Question

I am new to Assembly, but not new to programming and how it works in general. Because I know quite a few languages already, I decided to make my first ASM program calculate the SHA-1 hash of input files. I already have many of the separate parts worked out, but there are two parts I am having trouble with; one I will ask for help with here.

Basically, I am confused over the concepts of Big and Little endian numbers and how to represent them in ASM. To quote from the SHA-1 Psuedocode:

append length of message (before pre-processing), in bits, 
as 64-bit big-endian integer

The only things that come to mind is to, for example represent the number 1 as:

[63 0's]1

Or

1[63 0's]

Or

[31 0's] 1 [32 0's]

Basically my thought so far is basically:

mov    eax,  message ;Move message to be hashed into eax
shl    eax,  64      ;Shift message over enough bits to store 64-bits
or     eax,  msglen  ;Adds message length to the very end of message

This is most likely wrong, as this all uses 32-bit registers therefore shifting left 64-bits would create a huge overflow but that is just another problem I have yet to work out.

I think I’ve gone way off track, but basically my question is is the above code a correct way to append a big-endian integer? Or do I have it backwards, or am I just completely off on everything, etc. Not sure if this makes sense, too tired.

Answer 1

The binary representation of the number 1 as a big-endian 64-bit integer is

<- Start of message                                   End of message ->

00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000001

^ big (most significant) end