A simple raytracer test. A steady profile with a low amount of front-end misses and a higher retirement rate.
AMD metrics show a high amount of floating point code.
elapsed 193.229
on_cpu 0.920 # 14.72 / 16 cores
utime 2843.215
stime 1.199
nvcsw 2880 # 0.40%
nivcsw 712240 # 99.60%
inblock 120 # 0.62/sec
onblock 146936 # 760.42/sec
cpu-clock 2844460816089 # 2844.461 seconds
task-clock 2844483262894 # 2844.483 seconds
page faults 104687 # 36.804/sec
context switches 715888 # 251.676/sec
cpu migrations 3072 # 1.080/sec
major page faults 6 # 0.002/sec
minor page faults 104681 # 36.801/sec
alignment faults 0 # 0.000/sec
emulation faults 0 # 0.000/sec
branches 1079231734603 # 44.256 branches per 1000 inst
branch misses 3273560627 # 0.30% branch miss
conditional 1063385010947 # 43.606 conditional branches per 1000 inst
indirect 3067754574 # 0.126 indirect branches per 1000 inst
cpu-cycles 12321875960673 # 4.02 GHz
instructions 24380866354642 # 1.98 IPC
slots 24668946439356 #
retiring 8148298916767 # 33.0% (55.8%)
-- ucode 52074297 # 0.0%
-- fastpath 8148246842470 # 33.0%
frontend 498872246150 # 2.0% ( 3.4%)
-- latency 57738316554 # 0.2%
-- bandwidth 441133929596 # 1.8%
backend 5915123949013 # 24.0% (40.5%)
-- cpu 5834112788473 # 23.6%
-- memory 81011160540 # 0.3%
speculation 42870363857 # 0.2% ( 0.3%)
-- branch mispredict 42751660155 # 0.2%
-- pipeline restart 118703702 # 0.0%
smt-contention 10063754939603 # 40.8% ( 0.0%)
cpu-cycles 12315039672751 # 4.02 GHz
instructions 24380032728302 # 1.98 IPC
instructions 8129644074387 # 0.262 l2 access per 1000 inst
l2 hit from l1 1973886535 # 1.48% l2 miss
l2 miss from l1 21874035 #
l2 hit from l2 pf 143089209 #
l3 hit from l2 pf 6875755 #
l3 miss from l2 pf 2809825 #
instructions 8129337578296 # 324.015 float per 1000 inst
float 512 86 # 0.000 AVX-512 per 1000 inst
float 256 244 # 0.000 AVX-256 per 1000 inst
float 128 2634028430835 # 324.015 AVX-128 per 1000 inst
float MMX 0 # 0.000 MMX per 1000 inst
float scalar 0 # 0.000 scalar per 1000 inst
Intel metrics show a small code that barely makes L2 accesses.
elapsed 432.621
on_cpu 0.931 # 14.90 / 16 cores
utime 6447.098
stime 0.678
nvcsw 5005 # 0.31%
nivcsw 1590351 # 99.69%
inblock 680 # 1.57/sec
onblock 293032 # 677.34/sec
cpu-clock 6447770583077 # 6447.771 seconds
task-clock 6447786296295 # 6447.786 seconds
page faults 136778 # 21.213/sec
context switches 1597343 # 247.735/sec
cpu migrations 8986 # 1.394/sec
major page faults 40 # 0.006/sec
minor page faults 136738 # 21.207/sec
alignment faults 0 # 0.000/sec
emulation faults 0 # 0.000/sec
branches 2157151553481 # 44.232 branches per 1000 inst
branch misses 8271519627 # 0.38% branch miss
conditional 2157151612777 # 44.232 conditional branches per 1000 inst
indirect 714873005231 # 14.658 indirect branches per 1000 inst
slots 16079412619640 #
retiring 11921841202251 # 74.1% (74.1%)
-- ucode 2582238943 # 0.0%
-- fastpath 11919258963308 # 74.1%
frontend 3508733895595 # 21.8% (21.8%)
-- latency 2568996546907 # 16.0%
-- bandwidth 939737348688 # 5.8%
backend 469157121710 # 2.9% ( 2.9%)
-- cpu 396760153670 # 2.5%
-- memory 72396968040 # 0.5%
speculation 174707587812 # 1.1% ( 1.1%)
-- branch mispredict 173535377060 # 1.1%
-- pipeline restart 1172210752 # 0.0%
smt-contention 0 # 0.0% ( 0.0%)
cpu-cycles 6724054057954 # 1.91 GHz
instructions 15297307644937 # 2.28 IPC
l2 access 794762889 # 0.065 l2 access per 1000 inst
l2 miss 84294824 # 10.61% l2 missProcess profile shows time in c-ray-mt
1053 processes
771 c-ray-mt 683368.56 64.02
38 vulkaninfo 0.76 0.96
4 vulkani:disk$0 0.08 0.11
6 php 0.06 0.08
6 clang 0.05 0.02
2 llvmpipe-0 0.04 0.06
2 llvmpipe-1 0.04 0.06
2 llvmpipe-10 0.04 0.06
2 llvmpipe-11 0.04 0.06
2 llvmpipe-12 0.04 0.06
2 llvmpipe-13 0.04 0.06
2 llvmpipe-14 0.04 0.06
2 llvmpipe-15 0.04 0.06
2 llvmpipe-3 0.04 0.06
2 llvmpipe-4 0.04 0.06
2 llvmpipe-5 0.04 0.06
2 llvmpipe-6 0.04 0.06
2 llvmpipe-7 0.04 0.06
2 llvmpipe-8 0.04 0.06
2 llvmpipe-9 0.04 0.06
2 llvmpipe-2 0.04 0.05
1 lspci 0.00 0.03
91 sh 0.00 0.00
13 gcc 0.00 0.00
9 stty 0.00 0.00
8 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 glxinfo 0.00 0.00
5 gmain 0.00 0.00
5 phoronix-test-s 0.00 0.00
3 c-ray 0.00 0.00
3 dconf worker 0.00 0.00
2 cc 0.00 0.00
2 clinfo 0.00 0.00
2 grep 0.00 0.00
2 lscpu 0.00 0.00
2 setterm 0.00 0.00
2 uname 0.00 0.00
2 which 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 ifconfig 0.00 0.00
1 ip 0.00 0.00
1 lsmod 0.00 0.00
1 mktemp 0.00 0.00
1 ps 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 systemctl 0.00 0.00
1 template.sh 0.00 0.00
1 wc 0.00 0.00
0 processes running
267 maximum processes
Looks like many ~256 processes all launched simulataneously.