gdbserver/ChangeLog?

[Daniel Jacobowitz]

On Thu, Aug 08, 2002 at 10:34:56AM -0400, Andrew Cagney wrote:
> >On Thu, Aug 08, 2002 at 10:22:43AM -0400, Andrew Cagney wrote:
> >
> >>>On Thu, Aug 08, 2002 at 10:13:22AM -0400, Andrew Cagney wrote:
> >>>
> >
> >>>>Daniel,
> >>>>
> >>>>Should there be a gdbserver/ChangeLog?  I was recently doing some code 
> >>>>archology and was finding that the presence of gdbserver entries in 
> >>>>gdb/ChangeLog was making things mighty confusing.  Often the same 
> >>>>function/variable appears but in a totally different context.
> >
> >>>
> >>>
> >>>Hmm, I think that's a good idea.  Is it worth moving entries from the
> >>>old ChangeLogs, or just starting a new one for future changes?
> >
> >>
> >>The ``rewriting history'' problem.  I don't know.
> >>
> >>I'm tempted with suggesting that entries in this years changelog being 
> >>moved but I don't know how independant they are.
> >
> >
> >That sounds good to me.  I'll give it a day or two and then move them
> >over.
> 
> Hmm, a better starting point would be date the last branch was cut 
> (which reminds me, the release process doesn't stamp the changelog when 
> that is done ...).  Otherwize a diff of the changelogs between two 
> releases won't make much sense -> it will contain a lot of noise.

Yeah, you're right.

> The other more pragmatic aproach is to start it today, in 4 months, the 
> problem will go away anyway.

There've been significant changes since the 5.2 release; I'd rather do
it from there.

-- 
Daniel Jacobowitz                           Carnegie Mellon University
MontaVista Software                         Debian GNU/Linux Developer