Testing gzip with an archive of the Linux source tree. Relatively high frontend time and also higher than average speculation. Also interesting that none tests of different compression tools use the same metrics and workload so not easy to compare between tools.
AMD metrics show this is a single threaded test with on_cpu = 0.97. Branch misprediction is ~10%
elapsed 114.506
on_cpu 0.061 # 0.97 / 16 cores
utime 91.441
stime 19.796
nvcsw 177298 # 98.96%
nivcsw 1867 # 1.04%
inblock 40 # 0.35/sec
onblock 7227360 # 63117.86/sec
cpu-clock 111006148726 # 111.006 seconds
task-clock 111095628310 # 111.096 seconds
page faults 153331 # 1380.171/sec
context switches 179526 # 1615.959/sec
cpu migrations 440 # 3.961/sec
major page faults 0 # 0.000/sec
minor page faults 153331 # 1380.171/sec
alignment faults 0 # 0.000/sec
emulation faults 0 # 0.000/sec
branches 137063645615 # 194.883 branches per 1000 inst
branch misses 6598040560 # 4.81% branch miss
conditional 118795025229 # 168.908 conditional branches per 1000 inst
indirect 130980107 # 0.186 indirect branches per 1000 inst
cpu-cycles 523296982865 # 0.29 GHz
instructions 704108543986 # 1.35 IPC
slots 1046083290948 #
retiring 222296651323 # 21.3% (21.3%)
-- ucode 133972432 # 0.0%
-- fastpath 222162678891 # 21.2%
frontend 334207248998 # 31.9% (32.0%)
-- latency 185991431652 # 17.8%
-- bandwidth 148215817346 # 14.2%
backend 384933199811 # 36.8% (36.8%)
-- cpu 79870268435 # 7.6%
-- memory 305062931376 # 29.2%
speculation 104151488257 # 10.0% (10.0%)
-- branch mispredict 104110788174 # 10.0%
-- pipeline restart 40700083 # 0.0%
smt-contention 494316794 # 0.0% ( 0.0%)
cpu-cycles 520290545671 # 0.27 GHz
instructions 700630953700 # 1.35 IPC
instructions 233734844896 # 65.940 l2 access per 1000 inst
l2 hit from l1 10143531840 # 1.32% l2 miss
l2 miss from l1 101054348 #
l2 hit from l2 pf 5165956737 #
l3 hit from l2 pf 52180456 #
l3 miss from l2 pf 50850081 #
instructions 234427107073 # 1.198 float per 1000 inst
float 512 69 # 0.000 AVX-512 per 1000 inst
float 256 62 # 0.000 AVX-256 per 1000 inst
float 128 280776469 # 1.198 AVX-128 per 1000 inst
float MMX 0 # 0.000 MMX per 1000 inst
float scalar 0 # 0.000 scalar per 1000 instIntel metrics
elapsed 135.102
on_cpu 0.060 # 0.95 / 16 cores
utime 115.671
stime 13.159
nvcsw 180106 # 99.24%
nivcsw 1374 # 0.76%
inblock 424 # 3.14/sec
onblock 7226968 # 53492.83/sec
cpu-clock 127195184462 # 127.195 seconds
task-clock 127350251510 # 127.350 seconds
page faults 148395 # 1165.251/sec
context switches 181959 # 1428.808/sec
cpu migrations 819 # 6.431/sec
major page faults 0 # 0.000/sec
minor page faults 148395 # 1165.251/sec
alignment faults 0 # 0.000/sec
emulation faults 0 # 0.000/sec
branches 133636627064 # 191.427 branches per 1000 inst
branch misses 4859420388 # 3.64% branch miss
conditional 133636639640 # 191.427 conditional branches per 1000 inst
indirect 299381650 # 0.429 indirect branches per 1000 inst
slots 2816999297276 #
retiring 640641585470 # 22.7% (22.7%)
-- ucode 29076199603 # 1.0%
-- fastpath 611565385867 # 21.7%
frontend 534785879010 # 19.0% (19.0%)
-- latency 175651233196 # 6.2%
-- bandwidth 359134645814 # 12.7%
backend 1033672078961 # 36.7% (36.7%)
-- cpu 541860874031 # 19.2%
-- memory 491811204930 # 17.5%
speculation 602880895949 # 21.4% (21.4%)
-- branch mispredict 602043399503 # 21.4%
-- pipeline restart 837496446 # 0.0%
smt-contention 0 # 0.0% ( 0.0%)
cpu-cycles 470155455895 # 0.22 GHz
instructions 698463029008 # 1.49 IPC
l2 access 18736530988 # 26.854 l2 access per 1000 inst
l2 miss 995215530 # 5.31% l2 missProcess tree information shows only 4 instances of gzip and overall very short runtime the rest is mostly test suite overhead…
378 processes
4 gzip 88.30 1.40
64 clinfo 10.88 3.20
4 tar 1.40 9.06
38 vulkaninfo 0.57 1.48
2 cp 0.23 4.22
6 glxinfo:gdrv0 0.14 0.07
6 php 0.08 0.17
4 vulkani:disk$0 0.06 0.15
2 glxinfo 0.06 0.03
2 glxinfo:cs0 0.06 0.03
2 glxinfo:disk$0 0.06 0.03
2 glxinfo:sh0 0.06 0.03
2 glxinfo:shlo0 0.06 0.03
7 rm 0.05 2.99
2 llvmpipe-0 0.03 0.08
2 llvmpipe-1 0.03 0.08
2 llvmpipe-10 0.03 0.08
2 llvmpipe-11 0.03 0.08
2 llvmpipe-12 0.03 0.08
2 llvmpipe-13 0.03 0.08
2 llvmpipe-14 0.03 0.08
2 llvmpipe-15 0.03 0.08
2 llvmpipe-2 0.03 0.08
2 llvmpipe-3 0.03 0.08
2 llvmpipe-4 0.03 0.08
2 llvmpipe-5 0.03 0.08
2 llvmpipe-6 0.03 0.08
2 llvmpipe-7 0.03 0.08
2 llvmpipe-8 0.03 0.08
2 llvmpipe-9 0.03 0.08
6 clang 0.03 0.04
1 lspci 0.00 0.03
95 sh 0.00 0.00
12 gcc 0.00 0.00
9 stty 0.00 0.00
8 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 gmain 0.00 0.00
5 phoronix-test-s 0.00 0.00
4 bash 0.00 0.00
3 compress-gzip 0.00 0.00
3 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 cc 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
0 processes running
47 maximum processesThe core part of the benchmark is as follows
30297) compress-gzip start=13.71 finish=43.18
30298) tar start=13.71 finish=43.18
30299) sh start=13.71 finish=43.18
30300) gzip start=13.71 finish=43.18
30301) sh start=43.18 finish=43.24
30302) bash start=43.18 finish=43.24
30303) rm start=43.18 finish=43.23
30304) compress-gzip start=47.24 finish=76.22
30305) tar start=47.24 finish=76.22
30306) sh start=47.25 finish=76.22
30307) gzip start=47.25 finish=76.21