New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
nvcuvid.h: No such file or directory #12
Comments
Hi Luyang, I have heard that CUDA 9 changed the setup of the SDK, as mentioned in issue #11. If you could confirm that the header is not present on your system, and furthermore that it is installed as part of a CUDA 8 SDK, that would clarify. Of course, patches would be welcome too! |
CUDA >= 9.0 no longer ships nvcuvid.h so depending on the value of CUDART_VERSION we include dynlink_nvcuvid.h instead. Fixes NVIDIA#12.
I can confirm that nvcuvid.h used to be shipped in CUDA < 9.0 but is gone now. Starting with CUDA 9.0 there's a dynlink_nvcuvid.h instead which appears to be API compatible. I've created a pull request to use this depending on the CUDA version. I cannot test decoding, unfortunately, but encoding still works fine with either CUDA 8.0 or 9.1 on a Quadro K2000M and GTX 1080. |
Hi! I have a related question. I have downloaded the latest video SDK. And for NVDEC there is no dynlink_nvcuvid anymore but just nvcuvid. And there is a also a static lib for Windows only. What does that mean? How do I link on Linux? Do the latest drivers for linux come with static lib for nvcuvid? |
I've just pushed a redesigned NvPipe which works with CUDA 9 and is based on the reusable NvCodec classes in the Video Codec SDK. It uses the nvcuvid.h header and compile-time linking against libnvcuvid. |
I cloned this ffmpeg_libyami repo and copied the 2 cuda headers directly under /usr/include/ https://github.com/intel/ffmpeg_libyami/tree/master/compat/cuda And then I installed all the ffmpeg -dev libraries with apt-get. The hack worked. The compilation found /usr/include/nvcuid.h and proceeded the compination through to the end. |
Hi,
I got the error when compiling this project on Ubuntu 16.04: nvcuvid.h: No such file or directory. Could you let me know how to solve this issue? @tfogal
The detail logs shown on the terminal is here:
-- The C compiler identification is GNU 5.4.0
-- The CXX compiler identification is GNU 5.4.0
-- Check for working C compiler: /usr/bin/cc
-- Check for working C compiler: /usr/bin/cc -- works
-- Detecting C compiler ABI info
-- Detecting C compiler ABI info - done
-- Detecting C compile features
-- Detecting C compile features - done
-- Check for working CXX compiler: /usr/bin/c++
-- Check for working CXX compiler: /usr/bin/c++ -- works
-- Detecting CXX compiler ABI info
-- Detecting CXX compiler ABI info - done
-- Detecting CXX compile features
-- Detecting CXX compile features - done
-- Found CUDA: /usr/local/cuda (found version "9.0")
-- Found PkgConfig: /usr/bin/pkg-config (found version "0.29.1")
-- Could NOT find EGL (missing: EGL_LIBRARY EGL_opengl_LIBRARY EGL_INCLUDE_DIR)
-- Could NOT find GLEW (missing: GLEW_INCLUDE_DIR GLEW_LIBRARY)
-- Configuring done
-- Generating done
-- Build files have been written to: /home/luyang/Documents/NvPipe-master/build
luyang@luyang-ubuntu:~/Documents/NvPipe-master/build$ make
[ 12%] Building NVCC (Device) object CMakeFiles/nvpipe.dir/convert.o
Scanning dependencies of target nvpipe
[ 25%] Building C object CMakeFiles/nvpipe.dir/debug.c.o
[ 37%] Building C object CMakeFiles/nvpipe.dir/decode.c.o
/home/luyang/Documents/NvPipe-master/decode.c:44:21: fatal error: nvcuvid.h: No such file or directory
compilation terminated.
CMakeFiles/nvpipe.dir/build.make:93: recipe for target 'CMakeFiles/nvpipe.dir/decode.c.o' failed
make[2]: *** [CMakeFiles/nvpipe.dir/decode.c.o] Error 1
CMakeFiles/Makefile2:67: recipe for target 'CMakeFiles/nvpipe.dir/all' failed
make[1]: *** [CMakeFiles/nvpipe.dir/all] Error 2
Makefile:127: recipe for target 'all' failed
make: *** [all] Error 2
The text was updated successfully, but these errors were encountered: