This is the mail archive of the binutils@sources.redhat.com mailing list for the binutils project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: Adding support for AS argument "-force-long-branchs" for x86arch


Hi Ian,

> Wait, this patch needs some more thought.  First of all, it's not
> the case that -force-long-branches was treated as -f.  It was
> actually treated as
>     -f -o rce-long-branches

True.

> That is, short options may be concatenated following a single dash.
> So this patch is actually a behaviour change, in that it will no
> longer work to do,  e.g.
>
>     -fW
>
> I don't think we should casually change the behaviour of short options
> in this way.  Being able to concatenate short options is standard Unix
> behaviour mandated by POSIX.2.
>
> Furthermore, if we are going to change it for -f, then we should
> logically change it for all of the short options, and indeed we should
> eliminate std_shortopts and md_shortopts entirely.  I don't think that
> would be a good idea at all, as it might break existing scripts.

I agree - I will restore the original behaviour.

I do feel that it can be very confusing for the user when they are
confronted by this situation.  But since the POSIX specs call for it,
then there is nothing that we can do.  *sigh*

Cheers
        Nick

gas/ChangeLog        
2003-10-05  Nick Clifton  <nickc@redhat.com>

	* as.c (parse_args): Revert patch to change handling of -f
	option.  Its behaviour is mandated by the POSIX 2 spec.

                


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]