Re: [PATCH v4 5/5] drm/msm/dsi: Add support for RGB101010 pixel format

[Dmitry Baryshkov]

On Mon, Mar 23, 2026 at 03:28:59PM +0000, Alexander Koskovich wrote:
> On Saturday, March 21st, 2026 at 3:52 AM, Alexander Koskovich <akoskovich@xxxxx> wrote:
> 
> >  
> > +	if (dsi->format == MIPI_DSI_FMT_RGB101010 &&
> > +	    !msm_dsi_host_version_geq(msm_host, MSM_DSI_VER_MAJOR_6G,
> > +				     MSM_DSI_6G_VER_MINOR_V2_1_0)) {
> > +		DRM_DEV_ERROR(&msm_host->pdev->dev,
> > +			      "RGB101010 not supported on this DSI controller\n");
> > +		return -EINVAL;
> > +	}
> > +
> >
> 
> Was tipped off to this bit downstream where if DSC is enabled and panel uses
> RGB101010, it overrides back to RGB888:
> https://git.codelinaro.org/clo/la/platform/vendor/opensource/display-drivers/-/blob/DISPLAY.LA.4.0.r2-10000-lanai.0/msm/dsi/dsi_display.c?ref_type=tags#L7209
> 
> Only does this if widebus isn't also enabled though. Should we do this as well
> upstream?

I'd suggest adding a bit fat warning here if widebus is not available,
add a comment about downstream switching to RGB888 and letting it land
this way. I'm more than sure that there would be some additional tricks
if one uses RGB101010 on SDM845 or SDM670, so it's better to let people
having those panels face the need to research those rather than making
them believe that the panels should work.

-- 
With best wishes
Dmitry