[cfe-dev] patch for generalizing stdint for n-bit bytes

[Chris Lattner]

On Nov 11, 2009, at 5:01 AM, Ken Dyck wrote:
>
> Stage #2a attached. This adds definitions for types of 8-bit multiples
> from 8 to 64 to stdint.h and rationalizes the selection of types for  
> the
> exact-width definitions in InitPreprocessor.cpp.
>
> Only minimal changes have been made to the test cases:
> - A 'signed' specifier is added to the type definitions where it was
> unspecified before
> - int32_t on PIC16 and MSP430 is now defined as 'long int' instead of
> 'long long' and int64_t on SystemZ is now 'long int' instead of 'long
> long int'. These are all changes in types of equal width.
>
> In stage #2b, I plan to parameterize the suffixes used by the
> constant-generating macros in stdint.h, and use those macros to define
> the limits of the integer types.

Looks great to me, committed as r86977!

-Chris