generate VHDL test benches from an assembly file

gentest.c

#include <stdio.h>
#include <stdlib.h>

typedef char byte;
typedef unsigned int word;

char* header = 
"-- TESTCASE AUTOGENERATED by gentest\n"
"library ieee;\n"
"use ieee.std_logic_1164.all;\n"
"use ieee.numeric_std.all;\n"
"use work.MyTypes.all;\n"
"\n"
"entity cpu_tb is\n"
"end cpu_tb;\n"
"\n"
"architecture cpu_tb_arc of cpu_tb is\n"
"    signal clock: std_logic := '0';\n"
"    constant clock_period: time := 2 ns;\n"
"    signal memory: mem(127 downto 0) := (\n";

char* footer = "begin\n"
"    uut: entity work.cpu generic map (memory) port map (clock);\n"
"\n"
"    clock <= not clock after clock_period/2;\n"
"\n"
"end cpu_tb_arc;\n";


byte* get_file_as_bytes(char* fname) {
    FILE *fileptr;
    char *buffer;
    long filelen;
    
    fileptr = fopen(fname, "rb");  // Open the file in binary mode
    fseek(fileptr, 0, SEEK_END);          // Jump to the end of the file
    filelen = ftell(fileptr);             // Get the current byte offset in the file
    rewind(fileptr);                      // Jump back to the beginning of the file
    
    buffer = (char *)malloc(filelen * sizeof(char)); // Enough memory for the file
    fread(buffer, filelen, 1, fileptr); // Read in the entire file
    fclose(fileptr); // Close the file

    return buffer;
}

int main(int argc, char** argv) {
    
    // first argument contains assembly file to generate testcase for
    // second argument is output vhdl filename
    if (argc < 3) {
        printf("Incorrect format. Ending.");
        return 1;
    }

    char* fname = argv[1];
    // compile the file
    char* cmd = (char*)malloc(200);
    sprintf(cmd, "/Applications/ARM/bin/arm-none-eabi-as -march=armv4t -mbig-endian %s", fname);
    int result = system(cmd);
    if (result != 0) {
        printf("Could not compile file. Exiting.");
        return 1;
    }
    free(cmd);

    
    // post compilation, open the executable as a byte array
    byte* fbytes = get_file_as_bytes("a.out");

    // open output file for writing
    FILE *outfile;
    outfile = fopen(argv[2], "w");
    fprintf(outfile, header);

    // compiled as an ELF on my machine; commands start from 13th word, in 
    // big endian order (big endian makes a hexdump easier to read)
    for (int i = 13*4,j=0; ; i+=4,j++) {
        word cmd = ((word)fbytes[i] << 24) | 
                   ((word)fbytes[i+1] << 16) | 
                   ((word)fbytes[i+2] << 8) | 
                   ((word)fbytes[i+3]);
        if (cmd == (((word)0x41)<<24)) { // ELF command signifying end of instructions
            break;
        }
        else {
            fprintf(outfile, "        %d => X\"%02hhX%02hhX%02hhX%02hhX\",\n", j, fbytes[i], fbytes[i+1], fbytes[i+2], fbytes[i+3]);
        }
    }
    fprintf(outfile, "        others => X\"00000000\" );\n");
    fprintf(outfile, footer);

    free(fbytes);

    return 0;
}