Tony van der Hoff tony at
Fri Dec 21 11:08:23 PST 2001

On 21 Dec 2001, in message <70086bec4a.peter at>,
Peter Naulls <peter at> wrote:

> In message <c04669ec4a.Tony at>
>           Tony van der Hoff <tony at> wrote:
> > On 21 Dec 2001, in message <m3n10cd6k5.fsf at>,
> > Nick Burrett <nick at> wrote:
> > 
> > > It was a last minute inclusion on my behalf, using the
> > > standard Unix zip package.  Totally overlooked that issue.
> > > 
> > > Nick.
> > > 
> > 
> > The library files are all typed Text - 0xFFF, not Data.
> This has been the case for sometime (not sure how long exactly).
> GCC's output is typed text, unless it's an absolute, etc.
> Sure, it could be changed back, but I can't see any strong argument
> to do so.
Not in arm_riscos.2_95_2, as in the previiew 3 release, but it's no
real problem; it just looks a bit odd IMO.

> > Is there/should there be a path pointing at the library directory, for
> > passing to a linker, something like GCC:o.unixlib?
> Maybe.  The only instance I know of where you'd really want to directly
> reference the library is when compiling Norcroft against unixlib.
> Perhaps then you should be using the stand along Unixlib?
Well, it's not just UnixLib I'm after. That just happens to be where
all libraries now seem to live. Anyway, is there a (newly-released)
stand-alone UnixLib?

> THB, I don't know - what were you hoping to do?
I must have missed something, I thought everybody did it like this; perhaps
there's a better way; maybe using DRLink instead of Acorn's link:

My makefiles contain a list of libs that my project needs to link against. I
pass that list to Link together with the file list.

If I need, say libiostream, it's a lot simpler to specify
GCC:o.libiostream, where <GCC$Path> resolves to
ADFS::Tonys.$.tools.!gcc.bin.arm-riscos-aof.2_95_4.apcs26.unixlib., which is
set by !gcc.!run, than it is to type that lot in, especially if I need
multiple libs from that place.

Evidently I added such a path to my previous installation; that's no
hardship, I just wondered whether it was a deliberate omission, or an
oversight :-)

Tony van der Hoff         | MailTo:tony at
                          | MailTo:avanderhoff at
Buckinghamshire, England  |

More information about the gcc mailing list