[Lts-dev] 4.4.96/27e68f1b: regressions detected in project lkft/linux-stable-4.4-oe

Milosz Wasilewski milosz.wasilewski at linaro.org
Thu Nov 2 21:14:42 UTC 2017


On 2 November 2017 at 18:19, Greg KH <gregkh at google.com> wrote:
> On Thu, Nov 02, 2017 at 12:54:55PM +0000, Linaro QA wrote:
>> Summary
>> ------------------------------------------------------------------------
>>
>> kernel: 4.4.96
>> git repo: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable.git
>> git branch: linux-4.4.y
>> git commit: 27e68f1bca91a7f69808c8641f03c66cca2bb365
>> git describe: v4.4.96
>> Test details: https://qa-reports.linaro.org/lkft/linux-stable-4.4-oe/build/v4.4.96
>>
>> Regressions (compared to build v4.4.95)
>
> Is this real?

On the first look they're nothing to worry about. All look like timing
issue due to poorer than expected performance. On the other hand, they
didn't show up before (getrandom02, perf_event_open02). I know Naresh
did some research on getrandom02 and it doesn't fail always:
https://bugs.linaro.org/show_bug.cgi?id=3401#c1. I suspect it's the
same for other x15 cases.

The leapsec_timer case is most likely the setup issue - some other
process than the test updates the clock.

milosz


More information about the Kernel-build-reports mailing list