Boot Speed Testing Results

[Pete Graner]

On 12/12/2011 06:21 PM, Martin Owens wrote:
> On Mon, 2011-12-12 at 16:17 -0500, Pete Graner wrote:
>> As far as benchmarking, load&  stress tests, they are on the long
>> term
>> roadmap for QA but not for this cycle.
>
> Do we have a facility to test workingness? My computer doesn't register
> the PCI bus because of a race condition in newer kernels and it'd be
> nice to know if I test daily ISOs so it can report the continued
> problem?
>
> Martin,

Yes (however I'll caveat this due to your case). We do daily ISO 
testing. These test the ISO to ensure it boots and comes up to a desktop.

If its broke the ISO is rejected the issue fixed or reverted and then 
its respun and the process is repeated. This is done virtualized, 
however boot speed testing also validates the virtualized testing buy 
doing it on a limited set of bare metal.

So that brings us to your specific case. This is something that the 
kernel team needs to have brought to their attention, its most likely hw 
specific since its not been seen in the dailies.

We'll need a bug filed subscribe Joe Salisbury (cc'd) and drop him a 
quick email letting him know so he can get it in front of a kernel dev 
so we can figure out how this has escaped testing.

Thanks

~pete
-- 
Pete Graner - Ubuntu Kernel & QA Team Manager - <pgraner at canonical.com>
Canonical Ltd. - http://www.canonical.com/