This is the mail archive of the
mailing list for the GCC project.
Re: Patch to implement IA64 __fpreg
On Tue, 28 Jun 2005, Richard Henderson wrote:
> On Tue, Jun 28, 2005 at 05:09:28PM +0000, Joseph S. Myers wrote:
> > Even with the full version of __fpreg there are still operations which are
> > not permitted and so need to be diagnosed, with error not sorry, such as
> > division and conversions to or from __float128.
> Sure, but again the decision between error and sorry would be
> under control of the target hook.
The target hooks can call sorry then return NULL. I'm just not convinced
it's worthwhile to implement the distinction between features not in the
current GNU __fpreg but in HP __fpreg and features not in HP __fpreg
The hook design follows TARGET_INVALID_ARG_FOR_UNPROTOTYPED_FN which
established the principle of having such hooks for back ends to control
what uses front ends allow of target-specific types.
Joseph S. Myers http://www.srcf.ucam.org/~jsm28/gcc/
email@example.com (personal mail)
firstname.lastname@example.org (CodeSourcery mail)
email@example.com (Bugzilla assignments and CCs)