Stefan Roese wrote:
You have a point with your variable usage "restriction", and Wolfgang has a point with the "readability" of your patch as it doesn't really tell what is fixed without read the current source. Could be that your implementation is the "better" one, but the patch Wolfgang applied was just less intrusive.
Well, I can't argue with that.
No, we shouldn't "fork" the code here. Let's focus on the other open issues.
Since you bring it up, I submitted a patch for the 5xxx platform many months ago. How do I get that applied? The patch is here:
http://sourceforge.net/mailarchive/message.php?msg_id=117131648887-git-send-...
You mention other bugs introduced by the libfdt code above. ;-)
Jerry is already aware of them and working on fixes.