Re: [Nasm-devel] Bug report 2010180 (PUBDEF)

[H. P. A. <hp...@zy...>]

H. Peter Anvin wrote:
> 
> Ah, yes, now I remember: if we're going to do this, we need to 
> distinguish not just __ABSOLUTE__, but any nobits section.  Similarly, 
> distinguishing data sections (which want db 0) and code sections (which 
> want code) would be nice, too.
> 

I think for now we'll stick with "align" vs "alignb" (alignb in nobits 
contexts), unless someone has a really bright idea how to deal with 
that.  It seems that it would pretty much require the backends to feed 
back to the preprocessor the attributes of a segment... messy.

	-hpa