# agrim-p777/conda-package-genn architecture > Architecture diagram of the GitHub repository agrim-p777/conda-package-genn, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-06-26). It is an AI-made overview: check details against the source. - Interactive diagram: https://gitdiagram.com/agrim-p777/conda-package-genn - Repository: https://github.com/agrim-p777/conda-package-genn ## Overview 1. **What this repository is** This repo is not the GeNN source code itself; it is the **packaging and release architecture** for distributing **PyGeNN** as Conda packages. The goal is to turn a source-built C++/CUDA scientific library into installable binaries for end users. 2. **Two primary product lines** The architecture splits into two top-level package families: - **`pygenn-cpu`**: CPU-only build, no CUDA dependency - **`pygenn-cuda`**: GPU-enabled build, depends on CUDA toolchain/runtime packages This is the main product boundary in the repo. 3. **Conda recipe as the central build unit** Each package family is built from a Conda recipe centered on: - `meta.yaml` for package metadata, dependency pins, outputs, and build commands - `conda_build_config.yaml` for variant selection and version pinning The recipes are the authoritative architecture layer; everything else supports them. 4. **Build configuration is platform- and version-aware** The tree contains multiple recipe variants for **Linux, Windows, and macOS**, plus CUDA version splits such as **CUDA 11.7** and **CUDA 12.x / 12.4**. That indicates the repo is organized around a **matrix of build targets**, not a single universal build script. 5. **CPU vs CUDA have different dependency boundaries** The CPU package is a simpler runtime distribution of the Python bindings plus the compiled GeNN backend libraries. The CUDA package adds the full GPU compilation/runtime boundary: modular CUDA Conda packages, CUDA path handling, and post-install guidance. This is the key architectural distinction between the two deliverables. 6. **Post-install behavior is part of the package design** The CUDA recipe includes **`post-link.sh`** and **`post-link.bat`**, which means installation is not purely file copy based. The package also needs to communicate runtime environment setup, especially `CUDA_PATH`, so install-time hooks are used as part of the user-facing architecture. 7. **Platform-specific build adaptations** The presence of separate folders like `pygenn-cpu (mac01)`, `pygenn-cpu (mac02)`, `pygenn-win-cpu`, `pygenn-win-cuda (nmake)`, and Linux-specific recipe dirs shows the project had to encode **OS-specific build mechanics**: - macOS dynamic library handling - Windows build-system differences, including NMake/MSBuild - Linux recipes with simpler native toolchain assumptions These are not just alternate configs; they represent distinct packaging solutions per platform. 8. **Runtime artifact handling is important** The README highlights fixes for macOS `.dylib` packaging and Windows CUDA path resolution, which means the package boundary includes: - compiled shared libraries - Python extension/modules - platform loader behavior The repo therefore bridges **Python packaging** and **native binary distribution**. 9. **Tests are smoke tests, not application tests** The `tests/` directories contain smoke tests for both CPU and CUDA packages: - `run_pygenn_cpu_smoketest.py` - `run_pygenn_cuda_smoketest.py` - `test.py` These likely validate that the installed package can be imported and that the backend loads correctly in a Conda environment. The tests are aimed at packaging integrity rather than model correctness. 10. **Conda-Forge publication pipeline is explicit** The `Conda_Forge/` directory mirrors the package families, indicating a separate layout for **Conda-Forge submission and CI-ready recipes**. This repo supports the path from local packaging work to staged/official Conda-Forge recipes, so the architecture includes a **recipe authoring layer** and a **distribution layer**. 11. **Documentation and assets are supporting, not operational** `README.md` and the `assets/` directory are documentation/branding support. They explain the packaging rationale, but they are not part of the build or runtime path. 12. **Overall data flow** The high-level flow is: GeNN source build inputs and platform toolchains → Conda recipe selection (`meta.yaml` + `conda_build_config.yaml`) → platform/CUDA-specific build → packaged Python bindings and native libraries → post-link environment guidance → smoke tests in Conda environments. That is the architectural story this repo encodes. Main entry points: [`pygenn-cpu/meta.yaml`](https://github.com/agrim-p777/conda-package-genn/blob/main/pygenn-cpu/meta.yaml) (CPU meta), [`pygenn-cuda/meta.yaml`](https://github.com/agrim-p777/conda-package-genn/blob/main/pygenn-cuda/meta.yaml) (CUDA meta), [`pygenn-cuda`](https://github.com/agrim-p777/conda-package-genn/tree/main/pygenn-cuda) (CUDA package), [`pygenn-cpu`](https://github.com/agrim-p777/conda-package-genn/tree/main/pygenn-cpu) (CPU package) ## Diagram (Mermaid) ```mermaid flowchart TD subgraph group_docs["Docs"] node_readme["README
docs
[README.md]"] node_assets["Assets
docs"] end subgraph group_recipes["Core recipes"] node_cpu_root["CPU package
recipe family"] node_cuda_root["CUDA package
recipe family"] node_cpu_meta["CPU meta
conda recipe
[meta.yaml]"] node_cpu_cfg["CPU config
variant config"] node_cuda_meta["CUDA meta
conda recipe
[meta.yaml]"] node_cuda_cfg["CUDA config
variant config"] node_cuda_postlink["Post-link
install hook
[post-link.sh]"] end subgraph group_variants["Platform variants"] node_linux_cpu["Linux CPU
platform recipe"] node_linux_cuda117["Linux CUDA 11.7
platform recipe"] node_linux_cuda12x["Linux CUDA 12.x
platform recipe"] node_linux_cuda124["Linux CUDA 12.4
platform recipe"] node_win_cpu["Windows CPU
platform recipe"] node_win_cuda_nmake["Win CUDA NMake
platform recipe"] node_win_cuda124["Win CUDA 12.4
platform recipe"] node_mac_cpu1["macOS CPU A
platform recipe"] node_mac_cpu2["macOS CPU B
platform recipe"] node_mac_winlinux["Win/Linux CPU
platform recipe"] end subgraph group_forge["Conda-Forge"] node_forge_cpu["Forge CPU
forge recipe"] node_forge_cuda["Forge CUDA
forge recipe"] end subgraph group_checks["Validation"] node_smoketest_cpu["CPU smoke
smoke test"] node_smoketest_cuda["CUDA smoke
smoke test"] node_root_smoketest["Smoke suite
smoke tests
[test.py]"] end node_readme -->|"describes"| node_cpu_root node_readme -->|"describes"| node_cuda_root node_assets -.->|"supports"| node_readme node_cpu_root -->|"centers on"| node_cpu_meta node_cpu_root -->|"variants by"| node_cpu_cfg node_cuda_root -->|"centers on"| node_cuda_meta node_cuda_root -->|"variants by"| node_cuda_cfg node_cuda_root -->|"installs"| node_cuda_postlink node_cpu_meta -.->|"specializes"| node_linux_cpu node_cpu_meta -.->|"specializes"| node_win_cpu node_cpu_meta -.->|"specializes"| node_mac_cpu1 node_cpu_meta -.->|"specializes"| node_mac_cpu2 node_cpu_meta -.->|"specializes"| node_mac_winlinux node_cuda_meta -.->|"specializes"| node_linux_cuda117 node_cuda_meta -.->|"specializes"| node_linux_cuda12x node_cuda_meta -.->|"specializes"| node_linux_cuda124 node_cuda_meta -.->|"specializes"| node_win_cuda_nmake node_cuda_meta -.->|"specializes"| node_win_cuda124 node_forge_cpu -.->|"mirrors"| node_cpu_root node_forge_cuda -.->|"mirrors"| node_cuda_root node_cpu_root -->|"validated by"| node_smoketest_cpu node_cuda_root -->|"validated by"| node_smoketest_cuda node_smoketest_cpu -.->|"feeds"| node_root_smoketest node_smoketest_cuda -.->|"feeds"| node_root_smoketest node_cuda_postlink -.->|"supports"| node_smoketest_cuda click node_readme "https://github.com/agrim-p777/conda-package-genn/blob/main/README.md" click node_assets "https://github.com/agrim-p777/conda-package-genn/tree/main/assets" click node_cpu_root "https://github.com/agrim-p777/conda-package-genn/tree/main/pygenn-cpu" click node_cuda_root "https://github.com/agrim-p777/conda-package-genn/tree/main/pygenn-cuda" click node_cpu_meta "https://github.com/agrim-p777/conda-package-genn/blob/main/pygenn-cpu/meta.yaml" click node_cpu_cfg "https://github.com/agrim-p777/conda-package-genn/blob/main/pygenn-cpu/conda_build_config.yaml" click node_cuda_meta "https://github.com/agrim-p777/conda-package-genn/blob/main/pygenn-cuda/meta.yaml" click node_cuda_cfg "https://github.com/agrim-p777/conda-package-genn/blob/main/pygenn-cuda/conda_build_config.yaml" click node_cuda_postlink "https://github.com/agrim-p777/conda-package-genn/blob/main/pygenn-cuda/post-link.sh" click node_forge_cpu "https://github.com/agrim-p777/conda-package-genn/tree/main/Conda_Forge/pygenn-cpu" click node_forge_cuda "https://github.com/agrim-p777/conda-package-genn/tree/main/Conda_Forge/pygenn-cuda" click node_linux_cpu "https://github.com/agrim-p777/conda-package-genn/tree/main/pygenn-linux-cpu" click node_linux_cuda117 "https://github.com/agrim-p777/conda-package-genn/blob/main/pygenn-linux-cuda11.7" click node_linux_cuda12x "https://github.com/agrim-p777/conda-package-genn/blob/main/pygenn-linux-cuda12.x" click node_linux_cuda124 "https://github.com/agrim-p777/conda-package-genn/blob/main/pygenn-linux-cuda12.4" click node_win_cpu "https://github.com/agrim-p777/conda-package-genn/tree/main/pygenn-win-cpu" click node_win_cuda_nmake "https://github.com/agrim-p777/conda-package-genn/tree/main/pygenn-win-cuda (nmake)" click node_win_cuda124 "https://github.com/agrim-p777/conda-package-genn/blob/main/pygenn-win-cuda12.4" click node_mac_cpu1 "https://github.com/agrim-p777/conda-package-genn/tree/main/pygenn-cpu (mac01)" click node_mac_cpu2 "https://github.com/agrim-p777/conda-package-genn/tree/main/pygenn-cpu (mac02)" click node_mac_winlinux "https://github.com/agrim-p777/conda-package-genn/tree/main/pygenn-cpu (winlinux)" click node_smoketest_cpu "https://github.com/agrim-p777/conda-package-genn/blob/main/Conda_Forge/pygenn-cpu/tests/run_pygenn_cpu_smoketest.py" click node_smoketest_cuda "https://github.com/agrim-p777/conda-package-genn/blob/main/Conda_Forge/pygenn-cuda/tests/run_pygenn_cuda_smoketest.py" click node_root_smoketest "https://github.com/agrim-p777/conda-package-genn/blob/main/tests/test.py" classDef toneNeutral fill:#f8fafc,stroke:#334155,stroke-width:1.5px,color:#0f172a classDef toneBlue fill:#dbeafe,stroke:#2563eb,stroke-width:1.5px,color:#172554 classDef toneAmber fill:#fef3c7,stroke:#d97706,stroke-width:1.5px,color:#78350f classDef toneMint fill:#dcfce7,stroke:#16a34a,stroke-width:1.5px,color:#14532d classDef toneRose fill:#ffe4e6,stroke:#e11d48,stroke-width:1.5px,color:#881337 classDef toneIndigo fill:#e0e7ff,stroke:#4f46e5,stroke-width:1.5px,color:#312e81 classDef toneTeal fill:#ccfbf1,stroke:#0f766e,stroke-width:1.5px,color:#134e4a class node_readme,node_assets toneBlue class node_cpu_root,node_cuda_root,node_cpu_meta,node_cpu_cfg,node_cuda_meta,node_cuda_cfg,node_cuda_postlink toneAmber class node_linux_cpu,node_linux_cuda117,node_linux_cuda12x,node_linux_cuda124,node_win_cpu,node_win_cuda_nmake,node_win_cuda124,node_mac_cpu1,node_mac_cpu2,node_mac_winlinux toneMint class node_forge_cpu,node_forge_cuda toneRose class node_smoketest_cpu,node_smoketest_cuda,node_root_smoketest toneIndigo ``` ## Components ### Docs Narrative and visual support files. - [README](https://github.com/agrim-p777/conda-package-genn/blob/main/README.md): docs · Explains the packaging goals, platform splits, and release story. · `README.md` - [Assets](https://github.com/agrim-p777/conda-package-genn/tree/main/assets): docs · Supporting visual and branding files. · `assets` ### Core recipes Primary conda packaging definitions for the CPU and CUDA product lines. - [CPU package](https://github.com/agrim-p777/conda-package-genn/tree/main/pygenn-cpu): recipe family · Top-level CPU-only packaging path for PyGeNN. · `pygenn-cpu` - [CUDA package](https://github.com/agrim-p777/conda-package-genn/tree/main/pygenn-cuda): recipe family · Top-level GPU-enabled packaging path for PyGeNN. · `pygenn-cuda` - [CPU meta](https://github.com/agrim-p777/conda-package-genn/blob/main/pygenn-cpu/meta.yaml): conda recipe · Authoritative CPU package metadata, dependencies, and build commands. · `pygenn-cpu/meta.yaml` - [CPU config](https://github.com/agrim-p777/conda-package-genn/blob/main/pygenn-cpu/conda_build_config.yaml): variant config · CPU build variant and version pin selection. · `pygenn-cpu/conda_build_config.yaml` - [CUDA meta](https://github.com/agrim-p777/conda-package-genn/blob/main/pygenn-cuda/meta.yaml): conda recipe · Authoritative CUDA package metadata, dependencies, and build commands. · `pygenn-cuda/meta.yaml` - [CUDA config](https://github.com/agrim-p777/conda-package-genn/blob/main/pygenn-cuda/conda_build_config.yaml): variant config · CUDA build variant and version pin selection. · `pygenn-cuda/conda_build_config.yaml` - [Post-link](https://github.com/agrim-p777/conda-package-genn/blob/main/pygenn-cuda/post-link.sh): install hook · Install-time hooks for CUDA environment setup on Unix and Windows. · `pygenn-cuda/post-link.sh` ### Platform variants OS- and CUDA-specific recipe branches and build adaptations. - [Linux CPU](https://github.com/agrim-p777/conda-package-genn/tree/main/pygenn-linux-cpu): platform recipe · Linux-specific CPU packaging branch. · `pygenn-linux-cpu` - [Linux CUDA 11.7](https://github.com/agrim-p777/conda-package-genn/blob/main/pygenn-linux-cuda11.7): platform recipe · Linux CUDA packaging branch pinned to CUDA 11.7. · `pygenn-linux-cuda11.7` - [Linux CUDA 12.x](https://github.com/agrim-p777/conda-package-genn/blob/main/pygenn-linux-cuda12.x): platform recipe · Linux CUDA packaging branch for the CUDA 12 family. · `pygenn-linux-cuda12.x` - [Linux CUDA 12.4](https://github.com/agrim-p777/conda-package-genn/blob/main/pygenn-linux-cuda12.4): platform recipe · Linux CUDA packaging branch pinned to CUDA 12.4. · `pygenn-linux-cuda12.4` - [Windows CPU](https://github.com/agrim-p777/conda-package-genn/tree/main/pygenn-win-cpu): platform recipe · Windows-specific CPU packaging branch. · `pygenn-win-cpu` - [Win CUDA NMake](https://github.com/agrim-p777/conda-package-genn/tree/HEAD/pygenn-win-cuda%20(nmake)): platform recipe · Windows CUDA packaging branch using NMake-oriented build mechanics. · `pygenn-win-cuda (nmake)` - [Win CUDA 12.4](https://github.com/agrim-p777/conda-package-genn/blob/main/pygenn-win-cuda12.4): platform recipe · Windows CUDA packaging branch pinned to CUDA 12.4. · `pygenn-win-cuda12.4` - [macOS CPU A](https://github.com/agrim-p777/conda-package-genn/tree/HEAD/pygenn-cpu%20(mac01)): platform recipe · macOS CPU packaging branch for one recipe variant. · `pygenn-cpu (mac01)` - [macOS CPU B](https://github.com/agrim-p777/conda-package-genn/tree/HEAD/pygenn-cpu%20(mac02)): platform recipe · macOS CPU packaging branch with helper build scripts and loader handling. · `pygenn-cpu (mac02)` - [Win/Linux CPU](https://github.com/agrim-p777/conda-package-genn/tree/HEAD/pygenn-cpu%20(winlinux)): platform recipe · Shared CPU packaging branch for Windows and Linux. · `pygenn-cpu (winlinux)` ### Conda-Forge Recipe layout for Conda-Forge publication and CI. - [Forge CPU](https://github.com/agrim-p777/conda-package-genn/tree/main/Conda_Forge/pygenn-cpu): forge recipe · Conda-Forge aligned CPU recipe layout. · `Conda_Forge/pygenn-cpu` - [Forge CUDA](https://github.com/agrim-p777/conda-package-genn/tree/main/Conda_Forge/pygenn-cuda): forge recipe · Conda-Forge aligned CUDA recipe layout. · `Conda_Forge/pygenn-cuda` ### Validation Smoke tests that verify installed package behavior. - [CPU smoke](https://github.com/agrim-p777/conda-package-genn/blob/main/Conda_Forge/pygenn-cpu/tests/run_pygenn_cpu_smoketest.py): smoke test · Installed-package smoke test for the CPU build. · `Conda_Forge/pygenn-cpu/tests/run_pygenn_cpu_smoketest.py` - [CUDA smoke](https://github.com/agrim-p777/conda-package-genn/blob/main/Conda_Forge/pygenn-cuda/tests/run_pygenn_cuda_smoketest.py): smoke test · Installed-package smoke test for the CUDA build. · `Conda_Forge/pygenn-cuda/tests/run_pygenn_cuda_smoketest.py` - [Smoke suite](https://github.com/agrim-p777/conda-package-genn/blob/main/tests/test.py): smoke tests · Repository-level smoke testing entry point. · `tests/test.py` ## Connections - README → CPU package: describes · Documents the CPU packaging line. - README → CUDA package: describes · Documents the CUDA packaging line. - Assets → README: supports · Branding and illustrative support for the docs. - CPU package → CPU meta: centers on · The CPU package is defined by its recipe metadata. - CPU package → CPU config: variants by · CPU build matrix is selected through conda-build configuration. - CUDA package → CUDA meta: centers on · The CUDA package is defined by its recipe metadata. - CUDA package → CUDA config: variants by · CUDA build matrix is selected through conda-build configuration. - CUDA package → Post-link: installs · CUDA packaging includes install-time environment setup. - CPU meta → Linux CPU: specializes · Linux-specific CPU packaging branch. - CPU meta → Windows CPU: specializes · Windows-specific CPU packaging branch. - CPU meta → macOS CPU A: specializes · macOS CPU packaging branch. - CPU meta → macOS CPU B: specializes · Alternate macOS CPU packaging branch. - CPU meta → Win/Linux CPU: specializes · Shared Windows/Linux CPU branch. - CUDA meta → Linux CUDA 11.7: specializes · Linux CUDA 11.7 branch. - CUDA meta → Linux CUDA 12.x: specializes · Linux CUDA 12.x branch. - CUDA meta → Linux CUDA 12.4: specializes · Linux CUDA 12.4 branch. - CUDA meta → Win CUDA NMake: specializes · Windows CUDA branch using NMake build mechanics. - CUDA meta → Win CUDA 12.4: specializes · Windows CUDA 12.4 branch. - Forge CPU → CPU package: mirrors · Conda-Forge recipe layout mirrors the CPU package family. - Forge CUDA → CUDA package: mirrors · Conda-Forge recipe layout mirrors the CUDA package family. - CPU package → CPU smoke: validated by · CPU packaging is checked with an installed-environment smoke test. - CUDA package → CUDA smoke: validated by · CUDA packaging is checked with an installed-environment smoke test. - CPU smoke → Smoke suite: feeds · Repository-level test entry point includes the CPU smoke check. - CUDA smoke → Smoke suite: feeds · Repository-level test entry point includes the CUDA smoke check. - Post-link → CUDA smoke: supports · Install-time setup enables the CUDA smoke test to load correctly.