subtracting negative numbers in assembly

Question

I know in order to turn a number into a negative in assembly, I need to turn it into two's complement by xor'ing it then adding one.

I've done this, but my output is still incorrect, can someone please explain to me where I have made a mistake?

The object is to subtract -val1 - val2 - val3

Here is the bulk of my code

.data val1 dword 10000h val2 dword 40000h val3 dword 20000h finalVal dword ?

.code main PROC

mov eax,val1            ; start with 10000h
xor eax,val1            ; xor eax to make it almost negative
add eax,1h              ; now add one to make it negative
sub eax,val2            ; subtract 40000h
sub eax,val3            ; subtract 20000h
mov finalVal,eax        ; store the result (30000h)
call    DumpRegs        ; display the registers

exit

https://ideone.com/hulsdU

The final result should be EAX=FFF90000 but currently my output is EAX=FFFA0001

note that I am using 32 bit numbers, and running this on an x86 processor.

Thank you for any help

Answer 1

This won't work:

mov eax,val1
xor eax,val1
  ; (a XOR a) is always 0. Using the line above you XOR
  ; val1 with val1 so EAX will always be 0 here!
add eax,1h
  ; (Ok. EAX will be 1 here...)
sub eax,val2
  ...

If you want to negate a number you can use the "NEG" instruction:

neg eax

If you want to do this using the ones complement you use the "NOT" instruction:

not eax
add eax, 1h

Or:

not eax
inc eax