Re: Various libgloss problems after recent changes

On Fri, Feb 10, 2023 at 6:21 AM Richard Earnshaw <
[email protected]> wrote:

>
>
> On 09/02/2023 22:54, Jeff Johnston wrote:
> > Hi Jeff,
> >
> > Want to try out the attached patch?
> >
> > -- Jeff J.
> >
> >
> > On Sun, Feb 5, 2023 at 7:48 PM Jeff Law <[email protected]> wrote:
> >
> >> The fr30-elf and xstormy16-elf ports have been failing to build for
> >> about a month with this error:
> >>
> >>    > fr30-elf-as  -I. -I../../../../..//newlib-cygwin/libgloss/fr30/..
> >> -I./.. -idirafter
> >> ../../../../..//newlib-cygwin/libgloss/fr30/../../include  -o crt0.o
> >> ../../../../..//newlib-cygwin/libgloss/fr30/crt0.s
> >>> fr30-elf-as: unrecognized option '-idirafter'
> >>
> >>
> >> For a full log:
> >>
> >> http://law-sandy.freeddns.org:8080/job/fr30-elf/1984/console
> >>
> >> Hopefully this is easy to fix so we can get those ports under regular
> >> testing again.
> >>
> >> Jeff
> >>
> >>
>
> I'm slightly confused.  Why would some ports have -idirafter and some
> not, unless they're not using GCC (or a very old one).
>
> So wouldn't a autoconf test for -idirafter be a better approach rather
> than hard-coding it.
>

Yes, I should have picked a better flag name - I really just wanted to
exclude the 2 platforms that don't need the -idirafter statement in the
first place.
A better way would be to simply add a flag to indicate specifically that a
config wants to include files in top level include.


> Note that if it's due to the option not being supported in old versions
> of gcc, then that's a bigger problem because the reason the option was
> added in the first place (IIRC) was to support moving arm-acle-compat.h
> out of the newlib subdir so it could be used with libgloss.  And
> arm-acle-compat.h is there to support older versions of GCC.
>
> And that brings up another question.  Just how old a version of GCC do
> we want to support these days?  ACLE support was added to GCC 10 years
> ago, in GCC-4.8. If we can drop support for compilers older than that,
> we can drop using the compatibility header and get rid of the need for
> -idirafter entirely.
>
> R.
>
>