arm-aout vs arm-elf

Quality Quorum qqi@world.std.com
Mon Oct 15 08:46:00 GMT 2001


On Mon, 15 Oct 2001, Richard Earnshaw wrote:

> >
> >
> > On Mon, 15 Oct 2001, Philip Blundell wrote:
> >
> > > >If I do  xxxxx/confiugre --target=arm-elf ...,  I will get apcs-32,
> > > >correct settings of cpp defaults and properly configured multilib.
> > > >
> > > >If I do xxxx/configure --target=arm-aout ..., I will get apcs-26,
> > > >hardlfoat default for cpp and no multilib, BTW it will happen regardless
> > > >of `with-cpu' settings.
> > > >
> > > >I do not think that this is desired outcome.
> > >
> > > Why not?  Those two targets are not supposed to be in any way identical.
> > >
> > > The arm-aout target has existed since the beginning of time, and its default
> > > options were appropriate to the time it was created.  On the other hand,
> > > arm-elf is much newer and its default options reflect that.
> >
> > I do not think that this is acceptable, difference between aout and
> > elf should be in output file format only - at least it has to be a big
> > warning about these problems printed by configure.
>
> It would be far less acceptable to a user of, say, gcc-2.8 who upgraded to
> gcc-2.95 if he found that the same configuration started targetting a
> different processor.

I still do think that there are better ways to address the issue, at
least there should be a statement in arm/README.


> >
> > BTW, is it possible to configure arm-aout for apcs-32/soft-float by
> > providing --with-xxxx parameters to configure ?
>
> No, the existing configure options for changing floating point selection
> during configure are not used by the ARM port; nor do I expect them to be,
> since in the near future this will become a four-way choice (minimum) and
> the existing option is at most soft/hard.
>
> R.
>
>

Thanks,

Aleksey



------
Want more information?  See the CrossGCC FAQ, http://www.objsw.com/CrossGCC/
Want to unsubscribe? Send a note to crossgcc-unsubscribe@sourceware.cygnus.com



More information about the crossgcc mailing list