[Lucid SRU] UBUNTU: SAUCE: dell-laptop: fire SMI when toggling hardware killswitch

[Keng-Yü Lin]

2010/6/9 Chase Douglas <chase.douglas at canonical.com>:
> On Wed, 2010-06-09 at 15:08 +0200, Stefan Bader wrote:
>> On 06/09/2010 12:05 PM, Keng-Yü Lin wrote:
>> >   There are several sauce patches for dell-laptop.c in lucid tree. The
>> > code difference between lucid tree and stable tree makes it difficult
>> > to fix in a simple patch.
>>
>> Ok, sorry I missed that note about the SAUCE patch in the initial request. But
>> it would bring up the question to me, why the other patch has not gone upstream.
>> Was it tried but rejected? If so, why?
>> Is the same patch not needed for Maverick? Is the other patch in there? This
>> sounds a bit like being simpler to forget things on the way forward.
>
> I think way back in the early lucid development some patches were
> applied to our tree that did not come from upstream. Matthew Garrett,
> the upstream maintainer, took the patches and reworked them some.
> However, the damage is done now, so we can't converge the two without a
> wholesale replacement of parts of the lucid driver.
>
> That said, we need to be sure that anything we put into Lucid is also
> pushed upstream in a form that is applicable to their sources. In fact,
> it may make the most sense to push changes to this driver upstream
> first, and then propose a reworked version for lucid once it is
> accepted.
>
> -- Chase
>
>

dug back to the history found that several of these SAUCE patches
exist since karmic. much the same discussion two month ago while Chase
sent a patch on the same driver. Some explanation in that thread by
Mario Limonciello:

  "As I recall, during karmic we pulled a patch that didn't make it
upstream in the form we have it, and a bunch of those other patches
layered on top of it. I didn't have any type of notification that a
different form had finally landed until Matthew started pulling some
of the patches we had in too."

I tested maverick kernel (2.6.32-2.2) on the machine and this bug is
there. This sauce patch here does not fit in maverick or upstream. I
think I will work another patch for it without depending on the
previous sauce patch code.

agreed that it's a better process to fix upstream first and then gets
backported somehow. But I encountered this issue while working on some
enablement for lucid. This is why the patch for lucid comes first.


  Regards,
-kengyu