On Sat, Jun 26, 2004 at 04:03:33PM -0400, Richard Kenner wrote: > GCC used to do this in part. We'd use a separate register allocator, > stupid.c, for the -O0 case. This was eliminated in GCC 3 and we took a > substantial hit on -O0 compilation speed as a result. No, we didn't. Go back and look at the measurements done at the time. r~