OMAP-L138 framebuffer with 8bpp

Hi,

Thanks for your post.

Usually a palette entry would be selected from the look-up palette by the pixel data and once it is selected, it's contents would be sent to the gray-scaler/serializer and If it is monochrome data, it would be encoded as 4 bits passive (STN) mode.

These 4-bit values would be used to select one of the 16 intensity levels, as shown in Table 24-7, section 24.2.5.4.1 in the OMAPl138 TRM as below. A patented algorithm would be used during this processing to provide an optimized intensity value that matches the eye's visual perception of color/gray gradations.

For more details, please refer entry palette/frame buffer format (1,2,4, 12,16 BPP) in sections 24.2.5.1 & 24.2.5.2 from omapl138 TRM below:

www.ti.com/.../spruh77a.pdf

Thanks & regards,

Sivaraj K

---------------------------------------------------------------------------------
Please click the Verify Answer button on this post if it answers your question.
---------------------------------------------------------------------------------

Hi Sivaraj!

Thank you for your replay of my post.
I know and I've read the OMAP-L138 TRM's sections regarding the LCDC.

But under Linux I don't program or set the controller registers directly.
Instead I've to use a device driver.

The device driver for the framebuffer is "da8xx-fb.c" located under "/drivers/video/fbdev/" in the Linux source tree.
This driver configures a framebuffer and the LCDC registers.

In my board file I have to register a device with the function da8xx_register_lcdc():
...
ret = da8xx_register_lcdc(&wvga_16bpp_pdata);
if (ret) {
pr_warn("da8xx_register_lcdc(&wvga_16bpp_pdata) failed: %d\n", ret);
return;
}
...

The structurte for the platform data is:

struct da8xx_lcdc_platform_data wvga_16bpp_pdata = {
.manu_name = "ti",
.controller_data = &lcd_cfg_16bpp,
.type = "DVI_WVGA",
};

An element of this structure is "controller_data" and I've set this to:

static struct lcd_ctrl_config lcd_cfg_16bpp = {
.panel_shade = COLOR_ACTIVE,
.ac_bias = 255,
.ac_bias_intrpt = 0,
.dma_burst_sz = 16,
.bpp = 16,
.fdd = 0,
.tft_alt_mode = 0,
.stn_565_mode = 0,
.mono_8bit_mode = 0,
.sync_edge = 0,
.raster_order = 0,
.fifo_th = 6,
};

With this parameters I get a working display (TFT display with WVGA (800x400 pixel) resolution) and a framebuffer with a format of 16bpp.

Trying to set the element "bpp" to the value of "8" results in a not working display.

My questions are:
Is the device driver "da8xx-fb" able to set and configure a framebuffer with a format of 8bpp?
How can I set the parameters so that I get a framebuffer with a format of 8bpp?
How I've to set an appropriate palette?
Do you know someone who is familiar with the Linux device driver "da8xx-fb"?

Best regards,
Jan-Marc Stranz.