diff mbox series

Revert "drm/msm/dp: set self refresh aware based on PSR support"

Message ID 20230523151646.28366-1-johan+linaro@kernel.org
State New
Headers
Series Revert "drm/msm/dp: set self refresh aware based on PSR support" |

Commit Message

Johan Hovold May 23, 2023, 3:16 p.m. UTC 
  This reverts commit 1844e680d56bb0c4e0489138f2b7ba2dc1c988e3.

PSR support clearly is not ready for mainline and specifically breaks
virtual terminals which are no longer updated when PSR is enabled (e.g.
no keyboard input is echoed, no cursor blink).

Disable PSR support for now by reverting commit 1844e680d56b
("drm/msm/dp: set self refresh aware based on PSR support").

Cc: Vinod Polimera <quic_vpolimer@quicinc.com>
Cc: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Signed-off-by: Johan Hovold <johan+linaro@kernel.org>
---

Bjorn reported that PSR support broke virtual terminals two months ago, 
but this is still broken in 6.4-rc3:

	https://lore.kernel.org/lkml/20230326162723.3lo6pnsfdwzsvbhj@ripper/

despite the following series that claimed to address this:

	https://lore.kernel.org/lkml/1680271114-1534-1-git-send-email-quic_vpolimer@quicinc.com

Let's revert until this has been fixed properly.

Johan


 drivers/gpu/drm/msm/dp/dp_drm.c | 2 --
 1 file changed, 2 deletions(-)

Comments

Johan Hovold June 5, 2023, 10:02 a.m. UTC | #1 
[ +CC: Thorsten and regzbot so they can help with tracking this
regression ]

#regzbot introduced: v6.3..v6.4-rc1

On Tue, May 23, 2023 at 05:16:46PM +0200, Johan Hovold wrote:
> This reverts commit 1844e680d56bb0c4e0489138f2b7ba2dc1c988e3.
> 
> PSR support clearly is not ready for mainline and specifically breaks
> virtual terminals which are no longer updated when PSR is enabled (e.g.
> no keyboard input is echoed, no cursor blink).
> 
> Disable PSR support for now by reverting commit 1844e680d56b
> ("drm/msm/dp: set self refresh aware based on PSR support").
> 
> Cc: Vinod Polimera <quic_vpolimer@quicinc.com>
> Cc: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
> Signed-off-by: Johan Hovold <johan+linaro@kernel.org>
> ---
> 
> Bjorn reported that PSR support broke virtual terminals two months ago, 
> but this is still broken in 6.4-rc3:
> 
> 	https://lore.kernel.org/lkml/20230326162723.3lo6pnsfdwzsvbhj@ripper/
> 
> despite the following series that claimed to address this:
> 
> 	https://lore.kernel.org/lkml/1680271114-1534-1-git-send-email-quic_vpolimer@quicinc.com
> 
> Let's revert until this has been fixed properly.

Virtual terminals are still broken with 6.4-rc5 on the Lenovo ThinkPad
X13s two weeks after I reported this, and there has been no indication
of any progress in the other related thread:

	https://lore.kernel.org/lkml/ZHYPHnWoDbXB-fqe@hovoldconsulting.com

Seems like it is time to merge this revert to get this sorted.

Rob, Abhinav, Dmitry, can either of you merge this one and get it into
6.4-rc6?

Johan
Dmitry Baryshkov June 5, 2023, 10:05 a.m. UTC | #2 
On Mon, 5 Jun 2023 at 13:02, Johan Hovold <johan@kernel.org> wrote:
>
> [ +CC: Thorsten and regzbot so they can help with tracking this
> regression ]
>
> #regzbot introduced: v6.3..v6.4-rc1
>
> On Tue, May 23, 2023 at 05:16:46PM +0200, Johan Hovold wrote:
> > This reverts commit 1844e680d56bb0c4e0489138f2b7ba2dc1c988e3.
> >
> > PSR support clearly is not ready for mainline and specifically breaks
> > virtual terminals which are no longer updated when PSR is enabled (e.g.
> > no keyboard input is echoed, no cursor blink).
> >
> > Disable PSR support for now by reverting commit 1844e680d56b
> > ("drm/msm/dp: set self refresh aware based on PSR support").
> >
> > Cc: Vinod Polimera <quic_vpolimer@quicinc.com>
> > Cc: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
> > Signed-off-by: Johan Hovold <johan+linaro@kernel.org>
> > ---
> >
> > Bjorn reported that PSR support broke virtual terminals two months ago,
> > but this is still broken in 6.4-rc3:
> >
> >       https://lore.kernel.org/lkml/20230326162723.3lo6pnsfdwzsvbhj@ripper/
> >
> > despite the following series that claimed to address this:
> >
> >       https://lore.kernel.org/lkml/1680271114-1534-1-git-send-email-quic_vpolimer@quicinc.com
> >
> > Let's revert until this has been fixed properly.
>
> Virtual terminals are still broken with 6.4-rc5 on the Lenovo ThinkPad
> X13s two weeks after I reported this, and there has been no indication
> of any progress in the other related thread:
>
>         https://lore.kernel.org/lkml/ZHYPHnWoDbXB-fqe@hovoldconsulting.com
>
> Seems like it is time to merge this revert to get this sorted.
>
> Rob, Abhinav, Dmitry, can either of you merge this one and get it into
> 6.4-rc6?

Rob sent the pull request few hours ago, see
https://lore.kernel.org/dri-devel/CAF6AEGuHujkFjRa6ys36Uyh0KUr4Hd16u1EMqJo8tOZ3ifVubQ@mail.gmail.com/
Johan Hovold June 5, 2023, 10:18 a.m. UTC | #3 
On Mon, Jun 05, 2023 at 01:05:36PM +0300, Dmitry Baryshkov wrote:
> On Mon, 5 Jun 2023 at 13:02, Johan Hovold <johan@kernel.org> wrote:

> > Virtual terminals are still broken with 6.4-rc5 on the Lenovo ThinkPad
> > X13s two weeks after I reported this, and there has been no indication
> > of any progress in the other related thread:
> >
> >         https://lore.kernel.org/lkml/ZHYPHnWoDbXB-fqe@hovoldconsulting.com
> >
> > Seems like it is time to merge this revert to get this sorted.
> >
> > Rob, Abhinav, Dmitry, can either of you merge this one and get it into
> > 6.4-rc6?
> 
> Rob sent the pull request few hours ago, see
> https://lore.kernel.org/dri-devel/CAF6AEGuHujkFjRa6ys36Uyh0KUr4Hd16u1EMqJo8tOZ3ifVubQ@mail.gmail.com/

Ok, so you guys went with the module parameter hack. Whatever. As long
as the regression is finally fixed.

Next time, some visibility into your process would be appreciated to
avoid unnecessary work.

Johan
Linux regression tracking (Thorsten Leemhuis) June 6, 2023, 10:32 a.m. UTC | #4 
On 05.06.23 12:18, Johan Hovold wrote:
> On Mon, Jun 05, 2023 at 01:05:36PM +0300, Dmitry Baryshkov wrote:
>> On Mon, 5 Jun 2023 at 13:02, Johan Hovold <johan@kernel.org> wrote:
> 
>>> Virtual terminals are still broken with 6.4-rc5 on the Lenovo ThinkPad
>>> X13s two weeks after I reported this, and there has been no indication
>>> of any progress in the other related thread:
>>>
>>>         https://lore.kernel.org/lkml/ZHYPHnWoDbXB-fqe@hovoldconsulting.com
>>>
>>> Seems like it is time to merge this revert to get this sorted.

BTW, thx for bringing this to my attention!

>>> Rob, Abhinav, Dmitry, can either of you merge this one and get it into
>>> 6.4-rc6?
>>
>> Rob sent the pull request few hours ago, see
>> https://lore.kernel.org/dri-devel/CAF6AEGuHujkFjRa6ys36Uyh0KUr4Hd16u1EMqJo8tOZ3ifVubQ@mail.gmail.com/
> 
> Ok, so you guys went with the module parameter hack. Whatever. As long
> as the regression is finally fixed.

Yup. Let me tell regzbot about the fix:

#regzbot fix: drm/msm/dp: add module parameter for PSR
#regzbot ignore-activity

> Next time, some visibility into your process would be appreciated to
> avoid unnecessary work.

Yeah, that's something we IMHO sooner or later need to improve for all
of kernel development -- among others to give people that find existing
bug reports a chance to find patches that were posted or applied to
address the issue (and of course reporters also, like in this case).

Ciao, Thorsten (wearing his 'the Linux kernel's regression tracker' hat)
--
Everything you wanna know about Linux kernel regression tracking:
https://linux-regtracking.leemhuis.info/about/#tldr
That page also explains what to do if mails like this annoy you.
diff mbox series

Patch

diff --git a/drivers/gpu/drm/msm/dp/dp_drm.c b/drivers/gpu/drm/msm/dp/dp_drm.c
index 785d76639497..029e08c5bb06 100644
--- a/drivers/gpu/drm/msm/dp/dp_drm.c
+++ b/drivers/gpu/drm/msm/dp/dp_drm.c
@@ -117,8 +117,6 @@  static int edp_bridge_atomic_check(struct drm_bridge *drm_bridge,
 	if (WARN_ON(!conn_state))
 		return -ENODEV;
 
-	conn_state->self_refresh_aware = dp->psr_supported;
-
 	if (!conn_state->crtc || !crtc_state)
 		return 0;