Ubuntu 22.04 : OME Transcoding using GPU #1115

Pernifloss · 2023-02-10T08:31:09Z

Pernifloss
Feb 10, 2023

Hello !

I'm trying to use OME ABR with GPU acceleration :
I installed OME on my Ubuntu 22.04

I can stream but i can't access it.

Here all info i have:

The error log from OME when i start a stream :

[2023-02-10 08:23:15.585] I [SPRTMP-T1935:101733] Provider | stream.cpp:49   | Unknown/(102) has been started stream
[2023-02-10 08:23:15.585] I [SPRTMP-T1935:101733] RTMPProvider | rtmp_provider.cpp:152  | A RTMP client has connected from <ClientSocket: 0x7f04f7643c10, #4, Connected, TCP, Nonblocking, 95.178.88.86:47271>
[2023-02-10 08:23:17.067] I [SPRTMP-T1935:101733] MediaRouter | mediarouter_application.cpp:255  | [#default#app/stream(102)] Trying to create a stream
[2023-02-10 08:23:17.067] I [SPRTMP-T1935:101733] MediaRouter | mediarouter_stream.cpp:54   | [#default#app/stream(102)] Trying to create a mediarouter stream
[2023-02-10 08:23:17.067] I [SPRTMP-T1935:101733] Monitor | application_metrics.cpp:57   | Create StreamMetrics(stream/d1021ddd-6296-4252-8634-9764a9d12c13/default/#default#app/stream/i) for monitoring
[2023-02-10 08:23:17.068] I [SPRTMP-T1935:101733] Transcoder | transcoder_stream.cpp:626  | [#default#app/stream(102)] Output stream has been created. [#default#app/stream(534721361)]
[2023-02-10 08:23:17.083] I [SPRTMP-T1935:101733] MediaRouter | mediarouter_application.cpp:255  | [#default#app/stream(534721361)] Trying to create a stream
[2023-02-10 08:23:17.083] I [SPRTMP-T1935:101733] MediaRouter | mediarouter_stream.cpp:54   | [#default#app/stream(534721361)] Trying to create a mediarouter stream
[2023-02-10 08:23:17.083] I [SPRTMP-T1935:101733] Monitor | application_metrics.cpp:57   | Create StreamMetrics(stream/d1021ddd-6296-4252-8634-9764a9d12c13/default/#default#app/stream/o) for monitoring
[2023-02-10 08:23:17.083] I [SPRTMP-T1935:101733] Transcoder | transcoder_stream.cpp:92   | [#default#app/stream(102)] Transcoder stream has been started.
[2023-02-10 08:23:17.129] I [InboundWorker:101734] MediaRouter | mediarouter_application.cpp:779  | [#default#app/stream(102)] Stream has been created 
[Stream Info]
id(102), msid(0), output(stream), SourceType(Rtmp), RepresentationType(Source), Created Time (Fri Feb 10 08:23:15 2023) UUID(d1021ddd-6296-4252-8634-9764a9d12c13/default/#default#app/stream/i)

        Video Track #0: Name(video_0) Bitrate(2.00Mb) Codec(1,H264,Auto) BSF(H264_AVCC) Resolution(1920x1080) Framerate(30.00fps) KeyInterval(0) BFrames(0) timebase(1/1000)
        Audio Track #1: Name(audio_0) Bitrate(160.00Kb) Codec(6,AAC,Auto) BSF(AAC_RAW) Samplerate(48.0K) Format(s16, 16) Channel(stereo, 2) timebase(1/1000)
        Data  Track #302479339: Name(thHbrpm6) Codec(0,Unknown,Auto) BSF(Unknown) timebase(1/1000)
[2023-02-10 08:23:17.190] W [InboundWorker:101734] MediaRouter | mediarouter_stream.cpp:795  | [#default#app/stream(102)] b-frame has been detected in the 0 track of inbound stream
[2023-02-10 08:23:17.190] W [InboundWorker:101734] MediaRouter | mediarouter_stream.cpp:795  | [#default#app/stream(102)] b-frame has been detected in the 0 track of inbound stream
[2023-02-10 08:23:17.191] W [InboundWorker:101734] MediaRouter | mediarouter_stream.cpp:795  | [#default#app/stream(102)] b-frame has been detected in the 0 track of inbound stream
[2023-02-10 08:23:17.190] W [OutboundWorker:101735] MediaRouter | mediarouter_stream.cpp:795  | [#default#app/stream(534721361)] b-frame has been detected in the 0 track of outbound stream
[2023-02-10 08:23:17.191] W [OutboundWorker:101735] MediaRouter | mediarouter_stream.cpp:795  | [#default#app/stream(534721361)] b-frame has been detected in the 0 track of outbound stream
[2023-02-10 08:23:17.191] I [Decaac:101854] Transcoder | decoder_aac.cpp:221  | [#default#app/stream(102)] input track information: [audio] aac (LC), 48000 Hz, stereo, fltp, 149 kbps, timebase: 1/1000, frame_size: 1024
[2023-02-10 08:23:17.191] I [Decaac:101854] Transcoder | transcoder_encoder.cpp:65   | [#3] hwardware acceleration of the encoder is enabled. The library to be used will be Auto
[2023-02-10 08:23:17.192] I [Decaac:101854] Transcoder | filter_resampler.cpp:146  | Resampler is enabled for track #1 using parameters. input: time_base=1/1000:sample_rate=48000:sample_fmt=fltp:channel_layout=stereo / outputs: asettb=1/48000,aresample=async=1000,aresample=48000,aformat=sample_fmts=s16:channel_layouts=stereo,asetnsamples=n=960
[2023-02-10 08:23:17.238] W [InboundWorker:101734] MediaRouter | mediarouter_stream.cpp:795  | [#default#app/stream(102)] b-frame has been detected in the 0 track of inbound stream
[2023-02-10 08:23:17.238] W [OutboundWorker:101735] MediaRouter | mediarouter_stream.cpp:795  | [#default#app/stream(534721361)] b-frame has been detected in the 0 track of outbound stream
[2023-02-10 08:23:17.250] W [InboundWorker:101734] MediaRouter | mediarouter_stream.cpp:795  | [#default#app/stream(102)] b-frame has been detected in the 0 track of inbound stream
[2023-02-10 08:23:17.251] W [OutboundWorker:101735] MediaRouter | mediarouter_stream.cpp:795  | [#default#app/stream(534721361)] b-frame has been detected in the 0 track of outbound stream
[2023-02-10 08:23:17.260] W [InboundWorker:101734] MediaRouter | mediarouter_stream.cpp:795  | [#default#app/stream(102)] b-frame has been detected in the 0 track of inbound stream
[2023-02-10 08:23:17.260] W [OutboundWorker:101735] MediaRouter | mediarouter_stream.cpp:795  | [#default#app/stream(534721361)] b-frame has been detected in the 0 track of outbound stream
[2023-02-10 08:23:17.270] I [Dech264NV:101853] Transcoder | decoder_avc_nv.cpp:233  | [#default#app/stream(102)] input stream information: [video] h264 (High 4.0), cuda, 1920x1080 [SAR 1:1 DAR 16:9], 60 fps, 1953 kbps, timebase: 1/60, frame_size: 0
[2023-02-10 08:23:17.300] W [InboundWorker:101734] MediaRouter | mediarouter_stream.cpp:795  | [#default#app/stream(102)] b-frame has been detected in the 0 track of inbound stream
[2023-02-10 08:23:17.301] W [OutboundWorker:101735] MediaRouter | mediarouter_stream.cpp:795  | [#default#app/stream(534721361)] b-frame has been detected in the 0 track of outbound stream
[2023-02-10 08:23:17.304] I [Dech264NV:101853] Transcoder | transcoder_encoder.cpp:65   | [#1] hwardware acceleration of the encoder is enabled. The library to be used will be Auto
[2023-02-10 08:23:17.323] W [InboundWorker:101734] MediaRouter | mediarouter_stream.cpp:795  | [#default#app/stream(102)] b-frame has been detected in the 0 track of inbound stream
[2023-02-10 08:23:17.324] W [OutboundWorker:101735] MediaRouter | mediarouter_stream.cpp:795  | [#default#app/stream(534721361)] b-frame has been detected in the 0 track of outbound stream
[2023-02-10 08:23:17.366] W [InboundWorker:101734] MediaRouter | mediarouter_stream.cpp:795  | [#default#app/stream(102)] b-frame has been detected in the 0 track of inbound stream
[2023-02-10 08:23:17.366] W [OutboundWorker:101735] MediaRouter | mediarouter_stream.cpp:795  | [#default#app/stream(534721361)] b-frame has been detected in the 0 track of outbound stream
[2023-02-10 08:23:17.407] W [InboundWorker:101734] MediaRouter | mediarouter_stream.cpp:795  | [#default#app/stream(102)] b-frame has been detected in the 0 track of inbound stream
[2023-02-10 08:23:17.407] W [OutboundWorker:101735] MediaRouter | mediarouter_stream.cpp:795  | [#default#app/stream(534721361)] b-frame has been detected in the 0 track of outbound stream
[2023-02-10 08:23:17.431] W [OutboundWorker:101735] MediaRouter | mediarouter_stream.cpp:795  | [#default#app/stream(534721361)] b-frame has been detected in the 0 track of outbound stream
[2023-02-10 08:23:19.273] E [Dech264NV:101853] FFmpeg | third_parties.cpp:111  | [AVFilter: 0x7f04fd60a440] cu->cuModuleLoadData(cu_module, buf) failed
[2023-02-10 08:23:19.273] E [Dech264NV:101853] FFmpeg | third_parties.cpp:111  | [AVFilter: 0x7f04fd60a440]  -> CUDA_ERROR_INVALID_PTX: a PTX JIT compilation failed
[2023-02-10 08:23:19.273] E [Dech264NV:101853] FFmpeg | third_parties.cpp:111  | [AVFilter: 0x7f04fd60a440] 
[2023-02-10 08:23:19.273] E [Dech264NV:101853] FFmpeg | third_parties.cpp:111  | [AVFilter: 0x7f04fd60a440] Failed to configure output pad on Parsed_scale_cuda_2
[2023-02-10 08:23:19.273] E [Dech264NV:101853] Transcoder | filter_rescaler.cpp:162  | Could not validate filter graph for rescaling: -542398533
[2023-02-10 08:23:19.273] E [Dech264NV:101853] Transcoder | transcoder_filter.cpp:63   | Could not craete filter
[2023-02-10 08:23:19.273] E [Dech264NV:101853] Transcoder | transcoder_stream.cpp:1457 | [#default#app/stream(102)] `ed to create filter. filterId: 1

And the log when i try to access the stream :

[2023-02-10 08:24:53.038] W [SPllhls-T80:101724] LLHLS Publisher | llhls_publisher.cpp:340  | (#default#app/stream) stream has not started.
[2023-02-10 08:24:53.039] E [SPllhls-T80:101724] HttpServer | http_exchange.cpp:43   | 
[Client] HttpConnection(0x7f0520401410) : HTTP/1.1 <ClientSocket: 0x7f0520401010, #62, Connected, TCP, Nonblocking, {my_ip}:3269> TLS(Disabled) [Duration] 10011 
[Request] URL(http://{server_ip}/app/stream/llhls.m3u8) Method(GET) Version(1.1) Request Time(2023-02-10T08:24:43.027+00:00)
[Response] Status(404) Sent Size(0) Response Time(2023-02-10T08:24:53.039+00:00)

My conf Server.xml OutputProfiles :

     <OutputProfiles>
                        <HardwareAcceleration>true</HardwareAcceleration>
                        <OutputProfile>
                            <Name>bypass_stream</Name>
                            <OutputStreamName>${OriginStreamName}</OutputStreamName>
                            <!--LLHLS URL : https://domain/app/stream/abr.m3u8 -->
                            <Playlist>
                                <Name>For LLHLS</Name>
                                <FileName>abr</FileName>
                                <Options> <!-- Optinal -->
                                    <!--
                                    Automatically switch rendition in WebRTC ABR
                                    [Default] : true
                                    -->
                                    <WebRtcAutoAbr>true</WebRtcAutoAbr>
                                </Options>
                                <Rendition>
                                    <Name>Bypass</Name>
                                    <Video>bypass_video</Video>
                                    <Audio>bypass_audio</Audio>
                                </Rendition>
                                <Rendition>
                                    <Name>FHD</Name>
                                    <Video>video_540</Video>
                                    <Audio>bypass_audio</Audio>
                                </Rendition>
                                <Rendition>
                                    <Name>HD</Name>
                                    <Video>video_720</Video>
                                    <Audio>bypass_audio</Audio>
                                </Rendition>
                            </Playlist>
                            <!--LLHLS URL : https://domain/app/stream/llhls.m3u8 -->
                            <Playlist>
                                <Name>Change Default</Name>
                                <FileName>llhls</FileName>
                                <Rendition>
                                    <Name>HD</Name>
                                    <Video>video_720</Video>
                                    <Audio>bypass_audio</Audio>
                                </Rendition>
                            </Playlist>
                            <Encodes>
                                <Audio>
                                    <Name>bypass_audio</Name>
                                    <Bypass>true</Bypass>
                                </Audio>
                                <Video>
                                    <Name>bypass_video</Name>
                                    <Bypass>true</Bypass>
                                </Video>
                                <Audio>
                                    <Codec>opus</Codec>~
                                    <Bitrate>128000</Bitrate>
                                    <Samplerate>48000</Samplerate>
                                    <Channel>2</Channel>
                                </Audio>
                                <Video>
                                    <Name>video_540</Name>
                                    <Codec>h264</Codec>
                                    <Bitrate>1024000</Bitrate>
                                    <Framerate>30</Framerate>
                                    <Width>1024</Width>
                                    <Height>576</Height>
                                    <Preset>faster</Preset>
                                </Video>
                                <Video>
                                    <Name>video_720</Name>
                                    <Codec>h264</Codec>
                                    <Bitrate>2024000</Bitrate>
                                    <Framerate>30</Framerate>
                                    <Width>1280</Width>
                                    <Height>720</Height>
                                    <Preset>faster</Preset>
                                </Video>
                            </Encodes>
                        </OutputProfile>

nvidia-smi

+-----------------------------------------------------------------------------+
| NVIDIA-SMI 525.85.12    Driver Version: 525.85.12    CUDA Version: 12.0     |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|                               |                      |               MIG M. |
|===============================+======================+======================|
|   0  Tesla M6            On   | 00000000:21:00.0 Off |                  Off |
| N/A   40C    P0    25W / 100W |    136MiB /  8192MiB |      1%      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+
                                                                               
+-----------------------------------------------------------------------------+
| Processes:                                                                  |
|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |
|        ID   ID                                                   Usage      |
|=============================================================================|
|    0   N/A  N/A    101706      C   /usr/bin/OvenMediaEngine          133MiB |
+-----------------------------------------------------------------------------+

nvcc --version

nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2023 NVIDIA Corporation
Built on Fri_Jan__6_16:45:21_PST_2023
Cuda compilation tools, release 12.0, V12.0.140
Build cuda_12.0.r12.0/compiler.32267302_0

Keukhan · 2023-02-15T15:15:00Z

Keukhan
Feb 15, 2023
Maintainer Sponsor

@Pernifloss

Sorry for the late reply.
It is expected that CUDA's Scale PTX module loading failure is the cause. However, it was not possible to reproduce the problem in the same OS environment.

Are you using a physical server? Or are you using GPU instances in Cloud?

I look forward to your reply.

0 replies

Pernifloss · 2023-02-15T16:05:32Z

Pernifloss
Feb 15, 2023
Author

Hello !

I'm using a dedicated server from serverroom.

Thanks.

0 replies

Pernifloss · 2023-02-23T07:36:38Z

Pernifloss
Feb 23, 2023
Author

Hello, do you have any advice for me?

Maybe I didn't make the project properly ?

Do I just have to do

sudo apt-get update
cd OvenMediaEngine-master/src
make release
sudo make install
systemctl start ovenmediaengine
# If you want automatically start on boot
systemctl enable ovenmediaengine.service

Once I installed driver and nvcc?

Possibly I missed a step?
Do you have any tips? I really want to try out ABR on GPU to see if I can use OME in a production environment.

I can use docker if needed, but I didn't manage to so far.

Thanks in advance.

0 replies

Maypeur · 2023-03-02T14:49:54Z

Maypeur
Mar 2, 2023

Same here, i compiled everything well but same error :

[2023-03-02 14:25:27.191] I [Dech264NV:1978] Transcoder | decoder_avc_nv.cpp:233  | [#default#app/stream2(100)] input stream information: [video] h264 (High 4.0), cuda, 1920x1080 [SAR 1:1 DAR 16:9], 60 fps, 1953 kbps, timebase: 1/60, frame_size: 0
[2023-03-02 14:25:27.197] I [Dech264NV:1978] Transcoder | transcoder_encoder.cpp:65   | [#1] hwardware acceleration of the encoder is enabled. The library to be used will be Auto
[2023-03-02 14:25:27.243] I [Dech264NV:1978] Transcoder | transcoder_encoder.cpp:65   | [#2] hwardware acceleration of the encoder is enabled. The library to be used will be Auto
[2023-03-02 14:25:29.131] E [Dech264NV:1978] FFmpeg | third_parties.cpp:111  | [AVFilter: 0x7f20c39f0400] cu->cuModuleLoadData(cu_module, buf) failed
[2023-03-02 14:25:29.131] E [Dech264NV:1978] FFmpeg | third_parties.cpp:111  | [AVFilter: 0x7f20c39f0400]  -> CUDA_ERROR_INVALID_PTX: a PTX JIT compilation failed
[2023-03-02 14:25:29.131] E [Dech264NV:1978] FFmpeg | third_parties.cpp:111  | [AVFilter: 0x7f20c39f0400]
[2023-03-02 14:25:29.131] E [Dech264NV:1978] FFmpeg | third_parties.cpp:111  | [AVFilter: 0x7f20c39f0400] Failed to configure output pad on Parsed_scale_cuda_2
[2023-03-02 14:25:29.131] E [Dech264NV:1978] Transcoder | filter_rescaler.cpp:162  | Could not validate filter graph for rescaling: -542398533
[2023-03-02 14:25:29.132] E [Dech264NV:1978] Transcoder | transcoder_filter.cpp:63   | Could not craete filter
[2023-03-02 14:25:29.132] E [Dech264NV:1978] Transcoder | transcoder_stream.cpp:773  | [#default#app/stream2(100)] Failed to create filter. Filter(1)
[2023-03-02 14:25:30.528] E [Dech264NV:1978] FFmpeg | third_parties.cpp:111  | [AVFilter: 0x7f20c39f04c0] cu->cuModuleLoadData(cu_module, buf) failed
[2023-03-02 14:25:30.528] E [Dech264NV:1978] FFmpeg | third_parties.cpp:111  | [AVFilter: 0x7f20c39f04c0]  -> CUDA_ERROR_INVALID_PTX: a PTX JIT compilation failed
[2023-03-02 14:25:30.528] E [Dech264NV:1978] FFmpeg | third_parties.cpp:111  | [AVFilter: 0x7f20c39f04c0]
[2023-03-02 14:25:30.528] E [Dech264NV:1978] FFmpeg | third_parties.cpp:111  | [AVFilter: 0x7f20c39f04c0] Failed to configure output pad on Parsed_scale_cuda_2
[2023-03-02 14:25:30.528] E [Dech264NV:1978] Transcoder | filter_rescaler.cpp:162  | Could not validate filter graph for rescaling: -542398533
[2023-03-02 14:25:30.528] E [Dech264NV:1978] Transcoder | transcoder_filter.cpp:63   | Could not craete filter
[2023-03-02 14:25:30.528] E [Dech264NV:1978] Transcoder | transcoder_stream.cpp:773  | [#default#app/stream2(100)] Failed to create filter. Filter(2)


+-----------------------------------------------------------------------------+
| NVIDIA-SMI 510.108.03   Driver Version: 510.108.03   CUDA Version: 11.6     |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|                               |                      |               MIG M. |
|===============================+======================+======================|
|   0  Tesla M6            Off  | 00000000:21:00.0 Off |                  Off |
| N/A   48C    P0    26W / 100W |    178MiB /  8192MiB |      2%      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+

+-----------------------------------------------------------------------------+
| Processes:                                                                  |
|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |
|        ID   ID                                                   Usage      |
|=============================================================================|
|    0   N/A  N/A      1868      C   /usr/bin/OvenMediaEngine          175MiB |
+-----------------------------------------------------------------------------+

0 replies

Maypeur · 2023-03-02T16:03:52Z

Maypeur
Mar 2, 2023

More debug info :

[2023-03-02 15:51:42.522] D [Dech264NV:3360] FFmpeg | third_parties.cpp:125  | [AVCodecContext: 0x7f696eae9c00] Loaded Nvenc version 11.1
[2023-03-02 15:51:42.522] D [Dech264NV:3360] FFmpeg | third_parties.cpp:125  | [AVCodecContext: 0x7f696eae9c00] Nvenc initialized successfully
[2023-03-02 15:51:42.525] D [Dech264NV:3360] FFmpeg | third_parties.cpp:125  | [AVCodecContext: 0x7f696eae9c00] Calling dl_fn->cuda_dl->cuCtxPushCurrent(ctx->cu_context)
[2023-03-02 15:51:42.540] D [Dech264NV:3360] FFmpeg | third_parties.cpp:125  | [AVCodecContext: 0x7f696eae9c00] Calling dl_fn->cuda_dl->cuCtxPopCurrent(&dummy)
[2023-03-02 15:51:42.540] D [Dech264NV:3360] FFmpeg | third_parties.cpp:125  | [AVCodecContext: 0x7f696eae9c00] Calling dl_fn->cuda_dl->cuCtxPushCurrent(ctx->cu_context)
[2023-03-02 15:51:42.542] D [Resampler:3363] FFmpeg | third_parties.cpp:125  | [AVFilter: 0x7f696e0d9300] tb:1/1000 pts:149 -> tb:1/48000 pts:7152
[2023-03-02 15:51:42.561] D [Resampler:3363] FFmpeg | third_parties.cpp:125  | [AVFilter: 0x7f696e0d9300] tb:1/1000 pts:170 -> tb:1/48000 pts:8160
[2023-03-02 15:51:42.574] D [Dech264NV:3360] FFmpeg | third_parties.cpp:125  | [AVCodecContext: 0x7f696eae9c00] Calling dl_fn->cuda_dl->cuCtxPopCurrent(&dummy)
[2023-03-02 15:51:42.574] D [Dech264NV:3360] FFmpeg | third_parties.cpp:125  | [AVFilter: 0x7f696eb1d780] Setting 'video_size' to value '1920x1080'
[2023-03-02 15:51:42.574] D [Dech264NV:3360] FFmpeg | third_parties.cpp:125  | [AVFilter: 0x7f696eb1d780] Setting 'pix_fmt' to value '23'
[2023-03-02 15:51:42.574] D [Dech264NV:3360] FFmpeg | third_parties.cpp:125  | [AVFilter: 0x7f696eb1d780] Setting 'time_base' to value '1/1000'
[2023-03-02 15:51:42.574] D [Dech264NV:3360] FFmpeg | third_parties.cpp:125  | [AVFilter: 0x7f696eb1d780] Setting 'pixel_aspect' to value '1/1'
[2023-03-02 15:51:42.574] D [Dech264NV:3360] FFmpeg | third_parties.cpp:125  | [AVFilter: 0x7f696eb1d780] w:1920 h:1080 pixfmt:nv12 tb:1/1000 fr:0/1 sar:1/1
[2023-03-02 15:51:42.574] D [Dech264NV:3360] FFmpeg | third_parties.cpp:125  | [AVFilter: 0x7f696eb1d900] Setting 'fps' to value '30.00'
[2023-03-02 15:51:42.574] D [Dech264NV:3360] FFmpeg | third_parties.cpp:125  | [AVFilter: 0x7f696eb1d900] Setting 'round' to value 'near'


[2023-03-02 15:51:44.295] D [Dech264NV:3360] FFmpeg | third_parties.cpp:125  | [AVFilter: 0x7f696eb8de40] Calling cu->cuModuleLoadData(cu_module, buf)
[2023-03-02 15:51:44.295] E [Dech264NV:3360] FFmpeg | third_parties.cpp:111  | [AVFilter: 0x7f696eb8de40] cu->cuModuleLoadData(cu_module, buf) failed
[2023-03-02 15:51:44.295] E [Dech264NV:3360] FFmpeg | third_parties.cpp:111  | [AVFilter: 0x7f696eb8de40]  -> CUDA_ERROR_INVALID_PTX: a PTX JIT compilation failed
[2023-03-02 15:51:44.295] E [Dech264NV:3360] FFmpeg | third_parties.cpp:111  | [AVFilter: 0x7f696eb8de40]
[2023-03-02 15:51:44.295] D [Dech264NV:3360] FFmpeg | third_parties.cpp:125  | [AVFilter: 0x7f696eb8de40] Calling cu->cuCtxPopCurrent(&dummy)
[2023-03-02 15:51:44.295] E [Dech264NV:3360] FFmpeg | third_parties.cpp:111  | [AVFilter: 0x7f696eb8de40] Failed to configure output pad on Parsed_scale_cuda_2
[2023-03-02 15:51:44.295] E [Dech264NV:3360] Transcoder | filter_rescaler.cpp:162  | Could not validate filter graph for rescaling: -542398533
[2023-03-02 15:51:44.295] E [Dech264NV:3360] Transcoder | transcoder_filter.cpp:63   | Could not craete filter
[2023-03-02 15:51:44.295] E [Dech264NV:3360] Transcoder | transcoder_stream.cpp:773  | [#default#app/stream(102)] Failed to create filter. Filter(2)
[2023-03-02 15:51:44.295] D [Dech264NV:3360] FFmpeg | third_parties.cpp:125  | [AVHWDeviceContext: 0x7f696ea5d380] Calling cu->cuCtxPushCurrent(hwctx->cuda_ctx)
[2023-03-02 15:51:44.296] D [Dech264NV:3360] FFmpeg | third_parties.cpp:125  | [AVHWDeviceContext: 0x7f696ea5d380] Calling cu->cuMemFree((CUdeviceptr)data)
[2023-03-02 15:51:44.296] D [Dech264NV:3360] FFmpeg | third_parties.cpp:125  | [AVHWDeviceContext: 0x7f696ea5d380] Calling cu->cuCtxPopCurrent(&dummy)
[2023-03-02 15:51:44.297] D [Resampler:3363] FFmpeg | third_parties.cpp:125  | [AVFilter: 0x7f696e0d9300] tb:1/1000 pts:1920 -> tb:1/48000 pts:92160
[2023-03-02 15:51:44.331] D [Dech264NV:3360] FFmpeg | third_parties.cpp:125  | [AVHWDeviceContext: 0x7f696ea5d380] Calling cu->cuCtxDestroy(hwctx->cuda_ctx)
[2023-03-02 15:51:44.331] D [Dech264NV:3360] FFmpeg | third_parties.cpp:125  | [AVFilter: 0x7f696eb1d900] 0 frames in, 0 frames out; 0 frames dropped, 0 frames duplicated.

0 replies

Maypeur · 2023-03-02T16:09:02Z

Maypeur
Mar 2, 2023

From 1920x1080 to 1280x720

[2023-03-02 16:06:39.632] D [Dech264NV:3482] FFmpeg | third_parties.cpp:125  | [AVFilter: 0x7f35770b0bc0] Setting 'w' to value '1280'
[2023-03-02 16:06:39.632] D [Dech264NV:3482] FFmpeg | third_parties.cpp:125  | [AVFilter: 0x7f35770b0bc0] Setting 'h' to value '720'
[2023-03-02 16:06:39.633] D [Dech264NV:3482] FFmpeg | third_parties.cpp:125  | [AVFilter: 0x7f35770b0e00] Setting 'expr' to value '1/90000'
[2023-03-02 16:06:39.633] D [Dech264NV:3482] FFmpeg | third_parties.cpp:125  | [AVFilterGraph: 0x7f357705b580] query_formats: 7 queried, 6 merged, 0 already done, 0 delayed
[2023-03-02 16:06:39.633] D [Dech264NV:3482] FFmpeg | third_parties.cpp:125  | [AVFilter: 0x7f357244d4c0] fps=30/1
[2023-03-02 16:06:39.633] D [Dech264NV:3482] FFmpeg | third_parties.cpp:125  | [AVHWDeviceContext: 0x7f357278bf00] Calling cu->cuDeviceGetAttribute(&priv->tex_alignment, 14 , hwctx->internal->cuda_device)
[2023-03-02 16:06:39.633] D [Dech264NV:3482] FFmpeg | third_parties.cpp:125  | [AVHWFramesContext: 0x7f3575fc4c80] CUDA texture alignment: 512
[2023-03-02 16:06:39.633] D [Dech264NV:3482] FFmpeg | third_parties.cpp:125  | [AVHWDeviceContext: 0x7f357278bf00] Calling cu->cuDeviceGetAttribute(&priv->tex_alignment, 14 , hwctx->internal->cuda_device)
[2023-03-02 16:06:39.633] D [Dech264NV:3482] FFmpeg | third_parties.cpp:125  | [AVHWFramesContext: 0x7f3575fc4d80] CUDA texture alignment: 512
[2023-03-02 16:06:39.633] D [Dech264NV:3482] FFmpeg | third_parties.cpp:125  | [AVHWDeviceContext: 0x7f357278bf00] Calling cu->cuCtxPushCurrent(hwctx->cuda_ctx)
[2023-03-02 16:06:39.633] D [Dech264NV:3482] FFmpeg | third_parties.cpp:125  | [AVHWDeviceContext: 0x7f357278bf00] Calling cu->cuMemAlloc(&data, size)
[2023-03-02 16:06:39.633] D [Dech264NV:3482] FFmpeg | third_parties.cpp:125  | [AVHWDeviceContext: 0x7f357278bf00] Calling cu->cuCtxPopCurrent(&dummy)
[2023-03-02 16:06:39.633] D [Dech264NV:3482] FFmpeg | third_parties.cpp:125  | [AVFilter: 0x7f35770b0bc0] w:1920 h:1080 fmt:nv12 -> w:1280 h:720 fmt:nv12
[2023-03-02 16:06:39.633] D [Dech264NV:3482] FFmpeg | third_parties.cpp:125  | [AVFilter: 0x7f35770b0bc0] Calling cu->cuCtxPushCurrent(cuda_ctx)
[2023-03-02 16:06:40.006] D [Dech264NV:3482] FFmpeg | third_parties.cpp:125  | [AVFilter: 0x7f35770b0bc0] Calling cu->cuModuleLoadData(cu_module, buf)
[2023-03-02 16:06:40.006] E [Dech264NV:3482] FFmpeg | third_parties.cpp:111  | [AVFilter: 0x7f35770b0bc0] cu->cuModuleLoadData(cu_module, buf) failed
[2023-03-02 16:06:40.006] E [Dech264NV:3482] FFmpeg | third_parties.cpp:111  | [AVFilter: 0x7f35770b0bc0]  -> CUDA_ERROR_INVALID_PTX: a PTX JIT compilation failed
[2023-03-02 16:06:40.006] E [Dech264NV:3482] FFmpeg | third_parties.cpp:111  | [AVFilter: 0x7f35770b0bc0]
[2023-03-02 16:06:40.006] D [Dech264NV:3482] FFmpeg | third_parties.cpp:125  | [AVFilter: 0x7f35770b0bc0] Calling cu->cuCtxPopCurrent(&dummy)
[2023-03-02 16:06:40.006] E [Dech264NV:3482] FFmpeg | third_parties.cpp:111  | [AVFilter: 0x7f35770b0bc0] Failed to configure output pad on Parsed_scale_cuda_2
[2023-03-02 16:06:40.006] E [Dech264NV:3482] Transcoder | filter_rescaler.cpp:162  | Could not validate filter graph for rescaling: -542398533
[2023-03-02 16:06:40.006] E [Dech264NV:3482] Transcoder | transcoder_filter.cpp:63   | Could not craete filter
[2023-03-02 16:06:40.006] E [Dech264NV:3482] Transcoder | transcoder_stream.cpp:773  | [#default#app/stream(101)] Failed to create filter. Filter(1)
[2023-03-02 16:06:40.006] D [Dech264NV:3482] FFmpeg | third_parties.cpp:125  | [AVHWDeviceContext: 0x7f357278bf00] Calling cu->cuCtxPushCurrent(hwctx->cuda_ctx)
[2023-03-02 16:06:41.007] D [Dech264NV:3482] FFmpeg | third_parties.cpp:125  | [AVHWDeviceContext: 0x7f357278bf00] Calling cu->cuMemFree((CUdeviceptr)data)
[2023-03-02 16:06:41.007] D [Dech264NV:3482] FFmpeg | third_parties.cpp:125  | [AVHWDeviceContext: 0x7f357278bf00] Calling cu->cuCtxPopCurrent(&dummy)
[2023-03-02 16:06:41.037] D [Dech264NV:3482] FFmpeg | third_parties.cpp:125  | [AVHWDeviceContext: 0x7f357278bf00] Calling cu->cuCtxDestroy(hwctx->cuda_ctx)
[2023-03-02 16:06:41.037] D [Dech264NV:3482] FFmpeg | third_parties.cpp:125  | [AVFilter: 0x7f357244d4c0] 0 frames in, 0 frames out; 0 frames dropped, 0 frames duplicated.

0 replies

Keukhan · 2023-03-02T16:18:32Z

Keukhan
Mar 2, 2023
Maintainer Sponsor

@Maypeur

What is your test server environment? Are you using a GPU in the cloud? Like @Pernifloss , the same problem is occurring on the Tesla M6 model.

I'm sorry. I don't have enough time to solve this problem right now, but I'll check it out as soon as possible.

Wait for the good news.

Thanks.

0 replies

Maypeur · 2023-03-02T16:22:52Z

Maypeur
Mar 2, 2023

Thks for your feedback.
It is the same server as @Pernifloss
Do not hesitate to ask for test or give us a way for changing any parameters on stream or in .cpp files.

0 replies

fairbairn · 2023-05-05T00:24:21Z

fairbairn
May 5, 2023

Similar issue.

We were trying to build a docker container to run on an Ubuntu 20.04 host with NVIDIA T4 530 and CUDA 12.1, and the ffmpeg said the CUDA version was out of sync (within the container).

So now we're attempting to compile from source to run natively on the host, but not expecting it to work based on the above.

1 reply

fairbairn May 5, 2023

Quick update - we have not yet tackled the update to the Dockerfile to make this work, but we were able to compile OME from source on our existing Ubuntu 22.04 host with NVIDIA 530 and CUDA 12, and it worked.

Our next task is to go through the build steps and see if we can modify the Dockerfile to work specifically for this deployment, representing a common LTS of Ubuntu, the latest NVIDIA, and the latest CUDA.

nums · 2023-06-27T14:17:38Z

nums
Jun 27, 2023

On my side, I'm trying to compile OvenMediaEngine in a Docker with:

ubuntu 22.04
NVIDIA-SMI 525.85.12
CUDA Version: 12.0

My host is "Ubuntu Jammy GPU OS 12" with a Geoforce RTX 3070

and I'm falling with this error: ERROR: failed checking for nvcc

I have tried to update NVCC_HDR_VERSION to 12.0.16.0 but issue still there.
And I have to tried to downgrade GCC version but the script failed before.

When I test with a apt install ffmpeg the encoding is working well with h264_nvenc.

Any suggestions ?

Thanks

2 replies

mpisat Jul 1, 2023

did you install cuda-toolkit-12-0 and nvidia-cuda-toolkit with apt?

mpisat Jul 1, 2023

also I recommend nvidia-530 driver instead of 525, which uses cuda-toolkit-12-1

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Ubuntu 22.04 : OME Transcoding using GPU #1115

{{title}}

Replies: 10 comments 3 replies

{{title}}

{{title}}

{{title}}

{{title}}

{{title}}

{{title}}

{{title}}

{{title}}

{{title}}

{{title}}

{{editor}}'s edit

{{editor}}'s edit

{{title}}

{{title}}

{{title}}

Select a reply

Ubuntu 22.04 : OME Transcoding using GPU #1115

Replies: 10 comments · 3 replies

Keukhan Feb 15, 2023 Maintainer Sponsor

Pernifloss Feb 15, 2023 Author

Pernifloss Feb 23, 2023 Author

Keukhan Mar 2, 2023 Maintainer Sponsor

Replies: 10 comments 3 replies

Keukhan
Feb 15, 2023
Maintainer Sponsor

Pernifloss
Feb 15, 2023
Author

Pernifloss
Feb 23, 2023
Author

Keukhan
Mar 2, 2023
Maintainer Sponsor