y-cruncher is a program that calculates many digits of Pi. My running seems to hang. Initially it hung on both AMD and Intel, but at different places. The Intel version hung at 1B digits and the AMD at 5B digits. The y-cruncher page also reports a hang subsequently fixed but their description of symptoms is different from what I see. In any case, I only have AMD metrics here.
AMD metrics show this is very much a backend bound application with both CPU and memory contributing, Almost no frontend stalls. There is a small amount of floating point code and a reasonable L2 rate with some misses.
elapsed 679.121
on_cpu 0.920 # 14.72 / 16 cores
utime 9967.348
stime 31.191
nvcsw 360008 # 42.57%
nivcsw 485694 # 57.43%
inblock 1280448 # 1885.45/sec
onblock 29298368 # 43141.61/sec
cpu-clock 10001201726708 # 10001.202 seconds
task-clock 10001591896343 # 10001.592 seconds
page faults 184372 # 18.434/sec
context switches 848898 # 84.876/sec
cpu migrations 171430 # 17.140/sec
major page faults 7089 # 0.709/sec
minor page faults 177282 # 17.725/sec
alignment faults 0 # 0.000/sec
emulation faults 0 # 0.000/sec
branches 760807352038 # 22.662 branches per 1000 inst
branch misses 4713539070 # 0.62% branch miss
conditional 587069670473 # 17.487 conditional branches per 1000 inst
indirect 9994645887 # 0.298 indirect branches per 1000 inst
cpu-cycles 40453340253892 # 3.71 GHz
instructions 33575761816393 # 0.83 IPC
slots 80902865789910 #
retiring 11902956182674 # 14.7% (17.2%)
-- ucode 4585744763 # 0.0%
-- fastpath 11898370437911 # 14.7%
frontend 2041876738557 # 2.5% ( 3.0%)
-- latency 1659454106610 # 2.1%
-- bandwidth 382422631947 # 0.5%
backend 54990294148872 # 68.0% (79.6%)
-- cpu 24297716954409 # 30.0%
-- memory 30692577194463 # 37.9%
speculation 115427066513 # 0.1% ( 0.2%)
-- branch mispredict 72716495043 # 0.1%
-- pipeline restart 42710571470 # 0.1%
smt-contention 11852224998687 # 14.6% ( 0.0%)
cpu-cycles 40459239031625 # 3.71 GHz
instructions 33573330939113 # 0.83 IPC
instructions 11192891062647 # 106.431 l2 access per 1000 inst
l2 hit from l1 897172713903 # 11.75% l2 miss
l2 miss from l1 64301791449 #
l2 hit from l2 pf 218454217312 #
l3 hit from l2 pf 48863370442 #
l3 miss from l2 pf 26783988521 #
instructions 11187826293958 # 22.845 float per 1000 inst
float 512 60 # 0.000 AVX-512 per 1000 inst
float 256 96781510168 # 8.651 AVX-256 per 1000 inst
float 128 158800636747 # 14.194 AVX-128 per 1000 inst
float MMX 0 # 0.000 MMX per 1000 inst
float scalar 0 # 0.000 scalar per 1000 inst
Process statistics don’t seem to get much system or user time.
548 processes
64 clinfo 10.88 3.46
38 vulkaninfo 0.94 0.96
6 php 0.16 0.11
4 vulkani:disk$0 0.10 0.11
6 glxinfo:gdrv0 0.08 0.09
2 llvmpipe-0 0.05 0.06
2 llvmpipe-1 0.05 0.06
2 llvmpipe-10 0.05 0.06
2 llvmpipe-11 0.05 0.06
2 llvmpipe-12 0.05 0.06
2 llvmpipe-13 0.05 0.06
2 llvmpipe-14 0.05 0.06
2 llvmpipe-15 0.05 0.06
2 llvmpipe-2 0.05 0.06
2 llvmpipe-3 0.05 0.06
2 llvmpipe-4 0.05 0.06
2 llvmpipe-5 0.05 0.06
2 llvmpipe-6 0.05 0.06
2 llvmpipe-7 0.05 0.06
2 llvmpipe-8 0.05 0.06
2 llvmpipe-9 0.05 0.06
2 glxinfo 0.05 0.04
2 glxinfo:cs0 0.05 0.04
2 glxinfo:disk$0 0.05 0.04
2 glxinfo:shlo0 0.05 0.04
2 glxinfo:sh0 0.04 0.03
6 clang 0.03 0.04
1 lspci 0.01 0.02
189 22-ZN4 ~ Kizuna 0.00 0.00
91 sh 0.00 0.00
12 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
6 y-cruncher 0.00 0.00
5 gmain 0.00 0.00
5 phoronix-test-s 0.00 0.00
4 sed 0.00 0.00
3 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 cc 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 ps 0.00 0.00
1 qdbus 0.00 0.00
1 readlink 0.00 0.00
1 realpath 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
1 xrandr 0.00 0.00
0 processes running
76 maximum processesThe core computation starts many threads. One interesting thing is the program runs a binary named “ZN4” and the archive seems to ship with a static set of binaries. This suggests that perhaps these are already hardwired for Zen4 and other cores?
430509) y-cruncher cpu=13 start=5.87 finish=224.89
430510) y-cruncher cpu=15 start=5.88 finish=224.89
430512) sh cpu=11 start=5.88 finish=224.89
430513) 22-ZN4 ~ Kizuna cpu=0 start=5.88 finish=224.89
430514) 22-ZN4 ~ Kizuna cpu=0 start=5.88 finish=224.89
430515) 22-ZN4 ~ Kizuna cpu=0 start=5.88 finish=224.89
430516) 22-ZN4 ~ Kizuna cpu=0 start=5.88 finish=224.89
430517) 22-ZN4 ~ Kizuna cpu=0 start=5.88 finish=224.88
430518) 22-ZN4 ~ Kizuna cpu=0 start=5.88 finish=224.88
430519) 22-ZN4 ~ Kizuna cpu=0 start=5.88 finish=224.88
430520) 22-ZN4 ~ Kizuna cpu=0 start=5.88 finish=224.88
430521) 22-ZN4 ~ Kizuna cpu=0 start=5.88 finish=224.88
430522) 22-ZN4 ~ Kizuna cpu=0 start=5.88 finish=224.88
430523) 22-ZN4 ~ Kizuna cpu=0 start=5.88 finish=224.87
430524) 22-ZN4 ~ Kizuna cpu=0 start=5.88 finish=224.87
430525) 22-ZN4 ~ Kizuna cpu=0 start=5.88 finish=224.87
430526) 22-ZN4 ~ Kizuna cpu=0 start=5.88 finish=224.87
430527) 22-ZN4 ~ Kizuna cpu=0 start=5.88 finish=224.87
430528) 22-ZN4 ~ Kizuna cpu=0 start=5.88 finish=224.87
430529) 22-ZN4 ~ Kizuna cpu=0 start=5.88 finish=224.86
430530) 22-ZN4 ~ Kizuna cpu=0 start=5.88 finish=224.86
430531) 22-ZN4 ~ Kizuna cpu=0 start=5.88 finish=224.86
430532) 22-ZN4 ~ Kizuna cpu=0 start=5.88 finish=224.86
430533) 22-ZN4 ~ Kizuna cpu=0 start=5.88 finish=224.85
430534) 22-ZN4 ~ Kizuna cpu=0 start=5.88 finish=224.85
430535) 22-ZN4 ~ Kizuna cpu=0 start=5.88 finish=224.85
430536) 22-ZN4 ~ Kizuna cpu=0 start=5.88 finish=224.85
430537) 22-ZN4 ~ Kizuna cpu=0 start=5.88 finish=224.85
430538) 22-ZN4 ~ Kizuna cpu=0 start=5.88 finish=224.84
430539) 22-ZN4 ~ Kizuna cpu=0 start=5.88 finish=224.84
430540) 22-ZN4 ~ Kizuna cpu=0 start=5.88 finish=224.84
430541) 22-ZN4 ~ Kizuna cpu=0 start=5.88 finish=224.84
430542) 22-ZN4 ~ Kizuna cpu=0 start=5.88 finish=224.84
430543) 22-ZN4 ~ Kizuna cpu=0 start=5.88 finish=224.83
430544) 22-ZN4 ~ Kizuna cpu=0 start=5.88 finish=224.83
430545) 22-ZN4 ~ Kizuna cpu=0 start=6.06 finish=223.51
430546) 22-ZN4 ~ Kizuna cpu=0 start=6.06 finish=223.51
430547) 22-ZN4 ~ Kizuna cpu=0 start=6.06 finish=223.51
430548) 22-ZN4 ~ Kizuna cpu=0 start=6.06 finish=223.51
430549) 22-ZN4 ~ Kizuna cpu=0 start=6.06 finish=223.51
430550) 22-ZN4 ~ Kizuna cpu=0 start=6.06 finish=223.51
430551) 22-ZN4 ~ Kizuna cpu=0 start=6.06 finish=223.51
430552) 22-ZN4 ~ Kizuna cpu=0 start=6.06 finish=223.51
430553) 22-ZN4 ~ Kizuna cpu=0 start=6.06 finish=223.51
430554) 22-ZN4 ~ Kizuna cpu=0 start=6.06 finish=223.51
430555) 22-ZN4 ~ Kizuna cpu=0 start=6.06 finish=223.51
430556) 22-ZN4 ~ Kizuna cpu=0 start=6.06 finish=223.50
430557) 22-ZN4 ~ Kizuna cpu=0 start=6.06 finish=223.50
430558) 22-ZN4 ~ Kizuna cpu=0 start=6.06 finish=223.50
430559) 22-ZN4 ~ Kizuna cpu=0 start=6.06 finish=223.50
430560) 22-ZN4 ~ Kizuna cpu=0 start=6.06 finish=223.50
430561) 22-ZN4 ~ Kizuna cpu=0 start=6.06 finish=223.50
430562) 22-ZN4 ~ Kizuna cpu=0 start=6.06 finish=223.50
430563) 22-ZN4 ~ Kizuna cpu=0 start=6.06 finish=223.50
430564) 22-ZN4 ~ Kizuna cpu=0 start=6.06 finish=223.50
430565) 22-ZN4 ~ Kizuna cpu=0 start=6.06 finish=223.50
430566) 22-ZN4 ~ Kizuna cpu=0 start=6.06 finish=223.50
430567) 22-ZN4 ~ Kizuna cpu=0 start=6.06 finish=223.50
430568) 22-ZN4 ~ Kizuna cpu=0 start=6.06 finish=223.50
430569) 22-ZN4 ~ Kizuna cpu=0 start=6.06 finish=223.50
430570) 22-ZN4 ~ Kizuna cpu=0 start=6.06 finish=223.50
430571) 22-ZN4 ~ Kizuna cpu=0 start=6.06 finish=223.50
430572) 22-ZN4 ~ Kizuna cpu=0 start=6.06 finish=223.50
430573) 22-ZN4 ~ Kizuna cpu=0 start=6.06 finish=223.50
430574) 22-ZN4 ~ Kizuna cpu=0 start=6.06 finish=223.50
430575) 22-ZN4 ~ Kizuna cpu=0 start=6.06 finish=223.50
430511) sed cpu=6 start=5.88 finish=224.89