[v2,0/3] Add lcore poll busyness telemetry

  Currently, there is no way to measure lcore busyness in a passive way,
without any modifications to the application. This patchset adds a new EAL
API that will be able to passively track core busyness. As part of the set,
new telemetry endpoints are added to read the generate metrics.

Anatoly Burakov (2):
  eal: add lcore poll busyness telemetry
  eal: add cpuset lcore telemetry entries

Kevin Laatz (1):
  doc: add howto guide for lcore poll busyness

 config/meson.build                          |   1 +
 config/rte_config.h                         |   1 +
 doc/guides/howto/lcore_busyness.rst         |  79 +++++
 lib/bbdev/rte_bbdev.h                       |  17 +-
 lib/compressdev/rte_compressdev.c           |   2 +
 lib/cryptodev/rte_cryptodev.h               |   2 +
 lib/distributor/rte_distributor.c           |  21 +-
 lib/distributor/rte_distributor_single.c    |  14 +-
 lib/dmadev/rte_dmadev.h                     |  15 +-
 lib/eal/common/eal_common_lcore_telemetry.c | 340 ++++++++++++++++++++
 lib/eal/common/meson.build                  |   1 +
 lib/eal/include/rte_lcore.h                 |  80 +++++
 lib/eal/meson.build                         |   3 +
 lib/eal/version.map                         |   7 +
 lib/ethdev/rte_ethdev.h                     |   2 +
 lib/eventdev/rte_eventdev.h                 |  10 +-
 lib/rawdev/rte_rawdev.c                     |   5 +-
 lib/regexdev/rte_regexdev.h                 |   5 +-
 lib/ring/rte_ring_elem_pvt.h                |   1 +
 meson_options.txt                           |   2 +
 20 files changed, 584 insertions(+), 24 deletions(-)
 create mode 100644 doc/guides/howto/lcore_busyness.rst
 create mode 100644 lib/eal/common/eal_common_lcore_telemetry.c
  

> From: Kevin Laatz [mailto:kevin.laatz@intel.com]
> Sent: Wednesday, 24 August 2022 18.25
> 
> Currently, there is no way to measure lcore busyness in a passive way,
> without any modifications to the application. This patchset adds a new
> EAL
> API that will be able to passively track core busyness. As part of the
> set,
> new telemetry endpoints are added to read the generate metrics.
> 
> Anatoly Burakov (2):
>   eal: add lcore poll busyness telemetry
>   eal: add cpuset lcore telemetry entries
> 
> Kevin Laatz (1):
>   doc: add howto guide for lcore poll busyness
> 

At first glance, reading the commit messages, I thought you had ignored the feedback.
However, I see that V2 adds "poll" in front of "busyness".

It's just missing in many locations in the code, e.g. in eal_common_lcore_telemetry.c, the telemetry command is still registered as "/eal/lcore/busyness".

I recommend that you search for "business" in the files, and add "poll" where it is missing.

Perhaps one of your testers can do it for you... I know from experience that it can be difficult spotting details like this in your own code, whereas many test people are good at this stuff. There's even a company here in Denmark offering autistics as testers - they are absolutely brilliant at tasks like this!

-Morten
  

On 25/08/2022 08:47, Morten Brørup wrote:
>> From: Kevin Laatz [mailto:kevin.laatz@intel.com]
>> Sent: Wednesday, 24 August 2022 18.25
>>
>> Currently, there is no way to measure lcore busyness in a passive way,
>> without any modifications to the application. This patchset adds a new
>> EAL
>> API that will be able to passively track core busyness. As part of the
>> set,
>> new telemetry endpoints are added to read the generate metrics.
>>
>> Anatoly Burakov (2):
>>    eal: add lcore poll busyness telemetry
>>    eal: add cpuset lcore telemetry entries
>>
>> Kevin Laatz (1):
>>    doc: add howto guide for lcore poll busyness
>>
> At first glance, reading the commit messages, I thought you had ignored the feedback.
> However, I see that V2 adds "poll" in front of "busyness".

I'll make changes to the message to clarify this a bit better.

> It's just missing in many locations in the code, e.g. in eal_common_lcore_telemetry.c, the telemetry command is still registered as "/eal/lcore/busyness".
>
Seems some builds are failing in the CI so I'll send a v3 with fixes 
shortly - I'll do another sweep before sending to make sure there are no 
omissions.

-Kevin