Prefer nvidia channel for conda builds

[malfet]

Fixes #5635

[facebook-github-bot]

💊 CI failures summary and remediations

As of commit 876bb30 (more details on the Dr. CI page):

---

• 3/3 failures introduced in this PR

---

🕵️ 3 new failures recognized by patterns

The following CI failures do not appear to be due to upstream breakages:

binary_linux_conda_py3.8_cu115 (1/3)

Step: "packaging/build_conda.sh" (full log | diagnosis details | 🔁 rerun)

$SRC_DIR/torchvision/csrc/io/decoder/decoder.cp...tFormat*’ to ‘AVInputFormat*’ [-fpermissive]

          ^~~~~~~~~~~~~~~
$SRC_DIR/torchvision/csrc/io/decoder/decoder.cpp:33:10: note: suggested alternative: ‘AV_LOG_ERROR’
     case AV_LOCK_DESTROY:
          ^~~~~~~~~~~~~~~
          AV_LOG_ERROR
$SRC_DIR/torchvision/csrc/io/decoder/decoder.cpp: In lambda function:
$SRC_DIR/torchvision/csrc/io/decoder/decoder.cpp:206:5: error: ‘av_lockmgr_register’ was not declared in this scope
     av_lockmgr_register(&ffmpeg_lock);
     ^~~~~~~~~~~~~~~~~~~
$SRC_DIR/torchvision/csrc/io/decoder/decoder.cpp: In member function ‘virtual bool ffmpeg::Decoder::init(const ffmpeg::DecoderParameters&, ffmpeg::DecoderInCallback&&, std::vector<ffmpeg::DecoderMetadata>*)’:
$SRC_DIR/torchvision/csrc/io/decoder/decoder.cpp:280:33: error: invalid conversion from ‘const AVInputFormat*’ to ‘AVInputFormat*’ [-fpermissive]
       fmt = av_find_input_format(fmtName);
             ~~~~~~~~~~~~~~~~~~~~^~~~~~~~~
$SRC_DIR/torchvision/csrc/io/decoder/decoder.cpp: In member function ‘int ffmpeg::Decoder::getFrame(size_t)’:
$SRC_DIR/torchvision/csrc/io/decoder/decoder.cpp:509:27: warning: ‘void av_init_packet(AVPacket*)’ is deprecated [-Wdeprecated-declarations]
   av_init_packet(&avPacket);
                           ^
In file included from $BUILD_PREFIX/include/libavcodec/avcodec.h:45:0,
                 from $SRC_DIR/torchvision/csrc/io/decoder/defs.h:12,
                 from $SRC_DIR/torchvision/csrc/io/decoder/seekable_buffer.h:3,
                 from $SRC_DIR/torchvision/csrc/io/decoder/decoder.h:5,

binary_linux_conda_py3.10_cu115 (2/3)

Step: "packaging/build_conda.sh" (full log | diagnosis details | 🔁 rerun)

$SRC_DIR/torchvision/csrc/io/decoder/decoder.cp...tFormat*’ to ‘AVInputFormat*’ [-fpermissive]

          ^~~~~~~~~~~~~~~
$SRC_DIR/torchvision/csrc/io/decoder/decoder.cpp:33:10: note: suggested alternative: ‘AV_LOG_ERROR’
     case AV_LOCK_DESTROY:
          ^~~~~~~~~~~~~~~
          AV_LOG_ERROR
$SRC_DIR/torchvision/csrc/io/decoder/decoder.cpp: In lambda function:
$SRC_DIR/torchvision/csrc/io/decoder/decoder.cpp:206:5: error: ‘av_lockmgr_register’ was not declared in this scope
     av_lockmgr_register(&ffmpeg_lock);
     ^~~~~~~~~~~~~~~~~~~
$SRC_DIR/torchvision/csrc/io/decoder/decoder.cpp: In member function ‘virtual bool ffmpeg::Decoder::init(const ffmpeg::DecoderParameters&, ffmpeg::DecoderInCallback&&, std::vector<ffmpeg::DecoderMetadata>*)’:
$SRC_DIR/torchvision/csrc/io/decoder/decoder.cpp:280:33: error: invalid conversion from ‘const AVInputFormat*’ to ‘AVInputFormat*’ [-fpermissive]
       fmt = av_find_input_format(fmtName);
             ~~~~~~~~~~~~~~~~~~~~^~~~~~~~~
$SRC_DIR/torchvision/csrc/io/decoder/decoder.cpp: In member function ‘int ffmpeg::Decoder::getFrame(size_t)’:
$SRC_DIR/torchvision/csrc/io/decoder/decoder.cpp:509:27: warning: ‘void av_init_packet(AVPacket*)’ is deprecated [-Wdeprecated-declarations]
   av_init_packet(&avPacket);
                           ^
In file included from $BUILD_PREFIX/include/libavcodec/avcodec.h:45:0,
                 from $SRC_DIR/torchvision/csrc/io/decoder/defs.h:12,
                 from $SRC_DIR/torchvision/csrc/io/decoder/seekable_buffer.h:3,
                 from $SRC_DIR/torchvision/csrc/io/decoder/decoder.h:5,

binary_linux_conda_py3.7_cu115 (3/3)

Step: "packaging/build_conda.sh" (full log | diagnosis details | 🔁 rerun)

error_prefix='Error compiling objects for extension')

    self._build_extensions_serial()
  File "/opt/conda/conda-bld/torchvision_1647881675966/_h_env_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_p/lib/python3.7/site-packages/setuptools/_distutils/command/build_ext.py", line 473, in _build_extensions_serial
    self.build_extension(ext)
  File "/opt/conda/conda-bld/torchvision_1647881675966/_h_env_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_p/lib/python3.7/site-packages/setuptools/command/build_ext.py", line 202, in build_extension
    _build_ext.build_extension(self, ext)
  File "/opt/conda/conda-bld/torchvision_1647881675966/_h_env_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_p/lib/python3.7/site-packages/setuptools/_distutils/command/build_ext.py", line 534, in build_extension
    depends=ext.depends)
  File "/opt/conda/conda-bld/torchvision_1647881675966/_h_env_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_p/lib/python3.7/site-packages/torch/utils/cpp_extension.py", line 593, in unix_wrap_ninja_compile
    with_cuda=with_cuda)
  File "/opt/conda/conda-bld/torchvision_1647881675966/_h_env_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_p/lib/python3.7/site-packages/torch/utils/cpp_extension.py", line 1473, in _write_ninja_file_and_compile_objects
    error_prefix='Error compiling objects for extension')
  File "/opt/conda/conda-bld/torchvision_1647881675966/_h_env_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_p/lib/python3.7/site-packages/torch/utils/cpp_extension.py", line 1805, in _run_ninja_build
    raise RuntimeError(message) from e
RuntimeError: Error compiling objects for extension
Traceback (most recent call last):
  File "/opt/conda/bin/conda-build", line 11, in <module>
    sys.exit(main())
  File "/opt/conda/lib/python3.9/site-packages/conda_build/cli/main_build.py", line 488, in main
    execute(sys.argv[1:])
  File "/opt/conda/lib/python3.9/site-packages/conda_build/cli/main_build.py", line 477, in execute
    outputs = api.build(args.recipe, post=args.post, test_run_post=args.test_run_post,

---

This comment was automatically generated by Dr. CI (expand for details).

Please report bugs/suggestions to the (internal) Dr. CI Users group.

Click here to manually regenerate this comment.

[datumbox]

Thanks @malfet.

Is this a temporary update to fix the issue or does it mean we will change our recommended channel on the download page at pytorch.org?

[datumbox]

@malfet The issue persists on the cmake_linux_gpu job:
ImportError: libcupti.so.11.3: cannot open shared object file: No such file or directory

The binary_linux_conda_*_cu115 jobs are failing because they use ffmpeg-5.0.0 which deprecates some functionality that we currently use. There is another PR that attempts to fix this at #5644.

I think we should merge this after the cmake job is also fixed and then resolve the rest of the jobs on the other PR.

1 job still failing

[datumbox]

LGTM, thanks for patching cmake.

Let's merge on "green-ish" CI.

[github-actions]

Hey @malfet!

You merged this PR, but no labels were added. The list of valid labels is available at https://github.com/pytorch/vision/blob/main/.github/process_commit.py