Visible to Intel only — GUID: GUID-F3FCD3B3-7FA5-4A62-993A-A4170F7A4625
Why is FPGA Compilation Different?
Types of SYCL* FPGA Compilation
FPGA Compilation Flags
Emulate and Debug Your Design
Evaluate Your Kernel Through Simulation
Device Selectors for FPGA
FPGA IP Authoring Flow
Fast Recompile for FPGA
Generate Multiple FPGA Images (Linux only)
FPGA BSPs and Boards
Targeting Multiple Homogeneous FPGA Devices
Targeting Multiple Platforms
FPGA-CPU Interaction
FPGA Performance Optimization
Use of RTL Libraries for FPGA
Use SYCL Shared Library With Third-Party Applications
FPGA Workflows in IDEs
Intel oneAPI DPC++ Library (oneDPL)
Intel oneAPI Math Kernel Library (oneMKL)
Intel oneAPI Threading Building Blocks (oneTBB)
Intel oneAPI Data Analytics Library (oneDAL)
Intel oneAPI Collective Communications Library (oneCCL)
Intel oneAPI Deep Neural Network Library (oneDNN)
Intel oneAPI Video Processing Library (oneVPL)
Other Libraries
Visible to Intel only — GUID: GUID-F3FCD3B3-7FA5-4A62-993A-A4170F7A4625
Compile and Emulate Your Design
To compile and emulate your FPGA kernel design, perform the following steps:
Modify the host part of your program to declare the ext::intel::fpga_emulator_selector device selector. Use this device_selector when instantiating a device queue for enqueuing your FPGA device kernel.
Compile your design by including the -fintelfpga option in your icpx command to generate an executable.
Run the resulting executable:
For Windows:
Define the number of emulated devices by invoking the following command:
set CL_CONFIG_CPU_EMULATE_DEVICES=<number_of_devices>
Run the executable.
Invoke the following command to unset the variable:
set CL_CONFIG_CPU_EMULATE_DEVICES=
For Linux, invoke the following command:
env CL_CONFIG_CPU_EMULATE_DEVICES=<number_of_devices> <executable_filename>
This command specifies the number of identical emulation devices that the emulator must provide.
TIP:If you want to use only one emulator device, you need not set the CL_CONFIG_CPU_EMULATE_DEVICES environment variable.
NOTE:
- The Intel® FPGA Emulation Platform for OpenCL™ does not provide access to physical boards. Only the emulated devices are available.
The emulator is built with GCC 7.4.0 as part of the Intel® oneAPI DPC++/C++ Compiler. When running the executable for an emulated FPGA device, the version of libstdc++.so must be at least that of GCC 7.4.0. In other words, the LD_LIBRARY_PATH environment variable must ensure that the correct version of libstdc++.so is found.
If the correct version of libstdc++.so is not found, the call to clGetPlatformIDs function fails to load the FPGA emulator platform and returns CL_PLATFORM_NOT_FOUND_KHR (error code -1001). Depending on which version of libstdc++.so is found, the call to clGetPlatformIDs may succeed, but a later call to the clCreateContext function may fail with CL_DEVICE_NOT_AVAILABLE (error code -2).
If the LD_LIBRARY_PATH does not point to a compatible libstdc++.so, use the following syntax to invoke the host program:
env LD_LIBRARY_PATH=<path to compatible libstdc++.so>:$LD_LIBRARY_PATH <executable> [executable arguments]
To enable debugging of kernel code, optimizations are disabled by default for the FPGA emulator. This can lead to sub-optimal execution speed when emulating kernel code. You can pass the -g0 flag to the icpx compile command to disable debugging and enable optimizations. This enables faster emulator execution.
When targeting the FPGA emulator device, use the -O2 compiler flag to turn on optimizations and speed up the emulation. To turn off optimizations (for example, to facilitate debugging), pass -O0.