Measuring the time to compile Erlang/OTP. The number of runnable processes jumps high and CPU core usage is variable.
Topdown profile has a higher rate of topdown stalls with frontend stalls and retirement roughly similar.
AMD metrics show on average half the cores. Frontend stalls are the highest. There is not much floating point.
elapsed 374.136
on_cpu 0.520 # 8.33 / 16 cores
utime 2600.996
stime 513.696
nvcsw 7826857 # 26.55%
nivcsw 21651959 # 73.45%
inblock 24 # 0.06/sec
onblock 7827320 # 20921.03/sec
cpu-clock 3122273753608 # 3122.274 seconds
task-clock 3122941278301 # 3122.941 seconds
page faults 88926824 # 28475.343/sec
context switches 29413373 # 9418.484/sec
cpu migrations 517461 # 165.697/sec
major page faults 2910 # 0.932/sec
minor page faults 88923651 # 28474.327/sec
alignment faults 0 # 0.000/sec
emulation faults 0 # 0.000/sec
branches 2646561802828 # 192.873 branches per 1000 inst
branch misses 91839791917 # 3.47% branch miss
conditional 1924017982101 # 140.216 conditional branches per 1000 inst
indirect 112928353961 # 8.230 indirect branches per 1000 inst
cpu-cycles 11752414130889 # 1.98 GHz
instructions 13299671577609 # 1.13 IPC
slots 24883400890320 #
retiring 4545808476867 # 18.3% (22.4%)
-- ucode 19058779784 # 0.1%
-- fastpath 4526749697083 # 18.2%
frontend 8930881982420 # 35.9% (43.9%)
-- latency 6557542868286 # 26.4%
-- bandwidth 2373339114134 # 9.5%
backend 6017703339481 # 24.2% (29.6%)
-- cpu 853507402202 # 3.4%
-- memory 5164195937279 # 20.8%
speculation 831290685150 # 3.3% ( 4.1%)
-- branch mispredict 815429325807 # 3.3%
-- pipeline restart 15861359343 # 0.1%
smt-contention 4557505158750 # 18.3% ( 0.0%)
cpu-cycles 11730454882014 # 1.97 GHz
instructions 13294674228204 # 1.13 IPC
instructions 4558616129744 # 36.897 l2 access per 1000 inst
l2 hit from l1 144877796872 # 19.30% l2 miss
l2 miss from l1 19728155814 #
l2 hit from l2 pf 10591886123 #
l3 hit from l2 pf 6041907329 #
l3 miss from l2 pf 6689503149 #
instructions 4551855078110 # 31.314 float per 1000 inst
float 512 27337 # 0.000 AVX-512 per 1000 inst
float 256 189660 # 0.000 AVX-256 per 1000 inst
float 128 142535017390 # 31.314 AVX-128 per 1000 inst
float MMX 0 # 0.000 MMX per 1000 inst
float scalar 5847 # 0.000 scalar per 1000 inst
instructions 2705980 #
opcache 1009880 # 373.203 opcache per 1000 inst
opcache miss 542791 # 53.7% opcache miss rate
l1 dTLB miss 6952 # 2.569 L1 dTLB per 1000 inst
l2 dTLB miss 1329 # 0.491 L2 dTLB per 1000 inst
instructions 2721500 #
icache 1326721 # 487.496 icache per 1000 inst
icache miss 112354 # 8.5% icache miss rate
l1 iTLB miss 9 # 0.003 L1 iTLB per 1000 inst
l2 iTLB miss 0 # 0.000 L2 iTLB per 1000 inst
tlb flush 19 # 0.007 TLB flush per 1000 inst
Intel metrics
elapsed 395.889
on_cpu 0.533 # 8.53 / 16 cores
utime 3054.027
stime 323.352
nvcsw 7022544 # 24.22%
nivcsw 21975293 # 75.78%
inblock 71304 # 180.11/sec
onblock 7816912 # 19745.22/sec
cpu-clock 3380383176672 # 3380.383 seconds
task-clock 3381121509150 # 3381.122 seconds
page faults 88938503 # 26304.439/sec
context switches 28933721 # 8557.433/sec
cpu migrations 542782 # 160.533/sec
major page faults 1469 # 0.434/sec
minor page faults 88936766 # 26303.925/sec
alignment faults 0 # 0.000/sec
emulation faults 0 # 0.000/sec
branches 2571256188919 # 189.076 branches per 1000 inst
branch misses 61958336618 # 2.41% branch miss
conditional 2571261108567 # 189.076 conditional branches per 1000 inst
indirect 571366916746 # 42.015 indirect branches per 1000 inst
slots 23248461525260 #
retiring 7749937132128 # 33.3% (33.3%)
-- ucode 552996012166 # 2.4%
-- fastpath 7196941119962 # 31.0%
frontend 7269498380742 # 31.3% (31.3%)
-- latency 4040354222203 # 17.4%
-- bandwidth 3229144158539 # 13.9%
backend 5019972335993 # 21.6% (21.6%)
-- cpu 3131106952317 # 13.5%
-- memory 1888865383676 # 8.1%
speculation 3276124690314 # 14.1% (14.1%) high
-- branch mispredict 3156167726549 # 13.6%
-- pipeline restart 119956963765 # 0.5%
smt-contention 0 # 0.0% ( 0.0%)
cpu-cycles 8559345730142 # 1.37 GHz
instructions 11195665887023 # 1.31 IPC
l2 access 285577713652 # 35.058 l2 access per 1000 inst
l2 miss 78765912679 # 27.58% l2 miss
cpu-cycles 6242576212630 # 23.8% memory latency
load stalls 1419074135749 # 6.3% l1 bound
l1 miss 1024870800867 # 6.3% l2 bound
l2 miss 629646656380 # 2.3% l3 bound
l3 miss 486990186219 # 7.8% dram bound
store_stalls 65398513932 # 1.0% store bound
Process overview has a set of processes running to take user time. Over 150,000 processes total and it looks like we missed some.
153290 processes
3948 beam.smp 1349.21 277.09
3948 sys_sig_dispatc 1349.16 277.07
3948 sys_msg_dispatc 1349.11 277.04
3948 1_scheduler 1349.03 277.02
3948 1_dirty_cpu_sch 1348.95 277.00
3948 1_dirty_io_sche 1348.90 276.98
3948 2_dirty_io_sche 1348.84 276.98
3948 3_dirty_io_sche 1348.82 276.97
3948 4_dirty_io_sche 1348.80 276.95
3948 5_dirty_io_sche 1348.73 276.95
3948 6_dirty_io_sche 1348.67 276.91
3948 7_dirty_io_sche 1348.60 276.90
3948 8_dirty_io_sche 1348.56 276.88
3948 9_dirty_io_sche 1348.53 276.88
3948 10_dirty_io_sch 1348.47 276.87
3948 1_aux 1348.44 276.84
3948 0_poller 1348.39 276.83
3948 async_1 1347.00 276.47
186 javac 1089.11 40.06
96 Finalizer 544.81 20.08
96 Common-Cleaner 544.79 20.09
3795 cc1 353.47 31.04
288 cc1plus 157.15 20.61
2937 as 25.94 2.92
68 clinfo 16.54 5.98
30 10_scheduler 7.89 2.01
30 11_scheduler 7.89 2.01
30 2_scheduler 7.89 2.01
30 3_scheduler 7.89 2.01
30 4_scheduler 7.89 2.01
30 5_scheduler 7.89 2.01
30 6_scheduler 7.89 2.01
30 7_scheduler 7.89 2.01
30 8_scheduler 7.89 2.01
30 9_scheduler 7.89 2.01
30 12_scheduler 7.88 2.01
30 13_scheduler 7.88 2.01
30 14_scheduler 7.87 2.01
30 15_scheduler 7.87 2.01
30 16_scheduler 7.87 2.01
30 2_dirty_cpu_sch 7.87 2.01
30 3_dirty_cpu_sch 7.87 2.01
30 4_dirty_cpu_sch 7.87 2.01
30 5_dirty_cpu_sch 7.87 2.01
30 6_dirty_cpu_sch 7.87 2.01
30 10_dirty_cpu_sc 7.87 2.00
30 11_dirty_cpu_sc 7.87 2.00
30 12_dirty_cpu_sc 7.87 2.00
30 13_dirty_cpu_sc 7.87 2.00
30 14_dirty_cpu_sc 7.87 2.00
30 15_dirty_cpu_sc 7.87 2.00
30 16_dirty_cpu_sc 7.87 2.00
30 7_dirty_cpu_sch 7.87 2.00
30 8_dirty_cpu_sch 7.87 2.00
30 9_dirty_cpu_sch 7.87 2.00
9 yielding_c_fun 2.93 0.54
930 ld 2.65 0.90
3 gzip 2.05 0.10
93 G1 Main Marker 2.00 0.00
93 G1 Young RemSet 2.00 0.00
93 VM Periodic Tas 2.00 0.00
6670 bash 1.15 3.70
38 vulkaninfo 1.15 1.33
801 make 0.88 0.44
6 jar 0.54 0.10
72 perl 0.46 0.00
6 php 0.14 0.22
4 vulkani:disk$0 0.13 0.14
6 glxinfo:gdrv0 0.12 0.06
6 glxinfo:gl0 0.12 0.06
3 tar 0.07 1.33
2 llvmpipe-0 0.07 0.07
2 llvmpipe-1 0.07 0.07
2 llvmpipe-10 0.07 0.07
2 llvmpipe-11 0.07 0.07
2 llvmpipe-12 0.07 0.07
2 llvmpipe-13 0.07 0.07
2 llvmpipe-14 0.07 0.07
2 llvmpipe-15 0.07 0.07
2 llvmpipe-2 0.07 0.07
2 llvmpipe-3 0.07 0.07
2 llvmpipe-4 0.07 0.07
2 llvmpipe-5 0.07 0.07
2 llvmpipe-6 0.07 0.07
2 llvmpipe-7 0.07 0.07
2 llvmpipe-8 0.07 0.07
2 llvmpipe-9 0.07 0.07
6 clang 0.07 0.05
2 glxinfo 0.07 0.02
2 glxinfo:cs0 0.06 0.02
2 glxinfo:disk$0 0.06 0.02
2 glxinfo:sh0 0.06 0.02
2 glxinfo:shlo0 0.06 0.02
42 flex 0.06 0.00
5506 rm 0.03 1.00
7645 sh 0.03 0.07
3259 gcc 0.03 0.05
3221 sed 0.03 0.00
3 rocminfo 0.03 0.00
24 ranlib 0.02 0.80
24 ar 0.02 0.78
307 configure 0.02 0.00
12 m4 0.01 0.00
804 C2 CompilerThre 0.00 4156.98
426 C1 CompilerThre 0.00 2211.08
96 GC Thread#0 0.00 544.82
96 G1 Conc#0 0.00 544.81
96 Service Thread 0.00 544.81
96 VM Thread 0.00 544.81
96 Reference Handl 0.00 544.80
96 Sweeper thread 0.00 544.79
96 Signal Dispatch 0.00 544.61
93 G1 Conc#1 0.00 544.52
93 G1 Conc#2 0.00 544.52
93 GC Thread#1 0.00 544.51
93 G1 Refine#0 0.00 539.18
924 install 0.00 0.09
1 lspci 0.00 0.02
11774 dirname 0.00 0.00
5148 basename 0.00 0.00
3948 erl_child_setup 0.00 0.00
3885 dyn_erl 0.00 0.00
3573 cat 0.00 0.00
2475 config.sub 0.00 0.00
1044 mkdir 0.00 0.00
930 collect2 0.00 0.00
817 grep 0.00 0.00
786 awk 0.00 0.00
552 mv 0.00 0.00
453 expr 0.00 0.00
294 uname 0.00 0.00
276 g++ 0.00 0.00
261 cp 0.00 0.00
192 ls 0.00 0.00
119 conftest 0.00 0.00
114 ln 0.00 0.00
92 cc 0.00 0.00
82 sort 0.00 0.00
81 hostname 0.00 0.00
63 chmod 0.00 0.00
61 mktemp 0.00 0.00
60 inet_gethost 0.00 0.00
57 rmdir 0.00 0.00
54 arch 0.00 0.00
45 diff 0.00 0.00
45 find 0.00 0.00
42 tr 0.00 0.00
27 getconf 0.00 0.00
15 config.guess 0.00 0.00
12 gen_git_version 0.00 0.00
12 touch 0.00 0.00
11 gsettings 0.00 0.00
9 echo 0.00 0.00
9 otp_build 0.00 0.00
8 stat 0.00 0.00
8 systemd-detect- 0.00 0.00
6 llvm-link 0.00 0.00
6 pkg-config 0.00 0.00
5 phoronix-test-s 0.00 0.00
4 date 0.00 0.00
3 build-erlang 0.00 0.00
3 git 0.00 0.00
3 gmain 0.00 0.00
3 snmp-v2tov1 0.00 0.00
2 dconf worker 0.00 0.00
2 lscpu 0.00 0.00
2 which 0.00 0.00
2 xset 0.00 0.00
1 dmesg 0.00 0.00
1 dmidecode 0.00 0.00
1 ifconfig 0.00 0.00
1 ip 0.00 0.00
1 lsmod 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 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