This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH] include _mingw.h in stddef.h
- From: Kai Tietz <ktietz70 at googlemail dot com>
- To: Ross Ridge <rridge at csclub dot uwaterloo dot ca>
- Cc: gcc-patches at gcc dot gnu dot org
- Date: Sat, 19 Sep 2009 15:59:50 +0200
- Subject: Re: [PATCH] include _mingw.h in stddef.h
- References: <20090918160240.1E34373F75@caffeine.csclub.uwaterloo.ca>
2009/9/18 Ross Ridge <rridge@csclub.uwaterloo.ca>:
> Ross Ridge writes:
>>If support is added to GCC for these Microsoft types, I think it should
>>be as real types. ?That would make the types fully compatibile with MSC.
>
> Kai Tietz writes:
>>I agree that it would be best to have here builtin macros for those
>>type names. The builtin types aren't usable here, as they would be
>>typedefs and so not combinable with signed/unsigned.
>
> No, I'm saying that if anything is done, they should be made reserved
> words and synonymous with char, short, int and long long as appropriate,
> not typedefs or macros. ?That's what's necessary for a fully compatible
> implementation. ?Otherwise the macros are just a hack, and having the
> user do -D__int8=char, if that's what they want, is just as good of hack
> as anything else.
Well, to add here for C and C++ frontends the support of types __int8,
__int16, __int32, __int64, and for 64-bit version the type __int128
would be fine. But I have doubts that a target specific override of
RID_... and checks would be acceptable. But if you want my vote you
have to prepare a patch for this.
Kai
--
| (\_/) This is Bunny. Copy and paste
| (='.'=) Bunny into your signature to help
| (")_(") him gain world domination