[U-Boot] [PATCH] Fix Freescale link scripts for newer GCCs

Wolfgang Denk wd at denx.de
Wed Feb 18 22:54:34 CET 2009


Dear Jon,

In message <499C58FA.20507 at freescale.com> you wrote:
> 
> > Trent - you have  obviously  already  spend  a  lot  of  effort  into
> > analyzing  this  code - most probably more time than any other of us.
> > Based on your experience and the effort you already spent I think  it
> > would be best if you could provided a patch to fix all boards.
> 
> Just as a clarification, are you asking Trent to *fix* it
> for all boards, or to generate a patch that will apply to
> all boards?

He. Of course I can only ask to provide patch that, to  the  best  of
Trent's  experience,  is  supposed  to  fix  the problems. I am aware
thatthere is a chance that it will not work on some boards, but  then
it's  up to the board maintainers to complain and to help fixing this
(rpobabmly small) number of remaining issues.

> Specifically, I recall Trent's hesitation was he doesn't have
> access to most boards (not even all PPC boards), and he wasn't
> even sure it was a problem on some other ARCHes in the first place.

Good point. But which other options do we have?

We can:

1) apply the patch to a small subset of the boards that are known to
   be affected (at least all PowerPC boards)

2) apply the patch to all boards that are known to be affected (all
   PowerPC boards)

3) apply the patch to all boards that are most probably affected (all
   boards)

If we go for 1) or even 2), ther eis a pretty high  chance  that  the
problem  will  byte  us later, and until somebody remembers that this
was raised (and  fixed)  before  in  another  architecture,  lots  of
efforts will be wasted.

However, if we now apply a patch that breaks a number  of  boards  we
will quickly learn where the change works and where not.

In my point of view the risk of a potential damage involved with 3) is
smaller than the probability for damage and the amount of effort cause
by 2) and especially by 1).

I don;t think such a patch should go into the current release, but
it's a good candidate for the "next" branch and thus for v2009.05.

> But, if you are asking for an off-the-cuff patch that can be
> applied to all boards without testing by Trent, then I am
> confident he would be willing to help here.

I really hope Trent will help out. 

Best regards,

Wolfgang Denk

-- 
DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de
"We don't have to protect the environment -- the Second Coming is  at
hand."                                                   - James Watt


More information about the U-Boot mailing list