2018-08-17 17:35:04

by Ayan Halder

Subject: [PATCH] drm: Use horizontal and vertical chroma subsampling factor while calculating offsets in the physical address of framebuffer

For multi-planar formats, while calculating offsets in planes with index greater than 0
(ie second plane, third plane, etc), one needs to divide (src_x * cpp) with horizontal
chroma subsampling factor and (src_y * pitch) with vertical chroma subsampling factor.

The reason being that the planes contain subsampled (ie reduced) data (by a factor of 2) and thus the
while calculating the byte position coresponding to the x and y co-ordinates, one needs to be
divide it with the sampling factor.

Signed-off-by: Ayan Kumar halder <[email protected]>
---
drivers/gpu/drm/drm_fb_cma_helper.c | 11 +++++++++--
1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/drm_fb_cma_helper.c b/drivers/gpu/drm/drm_fb_cma_helper.c
index b127061..47e0e2f 100644
--- a/drivers/gpu/drm/drm_fb_cma_helper.c
+++ b/drivers/gpu/drm/drm_fb_cma_helper.c
@@ -86,14 +86,21 @@ dma_addr_t drm_fb_cma_get_gem_addr(struct drm_framebuffer *fb,
{
struct drm_gem_cma_object *obj;
dma_addr_t paddr;
+ u8 h_div = 1, v_div = 1;

obj = drm_fb_cma_get_gem_obj(fb, plane);
if (!obj)
return 0;

paddr = obj->paddr + fb->offsets[plane];
- paddr += fb->format->cpp[plane] * (state->src_x >> 16);
- paddr += fb->pitches[plane] * (state->src_y >> 16);
+
+ if (plane > 0) {
+ h_div = fb->format->hsub;
+ v_div = fb->format->vsub;
+ }
+
+ paddr += (fb->format->cpp[plane] * (state->src_x >> 16)) / h_div;
+ paddr += (fb->pitches[plane] * (state->src_y >> 16)) / v_div;

return paddr;
}
--
2.7.4

2018-08-20 11:05:32

by Liviu Dudau

Subject: Re: [PATCH] drm: Use horizontal and vertical chroma subsampling factor while calculating offsets in the physical address of framebuffer

On Fri, Aug 17, 2018 at 06:33:04PM +0100, Ayan Kumar Halder wrote:
> For multi-planar formats, while calculating offsets in planes with index greater than 0
> (ie second plane, third plane, etc), one needs to divide (src_x * cpp) with horizontal
> chroma subsampling factor and (src_y * pitch) with vertical chroma subsampling factor.
>
> The reason being that the planes contain subsampled (ie reduced) data (by a factor of 2) and thus the

drop the extraneous "the" at the end of the line.

> while calculating the byte position coresponding to the x and y co-ordinates, one needs to be

and drop the extraneous "be" at the end of this line.

> divide it with the sampling factor.
>
> Signed-off-by: Ayan Kumar halder <[email protected]>

Otherwise, it looks good to me!

Reviewed-by: Liviu Dudau <[email protected]>

Best regards,
Liviu

> ---
> drivers/gpu/drm/drm_fb_cma_helper.c | 11 +++++++++--
> 1 file changed, 9 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/gpu/drm/drm_fb_cma_helper.c b/drivers/gpu/drm/drm_fb_cma_helper.c
> index b127061..47e0e2f 100644
> --- a/drivers/gpu/drm/drm_fb_cma_helper.c
> +++ b/drivers/gpu/drm/drm_fb_cma_helper.c
> @@ -86,14 +86,21 @@ dma_addr_t drm_fb_cma_get_gem_addr(struct drm_framebuffer *fb,
> {
> struct drm_gem_cma_object *obj;
> dma_addr_t paddr;
> + u8 h_div = 1, v_div = 1;
>
> obj = drm_fb_cma_get_gem_obj(fb, plane);
> if (!obj)
> return 0;
>
> paddr = obj->paddr + fb->offsets[plane];
> - paddr += fb->format->cpp[plane] * (state->src_x >> 16);
> - paddr += fb->pitches[plane] * (state->src_y >> 16);
> +
> + if (plane > 0) {
> + h_div = fb->format->hsub;
> + v_div = fb->format->vsub;
> + }
> +
> + paddr += (fb->format->cpp[plane] * (state->src_x >> 16)) / h_div;
> + paddr += (fb->pitches[plane] * (state->src_y >> 16)) / v_div;
>
> return paddr;
> }
> --
> 2.7.4
>

--
====================
| I would like to |
| fix the world, |
| but they're not |
| giving me the |
\ source code! /
---------------
¯\_(ツ)_/¯

2018-08-20 12:41:16

by Ayan Halder

Subject: Re: [PATCH] drm: Use horizontal and vertical chroma subsampling factor while calculating offsets in the physical address of framebuffer

On Mon, Aug 20, 2018 at 12:03:19PM +0100, Liviu Dudau wrote:
> On Fri, Aug 17, 2018 at 06:33:04PM +0100, Ayan Kumar Halder wrote:
> > For multi-planar formats, while calculating offsets in planes with index greater than 0
> > (ie second plane, third plane, etc), one needs to divide (src_x * cpp) with horizontal
> > chroma subsampling factor and (src_y * pitch) with vertical chroma subsampling factor.
> >
> > The reason being that the planes contain subsampled (ie reduced) data (by a factor of 2) and thus the
>
> drop the extraneous "the" at the end of the line.
>
> > while calculating the byte position coresponding to the x and y co-ordinates, one needs to be
>
> and drop the extraneous "be" at the end of this line.
>
> > divide it with the sampling factor.
> >
> > Signed-off-by: Ayan Kumar halder <[email protected]>
>
> Otherwise, it looks good to me!
>
> Reviewed-by: Liviu Dudau <[email protected]>
>
Thanks Liviu. I will update the commit message and push this in
drm-misc-next.

> Best regards,
> Liviu
>
> > ---
> > drivers/gpu/drm/drm_fb_cma_helper.c | 11 +++++++++--
> > 1 file changed, 9 insertions(+), 2 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/drm_fb_cma_helper.c b/drivers/gpu/drm/drm_fb_cma_helper.c
> > index b127061..47e0e2f 100644
> > --- a/drivers/gpu/drm/drm_fb_cma_helper.c
> > +++ b/drivers/gpu/drm/drm_fb_cma_helper.c
> > @@ -86,14 +86,21 @@ dma_addr_t drm_fb_cma_get_gem_addr(struct drm_framebuffer *fb,
> > {
> > struct drm_gem_cma_object *obj;
> > dma_addr_t paddr;
> > + u8 h_div = 1, v_div = 1;
> >
> > obj = drm_fb_cma_get_gem_obj(fb, plane);
> > if (!obj)
> > return 0;
> >
> > paddr = obj->paddr + fb->offsets[plane];
> > - paddr += fb->format->cpp[plane] * (state->src_x >> 16);
> > - paddr += fb->pitches[plane] * (state->src_y >> 16);
> > +
> > + if (plane > 0) {
> > + h_div = fb->format->hsub;
> > + v_div = fb->format->vsub;
> > + }
> > +
> > + paddr += (fb->format->cpp[plane] * (state->src_x >> 16)) / h_div;
> > + paddr += (fb->pitches[plane] * (state->src_y >> 16)) / v_div;
> >
> > return paddr;
> > }
> > --
> > 2.7.4
> >
>
> --
> ====================
> | I would like to |
> | fix the world, |
> | but they're not |
> | giving me the |
> \ source code! /
> ---------------
> ??\_(???)_/??
> _______________________________________________
> dri-devel mailing list
> [email protected]
> https://lists.freedesktop.org/mailman/listinfo/dri-devel

2018-08-20 14:37:14

by Liviu Dudau

Subject: Re: [PATCH] drm: Use horizontal and vertical chroma subsampling factor while calculating offsets in the physical address of framebuffer

On Mon, Aug 20, 2018 at 01:38:59PM +0100, Ayan Halder wrote:
> On Mon, Aug 20, 2018 at 12:03:19PM +0100, Liviu Dudau wrote:
> > On Fri, Aug 17, 2018 at 06:33:04PM +0100, Ayan Kumar Halder wrote:
> > > For multi-planar formats, while calculating offsets in planes with index greater than 0
> > > (ie second plane, third plane, etc), one needs to divide (src_x * cpp) with horizontal
> > > chroma subsampling factor and (src_y * pitch) with vertical chroma subsampling factor.
> > >
> > > The reason being that the planes contain subsampled (ie reduced) data (by a factor of 2) and thus the
> >
> > drop the extraneous "the" at the end of the line.
> >
> > > while calculating the byte position coresponding to the x and y co-ordinates, one needs to be
> >
> > and drop the extraneous "be" at the end of this line.
> >
> > > divide it with the sampling factor.
> > >
> > > Signed-off-by: Ayan Kumar halder <[email protected]>
> >
> > Otherwise, it looks good to me!
> >
> > Reviewed-by: Liviu Dudau <[email protected]>
> >
> Thanks Liviu. I will update the commit message and push this in
> drm-misc-next.

Worth waiting for another review before pushing.

Best regards,
Liviu

>
> > Best regards,
> > Liviu
> >
> > > ---
> > > drivers/gpu/drm/drm_fb_cma_helper.c | 11 +++++++++--
> > > 1 file changed, 9 insertions(+), 2 deletions(-)
> > >
> > > diff --git a/drivers/gpu/drm/drm_fb_cma_helper.c b/drivers/gpu/drm/drm_fb_cma_helper.c
> > > index b127061..47e0e2f 100644
> > > --- a/drivers/gpu/drm/drm_fb_cma_helper.c
> > > +++ b/drivers/gpu/drm/drm_fb_cma_helper.c
> > > @@ -86,14 +86,21 @@ dma_addr_t drm_fb_cma_get_gem_addr(struct drm_framebuffer *fb,
> > > {
> > > struct drm_gem_cma_object *obj;
> > > dma_addr_t paddr;
> > > + u8 h_div = 1, v_div = 1;
> > >
> > > obj = drm_fb_cma_get_gem_obj(fb, plane);
> > > if (!obj)
> > > return 0;
> > >
> > > paddr = obj->paddr + fb->offsets[plane];
> > > - paddr += fb->format->cpp[plane] * (state->src_x >> 16);
> > > - paddr += fb->pitches[plane] * (state->src_y >> 16);
> > > +
> > > + if (plane > 0) {
> > > + h_div = fb->format->hsub;
> > > + v_div = fb->format->vsub;
> > > + }
> > > +
> > > + paddr += (fb->format->cpp[plane] * (state->src_x >> 16)) / h_div;
> > > + paddr += (fb->pitches[plane] * (state->src_y >> 16)) / v_div;
> > >
> > > return paddr;
> > > }
> > > --
> > > 2.7.4
> > >

--
====================
| I would like to |
| fix the world, |
| but they're not |
| giving me the |
\ source code! /
---------------
¯\_(ツ)_/¯