Hi Richard,
> On Tue, Apr 22, 2025 at 12:31 PM Sam James <s...@gentoo.org> wrote:
>>
>> Jakub Jelinek <ja...@redhat.com> writes:
>>
>> > On Mon, Apr 21, 2025 at 10:52:22AM +0200, Richard Biener wrote:
>> >> On Fri, Apr 18, 2025 at 8:10 PM Jakub Jelinek <ja...@redhat.com> wrote:
>> >> >
>> >> > On Fri, Apr 18, 2025 at 06:04:29PM +0200, Rainer Orth wrote:
>> >> > > That's one option, but maybe it's better the other way round: instead
>> >> > > of
>> >> > > excluding known-bad targets, restrict cobol to known-good ones
>> >> > > (i.e. x86_64-*-linux* and aarch64-*-linux*) instead.
>> >> > >
>> >> > > I've been using the following for this (should be retested for
>> >> > > safety).
>> >> >
>> >> > I admit I don't really know what works and what doesn't out of the
>> >> > box now,
>> >> > but your patch looks reasonable to me for 15 branch.
>> >> >
>> >> > Richard, Robert and/or James, do you agree?
>> >>
>> >> I agree to restrict =all to enable cobol only for known-good platform
>> >> triples.
>> >> But IIRC that's what libgcobol configure.tgt does - IIRC intent was to
>> >> allow
>> >
>> > But libgcobol configure.tgt does it only for the target case.
>> > Much more important right now is the host whitelist case.
>> > The code in the FE is still not sufficiently portable and so even
>> > i686-solaris -> x86_64-linux --enable-languages=all cross just won't
>> > build out
>> > of the box (and even if it would, it wouldn't work properly).
>> > Or x86_64-freebsd -> x86_64-linux might not either.
>> >
>> > So, I think we want Rainer's patch here.
>> >
>>
>> I thought about it further after discussion on IRC and I agree. Rainer's
>> looks good.
>
> OK, let's do that for GCC 15 then, we can add to the whilelist when we figure
> additional known-good cases for 15.2.
what about trunk then? Right now, cobol still doesn't build there on
Solaris/amd64 because 3 patches are missing:
cobol: Don't require GLOB_BRACE etc. [PR119217]
https://gcc.gnu.org/pipermail/gcc-patches/2025-April/680675.html
Still unreviewed.
cobol: Initialize regmatch_t portably [PR119217]
https://gcc.gnu.org/pipermail/gcc-patches/2025-April/680676.html
Unclear how exactly to handle this.
cobol: Allow for undefined NAME_MAX [PR119217]
https://gcc.gnu.org/pipermail/gcc-patches/2025-April/680682.html
Should go for a variant simply replacing NAME_MAX by its Linux
<limits.h> value (255) until the whole struct cbl_function_t.name thing
can be resolved. Already tested on Linux/x86_64 and Solaris/amd64.
libgcobol does build on Solaris (at least on trunk) since
commit a619a128c992b2121a862b8470960ae751d25db6
Author: Rainer Orth <r...@cebitec.uni-bielefeld.de>
Date: Mon Apr 21 15:59:14 2025 +0200
libgcobol: Check for struct tm tm_zone
Rainer
--
-----------------------------------------------------------------------------
Rainer Orth, Center for Biotechnology, Bielefeld University
