[PATCH] V4L: mx3_camera: select VIDEOBUF2_DMA_CONTIG instead of VIDEOBUF_DMA_CONTIG

[PATCH] V4L: mx3_camera: select VIDEOBUF2_DMA_CONTIG instead of VIDEOBUF_DMA_CONTIG

[Uwe Kleine-König]

Since commit

	379fa5d ([media] V4L: mx3_camera: convert to videobuf2)

mx3_camera uses videobuf2, but that commit didn't upgrade the select
resulting in the following build failure:

	drivers/built-in.o: In function `mx3_camera_init_videobuf':
	clkdev.c:(.text+0x86580): undefined reference to `vb2_dma_contig_memops'
	drivers/built-in.o: In function `mx3_camera_probe':
	clkdev.c:(.devinit.text+0x3548): undefined reference to `vb2_dma_contig_init_ctx'
	clkdev.c:(.devinit.text+0x3578): undefined reference to `vb2_dma_contig_cleanup_ctx'
	drivers/built-in.o: In function `mx3_camera_remove':
	clkdev.c:(.devexit.text+0x674): undefined reference to `vb2_dma_contig_cleanup_ctx'
	make[2]: *** [.tmp_vmlinux1] Error 1
	make[1]: *** [sub-make] Error 2
	make: *** [all] Error 2

Signed-off-by: Uwe Kleine-K?nig <u.kleine-koenig@pengutronix.de>
---
Hello,

does someone has a hint how to fix gcc not to believe the undefined
references to be in clkdev.c?

Best regards
Uwe

 drivers/media/video/Kconfig |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/drivers/media/video/Kconfig b/drivers/media/video/Kconfig
index 4498b94..00f51dd 100644
--- a/drivers/media/video/Kconfig
+++ b/drivers/media/video/Kconfig
@@ -875,7 +875,7 @@ config MX3_VIDEO
 config VIDEO_MX3
 	tristate "i.MX3x Camera Sensor Interface driver"
 	depends on VIDEO_DEV && MX3_IPU && SOC_CAMERA
-	select VIDEOBUF_DMA_CONTIG
+	select VIDEOBUF2_DMA_CONTIG
 	select MX3_VIDEO
 	---help---
 	  This is a v4l2 driver for the i.MX3x Camera Sensor Interface
-- 
1.7.2.3

[PATCH] V4L: mx3_camera: select VIDEOBUF2_DMA_CONTIG instead of VIDEOBUF_DMA_CONTIG

[Uwe Kleine-König]

Hi Guennadi,

On Thu, Apr 07, 2011 at 10:50:43AM +0200, Uwe Kleine-K?nig wrote:
> Since commit
> 
> 	379fa5d ([media] V4L: mx3_camera: convert to videobuf2)
> 
> mx3_camera uses videobuf2, but that commit didn't upgrade the select
> resulting in the following build failure:
> 
> 	drivers/built-in.o: In function `mx3_camera_init_videobuf':
> 	clkdev.c:(.text+0x86580): undefined reference to `vb2_dma_contig_memops'
> 	drivers/built-in.o: In function `mx3_camera_probe':
> 	clkdev.c:(.devinit.text+0x3548): undefined reference to `vb2_dma_contig_init_ctx'
> 	clkdev.c:(.devinit.text+0x3578): undefined reference to `vb2_dma_contig_cleanup_ctx'
> 	drivers/built-in.o: In function `mx3_camera_remove':
> 	clkdev.c:(.devexit.text+0x674): undefined reference to `vb2_dma_contig_cleanup_ctx'
> 	make[2]: *** [.tmp_vmlinux1] Error 1
> 	make[1]: *** [sub-make] Error 2
> 	make: *** [all] Error 2
> 
> Signed-off-by: Uwe Kleine-K?nig <u.kleine-koenig@pengutronix.de>
> ---
I guess the only problem with this is -ENOTIME on your side?

> does someone has a hint how to fix gcc not to believe the undefined
> references to be in clkdev.c?
I got a hint that might be related to ccache. Didn't look into it yet,
though.

Best regards
Uwe

-- 
Pengutronix e.K.                           | Uwe Kleine-K?nig            |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |

[PATCH] V4L: mx3_camera: select VIDEOBUF2_DMA_CONTIG instead of VIDEOBUF_DMA_CONTIG

[Guennadi Liakhovetski]

On Mon, 18 Apr 2011, Uwe Kleine-K?nig wrote:

> Hi Guennadi,
> 
> On Thu, Apr 07, 2011 at 10:50:43AM +0200, Uwe Kleine-K?nig wrote:
> > Since commit
> > 
> > 	379fa5d ([media] V4L: mx3_camera: convert to videobuf2)
> > 
> > mx3_camera uses videobuf2, but that commit didn't upgrade the select
> > resulting in the following build failure:
> > 
> > 	drivers/built-in.o: In function `mx3_camera_init_videobuf':
> > 	clkdev.c:(.text+0x86580): undefined reference to `vb2_dma_contig_memops'
> > 	drivers/built-in.o: In function `mx3_camera_probe':
> > 	clkdev.c:(.devinit.text+0x3548): undefined reference to `vb2_dma_contig_init_ctx'
> > 	clkdev.c:(.devinit.text+0x3578): undefined reference to `vb2_dma_contig_cleanup_ctx'
> > 	drivers/built-in.o: In function `mx3_camera_remove':
> > 	clkdev.c:(.devexit.text+0x674): undefined reference to `vb2_dma_contig_cleanup_ctx'
> > 	make[2]: *** [.tmp_vmlinux1] Error 1
> > 	make[1]: *** [sub-make] Error 2
> > 	make: *** [all] Error 2
> > 
> > Signed-off-by: Uwe Kleine-K?nig <u.kleine-koenig@pengutronix.de>
> > ---
> I guess the only problem with this is -ENOTIME on your side?

It's been pushed upstream almost 2 weeks ago:

http://article.gmane.org/gmane.linux.drivers.video-input-infrastructure/31352

Thanks
Guennadi

> 
> > does someone has a hint how to fix gcc not to believe the undefined
> > references to be in clkdev.c?
> I got a hint that might be related to ccache. Didn't look into it yet,
> though.
> 
> Best regards
> Uwe
> 
> -- 
> Pengutronix e.K.                           | Uwe Kleine-K?nig            |
> Industrial Linux Solutions                 | http://www.pengutronix.de/  |
> 

---
Guennadi Liakhovetski, Ph.D.
Freelance Open-Source Software Developer
http://www.open-technology.de/

[PATCH] V4L: mx3_camera: select VIDEOBUF2_DMA_CONTIG instead of VIDEOBUF_DMA_CONTIG

[Robert Schwebel]

Uwe,

On Mon, Apr 18, 2011 at 10:14:56AM +0200, Guennadi Liakhovetski wrote:
> It's been pushed upstream almost 2 weeks ago:
>
> http://article.gmane.org/gmane.linux.drivers.video-input-infrastructure/31352

As our autobuilder does still trigger, I assume that the configs have to
be refreshed and it may be an issue on our side. Can you take care of
that?

rsc
-- 
Pengutronix e.K.                           |                             |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |
Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |

[PATCH] V4L: mx3_camera: select VIDEOBUF2_DMA_CONTIG instead of VIDEOBUF_DMA_CONTIG

[Russell King - ARM Linux]

On Mon, Apr 18, 2011 at 10:20:49AM +0200, Robert Schwebel wrote:
> Uwe,
> 
> On Mon, Apr 18, 2011 at 10:14:56AM +0200, Guennadi Liakhovetski wrote:
> > It's been pushed upstream almost 2 weeks ago:
> >
> > http://article.gmane.org/gmane.linux.drivers.video-input-infrastructure/31352
> 
> As our autobuilder does still trigger, I assume that the configs have to
> be refreshed and it may be an issue on our side. Can you take care of
> that?

Just take a look at what's in mainline:

config VIDEO_MX3
        tristate "i.MX3x Camera Sensor Interface driver"
        depends on VIDEO_DEV && MX3_IPU && SOC_CAMERA
        select VIDEOBUF_DMA_CONTIG
        select MX3_VIDEO
        ---help---
          This is a v4l2 driver for the i.MX3x Camera Sensor Interface

and you'll see that it hasn't made it there yet.  If I search for
'linuxtv.org' in the history post 2.6.39-rc2, there's no sign of it.

[PATCH] V4L: mx3_camera: select VIDEOBUF2_DMA_CONTIG instead of VIDEOBUF_DMA_CONTIG

[Uwe Kleine-König]

Hi Robert,

On Mon, Apr 18, 2011 at 10:20:49AM +0200, Robert Schwebel wrote:
> On Mon, Apr 18, 2011 at 10:14:56AM +0200, Guennadi Liakhovetski wrote:
> > It's been pushed upstream almost 2 weeks ago:
> >
> > http://article.gmane.org/gmane.linux.drivers.video-input-infrastructure/31352
> 
> As our autobuilder does still trigger, I assume that the configs have to
> be refreshed and it may be an issue on our side. Can you take care of
> that?
The problem is not our config, but that the change didn't hit next
(next-20110418) yet.

git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-2.6 (master)
(which I guess is the tree targeted by Guennadi's pull request)
is currently at

	78fca1b9 (Merge branch 'release' of git://git.kernel.org/pub/scm/linux/kernel/git/aegl/linux-2.6)

which doesn't contain Guennadi's pull request. Also the other branches
in Mauro's repo don't contain it, in fact the repository is fully merged
into Linus tree.

Best regards
Uwe

-- 
Pengutronix e.K.                           | Uwe Kleine-K?nig            |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |