derek.holloway
7502bf64a4
CodeQL / Analyze (actions) (push) Canceled after 0s
CodeQL / Analyze (python) (push) Canceled after 0s
pre-commit / pre-commit (push) Canceled after 0s
Multi-Arch CI / setup (push) Canceled after 0s
Publish build_manylinux_x86_64 images / publish_build_manylinux_x86_64 (push) Canceled after 0s
Setup (Multi-Arch) / setup (push) Canceled after 0s
Multi-Arch CI / Linux::${{ fromJSON(needs.setup.outputs.linux_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubi10 image / publish_no_rocm_image_ubi10 (push) Canceled after 0s
Multi-Arch CI / Windows::${{ fromJSON(needs.setup.outputs.windows_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Multi-Arch CI / Manifest Diff (push) Canceled after 0s
TheRock Manifest Diff Report / Generate Manifest Diff Report (push) Canceled after 0s
Multi-Arch CI / CI Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_ocl_rt images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_rocgdb images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Unit Tests / Unit Tests :: ubuntu-24.04 (push) Canceled after 0s
Unit Tests / Unit Tests :: windows-2022 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_media (push) Canceled after 0s
Unit Tests / Unit Tests Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_openmpi (push) Canceled after 0s
90 lines
2.4 KiB
Markdown
90 lines
2.4 KiB
Markdown
# Building
|
|
|
|
## Prerequisites
|
|
|
|
- CMake 3.22+
|
|
- C++20 compiler (GCC 13+, Clang 16+)
|
|
- Python 3.10+ (for ISA code generation only)
|
|
- ROCm toolchain (optional, for HIP test kernels and daemon tests)
|
|
|
|
Third-party dependencies (Google Test, FlatBuffers) are fetched
|
|
automatically via CMake `FetchContent`.
|
|
|
|
## Quick start
|
|
|
|
```bash
|
|
cmake -B build -G Ninja -DCMAKE_BUILD_TYPE=Release
|
|
cmake --build build
|
|
```
|
|
|
|
## CMake options
|
|
|
|
| Option | Default | Description |
|
|
|---|---|---|
|
|
| `RJ_ENABLE_ASAN` | `OFF` | Enable AddressSanitizer |
|
|
| `RJ_ENABLE_UBSAN` | `OFF` | Enable UndefinedBehaviorSanitizer |
|
|
| `RJ_ENABLE_TSAN` | `OFF` | Enable ThreadSanitizer |
|
|
| `RJ_ENABLE_MSAN` | `OFF` | Enable MemorySanitizer |
|
|
| `RJ_SANITIZER_RUNTIME` | `AUTO` | Select `AUTO`, `SHARED`, or `STATIC` sanitizer runtime linkage |
|
|
| `RJ_CLANG_TIDY` | `OFF` | Enable clang-tidy static analysis |
|
|
| `LTO` | `OFF` | Enable link-time optimization for Release/RelWithDebInfo |
|
|
|
|
### Sanitizer builds
|
|
|
|
```bash
|
|
# AddressSanitizer
|
|
cmake -B build -G Ninja -DCMAKE_BUILD_TYPE=Debug -DRJ_ENABLE_ASAN=1
|
|
|
|
# AddressSanitizer + UndefinedBehaviorSanitizer
|
|
cmake -B build -G Ninja -DCMAKE_BUILD_TYPE=Debug -DRJ_ENABLE_ASAN=1 -DRJ_ENABLE_UBSAN=1
|
|
|
|
# ThreadSanitizer
|
|
cmake -B build -G Ninja -DCMAKE_BUILD_TYPE=Debug -DRJ_ENABLE_TSAN=1
|
|
|
|
# MemorySanitizer
|
|
cmake -B build -G Ninja -DCMAKE_BUILD_TYPE=Debug -DRJ_ENABLE_MSAN=1
|
|
|
|
# UndefinedBehaviorSanitizer
|
|
cmake -B build -G Ninja -DCMAKE_BUILD_TYPE=Debug -DRJ_ENABLE_UBSAN=1
|
|
```
|
|
|
|
`AUTO` uses shared runtimes for ASan, UBSan, and TSan, and the only
|
|
supported static runtime for MSan. `SHARED` rejects MSan explicitly because
|
|
Clang does not provide a shared MSan runtime.
|
|
|
|
### Static analysis
|
|
|
|
```bash
|
|
cmake -B build -G Ninja -DRJ_CLANG_TIDY=ON
|
|
```
|
|
|
|
## Formatting
|
|
|
|
The repo uses pre-commit hooks for formatting (clang-format for C++,
|
|
black for Python, gersemi for CMake). The config is at the repo root
|
|
(`rocm-systems/.pre-commit-config.yaml`).
|
|
|
|
```bash
|
|
pip install pre-commit
|
|
pre-commit install
|
|
pre-commit run --all-files
|
|
```
|
|
|
|
## Container setup for PyTorch
|
|
|
|
For running PyTorch workloads, use a persistent container with
|
|
ROCm and PyTorch pre-installed:
|
|
|
|
```bash
|
|
docker run -it --name rocjitsu-dev \
|
|
-v $PWD:/workspace \
|
|
rocm/pytorch:latest bash
|
|
|
|
# Inside the container, build rocjitsu and run:
|
|
cd /workspace
|
|
cmake -B build -G Ninja -DCMAKE_BUILD_TYPE=Release
|
|
cmake --build build
|
|
rocjitsu --daemon --config configs/gfx950_cdna4_kmd.json -- \
|
|
python3 -c "import torch; print(torch.randn(4,4,device='cuda'))"
|
|
```
|