JPEG Huffman Decoder

Jpeg Huffman Decoder example resides in L1/tests/jpegDec directory. The tutorial provides a step-by-step guide that covers commands for building and running IP.

Executable Usage

  • Work Directory(Step 1)

The steps for library download and environment setup can be found in l1_vitis_codec. For getting the design,

cd L1/tests/jpegDec
  • Run and Build IP(Step 2)

Run the following make command to build your IP targeting a specific device. Please be noticed that this process will take a long time, maybe couple of hours.

make run DEVICE=xilinx_u50_gen3x16_xdma_201920_3 VIVADO_IMPL=1

Usage: host.exe -[-xclbin -JPEGFile]
       -JPEGFile:  the path point to input *.jpg

Note: Default arguments are set in Makefile, you can use other Pictures & Performance listed in the table.

  • Example output(Step 2)
...
Time resolution is 1 ps
...
RTL Simulation : 0 / 1 [n/a] @ "106000"
RTL Simulation : 1 / 1 [n/a] @ "345801000"
...
INFO: [COSIM 212-316] Starting C post checking ...
INFO: [COSIM 212-1000] *** C/RTL co-simulation finished: PASS ***
...
Implementation tool: Xilinx Vivado v.2021.2
Project:             test.prj
Solution:            solution1
...
#=== Post-Implementation Resource usage ===
SLICE:            0
LUT:           7939
FF:            8043
DSP:             12
BRAM:             5
URAM:             0
LATCH:            0
SRL:            717
CLB:           1748

#=== Final timing ===
CP required:                     2.500
CP achieved post-synthesis:      4.564
CP achieved post-implementation: 3.703

Profiling

The hardware resource utilizations are listed in the following table. Different tool versions may result slightly different resource.

Table 1 IP resources for jpeg Huffman Decoder with jfif parser and huffman decoder
IP BRAM URAM DSP FF LUT Frequency(MHz)
kernel_parser_decoder 5 0 12 7939 8043 270

Table 2 : Jpeg Huffman Decoder IP profiling

Jpeg huffman Decoder IP profiling

Note

1. MAX_DEC_PIX is for benchmark. If testcase image is larger than 20M, the value of MAX_DEC_PIX should be enlarged following the size of image.
2. MAXCMP_BC is for benchmark. If testcase image is larger than 20M, the value of MAXCMP_BC should be enlarged following the size of image.