Quick running test of 15 glib functions, mostly math routines and also pthread creation. These run single-threaded. This benchmark is an outlier with Intel CPU reporting ~2X faster than AMD while almost every case has AMD faster and a median of AMD as 1.25x faster than my Intel CPU
Looks like these are single-threaded tests also not using the CPU much during that time.
Topdown profile shows across the map but generally frontend bound is highest.
AMD metrics confirm a low on-core amount of less than 0.25. This is floating point code with 1/6 of the instructions as branches.
elapsed 361.394
on_cpu 0.014 # 0.22 / 16 cores
utime 79.680
stime 1.329
nvcsw 2624 # 86.03%
nivcsw 426 # 13.97%
inblock 0 # 0.00/sec
onblock 590512 # 1633.98/sec
cpu-clock 81084003036 # 81.084 seconds
task-clock 81091329790 # 81.091 seconds
page faults 171284 # 2112.236/sec
context switches 4566 # 56.307/sec
cpu migrations 308 # 3.798/sec
major page faults 2 # 0.025/sec
minor page faults 171282 # 2112.211/sec
alignment faults 0 # 0.000/sec
emulation faults 0 # 0.000/sec
branches 143004053928 # 168.158 branches per 1000 inst
branch misses 147238834 # 0.10% branch miss
conditional 76874647961 # 90.397 conditional branches per 1000 inst
indirect 18182592456 # 21.381 indirect branches per 1000 inst
cpu-cycles 296276857093 # 0.05 GHz
instructions 835134997575 # 2.82 IPC
slots 595808733960 #
retiring 292824575828 # 49.1% (49.2%)
-- ucode 489790524 # 0.1%
-- fastpath 292334785304 # 49.1%
frontend 75678185011 # 12.7% (12.7%)
-- latency 33401593776 # 5.6%
-- bandwidth 42276591235 # 7.1%
backend 225177966311 # 37.8% (37.8%)
-- cpu 177289153913 # 29.8%
-- memory 47888812398 # 8.0%
speculation 1976038812 # 0.3% ( 0.3%) low
-- branch mispredict 1551074100 # 0.3%
-- pipeline restart 424964712 # 0.1%
smt-contention 151682832 # 0.0% ( 0.0%)
cpu-cycles 308438408523 # 0.05 GHz
instructions 873920649612 # 2.83 IPC
instructions 291987017136 # 0.711 l2 access per 1000 inst
l2 hit from l1 186290923 # 15.32% l2 miss
l2 miss from l1 20387235 #
l2 hit from l2 pf 9865651 #
l3 hit from l2 pf 4957233 #
l3 miss from l2 pf 6466175 #
instructions 291912058167 # 241.214 float per 1000 inst
float 512 99 # 0.000 AVX-512 per 1000 inst
float 256 686 # 0.000 AVX-256 per 1000 inst
float 128 70413333814 # 241.214 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 348.282
on_cpu 0.014 # 0.23 / 16 cores
utime 78.891
stime 0.979
nvcsw 2392 # 86.26%
nivcsw 381 # 13.74%
inblock 8 # 0.02/sec
onblock 579000 # 1662.44/sec
cpu-clock 79962687901 # 79.963 seconds
task-clock 79973477567 # 79.973 seconds
page faults 159208 # 1990.760/sec
context switches 4242 # 53.043/sec
cpu migrations 348 # 4.351/sec
major page faults 0 # 0.000/sec
minor page faults 159208 # 1990.760/sec
alignment faults 0 # 0.000/sec
emulation faults 0 # 0.000/sec
branches 184610022112 # 183.279 branches per 1000 inst
branch misses 48479139 # 0.03% branch miss
conditional 184610037632 # 183.279 conditional branches per 1000 inst
indirect 24923776264 # 24.744 indirect branches per 1000 inst
slots 1796340876272 #
retiring 1014873631013 # 56.5% (56.5%) high
-- ucode 80587003382 # 4.5%
-- fastpath 934286627631 # 52.0%
frontend 48346621350 # 2.7% ( 2.7%) low
-- latency 4806798326 # 0.3%
-- bandwidth 43539823024 # 2.4%
backend 727802271913 # 40.5% (40.5%)
-- cpu 547583708462 # 30.5%
-- memory 180218563451 # 10.0%
speculation 5601582618 # 0.3% ( 0.3%) low
-- branch mispredict 5416805254 # 0.3%
-- pipeline restart 184777364 # 0.0%
smt-contention 0 # 0.0% ( 0.0%)
cpu-cycles 356432878563 # 0.05 GHz
instructions 1213906374887 # 3.41 IPC high
l2 access 416472360 # 0.343 l2 access per 1000 inst
l2 miss 129752775 # 31.16% l2 miss
Process overview shows largest proportion in the linker, ld.so and glibc-bench having zero time.
470 processes
47 ld.so 76.06 0.00
68 clinfo 17.19 8.32
38 vulkaninfo 1.51 0.76
4 vulkani:disk$0 0.16 0.08
6 php 0.12 0.59
6 glxinfo:gdrv0 0.12 0.08
6 glxinfo:gl0 0.12 0.08
2 llvmpipe-0 0.08 0.04
2 llvmpipe-1 0.08 0.04
2 llvmpipe-10 0.08 0.04
2 llvmpipe-11 0.08 0.04
2 llvmpipe-12 0.08 0.04
2 llvmpipe-13 0.08 0.04
2 llvmpipe-14 0.08 0.04
2 llvmpipe-15 0.08 0.04
2 llvmpipe-2 0.08 0.04
2 llvmpipe-3 0.08 0.04
2 llvmpipe-4 0.08 0.04
2 llvmpipe-5 0.08 0.04
2 llvmpipe-6 0.08 0.04
2 llvmpipe-7 0.08 0.04
2 llvmpipe-8 0.08 0.04
2 llvmpipe-9 0.08 0.04
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 0.05 0.07
3 rocminfo 0.03 0.00
1 lspci 0.00 0.03
1 ps 0.00 0.01
110 sh 0.00 0.00
47 glibc-bench 0.00 0.00
13 gcc 0.00 0.00
11 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 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 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
Process overview shows the “ld.so” is the proxy for where the tests are running.
1103623) glibc-bench cpu=2 start=5.48 finish=6.62
1103624) ld.so cpu=3 start=5.48 finish=6.62
1103625) glibc-bench cpu=1 start=10.63 finish=11.77
1103626) ld.so cpu=10 start=10.63 finish=11.77
1103629) glibc-bench cpu=10 start=15.77 finish=16.91
1103630) ld.so cpu=11 start=15.77 finish=16.91
1103631) glibc-bench cpu=1 start=20.91 finish=22.06
1103632) ld.so cpu=2 start=20.92 finish=22.05
1103633) sh cpu=1 start=22.06 finish=22.06
1103634) sh cpu=11 start=22.06 finish=22.06
1103635) glibc-bench cpu=1 start=32.21 finish=36.35
1103636) ld.so cpu=10 start=32.22 finish=36.35
1103637) glibc-bench cpu=9 start=40.35 finish=44.48
1103638) ld.so cpu=10 start=40.35 finish=44.48
1103639) glibc-bench cpu=9 start=48.49 finish=52.62
1103640) ld.so cpu=2 start=48.49 finish=52.62
1103641) sh cpu=11 start=52.62 finish=52.63
1103642) sh cpu=12 start=52.62 finish=52.63
1103643) glibc-bench cpu=2 start=62.79 finish=63.91
1103644) ld.so cpu=3 start=62.79 finish=63.91
1103645) glibc-bench cpu=10 start=67.92 finish=69.05
1103646) ld.so cpu=11 start=67.92 finish=69.05
1103647) glibc-bench cpu=9 start=73.05 finish=74.18
1103648) ld.so cpu=10 start=73.05 finish=74.18
1103650) sh cpu=10 start=74.18 finish=74.18
1103651) sh cpu=3 start=74.18 finish=74.18
1103652) glibc-bench cpu=1 start=84.34 finish=86.49
1103653) ld.so cpu=10 start=84.35 finish=86.49
1103654) glibc-bench cpu=9 start=90.49 finish=92.64
1103655) ld.so cpu=10 start=90.50 finish=92.64
1103656) glibc-bench cpu=9 start=96.64 finish=98.78
1103657) ld.so cpu=10 start=96.64 finish=98.78