Re: [PATCH v2 00/17] Introduce runtime modifiable Energy Model

From: Lukasz Luba
Date: Mon Jul 03 2023 - 12:35:13 EST




On 5/30/23 12:07, Dietmar Eggemann wrote:
On 12/05/2023 11:57, Lukasz Luba wrote:
Hi all,

This patch set adds a new feature which allows to modify Energy Model (EM)
power values at runtime. It will allow to better reflect power model of
a recent SoCs and silicon. Different characteristics of the power usage
can be leveraged and thus better decisions made during task placement in EAS.

It's part of feature set know as Dynamic Energy Model. It has been presented
and discussed recently at OSPM2023 [3]. This patch set implements the 1st
improvement for the EM.

Why is the feature set called Dynamic Energy Model?

Dynamic Energy Model:

Runtime modifiable EM

Proper CPU performance state evaluation

CPU idle wakeup costs

CPU capacity as new EM data

Didn't this `Dynamic` stand for the modifiability of the EM only?

The 'modifiability' is the main feature, but not the last one.

The 2nd: 'Proper CPU performance state evaluation' is also
a 'dynamic' thing, which is related to the currently set CPU
frequency (for Mid or Big) and the dependent Little's and L3
cache frequency. As you know that frequency can change in time,
so it's 'dynamic' situation and will lands to the 'Dynamic EM'.

(The 2 below need more thinking and experiments)
The 3rd can be also quite dynamic. You might change the wake-up cost
for in the EM if you want to avoid waking up big cores in some
workloads. You might pay penalty for bigger latency, because tasks
would be more queued on Mids/Littles, but that's for power
saving scenario.

The 4th is about CPU capacity. We still have to conduct more
investigations, but it might be useful to change the EM
and provide a new CPU capacity from it to the OS. In the 3-gear
SoC the CPUs might have quite different capacity in different
scenarios (workloads). This could be also a 'dynamic' thing,
e.g. triggered by middle-ware for a long running video call.


The concepts:
1. The CPU power usage can vary due to the workload that it's running or due
to the temperature of the SoC. The same workload can use more power when the
temperature of the silicon has increased (e.g. due to hot GPU or ISP).
In such situation or EM can be adjusted and reflect the fact of increased
power usage. That power increase is due to a factor called static power
(sometimes called simply: leakage). The CPUs in recent SoCs are different.
We have heterogeneous SoCs with 3 (or even 4) different microarchitectures.
They are also built differently with High Performance (HP) cells or
Low Power (LP) cells. They are affected by the temperature increase
differently: HP cells have bigger leakage. The SW model can leverage that
knowledge.
2. It is also possible to change the EM to better reflect the currently
running workload. Usually the EM is derived from some average power values
taken from experiments with benchmark (e.g. Dhrystone). The model derived
from such scenario might not represent properly the workloads usually running
on the device. Therefore, runtime modification of the EM allows to switch to
a different model, when there is a need.
3. The EM can be adjusted after boot, when all the modules are loaded and
more information about the SoC is available e.g. chip binning. This would help
to better reflect the silicon characteristics. Thus, this EM modification
API allows it now. It wasn't possible in the past and the EM had to be
'set in stone'.

Some design details:
The internal mechanisms for the memory allocation are handled internally in the
EM. Kernel modules can just call the new API to update the EM data and the
new memory would be provided and owned by the EM. The EM memory is used by
EAS, which impacts those design decisions. The EM writers are protected by
a mutex. This new runtime modified EM table is protected using RCU mechanism,
which fits the current EAS hot path (which already uses RCU read lock).
The unregister API handles only non-CPU (e.g. GPU, ISP) devices and uses the
same mutex as EM modifiers to make sure the memory is safely freed.

This only mentions memory allocation and locking? A global design
overview, containing e.g.

Why 2 tables, modifiable (a) and default (b)?

Why does only EAS use (a)?

(a) and (b) being the same performance state table until first call to
modify (a) ()

as an introduction into the patches would be more helpful here.

More detailed explanation and background can be found in presentations
during LPC2022 [1][2] or in the documentation patches.

I checked 15/17 as well but could find any of this information there either.

[...]

For the above two comments: Yes, I see your point and will address that
in the next v3.

Thank you fro your valuable comments and time for reviewing it!

Lukasz