A simple test to measure context switch times. These show 304 with CraftyOS, 342 with Ubuntu on AMD and 210 with Ubuntu on Intel. This is a quick running single-threaded test.
Topdown profile has just a few data points
AMD metrics show only 20 seconds total and 0.29 cores, so also a good portion is test overhead.
elapsed 20.054
on_cpu 0.018 # 0.29 / 16 cores
utime 4.042
stime 1.866
nvcsw 1983 # 87.51%
nivcsw 283 # 12.49%
inblock 16 # 0.80/sec
onblock 12544 # 625.52/sec
cpu-clock 5923191652 # 5.923 seconds
task-clock 5925358739 # 5.925 seconds
page faults 140133 # 23649.707/sec
context switches 2208 # 372.636/sec
cpu migrations 256 # 43.204/sec
major page faults 2 # 0.338/sec
minor page faults 140131 # 23649.370/sec
alignment faults 0 # 0.000/sec
emulation faults 0 # 0.000/sec
branches 3629543676 # 199.805 branches per 1000 inst
branch misses 371673029 # 10.24% branch miss
conditional 1807110712 # 99.481 conditional branches per 1000 inst
indirect 75718384 # 4.168 indirect branches per 1000 inst
cpu-cycles 19927118685 # 0.06 GHz
instructions 18009441706 # 0.90 IPC
slots 41533810722 #
retiring 8878414099 # 21.4% (21.4%)
-- ucode 98614793 # 0.2%
-- fastpath 8779799306 # 21.1%
frontend 26228231198 # 63.1% (63.2%) high
-- latency 22577908482 # 54.4%
-- bandwidth 3650322716 # 8.8%
backend 5866295650 # 14.1% (14.1%) low
-- cpu 1052188093 # 2.5%
-- memory 4814107557 # 11.6%
speculation 536345946 # 1.3% ( 1.3%)
-- branch mispredict 535403653 # 1.3%
-- pipeline restart 942293 # 0.0%
smt-contention 24259787 # 0.1% ( 0.0%)
cpu-cycles 19881119763 # 0.06 GHz
instructions 17840215116 # 0.90 IPC
instructions 6060750171 # 17.474 l2 access per 1000 inst
l2 hit from l1 89913971 # 19.00% l2 miss
l2 miss from l1 11199606 #
l2 hit from l2 pf 7069515 #
l3 hit from l2 pf 4232960 #
l3 miss from l2 pf 4690054 #
instructions 6068369854 # 7.045 float per 1000 inst
float 512 67 # 0.000 AVX-512 per 1000 inst
float 256 8 # 0.000 AVX-256 per 1000 inst
float 128 42754042 # 7.045 AVX-128 per 1000 inst
float MMX 0 # 0.000 MMX per 1000 inst
float scalar 0 # 0.000 scalar per 1000 inst
instructions 18108100860 #
opcache 6120325302 # 337.988 opcache per 1000 inst
opcache miss 2693214500 # 44.0% opcache miss rate
l1 dTLB miss 25740136 # 1.421 L1 dTLB per 1000 inst
l2 dTLB miss 4195826 # 0.232 L2 dTLB per 1000 inst
instructions 18080295033 #
icache 6535414495 # 361.466 icache per 1000 inst
icache miss 149961081 # 2.3% icache miss rate
l1 iTLB miss 7771605 # 0.430 L1 iTLB per 1000 inst
l2 iTLB miss 0 # 0.000 L2 iTLB per 1000 inst
tlb flush 14969 # 0.001 TLB flush per 1000 inst
Intel metrics
elapsed 18.250
on_cpu 0.013 # 0.21 / 16 cores
utime 2.545
stime 1.207
nvcsw 1883 # 92.85%
nivcsw 145 # 7.15%
inblock 368 # 20.16/sec
onblock 1312 # 71.89/sec
cpu-clock 3762909328 # 3.763 seconds
task-clock 3765381454 # 3.765 seconds
page faults 136945 # 36369.489/sec
context switches 1949 # 517.610/sec
cpu migrations 247 # 65.598/sec
major page faults 2 # 0.531/sec
minor page faults 136943 # 36368.958/sec
alignment faults 0 # 0.000/sec
emulation faults 0 # 0.000/sec
branches 2890931293 # 177.255 branches per 1000 inst
branch misses 18691925 # 0.65% branch miss
conditional 2890943133 # 177.256 conditional branches per 1000 inst
indirect 80373303 # 4.928 indirect branches per 1000 inst
slots 77725376588 #
retiring 20898113762 # 26.9% (26.9%)
-- ucode 6193687965 # 8.0%
-- fastpath 14704425797 # 18.9%
frontend 22874502550 # 29.4% (29.4%)
-- latency 12004279543 # 15.4%
-- bandwidth 10870223007 # 14.0%
backend 31453841215 # 40.5% (40.5%)
-- cpu 26097993769 # 33.6%
-- memory 5355847446 # 6.9%
speculation 2567763043 # 3.3% ( 3.3%)
-- branch mispredict 2409561026 # 3.1%
-- pipeline restart 158202017 # 0.2%
smt-contention 0 # 0.0% ( 0.0%)
cpu-cycles 12974231376 # 0.05 GHz
instructions 15984695138 # 1.23 IPC
l2 access 250615984 # 15.719 l2 access per 1000 inst
l2 miss 78419869 # 31.29% l2 miss
cpu-cycles 13129389639 # 11.0% memory latency
load stalls 1364727104 # 5.0% l1 bound
l1 miss 703062088 # 2.7% l2 bound
l2 miss 350707746 # 0.8% l3 bound
l3 miss 251978924 # 1.9% dram bound
store_stalls 80035758 # 0.6% store bound
Process overview shows more time in Phoronix test suite routines than in the ctx_clock executable
354 processes
68 clinfo 18.85 6.28
3 ctx_clock 3.04 1.11
38 vulkaninfo 1.33 1.33
4 vulkani:disk$0 0.14 0.14
6 glxinfo:gdrv0 0.11 0.11
6 glxinfo:gl0 0.11 0.11
6 php 0.08 0.06
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
2 glxinfo 0.05 0.05
2 glxinfo:cs0 0.05 0.05
2 glxinfo:disk$0 0.05 0.05
2 glxinfo:sh0 0.05 0.05
2 glxinfo:shlo0 0.05 0.05
6 clang 0.04 0.08
1 lspci 0.01 0.01
3 rocminfo 0.00 0.01
1 ps 0.00 0.01
82 sh 0.00 0.00
13 gcc 0.00 0.00
12 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 ctx-clock 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 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 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 structure shows ~1.4 seconds per invocation
875082) ctx-clock cpu=1 start=5.60 finish=6.97
875083) ctx_clock cpu=4 start=5.60 finish=6.97
875084) ctx-clock cpu=8 start=10.97 finish=12.38
875085) ctx_clock cpu=9 start=10.98 finish=12.38
875088) ctx-clock cpu=0 start=16.39 finish=17.79
875089) ctx_clock cpu=9 start=16.39 finish=17.79