FFMPEG Using GTX titan GeForce GTX TITAN X(M200) fail?no CUDA-capable device is detected?

NVIDIA-SMI 384.111 Driver Version: 384.111
CUDA:9.0
FFMPEG:3.4.6
nv-codec-headers:SDK 8.0
I have 4 GeForce GTX TITAN X(M200).I found in the Support Matrix,GeForce GTX TITAN X(M200) support H.264.

info:
/usr/local/ffmpeg_gpu/bin/ffmpeg -y -vsync 0 -hwaccel cuvid -c:v h264_cuvid -i nanjing7_cut.mp4 -c:a copy -c:v h264_nvenc -b:v 5M nanjing8_cut.mp4

Input #0, mov,mp4,m4a,3gp,3g2,mj2, from ‘nanjing7_cut.mp4’:
Metadata:
major_brand : isom
minor_version : 512
compatible_brands: isomiso2avc1mp41
encoder : Lavf55.19.104
Duration: 00:00:08.26, start: 0.000000, bitrate: 13210 kb/s
Stream #0:0(und): Video: h264 (High) (avc1 / 0x31637661), yuvj420p(pc, bt470bg/bt470bg/smpte170m), 1920x1080, 13207 kb/s, 50 fps, 50 tbr, 12800 tbn, 50 tbc (default)
Metadata:
handler_name : VideoHandler
[h264_cuvid @ 0x24529c0] ctx->cvdl->cuvidGetDecoderCaps(&ctx->caps8) failed -> CUDA_ERROR_NO_DEVICE: no CUDA-capable device is detected
[h264_cuvid @ 0x24529c0] ctx->cvdl->cuvidGetDecoderCaps(&ctx->caps10) failed -> CUDA_ERROR_NO_DEVICE: no CUDA-capable device is detected
[h264_cuvid @ 0x24529c0] ctx->cvdl->cuvidGetDecoderCaps(&ctx->caps12) failed -> CUDA_ERROR_NO_DEVICE: no CUDA-capable device is detected

[root@abbe8c1bc066 cuda-9.0]# lspci | grep NVIDIA
02:00.0 VGA compatible controller: NVIDIA Corporation GM200 [GeForce GTX TITAN X] (rev a1)
02:00.1 Audio device: NVIDIA Corporation GM200 High Definition Audio (rev a1)
03:00.0 VGA compatible controller: NVIDIA Corporation GM200 [GeForce GTX TITAN X] (rev a1)
03:00.1 Audio device: NVIDIA Corporation GM200 High Definition Audio (rev a1)
82:00.0 VGA compatible controller: NVIDIA Corporation GM200 [GeForce GTX TITAN X] (rev a1)
82:00.1 Audio device: NVIDIA Corporation GM200 High Definition Audio (rev a1)
83:00.0 VGA compatible controller: NVIDIA Corporation GM200 [GeForce GTX TITAN X] (rev a1)
83:00.1 Audio device: NVIDIA Corporation GM200 High Definition Audio (rev a1)

[root@abbe8c1bc066 ffmpeg-3.4.6]# ldd /usr/local/ffmpeg_gpu/bin/ffmpeg
linux-vdso.so.1 => (0x00007fffd939b000)
libavdevice.so.57 => /usr/local/ffmpeg_gpu/lib/libavdevice.so.57 (0x00007f5c70754000)
libavfilter.so.6 => /usr/local/ffmpeg_gpu/lib/libavfilter.so.6 (0x00007f5c702dc000)
libavformat.so.57 => /usr/local/ffmpeg_gpu/lib/libavformat.so.57 (0x00007f5c6fead000)
libavcodec.so.57 => /usr/local/ffmpeg_gpu/lib/libavcodec.so.57 (0x00007f5c6e5b0000)
libpostproc.so.54 => /usr/local/ffmpeg_gpu/lib/libpostproc.so.54 (0x00007f5c6e38f000)
libswresample.so.2 => /usr/local/ffmpeg_gpu/lib/libswresample.so.2 (0x00007f5c6e172000)
libswscale.so.4 => /usr/local/ffmpeg_gpu/lib/libswscale.so.4 (0x00007f5c6deeb000)
libavutil.so.55 => /usr/local/ffmpeg_gpu/lib/libavutil.so.55 (0x00007f5c6dc6c000)
libm.so.6 => /lib64/libm.so.6 (0x00007f5c6d96a000)
libpthread.so.0 => /lib64/libpthread.so.0 (0x00007f5c6d74e000)
libc.so.6 => /lib64/libc.so.6 (0x00007f5c6d380000)
libxcb.so.1 => /lib64/libxcb.so.1 (0x00007f5c6d158000)
libxcb-shm.so.0 => /lib64/libxcb-shm.so.0 (0x00007f5c6cf54000)
libxcb-xfixes.so.0 => /lib64/libxcb-xfixes.so.0 (0x00007f5c6cd4b000)
libxcb-shape.so.0 => /lib64/libxcb-shape.so.0 (0x00007f5c6cb47000)
libnppig.so.9.0 => /usr/local/cuda/lib64/libnppig.so.9.0 (0x00007f5c6aee1000)
libnppicc.so.9.0 => /usr/local/cuda/lib64/libnppicc.so.9.0 (0x00007f5c6a8d5000)
libcuda.so.1 => /usr/local/nvidia/lib64/libcuda.so.1 (0x00007f5c69a57000)
libbz2.so.1 => /lib64/libbz2.so.1 (0x00007f5c69847000)
libz.so.1 => /lib64/libz.so.1 (0x00007f5c69630000)
libdl.so.2 => /lib64/libdl.so.2 (0x00007f5c6942c000)
liblzma.so.5 => /lib64/liblzma.so.5 (0x00007f5c69206000)
/lib64/ld-linux-x86-64.so.2 (0x00007f5c7097d000)
libXau.so.6 => /lib64/libXau.so.6 (0x00007f5c69001000)
libnppc.so.9.0 => /usr/local/cuda/lib64/libnppc.so.9.0 (0x00007f5c68d89000)
librt.so.1 => /lib64/librt.so.1 (0x00007f5c68b80000)
libstdc++.so.6 => /lib64/libstdc++.so.6 (0x00007f5c68879000)
libgcc_s.so.1 => /lib64/libgcc_s.so.1 (0x00007f5c68663000)
libnvidia-fatbinaryloader.so.384.111 => /usr/local/nvidia/lib64/libnvidia-fatbinaryloader.so.384.111 (0x00007f5c68410000)

Hi.
We are not able to reproduce this issue based on the information you provided. Is there anything else needed to reproduce this issue, that might be missed in the comment?

Please capture nvidia-bug-report by running script nvidia-bug-report.sh with root privileges after issue is reproduced and share it. The script should be already available on your system as a part of NVIDIA display driver installation.

Can you run some more isolation experiments like running with only one GPU connected to the system, does issue reproduce with SDK samples on latest supported driver and SDK?

Thanks.

Hi yangcheng830117,

Do you still see this issue? We are not be able to make any progress on this report until further details are available on how to reproduce this issue.

Thanks.

i meet this error too

RTX2080, cuda 10.1, nvidia-440.33.01, and build ffmpeg 4.2 from source,

stream webcam to rtsp with nvenc

env LD_LIBRARY_PATH=./build/lib/ ./build/bin/ffmpeg -hwaccel cuvid -f v4l2 -i /dev/video0 -c:v h264_nvenc -rtsp_transport tcp -f rtsp rtsp://127.0.0.1:1235

and ffplay decode stream from rtsp

env LD_LIBRARY_PATH=./build/lib/ ./build/bin/ffplay -vcodec h264_cuvid -rtsp_transport tcp  -rtsp_flags listen rtsp://127.0.0.1:1235

ffaply throw errors

ffplay version n4.2.1-9-g82a3a62 Copyright (c) 2003-2019 the FFmpeg developers
  built with gcc 5.4.0 (Ubuntu 5.4.0-6ubuntu1~16.04.11) 20160609
  configuration: --enable-cuda-nvcc --enable-cuvid --enable-nvenc --enable-nonfree --enable-libnpp --extra-cflags=-I/usr/local/cuda-10.1/include --extra-ldflags=-L/usr/local/cuda-10.1/lib64 --prefix=/opt/ffmpeg/build/ --enable-pic --enable-shared --disable-static --enable-sdl2
  libavutil      56. 31.100 / 56. 31.100
  libavcodec     58. 54.100 / 58. 54.100
  libavformat    58. 29.100 / 58. 29.100
  libavdevice    58.  8.100 / 58.  8.100
  libavfilter     7. 57.100 /  7. 57.100
  libswscale      5.  5.100 /  5.  5.100
  libswresample   3.  5.100 /  3.  5.100
Input #0, rtsp, from 'rtsp://127.0.0.1:1235': 0KB sq=    0B f=0/0   
  Metadata:
    title           : No Name
  Duration: N/A, start: 0.157000, bitrate: N/A
    Stream #0:0: Video: h264 (High 4:4:4 Predictive), yuv444p(progressive), 640x480 [SAR 1:1 DAR 4:3], 30 fps, 30 tbr, 90k tbn, 60 tbc
[h264_cuvid @ 0x7fc91401f600] ctx->cvdl->cuvidCreateDecoder(&ctx->cudecoder, &cuinfo) failed -> CUDA_ERROR_NO_DEVICE: no CUDA-capable device is detected
[h264_cuvid @ 0x7fc91401f600] ctx->cvdl->cuvidDecodePicture(ctx->cudecoder, picparams) failed -> CUDA_ERROR_INVALID_HANDLE: invalid resource handle
[h264_cuvid @ 0x7fc91401f600] cuvid decode callback error
    Last message repeated 1 times
[h264_cuvid @ 0x7fc91401f600] ctx->cvdl->cuvidDecodePicture(ctx->cudecoder, picparams) failed -> CUDA_ERROR_INVALID_HANDLE: invalid resource handle
[h264_cuvid @ 0x7fc91401f600] ctx->cvdl->cuvidDecodePicture(ctx->cudecoder, picparams) failed -> CUDA_ERROR_INVALID_HANDLE: invalid resource handle
[h264_cuvid @ 0x7fc91401f600] cuvid decode callback error
[h264_cuvid @ 0x7fc91401f600] ctx->cvdl->cuvidDecodePicture(ctx->cudecoder, picparams) failed -> CUDA_ERROR_INVALID_HANDLE: invalid resource handle
[h264_cuvid @ 0x7fc91401f600] cuvid decode callback error
[h264_cuvid @ 0x7fc91401f600] ctx->cvdl->cuvidDecodePicture(ctx->cudecoder, picparams) failed -> CUDA_ERROR_INVALID_HANDLE: invalid resource handle
[h264_cuvid @ 0x7fc91401f600] cuvid decode callback error

This thing is the worst. Even in Jetson Xavier NX after building FFmpeg seeing the issue. To NVIDIA, please maintain proper quality. I have followed the instruction that has been provided on the site. Spend 2 days on this and no success. Moreover, those encoding decoding libraries are essential for any AI work. That should come by default with the OS itself. Why do I have to manually copy libnvcuvid.so