Re: [dpdk-ci] Understanding results of patchwork

[Aaron Conole <aconole@redhat.com>]

Ferruh Yigit <ferruh.yigit@intel.com> writes:

> On 4/15/2020 5:42 PM, David Marchand wrote:
>> On Wed, Apr 15, 2020 at 6:34 PM Ferruh Yigit <ferruh.yigit@intel.com> wrote:
>>>
>>> On 4/15/2020 3:40 PM, Ajit Khaparde wrote:
>>>> Ferruh,
>>>> I am trying to apply this patchset.
>>>> https://patchwork.dpdk.org/project/dpdk/list/?series=9386
>>>>
>>>> So I am trying to check if the patches passed all the checks.
>>>> I can see only the checkpatch for individual patches and the
>>>> travis result - presumably for the complete set.
>>>>
>>>> https://travis-ci.com/ovsrobot/dpdk/builds/160349438
>>>>
>>>> For previous versions, I had seen results for individual patches and performance
>>>> results also. Is it enough to apply the patches based on just the travis and
>>>> checkpatch results?
>>>>
>>>
>>> Hi Ajit,
>>>
>>> We expect other checks, cc'ed David too.
>> 
>> Better to Cc: ci@dpdk.org from my pov.
>
> Agree, cc'ed now.
>
>> 
>> 
>>>
>>> Intel-compilation seems not sending the reports for a day, last one is from
>>> yesterday, it may be stuck in some set, etc.. @Zhaoyan, can you please check it?
>>>
>>> The community lab one is still on pending state, it may be because it is a big
>>> patchset.
>>> https://lab.dpdk.org/results/dashboard/patchsets/10445/
>>>
>>>
>>> And @David, @Aaron, does travis checks patchset only or does it checks each
>>> patch in set?
>> 
>> Travis checks the whole patchset at the last patch.
>
> Thanks, so we are missing patch by patch checks in our CI checks, since as far
> as I know Intel check is also for patchset.

For now, we only check the series as a whole in travis to save on
time.  It's expected that someone / something else would do
patch-at-a-time for the series, to preserve the ability to bisect
through series.

>> The dpdk.org server sends checkpatch reports for each patch.
>>