Re: stable-rc build: 0 warnings 1 failures (stable-rc/v4.9.53-105-g0e59436)

Re: stable-rc build: 0 warnings 1 failures (stable-rc/v4.9.53-105-g0e59436)

[Arnd Bergmann]

On Fri, Oct 6, 2017 at 12:03 PM, Olof's autobuilder <build@lixom.net> wrote:
> Here are the build results from automated periodic testing.
>
> The tree being built was stable-rc, found at:
>
> https://git.kernel.org/cgit/linux/kernel/git/stable/linux-stable-rc.git/
>
> Topmost commit:
>
> 0e59436 Linux 4.9.54-rc1

>
>         arm64.allmodconfig:
> /work/build/batch/arch/arm64/mm/kasan_init.c:156:95: error: implicit declaration of function 'lm_alias'; did you mean 'mm_alloc'? [-Werror=implicit-function-declaration]

Caused by

e773d137b255 ("arm64: kasan: avoid bad virt_to_pfn()")
(backported from b0de0ccc8b9ed)

Presumably requires a backport of

568c5fe5a54f ("mm: Introduce lm_alias")

Mark or Laura, can you confirm that this is the correct fix?
I see additional uses of lm_alias in mainline, but only this one
backported to stable, not sure if only this one is required there,
or we want the others as well.

       Arnd

Re: stable-rc build: 0 warnings 1 failures (stable-rc/v4.9.53-105-g0e59436)

[Laura Abbott]

On 10/06/2017 04:27 AM, Arnd Bergmann wrote:
> On Fri, Oct 6, 2017 at 12:03 PM, Olof's autobuilder <build@lixom.net> wrote:
>> Here are the build results from automated periodic testing.
>>
>> The tree being built was stable-rc, found at:
>>
>> https://git.kernel.org/cgit/linux/kernel/git/stable/linux-stable-rc.git/
>>
>> Topmost commit:
>>
>> 0e59436 Linux 4.9.54-rc1
> 
>>
>>         arm64.allmodconfig:
>> /work/build/batch/arch/arm64/mm/kasan_init.c:156:95: error: implicit declaration of function 'lm_alias'; did you mean 'mm_alloc'? [-Werror=implicit-function-declaration]
> 
> Caused by
> 
> e773d137b255 ("arm64: kasan: avoid bad virt_to_pfn()")
> (backported from b0de0ccc8b9ed)
> 
> Presumably requires a backport of
> 
> 568c5fe5a54f ("mm: Introduce lm_alias")
> 
> Mark or Laura, can you confirm that this is the correct fix?
> I see additional uses of lm_alias in mainline, but only this one
> backported to stable, not sure if only this one is required there,
> or we want the others as well.
> 
>        Arnd
> 


I don't think backporting lm_alias is the correct fix. That was
introduced as part of DEBUG_VIRTUAL for arm64 which isn't present
at all for 4.9. If DEBUG_VIRTUAL isn't present, there should be
no need for this fix since it's handled by existing code. I
think the patch should be dropped assuming there isn't another
issue with kasan.

Thanks,
Laura

Re: stable-rc build: 0 warnings 1 failures (stable-rc/v4.9.53-105-g0e59436)

[gregkh]

On Fri, Oct 06, 2017 at 08:14:40AM -0700, Laura Abbott wrote:
> On 10/06/2017 04:27 AM, Arnd Bergmann wrote:
> > On Fri, Oct 6, 2017 at 12:03 PM, Olof's autobuilder <build@lixom.net> wrote:
> >> Here are the build results from automated periodic testing.
> >>
> >> The tree being built was stable-rc, found at:
> >>
> >> https://git.kernel.org/cgit/linux/kernel/git/stable/linux-stable-rc.git/
> >>
> >> Topmost commit:
> >>
> >> 0e59436 Linux 4.9.54-rc1
> > 
> >>
> >>         arm64.allmodconfig:
> >> /work/build/batch/arch/arm64/mm/kasan_init.c:156:95: error: implicit declaration of function 'lm_alias'; did you mean 'mm_alloc'? [-Werror=implicit-function-declaration]
> > 
> > Caused by
> > 
> > e773d137b255 ("arm64: kasan: avoid bad virt_to_pfn()")
> > (backported from b0de0ccc8b9ed)
> > 
> > Presumably requires a backport of
> > 
> > 568c5fe5a54f ("mm: Introduce lm_alias")
> > 
> > Mark or Laura, can you confirm that this is the correct fix?
> > I see additional uses of lm_alias in mainline, but only this one
> > backported to stable, not sure if only this one is required there,
> > or we want the others as well.
> > 
> >        Arnd
> > 
> 
> 
> I don't think backporting lm_alias is the correct fix. That was
> introduced as part of DEBUG_VIRTUAL for arm64 which isn't present
> at all for 4.9. If DEBUG_VIRTUAL isn't present, there should be
> no need for this fix since it's handled by existing code. I
> think the patch should be dropped assuming there isn't another
> issue with kasan.

I've just dropped it now, sorry for this.

thanks,

greg k-h