Running Software Emulation - 2023.2 English

Vitis Tutorials: Hardware Acceleration (XD099)

Document ID
XD099
Release Date
2023-11-13
Version
2023.2 English

To build and run the kernel in software emulation mode, please run the following bash command:

make run TARGET=sw_emu

It will build an xclbin file to be used in software emulation mode only and launch an emulation run. Once the emulation finishes, you should get a console output similar to the one below. The output given below is for random input image case:

----------------------------------------------------------------------------

Xilinx 2D Filter Example Application (Randomized Input Version)

FPGA binary       : ./fpgabinary.sw_emu.xclbin
Number of runs    : 1
Image width       : 1920
Image height      : 1080
Filter type       : 5
Max requests      : 12
Compare perf.     : 1

Programming FPGA device
Generating a random 1920x1080 input image
Running FPGA accelerator on 1 images
  finished Filter2DRequest
  finished Filter2DRequest
  finished Filter2DRequest
Running Software version
Comparing results

Test PASSED: Output matches reference
----------------------------------------------------------------------------

If the input image is to be used for processing, please set OpenCV paths and INPUT_TYPE as an empty string in make_options.mk file and run it again. Following is the expected console output:

----------------------------------------------------------------------------

Xilinx 2D Filter Example Application

FPGA binary       : ./fpgabinary.sw_emu.xclbin
Input image       : ../test_images/picadilly_1080p.bmp
Number of runs    : 1
Filter type       : 3
Max requests      : 12
Compare perf.     : 1

Programming FPGA device
Running FPGA accelerator on 1 images
  finished Filter2DRequest
  finished Filter2DRequest
  finished Filter2DRequest
Running Software version
Comparing results

Test PASSED: Output matches reference
----------------------------------------------------------------------------

The input and the output images are shown below for filter type selection set to 3 which performs edge detection:

missing image missing image