It doesn't behave like 'normal' assemblers. For example:
ax, bx, cx etc are synonyms for eax, ebx, ecx etc based on the 16 or 32 bit mode
segment and size prefixes must be specified manually via cseg, dseg, o16, o32 etc rather than inferred from the size of the registers
likewise, instructions which change behaviour according to mode are generated incorrectly if you use the wrong name for the mode (like cwd and friends)
it's very easy to generate invalid instructions --- invalid configurations aren't checked for
I suspect the 32-bitness was a added as a quick hack on top of the 16-bit assembler. But it's painful to use and really needs to be rewritten.
It doesn't behave like 'normal' assemblers. For example:
I suspect the 32-bitness was a added as a quick hack on top of the 16-bit assembler. But it's painful to use and really needs to be rewritten.