Re: [LKP] [net] 19f92a030c: apachebench.requests_per_second -37.9% regression

From: Feng Tang
Date: Wed Nov 13 2019 - 05:35:07 EST

Next message: Loic PALLARDY: "RE: [PATCH v2 1/1] remoteproc: add support for co-processor booted before kernel"
Previous message: Andrey Ryabinin: "KSM WARN_ON_ONCE(page_mapped(page)) in remove_stable_node()"
In reply to: kernel test robot: "[net] 19f92a030c: apachebench.requests_per_second -37.9% regression"
Next in thread: Eric Dumazet: "Re: [LKP] [net] 19f92a030c: apachebench.requests_per_second -37.9% regression"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

Hi Eric,

On Fri, Nov 08, 2019 at 04:35:13PM +0800, kernel test robot wrote:
> Greeting,
>
> FYI, we noticed a -37.9% regression of apachebench.requests_per_second due to commit:
>
> commit: 19f92a030ca6d772ab44b22ee6a01378a8cb32d4 ("net: increase SOMAXCONN to 4096")
> https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master

Any thought on this? The test is actually:

sysctl -w net.ipv4.tcp_syncookies=0
enable_apache_mod auth_basic authn_core authn_file authz_core authz_host authz_user access_compat
systemctl restart apache2
ab -k -q -t 300 -n 1000000 -c 4000 127.0.0.1/

And some info about apachebench result is:

w/o patch:

Connection Times (ms)
min mean[+/-sd] median max
Connect: 0 0 19.5 0 7145
Processing: 0 4 110.0 3 21647
Waiting: 0 2 92.4 1 21646
Total: 0 4 121.1 3 24762

w/ patch:

Connection Times (ms)
min mean[+/-sd] median max
Connect: 0 0 43.2 0 7143
Processing: 0 19 640.4 3 38708
Waiting: 0 24 796.5 1 38708
Total: 0 19 657.5 3 39725

Thanks,
Feng

>
> in testcase: apachebench
> on test machine: 16 threads Intel(R) Xeon(R) CPU D-1541 @ 2.10GHz with 48G memory
> with following parameters:
>
> runtime: 300s
> concurrency: 4000
> cluster: cs-localhost
> cpufreq_governor: performance
> ucode: 0x7000019
>
> test-description: apachebench is a tool for benchmarking your Apache Hypertext Transfer Protocol (HTTP) server.
> test-url: https://httpd.apache.org/docs/2.4/programs/ab.html
>
> In addition to that, the commit also has significant impact on the following tests:
>
> +------------------+------------------------------------------------------------------+
> | testcase: change | apachebench: apachebench.requests_per_second -37.5% regression |
> | test machine | 16 threads Intel(R) Xeon(R) CPU D-1541 @ 2.10GHz with 48G memory |
> | test parameters | cluster=cs-localhost |
> | | concurrency=8000 |
> | | cpufreq_governor=performance |
> | | runtime=300s |
> | | ucode=0x7000019 |
> +------------------+------------------------------------------------------------------+
>
>
> If you fix the issue, kindly add following tag
> Reported-by: kernel test robot <rong.a.chen@xxxxxxxxx>
>
>
> 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/concurrency/cpufreq_governor/kconfig/rootfs/runtime/tbox_group/testcase/ucode:
> cs-localhost/gcc-7/4000/performance/x86_64-rhel-7.6/debian-x86_64-2019-09-23.cgz/300s/lkp-bdw-de1/apachebench/0x7000019
>
> commit:
> 6d6f0383b6 ("netdevsim: Fix use-after-free during device dismantle")
> 19f92a030c ("net: increase SOMAXCONN to 4096")
>
> 6d6f0383b697f004 19f92a030ca6d772ab44b22ee6a
> ---------------- ---------------------------
> %stddev %change %stddev
> \ | \
> 22640 Â 4% +71.1% 38734 apachebench.connection_time.processing.max
> 24701 +60.9% 39743 apachebench.connection_time.total.max
> 22639 Â 4% +71.1% 38734 apachebench.connection_time.waiting.max
> 24701 +15042.0 39743 apachebench.max_latency.100%
> 40454 -37.9% 25128 apachebench.requests_per_second
> 25.69 +58.8% 40.79 apachebench.time.elapsed_time
> 25.69 +58.8% 40.79 apachebench.time.elapsed_time.max
> 79.00 -37.0% 49.75 apachebench.time.percent_of_cpu_this_job_got
> 98.88 +61.0% 159.18 apachebench.time_per_request
> 434631 -37.9% 269889 apachebench.transfer_rate
> 1.5e+08 Â 18% +109.5% 3.141e+08 Â 27% cpuidle.C3.time
> 578957 Â 7% +64.1% 949934 Â 12% cpuidle.C3.usage
> 79085 Â 4% +24.8% 98720 meminfo.AnonHugePages
> 41176 +14.2% 47013 meminfo.PageTables
> 69429 -34.9% 45222 meminfo.max_used_kB
> 63.48 +12.7 76.15 mpstat.cpu.all.idle%
> 2.42 Â 2% -0.9 1.56 mpstat.cpu.all.soft%
> 15.30 -5.2 10.13 mpstat.cpu.all.sys%
> 18.80 -6.6 12.16 mpstat.cpu.all.usr%
> 65.00 +17.7% 76.50 vmstat.cpu.id
> 17.00 -35.3% 11.00 vmstat.cpu.us
> 7.00 Â 24% -50.0% 3.50 Â 14% vmstat.procs.r
> 62957 -33.3% 42012 vmstat.system.cs
> 33174 -1.4% 32693 vmstat.system.in
> 5394 Â 5% +16.3% 6272 Â 6% sched_debug.cfs_rq:/.min_vruntime.stddev
> 5396 Â 5% +16.3% 6275 Â 6% sched_debug.cfs_rq:/.spread0.stddev
> 33982 Â 48% -83.3% 5676 Â 47% sched_debug.cpu.avg_idle.min
> 26.75 Â 77% +169.8% 72.17 Â 41% sched_debug.cpu.sched_count.avg
> 212.00 Â 90% +168.2% 568.50 Â 50% sched_debug.cpu.sched_count.max
> 52.30 Â 89% +182.5% 147.73 Â 48% sched_debug.cpu.sched_count.stddev
> 11.33 Â 80% +193.9% 33.30 Â 42% sched_debug.cpu.sched_goidle.avg
> 104.50 Â 92% +170.6% 282.75 Â 50% sched_debug.cpu.sched_goidle.max
> 26.18 Â 90% +183.9% 74.31 Â 48% sched_debug.cpu.sched_goidle.stddev
> 959.00 -32.0% 652.00 turbostat.Avg_MHz
> 39.01 -11.2 27.79 turbostat.Busy%
> 1.46 Â 7% -0.5 0.96 Â 5% turbostat.C1%
> 9.58 Â 4% -3.2 6.38 turbostat.C1E%
> 578646 Â 7% +64.1% 949626 Â 12% turbostat.C3
> 940073 +51.1% 1420298 turbostat.IRQ
> 2.20 Â 22% +159.7% 5.71 Â 12% turbostat.Pkg%pc2
> 31.22 -17.2% 25.86 turbostat.PkgWatt
> 4.74 -7.5% 4.39 turbostat.RAMWatt
> 93184 -1.6% 91678 proc-vmstat.nr_active_anon
> 92970 -1.8% 91314 proc-vmstat.nr_anon_pages
> 288405 +1.0% 291286 proc-vmstat.nr_file_pages
> 8307 +6.3% 8831 proc-vmstat.nr_kernel_stack
> 10315 +14.2% 11783 proc-vmstat.nr_page_table_pages
> 21499 +6.0% 22798 proc-vmstat.nr_slab_unreclaimable
> 284131 +1.0% 286977 proc-vmstat.nr_unevictable
> 93184 -1.6% 91678 proc-vmstat.nr_zone_active_anon
> 284131 +1.0% 286977 proc-vmstat.nr_zone_unevictable
> 198874 Â 2% +43.7% 285772 Â 16% proc-vmstat.numa_hit
> 198874 Â 2% +43.7% 285772 Â 16% proc-vmstat.numa_local
> 249594 Â 3% +59.6% 398267 Â 13% proc-vmstat.pgalloc_normal
> 1216885 +12.7% 1371283 Â 3% proc-vmstat.pgfault
> 179705 Â 16% +82.9% 328634 Â 21% proc-vmstat.pgfree
> 346.25 Â 5% +133.5% 808.50 Â 2% slabinfo.TCPv6.active_objs
> 346.25 Â 5% +134.4% 811.75 Â 2% slabinfo.TCPv6.num_objs
> 22966 +15.6% 26559 slabinfo.anon_vma.active_objs
> 23091 +15.5% 26664 slabinfo.anon_vma.num_objs
> 69747 +16.1% 81011 slabinfo.anon_vma_chain.active_objs
> 1094 +15.9% 1269 slabinfo.anon_vma_chain.active_slabs
> 70092 +15.9% 81259 slabinfo.anon_vma_chain.num_objs
> 1094 +15.9% 1269 slabinfo.anon_vma_chain.num_slabs
> 1649 +12.9% 1861 slabinfo.cred_jar.active_objs
> 1649 +12.9% 1861 slabinfo.cred_jar.num_objs
> 4924 +20.0% 5907 slabinfo.pid.active_objs
> 4931 +19.9% 5912 slabinfo.pid.num_objs
> 266.50 Â 3% +299.2% 1063 slabinfo.request_sock_TCP.active_objs
> 266.50 Â 3% +299.2% 1063 slabinfo.request_sock_TCP.num_objs
> 11.50 Â 4% +1700.0% 207.00 Â 4% slabinfo.tw_sock_TCPv6.active_objs
> 11.50 Â 4% +1700.0% 207.00 Â 4% slabinfo.tw_sock_TCPv6.num_objs
> 41682 +16.0% 48360 slabinfo.vm_area_struct.active_objs
> 1046 +15.7% 1211 slabinfo.vm_area_struct.active_slabs
> 41879 +15.7% 48468 slabinfo.vm_area_struct.num_objs
> 1046 +15.7% 1211 slabinfo.vm_area_struct.num_slabs
> 4276 Â 2% +10.0% 4705 Â 2% slabinfo.vmap_area.num_objs
> 21.25 Â 27% +3438.8% 752.00 Â 83% interrupts.36:IR-PCI-MSI.2621443-edge.eth0-TxRx-2
> 21.25 Â 20% +1777.6% 399.00 Â155% interrupts.38:IR-PCI-MSI.2621445-edge.eth0-TxRx-4
> 54333 +54.3% 83826 interrupts.CPU0.LOC:Local_timer_interrupts
> 54370 +54.6% 84072 interrupts.CPU1.LOC:Local_timer_interrupts
> 21.25 Â 27% +3438.8% 752.00 Â 83% interrupts.CPU10.36:IR-PCI-MSI.2621443-edge.eth0-TxRx-2
> 54236 +54.7% 83925 interrupts.CPU10.LOC:Local_timer_interrupts
> 54223 +54.3% 83655 interrupts.CPU11.LOC:Local_timer_interrupts
> 377.75 Â 21% +27.1% 480.25 Â 10% interrupts.CPU11.RES:Rescheduling_interrupts
> 21.25 Â 20% +1777.6% 399.00 Â155% interrupts.CPU12.38:IR-PCI-MSI.2621445-edge.eth0-TxRx-4
> 54279 +54.1% 83646 interrupts.CPU12.LOC:Local_timer_interrupts
> 53683 +55.3% 83365 interrupts.CPU13.LOC:Local_timer_interrupts
> 53887 +55.7% 83903 interrupts.CPU14.LOC:Local_timer_interrupts
> 54156 +54.7% 83803 interrupts.CPU15.LOC:Local_timer_interrupts
> 54041 +55.1% 83806 interrupts.CPU2.LOC:Local_timer_interrupts
> 54042 +55.4% 83991 interrupts.CPU3.LOC:Local_timer_interrupts
> 54081 +55.2% 83938 interrupts.CPU4.LOC:Local_timer_interrupts
> 54322 +54.9% 84166 interrupts.CPU5.LOC:Local_timer_interrupts
> 53586 +56.5% 83849 interrupts.CPU6.LOC:Local_timer_interrupts
> 54049 +55.2% 83892 interrupts.CPU7.LOC:Local_timer_interrupts
> 54056 +54.9% 83751 interrupts.CPU8.LOC:Local_timer_interrupts
> 53862 +54.7% 83331 interrupts.CPU9.LOC:Local_timer_interrupts
> 865212 +55.0% 1340925 interrupts.LOC:Local_timer_interrupts
> 16477 Â 4% +32.2% 21779 Â 8% softirqs.CPU0.TIMER
> 18508 Â 15% +39.9% 25891 Â 17% softirqs.CPU1.TIMER
> 16625 Â 8% +21.3% 20166 Â 7% softirqs.CPU10.TIMER
> 5906 Â 21% +62.5% 9597 Â 13% softirqs.CPU12.SCHED
> 17474 Â 12% +29.4% 22610 Â 7% softirqs.CPU12.TIMER
> 7680 Â 11% +20.4% 9246 Â 14% softirqs.CPU13.SCHED
> 45558 Â 36% -37.8% 28320 Â 25% softirqs.CPU14.NET_RX
> 15365 Â 4% +40.7% 21622 Â 5% softirqs.CPU14.TIMER
> 8084 Â 4% +18.7% 9599 Â 12% softirqs.CPU15.RCU
> 16433 Â 4% +41.2% 23203 Â 14% softirqs.CPU2.TIMER
> 8436 Â 7% +19.9% 10117 Â 10% softirqs.CPU3.RCU
> 15992 Â 3% +48.5% 23742 Â 18% softirqs.CPU3.TIMER
> 17389 Â 14% +38.7% 24116 Â 11% softirqs.CPU4.TIMER
> 17749 Â 13% +42.2% 25235 Â 15% softirqs.CPU5.TIMER
> 16528 Â 9% +28.3% 21200 Â 2% softirqs.CPU6.TIMER
> 8321 Â 8% +31.3% 10929 Â 5% softirqs.CPU7.RCU
> 18024 Â 8% +28.8% 23212 Â 5% softirqs.CPU7.TIMER
> 15717 Â 5% +27.1% 19983 Â 7% softirqs.CPU8.TIMER
> 7383 Â 11% +30.5% 9632 Â 9% softirqs.CPU9.SCHED
> 16342 Â 18% +41.0% 23037 Â 10% softirqs.CPU9.TIMER
> 148013 +10.2% 163086 Â 2% softirqs.RCU
> 112139 +28.0% 143569 softirqs.SCHED
> 276690 Â 3% +30.4% 360747 Â 3% softirqs.TIMER
> 1.453e+09 -36.2% 9.273e+08 perf-stat.i.branch-instructions
> 67671486 -35.9% 43396843 Â 2% perf-stat.i.branch-misses
> 5.188e+08 -35.3% 3.357e+08 perf-stat.i.cache-misses
> 5.188e+08 -35.3% 3.357e+08 perf-stat.i.cache-references
> 71149 -36.0% 45536 perf-stat.i.context-switches
> 2.92 Â 6% +38.4% 4.04 Â 5% perf-stat.i.cpi
> 1.581e+10 -33.8% 1.046e+10 perf-stat.i.cpu-cycles
> 1957 Â 6% -35.4% 1264 Â 5% perf-stat.i.cpu-migrations
> 40.76 Â 2% +33.0% 54.21 perf-stat.i.cycles-between-cache-misses
> 0.64 Â 2% +0.2 0.86 Â 2% perf-stat.i.dTLB-load-miss-rate%
> 10720126 Â 2% -34.0% 7071299 perf-stat.i.dTLB-load-misses
> 2.05e+09 -35.8% 1.315e+09 perf-stat.i.dTLB-loads
> 0.16 +0.0 0.18 Â 5% perf-stat.i.dTLB-store-miss-rate%
> 1688635 -31.7% 1153595 perf-stat.i.dTLB-store-misses
> 1.17e+09 -33.5% 7.777e+08 perf-stat.i.dTLB-stores
> 61.00 Â 6% +8.7 69.70 Â 2% perf-stat.i.iTLB-load-miss-rate%
> 13112146 Â 10% -38.5% 8061589 Â 13% perf-stat.i.iTLB-load-misses
> 9827689 Â 3% -37.1% 6184316 perf-stat.i.iTLB-loads
> 7e+09 -36.2% 4.469e+09 perf-stat.i.instructions
> 0.42 Â 2% -22.2% 0.33 Â 3% perf-stat.i.ipc
> 45909 -32.4% 31036 perf-stat.i.minor-faults
> 45909 -32.4% 31036 perf-stat.i.page-faults
> 2.26 +3.6% 2.34 perf-stat.overall.cpi
> 30.47 +2.3% 31.16 perf-stat.overall.cycles-between-cache-misses
> 0.44 -3.5% 0.43 perf-stat.overall.ipc
> 1.398e+09 -35.3% 9.049e+08 perf-stat.ps.branch-instructions
> 65124290 -35.0% 42353120 Â 2% perf-stat.ps.branch-misses
> 4.993e+08 -34.4% 3.276e+08 perf-stat.ps.cache-misses
> 4.993e+08 -34.4% 3.276e+08 perf-stat.ps.cache-references
> 68469 -35.1% 44440 perf-stat.ps.context-switches
> 1.521e+10 -32.9% 1.021e+10 perf-stat.ps.cpu-cycles
> 1884 Â 6% -34.5% 1234 Â 5% perf-stat.ps.cpu-migrations
> 10314734 Â 2% -33.1% 6899548 perf-stat.ps.dTLB-load-misses
> 1.973e+09 -34.9% 1.283e+09 perf-stat.ps.dTLB-loads
> 1624883 -30.7% 1125668 perf-stat.ps.dTLB-store-misses
> 1.126e+09 -32.6% 7.589e+08 perf-stat.ps.dTLB-stores
> 12617676 Â 10% -37.7% 7866427 Â 13% perf-stat.ps.iTLB-load-misses
> 9458687 Â 3% -36.2% 6036594 perf-stat.ps.iTLB-loads
> 6.736e+09 -35.3% 4.361e+09 perf-stat.ps.instructions
> 44179 -31.4% 30288 perf-stat.ps.minor-faults
> 30791 +1.4% 31223 perf-stat.ps.msec
> 44179 -31.4% 30288 perf-stat.ps.page-faults
> 21.96 Â 69% -22.0 0.00 perf-profile.calltrace.cycles-pp.__x64_sys_execve.do_syscall_64.entry_SYSCALL_64_after_hwframe
> 21.96 Â 69% -22.0 0.00 perf-profile.calltrace.cycles-pp.__do_execve_file.__x64_sys_execve.do_syscall_64.entry_SYSCALL_64_after_hwframe
> 18.63 Â 89% -18.6 0.00 perf-profile.calltrace.cycles-pp.search_binary_handler.__do_execve_file.__x64_sys_execve.do_syscall_64.entry_SYSCALL_64_after_hwframe
> 18.63 Â 89% -18.6 0.00 perf-profile.calltrace.cycles-pp.load_elf_binary.search_binary_handler.__do_execve_file.__x64_sys_execve.do_syscall_64
> 13.15 Â 67% -8.2 5.00 Â173% perf-profile.calltrace.cycles-pp.intel_idle.cpuidle_enter_state.cpuidle_enter.do_idle.cpu_startup_entry
> 8.81 Â133% -8.0 0.83 Â173% perf-profile.calltrace.cycles-pp.ret_from_fork
> 13.15 Â 67% -7.3 5.83 Â173% perf-profile.calltrace.cycles-pp.secondary_startup_64
> 13.15 Â 67% -7.3 5.83 Â173% perf-profile.calltrace.cycles-pp.start_secondary.secondary_startup_64
> 13.15 Â 67% -7.3 5.83 Â173% perf-profile.calltrace.cycles-pp.cpu_startup_entry.start_secondary.secondary_startup_64
> 13.15 Â 67% -7.3 5.83 Â173% perf-profile.calltrace.cycles-pp.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64
> 13.15 Â 67% -7.3 5.83 Â173% perf-profile.calltrace.cycles-pp.cpuidle_enter.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64
> 13.15 Â 67% -7.3 5.83 Â173% perf-profile.calltrace.cycles-pp.cpuidle_enter_state.cpuidle_enter.do_idle.cpu_startup_entry.start_secondary
> 6.70 Â100% -6.7 0.00 perf-profile.calltrace.cycles-pp.__clear_user.load_elf_binary.search_binary_handler.__do_execve_file.__x64_sys_execve
> 4.79 Â108% -4.8 0.00 perf-profile.calltrace.cycles-pp.free_pgtables.exit_mmap.mmput.do_exit.do_group_exit
> 4.79 Â108% -4.8 0.00 perf-profile.calltrace.cycles-pp.unlink_file_vma.free_pgtables.exit_mmap.mmput.do_exit
> 4.79 Â108% -4.0 0.83 Â173% perf-profile.calltrace.cycles-pp.mmput.do_exit.do_group_exit.get_signal.do_signal
> 4.79 Â108% -4.0 0.83 Â173% perf-profile.calltrace.cycles-pp.exit_mmap.mmput.do_exit.do_group_exit.get_signal
> 5.24 Â112% -0.2 5.00 Â173% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.ioctl.perf_evsel__enable.evsel__enable.evlist__enable
> 5.24 Â112% -0.2 5.00 Â173% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.ioctl.perf_evsel__enable.evsel__enable
> 5.24 Â112% -0.2 5.00 Â173% perf-profile.calltrace.cycles-pp.event_function_call.perf_event_for_each_child._perf_ioctl.perf_ioctl.do_vfs_ioctl
> 5.24 Â112% -0.2 5.00 Â173% perf-profile.calltrace.cycles-pp.smp_call_function_single.event_function_call.perf_event_for_each_child._perf_ioctl.perf_ioctl
> 5.24 Â112% -0.2 5.00 Â173% perf-profile.calltrace.cycles-pp.__libc_start_main
> 5.24 Â112% -0.2 5.00 Â173% perf-profile.calltrace.cycles-pp.main.__libc_start_main
> 5.24 Â112% -0.2 5.00 Â173% perf-profile.calltrace.cycles-pp.run_builtin.main.__libc_start_main
> 5.24 Â112% -0.2 5.00 Â173% perf-profile.calltrace.cycles-pp.cmd_record.run_builtin.main.__libc_start_main
> 5.24 Â112% -0.2 5.00 Â173% perf-profile.calltrace.cycles-pp.evlist__enable.cmd_record.run_builtin.main.__libc_start_main
> 5.24 Â112% -0.2 5.00 Â173% perf-profile.calltrace.cycles-pp.evsel__enable.evlist__enable.cmd_record.run_builtin.main
> 5.24 Â112% -0.2 5.00 Â173% perf-profile.calltrace.cycles-pp.perf_evsel__enable.evsel__enable.evlist__enable.cmd_record.run_builtin
> 5.24 Â112% -0.2 5.00 Â173% perf-profile.calltrace.cycles-pp.ioctl.perf_evsel__enable.evsel__enable.evlist__enable.cmd_record
> 5.24 Â112% -0.2 5.00 Â173% perf-profile.calltrace.cycles-pp.__x64_sys_ioctl.do_syscall_64.entry_SYSCALL_64_after_hwframe.ioctl.perf_evsel__enable
> 5.24 Â112% -0.2 5.00 Â173% perf-profile.calltrace.cycles-pp.ksys_ioctl.__x64_sys_ioctl.do_syscall_64.entry_SYSCALL_64_after_hwframe.ioctl
> 5.24 Â112% -0.2 5.00 Â173% perf-profile.calltrace.cycles-pp.do_vfs_ioctl.ksys_ioctl.__x64_sys_ioctl.do_syscall_64.entry_SYSCALL_64_after_hwframe
> 5.24 Â112% -0.2 5.00 Â173% perf-profile.calltrace.cycles-pp.perf_ioctl.do_vfs_ioctl.ksys_ioctl.__x64_sys_ioctl.do_syscall_64
> 5.24 Â112% -0.2 5.00 Â173% perf-profile.calltrace.cycles-pp._perf_ioctl.perf_ioctl.do_vfs_ioctl.ksys_ioctl.__x64_sys_ioctl
> 5.24 Â112% -0.2 5.00 Â173% perf-profile.calltrace.cycles-pp.perf_event_for_each_child._perf_ioctl.perf_ioctl.do_vfs_ioctl.ksys_ioctl
> 4.79 Â108% +8.5 13.33 Â173% perf-profile.calltrace.cycles-pp.exit_to_usermode_loop.do_syscall_64.entry_SYSCALL_64_after_hwframe
> 4.79 Â108% +8.5 13.33 Â173% perf-profile.calltrace.cycles-pp.do_signal.exit_to_usermode_loop.do_syscall_64.entry_SYSCALL_64_after_hwframe
> 4.79 Â108% +8.5 13.33 Â173% perf-profile.calltrace.cycles-pp.get_signal.do_signal.exit_to_usermode_loop.do_syscall_64.entry_SYSCALL_64_after_hwframe
> 4.79 Â108% +8.5 13.33 Â173% perf-profile.calltrace.cycles-pp.do_group_exit.get_signal.do_signal.exit_to_usermode_loop.do_syscall_64
> 4.79 Â108% +8.5 13.33 Â173% perf-profile.calltrace.cycles-pp.do_exit.do_group_exit.get_signal.do_signal.exit_to_usermode_loop
> 12.92 Â100% +12.1 25.00 Â173% perf-profile.calltrace.cycles-pp.__x64_sys_exit_group.do_syscall_64.entry_SYSCALL_64_after_hwframe
> 12.92 Â100% +12.1 25.00 Â173% perf-profile.calltrace.cycles-pp.do_group_exit.__x64_sys_exit_group.do_syscall_64.entry_SYSCALL_64_after_hwframe
> 12.92 Â100% +12.1 25.00 Â173% perf-profile.calltrace.cycles-pp.do_exit.do_group_exit.__x64_sys_exit_group.do_syscall_64.entry_SYSCALL_64_after_hwframe
> 11.25 Â101% +13.8 25.00 Â173% perf-profile.calltrace.cycles-pp.mmput.do_exit.do_group_exit.__x64_sys_exit_group.do_syscall_64
> 9.58 Â108% +15.4 25.00 Â173% perf-profile.calltrace.cycles-pp.exit_mmap.mmput.do_exit.do_group_exit.__x64_sys_exit_group
> 21.96 Â 69% -22.0 0.00 perf-profile.children.cycles-pp.__x64_sys_execve
> 21.96 Â 69% -22.0 0.00 perf-profile.children.cycles-pp.__do_execve_file
> 18.63 Â 89% -18.6 0.00 perf-profile.children.cycles-pp.search_binary_handler
> 18.63 Â 89% -18.6 0.00 perf-profile.children.cycles-pp.load_elf_binary
> 13.15 Â 67% -8.2 5.00 Â173% perf-profile.children.cycles-pp.intel_idle
> 8.81 Â133% -8.0 0.83 Â173% perf-profile.children.cycles-pp.ret_from_fork
> 13.15 Â 67% -7.3 5.83 Â173% perf-profile.children.cycles-pp.secondary_startup_64
> 13.15 Â 67% -7.3 5.83 Â173% perf-profile.children.cycles-pp.start_secondary
> 13.15 Â 67% -7.3 5.83 Â173% perf-profile.children.cycles-pp.cpu_startup_entry
> 13.15 Â 67% -7.3 5.83 Â173% perf-profile.children.cycles-pp.do_idle
> 13.15 Â 67% -7.3 5.83 Â173% perf-profile.children.cycles-pp.cpuidle_enter
> 13.15 Â 67% -7.3 5.83 Â173% perf-profile.children.cycles-pp.cpuidle_enter_state
> 6.70 Â100% -6.7 0.00 perf-profile.children.cycles-pp.__clear_user
> 6.46 Â100% -6.5 0.00 perf-profile.children.cycles-pp.handle_mm_fault
> 4.79 Â108% -4.8 0.00 perf-profile.children.cycles-pp.page_fault
> 4.79 Â108% -4.8 0.00 perf-profile.children.cycles-pp.do_page_fault
> 4.79 Â108% -4.8 0.00 perf-profile.children.cycles-pp.__do_page_fault
> 4.79 Â108% -4.8 0.00 perf-profile.children.cycles-pp.free_pgtables
> 4.79 Â108% -4.8 0.00 perf-profile.children.cycles-pp.unlink_file_vma
> 5.24 Â112% -0.2 5.00 Â173% perf-profile.children.cycles-pp.__libc_start_main
> 5.24 Â112% -0.2 5.00 Â173% perf-profile.children.cycles-pp.main
> 5.24 Â112% -0.2 5.00 Â173% perf-profile.children.cycles-pp.run_builtin
> 5.24 Â112% -0.2 5.00 Â173% perf-profile.children.cycles-pp.cmd_record
> 5.24 Â112% -0.2 5.00 Â173% perf-profile.children.cycles-pp.evlist__enable
> 5.24 Â112% -0.2 5.00 Â173% perf-profile.children.cycles-pp.evsel__enable
> 5.24 Â112% -0.2 5.00 Â173% perf-profile.children.cycles-pp.perf_evsel__enable
> 5.24 Â112% -0.2 5.00 Â173% perf-profile.children.cycles-pp.ioctl
> 5.24 Â112% -0.2 5.00 Â173% perf-profile.children.cycles-pp.__x64_sys_ioctl
> 5.24 Â112% -0.2 5.00 Â173% perf-profile.children.cycles-pp.ksys_ioctl
> 5.24 Â112% -0.2 5.00 Â173% perf-profile.children.cycles-pp.do_vfs_ioctl
> 5.24 Â112% -0.2 5.00 Â173% perf-profile.children.cycles-pp.perf_ioctl
> 5.24 Â112% -0.2 5.00 Â173% perf-profile.children.cycles-pp._perf_ioctl
> 5.24 Â112% -0.2 5.00 Â173% perf-profile.children.cycles-pp.perf_event_for_each_child
> 4.79 Â108% +8.5 13.33 Â173% perf-profile.children.cycles-pp.exit_to_usermode_loop
> 4.79 Â108% +8.5 13.33 Â173% perf-profile.children.cycles-pp.do_signal
> 4.79 Â108% +8.5 13.33 Â173% perf-profile.children.cycles-pp.get_signal
> 5.24 Â112% +8.9 14.17 Â173% perf-profile.children.cycles-pp.event_function_call
> 5.24 Â112% +8.9 14.17 Â173% perf-profile.children.cycles-pp.smp_call_function_single
> 12.92 Â100% +12.1 25.00 Â173% perf-profile.children.cycles-pp.__x64_sys_exit_group
> 13.15 Â 67% -8.2 5.00 Â173% perf-profile.self.cycles-pp.intel_idle
> 6.46 Â100% -6.5 0.00 perf-profile.self.cycles-pp.unmap_page_range
> 5.24 Â112% +8.9 14.17 Â173% perf-profile.self.cycles-pp.smp_call_function_single
>

Next message: Loic PALLARDY: "RE: [PATCH v2 1/1] remoteproc: add support for co-processor booted before kernel"
Previous message: Andrey Ryabinin: "KSM WARN_ON_ONCE(page_mapped(page)) in remove_stable_node()"
In reply to: kernel test robot: "[net] 19f92a030c: apachebench.requests_per_second -37.9% regression"
Next in thread: Eric Dumazet: "Re: [LKP] [net] 19f92a030c: apachebench.requests_per_second -37.9% regression"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]