onednn is a neural network library, in this test convolution. Except for Stream, this is the most memory-bound application I observe. There are a few data points of front-end misses but overall frontend is low.
AMD metrics show only 50% on cpu. Not much floating point (despite this being listed as f32 and u8/i8 test. Very low IPC and high L2 misses.
elapsed 71.340
on_cpu 0.508 # 8.13 / 16 cores
utime 577.606
stime 2.283
nvcsw 2933 # 21.82%
nivcsw 10507 # 78.18%
inblock 7824 # 109.67/sec
onblock 1808 # 25.34/sec
cpu-clock 580267448241 # 580.267 seconds
task-clock 580293281735 # 580.293 seconds
page faults 917435 # 1580.985/sec
context switches 13603 # 23.442/sec
cpu migrations 355 # 0.612/sec
major page faults 101 # 0.174/sec
minor page faults 917334 # 1580.811/sec
alignment faults 0 # 0.000/sec
emulation faults 0 # 0.000/sec
branches 14973262484 # 27.616 branches per 1000 inst
branch misses 247963050 # 1.66% branch miss
conditional 12771157805 # 23.555 conditional branches per 1000 inst
indirect 148342180 # 0.274 indirect branches per 1000 inst
cpu-cycles 2780469655011 # 2.44 GHz
instructions 543789492995 # 0.20 IPC
slots 5553310978572 #
retiring 189704440465 # 3.4% ( 3.6%)
-- ucode 965399781 # 0.0%
-- fastpath 188739040684 # 3.4%
frontend 200601330078 # 3.6% ( 3.8%)
-- latency 172522010298 # 3.1%
-- bandwidth 28079319780 # 0.5%
backend 4904067730753 # 88.3% (92.6%)
-- cpu 843972890978 # 15.2%
-- memory 4060094839775 # 73.1%
speculation 1460554868 # 0.0% ( 0.0%)
-- branch mispredict 1317644471 # 0.0%
-- pipeline restart 142910397 # 0.0%
smt-contention 257472348837 # 4.6% ( 0.0%)
cpu-cycles 2780748578640 # 2.29 GHz
instructions 542814219642 # 0.20 IPC
instructions 181076126057 # 50.266 l2 access per 1000 inst
l2 hit from l1 6471351931 # 44.79% l2 miss
l2 miss from l1 2012992425 #
l2 hit from l2 pf 566893824 #
l3 hit from l2 pf 538088144 #
l3 miss from l2 pf 1525658477 #
instructions 181285666295 # 4.541 float per 1000 inst
float 512 62 # 0.000 AVX-512 per 1000 inst
float 256 314 # 0.000 AVX-256 per 1000 inst
float 128 823292867 # 4.541 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 also confirm this is a short-running workload.
elapsed 70.776
on_cpu 0.515 # 8.24 / 16 cores
utime 581.506
stime 1.509
nvcsw 2728 # 29.18%
nivcsw 6621 # 70.82%
inblock 35816 # 506.05/sec
onblock 1888 # 26.68/sec
cpu-clock 583151242836 # 583.151 seconds
task-clock 583163697663 # 583.164 seconds
page faults 913034 # 1565.656/sec
context switches 9520 # 16.325/sec
cpu migrations 476 # 0.816/sec
major page faults 197 # 0.338/sec
minor page faults 912837 # 1565.319/sec
alignment faults 0 # 0.000/sec
emulation faults 0 # 0.000/sec
branches 24742585163 # 27.518 branches per 1000 inst
branch misses 55106113 # 0.22% branch miss
conditional 24742600267 # 27.518 conditional branches per 1000 inst
indirect 8415132396 # 9.359 indirect branches per 1000 inst
slots 3324114012128 #
retiring 466162249004 # 14.0% (14.0%)
-- ucode 10119389453 # 0.3%
-- fastpath 456042859551 # 13.7%
frontend 180764613017 # 5.4% ( 5.4%)
-- latency 156300290055 # 4.7%
-- bandwidth 24464322962 # 0.7%
backend 2669530026167 # 80.3% (80.3%)
-- cpu 436293792636 # 13.1%
-- memory 2233236233531 # 67.2%
speculation 5291484831 # 0.2% ( 0.2%)
-- branch mispredict 3942250602 # 0.1%
-- pipeline restart 1349234229 # 0.0%
smt-contention 0 # 0.0% ( 0.0%)
cpu-cycles 1100899965042 # 0.97 GHz
instructions 454783623331 # 0.41 IPC
l2 access 10699740373 # 23.529 l2 access per 1000 inst
l2 miss 6241519885 # 58.33% l2 miss
Process shows time spent in benchdnn
454 processes
96 benchdnn 9231.20 22.88
64 clinfo 10.88 3.96
38 vulkaninfo 0.76 1.14
6 glxinfo:gdrv0 0.11 0.09
4 vulkani:disk$0 0.08 0.12
6 php 0.06 0.09
6 clang 0.05 0.03
2 glxinfo 0.05 0.03
2 glxinfo:cs0 0.05 0.03
2 glxinfo:disk$0 0.05 0.03
2 glxinfo:sh0 0.05 0.03
2 glxinfo:shlo0 0.05 0.03
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-2 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
1 lspci 0.01 0.03
90 sh 0.00 0.00
13 gcc 0.00 0.00
10 gsettings 0.00 0.00
9 stty 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 onednn 0.00 0.00
5 phoronix-test-s 0.00 0.00
4 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 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
1 xrandr 0.00 0.00
1 processes running
48 maximum processes
Computation core starts benchmark threads on all cpus.
454702) onednn cpu=0 start=15.52 finish=21.76
454703) benchdnn cpu=8 start=15.53 finish=21.75
454704) benchdnn cpu=0 start=15.56 finish=21.75
454705) benchdnn cpu=9 start=15.56 finish=21.75
454706) benchdnn cpu=1 start=15.56 finish=21.75
454707) benchdnn cpu=10 start=15.56 finish=21.75
454708) benchdnn cpu=2 start=15.56 finish=21.75
454709) benchdnn cpu=11 start=15.56 finish=21.75
454710) benchdnn cpu=3 start=15.56 finish=21.75
454711) benchdnn cpu=12 start=15.56 finish=21.75
454712) benchdnn cpu=4 start=15.56 finish=21.75
454713) benchdnn cpu=13 start=15.56 finish=21.75
454714) benchdnn cpu=5 start=15.56 finish=21.75
454715) benchdnn cpu=14 start=15.56 finish=21.75
454716) benchdnn cpu=6 start=15.56 finish=21.75
454717) benchdnn cpu=15 start=15.56 finish=21.75
454718) benchdnn cpu=7 start=15.56 finish=21.75