#include <stdio.h>
#include <cuda.h>
#include "secp256k1.h"
#include "sha256.h"
#include "ripemd160.h"

__global__ void hash(stuff) {

    secp(stuff);
    sha256(stuff);
    ripemd160(stuff);
}

int main() {
     allocate host memory array
     allocate cuda memory
     open file and put in host array

     cuda memcpy host to device

    hash<<<1,1>>>(stuff);

    cuda memcpy device to host

   printf("%sX\n", stuff);
}

now just add incremental threading and actual code.

#include <stdio.h>
#include <cuda.h>
#include "secp256k1.h"
#include "sha256.h"
#include "ripemd160.h"

__global__ void hash(stuff) {

    secp(stuff)
    sha256(stuff)
    ripemd160(stuff)
}

int main() {
     allocate host memory array
     allocate cuda memory
     open file and put in host array

     cuda memcpy host to device

    hash<<<1,1>>>(stuff);

    cuda memcpy device to host

   printf("%s\n", stuff);
}

#include <stdio.h>
#include <cuda.h>
#include "secp256k1.h"
#include "sha256.h"
#include "ripemd160.h"

__global__ hash() {


}