Update LlamaDemo AAR build docs (pytorch#10086)

Author: kirklandsign

examples/demo-apps/android/LlamaDemo/docs/delegates/mediatek_README.md

@@ -126,13 +126,31 @@ The Mediatek runner (`examples/mediatek/executor_runner/mtk_llama_runner.cpp`) c
 
 
 ## Build AAR Library
+1. Open a terminal window and navigate to the root directory of the executorch
+2. Set the following environment variables:
+```sh
+export ANDROID_NDK=<path_to_android_ndk>
+export ANDROID_ABIS=arm64-v8a
+export NEURON_BUFFER_ALLOCATOR_LIB=<path_to_neuron_buffer_allocator_lib>
+```
+*Note: <path_to_android_ndk> is the root for the NDK, which is usually under ~/Library/Android/sdk/ndk/XX.Y.ZZZZZ for macOS, and contains NOTICE and README.md. We use <path_to_android_ndk>/build/cmake/android.toolchain.cmake for CMake to cross-compile.*
 
-Next we need to build and compile the MediaTek backend and MediaTek Llama runner. By setting  `NEURON_BUFFER_ALLOCATOR_LIB`, the script will build the MediaTek backend.
+3. Create a directory to hold the AAR
+```sh
+mkdir -p aar-out
+export BUILD_AAR_DIR=aar-out
 ```
+
+4. Run the following command to build the AAR:
+```sh
 sh scripts/build_android_library.sh
 ```
 
-**Output**: This will generate an .aar file that is already imported into the expected directory for the Android app. It will live in `examples/demo-apps/android/Llamademo/app/libs`.
+5. Copy the AAR to the app:
+```sh
+mkdir -p examples/demo-apps/android/LlamaDemo/app/libs
+cp aar-out/executorch.aar examples/demo-apps/android/LlamaDemo/app/libs/executorch.aar
+```
 
 If you were to unzip the .aar file or open it in Android Studio, verify it contains the following related to MediaTek backend:
 * libneuron_buffer_allocator.so

examples/demo-apps/android/LlamaDemo/docs/delegates/qualcomm_README.md

@@ -178,30 +178,38 @@ adb push tokenizer.bin /data/local/tmp/llama
 
 
 ## Build AAR Library
-Open a terminal window and navigate to the root directory of the executorch.
+1. Open a terminal window and navigate to the root directory of the executorch
 Set the following environment variables:
-```
+```sh
 export ANDROID_NDK=<path_to_android_ndk>
-export ANDROID_ABI=arm64-v8a
-```
-Note: <path_to_android_ndk> is the root for the NDK, which is usually under ~/Library/Android/sdk/ndk/XX.Y.ZZZZZ for macOS, and contains NOTICE and README.md. We use <path_to_android_ndk>/build/cmake/android.toolchain.cmake for CMake to cross-compile.
-Build the Android Java extension code:
+export ANDROID_ABIS=arm64-v8a
+export QNN_SDK_ROOT=<path_to_qnn_sdk>
 ```
-pushd extension/android
-./gradlew build
-popd
+
+*Note: <path_to_android_ndk> is the root for the NDK, which is usually under ~/Library/Android/sdk/ndk/XX.Y.ZZZZZ for macOS, and contains NOTICE and README.md. We use <path_to_android_ndk>/build/cmake/android.toolchain.cmake for CMake to cross-compile.*
+
+3. Create a directory to hold the AAR
+```sh
+mkdir -p aar-out
+export BUILD_AAR_DIR=aar-out
 ```
-Run the following command set up the required JNI library:
+
+4. Run the following command to build the AAR:
+```sh
+sh scripts/build_android_library.sh
 ```
-pushd examples/demo-apps/android/LlamaDemo
-./gradlew :app:setupQnn
-popd
+
+5. Copy the AAR to the app:
+```sh
+mkdir -p examples/demo-apps/android/LlamaDemo/app/libs
+cp aar-out/executorch.aar examples/demo-apps/android/LlamaDemo/app/libs/executorch.aar
 ```
+
 Alternative you can also just run the shell script directly as in the root directory:
-```
+```sh
 sh examples/demo-apps/android/LlamaDemo/setup-with-qnn.sh
 ```
-This is running the shell script which configures the required core ExecuTorch, Llama2/3, and Android libraries, builds them, and copies them to jniLibs.
+This is running the shell script which configures the required core ExecuTorch, Llama2/3, and Android libraries, builds them into AAR, and copies it to the app.
 Note: If you are building the Android app mentioned in the next section on a separate machine (i.e. MacOS but building and exporting for QNN backend on Linux), make sure you copy the aar file generated from setup-with-qnn script to “examples/demo-apps/android/LlamaDemo/app/libs” before building the Android app.
 
 

examples/demo-apps/android/LlamaDemo/docs/delegates/xnnpack_README.md

@@ -132,30 +132,35 @@ adb push tokenizer.bin /data/local/tmp/llama
 ## Build AAR Library
 1. Open a terminal window and navigate to the root directory of the executorch
 2. Set the following environment variables:
-```
+```sh
 export ANDROID_NDK=<path_to_android_ndk>
-export ANDROID_ABI=arm64-v8a
+export ANDROID_ABIS=arm64-v8a
 ```
 *Note: <path_to_android_ndk> is the root for the NDK, which is usually under ~/Library/Android/sdk/ndk/XX.Y.ZZZZZ for macOS, and contains NOTICE and README.md. We use <path_to_android_ndk>/build/cmake/android.toolchain.cmake for CMake to cross-compile.*
 
-3. Build the Android Java extension code:
-```
-pushd extension/android
-./gradlew build
-popd
+3. Create a directory to hold the AAR
+```sh
+mkdir -p aar-out
+export BUILD_AAR_DIR=aar-out
 ```
-4. Run the following command set up the required JNI library:
+
+4. Run the following command to build the AAR:
+```sh
+sh scripts/build_android_library.sh
 ```
-pushd examples/demo-apps/android/LlamaDemo
-./gradlew :app:setup
-popd
+
+5. Copy the AAR to the app:
+```sh
+mkdir -p examples/demo-apps/android/LlamaDemo/app/libs
+cp aar-out/executorch.aar examples/demo-apps/android/LlamaDemo/app/libs/executorch.aar
 ```
+
 Alternative you can also just run the shell script directly as in the root directory:
-```
+```sh
 sh examples/demo-apps/android/LlamaDemo/setup.sh
 ```
 
-This is running the shell script which configures the required core ExecuTorch, Llama2/3, and Android libraries, builds them, and copies them to jniLibs.
+This is running the shell script which configures the required core ExecuTorch, Llama2/3, and Android libraries, builds them into AAR, and copies it to the app.
 
 **Output**: The executorch.aar file will be generated in a newly created folder in the example/demo-apps/android/LlamaDemo/app/libs directory. This is the path that the Android app expects it to be in.