[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Debian non-x86 kernel arches



On Sat, May 22, 2004 at 04:20:01PM -0500, Troy Benjegerdes wrote:
> >  
> > > I'd like to propose we attempt to build x86, amd64, ppc, and ia64
> > > kernels from the same source tree.
> > 
> > Your proposal has first- and second-class archs in it as well.  And it
> > means that several machines would be required in order to build all
> > binary packages from that one source package, even if just a small
> > arch-specific change was made.  I won't even speak of a sufficiently
> > large testbed, which for PowerPC alone involves half a dozen different
> > machines at the moment.
> 
> Testing is probably the biggest issue. How is this currently handled?

You upload and when you get no feedback, you suppose it is working :)

Seriously, the problem for powerpc is that most users are used to, like
you, build their own kernels, and it has been quite difficult to get
testing done from more than a handfull of people. Added to that the
tendency of pmac users to write HOWTO on how to circumvent problems in
stead of filling bug report and helping fi them, ...

> How do you verify a new package is okay? Is there any way to automate
> this.. or at least make it semi-automated? Obviously if if something
> fails, a machine probably needs to be reset. But if everything is
> working (which should be the normal case), what would prevent an
> automated system to boot the latest kernel and run some set of tests
> after bootup?

I doubt that is possible. The amount of different machines that run
powerpc is too diverse for testing before releasing a kernel, but this
is the same with every other debian package. You test it locally,
releases it, and then wait for user feedback and bug reports to solve
problems that may arrise and you have no way of testing. 

Friendly,

Sven Luther



Reply to: