cvs commit: src/sys/sys pbioio.h src/sys/i386/isa pbio.c src/sys/conf files.i386 src/sys/i386/conf NOTES

[Marcel Moolenaar]

On Thu, Oct 07, 2004 at 01:58:13PM -0600, M. Warner Losh wrote:
> : > : > 
> : > : >   Modified files:
> : > : >     sys/conf             files.i386 
> : > : >     sys/i386/conf        NOTES 
> : > : >   Added files:
> : > : >     sys/sys              pbioio.h 
> : > : >     sys/i386/isa         pbio.c 
> : > : >   Log:
> : > : >   Port pbio to HEAD.
> : > : 
> : > : I appreciate your speed, but don't you think that pbioio.h is pretty
> : > : MD given that the driver only exists on i386. Wouldn't <machine/pbioio.h>
> : > : be a better place?
> : > 
> : > I think that pbioio.h is MI.  What makes you think that it is pretty
> : > MD?
> : 
> : Because pbio(4) only exists on i386. Hence, all other platforms now
> : have <sys/pbioio.h> that serves no purpose and can only cause
> : autoconfiguration (ala autoconf) getting confused. Since pbioio.h
> : refers to hardware, it nature is MD, even though the definitions
> : (types and interfaces) may appear MI. So, <machine/bpioio.h> seems 
> : like a good place.
> 
> But we've had lots of problems in the past where MD drivers became MI
> drivers with 10 identical copies of machine/fooio.h.  The driver
> likely copy become MI relatively easily.

Speculation. Isn't it better to cross that bridge when you come to it?
A repo-copy when the work has actually been done seems easy enough.
Don't get me wrong. For all I care this driver is the next best thing
since the invention of a round wheel, but the fact is that all people
involved seem to worry about i386 only and even made it an isa only
driver (based on it's location not on source code). That's a pretty
limited scope to begin with and it doesn't exactly give me the warm
fuzzies that the hardware is anywhere near MI nor that the driver grows
MI. Hence my concern...

Anyway: I'm not asking for a backout and you're not of the opinion
that <machine/bpioio.h> is better, so we can close this case anytime
we like to. Feel free...

-- 
 Marcel Moolenaar	  USPA: A-39004		 marcel at xcllnt.net