Re: L4_Word_t vs. void*
Espen Skoglund <[email protected]>
| Newsgroups | gmane.comp.micro-kernel.l4.l4ka.general |
|---|---|
| Message-ID | <[email protected]> |
[Joshua Haberman] > I notice that the L4 headers use L4_Word_t (and never void*) to > refer to addresses. Is there a reason why L4_Word_t is a better > choice? Should I avoid using void* in my own L4 programs? > I imagine there are 32/64 bit issues here (and perhaps even > 32-on-64) that I don't fully understand. Can someone enlighten me? Difficult to give one single good reason for this. Much of it has historic reasons. It's also not so much a 32/64 bit issue as it is a question about how we use the addresses. For the most part, the addresses are not really pointers to objects (as in C++ objects). Describing them as void * can as such be misleading. It is also easier to use Word for calculating offsets, etc. Just because we don't use void * much in the API specification does not mean that you should necessarily try to avoid it in your own programs. eSk