Benchmark for the ffmpeg multimedia framework, working with various video and image workloads. Eight different workloads with slightly different characteristics but moderate retirement rate.
AMD metrics show we’re spending ~3.62 of cores, with a floating point app and some workloads with more branch prediction issues.
elapsed 3516.551
on_cpu 0.226 # 3.62 / 16 cores
utime 12390.943
stime 340.884
nvcsw 15578181 # 97.24%
nivcsw 442087 # 2.76%
inblock 544 # 0.15/sec
onblock 2240680 # 637.18/sec
cpu-clock 12698964696026 # 12698.965 seconds
task-clock 12707598404585 # 12707.598 seconds
page faults 79503821 # 6256.400/sec
context switches 16034984 # 1261.842/sec
cpu migrations 1151669 # 90.628/sec
major page faults 209 # 0.016/sec
minor page faults 79503612 # 6256.384/sec
alignment faults 0 # 0.000/sec
emulation faults 0 # 0.000/sec
branches 7203248030886 # 75.632 branches per 1000 inst
branch misses 240049896958 # 3.33% branch miss
conditional 4533988502930 # 47.606 conditional branches per 1000 inst
indirect 523019511730 # 5.492 indirect branches per 1000 inst
cpu-cycles 50141302219917 # 0.90 GHz
instructions 95031760041237 # 1.90 IPC
slots 101058172691034 #
retiring 33343174547595 # 33.0% (38.6%)
-- ucode 369097077701 # 0.4%
-- fastpath 32974077469894 # 32.6%
frontend 18316326101792 # 18.1% (21.2%)
-- latency 11796656848290 # 11.7%
-- bandwidth 6519669253502 # 6.5%
backend 28902503738660 # 28.6% (33.5%)
-- cpu 9238344511430 # 9.1%
-- memory 19664159227230 # 19.5%
speculation 5741462680659 # 5.7% ( 6.7%)
-- branch mispredict 5493902043181 # 5.4%
-- pipeline restart 247560637478 # 0.2%
smt-contention 14753369236746 # 14.6% ( 0.0%)
cpu-cycles 50150039568465 # 0.90 GHz
instructions 95013999407440 # 1.89 IPC
instructions 31732577834943 # 38.418 l2 access per 1000 inst
l2 hit from l1 1041329163186 # 10.60% l2 miss
l2 miss from l1 77034115956 #
l2 hit from l2 pf 125542444785 #
l3 hit from l2 pf 32137285724 #
l3 miss from l2 pf 20098536952 #
instructions 31712371266847 # 162.036 float per 1000 inst
float 512 1031 # 0.000 AVX-512 per 1000 inst
float 256 8708556624 # 0.275 AVX-256 per 1000 inst
float 128 5129851498771 # 161.762 AVX-128 per 1000 inst
float MMX 0 # 0.000 MMX per 1000 inst
float scalar 431 # 0.000 scalar per 1000 inst
Intel metrics
elapsed 4537.973
on_cpu 0.224 # 3.59 / 16 cores
utime 15956.113
stime 313.231
nvcsw 17077668 # 93.39%
nivcsw 1209684 # 6.61%
inblock 872 # 0.19/sec
onblock 2240792 # 493.79/sec
cpu-clock 16173096823697 # 16173.097 seconds
task-clock 16184087266515 # 16184.087 seconds
page faults 79247994 # 4896.661/sec
context switches 18307277 # 1131.190/sec
cpu migrations 2784557 # 172.055/sec
major page faults 105 # 0.006/sec
minor page faults 79247889 # 4896.655/sec
alignment faults 0 # 0.000/sec
emulation faults 0 # 0.000/sec
branches 7230167358835 # 74.859 branches per 1000 inst
branch misses 237818000973 # 3.29% branch miss
conditional 7230168530419 # 74.859 conditional branches per 1000 inst
indirect 2079543575777 # 21.531 indirect branches per 1000 inst
slots 152043588937604 #
retiring 67100901539543 # 44.1% (44.1%)
-- ucode 4640783240152 # 3.1%
-- fastpath 62460118299391 # 41.1%
frontend 27983850905441 # 18.4% (18.4%)
-- latency 12653607703945 # 8.3%
-- bandwidth 15330243201496 # 10.1%
backend 33676837503150 # 22.1% (22.1%)
-- cpu 21959791561468 # 14.4%
-- memory 11717045941682 # 7.7%
speculation 24216123543815 # 15.9% (15.9%)
-- branch mispredict 23560593854272 # 15.5%
-- pipeline restart 655529689543 # 0.4%
smt-contention 0 # 0.0% ( 0.0%)
cpu-cycles 57065942417550 # 0.79 GHz
instructions 123132261252145 # 2.16 IPC
l2 access 1879230062109 # 29.384 l2 access per 1000 inst
l2 miss 324166645172 # 17.25% l2 miss
Process tree shows different av:: encoders being used for different tests. A lot of short-lived processes.
36594 processes
10914 ffmpeg 185063.52 3556.32
1080 av:h264:df15 12462.15 276.74
1080 av:h264:df14 12462.13 276.67
1080 av:h264:df13 12462.08 276.64
1080 av:h264:df12 12462.03 276.58
1080 av:h264:df11 12461.98 276.51
1080 av:h264:df10 12461.94 276.41
1080 av:h264:df9 12461.87 276.33
1080 av:h264:df8 12461.82 276.29
1080 av:h264:df7 12461.71 276.27
1080 av:h264:df6 12461.62 276.22
1080 av:h264:df5 12461.59 276.11
1080 av:h264:df4 12461.48 276.06
1080 av:h264:df3 12461.42 275.99
1080 av:h264:df2 12461.36 275.95
1080 av:h264:df1 12461.30 275.91
1080 av:h264:df0 12461.21 275.80
900 dec0:0:h264 9524.23 211.29
900 dmx0:matroska,w 9309.61 207.19
360 mux0:matroska 7341.88 105.99
540 mux0:null 4586.64 125.17
360 dmx1:matroska,w 1036.13 65.92
180 av:hevc:df15 610.90 38.27
180 av:hevc:df14 610.90 38.23
180 av:hevc:df12 610.89 38.23
180 av:hevc:df13 610.89 38.23
180 av:hevc:df11 610.88 38.22
180 av:hevc:df10 610.87 38.22
180 av:hevc:df9 610.87 38.21
180 av:hevc:df7 610.87 38.19
180 av:hevc:df8 610.87 38.19
180 av:hevc:df4 610.87 38.17
180 av:hevc:df5 610.87 38.17
180 av:hevc:df6 610.87 38.17
180 av:hevc:df2 610.86 38.16
180 av:hevc:df3 610.86 38.16
180 av:hevc:df1 610.85 38.16
180 av:hevc:df0 610.84 38.15
180 dec1:0:hevc 607.72 36.37
180 dec1:0:h264 522.94 34.05
1350 ffprobe 427.14 4.24
64 clinfo 10.88 3.52
25 python3 1.38 0.75
38 vulkaninfo 0.95 0.77
6 php 0.16 0.68
4 vulkani:disk$0 0.10 0.09
6 glxinfo:gdrv0 0.05 0.10
2 llvmpipe-0 0.05 0.05
2 llvmpipe-1 0.05 0.05
2 llvmpipe-10 0.05 0.05
2 llvmpipe-11 0.05 0.05
2 llvmpipe-12 0.05 0.05
2 llvmpipe-13 0.05 0.05
2 llvmpipe-14 0.05 0.05
2 llvmpipe-15 0.05 0.05
2 llvmpipe-2 0.05 0.05
2 llvmpipe-3 0.05 0.05
2 llvmpipe-4 0.05 0.05
2 llvmpipe-5 0.05 0.05
2 llvmpipe-6 0.05 0.05
2 llvmpipe-7 0.05 0.05
2 llvmpipe-8 0.05 0.05
2 llvmpipe-9 0.05 0.05
2 glxinfo 0.04 0.04
2 glxinfo:cs0 0.04 0.04
2 glxinfo:disk$0 0.04 0.04
2 glxinfo:shlo0 0.04 0.04
2 glxinfo:sh0 0.03 0.04
6 clang 0.02 0.04
1 lspci 0.00 0.02
464 sh 0.00 0.00
13 gcc 0.00 0.00
9 gsettings 0.00 0.00
9 stty 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
4 gmain 0.00 0.00
3 dconf worker 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 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
65 maximum processes
An example test shows parallelism might be more limited by processes than by available cores.
238602) ffmpeg cpu=7 start=5.64 finish=41.62
238603) python3 cpu=6 start=5.64 finish=41.62
238604) ffprobe cpu=1 start=5.65 finish=5.67
238605) ffprobe cpu=2 start=5.67 finish=5.97
238606) ffprobe cpu=6 start=5.97 finish=5.99
238607) ffmpeg cpu=6 start=5.99 finish=7.02
238608) av:h264:df0 cpu=9 start=6.00 finish=7.02
238609) av:h264:df1 cpu=10 start=6.00 finish=7.02
238610) av:h264:df2 cpu=11 start=6.00 finish=7.02
238611) av:h264:df3 cpu=12 start=6.00 finish=7.02
238612) av:h264:df4 cpu=14 start=6.00 finish=7.02
238613) av:h264:df5 cpu=7 start=6.00 finish=7.02
238614) av:h264:df6 cpu=5 start=6.00 finish=7.02
238615) av:h264:df7 cpu=9 start=6.00 finish=7.02
238616) av:h264:df8 cpu=10 start=6.00 finish=7.02
238617) av:h264:df9 cpu=11 start=6.00 finish=7.02
238618) av:h264:df10 cpu=12 start=6.00 finish=7.02
238619) av:h264:df11 cpu=8 start=6.00 finish=7.02
238620) av:h264:df12 cpu=9 start=6.00 finish=7.02
238621) av:h264:df13 cpu=5 start=6.00 finish=7.02
238622) av:h264:df14 cpu=9 start=6.00 finish=7.02
238623) av:h264:df15 cpu=10 start=6.01 finish=7.02
238624) dec0:0:h264 cpu=7 start=6.01 finish=6.98
238625) dmx0:matroska,w cpu=1 start=6.01 finish=6.97
238626) mux0:matroska cpu=5 start=6.04 finish=7.02
238627) sh cpu=9 start=7.03 finish=7.23
238628) ffmpeg cpu=13 start=7.03 finish=7.22
238629) av:h264:df0 cpu=0 start=7.05 finish=7.22
238630) av:h264:df1 cpu=15 start=7.05 finish=7.22
238631) av:h264:df2 cpu=12 start=7.05 finish=7.22
238632) av:h264:df3 cpu=2 start=7.05 finish=7.22
238633) av:h264:df4 cpu=14 start=7.05 finish=7.22
238634) av:h264:df5 cpu=5 start=7.06 finish=7.22
238635) av:h264:df6 cpu=11 start=7.06 finish=7.22
238636) av:h264:df7 cpu=15 start=7.06 finish=7.22
238637) av:h264:df8 cpu=0 start=7.06 finish=7.22
238638) av:h264:df9 cpu=11 start=7.06 finish=7.22
238639) av:h264:df10 cpu=7 start=7.06 finish=7.22
238640) av:h264:df11 cpu=5 start=7.06 finish=7.22
238641) av:h264:df12 cpu=3 start=7.06 finish=7.22
238642) av:h264:df13 cpu=1 start=7.06 finish=7.22
238643) av:h264:df14 cpu=12 start=7.06 finish=7.22
238644) av:h264:df15 cpu=12 start=7.06 finish=7.22
238645) dec0:0:h264 cpu=6 start=7.06 finish=7.21
238646) av:h264:df0 cpu=3 start=7.06 finish=7.22
238647) av:h264:df1 cpu=10 start=7.06 finish=7.22
238648) av:h264:df2 cpu=4 start=7.06 finish=7.22
238649) av:h264:df3 cpu=9 start=7.06 finish=7.22
238650) av:h264:df4 cpu=3 start=7.06 finish=7.22
238651) av:h264:df5 cpu=7 start=7.06 finish=7.22
238652) av:h264:df6 cpu=2 start=7.06 finish=7.22
238653) av:h264:df7 cpu=12 start=7.06 finish=7.22
238654) av:h264:df8 cpu=10 start=7.06 finish=7.22
238655) av:h264:df9 cpu=0 start=7.06 finish=7.22
238656) av:h264:df10 cpu=14 start=7.06 finish=7.22
238657) av:h264:df11 cpu=2 start=7.06 finish=7.22
238658) av:h264:df12 cpu=10 start=7.06 finish=7.22
238659) av:h264:df13 cpu=7 start=7.06 finish=7.22
238660) av:h264:df14 cpu=11 start=7.06 finish=7.22
238661) av:h264:df15 cpu=14 start=7.06 finish=7.22
238662) dec1:0:h264 cpu=5 start=7.06 finish=7.21
238663) dmx0:matroska,w cpu=6 start=7.07 finish=7.20
238664) dmx1:matroska,w cpu=5 start=7.10 finish=7.20
238665) ffmpeg cpu=6 start=7.13 finish=7.22
238666) ffmpeg cpu=2 start=7.13 finish=7.22
238667) ffmpeg cpu=8 start=7.13 finish=7.22
238668) ffmpeg cpu=11 start=7.13 finish=7.22
238669) ffmpeg cpu=9 start=7.13 finish=7.22
238670) ffmpeg cpu=12 start=7.13 finish=7.22
238671) ffmpeg cpu=5 start=7.13 finish=7.22
238672) ffmpeg cpu=0 start=7.13 finish=7.22
238673) ffmpeg cpu=1 start=7.13 finish=7.22
238674) ffmpeg cpu=14 start=7.13 finish=7.22
238675) ffmpeg cpu=10 start=7.13 finish=7.22
238676) ffmpeg cpu=15 start=7.13 finish=7.22
238677) ffmpeg cpu=4 start=7.13 finish=7.22
238678) ffmpeg cpu=3 start=7.13 finish=7.22
238679) ffmpeg cpu=7 start=7.13 finish=7.22
238680) mux0:null cpu=11 start=7.13 finish=7.21
238681) ffprobe cpu=12 start=7.23 finish=7.25
238682) ffprobe cpu=11 start=7.25 finish=7.25
238683) ffprobe cpu=12 start=7.25 finish=7.45
238684) ffprobe cpu=6 start=7.45 finish=7.46
238685) ffmpeg cpu=5 start=7.46 finish=7.98