[HW Accel Support]: Having quite a hard time getting hardware acceleration working #11213
Replies: 4 comments 2 replies
-
If you are copying there is nothing to hwaccel on go2rtc I'd suggest letting Frigate run longer to collect logs |
Beta Was this translation helpful? Give feedback.
-
|
Heres an output showing the GPUs capabilities root@729f74eeb8fe:/opt/frigate# ffmpeg -decoders | grep cuvid Here's a new error message I get when using a camera configured strictly with FFMPEG and hardware decoding. This is the first time I've seen something relevant to decoding in the logs. 024-05-03 07:29:40.859997437 [2024-05-03 07:29:40] frigate.video ERROR : wyze: Unable to read frames from ffmpeg process. I suppose this is what it looks like when the ffmpeg library is not compatible with my GPUs drivers, which I'll point out are not updatable. |
Beta Was this translation helpful? Give feedback.
-
|
This is usually a problem with the host installation. It says it supports h264 but then
The built in ffmpeg fully supports nvidia gpus |
Beta Was this translation helpful? Give feedback.
-
|
Looking at past GitHub issues, I seem to have missed thr manual post install steps for Cuda. I’ll run the required actions there and see where I land. |
Beta Was this translation helpful? Give feedback.
-
|
Still no dice and I updated to driver version 535, loosing my sanity here. |
Beta Was this translation helpful? Give feedback.
-
|
This issue is fixed, I had to change the mdevctl type to the one below, others may work but type 156 was causing issues for me, also might not be important, but I did also change to Host vs AES x86 for CPU type in Proxmox. Decoding was working with that setting before but other issues here seem to indicate host cpu is best. nvidia-49 |
Beta Was this translation helpful? Give feedback.
-
Describe the problem you are having
NVIDIA SMI Shows in the Frigate Docker Container but nothing I've tired config wise seems to cause frigate to use the GPU for encoding and or decoding. Note the couple of commented lines under the front door config. I've tried go2rtc and FFgmpeg based hardware acceleration. Every config I've tried either results in no usage of the GPU, This is when I'm using go2rtc with the hardware option.
Or when I've tried to use hardware acceleration in FFmpeg it seems to just prevent any of my streams from loading.
Question I have,
Since I'm using go2rtc in copy mode would someone typically specify hardware encoding at the Ffmpeg level?
Is there some circumstance where I would not even benefit from hardware acceleration?
Also here is a screenshot of Nvidia-smi in the docker container (Neither the host or the docker SMI command shows running FFMPEG processes no matter how I've configured the cameras yet)
Currently running in a Proxmox VM with Ubuntu Guest 22.04 and Tesla P40 passed though to frigate with VGPU.
Nvidia Container toolkit is installed as well as the cuda toolkit.
Version
0.13.2-6476F8A
Frigate config file
docker-compose file or Docker CLI command
Relevant log output
FFprobe output from your camera
ffprobe version n5.1-2-g915ef932a3-20220731 Copyright (c) 2007-2022 the FFmpeg developers built with gcc 12.1.0 (crosstool-NG 1.25.0.55_3defb7b) configuration: --prefix=/ffbuild/prefix --pkg-config-flags=--static --pkg-config=pkg-config --cross-prefix=x86_64-ffbuild-linux-gnu- --arch=x86_64 --target-os=linux --enable-gpl --enable-version3 --disable-debug --enable-iconv --enable-libxml2 --enable-zlib --enable-libfreetype --enable-libfribidi --enable-gmp --enable-lzma --enable-fontconfig --enable-libvorbis --enable-opencl --enable-libpulse --enable-libvmaf --enable-libxcb --enable-xlib --enable-amf --enable-libaom --enable-libaribb24 --enable-avisynth --enable-libdav1d --enable-libdavs2 --disable-libfdk-aac --enable-ffnvcodec --enable-cuda-llvm --enable-frei0r --enable-libgme --enable-libass --enable-libbluray --enable-libjxl --enable-libmp3lame --enable-libopus --enable-mbedtls --enable-librist --enable-libtheora --enable-libvpx --enable-libwebp --enable-lv2 --enable-libmfx --enable-libopencore-amrnb --enable-libopencore-amrwb --enable-libopenh264 --enable-libopenjpeg --enable-libopenmpt --enable-librav1e --enable-librubberband --disable-schannel --enable-sdl2 --enable-libsoxr --enable-libsrt --enable-libsvtav1 --enable-libtwolame --enable-libuavs3d --enable-libdrm --enable-vaapi --enable-libvidstab --enable-vulkan --enable-libshaderc --enable-libplacebo --enable-libx264 --enable-libx265 --enable-libxavs2 --enable-libxvid --enable-libzimg --enable-libzvbi --extra-cflags=-DLIBTWOLAME_STATIC --extra-cxxflags= --extra-ldflags=-pthread --extra-ldexeflags=-pie --extra-libs='-ldl -lgomp' --extra-version=20220731 libavutil 57. 28.100 / 57. 28.100 libavcodec 59. 37.100 / 59. 37.100 libavformat 59. 27.100 / 59. 27.100 libavdevice 59. 7.100 / 59. 7.100 libavfilter 8. 44.100 / 8. 44.100 libswscale 6. 7.100 / 6. 7.100 libswresample 4. 7.100 / 4. 7.100 libpostproc 56. 6.100 / 56. 6.100 [http @ 0x55c69b5c5a80] HTTP error 500 Internal Server Error http://192.168.40.135/flv?port=1935: Server returned 5XX Server Error replyOperating system
Debian
Install method
Docker Compose
Network connection
Wired
Camera make and model
RLC411 and RLC511W
Any other information that may be helpful
No response
Beta Was this translation helpful? Give feedback.
All reactions