diff mbox series

ARM: s3c: Fix a build error after the s3c24xx dma driver was removed

Message ID 20221118215401.505480-1-u.kleine-koenig@pengutronix.de
State New
Headers show
Series ARM: s3c: Fix a build error after the s3c24xx dma driver was removed | expand

Commit Message

Uwe Kleine-König Nov. 18, 2022, 9:54 p.m. UTC
The linux/platform_data/dma-s3c24xx.h header file was removed. It didn't
declare or define any symbol needed in devs.c though, so the #include
can just be dropped.

Fixes: cccc46ae3623 ("dmaengine: remove s3c24xx driver")
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
---
 arch/arm/mach-s3c/devs.c | 1 -
 1 file changed, 1 deletion(-)

Comments

Krzysztof Kozlowski Nov. 20, 2022, 10:31 a.m. UTC | #1
On 18/11/2022 22:54, Uwe Kleine-König wrote:
> The linux/platform_data/dma-s3c24xx.h header file was removed. It didn't
> declare or define any symbol needed in devs.c though, so the #include
> can just be dropped.
> 
> Fixes: cccc46ae3623 ("dmaengine: remove s3c24xx driver")
> Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
> ---

The file was not removed... or it should not have been yet. The s3c24xx
dma driver removal should be part of Arnd series taken via SoC ARM.

Best regards,
Krzysztof
Krzysztof Kozlowski Nov. 20, 2022, 11:22 a.m. UTC | #2
On 20/11/2022 11:31, Krzysztof Kozlowski wrote:
> On 18/11/2022 22:54, Uwe Kleine-König wrote:
>> The linux/platform_data/dma-s3c24xx.h header file was removed. It didn't
>> declare or define any symbol needed in devs.c though, so the #include
>> can just be dropped.
>>
>> Fixes: cccc46ae3623 ("dmaengine: remove s3c24xx driver")
>> Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
>> ---
> 
> The file was not removed... or it should not have been yet. The s3c24xx
> dma driver removal should be part of Arnd series taken via SoC ARM.

Also: I don't think the fix is correct. The header was used in mach-s3c,
so removing the header should still cause broken build.

It also fails in other places:

arch/arm/mach-s3c/s3c24xx.c:21:10: fatal error:
linux/platform_data/dma-s3c24xx.h: No such file or directory
   21 | #include <linux/platform_data/dma-s3c24xx.h>
      |          ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
compilation terminated.

I think that commit should be just dropped instead.

Best regards,
Krzysztof
Uwe Kleine-König Nov. 25, 2022, 8:51 a.m. UTC | #3
Hello,

On Sun, Nov 20, 2022 at 12:22:31PM +0100, Krzysztof Kozlowski wrote:
> On 20/11/2022 11:31, Krzysztof Kozlowski wrote:
> > On 18/11/2022 22:54, Uwe Kleine-König wrote:
> >> The linux/platform_data/dma-s3c24xx.h header file was removed. It didn't
> >> declare or define any symbol needed in devs.c though, so the #include
> >> can just be dropped.
> >>
> >> Fixes: cccc46ae3623 ("dmaengine: remove s3c24xx driver")
> >> Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
> >> ---
> > 
> > The file was not removed... or it should not have been yet. The s3c24xx
> > dma driver removal should be part of Arnd series taken via SoC ARM.

The patch enters next with the merge of

	git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/dmaengine.git next

Ah, the patch that became cccc46ae3623 (i.e. patch #14) is part of a
bigger series. Its patch #1 removes s3c24xx.c (which you pointed out to be still
broken) and patch #2 includes the change I suggested here.

> I think that commit should be just dropped instead.

+1

BTW, cccc46ae3623 is included in next since next-20221107 and breaks
(at least) arm/s3c2410_defconfig. So I would consider reverting
cccc46ae3623 a fix. (Added linux-next to Cc:)

Best regards
Uwe
Uwe Kleine-König Dec. 2, 2022, 11:20 a.m. UTC | #4
Hello,

I made the subject a bit more expressive.

On Fri, Nov 25, 2022 at 09:52:51AM +0100, Krzysztof Kozlowski wrote:
> On 25/11/2022 09:51, Uwe Kleine-König wrote:
> > On Sun, Nov 20, 2022 at 12:22:31PM +0100, Krzysztof Kozlowski wrote:
> >> On 20/11/2022 11:31, Krzysztof Kozlowski wrote:
> >>> On 18/11/2022 22:54, Uwe Kleine-König wrote:
> >>>> The linux/platform_data/dma-s3c24xx.h header file was removed. It didn't
> >>>> declare or define any symbol needed in devs.c though, so the #include
> >>>> can just be dropped.
> >>>>
> >>>> Fixes: cccc46ae3623 ("dmaengine: remove s3c24xx driver")
> >>>> Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
> >>>> ---
> >>>
> >>> The file was not removed... or it should not have been yet. The s3c24xx
> >>> dma driver removal should be part of Arnd series taken via SoC ARM.
> > 
> > The patch enters next with the merge of
> > 
> > 	git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/dmaengine.git next
> > 
> > Ah, the patch that became cccc46ae3623 (i.e. patch #14) is part of a
> > bigger series. Its patch #1 removes s3c24xx.c (which you pointed out to be still
> > broken) and patch #2 includes the change I suggested here.
> > 
> >> I think that commit should be just dropped instead.
> > 
> > +1
> > 
> > BTW, cccc46ae3623 is included in next since next-20221107 and breaks
> > (at least) arm/s3c2410_defconfig. So I would consider reverting
> > cccc46ae3623 a fix. (Added linux-next to Cc:)
> 
> Yes. The build failure of next was reported already by kernel test robot.
> 
> Vinod, can we drop this patch?

That patch is still contained in the next tags of this week. :-(

Best regards
Uwe
Uwe Kleine-König Dec. 5, 2022, 8:56 a.m. UTC | #5
Hello Vinod,

On Fri, Dec 02, 2022 at 05:12:53PM +0530, Vinod Koul wrote:
> On 02-12-22, 12:20, Uwe Kleine-König wrote:
> > I made the subject a bit more expressive.

I wonder what I intended to write there, something like "Regression in
dmaengine/next in arm/s3c2410_defconfig", sorry to make it look as if
you already fixed the regression :-)

> > On Fri, Nov 25, 2022 at 09:52:51AM +0100, Krzysztof Kozlowski wrote:
> > > On 25/11/2022 09:51, Uwe Kleine-König wrote:
> > > > On Sun, Nov 20, 2022 at 12:22:31PM +0100, Krzysztof Kozlowski wrote:
> > > >> On 20/11/2022 11:31, Krzysztof Kozlowski wrote:
> > > >>> On 18/11/2022 22:54, Uwe Kleine-König wrote:
> > > >>>> The linux/platform_data/dma-s3c24xx.h header file was removed. It didn't
> > > >>>> declare or define any symbol needed in devs.c though, so the #include
> > > >>>> can just be dropped.
> > > >>>>
> > > >>>> Fixes: cccc46ae3623 ("dmaengine: remove s3c24xx driver")
> > > >>>> Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
> > > >>>> ---
> > > >>>
> > > >>> The file was not removed... or it should not have been yet. The s3c24xx
> > > >>> dma driver removal should be part of Arnd series taken via SoC ARM.
> > > > 
> > > > The patch enters next with the merge of
> > > > 
> > > > 	git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/dmaengine.git next
> > > > 
> > > > Ah, the patch that became cccc46ae3623 (i.e. patch #14) is part of a
> > > > bigger series. Its patch #1 removes s3c24xx.c (which you pointed out to be still
> > > > broken) and patch #2 includes the change I suggested here.
> > > > 
> > > >> I think that commit should be just dropped instead.
> > > > 
> > > > +1
> > > > 
> > > > BTW, cccc46ae3623 is included in next since next-20221107 and breaks
> > > > (at least) arm/s3c2410_defconfig. So I would consider reverting
> > > > cccc46ae3623 a fix. (Added linux-next to Cc:)
> > > 
> > > Yes. The build failure of next was reported already by kernel test robot.
> > > 
> > > Vinod, can we drop this patch?
> > 
> > That patch is still contained in the next tags of this week. :-(
> 
> Apologies for delay, I was moving...
> 
> I have reverted and pushed now

Thanks!

Best regards
Uwe
diff mbox series

Patch

diff --git a/arch/arm/mach-s3c/devs.c b/arch/arm/mach-s3c/devs.c
index 9ac07c023adf..a31d1c3038e8 100644
--- a/arch/arm/mach-s3c/devs.c
+++ b/arch/arm/mach-s3c/devs.c
@@ -29,7 +29,6 @@ 
 #include <linux/sizes.h>
 #include <linux/platform_data/s3c-hsudc.h>
 #include <linux/platform_data/s3c-hsotg.h>
-#include <linux/platform_data/dma-s3c24xx.h>
 
 #include <linux/platform_data/media/s5p_hdmi.h>