from multiprocessing import Process, Array, cpu_count
from functools import reduce
import sys

LENGTH = int(sys.argv[1])
N_PROC = int(sys.argv[2])

def sum_up_range(start, end, output, pos):
    sum = 0
    while start < end:
        sum += start
        start += 1

    output[pos] = sum

# create shared memory for N_PROC
partialSums = Array('l', range(N_PROC))
CHUNK_SIZE = LENGTH//N_PROC
print(f"Chunk size is {CHUNK_SIZE}, using {N_PROC} processes")

children = []
for i in range(N_PROC-1):
    p = Process(target=sum_up_range, args=(i * CHUNK_SIZE, (i+1) * CHUNK_SIZE, partialSums, i))
    p.start()
    children.append(p)

# the parent process does the last slice
sum_up_range((N_PROC-1) * CHUNK_SIZE, N_PROC * CHUNK_SIZE, partialSums, N_PROC-1)

# wait for children
for p in children:
    p.join()

print(sum(partialSums))