On Fri, 20 Jan 2012 18:55:08 -0500
Konrad Rzeszutek Wilk [off-list ref] wrote:

On Fri, Jan 20, 2012 at 03:11:49PM -0800, Andrew Morton wrote:

quoted

On Fri, 20 Jan 2012 11:09:38 -0500
Konrad Rzeszutek Wilk [off-list ref] wrote:

quoted

quoted

quoted

Did you test this patch with a large amount of minors?

Sorry I didn't do runtime test.

Please do.

The poor guy probably doesn't know how to test it and surely it would
be quite a lot of work for him to do so.

Overall, it would be much more efficient if the tester of this code is
someone who is set up to easily apply the patch and test it.  ie: the
code maintainer(s).

<grins>

This patch aside - Andrew how do you deal with a large amount of patches
and make sure they are tested?

Not very well :(

Is there a weekly Test Tuesday where you 
kick off your automated tests and dilligently look for any variations?
Or is it more of compile the kernel with X features and run it for a week
doing normal (and abnormal!) things to see if it falls over?

I build all the patches I have along with all of linux-next and boot
the thing, then use the resulting kernel for a few hours compilation
testing, then shove it all into linux-next where I naively hope that
someone else is testing things.

The coverage which is obtained this way is pretty poor.