Skip to content
Permalink

Comparing changes

Choose two branches to see what’s changed or to start a new pull request. If you need to, you can also or learn more about diff comparisons.

Open a pull request

Create a new pull request by comparing changes across two branches. If you need to, you can also . Learn more about diff comparisons here.
base repository: AmusementClub/vs-mlrt
Failed to load repositories. Confirm that selected base ref is valid, then try again.
Loading
base: v15.6
Choose a base ref
...
head repository: AmusementClub/vs-mlrt
Failed to load repositories. Confirm that selected head ref is valid, then try again.
Loading
compare: master
Choose a head ref

Commits on Nov 2, 2024

  1. vsmigx: allow num_streams > 1 (#113)

    Co-authored-by: Abi Hafshin <abi.hafshin@online-pajak.com>
    abihf and Abi Hafshin authored Nov 2, 2024
    Copy the full SHA
    afb834c View commit details
  2. Copy the full SHA
    f1b5ab9 View commit details
  3. Copy the full SHA
    6ba5495 View commit details

Commits on Nov 18, 2024

  1. Copy the full SHA
    092906a View commit details

Commits on Nov 19, 2024

  1. Copy the full SHA
    c684113 View commit details

Commits on Nov 29, 2024

  1. Add support for MLprogram in ort_coreml (#116)

    It enables fp16 computation on ANE, instead of allocating all to CPU. However, the MLprogram is not well-supported currently, supporting much less EPs than regular NeuralNetwork.
    yuygfgg authored Nov 29, 2024
    Copy the full SHA
    a2b1a88 View commit details

Commits on Nov 30, 2024

  1. Copy the full SHA
    ac25053 View commit details

Commits on Dec 3, 2024

  1. Copy the full SHA
    9a7dfac View commit details
  2. Copy the full SHA
    a35daea View commit details
  3. Copy the full SHA
    03b2f9f View commit details

Commits on Dec 22, 2024

  1. Copy the full SHA
    11df3b8 View commit details
  2. Copy the full SHA
    9a26e93 View commit details
  3. Copy the full SHA
    563b8f7 View commit details

Commits on Dec 23, 2024

  1. Copy the full SHA
    2adfbab View commit details

Commits on Jan 17, 2025

  1. Copy the full SHA
    a2a67ff View commit details
  2. Copy the full SHA
    3a0d913 View commit details

Commits on Jan 18, 2025

  1. Copy the full SHA
    d2538f2 View commit details
  2. Copy the full SHA
    1fd7267 View commit details
  3. Copy the full SHA
    b3d6425 View commit details

Commits on Jan 24, 2025

  1. Copy the full SHA
    e3ea9d9 View commit details
  2. Copy the full SHA
    ba63665 View commit details
  3. scripts/vsmlrt.py: add options tiling_optimization_level and `l2_l…

    …imit_for_tiling` to the trt backend
    WolframRhodium committed Jan 24, 2025
    Copy the full SHA
    0ac93bf View commit details
  4. Copy the full SHA
    6d5693c View commit details
  5. Copy the full SHA
    09ca939 View commit details
  6. Copy the full SHA
    9586434 View commit details
  7. Copy the full SHA
    d8a8de5 View commit details

Commits on Jan 25, 2025

  1. Copy the full SHA
    2bf8572 View commit details
  2. Copy the full SHA
    905222d View commit details
  3. Copy the full SHA
    845f2fb View commit details

Commits on Feb 5, 2025

  1. Copy the full SHA
    9ffa6ec View commit details
11 changes: 7 additions & 4 deletions .github/workflows/linux-migx.yml
Original file line number Diff line number Diff line change
@@ -9,7 +9,7 @@ on:

jobs:
build-linux:
runs-on: ubuntu-22.04
runs-on: ubuntu-24.04

defaults:
run:
@@ -33,16 +33,17 @@ jobs:
- name: Setup HIP and MIGraphX
run: |
wget https://repo.radeon.com/rocm/rocm.gpg.key -O - | gpg --dearmor | sudo tee /etc/apt/keyrings/rocm.gpg > /dev/null
echo "deb [arch=amd64 signed-by=/etc/apt/keyrings/rocm.gpg] https://repo.radeon.com/rocm/apt/6.1 jammy main" | sudo tee --append /etc/apt/sources.list.d/rocm.list
echo "deb [arch=amd64 signed-by=/etc/apt/keyrings/rocm.gpg] https://repo.radeon.com/rocm/apt/6.3.1 noble main" | sudo tee --append /etc/apt/sources.list.d/rocm.list
echo -e 'Package: *\nPin: release o=repo.radeon.com\nPin-Priority: 600' | sudo tee /etc/apt/preferences.d/rocm-pin-600
sudo apt update
sudo apt install hip-runtime-amd rocm-device-libs migraphx-dev
sudo apt install hip-runtime-amd rocm-device-libs migraphx-dev hipcc
ls -R /opt/rocm
- name: Configure
run: cmake -S . -B build -G Ninja -Wno-dev -LA
-D CMAKE_BUILD_TYPE=Release
-D VAPOURSYNTH_INCLUDE_DIRECTORY="`pwd`/vapoursynth/include"
-D CMAKE_CXX_COMPILER=g++-12
-D CMAKE_CXX_COMPILER=g++-13
-D CMAKE_CXX_FLAGS="-Wall -ffast-math -march=x86-64-v3"
-D migraphx_DIR=/opt/rocm/lib/cmake/migraphx
-D MIOpen_DIR=/opt/rocm/lib/cmake/miopen
@@ -51,6 +52,8 @@ jobs:
-D amd_comgr_DIR=/opt/rocm/lib/cmake/amd_comgr
-D hsa-runtime64_DIR=/opt/rocm/lib/cmake/hsa-runtime64
-D rocblas_DIR=/opt/rocm/lib/cmake/rocblas
-D hipblaslt_DIR=/opt/rocm/lib/cmake/hipblaslt
-D CMAKE_PREFIX_PATH=/opt/rocm/lib/cmake

- name: Build
run: cmake --build build --verbose
140 changes: 140 additions & 0 deletions .github/workflows/linux-ov-arm64.yml
Original file line number Diff line number Diff line change
@@ -0,0 +1,140 @@
name: Build (Linux-OV, ARM64)

on:
push:
paths:
- 'vsov/**'
- '.github/workflows/linux-ov-arm64.yml'
workflow_dispatch:

jobs:
build-linux:
runs-on: ubuntu-24.04-arm

defaults:
run:
working-directory: vsov

steps:
- name: Checkout repo
uses: actions/checkout@v4
with:
fetch-depth: 0

- name: Setup Ninja
run: pip install ninja

- name: Cache protobuf
id: cache-protobuf
uses: actions/cache@v4
with:
path: vsov/protobuf/install
key: ${{ runner.os }}-vsov-protobuf-arm64-v1

- name: Checkout protobuf
uses: actions/checkout@v4
if: steps.cache-protobuf.outputs.cache-hit != 'true'
with:
repository: protocolbuffers/protobuf
# follows protobuf in https://github.com/openvinotoolkit/openvino/tree/2024.6.0/thirdparty/protobuf
# if you change this, remember to bump the version of the cache key.
ref: f0dc78d7e6e331b8c6bb2d5283e06aa26883ca7c
fetch-depth: 1
path: vsov/protobuf

- name: Configure protobuf
if: steps.cache-protobuf.outputs.cache-hit != 'true'
run: cmake -S protobuf/cmake -B protobuf/build_rel -G Ninja -LA
-D CMAKE_BUILD_TYPE=Release
-D CMAKE_POSITION_INDEPENDENT_CODE=ON
-D protobuf_BUILD_SHARED_LIBS=OFF -D protobuf_BUILD_TESTS=OFF

- name: Build protobuf
if: steps.cache-protobuf.outputs.cache-hit != 'true'
run: cmake --build protobuf/build_rel --verbose

- name: Install protobuf
if: steps.cache-protobuf.outputs.cache-hit != 'true'
run: cmake --install protobuf/build_rel --prefix protobuf/install

- name: Cache onnx
id: cache-onnx
uses: actions/cache@v4
with:
path: vsov/onnx/install
key: ${{ runner.os }}-vsov-onnx-arm64-v1

- name: Checkout onnx
if: steps.cache-onnx.outputs.cache-hit != 'true'
uses: actions/checkout@v4
with:
repository: onnx/onnx
# follows onnx in https://github.com/openvinotoolkit/openvino/tree/2024.6.0/thirdparty/onnx
# if you change this, remember to bump the version of the cache key.
ref: b8baa8446686496da4cc8fda09f2b6fe65c2a02c
fetch-depth: 1
path: vsov/onnx

- name: Configure onnx
if: steps.cache-onnx.outputs.cache-hit != 'true'
run: cmake -S onnx -B onnx/build -G Ninja -LA
-D CMAKE_BUILD_TYPE=Release
-D CMAKE_POSITION_INDEPENDENT_CODE=ON
-D Protobuf_PROTOC_EXECUTABLE=protobuf/install/bin/protoc
-D Protobuf_LITE_LIBRARY=protobuf/install/lib
-D Protobuf_LIBRARIES=protobuf/install/lib
-D ONNX_USE_LITE_PROTO=ON -D ONNX_USE_PROTOBUF_SHARED_LIBS=OFF
-D ONNX_GEN_PB_TYPE_STUBS=OFF -D ONNX_ML=0
-D ONNX_USE_MSVC_STATIC_RUNTIME=1

- name: Build onnx
if: steps.cache-onnx.outputs.cache-hit != 'true'
run: cmake --build onnx/build --verbose

- name: Install onnx
if: steps.cache-onnx.outputs.cache-hit != 'true'
run: cmake --install onnx/build --prefix onnx/install

- name: Download VapourSynth headers
run: |
wget -q -O vs.zip https://github.com/vapoursynth/vapoursynth/archive/refs/tags/R57.zip
unzip -q vs.zip
mv vapoursynth*/ vapoursynth
- name: Setup OpenVINO
run: |
curl -L -o ov.tgz https://storage.openvinotoolkit.org/repositories/openvino/packages/2024.6/linux/l_openvino_toolkit_ubuntu20_2024.6.0.17404.4c0f47d2335_arm64.tgz
tar -xf ov.tgz
mv l_openvino_* openvino -v
- name: Configure
run: cmake -S . -B build -G Ninja -LA
-D CMAKE_BUILD_TYPE=Release
-D CMAKE_CXX_FLAGS="-Wall -ffast-math"
-D VAPOURSYNTH_INCLUDE_DIRECTORY="`pwd`/vapoursynth/include"
-D OpenVINO_DIR=openvino/runtime/cmake
-D ENABLE_VISUALIZATION=ON
-D WIN32_SHARED_OPENVINO=ON
-D protobuf_DIR=protobuf/install/lib/cmake/protobuf
-D ONNX_DIR=onnx/install/lib/cmake/ONNX

- name: Build
run: cmake --build build --verbose

- name: Install
run: cmake --install build --prefix install

- name: Prepare for upload
run: |
mkdir artifact
cp -v install/lib/*.so artifact
- name: Describe
run: git describe --tags --long

- name: Upload
uses: actions/upload-artifact@v4
with:
name: VSOV-Linux-ARM64
path: vsov/artifact

10 changes: 5 additions & 5 deletions .github/workflows/linux-ov.yml
Original file line number Diff line number Diff line change
@@ -36,9 +36,9 @@ jobs:
if: steps.cache-protobuf.outputs.cache-hit != 'true'
with:
repository: protocolbuffers/protobuf
# follows protobuf in https://github.com/openvinotoolkit/openvino/tree/2024.1.0/thirdparty/protobuf
# follows protobuf in https://github.com/openvinotoolkit/openvino/tree/2024.6.0/thirdparty/protobuf
# if you change this, remember to bump the version of the cache key.
ref: fe271ab76f2ad2b2b28c10443865d2af21e27e0e
ref: f0dc78d7e6e331b8c6bb2d5283e06aa26883ca7c
fetch-depth: 1
path: vsov/protobuf

@@ -69,9 +69,9 @@ jobs:
uses: actions/checkout@v4
with:
repository: onnx/onnx
# follows onnx in https://github.com/openvinotoolkit/openvino/tree/2024.1.0/thirdparty/onnx
# follows onnx in https://github.com/openvinotoolkit/openvino/tree/2024.6.0/thirdparty/onnx
# if you change this, remember to bump the version of the cache key.
ref: b86cc54efce19530fb953e4b21f57e6b3888534c
ref: b8baa8446686496da4cc8fda09f2b6fe65c2a02c
fetch-depth: 1
path: vsov/onnx

@@ -103,7 +103,7 @@ jobs:
- name: Setup OpenVINO
run: |
curl -L -o ov.tgz https://storage.openvinotoolkit.org/repositories/openvino/packages/2024.1/linux/l_openvino_toolkit_ubuntu22_2024.1.0.15008.f4afc983258_x86_64.tgz
curl -L -o ov.tgz https://storage.openvinotoolkit.org/repositories/openvino/packages/2024.6/linux/l_openvino_toolkit_ubuntu24_2024.6.0.17404.4c0f47d2335_x86_64.tgz
tar -xf ov.tgz
mv l_openvino_* openvino -v
70 changes: 70 additions & 0 deletions .github/workflows/linux-trt-arm64.yml
Original file line number Diff line number Diff line change
@@ -0,0 +1,70 @@
name: Build (Linux-TRT, ARM64)

on:
push:
paths:
- 'vstrt/**'
- '.github/workflows/linux-trt-arm64.yml'
workflow_dispatch:

jobs:
build-linux:
runs-on: ubuntu-24.04-arm

defaults:
run:
working-directory: vstrt

steps:
- name: Checkout repo
uses: actions/checkout@v4
with:
fetch-depth: 0

- name: Setup Ninja
run: pip install ninja --break-system-packages

- name: Download VapourSynth headers
run: |
wget -q -O vs.zip https://github.com/vapoursynth/vapoursynth/archive/refs/tags/R57.zip
unzip -q vs.zip
mv vapoursynth*/ vapoursynth
- name: Setup CUDA and TensorRT
run: |
wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2404/sbsa/cuda-keyring_1.1-1_all.deb
sudo dpkg -i cuda-keyring_1.1-1_all.deb
sudo apt-get update
export TRT_VER=10.8.0.43-1+cuda12.8
sudo apt-get install -y --no-install-recommends cuda-nvcc-12-8 cuda-cudart-dev-12-8 libnvinfer-dev=${TRT_VER} libnvinfer-headers-dev=${TRT_VER} libnvinfer10=${TRT_VER}
echo "PATH=/usr/local/cuda/bin${PATH:+:${PATH}}" >> $GITHUB_ENV
echo "CUDA_PATH=/usr/local/cuda" >> $GITHUB_ENV
echo "LD_LIBRARY_PATH=/usr/local/cuda/lib${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}" >> $GITHUB_ENV
- name: Configure
run: cmake -S . -B build -G Ninja -LA
-D CMAKE_BUILD_TYPE=Release
-D VAPOURSYNTH_INCLUDE_DIRECTORY="`pwd`/vapoursynth/include"
-D CMAKE_CXX_FLAGS="-Wall -ffast-math"

- name: Build
run: cmake --build build --verbose

- name: Install
run: cmake --install build --prefix install

- name: Prepare for upload
run: |
mkdir artifact
cp -v install/lib/*.so artifact
- name: Describe
run: git describe --tags --long

- name: Upload
uses: actions/upload-artifact@v4
with:
name: VSTRT-Linux-ARM64
path: vstrt/artifact
overwrite: true

5 changes: 2 additions & 3 deletions .github/workflows/linux-trt.yml
Original file line number Diff line number Diff line change
@@ -35,8 +35,8 @@ jobs:
wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2404/x86_64/cuda-keyring_1.1-1_all.deb
sudo dpkg -i cuda-keyring_1.1-1_all.deb
sudo apt-get update
export TRT_VER=10.6.0.26-1+cuda12.6
sudo apt-get install -y --no-install-recommends cuda-nvcc-12-6 cuda-cudart-dev-12-6 libnvinfer-dev=${TRT_VER} libnvinfer-headers-dev=${TRT_VER} libnvinfer10=${TRT_VER}
export TRT_VER=10.8.0.43-1+cuda12.8
sudo apt-get install -y --no-install-recommends cuda-nvcc-12-8 cuda-cudart-dev-12-8 libnvinfer-dev=${TRT_VER} libnvinfer-headers-dev=${TRT_VER} libnvinfer10=${TRT_VER}
echo "PATH=/usr/local/cuda/bin${PATH:+:${PATH}}" >> $GITHUB_ENV
echo "CUDA_PATH=/usr/local/cuda" >> $GITHUB_ENV
echo "LD_LIBRARY_PATH=/usr/local/cuda/lib${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}" >> $GITHUB_ENV
@@ -45,7 +45,6 @@ jobs:
run: cmake -S . -B build -G Ninja -LA
-D CMAKE_BUILD_TYPE=Release
-D VAPOURSYNTH_INCLUDE_DIRECTORY="`pwd`/vapoursynth/include"
-D CMAKE_CXX_COMPILER=g++-13
-D CMAKE_CXX_FLAGS="-Wall -ffast-math -march=x86-64-v3"

- name: Build
2 changes: 1 addition & 1 deletion .github/workflows/macos-ort.yml
Original file line number Diff line number Diff line change
@@ -103,7 +103,7 @@ jobs:
- name: Setup ONNX Runtime
run: |
curl -L -o ort.tgz https://github.com/microsoft/onnxruntime/releases/download/v1.19.2/onnxruntime-osx-arm64-1.19.2.tgz
curl -L -o ort.tgz https://github.com/microsoft/onnxruntime/releases/download/v1.20.0/onnxruntime-osx-arm64-1.20.0.tgz
tar -xf ort.tgz
mv onnxruntime-* onnxruntime
14 changes: 7 additions & 7 deletions .github/workflows/windows-cuda-dependency.yml
Original file line number Diff line number Diff line change
@@ -19,7 +19,7 @@ on:

jobs:
build-windows:
runs-on: windows-2022
runs-on: windows-2025
outputs:
runID: ${{ steps.output.outputs.runID }}

@@ -42,7 +42,7 @@ jobs:
- name: Download TensorRT library
run: |
curl -L -o trt.zip https://developer.nvidia.com/downloads/compute/machine-learning/tensorrt/10.6.0/zip/TensorRT-10.6.0.26.Windows.win10.cuda-12.6.zip
curl -L -o trt.zip https://developer.download.nvidia.com/compute/machine-learning/tensorrt/10.8.0/zip/TensorRT-10.8.0.43.Windows.win10.cuda-12.8.zip
- name: Extract TensorRT library
run: |
@@ -55,14 +55,14 @@ jobs:
- name: Download CUDA Libraries
shell: cmd
run: |
curl -s -o cuda_installer.exe -L https://developer.download.nvidia.com/compute/cuda/12.6.2/network_installers/cuda_12.6.2_windows_network.exe
cuda_installer.exe -s cudart_12.6 cublas_12.6 cufft_12.6 cupti_12.6 nvrtc_12.6
curl -s -o cuda_installer.exe -L https://developer.download.nvidia.com/compute/cuda/12.8.0/network_installers/cuda_12.8.0_windows_network.exe
cuda_installer.exe -s cudart_12.8 cublas_12.8 cufft_12.8 cupti_12.8 nvrtc_12.8
- name: Move CUDA Libraries
shell: cmd
run: |
move "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.6\extras\CUPTI\lib64\cupti*.dll" vsmlrt-cuda
move "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.6\bin\*.dll" vsmlrt-cuda
move "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8\extras\CUPTI\lib64\cupti*.dll" vsmlrt-cuda
move "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8\bin\*.dll" vsmlrt-cuda
del vsmlrt-cuda\cudart32*.dll
- name: Setup VC commands
@@ -93,7 +93,7 @@ jobs:
- name: Compress
run: |
7z a -t7z -mx=3 vsmlrt-cuda.7z vsmlrt-cuda
7z a -t7z -bb3 -mx=9 vsmlrt-cuda.7z vsmlrt-cuda
- name: Upload
uses: actions/upload-artifact@v4
Loading