lkml.org 
[lkml]   [2022]   [Jun]   [24]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
Date
SubjectRe: [PATCH v1 2/3] drm/msm/dp: decoupling dp->id out of dp controller_id at scxxxx_dp_cfg table
Quoting Stephen Boyd (2022-06-24 17:11:01)
> Quoting Abhinav Kumar (2022-06-24 17:03:37)
> >
> > So let me try to explain this as this is what i understood from the
> > patch and how kuogee explained me.
> >
> > The ordering of the array still matters here and thats what he is trying
> > to address with the second change.
>
> The order of the array should not matter. That's the problem.

It seems like somewhere else the order of the array matters, presumably
while setting up encoders?

>
> >
> > So as per him, he tried to swap the order of entries like below and that
> > did not work and that is incorrect behavior because he still retained
> > the MSM_DP_CONTROLLER_x field for the table like below:
> >
> > diff --git a/drivers/gpu/drm/msm/dp/dp_display.c
> > b/drivers/gpu/drm/msm/dp/dp_display.c
> > index dcd80c8a794c..7816e82452ca 100644
> > --- a/drivers/gpu/drm/msm/dp/dp_display.c
> > +++ b/drivers/gpu/drm/msm/dp/dp_display.c
> > @@ -140,8 +140,8 @@ static const struct msm_dp_config sc7180_dp_cfg = {
> >
> > static const struct msm_dp_config sc7280_dp_cfg = {
> > .descs = (const struct msm_dp_desc[]) {
> > - [MSM_DP_CONTROLLER_0] = { .io_start = 0x0ae90000,
> > .connector_type = DRM_MODE_CONNECTOR_DisplayPort, .wide_bus_en = true },
> > [MSM_DP_CONTROLLER_1] = { .io_start = 0x0aea0000,
> > .connector_type = DRM_MODE_CONNECTOR_eDP, .wide_bus_en = true },
> > + [MSM_DP_CONTROLLER_0] = { .io_start = 0x0ae90000,
> > .connector_type = DRM_MODE_CONNECTOR_DisplayPort, .wide_bus_en = true },
> > },
> > .num_descs = 2,
> > };
> >
> >
> > The reason order is important is because in this function below, even
> > though it matches the address to find which one to use it loops through
> > the array and so the value of *id will change depending on which one is
> > located where.
> >
> > static const struct msm_dp_desc *dp_display_get_desc(struct
> > platform_device *pdev,
> > unsigned int *id)
>
> Thanks! We should fix this function to not overwrite the id.
>

Ah nevermind. I mixed up dp->id and h_tile_instance thinking one was
overwriting the other but that doesn't make any sense.

\
 
 \ /
  Last update: 2022-06-25 02:24    [W:0.117 / U:1.836 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site