# Peasantmultiplication: 
# %rdi and %rsi to be multiplied and return result in %rax
# INPUT %rdi and %rsi
# OUTPUT %rax
# Other internal register: r8 and r9
multiply:
    # Rax to zero
    irmovq $0, %rax
    
    # r8 is place for the bit mask
    irmovq $1, %r8
    
    # Bitmask in r8 is compared against parameter in rsi, rdi is bit shifted and added to sum depending about comparison
    
    pmloop:
        # In this loop:
        # Check if each bit from right to left is 1
        # If yes, then add rdi to rax
        
        # r9 helps to store comparison result temporally
        rrmovq %r8, %r9
        
        # Get single bit with mask
        andq %rsi, %r9
        
        # Check if result is equal to mask
        subq %r8, %r9
        
        # If zero, continue to next round
        jne pmloopzero
        
        # Add current number to end sum if bit was 1 and continue normally
        addq %rdi, %rax
        
        pmloopzero:
        # Left bit shift to number rdi and mask r8

        addq %rdi, %rdi
        addq %r8, %r8
        
        # Check ending condition rdi has overflowed to negative side
        # Use r9 as temporal comparison register
        irmovq $0, %r9
        addq %r8, %r9
        
        jle pmend
        
        jmp pmloop
    
    
    pmend:
    ret