April 2019 - LKP - Ml01.01.Org

[tcp] 8b27dae5a2: netperf.Throughput_Mbps -25.7% regression

by kernel test robot

Greeting, FYI, we noticed a -25.7% regression of netperf.Throughput_Mbps due to commit: commit: 8b27dae5a2e89a61c46c6dbc76c040c0e6d0ed4c ("tcp: add one skb cache for rx") https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git master in testcase: netperf on test machine: 104 threads Skylake with 192G memory with following parameters: ip: ipv4 runtime: 900s nr_threads: 200% cluster: cs-localhost test: TCP_STREAM 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_Mbps -25.6% regression | | test machine | 104 threads Skylake with 192G memory | | test parameters | cluster=cs-localhost | | | cpufreq_governor=performance | | | ip=ipv4 | | | nr_threads=200% | | | runtime=900s | | | test=TCP_MAERTS | +------------------+----------------------------------------------------+ Details are as below: --------------------------------------------------------------------------------------------------> To reproduce: git clone https://github.com/intel/lkp-tests.git cd lkp-tests bin/lkp install job.yaml # job file is attached in this email bin/lkp run job.yaml ========================================================================================= cluster/compiler/cpufreq_governor/ip/kconfig/nr_threads/rootfs/runtime/tbox_group/test/testcase: cs-localhost/gcc-7/performance/ipv4/x86_64-rhel-7.6/200%/debian-x86_64-2018-04-03.cgz/900s/lkp-skl-fpga01/TCP_STREAM/netperf commit: 472c2e07ee ("tcp: add one skb cache for tx") 8b27dae5a2 ("tcp: add one skb cache for rx") 472c2e07eef04514 8b27dae5a2e89a61c46c6dbc76c ---------------- --------------------------- fail:runs %reproduction fail:runs | | | 1:4 -25% :4 dmesg.WARNING:at#for_ip_interrupt_entry/0x :4 25% 1:4 dmesg.WARNING:at_ip_ip_finish_output2/0x %stddev %change %stddev \ | \ 3939 -25.7% 2925 netperf.Throughput_Mbps 819330 -25.7% 608425 netperf.Throughput_total_Mbps 3.725e+09 -25.8% 2.764e+09 netperf.time.involuntary_context_switches 7399 -1.8% 7266 netperf.time.percent_of_cpu_this_job_got 65271 -1.0% 64597 netperf.time.system_time 1529 -34.0% 1009 netperf.time.user_time 5.626e+09 -25.7% 4.178e+09 netperf.workload 12.36 ± 4% +6.5 18.84 mpstat.cpu.all.soft% 2.95 ± 2% -0.8 2.15 mpstat.cpu.all.usr% 9.557e+08 +13.2% 1.082e+09 numa-numastat.node0.local_node 9.557e+08 +13.2% 1.082e+09 numa-numastat.node0.numa_hit 4.845e+08 ± 2% +11.6% 5.407e+08 numa-vmstat.node0.numa_hit 4.845e+08 ± 2% +11.6% 5.406e+08 numa-vmstat.node0.numa_local 2662 ± 2% +4.2% 2774 turbostat.Avg_MHz 1.87 ± 86% -66.1% 0.63 ± 9% turbostat.CPU%c1 94.75 ± 2% +2.4% 97.00 vmstat.cpu.sy 8145843 ± 2% -24.7% 6134835 vmstat.system.cs 224329 -0.8% 222470 vmstat.system.in 1.884e+09 +10.9% 2.09e+09 proc-vmstat.numa_hit 1.884e+09 +10.9% 2.089e+09 proc-vmstat.numa_local 1.506e+10 +10.9% 1.671e+10 proc-vmstat.pgalloc_normal 1.506e+10 +10.9% 1.671e+10 proc-vmstat.pgfree 2104 ± 8% +86.4% 3922 ± 58% sched_debug.cpu.avg_idle.min 35810547 -25.7% 26600377 sched_debug.cpu.nr_switches.avg 37423347 -25.1% 28047296 sched_debug.cpu.nr_switches.max 32860210 -25.5% 24491190 sched_debug.cpu.nr_switches.min 968.00 ± 4% -18.2% 792.00 ± 7% slabinfo.kmalloc-rcl-128.active_objs 968.00 ± 4% -18.2% 792.00 ± 7% slabinfo.kmalloc-rcl-128.num_objs 16611 ± 2% +6.1% 17628 ± 4% slabinfo.skbuff_fclone_cache.active_objs 16827 ± 2% +5.7% 17794 ± 4% slabinfo.skbuff_fclone_cache.num_objs 18.78 ± 5% +28.9% 24.21 perf-stat.i.MPKI 1.942e+10 ± 2% -21.1% 1.531e+10 perf-stat.i.branch-instructions 2.08 ± 6% -0.1 1.97 perf-stat.i.branch-miss-rate% 3.892e+08 ± 2% -22.6% 3.014e+08 perf-stat.i.branch-misses 1.767e+09 ± 2% +4.3% 1.842e+09 perf-stat.i.cache-references 8171306 ± 2% -24.7% 6151961 perf-stat.i.context-switches 2.88 ± 2% +31.0% 3.78 perf-stat.i.cpi 2.761e+11 ± 2% +4.2% 2.876e+11 perf-stat.i.cpu-cycles 1.10 ± 3% -0.1 0.97 perf-stat.i.dTLB-load-miss-rate% 3.254e+08 ± 4% -32.1% 2.208e+08 perf-stat.i.dTLB-load-misses 2.893e+10 ± 2% -21.9% 2.26e+10 perf-stat.i.dTLB-loads 58990104 ± 5% -25.8% 43766302 perf-stat.i.dTLB-store-misses 1.658e+10 ± 2% -23.6% 1.266e+10 perf-stat.i.dTLB-stores 55.76 -3.4 52.34 perf-stat.i.iTLB-load-miss-rate% 63191122 ± 2% -29.1% 44815478 perf-stat.i.iTLB-load-misses 49413835 ± 2% -17.7% 40661018 perf-stat.i.iTLB-loads 9.688e+10 ± 2% -21.5% 7.605e+10 perf-stat.i.instructions 1607 ± 3% +8.1% 1737 perf-stat.i.instructions-per-iTLB-miss 0.35 -24.3% 0.27 perf-stat.i.ipc 86243 ± 29% +52.4% 131400 ± 18% perf-stat.i.node-stores 18.24 +32.8% 24.23 perf-stat.overall.MPKI 2.00 -0.0 1.97 perf-stat.overall.branch-miss-rate% 2.85 +32.7% 3.78 perf-stat.overall.cpi 1.11 ± 3% -0.1 0.97 perf-stat.overall.dTLB-load-miss-rate% 56.12 -3.7 52.43 perf-stat.overall.iTLB-load-miss-rate% 1533 ± 2% +10.7% 1697 perf-stat.overall.instructions-per-iTLB-miss 0.35 -24.6% 0.26 perf-stat.overall.ipc 15748 +4.3% 16430 perf-stat.overall.path-length 1.939e+10 ± 2% -21.1% 1.53e+10 perf-stat.ps.branch-instructions 3.887e+08 ± 2% -22.5% 3.011e+08 perf-stat.ps.branch-misses 1.764e+09 ± 2% +4.3% 1.84e+09 perf-stat.ps.cache-references 8161286 ± 2% -24.7% 6144845 perf-stat.ps.context-switches 2.757e+11 ± 2% +4.2% 2.872e+11 perf-stat.ps.cpu-cycles 3.25e+08 ± 4% -32.1% 2.206e+08 perf-stat.ps.dTLB-load-misses 2.889e+10 ± 2% -21.8% 2.258e+10 perf-stat.ps.dTLB-loads 58916703 ± 5% -25.8% 43714633 perf-stat.ps.dTLB-store-misses 1.656e+10 ± 2% -23.6% 1.265e+10 perf-stat.ps.dTLB-stores 63112575 ± 2% -29.1% 44762927 perf-stat.ps.iTLB-load-misses 49353553 ± 2% -17.7% 40614004 perf-stat.ps.iTLB-loads 9.676e+10 ± 2% -21.5% 7.596e+10 perf-stat.ps.instructions 86182 ± 29% +52.3% 131291 ± 18% perf-stat.ps.node-stores 8.86e+13 -22.5% 6.864e+13 perf-stat.total.instructions 136251 ± 8% -18.5% 110991 ± 10% interrupts.CPU1.RES:Rescheduling_interrupts 128756 ± 3% -18.6% 104869 ± 6% interrupts.CPU10.RES:Rescheduling_interrupts 119826 ± 6% -12.1% 105329 ± 4% interrupts.CPU100.RES:Rescheduling_interrupts 121133 ± 10% -17.9% 99457 interrupts.CPU101.RES:Rescheduling_interrupts 9051 ± 3% +14.1% 10328 interrupts.CPU102.CAL:Function_call_interrupts 131446 ± 7% -17.8% 108104 ± 9% interrupts.CPU102.RES:Rescheduling_interrupts 124627 ± 2% -16.6% 103971 ± 9% interrupts.CPU12.RES:Rescheduling_interrupts 131072 ± 11% -19.9% 105033 ± 4% interrupts.CPU13.RES:Rescheduling_interrupts 127845 ± 8% -17.0% 106112 ± 9% interrupts.CPU14.RES:Rescheduling_interrupts 126159 ± 10% -20.6% 100111 ± 11% interrupts.CPU15.RES:Rescheduling_interrupts 132232 ± 5% -17.9% 108535 ± 7% interrupts.CPU17.RES:Rescheduling_interrupts 127157 ± 4% -17.3% 105165 ± 11% interrupts.CPU18.RES:Rescheduling_interrupts 127948 ± 2% -15.5% 108165 ± 11% interrupts.CPU19.RES:Rescheduling_interrupts 122716 ± 6% -14.0% 105519 ± 11% interrupts.CPU23.RES:Rescheduling_interrupts 130754 ± 5% -20.3% 104222 ± 9% interrupts.CPU24.RES:Rescheduling_interrupts 126283 ± 4% -16.6% 105320 ± 6% interrupts.CPU27.RES:Rescheduling_interrupts 8997 ± 3% -12.4% 7883 ± 9% interrupts.CPU28.CAL:Function_call_interrupts 128467 ± 5% -20.9% 101674 ± 11% interrupts.CPU29.RES:Rescheduling_interrupts 133914 ± 6% -26.0% 99089 ± 10% interrupts.CPU3.RES:Rescheduling_interrupts 8987 ± 3% -11.9% 7922 ± 7% interrupts.CPU31.CAL:Function_call_interrupts 115389 ± 9% -17.7% 94962 ± 5% interrupts.CPU31.RES:Rescheduling_interrupts 8830 ± 5% -12.1% 7759 ± 4% interrupts.CPU32.CAL:Function_call_interrupts 134198 ± 6% -19.6% 107954 ± 6% interrupts.CPU33.RES:Rescheduling_interrupts 8940 ± 3% -8.8% 8154 ± 4% interrupts.CPU34.CAL:Function_call_interrupts 123678 ± 12% -17.1% 102582 ± 2% interrupts.CPU34.RES:Rescheduling_interrupts 8969 ± 3% -9.2% 8143 ± 4% interrupts.CPU35.CAL:Function_call_interrupts 118372 ± 8% -10.9% 105507 ± 9% interrupts.CPU35.RES:Rescheduling_interrupts 8958 ± 3% -9.2% 8131 ± 4% interrupts.CPU36.CAL:Function_call_interrupts 132196 ± 3% -16.2% 110761 ± 7% interrupts.CPU4.RES:Rescheduling_interrupts 123005 ± 6% -17.2% 101838 ± 3% interrupts.CPU40.RES:Rescheduling_interrupts 132802 ± 15% -23.6% 101521 ± 11% interrupts.CPU43.RES:Rescheduling_interrupts 131107 ± 4% -25.6% 97566 ± 3% interrupts.CPU44.RES:Rescheduling_interrupts 127673 ± 8% -24.0% 97028 ± 6% interrupts.CPU47.RES:Rescheduling_interrupts 123709 ± 6% -14.3% 106030 ± 12% interrupts.CPU48.RES:Rescheduling_interrupts 123709 ± 7% -12.4% 108355 ± 6% interrupts.CPU49.RES:Rescheduling_interrupts 135382 ± 5% -20.7% 107334 ± 3% interrupts.CPU51.RES:Rescheduling_interrupts 130424 ± 4% -17.4% 107704 ± 6% interrupts.CPU52.RES:Rescheduling_interrupts 129234 ± 8% -18.6% 105171 ± 10% interrupts.CPU53.RES:Rescheduling_interrupts 131374 ± 6% -18.8% 106699 ± 6% interrupts.CPU54.RES:Rescheduling_interrupts 126141 ± 10% -14.7% 107626 ± 11% interrupts.CPU57.RES:Rescheduling_interrupts 133750 ± 9% -19.9% 107102 interrupts.CPU6.RES:Rescheduling_interrupts 119663 ± 5% -16.7% 99633 ± 3% interrupts.CPU60.RES:Rescheduling_interrupts 121078 ± 7% -13.6% 104670 ± 9% interrupts.CPU61.RES:Rescheduling_interrupts 121662 ± 5% -15.3% 102992 ± 10% interrupts.CPU63.RES:Rescheduling_interrupts 118130 ± 2% -15.1% 100310 ± 5% interrupts.CPU65.RES:Rescheduling_interrupts 128075 ± 6% -18.4% 104495 ± 8% interrupts.CPU67.RES:Rescheduling_interrupts 124700 ± 6% -15.7% 105149 ± 4% interrupts.CPU68.RES:Rescheduling_interrupts 119607 ± 9% -12.7% 104432 ± 6% interrupts.CPU7.RES:Rescheduling_interrupts 119288 ± 4% -11.8% 105251 ± 6% interrupts.CPU71.RES:Rescheduling_interrupts 125678 ± 7% -18.7% 102141 ± 8% interrupts.CPU72.RES:Rescheduling_interrupts 130174 ± 5% -19.8% 104394 ± 3% interrupts.CPU74.RES:Rescheduling_interrupts 9215 ± 2% +9.8% 10119 ± 4% interrupts.CPU76.CAL:Function_call_interrupts 113784 ± 8% -10.6% 101686 ± 7% interrupts.CPU78.RES:Rescheduling_interrupts 122103 ± 5% -16.2% 102305 ± 5% interrupts.CPU79.RES:Rescheduling_interrupts 122242 ± 3% -23.8% 93153 ± 8% interrupts.CPU81.RES:Rescheduling_interrupts 9154 ± 3% +10.3% 10092 ± 4% interrupts.CPU82.CAL:Function_call_interrupts 116414 ± 3% -18.8% 94476 ± 9% interrupts.CPU82.RES:Rescheduling_interrupts 128558 ± 12% -24.5% 97067 ± 2% interrupts.CPU83.RES:Rescheduling_interrupts 9147 ± 3% +10.2% 10078 ± 4% interrupts.CPU84.CAL:Function_call_interrupts 122842 ± 9% -19.8% 98521 ± 2% interrupts.CPU84.RES:Rescheduling_interrupts 9081 ± 3% +10.9% 10071 ± 4% interrupts.CPU85.CAL:Function_call_interrupts 9130 ± 2% +10.2% 10065 ± 4% interrupts.CPU86.CAL:Function_call_interrupts 9068 ± 3% +10.9% 10057 ± 5% interrupts.CPU87.CAL:Function_call_interrupts 9068 ± 3% +10.8% 10050 ± 5% interrupts.CPU88.CAL:Function_call_interrupts 122119 ± 5% -20.1% 97576 ± 3% interrupts.CPU88.RES:Rescheduling_interrupts 9065 ± 3% +12.8% 10223 ± 2% interrupts.CPU89.CAL:Function_call_interrupts 124799 ± 6% -20.4% 99390 ± 8% interrupts.CPU89.RES:Rescheduling_interrupts 130362 ± 2% -12.6% 113981 ± 5% interrupts.CPU9.RES:Rescheduling_interrupts 8902 +12.7% 10035 ± 5% interrupts.CPU90.CAL:Function_call_interrupts 9062 ± 3% +10.7% 10028 ± 5% interrupts.CPU91.CAL:Function_call_interrupts 119442 ± 8% -24.1% 90611 ± 2% interrupts.CPU91.RES:Rescheduling_interrupts 9058 ± 3% +10.7% 10023 ± 5% interrupts.CPU92.CAL:Function_call_interrupts 117081 ± 10% -21.2% 92294 ± 12% interrupts.CPU92.RES:Rescheduling_interrupts 9058 ± 3% +10.6% 10017 ± 5% interrupts.CPU93.CAL:Function_call_interrupts 130977 ± 17% -27.7% 94702 ± 6% interrupts.CPU94.RES:Rescheduling_interrupts 9055 ± 3% +10.5% 10003 ± 5% interrupts.CPU95.CAL:Function_call_interrupts 116952 ± 4% -12.4% 102488 ± 10% interrupts.CPU95.RES:Rescheduling_interrupts 9054 ± 3% +10.4% 9995 ± 5% interrupts.CPU96.CAL:Function_call_interrupts 9054 ± 3% +10.3% 9990 ± 5% interrupts.CPU97.CAL:Function_call_interrupts 118634 ± 5% -12.7% 103578 ± 9% interrupts.CPU98.RES:Rescheduling_interrupts 116356 ± 10% -16.5% 97115 ± 7% interrupts.CPU99.RES:Rescheduling_interrupts 12787758 ± 2% -14.8% 10896246 interrupts.RES:Rescheduling_interrupts 10.84 -10.8 0.00 perf-profile.calltrace.cycles-pp.tcp_ack.tcp_rcv_established.tcp_v4_do_rcv.__release_sock.release_sock 10.56 -10.6 0.00 perf-profile.calltrace.cycles-pp.tcp_clean_rtx_queue.tcp_ack.tcp_rcv_established.tcp_v4_do_rcv.__release_sock 10.16 -10.2 0.00 perf-profile.calltrace.cycles-pp.__kfree_skb.tcp_clean_rtx_queue.tcp_ack.tcp_rcv_established.tcp_v4_do_rcv 10.07 -10.1 0.00 perf-profile.calltrace.cycles-pp.skb_release_data.__kfree_skb.tcp_clean_rtx_queue.tcp_ack.tcp_rcv_established 9.71 -9.7 0.00 perf-profile.calltrace.cycles-pp.__free_pages_ok.skb_release_data.__kfree_skb.tcp_clean_rtx_queue.tcp_ack 9.57 -9.6 0.00 perf-profile.calltrace.cycles-pp.free_one_page.__free_pages_ok.skb_release_data.__kfree_skb.tcp_clean_rtx_queue 21.28 -6.8 14.49 perf-profile.calltrace.cycles-pp.__x64_sys_recvfrom.do_syscall_64.entry_SYSCALL_64_after_hwframe 21.19 -6.8 14.43 perf-profile.calltrace.cycles-pp.__sys_recvfrom.__x64_sys_recvfrom.do_syscall_64.entry_SYSCALL_64_after_hwframe 20.57 -6.6 14.00 perf-profile.calltrace.cycles-pp.inet_recvmsg.__sys_recvfrom.__x64_sys_recvfrom.do_syscall_64.entry_SYSCALL_64_after_hwframe 20.46 -6.5 13.93 perf-profile.calltrace.cycles-pp.tcp_recvmsg.inet_recvmsg.__sys_recvfrom.__x64_sys_recvfrom.do_syscall_64 16.37 -3.0 13.34 perf-profile.calltrace.cycles-pp.release_sock.tcp_sendmsg.sock_sendmsg.__sys_sendto.__x64_sys_sendto 16.18 -3.0 13.21 perf-profile.calltrace.cycles-pp.__release_sock.release_sock.tcp_sendmsg.sock_sendmsg.__sys_sendto 7.01 ± 4% -2.4 4.61 perf-profile.calltrace.cycles-pp.tcp_v4_do_rcv.tcp_v4_rcv.ip_protocol_deliver_rcu.ip_local_deliver_finish.ip_local_deliver 6.90 ± 5% -2.4 4.54 perf-profile.calltrace.cycles-pp.tcp_rcv_established.tcp_v4_do_rcv.tcp_v4_rcv.ip_protocol_deliver_rcu.ip_local_deliver_finish 7.71 ± 2% -2.2 5.51 perf-profile.calltrace.cycles-pp.skb_copy_datagram_iter.tcp_recvmsg.inet_recvmsg.__sys_recvfrom.__x64_sys_recvfrom 7.69 ± 2% -2.2 5.49 perf-profile.calltrace.cycles-pp.__skb_datagram_iter.skb_copy_datagram_iter.tcp_recvmsg.inet_recvmsg.__sys_recvfrom 6.19 ± 5% -2.1 4.10 perf-profile.calltrace.cycles-pp.sock_def_readable.tcp_rcv_established.tcp_v4_do_rcv.tcp_v4_rcv.ip_protocol_deliver_rcu 6.04 ± 5% -2.0 4.00 perf-profile.calltrace.cycles-pp.__wake_up_common_lock.sock_def_readable.tcp_rcv_established.tcp_v4_do_rcv.tcp_v4_rcv 5.89 ± 6% -2.0 3.90 perf-profile.calltrace.cycles-pp.__wake_up_common.__wake_up_common_lock.sock_def_readable.tcp_rcv_established.tcp_v4_do_rcv 7.03 ± 2% -2.0 5.04 perf-profile.calltrace.cycles-pp._copy_to_iter.__skb_datagram_iter.skb_copy_datagram_iter.tcp_recvmsg.inet_recvmsg 5.71 ± 6% -1.9 3.78 perf-profile.calltrace.cycles-pp.try_to_wake_up.__wake_up_common.__wake_up_common_lock.sock_def_readable.tcp_rcv_established 13.46 -1.9 11.54 perf-profile.calltrace.cycles-pp.tcp_v4_do_rcv.__release_sock.release_sock.tcp_sendmsg.sock_sendmsg 6.86 ± 2% -1.9 4.93 perf-profile.calltrace.cycles-pp.copyout._copy_to_iter.__skb_datagram_iter.skb_copy_datagram_iter.tcp_recvmsg 6.72 ± 2% -1.9 4.83 ± 2% perf-profile.calltrace.cycles-pp.copy_user_enhanced_fast_string.copyout._copy_to_iter.__skb_datagram_iter.skb_copy_datagram_iter 13.26 -1.8 11.41 perf-profile.calltrace.cycles-pp.tcp_rcv_established.tcp_v4_do_rcv.__release_sock.release_sock.tcp_sendmsg 5.49 -1.8 3.67 perf-profile.calltrace.cycles-pp.sk_wait_data.tcp_recvmsg.inet_recvmsg.__sys_recvfrom.__x64_sys_recvfrom 2.62 ± 14% -1.8 0.82 ± 5% perf-profile.calltrace.cycles-pp.select_task_rq_fair.try_to_wake_up.__wake_up_common.__wake_up_common_lock.sock_def_readable 5.29 -1.7 3.61 perf-profile.calltrace.cycles-pp.__tcp_transmit_skb.tcp_recvmsg.inet_recvmsg.__sys_recvfrom.__x64_sys_recvfrom 4.80 -1.6 3.17 perf-profile.calltrace.cycles-pp.wait_woken.sk_wait_data.tcp_recvmsg.inet_recvmsg.__sys_recvfrom 4.59 -1.6 3.02 perf-profile.calltrace.cycles-pp.schedule_timeout.wait_woken.sk_wait_data.tcp_recvmsg.inet_recvmsg 4.50 -1.5 2.96 perf-profile.calltrace.cycles-pp.schedule.schedule_timeout.wait_woken.sk_wait_data.tcp_recvmsg 4.41 -1.5 2.90 perf-profile.calltrace.cycles-pp.__sched_text_start.schedule.schedule_timeout.wait_woken.sk_wait_data 4.81 -1.5 3.31 perf-profile.calltrace.cycles-pp.__ip_queue_xmit.__tcp_transmit_skb.tcp_recvmsg.inet_recvmsg.__sys_recvfrom 10.13 -1.5 8.63 perf-profile.calltrace.cycles-pp._copy_from_iter_full.tcp_sendmsg_locked.tcp_sendmsg.sock_sendmsg.__sys_sendto 9.79 -1.4 8.36 perf-profile.calltrace.cycles-pp.copyin._copy_from_iter_full.tcp_sendmsg_locked.tcp_sendmsg.sock_sendmsg 2.14 -1.4 0.73 perf-profile.calltrace.cycles-pp.ttwu_do_activate.try_to_wake_up.__wake_up_common.__wake_up_common_lock.sock_def_readable 4.34 -1.4 2.96 perf-profile.calltrace.cycles-pp.ip_output.__ip_queue_xmit.__tcp_transmit_skb.tcp_recvmsg.inet_recvmsg 9.60 -1.4 8.23 perf-profile.calltrace.cycles-pp.copy_user_enhanced_fast_string.copyin._copy_from_iter_full.tcp_sendmsg_locked.tcp_sendmsg 2.10 ± 34% -1.4 0.73 ± 6% perf-profile.calltrace.cycles-pp.select_idle_sibling.select_task_rq_fair.try_to_wake_up.__wake_up_common.__wake_up_common_lock 2.08 -1.4 0.71 perf-profile.calltrace.cycles-pp.enqueue_task_fair.ttwu_do_activate.try_to_wake_up.__wake_up_common.__wake_up_common_lock 3.88 ± 2% -1.2 2.71 perf-profile.calltrace.cycles-pp.tcp_write_xmit.tcp_sendmsg_locked.tcp_sendmsg.sock_sendmsg.__sys_sendto 3.71 -1.2 2.55 perf-profile.calltrace.cycles-pp.ip_finish_output2.ip_output.__ip_queue_xmit.__tcp_transmit_skb.tcp_recvmsg 3.56 ± 2% -1.1 2.49 perf-profile.calltrace.cycles-pp.__tcp_transmit_skb.tcp_write_xmit.tcp_sendmsg_locked.tcp_sendmsg.sock_sendmsg 3.32 ± 2% -1.0 2.34 perf-profile.calltrace.cycles-pp.__ip_queue_xmit.__tcp_transmit_skb.tcp_write_xmit.tcp_sendmsg_locked.tcp_sendmsg 1.56 -0.9 0.61 ± 2% perf-profile.calltrace.cycles-pp.__dev_queue_xmit.ip_finish_output2.ip_output.__ip_queue_xmit.__tcp_transmit_skb 3.14 ± 2% -0.9 2.22 perf-profile.calltrace.cycles-pp.ip_output.__ip_queue_xmit.__tcp_transmit_skb.tcp_write_xmit.tcp_sendmsg_locked 1.77 -0.6 1.16 perf-profile.calltrace.cycles-pp.dequeue_task_fair.__sched_text_start.schedule.schedule_timeout.wait_woken 7.43 ± 2% -0.6 6.87 perf-profile.calltrace.cycles-pp.__tcp_push_pending_frames.tcp_sendmsg_locked.tcp_sendmsg.sock_sendmsg.__sys_sendto 7.41 ± 2% -0.6 6.85 perf-profile.calltrace.cycles-pp.tcp_write_xmit.__tcp_push_pending_frames.tcp_sendmsg_locked.tcp_sendmsg.sock_sendmsg 1.39 -0.5 0.87 perf-profile.calltrace.cycles-pp.exit_to_usermode_loop.do_syscall_64.entry_SYSCALL_64_after_hwframe 1.34 -0.5 0.84 perf-profile.calltrace.cycles-pp.schedule.exit_to_usermode_loop.do_syscall_64.entry_SYSCALL_64_after_hwframe 1.30 -0.5 0.81 perf-profile.calltrace.cycles-pp.__sched_text_start.schedule.exit_to_usermode_loop.do_syscall_64.entry_SYSCALL_64_after_hwframe 1.09 -0.4 0.72 perf-profile.calltrace.cycles-pp.__alloc_skb.sk_stream_alloc_skb.tcp_sendmsg_locked.tcp_sendmsg.sock_sendmsg 1.20 -0.4 0.84 perf-profile.calltrace.cycles-pp.__switch_to 1.09 -0.3 0.76 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64 6.67 ± 2% -0.3 6.38 perf-profile.calltrace.cycles-pp.__tcp_transmit_skb.tcp_write_xmit.__tcp_push_pending_frames.tcp_sendmsg_locked.tcp_sendmsg 0.83 -0.3 0.56 perf-profile.calltrace.cycles-pp.kmem_cache_alloc_node.__alloc_skb.sk_stream_alloc_skb.tcp_sendmsg_locked.tcp_sendmsg 0.84 -0.3 0.57 perf-profile.calltrace.cycles-pp.switch_mm_irqs_off.__sched_text_start.schedule.schedule_timeout.wait_woken 0.95 -0.3 0.69 perf-profile.calltrace.cycles-pp.syscall_return_via_sysret 0.75 ± 2% -0.2 0.52 perf-profile.calltrace.cycles-pp._cond_resched.__release_sock.release_sock.tcp_sendmsg.sock_sendmsg 0.73 ± 2% -0.2 0.51 perf-profile.calltrace.cycles-pp.preempt_schedule_common._cond_resched.__release_sock.release_sock.tcp_sendmsg 9.41 +1.1 10.47 perf-profile.calltrace.cycles-pp._raw_spin_lock.free_one_page.__free_pages_ok.skb_release_data.__kfree_skb 9.29 +1.1 10.39 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock.free_one_page.__free_pages_ok.skb_release_data 95.31 +1.4 96.67 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe 95.15 +1.4 96.56 perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe 11.02 +7.2 18.20 perf-profile.calltrace.cycles-pp.sk_stream_alloc_skb.tcp_sendmsg_locked.tcp_sendmsg.sock_sendmsg.__sys_sendto 12.28 ± 2% +7.2 19.51 perf-profile.calltrace.cycles-pp.__local_bh_enable_ip.ip_finish_output2.ip_output.__ip_queue_xmit.__tcp_transmit_skb 12.21 ± 2% +7.3 19.47 perf-profile.calltrace.cycles-pp.do_softirq.__local_bh_enable_ip.ip_finish_output2.ip_output.__ip_queue_xmit 12.00 ± 2% +7.3 19.34 perf-profile.calltrace.cycles-pp.do_softirq_own_stack.do_softirq.__local_bh_enable_ip.ip_finish_output2.ip_output 11.98 ± 2% +7.3 19.32 perf-profile.calltrace.cycles-pp.__softirqentry_text_start.do_softirq_own_stack.do_softirq.__local_bh_enable_ip.ip_finish_output2 11.74 ± 2% +7.4 19.14 perf-profile.calltrace.cycles-pp.net_rx_action.__softirqentry_text_start.do_softirq_own_stack.do_softirq.__local_bh_enable_ip 11.43 ± 2% +7.4 18.87 perf-profile.calltrace.cycles-pp.process_backlog.net_rx_action.__softirqentry_text_start.do_softirq_own_stack.do_softirq 11.37 ± 2% +7.5 18.83 perf-profile.calltrace.cycles-pp.ip_finish_output2.ip_output.__ip_queue_xmit.__tcp_transmit_skb.tcp_write_xmit 9.15 +7.5 16.67 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock.free_one_page.__free_pages_ok.___pskb_trim 9.26 +7.5 16.80 perf-profile.calltrace.cycles-pp._raw_spin_lock.free_one_page.__free_pages_ok.___pskb_trim.sk_stream_alloc_skb 9.43 +7.6 17.01 perf-profile.calltrace.cycles-pp.free_one_page.__free_pages_ok.___pskb_trim.sk_stream_alloc_skb.tcp_sendmsg_locked 9.79 +7.6 17.37 perf-profile.calltrace.cycles-pp.___pskb_trim.sk_stream_alloc_skb.tcp_sendmsg_locked.tcp_sendmsg.sock_sendmsg 10.97 ± 2% +7.6 18.56 perf-profile.calltrace.cycles-pp.__netif_receive_skb_one_core.process_backlog.net_rx_action.__softirqentry_text_start.do_softirq_own_stack 9.59 +7.6 17.22 perf-profile.calltrace.cycles-pp.__free_pages_ok.___pskb_trim.sk_stream_alloc_skb.tcp_sendmsg_locked.tcp_sendmsg 10.64 ± 3% +7.7 18.30 perf-profile.calltrace.cycles-pp.ip_rcv.__netif_receive_skb_one_core.process_backlog.net_rx_action.__softirqentry_text_start 10.11 ± 3% +7.8 17.94 perf-profile.calltrace.cycles-pp.ip_local_deliver.ip_rcv.__netif_receive_skb_one_core.process_backlog.net_rx_action 9.96 ± 3% +7.9 17.82 perf-profile.calltrace.cycles-pp.ip_local_deliver_finish.ip_local_deliver.ip_rcv.__netif_receive_skb_one_core.process_backlog 3.91 ± 2% +7.9 11.79 perf-profile.calltrace.cycles-pp.__tcp_push_pending_frames.tcp_rcv_established.tcp_v4_do_rcv.__release_sock.release_sock 3.91 ± 2% +7.9 11.79 perf-profile.calltrace.cycles-pp.tcp_write_xmit.__tcp_push_pending_frames.tcp_rcv_established.tcp_v4_do_rcv.__release_sock 9.90 ± 3% +7.9 17.79 perf-profile.calltrace.cycles-pp.ip_protocol_deliver_rcu.ip_local_deliver_finish.ip_local_deliver.ip_rcv.__netif_receive_skb_one_core 9.71 ± 3% +8.0 17.67 perf-profile.calltrace.cycles-pp.tcp_v4_rcv.ip_protocol_deliver_rcu.ip_local_deliver_finish.ip_local_deliver.ip_rcv 3.58 ± 2% +8.0 11.60 perf-profile.calltrace.cycles-pp.__tcp_transmit_skb.tcp_write_xmit.__tcp_push_pending_frames.tcp_rcv_established.tcp_v4_do_rcv 3.34 ± 2% +8.1 11.45 perf-profile.calltrace.cycles-pp.__ip_queue_xmit.__tcp_transmit_skb.tcp_write_xmit.__tcp_push_pending_frames.tcp_rcv_established 9.05 ± 2% +8.1 17.17 perf-profile.calltrace.cycles-pp.ip_output.__ip_queue_xmit.__tcp_transmit_skb.tcp_write_xmit.__tcp_push_pending_frames 18.39 +8.7 27.12 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.get_page_from_freelist.__alloc_pages_nodemask.skb_page_frag_refill 18.58 +8.7 27.31 perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.get_page_from_freelist.__alloc_pages_nodemask.skb_page_frag_refill.sk_page_frag_refill 19.93 +8.7 28.67 perf-profile.calltrace.cycles-pp.sk_page_frag_refill.tcp_sendmsg_locked.tcp_sendmsg.sock_sendmsg.__sys_sendto 19.92 +8.7 28.66 perf-profile.calltrace.cycles-pp.skb_page_frag_refill.sk_page_frag_refill.tcp_sendmsg_locked.tcp_sendmsg.sock_sendmsg 19.70 +8.7 28.45 perf-profile.calltrace.cycles-pp.get_page_from_freelist.__alloc_pages_nodemask.skb_page_frag_refill.sk_page_frag_refill.tcp_sendmsg_locked 19.77 +8.8 28.54 perf-profile.calltrace.cycles-pp.__alloc_pages_nodemask.skb_page_frag_refill.sk_page_frag_refill.tcp_sendmsg_locked.tcp_sendmsg 72.20 +8.8 81.00 perf-profile.calltrace.cycles-pp.__x64_sys_sendto.do_syscall_64.entry_SYSCALL_64_after_hwframe 72.09 +8.8 80.92 perf-profile.calltrace.cycles-pp.__sys_sendto.__x64_sys_sendto.do_syscall_64.entry_SYSCALL_64_after_hwframe 71.49 +9.0 80.48 perf-profile.calltrace.cycles-pp.sock_sendmsg.__sys_sendto.__x64_sys_sendto.do_syscall_64.entry_SYSCALL_64_after_hwframe 71.15 +9.1 80.24 perf-profile.calltrace.cycles-pp.tcp_sendmsg.sock_sendmsg.__sys_sendto.__x64_sys_sendto.do_syscall_64 0.00 +10.6 10.61 perf-profile.calltrace.cycles-pp.free_one_page.__free_pages_ok.skb_release_data.__kfree_skb.tcp_v4_rcv 0.00 +10.7 10.72 perf-profile.calltrace.cycles-pp.__free_pages_ok.skb_release_data.__kfree_skb.tcp_v4_rcv.ip_protocol_deliver_rcu 0.00 +11.0 10.98 perf-profile.calltrace.cycles-pp.skb_release_data.__kfree_skb.tcp_v4_rcv.ip_protocol_deliver_rcu.ip_local_deliver_finish 0.00 +11.1 11.11 perf-profile.calltrace.cycles-pp.__kfree_skb.tcp_v4_rcv.ip_protocol_deliver_rcu.ip_local_deliver_finish.ip_local_deliver 54.40 +12.2 66.63 perf-profile.calltrace.cycles-pp.tcp_sendmsg_locked.tcp_sendmsg.sock_sendmsg.__sys_sendto.__x64_sys_sendto 32407497 -26.7% 23748191 softirqs.CPU0.NET_RX 31936171 ± 3% -23.8% 24347695 softirqs.CPU1.NET_RX 31995720 ± 3% -23.5% 24474810 softirqs.CPU10.NET_RX 31851135 -28.4% 22807522 softirqs.CPU100.NET_RX 31165470 ± 2% -27.9% 22460444 ± 2% softirqs.CPU101.NET_RX 30427682 ± 3% -27.8% 21964060 ± 2% softirqs.CPU102.NET_RX 31822429 -30.3% 22190016 ± 2% softirqs.CPU103.NET_RX 31794681 ± 3% -24.0% 24170873 ± 2% softirqs.CPU11.NET_RX 31337588 ± 3% -25.0% 23487789 ± 2% softirqs.CPU12.NET_RX 30982384 ± 3% -21.6% 24285912 softirqs.CPU13.NET_RX 31985249 ± 2% -24.5% 24135878 ± 2% softirqs.CPU14.NET_RX 31569639 ± 3% -23.4% 24173977 ± 2% softirqs.CPU15.NET_RX 31672651 ± 2% -23.7% 24152941 ± 2% softirqs.CPU16.NET_RX 31845476 ± 2% -23.4% 24406051 softirqs.CPU17.NET_RX 32237888 -24.4% 24378972 softirqs.CPU18.NET_RX 32390163 -25.7% 24060765 softirqs.CPU19.NET_RX 32518154 -26.3% 23981309 ± 2% softirqs.CPU2.NET_RX 31625849 ± 2% -26.0% 23391632 ± 2% softirqs.CPU20.NET_RX 31800991 ± 2% -24.2% 24096267 ± 2% softirqs.CPU21.NET_RX 32165283 ± 2% -23.9% 24484882 softirqs.CPU22.NET_RX 32096347 ± 3% -24.6% 24198018 ± 2% softirqs.CPU23.NET_RX 32606868 -24.9% 24498919 softirqs.CPU24.NET_RX 31333059 ± 2% -23.5% 23972392 ± 2% softirqs.CPU25.NET_RX 29740104 ± 3% -23.9% 22620846 softirqs.CPU26.NET_RX 8933 ± 31% -24.6% 6736 ± 7% softirqs.CPU26.SCHED 31426927 -28.2% 22574373 ± 2% softirqs.CPU27.NET_RX 31494409 -27.6% 22811958 softirqs.CPU28.NET_RX 31749609 -29.8% 22274240 ± 2% softirqs.CPU29.NET_RX 31359195 ± 3% -23.3% 24058233 softirqs.CPU3.NET_RX 31891532 -28.5% 22801967 softirqs.CPU30.NET_RX 31022930 ± 3% -26.6% 22780850 softirqs.CPU31.NET_RX 31633941 -28.0% 22773366 softirqs.CPU32.NET_RX 31064582 ± 3% -28.2% 22291648 ± 2% softirqs.CPU33.NET_RX 30670424 ± 2% -26.0% 22706259 softirqs.CPU34.NET_RX 31155193 ± 3% -28.3% 22336447 ± 2% softirqs.CPU35.NET_RX 31767721 -29.3% 22462506 ± 2% softirqs.CPU36.NET_RX 31702765 -27.9% 22848963 softirqs.CPU37.NET_RX 31618979 -28.8% 22498908 ± 2% softirqs.CPU38.NET_RX 8157 ± 30% -22.3% 6342 softirqs.CPU38.SCHED 31178503 ± 2% -28.8% 22206116 ± 2% softirqs.CPU39.NET_RX 31543118 ± 3% -22.7% 24378208 softirqs.CPU4.NET_RX 29993167 ± 3% -26.6% 22006508 softirqs.CPU40.NET_RX 31389405 ± 2% -27.4% 22777491 softirqs.CPU41.NET_RX 31225889 ± 3% -28.1% 22466787 ± 2% softirqs.CPU42.NET_RX 31122849 ± 3% -27.5% 22548782 softirqs.CPU43.NET_RX 30449949 ± 3% -26.7% 22314973 ± 2% softirqs.CPU44.NET_RX 30708016 ± 3% -26.5% 22575709 softirqs.CPU45.NET_RX 30869514 ± 2% -27.1% 22517203 softirqs.CPU46.NET_RX 30349340 ± 3% -26.7% 22238899 ± 2% softirqs.CPU47.NET_RX 31690538 -27.9% 22844885 softirqs.CPU48.NET_RX 31442996 -27.6% 22774395 softirqs.CPU49.NET_RX 32539333 -24.8% 24475876 softirqs.CPU5.NET_RX 31777443 -28.1% 22839894 softirqs.CPU50.NET_RX 30319039 ± 4% -25.8% 22488386 softirqs.CPU51.NET_RX 32358234 -24.7% 24364666 softirqs.CPU52.NET_RX 31051127 ± 3% -23.5% 23764585 ± 2% softirqs.CPU53.NET_RX 31368731 ± 3% -22.3% 24384593 softirqs.CPU54.NET_RX 32739188 -26.5% 24063678 ± 2% softirqs.CPU55.NET_RX 32587893 -25.0% 24425586 softirqs.CPU56.NET_RX 31762403 ± 3% -24.6% 23947521 softirqs.CPU57.NET_RX 32342624 -25.5% 24095598 softirqs.CPU58.NET_RX 31850813 ± 3% -23.7% 24310413 softirqs.CPU59.NET_RX 32296552 -26.1% 23882250 ± 2% softirqs.CPU6.NET_RX 31208970 ± 3% -23.9% 23735529 ± 2% softirqs.CPU60.NET_RX 32655061 -25.7% 24259328 softirqs.CPU61.NET_RX 31834306 ± 3% -23.2% 24436288 softirqs.CPU62.NET_RX 32492263 -25.7% 24146807 ± 2% softirqs.CPU63.NET_RX 32160041 -24.6% 24237812 softirqs.CPU64.NET_RX 32146792 -25.1% 24083142 softirqs.CPU65.NET_RX 31886509 ± 3% -24.3% 24153287 ± 2% softirqs.CPU66.NET_RX 31124000 ± 2% -21.2% 24534615 softirqs.CPU67.NET_RX 31606731 ± 3% -23.7% 24115056 ± 2% softirqs.CPU68.NET_RX 31809649 ± 4% -27.3% 23140437 softirqs.CPU69.NET_RX 31832502 ± 3% -23.8% 24256886 softirqs.CPU7.NET_RX 31934664 ± 2% -23.5% 24445503 softirqs.CPU70.NET_RX 32283512 -24.5% 24374573 softirqs.CPU71.NET_RX 31559429 ± 3% -23.0% 24304361 softirqs.CPU72.NET_RX 32519181 -26.5% 23915099 ± 2% softirqs.CPU73.NET_RX 32064639 ± 3% -23.5% 24531236 softirqs.CPU74.NET_RX 32552635 -25.2% 24358616 softirqs.CPU75.NET_RX 32574031 -25.1% 24391309 softirqs.CPU76.NET_RX 29891561 -18.3% 24435768 softirqs.CPU77.NET_RX 30558797 ± 3% -27.1% 22290753 ± 2% softirqs.CPU78.NET_RX 30913524 ± 3% -27.1% 22542484 softirqs.CPU79.NET_RX 32338939 -25.2% 24184482 softirqs.CPU8.NET_RX 29930689 ± 3% -24.3% 22649621 softirqs.CPU80.NET_RX 31789053 -29.3% 22471154 ± 2% softirqs.CPU81.NET_RX 31080841 ± 3% -28.0% 22371709 ± 2% softirqs.CPU82.NET_RX 31004904 ± 3% -28.0% 22335427 softirqs.CPU83.NET_RX 30902184 ± 2% -26.4% 22746371 softirqs.CPU84.NET_RX 31508332 -28.4% 22567097 softirqs.CPU85.NET_RX 31437975 -29.0% 22331821 ± 2% softirqs.CPU86.NET_RX 31025912 ± 3% -26.1% 22913093 softirqs.CPU87.NET_RX 30270422 ± 2% -25.2% 22636977 ± 2% softirqs.CPU88.NET_RX 30385565 ± 4% -26.0% 22499521 ± 2% softirqs.CPU89.NET_RX 31577940 ± 3% -22.4% 24489124 softirqs.CPU9.NET_RX 30152844 ± 3% -24.7% 22717462 softirqs.CPU90.NET_RX 29900183 ± 2% -24.9% 22460973 softirqs.CPU91.NET_RX 31952707 -28.9% 22715130 softirqs.CPU92.NET_RX 30820437 ± 3% -26.6% 22619907 ± 2% softirqs.CPU93.NET_RX 30627580 ± 3% -25.6% 22790846 softirqs.CPU94.NET_RX 31367872 -27.7% 22670545 softirqs.CPU95.NET_RX 31045007 ± 3% -26.5% 22818368 softirqs.CPU96.NET_RX 30842779 ± 3% -28.2% 22143073 ± 2% softirqs.CPU97.NET_RX 30173663 ± 3% -24.8% 22686608 softirqs.CPU98.NET_RX 30750855 ± 3% -25.6% 22870339 softirqs.CPU99.NET_RX 3.273e+09 -25.8% 2.43e+09 softirqs.NET_RX netperf.Throughput_Mbps 4000 +-+------------------------------------------------------------------+ | | 3500 +-+ | 3000 +-+ | O O O OO OOO OO OO OOO OO OO OO OOO OO OO OO | 2500 +-+ | | | 2000 +-+ | | | 1500 +-+ | 1000 +-+ | | | 500 +-+ | | | 0 +O+O---O-------------------------------------------------------------+ netperf.Throughput_total_Mbps 900000 +-+----------------------------------------------------------------+ |+.++.+++.++.+++.++.+++.++.+++.++.+++.+++.++.+++.++.+++.++.+++.++.+| 800000 +-+ | 700000 +-+ | | O | 600000 O-+ O O O OO OOO OO OOO OO OOO O OOO OOO OO O | 500000 +-+ | | | 400000 +-+ | 300000 +-+ | | | 200000 +-+ | 100000 +-+ | | | 0 +O+O---O-----------------------------------------------------------+ netperf.workload 6e+09 +-+-----------------------------------------------------------------+ |+.++.++.+++.++.+++.++.++.+++.++.+++.++.++.+++.++.+++.++.++.+++.++.+| 5e+09 +-+ | | | O O O OOO OO OOO OO OO OOO OO OOO OO OO OOO | 4e+09 +-+ | | | 3e+09 +-+ | | | 2e+09 +-+ | | | | | 1e+09 +-+ | | | 0 +O+O---O------------------------------------------------------------+ netperf.time.user_time 2000 +-+------------------------------------------------------------------+ 1800 +-+ + + | | : : + : | 1600 +-+ +. + .++.++. + : +.+ .++. .++.++. + .++ + .++.+| 1400 +-+++.++.+ + + + + + +++ + +.++.++ +.++ | | | 1200 +-+ | 1000 O-+ O OO OOO OO OO OOO OO OO OO OOO OO OO OO | 800 +-+ | | | 600 +-+ | 400 +-+ | | | 200 +-+ | 0 +O+O--OO-------------------------------------------------------------+ netperf.time.percent_of_cpu_this_job_got 8000 +-+------------------------------------------------------------------+ O+.+O.++.OO.OOO.OO.OO.OOO.OO.OO.OO.OOO.OO.OO.OO+.++.++.++.+++.++.++.+| 7000 +-+ | 6000 +-+ | | | 5000 +-+ | | | 4000 +-+ | | | 3000 +-+ | 2000 +-+ | | | 1000 +-+ | | | 0 +O+O--OO-------------------------------------------------------------+ netperf.time.involuntary_context_switches 4e+09 +-+---------------------------------------------------------------+ |+.++.+++.+++.++.+++.+++.++.+++.+++.++.+++.+++.++.+++.+++.++.+++.+| 3.5e+09 +-+ | 3e+09 +-+ | O O O OOO OO OOO OOO OO OOO OOO OO OOO OO | 2.5e+09 +-+ | | | 2e+09 +-+ | | | 1.5e+09 +-+ | 1e+09 +-+ | | | 5e+08 +-+ | | | 0 +O+O--OO----------------------------------------------------------+ [*] bisect-good sample [O] bisect-bad sample *************************************************************************************************** lkp-skl-fpga01: 104 threads Skylake with 192G memory ========================================================================================= cluster/compiler/cpufreq_governor/ip/kconfig/nr_threads/rootfs/runtime/tbox_group/test/testcase: cs-localhost/gcc-7/performance/ipv4/x86_64-rhel-7.6/200%/debian-x86_64-2018-04-03.cgz/900s/lkp-skl-fpga01/TCP_MAERTS/netperf commit: 472c2e07ee ("tcp: add one skb cache for tx") 8b27dae5a2 ("tcp: add one skb cache for rx") 472c2e07eef04514 8b27dae5a2e89a61c46c6dbc76c ---------------- --------------------------- %stddev %change %stddev \ | \ 3933 -25.6% 2926 netperf.Throughput_Mbps 818124 -25.6% 608758 netperf.Throughput_total_Mbps 57423 ± 19% -41.1% 33817 netperf.time.involuntary_context_switches 3010 +3.7% 3122 netperf.time.percent_of_cpu_this_job_got 25432 +3.9% 26424 netperf.time.system_time 3.735e+09 -25.6% 2.778e+09 netperf.time.voluntary_context_switches 5.618e+09 -25.6% 4.18e+09 netperf.workload 194721 ± 22% +41.3% 275069 ± 2% cpuidle.C6.usage 13.00 ± 7% +5.9 18.86 mpstat.cpu.all.soft% 3.39 ± 5% -1.2 2.23 mpstat.cpu.all.usr% 9.539e+08 +13.2% 1.08e+09 numa-numastat.node0.local_node 9.539e+08 +13.2% 1.08e+09 numa-numastat.node0.numa_hit 4.767e+08 +13.1% 5.389e+08 numa-vmstat.node0.numa_hit 4.767e+08 +13.1% 5.389e+08 numa-vmstat.node0.numa_local 95.75 +1.3% 97.00 vmstat.cpu.sy 8245666 -25.5% 6142829 vmstat.system.cs 2696 +2.9% 2775 turbostat.Avg_MHz 0.89 ± 5% -32.4% 0.60 turbostat.CPU%c1 108.00 -1.6% 106.23 turbostat.RAMWatt 1.881e+09 +11.2% 2.091e+09 proc-vmstat.numa_hit 1.881e+09 +11.2% 2.091e+09 proc-vmstat.numa_local 1.504e+10 +11.2% 1.672e+10 proc-vmstat.pgalloc_normal 1.504e+10 +11.2% 1.672e+10 proc-vmstat.pgfree 4053 ± 15% +19.6% 4848 ± 13% slabinfo.eventpoll_pwq.active_objs 4053 ± 15% +19.6% 4848 ± 13% slabinfo.eventpoll_pwq.num_objs 545.50 ± 11% -17.1% 452.00 ± 11% slabinfo.kernfs_iattrs_cache.active_objs 545.50 ± 11% -17.1% 452.00 ± 11% slabinfo.kernfs_iattrs_cache.num_objs 12.31 ± 12% +25.1% 15.40 sched_debug.cfs_rq:/.load_avg.stddev 1449447 ± 23% -29.2% 1026327 ± 22% sched_debug.cfs_rq:/.min_vruntime.stddev 0.78 ± 66% +155.6% 2.00 ± 11% sched_debug.cfs_rq:/.removed.load_avg.avg 5.98 ± 60% +84.9% 11.05 ± 4% sched_debug.cfs_rq:/.removed.load_avg.stddev 35.94 ± 66% +156.2% 92.09 ± 11% sched_debug.cfs_rq:/.removed.runnable_sum.avg 274.98 ± 60% +85.4% 509.82 ± 4% sched_debug.cfs_rq:/.removed.runnable_sum.stddev 0.39 ± 66% +115.0% 0.84 ± 11% sched_debug.cfs_rq:/.removed.util_avg.avg 1.75 ± 14% +32.1% 2.31 ± 12% sched_debug.cfs_rq:/.runnable_load_avg.min 98724 ± 85% -66.3% 33282 ± 85% sched_debug.cfs_rq:/.runnable_weight.max -3346961 -44.2% -1868641 sched_debug.cfs_rq:/.spread0.min 1449347 ± 23% -29.2% 1026220 ± 22% sched_debug.cfs_rq:/.spread0.stddev 156728 ± 15% +49.0% 233530 ± 14% sched_debug.cpu.avg_idle.avg 260031 ± 9% +33.3% 346707 ± 10% sched_debug.cpu.avg_idle.stddev 20.54 ± 5% -10.2% 18.44 ± 2% sched_debug.cpu.clock.stddev 20.54 ± 5% -10.2% 18.44 ± 2% sched_debug.cpu.clock_task.stddev 35742943 -25.6% 26602562 sched_debug.cpu.nr_switches.avg 37162589 -24.6% 28014480 sched_debug.cpu.nr_switches.max 32371186 -24.3% 24512619 sched_debug.cpu.nr_switches.min 18.74 +29.7% 24.31 perf-stat.i.MPKI 1.938e+10 -20.9% 1.534e+10 perf-stat.i.branch-instructions 2.05 -0.1 1.98 perf-stat.i.branch-miss-rate% 3.961e+08 -23.4% 3.033e+08 perf-stat.i.branch-misses 0.74 ± 31% -0.4 0.38 ± 18% perf-stat.i.cache-miss-rate% 12929011 ± 32% -49.1% 6575616 ± 19% perf-stat.i.cache-misses 1.813e+09 +2.3% 1.854e+09 perf-stat.i.cache-references 8267879 -25.5% 6157361 perf-stat.i.context-switches 2.89 +30.6% 3.77 perf-stat.i.cpi 2.794e+11 +2.9% 2.876e+11 perf-stat.i.cpu-cycles 23942 ± 24% +97.5% 47291 ± 21% perf-stat.i.cycles-between-cache-misses 1.13 -0.1 0.99 perf-stat.i.dTLB-load-miss-rate% 3.331e+08 -31.8% 2.274e+08 perf-stat.i.dTLB-load-misses 2.9e+10 -22.0% 2.263e+10 perf-stat.i.dTLB-loads 60002134 ± 2% -26.3% 44193691 ± 2% perf-stat.i.dTLB-store-misses 1.664e+10 -23.6% 1.271e+10 perf-stat.i.dTLB-stores 58.26 -4.7 53.57 perf-stat.i.iTLB-load-miss-rate% 70291090 -31.0% 48469793 perf-stat.i.iTLB-load-misses 50169394 -16.5% 41890643 perf-stat.i.iTLB-loads 9.677e+10 -21.2% 7.624e+10 perf-stat.i.instructions 1410 +13.6% 1602 perf-stat.i.instructions-per-iTLB-miss 0.35 -23.4% 0.27 perf-stat.i.ipc 2150902 ± 41% -42.7% 1233268 ± 27% perf-stat.i.node-load-misses 652857 ± 24% -39.8% 393122 ± 15% perf-stat.i.node-loads 77.08 ± 11% -20.6 56.48 ± 14% perf-stat.i.node-store-miss-rate% 386848 ± 40% -64.7% 136565 ± 27% perf-stat.i.node-store-misses 18.74 +29.8% 24.32 perf-stat.overall.MPKI 2.04 -0.1 1.98 perf-stat.overall.branch-miss-rate% 0.71 ± 32% -0.4 0.35 ± 20% perf-stat.overall.cache-miss-rate% 2.89 +30.6% 3.77 perf-stat.overall.cpi 23413 ± 23% +95.0% 45644 ± 21% perf-stat.overall.cycles-between-cache-misses 1.14 -0.1 0.99 perf-stat.overall.dTLB-load-miss-rate% 58.35 -4.7 53.64 perf-stat.overall.iTLB-load-miss-rate% 1376 +14.3% 1573 perf-stat.overall.instructions-per-iTLB-miss 0.35 -23.4% 0.27 perf-stat.overall.ipc 75.96 ± 13% -20.1 55.89 ± 14% perf-stat.overall.node-store-miss-rate% 15545 +5.9% 16457 perf-stat.overall.path-length 1.936e+10 -20.9% 1.532e+10 perf-stat.ps.branch-instructions 3.956e+08 -23.4% 3.03e+08 perf-stat.ps.branch-misses 12917167 ± 32% -49.1% 6568704 ± 19% perf-stat.ps.cache-misses 1.811e+09 +2.3% 1.852e+09 perf-stat.ps.cache-references 8258274 -25.5% 6150289 perf-stat.ps.context-switches 2.791e+11 +2.9% 2.872e+11 perf-stat.ps.cpu-cycles 3.328e+08 -31.8% 2.271e+08 perf-stat.ps.dTLB-load-misses 2.897e+10 -22.0% 2.26e+10 perf-stat.ps.dTLB-loads 59931265 ± 2% -26.3% 44142618 ± 2% perf-stat.ps.dTLB-store-misses 1.662e+10 -23.6% 1.269e+10 perf-stat.ps.dTLB-stores 70208114 -31.0% 48413951 perf-stat.ps.iTLB-load-misses 50111323 -16.5% 41842519 perf-stat.ps.iTLB-loads 9.666e+10 -21.2% 7.616e+10 perf-stat.ps.instructions 2148559 ± 41% -42.7% 1231893 ± 27% perf-stat.ps.node-load-misses 653248 ± 24% -39.9% 392846 ± 15% perf-stat.ps.node-loads 386418 ± 40% -64.7% 136417 ± 27% perf-stat.ps.node-store-misses 8.733e+13 -21.2% 6.879e+13 perf-stat.total.instructions 887.50 ± 21% +961.1% 9417 ±122% interrupts.39:PCI-MSI.67633154-edge.eth0-TxRx-1 906772 +4.4% 946334 interrupts.CAL:Function_call_interrupts 128451 ± 3% -20.1% 102684 ± 11% interrupts.CPU1.RES:Rescheduling_interrupts 122297 ± 12% -20.1% 97767 ± 5% interrupts.CPU100.RES:Rescheduling_interrupts 114996 ± 8% -12.8% 100315 ± 5% interrupts.CPU101.RES:Rescheduling_interrupts 8676 ± 2% +14.9% 9972 ± 3% interrupts.CPU102.CAL:Function_call_interrupts 145573 ± 15% -30.4% 101298 ± 9% interrupts.CPU102.RES:Rescheduling_interrupts 8795 +13.2% 9956 ± 3% interrupts.CPU103.CAL:Function_call_interrupts 129834 ± 7% -15.4% 109820 ± 9% interrupts.CPU11.RES:Rescheduling_interrupts 122266 ± 5% -18.3% 99893 ± 4% interrupts.CPU13.RES:Rescheduling_interrupts 133695 ± 4% -23.7% 101984 ± 5% interrupts.CPU14.RES:Rescheduling_interrupts 115773 ± 7% -18.6% 94202 ± 8% interrupts.CPU18.RES:Rescheduling_interrupts 120005 ± 3% -15.7% 101149 ± 8% interrupts.CPU27.RES:Rescheduling_interrupts 119584 ± 3% -19.5% 96286 ± 6% interrupts.CPU29.RES:Rescheduling_interrupts 127064 ± 7% -15.1% 107858 ± 5% interrupts.CPU30.RES:Rescheduling_interrupts 887.50 ± 21% +961.1% 9417 ±122% interrupts.CPU31.39:PCI-MSI.67633154-edge.eth0-TxRx-1 118974 ± 11% -17.1% 98609 ± 7% interrupts.CPU33.RES:Rescheduling_interrupts 125463 ± 13% -23.9% 95477 ± 5% interrupts.CPU34.RES:Rescheduling_interrupts 125126 ± 16% -21.7% 97997 ± 7% interrupts.CPU35.RES:Rescheduling_interrupts 133035 ± 12% -28.2% 95517 ± 10% interrupts.CPU37.RES:Rescheduling_interrupts 120167 ± 11% -24.5% 90782 ± 8% interrupts.CPU38.RES:Rescheduling_interrupts 125040 ± 3% -19.8% 100223 ± 7% interrupts.CPU42.RES:Rescheduling_interrupts 119768 ± 15% -16.6% 99882 interrupts.CPU43.RES:Rescheduling_interrupts 125986 ± 7% -18.9% 102226 ± 8% interrupts.CPU45.RES:Rescheduling_interrupts 124516 ± 10% -15.7% 104961 ± 5% interrupts.CPU48.RES:Rescheduling_interrupts 125647 ± 5% -22.8% 96970 ± 3% interrupts.CPU49.RES:Rescheduling_interrupts 131345 ± 5% -17.2% 108796 ± 3% interrupts.CPU51.RES:Rescheduling_interrupts 127858 ± 3% -18.9% 103683 ± 5% interrupts.CPU53.RES:Rescheduling_interrupts 125241 ± 15% -18.7% 101797 ± 5% interrupts.CPU54.RES:Rescheduling_interrupts 133398 ± 7% -22.1% 103961 ± 3% interrupts.CPU56.RES:Rescheduling_interrupts 129732 ± 10% -24.2% 98324 ± 4% interrupts.CPU6.RES:Rescheduling_interrupts 132858 ± 12% -20.5% 105623 ± 9% interrupts.CPU60.RES:Rescheduling_interrupts 123442 ± 5% -20.7% 97934 ± 4% interrupts.CPU64.RES:Rescheduling_interrupts 132863 ± 3% -20.2% 106040 ± 4% interrupts.CPU65.RES:Rescheduling_interrupts 9002 +10.7% 9967 ± 2% interrupts.CPU66.CAL:Function_call_interrupts 8999 +10.7% 9958 ± 2% interrupts.CPU67.CAL:Function_call_interrupts 8774 ± 5% +13.4% 9951 ± 2% interrupts.CPU68.CAL:Function_call_interrupts 8991 +10.6% 9945 ± 2% interrupts.CPU69.CAL:Function_call_interrupts 125949 ± 9% -15.0% 107048 ± 4% interrupts.CPU69.RES:Rescheduling_interrupts 123443 ± 8% -15.7% 104008 ± 8% interrupts.CPU7.RES:Rescheduling_interrupts 8987 +10.6% 9938 ± 2% interrupts.CPU70.CAL:Function_call_interrupts 8983 +10.5% 9927 ± 2% interrupts.CPU71.CAL:Function_call_interrupts 122925 ± 12% -18.0% 100786 ± 8% interrupts.CPU71.RES:Rescheduling_interrupts 8925 +11.0% 9910 ± 2% interrupts.CPU73.CAL:Function_call_interrupts 126258 ± 7% -19.4% 101773 ± 7% interrupts.CPU73.RES:Rescheduling_interrupts 129281 ± 6% -18.1% 105945 ± 5% interrupts.CPU74.RES:Rescheduling_interrupts 8969 +10.3% 9893 ± 2% interrupts.CPU75.CAL:Function_call_interrupts 8966 +10.3% 9887 ± 2% interrupts.CPU76.CAL:Function_call_interrupts 8965 +10.1% 9870 ± 2% interrupts.CPU77.CAL:Function_call_interrupts 129057 ± 9% -24.7% 97240 ± 3% interrupts.CPU77.RES:Rescheduling_interrupts 8683 ± 5% +13.7% 9874 ± 2% interrupts.CPU78.CAL:Function_call_interrupts 129114 ± 9% -27.6% 93483 ± 4% interrupts.CPU78.RES:Rescheduling_interrupts 8685 ± 4% +13.6% 9865 ± 2% interrupts.CPU79.CAL:Function_call_interrupts 121386 ± 7% -18.7% 98627 ± 5% interrupts.CPU8.RES:Rescheduling_interrupts 8956 +10.1% 9864 ± 2% interrupts.CPU80.CAL:Function_call_interrupts 125015 ± 11% -24.1% 94918 ± 5% interrupts.CPU80.RES:Rescheduling_interrupts 8633 ± 7% +14.2% 9856 ± 2% interrupts.CPU81.CAL:Function_call_interrupts 132492 ± 4% -20.7% 105014 ± 4% interrupts.CPU81.RES:Rescheduling_interrupts 8857 ± 2% +11.2% 9848 ± 2% interrupts.CPU82.CAL:Function_call_interrupts 111387 ± 6% -13.8% 96026 ± 8% interrupts.CPU82.RES:Rescheduling_interrupts 112445 ± 6% -9.6% 101699 ± 7% interrupts.CPU83.RES:Rescheduling_interrupts 121973 ± 7% -21.1% 96217 ± 6% interrupts.CPU84.RES:Rescheduling_interrupts 8883 +10.6% 9822 ± 2% interrupts.CPU85.CAL:Function_call_interrupts 115045 ± 6% -16.3% 96249 ± 4% interrupts.CPU85.RES:Rescheduling_interrupts 8668 ± 3% +12.6% 9762 interrupts.CPU86.CAL:Function_call_interrupts 8579 ± 3% +14.3% 9806 ± 2% interrupts.CPU87.CAL:Function_call_interrupts 8719 ± 2% +12.4% 9797 ± 2% interrupts.CPU88.CAL:Function_call_interrupts 8769 +11.6% 9788 ± 2% interrupts.CPU89.CAL:Function_call_interrupts 131248 ± 9% -27.5% 95130 ± 9% interrupts.CPU9.RES:Rescheduling_interrupts 8766 +11.6% 9780 ± 2% interrupts.CPU90.CAL:Function_call_interrupts 121573 ± 6% -14.4% 104044 ± 8% interrupts.CPU90.RES:Rescheduling_interrupts 8762 +11.5% 9768 ± 2% interrupts.CPU91.CAL:Function_call_interrupts 125741 ± 6% -21.2% 99116 ± 10% interrupts.CPU91.RES:Rescheduling_interrupts 8764 +11.3% 9754 ± 2% interrupts.CPU92.CAL:Function_call_interrupts 125340 ± 7% -21.1% 98921 ± 6% interrupts.CPU92.RES:Rescheduling_interrupts 8620 ± 4% +13.0% 9743 ± 2% interrupts.CPU93.CAL:Function_call_interrupts 129358 ± 10% -25.0% 97082 ± 6% interrupts.CPU93.RES:Rescheduling_interrupts 8753 +11.2% 9730 ± 2% interrupts.CPU94.CAL:Function_call_interrupts 8750 +11.1% 9719 ± 2% interrupts.CPU95.CAL:Function_call_interrupts 115441 ± 6% -15.6% 97484 ± 8% interrupts.CPU95.RES:Rescheduling_interrupts 121213 ± 5% -17.0% 100662 ± 12% interrupts.CPU99.RES:Rescheduling_interrupts 12650754 ± 3% -14.8% 10783909 interrupts.RES:Rescheduling_interrupts 176.50 ± 59% -79.7% 35.75 ± 47% interrupts.TLB:TLB_shootdowns 10.98 -11.0 0.00 perf-profile.calltrace.cycles-pp.tcp_ack.tcp_rcv_established.tcp_v4_do_rcv.__release_sock.release_sock 10.70 -10.7 0.00 perf-profile.calltrace.cycles-pp.tcp_clean_rtx_queue.tcp_ack.tcp_rcv_established.tcp_v4_do_rcv.__release_sock 10.29 -10.3 0.00 perf-profile.calltrace.cycles-pp.__kfree_skb.tcp_clean_rtx_queue.tcp_ack.tcp_rcv_established.tcp_v4_do_rcv 10.20 -10.2 0.00 perf-profile.calltrace.cycles-pp.skb_release_data.__kfree_skb.tcp_clean_rtx_queue.tcp_ack.tcp_rcv_established 9.82 -9.8 0.00 perf-profile.calltrace.cycles-pp.__free_pages_ok.skb_release_data.__kfree_skb.tcp_clean_rtx_queue.tcp_ack 9.68 -9.7 0.00 perf-profile.calltrace.cycles-pp.free_one_page.__free_pages_ok.skb_release_data.__kfree_skb.tcp_clean_rtx_queue 21.27 -6.8 14.52 perf-profile.calltrace.cycles-pp.__x64_sys_recvfrom.do_syscall_64.entry_SYSCALL_64_after_hwframe 21.18 -6.7 14.46 perf-profile.calltrace.cycles-pp.__sys_recvfrom.__x64_sys_recvfrom.do_syscall_64.entry_SYSCALL_64_after_hwframe 20.54 -6.5 14.04 perf-profile.calltrace.cycles-pp.inet_recvmsg.__sys_recvfrom.__x64_sys_recvfrom.do_syscall_64.entry_SYSCALL_64_after_hwframe 20.43 -6.5 13.96 perf-profile.calltrace.cycles-pp.tcp_recvmsg.inet_recvmsg.__sys_recvfrom.__x64_sys_recvfrom.do_syscall_64 16.41 -3.1 13.34 perf-profile.calltrace.cycles-pp.release_sock.tcp_sendmsg.sock_sendmsg.__sys_sendto.__x64_sys_sendto 16.22 -3.0 13.21 perf-profile.calltrace.cycles-pp.__release_sock.release_sock.tcp_sendmsg.sock_sendmsg.__sys_sendto 7.69 -2.2 5.51 perf-profile.calltrace.cycles-pp.skb_copy_datagram_iter.tcp_recvmsg.inet_recvmsg.__sys_recvfrom.__x64_sys_recvfrom 7.67 -2.2 5.50 perf-profile.calltrace.cycles-pp.__skb_datagram_iter.skb_copy_datagram_iter.tcp_recvmsg.inet_recvmsg.__sys_recvfrom 6.75 ± 6% -2.2 4.60 ± 2% perf-profile.calltrace.cycles-pp.tcp_v4_do_rcv.tcp_v4_rcv.ip_protocol_deliver_rcu.ip_local_deliver_finish.ip_local_deliver 13.64 -2.1 11.53 perf-profile.calltrace.cycles-pp.tcp_v4_do_rcv.__release_sock.release_sock.tcp_sendmsg.sock_sendmsg 13.43 -2.0 11.40 perf-profile.calltrace.cycles-pp.tcp_rcv_established.tcp_v4_do_rcv.__release_sock.release_sock.tcp_sendmsg 7.00 -2.0 5.04 perf-profile.calltrace.cycles-pp._copy_to_iter.__skb_datagram_iter.skb_copy_datagram_iter.tcp_recvmsg.inet_recvmsg 6.83 -1.9 4.93 perf-profile.calltrace.cycles-pp.copyout._copy_to_iter.__skb_datagram_iter.skb_copy_datagram_iter.tcp_recvmsg 6.70 -1.9 4.83 perf-profile.calltrace.cycles-pp.copy_user_enhanced_fast_string.copyout._copy_to_iter.__skb_datagram_iter.skb_copy_datagram_iter 6.39 ± 3% -1.9 4.53 perf-profile.calltrace.cycles-pp.tcp_rcv_established.tcp_v4_do_rcv.tcp_v4_rcv.ip_protocol_deliver_rcu.ip_local_deliver_finish 5.43 -1.7 3.69 perf-profile.calltrace.cycles-pp.sk_wait_data.tcp_recvmsg.inet_recvmsg.__sys_recvfrom.__x64_sys_recvfrom 5.31 -1.7 3.59 perf-profile.calltrace.cycles-pp.__tcp_transmit_skb.tcp_recvmsg.inet_recvmsg.__sys_recvfrom.__x64_sys_recvfrom 5.68 ± 3% -1.6 4.11 ± 2% perf-profile.calltrace.cycles-pp.sock_def_readable.tcp_rcv_established.tcp_v4_do_rcv.tcp_v4_rcv.ip_protocol_deliver_rcu 4.76 -1.6 3.19 perf-profile.calltrace.cycles-pp.wait_woken.sk_wait_data.tcp_recvmsg.inet_recvmsg.__sys_recvfrom 4.84 -1.5 3.30 perf-profile.calltrace.cycles-pp.__ip_queue_xmit.__tcp_transmit_skb.tcp_recvmsg.inet_recvmsg.__sys_recvfrom 4.56 -1.5 3.04 perf-profile.calltrace.cycles-pp.schedule_timeout.wait_woken.sk_wait_data.tcp_recvmsg.inet_recvmsg 5.51 ± 3% -1.5 4.01 ± 2% perf-profile.calltrace.cycles-pp.__wake_up_common_lock.sock_def_readable.tcp_rcv_established.tcp_v4_do_rcv.tcp_v4_rcv 4.48 -1.5 2.99 perf-profile.calltrace.cycles-pp.schedule.schedule_timeout.wait_woken.sk_wait_data.tcp_recvmsg 4.39 -1.5 2.92 perf-profile.calltrace.cycles-pp.__sched_text_start.schedule.schedule_timeout.wait_woken.sk_wait_data 5.36 ± 3% -1.5 3.90 ± 2% perf-profile.calltrace.cycles-pp.__wake_up_common.__wake_up_common_lock.sock_def_readable.tcp_rcv_established.tcp_v4_do_rcv 5.19 ± 3% -1.4 3.78 ± 2% perf-profile.calltrace.cycles-pp.try_to_wake_up.__wake_up_common.__wake_up_common_lock.sock_def_readable.tcp_rcv_established 4.37 -1.4 2.96 perf-profile.calltrace.cycles-pp.ip_output.__ip_queue_xmit.__tcp_transmit_skb.tcp_recvmsg.inet_recvmsg 2.14 -1.4 0.73 perf-profile.calltrace.cycles-pp.ttwu_do_activate.try_to_wake_up.__wake_up_common.__wake_up_common_lock.sock_def_readable 10.04 -1.4 8.65 perf-profile.calltrace.cycles-pp._copy_from_iter_full.tcp_sendmsg_locked.tcp_sendmsg.sock_sendmsg.__sys_sendto 2.08 -1.4 0.70 perf-profile.calltrace.cycles-pp.enqueue_task_fair.ttwu_do_activate.try_to_wake_up.__wake_up_common.__wake_up_common_lock 9.73 -1.3 8.42 perf-profile.calltrace.cycles-pp.copyin._copy_from_iter_full.tcp_sendmsg_locked.tcp_sendmsg.sock_sendmsg 9.56 -1.3 8.29 perf-profile.calltrace.cycles-pp.copy_user_enhanced_fast_string.copyin._copy_from_iter_full.tcp_sendmsg_locked.tcp_sendmsg 3.73 -1.2 2.55 perf-profile.calltrace.cycles-pp.ip_finish_output2.ip_output.__ip_queue_xmit.__tcp_transmit_skb.tcp_recvmsg 3.78 ± 2% -1.1 2.68 perf-profile.calltrace.cycles-pp.tcp_write_xmit.tcp_sendmsg_locked.tcp_sendmsg.sock_sendmsg.__sys_sendto 1.86 ± 30% -1.0 0.84 ± 4% perf-profile.calltrace.cycles-pp.select_task_rq_fair.try_to_wake_up.__wake_up_common.__wake_up_common_lock.sock_def_readable 3.46 ± 2% -1.0 2.46 perf-profile.calltrace.cycles-pp.__tcp_transmit_skb.tcp_write_xmit.tcp_sendmsg_locked.tcp_sendmsg.sock_sendmsg 1.51 ± 6% -0.9 0.61 perf-profile.calltrace.cycles-pp.__dev_queue_xmit.ip_finish_output2.ip_output.__ip_queue_xmit.__tcp_transmit_skb 3.22 ± 2% -0.9 2.33 perf-profile.calltrace.cycles-pp.__ip_queue_xmit.__tcp_transmit_skb.tcp_write_xmit.tcp_sendmsg_locked.tcp_sendmsg 3.02 ± 2% -0.8 2.21 perf-profile.calltrace.cycles-pp.ip_output.__ip_queue_xmit.__tcp_transmit_skb.tcp_write_xmit.tcp_sendmsg_locked 1.78 -0.6 1.16 perf-profile.calltrace.cycles-pp.dequeue_task_fair.__sched_text_start.schedule.schedule_timeout.wait_woken 7.18 ± 2% -0.6 6.58 ± 2% perf-profile.calltrace.cycles-pp.__tcp_push_pending_frames.tcp_sendmsg_locked.tcp_sendmsg.sock_sendmsg.__sys_sendto 7.16 ± 2% -0.6 6.56 ± 2% perf-profile.calltrace.cycles-pp.tcp_write_xmit.__tcp_push_pending_frames.tcp_sendmsg_locked.tcp_sendmsg.sock_sendmsg 1.37 -0.5 0.88 perf-profile.calltrace.cycles-pp.exit_to_usermode_loop.do_syscall_64.entry_SYSCALL_64_after_hwframe 1.33 -0.5 0.84 perf-profile.calltrace.cycles-pp.schedule.exit_to_usermode_loop.do_syscall_64.entry_SYSCALL_64_after_hwframe 1.29 -0.5 0.82 perf-profile.calltrace.cycles-pp.__sched_text_start.schedule.exit_to_usermode_loop.do_syscall_64.entry_SYSCALL_64_after_hwframe 0.71 -0.5 0.25 ±100% perf-profile.calltrace.cycles-pp.__sched_text_start.preempt_schedule_common._cond_resched.__release_sock.release_sock 1.20 -0.4 0.82 perf-profile.calltrace.cycles-pp.__switch_to 1.09 -0.4 0.72 perf-profile.calltrace.cycles-pp.__alloc_skb.sk_stream_alloc_skb.tcp_sendmsg_locked.tcp_sendmsg.sock_sendmsg 1.09 -0.3 0.77 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64 6.42 ± 2% -0.3 6.09 ± 2% perf-profile.calltrace.cycles-pp.__tcp_transmit_skb.tcp_write_xmit.__tcp_push_pending_frames.tcp_sendmsg_locked.tcp_sendmsg 0.83 -0.3 0.56 perf-profile.calltrace.cycles-pp.kmem_cache_alloc_node.__alloc_skb.sk_stream_alloc_skb.tcp_sendmsg_locked.tcp_sendmsg 0.84 -0.3 0.58 ± 3% perf-profile.calltrace.cycles-pp.switch_mm_irqs_off.__sched_text_start.schedule.schedule_timeout.wait_woken 0.94 -0.3 0.68 perf-profile.calltrace.cycles-pp.syscall_return_via_sysret 0.74 -0.2 0.53 perf-profile.calltrace.cycles-pp._cond_resched.__release_sock.release_sock.tcp_sendmsg.sock_sendmsg 0.73 -0.2 0.52 perf-profile.calltrace.cycles-pp.preempt_schedule_common._cond_resched.__release_sock.release_sock.tcp_sendmsg 9.51 +0.7 10.17 perf-profile.calltrace.cycles-pp._raw_spin_lock.free_one_page.__free_pages_ok.skb_release_data.__kfree_skb 9.38 +0.7 10.08 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock.free_one_page.__free_pages_ok.skb_release_data 95.32 +1.4 96.69 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe 95.16 +1.4 96.58 perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe 11.17 +7.3 18.47 perf-profile.calltrace.cycles-pp.sk_stream_alloc_skb.tcp_sendmsg_locked.tcp_sendmsg.sock_sendmsg.__sys_sendto 11.81 ± 2% +7.4 19.18 perf-profile.calltrace.cycles-pp.__local_bh_enable_ip.ip_finish_output2.ip_output.__ip_queue_xmit.__tcp_transmit_skb 11.74 ± 2% +7.4 19.15 perf-profile.calltrace.cycles-pp.do_softirq.__local_bh_enable_ip.ip_finish_output2.ip_output.__ip_queue_xmit 11.52 ± 2% +7.5 18.98 perf-profile.calltrace.cycles-pp.__softirqentry_text_start.do_softirq_own_stack.do_softirq.__local_bh_enable_ip.ip_finish_output2 11.55 ± 2% +7.5 19.01 perf-profile.calltrace.cycles-pp.do_softirq_own_stack.do_softirq.__local_bh_enable_ip.ip_finish_output2.ip_output 11.27 ± 2% +7.5 18.81 perf-profile.calltrace.cycles-pp.net_rx_action.__softirqentry_text_start.do_softirq_own_stack.do_softirq.__local_bh_enable_ip 10.95 ± 2% +7.6 18.55 perf-profile.calltrace.cycles-pp.process_backlog.net_rx_action.__softirqentry_text_start.do_softirq_own_stack.do_softirq 9.28 +7.6 16.91 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock.free_one_page.__free_pages_ok.___pskb_trim 9.39 +7.7 17.05 perf-profile.calltrace.cycles-pp._raw_spin_lock.free_one_page.__free_pages_ok.___pskb_trim.sk_stream_alloc_skb 10.83 ± 2% +7.7 18.52 perf-profile.calltrace.cycles-pp.ip_finish_output2.ip_output.__ip_queue_xmit.__tcp_transmit_skb.tcp_write_xmit 9.57 +7.7 17.27 perf-profile.calltrace.cycles-pp.free_one_page.__free_pages_ok.___pskb_trim.sk_stream_alloc_skb.tcp_sendmsg_locked 9.94 +7.7 17.64 perf-profile.calltrace.cycles-pp.___pskb_trim.sk_stream_alloc_skb.tcp_sendmsg_locked.tcp_sendmsg.sock_sendmsg 10.49 ± 2% +7.7 18.23 perf-profile.calltrace.cycles-pp.__netif_receive_skb_one_core.process_backlog.net_rx_action.__softirqentry_text_start.do_softirq_own_stack 9.73 +7.8 17.49 perf-profile.calltrace.cycles-pp.__free_pages_ok.___pskb_trim.sk_stream_alloc_skb.tcp_sendmsg_locked.tcp_sendmsg 10.16 ± 2% +7.8 17.98 perf-profile.calltrace.cycles-pp.ip_rcv.__netif_receive_skb_one_core.process_backlog.net_rx_action.__softirqentry_text_start 3.79 +8.0 11.77 perf-profile.calltrace.cycles-pp.__tcp_push_pending_frames.tcp_rcv_established.tcp_v4_do_rcv.__release_sock.release_sock 3.78 +8.0 11.77 perf-profile.calltrace.cycles-pp.tcp_write_xmit.__tcp_push_pending_frames.tcp_rcv_established.tcp_v4_do_rcv.__release_sock 9.61 ± 2% +8.0 17.61 perf-profile.calltrace.cycles-pp.ip_local_deliver.ip_rcv.__netif_receive_skb_one_core.process_backlog.net_rx_action 9.45 ± 2% +8.1 17.50 perf-profile.calltrace.cycles-pp.ip_local_deliver_finish.ip_local_deliver.ip_rcv.__netif_receive_skb_one_core.process_backlog 9.40 ± 2% +8.1 17.47 perf-profile.calltrace.cycles-pp.ip_protocol_deliver_rcu.ip_local_deliver_finish.ip_local_deliver.ip_rcv.__netif_receive_skb_one_core 9.21 ± 2% +8.1 17.35 perf-profile.calltrace.cycles-pp.tcp_v4_rcv.ip_protocol_deliver_rcu.ip_local_deliver_finish.ip_local_deliver.ip_rcv 3.44 +8.1 11.58 perf-profile.calltrace.cycles-pp.__tcp_transmit_skb.tcp_write_xmit.__tcp_push_pending_frames.tcp_rcv_established.tcp_v4_do_rcv 8.65 ± 2% +8.2 16.88 perf-profile.calltrace.cycles-pp.ip_output.__ip_queue_xmit.__tcp_transmit_skb.tcp_write_xmit.__tcp_push_pending_frames 3.21 +8.2 11.44 perf-profile.calltrace.cycles-pp.__ip_queue_xmit.__tcp_transmit_skb.tcp_write_xmit.__tcp_push_pending_frames.tcp_rcv_established 18.55 ± 2% +8.6 27.14 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.get_page_from_freelist.__alloc_pages_nodemask.skb_page_frag_refill 20.09 ± 2% +8.6 28.68 perf-profile.calltrace.cycles-pp.sk_page_frag_refill.tcp_sendmsg_locked.tcp_sendmsg.sock_sendmsg.__sys_sendto 19.87 ± 2% +8.6 28.45 perf-profile.calltrace.cycles-pp.get_page_from_freelist.__alloc_pages_nodemask.skb_page_frag_refill.sk_page_frag_refill.tcp_sendmsg_locked 20.08 ± 2% +8.6 28.67 perf-profile.calltrace.cycles-pp.skb_page_frag_refill.sk_page_frag_refill.tcp_sendmsg_locked.tcp_sendmsg.sock_sendmsg 18.74 ± 2% +8.6 27.34 perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.get_page_from_freelist.__alloc_pages_nodemask.skb_page_frag_refill.sk_page_frag_refill 19.94 ± 2% +8.6 28.53 perf-profile.calltrace.cycles-pp.__alloc_pages_nodemask.skb_page_frag_refill.sk_page_frag_refill.tcp_sendmsg_locked.tcp_sendmsg 72.24 +8.7 80.98 perf-profile.calltrace.cycles-pp.__x64_sys_sendto.do_syscall_64.entry_SYSCALL_64_after_hwframe 72.13 +8.8 80.91 perf-profile.calltrace.cycles-pp.__sys_sendto.__x64_sys_sendto.do_syscall_64.entry_SYSCALL_64_after_hwframe 71.52 +9.0 80.48 perf-profile.calltrace.cycles-pp.sock_sendmsg.__sys_sendto.__x64_sys_sendto.do_syscall_64.entry_SYSCALL_64_after_hwframe 71.18 +9.1 80.25 perf-profile.calltrace.cycles-pp.tcp_sendmsg.sock_sendmsg.__sys_sendto.__x64_sys_sendto.do_syscall_64 0.00 +10.3 10.29 perf-profile.calltrace.cycles-pp.free_one_page.__free_pages_ok.skb_release_data.__kfree_skb.tcp_v4_rcv 0.00 +10.4 10.41 perf-profile.calltrace.cycles-pp.__free_pages_ok.skb_release_data.__kfree_skb.tcp_v4_rcv.ip_protocol_deliver_rcu 0.00 +10.7 10.66 perf-profile.calltrace.cycles-pp.skb_release_data.__kfree_skb.tcp_v4_rcv.ip_protocol_deliver_rcu.ip_local_deliver_finish 0.00 +10.8 10.81 perf-profile.calltrace.cycles-pp.__kfree_skb.tcp_v4_rcv.ip_protocol_deliver_rcu.ip_local_deliver_finish.ip_local_deliver 54.39 +12.2 66.64 perf-profile.calltrace.cycles-pp.tcp_sendmsg_locked.tcp_sendmsg.sock_sendmsg.__sys_sendto.__x64_sys_sendto 31487988 ± 3% -23.2% 24183132 softirqs.CPU0.NET_RX 32151929 -26.1% 23765617 ± 2% softirqs.CPU1.NET_RX 31962684 ± 3% -23.5% 24458409 softirqs.CPU10.NET_RX 31110228 ± 3% -26.2% 22964939 softirqs.CPU100.NET_RX 31626866 -27.6% 22906397 softirqs.CPU101.NET_RX 29779153 ± 3% -22.9% 22969929 softirqs.CPU102.NET_RX 31513116 -27.3% 22920216 softirqs.CPU103.NET_RX 31713243 ± 2% -24.9% 23826688 ± 2% softirqs.CPU11.NET_RX 32072474 -26.1% 23696606 ± 2% softirqs.CPU12.NET_RX 32279430 -24.8% 24260634 softirqs.CPU13.NET_RX 31697375 ± 3% -23.0% 24406003 softirqs.CPU14.NET_RX 32278026 -24.7% 24317293 softirqs.CPU15.NET_RX 31162122 ± 3% -21.7% 24396540 softirqs.CPU16.NET_RX 32209863 -24.2% 24418327 softirqs.CPU17.NET_RX 32329002 -25.6% 24054387 ± 2% softirqs.CPU18.NET_RX 32127509 -24.5% 24258223 softirqs.CPU19.NET_RX 32217120 -25.2% 24089001 softirqs.CPU2.NET_RX 31415502 ± 3% -22.5% 24349881 softirqs.CPU20.NET_RX 32186243 -25.0% 24137967 softirqs.CPU21.NET_RX 32271735 -26.9% 23586468 ± 2% softirqs.CPU22.NET_RX 31871656 ± 3% -23.8% 24292511 softirqs.CPU23.NET_RX 32321894 -25.2% 24177260 softirqs.CPU24.NET_RX 31969188 -24.7% 24057006 softirqs.CPU25.NET_RX 30938777 -27.7% 22360933 ± 2% softirqs.CPU26.NET_RX 30118051 ± 3% -25.9% 22321500 ± 2% softirqs.CPU27.NET_RX 31079253 -28.1% 22337317 ± 2% softirqs.CPU28.NET_RX 31429166 -27.6% 22746093 ± 2% softirqs.CPU29.NET_RX 30485045 ± 3% -20.5% 24221128 ± 2% softirqs.CPU3.NET_RX 30912223 ± 2% -26.8% 22622183 ± 2% softirqs.CPU30.NET_RX 31357027 -27.6% 22717601 softirqs.CPU31.NET_RX 31052178 -27.3% 22565776 ± 2% softirqs.CPU32.NET_RX 31101262 -28.2% 22323348 ± 2% softirqs.CPU33.NET_RX 31331660 -27.7% 22656208 softirqs.CPU34.NET_RX 30833278 ± 3% -25.4% 22998777 softirqs.CPU35.NET_RX 30738161 ± 3% -25.2% 22998898 softirqs.CPU36.NET_RX 30881460 ± 4% -26.6% 22671426 softirqs.CPU37.NET_RX 31140748 -27.4% 22621903 softirqs.CPU38.NET_RX 31218654 -27.3% 22684067 softirqs.CPU39.NET_RX 32583070 -24.8% 24513384 softirqs.CPU4.NET_RX 31462139 -28.4% 22517351 softirqs.CPU40.NET_RX 31566697 -27.9% 22768461 softirqs.CPU41.NET_RX 30975279 ± 2% -27.2% 22555119 softirqs.CPU42.NET_RX 30718280 ± 3% -26.2% 22661320 ± 2% softirqs.CPU43.NET_RX 31283332 -27.3% 22740697 softirqs.CPU44.NET_RX 30983467 -26.4% 22794814 softirqs.CPU45.NET_RX 30658730 ± 4% -28.4% 21946934 ± 2% softirqs.CPU46.NET_RX 31181856 -26.5% 22929226 softirqs.CPU47.NET_RX 31037938 ± 2% -26.9% 22691892 ± 2% softirqs.CPU48.NET_RX 30911500 ± 2% -26.3% 22777693 softirqs.CPU49.NET_RX 31802923 ± 3% -23.9% 24204592 softirqs.CPU5.NET_RX 31517206 -28.1% 22648510 softirqs.CPU50.NET_RX 30660453 ± 3% -27.2% 22327925 ± 2% softirqs.CPU51.NET_RX 31819962 ± 2% -23.9% 24206194 softirqs.CPU52.NET_RX 32142649 -26.4% 23668373 ± 2% softirqs.CPU53.NET_RX 32149559 -24.2% 24360340 softirqs.CPU54.NET_RX 32425853 -26.4% 23854082 ± 2% softirqs.CPU55.NET_RX 31817157 ± 2% -24.3% 24096785 softirqs.CPU56.NET_RX 31640490 ± 3% -24.4% 23920278 ± 2% softirqs.CPU57.NET_RX 31650729 ± 3% -23.1% 24327421 softirqs.CPU58.NET_RX 31269148 ± 3% -22.3% 24307789 softirqs.CPU59.NET_RX 31030416 ± 4% -22.0% 24203694 softirqs.CPU6.NET_RX 32255424 -25.2% 24141973 ± 2% softirqs.CPU60.NET_RX 32245715 -25.1% 24167151 ± 2% softirqs.CPU61.NET_RX 31763163 ± 2% -23.1% 24415760 softirqs.CPU62.NET_RX 31632095 ± 3% -22.7% 24451397 softirqs.CPU63.NET_RX 31490930 ± 2% -24.2% 23879322 softirqs.CPU64.NET_RX 31516754 ± 3% -23.5% 24108737 softirqs.CPU65.NET_RX 32359236 -24.5% 24440756 softirqs.CPU66.NET_RX 31784935 ± 3% -23.9% 24195033 ± 2% softirqs.CPU67.NET_RX 30967426 ± 4% -21.8% 24208434 softirqs.CPU68.NET_RX 32208978 -27.9% 23223363 softirqs.CPU69.NET_RX 31534273 ± 3% -23.6% 24083955 softirqs.CPU7.NET_RX 31800791 ± 2% -23.3% 24375677 softirqs.CPU70.NET_RX 31511220 ± 3% -23.1% 24225232 softirqs.CPU71.NET_RX 32055308 -25.2% 23985501 ± 2% softirqs.CPU72.NET_RX 32255683 -24.3% 24425926 softirqs.CPU73.NET_RX 31808226 ± 2% -23.7% 24280351 ± 2% softirqs.CPU74.NET_RX 31994746 ± 3% -25.5% 23828432 ± 2% softirqs.CPU75.NET_RX 31732779 ± 3% -24.0% 24121773 ± 2% softirqs.CPU76.NET_RX 31249558 ± 2% -23.8% 23816726 ± 2% softirqs.CPU77.NET_RX 29102709 ± 3% -24.1% 22087804 ± 2% softirqs.CPU78.NET_RX 31050911 -26.3% 22879670 softirqs.CPU79.NET_RX 32385972 -25.5% 24114907 softirqs.CPU8.NET_RX 31250746 -26.8% 22865137 softirqs.CPU80.NET_RX 30970512 ± 3% -27.5% 22460428 ± 2% softirqs.CPU81.NET_RX 30769035 ± 3% -26.4% 22640114 softirqs.CPU82.NET_RX 31163906 -28.0% 22426881 ± 2% softirqs.CPU83.NET_RX 30833151 ± 2% -26.9% 22535706 softirqs.CPU84.NET_RX 31213448 -27.8% 22530613 ± 2% softirqs.CPU85.NET_RX 31299014 -27.8% 22598945 ± 2% softirqs.CPU86.NET_RX 31380574 -26.8% 22966556 softirqs.CPU87.NET_RX 31419156 -26.9% 22966072 softirqs.CPU88.NET_RX 31485946 -29.0% 22360125 ± 2% softirqs.CPU89.NET_RX 32266157 -25.9% 23914440 ± 2% softirqs.CPU9.NET_RX 30336933 ± 3% -25.9% 22489963 softirqs.CPU90.NET_RX 30595348 ± 3% -26.1% 22614928 softirqs.CPU91.NET_RX 30924249 ± 2% -26.0% 22894665 softirqs.CPU92.NET_RX 30593902 ± 2% -24.8% 23001426 softirqs.CPU93.NET_RX 31556110 -29.2% 22330354 ± 2% softirqs.CPU94.NET_RX 30668287 ± 3% -29.0% 21759470 softirqs.CPU95.NET_RX 31245897 -26.9% 22847267 softirqs.CPU96.NET_RX 31136421 -28.7% 22213048 ± 2% softirqs.CPU97.NET_RX 30013471 ± 4% -23.9% 22844117 softirqs.CPU98.NET_RX 31291344 -28.5% 22378868 ± 2% softirqs.CPU99.NET_RX 3.269e+09 -25.6% 2.431e+09 softirqs.NET_RX Disclaimer: Results have been estimated based on internal Intel analysis and are provided for informational purposes only. Any difference in system hardware or software design or configuration may affect actual performance. Thanks, Rong Chen

3 years

3
5
0 / 0

Re: [LKP] [rcu] kernel BUG at include/linux/pagemap.h:149!

by Frederic Weisbecker

On Fri, Sep 11, 2015 at 10:19:47AM +0800, Boqun Feng wrote: > Subject: [PATCH 01/27] rcu: Don't disable preemption for Tiny and Tree RCU > readers > > Because preempt_disable() maps to barrier() for non-debug builds, > it forces the compiler to spill and reload registers. Because Tree > RCU and Tiny RCU now only appear in CONFIG_PREEMPT=n builds, these > barrier() instances generate needless extra code for each instance of > rcu_read_lock() and rcu_read_unlock(). This extra code slows down Tree > RCU and bloats Tiny RCU. > > This commit therefore removes the preempt_disable() and preempt_enable() > from the non-preemptible implementations of __rcu_read_lock() and > __rcu_read_unlock(), respectively. > > For debug purposes, preempt_disable() and preempt_enable() are still > kept if CONFIG_PREEMPT_COUNT=y, which makes the detection of sleeping > inside atomic sections still work in non-preemptible kernels. > > Signed-off-by: Boqun Feng <boqun.feng(a)gmail.com> > Signed-off-by: Paul E. McKenney <paulmck(a)linux.vnet.ibm.com> > --- > include/linux/rcupdate.h | 6 ++++-- > include/linux/rcutiny.h | 1 + > kernel/rcu/tree.c | 9 +++++++++ > 3 files changed, 14 insertions(+), 2 deletions(-) > > diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h > index d63bb77..6c3cece 100644 > --- a/include/linux/rcupdate.h > +++ b/include/linux/rcupdate.h > @@ -297,12 +297,14 @@ void synchronize_rcu(void); > > static inline void __rcu_read_lock(void) > { > - preempt_disable(); > + if (IS_ENABLED(CONFIG_PREEMPT_COUNT)) > + preempt_disable(); preempt_disable() is a no-op when !CONFIG_PREEMPT_COUNT, right? Or rather it's a barrier(), which is anyway implied by rcu_read_lock(). So perhaps we can get rid of the IS_ENABLED() check?

3 years

4
4
0 / 0

[btrfs] 302167c50b: fio.write_bw_MBps -12.4% regression

by kernel test robot

Greeting, FYI, we noticed a -12.4% regression of fio.write_bw_MBps due to commit: commit: 302167c50b32e7fccc98994a91d40ddbbab04e52 ("btrfs: don't end the transaction for delayed refs in throttle") https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git pending-fixes in testcase: fio-basic on test machine: 88 threads Intel(R) Xeon(R) CPU E5-2699 v4 @ 2.20GHz with 64G memory with following parameters: runtime: 300s nr_task: 8t disk: 1SSD fs: btrfs rw: randwrite bs: 4k ioengine: sync test_size: 400g cpufreq_governor: performance ucode: 0xb00002e test-description: Fio is a tool that will spawn a number of threads or processes doing a particular type of I/O action as specified by the user. test-url: https://github.com/axboe/fio Details are as below: --------------------------------------------------------------------------------------------------> To reproduce: git clone https://github.com/intel/lkp-tests.git cd lkp-tests bin/lkp install job.yaml # job file is attached in this email bin/lkp run job.yaml ========================================================================================= bs/compiler/cpufreq_governor/disk/fs/ioengine/kconfig/nr_task/rootfs/runtime/rw/tbox_group/test_size/testcase/ucode: 4k/gcc-7/performance/1SSD/btrfs/sync/x86_64-rhel-7.2/8t/debian-x86_64-2018-04-03.cgz/300s/randwrite/lkp-bdw-ep3b/400g/fio-basic/0xb00002e commit: a627947076 ("Btrfs: fix deadlock when allocating tree block during leaf/node split") 302167c50b ("btrfs: don't end the transaction for delayed refs in throttle") a6279470762c19ba 302167c50b32e7fccc98994a91 ---------------- -------------------------- fail:runs %reproduction fail:runs | | | :4 25% 1:4 dmesg.WARNING:at#for_ip_interrupt_entry/0x %stddev %change %stddev \ | \ 0.02 ± 4% -0.0 0.01 fio.latency_100ms% 41.36 ± 2% -14.7 26.66 ± 12% fio.latency_100us% 0.85 ± 6% +0.3 1.14 ± 14% fio.latency_10us% 0.01 +0.0 0.02 ± 3% fio.latency_2000ms% 0.02 ± 18% -0.0 0.01 ± 5% fio.latency_20ms% 0.50 ± 11% +0.1 0.56 ± 11% fio.latency_20us% 0.03 ± 11% -0.0 0.01 ± 10% fio.latency_250ms% 8.90 ± 5% -2.1 6.80 ± 3% fio.latency_250us% 0.03 ± 7% -0.0 0.02 ± 7% fio.latency_500ms% 0.03 ± 15% -0.0 0.01 fio.latency_50ms% 41.49 ± 3% +16.2 57.73 ± 5% fio.latency_50us% 44895412 ± 2% -12.5% 39295860 fio.time.file_system_outputs 36.25 ± 3% -16.6% 30.25 fio.time.percent_of_cpu_this_job_got 98.06 ± 3% -18.2% 80.23 fio.time.system_time 5558064 ± 2% -12.7% 4851975 fio.time.voluntary_context_switches 5610728 ± 2% -12.5% 4909544 fio.workload 72.97 ± 2% -12.4% 63.91 fio.write_bw_MBps 427.18 ± 2% +14.2% 487.93 fio.write_clat_mean_us 13691 ± 2% +43.7% 19669 fio.write_clat_stddev 18680 ± 2% -12.4% 16360 fio.write_iops 0.97 -0.7 0.30 ± 2% mpstat.cpu.iowait% 3.94 ± 3% -1.5 2.40 mpstat.cpu.sys% 2875717 -13.4% 2489058 softirqs.BLOCK 5107622 ± 3% +27.5% 6510241 ± 4% softirqs.RCU 30695 ± 15% -30.2% 21424 ± 11% numa-meminfo.node0.Writeback 179069 ± 19% +134.0% 419038 ± 20% numa-meminfo.node1.Active 36182 ±105% +701.8% 290125 ± 30% numa-meminfo.node1.Active(file) 1.096e+09 ± 3% -22.2% 8.531e+08 ± 7% cpuidle.C1.time 57940399 -34.0% 38218420 ± 4% cpuidle.C1.usage 13565831 ± 7% -67.4% 4420507 ± 16% cpuidle.POLL.time 4064467 ± 5% -72.0% 1136676 ± 12% cpuidle.POLL.usage 124.33 ± 2% -59.2% 50.74 ± 3% iostat.sda.avgqu-sz 18410 -13.2% 15975 iostat.sda.w/s 300245 -21.0% 237217 iostat.sda.wkB/s 9.15 ± 10% -42.0% 5.31 ± 19% iostat.sda.wrqm/s 300252 -21.0% 237234 vmstat.io.bo 1.00 -100.0% 0.00 vmstat.procs.b 3.00 -33.3% 2.00 vmstat.procs.r 392814 -36.9% 247683 vmstat.system.cs 12975351 -10.0% 11683920 meminfo.Inactive 12742134 -10.1% 11450539 meminfo.Inactive(file) 1336423 -10.4% 1197060 meminfo.SUnreclaim 36875 ± 15% -35.8% 23682 ± 8% meminfo.Writeback 97963 ± 4% -9.3% 88890 ± 2% meminfo.max_used_kB 9315760 ± 11% -24.4% 7044222 ± 9% numa-vmstat.node0.nr_dirtied 7593 ± 15% -30.2% 5301 ± 8% numa-vmstat.node0.nr_writeback 9253810 ± 11% -24.4% 6992866 ± 9% numa-vmstat.node0.nr_written 9053 ±105% +699.4% 72375 ± 30% numa-vmstat.node1.nr_active_file 9053 ±105% +699.4% 72375 ± 30% numa-vmstat.node1.nr_zone_active_file 197.50 ± 2% -20.8% 156.50 ± 4% turbostat.Avg_MHz 7.59 ± 4% -1.1 6.45 ± 7% turbostat.Busy% 57935368 -34.0% 38214519 ± 4% turbostat.C1 3.97 ± 3% -0.9 3.10 ± 7% turbostat.C1% 117.34 ± 5% -10.4% 105.14 ± 3% turbostat.PkgWatt 6.93 -5.8% 6.53 ± 3% turbostat.RAMWatt 23703837 -21.2% 18668822 proc-vmstat.nr_dirtied 11565487 +2.6% 11866577 proc-vmstat.nr_free_pages 3186566 -10.0% 2867899 proc-vmstat.nr_inactive_file 14987 -2.0% 14683 proc-vmstat.nr_kernel_stack 203124 -2.2% 198730 proc-vmstat.nr_slab_reclaimable 334281 -10.4% 299452 proc-vmstat.nr_slab_unreclaimable 23643508 -21.2% 18622029 proc-vmstat.nr_written 3186566 -10.0% 2867899 proc-vmstat.nr_zone_inactive_file 9200220 ± 4% -16.8% 7655217 ± 2% proc-vmstat.numa_hit 9182883 ± 4% -16.8% 7637938 ± 2% proc-vmstat.numa_local 15866899 ± 3% -34.3% 10421136 ± 2% proc-vmstat.pgalloc_normal 15347481 -37.3% 9620050 ± 3% proc-vmstat.pgfree 94578712 -21.2% 74490196 proc-vmstat.pgpgout 1.653e+09 -28.2% 1.188e+09 ± 2% perf-stat.i.branch-instructions 16239810 ± 6% -20.2% 12960638 ± 7% perf-stat.i.cache-misses 1.771e+08 ± 4% -21.6% 1.389e+08 ± 6% perf-stat.i.cache-references 397106 -37.0% 250140 perf-stat.i.context-switches 1.75e+10 ± 5% -21.7% 1.37e+10 ± 6% perf-stat.i.cpu-cycles 8.56 ± 17% -55.8% 3.79 ± 15% perf-stat.i.cpu-migrations 2.408e+09 -24.3% 1.823e+09 ± 2% perf-stat.i.dTLB-loads 1.351e+09 ± 6% -18.8% 1.097e+09 ± 2% perf-stat.i.dTLB-stores 6077563 ± 3% -14.6% 5188983 ± 6% perf-stat.i.iTLB-loads 8.756e+09 -25.6% 6.518e+09 perf-stat.i.instructions 48.01 ± 18% +12.6 60.57 ± 7% perf-stat.i.node-load-miss-rate% 2697176 ± 11% -36.8% 1705410 ± 12% perf-stat.i.node-loads 50.90 ± 16% +12.8 63.72 ± 5% perf-stat.overall.node-load-miss-rate% 486504 ± 2% -15.1% 412869 ± 2% perf-stat.overall.path-length 1.648e+09 -28.2% 1.184e+09 ± 2% perf-stat.ps.branch-instructions 16185048 ± 6% -20.2% 12917198 ± 7% perf-stat.ps.cache-misses 1.765e+08 ± 4% -21.6% 1.384e+08 ± 6% perf-stat.ps.cache-references 395744 -37.0% 249290 perf-stat.ps.context-switches 1.744e+10 ± 5% -21.7% 1.365e+10 ± 6% perf-stat.ps.cpu-cycles 8.54 ± 17% -55.7% 3.78 ± 15% perf-stat.ps.cpu-migrations 2.4e+09 -24.3% 1.817e+09 ± 2% perf-stat.ps.dTLB-loads 1.347e+09 ± 6% -18.8% 1.094e+09 ± 2% perf-stat.ps.dTLB-stores 6056751 ± 3% -14.6% 5171616 ± 6% perf-stat.ps.iTLB-loads 8.727e+09 -25.6% 6.497e+09 perf-stat.ps.instructions 2688159 ± 11% -36.8% 1699709 ± 12% perf-stat.ps.node-loads 2.729e+12 -25.7% 2.026e+12 perf-stat.total.instructions 7679 ± 2% -37.9% 4771 ± 7% sched_debug.cfs_rq:/.exec_clock.avg 25109 ± 10% -20.3% 20001 ± 12% sched_debug.cfs_rq:/.exec_clock.max 6099 ± 20% -24.1% 4629 ± 7% sched_debug.cfs_rq:/.exec_clock.stddev 96721 ± 8% -43.2% 54939 ± 37% sched_debug.cfs_rq:/.load.avg 243210 ± 4% -27.0% 177643 ± 21% sched_debug.cfs_rq:/.load.stddev 105.27 ± 15% -43.2% 59.81 ± 22% sched_debug.cfs_rq:/.load_avg.avg 197.18 ± 11% -21.2% 155.31 ± 8% sched_debug.cfs_rq:/.load_avg.stddev 0.13 ± 6% -31.5% 0.09 ± 25% sched_debug.cfs_rq:/.nr_running.avg 49.64 ± 12% -49.3% 25.18 ± 28% sched_debug.cfs_rq:/.runnable_load_avg.avg 689.54 ± 4% -9.3% 625.71 ± 5% sched_debug.cfs_rq:/.runnable_load_avg.max 142.56 ± 7% -26.4% 104.98 ± 12% sched_debug.cfs_rq:/.runnable_load_avg.stddev 97240 ± 8% -46.4% 52094 ± 33% sched_debug.cfs_rq:/.runnable_weight.avg 243593 ± 4% -28.5% 174272 ± 19% sched_debug.cfs_rq:/.runnable_weight.stddev 147.89 ± 8% -27.2% 107.65 ± 13% sched_debug.cfs_rq:/.util_avg.avg 192.27 ± 8% -20.2% 153.44 ± 8% sched_debug.cfs_rq:/.util_avg.stddev 43.61 ± 16% -60.4% 17.27 ± 44% sched_debug.cfs_rq:/.util_est_enqueued.avg 493.75 ± 16% -44.6% 273.67 ± 33% sched_debug.cfs_rq:/.util_est_enqueued.max 120.70 ± 13% -52.0% 57.95 ± 35% sched_debug.cfs_rq:/.util_est_enqueued.stddev 26.69 ± 32% -43.1% 15.20 ± 13% sched_debug.cpu.cpu_load[0].avg 107.63 ± 13% -22.0% 84.01 ± 8% sched_debug.cpu.cpu_load[0].stddev 28.23 ± 30% -46.4% 15.13 ± 10% sched_debug.cpu.cpu_load[1].avg 96.80 ± 14% -19.5% 77.96 ± 4% sched_debug.cpu.cpu_load[1].stddev 28.35 ± 27% -50.8% 13.93 ± 13% sched_debug.cpu.cpu_load[2].avg 26.83 ± 28% -54.8% 12.13 ± 16% sched_debug.cpu.cpu_load[3].avg 76.35 ± 21% -27.2% 55.61 ± 9% sched_debug.cpu.cpu_load[3].stddev 24.61 ± 29% -58.0% 10.35 ± 18% sched_debug.cpu.cpu_load[4].avg 67.78 ± 23% -29.6% 47.73 ± 16% sched_debug.cpu.cpu_load[4].stddev 217.01 ± 9% -29.1% 153.85 ± 11% sched_debug.cpu.curr->pid.avg 65004 ± 18% -52.3% 31025 ± 31% sched_debug.cpu.load.avg 200774 ± 8% -31.1% 138243 ± 19% sched_debug.cpu.load.stddev 0.09 ± 12% -33.7% 0.06 ± 17% sched_debug.cpu.nr_running.avg 0.27 ± 5% -16.5% 0.23 ± 9% sched_debug.cpu.nr_running.stddev 735069 -32.2% 498554 sched_debug.cpu.nr_switches.avg 2860144 ± 11% -27.4% 2076064 ± 13% sched_debug.cpu.nr_switches.max 665483 ± 24% -31.6% 455234 ± 9% sched_debug.cpu.nr_switches.stddev 0.13 ± 7% -30.1% 0.09 ± 12% sched_debug.cpu.nr_uninterruptible.avg 735117 -32.2% 498430 sched_debug.cpu.sched_count.avg 2858539 ± 11% -27.4% 2076509 ± 13% sched_debug.cpu.sched_count.max 665356 ± 24% -31.6% 454947 ± 9% sched_debug.cpu.sched_count.stddev 366543 -32.2% 248579 sched_debug.cpu.sched_goidle.avg 1428344 ± 11% -27.4% 1036752 ± 13% sched_debug.cpu.sched_goidle.max 332365 ± 24% -31.6% 227301 ± 9% sched_debug.cpu.sched_goidle.stddev 368002 -32.2% 249386 sched_debug.cpu.ttwu_count.avg 3059342 -9.8% 2760232 slabinfo.Acpi-State.active_objs 60835 -10.0% 54758 slabinfo.Acpi-State.active_slabs 3102644 -10.0% 2792672 slabinfo.Acpi-State.num_objs 60835 -10.0% 54758 slabinfo.Acpi-State.num_slabs 40884 ± 7% -42.6% 23477 ± 21% slabinfo.avc_xperms_data.active_objs 323.25 ± 7% -41.8% 188.00 ± 21% slabinfo.avc_xperms_data.active_slabs 41459 ± 7% -41.8% 24144 ± 21% slabinfo.avc_xperms_data.num_objs 323.25 ± 7% -41.8% 188.00 ± 21% slabinfo.avc_xperms_data.num_slabs 1524 ± 18% -25.4% 1136 ± 11% slabinfo.biovec-128.active_objs 1536 ± 18% -24.8% 1155 ± 11% slabinfo.biovec-128.num_objs 1681 ± 7% -20.8% 1331 ± 13% slabinfo.biovec-64.active_objs 1681 ± 7% -20.8% 1331 ± 13% slabinfo.biovec-64.num_objs 2654 ± 10% -56.1% 1166 ± 13% slabinfo.biovec-max.active_objs 671.00 ± 10% -55.3% 300.00 ± 12% slabinfo.biovec-max.active_slabs 2685 ± 10% -55.3% 1201 ± 12% slabinfo.biovec-max.num_objs 671.00 ± 10% -55.3% 300.00 ± 12% slabinfo.biovec-max.num_slabs 21641 ± 9% -12.3% 18989 ± 7% slabinfo.btrfs_delayed_ref_head.active_objs 22866 ± 8% -10.1% 20556 ± 7% slabinfo.btrfs_delayed_ref_head.num_objs 67913 ± 4% -12.5% 59451 ± 3% slabinfo.btrfs_extent_buffer.active_objs 1237 ± 4% -14.7% 1055 ± 3% slabinfo.btrfs_extent_buffer.active_slabs 71775 ± 4% -14.7% 61246 ± 3% slabinfo.btrfs_extent_buffer.num_objs 1237 ± 4% -14.7% 1055 ± 3% slabinfo.btrfs_extent_buffer.num_slabs 6184518 -10.1% 5562477 slabinfo.btrfs_extent_map.active_objs 110462 -10.1% 99345 slabinfo.btrfs_extent_map.active_slabs 6185888 -10.1% 5563352 slabinfo.btrfs_extent_map.num_objs 110462 -10.1% 99345 slabinfo.btrfs_extent_map.num_slabs 26097 ± 3% -27.1% 19016 ± 9% slabinfo.btrfs_ordered_extent.active_objs 673.75 ± 4% -26.8% 493.50 ± 9% slabinfo.btrfs_ordered_extent.active_slabs 26301 ± 4% -26.8% 19264 ± 9% slabinfo.btrfs_ordered_extent.num_objs 673.75 ± 4% -26.8% 493.50 ± 9% slabinfo.btrfs_ordered_extent.num_slabs 13863 ± 5% -39.9% 8328 ± 17% slabinfo.btrfs_path.active_objs 387.25 ± 5% -39.4% 234.50 ± 16% slabinfo.btrfs_path.active_slabs 13954 ± 5% -39.3% 8467 ± 16% slabinfo.btrfs_path.num_objs 387.25 ± 5% -39.4% 234.50 ± 16% slabinfo.btrfs_path.num_slabs 13884 ± 9% -25.6% 10330 ± 15% slabinfo.kmalloc-128.active_objs 439.75 ± 8% -24.7% 331.25 ± 15% slabinfo.kmalloc-128.active_slabs 14089 ± 8% -24.6% 10617 ± 15% slabinfo.kmalloc-128.num_objs 439.75 ± 8% -24.7% 331.25 ± 15% slabinfo.kmalloc-128.num_slabs 1554 ± 3% -10.8% 1386 ± 5% slabinfo.kmalloc-rcl-96.active_objs 1554 ± 3% -10.8% 1386 ± 5% slabinfo.kmalloc-rcl-96.num_objs 10158 ± 8% -28.3% 7284 ± 15% slabinfo.mnt_cache.active_objs 10369 ± 8% -26.9% 7581 ± 14% slabinfo.mnt_cache.num_objs 1660 ± 7% -15.2% 1408 ± 11% slabinfo.skbuff_fclone_cache.active_objs 1660 ± 7% -15.2% 1408 ± 11% slabinfo.skbuff_fclone_cache.num_objs 17.20 ± 15% -10.1 7.14 ± 5% perf-profile.calltrace.cycles-pp.btrfs_mark_extent_written.btrfs_finish_ordered_io.normal_work_helper.process_one_work.worker_thread 19.67 ± 13% -9.6 10.08 ± 7% perf-profile.calltrace.cycles-pp.btrfs_finish_ordered_io.normal_work_helper.process_one_work.worker_thread.kthread 14.18 ± 16% -9.5 4.73 ± 6% perf-profile.calltrace.cycles-pp.btrfs_search_slot.btrfs_mark_extent_written.btrfs_finish_ordered_io.normal_work_helper.process_one_work 20.52 ± 13% -9.1 11.40 ± 5% perf-profile.calltrace.cycles-pp.normal_work_helper.process_one_work.worker_thread.kthread.ret_from_fork 27.59 ± 9% -8.7 18.88 ± 4% perf-profile.calltrace.cycles-pp.ret_from_fork 27.59 ± 9% -8.7 18.88 ± 4% perf-profile.calltrace.cycles-pp.kthread.ret_from_fork 24.79 ± 10% -6.3 18.45 ± 4% perf-profile.calltrace.cycles-pp.process_one_work.worker_thread.kthread.ret_from_fork 25.03 ± 9% -6.2 18.79 ± 4% perf-profile.calltrace.cycles-pp.worker_thread.kthread.ret_from_fork 5.57 ± 21% -4.2 1.36 ± 7% perf-profile.calltrace.cycles-pp.btrfs_lock_root_node.btrfs_search_slot.btrfs_mark_extent_written.btrfs_finish_ordered_io.normal_work_helper 5.55 ± 21% -4.2 1.35 ± 7% perf-profile.calltrace.cycles-pp.btrfs_tree_lock.btrfs_lock_root_node.btrfs_search_slot.btrfs_mark_extent_written.btrfs_finish_ordered_io 4.87 ± 20% -3.6 1.31 ± 10% perf-profile.calltrace.cycles-pp.btrfs_read_lock_root_node.btrfs_search_slot.btrfs_mark_extent_written.btrfs_finish_ordered_io.normal_work_helper 4.84 ± 20% -3.6 1.28 ± 10% perf-profile.calltrace.cycles-pp.btrfs_tree_read_lock.btrfs_read_lock_root_node.btrfs_search_slot.btrfs_mark_extent_written.btrfs_finish_ordered_io 3.84 ± 24% -3.1 0.75 ± 7% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.prepare_to_wait_event.btrfs_tree_lock.btrfs_lock_root_node.btrfs_search_slot 3.76 ± 24% -3.0 0.72 ± 7% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.prepare_to_wait_event.btrfs_tree_lock.btrfs_lock_root_node 3.60 ± 22% -2.8 0.81 ± 9% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.prepare_to_wait_event.btrfs_tree_read_lock.btrfs_read_lock_root_node.btrfs_search_slot 3.54 ± 22% -2.7 0.79 ± 10% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.prepare_to_wait_event.btrfs_tree_read_lock.btrfs_read_lock_root_node 3.47 ± 19% -2.7 0.80 ± 6% perf-profile.calltrace.cycles-pp.prepare_to_wait_event.btrfs_tree_lock.btrfs_lock_root_node.btrfs_search_slot.btrfs_mark_extent_written 3.25 ± 17% -2.4 0.85 ± 10% perf-profile.calltrace.cycles-pp.prepare_to_wait_event.btrfs_tree_read_lock.btrfs_read_lock_root_node.btrfs_search_slot.btrfs_mark_extent_written 1.85 ± 8% -1.2 0.65 ± 3% perf-profile.calltrace.cycles-pp.unlock_up.btrfs_search_slot.btrfs_mark_extent_written.btrfs_finish_ordered_io.normal_work_helper 1.83 ± 9% -1.2 0.63 ± 4% perf-profile.calltrace.cycles-pp.__wake_up_common_lock.unlock_up.btrfs_search_slot.btrfs_mark_extent_written.btrfs_finish_ordered_io 1.45 ± 17% -1.2 0.26 ±100% perf-profile.calltrace.cycles-pp.btrfs_search_slot.setup_leaf_for_split.btrfs_duplicate_item.btrfs_mark_extent_written.btrfs_finish_ordered_io 1.71 ± 8% -1.1 0.60 ± 5% perf-profile.calltrace.cycles-pp.__wake_up_common.__wake_up_common_lock.unlock_up.btrfs_search_slot.btrfs_mark_extent_written 1.69 ± 9% -1.1 0.59 ± 6% perf-profile.calltrace.cycles-pp.autoremove_wake_function.__wake_up_common.__wake_up_common_lock.unlock_up.btrfs_search_slot 1.63 ± 9% -1.1 0.57 ± 7% perf-profile.calltrace.cycles-pp.try_to_wake_up.autoremove_wake_function.__wake_up_common.__wake_up_common_lock.unlock_up 2.12 ± 13% -0.7 1.43 ± 5% perf-profile.calltrace.cycles-pp.setup_leaf_for_split.btrfs_duplicate_item.btrfs_mark_extent_written.btrfs_finish_ordered_io.normal_work_helper 2.75 ± 10% -0.7 2.09 ± 5% perf-profile.calltrace.cycles-pp.btrfs_duplicate_item.btrfs_mark_extent_written.btrfs_finish_ordered_io.normal_work_helper.process_one_work 0.76 ± 5% -0.2 0.57 ± 8% perf-profile.calltrace.cycles-pp.schedule_idle.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64 0.73 ± 5% -0.2 0.56 ± 8% perf-profile.calltrace.cycles-pp.__schedule.schedule_idle.do_idle.cpu_startup_entry.start_secondary 0.99 ± 7% -0.1 0.87 ± 6% perf-profile.calltrace.cycles-pp.__btrfs_cow_block.btrfs_cow_block.btrfs_search_slot.btrfs_mark_extent_written.btrfs_finish_ordered_io 0.99 ± 7% -0.1 0.87 ± 7% perf-profile.calltrace.cycles-pp.btrfs_cow_block.btrfs_search_slot.btrfs_mark_extent_written.btrfs_finish_ordered_io.normal_work_helper 0.88 ± 10% +0.3 1.21 ± 13% perf-profile.calltrace.cycles-pp.get_next_timer_interrupt.tick_nohz_next_event.tick_nohz_get_sleep_length.menu_select.do_idle 0.27 ±100% +0.4 0.62 ± 10% perf-profile.calltrace.cycles-pp.setup_items_for_insert.btrfs_insert_empty_items.btrfs_csum_file_blocks.add_pending_csums.btrfs_finish_ordered_io 0.99 ± 10% +0.4 1.38 ± 7% perf-profile.calltrace.cycles-pp.btrfs_insert_empty_items.btrfs_csum_file_blocks.add_pending_csums.btrfs_finish_ordered_io.normal_work_helper 1.71 ± 17% +0.5 2.17 ± 13% perf-profile.calltrace.cycles-pp.tick_nohz_get_sleep_length.menu_select.do_idle.cpu_startup_entry.start_secondary 0.29 ±100% +0.5 0.76 ± 12% perf-profile.calltrace.cycles-pp.__next_timer_interrupt.get_next_timer_interrupt.tick_nohz_next_event.tick_nohz_get_sleep_length.menu_select 1.50 ± 7% +0.5 2.00 ± 9% perf-profile.calltrace.cycles-pp.add_pending_csums.btrfs_finish_ordered_io.normal_work_helper.process_one_work.worker_thread 1.49 ± 7% +0.5 2.00 ± 9% perf-profile.calltrace.cycles-pp.btrfs_csum_file_blocks.add_pending_csums.btrfs_finish_ordered_io.normal_work_helper.process_one_work 0.31 ±103% +0.5 0.83 ± 12% perf-profile.calltrace.cycles-pp.scheduler_tick.update_process_times.tick_sched_handle.tick_sched_timer.__hrtimer_run_queues 0.15 ±173% +0.6 0.75 ± 6% perf-profile.calltrace.cycles-pp.btrfs_search_slot.btrfs_insert_empty_items.btrfs_csum_file_blocks.add_pending_csums.btrfs_finish_ordered_io 0.14 ±173% +0.6 0.75 ± 12% perf-profile.calltrace.cycles-pp.split_leaf.setup_leaf_for_split.btrfs_duplicate_item.btrfs_mark_extent_written.btrfs_finish_ordered_io 0.29 ±100% +0.6 0.91 ± 27% perf-profile.calltrace.cycles-pp.clockevents_program_event.hrtimer_interrupt.smp_apic_timer_interrupt.apic_timer_interrupt.cpuidle_enter_state 1.08 ± 18% +0.6 1.71 ± 12% perf-profile.calltrace.cycles-pp.update_process_times.tick_sched_handle.tick_sched_timer.__hrtimer_run_queues.hrtimer_interrupt 0.00 +0.7 0.66 ± 13% perf-profile.calltrace.cycles-pp.push_leaf_right.split_leaf.setup_leaf_for_split.btrfs_duplicate_item.btrfs_mark_extent_written 1.21 ± 19% +0.7 1.92 ± 12% perf-profile.calltrace.cycles-pp.tick_sched_handle.tick_sched_timer.__hrtimer_run_queues.hrtimer_interrupt.smp_apic_timer_interrupt 1.37 ± 20% +0.8 2.20 ± 12% perf-profile.calltrace.cycles-pp.tick_sched_timer.__hrtimer_run_queues.hrtimer_interrupt.smp_apic_timer_interrupt.apic_timer_interrupt 3.28 ± 11% +1.1 4.33 ± 10% perf-profile.calltrace.cycles-pp.menu_select.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64 0.00 +1.1 1.13 ± 24% perf-profile.calltrace.cycles-pp.__filemap_fdatawrite_range.btrfs_write_marked_extents.btrfs_write_and_wait_transaction.btrfs_commit_transaction.flush_space 0.00 +1.1 1.14 ± 24% perf-profile.calltrace.cycles-pp.btrfs_write_marked_extents.btrfs_write_and_wait_transaction.btrfs_commit_transaction.flush_space.btrfs_async_reclaim_metadata_space 0.00 +1.1 1.15 ± 23% perf-profile.calltrace.cycles-pp.btrfs_write_and_wait_transaction.btrfs_commit_transaction.flush_space.btrfs_async_reclaim_metadata_space.process_one_work 2.54 ± 17% +1.3 3.81 ± 12% perf-profile.calltrace.cycles-pp.__hrtimer_run_queues.hrtimer_interrupt.smp_apic_timer_interrupt.apic_timer_interrupt.cpuidle_enter_state 0.00 +1.4 1.35 ± 16% perf-profile.calltrace.cycles-pp.btrfs_run_delayed_refs.btrfs_commit_transaction.flush_space.btrfs_async_reclaim_metadata_space.process_one_work 0.00 +1.4 1.35 ± 16% perf-profile.calltrace.cycles-pp.__btrfs_run_delayed_refs.btrfs_run_delayed_refs.btrfs_commit_transaction.flush_space.btrfs_async_reclaim_metadata_space 3.59 ± 12% +1.9 5.50 ± 8% perf-profile.calltrace.cycles-pp.hrtimer_interrupt.smp_apic_timer_interrupt.apic_timer_interrupt.cpuidle_enter_state.do_idle 0.15 ±173% +2.5 2.67 ± 14% perf-profile.calltrace.cycles-pp.btrfs_async_reclaim_metadata_space.process_one_work.worker_thread.kthread.ret_from_fork 0.15 ±173% +2.5 2.67 ± 14% perf-profile.calltrace.cycles-pp.flush_space.btrfs_async_reclaim_metadata_space.process_one_work.worker_thread.kthread 0.00 +2.7 2.67 ± 14% perf-profile.calltrace.cycles-pp.btrfs_commit_transaction.flush_space.btrfs_async_reclaim_metadata_space.process_one_work.worker_thread 6.62 ± 16% +3.0 9.58 ± 5% perf-profile.calltrace.cycles-pp.smp_apic_timer_interrupt.apic_timer_interrupt.cpuidle_enter_state.do_idle.cpu_startup_entry 7.43 ± 10% +3.3 10.76 ± 2% perf-profile.calltrace.cycles-pp.apic_timer_interrupt.cpuidle_enter_state.do_idle.cpu_startup_entry.start_secondary 52.83 ± 4% +4.2 57.01 ± 4% perf-profile.calltrace.cycles-pp.intel_idle.cpuidle_enter_state.do_idle.cpu_startup_entry.start_secondary 62.44 ± 3% +7.2 69.61 ± 3% perf-profile.calltrace.cycles-pp.cpuidle_enter_state.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64 69.18 ± 3% +8.3 77.44 perf-profile.calltrace.cycles-pp.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64 69.26 ± 3% +8.3 77.52 perf-profile.calltrace.cycles-pp.start_secondary.secondary_startup_64 69.25 ± 3% +8.3 77.52 perf-profile.calltrace.cycles-pp.cpu_startup_entry.start_secondary.secondary_startup_64 69.97 ± 3% +8.8 78.74 perf-profile.calltrace.cycles-pp.secondary_startup_64 fio.write_clat_stddev 21000 +-+-----------------------------------------------------------------+ | O O O O | 20000 O-O O O O O O O O O O O O O O O O O O O | 19000 +-+ O O O | | | 18000 +-+ | 17000 +-+ | | | 16000 +-+ | 15000 +-+ | | | 14000 +-+ .+ .+.. .+. .+. .+. .+.+.+.+.+.. .+.| 13000 +-+. .+. + .+.+.+.+.+ + +.+ + +..+.+.+ +.+.+ | | + + | 12000 +-+-----------------------------------------------------------------+ fio.latency_2000ms_ 0.019 O-+-----------------------------O---O------------O------------------+ | O O O O O O O O | 0.018 +-+ O O | 0.017 +-+ O O O O | | O O O O O O O | 0.016 +-O | 0.015 +-+ O | | | 0.014 +-+ | 0.013 +-+ | | | 0.012 +-+ | 0.011 +-+ | | | 0.01 +-+-----------------------------------------------------------------+ fio.time.voluntary_context_switches 5.8e+06 +-+---------------------------------------------------------------+ | + + | 5.6e+06 +-+ + +.. : + .+. .+ : : | |. + : + : : +. + + : : : +.| | + : + + : + +. .+. + : .+. .+.+.+ :+ | 5.4e+06 +-+ +.+ + + + + +.+.+..+.+ + | | | 5.2e+06 +-+ | | | 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 | 4.8e+06 +-+ O O | | | 4.6e+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, Rong Chen

3 years, 1 month

2
3
0 / 0

[ext4] 345c0dbf3a: xfstests.ext4.303.fail

by kernel test robot

FYI, we noticed the following commit (built with gcc-7): commit: 345c0dbf3a30872d9b204db96b5857cd00808cae ("ext4: protect journal inode's blocks using block_validity") https://git.kernel.org/cgit/linux/kernel/git/tytso/ext4.git dev in testcase: xfstests with following parameters: disk: 4HDD fs: ext4 test: ext4-run test-description: xfstests is a regression test suite for xfs and other files ystems. test-url: git://git.kernel.org/pub/scm/fs/xfs/xfstests-dev.git on test machine: qemu-system-x86_64 -enable-kvm -cpu SandyBridge -smp 2 -m 2G caused below changes (please refer to attached dmesg/kmsg for entire log/backtrace): If you fix the issue, kindly add following tag Reported-by: kernel test robot <rong.a.chen(a)intel.com> 2019-04-26 12:48:55 export TEST_DIR=/fs/vda 2019-04-26 12:48:55 export TEST_DEV=/dev/vda 2019-04-26 12:48:55 export FSTYP=ext4 2019-04-26 12:48:55 export SCRATCH_MNT=/fs/scratch 2019-04-26 12:48:55 mkdir /fs/scratch -p 2019-04-26 12:48:55 export SCRATCH_DEV=/dev/vdd 2019-04-26 12:48:55 sed "s:^:ext4/:" /lkp/lkp/src/pack/xfstests-addon/tests/ext4-run | grep -F -f merged_ignored_files 2019-04-26 12:48:55 sed "s:^:ext4/:" /lkp/lkp/src/pack/xfstests-addon/tests/ext4-run | grep -v -F -f merged_ignored_files 2019-04-26 12:48:55 ./check ext4/001 ext4/002 ext4/003 ext4/004 ext4/005 ext4/010 ext4/011 ext4/012 ext4/013 ext4/014 ext4/015 ext4/016 ext4/017 ext4/018 ext4/019 ext4/020 ext4/021 ext4/022 ext4/023 ext4/024 ext4/025 ext4/026 ext4/027 ext4/028 ext4/032 ext4/033 ext4/034 ext4/271 ext4/301 ext4/302 ext4/303 ext4/305 ext4/306 ext4/307 ext4/308 FSTYP -- ext4 PLATFORM -- Linux/x86_64 vm-snb-2G-405 5.1.0-rc3-00007-g345c0dbf MKFS_OPTIONS -- /dev/vdd MOUNT_OPTIONS -- -o acl,user_xattr /dev/vdd /fs/scratch ext4/001 - output mismatch (see /lkp/benchmarks/xfstests/results//ext4/001.out.bad) --- tests/ext4/001.out 2019-04-25 09:04:55.000000000 +0800 +++ /lkp/benchmarks/xfstests/results//ext4/001.out.bad 2019-04-26 12:49:25.516490198 +0800 @@ -1,79 +1,45 @@ QA output created by 001 1. into a hole -0: [0..127]: hole -1: [128..383]: unwritten -2: [384..639]: hole +0: [0..639]: hole 1aca77e2188f52a62674fe8a873bdaba ... (Run 'diff -u /lkp/benchmarks/xfstests/tests/ext4/001.out /lkp/benchmarks/xfstests/results//ext4/001.out.bad' to see the entire diff) ext4/002 - output mismatch (see /lkp/benchmarks/xfstests/results//ext4/002.out.bad) --- tests/ext4/002.out 2019-04-25 09:04:55.000000000 +0800 +++ /lkp/benchmarks/xfstests/results//ext4/002.out.bad 2019-04-26 12:49:43.719490198 +0800 @@ -1,7 +1,13 @@ QA output created by 002 -Blocks modified: [0 - 0] -Blocks modified: [0 - 0] -Blocks modified: [0 - 9] -Blocks modified: [0 - 9] -Blocks modified: [64 - 64] -Blocks modified: [64 - 64] ... (Run 'diff -u /lkp/benchmarks/xfstests/tests/ext4/002.out /lkp/benchmarks/xfstests/results//ext4/002.out.bad' to see the entire diff) ext4/003 8s ext4/004 12s ext4/005 14s ext4/010 55s ext4/011 56s ext4/012 39s ext4/013 47s ext4/014 52s ext4/015 44s ext4/016 42s ext4/017 42s ext4/018 [failed, exit status 1]- output mismatch (see /lkp/benchmarks/xfstests/results//ext4/018.out.bad) --- tests/ext4/018.out 2019-04-25 09:04:55.000000000 +0800 +++ /lkp/benchmarks/xfstests/results//ext4/018.out.bad 2019-04-26 12:57:00.169490198 +0800 @@ -4,9 +4,8 @@ + make some files + check fs + corrupt image +./tests/ext4/018: line 59: 9287 Directory ACL: 0: syntax error in expression (error token is "Directory ACL: 0") + mount image + modify attrs -+ repair fs ... (Run 'diff -u /lkp/benchmarks/xfstests/tests/ext4/018.out /lkp/benchmarks/xfstests/results//ext4/018.out.bad' to see the entire diff) ext4/019 21s ext4/020 19s ext4/021 40s ext4/022 24s ext4/023 [failed, exit status 1]- output mismatch (see /lkp/benchmarks/xfstests/results//ext4/023.out.bad) --- tests/ext4/023.out 2019-04-25 09:04:55.000000000 +0800 +++ /lkp/benchmarks/xfstests/results//ext4/023.out.bad 2019-04-26 13:08:07.375490198 +0800 @@ -1,3 +1,2 @@ QA output created by 023 Format and populate -Mount ... (Run 'diff -u /lkp/benchmarks/xfstests/tests/ext4/023.out /lkp/benchmarks/xfstests/results//ext4/023.out.bad' to see the entire diff) ext4/024 27s ext4/025 21s ext4/026 [not run] mkfs.ext4 doesn't support ea_inode feature ext4/027 17s ext4/028 [failed, exit status 1]- output mismatch (see /lkp/benchmarks/xfstests/results//ext4/028.out.bad) --- tests/ext4/028.out 2019-04-25 09:04:55.000000000 +0800 +++ /lkp/benchmarks/xfstests/results//ext4/028.out.bad 2019-04-26 13:09:34.968490198 +0800 @@ -1,3 +1,2 @@ QA output created by 028 Format and mount -Compare fsmap ... (Run 'diff -u /lkp/benchmarks/xfstests/tests/ext4/028.out /lkp/benchmarks/xfstests/results//ext4/028.out.bad' to see the entire diff) ext4/032 83s ext4/033 52s ext4/034 15s ext4/271 1s ext4/301 55s ext4/302 _check_generic_filesystem: filesystem on /dev/vdd is inconsistent (see /lkp/benchmarks/xfstests/results//ext4/302.full for details) ext4/303 [failed, exit status 1]- output mismatch (see /lkp/benchmarks/xfstests/results//ext4/303.out.bad) --- tests/ext4/303.out 2019-04-25 09:04:55.000000000 +0800 +++ /lkp/benchmarks/xfstests/results//ext4/303.out.bad 2019-04-26 13:14:26.046490198 +0800 @@ -2,3 +2,5 @@ Start defragment activity +failed: '/usr/bin/fio /tmp/8398.fio' +(see /lkp/benchmarks/xfstests/results//ext4/303.full for details) ... (Run 'diff -u /lkp/benchmarks/xfstests/tests/ext4/303.out /lkp/benchmarks/xfstests/results//ext4/303.out.bad' to see the entire diff) ext4/305 19s ext4/306 11s ext4/307 - output mismatch (see /lkp/benchmarks/xfstests/results//ext4/307.out.bad) --- tests/ext4/307.out 2019-04-25 09:04:55.000000000 +0800 +++ /lkp/benchmarks/xfstests/results//ext4/307.out.bad 2019-04-26 13:15:02.522490198 +0800 @@ -1,6 +1,7 @@ QA output created by 307 Run fsstress +./tests/ext4/307: line 34: gawk: command not found Allocate donor file Perform compacting Check data ... (Run 'diff -u /lkp/benchmarks/xfstests/tests/ext4/307.out /lkp/benchmarks/xfstests/results//ext4/307.out.bad' to see the entire diff) ext4/308 13s Ran: ext4/001 ext4/002 ext4/003 ext4/004 ext4/005 ext4/010 ext4/011 ext4/012 ext4/013 ext4/014 ext4/015 ext4/016 ext4/017 ext4/018 ext4/019 ext4/020 ext4/021 ext4/022 ext4/023 ext4/024 ext4/025 ext4/026 ext4/027 ext4/028 ext4/032 ext4/033 ext4/034 ext4/271 ext4/301 ext4/302 ext4/303 ext4/305 ext4/306 ext4/307 ext4/308 Not run: ext4/026 Failures: ext4/001 ext4/002 ext4/018 ext4/023 ext4/028 ext4/302 ext4/303 ext4/307 Failed 8 of 35 tests To reproduce: # build kernel cd linux cp config-5.1.0-rc3-00007-g345c0dbf .config make HOSTCC=gcc-7 CC=gcc-7 ARCH=x86_64 olddefconfig make HOSTCC=gcc-7 CC=gcc-7 ARCH=x86_64 prepare make HOSTCC=gcc-7 CC=gcc-7 ARCH=x86_64 modules_prepare make HOSTCC=gcc-7 CC=gcc-7 ARCH=x86_64 SHELL=/bin/bash make HOSTCC=gcc-7 CC=gcc-7 ARCH=x86_64 bzImage git clone https://github.com/intel/lkp-tests.git cd lkp-tests find lib/ | cpio -o -H newc --quiet | gzip > modules.cgz bin/lkp qemu -k <bzImage> -m modules.cgz job-script # job-script is attached in this email Thanks, Rong Chen

3 years, 1 month

3
2
0 / 0

[function_graph] 02519fa3c4: BUG:kernel_reboot-without-warning_in_early-boot_stage, last_printk:early_console_in_setup_code

by kernel test robot

FYI, we noticed the following commit (built with gcc-7): commit: 02519fa3c43164aa8ee0ebbe1b464a568130a36f ("function_graph: Use a ftrace_graph_ret_stub() for return") https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git master in testcase: trinity with following parameters: runtime: 300s test-description: Trinity is a linux system call fuzz tester. test-url: http://codemonkey.org.uk/projects/trinity/ on test machine: qemu-system-x86_64 -enable-kvm -cpu SandyBridge -smp 2 -m 2G caused below changes (please refer to attached dmesg/kmsg for entire log/backtrace): +-----------------------------------------------------------------------------------------------+------------+------------+ | | 52fde6e70c | 02519fa3c4 | +-----------------------------------------------------------------------------------------------+------------+------------+ | boot_successes | 4 | 0 | | boot_failures | 0 | 4 | | BUG:kernel_reboot-without-warning_in_early-boot_stage,last_printk:early_console_in_setup_code | 0 | 4 | +-----------------------------------------------------------------------------------------------+------------+------------+ If you fix the issue, kindly add following tag Reported-by: kernel test robot <lkp(a)intel.com> early console in setup code BUG: kernel reboot-without-warning in early-boot stage, last printk: early console in setup code Linux version 5.1.0-rc3-00023-g02519fa #1 Command line: ip=::::vm-snb-quantal-ia32-804::dhcp root=/dev/ram0 user=lkp job=/lkp/jobs/scheduled/vm-snb-quantal-ia32-804/trinity-300s-quantal-core-i386-2019-04-26.cgz-02519fa3c43-20190429-62375-116iczn-1.yaml ARCH=x86_64 kconfig=x86_64-randconfig-s3-04261012 branch=linux-devel/devel-hourly-2019042607 commit=02519fa3c43164aa8ee0ebbe1b464a568130a36f BOOT_IMAGE=/pkg/linux/x86_64-randconfig-s3-04261012/gcc-7/02519fa3c43164aa8ee0ebbe1b464a568130a36f/vmlinuz-5.1.0-rc3-00023-g02519fa max_uptime=1500 RESULT_ROOT=/result/trinity/300s/vm-snb-quantal-ia32/quantal-core-i386-2019-04-26.cgz/x86_64-randconfig-s3-04261012/gcc-7/02519fa3c43164aa8ee0ebbe1b464a568130a36f/3 LKP_SERVER=inn debug apic=debug sysrq_always_enabled rcupdate.rcu_cpu_stall_timeout=100 net.ifnames=0 printk.devkmsg=on panic=-1 softlockup_panic=1 nmi_watchdog=panic oops=panic load_ramdisk=2 prompt_ramdisk=0 drbd.minor_count=8 systemd.log_level=err ignore_loglevel console=tty0 earlyprintk=ttyS0,115200 console=ttyS0,115200 vga=normal rw rcuperf.shutdown=0 Elapsed time: 30 To reproduce: # build kernel cd linux cp config-5.1.0-rc3-00023-g02519fa .config make HOSTCC=gcc-7 CC=gcc-7 ARCH=x86_64 olddefconfig make HOSTCC=gcc-7 CC=gcc-7 ARCH=x86_64 prepare make HOSTCC=gcc-7 CC=gcc-7 ARCH=x86_64 modules_prepare make HOSTCC=gcc-7 CC=gcc-7 ARCH=x86_64 SHELL=/bin/bash make HOSTCC=gcc-7 CC=gcc-7 ARCH=x86_64 bzImage git clone https://github.com/intel/lkp-tests.git cd lkp-tests bin/lkp qemu -k <bzImage> job-script # job-script is attached in this email Thanks, lkp

3 years, 2 months

2
1
0 / 0

[x86/unwind] 0830cf62f5: BUG:KASAN:stack-out-of-bounds_in_u

by kernel test robot

FYI, we noticed the following commit (built with gcc-7): commit: 0830cf62f5290b2f878faacc2b6f32e77bc2ea12 ("x86/unwind: Add hardcoded ORC entry for NULL") https://git.kernel.org/cgit/linux/kernel/git/stable/linux-stable-rc.git linux-5.0.y in testcase: trinity with following parameters: runtime: 300s test-description: Trinity is a linux system call fuzz tester. test-url: http://codemonkey.org.uk/projects/trinity/ on test machine: qemu-system-x86_64 -enable-kvm -cpu SandyBridge -smp 2 -m 2G caused below changes (please refer to attached dmesg/kmsg for entire log/backtrace): +------------------------------------+------------+------------+ | | 0312f3032e | 0830cf62f5 | +------------------------------------+------------+------------+ | boot_successes | 66 | 52 | | boot_failures | 0 | 14 | | BUG:KASAN:stack-out-of-bounds_in_u | 0 | 14 | | RIP:__x86_indirect_thunk_rdx | 0 | 14 | +------------------------------------+------------+------------+ If you fix the issue, kindly add following tag Reported-by: kernel test robot <rong.a.chen(a)intel.com> [ 176.470970] BUG: KASAN: stack-out-of-bounds in unwind_next_frame+0x1361/0x1b20 [ 176.473005] Read of size 8 at addr ffff88805723f878 by task trinity-main/605 [ 176.474776] [ 176.475424] CPU: 1 PID: 605 Comm: trinity-main Not tainted 5.0.4-00048-g0830cf6 #1 [ 176.477513] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.10.2-1 04/01/2014 [ 176.479754] Call Trace: [ 176.480590] <IRQ> [ 176.481344] dump_stack+0x5b/0x8b [ 176.482358] ? unwind_next_frame+0x1361/0x1b20 [ 176.483598] print_address_description+0x6a/0x290 [ 176.484893] ? unwind_next_frame+0x1361/0x1b20 [ 176.486126] ? unwind_next_frame+0x1361/0x1b20 [ 176.487370] kasan_report+0x139/0x199 [ 176.488450] ? unwind_next_frame+0x1361/0x1b20 [ 176.489690] unwind_next_frame+0x1361/0x1b20 [ 176.490893] ? entry_SYSCALL_64_after_hwframe+0x44/0xa9 [ 176.492288] ? unwind_get_return_address_ptr+0xb0/0xb0 [ 176.493662] ? rcu_dynticks_curr_cpu_in_eqs+0x54/0xb0 [ 176.495016] ? rcu_is_watching+0xc/0x20 [ 176.496131] ? rcu_is_watching+0xc/0x20 [ 176.497251] ? kernel_text_address+0x68/0x90 [ 176.498454] __save_stack_trace+0x73/0xd0 [ 176.499607] ? entry_SYSCALL_64_after_hwframe+0x44/0xa9 [ 176.500995] save_stack+0x32/0xb0 [ 176.502003] ? __kasan_slab_free+0x130/0x180 [ 176.503204] ? kfree+0xaa/0x1e0 [ 176.504176] ? rcu_process_callbacks+0x4b5/0xd00 [ 176.505467] ? __do_softirq+0x1bc/0x6d9 [ 176.506590] ? irq_exit+0x10f/0x130 [ 176.507644] ? smp_apic_timer_interrupt+0x176/0x400 [ 176.508994] ? apic_timer_interrupt+0xf/0x20 [ 176.510220] ? __x86_indirect_thunk_rcx+0x20/0x20 [ 176.511515] ? entry_SYSCALL_64_after_hwframe+0x44/0xa9 [ 176.512935] ? check_preempt_wakeup+0x2b4/0x690 [ 176.514233] ? probe_sched_switch+0x30/0x30 [ 176.515442] ? tracing_record_taskinfo_skip+0x56/0x70 [ 176.516824] ? tracing_record_taskinfo+0x14/0x1a0 [ 176.518156] ? ttwu_do_wakeup+0x3a1/0x530 [ 176.519304] ? _raw_spin_unlock_irqrestore+0x18/0x30 [ 176.520561] ? try_to_wake_up+0xc5/0x11e0 [ 176.521635] ? __migrate_task+0x140/0x140 [ 176.522705] ? _raw_spin_lock_irqsave+0x84/0xd0 [ 176.523957] ? _raw_spin_lock_irq+0xd0/0xd0 [ 176.525146] __kasan_slab_free+0x130/0x180 [ 176.526311] ? rcu_process_callbacks+0x4b5/0xd00 [ 176.527578] kfree+0xaa/0x1e0 [ 176.528515] rcu_process_callbacks+0x4b5/0xd00 [ 176.529708] ? rcu_read_unlock_special+0xf0/0xf0 [ 176.530908] ? sched_clock_cpu+0x31/0x1e0 [ 176.531934] __do_softirq+0x1bc/0x6d9 [ 176.532851] irq_exit+0x10f/0x130 [ 176.533709] smp_apic_timer_interrupt+0x176/0x400 [ 176.534800] apic_timer_interrupt+0xf/0x20 [ 176.535789] </IRQ> [ 176.536434] RIP: 0010:__x86_indirect_thunk_rdx+0x0/0x20 [ 176.537618] Code: 84 00 00 00 00 00 0f 1f 40 00 e8 07 00 00 00 f3 90 0f ae e8 eb f9 48 89 0c 24 c3 66 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 40 00 <e8> 07 00 00 00 f3 90 0f ae e8 eb f9 48 89 14 24 c3 66 66 2e 0f 1f [ 176.541507] RSP: 0018:ffff88805723f7f0 EFLAGS: 00000297 ORIG_RAX: ffffffffffffff13 [ 176.543287] RAX: 0000000000000005 RBX: ffff88805723f8c8 RCX: 0000000000000000 [ 176.545047] RDX: ffffffff968efe39 RSI: 0000000000000001 RDI: 0000000000000001 [ 176.546782] RBP: 1ffff1100ae47f06 R08: ffffffff9d146598 R09: ffffffff9d14659c [ 176.548526] R10: 00000000000ebfb5 R11: ffff88805723f8fd R12: 0000000000000001 [ 176.550261] R13: ffff88805723f910 R14: ffff88805723f900 R15: ffff88805723f918 [ 176.554447] ? unwind_next_frame+0x8b9/0x1b20 [ 176.555638] entry_SYSCALL_64_after_hwframe+0x44/0xa9 [ 176.556907] RIP: 5bd3e400:0x2 [ 176.557788] Code: Bad RIP value. [ 176.558712] RSP: 5723f940:ffff88805723f950 EFLAGS: 1010000000000 ORIG_RAX: 0000000000000000 [ 176.560829] RAX: ffff88805723f8d0 RBX: ffff88805723f8d8 RCX: ffff88805723ff58 [ 176.562513] RDX: 0000000000000001 RSI: ffff888057238000 RDI: ffff888057240000 [ 176.564191] RBP: ffffffff9cbbef86 R08: ffffffff968ef580 R09: ffffffff9b9d951c [ 176.565852] R10: 0000000041b58ab3 R11: ffffffff96a0e86e R12: ffff88805723f8fd [ 176.567548] R13: ffffffff9cbbef82 R14: ffff88805723f8fd R15: ffff88805723ff58 [ 176.569293] ? __kernel_text_address+0xe/0x30 [ 176.570492] ? unwind_get_return_address_ptr+0xb0/0xb0 [ 176.571848] ? __save_stack_trace+0x73/0xd0 [ 176.573048] ? entry_SYSCALL_64_after_hwframe+0x44/0xa9 [ 176.574368] ? save_stack+0x32/0xb0 [ 176.575362] ? __kasan_kmalloc+0xa0/0xd0 [ 176.576631] ? kmem_cache_alloc+0xb7/0x1b0 [ 176.577739] ? anon_vma_fork+0xcf/0x5b0 [ 176.578800] ? copy_process+0x4f25/0x5a90 [ 176.580208] ? _do_fork+0x13f/0x840 [ 176.581224] ? do_syscall_64+0x96/0x8fb [ 176.582300] ? entry_SYSCALL_64_after_hwframe+0x44/0xa9 [ 176.583654] ? kmem_cache_alloc+0xb7/0x1b0 [ 176.584804] ? vm_area_dup+0x1e/0x180 [ 176.585884] ? copy_process+0x4b1b/0x5a90 [ 176.587146] ? _do_fork+0x13f/0x840 [ 176.588159] ? do_syscall_64+0x96/0x8fb [ 176.589221] ? entry_SYSCALL_64_after_hwframe+0x44/0xa9 [ 176.590549] ? copy_page_range+0xe56/0x1ad0 [ 176.591711] ? kasan_unpoison_shadow+0x30/0x40 [ 176.592933] ? __kasan_kmalloc+0xa0/0xd0 [ 176.594114] ? kasan_unpoison_shadow+0x30/0x40 [ 176.595168] ? __kasan_kmalloc+0xa0/0xd0 [ 176.596302] ? anon_vma_fork+0xcf/0x5b0 [ 176.597241] ? kmem_cache_alloc+0xb7/0x1b0 [ 176.598225] ? anon_vma_fork+0xcf/0x5b0 [ 176.599168] ? copy_process+0x4f25/0x5a90 [ 176.600255] ? __cleanup_sighand+0x40/0x40 [ 176.601240] ? __might_fault+0x87/0xb0 To reproduce: # build kernel cd linux cp config-5.0.4-00048-g0830cf6 .config make HOSTCC=gcc-7 CC=gcc-7 ARCH=x86_64 olddefconfig make HOSTCC=gcc-7 CC=gcc-7 ARCH=x86_64 prepare make HOSTCC=gcc-7 CC=gcc-7 ARCH=x86_64 modules_prepare make HOSTCC=gcc-7 CC=gcc-7 ARCH=x86_64 SHELL=/bin/bash make HOSTCC=gcc-7 CC=gcc-7 ARCH=x86_64 bzImage git clone https://github.com/intel/lkp-tests.git cd lkp-tests bin/lkp qemu -k <bzImage> job-script # job-script is attached in this email Thanks, Rong Chen

3 years, 2 months

2
1
0 / 0

[drm/bochs] 7a0483ac4f: BUG:unable_to_handle_kernel

by kernel test robot

FYI, we noticed the following commit (built with gcc-6): commit: 7a0483ac4ffca4998945c159b28afdde8353cc84 ("drm/bochs: switch to generic drm fbdev emulation") https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master in testcase: trinity with following parameters: runtime: 300s test-description: Trinity is a linux system call fuzz tester. test-url: http://codemonkey.org.uk/projects/trinity/ on test machine: qemu-system-x86_64 -enable-kvm -cpu SandyBridge -smp 2 -m 2G caused below changes (please refer to attached dmesg/kmsg for entire log/backtrace): +-----------------------------------------------------+------------+------------+ | | 132a45340a | 7a0483ac4f | +-----------------------------------------------------+------------+------------+ | boot_successes | 48 | 6 | | boot_failures | 0 | 44 | | BUG:unable_to_handle_kernel | 0 | 44 | | Oops:#[##] | 0 | 44 | | RIP:restore_fbdev_mode | 0 | 44 | | WARNING:at_kernel/locking/lockdep.c:#lock_downgrade | 0 | 14 | | RIP:lock_downgrade | 0 | 14 | | Kernel_panic-not_syncing:Fatal_exception | 0 | 44 | +-----------------------------------------------------+------------+------------+ If you fix the issue, kindly add following tag Reported-by: kernel test robot <lkp(a)intel.com> [ 9.996031] BUG: unable to handle kernel NULL pointer dereference at 0000000000000020 [ 9.996169] ------------[ cut here ]------------ [ 9.996991] #PF error: [normal kernel read fault] [ 9.997548] downgrading a read lock [ 9.997571] WARNING: CPU: 0 PID: 479 at kernel/locking/lockdep.c:3553 lock_downgrade+0xd1/0x190 [ 9.998095] PGD 800000001c3f1067 P4D 800000001c3f1067 PUD 1c148067 PMD 0 [ 9.998516] Modules linked in: [ 9.999778] Oops: 0000 [#1] PREEMPT SMP PTI [ 10.000594] CPU: 0 PID: 479 Comm: mkdir Not tainted 5.0.0-rc1-00140-g7a0483a #1 [ 10.000967] CPU: 1 PID: 239 Comm: plymouthd Not tainted 5.0.0-rc1-00140-g7a0483a #1 [ 10.001478] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.10.2-1 04/01/2014 [ 10.002351] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.10.2-1 04/01/2014 [ 10.003269] RIP: 0010:lock_downgrade+0xd1/0x190 [ 10.004274] RIP: 0010:restore_fbdev_mode+0x1b/0x1e0 [ 10.005272] Code: 8b 54 24 0c 89 93 28 0b 00 00 48 8b 08 48 89 8b 20 0b 00 00 f6 40 32 03 74 1a 48 c7 c7 10 56 53 82 48 89 04 24 e8 8f 60 fa ff <0f> 0b 8b 54 24 0c 48 8b 04 24 0f b6 48 32 4c 89 60 08 44 89 ee 48 [ 10.005824] Code: 41 5c 41 5d 41 5e 41 5f c3 66 0f 1f 44 00 00 41 57 41 56 41 55 41 54 49 89 fc 55 53 48 83 ec 08 e8 2a f8 ba ff 4d 8b 6c 24 40 <49> 8b 55 20 41 8b 45 50 25 00 00 01 00 85 82 78 01 00 00 0f 85 47 [ 10.006415] RSP: 0018:ffffc9000066fe28 EFLAGS: 00010096 [ 10.008634] RSP: 0018:ffffc9000045bdb0 EFLAGS: 00010293 [ 10.010881] RAX: 0000000000000017 RBX: ffff88802cabc000 RCX: 0000000000000001 [ 10.011513] RAX: 0000000000000000 RBX: ffff88801c1e6d40 RCX: 0000000000000000 [ 10.012142] RDX: ffff88802cabc000 RSI: ffffffff81132f29 RDI: 00000000ffffffff [ 10.012999] RDX: ffff88802f834000 RSI: ffffffff815d0886 RDI: ffff88801c1e6d40 [ 10.013855] RBP: ffff88802d830128 R08: 00000002c324869d R09: ffffc9000066fdb8 [ 10.014712] RBP: 0000000000000000 R08: 0000000000000000 R09: 0000000000000001 [ 10.015571] R10: 0000000000000001 R11: 0000000000000000 R12: ffffffff811f1db8 [ 10.016430] R10: ffffc9000045bdf0 R11: 0000000000000001 R12: ffff88801c1e6d40 [ 10.017283] R13: 0000000000000003 R14: 0000000000000246 R15: 00007fa7532b9000 [ 10.018145] R13: 0000000000000000 R14: ffff88801c255278 R15: ffffffff815d2680 [ 10.019010] FS: 00007fa7532b47c0(0000) GS:ffff888029600000(0000) knlGS:0000000000000000 [ 10.019885] FS: 00007f8ba8290700(0000) GS:ffff888029a00000(0000) knlGS:0000000000000000 [ 10.020747] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 10.021715] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 10.022684] CR2: 00007fa7532b8000 CR3: 000000002d86a000 CR4: 00000000000406f0 [ 10.023390] CR2: 0000000000000020 CR3: 000000001c16e000 CR4: 00000000000406e0 [ 10.024078] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 10.024927] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 10.025781] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 10.026632] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 10.027485] Call Trace: [ 10.028342] Call Trace: [ 10.029195] downgrade_write+0x12/0x80 [ 10.029501] ? drm_fb_helper_lastclose+0x20/0x20 [ 10.029820] __do_munmap+0x3d8/0x5c0 [ 10.030273] drm_fb_helper_restore_fbdev_mode_unlocked+0x4a/0xa0 [ 10.030835] __vm_munmap+0x67/0xc0 [ 10.031268] drm_fbdev_client_restore+0x11/0x20 [ 10.031969] __x64_sys_munmap+0x1b/0x20 [ 10.032390] drm_client_dev_restore+0xa8/0xe0 [ 10.032936] do_syscall_64+0xb5/0x200 [ 10.033405] drm_release+0xc0/0x120 [ 10.033933] entry_SYSCALL_64_after_hwframe+0x44/0xa9 [ 10.034381] ? drm_lastclose+0x100/0x100 [ 10.034806] RIP: 0033:0x7fa752bab897 [ 10.035422] __fput+0x15c/0x250 [ 10.035904] Code: f0 ff ff 73 01 c3 48 8b 0d a6 75 2c 00 31 d2 48 29 c2 64 89 11 48 83 c8 ff eb ea 90 90 90 90 90 90 90 90 b8 0b 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 79 75 2c 00 31 d2 48 29 c2 64 [ 10.036344] task_work_run+0x9d/0xd0 [ 10.036726] RSP: 002b:00007ffc66b50a38 EFLAGS: 00000206 ORIG_RAX: 000000000000000b [ 10.038945] exit_to_usermode_loop+0x7a/0xc0 [ 10.039383] RAX: ffffffffffffffda RBX: 000000000060c010 RCX: 00007fa752bab897 [ 10.040280] do_syscall_64+0x1c9/0x200 [ 10.040804] RDX: 0000000000000000 RSI: 0000000000001000 RDI: 00007fa7532b8000 [ 10.041662] entry_SYSCALL_64_after_hwframe+0x44/0xa9 [ 10.042113] RBP: 0000000000000000 R08: 0000000000000010 R09: 00000000ffffffff [ 10.042964] RIP: 0033:0x7f8ba7973040 [ 10.043574] R10: 00007ffc66b50840 R11: 0000000000000206 R12: 0000000000000000 [ 10.044426] Code: 40 75 0b 31 c0 48 83 c4 08 e9 0c ff ff ff 48 8d 3d c5 99 09 00 e8 a0 3f 02 00 83 3d 9d 71 2d 00 00 75 10 b8 03 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 31 c3 48 83 ec 08 e8 9e b1 01 00 48 89 04 24 [ 10.044860] R13: 00007ffc66b50bb8 R14: 00007fa753096d00 R15: 0000000000000000 [ 10.045722] RSP: 002b:00007ffe098cf568 EFLAGS: 00000246 ORIG_RAX: 0000000000000003 [ 10.047994] ---[ end trace 2efaa9fea56af290 ]--- To reproduce: # build kernel cd linux cp config-5.0.0-rc1-00140-g7a0483a .config make HOSTCC=gcc-6 CC=gcc-6 ARCH=x86_64 olddefconfig make HOSTCC=gcc-6 CC=gcc-6 ARCH=x86_64 prepare make HOSTCC=gcc-6 CC=gcc-6 ARCH=x86_64 modules_prepare make HOSTCC=gcc-6 CC=gcc-6 ARCH=x86_64 SHELL=/bin/bash make HOSTCC=gcc-6 CC=gcc-6 ARCH=x86_64 bzImage git clone https://github.com/intel/lkp-tests.git cd lkp-tests bin/lkp qemu -k <bzImage> job-script # job-script is attached in this email Thanks, lkp

3 years, 2 months

1
0
0 / 0

[futex] 5a07168d8d: will-it-scale.per_thread_ops -2.7% regression

by kernel test robot

Greeting, FYI, we noticed a -2.7% regression of will-it-scale.per_thread_ops due to commit: commit: 5a07168d8d89b00fe1760120714378175b3ef992 ("futex: Ensure that futex address is aligned in handle_futex_death()") https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master in testcase: will-it-scale on test machine: 88 threads Intel(R) Xeon(R) CPU E5-2699 v4 @ 2.20GHz with 64G memory with following parameters: nr_task: 100% mode: thread test: futex3 cpufreq_governor: performance ucode: 0xb00002e 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 In addition to that, the commit also has significant impact on the following tests: +------------------+----------------------------------------------------------------------+ | testcase: change | will-it-scale: will-it-scale.per_process_ops -1.6% regression | | test machine | 88 threads Intel(R) Xeon(R) CPU E5-2699 v4 @ 2.20GHz with 64G memory | | test parameters | cpufreq_governor=performance | | | mode=process | | | nr_task=100% | | | test=futex3 | | | ucode=0xb00002e | +------------------+----------------------------------------------------------------------+ Details are as below: --------------------------------------------------------------------------------------------------> To reproduce: git clone https://github.com/intel/lkp-tests.git cd lkp-tests bin/lkp install job.yaml # job file is attached in this email bin/lkp run job.yaml ========================================================================================= compiler/cpufreq_governor/kconfig/mode/nr_task/rootfs/tbox_group/test/testcase/ucode: gcc-7/performance/x86_64-rhel-7.6/thread/100%/debian-x86_64-2018-04-03.cgz/lkp-bdw-ep3b/futex3/will-it-scale/0xb00002e commit: 82efcab3b9 ("workqueue: Only unregister a registered lockdep key") 5a07168d8d ("futex: Ensure that futex address is aligned in handle_futex_death()") 82efcab3b9f3ef59 5a07168d8d89b00fe1760120714 ---------------- --------------------------- %stddev %change %stddev \ | \ 2979618 -2.7% 2898944 will-it-scale.per_thread_ops 16627 +2.4% 17021 will-it-scale.time.system_time 9856 ± 3% -4.0% 9460 will-it-scale.time.user_time 2.622e+08 -2.7% 2.551e+08 will-it-scale.workload 23771 ± 85% +95.5% 46469 ± 41% numa-meminfo.node0.Shmem 5942 ± 85% +95.6% 11624 ± 41% numa-vmstat.node0.nr_shmem 1216 +2.3% 1244 proc-vmstat.nr_page_table_pages 8525 ± 4% -7.7% 7870 ± 5% slabinfo.kmalloc-512.active_objs 8614 ± 4% -7.8% 7940 ± 6% slabinfo.kmalloc-512.num_objs 61.50 +2.4% 63.00 vmstat.cpu.sy 36.50 ± 2% -4.1% 35.00 vmstat.cpu.us 27376 ± 3% -4.8% 26064 ± 3% softirqs.CPU42.RCU 25222 ± 20% -14.4% 21590 ± 4% softirqs.CPU82.RCU 22490 ± 5% -3.9% 21617 ± 4% softirqs.CPU86.RCU 250320 ± 32% +69.7% 424902 ± 8% numa-numastat.node0.local_node 265927 ± 33% +63.3% 434388 ± 9% numa-numastat.node0.numa_hit 353210 ± 23% -49.9% 177033 ± 20% numa-numastat.node1.local_node 366101 ± 24% -46.4% 196058 ± 21% numa-numastat.node1.numa_hit 33.44 ± 6% -14.4% 28.63 ± 16% sched_debug.cfs_rq:/.load_avg.stddev 155.87 ± 24% -43.6% 87.89 ± 81% sched_debug.cfs_rq:/.removed.runnable_sum.avg 1083 ± 13% -39.3% 658.24 ± 73% sched_debug.cfs_rq:/.removed.runnable_sum.stddev 1.48 ± 19% -42.9% 0.84 ± 76% sched_debug.cfs_rq:/.removed.util_avg.avg 10.53 ± 8% -38.0% 6.53 ± 71% sched_debug.cfs_rq:/.removed.util_avg.stddev 2.29 ± 7% -4.9% 2.18 ± 8% sched_debug.cfs_rq:/.runnable_load_avg.stddev 401.30 ± 59% +98.8% 797.72 ± 4% sched_debug.cfs_rq:/.util_est_enqueued.avg 1.763e+10 -2.7% 1.715e+10 perf-stat.i.branch-instructions 2.651e+08 -2.7% 2.579e+08 perf-stat.i.branch-misses 1.97 +2.8% 2.02 perf-stat.i.cpi 3.073e+10 -2.7% 2.99e+10 perf-stat.i.dTLB-loads 2.31e+10 -2.7% 2.247e+10 perf-stat.i.dTLB-stores 1.24e+11 -2.7% 1.207e+11 perf-stat.i.instructions 0.51 -2.7% 0.49 perf-stat.i.ipc 1.97 +2.8% 2.02 perf-stat.overall.cpi 0.51 -2.7% 0.49 perf-stat.overall.ipc 1.757e+10 -2.7% 1.709e+10 perf-stat.ps.branch-instructions 2.643e+08 -2.7% 2.57e+08 perf-stat.ps.branch-misses 3.063e+10 -2.7% 2.98e+10 perf-stat.ps.dTLB-loads 2.302e+10 -2.7% 2.24e+10 perf-stat.ps.dTLB-stores 1.236e+11 -2.7% 1.202e+11 perf-stat.ps.instructions 3.732e+13 -2.6% 3.635e+13 perf-stat.total.instructions 38.08 -1.0 37.05 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64.syscall 27.53 -0.4 27.12 perf-profile.calltrace.cycles-pp.syscall_return_via_sysret.syscall 3.28 -0.1 3.16 perf-profile.calltrace.cycles-pp.testcase 2.53 +0.1 2.66 perf-profile.calltrace.cycles-pp.get_futex_key_refs.get_futex_key.futex_wake.do_futex.__x64_sys_futex 6.02 +0.3 6.28 perf-profile.calltrace.cycles-pp.get_futex_key.futex_wake.do_futex.__x64_sys_futex.do_syscall_64 3.05 ± 2% +0.7 3.71 perf-profile.calltrace.cycles-pp.hash_futex.futex_wake.do_futex.__x64_sys_futex.do_syscall_64 13.11 +0.9 14.04 perf-profile.calltrace.cycles-pp.futex_wake.do_futex.__x64_sys_futex.do_syscall_64.entry_SYSCALL_64_after_hwframe 16.74 +1.0 17.75 perf-profile.calltrace.cycles-pp.do_futex.__x64_sys_futex.do_syscall_64.entry_SYSCALL_64_after_hwframe.syscall 29.05 +1.6 30.63 perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.syscall 30.35 +1.6 31.95 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.syscall 22.91 +1.6 24.54 perf-profile.calltrace.cycles-pp.__x64_sys_futex.do_syscall_64.entry_SYSCALL_64_after_hwframe.syscall 33.87 -0.9 32.98 perf-profile.children.cycles-pp.entry_SYSCALL_64 31.87 -0.5 31.32 perf-profile.children.cycles-pp.syscall_return_via_sysret 2.36 -0.1 2.27 perf-profile.children.cycles-pp.testcase 98.54 +0.0 98.58 perf-profile.children.cycles-pp.syscall 2.54 +0.1 2.67 perf-profile.children.cycles-pp.get_futex_key_refs 6.22 +0.3 6.51 perf-profile.children.cycles-pp.get_futex_key 3.07 ± 2% +0.7 3.76 perf-profile.children.cycles-pp.hash_futex 13.47 +1.0 14.47 perf-profile.children.cycles-pp.futex_wake 16.91 +1.1 18.00 perf-profile.children.cycles-pp.do_futex 31.04 +1.5 32.57 perf-profile.children.cycles-pp.entry_SYSCALL_64_after_hwframe 28.59 +1.6 30.21 perf-profile.children.cycles-pp.do_syscall_64 22.58 +1.7 24.25 perf-profile.children.cycles-pp.__x64_sys_futex 29.64 -0.8 28.88 perf-profile.self.cycles-pp.entry_SYSCALL_64 31.84 -0.5 31.30 perf-profile.self.cycles-pp.syscall_return_via_sysret 5.62 -0.2 5.41 perf-profile.self.cycles-pp.syscall 3.05 -0.1 2.92 perf-profile.self.cycles-pp.entry_SYSCALL_64_after_hwframe 1.41 -0.1 1.35 perf-profile.self.cycles-pp.testcase 4.02 +0.1 4.09 perf-profile.self.cycles-pp.futex_wake 3.42 +0.1 3.51 perf-profile.self.cycles-pp.do_futex 2.50 +0.1 2.61 perf-profile.self.cycles-pp.get_futex_key_refs 3.51 ± 3% +0.2 3.72 ± 2% perf-profile.self.cycles-pp.get_futex_key 5.23 +0.6 5.82 perf-profile.self.cycles-pp.__x64_sys_futex 3.04 ± 2% +0.6 3.66 perf-profile.self.cycles-pp.hash_futex 1367 ± 95% -68.3% 433.67 ± 81% interrupts.36:PCI-MSI.3145733-edge.eth0-TxRx-4 503.50 ± 39% -45.6% 274.00 ± 55% interrupts.39:PCI-MSI.3145736-edge.eth0-TxRx-7 4918 ± 34% +60.4% 7889 interrupts.CPU14.NMI:Non-maskable_interrupts 4918 ± 34% +60.4% 7889 interrupts.CPU14.PMI:Performance_monitoring_interrupts 1367 ± 95% -68.3% 433.67 ± 81% interrupts.CPU15.36:PCI-MSI.3145733-edge.eth0-TxRx-4 200.75 ± 39% +197.7% 597.67 ± 13% interrupts.CPU15.RES:Rescheduling_interrupts 204.50 ± 80% +251.1% 718.00 ± 52% interrupts.CPU16.RES:Rescheduling_interrupts 503.50 ± 39% -45.6% 274.00 ± 55% interrupts.CPU18.39:PCI-MSI.3145736-edge.eth0-TxRx-7 4912 -22.9% 3788 ± 32% interrupts.CPU19.CAL:Function_call_interrupts 1753 ± 3% -10.6% 1567 ± 11% interrupts.CPU21.TLB:TLB_shootdowns 4909 -21.2% 3869 ± 31% interrupts.CPU22.CAL:Function_call_interrupts 473.25 ± 46% -83.3% 79.00 ± 37% interrupts.CPU24.RES:Rescheduling_interrupts 698.75 ± 55% -63.9% 252.00 ±106% interrupts.CPU26.RES:Rescheduling_interrupts 382.50 ± 53% -64.9% 134.33 ± 47% interrupts.CPU28.RES:Rescheduling_interrupts 342.00 ± 48% +146.4% 842.67 ± 26% interrupts.CPU3.RES:Rescheduling_interrupts 572.25 ± 97% -77.6% 128.00 ± 49% interrupts.CPU33.RES:Rescheduling_interrupts 374.75 ± 62% -79.9% 75.33 ± 50% interrupts.CPU34.RES:Rescheduling_interrupts 1306 ± 80% -91.8% 107.67 ± 25% interrupts.CPU35.RES:Rescheduling_interrupts 1377 ± 57% -61.6% 529.33 ±128% interrupts.CPU38.RES:Rescheduling_interrupts 427.00 ± 67% -73.2% 114.33 ± 51% interrupts.CPU39.RES:Rescheduling_interrupts 343.25 ±117% +794.5% 3070 ± 36% interrupts.CPU4.RES:Rescheduling_interrupts 435.25 ± 76% -86.4% 59.00 ± 47% interrupts.CPU40.RES:Rescheduling_interrupts 15.50 ± 33% +5132.3% 811.00 ±135% interrupts.CPU46.RES:Rescheduling_interrupts 22.50 ± 58% +6574.1% 1501 ±102% interrupts.CPU53.RES:Rescheduling_interrupts 27.00 ± 52% +685.2% 212.00 ±102% interrupts.CPU54.RES:Rescheduling_interrupts 15.00 ± 48% +2224.4% 348.67 ±123% interrupts.CPU62.RES:Rescheduling_interrupts 7918 -33.4% 5276 ± 35% interrupts.CPU70.NMI:Non-maskable_interrupts 7918 -33.4% 5276 ± 35% interrupts.CPU70.PMI:Performance_monitoring_interrupts 82.50 ± 95% -68.1% 26.33 ± 39% interrupts.CPU70.RES:Rescheduling_interrupts 7903 -33.5% 5258 ± 35% interrupts.CPU71.NMI:Non-maskable_interrupts 7903 -33.5% 5258 ± 35% interrupts.CPU71.PMI:Performance_monitoring_interrupts 287.00 ±123% -91.4% 24.67 ± 5% interrupts.CPU71.RES:Rescheduling_interrupts 6921 ± 24% -24.1% 5256 ± 34% interrupts.CPU74.NMI:Non-maskable_interrupts 6921 ± 24% -24.1% 5256 ± 34% interrupts.CPU74.PMI:Performance_monitoring_interrupts 6898 ± 24% -24.1% 5238 ± 35% interrupts.CPU75.NMI:Non-maskable_interrupts 6898 ± 24% -24.1% 5238 ± 35% interrupts.CPU75.PMI:Performance_monitoring_interrupts 61.50 ± 72% -86.4% 8.33 ± 40% interrupts.CPU75.RES:Rescheduling_interrupts 6919 ± 24% -24.1% 5252 ± 35% interrupts.CPU78.NMI:Non-maskable_interrupts 6919 ± 24% -24.1% 5252 ± 35% interrupts.CPU78.PMI:Performance_monitoring_interrupts 593.25 ± 67% +188.3% 1710 ± 38% interrupts.CPU8.RES:Rescheduling_interrupts 63.00 ± 77% -63.5% 23.00 ± 57% interrupts.CPU82.RES:Rescheduling_interrupts 280.25 ±136% -93.3% 18.67 ± 28% interrupts.CPU83.RES:Rescheduling_interrupts 57.00 ± 60% -83.0% 9.67 ± 48% interrupts.CPU84.RES:Rescheduling_interrupts will-it-scale.per_thread_ops 3.5e+06 +-+---------------------------------------------------------------+ | | 3e+06 O-++..O O O O.O..O O O.+..O..O O +..O..+..+ +..+..+.+..| | : : : : : : : : : | 2.5e+06 +-+ : :: : : : : : : : | | : : : : : : : : : : | 2e+06 +-+ : : : : : : : : : : | | : : : : : : : : : : | 1.5e+06 +-+ : : : : : : : : : : | | : : : : : : : : : : | 1e+06 +-+ : : : : : : : : : : | | :: : : : : : : :: | 500000 +-+ :: :: :: : :: | | : : : : : | 0 +-+O----------O---------------O----------O------------------------+ will-it-scale.workload 3e+08 +-+---------------------------------------------------------------+ | | 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 +-+O----------O---------------O----------O------------------------+ [*] bisect-good sample [O] bisect-bad sample *************************************************************************************************** lkp-bdw-ep3b: 88 threads Intel(R) Xeon(R) CPU E5-2699 v4 @ 2.20GHz with 64G memory ========================================================================================= compiler/cpufreq_governor/kconfig/mode/nr_task/rootfs/tbox_group/test/testcase/ucode: gcc-7/performance/x86_64-rhel-7.6/process/100%/debian-x86_64-2018-04-03.cgz/lkp-bdw-ep3b/futex3/will-it-scale/0xb00002e commit: 82efcab3b9 ("workqueue: Only unregister a registered lockdep key") 5a07168d8d ("futex: Ensure that futex address is aligned in handle_futex_death()") 82efcab3b9f3ef59 5a07168d8d89b00fe1760120714 ---------------- --------------------------- fail:runs %reproduction fail:runs | | | 1:4 -25% :4 dmesg.WARNING:at#for_ip_interrupt_entry/0x 1:4 -25% :4 kmsg.DHCP/BOOTP:Reply_not_for_us_on_eth#,op[#]xid[#] %stddev %change %stddev \ | \ 2967197 -1.6% 2921017 will-it-scale.per_process_ops 2.611e+08 -1.6% 2.57e+08 will-it-scale.workload 6581105 ± 7% -10.8% 5868555 meminfo.DirectMap2M 0.00 ± 36% +0.0 0.00 ± 64% mpstat.cpu.all.soft% 61.00 +1.6% 62.00 vmstat.cpu.sy 36.00 -2.8% 35.00 vmstat.cpu.us 1355725 ± 87% -74.5% 345713 ± 10% cpuidle.C1.time 88925 ±129% -94.6% 4801 ± 25% cpuidle.C1.usage 86713 ±104% -88.5% 9950 ± 9% cpuidle.C1E.usage 86486 ±132% -95.3% 4102 ± 26% turbostat.C1 82307 ±107% -88.9% 9119 ± 12% turbostat.C1E 0.36 ± 19% +34.9% 0.49 ± 21% turbostat.CPU%c1 26465 ± 22% +68.6% 44634 ± 19% numa-vmstat.node0.nr_active_anon 26423 ± 22% +49.4% 39476 ± 16% numa-vmstat.node0.nr_anon_pages 26465 ± 22% +68.6% 44634 ± 19% numa-vmstat.node0.nr_zone_active_anon 51873 ± 13% -35.6% 33430 ± 25% numa-vmstat.node1.nr_active_anon 51873 ± 13% -35.6% 33430 ± 25% numa-vmstat.node1.nr_zone_active_anon 758.00 ± 41% +33.3% 1010 ± 29% interrupts.32:PCI-MSI.3145729-edge.eth0-TxRx-0 758.00 ± 41% +33.3% 1010 ± 29% interrupts.CPU11.32:PCI-MSI.3145729-edge.eth0-TxRx-0 1436 ± 13% -52.7% 678.75 ± 66% interrupts.CPU22.RES:Rescheduling_interrupts 16.67 ±107% +810.5% 151.75 ± 86% interrupts.CPU7.RES:Rescheduling_interrupts 28.33 ± 21% +764.7% 245.00 ± 87% interrupts.CPU80.RES:Rescheduling_interrupts 23.67 ±108% +300.4% 94.75 ± 93% interrupts.CPU9.RES:Rescheduling_interrupts 1936 ± 4% -19.1% 1567 ± 6% slabinfo.UNIX.active_objs 1936 ± 4% -19.1% 1567 ± 6% slabinfo.UNIX.num_objs 3665 ± 2% -15.6% 3094 ± 5% slabinfo.sock_inode_cache.active_objs 3665 ± 2% -15.6% 3094 ± 5% slabinfo.sock_inode_cache.num_objs 1397 ± 7% -13.0% 1215 ± 8% slabinfo.task_group.active_objs 1397 ± 7% -13.0% 1215 ± 8% slabinfo.task_group.num_objs 105945 ± 22% +68.5% 178555 ± 19% numa-meminfo.node0.Active 105854 ± 22% +68.6% 178487 ± 19% numa-meminfo.node0.Active(anon) 60862 ± 41% +87.3% 114016 ± 14% numa-meminfo.node0.AnonHugePages 105703 ± 22% +49.3% 157859 ± 16% numa-meminfo.node0.AnonPages 207473 ± 13% -35.5% 133847 ± 25% numa-meminfo.node1.Active 207428 ± 13% -35.5% 133779 ± 25% numa-meminfo.node1.Active(anon) 112468 ± 23% -47.0% 59623 ± 27% numa-meminfo.node1.AnonHugePages 22164 +12.9% 25013 ± 7% softirqs.CPU10.RCU 23983 ± 9% +8.1% 25924 ± 6% softirqs.CPU19.RCU 98574 ± 5% +7.1% 105614 ± 7% softirqs.CPU21.TIMER 25231 ± 7% +7.7% 27186 ± 6% softirqs.CPU30.RCU 28403 ± 5% +29.2% 36687 ± 10% softirqs.CPU36.RCU 27015 ± 6% +6.3% 28706 ± 4% softirqs.CPU42.RCU 22681 ± 4% +18.4% 26865 ± 19% softirqs.CPU5.RCU 21389 ± 8% +14.8% 24549 ± 5% softirqs.CPU54.RCU 98111 ± 5% +3.8% 101866 ± 4% softirqs.CPU57.TIMER 23135 ± 3% +8.6% 25116 ± 4% softirqs.CPU6.RCU 97209 ± 5% +39.9% 136028 ± 25% softirqs.CPU65.TIMER 24333 ± 3% +15.7% 28162 ± 14% softirqs.CPU7.RCU 21320 ± 7% +7.0% 22804 ± 8% softirqs.CPU78.RCU 23119 ± 3% +18.7% 27451 ± 17% softirqs.CPU9.RCU 1.741e+10 -1.3% 1.719e+10 perf-stat.i.branch-instructions 2.62e+08 -1.3% 2.586e+08 perf-stat.i.branch-misses 3.036e+10 -1.3% 2.997e+10 perf-stat.i.dTLB-loads 2.282e+10 -1.3% 2.253e+10 perf-stat.i.dTLB-stores 3.765e+08 ± 4% -9.3% 3.415e+08 perf-stat.i.iTLB-load-misses 1.225e+11 -1.3% 1.209e+11 perf-stat.i.instructions 333.48 ± 4% +6.9% 356.47 perf-stat.i.instructions-per-iTLB-miss 0.50 -1.4% 0.50 perf-stat.i.ipc 1.98 +1.6% 2.01 perf-stat.overall.cpi 326.21 ± 5% +8.6% 354.20 perf-stat.overall.instructions-per-iTLB-miss 0.51 -1.5% 0.50 perf-stat.overall.ipc 1.735e+10 -1.3% 1.713e+10 perf-stat.ps.branch-instructions 2.611e+08 -1.3% 2.577e+08 perf-stat.ps.branch-misses 3.025e+10 -1.3% 2.987e+10 perf-stat.ps.dTLB-loads 2.274e+10 -1.3% 2.245e+10 perf-stat.ps.dTLB-stores 3.753e+08 ± 4% -9.3% 3.404e+08 perf-stat.ps.iTLB-load-misses 1.221e+11 -1.3% 1.205e+11 perf-stat.ps.instructions 3.727e+13 -1.5% 3.67e+13 perf-stat.total.instructions 38.07 -0.4 37.66 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64.syscall 28.03 -0.3 27.68 perf-profile.calltrace.cycles-pp.syscall_return_via_sysret.syscall 3.35 -0.1 3.22 perf-profile.calltrace.cycles-pp.hash_futex.futex_wake.do_futex.__x64_sys_futex.do_syscall_64 3.24 -0.1 3.14 perf-profile.calltrace.cycles-pp.testcase 97.84 +0.1 97.92 perf-profile.calltrace.cycles-pp.syscall 16.28 +0.5 16.82 perf-profile.calltrace.cycles-pp.do_futex.__x64_sys_futex.do_syscall_64.entry_SYSCALL_64_after_hwframe.syscall 28.49 +0.9 29.34 perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.syscall 29.73 +0.9 30.62 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.syscall 22.21 +0.9 23.12 perf-profile.calltrace.cycles-pp.__x64_sys_futex.do_syscall_64.entry_SYSCALL_64_after_hwframe.syscall 32.37 -0.4 31.94 perf-profile.children.cycles-pp.syscall_return_via_sysret 33.97 -0.3 33.66 perf-profile.children.cycles-pp.entry_SYSCALL_64 3.37 -0.1 3.24 perf-profile.children.cycles-pp.hash_futex 2.33 -0.1 2.25 perf-profile.children.cycles-pp.testcase 98.56 +0.1 98.61 perf-profile.children.cycles-pp.syscall 16.39 +0.6 16.98 perf-profile.children.cycles-pp.do_futex 30.45 +0.9 31.31 perf-profile.children.cycles-pp.entry_SYSCALL_64_after_hwframe 27.99 +0.9 28.88 perf-profile.children.cycles-pp.do_syscall_64 21.88 +1.0 22.85 perf-profile.children.cycles-pp.__x64_sys_futex 32.33 -0.4 31.90 perf-profile.self.cycles-pp.syscall_return_via_sysret 29.86 -0.2 29.63 perf-profile.self.cycles-pp.entry_SYSCALL_64 5.46 -0.1 5.33 perf-profile.self.cycles-pp.syscall 3.28 -0.1 3.18 perf-profile.self.cycles-pp.hash_futex 3.11 -0.1 3.03 perf-profile.self.cycles-pp.entry_SYSCALL_64_after_hwframe 2.57 -0.1 2.52 perf-profile.self.cycles-pp.get_futex_key_refs 1.39 -0.1 1.33 perf-profile.self.cycles-pp.testcase 4.00 +0.1 4.10 perf-profile.self.cycles-pp.futex_wake 5.06 +0.4 5.47 perf-profile.self.cycles-pp.__x64_sys_futex 2.62 +0.6 3.19 perf-profile.self.cycles-pp.do_futex Disclaimer: Results have been estimated based on internal Intel analysis and are provided for informational purposes only. Any difference in system hardware or software design or configuration may affect actual performance. Thanks, Rong Chen

3 years, 2 months

1
0
0 / 0

[drm/i915/fbdev] 09ded8af57: dmesg.RIP:drm_setup_crtcs[drm_kms_helper]

by kernel test robot

FYI, we noticed the following commit (built with gcc-7): commit: 09ded8af57bcef7287b8242087d3e7556380de62 ("drm/i915/fbdev: Move intel_fb_initial_config() to fbdev helper") https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git master in testcase: plzip with following parameters: nr_threads: 100% cpufreq_governor: performance ucode: 0x12 on test machine: 144 threads Intel(R) Xeon(R) CPU E7-8890 v3 @ 2.50GHz with 512G memory caused below changes (please refer to attached dmesg/kmsg for entire log/backtrace): +-------------------------------------------------+------------+------------+ | | e33898a207 | 09ded8af57 | +-------------------------------------------------+------------+------------+ | boot_successes | 14 | 4 | | boot_failures | 3 | | | BUG:kernel_reboot-without-warning_in_test_stage | 3 | | | dmesg.RIP:drm_setup_crtcs[drm_kms_helper] | 0 | 4 | +-------------------------------------------------+------------+------------+ If you fix the issue, kindly add following tag Reported-by: kernel test robot <rong.a.chen(a)intel.com> [ 35.625371] BUG: unable to handle kernel NULL pointer dereference at 0000000000000010 [ 35.634549] #PF error: [normal kernel read fault] [ 35.639800] PGD 0 P4D 0 [ 35.639805] Oops: 0000 [#1] SMP PTI [ 35.646527] CPU: 53 PID: 1179 Comm: systemd-udevd Not tainted 5.1.0-rc2-01104-g09ded8a #1 [ 35.655659] Hardware name: Intel Corporation BRICKLAND/BRICKLAND, BIOS BRHSXSD1.86B.0067.R02.1507221722 07/22/2015 [ 35.667239] RIP: 0010:drm_setup_crtcs+0x3eb/0x1020 [drm_kms_helper] [ 35.674236] Code: 44 24 60 48 8b 04 24 4c 01 f0 80 38 00 0f 84 a3 07 00 00 41 8b 8d 1c 03 00 00 83 f9 01 0f 84 5f 09 00 00 49 8b 95 e0 03 00 00 <48> 83 7a 10 00 0f 84 e0 01 00 00 48 8b 52 08 48 85 d2 0f 84 d1 01 [ 35.695199] RSP: 0000:ffffc9000f9938c0 EFLAGS: 00010297 [ 35.701035] RAX: ffff88c083da2198 RBX: ffff88c083da2180 RCX: 0000000000000000 [ 35.709000] RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffff88c083da21a1 [ 35.716966] RBP: 0000000000000000 R08: ffff88e083ab7d48 R09: 0000000000000000 [ 35.724934] R10: 0000000000000000 R11: ffff889fed53ec18 R12: 0000000000000001 [ 35.732894] R13: ffff888103a92800 R14: 0000000000000000 R15: 0000000000000001 [ 35.740863] FS: 00007fe6402648c0(0000) GS:ffff88dfff640000(0000) knlGS:0000000000000000 [ 35.749898] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 35.756312] CR2: 0000000000000010 CR3: 000000807c142003 CR4: 00000000001606e0 [ 35.764278] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 35.772235] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 35.780201] Call Trace: [ 35.782946] __drm_fb_helper_initial_config_and_unlock+0x46/0x540 [drm_kms_helper] [ 35.791404] mgag200_fbdev_init+0xc6/0xe0 [mgag200] [ 35.796855] mgag200_modeset_init+0x150/0x1b0 [mgag200] [ 35.802694] mgag200_driver_load+0x359/0x4d0 [mgag200] [ 35.808470] drm_dev_register+0x11c/0x1b0 [drm] [ 35.813547] drm_get_pci_dev+0x9d/0x180 [drm] [ 35.814299] ata5: SATA link up 1.5 Gbps (SStatus 113 SControl 300) [ 35.818417] local_pci_probe+0x42/0x90 [ 35.818427] ? _cond_resched+0x19/0x30 [ 35.827168] ata5.00: ATAPI: TEAC DV-W28S-W, 1.0A, max UDMA/100 [ 35.829511] pci_device_probe+0x141/0x1b0 [ 35.829521] really_probe+0xf8/0x3e0 [ 35.835771] ata5.00: configured for UDMA/100 [ 35.840505] driver_probe_device+0x10f/0x120 [ 35.840511] device_driver_attach+0x50/0x60 [ 35.847061] scsi 5:0:0:0: CD-ROM TEAC DV-W28S-W 1.0A PQ: 0 ANSI: 5 [ 35.848979] __driver_attach+0x9a/0x140 [ 35.848983] ? device_driver_attach+0x60/0x60 [ 35.848987] bus_for_each_dev+0x76/0xc0 [ 35.848996] ? klist_add_tail+0x3b/0x70 [ 35.889929] bus_add_driver+0x141/0x210 [ 35.894214] ? 0xffffffffc078d000 [ 35.897914] driver_register+0x5b/0xe0 [ 35.902093] ? 0xffffffffc078d000 [ 35.905805] do_one_initcall+0x46/0x1e4 [ 35.910089] ? _cond_resched+0x19/0x30 [ 35.914279] ? kmem_cache_alloc_trace+0x3b/0x1d0 [ 35.919439] do_init_module+0x5b/0x210 [ 35.923630] load_module+0x1838/0x1f00 [ 35.927822] ? ima_post_read_file+0xe2/0x120 [ 35.932592] ? __do_sys_finit_module+0xe9/0x110 [ 35.937650] __do_sys_finit_module+0xe9/0x110 [ 35.942508] do_syscall_64+0x5b/0x1a0 [ 35.946595] entry_SYSCALL_64_after_hwframe+0x44/0xa9 [ 35.952226] RIP: 0033:0x7fe63f0e1229 [ 35.956210] Code: 00 f3 c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 40 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 3f 4c 2b 00 f7 d8 64 89 01 48 [ 35.977171] RSP: 002b:00007fff141d8258 EFLAGS: 00000246 ORIG_RAX: 0000000000000139 [ 35.985624] RAX: ffffffffffffffda RBX: 000055dca71ebe70 RCX: 00007fe63f0e1229 [ 35.993580] RDX: 0000000000000000 RSI: 00007fe63f9fa265 RDI: 0000000000000015 [ 36.001545] RBP: 00007fe63f9fa265 R08: 0000000000000000 R09: 00007fff141d87d0 [ 36.009511] R10: 0000000000000015 R11: 0000000000000246 R12: 0000000000000000 [ 36.017477] R13: 000055dca71ff2f0 R14: 0000000000020000 R15: 000055dca5594cbc [ 36.025436] Modules linked in: irqbypass mgag200(+) ttm snd_pcm crct10dif_pclmul crc32_pclmul crc32c_intel snd_timer ipmi_ssif ghash_clmulni_intel drm_kms_helper ahci snd syscopyarea aesni_intel sysfillrect sysimgblt crypto_simd fb_sys_fops libahci soundcore cryptd mpt3sas(+) glue_helper pcspkr drm joydev libata ipmi_si raid_class lpc_ich i2c_i801 scsi_transport_sas ipmi_devintf wmi ipmi_msghandler acpi_pad pcc_cpufreq ip_tables [ 36.067874] CR2: 0000000000000010 [ 36.071721] ---[ end trace d29a594eb030ca5d ]--- To reproduce: git clone https://github.com/intel/lkp-tests.git cd lkp-tests bin/lkp install job.yaml # job file is attached in this email bin/lkp run job.yaml Thanks, Rong Chen

3 years, 2 months

2
1
0 / 0

[cgroup] 926250338c: WARNING:at_kernel/cgroup/cgroup.c:#cgroup_exit

by kernel test robot

FYI, we noticed the following commit (built with gcc-7): commit: 926250338c94ed8fd19677f861fef00bf6de7af7 ("cgroup: get rid of cgroup_freezer_frozen_exit()") https://github.com/rgushchin/linux.git freezer.10 in testcase: libhugetlbfs-test with following parameters: pagesize: 2MB on test machine: qemu-system-x86_64 -enable-kvm -cpu SandyBridge -smp 2 -m 2G caused below changes (please refer to attached dmesg/kmsg for entire log/backtrace): +-------------------------------------------------+------------+------------+ | | 8ad29ba936 | 926250338c | +-------------------------------------------------+------------+------------+ | boot_successes | 4 | 1 | | boot_failures | 0 | 8 | | WARNING:at_kernel/cgroup/cgroup.c:#cgroup_exit | 0 | 4 | | RIP:cgroup_exit | 0 | 4 | | BUG:kernel_reboot-without-warning_in_test_stage | 0 | 2 | | BUG:kernel_in_stage | 0 | 2 | +-------------------------------------------------+------------+------------+ If you fix the issue, kindly add following tag Reported-by: kernel test robot <rong.a.chen(a)intel.com> [ 44.576072] WARNING: CPU: 1 PID: 3028 at kernel/cgroup/cgroup.c:5932 cgroup_exit+0x148/0x160 [ 44.577724] Modules linked in: crct10dif_pclmul crc32_pclmul crc32c_intel ghash_clmulni_intel sr_mod cdrom bochs_drm sg ttm ata_generic pata_acpi ppdev drm_kms_helper snd_pcm syscopyarea aesni_intel snd_timer sysfillrect sysimgblt snd crypto_simd cryptd glue_helper soundcore fb_sys_fops joydev drm serio_raw pcspkr ata_piix libata i2c_piix4 floppy parport_pc parport ip_tables [ 44.583106] CPU: 1 PID: 3028 Comm: ptrace-write-hu Not tainted 5.1.0-rc3-00053-g9262503 #5 [ 44.584600] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.10.2-1 04/01/2014 [ 44.586116] RIP: 0010:cgroup_exit+0x148/0x160 [ 44.587135] Code: 0f 84 50 ff ff ff 48 8b 85 c8 0c 00 00 48 8b 78 70 e8 ec 2e 00 00 e9 3b ff ff ff f0 ff 43 60 0f 88 72 21 89 00 e9 48 ff ff ff <0f> 0b e9 1b ff ff ff e8 3c 73 f4 ff 66 90 66 2e 0f 1f 84 00 00 00 [ 44.590113] RSP: 0018:ffffb25702dcfd30 EFLAGS: 00010002 [ 44.591167] RAX: ffff96a7fee32410 RBX: ffff96a7ff1d6000 RCX: dead000000000200 [ 44.592446] RDX: ffff96a7ff1d6080 RSI: ffff96a7fec75290 RDI: ffff96a7fec75290 [ 44.593715] RBP: ffff96a7fec745c0 R08: ffff96a7fec74658 R09: 0000000000000000 [ 44.594985] R10: 0000000000000000 R11: 0000000000000001 R12: ffff96a7fec75101 [ 44.596266] R13: ffff96a7fec745c0 R14: ffff96a7ff3bde30 R15: ffff96a7fec75130 [ 44.597550] FS: 0000000000000000(0000) GS:ffff96a7dd700000(0000) knlGS:0000000000000000 [ 44.598950] CS: 0010 DS: 002b ES: 002b CR0: 0000000080050033 [ 44.600098] CR2: 00000000f7a00000 CR3: 000000000d20e000 CR4: 00000000000406e0 [ 44.601417] Call Trace: [ 44.602777] do_exit+0x337/0xc40 [ 44.603677] do_group_exit+0x3a/0xa0 [ 44.604610] get_signal+0x12e/0x8d0 [ 44.605533] ? __switch_to_asm+0x40/0x70 [ 44.606503] do_signal+0x36/0x650 [ 44.607409] ? __switch_to_asm+0x40/0x70 [ 44.608383] ? __schedule+0x267/0x860 [ 44.609329] exit_to_usermode_loop+0x89/0xf0 [ 44.610349] do_fast_syscall_32+0x251/0x2e3 [ 44.611357] entry_SYSENTER_compat+0x7f/0x91 [ 44.612376] ---[ end trace e4ca5cfc4b7f7964 ]--- To reproduce: # build kernel cd linux cp config-5.1.0-rc3-00053-g9262503 .config make HOSTCC=gcc-7 CC=gcc-7 ARCH=x86_64 olddefconfig make HOSTCC=gcc-7 CC=gcc-7 ARCH=x86_64 prepare make HOSTCC=gcc-7 CC=gcc-7 ARCH=x86_64 modules_prepare make HOSTCC=gcc-7 CC=gcc-7 ARCH=x86_64 SHELL=/bin/bash make HOSTCC=gcc-7 CC=gcc-7 ARCH=x86_64 bzImage git clone https://github.com/intel/lkp-tests.git cd lkp-tests bin/lkp qemu -k <bzImage> job-script # job-script is attached in this email Thanks, Rong Chen

3 years, 2 months

1
0
0 / 0

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

LKP April 2019