[lno] [RFC] if-conversion and auto vectorizer

Richard Henderson rth@redhat.com
Tue Mar 16 18:32:00 GMT 2004


On Tue, Mar 16, 2004 at 07:27:02AM -0500, Robert Dewar wrote:
> I don't understand the argument for C89. Why prevent a useful
> optimization in order to enable incorrect code that just happens 
> to work to keep working.

For the same reasons that we preserve a set of rules that are
beyond C89 such that it's possible to write kernels.


r~



More information about the Gcc mailing list