[HLS] Call of queuing function in pipeline simulation

I. function introduction

①

ihc_hls_enqueue(void *retptr, void *funcptr,/*function arguments*/)

Parameters
retptr： Return value
funcptr： To be called HLS component

This function is right for HLS A call of the component is queued . The return value is stored in the first argument , The argument should be a pointer to the return type . Calling ihc_hls_component_run_all() Before , The component will not run .

②

ihc_hls_enqueue_noret(void* funcptr,/*function arguments*/)

Parameters
funcptr： To be called HLS component

This function is right for HLS A call of the component is queued . When HLS The return type of the component is void when , This function should be used . Calling ihc_hls_component_run_all() Before , The component will not run .

③

ihc_hls_component_run_all(void* funcptr)

Parameters
funcptr： To be called HLS component

This function accepts a point HLS Pointer to component function . All queued calls of the component will be pushed at run time , When the component can accept new calls ,HDL The simulator can run at the fastest speed .

2. Compilation instructions

①

i++ -march=x86-64 <.c  perhaps  .cpp file > -v

This step is just to use g++ or i++ The compiler performs function verification

②

i++ -march=CycloneV <.c  perhaps  .cpp file > -v -ghdl

compile HLS, Generate IP, Check the reports generated by the compiler , Verify the design in simulation

③
Run the generated under the directory exe Executable file , If not running , There will be no waveform in the next simulation

④

vsim a.prj/verification/vsim.wlf

3. Adder

Code design ：

#include "HLS/stdio.h"
#include "HLS/hls.h"
#include "assert.h"
#include "stdlib.h"

component int add(int a,int b)
{
    
    return a+b;
}

int main()
{
    
    srand(0);
    int i,x,y,z;
    #if 0
    for ( i = 0; i < 10; i++)
    {
    
        x = rand() % 10;
        y = rand() % 10;
        ihc_hls_enqueue(&z,&add,x,y);
    }
    ihc_hls_component_run_all(add);
    return 0;
    
    #else 
    for ( i = 0; i < 10; i++)
    {
    
        x = rand() % 10;
        y = rand() % 10;
        z = add(x,y);
        printf("%d = %d + %d\n",z,x,y);
        assert(z == x+y);
    }
    return 0;

    #endif
}

1. No assembly line

It can be seen from the waveform that Is a function call , Wait for the result to be output once , Then you can call the function again

2. Assembly line

Call functions accumulatively first acc, Waiting ihc_hls_component_run_all Function call play , Queue function before one run .

4. Multiplier

Code design ：

#include "HLS/stdio.h"
#include "HLS/hls.h"
#include "assert.h"
#include "stdlib.h"

component int multiply(int a,int b)
{
    
    return a*b;
}

int main()
{
    
    srand(0);
    int i,x[10]={
    0},y[10]={
    0},z[10]={
    0};
    #if 0

    for ( i = 0; i < 10; i++)
    {
    
        x[i] = rand() % 10;
        y[i] = rand() % 10;
        z[i] = multiply(x[i],y[i]);
        printf("%d = %d * %d\n",z[i],x[i],y[i]);
    }
    return 0;
    #else
    for ( i = 0; i < 10; i++)
    {
    
        x[i] = rand() % 10;
        y[i] = rand() % 10;
        
        ihc_hls_enqueue(&z[i],&multiply,x[i],y[i]);
    }
    ihc_hls_component_run_all(multiply);
    return 0;
    #endif
}

1. No assembly line

2. Assembly line

3. Report analysis

Follow the path to find the report analysis

Open it to see the resource usage

Finally, comparing the reports of the two methods, it is found that the resources consumed are the same , Only here is the difference

Reference article ：https://blog.csdn.net/qq_42585108/article/details/120614492