Wide Memory Read/Write

This is simple example of vector addition to demonstrate Wide Memory Access using ap_uint<512> data type. Based on input argument type, V++ compiler will figure our the memory datawidth between Global Memory and Kernel. For this example, ap_uint<512> datatype is used, so Memory datawidth will be 16 x (integer bit size) = 16 x 32 = 512 bit.

KEY CONCEPTS: Kernel to DDR, wide memory access, burst read and write

KEYWORDS: ap_uint<>, ap_int.h

This example demonstrates the use of ap_uint<> datatype which enables V++ to figure out the memory datawidth for the kernel to global memory transfers.

For efficient use of this bandwidth, ap_uint<512> is used as interface between kernels and global memory allows for maximum of 512 bits to be transferred concurrently. User must include ap_int.h to use these datatypes.

#define DATAWIDTH 512
typedef ap_uint<DATAWIDTH> uint512_dt;

uint512_dt v1_local[BUFFER_SIZE];
uint512_dt result_local[BUFFER_SIZE];

Using 512 bits user can enable the burst read and write for the data by using #pragma HLS PIPELINE. Auto-pipeline is going to apply pipeline to these loops.

v1_rd:
       for (int j = 0; j < chunk_size; j++) {
          #pragma HLS LOOP_TRIPCOUNT min=c_chunk_sz max=c_chunk_sz
           v1_local[j] = in1[i + j];
       }

EXCLUDED PLATFORMS:

• Embedded ZC702

• All NoDMA Platforms, i.e u50 nodma etc

DESIGN FILES

Application code is located in the src directory. Accelerator binary files will be compiled to the xclbin directory. The xclbin directory is required by the Makefile and its contents will be filled during compilation. A listing of all the files in this example is shown below

src/host.cpp
src/vadd.cpp

Access these files in the github repo by clicking here.

COMMAND LINE ARGUMENTS

Once the environment has been configured, the application can be executed by

./wide_mem_rw <vadd XCLBIN>