[Freemint-list] FreeMiNT for 21st century
Peter Slegg
p.slegg at scubadivers.co.uk
Sat Jan 14 00:12:14 MSK 2017
> Date: Fri, 13 Jan 2017 10:39:28 +0000
> From: Alan Hourihane <alanh at fairlite.co.uk>
> Subject: Re: [Freemint-list] FreeMiNT for 21st century
> To: freemint-list at mail.atariforge.org
> Message-ID: <029fc05b-43c0-82c0-b115-0a8509783703 at fairlite.co.uk>
> On 13/01/17 00:39, Miro Krop??ek wrote:
> >
> > On 13 January 2017 at 07:54, Peter Slegg <p.slegg at scubadivers.co.uk
> > <mailto:p.slegg at scubadivers.co.uk>> wrote:
> >
> > I had assumed the builds were automated a long time ago or was it
> > just the kernel ?
> >
> > They are, it's just a set of scripts after all.
> >
> > There are several problems with current state of things, though:
> >
> > * if something breaks, you'll get to know about it only the next day
> >
>
> We could enable post-commit scripts where the build takes place after
> every commit, then you'd know about it immediately.
>
> > * to even get to know about it you must convince Alan to commit it first
> >
>
> That takes little convincing. I'm all for more people having access to
> the CVS (or git).
>
> > * if you commit something terrible, Alan will be very angry ;)
> >
>
> Mmm? We should be at least building the patches we commit so we don't
> break the build. Testing is a different matter as there may be subtle
> issues introduced. There aren't that many of us left to do that either.
>
Are there many unit test scripts or is this tricky with kernel code ?
Easier with xaaes I would imagine.
Peter
More information about the Freemint-list
mailing list