Testing an interprocess communication primitives. There are 24 different subtests including
- TCP socket with different sized messages
- Pipes with different sized messages
- FIFO pipes with different sized messages
- UNIX domain sockets with different sized messages
Topdown summary shows mostly two processes running but not spending much time and also a moderate level of interrupts. For example the initial section is TCP sockets with increasing message sizes.
Topdown profile shows frontend time as the largest portion.
AMD metrics show little time is spent on the CPU, approximately 1/2 of one core. This is not floating point code. There is a high opcache miss rate.
elapsed 1020.621
on_cpu 0.032 # 0.51 / 16 cores
utime 44.062
stime 474.548
nvcsw 289753 # 99.13%
nivcsw 2532 # 0.87%
inblock 48 # 0.05/sec
onblock 16584 # 16.25/sec
cpu-clock 1001082661388 # 1001.083 seconds
task-clock 1002538150096 # 1002.538 seconds
page faults 196593 # 196.095/sec
context switches 7091548 # 7073.594/sec
cpu migrations 518 # 0.517/sec
major page faults 2 # 0.002/sec
minor page faults 196591 # 196.093/sec
alignment faults 0 # 0.000/sec
emulation faults 0 # 0.000/sec
branches 741624850927 # 203.483 branches per 1000 inst
branch misses 96937340773 # 13.07% branch miss
conditional 335512707016 # 92.056 conditional branches per 1000 inst
indirect 8794435030 # 2.413 indirect branches per 1000 inst
cpu-cycles 3938011140824 # 0.27 GHz
instructions 3335477922804 # 0.85 IPC
slots 7849808157270 #
retiring 1395394811656 # 17.8% (17.8%)
-- ucode 10805555686 # 0.1%
-- fastpath 1384589255970 # 17.6%
frontend 5233357070230 # 66.7% (66.7%) high
-- latency 4401810815286 # 56.1%
-- bandwidth 831546254944 # 10.6%
backend 1170512139286 # 14.9% (14.9%) low
-- cpu 308721940644 # 3.9%
-- memory 861790198642 # 11.0%
speculation 49986128184 # 0.6% ( 0.6%) low
-- branch mispredict 49821138730 # 0.6%
-- pipeline restart 164989454 # 0.0%
smt-contention 556982258 # 0.0% ( 0.0%)
cpu-cycles 5573850138090 # 0.29 GHz
instructions 4746679838611 # 0.85 IPC
instructions 1578587483573 # 45.278 l2 access per 1000 inst
l2 hit from l1 49877915552 # 26.93% l2 miss
l2 miss from l1 11601954748 #
l2 hit from l2 pf 13953015145 #
l3 hit from l2 pf 6995462328 #
l3 miss from l2 pf 649138092 #
instructions 1578938514540 # 11.084 float per 1000 inst
float 512 180 # 0.000 AVX-512 per 1000 inst
float 256 604 # 0.000 AVX-256 per 1000 inst
float 128 17500402570 # 11.084 AVX-128 per 1000 inst
float MMX 0 # 0.000 MMX per 1000 inst
float scalar 0 # 0.000 scalar per 1000 inst
instructions 3387121496957 #
opcache 1290539350027 # 381.014 opcache per 1000 inst
opcache miss 694249272052 # 53.8% opcache miss rate
l1 dTLB miss 571148055 # 0.169 L1 dTLB per 1000 inst
l2 dTLB miss 24800609 # 0.007 L2 dTLB per 1000 inst
instructions 3652008286708 #
icache 1811405175336 # 496.002 icache per 1000 inst
icache miss 33218736875 # 1.8% icache miss rate
l1 iTLB miss 8661876 # 0.002 L1 iTLB per 1000 inst
l2 iTLB miss 0 # 0.000 L2 iTLB per 1000 inst
tlb flush 27911 # 0.000 TLB flush per 1000 inst
Intel metrics show L3 as taking the largest amount of memory and surprisingly a lot more backend bound than AMD.
elapsed 1411.002
on_cpu 0.030 # 0.48 / 16 cores
utime 46.769
stime 630.786
nvcsw 521847 # 99.48%
nivcsw 2734 # 0.52%
inblock 536 # 0.38/sec
onblock 6192 # 4.39/sec
cpu-clock 1351750378635 # 1351.750 seconds
task-clock 1351993088667 # 1351.993 seconds
page faults 211379 # 156.346/sec
context switches 2150650 # 1590.726/sec
cpu migrations 1780 # 1.317/sec
major page faults 5 # 0.004/sec
minor page faults 211374 # 156.343/sec
alignment faults 0 # 0.000/sec
emulation faults 0 # 0.000/sec
branches 854112014822 # 164.960 branches per 1000 inst
branch misses 2537777092 # 0.30% branch miss
conditional 854112041478 # 164.960 conditional branches per 1000 inst
indirect 13809268745 # 2.667 indirect branches per 1000 inst
slots 31318484719550 #
retiring 7356339156164 # 23.5% (23.5%)
-- ucode 1450912164180 # 4.6%
-- fastpath 5905426991984 # 18.9%
frontend 5359358888516 # 17.1% (17.1%)
-- latency 2698193464442 # 8.6%
-- bandwidth 2661165424074 # 8.5%
backend 18030808250048 # 57.6% (57.6%)
-- cpu 6672071938072 # 21.3%
-- memory 11358736311976 # 36.3%
speculation 587508667689 # 1.9% ( 1.9%)
-- branch mispredict 445657118471 # 1.4%
-- pipeline restart 141851549218 # 0.5%
smt-contention 0 # 0.0% ( 0.0%)
cpu-cycles 6457245889546 # 0.23 GHz
instructions 7275449365505 # 1.13 IPC
l2 access 143093585864 # 19.677 l2 access per 1000 inst
l2 miss 85516676526 # 59.76% l2 miss
cpu-cycles 6129827611475 # 40.2% memory latency
load stalls 2378116952695 # 7.6% l1 bound
l1 miss 1910108431750 # 7.2% l2 bound
l2 miss 1468707911024 # 23.9% l3 bound
l3 miss 1043305120 # 0.0% dram bound
store_stalls 84585107510 # 1.4% store bound
Process summary shows different programs for processes
688 processes
64 pipe 27.56 274.22
60 fifo 26.79 263.40
36 tcp 20.19 245.40
36 socketpair 19.53 326.57
68 clinfo 17.53 5.31
38 vulkaninfo 0.96 1.53
6 php 0.18 0.56
4 vulkani:disk$0 0.10 0.17
6 clang 0.07 0.05
6 glxinfo:gdrv0 0.05 0.13
6 glxinfo:gl0 0.05 0.13
2 llvmpipe-0 0.05 0.09
2 llvmpipe-1 0.05 0.09
2 llvmpipe-10 0.05 0.09
2 llvmpipe-11 0.05 0.09
2 llvmpipe-12 0.05 0.09
2 llvmpipe-13 0.05 0.09
2 llvmpipe-14 0.05 0.09
2 llvmpipe-2 0.05 0.09
2 llvmpipe-3 0.05 0.09
2 llvmpipe-4 0.05 0.09
2 llvmpipe-5 0.05 0.09
2 llvmpipe-6 0.05 0.09
2 llvmpipe-7 0.05 0.09
2 llvmpipe-9 0.05 0.09
2 llvmpipe-15 0.05 0.08
2 llvmpipe-8 0.05 0.08
2 glxinfo 0.03 0.05
2 glxinfo:cs0 0.03 0.05
2 glxinfo:disk$0 0.03 0.05
2 glxinfo:sh0 0.03 0.05
2 glxinfo:shlo0 0.03 0.05
3 rocminfo 0.03 0.00
1 lspci 0.00 0.02
128 sh 0.00 0.00
98 ipc-benchmark 0.00 0.00
13 gcc 0.00 0.00
10 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
3 dconf worker 0.00 0.00
3 gmain 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 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 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
An example computation block
829619) ipc-benchmark cpu=4 start=26.57 finish=32.98
829620) tcp cpu=5 start=26.57 finish=32.98
829621) tcp cpu=14 start=26.57 finish=32.98