Suspend2 isn't invasive.

[Nigel Cunningham]

Hi Ben.

On Thu, 2006-11-30 at 15:16 -0500, Ben Collins wrote:
> On Thu, 2006-11-30 at 17:15 +1100, Nigel Cunningham wrote:
> > Hi.
> > 
> > Sorry for not continuing the existing thread, but I've only just
> > subscribed, after seeing Ben's assertion.
> > 
> > Suspend2 isn't normally an invasive patch. At the moment it's a little
> > more invasive because I'm carrying the addition of linux/freezer.h that
> > will go into Linus' tree shortly, so let me prove a diffstat from 2.2.8.
> 
> Your definition of invasive and mine are two different things.
> 
> Mine includes things like integrating fixes from your code into a tree
> that has an existing suspend2 patch applied. So incremental patching is
> required, and this increases the amount of merging I have to do (even if
> it's not a conflict, it creates merges when I sync with Linus).

If you have fixes, please give them to me and I'll include them
immediately. Matt doesn't seem to be aware of it, but I'm maintaining a
Suspend2-against-Ubuntu branch on git.kernel.org, as well as one against
Linus' tree. These are normally updated on a daily basis, but I've been
a bit behind this week due to doing a milking course.

> My statement wasn't a stab at suspend2, as much as it was a declaration
> about our development criteria. The major point being that I like to
> keep changes to stock kernel code to a bare minimum, unless it is
> required. Since suspend2 cannot meet this criteria, and it's benefit
> does not outweigh that, I wont include it.

Understood. I would think that the benefits would far outweigh not
having it - could we perhaps discuss that side some more?

Regards,

Nigel