3.9-rc1 regression in arm dtb build

[Pavel Machek]

On Sun 2013-03-10 22:05:46, Olof Johansson wrote:
> On Thu, Mar 07, 2013 at 01:50:54PM -0700, Stephen Warren wrote:
> > On 03/07/2013 07:45 AM, Pavel Machek wrote:
> > > Hi!
> > >>> commit 499cd8298628eeabf0eb5eb6525d4faa0eec80d8
> > >>> Author: Grant Likely <grant.likely at secretlab.ca>
> > >>> Date:   Tue Nov 27 16:29:11 2012 -0700
> > > ...
> > >>> 1) it is inconsistent with 3.8, making switching between 3.9-rc1 and
> > >>> 3.8 tricky
> > >>
> > >> It's pretty easy to locate the DTB by automatically looking in
> > >> arch/*/boot/dts first, then if the file doesn't exist there, looking in
> > >> arch/*/boot instead as a legacy fallback.
> > > 
> > > Unfortunately, that does not work. See 2) below. Stale files even
> > > survive "make clean".
> > 
> > Oh right. I guess it works fine when having built 3.7 and then building
> > 3.8, since you'd look in the new location first and only fall back to
> > the old location if missing. However, if you then switch back to 3.7 to
> > test something, I guess that doesn't work so well.
> > 
> > I tend to always build the dtbs target, so I wasn't affected by this.
> > And, often run "git clean -f -d -x" too.
> > 
> > Grant, what's your call on this issue; should I/you look into ways to
> > clean the files better, or do you want to revert this?
> 
> I'm not Grant, but I think we're better off cleaning up the current situation
> instead of reverting. Otherwise 3.8 will be the oddball weird release; it's
> better to get people moved over to the new location.

Well... it seems that noone is willing to fix the patch, that new
location has no advantages, and that it is incompatible with
powerpc. Maybe oddball release is less evil here, and we can still fix
it up in -stable...
									Pavel
-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html