mbox series

[RESEND,0/5] drm/mxsfb: Fix runtime PM for unpowering lcdif block

Message ID cover.1537191359.git.leonard.crestez@nxp.com (mailing list archive)
Headers show
Series drm/mxsfb: Fix runtime PM for unpowering lcdif block | expand

Message

Leonard Crestez Sept. 17, 2018, 1:42 p.m. UTC
Adding lcdif nodes to a power domain currently doesn't work, it results
in black/corrupted screens or hangs. While the driver does enable
runtime pm it does not deal correctly with the block being unpowered.

---

All patches in this series have review tags from a while ago and I
tested them again on top of next-20180913. No changes since last
version: https://lkml.org/lkml/2018/8/27/299

This series stalled so I reached out to Marek on IRC and he was
surprised to be listed as maintainer and asked me to resend and add
Daniel Vetter.

Perhaps it would help to clarify that the pengutronix people should feel
free to push patches in this area?

Right now drm/imx is mostly for IPUv3 but there are other display output
paths on imx, such as the LCDIF supported by this driver. This LCDIF
block is included on imx8 so still quite relevant.

Leonard Crestez (5):
  drm/mxsfb: Move axi clk enable/disable to crtc enable/disable
  drm/mxsfb: Fix initial corrupt frame when activating display
  drm/mxsfb: Add pm_runtime calls to pipe_enable/disable
  drm/mxsfb: Add PM_SLEEP support
  drm/mxsfb: Switch to drm_atomic_helper_commit_tail_rpm

 drivers/gpu/drm/mxsfb/mxsfb_crtc.c | 53 +++++++++++++++++++-----------
 drivers/gpu/drm/mxsfb/mxsfb_drv.c  | 40 ++++++++++++++++++++++
 2 files changed, 74 insertions(+), 19 deletions(-)

Comments

Leonard Crestez Sept. 20, 2018, 3:59 p.m. UTC | #1
On Mon, 2018-09-17 at 16:37 -0700, Stefan Agner wrote:
> On 17.09.2018 12:16, Sean Paul wrote:
> > On Mon, Sep 17, 2018 at 04:42:10PM +0300, Leonard Crestez wrote:
> > > Adding lcdif nodes to a power domain currently doesn't work, it results
> > > in black/corrupted screens or hangs. While the driver does enable
> > > runtime pm it does not deal correctly with the block being unpowered.
> > > 
> > > All patches in this series have review tags from a while ago and I
> > > tested them again on top of next-20180913. No changes
> > > 
> > > This series stalled so I reached out to Marek on IRC and he was
> > > surprised to be listed as maintainer
> > 
> > Hopefully not too surprised since Marek added themself to MAINTAINERS when
> > adding the driver :-)
> 
> There have been some confusion about the DRM development processes
> around the mxsfb already in the past.
>
> > I suppose we should probably move this to drm-misc since it qualifies as a
> > "small driver" and needs a home. Looking through git history shows the last
> > mxsfb-specific change was back in 02/17. Everything else has been drm-wide
> > refactors. Thoughts?
> > 
> > Marek/Leonard: Care to sign up to be listed as a reviewers?
> 
> drm-misc seems to make sense. I volunteer to be listed as reviewer or co-maintainer.

This gets +1 from me, you were very helpful during review.

Next would be for one of the drm-misc maintainers to post a change for
the MAINTAINER file and also merge my series?

--
Regards,
Leonard
Sean Paul Sept. 20, 2018, 4:39 p.m. UTC | #2
On Thu, Sep 20, 2018 at 11:59 AM Leonard Crestez
<leonard.crestez@nxp.com> wrote:
>
> On Mon, 2018-09-17 at 16:37 -0700, Stefan Agner wrote:
> > On 17.09.2018 12:16, Sean Paul wrote:
> > > On Mon, Sep 17, 2018 at 04:42:10PM +0300, Leonard Crestez wrote:
> > > > Adding lcdif nodes to a power domain currently doesn't work, it results
> > > > in black/corrupted screens or hangs. While the driver does enable
> > > > runtime pm it does not deal correctly with the block being unpowered.
> > > >
> > > > All patches in this series have review tags from a while ago and I
> > > > tested them again on top of next-20180913. No changes
> > > >
> > > > This series stalled so I reached out to Marek on IRC and he was
> > > > surprised to be listed as maintainer
> > >
> > > Hopefully not too surprised since Marek added themself to MAINTAINERS when
> > > adding the driver :-)
> >
> > There have been some confusion about the DRM development processes
> > around the mxsfb already in the past.
> >
> > > I suppose we should probably move this to drm-misc since it qualifies as a
> > > "small driver" and needs a home. Looking through git history shows the last
> > > mxsfb-specific change was back in 02/17. Everything else has been drm-wide
> > > refactors. Thoughts?
> > >
> > > Marek/Leonard: Care to sign up to be listed as a reviewers?
> >
> > drm-misc seems to make sense. I volunteer to be listed as reviewer or co-maintainer.
>
> This gets +1 from me, you were very helpful during review.
>
> Next would be for one of the drm-misc maintainers to post a change for
> the MAINTAINER file and also merge my series?
>

I've posted the MAINTAINERS patch here:
https://lists.freedesktop.org/archives/dri-devel/2018-September/190414.html

Once I get some Acks, we can merge that and one of Stefan or Marek can apply the
patchset to drm-misc.

Sean

> --
> Regards,
> Leonard
Sean Paul Sept. 26, 2018, 8:13 a.m. UTC | #3
On Mon, Sep 17, 2018 at 04:42:10PM +0300, Leonard Crestez wrote:
> Adding lcdif nodes to a power domain currently doesn't work, it results
> in black/corrupted screens or hangs. While the driver does enable
> runtime pm it does not deal correctly with the block being unpowered.
> 
> ---
> 
> All patches in this series have review tags from a while ago and I
> tested them again on top of next-20180913. No changes since last
> version: https://lkml.org/lkml/2018/8/27/299
> 
> This series stalled so I reached out to Marek on IRC and he was
> surprised to be listed as maintainer and asked me to resend and add
> Daniel Vetter.
> 
> Perhaps it would help to clarify that the pengutronix people should feel
> free to push patches in this area?
> 
> Right now drm/imx is mostly for IPUv3 but there are other display output
> paths on imx, such as the LCDIF supported by this driver. This LCDIF
> block is included on imx8 so still quite relevant.
> 
> Leonard Crestez (5):
>   drm/mxsfb: Move axi clk enable/disable to crtc enable/disable
>   drm/mxsfb: Fix initial corrupt frame when activating display
>   drm/mxsfb: Add pm_runtime calls to pipe_enable/disable
>   drm/mxsfb: Add PM_SLEEP support
>   drm/mxsfb: Switch to drm_atomic_helper_commit_tail_rpm

The whole set is

Reviewed-by: Sean Paul <seanpaul@chromium.org>


> 
>  drivers/gpu/drm/mxsfb/mxsfb_crtc.c | 53 +++++++++++++++++++-----------
>  drivers/gpu/drm/mxsfb/mxsfb_drv.c  | 40 ++++++++++++++++++++++
>  2 files changed, 74 insertions(+), 19 deletions(-)
> 
> -- 
> 2.17.1
>
Stefan Agner Sept. 26, 2018, 8:12 p.m. UTC | #4
On 20.09.2018 18:39, Sean Paul wrote:
> On Thu, Sep 20, 2018 at 11:59 AM Leonard Crestez
> <leonard.crestez@nxp.com> wrote:
>>
>> On Mon, 2018-09-17 at 16:37 -0700, Stefan Agner wrote:
>> > On 17.09.2018 12:16, Sean Paul wrote:
>> > > On Mon, Sep 17, 2018 at 04:42:10PM +0300, Leonard Crestez wrote:
>> > > > Adding lcdif nodes to a power domain currently doesn't work, it results
>> > > > in black/corrupted screens or hangs. While the driver does enable
>> > > > runtime pm it does not deal correctly with the block being unpowered.
>> > > >
>> > > > All patches in this series have review tags from a while ago and I
>> > > > tested them again on top of next-20180913. No changes
>> > > >
>> > > > This series stalled so I reached out to Marek on IRC and he was
>> > > > surprised to be listed as maintainer
>> > >
>> > > Hopefully not too surprised since Marek added themself to MAINTAINERS when
>> > > adding the driver :-)
>> >
>> > There have been some confusion about the DRM development processes
>> > around the mxsfb already in the past.
>> >
>> > > I suppose we should probably move this to drm-misc since it qualifies as a
>> > > "small driver" and needs a home. Looking through git history shows the last
>> > > mxsfb-specific change was back in 02/17. Everything else has been drm-wide
>> > > refactors. Thoughts?
>> > >
>> > > Marek/Leonard: Care to sign up to be listed as a reviewers?
>> >
>> > drm-misc seems to make sense. I volunteer to be listed as reviewer or co-maintainer.
>>
>> This gets +1 from me, you were very helpful during review.
>>
>> Next would be for one of the drm-misc maintainers to post a change for
>> the MAINTAINER file and also merge my series?
>>
> 
> I've posted the MAINTAINERS patch here:
> https://lists.freedesktop.org/archives/dri-devel/2018-September/190414.html
> 
> Once I get some Acks, we can merge that and one of Stefan or Marek can apply the
> patchset to drm-misc.

Applied, fixed a white space issue in patch 4 and pushed to drm-misc.

Thanks Sean for helping out here!

And thanks Leonard for working on this and push for its inclusion!

--
Stefan

> 
> Sean
> 
>> --
>> Regards,
>> Leonard