linux-next: build failure after merge of the v4l-dvb tree

linux-next: build failure after merge of the v4l-dvb tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 838 bytes --]

Hi Mauro,

After merging the v4l-dvb tree, today's linux-next build (x86_64
allmodconfig) failed like this:

drivers/staging/dt3155v4l/dt3155v4l.c: In function 'dt3155_start_acq':
drivers/staging/dt3155v4l/dt3155v4l.c:210:2: error: implicit declaration of function 'vb2_dma_contig_plane_paddr'
drivers/staging/dt3155v4l/dt3155v4l.c: At top level:
drivers/staging/dt3155v4l/dt3155v4l.c:321:2: warning: initialization from incompatible pointer type
drivers/staging/dt3155v4l/dt3155v4l.c:325:2: warning: initialization from incompatible pointer type

Casued by commit ba7fcb0c9549 ("[media] media: vb2: dma contig allocator:
use dma_addr instread of paddr").

I have used the v4l-dvb tree from next-20110921 for today.
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/

[-- Attachment #2: Type: application/pgp-signature, Size: 490 bytes --]

[PATCH] staging: dt3155v4l: fix build break

[Marek Szyprowski]

This patch fixes build break caused by commit ba7fcb0c9549 ("[media] media: vb2: dma
contig allocator: use dma_addr instread of paddr").

Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
---
 drivers/staging/dt3155v4l/dt3155v4l.c |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/staging/dt3155v4l/dt3155v4l.c b/drivers/staging/dt3155v4l/dt3155v4l.c
index 05aa41c..0ede5d1 100644
--- a/drivers/staging/dt3155v4l/dt3155v4l.c
+++ b/drivers/staging/dt3155v4l/dt3155v4l.c
@@ -207,7 +207,7 @@ dt3155_start_acq(struct dt3155_priv *pd)
 	struct vb2_buffer *vb = pd->curr_buf;
 	dma_addr_t dma_addr;
 
-	dma_addr = vb2_dma_contig_plane_paddr(vb, 0);
+	dma_addr = vb2_dma_contig_plane_dma_addr(vb, 0);
 	iowrite32(dma_addr, pd->regs + EVEN_DMA_START);
 	iowrite32(dma_addr + img_width, pd->regs + ODD_DMA_START);
 	iowrite32(img_width, pd->regs + EVEN_DMA_STRIDE);
@@ -374,7 +374,7 @@ dt3155_irq_handler_even(int irq, void *dev_id)
 	ivb = list_first_entry(&ipd->dmaq, typeof(*ivb), done_entry);
 	list_del(&ivb->done_entry);
 	ipd->curr_buf = ivb;
-	dma_addr = vb2_dma_contig_plane_paddr(ivb, 0);
+	dma_addr = vb2_dma_contig_plane_dma_addr(ivb, 0);
 	iowrite32(dma_addr, ipd->regs + EVEN_DMA_START);
 	iowrite32(dma_addr + img_width, ipd->regs + ODD_DMA_START);
 	iowrite32(img_width, ipd->regs + EVEN_DMA_STRIDE);
-- 
1.7.1.569.g6f426

Re: linux-next: build failure after merge of the v4l-dvb tree

[Mauro Carvalho Chehab]

Em 22-09-2011 00:12, Stephen Rothwell escreveu:
> Hi Mauro,
> 
> After merging the v4l-dvb tree, today's linux-next build (x86_64
> allmodconfig) failed like this:
> 
> drivers/staging/dt3155v4l/dt3155v4l.c: In function 'dt3155_start_acq':
> drivers/staging/dt3155v4l/dt3155v4l.c:210:2: error: implicit declaration of function 'vb2_dma_contig_plane_paddr'
> drivers/staging/dt3155v4l/dt3155v4l.c: At top level:
> drivers/staging/dt3155v4l/dt3155v4l.c:321:2: warning: initialization from incompatible pointer type
> drivers/staging/dt3155v4l/dt3155v4l.c:325:2: warning: initialization from incompatible pointer type
> 
> Casued by commit ba7fcb0c9549 ("[media] media: vb2: dma contig allocator:
> use dma_addr instread of paddr").
> 
> I have used the v4l-dvb tree from next-20110921 for today.

Hi Stephen,

I think that the right thing to do is to disable the building of this
staging driver, if it is not using the current media core API's.

Greg,

If possible, I prefer to merge those staging drivers for the media
subsystem via my tree, as it helps me to solve in advance conflicts
like that. Maybe we could add a drivers/staging/media tree or
drivers/media/staging, and move those beasts to there.

What do you think?

Thanks,
Mauro

Re: linux-next: build failure after merge of the v4l-dvb tree

[Greg KH]

On Thu, Sep 22, 2011 at 09:46:59AM -0300, Mauro Carvalho Chehab wrote:
> Em 22-09-2011 00:12, Stephen Rothwell escreveu:
> > Hi Mauro,
> > 
> > After merging the v4l-dvb tree, today's linux-next build (x86_64
> > allmodconfig) failed like this:
> > 
> > drivers/staging/dt3155v4l/dt3155v4l.c: In function 'dt3155_start_acq':
> > drivers/staging/dt3155v4l/dt3155v4l.c:210:2: error: implicit declaration of function 'vb2_dma_contig_plane_paddr'
> > drivers/staging/dt3155v4l/dt3155v4l.c: At top level:
> > drivers/staging/dt3155v4l/dt3155v4l.c:321:2: warning: initialization from incompatible pointer type
> > drivers/staging/dt3155v4l/dt3155v4l.c:325:2: warning: initialization from incompatible pointer type
> > 
> > Casued by commit ba7fcb0c9549 ("[media] media: vb2: dma contig allocator:
> > use dma_addr instread of paddr").
> > 
> > I have used the v4l-dvb tree from next-20110921 for today.
> 
> Hi Stephen,
> 
> I think that the right thing to do is to disable the building of this
> staging driver, if it is not using the current media core API's.
> 
> Greg,
> 
> If possible, I prefer to merge those staging drivers for the media
> subsystem via my tree, as it helps me to solve in advance conflicts
> like that. Maybe we could add a drivers/staging/media tree or
> drivers/media/staging, and move those beasts to there.
> 
> What do you think?

Yes, I would _LOVE_ a drivers/staging/media/ directory so that I know
which drivers I am supposed to be ignoring and leaving to you :)

Just let me know which ones I should move into there and I will do that
today and let you maintain them all and just forward any emails to you.
Which will keep issues like this from happening in the future.

thanks,

greg k-h

Re: linux-next: build failure after merge of the v4l-dvb tree

[Mauro Carvalho Chehab]

Em 22-09-2011 17:19, Greg KH escreveu:
> On Thu, Sep 22, 2011 at 09:46:59AM -0300, Mauro Carvalho Chehab wrote:
>> Em 22-09-2011 00:12, Stephen Rothwell escreveu:
>>> Hi Mauro,
>>>
>>> After merging the v4l-dvb tree, today's linux-next build (x86_64
>>> allmodconfig) failed like this:
>>>
>>> drivers/staging/dt3155v4l/dt3155v4l.c: In function 'dt3155_start_acq':
>>> drivers/staging/dt3155v4l/dt3155v4l.c:210:2: error: implicit declaration of function 'vb2_dma_contig_plane_paddr'
>>> drivers/staging/dt3155v4l/dt3155v4l.c: At top level:
>>> drivers/staging/dt3155v4l/dt3155v4l.c:321:2: warning: initialization from incompatible pointer type
>>> drivers/staging/dt3155v4l/dt3155v4l.c:325:2: warning: initialization from incompatible pointer type
>>>
>>> Casued by commit ba7fcb0c9549 ("[media] media: vb2: dma contig allocator:
>>> use dma_addr instread of paddr").
>>>
>>> I have used the v4l-dvb tree from next-20110921 for today.
>>
>> Hi Stephen,
>>
>> I think that the right thing to do is to disable the building of this
>> staging driver, if it is not using the current media core API's.
>>
>> Greg,
>>
>> If possible, I prefer to merge those staging drivers for the media
>> subsystem via my tree, as it helps me to solve in advance conflicts
>> like that. Maybe we could add a drivers/staging/media tree or
>> drivers/media/staging, and move those beasts to there.
>>
>> What do you think?
> 
> Yes, I would _LOVE_ a drivers/staging/media/ directory so that I know
> which drivers I am supposed to be ignoring and leaving to you :)

:)

If you receive any new drivers like the dt3155v4l that include
one of those files:
	linux/videodev2.h
	media/*.h
	linux/dvb/*.h
They're dependent on the media core. So, could you please let me know about
its addition at staging, by either forwarding it to me or adding it at the 
new staging/media c/c me on your reply?

> Just let me know which ones I should move into there and I will do that
> today and let you maintain them all and just forward any emails to you.
> Which will keep issues like this from happening in the future.

At the current linux-next tree, the drivers are media stuff:
	staging/easycap
	staging/go7007
	staging/solo6x10
	staging/cx25821
	staging/lirc
	staging/cxd2099
	staging/altera-stapl
	staging/dt3155v4l

>From the above:
	 staging/tm6000 can be moved to /drivers/media/video, after a final checkpatch
cleanup patch I've made. However, I have some changesets for it on my -next tree,
so it is probably wiser if I move it on my tree instead on yours.

	staging/cx25281 can also be moved to drivers/media/video after some codingstyle
fixups. I think you're received some such patches recently. Not sure if they cover the
remaining checkpatch.pl significative complains.

	staging/cxd2099 has one problem: it exposes a wrong DVB API for the conditional
access module, due to the lack of a proper API definition. It is a driver optionally used 
by some DVB cards, so moving it may require some include fixups. I don't want to move it
out of staging while we don't have a solution for the API issue. I'm planning to discuss 
it during the media KS workshop in Plague.

	staging/altera-stapl can be moved to drivers/misc. One include needs to be fixup
when moving it, as one DVB driver uses it for a couple boards that come with this
companion chipset.

If you prefer, I can do the git mv patches for those weird stuff.

> 
> thanks,
> 
> greg k-h

Thanks,
Mauro

Re: linux-next: build failure after merge of the v4l-dvb tree

[Greg KH]

On Fri, Sep 23, 2011 at 11:27:02AM -0300, Mauro Carvalho Chehab wrote:
> Em 22-09-2011 17:19, Greg KH escreveu:
> > On Thu, Sep 22, 2011 at 09:46:59AM -0300, Mauro Carvalho Chehab wrote:
> >> Em 22-09-2011 00:12, Stephen Rothwell escreveu:
> >>> Hi Mauro,
> >>>
> >>> After merging the v4l-dvb tree, today's linux-next build (x86_64
> >>> allmodconfig) failed like this:
> >>>
> >>> drivers/staging/dt3155v4l/dt3155v4l.c: In function 'dt3155_start_acq':
> >>> drivers/staging/dt3155v4l/dt3155v4l.c:210:2: error: implicit declaration of function 'vb2_dma_contig_plane_paddr'
> >>> drivers/staging/dt3155v4l/dt3155v4l.c: At top level:
> >>> drivers/staging/dt3155v4l/dt3155v4l.c:321:2: warning: initialization from incompatible pointer type
> >>> drivers/staging/dt3155v4l/dt3155v4l.c:325:2: warning: initialization from incompatible pointer type
> >>>
> >>> Casued by commit ba7fcb0c9549 ("[media] media: vb2: dma contig allocator:
> >>> use dma_addr instread of paddr").
> >>>
> >>> I have used the v4l-dvb tree from next-20110921 for today.
> >>
> >> Hi Stephen,
> >>
> >> I think that the right thing to do is to disable the building of this
> >> staging driver, if it is not using the current media core API's.
> >>
> >> Greg,
> >>
> >> If possible, I prefer to merge those staging drivers for the media
> >> subsystem via my tree, as it helps me to solve in advance conflicts
> >> like that. Maybe we could add a drivers/staging/media tree or
> >> drivers/media/staging, and move those beasts to there.
> >>
> >> What do you think?
> > 
> > Yes, I would _LOVE_ a drivers/staging/media/ directory so that I know
> > which drivers I am supposed to be ignoring and leaving to you :)
> 
> :)
> 
> If you receive any new drivers like the dt3155v4l that include
> one of those files:
> 	linux/videodev2.h
> 	media/*.h
> 	linux/dvb/*.h
> They're dependent on the media core. So, could you please let me know about
> its addition at staging, by either forwarding it to me or adding it at the 
> new staging/media c/c me on your reply?
> 
> > Just let me know which ones I should move into there and I will do that
> > today and let you maintain them all and just forward any emails to you.
> > Which will keep issues like this from happening in the future.
> 
> At the current linux-next tree, the drivers are media stuff:
> 	staging/easycap
> 	staging/go7007
> 	staging/solo6x10
> 	staging/cx25821
> 	staging/lirc
> 	staging/cxd2099
> 	staging/altera-stapl
> 	staging/dt3155v4l
> 
> >From the above:
> 	 staging/tm6000 can be moved to /drivers/media/video, after a final checkpatch
> cleanup patch I've made. However, I have some changesets for it on my -next tree,
> so it is probably wiser if I move it on my tree instead on yours.
> 
> 	staging/cx25281 can also be moved to drivers/media/video after some codingstyle
> fixups. I think you're received some such patches recently. Not sure if they cover the
> remaining checkpatch.pl significative complains.
> 
> 	staging/cxd2099 has one problem: it exposes a wrong DVB API for the conditional
> access module, due to the lack of a proper API definition. It is a driver optionally used 
> by some DVB cards, so moving it may require some include fixups. I don't want to move it
> out of staging while we don't have a solution for the API issue. I'm planning to discuss 
> it during the media KS workshop in Plague.
> 
> 	staging/altera-stapl can be moved to drivers/misc. One include needs to be fixup
> when moving it, as one DVB driver uses it for a couple boards that come with this
> companion chipset.
> 
> If you prefer, I can do the git mv patches for those weird stuff.

That would be easiest, if you want to create drivers/staging/media/ and
move what you want to maintain into it, please do so.

thanks,

greg k-h