Re: [PATCH/RFC 05/12] drm: rcar-du: lvds: Add data swap support

[Laurent Pinchart]

Hi Fabrizio,

On Mon, Aug 05, 2019 at 09:32:42AM +0000, Fabrizio Castro wrote:
> On 02 August 2019 09:06 Laurent Pinchart wrote:
> > On Fri, Aug 02, 2019 at 08:34:02AM +0100, Fabrizio Castro wrote:
> > > When in vertical stripe mode of operation, there is the option
> > > of swapping even data and odd data on the two LVDS interfaces
> > > used to drive the video output.
> > > Add data swap support by exposing a new DT property named
> > > "renesas,swap-data".
> > >
> > > Signed-off-by: Fabrizio Castro <fabrizio.castro@xxxxxxxxxxxxxx>
> > > ---
> > >  drivers/gpu/drm/rcar-du/rcar_lvds.c | 23 ++++++++++++++++-------
> > >  1 file changed, 16 insertions(+), 7 deletions(-)
> > >
> > > diff --git a/drivers/gpu/drm/rcar-du/rcar_lvds.c b/drivers/gpu/drm/rcar-du/rcar_lvds.c
> > > index 3aeaf9e..c306fab 100644
> > > --- a/drivers/gpu/drm/rcar-du/rcar_lvds.c
> > > +++ b/drivers/gpu/drm/rcar-du/rcar_lvds.c
> > > @@ -69,6 +69,7 @@ struct rcar_lvds {
> > >
> > >  	struct drm_bridge *companion;
> > >  	bool dual_link;
> > > +	bool stripe_swap_data;
> > >  };
> > >
> > >  #define bridge_to_rcar_lvds(bridge) \
> > > @@ -439,12 +440,16 @@ static void rcar_lvds_enable(struct drm_bridge *bridge)
> > >  	rcar_lvds_write(lvds, LVDCHCR, lvdhcr);
> > >
> > >  	if (lvds->info->quirks & RCAR_LVDS_QUIRK_DUAL_LINK) {
> > > -		/*
> > > -		 * Configure vertical stripe based on the mode of operation of
> > > -		 * the connected device.
> > > -		 */
> > > -		rcar_lvds_write(lvds, LVDSTRIPE,
> > > -				lvds->dual_link ? LVDSTRIPE_ST_ON : 0);
> > > +		u32 lvdstripe = 0;
> > > +
> > > +		if (lvds->dual_link)
> > > +			/*
> > > +			 * Configure vertical stripe based on the mode of
> > > +			 * operation of the connected device.
> > > +			 */
> > > +			lvdstripe = LVDSTRIPE_ST_ON | (lvds->stripe_swap_data ?
> > > +						       LVDSTRIPE_ST_SWAP : 0);
> > 
> > Would the following be simpler ?
> > 
> > 		lvdstripe = (lvds->dual_link ? LVDSTRIPE_ST_ON : 0)
> > 			  | (lvds->stripe_swap_data ? LVDSTRIPE_ST_SWAP : 0);
> > 
> > > +		rcar_lvds_write(lvds, LVDSTRIPE, lvdstripe);
> 
> I actually think I need to rework this patch slightly, because the user manual
> says that ST_SWAP is reserved for LVD1STRIPE, so I need to make sure we
> don't set it for LVDS1.
> 
> So perhaps, this could translate to something like:
> If (lvds->dual_link)
> 	lvdstripe = LVDSTRIPE_ST_ON | (<swap-whatever> && lvds->companion) ? LVDSTRIPE_ST_SWAP : 0);
> 
> I don't think we should be setting LVDSTRIPE_ST_SWAP without LVDSTRIPE_ST_ON, this solution
> would allow us to test lvds->dual_link only once, and will prevent us from setting LVDSTRIPE_ST_SWAP if
> LVDSTRIPE_ST_ON is not set or if we are touching LVDS1.
> 
> What do you think?

I was thinking that lvds->stripe_swap_data should only be set when
lvds->dual_link is set and lvds->companion is non-NULL, so both are
roughly equivalent. It's a detail anyway, it doesn't matter too much.

> > >  	}
> > >
> > >  	/*
> > > @@ -770,8 +775,12 @@ static int rcar_lvds_parse_dt(struct rcar_lvds *lvds)
> > >  		}
> > >  	}
> > >
> > > -	if (lvds->dual_link)
> > > +	if (lvds->dual_link) {
> > > +		lvds->stripe_swap_data = of_property_read_bool(
> > > +						lvds->dev->of_node,
> > > +						"renesas,swap-data");
> > >  		ret = rcar_lvds_parse_dt_companion(lvds);
> > > +	}
> > 
> > As explained in the review of the corresponding DT bindings, I think
> > this should be queried from the remote device rather than specified in
> > DT.
> > 
> > >
> > >  done:
> > >  	of_node_put(local_output);

-- 
Regards,

Laurent Pinchart