Tesseract is a GPU focused game that tries a set of scenes at different resolutions and reports frames/sec. As such unclear how much this really tests the CPU so don’t expect to run more of these game applications. The progression below uses increasing resolutions and presumably slower waits for GPU. Just a few processes are runable.
Running on a system with discrete GPU and ROCm running we can also used the amdsmi library to monitor business of AMD graphics and memory. In yellow below we can see the graphics is kept busier than CPU and the UMC in blue is also shown. Unfortunately, this uses the amdgpu driver and only seems to work platforms with ROCm.
Topdown metrics shows mix of frontend stalls and then retiring and backend.
AMD metrics show only 0.60 on cpu so most all is a GPU application time. There is a moderate L2 access with a higher miss rate.
elapsed 940.138
on_cpu 0.037 # 0.60 / 16 cores
utime 448.137
stime 114.663
nvcsw 5006246 # 99.31%
nivcsw 34548 # 0.69%
inblock 760 # 0.81/sec
onblock 15352 # 16.33/sec
cpu-clock 555072782963 # 555.073 seconds
task-clock 557387113153 # 557.387 seconds
page faults 1091959 # 1959.068/sec
context switches 5045214 # 9051.544/sec
cpu migrations 6429 # 11.534/sec
major page faults 783 # 1.405/sec
minor page faults 1091176 # 1957.663/sec
alignment faults 0 # 0.000/sec
emulation faults 0 # 0.000/sec
branches 432496189191 # 164.537 branches per 1000 inst
branch misses 12518545903 # 2.89% branch miss
conditional 312883554453 # 119.032 conditional branches per 1000 inst
indirect 16879982752 # 6.422 indirect branches per 1000 inst
cpu-cycles 2297827692401 # 0.10 GHz
instructions 4134264086743 # 1.80 IPC
slots 4592679099624 #
retiring 1399510657504 # 30.5% (30.6%)
-- ucode 5332102804 # 0.1%
-- fastpath 1394178554700 # 30.4%
frontend 1631768830655 # 35.5% (35.7%)
-- latency 1193385343362 # 26.0%
-- bandwidth 438383487293 # 9.5%
backend 1228985655369 # 26.8% (26.9%)
-- cpu 195338081902 # 4.3%
-- memory 1033647573467 # 22.5%
speculation 306018898077 # 6.7% ( 6.7%)
-- branch mispredict 301962227802 # 6.6%
-- pipeline restart 4056670275 # 0.1%
smt-contention 26346755888 # 0.6% ( 0.0%)
cpu-cycles 2081086819075 # 0.10 GHz
instructions 3702405530816 # 1.78 IPC
instructions 1229550171162 # 41.165 l2 access per 1000 inst
l2 hit from l1 47072177297 # 25.67% l2 miss
l2 miss from l1 10767695677 #
l2 hit from l2 pf 1316928266 #
l3 hit from l2 pf 2055888447 #
l3 miss from l2 pf 169576582 #
instructions 1224949261309 # 63.013 float per 1000 inst
float 512 127 # 0.000 AVX-512 per 1000 inst
float 256 158978 # 0.000 AVX-256 per 1000 inst
float 128 77187093384 # 63.012 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 with a somewhat different iGPU that seems ~2x slower frame rate.
elapsed 916.368
on_cpu 0.044 # 0.71 / 16 cores
utime 560.936
stime 85.333
nvcsw 1113258 # 99.18%
nivcsw 9227 # 0.82%
inblock 16088 # 17.56/sec
onblock 9976 # 10.89/sec
cpu-clock 639362215876 # 639.362 seconds
task-clock 640526730187 # 640.527 seconds
page faults 1119966 # 1748.508/sec
context switches 1126776 # 1759.140/sec
cpu migrations 8294 # 12.949/sec
major page faults 109 # 0.170/sec
minor page faults 1119857 # 1748.338/sec
alignment faults 0 # 0.000/sec
emulation faults 0 # 0.000/sec
branches 386646868699 # 170.775 branches per 1000 inst
branch misses 6406907616 # 1.66% branch miss
conditional 386646903227 # 170.775 conditional branches per 1000 inst
indirect 13807109676 # 6.098 indirect branches per 1000 inst
slots 7765648184300 #
retiring 2874712201531 # 37.0% (37.0%)
-- ucode 217635797671 # 2.8%
-- fastpath 2657076403860 # 34.2%
frontend 1908382930340 # 24.6% (24.6%)
-- latency 858073053677 # 11.0%
-- bandwidth 1050309876663 # 13.5%
backend 1650754897442 # 21.3% (21.3%)
-- cpu 768652337590 # 9.9%
-- memory 882102559852 # 11.4%
speculation 1405928996655 # 18.1% (18.1%)
-- branch mispredict 1385587146745 # 17.8%
-- pipeline restart 20341849910 # 0.3%
smt-contention 0 # 0.0% ( 0.0%)
cpu-cycles 1319100239590 # 0.09 GHz
instructions 2254013924600 # 1.71 IPC
l2 access 92921018270 # 41.629 l2 access per 1000 inst
l2 miss 37922616224 # 40.81% l2 miss
Process summary shows both gdrv (driver?) and several client and disk processes.
1005 processes
96 linux_64:gdrv0 433.33 94.69
24 linux_64_client 430.57 93.18
24 SDLTimer 430.53 93.16
24 linux_64_c:cs0 430.53 93.15
24 linux_6:disk$0 430.52 93.15
24 linux_64:shlo0 430.51 93.15
24 linux_64_c:sh1 430.51 93.14
24 linux_64_c:sh10 430.51 93.14
24 linux_64_c:sh2 430.51 93.14
24 linux_64_c:sh3 430.51 93.14
24 linux_64_c:sh4 430.51 93.14
24 linux_64_c:sh5 430.51 93.14
24 linux_64_c:sh7 430.51 93.14
24 linux_64_c:sh8 430.51 93.14
24 linux_64_c:sh0 430.50 93.14
24 linux_64_c:sh11 430.50 93.14
24 linux_64_c:sh6 430.50 93.14
24 linux_64_c:sh9 430.50 93.14
24 PulseHotplug 430.47 93.11
24 SDLAudioP2 430.46 93.11
68 clinfo 19.06 7.30
38 vulkaninfo 1.14 1.53
6 glxinfo:gdrv0 0.17 0.10
6 php 0.13 0.26
4 vulkani:disk$0 0.12 0.17
2 glxinfo 0.08 0.04
2 glxinfo:cs0 0.08 0.04
2 glxinfo:disk$0 0.08 0.04
6 clang 0.07 0.05
2 glxinfo:sh0 0.07 0.04
2 glxinfo:shlo0 0.07 0.04
2 llvmpipe-0 0.06 0.09
2 llvmpipe-1 0.06 0.09
2 llvmpipe-10 0.06 0.09
2 llvmpipe-11 0.06 0.09
2 llvmpipe-12 0.06 0.09
2 llvmpipe-13 0.06 0.09
2 llvmpipe-14 0.06 0.09
2 llvmpipe-15 0.06 0.09
2 llvmpipe-2 0.06 0.09
2 llvmpipe-3 0.06 0.09
2 llvmpipe-4 0.06 0.09
2 llvmpipe-5 0.06 0.09
2 llvmpipe-6 0.06 0.09
2 llvmpipe-7 0.06 0.09
2 llvmpipe-8 0.06 0.09
2 llvmpipe-9 0.06 0.09
3 rocminfo 0.00 0.03
1 lspci 0.00 0.03
1 ps 0.00 0.01
97 sh 0.00 0.00
50 uname 0.00 0.00
24 rm 0.00 0.00
24 tesseract 0.00 0.00
14 gsettings 0.00 0.00
12 gcc 0.00 0.00
9 systemd-detect- 0.00 0.00
8 stat 0.00 0.00
6 llvm-link 0.00 0.00
5 phoronix-test-s 0.00 0.00
2 gmain 0.00 0.00
2 lscpu 0.00 0.00
2 which 0.00 0.00
2 xrandr 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 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
0 processes running
47 maximum processes
Following is the processes for one iteration
290837) tesseract cpu=11 start=5.72 finish=37.33
290838) rm cpu=4 start=5.72 finish=5.72
290839) linux_64_client cpu=4 start=5.72 finish=37.31
290840) uname cpu=1 start=5.72 finish=5.72
290841) uname cpu=10 start=5.72 finish=5.73
290842) SDLTimer cpu=4 start=5.74 finish=37.31
290843) linux_64_c:cs0 cpu=9 start=5.78 finish=37.31
290844) linux_6:disk$0 cpu=10 start=5.78 finish=37.31
290845) linux_64_c:sh0 cpu=14 start=5.78 finish=37.31
290846) linux_64:shlo0 cpu=15 start=5.78 finish=37.31
290847) linux_64:gdrv0 cpu=14 start=5.79 finish=5.80
290848) linux_64:gdrv0 cpu=14 start=5.80 finish=5.80
290850) PulseHotplug cpu=15 start=5.80 finish=37.30
290851) linux_64:gdrv0 cpu=4 start=5.80 finish=5.80
290852) linux_64:gdrv0 cpu=7 start=5.81 finish=37.31
290853) linux_64_c:sh1 cpu=3 start=5.82 finish=37.31
290854) linux_64_c:sh2 cpu=12 start=5.82 finish=37.31
290855) linux_64_c:sh3 cpu=2 start=5.82 finish=37.31
290856) linux_64_c:sh4 cpu=6 start=5.82 finish=37.31
290857) SDLAudioP2 cpu=10 start=5.88 finish=37.30
290858) linux_64_c:sh5 cpu=12 start=5.91 finish=37.31
290859) linux_64_c:sh6 cpu=6 start=5.91 finish=37.31
290860) linux_64_c:sh7 cpu=5 start=5.91 finish=37.31
290861) linux_64_c:sh8 cpu=5 start=5.91 finish=37.31
290862) linux_64_c:sh9 cpu=15 start=5.91 finish=37.31
290863) linux_64_c:sh10 cpu=6 start=5.92 finish=37.31
290864) linux_64_c:sh11 cpu=1 start=5.92 finish=37.31