1. Messages
  2. Some questions about Azure CI

Re: Some questions about Azure CI

Michael Kubacki
 

My understanding is that you were referring to the document on my fork being difficult to access and you shared a link to your CI results that does not have the message pointing to the wiki page yet.

This is all correct. The wiki content has moved from my fork to the official wiki page and the patch that adds the message to the CI results has not been merged yet. It is posted on the mailing list here and waiting for reviews:

https://edk2.groups.io/g/devel/message/89098

---

As for your build failing between yesterday and today, note that the builds are based on different commits.

Yesterday:
https://github.com/loongson/edk2/commit/8d100910a2ad98b71acaabc0dcdd70be7e0eba63

Today:
https://github.com/loongson/edk2/commit/6ca5d9d6c55ca773a652fe52508e980c7aa372d3

The commits have different content. In particular, I took a quick look at a build that is failing now (Build_GCC5 TARGET_MDEMODULE_DEBUG) but previously succeeded.

The compilation step is failing for reasons such as the following:

INFO - /home/vsts/work/1/s/MdePkg/Include/LoongArch64/ProcessorBind.h:35:24: error: expected ‘=’, ‘,’, ‘;’, ‘asm’ or ‘__attribute__’ before ‘UINT16’
INFO - 35 | typedef unsigned shor UINT16;

Notice, that there is a typo introduced in the new commit. "shor" versus "short", this is what is causing compilation failure.

Here is a comparison of those two commits showing the line in ProcessorBind.h that is causing the problem:

https://github.com/loongson/edk2/compare/8d10091..6ca5d9d#diff-50b24eb92c5785d5c70097a70104040fcacfb70225dd8196c1a97ef3bbb305c8R35-R36

Regards,
Michael

On 4/20/2022 7:08 AM, Chao Li wrote:
Hi Micheal,
Do you know if anyone has modified the Azure CI today? I did a CI test today, and many modules reported errors, but yesterday, only two modules reported errors, I don't know what happened. Please refer following two links:
Today:
https://dev.azure.com/kilaterlee/LoongArch_edk2/_build/results?buildId=32&view=results <https://link.getmailspring.com/link/3FB8CF0A-BF1B-4A0A-924B-3B5F6D7EB834@getmailspring.com/0?redirect=https%3A%2F%2Fdev.azure.com%2Fkilaterlee%2FLoongArch_edk2%2F_build%2Fresults%3FbuildId%3D32%26view%3Dresults&recipient=bWlrdWJhY2tAbGludXgubWljcm9zb2Z0LmNvbQ%3D%3D>
Yesterday:
https://dev.azure.com/kilaterlee/LoongArch_edk2/_build/results?buildId=31&view=results <https://link.getmailspring.com/link/3FB8CF0A-BF1B-4A0A-924B-3B5F6D7EB834@getmailspring.com/1?redirect=https%3A%2F%2Fdev.azure.com%2Fkilaterlee%2FLoongArch_edk2%2F_build%2Fresults%3FbuildId%3D31%26view%3Dresults&recipient=bWlrdWJhY2tAbGludXgubWljcm9zb2Z0LmNvbQ%3D%3D>
--
Thanks,
Chao
------------------------
On 4月 20 2022, at 6:10 晚上, Chao Li <lichao@...> wrote:
Hi Micheal,
Sorry for late reply, I'm busy with other things, focusing on
modifying our code, refer to the detailed error log file.
I am learning how to get the detailed log from this link:
https://github.com/makubacki/tianocore.github.io/blob/add_ci_uncrustify_instructions/EDK-II-Code-Formatting.md#how-to-find-uncrustify-formatting-errors-in-continuous-integration-ci
<https://link.getmailspring.com/link/3FB8CF0A-BF1B-4A0A-924B-3B5F6D7EB834@getmailspring.com/2?redirect=https%3A%2F%2Flink.getmailspring.com%2Flink%2FCA5CE175-A392-4D5E-9BE9-392B7A97BB32%40getmailspring.com%2F0%3Fredirect%3Dhttps%253A%252F%252Fgithub.com%252Fmakubacki%252Ftianocore.github.io%252Fblob%252Fadd_ci_uncrustify_instructions%252FEDK-II-Code-Formatting.md%2523how-to-find-uncrustify-formatting-errors-in-continuous-integration-ci%26recipient%3DZGV2ZWxAZWRrMi5ncm91cHMuaW8%253D&recipient=bWlrdWJhY2tAbGludXgubWljcm9zb2Z0LmNvbQ%3D%3D>.
I didn't find the detailed log link in the Azure error log, it looks
like:
https://dev.azure.com/kilaterlee/LoongArch_edk2/_build/results?buildId=31&view=logs&j=9701361e-2546-5093-77e5-58c75d95b7ce&t=27131ef8-75e1-57c3-3236-d806181f2f1f
<https://link.getmailspring.com/link/3FB8CF0A-BF1B-4A0A-924B-3B5F6D7EB834@getmailspring.com/3?redirect=https%3A%2F%2Flink.getmailspring.com%2Flink%2FCA5CE175-A392-4D5E-9BE9-392B7A97BB32%40getmailspring.com%2F1%3Fredirect%3Dhttps%253A%252F%252Fdev.azure.com%252Fkilaterlee%252FLoongArch_edk2%252F_build%252Fresults%253FbuildId%253D31%2526view%253Dlogs%2526j%253D9701361e-2546-5093-77e5-58c75d95b7ce%2526t%253D27131ef8-75e1-57c3-3236-d806181f2f1f%26recipient%3DZGV2ZWxAZWRrMi5ncm91cHMuaW8%253D&recipient=bWlrdWJhY2tAbGludXgubWljcm9zb2Z0LmNvbQ%3D%3D>.
--
Thanks,
Chao
------------------------
On 4月 20 2022, at 9:56 上午, "Michael Kubacki"
<mikuback@...> wrote:
Hi Chao,
Are you referring to the change I made to put the link to the
instructions in the build log. Here's an example of what that
will look
like:
https://github.com/makubacki/tianocore.github.io/blob/temp_show_uncrustify_ci_link_in_build_log/images/edk-ii-code-formatting/temp-example-instruction-link-in-build-log.png?raw=true
(some of the words running together are not in the actual change)
I'm open to moving it elsewhere but developers will need to know
how to
at least find that or they will have trouble fixing build issues.
Regards,
Michael
On 4/19/2022 9:46 PM, Chao Li wrote:
> Hi Micheal,
> Using the way your provided the link, I found the detail log,
thanks,
> but this way is not friendly to newbies, because they don't
know how to
> find the tutorial.:)
>
> --
> Thanks,
> Chao
> ------------------------
>
>
> On 4月 20 2022, at 7:43 早上, "Michael Kubacki"
> <mikuback@...> wrote:
>
> The wiki update is here:
>
https://github.com/makubacki/tianocore.github.io/blob/add_ci_uncrustify_instructions/EDK-II-Code-Formatting.md#how-to-find-uncrustify-formatting-errors-in-continuous-integration-ci
>
> I am waiting for permissions to be updated so I can update
the actual
> wiki page and then I'll send a patch for the plugin change
with the link
> to that section.
>
> Regards,
> Michael
>
> On 4/19/2022 6:15 PM, Michael Kubacki wrote:
> > I agree that is ideal but I'm not aware of how to do that
at the
> moment.
> >
> > Thanks,
> > Michael
> >
> > On 4/19/2022 4:52 PM, Michael D Kinney wrote:
> >> Hi Michael,
> >>
> >> Can the build log provide a direct link to the log file
artifact
> if an
> >> error is detected?
> >>
> >> Mike
> >>
> >>> -----Original Message-----
> >>> From: devel@edk2.groups.io <devel@edk2.groups.io> On
Behalf Of
> >>> Michael Kubacki
> >>> Sent: Tuesday, April 19, 2022 12:42 PM
> >>> To: devel@edk2.groups.io; quic_rcran@...;
> >>> spbrogan@...; Chao Li <lichao@...>
> >>> Subject: Re: [edk2-devel] Some questions about Azure CI
> >>>
> >>> Hi Rebecca,
> >>>
> >>> We are trying to keep the results reporting experience
> consistent with
> >>> other plugins and prevent an overwhelming amount of
information
> being
> >>> printed to the build log.
> >>>
> >>> In case other errors are present, providing high-level
> information from
> >>> each plugin can help point the user in the right direction to
> get more
> >>> detail.
> >>>
> >>> I completely understand the concern though, so I'm
planning to make
> >>> another step toward providing more information about how to
> debug an
> >>> issue, when it occurs. That is to put the step-by-step
> information about
> >>> where to find the file diff into the EDK II Code Formatting
> wiki page
> >>> and then if a failure occurs, print a link to that section of
> the wiki
> >>> page. I am hoping this will provide sufficient information to
> get to the
> >>> file diff at the point of failure.
> >>>
> >>> I should be able to send these patches later today.
> >>>
> >>> Regards,
> >>> Michael
> >>>
> >>> On 4/19/2022 12:43 PM, Rebecca Cran wrote:
> >>>> Since people are going to keep running into this, could we
> just output
> >>>> the file diff to the console? That would avoid having to go
> hunting for
> >>>> the log file.
> >>>>
> >>>
> >>>
> >>>
> >>>
> >>
> >>
> >>
> >>
> >>
> >>
>
>
> Sent from Mailspring
>
Sent from Mailspring
Join devel@edk2.groups.io to automatically receive all group messages.