Document prebuilt wheel install path

Adds a wheel-first install section to docs/buildingRyzenLin.md so users
can `pip install mlir_air` instead of building from source. The previous
"Quick Start" was a source build using wheel dependencies — renamed to
clarify. Also populates GitHub release descriptions for the
latest-air-wheels and latest-air-wheels-no-rtti tags so the release page
itself shows install instructions and explains the RTTI vs no-RTTI
variants (the no-RTTI variant exists for downstream LLVM-no-RTTI
integration).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

Author: erwei-xilinx

.github/workflows/buildAIRWheels.yml

@@ -263,6 +263,35 @@ jobs:
           allowUpdates: true
           replacesArtifacts: true
           makeLatest: ${{ github.event_name == 'push' }}
+          body: |
+            ## MLIR-AIR Wheels
+
+            **Latest commit:** `${{ needs.get-air-project-commit.outputs.AIR_PROJECT_COMMIT }}`
+            **Platforms:** Linux x86_64 (Python 3.10–3.14), Windows AMD64 (Python 3.10–3.12)
+
+            ### Installation
+
+            ```bash
+            # MLIR-AIR (default: RTTI enabled)
+            pip install mlir_air -f https://github.com/Xilinx/mlir-air/releases/expanded_assets/latest-air-wheels
+
+            # MLIR-AIE
+            pip install mlir_aie -f https://github.com/Xilinx/mlir-aie/releases/expanded_assets/latest-wheels-3
+
+            # Peano (llvm-aie) — AIE backend compiler
+            pip install llvm-aie -f https://github.com/Xilinx/llvm-aie/releases/expanded_assets/nightly
+            ```
+
+            ### RTTI / no-RTTI Variants
+
+            Both RTTI-enabled and RTTI-disabled wheels are released for downstream integration convenience:
+
+            - [`latest-air-wheels`](https://github.com/Xilinx/mlir-air/releases/tag/latest-air-wheels) — RTTI **on**. Default; use this for standalone development.
+            - [`latest-air-wheels-no-rtti`](https://github.com/Xilinx/mlir-air/releases/tag/latest-air-wheels-no-rtti) — RTTI **off**. For integrating MLIR-AIR into a downstream project whose LLVM is built with `-DLLVM_ENABLE_RTTI=OFF`. Pair with `mlir_aie` from [`latest-wheels-no-rtti`](https://github.com/Xilinx/mlir-aie/releases/tag/latest-wheels-no-rtti).
+
+            When mixing wheels into a downstream build, all three of `mlir`, `mlir_aie`, and `mlir_air` must agree on RTTI.
+
+            Full setup guide: [docs/buildingRyzenLin.md](https://github.com/Xilinx/mlir-air/blob/main/docs/buildingRyzenLin.md#install-prebuilt-wheels-recommended)
 
   build-windows:
     name: Build and upload mlir_air wheels (Windows)
@@ -419,3 +448,32 @@ jobs:
           allowUpdates: true
           replacesArtifacts: true
           makeLatest: ${{ github.event_name == 'push' }}
+          body: |
+            ## MLIR-AIR Wheels
+
+            **Latest commit:** `${{ needs.get-air-project-commit.outputs.AIR_PROJECT_COMMIT }}`
+            **Platforms:** Linux x86_64 (Python 3.10–3.14), Windows AMD64 (Python 3.10–3.12)
+
+            ### Installation
+
+            ```bash
+            # MLIR-AIR (default: RTTI enabled)
+            pip install mlir_air -f https://github.com/Xilinx/mlir-air/releases/expanded_assets/latest-air-wheels
+
+            # MLIR-AIE
+            pip install mlir_aie -f https://github.com/Xilinx/mlir-aie/releases/expanded_assets/latest-wheels-3
+
+            # Peano (llvm-aie) — AIE backend compiler
+            pip install llvm-aie -f https://github.com/Xilinx/llvm-aie/releases/expanded_assets/nightly
+            ```
+
+            ### RTTI / no-RTTI Variants
+
+            Both RTTI-enabled and RTTI-disabled wheels are released for downstream integration convenience:
+
+            - [`latest-air-wheels`](https://github.com/Xilinx/mlir-air/releases/tag/latest-air-wheels) — RTTI **on**. Default; use this for standalone development.
+            - [`latest-air-wheels-no-rtti`](https://github.com/Xilinx/mlir-air/releases/tag/latest-air-wheels-no-rtti) — RTTI **off**. For integrating MLIR-AIR into a downstream project whose LLVM is built with `-DLLVM_ENABLE_RTTI=OFF`. Pair with `mlir_aie` from [`latest-wheels-no-rtti`](https://github.com/Xilinx/mlir-aie/releases/tag/latest-wheels-no-rtti).
+
+            When mixing wheels into a downstream build, all three of `mlir`, `mlir_aie`, and `mlir_air` must agree on RTTI.
+
+            Full setup guide: [docs/buildingRyzenLin.md](https://github.com/Xilinx/mlir-air/blob/main/docs/buildingRyzenLin.md#install-prebuilt-wheels-recommended)

docs/buildingRyzenLin.md

@@ -1,6 +1,78 @@
 # Getting Started and Running on Linux Ryzen™ AI
 
-## Quick Start: Build with Prebuilt Wheels
+## Install Prebuilt Wheels (Recommended)
+
+The fastest way to get MLIR-AIR is to install the prebuilt wheel — no source build, no CMake, no LLVM clone. Use this path unless you need to modify MLIR-AIR itself or run on an unsupported Python/platform.
+
+### Prerequisites
+
+- **Python 3.10–3.14** on Linux x86_64, or **Python 3.10–3.12** on Windows AMD64
+- **pip**
+- **XRT** (optional, required only for running on hardware) — see [mlir-aie's XRT install instructions](https://github.com/Xilinx/mlir-aie#install-the-xdna-driver-and-xrt)
+
+### Steps
+
+1. **Create a virtual environment:**
+   ```bash
+   python3 -m venv airenv
+   source airenv/bin/activate
+   pip install --upgrade pip
+   ```
+
+2. **Install MLIR-AIR and its companion wheels:**
+   ```bash
+   # MLIR-AIR (default: RTTI-enabled)
+   pip install mlir_air -f https://github.com/Xilinx/mlir-air/releases/expanded_assets/latest-air-wheels
+
+   # MLIR-AIE
+   pip install mlir_aie -f https://github.com/Xilinx/mlir-aie/releases/expanded_assets/latest-wheels-3
+
+   # Peano (llvm-aie) — AIE backend compiler
+   pip install llvm-aie -f https://github.com/Xilinx/llvm-aie/releases/expanded_assets/nightly
+   ```
+
+3. **Set up environment variables** (paths derived from `pip show`):
+   ```bash
+   export MLIR_AIR_INSTALL_DIR=$(python3 -m pip show mlir_air | grep ^Location: | awk '{print $2}')/mlir_air
+   export MLIR_AIE_INSTALL_DIR=$(python3 -m pip show mlir_aie | grep ^Location: | awk '{print $2}')/mlir_aie
+   export PEANO_INSTALL_DIR=$(python3 -m pip show llvm-aie | grep ^Location: | awk '{print $2}')/llvm-aie
+
+   export PATH=$MLIR_AIR_INSTALL_DIR/bin:$MLIR_AIE_INSTALL_DIR/bin:$PATH
+   export PYTHONPATH=$MLIR_AIR_INSTALL_DIR/python:$MLIR_AIE_INSTALL_DIR/python:$PYTHONPATH
+   export LD_LIBRARY_PATH=$MLIR_AIR_INSTALL_DIR/lib:$MLIR_AIE_INSTALL_DIR/lib:$LD_LIBRARY_PATH
+
+   # Only if you have XRT installed:
+   source /opt/xilinx/xrt/setup.sh
+   ```
+
+4. **Verify the install:**
+   ```bash
+   air-opt --help
+   ```
+
+You can now jump to [Running a Quick Example](#running-a-quick-example) below.
+
+### Choosing a Wheel Release
+
+| Tag | When to use |
+|-----|-------------|
+| [`latest-air-wheels`](https://github.com/Xilinx/mlir-air/releases/tag/latest-air-wheels) | Default. RTTI enabled. Use this for standalone development. |
+| [`latest-air-wheels-no-rtti`](https://github.com/Xilinx/mlir-air/releases/tag/latest-air-wheels-no-rtti) | For integrating MLIR-AIR into a downstream project whose LLVM is built with `-DLLVM_ENABLE_RTTI=OFF`. Pair with `mlir_aie` from `latest-wheels-no-rtti`. |
+| [`v*.*.*`](https://github.com/Xilinx/mlir-air/releases) (e.g. `v1.0.0`) | Pinned tagged release, for reproducible builds: `pip install mlir_air==<ver> -f https://github.com/Xilinx/mlir-air/releases/expanded_assets/v<ver>` |
+
+We release **both RTTI and no-RTTI variants** so downstream projects can match their LLVM build configuration without having to rebuild MLIR-AIR from source. When mixing wheels into a downstream build, all three (`mlir`, `mlir_aie`, `mlir_air`) must agree on RTTI.
+
+### Running Examples Against a Wheel Install
+
+When checking out the MLIR-AIR repo to run programming examples, sync to the commit hash embedded in the wheel filename (`mlir_air-0.0.1.{DATETIME}+{HASH}-...`) so the example sources match the installed compiler.
+
+If the wheel install path doesn't fit your needs, see the source-build instructions below.
+
+---
+
+## Build From Source Using Prebuilt Dependencies
+
+This path builds MLIR-AIR from source but installs LLVM, MLIR-AIE, and Peano from prebuilt wheels — useful when you're hacking on MLIR-AIR itself but don't want to also rebuild its dependencies.
 
 ### Prerequisites