Re: Gforth and gcc "progress"
- From: Bernd Paysan <bernd.paysan@xxxxxx>
- Date: Tue, 26 Jun 2007 16:01:35 +0200
Andrew Haley wrote:
On Alpha it was not.
OK, but that has nothing at all to do with gcc: the sizes of types is
determined by the ABI. gcc implements the ABI of whatever platform it
runs on.
long long wasn't part of the ABI. long long back then was a GCC extension,
not supported by any other compiler. C99 put long long into the standard,
and therefore now, it's part of the ABI, and you have to take C99's weasel
wording (courtesy to other vendors which have rendered the original C
typing system completely useless by now, like having no C90 integer type
which can hold a pointer - the IL32P64 model).
I'm ok when int128_t is implemented in a useful way; I don't care about how
the pointer type is called, as long as I can check it with autoconf.
--
Bernd Paysan
"If you want it done right, you have to do it yourself"
http://www.jwdt.com/~paysan/
.
- References:
- Gforth and gcc "progress"
- From: Anton Ertl
- Re: Gforth and gcc "progress"
- From: slava@xxxxxxxxx
- Re: Gforth and gcc "progress"
- From: Anton Ertl
- Re: Gforth and gcc "progress"
- From: Andrew Haley
- Re: Gforth and gcc "progress"
- From: Anton Ertl
- Re: Gforth and gcc "progress"
- From: Andrew Haley
- Gforth and gcc "progress"
- Prev by Date: Re: MPE 64-bit bench
- Next by Date: Re: Gforth and gcc "progress"
- Previous by thread: Re: Gforth and gcc "progress"
- Next by thread: Re: Gforth and gcc "progress"
- Index(es):
Relevant Pages
|
