> I'm not only loading the structure I'm using it all over the code. I > don't have clean cut. You can blame me for that but that's another > discussion. I actually thought that was intentional to reduce the number of local variables. > We are going around the same issue all gain. I already agreed that > your approach will be correct if we had more code of your example type > then setting bits. But we are mostly setting bits. Going from the definitions in the header files that's not true. I'm too lazy to do actual stats though. Yes, there are many bits, but very often you have two/three-bit values intermingled with flags. > This is really arguing about taste and feel. Sort of. Maybe all I'm saying is that you're (artificially) limiting the number of people who can work on your drivers because all non-Intel drivers use the opposing style of defining things in native byte-order while you're hiding the endian conversion in the defines rather than making it explicit in the code which is much easier to understand. johannes
Attachment:
signature.asc
Description: This is a digitally signed message part
