[PATCH 0/2] omap_vout: remove cpu_is_* uses

[PATCH 0/2] omap_vout: remove cpu_is_* uses

[Tomi Valkeinen]

Hi,

This patch removes use of cpu_is_* funcs from omap_vout, and uses omapdss's
version instead. The other patch removes an unneeded plat/dma.h include.

These are based on current omapdss master branch, which has the omapdss version
code. The omapdss version code is queued for v3.8. I'm not sure which is the
best way to handle these patches due to the dependency to omapdss. The easiest
option is to merge these for 3.9.

There's still the OMAP DMA use in omap_vout_vrfb.c, which is the last OMAP
dependency in the omap_vout driver. I'm not going to touch that, as it doesn't
look as trivial as this cpu_is_* removal, and I don't have much knowledge of
the omap_vout driver.

Compiled, but not tested.

 Tomi

Tomi Valkeinen (2):
  [media] omap_vout: use omapdss's version instead of cpu_is_*
  [media] omap_vout: remove extra include

 drivers/media/platform/omap/omap_vout.c    |    4 +--
 drivers/media/platform/omap/omap_voutlib.c |   38 ++++++++++++++++++++--------
 drivers/media/platform/omap/omap_voutlib.h |    3 +++
 3 files changed, 32 insertions(+), 13 deletions(-)

-- 
1.7.10.4

[PATCH 1/2] [media] omap_vout: use omapdss's version instead of cpu_is_*

[Tomi Valkeinen]

cpu_is_* class functions create a dependency to OMAP platform code.
omapdss driver, which omap_vout uses, exposes a function to get the
version of the DSS hardware.

To remove the dependency to OMAP platform code this patch changes
omap_vout to use the omapdss version. For most of the checks, the ones
dealing with DSS differences, this is actually more correct than using
cpu_is_* functions. For the check whether VRFB is available or not this
is not really correct, but still works fine.

Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
---
 drivers/media/platform/omap/omap_vout.c    |    3 +--
 drivers/media/platform/omap/omap_voutlib.c |   38 ++++++++++++++++++++--------
 drivers/media/platform/omap/omap_voutlib.h |    3 +++
 3 files changed, 32 insertions(+), 12 deletions(-)

diff --git a/drivers/media/platform/omap/omap_vout.c b/drivers/media/platform/omap/omap_vout.c
index 3ff94a3..7b1afc8 100644
--- a/drivers/media/platform/omap/omap_vout.c
+++ b/drivers/media/platform/omap/omap_vout.c
@@ -44,7 +44,6 @@
 #include <media/v4l2-device.h>
 #include <media/v4l2-ioctl.h>
 
-#include <plat/cpu.h>
 #include <plat/dma.h>
 #include <video/omapvrfb.h>
 #include <video/omapdss.h>
@@ -2064,7 +2063,7 @@ static int __init omap_vout_create_video_devices(struct platform_device *pdev)
 		vout->vid_info.id = k + 1;
 
 		/* Set VRFB as rotation_type for omap2 and omap3 */
-		if (cpu_is_omap24xx() || cpu_is_omap34xx())
+		if (omap_vout_dss_omap24xx() || omap_vout_dss_omap34xx())
 			vout->vid_info.rotation_type = VOUT_ROT_VRFB;
 
 		/* Setup the default configuration for the video devices
diff --git a/drivers/media/platform/omap/omap_voutlib.c b/drivers/media/platform/omap/omap_voutlib.c
index 115408b..80b0d88 100644
--- a/drivers/media/platform/omap/omap_voutlib.c
+++ b/drivers/media/platform/omap/omap_voutlib.c
@@ -26,7 +26,7 @@
 
 #include <linux/dma-mapping.h>
 
-#include <plat/cpu.h>
+#include <video/omapdss.h>
 
 #include "omap_voutlib.h"
 
@@ -124,7 +124,7 @@ int omap_vout_new_window(struct v4l2_rect *crop,
 	win->chromakey = new_win->chromakey;
 
 	/* Adjust the cropping window to allow for resizing limitation */
-	if (cpu_is_omap24xx()) {
+	if (omap_vout_dss_omap24xx()) {
 		/* For 24xx limit is 8x to 1/2x scaling. */
 		if ((crop->height/win->w.height) >= 2)
 			crop->height = win->w.height * 2;
@@ -140,7 +140,7 @@ int omap_vout_new_window(struct v4l2_rect *crop,
 			if (crop->height != win->w.height)
 				crop->width = 768;
 		}
-	} else if (cpu_is_omap34xx()) {
+	} else if (omap_vout_dss_omap34xx()) {
 		/* For 34xx limit is 8x to 1/4x scaling. */
 		if ((crop->height/win->w.height) >= 4)
 			crop->height = win->w.height * 4;
@@ -196,7 +196,7 @@ int omap_vout_new_crop(struct v4l2_pix_format *pix,
 	if (try_crop.width <= 0 || try_crop.height <= 0)
 		return -EINVAL;
 
-	if (cpu_is_omap24xx()) {
+	if (omap_vout_dss_omap24xx()) {
 		if (try_crop.height != win->w.height) {
 			/* If we're resizing vertically, we can't support a
 			 * crop width wider than 768 pixels.
@@ -207,9 +207,9 @@ int omap_vout_new_crop(struct v4l2_pix_format *pix,
 	}
 	/* vertical resizing */
 	vresize = (1024 * try_crop.height) / win->w.height;
-	if (cpu_is_omap24xx() && (vresize > 2048))
+	if (omap_vout_dss_omap24xx() && (vresize > 2048))
 		vresize = 2048;
-	else if (cpu_is_omap34xx() && (vresize > 4096))
+	else if (omap_vout_dss_omap34xx() && (vresize > 4096))
 		vresize = 4096;
 
 	win->w.height = ((1024 * try_crop.height) / vresize) & ~1;
@@ -226,9 +226,9 @@ int omap_vout_new_crop(struct v4l2_pix_format *pix,
 	}
 	/* horizontal resizing */
 	hresize = (1024 * try_crop.width) / win->w.width;
-	if (cpu_is_omap24xx() && (hresize > 2048))
+	if (omap_vout_dss_omap24xx() && (hresize > 2048))
 		hresize = 2048;
-	else if (cpu_is_omap34xx() && (hresize > 4096))
+	else if (omap_vout_dss_omap34xx() && (hresize > 4096))
 		hresize = 4096;
 
 	win->w.width = ((1024 * try_crop.width) / hresize) & ~1;
@@ -243,7 +243,7 @@ int omap_vout_new_crop(struct v4l2_pix_format *pix,
 		if (try_crop.width == 0)
 			try_crop.width = 2;
 	}
-	if (cpu_is_omap24xx()) {
+	if (omap_vout_dss_omap24xx()) {
 		if ((try_crop.height/win->w.height) >= 2)
 			try_crop.height = win->w.height * 2;
 
@@ -258,7 +258,7 @@ int omap_vout_new_crop(struct v4l2_pix_format *pix,
 			if (try_crop.height != win->w.height)
 				try_crop.width = 768;
 		}
-	} else if (cpu_is_omap34xx()) {
+	} else if (omap_vout_dss_omap34xx()) {
 		if ((try_crop.height/win->w.height) >= 4)
 			try_crop.height = win->w.height * 4;
 
@@ -337,3 +337,21 @@ void omap_vout_free_buffer(unsigned long virtaddr, u32 buf_size)
 	}
 	free_pages((unsigned long) virtaddr, order);
 }
+
+bool omap_vout_dss_omap24xx(void)
+{
+	return omapdss_get_version() == OMAPDSS_VER_OMAP24xx;
+}
+
+bool omap_vout_dss_omap34xx(void)
+{
+	switch (omapdss_get_version()) {
+	case OMAPDSS_VER_OMAP34xx_ES1:
+	case OMAPDSS_VER_OMAP34xx_ES3:
+	case OMAPDSS_VER_OMAP3630:
+	case OMAPDSS_VER_AM35xx:
+		return true;
+	default:
+		return false;
+	}
+}
diff --git a/drivers/media/platform/omap/omap_voutlib.h b/drivers/media/platform/omap/omap_voutlib.h
index e51750a..f9d1c07 100644
--- a/drivers/media/platform/omap/omap_voutlib.h
+++ b/drivers/media/platform/omap/omap_voutlib.h
@@ -32,5 +32,8 @@ void omap_vout_new_format(struct v4l2_pix_format *pix,
 		struct v4l2_window *win);
 unsigned long omap_vout_alloc_buffer(u32 buf_size, u32 *phys_addr);
 void omap_vout_free_buffer(unsigned long virtaddr, u32 buf_size);
+
+bool omap_vout_dss_omap24xx(void);
+bool omap_vout_dss_omap34xx(void);
 #endif	/* #ifndef OMAP_VOUTLIB_H */
 
-- 
1.7.10.4

[PATCH 2/2] [media] omap_vout: remove extra include

[Tomi Valkeinen]

Remove including plat/dma.h which is not needed.

Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
---
 drivers/media/platform/omap/omap_vout.c |    1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/media/platform/omap/omap_vout.c b/drivers/media/platform/omap/omap_vout.c
index 7b1afc8..a2cc634 100644
--- a/drivers/media/platform/omap/omap_vout.c
+++ b/drivers/media/platform/omap/omap_vout.c
@@ -44,7 +44,6 @@
 #include <media/v4l2-device.h>
 #include <media/v4l2-ioctl.h>
 
-#include <plat/dma.h>
 #include <video/omapvrfb.h>
 #include <video/omapdss.h>
 
-- 
1.7.10.4

Re: [PATCH 0/2] omap_vout: remove cpu_is_* uses

[Tony Lindgren]

* Tomi Valkeinen <tomi.valkeinen@ti.com> [121112 05:35]:
> Hi,
> 
> This patch removes use of cpu_is_* funcs from omap_vout, and uses omapdss's
> version instead. The other patch removes an unneeded plat/dma.h include.
> 
> These are based on current omapdss master branch, which has the omapdss version
> code. The omapdss version code is queued for v3.8. I'm not sure which is the
> best way to handle these patches due to the dependency to omapdss. The easiest
> option is to merge these for 3.9.
> 
> There's still the OMAP DMA use in omap_vout_vrfb.c, which is the last OMAP
> dependency in the omap_vout driver. I'm not going to touch that, as it doesn't
> look as trivial as this cpu_is_* removal, and I don't have much knowledge of
> the omap_vout driver.
> 
> Compiled, but not tested.

Thanks for fixing this. Can you please resend with the selected
people cc:d from output of:

$ scripts/get_maintainer.pl -f drivers/media/platform/omap

So we get the necessary acks.

Regards,

Tony 

Re: [PATCH 0/2] omap_vout: remove cpu_is_* uses

[Laurent Pinchart]

Hi Tomi,

On Monday 12 November 2012 15:33:38 Tomi Valkeinen wrote:
> Hi,
> 
> This patch removes use of cpu_is_* funcs from omap_vout, and uses omapdss's
> version instead. The other patch removes an unneeded plat/dma.h include.
> 
> These are based on current omapdss master branch, which has the omapdss
> version code. The omapdss version code is queued for v3.8. I'm not sure
> which is the best way to handle these patches due to the dependency to
> omapdss. The easiest option is to merge these for 3.9.
> 
> There's still the OMAP DMA use in omap_vout_vrfb.c, which is the last OMAP
> dependency in the omap_vout driver. I'm not going to touch that, as it
> doesn't look as trivial as this cpu_is_* removal, and I don't have much
> knowledge of the omap_vout driver.
> 
> Compiled, but not tested.

Tested on a Beagleboard-xM.

Tested-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>

The patches depend on unmerged OMAP DSS patches. Would you like to push this 
series through linuxtv or through your DSS tree ? The later might be easier, 
depending on when the required DSS patches will hit mainline.

> Tomi Valkeinen (2):
>   [media] omap_vout: use omapdss's version instead of cpu_is_*
>   [media] omap_vout: remove extra include
> 
>  drivers/media/platform/omap/omap_vout.c    |    4 +--
>  drivers/media/platform/omap/omap_voutlib.c |   38 ++++++++++++++++++-------
>  drivers/media/platform/omap/omap_voutlib.h |    3 +++
>  3 files changed, 32 insertions(+), 13 deletions(-)

-- 
Regards,

Laurent Pinchart

Re: [PATCH 0/2] omap_vout: remove cpu_is_* uses

[Tomi Valkeinen]

[-- Attachment #1: Type: text/plain, Size: 1852 bytes --]

On 2012-11-28 17:13, Laurent Pinchart wrote:
> Hi Tomi,
> 
> On Monday 12 November 2012 15:33:38 Tomi Valkeinen wrote:
>> Hi,
>>
>> This patch removes use of cpu_is_* funcs from omap_vout, and uses omapdss's
>> version instead. The other patch removes an unneeded plat/dma.h include.
>>
>> These are based on current omapdss master branch, which has the omapdss
>> version code. The omapdss version code is queued for v3.8. I'm not sure
>> which is the best way to handle these patches due to the dependency to
>> omapdss. The easiest option is to merge these for 3.9.
>>
>> There's still the OMAP DMA use in omap_vout_vrfb.c, which is the last OMAP
>> dependency in the omap_vout driver. I'm not going to touch that, as it
>> doesn't look as trivial as this cpu_is_* removal, and I don't have much
>> knowledge of the omap_vout driver.
>>
>> Compiled, but not tested.
> 
> Tested on a Beagleboard-xM.
> 
> Tested-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
> Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>

Thanks.

> The patches depend on unmerged OMAP DSS patches. Would you like to push this 
> series through linuxtv or through your DSS tree ? The later might be easier, 
> depending on when the required DSS patches will hit mainline.

The DSS patches will be merged for 3.8. I can take this via the omapdss
tree, as there probably won't be any conflicts with other v4l2 stuff.

Or, we can just delay these until 3.9. These patches remove omap
platform dependencies, helping the effort to get common ARM kernel.
However, as there's still the VRFB code in the omap_vout driver, the
dependency remains. Thus, in way, these patches alone don't help
anything, and we could delay these for 3.9 and hope that
omap_vout_vrfb.c gets converted also for that merge window.

 Tomi



[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 899 bytes --]

Re: [PATCH 0/2] omap_vout: remove cpu_is_* uses

[Laurent Pinchart]

[-- Attachment #1: Type: text/plain, Size: 2007 bytes --]

Hi Tomi,

On Thursday 29 November 2012 11:30:28 Tomi Valkeinen wrote:
> On 2012-11-28 17:13, Laurent Pinchart wrote:
> > On Monday 12 November 2012 15:33:38 Tomi Valkeinen wrote:
> >> Hi,
> >> 
> >> This patch removes use of cpu_is_* funcs from omap_vout, and uses
> >> omapdss's version instead. The other patch removes an unneeded plat/dma.h
> >> include.
> >> 
> >> These are based on current omapdss master branch, which has the omapdss
> >> version code. The omapdss version code is queued for v3.8. I'm not sure
> >> which is the best way to handle these patches due to the dependency to
> >> omapdss. The easiest option is to merge these for 3.9.
> >> 
> >> There's still the OMAP DMA use in omap_vout_vrfb.c, which is the last
> >> OMAP dependency in the omap_vout driver. I'm not going to touch that, as
> >> it doesn't look as trivial as this cpu_is_* removal, and I don't have
> >> much knowledge of the omap_vout driver.
> >> 
> >> Compiled, but not tested.
> > 
> > Tested on a Beagleboard-xM.
> > 
> > Tested-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
> > Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
> 
> Thanks.
> 
> > The patches depend on unmerged OMAP DSS patches. Would you like to push
> > this series through linuxtv or through your DSS tree ? The later might be
> > easier, depending on when the required DSS patches will hit mainline.
> 
> The DSS patches will be merged for 3.8. I can take this via the omapdss
> tree, as there probably won't be any conflicts with other v4l2 stuff.
> 
> Or, we can just delay these until 3.9. These patches remove omap
> platform dependencies, helping the effort to get common ARM kernel.
> However, as there's still the VRFB code in the omap_vout driver, the
> dependency remains. Thus, in way, these patches alone don't help
> anything, and we could delay these for 3.9 and hope that
> omap_vout_vrfb.c gets converted also for that merge window.

OK, I'll queue them for v3.9 then.

-- 
Regards,

Laurent Pinchart

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 490 bytes --]

Re: [PATCH 0/2] omap_vout: remove cpu_is_* uses

[Tony Lindgren]

* Laurent Pinchart <laurent.pinchart@ideasonboard.com> [121129 01:37]:
> Hi Tomi,
> 
> On Thursday 29 November 2012 11:30:28 Tomi Valkeinen wrote:
> > On 2012-11-28 17:13, Laurent Pinchart wrote:
> > > On Monday 12 November 2012 15:33:38 Tomi Valkeinen wrote:
> > >> Hi,
> > >> 
> > >> This patch removes use of cpu_is_* funcs from omap_vout, and uses
> > >> omapdss's version instead. The other patch removes an unneeded plat/dma.h
> > >> include.
> > >> 
> > >> These are based on current omapdss master branch, which has the omapdss
> > >> version code. The omapdss version code is queued for v3.8. I'm not sure
> > >> which is the best way to handle these patches due to the dependency to
> > >> omapdss. The easiest option is to merge these for 3.9.
> > >> 
> > >> There's still the OMAP DMA use in omap_vout_vrfb.c, which is the last
> > >> OMAP dependency in the omap_vout driver. I'm not going to touch that, as
> > >> it doesn't look as trivial as this cpu_is_* removal, and I don't have
> > >> much knowledge of the omap_vout driver.
> > >> 
> > >> Compiled, but not tested.
> > > 
> > > Tested on a Beagleboard-xM.
> > > 
> > > Tested-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
> > > Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
> > 
> > Thanks.
> > 
> > > The patches depend on unmerged OMAP DSS patches. Would you like to push
> > > this series through linuxtv or through your DSS tree ? The later might be
> > > easier, depending on when the required DSS patches will hit mainline.
> > 
> > The DSS patches will be merged for 3.8. I can take this via the omapdss
> > tree, as there probably won't be any conflicts with other v4l2 stuff.
> > 
> > Or, we can just delay these until 3.9. These patches remove omap
> > platform dependencies, helping the effort to get common ARM kernel.
> > However, as there's still the VRFB code in the omap_vout driver, the
> > dependency remains. Thus, in way, these patches alone don't help
> > anything, and we could delay these for 3.9 and hope that
> > omap_vout_vrfb.c gets converted also for that merge window.
> 
> OK, I'll queue them for v3.9 then.

Please rather queue the cpu_is_omap removal to v3.8 so I can
remove plat/cpu.h for omap2+.

Regards,

Tony

Re: [PATCH 0/2] omap_vout: remove cpu_is_* uses

[Laurent Pinchart]

Hi Tony,

On Thursday 29 November 2012 08:29:27 Tony Lindgren wrote:
> * Laurent Pinchart <laurent.pinchart@ideasonboard.com> [121129 01:37]:
> > On Thursday 29 November 2012 11:30:28 Tomi Valkeinen wrote:
> > > On 2012-11-28 17:13, Laurent Pinchart wrote:
> > > > On Monday 12 November 2012 15:33:38 Tomi Valkeinen wrote:
> > > >> Hi,
> > > >> 
> > > >> This patch removes use of cpu_is_* funcs from omap_vout, and uses
> > > >> omapdss's version instead. The other patch removes an unneeded
> > > >> plat/dma.h include.
> > > >> 
> > > >> These are based on current omapdss master branch, which has the
> > > >> omapdss version code. The omapdss version code is queued for v3.8.
> > > >> I'm not sure which is the best way to handle these patches due to the
> > > >> dependency to omapdss. The easiest option is to merge these for 3.9.
> > > >> 
> > > >> There's still the OMAP DMA use in omap_vout_vrfb.c, which is the last
> > > >> OMAP dependency in the omap_vout driver. I'm not going to touch that,
> > > >> as it doesn't look as trivial as this cpu_is_* removal, and I don't
> > > >> have much knowledge of the omap_vout driver.
> > > >> 
> > > >> Compiled, but not tested.
> > > > 
> > > > Tested on a Beagleboard-xM.
> > > > 
> > > > Tested-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
> > > > Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
> > > 
> > > Thanks.
> > > 
> > > > The patches depend on unmerged OMAP DSS patches. Would you like to
> > > > push this series through linuxtv or through your DSS tree ? The later
> > > > might be easier, depending on when the required DSS patches will hit
> > > > mainline.
> > > 
> > > The DSS patches will be merged for 3.8. I can take this via the omapdss
> > > tree, as there probably won't be any conflicts with other v4l2 stuff.
> > > 
> > > Or, we can just delay these until 3.9. These patches remove omap
> > > platform dependencies, helping the effort to get common ARM kernel.
> > > However, as there's still the VRFB code in the omap_vout driver, the
> > > dependency remains. Thus, in way, these patches alone don't help
> > > anything, and we could delay these for 3.9 and hope that
> > > omap_vout_vrfb.c gets converted also for that merge window.
> > 
> > OK, I'll queue them for v3.9 then.
> 
> Please rather queue the cpu_is_omap removal to v3.8 so I can
> remove plat/cpu.h for omap2+.

In that case the patches should go through the DSS tree. Mauro, are you fine 
with that ?

-- 
Regards,

Laurent Pinchart

Re: [PATCH 0/2] omap_vout: remove cpu_is_* uses

[Mauro Carvalho Chehab]

Em Thu, 29 Nov 2012 17:39:45 +0100
Laurent Pinchart <laurent.pinchart@ideasonboard.com> escreveu:

> Hi Tony,
> 
> On Thursday 29 November 2012 08:29:27 Tony Lindgren wrote:
> > * Laurent Pinchart <laurent.pinchart@ideasonboard.com> [121129 01:37]:
> > > On Thursday 29 November 2012 11:30:28 Tomi Valkeinen wrote:
> > > > On 2012-11-28 17:13, Laurent Pinchart wrote:
> > > > > On Monday 12 November 2012 15:33:38 Tomi Valkeinen wrote:
> > > > >> Hi,
> > > > >> 
> > > > >> This patch removes use of cpu_is_* funcs from omap_vout, and uses
> > > > >> omapdss's version instead. The other patch removes an unneeded
> > > > >> plat/dma.h include.
> > > > >> 
> > > > >> These are based on current omapdss master branch, which has the
> > > > >> omapdss version code. The omapdss version code is queued for v3.8.
> > > > >> I'm not sure which is the best way to handle these patches due to the
> > > > >> dependency to omapdss. The easiest option is to merge these for 3.9.
> > > > >> 
> > > > >> There's still the OMAP DMA use in omap_vout_vrfb.c, which is the last
> > > > >> OMAP dependency in the omap_vout driver. I'm not going to touch that,
> > > > >> as it doesn't look as trivial as this cpu_is_* removal, and I don't
> > > > >> have much knowledge of the omap_vout driver.
> > > > >> 
> > > > >> Compiled, but not tested.
> > > > > 
> > > > > Tested on a Beagleboard-xM.
> > > > > 
> > > > > Tested-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
> > > > > Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
> > > > 
> > > > Thanks.
> > > > 
> > > > > The patches depend on unmerged OMAP DSS patches. Would you like to
> > > > > push this series through linuxtv or through your DSS tree ? The later
> > > > > might be easier, depending on when the required DSS patches will hit
> > > > > mainline.
> > > > 
> > > > The DSS patches will be merged for 3.8. I can take this via the omapdss
> > > > tree, as there probably won't be any conflicts with other v4l2 stuff.
> > > > 
> > > > Or, we can just delay these until 3.9. These patches remove omap
> > > > platform dependencies, helping the effort to get common ARM kernel.
> > > > However, as there's still the VRFB code in the omap_vout driver, the
> > > > dependency remains. Thus, in way, these patches alone don't help
> > > > anything, and we could delay these for 3.9 and hope that
> > > > omap_vout_vrfb.c gets converted also for that merge window.
> > > 
> > > OK, I'll queue them for v3.9 then.
> > 
> > Please rather queue the cpu_is_omap removal to v3.8 so I can
> > remove plat/cpu.h for omap2+.
> 
> In that case the patches should go through the DSS tree. Mauro, are you fine 
> with that ?

Sure.

For both patches:

Acked-by: Mauro Carvalho Chehab <mchehab@redhat.com>

Regards,
Mauro

Re: [PATCH 0/2] omap_vout: remove cpu_is_* uses

[Tomi Valkeinen]

[-- Attachment #1: Type: text/plain, Size: 521 bytes --]

On 2012-11-29 19:05, Mauro Carvalho Chehab wrote:
> Em Thu, 29 Nov 2012 17:39:45 +0100
> Laurent Pinchart <laurent.pinchart@ideasonboard.com> escreveu:

>>> Please rather queue the cpu_is_omap removal to v3.8 so I can
>>> remove plat/cpu.h for omap2+.
>>
>> In that case the patches should go through the DSS tree. Mauro, are you fine 
>> with that ?
> 
> Sure.
> 
> For both patches:
> 
> Acked-by: Mauro Carvalho Chehab <mchehab@redhat.com>

Okay, thanks, I'll apply them to omapdss tree.

 Tomi



[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 899 bytes --]