Port C++ code to rely on torch stable ABI API. by NicolasHug · Pull Request #1188 · meta-pytorch/torchcodec

NicolasHug · 2026-01-23T11:42:15Z

No description provided.

NicolasHug · 2026-01-23T14:47:21Z

test/third-party-interface/ThirdPartyInterfaceTest.cpp

 class DummyDeviceInterface : public DeviceInterface {
 public:
-  DummyDeviceInterface(const torch::Device& device) : DeviceInterface(device) {}
+  DummyDeviceInterface(const StableDevice& device) : DeviceInterface(device) {}


@dvrogozh , heads up that this is something we might want to do in the near future.

The general context is that we want to rely on torch stable ABI so that we (or you) don't have to synchronize torchcodec releases with torch releases 6X a year.

I think it means that the third-party extensions will have to pass-in "Stable" types - you might still be able to use the non-stable ones internally as long as you convert them.
Let me know if you have any concern about that.

Note that the PR is in a very, very WIP state at the moment. Plenty of things are likely to change. But this is the direction we're heading to.

@NicolasHug , thank your for heads up. I will start looking into the story, but "stable" sounds very appealing. Once PR will be in a closer to merge stage I would love to give it a try on my side (create respective PR for xpu and see how it goes).

NicolasHug · 2026-01-23T14:47:50Z

src/torchcodec/share/cmake/TorchCodec/TorchCodecConfig.cmake

+      # Set PyTorch stable ABI target version to 2.11 for consumers.
+      # This ensures third-party code uses stable ABI types compatible with
+      # the library. Format: 0x MAJOR(2) MINOR(2) PATCH(2) TWEAK(2) zeros(8)
+      INTERFACE_COMPILE_DEFINITIONS "TORCH_TARGET_VERSION=0x020b000000000000"


TODO: I'm not sure we'll want to enforce that on third-parties?

CC @dvrogozh

src/torchcodec/_core/__init__.py

NicolasHug · 2026-01-23T16:32:33Z

src/torchcodec/_core/AVIOTensorContext.h

 // For Decoding: enables users to pass in the entire video or audio as bytes.
 // Our read and seek functions then traverse the bytes in memory.
-class AVIOFromTensorContext : public AVIOContextHolder {
+class TORCHCODEC_API AVIOFromTensorContext : public AVIOContextHolder {


TODO: understand why that's needed. Ideally it shouldn't.

src/torchcodec/_core/BetaCudaDeviceInterface.cpp

NicolasHug · 2026-01-23T16:33:15Z

src/torchcodec/_core/BetaCudaDeviceInterface.cpp

+  if (!device_.has_index()) {
+    int resolvedIndex = getDeviceIndex(device_);
+    device_.set_index(static_cast<StableDeviceIndex>(resolvedIndex));
+  }


TODO above: this was causing a hang - investigate.

NicolasHug · 2026-01-23T16:35:56Z

src/torchcodec/_core/CMakeLists.txt

+    target_include_directories(${library_name}
+        BEFORE PRIVATE
+        ${pybind11_INCLUDE_DIRS}
+    )


TODO investigate above.

NicolasHug · 2026-01-23T16:36:25Z

src/torchcodec/_core/CMakeLists.txt

+            "-Wno-attributes"
+        )
+    endif()
+


TODO on all these

NicolasHug · 2026-01-23T16:37:04Z

src/torchcodec/_core/CpuDeviceInterface.cpp

+    // For a tensor of shape [numChannels, numSamples], we need to calculate
+    // offsets
+    outputBuffers[i] = static_cast<uint8_t*>(lastSamples.mutable_data_ptr()) +
+        i * numRemainingSamples * sizeof(float);


NicolasHug · 2026-01-23T16:38:03Z

src/torchcodec/_core/custom_ops.cpp

+  // Note: impl_abstract_pystub is not available in stable ABI.
+  // The Python stubs are handled via Python-side registration instead.


src/torchcodec/_core/custom_ops.cpp

NicolasHug · 2026-01-23T16:38:37Z

src/torchcodec/_core/custom_ops.cpp

+      "_add_video_stream(Tensor(a!) decoder, *, int? num_threads=None, str? dimension_order=None, int? stream_index=None, str device=\"cpu\", str device_variant=\"ffmpeg\", str transform_specs=\"\", Tensor? custom_frame_mappings_all_frames=None, Tensor? custom_frame_mappings_is_key_frame=None, Tensor? custom_frame_mappings_duration=None, str? color_conversion_library=None) -> ()");
  m.def(
-      "add_video_stream(Tensor(a!) decoder, *, int? num_threads=None, str? dimension_order=None, int? stream_index=None, str device=\"cpu\", str device_variant=\"ffmpeg\", str transform_specs=\"\", (Tensor, Tensor, Tensor)? custom_frame_mappings=None) -> ()");
+      "add_video_stream(Tensor(a!) decoder, *, int? num_threads=None, str? dimension_order=None, int? stream_index=None, str device=\"cpu\", str device_variant=\"ffmpeg\", str transform_specs=\"\", Tensor? custom_frame_mappings_all_frames=None, Tensor? custom_frame_mappings_is_key_frame=None, Tensor? custom_frame_mappings_duration=None) -> ()");


TODO: can't pass tuples anymore?

NicolasHug · 2026-01-23T16:39:15Z

src/torchcodec/_core/custom_ops.cpp

+    StableTensor& decoder,
    std::optional<int64_t> num_threads = std::nullopt,
-    std::optional<std::string_view> dimension_order = std::nullopt,
+    std::optional<std::string> dimension_order = std::nullopt,


TODO we went from string_view to string everywhere, is that OK? Is that needed?

src/torchcodec/_core/DeviceInterface.cpp

src/torchcodec/_core/ops.py

…_abi

This PR adds `deleter` support to `torch::stable::from_blob` by adding a new `aoti_torch_create_tensor_from_blob_v3`. We need it to cleanly port TorchCodec to the stable ABI in meta-pytorch/torchcodec#1188 There's a bit of scaffolding, especially for the tests where I had to create a new `test/cpp_extensions/libtorch_agn_2_11_extension` folder. Most of it is just copy/paste from `test/cpp_extensions/libtorch_agn_2_11_extension/libtorch_agn_2_10/__init__.py` Pull Request resolved: #173371 Approved by: https://github.com/janeyx99

…o stable_abi

pytorch-bot · 2026-02-04T13:43:56Z

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/meta-pytorch/torchcodec/1188

📄 Preview Python docs built from this PR

Note: Links to docs will display an error until the docs builds have been completed.

❌ 20 New Failures, 10 Unrelated Failures

As of commit aa87b60 with merge base 6f10da0 ():

NEW FAILURES - The following jobs have failed:

Build and test Linux CUDA aarch64 wheels / install-and-test (3.10, 12.6, 4.4.2) (gh)
Process completed with exit code 1.
Build and test Linux CUDA aarch64 wheels / install-and-test (3.10, 12.6, 5.1.2) (gh)
Process completed with exit code 1.
Build and test Linux CUDA aarch64 wheels / install-and-test (3.10, 12.6, 6.1.1) (gh)
Process completed with exit code 1.
Build and test Linux CUDA aarch64 wheels / install-and-test (3.10, 12.6, 7.0.1) (gh)
Process completed with exit code 1.
Build and test Linux CUDA aarch64 wheels / install-and-test (3.10, 12.6, 8.0) (gh)
Process completed with exit code 1.
Build and test Linux CUDA wheels and docs / build-docs (gh)
Process completed with exit code 1.
Build and test Linux CUDA wheels and docs / install-and-test (3.10, 12.6, 4.4.2) (gh)
Process completed with exit code 1.
Build and test Linux CUDA wheels and docs / install-and-test (3.10, 12.6, 6) (gh)
Process completed with exit code 1.
Build and test Linux CUDA wheels and docs / install-and-test (3.10, 12.6, 7) (gh)
Process completed with exit code 1.
Build and test Linux CUDA wheels and docs / install-and-test (3.10, 12.6, 8.0) (gh)
Process completed with exit code 1.
Build and test Linux CUDA wheels and docs / install-and-test (3.10, 13.0, 4.4.2) (gh)
Process completed with exit code 1.
Build and test Linux CUDA wheels and docs / install-and-test (3.10, 13.0, 6) (gh)
Process completed with exit code 1.
Build and test Linux CUDA wheels and docs / install-and-test (3.10, 13.0, 7) (gh)
Process completed with exit code 1.
Build and test Linux CUDA wheels and docs / install-and-test (3.10, 13.0, 8.0) (gh)
Process completed with exit code 1.
Build and test MacOS wheel / install-and-test (3.10, 4.4.2) (gh)
test/test_transform_ops.py::TestCoreVideoDecoderTransformOps::test_crop_transform_fails
Build and test MacOS wheel / install-and-test (3.10, 5.1.2) (gh)
test/test_transform_ops.py::TestCoreVideoDecoderTransformOps::test_crop_transform_fails
Build and test MacOS wheel / install-and-test (3.10, 6.1.1) (gh)
test/test_transform_ops.py::TestCoreVideoDecoderTransformOps::test_crop_transform_fails
Build and test MacOS wheel / install-and-test (3.10, 7.0.1) (gh)
test/test_transform_ops.py::TestCoreVideoDecoderTransformOps::test_crop_transform_fails
Build and test MacOS wheel / install-and-test (3.10, 8.0) (gh)
test/test_transform_ops.py::TestCoreVideoDecoderTransformOps::test_crop_transform_fails
Lint / mypy (3.12) (gh)
Process completed with exit code 1.

BROKEN TRUNK - The following jobs failed but were present on the merge base:

👉 Rebase onto the `viable/strict` branch to avoid these failures

Build and test Linux wheel / install-and-test (3.10, 4.4.2) (gh) (trunk failure)
Process completed with exit code 1.
Build and test Linux wheel / install-and-test (3.10, 5.1.2) (gh) (trunk failure)
Process completed with exit code 1.
Build and test Linux wheel / install-and-test (3.10, 6.1.1) (gh) (trunk failure)
Build and test Linux wheel / install-and-test (3.10, 7.0.1) (gh) (trunk failure)
Process completed with exit code 1.
Build and test Linux wheel / install-and-test (3.10, 8.0) (gh) (trunk failure)
Process completed with exit code 1.
Build and test Linux wheel / install-and-test-third-party-interface (3.10, 8.0) (gh) (trunk failure)
Process completed with exit code 1.
Build and test Windows wheel / install-and-test (3.10, 4.4.2) (gh) (trunk failure)
Process completed with exit code 1.
Build and test Windows wheel / install-and-test (3.10, 6.1.1) (gh) (trunk failure)
Process completed with exit code 1.
Build and test Windows wheel / install-and-test (3.10, 7.0.1) (gh) (trunk failure)
Process completed with exit code 1.
Build and test Windows wheel / install-and-test (3.10, 8.0) (gh) (trunk failure)
Process completed with exit code 1.

This comment was automatically generated by Dr. CI and updates every 15 minutes.

NicolasHug · 2026-02-04T13:55:05Z

src/torchcodec/_core/DeviceInterface.cpp

+      StableIntArrayRef(strides.data(), strides.size()),
+      StableDevice(kStableCPU),
+      kStableUInt8,
+      avFrameDeleter);


…o stable_abi

…_abi

NicolasHug added 7 commits January 22, 2026 17:58

WIP

5025986

It compiles. And crashes.

68db6f2

Compiles, import is OK, segfaults on tests

04b3e5c

Some tests are passing

f1bb56f

Really weird fix to fix hanging issue

ace2198

More stufffffff

0a5dc8d

Lint

0b7b31a

meta-cla bot added the CLA Signed This label is managed by the Meta Open Source bot. label Jan 23, 2026

NicolasHug added 6 commits January 23, 2026 13:18

Fix stuff?

3f0219a

lint

22d811a

more

8e10fab

refact C++ test

b0e00b7

Fix Windows?

b86c98b

lint

72977df