OMAPFB problem on 3430SDP

OMAPFB problem on 3430SDP

[Tomi Valkeinen]

Hi,

When running linux-omap kernel on 3430SDP board (version VG5.0.0), the
picture on the LCD is very green. It seems to show only green color
component, or something like that. TI's own kernel works ok.

I've been going through all the DSS registers with Trace32, comparing
them between working and non-working kernels, but I haven't been able to
find differences.

Does linux-omap kernel work ok on earlier SDP versions? This one has
750-2081-102(B) as a display board, (I think the earlier ones have 101).
Any ideas what could be missing?

 Tomi

Re: OMAPFB problem on 3430SDP

[iqbal]

Tomi,

> When running linux-omap kernel on 3430SDP board (version VG5.0.0), the
> picture on the LCD is very green. It seems to show only green color
> component, or something like that. TI's own kernel works ok.
>
> I've been going through all the DSS registers with Trace32, comparing
> them between working and non-working kernels, but I haven't been able to
> find differences.
>
> Does linux-omap kernel work ok on earlier SDP versions? This one has
> 750-2081-102(B) as a display board, (I think the earlier ones have 101).
> Any ideas what could be missing?
>
This problem is seen on the older SDP versions too, all have 750-2081-102(B)
display. Something to do with linux-omap code, I am sure TI's 2.6.24 code
has a fix.

Regards
Iqbal
>
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-omap" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>

omapfb problem on 3430sdp

[stanley.miao]

Hardware: 
	Mainboard: 750-2094-001(C)
	Display board: 750-2081-101(B)

Software: 
	the lastest kernel from linux-omap-2.6 and from omapzoom.org.

The linux bootup logo is green, and LCD is dark. It looks like backlight
disabled. 

I commented the following two sentences in lcd_2430.c, there is still
the same thing. It seems these two sentences doesn't work.

 96         omap_set_gpio_dataout(enable_gpio, 1);
 97         omap_set_gpio_dataout(backlight_gpio, 1);

Both the the latest kernels from omapzoom.org and linux-omap-2.6 have
this problem.  However, the old version in linux-omap-2.6, 2.6.21,
doesn't have this problem.

Stanley.

Re: omapfb problem on 3430sdp

[Felipe Balbi]

On Wed, Aug 27, 2008 at 05:28:36PM +0800, ext stanley.miao wrote:
> Hardware: 
> 	Mainboard: 750-2094-001(C)
> 	Display board: 750-2081-101(B)
> 
> Software: 
> 	the lastest kernel from linux-omap-2.6 and from omapzoom.org.
> 
> The linux bootup logo is green, and LCD is dark. It looks like backlight
> disabled. 
> 
> I commented the following two sentences in lcd_2430.c, there is still
> the same thing. It seems these two sentences doesn't work.
> 
>  96         omap_set_gpio_dataout(enable_gpio, 1);
>  97         omap_set_gpio_dataout(backlight_gpio, 1);
> 
> Both the the latest kernels from omapzoom.org and linux-omap-2.6 have
> this problem.  However, the old version in linux-omap-2.6, 2.6.21,
> doesn't have this problem.

Cool could you bisect and try to find the commit that broke it ?

$ git bisect start
$ git bisect good v2.6.21-omap1
$ git bisect bad

Should help you. Read git help bisect for more information. When you
find the exact commit that broke lcd_2430.c, report here with the commit
id.

Just a note, if you can't build the bisect point, reset to some points
before and try again until you find a commit that builds, do not mark
them as bad.

-- 
balbi

Re: omapfb problem on 3430sdp

[stanley.miao]

I updated the kernel again. Now the kernel from omapzoom.org is OK, but
the kernel from "source.mvista.com/git/linux-omap-2.6.git" still can't
work.

The LCD drivers in omapzoom.org and in linux-omap-2.6 use different
files.  the LCD driver in omapzoom.org is omap_disp_out.c and omap_fb.c
while the driver in linux-omap-2.6 is lcd_2430sdp.c and omapfb_main.c.

The following MACRO are defined differently. Can anyone tell me why ?

In the file omap_disp_out.c in omapzoom.org :

 113 #define LCD_PANEL_ENABLE_GPIO           5
 114 #define LCD_PANEL_BACKLIGHT_GPIO        8

In the file lcd_2430sdp.c in linux-omap-2.6:

 35 #define SDP3430_LCD_PANEL_BACKLIGHT_GPIO        24
 36 #define SDP3430_LCD_PANEL_ENABLE_GPIO           28


Why are they defined differently?

Thanks
Stanley.


On Wed, 2008-08-27 at 12:43 +0300, Felipe Balbi wrote:
> On Wed, Aug 27, 2008 at 05:28:36PM +0800, ext stanley.miao wrote:
> > Hardware: 
> > 	Mainboard: 750-2094-001(C)
> > 	Display board: 750-2081-101(B)
> > 
> > Software: 
> > 	the lastest kernel from linux-omap-2.6 and from omapzoom.org.
> > 
> > The linux bootup logo is green, and LCD is dark. It looks like backlight
> > disabled. 
> > 
> > I commented the following two sentences in lcd_2430.c, there is still
> > the same thing. It seems these two sentences doesn't work.
> > 
> >  96         omap_set_gpio_dataout(enable_gpio, 1);
> >  97         omap_set_gpio_dataout(backlight_gpio, 1);
> > 
> > Both the the latest kernels from omapzoom.org and linux-omap-2.6 have
> > this problem.  However, the old version in linux-omap-2.6, 2.6.21,
> > doesn't have this problem.
> 
> Cool could you bisect and try to find the commit that broke it ?
> 
> $ git bisect start
> $ git bisect good v2.6.21-omap1
> $ git bisect bad
> 
> Should help you. Read git help bisect for more information. When you
> find the exact commit that broke lcd_2430.c, report here with the commit
> id.
> 
> Just a note, if you can't build the bisect point, reset to some points
> before and try again until you find a commit that builds, do not mark
> them as bad.
> 

Re: omapfb problem on 3430sdp

[Felipe Balbi]

On Wed, Aug 27, 2008 at 09:03:47PM +0800, ext stanley.miao wrote:
> I updated the kernel again. Now the kernel from omapzoom.org is OK, but
> the kernel from "source.mvista.com/git/linux-omap-2.6.git" still can't
> work.
> 
> The LCD drivers in omapzoom.org and in linux-omap-2.6 use different
> files.  the LCD driver in omapzoom.org is omap_disp_out.c and omap_fb.c
> while the driver in linux-omap-2.6 is lcd_2430sdp.c and omapfb_main.c.
> 
> The following MACRO are defined differently. Can anyone tell me why ?
> 
> In the file omap_disp_out.c in omapzoom.org :
> 
>  113 #define LCD_PANEL_ENABLE_GPIO           5
>  114 #define LCD_PANEL_BACKLIGHT_GPIO        8
> 
> In the file lcd_2430sdp.c in linux-omap-2.6:
> 
>  35 #define SDP3430_LCD_PANEL_BACKLIGHT_GPIO        24
>  36 #define SDP3430_LCD_PANEL_ENABLE_GPIO           28

different board revisions, maybe ?

I think mvista.com will have support for es2.0, although it would be
nice to check the revision and change the gpio numbers based on that.

-- 
balbi

Re: omapfb problem on 3430sdp

[iqbal]

Stanley,

>> In the file omap_disp_out.c in omapzoom.org :
>>
>>  113 #define LCD_PANEL_ENABLE_GPIO           5
>>  114 #define LCD_PANEL_BACKLIGHT_GPIO        8
>>
>> In the file lcd_2430sdp.c in linux-omap-2.6:
>>
>>  35 #define SDP3430_LCD_PANEL_BACKLIGHT_GPIO        24
>>  36 #define SDP3430_LCD_PANEL_ENABLE_GPIO           28

Both 5-8 and 24-28 GPIO combinations work fine on 3430sdp ES 2.0.
According to TRM 5-8 are the right ones for ES 2.0. One of my patches on
linux-omap had these changes but it didn't through.

>
> different board revisions, maybe ?
>
> I think mvista.com will have support for es2.0, although it would be
> nice to check the revision and change the gpio numbers based on that.
>
> --
> balbi
> --
> To unsubscribe from this list: send the line "unsubscribe linux-omap" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>
>

Re: omapfb problem on 3430sdp

[iqbal]

> The linux bootup logo is green, and LCD is dark. It looks like backlight
> disabled.
>
> I commented the following two sentences in lcd_2430.c, there is still
> the same thing. It seems these two sentences doesn't work.
>
>  96         omap_set_gpio_dataout(enable_gpio, 1);
>  97         omap_set_gpio_dataout(backlight_gpio, 1);
>
> Both the the latest kernels from omapzoom.org and linux-omap-2.6 have
> this problem.  However, the old version in linux-omap-2.6, 2.6.21,
> doesn't have this problem.
>
> Stanley.

If I use this piece of code in panel suspend and resume, I see that the panel
backlight is going off on suspend and comes back while calling resume. Give a
try .

 static int sdp2430_panel_suspend(struct platform_device *pdev, pm_message_t mesg)
 {
+       omap_set_gpio_dataout(backlight_gpio, 0);
+       twl4030_i2c_write_u8(TWL4030_MODULE_PM_RECEIVER,0x0,
+                               TWL4030_VPLL2_DEDICATED);
+       twl4030_i2c_write_u8(TWL4030_MODULE_PM_RECEIVER,0x0,
+                               TWL4030_VPLL2_DEV_GRP);
+       mdelay(4);
        return 0;
 }

 static int sdp2430_panel_resume(struct platform_device *pdev)
 {
+       if (lcd_in_use)
+       return 0;
+       twl4030_i2c_write_u8(TWL4030_MODULE_PM_RECEIVER,
+                       ENABLE_VPLL2_DEDICATED,TWL4030_VPLL2_DEDICATED);
+       twl4030_i2c_write_u8(TWL4030_MODULE_PM_RECEIVER,
+                       ENABLE_VPLL2_DEV_GRP,TWL4030_VPLL2_DEV_GRP);
+       omap_set_gpio_dataout(backlight_gpio, 1);
+
        return 0;
 }

>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-omap" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>
>

Re: omapfb problem on 3430sdp

[stanley.miao]

I have tried. It is not backlight's problem.

I found that only green color was showed, red and blue were filtered. So
the picture looked green and dark.

The color mode is RGB565. I wrote 0x07e0, 0xffe0, and 0xffff to /dev/fb0
respectively. All of them are showed to the same color, it's green.

I am using the latest code from source.mvista.com on omap3430sdp. Does
anybody encounter this problem ?

Thanks
Stanley.


On Wed, 2008-08-27 at 19:57 +0530, iqbal wrote:
> > The linux bootup logo is green, and LCD is dark. It looks like backlight
> > disabled.
> >
> > I commented the following two sentences in lcd_2430.c, there is still
> > the same thing. It seems these two sentences doesn't work.
> >
> >  96         omap_set_gpio_dataout(enable_gpio, 1);
> >  97         omap_set_gpio_dataout(backlight_gpio, 1);
> >
> > Both the the latest kernels from omapzoom.org and linux-omap-2.6 have
> > this problem.  However, the old version in linux-omap-2.6, 2.6.21,
> > doesn't have this problem.
> >
> > Stanley.
> 
> If I use this piece of code in panel suspend and resume, I see that the panel
> backlight is going off on suspend and comes back while calling resume. Give a
> try .
> 
>  static int sdp2430_panel_suspend(struct platform_device *pdev, pm_message_t mesg)
>  {
> +       omap_set_gpio_dataout(backlight_gpio, 0);
> +       twl4030_i2c_write_u8(TWL4030_MODULE_PM_RECEIVER,0x0,
> +                               TWL4030_VPLL2_DEDICATED);
> +       twl4030_i2c_write_u8(TWL4030_MODULE_PM_RECEIVER,0x0,
> +                               TWL4030_VPLL2_DEV_GRP);
> +       mdelay(4);
>         return 0;
>  }
> 
>  static int sdp2430_panel_resume(struct platform_device *pdev)
>  {
> +       if (lcd_in_use)
> +       return 0;
> +       twl4030_i2c_write_u8(TWL4030_MODULE_PM_RECEIVER,
> +                       ENABLE_VPLL2_DEDICATED,TWL4030_VPLL2_DEDICATED);
> +       twl4030_i2c_write_u8(TWL4030_MODULE_PM_RECEIVER,
> +                       ENABLE_VPLL2_DEV_GRP,TWL4030_VPLL2_DEV_GRP);
> +       omap_set_gpio_dataout(backlight_gpio, 1);
> +
>         return 0;
>  }
> 
> >
> > --
> > To unsubscribe from this list: send the line "unsubscribe linux-omap" in
> > the body of a message to majordomo@vger.kernel.org
> > More majordomo info at  http://vger.kernel.org/majordomo-info.html
> >
> >
> 
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-omap" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: omapfb problem on 3430sdp

[arun c]

On Sun, Aug 31, 2008 at 9:31 PM, stanley.miao
<stanley.miao@windriver.com> wrote:
>
> On Fri, 2008-08-29 at 21:18 +0530, arun c wrote:
>> Hi stanley,
>>
>> On Fri, Aug 29, 2008 at 5:58 PM, stanley.miao
>> <stanley.miao@windriver.com> wrote:
>> >
>> > I have tried. It is not backlight's problem.
>>
>> Try this out
>>
>> Boot with the working kernel after that soft reset the board and try
>> the non working image(git) and see what happens.
>>
>> If no success
>
> If yes, what should I do ? You seems to encounter this problem ever.
>
> After soft reset with the working kernel, the non-working kernel will
> work. It look like the initial process is incomplete.

Can you check DSS DATA pin voltages for both the kernel?

Procedure:
Write various data (white, red ,green ,blue etc.) and verify
the pin voltages.

>
> Thanks
> Stanley
>
>>
>> Try the below values in lcd_2430sdp.c
>>
>> #define ENABLE_VAUX3_DEDICATED  0x05
>> #define ENABLE_VAUX3_DEV_GRP    0xE0
>>
>> >
>> > I found that only green color was showed, red and blue were filtered. So
>> > the picture looked green and dark.
>> >
>> > The color mode is RGB565. I wrote 0x07e0, 0xffe0, and 0xffff to /dev/fb0
>> > respectively. All of them are showed to the same color, it's green.
>> >
>> > I am using the latest code from source.mvista.com on omap3430sdp. Does
>> > anybody encounter this problem ?
>> >
>> > Thanks
>> > Stanley.
>> >
>> >
>> > On Wed, 2008-08-27 at 19:57 +0530, iqbal wrote:
>> >> > The linux bootup logo is green, and LCD is dark. It looks like backlight
>> >> > disabled.
>> >> >
>> >> > I commented the following two sentences in lcd_2430.c, there is still
>> >> > the same thing. It seems these two sentences doesn't work.
>> >> >
>> >> >  96         omap_set_gpio_dataout(enable_gpio, 1);
>> >> >  97         omap_set_gpio_dataout(backlight_gpio, 1);
>> >> >
>> >> > Both the the latest kernels from omapzoom.org and linux-omap-2.6 have
>> >> > this problem.  However, the old version in linux-omap-2.6, 2.6.21,
>> >> > doesn't have this problem.
>> >> >
>> >> > Stanley.
>> >>
>> >> If I use this piece of code in panel suspend and resume, I see that the panel
>> >> backlight is going off on suspend and comes back while calling resume. Give a
>> >> try .
>> >>
>> >>  static int sdp2430_panel_suspend(struct platform_device *pdev, pm_message_t mesg)
>> >>  {
>> >> +       omap_set_gpio_dataout(backlight_gpio, 0);
>> >> +       twl4030_i2c_write_u8(TWL4030_MODULE_PM_RECEIVER,0x0,
>> >> +                               TWL4030_VPLL2_DEDICATED);
>> >> +       twl4030_i2c_write_u8(TWL4030_MODULE_PM_RECEIVER,0x0,
>> >> +                               TWL4030_VPLL2_DEV_GRP);
>> >> +       mdelay(4);
>> >>         return 0;
>> >>  }
>> >>
>> >>  static int sdp2430_panel_resume(struct platform_device *pdev)
>> >>  {
>> >> +       if (lcd_in_use)
>> >> +       return 0;
>> >> +       twl4030_i2c_write_u8(TWL4030_MODULE_PM_RECEIVER,
>> >> +                       ENABLE_VPLL2_DEDICATED,TWL4030_VPLL2_DEDICATED);
>> >> +       twl4030_i2c_write_u8(TWL4030_MODULE_PM_RECEIVER,
>> >> +                       ENABLE_VPLL2_DEV_GRP,TWL4030_VPLL2_DEV_GRP);
>> >> +       omap_set_gpio_dataout(backlight_gpio, 1);
>> >> +
>> >>         return 0;
>> >>  }
>> >>
>> >> >
>> >> > --
>> >> > To unsubscribe from this list: send the line "unsubscribe linux-omap" in
>> >> > the body of a message to majordomo@vger.kernel.org
>> >> > More majordomo info at  http://vger.kernel.org/majordomo-info.html
>> >> >
>> >> >
>> >>
>> >>
>> >> --
>> >> To unsubscribe from this list: send the line "unsubscribe linux-omap" in
>> >> the body of a message to majordomo@vger.kernel.org
>> >> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>> > --
>> > To unsubscribe from this list: send the line "unsubscribe linux-omap" in
>> > the body of a message to majordomo@vger.kernel.org
>> > More majordomo info at  http://vger.kernel.org/majordomo-info.html
>> >
>