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 to emit_tinfo_decl

From: Jason Merrill <jason at redhat dot com>
To: Richard Henderson <rth at redhat dot com>
Cc: Nathan Sidwell <nathan at codesourcery dot com>,John David Anglin <dave at hiauly1 dot hia dot nrc dot ca>, gcc-patches at gcc dot gnu dot org
Date: Fri, 17 May 2002 21:17:31 +0100
Subject: Re: C++ PATCH to emit_tinfo_decl
References: <200205132140.g4DLeJVr000285@hiauly1.hia.nrc.ca><wvlwuu4em5g.fsf@prospero.cambridge.redhat.com><20020516105224.C25830@redhat.com><3CE3FC76.EFD386AF@codesourcery.com><20020516161307.A25975@redhat.com>

>>>>> "Richard" == Richard Henderson <rth@redhat.com> writes:

> On Thu, May 16, 2002 at 07:37:42PM +0100, Nathan Sidwell wrote:
>> the sizeof the typeid struct for T depends on whether T has
>> no bases, one single non-virtual base, or is more complicated.

> In which case, it ought to be incomplete itself, correct?
> IIRC, incomplete types have size zero, and small data opts
> are supposed to ignore those.  Fibbing with size 8 is surely
> the worst thing that can be done.

Agreed.

> That's not really the best of ideas.  Previously, we put small comdat
> variables in .gnu.linkonce.s.*, which gets merged with .sdata.

Ah, I hadn't realized that we were doing this.  In that case, I'll go ahead
and fix the lie.

Jason

References:
- Re: C++ PATCH to emit_tinfo_decl
  - From: John David Anglin
- Re: C++ PATCH to emit_tinfo_decl
  - From: Jason Merrill
- Re: C++ PATCH to emit_tinfo_decl
  - From: Richard Henderson
- Re: C++ PATCH to emit_tinfo_decl
  - From: Nathan Sidwell
- Re: C++ PATCH to emit_tinfo_decl
  - From: Richard Henderson

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