On Sun, Oct 13, 2002 at 07:48:22PM +0200, Felix K�hling wrote: > Appart from that, I think that radeonUpdateViewportOffset is buggy. The > aliasing of int and float is inconsistent. The comparison is done > without aliasing, in the subsequent assignment the floats are aliased to > ints. I attached a patch that is a workaround for the gcc-3.2 problem at > the same time. I guess it's the inconsistent aliasing that confused gcc.
I believe that ANSI C is quite clear on this. IIRC, if float and int are mixed in a arithmetic operation, the int is always promoted to float. I can ask the compiler folks at the other end of the floor abou this later today. It's nice having a bunch of Linux engineers all gathered in one place. :) -- Smile! http://antwrp.gsfc.nasa.gov/apod/ap990315.html ------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Welcome to geek heaven. http://thinkgeek.com/sf _______________________________________________ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
