| 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: <20200927004119.GR28663@shao2-debian>
Date: Sun, 27 Sep 2020 08:41:19 +0800
From: kernel test robot <rong.a.chen@...el.com>
To: Peter Xu <peterx@...hat.com>
Cc: linux-mm@...ck.org, linux-kernel@...r.kernel.org,
Jason Gunthorpe <jgg@...pe.ca>,
Andrew Morton <akpm@...ux-foundation.org>,
Jan Kara <jack@...e.cz>, Michal Hocko <mhocko@...e.com>,
Kirill Tkhai <ktkhai@...tuozzo.com>,
Kirill Shutemov <kirill@...temov.name>,
Hugh Dickins <hughd@...gle.com>, Peter Xu <peterx@...hat.com>,
Christoph Hellwig <hch@....de>,
Andrea Arcangeli <aarcange@...hat.com>,
John Hubbard <jhubbard@...dia.com>,
Oleg Nesterov <oleg@...hat.com>,
Leon Romanovsky <leonro@...dia.com>,
Linus Torvalds <torvalds@...ux-foundation.org>,
Jann Horn <jannh@...gle.com>, 0day robot <lkp@...el.com>,
lkp@...ts.01.org, ying.huang@...el.com, feng.tang@...el.com,
zhengjun.xing@...el.com
Subject: [mm] 698ac7610f: will-it-scale.per_thread_ops 8.2% improvement
Greeting,
FYI, we noticed a 8.2% improvement of will-it-scale.per_thread_ops due to commit:
commit: 698ac7610f7928ddfa44a0736e89d776579d8b82 ("[PATCH 1/5] mm: Introduce mm_struct.has_pinned")
url: https://github.com/0day-ci/linux/commits/Peter-Xu/mm-Break-COW-for-pinned-pages-during-fork/20200922-052211
base: https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git bcf876870b95592b52519ed4aafcf9d95999bc9c
in testcase: will-it-scale
on test machine: 96 threads Intel(R) Xeon(R) CPU @ 2.30GHz with 128G memory
with following parameters:
nr_task: 100%
mode: thread
test: mmap2
cpufreq_governor: performance
ucode: 0x4002f01
test-description: Will It Scale takes a testcase and runs it from 1 through to n parallel copies to see if the testcase will scale. It builds both a process and threads based test in order to see any differences between the two.
test-url: https://github.com/antonblanchard/will-it-scale
Details are as below:
-------------------------------------------------------------------------------------------------->
To reproduce:
git clone https://github.com/intel/lkp-tests.git
cd lkp-tests
bin/lkp install job.yaml # job file is attached in this email
bin/lkp run job.yaml
=========================================================================================
compiler/cpufreq_governor/kconfig/mode/nr_task/rootfs/tbox_group/test/testcase/ucode:
gcc-9/performance/x86_64-rhel-8.3/thread/100%/debian-10.4-x86_64-20200603.cgz/lkp-csl-2sp4/mmap2/will-it-scale/0x4002f01
commit:
v5.8
698ac7610f ("mm: Introduce mm_struct.has_pinned")
v5.8 698ac7610f7928ddfa44a0736e8
---------------- ---------------------------
%stddev %change %stddev
\ | \
2003 +8.2% 2168 will-it-scale.per_thread_ops
192350 +8.3% 208245 will-it-scale.workload
2643 ± 33% -36.3% 1683 meminfo.Active(file)
3.88 ± 2% -0.6 3.25 ± 2% mpstat.cpu.all.idle%
0.00 ± 3% -0.0 0.00 ± 8% mpstat.cpu.all.iowait%
307629 ± 3% +10.5% 340075 ± 3% numa-numastat.node0.local_node
15503 ± 60% +60.2% 24839 ± 25% numa-numastat.node1.other_node
161670 -58.7% 66739 ± 4% vmstat.system.cs
209406 -3.9% 201176 vmstat.system.in
364.00 ± 23% -53.8% 168.00 slabinfo.pid_namespace.active_objs
364.00 ± 23% -53.8% 168.00 slabinfo.pid_namespace.num_objs
985.50 ± 11% +14.6% 1129 ± 8% slabinfo.task_group.active_objs
985.50 ± 11% +14.6% 1129 ± 8% slabinfo.task_group.num_objs
660.25 ± 33% -36.3% 420.25 proc-vmstat.nr_active_file
302055 +2.4% 309211 proc-vmstat.nr_file_pages
281010 +2.6% 288385 proc-vmstat.nr_unevictable
660.25 ± 33% -36.3% 420.25 proc-vmstat.nr_zone_active_file
281010 +2.6% 288385 proc-vmstat.nr_zone_unevictable
20640832 ± 16% +40.0% 28888446 ± 6% cpuidle.C1.time
1743036 ± 6% -54.5% 792376 ± 4% cpuidle.C1.usage
5.048e+08 ± 54% -98.3% 8642335 ± 2% cpuidle.C6.time
706531 ± 51% -94.9% 36224 cpuidle.C6.usage
38313880 -56.5% 16666274 ± 5% cpuidle.POLL.time
18289550 -59.1% 7488947 ± 5% cpuidle.POLL.usage
302.94 ± 5% -32.9% 203.13 ± 6% sched_debug.cfs_rq:/.exec_clock.stddev
31707 ± 6% -43.6% 17867 ± 6% sched_debug.cfs_rq:/.min_vruntime.stddev
0.77 ± 22% +41.1% 1.09 ± 4% sched_debug.cfs_rq:/.nr_spread_over.avg
163292 ± 15% -75.8% 39543 ± 24% sched_debug.cfs_rq:/.spread0.avg
220569 ± 13% -65.9% 75287 ± 16% sched_debug.cfs_rq:/.spread0.max
-1903 +1952.7% -39073 sched_debug.cfs_rq:/.spread0.min
31680 ± 6% -43.7% 17850 ± 6% sched_debug.cfs_rq:/.spread0.stddev
698820 ± 2% -28.4% 500526 ± 3% sched_debug.cpu.avg_idle.avg
1100275 ± 3% -7.2% 1020875 ± 3% sched_debug.cpu.avg_idle.max
250869 -58.4% 104239 ± 4% sched_debug.cpu.nr_switches.avg
766741 ± 25% -64.8% 269919 ± 10% sched_debug.cpu.nr_switches.max
111347 ± 16% -50.8% 54786 ± 11% sched_debug.cpu.nr_switches.min
108077 ± 11% -67.3% 35316 ± 8% sched_debug.cpu.nr_switches.stddev
262769 -59.1% 107346 ± 4% sched_debug.cpu.sched_count.avg
800567 ± 25% -65.9% 272755 ± 10% sched_debug.cpu.sched_count.max
115870 ± 15% -51.6% 56034 ± 11% sched_debug.cpu.sched_count.min
112678 ± 11% -67.8% 36309 ± 9% sched_debug.cpu.sched_count.stddev
122760 -59.2% 50082 ± 4% sched_debug.cpu.sched_goidle.avg
372289 ± 25% -65.9% 126854 ± 10% sched_debug.cpu.sched_goidle.max
53911 ± 15% -51.7% 26040 ± 11% sched_debug.cpu.sched_goidle.min
52986 ± 12% -67.7% 17092 ± 9% sched_debug.cpu.sched_goidle.stddev
138914 -59.1% 56816 ± 4% sched_debug.cpu.ttwu_count.avg
168089 -57.1% 72151 ± 4% sched_debug.cpu.ttwu_count.max
123046 -61.1% 47837 ± 4% sched_debug.cpu.ttwu_count.min
11828 ± 15% -39.8% 7114 ± 4% sched_debug.cpu.ttwu_count.stddev
3050 ± 6% -34.3% 2004 ± 6% sched_debug.cpu.ttwu_local.max
383.97 ± 6% -30.1% 268.54 ± 12% sched_debug.cpu.ttwu_local.stddev
51.21 -0.4 50.79 perf-profile.calltrace.cycles-pp.rwsem_down_write_slowpath.down_write_killable.vm_mmap_pgoff.ksys_mmap_pgoff.do_syscall_64
51.24 -0.4 50.82 perf-profile.calltrace.cycles-pp.down_write_killable.vm_mmap_pgoff.ksys_mmap_pgoff.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.84 ± 8% -0.1 0.76 ± 4% perf-profile.calltrace.cycles-pp.secondary_startup_64
0.83 ± 8% -0.1 0.74 ± 4% perf-profile.calltrace.cycles-pp.cpu_startup_entry.start_secondary.secondary_startup_64
0.83 ± 8% -0.1 0.74 ± 4% perf-profile.calltrace.cycles-pp.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64
0.83 ± 8% -0.1 0.74 ± 4% perf-profile.calltrace.cycles-pp.start_secondary.secondary_startup_64
47.27 +0.5 47.79 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.__munmap
47.27 +0.5 47.79 perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.__munmap
47.26 +0.5 47.78 perf-profile.calltrace.cycles-pp.__x64_sys_munmap.do_syscall_64.entry_SYSCALL_64_after_hwframe.__munmap
47.25 +0.5 47.78 perf-profile.calltrace.cycles-pp.__vm_munmap.__x64_sys_munmap.do_syscall_64.entry_SYSCALL_64_after_hwframe.__munmap
47.28 +0.5 47.81 perf-profile.calltrace.cycles-pp.__munmap
46.75 +0.5 47.30 perf-profile.calltrace.cycles-pp.rwsem_down_write_slowpath.down_write_killable.__vm_munmap.__x64_sys_munmap.do_syscall_64
46.24 +0.5 46.79 perf-profile.calltrace.cycles-pp.osq_lock.rwsem_optimistic_spin.rwsem_down_write_slowpath.down_write_killable.__vm_munmap
46.77 +0.6 47.33 perf-profile.calltrace.cycles-pp.down_write_killable.__vm_munmap.__x64_sys_munmap.do_syscall_64.entry_SYSCALL_64_after_hwframe
46.70 +0.6 47.26 perf-profile.calltrace.cycles-pp.rwsem_optimistic_spin.rwsem_down_write_slowpath.down_write_killable.__vm_munmap.__x64_sys_munmap
0.84 ± 8% -0.1 0.76 ± 4% perf-profile.children.cycles-pp.secondary_startup_64
0.84 ± 8% -0.1 0.76 ± 4% perf-profile.children.cycles-pp.cpu_startup_entry
0.84 ± 8% -0.1 0.76 ± 4% perf-profile.children.cycles-pp.do_idle
0.83 ± 8% -0.1 0.74 ± 4% perf-profile.children.cycles-pp.start_secondary
0.11 ± 4% -0.1 0.04 ± 57% perf-profile.children.cycles-pp.__sched_text_start
0.14 ± 3% -0.1 0.08 ± 6% perf-profile.children.cycles-pp.rwsem_wake
0.10 ± 4% -0.0 0.06 ± 9% perf-profile.children.cycles-pp.wake_up_q
0.10 ± 4% -0.0 0.05 perf-profile.children.cycles-pp.try_to_wake_up
0.07 ± 7% -0.0 0.05 perf-profile.children.cycles-pp._raw_spin_lock_irqsave
0.24 ± 2% +0.0 0.26 perf-profile.children.cycles-pp.mmap_region
0.42 ± 2% +0.0 0.45 perf-profile.children.cycles-pp.do_mmap
0.67 +0.0 0.71 ± 2% perf-profile.children.cycles-pp.rwsem_spin_on_owner
97.96 +0.1 98.09 perf-profile.children.cycles-pp.rwsem_down_write_slowpath
98.02 +0.1 98.14 perf-profile.children.cycles-pp.down_write_killable
97.86 +0.2 98.02 perf-profile.children.cycles-pp.rwsem_optimistic_spin
96.90 +0.2 97.07 perf-profile.children.cycles-pp.osq_lock
47.26 +0.5 47.78 perf-profile.children.cycles-pp.__x64_sys_munmap
47.28 +0.5 47.81 perf-profile.children.cycles-pp.__munmap
47.25 +0.5 47.78 perf-profile.children.cycles-pp.__vm_munmap
0.24 -0.0 0.19 ± 4% perf-profile.self.cycles-pp.rwsem_optimistic_spin
0.06 -0.0 0.03 ±100% perf-profile.self.cycles-pp._raw_spin_lock_irqsave
0.09 +0.0 0.10 perf-profile.self.cycles-pp.find_vma
0.65 +0.0 0.70 ± 2% perf-profile.self.cycles-pp.rwsem_spin_on_owner
96.31 +0.2 96.53 perf-profile.self.cycles-pp.osq_lock
0.66 -21.0% 0.52 ± 2% perf-stat.i.MPKI
0.11 +0.0 0.11 perf-stat.i.branch-miss-rate%
14774053 +4.3% 15410076 perf-stat.i.branch-misses
41.17 +0.9 42.09 perf-stat.i.cache-miss-rate%
19999033 -18.7% 16264313 ± 3% perf-stat.i.cache-misses
48696879 -20.4% 38779964 ± 2% perf-stat.i.cache-references
162553 -58.9% 66790 ± 4% perf-stat.i.context-switches
157.36 -3.9% 151.20 perf-stat.i.cpu-migrations
14271 +23.9% 17676 ± 3% perf-stat.i.cycles-between-cache-misses
0.00 ± 8% +0.0 0.00 ± 4% perf-stat.i.dTLB-load-miss-rate%
155048 ± 5% +46.3% 226826 ± 4% perf-stat.i.dTLB-load-misses
0.00 ± 2% +0.0 0.01 perf-stat.i.dTLB-store-miss-rate%
16761 +19.2% 19972 ± 2% perf-stat.i.dTLB-store-misses
4.118e+08 -15.7% 3.47e+08 perf-stat.i.dTLB-stores
93.99 +1.4 95.42 perf-stat.i.iTLB-load-miss-rate%
4103535 ± 3% -6.7% 3827881 perf-stat.i.iTLB-load-misses
263239 ± 4% -28.6% 188037 ± 2% perf-stat.i.iTLB-loads
18498 ± 3% +7.5% 19885 perf-stat.i.instructions-per-iTLB-miss
0.31 +226.8% 1.01 ± 3% perf-stat.i.metric.K/sec
88.51 +1.8 90.30 perf-stat.i.node-load-miss-rate%
8104644 -13.5% 7009101 perf-stat.i.node-load-misses
1047577 -28.3% 751122 perf-stat.i.node-loads
3521008 -13.1% 3058055 perf-stat.i.node-store-misses
0.64 -20.6% 0.51 ± 2% perf-stat.overall.MPKI
0.10 +0.0 0.10 perf-stat.overall.branch-miss-rate%
41.07 +0.9 41.92 perf-stat.overall.cache-miss-rate%
14273 +23.7% 17662 ± 3% perf-stat.overall.cycles-between-cache-misses
0.00 ± 5% +0.0 0.00 ± 5% perf-stat.overall.dTLB-load-miss-rate%
0.00 ± 2% +0.0 0.01 ± 2% perf-stat.overall.dTLB-store-miss-rate%
93.95 +1.3 95.29 perf-stat.overall.iTLB-load-miss-rate%
18507 ± 3% +7.5% 19890 perf-stat.overall.instructions-per-iTLB-miss
88.55 +1.8 90.32 perf-stat.overall.node-load-miss-rate%
1.191e+08 -7.0% 1.107e+08 perf-stat.overall.path-length
14739033 +4.2% 15364552 perf-stat.ps.branch-misses
19930054 -18.7% 16209879 ± 3% perf-stat.ps.cache-misses
48535919 -20.3% 38662093 ± 2% perf-stat.ps.cache-references
161841 -58.9% 66477 ± 4% perf-stat.ps.context-switches
157.06 -3.9% 150.89 perf-stat.ps.cpu-migrations
154906 ± 5% +46.1% 226269 ± 5% perf-stat.ps.dTLB-load-misses
16730 +19.1% 19927 ± 2% perf-stat.ps.dTLB-store-misses
4.104e+08 -15.7% 3.459e+08 perf-stat.ps.dTLB-stores
4089203 ± 3% -6.7% 3814320 perf-stat.ps.iTLB-load-misses
263134 ± 4% -28.4% 188441 ± 2% perf-stat.ps.iTLB-loads
8076488 -13.5% 6984923 perf-stat.ps.node-load-misses
1044002 -28.3% 748425 perf-stat.ps.node-loads
3509068 -13.1% 3048190 perf-stat.ps.node-store-misses
2270 ±170% -98.9% 24.50 ±166% interrupts.36:PCI-MSI.31981569-edge.i40e-eth0-TxRx-0
3730764 ± 2% -62.0% 1416300 ± 4% interrupts.CAL:Function_call_interrupts
2270 ±170% -98.9% 24.25 ±168% interrupts.CPU0.36:PCI-MSI.31981569-edge.i40e-eth0-TxRx-0
111116 ± 29% -67.4% 36227 ± 8% interrupts.CPU0.CAL:Function_call_interrupts
11091 ± 38% -62.2% 4195 ± 12% interrupts.CPU0.RES:Rescheduling_interrupts
48001 ± 26% -53.7% 22228 ± 14% interrupts.CPU1.CAL:Function_call_interrupts
4189 ± 29% -43.2% 2378 ± 5% interrupts.CPU1.RES:Rescheduling_interrupts
34024 ± 32% -44.7% 18804 ± 4% interrupts.CPU10.CAL:Function_call_interrupts
26171 ± 19% -26.1% 19334 ± 3% interrupts.CPU11.CAL:Function_call_interrupts
2370 ± 13% -20.8% 1876 ± 18% interrupts.CPU11.RES:Rescheduling_interrupts
30568 ± 23% -45.9% 16537 ± 4% interrupts.CPU12.CAL:Function_call_interrupts
3094 ± 27% -46.9% 1643 ± 8% interrupts.CPU12.RES:Rescheduling_interrupts
514.50 ± 10% +35.6% 697.75 ± 16% interrupts.CPU12.TLB:TLB_shootdowns
29819 ± 32% -41.7% 17393 interrupts.CPU13.CAL:Function_call_interrupts
35364 ± 38% -49.6% 17819 ± 9% interrupts.CPU14.CAL:Function_call_interrupts
694.75 ± 11% +23.2% 856.00 ± 9% interrupts.CPU14.TLB:TLB_shootdowns
38361 ± 48% -49.8% 19273 ± 9% interrupts.CPU16.CAL:Function_call_interrupts
30069 ± 23% -31.6% 20559 ± 3% interrupts.CPU17.CAL:Function_call_interrupts
809.75 ± 7% -19.5% 651.75 ± 13% interrupts.CPU17.TLB:TLB_shootdowns
28245 ± 29% -33.1% 18894 ± 6% interrupts.CPU18.CAL:Function_call_interrupts
33560 ± 23% -48.2% 17369 ± 6% interrupts.CPU19.CAL:Function_call_interrupts
2863 ± 19% -40.3% 1709 ± 13% interrupts.CPU19.RES:Rescheduling_interrupts
47118 ± 32% -55.7% 20868 ± 3% interrupts.CPU2.CAL:Function_call_interrupts
3897 ± 38% -48.9% 1991 ± 7% interrupts.CPU2.RES:Rescheduling_interrupts
34735 ± 29% -41.7% 20246 ± 9% interrupts.CPU20.CAL:Function_call_interrupts
37232 ± 23% -46.6% 19883 ± 12% interrupts.CPU21.CAL:Function_call_interrupts
32345 ± 16% -38.6% 19845 ± 6% interrupts.CPU22.CAL:Function_call_interrupts
34083 ± 22% -43.4% 19301 ± 6% interrupts.CPU23.CAL:Function_call_interrupts
61308 ± 16% -76.3% 14529 ± 13% interrupts.CPU24.CAL:Function_call_interrupts
6610 ± 26% -76.3% 1568 ± 11% interrupts.CPU24.RES:Rescheduling_interrupts
51384 ± 32% -75.0% 12848 ± 12% interrupts.CPU25.CAL:Function_call_interrupts
4643 ± 39% -70.6% 1366 ± 9% interrupts.CPU25.RES:Rescheduling_interrupts
48788 ± 25% -71.7% 13826 ± 17% interrupts.CPU26.CAL:Function_call_interrupts
4076 ± 32% -70.5% 1203 ± 18% interrupts.CPU26.RES:Rescheduling_interrupts
45702 ± 14% -70.7% 13369 ± 12% interrupts.CPU27.CAL:Function_call_interrupts
3614 ± 21% -67.3% 1180 ± 16% interrupts.CPU27.RES:Rescheduling_interrupts
51216 ± 14% -71.6% 14546 ± 15% interrupts.CPU28.CAL:Function_call_interrupts
4395 ± 24% -67.9% 1410 ± 21% interrupts.CPU28.RES:Rescheduling_interrupts
614.25 ± 18% +33.3% 818.75 ± 17% interrupts.CPU28.TLB:TLB_shootdowns
44945 ± 23% -66.5% 15059 ± 14% interrupts.CPU29.CAL:Function_call_interrupts
3994 ± 34% -68.2% 1271 ± 10% interrupts.CPU29.RES:Rescheduling_interrupts
39154 ± 24% -41.6% 22857 ± 6% interrupts.CPU3.CAL:Function_call_interrupts
45674 ± 11% -68.3% 14470 ± 8% interrupts.CPU30.CAL:Function_call_interrupts
4097 ± 23% -68.8% 1278 ± 10% interrupts.CPU30.RES:Rescheduling_interrupts
51890 ± 13% -72.6% 14227 ± 16% interrupts.CPU31.CAL:Function_call_interrupts
4557 ± 26% -71.4% 1305 ± 21% interrupts.CPU31.RES:Rescheduling_interrupts
41324 ± 23% -76.0% 9933 ± 11% interrupts.CPU32.CAL:Function_call_interrupts
3284 ± 33% -73.4% 873.75 ± 15% interrupts.CPU32.RES:Rescheduling_interrupts
39758 ± 31% -74.5% 10120 ± 17% interrupts.CPU33.CAL:Function_call_interrupts
3373 ± 42% -74.2% 869.00 ± 15% interrupts.CPU33.RES:Rescheduling_interrupts
513.00 ± 27% +46.0% 748.75 ± 16% interrupts.CPU33.TLB:TLB_shootdowns
40015 ± 14% -72.8% 10885 ± 8% interrupts.CPU34.CAL:Function_call_interrupts
3402 ± 25% -68.2% 1080 ± 13% interrupts.CPU34.RES:Rescheduling_interrupts
635.25 ± 22% +49.3% 948.25 ± 13% interrupts.CPU34.TLB:TLB_shootdowns
45251 ± 19% -75.2% 11204 ± 17% interrupts.CPU35.CAL:Function_call_interrupts
3731 ± 31% -73.4% 992.50 ± 20% interrupts.CPU35.RES:Rescheduling_interrupts
43390 ± 11% -78.3% 9434 ± 15% interrupts.CPU36.CAL:Function_call_interrupts
3536 ± 23% -77.3% 803.75 ± 14% interrupts.CPU36.RES:Rescheduling_interrupts
39820 ± 11% -75.9% 9613 ± 10% interrupts.CPU37.CAL:Function_call_interrupts
2987 ± 21% -70.8% 873.25 ± 9% interrupts.CPU37.RES:Rescheduling_interrupts
42969 ± 32% -76.6% 10068 ± 17% interrupts.CPU38.CAL:Function_call_interrupts
3202 ± 36% -74.4% 818.50 ± 27% interrupts.CPU38.RES:Rescheduling_interrupts
35571 ± 16% -72.4% 9822 ± 9% interrupts.CPU39.CAL:Function_call_interrupts
2986 ± 24% -73.9% 778.75 ± 15% interrupts.CPU39.RES:Rescheduling_interrupts
45001 ± 21% -48.2% 23317 ± 7% interrupts.CPU4.CAL:Function_call_interrupts
3689 ± 24% -43.6% 2080 ± 2% interrupts.CPU4.RES:Rescheduling_interrupts
39302 ± 21% -73.4% 10463 ± 8% interrupts.CPU40.CAL:Function_call_interrupts
2968 ± 32% -71.4% 848.50 ± 18% interrupts.CPU40.RES:Rescheduling_interrupts
40826 ± 19% -75.3% 10070 ± 10% interrupts.CPU41.CAL:Function_call_interrupts
3321 ± 30% -70.9% 967.25 ± 10% interrupts.CPU41.RES:Rescheduling_interrupts
700.25 ± 18% +26.2% 883.75 ± 17% interrupts.CPU41.TLB:TLB_shootdowns
35368 ± 11% -73.7% 9308 ± 15% interrupts.CPU42.CAL:Function_call_interrupts
2839 ± 12% -70.3% 844.50 ± 13% interrupts.CPU42.RES:Rescheduling_interrupts
45459 ± 25% -78.7% 9687 ± 11% interrupts.CPU43.CAL:Function_call_interrupts
3703 ± 29% -74.1% 959.50 ± 16% interrupts.CPU43.RES:Rescheduling_interrupts
41495 ± 15% -77.1% 9522 ± 12% interrupts.CPU44.CAL:Function_call_interrupts
3153 ± 28% -75.0% 789.75 ± 15% interrupts.CPU44.RES:Rescheduling_interrupts
38501 ± 26% -72.5% 10601 ± 14% interrupts.CPU45.CAL:Function_call_interrupts
3024 ± 38% -73.8% 791.00 ± 19% interrupts.CPU45.RES:Rescheduling_interrupts
39083 ± 35% -73.6% 10323 ± 18% interrupts.CPU46.CAL:Function_call_interrupts
3173 ± 37% -73.9% 829.75 ± 24% interrupts.CPU46.RES:Rescheduling_interrupts
44486 ± 20% -75.3% 10968 ± 15% interrupts.CPU47.CAL:Function_call_interrupts
3773 ± 34% -76.4% 890.25 ± 24% interrupts.CPU47.RES:Rescheduling_interrupts
34967 ± 42% -51.0% 17117 ± 10% interrupts.CPU48.CAL:Function_call_interrupts
31969 ± 38% -51.7% 15432 ± 12% interrupts.CPU49.CAL:Function_call_interrupts
33786 ± 12% -29.2% 23918 ± 5% interrupts.CPU5.CAL:Function_call_interrupts
3014 ± 16% -33.2% 2012 ± 9% interrupts.CPU5.RES:Rescheduling_interrupts
30514 ± 29% -46.4% 16343 ± 6% interrupts.CPU51.CAL:Function_call_interrupts
34448 ± 26% -48.7% 17686 ± 4% interrupts.CPU52.CAL:Function_call_interrupts
2811 ± 34% -42.0% 1631 ± 4% interrupts.CPU52.RES:Rescheduling_interrupts
30848 ± 33% -47.9% 16059 ± 3% interrupts.CPU54.CAL:Function_call_interrupts
31017 ± 22% -52.7% 14676 ± 7% interrupts.CPU55.CAL:Function_call_interrupts
2501 ± 41% -41.8% 1455 ± 10% interrupts.CPU55.RES:Rescheduling_interrupts
28249 ± 23% -46.9% 14997 ± 10% interrupts.CPU56.CAL:Function_call_interrupts
2113 ± 18% -36.0% 1352 ± 15% interrupts.CPU56.RES:Rescheduling_interrupts
27658 ± 20% -49.3% 14034 ± 3% interrupts.CPU57.CAL:Function_call_interrupts
26559 ± 34% -40.6% 15778 ± 11% interrupts.CPU58.CAL:Function_call_interrupts
27984 ± 27% -39.9% 16815 ± 12% interrupts.CPU59.CAL:Function_call_interrupts
35098 ± 33% -37.5% 21921 interrupts.CPU6.CAL:Function_call_interrupts
3073 ± 37% -40.6% 1825 ± 8% interrupts.CPU6.RES:Rescheduling_interrupts
29248 ± 38% -48.2% 15149 ± 6% interrupts.CPU60.CAL:Function_call_interrupts
30880 ± 33% -52.3% 14722 ± 10% interrupts.CPU61.CAL:Function_call_interrupts
31218 ± 43% -51.5% 15152 ± 3% interrupts.CPU62.CAL:Function_call_interrupts
29210 ± 40% -46.5% 15627 ± 6% interrupts.CPU63.CAL:Function_call_interrupts
26813 ± 15% -39.0% 16343 ± 13% interrupts.CPU64.CAL:Function_call_interrupts
24791 ± 14% -32.6% 16704 ± 10% interrupts.CPU67.CAL:Function_call_interrupts
29638 ± 33% -42.9% 16914 ± 9% interrupts.CPU68.CAL:Function_call_interrupts
36247 ± 33% -48.5% 18670 ± 8% interrupts.CPU69.CAL:Function_call_interrupts
30379 ± 24% -30.6% 21096 ± 5% interrupts.CPU7.CAL:Function_call_interrupts
3027 ± 25% -32.0% 2059 ± 3% interrupts.CPU7.RES:Rescheduling_interrupts
31064 ± 25% -42.8% 17774 interrupts.CPU70.CAL:Function_call_interrupts
52949 ± 14% -77.0% 12198 ± 10% interrupts.CPU72.CAL:Function_call_interrupts
4057 ± 23% -75.7% 985.00 ± 8% interrupts.CPU72.RES:Rescheduling_interrupts
42694 ± 22% -73.6% 11281 ± 16% interrupts.CPU73.CAL:Function_call_interrupts
3318 ± 39% -74.3% 851.75 ± 16% interrupts.CPU73.RES:Rescheduling_interrupts
49143 ± 24% -76.1% 11756 ± 12% interrupts.CPU74.CAL:Function_call_interrupts
3606 ± 31% -73.8% 946.00 ± 12% interrupts.CPU74.RES:Rescheduling_interrupts
50587 ± 24% -72.5% 13930 ± 20% interrupts.CPU75.CAL:Function_call_interrupts
3655 ± 36% -71.1% 1056 ± 12% interrupts.CPU75.RES:Rescheduling_interrupts
57791 ± 21% -78.4% 12488 ± 11% interrupts.CPU76.CAL:Function_call_interrupts
5109 ± 37% -79.7% 1037 ± 20% interrupts.CPU76.RES:Rescheduling_interrupts
52455 ± 26% -75.4% 12922 ± 5% interrupts.CPU77.CAL:Function_call_interrupts
3997 ± 37% -73.9% 1043 ± 14% interrupts.CPU77.RES:Rescheduling_interrupts
49188 ± 21% -74.5% 12521 ± 10% interrupts.CPU78.CAL:Function_call_interrupts
3867 ± 42% -74.5% 986.25 ± 18% interrupts.CPU78.RES:Rescheduling_interrupts
45517 ± 19% -72.6% 12484 ± 19% interrupts.CPU79.CAL:Function_call_interrupts
3369 ± 34% -71.9% 946.25 ± 20% interrupts.CPU79.RES:Rescheduling_interrupts
30702 ± 22% -39.9% 18462 ± 9% interrupts.CPU8.CAL:Function_call_interrupts
2580 ± 28% -39.9% 1550 ± 10% interrupts.CPU8.RES:Rescheduling_interrupts
35561 ± 30% -69.8% 10728 ± 12% interrupts.CPU80.CAL:Function_call_interrupts
2675 ± 44% -70.6% 787.50 ± 7% interrupts.CPU80.RES:Rescheduling_interrupts
38762 ± 33% -73.3% 10349 ± 18% interrupts.CPU81.CAL:Function_call_interrupts
2892 ± 48% -70.5% 853.25 ± 20% interrupts.CPU81.RES:Rescheduling_interrupts
46500 ± 39% -80.2% 9203 ± 6% interrupts.CPU82.CAL:Function_call_interrupts
3726 ± 41% -83.3% 622.75 ± 12% interrupts.CPU82.RES:Rescheduling_interrupts
42125 ± 25% -76.0% 10103 ± 7% interrupts.CPU83.CAL:Function_call_interrupts
3275 ± 40% -75.4% 804.50 ± 6% interrupts.CPU83.RES:Rescheduling_interrupts
37359 ± 28% -74.7% 9436 ± 7% interrupts.CPU84.CAL:Function_call_interrupts
2762 ± 45% -71.1% 797.50 ± 17% interrupts.CPU84.RES:Rescheduling_interrupts
38900 ± 13% -76.2% 9272 ± 8% interrupts.CPU85.CAL:Function_call_interrupts
2704 ± 27% -77.0% 622.25 ± 10% interrupts.CPU85.RES:Rescheduling_interrupts
40662 ± 24% -77.2% 9274 ± 14% interrupts.CPU86.CAL:Function_call_interrupts
3139 ± 39% -79.5% 643.00 ± 28% interrupts.CPU86.RES:Rescheduling_interrupts
33538 ± 23% -71.7% 9484 ± 14% interrupts.CPU87.CAL:Function_call_interrupts
2406 ± 40% -73.5% 638.25 ± 21% interrupts.CPU87.RES:Rescheduling_interrupts
36240 ± 26% -73.8% 9499 ± 10% interrupts.CPU88.CAL:Function_call_interrupts
2450 ± 39% -70.5% 721.75 ± 5% interrupts.CPU88.RES:Rescheduling_interrupts
41267 ± 29% -77.1% 9463 ± 11% interrupts.CPU89.CAL:Function_call_interrupts
3286 ± 34% -73.2% 879.50 ± 17% interrupts.CPU89.RES:Rescheduling_interrupts
36038 ± 18% -50.6% 17796 ± 3% interrupts.CPU9.CAL:Function_call_interrupts
3140 ± 28% -48.3% 1622 ± 9% interrupts.CPU9.RES:Rescheduling_interrupts
38534 ± 25% -77.5% 8675 ± 9% interrupts.CPU90.CAL:Function_call_interrupts
3008 ± 27% -79.1% 629.50 ± 16% interrupts.CPU90.RES:Rescheduling_interrupts
38422 ± 29% -77.2% 8741 ± 14% interrupts.CPU91.CAL:Function_call_interrupts
3095 ± 51% -78.7% 658.75 ± 14% interrupts.CPU91.RES:Rescheduling_interrupts
38120 ± 45% -73.6% 10059 ± 10% interrupts.CPU92.CAL:Function_call_interrupts
2711 ± 61% -73.3% 722.75 ± 10% interrupts.CPU92.RES:Rescheduling_interrupts
37155 ± 19% -74.1% 9628 ± 12% interrupts.CPU93.CAL:Function_call_interrupts
2724 ± 32% -70.4% 806.00 ± 32% interrupts.CPU93.RES:Rescheduling_interrupts
43458 ± 15% -77.1% 9936 ± 10% interrupts.CPU94.CAL:Function_call_interrupts
2832 ± 25% -76.8% 655.75 ± 18% interrupts.CPU94.RES:Rescheduling_interrupts
54226 ± 22% -76.8% 12596 ± 17% interrupts.CPU95.CAL:Function_call_interrupts
4437 ± 37% -80.6% 860.50 ± 11% interrupts.CPU95.RES:Rescheduling_interrupts
302853 ± 2% -58.2% 126676 ± 2% interrupts.RES:Rescheduling_interrupts
will-it-scale.per_thread_ops
2300 +--------------------------------------------------------------------+
| |
2250 |-+ O O O O O |
2200 |-+ O O O O |
| O O O O O O O O |
2150 |-O O O O O O |
| O O O O |
2100 |-+ + |
| : + + +.. .+. |
2050 |.+ : + .+.. + : : + +. +.. + |
2000 |-+..+ + + : : + + + |
| : : + + |
1950 |-+ +.+.. +..+ |
| + |
1900 +--------------------------------------------------------------------+
[*] bisect-good sample
[O] bisect-bad sample
Disclaimer:
Results have been estimated based on internal Intel analysis and are provided
for informational purposes only. Any difference in system hardware or software
design or configuration may affect actual performance.
Thanks,
Rong Chen
View attachment "config-5.8.0-00001-g698ac7610f7928" of type "text/plain" (169434 bytes)
View attachment "job-script" of type "text/plain" (7989 bytes)
View attachment "job.yaml" of type "text/plain" (5431 bytes)
View attachment "reproduce" of type "text/plain" (336 bytes)
Powered by blists - more mailing lists