On Tue, Dec 15, 2009 at 04:10:22PM -0800, Jon Aimone wrote: > Bill Sommerfeld perorated, on or about 12/15/09 15:25: > >It's IMHO very important to distinguish between things which are private > >by design and things which are private because we haven't gotten around > >to defining a public interface. > > > >Folks who are not "on the inside" often treat a "no, you can't do it > >that way right now" answer as a "no, you can't do that ever".
+1 > I could not agree more. Hence my original request and the subsequent > RFE... I'd much rather pull for an official interface than hack around > or duplicate previous art. > > My only concern now is the fact the RFE was quickly moved to bugster as > a P4 RFE... and I'm wondering what kind of miracle it will take to get > activity on a P4... If you can't get someone else to do it, you might do it yourself. The simplest fix would be to file a self-review or fasttrack ARC case documenting the OPROMDEV2PROMNAME and OPROMPROM2DEVNAME ioctls of the openprom(7D) driver. That's not the cleanest fix, but hey, it's just docs, can't beat that! The better, less simple way would be to add di_dev2prom() and di_prom2dev() to libdevinfo as simple wrappers of those internal functions. A simple ARC case and simple code fix. You could also do both. Nico --
