Eight different built-in benchmarks for opencv operations. Looks like a mixture of single-threaded and multi-core tests. Some of them take some time to stabilize so get run more than three times.
Topdown shows a mixture of profiles depending on the test.
AMD metrics reflect mixture of on-core behavior. Generally floating point code with a small amount of L2 access and moderate branches.
elapsed 4652.674
on_cpu 0.387 # 6.20 / 16 cores
utime 22478.001
stime 6365.386
nvcsw 4855478 # 73.46%
nivcsw 1753786 # 26.54%
inblock 8 # 0.00/sec
onblock 117920 # 25.34/sec
cpu-clock 28815667699825 # 28815.668 seconds
task-clock 28821382684850 # 28821.383 seconds
page faults 307497285 # 10669.068/sec
context switches 6632219 # 230.115/sec
cpu migrations 37169 # 1.290/sec
major page faults 294 # 0.010/sec
minor page faults 307496991 # 10669.058/sec
alignment faults 0 # 0.000/sec
emulation faults 0 # 0.000/sec
branches 16403069524758 # 106.832 branches per 1000 inst
branch misses 828455317656 # 5.05% branch miss
conditional 11923539266664 # 77.657 conditional branches per 1000 inst
indirect 422732816500 # 2.753 indirect branches per 1000 inst
cpu-cycles 134335559445133 # 1.71 GHz
instructions 177139632100759 # 1.32 IPC
slots 268758248314950 #
retiring 61647781137976 # 22.9% (29.1%)
-- ucode 124149216346 # 0.0%
-- fastpath 61523631921630 # 22.9%
frontend 54787629853380 # 20.4% (25.9%)
-- latency 41293315615422 # 15.4%
-- bandwidth 13494314237958 # 5.0%
backend 91911217413741 # 34.2% (43.4%)
-- cpu 34999936609993 # 13.0%
-- memory 56911280803748 # 21.2%
speculation 3311643758552 # 1.2% ( 1.6%)
-- branch mispredict 3291586145649 # 1.2%
-- pipeline restart 20057612903 # 0.0%
smt-contention 57099414604412 # 21.2% ( 0.0%)
cpu-cycles 91565832999745 # 1.60 GHz
instructions 125015233884237 # 1.37 IPC
instructions 41670344100784 # 34.699 l2 access per 1000 inst
l2 hit from l1 886460797566 # 20.52% l2 miss
l2 miss from l1 64299878440 #
l2 hit from l2 pf 327036437902 #
l3 hit from l2 pf 175255051987 #
l3 miss from l2 pf 57158618370 #
instructions 41599072769052 # 265.827 float per 1000 inst
float 512 93 # 0.000 AVX-512 per 1000 inst
float 256 4009590064 # 0.096 AVX-256 per 1000 inst
float 128 11054155455194 # 265.731 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
elapsed 8929.748
on_cpu 0.302 # 4.83 / 16 cores
utime 35037.226
stime 8114.568
nvcsw 16144279 # 65.79%
nivcsw 8396621 # 34.21%
inblock 92520 # 10.36/sec
onblock 84432 # 9.46/sec
cpu-clock 42918505907879 # 42918.506 seconds
task-clock 42933876988368 # 42933.877 seconds
page faults 789930954 # 18398.780/sec
context switches 24585261 # 572.631/sec
cpu migrations 714157 # 16.634/sec
major page faults 810 # 0.019/sec
minor page faults 789930144 # 18398.761/sec
alignment faults 0 # 0.000/sec
emulation faults 0 # 0.000/sec
branches 26769265019370 # 123.191 branches per 1000 inst
branch misses 171790003420 # 0.64% branch miss
conditional 26769265226570 # 123.191 conditional branches per 1000 inst
indirect 6811153187756 # 31.345 indirect branches per 1000 inst
slots 316190860328552 #
retiring 152939542929502 # 48.4% (48.4%)
-- ucode 15947069922576 # 5.0%
-- fastpath 136992473006926 # 43.3%
frontend 41215234315610 # 13.0% (13.0%)
-- latency 23077500486912 # 7.3%
-- bandwidth 18137733828698 # 5.7%
backend 105108579100965 # 33.2% (33.2%)
-- cpu 48161090633115 # 15.2%
-- memory 56947488467850 # 18.0%
speculation 20264503084589 # 6.4% ( 6.4%)
-- branch mispredict 19084866553039 # 6.0%
-- pipeline restart 1179636531550 # 0.4%
smt-contention 0 # 0.0% ( 0.0%)
cpu-cycles 169947932831273 # 0.98 GHz
instructions 307478565821240 # 1.81 IPC
l2 access 4221508301051 # 24.988 l2 access per 1000 inst
l2 miss 1732467749070 # 41.04% l2 miss
Process summary, this crashed partway through so not a full reflection. From the names, looks like a different executable for each test.
5843 processes
1200 opencv_perf_gap 1427508.34 348263.04
3840 opencv_perf_vid 348468.47 88338.60
48 opencv_perf_img 67596.32 17327.20
240 opencv_perf_obj 27428.01 12202.04
48 opencv_perf_dnn 18839.04 1405.28
48 opencv_perf_fea 18809.28 9131.84
48 opencv_perf_sti 14596.96 8618.88
64 opencv_perf_cor 8730.40 6199.36
34 clinfo 9.59 3.66
19 vulkaninfo 0.95 0.38
2 vulkani:disk$0 0.10 0.04
6 clang 0.09 0.03
3 glxinfo:gdrv0 0.07 0.06
1 llvmpipe-0 0.05 0.02
1 llvmpipe-1 0.05 0.02
1 llvmpipe-10 0.05 0.02
1 llvmpipe-11 0.05 0.02
1 llvmpipe-12 0.05 0.02
1 llvmpipe-13 0.05 0.02
1 llvmpipe-14 0.05 0.02
1 llvmpipe-15 0.05 0.02
1 llvmpipe-2 0.05 0.02
1 llvmpipe-3 0.05 0.02
1 llvmpipe-4 0.05 0.02
1 llvmpipe-5 0.05 0.02
1 llvmpipe-6 0.05 0.02
1 llvmpipe-7 0.05 0.02
1 llvmpipe-8 0.05 0.02
1 llvmpipe-9 0.05 0.02
1 glxinfo 0.03 0.02
1 glxinfo:cs0 0.03 0.02
1 glxinfo:disk$0 0.03 0.02
1 glxinfo:sh0 0.03 0.02
1 glxinfo:shlo0 0.03 0.02
1 ps 0.00 0.01
72 sh 0.00 0.00
49 opencv 0.00 0.00
13 gcc 0.00 0.00
8 systemd-detect- 0.00 0.00
7 stat 0.00 0.00
6 gsettings 0.00 0.00
6 llvm-link 0.00 0.00
5 gmain 0.00 0.00
4 dconf worker 0.00 0.00
4 phoronix-test-s 0.00 0.00
2 which 0.00 0.00
1 cc 0.00 0.00
1 date 0.00 0.00
1 dirname 0.00 0.00
1 grep 0.00 0.00
1 ifconfig 0.00 0.00
1 ip 0.00 0.00
1 lscpu 0.00 0.00
1 mktemp 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
1 xset 0.00 0.00
26 processes running
47 maximum processes
A typical launch pattern with one process per core.
1945253) opencv cpu=0 start=142.15 finish=208.66
1945254) opencv_perf_cor cpu=9 start=142.15 finish=208.65
1945255) opencv_perf_cor cpu=11 start=142.17 finish=208.65
1945256) opencv_perf_cor cpu=12 start=142.17 finish=208.65
1945257) opencv_perf_cor cpu=4 start=142.17 finish=208.65
1945258) opencv_perf_cor cpu=5 start=142.17 finish=208.65
1945259) opencv_perf_cor cpu=8 start=142.17 finish=208.65
1945260) opencv_perf_cor cpu=14 start=142.17 finish=208.65
1945261) opencv_perf_cor cpu=10 start=142.17 finish=208.65
1945262) opencv_perf_cor cpu=0 start=142.17 finish=208.65
1945263) opencv_perf_cor cpu=13 start=142.17 finish=208.65
1945264) opencv_perf_cor cpu=6 start=142.17 finish=208.65
1945265) opencv_perf_cor cpu=15 start=142.17 finish=208.65
1945266) opencv_perf_cor cpu=7 start=142.17 finish=208.65
1945267) opencv_perf_cor cpu=2 start=142.17 finish=208.65
1945268) opencv_perf_cor cpu=3 start=142.17 finish=208.65
1945269) opencv_perf_cor cpu=1 start=142.17 finish=208.65