[dpdk-dev] [PATCH v17 00/11] Add PMD power management

[Burakov, Anatoly]

On 18-Jan-21 4:06 PM, Thomas Monjalon wrote:
> 18/01/2021 16:45, Burakov, Anatoly:
>> On 18-Jan-21 3:24 PM, David Marchand wrote:
>>> On Thu, Jan 14, 2021 at 3:46 PM Anatoly Burakov
>>> <anatoly.burakov at intel.com> wrote:
>>>>
>>>> This patchset proposes a simple API for Ethernet drivers to cause the
>>>> CPU to enter a power-optimized state while waiting for packets to
>>>> arrive. There are multiple proposed mechanisms to achieve said power
>>>> savings: simple frequency scaling, idle loop, and monitoring the Rx
>>>> queue for incoming packages. The latter is achieved through cooperation
>>>> with the NIC driver that will allow us to know address of wake up event,
>>>> and wait for writes on that address.
> [...]
>>>> Why are we putting it into ethdev as opposed to leaving this up to the
>>>> application? Our customers specifically requested a way to do it with
>>>> minimal changes to the application code. The current approach allows to
>>>> just flip a switch and automatically have power savings.
> 
> The customer laziness is usually a bad justification :)
> I think we could achieve the same with not too much code
> on application side.

Yes, we could. Customers could basically take this patch and reimplement 
it inside their application, and get the same benefits (with also added 
benefit of having knowledge about their queue/core mapping, and so being 
able to use the PAUSE or SCALE schemes for more than one queue).

However, i still think it's a valid use case - if we can do it that way 
and have a ready-made power management story, why not?

> And I'm not sure hiding queue management is sane.
> Remember this rule: application must remain in control.
>

The application can still be in control by just not using the API and 
implementing things manually instead. Nothing is being taken away from 
the ability of application to be in control.

> [...]
>>> SPDK build is still broken.
>>> http://mails.dpdk.org/archives/test-report/2021-January/174840.html
> [...]
>>> I guess this is because of the added dependency of rte_ethdev to rte_power.
>>> Afaics, SPDK does not use pkg-config:
>>> https://github.com/spdk/spdk/blob/master/lib/env_dpdk/env.mk#L53
>>
>> Sooo... this is an SPDK issue then? Because i can't see any way of
>> fixing the issue on DPDK side.
> 
> Yes SPDK should not skip pkg-config.
> But it raises 2 question:
> 	- are we breaking ABI compatibility?

Good question. Does including an extra intra-DPDK dependency count as 
ABI break? I was under impression that we didn't want DPDK to be 
distributed as individual libraries but rather would like it to be used 
as a whole, so if internal dependencies between components change, it's 
not a big deal (unless a third-party build system is used that 
explicitly specifies dependencies rather than using pkg-config).

> 	- is ethdev management expected for librte_power?
> 
> It makes me wonder whether we should host the few functions mixing
> librte_ethdev and librte_power somewhere else.
> The question is where?
> 

That could be another possibility. We could put this into a separate 
library, but IMO it would serve no purpose other than avoiding adding a 
dependency on *internal* component to librte_power. I'm not sure it's a 
worthy trade off.

-- 
Thanks,
Anatoly