linux-next: build failure after merge of the staging.current tree

linux-next: build failure after merge of the staging.current tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 492 bytes --]

Hi Greg,

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

drivers/media/video/cx23885/cx23885-cards.c:28:28: fatal error: staging/altera.h: No such file or directory

Caused by commit 85ab9ee946da ("Staging: altera: move .h file to proper
place").

I have used the staging.current tree from next-20110607 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 --]

Re: linux-next: build failure after merge of the staging.current tree

[Greg KH]

On Wed, Jun 08, 2011 at 11:41:30AM +1000, Stephen Rothwell wrote:
> Hi Greg,
> 
> After merging the staging.current tree, today's linux-next build (x86_64
> allmodconfig) failed like this:
> 
> drivers/media/video/cx23885/cx23885-cards.c:28:28: fatal error: staging/altera.h: No such file or directory
> 
> Caused by commit 85ab9ee946da ("Staging: altera: move .h file to proper
> place").
> 
> I have used the staging.current tree from next-20110607 for today.

Ok, sorry about that.

Mauro, Igor, why would another driver need to reference a .h file of a
staging driver?  I'll patch it right now to point to the proper place
(down in the staging tree), but it would be good to resolve this
"properly" somehow.

Next time, please never include a staging driver that isn't
self-contained, and don't create include/staging/ that's not acceptable.

thanks,

greg k-h

Re: linux-next: build failure after merge of the staging.current tree

[Mauro Carvalho Chehab]

Hi Greg,

Em 08-06-2011 13:11, Greg KH escreveu:
> On Wed, Jun 08, 2011 at 11:41:30AM +1000, Stephen Rothwell wrote:
>> Hi Greg,
>>
>> After merging the staging.current tree, today's linux-next build (x86_64
>> allmodconfig) failed like this:
>>
>> drivers/media/video/cx23885/cx23885-cards.c:28:28: fatal error: staging/altera.h: No such file or directory
>>
>> Caused by commit 85ab9ee946da ("Staging: altera: move .h file to proper
>> place").
>>
>> I have used the staging.current tree from next-20110607 for today.
> 
> Ok, sorry about that.
> 
> Mauro, Igor, why would another driver need to reference a .h file of a
> staging driver?  I'll patch it right now to point to the proper place
> (down in the staging tree), but it would be good to resolve this
> "properly" somehow.
> 
> Next time, please never include a staging driver that isn't
> self-contained, and don't create include/staging/ that's not acceptable.

The altera driver were discussed for about two kernel cycles. When it was
about to be merged upstrem, two people complained about having a FPGA firmware 
driver inside the Linux Kernel. However, this driver is essential in order to load
the firmware for a DVB device. So, to avoid postponing the merge for yet another
kernel cycle, I've moved the offending part to staging, and asked for those
two people to provide us a new solution up to the next merge window, otherwise, it
would be moved it to the proper place.

Basically, this driver is used to load the FPGA firmware for the CI module that 
de-scrambles DVB-C/DVB-T/DVB-S streams, used by some cx23885 devices.

Basically, cx23885 gets the firmware file via request_firmware() and uses the
altera module to load it. After loaded, the altera-ci driver implements the logic
to talk with the CI hardware.

There's no Coding Style or any other troubles on this driver that I'm aware,
except that the altera driver also allows loading FPGA firmwares via LPT port.

The main argue is that there are some userspace tools available to program a
FPGA hardware, but the problem is that the cx23885 needs to command the firmware
load, as it needs to occur when the device is booted (or resumed).

There were some discussions about that at LKML:
	https://lkml.org/lkml/2011/3/7/270

>From my side, I think that the driver is doing the right approach. Using a different
logic will make harder to sync userspace with kernelspace and will require an
specialized userspace application for just a few device types (just two of the
30 boards supported by cx23885 currently needs it).

Igor,

Could you please prepare a patch against linux-next moving it to the proper place?
It is probably better to apply such patch at Greg's tree, as he has two patches
touching at the altera include files. Also, please get rid of the parallel port
support, as you don't need it for cx23885.

Thanks,
Mauro.

Re: linux-next: build failure after merge of the staging.current tree

[Greg KH]

On Wed, Jun 08, 2011 at 11:41:30AM +1000, Stephen Rothwell wrote:
> Hi Greg,
> 
> After merging the staging.current tree, today's linux-next build (x86_64
> allmodconfig) failed like this:
> 
> drivers/media/video/cx23885/cx23885-cards.c:28:28: fatal error: staging/altera.h: No such file or directory
> 
> Caused by commit 85ab9ee946da ("Staging: altera: move .h file to proper
> place").

Now resolved, it should be fine for you to pull from for the next
linux-next release.

greg k-h

linux-next: build failure after merge of the staging.current tree

[Stephen Rothwell]

Hi Greg,

After merging the staging.current tree, today's linux-next build (powerpc
allyesconfig) failed like this:

drivers/staging/android/ion/ion_cma_heap.c: In function 'ion_cma_allocate':
drivers/staging/android/ion/ion_cma_heap.c:38:14: error: 'CONFIG_CMA_ALIGNMENT' undeclared (first use in this function)
  if (align > CONFIG_CMA_ALIGNMENT)
              ^

Caused by commit

  d98e6dbf42f7 ("staging: ion: Fix ion_cma_heap allocations")

CONFIG_CMA_ALIGNMENT (and CONFIG_DMA_CMA) is not set for this build.

I have reverted that commit for today.

-- 
Cheers,
Stephen Rothwell

Re: linux-next: build failure after merge of the staging.current tree

[Greg KH]

On Thu, Dec 14, 2017 at 05:00:33PM +1100, Stephen Rothwell wrote:
> Hi Greg,
> 
> After merging the staging.current tree, today's linux-next build (powerpc
> allyesconfig) failed like this:
> 
> drivers/staging/android/ion/ion_cma_heap.c: In function 'ion_cma_allocate':
> drivers/staging/android/ion/ion_cma_heap.c:38:14: error: 'CONFIG_CMA_ALIGNMENT' undeclared (first use in this function)
>   if (align > CONFIG_CMA_ALIGNMENT)
>               ^
> 
> Caused by commit
> 
>   d98e6dbf42f7 ("staging: ion: Fix ion_cma_heap allocations")
> 
> CONFIG_CMA_ALIGNMENT (and CONFIG_DMA_CMA) is not set for this build.
> 
> I have reverted that commit for today.

Thanks, and sorry about that.  John is working on a fix...

greg k-h

linux-next: build failure after merge of the staging.current tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 2194 bytes --]

Hi all,

After merging the staging.current tree, today's linux-next build (powerpc
allyesconfig) failed like this:

ld: drivers/staging/gpib/tnt4882/mite.o: in function `mite_dma_disarm':
mite.c:(.text+0x1240): multiple definition of `mite_dma_disarm'; drivers/comedi/drivers/mite.o:mite.c:(.text+0x1ac0): first defined here

Exposed by commit

  84b172cea4a2 ("staging: gpib: refer to correct config symbol in tnt4882 Makefile")

It is not, of course, the real cause of the build failure.  The duplicate
version of mite_dma_disarm() in drivers/staging/gpib/tnt4882/mite.c is
not actually used anywhere, so I applied the following patch for today.

From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Wed, 8 Jan 2025 21:07:23 +1100
Subject: [PATCH] fix for "staging: gpib: refer to correct config symbol in
 tnt4882 Makefile"

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
 drivers/staging/gpib/tnt4882/mite.c | 9 ---------
 drivers/staging/gpib/tnt4882/mite.h | 1 -
 2 files changed, 10 deletions(-)

diff --git a/drivers/staging/gpib/tnt4882/mite.c b/drivers/staging/gpib/tnt4882/mite.c
index 90341f8c394e..22035e53ec35 100644
--- a/drivers/staging/gpib/tnt4882/mite.c
+++ b/drivers/staging/gpib/tnt4882/mite.c
@@ -170,15 +170,6 @@ int mite_dma_tcr(struct mite_struct *mite)
 	return tcr;
 }
 
-void mite_dma_disarm(struct mite_struct *mite)
-{
-	int chor;
-
-	/* disarm */
-	chor = CHOR_ABORT;
-	writel(chor, mite->mite_io_addr + CHAN_OFFSET(0) + MITE_CHOR);
-}
-
 void mite_dump_regs(struct mite_struct *mite)
 {
 	void *addr = 0;
diff --git a/drivers/staging/gpib/tnt4882/mite.h b/drivers/staging/gpib/tnt4882/mite.h
index 2ecd83379974..dbfc59468228 100644
--- a/drivers/staging/gpib/tnt4882/mite.h
+++ b/drivers/staging/gpib/tnt4882/mite.h
@@ -64,7 +64,6 @@ void mite_list_devices(void);
 int mite_dma_tcr(struct mite_struct *mite);
 
 void mite_dma_arm(struct mite_struct *mite);
-void mite_dma_disarm(struct mite_struct *mite);
 
 void mite_dump_regs(struct mite_struct *mite);
 void mite_setregs(struct mite_struct *mite, unsigned long ll_start, int chan, int dir);
-- 
2.45.2

-- 
Cheers,
Stephen Rothwell

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: linux-next: build failure after merge of the staging.current tree

[Greg KH]

On Wed, Jan 08, 2025 at 09:42:40PM +1100, Stephen Rothwell wrote:
> Hi all,
> 
> After merging the staging.current tree, today's linux-next build (powerpc
> allyesconfig) failed like this:
> 
> ld: drivers/staging/gpib/tnt4882/mite.o: in function `mite_dma_disarm':
> mite.c:(.text+0x1240): multiple definition of `mite_dma_disarm'; drivers/comedi/drivers/mite.o:mite.c:(.text+0x1ac0): first defined here
> 
> Exposed by commit
> 
>   84b172cea4a2 ("staging: gpib: refer to correct config symbol in tnt4882 Makefile")
> 
> It is not, of course, the real cause of the build failure.  The duplicate
> version of mite_dma_disarm() in drivers/staging/gpib/tnt4882/mite.c is
> not actually used anywhere, so I applied the following patch for today.
> 
> From: Stephen Rothwell <sfr@canb.auug.org.au>
> Date: Wed, 8 Jan 2025 21:07:23 +1100
> Subject: [PATCH] fix for "staging: gpib: refer to correct config symbol in
>  tnt4882 Makefile"
> 
> Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
> ---
>  drivers/staging/gpib/tnt4882/mite.c | 9 ---------
>  drivers/staging/gpib/tnt4882/mite.h | 1 -
>  2 files changed, 10 deletions(-)
> 
> diff --git a/drivers/staging/gpib/tnt4882/mite.c b/drivers/staging/gpib/tnt4882/mite.c
> index 90341f8c394e..22035e53ec35 100644
> --- a/drivers/staging/gpib/tnt4882/mite.c
> +++ b/drivers/staging/gpib/tnt4882/mite.c
> @@ -170,15 +170,6 @@ int mite_dma_tcr(struct mite_struct *mite)
>  	return tcr;
>  }
>  
> -void mite_dma_disarm(struct mite_struct *mite)
> -{
> -	int chor;
> -
> -	/* disarm */
> -	chor = CHOR_ABORT;
> -	writel(chor, mite->mite_io_addr + CHAN_OFFSET(0) + MITE_CHOR);
> -}
> -
>  void mite_dump_regs(struct mite_struct *mite)
>  {
>  	void *addr = 0;
> diff --git a/drivers/staging/gpib/tnt4882/mite.h b/drivers/staging/gpib/tnt4882/mite.h
> index 2ecd83379974..dbfc59468228 100644
> --- a/drivers/staging/gpib/tnt4882/mite.h
> +++ b/drivers/staging/gpib/tnt4882/mite.h
> @@ -64,7 +64,6 @@ void mite_list_devices(void);
>  int mite_dma_tcr(struct mite_struct *mite);
>  
>  void mite_dma_arm(struct mite_struct *mite);
> -void mite_dma_disarm(struct mite_struct *mite);
>  
>  void mite_dump_regs(struct mite_struct *mite);
>  void mite_setregs(struct mite_struct *mite, unsigned long ll_start, int chan, int dir);
> -- 
> 2.45.2
> 
> -- 
> Cheers,
> Stephen Rothwell


Ah, I just beat you to it:
	https://lore.kernel.org/r/2025010809-padding-survive-91b3@gregkh

I'll queue this up later today, thanks!

greg k-h