[lkp] [inet] 14135f30e3: netperf.Throughput_tps 6.7% improvement
by kernel test robot
Greeting,
FYI, we noticed a 6.7% improvement of netperf.Throughput_tps due to commit:
commit 14135f30e33ce37b22529f73660d7369cf424375 ("inet: fix sleeping inside inet_wait_for_connect()")
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
in testcase: netperf
on test machine: 16 threads Nehalem-EP with 5G memory
with following parameters:
ip: ipv4
runtime: 300s
nr_threads: 200%
cluster: cs-localhost
test: TCP_CRR
cpufreq_governor: performance
test-description: Netperf is a benchmark that can be use to measure various aspect of networking performance.
test-url: http://www.netperf.org/netperf/
In addition to that, the commit also has significant impact on the following tests:
+------------------+----------------------------------------------------------------------+
| testcase: change | netperf: netperf.Throughput_tps 7.0% improvement |
| test machine | 48 threads Intel(R) Xeon(R) CPU E5-2697 v2 @ 2.70GHz with 64G memory |
| test parameters | cluster=cs-localhost |
| | cpufreq_governor=performance |
| | ip=ipv4 |
| | nr_threads=200% |
| | runtime=300s |
| | test=TCP_CRR |
+------------------+----------------------------------------------------------------------+
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
=========================================================================================
cluster/compiler/cpufreq_governor/ip/kconfig/nr_threads/rootfs/runtime/tbox_group/test/testcase:
cs-localhost/gcc-6/performance/ipv4/x86_64-rhel-7.2/200%/debian-x86_64-2016-08-31.cgz/300s/lkp-ne02/TCP_CRR/netperf
commit:
269ebce453 ("xen-netfront: cast grant table reference first to type int")
14135f30e3 ("inet: fix sleeping inside inet_wait_for_connect()")
269ebce4531b8edc 14135f30e33ce37b22529f7366
---------------- --------------------------
%stddev %change %stddev
\ | \
4725 ± 0% +6.7% 5043 ± 0% netperf.Throughput_tps
47071067 ± 0% -95.7% 2034229 ± 2% netperf.time.involuntary_context_switches
1089 ± 0% +7.9% 1175 ± 0% netperf.time.percent_of_cpu_this_job_got
3170 ± 0% +8.4% 3437 ± 0% netperf.time.system_time
153.32 ± 0% -2.7% 149.21 ± 1% netperf.time.user_time
45002630 ± 0% +6.8% 48057080 ± 0% netperf.time.voluntary_context_switches
8179 ± 9% +6.7% 8728 ± 9% numa-meminfo.node0.Mapped
617750 ± 0% -44.1% 345350 ± 0% vmstat.system.cs
401.00 ±173% +300.6% 1606 ± 0% meminfo.Mlocked
23124 ± 0% +11.5% 25777 ± 9% meminfo.Shmem
401.00 ±173% +300.6% 1606 ± 0% meminfo.Unevictable
100.00 ±173% +301.2% 401.25 ± 0% proc-vmstat.nr_mlock
5776 ± 0% +11.4% 6435 ± 9% proc-vmstat.nr_shmem
100.00 ±173% +301.2% 401.25 ± 0% proc-vmstat.nr_unevictable
100.00 ±173% +301.2% 401.25 ± 0% proc-vmstat.nr_zone_unevictable
9000 ± 1% +12.7% 10148 ± 10% proc-vmstat.pgactivate
2046 ± 9% +6.6% 2181 ± 9% numa-vmstat.node0.nr_mapped
46.00 ±173% +336.4% 200.75 ± 11% numa-vmstat.node0.nr_mlock
46.00 ±173% +336.4% 200.75 ± 11% numa-vmstat.node0.nr_unevictable
46.00 ±173% +336.4% 200.75 ± 11% numa-vmstat.node0.nr_zone_unevictable
54.00 ±173% +270.4% 200.00 ± 12% numa-vmstat.node1.nr_mlock
54.00 ±173% +270.4% 200.00 ± 12% numa-vmstat.node1.nr_unevictable
54.00 ±173% +270.4% 200.00 ± 12% numa-vmstat.node1.nr_zone_unevictable
8.74 ± 0% +3.9% 9.09 ± 0% perf-stat.branch-miss-rate%
8.07e+10 ± 0% +3.7% 8.37e+10 ± 0% perf-stat.branch-misses
1.01 ± 0% +10.7% 1.11 ± 0% perf-stat.cache-miss-rate%
1.662e+09 ± 1% +8.5% 1.803e+09 ± 0% perf-stat.cache-misses
1.654e+11 ± 0% -2.0% 1.621e+11 ± 0% perf-stat.cache-references
1.896e+08 ± 0% -44.1% 1.06e+08 ± 0% perf-stat.context-switches
34044 ± 3% +21.6% 41402 ± 7% perf-stat.cpu-migrations
0.23 ± 0% -2.2% 0.23 ± 0% perf-stat.iTLB-load-miss-rate%
1.196e+10 ± 0% -2.1% 1.171e+10 ± 0% perf-stat.iTLB-load-misses
425.19 ± 0% +2.2% 434.62 ± 0% perf-stat.instructions-per-iTLB-miss
1.053e+09 ± 1% +8.8% 1.146e+09 ± 0% perf-stat.node-loads
5.711e+08 ± 0% +8.5% 6.195e+08 ± 1% perf-stat.node-stores
571225 ± 8% +126.5% 1293742 ± 6% sched_debug.cfs_rq:/.MIN_vruntime.avg
757172 ± 1% +11.7% 845694 ± 2% sched_debug.cfs_rq:/.MIN_vruntime.stddev
116.03 ± 2% +20.9% 140.33 ± 1% sched_debug.cfs_rq:/.load_avg.avg
370.21 ± 16% +33.9% 495.88 ± 9% sched_debug.cfs_rq:/.load_avg.max
73.12 ± 8% +25.1% 91.50 ± 11% sched_debug.cfs_rq:/.load_avg.min
71.20 ± 18% +34.8% 95.97 ± 11% sched_debug.cfs_rq:/.load_avg.stddev
571225 ± 8% +126.5% 1293743 ± 6% sched_debug.cfs_rq:/.max_vruntime.avg
757172 ± 1% +11.7% 845694 ± 2% sched_debug.cfs_rq:/.max_vruntime.stddev
1578370 ± 1% +19.9% 1891826 ± 0% sched_debug.cfs_rq:/.min_vruntime.avg
1424031 ± 3% +19.5% 1701463 ± 4% sched_debug.cfs_rq:/.min_vruntime.min
1.27 ± 3% +21.1% 1.54 ± 2% sched_debug.cfs_rq:/.nr_running.avg
71.26 ± 1% +29.2% 92.07 ± 1% sched_debug.cfs_rq:/.runnable_load_avg.avg
111.04 ± 8% +32.3% 146.88 ± 13% sched_debug.cfs_rq:/.runnable_load_avg.max
17.81 ± 12% +77.3% 31.59 ± 14% sched_debug.cfs_rq:/.runnable_load_avg.stddev
10951 ± 28% +81.0% 19827 ± 31% sched_debug.cpu.avg_idle.min
70.46 ± 2% +24.7% 87.87 ± 2% sched_debug.cpu.cpu_load[0].avg
109.79 ± 9% +32.6% 145.58 ± 13% sched_debug.cpu.cpu_load[0].max
18.71 ± 13% +74.6% 32.68 ± 11% sched_debug.cpu.cpu_load[0].stddev
70.80 ± 2% +24.6% 88.25 ± 1% sched_debug.cpu.cpu_load[1].avg
104.46 ± 8% +31.8% 137.62 ± 11% sched_debug.cpu.cpu_load[1].max
14.18 ± 16% +67.2% 23.70 ± 19% sched_debug.cpu.cpu_load[1].stddev
70.58 ± 1% +26.0% 88.92 ± 0% sched_debug.cpu.cpu_load[2].avg
69.97 ± 1% +28.4% 89.87 ± 0% sched_debug.cpu.cpu_load[3].avg
68.27 ± 1% +33.4% 91.10 ± 0% sched_debug.cpu.cpu_load[4].avg
5956665 ± 0% -43.0% 3394350 ± 0% sched_debug.cpu.nr_switches.avg
7912513 ± 6% -16.0% 6649061 ± 1% sched_debug.cpu.nr_switches.max
5664910 ± 0% -46.4% 3034128 ± 0% sched_debug.cpu.nr_switches.min
576345 ± 16% +50.5% 867213 ± 3% sched_debug.cpu.nr_switches.stddev
19.13 ± 0% -31.6% 13.09 ± 0% perf-profile.calltrace.cycles-pp.SYSC_recvfrom.sys_recvfrom.entry_SYSCALL_64_fastpath
13.43 ± 1% +34.5% 18.07 ± 0% perf-profile.calltrace.cycles-pp.SYSC_sendto.sys_sendto.entry_SYSCALL_64_fastpath
0.00 ± -1% +Inf% 1.45 ± 2% perf-profile.calltrace.cycles-pp.__release_sock.release_sock.tcp_sendmsg.inet_sendmsg.sock_sendmsg
1.47 ± 2% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.__schedule.schedule.schedule_timeout.__inet_stream_connect.inet_stream_connect
2.63 ± 1% -16.5% 2.20 ± 2% perf-profile.calltrace.cycles-pp.__schedule.schedule.schedule_timeout.sk_wait_data.tcp_recvmsg
10.83 ± 0% -45.2% 5.93 ± 0% perf-profile.calltrace.cycles-pp.__tcp_ack_snd_check.tcp_rcv_established.tcp_v4_do_rcv.tcp_prequeue_process.tcp_recvmsg
0.00 ± -1% +Inf% 1.47 ± 2% perf-profile.calltrace.cycles-pp.__tcp_ack_snd_check.tcp_rcv_established.tcp_v4_do_rcv.tcp_v4_rcv.ip_local_deliver_finish
10.49 ± 1% +28.5% 13.48 ± 0% perf-profile.calltrace.cycles-pp.__tcp_push_pending_frames.tcp_push.tcp_sendmsg.inet_sendmsg.sock_sendmsg
1.82 ± 3% +10.2% 2.00 ± 0% perf-profile.calltrace.cycles-pp.__wake_up_common.__wake_up_sync_key.sock_def_readable.tcp_child_process.tcp_v4_rcv
2.51 ± 1% -40.1% 1.51 ± 2% perf-profile.calltrace.cycles-pp.__wake_up_common.__wake_up_sync_key.tcp_prequeue.tcp_v4_rcv.ip_local_deliver_finish
1.86 ± 3% +11.0% 2.06 ± 1% perf-profile.calltrace.cycles-pp.__wake_up_sync_key.sock_def_readable.tcp_child_process.tcp_v4_rcv.ip_local_deliver_finish
2.62 ± 2% -39.8% 1.58 ± 2% perf-profile.calltrace.cycles-pp.__wake_up_sync_key.tcp_prequeue.tcp_v4_rcv.ip_local_deliver_finish.ip_local_deliver
2.17 ± 1% -22.2% 1.69 ± 1% perf-profile.calltrace.cycles-pp.activate_task.ttwu_do_activate.try_to_wake_up.default_wake_function.autoremove_wake_function
1.75 ± 3% +10.3% 1.92 ± 0% perf-profile.calltrace.cycles-pp.autoremove_wake_function.__wake_up_common.__wake_up_sync_key.sock_def_readable.tcp_child_process
2.42 ± 2% -40.4% 1.44 ± 3% perf-profile.calltrace.cycles-pp.autoremove_wake_function.__wake_up_common.__wake_up_sync_key.tcp_prequeue.tcp_v4_rcv
1.18 ± 2% -23.9% 0.90 ± 2% perf-profile.calltrace.cycles-pp.deactivate_task.__schedule.schedule.schedule_timeout.sk_wait_data
1.73 ± 3% +10.2% 1.91 ± 0% perf-profile.calltrace.cycles-pp.default_wake_function.autoremove_wake_function.__wake_up_common.__wake_up_sync_key.sock_def_readable
2.38 ± 2% -40.7% 1.42 ± 3% perf-profile.calltrace.cycles-pp.default_wake_function.autoremove_wake_function.__wake_up_common.__wake_up_sync_key.tcp_prequeue
1.59 ± 2% -10.5% 1.42 ± 1% perf-profile.calltrace.cycles-pp.dequeue_task_fair.deactivate_task.__schedule.schedule.schedule_timeout
1.74 ± 2% -26.2% 1.28 ± 2% perf-profile.calltrace.cycles-pp.dev_hard_start_xmit.__dev_queue_xmit.dev_queue_xmit.ip_finish_output2.ip_finish_output
1.48 ± 1% -55.9% 0.66 ± 3% perf-profile.calltrace.cycles-pp.enqueue_entity.enqueue_task_fair.activate_task.ttwu_do_activate.try_to_wake_up
1.98 ± 1% -21.8% 1.55 ± 1% perf-profile.calltrace.cycles-pp.enqueue_task_fair.activate_task.ttwu_do_activate.try_to_wake_up.default_wake_function
2.38 ± 2% +9.8% 2.61 ± 1% perf-profile.calltrace.cycles-pp.inet_csk_accept.inet_accept.SYSC_accept4.sys_accept.entry_SYSCALL_64_fastpath
18.30 ± 0% -32.7% 12.30 ± 0% perf-profile.calltrace.cycles-pp.inet_recvmsg.sock_recvmsg.SYSC_recvfrom.sys_recvfrom.entry_SYSCALL_64_fastpath
12.87 ± 1% +36.1% 17.52 ± 0% perf-profile.calltrace.cycles-pp.inet_sendmsg.sock_sendmsg.SYSC_sendto.sys_sendto.entry_SYSCALL_64_fastpath
8.83 ± 1% -36.9% 5.57 ± 0% perf-profile.calltrace.cycles-pp.ip_local_out.ip_queue_xmit.tcp_transmit_skb.tcp_send_ack.__tcp_ack_snd_check
18.18 ± 0% +13.7% 20.66 ± 0% perf-profile.calltrace.cycles-pp.ip_local_out.ip_queue_xmit.tcp_transmit_skb.tcp_write_xmit.__tcp_push_pending_frames
17.55 ± 0% -15.0% 14.91 ± 0% perf-profile.calltrace.cycles-pp.ip_output.ip_local_out.ip_queue_xmit.tcp_transmit_skb.tcp_send_ack
17.72 ± 0% +14.3% 20.25 ± 0% perf-profile.calltrace.cycles-pp.ip_output.ip_local_out.ip_queue_xmit.tcp_transmit_skb.tcp_write_xmit
9.07 ± 1% -36.1% 5.79 ± 0% perf-profile.calltrace.cycles-pp.ip_queue_xmit.tcp_transmit_skb.tcp_send_ack.__tcp_ack_snd_check.tcp_rcv_established
8.42 ± 1% +20.6% 10.15 ± 1% perf-profile.calltrace.cycles-pp.ip_queue_xmit.tcp_transmit_skb.tcp_write_xmit.__tcp_push_pending_frames.tcp_push
0.00 ± -1% +Inf% 1.57 ± 3% perf-profile.calltrace.cycles-pp.release_sock.tcp_sendmsg.inet_sendmsg.sock_sendmsg.SYSC_sendto
1.59 ± 3% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.schedule.schedule_timeout.__inet_stream_connect.inet_stream_connect.SYSC_connect
1.78 ± 2% +9.8% 1.96 ± 2% perf-profile.calltrace.cycles-pp.schedule.schedule_timeout.inet_csk_accept.inet_accept.SYSC_accept4
2.81 ± 1% -16.8% 2.34 ± 2% perf-profile.calltrace.cycles-pp.schedule.schedule_timeout.sk_wait_data.tcp_recvmsg.inet_recvmsg
1.63 ± 3% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.schedule_timeout.__inet_stream_connect.inet_stream_connect.SYSC_connect.sys_connect
2.87 ± 1% -16.7% 2.39 ± 2% perf-profile.calltrace.cycles-pp.schedule_timeout.sk_wait_data.tcp_recvmsg.inet_recvmsg.sock_recvmsg
0.55 ± 3% +111.9% 1.16 ± 1% perf-profile.calltrace.cycles-pp.security_sock_rcv_skb.sk_filter_trim_cap.tcp_v4_rcv.ip_local_deliver_finish.ip_local_deliver
0.00 ± -1% +Inf% 0.78 ± 32% perf-profile.calltrace.cycles-pp.selinux_socket_sock_rcv_skb.security_sock_rcv_skb.sk_filter_trim_cap.tcp_v4_rcv.ip_local_deliver_finish
0.63 ± 4% +135.9% 1.48 ± 14% perf-profile.calltrace.cycles-pp.sk_filter_trim_cap.tcp_v4_rcv.ip_local_deliver_finish.ip_local_deliver.ip_rcv_finish
3.35 ± 0% -19.7% 2.69 ± 1% perf-profile.calltrace.cycles-pp.sk_wait_data.tcp_recvmsg.inet_recvmsg.sock_recvmsg.SYSC_recvfrom
1.02 ± 1% +18.2% 1.20 ± 0% perf-profile.calltrace.cycles-pp.sock_alloc_file.SYSC_accept4.sys_accept.entry_SYSCALL_64_fastpath
1.90 ± 2% +11.1% 2.11 ± 1% perf-profile.calltrace.cycles-pp.sock_def_readable.tcp_child_process.tcp_v4_rcv.ip_local_deliver_finish.ip_local_deliver
18.68 ± 0% -32.1% 12.69 ± 0% perf-profile.calltrace.cycles-pp.sock_recvmsg.SYSC_recvfrom.sys_recvfrom.entry_SYSCALL_64_fastpath
13.13 ± 0% +35.5% 17.80 ± 0% perf-profile.calltrace.cycles-pp.sock_sendmsg.SYSC_sendto.sys_sendto.entry_SYSCALL_64_fastpath
19.21 ± 0% -31.5% 13.16 ± 0% perf-profile.calltrace.cycles-pp.sys_recvfrom.entry_SYSCALL_64_fastpath
13.47 ± 1% +34.6% 18.13 ± 0% perf-profile.calltrace.cycles-pp.sys_sendto.entry_SYSCALL_64_fastpath
1.01 ± 3% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.tcp_ack.tcp_rcv_established.tcp_v4_do_rcv.tcp_v4_rcv.ip_local_deliver_finish
1.25 ± 3% -21.4% 0.98 ± 4% perf-profile.calltrace.cycles-pp.tcp_finish_connect.tcp_rcv_state_process.tcp_v4_do_rcv.__release_sock.release_sock
3.26 ± 1% -39.2% 1.98 ± 1% perf-profile.calltrace.cycles-pp.tcp_prequeue.tcp_v4_rcv.ip_local_deliver_finish.ip_local_deliver.ip_rcv_finish
13.38 ± 0% -42.8% 7.65 ± 0% perf-profile.calltrace.cycles-pp.tcp_prequeue_process.tcp_recvmsg.inet_recvmsg.sock_recvmsg.SYSC_recvfrom
10.53 ± 1% +28.4% 13.52 ± 0% perf-profile.calltrace.cycles-pp.tcp_push.tcp_sendmsg.inet_sendmsg.sock_sendmsg.SYSC_sendto
0.00 ± -1% +Inf% 1.29 ± 3% perf-profile.calltrace.cycles-pp.tcp_rcv_established.tcp_v4_do_rcv.__release_sock.release_sock.tcp_sendmsg
11.21 ± 0% -35.5% 7.23 ± 0% perf-profile.calltrace.cycles-pp.tcp_rcv_established.tcp_v4_do_rcv.tcp_prequeue_process.tcp_recvmsg.inet_recvmsg
1.46 ± 2% +50.7% 2.21 ± 2% perf-profile.calltrace.cycles-pp.tcp_rcv_established.tcp_v4_do_rcv.tcp_v4_rcv.ip_local_deliver_finish.ip_local_deliver
18.07 ± 0% -33.4% 12.04 ± 0% perf-profile.calltrace.cycles-pp.tcp_recvmsg.inet_recvmsg.sock_recvmsg.SYSC_recvfrom.sys_recvfrom
10.41 ± 0% -46.4% 5.59 ± 0% perf-profile.calltrace.cycles-pp.tcp_send_ack.__tcp_ack_snd_check.tcp_rcv_established.tcp_v4_do_rcv.tcp_prequeue_process
0.00 ± -1% +Inf% 1.39 ± 2% perf-profile.calltrace.cycles-pp.tcp_send_ack.__tcp_ack_snd_check.tcp_rcv_established.tcp_v4_do_rcv.tcp_v4_rcv
12.41 ± 0% +36.7% 16.96 ± 0% perf-profile.calltrace.cycles-pp.tcp_sendmsg.inet_sendmsg.sock_sendmsg.SYSC_sendto.sys_sendto
9.61 ± 1% -34.7% 6.28 ± 0% perf-profile.calltrace.cycles-pp.tcp_transmit_skb.tcp_send_ack.__tcp_ack_snd_check.tcp_rcv_established.tcp_v4_do_rcv
9.25 ± 0% +17.4% 10.86 ± 0% perf-profile.calltrace.cycles-pp.tcp_transmit_skb.tcp_write_xmit.__tcp_push_pending_frames.tcp_push.tcp_sendmsg
0.00 ± -1% +Inf% 1.36 ± 3% perf-profile.calltrace.cycles-pp.tcp_v4_do_rcv.__release_sock.release_sock.tcp_sendmsg.inet_sendmsg
9.60 ± 0% -42.4% 5.53 ± 0% perf-profile.calltrace.cycles-pp.tcp_v4_do_rcv.tcp_prequeue_process.tcp_recvmsg.inet_recvmsg.sock_recvmsg
12.92 ± 0% +10.3% 14.25 ± 0% perf-profile.calltrace.cycles-pp.tcp_v4_do_rcv.tcp_v4_rcv.ip_local_deliver_finish.ip_local_deliver.ip_rcv_finish
10.35 ± 1% +28.8% 13.33 ± 0% perf-profile.calltrace.cycles-pp.tcp_write_xmit.__tcp_push_pending_frames.tcp_push.tcp_sendmsg.inet_sendmsg
3.99 ± 1% -19.9% 3.20 ± 1% perf-profile.calltrace.cycles-pp.try_to_wake_up.default_wake_function.autoremove_wake_function.__wake_up_common.__wake_up_sync_key
2.75 ± 1% -19.5% 2.21 ± 1% perf-profile.calltrace.cycles-pp.ttwu_do_activate.try_to_wake_up.default_wake_function.autoremove_wake_function.__wake_up_common
19.16 ± 0% -31.5% 13.11 ± 0% perf-profile.children.cycles-pp.SYSC_recvfrom
13.44 ± 1% +34.6% 18.08 ± 0% perf-profile.children.cycles-pp.SYSC_sendto
1.62 ± 3% +16.2% 1.89 ± 1% perf-profile.children.cycles-pp.__kfree_skb
14.52 ± 1% +14.0% 16.55 ± 0% perf-profile.children.cycles-pp.__release_sock
5.87 ± 0% -29.8% 4.12 ± 1% perf-profile.children.cycles-pp.__schedule
10.89 ± 0% -31.5% 7.46 ± 0% perf-profile.children.cycles-pp.__tcp_ack_snd_check
23.51 ± 0% +16.6% 27.41 ± 0% perf-profile.children.cycles-pp.__tcp_push_pending_frames
4.75 ± 1% -23.3% 3.64 ± 0% perf-profile.children.cycles-pp.__wake_up_common
4.54 ± 1% -18.9% 3.68 ± 0% perf-profile.children.cycles-pp.__wake_up_sync_key
2.21 ± 1% -21.9% 1.73 ± 1% perf-profile.children.cycles-pp.activate_task
4.57 ± 1% -25.3% 3.42 ± 1% perf-profile.children.cycles-pp.autoremove_wake_function
1.92 ± 2% -11.5% 1.69 ± 1% perf-profile.children.cycles-pp.deactivate_task
4.50 ± 1% -23.6% 3.44 ± 1% perf-profile.children.cycles-pp.default_wake_function
1.30 ± 3% -14.8% 1.11 ± 2% perf-profile.children.cycles-pp.dequeue_entity
1.64 ± 2% -10.8% 1.46 ± 1% perf-profile.children.cycles-pp.dequeue_task_fair
1.53 ± 1% -25.2% 1.14 ± 1% perf-profile.children.cycles-pp.enqueue_entity
2.05 ± 1% -21.6% 1.61 ± 1% perf-profile.children.cycles-pp.enqueue_task_fair
2.39 ± 2% +9.8% 2.62 ± 1% perf-profile.children.cycles-pp.inet_csk_accept
18.34 ± 0% -32.7% 12.33 ± 0% perf-profile.children.cycles-pp.inet_recvmsg
12.88 ± 1% +36.3% 17.56 ± 0% perf-profile.children.cycles-pp.inet_sendmsg
2.00 ± 3% -43.4% 1.13 ± 1% perf-profile.children.cycles-pp.pick_next_task_fair
15.19 ± 1% +13.6% 17.26 ± 0% perf-profile.children.cycles-pp.release_sock
6.35 ± 0% -30.2% 4.43 ± 1% perf-profile.children.cycles-pp.schedule
6.40 ± 0% -30.6% 4.44 ± 1% perf-profile.children.cycles-pp.schedule_timeout
3.36 ± 0% -19.8% 2.70 ± 1% perf-profile.children.cycles-pp.sk_wait_data
1.84 ± 1% +13.9% 2.09 ± 0% perf-profile.children.cycles-pp.sock_alloc_file
18.70 ± 0% -32.0% 12.71 ± 0% perf-profile.children.cycles-pp.sock_recvmsg
13.14 ± 0% +35.6% 17.81 ± 0% perf-profile.children.cycles-pp.sock_sendmsg
19.27 ± 0% -31.5% 13.21 ± 0% perf-profile.children.cycles-pp.sys_recvfrom
13.52 ± 1% +34.4% 18.18 ± 0% perf-profile.children.cycles-pp.sys_sendto
4.30 ± 2% +9.1% 4.69 ± 0% perf-profile.children.cycles-pp.tcp_data_queue
1.27 ± 3% -20.8% 1.00 ± 4% perf-profile.children.cycles-pp.tcp_finish_connect
3.43 ± 1% -36.5% 2.18 ± 1% perf-profile.children.cycles-pp.tcp_prequeue
13.40 ± 0% -42.8% 7.67 ± 0% perf-profile.children.cycles-pp.tcp_prequeue_process
10.62 ± 1% +28.4% 13.64 ± 0% perf-profile.children.cycles-pp.tcp_push
12.69 ± 0% -15.3% 10.74 ± 0% perf-profile.children.cycles-pp.tcp_rcv_established
1.57 ± 2% +18.7% 1.87 ± 2% perf-profile.children.cycles-pp.tcp_rearm_rto
18.12 ± 0% -33.2% 12.10 ± 0% perf-profile.children.cycles-pp.tcp_recvmsg
21.06 ± 0% -13.3% 18.25 ± 0% perf-profile.children.cycles-pp.tcp_send_ack
12.44 ± 0% +36.7% 17.01 ± 0% perf-profile.children.cycles-pp.tcp_sendmsg
23.25 ± 0% +16.7% 27.13 ± 0% perf-profile.children.cycles-pp.tcp_write_xmit
4.30 ± 1% -23.4% 3.29 ± 1% perf-profile.children.cycles-pp.try_to_wake_up
2.79 ± 1% -19.5% 2.25 ± 1% perf-profile.children.cycles-pp.ttwu_do_activate
1.26 ± 2% +15.9% 1.46 ± 3% perf-profile.self.cycles-pp.tcp_ack
perf-stat.context-switches
2e+08 ++----------------------------------------------------------------+
1.8e+08 ++ *..*.*..*.*..*.*..*.*..*.*..*..*.*..*.*..*.*..* |
| : |
1.6e+08 ++ : |
1.4e+08 ++ : |
| : |
1.2e+08 ++ : |
1e+08 O+ O:O O O O O O O O O O O O O O O O O O O O O O O O O
8e+07 ++ : |
| : |
6e+07 ++ : |
4e+07 ++ : |
| : |
2e+07 ++ : |
0 *+-*--------------------------------------------------------------+
netperf.Throughput_tps
6000 ++-------------------------------------------------------------------+
| |
5000 ++ O O O O O O O O O O O O O O O O O O O O O O
O O O..O..O.*..*..*.*..*..*.*..*..*.*..*.*..*..*.*..* |
| : |
4000 ++ : |
| : |
3000 ++ : |
| : |
2000 ++ : |
| : |
| : |
1000 ++ : |
| : |
0 *+-*-----------------------------------------------------------------+
netperf.time.system_time
3500 O+-O-O--O--O-O--O--O-O--O--O-O--O--O-O--O-O--O--O-O--O--O-O--O--O-O--O
| *..*..*.*..*..*.*..*..*.*..*..*.*..*.*..*..*.*..* |
3000 ++ : |
| : |
2500 ++ : |
| : |
2000 ++ : |
| : |
1500 ++ : |
| : |
1000 ++ : |
| : |
500 ++ : |
| : |
0 *+-*-----------------------------------------------------------------+
netperf.time.percent_of_cpu_this_job_got
1200 O+-O-O--O--O-O--O--O-O--O--O-O--O--O-O--O-O--O--O-O--O--O-O--O--O-O--O
| *..*..*.*..*..*.*..*..*.*..*..*.*..*.*..*..*.*..* |
1000 ++ : |
| : |
| : |
800 ++ : |
| : |
600 ++ : |
| : |
400 ++ : |
| : |
| : |
200 ++ : |
| : |
0 *+-*-----------------------------------------------------------------+
netperf.time.voluntary_context_switches
5e+07 ++----------------------------------------------------------------O
4.5e+07 O+ O O..O.O..O.O O.O..O.O..O.O..O..O.O..O.O..O.O..O O O O O O |
| : *..* |
4e+07 ++ : |
3.5e+07 ++ : |
| : |
3e+07 ++ : |
2.5e+07 ++ : |
2e+07 ++ : |
| : |
1.5e+07 ++ : |
1e+07 ++ : |
| : |
5e+06 ++ : |
0 *+-*--------------------------------------------------------------+
netperf.time.involuntary_context_switches
5e+07 ++----------------------------------------------------------------+
4.5e+07 ++ *..*.*..*.*..*.*..*.*..*.*..*..*.*..*.*..*.*..* |
| : |
4e+07 ++ : |
3.5e+07 ++ : |
| : |
3e+07 ++ : |
2.5e+07 ++ : |
2e+07 ++ : |
| : |
1.5e+07 ++ : |
1e+07 ++ : |
| : |
5e+06 O+ O O O O O O O O O O O O O O O O O O O O O O O O O O
0 *+-*--------------------------------------------------------------+
vmstat.system.cs
700000 ++-----------------------------------------------------------------+
| |
600000 ++ *..*.*..*.*..*..*.*..*.*..*..*.*..*.*..*.*..*..* |
| : |
| : |
500000 ++ : |
| : |
400000 ++ : |
| O:O O O O O O O O O O O O O O O O O O O O O O O O
300000 O+ : O |
| : |
| : |
200000 ++ : |
| : |
100000 *+-*---------------------------------------------------------------+
[*] bisect-good sample
[O] bisect-bad sample
***************************************************************************************************
ivb43: 48 threads Intel(R) Xeon(R) CPU E5-2697 v2 @ 2.70GHz with 64G memory
=========================================================================================
cluster/compiler/cpufreq_governor/ip/kconfig/nr_threads/rootfs/runtime/tbox_group/test/testcase:
cs-localhost/gcc-6/performance/ipv4/x86_64-rhel-7.2/200%/debian-x86_64-2016-08-31.cgz/300s/ivb43/TCP_CRR/netperf
commit:
269ebce453 ("xen-netfront: cast grant table reference first to type int")
14135f30e3 ("inet: fix sleeping inside inet_wait_for_connect()")
269ebce4531b8edc 14135f30e33ce37b22529f7366
---------------- --------------------------
fail:runs %reproduction fail:runs
| | |
%stddev %change %stddev
\ | \
4801 ± 0% +7.0% 5140 ± 0% netperf.Throughput_tps
1.386e+08 ± 0% -99.7% 460648 ± 2% netperf.time.involuntary_context_switches
3307 ± 0% +7.3% 3549 ± 0% netperf.time.percent_of_cpu_this_job_got
9612 ± 0% +7.9% 10373 ± 0% netperf.time.system_time
483.92 ± 0% -4.9% 460.02 ± 0% netperf.time.user_time
1.382e+08 ± 0% +7.0% 1.48e+08 ± 0% netperf.time.voluntary_context_switches
122730 ± 13% -14.1% 105402 ± 3% cpuidle.C1E-IVT.time
293198 ± 3% +11.0% 325579 ± 2% meminfo.Committed_AS
21.50 ± 34% -80.2% 4.25 ±159% numa-numastat.node1.other_node
1814908 ± 0% -46.0% 979599 ± 0% vmstat.system.cs
306.00 ± 52% -64.7% 108.00 ± 0% slabinfo.flow_cache.active_objs
306.00 ± 52% -64.7% 108.00 ± 0% slabinfo.flow_cache.num_objs
18414 ± 78% +157.3% 47389 ± 25% numa-meminfo.node0.AnonHugePages
9635 ± 20% +12.8% 10864 ± 20% numa-meminfo.node0.Mapped
38777 ± 37% -71.0% 11246 ±116% numa-meminfo.node1.AnonHugePages
68075 ± 26% -49.2% 34558 ± 44% numa-meminfo.node1.AnonPages
875482 ± 4% +10.1% 963960 ± 3% numa-vmstat.node0.numa_hit
875475 ± 4% +10.1% 963939 ± 3% numa-vmstat.node0.numa_local
17027 ± 26% -49.2% 8641 ± 44% numa-vmstat.node1.nr_anon_pages
18.50 ± 33% -79.7% 3.75 ±173% numa-vmstat.node1.numa_other
2.40 ± 0% -1.2% 2.37 ± 0% perf-stat.branch-miss-rate%
6.743e+10 ± 0% -1.1% 6.668e+10 ± 0% perf-stat.branch-misses
1.33 ± 0% +7.6% 1.43 ± 1% perf-stat.cache-miss-rate%
5.166e+11 ± 0% -5.2% 4.899e+11 ± 0% perf-stat.cache-references
5.572e+08 ± 0% -46.0% 3.007e+08 ± 0% perf-stat.context-switches
42492 ± 3% +16.9% 49690 ± 4% perf-stat.cpu-migrations
0.14 ± 2% -17.9% 0.11 ± 5% perf-stat.dTLB-store-miss-rate%
4.345e+09 ± 2% -17.4% 3.59e+09 ± 5% perf-stat.dTLB-store-misses
1.117e+10 ± 2% -9.0% 1.017e+10 ± 5% perf-stat.iTLB-load-misses
4.551e+09 ± 0% +3.0% 4.687e+09 ± 1% perf-stat.node-loads
1638442 ± 4% +128.1% 3737951 ± 5% sched_debug.cfs_rq:/.MIN_vruntime.avg
5420046 ± 5% +10.7% 5999630 ± 0% sched_debug.cfs_rq:/.MIN_vruntime.max
2219277 ± 1% +22.3% 2714119 ± 2% sched_debug.cfs_rq:/.MIN_vruntime.stddev
26.08 ± 0% +22.4% 31.92 ± 5% sched_debug.cfs_rq:/.load_avg.min
1638442 ± 4% +128.1% 3737951 ± 5% sched_debug.cfs_rq:/.max_vruntime.avg
5420046 ± 5% +10.7% 5999630 ± 0% sched_debug.cfs_rq:/.max_vruntime.max
2219277 ± 1% +22.3% 2714119 ± 2% sched_debug.cfs_rq:/.max_vruntime.stddev
4765179 ± 0% +20.9% 5758756 ± 0% sched_debug.cfs_rq:/.min_vruntime.avg
4303812 ± 2% +27.4% 5483885 ± 0% sched_debug.cfs_rq:/.min_vruntime.min
260120 ± 21% -60.8% 101919 ± 17% sched_debug.cfs_rq:/.min_vruntime.stddev
1.17 ± 0% +20.9% 1.41 ± 1% sched_debug.cfs_rq:/.nr_running.avg
23.84 ± 1% +24.8% 29.76 ± 1% sched_debug.cfs_rq:/.runnable_load_avg.avg
41.46 ± 5% +44.3% 59.83 ± 28% sched_debug.cfs_rq:/.runnable_load_avg.max
6.06 ± 4% +83.6% 11.13 ± 15% sched_debug.cfs_rq:/.runnable_load_avg.stddev
-509334 ±-14% -56.5% -221562 ±-56% sched_debug.cfs_rq:/.spread0.min
260093 ± 21% -60.8% 101925 ± 17% sched_debug.cfs_rq:/.spread0.stddev
494239 ± 29% -41.8% 287579 ± 26% sched_debug.cpu.avg_idle.max
92115 ± 24% -41.3% 54084 ± 14% sched_debug.cpu.avg_idle.stddev
23.44 ± 1% +26.0% 29.52 ± 1% sched_debug.cpu.cpu_load[0].avg
35.21 ± 2% +44.9% 51.00 ± 8% sched_debug.cpu.cpu_load[0].max
5.76 ± 3% +79.3% 10.32 ± 5% sched_debug.cpu.cpu_load[0].stddev
24.07 ± 3% +24.6% 29.99 ± 0% sched_debug.cpu.cpu_load[1].avg
23.79 ± 1% +27.7% 30.39 ± 0% sched_debug.cpu.cpu_load[2].avg
23.33 ± 0% +31.3% 30.64 ± 0% sched_debug.cpu.cpu_load[3].avg
38.96 ± 19% +33.4% 51.96 ± 7% sched_debug.cpu.cpu_load[3].max
22.41 ± 0% +44.5% 32.38 ± 0% sched_debug.cpu.cpu_load[4].avg
34.88 ± 9% +53.9% 53.67 ± 5% sched_debug.cpu.cpu_load[4].max
3.44 ± 11% +65.0% 5.68 ± 12% sched_debug.cpu.cpu_load[4].stddev
5739095 ± 0% -46.0% 3096722 ± 0% sched_debug.cpu.nr_switches.avg
5958477 ± 0% -44.6% 3300572 ± 1% sched_debug.cpu.nr_switches.max
5585409 ± 0% -46.3% 2996720 ± 0% sched_debug.cpu.nr_switches.min
78557 ± 8% -21.5% 61656 ± 12% sched_debug.cpu.nr_switches.stddev
0.03 ± 25% -78.7% 0.01 ±173% sched_debug.rt_rq:/.rt_nr_running.stddev
0.05 ± 7% -37.7% 0.03 ± 13% sched_debug.rt_rq:/.rt_time.avg
1.78 ± 7% -26.5% 1.31 ± 13% sched_debug.rt_rq:/.rt_time.max
0.26 ± 6% -27.6% 0.19 ± 13% sched_debug.rt_rq:/.rt_time.stddev
5.44 ± 0% +10.4% 6.00 ± 1% perf-profile.calltrace.cycles-pp.SYSC_accept4.sys_accept.entry_SYSCALL_64_fastpath
18.98 ± 0% -32.6% 12.80 ± 0% perf-profile.calltrace.cycles-pp.SYSC_recvfrom.sys_recvfrom.entry_SYSCALL_64_fastpath
13.18 ± 0% +33.3% 17.56 ± 0% perf-profile.calltrace.cycles-pp.SYSC_sendto.sys_sendto.entry_SYSCALL_64_fastpath
2.55 ± 1% +10.1% 2.81 ± 2% perf-profile.calltrace.cycles-pp.__dentry_kill.dput.__fput.____fput.task_work_run
0.81 ± 2% +13.8% 0.93 ± 7% perf-profile.calltrace.cycles-pp.__destroy_inode.destroy_inode.evict.iput.dentry_unlink_inode
0.00 ± -1% +Inf% 1.30 ± 2% perf-profile.calltrace.cycles-pp.__release_sock.release_sock.tcp_sendmsg.inet_sendmsg.sock_sendmsg
1.41 ± 1% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.__schedule.schedule.schedule_timeout.__inet_stream_connect.inet_stream_connect
1.69 ± 1% +14.6% 1.94 ± 0% perf-profile.calltrace.cycles-pp.__schedule.schedule.schedule_timeout.inet_csk_accept.inet_accept
2.71 ± 1% -17.9% 2.22 ± 2% perf-profile.calltrace.cycles-pp.__schedule.schedule.schedule_timeout.sk_wait_data.tcp_recvmsg
10.58 ± 0% -46.9% 5.62 ± 0% perf-profile.calltrace.cycles-pp.__tcp_ack_snd_check.tcp_rcv_established.tcp_v4_do_rcv.tcp_prequeue_process.tcp_recvmsg
0.00 ± -1% +Inf% 1.36 ± 0% perf-profile.calltrace.cycles-pp.__tcp_ack_snd_check.tcp_rcv_established.tcp_v4_do_rcv.tcp_v4_rcv.ip_local_deliver_finish
10.23 ± 0% +28.1% 13.11 ± 0% perf-profile.calltrace.cycles-pp.__tcp_push_pending_frames.tcp_push.tcp_sendmsg.inet_sendmsg.sock_sendmsg
1.82 ± 1% +12.7% 2.05 ± 1% perf-profile.calltrace.cycles-pp.__wake_up_common.__wake_up_sync_key.sock_def_readable.tcp_child_process.tcp_v4_rcv
2.46 ± 0% -40.2% 1.47 ± 1% perf-profile.calltrace.cycles-pp.__wake_up_common.__wake_up_sync_key.tcp_prequeue.tcp_v4_rcv.ip_local_deliver_finish
1.87 ± 1% +12.9% 2.11 ± 1% perf-profile.calltrace.cycles-pp.__wake_up_sync_key.sock_def_readable.tcp_child_process.tcp_v4_rcv.ip_local_deliver_finish
2.58 ± 0% -39.8% 1.55 ± 0% perf-profile.calltrace.cycles-pp.__wake_up_sync_key.tcp_prequeue.tcp_v4_rcv.ip_local_deliver_finish.ip_local_deliver
2.22 ± 1% -20.9% 1.75 ± 1% perf-profile.calltrace.cycles-pp.activate_task.ttwu_do_activate.try_to_wake_up.default_wake_function.autoremove_wake_function
1.76 ± 1% +12.7% 1.98 ± 1% perf-profile.calltrace.cycles-pp.autoremove_wake_function.__wake_up_common.__wake_up_sync_key.sock_def_readable.tcp_child_process
2.38 ± 0% -40.4% 1.42 ± 1% perf-profile.calltrace.cycles-pp.autoremove_wake_function.__wake_up_common.__wake_up_sync_key.tcp_prequeue.tcp_v4_rcv
1.23 ± 1% -27.1% 0.90 ± 3% perf-profile.calltrace.cycles-pp.deactivate_task.__schedule.schedule.schedule_timeout.sk_wait_data
1.73 ± 2% +12.7% 1.96 ± 1% perf-profile.calltrace.cycles-pp.default_wake_function.autoremove_wake_function.__wake_up_common.__wake_up_sync_key.sock_def_readable
2.34 ± 0% -40.5% 1.39 ± 1% perf-profile.calltrace.cycles-pp.default_wake_function.autoremove_wake_function.__wake_up_common.__wake_up_sync_key.tcp_prequeue
2.28 ± 1% +10.4% 2.52 ± 2% perf-profile.calltrace.cycles-pp.dentry_unlink_inode.__dentry_kill.dput.__fput.____fput
1.63 ± 1% -12.7% 1.43 ± 2% perf-profile.calltrace.cycles-pp.dequeue_task_fair.deactivate_task.__schedule.schedule.schedule_timeout
1.17 ± 2% +13.6% 1.33 ± 6% perf-profile.calltrace.cycles-pp.destroy_inode.evict.iput.dentry_unlink_inode.__dentry_kill
1.69 ± 1% -26.6% 1.24 ± 3% perf-profile.calltrace.cycles-pp.dev_hard_start_xmit.__dev_queue_xmit.dev_queue_xmit.ip_finish_output2.ip_finish_output
1.54 ± 1% -55.6% 0.68 ± 2% perf-profile.calltrace.cycles-pp.enqueue_entity.enqueue_task_fair.activate_task.ttwu_do_activate.try_to_wake_up
2.00 ± 2% -21.3% 1.58 ± 1% perf-profile.calltrace.cycles-pp.enqueue_task_fair.activate_task.ttwu_do_activate.try_to_wake_up.default_wake_function
1.69 ± 1% +11.7% 1.89 ± 4% perf-profile.calltrace.cycles-pp.evict.iput.dentry_unlink_inode.__dentry_kill.dput
2.69 ± 0% +12.5% 3.02 ± 0% perf-profile.calltrace.cycles-pp.inet_accept.SYSC_accept4.sys_accept.entry_SYSCALL_64_fastpath
2.39 ± 0% +13.1% 2.70 ± 1% perf-profile.calltrace.cycles-pp.inet_csk_accept.inet_accept.SYSC_accept4.sys_accept.entry_SYSCALL_64_fastpath
18.11 ± 0% -34.0% 11.95 ± 1% perf-profile.calltrace.cycles-pp.inet_recvmsg.sock_recvmsg.SYSC_recvfrom.sys_recvfrom.entry_SYSCALL_64_fastpath
12.60 ± 0% +35.0% 17.01 ± 0% perf-profile.calltrace.cycles-pp.inet_sendmsg.sock_sendmsg.SYSC_sendto.sys_sendto.entry_SYSCALL_64_fastpath
8.52 ± 0% -39.5% 5.15 ± 0% perf-profile.calltrace.cycles-pp.ip_local_out.ip_queue_xmit.tcp_transmit_skb.tcp_send_ack.__tcp_ack_snd_check
17.80 ± 0% +13.7% 20.25 ± 0% perf-profile.calltrace.cycles-pp.ip_local_out.ip_queue_xmit.tcp_transmit_skb.tcp_write_xmit.__tcp_push_pending_frames
17.12 ± 0% -15.4% 14.48 ± 0% perf-profile.calltrace.cycles-pp.ip_output.ip_local_out.ip_queue_xmit.tcp_transmit_skb.tcp_send_ack
17.35 ± 0% +14.3% 19.84 ± 0% perf-profile.calltrace.cycles-pp.ip_output.ip_local_out.ip_queue_xmit.tcp_transmit_skb.tcp_write_xmit
8.76 ± 0% -38.7% 5.37 ± 0% perf-profile.calltrace.cycles-pp.ip_queue_xmit.tcp_transmit_skb.tcp_send_ack.__tcp_ack_snd_check.tcp_rcv_established
8.15 ± 0% +21.4% 9.89 ± 0% perf-profile.calltrace.cycles-pp.ip_queue_xmit.tcp_transmit_skb.tcp_write_xmit.__tcp_push_pending_frames.tcp_push
2.07 ± 1% +11.0% 2.30 ± 3% perf-profile.calltrace.cycles-pp.iput.dentry_unlink_inode.__dentry_kill.dput.__fput
0.00 ± -1% +Inf% 1.38 ± 2% perf-profile.calltrace.cycles-pp.release_sock.tcp_sendmsg.inet_sendmsg.sock_sendmsg.SYSC_sendto
1.53 ± 2% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.schedule.schedule_timeout.__inet_stream_connect.inet_stream_connect.SYSC_connect
1.78 ± 1% +14.4% 2.03 ± 0% perf-profile.calltrace.cycles-pp.schedule.schedule_timeout.inet_csk_accept.inet_accept.SYSC_accept4
2.85 ± 1% -18.0% 2.34 ± 2% perf-profile.calltrace.cycles-pp.schedule.schedule_timeout.sk_wait_data.tcp_recvmsg.inet_recvmsg
1.59 ± 1% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.schedule_timeout.__inet_stream_connect.inet_stream_connect.SYSC_connect.sys_connect
1.82 ± 0% +14.3% 2.08 ± 0% perf-profile.calltrace.cycles-pp.schedule_timeout.inet_csk_accept.inet_accept.SYSC_accept4.sys_accept
2.93 ± 1% -18.2% 2.39 ± 2% perf-profile.calltrace.cycles-pp.schedule_timeout.sk_wait_data.tcp_recvmsg.inet_recvmsg.sock_recvmsg
0.13 ±173% +627.5% 0.93 ± 24% perf-profile.calltrace.cycles-pp.security_sock_rcv_skb.sk_filter_trim_cap.tcp_v4_rcv.ip_local_deliver_finish.ip_local_deliver
0.53 ± 4% +124.9% 1.20 ± 1% perf-profile.calltrace.cycles-pp.sk_filter_trim_cap.tcp_v4_rcv.ip_local_deliver_finish.ip_local_deliver.ip_rcv_finish
3.42 ± 0% -20.7% 2.71 ± 2% perf-profile.calltrace.cycles-pp.sk_wait_data.tcp_recvmsg.inet_recvmsg.sock_recvmsg.SYSC_recvfrom
1.14 ± 2% +15.8% 1.32 ± 1% perf-profile.calltrace.cycles-pp.sock_alloc_file.SYSC_accept4.sys_accept.entry_SYSCALL_64_fastpath
1.91 ± 1% +12.8% 2.16 ± 1% perf-profile.calltrace.cycles-pp.sock_def_readable.tcp_child_process.tcp_v4_rcv.ip_local_deliver_finish.ip_local_deliver
18.52 ± 0% -33.3% 12.36 ± 1% perf-profile.calltrace.cycles-pp.sock_recvmsg.SYSC_recvfrom.sys_recvfrom.entry_SYSCALL_64_fastpath
12.86 ± 0% +34.5% 17.30 ± 0% perf-profile.calltrace.cycles-pp.sock_sendmsg.SYSC_sendto.sys_sendto.entry_SYSCALL_64_fastpath
5.55 ± 0% +10.5% 6.13 ± 1% perf-profile.calltrace.cycles-pp.sys_accept.entry_SYSCALL_64_fastpath
19.05 ± 0% -32.4% 12.88 ± 1% perf-profile.calltrace.cycles-pp.sys_recvfrom.entry_SYSCALL_64_fastpath
13.20 ± 0% +33.4% 17.62 ± 0% perf-profile.calltrace.cycles-pp.sys_sendto.entry_SYSCALL_64_fastpath
0.99 ± 0% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.tcp_ack.tcp_rcv_established.tcp_v4_do_rcv.tcp_v4_rcv.ip_local_deliver_finish
3.13 ± 0% +9.8% 3.44 ± 1% perf-profile.calltrace.cycles-pp.tcp_child_process.tcp_v4_rcv.ip_local_deliver_finish.ip_local_deliver.ip_rcv_finish
1.23 ± 0% -19.5% 0.99 ± 0% perf-profile.calltrace.cycles-pp.tcp_finish_connect.tcp_rcv_state_process.tcp_v4_do_rcv.__release_sock.release_sock
3.21 ± 0% -40.3% 1.92 ± 0% perf-profile.calltrace.cycles-pp.tcp_prequeue.tcp_v4_rcv.ip_local_deliver_finish.ip_local_deliver.ip_rcv_finish
13.07 ± 0% -44.3% 7.28 ± 0% perf-profile.calltrace.cycles-pp.tcp_prequeue_process.tcp_recvmsg.inet_recvmsg.sock_recvmsg.SYSC_recvfrom
10.28 ± 0% +28.0% 13.15 ± 0% perf-profile.calltrace.cycles-pp.tcp_push.tcp_sendmsg.inet_sendmsg.sock_sendmsg.SYSC_sendto
0.00 ± -1% +Inf% 1.14 ± 3% perf-profile.calltrace.cycles-pp.tcp_rcv_established.tcp_v4_do_rcv.__release_sock.release_sock.tcp_sendmsg
10.93 ± 0% -37.1% 6.88 ± 0% perf-profile.calltrace.cycles-pp.tcp_rcv_established.tcp_v4_do_rcv.tcp_prequeue_process.tcp_recvmsg.inet_recvmsg
1.47 ± 1% +47.8% 2.17 ± 1% perf-profile.calltrace.cycles-pp.tcp_rcv_established.tcp_v4_do_rcv.tcp_v4_rcv.ip_local_deliver_finish.ip_local_deliver
17.88 ± 0% -34.6% 11.69 ± 1% perf-profile.calltrace.cycles-pp.tcp_recvmsg.inet_recvmsg.sock_recvmsg.SYSC_recvfrom.sys_recvfrom
10.11 ± 0% -48.3% 5.22 ± 0% perf-profile.calltrace.cycles-pp.tcp_send_ack.__tcp_ack_snd_check.tcp_rcv_established.tcp_v4_do_rcv.tcp_prequeue_process
0.00 ± -1% +Inf% 1.30 ± 0% perf-profile.calltrace.cycles-pp.tcp_send_ack.__tcp_ack_snd_check.tcp_rcv_established.tcp_v4_do_rcv.tcp_v4_rcv
12.17 ± 0% +35.1% 16.45 ± 0% perf-profile.calltrace.cycles-pp.tcp_sendmsg.inet_sendmsg.sock_sendmsg.SYSC_sendto.sys_sendto
9.33 ± 0% -37.2% 5.87 ± 0% perf-profile.calltrace.cycles-pp.tcp_transmit_skb.tcp_send_ack.__tcp_ack_snd_check.tcp_rcv_established.tcp_v4_do_rcv
9.03 ± 0% +17.5% 10.61 ± 0% perf-profile.calltrace.cycles-pp.tcp_transmit_skb.tcp_write_xmit.__tcp_push_pending_frames.tcp_push.tcp_sendmsg
0.00 ± -1% +Inf% 1.20 ± 2% perf-profile.calltrace.cycles-pp.tcp_v4_do_rcv.__release_sock.release_sock.tcp_sendmsg.inet_sendmsg
9.46 ± 1% -43.6% 5.34 ± 0% perf-profile.calltrace.cycles-pp.tcp_v4_do_rcv.tcp_prequeue_process.tcp_recvmsg.inet_recvmsg.sock_recvmsg
10.10 ± 0% +28.1% 12.94 ± 0% perf-profile.calltrace.cycles-pp.tcp_write_xmit.__tcp_push_pending_frames.tcp_push.tcp_sendmsg.inet_sendmsg
3.99 ± 0% -18.5% 3.25 ± 0% perf-profile.calltrace.cycles-pp.try_to_wake_up.default_wake_function.autoremove_wake_function.__wake_up_common.__wake_up_sync_key
2.75 ± 1% -18.7% 2.24 ± 1% perf-profile.calltrace.cycles-pp.ttwu_do_activate.try_to_wake_up.default_wake_function.autoremove_wake_function.__wake_up_common
5.44 ± 0% +10.4% 6.01 ± 1% perf-profile.children.cycles-pp.SYSC_accept4
19.02 ± 0% -32.5% 12.83 ± 0% perf-profile.children.cycles-pp.SYSC_recvfrom
13.18 ± 0% +33.4% 17.58 ± 0% perf-profile.children.cycles-pp.SYSC_sendto
2.57 ± 1% +9.9% 2.83 ± 2% perf-profile.children.cycles-pp.__dentry_kill
0.84 ± 3% +13.4% 0.95 ± 7% perf-profile.children.cycles-pp.__destroy_inode
1.51 ± 0% +14.0% 1.73 ± 2% perf-profile.children.cycles-pp.__kfree_skb
14.36 ± 0% +14.1% 16.39 ± 0% perf-profile.children.cycles-pp.__release_sock
5.88 ± 0% -28.3% 4.22 ± 1% perf-profile.children.cycles-pp.__schedule
0.98 ± 1% +11.0% 1.09 ± 1% perf-profile.children.cycles-pp.__sk_free
10.64 ± 0% -33.7% 7.05 ± 0% perf-profile.children.cycles-pp.__tcp_ack_snd_check
23.05 ± 0% +15.7% 26.68 ± 0% perf-profile.children.cycles-pp.__tcp_push_pending_frames
4.69 ± 0% -22.3% 3.64 ± 1% perf-profile.children.cycles-pp.__wake_up_common
4.48 ± 0% -17.8% 3.68 ± 1% perf-profile.children.cycles-pp.__wake_up_sync_key
2.24 ± 1% -20.8% 1.78 ± 1% perf-profile.children.cycles-pp.activate_task
4.54 ± 0% -24.4% 3.43 ± 0% perf-profile.children.cycles-pp.autoremove_wake_function
1.95 ± 1% -12.5% 1.71 ± 2% perf-profile.children.cycles-pp.deactivate_task
4.44 ± 0% -22.4% 3.45 ± 1% perf-profile.children.cycles-pp.default_wake_function
2.31 ± 1% +10.4% 2.55 ± 3% perf-profile.children.cycles-pp.dentry_unlink_inode
1.31 ± 1% -16.7% 1.09 ± 3% perf-profile.children.cycles-pp.dequeue_entity
1.68 ± 1% -13.1% 1.46 ± 2% perf-profile.children.cycles-pp.dequeue_task_fair
1.21 ± 2% +13.8% 1.38 ± 5% perf-profile.children.cycles-pp.destroy_inode
1.57 ± 1% -26.3% 1.16 ± 2% perf-profile.children.cycles-pp.enqueue_entity
2.05 ± 2% -20.9% 1.62 ± 1% perf-profile.children.cycles-pp.enqueue_task_fair
1.73 ± 2% +11.3% 1.92 ± 3% perf-profile.children.cycles-pp.evict
2.69 ± 0% +12.5% 3.02 ± 0% perf-profile.children.cycles-pp.inet_accept
2.40 ± 0% +13.3% 2.72 ± 1% perf-profile.children.cycles-pp.inet_csk_accept
18.16 ± 0% -34.0% 11.99 ± 1% perf-profile.children.cycles-pp.inet_recvmsg
12.60 ± 0% +35.4% 17.06 ± 0% perf-profile.children.cycles-pp.inet_sendmsg
2.12 ± 1% +10.9% 2.36 ± 3% perf-profile.children.cycles-pp.iput
1.91 ± 2% -41.2% 1.12 ± 1% perf-profile.children.cycles-pp.pick_next_task_fair
14.89 ± 0% +13.6% 16.92 ± 0% perf-profile.children.cycles-pp.release_sock
6.29 ± 0% -28.9% 4.47 ± 1% perf-profile.children.cycles-pp.schedule
6.41 ± 0% -29.4% 4.52 ± 1% perf-profile.children.cycles-pp.schedule_timeout
1.15 ± 4% +8.3% 1.24 ± 2% perf-profile.children.cycles-pp.sel_netport_sid
1.11 ± 1% +10.8% 1.23 ± 1% perf-profile.children.cycles-pp.sk_free
3.43 ± 0% -20.8% 2.71 ± 2% perf-profile.children.cycles-pp.sk_wait_data
1.98 ± 1% +9.9% 2.18 ± 1% perf-profile.children.cycles-pp.sock_alloc_file
18.52 ± 0% -33.2% 12.37 ± 1% perf-profile.children.cycles-pp.sock_recvmsg
12.86 ± 0% +34.6% 17.30 ± 0% perf-profile.children.cycles-pp.sock_sendmsg
5.55 ± 0% +10.5% 6.13 ± 1% perf-profile.children.cycles-pp.sys_accept
19.10 ± 0% -32.4% 12.91 ± 1% perf-profile.children.cycles-pp.sys_recvfrom
13.23 ± 0% +33.4% 17.65 ± 0% perf-profile.children.cycles-pp.sys_sendto
3.14 ± 0% +9.8% 3.44 ± 1% perf-profile.children.cycles-pp.tcp_child_process
4.51 ± 0% +10.4% 4.98 ± 0% perf-profile.children.cycles-pp.tcp_data_queue
1.26 ± 0% -19.3% 1.02 ± 0% perf-profile.children.cycles-pp.tcp_finish_connect
3.40 ± 0% -37.8% 2.11 ± 0% perf-profile.children.cycles-pp.tcp_prequeue
13.08 ± 0% -44.2% 7.29 ± 0% perf-profile.children.cycles-pp.tcp_prequeue_process
10.38 ± 0% +27.8% 13.26 ± 0% perf-profile.children.cycles-pp.tcp_push
12.41 ± 0% -17.7% 10.21 ± 0% perf-profile.children.cycles-pp.tcp_rcv_established
1.56 ± 1% +15.4% 1.80 ± 0% perf-profile.children.cycles-pp.tcp_rearm_rto
17.92 ± 0% -34.6% 11.73 ± 1% perf-profile.children.cycles-pp.tcp_recvmsg
20.52 ± 0% -13.9% 17.68 ± 0% perf-profile.children.cycles-pp.tcp_send_ack
12.19 ± 0% +35.4% 16.49 ± 0% perf-profile.children.cycles-pp.tcp_sendmsg
22.83 ± 0% +15.7% 26.42 ± 0% perf-profile.children.cycles-pp.tcp_write_xmit
4.30 ± 0% -22.1% 3.35 ± 1% perf-profile.children.cycles-pp.try_to_wake_up
2.77 ± 1% -18.7% 2.25 ± 1% perf-profile.children.cycles-pp.ttwu_do_activate
1.22 ± 0% +12.5% 1.38 ± 1% perf-profile.self.cycles-pp.tcp_ack
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
3 years, 8 months
- 1
- 0
[lkp] [fs] 27a23abdf6: vm-scalability.throughput 44.7% improvement
by kernel test robot
Greeting,
FYI, we noticed a 44.7% improvement of vm-scalability.throughput due to commit:
commit 27a23abdf63184c5853c6e5004cf584b880d26fb ("fs: only set S_VERSION when updating times if it has been queried")
git://git.samba.org/jlayton/linux iversion
in testcase: vm-scalability
on test machine: 4 threads Intel(R) Core(TM) i3-3220 CPU @ 3.30GHz with 8G memory
with following parameters:
runtime: 300s
size: 1T
test: msync-mt
cpufreq_governor: performance
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/
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
=========================================================================================
compiler/cpufreq_governor/kconfig/rootfs/runtime/size/tbox_group/test/testcase:
gcc-6/performance/x86_64-rhel-7.2/debian-x86_64-2016-08-31.cgz/300s/1T/lkp-ivb-d02/msync-mt/vm-scalability
commit:
5846f3b347 ("fs: ensure we set the INODE_I_VERSION_QUERIED flag when loading from disk")
27a23abdf6 ("fs: only set S_VERSION when updating times if it has been queried")
5846f3b347857111 27a23abdf63184c5853c6e5004
---------------- --------------------------
fail:runs %reproduction fail:runs
| | |
%stddev %change %stddev
\ | \
1875009 ± 0% +44.7% 2713468 ± 0% vm-scalability.throughput
7.198e+08 ± 0% +42.4% 1.025e+09 ± 0% vm-scalability.time.file_system_outputs
1995780 ± 5% +11.1% 2217616 ± 3% vm-scalability.time.involuntary_context_switches
2.354e+08 ± 0% +12.3% 2.643e+08 ± 0% vm-scalability.time.minor_page_faults
253.75 ± 0% -6.7% 236.67 ± 0% vm-scalability.time.percent_of_cpu_this_job_got
1634 ± 0% -26.4% 1202 ± 0% vm-scalability.time.system_time
627.39 ± 0% +43.6% 900.65 ± 0% vm-scalability.time.user_time
1.337e+08 ± 1% -23.2% 1.027e+08 ± 0% vm-scalability.time.voluntary_context_switches
1.432e+08 ± 0% +63.4% 2.341e+08 ± 0% interrupts.CAL:Function_call_interrupts
930176 ± 0% -10.4% 833574 ± 0% softirqs.SCHED
1130 ± 8% +14.8% 1297 ± 6% slabinfo.cred_jar.num_objs
5232 ± 7% +26.1% 6598 ± 3% slabinfo.kmalloc-64.active_objs
5232 ± 7% +27.9% 6693 ± 5% slabinfo.kmalloc-64.num_objs
403566 ± 0% +42.7% 576015 ± 0% vmstat.io.bo
302951 ± 1% -23.0% 233307 ± 0% vmstat.system.cs
172265 ± 0% +59.8% 275353 ± 0% vmstat.system.in
80.29 ± 0% +1.3% 81.31 ± 0% turbostat.%Busy
2643 ± 0% +1.3% 2677 ± 0% turbostat.Avg_MHz
16.17 ± 1% -16.2% 13.55 ± 1% turbostat.CPU%c1
0.04 ± 22% +157.8% 0.10 ± 39% turbostat.CPU%c3
3.50 ± 5% +43.9% 5.04 ± 5% turbostat.CPU%c6
3.58e+08 ± 1% -51.4% 1.739e+08 ± 1% cpuidle.C1-IVB.time
1.092e+08 ± 1% -39.0% 66666567 ± 1% cpuidle.C1-IVB.usage
3.014e+08 ± 2% +47.7% 4.452e+08 ± 1% cpuidle.C6-IVB.time
390381 ± 1% +39.5% 544769 ± 0% cpuidle.C6-IVB.usage
12016507 ± 11% -29.8% 8437750 ± 15% cpuidle.POLL.time
2631079 ± 4% -21.5% 2065701 ± 8% cpuidle.POLL.usage
1.932e+12 ± 0% +19.9% 2.317e+12 ± 0% perf-stat.branch-instructions
0.37 ± 0% -35.1% 0.24 ± 0% perf-stat.branch-miss-rate%
7.219e+09 ± 0% -22.2% 5.618e+09 ± 1% perf-stat.branch-misses
26.99 ± 0% +44.7% 39.05 ± 0% perf-stat.cache-miss-rate%
1.937e+10 ± 1% +43.0% 2.771e+10 ± 0% perf-stat.cache-misses
2.701e+08 ± 1% -23.2% 2.076e+08 ± 0% perf-stat.context-switches
2275837 ± 3% -6.0% 2138633 ± 4% perf-stat.cpu-migrations
0.42 ± 22% -22.4% 0.32 ± 8% perf-stat.dTLB-load-miss-rate%
1.974e+12 ± 0% +10.9% 2.189e+12 ± 0% perf-stat.dTLB-loads
1.125e+12 ± 0% +7.6% 1.21e+12 ± 0% perf-stat.dTLB-stores
41.66 ± 1% +18.9% 49.53 ± 0% perf-stat.iTLB-load-miss-rate%
1.246e+09 ± 1% -6.6% 1.164e+09 ± 0% perf-stat.iTLB-load-misses
1.744e+09 ± 1% -32.0% 1.186e+09 ± 0% perf-stat.iTLB-loads
8.184e+12 ± 0% +15.8% 9.477e+12 ± 0% perf-stat.instructions
6570 ± 1% +24.0% 8144 ± 0% perf-stat.instructions-per-iTLB-miss
0.85 ± 0% +15.3% 0.98 ± 0% perf-stat.ipc
2.291e+08 ± 0% +11.4% 2.552e+08 ± 0% perf-stat.minor-faults
2.291e+08 ± 0% +11.4% 2.552e+08 ± 0% perf-stat.page-faults
18612 ±173% -29.0% 13211 ± 70% latency_stats.avg.balance_dirty_pages.balance_dirty_pages_ratelimited.handle_mm_fault.__do_page_fault.do_page_fault.page_fault
3448 ±173% +3084.5% 109808 ±126% latency_stats.avg.call_rwsem_down_write_failed.path_openat.do_filp_open.do_sys_open.SyS_open.entry_SYSCALL_64_fastpath
164779 ± 45% +937.8% 1710049 ±132% latency_stats.avg.max
9316788 ± 2% -74.2% 2401982 ± 1% latency_stats.hits.call_rwsem_down_write_failed.xfs_ilock.xfs_file_iomap_begin_delay.xfs_file_iomap_begin.iomap_apply.iomap_page_mkwrite.xfs_filemap_page_mkwrite.do_page_mkwrite.handle_mm_fault.__do_page_fault.do_page_fault.page_fault
60275 ± 11% -98.8% 723.67 ± 15% latency_stats.hits.call_rwsem_down_write_failed.xfs_ilock.xfs_vn_update_time.file_update_time.xfs_filemap_page_mkwrite.do_page_mkwrite.do_wp_page.handle_mm_fault.__do_page_fault.do_page_fault.page_fault
10460952 ± 2% -99.5% 54387 ± 1% latency_stats.hits.call_rwsem_down_write_failed.xfs_ilock.xfs_vn_update_time.file_update_time.xfs_filemap_page_mkwrite.do_page_mkwrite.handle_mm_fault.__do_page_fault.do_page_fault.page_fault
6644 ±102% +1139.5% 82360 ± 83% latency_stats.max.balance_dirty_pages.balance_dirty_pages_ratelimited.do_wp_page.handle_mm_fault.__do_page_fault.do_page_fault.page_fault
50765 ±173% -73.2% 13583 ± 70% latency_stats.max.balance_dirty_pages.balance_dirty_pages_ratelimited.handle_mm_fault.__do_page_fault.do_page_fault.page_fault
3448 ±173% +3084.5% 109808 ±126% latency_stats.max.call_rwsem_down_write_failed.path_openat.do_filp_open.do_sys_open.SyS_open.entry_SYSCALL_64_fastpath
21289 ± 32% -60.8% 8345 ± 21% latency_stats.max.call_rwsem_down_write_failed.xfs_ilock.xfs_vn_update_time.file_update_time.xfs_filemap_page_mkwrite.do_page_mkwrite.do_wp_page.handle_mm_fault.__do_page_fault.do_page_fault.page_fault
188896 ± 35% +920.0% 1926824 ±109% latency_stats.max.max
1584 ± 83% +2121.8% 35210 ±124% latency_stats.max.pipe_write.__vfs_write.vfs_write.SyS_write.entry_SYSCALL_64_fastpath
38120 ± 9% +699.6% 304826 ± 87% latency_stats.max.wait_on_page_bit_killable.__lock_page_or_retry.filemap_fault.xfs_filemap_fault.__do_fault.handle_mm_fault.__do_page_fault.do_page_fault.page_fault
245991 ±101% +2594.7% 6628814 ± 50% latency_stats.sum.balance_dirty_pages.balance_dirty_pages_ratelimited.do_wp_page.handle_mm_fault.__do_page_fault.do_page_fault.page_fault
2959446 ±173% -70.2% 881433 ±100% latency_stats.sum.balance_dirty_pages.balance_dirty_pages_ratelimited.handle_mm_fault.__do_page_fault.do_page_fault.page_fault
1.106e+08 ± 2% -84.2% 17511262 ± 2% latency_stats.sum.call_rwsem_down_read_failed.xfs_ilock.xfs_ilock_data_map_shared.__xfs_get_blocks.xfs_get_blocks.do_mpage_readpage.mpage_readpages.xfs_vm_readpages.__do_page_cache_readahead.ondemand_readahead.page_cache_async_readahead.filemap_fault
3448 ±173% +3084.5% 109808 ±126% latency_stats.sum.call_rwsem_down_write_failed.path_openat.do_filp_open.do_sys_open.SyS_open.entry_SYSCALL_64_fastpath
1052706 ± 10% -81.3% 196388 ± 10% latency_stats.sum.call_rwsem_down_write_failed.xfs_ilock.xfs_file_iomap_begin_delay.xfs_file_iomap_begin.iomap_apply.iomap_page_mkwrite.xfs_filemap_page_mkwrite.do_page_mkwrite.do_wp_page.handle_mm_fault.__do_page_fault.do_page_fault
1724048 ± 6% -98.1% 33076 ± 18% latency_stats.sum.call_rwsem_down_write_failed.xfs_ilock.xfs_vn_update_time.file_update_time.xfs_filemap_page_mkwrite.do_page_mkwrite.do_wp_page.handle_mm_fault.__do_page_fault.do_page_fault.page_fault
65197018 ± 2% -97.9% 1361430 ± 7% latency_stats.sum.call_rwsem_down_write_failed.xfs_ilock.xfs_vn_update_time.file_update_time.xfs_filemap_page_mkwrite.do_page_mkwrite.handle_mm_fault.__do_page_fault.do_page_fault.page_fault
4850 ± 74% +718.5% 39697 ±104% latency_stats.sum.pipe_write.__vfs_write.vfs_write.SyS_write.entry_SYSCALL_64_fastpath
1910 ± 53% +157.6% 4919 ± 48% proc-vmstat.allocstall_movable
12328 ± 84% -84.8% 1868 ±141% proc-vmstat.compact_isolated
6519 ± 85% -84.8% 988.00 ±141% proc-vmstat.compact_migrate_scanned
8597 ± 7% +50.6% 12950 ± 8% proc-vmstat.kswapd_high_wmark_hit_quickly
8549 ± 5% +53.8% 13146 ± 15% proc-vmstat.kswapd_low_wmark_hit_quickly
89960322 ± 0% +42.4% 1.281e+08 ± 0% proc-vmstat.nr_dirtied
242.25 ± 43% +262.6% 878.33 ± 71% proc-vmstat.nr_pages_scanned
781.75 ± 54% +3.4e+06% 26492929 ± 83% proc-vmstat.nr_vmscan_immediate_reclaim
169.25 ±148% +85080.3% 144167 ± 75% proc-vmstat.nr_vmscan_write
89934329 ± 0% +42.5% 1.281e+08 ± 0% proc-vmstat.nr_written
1.529e+08 ± 0% +41.3% 2.16e+08 ± 0% proc-vmstat.numa_hit
1.529e+08 ± 0% +41.3% 2.16e+08 ± 0% proc-vmstat.numa_local
18374 ± 6% +46.0% 26827 ± 10% proc-vmstat.pageoutrun
91616308 ± 0% +42.2% 1.302e+08 ± 1% proc-vmstat.pgactivate
1.156e+08 ± 5% +48.9% 1.721e+08 ± 9% proc-vmstat.pgalloc_normal
57204038 ± 2% +44.4% 82585962 ± 1% proc-vmstat.pgdeactivate
2.805e+08 ± 0% +17.0% 3.282e+08 ± 0% proc-vmstat.pgfault
1.518e+08 ± 1% +42.1% 2.158e+08 ± 0% proc-vmstat.pgfree
6049 ± 85% -85.7% 868.00 ±141% proc-vmstat.pgmigrate_success
3632 ± 3% +15.7% 4201 ± 0% proc-vmstat.pgpgin
3.598e+08 ± 0% +42.4% 5.125e+08 ± 0% proc-vmstat.pgpgout
57208615 ± 2% +44.4% 82590561 ± 1% proc-vmstat.pgrefill
1342923 ± 31% +119.4% 2945897 ± 8% proc-vmstat.pgrotated
39633539 ± 3% +129.5% 90958270 ± 29% proc-vmstat.pgscan_direct
1.971e+08 ± 3% +46.1% 2.881e+08 ± 3% proc-vmstat.pgscan_kswapd
1054030 ± 28% +73.8% 1831908 ± 26% proc-vmstat.pgsteal_direct
96725698 ± 2% +45.7% 1.409e+08 ± 3% proc-vmstat.pgsteal_kswapd
27324643 ± 2% +44.2% 39400907 ± 1% proc-vmstat.workingset_refault
24191 ± 25% -56.7% 10478 ± 49% sched_debug.cfs_rq:/.min_vruntime.stddev
36.57 ± 16% +53.9% 56.29 ± 11% sched_debug.cfs_rq:/.runnable_load_avg.min
-11184 ±-71% -89.4% -1190 ±-188% sched_debug.cfs_rq:/.spread0.avg
-51996 ±-33% -66.7% -17324 ±-64% sched_debug.cfs_rq:/.spread0.min
24192 ± 25% -56.7% 10479 ± 49% sched_debug.cfs_rq:/.spread0.stddev
106.12 ± 14% +40.1% 148.72 ± 8% sched_debug.cfs_rq:/.util_avg.stddev
161360 ± 11% +44.6% 233250 ± 16% sched_debug.cpu.avg_idle.avg
248824 ± 8% +53.0% 380730 ± 18% sched_debug.cpu.avg_idle.max
61098 ± 8% +61.4% 98618 ± 26% sched_debug.cpu.avg_idle.stddev
43.68 ± 12% +96.6% 85.89 ± 35% sched_debug.cpu.cpu_load[1].min
46.10 ± 13% +121.0% 101.87 ± 25% sched_debug.cpu.cpu_load[2].min
55.13 ± 13% +105.7% 113.42 ± 16% sched_debug.cpu.cpu_load[3].min
60.32 ± 15% +96.9% 118.78 ± 19% sched_debug.cpu.cpu_load[4].min
3868 ± 11% +26.0% 4872 ± 12% sched_debug.cpu.nr_load_updates.stddev
32143238 ± 1% -22.7% 24857641 ± 0% sched_debug.cpu.nr_switches.avg
34824253 ± 1% -21.3% 27397413 ± 3% sched_debug.cpu.nr_switches.max
30678305 ± 1% -24.8% 23081497 ± 0% sched_debug.cpu.nr_switches.min
1095 ± 17% -32.0% 745.24 ± 3% sched_debug.cpu.nr_uninterruptible.max
824.46 ± 12% -24.2% 624.90 ± 9% sched_debug.cpu.nr_uninterruptible.stddev
32194984 ± 1% -22.6% 24920945 ± 0% sched_debug.cpu.sched_count.avg
34874301 ± 1% -21.3% 27459859 ± 3% sched_debug.cpu.sched_count.max
30731176 ± 1% -24.7% 23147150 ± 0% sched_debug.cpu.sched_count.min
15048852 ± 1% -25.3% 11247844 ± 0% sched_debug.cpu.sched_goidle.avg
15511954 ± 0% -24.1% 11770336 ± 0% sched_debug.cpu.sched_goidle.max
14785332 ± 1% -26.4% 10883202 ± 0% sched_debug.cpu.sched_goidle.min
16733610 ± 1% -21.8% 13092776 ± 0% sched_debug.cpu.ttwu_count.avg
18176852 ± 1% -20.7% 14412533 ± 2% sched_debug.cpu.ttwu_count.max
15905821 ± 2% -23.5% 12168901 ± 0% sched_debug.cpu.ttwu_count.min
518505 ± 5% -11.2% 460302 ± 1% sched_debug.cpu.ttwu_local.min
0.99 ± 8% -14.6% 0.85 ± 16% sched_debug.rt_rq:/.rt_time.avg
3.96 ± 8% -14.7% 3.37 ± 16% sched_debug.rt_rq:/.rt_time.max
1.71 ± 8% -14.8% 1.46 ± 16% sched_debug.rt_rq:/.rt_time.stddev
2.60 ± 6% -37.4% 1.63 ± 6% perf-profile.calltrace.cycles-pp.__account_scheduler_latency.enqueue_entity.enqueue_task_fair.activate_task.ttwu_do_activate
1.08 ± 21% +61.1% 1.74 ± 16% perf-profile.calltrace.cycles-pp.__blk_mq_complete_request.blk_mq_complete_request.loop_queue_work.kthread_worker_fn.kthread
7.15 ± 3% +22.1% 8.74 ± 9% perf-profile.calltrace.cycles-pp.__do_fault.handle_mm_fault.__do_page_fault.do_page_fault.page_fault
34.48 ± 25% -71.4% 9.87 ± 71% perf-profile.calltrace.cycles-pp.__do_page_fault.do_page_fault.page_fault
2.65 ± 21% +57.3% 4.16 ± 16% perf-profile.calltrace.cycles-pp.__generic_file_write_iter.generic_file_write_iter.vfs_iter_write.lo_write_bvec.loop_queue_work
1.23 ± 14% -47.5% 0.64 ± 9% perf-profile.calltrace.cycles-pp.__lock_page.iomap_page_mkwrite.xfs_filemap_page_mkwrite.do_page_mkwrite.do_wp_page
1.18 ± 18% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.__lock_page.iomap_page_mkwrite.xfs_filemap_page_mkwrite.do_page_mkwrite.handle_mm_fault
1.65 ± 8% +65.3% 2.73 ± 7% perf-profile.calltrace.cycles-pp.__lock_page_or_retry.filemap_fault.xfs_filemap_fault.__do_fault.handle_mm_fault
1.11 ± 27% +63.8% 1.83 ± 6% perf-profile.calltrace.cycles-pp.__sched_text_start.bit_wait_io.__wait_on_bit.wait_on_page_bit_killable.__lock_page_or_retry
1.78 ± 14% -79.4% 0.37 ± 71% perf-profile.calltrace.cycles-pp.__sched_text_start.bit_wait_io.__wait_on_bit_lock.__lock_page.iomap_page_mkwrite
1.71 ± 5% -23.1% 1.31 ± 11% perf-profile.calltrace.cycles-pp.__schedule.schedule.schedule_preempt_disabled.cpu_startup_entry.start_secondary
2.07 ± 6% -25.1% 1.55 ± 3% perf-profile.calltrace.cycles-pp.__tick_nohz_idle_enter.tick_nohz_idle_enter.cpu_startup_entry.start_secondary
1.52 ± 8% +54.1% 2.35 ± 7% perf-profile.calltrace.cycles-pp.__wait_on_bit.wait_on_page_bit_killable.__lock_page_or_retry.filemap_fault.xfs_filemap_fault
2.22 ± 13% -71.5% 0.63 ± 11% perf-profile.calltrace.cycles-pp.__wait_on_bit_lock.__lock_page.iomap_page_mkwrite.xfs_filemap_page_mkwrite.do_page_mkwrite
2.38 ± 12% -41.4% 1.39 ± 35% perf-profile.calltrace.cycles-pp.__wake_up.__wake_up_bit.unlock_page.do_wp_page.handle_mm_fault
2.42 ± 11% -41.8% 1.41 ± 34% perf-profile.calltrace.cycles-pp.__wake_up_bit.unlock_page.do_wp_page.handle_mm_fault.__do_page_fault
1.94 ± 10% -32.2% 1.31 ± 35% perf-profile.calltrace.cycles-pp.__wake_up_common.__wake_up.__wake_up_bit.unlock_page.do_wp_page
6.09 ± 15% +42.3% 8.67 ± 14% perf-profile.calltrace.cycles-pp.__writeback_inodes_wb.wb_writeback.wb_workfn.process_one_work.worker_thread
6.09 ± 15% +42.3% 8.67 ± 14% perf-profile.calltrace.cycles-pp.__writeback_single_inode.writeback_sb_inodes.__writeback_inodes_wb.wb_writeback.wb_workfn
1.85 ± 3% -45.6% 1.01 ± 24% perf-profile.calltrace.cycles-pp.__xfs_get_blocks.xfs_get_blocks.do_mpage_readpage.mpage_readpages.xfs_vm_readpages
5.86 ± 4% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.__xfs_trans_commit.xfs_trans_commit.xfs_vn_update_time.file_update_time.xfs_filemap_page_mkwrite
1.03 ± 12% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp._raw_spin_lock.handle_mm_fault.__do_page_fault.do_page_fault.page_fault
4.12 ± 3% -14.3% 3.53 ± 7% perf-profile.calltrace.cycles-pp.activate_task.ttwu_do_activate.try_to_wake_up.default_wake_function.wake_bit_function
1.01 ± 22% +65.8% 1.67 ± 15% perf-profile.calltrace.cycles-pp.bio_endio.blk_update_request.blk_mq_end_request.__blk_mq_complete_request.blk_mq_complete_request
1.13 ± 27% +62.8% 1.84 ± 6% perf-profile.calltrace.cycles-pp.bit_wait_io.__wait_on_bit.wait_on_page_bit_killable.__lock_page_or_retry.filemap_fault
1.81 ± 14% -79.2% 0.38 ± 70% perf-profile.calltrace.cycles-pp.bit_wait_io.__wait_on_bit_lock.__lock_page.iomap_page_mkwrite.xfs_filemap_page_mkwrite
1.08 ± 21% +61.1% 1.74 ± 16% perf-profile.calltrace.cycles-pp.blk_mq_complete_request.loop_queue_work.kthread_worker_fn.kthread.ret_from_fork
1.08 ± 21% +60.9% 1.73 ± 16% perf-profile.calltrace.cycles-pp.blk_mq_end_request.__blk_mq_complete_request.blk_mq_complete_request.loop_queue_work.kthread_worker_fn
1.04 ± 22% +63.1% 1.70 ± 16% perf-profile.calltrace.cycles-pp.blk_update_request.blk_mq_end_request.__blk_mq_complete_request.blk_mq_complete_request.loop_queue_work
3.54 ± 11% -49.5% 1.79 ± 1% perf-profile.calltrace.cycles-pp.call_cpuidle.cpu_startup_entry.rest_init.start_kernel.x86_64_start_reservations
11.68 ± 9% -27.8% 8.43 ± 2% perf-profile.calltrace.cycles-pp.call_cpuidle.cpu_startup_entry.start_secondary
1.61 ± 17% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.call_rwsem_down_write_failed.down_write.xfs_ilock.xfs_file_iomap_begin_delay.xfs_file_iomap_begin
1.93 ± 16% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.call_rwsem_down_write_failed.down_write.xfs_ilock.xfs_vn_update_time.file_update_time
0.89 ± 29% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.call_rwsem_wake.up_write.xfs_iunlock.xfs_inode_item_unlock.xfs_trans_free_items
3.70 ± 15% +46.9% 5.44 ± 16% perf-profile.calltrace.cycles-pp.clear_page_dirty_for_io.write_cache_pages.xfs_vm_writepages.do_writepages.__writeback_single_inode
6.09 ± 13% -40.6% 3.62 ± 5% perf-profile.calltrace.cycles-pp.cpu_startup_entry.rest_init.start_kernel.x86_64_start_reservations.x86_64_start_kernel
19.48 ± 8% -25.5% 14.51 ± 1% perf-profile.calltrace.cycles-pp.cpu_startup_entry.start_secondary
3.52 ± 11% -49.5% 1.78 ± 0% perf-profile.calltrace.cycles-pp.cpuidle_enter.call_cpuidle.cpu_startup_entry.rest_init.start_kernel
11.63 ± 9% -27.8% 8.40 ± 2% perf-profile.calltrace.cycles-pp.cpuidle_enter.call_cpuidle.cpu_startup_entry.start_secondary
3.45 ± 11% -49.2% 1.75 ± 0% perf-profile.calltrace.cycles-pp.cpuidle_enter_state.cpuidle_enter.call_cpuidle.cpu_startup_entry.rest_init
11.40 ± 9% -28.1% 8.20 ± 2% perf-profile.calltrace.cycles-pp.cpuidle_enter_state.cpuidle_enter.call_cpuidle.cpu_startup_entry.start_secondary
34.72 ± 25% -71.2% 9.99 ± 71% perf-profile.calltrace.cycles-pp.do_page_fault.page_fault
10.85 ± 11% -61.1% 4.22 ± 33% perf-profile.calltrace.cycles-pp.do_page_mkwrite.do_wp_page.handle_mm_fault.__do_page_fault.do_page_fault
15.75 ± 4% -56.3% 6.88 ± 7% perf-profile.calltrace.cycles-pp.do_page_mkwrite.handle_mm_fault.__do_page_fault.do_page_fault.page_fault
16.95 ± 59% +145.4% 41.59 ± 20% perf-profile.calltrace.cycles-pp.do_unit
0.63 ± 2% +47.7% 0.93 ± 32% perf-profile.calltrace.cycles-pp.do_unlinkat.sys_unlinkat.entry_SYSCALL_64_fastpath.unlinkat
13.92 ± 11% -54.6% 6.32 ± 32% perf-profile.calltrace.cycles-pp.do_wp_page.handle_mm_fault.__do_page_fault.do_page_fault.page_fault
6.09 ± 15% +42.3% 8.67 ± 14% perf-profile.calltrace.cycles-pp.do_writepages.__writeback_single_inode.writeback_sb_inodes.__writeback_inodes_wb.wb_writeback
1.96 ± 15% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.down_write.xfs_ilock.xfs_file_iomap_begin_delay.xfs_file_iomap_begin.iomap_apply
2.17 ± 15% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.down_write.xfs_ilock.xfs_vn_update_time.file_update_time.xfs_filemap_page_mkwrite
0.77 ± 24% +69.1% 1.29 ± 17% perf-profile.calltrace.cycles-pp.end_buffer_async_write.xfs_destroy_ioend.xfs_end_bio.bio_endio.blk_update_request
0.19 ±173% +405.6% 0.97 ± 15% perf-profile.calltrace.cycles-pp.end_page_writeback.end_buffer_async_write.xfs_destroy_ioend.xfs_end_bio.bio_endio
0.63 ± 2% +47.7% 0.93 ± 32% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_fastpath.unlinkat
0.63 ± 2% +47.7% 0.93 ± 32% perf-profile.calltrace.cycles-pp.evict.iput.do_unlinkat.sys_unlinkat.entry_SYSCALL_64_fastpath
5.19 ± 12% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.file_update_time.xfs_filemap_page_mkwrite.do_page_mkwrite.do_wp_page.handle_mm_fault
7.63 ± 4% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.file_update_time.xfs_filemap_page_mkwrite.do_page_mkwrite.handle_mm_fault.__do_page_fault
6.58 ± 3% +22.9% 8.09 ± 9% perf-profile.calltrace.cycles-pp.filemap_fault.xfs_filemap_fault.__do_fault.handle_mm_fault.__do_page_fault
2.68 ± 16% +49.1% 4.00 ± 15% perf-profile.calltrace.cycles-pp.flush_tlb_page.ptep_clear_flush.page_mkclean_one.rmap_walk_file.rmap_walk
2.75 ± 21% +56.9% 4.32 ± 16% perf-profile.calltrace.cycles-pp.generic_file_write_iter.vfs_iter_write.lo_write_bvec.loop_queue_work.kthread_worker_fn
2.52 ± 21% +56.7% 3.96 ± 16% perf-profile.calltrace.cycles-pp.generic_perform_write.__generic_file_write_iter.generic_file_write_iter.vfs_iter_write.lo_write_bvec
32.93 ± 25% -72.2% 9.15 ± 71% perf-profile.calltrace.cycles-pp.handle_mm_fault.__do_page_fault.do_page_fault.page_fault
14.29 ± 4% -32.9% 9.59 ± 3% perf-profile.calltrace.cycles-pp.intel_idle.cpuidle_enter_state.cpuidle_enter.call_cpuidle.cpu_startup_entry
5.15 ± 11% -28.3% 3.69 ± 32% perf-profile.calltrace.cycles-pp.iomap_page_mkwrite.xfs_filemap_page_mkwrite.do_page_mkwrite.do_wp_page.handle_mm_fault
7.42 ± 5% -18.0% 6.09 ± 7% perf-profile.calltrace.cycles-pp.iomap_page_mkwrite.xfs_filemap_page_mkwrite.do_page_mkwrite.handle_mm_fault.__do_page_fault
2.63 ± 6% +22.2% 3.21 ± 9% perf-profile.calltrace.cycles-pp.iomap_page_mkwrite_actor.iomap_apply.iomap_page_mkwrite.xfs_filemap_page_mkwrite.do_page_mkwrite
1.83 ± 22% +62.7% 2.98 ± 15% perf-profile.calltrace.cycles-pp.iov_iter_copy_from_user_atomic.generic_perform_write.__generic_file_write_iter.generic_file_write_iter.vfs_iter_write
0.63 ± 2% +47.7% 0.93 ± 32% perf-profile.calltrace.cycles-pp.iput.do_unlinkat.sys_unlinkat.entry_SYSCALL_64_fastpath.unlinkat
4.46 ± 4% +51.9% 6.78 ± 11% perf-profile.calltrace.cycles-pp.kswapd.kthread.ret_from_fork
14.70 ± 5% +49.9% 22.03 ± 7% perf-profile.calltrace.cycles-pp.kthread.ret_from_fork
4.10 ± 20% +58.8% 6.51 ± 16% perf-profile.calltrace.cycles-pp.kthread_worker_fn.kthread.ret_from_fork
2.87 ± 20% +59.0% 4.56 ± 17% perf-profile.calltrace.cycles-pp.lo_write_bvec.loop_queue_work.kthread_worker_fn.kthread.ret_from_fork
4.06 ± 20% +58.9% 6.45 ± 16% perf-profile.calltrace.cycles-pp.loop_queue_work.kthread_worker_fn.kthread.ret_from_fork
1.73 ± 21% +64.4% 2.84 ± 14% perf-profile.calltrace.cycles-pp.memcpy_erms.iov_iter_copy_from_user_atomic.generic_perform_write.__generic_file_write_iter.generic_file_write_iter
2.60 ± 15% +47.8% 3.85 ± 15% perf-profile.calltrace.cycles-pp.native_flush_tlb_others.flush_tlb_page.ptep_clear_flush.page_mkclean_one.rmap_walk_file
0.00 ± -1% +Inf% 1.04 ± 16% perf-profile.calltrace.cycles-pp.native_irq_return_iret.do_unit
34.86 ± 25% -71.2% 10.03 ± 71% perf-profile.calltrace.cycles-pp.page_fault
3.20 ± 16% +48.3% 4.74 ± 16% perf-profile.calltrace.cycles-pp.page_mkclean.clear_page_dirty_for_io.write_cache_pages.xfs_vm_writepages.do_writepages
2.91 ± 16% +49.3% 4.34 ± 15% perf-profile.calltrace.cycles-pp.page_mkclean_one.rmap_walk_file.rmap_walk.page_mkclean.clear_page_dirty_for_io
0.91 ± 11% +52.4% 1.39 ± 13% perf-profile.calltrace.cycles-pp.page_referenced.shrink_page_list.shrink_inactive_list.shrink_node_memcg.shrink_node
6.09 ± 15% +42.5% 8.68 ± 14% perf-profile.calltrace.cycles-pp.process_one_work.worker_thread.kthread.ret_from_fork
2.75 ± 16% +48.1% 4.07 ± 15% perf-profile.calltrace.cycles-pp.ptep_clear_flush.page_mkclean_one.rmap_walk_file.rmap_walk.page_mkclean
6.17 ± 13% -40.5% 3.67 ± 5% perf-profile.calltrace.cycles-pp.rest_init.start_kernel.x86_64_start_reservations.x86_64_start_kernel
14.70 ± 5% +49.9% 22.03 ± 7% perf-profile.calltrace.cycles-pp.ret_from_fork
3.17 ± 15% +48.7% 4.71 ± 15% perf-profile.calltrace.cycles-pp.rmap_walk.page_mkclean.clear_page_dirty_for_io.write_cache_pages.xfs_vm_writepages
0.82 ± 10% +55.2% 1.28 ± 14% perf-profile.calltrace.cycles-pp.rmap_walk.page_referenced.shrink_page_list.shrink_inactive_list.shrink_node_memcg
0.66 ± 8% +46.0% 0.96 ± 10% perf-profile.calltrace.cycles-pp.rmap_walk.try_to_unmap.shrink_page_list.shrink_inactive_list.shrink_node_memcg
3.11 ± 15% +49.3% 4.65 ± 15% perf-profile.calltrace.cycles-pp.rmap_walk_file.rmap_walk.page_mkclean.clear_page_dirty_for_io.write_cache_pages
0.72 ± 9% +57.1% 1.12 ± 14% perf-profile.calltrace.cycles-pp.rmap_walk_file.rmap_walk.page_referenced.shrink_page_list.shrink_inactive_list
1.58 ± 17% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.rwsem_down_write_failed.call_rwsem_down_write_failed.down_write.xfs_ilock.xfs_file_iomap_begin_delay
1.77 ± 21% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.rwsem_down_write_failed.call_rwsem_down_write_failed.down_write.xfs_ilock.xfs_vn_update_time
0.87 ± 28% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.rwsem_wake.call_rwsem_wake.up_write.xfs_iunlock.xfs_inode_item_unlock
1.65 ± 20% -47.6% 0.86 ± 18% perf-profile.calltrace.cycles-pp.save_stack_trace_tsk.__account_scheduler_latency.enqueue_entity.enqueue_task_fair.activate_task
1.83 ± 4% -23.5% 1.40 ± 10% perf-profile.calltrace.cycles-pp.schedule.schedule_preempt_disabled.cpu_startup_entry.start_secondary
1.12 ± 34% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.schedule.schedule_timeout.__sched_text_start.bit_wait_io.__wait_on_bit_lock
1.91 ± 5% -23.0% 1.47 ± 11% perf-profile.calltrace.cycles-pp.schedule_preempt_disabled.cpu_startup_entry.start_secondary
0.73 ± 62% +115.3% 1.57 ± 8% perf-profile.calltrace.cycles-pp.schedule_timeout.__sched_text_start.bit_wait_io.__wait_on_bit.wait_on_page_bit_killable
1.29 ± 30% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.schedule_timeout.__sched_text_start.bit_wait_io.__wait_on_bit_lock.__lock_page
4.07 ± 5% +50.1% 6.10 ± 11% perf-profile.calltrace.cycles-pp.shrink_inactive_list.shrink_node_memcg.shrink_node.kswapd.kthread
4.46 ± 4% +52.0% 6.77 ± 11% perf-profile.calltrace.cycles-pp.shrink_node.kswapd.kthread.ret_from_fork
4.42 ± 4% +52.0% 6.72 ± 11% perf-profile.calltrace.cycles-pp.shrink_node_memcg.shrink_node.kswapd.kthread.ret_from_fork
3.46 ± 5% +50.6% 5.21 ± 11% perf-profile.calltrace.cycles-pp.shrink_page_list.shrink_inactive_list.shrink_node_memcg.shrink_node.kswapd
2.55 ± 15% +46.9% 3.75 ± 15% perf-profile.calltrace.cycles-pp.smp_call_function_many.native_flush_tlb_others.flush_tlb_page.ptep_clear_flush.page_mkclean_one
6.17 ± 13% -40.5% 3.67 ± 5% perf-profile.calltrace.cycles-pp.start_kernel.x86_64_start_reservations.x86_64_start_kernel
19.74 ± 8% -25.6% 14.69 ± 2% perf-profile.calltrace.cycles-pp.start_secondary
0.63 ± 2% +47.7% 0.93 ± 32% perf-profile.calltrace.cycles-pp.sys_unlinkat.entry_SYSCALL_64_fastpath.unlinkat
2.17 ± 7% -25.4% 1.62 ± 3% perf-profile.calltrace.cycles-pp.tick_nohz_idle_enter.cpu_startup_entry.start_secondary
1.19 ± 13% -30.5% 0.82 ± 7% perf-profile.calltrace.cycles-pp.tick_nohz_idle_exit.cpu_startup_entry.start_secondary
0.87 ± 14% -35.8% 0.56 ± 2% perf-profile.calltrace.cycles-pp.tick_nohz_restart_sched_tick.tick_nohz_idle_exit.cpu_startup_entry.start_secondary
1.76 ± 7% -26.7% 1.29 ± 3% perf-profile.calltrace.cycles-pp.tick_nohz_stop_sched_tick.__tick_nohz_idle_enter.tick_nohz_idle_enter.cpu_startup_entry.start_secondary
0.63 ± 2% +47.7% 0.93 ± 32% perf-profile.calltrace.cycles-pp.truncate_inode_pages_final.evict.iput.do_unlinkat.sys_unlinkat
0.62 ± 2% +47.3% 0.92 ± 31% perf-profile.calltrace.cycles-pp.truncate_inode_pages_range.truncate_inode_pages_final.evict.iput.do_unlinkat
0.68 ± 8% +46.7% 0.99 ± 10% perf-profile.calltrace.cycles-pp.try_to_unmap.shrink_page_list.shrink_inactive_list.shrink_node_memcg.shrink_node
1.13 ± 38% -78.2% 0.25 ±141% perf-profile.calltrace.cycles-pp.try_to_wake_up.wake_up_q.rwsem_wake.call_rwsem_wake.up_write
4.37 ± 2% -14.2% 3.75 ± 7% perf-profile.calltrace.cycles-pp.ttwu_do_activate.try_to_wake_up.default_wake_function.wake_bit_function.__wake_up_common
0.63 ± 2% +47.7% 0.93 ± 32% perf-profile.calltrace.cycles-pp.unlinkat
2.45 ± 11% -41.1% 1.44 ± 34% perf-profile.calltrace.cycles-pp.unlock_page.do_wp_page.handle_mm_fault.__do_page_fault.do_page_fault
1.06 ± 23% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.up_write.xfs_iunlock.xfs_inode_item_unlock.xfs_trans_free_items.xfs_log_commit_cil
2.81 ± 20% +55.9% 4.37 ± 16% perf-profile.calltrace.cycles-pp.vfs_iter_write.lo_write_bvec.loop_queue_work.kthread_worker_fn.kthread
1.57 ± 8% +53.9% 2.42 ± 7% perf-profile.calltrace.cycles-pp.wait_on_page_bit_killable.__lock_page_or_retry.filemap_fault.xfs_filemap_fault.__do_fault
1.16 ± 39% -78.5% 0.25 ±141% perf-profile.calltrace.cycles-pp.wake_up_q.rwsem_wake.call_rwsem_wake.up_write.xfs_iunlock
6.09 ± 15% +42.3% 8.67 ± 14% perf-profile.calltrace.cycles-pp.wb_workfn.process_one_work.worker_thread.kthread.ret_from_fork
6.09 ± 15% +42.3% 8.67 ± 14% perf-profile.calltrace.cycles-pp.wb_writeback.wb_workfn.process_one_work.worker_thread.kthread
6.09 ± 15% +42.5% 8.69 ± 14% perf-profile.calltrace.cycles-pp.worker_thread.kthread.ret_from_fork
6.08 ± 15% +42.5% 8.66 ± 14% perf-profile.calltrace.cycles-pp.write_cache_pages.xfs_vm_writepages.do_writepages.__writeback_single_inode.writeback_sb_inodes
6.09 ± 15% +42.3% 8.67 ± 14% perf-profile.calltrace.cycles-pp.writeback_sb_inodes.__writeback_inodes_wb.wb_writeback.wb_workfn.process_one_work
6.17 ± 13% -40.5% 3.67 ± 5% perf-profile.calltrace.cycles-pp.x86_64_start_kernel
6.17 ± 13% -40.5% 3.67 ± 5% perf-profile.calltrace.cycles-pp.x86_64_start_reservations.x86_64_start_kernel
0.98 ± 22% +67.0% 1.64 ± 16% perf-profile.calltrace.cycles-pp.xfs_destroy_ioend.xfs_end_bio.bio_endio.blk_update_request.blk_mq_end_request
1.98 ± 15% +37.5% 2.72 ± 11% perf-profile.calltrace.cycles-pp.xfs_do_writepage.write_cache_pages.xfs_vm_writepages.do_writepages.__writeback_single_inode
0.99 ± 22% +67.2% 1.65 ± 16% perf-profile.calltrace.cycles-pp.xfs_end_bio.bio_endio.blk_update_request.blk_mq_end_request.__blk_mq_complete_request
6.07 ± 4% -33.0% 4.07 ± 8% perf-profile.calltrace.cycles-pp.xfs_file_iomap_begin.iomap_apply.iomap_page_mkwrite.xfs_filemap_page_mkwrite.do_page_mkwrite
5.80 ± 4% -35.0% 3.77 ± 8% perf-profile.calltrace.cycles-pp.xfs_file_iomap_begin_delay.xfs_file_iomap_begin.iomap_apply.iomap_page_mkwrite.xfs_filemap_page_mkwrite
7.04 ± 3% +21.8% 8.58 ± 9% perf-profile.calltrace.cycles-pp.xfs_filemap_fault.__do_fault.handle_mm_fault.__do_page_fault.do_page_fault
10.78 ± 11% -61.3% 4.17 ± 33% perf-profile.calltrace.cycles-pp.xfs_filemap_page_mkwrite.do_page_mkwrite.do_wp_page.handle_mm_fault.__do_page_fault
15.67 ± 4% -56.7% 6.79 ± 7% perf-profile.calltrace.cycles-pp.xfs_filemap_page_mkwrite.do_page_mkwrite.handle_mm_fault.__do_page_fault.do_page_fault
1.87 ± 4% -44.6% 1.04 ± 24% perf-profile.calltrace.cycles-pp.xfs_get_blocks.do_mpage_readpage.mpage_readpages.xfs_vm_readpages.__do_page_cache_readahead
2.01 ± 16% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.xfs_ilock.xfs_file_iomap_begin_delay.xfs_file_iomap_begin.iomap_apply.iomap_page_mkwrite
2.23 ± 15% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.xfs_ilock.xfs_vn_update_time.file_update_time.xfs_filemap_page_mkwrite.do_page_mkwrite
1.25 ± 3% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.xfs_inode_item_unlock.xfs_trans_free_items.xfs_log_commit_cil.__xfs_trans_commit.xfs_trans_commit
1.11 ± 22% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.xfs_iunlock.xfs_inode_item_unlock.xfs_trans_free_items.xfs_log_commit_cil.__xfs_trans_commit
4.90 ± 4% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.xfs_log_commit_cil.__xfs_trans_commit.xfs_trans_commit.xfs_vn_update_time.file_update_time
0.95 ± 50% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.xfs_log_reserve.xfs_trans_reserve.xfs_trans_alloc.xfs_vn_update_time.file_update_time
1.23 ± 14% +37.7% 1.70 ± 12% perf-profile.calltrace.cycles-pp.xfs_start_page_writeback.xfs_do_writepage.write_cache_pages.xfs_vm_writepages.do_writepages
2.81 ± 9% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.xfs_trans_alloc.xfs_vn_update_time.file_update_time.xfs_filemap_page_mkwrite.do_page_mkwrite
5.99 ± 4% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.xfs_trans_commit.xfs_vn_update_time.file_update_time.xfs_filemap_page_mkwrite.do_page_mkwrite
1.48 ± 4% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.xfs_trans_free_items.xfs_log_commit_cil.__xfs_trans_commit.xfs_trans_commit.xfs_vn_update_time
1.62 ± 20% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.xfs_trans_reserve.xfs_trans_alloc.xfs_vn_update_time.file_update_time.xfs_filemap_page_mkwrite
6.09 ± 15% +42.3% 8.67 ± 14% perf-profile.calltrace.cycles-pp.xfs_vm_writepages.do_writepages.__writeback_single_inode.writeback_sb_inodes.__writeback_inodes_wb
4.98 ± 12% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.xfs_vn_update_time.file_update_time.xfs_filemap_page_mkwrite.do_page_mkwrite.do_wp_page
7.29 ± 4% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.xfs_vn_update_time.file_update_time.xfs_filemap_page_mkwrite.do_page_mkwrite.handle_mm_fault
3.72 ± 6% -27.8% 2.69 ± 4% perf-profile.children.cycles-pp.__account_scheduler_latency
1.39 ± 23% +55.6% 2.17 ± 19% perf-profile.children.cycles-pp.__alloc_pages_nodemask
0.72 ± 38% +67.0% 1.21 ± 26% perf-profile.children.cycles-pp.__alloc_pages_slowpath
1.08 ± 21% +61.1% 1.74 ± 16% perf-profile.children.cycles-pp.__blk_mq_complete_request
1.30 ± 9% +19.0% 1.55 ± 11% perf-profile.children.cycles-pp.__block_commit_write
1.25 ± 2% +32.3% 1.65 ± 6% perf-profile.children.cycles-pp.__block_write_begin_int
9.79 ± 8% +24.3% 12.16 ± 11% perf-profile.children.cycles-pp.__do_fault
46.18 ± 3% -33.1% 30.87 ± 7% perf-profile.children.cycles-pp.__do_page_fault
2.66 ± 21% +56.9% 4.17 ± 16% perf-profile.children.cycles-pp.__generic_file_write_iter
2.62 ± 4% -46.6% 1.40 ± 16% perf-profile.children.cycles-pp.__lock_page
1.66 ± 8% +65.3% 2.74 ± 8% perf-profile.children.cycles-pp.__lock_page_or_retry
1.29 ± 10% -22.0% 1.01 ± 9% perf-profile.children.cycles-pp.__might_sleep
1.18 ± 24% +49.8% 1.76 ± 22% perf-profile.children.cycles-pp.__page_cache_alloc
0.81 ± 3% +20.5% 0.97 ± 5% perf-profile.children.cycles-pp.__radix_tree_lookup
2.33 ± 6% -29.3% 1.64 ± 4% perf-profile.children.cycles-pp.__save_stack_trace
6.04 ± 3% -22.4% 4.69 ± 8% perf-profile.children.cycles-pp.__schedule
2.80 ± 1% -24.8% 2.11 ± 3% perf-profile.children.cycles-pp.__tick_nohz_idle_enter
1.53 ± 8% +54.3% 2.36 ± 7% perf-profile.children.cycles-pp.__wait_on_bit
2.57 ± 5% -38.7% 1.58 ± 16% perf-profile.children.cycles-pp.__wait_on_bit_lock
7.02 ± 3% -13.6% 6.06 ± 6% perf-profile.children.cycles-pp.__wake_up
7.23 ± 3% -13.4% 6.26 ± 5% perf-profile.children.cycles-pp.__wake_up_bit
6.16 ± 3% -9.3% 5.58 ± 6% perf-profile.children.cycles-pp.__wake_up_common
6.09 ± 15% +42.3% 8.67 ± 14% perf-profile.children.cycles-pp.__writeback_inodes_wb
6.09 ± 15% +42.3% 8.67 ± 14% perf-profile.children.cycles-pp.__writeback_single_inode
2.08 ± 3% -32.5% 1.40 ± 2% perf-profile.children.cycles-pp.__xfs_get_blocks
5.88 ± 4% -99.0% 0.06 ± 23% perf-profile.children.cycles-pp.__xfs_trans_commit
2.92 ± 6% -28.1% 2.10 ± 6% perf-profile.children.cycles-pp._raw_spin_lock
2.11 ± 2% -21.6% 1.66 ± 10% perf-profile.children.cycles-pp._raw_spin_lock_irqsave
5.86 ± 2% -27.9% 4.22 ± 6% perf-profile.children.cycles-pp.activate_task
1.27 ± 12% +35.2% 1.72 ± 15% perf-profile.children.cycles-pp.add_to_page_cache_lru
1.46 ± 22% +55.2% 2.27 ± 18% perf-profile.children.cycles-pp.alloc_pages_current
1.01 ± 22% +65.8% 1.67 ± 15% perf-profile.children.cycles-pp.bio_endio
1.08 ± 21% +60.7% 1.74 ± 16% perf-profile.children.cycles-pp.blk_mq_complete_request
1.08 ± 21% +60.9% 1.73 ± 16% perf-profile.children.cycles-pp.blk_mq_end_request
1.04 ± 22% +63.1% 1.70 ± 16% perf-profile.children.cycles-pp.blk_update_request
1.31 ± 8% +18.6% 1.56 ± 12% perf-profile.children.cycles-pp.block_commit_write
15.23 ± 4% -32.8% 10.23 ± 2% perf-profile.children.cycles-pp.call_cpuidle
1.04 ± 9% +63.4% 1.70 ± 24% perf-profile.children.cycles-pp.call_function_interrupt
4.10 ± 6% -93.8% 0.25 ± 9% perf-profile.children.cycles-pp.call_rwsem_down_write_failed
2.82 ± 4% -64.2% 1.01 ± 10% perf-profile.children.cycles-pp.call_rwsem_wake
4.38 ± 15% +44.2% 6.31 ± 14% perf-profile.children.cycles-pp.clear_page_dirty_for_io
25.57 ± 3% -29.1% 18.14 ± 2% perf-profile.children.cycles-pp.cpu_startup_entry
15.17 ± 4% -32.8% 10.19 ± 2% perf-profile.children.cycles-pp.cpuidle_enter
14.86 ± 4% -33.1% 9.95 ± 2% perf-profile.children.cycles-pp.cpuidle_enter_state
0.65 ± 6% +54.5% 1.00 ± 6% perf-profile.children.cycles-pp.create_empty_buffers
0.83 ± 4% +51.1% 1.25 ± 5% perf-profile.children.cycles-pp.create_page_buffers
2.12 ± 4% -23.9% 1.61 ± 7% perf-profile.children.cycles-pp.deactivate_task
5.97 ± 3% -9.3% 5.42 ± 6% perf-profile.children.cycles-pp.default_wake_function
1.62 ± 6% -24.1% 1.23 ± 5% perf-profile.children.cycles-pp.dequeue_entity
1.86 ± 4% -24.0% 1.41 ± 5% perf-profile.children.cycles-pp.dequeue_task_fair
46.50 ± 3% -32.9% 31.22 ± 8% perf-profile.children.cycles-pp.do_page_fault
26.60 ± 4% -58.3% 11.10 ± 11% perf-profile.children.cycles-pp.do_page_mkwrite
0.68 ± 39% +66.1% 1.13 ± 27% perf-profile.children.cycles-pp.do_try_to_free_pages
16.95 ± 59% +145.4% 41.59 ± 20% perf-profile.children.cycles-pp.do_unit
0.63 ± 2% +47.7% 0.93 ± 32% perf-profile.children.cycles-pp.do_unlinkat
13.93 ± 11% -54.5% 6.34 ± 32% perf-profile.children.cycles-pp.do_wp_page
6.50 ± 14% +39.2% 9.04 ± 12% perf-profile.children.cycles-pp.do_writepages
1.96 ± 3% -13.9% 1.69 ± 3% perf-profile.children.cycles-pp.down_read
4.68 ± 6% -88.8% 0.53 ± 13% perf-profile.children.cycles-pp.down_write
0.77 ± 24% +69.3% 1.30 ± 17% perf-profile.children.cycles-pp.end_buffer_async_write
0.53 ± 27% +84.9% 0.98 ± 15% perf-profile.children.cycles-pp.end_page_writeback
5.34 ± 2% -27.7% 3.86 ± 7% perf-profile.children.cycles-pp.enqueue_entity
5.53 ± 2% -27.2% 4.03 ± 7% perf-profile.children.cycles-pp.enqueue_task_fair
0.63 ± 2% +47.7% 0.93 ± 32% perf-profile.children.cycles-pp.evict
12.92 ± 5% -96.3% 0.48 ± 11% perf-profile.children.cycles-pp.file_update_time
9.21 ± 8% +25.0% 11.52 ± 11% perf-profile.children.cycles-pp.filemap_fault
0.81 ± 8% +48.1% 1.20 ± 21% perf-profile.children.cycles-pp.flush_smp_call_function_queue
2.74 ± 15% +48.5% 4.07 ± 14% perf-profile.children.cycles-pp.flush_tlb_page
2.76 ± 20% +56.6% 4.33 ± 16% perf-profile.children.cycles-pp.generic_file_write_iter
2.54 ± 21% +56.1% 3.96 ± 16% perf-profile.children.cycles-pp.generic_perform_write
0.96 ± 9% +47.9% 1.42 ± 22% perf-profile.children.cycles-pp.generic_smp_call_function_single_interrupt
1.06 ± 1% -22.8% 0.82 ± 9% perf-profile.children.cycles-pp.get_next_timer_interrupt
47.30 ± 3% -31.1% 32.58 ± 3% perf-profile.children.cycles-pp.handle_mm_fault
1.14 ± 8% -29.7% 0.80 ± 7% perf-profile.children.cycles-pp.hrtimer_start_range_ns
14.32 ± 4% -32.8% 9.62 ± 3% perf-profile.children.cycles-pp.intel_idle
8.94 ± 4% -14.6% 7.64 ± 9% perf-profile.children.cycles-pp.iomap_apply
12.58 ± 4% -22.2% 9.79 ± 10% perf-profile.children.cycles-pp.iomap_page_mkwrite
2.65 ± 6% +24.3% 3.29 ± 8% perf-profile.children.cycles-pp.iomap_page_mkwrite_actor
1.83 ± 22% +62.6% 2.98 ± 15% perf-profile.children.cycles-pp.iov_iter_copy_from_user_atomic
0.63 ± 2% +47.7% 0.93 ± 32% perf-profile.children.cycles-pp.iput
1.25 ± 5% -53.2% 0.59 ± 4% perf-profile.children.cycles-pp.kmem_cache_alloc
1.53 ± 8% -100.0% 0.00 ± -1% perf-profile.children.cycles-pp.kmem_zone_alloc
4.46 ± 4% +51.9% 6.78 ± 11% perf-profile.children.cycles-pp.kswapd
14.70 ± 5% +49.9% 22.03 ± 7% perf-profile.children.cycles-pp.kthread
4.10 ± 20% +58.8% 6.51 ± 16% perf-profile.children.cycles-pp.kthread_worker_fn
2.87 ± 20% +59.1% 4.57 ± 17% perf-profile.children.cycles-pp.lo_write_bvec
4.06 ± 20% +58.9% 6.45 ± 16% perf-profile.children.cycles-pp.loop_queue_work
1.00 ± 7% +27.3% 1.27 ± 9% perf-profile.children.cycles-pp.mark_buffer_dirty
1.75 ± 21% +64.1% 2.88 ± 14% perf-profile.children.cycles-pp.memcpy_erms
2.80 ± 15% +47.1% 4.12 ± 13% perf-profile.children.cycles-pp.native_flush_tlb_others
1.47 ± 3% +29.2% 1.90 ± 13% perf-profile.children.cycles-pp.native_irq_return_iret
2.63 ± 7% -34.2% 1.73 ± 10% perf-profile.children.cycles-pp.native_queued_spin_lock_slowpath
46.66 ± 3% -32.8% 31.34 ± 8% perf-profile.children.cycles-pp.page_fault
3.77 ± 16% +45.5% 5.48 ± 14% perf-profile.children.cycles-pp.page_mkclean
3.15 ± 15% +48.1% 4.67 ± 13% perf-profile.children.cycles-pp.page_mkclean_one
1.38 ± 12% +46.9% 2.03 ± 21% perf-profile.children.cycles-pp.page_referenced
1.37 ± 7% -25.8% 1.02 ± 9% perf-profile.children.cycles-pp.pick_next_task_fair
6.09 ± 15% +42.5% 8.68 ± 14% perf-profile.children.cycles-pp.process_one_work
2.81 ± 16% +47.3% 4.14 ± 14% perf-profile.children.cycles-pp.ptep_clear_flush
6.17 ± 13% -40.5% 3.67 ± 5% perf-profile.children.cycles-pp.rest_init
14.70 ± 5% +49.8% 22.03 ± 7% perf-profile.children.cycles-pp.ret_from_fork
5.59 ± 7% +45.7% 8.15 ± 3% perf-profile.children.cycles-pp.rmap_walk
5.29 ± 8% +46.4% 7.75 ± 4% perf-profile.children.cycles-pp.rmap_walk_file
4.01 ± 6% -93.9% 0.24 ± 6% perf-profile.children.cycles-pp.rwsem_down_write_failed
2.14 ± 10% -96.3% 0.08 ± 0% perf-profile.children.cycles-pp.rwsem_spin_on_owner
2.78 ± 4% -64.3% 0.99 ± 11% perf-profile.children.cycles-pp.rwsem_wake
2.75 ± 6% -29.9% 1.93 ± 3% perf-profile.children.cycles-pp.save_stack_trace_tsk
6.35 ± 3% -23.0% 4.89 ± 8% perf-profile.children.cycles-pp.schedule
2.63 ± 4% -26.1% 1.95 ± 10% perf-profile.children.cycles-pp.schedule_preempt_disabled
0.48 ± 11% +61.8% 0.78 ± 32% perf-profile.children.cycles-pp.shrink_active_list
4.55 ± 9% +53.0% 6.96 ± 13% perf-profile.children.cycles-pp.shrink_inactive_list
5.14 ± 8% +53.9% 7.91 ± 14% perf-profile.children.cycles-pp.shrink_node
5.10 ± 8% +53.7% 7.84 ± 13% perf-profile.children.cycles-pp.shrink_node_memcg
3.78 ± 8% +49.7% 5.65 ± 12% perf-profile.children.cycles-pp.shrink_page_list
0.97 ± 9% +61.2% 1.56 ± 22% perf-profile.children.cycles-pp.smp_call_function_interrupt
2.75 ± 15% +46.6% 4.03 ± 13% perf-profile.children.cycles-pp.smp_call_function_many
1.16 ± 8% +44.6% 1.67 ± 21% perf-profile.children.cycles-pp.smp_call_function_single_interrupt
6.17 ± 13% -40.5% 3.67 ± 5% perf-profile.children.cycles-pp.start_kernel
19.74 ± 8% -25.6% 14.69 ± 2% perf-profile.children.cycles-pp.start_secondary
0.63 ± 2% +47.7% 0.93 ± 32% perf-profile.children.cycles-pp.sys_unlinkat
2.93 ± 1% -25.3% 2.19 ± 3% perf-profile.children.cycles-pp.tick_nohz_idle_enter
1.55 ± 7% -29.0% 1.10 ± 8% perf-profile.children.cycles-pp.tick_nohz_idle_exit
1.12 ± 7% -32.4% 0.76 ± 3% perf-profile.children.cycles-pp.tick_nohz_restart_sched_tick
2.39 ± 2% -25.9% 1.77 ± 4% perf-profile.children.cycles-pp.tick_nohz_stop_sched_tick
0.63 ± 2% +47.7% 0.93 ± 32% perf-profile.children.cycles-pp.truncate_inode_pages_final
0.62 ± 2% +47.3% 0.92 ± 31% perf-profile.children.cycles-pp.truncate_inode_pages_range
0.68 ± 39% +66.1% 1.13 ± 27% perf-profile.children.cycles-pp.try_to_free_pages
0.69 ± 8% +48.9% 1.02 ± 10% perf-profile.children.cycles-pp.try_to_unmap
8.49 ± 1% -25.7% 6.30 ± 5% perf-profile.children.cycles-pp.try_to_wake_up
6.24 ± 2% -28.1% 4.48 ± 7% perf-profile.children.cycles-pp.ttwu_do_activate
0.63 ± 2% +47.7% 0.93 ± 32% perf-profile.children.cycles-pp.unlinkat
7.55 ± 2% -11.8% 6.66 ± 5% perf-profile.children.cycles-pp.unlock_page
1.52 ± 4% -21.7% 1.19 ± 6% perf-profile.children.cycles-pp.up_read
2.44 ± 7% -61.5% 0.94 ± 8% perf-profile.children.cycles-pp.up_write
2.81 ± 20% +56.0% 4.39 ± 16% perf-profile.children.cycles-pp.vfs_iter_write
1.58 ± 8% +53.3% 2.43 ± 7% perf-profile.children.cycles-pp.wait_on_page_bit_killable
6.05 ± 3% -9.4% 5.48 ± 6% perf-profile.children.cycles-pp.wake_bit_function
2.62 ± 3% -64.5% 0.93 ± 9% perf-profile.children.cycles-pp.wake_up_q
6.09 ± 15% +42.3% 8.67 ± 14% perf-profile.children.cycles-pp.wb_workfn
6.09 ± 15% +42.3% 8.67 ± 14% perf-profile.children.cycles-pp.wb_writeback
6.09 ± 15% +42.5% 8.69 ± 14% perf-profile.children.cycles-pp.worker_thread
6.49 ± 14% +39.2% 9.03 ± 12% perf-profile.children.cycles-pp.write_cache_pages
6.09 ± 15% +42.3% 8.67 ± 14% perf-profile.children.cycles-pp.writeback_sb_inodes
6.17 ± 13% -40.5% 3.67 ± 5% perf-profile.children.cycles-pp.x86_64_start_kernel
6.17 ± 13% -40.5% 3.67 ± 5% perf-profile.children.cycles-pp.x86_64_start_reservations
0.98 ± 22% +67.0% 1.64 ± 16% perf-profile.children.cycles-pp.xfs_destroy_ioend
2.10 ± 13% +34.8% 2.84 ± 10% perf-profile.children.cycles-pp.xfs_do_writepage
0.99 ± 22% +67.2% 1.65 ± 16% perf-profile.children.cycles-pp.xfs_end_bio
6.08 ± 4% -31.4% 4.17 ± 9% perf-profile.children.cycles-pp.xfs_file_iomap_begin
5.82 ± 4% -33.9% 3.85 ± 8% perf-profile.children.cycles-pp.xfs_file_iomap_begin_delay
9.71 ± 7% +23.8% 12.03 ± 10% perf-profile.children.cycles-pp.xfs_filemap_fault
26.49 ± 4% -58.6% 10.98 ± 11% perf-profile.children.cycles-pp.xfs_filemap_page_mkwrite
2.12 ± 4% -31.4% 1.45 ± 3% perf-profile.children.cycles-pp.xfs_get_blocks
6.06 ± 6% -74.5% 1.55 ± 6% perf-profile.children.cycles-pp.xfs_ilock
1.04 ± 2% -100.0% 0.00 ± -1% perf-profile.children.cycles-pp.xfs_inode_item_format
1.40 ± 4% -100.0% 0.00 ± -1% perf-profile.children.cycles-pp.xfs_inode_item_unlock
3.60 ± 4% -58.4% 1.50 ± 8% perf-profile.children.cycles-pp.xfs_iunlock
4.95 ± 4% -98.9% 0.05 ± 8% perf-profile.children.cycles-pp.xfs_log_commit_cil
1.60 ± 9% -100.0% 0.00 ± -1% perf-profile.children.cycles-pp.xfs_log_reserve
1.32 ± 13% +33.8% 1.77 ± 11% perf-profile.children.cycles-pp.xfs_start_page_writeback
2.83 ± 9% -98.6% 0.04 ± 73% perf-profile.children.cycles-pp.xfs_trans_alloc
6.01 ± 4% -98.9% 0.07 ± 25% perf-profile.children.cycles-pp.xfs_trans_commit
1.78 ± 4% -100.0% 0.00 ± -1% perf-profile.children.cycles-pp.xfs_trans_free_items
1.84 ± 8% -100.0% 0.00 ± -1% perf-profile.children.cycles-pp.xfs_trans_reserve
6.50 ± 14% +39.2% 9.04 ± 12% perf-profile.children.cycles-pp.xfs_vm_writepages
12.31 ± 5% -98.7% 0.16 ± 20% perf-profile.children.cycles-pp.xfs_vn_update_time
0.81 ± 3% +20.5% 0.97 ± 5% perf-profile.self.cycles-pp.__radix_tree_lookup
0.98 ± 6% -24.6% 0.74 ± 10% perf-profile.self.cycles-pp.__schedule
1.45 ± 6% -12.0% 1.28 ± 6% perf-profile.self.cycles-pp._raw_spin_lock
1.45 ± 15% +36.6% 1.98 ± 20% perf-profile.self.cycles-pp.do_mpage_readpage
4.59 ± 59% +295.3% 18.14 ± 20% perf-profile.self.cycles-pp.do_unit
14.32 ± 4% -32.8% 9.62 ± 3% perf-profile.self.cycles-pp.intel_idle
1.75 ± 21% +64.1% 2.88 ± 14% perf-profile.self.cycles-pp.memcpy_erms
1.47 ± 3% +29.2% 1.90 ± 13% perf-profile.self.cycles-pp.native_irq_return_iret
2.63 ± 7% -34.2% 1.73 ± 10% perf-profile.self.cycles-pp.native_queued_spin_lock_slowpath
2.14 ± 10% -96.3% 0.08 ± 0% perf-profile.self.cycles-pp.rwsem_spin_on_owner
1.55 ± 13% +65.5% 2.57 ± 12% perf-profile.self.cycles-pp.smp_call_function_many
0.82 ± 7% +20.7% 0.99 ± 5% perf-profile.self.cycles-pp.up_read
perf-stat.instructions
1e+13 ++-------------O-------O----O---O-O---------------------------------+
9e+12 O+O O O O O O O O O O O O O O O O O O |
*. .*.*..*.*.*.*.*. .*. .*.*. .*.*.*.*. .*.*.*.*.* *. .*. |
8e+12 ++* * *. * *. : : *..* *.*.*
7e+12 ++ : : |
| : : |
6e+12 ++ : : |
5e+12 ++ : : |
4e+12 ++ : : |
| : : |
3e+12 ++ : : |
2e+12 ++ :: |
| : |
1e+12 ++ : |
0 ++---------------------------------------------------*--------------+
perf-stat.cache-misses
3e+10 ++----------------------------------------------------------------+
O O O O O O O O O O O O O O O O O O O O O O O O |
2.5e+10 ++ |
| |
| *. |
2e+10 *+*.*.*.*.*.*.*..*. .*.*.*.*.*.*.*.*.*.*.*.*.*.*..* : *.*.*.*.*.*
| * : : |
1.5e+10 ++ : : |
| : : |
1e+10 ++ : : |
| : : |
| : : |
5e+09 ++ : |
| : |
0 ++--------------------------------------------------*-------------+
perf-stat.branch-instructions
2.5e+12 ++----------------------------------------------------------------+
O O O O O O O O O O O O O O O O O O O O O O O O |
| |
2e+12 *+*.*.*.*.*.*.*..*. .*.*.*.*.*.*.*.*.*.*.*.*.*.*..* *.*.*.*.*.*.*
| * : : |
| : : |
1.5e+12 ++ : : |
| : : |
1e+12 ++ : : |
| : : |
| : : |
5e+11 ++ : : |
| : |
| : |
0 ++--------------------------------------------------*-------------+
perf-stat.branch-misses
8e+09 ++---------------*--------------------------------------------------+
|.*. .*.*.* + .*.*..*. .*. .*.. .*.*.*.*.|
7e+09 *+ *.*..* * * *.*.*.* *.*.*.*.* *.*. *
6e+09 ++ : : |
O O O O O O O O O O O O O O O O O O O O O O O O : : |
5e+09 ++ : : |
| : : |
4e+09 ++ : : |
| : : |
3e+09 ++ : : |
2e+09 ++ : : |
| :: |
1e+09 ++ : |
| : |
0 ++---------------------------------------------------*--------------+
perf-stat.context-switches
3.5e+08 ++----------------------------------------------------------------+
| |
3e+08 ++*. .*. .* .*. .*. .*. |
* *.*.*.* *. + .*. .* *.*.*.* *.*.*.*..* *.* *.*.*.*
2.5e+08 ++ *.* * : : |
O O O O O O : : |
2e+08 ++O O O O O O O O O O O O O O O O O O : : |
| : : |
1.5e+08 ++ : : |
| : : |
1e+08 ++ : : |
| : : |
5e+07 ++ : |
| : |
0 ++--------------------------------------------------*-------------+
perf-stat.cache-miss-rate_
40 O+O-O----O-O-O--O-O-O-O-O----O-O-O--O-O-O-O-O--O-O-O-------------------+
| O O |
35 ++ |
30 ++ |
*. .*..*.*.*. .*. .*. .*..*.*.*. .*..*.*.*.* *. .*.*
25 ++* *. *.*.*.*..* * * : : *.*.*..* |
| : : |
20 ++ : : |
| : : |
15 ++ : : |
10 ++ : : |
| : : |
5 ++ :: |
| : |
0 ++-----------------------------------------------------*---------------+
perf-stat.branch-miss-rate_
0.45 ++-------------------------------------------------------------------+
| .*. |
0.4 ++*. .*.*.* *..*.*.*. .*. .*. .*..*.*.*.*
0.35 *+ *.*..* * *..*.*.* *.*.*..*.* *.* |
| : : |
0.3 ++ : : |
0.25 O+ O O O O O O O O : : |
| O O O O O O O O O O O O O O O : : |
0.2 ++ : : |
0.15 ++ : : |
| : : |
0.1 ++ : : |
0.05 ++ : |
| : |
0 ++----------------------------------------------------*--------------+
perf-stat.ipc
1.2 ++--------------------------------------------------------------------+
| |
1 ++ O O O O O O O O O O O O O O |
O O O O O O O O O O |
*.*.*..*.*.*.*.*..*.*.*.*.*.*..*.*.*.*.*..*.*.*.*.*..* *.*.*.*..*.*.*
0.8 ++ : : |
| : : |
0.6 ++ : : |
| : : |
0.4 ++ : : |
| : : |
| : : |
0.2 ++ : |
| : |
0 ++-----------------------------------------------------*--------------+
vm-scalability.time.user_time
1000 ++-------------------------------------------------------------------+
900 ++O O O O O O O O O O O O O |
O O O O O O O O O O O |
800 ++ |
700 ++ |
*. .*.*..*.*. .*. *. .*.*. .*..*.*.*. .*.*.*..*.* *.*. .*.*.*.*
600 ++* * * : * * * : : *. |
500 ++ : : : : |
400 ++ : : : : |
| : : : : |
300 ++ : : : : |
200 ++ : : : : |
| :: : |
100 ++ : : |
0 ++-----------------*----------------------------------*--------------+
vm-scalability.time.system_time
1800 ++-------------------------------------------------------------------+
|.*.*.*..*.*.*.*.* *.*.*. .*.*..*.*. .*.*.*.*..*.* *.*.*..*.*.*.*
1600 *+ : : * * : : |
1400 ++ : : : : |
O O O O : :O O O O : : |
1200 ++O O O O O:O : O O O O O O O O O O : : |
1000 ++ : : : : |
| : : : : |
800 ++ : : : : |
600 ++ : : : : |
| : : : : |
400 ++ :: : |
200 ++ : : |
| : : |
0 ++-----------------*----------------------------------*--------------+
vm-scalability.time.percent_of_cpu_this_job_got
300 ++--------------------------------------------------------------------+
| |
250 *+*.*..*.*.*.*.*..* *.*.*.*..*.*.*.*.*..*.*.*.*.*..* *.*.*.*..*.*.*
O O O O O O O O O O O O O O O O O O O O O O O O : : |
| : : : : |
200 ++ : : : : |
| : : : : |
150 ++ : : : : |
| : : : : |
100 ++ : : : : |
| : : : : |
| :: :: |
50 ++ : : |
| : : |
0 ++------------------*----------------------------------*--------------+
vm-scalability.time.minor_page_faults
3e+08 ++----------------------------------------------------------------+
| O O O O O O O O O O O |
2.5e+08 O+ O O O O O O O O O O O O |
*.*.*.*.*.*.*.*..* *.*.*.*.*.*.*.*.*.*.*.*.*.*..* *.*.*.*.*.*.*
| : : : : |
2e+08 ++ : : : : |
| : : : : |
1.5e+08 ++ : : : : |
| : : : : |
1e+08 ++ : : : : |
| : : : : |
| : : : : |
5e+07 ++ : : |
| : : |
0 ++-----------------*--------------------------------*-------------+
vm-scalability.time.voluntary_context_switches
1.6e+08 ++----------------------------------------------------------------+
| *. *. .* *. *. .*. |
1.4e+08 ++ *.*.*. + *. : .*. + *.*.*. + *.*.*. .* *.* *.*.*.|
1.2e+08 *+ * : * *.* * *. : : *
| : : : : |
1e+08 O+O O O O O O O O:O O O O O O O O O O O O O O : : |
| : : O : : |
8e+07 ++ : : : : |
| : : : : |
6e+07 ++ : : : : |
4e+07 ++ : : : : |
| :: :: |
2e+07 ++ : : |
| : : |
0 ++-----------------*--------------------------------*-------------+
vm-scalability.time.file_system_outputs
1.2e+09 ++----------------------------------------------------------------+
| |
1e+09 ++O O O O O O O O O O O O O O O O O O O O O O O |
O |
| |
8e+08 *+ .*.*.*. .*.. *. .*.*. .*.*.*.*. .*.*.*.*.. *. |
| * *.* * : * * * * : *.*.*.*.*.*
6e+08 ++ : : : : |
| : : : : |
4e+08 ++ : : : : |
| : : : : |
| : : : : |
2e+08 ++ : : |
| : : |
0 ++-----------------*--------------------------------*-------------+
vm-scalability.throughput
3e+06 ++----------------------------------------------------------------+
| O O O O O O O O O O O O O O O O O O O O O |
2.5e+06 O+ O O |
| |
| O |
2e+06 *+ .*.*.*.*. .*.. .*.*.*.*.*. .*.*.*.*. .*.*.*.*..* *.*. .*.*.*.*
| * * * * * : : * |
1.5e+06 ++ : : |
| : : |
1e+06 ++ : : |
| : : |
| : : |
500000 ++ : |
| : |
0 ++--------------------------------------------------*-------------+
interrupts.CAL:Function_call_interrupts
3e+08 ++----------------------------------------------------------------+
| |
2.5e+08 ++ O O |
| O O O O O O O O O O O O O O O O O O O O O |
O |
2e+08 ++ |
| |
1.5e+08 *+ .*.*.*.*. .*.. *. .*.*. .*.*.*.*. .*.*.*.*..* *.*. .*.*.*.*
| * * : * * * : : * |
1e+08 ++ * : : : |
| : : : : |
| : : : : |
5e+07 ++ : : : : |
| : : |
0 ++-----------------*------------------------------O-*-------------+
vmstat.io.bo
700000 ++-----------------------------------------------------------------+
| |
600000 ++O O O O O O O O O O O O O O |
O O O O O O O O O O |
500000 ++ |
| |
400000 *+*.*.*.*.*..*.*.*.*.*.*.*.*.*.*..*.*.*.*.*.*.*.*.O *..*.*.*.*.*.*
| : : |
300000 ++ : : |
| : : |
200000 ++ : : |
| : : |
100000 ++ :: |
| : |
0 ++--------------------------------------------------*--------------+
vmstat.system.in
350000 ++-----------------------------------------------------------------+
| |
300000 ++ O O O |
| O O O O O O O O O O O O O O O O O O O O |
250000 O+ |
| O |
200000 ++ |
*. .*.*.*.*.. .*. .*.*. .*.*. .*..*.*.*. .*.*.*.*.* *..*. .*.*.*.*
150000 ++* * * * * * : : * |
| : : |
100000 ++ : : |
| : : |
50000 ++ : : |
| : |
0 ++--------------------------------------------------*--------------+
vmstat.system.cs
350000 ++*--------------*-----------*----------*--------------------------+
|+ + .*. + + + + + + .* .*. |
300000 *+ *.*.*.*. * *.*.*.*.* *..*.*.* *.*.*.* : *..* *.*.*.*
| O : |
250000 O+ O O O O O O O : : |
| O O O O O O O O O O O O O O O O : : |
200000 ++ : : |
| : : |
150000 ++ : : |
| : : |
100000 ++ : : |
| :: |
50000 ++ : |
| : |
0 ++--------------------------------------------------*--------------+
[*] 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,
Xiaolong
3 years, 8 months
- 1
- 0
[lkp] [rcu] a7410f28ce: INFO: rcu_sched detected expedited stalls on CPUs/tasks: { 0-... } 25089 jiffies s: 9 root: 0x1/.
by kernel test robot
FYI, we noticed the following commit:
https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git master
commit a7410f28ceb566bda840b4afc278747c63383fb6 ("rcu: update: Make RCU_EXPEDITE_BOOT be the default")
in testcase: boot
on test machine: qemu-system-x86_64 -enable-kvm -cpu host -smp 2 -m 1G
caused below changes:
+-----------------------------------------------------------------------------------+------------+------------+
| | 9aa96e24ec | a7410f28ce |
+-----------------------------------------------------------------------------------+------------+------------+
| boot_successes | 3 | 7 |
| boot_failures | 17 | 17 |
| INFO:rcu_sched_self-detected_stall_on_CPU | 17 | 14 |
| calltrace:mark_rodata_ro | 17 | 14 |
| INFO:rcu_sched_detected_stalls_on_CPUs/tasks | 15 | 14 |
| INFO:rcu_sched_detected_expedited_stalls_on_CPUs/tasks:{#-...}#jiffies_s:#root:#/ | 0 | 15 |
+-----------------------------------------------------------------------------------+------------+------------+
[ 38.504100] Write protecting the kernel read-only data: 47104k
[ 38.511102] Freeing unused kernel memory: 720K (ffff88000314c000 - ffff880003200000)
[ 38.547480] Freeing unused kernel memory: 1032K (ffff880003cfe000 - ffff880003e00000)
[ 149.710631] INFO: rcu_sched detected expedited stalls on CPUs/tasks: { 0-... } 25089 jiffies s: 9 root: 0x1/.
[ 149.720353] blocking rcu_node structures:
[ 149.722359] Task dump for CPU 0:
[ 149.724094] swapper/0 R running task 0 1 0 0x00000008
[ 149.726939] ffff88002ccc98c0 ffff88002d8202c0 ffff88002cccc9c0 ffffffff83f30e40
[ 149.730972] ffff88002ccc9b30 ffff88002ccd7c80 ffffffff8313afd3 ffffffff83c43f92
[ 149.735011] ffffffff81e50101 1ffff1000599af77 0000000000000000 0000000041b58ab3
[ 149.739029] Call Trace:
[ 149.740574] [<ffffffff8313afd3>] ? __schedule+0xad3/0xb02
[ 149.742801] [<ffffffff81e50101>] ? hide_cursor+0x175/0x20d
[ 149.745015] [<ffffffff8313a500>] ? cb_alloc+0x264/0x264
To reproduce:
git clone git://git.kernel.org/pub/scm/linux/kernel/git/wfg/lkp-tests.git
cd lkp-tests
bin/lkp qemu -k <bzImage> job-script # job-script is attached in this email
Thanks,
Xiaolong
3 years, 8 months
- 3
- 3
[lkp] [mfd] fe517a5ae2: kernel_BUG_at_mm/slab.c
by kernel test robot
FYI, we noticed the following commit:
commit fe517a5ae2b2b9da7fe19abb41618a9189f80668 ("mfd: twl-core: make driver DT only")
url: https://github.com/0day-ci/linux/commits/Nicolae-Rosia/mfd-twl-core-make-...
base: https://git.kernel.org/pub/scm/linux/kernel/git/lee/mfd.git for-mfd-next
in testcase: boot
on test machine: qemu-system-x86_64 -enable-kvm -m 320M
caused below changes:
+----------------------------------------------+------------+------------+
| | 77f298a18f | fe517a5ae2 |
+----------------------------------------------+------------+------------+
| boot_successes | 0 | 0 |
| boot_failures | 12 | 12 |
| calltrace:async_run_entry_fn | 12 | 12 |
| WARNING:at_include/linux/kref.h:#kobject_get | 12 | 12 |
| calltrace:SyS_open | 12 | 4 |
| calltrace:compat_SyS_open | 8 | |
| WARNING:at_fs/sysfs/dir.c:#sysfs_warn_dup | 1 | |
| calltrace:scsi_debug_init | 1 | 8 |
| kernel_BUG_at_mm/slab.c | 0 | 8 |
| invalid_opcode:#[##]PREEMPT_SMP | 0 | 8 |
| RIP:free_block | 0 | 8 |
| Kernel_panic-not_syncing:Fatal_exception | 0 | 8 |
+----------------------------------------------+------------+------------+
[ 2.448641] sd 0:0:0:0: [sda] Attached SCSI disk
[ 2.450150] sd 0:0:0:0: [sda] Synchronizing SCSI cache
[ 2.459747] slab: double free detected in cache 'kmalloc-32', objp ffff8800101b5560
[ 2.461399] ------------[ cut here ]------------
[ 2.461953] kernel BUG at mm/slab.c:2637!
[ 2.461953] invalid opcode: 0000 [#1] PREEMPT SMP
[ 2.461953] CPU: 0 PID: 1 Comm: swapper/0 Tainted: G W 4.9.0-rc1-00025-gfe517a5 #1
[ 2.461953] task: ffff8800001da040 task.stack: ffffc900000d0000
[ 2.461953] RIP: 0010:[<ffffffffb4e852bc>] [<ffffffffb4e852bc>] free_block+0x2bc/0x2c0
[ 2.461953] RSP: 0000:ffffc900000d3a70 EFLAGS: 00010082
[ 2.461953] RAX: 0000000000000047 RBX: ffff880013a1a448 RCX: 0000000000000000
[ 2.461953] RDX: 0000000000000047 RSI: 0000000000000000 RDI: ffffffffb63f8678
[ 2.461953] RBP: ffffc900000d3ad0 R08: 0000000000000001 R09: 0000000000000000
[ 2.461953] R10: 0000000000000023 R11: 0000000000000001 R12: ffff8800101b5560
[ 2.461953] R13: ffff880000098100 R14: ffff880013336f98 R15: ffff880000090180
[ 2.461953] FS: 0000000000000000(0000) GS:ffff880013a00000(0000) knlGS:0000000000000000
[ 2.461953] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 2.461953] CR2: 0000000000000000 CR3: 0000000006214000 CR4: 00000000000006b0
[ 2.461953] Stack:
[ 2.461953] ffffc900000d3b18 ffff8800000901d8 ffff8800000901b8 00000000b4d70f3d
[ 2.461953] ffff880013a1a460 ffff8800000901b8 ffff880013336f98 ffff880000098100
[ 2.461953] 0000000000000010 ffff8800101b5160 0000000000000000 0000000000000000
[ 2.461953] Call Trace:
[ 2.461953] [<ffffffffb4e8b155>] ___cache_free+0x715/0x8a0
[ 2.461953] [<ffffffffb4e8baab>] kfree+0x2ab/0x300
[ 2.461953] [<ffffffffb5368517>] blk_mq_free_tag_set+0x57/0x80
[ 2.461953] [<ffffffffb55a0a85>] scsi_mq_destroy_tags+0x15/0x20
[ 2.461953] [<ffffffffb5596fae>] scsi_host_dev_release+0xbe/0x140
[ 2.461953] [<ffffffffb55205d2>] device_release+0x32/0x90
[ 2.461953] [<ffffffffb53814f2>] kobject_put+0x52/0xa0
[ 2.461953] [<ffffffffb5521a07>] put_device+0x17/0x20
[ 2.461953] [<ffffffffb5597cd5>] scsi_host_put+0x15/0x20
[ 2.461953] [<ffffffffb561322b>] sdebug_driver_remove+0x7b/0xa0
[ 2.461953] [<ffffffffb55256c6>] driver_probe_device+0xe6/0x390
[ 2.461953] [<ffffffffb5525a95>] __device_attach_driver+0x75/0xa0
[ 2.461953] [<ffffffffb5525a20>] ? __driver_attach+0xb0/0xb0
[ 2.461953] [<ffffffffb5523d2d>] bus_for_each_drv+0x5d/0x90
[ 2.461953] [<ffffffffb55254bd>] __device_attach+0xbd/0x110
[ 2.461953] [<ffffffffb5525b13>] device_initial_probe+0x13/0x20
[ 2.461953] [<ffffffffb5523fc3>] bus_probe_device+0xa3/0xb0
[ 2.461953] [<ffffffffb5521643>] device_add+0x443/0x680
[ 2.461953] [<ffffffffb55219ca>] device_register+0x1a/0x20
[ 2.461953] [<ffffffffb5612029>] sdebug_add_adapter+0xf9/0x1c0
[ 2.461953] [<ffffffffb6569809>] scsi_debug_init+0x62c/0x7b1
[ 2.461953] [<ffffffffb6569120>] ? init_sg+0x145/0x145
[ 2.461953] [<ffffffffb65691dd>] ? osd_uld_init+0xbd/0xbd
[ 2.461953] [<ffffffffb6507236>] do_one_initcall+0x99/0x16c
[ 2.461953] [<ffffffffb4d36600>] ? parse_args+0x210/0x3c0
[ 2.461953] [<ffffffffb65074a9>] kernel_init_freeable+0x1a0/0x22a
[ 2.461953] [<ffffffffb5b9dd20>] ? rest_init+0xd0/0xd0
[ 2.461953] [<ffffffffb5b9dd2e>] kernel_init+0xe/0x100
[ 2.461953] [<ffffffffb5bab4f5>] ret_from_fork+0x25/0x30
[ 2.461953] Code: ca 48 83 e0 fc 48 01 d0 4c 8b 48 20 4d 89 ce 41 83 e6 01 e9 15 fe ff ff 49 8b 75 58 4c 89 e2 48 c7 c7 a0 2c 03 b6 e8 4c d4 f8 ff <0f> 0b 66 90 0f 1f 44 00 00 8b 06 85 c0 75 01 c3 55 48 63 c2 48
[ 2.461953] RIP [<ffffffffb4e852bc>] free_block+0x2bc/0x2c0
[ 2.461953] RSP <ffffc900000d3a70>
[ 2.461953] ---[ end trace e31ed8a903d94b52 ]---
[ 2.461953] Kernel panic - not syncing: Fatal exception
[ 2.461953] Kernel Offset: 0x33c00000 from 0xffffffff81000000 (relocation range: 0xffffffff80000000-0xffffffffbfffffff)
Elapsed time: 10
To reproduce:
git clone git://git.kernel.org/pub/scm/linux/kernel/git/wfg/lkp-tests.git
cd lkp-tests
bin/lkp qemu -k <bzImage> job-script # job-script is attached in this email
Thanks,
Xiaolong
3 years, 8 months
- 1
- 0
[lkp] [mm] 190149375b: [No primary change] phoronix-test-suite.time.file_system_inputs -28.4% decrease
by kernel test robot
Greeting,
There is no primary kpi change in this test, below is the data collected through multiple monitors running background just for your information.
commit: 190149375bd87bc9578c7f8b701337a032d29acc ("mm: ovl: copy-up on MAP_SHARED")
https://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/vfs.git overlayfs-constino
in testcase: phoronix-test-suite
on test machine: 4 threads Intel(R) Core(TM) i5-2500K CPU @ 3.30GHz with 6G memory
with following parameters:
need_x: true
test: unigine-tropics-1.6.2
cpufreq_governor: performance
test-description: The Phoronix Test Suite is the most comprehensive testing and benchmarking platform available that provides an extensible framework for which new tests can be easily added.
test-url: http://www.phoronix-test-suite.com/
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
=========================================================================================
compiler/cpufreq_governor/kconfig/need_x/rootfs/tbox_group/test/testcase:
gcc-6/performance/x86_64-rhel-7.2/true/debian-full-x86_64/snb-drag/unigine-tropics-1.6.2/phoronix-test-suite
commit:
6dda996439 ("vfs: export filp_clone_open()")
190149375b ("mm: ovl: copy-up on MAP_SHARED")
6dda9964397d35dd 190149375bd87bc9578c7f8b70
---------------- --------------------------
fail:runs %reproduction fail:runs
| | |
%stddev %change %stddev
\ | \
57802 ± 0% -28.4% 41384 ± 0% phoronix-test-suite.time.file_system_inputs
0.98 ± 2% +4.0% 1.02 ± 1% perf-stat.branch-miss-rate%
0.13 ± 2% -7.4% 0.12 ± 6% perf-stat.dTLB-load-miss-rate%
153.00 ± 1% -68.0% 49.00 ± 0% perf-stat.major-faults
717.25 ± 1% -18.6% 584.00 ± 3% vmstat.io.bi
573561 ± 0% +112.9% 1221203 ± 0% vmstat.memory.cache
5206446 ± 0% -12.5% 4556087 ± 0% vmstat.memory.free
12611 ± 0% -100.0% 0.00 ± -1% latency_stats.max.wait_on_page_bit_killable.__lock_page_or_retry.filemap_fault.__do_fault.handle_mm_fault.__do_page_fault.do_page_fault.page_fault.generic_perform_write.__generic_file_write_iter.generic_file_write_iter.__vfs_write
0.00 ± -1% +Inf% 12605 ± 0% latency_stats.max.wait_on_page_bit_killable.generic_file_read_iter.generic_file_splice_read.do_splice_to.splice_direct_to_actor.do_splice_direct.ovl_copy_up_one.[overlay].ovl_copy_up.[overlay].ovl_open_maybe_copy_up.[overlay].ovl_d_real.[overlay].vm_mmap_pgoff.SyS_mmap_pgoff
13227 ± 0% -100.0% 0.00 ± -1% latency_stats.sum.wait_on_page_bit_killable.__lock_page_or_retry.filemap_fault.__do_fault.handle_mm_fault.__do_page_fault.do_page_fault.page_fault.generic_perform_write.__generic_file_write_iter.generic_file_write_iter.__vfs_write
0.00 ± -1% +Inf% 53556 ± 4% latency_stats.sum.wait_on_page_bit_killable.generic_file_read_iter.generic_file_splice_read.do_splice_to.splice_direct_to_actor.do_splice_direct.ovl_copy_up_one.[overlay].ovl_copy_up.[overlay].ovl_open_maybe_copy_up.[overlay].ovl_d_real.[overlay].vm_mmap_pgoff.SyS_mmap_pgoff
2537 ± 1% +15.8% 2938 ± 5% slabinfo.kmalloc-96.active_objs
2537 ± 1% +19.9% 3041 ± 7% slabinfo.kmalloc-96.num_objs
6072 ± 0% +54.9% 9404 ± 0% slabinfo.radix_tree_node.active_objs
216.25 ± 0% +55.1% 335.50 ± 0% slabinfo.radix_tree_node.active_slabs
6072 ± 0% +54.9% 9404 ± 0% slabinfo.radix_tree_node.num_objs
216.25 ± 0% +55.1% 335.50 ± 0% slabinfo.radix_tree_node.num_slabs
6090 ± 13% -20.8% 4822 ± 5% sched_debug.cfs_rq:/.min_vruntime.min
467.97 ±119% -245.4% -680.29 ±-47% sched_debug.cfs_rq:/.spread0.avg
1584 ± 49% -62.9% 587.38 ± 76% sched_debug.cfs_rq:/.spread0.max
-376.35 ±-145% +453.5% -2083 ±-51% sched_debug.cfs_rq:/.spread0.min
690982 ± 33% +37.4% 949101 ± 5% sched_debug.cpu.avg_idle.max
7.02 ±113% +387.3% 34.19 ± 77% sched_debug.cpu.cpu_load[1].stddev
4.89 ± 95% +493.2% 28.99 ± 76% sched_debug.cpu.cpu_load[2].stddev
3.75 ± 53% +426.6% 19.73 ± 80% sched_debug.cpu.cpu_load[3].stddev
173774 ± 0% +73.1% 300806 ± 3% meminfo.Active
130009 ± 0% +58.0% 205413 ± 2% meminfo.Active(anon)
43764 ± 0% +118.0% 95392 ± 5% meminfo.Active(file)
573373 ± 0% +113.0% 1221012 ± 0% meminfo.Cached
326943 ± 0% +158.6% 845442 ± 0% meminfo.Committed_AS
526737 ± 0% +98.8% 1047372 ± 0% meminfo.Inactive
118349 ± 0% +373.9% 560807 ± 0% meminfo.Inactive(anon)
408387 ± 0% +19.1% 486565 ± 1% meminfo.Inactive(file)
31326 ± 0% +13.7% 35611 ± 0% meminfo.Mapped
5461194 ± 0% -9.5% 4941511 ± 0% meminfo.MemAvailable
5206538 ± 0% -12.5% 4556061 ± 0% meminfo.MemFree
123330 ± 0% +419.9% 641163 ± 0% meminfo.Shmem
32552 ± 0% +57.9% 51412 ± 2% proc-vmstat.nr_active_anon
10960 ± 0% +117.6% 23851 ± 5% proc-vmstat.nr_active_file
138077 ± 0% -9.4% 125078 ± 0% proc-vmstat.nr_dirty_background_threshold
276493 ± 0% -9.4% 250463 ± 0% proc-vmstat.nr_dirty_threshold
143895 ± 0% +112.5% 305791 ± 0% proc-vmstat.nr_file_pages
1301586 ± 0% -12.5% 1138980 ± 0% proc-vmstat.nr_free_pages
29586 ± 0% +373.9% 140200 ± 0% proc-vmstat.nr_inactive_anon
102100 ± 0% +19.1% 121647 ± 1% proc-vmstat.nr_inactive_file
8081 ± 0% +13.3% 9153 ± 0% proc-vmstat.nr_mapped
30833 ± 0% +419.9% 160291 ± 0% proc-vmstat.nr_shmem
32552 ± 0% +57.9% 51412 ± 2% proc-vmstat.nr_zone_active_anon
10960 ± 0% +117.6% 23851 ± 5% proc-vmstat.nr_zone_active_file
29586 ± 0% +373.9% 140200 ± 0% proc-vmstat.nr_zone_inactive_anon
102100 ± 0% +19.1% 121647 ± 1% proc-vmstat.nr_zone_inactive_file
4907 ± 1% +22.7% 6021 ± 2% proc-vmstat.pgactivate
40107 ± 0% -20.8% 31745 ± 0% proc-vmstat.pgpgin
perf-stat.major-faults
160 *+--------*----*-------*-*--*--------------*------*------------*------+
| *.*..* * *.*. *..*.*..*.*. *..* *.*..*.*. *..*.*
140 ++ |
| |
| |
120 ++ |
| |
100 ++ |
| |
80 ++ |
| |
| |
60 ++ O O O |
O O O O O O O O O O O O O O O O O O O O O O O O |
40 ++--------------------------------------------------------------------+
phoronix-test-suite.time.file_system_inputs
58000 *+*--*-*--*-*--*-*-*--*-*--*-*--*-*-*--*-*--*-*--*-*-*--*-*--*-*--*-*
| |
56000 ++ |
54000 ++ |
| |
52000 ++ |
50000 ++ |
| |
48000 ++ |
46000 ++ |
| |
44000 ++ |
42000 ++ |
O O O O O O O O O O O O O O O O O O O O O O O O O O O |
40000 ++------------------------------------------------------------------+
[*] 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,
Xiaolong
3 years, 8 months
- 1
- 0
f8bfff5e6d: WARNING: CPU: 0 PID: 1 at drivers/base/power/common.c:150 dev_pm_domain_set
by kernel test robot
Greetings,
0day kernel testing robot got the below dmesg and the first bad commit is
https://git.linaro.org/people/vireshk/linux opp/genpd-performance-state
commit f8bfff5e6d937251e39602cfd7fb5cd3478934b1
Author: Viresh Kumar <viresh.kumar(a)linaro.org>
AuthorDate: Wed Nov 23 11:01:35 2016 +0530
Commit: Viresh Kumar <viresh.kumar(a)linaro.org>
CommitDate: Fri Nov 25 15:09:07 2016 +0530
test-code
Signed-off-by: Viresh Kumar <viresh.kumar(a)linaro.org>
+-----------------------------------------------------------+------------+------------+------------+
| | 482d49f33a | f8bfff5e6d | 1cef1f2165 |
+-----------------------------------------------------------+------------+------------+------------+
| boot_successes | 85 | 0 | 0 |
| boot_failures | 3 | 22 | 13 |
| BUG:kernel_reboot-without-warning_in_test_stage | 3 | | |
| WARNING:at_drivers/base/power/common.c:#dev_pm_domain_set | 0 | 22 | 13 |
| calltrace:genpd_test_init | 0 | 22 | |
| Oops | 0 | 0 | 10 |
| Kernel_panic-not_syncing:Fatal_exception | 0 | 0 | 9 |
+-----------------------------------------------------------+------------+------------+------------+
[ 3.870779] watchdog: Software Watchdog: a legacy watchdog module is probably present.
[ 3.870857] softdog: initialized. soft_noboot=0 soft_margin=60 sec soft_panic=0 (nowayout=0)
[ 3.870871] ------------[ cut here ]------------
[ 3.870876] WARNING: CPU: 0 PID: 1 at drivers/base/power/common.c:150 dev_pm_domain_set+0x51/0x60
[ 3.870877] PM domains can only be changed for unbound devices
[ 3.870878] Modules linked in:
[ 3.870880] CPU: 0 PID: 1 Comm: swapper/0 Not tainted 4.9.0-rc6-00134-gf8bfff5 #1
[ 3.870884] ffff8800001afd70 ffffffff81348c02 ffff8800001afdc0 0000000000000000
[ 3.870886] ffff8800001afdb0 ffffffff81063c4b 00000096810a7510 ffff88001fa0bea8
[ 3.870887] ffffffff81b95120 0000000000000000 ffffffff81daa707 5ad7850f274f2758
[ 3.870888] Call Trace:
[ 3.870892] [<ffffffff81348c02>] dump_stack+0x63/0x81
[ 3.870894] [<ffffffff81063c4b>] __warn+0xbb/0xe0
[ 3.870898] [<ffffffff81daa707>] ? do_early_param+0x8f/0x8f
[ 3.870900] [<ffffffff81063cba>] warn_slowpath_fmt+0x4a/0x50
[ 3.870901] [<ffffffff81daa707>] ? do_early_param+0x8f/0x8f
[ 3.870903] [<ffffffff8141cb71>] dev_pm_domain_set+0x51/0x60
[ 3.870906] [<ffffffff813a4367>] acpi_dev_pm_attach+0x6c/0xa2
[ 3.870908] [<ffffffff8141ca0d>] dev_pm_domain_attach+0xd/0x20
[ 3.870911] [<ffffffff81ddae68>] genpd_test_init+0x83/0x8b
[ 3.870913] [<ffffffff81ddade5>] ? cpufreq_gov_dbs_init+0x12/0x12
[ 3.870916] [<ffffffff81000457>] do_one_initcall+0x87/0x140
[ 3.870918] [<ffffffff81daa707>] ? do_early_param+0x8f/0x8f
[ 3.870919] [<ffffffff81dab07e>] kernel_init_freeable+0x10d/0x1a1
[ 3.870922] [<ffffffff81526200>] ? rest_init+0xd0/0xd0
[ 3.870923] [<ffffffff81526209>] kernel_init+0x9/0xf0
[ 3.870925] [<ffffffff8152dd25>] ret_from_fork+0x25/0x30
[ 3.870929] ---[ end trace 8b8714336b5a3086 ]---
[ 3.871370] fake-fmc-carrier: mezzanine 0
git bisect start 1cef1f2165ac5a0180d97805595b2eaa49ed355c 9c763584b7c8911106bb77af7e648bef09af9d80 --
git bisect bad a466b00806baf47f713f49c40d6211f5bf569d5f # 20:30 0- 10 Merge 'linux-review/Arnd-Bergmann/mvpp2-use-correct-size-for-memset/20161125-092232' into devel-spot-201611251856
git bisect bad f081b9f87c030250e90b375660d4f9871b9ab8fd # 20:49 0- 7 Merge 'linux-review/mengdong-lin-linux-intel-com/ASoC-topology-Allow-a-widget-to-have-multiple-enum-controls/20161125-164058' into devel-spot-201611251856
git bisect bad 3bf91d05d1e5a5f5e304790785531c71b9b7cba7 # 20:59 0- 1 Merge 'linux-review/Nabendu-Maiti/drm-i915-Move-number-of-scalers-initialization-to-runtime-init/20161125-173725' into devel-spot-201611251856
git bisect bad 425df59ec2cc35e0d8918695284853381eb779b4 # 21:08 0- 10 Merge 'linuxtv-media/master' into devel-spot-201611251856
git bisect good 4264d429e75198947228a493c80a2874723f16b4 # 21:22 21+ 0 Merge 'kdave-btrfs-devel/misc-next' into devel-spot-201611251856
git bisect bad 84175361060b494f752fb98a7780614e688a9050 # 21:31 0- 11 Merge 'vireshk/opp/genpd-performance-state' into devel-spot-201611251856
git bisect good eb4c7db93df01a59598be6c3cca6ae2b7ee0088b # 21:58 22+ 3 Merge branch 'pm-cpufreq' into linux-next
git bisect good 7607e6f8ddd3fb39929d3ee1b78ecf4bd28cef80 # 22:08 22+ 1 Merge branch 'pm-cpuidle' into linux-next
git bisect good 3d56e03162ca0f97a02e8f104bcebadf9601b128 # 22:22 22+ 1 PM / OPP: Add infrastructure to manage multiple regulators
git bisect good c23ec997d1101fc6bebc58bd238d2f7ff4dca421 # 22:43 20+ 2 PM / QOS: Add default case to the switch
git bisect good 4e5c95c926e147659642c9fc5f6b2339e7304318 # 22:57 22+ 0 PM / domain: Add set_performance_state() callback
git bisect good 482d49f33a174ae1e05a813e1290b28f92efa7d4 # 23:11 22+ 1 exynos OPP-v2 support
git bisect bad f8bfff5e6d937251e39602cfd7fb5cd3478934b1 # 23:21 0- 3 test-code
# first bad commit: [f8bfff5e6d937251e39602cfd7fb5cd3478934b1] test-code
git bisect good 482d49f33a174ae1e05a813e1290b28f92efa7d4 # 23:27 60+ 3 exynos OPP-v2 support
# extra tests with CONFIG_DEBUG_INFO_REDUCED
git bisect bad f8bfff5e6d937251e39602cfd7fb5cd3478934b1 # 23:43 0- 1 test-code
# extra tests on HEAD of linux-devel/devel-spot-201611251856
git bisect bad 1cef1f2165ac5a0180d97805595b2eaa49ed355c # 23:43 0- 13 0day head guard for 'devel-spot-201611251856'
# extra tests on tree/branch vireshk/opp/genpd-performance-state
git bisect bad f8bfff5e6d937251e39602cfd7fb5cd3478934b1 # 23:47 0- 22 test-code
# extra tests with first bad commit reverted
git bisect good c3b87e88fcc3e4026e0ef2949c1c55f63c81e46c # 01:25 63+ 1 Revert "test-code"
# extra tests on tree/branch linus/master
git bisect good 16ae16c6e5616c084168740990fc508bda6655d4 # 01:36 65+ 2 Merge tag 'mmc-v4.9-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/mmc
# extra tests on tree/branch linux-next/master
git bisect good 09bfc750ec237029d69ebb749602b6f1304d028a # 01:55 61+ 2 Add linux-next specific files for 20161125
---
0-DAY kernel test infrastructure Open Source Technology Center
https://lists.01.org/pipermail/lkp Intel Corporation
3 years, 8 months
- 3
- 3
[lkp] [mremap] 5d1904204c: will-it-scale.per_thread_ops -13.1% regression
by kernel test robot
Greeting,
FYI, we noticed a -13.1% regression of will-it-scale.per_thread_ops due to commit:
commit 5d1904204c99596b50a700f092fe49d78edba400 ("mremap: fix race between mremap() and page cleanning")
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
in testcase: will-it-scale
on test machine: 12 threads Intel(R) Core(TM) i7 CPU X 980 @ 3.33GHz with 6G memory
with following parameters:
test: eventfd1
cpufreq_governor: performance
test-description: Will It Scale takes a testcase and runs it from 1 through to n parallel copies to see if the testcase will scale. It builds both a process and threads based test in order to see any differences between the two.
test-url: https://github.com/antonblanchard/will-it-scale
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
=========================================================================================
compiler/cpufreq_governor/kconfig/rootfs/tbox_group/test/testcase:
gcc-6/performance/x86_64-rhel-7.2/debian-x86_64-2016-08-31.cgz/wsm/eventfd1/will-it-scale
commit:
961b708e95 (" fixes for amdgpu, and a bunch of arm drivers.")
5d1904204c ("mremap: fix race between mremap() and page cleanning")
961b708e95181041 5d1904204c99596b50a700f092
---------------- --------------------------
fail:runs %reproduction fail:runs
| | |
%stddev %change %stddev
\ | \
2459656 ± 0% -13.1% 2137017 ± 1% will-it-scale.per_thread_ops
2865527 ± 3% +4.2% 2986100 ± 0% will-it-scale.per_process_ops
0.62 ± 11% -13.2% 0.54 ± 1% will-it-scale.scalability
893.40 ± 0% +1.3% 905.24 ± 0% will-it-scale.time.system_time
169.92 ± 0% -7.0% 158.09 ± 0% will-it-scale.time.user_time
176943 ± 6% +26.1% 223131 ± 11% cpuidle.C1E-NHM.time
10.00 ± 6% -10.9% 8.91 ± 4% turbostat.CPU%c6
30508 ± 1% +3.4% 31541 ± 0% vmstat.system.cs
27239 ± 0% +1.5% 27650 ± 0% vmstat.system.in
2.03 ± 2% -11.6% 1.80 ± 6% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64
4.11 ± 1% -12.0% 3.61 ± 4% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_swapgs
1.70 ± 3% -13.8% 1.46 ± 5% perf-profile.children.cycles-pp.__fget_light
2.03 ± 2% -11.6% 1.80 ± 6% perf-profile.children.cycles-pp.entry_SYSCALL_64
4.11 ± 1% -12.0% 3.61 ± 4% perf-profile.children.cycles-pp.entry_SYSCALL_64_after_swapgs
12.79 ± 1% -10.0% 11.50 ± 6% perf-profile.children.cycles-pp.selinux_file_permission
1.70 ± 3% -13.8% 1.46 ± 5% perf-profile.self.cycles-pp.__fget_light
2.03 ± 2% -11.6% 1.80 ± 6% perf-profile.self.cycles-pp.entry_SYSCALL_64
4.11 ± 1% -12.0% 3.61 ± 4% perf-profile.self.cycles-pp.entry_SYSCALL_64_after_swapgs
5.85 ± 2% -12.5% 5.12 ± 5% perf-profile.self.cycles-pp.selinux_file_permission
1.472e+12 ± 0% -5.5% 1.392e+12 ± 0% perf-stat.branch-instructions
0.89 ± 0% -6.0% 0.83 ± 0% perf-stat.branch-miss-rate%
1.303e+10 ± 0% -11.1% 1.158e+10 ± 0% perf-stat.branch-misses
5.534e+08 ± 4% -6.9% 5.151e+08 ± 1% perf-stat.cache-references
9347877 ± 1% +3.4% 9663609 ± 0% perf-stat.context-switches
2.298e+12 ± 0% -5.6% 2.168e+12 ± 0% perf-stat.dTLB-loads
1.525e+12 ± 1% -5.4% 1.442e+12 ± 0% perf-stat.dTLB-stores
7.795e+12 ± 0% -5.5% 7.363e+12 ± 0% perf-stat.iTLB-loads
6.694e+12 ± 1% -4.5% 6.391e+12 ± 2% perf-stat.instructions
0.93 ± 0% -5.5% 0.88 ± 0% perf-stat.ipc
119024 ± 5% -11.3% 105523 ± 8% sched_debug.cfs_rq:/.exec_clock.max
5933459 ± 19% +24.5% 7385120 ± 3% sched_debug.cpu.nr_switches.max
1684848 ± 15% +20.6% 2032107 ± 3% sched_debug.cpu.nr_switches.stddev
5929704 ± 19% +24.5% 7382036 ± 3% sched_debug.cpu.sched_count.max
1684318 ± 15% +20.6% 2031701 ± 3% sched_debug.cpu.sched_count.stddev
2826278 ± 18% +30.4% 3684493 ± 3% sched_debug.cpu.sched_goidle.max
804195 ± 14% +26.2% 1014783 ± 3% sched_debug.cpu.sched_goidle.stddev
2969365 ± 19% +24.3% 3692180 ± 3% sched_debug.cpu.ttwu_count.max
843614 ± 15% +20.5% 1016263 ± 3% sched_debug.cpu.ttwu_count.stddev
2963657 ± 19% +24.4% 3687897 ± 3% sched_debug.cpu.ttwu_local.max
843104 ± 15% +20.5% 1016333 ± 3% sched_debug.cpu.ttwu_local.stddev
will-it-scale.time.user_time
172 ++--------------------*--------*---*----------------------------------+
170 ++..*....*...*....*. *.. . ..*.... ..*...*.... |
*. *....*. *. *...*
168 ++ |
166 ++ |
| |
164 ++ |
162 ++ |
160 ++ |
| O O |
158 ++ O O O |
156 ++ O |
O O |
154 ++ O O |
152 ++--O-----------------------------------------------------------------+
will-it-scale.time.system_time
912 ++--------------------------------------------------------------------+
910 ++ O O |
O O O |
908 ++ O |
906 ++ O O O |
904 ++ O O |
902 ++ |
| |
900 ++ |
898 ++ |
896 ++ |
894 ++ ..*
*...*.... ..*.... ..*....*...*....*...*...*....*. |
892 ++ *. *...*...*....*...*. |
890 ++--------------------------------------------------------------------+
will-it-scale.per_thread_ops
2.55e+06 ++---------------------------------------------------------------+
2.5e+06 ++ .*...*.. .*...*... ..*. .*.. |
|.. . .. *...*....*. .. .*... .. . |
2.45e+06 *+ * .. * *...*...*
2.4e+06 ++ * |
2.35e+06 ++ |
2.3e+06 ++ |
| |
2.25e+06 ++ |
2.2e+06 ++ O |
2.15e+06 ++ O O O O |
2.1e+06 ++ O O |
O O O |
2.05e+06 ++ O |
2e+06 ++---------------------------------------------------------------+
[*] 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,
Xiaolong
3 years, 8 months
- 2
- 1
[lkp] [sched/autogroup] 89fc4bcc4c: unixbench.score -3.9% regression
by kernel test robot
Greeting,
FYI, we noticed a -3.9% regression of unixbench.score due to commit:
commit: 89fc4bcc4c2cdb5a801574e6beaa6dcdb7df108d ("sched/autogroup: a zombie thread must not use autogroup->tg")
url: https://github.com/0day-ci/linux/commits/Oleg-Nesterov/sched-autogroup-us...
in testcase: unixbench
on test machine: 8 threads Intel(R) Core(TM) i7 CPU 870 @ 2.93GHz with 4G memory
with following parameters:
runtime: 300s
nr_task: 100%
test: shell8
test-description: UnixBench is the original BYTE UNIX benchmark suite aims to test performance of Unix-like system.
test-url: https://github.com/kdlucas/byte-unixbench
In addition to that, the commit also has significant impact on the following tests:
+------------------+---------------------------------------------+
| testcase: change | unixbench: unixbench.score -2.0% regression |
| test machine | 8 threads Ivy Bridge with 16G memory |
| test parameters | cpufreq_governor=performance |
| | nr_task=1 |
| | runtime=300s |
| | test=shell8 |
+------------------+---------------------------------------------+
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
=========================================================================================
compiler/kconfig/nr_task/rootfs/runtime/tbox_group/test/testcase:
gcc-6/x86_64-rhel-7.2/100%/debian-x86_64-2016-08-31.cgz/300s/nhm-white2/shell8/unixbench
commit:
ece6c5071c ("sched/autogroup: autogroup_move_group() must never skip sched_move_task()")
89fc4bcc4c ("sched/autogroup: a zombie thread must not use autogroup->tg")
ece6c5071c704da1 89fc4bcc4c2cdb5a801574e6be
---------------- --------------------------
%stddev %change %stddev
\ | \
10387 ± 0% -3.9% 9986 ± 0% unixbench.score
8927670 ± 1% +22.5% 10936936 ± 1% unixbench.time.involuntary_context_switches
3.91e+08 ± 0% -3.8% 3.763e+08 ± 0% unixbench.time.minor_page_faults
721.25 ± 1% -11.9% 635.25 ± 14% unixbench.time.percent_of_cpu_this_job_got
1805 ± 0% -5.7% 1702 ± 0% unixbench.time.system_time
2818 ± 0% -6.0% 2648 ± 0% unixbench.time.user_time
12496048 ± 0% -3.0% 12117388 ± 0% unixbench.time.voluntary_context_switches
81520 ± 2% +208.7% 251690 ± 3% interrupts.CAL:Function_call_interrupts
16623 ± 2% +14.3% 18998 ± 6% vmstat.system.in
1662380 ± 0% -12.1% 1460930 ± 0% softirqs.RCU
453583 ± 0% +70.0% 770974 ± 0% softirqs.SCHED
92.73 ± 1% -10.5% 83.02 ± 12% turbostat.%Busy
2709 ± 1% -10.6% 2420 ± 13% turbostat.Avg_MHz
2.30 ± 5% +219.9% 7.37 ± 11% turbostat.CPU%c1
23674690 ± 0% +194.4% 69707060 ± 1% cpuidle.C1-NHM.time
645225 ± 3% +132.4% 1499655 ± 1% cpuidle.C1-NHM.usage
37236858 ± 0% +316.0% 1.549e+08 ± 0% cpuidle.C1E-NHM.time
592875 ± 0% +278.5% 2243821 ± 0% cpuidle.C1E-NHM.usage
501730 ± 12% +259.0% 1801369 ± 48% cpuidle.C3-NHM.usage
91486 ± 17% +120.4% 201615 ± 13% cpuidle.POLL.usage
1844715 ± 99% +865.5% 17811055 ±173% latency_stats.avg.max
1769825 ±107% +906.2% 17807455 ±173% latency_stats.avg.nfs_wait_on_request.nfs_writepage_setup.nfs_updatepage.nfs_write_end.generic_perform_write.nfs_file_write.__vfs_write.vfs_write.SyS_write.entry_SYSCALL_64_fastpath
887543 ± 85% +355.3% 4040659 ±173% latency_stats.avg.wait_on_page_bit.__filemap_fdatawait_range.filemap_fdatawait_keep_errors.sync_inodes_sb.sync_inodes_one_sb.iterate_supers.sys_sync.entry_SYSCALL_64_fastpath
4785596 ± 77% +1379.0% 70777124 ±173% latency_stats.max.max
2830738 ±110% +2399.8% 70762618 ±173% latency_stats.max.nfs_wait_on_request.nfs_writepage_setup.nfs_updatepage.nfs_write_end.generic_perform_write.nfs_file_write.__vfs_write.vfs_write.SyS_write.entry_SYSCALL_64_fastpath
4709919 ± 80% +1331.4% 67417519 ±173% latency_stats.max.wait_on_page_bit.__filemap_fdatawait_range.filemap_fdatawait_keep_errors.sync_inodes_sb.sync_inodes_one_sb.iterate_supers.sys_sync.entry_SYSCALL_64_fastpath
5845821 ±139% +1118.3% 71222216 ±173% latency_stats.sum.nfs_wait_on_request.nfs_writepage_setup.nfs_updatepage.nfs_write_end.generic_perform_write.nfs_file_write.__vfs_write.vfs_write.SyS_write.entry_SYSCALL_64_fastpath
2583 ± 97% +315.5% 10733 ± 66% latency_stats.sum.pipe_wait.wait_for_partner.fifo_open.do_dentry_open.vfs_open.path_openat.do_filp_open.do_sys_open.SyS_open.entry_SYSCALL_64_fastpath
9358983 ± 88% +633.9% 68687358 ±173% latency_stats.sum.wait_on_page_bit.__filemap_fdatawait_range.filemap_fdatawait_keep_errors.sync_inodes_sb.sync_inodes_one_sb.iterate_supers.sys_sync.entry_SYSCALL_64_fastpath
1.933e+12 ± 0% -2.9% 1.876e+12 ± 0% perf-stat.branch-instructions
3.20 ± 0% +1.5% 3.25 ± 0% perf-stat.branch-miss-rate%
6.186e+10 ± 0% -1.5% 6.093e+10 ± 0% perf-stat.branch-misses
13.03 ± 0% -9.9% 11.74 ± 0% perf-stat.cache-miss-rate%
1.762e+10 ± 0% -11.1% 1.567e+10 ± 0% perf-stat.cache-misses
1.352e+11 ± 0% -1.3% 1.335e+11 ± 0% perf-stat.cache-references
31402862 ± 1% +17.8% 37004055 ± 1% perf-stat.context-switches
1.375e+13 ± 0% -4.5% 1.312e+13 ± 1% perf-stat.cpu-cycles
5617950 ± 0% +4.7% 5884318 ± 0% perf-stat.cpu-migrations
3.14e+12 ± 0% -2.4% 3.065e+12 ± 0% perf-stat.dTLB-loads
1.876e+12 ± 0% -3.4% 1.813e+12 ± 0% perf-stat.dTLB-stores
0.02 ± 0% +2.9% 0.02 ± 0% perf-stat.iTLB-load-miss-rate%
9.881e+12 ± 0% -3.0% 9.585e+12 ± 0% perf-stat.iTLB-loads
9.866e+12 ± 0% -3.2% 9.55e+12 ± 0% perf-stat.instructions
4399 ± 0% -3.0% 4267 ± 0% perf-stat.instructions-per-iTLB-miss
3.832e+08 ± 0% -3.7% 3.688e+08 ± 0% perf-stat.minor-faults
4.846e+09 ± 0% -12.9% 4.222e+09 ± 0% perf-stat.node-loads
1.244e+10 ± 0% -10.2% 1.117e+10 ± 0% perf-stat.node-stores
3.832e+08 ± 0% -3.7% 3.688e+08 ± 0% perf-stat.page-faults
20607 ± 20% -55.1% 9254 ± 35% sched_debug.cfs_rq:/.load.min
96.01 ± 18% +372.0% 453.11 ± 22% sched_debug.cfs_rq:/.load_avg.avg
198.84 ± 68% +503.0% 1199 ± 46% sched_debug.cfs_rq:/.load_avg.max
69.05 ± 1% +281.9% 263.72 ± 14% sched_debug.cfs_rq:/.load_avg.min
42.12 ±106% +639.4% 311.43 ± 57% sched_debug.cfs_rq:/.load_avg.stddev
6615955 ± 1% +29.5% 8566983 ± 11% sched_debug.cfs_rq:/.min_vruntime.avg
6696252 ± 1% +29.3% 8658474 ± 10% sched_debug.cfs_rq:/.min_vruntime.max
6532857 ± 1% +29.1% 8432193 ± 11% sched_debug.cfs_rq:/.min_vruntime.min
55051 ± 11% +30.9% 72050 ± 11% sched_debug.cfs_rq:/.min_vruntime.stddev
0.86 ± 11% -43.4% 0.49 ± 33% sched_debug.cfs_rq:/.nr_running.min
0.09 ± 82% +137.2% 0.22 ± 10% sched_debug.cfs_rq:/.nr_running.stddev
1.65 ± 5% +82.9% 3.01 ± 17% sched_debug.cfs_rq:/.nr_spread_over.avg
0.10 ±103% +94896.3% 91.76 ± 28% sched_debug.cfs_rq:/.removed_load_avg.avg
0.77 ±103% +74775.2% 578.58 ± 24% sched_debug.cfs_rq:/.removed_load_avg.max
0.26 ±103% +80039.9% 204.80 ± 24% sched_debug.cfs_rq:/.removed_load_avg.stddev
0.08 ±101% +17266.8% 14.31 ± 38% sched_debug.cfs_rq:/.removed_util_avg.avg
0.66 ±101% +15678.9% 104.00 ± 32% sched_debug.cfs_rq:/.removed_util_avg.max
0.22 ±101% +15946.5% 34.98 ± 33% sched_debug.cfs_rq:/.removed_util_avg.stddev
55.98 ± 11% -36.5% 35.52 ± 28% sched_debug.cfs_rq:/.runnable_load_avg.min
15462 ±276% +643.4% 114940 ± 35% sched_debug.cfs_rq:/.spread0.avg
95736 ± 52% +115.6% 206391 ± 21% sched_debug.cfs_rq:/.spread0.max
-67604 ±-54% -70.7% -19821 ±-69% sched_debug.cfs_rq:/.spread0.min
55019 ± 11% +30.9% 72029 ± 11% sched_debug.cfs_rq:/.spread0.stddev
28.46 ± 13% +154.0% 72.28 ± 4% sched_debug.cfs_rq:/.util_avg.stddev
684011 ± 6% -5.5% 646663 ± 6% sched_debug.cpu.avg_idle.max
41.32 ± 21% -57.3% 17.65 ± 27% sched_debug.cpu.cpu_load[0].min
13568 ± 6% +21.9% 16539 ± 6% sched_debug.cpu.curr->pid.max
8211 ± 26% -83.2% 1375 ± 80% sched_debug.cpu.curr->pid.min
1874 ± 29% +206.5% 5744 ± 9% sched_debug.cpu.curr->pid.stddev
18868 ± 22% -56.9% 8124 ± 19% sched_debug.cpu.load.min
0.00 ± 3% +59.2% 0.00 ± 40% sched_debug.cpu.next_balance.stddev
1.32 ± 25% -61.3% 0.51 ± 26% sched_debug.cpu.nr_running.min
2159214 ± 2% +25.3% 2705719 ± 9% sched_debug.cpu.nr_switches.avg
4108583 ± 21% +48.1% 6084651 ± 11% sched_debug.cpu.nr_switches.max
1431399 ± 2% +29.1% 1847314 ± 11% sched_debug.cpu.nr_switches.min
941738 ± 28% +45.2% 1367231 ± 12% sched_debug.cpu.nr_switches.stddev
276.77 ± 22% +44.7% 400.52 ± 12% sched_debug.cpu.nr_uninterruptible.max
-378.68 ±-19% +46.6% -554.99 ± -4% sched_debug.cpu.nr_uninterruptible.min
196.86 ± 3% +54.5% 304.21 ± 8% sched_debug.cpu.nr_uninterruptible.stddev
2156505 ± 2% +25.3% 2703135 ± 9% sched_debug.cpu.sched_count.avg
4105269 ± 21% +48.1% 6081846 ± 11% sched_debug.cpu.sched_count.max
1429593 ± 2% +29.1% 1844965 ± 11% sched_debug.cpu.sched_count.min
941292 ± 28% +45.2% 1367022 ± 12% sched_debug.cpu.sched_count.stddev
114854 ± 3% +189.5% 332496 ± 10% sched_debug.cpu.sched_goidle.avg
300694 ± 10% +64.0% 493255 ± 14% sched_debug.cpu.sched_goidle.max
72263 ± 2% +283.2% 276909 ± 9% sched_debug.cpu.sched_goidle.min
2066555 ± 21% +42.4% 2942549 ± 11% sched_debug.cpu.ttwu_count.max
480095 ± 28% +44.7% 694825 ± 11% sched_debug.cpu.ttwu_count.stddev
1877807 ± 23% +38.5% 2600359 ± 12% sched_debug.cpu.ttwu_local.max
475741 ± 29% +46.8% 698192 ± 12% sched_debug.cpu.ttwu_local.stddev
2.26 ± 2% -9.5% 2.04 ± 1% perf-profile.calltrace.cycles-pp._IO_default_xsputn
10.70 ± 1% -9.4% 9.70 ± 1% perf-profile.calltrace.cycles-pp.__vsnprintf_chk
1.01 ± 5% -10.7% 0.90 ± 5% perf-profile.calltrace.cycles-pp.alloc_pages_vma.handle_mm_fault.__do_page_fault.do_page_fault.page_fault
0.00 ± -1% +Inf% 1.01 ± 8% perf-profile.calltrace.cycles-pp.call_cpuidle.cpu_startup_entry.rest_init.start_kernel.x86_64_start_reservations
1.45 ± 8% +374.8% 6.87 ± 4% perf-profile.calltrace.cycles-pp.call_cpuidle.cpu_startup_entry.start_secondary
1.34 ± 4% -10.0% 1.21 ± 1% perf-profile.calltrace.cycles-pp.copy_page_range.copy_process._do_fork.sys_clone.do_syscall_64
2.27 ± 1% -9.8% 2.04 ± 4% perf-profile.calltrace.cycles-pp.copy_strings.do_execveat_common.sys_execve.do_syscall_64.return_from_SYSCALL_64
0.00 ± -1% +Inf% 1.10 ± 8% perf-profile.calltrace.cycles-pp.cpu_startup_entry.rest_init.start_kernel.x86_64_start_reservations.x86_64_start_kernel
1.62 ± 8% +362.7% 7.47 ± 4% perf-profile.calltrace.cycles-pp.cpu_startup_entry.start_secondary
0.00 ± -1% +Inf% 1.00 ± 8% perf-profile.calltrace.cycles-pp.cpuidle_enter.call_cpuidle.cpu_startup_entry.rest_init.start_kernel
1.45 ± 8% +374.6% 6.87 ± 4% perf-profile.calltrace.cycles-pp.cpuidle_enter.call_cpuidle.cpu_startup_entry.start_secondary
0.00 ± -1% +Inf% 0.98 ± 9% perf-profile.calltrace.cycles-pp.cpuidle_enter_state.cpuidle_enter.call_cpuidle.cpu_startup_entry.rest_init
1.43 ± 8% +377.1% 6.81 ± 4% perf-profile.calltrace.cycles-pp.cpuidle_enter_state.cpuidle_enter.call_cpuidle.cpu_startup_entry.start_secondary
2.54 ± 3% -9.9% 2.29 ± 3% perf-profile.calltrace.cycles-pp.do_mprotect_pkey.sys_mprotect.entry_SYSCALL_64_fastpath
1.01 ± 7% -16.8% 0.84 ± 4% perf-profile.calltrace.cycles-pp.do_wait.sys_wait4.entry_SYSCALL_64_fastpath.wait4
1.21 ± 5% -14.5% 1.03 ± 4% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_fastpath.wait4
1.28 ± 2% -9.4% 1.16 ± 4% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_fastpath.write
2.21 ± 3% -9.5% 2.00 ± 1% perf-profile.calltrace.cycles-pp.free_pages_and_swap_cache.tlb_flush_mmu_free.tlb_finish_mmu.exit_mmap.mmput
1.23 ± 2% -12.0% 1.08 ± 3% perf-profile.calltrace.cycles-pp.free_pgtables.exit_mmap.mmput.flush_old_exec.load_elf_binary
1.42 ± 8% +443.5% 7.69 ± 3% perf-profile.calltrace.cycles-pp.intel_idle.cpuidle_enter_state.cpuidle_enter.call_cpuidle.cpu_startup_entry
2.21 ± 4% -9.3% 2.01 ± 3% perf-profile.calltrace.cycles-pp.mprotect_fixup.do_mprotect_pkey.sys_mprotect.entry_SYSCALL_64_fastpath
0.00 ± -1% +Inf% 1.11 ± 8% perf-profile.calltrace.cycles-pp.rest_init.start_kernel.x86_64_start_reservations.x86_64_start_kernel
0.00 ± -1% +Inf% 1.11 ± 8% perf-profile.calltrace.cycles-pp.start_kernel.x86_64_start_reservations.x86_64_start_kernel
1.63 ± 8% +361.4% 7.50 ± 4% perf-profile.calltrace.cycles-pp.start_secondary
1.50 ± 3% -7.2% 1.39 ± 2% perf-profile.calltrace.cycles-pp.sys_access.entry_SYSCALL_64_fastpath
2.56 ± 3% -9.9% 2.31 ± 3% perf-profile.calltrace.cycles-pp.sys_mprotect.entry_SYSCALL_64_fastpath
1.03 ± 6% -16.0% 0.87 ± 5% perf-profile.calltrace.cycles-pp.sys_wait4.entry_SYSCALL_64_fastpath.wait4
1.19 ± 1% -8.8% 1.09 ± 4% perf-profile.calltrace.cycles-pp.sys_write.entry_SYSCALL_64_fastpath.write
1.54 ± 2% -11.9% 1.35 ± 3% perf-profile.calltrace.cycles-pp.tlb_finish_mmu.exit_mmap.mmput.do_exit.do_group_exit
1.48 ± 2% -12.0% 1.30 ± 3% perf-profile.calltrace.cycles-pp.tlb_flush_mmu_free.tlb_finish_mmu.exit_mmap.mmput.do_exit
7.09 ± 1% -8.5% 6.49 ± 1% perf-profile.calltrace.cycles-pp.vfprintf.__vsnprintf_chk
1.15 ± 2% -8.7% 1.04 ± 3% perf-profile.calltrace.cycles-pp.vfs_write.sys_write.entry_SYSCALL_64_fastpath.write
1.32 ± 4% -15.3% 1.12 ± 4% perf-profile.calltrace.cycles-pp.wait4
1.33 ± 2% -10.1% 1.20 ± 4% perf-profile.calltrace.cycles-pp.write
0.00 ± -1% +Inf% 1.11 ± 8% perf-profile.calltrace.cycles-pp.x86_64_start_kernel
0.00 ± -1% +Inf% 1.11 ± 8% perf-profile.calltrace.cycles-pp.x86_64_start_reservations.x86_64_start_kernel
2.78 ± 1% -9.5% 2.52 ± 1% perf-profile.children.cycles-pp._IO_default_xsputn
4.04 ± 0% -9.6% 3.65 ± 1% perf-profile.children.cycles-pp.__alloc_pages_nodemask
10.70 ± 1% -9.4% 9.70 ± 1% perf-profile.children.cycles-pp.__vsnprintf_chk
2.48 ± 3% -9.1% 2.25 ± 2% perf-profile.children.cycles-pp.alloc_pages_current
2.00 ± 3% -10.1% 1.80 ± 3% perf-profile.children.cycles-pp.alloc_pages_vma
1.66 ± 6% +376.3% 7.88 ± 3% perf-profile.children.cycles-pp.call_cpuidle
1.37 ± 4% -9.9% 1.23 ± 1% perf-profile.children.cycles-pp.copy_page_range
2.31 ± 1% -9.9% 2.08 ± 4% perf-profile.children.cycles-pp.copy_strings
1.85 ± 5% +364.3% 8.58 ± 3% perf-profile.children.cycles-pp.cpu_startup_entry
1.66 ± 6% +376.0% 7.88 ± 3% perf-profile.children.cycles-pp.cpuidle_enter
1.63 ± 6% +377.8% 7.80 ± 3% perf-profile.children.cycles-pp.cpuidle_enter_state
2.55 ± 3% -9.7% 2.30 ± 3% perf-profile.children.cycles-pp.do_mprotect_pkey
1.03 ± 7% -16.7% 0.86 ± 4% perf-profile.children.cycles-pp.do_wait
2.35 ± 3% -9.0% 2.14 ± 1% perf-profile.children.cycles-pp.free_pages_and_swap_cache
1.61 ± 6% +376.2% 7.69 ± 3% perf-profile.children.cycles-pp.intel_idle
2.23 ± 4% -9.6% 2.02 ± 3% perf-profile.children.cycles-pp.mprotect_fixup
1.98 ± 3% -8.1% 1.83 ± 1% perf-profile.children.cycles-pp.path_lookupat
1.35 ± 3% -11.4% 1.20 ± 5% perf-profile.children.cycles-pp.pte_alloc_one
0.23 ± 20% +378.5% 1.11 ± 8% perf-profile.children.cycles-pp.rest_init
0.23 ± 20% +378.5% 1.11 ± 8% perf-profile.children.cycles-pp.start_kernel
1.63 ± 8% +361.4% 7.50 ± 4% perf-profile.children.cycles-pp.start_secondary
2.57 ± 3% -10.0% 2.31 ± 3% perf-profile.children.cycles-pp.sys_mprotect
1.05 ± 7% -15.5% 0.89 ± 4% perf-profile.children.cycles-pp.sys_wait4
1.40 ± 2% -8.6% 1.27 ± 4% perf-profile.children.cycles-pp.sys_write
2.60 ± 2% -8.4% 2.39 ± 1% perf-profile.children.cycles-pp.tlb_finish_mmu
2.39 ± 3% -9.3% 2.17 ± 1% perf-profile.children.cycles-pp.tlb_flush_mmu_free
7.54 ± 2% -7.9% 6.94 ± 1% perf-profile.children.cycles-pp.vfprintf
1.32 ± 4% -15.3% 1.12 ± 4% perf-profile.children.cycles-pp.wait4
1.21 ± 3% -9.1% 1.11 ± 1% perf-profile.children.cycles-pp.walk_component
1.45 ± 1% -9.0% 1.32 ± 4% perf-profile.children.cycles-pp.write
0.23 ± 20% +378.5% 1.11 ± 8% perf-profile.children.cycles-pp.x86_64_start_kernel
0.23 ± 20% +378.5% 1.11 ± 8% perf-profile.children.cycles-pp.x86_64_start_reservations
2.75 ± 1% -9.7% 2.48 ± 1% perf-profile.self.cycles-pp._IO_default_xsputn
1.58 ± 2% -12.2% 1.38 ± 3% perf-profile.self.cycles-pp.handle_mm_fault
1.61 ± 6% +376.9% 7.68 ± 3% perf-profile.self.cycles-pp.intel_idle
2.65 ± 1% -11.8% 2.34 ± 1% perf-profile.self.cycles-pp.page_fault
7.05 ± 2% -8.2% 6.48 ± 1% perf-profile.self.cycles-pp.vfprintf
perf-stat.instructions
1e+13 ++**-*-**-*-**-*-**-*-**-*-**-*-**-**-*-**-*-**-*-**-*-**-*-**-*-**-*
9e+12 O+OO O OO O OO O OO O OO O OO O OO |
| : |
8e+12 ++: |
7e+12 ++ |
|: |
6e+12 ++ |
5e+12 ++ |
4e+12 ++ |
|: |
3e+12 ++ |
2e+12 ++ |
| |
1e+12 ++ |
0 *+------------------------------------------------------------------+
perf-stat.cache-misses
1.8e+10 ++-*-**-*-**-**-*-**-*-**-**-*----**---**-*----**---**-**-*-**-**-*
| * ** * *.* * |
1.6e+10 O+OO OO O OO OO O OO O OO OO O OO |
1.4e+10 ++: |
| : |
1.2e+10 ++ |
1e+10 ++ |
|: |
8e+09 ++ |
6e+09 ++ |
|: |
4e+09 ++ |
2e+09 ++ |
| |
0 *+----------------------------------------------------------------+
perf-stat.dTLB-stores
2e+12 ++----------------------------------------------------------------+
1.8e+12 O+OO.OO.O.OO.OO.O.OO.O.OO.OO.O.OO.**.*.**.**.*.**.*.**.**.*.**.**.*
| : |
1.6e+12 ++: |
1.4e+12 ++: |
|: |
1.2e+12 ++ |
1e+12 ++ |
8e+11 ++ |
|: |
6e+11 ++ |
4e+11 ++ |
| |
2e+11 ++ |
0 *+----------------------------------------------------------------+
perf-stat.node-loads
5e+09 ++-*-**-*--*--*--------------*----**---**-*----**---**-**-*-**-**-*
4.5e+09 ++* * * *.**.*.**.** ** * *.* * |
O OO OO O OO OO O OO O OO OO O OO |
4e+09 ++: |
3.5e+09 ++: |
|: |
3e+09 ++ |
2.5e+09 ++ |
2e+09 ++ |
|: |
1.5e+09 ++ |
1e+09 ++ |
| |
5e+08 ++ |
0 *+----------------------------------------------------------------+
perf-stat.node-stores
1.4e+10 ++----------------------------------------------------------------+
| *. .* .* |
1.2e+10 ++**.* *.**.**.*.**.*.**.**.*.** *.*.**.**.* *.*.**.**.*.**.**.*
O OO OO O OO OO O OO O OO OO O OO |
1e+10 ++: |
| : |
8e+09 ++ |
|: |
6e+09 ++ |
|: |
4e+09 ++ |
|: |
2e+09 ++ |
| |
0 *+----------------------------------------------------------------+
perf-stat.page-faults
4e+08 ++----------------------------------------------------------------+
O OO.OO.O.OO.OO.O.OO.O.OO.OO.O.OO.**.*.**.**.*.**.*.**.**.*.**.**.*
3.5e+08 ++: |
3e+08 ++: |
| : |
2.5e+08 ++ |
|: |
2e+08 ++ |
|: |
1.5e+08 ++ |
1e+08 ++ |
| |
5e+07 ++ |
| |
0 *+----------------------------------------------------------------+
perf-stat.minor-faults
4e+08 ++----------------------------------------------------------------+
O OO.OO.O.OO.OO.O.OO.O.OO.OO.O.OO.**.*.**.**.*.**.*.**.**.*.**.**.*
3.5e+08 ++: |
3e+08 ++: |
| : |
2.5e+08 ++ |
|: |
2e+08 ++ |
|: |
1.5e+08 ++ |
1e+08 ++ |
| |
5e+07 ++ |
| |
0 *+----------------------------------------------------------------+
perf-stat.cache-miss-rate_
14 ++---------------------------------------------------------------------+
| **.*.*.**.*.*.**.*.*.**.*.**.*.*.**.*.*.**.*.**.*.*.**.*.*.**.*.*.**.*
12 O+OO O O OO O O OO O O OO O OO O O O |
| : |
10 ++: |
|: |
8 ++ |
|: |
6 ++ |
|: |
4 ++ |
| |
2 ++ |
| |
0 *+---------------------------------------------------------------------+
turbostat.Avg_MHz
3000 ++-------------------------------------------------------------------+
| **.*.**.*.**.*.*.**.*.**.*.**.*.**.*.**.*.**.*.**.*.*.**.*.**.*.**.|
2500 O+OO O OO O OO O O OO O O O OO O *
| : |
| : |
2000 ++ O O |
|: |
1500 ++ |
|: |
1000 ++ |
|: |
| |
500 ++ |
| |
0 *+-------------------------------------------------------------------+
turbostat._Busy
100 ++--------------------------------------------------------------------+
90 ++OO.O.O.*O.*.**.*.**.*.*.**.*.**.*.*.**.*.**.*.*.**.*.**.*.**.*.*.**.*
O : O O OO O OO O O O O O O |
80 ++: |
70 ++: |
|: O O |
60 ++ |
50 ++ |
40 ++ |
|: |
30 ++ |
20 ++ |
| |
10 ++ |
0 *+--------------------------------------------------------------------+
unixbench.score
12000 ++------------------------------------------------------------------+
| |
10000 O+OO.O.OO.O.OO.O.OO.O.OO.O.OO.O.OO.**.*.**.*.**.*.**.*.**.*.**.*.**.*
| : |
| : |
8000 ++: |
|: |
6000 ++ |
|: |
4000 ++ |
|: |
|: |
2000 ++ |
| |
0 *+------------------------------------------------------------------+
interrupts.CAL:Function_call_interrupts
300000 ++-----------------------------------------------------------------+
| |
250000 ++ O OO O O O |
O OO O OO OO O OO O O O O |
| |
200000 ++ |
| |
150000 ++ |
| |
100000 ++* .*.* .* .*.* .* |
| :*.*.**.**.*.**.*.**.*.**.** *.* *.** *.* *.*.**.**.*.**.*
|: |
50000 ++ |
| |
0 *+-----------------------------------------------------------------+
unixbench.time.user_time
3000 ++-------------------------------------------------------------------+
| **.*.**.*.**.*.*.**.*.**.*.**.*.**.*.**.*.**.*.**.*.*.**.*.**.*.**.*
2500 O+OO O OO O OO O O OO O OO O OO O O |
| : |
| : |
2000 ++ |
|: |
1500 ++ |
|: |
1000 ++ |
|: |
| |
500 ++ |
| |
0 *+-------------------------------------------------------------------+
unixbench.time.system_time
2000 ++-------------------------------------------------------------------+
1800 ++ *.*.**.*.**.*.*.**.*.**.*.**.*.**.*.**.*.**.*.**.*.*.**.*.**.*.**.*
O OO O OO O OO O O OO O OO O OO O O |
1600 ++: |
1400 ++: |
| : |
1200 ++ |
1000 ++ |
800 ++ |
|: |
600 ++ |
400 ++ |
| |
200 ++ |
0 *+-------------------------------------------------------------------+
unixbench.time.percent_of_cpu_this_job_got
800 ++--------------------------------------------------------------------+
| **.*.*.**.*.**.*.**.*.*.**.*.**.*.*.**.*.**.*.*.**.*.**.*.**.*.*.**.|
700 O+OO O O OO O OO O OO O O O O O O *
600 ++: |
| : |
500 ++ |
|: O O |
400 ++ |
|: |
300 ++ |
200 ++ |
| |
100 ++ |
| |
0 *+--------------------------------------------------------------------+
unixbench.time.minor_page_faults
4e+08 ++**-**-*-**-**-*-**-*-**-**-*-**-**-*-**-**-*-**-*-**-**-*-**-**-*
O OO OO O OO OO O OO O OO OO O OO |
3.5e+08 ++: |
3e+08 ++: |
|: |
2.5e+08 ++ |
|: |
2e+08 ++ |
|: |
1.5e+08 ++ |
1e+08 ++ |
| |
5e+07 ++ |
| |
0 *+----------------------------------------------------------------+
unixbench.time.voluntary_context_switches
1.4e+07 ++----------------------------------------------------------------+
| * .*.**. .*. .**.*. .*. .*. .**.|
1.2e+07 O+OO.OO.O.OO.OO.O.OO.O.OO.OO O OO ** ** ** **.** ** *
| : |
1e+07 ++: |
|: |
8e+06 ++ |
|: |
6e+06 ++ |
|: |
4e+06 ++ |
| |
2e+06 ++ |
| |
0 *+----------------------------------------------------------------+
[*] bisect-good sample
[O] bisect-bad sample
***************************************************************************************************
lkp-ivb-d01: 8 threads Ivy Bridge with 16G memory
=========================================================================================
compiler/cpufreq_governor/kconfig/nr_task/rootfs/runtime/tbox_group/test/testcase:
gcc-6/performance/x86_64-rhel-7.2/1/debian-x86_64-2016-08-31.cgz/300s/lkp-ivb-d01/shell8/unixbench
commit:
ece6c5071c ("sched/autogroup: autogroup_move_group() must never skip sched_move_task()")
89fc4bcc4c ("sched/autogroup: a zombie thread must not use autogroup->tg")
ece6c5071c704da1 89fc4bcc4c2cdb5a801574e6be
---------------- --------------------------
%stddev %change %stddev
\ | \
10898 ± 0% -2.0% 10676 ± 0% unixbench.score
7573674 ± 0% +2.7% 7780529 ± 0% unixbench.time.involuntary_context_switches
4.105e+08 ± 0% -2.0% 4.021e+08 ± 0% unixbench.time.minor_page_faults
493.00 ± 0% -2.4% 481.00 ± 0% unixbench.time.percent_of_cpu_this_job_got
881.69 ± 0% -2.0% 864.27 ± 0% unixbench.time.system_time
2228 ± 0% -2.6% 2170 ± 0% unixbench.time.user_time
769259 ± 0% +10.4% 849294 ± 0% cpuidle.C6-IVB.usage
1342895 ± 0% +1.2% 1359491 ± 0% interrupts.CAL:Function_call_interrupts
10600 ± 55% -77.3% 2403 ± 77% latency_stats.sum.pipe_wait.wait_for_partner.fifo_open.do_dentry_open.vfs_open.path_openat.do_filp_open.do_sys_open.SyS_open.entry_SYSCALL_64_fastpath
64.07 ± 0% -2.3% 62.61 ± 0% turbostat.%Busy
2365 ± 0% -2.3% 2311 ± 0% turbostat.Avg_MHz
74350 ± 0% +1.1% 75144 ± 0% vmstat.system.cs
18260 ± 0% +2.2% 18669 ± 0% vmstat.system.in
2.041e+12 ± 0% -2.0% 2e+12 ± 0% perf-stat.branch-instructions
4.94e+10 ± 0% -1.7% 4.856e+10 ± 0% perf-stat.branch-misses
1.416e+11 ± 0% -1.2% 1.4e+11 ± 0% perf-stat.cache-references
47059027 ± 0% +1.1% 47560897 ± 0% perf-stat.context-switches
1.187e+13 ± 0% -2.3% 1.16e+13 ± 0% perf-stat.cpu-cycles
8735006 ± 0% -3.1% 8465307 ± 0% perf-stat.cpu-migrations
2.619e+12 ± 0% -1.9% 2.569e+12 ± 0% perf-stat.dTLB-loads
0.05 ± 2% +6.0% 0.06 ± 2% perf-stat.dTLB-store-miss-rate%
1.773e+12 ± 0% -1.7% 1.743e+12 ± 0% perf-stat.dTLB-stores
4.194e+09 ± 0% -3.5% 4.047e+09 ± 0% perf-stat.iTLB-load-misses
1.039e+13 ± 0% -2.0% 1.018e+13 ± 0% perf-stat.instructions
2476 ± 0% +1.6% 2516 ± 0% perf-stat.instructions-per-iTLB-miss
4.018e+08 ± 0% -2.0% 3.936e+08 ± 0% perf-stat.minor-faults
4.018e+08 ± 0% -2.0% 3.936e+08 ± 0% perf-stat.page-faults
86.97 ± 97% +469.0% 494.82 ± 8% sched_debug.cfs_rq:/.load_avg.avg
26.52 ± 2% +1021.0% 297.32 ± 2% sched_debug.cfs_rq:/.load_avg.min
3864291 ± 0% -17.9% 3172471 ± 0% sched_debug.cfs_rq:/.min_vruntime.avg
3894351 ± 0% -18.0% 3194754 ± 0% sched_debug.cfs_rq:/.min_vruntime.max
3792628 ± 0% -17.8% 3117386 ± 0% sched_debug.cfs_rq:/.min_vruntime.min
34857 ± 2% -23.5% 26683 ± 4% sched_debug.cfs_rq:/.min_vruntime.stddev
1.14 ± 7% +86.7% 2.12 ± 4% sched_debug.cfs_rq:/.nr_spread_over.avg
2.05 ± 19% +76.7% 3.61 ± 8% sched_debug.cfs_rq:/.nr_spread_over.max
0.43 ± 28% +98.5% 0.85 ± 7% sched_debug.cfs_rq:/.nr_spread_over.stddev
0.13 ± 90% +96617.8% 123.64 ± 43% sched_debug.cfs_rq:/.removed_load_avg.avg
0.82 ± 91% +74833.3% 613.09 ± 32% sched_debug.cfs_rq:/.removed_load_avg.max
0.29 ± 93% +82941.7% 237.79 ± 36% sched_debug.cfs_rq:/.removed_load_avg.stddev
0.10 ± 89% +24402.8% 25.06 ± 51% sched_debug.cfs_rq:/.removed_util_avg.avg
0.64 ± 93% +21882.1% 139.89 ± 45% sched_debug.cfs_rq:/.removed_util_avg.max
0.22 ± 91% +21951.5% 49.53 ± 46% sched_debug.cfs_rq:/.removed_util_avg.stddev
34858 ± 2% -23.5% 26684 ± 4% sched_debug.cfs_rq:/.spread0.stddev
682.07 ± 0% +42.8% 973.90 ± 1% sched_debug.cfs_rq:/.util_avg.avg
726.77 ± 0% +57.8% 1147 ± 3% sched_debug.cfs_rq:/.util_avg.max
614.93 ± 1% +33.1% 818.57 ± 0% sched_debug.cfs_rq:/.util_avg.min
35.14 ± 8% +176.4% 97.13 ± 8% sched_debug.cfs_rq:/.util_avg.stddev
83858 ± 5% -11.3% 74357 ± 5% sched_debug.cpu.avg_idle.min
0.57 ± 2% +23.7% 0.70 ± 17% sched_debug.cpu.clock.stddev
0.57 ± 2% +23.7% 0.70 ± 17% sched_debug.cpu.clock_task.stddev
18.48 ± 5% +15.6% 21.36 ± 2% sched_debug.cpu.cpu_load[1].min
20.41 ± 3% +13.3% 23.11 ± 2% sched_debug.cpu.cpu_load[2].min
22.27 ± 3% +11.6% 24.86 ± 1% sched_debug.cpu.cpu_load[3].min
23.66 ± 3% +10.0% 26.02 ± 1% sched_debug.cpu.cpu_load[4].min
10776 ± 8% -16.1% 9042 ± 10% sched_debug.cpu.curr->pid.avg
19334 ± 4% -25.3% 14436 ± 13% sched_debug.cpu.curr->pid.max
1044 ± 84% -90.0% 104.89 ± 61% sched_debug.cpu.curr->pid.min
7957 ± 4% -21.4% 6253 ± 15% sched_debug.cpu.curr->pid.stddev
351250 ± 0% -9.6% 317438 ± 0% sched_debug.cpu.ttwu_local.min
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
3 years, 8 months
- 1
- 0
[lkp] [locking/mutex] b341afb325: fsmark.files_per_sec +19.1% improvement
by kernel test robot
Greeting,
FYI, we noticed a +19.1% improvement of fsmark.files_per_sec due to commit:
commit b341afb325eb390f707a82cbefd65cda887302ab ("locking/mutex: Enable optimistic spinning of woken waiter")
https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git master
in testcase: fsmark
on test machine: 48 threads Intel(R) Xeon(R) CPU E5-2697 v2 @ 2.70GHz with 64G memory
with following parameters:
iterations: 1x
nr_threads: 64t
disk: 1BRD_48G
fs: f2fs
filesize: 4M
test_size: 40G
sync_method: fsyncBeforeClose
cpufreq_governor: performance
test-description: The fsmark is a file system benchmark to test synchronous write workloads, for example, mail servers workload.
test-url: https://sourceforge.net/projects/fsmark/
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
=========================================================================================
compiler/cpufreq_governor/disk/filesize/fs/iterations/kconfig/nr_threads/rootfs/sync_method/tbox_group/test_size/testcase:
gcc-6/performance/1BRD_48G/4M/f2fs/1x/x86_64-rhel-7.2/64t/debian-x86_64-2016-08-31.cgz/fsyncBeforeClose/ivb44/40G/fsmark
commit:
a40ca56577 ("locking/mutex: Simplify some ww_mutex code in __mutex_lock_common()")
b341afb325 ("locking/mutex: Enable optimistic spinning of woken waiter")
a40ca56577f628eb b341afb325eb390f707a82cbef
---------------- --------------------------
%stddev %change %stddev
\ | \
299959 ± 1% +61.1% 483129 ± 7% fsmark.app_overhead
204.83 ± 0% +19.1% 243.90 ± 0% fsmark.files_per_sec
50.71 ± 0% -16.7% 42.26 ± 0% fsmark.time.elapsed_time
50.71 ± 0% -16.7% 42.26 ± 0% fsmark.time.elapsed_time.max
2567 ± 0% +521.5% 15955 ± 6% fsmark.time.involuntary_context_switches
17572 ± 0% +150.0% 43937 ± 6% fsmark.time.minor_page_faults
404.50 ± 0% +849.9% 3842 ± 2% fsmark.time.percent_of_cpu_this_job_got
204.92 ± 0% +692.5% 1623 ± 2% fsmark.time.system_time
1549367 ± 0% -71.3% 445198 ± 3% fsmark.time.voluntary_context_switches
39996 ± 1% +130.9% 92352 ± 10% interrupts.CAL:Function_call_interrupts
906.75 ± 14% -43.3% 513.75 ± 43% sched_debug.cfs_rq:/.load_avg.max
139.36 ± 17% -31.6% 95.26 ± 22% sched_debug.cfs_rq:/.load_avg.stddev
1531 ± 24% +97.0% 3016 ± 51% sched_debug.cfs_rq:/.min_vruntime.min
103310 ± 3% +215.5% 325987 ± 3% softirqs.RCU
97813 ± 1% +99.5% 195172 ± 2% softirqs.SCHED
332526 ± 0% +199.5% 995868 ± 1% softirqs.TIMER
793049 ± 0% +18.4% 939339 ± 0% vmstat.io.bo
4.00 ± 0% +812.5% 36.50 ± 4% vmstat.procs.r
60841 ± 0% -64.7% 21488 ± 3% vmstat.system.cs
51391 ± 1% +32.2% 67964 ± 2% vmstat.system.in
183585 ± 0% +20.9% 221920 ± 2% meminfo.Active
116626 ± 1% +33.4% 155560 ± 3% meminfo.Active(anon)
9805 ± 0% +54.0% 15101 ± 4% meminfo.Inactive(anon)
19877 ± 1% +28.2% 25481 ± 0% meminfo.Mapped
16060 ± 8% +274.7% 60179 ± 9% meminfo.Shmem
3894720 ± 0% +11.0% 4322366 ± 0% slabinfo.ext4_io_end.active_objs
60855 ± 0% +11.0% 67537 ± 0% slabinfo.ext4_io_end.active_slabs
3894781 ± 0% +11.0% 4322428 ± 0% slabinfo.ext4_io_end.num_objs
60855 ± 0% +11.0% 67537 ± 0% slabinfo.ext4_io_end.num_slabs
117.00 ± 13% +223.1% 378.00 ± 35% slabinfo.flow_cache.active_objs
117.00 ± 13% +223.1% 378.00 ± 35% slabinfo.flow_cache.num_objs
145334 ± 2% -28.7% 103657 ± 3% cpuidle.C1-IVT.usage
23479479 ± 1% -52.5% 11162178 ± 6% cpuidle.C1E-IVT.time
134538 ± 1% -51.8% 64800 ± 5% cpuidle.C1E-IVT.usage
84096016 ± 0% -58.5% 34903789 ± 6% cpuidle.C3-IVT.time
247060 ± 0% -52.1% 118257 ± 6% cpuidle.C3-IVT.usage
2.099e+09 ± 0% -82.3% 3.72e+08 ± 9% cpuidle.C6-IVT.time
3200411 ± 0% -81.0% 607480 ± 5% cpuidle.C6-IVT.usage
667237 ± 4% -76.2% 158508 ± 11% cpuidle.POLL.time
619.50 ± 1% -50.4% 307.50 ± 11% cpuidle.POLL.usage
10.36 ± 0% +665.0% 79.28 ± 1% turbostat.%Busy
304.75 ± 0% +679.3% 2375 ± 2% turbostat.Avg_MHz
45.57 ± 0% -63.5% 16.65 ± 8% turbostat.CPU%c1
1.60 ± 1% -72.1% 0.45 ± 6% turbostat.CPU%c3
42.47 ± 0% -91.5% 3.63 ± 5% turbostat.CPU%c6
86.59 ± 1% +75.3% 151.79 ± 1% turbostat.CorWatt
3.89 ± 2% -82.5% 0.68 ± 4% turbostat.Pkg%pc2
118.26 ± 1% +55.8% 184.27 ± 1% turbostat.PkgWatt
6.45 ± 0% +12.4% 7.25 ± 1% turbostat.RAMWatt
11099 ± 6% +726.2% 91700 ± 8% latency_stats.hits.call_rwsem_down_read_failed.get_node_info.[f2fs].write_data_page.[f2fs].do_write_data_page.[f2fs].f2fs_write_data_page.[f2fs].f2fs_write_cache_pages.[f2fs].f2fs_write_data_pages.[f2fs].do_writepages.__filemap_fdatawrite_range.filemap_write_and_wait_range.f2fs_do_sync_file.[f2fs].f2fs_sync_file.[f2fs]
1509342 ± 0% -79.7% 307119 ± 2% latency_stats.hits.do_write_page.[f2fs].write_data_page.[f2fs].do_write_data_page.[f2fs].f2fs_write_data_page.[f2fs].f2fs_write_cache_pages.[f2fs].f2fs_write_data_pages.[f2fs].do_writepages.__filemap_fdatawrite_range.filemap_write_and_wait_range.f2fs_do_sync_file.[f2fs].f2fs_sync_file.[f2fs].vfs_fsync_range
1509342 ± 0% -79.7% 307119 ± 2% latency_stats.hits.max
2782 ± 48% +331.5% 12008 ±110% latency_stats.max.call_rwsem_down_read_failed.f2fs_write_data_page.[f2fs].f2fs_write_cache_pages.[f2fs].f2fs_write_data_pages.[f2fs].do_writepages.__filemap_fdatawrite_range.filemap_write_and_wait_range.f2fs_do_sync_file.[f2fs].f2fs_sync_file.[f2fs].vfs_fsync_range.do_fsync.SyS_fsync
2713 ± 50% +338.5% 11897 ±112% latency_stats.max.call_rwsem_down_write_failed.block_operations.[f2fs].write_checkpoint.[f2fs].f2fs_sync_fs.[f2fs].f2fs_do_sync_file.[f2fs].f2fs_sync_file.[f2fs].vfs_fsync_range.do_fsync.SyS_fsync.entry_SYSCALL_64_fastpath
557625 ± 35% +1628.8% 9640132 ± 14% latency_stats.sum.call_rwsem_down_read_failed.get_node_info.[f2fs].write_data_page.[f2fs].do_write_data_page.[f2fs].f2fs_write_data_page.[f2fs].f2fs_write_cache_pages.[f2fs].f2fs_write_data_pages.[f2fs].do_writepages.__filemap_fdatawrite_range.filemap_write_and_wait_range.f2fs_do_sync_file.[f2fs].f2fs_sync_file.[f2fs]
259.25 ± 26% +11491.0% 30049 ± 15% latency_stats.sum.call_rwsem_down_write_failed.get_node_info.[f2fs].new_node_page.[f2fs].new_inode_page.[f2fs].init_inode_metadata.[f2fs].f2fs_add_regular_entry.[f2fs].__f2fs_do_add_link.[f2fs].__f2fs_add_link.[f2fs].f2fs_create.[f2fs].path_openat.do_filp_open.do_sys_open
1961 ± 78% +2287.9% 46839 ± 26% latency_stats.sum.call_rwsem_down_write_failed.set_node_addr.[f2fs].f2fs_write_node_page.[f2fs].fsync_node_pages.[f2fs].f2fs_do_sync_file.[f2fs].f2fs_sync_file.[f2fs].vfs_fsync_range.do_fsync.SyS_fsync.entry_SYSCALL_64_fastpath
18361 ± 34% +437.4% 98672 ± 17% latency_stats.sum.call_rwsem_down_write_failed.set_node_addr.[f2fs].new_node_page.[f2fs].get_dnode_of_data.[f2fs].f2fs_map_blocks.[f2fs].f2fs_preallocate_blocks.[f2fs].f2fs_file_write_iter.[f2fs].__vfs_write.vfs_write.SyS_write.entry_SYSCALL_64_fastpath
5150 ± 7% +698.4% 41117 ± 12% latency_stats.sum.call_rwsem_down_write_failed.set_node_addr.[f2fs].new_node_page.[f2fs].new_inode_page.[f2fs].init_inode_metadata.[f2fs].f2fs_add_regular_entry.[f2fs].__f2fs_do_add_link.[f2fs].__f2fs_add_link.[f2fs].f2fs_create.[f2fs].path_openat.do_filp_open.do_sys_open
2142 ± 86% +1589.8% 36203 ± 19% latency_stats.sum.wait_on_page_bit.__migration_entry_wait.migration_entry_wait.do_swap_page.handle_mm_fault.__do_page_fault.do_page_fault.page_fault
29157 ± 1% +33.3% 38879 ± 3% proc-vmstat.nr_active_anon
2451 ± 0% +54.0% 3774 ± 4% proc-vmstat.nr_inactive_anon
4966 ± 1% +28.2% 6367 ± 0% proc-vmstat.nr_mapped
4014 ± 8% +274.7% 15044 ± 9% proc-vmstat.nr_shmem
29157 ± 1% +33.3% 38879 ± 3% proc-vmstat.nr_zone_active_anon
2451 ± 0% +54.0% 3774 ± 4% proc-vmstat.nr_zone_inactive_anon
6062 ± 1% +425.4% 31852 ± 9% proc-vmstat.numa_hint_faults
3116 ± 8% +549.0% 20226 ± 13% proc-vmstat.numa_hint_faults_local
1183 ± 11% +208.4% 3649 ± 15% proc-vmstat.numa_pages_migrated
9698 ± 0% +267.2% 35609 ± 8% proc-vmstat.numa_pte_updates
14231 ± 3% +84.8% 26296 ± 7% proc-vmstat.pgactivate
137068 ± 2% +15.6% 158474 ± 3% proc-vmstat.pgfault
1183 ± 11% +208.4% 3649 ± 15% proc-vmstat.pgmigrate_success
9732128 ± 1% +14.2% 11110432 ± 3% proc-vmstat.slabs_scanned
7.388e+10 ± 1% +517.0% 4.558e+11 ± 1% perf-stat.branch-instructions
1.52 ± 1% -93.8% 0.09 ± 2% perf-stat.branch-miss-rate%
1.122e+09 ± 0% -61.9% 4.276e+08 ± 0% perf-stat.branch-misses
39.97 ± 1% +18.3% 47.30 ± 0% perf-stat.cache-miss-rate%
2.426e+09 ± 1% -7.3% 2.249e+09 ± 0% perf-stat.cache-misses
6.07e+09 ± 1% -21.7% 4.755e+09 ± 0% perf-stat.cache-references
3215250 ± 0% -70.4% 951966 ± 3% perf-stat.context-switches
7.822e+11 ± 1% +533.1% 4.952e+12 ± 2% perf-stat.cpu-cycles
99660 ± 1% +41.6% 141112 ± 6% perf-stat.cpu-migrations
2.07 ± 0% -80.2% 0.41 ± 0% perf-stat.dTLB-load-miss-rate%
9.147e+10 ± 0% +415.4% 4.714e+11 ± 1% perf-stat.dTLB-loads
37898605 ± 5% -20.2% 30260554 ± 9% perf-stat.dTLB-store-misses
4.826e+10 ± 0% -14.9% 4.106e+10 ± 2% perf-stat.dTLB-stores
56.26 ± 0% -31.1% 38.79 ± 2% perf-stat.iTLB-load-miss-rate%
64645829 ± 0% -75.9% 15565993 ± 3% perf-stat.iTLB-load-misses
50256497 ± 1% -51.1% 24558159 ± 1% perf-stat.iTLB-loads
3.501e+11 ± 1% +379.9% 1.68e+12 ± 1% perf-stat.instructions
5415 ± 0% +1895.0% 108046 ± 2% perf-stat.instructions-per-iTLB-miss
0.45 ± 0% -24.2% 0.34 ± 0% perf-stat.ipc
125817 ± 1% +16.9% 147088 ± 2% perf-stat.minor-faults
36.39 ± 0% +10.8% 40.31 ± 0% perf-stat.node-load-miss-rate%
4.65e+08 ± 1% +42.1% 6.609e+08 ± 0% perf-stat.node-load-misses
8.13e+08 ± 0% +20.4% 9.787e+08 ± 0% perf-stat.node-loads
17.33 ± 1% +32.5% 22.96 ± 0% perf-stat.node-store-miss-rate%
2.49e+08 ± 1% +55.2% 3.863e+08 ± 0% perf-stat.node-store-misses
1.187e+09 ± 0% +9.2% 1.296e+09 ± 0% perf-stat.node-stores
125821 ± 1% +16.9% 147109 ± 2% perf-stat.page-faults
1.85 ± 2% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.__alloc_pages_nodemask.alloc_pages_current.brd_insert_page.brd_do_bvec.brd_make_request
44.41 ± 1% +106.4% 91.67 ± 0% perf-profile.calltrace.cycles-pp.__filemap_fdatawrite_range.filemap_write_and_wait_range.f2fs_do_sync_file.f2fs_sync_file.vfs_fsync_range
5.42 ± 5% -77.3% 1.23 ± 2% perf-profile.calltrace.cycles-pp.__generic_file_write_iter.f2fs_file_write_iter.__vfs_write.vfs_write.sys_write
2.05 ± 3% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.__irqentry_text_start.cpuidle_enter.call_cpuidle.cpu_startup_entry.start_secondary
23.88 ± 2% +251.1% 83.84 ± 1% perf-profile.calltrace.cycles-pp.__mutex_lock_slowpath.mutex_lock.do_write_page.write_data_page.do_write_data_page
2.41 ± 4% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.__mutex_unlock_slowpath.mutex_unlock.do_write_page.write_data_page.do_write_data_page
5.68 ± 5% -39.2% 3.46 ± 12% perf-profile.calltrace.cycles-pp.__percpu_counter_sum.f2fs_balance_fs.f2fs_write_data_page.f2fs_write_cache_pages.f2fs_write_data_pages
4.59 ± 1% -74.9% 1.16 ± 1% perf-profile.calltrace.cycles-pp.__submit_merged_bio.f2fs_submit_page_mbio.do_write_page.write_data_page.do_write_data_page
6.07 ± 6% -77.0% 1.40 ± 2% perf-profile.calltrace.cycles-pp.__vfs_write.vfs_write.sys_write.entry_SYSCALL_64_fastpath
0.94 ± 6% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.try_to_wake_up.wake_up_q.__mutex_unlock_slowpath.mutex_unlock
1.88 ± 2% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.alloc_pages_current.brd_insert_page.brd_do_bvec.brd_make_request.generic_make_request
1.02 ± 1% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.allocate_data_block.do_write_page.write_data_page.do_write_data_page.f2fs_write_data_page
4.00 ± 1% -75.8% 0.97 ± 1% perf-profile.calltrace.cycles-pp.brd_do_bvec.brd_make_request.generic_make_request.submit_bio.__submit_merged_bio
1.98 ± 2% -73.8% 0.52 ± 2% perf-profile.calltrace.cycles-pp.brd_insert_page.brd_do_bvec.brd_make_request.generic_make_request.submit_bio
4.57 ± 1% -74.8% 1.15 ± 1% perf-profile.calltrace.cycles-pp.brd_make_request.generic_make_request.submit_bio.__submit_merged_bio.f2fs_submit_page_mbio
1.19 ± 5% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.call_cpuidle.cpu_startup_entry.rest_init.start_kernel.x86_64_start_reservations
44.17 ± 1% -85.7% 6.30 ± 9% perf-profile.calltrace.cycles-pp.call_cpuidle.cpu_startup_entry.start_secondary
1.53 ± 1% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.clear_page_c_e.__alloc_pages_nodemask.alloc_pages_current.brd_insert_page.brd_do_bvec
2.16 ± 1% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.copy_user_enhanced_fast_string.generic_perform_write.__generic_file_write_iter.f2fs_file_write_iter.__vfs_write
1.24 ± 6% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.cpu_startup_entry.rest_init.start_kernel.x86_64_start_reservations.x86_64_start_kernel
46.67 ± 1% -86.1% 6.47 ± 9% perf-profile.calltrace.cycles-pp.cpu_startup_entry.start_secondary
1.19 ± 5% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.cpuidle_enter.call_cpuidle.cpu_startup_entry.rest_init.start_kernel
44.14 ± 1% -85.7% 6.30 ± 9% perf-profile.calltrace.cycles-pp.cpuidle_enter.call_cpuidle.cpu_startup_entry.start_secondary
41.89 ± 1% -85.1% 6.23 ± 9% perf-profile.calltrace.cycles-pp.cpuidle_enter_state.cpuidle_enter.call_cpuidle.cpu_startup_entry.start_secondary
44.58 ± 1% +105.7% 91.70 ± 0% perf-profile.calltrace.cycles-pp.do_fsync.sys_fsync.entry_SYSCALL_64_fastpath
36.26 ± 1% +140.9% 87.35 ± 0% perf-profile.calltrace.cycles-pp.do_write_data_page.f2fs_write_data_page.f2fs_write_cache_pages.f2fs_write_data_pages.do_writepages
33.38 ± 1% +158.5% 86.31 ± 0% perf-profile.calltrace.cycles-pp.do_write_page.write_data_page.do_write_data_page.f2fs_write_data_page.f2fs_write_cache_pages
44.41 ± 1% +106.4% 91.67 ± 0% perf-profile.calltrace.cycles-pp.do_writepages.__filemap_fdatawrite_range.filemap_write_and_wait_range.f2fs_do_sync_file.f2fs_sync_file
50.94 ± 1% +82.9% 93.18 ± 0% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_fastpath
6.19 ± 4% -40.3% 3.69 ± 11% perf-profile.calltrace.cycles-pp.f2fs_balance_fs.f2fs_write_data_page.f2fs_write_cache_pages.f2fs_write_data_pages.do_writepages
44.58 ± 1% +105.7% 91.70 ± 0% perf-profile.calltrace.cycles-pp.f2fs_do_sync_file.f2fs_sync_file.vfs_fsync_range.do_fsync.sys_fsync
6.04 ± 6% -77.0% 1.39 ± 2% perf-profile.calltrace.cycles-pp.f2fs_file_write_iter.__vfs_write.vfs_write.sys_write.entry_SYSCALL_64_fastpath
1.01 ± 11% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.f2fs_set_data_page_dirty.set_page_dirty.f2fs_write_end.generic_perform_write.__generic_file_write_iter
5.23 ± 0% -74.4% 1.34 ± 0% perf-profile.calltrace.cycles-pp.f2fs_submit_page_mbio.do_write_page.write_data_page.do_write_data_page.f2fs_write_data_page
44.58 ± 1% +105.7% 91.70 ± 0% perf-profile.calltrace.cycles-pp.f2fs_sync_file.vfs_fsync_range.do_fsync.sys_fsync.entry_SYSCALL_64_fastpath
1.17 ± 6% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.f2fs_update_data_blkaddr.write_data_page.do_write_data_page.f2fs_write_data_page.f2fs_write_cache_pages
1.03 ± 7% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.f2fs_update_extent_cache.f2fs_update_data_blkaddr.write_data_page.do_write_data_page.f2fs_write_data_page
0.94 ± 6% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.f2fs_update_extent_tree_range.f2fs_update_extent_cache.f2fs_update_data_blkaddr.write_data_page.do_write_data_page
1.14 ± 3% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.f2fs_write_begin.generic_perform_write.__generic_file_write_iter.f2fs_file_write_iter.__vfs_write
44.38 ± 1% +106.5% 91.67 ± 0% perf-profile.calltrace.cycles-pp.f2fs_write_cache_pages.f2fs_write_data_pages.do_writepages.__filemap_fdatawrite_range.filemap_write_and_wait_range
43.43 ± 1% +110.5% 91.43 ± 0% perf-profile.calltrace.cycles-pp.f2fs_write_data_page.f2fs_write_cache_pages.f2fs_write_data_pages.do_writepages.__filemap_fdatawrite_range
44.41 ± 1% +106.4% 91.67 ± 0% perf-profile.calltrace.cycles-pp.f2fs_write_data_pages.do_writepages.__filemap_fdatawrite_range.filemap_write_and_wait_range.f2fs_do_sync_file
1.91 ± 16% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.f2fs_write_end.generic_perform_write.__generic_file_write_iter.f2fs_file_write_iter.__vfs_write
44.41 ± 1% +106.4% 91.67 ± 0% perf-profile.calltrace.cycles-pp.filemap_write_and_wait_range.f2fs_do_sync_file.f2fs_sync_file.vfs_fsync_range.do_fsync
4.59 ± 1% -74.8% 1.16 ± 1% perf-profile.calltrace.cycles-pp.generic_make_request.submit_bio.__submit_merged_bio.f2fs_submit_page_mbio.do_write_page
5.38 ± 5% -77.6% 1.21 ± 2% perf-profile.calltrace.cycles-pp.generic_perform_write.__generic_file_write_iter.f2fs_file_write_iter.__vfs_write.vfs_write
1.05 ± 4% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.grab_cache_page_write_begin.f2fs_write_begin.generic_perform_write.__generic_file_write_iter.f2fs_file_write_iter
42.33 ± 1% -85.3% 6.21 ± 9% perf-profile.calltrace.cycles-pp.intel_idle.cpuidle_enter_state.cpuidle_enter.call_cpuidle.cpu_startup_entry
24.05 ± 2% +249.1% 83.94 ± 1% perf-profile.calltrace.cycles-pp.mutex_lock.do_write_page.write_data_page.do_write_data_page.f2fs_write_data_page
22.98 ± 2% +264.0% 83.65 ± 1% perf-profile.calltrace.cycles-pp.mutex_optimistic_spin.__mutex_lock_slowpath.mutex_lock.do_write_page.write_data_page
5.49 ± 1% -61.0% 2.14 ± 1% perf-profile.calltrace.cycles-pp.mutex_spin_on_owner.mutex_optimistic_spin.__mutex_lock_slowpath.mutex_lock.do_write_page
2.70 ± 4% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.mutex_unlock.do_write_page.write_data_page.do_write_data_page.f2fs_write_data_page
17.06 ± 4% +374.7% 81.01 ± 1% perf-profile.calltrace.cycles-pp.osq_lock.mutex_optimistic_spin.__mutex_lock_slowpath.mutex_lock.do_write_page
1.04 ± 4% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.pagecache_get_page.grab_cache_page_write_begin.f2fs_write_begin.generic_perform_write.__generic_file_write_iter
1.25 ± 5% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.rest_init.start_kernel.x86_64_start_reservations.x86_64_start_kernel
1.04 ± 11% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.set_page_dirty.f2fs_write_end.generic_perform_write.__generic_file_write_iter.f2fs_file_write_iter
1.99 ± 3% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.smp_apic_timer_interrupt.__irqentry_text_start.cpuidle_enter.call_cpuidle.cpu_startup_entry
1.25 ± 5% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.start_kernel.x86_64_start_reservations.x86_64_start_kernel
46.80 ± 1% -86.1% 6.48 ± 9% perf-profile.calltrace.cycles-pp.start_secondary
4.59 ± 1% -74.8% 1.16 ± 1% perf-profile.calltrace.cycles-pp.submit_bio.__submit_merged_bio.f2fs_submit_page_mbio.do_write_page.write_data_page
44.58 ± 1% +105.7% 91.70 ± 0% perf-profile.calltrace.cycles-pp.sys_fsync.entry_SYSCALL_64_fastpath
6.16 ± 6% -77.1% 1.41 ± 2% perf-profile.calltrace.cycles-pp.sys_write.entry_SYSCALL_64_fastpath
2.04 ± 6% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.try_to_wake_up.wake_up_q.__mutex_unlock_slowpath.mutex_unlock.do_write_page
44.58 ± 1% +105.7% 91.70 ± 0% perf-profile.calltrace.cycles-pp.vfs_fsync_range.do_fsync.sys_fsync.entry_SYSCALL_64_fastpath
6.14 ± 6% -77.0% 1.41 ± 2% perf-profile.calltrace.cycles-pp.vfs_write.sys_write.entry_SYSCALL_64_fastpath
2.12 ± 6% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.wake_up_q.__mutex_unlock_slowpath.mutex_unlock.do_write_page.write_data_page
35.05 ± 1% +148.0% 86.94 ± 0% perf-profile.calltrace.cycles-pp.write_data_page.do_write_data_page.f2fs_write_data_page.f2fs_write_cache_pages.f2fs_write_data_pages
1.25 ± 5% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.x86_64_start_kernel
1.25 ± 5% -100.0% 0.00 ± -1% perf-profile.calltrace.cycles-pp.x86_64_start_reservations.x86_64_start_kernel
2.45 ± 1% -74.9% 0.61 ± 2% perf-profile.children.cycles-pp.__alloc_pages_nodemask
44.41 ± 1% +106.4% 91.67 ± 0% perf-profile.children.cycles-pp.__filemap_fdatawrite_range
5.42 ± 5% -77.0% 1.25 ± 2% perf-profile.children.cycles-pp.__generic_file_write_iter
2.46 ± 3% -83.7% 0.40 ± 5% perf-profile.children.cycles-pp.__irqentry_text_start
23.90 ± 2% +250.8% 83.85 ± 1% perf-profile.children.cycles-pp.__mutex_lock_slowpath
2.41 ± 4% -85.9% 0.34 ± 5% perf-profile.children.cycles-pp.__mutex_unlock_slowpath
5.73 ± 5% -39.3% 3.48 ± 12% perf-profile.children.cycles-pp.__percpu_counter_sum
1.13 ± 5% -82.3% 0.20 ± 3% perf-profile.children.cycles-pp.__schedule
4.90 ± 1% -74.8% 1.23 ± 1% perf-profile.children.cycles-pp.__submit_merged_bio
6.23 ± 6% -76.9% 1.44 ± 2% perf-profile.children.cycles-pp.__vfs_write
1.90 ± 15% -83.3% 0.32 ± 5% perf-profile.children.cycles-pp._raw_spin_lock
3.15 ± 8% -31.6% 2.16 ± 18% perf-profile.children.cycles-pp._raw_spin_lock_irqsave
0.96 ± 5% -93.0% 0.07 ± 6% perf-profile.children.cycles-pp.activate_task
2.50 ± 1% -74.4% 0.64 ± 2% perf-profile.children.cycles-pp.alloc_pages_current
1.02 ± 1% -57.2% 0.44 ± 5% perf-profile.children.cycles-pp.allocate_data_block
4.26 ± 1% -75.7% 1.04 ± 1% perf-profile.children.cycles-pp.brd_do_bvec
2.10 ± 1% -73.7% 0.55 ± 2% perf-profile.children.cycles-pp.brd_insert_page
4.87 ± 1% -74.9% 1.22 ± 1% perf-profile.children.cycles-pp.brd_make_request
45.37 ± 1% -85.8% 6.43 ± 9% perf-profile.children.cycles-pp.call_cpuidle
1.65 ± 0% -76.3% 0.39 ± 3% perf-profile.children.cycles-pp.clear_page_c_e
2.17 ± 1% -78.5% 0.46 ± 1% perf-profile.children.cycles-pp.copy_user_enhanced_fast_string
47.91 ± 1% -86.2% 6.60 ± 9% perf-profile.children.cycles-pp.cpu_startup_entry
45.34 ± 1% -85.8% 6.43 ± 9% perf-profile.children.cycles-pp.cpuidle_enter
42.80 ± 1% -85.2% 6.34 ± 9% perf-profile.children.cycles-pp.cpuidle_enter_state
44.58 ± 1% +105.7% 91.70 ± 0% perf-profile.children.cycles-pp.do_fsync
36.26 ± 1% +140.9% 87.35 ± 0% perf-profile.children.cycles-pp.do_write_data_page
33.39 ± 1% +158.5% 86.31 ± 0% perf-profile.children.cycles-pp.do_write_page
44.41 ± 1% +106.4% 91.67 ± 0% perf-profile.children.cycles-pp.do_writepages
51.27 ± 1% +81.9% 93.26 ± 0% perf-profile.children.cycles-pp.entry_SYSCALL_64_fastpath
6.23 ± 4% -40.5% 3.71 ± 11% perf-profile.children.cycles-pp.f2fs_balance_fs
44.58 ± 1% +105.7% 91.70 ± 0% perf-profile.children.cycles-pp.f2fs_do_sync_file
6.04 ± 6% -77.0% 1.39 ± 2% perf-profile.children.cycles-pp.f2fs_file_write_iter
1.01 ± 11% -76.9% 0.23 ± 5% perf-profile.children.cycles-pp.f2fs_set_data_page_dirty
5.24 ± 0% -74.5% 1.34 ± 0% perf-profile.children.cycles-pp.f2fs_submit_page_mbio
44.58 ± 1% +105.7% 91.70 ± 0% perf-profile.children.cycles-pp.f2fs_sync_file
1.18 ± 6% -69.4% 0.36 ± 4% perf-profile.children.cycles-pp.f2fs_update_data_blkaddr
1.11 ± 5% -68.3% 0.35 ± 3% perf-profile.children.cycles-pp.f2fs_update_extent_cache
0.95 ± 6% -68.2% 0.30 ± 3% perf-profile.children.cycles-pp.f2fs_update_extent_tree_range
1.14 ± 3% -71.9% 0.32 ± 2% perf-profile.children.cycles-pp.f2fs_write_begin
44.38 ± 1% +106.5% 91.67 ± 0% perf-profile.children.cycles-pp.f2fs_write_cache_pages
43.43 ± 1% +110.5% 91.43 ± 0% perf-profile.children.cycles-pp.f2fs_write_data_page
44.41 ± 1% +106.4% 91.67 ± 0% perf-profile.children.cycles-pp.f2fs_write_data_pages
1.91 ± 16% -80.2% 0.38 ± 4% perf-profile.children.cycles-pp.f2fs_write_end
44.41 ± 1% +106.4% 91.67 ± 0% perf-profile.children.cycles-pp.filemap_write_and_wait_range
4.89 ± 1% -74.8% 1.23 ± 1% perf-profile.children.cycles-pp.generic_make_request
5.38 ± 5% -77.3% 1.22 ± 2% perf-profile.children.cycles-pp.generic_perform_write
1.06 ± 4% -71.4% 0.30 ± 1% perf-profile.children.cycles-pp.grab_cache_page_write_begin
1.01 ± 4% -73.7% 0.27 ± 6% perf-profile.children.cycles-pp.hrtimer_interrupt
42.35 ± 1% -85.1% 6.33 ± 9% perf-profile.children.cycles-pp.intel_idle
0.97 ± 5% -89.4% 0.10 ± 4% perf-profile.children.cycles-pp.irq_exit
1.06 ± 5% -75.0% 0.27 ± 6% perf-profile.children.cycles-pp.local_apic_timer_interrupt
24.40 ± 2% +244.7% 84.09 ± 1% perf-profile.children.cycles-pp.mutex_lock
23.00 ± 2% +263.7% 83.65 ± 1% perf-profile.children.cycles-pp.mutex_optimistic_spin
5.50 ± 1% -61.0% 2.14 ± 1% perf-profile.children.cycles-pp.mutex_spin_on_owner
2.80 ± 5% -81.9% 0.51 ± 3% perf-profile.children.cycles-pp.mutex_unlock
3.15 ± 19% -39.1% 1.92 ± 20% perf-profile.children.cycles-pp.native_queued_spin_lock_slowpath
17.07 ± 4% +374.6% 81.01 ± 1% perf-profile.children.cycles-pp.osq_lock
1.32 ± 3% -71.5% 0.38 ± 2% perf-profile.children.cycles-pp.pagecache_get_page
1.25 ± 5% -89.4% 0.13 ± 18% perf-profile.children.cycles-pp.rest_init
1.22 ± 5% -82.8% 0.21 ± 3% perf-profile.children.cycles-pp.schedule
1.15 ± 7% -86.3% 0.16 ± 5% perf-profile.children.cycles-pp.schedule_preempt_disabled
1.11 ± 10% -77.0% 0.26 ± 6% perf-profile.children.cycles-pp.set_page_dirty
2.40 ± 2% -83.5% 0.39 ± 6% perf-profile.children.cycles-pp.smp_apic_timer_interrupt
1.25 ± 5% -89.4% 0.13 ± 18% perf-profile.children.cycles-pp.start_kernel
46.80 ± 1% -86.1% 6.48 ± 9% perf-profile.children.cycles-pp.start_secondary
4.89 ± 1% -74.9% 1.23 ± 1% perf-profile.children.cycles-pp.submit_bio
44.58 ± 1% +105.7% 91.70 ± 0% perf-profile.children.cycles-pp.sys_fsync
6.33 ± 6% -76.9% 1.46 ± 2% perf-profile.children.cycles-pp.sys_write
2.12 ± 5% -90.5% 0.20 ± 4% perf-profile.children.cycles-pp.try_to_wake_up
1.03 ± 5% -93.0% 0.07 ± 5% perf-profile.children.cycles-pp.ttwu_do_activate
44.58 ± 1% +105.7% 91.70 ± 0% perf-profile.children.cycles-pp.vfs_fsync_range
6.30 ± 6% -76.9% 1.45 ± 2% perf-profile.children.cycles-pp.vfs_write
2.17 ± 5% -90.2% 0.21 ± 3% perf-profile.children.cycles-pp.wake_up_q
35.06 ± 1% +148.0% 86.94 ± 0% perf-profile.children.cycles-pp.write_data_page
1.25 ± 5% -89.4% 0.13 ± 18% perf-profile.children.cycles-pp.x86_64_start_kernel
1.25 ± 5% -89.4% 0.13 ± 18% perf-profile.children.cycles-pp.x86_64_start_reservations
2.71 ± 3% -56.0% 1.19 ± 3% perf-profile.self.cycles-pp.__percpu_counter_sum
1.10 ± 8% -79.2% 0.23 ± 5% perf-profile.self.cycles-pp._raw_spin_lock
1.49 ± 4% -75.9% 0.36 ± 3% perf-profile.self.cycles-pp.brd_do_bvec
1.65 ± 0% -76.3% 0.39 ± 3% perf-profile.self.cycles-pp.clear_page_c_e
2.17 ± 1% -78.5% 0.46 ± 1% perf-profile.self.cycles-pp.copy_user_enhanced_fast_string
42.32 ± 1% -85.1% 6.33 ± 9% perf-profile.self.cycles-pp.intel_idle
5.50 ± 1% -61.0% 2.14 ± 1% perf-profile.self.cycles-pp.mutex_spin_on_owner
3.15 ± 19% -39.1% 1.92 ± 20% perf-profile.self.cycles-pp.native_queued_spin_lock_slowpath
17.07 ± 4% +374.6% 81.01 ± 1% perf-profile.self.cycles-pp.osq_lock
perf-stat.context-switches
4.5e+06 ++*--*-----------------------*--*----*--*-------*------*-*--------+
* * * * * ** * *.*.** *.**.* : |
4e+06 ++ : : : |
3.5e+06 ++ : : : |
| * .* .* : *.* .* .*
3e+06 ++ *.* .** *.* .* * * * |
| * * |
2.5e+06 ++ |
| |
2e+06 ++ |
1.5e+06 ++ |
| |
1e+06 O+OO OO OO OO OO OO O O OO OO O |
| O OO OO OO |
500000 ++----------------------------------------------------------------+
perf-stat.branch-miss-rate_
1.8 ++--------------------------------------------------------------------+
| * |
1.6 ++ :*.*.**.*
1.4 ++ : |
| : |
1.2 ++ : |
1 ++ : |
| : |
0.8 ++ : |
0.6 ++ : |
| : |
0.4 ++ : |
0.2 ++ : |
O.OO.O.OO.OO.O.OO.O.OO.OO.O.OO.O.OO.OO.O.OO.*.**.**.*.**.*.** |
0 ++--------------------------------------------------------------------+
fsmark.time.voluntary_context_switches
2.2e+06 ++*--*-----------------------*--*----*--*-------*------*-*--------+
* * * * * ** * *.*. * *.**.* : |
2e+06 ++ : : * : |
1.8e+06 ++ : : : |
| * : *.* |
1.6e+06 ++ *.* **.**. .*: *.**.*
1.4e+06 ++ :+ * .* * |
| * * |
1.2e+06 ++ |
1e+06 ++ |
| |
800000 ++ |
600000 ++ |
O OO OO OO OO O O O |
400000 ++-------------O--O-OO-O-OO-OO-OO-OO-OO---------------------------+
[*] 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,
Xiaolong
3 years, 8 months
- 1
- 0
[mm] 2d66cccd73: kernel BUG at mm/slub.c:889!
by kernel test robot
Greetings,
0day kernel testing robot got the below dmesg and the first bad commit is
https://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-rcu.git rcu/next
commit 2d66cccd73436ac9985a08e5c2f82e4344f72264
Author: Paul E. McKenney <paulmck(a)linux.vnet.ibm.com>
AuthorDate: Fri Nov 25 12:48:10 2016 -0800
Commit: Paul E. McKenney <paulmck(a)linux.vnet.ibm.com>
CommitDate: Sat Nov 26 13:10:36 2016 -0800
mm: Prevent __alloc_pages_nodemask() RCU CPU stall warnings
This commit adds a cond_resched_rcu_qs() call in __alloc_pages_nodemask()
(which is invoked indirectly from alloc_page_interleave()) in order to
prevent RCU CPU stall warnings. More importantly, this change allows
RCU grace periods to complete despite "longer than average" loops in
__alloc_pages_nodemask().
Note that this is a case of "treating the symptoms". In addition,
it is quite possible that the cond_resched_rcu_qs() needs to be
in some lower-level function that might have been inlined into
__alloc_pages_nodemask(). In short, this one is experimental in nature.
Reported-by: Boris Zhmurov <bb(a)kernelpanic.ru>
Signed-off-by: Paul E. McKenney <paulmck(a)linux.vnet.ibm.com>
Cc: Michal Hocko <mhocko(a)kernel.org>
Cc: Andrew Morton <akpm(a)linux-foundation.org>
Cc: Johannes Weiner <hannes(a)cmpxchg.org>
Cc: Vlastimil Babka <vbabka(a)suse.cz>
Cc: <linux-mm(a)kvack.org>
+-------------------------------------------------+------------+------------+------------+
| | 34c53f5cd3 | 2d66cccd73 | 400d17262d |
+-------------------------------------------------+------------+------------+------------+
| boot_successes | 79 | 0 | 0 |
| boot_failures | 12 | 26 | 21 |
| BUG:kernel_reboot-without-warning_in_test_stage | 9 | | |
| BUG:unable_to_handle_kernel | 3 | | |
| Oops | 3 | | |
| EIP_is_at_tty_wakeup | 2 | | |
| kernel_BUG_at_mm/slub.c | 0 | 26 | 21 |
| invalid_opcode:#[##]SMP | 0 | 26 | 21 |
| EIP_is_at_check_slab | 0 | 26 | 21 |
| calltrace:cryptomgr_test | 0 | 4 | 6 |
| Kernel_panic-not_syncing:Fatal_exception | 0 | 26 | 21 |
| calltrace:SyS_newlstat | 0 | 1 | |
| calltrace:populate_rootfs | 0 | 1 | |
| calltrace:init_o2nm | 0 | 19 | 11 |
| calltrace:pty_init | 0 | 1 | 2 |
| calltrace:sock_diag_init | 0 | 1 | 1 |
| calltrace:brd_init | 0 | 0 | 1 |
+-------------------------------------------------+------------+------------+------------+
[ 3.892754] ocfs2: Registered cluster interface user
[ 3.895978] OCFS2 User DLM kernel interface loaded
[ 3.898502] ------------[ cut here ]------------
[ 3.900676] kernel BUG at mm/slub.c:889!
[ 3.903302] invalid opcode: 0000 [#1] SMP
[ 3.905231] CPU: 0 PID: 1 Comm: swapper/0 Not tainted 4.9.0-rc1-00047-g2d66ccc #1
[ 3.908734] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Debian-1.8.2-1 04/01/2014
[ 3.912577] task: cf4ca840 task.stack: cf4c4000
[ 3.914567] EIP: 0060:[<c57eec12>] EFLAGS: 00210246 CPU: 0
[ 3.916808] EIP is at check_slab+0x32/0x162
[ 3.918808] EAX: 00200246 EBX: cfc5c2e0 ECX: cd817d08 EDX: cfc5c2e0
[ 3.921428] ESI: cf40ac40 EDI: cfc5c2e0 EBP: cf4c5e04 ESP: cf4c5df4
[ 3.923823] DS: 007b ES: 007b FS: 00d8 GS: 0000 SS: 0068
[ 3.926030] CR0: 80050033 CR2: 00000000 CR3: 0780a000 CR4: 001406d0
[ 3.928546] Stack:
[ 3.929961] cd817d08 cf40ac40 cfc5c2e0 cfc5c2e0 cf4c5e1c c57f0adb cd817d08 cd817d08
[ 3.934662] cf40ac40 cfc5c2e0 cf4c5e70 c57f240b c5e54547 c5e54547 02089220 c5e54547
[ 3.959494] 02089220 cf406820 cfaa3b70 00000000 80130013 cf4c5e50 80130013 cf4c5e58
[ 3.965689] Call Trace:
[ 3.966943] [<c57f0adb>] alloc_debug_processing+0x52/0x1f8
[ 3.972845] [<c57f240b>] ___slab_alloc+0x5cf/0x5f6
[ 3.974757] [<c5e54547>] ? __debug_object_init+0x96/0x5f0
[ 3.980653] [<c5e54547>] ? __debug_object_init+0x96/0x5f0
[ 3.982692] [<c5e54547>] ? __debug_object_init+0x96/0x5f0
[ 3.988588] [<c569547a>] ? irq_exit+0xef/0xff
[ 3.990215] [<c57f246d>] __slab_alloc+0x3b/0x69
[ 3.996157] [<c57ec072>] ? arch_local_irq_save+0x2d/0x40
[ 3.998025] [<c57f246d>] ? __slab_alloc+0x3b/0x69
[ 4.003834] [<c5e54547>] ? __debug_object_init+0x96/0x5f0
[ 4.005645] [<c57f260a>] kmem_cache_alloc+0x99/0x19a
[ 4.012128] [<c5e54547>] ? __debug_object_init+0x96/0x5f0
[ 4.016292] [<c5e54547>] __debug_object_init+0x96/0x5f0
[ 4.020495] [<c5e54ad0>] debug_object_init+0x2f/0x3f
[ 4.024645] [<c571289a>] debug_timer_init+0x1b/0x2b
[ 4.028615] [<c5713523>] init_timer_key+0x1d/0x44
[ 4.032422] [<c5beebab>] o2net_init+0x188/0x328
[ 4.036212] [<c77041d7>] init_o2nm+0x42/0x195
[ 4.039552] [<c7704195>] ? init_dlmfs_fs+0x190/0x190
[ 4.042550] [<c76c2943>] do_one_initcall+0x10e/0x21a
[ 4.046650] [<c56b8f75>] ? parse_args+0x487/0x637
[ 4.051673] [<c76c2dfa>] ? kernel_init_freeable+0x3ab/0x54f
[ 4.056178] [<c76c2e24>] kernel_init_freeable+0x3d5/0x54f
[ 4.060339] [<c6a8cd93>] ? rest_init+0x107/0x107
[ 4.064473] [<c6a8cda9>] kernel_init+0x16/0x248
[ 4.068677] [<c6a966b7>] ret_from_fork+0x1b/0x28
[ 4.072586] Code: 89 c6 83 05 c0 0c 91 c7 01 89 d3 83 15 c4 0c 91 c7 00 e8 ab d3 ff ff 0f ba e0 09 73 1e 83 05 c8 0c 91 c7 01 83 15 cc 0c 91 c7 00 <0f> 0b 83 05 d8 0c 91 c7 01 83 15 dc 0c 91 c7 00 83 05 d0 0c 91
[ 4.104457] EIP: [<c57eec12>]
[ 4.106353] check_slab+0x32/0x162
[ 4.110351] SS:ESP 0068:cf4c5df4
[ 4.114329] ---[ end trace dd772c7c561f31f1 ]---
[ 4.118334] Kernel panic - not syncing: Fatal exception
git bisect start 400d17262dbf73320d5edc8e7238c899cfe68f3f 9c763584b7c8911106bb77af7e648bef09af9d80 --
git bisect good 02cc14a714bc1dc19d3596ea72fd23ab89d3b612 # 05:42 22+ 3 0day base guard for 'devel-catchup-201611270517'
git bisect bad 9db4f6016e94155f5c4f1caaee35b6a45265f57b # 05:46 0- 22 Merge 'rcu/rcu/next' into devel-catchup-201611270517
git bisect good f0b35a041787b98b165892fcc8e0a1fe821b4c60 # 05:55 21+ 7 rcu: Abstract dynticks extended quiescent state enter/exit operations
git bisect good 416b6b9a878141238d34819f60d0703bd921d8da # 06:06 22+ 2 torture: Update RCU test scenario documentation
git bisect good eb2679bc7d58aca8d6451488caf5e995697427be # 06:14 21+ 1 rcu: Remove unused but set variable
git bisect good d2db185bfee894c573faebed93461e9938bdbb61 # 06:22 22+ 2 rcu: Remove short-term CPU kicking
git bisect bad 2d66cccd73436ac9985a08e5c2f82e4344f72264 # 06:42 0- 17 mm: Prevent __alloc_pages_nodemask() RCU CPU stall warnings
git bisect good 34c53f5cd399801b083047cc9cf2ad3ed17c3144 # 06:51 22+ 2 mm: Prevent shrink_node_memcg() RCU CPU stall warnings
# first bad commit: [2d66cccd73436ac9985a08e5c2f82e4344f72264] mm: Prevent __alloc_pages_nodemask() RCU CPU stall warnings
git bisect good 34c53f5cd399801b083047cc9cf2ad3ed17c3144 # 06:55 65+ 12 mm: Prevent shrink_node_memcg() RCU CPU stall warnings
# extra tests with CONFIG_DEBUG_INFO_REDUCED
git bisect bad 2d66cccd73436ac9985a08e5c2f82e4344f72264 # 07:04 0- 26 mm: Prevent __alloc_pages_nodemask() RCU CPU stall warnings
# extra tests on HEAD of linux-devel/devel-catchup-201611270517
git bisect bad 400d17262dbf73320d5edc8e7238c899cfe68f3f # 07:04 0- 21 0day head guard for 'devel-catchup-201611270517'
# extra tests on tree/branch rcu/rcu/next
git bisect bad 2d66cccd73436ac9985a08e5c2f82e4344f72264 # 07:05 0- 24 mm: Prevent __alloc_pages_nodemask() RCU CPU stall warnings
# extra tests with first bad commit reverted
git bisect good ecc15013f1621ac46efefdaa4d242266367a7c43 # 07:33 66+ 10 Revert "mm: Prevent __alloc_pages_nodemask() RCU CPU stall warnings"
# extra tests on tree/branch linus/master
git bisect good a0d60e62ea5c88a9823410e9d0929a513e29dea2 # 07:38 65+ 6 Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net
# extra tests on tree/branch linux-next/master
git bisect good 09bfc750ec237029d69ebb749602b6f1304d028a # 07:46 66+ 14 Add linux-next specific files for 20161125
---
0-DAY kernel test infrastructure Open Source Technology Center
https://lists.01.org/pipermail/lkp Intel Corporation
3 years, 8 months
- 1
- 0