Re: PATCH: DragonFly support, attempt #3

[Nick Clifton <nickc at redhat dot com>]

Hi John,

> Sorry about the missing te-dragonfly.h file. My "git fu" isn't as good
> as it should be. I have regenerated the entire patch set and attached
> them here along with the same commit messages as before for convenience.
> I think the entire attached patchset is "mainstream" binutils.

Thanks - I have now committed these patches to the mainline.

> By the way, can you take a closer look at ldlex.l change? I've gotten
> feedback that the proper fix for the duplicate YY_NO_UNPUT macro
> definition is just to remove the line "#define YY_NO_UNPUT" since the
> first line "%option nounput" always defines YY_NO_UNPUT. If you agree
> with that feedback, you'd want to change the ldlex.l patch although
> leaving it as it is doesn't do any harm.

It probably is sufficient to just remove the definition of YY_NO_UNPUT, 
but I prefer to be paranoid and assume that LD might be built with a 
broken implementation of lex that does not honour the "%option nounput", 
and so it is safer to include the definition.

Cheers
  Nick