[1/3] ARM: dts: am335x-evm: Add blue-and-red-wiring -property to lcdc node

Commit Message

Jyri Sarha Sept. 16, 2016, 11:50 a.m. UTC

Add blue-and-red-wiring -property to lcdc node. The am335x-evm has
blue and red wires crossed to get 24-bit RGB (and 16-bit BGR)
support. After this patch am335x-evm supports BGR565, RGB888, and
XRGB8888 color formats. See details in
Documentation/devicetree/bindings/display/tilcdc/tilcdc.txt.

Signed-off-by: Jyri Sarha <[e-mail]>
Reviewed-by: Tomi Valkeinen <[e-mail]>
---
 arch/arm/boot/dts/am335x-evm.dts | 2 ++
 1 file changed, 2 insertions(+)

Comments

Nishanth Menon Sept. 16, 2016, 10:17 p.m. UTC | #1

On 09/16/2016 06:50 AM, Jyri Sarha wrote:
[..]
> +	blue-and-red-wiring = "crossed";

Dumb question: Why is this a Linux generic property - does it make 
sense to have "blue-and-red-wiring" as a Linux generic property? I 
mean, it reminds me little of those movies "cut the blue or red wire 
to diffuse the bomb" ;)

"The am335x-evm has blue and red wires crossed to get 24-bit RGB (and 
16-bit BGR". I mean, there has to be some sort of better description 
specific to lcdc if some product used yellow and green wires for the 
same signals crossed?

Jyri Sarha Sept. 19, 2016, 7:55 a.m. UTC | #2

On 09/17/16 01:17, Nishanth Menon wrote:
> On 09/16/2016 06:50 AM, Jyri Sarha wrote:
> [..]
>> +    blue-and-red-wiring = "crossed";
> 
> Dumb question: Why is this a Linux generic property - does it make sense
> to have "blue-and-red-wiring" as a Linux generic property? I mean, it
> reminds me little of those movies "cut the blue or red wire to diffuse
> the bomb" ;)

It is a property of "ti,tilcdc" node. I did not think I should add a ti
specific prefix for it. And after all if some other SoC or display chip
has the same bug (eg. blue and red colour components swapped between two
different colour modes), they could use the same property.

Of course in theory, if such bugs were common, a generic property that
would describe how connections between different display data wires are
drawn could make sense, but AFAIK am335x is the only one out there with
such a bug [1].

> 
> "The am335x-evm has blue and red wires crossed to get 24-bit RGB (and
> 16-bit BGR". I mean, there has to be some sort of better description
> specific to lcdc if some product used yellow and green wires for the
> same signals crossed?
> 

The property and why it is there is described in detail in the binding
document referred to in the commit description.

The functional patches, with the binding document can be found here:
http://www.spinics.net/lists/devicetree/msg141246.html

[1] For more details see section 3.1.1 in AM335x Silicon Errata:
http://www.ti.com/general/docs/lit/getliterature.tsp?baseLiteratureNumber=sprz360

Best regards,
Jyri

Nishanth Menon Sept. 19, 2016, 2:47 p.m. UTC | #3

On 09/19/2016 02:55 AM, Jyri Sarha wrote:
[...]
> The functional patches, with the binding document can be found here:
> http://www.spinics.net/lists/devicetree/msg141246.html
>
> [1] For more details see section 3.1.1 in AM335x Silicon Errata:
> http://www.ti.com/general/docs/lit/getliterature.tsp?baseLiteratureNumber=sprz360
>

Thanks for the detailed explanation. 
http://www.spinics.net/lists/devicetree/msg141248.html helped explain 
the same.

Patch

diff --git a/arch/arm/boot/dts/am335x-evm.dts b/arch/arm/boot/dts/am335x-evm.dts
index 5d28712..c3cbce0 100644
--- a/arch/arm/boot/dts/am335x-evm.dts
+++ b/arch/arm/boot/dts/am335x-evm.dts
@@ -497,6 +497,8 @@ 
 
 &lcdc {
 	status = "okay";
+
+	blue-and-red-wiring = "crossed";
 };
 
 &elm {