[Ksummit-2012-discuss] [ATTEND] writeback and kernel testing

Mel Gorman mgorman at suse.de
Tue Jun 19 16:12:55 UTC 2012

On Sat, Jun 16, 2012 at 06:29:49PM -0700, H. Peter Anvin wrote:
> On 06/16/2012 06:18 PM, Guenter Roeck wrote:
> > On Sat, Jun 16, 2012 at 10:50:41PM +0800, Fengguang Wu wrote:
> > [ ... ]
> >>
> >> Yeah, that should be a pretty common impression. The hardware
> >> capability and the fruit of catching one bug per day turns out to be
> >> much higher than my expectation :-)
> >>
> > How many servers do you actually need to compile 25k kernels per day ?
> > With my little system (i7-2600, SSD) I get about 1,000 randconfig kernel builds
> > per 24 hours, so it should not be too many.
> > 
> It would be nice to have a "kernel testing appliance" -- rather than a
> bunch of people writing similar stuff have something that anyone with
> spare hardware can put on a system. 

There are probably a few of these floating around. MM Tests is another one
although it's not something I've advertised very well and the last public
release was a long time ago. It is self-contained and for the most part it
is able to download and build whatever it needs. There are some package
pre-requisities but I have a list for openSUSE that could be converted
for other distributions. I have not released recently as I have a ton of
historical data I wanted to dig through to make sure the recent tests are
really measuring the right thing.

It is aimed at testing a single kernel configuration and comparing
different kernel versions rather than different kernel configurations.
This is because some of the tests are very long lived.

> Bonus points if it actually can
> drive a physical test box, too. 

This is trickier because it depends on what distribution you are using.
My own setup is partially automated but parts of that automation
makes assumptions on what the bootloader configuration looks like.
Capturing serial console output is fairly easy because software like
conmux exist. Automatic power management would be a pain because there
are so many ways it can be achieved but it's not something I put any
effort into as too many of my test machines have no remote power

Mel Gorman

More information about the Ksummit-2012-discuss mailing list