mnn is a neural network framework. This test tries eight different models and creates a geographic mean. The process overview suggests consistently spawning threads on all cores.
Topdown overview suggests certain phases of backend activity or frontend activity and a consistent retirement rate.
AMD metrics
elapsed 350.487
on_cpu 0.942 # 15.07 / 16 cores
utime 3132.874
stime 2147.667
nvcsw 2652 # 5.72%
nivcsw 43705 # 94.28%
inblock 8 # 0.02/sec
onblock 12752 # 36.38/sec
cpu-clock 5280985842856 # 5280.986 seconds
task-clock 5281014226031 # 5281.014 seconds
page faults 1210146 # 229.150/sec
context switches 47938 # 9.077/sec
cpu migrations 286 # 0.054/sec
major page faults 2 # 0.000/sec
minor page faults 1210144 # 229.150/sec
alignment faults 0 # 0.000/sec
emulation faults 0 # 0.000/sec
branches 2031821929040 # 120.505 branches per 1000 inst
branch misses 253888931138 # 12.50% branch miss
conditional 970120823455 # 57.537 conditional branches per 1000 inst
indirect 20899090234 # 1.239 indirect branches per 1000 inst
cpu-cycles 110973875070917 # 3.88 GHz
instructions 86300390156541 # 0.78 IPC
slots 221938347605922 #
retiring 32263730366925 # 14.5% (17.3%)
-- ucode 168671411473 # 0.1%
-- fastpath 32095058955452 # 14.5%
frontend 71461075367125 # 32.2% (38.2%)
-- latency 59166100322928 # 26.7%
-- bandwidth 12294975044197 # 5.5%
backend 82836224092829 # 37.3% (44.3%)
-- cpu 36711151611261 # 16.5%
-- memory 46125072481568 # 20.8%
speculation 366713033149 # 0.2% ( 0.2%)
-- branch mispredict 362656308286 # 0.2%
-- pipeline restart 4056724863 # 0.0%
smt-contention 35010489172413 # 15.8% ( 0.0%)
cpu-cycles 111531376124456 # 3.86 GHz
instructions 86619805970825 # 0.78 IPC
instructions 28881985494088 # 73.009 l2 access per 1000 inst
l2 hit from l1 1518505686348 # 21.90% l2 miss
l2 miss from l1 198514190460 #
l2 hit from l2 pf 326870900178 #
l3 hit from l2 pf 224827579693 #
l3 miss from l2 pf 38451066229 #
instructions 28860405033995 # 19.572 float per 1000 inst
float 512 54 # 0.000 AVX-512 per 1000 inst
float 256 658 # 0.000 AVX-256 per 1000 inst
float 128 564863466718 # 19.572 AVX-128 per 1000 inst
float MMX 0 # 0.000 MMX per 1000 inst
float scalar 1 # 0.000 scalar per 1000 inst
Intel metrics
elapsed 2678.159
on_cpu 0.975 # 15.59 / 16 cores
utime 25134.666
stime 16629.070
nvcsw 4543 # 0.15%
nivcsw 3013644 # 99.85%
inblock 1368 # 0.51/sec
onblock 1880 # 0.70/sec
cpu-clock 41764931327447 # 41764.931 seconds
task-clock 41765074209486 # 41765.074 seconds
page faults 4760326 # 113.979/sec
context switches 3031380 # 72.582/sec
cpu migrations 516 # 0.012/sec
major page faults 1 # 0.000/sec
minor page faults 4760325 # 113.979/sec
alignment faults 0 # 0.000/sec
emulation faults 0 # 0.000/sec
branches 19697851320338 # 108.948 branches per 1000 inst
branch misses 11220044229 # 0.06% branch miss
conditional 19697851339346 # 108.948 conditional branches per 1000 inst
indirect 5502668904687 # 30.435 indirect branches per 1000 inst
slots 46191558811640 #
retiring 26812914980896 # 58.0% (58.0%)
-- ucode 3186273037326 # 6.9%
-- fastpath 23626641943570 # 51.1%
frontend 12488108841259 # 27.0% (27.0%)
-- latency 7466005891595 # 16.2%
-- bandwidth 5022102949664 # 10.9%
backend 6407985278786 # 13.9% (13.9%)
-- cpu 2050555342763 # 4.4%
-- memory 4357429936023 # 9.4%
speculation 385626559145 # 0.8% ( 0.8%)
-- branch mispredict 263567987217 # 0.6%
-- pipeline restart 122058571928 # 0.3%
smt-contention 0 # 0.0% ( 0.0%)
cpu-cycles 70900703783603 # 1.63 GHz
instructions 116275737814803 # 1.64 IPC
l2 access 2953834438475 # 28.835 l2 access per 1000 inst
l2 miss 650131037736 # 22.01% l2 miss
Process overview shows the application is named benchmark.out
597 processes
240 benchmark.out 255571.35 185286.88
68 clinfo 18.83 7.01
38 vulkaninfo 1.32 1.33
6 glxinfo:gdrv0 0.22 0.04
6 php 0.15 0.25
4 vulkani:disk$0 0.14 0.14
2 glxinfo 0.10 0.02
2 glxinfo:cs0 0.10 0.02
2 glxinfo:disk$0 0.10 0.02
2 glxinfo:sh0 0.10 0.02
2 glxinfo:shlo0 0.10 0.02
2 llvmpipe-0 0.07 0.07
2 llvmpipe-1 0.07 0.07
2 llvmpipe-10 0.07 0.07
2 llvmpipe-11 0.07 0.07
2 llvmpipe-12 0.07 0.07
2 llvmpipe-13 0.07 0.07
2 llvmpipe-14 0.07 0.07
2 llvmpipe-15 0.07 0.07
2 llvmpipe-2 0.07 0.07
2 llvmpipe-3 0.07 0.07
2 llvmpipe-4 0.07 0.07
2 llvmpipe-5 0.07 0.07
2 llvmpipe-6 0.07 0.07
2 llvmpipe-7 0.07 0.07
2 llvmpipe-8 0.07 0.07
2 llvmpipe-9 0.07 0.07
6 clang 0.06 0.06
3 rocminfo 0.01 0.03
1 lspci 0.00 0.02
1 ps 0.00 0.01
82 sh 0.00 0.00
15 mnn 0.00 0.00
13 gcc 0.00 0.00
11 gsettings 0.00 0.00
8 stat 0.00 0.00
8 systemd-detect- 0.00 0.00
6 llvm-link 0.00 0.00
5 phoronix-test-s 0.00 0.00
3 gmain 0.00 0.00
2 cc 0.00 0.00
2 dconf worker 0.00 0.00
2 lscpu 0.00 0.00
2 uname 0.00 0.00
2 which 0.00 0.00
2 xset 0.00 0.00
1 date 0.00 0.00
1 dirname 0.00 0.00
1 dmesg 0.00 0.00
1 dmidecode 0.00 0.00
1 grep 0.00 0.00
1 ifconfig 0.00 0.00
1 ip 0.00 0.00
1 lsmod 0.00 0.00
1 mktemp 0.00 0.00
1 qdbus 0.00 0.00
1 readlink 0.00 0.00
1 realpath 0.00 0.00
1 sed 0.00 0.00
1 sort 0.00 0.00
1 stty 0.00 0.00
1 systemctl 0.00 0.00
1 template.sh 0.00 0.00
1 wc 0.00 0.00
1 xrandr 0.00 0.00
0 processes running
47 maximum processes
Core computation blocks
12168) mnn cpu=2 start=6.75 finish=128.36
12169) benchmark.out cpu=12 start=6.75 finish=128.35
12170) benchmark.out cpu=13 start=6.90 finish=128.35
12171) benchmark.out cpu=5 start=6.90 finish=128.35
12172) benchmark.out cpu=0 start=6.90 finish=128.35
12173) benchmark.out cpu=9 start=6.90 finish=128.35
12174) benchmark.out cpu=11 start=6.90 finish=128.35
12175) benchmark.out cpu=7 start=6.90 finish=128.35
12176) benchmark.out cpu=3 start=6.90 finish=128.35
12177) benchmark.out cpu=14 start=6.90 finish=128.35
12178) benchmark.out cpu=8 start=6.90 finish=128.35
12179) benchmark.out cpu=1 start=6.90 finish=128.35
12180) benchmark.out cpu=15 start=6.90 finish=128.35
12181) benchmark.out cpu=10 start=6.90 finish=128.35
12182) benchmark.out cpu=4 start=6.90 finish=128.35
12183) benchmark.out cpu=2 start=6.90 finish=128.35
12184) benchmark.out cpu=14 start=6.90 finish=128.35