If you need to fill the alignment area with some other values, you can combine align with virtual to get the size of alignment needed and then create the alignment yourself, like:

Code:

    virtual
        align 16
        a = $ - $$
    end virtual
    db a dup 0    

macro align ment*,dat {
     local a
     if ~ dat eq
         virtual
                     align ment
                  a=$-$$
              end virtual
         db a dup dat
        else
                align ment
  end if
}    

But it fails to emulate the standard "align" directive properly. You should just make it fall back to default "align" directive when "dat" is empty.

If you need to fill
the alignment area with some other values, you can combine align
with virtual to get the size of alignment needed and then create the alignment yourself, like:

Code:

 virtual
 align 16
 a = $ - $$
 end virtual
    

> 2. There are some aligns inside the binary and they contain random garbage
> rather than ZERO's. Is it documented anywhere that the aligns are expected
> and how they are supposed to work ?

Yes. These may be filler bytes used for segment alignment. They aren't written, there's just a file positioning, so the values may be OS-dependent.

This is expected behavior.

http://www.bttr-software.de/forum/forum_entry.php?id=8979 wrote:

> 2. There are some aligns inside the binary and they contain random garbage
> rather than ZERO's. Is it documented anywhere that the aligns are expected
> and how they are supposed to work ?

Yes. These may be filler bytes used for segment alignment. They aren't written, there's just a file positioning, so the values may be OS-dependent.

This is expected behavior.

Ugh, that's NASTY - I once had pieces of source code show up in an executable because the compiler didn't zero out allocated memory...