[PATCH][Ada] PR ada/5911 Enable multilib build for Ada

Joel Sherrill joel.sherrill@oarcorp.com
Tue Sep 2 13:44:00 GMT 2008


Arnaud Charlet wrote:
>> The following patch has been tested on x86_64-linux trunk
>> with this patch to correct an unrelated dwarf2out.c Ada bootstrap issue:
>>
>> http://gcc.gnu.org/ml/gcc-patches/2008-08/msg02353.html
>>
>> Testing is currently not possible on sparc-solaris/linux because
>> bootstrap for C is broken:
>>
>> http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37279
>>     
>
> Since builds are broken in several ways, I'd suggest waiting that things
> stabilize, otherwise people won't be able to test your change independently and
> detect related troubles easily.
>
>   
FWIW the sparc-rtems ACATS results took a nosedive also.
There are ~120 broken tests which end with an alignment
exception.  These were not broken on 24 August.

BAD: http://gcc.gnu.org/ml/gcc-testresults/2008-09/msg00106.html

GOOD: http://gcc.gnu.org/ml/gcc-testresults/2008-08/msg02355.html

Sorry the gap between those two is a week.  I was moving
my testing to cfarm and didn't get a full run until this weekend.
>> I will need help for testing and completing the TODO part to support
>> more platforms, Eric volunteered for sparc and Joel for RTEMS.
>>     
>
> Well, the most problematic platforms will likely be powerpc-*, since there
> are lots of multilib variants for ppc, so testing should really be done on
> a powerpc multilib target (RTEMS maybe ?) before committing this change.
>
>   
I built a powerpc, sparc, and i386 for an earlier version of this
patch but am happy to test it again.

m68k was the problem case and I think Paolo addressed something
there.

Laurent/Arnaud.. do you want me to apply this patch and build
the RTEMS targets now or wait until sparc is OK?

--joel



More information about the Gcc-patches mailing list