Am Fr, 2004-08-06 um 18.38 schrieb Bill Studenmund:
> On Fri, Aug 06, 2004 at 04:13:58PM +0100, phlox wrote:
> > Greetings.
> > I'm developing a character device driver for NetBSD that will allow the
> > microcode update in the IA-32 architecture, but I have some design questions
> > that I would like to clarify:
> >     * What devices should I use for SMP machines ? Should I create only
> >       /dev/microcodei (and then use ioctl() to select the CPU) 
> >       or /dev/microcode[0-3] (for a 4-way SMP system) ? 
> >       (I think the last follows UNIX semantics better).
> 
> I'd say just one device will be fine. The important thing is that you 
> won't have concurrent, multi-user access to the different devices; they 
> aren't tty's that folks are logged-onto. Also, once you've done the 
> update, you don't need the devices anymore.
> 
> Further, there are systems out there with much more than 4 CPUs on them. 
> Higher-end iron can have 16 or 64 heads, and really-high end stuff can 
> have 512 or 1024 heads. It would be really really messy to have to have a 
> node for each one.
> 
> >     * I've been seeing the lkm code and I found that the loaded lkm list is an
> >       array (lkmods[MAXLKMS]). Why isn't it a linked list, since this would
> >       allow better cleanness ?
> 
> Laziness probably.
> 
> Take care,
> 
> Bill

I like the second, because of the "UNIX semantics". What about a
directory ?