next/master boot: 296 boots: 62 failed, 230 passed with 4 conflicts (next-20171113)

next/master boot: 296 boots: 62 failed, 230 passed with 4 conflicts (next-20171113)

[Guillaume Tucker]

On 13/11/17 19:02, kernelci.org bot wrote:
> next/master boot: 296 boots: 62 failed, 230 passed with 4 conflicts (next-20171113)
>
> Full Boot Summary: https://kernelci.org/boot/all/job/next/branch/master/kernel/next-20171113/
> Full Build Summary: https://kernelci.org/build/next/branch/master/kernel/next-20171113/
>
> Tree: next
> Branch: master
> Git Describe: next-20171113
> Git Commit: c348a99ee55feac43b5b62a5957c6d8e2b6c3abe
> Git URL: http://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git
> Tested: 52 unique boards, 17 SoC families, 33 builds out of 213
>
> Boot Regressions Detected:
>
> arm:
>
[...]
>     multi_v7_defconfig:
[...]
>         tegra124-nyan-big:
>             lab-collabora: failing since 9 days (last pass: next-20171102 - first fail: next-20171103)


There are several things failing on the tegra124-nyan-big, at
least I've isolated one with my latest bisection run:

6c78935777d12ead2d32adf3eb525a24faf02d04 is the first bad commit
commit 6c78935777d12ead2d32adf3eb525a24faf02d04
Author: Kees Cook <keescook@chromium.org>
Date:   Fri Nov 10 16:34:52 2017 +0100

     video: fbdev: Convert timers to use timer_setup()


These boot tests are with multi_v7_defconfig with CONFIG_MODULES
and CONFIG_DRM_NOUVEAU disabled:

* the first one on the revision mentioned above, fails:
   https://lava.collabora.co.uk/scheduler/job/987651

* then the same but with the commit reverted, passes:
   https://lava.collabora.co.uk/scheduler/job/987652

This was found using the still experimental automated bisection
tool for kernelci.org, I'm not sure yet how reliable the results
are.  I think the manual check with these 2 boots proves it, but
I haven't really investigated further than that.

So I then did the same tests at the top of the branch, on the
next-20171113 tag:

* next-20171113, fails but differently:
   https://lava.collabora.co.uk/scheduler/job/987653

* next-20171113 with the commit reverted, still fails:
   https://lava.collabora.co.uk/scheduler/job/987654

It's hard to tell whether the commit I mentioned got fixed
in-between, I guess it would be best to not have any boot failure
in any case.

The next thing I'll try is start a bisection for the other
failure see on next-20171113 tag, with the commit above reverted
and see if it leads anywhere...

Hope this helps!

Guillaume

> ---
> For more info write to <info@kernelci.org>
>
>
>
> _______________________________________________
> Kernel-build-reports mailing list
> Kernel-build-reports at lists.linaro.org
> https://lists.linaro.org/mailman/listinfo/kernel-build-reports
>

next/master boot: 296 boots: 62 failed, 230 passed with 4 conflicts (next-20171113)

[Kees Cook]

On Tue, Nov 14, 2017 at 9:10 AM, Guillaume Tucker
<guillaume.tucker@collabora.com> wrote:
> On 13/11/17 19:02, kernelci.org bot wrote:
>>
>> next/master boot: 296 boots: 62 failed, 230 passed with 4 conflicts
>> (next-20171113)
>>
>> Full Boot Summary:
>> https://kernelci.org/boot/all/job/next/branch/master/kernel/next-20171113/
>> Full Build Summary:
>> https://kernelci.org/build/next/branch/master/kernel/next-20171113/
>>
>> Tree: next
>> Branch: master
>> Git Describe: next-20171113
>> Git Commit: c348a99ee55feac43b5b62a5957c6d8e2b6c3abe
>> Git URL:
>> http://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git
>> Tested: 52 unique boards, 17 SoC families, 33 builds out of 213
>>
>> Boot Regressions Detected:
>>
>> arm:
>>
> [...]
>>
>>     multi_v7_defconfig:
>
> [...]
>>
>>         tegra124-nyan-big:
>>             lab-collabora: failing since 9 days (last pass: next-20171102
>> - first fail: next-20171103)
>
>
>
> There are several things failing on the tegra124-nyan-big, at
> least I've isolated one with my latest bisection run:
>
> 6c78935777d12ead2d32adf3eb525a24faf02d04 is the first bad commit
> commit 6c78935777d12ead2d32adf3eb525a24faf02d04
> Author: Kees Cook <keescook@chromium.org>
> Date:   Fri Nov 10 16:34:52 2017 +0100
>
>     video: fbdev: Convert timers to use timer_setup()
>
>
> These boot tests are with multi_v7_defconfig with CONFIG_MODULES
> and CONFIG_DRM_NOUVEAU disabled:
>
> * the first one on the revision mentioned above, fails:
>   https://lava.collabora.co.uk/scheduler/job/987651
>
> * then the same but with the commit reverted, passes:
>   https://lava.collabora.co.uk/scheduler/job/987652
>
> This was found using the still experimental automated bisection
> tool for kernelci.org, I'm not sure yet how reliable the results
> are.  I think the manual check with these 2 boots proves it, but
> I haven't really investigated further than that.
>
> So I then did the same tests at the top of the branch, on the
> next-20171113 tag:
>
> * next-20171113, fails but differently:
>   https://lava.collabora.co.uk/scheduler/job/987653
>
> * next-20171113 with the commit reverted, still fails:
>   https://lava.collabora.co.uk/scheduler/job/987654
>
> It's hard to tell whether the commit I mentioned got fixed
> in-between, I guess it would be best to not have any boot failure
> in any case.
>
> The next thing I'll try is start a bisection for the other
> failure see on next-20171113 tag, with the commit above reverted
> and see if it leads anywhere...
>
> Hope this helps!

Thanks for the report! I believe this is fixed with:

https://marc.info/?l=linux-fbdev&m=151056635200583&w=2

and that commit appears in next-20171114. Please let me know if that
doesn't fix it, though!

-Kees

>
> Guillaume
>
>> ---
>> For more info write to <info@kernelci.org>
>>
>>
>>
>> _______________________________________________
>> Kernel-build-reports mailing list
>> Kernel-build-reports at lists.linaro.org
>> https://lists.linaro.org/mailman/listinfo/kernel-build-reports
>>
>



-- 
Kees Cook
Pixel Security

next/master boot: 296 boots: 62 failed, 230 passed with 4 conflicts (next-20171113)

[Guillaume Tucker]

On 14/11/17 19:53, Kees Cook wrote:
> On Tue, Nov 14, 2017 at 9:10 AM, Guillaume Tucker
> <guillaume.tucker@collabora.com> wrote:
>> On 13/11/17 19:02, kernelci.org bot wrote:
>>>
>>> next/master boot: 296 boots: 62 failed, 230 passed with 4 conflicts
>>> (next-20171113)
>>>
>>> Full Boot Summary:
>>> https://kernelci.org/boot/all/job/next/branch/master/kernel/next-20171113/
>>> Full Build Summary:
>>> https://kernelci.org/build/next/branch/master/kernel/next-20171113/
>>>
>>> Tree: next
>>> Branch: master
>>> Git Describe: next-20171113
>>> Git Commit: c348a99ee55feac43b5b62a5957c6d8e2b6c3abe
>>> Git URL:
>>> http://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git
>>> Tested: 52 unique boards, 17 SoC families, 33 builds out of 213
>>>
>>> Boot Regressions Detected:
>>>
>>> arm:
>>>
>> [...]
>>>
>>>     multi_v7_defconfig:
>>
>> [...]
>>>
>>>         tegra124-nyan-big:
>>>             lab-collabora: failing since 9 days (last pass: next-20171102
>>> - first fail: next-20171103)
>>
>>
>>
>> There are several things failing on the tegra124-nyan-big, at
>> least I've isolated one with my latest bisection run:
>>
>> 6c78935777d12ead2d32adf3eb525a24faf02d04 is the first bad commit
>> commit 6c78935777d12ead2d32adf3eb525a24faf02d04
>> Author: Kees Cook <keescook@chromium.org>
>> Date:   Fri Nov 10 16:34:52 2017 +0100
>>
>>     video: fbdev: Convert timers to use timer_setup()
>>
>>
>> These boot tests are with multi_v7_defconfig with CONFIG_MODULES
>> and CONFIG_DRM_NOUVEAU disabled:
>>
>> * the first one on the revision mentioned above, fails:
>>   https://lava.collabora.co.uk/scheduler/job/987651
>>
>> * then the same but with the commit reverted, passes:
>>   https://lava.collabora.co.uk/scheduler/job/987652
>>
>> This was found using the still experimental automated bisection
>> tool for kernelci.org, I'm not sure yet how reliable the results
>> are.  I think the manual check with these 2 boots proves it, but
>> I haven't really investigated further than that.
>>
>> So I then did the same tests at the top of the branch, on the
>> next-20171113 tag:
>>
>> * next-20171113, fails but differently:
>>   https://lava.collabora.co.uk/scheduler/job/987653
>>
>> * next-20171113 with the commit reverted, still fails:
>>   https://lava.collabora.co.uk/scheduler/job/987654
>>
>> It's hard to tell whether the commit I mentioned got fixed
>> in-between, I guess it would be best to not have any boot failure
>> in any case.
>>
>> The next thing I'll try is start a bisection for the other
>> failure see on next-20171113 tag, with the commit above reverted
>> and see if it leads anywhere...
>>
>> Hope this helps!
>
> Thanks for the report! I believe this is fixed with:
>
> https://marc.info/?l=linux-fbdev&m=151056635200583&w=2
>
> and that commit appears in next-20171114. Please let me know if that
> doesn't fix it, though!

Thanks, I've applied this patch in-place and confirmed it did fix
the issue:

   https://lava.collabora.co.uk/scheduler/job/988891

Will move on to next-20171114 now.

Guillaume

>>> ---
>>> For more info write to <info@kernelci.org>
>>>
>>>
>>>
>>> _______________________________________________
>>> Kernel-build-reports mailing list
>>> Kernel-build-reports at lists.linaro.org
>>> https://lists.linaro.org/mailman/listinfo/kernel-build-reports
>>>
>>
>
>
>