Re: 4.2 Backport fix for PR libfortran/31196

[Steve Kargl <sgk at troutmask dot apl dot washington dot edu>]

On Sun, Apr 15, 2007 at 06:23:08PM +0200, FX Coudert wrote:
> Hi Thomas,
> 
> >can I backport http://gcc.gnu.org/viewcvs? 
> >root=gcc&view=rev&rev=123195 to 4.2?
> 
> I think you don't need Mark's approval. The gcc.gnu.org front page  
> says "regression fixes & docs only", which IIRC this PR is (a  
> regression wrt g77).

The PR deals with RESHAPE, so it can't be a regression wrt to g77.

> Furthermore, I haven't seen any news of a release since 2007-03-22  
> (http://gcc.gnu.org/ml/gcc/2007-03/msg00865.html), which is more than  
> 3 weeks ago, so I guess things have slowed down.

I do, however, believe that Thomas should do the back port and
send markm a request for inclusion.

After watching the slow death of 4.2 and its potential impact on
gfortran, I've concluded that the GCC development model is fundamentally
flawed.  The 4.2 branch was clearly branched too early in the development
cycle.

IMHO, there should a current stable release branch (aka 4.1.x) and
mainline.  When a new release (aka 4.2) is set to occur, mainline
is branched and a moratorium on commits to mainline is then put 
into place.  No one is allowed to commit mainline (ie., SC, global
write maintainers, zip, nada, no one).  Everyone's attention should
be squarely on getting 4.2 out of the door, which should only take
a week or two (barring any catastrophic showstopper bug).
 
Yes, I know working on that new cool feature is much more interesting
than fixing bugs.  Yes, I know getting all contributors to buy into
a new development is tantamount to herding cats.  But, at this point
in time 4.2 appears to be dead.

-- 
Steve