freebsd stable - Mar 2015 - 10-STABLE suspend failure

Adrian,

Thanks! I can confirm that rolling back the intel_opregion commit 279961
restores suspend. I only rolled back this commit, so there is no doubt this
is the culprit. Of course, rolling it back also breaks brightness control.

Does it look like there is a reasonable explanation for this? Without
looking at the Linux code (which I probably would not understand anyway),
I'd guess something is missing that is needed for suspend to work.

Kevin Oberman, Network Engineer, Retired
E-mail: rkoberman at gmail.com

On Wed, Mar 25, 2015 at 1:13 PM, Adrian Chadd <adrian at freebsd.org>
wrote:
> Hi,
>
> that GEM idle fail is problematic.
>
> Try backing it out say, 1 month or so? I think that's before the drm
> merge from head to stable/10.
>
>
> -a
>

I honestly have no idea, and this is kind of why I didn't want to
commit it to stable/10 without it getting a whole lot more testing.

Does -HEAD also fail to suspend for you?


-a