mcdonald@uxe.cso.uiuc.edu (01/30/89)
>The bottom line when comparing code size is the size of the code >segment in an individual OBJ file. Sorry, I would think that the bottom line would be the size of the code when it got put in memory, ready to execute. If people, particularly compiler reviewers for magazines, keep worrying about .obj or .exe sizes, some compiler vendor is going to compress their .objs and/or .exe's and expand them at run time: The actual code segment in the .exe would consist of the decoder, which would decode the real code and data segments. That would really zap the benchmarkers. On further consideration, considering how much faster Pkarc is than copy, maybe it might be a good idea. On even further consideration, doesn't one of zoo's friends do this?