Re: [EXTERNAL] Re: [PATCH v2 0/3] testing: Build WHPX enabled binaries

[Philippe Mathieu-Daudé]

Hi Sunil,

On 8/1/20 1:31 AM, Sunil Muthuswamy wrote:
>> Hi Justin, Sunil,
> 
> Justin has moved to a different team is no longer working with WHPX. Moving 
> him
> to bcc.

OK. Does that mean you are the new responsible of updating the ticket
regarding the WHPX headers and their license?

> 
>>
>> On 5/20/20 12:26 PM, Philippe Mathieu-Daudé wrote:
>>> +launchpad ticket
>>>
>>> On 9/20/19 6:53 PM, Justin Terry (VM) wrote:
>>>> Hey Phil,
>>>>
>>>> I have contacted our legal department for guidance on this specific
>>>> use case and will update you when I hear back. Thank you for your
>>>> patience.
>>
>> I recently understood legal changes can be very complex, thus it is
>> implicit it can take years before getting updates.
>>
>> Since the project is still actively developed, maybe you could provide
>> a Azure CI job to build a WHPX binary. We don't need to have access to
>> the binary, just to the exit status (success/fail) and build logs.
>>
>> Do you think it is doable?
>>
>> Thanks,
>>
>> Phil.
>>
> The ask generally sounds reasonable. But, can you help me understand the full
> scope of the ask. Few questions:
> 1. Stefan has a CI pipeline to build WHPX.

Great! I didn't know Stefan already did it :)
Can you share the URL please, so we can integrate it with mainstream CI?

> What's the benefit of having another CI
> job, that doesn't export the binary, but, just the status?

As usual, we do not want to circumvent the license. IANAL but IIUC we
can not force a CI job to accept the EULA when installing it, even to
test it. So the best we can do is check if the build succeeded (exit
status).

> 2. Which branch is the CI pipeline expected to build?

'master', to be sure no regressions are introduced.

> 3. Is the expectation also that it will build WHPX patches that are submitted 
> to the
> WHPX branch?

You describe a "downstream CI" testing, which is out of scope of the
community public CI.

Regards,

Phil.