A test of file system operations related to what is used to maintain kernel trees. This has three subtests: compile, initial create and read compiled tree. These tests run quickly and look to be single-threaded.
Topdown profile shows frontend stalls as highest with relatively low backend stalls.
AMD metrics show little floating point, a high opcache miss rate and an 11% icache miss rate with 1/5 of the instructions as branches. Frontend latency is high. There is a high amount of output blocks and almost no input blocks.
elapsed 89.996
on_cpu 0.044 # 0.71 / 16 cores
utime 7.809
stime 56.158
nvcsw 2509 # 24.27%
nivcsw 7829 # 75.73%
inblock 8 # 0.09/sec
onblock 65538520 # 728234.48/sec
cpu-clock 63999083442 # 63.999 seconds
task-clock 64004384231 # 64.004 seconds
page faults 178065 # 2782.075/sec
context switches 10580 # 165.301/sec
cpu migrations 394 # 6.156/sec
major page faults 2 # 0.031/sec
minor page faults 178063 # 2782.044/sec
alignment faults 0 # 0.000/sec
emulation faults 0 # 0.000/sec
branches 63013567795 # 198.359 branches per 1000 inst
branch misses 5940612790 # 9.43% branch miss
conditional 35303103205 # 111.130 conditional branches per 1000 inst
indirect 1192983745 # 3.755 indirect branches per 1000 inst
cpu-cycles 290958943660 # 0.20 GHz
instructions 317320090766 # 1.09 IPC
slots 583966234098 #
retiring 116929459720 # 20.0% (20.0%)
-- ucode 571788094 # 0.1%
-- fastpath 116357671626 # 19.9%
frontend 401304724510 # 68.7% (68.7%) high
-- latency 336924894132 # 57.7%
-- bandwidth 64379830378 # 11.0%
backend 62079412727 # 10.6% (10.6%) low
-- cpu 13257356583 # 2.3%
-- memory 48822056144 # 8.4%
speculation 3454155649 # 0.6% ( 0.6%) low
-- branch mispredict 3447845685 # 0.6%
-- pipeline restart 6309964 # 0.0%
smt-contention 198173313 # 0.0% ( 0.0%)
cpu-cycles 289259190827 # 0.20 GHz
instructions 315675605246 # 1.09 IPC
instructions 105762868493 # 68.909 l2 access per 1000 inst
l2 hit from l1 6720971259 # 6.82% l2 miss
l2 miss from l1 241036077 #
l2 hit from l2 pf 311085299 #
l3 hit from l2 pf 88085211 #
l3 miss from l2 pf 167888574 #
instructions 105756026935 # 12.301 float per 1000 inst
float 512 74 # 0.000 AVX-512 per 1000 inst
float 256 668 # 0.000 AVX-256 per 1000 inst
float 128 1300940242 # 12.301 AVX-128 per 1000 inst
float MMX 0 # 0.000 MMX per 1000 inst
float scalar 21 # 0.000 scalar per 1000 inst
instructions 317305283501 #
opcache 96197134370 # 303.169 opcache per 1000 inst
opcache miss 57752090686 # 60.0% opcache miss rate
l1 dTLB miss 978514157 # 3.084 L1 dTLB per 1000 inst
l2 dTLB miss 13665626 # 0.043 L2 dTLB per 1000 inst
instructions 318170252018 #
icache 124538457131 # 391.421 icache per 1000 inst
icache miss 13704861327 # 11.0% icache miss rate
l1 iTLB miss 112977480 # 0.355 L1 iTLB per 1000 inst
l2 iTLB miss 0 # 0.000 L2 iTLB per 1000 inst
tlb flush 24981 # 0.000 TLB flush per 1000 inst
Intel metrics show L2 as the largest part of memory bound aspects.
elapsed 103.155
on_cpu 0.025 # 0.41 / 16 cores
utime 7.438
stime 34.585
nvcsw 6535 # 75.58%
nivcsw 2112 # 24.42%
inblock 1336 # 12.95/sec
onblock 65527288 # 635229.35/sec
cpu-clock 42023931539 # 42.024 seconds
task-clock 42032493432 # 42.032 seconds
page faults 167919 # 3994.981/sec
context switches 8957 # 213.097/sec
cpu migrations 491 # 11.681/sec
major page faults 0 # 0.000/sec
minor page faults 167919 # 3994.981/sec
alignment faults 0 # 0.000/sec
emulation faults 0 # 0.000/sec
branches 50807657927 # 172.266 branches per 1000 inst
branch misses 135894069 # 0.27% branch miss
conditional 50807670919 # 172.266 conditional branches per 1000 inst
indirect 1194939709 # 4.052 indirect branches per 1000 inst
slots 773415976724 #
retiring 331308877913 # 42.8% (42.8%)
-- ucode 35167431992 # 4.5%
-- fastpath 296141445921 # 38.3%
frontend 233006981567 # 30.1% (30.1%)
-- latency 80228342919 # 10.4%
-- bandwidth 152778638648 # 19.8%
backend 181914046624 # 23.5% (23.5%)
-- cpu 69711483930 # 9.0%
-- memory 112202562694 # 14.5%
speculation 39111851292 # 5.1% ( 5.1%)
-- branch mispredict 32350542141 # 4.2%
-- pipeline restart 6761309151 # 0.9%
smt-contention 0 # 0.0% ( 0.0%)
cpu-cycles 129783140930 # 0.08 GHz
instructions 294812371436 # 2.27 IPC
l2 access 16882455472 # 57.333 l2 access per 1000 inst
l2 miss 2140286539 # 12.68% l2 miss
cpu-cycles 129742170073 # 23.8% memory latency
load stalls 25972698653 # 4.1% l1 bound
l1 miss 20702278744 # 10.4% l2 bound
l2 miss 7150731886 # 1.4% l3 bound
l3 miss 5283474262 # 4.1% dram bound
store_stalls 4878082314 # 3.8% store bound
The process overview shows this driven from python with system time as the largest portion.
390 processes
68 clinfo 17.19 8.64
3 python2 6.44 55.32
38 vulkaninfo 1.14 1.80
6 glxinfo:gdrv0 0.13 0.07
6 glxinfo:gl0 0.13 0.07
4 vulkani:disk$0 0.12 0.19
6 php 0.09 0.06
2 glxinfo 0.07 0.03
2 glxinfo:cs0 0.07 0.03
2 glxinfo:disk$0 0.07 0.03
2 glxinfo:sh0 0.07 0.03
2 glxinfo:shlo0 0.07 0.03
2 llvmpipe-0 0.06 0.10
2 llvmpipe-1 0.06 0.10
2 llvmpipe-10 0.06 0.10
2 llvmpipe-11 0.06 0.10
2 llvmpipe-12 0.06 0.10
2 llvmpipe-13 0.06 0.10
2 llvmpipe-14 0.06 0.10
2 llvmpipe-15 0.06 0.10
2 llvmpipe-2 0.06 0.10
2 llvmpipe-3 0.06 0.10
2 llvmpipe-4 0.06 0.10
2 llvmpipe-5 0.06 0.10
2 llvmpipe-6 0.06 0.10
2 llvmpipe-7 0.06 0.10
2 llvmpipe-8 0.06 0.10
2 llvmpipe-9 0.06 0.10
6 clang 0.05 0.07
3 rocminfo 0.03 0.00
1 lspci 0.01 0.02
12 sync 0.00 0.13
96 sh 0.00 0.00
12 gcc 0.00 0.00
12 gsettings 0.00 0.00
9 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
5 which 0.00 0.00
3 compilebench 0.00 0.00
3 gmain 0.00 0.00
3 mkdir 0.00 0.00
3 rm 0.00 0.00
2 lscpu 0.00 0.00
2 uname 0.00 0.00
2 xset 0.00 0.00
1 cc 0.00 0.00
1 date 0.00 0.00
1 dconf worker 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 python 0.00 0.00
1 python3 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
Computation blocks look as follows
12442) compilebench cpu=9 start=6.64 finish=29.93
12443) rm cpu=11 start=6.64 finish=6.64
12444) mkdir cpu=15 start=6.64 finish=6.64
12445) which cpu=12 start=6.64 finish=6.64
12446) python2 cpu=1 start=6.64 finish=29.92
12448) sh cpu=4 start=15.08 finish=16.91
12449) sync cpu=5 start=15.09 finish=16.91
12450) sh cpu=12 start=22.44 finish=23.04
12451) sync cpu=5 start=22.44 finish=23.04
12452) sh cpu=12 start=23.99 finish=24.03
12453) sync cpu=5 start=23.99 finish=24.03
12454) sh cpu=10 start=29.87 finish=29.92
12455) sync cpu=3 start=29.87 finish=29.92