On Fri, 2005-06-03 at 22:30 -0700, Jonathan Day wrote:
> For the three people in hiding from all the other
> mailing lists discussing this, ha! You can't escape.
> Apparently, a number of kernel developers and some
> IBMers (or maybe IBMers who are kernel developers,

Both, actually.  It was started by a team of interns that did a quick
little two-month internship that worked on just the automation harness
alone.  They had some guidance from people that do work on the kernel
full time, like me.  In addition to my kernel work, I took over
developing it after their internships ended.

> I'm
> not sure) have been working on a test suite for the
> Linux kernel that seems to test a bunch of kernel
> internals for errors and timings.

Well, it makes sure that some things don't regress performance-wise, but
it doesn't really check for timings.  The most important thing is that
it's automated, and doesn't need any user intervention when things
crash.

> The problem, as I see it, is that there are now an
> awful lot of code validators, API conformance tests,
> benchmarking utilities, etc, but no real clear picture
> of what program does what, how the output of the
> different programs relate (if at all), what is of
> actual practical use to a developer, what is of actual
> practical interest to a user

The IBM thing isn't all that important in the grand scheme of things.
IT just allows things to be automated with little user intervention, and
to produce a "big picture" at the end of it all.

-- Dave

_______________________________________________
PDXLUG mailing list
[email protected]
http://lists.pdxlug.org/mailman/listinfo/pdxlug
IRC: irc.freenode.net #pdxlug

Reply via email to