nab is a SPEC CPU(R) benchmark described here and written in C. The workload runs on all logical cores.
Topdown profile shows medium level backend-bound stalls with a ~35% retirement rate. There is period at end with different characteristic.
AMD metrics on 7840 processor show more CPU stalls than memory stalls. Overall memory is only 50 L2 access per 1000 instructions and ~5% miss rate.
elapsed 581.066
on_cpu 0.970 # 15.53 / 16 cores
utime 9013.279
stime 8.575
nvcsw 13853 # 13.67%
nivcsw 87522 # 86.33%
inblock 0 # 0.00/sec
onblock 13480 # 23.20/sec
cpu-clock 9022371301663 # 9022.371 seconds
task-clock 9022438159718 # 9022.438 seconds
page faults 2719530 # 301.419/sec
context switches 100813 # 11.174/sec
cpu migrations 151 # 0.017/sec
major page faults 761 # 0.084/sec
minor page faults 2718769 # 301.334/sec
alignment faults 0 # 0.000/sec
emulation faults 0 # 0.000/sec
branches 3705536285749 # 83.289 branches per 1000 inst
branch misses 48320502361 # 1.30% branch miss
conditional 3211079075725 # 72.175 conditional branches per 1000 inst
indirect 78933292050 # 1.774 indirect branches per 1000 inst
cpu-cycles 35173554892555 # 3.77 GHz
instructions 44502941930878 # 1.27 IPC
slots 70335260112720 #
retiring 17957481268465 # 25.5% (36.9%)
-- ucode 49152961542 # 0.1%
-- fastpath 17908328306923 # 25.5%
frontend 4243956072338 # 6.0% ( 8.7%)
-- latency 3490353506760 # 5.0%
-- bandwidth 753602565578 # 1.1%
backend 25363097670219 # 36.1% (52.1%)
-- cpu 16028170538565 # 22.8%
-- memory 9334927131654 # 13.3%
speculation 1119054942785 # 1.6% ( 2.3%)
-- branch mispredict 1084377989807 # 1.5%
-- pipeline restart 34676952978 # 0.0%
smt-contention 21651631104051 # 30.8% ( 0.0%)
cpu-cycles 35197471408100 # 3.77 GHz
instructions 44501705968789 # 1.26 IPC
instructions 14829986452623 # 52.578 l2 access per 1000 inst
l2 hit from l1 563992892841 # 4.94% l2 miss
l2 miss from l1 11308778556 #
l2 hit from l2 pf 188511013533 #
l3 hit from l2 pf 10720306269 #
l3 miss from l2 pf 16503721139 #
instructions 14826609038424 # 318.530 float per 1000 inst
float 512 249 # 0.000 AVX-512 per 1000 inst
float 256 24571061449 # 1.657 AVX-256 per 1000 inst
float 128 4698147760247 # 316.873 AVX-128 per 1000 inst
float MMX 0 # 0.000 MMX per 1000 inst
float scalar 0 # 0.000 scalar per 1000 inst
instructions 44489194256996 #
opcache 4756323710148 # 106.910 opcache per 1000 inst
opcache miss 66379189344 # 1.4% opcache miss rate
l1 dTLB miss 41837658678 # 0.940 L1 dTLB per 1000 inst
l2 dTLB miss 1090929870 # 0.025 L2 dTLB per 1000 inst
instructions 44489204211481 #
icache 80661600392 # 1.813 icache per 1000 inst
icache miss 12168515583 # 15.1% icache miss rate
l1 iTLB miss 173686412 # 0.004 L1 iTLB per 1000 inst
l2 iTLB miss 0 # 0.000 L2 iTLB per 1000 inst
tlb flush 182039 # 0.000 TLB flush per 1000 inst
Process overview shows time spent in nab_r_base.mev-
581 processes
48 nab_r_base.mev- 9011.77 4.93
69 specperl 8.59 1.55
1 clang 0.01 0.00
11 ps 0.00 0.02
1 lsb_release 0.00 0.01
173 sh 0.00 0.00
54 specrxp 0.00 0.00
48 bash 0.00 0.00
41 specinvoke 0.00 0.00
21 grep 0.00 0.00
20 cat 0.00 0.00
12 uniq 0.00 0.00
11 sort 0.00 0.00
10 expand 0.00 0.00
6 pwd 0.00 0.00
5 basename 0.00 0.00
5 specmake 0.00 0.00
5 systemctl 0.00 0.00
4 specpp 0.00 0.00
4 uname 0.00 0.00
3 dirname 0.00 0.00
3 dmidecode 0.00 0.00
3 lscpu 0.00 0.00
2 df 0.00 0.00
2 dpkg 0.00 0.00
2 rm 0.00 0.00
2 runcpu 0.00 0.00
2 specsha512sum 0.00 0.00
2 specxz 0.00 0.00
2 who 0.00 0.00
1 cpupower 0.00 0.00
1 head 0.00 0.00
1 logname 0.00 0.00
1 ls 0.00 0.00
1 numactl 0.00 0.00
1 sysctl 0.00 0.00
1 w 0.00 0.00
1 wc 0.00 0.00
1 which 0.00 0.00
0 processes running
53 maximum processes
specinvoke starts separate processes on each logical core.
452134) specinvoke cpu=1 start=3.26 finish=195.57
452136) sh cpu=0 start=3.26 finish=191.47
452143) bash cpu=0 start=3.26 finish=191.47
452167) nab_r_base.mev- cpu=0 start=3.26 finish=191.46
452137) sh cpu=10 start=3.26 finish=191.13
452144) bash cpu=1 start=3.26 finish=191.13
452165) nab_r_base.mev- cpu=1 start=3.26 finish=191.12
452138) sh cpu=10 start=3.26 finish=191.07
452147) bash cpu=2 start=3.26 finish=191.07
452170) nab_r_base.mev- cpu=2 start=3.27 finish=191.06
452139) sh cpu=9 start=3.26 finish=191.56
452150) bash cpu=3 start=3.26 finish=191.56
452174) nab_r_base.mev- cpu=3 start=3.27 finish=191.54
452140) sh cpu=9 start=3.26 finish=195.48
452154) bash cpu=4 start=3.26 finish=195.48
452173) nab_r_base.mev- cpu=4 start=3.27 finish=195.47
452141) sh cpu=5 start=3.26 finish=189.13
452152) bash cpu=5 start=3.26 finish=189.13
452172) nab_r_base.mev- cpu=5 start=3.27 finish=189.11
452142) sh cpu=7 start=3.26 finish=191.44
452149) bash cpu=6 start=3.26 finish=191.44
452171) nab_r_base.mev- cpu=6 start=3.27 finish=191.43
452145) sh cpu=11 start=3.26 finish=191.33
452156) bash cpu=7 start=3.26 finish=191.33
452175) nab_r_base.mev- cpu=7 start=3.27 finish=191.31
452146) sh cpu=2 start=3.26 finish=191.26
452158) bash cpu=8 start=3.26 finish=191.26
452177) nab_r_base.mev- cpu=8 start=3.27 finish=191.25
452148) sh cpu=10 start=3.26 finish=191.28
452160) bash cpu=9 start=3.26 finish=191.28
452179) nab_r_base.mev- cpu=9 start=3.27 finish=191.27
452151) sh cpu=5 start=3.26 finish=190.86
452162) bash cpu=10 start=3.26 finish=190.86
452178) nab_r_base.mev- cpu=10 start=3.27 finish=190.84
452153) sh cpu=8 start=3.26 finish=191.28
452163) bash cpu=11 start=3.26 finish=191.28
452176) nab_r_base.mev- cpu=11 start=3.27 finish=191.26
452155) sh cpu=11 start=3.26 finish=195.57
452164) bash cpu=12 start=3.26 finish=195.57
452180) nab_r_base.mev- cpu=12 start=3.27 finish=195.56
452157) sh cpu=5 start=3.26 finish=189.76
452166) bash cpu=13 start=3.26 finish=189.76
452181) nab_r_base.mev- cpu=13 start=3.27 finish=189.75
452159) sh cpu=13 start=3.26 finish=191.16
452168) bash cpu=14 start=3.26 finish=191.16
452182) nab_r_base.mev- cpu=14 start=3.27 finish=191.14
452161) sh cpu=11 start=3.26 finish=191.66
452169) bash cpu=15 start=3.27 finish=191.66
452183) nab_r_base.mev- cpu=15 start=3.27 finish=191.65