[mainline] PATCH to diagnostic.[hc]

Gabriel Dos Reis gdr@integrable-solutions.net
Wed Apr 30 16:09:00 GMT 2003


"Joseph S. Myers" <jsm28@cam.ac.uk> writes:

| On Wed, 30 Apr 2003, Gabriel Dos Reis wrote:
| 
| > | A C90 compiler (e.g. older GCC), not necessarily a C90 library (except to
| > | the extent to which libiberty emulates one - and though there have been
| > | discussions of putting a *printf reimplementation in libiberty, it hasn't
| > | happened yet).
| > 
| > I'm not sure we do really want to make that pedantic distinction.
| > Was distinction  the SC voted on?
| > I do not believe it would bring any value over the previous K+R
| > requirement.  
| 
| It's what I understood from the previous discussions - the actual
| announcement said nothing about the details [1].  It fits in with the
| principle that libiberty (on the host and build systems) and fixincludes
| (on the target) make old libraries largely invisible whereas old compilers
| have an impact on all the code, and systems with old compilers can be
| bootstrapped via an older GCC.

I'll make the change and leave the library issue for latter, although
I do think that we should also remove the library restrictions (by
putting appropriate bits in libiberty if necessary).

-- Gaby



More information about the Gcc-patches mailing list