[cfe-dev] Predefined stdint macros (i.e. __UINT8_TYPE__, etc.)
Paul A. Bristow
pbristow at hetp.u-net.com
Fri Jun 13 01:30:58 PDT 2014
From: cfe-dev-bounces at cs.uiuc.edu [mailto:cfe-dev-bounces at cs.uiuc.edu] On Behalf Of Chandler Carruth
Sent: 11 June 2014 20:15
To: Ben Smith
Cc: clang-dev Developers
Subject: Re: [cfe-dev] Predefined stdint macros (i.e. __UINT8_TYPE__, etc.)
On Wed, Jun 11, 2014 at 8:10 PM, Ben Smith <binji at chromium.org> wrote:
Hello cfe-dev,
Clang defines a subset of the predefined stdint macros that GCC does. In particular, GCC defines the following macros many of which are not defined in Clang:
__U?INT{_,_FAST,_LEAST}{8,16,32,64}_{MAX,TYPE}__
__U?INT{PTR,MAX}_{MAX,TYPE}__
Some of these macros are used in newlib's stdint.h, if they exist. It seems that it hasn't been thoroughly tested with Clang, however, as some macros are assumed to exist.
Is it worth adding these to Clang?
Yes, we should try to be compatible here, and generally it seems reasonable to define the fully expanded set of these.
Yes please – these are very helpful – and these are very likely to become C++Standard,
and also the floating-point specified-width versions proposed
www.open-std.org/jtc1/sc22/wg21/docs/papers/2013/ <http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2013/n3626.pdf> n3626.pdf
(and the corresponding C macros in the identical Wg14 paper n1703)
A fully working prototype using GCC developed by Christopher Kormanyos shows that these can be very useful, especially the floating-point types to facilitate 128-bit FP with C++, but also with single-board ‘bare–metal’ embedded systems.
Christopher Kormanyos will present an updated version showing how this is useful in practice at Rapperswil.
Paul
---
Paul A. Bristow
Prizet Farmhouse
Kendal UK LA8 8AB
+44 01539 561830
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/cfe-dev/attachments/20140613/9a41e6d3/attachment.html>
More information about the cfe-dev
mailing list