Warning for different pointer signedness

Eric Botcazou ebotcazou@libertysurf.fr
Mon Dec 20 22:31:00 GMT 2004


> Well, there is an infinite amount of "real-life correct" codes that fail
> type checking; do you propose a clear guideline that the compiler
> should follow to accept them?

Do not periodically let the kernel folks think you introduce random changes to 
make their life harder.

-- 
Eric Botcazou



More information about the Gcc mailing list