Re: [PATCH 4.4 39/53] PCI/PM: Add needs_resume flag to avoid suspend complete optimization

[Imre Deak]

On Wed, Jun 07, 2017 at 11:49:31AM +0200, Greg Kroah-Hartman wrote:
> On Tue, Jun 06, 2017 at 05:19:30PM +0100, Ben Hutchings wrote:
> > On Mon, 2017-06-05 at 18:17 +0200, Greg Kroah-Hartman wrote:
> > > 4.4-stable review patch.  If anyone has any objections, please let me know.
> > > 
> > > ------------------
> > > 
> > > From: Imre Deak <imre.deak@xxxxxxxxx>
> > > 
> > > commit 4d071c3238987325b9e50e33051a40d1cce311cc upstream.
> > > 
> > > Some drivers - like i915 - may not support the system suspend direct
> > > complete optimization due to differences in their runtime and system
> > > suspend sequence.  Add a flag that when set resumes the device before
> > > calling the driver's system suspend handlers which effectively disables
> > > the optimization.
> > > 
> > > Needed by a future patch fixing suspend/resume on i915.
> > 
> > The patch for i915 isn't in this series, or even in mainline yet, so I
> > don't think this belongs in stable yet.  Also, based on the commit
> > message (commit adfdf85d795f "drm/i915: Prevent the system suspend
> > complete optimization") it appears that only 4.8 and later are affected
> > by the bug.
> 
> Thanks for digging into this, I was kind of curious about it, but I
> thought it was fixing an old issue.  I've now dropped this from all
> stable queues.

4d071c3 and adfdf85 were pushed to separate trees (pci and drm) and the
corresponding drm pull request from Dave will only be sent to Linus by
this Sunday.

Yes, the fix is needed for v4.8, I forgot to add the tag for this to
4d071c3 and added the wrong 4.10 tag to adfdf85 (in confusion thinking
that I need to care only about stable releases and not longterm ones as
shown at https://www.kernel.org) :/

I'll resend the two patches with the correct tags once both patches are
in Linus' tree.

--Imre