[<prev] [next>] [day] [month] [year] [list]
Message-ID: <20200812091727.GF7488@shao2-debian>
Date: Wed, 12 Aug 2020 17:17:27 +0800
From: kernel test robot <rong.a.chen@...el.com>
To: Linus Torvalds <torvalds@...ux-foundation.org>
Cc: Kirill Smelkov <kirr@...edi.com>,
Eic Dumazet <edumazet@...gle.com>,
Al Viro <viro@...iv.linux.org.uk>,
Alan Stern <stern@...land.harvard.edu>,
Marco Elver <elver@...gle.com>,
Andrea Parri <parri.andrea@...il.com>,
Paul McKenney <paulmck@...nel.org>,
LKML <linux-kernel@...r.kernel.org>, lkp@...ts.01.org
Subject: [vfs] 0be0ee7181: netperf.Throughput_tps 3.0% improvement
Greeting,
FYI, we noticed a 3.0% improvement of netperf.Throughput_tps due to commit:
commit: 0be0ee71816b2b6725e2b4f32ad6726c9d729777 ("vfs: properly and reliably lock f_pos in fdget_pos()")
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master
in testcase: netperf
on test machine: 16 threads Intel(R) Xeon(R) E-2278G CPU @ 3.40GHz with 32G memory
with following parameters:
ip: ipv4
runtime: 300s
nr_threads: 50%
cluster: cs-localhost
test: UDP_RR
cpufreq_governor: performance
ucode: 0xd6
test-description: Netperf is a benchmark that can be use to measure various aspect of networking performance.
test-url: http://www.netperf.org/netperf/
Details are as below:
-------------------------------------------------------------------------------------------------->
To reproduce:
git clone https://github.com/intel/lkp-tests.git
cd lkp-tests
bin/lkp install job.yaml # job file is attached in this email
bin/lkp run job.yaml
=========================================================================================
cluster/compiler/cpufreq_governor/ip/kconfig/nr_threads/rootfs/runtime/tbox_group/test/testcase/ucode:
cs-localhost/gcc-9/performance/ipv4/x86_64-rhel-8.3/50%/debian-10.4-x86_64-20200603.cgz/300s/lkp-cfl-e1/UDP_RR/netperf/0xd6
commit:
d8e464ecc1 ("vfs: mark pipes and sockets as stream-like file descriptors")
0be0ee7181 ("vfs: properly and reliably lock f_pos in fdget_pos()")
d8e464ecc17b4444 0be0ee71816b2b6725e2b4f32ad
---------------- ---------------------------
%stddev %change %stddev
\ | \
863649 +3.0% 889801 netperf.Throughput_total_tps
107956 +3.0% 111225 netperf.Throughput_tps
60.93 +7.2% 65.30 ± 2% netperf.time.user_time
2.591e+08 +3.0% 2.669e+08 netperf.time.voluntary_context_switches
2.591e+08 +3.0% 2.669e+08 netperf.workload
3407622 +3.0% 3510829 vmstat.system.cs
55647 ± 9% +30.6% 72702 ± 7% sched_debug.cpu.avg_idle.avg
1.79 ± 7% -16.3% 1.50 ± 7% sched_debug.cpu.nr_running.max
37625 ± 4% +6.8% 40188 ± 2% slabinfo.vm_area_struct.active_objs
37657 ± 4% +6.8% 40221 ± 2% slabinfo.vm_area_struct.num_objs
737.50 -1.6% 726.00 proc-vmstat.nr_page_table_pages
6418 ± 2% +3.3% 6632 proc-vmstat.nr_shmem
7186 ± 3% +6.4% 7643 ± 2% proc-vmstat.pgactivate
32091 ± 4% -53.1% 15050 ± 50% cpuidle.C1E.usage
416466 ± 11% +1676.3% 7397759 ± 61% cpuidle.C3.time
1866 ± 51% +2561.5% 49677 ± 50% cpuidle.C3.usage
13374705 ± 27% -96.0% 532246 ± 22% cpuidle.C8.time
14370 ± 26% -96.1% 566.75 ± 22% cpuidle.C8.usage
824.75 ± 44% +2081.8% 17994 ± 36% interrupts.132:IR-PCI-MSI.2097153-edge.eth1-TxRx-0
88.25 ± 18% -30.6% 61.25 ± 33% interrupts.CPU0.CAL:Function_call_interrupts
3558 ± 7% +47.0% 5231 ± 25% interrupts.CPU0.RES:Rescheduling_interrupts
824.75 ± 44% +2081.8% 17994 ± 36% interrupts.CPU1.132:IR-PCI-MSI.2097153-edge.eth1-TxRx-0
3608 ± 9% +71.4% 6185 ± 34% interrupts.CPU1.RES:Rescheduling_interrupts
3215 ± 7% +52.4% 4899 ± 32% interrupts.CPU10.RES:Rescheduling_interrupts
3925 ± 2% +43.6% 5638 ± 16% interrupts.CPU11.RES:Rescheduling_interrupts
3293 ± 5% +47.8% 4868 ± 30% interrupts.CPU13.RES:Rescheduling_interrupts
3122 ± 3% +54.3% 4816 ± 32% interrupts.CPU14.RES:Rescheduling_interrupts
3224 ± 7% +52.7% 4922 ± 30% interrupts.CPU15.RES:Rescheduling_interrupts
4584 ± 3% +36.2% 6244 ± 14% interrupts.CPU3.RES:Rescheduling_interrupts
4141 ± 9% +35.3% 5601 ± 20% interrupts.CPU4.RES:Rescheduling_interrupts
3383 ± 7% +44.8% 4898 ± 30% interrupts.CPU5.RES:Rescheduling_interrupts
112.50 ± 26% -31.8% 76.75 ± 14% interrupts.CPU7.CAL:Function_call_interrupts
3530 ± 5% +43.8% 5077 ± 26% interrupts.CPU7.RES:Rescheduling_interrupts
3286 ± 7% +45.1% 4770 ± 33% interrupts.CPU8.RES:Rescheduling_interrupts
3105 ± 6% +52.3% 4729 ± 34% interrupts.CPU9.RES:Rescheduling_interrupts
59039 ± 6% +42.0% 83820 ± 26% interrupts.RES:Rescheduling_interrupts
94.11 ± 3% -7.6% 86.98 perf-stat.i.MPKI
5.753e+09 +2.8% 5.917e+09 perf-stat.i.branch-instructions
1.87 +0.0 1.92 perf-stat.i.branch-miss-rate%
1.068e+08 +5.6% 1.128e+08 perf-stat.i.branch-misses
0.08 ± 5% +0.0 0.09 ± 3% perf-stat.i.cache-miss-rate%
2.695e+09 ± 2% -4.9% 2.564e+09 perf-stat.i.cache-references
3431432 +3.0% 3535370 perf-stat.i.context-switches
2.13 -2.8% 2.07 perf-stat.i.cpi
8.248e+09 +3.0% 8.492e+09 perf-stat.i.dTLB-loads
0.00 ± 6% -0.0 0.00 ± 9% perf-stat.i.dTLB-store-miss-rate%
23161 ± 7% -32.9% 15547 perf-stat.i.dTLB-store-misses
4.99e+09 +3.0% 5.139e+09 perf-stat.i.dTLB-stores
87.07 +2.0 89.10 perf-stat.i.iTLB-load-miss-rate%
47009128 +14.3% 53717099 perf-stat.i.iTLB-load-misses
2.857e+10 +2.9% 2.941e+10 perf-stat.i.instructions
692.82 -12.8% 604.01 perf-stat.i.instructions-per-iTLB-miss
0.47 +2.8% 0.49 perf-stat.i.ipc
0.94 ± 12% +59.3% 1.49 perf-stat.i.metric.K/sec
1356 +2.0% 1382 perf-stat.i.metric.M/sec
94.34 ± 3% -7.6% 87.20 perf-stat.overall.MPKI
1.86 +0.1 1.91 perf-stat.overall.branch-miss-rate%
0.01 ± 2% +0.0 0.01 ± 2% perf-stat.overall.cache-miss-rate%
2.13 -2.8% 2.07 perf-stat.overall.cpi
0.00 ± 7% -0.0 0.00 ± 2% perf-stat.overall.dTLB-store-miss-rate%
87.15 +2.0 89.14 perf-stat.overall.iTLB-load-miss-rate%
607.94 -9.9% 547.59 perf-stat.overall.instructions-per-iTLB-miss
0.47 +2.8% 0.48 perf-stat.overall.ipc
0.00 ± 17% -0.0 0.00 ± 33% perf-stat.overall.node-store-miss-rate%
5.734e+09 +2.8% 5.897e+09 perf-stat.ps.branch-instructions
1.064e+08 +5.6% 1.124e+08 perf-stat.ps.branch-misses
2.686e+09 ± 2% -4.9% 2.556e+09 perf-stat.ps.cache-references
3419895 +3.0% 3523472 perf-stat.ps.context-switches
8.22e+09 +3.0% 8.464e+09 perf-stat.ps.dTLB-loads
23095 ± 7% -32.8% 15523 perf-stat.ps.dTLB-store-misses
4.973e+09 +3.0% 5.122e+09 perf-stat.ps.dTLB-stores
46851194 +14.3% 53536737 perf-stat.ps.iTLB-load-misses
2.848e+10 +2.9% 2.932e+10 perf-stat.ps.instructions
8.605e+12 +2.8% 8.844e+12 perf-stat.total.instructions
33.77 -1.3 32.45 perf-profile.calltrace.cycles-pp.sock_sendmsg.__sys_sendto.__x64_sys_sendto.do_syscall_64.entry_SYSCALL_64_after_hwframe
32.93 -1.3 31.66 perf-profile.calltrace.cycles-pp.udp_sendmsg.sock_sendmsg.__sys_sendto.__x64_sys_sendto.do_syscall_64
34.88 -1.0 33.90 perf-profile.calltrace.cycles-pp.__sys_sendto.__x64_sys_sendto.do_syscall_64.entry_SYSCALL_64_after_hwframe
35.02 -1.0 34.04 perf-profile.calltrace.cycles-pp.__x64_sys_sendto.do_syscall_64.entry_SYSCALL_64_after_hwframe
22.12 -0.9 21.25 perf-profile.calltrace.cycles-pp.ip_send_skb.udp_send_skb.udp_sendmsg.sock_sendmsg.__sys_sendto
22.43 -0.9 21.58 perf-profile.calltrace.cycles-pp.udp_send_skb.udp_sendmsg.sock_sendmsg.__sys_sendto.__x64_sys_sendto
21.59 -0.8 20.76 perf-profile.calltrace.cycles-pp.ip_output.ip_send_skb.udp_send_skb.udp_sendmsg.sock_sendmsg
19.99 -0.7 19.32 perf-profile.calltrace.cycles-pp.ip_finish_output2.ip_output.ip_send_skb.udp_send_skb.udp_sendmsg
13.47 -0.6 12.83 perf-profile.calltrace.cycles-pp.__netif_receive_skb_one_core.process_backlog.net_rx_action.__softirqentry_text_start.do_softirq_own_stack
15.55 -0.6 14.93 perf-profile.calltrace.cycles-pp.__softirqentry_text_start.do_softirq_own_stack.do_softirq.__local_bh_enable_ip.ip_finish_output2
14.20 -0.6 13.59 perf-profile.calltrace.cycles-pp.process_backlog.net_rx_action.__softirqentry_text_start.do_softirq_own_stack.do_softirq
12.74 -0.6 12.13 perf-profile.calltrace.cycles-pp.ip_rcv.__netif_receive_skb_one_core.process_backlog.net_rx_action.__softirqentry_text_start
14.69 -0.6 14.09 perf-profile.calltrace.cycles-pp.net_rx_action.__softirqentry_text_start.do_softirq_own_stack.do_softirq.__local_bh_enable_ip
16.00 -0.6 15.42 perf-profile.calltrace.cycles-pp.do_softirq.__local_bh_enable_ip.ip_finish_output2.ip_output.ip_send_skb
15.63 -0.6 15.07 perf-profile.calltrace.cycles-pp.do_softirq_own_stack.do_softirq.__local_bh_enable_ip.ip_finish_output2.ip_output
16.15 -0.6 15.60 perf-profile.calltrace.cycles-pp.__local_bh_enable_ip.ip_finish_output2.ip_output.ip_send_skb.udp_send_skb
10.69 -0.5 10.20 perf-profile.calltrace.cycles-pp.udp_unicast_rcv_skb.__udp4_lib_rcv.ip_protocol_deliver_rcu.ip_local_deliver_finish.ip_local_deliver
10.50 -0.5 10.01 perf-profile.calltrace.cycles-pp.udp_queue_rcv_one_skb.udp_unicast_rcv_skb.__udp4_lib_rcv.ip_protocol_deliver_rcu.ip_local_deliver_finish
11.82 -0.5 11.37 perf-profile.calltrace.cycles-pp.ip_protocol_deliver_rcu.ip_local_deliver_finish.ip_local_deliver.ip_rcv.__netif_receive_skb_one_core
4.93 -0.4 4.48 perf-profile.calltrace.cycles-pp.__ip_append_data.ip_make_skb.udp_sendmsg.sock_sendmsg.__sys_sendto
12.10 -0.4 11.67 perf-profile.calltrace.cycles-pp.ip_local_deliver.ip_rcv.__netif_receive_skb_one_core.process_backlog.net_rx_action
53.15 -0.4 52.73 perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe
53.56 -0.4 53.13 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe
11.53 -0.4 11.12 perf-profile.calltrace.cycles-pp.__udp4_lib_rcv.ip_protocol_deliver_rcu.ip_local_deliver_finish.ip_local_deliver.ip_rcv
11.89 -0.4 11.51 perf-profile.calltrace.cycles-pp.ip_local_deliver_finish.ip_local_deliver.ip_rcv.__netif_receive_skb_one_core.process_backlog
9.49 -0.4 9.13 perf-profile.calltrace.cycles-pp.__udp_enqueue_schedule_skb.udp_queue_rcv_one_skb.udp_unicast_rcv_skb.__udp4_lib_rcv.ip_protocol_deliver_rcu
7.77 -0.3 7.47 perf-profile.calltrace.cycles-pp.autoremove_wake_function.__wake_up_common.__wake_up_common_lock.sock_def_readable.__udp_enqueue_schedule_skb
8.83 -0.3 8.53 perf-profile.calltrace.cycles-pp.sock_def_readable.__udp_enqueue_schedule_skb.udp_queue_rcv_one_skb.udp_unicast_rcv_skb.__udp4_lib_rcv
6.86 -0.3 6.56 perf-profile.calltrace.cycles-pp.ip_make_skb.udp_sendmsg.sock_sendmsg.__sys_sendto.__x64_sys_sendto
8.52 -0.3 8.23 perf-profile.calltrace.cycles-pp.__wake_up_common_lock.sock_def_readable.__udp_enqueue_schedule_skb.udp_queue_rcv_one_skb.udp_unicast_rcv_skb
7.61 -0.3 7.32 perf-profile.calltrace.cycles-pp.try_to_wake_up.autoremove_wake_function.__wake_up_common.__wake_up_common_lock.sock_def_readable
7.99 -0.3 7.70 perf-profile.calltrace.cycles-pp.__wake_up_common.__wake_up_common_lock.sock_def_readable.__udp_enqueue_schedule_skb.udp_queue_rcv_one_skb
2.90 -0.2 2.68 ± 2% perf-profile.calltrace.cycles-pp.sock_alloc_send_pskb.__ip_append_data.ip_make_skb.udp_sendmsg.sock_sendmsg
0.76 ± 2% -0.2 0.56 ± 8% perf-profile.calltrace.cycles-pp.validate_xmit_skb.__dev_queue_xmit.ip_finish_output2.ip_output.ip_send_skb
0.85 ± 2% -0.2 0.68 ± 4% perf-profile.calltrace.cycles-pp._copy_from_iter_full.ip_generic_getfrag.__ip_append_data.ip_make_skb.udp_sendmsg
3.23 -0.2 3.08 perf-profile.calltrace.cycles-pp.__dev_queue_xmit.ip_finish_output2.ip_output.ip_send_skb.udp_send_skb
1.29 -0.1 1.17 ± 2% perf-profile.calltrace.cycles-pp.ip_generic_getfrag.__ip_append_data.ip_make_skb.udp_sendmsg.sock_sendmsg
1.03 -0.1 0.98 perf-profile.calltrace.cycles-pp._copy_to_iter.udp_recvmsg.inet_recvmsg.__sys_recvfrom.__x64_sys_recvfrom
1.38 ± 2% -0.1 1.33 perf-profile.calltrace.cycles-pp.select_task_rq_fair.try_to_wake_up.autoremove_wake_function.__wake_up_common.__wake_up_common_lock
0.66 -0.0 0.61 perf-profile.calltrace.cycles-pp.__check_object_size.udp_recvmsg.inet_recvmsg.__sys_recvfrom.__x64_sys_recvfrom
0.55 +0.0 0.58 perf-profile.calltrace.cycles-pp.check_preempt_curr.ttwu_do_wakeup.try_to_wake_up.autoremove_wake_function.__wake_up_common
0.78 +0.1 0.83 ± 3% perf-profile.calltrace.cycles-pp.__kmalloc_reserve.__alloc_skb.alloc_skb_with_frags.sock_alloc_send_pskb.__ip_append_data
0.69 ± 5% +0.1 0.74 perf-profile.calltrace.cycles-pp.select_idle_sibling.select_task_rq_fair.try_to_wake_up.autoremove_wake_function.__wake_up_common
0.54 ± 2% +0.1 0.61 ± 4% perf-profile.calltrace.cycles-pp.enqueue_to_backlog.netif_rx_internal.netif_rx.loopback_xmit.dev_hard_start_xmit
0.94 ± 2% +0.1 1.03 perf-profile.calltrace.cycles-pp.move_addr_to_user.__sys_recvfrom.__x64_sys_recvfrom.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.73 ± 3% +0.1 0.83 perf-profile.calltrace.cycles-pp.__consume_stateless_skb.udp_recvmsg.inet_recvmsg.__sys_recvfrom.__x64_sys_recvfrom
1.75 +0.1 1.87 ± 2% perf-profile.calltrace.cycles-pp.loopback_xmit.dev_hard_start_xmit.__dev_queue_xmit.ip_finish_output2.ip_output
1.86 +0.1 1.99 ± 2% perf-profile.calltrace.cycles-pp.dev_hard_start_xmit.__dev_queue_xmit.ip_finish_output2.ip_output.ip_send_skb
0.88 ± 3% +0.1 1.02 ± 2% perf-profile.calltrace.cycles-pp.netif_rx_internal.netif_rx.loopback_xmit.dev_hard_start_xmit.__dev_queue_xmit
7.71 +0.2 7.87 perf-profile.calltrace.cycles-pp.schedule.schedule_timeout.__skb_wait_for_more_packets.__skb_recv_udp.udp_recvmsg
0.89 ± 2% +0.2 1.05 ± 2% perf-profile.calltrace.cycles-pp.netif_rx.loopback_xmit.dev_hard_start_xmit.__dev_queue_xmit.ip_finish_output2
1.59 +0.2 1.76 perf-profile.calltrace.cycles-pp.__ip_make_skb.ip_make_skb.udp_sendmsg.sock_sendmsg.__sys_sendto
10.88 +0.3 11.15 perf-profile.calltrace.cycles-pp.__skb_recv_udp.udp_recvmsg.inet_recvmsg.__sys_recvfrom.__x64_sys_recvfrom
43.59 +0.3 43.93 perf-profile.calltrace.cycles-pp.secondary_startup_64
17.36 +0.5 17.82 perf-profile.calltrace.cycles-pp.__sys_recvfrom.__x64_sys_recvfrom.do_syscall_64.entry_SYSCALL_64_after_hwframe
17.50 +0.5 17.97 perf-profile.calltrace.cycles-pp.__x64_sys_recvfrom.do_syscall_64.entry_SYSCALL_64_after_hwframe
15.08 +0.5 15.57 perf-profile.calltrace.cycles-pp.udp_recvmsg.inet_recvmsg.__sys_recvfrom.__x64_sys_recvfrom.do_syscall_64
15.32 +0.5 15.84 perf-profile.calltrace.cycles-pp.inet_recvmsg.__sys_recvfrom.__x64_sys_recvfrom.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.00 +0.6 0.57 ± 4% perf-profile.calltrace.cycles-pp.kmem_cache_alloc_node.__alloc_skb.alloc_skb_with_frags.sock_alloc_send_pskb.__ip_append_data
22.65 +0.7 23.32 perf-profile.calltrace.cycles-pp.poll_idle.cpuidle_enter_state.cpuidle_enter.do_idle.cpu_startup_entry
28.04 +1.0 29.05 perf-profile.calltrace.cycles-pp.cpuidle_enter.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64
27.99 +1.0 29.02 perf-profile.calltrace.cycles-pp.cpuidle_enter_state.cpuidle_enter.do_idle.cpu_startup_entry.start_secondary
33.79 -1.3 32.46 perf-profile.children.cycles-pp.sock_sendmsg
32.97 -1.3 31.69 perf-profile.children.cycles-pp.udp_sendmsg
35.03 -1.0 34.05 perf-profile.children.cycles-pp.__x64_sys_sendto
34.89 -1.0 33.91 perf-profile.children.cycles-pp.__sys_sendto
22.13 -0.9 21.26 perf-profile.children.cycles-pp.ip_send_skb
22.44 -0.9 21.58 perf-profile.children.cycles-pp.udp_send_skb
21.59 -0.8 20.78 perf-profile.children.cycles-pp.ip_output
20.00 -0.7 19.34 perf-profile.children.cycles-pp.ip_finish_output2
13.49 -0.6 12.84 perf-profile.children.cycles-pp.__netif_receive_skb_one_core
12.76 -0.6 12.14 perf-profile.children.cycles-pp.ip_rcv
15.60 -0.6 14.98 perf-profile.children.cycles-pp.__softirqentry_text_start
14.21 -0.6 13.60 perf-profile.children.cycles-pp.process_backlog
14.70 -0.6 14.10 perf-profile.children.cycles-pp.net_rx_action
16.01 -0.6 15.43 perf-profile.children.cycles-pp.do_softirq
15.65 -0.6 15.08 perf-profile.children.cycles-pp.do_softirq_own_stack
16.37 -0.5 15.86 perf-profile.children.cycles-pp.__local_bh_enable_ip
10.53 -0.5 10.03 perf-profile.children.cycles-pp.udp_queue_rcv_one_skb
10.69 -0.5 10.20 perf-profile.children.cycles-pp.udp_unicast_rcv_skb
11.84 -0.5 11.39 perf-profile.children.cycles-pp.ip_protocol_deliver_rcu
4.94 -0.5 4.49 perf-profile.children.cycles-pp.__ip_append_data
53.23 -0.4 52.79 perf-profile.children.cycles-pp.do_syscall_64
53.59 -0.4 53.17 perf-profile.children.cycles-pp.entry_SYSCALL_64_after_hwframe
11.55 -0.4 11.14 perf-profile.children.cycles-pp.__udp4_lib_rcv
12.11 -0.4 11.70 perf-profile.children.cycles-pp.ip_local_deliver
11.89 -0.4 11.52 perf-profile.children.cycles-pp.ip_local_deliver_finish
9.50 -0.4 9.13 perf-profile.children.cycles-pp.__udp_enqueue_schedule_skb
8.84 -0.3 8.53 perf-profile.children.cycles-pp.sock_def_readable
7.78 -0.3 7.47 perf-profile.children.cycles-pp.autoremove_wake_function
6.88 -0.3 6.58 perf-profile.children.cycles-pp.ip_make_skb
7.62 -0.3 7.33 perf-profile.children.cycles-pp.try_to_wake_up
8.52 -0.3 8.23 perf-profile.children.cycles-pp.__wake_up_common_lock
7.99 -0.3 7.71 perf-profile.children.cycles-pp.__wake_up_common
0.79 ± 2% -0.2 0.57 ± 8% perf-profile.children.cycles-pp.validate_xmit_skb
2.90 -0.2 2.69 ± 2% perf-profile.children.cycles-pp.sock_alloc_send_pskb
0.88 ± 2% -0.2 0.70 ± 4% perf-profile.children.cycles-pp._copy_from_iter_full
3.26 -0.2 3.09 perf-profile.children.cycles-pp.__dev_queue_xmit
0.48 ± 5% -0.1 0.33 ± 5% perf-profile.children.cycles-pp.sock_recvmsg
0.42 ± 3% -0.1 0.28 ± 9% perf-profile.children.cycles-pp.nf_hook_slow
0.44 ± 6% -0.1 0.30 ± 6% perf-profile.children.cycles-pp.security_socket_recvmsg
0.16 ± 9% -0.1 0.03 ±100% perf-profile.children.cycles-pp.netdev_core_pick_tx
1.30 -0.1 1.17 ± 2% perf-profile.children.cycles-pp.ip_generic_getfrag
0.65 -0.1 0.52 ± 2% perf-profile.children.cycles-pp.aa_sk_perm
0.29 ± 6% -0.1 0.17 ± 5% perf-profile.children.cycles-pp.ip_rcv_finish
0.29 ± 3% -0.1 0.17 ± 2% perf-profile.children.cycles-pp.rb_next
0.22 ± 11% -0.1 0.11 ± 14% perf-profile.children.cycles-pp.xfrm_lookup_with_ifid
0.23 ± 11% -0.1 0.13 ± 14% perf-profile.children.cycles-pp.xfrm_lookup_route
0.37 ± 3% -0.1 0.27 ± 5% perf-profile.children.cycles-pp.copyin
0.13 ± 5% -0.1 0.03 ±100% perf-profile.children.cycles-pp.cpuidle_not_available
0.40 ± 4% -0.1 0.30 perf-profile.children.cycles-pp.__ip_local_out
3.82 -0.1 3.73 perf-profile.children.cycles-pp.ttwu_do_activate
3.76 -0.1 3.68 perf-profile.children.cycles-pp.activate_task
0.20 ± 7% -0.1 0.11 ± 9% perf-profile.children.cycles-pp.pm_qos_request
0.44 ± 2% -0.1 0.35 ± 9% perf-profile.children.cycles-pp.__ksize
0.23 ± 8% -0.1 0.15 ± 8% perf-profile.children.cycles-pp.rcu_qs
0.39 ± 4% -0.1 0.31 ± 10% perf-profile.children.cycles-pp.__ip_finish_output
0.27 ± 6% -0.1 0.19 ± 7% perf-profile.children.cycles-pp.rcu_note_context_switch
0.28 ± 8% -0.1 0.21 ± 2% perf-profile.children.cycles-pp.__hrtimer_next_event_base
0.17 ± 7% -0.1 0.10 ± 4% perf-profile.children.cycles-pp.ip_rcv_finish_core
0.11 ± 4% -0.1 0.05 ± 58% perf-profile.children.cycles-pp.apparmor_ip_postroute
0.21 ± 6% -0.1 0.15 ± 3% perf-profile.children.cycles-pp.ipv4_pktinfo_prepare
0.14 ± 20% -0.1 0.08 ± 34% perf-profile.children.cycles-pp.tick_check_broadcast_expired
0.43 ± 5% -0.1 0.36 ± 2% perf-profile.children.cycles-pp.ip_local_out
0.13 ± 16% -0.1 0.07 ± 12% perf-profile.children.cycles-pp.tick_nohz_idle_got_tick
0.45 ± 6% -0.1 0.39 ± 4% perf-profile.children.cycles-pp.security_socket_sendmsg
0.23 ± 3% -0.1 0.17 ± 10% perf-profile.children.cycles-pp.account_entity_dequeue
0.11 ± 7% -0.1 0.06 ± 15% perf-profile.children.cycles-pp.skb_clone_tx_timestamp
0.51 ± 5% -0.1 0.46 perf-profile.children.cycles-pp.irqtime_account_irq
1.05 -0.0 1.00 perf-profile.children.cycles-pp._copy_to_iter
0.14 ± 8% -0.0 0.09 ± 21% perf-profile.children.cycles-pp.apparmor_ipv4_postroute
0.14 ± 5% -0.0 0.10 ± 11% perf-profile.children.cycles-pp.raw_local_deliver
0.10 ± 8% -0.0 0.05 ± 9% perf-profile.children.cycles-pp.__sk_mem_raise_allocated
0.25 ± 5% -0.0 0.21 ± 5% perf-profile.children.cycles-pp.security_sock_rcv_skb
0.43 -0.0 0.39 ± 3% perf-profile.children.cycles-pp.account_entity_enqueue
0.24 ± 5% -0.0 0.21 ± 5% perf-profile.children.cycles-pp.menu_reflect
0.17 ± 4% -0.0 0.14 ± 3% perf-profile.children.cycles-pp.ip_rcv_core
0.08 ± 10% -0.0 0.06 ± 7% perf-profile.children.cycles-pp.skb_pull_rcsum
0.12 ± 4% -0.0 0.10 ± 7% perf-profile.children.cycles-pp.deactivate_task
0.17 ± 2% -0.0 0.15 ± 5% perf-profile.children.cycles-pp.update_irq_load_avg
0.10 ± 11% -0.0 0.07 ± 15% perf-profile.children.cycles-pp.skb_network_protocol
0.15 -0.0 0.14 ± 3% perf-profile.children.cycles-pp.dst_release
0.08 ± 5% +0.0 0.10 ± 5% perf-profile.children.cycles-pp.pm_qos_read_value
0.15 +0.0 0.17 ± 2% perf-profile.children.cycles-pp.__calc_delta
0.15 ± 5% +0.0 0.17 ± 2% perf-profile.children.cycles-pp.ip4_datagram_release_cb
0.10 ± 8% +0.0 0.12 ± 3% perf-profile.children.cycles-pp.perf_exclude_event
0.11 ± 7% +0.0 0.13 ± 6% perf-profile.children.cycles-pp.kfree
0.56 ± 2% +0.0 0.58 perf-profile.children.cycles-pp.check_preempt_curr
0.15 ± 7% +0.0 0.17 ± 5% perf-profile.children.cycles-pp.__update_idle_core
0.17 ± 7% +0.0 0.20 ± 5% perf-profile.children.cycles-pp.set_next_task_idle
0.19 ± 6% +0.0 0.21 ± 2% perf-profile.children.cycles-pp.ip_setup_cork
0.07 ± 7% +0.0 0.09 ± 8% perf-profile.children.cycles-pp.security_sk_classify_flow
0.27 ± 4% +0.0 0.30 ± 2% perf-profile.children.cycles-pp.__check_heap_object
0.69 ± 3% +0.0 0.73 ± 2% perf-profile.children.cycles-pp.tick_nohz_idle_exit
0.09 ± 17% +0.0 0.12 ± 3% perf-profile.children.cycles-pp.tick_nohz_idle_retain_tick
0.70 ± 2% +0.0 0.74 ± 2% perf-profile.children.cycles-pp.__kmalloc_node_track_caller
0.46 ± 3% +0.0 0.49 ± 2% perf-profile.children.cycles-pp.tick_nohz_idle_enter
0.15 ± 2% +0.0 0.19 ± 4% perf-profile.children.cycles-pp.__skb_try_recv_from_queue
0.01 ±173% +0.0 0.05 ± 9% perf-profile.children.cycles-pp.receiver_wake_function
0.46 ± 2% +0.0 0.50 ± 4% perf-profile.children.cycles-pp.__fget_light
0.61 ± 2% +0.0 0.66 ± 2% perf-profile.children.cycles-pp.copy_user_generic_unrolled
0.40 ± 3% +0.1 0.45 ± 5% perf-profile.children.cycles-pp.release_sock
0.27 ± 3% +0.1 0.33 ± 2% perf-profile.children.cycles-pp.ipv4_mtu
0.08 ± 15% +0.1 0.14 ± 19% perf-profile.children.cycles-pp.get_cpu_device
0.05 ± 67% +0.1 0.11 ± 9% perf-profile.children.cycles-pp.skb_consume_udp
0.34 ± 6% +0.1 0.40 ± 3% perf-profile.children.cycles-pp.cpus_share_cache
0.31 ± 7% +0.1 0.37 ± 6% perf-profile.children.cycles-pp.kmem_cache_free
0.26 ± 4% +0.1 0.32 ± 3% perf-profile.children.cycles-pp.ktime_get_with_offset
0.00 +0.1 0.06 ± 13% perf-profile.children.cycles-pp.memcg_kmem_put_cache
0.55 ± 2% +0.1 0.62 ± 4% perf-profile.children.cycles-pp.enqueue_to_backlog
0.24 ± 7% +0.1 0.32 ± 2% perf-profile.children.cycles-pp.switch_fpu_return
0.96 ± 2% +0.1 1.05 perf-profile.children.cycles-pp.move_addr_to_user
0.00 +0.1 0.09 ± 7% perf-profile.children.cycles-pp.inet_send_prepare
0.74 ± 3% +0.1 0.84 perf-profile.children.cycles-pp.__consume_stateless_skb
0.32 ± 2% +0.1 0.42 ± 8% perf-profile.children.cycles-pp._copy_from_user
0.49 ± 4% +0.1 0.59 ± 4% perf-profile.children.cycles-pp.kmem_cache_alloc_node
0.59 ± 2% +0.1 0.71 perf-profile.children.cycles-pp.___perf_sw_event
1.76 +0.1 1.88 ± 2% perf-profile.children.cycles-pp.loopback_xmit
0.57 +0.1 0.70 ± 3% perf-profile.children.cycles-pp.sockfd_lookup_light
1.86 +0.1 2.00 ± 2% perf-profile.children.cycles-pp.dev_hard_start_xmit
0.88 ± 3% +0.1 1.03 ± 2% perf-profile.children.cycles-pp.netif_rx_internal
7.72 +0.2 7.88 perf-profile.children.cycles-pp.schedule
0.90 ± 3% +0.2 1.05 ± 2% perf-profile.children.cycles-pp.netif_rx
0.39 ± 2% +0.2 0.55 ± 7% perf-profile.children.cycles-pp.move_addr_to_kernel
1.59 +0.2 1.77 perf-profile.children.cycles-pp.__ip_make_skb
10.91 +0.3 11.18 perf-profile.children.cycles-pp.__skb_recv_udp
43.59 +0.3 43.93 perf-profile.children.cycles-pp.secondary_startup_64
43.59 +0.3 43.93 perf-profile.children.cycles-pp.cpu_startup_entry
43.50 +0.3 43.84 perf-profile.children.cycles-pp.do_idle
17.52 +0.5 17.99 perf-profile.children.cycles-pp.__x64_sys_recvfrom
17.38 +0.5 17.84 perf-profile.children.cycles-pp.__sys_recvfrom
15.10 +0.5 15.59 perf-profile.children.cycles-pp.udp_recvmsg
15.33 +0.5 15.87 perf-profile.children.cycles-pp.inet_recvmsg
22.67 +0.7 23.34 perf-profile.children.cycles-pp.poll_idle
29.86 +0.9 30.79 perf-profile.children.cycles-pp.cpuidle_enter
29.81 +1.0 30.76 perf-profile.children.cycles-pp.cpuidle_enter_state
0.41 ± 4% -0.2 0.17 ± 4% perf-profile.self.cycles-pp.sock_alloc_send_pskb
0.43 ± 6% -0.2 0.21 ± 5% perf-profile.self.cycles-pp.validate_xmit_skb
1.07 ± 4% -0.2 0.88 ± 3% perf-profile.self.cycles-pp.do_idle
1.06 -0.1 0.92 ± 3% perf-profile.self.cycles-pp.menu_select
0.64 ± 9% -0.1 0.49 ± 2% perf-profile.self.cycles-pp.ip_route_output_key_hash_rcu
0.50 ± 2% -0.1 0.38 ± 4% perf-profile.self.cycles-pp.aa_sk_perm
0.18 ± 4% -0.1 0.08 ± 10% perf-profile.self.cycles-pp.tick_nohz_get_sleep_length
0.21 ± 12% -0.1 0.11 ± 16% perf-profile.self.cycles-pp.xfrm_lookup_with_ifid
0.27 ± 4% -0.1 0.16 ± 2% perf-profile.self.cycles-pp.rb_next
0.27 ± 4% -0.1 0.18 ± 4% perf-profile.self.cycles-pp.__alloc_skb
0.20 ± 5% -0.1 0.10 ± 8% perf-profile.self.cycles-pp.__ip_local_out
0.12 ± 6% -0.1 0.03 ±100% perf-profile.self.cycles-pp.cpuidle_not_available
0.30 ± 8% -0.1 0.21 ± 14% perf-profile.self.cycles-pp.__ip_finish_output
0.19 ± 5% -0.1 0.11 ± 7% perf-profile.self.cycles-pp.pm_qos_request
0.23 ± 7% -0.1 0.14 ± 10% perf-profile.self.cycles-pp.rcu_qs
0.54 ± 4% -0.1 0.46 ± 3% perf-profile.self.cycles-pp.select_task_rq_fair
0.19 ± 6% -0.1 0.11 ± 3% perf-profile.self.cycles-pp.copyin
0.40 ± 7% -0.1 0.33 ± 6% perf-profile.self.cycles-pp.try_to_wake_up
0.43 ± 2% -0.1 0.35 ± 9% perf-profile.self.cycles-pp.__ksize
0.28 ± 6% -0.1 0.20 ± 4% perf-profile.self.cycles-pp._copy_from_iter_full
0.23 ± 9% -0.1 0.16 ± 5% perf-profile.self.cycles-pp.irqtime_account_irq
0.14 ± 21% -0.1 0.07 ± 30% perf-profile.self.cycles-pp.tick_check_broadcast_expired
0.13 ± 15% -0.1 0.07 ± 7% perf-profile.self.cycles-pp.tick_nohz_idle_got_tick
0.11 ± 6% -0.1 0.05 ± 58% perf-profile.self.cycles-pp.apparmor_ip_postroute
0.16 ± 7% -0.1 0.10 ± 4% perf-profile.self.cycles-pp.ip_rcv_finish_core
0.20 ± 9% -0.1 0.14 ± 5% perf-profile.self.cycles-pp.ip_rcv
0.22 ± 3% -0.1 0.17 ± 10% perf-profile.self.cycles-pp.account_entity_dequeue
0.11 ± 15% -0.1 0.06 ± 11% perf-profile.self.cycles-pp.ip_rcv_finish
0.11 ± 8% -0.1 0.06 ± 15% perf-profile.self.cycles-pp.skb_clone_tx_timestamp
0.21 ± 7% -0.0 0.17 ± 9% perf-profile.self.cycles-pp.tick_nohz_next_event
0.18 ± 6% -0.0 0.13 ± 5% perf-profile.self.cycles-pp.nf_hook_slow
0.17 ± 4% -0.0 0.13 ± 10% perf-profile.self.cycles-pp.ip_make_skb
0.16 ± 5% -0.0 0.12 perf-profile.self.cycles-pp.security_sock_rcv_skb
0.14 ± 15% -0.0 0.10 ± 10% perf-profile.self.cycles-pp.inet_sendmsg
0.21 ± 7% -0.0 0.17 ± 7% perf-profile.self.cycles-pp.udp_queue_rcv_one_skb
0.10 ± 5% -0.0 0.05 ± 9% perf-profile.self.cycles-pp.__sk_mem_raise_allocated
0.12 ± 8% -0.0 0.09 ± 17% perf-profile.self.cycles-pp.apparmor_ipv4_postroute
0.62 ± 2% -0.0 0.59 ± 2% perf-profile.self.cycles-pp.__ip_append_data
0.21 ± 4% -0.0 0.17 ± 4% perf-profile.self.cycles-pp.__netif_receive_skb_one_core
0.40 ± 3% -0.0 0.37 ± 2% perf-profile.self.cycles-pp.account_entity_enqueue
0.18 ± 3% -0.0 0.15 ± 7% perf-profile.self.cycles-pp._copy_to_iter
0.13 ± 8% -0.0 0.10 ± 9% perf-profile.self.cycles-pp.raw_local_deliver
0.24 ± 6% -0.0 0.21 ± 2% perf-profile.self.cycles-pp.sock_wfree
0.08 ± 10% -0.0 0.05 ± 8% perf-profile.self.cycles-pp.skb_pull_rcsum
0.17 ± 4% -0.0 0.14 ± 3% perf-profile.self.cycles-pp.ip_rcv_core
0.08 ± 13% -0.0 0.06 ± 11% perf-profile.self.cycles-pp.skb_network_protocol
0.17 ± 4% -0.0 0.14 ± 9% perf-profile.self.cycles-pp.update_irq_load_avg
0.13 ± 8% -0.0 0.11 ± 6% perf-profile.self.cycles-pp.cpu_startup_entry
0.22 ± 3% -0.0 0.20 ± 2% perf-profile.self.cycles-pp.ip_local_deliver
0.12 ± 5% -0.0 0.10 ± 7% perf-profile.self.cycles-pp.deactivate_task
0.10 ± 5% +0.0 0.11 ± 3% perf-profile.self.cycles-pp.perf_exclude_event
0.08 ± 5% +0.0 0.10 ± 5% perf-profile.self.cycles-pp.pm_qos_read_value
0.07 ± 7% +0.0 0.08 ± 5% perf-profile.self.cycles-pp._cond_resched
0.10 ± 8% +0.0 0.12 ± 7% perf-profile.self.cycles-pp.kfree
0.32 ± 2% +0.0 0.34 perf-profile.self.cycles-pp.switch_mm_irqs_off
0.06 ± 6% +0.0 0.09 ± 5% perf-profile.self.cycles-pp.security_sk_classify_flow
0.12 ± 10% +0.0 0.15 ± 3% perf-profile.self.cycles-pp.ktime_get
0.27 +0.0 0.30 ± 2% perf-profile.self.cycles-pp.move_addr_to_user
0.23 ± 6% +0.0 0.26 ± 3% perf-profile.self.cycles-pp.copy_user_enhanced_fast_string
0.14 ± 8% +0.0 0.17 ± 6% perf-profile.self.cycles-pp.__update_idle_core
0.08 ± 14% +0.0 0.11 ± 4% perf-profile.self.cycles-pp.schedule_idle
0.35 +0.0 0.39 ± 4% perf-profile.self.cycles-pp.__softirqentry_text_start
0.08 ± 6% +0.0 0.11 ± 4% perf-profile.self.cycles-pp.__skb_try_recv_from_queue
0.26 ± 5% +0.0 0.29 ± 3% perf-profile.self.cycles-pp.__check_heap_object
0.35 ± 8% +0.0 0.39 ± 5% perf-profile.self.cycles-pp.__sys_sendto
0.09 ± 17% +0.0 0.12 ± 3% perf-profile.self.cycles-pp.tick_nohz_idle_retain_tick
0.13 ± 13% +0.0 0.17 perf-profile.self.cycles-pp.__hrtimer_next_event_base
0.01 ±173% +0.0 0.05 ± 9% perf-profile.self.cycles-pp.receiver_wake_function
0.22 ± 3% +0.0 0.26 ± 9% perf-profile.self.cycles-pp.loopback_xmit
0.14 ± 12% +0.0 0.18 ± 10% perf-profile.self.cycles-pp.kmem_cache_free
0.09 ± 11% +0.0 0.14 ± 8% perf-profile.self.cycles-pp.do_softirq_own_stack
0.14 ± 5% +0.0 0.19 ± 6% perf-profile.self.cycles-pp.ip_route_output_key_hash
0.21 ± 5% +0.0 0.26 ± 7% perf-profile.self.cycles-pp.check_preempt_curr
0.45 +0.0 0.50 ± 4% perf-profile.self.cycles-pp.__fget_light
0.26 ± 3% +0.0 0.31 ± 2% perf-profile.self.cycles-pp.ipv4_mtu
0.22 ± 3% +0.1 0.27 ± 12% perf-profile.self.cycles-pp.newidle_balance
0.59 ± 4% +0.1 0.64 ± 4% perf-profile.self.cycles-pp.ip_finish_output2
0.11 ± 24% +0.1 0.16 ± 4% perf-profile.self.cycles-pp.tick_nohz_idle_enter
0.08 ± 15% +0.1 0.13 ± 22% perf-profile.self.cycles-pp.get_cpu_device
0.10 ± 17% +0.1 0.16 ± 2% perf-profile.self.cycles-pp.ktime_get_with_offset
0.06 ± 6% +0.1 0.12 ± 17% perf-profile.self.cycles-pp._copy_from_user
0.14 ± 6% +0.1 0.19 ± 8% perf-profile.self.cycles-pp.tick_nohz_tick_stopped
0.05 ± 64% +0.1 0.11 ± 10% perf-profile.self.cycles-pp.skb_consume_udp
0.00 +0.1 0.06 ± 7% perf-profile.self.cycles-pp.ip_route_output_flow
0.23 ± 6% +0.1 0.29 ± 6% perf-profile.self.cycles-pp.inet_recvmsg
0.32 ± 7% +0.1 0.38 ± 3% perf-profile.self.cycles-pp.cpus_share_cache
0.32 ± 4% +0.1 0.39 ± 7% perf-profile.self.cycles-pp.enqueue_to_backlog
0.00 +0.1 0.07 ± 13% perf-profile.self.cycles-pp.ip_local_out
0.54 ± 2% +0.1 0.60 ± 2% perf-profile.self.cycles-pp.__skb_wait_for_more_packets
0.09 ± 13% +0.1 0.16 ± 5% perf-profile.self.cycles-pp.sockfd_lookup_light
0.35 ± 4% +0.1 0.42 perf-profile.self.cycles-pp.__local_bh_enable_ip
0.10 ± 12% +0.1 0.17 ± 4% perf-profile.self.cycles-pp.alloc_skb_with_frags
0.29 ± 7% +0.1 0.36 ± 3% perf-profile.self.cycles-pp.kmem_cache_alloc_node
0.57 ± 4% +0.1 0.65 ± 2% perf-profile.self.cycles-pp.udp_recvmsg
0.00 +0.1 0.08 ± 8% perf-profile.self.cycles-pp.inet_send_prepare
0.38 ± 4% +0.1 0.46 perf-profile.self.cycles-pp.ip_output
0.20 ± 7% +0.1 0.28 ± 2% perf-profile.self.cycles-pp.switch_fpu_return
0.15 ± 8% +0.1 0.24 ± 6% perf-profile.self.cycles-pp.cpuidle_enter_state
0.03 ±100% +0.1 0.11 ± 7% perf-profile.self.cycles-pp.ip_local_deliver_finish
0.47 ± 5% +0.1 0.57 perf-profile.self.cycles-pp.___perf_sw_event
0.30 ± 3% +0.1 0.45 ± 4% perf-profile.self.cycles-pp.__ip_make_skb
22.44 +0.7 23.11 perf-profile.self.cycles-pp.poll_idle
netperf.Throughput_tps
118000 +------------------------------------------------------------------+
| O |
116000 |-+ O O O |
| O O |
| O O O |
114000 |-+ |
| O |
112000 |-+ O O O O |
| O O O |
110000 |-+ O |
| O O O O O |
| +..+..+..+..+ + |
108000 |-.+.. + + .. |
|. + + |
106000 +------------------------------------------------------------------+
[*] bisect-good sample
[O] bisect-bad sample
Disclaimer:
Results have been estimated based on internal Intel analysis and are provided
for informational purposes only. Any difference in system hardware or software
design or configuration may affect actual performance.
Thanks,
Rong Chen
View attachment "config-5.4.0-00002-g0be0ee71816b2" of type "text/plain" (152100 bytes)
View attachment "job-script" of type "text/plain" (7741 bytes)
View attachment "job.yaml" of type "text/plain" (5280 bytes)
View attachment "reproduce" of type "text/plain" (675 bytes)
Powered by blists - more mailing lists