The header files linux/mem_encrypt.h is included twice in svm.c,
so one inclusion of each can be removed.

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=11750
Signed-off-by: Yang Li <yang.lee@linux.alibaba.com>
---
 arch/powerpc/platforms/pseries/svm.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/arch/powerpc/platforms/pseries/svm.c b/arch/powerpc/platforms/pseries/svm.c
index c5d0f92c7969..384c9dc1899a 100644
--- a/arch/powerpc/platforms/pseries/svm.c
+++ b/arch/powerpc/platforms/pseries/svm.c
@@ -10,7 +10,6 @@
 #include <linux/memblock.h>
 #include <linux/mem_encrypt.h>
 #include <linux/cc_platform.h>
-#include <linux/mem_encrypt.h>
 #include <asm/machdep.h>
 #include <asm/svm.h>
 #include <asm/swiotlb.h>
-- 
2.32.0.3.g01195cf9f

Yang Li <yang.lee@linux.alibaba.com> writes:
> The header files linux/mem_encrypt.h is included twice in svm.c,
> so one inclusion of each can be removed.
>
> Reported-by: Abaci Robot <abaci@linux.alibaba.com>
> Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=11750
> Signed-off-by: Yang Li <yang.lee@linux.alibaba.com>
> ---
>  arch/powerpc/platforms/pseries/svm.c | 1 -
>  1 file changed, 1 deletion(-)

The two includes only appear in linux-next, and they both come from
different trees. They are required in each tree to avoid breaking the
build.

So no one can merge this patch until the two trees are merged into mainline.

cheers

> diff --git a/arch/powerpc/platforms/pseries/svm.c b/arch/powerpc/platforms/pseries/svm.c
> index c5d0f92c7969..384c9dc1899a 100644
> --- a/arch/powerpc/platforms/pseries/svm.c
> +++ b/arch/powerpc/platforms/pseries/svm.c
> @@ -10,7 +10,6 @@
>  #include <linux/memblock.h>
>  #include <linux/mem_encrypt.h>
>  #include <linux/cc_platform.h>
> -#include <linux/mem_encrypt.h>
>  #include <asm/machdep.h>
>  #include <asm/svm.h>
>  #include <asm/swiotlb.h>
> -- 
> 2.32.0.3.g01195cf9f

Le 07/11/2024 à 12:29, Michael Ellerman a écrit :
> Yang Li <yang.lee@linux.alibaba.com> writes:
>> The header files linux/mem_encrypt.h is included twice in svm.c,
>> so one inclusion of each can be removed.
>>
>> Reported-by: Abaci Robot <abaci@linux.alibaba.com>
>> Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=11750
>> Signed-off-by: Yang Li <yang.lee@linux.alibaba.com>
>> ---
>>   arch/powerpc/platforms/pseries/svm.c | 1 -
>>   1 file changed, 1 deletion(-)
> 
> The two includes only appear in linux-next, and they both come from
> different trees. They are required in each tree to avoid breaking the
> build.
> 
> So no one can merge this patch until the two trees are merged into mainline.

But can't those two trees coordinate the patches so that the include 
goes at the same place avoiding duplication at merge ?

Christophe

Christophe Leroy <christophe.leroy@csgroup.eu> writes:
> Le 07/11/2024 à 12:29, Michael Ellerman a écrit :
>> Yang Li <yang.lee@linux.alibaba.com> writes:
>>> The header files linux/mem_encrypt.h is included twice in svm.c,
>>> so one inclusion of each can be removed.
>>>
>>> Reported-by: Abaci Robot <abaci@linux.alibaba.com>
>>> Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=11750
>>> Signed-off-by: Yang Li <yang.lee@linux.alibaba.com>
>>> ---
>>>   arch/powerpc/platforms/pseries/svm.c | 1 -
>>>   1 file changed, 1 deletion(-)
>> 
>> The two includes only appear in linux-next, and they both come from
>> different trees. They are required in each tree to avoid breaking the
>> build.
>> 
>> So no one can merge this patch until the two trees are merged into mainline.
>
> But can't those two trees coordinate the patches so that the include 
> goes at the same place avoiding duplication at merge ?

Yes that would work.

Except that in this case it's too late because the commits have already
been applied to both trees for over a week - neither maintainer is going
to want to rebase for something trivial like a duplicated header.

I could apply a patch to my tree to move the include to the same line as
the commit in the DMA tree, but even that seems like overkill for a
duplicated header.

I'll try and remember to apply this once the trees are merged in
mainline. But if not the bot that detected it in the first place can
just detect it again and repost.

cheers