| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <20190110132539.GL17624@shao2-debian>
Date: Thu, 10 Jan 2019 21:25:39 +0800
From: kernel test robot <rong.a.chen@...el.com>
To: Michal Hocko <mhocko@...e.com>
Cc: Linus Torvalds <torvalds@...ux-foundation.org>,
Jan Kara <jack@...e.cz>,
"Kirill A. Shutemov" <kirill.shutemov@...ux.intel.com>,
David Hildenbrand <david@...hat.com>,
Hugh Dickins <hughd@...gle.com>,
William Kucharski <william.kucharski@...cle.com>,
Oscar Salvador <osalvador@...e.de>,
Pavel Tatashin <pasha.tatashin@...een.com>,
Andrew Morton <akpm@...ux-foundation.org>,
LKML <linux-kernel@...r.kernel.org>, lkp@...org
Subject: [LKP] [mm, fault_around] e0975b2aae: vm-scalability.throughput 7.2%
improvement
Greeting,
FYI, we noticed a 7.2% improvement of vm-scalability.throughput due to commit:
commit: e0975b2aae0e669f995f7d5f11db25c3080ae11c ("mm, fault_around: do not take a reference to a locked page")
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master
in testcase: vm-scalability
on test machine: 144 threads Intel(R) Xeon(R) CPU E7-8890 v3 @ 2.50GHz with 512G memory
with following parameters:
runtime: 300s
size: 2T
test: shm-pread-seq
cpufreq_governor: performance
ucode: 0x12
test-description: The motivation behind this suite is to exercise functions and regions of the mm/ of the Linux kernel which are of interest to us.
test-url: https://git.kernel.org/cgit/linux/kernel/git/wfg/vm-scalability.git/
In addition to that, the commit also has significant impact on the following tests:
+------------------+------------------------------------------------------------------------+
| testcase: change | vm-scalability: vm-scalability.throughput 7.8% improvement |
| test machine | 104 threads Skylake with 192G memory |
| test parameters | cpufreq_governor=performance |
| | runtime=300s |
| | size=2T |
| | test=shm-xread-seq-mt |
+------------------+------------------------------------------------------------------------+
| testcase: change | vm-scalability: vm-scalability.throughput 8.2% improvement |
| test machine | 144 threads Intel(R) Xeon(R) CPU E7-8890 v3 @ 2.50GHz with 512G memory |
| test parameters | cpufreq_governor=performance |
| | runtime=300s |
| | size=2T |
| | test=shm-xread-seq |
| | ucode=0x12 |
+------------------+------------------------------------------------------------------------+
| testcase: change | vm-scalability: vm-scalability.throughput 27.4% improvement |
| test machine | 104 threads Skylake with 192G memory |
| test parameters | cpufreq_governor=performance |
| | runtime=300s |
| | size=2T |
| | test=shm-pread-seq |
+------------------+------------------------------------------------------------------------+
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/rootfs/runtime/size/tbox_group/test/testcase/ucode:
gcc-7/performance/x86_64-rhel-7.2/debian-x86_64-2018-04-03.cgz/300s/2T/lkp-hsx04/shm-pread-seq/vm-scalability/0x12
commit:
bb8965bd82 ("mm, memory_hotplug: deobfuscate migration part of offlining")
e0975b2aae ("mm, fault_around: do not take a reference to a locked page")
bb8965bd82fd4ed4 e0975b2aae0e669f995f7d5f11
---------------- --------------------------
%stddev %change %stddev
\ | \
205797 +7.2% 220550 vm-scalability.median
29634759 +7.2% 31759332 vm-scalability.throughput
191779 +8.8% 208653 vm-scalability.time.involuntary_context_switches
28835128 +15.5% 33292471 vm-scalability.time.maximum_resident_set_size
2.885e+08 +32.5% 3.822e+08 vm-scalability.time.minor_page_faults
10463 +3.6% 10841 vm-scalability.time.percent_of_cpu_this_job_got
5289 +18.8% 6283 vm-scalability.time.user_time
83737450 +24.2% 1.04e+08 vm-scalability.time.voluntary_context_switches
8.906e+09 +7.2% 9.548e+09 vm-scalability.workload
11.36 +1.9 13.28 mpstat.cpu.usr%
9659 ± 4% +102.3% 19541 ± 42% softirqs.NET_RX
501.91 +1.3% 508.56 pmeter.Average_Active_Power
59044 +5.8% 62450 pmeter.performance_per_watt
47171145 ± 6% +15.5% 54471732 ± 4% cpuidle.C1E.usage
1.12e+08 ± 6% +51.6% 1.698e+08 cpuidle.POLL.time
22957109 ± 3% +72.3% 39565949 cpuidle.POLL.usage
16265232 +14.1% 18566370 vmstat.memory.cache
34.00 ± 2% -13.2% 29.50 ± 3% vmstat.procs.b
512381 +22.3% 626595 ± 2% vmstat.system.cs
273428 ± 16% +11.7% 305449 vmstat.system.in
5647 ± 10% -17.0% 4687 ± 9% slabinfo.kmalloc-rcl-64.active_objs
5647 ± 10% -17.0% 4687 ± 9% slabinfo.kmalloc-rcl-64.num_objs
77126 +12.0% 86346 slabinfo.radix_tree_node.active_objs
1380 +12.0% 1546 slabinfo.radix_tree_node.active_slabs
77344 +12.0% 86622 slabinfo.radix_tree_node.num_objs
1380 +12.0% 1546 slabinfo.radix_tree_node.num_slabs
894.50 +3.3% 923.75 turbostat.Avg_MHz
47169644 ± 6% +15.5% 54468669 ± 4% turbostat.C1E
13.24 ± 5% -1.5 11.72 ± 3% turbostat.C1E%
23.60 -12.6% 20.63 turbostat.CPU%c1
206.09 +1.7% 209.60 turbostat.PkgWatt
67.50 +2.9% 69.44 turbostat.RAMWatt
2951760 ± 2% +15.4% 3405759 numa-numastat.node0.local_node
2972843 ± 2% +15.3% 3428397 numa-numastat.node0.numa_hit
3023383 +15.8% 3500524 ± 2% numa-numastat.node1.local_node
3040209 +15.8% 3521706 ± 2% numa-numastat.node1.numa_hit
3044343 +18.6% 3610848 numa-numastat.node2.local_node
3060544 +18.5% 3626820 numa-numastat.node2.numa_hit
3114211 ± 2% +17.6% 3662636 ± 3% numa-numastat.node3.local_node
3141243 ± 2% +17.3% 3684146 ± 2% numa-numastat.node3.numa_hit
1274575 ± 6% +26.9% 1616942 ± 6% meminfo.Active
1274375 ± 6% +26.9% 1616744 ± 6% meminfo.Active(anon)
16192488 +14.2% 18493780 meminfo.Cached
14943953 +16.0% 17331376 meminfo.Committed_AS
14272272 +13.8% 16236839 meminfo.Inactive
14270911 +13.8% 16235487 meminfo.Inactive(anon)
14026514 +14.1% 15998032 meminfo.Mapped
21736223 +13.5% 24675545 meminfo.Memused
4150612 +15.1% 4776179 meminfo.PageTables
15291010 +15.1% 17597113 meminfo.Shmem
121327 +12.9% 136955 meminfo.max_used_kB
319303 ± 7% +26.5% 404003 ± 6% proc-vmstat.nr_active_anon
4048119 +14.2% 4622041 proc-vmstat.nr_file_pages
3566764 +13.8% 4057370 proc-vmstat.nr_inactive_anon
3506031 +14.0% 3998186 proc-vmstat.nr_mapped
1035762 +15.1% 1192254 proc-vmstat.nr_page_table_pages
3822486 +15.0% 4397612 proc-vmstat.nr_shmem
29593 +4.3% 30875 proc-vmstat.nr_slab_reclaimable
319303 ± 7% +26.5% 404003 ± 6% proc-vmstat.nr_zone_active_anon
3566764 +13.8% 4057370 proc-vmstat.nr_zone_inactive_anon
267485 ± 2% +7.5% 287493 proc-vmstat.numa_hint_faults
138073 +8.4% 149728 proc-vmstat.numa_hint_faults_local
12242939 +16.7% 14287749 proc-vmstat.numa_hit
12161618 +16.8% 14206439 proc-vmstat.numa_local
7215166 +15.5% 8330196 proc-vmstat.pgactivate
12329910 +16.6% 14375339 proc-vmstat.pgalloc_normal
2.896e+08 +32.4% 3.833e+08 proc-vmstat.pgfault
12031529 ± 2% +18.6% 14269136 proc-vmstat.pgfree
46882139 ± 7% -1.7e+07 30076309 ± 2% syscalls.sys_close.noise.100%
66759694 ± 5% -1.6e+07 50769186 syscalls.sys_close.noise.2%
62808302 ± 6% -1.6e+07 46659603 ± 2% syscalls.sys_close.noise.25%
66494384 ± 5% -1.6e+07 50475741 ± 2% syscalls.sys_close.noise.5%
58018236 ± 6% -1.7e+07 41364386 syscalls.sys_close.noise.50%
52585595 ± 7% -1.7e+07 35436324 ± 2% syscalls.sys_close.noise.75%
2.584e+08 ± 5% -1.3e+08 1.307e+08 ± 59% syscalls.sys_mmap.noise.100%
4.64e+08 ± 2% -1.8e+08 2.869e+08 ± 58% syscalls.sys_mmap.noise.2%
4.365e+08 ± 2% -1.7e+08 2.647e+08 ± 58% syscalls.sys_mmap.noise.25%
4.627e+08 ± 2% -1.8e+08 2.858e+08 ± 58% syscalls.sys_mmap.noise.5%
3.821e+08 ± 2% -1.6e+08 2.22e+08 ± 58% syscalls.sys_mmap.noise.50%
3.158e+08 ± 3% -1.4e+08 1.728e+08 ± 58% syscalls.sys_mmap.noise.75%
9.457e+08 ± 53% -3.6e+08 5.889e+08 ± 35% syscalls.sys_write.noise.100%
9.75e+08 ± 52% -3.6e+08 6.166e+08 ± 33% syscalls.sys_write.noise.2%
9.638e+08 ± 52% -3.6e+08 6.039e+08 ± 34% syscalls.sys_write.noise.25%
9.744e+08 ± 52% -3.6e+08 6.158e+08 ± 34% syscalls.sys_write.noise.5%
9.563e+08 ± 53% -3.6e+08 5.973e+08 ± 35% syscalls.sys_write.noise.50%
9.504e+08 ± 53% -3.6e+08 5.923e+08 ± 35% syscalls.sys_write.noise.75%
411535 ± 32% +35.9% 559080 ± 15% sched_debug.cfs_rq:/.spread0.max
273311 -13.2% 237211 sched_debug.cpu.avg_idle.avg
25367 ± 8% -25.8% 18829 ± 4% sched_debug.cpu.avg_idle.min
0.75 ± 36% +72.2% 1.29 ± 5% sched_debug.cpu.cpu_load[3].min
0.75 ± 36% +72.2% 1.29 ± 5% sched_debug.cpu.cpu_load[4].min
578316 +23.9% 716714 sched_debug.cpu.nr_switches.avg
603487 +24.4% 750710 sched_debug.cpu.nr_switches.max
523263 ± 2% +18.9% 621971 ± 3% sched_debug.cpu.nr_switches.min
13300 ± 2% +39.5% 18551 ± 7% sched_debug.cpu.nr_switches.stddev
577.58 ± 5% -15.1% 490.46 ± 6% sched_debug.cpu.nr_uninterruptible.max
85.33 ± 3% -10.9% 76.04 ± 4% sched_debug.cpu.nr_uninterruptible.stddev
579069 +24.0% 717916 sched_debug.cpu.sched_count.avg
619427 +24.5% 771344 sched_debug.cpu.sched_count.max
523432 ± 2% +19.0% 622912 ± 3% sched_debug.cpu.sched_count.min
13457 ± 3% +38.5% 18638 ± 6% sched_debug.cpu.sched_count.stddev
287779 +24.0% 356747 sched_debug.cpu.sched_goidle.avg
298839 +24.7% 372681 sched_debug.cpu.sched_goidle.max
260244 ± 2% +19.0% 309760 ± 3% sched_debug.cpu.sched_goidle.min
6507 ± 3% +40.2% 9122 ± 7% sched_debug.cpu.sched_goidle.stddev
296998 +25.0% 371368 sched_debug.cpu.ttwu_count.avg
302223 +25.4% 378931 sched_debug.cpu.ttwu_count.max
270019 ± 2% +19.9% 323707 ± 3% sched_debug.cpu.ttwu_count.min
3491 ± 4% +57.1% 5485 ± 5% sched_debug.cpu.ttwu_count.stddev
628.74 +18.7% 746.13 ± 2% sched_debug.cpu.ttwu_local.avg
1289 +13.2% 1459 ± 5% sched_debug.cpu.ttwu_local.max
508.75 ± 2% +16.3% 591.75 ± 6% sched_debug.cpu.ttwu_local.min
81.04 ± 3% +12.4% 91.06 ± 5% sched_debug.cpu.ttwu_local.stddev
286567 ± 19% +28.2% 367248 ± 15% numa-meminfo.node0.Active
286518 ± 19% +28.2% 367204 ± 15% numa-meminfo.node0.Active(anon)
4015453 ± 3% +11.4% 4473419 numa-meminfo.node0.FilePages
3541191 ± 3% +10.7% 3920092 numa-meminfo.node0.Inactive
3540854 ± 3% +10.7% 3919353 numa-meminfo.node0.Inactive(anon)
3473113 +12.8% 3916778 numa-meminfo.node0.Mapped
5489928 ± 3% +11.3% 6110847 numa-meminfo.node0.MemUsed
1038778 +14.9% 1193144 numa-meminfo.node0.PageTables
3795283 ± 3% +11.9% 4248158 numa-meminfo.node0.Shmem
4046773 ± 2% +14.4% 4628502 ± 3% numa-meminfo.node1.FilePages
3561183 ± 2% +14.3% 4070136 ± 4% numa-meminfo.node1.Inactive
3560405 ± 2% +14.3% 4069750 ± 4% numa-meminfo.node1.Inactive(anon)
3438895 +14.8% 3948826 numa-meminfo.node1.Mapped
5457819 +12.7% 6150154 ± 4% numa-meminfo.node1.MemUsed
1037953 +15.2% 1195453 numa-meminfo.node1.PageTables
3814280 ± 2% +15.5% 4406865 ± 4% numa-meminfo.node1.Shmem
4061718 ± 3% +14.3% 4643314 ± 2% numa-meminfo.node2.FilePages
3578353 ± 3% +13.8% 4070403 ± 2% numa-meminfo.node2.Inactive
3578107 ± 3% +13.8% 4070312 ± 2% numa-meminfo.node2.Inactive(anon)
3518799 +14.1% 4015498 numa-meminfo.node2.Mapped
5391913 ± 2% +13.8% 6136705 numa-meminfo.node2.MemUsed
1034849 +15.1% 1191480 numa-meminfo.node2.PageTables
3834040 ± 3% +15.1% 4411328 ± 2% numa-meminfo.node2.Shmem
4079904 +16.1% 4734755 ± 5% numa-meminfo.node3.FilePages
3598048 +15.8% 4164785 ± 4% numa-meminfo.node3.Inactive
3598046 +15.7% 4164649 ± 4% numa-meminfo.node3.Inactive(anon)
3602664 +14.0% 4106006 ± 2% numa-meminfo.node3.Mapped
5399513 ± 2% +15.9% 6258841 ± 5% numa-meminfo.node3.MemUsed
1032679 +15.3% 1190968 numa-meminfo.node3.PageTables
3857708 ± 2% +17.1% 4515923 ± 5% numa-meminfo.node3.Shmem
71593 ± 19% +28.7% 92155 ± 16% numa-vmstat.node0.nr_active_anon
1003120 ± 3% +11.6% 1119645 numa-vmstat.node0.nr_file_pages
884514 ± 3% +10.9% 980766 numa-vmstat.node0.nr_inactive_anon
867640 +13.0% 980155 numa-vmstat.node0.nr_mapped
260068 +14.7% 298393 numa-vmstat.node0.nr_page_table_pages
948077 ± 3% +12.2% 1063329 numa-vmstat.node0.nr_shmem
71592 ± 19% +28.7% 92154 ± 16% numa-vmstat.node0.nr_zone_active_anon
884514 ± 3% +10.9% 980766 numa-vmstat.node0.nr_zone_inactive_anon
1947758 ± 5% +14.2% 2225061 ± 2% numa-vmstat.node0.numa_hit
1926345 ± 5% +14.3% 2202118 ± 3% numa-vmstat.node0.numa_local
1010900 ± 2% +14.6% 1158444 ± 3% numa-vmstat.node1.nr_file_pages
889362 ± 2% +14.5% 1018390 ± 4% numa-vmstat.node1.nr_inactive_anon
859119 +15.0% 988130 numa-vmstat.node1.nr_mapped
259869 +15.1% 299005 numa-vmstat.node1.nr_page_table_pages
952777 ± 2% +15.8% 1103035 ± 3% numa-vmstat.node1.nr_shmem
889362 ± 2% +14.5% 1018390 ± 4% numa-vmstat.node1.nr_zone_inactive_anon
1891781 ± 4% +13.6% 2148479 ± 4% numa-vmstat.node1.numa_hit
1800381 ± 4% +14.0% 2053014 ± 4% numa-vmstat.node1.numa_local
1014637 ± 3% +14.5% 1162201 ± 2% numa-vmstat.node2.nr_file_pages
893784 ± 3% +14.0% 1018557 ± 2% numa-vmstat.node2.nr_inactive_anon
879122 +14.3% 1004884 numa-vmstat.node2.nr_mapped
259099 +15.1% 298139 numa-vmstat.node2.nr_page_table_pages
957718 ± 3% +15.3% 1104205 ± 2% numa-vmstat.node2.nr_shmem
893784 ± 3% +14.0% 1018557 ± 2% numa-vmstat.node2.nr_zone_inactive_anon
1919426 ± 2% +15.0% 2206628 ± 5% numa-vmstat.node2.numa_hit
1828624 ± 2% +15.7% 2116391 ± 5% numa-vmstat.node2.numa_local
1019191 +16.3% 1185073 ± 4% numa-vmstat.node3.nr_file_pages
898768 +16.0% 1042162 ± 4% numa-vmstat.node3.nr_inactive_anon
900011 +14.2% 1027459 ± 2% numa-vmstat.node3.nr_mapped
258567 +15.3% 298137 numa-vmstat.node3.nr_page_table_pages
963643 +17.3% 1130365 ± 5% numa-vmstat.node3.nr_shmem
898768 +16.0% 1042162 ± 4% numa-vmstat.node3.nr_zone_inactive_anon
1891488 +18.0% 2231718 ± 5% numa-vmstat.node3.numa_hit
1790050 ± 2% +19.3% 2135992 ± 6% numa-vmstat.node3.numa_local
0.64 ± 5% +0.1 0.75 perf-profile.calltrace.cycles-pp.menu_select.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64
1.78 ± 5% +0.1 1.90 perf-profile.calltrace.cycles-pp.__wake_up_common.wake_up_page_bit.filemap_map_pages.__handle_mm_fault.handle_mm_fault
0.70 ± 8% +0.2 0.85 ± 2% perf-profile.calltrace.cycles-pp.save_stack_trace_tsk.__account_scheduler_latency.enqueue_entity.enqueue_task_fair.ttwu_do_activate
0.59 ± 4% +0.2 0.75 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.finish_wait.__lock_page.find_lock_entry
0.65 ± 8% +0.2 0.81 perf-profile.calltrace.cycles-pp.__save_stack_trace.save_stack_trace_tsk.__account_scheduler_latency.enqueue_entity.enqueue_task_fair
0.62 ± 4% +0.2 0.79 perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.finish_wait.__lock_page.find_lock_entry.shmem_getpage_gfp
0.93 ± 6% +0.2 1.11 perf-profile.calltrace.cycles-pp.try_to_wake_up.autoremove_wake_function.__wake_up_common.wake_up_page_bit.filemap_map_pages
0.64 ± 3% +0.2 0.82 perf-profile.calltrace.cycles-pp.finish_wait.__lock_page.find_lock_entry.shmem_getpage_gfp.shmem_fault
1.00 ± 5% +0.2 1.18 perf-profile.calltrace.cycles-pp.autoremove_wake_function.__wake_up_common.wake_up_page_bit.filemap_map_pages.__handle_mm_fault
0.81 ± 3% +0.2 1.01 ± 3% perf-profile.calltrace.cycles-pp.__sched_text_start.schedule.io_schedule.__lock_page.find_lock_entry
0.61 ± 6% +0.2 0.81 ± 3% perf-profile.calltrace.cycles-pp.__wake_up_common.wake_up_page_bit.__handle_mm_fault.handle_mm_fault.__do_page_fault
0.97 ± 4% +0.2 1.17 perf-profile.calltrace.cycles-pp.find_get_entry.find_lock_entry.shmem_getpage_gfp.shmem_fault.__do_fault
0.83 ± 3% +0.2 1.03 ± 3% perf-profile.calltrace.cycles-pp.schedule.io_schedule.__lock_page.find_lock_entry.shmem_getpage_gfp
0.83 ± 3% +0.2 1.04 ± 3% perf-profile.calltrace.cycles-pp.io_schedule.__lock_page.find_lock_entry.shmem_getpage_gfp.shmem_fault
0.97 ± 7% +0.3 1.26 ± 3% perf-profile.calltrace.cycles-pp.__account_scheduler_latency.enqueue_entity.enqueue_task_fair.ttwu_do_activate.sched_ttwu_pending
1.28 ± 6% +0.4 1.64 ± 2% perf-profile.calltrace.cycles-pp.enqueue_entity.enqueue_task_fair.ttwu_do_activate.sched_ttwu_pending.do_idle
1.36 ± 7% +0.4 1.72 ± 2% perf-profile.calltrace.cycles-pp.enqueue_task_fair.ttwu_do_activate.sched_ttwu_pending.do_idle.cpu_startup_entry
1.38 ± 6% +0.4 1.75 ± 2% perf-profile.calltrace.cycles-pp.ttwu_do_activate.sched_ttwu_pending.do_idle.cpu_startup_entry.start_secondary
1.61 ± 6% +0.4 2.04 ± 2% perf-profile.calltrace.cycles-pp.sched_ttwu_pending.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64
0.00 +0.5 0.53 ± 4% perf-profile.calltrace.cycles-pp.try_to_wake_up.autoremove_wake_function.__wake_up_common.wake_up_page_bit.__handle_mm_fault
0.12 ±173% +0.5 0.66 ± 2% perf-profile.calltrace.cycles-pp.finish_fault.__handle_mm_fault.handle_mm_fault.__do_page_fault.do_page_fault
0.00 +0.6 0.56 ± 3% perf-profile.calltrace.cycles-pp.autoremove_wake_function.__wake_up_common.wake_up_page_bit.__handle_mm_fault.handle_mm_fault
0.00 +0.6 0.64 ± 2% perf-profile.calltrace.cycles-pp.alloc_set_pte.finish_fault.__handle_mm_fault.handle_mm_fault.__do_page_fault
5.88 ± 5% +0.7 6.56 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irq.__lock_page.find_lock_entry.shmem_getpage_gfp
6.06 ± 5% +0.7 6.79 perf-profile.calltrace.cycles-pp._raw_spin_lock_irq.__lock_page.find_lock_entry.shmem_getpage_gfp.shmem_fault
8.63 ± 4% +1.3 9.90 perf-profile.calltrace.cycles-pp.__lock_page.find_lock_entry.shmem_getpage_gfp.shmem_fault.__do_fault
5.63 ± 5% +1.4 7.00 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.wake_up_page_bit.__handle_mm_fault.handle_mm_fault
5.72 ± 5% +1.4 7.13 perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.wake_up_page_bit.__handle_mm_fault.handle_mm_fault.__do_page_fault
9.99 ± 4% +1.6 11.61 perf-profile.calltrace.cycles-pp.find_lock_entry.shmem_getpage_gfp.shmem_fault.__do_fault.__handle_mm_fault
6.50 ± 4% +1.7 8.17 perf-profile.calltrace.cycles-pp.wake_up_page_bit.__handle_mm_fault.handle_mm_fault.__do_page_fault.do_page_fault
10.24 ± 4% +1.7 11.95 perf-profile.calltrace.cycles-pp.shmem_getpage_gfp.shmem_fault.__do_fault.__handle_mm_fault.handle_mm_fault
10.30 ± 4% +1.7 12.04 perf-profile.calltrace.cycles-pp.shmem_fault.__do_fault.__handle_mm_fault.handle_mm_fault.__do_page_fault
10.35 ± 4% +1.8 12.10 perf-profile.calltrace.cycles-pp.__do_fault.__handle_mm_fault.handle_mm_fault.__do_page_fault.do_page_fault
69.78 ± 5% +2.4 72.22 perf-profile.calltrace.cycles-pp.__handle_mm_fault.handle_mm_fault.__do_page_fault.do_page_fault.page_fault
18.70 ± 5% +3.2 21.90 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.wake_up_page_bit.filemap_map_pages.__handle_mm_fault
19.11 ± 5% +3.2 22.33 perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.wake_up_page_bit.filemap_map_pages.__handle_mm_fault.handle_mm_fault
21.54 ± 5% +3.5 25.05 perf-profile.calltrace.cycles-pp.wake_up_page_bit.filemap_map_pages.__handle_mm_fault.handle_mm_fault.__do_page_fault
54.16 ± 57% +18.2 72.39 perf-profile.calltrace.cycles-pp.handle_mm_fault.__do_page_fault.do_page_fault.page_fault
54.57 ± 57% +18.4 73.00 perf-profile.calltrace.cycles-pp.__do_page_fault.do_page_fault.page_fault
54.65 ± 57% +18.5 73.12 perf-profile.calltrace.cycles-pp.do_page_fault.page_fault
54.66 ± 57% +18.5 73.15 perf-profile.calltrace.cycles-pp.page_fault
1.982e+10 +13.8% 2.256e+10 perf-stat.i.branch-instructions
54498559 +18.2% 64421735 ± 2% perf-stat.i.branch-misses
72537292 +15.9% 84041108 perf-stat.i.cache-misses
2.637e+08 +14.6% 3.022e+08 perf-stat.i.cache-references
528852 +22.1% 645921 ± 2% perf-stat.i.context-switches
2.13 ± 3% -7.4% 1.97 ± 3% perf-stat.i.cpi
1.309e+11 +3.1% 1.349e+11 perf-stat.i.cpu-cycles
0.11 ± 6% +0.0 0.13 ± 7% perf-stat.i.dTLB-load-miss-rate%
16848866 ± 4% +33.7% 22519926 ± 6% perf-stat.i.dTLB-load-misses
1.688e+10 +13.8% 1.922e+10 perf-stat.i.dTLB-loads
495556 ± 5% +25.2% 620200 ± 2% perf-stat.i.dTLB-store-misses
3.06e+09 +15.9% 3.546e+09 perf-stat.i.dTLB-stores
29.46 ± 2% +1.9 31.36 ± 3% perf-stat.i.iTLB-load-miss-rate%
2063983 ± 3% +29.2% 2666300 perf-stat.i.iTLB-load-misses
5683347 ± 2% +16.4% 6616736 ± 2% perf-stat.i.iTLB-loads
6.558e+10 +13.9% 7.471e+10 perf-stat.i.instructions
0.50 +10.0% 0.55 perf-stat.i.ipc
911272 +30.3% 1187057 perf-stat.i.minor-faults
44168476 +18.1% 52140968 perf-stat.i.node-load-misses
1432172 +16.9% 1674534 perf-stat.i.node-loads
83.66 -1.4 82.24 perf-stat.i.node-store-miss-rate%
22604189 +10.5% 24982847 perf-stat.i.node-store-misses
4317350 +21.1% 5229582 perf-stat.i.node-stores
911272 +30.3% 1187057 perf-stat.i.page-faults
2.00 -9.5% 1.81 perf-stat.overall.cpi
0.10 ± 3% +0.0 0.12 ± 6% perf-stat.overall.dTLB-load-miss-rate%
0.02 ± 5% +0.0 0.02 ± 3% perf-stat.overall.dTLB-store-miss-rate%
26.64 +2.1 28.73 perf-stat.overall.iTLB-load-miss-rate%
31807 ± 3% -11.9% 28021 perf-stat.overall.instructions-per-iTLB-miss
0.50 +10.5% 0.55 perf-stat.overall.ipc
83.96 -1.3 82.69 perf-stat.overall.node-store-miss-rate%
2339 +8.0% 2526 perf-stat.overall.path-length
6.298e+12 +15.7% 7.284e+12 perf-stat.total.branch-instructions
1.732e+10 +20.1% 2.081e+10 ± 2% perf-stat.total.branch-misses
2.305e+10 +17.7% 2.714e+10 perf-stat.total.cache-misses
8.378e+10 +16.5% 9.757e+10 perf-stat.total.cache-references
1.68e+08 +24.1% 2.085e+08 perf-stat.total.context-switches
4.159e+13 +4.8% 4.358e+13 perf-stat.total.cpu-cycles
5.353e+09 ± 3% +35.8% 7.269e+09 ± 5% perf-stat.total.dTLB-load-misses
5.364e+12 +15.7% 6.206e+12 perf-stat.total.dTLB-loads
1.575e+08 ± 5% +27.2% 2.004e+08 ± 3% perf-stat.total.dTLB-store-misses
9.721e+11 +17.8% 1.145e+12 perf-stat.total.dTLB-stores
6.558e+08 ± 3% +31.3% 8.611e+08 perf-stat.total.iTLB-load-misses
1.806e+09 ± 3% +18.3% 2.136e+09 perf-stat.total.iTLB-loads
2.084e+13 +15.8% 2.412e+13 perf-stat.total.instructions
2.896e+08 +32.4% 3.833e+08 perf-stat.total.minor-faults
1.403e+10 +20.0% 1.684e+10 perf-stat.total.node-load-misses
4.551e+08 +18.9% 5.409e+08 perf-stat.total.node-loads
7.182e+09 +12.3% 8.067e+09 perf-stat.total.node-store-misses
1.372e+09 +23.1% 1.689e+09 perf-stat.total.node-stores
2.896e+08 +32.4% 3.833e+08 perf-stat.total.page-faults
vm-scalability.time.user_time
8000 +-+------------------------------------------------------------------+
| O |
7500 +-+ |
| |
| |
7000 +-+ |
| |
6500 +-+ |
O OO O O O O OO O O O OO O O O O OO |
6000 +-+ + |
| :: |
| +. : : |
5500 +-+.++. .+. .+. .++. .+.+. .+.+. : +. .+. : +. |
|.+ + +.+.++.+ + +.+ ++.+ + + ++ +.+.++.+.|
5000 +-+------------------------------------------------------------------+
vm-scalability.time.percent_of_cpu_this_job_got
11000 +-+-----------------------------------------------------------------+
| |
10900 +-+ O O O |
O O O O O OO O OO O O O O |
10800 +-+ O O O |
| O |
10700 +-+ |
| |
10600 +-+ |
| .+ .+. .+ |
10500 +-+ .+ + + .+ : + + :.+.+.+. |
|.+.++.+.+ + + + + + : .++.+.+. .++.+.+. .+ + ++.+.|
10400 +-+ ++ + + + + |
| |
10300 +-+-----------------------------------------------------------------+
vm-scalability.time.maximum_resident_set_size
3.4e+07 +-O---------O-----------------------------------------------------+
O O O O O O OO O O O |
3.3e+07 +-+ O OO OO OO O |
| |
| |
3.2e+07 +-+ |
| |
3.1e+07 +-+ |
| |
3e+07 +-+ |
| |
| +. .+ + +. .++. .+ .+. .+. |
2.9e+07 +-+ +.+ +. + :+ + +.+ +.+.++ +.++ +.++.+. .++.+.+.++.|
| +.+ + + |
2.8e+07 +-+---------------------------------------------------------------+
vm-scalability.time.minor_page_faults
3.9e+08 +-+---------------------------------------------------------------+
3.8e+08 O-OO O O OO O O OO O O OO O O OO O O |
| |
3.7e+08 +-+ |
3.6e+08 +-+ |
3.5e+08 +-+ |
3.4e+08 +-+ |
| |
3.3e+08 +-+ |
3.2e+08 +-+ |
3.1e+08 +-+ |
3e+08 +-+ |
| .+ |
2.9e+08 +-++.+.+.++.+.+.++.+.+ +.+.+.++.+.++.+.+.++.+.+.++.+.+.++.+.+.++.|
2.8e+08 +-+---------------------------------------------------------------+
vm-scalability.time.voluntary_context_switches
1.1e+08 +-+--------------------------------------------------------------+
| |
1.05e+08 +-+O O O O O |
O O O O OO O O OO O O O O |
| O |
1e+08 +-O |
| |
9.5e+07 +-+ |
| |
9e+07 +-+ |
| |
| |
8.5e+07 +-++.+.+. .+.+ .+.+.++. .++.+.+.++.+.++.+.+ .++. .+ .+.+.++.|
| ++ + + +.+.+ + + |
8e+07 +-+--------------------------------------------------------------+
vm-scalability.time.involuntary_context_switches
230000 +-+----------------------------------------------------------------+
| O |
225000 +-+ |
220000 +-+ |
| O |
215000 +-+ O O O |
210000 +-+ |
| O O O O O O O O |
205000 O-O O O O O O O |
200000 +-+ |
| |
195000 +-+ +. .+. + + +.+. +. +. .+.++. |
190000 +-+ .+. +.+.+.+ + + + :+ + + +. .+ + +.|
| +.++.+ + +.+.+.+ +.+.++ + |
185000 +-+----------------------------------------------------------------+
syscalls.sys_close.noise.100_
5.5e+07 +-+---------------------------------------------------------------+
| + |
5e+07 +-+ : |
| + +. ++ : : |
| + +. .+. .+ + .+ + : + +. + + : : .|
4.5e+07 +-+ + + ++ + .+.+.++. .+ :+ + + + + : + + + :+ |
| + + + + + + + +.+ + |
4e+07 +-+ + |
| |
3.5e+07 +-+ |
| |
| O |
3e+07 +-+O O O O O O O O O O |
| O O O O O O O O |
2.5e+07 O-+------O--------------------------------------------------------+
syscalls.sys_close.noise.75_
6e+07 +-+---------------------------------------------------------------+
| + |
5.5e+07 +-+ :: |
| + +.+ ++ : : |
| + +. .+. .+ +. .+ + : + + + + : : +|
5e+07 +-+ + + ++ + .+.+.++.+.+ :+ ++ + + : + + + ++ |
| + + + + + +.+ |
4.5e+07 +-+ + |
| |
4e+07 +-+ |
| |
| O O O |
3.5e+07 +-OO O O O O O O O O |
| O O O O O O O |
3e+07 O-+---------------------------------------------------------------+
syscalls.sys_close.noise.50_
6.5e+07 +-+---------------------------------------------------------------+
| : |
6e+07 +-+ : : |
| + + + + +.+ ++. : : .|
| :: +. .+. +: +. + : + : + + + + :+ |
5.5e+07 +-+ : + ++ + .+.+.++.+.+ :+ ++ : + : + + + |
| + + + + + +.+ |
5e+07 +-+ + |
| |
4.5e+07 +-+ |
| |
| O O OO O O |
4e+07 +-OO O O O O O O |
O O O O O O O |
3.5e+07 +-+---------------------------------------------------------------+
syscalls.sys_close.noise.25_
7e+07 +-+---------------------------------------------------------------+
| + |
6.5e+07 +-+ + : : |
| + + + + +. + + : : |
| :: + .+. +: +. + : + : + +. + + : +|
6e+07 +-+ : + + + + .+.+ .+.+ :+ ++ : + : : + ++ |
| + + : +.+ + + + +. : |
5.5e+07 +-+ : : + |
| + |
5e+07 +-+ |
| O |
| O O O O O |
4.5e+07 +-OO O O O OO O O O O |
O O O O |
4e+07 +-+---------------------------------------------------------------+
syscalls.sys_close.noise.5_
7.5e+07 +-+---------------------------------------------------------------+
| + |
7e+07 +-+ : : |
| + + + + +. ++. : : .|
6.5e+07 +-+:+ .+. .+. + :.+. + + .+ : + +. + + :+ |
|.+ + ++ + .+.+.++.+.+ + ++ + : + + + |
6e+07 +-+ + + +.+ |
| + |
5.5e+07 +-+ |
| OO |
5e+07 +-+O O O O O |
| O O O O OO O O OO O O |
4.5e+07 O-+ O |
| |
4e+07 +-+---------------------------------------------------------------+
syscalls.sys_close.noise.2_
7.5e+07 +-+---------------------------------------------------------------+
| + |
7e+07 +-+ : |
| +. ++ : : |
| + + + + + + ++ : +. + + : : .|
6.5e+07 +-+:+ + + + + .+. + :+ + + + + : + + + :+ |
|.+ + ++ + +.+ ++.+.+ + ++ + : + + |
6e+07 +-+ + : +.+ |
| + |
5.5e+07 +-+ |
| |
| O OO |
5e+07 +-OO O O O O O O O |
| O O OO O O O |
4.5e+07 O-+------O--------------------------------------------------------+
vm-scalability.throughput
3.25e+07 +-+--------------------------------------------------------------+
| |
3.2e+07 O-O O O O OO O OO |
| O O O OO O O O O OO |
3.15e+07 +-+ |
| |
3.1e+07 +-+ |
| |
3.05e+07 +-+ |
| |
3e+07 +-+ + +. +. |
| +.+. + : +. + + +. .+.+. .++. .++.+.+. +. .+ .+.+ .|
2.95e+07 +-+ + +.+.+ + ++ ++.+ + + + +.+ + |
| |
2.9e+07 +-+--------------------------------------------------------------+
vm-scalability.median
224000 +-+----------------------------------------------------------------+
222000 +-O O O O O O |
O OO O O O O O O O O |
220000 +-+ O O O O |
218000 +-+ |
| |
216000 +-+ |
214000 +-+ |
212000 +-+ |
| |
210000 +-+ |
208000 +-+ + +. .+. |
| .++. + + + : + + .+.+. .+.+ .+.+ |
206000 +-+ + ++. + +: +.+ ++.+ +.+ +.+.+.++.+.+.++.+.|
204000 +-+----------------------------------------------------------------+
vm-scalability.workload
9.7e+09 +-+---------------------------------------------------------------+
| O O O O OO O |
9.6e+09 O-+O O O O O O OO O |
9.5e+09 +-+ O O O O |
| |
9.4e+09 +-+ |
9.3e+09 +-+ |
| |
9.2e+09 +-+ |
9.1e+09 +-+ |
| |
9e+09 +-+ .+ +. .++ .+. |
8.9e+09 +-++.+.+ +. + + + + .++.+. .+.+.++ +.++.+. .+ .+.+ .|
|.+ +. + + +.+ ++ + +.+ + |
8.8e+09 +-+---------------------------------------------------------------+
pmeter.performance_per_watt
63500 +-+-----------------------------------------------------------------+
63000 O-O O |
| O OO O O O O |
62500 +-+ OO O O O OO O |
62000 +-+ O O O |
| |
61500 +-+ |
61000 +-+ |
60500 +-+ |
| |
60000 +-+ + |
59500 +-+ .+ + .+.+ .+ +: + |
| .++.+ + + : + + +.+.+. +.+.+.+ +. .+ :+ + .++. .|
59000 +-+ +.+ + + : +.+ +.+ + + +.+ + |
58500 +-+-----------------------------------------------------------------+
[*] bisect-good sample
[O] bisect-bad sample
***************************************************************************************************
lkp-skl-fpga01: 104 threads Skylake with 192G memory
=========================================================================================
compiler/cpufreq_governor/kconfig/rootfs/runtime/size/tbox_group/test/testcase:
gcc-7/performance/x86_64-rhel-7.2/debian-x86_64-2018-04-03.cgz/300s/2T/lkp-skl-fpga01/shm-xread-seq-mt/vm-scalability
commit:
bb8965bd82 ("mm, memory_hotplug: deobfuscate migration part of offlining")
e0975b2aae ("mm, fault_around: do not take a reference to a locked page")
bb8965bd82fd4ed4 e0975b2aae0e669f995f7d5f11
---------------- --------------------------
%stddev %change %stddev
\ | \
144234 +7.8% 155452 vm-scalability.median
15000404 +7.8% 16167083 vm-scalability.throughput
43249 +3.5% 44781 vm-scalability.time.involuntary_context_switches
31039428 +10.9% 34430051 vm-scalability.time.maximum_resident_set_size
1.201e+08 +10.7% 1.329e+08 vm-scalability.time.minor_page_faults
1941 +2.2% 1984 vm-scalability.time.percent_of_cpu_this_job_got
1158 +12.1% 1298 vm-scalability.time.user_time
1.265e+08 +12.9% 1.428e+08 vm-scalability.time.voluntary_context_switches
4.507e+09 +7.8% 4.859e+09 vm-scalability.workload
6890003 ± 8% +15.7% 7968461 ± 2% cpuidle.POLL.usage
3.43 +0.4 3.84 mpstat.cpu.usr%
16288082 +11.4% 18140991 vmstat.memory.cache
811657 +14.0% 925028 vmstat.system.cs
562.75 +3.6% 582.75 turbostat.Avg_MHz
22.25 ± 3% +3.1 25.36 ± 5% turbostat.C1%
245.19 +1.3% 248.44 turbostat.PkgWatt
5678487 +17.6% 6679342 numa-numastat.node0.local_node
5683695 +17.7% 6687946 numa-numastat.node0.numa_hit
5582601 +19.8% 6687333 numa-numastat.node1.local_node
5597825 +19.7% 6699230 numa-numastat.node1.numa_hit
16243417 +11.3% 18079338 meminfo.Cached
16295961 +11.5% 18172954 meminfo.Committed_AS
15317570 +12.0% 17161623 meminfo.Inactive
15316414 +12.0% 17160456 meminfo.Inactive(anon)
15041014 +11.9% 16838110 meminfo.Mapped
17282072 +10.7% 19127322 meminfo.Memused
33914 +10.5% 37473 meminfo.PageTables
15342378 +12.0% 17183105 meminfo.Shmem
106698 +11.2% 118601 meminfo.max_used_kB
8292175 +10.1% 9127867 ± 2% numa-meminfo.node0.FilePages
7819565 +10.8% 8667082 ± 2% numa-meminfo.node0.Inactive
7819192 +10.8% 8666496 ± 2% numa-meminfo.node0.Inactive(anon)
7673978 +11.0% 8519916 ± 3% numa-meminfo.node0.Mapped
7837016 +10.8% 8680949 ± 2% numa-meminfo.node0.Shmem
7944534 ± 2% +12.6% 8942310 numa-meminfo.node1.FilePages
7490327 ± 2% +13.3% 8484321 numa-meminfo.node1.Inactive
7489544 ± 2% +13.3% 8483741 numa-meminfo.node1.Inactive(anon)
7359777 +12.9% 8305653 numa-meminfo.node1.Mapped
8433516 ± 2% +12.4% 9477657 numa-meminfo.node1.MemUsed
16072 ± 6% +16.5% 18722 ± 5% numa-meminfo.node1.PageTables
7497615 ± 2% +13.3% 8491945 numa-meminfo.node1.Shmem
2071045 +10.3% 2284635 ± 2% numa-vmstat.node0.nr_file_pages
1952804 +11.1% 2169301 ± 2% numa-vmstat.node0.nr_inactive_anon
1916283 +11.3% 2133131 ± 3% numa-vmstat.node0.nr_mapped
1957255 +11.0% 2172906 ± 2% numa-vmstat.node0.nr_shmem
1952803 +11.1% 2169301 ± 2% numa-vmstat.node0.nr_zone_inactive_anon
3503566 +12.8% 3951676 numa-vmstat.node0.numa_hit
3498309 ± 2% +12.7% 3942970 numa-vmstat.node0.numa_local
1984140 ± 2% +12.8% 2238185 numa-vmstat.node1.nr_file_pages
1870387 ± 2% +13.5% 2123528 numa-vmstat.node1.nr_inactive_anon
1837863 +13.1% 2079499 numa-vmstat.node1.nr_mapped
4004 ± 6% +16.9% 4680 ± 5% numa-vmstat.node1.nr_page_table_pages
1872410 ± 2% +13.5% 2125594 numa-vmstat.node1.nr_shmem
1870387 ± 2% +13.5% 2123528 numa-vmstat.node1.nr_zone_inactive_anon
3278102 ± 3% +17.4% 3847219 numa-vmstat.node1.numa_hit
3114333 ± 3% +18.4% 3687396 ± 2% numa-vmstat.node1.numa_local
4458368 -1.0% 4412087 proc-vmstat.nr_dirty_background_threshold
8927638 -1.0% 8834963 proc-vmstat.nr_dirty_threshold
4057685 +11.4% 4518643 proc-vmstat.nr_file_pages
44867224 -1.0% 44403734 proc-vmstat.nr_free_pages
3825717 +12.1% 4288625 proc-vmstat.nr_inactive_anon
3756585 +12.0% 4208113 proc-vmstat.nr_mapped
8471 +10.3% 9342 proc-vmstat.nr_page_table_pages
3832166 +12.1% 4294323 proc-vmstat.nr_shmem
26805 +4.1% 27891 proc-vmstat.nr_slab_reclaimable
3825717 +12.1% 4288625 proc-vmstat.nr_zone_inactive_anon
11306437 +18.6% 13411049 proc-vmstat.numa_hit
11285983 +18.6% 13390541 proc-vmstat.numa_local
9922 ± 2% -14.3% 8503 ± 7% proc-vmstat.pgactivate
11384788 +18.5% 13491951 proc-vmstat.pgalloc_normal
1.209e+08 +10.6% 1.337e+08 proc-vmstat.pgfault
11251068 +19.6% 13459118 proc-vmstat.pgfree
6068 ± 26% +76.5% 10712 ± 19% sched_debug.cfs_rq:/.MIN_vruntime.avg
6068 ± 26% +76.5% 10712 ± 19% sched_debug.cfs_rq:/.max_vruntime.avg
1.11 ± 27% +72.0% 1.90 ± 16% sched_debug.cfs_rq:/.removed.util_avg.avg
8.67 ± 6% +39.6% 12.10 ± 10% sched_debug.cfs_rq:/.removed.util_avg.stddev
747207 ± 8% -29.4% 527807 ± 16% sched_debug.cpu.avg_idle.max
103865 ± 5% -20.7% 82331 ± 7% sched_debug.cpu.avg_idle.stddev
5.21 ± 11% -13.9% 4.49 ± 7% sched_debug.cpu.cpu_load[1].stddev
1209506 +12.8% 1364777 sched_debug.cpu.nr_switches.avg
1227033 +12.5% 1380943 sched_debug.cpu.nr_switches.max
1167455 +14.4% 1335633 sched_debug.cpu.nr_switches.min
1208675 +12.8% 1363928 sched_debug.cpu.sched_count.avg
1223481 +12.6% 1377943 sched_debug.cpu.sched_count.max
1166534 +14.4% 1334716 sched_debug.cpu.sched_count.min
603965 +12.8% 681563 sched_debug.cpu.sched_goidle.avg
611293 +12.6% 688570 sched_debug.cpu.sched_goidle.max
582920 +14.4% 666940 sched_debug.cpu.sched_goidle.min
604448 +12.8% 682101 sched_debug.cpu.ttwu_count.avg
620456 +12.7% 699432 sched_debug.cpu.ttwu_count.max
577893 ± 2% +14.2% 659765 sched_debug.cpu.ttwu_count.min
1.404e+10 +11.4% 1.563e+10 perf-stat.i.branch-instructions
33475384 +10.7% 37068384 perf-stat.i.branch-misses
2.359e+08 ± 3% +7.3% 2.53e+08 perf-stat.i.cache-references
821488 +13.8% 934814 perf-stat.i.context-switches
5.84e+10 +3.6% 6.048e+10 perf-stat.i.cpu-cycles
2388 +10.6% 2640 ± 2% perf-stat.i.cpu-migrations
2559882 ± 2% +15.0% 2944062 ± 5% perf-stat.i.dTLB-load-misses
1.155e+10 +11.3% 1.285e+10 perf-stat.i.dTLB-loads
459268 +11.0% 509969 perf-stat.i.dTLB-store-misses
2.446e+09 +11.3% 2.723e+09 perf-stat.i.dTLB-stores
821295 +10.4% 906622 perf-stat.i.iTLB-load-misses
6623634 +9.4% 7245175 perf-stat.i.iTLB-loads
4.496e+10 +11.2% 5e+10 perf-stat.i.instructions
0.76 +7.9% 0.82 perf-stat.i.ipc
391866 +11.5% 436886 perf-stat.i.minor-faults
4317157 +9.8% 4742283 perf-stat.i.node-store-misses
391866 +11.5% 436886 perf-stat.i.page-faults
1.30 -6.9% 1.21 perf-stat.overall.cpi
0.77 +7.4% 0.83 perf-stat.overall.ipc
3077 +2.3% 3148 perf-stat.overall.path-length
4.329e+12 +10.5% 4.783e+12 perf-stat.total.branch-instructions
1.033e+10 +9.8% 1.134e+10 perf-stat.total.branch-misses
7.276e+10 ± 3% +6.4% 7.742e+10 perf-stat.total.cache-references
2.534e+08 +12.9% 2.86e+08 perf-stat.total.context-switches
1.802e+13 +2.7% 1.851e+13 perf-stat.total.cpu-cycles
736807 +9.6% 807746 perf-stat.total.cpu-migrations
7.897e+08 ± 2% +14.1% 9.009e+08 ± 6% perf-stat.total.dTLB-load-misses
3.562e+12 +10.4% 3.931e+12 perf-stat.total.dTLB-loads
1.417e+08 +10.1% 1.56e+08 perf-stat.total.dTLB-store-misses
7.544e+11 +10.4% 8.331e+11 perf-stat.total.dTLB-stores
2.534e+08 ± 2% +9.5% 2.774e+08 perf-stat.total.iTLB-load-misses
2.043e+09 +8.5% 2.217e+09 perf-stat.total.iTLB-loads
1.387e+13 +10.3% 1.53e+13 perf-stat.total.instructions
1.209e+08 +10.6% 1.337e+08 perf-stat.total.minor-faults
1.332e+09 +9.0% 1.451e+09 perf-stat.total.node-store-misses
1.209e+08 +10.6% 1.337e+08 perf-stat.total.page-faults
42.55 -0.9 41.64 perf-profile.calltrace.cycles-pp.__handle_mm_fault.handle_mm_fault.__do_page_fault.do_page_fault.page_fault
42.63 -0.9 41.73 perf-profile.calltrace.cycles-pp.handle_mm_fault.__do_page_fault.do_page_fault.page_fault
42.99 -0.9 42.13 perf-profile.calltrace.cycles-pp.do_page_fault.page_fault
43.00 -0.9 42.14 perf-profile.calltrace.cycles-pp.page_fault
42.96 -0.9 42.11 perf-profile.calltrace.cycles-pp.__do_page_fault.do_page_fault.page_fault
20.69 -0.8 19.89 perf-profile.calltrace.cycles-pp.filemap_map_pages.__handle_mm_fault.handle_mm_fault.__do_page_fault.do_page_fault
16.44 -0.7 15.79 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock.alloc_set_pte.finish_fault.__handle_mm_fault
16.59 -0.6 15.95 perf-profile.calltrace.cycles-pp.alloc_set_pte.finish_fault.__handle_mm_fault.handle_mm_fault.__do_page_fault
16.55 -0.6 15.90 perf-profile.calltrace.cycles-pp._raw_spin_lock.alloc_set_pte.finish_fault.__handle_mm_fault.handle_mm_fault
16.61 -0.6 15.98 perf-profile.calltrace.cycles-pp.finish_fault.__handle_mm_fault.handle_mm_fault.__do_page_fault.do_page_fault
0.83 +0.1 0.89 ± 2% perf-profile.calltrace.cycles-pp.menu_select.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64
0.54 ± 2% +0.1 0.60 perf-profile.calltrace.cycles-pp.__sched_text_start.schedule_idle.do_idle.cpu_startup_entry.start_secondary
0.71 ± 2% +0.1 0.78 ± 2% perf-profile.calltrace.cycles-pp.__save_stack_trace.save_stack_trace_tsk.__account_scheduler_latency.enqueue_entity.enqueue_task_fair
0.56 ± 2% +0.1 0.63 perf-profile.calltrace.cycles-pp.schedule_idle.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64
0.76 ± 2% +0.1 0.82 ± 2% perf-profile.calltrace.cycles-pp.save_stack_trace_tsk.__account_scheduler_latency.enqueue_entity.enqueue_task_fair.ttwu_do_activate
1.56 +0.1 1.66 perf-profile.calltrace.cycles-pp.autoremove_wake_function.__wake_up_common.wake_up_page_bit.__handle_mm_fault.handle_mm_fault
0.75 ± 6% +0.1 0.85 ± 3% perf-profile.calltrace.cycles-pp.poll_idle.cpuidle_enter_state.do_idle.cpu_startup_entry.start_secondary
1.48 +0.1 1.59 perf-profile.calltrace.cycles-pp.try_to_wake_up.autoremove_wake_function.__wake_up_common.wake_up_page_bit.__handle_mm_fault
1.71 +0.1 1.82 perf-profile.calltrace.cycles-pp.__wake_up_common.wake_up_page_bit.__handle_mm_fault.handle_mm_fault.__do_page_fault
1.84 +0.1 1.96 perf-profile.calltrace.cycles-pp.wake_up_page_bit.__handle_mm_fault.handle_mm_fault.__do_page_fault.do_page_fault
1.28 +0.2 1.44 perf-profile.calltrace.cycles-pp.try_to_wake_up.autoremove_wake_function.__wake_up_common.wake_up_page_bit.filemap_map_pages
1.23 ± 2% +0.2 1.40 perf-profile.calltrace.cycles-pp.io_schedule.__lock_page.find_lock_entry.shmem_getpage_gfp.shmem_fault
1.22 ± 2% +0.2 1.39 perf-profile.calltrace.cycles-pp.schedule.io_schedule.__lock_page.find_lock_entry.shmem_getpage_gfp
1.20 ± 2% +0.2 1.37 perf-profile.calltrace.cycles-pp.__sched_text_start.schedule.io_schedule.__lock_page.find_lock_entry
1.35 +0.2 1.52 ± 2% perf-profile.calltrace.cycles-pp.autoremove_wake_function.__wake_up_common.wake_up_page_bit.filemap_map_pages.__handle_mm_fault
0.39 ± 57% +0.2 0.56 perf-profile.calltrace.cycles-pp.unwind_next_frame.__save_stack_trace.save_stack_trace_tsk.__account_scheduler_latency.enqueue_entity
1.51 +0.2 1.71 perf-profile.calltrace.cycles-pp.__wake_up_common.wake_up_page_bit.filemap_map_pages.__handle_mm_fault.handle_mm_fault
1.98 ± 3% +0.2 2.21 perf-profile.calltrace.cycles-pp.__lock_page.find_lock_entry.shmem_getpage_gfp.shmem_fault.__do_fault
2.27 ± 3% +0.2 2.51 perf-profile.calltrace.cycles-pp.find_lock_entry.shmem_getpage_gfp.shmem_fault.__do_fault.__handle_mm_fault
1.68 +0.2 1.92 perf-profile.calltrace.cycles-pp.wake_up_page_bit.filemap_map_pages.__handle_mm_fault.handle_mm_fault.__do_page_fault
2.56 ± 3% +0.3 2.83 perf-profile.calltrace.cycles-pp.shmem_getpage_gfp.shmem_fault.__do_fault.__handle_mm_fault.handle_mm_fault
2.60 ± 3% +0.3 2.88 perf-profile.calltrace.cycles-pp.shmem_fault.__do_fault.__handle_mm_fault.handle_mm_fault.__do_page_fault
2.61 ± 3% +0.3 2.90 perf-profile.calltrace.cycles-pp.__do_fault.__handle_mm_fault.handle_mm_fault.__do_page_fault.do_page_fault
51.38 +0.4 51.78 perf-profile.calltrace.cycles-pp.cpuidle_enter_state.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64
55.44 +0.7 56.13 perf-profile.calltrace.cycles-pp.cpu_startup_entry.start_secondary.secondary_startup_64
55.44 +0.7 56.13 perf-profile.calltrace.cycles-pp.start_secondary.secondary_startup_64
55.42 +0.7 56.12 perf-profile.calltrace.cycles-pp.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64
56.01 +0.8 56.77 perf-profile.calltrace.cycles-pp.secondary_startup_64
***************************************************************************************************
lkp-hsx04: 144 threads Intel(R) Xeon(R) CPU E7-8890 v3 @ 2.50GHz with 512G memory
=========================================================================================
compiler/cpufreq_governor/kconfig/rootfs/runtime/size/tbox_group/test/testcase/ucode:
gcc-7/performance/x86_64-rhel-7.2/debian-x86_64-2018-04-03.cgz/300s/2T/lkp-hsx04/shm-xread-seq/vm-scalability/0x12
commit:
bb8965bd82 ("mm, memory_hotplug: deobfuscate migration part of offlining")
e0975b2aae ("mm, fault_around: do not take a reference to a locked page")
bb8965bd82fd4ed4 e0975b2aae0e669f995f7d5f11
---------------- --------------------------
%stddev %change %stddev
\ | \
205542 +8.2% 222454 vm-scalability.median
29598225 +8.2% 32033423 vm-scalability.throughput
324.14 +1.4% 328.72 vm-scalability.time.elapsed_time
324.14 +1.4% 328.72 vm-scalability.time.elapsed_time.max
193448 ± 2% +6.2% 205501 vm-scalability.time.involuntary_context_switches
28728244 +17.7% 33815938 vm-scalability.time.maximum_resident_set_size
2.874e+08 +33.3% 3.83e+08 vm-scalability.time.minor_page_faults
10456 +4.0% 10874 vm-scalability.time.percent_of_cpu_this_job_got
28629 +2.2% 29255 vm-scalability.time.system_time
5264 +23.3% 6493 ± 3% vm-scalability.time.user_time
83497287 +22.5% 1.023e+08 vm-scalability.time.voluntary_context_switches
8.891e+09 +8.2% 9.623e+09 vm-scalability.workload
476556 +4.0% 495795 ± 2% interrupts.CAL:Function_call_interrupts
8.13 ± 2% -2.8 5.36 ± 16% mpstat.cpu.iowait%
11.37 +2.1 13.45 ± 2% mpstat.cpu.usr%
502.22 +0.9% 506.74 pmeter.Average_Active_Power
58934 +7.3% 63216 pmeter.performance_per_watt
16260145 +14.7% 18650484 vmstat.memory.cache
34.75 ± 2% -18.0% 28.50 ± 3% vmstat.procs.b
513358 +17.8% 604774 ± 2% vmstat.system.cs
2.848e+09 ± 7% -10.8% 2.541e+09 ± 5% cpuidle.C1.time
37012302 ± 5% +17.4% 43434663 ± 4% cpuidle.C1.usage
1.078e+08 +54.8% 1.67e+08 cpuidle.POLL.time
23357882 +66.1% 38802045 cpuidle.POLL.usage
37010133 ± 5% +17.4% 43432135 ± 4% turbostat.C1
6.08 ± 7% -0.9 5.22 ± 5% turbostat.C1%
13.18 ± 6% -1.9 11.24 ± 6% turbostat.C1E%
23.71 -13.5% 20.52 ± 2% turbostat.CPU%c1
0.82 ± 5% +205.5% 2.50 ± 58% turbostat.Pkg%pc2
2994335 ± 2% +18.4% 3544424 numa-numastat.node0.local_node
3011367 ± 2% +18.3% 3561914 numa-numastat.node0.numa_hit
2977869 ± 2% +16.7% 3476554 ± 2% numa-numastat.node1.local_node
2994091 +16.8% 3495935 ± 2% numa-numastat.node1.numa_hit
3035528 ± 2% +17.9% 3579010 ± 2% numa-numastat.node2.local_node
3059054 ± 2% +17.8% 3603731 numa-numastat.node2.numa_hit
3106731 +20.1% 3731869 ± 2% numa-numastat.node3.local_node
3130558 +19.8% 3751035 ± 2% numa-numastat.node3.numa_hit
3998 ± 4% +28.7% 5147 ± 8% slabinfo.kmalloc-rcl-64.active_objs
3998 ± 4% +28.7% 5147 ± 8% slabinfo.kmalloc-rcl-64.num_objs
77270 +12.3% 86793 slabinfo.radix_tree_node.active_objs
1382 +12.6% 1556 slabinfo.radix_tree_node.active_slabs
77452 +12.6% 87183 slabinfo.radix_tree_node.num_objs
1382 +12.6% 1556 slabinfo.radix_tree_node.num_slabs
1114 ± 10% +21.5% 1354 ± 6% slabinfo.task_group.active_objs
1114 ± 10% +21.5% 1354 ± 6% slabinfo.task_group.num_objs
1241132 ± 7% +31.0% 1625578 ± 5% meminfo.Active
1240916 ± 7% +31.0% 1625376 ± 5% meminfo.Active(anon)
16209000 +14.6% 18580090 meminfo.Cached
14991052 +16.3% 17441268 ± 2% meminfo.Committed_AS
14323027 +13.9% 16315151 meminfo.Inactive
14321660 +13.9% 16313809 meminfo.Inactive(anon)
14074933 +14.2% 16076743 meminfo.Mapped
21745722 +14.0% 24789279 meminfo.Memused
34.25 ± 14% +2647.4% 941.00 ± 85% meminfo.Mlocked
4148946 +15.6% 4794960 meminfo.PageTables
15307509 +15.5% 17683405 ± 2% meminfo.Shmem
121679 +12.0% 136300 meminfo.max_used_kB
46095947 -1.4e+07 32016211 ± 7% syscalls.sys_close.noise.100%
65570785 -1.3e+07 52869552 ± 4% syscalls.sys_close.noise.2%
61504238 -1.3e+07 48872078 ± 4% syscalls.sys_close.noise.25%
65292267 -1.3e+07 52569436 ± 4% syscalls.sys_close.noise.5%
56919490 -1.3e+07 43563196 ± 4% syscalls.sys_close.noise.50%
51551093 -1.4e+07 37234904 ± 5% syscalls.sys_close.noise.75%
2.394e+08 ± 8% -8.8e+07 1.518e+08 ± 58% syscalls.sys_mmap.noise.100%
4.477e+08 ± 3% -1.4e+08 3.057e+08 ± 57% syscalls.sys_mmap.noise.2%
4.203e+08 ± 4% -1.4e+08 2.845e+08 ± 57% syscalls.sys_mmap.noise.25%
4.464e+08 ± 3% -1.4e+08 3.048e+08 ± 57% syscalls.sys_mmap.noise.5%
3.66e+08 ± 5% -1.2e+08 2.461e+08 ± 58% syscalls.sys_mmap.noise.50%
2.998e+08 ± 7% -1e+08 1.979e+08 ± 58% syscalls.sys_mmap.noise.75%
309458 ± 7% +30.9% 405196 ± 5% proc-vmstat.nr_active_anon
4047233 +14.8% 4645438 proc-vmstat.nr_file_pages
3575895 +14.1% 4079756 proc-vmstat.nr_inactive_anon
3514460 +14.4% 4020372 proc-vmstat.nr_mapped
8.75 ± 14% +2580.0% 234.50 ± 85% proc-vmstat.nr_mlock
1037695 +15.6% 1199318 ± 2% proc-vmstat.nr_page_table_pages
3821596 +15.7% 4421008 proc-vmstat.nr_shmem
29395 +5.0% 30877 proc-vmstat.nr_slab_reclaimable
309458 ± 7% +30.9% 405196 ± 5% proc-vmstat.nr_zone_active_anon
3575895 +14.1% 4079756 proc-vmstat.nr_zone_inactive_anon
265512 ± 2% +5.8% 280788 proc-vmstat.numa_hint_faults
138862 +5.4% 146426 proc-vmstat.numa_hint_faults_local
12221647 +18.2% 14441042 proc-vmstat.numa_hit
12140316 +18.3% 14359778 proc-vmstat.numa_local
35834 ± 3% +30.5% 46771 ± 21% proc-vmstat.numa_pages_migrated
7189152 +17.7% 8460359 proc-vmstat.pgactivate
12306840 +18.1% 14537789 proc-vmstat.pgalloc_normal
2.885e+08 +33.1% 3.842e+08 proc-vmstat.pgfault
12117889 +18.8% 14391554 proc-vmstat.pgfree
35834 ± 3% +30.5% 46771 ± 21% proc-vmstat.pgmigrate_success
12405 ± 61% -62.2% 4686 ± 3% sched_debug.cfs_rq:/.load.stddev
0.26 ± 15% +34.5% 0.35 ± 8% sched_debug.cfs_rq:/.nr_spread_over.avg
2.62 ± 34% +55.6% 4.08 ± 14% sched_debug.cfs_rq:/.nr_spread_over.max
0.50 ± 17% +38.7% 0.70 ± 7% sched_debug.cfs_rq:/.nr_spread_over.stddev
12263 ± 62% -62.9% 4555 ± 3% sched_debug.cfs_rq:/.runnable_weight.stddev
1059 ± 20% +32.2% 1400 ± 5% sched_debug.cfs_rq:/.util_est_enqueued.max
268517 -11.6% 237409 sched_debug.cpu.avg_idle.avg
32990 ± 15% -41.4% 19346 ± 7% sched_debug.cpu.avg_idle.min
576975 +22.2% 705245 sched_debug.cpu.nr_switches.avg
602543 +22.4% 737681 sched_debug.cpu.nr_switches.max
501348 ± 3% +24.2% 622609 ± 6% sched_debug.cpu.nr_switches.min
14153 ± 3% +19.1% 16862 ± 7% sched_debug.cpu.nr_switches.stddev
0.23 ± 9% -18.3% 0.19 ± 5% sched_debug.cpu.nr_uninterruptible.avg
565.46 ± 4% -13.9% 487.04 ± 3% sched_debug.cpu.nr_uninterruptible.max
84.95 ± 2% -11.3% 75.34 ± 2% sched_debug.cpu.nr_uninterruptible.stddev
577838 +22.3% 706482 sched_debug.cpu.sched_count.avg
617235 ± 2% +22.7% 757210 ± 2% sched_debug.cpu.sched_count.max
501575 ± 3% +24.0% 621953 ± 6% sched_debug.cpu.sched_count.min
14240 ± 2% +20.5% 17166 ± 10% sched_debug.cpu.sched_count.stddev
287094 +22.3% 351089 sched_debug.cpu.sched_goidle.avg
298687 +22.7% 366344 sched_debug.cpu.sched_goidle.max
249316 ± 4% +24.0% 309274 ± 6% sched_debug.cpu.sched_goidle.min
6977 ± 2% +19.7% 8349 ± 8% sched_debug.cpu.sched_goidle.stddev
296478 +23.3% 365511 sched_debug.cpu.ttwu_count.avg
301862 +23.6% 373218 sched_debug.cpu.ttwu_count.max
257184 ± 3% +25.0% 321458 ± 5% sched_debug.cpu.ttwu_count.min
646.76 ± 4% +12.4% 726.73 sched_debug.cpu.ttwu_local.avg
523.38 ± 5% +11.9% 585.92 ± 6% sched_debug.cpu.ttwu_local.min
318414 ± 16% +54.2% 491071 ± 4% numa-meminfo.node0.Active
318296 ± 16% +54.3% 490999 ± 4% numa-meminfo.node0.Active(anon)
47591 ±116% +154.6% 121156 ± 20% numa-meminfo.node0.AnonHugePages
68246 ± 90% +125.7% 153999 ± 16% numa-meminfo.node0.AnonPages
3489880 +14.6% 3999849 numa-meminfo.node0.Mapped
5645899 ± 3% +12.5% 6352394 ± 3% numa-meminfo.node0.MemUsed
1041071 +14.9% 1196264 ± 2% numa-meminfo.node0.PageTables
3874897 ± 5% +13.6% 4402717 ± 3% numa-meminfo.node0.Shmem
3946112 ± 3% +15.3% 4549030 ± 3% numa-meminfo.node1.FilePages
3492539 ± 4% +14.6% 4003083 ± 4% numa-meminfo.node1.Inactive
3492487 ± 4% +14.6% 4002545 ± 4% numa-meminfo.node1.Inactive(anon)
3432678 +14.9% 3945042 ± 2% numa-meminfo.node1.Mapped
5297542 ± 4% +13.1% 5991649 ± 3% numa-meminfo.node1.MemUsed
1036995 +15.6% 1198973 ± 2% numa-meminfo.node1.PageTables
3729341 ± 3% +16.2% 4332909 ± 4% numa-meminfo.node1.Shmem
4055225 ± 2% +14.8% 4656336 ± 5% numa-meminfo.node2.FilePages
3583032 ± 2% +14.2% 4090753 ± 5% numa-meminfo.node2.Inactive
3582386 ± 2% +14.2% 4090663 ± 5% numa-meminfo.node2.Inactive(anon)
3523884 +14.4% 4031980 ± 2% numa-meminfo.node2.Mapped
5391465 +14.3% 6164631 ± 4% numa-meminfo.node2.MemUsed
1036790 +15.7% 1199620 ± 2% numa-meminfo.node2.PageTables
3829375 ± 2% +15.7% 4432226 ± 5% numa-meminfo.node2.Shmem
299831 ± 5% +27.6% 382447 ± 12% numa-meminfo.node3.Active
299801 ± 5% +27.5% 382380 ± 12% numa-meminfo.node3.Active(anon)
4070305 +16.6% 4745844 ± 2% numa-meminfo.node3.FilePages
3604400 +15.4% 4158947 ± 2% numa-meminfo.node3.Inactive
3603902 +15.4% 4158364 ± 2% numa-meminfo.node3.Inactive(anon)
21785 ± 8% +58.1% 34443 ± 16% numa-meminfo.node3.KReclaimable
3608974 +13.7% 4101705 numa-meminfo.node3.Mapped
5391449 +16.3% 6268719 ± 2% numa-meminfo.node3.MemUsed
1035462 +15.7% 1197562 numa-meminfo.node3.PageTables
21785 ± 8% +58.1% 34443 ± 16% numa-meminfo.node3.SReclaimable
3852964 +17.0% 4507954 ± 2% numa-meminfo.node3.Shmem
64566 ± 8% +23.8% 79928 ± 14% numa-meminfo.node3.Slab
79778 ± 16% +54.2% 122987 ± 4% numa-vmstat.node0.nr_active_anon
17064 ± 90% +125.5% 38488 ± 16% numa-vmstat.node0.nr_anon_pages
873523 +14.4% 999522 numa-vmstat.node0.nr_mapped
2.25 ± 36% +2766.7% 64.50 ± 94% numa-vmstat.node0.nr_mlock
259906 +15.0% 298980 ± 2% numa-vmstat.node0.nr_page_table_pages
969772 ± 5% +13.5% 1100443 ± 3% numa-vmstat.node0.nr_shmem
79777 ± 16% +54.2% 122986 ± 4% numa-vmstat.node0.nr_zone_active_anon
2067000 ± 2% +13.4% 2344491 ± 4% numa-vmstat.node0.numa_hit
2048877 ± 2% +13.5% 2325856 ± 4% numa-vmstat.node0.numa_local
987562 ± 3% +15.1% 1136953 ± 3% numa-vmstat.node1.nr_file_pages
873956 ± 3% +14.4% 1000115 ± 4% numa-vmstat.node1.nr_inactive_anon
859098 +14.7% 985813 ± 2% numa-vmstat.node1.nr_mapped
1.25 ± 34% +5220.0% 66.50 ± 98% numa-vmstat.node1.nr_mlock
258893 +15.7% 299637 ± 2% numa-vmstat.node1.nr_page_table_pages
933369 ± 3% +16.0% 1082922 ± 4% numa-vmstat.node1.nr_shmem
873955 ± 3% +14.4% 1000115 ± 4% numa-vmstat.node1.nr_zone_inactive_anon
1847413 ± 2% +15.6% 2136169 ± 3% numa-vmstat.node1.numa_hit
1756489 ± 2% +16.3% 2042623 ± 3% numa-vmstat.node1.numa_local
1014921 ± 2% +14.7% 1163873 ± 5% numa-vmstat.node2.nr_file_pages
896507 ± 2% +14.0% 1022219 ± 5% numa-vmstat.node2.nr_inactive_anon
881991 +14.2% 1007615 ± 2% numa-vmstat.node2.nr_mapped
258854 +15.8% 299784 ± 2% numa-vmstat.node2.nr_page_table_pages
958458 ± 2% +15.6% 1107845 ± 5% numa-vmstat.node2.nr_shmem
896507 ± 2% +14.0% 1022220 ± 5% numa-vmstat.node2.nr_zone_inactive_anon
1848935 +20.4% 2225472 ± 3% numa-vmstat.node2.numa_hit
1751009 ± 2% +21.4% 2126386 ± 4% numa-vmstat.node2.numa_local
75162 ± 5% +27.5% 95825 ± 11% numa-vmstat.node3.nr_active_anon
1018668 +16.4% 1186179 ± 2% numa-vmstat.node3.nr_file_pages
901853 +15.2% 1039088 ± 2% numa-vmstat.node3.nr_inactive_anon
903255 +13.5% 1025128 numa-vmstat.node3.nr_mapped
258524 +15.7% 299185 numa-vmstat.node3.nr_page_table_pages
964333 +16.8% 1126707 ± 2% numa-vmstat.node3.nr_shmem
5447 ± 8% +58.1% 8611 ± 16% numa-vmstat.node3.nr_slab_reclaimable
75161 ± 5% +27.5% 95825 ± 11% numa-vmstat.node3.nr_zone_active_anon
901853 +15.2% 1039088 ± 2% numa-vmstat.node3.nr_zone_inactive_anon
1838509 ± 2% +26.0% 2317163 ± 4% numa-vmstat.node3.numa_hit
1740069 ± 2% +27.8% 2223959 ± 5% numa-vmstat.node3.numa_local
1.988e+10 +12.9% 2.244e+10 perf-stat.i.branch-instructions
52256410 ± 3% +18.8% 62062282 ± 3% perf-stat.i.branch-misses
72747442 +13.8% 82759538 perf-stat.i.cache-misses
2.69e+08 +11.5% 2.998e+08 ± 2% perf-stat.i.cache-references
530774 +17.7% 624467 ± 2% perf-stat.i.context-switches
3254 ± 4% -7.9% 2998 perf-stat.i.cpu-migrations
1.693e+10 +12.9% 1.911e+10 perf-stat.i.dTLB-loads
3.06e+09 +14.9% 3.517e+09 perf-stat.i.dTLB-stores
27.74 ± 3% +2.6 30.37 ± 2% perf-stat.i.iTLB-load-miss-rate%
2089877 +23.6% 2583091 ± 2% perf-stat.i.iTLB-load-misses
5829102 +9.8% 6401145 ± 2% perf-stat.i.iTLB-loads
6.573e+10 +13.0% 7.425e+10 perf-stat.i.instructions
0.50 +9.0% 0.55 perf-stat.i.ipc
913677 +28.0% 1169682 perf-stat.i.minor-faults
44371606 +15.6% 51310129 perf-stat.i.node-load-misses
1412092 +16.2% 1640849 perf-stat.i.node-loads
83.68 -1.8 81.93 perf-stat.i.node-store-miss-rate%
22602346 +8.9% 24605351 perf-stat.i.node-store-misses
4329404 +18.6% 5133802 ± 2% perf-stat.i.node-stores
913677 +28.0% 1169682 perf-stat.i.page-faults
4.09 -1.3% 4.04 perf-stat.overall.MPKI
27.05 +0.6 27.60 perf-stat.overall.cache-miss-rate%
2.00 -10.0% 1.80 perf-stat.overall.cpi
26.39 +2.4 28.75 perf-stat.overall.iTLB-load-miss-rate%
31454 -8.6% 28757 ± 2% perf-stat.overall.instructions-per-iTLB-miss
0.50 +11.1% 0.56 perf-stat.overall.ipc
83.92 -1.2 82.74 perf-stat.overall.node-store-miss-rate%
2334 +8.5% 2533 perf-stat.overall.path-length
6.276e+12 +17.4% 7.37e+12 perf-stat.total.branch-instructions
1.65e+10 ± 3% +23.6% 2.038e+10 ± 3% perf-stat.total.branch-misses
2.297e+10 +18.3% 2.718e+10 perf-stat.total.cache-misses
8.493e+10 +15.9% 9.846e+10 perf-stat.total.cache-references
1.676e+08 +22.4% 2.051e+08 perf-stat.total.context-switches
4.152e+13 +5.7% 4.389e+13 perf-stat.total.cpu-cycles
1027502 ± 4% -4.2% 984797 perf-stat.total.cpu-migrations
5.347e+12 +17.4% 6.276e+12 perf-stat.total.dTLB-loads
9.662e+11 +19.5% 1.155e+12 perf-stat.total.dTLB-stores
6.599e+08 +28.5% 8.482e+08 perf-stat.total.iTLB-load-misses
1.84e+09 +14.2% 2.102e+09 perf-stat.total.iTLB-loads
2.076e+13 +17.5% 2.438e+13 perf-stat.total.instructions
2.885e+08 +33.1% 3.841e+08 perf-stat.total.minor-faults
93311336 +4.1% 97124343 perf-stat.total.msec
1.401e+10 +20.3% 1.685e+10 perf-stat.total.node-load-misses
4.459e+08 +20.9% 5.389e+08 perf-stat.total.node-loads
7.137e+09 +13.2% 8.08e+09 perf-stat.total.node-store-misses
1.367e+09 +23.3% 1.686e+09 perf-stat.total.node-stores
2.885e+08 +33.1% 3.841e+08 perf-stat.total.page-faults
52.58 ± 57% -52.6 0.00 perf-profile.calltrace.cycles-pp.do_access
47.84 ± 57% -47.8 0.00 perf-profile.calltrace.cycles-pp.page_fault.do_access
47.54 ± 57% -47.5 0.00 perf-profile.calltrace.cycles-pp.do_page_fault.page_fault.do_access
47.47 ± 57% -47.5 0.00 perf-profile.calltrace.cycles-pp.__do_page_fault.do_page_fault.page_fault.do_access
47.11 ± 57% -47.1 0.00 perf-profile.calltrace.cycles-pp.handle_mm_fault.__do_page_fault.do_page_fault.page_fault.do_access
8.05 ± 57% -8.1 0.00 perf-profile.calltrace.cycles-pp.do_rw_once
0.94 ± 5% +0.1 1.05 perf-profile.calltrace.cycles-pp.page_add_file_rmap.alloc_set_pte.filemap_map_pages.__handle_mm_fault.handle_mm_fault
0.60 ± 5% +0.1 0.75 ± 3% perf-profile.calltrace.cycles-pp.menu_select.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64
1.58 ± 4% +0.2 1.75 perf-profile.calltrace.cycles-pp.alloc_set_pte.filemap_map_pages.__handle_mm_fault.handle_mm_fault.__do_page_fault
0.58 ± 7% +0.2 0.75 ± 3% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.finish_wait.__lock_page.find_lock_entry
0.60 ± 7% +0.2 0.79 ± 3% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.finish_wait.__lock_page.find_lock_entry.shmem_getpage_gfp
0.77 ± 7% +0.2 0.97 ± 3% perf-profile.calltrace.cycles-pp.__sched_text_start.schedule.io_schedule.__lock_page.find_lock_entry
0.62 ± 7% +0.2 0.82 ± 2% perf-profile.calltrace.cycles-pp.finish_wait.__lock_page.find_lock_entry.shmem_getpage_gfp.shmem_fault
0.79 ± 7% +0.2 0.99 ± 2% perf-profile.calltrace.cycles-pp.io_schedule.__lock_page.find_lock_entry.shmem_getpage_gfp.shmem_fault
0.78 ± 7% +0.2 0.98 ± 3% perf-profile.calltrace.cycles-pp.schedule.io_schedule.__lock_page.find_lock_entry.shmem_getpage_gfp
0.61 ± 4% +0.2 0.82 ± 3% perf-profile.calltrace.cycles-pp.__save_stack_trace.save_stack_trace_tsk.__account_scheduler_latency.enqueue_entity.enqueue_task_fair
0.89 ± 6% +0.2 1.10 perf-profile.calltrace.cycles-pp.try_to_wake_up.autoremove_wake_function.__wake_up_common.wake_up_page_bit.filemap_map_pages
1.67 ± 6% +0.2 1.89 perf-profile.calltrace.cycles-pp.__wake_up_common.wake_up_page_bit.filemap_map_pages.__handle_mm_fault.handle_mm_fault
0.65 ± 4% +0.2 0.87 ± 2% perf-profile.calltrace.cycles-pp.save_stack_trace_tsk.__account_scheduler_latency.enqueue_entity.enqueue_task_fair.ttwu_do_activate
0.95 ± 7% +0.2 1.18 perf-profile.calltrace.cycles-pp.autoremove_wake_function.__wake_up_common.wake_up_page_bit.filemap_map_pages.__handle_mm_fault
0.58 ± 6% +0.2 0.82 ± 2% perf-profile.calltrace.cycles-pp.__wake_up_common.wake_up_page_bit.__handle_mm_fault.handle_mm_fault.__do_page_fault
0.89 ± 5% +0.3 1.15 ± 2% perf-profile.calltrace.cycles-pp.find_get_entry.find_lock_entry.shmem_getpage_gfp.shmem_fault.__do_fault
0.94 ± 3% +0.3 1.25 perf-profile.calltrace.cycles-pp.__account_scheduler_latency.enqueue_entity.enqueue_task_fair.ttwu_do_activate.sched_ttwu_pending
1.23 ± 3% +0.4 1.61 perf-profile.calltrace.cycles-pp.enqueue_entity.enqueue_task_fair.ttwu_do_activate.sched_ttwu_pending.do_idle
1.30 ± 3% +0.4 1.70 perf-profile.calltrace.cycles-pp.enqueue_task_fair.ttwu_do_activate.sched_ttwu_pending.do_idle.cpu_startup_entry
1.33 ± 3% +0.4 1.73 perf-profile.calltrace.cycles-pp.ttwu_do_activate.sched_ttwu_pending.do_idle.cpu_startup_entry.start_secondary
1.56 ± 3% +0.5 2.01 perf-profile.calltrace.cycles-pp.sched_ttwu_pending.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64
0.12 ±173% +0.5 0.66 ± 3% perf-profile.calltrace.cycles-pp.finish_fault.__handle_mm_fault.handle_mm_fault.__do_page_fault.do_page_fault
0.00 +0.5 0.54 perf-profile.calltrace.cycles-pp.try_to_wake_up.autoremove_wake_function.__wake_up_common.wake_up_page_bit.__handle_mm_fault
0.00 +0.6 0.58 perf-profile.calltrace.cycles-pp.autoremove_wake_function.__wake_up_common.wake_up_page_bit.__handle_mm_fault.handle_mm_fault
0.00 +0.6 0.64 ± 3% perf-profile.calltrace.cycles-pp.alloc_set_pte.finish_fault.__handle_mm_fault.handle_mm_fault.__do_page_fault
5.53 ± 6% +1.0 6.48 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irq.__lock_page.find_lock_entry.shmem_getpage_gfp
5.70 ± 6% +1.0 6.71 perf-profile.calltrace.cycles-pp._raw_spin_lock_irq.__lock_page.find_lock_entry.shmem_getpage_gfp.shmem_fault
8.15 ± 6% +1.6 9.80 perf-profile.calltrace.cycles-pp.__lock_page.find_lock_entry.shmem_getpage_gfp.shmem_fault.__do_fault
5.27 ± 8% +1.8 7.07 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.wake_up_page_bit.__handle_mm_fault.handle_mm_fault
5.36 ± 8% +1.8 7.19 perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.wake_up_page_bit.__handle_mm_fault.handle_mm_fault.__do_page_fault
9.40 ± 6% +2.1 11.49 perf-profile.calltrace.cycles-pp.find_lock_entry.shmem_getpage_gfp.shmem_fault.__do_fault.__handle_mm_fault
9.66 ± 6% +2.2 11.82 perf-profile.calltrace.cycles-pp.shmem_getpage_gfp.shmem_fault.__do_fault.__handle_mm_fault.handle_mm_fault
6.08 ± 7% +2.2 8.24 perf-profile.calltrace.cycles-pp.wake_up_page_bit.__handle_mm_fault.handle_mm_fault.__do_page_fault.do_page_fault
9.71 ± 6% +2.2 11.91 perf-profile.calltrace.cycles-pp.shmem_fault.__do_fault.__handle_mm_fault.handle_mm_fault.__do_page_fault
9.76 ± 6% +2.2 11.98 perf-profile.calltrace.cycles-pp.__do_fault.__handle_mm_fault.handle_mm_fault.__do_page_fault.do_page_fault
17.66 ± 6% +4.2 21.87 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.wake_up_page_bit.filemap_map_pages.__handle_mm_fault
18.03 ± 6% +4.3 22.30 perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.wake_up_page_bit.filemap_map_pages.__handle_mm_fault.handle_mm_fault
20.32 ± 6% +4.7 25.01 perf-profile.calltrace.cycles-pp.wake_up_page_bit.filemap_map_pages.__handle_mm_fault.handle_mm_fault.__do_page_fault
65.06 ± 6% +7.2 72.27 perf-profile.calltrace.cycles-pp.__handle_mm_fault.handle_mm_fault.__do_page_fault.do_page_fault.page_fault
18.07 ±173% +54.4 72.44 perf-profile.calltrace.cycles-pp.handle_mm_fault.__do_page_fault.do_page_fault.page_fault
18.20 ±173% +54.9 73.08 perf-profile.calltrace.cycles-pp.__do_page_fault.do_page_fault.page_fault
18.23 ±173% +55.0 73.21 perf-profile.calltrace.cycles-pp.do_page_fault.page_fault
18.23 ±173% +55.0 73.23 perf-profile.calltrace.cycles-pp.page_fault
***************************************************************************************************
lkp-skl-fpga01: 104 threads Skylake with 192G memory
=========================================================================================
compiler/cpufreq_governor/kconfig/rootfs/runtime/size/tbox_group/test/testcase:
gcc-7/performance/x86_64-rhel-7.2/debian-x86_64-2018-04-03.cgz/300s/2T/lkp-skl-fpga01/shm-pread-seq/vm-scalability
commit:
bb8965bd82 ("mm, memory_hotplug: deobfuscate migration part of offlining")
e0975b2aae ("mm, fault_around: do not take a reference to a locked page")
bb8965bd82fd4ed4 e0975b2aae0e669f995f7d5f11
---------------- --------------------------
%stddev %change %stddev
\ | \
20.32 ± 2% +35.4% 27.52 ± 2% vm-scalability.free_time
206116 +27.4% 262500 vm-scalability.median
21436049 +27.4% 27299983 vm-scalability.throughput
325.40 +2.6% 334.02 vm-scalability.time.elapsed_time
325.40 +2.6% 334.02 vm-scalability.time.elapsed_time.max
91050 +7.3% 97658 vm-scalability.time.involuntary_context_switches
49659993 +34.2% 66628111 vm-scalability.time.maximum_resident_set_size
4.583e+08 +51.7% 6.953e+08 vm-scalability.time.minor_page_faults
8228 +3.0% 8473 vm-scalability.time.percent_of_cpu_this_job_got
24542 +2.7% 25195 vm-scalability.time.system_time
2232 +39.3% 3109 vm-scalability.time.user_time
1.427e+08 +40.7% 2.007e+08 vm-scalability.time.voluntary_context_switches
6.443e+09 +27.4% 8.208e+09 vm-scalability.workload
1.90 ± 10% -0.6 1.30 ± 4% mpstat.cpu.iowait%
6.41 ± 5% +2.4 8.82 mpstat.cpu.usr%
8954499 +39.3% 12473967 numa-numastat.node0.local_node
8959732 +39.2% 12474154 numa-numastat.node0.numa_hit
8983237 ± 3% +35.9% 12206926 numa-numastat.node1.local_node
8998587 ± 3% +35.9% 12227322 numa-numastat.node1.numa_hit
26660283 ± 5% +37.5% 36651992 vmstat.memory.cache
18.25 ± 5% -20.5% 14.50 ± 3% vmstat.procs.b
848025 ± 4% +39.4% 1182314 vmstat.system.cs
217673 ± 2% +2.7% 223468 vmstat.system.in
2.138e+09 ± 2% -31.9% 1.457e+09 cpuidle.C1.time
3.655e+09 ± 5% -18.1% 2.995e+09 ± 6% cpuidle.C1E.time
65726694 ± 2% +12.1% 73704022 cpuidle.C1E.usage
7.785e+08 +64.4% 1.28e+09 cpuidle.POLL.time
85175054 +84.5% 1.571e+08 cpuidle.POLL.usage
2238 ± 4% +6.4% 2382 turbostat.Avg_MHz
6.12 ± 3% -2.0 4.14 turbostat.C1%
65718547 ± 2% +12.1% 73693777 turbostat.C1E
10.48 ± 8% -2.0 8.51 ± 7% turbostat.C1E%
18.57 ± 10% -24.4% 14.04 ± 4% turbostat.CPU%c1
315.50 ± 2% +6.0% 334.33 turbostat.PkgWatt
141.19 ± 2% +9.0% 153.87 turbostat.RAMWatt
622.25 ± 5% +15.5% 718.50 ± 6% slabinfo.file_lock_cache.active_objs
622.25 ± 5% +15.5% 718.50 ± 6% slabinfo.file_lock_cache.num_objs
9963 ± 5% +11.5% 11113 slabinfo.kmalloc-512.active_objs
10048 ± 5% +10.8% 11130 slabinfo.kmalloc-512.num_objs
116705 ± 4% +33.8% 156158 slabinfo.radix_tree_node.active_objs
2087 ± 4% +33.7% 2791 slabinfo.radix_tree_node.active_slabs
116945 ± 4% +33.7% 156339 slabinfo.radix_tree_node.num_objs
2087 ± 4% +33.7% 2791 slabinfo.radix_tree_node.num_slabs
2125383 ± 6% +66.1% 3529691 ± 3% meminfo.Active
2125321 ± 6% +66.1% 3529627 ± 3% meminfo.Active(anon)
26609292 ± 5% +37.4% 36573172 meminfo.Cached
26054799 ± 5% +39.0% 36222963 ± 2% meminfo.Committed_AS
23838297 ± 5% +35.9% 32401152 meminfo.Inactive
23837149 ± 5% +35.9% 32399988 meminfo.Inactive(anon)
130912 ± 2% +17.3% 153597 meminfo.KReclaimable
23751791 ± 5% +36.0% 32313264 meminfo.Mapped
32763824 ± 5% +36.5% 44720970 meminfo.Memused
5120515 ± 5% +38.3% 7082028 meminfo.PageTables
130912 ± 2% +17.3% 153597 meminfo.SReclaimable
25708248 ± 5% +38.8% 35676400 meminfo.Shmem
184916 ± 5% +31.9% 243839 meminfo.max_used_kB
1089554 ± 6% +68.9% 1840059 ± 4% numa-meminfo.node0.Active
1089508 ± 6% +68.9% 1840039 ± 4% numa-meminfo.node0.Active(anon)
13263954 ± 4% +40.7% 18661298 ± 2% numa-meminfo.node0.FilePages
11855463 ± 4% +39.2% 16506417 ± 2% numa-meminfo.node0.Inactive
11854609 ± 4% +39.2% 16506036 ± 2% numa-meminfo.node0.Inactive(anon)
70171 ± 11% +27.2% 89286 ± 3% numa-meminfo.node0.KReclaimable
11790152 ± 4% +39.2% 16412928 ± 2% numa-meminfo.node0.Mapped
16357141 ± 4% +39.3% 22781798 ± 2% numa-meminfo.node0.MemUsed
2556757 ± 5% +38.7% 3545040 ± 2% numa-meminfo.node0.PageTables
70171 ± 11% +27.2% 89286 ± 3% numa-meminfo.node0.SReclaimable
12804300 ± 4% +42.1% 18200193 ± 3% numa-meminfo.node0.Shmem
151366 ± 9% +17.1% 177261 ± 3% numa-meminfo.node0.Slab
1047464 ± 9% +63.4% 1711298 numa-meminfo.node1.Active
1047448 ± 9% +63.4% 1711254 numa-meminfo.node1.Active(anon)
13339794 ± 7% +34.4% 17930482 numa-meminfo.node1.FilePages
11964624 ± 7% +32.8% 15890641 numa-meminfo.node1.Inactive
11964331 ± 7% +32.8% 15889858 numa-meminfo.node1.Inactive(anon)
11943581 ± 6% +33.1% 15896642 numa-meminfo.node1.Mapped
16393799 ± 6% +34.0% 21962322 numa-meminfo.node1.MemUsed
2554698 ± 5% +38.6% 3541528 numa-meminfo.node1.PageTables
12897375 ± 7% +35.6% 17493769 numa-meminfo.node1.Shmem
271474 ± 6% +68.8% 458341 ± 4% numa-vmstat.node0.nr_active_anon
3316562 ± 4% +40.6% 4662543 ± 2% numa-vmstat.node0.nr_file_pages
2965111 ± 4% +39.1% 4125374 ± 2% numa-vmstat.node0.nr_inactive_anon
2949108 ± 5% +39.1% 4102153 ± 2% numa-vmstat.node0.nr_mapped
640259 ± 5% +38.3% 885781 ± 2% numa-vmstat.node0.nr_page_table_pages
3201649 ± 4% +42.0% 4547267 ± 3% numa-vmstat.node0.nr_shmem
17542 ± 11% +27.2% 22316 ± 3% numa-vmstat.node0.nr_slab_reclaimable
271474 ± 6% +68.8% 458340 ± 4% numa-vmstat.node0.nr_zone_active_anon
2965112 ± 4% +39.1% 4125374 ± 2% numa-vmstat.node0.nr_zone_inactive_anon
5397618 ± 4% +38.6% 7479811 numa-vmstat.node0.numa_hit
5392259 ± 4% +38.7% 7479544 numa-vmstat.node0.numa_local
260949 ± 9% +63.3% 426216 numa-vmstat.node1.nr_active_anon
3335575 ± 7% +34.3% 4479908 numa-vmstat.node1.nr_file_pages
2992597 ± 7% +32.7% 3971339 numa-vmstat.node1.nr_inactive_anon
2987487 ± 6% +33.0% 3973080 numa-vmstat.node1.nr_mapped
639770 ± 5% +38.3% 884874 numa-vmstat.node1.nr_page_table_pages
3224971 ± 7% +35.5% 4370730 numa-vmstat.node1.nr_shmem
260949 ± 9% +63.3% 426216 numa-vmstat.node1.nr_zone_active_anon
2992598 ± 7% +32.7% 3971339 numa-vmstat.node1.nr_zone_inactive_anon
5406505 ± 4% +31.6% 7115266 numa-vmstat.node1.numa_hit
5242753 ± 4% +32.5% 6947359 numa-vmstat.node1.numa_local
533940 ± 6% +66.6% 889701 ± 3% proc-vmstat.nr_active_anon
4072439 -7.4% 3772149 proc-vmstat.nr_dirty_background_threshold
8154837 -7.4% 7553525 proc-vmstat.nr_dirty_threshold
6647427 ± 5% +37.7% 9151567 proc-vmstat.nr_file_pages
41002279 -7.3% 37994979 proc-vmstat.nr_free_pages
5951525 ± 5% +36.1% 8100698 proc-vmstat.nr_inactive_anon
5930348 ± 5% +36.2% 8079160 proc-vmstat.nr_mapped
1276136 ± 5% +38.8% 1771811 proc-vmstat.nr_page_table_pages
6421909 ± 5% +39.0% 8927112 proc-vmstat.nr_shmem
32718 ± 2% +17.4% 38412 proc-vmstat.nr_slab_reclaimable
533940 ± 6% +66.6% 889701 ± 3% proc-vmstat.nr_zone_active_anon
5951525 ± 5% +36.1% 8100698 proc-vmstat.nr_zone_inactive_anon
108055 ± 2% +22.0% 131864 ± 15% proc-vmstat.numa_hint_faults
50736 ± 5% +36.6% 69282 ± 26% proc-vmstat.numa_hint_faults_local
17983536 +37.5% 24728866 proc-vmstat.numa_hit
17962942 +37.6% 24708277 proc-vmstat.numa_local
137952 ± 20% +29.9% 179225 ± 7% proc-vmstat.numa_pte_updates
12438184 +34.1% 16680755 proc-vmstat.pgactivate
18076435 +37.4% 24832133 proc-vmstat.pgalloc_normal
4.592e+08 +51.6% 6.962e+08 proc-vmstat.pgfault
17945474 +36.7% 24523437 proc-vmstat.pgfree
133924 ± 33% +73.7% 232608 ± 21% sched_debug.cfs_rq:/.MIN_vruntime.avg
6885541 ± 34% +55.7% 10719317 sched_debug.cfs_rq:/.MIN_vruntime.max
926236 ± 30% +63.4% 1513255 ± 11% sched_debug.cfs_rq:/.MIN_vruntime.stddev
133924 ± 33% +73.7% 232608 ± 21% sched_debug.cfs_rq:/.max_vruntime.avg
6885541 ± 34% +55.7% 10719317 sched_debug.cfs_rq:/.max_vruntime.max
926236 ± 30% +63.4% 1513255 ± 11% sched_debug.cfs_rq:/.max_vruntime.stddev
0.11 ± 3% +23.2% 0.14 ± 11% sched_debug.cfs_rq:/.nr_spread_over.avg
0.32 ± 2% +14.5% 0.37 ± 4% sched_debug.cfs_rq:/.nr_spread_over.stddev
0.00 ± 5% -10.4% 0.00 ± 7% sched_debug.cpu.next_balance.stddev
1362381 +40.5% 1914499 sched_debug.cpu.nr_switches.avg
1412333 +39.9% 1975433 sched_debug.cpu.nr_switches.max
1310627 +40.5% 1841733 sched_debug.cpu.nr_switches.min
20464 ± 7% +26.0% 25793 ± 9% sched_debug.cpu.nr_switches.stddev
1363920 +40.5% 1916717 sched_debug.cpu.sched_count.avg
1414295 +39.7% 1976181 sched_debug.cpu.sched_count.max
1312389 +40.5% 1844419 sched_debug.cpu.sched_count.min
20433 ± 8% +24.9% 25526 ± 9% sched_debug.cpu.sched_count.stddev
680009 +40.6% 955918 sched_debug.cpu.sched_goidle.avg
704278 +40.0% 985646 sched_debug.cpu.sched_goidle.max
654441 +40.6% 919851 sched_debug.cpu.sched_goidle.min
10101 ± 7% +26.4% 12769 ± 9% sched_debug.cpu.sched_goidle.stddev
698872 +41.6% 989297 sched_debug.cpu.ttwu_count.avg
709489 +42.7% 1012313 sched_debug.cpu.ttwu_count.max
680871 +40.4% 956161 sched_debug.cpu.ttwu_count.min
4704 ± 15% +107.3% 9751 ± 9% sched_debug.cpu.ttwu_count.stddev
541.57 +22.7% 664.28 ± 3% sched_debug.cpu.ttwu_local.avg
468.38 ± 4% +21.4% 568.38 ± 3% sched_debug.cpu.ttwu_local.min
62.65 -5.8 56.84 perf-profile.calltrace.cycles-pp.filemap_map_pages.__handle_mm_fault.handle_mm_fault.__do_page_fault.do_page_fault
14.84 -2.3 12.57 perf-profile.calltrace.cycles-pp.intel_idle.cpuidle_enter_state.do_idle.cpu_startup_entry.start_secondary
78.30 -0.9 77.45 perf-profile.calltrace.cycles-pp.__handle_mm_fault.handle_mm_fault.__do_page_fault.do_page_fault.page_fault
78.40 -0.8 77.59 perf-profile.calltrace.cycles-pp.handle_mm_fault.__do_page_fault.do_page_fault.page_fault
78.79 -0.7 78.07 perf-profile.calltrace.cycles-pp.__do_page_fault.do_page_fault.page_fault
78.84 -0.7 78.15 perf-profile.calltrace.cycles-pp.do_page_fault.page_fault
78.86 -0.7 78.18 perf-profile.calltrace.cycles-pp.page_fault
17.84 -0.4 17.43 perf-profile.calltrace.cycles-pp.cpuidle_enter_state.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64
2.42 -0.2 2.23 perf-profile.calltrace.cycles-pp.unlock_page.filemap_map_pages.__handle_mm_fault.handle_mm_fault.__do_page_fault
0.61 ± 2% +0.2 0.81 ± 2% perf-profile.calltrace.cycles-pp.schedule.io_schedule.__lock_page.find_lock_entry.shmem_getpage_gfp
0.60 ± 2% +0.2 0.81 ± 2% perf-profile.calltrace.cycles-pp.__sched_text_start.schedule.io_schedule.__lock_page.find_lock_entry
0.62 ± 2% +0.2 0.83 perf-profile.calltrace.cycles-pp.io_schedule.__lock_page.find_lock_entry.shmem_getpage_gfp.shmem_fault
1.06 +0.3 1.32 perf-profile.calltrace.cycles-pp.find_get_entry.find_lock_entry.shmem_getpage_gfp.shmem_fault.__do_fault
1.08 +0.3 1.34 perf-profile.calltrace.cycles-pp.try_to_wake_up.autoremove_wake_function.__wake_up_common.wake_up_page_bit.filemap_map_pages
1.89 +0.3 2.15 perf-profile.calltrace.cycles-pp.__wake_up_common.wake_up_page_bit.filemap_map_pages.__handle_mm_fault.handle_mm_fault
0.54 +0.3 0.82 perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.finish_wait.__lock_page.find_lock_entry.shmem_getpage_gfp
1.15 +0.3 1.43 perf-profile.calltrace.cycles-pp.autoremove_wake_function.__wake_up_common.wake_up_page_bit.filemap_map_pages.__handle_mm_fault
0.64 +0.3 0.93 ± 4% perf-profile.calltrace.cycles-pp.enqueue_entity.enqueue_task_fair.ttwu_do_activate.sched_ttwu_pending.do_idle
0.56 +0.3 0.86 perf-profile.calltrace.cycles-pp.finish_wait.__lock_page.find_lock_entry.shmem_getpage_gfp.shmem_fault
0.71 +0.3 1.02 ± 3% perf-profile.calltrace.cycles-pp.ttwu_do_activate.sched_ttwu_pending.do_idle.cpu_startup_entry.start_secondary
0.70 +0.3 1.01 ± 3% perf-profile.calltrace.cycles-pp.enqueue_task_fair.ttwu_do_activate.sched_ttwu_pending.do_idle.cpu_startup_entry
0.62 +0.4 0.99 perf-profile.calltrace.cycles-pp.__wake_up_common.wake_up_page_bit.__handle_mm_fault.handle_mm_fault.__do_page_fault
0.87 +0.4 1.24 ± 2% perf-profile.calltrace.cycles-pp.sched_ttwu_pending.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64
0.39 ± 57% +0.4 0.77 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.finish_wait.__lock_page.find_lock_entry
0.00 +0.6 0.61 perf-profile.calltrace.cycles-pp.alloc_set_pte.finish_fault.__handle_mm_fault.handle_mm_fault.__do_page_fault
0.00 +0.6 0.62 ± 3% perf-profile.calltrace.cycles-pp.enqueue_entity.enqueue_task_fair.ttwu_do_activate.try_to_wake_up.autoremove_wake_function
0.00 +0.6 0.63 ± 2% perf-profile.calltrace.cycles-pp.finish_fault.__handle_mm_fault.handle_mm_fault.__do_page_fault.do_page_fault
0.00 +0.6 0.65 ± 3% perf-profile.calltrace.cycles-pp.enqueue_task_fair.ttwu_do_activate.try_to_wake_up.autoremove_wake_function.__wake_up_common
0.00 +0.7 0.66 ± 3% perf-profile.calltrace.cycles-pp.ttwu_do_activate.try_to_wake_up.autoremove_wake_function.__wake_up_common.wake_up_page_bit
0.00 +0.7 0.70 perf-profile.calltrace.cycles-pp.try_to_wake_up.autoremove_wake_function.__wake_up_common.wake_up_page_bit.__handle_mm_fault
0.00 +0.7 0.72 ± 4% perf-profile.calltrace.cycles-pp.__account_scheduler_latency.enqueue_entity.enqueue_task_fair.ttwu_do_activate.sched_ttwu_pending
0.00 +0.7 0.74 perf-profile.calltrace.cycles-pp.autoremove_wake_function.__wake_up_common.wake_up_page_bit.__handle_mm_fault.handle_mm_fault
4.42 +1.1 5.52 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irq.__lock_page.find_lock_entry.shmem_getpage_gfp
4.63 +1.2 5.81 perf-profile.calltrace.cycles-pp._raw_spin_lock_irq.__lock_page.find_lock_entry.shmem_getpage_gfp.shmem_fault
5.06 +1.5 6.59 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.wake_up_page_bit.__handle_mm_fault.handle_mm_fault
5.17 +1.6 6.75 perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.wake_up_page_bit.__handle_mm_fault.handle_mm_fault.__do_page_fault
6.94 +1.8 8.78 perf-profile.calltrace.cycles-pp.__lock_page.find_lock_entry.shmem_getpage_gfp.shmem_fault.__do_fault
2.84 +1.9 4.71 perf-profile.calltrace.cycles-pp.poll_idle.cpuidle_enter_state.do_idle.cpu_startup_entry.start_secondary
5.89 +2.0 7.92 perf-profile.calltrace.cycles-pp.wake_up_page_bit.__handle_mm_fault.handle_mm_fault.__do_page_fault.do_page_fault
8.60 +2.4 10.96 perf-profile.calltrace.cycles-pp.find_lock_entry.shmem_getpage_gfp.shmem_fault.__do_fault.__handle_mm_fault
8.87 +2.5 11.35 perf-profile.calltrace.cycles-pp.shmem_getpage_gfp.shmem_fault.__do_fault.__handle_mm_fault.handle_mm_fault
8.91 +2.5 11.46 perf-profile.calltrace.cycles-pp.shmem_fault.__do_fault.__handle_mm_fault.handle_mm_fault.__do_page_fault
8.95 +2.6 11.51 perf-profile.calltrace.cycles-pp.__do_fault.__handle_mm_fault.handle_mm_fault.__do_page_fault.do_page_fault
18.27 +4.1 22.38 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.wake_up_page_bit.filemap_map_pages.__handle_mm_fault
18.72 +4.1 22.86 perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.wake_up_page_bit.filemap_map_pages.__handle_mm_fault.handle_mm_fault
21.07 +4.5 25.60 perf-profile.calltrace.cycles-pp.wake_up_page_bit.filemap_map_pages.__handle_mm_fault.handle_mm_fault.__do_page_fault
2.153e+10 ± 5% +33.6% 2.877e+10 perf-stat.i.branch-instructions
72045123 ± 3% +40.9% 1.015e+08 perf-stat.i.branch-misses
1.684e+08 ± 4% +33.8% 2.253e+08 ± 2% perf-stat.i.cache-misses
6.465e+08 ± 4% +30.2% 8.416e+08 ± 2% perf-stat.i.cache-references
860783 ± 4% +39.5% 1200816 perf-stat.i.context-switches
3.82 ± 6% -15.0% 3.25 perf-stat.i.cpi
2.349e+11 ± 4% +6.4% 2.501e+11 perf-stat.i.cpu-cycles
7423479 ± 3% +44.5% 10724252 ± 3% perf-stat.i.dTLB-load-misses
1.808e+10 ± 5% +33.7% 2.416e+10 perf-stat.i.dTLB-loads
1493786 ± 4% +49.2% 2228789 perf-stat.i.dTLB-store-misses
3.863e+09 ± 4% +35.1% 5.218e+09 perf-stat.i.dTLB-stores
30.54 ± 6% +4.0 34.59 perf-stat.i.iTLB-load-miss-rate%
2835309 ± 3% +49.6% 4240399 perf-stat.i.iTLB-load-misses
7275768 ± 3% +15.1% 8372947 perf-stat.i.iTLB-loads
6.997e+10 ± 5% +33.9% 9.368e+10 perf-stat.i.instructions
0.30 ± 3% +26.9% 0.38 perf-stat.i.ipc
1384177 ± 4% +50.4% 2081545 perf-stat.i.minor-faults
30594295 ± 4% +37.4% 42030708 perf-stat.i.node-load-misses
1566415 ± 5% +51.6% 2374125 ± 6% perf-stat.i.node-loads
18082396 ± 4% +16.7% 21102487 perf-stat.i.node-store-misses
50456 ± 6% +28.0% 64588 perf-stat.i.node-stores
1384177 ± 4% +50.4% 2081545 perf-stat.i.page-faults
0.33 ± 2% +0.0 0.35 perf-stat.overall.branch-miss-rate%
26.04 +0.7 26.77 perf-stat.overall.cache-miss-rate%
3.36 -20.5% 2.67 perf-stat.overall.cpi
0.04 ± 2% +0.0 0.04 ± 3% perf-stat.overall.dTLB-load-miss-rate%
0.04 +0.0 0.04 perf-stat.overall.dTLB-store-miss-rate%
28.05 +5.6 33.62 perf-stat.overall.iTLB-load-miss-rate%
24662 -10.4% 22092 perf-stat.overall.instructions-per-iTLB-miss
0.30 +25.8% 0.37 perf-stat.overall.ipc
3601 +6.0% 3817 perf-stat.overall.path-length
7.142e+12 +34.7% 9.622e+12 perf-stat.total.branch-instructions
2.391e+10 +42.0% 3.396e+10 perf-stat.total.branch-misses
5.585e+10 +34.9% 7.536e+10 ± 2% perf-stat.total.cache-misses
2.145e+11 +31.2% 2.815e+11 ± 2% perf-stat.total.cache-references
2.857e+08 +40.6% 4.016e+08 perf-stat.total.context-switches
7.794e+13 +7.3% 8.364e+13 perf-stat.total.cpu-cycles
2.465e+09 ± 2% +45.5% 3.587e+09 ± 3% perf-stat.total.dTLB-load-misses
5.996e+12 +34.8% 8.082e+12 perf-stat.total.dTLB-loads
4.957e+08 +50.4% 7.455e+08 perf-stat.total.dTLB-store-misses
1.281e+12 +36.2% 1.745e+12 perf-stat.total.dTLB-stores
9.411e+08 +50.7% 1.418e+09 perf-stat.total.iTLB-load-misses
2.415e+09 +16.0% 2.801e+09 perf-stat.total.iTLB-loads
2.321e+13 +35.0% 3.133e+13 perf-stat.total.instructions
4.592e+08 +51.6% 6.962e+08 perf-stat.total.minor-faults
1.015e+10 +38.5% 1.406e+10 perf-stat.total.node-load-misses
5.196e+08 +52.8% 7.941e+08 ± 6% perf-stat.total.node-loads
5.999e+09 +17.7% 7.058e+09 perf-stat.total.node-store-misses
16729803 ± 3% +29.1% 21605974 perf-stat.total.node-stores
4.592e+08 +51.6% 6.962e+08 perf-stat.total.page-faults
Disclaimer:
Results have been estimated based on internal Intel analysis and are provided
for informational purposes only. Any difference in system hardware or software
design or configuration may affect actual performance.
Thanks,
Rong Chen
View attachment "config-4.20.0-06572-ge0975b2" of type "text/plain" (168788 bytes)
View attachment "job-script" of type "text/plain" (7592 bytes)
View attachment "job.yaml" of type "text/plain" (5036 bytes)
View attachment "reproduce" of type "text/plain" (904 bytes)
Powered by blists - more mailing lists