Subject: Re: NetBSD kernel "underbelly" testing
To: Bill Studenmund <firstname.lastname@example.org>
From: Matthew Orgass <email@example.com>
Date: 05/22/2004 22:21:11
On 2004-05-22 firstname.lastname@example.org wrote:
> On Fri, May 21, 2004 at 06:36:36PM -0400, Matthew Orgass wrote:
> > On 2004-05-21 email@example.com wrote:
> > > I've been giving thought to testing the kernel using a virtualisation
> > > system like Xen. If we presuppose that NetBSD, being so portable, has
> > > a minimum amount of machine dependent code, we can get a good
> > > approximation for all NetBSD ports by testing NetBSD/xen on a Xen
> > > system.
> > That depends on what exactly is being tested and what all xen can do.
> > The major factors that I can think of for MI code are 32/64-bit pointers,
> > alignment, and endien. Being able to cycle through all of those would be
> > helpful (at least endien testing is likely to be impossible on xen).
> I don't think that's what Ben meant. I don't think he's talking about
> testing NetBSD across a spectrum of hardware/architectures, for which
> ILP32 vs LP64, alignment, and endianness make sense. I think he's instead
> implying that since a lot of NetBSD is MI, that one simulator can cover a
> lot of bases. Ben, please correct me if I'm wrong!
I didn't mean to imply that it wouldn't be helpful even without doing
any of that, just that it would not give a good approximation of all
NetBSD ports. I agree that even just testing base xen would be helpful
and can uncover a great deal that applies to other ports as well. I think
it would be more helpful if as much as is reasonable of the test code is
MI and can, with emulator specific code, be attached to other emulators as
well (which may have been the plan to begin with).