Re: [Linux-fbdev-devel] [PATCH v4] Frame Buffer driver for TI DA8xx/OMAP-L1xx

[Krzysztof H. <krz...@po...>]

On Tue, 30 Jun 2009 01:50:13 -0400
"Rajashekhara, Sudhakar" <sud...@ti...> wrote:

> Adds LCD controller (LCDC) driver for TI's DA8xx/OMAP-L1xx architecture.
> LCDC specifications can be found at http://www.ti.com/litv/pdf/sprufm0a.
> 
> LCDC on DA8xx consists of two independent controllers, the Raster Controller
> and the LCD Interface Display Driver (LIDD) controller. LIDD further supports
> character and graphic displays.
> 
> This patch adds support for the graphic display (Sharp LQ035Q3DG01) found on
> the DA830 based EVM. The EVM details can be found at:
> http://support.spectrumdigital.com/boards/dskda830/revc/.
> 
> Signed-off-by: Sudhakar Rajashekhara <sud...@ti...>
> Signed-off-by: Pavel Kiryukhin <pki...@ru...>
> Signed-off-by: Steve Chen <sc...@mv...>
> ---
>  Since the previous version, fb_setcolreg function has been modified for
>  8 and 16 bit modes.
> 
>  drivers/video/Kconfig    |   11 +
>  drivers/video/Makefile   |    1 +
>  drivers/video/da8xx-fb.c |  897 ++++++++++++++++++++++++++++++++++++++++++++++
>  include/video/da8xx-fb.h |  106 ++++++
>  4 files changed, 1015 insertions(+), 0 deletions(-)
>  create mode 100644 drivers/video/da8xx-fb.c
>  create mode 100644 include/video/da8xx-fb.h
> 

> diff --git a/drivers/video/da8xx-fb.c b/drivers/video/da8xx-fb.c
> new file mode 100644
> index 0000000..3135234
> --- /dev/null
> +++ b/drivers/video/da8xx-fb.c
> @@ -0,0 +1,897 @@

(...)

> +
> +static int fb_setcolreg(unsigned regno, unsigned red, unsigned green,
> +			      unsigned blue, unsigned transp,
> +			      struct fb_info *info)
> +{
> +	struct da8xx_fb_par *par = info->par;
> +	unsigned short *palette = (unsigned short *)par->v_palette_base;
> +	u_short pal;
> +
> +	if (regno > 255)
> +		return 1;
> +
> +	if (info->fix.visual == FB_VISUAL_DIRECTCOLOR)
> +		return 1;
> +
> +	if (info->var.bits_per_pixel == 8) {
> +		red >>= 4;
> +		green >>= 4;
> +		blue >>= 4;
> +
> +		pal = (red & 0x0f00);
> +		pal |= (green & 0x00f0);
> +		pal |= (blue & 0x000f);
> +
> +		palette[regno] = pal;
> +

I am losing my patience. The shift value for the red component is right (probably as I do 
not know the format of this palette registers). It is however almost surely wrong for other
componets. Just put this simple example value to get almost white color:
red = 0xf000;
green = 0xf000;
blue = 0xf000;

The resulting palette value would be 0xf00. Is it close to white on your controller?

Another example: blue
red = 0;
green = 0;
blue = 0xff00;

The result : 0x0. It does not differ from the value for black colour (rgb = 0/0/0).

> +	} else if ((info->var.bits_per_pixel == 16) && regno < 16) {
> +		red >>= (16 - info->var.red.length);
> +		red <<= info->var.red.offset;
> +
> +		green >>= (16 - info->var.green.length);
> +		green <<= info->var.green.offset;
> +
> +		blue >>= (16 - info->var.blue.length);
> +		blue <<= info->var.blue.offset;
> +
> +		par->pseudo_palette[regno] = red | green | blue;
> +
> +		palette[0] = 0x4000;
> +	}

This one is OK.

Regards,
Krzysztof

----------------------------------------------------------------------
Wybierz najlepsze ubezpieczenie komunikacyjne w Ergo Hestia
http://link.interia.pl/f222a