Add Support for Arm-based CPU (#3)

* Create README.md

* Update README.md

* Add support for arm-based CPU

* Update README.md

---------

Co-authored-by: Rifky Bujana Bisri <[email protected]>

Author: althaafsn

CMakeLists.txt

@@ -15,6 +15,14 @@ set(CMAKE_CXX_STANDARD 17)
 set(CMAKE_CXX_STANDARD_REQUIRED ON)
 set(CMAKE_WINDOWS_EXPORT_ALL_SYMBOLS ON)
 
+# Detect if the architecture is ARM (AArch64)
+if(CMAKE_SYSTEM_PROCESSOR MATCHES "^(aarch64|ARM64|arm64|armv8|armv7)")
+    message(STATUS "Building on ARM: Enabling -fPIC")
+    set(CMAKE_POSITION_INDEPENDENT_CODE ON)
+else()
+    message(STATUS "Building on x86: No -fPIC required")
+endif()
+
 ucm_print_flags()
 
 # Source files
@@ -170,4 +178,4 @@ target_link_libraries(test_inference PRIVATE ${TARGET_NAME})
 # Put the test in the bin directory
 set_target_properties(test_inference PROPERTIES
   RUNTIME_OUTPUT_DIRECTORY "${CMAKE_BINARY_DIR}/bin"
-)
+)

README.md

@@ -0,0 +1,49 @@
+## 🚀 Setup
+
+To ensure compatibility, this project uses **llama.cpp version b4514**``.
+
+### 1️⃣ Clone the Repository
+
+```sh
+git clone https://github.com/genta-technology/inference-personal
+cd inference-personal
+```
+
+### 2️⃣ Setup `llama.cpp`
+
+```sh
+cd external
+git clone https://github.com/ggml-org/llama.cpp.git
+cd llama.cpp
+git checkout b4514
+git switch -c b4514
+git branch  # Ensure the branch is `b4514` or lower
+git submodule update --init --recursive
+git lfs install
+git lfs pull
+```
+
+## 🛠️ Build Instructions
+
+### 1️⃣ Create Build Directory
+
+```sh
+cd ../..
+mkdir build && cd build
+```
+
+### 2️⃣ Compile the Project
+
+```sh
+cmake ..
+make
+```
+
+## ✅ Running Tests
+
+After a successful build, run the test binary:
+
+```sh
+./bin/test
+```
+