On Wed, Sep 01, 2010 at 14:20:00, Caglar Akyuz wrote:
> On Wednesday 01 September 2010 09:30:53 pm Ambrose, Martin wrote:
> > On Wed, Sep 01, 2010 at 11:57:32, Caglar Akyuz wrote:
[snip]
> > If I am interpreting the above correctly then "*** 8" corresponds to
> > LCD_STAT.
>
> Correct.
>
> > In the working case bit 6 is set indicating the palette has been
> > loaded whereas this is not set in the failure case.
> >
> > This was definitely an area of change in the double buffer patch. Prior the
> > palette was loaded explicitly in the fb_probe function. In the double
> > buffer case this is loaded by the setcolreg function. Normally this is
> > called during kernel boot (I think) when the display is configured for the
> > boot logo.
> >
> > If it is not then I believe you can force a call through a user space ioctl
> > but I'm not sure what that is atm.
> >
> > Now if could be that the palette is being loaded but then there is an error
> > afterwards. Can you add a debug print to the lcd_blit function to see if
> > it is being called with LOAD_PALETTE?
> >
>
> It seems like there is a race in the driver. Attached patch makes the
> problem go away.
Looks like just adding the printf changed which side of the race condition you
are on.
Correct?
> Any ideas for a proper fix?
Sorry my breadth of knowledge here is pretty shallow. This patch was my first
exposure to the framebuffer architecture.
I think I would need to be able to recreate the problem to better analyze.
And, unfortunately, don't have a Hawkboard myself. If you or Roger
have a spare board I can take a more in-depth look but likely not until
beginning of Oct due to prior commits.
Regards,
Martin
> Thanks for the help,
> Caglar
>
> ---
>
> diff --git a/drivers/video/da8xx-fb.c b/drivers/video/da8xx-fb.c
> index 309269f..ffc2f43 100644
> --- a/drivers/video/da8xx-fb.c
> +++ b/drivers/video/da8xx-fb.c
> @@ -246,6 +246,7 @@ static void lcd_blit(int load_mode, struct da8xx_fb_par
> *par)
> lcdc_write(start, LCD_DMA_FRM_BUF_BASE_ADDR_1_REG);
> lcdc_write(end, LCD_DMA_FRM_BUF_CEILING_ADDR_1_REG);
> } else if (load_mode == LOAD_PALETTE) {
> + printk("*** %s: LOAD_PALETTE ***\n", __func__);
> start = par->p_palette_base;
> end = start + par->palette_sz - 1;
>
>
> > Regards,
> > Martin
> >
Regards,
Martin
_______________________________________________
Davinci-linux-open-source mailing list
[email protected]
http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source
