diff mbox

[v5,0/6] Enable NV12 support

Message ID 1dfe8d48-7b9a-64a0-ffda-856516cdfff9@linux.intel.com (mailing list archive)
State New, archived
Headers show

Commit Message

Maarten Lankhorst May 2, 2018, 8:58 a.m. UTC 
Hey,

Op 02-05-18 om 04:43 schreef Srinivas, Vidya:
>
>> -----Original Message-----
>> From: Kristian Høgsberg [mailto:hoegsberg@gmail.com]
>> Sent: Monday, April 30, 2018 11:58 PM
>> To: Srinivas, Vidya <vidya.srinivas@intel.com>
>> Cc: intel-gfx@lists.freedesktop.org
>> Subject: Re: [Intel-gfx] [PATCH v5 0/6] Enable NV12 support
>>
>> On Thu, Apr 19, 2018 at 3:34 AM Vidya Srinivas <vidya.srinivas@intel.com>
>> wrote:
>>
>>> Enabling NV12 support:
>>> - Framebuffer creation
>>> - Primary and Sprite plane support
>>> Patch series depend on Enable display workaround 827 patch mentioned
>>> below submitted by Maarten
>> How are we looking on being able to land these last NV12 patches?
> I think Maarten is working on this series related https://patchwork.freedesktop.org/series/42476/
> I guess all would go in together?

I did some testing and with just the nv12 series and my series I don't hit underruns.
I made the following changes to your series on top of mine:

skl_update_scaler check removed, was done in patch 6 instead.

Comments

Vidya Srinivas May 2, 2018, 10:16 a.m. UTC | #1 
> -----Original Message-----

> From: Maarten Lankhorst [mailto:maarten.lankhorst@linux.intel.com]

> Sent: Wednesday, May 2, 2018 2:29 PM

> To: Srinivas, Vidya <vidya.srinivas@intel.com>; Kristian Høgsberg

> <hoegsberg@gmail.com>

> Cc: intel-gfx@lists.freedesktop.org

> Subject: Re: [Intel-gfx] [PATCH v5 0/6] Enable NV12 support

> 

> Hey,

> 

> Op 02-05-18 om 04:43 schreef Srinivas, Vidya:

> >

> >> -----Original Message-----

> >> From: Kristian Høgsberg [mailto:hoegsberg@gmail.com]

> >> Sent: Monday, April 30, 2018 11:58 PM

> >> To: Srinivas, Vidya <vidya.srinivas@intel.com>

> >> Cc: intel-gfx@lists.freedesktop.org

> >> Subject: Re: [Intel-gfx] [PATCH v5 0/6] Enable NV12 support

> >>

> >> On Thu, Apr 19, 2018 at 3:34 AM Vidya Srinivas

> >> <vidya.srinivas@intel.com>

> >> wrote:

> >>

> >>> Enabling NV12 support:

> >>> - Framebuffer creation

> >>> - Primary and Sprite plane support

> >>> Patch series depend on Enable display workaround 827 patch mentioned

> >>> below submitted by Maarten

> >> How are we looking on being able to land these last NV12 patches?

> > I think Maarten is working on this series related

> > https://patchwork.freedesktop.org/series/42476/

> > I guess all would go in together?

> 

> I did some testing and with just the nv12 series and my series I don't hit

> underruns.


Thank you so much. I applied your series https://patchwork.freedesktop.org/series/42476/
And then applied the NV12 series https://patchwork.freedesktop.org/series/41674/
Somehow, I see that patch 6 doesn’t show the updated contents you have submitted under https://patchwork.freedesktop.org/patch/219915/
Not sure if only I don’t see the change (series show updated may 2nd but rev 3 contents are not reflecting in patch 6)

But, I manually made the changes you mentioned and I too did not see any underruns on GLK.
Couldn’t run the set on APL as we now have BXT removed.

Regards
Vidya

> I made the following changes to your series on top of mine:

> 

> skl_update_scaler check removed, was done in patch 6 instead.

> diff --git a/drivers/gpu/drm/i915/intel_display.c

> b/drivers/gpu/drm/i915/intel_display.c

> index ad13c797e946..0a651f26f827 100644

> --- a/drivers/gpu/drm/i915/intel_display.c

> +++ b/drivers/gpu/drm/i915/intel_display.c

> @@ -4861,8 +4861,7 @@ skl_update_scaler(struct intel_crtc_state

> *crtc_state, bool force_detach,

>  	}

> 

>  	if (plane_scaler_check && pixel_format == DRM_FORMAT_NV12 &&

> -	    (src_h < SKL_MIN_YUV_420_SRC_H || (src_w % 4) != 0 ||

> -	     (src_h % 4) != 0)) {

> +	    (src_h < SKL_MIN_YUV_420_SRC_H || src_w <

> SKL_MIN_YUV_420_SRC_W))

> +{

>  		DRM_DEBUG_KMS("NV12: src dimensions not met\n");

>  		return -EINVAL;

>  	}

> 

> And fixup sprite format.

> diff --git a/drivers/gpu/drm/i915/intel_sprite.c

> b/drivers/gpu/drm/i915/intel_sprite.c

> index dcba946af2ff..573ef6932633 100644

> --- a/drivers/gpu/drm/i915/intel_sprite.c

> +++ b/drivers/gpu/drm/i915/intel_sprite.c

> @@ -1011,6 +1011,7 @@ intel_check_sprite_plane(struct intel_plane

> *plane,

>  		src->y2 = (src_y + src_h) << 16;

> 

>  		if (intel_format_is_yuv(fb->format->format) &&

> +		    fb->format->format != DRM_FORMAT_NV12 &&

>  		    (src_x % 2 || src_w % 2)) {

>  			DRM_DEBUG_KMS("src x/w (%u, %u) must be a

> multiple of 2 for YUV planes\n",

>  				      src_x, src_w);
Vidya Srinivas May 3, 2018, 8:58 a.m. UTC | #2 
> -----Original Message-----

> From: Srinivas, Vidya

> Sent: Wednesday, May 2, 2018 3:47 PM

> To: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>; Kristian

> Høgsberg <hoegsberg@gmail.com>

> Cc: intel-gfx@lists.freedesktop.org

> Subject: RE: [Intel-gfx] [PATCH v5 0/6] Enable NV12 support

> 

> 

> 

> > -----Original Message-----

> > From: Maarten Lankhorst [mailto:maarten.lankhorst@linux.intel.com]

> > Sent: Wednesday, May 2, 2018 2:29 PM

> > To: Srinivas, Vidya <vidya.srinivas@intel.com>; Kristian Høgsberg

> > <hoegsberg@gmail.com>

> > Cc: intel-gfx@lists.freedesktop.org

> > Subject: Re: [Intel-gfx] [PATCH v5 0/6] Enable NV12 support

> >

> > Hey,

> >

> > Op 02-05-18 om 04:43 schreef Srinivas, Vidya:

> > >

> > >> -----Original Message-----

> > >> From: Kristian Høgsberg [mailto:hoegsberg@gmail.com]

> > >> Sent: Monday, April 30, 2018 11:58 PM

> > >> To: Srinivas, Vidya <vidya.srinivas@intel.com>

> > >> Cc: intel-gfx@lists.freedesktop.org

> > >> Subject: Re: [Intel-gfx] [PATCH v5 0/6] Enable NV12 support

> > >>

> > >> On Thu, Apr 19, 2018 at 3:34 AM Vidya Srinivas

> > >> <vidya.srinivas@intel.com>

> > >> wrote:

> > >>

> > >>> Enabling NV12 support:

> > >>> - Framebuffer creation

> > >>> - Primary and Sprite plane support Patch series depend on Enable

> > >>> display workaround 827 patch mentioned below submitted by Maarten

> > >> How are we looking on being able to land these last NV12 patches?

> > > I think Maarten is working on this series related

> > > https://patchwork.freedesktop.org/series/42476/

> > > I guess all would go in together?

> >

> > I did some testing and with just the nv12 series and my series I don't

> > hit underruns.

> 

> Thank you so much. I applied your series

> https://patchwork.freedesktop.org/series/42476/

> And then applied the NV12 series

> https://patchwork.freedesktop.org/series/41674/

> Somehow, I see that patch 6 doesn’t show the updated contents you have

> submitted under https://patchwork.freedesktop.org/patch/219915/

> Not sure if only I don’t see the change (series show updated may 2nd but rev

> 3 contents are not reflecting in patch 6)


What I meant to ask was - should I make the changes and update again?
Or could you please help push it from your side? Please let me know.
Thank you. Also, can all these go in now? I mean for merge :)

Regards
Vidya

> 

> But, I manually made the changes you mentioned and I too did not see any

> underruns on GLK.

> Couldn’t run the set on APL as we now have BXT removed.

> 

> Regards

> Vidya

> 

> > I made the following changes to your series on top of mine:

> >

> > skl_update_scaler check removed, was done in patch 6 instead.

> > diff --git a/drivers/gpu/drm/i915/intel_display.c

> > b/drivers/gpu/drm/i915/intel_display.c

> > index ad13c797e946..0a651f26f827 100644

> > --- a/drivers/gpu/drm/i915/intel_display.c

> > +++ b/drivers/gpu/drm/i915/intel_display.c

> > @@ -4861,8 +4861,7 @@ skl_update_scaler(struct intel_crtc_state

> > *crtc_state, bool force_detach,

> >  	}

> >

> >  	if (plane_scaler_check && pixel_format == DRM_FORMAT_NV12 &&

> > -	    (src_h < SKL_MIN_YUV_420_SRC_H || (src_w % 4) != 0 ||

> > -	     (src_h % 4) != 0)) {

> > +	    (src_h < SKL_MIN_YUV_420_SRC_H || src_w <

> > SKL_MIN_YUV_420_SRC_W))

> > +{

> >  		DRM_DEBUG_KMS("NV12: src dimensions not met\n");

> >  		return -EINVAL;

> >  	}

> >

> > And fixup sprite format.

> > diff --git a/drivers/gpu/drm/i915/intel_sprite.c

> > b/drivers/gpu/drm/i915/intel_sprite.c

> > index dcba946af2ff..573ef6932633 100644

> > --- a/drivers/gpu/drm/i915/intel_sprite.c

> > +++ b/drivers/gpu/drm/i915/intel_sprite.c

> > @@ -1011,6 +1011,7 @@ intel_check_sprite_plane(struct intel_plane

> > *plane,

> >  		src->y2 = (src_y + src_h) << 16;

> >

> >  		if (intel_format_is_yuv(fb->format->format) &&

> > +		    fb->format->format != DRM_FORMAT_NV12 &&

> >  		    (src_x % 2 || src_w % 2)) {

> >  			DRM_DEBUG_KMS("src x/w (%u, %u) must be a

> multiple of 2 for YUV

> > planes\n",

> >  				      src_x, src_w);
Maarten Lankhorst May 4, 2018, 1:15 p.m. UTC | #3 
Op 03-05-18 om 10:58 schreef Srinivas, Vidya:
>
>> -----Original Message-----
>> From: Srinivas, Vidya
>> Sent: Wednesday, May 2, 2018 3:47 PM
>> To: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>; Kristian
>> Høgsberg <hoegsberg@gmail.com>
>> Cc: intel-gfx@lists.freedesktop.org
>> Subject: RE: [Intel-gfx] [PATCH v5 0/6] Enable NV12 support
>>
>>
>>
>>> -----Original Message-----
>>> From: Maarten Lankhorst [mailto:maarten.lankhorst@linux.intel.com]
>>> Sent: Wednesday, May 2, 2018 2:29 PM
>>> To: Srinivas, Vidya <vidya.srinivas@intel.com>; Kristian Høgsberg
>>> <hoegsberg@gmail.com>
>>> Cc: intel-gfx@lists.freedesktop.org
>>> Subject: Re: [Intel-gfx] [PATCH v5 0/6] Enable NV12 support
>>>
>>> Hey,
>>>
>>> Op 02-05-18 om 04:43 schreef Srinivas, Vidya:
>>>>> -----Original Message-----
>>>>> From: Kristian Høgsberg [mailto:hoegsberg@gmail.com]
>>>>> Sent: Monday, April 30, 2018 11:58 PM
>>>>> To: Srinivas, Vidya <vidya.srinivas@intel.com>
>>>>> Cc: intel-gfx@lists.freedesktop.org
>>>>> Subject: Re: [Intel-gfx] [PATCH v5 0/6] Enable NV12 support
>>>>>
>>>>> On Thu, Apr 19, 2018 at 3:34 AM Vidya Srinivas
>>>>> <vidya.srinivas@intel.com>
>>>>> wrote:
>>>>>
>>>>>> Enabling NV12 support:
>>>>>> - Framebuffer creation
>>>>>> - Primary and Sprite plane support Patch series depend on Enable
>>>>>> display workaround 827 patch mentioned below submitted by Maarten
>>>>> How are we looking on being able to land these last NV12 patches?
>>>> I think Maarten is working on this series related
>>>> https://patchwork.freedesktop.org/series/42476/
>>>> I guess all would go in together?
>>> I did some testing and with just the nv12 series and my series I don't
>>> hit underruns.
>> Thank you so much. I applied your series
>> https://patchwork.freedesktop.org/series/42476/
>> And then applied the NV12 series
>> https://patchwork.freedesktop.org/series/41674/
>> Somehow, I see that patch 6 doesn’t show the updated contents you have
>> submitted under https://patchwork.freedesktop.org/patch/219915/
>> Not sure if only I don’t see the change (series show updated may 2nd but rev
>> 3 contents are not reflecting in patch 6)
> What I meant to ask was - should I make the changes and update again?
> Or could you please help push it from your side? Please let me know.
> Thank you. Also, can all these go in now? I mean for merge :)
We want to test again on top of drm-tip, so best just make the changes, and reorder in the order I specified earlier. :)

~Maarten
Vidya Srinivas May 5, 2018, 1:09 p.m. UTC | #4 
> -----Original Message-----

> From: Maarten Lankhorst [mailto:maarten.lankhorst@linux.intel.com]

> Sent: Friday, May 4, 2018 6:46 PM

> To: Srinivas, Vidya <vidya.srinivas@intel.com>; 'Kristian Høgsberg'

> <hoegsberg@gmail.com>

> Cc: 'intel-gfx@lists.freedesktop.org' <intel-gfx@lists.freedesktop.org>

> Subject: Re: [Intel-gfx] [PATCH v5 0/6] Enable NV12 support

> 

> Op 03-05-18 om 10:58 schreef Srinivas, Vidya:

> >

> >> -----Original Message-----

> >> From: Srinivas, Vidya

> >> Sent: Wednesday, May 2, 2018 3:47 PM

> >> To: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>; Kristian

> >> Høgsberg <hoegsberg@gmail.com>

> >> Cc: intel-gfx@lists.freedesktop.org

> >> Subject: RE: [Intel-gfx] [PATCH v5 0/6] Enable NV12 support

> >>

> >>

> >>

> >>> -----Original Message-----

> >>> From: Maarten Lankhorst [mailto:maarten.lankhorst@linux.intel.com]

> >>> Sent: Wednesday, May 2, 2018 2:29 PM

> >>> To: Srinivas, Vidya <vidya.srinivas@intel.com>; Kristian Høgsberg

> >>> <hoegsberg@gmail.com>

> >>> Cc: intel-gfx@lists.freedesktop.org

> >>> Subject: Re: [Intel-gfx] [PATCH v5 0/6] Enable NV12 support

> >>>

> >>> Hey,

> >>>

> >>> Op 02-05-18 om 04:43 schreef Srinivas, Vidya:

> >>>>> -----Original Message-----

> >>>>> From: Kristian Høgsberg [mailto:hoegsberg@gmail.com]

> >>>>> Sent: Monday, April 30, 2018 11:58 PM

> >>>>> To: Srinivas, Vidya <vidya.srinivas@intel.com>

> >>>>> Cc: intel-gfx@lists.freedesktop.org

> >>>>> Subject: Re: [Intel-gfx] [PATCH v5 0/6] Enable NV12 support

> >>>>>

> >>>>> On Thu, Apr 19, 2018 at 3:34 AM Vidya Srinivas

> >>>>> <vidya.srinivas@intel.com>

> >>>>> wrote:

> >>>>>

> >>>>>> Enabling NV12 support:

> >>>>>> - Framebuffer creation

> >>>>>> - Primary and Sprite plane support Patch series depend on Enable

> >>>>>> display workaround 827 patch mentioned below submitted by

> Maarten

> >>>>> How are we looking on being able to land these last NV12 patches?

> >>>> I think Maarten is working on this series related

> >>>> https://patchwork.freedesktop.org/series/42476/

> >>>> I guess all would go in together?

> >>> I did some testing and with just the nv12 series and my series I

> >>> don't hit underruns.

> >> Thank you so much. I applied your series

> >> https://patchwork.freedesktop.org/series/42476/

> >> And then applied the NV12 series

> >> https://patchwork.freedesktop.org/series/41674/

> >> Somehow, I see that patch 6 doesn’t show the updated contents you

> >> have submitted under https://patchwork.freedesktop.org/patch/219915/

> >> Not sure if only I don’t see the change (series show updated may 2nd

> >> but rev

> >> 3 contents are not reflecting in patch 6)

> > What I meant to ask was - should I make the changes and update again?

> > Or could you please help push it from your side? Please let me know.

> > Thank you. Also, can all these go in now? I mean for merge :)

> We want to test again on top of drm-tip, so best just make the changes, and

> reorder in the order I specified earlier. :)

> 

Thank you so much. I have changed the patch and ordering as per your suggestion
vnd floated them. Kindly have a check.

Regards
Vidya

> ~Maarten
diff mbox

Patch

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index ad13c797e946..0a651f26f827 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -4861,8 +4861,7 @@  skl_update_scaler(struct intel_crtc_state *crtc_state, bool force_detach,
 	}
 
 	if (plane_scaler_check && pixel_format == DRM_FORMAT_NV12 &&
-	    (src_h < SKL_MIN_YUV_420_SRC_H || (src_w % 4) != 0 ||
-	     (src_h % 4) != 0)) {
+	    (src_h < SKL_MIN_YUV_420_SRC_H || src_w < SKL_MIN_YUV_420_SRC_W)) {
 		DRM_DEBUG_KMS("NV12: src dimensions not met\n");
 		return -EINVAL;
 	}

And fixup sprite format.
diff --git a/drivers/gpu/drm/i915/intel_sprite.c b/drivers/gpu/drm/i915/intel_sprite.c
index dcba946af2ff..573ef6932633 100644
--- a/drivers/gpu/drm/i915/intel_sprite.c
+++ b/drivers/gpu/drm/i915/intel_sprite.c
@@ -1011,6 +1011,7 @@  intel_check_sprite_plane(struct intel_plane *plane,
 		src->y2 = (src_y + src_h) << 16;
 
 		if (intel_format_is_yuv(fb->format->format) &&
+		    fb->format->format != DRM_FORMAT_NV12 &&
 		    (src_x % 2 || src_w % 2)) {
 			DRM_DEBUG_KMS("src x/w (%u, %u) must be a multiple of 2 for YUV planes\n",
 				      src_x, src_w);