Assemble Error ... don't understand why ...

[m7xtuf]

Hi there,

   I have an assemble error but I have no idea why ... maybe all you experienced programmer can help me ...  With below's code, I got the error "cannot use 16-bit register with a 32-bit address" on the last line.  I am trying to copy the data from Array[] to AX register.

   Help !!!

----------------------------------------------------------------------

Code Select Expand



.data

Array       dw      25, 169, 144, 9, 49, 144
Count       dw      0

.code

start:

            lea     BX, Array
            mov     SI, Count
            add     SI, SI

            inc     Count

            mov     AX, [BX][SI]


Added code tags

[MichaelW]

The error message means just what it says. For 32-bit code addresses are 32-bit values, so indirect memory operands must use 32-bit registers. This code will assemble without error:

Code Select Expand


            lea     EBX, Array
            movzx   ESI, Count
            add     ESI, ESI
            inc     Count
            mov     AX, [EBX][ESI]


But in 32-bit code it would make more sense to define Count as a DWORD, and probably also Array. And since in 32-bit code indirect memory operands can include a scaling factor (of 1, 2, 4, or 8) you can avoid having to scale ESI separately. So the code could then be:

Code Select Expand


            lea     EBX, Array
            mov     ESI, Count
            inc     Count
            mov     EAX, [EBX+ESI*4]


[m7xtuf]

Thanx ... but is there a way to keep my code in 16-bit instead of 32-bit ...

[redskull]

Thanx ... but is there a way to keep my code in 16-bit instead of 32-bit ...

Use the 16-bit linker, and check out the 16-bit subfourm.  More specifically, use the small (or tiny) model directive.  Also, if you are more interesting in learning than producing code, things like emu8086 offer easy ways to get started.

-r