This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]
Other format:	[Raw text]

Re: [C++ PATCH] Rewrite convert_nontype_argument

From: Andrew Pinski <pinskia at physics dot uc dot edu>
To: Roger Sayle <roger at eyesopen dot com>
Cc: Mark Mitchell <mark at codesourcery dot com>, Zack Weinberg <zack at codesourcery dot com>, Jason Merrill <jason at redhat dot com>, Giovanni Bajo <giovannibajo at libero dot it>, <gcc-patches at gcc dot gnu dot org>
Date: Tue, 20 Jul 2004 19:15:49 -0700
Subject: Re: [C++ PATCH] Rewrite convert_nontype_argument
References: <Pine.LNX.4.44.0407201945520.5412-100000@www.eyesopen.com>

On Jul 20, 2004, at 6:55 PM, Roger Sayle wrote:

On Tue, 20 Jul 2004, Andrew Pinski wrote:

Well it also used for IMA reasons so it cannot be removed after fixing
the C++ front-end.


Could you explain why in a bit more detail?  I'm not happy that
we're now "stuck" with this performance degrading work-around for
reasons that weren't made explicit when the patch was proposed.

With IMA we have compatible structs which do not set the TYPE_MAIN_VARIANT to one other because of they are not variants to one another until after the front-end is finished with them.

What could be done is a pass which takes the compatible structs and make them one type thus being able to remove the language hook.

Thanks,
Andrew Pinski

Follow-Ups:
- Re: [C++ PATCH] Rewrite convert_nontype_argument
  - From: Roger Sayle

References:
- Re: [C++ PATCH] Rewrite convert_nontype_argument
  - From: Roger Sayle

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]