mkl_delphi - Intel MKL Wrapper for Delphi Integration

mkl_delphi is a dynamic library written in C, providing a foundation for integrating the Intel Math Kernel Library (MKL) into Delphi applications. It currently exposes the highly optimized CBLAS cblas_dgemm routine and serves as an open-source template for expanding Delphi-MKL interoperability in High-Performance Computing (HPC) environments.

Features

• Intel MKL Integration: Provides C binding to Intel MKL's industry-standard CBLAS cblas_dgemm routine, unlocking maximum hardware performance for dense linear algebra.
• Delphi Wrapper: Includes a meticulously crafted Delphi wrapper (MKL.pas) that utilizes $Z4 binary-compatible enumerations for seamless and safe C-to-Delphi communication.
• Extensible Architecture: Designed as a robust, open-source starting point, allowing developers to easily add wrappers for additional MKL functions as needed.
• High-Performance Computing: Enables hardware-optimized computations for improved performance in intense algebraic tasks.
• Cross-Language Compatibility: Distributed as a DLL, accessible from various programming environments.

Requirements & Development Environment

The library is developed, tested, and intended to be used in the following environment:

• Operating System: Windows 11 (64-bit)
• Compiler: Intel C++ Compiler (icx-cc)
• Toolchain: Microsoft Visual Studio Build Tools (nmake)
• Dependencies: Intel Math Kernel Library (MKL) for headers (mkl_cblas.h) and linked binaries.
• Delphi Integration: RAD Studio (Delphi 12.1 Community Edition)

Toolchain Setup and Installation Steps

Unlike standard GCC setups, compiling against Intel MKL on Windows requires the official Intel oneAPI toolkit and Microsoft's build tools. Follow these steps carefully:

1: Install MSVC Build Tools (for nmake)

• Navigate to the Visual Studio Downloads page.

• Scroll down to the "Tools for Visual Studio" section and download the installer for Build Tools for Visual Studio 2026 (the version year may vary depending on the current release).

• Open the downloaded vs_BuildTools.exe installer.

• In the workloads tab, check the box for Desktop development with C++.

• In the installation details column on the right, ensure that MSVC Build Tools for x64/x86 (latest) is checked.

• Click Install.

• Once installed, open the Developer Command Prompt for VS from your Windows start menu and type nmake /?. If the help text appears, the installation was successful.

2: Install Intel oneAPI Toolkit (for icx and MKL)

• Navigate to the Intel oneAPI Toolkit Download page.

• Click on "Continue as Guest..." to download the offline or online installer.

• Run the installer. When prompted to select components, ensure that both the Intel C++ Compiler (icx) and the Intel Math Kernel Library (MKL) are selected for installation.

• Proceed with the default installation paths.

Project Structure

mkl_delphi/
│
├── build/             # Compiled binaries and intermediate files
├── include/           # Public API headers (C)
│   └── mkl_delphi.h
├── interface/         # Delphi wrapper
│   └── MKL.pas
├── src/               # C source code
│   └── mkl_delphi.c
├── LICENSE            # License information
├── Makefile           # Build script for DLL
└── README.md          # Project documentation

Compilation Instructions

Once your build tools and Intel MKL are properly installed, follow these steps to compile the DLL:

1. Open a Command Prompt.

2. Set the VS2022INSTALLDIR variable so that Intel's setvars.bat can locate the MSVC Build Tools correctly:

set "VS2022INSTALLDIR=C:\Program Files (x86)\Microsoft Visual Studio\18\BuildTools"

Note: you can skip this step if you opened Developer Command Prompt instead.

3. Initialize the Intel oneAPI environment variables by running the setvars.bat script (adjust the path if you installed oneAPI in a custom directory):

call "C:\Program Files (x86)\Intel\oneAPI\setvars.bat"

4. Navigate to the root directory of this repository:

cd path\to\mkl_delphi

5. Run the build tool to compile the DLL:

nmake

This will generate mkl_delphi.dll in the build\ directory.

Using the Library in Delphi

1. Linking the DLL (Choose One Method)

To allow your Delphi application to load and communicate with the dynamic library, you must make mkl_delphi.dll (and the underlying MKL runtime DLLs) discoverable by Windows. You can achieve this using one of the following methods:

Method A: Add the Build Directory to the System Path (Recommended for Development)
During active development of the C library, you might need to recompile the DLL frequently. Copying the file manually after every compilation is tedious and error-prone.

1. Add the absolute path of your local mkl_delphi\build\ directory and the Intel MKL lib directory (e.g.: C:\Program Files (x86)\Intel\oneAPI\mkl\2026.0\lib) directly to your Windows system's PATH environment variable following the next steps for each directory:

• Press Win + S, type "env", and select Edit the system environment variables.

• In the System Properties window, click on the Environment Variables... button.

• Under System variables (bottom list), select the variable named Path and click Edit....

• Click New on the right side and type the absolute path to the directory.

• Click OK to close each of the three setup windows

Method B: Place the DLL in the Executable Folder (Recommended for Distribution)
If you are packaging, deploying, or sharing your compiled Delphi application:

1. Locate your Delphi project's output directory (usually named Win64\Debug or Win64\Release inside your project's folder structure).

2. Copy mkl_delphi.dll from mkl_delphi\build\ and paste it directly into that folder, alongside your compiled .exe file.

3. You must also include any required Intel MKL redistributable DLLs in this same folder.

2. Integrating the Wrapper in RAD Studio / Delphi

To use the library functions natively inside your Delphi Pascal source code, follow these steps:

1. Open your Delphi project in RAD Studio.

2. Go to the top menu and select Project > Options....

3. In the left panel, navigate to Building > Delphi Compiler.

4. In the main area, click the Search Path field.

5. Click on the ellipsis ... button and add the absolute path to the interface\ folder inside your repository (e.g.: C:\Projects\mkl_delphi\interface).

6. Click OK and then Save.

Example Usage:

uses MKL;

// Example: Multiply matrices A and B into C using Intel MKL's highly optimized DGEMM
// Operation: C := alpha*op(A)*op(B) + beta*C
cblas_dgemm(CblasRowMajor, CblasNoTrans, CblasNoTrans, 
            M, N, K, 
            1.0, @MatrixA[0], K, 
            @MatrixB[0], N, 
            0.0, @MatrixC[0], N);

Academic Citation

If you use this software in your research, please cite it using the metadata provided in the CITATION.cff file located in the root of this repository.

License

This project is licensed under the MIT License - see the LICENSE file for details.