3:02 a.m.
Greeting,
FYI, we noticed a +23.0% improvement of netperf.Throughput_Mbps due to commit:
commit 79404c5a5c66481aa55c0cae685e49e0f44a0479 ("mm: page_alloc: High-order per-cpu
page allocator")
https://git.kernel.org/pub/scm/linux/kernel/git/mel/linux.git
mm-pagealloc-highorder-percpu-v3r1
+------------------+-----------------------------------------------------------------+
| testcase: change | netperf: netperf.Throughput_Mbps 23.0% improvement |
| test machine | 8 threads Intel(R) Atom(TM) CPU C2750 @ 2.40GHz with 16G memory |
| test parameters | cluster=cs-localhost |
| | cpufreq_governor=performance |
| | ip=ipv4 |
| | nr_threads=25% |
| | runtime=300s |
| | send_size=10K |
| | test=SCTP_STREAM_MANY |
+------------------+-----------------------------------------------------------------+
test-description: The motivation behind this suite is to exercise functions and regions of
the mm/ of the Linux kernel which are of interest to us.
test-url: https://git.kernel.org/cgit/linux/kernel/git/wfg/vm-scalability.git/
In addition to that, the commit also has significant impact on the following tests:
Details are as below:
-------------------------------------------------------------------------------------------------->
To reproduce:
git clone git://git.kernel.org/pub/scm/linux/kernel/git/wfg/lkp-tests.git
cd lkp-tests
bin/lkp install job.yaml # job file is attached in this email
bin/lkp run job.yaml
***************************************************************************************************
lkp-avoton3: 8 threads Intel(R) Atom(TM) CPU C2750 @ 2.40GHz with 16G memory
=========================================================================================
cluster/compiler/cpufreq_governor/ip/kconfig/nr_threads/rootfs/runtime/send_size/tbox_group/test/testcase:
cs-localhost/gcc-6/performance/ipv4/x86_64-rhel-7.2/25%/debian-x86_64-2016-08-31.cgz/300s/10K/lkp-avoton3/SCTP_STREAM_MANY/netperf
commit:
v4.9-rc5
79404c5a5c ("mm: page_alloc: High-order per-cpu page allocator")
v4.9-rc5 79404c5a5c66481aa55c0cae68
---------------- --------------------------
fail:runs %reproduction fail:runs
| | |
%stddev %change %stddev
\ | \
4124 ± 6% +23.0% 5074 ± 1% netperf.Throughput_Mbps
165.20 ± 5% +8.2% 178.75 ± 1% netperf.time.percent_of_cpu_this_job_got
490.26 ± 5% +8.0% 529.39 ± 0% netperf.time.system_time
56253 ± 8% +20.0% 67530 ± 12% meminfo.DirectMap4k
4332800 ± 6% +22.8% 5319765 ± 1% softirqs.NET_RX
57290 ± 6% +22.7% 70278 ± 1% vmstat.system.cs
8725309 ± 7% +23.5% 10773756 ± 1% cpuidle.C1-AVN.usage
18968 ± 6% +22.3% 23200 ± 1% cpuidle.POLL.usage
323.15 ± 0% -100.0% 0.00 ± -1% uptime.boot
1899 ± 1% -100.0% 0.00 ± -1% uptime.idle
31.90 ± 5% +10.7% 35.31 ± 0% turbostat.%Busy
817.80 ± 7% +12.3% 918.25 ± 0% turbostat.Avg_MHz
38.00 ± 8% +20.4% 45.75 ± 3% turbostat.CoreTmp
39068988 ± 6% +22.9% 48019605 ± 1% proc-vmstat.numa_hit
39068960 ± 6% +22.9% 48019576 ± 1% proc-vmstat.numa_local
2.243e+08 ± 6% +23.1% 2.761e+08 ± 1% proc-vmstat.pgalloc_normal
2.243e+08 ± 6% -43.1% 1.276e+08 ± 1% proc-vmstat.pgfree
87247 ± 7% +11.9% 97636 ± 0% sched_debug.cfs_rq:/.min_vruntime.avg
46310 ± 11% +50.1% 69504 ± 2% sched_debug.cfs_rq:/.min_vruntime.min
1.47 ± 9% -18.8% 1.19 ± 16% sched_debug.cpu.clock.stddev
1.47 ± 9% -18.9% 1.19 ± 16% sched_debug.cpu.clock_task.stddev
55358 ± 9% +30.4% 72172 ± 3% sched_debug.cpu.nr_load_updates.min
12494 ± 9% -32.1% 8488 ± 17% sched_debug.cpu.nr_load_updates.stddev
1074035 ± 7% +23.3% 1324227 ± 0% sched_debug.cpu.nr_switches.avg
1577603 ± 10% +19.3% 1882212 ± 5% sched_debug.cpu.nr_switches.max
8.859e+10 ± 6% +17.2% 1.039e+11 ± 0% perf-stat.branch-instructions
4.18 ± 0% -4.5% 3.99 ± 0% perf-stat.branch-miss-rate%
3.699e+09 ± 5% +12.0% 4.145e+09 ± 0% perf-stat.branch-misses
1.183e+11 ± 8% +11.9% 1.324e+11 ± 0% perf-stat.cache-references
17412166 ± 6% +22.7% 21360182 ± 1% perf-stat.context-switches
1.94e+12 ± 7% +13.3% 2.198e+12 ± 0% perf-stat.cpu-cycles
18000 ± 19% -26.1% 13294 ± 2% perf-stat.cpu-migrations
0.91 ± 0% -4.2% 0.87 ± 0% perf-stat.iTLB-load-miss-rate%
4.303e+09 ± 5% +12.8% 4.853e+09 ± 1% perf-stat.iTLB-load-misses
4.698e+11 ± 6% +17.7% 5.528e+11 ± 1% perf-stat.iTLB-loads
4.609e+11 ± 5% +19.9% 5.526e+11 ± 1% perf-stat.instructions
107.14 ± 1% +6.3% 113.88 ± 0% perf-stat.instructions-per-iTLB-miss
0.24 ± 2% +5.6% 0.25 ± 0% perf-stat.ipc
7.80 ± 7% +32.7% 10.34 ± 4%
perf-profile.calltrace.cycles-pp.___sys_recvmsg.__sys_recvmsg.sys_recvmsg.entry_SYSCALL_64_fastpath
2.02 ± 12% -20.0% 1.61 ± 6%
perf-profile.calltrace.cycles-pp.__alloc_skb._sctp_make_chunk.sctp_make_datafrag_empty.sctp_datamsg_from_user.sctp_sendmsg
1.53 ± 14% -35.6% 0.98 ± 6%
perf-profile.calltrace.cycles-pp.__kmalloc_node_track_caller.__kmalloc_reserve.__alloc_skb._sctp_make_chunk.sctp_make_datafrag_empty
1.59 ± 13% -33.2% 1.06 ± 6%
perf-profile.calltrace.cycles-pp.__kmalloc_reserve.__alloc_skb._sctp_make_chunk.sctp_make_datafrag_empty.sctp_datamsg_from_user
7.17 ± 12% +29.0% 9.26 ± 4%
perf-profile.calltrace.cycles-pp.__local_bh_enable_ip.ip_finish_output2.ip_finish_output.ip_output.ip_local_out
6.76 ± 12% +28.6% 8.69 ± 3%
perf-profile.calltrace.cycles-pp.__netif_receive_skb.process_backlog.net_rx_action.__softirqentry_text_start.do_softirq_own_stack
6.73 ± 12% +28.3% 8.64 ± 3%
perf-profile.calltrace.cycles-pp.__netif_receive_skb_core.__netif_receive_skb.process_backlog.net_rx_action.__softirqentry_text_start
8.83 ± 8% +32.2% 11.67 ± 4%
perf-profile.calltrace.cycles-pp.__release_sock.release_sock.sctp_sendmsg.inet_sendmsg.sock_sendmsg
7.09 ± 12% +28.6% 9.11 ± 4%
perf-profile.calltrace.cycles-pp.__softirqentry_text_start.do_softirq_own_stack.do_softirq.__local_bh_enable_ip.ip_finish_output2
7.91 ± 7% +32.8% 10.51 ± 4%
perf-profile.calltrace.cycles-pp.__sys_recvmsg.sys_recvmsg.entry_SYSCALL_64_fastpath
0.70 ± 9% +123.3% 1.57 ± 4%
perf-profile.calltrace.cycles-pp._raw_spin_lock_bh.lock_sock_nested.sctp_skb_recv_datagram.sctp_recvmsg.inet_recvmsg
1.14 ± 17% -22.2% 0.89 ± 3%
perf-profile.calltrace.cycles-pp.consume_skb.sctp_chunk_put.sctp_chunk_free.sctp_outq_sack.sctp_cmd_interpreter
3.41 ± 16% -41.3% 2.00 ± 2%
perf-profile.calltrace.cycles-pp.copy_user_enhanced_fast_string.sctp_user_addto_chunk.sctp_datamsg_from_user.sctp_sendmsg.inet_sendmsg
2.82 ± 8% +25.4% 3.54 ± 9%
perf-profile.calltrace.cycles-pp.copy_user_enhanced_fast_string.skb_copy_datagram_iter.sctp_recvmsg.inet_recvmsg.sock_recvmsg
7.16 ± 12% +28.7% 9.21 ± 4%
perf-profile.calltrace.cycles-pp.do_softirq.__local_bh_enable_ip.ip_finish_output2.ip_finish_output.ip_output
7.12 ± 12% +28.6% 9.16 ± 4%
perf-profile.calltrace.cycles-pp.do_softirq_own_stack.do_softirq.__local_bh_enable_ip.ip_finish_output2.ip_finish_output
36.95 ± 10% +17.6% 43.43 ± 2%
perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_fastpath
1.00 ± 15% -59.8% 0.40 ± 58%
perf-profile.calltrace.cycles-pp.get_page_from_freelist.__alloc_pages_nodemask.kmalloc_large_node.__kmalloc_node_track_caller.__kmalloc_reserve
7.28 ± 7% +33.2% 9.69 ± 4%
perf-profile.calltrace.cycles-pp.inet_recvmsg.sock_recvmsg.___sys_recvmsg.__sys_recvmsg.sys_recvmsg
7.59 ± 12% +29.7% 9.85 ± 4%
perf-profile.calltrace.cycles-pp.ip_finish_output.ip_output.ip_local_out.ip_queue_xmit.sctp_v4_xmit
7.55 ± 12% +29.5% 9.78 ± 4%
perf-profile.calltrace.cycles-pp.ip_finish_output2.ip_finish_output.ip_output.ip_local_out.ip_queue_xmit
6.34 ± 12% +28.4% 8.14 ± 4%
perf-profile.calltrace.cycles-pp.ip_local_deliver.ip_rcv_finish.ip_rcv.__netif_receive_skb_core.__netif_receive_skb
6.29 ± 12% +28.4% 8.08 ± 4%
perf-profile.calltrace.cycles-pp.ip_local_deliver_finish.ip_local_deliver.ip_rcv_finish.ip_rcv.__netif_receive_skb_core
8.19 ± 14% +32.0% 10.82 ± 4%
perf-profile.calltrace.cycles-pp.ip_local_out.ip_queue_xmit.sctp_v4_xmit.sctp_packet_transmit.sctp_outq_flush
7.88 ± 14% +34.6% 10.61 ± 4%
perf-profile.calltrace.cycles-pp.ip_output.ip_local_out.ip_queue_xmit.sctp_v4_xmit.sctp_packet_transmit
8.48 ± 12% +29.3% 10.96 ± 4%
perf-profile.calltrace.cycles-pp.ip_queue_xmit.sctp_v4_xmit.sctp_packet_transmit.sctp_outq_flush.sctp_outq_uncork
6.54 ± 12% +28.7% 8.41 ± 4%
perf-profile.calltrace.cycles-pp.ip_rcv.__netif_receive_skb_core.__netif_receive_skb.process_backlog.net_rx_action
6.39 ± 12% +28.7% 8.23 ± 4%
perf-profile.calltrace.cycles-pp.ip_rcv_finish.ip_rcv.__netif_receive_skb_core.__netif_receive_skb.process_backlog
1.08 ± 8% +29.2% 1.39 ± 5%
perf-profile.calltrace.cycles-pp.kfree_skb.sctp_ulpevent_free.sctp_recvmsg.inet_recvmsg.sock_recvmsg
1.48 ± 14% -36.6% 0.94 ± 6%
perf-profile.calltrace.cycles-pp.kmalloc_large_node.__kmalloc_node_track_caller.__kmalloc_reserve.__alloc_skb._sctp_make_chunk
0.76 ± 9% +115.4% 1.63 ± 3%
perf-profile.calltrace.cycles-pp.lock_sock_nested.sctp_skb_recv_datagram.sctp_recvmsg.inet_recvmsg.sock_recvmsg
3.76 ± 6% +45.9% 5.49 ± 8%
perf-profile.calltrace.cycles-pp.memcpy_erms.sctp_outq_flush.sctp_outq_uncork.sctp_cmd_interpreter.sctp_do_sm
0.70 ± 9% +123.1% 1.56 ± 4%
perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_bh.lock_sock_nested.sctp_skb_recv_datagram.sctp_recvmsg
6.99 ± 12% +28.3% 8.96 ± 4%
perf-profile.calltrace.cycles-pp.net_rx_action.__softirqentry_text_start.do_softirq_own_stack.do_softirq.__local_bh_enable_ip
6.86 ± 12% +28.4% 8.81 ± 3%
perf-profile.calltrace.cycles-pp.process_backlog.net_rx_action.__softirqentry_text_start.do_softirq_own_stack.do_softirq
8.97 ± 8% +31.6% 11.81 ± 4%
perf-profile.calltrace.cycles-pp.release_sock.sctp_sendmsg.inet_sendmsg.sock_sendmsg.___sys_sendmsg
6.95 ± 7% +35.3% 9.40 ± 4%
perf-profile.calltrace.cycles-pp.sctp_assoc_bh_rcv.sctp_inq_push.sctp_backlog_rcv.__release_sock.release_sock
4.80 ± 13% +25.3% 6.01 ± 4%
perf-profile.calltrace.cycles-pp.sctp_assoc_bh_rcv.sctp_inq_push.sctp_rcv.ip_local_deliver_finish.ip_local_deliver
8.75 ± 8% +32.3% 11.58 ± 4%
perf-profile.calltrace.cycles-pp.sctp_backlog_rcv.__release_sock.release_sock.sctp_sendmsg.inet_sendmsg
1.27 ± 16% -17.3% 1.05 ± 2%
perf-profile.calltrace.cycles-pp.sctp_chunk_put.sctp_chunk_free.sctp_outq_sack.sctp_cmd_interpreter.sctp_do_sm
7.03 ± 8% +35.7% 9.53 ± 4%
perf-profile.calltrace.cycles-pp.sctp_cmd_interpreter.sctp_do_sm.sctp_assoc_bh_rcv.sctp_inq_push.sctp_backlog_rcv
3.99 ± 13% +25.2% 5.00 ± 4%
perf-profile.calltrace.cycles-pp.sctp_cmd_interpreter.sctp_do_sm.sctp_assoc_bh_rcv.sctp_inq_push.sctp_rcv
6.11 ± 11% +22.1% 7.46 ± 6%
perf-profile.calltrace.cycles-pp.sctp_cmd_interpreter.sctp_do_sm.sctp_primitive_SEND.sctp_sendmsg.inet_sendmsg
7.09 ± 13% -21.8% 5.55 ± 5%
perf-profile.calltrace.cycles-pp.sctp_datamsg_from_user.sctp_sendmsg.inet_sendmsg.sock_sendmsg.___sys_sendmsg
6.86 ± 7% +35.9% 9.31 ± 4%
perf-profile.calltrace.cycles-pp.sctp_do_sm.sctp_assoc_bh_rcv.sctp_inq_push.sctp_backlog_rcv.__release_sock
4.44 ± 13% +25.1% 5.56 ± 4%
perf-profile.calltrace.cycles-pp.sctp_do_sm.sctp_assoc_bh_rcv.sctp_inq_push.sctp_rcv.ip_local_deliver_finish
6.08 ± 11% +21.7% 7.40 ± 6%
perf-profile.calltrace.cycles-pp.sctp_do_sm.sctp_primitive_SEND.sctp_sendmsg.inet_sendmsg.sock_sendmsg
6.94 ± 7% +35.0% 9.37 ± 4%
perf-profile.calltrace.cycles-pp.sctp_inq_push.sctp_backlog_rcv.__release_sock.release_sock.sctp_sendmsg
4.90 ± 13% +25.4% 6.15 ± 4%
perf-profile.calltrace.cycles-pp.sctp_inq_push.sctp_rcv.ip_local_deliver_finish.ip_local_deliver.ip_rcv_finish
7.48 ± 7% +40.6% 10.52 ± 5%
perf-profile.calltrace.cycles-pp.sctp_outq_flush.sctp_outq_uncork.sctp_cmd_interpreter.sctp_do_sm.sctp_assoc_bh_rcv
7.32 ± 12% +23.7% 9.05 ± 5%
perf-profile.calltrace.cycles-pp.sctp_outq_flush.sctp_outq_uncork.sctp_cmd_interpreter.sctp_do_sm.sctp_primitive_SEND
7.55 ± 7% +40.5% 10.61 ± 5%
perf-profile.calltrace.cycles-pp.sctp_outq_uncork.sctp_cmd_interpreter.sctp_do_sm.sctp_assoc_bh_rcv.sctp_inq_push
7.42 ± 12% +23.5% 9.17 ± 5%
perf-profile.calltrace.cycles-pp.sctp_outq_uncork.sctp_cmd_interpreter.sctp_do_sm.sctp_primitive_SEND.sctp_sendmsg
9.64 ± 11% +29.3% 12.47 ± 3%
perf-profile.calltrace.cycles-pp.sctp_packet_transmit.sctp_outq_flush.sctp_outq_uncork.sctp_cmd_interpreter.sctp_do_sm
9.07 ± 12% +23.2% 11.17 ± 5%
perf-profile.calltrace.cycles-pp.sctp_primitive_SEND.sctp_sendmsg.inet_sendmsg.sock_sendmsg.___sys_sendmsg
6.01 ± 12% +27.9% 7.69 ± 4%
perf-profile.calltrace.cycles-pp.sctp_rcv.ip_local_deliver_finish.ip_local_deliver.ip_rcv_finish.ip_rcv
7.17 ± 7% +33.6% 9.58 ± 4%
perf-profile.calltrace.cycles-pp.sctp_recvmsg.inet_recvmsg.sock_recvmsg.___sys_recvmsg.__sys_recvmsg
1.96 ± 9% +57.6% 3.08 ± 3%
perf-profile.calltrace.cycles-pp.sctp_skb_recv_datagram.sctp_recvmsg.inet_recvmsg.sock_recvmsg.___sys_recvmsg
1.44 ± 8% +31.1% 1.89 ± 4%
perf-profile.calltrace.cycles-pp.sctp_ulpevent_free.sctp_recvmsg.inet_recvmsg.sock_recvmsg.___sys_recvmsg
3.68 ± 15% -37.4% 2.30 ± 2%
perf-profile.calltrace.cycles-pp.sctp_user_addto_chunk.sctp_datamsg_from_user.sctp_sendmsg.inet_sendmsg.sock_sendmsg
8.53 ± 12% +29.5% 11.04 ± 3%
perf-profile.calltrace.cycles-pp.sctp_v4_xmit.sctp_packet_transmit.sctp_outq_flush.sctp_outq_uncork.sctp_cmd_interpreter
3.01 ± 7% +24.6% 3.75 ± 9%
perf-profile.calltrace.cycles-pp.skb_copy_datagram_iter.sctp_recvmsg.inet_recvmsg.sock_recvmsg.___sys_recvmsg
0.96 ± 17% -28.3% 0.69 ± 1%
perf-profile.calltrace.cycles-pp.skb_release_all.consume_skb.sctp_chunk_put.sctp_chunk_free.sctp_outq_sack
0.86 ± 9% +28.2% 1.11 ± 5%
perf-profile.calltrace.cycles-pp.skb_release_all.kfree_skb.sctp_ulpevent_free.sctp_recvmsg.inet_recvmsg
7.42 ± 7% +33.2% 9.89 ± 4%
perf-profile.calltrace.cycles-pp.sock_recvmsg.___sys_recvmsg.__sys_recvmsg.sys_recvmsg.entry_SYSCALL_64_fastpath
7.92 ± 7% +32.8% 10.52 ± 4%
perf-profile.calltrace.cycles-pp.sys_recvmsg.entry_SYSCALL_64_fastpath
7.80 ± 7% +32.7% 10.35 ± 4%
perf-profile.children.cycles-pp.___sys_recvmsg
1.99 ± 12% -18.1% 1.63 ± 2%
perf-profile.children.cycles-pp.__alloc_pages_nodemask
0.77 ± 12% +39.6% 1.07 ± 5%
perf-profile.children.cycles-pp.__dev_queue_xmit
1.08 ± 16% -74.7% 0.27 ± 14%
perf-profile.children.cycles-pp.__free_pages_ok
7.43 ± 12% +28.9% 9.58 ± 4%
perf-profile.children.cycles-pp.__local_bh_enable_ip
6.78 ± 12% +28.5% 8.71 ± 3%
perf-profile.children.cycles-pp.__netif_receive_skb
6.76 ± 12% +28.3% 8.67 ± 4%
perf-profile.children.cycles-pp.__netif_receive_skb_core
8.85 ± 8% +32.0% 11.68 ± 4%
perf-profile.children.cycles-pp.__release_sock
1.37 ± 16% +20.6% 1.65 ± 4% perf-profile.children.cycles-pp.__schedule
7.18 ± 12% +28.3% 9.21 ± 4%
perf-profile.children.cycles-pp.__softirqentry_text_start
7.92 ± 7% +32.8% 10.51 ± 4%
perf-profile.children.cycles-pp.__sys_recvmsg
0.87 ± 10% +102.3% 1.76 ± 3%
perf-profile.children.cycles-pp._raw_spin_lock_bh
0.81 ± 11% +38.9% 1.12 ± 5%
perf-profile.children.cycles-pp.dev_queue_xmit
7.17 ± 12% +28.8% 9.23 ± 4% perf-profile.children.cycles-pp.do_softirq
7.16 ± 12% +28.7% 9.21 ± 4%
perf-profile.children.cycles-pp.do_softirq_own_stack
37.15 ± 10% +17.6% 43.68 ± 2%
perf-profile.children.cycles-pp.entry_SYSCALL_64_fastpath
1.51 ± 12% -25.1% 1.13 ± 3%
perf-profile.children.cycles-pp.get_page_from_freelist
7.28 ± 7% +33.2% 9.70 ± 4%
perf-profile.children.cycles-pp.inet_recvmsg
7.92 ± 12% +29.6% 10.27 ± 4%
perf-profile.children.cycles-pp.ip_finish_output
7.85 ± 12% +29.6% 10.17 ± 4%
perf-profile.children.cycles-pp.ip_finish_output2
6.35 ± 12% +28.5% 8.16 ± 4%
perf-profile.children.cycles-pp.ip_local_deliver
6.31 ± 12% +28.6% 8.11 ± 4%
perf-profile.children.cycles-pp.ip_local_deliver_finish
8.40 ± 12% +29.2% 10.85 ± 4%
perf-profile.children.cycles-pp.ip_local_out
8.23 ± 12% +29.1% 10.62 ± 4% perf-profile.children.cycles-pp.ip_output
8.49 ± 12% +29.3% 10.98 ± 3%
perf-profile.children.cycles-pp.ip_queue_xmit
6.57 ± 12% +28.4% 8.44 ± 4% perf-profile.children.cycles-pp.ip_rcv
6.42 ± 12% +28.5% 8.25 ± 4%
perf-profile.children.cycles-pp.ip_rcv_finish
1.08 ± 8% +29.2% 1.40 ± 5% perf-profile.children.cycles-pp.kfree_skb
1.00 ± 8% +90.4% 1.90 ± 3%
perf-profile.children.cycles-pp.lock_sock_nested
3.92 ± 5% +43.4% 5.62 ± 7% perf-profile.children.cycles-pp.memcpy_erms
0.83 ± 11% +89.7% 1.58 ± 3%
perf-profile.children.cycles-pp.native_queued_spin_lock_slowpath
7.02 ± 12% +28.2% 9.00 ± 3%
perf-profile.children.cycles-pp.net_rx_action
6.88 ± 12% +28.3% 8.83 ± 3%
perf-profile.children.cycles-pp.process_backlog
9.10 ± 8% +31.4% 11.95 ± 4%
perf-profile.children.cycles-pp.release_sock
1.47 ± 15% +21.1% 1.78 ± 4% perf-profile.children.cycles-pp.schedule
11.77 ± 9% +31.0% 15.43 ± 4%
perf-profile.children.cycles-pp.sctp_assoc_bh_rcv
8.77 ± 8% +32.2% 11.59 ± 4%
perf-profile.children.cycles-pp.sctp_backlog_rcv
17.20 ± 10% +28.2% 22.05 ± 4%
perf-profile.children.cycles-pp.sctp_cmd_interpreter
7.10 ± 13% -21.5% 5.57 ± 5%
perf-profile.children.cycles-pp.sctp_datamsg_from_user
17.44 ± 10% +28.1% 22.33 ± 4% perf-profile.children.cycles-pp.sctp_do_sm
11.88 ± 9% +30.9% 15.54 ± 4%
perf-profile.children.cycles-pp.sctp_inq_push
14.87 ± 9% +32.3% 19.66 ± 5%
perf-profile.children.cycles-pp.sctp_outq_flush
15.05 ± 9% +32.1% 19.88 ± 5%
perf-profile.children.cycles-pp.sctp_outq_uncork
9.66 ± 11% +29.2% 12.48 ± 3%
perf-profile.children.cycles-pp.sctp_packet_transmit
9.09 ± 12% +23.2% 11.20 ± 5%
perf-profile.children.cycles-pp.sctp_primitive_SEND
6.03 ± 12% +27.9% 7.71 ± 4% perf-profile.children.cycles-pp.sctp_rcv
7.17 ± 7% +33.7% 9.59 ± 4%
perf-profile.children.cycles-pp.sctp_recvmsg
1.96 ± 9% +57.5% 3.09 ± 3%
perf-profile.children.cycles-pp.sctp_skb_recv_datagram
1.45 ± 8% +31.1% 1.89 ± 4%
perf-profile.children.cycles-pp.sctp_ulpevent_free
3.70 ± 15% -37.1% 2.33 ± 2%
perf-profile.children.cycles-pp.sctp_user_addto_chunk
8.56 ± 12% +29.4% 11.07 ± 3%
perf-profile.children.cycles-pp.sctp_v4_xmit
3.02 ± 7% +24.5% 3.75 ± 9%
perf-profile.children.cycles-pp.skb_copy_datagram_iter
7.42 ± 7% +33.3% 9.89 ± 4%
perf-profile.children.cycles-pp.sock_recvmsg
7.93 ± 7% +32.8% 10.53 ± 4% perf-profile.children.cycles-pp.sys_recvmsg
3.92 ± 5% +43.4% 5.62 ± 7% perf-profile.self.cycles-pp.memcpy_erms
0.83 ± 11% +89.7% 1.58 ± 3%
perf-profile.self.cycles-pp.native_queued_spin_lock_slowpath
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,
Xiaolong