On Thu, Nov 29, 2018 at 12:10 PM Benjamin Tissoires <btissoir(a)redhat.com> wrote:
On Wed, Nov 28, 2018 at 6:48 PM Hans de Goede <hdegoede(a)redhat.com> wrote:
> On 28-11-18 18:18, Schmauss, Erik wrote:
> <big snip>
> >> Repeating myself I'm somewhat surprised by your and Bob's
> >> against my approach, since it gets acpica out of the game of having to
> >> what the LEN field means / when to use then LEN field and I expected that
> >> both you and Bob would actually be quite happy to no longer having to do
> >> that.
> > This is a good point..
> > Bob and I have talked it over. We'll accept your patch as long as your
surface 3 also
> > boots and behaves as expected with this patch
> I don't have a surface 3, I got involved in the whole surface 3 discussion
> because I was trying to help to get the battery monitoring support for it
> upstream, but I don't have one myself.
> Still I don't see how this can break the surface 3, since we simply end up
> potentially copying more data to the tmp-buffer then before my patch and
> if the Surface 3's opregion driver does not expect that data to be there
> it will simply ignore it.
> I will ask Benjamin Tissoires to test 4.20 + my patch on his Surface 3
> to double-check, but in the mean time I believe it is really time to get
> this fix into 4.20 now.
Thanks Hans for all of this work. And thanks everybody for taking care
of this problem.
I just tested the patch on top of 4.20-rc, and I can confirm that this
doesn't regress on the Surface 3.
So it is fine from my HW point of view to apply it.
Thank you for the confirmation.
I'm going to send a pull request including this patch later today.