On Mon, Oct 29, 2018 at 07:56:27PM +0530, Jagan Teki wrote:
> On Mon, Oct 29, 2018 at 2:52 PM Maxime Ripard <maxime.ripard@bootlin.com> wrote:
> >
> > On Fri, Oct 26, 2018 at 08:13:31PM +0530, Jagan Teki wrote:
> > > Add 10 bytes packet overhead for hblk where blank is set using
> > > a blanking packet like (4 bytes + 4 bytes + payload + 2 bytes)
> > >
> > > This is according to BSP code from BPI-M64-bsp
> > > (in drivers/video/sunxi/disp2/disp/de/lowlevel_sun50iw1/de_dsi.c)
> > > dsi_hblk = (ht-hspw)*dsi_pixel_bits[format]/8-(4+4+2);
> > >
> > > So, add 10 bytes packet overhead for DSI hblk.
> > >
> > > Tested on 2-lane, 4-lane MIPI-DSI LCD panels.
> > >
> > > Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
> > > Tested-by: Jagan Teki <jagan@amarulasolutions.com>
> > > ---
> > > Changes for v3:
> > > - new patch
> > > Changes for v2:
> > > - none
> > >
> > >  drivers/gpu/drm/sun4i/sun6i_mipi_dsi.c | 9 +++++++--
> > >  1 file changed, 7 insertions(+), 2 deletions(-)
> > >
> > > diff --git a/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.c b/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.c
> > > index 596e560263bf..cf42be1f1ba1 100644
> > > --- a/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.c
> > > +++ b/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.c
> > > @@ -461,7 +461,7 @@ static void sun6i_dsi_setup_timings(struct sun6i_dsi *dsi,
> > >  {
> > >       struct mipi_dsi_device *device = dsi->device;
> > >       unsigned int Bpp = mipi_dsi_pixel_format_to_bpp(device->format) / 8;
> > > -     u16 hbp, hfp, hsa, hblk, vblk;
> > > +     u16 hbp, hfp, hsa, hblk_max, hblk, vblk;
> > >       size_t bytes;
> > >       u8 *buffer;
> > >
> > > @@ -494,8 +494,13 @@ static void sun6i_dsi_setup_timings(struct sun6i_dsi *dsi,
> > >
> > >       /*
> > >        * hblk seems to be the line + porches length.
> > > +      * The blank is set using a blanking packet (4 bytes + 4 bytes +
> > > +      * payload + 2 bytes). So minimal size is 10 bytes
> > >        */
> > > -     hblk = (mode->htotal - (mode->hsync_end - mode->hsync_start)) * Bpp;
> > > +#define HBLK_PACKET_OVERHEAD 10
> > > +     hblk_max = (mode->htotal - (mode->hsync_end - mode->hsync_start)) * Bpp;
> > > +     hblk_max -= HBLK_PACKET_OVERHEAD;
> > > +     hblk = max((unsigned int)HBLK_PACKET_OVERHEAD, hblk_max);
> >
> > I'd rather use the same convention than the other assignments done
> > before in that function.
> 
> Yes, old code did use hsa. but it's added 10 bytes packet head instead
> of subtracting it.
> 
> hblk = mode->htotal * Bpp - hsa;
> => mode->htotal * Bpp - (mode->hsync_end - mode->hsync_start) * Bpp -
> HSA_PACKET_OVERHEAD);
> => (mode->htotal - (mode->hsync_end - mode->hsync_start)) * Bpp +
> HSA_PACKET_OVERHEAD;
> 
> And it should be
> (mode->htotal - (mode->hsync_end - mode->hsync_start)) * Bpp -
> HSA_PACKET_OVERHEAD;
> 
> This patch is simply doing the same by explicitly adding packet over
> ahead macro, which again used in hfp.

That's not my point.

The rest of the driver uses a construct that would be:

hblk = max((unsigned int)HBLK_PACKET_OVERHEAD,
           (mode->htotal - (mode->hsync_end - mode->hsync_start)) * Bpp - HBLK_PACKET_OVERHEAD);

We want to remain consistent.

Maxime

-- 
Maxime Ripard, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com