-pp.refill_obj_stock 0.94 +0.0 0.98 perf-profile.self.cycles-pp.sock_read_iter 0.68 +0.0 0.72 perf-profile.self.cycles-pp.__check_object_size 0.70 +0.0 0.74 perf-profile.self.cycles-pp.__cond_resched 0.89 +0.0 0.94 perf-profile.self.cycles-pp.obj_cgroup_charge 0.13 +0.1 0.18 ± 3% perf-profile.self.cycles-pp.__list_add_valid_or_report 0.73 +0.1 0.78 perf-profile.self.cycles-pp.__check_heap_object 1.43 +0.1 1.49 perf-profile.self.cycles-pp.vfs_read 1.11 +0.1 1.17 perf-profile.self.cycles-pp.kmem_cache_alloc_node 1.18 +0.1 1.24 perf-profile.self.cycles-pp.__alloc_skb 1.24 +0.1 1.30 perf-profile.self.cycles-pp.sock_write_iter 1.37 +0.1 1.44 perf-profile.self.cycles-pp.vfs_write 1.48 +0.1 1.56 perf-profile.self.cycles-pp.__kmem_cache_alloc_node 1.21 ± 2% +0.1 1.30 ± 2% perf-profile.self.cycles-pp.aa_sk_perm 1.49 +0.1 1.58 perf-profile.self.cycles-pp.entry_SYSRETQ_unsafe_stack 2.18 +0.1 2.28 perf-profile.self.cycles-pp.memcg_slab_post_alloc_hook 2.21 +0.1 2.32 perf-profile.self.cycles-pp.__kmem_cache_free 2.51 +0.1 2.63 perf-profile.self.cycles-pp.mod_objcg_state 2.34 +0.1 2.47 perf-profile.self.cycles-pp.check_heap_object 3.46 +0.1 3.59 ± 2% perf-profile.self.cycles-pp._raw_spin_lock 3.42 +0.1 3.57 perf-profile.self.cycles-pp.unix_stream_read_generic 3.59 +0.2 3.74 perf-profile.self.cycles-pp.__slab_free 0.86 +0.2 1.02 perf-profile.self.cycles-pp.___slab_alloc 4.32 +0.2 4.52 perf-profile.self.cycles-pp.kmem_cache_free 3.42 +0.2 3.64 ± 2% perf-profile.self.cycles-pp._copy_to_iter 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/wiki