Build Mac wheels in CI (#230)

Co-authored-by: Andrey Talman <[email protected]>
Co-authored-by: Huy Do <[email protected]>

.github/workflows/linux_wheel.yaml

@@ -1,4 +1,4 @@
-name: Build and test Linux wheels
+name: Build and test Linux wheel
 
 on:
   pull_request:
@@ -24,6 +24,7 @@ defaults:
     shell: bash -l -eo pipefail {0}
 
 jobs:
+
   generate-matrix:
     uses: pytorch/test-infra/.github/workflows/generate_binary_build_matrix.yml@main
     with:
@@ -34,11 +35,13 @@ jobs:
       with-xpu: disable
       with-rocm: disable
       with-cuda: disable
+      build-python-only: "disable"
+
   build:
     needs: generate-matrix
     strategy:
       fail-fast: false
-    name: Build and Upload wheel
+    name: Build and Upload Linux wheel
     uses: pytorch/test-infra/.github/workflows/build_wheels_linux.yml@main
     with:
       repository: pytorch/torchcodec
@@ -60,7 +63,6 @@ jobs:
       matrix:
         python-version: ['3.9']
         ffmpeg-version-for-tests: ['4.4.2', '5.1.2', '6.1.1', '7.0.1']
-    if: ${{ always() }}
     needs: build
     steps:
       - uses: actions/download-artifact@v3
@@ -121,7 +123,7 @@ jobs:
           ls
       - name: Smoke test
         run: |
-          python test/decoders/manual_smoke_test.py
+          python -X faulthandler test/decoders/manual_smoke_test.py
       - name: Run Python tests
         run: |
           pytest test -vvv

.github/workflows/macos_wheel.yaml

@@ -1,4 +1,4 @@
-name: Build and test MacOS
+name: Build and test MacOS wheel
 
 on:
   pull_request:
@@ -24,15 +24,54 @@ defaults:
     shell: bash -l -eo pipefail {0}
 
 jobs:
+
+  generate-matrix:
+    uses: pytorch/test-infra/.github/workflows/generate_binary_build_matrix.yml@main
+    with:
+      package-type: wheel
+      os: macos-arm64
+      test-infra-repository: pytorch/test-infra
+      test-infra-ref: main
+      with-xpu: disable
+      with-rocm: disable
+      with-cuda: disable
+      build-python-only: "disable"
+
+  build:
+    needs: generate-matrix
+    strategy:
+      fail-fast: false
+    name: Build and Upload Mac wheel
+    uses: pytorch/test-infra/.github/workflows/build_wheels_macos.yml@main
+    with:
+      repository: pytorch/torchcodec
+      ref: ""
+      test-infra-repository: pytorch/test-infra
+      test-infra-ref: main
+      build-matrix: ${{ needs.generate-matrix.outputs.matrix }}
+      post-script: packaging/post_build_script.sh
+      smoke-test-script: packaging/fake_smoke_test.py
+      runner-type: macos-m1-stable
+      package-name: torchcodec
+      trigger-event: ${{ github.event_name }}
+      build-platform: "python-build-package"
+      build-command: "BUILD_AGAINST_ALL_FFMPEG_FROM_S3=1 python -m build --wheel -vvv --no-isolation"
+
   install-and-test:
-    runs-on: macos-m1-stable
+    runs-on: macos-14-xlarge
     strategy:
       fail-fast: false
       matrix:
         python-version: ['3.9']
         ffmpeg-version-for-tests: ['4.4.2', '5.1.2', '6.1.1', '7.0.1']
-    if: ${{ always() }}
+    needs: build
     steps:
+      - name: Download wheel
+        uses: actions/download-artifact@v3
+        with:
+          name: pytorch_torchcodec__${{ matrix.python-version }}_cpu_
+          path: pytorch/torchcodec/dist/
+
       - name: Setup conda env
         uses: conda-incubator/setup-miniconda@v3
         with:
@@ -42,36 +81,49 @@ jobs:
           python-version: ${{ matrix.python-version }}
       - name: Update pip
         run: python -m pip install --upgrade pip
+
       - name: Install PyTorch
         run: |
           python -m pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cpu
-      - name: Check out repo
+
+      - name: Install torchcodec from the wheel
+        run: |
+          wheel_path=`find pytorch/torchcodec/dist -type f -name "*.whl"`
+          echo Installing $wheel_path
+          python -m pip install $wheel_path -vvv
+
+      - name: Check out torchcodec repo
         uses: actions/checkout@v3
-      - name: Install compile from source dependencies
+
+      - name: Install ffmpeg
         run: |
-          conda install cmake pkg-config -c conda-forge
+          conda install "ffmpeg=${{ matrix.ffmpeg-version-for-tests }}" -c conda-forge
+          ffmpeg -version
+
       - name: Install test dependencies
         run: |
           python -m pip install --pre torchvision --index-url https://download.pytorch.org/whl/nightly/cpu
-          # Ideally we would find a way to get those dependencies from pyproject.toml
           python -m pip install numpy pytest pillow
-      - name: Install torchcodec from source, building against non-GPL FFmpeg
-        run: |
-          BUILD_AGAINST_ALL_FFMPEG_FROM_S3=1 pip install -e ".[dev]" --no-build-isolation
-      - name: Install ffmpeg, post build
+
+      - name: Delete the src/ folder just for fun
         run: |
-          # Ideally we would have checked for that before installing the wheel,
-          # but we need to checkout the repo to access this file, and we don't
-          # want to checkout the repo before installing the wheel to avoid any
-          # side-effect. It's OK.
-          source packaging/helpers.sh
-          assert_ffmpeg_not_installed
+          # The only reason we checked-out the repo is to get access to the
+          # tests. We don't care about the rest. Out of precaution, we delete
+          # the src/ folder to be extra sure that we're running the code from
+          # the installed wheel rather than from the source.
+          # This is just to be extra cautious and very overkill because a)
+          # there's no way the `torchcodec` package from src/ can be found from
+          # the PythonPath: the main point of `src/` is precisely to protect
+          # against that and b) if we ever were to execute code from
+          # `src/torchcodec`, it would fail loudly because the built .so files
+          # aren't present there.
+          rm -r src/
+          ls -lh
 
-          conda install "ffmpeg=${{ matrix.ffmpeg-version-for-tests }}" -c conda-forge
-          ffmpeg -version
       - name: Smoke test
         run: |
-          python test/decoders/manual_smoke_test.py
+          python -X faulthandler test/decoders/manual_smoke_test.py
+
       - name: Run Python tests
         run: |
           pytest test -vvv

packaging/check_glibcxx.py

@@ -38,14 +38,15 @@
 
 MAX_ALLOWED = (3, 4, 19)
 
+symbol_matches = sys.argv[1].split("\n")
 all_symbols = set()
-for line in sys.argv[1].split("\n"):
+for line in symbol_matches:
     # We search for GLIBCXX_major.minor.micro
     if match := re.search(r"GLIBCXX_\d+\.\d+\.\d+", line):
         all_symbols.add(match.group(0))
 
 if not all_symbols:
-    raise ValueError("No GLIBCXX symbols found. Something is wrong.")
+    raise ValueError(f"No GLIBCXX symbols found in {symbol_matches}. Something is wrong.")
 
 all_versions = (symbol.split("_")[1].split(".") for symbol in all_symbols)
 all_versions = (tuple(int(v) for v in version) for version in all_versions)

packaging/post_build_script.sh

@@ -1,45 +1,57 @@
 #!/bin/bash
 
+set -ex
+
 source packaging/helpers.sh
 
 wheel_path=$(pwd)/$(find dist -type f -name "*.whl")
 echo "Wheel content:"
 unzip -l $wheel_path
 
-ffmpeg_versions=(4 5 6 7)
+unamestr=$(uname)
+if [[ "$unamestr" == 'Linux' ]]; then
+    ext="so"
+elif [[ "$unamestr" == 'Darwin' ]]; then
+    ext="dylib"
+else
+    echo "Unknown operating system: $unamestr"
+    exit 1
+fi
 
 # TODO: Make ffmpeg4 work with nvcc.
-if [ "$ENABLE_CUDA" -eq 1 ]; then
-    ffmpeg_versions=(5 6 7)
+if [[ "$ENABLE_CUDA" -eq 1 ]]; then
+  ffmpeg_versions=(5 6 7)
 fi
 
-for ffmpeg_major_version in ${ffmepg_versions[@]}; do
-    assert_in_wheel $wheel_path torchcodec/libtorchcodec${ffmpeg_major_version}.so
+for ffmpeg_major_version in ${ffmpeg_versions[@]}; do
+    assert_in_wheel $wheel_path torchcodec/libtorchcodec${ffmpeg_major_version}.${ext}
 done
-assert_not_in_wheel $wheel_path libtorchcodec.so
+assert_not_in_wheel $wheel_path libtorchcodec.${ext}
 
-for ffmpeg_so in libavcodec.so libavfilter.so libavformat.so libavutil.so libavdevice.so ; do
-    assert_not_in_wheel $wheel_path $ffmpeg_so
+for ffmpeg_ext in libavcodec.${ext} libavfilter.${ext} libavformat.${ext} libavutil.${ext} libavdevice.${ext} ; do
+    assert_not_in_wheel $wheel_path $ffmpeg_ext
 done
 
 assert_not_in_wheel $wheel_path "^test"
 assert_not_in_wheel $wheel_path "^doc"
 assert_not_in_wheel $wheel_path "^benchmarks"
 assert_not_in_wheel $wheel_path "^packaging"
 
-# See invoked python script below for details about this check.
-extracted_wheel_dir=$(mktemp -d)
-unzip -q $wheel_path -d $extracted_wheel_dir
-symbols_matches=$(find $extracted_wheel_dir | grep ".so$" | xargs objdump --syms | grep GLIBCXX_3.4.)
-python packaging/check_glibcxx.py "$symbols_matches"
+if [[ "$unamestr" == 'Linux' ]]; then
+    # See invoked python script below for details about this check.
+    extracted_wheel_dir=$(mktemp -d)
+    unzip -q $wheel_path -d $extracted_wheel_dir
+    symbols_matches=$(find $extracted_wheel_dir | grep ".so$" | xargs objdump --syms | grep GLIBCXX_3.4.)
+    python packaging/check_glibcxx.py "$symbols_matches"
 
-echo "ls dist"
-ls dist
+    echo "ls dist"
+    ls dist
 
-old="linux_x86_64"
-new="manylinux_2_17_x86_64.manylinux2014_x86_64"
-echo "Replacing ${old} with ${new} in wheel name"
-mv dist/*${old}*.whl $(echo dist/*${old}*.whl | sed "s/${old}/${new}/")
+    old="linux_x86_64"
+    new="manylinux_2_17_x86_64.manylinux2014_x86_64"
+    echo "Replacing ${old} with ${new} in wheel name"
+    mv dist/*${old}*.whl $(echo dist/*${old}*.whl | sed "s/${old}/${new}/")
+fi
 
 echo "ls dist"
 ls dist

src/torchcodec/decoders/_core/VideoDecoder.cpp

@@ -676,9 +676,6 @@ void VideoDecoder::maybeSeekToBeforeDesiredPts() {
     int64_t desiredPtsForStream = *maybeDesiredPts_ * streamInfo.timeBase.den;
     if (!canWeAvoidSeekingForStream(
             streamInfo, streamInfo.currentPts, desiredPtsForStream)) {
-      VLOG(5) << "Seeking is needed for streamIndex=" << streamIndex
-              << " desiredPts=" << desiredPtsForStream
-              << " currentPts=" << streamInfo.currentPts;
       mustSeek = true;
       break;
     }
@@ -727,13 +724,10 @@ VideoDecoder::RawDecodedOutput VideoDecoder::getDecodedOutputWithFilter(
   if (activeStreamIndices_.size() == 0) {
     throw std::runtime_error("No active streams configured.");
   }
-  VLOG(9) << "Starting getDecodedOutputWithFilter()";
   resetDecodeStats();
   if (maybeDesiredPts_.has_value()) {
-    VLOG(9) << "maybeDesiredPts_=" << *maybeDesiredPts_;
     maybeSeekToBeforeDesiredPts();
     maybeDesiredPts_ = std::nullopt;
-    VLOG(9) << "seeking done";
   }
   auto seekDone = std::chrono::high_resolution_clock::now();
   // Need to get the next frame or error from PopFrame.
@@ -748,13 +742,9 @@ VideoDecoder::RawDecodedOutput VideoDecoder::getDecodedOutputWithFilter(
       StreamInfo& streamInfo = streams_[streamIndex];
       ffmpegStatus =
           avcodec_receive_frame(streamInfo.codecContext.get(), frame.get());
-      VLOG(9) << "received frame" << " status=" << ffmpegStatus
-              << " streamIndex=" << streamInfo.stream->index;
       bool gotNonRetriableError =
           ffmpegStatus != AVSUCCESS && ffmpegStatus != AVERROR(EAGAIN);
       if (gotNonRetriableError) {
-        VLOG(9) << "Got non-retriable error from decoder: "
-                << getFFMPEGErrorStringFromErrorCode(ffmpegStatus);
         gotPermanentErrorOnAnyActiveStream = true;
         break;
       }
@@ -784,7 +774,6 @@ VideoDecoder::RawDecodedOutput VideoDecoder::getDecodedOutputWithFilter(
     UniqueAVPacket packet(av_packet_alloc());
     ffmpegStatus = av_read_frame(formatContext_.get(), packet.get());
     decodeStats_.numPacketsRead++;
-    VLOG(9) << "av_read_frame returned status: " << ffmpegStatus;
     if (ffmpegStatus == AVERROR_EOF) {
       // End of file reached. We must drain all codecs by sending a nullptr
       // packet.
@@ -807,8 +796,6 @@ VideoDecoder::RawDecodedOutput VideoDecoder::getDecodedOutputWithFilter(
           "Could not read frame from input file: " +
           getFFMPEGErrorStringFromErrorCode(ffmpegStatus));
     }
-    VLOG(9) << "Got packet: stream_index=" << packet->stream_index
-            << " pts=" << packet->pts << " size=" << packet->size;
     if (activeStreamIndices_.count(packet->stream_index) == 0) {
       // This packet is not for any of the active streams.
       continue;
@@ -846,10 +833,6 @@ VideoDecoder::RawDecodedOutput VideoDecoder::getDecodedOutputWithFilter(
       std::chrono::duration_cast<std::chrono::milliseconds>(seekDone - start);
   auto seekToDecodeDone = std::chrono::duration_cast<std::chrono::milliseconds>(
       decodeDone - seekDone);
-  VLOG(3) << "Got frame: stream_index=" << activeStream.stream->index
-          << " pts=" << frame->pts << " stats=" << decodeStats_
-          << " startToSeekDone=" << startToSeekDone.count() << "ms"
-          << " seekToDecodeDone=" << seekToDecodeDone.count() << "ms";
   RawDecodedOutput rawOutput;
   rawOutput.streamIndex = frameStreamIndex;
   rawOutput.frame = std::move(frame);

src/torchcodec/decoders/_core/VideoDecoderOps.cpp

@@ -292,12 +292,6 @@ bool _test_frame_pts_equality(
     int64_t frame_index,
     double pts_seconds_to_test) {
   auto videoDecoder = unwrapTensorToGetDecoder(decoder);
-  LOG(INFO) << "pts_seconds_to_test: " << std::setprecision(15)
-            << pts_seconds_to_test << std::endl;
-  LOG(INFO) << "frame pts  : " << std::setprecision(15)
-            << videoDecoder->getPtsSecondsForFrame(stream_index, frame_index)
-            << std::endl
-            << std::endl;
   return pts_seconds_to_test ==
       videoDecoder->getPtsSecondsForFrame(stream_index, frame_index);
 }