| 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: <20171204053538.GY21779@yexl-desktop>
Date: Mon, 4 Dec 2017 13:35:38 +0800
From: kernel test robot <xiaolong.ye@...el.com>
To: Shakeel Butt <shakeelb@...gle.com>
Cc: Stephen Rothwell <sfr@...b.auug.org.au>,
Vlastimil Babka <vbabka@...e.cz>,
Jérôme Glisse <jglisse@...hat.com>,
Huang Ying <ying.huang@...el.com>,
Tim Chen <tim.c.chen@...ux.intel.com>,
Michal Hocko <mhocko@...nel.org>,
Greg Thelen <gthelen@...gle.com>,
Johannes Weiner <hannes@...xchg.org>,
Balbir Singh <bsingharora@...il.com>,
Minchan Kim <minchan@...nel.org>, Shaohua Li <shli@...com>,
Jan Kara <jack@...e.cz>, Nicholas Piggin <npiggin@...il.com>,
Dan Williams <dan.j.williams@...el.com>,
Mel Gorman <mgorman@...e.de>, Hugh Dickins <hughd@...gle.com>,
Andrew Morton <akpm@...ux-foundation.org>,
LKML <linux-kernel@...r.kernel.org>, lkp@...org
Subject: [lkp-robot] [mm, mlock, vmscan] ae938e2990: reaim.jobs_per_min
-12.1% regression
Greeting,
FYI, we noticed a -12.1% regression of reaim.jobs_per_min due to commit:
commit: ae938e2990fbfb4e7ed92e4a6f494a47c418dde7 ("mm, mlock, vmscan: no more skipping pagevecs")
https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git master
in testcase: reaim
on test machine: 56 threads Intel(R) Xeon(R) CPU E5-2695 v3 @ 2.30GHz with 256G memory
with following parameters:
runtime: 300s
nr_task: 1000
test: page_test
cpufreq_governor: performance
test-description: REAIM is an updated and improved version of AIM 7 benchmark.
test-url: https://sourceforge.net/projects/re-aim-7/
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/nr_task/rootfs/runtime/tbox_group/test/testcase:
gcc-7/performance/x86_64-rhel-7.2/1000/debian-x86_64-2016-08-31.cgz/300s/lkp-hsw-ep5/page_test/reaim
commit:
9b6acff687 ("mm: use sc->priority for slab shrink targets")
ae938e2990 ("mm, mlock, vmscan: no more skipping pagevecs")
9b6acff687a42dd6 ae938e2990fbfb4e7ed92e4a6f
---------------- --------------------------
%stddev %change %stddev
\ | \
345296 -12.1% 303560 reaim.jobs_per_min
345.30 -12.1% 303.56 reaim.jobs_per_min_child
932.21 +14.3% 1065 reaim.child_systime
38.02 +1.9% 38.75 reaim.child_utime
352222 -12.4% 308566 reaim.max_jobs_per_min
17.38 +13.7% 19.77 reaim.parent_time
312.15 -1.8% 306.66 reaim.time.elapsed_time
312.15 -1.8% 306.66 reaim.time.elapsed_time.max
2.561e+09 -12.5% 2.241e+09 reaim.time.minor_page_faults
4973 +1.4% 5041 reaim.time.percent_of_cpu_this_job_got
608.53 -10.8% 542.73 reaim.time.user_time
118566 ± 6% +116.4% 256534 ± 5% reaim.time.voluntary_context_switches
294692 ± 4% +38.3% 407448 ± 3% interrupts.CAL:Function_call_interrupts
2.50 ± 44% +500.0% 15.00 ± 24% vmstat.procs.b
13.94 ± 22% -3.3 10.63 ± 4% mpstat.cpu.idle%
0.00 ± 41% +0.0 0.00 ± 26% mpstat.cpu.iowait%
12048 ± 5% -10.8% 10749 ± 5% slabinfo.kmalloc-96.active_objs
12186 ± 5% -9.7% 11008 ± 5% slabinfo.kmalloc-96.num_objs
5573497 ± 21% -19.6% 4480333 ± 4% cpuidle.C3.time
2.418e+09 ± 27% -27.7% 1.749e+09 ± 3% cpuidle.C6.time
2507881 ± 27% -27.6% 1816444 ± 3% cpuidle.C6.usage
1.385e+09 -12.9% 1.207e+09 numa-numastat.node0.local_node
1.385e+09 -12.9% 1.207e+09 numa-numastat.node0.numa_hit
1.35e+09 -12.7% 1.178e+09 numa-numastat.node1.local_node
1.35e+09 -12.7% 1.178e+09 numa-numastat.node1.numa_hit
245679 -19.5% 197748 ± 2% meminfo.Active
47203 -100.0% 2.75 ± 59% meminfo.Active(file)
1194766 -99.1% 10528 meminfo.Inactive
1183902 -100.0% 1.00 ±173% meminfo.Inactive(file)
484.00 ±146% +2.6e+05% 1235282 meminfo.Unevictable
2282 ± 3% +4.4% 2382 turbostat.Avg_MHz
19925 ± 23% -20.6% 15817 ± 4% turbostat.C3
2506786 ± 27% -27.6% 1815322 ± 3% turbostat.C6
13.25 ± 23% -3.2 10.10 ± 3% turbostat.C6%
4.01 ± 21% -22.8% 3.10 ± 3% turbostat.CPU%c1
9.21 ± 23% -24.1% 6.99 ± 3% turbostat.CPU%c6
6.38 ± 19% -27.4% 4.63 ± 6% turbostat.Pkg%pc2
23602 ± 6% -100.0% 1.75 ±102% numa-meminfo.node0.Active(file)
591666 -98.8% 7111 ± 50% numa-meminfo.node0.Inactive
586621 -100.0% 1.00 ±173% numa-meminfo.node0.Inactive(file)
273.25 ±151% +2.2e+05% 600760 numa-meminfo.node0.Unevictable
23598 ± 4% -100.0% 1.00 ±173% numa-meminfo.node1.Active(file)
603090 -99.4% 3372 ±107% numa-meminfo.node1.Inactive
5809 ± 71% -41.9% 3372 ±107% numa-meminfo.node1.Inactive(anon)
597280 -100.0% 0.00 numa-meminfo.node1.Inactive(file)
11447 ± 20% -12.8% 9985 ± 21% numa-meminfo.node1.Mapped
213.00 ±137% +3e+05% 634522 numa-meminfo.node1.Unevictable
11800 -100.0% 0.50 ±100% proc-vmstat.nr_active_file
295974 -100.0% 0.25 ±173% proc-vmstat.nr_inactive_file
120.75 ±146% +2.6e+05% 308820 proc-vmstat.nr_unevictable
11800 -100.0% 0.50 ±100% proc-vmstat.nr_zone_active_file
295974 -100.0% 0.25 ±173% proc-vmstat.nr_zone_inactive_file
120.75 ±146% +2.6e+05% 308820 proc-vmstat.nr_zone_unevictable
2881 ± 18% +1614.2% 49394 ± 5% proc-vmstat.numa_hint_faults
167.50 ± 43% +3023.0% 5231 ± 11% proc-vmstat.numa_hint_faults_local
2.735e+09 -12.8% 2.384e+09 proc-vmstat.numa_hit
2.735e+09 -12.8% 2.384e+09 proc-vmstat.numa_local
4519 ± 29% +901.6% 45262 ± 4% proc-vmstat.numa_pages_migrated
33100 ± 18% +1241.9% 444183 ± 2% proc-vmstat.numa_pte_updates
13053 ± 2% -34.9% 8499 ± 27% proc-vmstat.pgactivate
2.735e+09 -12.8% 2.385e+09 proc-vmstat.pgalloc_normal
2.562e+09 -12.5% 2.242e+09 proc-vmstat.pgfault
2.735e+09 -12.8% 2.385e+09 proc-vmstat.pgfree
4519 ± 29% +901.6% 45262 ± 4% proc-vmstat.pgmigrate_success
5900 ± 6% -100.0% 0.25 ±173% numa-vmstat.node0.nr_active_file
146655 -100.0% 0.25 ±173% numa-vmstat.node0.nr_inactive_file
68.25 ±151% +2.2e+05% 150189 numa-vmstat.node0.nr_unevictable
5900 ± 6% -100.0% 0.25 ±173% numa-vmstat.node0.nr_zone_active_file
146655 -100.0% 0.25 ±173% numa-vmstat.node0.nr_zone_inactive_file
68.25 ±151% +2.2e+05% 150189 numa-vmstat.node0.nr_zone_unevictable
7.105e+08 ± 4% -15.4% 6.011e+08 numa-vmstat.node0.numa_hit
7.105e+08 ± 4% -15.4% 6.011e+08 numa-vmstat.node0.numa_local
5899 ± 4% -100.0% 0.25 ±173% numa-vmstat.node1.nr_active_file
1453 ± 71% -41.4% 852.25 ±106% numa-vmstat.node1.nr_inactive_anon
149319 -100.0% 0.00 numa-vmstat.node1.nr_inactive_file
2988 ± 19% -14.2% 2565 ± 22% numa-vmstat.node1.nr_mapped
53.25 ±137% +3e+05% 158629 numa-vmstat.node1.nr_unevictable
5899 ± 4% -100.0% 0.25 ±173% numa-vmstat.node1.nr_zone_active_file
1453 ± 71% -41.4% 852.25 ±106% numa-vmstat.node1.nr_zone_inactive_anon
149319 -100.0% 0.00 numa-vmstat.node1.nr_zone_inactive_file
53.25 ±137% +3e+05% 158629 numa-vmstat.node1.nr_zone_unevictable
6.933e+08 ± 4% -15.1% 5.886e+08 numa-vmstat.node1.numa_hit
6.931e+08 ± 4% -15.1% 5.884e+08 numa-vmstat.node1.numa_local
3.922e+12 -4.7% 3.738e+12 perf-stat.branch-instructions
5.421e+09 ± 2% -7.7% 5.005e+09 ± 2% perf-stat.branch-misses
1.36 +0.0 1.41 perf-stat.cache-miss-rate%
3.368e+08 -2.5% 3.283e+08 perf-stat.cache-misses
2.48e+10 -5.8% 2.335e+10 perf-stat.cache-references
2.30 +5.8% 2.43 perf-stat.cpi
0.04 ± 24% -0.0 0.03 ± 3% perf-stat.dTLB-load-miss-rate%
1.99e+09 ± 24% -26.5% 1.463e+09 ± 3% perf-stat.dTLB-load-misses
4.693e+12 -5.7% 4.424e+12 perf-stat.dTLB-loads
1.281e+10 -12.9% 1.116e+10 perf-stat.dTLB-store-misses
2.394e+12 -12.2% 2.102e+12 perf-stat.dTLB-stores
5.02 ± 10% -2.0 3.05 ± 2% perf-stat.iTLB-load-miss-rate%
2.64e+08 ± 10% -43.3% 1.497e+08 ± 2% perf-stat.iTLB-load-misses
4.994e+09 -4.9% 4.751e+09 perf-stat.iTLB-loads
1.788e+13 -5.4% 1.692e+13 perf-stat.instructions
68514 ± 10% +65.0% 113070 ± 2% perf-stat.instructions-per-iTLB-miss
0.44 -5.5% 0.41 perf-stat.ipc
2.562e+09 -12.5% 2.242e+09 perf-stat.minor-faults
48.30 +3.3 51.64 perf-stat.node-load-miss-rate%
1.177e+08 +4.3% 1.228e+08 ± 2% perf-stat.node-load-misses
1.26e+08 ± 2% -8.8% 1.15e+08 ± 3% perf-stat.node-loads
27.21 +2.8 30.03 ± 2% perf-stat.node-store-miss-rate%
23994106 +5.8% 25385058 ± 3% perf-stat.node-store-misses
64182752 -7.9% 59114451 perf-stat.node-stores
2.562e+09 -12.5% 2.242e+09 perf-stat.page-faults
0.00 +1.3e+12% 12994 ±172% sched_debug.cfs_rq:/.MIN_vruntime.avg
0.00 +7.3e+13% 727685 ±172% sched_debug.cfs_rq:/.MIN_vruntime.max
0.00 ± 6% +1e+28% 96368 ±172% sched_debug.cfs_rq:/.MIN_vruntime.stddev
105.46 ± 5% -13.0% 91.74 ± 11% sched_debug.cfs_rq:/.exec_clock.stddev
22555 ± 10% -21.0% 17813 ± 8% sched_debug.cfs_rq:/.load.avg
2040 ± 15% +92.4% 3926 ± 5% sched_debug.cfs_rq:/.load.min
42230 ± 21% -49.5% 21343 ± 43% sched_debug.cfs_rq:/.load.stddev
0.00 +1.3e+12% 12994 ±172% sched_debug.cfs_rq:/.max_vruntime.avg
0.00 +7.3e+13% 727685 ±172% sched_debug.cfs_rq:/.max_vruntime.max
0.00 ± 6% +1e+28% 96368 ±172% sched_debug.cfs_rq:/.max_vruntime.stddev
11528062 -9.6% 10423060 sched_debug.cfs_rq:/.min_vruntime.avg
13730061 ± 3% -13.7% 11848070 sched_debug.cfs_rq:/.min_vruntime.max
1003571 ± 16% -40.5% 597366 ± 2% sched_debug.cfs_rq:/.min_vruntime.stddev
487.71 ± 11% -28.6% 348.33 ± 23% sched_debug.cfs_rq:/.runnable_load_avg.max
92.89 ± 6% -25.4% 69.26 ± 15% sched_debug.cfs_rq:/.runnable_load_avg.stddev
22416 ± 10% -22.7% 17324 ± 7% sched_debug.cfs_rq:/.runnable_weight.avg
1965 ± 13% +76.8% 3475 ± 12% sched_debug.cfs_rq:/.runnable_weight.min
42026 ± 21% -49.9% 21058 ± 42% sched_debug.cfs_rq:/.runnable_weight.stddev
2193210 ± 28% -39.4% 1328766 ± 13% sched_debug.cfs_rq:/.spread0.max
1003218 ± 16% -40.5% 596958 ± 2% sched_debug.cfs_rq:/.spread0.stddev
293.58 ± 11% +35.0% 396.46 ± 8% sched_debug.cfs_rq:/.util_avg.min
422.38 ± 14% -23.3% 323.96 ± 22% sched_debug.cpu.cpu_load[2].max
77.47 ± 10% -18.4% 63.21 ± 15% sched_debug.cpu.cpu_load[2].stddev
1.92 ± 18% +26.1% 2.42 ± 11% sched_debug.cpu.cpu_load[4].min
2120 ± 19% +85.2% 3926 ± 5% sched_debug.cpu.load.min
2.00 ± 18% +52.1% 3.04 ± 17% sched_debug.cpu.nr_running.min
0.00 ±110% +9666.7% 0.22 ± 26% sched_debug.cpu.nr_uninterruptible.avg
40.54 ± 16% +107.5% 84.12 ± 11% sched_debug.cpu.nr_uninterruptible.max
26.78 ± 8% +63.7% 43.84 ± 5% sched_debug.cpu.nr_uninterruptible.stddev
1532 ± 6% +41.8% 2173 ± 10% sched_debug.cpu.ttwu_count.min
0.01 ±173% +300.0% 0.02 sched_debug.rt_rq:/.rt_nr_migratory.stddev
0.01 ±173% +300.0% 0.02 sched_debug.rt_rq:/.rt_nr_running.stddev
51.11 ± 3% -23.9 27.16 ± 2% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_fastpath
50.92 ± 3% -23.9 27.03 ± 2% perf-profile.calltrace.cycles-pp.sys_brk.entry_SYSCALL_64_fastpath
45.62 ± 3% -21.1 24.49 ± 2% perf-profile.calltrace.cycles-pp.do_munmap.sys_brk.entry_SYSCALL_64_fastpath
45.21 ± 3% -20.9 24.31 ± 2% perf-profile.calltrace.cycles-pp.unmap_region.do_munmap.sys_brk.entry_SYSCALL_64_fastpath
17.64 +0.2 17.88 perf-profile.calltrace.cycles-pp.pagevec_lru_move_fn.lru_add_drain_cpu.lru_add_drain.unmap_region.do_munmap
17.12 +0.2 17.36 perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.pagevec_lru_move_fn.lru_add_drain_cpu.lru_add_drain.unmap_region
17.66 +0.2 17.91 perf-profile.calltrace.cycles-pp.lru_add_drain.unmap_region.do_munmap.sys_brk.entry_SYSCALL_64_fastpath
17.66 +0.2 17.91 perf-profile.calltrace.cycles-pp.lru_add_drain_cpu.lru_add_drain.unmap_region.do_munmap.sys_brk
17.04 +0.3 17.30 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.pagevec_lru_move_fn.lru_add_drain_cpu.lru_add_drain
28.18 +0.6 28.75 perf-profile.calltrace.cycles-pp.page_fault
37.65 +0.6 38.23 perf-profile.calltrace.cycles-pp.tlb_finish_mmu.unmap_region.do_munmap.sys_brk.entry_SYSCALL_64_fastpath
37.64 +0.6 38.22 perf-profile.calltrace.cycles-pp.arch_tlb_finish_mmu.tlb_finish_mmu.unmap_region.do_munmap.sys_brk
28.01 +0.6 28.59 perf-profile.calltrace.cycles-pp.__do_page_fault.do_page_fault.page_fault
28.03 +0.6 28.62 perf-profile.calltrace.cycles-pp.do_page_fault.page_fault
25.72 +0.8 26.52 perf-profile.calltrace.cycles-pp.handle_mm_fault.__do_page_fault.do_page_fault.page_fault
35.98 +0.8 36.81 perf-profile.calltrace.cycles-pp.tlb_flush_mmu_free.arch_tlb_finish_mmu.tlb_finish_mmu.unmap_region.do_munmap
25.37 +0.8 26.20 perf-profile.calltrace.cycles-pp.__handle_mm_fault.handle_mm_fault.__do_page_fault.do_page_fault.page_fault
35.79 +0.9 36.66 perf-profile.calltrace.cycles-pp.release_pages.tlb_flush_mmu_free.arch_tlb_finish_mmu.tlb_finish_mmu.unmap_region
33.30 +1.1 34.42 perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.release_pages.tlb_flush_mmu_free.arch_tlb_finish_mmu.tlb_finish_mmu
33.15 +1.1 34.29 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.release_pages.tlb_flush_mmu_free.arch_tlb_finish_mmu
16.68 +1.4 18.08 perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.pagevec_lru_move_fn.__lru_cache_add.__handle_mm_fault.handle_mm_fault
16.61 +1.4 18.01 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.pagevec_lru_move_fn.__lru_cache_add.__handle_mm_fault
18.13 +1.6 19.71 perf-profile.calltrace.cycles-pp.__lru_cache_add.__handle_mm_fault.handle_mm_fault.__do_page_fault.do_page_fault
17.97 +1.6 19.56 perf-profile.calltrace.cycles-pp.pagevec_lru_move_fn.__lru_cache_add.__handle_mm_fault.handle_mm_fault.__do_page_fault
11.40 ± 14% +21.5 32.89 perf-profile.calltrace.cycles-pp.unmap_region.do_munmap.sys_brk.entry_SYSCALL_64_fastpath.brk
11.52 ± 14% +21.6 33.15 perf-profile.calltrace.cycles-pp.do_munmap.sys_brk.entry_SYSCALL_64_fastpath.brk
12.88 ± 14% +23.8 36.64 perf-profile.calltrace.cycles-pp.sys_brk.entry_SYSCALL_64_fastpath.brk
12.90 ± 14% +23.8 36.70 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_fastpath.brk
13.42 ± 14% +24.6 38.03 perf-profile.calltrace.cycles-pp.brk
5.49 -0.5 4.98 perf-profile.children.cycles-pp.do_brk_flags
64.09 -0.1 63.95 perf-profile.children.cycles-pp.entry_SYSCALL_64_fastpath
63.91 -0.1 63.77 perf-profile.children.cycles-pp.sys_brk
17.71 +0.2 17.95 perf-profile.children.cycles-pp.lru_add_drain
17.70 +0.2 17.95 perf-profile.children.cycles-pp.lru_add_drain_cpu
57.15 +0.5 57.67 perf-profile.children.cycles-pp.do_munmap
28.20 +0.6 28.76 perf-profile.children.cycles-pp.page_fault
28.15 +0.6 28.72 perf-profile.children.cycles-pp.__do_page_fault
37.69 +0.6 38.27 perf-profile.children.cycles-pp.arch_tlb_finish_mmu
37.70 +0.6 38.28 perf-profile.children.cycles-pp.tlb_finish_mmu
28.12 +0.6 28.70 perf-profile.children.cycles-pp.do_page_fault
56.63 +0.6 57.22 perf-profile.children.cycles-pp.unmap_region
25.79 +0.8 26.58 perf-profile.children.cycles-pp.handle_mm_fault
36.17 +0.8 36.99 perf-profile.children.cycles-pp.release_pages
25.46 +0.8 26.29 perf-profile.children.cycles-pp.__handle_mm_fault
36.03 +0.8 36.86 perf-profile.children.cycles-pp.tlb_flush_mmu_free
18.15 +1.6 19.72 perf-profile.children.cycles-pp.__lru_cache_add
35.66 +1.8 37.50 perf-profile.children.cycles-pp.pagevec_lru_move_fn
67.18 +2.8 69.94 perf-profile.children.cycles-pp._raw_spin_lock_irqsave
66.91 +2.8 69.72 perf-profile.children.cycles-pp.native_queued_spin_lock_slowpath
13.42 ± 14% +24.6 38.03 perf-profile.children.cycles-pp.brk
66.91 +2.8 69.72 perf-profile.self.cycles-pp.native_queued_spin_lock_slowpath
reaim.child_systime
1080 +-+------------------------------------------------------------------+
| O O O O O O O O O O O O |
1060 +-+ O O |
1040 O-O O O O O O |
| |
1020 +-+ |
| |
1000 +-+ |
| |
980 +-+ |
960 +-+ |
| |
940 +-+.+.+.+. .+. .+. .+.+..+.+.+.+.+.+.+.+.+.+. |
| + + +.+.+.+.+.+ +.+.+.+.+.+.+.|
920 +-+------------------------------------------------------------------+
reaim.jobs_per_min
350000 +-+----------------------------------------------------------------+
345000 +-+ .+. .+.+.+.+.+.+.+.|
|.+.+.+.+.+.+ +.+.+.+.+.+.+.+. +.+.+.+. .+.+.+.+ |
340000 +-+ +.+ + |
335000 +-+ |
| |
330000 +-+ |
325000 +-+ |
320000 +-+ |
| |
315000 +-+ |
310000 +-+ |
O O O O O O O O |
305000 +-+ O O O O O O O O O OO O O |
300000 +-+----------------------------------------------------------------+
reaim.jobs_per_min_child
350 +-+-------------------------------------------------------------------+
345 +-+ .+. .+.+.+.+.+.+.+.|
|.+.+.+.+.+.+ +..+.+.+.+.+.+.+. .+.+.+.+. .+.+.+..+ |
340 +-+ +.+ + |
335 +-+ |
| |
330 +-+ |
325 +-+ |
320 +-+ |
| |
315 +-+ |
310 +-+ |
O O O O O O O O |
305 +-+ O O O O O O O O O O O O O |
300 +-+-------------------------------------------------------------------+
reaim.max_jobs_per_min
360000 +-+----------------------------------------------------------------+
355000 +-+ + +. .+. |
| .+. + + + +.+.+ +.+.|
350000 +-+.+.+.+ + +.+.+.+.+.+.+. .++.+.+.+.+.+.+.+.+ |
345000 +-+ +.+ |
340000 +-+ |
335000 +-+ |
| |
330000 +-+ |
325000 +-+ |
320000 +-+ |
315000 +-+ |
O O O O O O |
310000 +-+ O O O O O O O O O O O O O O |
305000 +-+--------------------------------O-------------------------------+
1.4e+08 +-+---------------------------------------------------------------+
| O |
1.2e+08 +-+ |
| |
1e+08 +-+ O O O |
| O O |
8e+07 +-+ O O O |
| O O OO O |
6e+07 O-+ O O O O |
| O |
4e+07 +-+ O |
| |
2e+07 +-+ |
| .+. .+. .+. .+ |
0 +-+---------------------------------------------------------------+
1.6e+08 +-+---------------------------------------------------------------+
| O |
1.4e+08 +-+ |
1.2e+08 +-+ |
| O O O O O |
1e+08 +-+ O O O |
| O O O O O |
8e+07 O-+ O O O O |
| O |
6e+07 +-+ O |
4e+07 +-+ |
| |
2e+07 +-+ |
| .+. .+. .+. .+ |
0 +-+---------------------------------------------------------------+
perf-stat.instructions
1.8e+13 +-+--------------------------------------------------------------+
|.+.+.+.+.++. .+.+.+.+.+.+.+ ++.+.+.+.+.+.+.+.+. .+ .+.+.+.+.|
1.78e+13 +-+ + + + |
| |
1.76e+13 +-+ |
| |
1.74e+13 +-+ |
| |
1.72e+13 +-+ |
| |
1.7e+13 +-+ O O |
| O O O O O O O O OO O O |
1.68e+13 O-O O O O OO |
| |
1.66e+13 +-+--------------------------------------------------------------+
perf-stat.branch-instructions
3.95e+12 +-+--------------------------------------------------------------+
|.+.+ +.++.+.+.+. .+. + ++.+ +.+.+.+ +. .+ .+.+. .|
3.9e+12 +-+ +.+.+ + + +.+ + |
| |
| |
3.85e+12 +-+ |
| |
3.8e+12 +-+ |
| |
3.75e+12 +-+ |
| O O O O O O O O OO O O |
| O O |
3.7e+12 O-O O O O OO |
| |
3.65e+12 +-+--------------------------------------------------------------+
perf-stat.dTLB-loads
4.85e+12 +-+--------------------------------------------------------------+
| +.+ |
4.8e+12 +-+ +.+. .+. + : |
4.75e+12 +-+ : +.+ + : |
|.+.+.+.+. +. : :+.+.+.+. .+.+.+.+. |
4.7e+12 +-+ + +.+ + + +.++.+.+.+.+.|
4.65e+12 +-+ |
| |
4.6e+12 +-+ |
4.55e+12 +-+ |
| |
4.5e+12 +-+ |
4.45e+12 O-O O O OO |
| O O O O O O O O O O O OO O O |
4.4e+12 +-+--------------------------------------------------------------+
perf-stat.dTLB-stores
2.5e+12 +-+--------------------------------------------------------------+
| +.+.+.+.+.+.+.+ |
2.45e+12 +-+ + + |
2.4e+12 +-+.+.+.+.++.+.+ ++.+.+.+.+.+.+.+.+.+.++.+.+.+.+.|
| |
2.35e+12 +-+ |
2.3e+12 +-+ |
| |
2.25e+12 +-+ |
2.2e+12 +-+ |
O O O O O OO |
2.15e+12 +-+ |
2.1e+12 +-+ O O O O O O O O O O OO O O |
| |
2.05e+12 +-+--------------------------------------------------------------+
perf-stat.iTLB-loads
5.8e+09 +-+---------------------------------------------------------------+
| |
5.6e+09 +-+ +.+.+.+.+.+.+.+ |
| : : |
| : : |
5.4e+09 +-+ : : |
| : : |
5.2e+09 +-+ : : |
| +.+.+.+ : |
5e+09 +-+ + : .+.+.+.|
|.+.+.+.+ +.+.+.+.+.+.+.+.++.+.+.+.+ |
| |
4.8e+09 +-+ O OO O O O O O O O |
O O O O O O O O |
4.6e+09 +-O---O-O---------------------------------------------------------+
perf-stat.page-faults
2.6e+09 +-+--------------------------------------------------------------+
|.+.+.+.+.++.+.+.+.+.+.+.+.+.+.+.++.+.+.+.+.+.+.+.+.+.++.+.+.+.+.|
2.55e+09 +-+ |
2.5e+09 +-+ |
| |
2.45e+09 +-+ |
| |
2.4e+09 +-+ |
| |
2.35e+09 +-+ |
2.3e+09 +-+ |
| |
2.25e+09 +-+ |
O O O O O OO O O O O O O O O O O OO O O |
2.2e+09 +-+--------------------------------------------------------------+
perf-stat.minor-faults
2.6e+09 +-+--------------------------------------------------------------+
|.+.+.+.+.++.+.+.+.+.+.+.+.+.+.+.++.+.+.+.+.+.+.+.+.+.++.+.+.+.+.|
2.55e+09 +-+ |
2.5e+09 +-+ |
| |
2.45e+09 +-+ |
| |
2.4e+09 +-+ |
| |
2.35e+09 +-+ |
2.3e+09 +-+ |
| |
2.25e+09 +-+ |
O O O O O OO O O O O O O O O O O OO O O |
2.2e+09 +-+--------------------------------------------------------------+
perf-stat.ipc
0.44 +-+-----------------------------------------------------------------+
| +. |
0.435 +-+ .+. + +.+.+.+.+.+.|
| .+.+.+ +.+.+.+.+.+.+.+.+. .+. .+ |
|.+.+.+ +.+.+.+ +.+.+.+ |
0.43 +-+ |
| |
0.425 +-+ |
| |
0.42 +-+ |
| |
| |
0.415 O-O O O O O O O O |
| O O O O O O O O O O O O |
0.41 +-+-----------------------------------------------------------------+
perf-stat.cpi
2.44 +-+------------------------------------------------------------------+
| O O O O O O O O O O O O |
2.42 O-O O O O O O O |
2.4 +-+ O |
| |
2.38 +-+ |
| |
2.36 +-+ |
| |
2.34 +-+ |
2.32 +-+ |
|.+.+.+.+. .+. .+..+.+.+.+.+.+.+.+.+.+ |
2.3 +-+ + +.+.+.+.+.+.+.+.+ + .+. .|
| +.+.+ +.+.+ |
2.28 +-+------------------------------------------------------------------+
reaim.time.percent_of_cpu_this_job_got
5050 +-+------------------------------------------------------------------+
| O O O O O |
5040 O-O O O O O O O O O O O O |
5030 +-+ O O O |
| |
5020 +-+ |
5010 +-+ |
| |
5000 +-+ |
4990 +-+ |
| |
4980 +-+.+.+. .+. .+. .+. .+.+. .+. .+.+. |
4970 +-+ +.+ +.+.+.+.+.+.+.+ +. + + + +. .+.+.+.+.+.+.|
| + |
4960 +-+------------------------------------------------------------------+
reaim.time.elapsed_time
318 +-+-------------------------------------------------------------------+
| |
316 +-+.+. +.+.+.+.+. .+.+.+. |
314 +-+ +.+. .+ + +.+ +..+ |
| + + .+..+.+.+.+.+.+ : |
312 +-+ + : .+.+.+.+.+.|
| +.+ |
310 +-+ |
| |
308 +-+ |
306 +-+ O O O O O O O O O O O O |
| O O |
304 O-O O |
| O O O O |
302 +-+-------------------------------------------------------------------+
reaim.time.elapsed_time.max
318 +-+-------------------------------------------------------------------+
| |
316 +-+.+. +.+.+.+.+. .+.+.+. |
314 +-+ +.+. .+ + +.+ +..+ |
| + + .+..+.+.+.+.+.+ : |
312 +-+ + : .+.+.+.+.+.|
| +.+ |
310 +-+ |
| |
308 +-+ |
306 +-+ O O O O O O O O O O O O |
| O O |
304 O-O O |
| O O O O |
302 +-+-------------------------------------------------------------------+
reaim.time.minor_page_faults
2.6e+09 +-+--------------------------------------------------------------+
|.+.+.+.+.++.+.+.+.+.+.+.+.+.+.+.++.+.+.+.+.+.+.+.+.+.++.+.+.+.+.|
2.55e+09 +-+ |
2.5e+09 +-+ |
| |
2.45e+09 +-+ |
| |
2.4e+09 +-+ |
| |
2.35e+09 +-+ |
2.3e+09 +-+ |
| |
2.25e+09 +-+ |
O O O O O OO O O O O O O O O O O OO O O |
2.2e+09 +-+--------------------------------------------------------------+
reaim.time.voluntary_context_switches
280000 +-+---------------------------O--------O---------------------------+
260000 +-+ O O O O O |
O O O O O O O O O O |
240000 +-+ O O O O |
220000 +-+ |
| |
200000 +-+ |
180000 +-+ |
160000 +-+ |
|.+.+. .+. .+.++.+. .+.+.+.+. .+ |
140000 +-+ + +.+.+.+.+ + + + : |
120000 +-+ + .+.+.+. + : +.+ +. .+.|
| + + :+ + + + |
100000 +-+ + + |
80000 +-+----------------------------------------------------------------+
[*] bisect-good sample
[O] bisect-bad sample
Disclaimer:
Results have been estimated based on internal Intel analysis and are provided
for informational purposes only. Any difference in system hardware or software
design or configuration may affect actual performance.
Thanks,
Xiaolong
View attachment "config-4.15.0-rc1-00083-gae938e2" of type "text/plain" (164439 bytes)
View attachment "job-script" of type "text/plain" (6787 bytes)
View attachment "job.yaml" of type "text/plain" (4493 bytes)
View attachment "reproduce" of type "text/plain" (1956 bytes)
Powered by blists - more mailing lists