A collection of four financial benchmarks. Looks like the first two are multi-threaded on half the cores and take the majority of the time.
Topdown profile shows backend stalls as high but also a reasonable retirement rate.
AMD metrics show low frontend stalls. This is floating point code with little L2 access. Most of the backend stalls are CPU-bound.
elapsed 1035.282
on_cpu 0.355 # 5.68 / 16 cores
utime 5785.973
stime 96.720
nvcsw 4591 # 22.08%
nivcsw 16206 # 77.92%
inblock 37784 # 36.50/sec
onblock 35664 # 34.45/sec
cpu-clock 5882831894422 # 5882.832 seconds
task-clock 5882851684544 # 5882.852 seconds
page faults 65455628 # 11126.513/sec
context switches 25671 # 4.364/sec
cpu migrations 5959 # 1.013/sec
major page faults 517 # 0.088/sec
minor page faults 65455111 # 11126.426/sec
alignment faults 0 # 0.000/sec
emulation faults 0 # 0.000/sec
branches 8328733911169 # 128.024 branches per 1000 inst
branch misses 14693691423 # 0.18% branch miss
conditional 5603552654609 # 86.134 conditional branches per 1000 inst
indirect 497478035210 # 7.647 indirect branches per 1000 inst
cpu-cycles 7488943530330 # 1.31 GHz
instructions 19758434866181 # 2.64 IPC
slots 14977782375228 #
retiring 6814177121877 # 45.5% (45.5%)
-- ucode 1365372062 # 0.0%
-- fastpath 6812811749815 # 45.5%
frontend 326471508892 # 2.2% ( 2.2%) low
-- latency 254726563758 # 1.7%
-- bandwidth 71744945134 # 0.5%
backend 7681737582299 # 51.3% (51.3%)
-- cpu 5100495206851 # 34.1%
-- memory 2581242375448 # 17.2%
speculation 149543048615 # 1.0% ( 1.0%) low
-- branch mispredict 124883703377 # 0.8%
-- pipeline restart 24659345238 # 0.2%
smt-contention 5848798338 # 0.0% ( 0.0%)
cpu-cycles 7482994801959 # 1.30 GHz
instructions 19738225800564 # 2.64 IPC
instructions 6588909045206 # 1.178 l2 access per 1000 inst
l2 hit from l1 5322755575 # 10.68% l2 miss
l2 miss from l1 184539829 #
l2 hit from l2 pf 1793129410 #
l3 hit from l2 pf 34443363 #
l3 miss from l2 pf 609523840 #
instructions 6579793505573 # 222.719 float per 1000 inst
float 512 83 # 0.000 AVX-512 per 1000 inst
float 256 946 # 0.000 AVX-256 per 1000 inst
float 128 1465443386798 # 222.719 AVX-128 per 1000 inst
float MMX 0 # 0.000 MMX per 1000 inst
float scalar 0 # 0.000 scalar per 1000 inst
instructions 19748403504221 #
opcache 2905533867944 # 147.128 opcache per 1000 inst
opcache miss 41922370639 # 1.4% opcache miss rate
l1 dTLB miss 234043952 # 0.012 L1 dTLB per 1000 inst
l2 dTLB miss 76014301 # 0.004 L2 dTLB per 1000 inst
instructions 19748634163037 #
icache 76406830087 # 3.869 icache per 1000 inst
icache miss 6662379260 # 8.7% icache miss rate
l1 iTLB miss 27224549 # 0.001 L1 iTLB per 1000 inst
l2 iTLB miss 0 # 0.000 L2 iTLB per 1000 inst
tlb flush 49125 # 0.000 TLB flush per 1000 inst
Intel metrics also show cpu-bound stalls for the backend.
elapsed 1179.262
on_cpu 0.303 # 4.84 / 16 cores
utime 5637.148
stime 73.328
nvcsw 58509 # 80.24%
nivcsw 14404 # 19.76%
inblock 2304912 # 1954.54/sec
onblock 2008 # 1.70/sec
cpu-clock 5710573919435 # 5710.574 seconds
task-clock 5710599096564 # 5710.599 seconds
page faults 68354687 # 11969.793/sec
context switches 78524 # 13.751/sec
cpu migrations 3880 # 0.679/sec
major page faults 55675 # 9.749/sec
minor page faults 68299012 # 11960.043/sec
alignment faults 0 # 0.000/sec
emulation faults 0 # 0.000/sec
branches 6289632555846 # 122.671 branches per 1000 inst
branch misses 4917247843 # 0.08% branch miss
conditional 6289632579782 # 122.671 conditional branches per 1000 inst
indirect 1077480290780 # 21.015 indirect branches per 1000 inst
slots 43206819481880 #
retiring 20365363088824 # 47.1% (47.1%)
-- ucode 1830818416962 # 4.2%
-- fastpath 18534544671862 # 42.9%
frontend 2632266679747 # 6.1% ( 6.1%)
-- latency 1354271430326 # 3.1%
-- bandwidth 1277995249421 # 3.0%
backend 19558375224339 # 45.3% (45.3%)
-- cpu 16552070779783 # 38.3%
-- memory 3006304444556 # 7.0%
speculation 576520337778 # 1.3% ( 1.3%)
-- branch mispredict 392102827742 # 0.9%
-- pipeline restart 184417510036 # 0.4%
smt-contention 0 # 0.0% ( 0.0%)
cpu-cycles 8481833511738 # 1.09 GHz
instructions 23656551048100 # 2.79 IPC
l2 access 9587069567 # 0.665 l2 access per 1000 inst
l2 miss 4576505692 # 47.74% l2 miss
cpu-cycles 5156699832810 # 10.2% memory latency
load stalls 511213801969 # 9.7% l1 bound
l1 miss 12522557086 # 0.1% l2 bound
l2 miss 4800858625 # 0.0% l3 bound
l3 miss 3349985261 # 0.1% dram bound
store_stalls 15141260509 # 0.3% store bound
Process overview shows mostly the bondEngine and repoEngine applications as taking the most time.
608 processes
24 bondsEngine.exe 8107.84 102.24
24 repoEngine.exe 5418.32 135.04
201 clinfo 54.56 21.08
9 monteCarloEngin 35.74 0.36
9 blackScholesAna 4.11 1.58
38 vulkaninfo 1.14 1.71
4 vulkani:disk$0 0.12 0.18
6 glxinfo:gdrv0 0.12 0.10
6 glxinfo:gl0 0.12 0.10
6 php 0.08 0.14
2 llvmpipe-0 0.06 0.09
2 llvmpipe-1 0.06 0.09
2 llvmpipe-10 0.06 0.09
2 llvmpipe-11 0.06 0.09
2 llvmpipe-12 0.06 0.09
2 llvmpipe-13 0.06 0.09
2 llvmpipe-14 0.06 0.09
2 llvmpipe-15 0.06 0.09
2 llvmpipe-2 0.06 0.09
2 llvmpipe-3 0.06 0.09
2 llvmpipe-4 0.06 0.09
2 llvmpipe-5 0.06 0.09
2 llvmpipe-6 0.06 0.09
2 llvmpipe-7 0.06 0.09
2 llvmpipe-8 0.06 0.09
2 llvmpipe-9 0.06 0.09
2 glxinfo 0.06 0.04
2 glxinfo:cs0 0.06 0.04
2 glxinfo:disk$0 0.06 0.04
2 glxinfo:sh0 0.06 0.04
2 glxinfo:shlo0 0.06 0.04
6 clang 0.04 0.06
3 rocminfo 0.00 0.03
1 lspci 0.00 0.03
92 sh 0.00 0.00
24 financebench 0.00 0.00
13 dirname 0.00 0.00
13 gcc 0.00 0.00
12 basename 0.00 0.00
9 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
4 gmain 0.00 0.00
3 dconf worker 0.00 0.00
2 cc 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 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 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
Computation blocks
950098) financebench cpu=14 start=6.49 finish=42.64
950099) dirname cpu=4 start=6.49 finish=6.49
950100) financebench cpu=15 start=6.49 finish=6.49
950101) repoEngine.exe cpu=2 start=6.49 finish=42.64
950102) basename cpu=4 start=6.49 finish=6.49
950105) repoEngine.exe cpu=11 start=13.65 finish=42.64
950106) repoEngine.exe cpu=4 start=13.65 finish=42.64
950107) repoEngine.exe cpu=13 start=13.65 finish=42.64
950108) repoEngine.exe cpu=6 start=13.65 finish=42.64
950109) repoEngine.exe cpu=15 start=13.65 finish=42.64
950110) repoEngine.exe cpu=1 start=13.65 finish=42.64
950111) repoEngine.exe cpu=0 start=13.65 finish=42.64