This is the mail archive of the
mailing list for the GCC project.
Re: [PATCH] Fix bug and pessimization for packed records in Ada
> Why don't you add a gcc_assert somewhere to abort if you find this
> semi-promoted state? Given that it's so invalid that SPARC generates
> wrong-code with it, I think it is better to have the compiler ICE if ever
> build it somehow.
Maybe. But the code in stor-layout.c is so intricate that I'd be very
> > The patch also gets rid of the set_lang_adjust_rli hook that was
> > added for Ada 3 years ago and silenced (but not removed) by Jason in
> I guess you need to poison it?
This was an old-style hook, with no #define.