Greywolf wrote:
   > # On Sun, 22 Dec 2002, Jaromir Dolecek wrote:
   > # 
   > # I don't see general need need for emulation 'init' function. Exactly
   > # one emulation can take advantage of it, and even the one runs
   > # perfectly fine the current way.
   > 
   > At the moment this may be true, sure.  But is this forward-thinking
   > at all?
   
   less bloat in generic code == good
   
   Mach emulation doesn't need to adjust any OS parameters or structures,
   so it doesn't need to run it's initialization on boot. It's perfectly
   fine to execute the stuff when first Mach process is executed, IMHO.
   
   So, we have one emulation which would take cosmetic advantage of such
   thing (e_init called on boot), and no emulation which would need
   it to work properly. I don't think it would be appropriate to add
   the framework in this situation.

i agree.


additionally, if mach_init() or whatever it's called is called
at boot time, it will allocate it's resources _always_ even for
people who never use it.  think of all the netbsd/i386 users that
will affect when COMPAT_MACH is enabled in GENERIC? :-)

of course, this wouldn't be an issue with really good emul LKM's :-)


.mrg.