From: Rafaël Carré <rafael.carre_at_gmail.com>
Date: Mon, 14 Jun 2010 15:38:58 +0200

On Mon, 14 Jun 2010 14:06:52 +0200
Frank Gevaerts <frank_at_gevaerts.be> wrote:

> On Mon, Jun 14, 2010 at 07:14:30PM +1000, Jonathan Gordon wrote:
> > I am equally annoyed at the change to ACTUAL TARGET CODE WHICH WAS
> > UNTESTED! http://svn.rockbox.org/viewvc.cgi?view=rev;revision=26832
> > If that happens in the future then send an email to the dev mailing
> > list (or the maintainer) or create a bug for it, NEVER commit code
> > to target which you personally cant test on when it is a hardware
> > specific change.
>
> I agree there. That sort of commit should NEVER go in without testing.

I would have preferred to let the warning in the table until someone
with a target checks the code, rather than reverting the commit which
highlighted the (valid) warning.

But I have the impression that anything which makes the table not 100%
green is not wanted by the team.

What should I have done?

-- 
Rafaël Carré

Received on 2010-06-14