Re: Regression notification from "TCWG CI"

On 2021-12-20 08:01, Jan Vrany via Gdb wrote:
> Hi there,
> 
> I've got two messages from "TCWG CI" about regression caused two
> of "my" commits (see one such message below).
> 
> I looked around trying to figure out what's the problem but could not
> find anything relevant. The closest indication of some error I found
> is:
> 
> --
> + ./jenkins-scripts/../bmk-scripts/merge-metric-csvs.py /home/tcwg-
> buildslave/workspace/tcwg_bmk_2/artifacts/build-
> f76d800be844dd2aa4aa8f189a3ace16c5e931bc/09-check_regression/results-
> perf-0.csv /home/tcwg-buildslave/workspace/tcwg_bmk_2/artifacts/build-
> f76d800be844dd2aa4aa8f189a3ace16c5e931bc/09-check_regression/results-
> vect-0.csv
> Traceback (most recent call last):
>   File "./jenkins-scripts/../bmk-scripts/merge-metric-csvs.py", line
> 17, in <module>
>     main()
>   File "./jenkins-scripts/../bmk-scripts/merge-metric-csvs.py", line
> 13, in main
>     result = pd.merge(df1, df2, on=["benchmark", "symbol"])
> ...
> --
> in
> https://ci.linaro.org/job/tcwg_bmk_ci_gnu_eabi-bisect-tcwg_bmk_stm32-gnu_eabi-master-arm_eabi-coremark-O3/14/artifact/artifacts/build-f76d800be844dd2aa4aa8f189a3ace16c5e931bc/09-check_regression/results.log
> 
> but this looks to me more like an error is CI job than error in the
> commit itself.
> 
> When following "repoduce steps" from that email, I got another error:
> 
> --
> ERROR: required parameter ssh_host not set
> --
> 
> 
> Could please someone point me to where I find out what's the
> regression?

It looks like a CI configuration problem, I would ignore it.  For the
record, I got this message a few times when I broke the build, and they
were correct all the times.  So so far it just looks like  blip.

Simon